text
stringlengths 254
1.16M
|
|---|
---
title: 'Child Tax Credit, Spending Patterns, and Mental Health: Mediation Analyses
of Data from the U.S. Census Bureau’s Household Pulse Survey during COVID-19'
authors:
- JungHo Park
- Sujin Kim
journal: International Journal of Environmental Research and Public Health
year: 2023
pmcid: PMC10002275
doi: 10.3390/ijerph20054425
license: CC BY 4.0
---
# Child Tax Credit, Spending Patterns, and Mental Health: Mediation Analyses of Data from the U.S. Census Bureau’s Household Pulse Survey during COVID-19
## Abstract
This study examined the relationship between the receipt of COVID-19 child tax credit and adult mental health problems in the United States, and we explored whether and the extent to which a wide range of spending patterns of the credit—15 patterns regarding basic necessities, child education, and household expenditure—mediated the relationship. We used COVID-19-specialized data from the U.S. Census Bureau’s Household Pulse Survey, a representative population sample ($$n = 98$$,026) of adult respondents (18 and older) who participated between 21 July 2021 and 11 July 2022. By conducting mediation analyses with logistic regression, we found relationships between the credit and lower levels of anxiety (odds ratio [OR] = 0.914; $95\%$ confidence interval [CI] = 0.879, 0.952). The OR was substantially mediated by spending on basic necessities such as food and housing costs (proportion mediated = $46\%$ and $44\%$, respectively). The mediating role was relatively moderate in the case of spending on child education and household expenditure. We also found that spending the credit on savings or investments reduces the effect of the child tax credit on anxiety (−$40\%$) while donations or giving to family were not a significant mediator. Findings on depression were consistent with anxiety. The child tax credit–depression relationships were substantially mediated by spending on food and housing (proportion mediated = $53\%$ and $70\%$). These mediation analyses suggested that different patterns of credit spending are important mediators of the relationship between the receipt of the child tax credit and mental illnesses. Public health approaches to improve adult mental health during and after the COVID-19 pandemic need to consider the notable mediating role of spending patterns.
## 1. Introduction
As the Coronavirus Disease 2019 (COVID-19) pandemic enters its fourth year, many countries around the globe are in a better place with dealing with the disease but still face a crisis in relation to people’s mental health [1,2,3,4]. The pandemic has increased the risks related to poor mental health—stress and fear, loss of income, job insecurity, etc.—whereas defensive aspects—outdoor activities, socialization, educational opportunities, accessibility to health care services, etc.—worsened markedly [5,6]. The accumulated scale of mental health problems is so enormous that it calls for comprehensive and all-encompassing societal support across the globe [6]. The Global Burden of Diseases, Injuries, and Risk Factors (GBD) 2020 study aligned by the World Health Organization (WHO) estimated that, during the COVID-19 pandemic, symptoms of depression and anxiety were the most prevalent mental disorders and they increased by $27.6\%$ and $25.6\%$, respectively, across the world [7].
In the United States, in response to the global health crisis, a wide range of economic and social safety net programs have been newly developed (e.g., stimulus check) and expanded (e.g., child tax credit, unemployment insurance). Daily conflicts between work and life domains among numerous American households, especially for those with kids, can be eased by programs and policies that recognize a variety of circumstances in which normal households find themselves [5,8,9,10]. Recent studies have reported that socioeconomic programs and policies have proved successful in helping to directly or indirectly improve the mental health of beneficiaries [11,12]. Beyond the general understanding of the link, however, the underlying mechanisms linking government support to mental health have yet to be examined.
## 1.1. Expansion of the U.S. Child Tax Credit Amid the Crisis of Mental Health during COVID-19
Considering the enormous socioeconomic difficulties among American households resulting from the COVID-19 pandemic, the American Rescue Plan Act of 2021 made critical expansions to the child tax credit (hereafter, CTC). CTC was first introduced by The Taxpayer Relief Act of 1997 in order to ease the economic burden on households with children. CTC is a refundable tax credit for dependent children. The tax amount of households with federal income tax to be paid is deducted, and cash benefits are also paid from the federal government.
An important element of the American Rescue Plan Act included an expansion of CTC with advance payments starting from July 2021, a “child allowance” which was expected to sharply elevate the level of child poverty [13]. The amount of CTC was increased from USD 2000 to USD 3600 for eligible children (aged 5 or younger) and USD 3000 for other eligible children (aged 17 or younger). To provide much-needed economic relief, the American Rescue Plan Act revised the credit to be entirely refundable and provided on a monthly basis for the initial six months in contrast with the previous rate of once per year. By the middle of July 2022, $88\%$ of households with children received payments of USD 250 (or USD 300) per child every month. The changes included expanding the qualification to age 17, making the CTC entirely refundable, raising the credit for wide lower- and middle-income taxpayers, with greater increases for younger kids, and disbursing half of these payments every month rather than after a household files their taxes, beginning in July 2021 [14].
As is the case with CTC in the U.S., many countries across the world expanded their child allowance (interchangeably child benefit, [15,16]) during and even before the COVID-19 pandemic. This included child benefit in the U.K., Canadian child benefits in Canada, child tax benefit in South Korea, and child allowance in Japan, among many others. A growing body of COVID-19 studies in the U.S. has also found that there are relationships between income supplements and mental health, such as stimulus checks [17], economic impact payment [18], unemployment insurance [19], supplemental nutrition assistance programs [20], and temporary assistance for needy families [21,22], among many others. Focusing on child mental health, studies identified the positive role of cash assistance. For example, at ages 25 and 30, participants from members of an American Indian tribe whose households received cash payments experienced less anxiety and depression, indicating that cashable payments to households may result in longer-term benefits for children’s mental wellbeing [23].
Yet, there is restricted knowledge of the fundamental processes between the receipt of CTC and adult mental health. On the one hand, the cash benefit may directly improve mental health simply because of the receipt of the credit. During the pandemic, policies that have provided greater financial assistance or reduced the risk of financial insecurity have also been shown to provide opportunities for good health or to directly improve health [22]. On the other hand, there may be indirect ways through which the receipt of CTC may improve mental health. CTC may also increase the perceived manageability of debt while easing mental health stress [24].
## 1.2. Diverse Spending Patterns of the Child Tax Credit
Maslow’s hierarchy of needs may explain the human motivation behind consumer behaviors and spending patterns, particularly in urgent situations such as the COVID-19 pandemic [25,26,27]. The theory describes five hierarchical levels—physiology, safety, love and belonging needs, esteem, and self-actualization—which may be related to a wide range of spending patterns of cashable CTC in the COVID-19 pandemic. Moreover, the COVID-19 crisis increased consumer fear and uncertainty in relation to spending decisions due to the loss of income and fears of contagion [26]. Therefore, consumers may focus more on satisfying basic needs than on fulfilling higher levels of needs [28].
Considerable evidence supports the primary uses of CTC for basic needs [29]. CTC allowed households to cover daily expenses, including housing cost, more and higher-quality food, clothing, and other necessities for their children [30,31]. The expanded monthly payments might help support adults in making ends meet when pandemic-induced inflation raises the prices of essential items [32]. Especially for the poorest parents with children, the added monthly income from the credit helps them secure urgent daily items [32].
In particular, previous literature found that missing routine meals due to economic issues is among the most serious hardships related to mental health problems [10]. CTC significantly lowered qualified households’ food insecurity and helped them afford balanced and healthy foods for their kids and pregnant women [33]. Since the distribution of CTC, food security has improved dramatically for all racial/ethnic subgroups, but particularly for Black and Latino people [34]. Although adult members of households with kids are more likely to suffer a lack of food, the households experienced a three percentage point reduction between the surveys performed before and after the credit payments [35]. Food insecurity is related to mental illnesses due to the consequent fear, depression, anxiety, stigma, and stress [36]. In case of Canada, lack of food was independently related to poor mental health in the early days of the COVID-19 pandemic [37].
As for expenses relating to child education during the pandemic, parents have faced unprecedented hardships due to limited transportation from and to school, isolation measures, and the closure of childcare centers and schools [38]. CTC recipients may spend the payments on educational materials and class activities for their kids. A recent study reported that the credit payments resulted in a significant difference in parents’ capacity to pay off basic school items and class activities for their kids, benefiting children’s health and educational opportunities [32]. Pre-pandemic studies also found that kids in households who benefited from income support showed better mental health in adulthood. This implies a positive and long-term benefit of CTC, which will be realized in the future [39,40].
In the case of household expenditure, CTC effectively and efficiently lowered financial burdens for qualified households, as demonstrated by their reduced credit card debt and lower risk of eviction. The significant expansions have enormously lowered child poverty, lifting an added 4.1 million kids above the nationwide poverty level by $40\%$ [41]. CTC also helped households save for emergencies, pay off debt [30], and work additional hours outside the home [32]. During the prevention and control of COVID-19, social support could help reduce a variety of symptoms of mental illnesses [42]. Additionally, a greater amount of debts were associated with a higher level of stress and, in turn, worse mental health [43].
These different spending patterns of CTC may be mediators in the relationship between the receipt of CTC and mental health during the pandemic. The National Child Tax Credit Survey showed that the credit payments have successfully lowered economic stress, with $70\%$ of survey participants stating that the payments made them much less stressed about household expenditure [41]. In contrast, a study found a series of mixed results regarding the effects of CTC on adult depression and parental stress [30].
Given the historical expansion of CTC and its substantial effects on American households, it is important to examine the relationships between the CTC monthly payments and mental health outcomes in the pandemic. Parental mental health plays a central role in securing the psychological well-being of the entire household, reducing parenting irritability, parental burnout, and verbal conflict between couples [44,45,46,47,48]. Furthermore, studies have revealed that deterioration of parental mental health occurs due to not only social distancing and closures [49,50,51] but also extended time spent on childcare and homeschooling [5]. Understanding the effects of public assistance, including CTC, on adult mental well-being is also critical for informing public health policies that better resolve mental health needs related to urgency and enhance the psychological well-being of adult parents and their household members.
Few population-based studies have focused on whether and the extent to which a wide range of spending patterns of the CTC monthly payments may mediate the relationship between CTC and mental health in the pandemic. Population-based research is critical since clinical study cases may not represent the entire population. Moreover, a wide range of daily spending patterns may be critical to draw health implications for the entire population.
We aim to contribute to an expanding body of literature that shows the importance of social policies on mental health, particularly during the pandemic [22]. This study used population-based and pooled cross-sectional datasets to analyze the relationship between the receipt of CTC and adult mental health problems. We also explored whether and the extent to which fifteen different patterns of the credit usage mediated the relationship between the receipt of the credit and mental health outcomes.
Figure 1 shows an overall conceptual framework of our study that attempts to link CTC (exposure) and mental health problems (outcome) with salient mediators of 15 different spending patterns of CTC. The following section describes the Household Pulse Survey data—primary data of the study—and variable definitions, along with the specification of mediation analysis models.
## 2.1. U.S. Census Bureau’s HPS Data during COVID-19
The Household Pulse Survey (HPS) is a nationally representative survey deployed by the U.S. Census Bureau and the U.S. National Center for Health Statistics (NCHS), as well as other federal organizations. It surveys the socioeconomic and health impacts of the pandemic on adult households in the U.S. HPS was conducted biweekly (or weekly in early periods of the pandemic) and largely consists of three phases starting from 23 April 2020: phase 1 (23 April–21 July 2020, survey weeks 1 to 12), phase 2 (19 August–26 October 2020, survey weeks 13 to 17), and phase 3 and following subphases (28 October 2020–ongoing, survey weeks 18 and later). We utilized a one-year portion of HPS (21 July 2021–11 July 2022, survey weeks 34 to 47) when new survey questions about CTC were introduced and became available for analysis. Note that we did not use data collected later than survey week 47 because CTC-related questions were not asked anymore. This study utilized the Public Use File (PUF) of HPS—microdata which are free to download—which provides survey answers from individual respondents (see Supplementary Table S1 for details about the sample size of PUF microdata by survey week and phase).
## 2.2.1. Outcome: Mental Health Problems
Two types of self-reported measures—Generalized Anxiety Disorder (GAD; GAD-2 [52]) and Major Depressive Disorder (MDD; PHQ-2 [53])—were adopted to identify the level of mental health of CTC recipients. The two questions measure the frequencies of the symptoms of anxiety and depression in the past week. The base question of GAD-2 and PHQ-2 is “in the past week, how often have you been distracted by any of the following difficulties?” The two subitems of GAD-2 are “feeling nervous, anxious or on edge” and “cannot stop or control worrying” while the items for PHQ-2 are “having little interest or pleasure in doing things” and “feeling down, depressed, or hopeless.” The responses from survey participants were coded by integers, such as not at all = 0, several days = 1, more than half the days = 2, and nearly every day = 3. Values for each item were summed and then categorized into binomial outcomes, such as four or higher points from GAD-2 as GAD and from PHQ-2 as MDD. The thresholds of PHQ-2 and GAD-2 have been validated for diagnosed GAD and MDD [52,53] (see Supplementary Table S2 for more information regarding survey questions and answers).
## 2.2.2. Exposure: Receipt of the Child Tax Credit
We specified a binary exposure variable using the following survey question: “In the last 4 weeks, did you receive a refund from your 2021 tax return?” with answer options of yes (=1) or no (=0). To narrow down the sample of our analysis, we dropped respondents who did not respond to another CTC-related question ‘Considering your spending of the CTC monthly payments, did you: (a) mostly spend it, (b) mostly save it, or (c) mostly use it to pay off debt’.
## 2.2.3. Mediator: Spending Patterns of the Child Tax Credit
We measured the spending patterns of CTC on the basis of a multiple-choice survey question: “What did you and your household mostly spend the “Child Tax Credit” portion of your refund on? Select all that apply”. Answer options of yes or no were available for the 15 different patterns of CTC spending, which were grouped into three types: CTC spending on basic necessities (food, rent or mortgage, and clothing), CTC spending on child education (childcare, schoolbooks and supplies, school tuition, tutoring services, afterschool programs, transportation for school, and recreational goods), and CTC spending on household expenditure (utilities and telecommunications, vehicle payments, paying off credit cards or debts, savings or investments, and donations or giving money to family). We ran the same mediation analysis 15 times using one mediating variable at a time to avoid multicollinearity between the mediators (see Pearson correlation coefficients between −0.001 and 0.3925 as shown in Supplementary Table S3).
## 2.2.4. Covariate: Characteristics of Survey Participants
We considered individual and household characteristics ranging from demographic attributes to social and economic statuses (SES), health insurance status, and location of residence. Demographic characteristics consisted of age, sex, race and ethnicity, marital status, count of kids, and number of household members. SES included education and household income. To control for health-related covariates, we considered the status of public and private health insurance. In addition, two sets of geographic identifiers were included in the model to control for the location of residence of survey respondents, such as 50 states and the Washington, D.C., and 15 largest metropolitan statistical areas (MSAs). These geographic variables were included to reflect differences in the mental health outcomes between distinct areas across the nation.
Table 1 shows descriptive statistics of variables of the entire cases, as well as separately for the diagnosis of anxiety and depression. The overall effect of the pandemic on mental health was not distributed equally across the population subgroups, which is in line with previous literature on relationships between public support and mental health [19,20,21,22,32,34,35,54].
## 2.3. Model Specification
We adopted Stata MP version 13.1 program (StataCorp, College Station, TX, USA) across analysis models We used logistic regression models for our mediation analysis. We used mediation analyses (paramed in Stata program; [55]) which were developed by VanderWeele [56]. The method allows researchers to decompose a total effect into direct and indirect effects on the basis of counterfactual framework. VanderWeele’s method can also address limitations of the traditional approach developed [57], which omits potential interrelationships between exposure variables and mediating variables. By using 500 bootstrapping resamples and producing $95\%$ bias-corrected confidence intervals, we estimated direct and indirect effects in the models. Additionally, we estimated how much of the total effect is mediated in the model using the following equation: odds ratio [indirect effect]/odds ratio [total effect] × $100\%$.
## 3.1. Spending Patterns of CTC and Anxiety
The first model result (CTC spent on food) at the top of Table 2 shows a negative relationship between the receipt of CTC and GAD (odds ratio [OR] of total effect = 0.914; $95\%$ confidence interval [CI] = 0.879, 0.952) after controlling for all covariates. The association between CTC and GAD was significantly mediated by the spending pattern of using CTC to purchase food by $46\%$ (OR of indirect effect = 0.958; $95\%$ CI = 0.938, 0.980). Additionally, CTC spending on housing costs (rent or mortgage) substantially mediated the association between CTC and GAD by $44\%$. In contrast, using CTC to buy clothing did not significantly mediate the association between CTC and GAD.
As for CTC spending related to child education, we find that the association between CTC and GAD is significantly mediated when the recipients use the credit to pay for childcare (OR of indirect effect = 0.988; $95\%$ CI = 0.981, 0.995, proportion mediated = $13\%$), school tuition (OR of indirect effect = 0.994; $95\%$ CI = 0.989, 0.999, proportion mediated = $7\%$), and transportation costs of traveling to and from school (OR of indirect effect = 0.988; $95\%$ CI = 0.981, 0.995, proportion mediated = $13\%$). Other spendings on child education—schoolbooks and supplies, tutoring services, afterschool programs, and recreational goods, etc.—are insignificant or weakly significant mediators in the association between CTC and GAD.
Most CTC spending on household expenditure emerges as a significant mediator in the relationship between CTC and GAD. Using CTC to pay for utilities and telecommunications significantly mediated the relationship between CTC and GAD by $36\%$. Similarly, CTC spending on vehicle payments and credit cards or debts mediated the relation between CTC and GAD by $13\%$ and $27\%$, respectively. Unlike all the other mediators in this article, the use of CTC on savings or future investments appears to be associated with a higher level of GAD (OR of indirect effect = 1.043; $95\%$ CI = 1.026, 1.060, proportion mediated = −$40\%$). Spending CTC on charitable donations or giving to family did not have a significant mediating role in the model.
## 3.2. Spending Patterns of CTC and Depression
Table 3 includes mediation analysis results with MDD as the outcome variable, which is mostly consistent with the model result with GAD. When using CTC spending on basic necessities as a mediator, we find negative associations between CTC and MDD across models with different mediators. The relationships between CTC and MDD were significantly and substantially mediated by CTC spending on food (proportion mediated = $53\%$) and housing costs (proportion mediated = $70\%$). Notably, more than half of the total effects of CTC on MDD were mediated by those spending patterns. CTC spending on clothing was not a significant mediator.
Among CTC spending on child education, we find a significant mediating role only in the cases of childcare spending (OR of indirect effect = 0.989; $95\%$ CI = 0.981, 0.997, proportion mediated = $13\%$) and cost of transportation to and from school (OR of indirect effect = 0.997; $95\%$ CI = 0.995, 1.000, proportion mediated = $4\%$). The other spendings on child education were not a significant mediator in the model; these included schoolbooks and supplies, school tuition, tutoring services, afterschool programs, and recreational goods.
Turning to CTC spending on household expenditure, we discovered significant mediating roles when the credit was spent on utilities and telecommunications (OR of indirect effect = 0.968; $95\%$ CI = 0.957, 0.980, proportion mediated = $46\%$), vehicle payments (OR of indirect effect = 0.990; $95\%$ CI = 0.984, 0.997, proportion mediated = $13\%$), and paying off credit cards or debts (OR of indirect effect = 0.976; $95\%$ CI = 0.963, 0.989, proportion mediated = $29\%$). As was the case for the GAD model, CTC spending on savings or investments was associated with a higher level of MDD (OR of indirect effect = 1.029; $95\%$ CI = 1.008, 1.051, proportion mediated = −$33\%$). Spending CTC on charitable donations or giving money to family was not a significant mediator in the model.
## 4.1. Key Findings of Mediation Analyses
Overall findings indicate that CTC recipients in the COVID-19 pandemic are at notably decreased risk of anxiety and depression and that a substantial proportion of this lowered risk stems from spending patterns of the credit on basic necessities, child education, and household expenditure. These findings correspond with previous results about a wide range of spending patterns of CTC and lowered level of mental illnesses among CTC recipients [24,30,31,32,34], suggesting that the risk of mental health problems is lowered for CTC recipients who spent the monthly credit on life essentials during the pandemic.
Our findings show significant associations of the receipt of CTC with anxiety and depression, in addition to the important mediating roles of spending patterns of the credit in the U.S. in the pandemic. The findings are consistent with early COVID-19 studies in that CTC plays a positive role in alleviating mental illnesses during the pandemic. Very few studies have explored whether a variety of spending patterns of CTC might be mediators of the relationship between the receipt of the credit and mental health outcomes. Furthermore, we have limited knowledge about the extent to which these relationships are mediated by the spending pattern of the credit. We found that some specific spending patterns were partial but significant mediators of the association between the receipt of CTC and mental health outcomes, while others were not.
We made three key discoveries about the relationships between the receipt of CTC, spending patterns, and mental health outcomes. First, we found the strongest mediators were spendings on food and housing as people’s most fundamental needs. In particular, the expanded CTC helps low- and moderate-income households to decrease their financial stresses as it allows them to buy daily necessities, increasing opportunities for children’s education [32]. Additionally, permanent payment of expanded benefits may reduce food insecurity during the pandemic situation [34] and help recipients secure stable housing [58]. The need for continued expansion of CTC in terms of mental health improvement is also found in other government programs such as EITC [59]. After expiration of CTC monthly payments, food insufficiency was increased in households with children [60] and they experienced poverty, especially in households including Black and Latinx children [32].
Second, we found that spending of CTC on childcare and transportation for school mediated the relationship between the receipt of CTC and mental health problems. Parents used the CTC to buy toys and engage in activities with their children; thus, the bond between parents and children improved. Additionally, parents purchased more food or high-quality food [41]. A third of spending CTC was school related [35] and educational opportunities increased [32]. In addition, our results showed that use of CTC for household expenditure was a significant mediator of the relationship. The cash form of CTC appears to play a crucial role in securing flexibility and diversity of uses of the credit. In terms of flexibility, a recent study supports the rising importance of flexibly designed mental health measures interventions for both kids and parents [48]. The current policy trend of low-income and middle-income countries is being adapted or expanded as cash transfer programs in order to overcome the pandemic crisis [11]. These cash transfer programs should not only deal with food insecurity but should also focus on addressing long-term mental health disruptions resulting from the ongoing pandemic. A study by Brookings suggests that CTC is an effective tool in terms of cutting child poverty in the short term and will help to increase household social mobility in the long term [61]. *More* generous cash transfer is a powerful tool, relieving the negative mental, physical, and behavioral health responses to stress created by unemployment and wage loss [22]. Continued and regular cash payment is important, as it alleviates stress [58].
Third, we found spending on savings and investments was related to a higher level of anxiety and depression, which indicates the opposite role against other mediators used in our study. This finding calls for policy makers to consider negative effects of CTC on recipient households and their consumption. We need to consider that the prolonged COVID-19 situation aggravated household’s economic burdens [45]. This is different from the situation in which people received public subsidies in the early stages of COVID-19; at that time, parents spent these subsidies on daily necessities and debts [35]. On the other hand, spending on saving or investment in this study was intended to prepare for financial difficulties due to prolonged COVID-19; thus, saving or investments were not positive acts for mental health. In addition, the expanded CTC may result in low-income earners making less of an effort to find work [62], which will have the opposite effect, leading to a decrease in income and household assets. In this situation, it is judged that savings and investment negatively affect mental health because CTC should be prioritized in food and housing expenditure, and savings or investments can be made later.
## 4.2. Limitations and Future Research
Our findings need to be considered in light of four limitations. First, the HPS did not include previous and chronic indicators of the survey participants’ mental health, and therefore we were not able to control for pre-pandemic existence of a diagnosis of mental illness. Second, as our study did not analyze the initial pandemic situation when CTC was not expanded, we could not compare how much the mental health of parents improved in response to CTC. Third, we did not consider temptation goods such as cigarettes or alcohol as a food purchasing item in the HPS checklist, as food purchasing only included groceries, eating out, and take out. Thus, we could not find whether temptation goods affected the improvement of mental health [63]. Fourth, we did not compare the characteristics of consumption patterns according to differences in household income and other socioeconomic characteristics which may interact with spending patterns of CTC. A number of studies found that low-income families became more vulnerable to food insecurity and economic problems during the pandemic [13,33,34,54]. Further studies may focus on lower-income families who are likely to continue and even expand spending CTC on essential items (e.g., food, rent, utility bills, children’s education) than on savings and investment as the pandemic enters its fourth year.
## 4.3. Policy Implications
We can provide three policy implications to help improve mental health among CTC recipients by considering the important mediating role of their spending patterns during the COVID-19 pandemic. First, the government should consider whether the expanded CTC should be made permanent to promote households’ health and reduce health differences between income groups [22]. Considering the limitation of the government budget, subsidies may result in financial burden in the long run. An unexpected discontinuity of CTC—particularly its expanded benefits—may disrupt the spending power of eligible households with children because they are likely to plan their spending on the basis of CTC benefits. In the midst of working toward a permanent expansion of CTC, community partners, lawmakers, and federal officials need to secure the continuity of CTC with regard to spending patterns [32].
Second, policy makers should design public financial assistance to be spent on buying healthy food and well-being products to improve recipients’ mental health. Previous studies showed that stress made consumers buy things on impulse [63,64,65]. In response to a historical crisis, some people even tend to buy hedonic and harmful products within a few weeks, as well as spending money on temptation goods [28]. These spending behaviors may not be sustainable and desirable in the long term, despite their temporary effect on stress. Especially in a pandemic, careful choice of healthy foods is important for the population’s health [66]. During the COVID-19 pandemic, studies found that some consumers preferred purchasing healthy foods to buying other non-healthy goods [67]. Thus, health policy makers should consider ways to promote the purchase of health-friendly goods, beyond simply increasing the number and size of public benefits.
Third, policy providers may suggest appropriate timing and duration of the assistance provided to population subgroups. We found that spending patterns (e.g., foods, children education, saving) had different effects on improving recipients’ mental health. Spending on foods and child education affected mental health positively, while spending on saving and debt had a negative effect on mental health. This was probably caused by differences in household income and other socioeconomic characteristics. Recent studies found disproportionate difficulties of daily spending among socioeconomically disadvantaged households [68,69].
## 5. Conclusions
We conducted a population-based study on the relationship between the receipt of CTC and mental health outcomes during the COVID-19 pandemic, with a special focus on the mediating roles of spending patterns of credit in the U.S. Key results support the hypothesis that the receipt of CTC is related to lower levels of anxiety and depression, and these relationships were partially mediated by spending patterns of the credit during the pandemic. Public health approaches to improve population mental health during and after the pandemic need to consider the important mediating roles of spending patterns of CTC.
## References
1. Hart J., Han W.-J.. **COVID-19 Experiences and Parental Mental Health**. *J. Soc. Soc. Work Res.* (2021.0) **12** 283-302. DOI: 10.1086/711613
2. Nelson B., Pettitt A., Flannery J., Allen N.. **Psychological and Epidemiological Predictors of COVID-19 Concern and Health-Related Behaviors**. *Natl. Libr. Med.* (2020.0) **15** 11
3. Rajkumar R.P.. **COVID-19 and Mental Health: A Review of the Existing Literature**. *Asian J. Psychiatry* (2020.0) **52** 102066. DOI: 10.1016/j.ajp.2020.102066
4. Twenge J.M., Joiner T.E.. **Mental Distress among U.S. Adults during the COVID-19 Pandemic**. *J. Clin. Psychol.* (2020.0) **76** 2170-2182. DOI: 10.1002/jclp.23064
5. Cheng Z., Mendolia S., Paloyo A.R., Savage D.A., Tani M.. **Working Parents, Financial Insecurity, and Childcare: Mental Health in the Time of COVID-19 in the UK**. *Rev. Econ. Househ.* (2021.0) **19** 123-144. DOI: 10.1007/s11150-020-09538-3
6. 6.
OECD
Tackling the Mental Health Impact of the COVID-19 Crisis: An Integrated, Whole-of-Society ResponseOECDParis, France2021. *Tackling the Mental Health Impact of the COVID-19 Crisis: An Integrated, Whole-of-Society Response* (2021.0)
7. **Mental Health and COVID-19: Early Evidence of the Pandemic’s Impact: Scientific Brief. 2022**
8. Chatterji S., McDougal L., Johns N., Ghule M., Rao N., Raj A.. **COVID-19-Related Financial Hardship, Job Loss, and Mental Health Symptoms: Findings from a Cross-Sectional Study in a Rural Agrarian Community in India**. *Int. J. Environ. Res. Public Health* (2021.0) **18**. DOI: 10.3390/ijerph18168647
9. Ikeda T., Igarashi A., Odani S., Murakami M., Tabuchi T.. **Health-Related Quality of Life during COVID-19 Pandemic: Assessing Impacts of Job Loss and Financial Support Programs in Japan**. *Appl. Res. Qual. Life* (2022.0) **17** 541-557. DOI: 10.1007/s11482-021-09918-6
10. Kiely K.M., Leach L.S., Olesen S.C., Butterworth P.. **How Financial Hardship Is Associated with the Onset of Mental Health Problems over Time**. *Soc. Psychiatry Psychiatr. Epidemiol.* (2015.0) **50** 909-918. DOI: 10.1007/s00127-015-1027-0
11. Bauer A., Garman E., McDaid D., Avendano M., Hessel P., Díaz Y., Araya R., Lund C., Malvasi P., Matijasevich A.. **Integrating Youth Mental Health into Cash Transfer Programmes in Response to the COVID-19 Crisis in Low-Income and Middle-Income Countries**. *Lancet Psychiatry* (2021.0) **8** 340-346. DOI: 10.1016/S2215-0366(20)30382-5
12. Ohrnberger J., Anselmi L., Fichera E., Sutton M.. **The Effect of Cash Transfers on Mental Health: Opening the Black Box—A Study from South Africa**. *Soc. Sci. Med.* (2020.0) **260** 113181. DOI: 10.1016/j.socscimed.2020.113181
13. Shafer P.R., Gutiérrez K.M., Ettinger de Cuba S., Bovell-Ammon A., Raifman J.. **Association of the Implementation of Child Tax Credit Advance Payments with Food Insufficiency in US Households**. *JAMA Netw. Open* (2022.0) **5** e2143296. DOI: 10.1001/jamanetworkopen.2021.43296
14. Crandall-Hollick M.L., Carter J.A., Boyle C.F.. *The Child Tax Credit: The Impact of the American Rescue Plan Act (ARPA P.L. 117-2) Expansion on Income and Poverty* (2021.0)
15. Ananat E., Glasner B., Hamilton C., Parolin Z.. *Effects of the Expanded Child Tax Credit on Employment Outcomes: Evidence from Real-World Data from April to December 2021* (2022.0)
16. Baker M., Messacar D., Stabile M.. *The Effects of Child Tax Benefits on Poverty and Labor Supply: Evidence from the Canada Child Benefit and Universal Child Care Benefit* (2021.0)
17. Chu L., Teng L.. **Does Stimulus Check Payment Improve People’s Mental Health in the COVID-19 Pandemic? Evidence from U.S. Household Pulse Survey**. *J. Ment. Health Policy Econ.* (2022.0) **25** 133-142. PMID: 36535911
18. Tsai J., Huang M., Rajan S.S., Elbogen E.B.. **Prospective Association between Receipt of the Economic Impact Payment and Mental Health Outcomes. J. Epidemiol**. *Commun. Health* (2022.0) **76** 285-292. DOI: 10.1136/jech-2021-216661
19. Berkowitz S.A., Basu S.. **Unemployment Insurance, Health-Related Social Needs, Health Care Access, and Mental Health During the COVID-19 Pandemic**. *JAMA Intern. Med.* (2021.0) **181** 699-702. DOI: 10.1001/jamainternmed.2020.7048
20. Wolfson J.A., Garcia T., Leung C.W.. **Food Insecurity Is Associated with Depression, Anxiety, and Stress: Evidence from the Early Days of the COVID-19 Pandemic in the United States**. *Health Equity* (2021.0) **5** 64-71. DOI: 10.1089/heq.2020.0059
21. Smith M.V., Callinan L.S., Posner C.S., Holmes S.C., Ebling R.. **Improving Maternal Mental Health as a Pathway to Economic Mobility in the TANF System**. *Psychiatr. Serv.* (2021.0) **72** 1139-1144. DOI: 10.1176/appi.ps.202000492
22. Dore E.C., Livingston M.D., Shafer P.R.. **Easing Cash Assistance Rules During COVID-19 Was Associated with Reduced Days of Poor Physical and Mental Health: Study Examines Impact of Easing Cash Assistance Rules during COVID-19 Had Participants’ Physical and Mental Health**. *Health Aff.* (2022.0) **41** 1590-1597. DOI: 10.1377/hlthaff.2022.00740
23. Sears R.. **Cash Payments to Families Lead to Long Term Improvements in Child Mental Health. Mad in America, 11 October 2022**
24. Shevlin M., Redican E., Hyland P., Butter S., McBride O., Hartman T.K., Murphy J., Vallières F., Bentall R.P.. **Perceived Manageability of Debt and Mental Health during the COVID-19 Pandemic: A UK Population Analysis**. *PLoS ONE* (2022.0) **17**. DOI: 10.1371/journal.pone.0274052
25. Maslow A.H.. **A Theory of Human Motivation**. *Psychol. Rev.* (1943.0) **50** 370-396. DOI: 10.1037/h0054346
26. Loxton M., Truskett R., Scarf B., Sindone L., Baldry G., Zhao Y.. **Consumer Behaviour during Crises: Preliminary Research on How Coronavirus Has Manifested Consumer Panic Buying, Herd Mentality, Changing Discretionary Spending and the Role of the Media in Influencing Behaviour**. *J. Risk Financ. Manag.* (2020.0) **13**. DOI: 10.3390/jrfm13080166
27. Afify A.S.. **Effect of COVID-19 Pandemic on Consumption Patterns among Egyptian Consumers and Its Relation to Maslow’s Hierarchy of Needs**. *Mod. Econ.* (2022.0) **13** 186-207. DOI: 10.4236/me.2022.132012
28. Forbes S.L.. **Post-Disaster Consumption: Analysis from the 2011 Christchurch Earthquake**. *Int. Rev. Retail. Distrib. Consum. Res.* (2017.0) **27** 28-42. DOI: 10.1080/09593969.2016.1247010
29. Luhby T.. **The Child Tax Credit May Be Helping Ease Economic Hardship after Just One Round. CNN Politics, 12 August 2021**
30. Hamilton L., Roll S., Despard M., Maag E., Chun Y., Brugger L., Grinstein-Weiss M.. *The Impacts of the 2021 Expanded Child Tax Credit on Family Employment, Nutrition, and Financial Well-Being: Findings from the Social Policy Institute’s Child Tax Credit Panel (Wave 2)* (2022.0) 173
31. Rummo P.E., Bragg M.A., Yi S.S.. **Supporting Equitable Food Access During National Emergencies—The Promise of Online Grocery Shopping and Food Delivery Services**. *JAMA Health Forum* (2020.0) **1** e200365. DOI: 10.1001/jamahealthforum.2020.0365
32. Kemp M.. *The Expanded Child Tax Credit Is Helping Families, but National Survey Shows Continued Outreach Remains Essential* (2022.0)
33. Javaid S., Tucker J.. **Women of Color Use Their Advance Child Tax Credit to Cover Food Costs. National Women’s Law Center, 7 September 2021**
34. Zippel C.. *After Child Tax Credit Payments Begin, Many More Families Have Enough to Eat* (2021.0)
35. Perez-Lopez D.J.. **Household Pulse Survey Collected Responses Just Before and Just After the Arrival of the First CTC Checks. Census Bureau’s Social, Economic and Housing Statistics Division, 11 August 2021**
36. Samuelson K.. **Food Stamp Work Requirements Increase Mental Health Care Use. Northwestern News, 28 July 2022**
37. Polsky J.Y., Gilmour H.. **Food Insecurity and Mental Health during the COVID-19 Pandemic**. *Health Rep.* (2020.0) **31** 3-11. DOI: 10.25318/82-003-x202001200001-eng
38. Dawes J., May T., McKinlay A., Fancourt D., Burton A.. **Impact of the COVID-19 Pandemic on the Mental Health and Wellbeing of Parents with Young Children: A Qualitative Interview Study**. *BMC Psychol.* (2021.0) **9**. DOI: 10.1186/s40359-021-00701-8
39. Costello E.J., Compton S.N., Keeler G., Angold A.. **Relationships between Poverty and Psychopathology: A Natural Experiment**. *JAMA* (2003.0) **290** 2023-2029. DOI: 10.1001/jama.290.15.2023
40. Costello E.J., Erkanli A., Copeland W., Angold A.. **Association of Family Income Supplements in Adolescence with Development of Psychiatric and Substance Use Disorders in Adulthood among an American Indian Population**. *JAMA* (2010.0) **303** 1954-1960. DOI: 10.1001/jama.2010.621
41. **2020 Mom. The Child Tax Credit: Tackling Child Poverty and Maternal Mental Health. 2020 Mom, 23 November 2021**
42. Huang C., Feng Q., Zhang B., Ren H., Liu Z., Kang Y., Zhang F.. **Income and Social Support Related with Mental Health during COVID-19 Outbreak in China. Medicine**. *Baltimore* (2022.0) **101** e29022. DOI: 10.1097/MD.0000000000029022
43. Simonse O., Van Dijk W.W., Van Dillen L.F., Van Dijk E.. **The Role of Financial Stress in Mental Health Changes during COVID-19**. *NPJ Ment. Health Res.* (2022.0) **1** 15. DOI: 10.1038/s44184-022-00016-5
44. Benfer E.A., Wiley L.F.. **Health Justice Strategies to Combat COVID-19: Protecting Vulnerable Communities During a Pan-demic**. *Health Aff. Forefr.* (2020.0). DOI: 10.1377/forefront.20200319.757883
45. Furlong M., McGilloway S., Mulligan C., Killion M.G., McGarr S., Grant A., Davidson G., Donaghy M.. **Covid-19 and Families with Parental Mental Illness: Crisis and Opportunity**. *Front. Psychiatry* (2021.0) **12** 567447. DOI: 10.3389/fpsyt.2021.567447
46. Gadermann A.C., Thomson K.C., Richardson C.G., Gagné M., McAuliffe C., Hirani S., Jenkins E.. **Examining the Impacts of the COVID-19 Pandemic on Family Mental Health in Canada: Findings from a National Cross-Sectional Study**. *BMJ Open* (2021.0) **11** e042871. DOI: 10.1136/bmjopen-2020-042871
47. Kerr M.L., Fanning K.A., Huynh T., Botto I., Kim C.N.. **Parents’ Self-Reported Psychological Impacts of COVID-19: Associations with Parental Burnout, Child Behavior, and Income**. *J. Pediatr. Psychol.* (2021.0) **46** 1162-1171. DOI: 10.1093/jpepsy/jsab089
48. Westrupp E.M., Bennett C., Berkowitz T., Youssef G.J., Toumbourou J.W., Tucker R., Andrews F.J., Evans S., Teague S.J., Karantzas G.C.. **Child, Parent, and Family Mental Health and Functioning in Australia during COVID-19: Comparison to Pre-Pandemic Data**. *Eur. Child Adolesc. Psychiatry* (2021.0) **32** 317-330. DOI: 10.1007/s00787-021-01861-z
49. Marroquín B., Vine V., Morgan R.. **Mental Health during the COVID-19 Pandemic: Effects of Stay-at-Home Policies, Social Distancing Behavior, and Social Resources**. *Psychiatry Res.* (2020.0) **293** 113419. DOI: 10.1016/j.psychres.2020.113419
50. **Mental Health and Psychosocial Considerations during the COVID-19 Outbreak. 18 March 2020**
51. Wu M., Xu W., Yao Y., Zhang L., Guo L., Fan J., Chen J.. **Mental Health Status of Students’ Parents during COVID-19 Pandemic and Its Influence Factors**. *Gen. Psych.* (2020.0) **33** e100250. DOI: 10.1136/gpsych-2020-100250
52. Plummer F., Manea L., Trepel D., McMillan D.. **Screening for Anxiety Disorders with the GAD-7 and GAD-2: A Systematic Review and Diagnostic Metaanalysis**. *Gen. Hosp. Psychiatry* (2016.0) **39** 24-31. DOI: 10.1016/j.genhosppsych.2015.11.005
53. Arroll B., Goodyear-Smith F., Crengle S., Gunn J., Kerse N., Fishman T., Falloon K., Hatcher S.. **Validation of PHQ-2 and PHQ-9 to Screen for Major Depression in the Primary Care Population**. *Ann. Fam. Med.* (2010.0) **8** 348-353. DOI: 10.1370/afm.1139
54. Banks J., Xu X.. **The Mental Health Effects of the First Two Months of Lockdown during the COVID-19 Pandemic in the UK**. *Fisc. Stud.* (2020.0) **41** 685-708. DOI: 10.1111/1475-5890.12239
55. Emsley R., Liu H.. **Stata Module to Perform Causal Mediation Analysis Using Parametric Regression Models. 2013**
56. Valeri L., Vanderweele T.J.. **Mediation Analysis Allowing for Exposure-Mediator Interactions and Causal Interpretation: Theoretical Assumptions and Implementation with SAS and SPSS Macros**. *Psychol. Methods* (2013.0) **18** 137-150. DOI: 10.1037/a0031034
57. Baron R.M., Kenny D.A.. **The Moderator–Mediator Variable Distinction in Social Psychological Research: Conceptual, Strategic, and Statistical Considerations**. *J. Personal. Soc. Psychol.* (1986.0) **51** 1173-1182. DOI: 10.1037/0022-3514.51.6.1173
58. Blount D., Minoff E.. **The Child Tax Credit & Family Economic Security: Findings from the Center for the Study of Social Policy’s Survey of Families with Children. 2022**
59. Collin D.F., Shields-Zeeman L.S., Batra A., White J.S., Tong M., Hamad R.. **The Effects of State Earned Income Tax Credits on Mental Health and Health Behaviors: A Quasi-Experimental Study**. *Soc. Sci. Med.* (2021.0) **276** 113274. DOI: 10.1016/j.socscimed.2020.113274
60. Bovell-Ammon A., McCann N., Mulugeta M., Ettinger de Cuba S., Raifman J., Shafer P.. **Food Insufficiency in Families with Children Increased after Expiration of Child Tax Credit Monthly Payments. 2022**
61. Grinstein-Weiss J.J., Hamilton L., Roll S., Grinstein-Weiss M.. *The New Child Tax Credit Does More than Just Cut Poverty* (2021.0)
62. Rachidi A.. **Why a Fully Refundable CTC Is the Wrong Policy for Working Families. American Enterprise Institute—AEI. 2021**
63. Dasso R., Fernandez F.. *Temptation Goods and Conditional Cash Transfers in Peru* (2014.0)
64. Burroughs J.E., Rindfleisch A.. **Materialism and Well-Being: A Conflicting Values Perspective**. *J. Consum. Res.* (2002.0) **29** 348-370. DOI: 10.1086/344429
65. Duhachek A.. **Coping: A Multidimensional, Hierarchical Framework of Responses to Stressful Consumption Episodes**. *J. Consum. Res.* (2005.0) **32** 41-53. DOI: 10.1086/426612
66. Halvatsiotis P., Kotanidou A., Tzannis K., Jahaj E., Magira E., Theodorakopoulou M., Konstandopoulou G., Gkeka E., Pourzitaki C., Kapravelos N.. **Demographic and Clinical Features of Critically Ill Patients with COVID-19 in Greece: The Burden of Diabetes and Obesity**. *Diabetes Res. Clin. Pract.* (2020.0) **166** 108331. DOI: 10.1016/j.diabres.2020.108331
67. Hesham F., Riadh H., Sihem N.K.. **What Have We Learned about the Effects of the COVID-19 Pandemic on Consumer Behavior?**. *Sustainability* (2021.0) **13**. DOI: 10.3390/su13084304
68. Bell T.. *The Covid Certainty: More Savings for the Rich, More Debt for the Poor* (2021.0)
69. Crosta A.D., Ceccato I., Marchetti D., Malva P.L., Maiella R., Cannito L., Cipi M., Mammarella N., Palumbo R., Verrocchio M.C.. **Psychological Factors and Consumer Behavior during the COVID-19 Pandemic**. *PLoS ONE* (2021.0) **16**. DOI: 10.1371/journal.pone.0256095
|
---
title: Association between Smoking and Periodontal Disease in South Korean Adults
authors:
- Ka-Yun Sim
- Yun Seo Jang
- Ye Seul Jang
- Nataliya Nerobkova
- Eun-Cheol Park
journal: International Journal of Environmental Research and Public Health
year: 2023
pmcid: PMC10002277
doi: 10.3390/ijerph20054423
license: CC BY 4.0
---
# Association between Smoking and Periodontal Disease in South Korean Adults
## Abstract
Smoking poses a threat to global public health. This study analyzed data from the 2016–2018 National Health and Nutrition Examination Survey to investigate smoking’s impact on periodontal health and identify potential risk factors associated with poor periodontal health in Korean adults. The final study population was 9178 patients, with 4161 men and 5017 women. The dependent variable was the Community Periodontal Index (CPI), to investigate periodontal disease risks. Smoking was the independent variable and was divided into three groups. The chi-squared test and multivariable logistic regression analyses were used in this study. Current smokers had a higher risk of periodontal disease than non-smokers (males OR: 1.78, $95\%$ CIs = 1.43–2.23, females OR: 1.44, $95\%$ CIs = 1.04–1.99). Age, educational level, and dental checkups affected periodontal disease. Men with a higher number of pack years had a higher risk of periodontal disease than non-smokers (OR: 1.84, $95\%$ CIs = 1.38–2.47). Men who quit smoking for less than five years had a higher risk of periodontal disease than non-smokers but lower than current smokers (current OR: 1.78, $95\%$ CIs = 1.43–2.23, ex OR: 1.42, $95\%$ CIs = 1.04–1.96). Those who had quit smoking for less than five years had a higher risk of periodontal disease than non-smokers but lower than current smokers (males OR: 1.42, $95\%$ CIs = 1.04–1.96, females OR: 1.11, $95\%$ CIs = 1.71–1.74). It is necessary to motivate smokers by educating them on the importance of early smoking cessation.
## 1. Introduction
Smoking is one of the biggest threats to public health [1]. According to the World Health Organization, more than 8 million people have been killed, including approximately 1.2 million deaths from exposure to secondhand smoke [2]. Moreover, since tobacco has more than 7000 toxic chemicals [3], smoking is associated with numerous preventable chronic diseases [4]. In Korea, the smoking rate has been decreasing; however, as of 2018, the prevalence of daily smoking among men in Korea reached $30.5\%$, the third-highest rate among the Organization for Economic Co-operation and Development (OECD) members [5]. The authorities have made intensive efforts to eliminate tobacco use by implementing strong and effective tobacco control policies and measures, such as cigarette tax hikes and media campaigns [4,6].
The association between smoking and various diseases, including major causes of death, has been well-established. A cohort study in the US reported that smokers had a higher risk of developing bladder cancer and pancreatic cancer than non-smokers [7]. Another study found that smokers were more likely to have elevated levels of blood insulin and triglycerides compared to non-smokers [8,9].
Smoking can negatively impact the oral cavity, particularly in non-inflammatory oral diseases [10]. Harmful substances in tobacco products, such as nicotine, can harm the gingival tissue, decrease blood flow to the gums, and compromise the immune system [11]. Tobacco use can increase susceptibility to oral infections, stain teeth, cause dryness in the mouth, and delay the healing of oral wounds [12].
Periodontal diseases are considered to be chronic destructive inflammatory diseases [13]. They are characterized by the destruction of the periodontal tissue, loss of adhesion to connective tissues, loss of alveolar bone, and the formation of pathological sacs around the teeth [14,15,16]. In addition, poor periodontal health is associated with systemic diseases, such as cancer, heart disease, and diabetes; therefore, management is important [17,18,19]. Previous studies have shown that smoking is associated with poor periodontal health, even among young adults [20]. Another study in Korea revealed that quitting smoking within a decade could potentially improve periodontal health for former smokers [21]. A study in the US, which used large-scale data, concluded that smoking is a significant risk factor for periodontitis and may account for more than $50\%$ of periodontitis in adults [22].
While previous studies have examined the association between smoking and periodontal diseases, additional evidence is needed to encourage healthy habits that promote smoking cessation. This study aimed to investigate the relationship between smoking and the risk of periodontal diseases in Korean adults, using a nationwide cross-sectional survey with a large sample size. Furthermore, this study aimed to provide more robust evidence for the importance of early smoking cessation by analyzing the relationship between smoking cessation in five-year intervals, which was more detailed than in previous studies.
## 2.1. Data
The data for this study were obtained from the 2016–2018 Korea National Health and Nutrition Examination Survey (KNHANES) and used a separate raw dataset (HNYN_OE). The KNHANES has been conducted by the Korea Disease Control and Prevention Agency (KDCA) since 1998 to investigate national statistics through a survey of the health level, health-related behavior, and nutritional status of 10,000 Koreans annually. The KDCA Research Ethics Review Board approved the data collection protocols for the KNHANES. The data are available for download from the KDCA website (https://knhanes.kdca.go.kr/knhanes/sub03/sub03_02_05.do, accessed on 1 January 2023). Thus, this study did not need extra approval from the ethics review board. The KNHANES is a self-reported survey using a stratified, two-stage, clustered sampling design conducted annually for South Koreans of all ages, divided into three age groups: (children: 1–11 years old, adolescents: 12–18 years old, and adults: 19 years or older).
## 2.2. Study Population
The total number of participants who completed the health examination survey for KNHANES 2016–2018 was 16,489 (7485 males and 9004 females). The exclusion criteria consisted of three categories: (a) under 19 years of age ($$n = 3299$$), (b) unable to perform oral examination due to tooth loss ($$n = 2581$$), and (c) missing values in health assessment or survey ($$n = 1440$$). The final study population was 9178, with 4161 men and 5017 women (Figure 1).
## 2.3. Variables
The dependent variable in this study was the Community Periodontal Index (CPI), used to measure the risk of periodontal disease. The oral health examinations were conducted by public health dentists and local public health dentists at the city and provincial levels under the supervision of the Korea Disease Control and Prevention Agency (KDCA). The risk to periodontal health was assessed by dividing the upper and lower jaws into three sections and recording the highest CPI score for each section. The CPI score was based on periodontal pocket depth, calculus attachment, and gingival bleeding measurements. The scores ranged from 0 to 4, with 0 indicating healthy, 1 indicating bleeding, 2 indicating dental calculus, 3 indicating a superficial periodontal pocket of 4–5 mm, and 4 indicating a deep periodontal pocket of 6 mm or more. Using the sum of the CPI scores, we assessed the risk of periodontal disease as the outcome variable.
The independent variable was the smoking status, classified into three groups: non-smokers, ex-smokers, and current smokers. Smoking status was based on the question, “Do you currently smoke cigarettes?”. We also used pack years and smoking cessation status as variables in the subgroup analysis. Pack years indicate the number of cigarettes a person has smoked in their lifetime, calculated by multiplying the total number of cigarettes smoked per day by the total number of years a person smoked.
The covariate variables were controlled for, as potential confounding factors. These included socioeconomic factors, such as sex, age, household income, and region, and factors related to health behaviors, such as current drinking status and physical activity. Oral health habits were also included as covariates. Teeth brushing frequency was investigated, based on the number of times teeth were brushed during the previous day, while dental checkup status was surveyed based on the question, “Did you have a dental checkup in the past 12 months?”.
## 2.4. Statistical Analysis
A chi-squared test was conducted to explore the general characteristics of the study population. *The* general characteristics of the final study population were represented as frequency and percentage. To assess the relationship between smoking and periodontal disease using the sum of the CPI scores in adults, we used multivariable logistic regression analysis with covariate adjustment. Subgroup analyses were performed to evaluate the relationship between pack years, smoking cessation status, and periodontal disease. All the results were presented as odds ratios (ORs) and $95\%$ confidence intervals (CIs). The analyses were performed using stratified sampling variables. All the estimates were estimated using weighted variables to generalize the data. SAS version 9.4 software (SAS Institute, Cary, NC, USA) was used for all the statistical analyses. Statistical significance was determined as a two-sided p-value of <0.05.
## 3. Results
Table 1 summarizes the characteristics of the study population, classified according to sex. Of the 9178 participants, 4161 were male ($45.3\%$), and 5017 were female ($54.7\%$). A total of 3042 ($73.1\%$) males and 3143 ($62.6\%$) females had periodontal disease risks, as expressed by the CPI. Among the males, 1484 ($35.7\%$) were current smokers, 1623 ($39.0\%$) were ex-smokers, and 1054 ($25.3\%$) were non-smokers. Among the females, 282 ($5.6\%$) were current smokers, 344 ($6.9\%$) were ex-smokers, and 4391 ($87.5\%$) were non-smokers.
Table 2 presents the multivariate logistic regression analysis results that explore the association between smoking and periodontal disease while adjusting for covariates. The smokers had a higher risk of periodontal disease than the non-smokers. While the ex-smokers were statistically insignificant, the current smokers were significant for males (OR: 1.78, $95\%$ CIs = 1.43–2.23) and females (OR: 1.44, $95\%$ CIs = 1.04–1.99). As age increased, the participants showed an elevated risk of periodontal disease. The participants with a middle school education or lower had a higher risk of periodontal disease than those with a college education (males OR: 1.63, $95\%$ CIs = 1.15–2.23, females OR: 1.59, $95\%$ CIs = 1.18–2.14). The individuals who did not receive dental checkups were likelier to have periodontal diseases (males OR: 1.62, $95\%$ CIs = 1.36–1.93, females OR: 1.60, $95\%$ CIs = 1.39–1.84).
Table 3 presents the results of the subgroup analysis for the independent variables stratified by smoking behavior. Most of the ex-smokers did not have significant results. The observed results were more significant in the males than in the females. The risk of periodontal disease generally increased with age in men who are current smokers but was not statistically significant in their 50s. The current smokers had a higher risk of periodontal disease in all education levels, and the risk was highest for those with middle school education or lower (OR: 3.15; $95\%$ CIs = 1.37–7.21). The current smokers had a risk of periodontal disease, regardless of their physical activity status (male, adequate: OR = 1.83, $95\%$ CIs = 1.35–2.50; inadequate: OR = 1.77, $95\%$ CIs = 1.30–2.42). Similarly, regardless of whether they received regular dental checkups, current smokers had a higher risk of periodontal disease (male, checkups: OR = 1.90, $95\%$ CIs = 1.40–2.59; no checkups: OR = 1.73, $95\%$ CIs = 1.29–2.33).
The results of the subgroup analysis, which were stratified by pack years and smoking cessation, are presented in Table 4. The males showed a statistically significant positive association. Those with a higher number of pack years had a higher risk of periodontal diseases than the non-smokers (over 20 pack years OR: 1.84, $95\%$ CIs = 1.38–2.47). Those who had quit smoking for less than five years had a higher risk of periodontal disease than the non-smokers but lower than the current smokers (males OR: 1.42, $95\%$ CIs = 1.04–1.96; females OR: 1.11, $95\%$ CIs = 1.71–1.74).
## 4. Discussion
Despite the reduction in smoking prevalence over the past 30 years, the total number of smokers has increased from 0.99 billion in 1990 to 1.14 billion in 2019 worldwide, due to population growth [23]. The American Academy of Periodontology has pointed out that smoking negatively impacts the healing and treatment of periodontitis [24]. The purpose of the study was two main issues. First, we used a nationwide survey with a large sample size to investigate the association between smoking and periodontal disease. Second, we attempted to support the importance of early smoking cessation by analyzing the relationship between smoking cessation in five-year intervals compared to the previous studies using ten-year intervals.
The mechanisms underlying the association between smoking and periodontal disease were the following. Smoking stimulates the establishment of pathogenic microflora, diminishes the immune host response, and elevates the release of inflammatory mediators [14,15,16,25]. As smokers are more likely to absorb pathogenic microorganisms than non-smokers, previous studies have reported an increase in particular pathogens in smokers, such as Actinobacillus actinomycetemcomitans and Bacteroides forsythus, although the pathogen levels may have varied, based on the methods used in the studies [14,26,27]. Smoking can affect host inflammatory and immune responses, such as the immunosuppressive effects of macrophages on cell-mediated immune responses, inhibition of human periodontal ligament fibroblast migration, and repression of alkaline phosphatase production by nicotine [28,29]. As with this mechanism, Table 2 shows current smokers had a higher risk of periodontal diseases than non-smokers. It supports previous studies’ results that smoking is a risk factor for oral health, even among young smokers [20]. Additionally, the results in Table 2 are in the same vein as previous studies, showing that smoking significantly influences periodontitis, using a large sample in the US [22].
Notable points in Table 2 were the results of age, education level, and dental checkup status variables. In the case of age, it was consistent with the results of previous studies that the prevalence of periodontal disease tends to increase as the age of participants increases. Previous studies in Brazil and India have reported that age increases affect the severity and prevalence of periodontal disease, regardless of gender [30,31]. The education level affected periodontal health. Middle school or lower education participants had a higher risk of periodontal disease than those with a college or higher education. As some studies have reported similar findings [22,32], education progressively decreases the risk of periodontal diseases. This finding implies that education regarding periodontal health is important. There was a higher risk of periodontal disease in people who did not undergo oral examinations, which supports previous studies that those who regularly underwent oral examinations had a lower risk of periodontal disease than those who did not [33,34].
In Table 3, male smokers in their 50s were not statistically significant. This counterintuitive finding can be explained by aging, which affects tooth loss [4,35]. Even if good physical activity habits and dental checkups were regularly undertaken, the current smokers had a higher risk of periodontal disease than the non-smokers. This result could explain that current smokers cannot avoid the risk of periodontal disease, even if they have good health habits.
As shown in Table 4, the men with high pack years had a higher risk of periodontal disease. However, the women showed statistically insignificant results. The results can be explained in the WHO Framework Convention on Tobacco Control (WHO FCTC) context. The WHO FCTC emphasized the need to consider gender when developing tobacco control strategies, as perceptions of smoking habits related to gender continue to differ depending on social contexts and cultural norms. Specifically, in Confucian Asian countries, there is still a tendency for views on female smoking to be more conservative than those on male smoking [36]. The smoking rate of women is also increasing in Korea. However, considering the social context, the data on the female smoking rate collected by voluntary reporting may not be accurate, due to the opposing views of some female smokers.
We found that ex-smokers with relatively short smoking cessation periods had a lower risk of periodontal disease than current smokers. This result can be compared to a previous study that reported the possibility of reversing the risk of periodontal disease if an individual quits smoking for ten years [15]. The sentence that smoking is harmful was too clear and simple, but the results of this study tried to support the importance of early smoking cessation. It could motivate smokers to quit smoking by revealing that people who quit smoking for a relatively short period, fewer than five years, have a lower risk of periodontal disease.
This study had several limitations. First, clarifying an inverse causal association was difficult since it was a cross-sectional study. Second, the KNHANES data were collected through a self-reported survey. The data on smoking behavior, health habits, and socioeconomic variables may not be accurately estimated. There was a possibility of recall bias. Third, it was impossible to identify the type of smoking, such as whether the participants used conventional cigarettes, e-cigarettes, or both. In addition, we could not use biological indicators, such as urine cotinine, in the subjects. Therefore, further studies are needed, considering these limitations.
Despite these limitations, our study has several strengths. The main strength of this study is the use of nationally representative, large, and high-quality data. KNHANES was conducted using a random cluster design, which can generalize the study’s results to the general population. Second, oral health examination datasets collected by public health doctors may effectively estimate periodontal disease risks. It was possible to estimate the risk of periodontal disease more precisely by using the CPI score through a doctor’s examination than by using the participant’s subjective oral symptom self-reported survey. Third, our study supported the importance of early smoking cessation. Compared to previous studies, we provided more proactive evidence for the importance of early smoking cessation.
## 5. Conclusions
This study demonstrated a strong association between smoking and periodontal disease in South Korean adults. Long-term smoking was closely related to poor periodontal health. The findings that even a relatively short period of smoking cessation, less than five years, had a positive impact on periodontal disease could be a powerful motivator for smokers. There is a need for effective tobacco control measures to reduce the prevalence of periodontitis. Tailored smoking cessation policies and educational interventions, highlighting the benefits of short-term smoking abstinence in reducing periodontal disease risk, could encourage current smokers to quit and ultimately improve public oral health.
## References
1. **Fact Sheets of Tobacco**
2. **IHME Global Burden of Disease Results-GBD Estimate of Oral Health**
3. Varghese C., Troisi G., Schotte K., Prasad V.M., Claire S.M.S.. **World No Tobacco Day 2019 puts the spotlight on lung health**. *J. Thorac. Dis.* (2019.0) **11** 2639-2642. DOI: 10.21037/jtd.2019.06.38
4. Jeong W., Choi D.W., Kim Y.K., Lee H.J., Lee S.A., Park E.C., Jang S.I.. **Associations of electronic and conventional cigarette use with periodontal disease in South Korean adults**. *J. Periodontol.* (2020.0) **91** 55-64. DOI: 10.1002/JPER.19-0060
5. **Organisation for Economic Co-operation and Development-Health Statistics 2021**
6. Lee S., Kimm H., Yun J.E., Jee S.H.. **Public Health Challenges of Electronic Cigarettes in South Korea**. *J. Prev. Med. Public Health* (2011.0) **44** 235-241. DOI: 10.3961/jpmph.2011.44.6.235
7. Inoue-Choi M., Hartge P., Liao L.M., Caporaso N., Freedman N.D.. **Association between long-term low-intensity cigarette smoking and incidence of smoking-related cancer in the national institutes of health-AARP cohort**. *Int. J. Cancer* (2017.0) **142** 271-280. DOI: 10.1002/ijc.31059
8. Huh Y., Han K., Choi M.-J., Kim J.H., Kim S.M., Nam G.E.. **Association of Smoking Status With the Risk of Type 2 Diabetes Among Young Adults: A Nationwide Cohort Study in South Korea**. *Nicotine Tob. Res.* (2022.0) **24** 1234-1240. DOI: 10.1093/ntr/ntac044
9. Will J.C., Galuska D.A., Ford E.S., Mokdad A., Calle E.E.. **Cigarette smoking and diabetes mellitus: Evidence of a positive association from a large prospective cohort study**. *Int. J. Epidemiol.* (2001.0) **30** 540-546. DOI: 10.1093/ije/30.3.540
10. Johnson G.K., Guthmiller J.M.. **The impact of cigarette smoking on periodontal disease and treatment**. *Periodontology 2000* (2007.0) **44** 178-194. DOI: 10.1111/j.1600-0757.2007.00212.x
11. Gajendra S., McIntosh S., Ghosh S.. **Effects of tobacco product use on oral health and the role of oral healthcare providers in cessation: A narrative review**. *Tob. Induc. Dis.* (2023.0) **21** 12. DOI: 10.18332/tid/157203
12. Agnihotri R., Gaur S.. **Implications of tobacco smoking on the oral health of older adults**. *Geriatr. Gerontol. Int.* (2014.0) **14** 526-540. DOI: 10.1111/ggi.12285
13. Hussein B.J., Khalil A.M., Atallah H.N., Ali N.A.. **Assessment of salivary interleukin-17 and vitamin d levels in smokers with severe periodontitis patients and healthy individuals**. *J. Pharm. Negat. Results* (2023.0) **2** 7147-7155
14. Kinane D., Chestnutt I.. **Smoking and periodontal disease**. *Crit. Rev. Oral Biol. Med.* (2000.0) **11** 356-365. DOI: 10.1177/10454411000110030501
15. Fiorini T., Musskopf M.L., Oppermann R.V., Susin C.. **Is There a Positive Effect of Smoking Cessation on Periodontal Health? A Systematic Review**. *J. Periodontol.* (2014.0) **85** 83-91. DOI: 10.1902/jop.2013.130047
16. Leite F.R., Nascimento G.G., Scheutz F., López R.. **Effect of Smoking on Periodontitis: A Systematic Review and Meta-regression**. *Am. J. Prev. Med.* (2018.0) **54** 831-841. DOI: 10.1016/j.amepre.2018.02.014
17. Han D.-H., Lim S., Kim J.-B.. **The Association of Smoking and Diabetes With Periodontitis in a Korean Population**. *J. Periodontol.* (2012.0) **83** 1397-1406. DOI: 10.1902/jop.2012.110686
18. Hong M.-H.. **TRelationship between Oral Health Behavior, Chronic Disease and Periodontal Disease in Middle and Older Adults According to Gender**. *J. Korea Acad.-Ind. Coop. Soc.* (2018.0) **19** 403-410
19. Petrenya N., Hopstock L.A., Holde G.E., Oscarson N., Jönsson B.. **Relationship between periodontitis and risk of cardiovascular disease: Insights from the Tromsø Study**. *J. Periodontol.* (2022.0) **93** 1353-1365. DOI: 10.1002/JPER.22-0004
20. Machuca G., Rosales I., LaCalle J.R., Machuca C., Bullon P.. **Effect of Cigarette Smoking on Periodontal Status of Healthy Young Adults**. *J. Periodontol.* (2000.0) **71** 73-78. DOI: 10.1902/jop.2000.71.1.73
21. Jang A.-Y., Lee J.-K., Shin J., Lee H.-Y.. **Association between Smoking and Periodontal Disease in Korean Adults: The Fifth Korea National Health and Nutrition Examination Survey (2010 and 2012)**. *Korean J. Fam. Med.* (2016.0) **37** 117-122. DOI: 10.4082/kjfm.2016.37.2.117
22. Tomar S.L., Asma S.. **Smoking-Attributable Periodontitis in the United States: Findings From NHANES III**. *J. Periodontol.* (2000.0) **71** 743-751. DOI: 10.1902/jop.2000.71.5.743
23. Reitsma M.B., Flor L.S., Mullany E.C., Gupta V., Hay S.I., Gakidou E.. **Spatial, temporal, and demographic patterns in prevalence of smoking tobacco use and initiation among young people in 204 countries and territories, 1990–2019**. *Lancet Public Health* (2021.0) **6** e472-e481. DOI: 10.1016/S2468-2667(21)00102-X
24. Research S.. **Position paper: Tobacco use and the periodontal patient**. *J. Periodontol.* (1999.0) **70** 1419-1427
25. Apatzidou D.A.. **The role of cigarette smoking in periodontal disease and treatment outcomes of dental implant therapy**. *Periodontology 2000* (2022.0) **90** 45-61. DOI: 10.1111/prd.12449
26. Haffajee A., Cugini M., Dibart S., Smith C., Kent Jr R., Socransky S.. **The effect of SRP on the clinical and microbiological parameters of periodontal diseases**. *J. Clin. Periodontol.* (1997.0) **24** 324-334. DOI: 10.1111/j.1600-051X.1997.tb00765.x
27. Zambon J., Grossi S., Machtei E., Ho A., Dunford R., Genco R.. **Cigarette Smoking Increases the Risk for Subgingival Infection With Periodontal Pathogens**. *J. Periodontol.* (1996.0) **67** 1050-1054. DOI: 10.1902/jop.1996.67.10s.1050
28. Giannopoulou C., Geinoz A., Cimasoni G.. **Effects of nicotine on periodontal ligament fibroblasts in vitro**. *J. Clin. Periodontol.* (1999.0) **26** 49-55. DOI: 10.1034/j.1600-051X.1999.260109.x
29. Arcavi L., Benowitz N.L.. **Cigarette smoking and infection**. *Arch. Intern. Med.* (2004.0) **164** 2206-2216. DOI: 10.1001/archinte.164.20.2206
30. Selvaraj S., Naing N.N., Wan-Arfah N., Djearamane S., Wong L.S., Subramaniyan V., Fuloria N.K., Sekar M., Fuloria S., de Abreu M.H.N.G.. **Epidemiological Factors of Periodontal Disease Among South Indian Adults**. *J. Multidiscip. Healthc.* (2022.0) **15** 1547-1557. DOI: 10.2147/JMDH.S374480
31. Figueiredo A., Soares S., Lopes H., dos Santos J.N., Ramalho L.M.P., Cangussu M.C., Cury P.R.. **Destructive periodontal disease in adult Indians from Northeast Brazil: Cross-sectional study of prevalence and risk indicators**. *J. Clin. Periodontol.* (2013.0) **40** 1001-1006. DOI: 10.1111/jcpe.12147
32. Zini A., Sgan-Cohen H.D., Marcenes W.. **Socio-economic position, smoking, and plaque: A pathway to severe chronic periodontitis**. *J. Clin. Periodontol.* (2010.0) **38** 229-235. DOI: 10.1111/j.1600-051X.2010.01689.x
33. Lang W.P., Farghaly M.M., Ronis D.L.. **The relation of preventive dental behaviors to periodontal health status**. *J. Clin. Periodontol.* (1994.0) **21** 194-198. DOI: 10.1111/j.1600-051X.1994.tb00303.x
34. Takami Y., Nakagaki H., Morita I., Tsuboi S., Takami S., Suzuki N., Niwa H., Ogura Y.. **Blood Test Values and Community Periodontal Index Scores in Medical Checkup Recipients**. *J. Periodontol.* (2003.0) **74** 1778-1784. DOI: 10.1902/jop.2003.74.12.1778
35. Ji M.G.. **Relationship between adults’ smoking realities and periodontal disease-2009 Korea National Health and Nutrition Examination Survey data**. *J. Korea Inst. Electron. Commun. Sci.* (2012.0) **7** 917-924
36. **WHO Framework Convention on Tobacco Control**
|
---
title: Effects of Long-Term Physical Activity and BCAA Availability on the Subcellular
Associations between Intramyocellular Lipids, Perilipins and PGC-1α
authors:
- Vasco Fachada
- Mika Silvennoinen
- Ulla-Maria Sahinaho
- Paavo Rahkila
- Riikka Kivelä
- Juha J. Hulmi
- Urho Kujala
- Heikki Kainulainen
journal: International Journal of Molecular Sciences
year: 2023
pmcid: PMC10002284
doi: 10.3390/ijms24054282
license: CC BY 4.0
---
# Effects of Long-Term Physical Activity and BCAA Availability on the Subcellular Associations between Intramyocellular Lipids, Perilipins and PGC-1α
## Abstract
Cellular skeletal muscle lipid metabolism is of paramount importance for metabolic health, specifically through its connection to branched-chain amino acids (BCAA) metabolism and through its modulation by exercise. In this study, we aimed at better understanding intramyocellular lipids (IMCL) and their related key proteins in response to physical activity and BCAA deprivation. By means of confocal microscopy, we examined IMCL and the lipid droplet coating proteins PLIN2 and PLIN5 in human twin pairs discordant for physical activity. Additionally, in order to study IMCLs, PLINs and their association to peroxisome proliferator-activated receptor gamma coactivator 1-alpha (PGC-1α) in cytosolic and nuclear pools, we mimicked exercise-induced contractions in C2C12 myotubes by electrical pulse stimulation (EPS), with or without BCAA deprivation. The life-long physically active twins displayed an increased IMCL signal in type I fibers when compared to their inactive twin pair. Moreover, the inactive twins showed a decreased association between PLIN2 and IMCL. Similarly, in the C2C12 cell line, PLIN2 dissociated from IMCL when myotubes were deprived of BCAA, especially when contracting. In addition, in myotubes, EPS led to an increase in nuclear PLIN5 signal and its associations with IMCL and PGC-1α. This study demonstrates how physical activity and BCAA availability affects IMCL and their associated proteins, providing further and novel evidence for the link between the BCAA, energy and lipid metabolisms.
## 1. Introduction
On top of being the largest organ in the human body, skeletal muscle has high energy demands, leading to elevated lipid turnover rates [1,2,3]. Despite being well established, the connection between skeletal muscle lipid metabolism and metabolic health is far from linear. On one hand, several metabolic diseases—such as insulin resistance—have been associated with physical inactivity and elevated intramyocellular lipids (IMCL). On the other hand, highly insulin sensitive individuals—such as endurance athletes—associate with even higher levels of IMCL [2,4]. It became ever more clear that mere levels of IMCL were not sufficient to explain skeletal muscle lipid metabolism efficiency. The perilipin protein family members (PLINs) are central agents in managing the fate of IMCL, and they are mostly known for regulating the access of lipolytic and lipogenic enzymes onto the surface of lipid droplets (LDs), thus protecting the cell against lipotoxicity and improving metabolic health [5,6,7,8,9,10].
The responses of human skeletal muscle LDs and PLINs to different exercise modalities have been well studied in a multitude of setups [11,12,13,14,15]. However, up to this date, the effects of long-term physical activity on IMCL and PLINs, amongst genetically similar individuals, remain largely unstudied. As a first aim of the present work, we propose to explore this gap. Therefore, through confocal microscopy and pixel-to-pixel intensity correlation analysis (ICA) [16], we examined IMCL, PLIN2 and PLIN5 and their associations in two main fiber types of twin pairs with discordant physical activity. We hypothesize that physically discordant twins will display different patterns of IMCL-PLIN association relative to previously studied setups.
The IMCL-PLINs dynamics are complex, and the role of PLINs themselves is not limited to hydrolysis or esterification of triacylglycerol (TAG) within LDs. For instance, PLIN5 has been shown to be necessary to transport monounsaturated fatty acids (MUFAs) into nuclei, in order to activate peroxisome proliferator-activated receptor gamma coactivator 1-alpha (PGC-1α) and, consequently, induce mitochondria biogenesis and fatty acid oxidation [17,18]. Additionally, PGC-1α is also an important bridge between the skeletal muscle lipid and branched-chain amino acid (BCAA) metabolisms, as it activates the latter through multiple nuclear receptors [19,20]. Importantly, unlike other amino acids that can be processed by the liver, BCAA are mostly catabolized in skeletal muscle [21,22]. Respectively, we have previously hypothesized that the connection between IMCL and metabolic health could further extend to the BCAA metabolism [22], as one important source of TAG in muscle comes from glyceroneogenesis [23]. Furthermore, inefficient skeletal muscle BCAA metabolism has been associated with impaired lipid metabolism and insulin resistance [21]. Finally, there is a known interplay between exercise and muscle BCAA metabolism [24,25]. However, studies establishing the relationship between intramyocellular BCAA and PLINs are essentially lacking. Therefore, it is important to investigate the impacts that BCAA availability may have on LD-PLINs regulation.
As a second aim, we investigated IMCL, PLIN2, PLIN5 and PGC-1α subcellular responses to exercise and BCAA availability. By combining electrical pulse stimulation (EPS)—an exercise-mimicking method [26]—with BCAA deprivation, we measured the optical density and performed ICA in different myotube compartments. We hypothesize that skeletal muscle PLIN5 and PGC-1α signals associate in response to EPS, especially within the nuclei. Moreover, we postulate that, besides LDs, PLIN5 or PGC-1α, PLIN2 may have unreported nuclear associations, which could be affected by EPS and/or BCAA deprivation.
## 2.1. Active Co-Twins Have Increased IMCL in Type I Fibers
Concerning the twin participants described in Table 1, type I fibers contained more IMCL than type II fibers, as expected ($p \leq 0.001$, Figure 1A–C). Interestingly, physically active twins had increased IMCL in type I fibers compared to their inactive co-twin ($p \leq 0.001$, Figure 1B,C), but there was no difference in type II fibers due to LTPA. The active twins demonstrated a significant difference in IMCL between fiber types ($p \leq 0.001$), which was not observed in the inactive co-twins ($$p \leq 0.064$$), as seen in Figure 1C.
As expected, PLIN5 associated IMCL was significantly higher in type I than in type II fibers ($$p \leq 0.001$$, Figure 2A,B), and with no differences between twin pairs (Figure 2B,C). Lastly, both PLIN5 mRNA levels and PLIN5 confocal mean signal remained unchanged between twin pairs (Figure S7).
## 2.2. Inactive Twins Show Decreased IMCL-PLIN2 Association
The inactive twin pairs show a decreased association between IMCL and PLIN2 ($$p \leq 0.008$$), mainly through a very significant decrease in the type II fibers ($p \leq 0.001$, Figure 3A–C). The latter happened despite no fiber type or LTPA differences in PLIN2 mean signal (Figure S1) or PLIN2 mRNA levels (Figure S7). Taken together, this shows that IMCL targeting by PLIN2 is clearly restrained in type II fibers of inactive twins.
## 2.3. PLIN5 Abounds in C2C12 Myotube Nuclei, PLIN2 Detected
Next, given the role in regulating energy metabolism and a known nuclear interaction with PLIN5, we studied PGC-1α and its association with IMCL and PLIN5 in different myotube compartments.
The compartmental analysis showed a significant contrast ($p \leq 0.001$) between cytosolic and nuclear signals for all markers within the myotubes. The most abundant nuclear signals were observed for IMCL and PLIN5. A much smaller proportion of PLIN2 ($p \leq 0.001$) was detected above the background and occasionally in a particle-like manner (Figure 4A,B).
## 2.4. PLIN2 Dissociates from IMCL upon BCAA Deprivation in Myotubes
Both IMCL and PLIN2 showed a mostly diffused signal in C2C12 myotubes. Occasionally, semi-spherical IMCL aggregates were visible as LDs (Video S1). Likewise, PLIN2 aggregates were common and often seen as dotted ring structures enveloping LDs (Figure 5A).
In addition to fiber type and exercise, BCAA can also affect IMCL metabolism, and this may interact with muscle contraction. Respectively, we observed a cytosolic decrease in the association between PLIN2 and IMCL after BCAA deprivation ($$p \leq 0.028$$, Figure S2), especially after EPS ($$p \leq 0.048$$, Figure 5B). The same combination (No BCAA|EPS) resulted in increased PLIN2 and PLIN5 association inside the nuclei ($$p \leq 0.030$$), and in a dependent manner ($$p \leq 0.033$$, Figure 5C).
Such events were independent from the overall PLIN2 signal, which remained unchanged after EPS and BCAA deprivation (Figure S2).
## 2.5. PLIN5 Moves to Myotube Nuclei upon Stimulation, Further Associating with IMCL and PGC-1α
The PLIN5 signal was mostly punctate and abundant, often immediately adjacent to or colocalizing with other markers. In turn, PGC-1α showed mostly a diffused signal, sometimes concentrating in differently shaped aggregates and often colocalizing with IMCL (Figure 6A).
Notably, PLIN5 signal increased in nuclei after EPS ($$p \leq 0.033$$, Figure 6B), where it associated with IMCL ($$p \leq 0.019$$, Figure 6D), especially under BCAA availability ($$p \leq 0.002$$, Figure S3). Likewise, under Normal BCAA|EPS, nuclear PLIN5 further associated with PGC-1α, and very significantly so ($$p \leq 0.009$$, Figure 6E).
Interestingly, BCAA deprivation alone was sufficient to decrease PGC-1α signal in myotube nuclei ($$p \leq 0.009$$, Figure 6C). It is worth noting that such effect was seen only in the nuclei, as neither a cytosolic confocal signal (Figure 6B) nor whole cell Western blots detected changes in PGC-1α protein concentration (Figure S8).
In the cytosol, the level of association between PLIN5 and PGC-1α was increased after BCAA deprivation ($$p \leq 0.026$$), as seen in (Figure 6E). The level of association between IMCL and PGC-1α was not altered with either EPS or BCAA deprivation (Figure S4).
## 3.1. Overview
This study examined the effects of physical activity on intramyocellular lipids and respective coating proteins in human twin pairs discordant for life-long physical activity. We found that, in physically active twins, the intramyocellular phenotype resembles that of athletes, namely in their type I fiber elevated lipid content, together with an enhanced lipid coating by PLIN2.
Secondly, we investigated myotube inter-compartmental responses to muscle contraction induced by EPS and to BCAA deprivation. We found that BCAA deprivation leads to a cytosolic dissociation between PLIN2 and IMCL, especially when combined with EPS. Importantly, we found that EPS leads to an increased presence of PLIN5 in nuclei, with increased association to PGC-1α, IMCL and PLIN2. Finally, we found that the signal of nuclear PGC-1α is abruptly decreased after BCAA deprivation.
## 3.2. Active Twins Resemble Athlete Phenotype
It has been shown that a healthy elevation of IMCL can be expected from not only athletes [2,4,12], but also from sedentary individuals who underwent a 6-week training period, especially in type I fibers [13]. Although the overall IMCL content was not different between twin pairs, we did observe significantly increased IMCL in type I fibers of active twins when compared to their inactive co-twins (Section 2.1). Our results suggest that the athlete paradox phenotype [2] may not be genetically determined and might be reached via life-long LTPA. Concomitantly, we have previously demonstrated that the active twins have improved skeletal muscle oxidative energy and lipid metabolism [24]. It should be noted that LTPA has recently been associated with slower epigenetic aging [27], suggesting that such mechanisms may be behind the results reported in our work.
Of the muscle PLINs, PLIN5 is probably the most studied member, and it is known for positively responding to exercise and high fat diet, both at the protein level and on IMCL association [8,9,12,28,29,30]. Interestingly, despite an obvious fiber type difference, LTPA led to no changes in PLIN5 signal or PLIN5 associated IMCL (Section 2.3). This may reflect the fact that intramyocellular physiological responses driven by LTPA could be distinct from those of more strenuous exercise programs in previous studies.
In addition, associating with efficient TAG storage and healthier profiles, intramyocellular PLIN2 has been shown to increase with exercise [13,31]. Although we did not register changes in PLIN2 signal, we did observe a significant decrease in IMCL-PLIN2 association in type II fibers of inactive twins (Section 2.2), suggesting an unhealthier phenotype. The hypothesis that lipotoxic signaling in skeletal muscle could originate from type II fibers is not new, as this fiber type is generally ill-equipped to metabolize lipids [32], especially in innermost regions of fibers ([33], Figure S5). Future studies should further explore the signaling impact of poorly PLIN coated-IMCL in glycolytic muscle fibers.
## 3.3. Myotubes Resembling Type II Fibers
Beyond PLINs [31], exercise and EPS are also expected to increase PGC-1α levels in skeletal muscle, including C2C12 myotubes [34,35,36,37]. Associated with mitochondrial biogenesis and fatty acid oxidation, as well as with glucose uptake and decreased glucose oxidation, PGC-1α is a rather lipolytic and glucogenic agent [38]. However, in the current study, EPS alone did not trigger significant cytosolic responses in the signal of PLIN2, PLIN5, PGC-1α or their association. Accordingly, from previous studies using the same protocol, we have observed a sharp glycolytic response in C2C12 myotubes [25] and only a modest lipolytic one [39]. More specifically, we had shown that EPS led to unchanged IMCL content, unaffected lipogenesis and decreased lipid oxidation [39].
One study has reported increased PGC-1α protein and unchanged lactate or pyruvate levels after using EPS [40]. Contrastingly, we have observed unchanged PGC-1α signal (this study) and increased lactate and pyruvate-derived products [25]. Interestingly, pyruvate is a known inhibitor of PGC-1α [41] and could be hindering a stronger lipolytic response in our setup. Conflicting results when studying lipid metabolism in C2C12 are not uncommon, as the generally glycolytic nature of this cell line can be increased with longer differentiation protocols [40,42]. Future research should focus on the same phenomena using different cell lines and culture parameters.
## 3.4. BCAA Necessary for PLIN2 Coating of IMCL
Our group has earlier demonstrated that the unhealthier profile of the inactive twins extends beyond an inefficient lipid metabolism, showing an associated downregulation of BCAA catabolism [24]. Furthermore, we have previously shown that BCAA deprivation decreases both lipid oxidation and lipogenesis in myotubes. In addition, when combined with EPS, BCAA deprivation also decreased the number of segmented LDs [39]. In the current work, the latter combination resulted in a dissociation between cytosolic IMCL and PLIN2 (Section 2.4), while the association between IMCL and PLIN5 remained unchanged (Section 2.5).
Often associating with TAG accumulation and protection against lipotoxicity derived-insulin resistance, PLIN2 is known to abound on the surface of LDs. There it can bind to both lipases and esterification enzymes, possibly having a more lipogenic role than PLIN5 [5,6,31,43]. Finally, there is evidence suggesting that BCAA facilitates TAG accumulation in muscle [44] and that EPS increases BCAA catabolism [25]. Together with our results, these data suggest that PLIN2 coating of IMCL is central to a healthy storage of lipid derivatives, which is improved by BCAA availability combined with physical activity, ultimately resulting in efficient BCAA and lipid metabolisms.
An efficient BCAA catabolism together with BCAA availability promotes ketonegenesis via several mitochondrial enzymes [45], some of which can further synthesize cholesterol for cellular needs [46]. It can be speculated that, in a scenario of impaired BCAA catabolism or BCAA deprivation, the source of ketone bodies and cholesterol could shift towards IMCL, once exposed and uncoated from PLIN2 (Figure 7A,C). It is known that C2C12 myotubes are able to produce ketone bodies, especially when contracting [25] and that cholesterol is a major constituent of IMCL [47].
## 3.5. Setting Transcription in Cytosol?
In this study, the individual signals from cytosolic PLIN5 and PGC-1α remained unaltered after BCAA deprivation. Nevertheless, the same experiment led to an increase in the cytosolic association between these markers (Section 2.5), often with very clearly colocalizing aggregates (Figure 7B). The same two proteins are known to interact, even if not directly binding [17]. Moreover, PGC-1α can interact with nuclear receptors, such as estrogen-related receptor α (ERRα) and sirtuin 1 (SIRT1), which have also been shown to be present in cytosolic pools [48,49] (Figure 7C). Interestingly, PGC-1α-mediated upregulation of BCAA metabolism does require ERRα [20], whose activation is dependent on cholesterol [50].
The relationship between cytosolic ERRα-PGC-1α-SIRT1-PLIN5-IMCL complexes and BCAA availability is unclear. It is possible that, through PLIN5 coordination, cytosolic IMCL could provide cholesterol and MUFAs to ERRα and SIRT1, respectively (Figure 7C), thus triggering the translocation to nuclei and consequent activation of transcription factors. In fact, we observed that, under Normal BCAA, EPS led to an increased triple association between cytosolic PGC-1α-PLIN5-IMCL (Figure S4). More research is needed to clarify the dynamics between cytosolic pools of IMCL, PLINs, and related transcription factor co-activators.
## 3.6. Nuclear Affairs
Gallardo-Montejano et al. have demonstrated that catecholamine and fasting causes phosphorylation of PLIN5 and its translocation to nuclei, promoting PGC-1α activation via SIRT1 disinhibition [17]. The same authors hypothesized that exercise would cause a similar response. Our results support this hypothesis, as we unprecedentedly showed that, in response to EPS, PLIN5 does enrich myotube nuclei, further associating with PGC-1α (Section 2.5). More recently, Natj et al. elegantly demonstrated that PLIN5’s role in activating SIRT1-PGC-1α occurs via MUFAs binding and chaperoning from LDs towards the nuclei [18]. Although we observed an increase in both IMCL-PLIN5 and PGC-1α-PLIN5 associations in the nuclei, we did not observe any changes in IMCL-PGC-1α nuclear association (Figure S4). This suggests that the delivery of MUFAs by PLIN5 does not require close proximity between nuclear IMCL and PGC-1α aggregates, and that nuclear IMCL-PLIN5 association possibly has further roles to be explored.
Independent of EPS, we have observed a very sharp decline in nuclear PGC-1α after BCAA deprivation (Section 2.5). This may indicate that the function of PGC-1α as a nuclear transcription factor co-activator becomes hampered by BCAA deprivation. This is consistent with the decreases in lipid oxidation and lipogenesis we have previously observed from the same setup and experiment [39]. The precise mechanism behind such strong effects cannot be demonstrated by our study, but BCAAs are known inducers of the mammalian target of rapamycin complex 1 (mTORC1) [51], which in turn is an important activator of PGC-1α [52] (Figure 7B).
Lastly, LDs and other PLINs have also been reported from nuclei. While nuclear PLIN3 and PLIN5 were shown abundant, nuclear PLIN2 has been reported as virtually absent and largely unresponsive [17,53]. Our results corroborate the latter observation (Section 2.3). However, we observed that, when electrically stimulated, PLIN2 increasingly associates with PLIN5 in myotube nuclei, especially if deprived from BCAA (Section 2.4). Although it is believed that PLIN2 and PLIN5 do not bind each other [17,54], their spatial correlation under such stimuli may indicate some level of indirect interaction, leading to possible changes in the metabolic status. Future research is needed on potential coordinating roles between the different PLINs within nuclei.
## 3.7. Limitations and Strengths
Although a strong model to study effects of physical activity even in small cohorts [24,55], the number of twins and respective fibers were relatively small, especially for studying such dynamic events which tend to be very variable between cells. Despite this limitation, concerning IMCL and PLINs, this is the first study showing the effects of life-long physical activity in genetically similar humans, bringing additional insights to the area.
While it is our strength that we have two very different models of physical activity/exercise, the C2C12 and twin models used in this study should not be directly compared. In addition, the current C2C12 experimental model may be too glycolytic for robust IMCL metabolism studies. However, knowing that undifferentiated myoblasts are too far removed from muscle biology, we were able to classify and measure the signal from myotubes exclusively (Figure S6), producing novel observations. Nevertheless, future studies should try to replicate the same observations in other cell models, such as FACS sorted myofibers or satellite cells.
One of the focuses of this work was to investigate if BCAA deprivation would affect IMCL, PLINs and PGC-1α, thus establishing this unexplored link. We have not, however, studied the effects of BCAA over-supplementation on these markers and, therefore, this should be addressed in future research.
Moreover, given the suboptical nature of diffused biomolecules, we preferred to focus on overall signal intensity rather than solely on thresholded objects—hence, for instance, the focus on the term IMCL, more inclusive and not limited to LDs or TAG. With the current instruments, we could have confidently segmented the brighter marker aggregates of about ø 1 μm, but by doing so, we would be discarding precious information coming from smaller or more diffused aggregates (Figure 7A).
In our study, we cannot strictly pertain to protein–protein or IMCL–protein binding interactions, something that could be aided by fluorescence resonance energy transfer, cell fractionation or immunoprecipitation. Instead, we conducted ICA on distinct tissue and intracellular compartments, allowing us to investigate the distribution and spatial correlation between multiple markers, some of which are not immunopercepitable. Even when two given biomolecules do not directly bind, the statistical correlation between their signals can inform us about regions where such markers could—either directly or indirectly—associate and co-act on given signaling phenomena [16,56].
## 4.1. Human Twin Pairs
A total of 8 participants from 4 same-sex twins pairs (2 male and 2 female) with discordant leisure time physical activity (LTPA) for 32 years were identified from the Finnish Twin Cohort (Table 1). Discordance was based on a series of structured questions concerning leisure activity and physical activity during journeys to and from work. The leisure time (metabolic equivalent (MET)) index was calculated by assigning a multiple of the resting metabolic rate to each form of physical activity (intensity × duration × frequency) and expressed as a sum score of leisure time MET hours per day. It is worth noting that the active twins’ average LTPA score (13.8, Table 1), roughly corresponds to 1 h of running per day, for more than three decades. On the other hand, the inactive twins are not sedentary and endured basic levels of LTPA.
The study participants were advised not to exercise vigorously during the morning and two days before both of their laboratory visits (one visit for clinical examinations including exercise tests and one visit for biopsy studies). Muscle tissue samples were taken after an overnight fast between 8:00 a.m. and 10:00 a.m. under local anesthesia after skin cooling and disinfection. Using a suction technique with a Bergström’s needle (ø 5 mm), the muscle biopsy was taken from *Vastus lateralis* at the midpoint between Trochanter major and the lateral joint line of the knee. The sample was then mounted transversely on cork with Tissue Tek™ (Miles, Elkhart, In, USA; Sakura, Cat. # 4583), and frozen rapidly (10–15 s) in isopentane (Fluka, Cat. # 59080), precooled to −160 °C in liquid nitrogen and stored at −80 °C.
For further details on participant description and recruiting procedures, see Leskinen et al. [ 2009, 2010] [24,55].
## 4.2. Myotube Experiments
Murine C2C12 myoblasts (American Type Culture Collection, ATCC, Manassas, VA, USA) were maintained in high glucose-containing Dulbecco’s Modified Eagle growth medium (GM) (4.5g·L−1, DMEM, #BE12-614F, Lonza, Basel, Switzerland) supplemented with $10\%$ (v/v) fetal bovine serum (FBS, #10270, Gibco, Rockville, MD, USA), 100U·mL−1 penicillin, 100μg·mL−1 streptomycin (P/S, #15140, Gibco) and 2 mM L-Glutamine (#17-605E, Lonza, Basel, Switzerland). Myoblasts were seeded on 6-well plates (NunclonTM Delta; Thermo Fisher Scientific, Waltham, MA, USA). When the myoblasts reached 95–$100\%$ confluence, the cells were rinsed with phosphate-buffered saline (PBS, pH 7.4), and the GM was replaced by differentiation medium (DM) containing high glucose DMEM, $2\%$ (v/v) horse serum (HS, 12449C, Sigma-Aldrich, St. Louis, MO, USA), 100U·mL−1 and 100μg·mL−1 P/S and 2 mM L-glutamine to promote differentiation into myotubes. Fresh DM was changed every second day. The cells were screened negative for mycoplasma contaminations, following manufacturer’s instructions (MycoSPY Master Mix Test Kit, M020, Biontex, München, Germany). The experiments were conducted on days 5–6 post differentiation and on a duplicated way.
The myotubes on 6-well plates were acclimatized to 0.1 mM oleic acid and 1 mM L-carnitine in normal BCAA DM on the day 4 post differentiation. On the next day, the electrodes were placed directly onto the wells. The electrical stimulation (1 Hz, 2 ms, 12 V) was applied to the cells using a C-Pace pulse generator (C-Pace EM, IonOptix, Milton, MA, USA) for 24 h at 37 °C with the same protocol as described earlier [25]. As described previously, EPS was paused after 22 h and target BCAA concentrations (no BCAA or normal BCAA) were employed to investigate the interactive effects of EPS and BCAA deprivation.
The BCAA deprivation experiments were carried out for 2 h at 37 °C in high-glucose BCAA-free DM (4.5 g · L−1, BioConcept, 1-26S289-I, Allschwill, Switzerland). The experimental groups were as follows: [1] cells supplemented with 0.8 mmol · L−1 of all BCAAs without EPS (Normal BCAA|Rest) or [2] with EPS (Normal BCAA|EPS), and [3] cells deprived (0.0 mmol · L−1) of all BCAA’s without EPS (No BCAA|Rest) or [4] with EPS (No BCAA|EPS).
## 4.3. Protein Extraction and Western Blotting
The C2C12 cells were harvested and Western blotting was conducted as previously described [25] with minor modifications. Briefly, 10 μg of total protein per samples were loaded on 4–$20\%$ Criterion TGX Stain-Free protein gels (#5678094, Bio-Rad Laboratories, Hercules, CA, USA) and samples were separated by SDS-PAGE.
To visualize proteins using stain-free technology, the gels were activated and the proteins were transferred to the PVDF membranes. Membranes were blocked with Intercept Blocking Buffer (#927-70001, LI-COR, Lincoln, NE, USA) followed by overnight incubation at 4 °C with primary antibody (PGC-1α, 1:10,000, ab191838, Abcam, Cambridge, UK) in Intercept Blocking Buffer diluted (v:v, 1:1) with Tris-buffered saline (TBS) with $0.1\%$ Tween-20.
Membranes were incubated with the horseradish peroxidase-conjugated secondary IgG antibody (anti-Rabbit, 1:4,0000) (Jackson ImmunoResearch Laboratories, West Grove, PA, USA) in Intercept Blocking Buffer diluted (v:v, 1:1) with TBS-$0.1\%$ Tween 20. Enhanced chemiluminescence (SuperSignal west femto maximum sensitivity substrate; Pierce Biotechnology, Rockford, IL, USA) and ChemiDoc MP device (Bio-Rad Laboratories) were together used for protein visualization.
Stain free (75-250 kDA area of the lanes) was used as a loading control and for the normalization of the results.
## 4.4. Gene-Expression Arrays
The RNA preparation, cRNA generation and microarray hybridization procedures were used as previously described [24]. In brief, Trizol-reagent (Invitrogen, Carlsbad, CA, USA) was used to isolate total RNA from the twin muscle biopsies, which were homogenized on FastPrep FP120 apparatus (MP Biomedicals, Illkirch, France).
An Illumina RNA amplification kit (Ambion, Austin, TX, USA) was used according to the manufacturer’s instructions to obtain biotinlabeled cRNA from 500 ng of total RNA.
Hybridizations to Illumina HumanWG-6 v3.0 Expression BeadChips (Illumina Inc., San Diego, CA, USA) containing probes for PLIN2 and PLIN5, were performed by the Finnish DNA Microarray Center at Turku Center for Biotechnology according to the Illumina BeadStation 500x manual.
Hybridized probes were detected with Cyanin-3-streptavidin (1μg·mL−1, Amersham Biosciences, GE Healthcare, Uppsala, Sweden) using Illumina BeadArray Reader (Illumina Inc.) and BeadStudio v3 software (Illumina Inc.).
*The* gene expression data and the raw data sets for skeletal muscle have been deposited in the GEO database (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE20319, accessed on 5 January 2023).
## 4.5. Histology
For each twin, two 8 μm cross sections were made in a cryostat at −25 °C (Leica CM 3000, Wetzlar, Germany) and collected onto 13 mm round coverslips. For the cell culture experiments, duplicate 6-well plates were used containing three 13 mm round coverslips per well. Each experimental group was measured from 18 coverslips. After the 24 h of EPS, the plates were removed from the incubator and the medium aspirated. In both the human and C2C12 experiments, the samples were immediately fixed in $4\%$ paraformaldehyde for 15 min at room temperature (RT). After washing for 3 × 5 min with PBS, the samples were blocked with $10\%$ goat serum (GS) in PBS-$0.05\%$ saponin (PBSap) for 30 min and then washed briefly with PBSap. Primary antibodies were diluted in $1\%$ GS-PBSap and incubated for 1 h at RT. A 3 × 10 min wash in PBSap ensued before incubating the secondary antibodies for 1 h at RT. Excess antibody was removed with another 3 × 10 min wash in PBSap. Finally, non-immuno stains were incubated for 30 min before 2 × 10 s washed with PBS. Thorough vial mixing and smooth rocking were ensured for every incubation step in order to grant an even stain.
For the twin studies, two different quadruple staining procedures took place, each sharing as common markers: LD540 for IMCL (0.1 μg· mL−1,57]), caveolin 3 for sarcolemma (2 μg· mL−1, PA1-066, Thermo Fisher Scientific) and slow myosin heavy chain (MyHC) for type I fibers (2 μg· mL−1, A4.951, DSHB, University of Iowa, IA, USA). While one section was further incubated for PLIN2 (1:200 dilution, GP47, Progen), the second section was instead incubated for PLIN5 (1:200 dilution, GP31, Progen). Respectively, the following secondary antibodies were used in combination: Alexa Fluor 405 Goat anti-Rabbit IgG (H+L), Alexa Fluor 594 Goat anti-Mouse IgG (H+L) and Alexa Fluor 488 Goat anti-Guinea Pig IgG (H+L) (Thermo Fisher Scientific).
For the C2C12 samples, a quintuple staining was performed, using 3 antibody markers: differentiated myotubes (5 μg· mL−1, MF-20, DSHB), PLIN5 (1:200 dilution, GP31, Progen), plus either PLIN2 (5 μg· mL−1, ab52356, Abcam) or PGC-1α (5 μg· mL−1, ab191838, Abcam). Respectively, the following secondary antibodies were used in combination: Alexa Fluor 647 Donkey anti-Mouse IgG (H+L), Alexa Fluor 488 Goat anti-Guinea Pig IgG (H+L) and Alexa Fluor 594 Donkey anti-Rabbit IgG (H+L) (Thermo Fisher Scientific). Additionally, IMCL (0.1 μg· mL−1, LD540, [57]) and nuclei (5 μg· mL−1, DAPI, Thermo Fisher Scientific) were stained in all coverslips. Cross reactivity was successfully ruled out by carefully controlling every antibody combination.
Every coverslip was mounted on microscopy slides using Mowiol with $2.5\%$ DABCO (Sigma-Aldrich) and left to dry for 24 h in the dark at 4 °C. Imaging took place within 48 h after mounting.
## 4.6. Image Acquisition
All image data were acquired on a LSM700 confocal microscope using the ZEN black software (Zeiss, Germany). Twin data were collected with a Plan-Apochromat 20x/0.8 objective (Zeiss, Germany), producing 2 images per participant, each image covering an area of 320.1 × 320.1 μm (voxel size = 0.31 × 0.31 × 2.4 μm). Cell data were collected with a Plan-Apochromat 63×/1.4 oil objective (Zeiss, Germany) from 3 random 203.2 × 203.2 μm confluent areas in each coverslip, producing 9 images per well and a total of 54 images per experimental group (voxel size = 0.1 × 0.1 × 1 μm).
Multi channel acquisition was achieved through the use of 4 laser lines (405, 488, 555 and 639 nm). Bleed-through was successfully avoided in each channel by manually configuring the secondary dichroic mirror position over two different photomultiplier tubes. Control samples incubated solely with secondary antibodies were used to set background values.
## 4.7. Image Analyses
For C2C12, the MF-20 signal was used to segment and analyze differentiated myotubes only. The nuclei detected within the segmented myotubes were also segmented, thus allowing compartmental analyses on the cytosolic versus nucleic markers. For the human cross sections, only intact and artifact-free fibers were segmented (Active = 19.3 ± 2.5 and Inactive = 15.5 ± 4.3. Mean ± SE). Segmentation was aided by machine learning algorithms using the Trainable Weka Segmentation tool [58] in Fiji [59]. Each analyzed fiber cross section was classified into either type I or type II, according to the detected and thresholded signal of slow myosin per cell area. The optical density of each marker was determined by measuring the mean value of the respective signal within each cell. The level of association between the different markers was determined through pixel-to-pixel intensity correlation analysis (ICA), by thresholding image data according to Costes et al. [ 16] in Fiji. In order to ensure a zero valued background, prior to analyses, all images were denoised and deconvoluted in Fiji using a theoretical point spread function separately for each channel.
## 4.8. Data Cleaning and Statistics
For the twin data, given the low number of participants, statistical cases are constituted of individual muscle fibers. Concerning the C2C12 data, from each coverslip, the 2 closest values per variable were averaged, while from each well, the values from the 2 closest coverslips were further averaged. To control for inter cell batch variability, all values were normalized against the control group (Normal BCAA|Rest). Finally, for both human and C2C12 studies, outliers were identified and removed via z-score (2 standard deviations).
Boxes in the boxplot figures depict interquartile ranges and medians, while whiskers represent the $95\%$ confidence interval, unless stated otherwise. The main effect significance is marked with # and combined group significance is marked with *, while interacting effects between independent variables are expressed with &. Normality was assessed with Shapiro–Wilk tests and group comparisons were preformed with either Mann–Whitney U tests or t-tests, depending on data distribution. Interacting effects were tested with a two-way ANOVA. Given the large number of human muscle fibers, significance levels were set at $p \leq 0.01$ and $p \leq 0.001.$ For C2C12, the significance levels were set at $p \leq 0.05$ and $p \leq 0.01.$ The ICA between the different markers were tested with a Manders split coefficent test after the thresholding step mentioned in Section 4.7.
Data crunching, statistics and boxplot visualization were performed in Python 3.9.0, with the packages NumPy [60], pandas [61], SciPy [62], statsmodels [63], seaborn [64] and matplotlib [65], respectively. All Fiji and Python routines can be found at https://github.com/seiryoku-zenyo/twinC2C12-studies, accessed on 1 August 2022.
Comparison and analysis of mRNA data were performed with the GEO2R functionality on the GEO database (https://www.ncbi.nlm.nih.gov/geo/geo2r/?acc=GSE20319, accessed on 5 January 2023).
Protein network analysis and visualization were performed with STRING [66] and Cytoscape [67] software.
## 5. Conclusions
The present study expands our basic knowledge on the known link between a functional BCAA metabolism, physical activity and an efficient lipid metabolism: specifically, via PLIN2’s function in coating IMCL upon BCAA availability and long-term physical activity. Moreover, we showed that PLIN5 has an ability to translocate to nuclei and associate with PGC-1α after contractions.
## References
1. Pedersen B.K.. **Muscle as a Secretory Organ**. *Comprehensive Physiology* (2013.0) 1337-1362. DOI: 10.1002/cphy.c120033
2. Goodpaster B.H., He J., Watkins S., Kelley D.E.. **Skeletal Muscle Lipid Content and Insulin Resistance: Evidence for a Paradox in Endurance-Trained Athletes**. *J. Clin. Endocrinol. Metab.* (2001.0) **86** 5755-5761. DOI: 10.1210/jcem.86.12.8075
3. Moro C., Bajpeyi S., Smith S.R.. **Determinants of intramyocellular triglyceride turnover: Implications for insulin sensitivity**. *Am. J. Physiol. Endocrinol. Metab.* (2008.0) **294** E203-E213. DOI: 10.1152/ajpendo.00624.2007
4. van Loon L.J.C., Koopman R., Manders R., van der Weegen W., van Kranenburg G.P., Keizer H.A.. **Intramyocellular lipid content in type 2 diabetes patients compared with overweight sedentary men and highly trained endurance athletes**. *Am. J. Physiol. Endocrinol. Metab.* (2004.0) **287** E558-E565. DOI: 10.1152/ajpendo.00464.2003
5. Sztalryd C., Brasaemle D.L.. **The perilipin family of lipid droplet proteins: Gatekeepers of intracellular lipolysis**. *Biochim. Biophys. Acta Mol. Cell Biol. Lipids* (2017.0) **1862** 1221-1232. DOI: 10.1016/j.bbalip.2017.07.009
6. MacPherson R.E., Peters S.J.. **Piecing together the puzzle of perilipin proteins and skeletal muscle lipolysis**. *Appl. Physiol. Nutr. Metab.* (2015.0) **40** 641-651. DOI: 10.1139/apnm-2014-0485
7. Prats C., Donsmark M., Qvortrup K., Londos C., Sztalryd C., Holm C., Galbo H., Ploug T.. **Decrease in intramuscular lipid droplets and translocation of HSL in response to muscle contraction and epinephrine**. *J. Lipid Res.* (2006.0) **47** 2392-2399. DOI: 10.1194/jlr.M600247-JLR200
8. Whytock K., Shepherd S., Wagenmakers A., Strauss J.. **Hormone sensitive lipase preferentially redistributes to perilipin-5 lipid droplets in human skeletal muscle during moderate-intensity exercise**. *J. Physiol.* (2018.0) **596** 2077-2090. DOI: 10.1113/JP275502
9. **Unraveling the roles of PLIN5: Linking cell biology to physiology**. *Trends Endocrinol. Metab.* (2015.0) **26** 144-152. DOI: 10.1016/j.tem.2015.01.005
10. Gemmink A., Bosma M., Kuijpers H.J., Hoeks J., Schaart G., van Zandvoort M.A., Schrauwen P., Hesselink M.K.. **Decoration of intramyocellular lipid droplets with PLIN5 modulates fasting-induced insulin resistance and lipotoxicity in humans**. *Diabetologia* (2016.0) **59** 1040-1048. DOI: 10.1007/s00125-016-3865-z
11. Gemmink A., Daemen S., Brouwers B., Hoeks J., Schaart G., Knoops K., Schrauwen P., Hesselink M.K.. **Decoration of myocellular lipid droplets with perilipins as a marker for in vivo lipid droplet dynamics: A super-resolution microscopy study in trained athletes and insulin resistant individuals**. *Biochim. Biophys. Acta Mol. Cell Biol. Lipids* (2021.0) **1866** 158852. DOI: 10.1016/j.bbalip.2020.158852
12. Gemmink A., Daemen S., Brouwers B., Huntjens P.R., Schaart G., Moonen-Kornips E., Jörgensen J., Hoeks J., Schrauwen P., Hesselink M.K.C.. **Dissociation of intramyocellular lipid storage and insulin resistance in trained athletes and type 2 diabetes patients; involvement of perilipin 5?**. *J. Physiol.* (2018.0) **596** 857-868. DOI: 10.1113/JP275182
13. Shepherd S.O., Cocks M., Tipton K.D., Ranasinghe A.M., Barker T.A., Burniston J.G., Wagenmakers A.J.M., Shaw C.S.. **Sprint interval and traditional endurance training increase net intramuscular triglyceride breakdown and expression of perilipin 2 and 5**. *J. Physiol.* (2013.0) **591** 657-675. DOI: 10.1113/jphysiol.2012.240952
14. Shaw C.S., Sherlock M., Stewart P.M., Wagenmakers A.J.. **Adipophilin distribution and colocalisation with lipid droplets in skeletal muscle**. *Histochem. Cell Biol.* (2009.0) **131** 575-581. DOI: 10.1007/s00418-009-0558-4
15. Koh H.C.E., Ørtenblad N., Winding K.M., Hellsten Y., Mortensen S.P., Nielsen J.. **High-intensity interval, but not endurance, training induces muscle fiber type-specific subsarcolemmal lipid droplet size reduction in type 2 diabetic patients**. *Am. J. Physiol. Endocrinol. Metab.* (2018.0) **315** E872-E884. DOI: 10.1152/ajpendo.00161.2018
16. Costes S.V., Daelemans D., Cho E.H., Dobbin Z., Pavlakis G., Lockett S.. **Automatic and quantitative measurement of protein-protein colocalization in live cells**. *Biophys. J.* (2004.0) **86** 3993-4003. DOI: 10.1529/biophysj.103.038422
17. Gallardo-Montejano V.I., Saxena G., Kusminski C.M., Yang C., McAfee J.L., Hahner L., Hoch K., Dubinsky W., Narkar V.A., Bickel P.E.. **Nuclear Perilipin 5 integrates lipid droplet lipolysis with PGC-1**. *Nat. Commun.* (2016.0) **7** 12723. DOI: 10.1038/ncomms12723
18. Najt C.P., Khan S.A., Heden T.D., Witthuhn B.A., Perez M., Heier J.L., Mead L.E., Franklin M.P., Karanja K.K., Graham M.J.. **Lipid Droplet-Derived Monounsaturated Fatty Acids Traffic via PLIN5 to Allosterically Activate SIRT1**. *Mol. Cell* (2020.0) **77** 810-824. DOI: 10.1016/j.molcel.2019.12.003
19. Hatazawa Y., Tadaishi M., Nagaike Y., Morita A., Ogawa Y., Ezaki O., Takai-Igarashi T., Kitaura Y., Shimomura Y., Kamei Y.. **PGC-1**. *PLoS ONE* (2014.0) **9**. DOI: 10.1371/journal.pone.0091006
20. Sjögren R.J., Rizo-Roca D., Chibalin A.V., Chorell E., Furrer R., Katayama S., Harada J., Karlsson H.K., Handschin C., Moritz T.. **Branched-chain amino acid metabolism is regulated by ERR**. *Diabetologia* (2021.0) **64** 2077-2091. DOI: 10.1007/s00125-021-05481-9
21. Lerin C., Goldfine A.B., Boes T., Liu M., Kasif S., Dreyfuss J.M., De Sousa-Coelho A.L., Daher G., Manoli I., Sysol J.R.. **Defects in muscle branched-chain amino acid oxidation contribute to impaired lipid metabolism**. *Mol. Metab.* (2016.0) **5** 926-936. DOI: 10.1016/j.molmet.2016.08.001
22. Kainulainen H., Hulmi J.J., Kujala U.M.. **Potential role of branched-chain amino acid catabolism in regulating fat oxidation**. *Exerc. Sport Sci. Rev.* (2013.0) **41** 194-200. DOI: 10.1097/JES.0b013e3182a4e6b6
23. Nye C.K., Hanson R.W., Kalhan S.C.. **Glyceroneogenesis is the dominant pathway for triglyceride glycerol synthesis in vivo in the rat**. *J. Biol. Chem.* (2008.0) **283** 27565-27574. DOI: 10.1074/jbc.M804393200
24. Leskinen T., Rinnankoski-Tuikka R., Rintala M., Seppänen-Laakso T., Pöllänen E., Alen M., Sipilä S., Kaprio J., Kovanen V., Rahkila P.. **Differences in Muscle and Adipose Tissue Gene Expression and Cardio-Metabolic Risk Factors in the Members of Physical Activity Discordant Twin Pairs**. *PLoS ONE* (2010.0) **5**. DOI: 10.1371/journal.pone.0012609
25. Lautaoja J.H., O’Connell T.M., Mäntyselkä S., Peräkylä J., Kainulainen H., Pekkala S., Permi P., Hulmi J.J.. **Higher glucose availability augments the metabolic responses of the C2C12 myotubes to exercise-like electrical pulse stimulation**. *Am. J. Physiol. Endocrinol. Metab.* (2021.0) **321** E229-E245. DOI: 10.1152/ajpendo.00133.2021
26. Nikolić N., Bakke S.S., Kase E.T., Rudberg I., Halle I.F., Rustan A.C., Thoresen G.H., Aas V.. **Correction: Electrical Pulse Stimulation of Cultured Human Skeletal Muscle Cells as an In Vitro Model of Exercise**. *PLoS ONE* (2013.0) **8**. DOI: 10.1371/annotation/270b432d-50ec-41f1-ad4d-ddd9f51f62a5
27. Kankaanpää A., Tolvanen A., Bollepalli S., Leskinen T., Kujala U.M., Kaprio J., Ollikainen M., Sillanpää E.. **Leisure-time and occupational physical activity associates differently with epigenetic aging**. *Med. Sci. Sport. Exerc.* (2021.0) **53** 487. DOI: 10.1249/MSS.0000000000002498
28. Ko K., Woo J., Bae J.Y., Roh H.T., Lee Y.H., Shin K.O.. **Exercise training improves intramuscular triglyceride lipolysis sensitivity in high-fat diet induced obese mice**. *Lipids Health Dis.* (2018.0) **17** 81. DOI: 10.1186/s12944-018-0730-8
29. Ramos S.V., MacPherson R.E.K., Turnbull P.C., Bott K.N., LeBlanc P., Ward W.E., Peters S.J.. **Higher PLIN5 but not PLIN3 content in isolated skeletal muscle mitochondria following acute in vivo contraction in rat hindlimb**. *Physiol. Rep.* (2014.0) **2** e12154. DOI: 10.14814/phy2.12154
30. Rinnankoski-Tuikka R., Hulmi J.J., Torvinen S., Silvennoinen M., Lehti M., Kivelä R., Reunanen H., Kujala U.M., Kainulainen H.. **Lipid droplet-associated proteins in high-fat fed mice with the effects of voluntary running and diet change**. *Metabolism* (2014.0) **63** 1031-1040. DOI: 10.1016/j.metabol.2014.05.010
31. Bosma M., Hesselink M.K., Sparks L.M., Timmers S., Ferraz M.J., Mattijssen F., van Beurden D., Schaart G., de Baets M.H., Verheyen F.K.. **Perilipin 2 improves insulin sensitivity in skeletal muscle despite elevated intramuscular lipid levels**. *Diabetes* (2012.0) **61** 2679-2690. DOI: 10.2337/db11-1402
32. Bonen A., Luiken J., Liu S., Dyck D., Kiens B., Kristiansen S., Turcotte L., Van Der Vusse G., Glatz J.. **Palmitate transport and fatty acid transporters in red and white muscles**. *Am. J. Physiol. Endocrinol. Metab.* (1998.0) **275** E471-E478. DOI: 10.1152/ajpendo.1998.275.3.E471
33. Fachada V., Rahkila P., Fachada N., Turpeinen T., Kujala U.M., Kainulainen H.. **Enlarged PLIN5-uncoated lipid droplets in inner regions of skeletal muscle type II fibers associate with type 2 diabetes**. *Acta Histochem.* (2022.0) **124** 151869. DOI: 10.1016/j.acthis.2022.151869
34. Yang B., Yu Q., Chang B., Guo Q., Xu S., Yi X., Cao S.. **MOTS-c interacts synergistically with exercise intervention to regulate PGC-1**. *Biochim. Biophys. Acta Mol. Basis Dis.* (2021.0) **1867** 166126. DOI: 10.1016/j.bbadis.2021.166126
35. Philp A., Belew M.Y., Evans A., Pham D., Sivia I., Chen A., Schenk S., Baar K.. **The PGC-1**. *Am. J. Physiol. Regul. Integr. Comp. Physiol.* (2011.0) **301** R864-R872. DOI: 10.1152/ajpregu.00232.2011
36. Melouane A., Yoshioka M., Kanzaki M., St-Amand J.. **Sparc, an EPS-induced gene, modulates the extracellular matrix and mitochondrial function via ILK/AMPK pathways in C2C12 cells**. *Life Sci.* (2019.0) **229** 277-287. DOI: 10.1016/j.lfs.2019.05.070
37. Lee I.H., Lee Y.J., Seo H., Kim Y.S., Nam J.O., Jeon B.D., Kwon T.D.. **Study of muscle contraction induced by electrical pulse stimulation and nitric oxide in C2C12 myotube cells**. *J. Exerc. Nutr. Biochem.* (2018.0) **22** 22. DOI: 10.20463/jenb.2018.0004
38. Finck B.N., Kelly D.P.. **PGC-1 coactivators: Inducible regulators of energy metabolism in health and disease**. *J. Clin. Investig.* (2006.0) **116** 615-622. DOI: 10.1172/JCI27794
39. Karvinen S., Fachada V., Sahinaho U.M., Pekkala S., Lautaoja J.H., Mäntyselkä S., Permi P., Hulmi J.J., Silvennoinen M., Kainulainen H.. **Branched-Chain Amino Acid Deprivation Decreases Lipid Oxidation and Lipogenesis in C2C12 Myotubes**. *Metabolites* (2022.0) **12**. DOI: 10.3390/metabo12040328
40. Son Y.H., Lee S.M., Lee S.H., Yoon J.H., Kang J.S., Yang Y.R., Kwon K.S.. **Comparative molecular analysis of endurance exercise in vivo with electrically stimulated in vitro myotube contraction**. *J. Appl. Physiol.* (2019.0) **127** 1742-1753. DOI: 10.1152/japplphysiol.00091.2019
41. Philp A., Perez-Schindler J., Green C., Hamilton D.L., Baar K.. **Pyruvate suppresses PGC1**. *Am. J. Physiol. Cell Physiol.* (2010.0) **299** C240-C250. DOI: 10.1152/ajpcell.00438.2009
42. Brown D.M., Parr T., Brameld J.M.. **Myosin heavy chain mRNA isoforms are expressed in two distinct cohorts during C2C12 myogenesis**. *J. Muscle Res. Cell Motil.* (2012.0) **32** 383-390. DOI: 10.1007/s10974-011-9267-4
43. McIntosh A.L., Senthivinayagam S., Moon K.C., Gupta S., Lwande J.S., Murphy C.C., Storey S.M., Atshaves B.P.. **Direct interaction of Plin2 with lipids on the surface of lipid droplets: A live cell FRET analysis**. *Am. J. Physiol. Cell Physiol.* (2012.0) **303** C728-C742. DOI: 10.1152/ajpcell.00448.2011
44. Zhao H., Zhang F., Sun D., Wang X., Zhang X., Zhang J., Yan F., Huang C., Xie H., Lin C.. **Branched-chain amino acids exacerbate obesity-related hepatic glucose and lipid metabolic disorders via attenuating Akt2 signaling**. *Diabetes* (2020.0) **69** 1164-1177. DOI: 10.2337/db19-0920
45. Kim S., Jeon J.M., Kwon O.K., Choe M.S., Yeo H.C., Peng X., Cheng Z., Lee M.Y., Lee S.. **Comparative Proteomic Analysis Reveals the Upregulation of Ketogenesis in Cardiomyocytes Differentiated from Induced Pluripotent Stem Cells**. *Proteomics* (2019.0) **19** 1800284. DOI: 10.1002/pmic.201800284
46. Halama A., Horsch M., Kastenmüller G., Möller G., Kumar P., Prehn C., Laumen H., Hauner H., Hrabĕ de Angelis M., Beckers J.. **Metabolic switch during adipogenesis: From branched chain amino acid catabolism to lipid synthesis**. *Arch. Biochem. Biophys.* (2016.0) **589** 93-107. DOI: 10.1016/j.abb.2015.09.013
47. Itabe H., Yamaguchi T., Nimura S., Sasabe N.. **Perilipins: A diversity of intracellular lipid droplet proteins**. *Lipids Health Dis.* (2017.0) **16** 83. DOI: 10.1186/s12944-017-0473-y
48. Rossi M., Colecchia D., Iavarone C., Strambi A., Piccioni F., Verrotti di Pianella A., Chiariello M.. **Extracellular Signal-regulated Kinase 8 (ERK8) Controls Estrogen-related Receptor**. *J. Biol. Chem.* (2011.0) **286** 8507-8522. DOI: 10.1074/jbc.M110.179523
49. Nasrin N., Kaushik V.K., Fortier E., Wall D., Pearson K.J., De Cabo R., Bordone L.. **JNK1 phosphorylates SIRT1 and promotes its enzymatic activity**. *PloS ONE* (2009.0) **4**. DOI: 10.1371/journal.pone.0008414
50. Wei W., Schwaid A.G., Wang X., Wang X., Chen S., Chu Q., Saghatelian A., Wan Y.. **Ligand activation of ERR**. *Cell Metab.* (2016.0) **23** 479-491. DOI: 10.1016/j.cmet.2015.12.010
51. Neishabouri S.H., Hutson S., Davoodi J.. **Chronic activation of mTOR complex 1 by branched chain amino acids and organ hypertrophy**. *Amino Acids* (2015.0) **47** 1167-1182. DOI: 10.1007/s00726-015-1944-y
52. Cunningham J.T., Rodgers J.T., Arlow D.H., Vazquez F., Mootha V.K., Puigserver P.. **mTOR controls mitochondrial oxidative function through a YY1–PGC-1**. *Nature* (2007.0) **450** 736-740. DOI: 10.1038/nature06322
53. Ohsaki Y., Kawai T., Yoshikawa Y., Cheng J., Jokitalo E., Fujimoto T.. **PML isoform II plays a critical role in nuclear lipid droplet formation**. *J. Cell Biol.* (2016.0) **212** 29-38. DOI: 10.1083/jcb.201507122
54. Gemmink A., Daemen S., Kuijpers H.J., Schaart G., Duimel H., López-Iglesias C., van Zandvoort M.A., Knoops K., Hesselink M.K.. **Super-resolution microscopy localizes perilipin 5 at lipid droplet-mitochondria interaction sites and at lipid droplets juxtaposing to perilipin 2**. *Biochim. Biophys. Acta Mol. Cell Biol. Lipids* (2018.0) **1863** 1423-1432. DOI: 10.1016/j.bbalip.2018.08.016
55. Leskinen T., Waller K., Mutikainen S., Aaltonen S., Ronkainen P.H.A., Alén M., Sipilä S., Kovanen V., Perhonen M., Pietiläinen K.H.. **Effects of 32-Year Leisure Time Physical Activity Discordance in Twin Pairs on Health (TWINACTIVE Study): Aims, Design and Results for Physical Fitness**. *Twin Res. Hum. Genet.* (2009.0) **12** 108-117. DOI: 10.1375/twin.12.1.108
56. Tan H.Y., Qiu Y.T., Sun H., Yan J.w., Zhang L.. **A lysosome-targeting dual-functional fluorescent probe for imaging intracellular viscosity and beta-amyloid**. *Chem. Commun.* (2019.0) **55** 2688-2691. DOI: 10.1039/C9CC00113A
57. Spandl J., White D.J., Peychl J., Thiele C.. **Live Cell Multicolor Imaging of Lipid Droplets with a New Dye, LD540**. *Traffic* (2009.0) **10** 1579-1584. DOI: 10.1111/j.1600-0854.2009.00980.x
58. Arganda-Carreras I., Kaynig V., Rueden C., Eliceiri K.W., Schindelin J., Cardona A., Sebastian Seung H.. **Trainable Weka Segmentation: A machine learning tool for microscopy pixel classification**. *Bioinformatics* (2017.0) **33** 2424-2426. DOI: 10.1093/bioinformatics/btx180
59. Schindelin J., Arganda-Carreras I., Frise E., Kaynig V., Longair M., Pietzsch T., Preibisch S., Rueden C., Saalfeld S., Schmid B.. **Fiji: An open-source platform for biological-image analysis**. *Nat. Methods* (2012.0) **9** 676-682. DOI: 10.1038/nmeth.2019
60. Harris C.R., Millman K.J., van der Walt S.J., Gommers R., Virtanen P., Cournapeau D., Wieser E., Taylor J., Berg S., Smith N.J.. **Array programming with NumPy**. *Nature* (2020.0) **585** 357-362. DOI: 10.1038/s41586-020-2649-2
61. **pandas-dev/pandas: Pandas, 2020**. DOI: 10.5281/zenodo.3509134
62. Virtanen P., Gommers R., Oliphant T.E., Haberland M., Reddy T., Cournapeau D., Burovski E., Peterson P., Weckesser W., Bright J.. **SciPy 1.0: Fundamental Algorithms for Scientific Computing in Python**. *Nat. Methods* (2020.0) **17** 261-272. DOI: 10.1038/s41592-019-0686-2
63. Seabold S., Perktold J.. **Statsmodels: Econometric and statistical modeling with python**. *Proceedings of the 9th Python in Science Conference*
64. Waskom M.. **The Seaborn Development Team. Mwaskom/Seaborn, 2020**. DOI: 10.5281/zenodo.592845
65. Hunter J.D.. **Matplotlib: A 2D graphics environment**. *Comput. Sci. Eng.* (2007.0) **9** 90-95. DOI: 10.1109/MCSE.2007.55
66. Szklarczyk D., Gable A.L., Lyon D., Junge A., Wyder S., Huerta-Cepas J., Simonovic M., Doncheva N.T., Morris J.H., Bork P.. **STRING v11: Protein–protein association networks with increased coverage, supporting functional discovery in genome-wide experimental datasets**. *Nucleic Acids Res.* (2018.0) **47** D607-D613. DOI: 10.1093/nar/gky1131
67. Shannon P., Markiel A., Ozier O., Baliga N.S., Wang J.T., Ramage D., Amin N., Schwikowski B., Ideker T.. **Cytoscape: A software environment for integrated models of biomolecular interaction networks**. *Genome Res.* (2003.0) **13** 2498-2504. DOI: 10.1101/gr.1239303
|
---
title: 'Enjoyed by Jack but Endured by Jill: An Exploratory Case Study Examining Differences
in Adolescent Design Preferences and Perceived Impacts of a Secondary Schoolyard'
authors:
- Gweneth Leigh
- Milica Muminovic
- Rachel Davey
journal: International Journal of Environmental Research and Public Health
year: 2023
pmcid: PMC10002286
doi: 10.3390/ijerph20054221
license: CC BY 4.0
---
# Enjoyed by Jack but Endured by Jill: An Exploratory Case Study Examining Differences in Adolescent Design Preferences and Perceived Impacts of a Secondary Schoolyard
## Abstract
The school grounds provide students opportunities for respite, relaxation and relief from daily stresses during breaks in the school day. However, it is unclear whether secondary schoolyard designs adequately support the diverse and evolving needs of adolescents, particularly at a time when they are experiencing rapid emotional and physical developmental change. To investigate this, quantitative methods were used to explore differences in perceptions of schoolyard attractiveness and restorative quality based on student gender and year level. A school-wide survey was administered to approximately 284 students in years 7 to 10 at a secondary school in Canberra, Australia. Results indicate significant declines in student perceptions of schoolyard attractiveness and restorative quality. Higher ratings of schoolyard likeability, accessibility, personal connection and restorative quality of ‘being away’ were associated with male students across all year levels. Further work is needed to explore how schoolyard environments can better support the design preferences and well-being needs of older and female students. Such information would help planners, designers and land managers develop schoolyard designs that are more equitable in their benefits to secondary school students of different genders and year levels.
## 1. Introduction
Schoolyard design shapes space in ways that affect how students physically, socially and emotionally engage with one another at school. In a time when peer opportunities for play continue to diminish [1], school grounds provide a place for students to connect with friends and mediate social stresses [2], escape from classroom structures, freely assert their identity and develop a sense of belonging [3]. However, to attain these benefits, schoolyard spaces need to be designed in ways that are inclusive and accessible to the needs of both individual students and the broader school community.
While the design of school buildings continues to evolve in response to environmental and pedagogical needs [4], the pace of change within the design of school grounds has been slow. For decades, most schoolyard design standards have remained focused on the provision of grassed sporting ovals, concrete courtyards and standardized play structures [5,6], with little emphasis on natural diversity [7]. Such spaces are traditionally designed from an adult perspective, guided more by priorities of safety and risk management than around student affordances for physical activity and well-being [3,8,9,10]. As students age, such spaces are increasingly regarded by them as ‘dull and frustrating’ [7].
Time outdoors provides positive health benefits to youth [3,11]. However, the links between schoolyard quality and adolescent health remains understudied, particularly at a time when their mental health issues are growing [12,13]. Students become less active in the schoolyard in secondary school [14]; some have attributed this to low school ground quality [15]. While secondary schoolyards provide students with a daily dose of green space, it is unclear whether its sport-dominated focus is equitable in the affordances provided to students of different genders and ages.
Mounting evidence demonstrates male and female students engage with schoolyard spaces differently. Quantitative tools mapping student physical activity levels reveal male students are more active during recess, while female students are more social and sedentary [2,16,17,18,19]. Several factors contribute towards these differences. In focus group discussions with adolescent girls in Denmark, Pawlowski et al. [ 2015] found that most did not go into the schoolyard due to ‘a lack of attractive outdoor play facilities’ [20]. In similar discussions at a Melbourne school with Year 6 students, Spark et al. [ 2019] found girls were excluded from active play spaces by male students [21].
Disparities in schoolyard use also exist based on student age, related in part to the physical and emotional changes associated with adolescence: the developmental needs of students entering Year 7 are quite different to students in Year 10. As students progress in secondary school, there is a shift from engaging in free play to an increased preference for more structured physical activities [5], although research shows this may be influenced by gender. In looking at student behaviors during school breaks, Raney et al. [ 2023] found that while older male students remained interested in sporting ovals and ball courts, such features did not sustain the interest of older female students [22].
Student perceptions of schoolyard design preferences and its well-being impacts is understudied. Although less documented [23,24,25,26], capturing student voice is recognized as an important attribute in creating effective schoolyard designs [27]. Using perceptions of attractiveness and restoration as indicators of design quality provide an important well-being metric based on user engagement with schoolyard spaces, a missing link among design rating systems that otherwise often focus on sustainability and operational costs [28].
According to Kaplan and Kaplan [1989], aesthetic preference is an indicator of the types of settings in which a person is able to best function [29]. Such places are associated with supporting restoration—the ability to rebuild or renew capabilities that help individuals to function in ways that are physical, cognitive and emotional [30,31]. Their Attention Restoration Theory asserts that time outdoors can help renew focus and alleviate mental fatigue [29]. Its four domains of focus—being away, fascination, extent and compatibility—provide a framework from which to build an understanding of the adequacy of outdoor settings to positively affect user well-being. Research into the attractiveness and perceived restorative quality of open space explores how design features affect the well-being benefits experienced by users [32,33,34,35,36,37]. However, its application to the environments of young people continues to be understudied [23].
This paper applies Kaplan and Kaplan’s restorative environment theory within a traditional school ground setting. It looks to build evidence on the restorative qualities of a secondary schoolyard as perceived and experienced by students, and to understand whether these student views are consistent across the school community. This was explored through the following two questions:Do differences exist in student perceptions of schoolyard attractiveness based on student year level and gender?Do differences exist in student perceptions of schoolyard restorative quality based on student year level and gender?
Schoolyard activities can make an important contribution to the physical, social and emotional development and well-being of students [38]. Schoolyard design quality is formative in these experiences and, as such, can influence the types of health benefits provided to users. Exploring ways to quantify these design impacts based on user experiences provides a metric for identifying where such spaces are performing well, and where additional investment is warranted.
## 2.1. Site Context
The study was structured as an exploratory case study [39] carried out at a public secondary school established in 1985 located in Canberra, Australia. The school consists of approximately 310 students in years 7–10. The school grounds cover approximately 16,000 square meters (51 square meters per child) and consist of a large grass amphitheater, paved ball courts, informal open space with grass and trees, and small shade structures near the canteen, shown in Figure 1. During the school day, students are provided two breaks, recess (thirty minutes) and lunch (forty minutes). Break times are characterized by free play supervised by teachers.
The school was recruited as part of a government initiative called It’s Your Move to co-design with students a multi-use outdoor space in the school grounds. This study was conducted during the conceptual design phase of the schoolyard intervention. Ethics approval was obtained from the Human Research Ethics Committee at the University of Canberra (HREC-6950) and from the ACT government (number RES-2104), followed by permission from the school principal. Participation by students was completely voluntary. A passive consent procedure was conducted through a notice to parents in the school newsletter one week prior to data collection. Students who decided to opt out were instructed to inform their teacher.
## 2.2. Data Collection
The total study population consisted of students aged 11 to 16 in years 7, 8, 9 and 10 (as classified by the Australian educational system). A paper survey was administered to students in each year level as a classroom activity immediately following recess. Collected data were used to investigate student perceptions of schoolyard attractiveness and its perceived restorative impact, with questions outlined further below and in Table 1. Participants provided their demographic information including age, year level and whether they identified as male or female. Survey responses did not require any identifiable student information and remained anonymous. A total of 284 students completed the survey; the teacher of each class collected the data. Each survey was assigned a numerical code during analysis and reporting.
## 2.2.1. Survey Instrument: Schoolyard Attractiveness
As part of the government schoolyard intervention, a design thinking workshop was facilitated by the ACT government and conducted in August 2020 with a selection of students to identify desired outcomes for the future renewed schoolyard. The top ten student design priorities from this workshop were summarized and incorporated into this study as a series of survey statements (e.g., ‘The design of the schoolyard provides a sense of belonging’ and ‘The design of this schoolyard is accessible by all’). Students undertaking the survey were to consider how each statement applied to their current schoolyard environment and circle the most suitable answer from a 5-point Likert scale, with 1 = Not at all and 5 = Completely true. Students also evaluated the likeability of their schoolyard based on a similar tool used by van Dijk-Wesselius et al. [ 2018] in studying student appreciation of the school grounds [40]. Schoolyard likeability was rated by students on a 10-point Likert scale from 1 = ‘I don’t like my schoolyard at all’ to 10 = ‘My schoolyard is fantastic, it could not be better’.
## 2.2.2. Survey Instrument: Schoolyard Restorative Quality
To determine the impact of the school ground design on their individual well-being, students rated the restorative value of the schoolyard. This was based on fifteen items derived from the ‘Perceived Restorative Components Scale for Children’ (PRCS-C II) developed by Bagot, Kuo and Allen [2007] [41]. Adapted from the Perceived Restorativeness Scale developed for adults by Hartig et al. [ 1997], this validated tool determines the restorative capacity of outdoor settings by measuring the qualities of person–environment transactions within these spaces [42]. Students responded to 15 measures based on the 5 restorative qualities defined by the Attention Restoration Theory [29,43]: Being away–physical, being away–psychological, fascination, extent and compatibility. Students were asked to consider how true each statement was to them and to circle the most suitable answer from a 5-point Likert scale, indicating the extent to which the statement described their experience in the schoolyard with 0 = Not at all and 4 = Completely true.
## 2.3. Data Analysis
Out of the 284 students who participated in the study, valid data were obtained from 252 students aged between 11 and 16 years (Mage = 13.59 years, SD = 1.16). Students were in year levels between 7 and 10; 52.8 per cent of total participants were male, as indicated in Table 2. Analyses were carried out using IBM SPSS Statistics (Version 27) for Windows.
Descriptive analyses were used to characterize study variables, including means and standard deviations. Unanswered survey questions were assigned a value of zero in the analysis. A principal component analysis (PCA) was conducted to identify broader themes within schoolyard attractiveness questions and to test internal consistencies for schoolyard restorative quality, followed by linear regressions using univariate analysis to examine differences in survey responses based on student year level and gender. This process is outlined further below. Factor loadings after rotation for both categories are provided in Supplementary Materials (Tables S1 and S2).
To identify broader themes within the category of Schoolyard Attractiveness, a PCA was conducted with oblique rotation. The Kaiser–Meyer–Olkin measure verified the sampling adequacy for the analysis, KMO = 0.930. An initial analysis was run to obtain eigenvalues for each factor in the data. As not all values were above 0.7, Kaiser’s criterion was deemed too strict and Joliffe’s criterion [44,45] was used instead, retaining all factors with eigenvalues more than 0.7. Two items—schoolyard likeability and accessibility—were removed for separate analysis, as their inclusion negatively impacted analysis reliability due to high eigenvalues but a lack of loading factors. Eight items were retained.
Factor analysis confirmed a bidimensional scale (70.47 per cent explained variance), reducing responses into two factors: ‘Personal connection’ and ‘Design adaptability’. The five design attributes clustered under the factor ‘Personal connection’ focused on those schoolyard impacts that enabled individuals to establish an emotional link with space through feelings of welcome, agency and safety (e.g., ‘The design of this schoolyard provides a sense of belonging’, ‘The design of this schoolyard promotes well-being’, ‘The design of this schoolyard is liberating’). The three attributes defining ‘Design adaptability’ focused on the ability of schoolyard spaces to create flexible environments to accommodate different needs and abilities of students (e.g., ‘The design of this schoolyard is diverse’ and ‘The design of this schoolyard provides age-appropriate play’). Items within each of these two factors were summed and a total schoolyard attractiveness score calculated for each; values could range from 5 to 25 for Personal connection and from 3 to 15 for Design adaptability. The scale for both components showed good reliability, with Cronbach’s alpha, ranging between 0.89 (Personal connection) and 0.75 (Design adaptability).
For the category of schoolyard restorative quality, a PCA was used to test internal consistencies on the 15 items using oblique rotation. The KMO measure verified the sampling adequacy for the analysis, KMO = 0.927; all KMO values for individual items were above the acceptable limit of 0.5 [46]. An initial analysis was run to obtain the eigenvalues for each factor in the data. Four factors had eigenvalues over Jolliffe’s criterion of 0.7, and in combination explained 71 per cent of the variance. Consequently, PRCS-C II measures were reduced from five to four components, with ‘Being away–psychological’ and ‘Being away–physical’ consolidated into one category, ‘Being away’, consisting of six items. Items for each of the three remaining components (fascination, compatibility and extent) align with the factorial structure of PRSC-C II, as identified by Bagot et al. [ 2007] [41]. These were summed and a total schoolyard perceived restorativeness score calculated. Values could range from 0 to 24 (Being away), 0 to 16 (Fascination), 0 to 8 (Compatibility) and 0 to 8 (Extent). Higher scores indicated higher perceived restorative quality. Reliability was good for all components, with Cronbach’s alpha ranging between 0.73 and 0.88 (Being away: 0.86, Fascination: 0.88, Compatibility: 0.81, Extent: 0.62).
Further investigations were conducted to identify the significance of initial findings of schoolyard attractiveness and restorative quality between sub-groups of students based on gender and year level. The normality of the variables was examined in a preliminary analysis using the Kolmogorov–Smirnov test. Residual plots were also conducted to evaluate the normal distribution of model residuals. As the distribution of responses to both schoolyard attractiveness and restorative quality were significantly non-normal, non-parametric versions of independent sample t-tests (Mann–Whitney tests) and analyses of variance (ANOVAs) were used to determine whether significant differences existed. These relationships were further explored through univariate linear regression analyses. The results are presented as unstandardized coefficients (b) with 95 per cent confidence intervals (CI). A p-value of 0.05 was used to indicate statistical significance.
## 3. Results
Unadjusted means and standard deviations for student responses to schoolyard attractiveness and restorative quality were calculated and are reported in Table 3. Students predominantly scored the schoolyard between the mid-to-lower range of scales. Based on the calculated means of total responses for each category, schoolyard accessibility was the highest rated (MAccessibility = 3.14 on a scale of 1 to 5), and schoolyard fascination was the lowest rated (MFascination = 5.08 on a scale of 1 to 16).
## 3.1. Significance of Student Year Level and Gender on Perceptions of Schoolyard Attractiveness
In terms of year level, perceptions of schoolyard attractiveness demonstrate significant declines by the time students reach Year 10; results are presented in Table 4. In comparison to Year 10 students, Year 7 students are associated with more positive ratings of schoolyard attractiveness across all categories, with likeability and adaptability both significant at the 1 per cent level; accessibility and personal connection are both significant at the 1 per cent level. These significant differences persist with Year 8 students, predicting more positive ratings of schoolyard likeability at the 5 per cent level ($b = 1.15$, $95\%$ CI 0.20 to 2.11), as well as Year 9 students at the 1 per cent level ($b = 1.52$, $95\%$ CI 0.57 to 2.47) in comparison to Year 10 students.
Significant differences also exist between male and female student perceptions of schoolyard attractiveness. Compared to females, male students predict more positive ratings of schoolyard accessibility at the 1 per cent level, with schoolyard likeability and personal connection at the 5 per cent level.
Further significant relationships are identified when combining the impact of both student gender and year level on perceptions of schoolyard attractiveness. Male students in Year 7 predict higher ratings of schoolyard likeability, accessibility and adaptability at the 5 per cent level when compared to female students in Year 10. This relationship is illustrated in Figure 2.
The differences in the means between male and female students for categories of adaptability, likeability and accessibility were nominal in Year 7, with adaptability at 7.4 per cent (MMaleAdapt7 = 9.23 and MFemaleAdapt7 = 9.91), likeability at 0.8 per cent (MMaleLike7 = 5.14 and MFemaleLike7 = 5.18), and accessibility at 1.8 per cent (MMaleAccess7 = 3.42 and MFemaleAccess7 = 3.36). However, in progressive year levels, the mean schoolyard attractiveness scores of female students decrease at a faster rate than for males. By Year 10, the means of schoolyard attractiveness between male and female students differ by 16.6 per cent for adaptability (MMaleAdapt10 = 7.77 and MFemaleAdapt10 = 6.48), 23.8 per cent for likeability (MMaleLike10 = 4.46 and MFemaleLike10 = 3.40), and 25.4 per cent for accessibility (MMaleAccess10 = 3.38 and MFemaleAccess10 = 2.52).
## 3.2. Significance of Student Year Level and Gender on Perceptions of Schoolyard Restorative Quality
Student perceptions of schoolyard restoration demonstrate significant declines for students between year levels 7 and 10; results are presented in Table 5. In comparison to the mean scores of Year 10 students, students in Year 7 are associated with more positive ratings of schoolyard restorative quality across all categories, with being away and extent both significant at the 1 per cent level and fascination and compatibility both significant at the 5 per cent level.
Significant differences based on student gender were also demonstrated, with male students positively predicting being away at a 5 per cent level of significance. This relationship is further illustrated in Figure 3. In Year 7, there is a 6.3 per cent difference in means between male and female students for the category ‘Being away’ (MMaleBeingAway7 = 15, MFemaleBeingAway7 = 14.05). By Year 10, this difference expands to 26.3 per cent (MMaleBeingAway10 = 13.46, MFemaleBeingAway10 = 9.92), as the scores of female students decreased at a faster rate than for males.
Although significant relationships were not identified when combining the impact of student gender and year level on perceptions of restorative quality, there are trends worth noting. In Year 7, male and female students rate restorative qualities similarly except for fascination. Year 7 females scored the schoolyard more than one quarter (26.7 per cent) less fascinating (MFemaleFasc7 = 5.27) than male students in the same year (MMaleFasc7 = 7.19). This evens out in Year 8, when male fascination scores fall to female Year 8 levels, a decrease of 32.9 per cent (MMaleFasc8 = 5.27). However, the restorative scores for Year 8 female students in all other categories continue to decline at rates faster than male students. Between Year 8 and Year 10, female students record drops in schoolyard restorative quality of 17.6 per cent for being away (MFemaleBeingAway8 = 12.04 and MFemaleBeingAway10 = 9.92), 21.7 per cent in fascination (MFemaleFasc8 = 5.11 and MFemaleFasc10 = 4.00) and 16.7 per cent drop in extent (MFemaleExt8 = 4.80 and MFemaleExt10 = 4.00). By comparison, male students between years 8 to 10 demonstrated falls in the same categories of 5.4 per cent (MMaleBeingAway8 = 14.23 and MMaleBeingAway10 = 13.46), 6.6 per cent (MMaleFasc8 = 5.27 and MMaleFasc10 = 4.92) and 4 per cent, respectively (MMaleExt8 = 5.45 and MMaleExt10 = 5.23).
## 4. Discussion
This study examined differences in student perceptions of schoolyard attractiveness and restorative quality based on year level and gender. It aimed to identify how the design of secondary schoolyards is perceived by students and understand whether these views are consistent across the school community. Findings show that students in younger year levels predict more favorable perceptions of schoolyard design and restorative quality compared to students in later year levels. Male students predict more positive perceptions of schoolyard likeability, accessibility and adaptability, as well as the restorative quality of ‘being away’ in contrast to female students. While these results are compatible with related research into student schoolyard behaviors, the findings reveal new insights into student perceptions of schoolyard design.
Previous schoolyard studies have often focused on measuring student use of the school grounds. This study was unique in using student opinion to assess the quality of these experiences. This was achieved through survey measures derived from an adult perspective (schoolyard restorativeness) and student-identified priorities (schoolyard attractiveness). These differences in student schoolyard perceptions may help explain differences in student schoolyard use.
A lack of diversity in secondary schoolyard spaces has been identified as a disincentive to motivate student use [5,47]. For most secondary schools, sporting grounds provide the scaffolding around which schoolyard spaces are frequently organized [5]. The site for this study is no different, with the majority (52.7 per cent) of spaces available to students at break times consisting of open grass, sporting fields and ball courts.
Previous research demonstrates that as students age, traditional schoolyard activities lose appeal [22]. Some of this has been attributed to the rapid social, physical and emotional changes associated with adolescence [48]. It has been proposed that the elevation of structured sports programming within secondary schoolyards fails to accommodate these evolving student needs [49]. Changes to body image, self-esteem and friendship groups can make those with less physical skills more reluctant to participate in sports [50].
During adolescence, the ability for teenagers to retreat from others and escape everyday pressures becomes an important part of their psychological development [51,52,53]. The freedom and independence of outdoor spaces are associated with places of calm and emotional restoration by teenagers [54,55]. Previous studies demonstrate that students have the desire to be active during recess, but find it difficult to achieve within existing designs [56], perceiving there are fewer options available to attract and engage those in higher year levels [57].
The student self-report measures in this study complement the findings of this previous research. The mean of Year 7 student schoolyard attractiveness scores are the highest among year levels; by Year 8, student scores begin a decline from which they do not recover. The positive ratings of Year 7 students could reflect the need of younger students to be physically active [58]; thus, they may appreciate schoolyards more [59].
The decline in student restorative quality scores as they age may also be due to a lack of perceived affordances within the existing schoolyard space. Research by Kaplan and Kaplan [1989] promotes that landscape preferences are indicators of places that best support user needs. The decrease in ratings of schoolyard attractiveness as students age may imply that its lack of appeal creates a lack of apparent restorative benefits to be derived from it.
The disparities in male and female perceptions of schoolyard attractiveness in this study raise the question as to whether schoolyard programming is more naturally aligned with the design preferences of male students. Between years 9 and 10 alone, female students recorded a 31.2 per cent decrease in schoolyard likeability (compared to a drop of 6.5 per cent for males) and a 20.1 per cent decline in schoolyard personal connection (compared to a 5.6 per cent decrease for males). If male students are more attracted across year levels to the sport-dominated design of the study site, then this attraction may also help explain the finding of their more positive perceptions of ‘being away’ during school breaks compared to female students.
The differences in male and female attitudes to play are well documented. Boys have traditionally been more physically active and reliant on sporting fields across year levels [47,60,61,62]. By contrast, female students have been documented as relying less on—and even avoiding—areas of turf and asphalt [61,63,64,65,66,67,68], with play more focused on sedentary games within small groups [22]. Their preferences are more aligned with non-competitive play activities that provide greater choice in activities over traditional sports, such as obstacle courses, trampolines, dancing and gymnastics [22,68,69,70]. Such findings suggest that the large land areas devoted to ball sports might actually come at the cost of minimizing female physical activity choices [22].
In contrast to boys, research shows girls spend much less time outdoors [71], while also demonstrating worsening mental health, particularly in terms of psychological distress and life satisfaction [72]. These disparities often extend into adulthood: compared to men, women are three times more likely to experience common mental health problems, particularly in countries considered gender equal [72]. Given the known health benefits of time outdoors, the findings of this study raise whether female students have ‘more to gain’ [73] from improvements to schoolyard spaces than male students.
Research demonstrates that the outdoor preferences of adolescents can be quite variable across ages [55]. Versatile schoolyard spaces that provide diverse, multi-sensory experiences have been shown to provide a greater choice of activities [10] and be more effective in reducing student stress and promoting creativity [38]. Considering the significant differences in student perceptions of schoolyard attractiveness and restorative quality found in this study, increasing the diversity of available schoolyard settings might be one way of increasing the benefits attained by students during outdoor breaks in the school day.
While this study makes contributions in the knowledge around adolescent design preferences and perceived impacts of schoolyards, it also has limitations. As an exploratory study, its limited scope and small sample size may constrain inferences from produced regression models. Given the different types of environments available in the schoolyard (open/paved/sheltered), measuring student perceptions within each would have provided context on which areas best supported user needs. However, due to practicalities of availability of students, time constraints and staff resourcing, this was not possible. It would be beneficial to expand and diversify the sample to a broader subset of schools of different sizes and locations to compare whether findings from this study hold to larger populations. Better monitoring of recess conditions during data collection would help limit confounding factors that may affect results, such as the impact of food during break times, peer effects and varying amounts of time spent by students outdoors prior to undertaking the survey. Given the quantitative nature of the study, results focus on broader patterns of age and gender differences. Qualitative studies that provide students the opportunity to share their perspectives and experiences would complement these findings and explore possible determinants of student schoolyard perceptions and preferences.
## 5. Conclusions
Adolescence is a critical period of social, emotional and physical development. The findings from this study are important for questioning whether existing schoolyard design standards positively support the changing needs of secondary students. Previous studies have addressed the adequacy of schoolyard design by measuring student physical activity levels and observing behavioral patterns. This study is unique by focusing on how students, as users, rate the quality of their schoolyard experiences.
Study findings reveal new insights into differences in schoolyard perceptions between younger and older students, as well as between male and female schoolyard users. Compared to Year 7 students, Year 10 students demonstrated significant declines in ratings across all categories of schoolyard attractiveness and restorative capacity. Female students are more negative than male students in their perceptions of schoolyard accessibility, likeability and personal connection. Male students more positively view the restorative design capacity of the schoolyard to support feelings of ‘being away’. These results suggest that schoolyard standards need revising to provide optimal environments that fulfill the design preferences of females and older students.
The schoolyard is a social landscape. Its design can help students build peer interactions, feel a sense of belonging and find respite during long days of classes. However, to help students thrive, its design needs to consider how issues of materiality, space, form and programming responds to the diverse needs of the broader student body. While the benefits of spending time outside are well known, less understood are the impacts of how its design quality affects user experiences. Exploring new metrics that account for student voice in assessing schoolyard quality can help validate where designs are working well, while providing the tools for calling it out when it is not.
## References
1. Gray P.. **The decline of play and the rise of psychopathology in children and adolescents**. *Am. J. Play.* (2011.0) **3** 443-463
2. Blatchford P., Baines E., Pellegrini A.. **The social context of school playground games: Sex and ethnic differences, and changes over time after entry to junior school**. *Br. J. Dev. Psychol.* (2003.0) **21** 481-505. DOI: 10.1348/026151003322535183
3. Bell A.C., Dyment J.E.. **Grounds for health: The intersection of green school grounds and health-promoting schools**. *Environ. Educ. Res.* (2008.0) **14** 77-90. DOI: 10.1080/13504620701843426
4. Lippman P.C.. *Evidence-Based Design of Elementary and Secondary Schools: A Responsive Approach to Creating Learning Environments* (2010.0)
5. Hyndman B.. **A qualitative investigation of Australian youth perceptions to enhance school physical activity: The Environmental Perceptions Investigation of Children’s Physical Activity (EPIC-PA) study**. *J. Phys. Act. Health* (2016.0) **13** 543-550. DOI: 10.1123/jpah.2015-0165
6. Scott M.M., Cohen D.A., Evenson K.R., Elder J., Catellier D., Ashwood J.S., Overton A.. **Weekend schoolyard accessibility, physical activity, and obesity: The Trial of Activity in Adolescent Girls (TAAG) study**. *Prev. Med.* (2007.0) **44** 398-403. DOI: 10.1016/j.ypmed.2006.12.010
7. Samborski S.. **Biodiverse or barren school grounds: Their effects on children**. *Child. Youth Environ.* (2010.0) **20** 67-115
8. Freeman C., Tranter P.. *Children and Their Urban Environment: Changing Worlds* (2012.0)
9. Francis M.. **Negotiating between children and adult design values in open space projects**. *Des. Stud.* (1988.0) **9** 67-75. DOI: 10.1016/0142-694X(88)90032-4
10. Rutkauskaite R., Gisladottir T., Pihu M., Kjonniksen L., Lounassalo I., Huovinen T., Gruodyte-Raciene R., Visagurskiene K., Olafson O., Kull M.. **Schoolyard Affordances for Physical Activity: A Pilot Study in 6 Nordic–Baltic Countries**. *Sustainability* (2021.0) **13**. DOI: 10.3390/su132111640
11. Hartig T., Mitchell R., De Vries S., Frumkin H.. **Nature and health**. *Annu. Rev. Public Health* (2014.0) **35** 207-228. DOI: 10.1146/annurev-publhealth-032013-182443
12. Hall S., Fildes J., Perrens B., Plummer J., Carlisle E., Cockayne N., Werner-Seidler A.. **Can We Talk?**. *2019.*
13. Bernard M.E., Stephanou A.. **Ecological levels of social and emotional wellbeing of young people**. *Child Indic. Res.* (2018.0) **11** 661-679. DOI: 10.1007/s12187-017-9466-7
14. Buttazzoni A., Dean J., Minaker L.. **Urban design and adolescent mental health: A qualitative examination of adolescent emotional responses to pedestrian- and transit-oriented design and cognitive architecture concepts**. *Health Place* (2022.0) **76** 102825. DOI: 10.1016/j.healthplace.2022.102825
15. Jansson M., Abdulah M., Eriksson A.. **Secondary school students’ perspectives and use of three school grounds of varying size, content and design**. *Urban For. Urban Green.* (2018.0) **30** 115-123. DOI: 10.1016/j.ufug.2018.01.015
16. Anthamatten P., Brink L., Lampe S., Greenwood E., Kingston B., Nigg C.. **An assessment of schoolyard renovation strategies to encourage children’s physical activity**. *Int. J. Behav. Nutr. Phys. Act.* (2011.0) **8** 27. DOI: 10.1186/1479-5868-8-27
17. Pawlowski C.S., Andersen H.B., Troelsen J., Schipperijn J.. **Children’s physical activity behavior during school recess: A pilot study using GPS, accelerometer, participant observation, and go-along interview**. *PLoS ONE* (2016.0) **11**. DOI: 10.1371/journal.pone.0148786
18. Ridgers N.D., Salmon J., Parrish A.-M., Stanley R.M., Okely A.D.. **Physical activity during school recess: A systematic review**. *Am. J. Prev. Med.* (2012.0) **43** 320-328. DOI: 10.1016/j.amepre.2012.05.019
19. Willenberg L.J., Ashbolt R., Holland D., Gibbs L., MacDougall C., Garrard J., Green J.B., Waters E.. **Increasing school playground physical activity: A mixed methods study combining environmental measures and children’s perspectives**. *J. Sci. Med. Sport* (2010.0) **13** 210-216. DOI: 10.1016/j.jsams.2009.02.011
20. Pawlowski C.S., Ergler C., Tjørnhøj-Thomsen T., Schipperijn J., Troelsen J.. **‘Like a soccer camp for boys’ A qualitative exploration of gendered activity patterns in children’s self-organized play during school recess**. *Eur. Phys. Educ. Rev.* (2015.0) **21** 275-291. DOI: 10.1177/1356336X14561533
21. Spark C., Porter L., de Kleyn L.. **‘We’re not very good at soccer’: Gender, space and competence in a Victorian primary school**. *Child. Geogr.* (2019.0) **17** 190-203. DOI: 10.1080/14733285.2018.1479513
22. Raney M.A., Daniel E., Jack N.. **Impact of urban schoolyard play zone diversity and nature-based design features on unstructured recess play behaviors**. *Landsc. Urban Plann.* (2023.0) **230** 104632. DOI: 10.1016/j.landurbplan.2022.104632
23. Bagot K.L., Allen F.C.L., Toukhsati S.. **Perceived restorativeness of children’s school playground environments: Nature, playground features and play period experiences**. *J. Environ. Psychol.* (2015.0) **41** 1-9. DOI: 10.1016/j.jenvp.2014.11.005
24. Samad Z.A., Macmillan S.. **The valuation of intangibles: Explored through primary school design**. *Proceedings of the CIB W096 Architectural Management, Special Meeting on Designing Value: New Directions in Architectural Management* 39-46
25. Rouse J.. **Measuring value or only cost: The need for new valuation methods**. *Designing Better Buildings* (2004.0) 55-71
26. Higgins S., Hall E., Wall K., Woolner P., McCaughey C.. *The Impact of School Environments: A Literature Review* (2005.0)
27. Kreutz A., Derr V., Chawla L.. **Fluid or fixed? Processes that facilitate or constrain a sense of inclusion in participatory schoolyard and park design**. *Landsc. J.* (2018.0) **37** 39-54. DOI: 10.3368/lj.37.1.39
28. Awadh O.. **Sustainability and green building rating systems: LEED, BREEAM, GSAS and Estidama critical analysis**. *J. Build. Eng.* (2017.0) **11** 25-29. DOI: 10.1016/j.jobe.2017.03.010
29. Kaplan R., Kaplan S.. *The Experience of Nature: A Psychological Perspective* (1989.0)
30. Berto R.. **The role of nature in coping with psycho-physiological stress: A literature review on restorativeness**. *Behav. Sci.* (2014.0) **4** 394-409. DOI: 10.3390/bs4040394
31. Abdulkarim D., Nasar J.L.. **Are livable elements also restorative?**. *J. Environ. Psychol.* (2014.0) **38** 29-38. DOI: 10.1016/j.jenvp.2013.12.003
32. Jorgensen A., Gobster P.H.. **Shades of green: Measuring the ecology of Urban green space in the context of human health and well-being**. *Nat. Cult.* (2010.0) **5** 338-363. DOI: 10.3167/nc.2010.050307
33. van den Berg A.E., Hartig T., Staats H.. **Preference for nature in urbanized societies: Stress, restoration, and the pursuit of sustainability**. *J. Soc. Iss.* (2007.0) **63** 79-96. DOI: 10.1111/j.1540-4560.2007.00497.x
34. Van den Berg A.E., Jorgensen A., Wilson E.R.. **Evaluating restoration in urban green spaces: Does setting type make a difference?**. *Landsc. Urban Plann.* (2014.0) **127** 173-181. DOI: 10.1016/j.landurbplan.2014.04.012
35. Velarde M.D.M.D.. **The link between landscape preferences and perceived restorativeness—Current research trends and suggestions for future studies**. *Environmental Psychology: New Developments* (2010.0) 235-242
36. Bedimo-Rung A.L., Mowen A.J., Cohen D.A.. **The significance of parks to physical activity and public health: A conceptual model**. *Am. J. Prev. Med.* (2005.0) **28** 159-168. DOI: 10.1016/j.amepre.2004.10.024
37. Maas J.J.. **Social contacts as a possible mechanism behind the relation between green space and health**. *Health Place* (2009.0) **15** 586-595. DOI: 10.1016/j.healthplace.2008.09.006
38. Chawla L., Keena K., Pevec I., Stanley E.. **Green schoolyards as havens from stress and resources for resilience in childhood and adolescence**. *Health Place* (2014.0) **28** 1-13. DOI: 10.1016/j.healthplace.2014.03.001
39. Yin R.K.. *Case Study Research: Design and Methods* (2009.0) **Volume 5**
40. Van Dijk-Wesselius J., Maas J., Hovinga D., Van Vugt M., Van den Berg A.. **The impact of greening schoolyards on the appreciation, and physical, cognitive and social-emotional well-being of schoolchildren: A prospective intervention study**. *Landsc. Urban Plann.* (2018.0) **180** 15-26. DOI: 10.1016/j.landurbplan.2018.08.003
41. Kathleen L.B., Frances E.K., Felicity C.L.A.. **Amendments to the Perceived Restorative Components Scale for Children (PRCS-C II)**. *Child. Youth Environ.* (2007.0) **17** 124-127
42. Hartig T., Kaiser F.G., Bowler P.A.. *Further Development of a Measure of Perceived Environmental Restorativeness* (1997.0) 18
43. Kaplan S.. **The urban forest as a source of psychological well-being**. *Urban Forest Landscapes: Integrating Multidisciplinary Perspectives* (1995.0) 100
44. Joliffe I.T., Morgan B.. **Principal component analysis and exploratory factor analysis**. *Stat. Methods Med. Res.* (1992.0) **1** 69-95. DOI: 10.1177/096228029200100105
45. Jolliffe I.T.. **Discarding variables in a principal component analysis. I: Artificial data**. *J. Roy. Stat. Soc. Ser. C (Appl. Stat.)* (1972.0) **21** 160-173. DOI: 10.2307/2346488
46. Field A.. *Discovering Statistics Using IBM SPSS Statistics* (2013.0)
47. Haug E., Torsheim T., Sallis J.F., Samdal O.. **The characteristics of the outdoor school environment associated with physical activity**. *Health Educ. Res.* (2008.0) **25** 248-256. DOI: 10.1093/her/cyn050
48. Garcia A.W., Pender N.J., Antonakos C.L., Ronis D.L.. **Changes in physical activity beliefs and behaviors of boys and girls across the transition to junior high school**. *J. Adolesc. Health* (1998.0) **22** 394-402. DOI: 10.1016/S1054-139X(97)00259-0
49. Stanley R.M., Boshoff K., Dollman J.. **Voices in the playground: A qualitative exploration of the barriers and facilitators of lunchtime play**. *J. Sci. Med. Sport* (2012.0) **15** 44-51. DOI: 10.1016/j.jsams.2011.08.002
50. Barbour A.C.. **The impact of playground design on the play behaviors of children with differing levels of physical competence**. *Early Child. Res. Q.* (1999.0) **14** 75-98. DOI: 10.1016/S0885-2006(99)80007-6
51. Owens P.E.. **Natural landscapes, gathering places, and prospect refuges: Characteristics of outdoor places valued by teens**. *Child. Environ. Q.* (1988.0) **5** 17-24
52. Dovey K.. **Refuge and imagination: Places of peace in childhood**. *Child. Environ. Q.* (1990.0) **7** 13-17
53. Sobel D.. *Children’s Special Places: Exploring the Role of Forts, Dens, and Bush Houses in Middle Childhood* (2002.0)
54. Korpela K., Hartig T.. **Restorative qualities of favorite places**. *J. Environ. Psychol.* (1996.0) **16** 221-233. DOI: 10.1006/jevp.1996.0018
55. Korpela K., Kyttä M., Hartig T.. **Restorative experience, self-regulation, and children’s place preferences**. *J. Environ. Psychol.* (2002.0) **22** 387-398. DOI: 10.1006/jevp.2002.0277
56. Högman J., Augustsson C., Hedström P.. **Let’s do those 60 minutes! Children’s perceived landscape for daily physical activity**. *Sport Educ. Soc.* (2020.0) **25** 395-408. DOI: 10.1080/13573322.2019.1610374
57. McEvoy E., MacPhail A., Enright E.. **Physical activity experiences of young people in an area of disadvantage: ‘There’s nothing there for big kids, like us’**. *Sport Educ. Soc.* (2016.0) **21** 1161-1175. DOI: 10.1080/13573322.2014.994176
58. Fjørtoft I.. **Landscape as playscape: The effects of natural environments on children’s play and motor development**. *Child. Youth Environ.* (2004.0) **14** 21-44
59. Jansson M., Gunnarsson A., Mårtensson F., Andersson S.. **Children’s perspectives on vegetation establishment: Implications for school ground greening**. *Urban For. Urban Green.* (2014.0) **13** 166-174. DOI: 10.1016/j.ufug.2013.09.003
60. Pagels P., Raustorp A., De Leon A.P., Mårtensson F., Kylin M., Boldemann C.. **A repeated measurement study investigating the impact of school outdoor environment upon physical activity across ages and seasons in Swedish second, fifth and eighth graders**. *BMC Public Health* (2014.0) **14**. DOI: 10.1186/1471-2458-14-803
61. Raney M.A., Hendry C.F., Yee S.A.. **Physical Activity and Social Behaviors of Urban Children in Green Playgrounds**. *Am. J. Prev. Med.* (2019.0) **56** 522-529. DOI: 10.1016/j.amepre.2018.11.004
62. Andersen H.B., Klinker C.D., Toftager M., Pawlowski C.S., Schipperijn J.. **Objectively measured differences in physical activity in five types of schoolyard area**. *Landsc. Urban Plann.* (2015.0) **134** 83-92. DOI: 10.1016/j.landurbplan.2014.10.005
63. Beighle A., Morgan C.F., Le Masurier G., Pangrazi R.P.. **Children’s physical activity during recess and outside of school**. *J. Sch. Health* (2006.0) **76** 516-520. DOI: 10.1111/j.1746-1561.2006.00151.x
64. Schmidt L.. *Outdoor Spaces–Jungle or Exercise Yard? A Study of Facilities, Children and Physical Activity at School* (2004.0) 179
65. Dyment J.E., Bell A.C., Lucas A.J.. **The relationship between school ground design and intensity of physical activity**. *Child. Geogr.* (2009.0) **7** 261-276. DOI: 10.1080/14733280903024423
66. Roberts S.J., Fairclough S.J., Ridgers N.D., Porteous C.. **An observational assessment of physical activity levels and social behaviour during elementary school recess**. *Health Educ. J.* (2013.0) **72** 254-262. DOI: 10.1177/0017896912439126
67. Paechter C., Clark S.. **Learning gender in primary school playgrounds: Findings from the Tomboy Identities Study**. *Pedagog. Cult. Soc.* (2007.0) **15** 317-331. DOI: 10.1080/14681360701602224
68. Pawlowski C.S., Andersen H.B., Arvidsen J., Schipperijn J.. **Changing recess geographies: Children’s perceptions of a schoolyard renovation project promoting physical activity**. *Child. Geogr.* (2019.0) **17** 664-675. DOI: 10.1080/14733285.2019.1582754
69. Lucas A.J., Dyment J.E.. **Where do children choose to play on the school ground? The influence of green design**. *Education 3-13* (2010.0) **38** 177-189. DOI: 10.1080/03004270903130812
70. Brady F.. **Children’s organized sports: A developmental perspective**. *J. Phys. Educ. Recreat. Danc.* (2004.0) **75** 35-41. DOI: 10.1080/07303084.2004.10608557
71. Larson L.R., Szczytko R., Bowers E.P., Stephens L.E., Stevenson K.T., Floyd M.F.. **Outdoor Time, Screen Time, and Connection to Nature: Troubling Trends Among Rural Youth?**. *Environ. Behav.* (2019.0) **51** 966-991. DOI: 10.1177/0013916518806686
72. Campbell O.L.K., Bann D., Patalay P.. **The gender gap in adolescent mental health: A cross-national investigation of 566,829 adolescents across 73 countries**. *SSM—Popul. Health* (2021.0) **13** 100742. DOI: 10.1016/j.ssmph.2021.100742
73. Sillman D., Rigolon A., Browning M.H., McAnirlin O.. **Do sex and gender modify the association between green space and physical health? A systematic review**. *Environ. Res.* (2022.0) **209** 112869. DOI: 10.1016/j.envres.2022.112869
|
---
title: Relationship between Behavioural Intention for Using Food Mobile Applications
and Obesity and Overweight among Adolescent Girls
authors:
- Rajaa A. Alyami
- Manal F. Alharbi
journal: International Journal of Environmental Research and Public Health
year: 2023
pmcid: PMC10002290
doi: 10.3390/ijerph20054432
license: CC BY 4.0
---
# Relationship between Behavioural Intention for Using Food Mobile Applications and Obesity and Overweight among Adolescent Girls
## Abstract
Changes in the body mass index (BMI) of children and adolescents have been linked to mobile usage, particularly food applications. This study aimed to investigate the relationship between food application usage and obesity and overweight among adolescent girls. This cross-sectional study was conducted among adolescent girls aged 16–18 years. Data were collected using a self-administered questionnaire from female high schools in five different regional offices across Riyadh City. The questionnaire included questions regarding demographic data (age and academic level), BMI and behavioural intention (BI) scale comprising three constructs: attitude towards behaviour, subjective norms and perceived behavioural control. Of the included 385 adolescent girls, $36.1\%$ were 17 years old, and $71.4\%$ had normal BMI. The overall mean BI scale score was 65.4 (SD 9.95). No significant differences were observed between overweight or obesity in relation to the overall BI score and its constructs. A high BI score was more associated with participants studying in the east educational office than those who were enrolled in the central educational office. Behavioural intention to use food applications greatly influenced the adolescent age group. Further investigations are necessary to determine the influence of food application services among individuals with high BMI.
## 1. Introduction
Childhood obesity is one of the most challenging global health issues of the 21st century [1]. It has been established that body weight during childhood directly impacts an individual’s lifelong health [2]. According to the World Obesity Atlas 2022, published by the World Obesity Federation, one billion people will be obese by 2030. This includes one in five women and one in seven men [3]. Globally in 2016, over 340 million children and adolescents aged up to 19 years were overweight or obese [4]. Adolescents, a group comprising individuals who are 10–19 years of age, are of interest as they undergo various physical, sexual, psychological and social developmental changes [5]. Contrary to popular belief that individuals in this age group are often healthy, adolescents face several health issues and are at high risk of being overweight and obese until their adulthood and developing noncommunicable diseases (NCDs) like diabetes and cardiovascular diseases at a young age [4,5]. The body mass index (BMI) is used to diagnose childhood overweight and obesity. Overweight is defined as having a BMI at or above the 85th percentile and below the 95th percentile for children and adolescents of the same age and sex, whereas obesity is defined as having a BMI at or above the 95th percentile for children and teens of the same age and sex [6]. Increasing levels of overweight and obesity pose some risks to the health and well-being of children and adolescents [7]. Technology has led to an increase in overweight and obesity [8]. As a type of technology used to deliver food online, food apps may pose serious health risks. Therefore, prevention of weight gain among children and adolescents requires taking food apps into consideration as risk factors. Further research is required to understand these implications.
## 1.1. Literature Review
Obesity is a major problem in the USA, with its prevalence being higher in adolescents than in children of other age groups. According to the Center for Disease Control and Prevention (CDC), the prevalence of obesity is $13.9\%$ among 2–5-year-olds, $18.4\%$ among 6–11-year-olds and $20.6\%$ among 12–19-year-olds [9]. It is known that being overweight and obese status are caused by a variety of factors, including eating patterns, lack of sleep or physical activity, certain medications, and genetic factors. [ 10]. A systematic review revealed the association between friendship networks and obesity-related behaviours among adolescents [11]. A study in Iraq revealed that overweight and obesity were higher among females than among male adolescents [12]. In Saudi Arabia, the overall prevalence figures of overweight and obesity were $13.4\%$ and $18.2\%$, respectively, and when compared with the WHO-based national prevalence rate of obesity reported in 2004 (≈$9.3\%$), the obesity rate has doubled in 10 years [13]. Also, the estimated prevalence rates of overweight and obesity among school-aged children were $19.6\%$ and $7.9\%$, respectively, and high rates were reported for adolescents ($26.6\%$ and $10.6\%$ for overweight and obesity, respectively) [13]. Many studies have identified overweight and obesity as health issues in Saudi Arabia [14,15,16,17] and observed a direct relationship between obesity and several factors, such as skipping breakfast, excessive calories, consumption patterns, and parental socioeconomic factors [16,17,18,19].
There are many factors that may influence the development of obesity, such as lack of physical activity and digital devices [20]. Digital device use has been associated with many evidence-based benefits, including early learning, exposure to many ideas and knowledge, enhanced social interactions and support and increased access to health promotion information and messages [21]. However, the use of such digital devices can compromise sleep, attention, and learning; increase the incidence of obesity and depression; and expose people to inaccurate, insensitive, or unsafe information [21]. In contrast, a recent study suggests that digital devices can empower parents who are concerned about their children’s overweight and obesity [22]. Studies have found an increased incidence of high BMI in children and adolescents who use digital devices [23,24,25,26,27], including ordering food online. Also, the use of online food ordering systems is increasing [21]. Currently, more than 1.2 billion people use online food delivery systems (OFDs) worldwide. By 2027, the total number of people using platform-to-consumer delivery systems will reach 1517.4 million [28]. Within the past four years, food orders placed through direct restaurant apps or third-party services have increased by $23\%$ in the USA [29]. The convenience of food applications can lead to adverse health outcomes. In addition to being easy to use, these apps provide access to menus and offers available at local restaurants. This demonstrates how technology has a significant impact on lifestyle and wellness. For instance, in the U.S., there is a lack of research to support how digital food ordering affects health and wellness from an individual or public health perspective [29], and also, studies that investigate if the users of food applications are overweight or obese or among overweight or obese adolescents. Additionally, an early report reported that the obesity prevalence is expected to rise from $12\%$ in 1992 to $41\%$ by 2022 for men and from $21\%$ to $78\%$ for women, according to World Health Organization projections for 1992 to 2022 in Saudi Arabia [30]. This highlights the critical gaps in the literature regarding the investigation of the relationship between food apps and obesity among adolescence. More research is needed to explain the underlying mechanisms and provide effective prevention strategies [20].
Considering food apps as risk factors in both age groups helps in applying preventive measures and decreasing weight gain among children and adolescents. In order to fully understand the potential health implications of online food delivery, further research is needed. The current research on the impact of digital use on lifelong health is lacking. As a result, and because of this percentage among female adolescents, it is imperative to predict and recognise target-oriented behaviour, take preventive measures, and raise awareness among young women about their health. Therefore, this study aimed to assess the intended behaviour of using food apps by applying the theory of planned behaviour (TPB).
## 1.2. The Conceptual Framework
The theory of planned behaviour is one of the social psychology theories that are widely used in health promotion activities. TPB can offer a reasonable explanation of the decision-making processes underlying both the intention for and engagement in self-care overweight/obesity-reducing behaviours [31]. It believes that people are rational and their decisions are based on the knowledge available to them. TPB explains and understands environmental and individual factors that influence behaviour. The important determinant of a person’s behaviour is their intention [31]. Three interrelated concepts are described in TBP, and these concepts can serve as factors that define the level of behavioural intention (BI) [32]. First, attitude towards the behaviour (ATB) refers to the individual’s positive or negative evaluation of performing the behaviour [32]. Attitude is perceived as a combination of feelings, beliefs, intentions and perceptions. Second, subjective norm (SN) is viewed as the social pressure upon a person to behave in a certain way [32]. Lastly, perceived behavioural control (PBC) is related to the perceived influence of factors on the behaviour; therefore, it may enhance or hamper certain behaviours [32]. Thus, the intention should be placed at the core of an individual’s behaviour to act upon the three concepts [32]. Accordingly, positive ATB and SN lead to high perceived control PBC and a strong individual intention to perform a positive behaviour [32].
For the current study, using food apps, the attitude towards an action or behaviour predicts that participants might believe that using an app is more convenient. SN focus on the surroundings of the individual, such as family, friends, beliefs, habits or social media advertising that probably influence their decision. The PBC means that adolescents respond to using apps as an easy way to fulfil their diet needs. Because the theory helps predict a positive or negative attitude towards an action, if the three concepts are positive or two of them are, the intention is increased. Therefore, adolescents will be more likely to have the behaviour of using food apps. See Figure 1.
Behavioural intention for using food apps may influence BMI. Attitude towards the behaviour, sociocultural factors and barriers associated with facilitators of adolescent behaviour influence behavioural intention to food app user behaviour. Therefore, recognising the food app intention behaviour as a factor that leads to overweight and obesity will be useful in developing prevention measures. Thus, this study aimed to investigate the influence of food app intention behaviour on adolescent girls in Saudi Arabia using the theory of planned behaviour (TPB). In addition, the current study hypothesised that the attitude towards the behaviour, subjective norm, and perceived behavioural control predict behavioural intention.
## 2.1. Design and Participants
This study adopted a quantitative descriptive design. This design was selected to test the hypothesis of the study. It is a deductive approach where concepts of obesity, overweight and food apps are downed to variables, and their relationship is tested. When an evidence-based conclusion is drawn, generalisations can be extended to a larger population. Given the cross-sectional design of our study, the variables and relationships among them were determined [33]. Data were collected from female high schools in five different educational regions across Riyadh City. This study included female students aged 16–18 years. The required sample size was determined as 383, which was determined on the basis of the estimated population size of 86,704 obtained from the Ministry of Education database that was last updated in June 2016 [34]. Calculations were made using Epinfo version 7.2.3, with a confidence level of $95\%$, a margin of error of $5\%$, an expected frequency of $50\%$ and a design effect of 1.0 in five clusters [35]. This sample size was increased by approximately $15\%$ [415] to compensate for any absenteeism, dropouts or incomplete questionnaire. Thus, the returned sample was 390 after excluding five incomplete questionnaires.
## 2.2. Data Collection
The study was conducted from January to March 2021. Quantitative data were collected by distributing a self-report questionnaire and performing physiologic measurements of the student’s weight and height to evaluate their BMI. In order to ensure the most desirable representative sample, probability sampling was conducted, which entails both clustering and stratified sampling techniques. The sample was divided into five clusters that are distributed based on five regions of Riyadh City. Each regional cluster involved one randomly selected high school. Moreover, stratified sampling was applied to the selection of classrooms. Since female high schools were accessible, all female students were grouped by level (first, second, and third). From each level, one class was selected randomly. All students in the selected classes were recruited [33]. Approximately 79 students in each regional sample cluster were enrolled. Data collection followed specific ethical protocols that involved an explanation of the purpose of the study to the participants and the distribution of questionnaires by researchers, ensuring voluntary participation and an agreement of participation secured. The questionnaires were handled confidentially, and all collected data were manually verified. BMI was measured using a formula after taking the participants’ weight and height. The CDC recommends BMI categorisation for children and teens between ages 2 and 20 years. Therefore, the BMI-for-age percentile growth charts were used. The CDC BMI categorisation for children and teens between the ages of 2 and 20 is as follows: underweight, <$5\%$; healthy weight, $5\%$–$85\%$; at risk of overweight, $85\%$–$95\%$ and overweight, >$95\%$ [36].
## 2.3. Measurements
A 24-item questionnaire that consists of close-ended questions was developed by the researchers. The self-administered questionnaire consisted of four parts. Each part assessed a certain variable. Part one assessed the sociodemographic characteristics of the adolescent. Part two assessed ATB with ten items (i.e., adolescent attitudes towards the use of food apps). Part three assessed SN with four items (i.e., adolescent sociocultural factors). Finally, part four assessed PBC with six items (i.e., barriers and facilitators of adolescent behaviours). In order to measure these variables, a 5-point Likert scale was used for all parts, with positively worded statements and various response options. These options, as a form of frequency that ranges from always to never, an agreement that ranges from strongly disagree to strongly agree and a level that ranges from very high to very low. Positive statements are scored 1–5. Each score item was reported individually [33].
## 2.4. Validity and Reliability
It is necessary to measure the validity and reliability of the research established scale. Content validity evaluates whether questions cover all aspects of the study and whether irrelevant questions will be removed [33]. An empirical method of testing content validity involves techniques to calculate the content validity index (CVI). The CVI was determined by experts ($$n = 7$$) from the field who evaluated each item of the questionnaire. The items were evaluated for their clarity, relatedness, representativeness and appropriateness, and their instructions were suited for the target group. Each item uses a four-point scale (from 1 as not relevant to 4 as very relevant) to determine whether the item is to be approved or rejected [37]. The rating scales were described on the item level (I-CVI) and scale level (S-CVI). I-CVI was measured by I-CVI = (agreed item)/(number of experts). S-CVI was determined by which the scoring average of the I-CVI for all items on the scale (S-CVI/Ave) and the item’s proportion on the scale that scored a scale of 3 or 4 by all experts (S-CVI/UA) [38,39]. In order to test the reliability of the established scale, the scale requires examination of the stability and internal consistency as the best and oldest technique used for consistency. Coefficient alpha interpretation constitutes the normal range of values that is between $0.00\%$ and $1.00\%$, and higher values indicate a greater internal consistency [33]. The established scale demonstrated properties of reliability for the adolescent age group. The internal consistency calculated using Cronbach’s alpha was found to be good (α = 0.84) for 20 items of the entire scale (BI), subscales ATB (α = 0.80) and PBC (α = 0.71). Other categories showed acceptable levels for the subscale SN (α = 0.66). Additionally, the average inter-item correlation calculated for determining the appropriate internal consistency reliability was good (0.42) and showed a positive and had a good item level, ranging between 0.20 and 0.57. The ideal range for item level is considered to be 0.15–0.50, and values over 0.2 are considered acceptable [40,41].
## 2.5. Statistical Analysis
The IBM SPSS for Windows, version 26.0 (IBM Corp., Armonk, NY, USA), was used to analyse all data. Descriptive statistics were presented using numbers, percentages and mean ± standard deviation. Sociodemographic characteristics and BI were compared using the Kruskal–Wallis test, whereas differences in the score of BI and its constructs according to the BMI level were analysed using the Mann–Whitney U-test. Furthermore, Spearman’s correlation coefficient was used to determine the correlation between the BI scale and its constructs. The normality test was conducted using the Shapiro–Wilk test and Kolmogorov–Smirnov test. A $p \leq 0.05$ was taken as significant.
## 3. Results
A total of 385 young girls completed the survey, and five incomplete questionnaires were excluded. As shown in Table 1, the most common age was 17 years ($36.1\%$). Girls enrolled from the north educational office constituted $20.5\%$, of which $36.6\%$ of them are currently in the second year. Additionally, participants with normal BMI were predominant ($71.4\%$). Table 2 shows the mean score of the BI constructs, which are composed of the ATB, SN and PBC. Regarding the ATB, the mean score was highest in the statement ‘I like to use the food app because it is easy and convenient’ (mean score, 3.76), followed by ‘Food apps give various food choices than home meals’ (mean score, 3.67) and ‘Meals that are ordered through food apps are more attractive than home meals’ (mean score, 3.56), whereas it was lowest in the statement ‘I used to order through food apps almost daily’ (mean score, 2.45). The overall mean score was 31.6 (SD 5.82). For the SN, the mean rating was the highest for the statement ‘Food apps advertisement is everywhere’ (mean score, 4.34), whereas it was lowest in the statement ‘Food apps were recommended by my friends’ (mean score, 3.12). The overall mean score for SN was 14.8 (SD 2.49). Finally, for the PBC, the mean score was highest for the statement ‘*It is* easy to order food through food apps because I have a device’ (mean score, 3.97), followed by ‘*It is* easy to order food through food apps because I have internet access all the time’, whereas it was the lowest in the statement ‘*It is* easy to order food through food apps because my mother is busy and cannot cook home meals’ (mean score, 2.30). The overall mean score of the PBC was 18.9 (SD 4.25) and the mean total BI score was 65.4 (SD 9.95). In Table 3, a positive and highly significant correlation was found between BI scores among its constructs, including ATB ($r = 0.870$), SN ($r = 0.682$) and PBC ($r = 0.750$). In addition, we noted a positive and highly significant correlation between attitude towards behaviour according to SN ($r = 0.469$) and PBC ($r = 0.394$). Finally, a positive and highly significant correlation was observed between SN and PBC ($r = 0.369$). In Table 4, no significant correlation was found between the BMI level and the total BI score, including its constructs such as ATB, SN and PBC ($p \leq 0.05$). In Table 5, a higher BI score was more associated with the east educational office, but it was the lowest in the central educational office ($H = 28.813$; $p \leq 0.001$). No significant differences were found in BI according to the age group ($$p \leq 0.099$$) and academic year level ($$p \leq 0.274$$). In Table 6, the post-hoc analysis indicates that the mean differences in BI scores were significant between the south educational office and the east educational office ($$p \leq 0.015$$). Moreover, we found significant differences between the central educational office and the north educational office ($$p \leq 0.002$$), west educational office ($$p \leq 0.007$$) and east educational office ($p \leq 0.001$).
## 4. Discussion
This study investigated the relationship between food apps among adolescent girls who are overweight and obese. To our knowledge, this is the first study in Saudi Arabia-Riyadh city that tested the influence of ordering meals through digital apps on the weight levels of female high school adolescents. We employed the TPB as a tool for measuring behaviour intentions in using digital food apps. The results of this study revealed that the overall BI score has a mean of 65.4 (SD 9.95). Regarding its constructs, the ATB mean score was 31.6, the SN mean score was 14.8, and the PBC mean score was 18.9. The overall scores of the BI and its constructs were above the average of the mean points, suggesting the high behavioural intention to use food apps among the participants. The continuous innovation of the digital world is reflected even in food consumption, and the increased behaviour of utilising food apps was evidently seen in our results. This study adds to the existing discussion on consumer behaviour in the context of digital food delivery in Saudi Arabia and uncovers the elements that could be used to predict people’s motivation to buy food through food delivery apps.
A positive and highly significant correlation was found between the overall BI score and its constructs, suggesting that the increase in the score of each ATB, SN and PBC construct correlated with the increase in the overall BI score. For instance, increasing adolescents’ attitudes, SN or PBC to ordering food through food apps correlated with an increase in overall behavioural intention. Consistent with our findings, Choyhirun et al. [ 2008] found that attitudes, SN and PBC explained up to $41.8\%$ of the variance in intentions [42]. The intentions were influenced most by PBC and then by attitudes and SN. The positive trend of behavioural intentions in using food applications among our youth may lead to overconsumption, which may result in unhealthy food consumption. This scenario may be in accordance with that of Lwin et al. [ 2017] as well as Andrews, Silk and Eneli [2010] [43,44]. According to their reports, the enhanced attitude of children towards eating healthy food is directly influenced by the guidance of parents, decreasing the intention to eat unhealthy food. However, parental mediation of TV advertising negatively affected healthy food attitudes to a greater extent [43,44]. Hence, parental guidance to children is imperative when ordering food through apps to avoid overconsumption of food and an unhealthy lifestyle.
Our results suggest that although participants who were overweight/obese had a slightly higher attitude in using food apps, adolescents with normal/underweight BMI had slightly higher SN, PBC and behavioural intentions; however, this did not yield significance ($p \leq 0.05$). In a study conducted in Thailand among 112 overweight ($$n = 52$$) and obese ($$n = 56$$) young adults, the overall mean TBP score increased significantly from baseline in the health dieting behaviour and SN following group counselling, concluding that group counselling was not inferior to individual counselling and that group counselling is a better option for healthy dieting management [45]. In Greece, between the normal-weight group and overweight/obese group, correlations between variables of TPB and behaviours (healthy eating and exercise) were higher in the normal-weight group than in the overweight/obese group, whereas attitude was a significant predictor for those with higher values in the normal-weight group [46]. On the contrary, several studies have reported a decrease in BMI after educational interventions. For example, Jejhooni et al. [ 2022] reported that before the educational intervention, no significant difference was found in the behavioural intention between the experimental group and control group; after six months of the training intervention, a significant increase was found in each of the TPB constructs, weight and BMI among the intervention group whereas the control group did not differ significantly after the educational intervention [47]. This has been concluded by Sanaeinasab et al. [ 2020], Mazloomy-Mahmoodabad et al. [ 2017] and Soorgi, Miri and Sharifzadeh [2015], revealing significant changes after educational intervention in behavioural intentions and BMI levels specifically towards the experimental group [48,49,50].
No significant differences in BI were found in relation to age and academic year level ($p \leq 0.05$). These findings are similar to those of Jeihooni et al. [ 2022] [47]. No significant difference was found in their study in TBP constructs at baseline between the experimental and control groups in terms of age and education. The study by Alfadda and Masood [2019] correlated overweight and obesity with high levels of parental socioeconomic status and urbanisation in Saudi society [18]. Although caution may be warranted, further investigations should be conducted to determine the effect of behavioural intentions on the sociodemographic data of the overweight and obese populations. One unanticipated finding was the difference between BI, and educational region office location was significant ($p \leq 0.001$). The east educational office was associated with a higher BI score, whereas the central educational office was associated with a lower BI score. A possible explanation for these results may be the lack of adequate data on the financial and employment status of the participants and their parents or living in a location that has a variety of restaurants, which causes a high intrinsic motivation to order using food application. A limitation of this study is that descriptive studies are not helpful in understanding the causes of the phenomena, as the survey method limits the ability to identify causes. Additionally, there is a lack of information regarding the use of food apps by boys in high schools. Despite these limitations, this study can help in testing a newly established scale. Although the scale revealed good reliability and validity characteristics, further explorative and validation studies should be considered to test the newly established scale. Further research is recommended to establish the influence of food app services among individuals with increased BMI levels. It would be a fruitful endeavour for the current findings to be repeated in different contexts for a better understanding of the factors that influence the intended behaviour of food app use in a larger age group. Moreover, this study can serve as a reference guide and baseline for research investigating such topics in the future.
## 5. Conclusions
This study set out to investigate the influence of food app intention behaviour on adolescent girls in Saudi Arabia using the theory of planned behaviour (TPB). This study has found that, generally, the behavioural intention to use food apps greatly influences our sample population. There was a high behavioural intention among the participants to use food apps, as indicated by the overall scores of the BI and its constructs. This was above the average of the mean points. As the digital world develops, so does food consumption, and our results demonstrate that more people are taking advantage of food apps to consume food. Through food delivery apps, we explore the elements that could be used to predict people’s motivations to buy food in the context of digital food delivery in Saudi Arabia. The results indicate that although adolescents with overweight/obese BMI had a slightly higher attitude towards using food apps, adolescents with normal/underweight BMI had slightly higher SN, PBC, and behavioural intentions. Further, evidence suggests that ATB, SN, PBC and overall BI were not directly related to the increased weight of the young population. However, the increase in the overall behavioural intention to use food apps could be associated with participants enrolled in the east educational office but less among those enrolled in the central educational office.
## References
1. **Childhood Overweight and Obesity**
2. Barton M.. **Childhood obesity: A life-long health risk**. *Acta Pharmacol. Sin.* (2012.0) **33** 189-193. DOI: 10.1038/aps.2011.204
3. **Globally Estimated to Be Living with Obesity by 2030**
4. **Obesity and Overweight**
5. **Adolesc Health**
6. **Defining Childhood Obesity**
7. Tabarés Seisdedos R.. **Health effects of overweight and obesity in 195 countries over 25 years**. *N. Engl. J. Med.* (2017.0) **377** 13-27. PMID: 28604169
8. Ali S.A., Al-Fayyadh H.R., Mohammed S.H., Ahmed S.R.. **A Descriptive Statistical Analysis of Overweight and Obesity Using Big Data**. *Proceedings of the 2022 International Congress on HORA* 1-6
9. **Childhood Obesity Facts**. (2020.0)
10. **What Causes Obesity & Overweight?**. (2021.0)
11. Alsayed N., Eldabi T., Lee H.. **Investigating obesity related behaviours in friendship networks among the youth: A systemized review**. *KnE Soc. Sci.* (2019.0) 109-127. DOI: 10.18502/kss.v3i25.5194
12. Hassan Farman W.M., M Majeed H.. **Overweight and obesity among children and adolescents in Baladrouz city**. *Diyala J. Med.* (2019.0) **16** 31-38. DOI: 10.26505/DJM.16024391108
13. Al-Hussaini A., Bashir M.S., Khormi M., AlTuraiki M., Alkhamis W., Alrajhi M., Halal T.. **Overweight and obesity among Saudi children and adolescents: Where do we stand today?**. *Saudi J. Gastroenterol.* (2019.0) **25** 229-235. DOI: 10.4103/sjg.SJG_617_18
14. Al-Agha A.E., Mabkhoot Y.M., Bahwirith A.S., Mohammed A.N., Ragbi R., Allhabi E., Dumyati B.K., Milyani A.. **Various causative factors and associated complications of childhood obesity in Jeddah, Western Region, Saudi Arabia**. *Ann. Afr. Med.* (2020.0) **19** 15-19. DOI: 10.4103/aam.aam_8_19
15. Al-Daghri N.M., Aljohani N.J., Al-Attas O.S., Al-Saleh Y., Alnaami A.M., Sabico S., Amer O.E., Alharbi M., Kumar S., Alokail M.S.. **Comparisons in childhood obesity and cardiometabolic risk factors among urban Saudi Arab adolescents in 2008 and 2013**. *Child Care Health Dev.* (2016.0) **42** 652-657. DOI: 10.1111/cch.12361
16. Alyousef S.M., Alhamidi S.A.. **Exploring the contributory factors toward childhood obesity and being overweight in Saudi Arabia and the Gulf States**. *J. Transcult. Nurs.* (2020.0) **31** 360-368. DOI: 10.1177/1043659619868286
17. Al Yazeedi B.A., Berry D.C.. **Childhood overweight and obesity is increasing in Gulf Cooperation Council Countries: A review of the literature**. *J. Transcult. Nurs.* (2019.0) **30** 603-615. DOI: 10.1177/1043659619829528
18. Alfadda A.A., Masood A.. **Now trending in Riyadh: Increased urbanization equals to more obesity in children and adolescents**. *Saudi J. Gastroenterol.* (2019.0) **25** 207-209. DOI: 10.4103/sjg.SJG_347_19
19. Alhilabi H.S., Payne A.. **The impact of skipping breakfast on the body weight of children and young people in Saudi Arabia; a systematic review**. *Arab. J. Nutr. Exerc. AJNE* (2018.0) **3** 67. DOI: 10.18502/ajne.v3i3.3588
20. Lehmann F., Varnaccia G., Zeiher J., Lange C., Jordan S.. **Influencing factors of obesity in school-age children and adolescents—A systematic review of the literature in the context of obesity monitoring**. *J. Health Monit.* (2020.0) **5** 2-23
21. Reid Chassiakos Y.L., Radesky J., Christakis D., Moreno M.A., Cross C., Hill D., Ameenuddin N., Hutchinson J., Levine A.. **Children and Adolescents and Digital Media**. *Pediatrics* (2016.0) **138** e20162593. DOI: 10.1542/peds.2016-2593
22. Fidjeland T.G., Øen K.G.. **Parents’ Experiences Using Digital Health Technologies in Paediatric Overweight and Obesity Support: An Integrative Review**. *Int. J. Environ. Res. Public Health* (2023.0) **20**. DOI: 10.3390/ijerph20010410
23. Alturki H.A., Brookes D.S., Davies P.S.. **Does spending more time on electronic screen devices determine the weight outcomes in obese and normal weight Saudi Arabian children?**. *Saudi Med. J.* (2020.0) **41** 79-87. DOI: 10.15537/smj.2020.1.24786
24. Haghjoo P., Siri G., Soleimani E., Farhangi M.A., Alesaeidi S.. **Screen time increases overweight and obesity risk among adolescents: A systematic review and dose-response meta-analysis**. *BMC Prim. Care* (2022.0) **23**. DOI: 10.1186/s12875-022-01761-4
25. Robinson T.N., Banda J.A., Hale L., Lu A.S., Fleming-Milici F., Calvert S.L., Wartella E.. **Screen media exposure and obesity in children and adolescents**. *Pediatrics* (2017.0) **140** S97-S101. DOI: 10.1542/peds.2016-1758K
26. Rocka A., Jasielska F., Madras D., Krawiec P., Pac-Kożuchowska E.. **The impact of digital screen time on dietary habits and physical activity in children and adolescents**. *Nutrients* (2022.0) **14**. DOI: 10.3390/nu14142985
27. Tripathi M., Mishra S.K.. **Screen time and adiposity among children and adolescents: A systematic review**. *J. Public Health* (2020.0) **28** 227-244. DOI: 10.1007/s10389-019-01043-x
28. **Statista Platform Delivery–Worldwide**
29. Stephens J., Miller H., Militello L.. **Food delivery apps and the negative health impacts for Americans**. *Front. Nutr.* (2020.0) **7** 14. DOI: 10.3389/fnut.2020.00014
30. Al-Quwaidhi A.J., Pearce M.S., Critchley J.A., Sobngwi E., O’flaherty M.. **Trends and future projections of the prevalence of adult obesity in Saudi Arabia, 1992–2022**. *East. Mediterr. Health J.* (2014.0) **20** 10. DOI: 10.26719/2014.20.10.589
31. Ho R.T.K., Sun X.Y.. **Overweight/Obesity-related attitudes and self-care behaviours: Evaluation and comparison of the protection motivation model and theory of planned behaviour**. *J. Pac. Rim Psychol.* (2016.0) **10**. DOI: 10.1017/prp.2015.5
32. Wills E.M.. **Grand nursing theories based on interactive process**. *Theor. Basis Nurs.* (2014.0) 159-191
33. Polit D.F., Beck C.T.. *Essentials of Nursing Research: Appraising Evidence for Nursing Practice* (2018.0)
34. **Document Center**
35. **Epinfo, 7.2.3 version**
36. **Defining Childhood Weight Status**. (2022.0)
37. Kyngäs H., Mikkonen K., Kääriäinen M.. *The Application of Content Analysis in Nursing Science Research* (2020.0)
38. Yusoff M.S.B.. **ABC of content validation and content validity index calculation**. *Educ. Med. J.* (2019.0) **11** 49-54. DOI: 10.21315/eimj2019.11.2.6
39. Del Greco L., Walop W., McCarthy R.H.. **Questionnaire development: 2. validity and reliability**. *CMAJ Can. Med. Assoc. J.* (1987.0) **136** 699-700. PMID: 3828925
40. Glen S.. **2018 June Average Inter-Item Correlation: Definition, Example. From StatisticsHowTo. Com: Elementary Statistics for the Rest of Us**
41. Shultz K.S., Whitney D.J., Zickar M.J.. *Measurement Theory in Action: Case Studies and Exercises* (2020.0)
42. Choyhirun T., Suchaxaya P., Chontawan R., Kantawang S.. **Predictors of eating behaviors for weight control among overweight early adolescents**. *Pac. Rim Int. J. Nurs. Res.* (2008.0) **12** 107-120
43. Lwin M.O., Shin W., Yee A.Z.H., Wardoyo R.J.. **A parental health education model of children’s food consumption: Influence on children’s attitudes, intention, and consumption of healthy and unhealthy foods**. *J. Health Commun.* (2017.0) **22** 403-412. DOI: 10.1080/10810730.2017.1302523
44. Andrews K.R., Silk K.S., Eneli I.U.. **Parents as Health Promoters: A Theory of Planned Behavior Perspective on the Prevention of Childhood Obesity**. *J. Health Commun.* (2010.0) **15** 95-107. DOI: 10.1080/10810730903460567
45. Phimarn W., Paktipat P., Pansiri K., Klabklang P., DuangJanchot P., Tongkul A.. **Effect of weight control counseling in overweight and obese young adults**. *Indian J. Pharm. Sci.* (2017.0) **79** 35-41. DOI: 10.4172/pharmaceutical-sciences.1000198
46. Psouni S., Hassandra M., Theodorakis Y.. **Exercise and healthy eating intentions and behaviors among normal weight and overweight/obese adults**. *Psychology* (2016.0) **07** 598-611. DOI: 10.4236/psych.2016.74062
47. Khani Jeihooni A., Layeghiasl M., Yari A., Rakhshani T.. **the effect of educational intervention based on the theory of planned behavior on improving physical and nutrition status of obese and overweight women**. *BMC Womens Health* (2022.0) **22**. DOI: 10.1186/s12905-022-01593-5
48. Sanaeinasab H., Saffari M., Dashtaki M.A., Pakpour A.H., Karimi Zarchi A., O’Garo K.N., Koenig H.G.. **A theory of planned behavior-based program to increase physical activity in overweight/obese military personnel: A randomised controlled trial**. *Appl. Psychol. Health Well Being* (2020.0) **12** 101-124. DOI: 10.1111/aphw.12175
49. Mazloomy-Mahmoodabad S.S., Navabi Z.S., Ahmadi A., Askarishahi M.. **The effect of educational intervention on weight loss in adolescents with overweight and obesity: Application of the theory of planned behavior**. *ARYA Atheroscler.* (2017.0) **13** 176-183. PMID: 29147128
50. Soorgi Z., Miri M., Sharifzadeh G.. **The impact of planned behavior theory-based education on female students’ obesity-related behaviors**. *Mod. Care J.* (2015.0) **12** e8672. DOI: 10.17795/modernc.8672
|
---
title: 'Galectin-3 and Blood Group: Binding Properties, Effects on Plasma Levels,
and Consequences for Prognostic Performance'
authors:
- Carolin Pozder
- Elles M. Screever
- A. Rogier van der Velde
- Herman H. Silljé
- Janne Suwijn
- Saskia de Rond
- Marcus E. Kleber
- Graciela Delgado
- Jan Jacob Schuringa
- Wiek H. van Gilst
- Wouter C. Meijers
- Winfried März
- Rudolf A. de Boer
journal: International Journal of Molecular Sciences
year: 2023
pmcid: PMC10002292
doi: 10.3390/ijms24054415
license: CC BY 4.0
---
# Galectin-3 and Blood Group: Binding Properties, Effects on Plasma Levels, and Consequences for Prognostic Performance
## Abstract
Previous studies have reported an association between ABO type blood group and cardiovascular (CV) events and outcomes. The precise mechanisms underpinning this striking observation remain unknown, although differences in von Willebrand factor (VWF) plasma levels have been proposed as an explanation. Recently, galectin-3 was identified as an endogenous ligand of VWF and red blood cells (RBCs) and, therefore, we aimed to explore the role of galectin-3 in different blood groups. Two in vitro assays were used to assess the binding capacity of galectin-3 to RBCs and VWF in different blood groups. Additionally, plasma levels of galectin-3 were measured in different blood groups in the Ludwigshafen Risk and Cardiovascular Health (LURIC) study (2571 patients hospitalized for coronary angiography) and validated in a community-based cohort of the Prevention of Renal and Vascular End-stage Disease (PREVEND) study (3552 participants). To determine the prognostic value of galectin-3 in different blood groups, logistic regression and cox regression models were used with all-cause mortality as the primary outcome. First, we demonstrated that galectin-3 has a higher binding capacity for RBCs and VWF in non-O blood groups, compared to blood group O. Additionally, LURIC patients with non-O blood groups had substantially lower plasma levels of galectin-3 (15.0, 14.9, and 14.0 μg/L in blood groups A, B, and AB, respectively, compared to 17.1 μg/L in blood group O, $p \leq 0.0001$). Finally, the independent prognostic value of galectin-3 for all-cause mortality showed a non-significant trend towards higher mortality in non-O blood groups. Although plasma galectin-3 levels are lower in non-O blood groups, the prognostic value of galectin-3 is also present in subjects with a non-O blood group. We conclude that physical interaction between galectin-3 and blood group epitopes may modulate galectin-3, which may affect its performance as a biomarker and its biological activity.
## 1. Introduction
Cardiovascular (CV) diseases account for $32\%$ of global deaths, and the prognosis of patients with CV disease, particularly in patients with heart failure, remains poor; it is, therefore, important to further investigate disease characteristics and identify risk factors that can serve as therapeutic targets [1,2]. Besides the classical risk factors of heart failure such as hypertension, smoking, dyslipidaemia, obesity, and diabetes mellitus, a sedentary habit, excessive alcohol intake, influenza, certain microbes, cardiotoxic drugs, chest radiation, and coronary artery disease also have to be considered [3,4]. However, the residual risk remains high, and we do not fully understand all factors contributing to CV disease development.
In the past years, the ABO blood group has been identified as a novel and intriguing risk factor for CV disease. Multiple studies have shown an association between non-O blood groups and the risk of different thromboembolic events [5], coronary heart disease [6], the size of a myocardial infarction after an acute coronary syndrome [7], increased mortality in patients with ischemic heart disease [8], and venous thrombosis [9]. The exact mechanisms behind these associations remain unclear to date, but as a possible common mechanism, variable levels and activity of the von Willebrand Factor (VWF) have been proposed. VWF is widely acknowledged as a key determinant in CV homeostasis and has been linked to thrombosis and CV events [10,11]. VWF was also found to be a binding partner of galectin-3 [12].
Galectin-3 is a carbohydrate-binding protein and has been shown to be involved in inflammation, cancer, and CV disease [13,14,15,16,17]. It was shown that galectin-3 is able to modulate VWF-mediated thrombus formation via a direct (physical) interaction with VWF [12]. A possible link between galectin-3 and blood group has been described previously—a genome-wide association study showed that the ABO gene locus was strongly associated with plasma galectin-3 levels [18]. This ABO locus appears to be a very pleiotropic locus that associates with several CV traits [19]. Building upon those findings, we hypothesized that ABO, galectin-3, and VWF would interact and, specifically, that the described associations between galectin-3 and CV outcome [20] can, at least partially, be explained by an interaction with the ABO blood group and VWF levels.
## 2.1. Study Population
The baseline characteristics of patients in the LURIC study are presented in Table 1. The mean age (SD) was 63 [10] years, and the majority of the population was male ($68\%$). Out of the population, 946 ($37\%$) of the patients had blood group O, 1219 ($47\%$) blood group A, 276 ($11\%$) blood group B, and 130 ($5\%$) blood group AB. Additionally, 495 ($19\%$) of the patients were smokers, and a medical history of hypertension and coronary artery disease were very common ($73\%$ and $77\%$, respectively). To validate our findings, we studied the relationship between galectin-3 and the blood group in a community-based cohort, for which we used the PREVEND study. Participants of the PREVEND cohort were younger (mean age 50 ± 12), and sex was equally distributed ($51\%$ male versus $49\%$ female). 1557 ($44\%$) had blood group O, 1606 ($45\%$) blood group A, 271 ($8\%$) blood group B, and 118 ($3\%$) blood group AB. Smoking was common ($46\%$), but hypertension was less abundant compared to the LURIC cohort ($30\%$), as expected (Supplemental Table S1).
## 2.2. Galectin-3 Plasma Levels Stratified by Blood Group
The LURIC cohort was stratified by blood group. Plasma levels of galectin-3 were significantly higher in blood group O compared to other blood groups ($p \leq 0.0001$ for all groups versus blood group O) (Table 1, Figure 1A). Furthermore, VWF levels were significantly lower in blood group O compared to other blood groups (Table 1). In the PREVEND cohort, galectin-3 levels were also significantly different among blood groups and showed the highest values in blood group O compared to other blood groups (Figure 1B, Supplemental Table S1). Moreover, subjects with homozygous blood groups showed a trend towards lower plasma levels of galectin-3 compared to subjects with heterozygous blood groups (Supplemental Figure S1).
## 2.3. Binding of Galectin-3 and Red Blood Cells
Galectin-3 is known to mediate the hemagglutination of red blood cells (RBCs). To further characterize a potential interaction between galectin-3, VWF, and blood group, two different in vitro assays were performed. The first assay was a hemagglutination assay, to examine the interaction between galectin-3 and the blood group, as displayed in Supplemental Figure S2. With this assay we showed that the binding of galectin-3 with RBCs was significantly different between blood groups, with RBCs from blood group O binding less galectin-3 compared to all other blood groups (Figure 2A,B).
## 2.4. Interaction between VWF and Galectin-3
Since galectin-3 has been presented as a partner for VWF, we assessed the binding of VWF with galectin-3 in different blood groups using a VWF-galectin-3 binding assay. Blood plasma with similar levels of VWF (as determined with ELISA) was equalized to similar concentrations with $0.9\%$ NaCl and incubated in a plate coated with galectin-3. Using VWF antibodies, the galectin-3-VWF binding was detected. This assay showed that the binding for galectin-3 to VWF was stronger in all non-O blood groups compared to blood group O (Figure 2C).
## 2.5. Prognostic Value of Galectin-3
We studied the prognostic value of galectin-3 in different blood groups in the LURIC study cohort. During a median follow-up time of 9.8 [8.6–10.4] years, 758 deaths ($29\%$) were observed. Using Cox regression analyses, galectin-3 remained a significant predictor for all-cause mortality, even after multivariate adjustment (HR 1.89 [1.28–2.79] and HR 2.19 [1.67–2.86] in blood group O and blood group non-O, respectively) (Table 2). The HR is higher in non-O blood groups, although galectin-3 plasma levels were lower in these patients (Figure 3A).
We also assessed the prognostic value of galectin-3 among different blood groups in the general population after adjustment for the same variables. In the PREVEND study, the median follow-up time was 12.6 [12.3–12.9] years, and 353 subjects ($10\%$) died during this period. The same trend was observed compared to the LURIC study: galectin-3 appeared to have a higher prognostic value regarding all-cause mortality in non-O blood groups (Figure 3B), although the p for interaction was non-significant. The blood group itself was not an independent predictor for outcome in both the LURIC and PREVEND study cohorts (Supplemental Table S2).
## 3. Discussion
We demonstrate that circulating galectin-3 levels in subjects with non-O blood groups are significantly lower compared to levels in subjects with blood group O. However, the prognostic value of galectin-3 is stronger in subjects with non-O blood groups. As a potential mechanism, we propose that VWF may mediate this, as circulating VWF and galectin-3 were inversely related. We demonstrate that galectin-3 binds stronger to RBCs and VWF of subjects with non-O blood groups compared to subjects with blood group O.
Accumulating evidence suggests that the ABO blood group is involved in the pathogenesis of CV disease and that non-O blood groups had the highest risk of CV disease [19,21]. Previous studies have shown that the presence of non-O blood groups is associated with worse outcomes compared to blood group O [21,22,23,24]. In a recent case-control study consisting of 165 centenarians and 5063 blood donors from the same geographical region, it was observed that among centenarians the prevalence of blood group O was higher ($56.4\%$ vs. $43.5\%$; $$p \leq 0.001$$) [25]. Besides studies that demonstrate a higher CV risk for non-O blood groups, there are a few studies that specifically found the highest risk, for blood group A and blood group AB [6,21,26,27,28,29]. For example, one recent Finnish study found the highest risk of ischaemic heart disease in a patient with blood group A with T1DM and microalbuminuria [27]. A Canadian study ($$n = 64$$,686) demonstrated that blood group AB is associated with an increased risk of thrombotic events in participants from Quebec [28].
The ABO(H) blood group is the most important blood group system and is determined by complex carbohydrate moieties at the extracellular surface of the RBC membrane [30]. The A and B alleles encode for either A- or B-glycosyltransferases that add N-acetylgalactosamine or D-galactose to the common H-glycan precursor backbone, respectively. In subjects with blood group O, no A- or B-transferase activity is present, resulting in the expression of the H-glycan backbone without an additional group [31]. Next to the expression of RBCs, these blood group epitopes and different antigens are also expressed on other cells, such as the vascular endothelium, epithelial cells, T-cells, B-cells, and platelets, and present on molecules such as VWF [32,33].
Several studies described the major effects of the ABO blood group on plasma levels of VWF: plasma VWF levels appear to be $25\%$ lower in the O blood group compared to non-O blood groups [6]. This implies that subjects with blood group O may experience a higher incidence of bleeding events, while subjects with non-O blood groups experience a higher incidence of thrombotic events [34,35]. The exact mechanisms underpinning these observations remain unclear, but this effect may be mediated by VWF. The effect of the ABO blood group on plasma levels of VWF seems to be the result of a direct effect of the ABO blood group [36]. The conversion of the blood group O determinant into other antigens of the ABO blood group was correlated with an increased capacity to modify the N-linked glycosylation of VWF [37]. Therefore, changes in VWF glycan composition also affect the biological activity of VWF and are not restricted to its plasma levels [38]. Carbohydrate structures on the surface of VWF play an important role in the life cycle of VWF. Galectin-3 is a carbohydrate-binding protein and has recently been identified as a new partner of VWF [12]. Furthermore, the affinity of transmembrane glycoproteins to the galectin-3 molecule is proportional to the number and branching of their N-glycans [39]. Therefore, we hypothesize that the biological activity of galectin-3 might also be directly regulated by the glycosylation of the molecule by the ABO blood group.
In agreement with previous studies, we confirmed that plasma VWF levels are ~$25\%$ higher in non-O blood groups. Additionally, we now show in two independent cohorts with different populations, that galectin-3 levels are significantly lower in non-O blood groups. Furthermore, we show that galectin-3 levels are lower in patients who had a heterozygous blood group. This inverse relationship between galectin-3 and VWF levels in different blood groups is an interesting phenomenon, potentially explained by the fact they are ligands of each other.
Numerous studies have assessed the prognostic value of galectin-3 in various cohorts [40,41,42]. We again corroborated these findings in the current study and herein confirm that galectin-3 is an independent predictor for all-cause mortality, particularly in subjects with non-O blood groups. The striking observation that galectin-3 has a strong prognostic value in non-O blood groups, although the group has lower galectin-3 values, should be explored in further detail. We speculate that the observed lower galectin-3 plasma values in the non-O blood group participants are caused by galectin-3 binding with blood group epitopes and that glycosylation might play a role in this.
In two different in vitro assays, we show a higher binding capacity of galectin-3 with RBCs and VWF in subjects with non-O blood groups, compared to blood group O. Binding preference of galectin-3 is most likely related to the extensive glycosylation of VWF, generating a clustered glycan surface, resembling the cell membrane [12]. These protein-glycan interactions between VWF and galectin-3 mainly consist of binding patterns with N-linked glycans rather than O-linked glycans, as has been shown previously [43]. Galectins regularly show a high affinity for glycans with longer poly-N-acetyllactosamine (poly-LacNAc) chains, given their higher binding capacity for N-linked glycans.
The higher hemagglutination activity in subjects with non-O blood groups is consistent with previous findings from erythrocyte binding and glycan microarray studies, suggesting that galectin-3 exhibits higher binding towards blood group A and B antigens compared to those bearing the H antigen [43,44,45,46]. While all galectins show a high affinity for β-galactosides, their recognition following terminal glycan modifications varies. The enhanced recognition of galectin-3 towards A and B blood group substitutions is potentially caused by unique subsides within the carbohydrate recognition domain (CRD) [43] and might play an evolutionary role. In fact, it enables the targeting of microbes that utilize blood group molecular mimicry [47]. Additionally, we hypothesize that stronger binding of galectin-3 with RBCs and VWF in non-O blood groups could explain lower levels of circulating galectin-3.
The prognostic value and absolute levels of biomarkers may differ between different subgroups in a study cohort, as previously observed for other biomarkers [48]. For instance, plasma levels differ between sexes, and also age, renal function, and the presence of diabetes are important determinants of hemoglobin level [49,50]. Even for the established cardiac marker NT-proBNP, important determinants exist leading to differences in circulating levels; renal failure tends to increase natriuretic peptide levels, whereas patients with obesity show lower levels of NT-proBNP [51,52]. Using a combination of biomarkers might improve risk prediction of clinical outcomes and, therefore, healthcare-related costs.
In conclusion, we postulate that the binding of galectin-3 to the A-, B-, and AB- blood group epitopes affects the circulating plasma levels and its biological activity, and thereby also its prognostic power for a given concentration. Future studies should provide more detailed data on this interaction and practical information on how to deal with this potential confounder.
## 4.1.1. LURIC
The Ludwigshafen Risk and Cardiovascular Health (LURIC) study consists of 3316 patients who were hospitalized for coronary angiography between 1997 and 2000. Indications for coronary angiography were chest pain or a positive non-invasive stress test suggestive of myocardial ischemia. Further methods and results have been described previously [53]. In total, galectin-3 values and blood group information were available for 2571 patients.
## 4.1.2. PREVEND
The Prevention of Renal and Vascular End-stage Disease (PREVEND) study is a prospective, observational, community-based study and was used to validate our findings [18,54]. The PREVEND study enrolled community-dwelling subjects during 1997–1998, and the study was designed to track the long-term development of cardiac, renal, and peripheral vascular disease. More details of the design of the study have been described previously [55,56]. Galectin-3 and blood group data were available in 3552 subjects.
In both studies, all participants provided informed consent, and the study procedures were conducted in accordance with the 1975 Declaration of Helsinki. The LURIC study was approved by the ethical committee of the Ärztekammer Rheinland-Pfalz, and the PREVEND study was approved by the ethical committee of the University Medical Center Groningen (UMCG).
## 4.2. Galectin-3 Measurements
In the LURIC study, galectin-3 levels were measured in plasma samples from the baseline. These samples were stored at −80 °C and were analysed using the ARCHITECT analyser (Abbott Diagnostics, Abbott Park, IL, USA). This automated assay uses the same antibodies and conjugates as in the manual assay and has a lower limit of detection of 1.01 ng/mL. Intra- and inter-assay variability are $3.2\%$ and $0.8\%$, respectively [57]. In the PREVEND study, blood was drawn at the baseline and anticoagulated with EDTA. Samples were stored at −80 °C until the time of analysis. Galectin-3 concentration was measured in plasma samples from the baseline using the BGM galectin-3 ELISA kit (BG Medicine Inc., Waltham, MA, USA). Intra- and inter-assay coefficients of this assay are $3.2\%$ and $5.6\%$, respectively. The assay has a lower limit of detection of 1.13 ng/mL and did not show cross-reactivity with collagens or other members of the galectin family [58].
## 4.3. Blood Group Determination
Blood group in LURIC was determined in the Haemostaseology Laboratory of the Ludwigshafen Cardiac Centre using a blood group antisera macroscopic agglutination assay (ABO- and Rh-blood group sera, Loxo GmbH, Dossenheim, Germany). In the PREVEND cohort, the ABO blood group was inferred from genotyping three single nucleotide polymorphisms (SNPs) on the ABO gene, namely rs8176719, rs8176746, and rs8176747. Using a combination of these SNPs, a blood group could be determined, as described previously [59].
## 4.4. Clinical Endpoints
In LURIC, mortality data were collected from local registries. Two independent and experienced clinicians, who were blinded for patient characteristics, reviewed information from death certificates, medical records from hospitals, and data from autopsies [20,60]. In PREVEND, mortality data were collected using the municipal register, and cause of death was obtained using the Prismant health care data system or Dutch Central Bureau of Statistics. Follow-up times ranged from the last follow-up or were censored on the date of the event or last contact, whatever occurred first.
## 4.5.1. Isolation of Red Blood Cells
Neonatal cord blood was obtained from healthy full-term pregnancies from donors from the obstetrics departments of the Martini Hospital Groningen and UMCG after informed consent was given. All donors were informed about the studies that were performed, as approved by the local Medical Ethical Committee of the UMCG. Furthermore, healthy volunteers from the research lab also provided blood specimens. Blood was collected in 10 mL EDTA tubes and 20 µL of blood was used to determine the ABO blood group using a Serafol ABO bedside test (Bio-Rad Laboratories BV, Veenendaal, the Netherlands). The remaining blood was centrifuged at 3500 rpm for 5 min. The buffy coat appeared as a dense white layer in the middle between the RBCs and plasma. Plasma and the buffy coat were removed from the tube. RBCs remained in the tube and were resuspended in PBS and again centrifuged at 2000 rpm for 5 min at 4 °C. This washing step was repeated 3 times. Subsequently, the remaining RBCs were diluted 12.5× in PBS-$3\%$ glutaraldehyde in a tube, and this was put on a rotating wheel for 1 h at room temperature. Afterwards, the cells were washed 5 times with PBS ($0.0025\%$ NaN3) and centrifuged at 2000 rpm for 2 min at 4 °C, and in the last step, cells were resuspended at 3–$4\%$ in PBS ($0.0025\%$ NaN3). Cells were stored at 4 °C for several days.
## 4.5.2. Hemagglutination Assay
RBCs were counted using a Fuchs-Rosenthal counting chamber. All cells were diluted to the lowest concentration of RBCs. We first calibrated our hemagglutination assay to determine the number of RBCs that were needed to show hemagglutination and to clearly distinguish between agglutinated and non-agglutinated cells. We tested 3 different concentrations of RBCs (5 µL/10 µL/15 µL of 2000 cells/µL) and 2 concentrations of galectin-3 (1 µM/2 µM). Following calibration, we used 15 µL RBCs/2 µM galectin-3 in the first well of a round-bottom, 96-well plate (Costar #3799, Corning Inc., Kennebunk, ME, USA). Next, 2 µM galectin-3 was serially diluted 1:1 into the next wells and 87,5 µL PBS was added to a total volume of 185 µL. Finally, 15 µL (2000 cells/µL) of RBCs were added to each well. The plate was incubated for 30 min at 4 °C and pictures were made using the ImageQuant LAS 4000 (GE Healthcare, Europe GmbH, Diegem, Belgium). Hemagglutination was assessed using ImageJ software (Version 1.50, National Institutes of Health, Bethesda, MD, USA), and the hemagglutination-index ((surface area of RBCs after incubation/surface area of the total well) × 100) (HA-index) was calculated.
## 4.5.3. Von Willebrand Factor ELISA
VWF was measured in human plasma using the VWF ELISA kit (Abcam, Cambridge, UK). This kit was designed for the quantitative measurement of human VWF in plasma, serum, and cell culture supernatants. Intra- and inter-assay coefficients of variation of this assay are $5\%$ and $7.1\%$, respectively. The lower level of detection is 2.5 mU/mL.
In LURIC, VWF was measured using the STA Liatest®VWF assay (Stago Diagnostica/Roche, Mannheim, Germany).
## 4.5.4. Galectin-3—von Willebrand Factor Binding Study
As previously described [12], an immunosorbent assay was performed in which a microtiter 96-well plate was coated with galectin-3 (5 µg/well) overnight at 4 °C. After washing 3 times with PBS ($0.1\%$ Tween-20) the plate was blocked for 2 h with PBS ($0.1\%$ Tween-20/$3\%$ BSA) at 37 °C. After washing 2 times with PBS ($0.1\%$ Tween-20), plasma of different blood groups was incubated in the wells for 1 h at 37 °C. After discarding the plasma, the plate was washed 2 times with PBS ($0.1\%$ Tween-20). Bound VWF was detected by adding 50 µL HRP-labelled polyclonal VWF antibody (1:1000; P0226, DAKO, Glostrup, Denmark). 50 µL 3,3′,5,5′-tetramethylbenzidine (TMB) was added to detect HRP activity, and after 10 min 50 µL of stop solution (H2SO4) was added to stop the reaction. The absorbance was measured using a microplate reader at a wavelength of 450 nm (BioTek Synergy H1, Winooski, VT, USA).
## 4.6. Statistical Analysis
Normally distributed variables are presented as means ± standard deviation (SD) or standard error of the mean (SEM). Non-normally distributed variables are expressed as medians [interquartile range (IQR)]. To compare normally distributed values across two groups, a two-sample t-test was performed, and to compare non-normally distributed values, we used the Wilcoxon rank-sum test. The comparison of categorical values was done using Pearson’s Chi-square test. Characteristics across four groups were compared using the ANOVA for continuous and normally distributed values and the Kruskal-Wallis test for continuous, non-normally distributed values. In a comparison of >1 group with a control group, we used the Kruskal-Wallis with a post hoc Dunn’s multiple comparisons tests.
Prior to analysis, galectin-3 was transformed logarithmically to obtain approximately normal distributions because of a skewed distribution as assessed by the Shapiro-Wilk test. To study the association of galectin-3 with all-cause mortality, Cox regression analysis and logistic regression analysis were performed with log-transformed galectin-3 as a continuous variable. The model was adjusted for age and sex and a multivariable model consisting of eGFR, smoking, systolic blood pressure, BMI, LDL-cholesterol, diabetes mellitus, lipid-lowering therapy, triglycerides, and CRP. This model is an established risk model for all-cause mortality in the LURIC study and has been used previously in other studies [20]. Results are stratified to blood group and summarized as hazard ratios, with $95\%$ confidence intervals (CI). For the interaction term, a p-value of <0.10 was considered to indicate statistical significance. For all other analyses, p-values <0.05 were considered to be statistically significant. Analyses were performed using STATA software version 14.2 and GraphPad Prism version 9.3.1 (GraphPad Software Inc., La Jolla, CA, USA).
## References
1. **Cardiovascular Diseases (CVDs)**
2. Gerber Y., Weston S.A., Redfield M.M., Chamberlain A.M., Manemann S.M., Jiang R., Killian J.M., Roger V.L.. **A contemporary appraisal of the heart failure epidemic in Olmsted County, Minnesota, 2000 to 2010**. *JAMA Intern. Med.* (2015.0) **175** 996-1004. DOI: 10.1001/jamainternmed.2015.0924
3. Vasan R.S., Sullivan L.M., Wilson P.W., Sempos C.T., Sundström J., Kannel W.B., Levy D., D’Agostino R.B.. **Relative importance of borderline and elevated levels of coronary heart disease risk factors**. *Ann. Intern. Med.* (2005.0) **142** 393-402. DOI: 10.7326/0003-4819-142-6-200503150-00005
4. McDonagh T.A., Metra M., Adamo M., Gardner R.S., Baumbach A., Böhm M., Burri H., Butler J., Čelutkienė J., Chioncel O.. **2021 ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure**. *Eur. Heart J.* (2021.0) **42** 3599-3726. DOI: 10.1093/eurheartj/ehab368
5. Groot H.E., Sierra L.E.V., Said M.A., Lipsic E., Karper J.C., van der Harst P.. **Genetically determined ABO blood group and its associations with health and disease**. *Arterioscler. Thromb. Vasc. Biol.* (2020.0) **40** 830-838. DOI: 10.1161/ATVBAHA.119.313658
6. He M., Wolpin B., Rexrode K., Manson J.E., Rimm E., Hu F.B., Qi L.. **ABO blood group and risk of coronary heart disease in two prospective cohort studies**. *Arterioscler. Thromb. Vasc. Biol.* (2012.0) **32** 2314-2320. DOI: 10.1161/ATVBAHA.112.248757
7. Ketch T.R., Turner S.J., Sacrinty M.T., Lingle K.C., Applegate R.J., Kutcher M.A., Sane D.C.. **ABO blood types: Influence on infarct size, procedural characteristics and prognosis**. *Thromb. Res.* (2008.0) **123** 200-205. DOI: 10.1016/j.thromres.2008.02.003
8. Carpeggiani C., Coceani M., Landi P., Michelassi C., L’abbate A.. **ABO blood group alleles: A risk factor for coronary artery disease. An angiographic study**. *Atherosclerosis* (2010.0) **211** 461-466. DOI: 10.1016/j.atherosclerosis.2010.03.012
9. Franchini M., Favaloro E.J., Targher G., Lippi G.. **ABO blood group, hypercoagulability, and cardiovascular and cancer risk**. *Crit. Rev. Clin. Lab. Sci.* (2012.0) **49** 137-149. DOI: 10.3109/10408363.2012.708647
10. Vischer U.M.. **von Willebrand factor, endothelial dysfunction, and cardiovascular disease**. *J. Thromb. Haemost.* (2006.0) **4** 1186-1193. DOI: 10.1111/j.1538-7836.2006.01949.x
11. Spiel A.O., Gilbert J.C., Jilma B.. **von Willebrand factor in cardiovascular disease: Focus on acute coronary syndromes**. *Circulation* (2008.0) **117** 1449-1459. DOI: 10.1161/CIRCULATIONAHA.107.722827
12. Saint-Lu N., Oortwijn B.D., Pegon J.N., Odouard S., Christophe O.D., de Groot P.G., Denis C.V., Lenting P.J.. **Identification of galectin-1 and galectin-3 as novel partners for von Willebrand factor**. *Arterioscler. Thromb. Vasc. Biol.* (2012.0) **32** 894-901. DOI: 10.1161/ATVBAHA.111.240309
13. Martínez-Martínez E., Calvier L., Fernández-Celis A., Rousseau E., Jurado-López R., Rossoni L.V., Jaisser F., Zannad F., Rossignol P., Cachofeiro V.. **Galectin-3 Blockade Inhibits Cardiac Inflammation and Fibrosis in Experimental Hyperaldosteronism and Hypertension**. *Hypertension* (2015.0) **66** 767-775. DOI: 10.1161/HYPERTENSIONAHA.115.05876
14. Thijssen V.L., Heusschen R., Caers J., Griffioen A.W.. **Galectin expression in cancer diagnosis and prognosis: A systematic review**. *Biochim. Biophys. Acta* (2015.0) **1855** 235-247. DOI: 10.1016/j.bbcan.2015.03.003
15. De Boer R.A., Voors A.A., Muntendam P., Van Gilst W.H., Van Veldhuisen D.J.. **Galectin-3: A novel mediator of heart failure development and progression**. *Eur. J. Heart Fail.* (2009.0) **11** 811-817. DOI: 10.1093/eurjhf/hfp097
16. Gehlken C., Suthahar N., Meijers W.C., de Boer R.A.. **Galectin-3 in Heart Failure: An Update of the Last 3 Years**. *Heart Fail. Clin.* (2018.0) **14** 75-92. DOI: 10.1016/j.hfc.2017.08.009
17. Meijers W.C., Maglione M., Bakker S.J.L., Oberhuber R., Kieneker L.M., de Jong S., Haubner B.J., Nagengast W.B., Lyon A.R., van der Vegt B.. **Heart failure stimulates tumor growth by circulating factors**. *Circulation* (2018.0) **138** 678-691. DOI: 10.1161/CIRCULATIONAHA.117.030816
18. De Boer R.A., van Veldhuisen D.J., Gansevoort R.T., Muller Kobold A.C., van Gilst W.H., Hillege H.L., Bakker S.J.L., van der Harst P.. **The fibrosis marker galectin-3 and outcome in the general population**. *J. Intern. Med.* (2012.0) **272** 55-64. DOI: 10.1111/j.1365-2796.2011.02476.x
19. Franchini M., Lippi G.. **The intriguing relationship between the ABO blood group, cardiovascular disease, and cancer**. *BMC Med.* (2015.0) **13**. DOI: 10.1186/s12916-014-0250-y
20. Drechsler C., Delgado G., Wanner C., Blouin K., Pilz S., Tomaschitz A., Kleber M.E., Dressel A., Willmes C., Krane V.. **Galectin-3, Renal Function, and Clinical Outcomes: Results from the LURIC and 4D Studies**. *J. Am. Soc. Nephrol.* (2015.0) **26** 2213-2221. DOI: 10.1681/ASN.2014010093
21. Gotsman I., Keren A., Zwas D.R., Lotan C., Admon D.. **Clinical Impact of ABO and Rhesus D Blood Type Groups in Patients with Chronic Heart Failure**. *Am. J. Cardiol.* (2018.0) **122** 413-419. DOI: 10.1016/j.amjcard.2018.04.018
22. Franchini M., Capra F., Targher G., Montagnana M., Lippi G.. **Relationship between ABO blood group and von Willebrand factor levels: From biology to clinical implications**. *Thromb. J.* (2017.0) **5** 14. DOI: 10.1186/1477-9560-5-14
23. Wolpin B.M., Chan A.T., Hartge P., Chanock S.J., Kraft P., Hunter D.J., Giovannucci E.L., Fuchs C.S.. **ABO blood group and the risk of pancreatic cancer**. *J. Natl. Cancer Inst.* (2009.0) **101** 424-431. DOI: 10.1093/jnci/djp020
24. Etemadi A., Kamangar F., Islami F., Poustchi H., Pourshams A., Brennan P., Boffetta P., Malekzadeh R., Dawsey S.M., Abnet C.C.. **Mortality and cancer in relation to ABO blood group phenotypes in the Golestan Cohort Study**. *BMC Med.* (2015.0) **13**. DOI: 10.1186/s12916-014-0237-8
25. Franchini M., Mengoli C., Bonfanti C., Rossi C., Lippi G.. **Genetic determinants of extreme longevity: The role of ABO blood group**. *Thromb. Haemost.* (2015.0) **115** 458-460. PMID: 26334169
26. Chen Z., Yang S.H., Xu H., Li J.J.. **ABO blood group system and the coronary artery disease: An updated systematic review and meta-analysis**. *Sci. Rep.* (2016.0) **6** 23250. DOI: 10.1038/srep23250
27. Parente E.B., Harjutsalo V., Lehto M., Forsblom C., Sandholm N., Groop P.H.. **Relationship between ABO blood groups and cardiovascular disease in type 1 diabetes according to diabetic nephropathy status**. *Cardiovasc. Diabetol.* (2020.0) **19** 68. DOI: 10.1186/s12933-020-01038-z
28. Blais C., Germain M., Delage G., Grégoire Y.. **The association between blood group and the risk of vascular disease in Quebec blood donors**. *Blood Transfus.* (2016.0) **14** 455-459. PMID: 27177404
29. Biswas S., Ghoshal P.K., Halder B., Mandal N.. **Distribution of ABO blood group and major cardiovascular risk factors with coronary heart disease**. *Biomed Res. Int.* (2013.0) **2013** 782941. DOI: 10.1155/2013/782941
30. Yamamoto F., Clausen H., White T., Marken J., Hakomori S.. **Molecular genetic basis of the histo-blood group ABO system**. *Nature* (1990.0) **345** 229-233. DOI: 10.1038/345229a0
31. Lowe J.B.. **The blood group-specific human glycosyltransferases**. *Baillieres. Clin. Haematol.* (1993.0) **6** 465-492. DOI: 10.1016/S0950-3536(05)80155-6
32. Franchini M., Liumbruno G.M.. **ABO blood group: Old dogma, new perspectives**. *Clin. Chem. Lab. Med.* (2013.0) **51** 1545-1553. DOI: 10.1515/cclm-2013-0168
33. Ewald D.R., Sumner S.C.J.. **Blood type biochemistry and human disease**. *Wiley Interdiscip. Rev. Syst. Biol. Med.* (2016.0) **8** 517-535. DOI: 10.1002/wsbm.1355
34. Reddy V.M., Daniel M., Bright E., Broad S.R., Moir A.A.. **Is there an association between blood group O and epistaxis?**. *J. Laryngol. Otol.* (2008.0) **122** 366-368. DOI: 10.1017/S0022215107008560
35. Zhang H., Mooney C.J., Reilly M.P.. **ABO blood groups and cardiovascular diseases**. *Int. J. Vasc. Med.* (2012.0) **2012** 641917. DOI: 10.1155/2012/641917
36. Souto J.C., Almasy L., Muñiz-Diaz E., Soria J.M., Borrell M., Bayén L., Mateo J., Madoz P., Stone W., Blangero J.. **Functional effects of the ABO locus polymorphism on plasma levels of von Willebrand factor, factor VIII, and activated partial thromboplastin time**. *Arterioscler. Thromb. Vasc. Biol.* (2000.0) **20** 2024-2028. DOI: 10.1161/01.ATV.20.8.2024
37. Matsui T., Titani K., Mizuochi T.. **Structures of the asparagine-linked oligosaccharide chains of human von Willebrand factor. Occurrence of blood group A, B, and H(O) structures**. *J. Biol. Chem.* (1992.0) **267** 8723-8731. DOI: 10.1016/S0021-9258(19)50338-6
38. Sarode R., Goldstein J., Sussman I.I., Nagel R.L., Tsai H.M.. **Role of A and B blood group antigens in the expression of adhesive activity of von Willebrand factor**. *Br. J. Haematol.* (2000.0) **109** 857-864. DOI: 10.1046/j.1365-2141.2000.02113.x
39. Nabi I.R., Shankar J., Dennis J.W.. **The galectin lattice at a glance**. *J. Cell Sci.* (2015.0) **128** 2213-2219. DOI: 10.1242/jcs.151159
40. Chen A., Hou W., Zhang Y., Chen Y., He B.. **Prognostic value of serum galectin-3 in patients with heart failure: A meta-analysis**. *Int. J. Cardiol.* (2015.0) **182** 168-170. DOI: 10.1016/j.ijcard.2014.12.137
41. Meijers W.C., Januzzi J.L., Defilippi C., Adourian A.S., Shah S.J., van Veldhuisen D.J., de Boer R.A.. **Elevated plasma galectin-3 is associated with near-term rehospitalization in heart failure: A pooled analysis of 3 clinical trials**. *Am. Heart J.* (2014.0) **167** 853-860.e4. DOI: 10.1016/j.ahj.2014.02.011
42. Meijers W.C., de Boer R.A., van Veldhuisen D.J., Jaarsma T., Hillege H.L., Maisel A.S., Di Somma S., Voors A.A., Peacock W.F.. **Biomarkers and low risk in heart failure. Data from COACH and TRIUMPH**. *Eur. J. Heart Fail.* (2015.0) **17** 1271-1282. DOI: 10.1002/ejhf.407
43. Stowell S.R., Arthur C.M., Mehta P., Slanina K.A., Blixt O., Leffler H., Smith D.F., Cummings R.D.. **Galectin-1, -2, and -3 exhibit differential recognition of sialylated glycans and blood group antigens**. *J. Biol. Chem.* (2008.0) **283** 10109-10123. DOI: 10.1074/jbc.M709545200
44. Feizi T., Solomon J.C., Yuen C.T., Jeng K.C., Frigeri L.G., Hsu D.K., Liu F.T.. **The adhesive specificity of the soluble human lectin, IgE-binding protein, toward lipid-linked oligosaccharides. Presence of the blood group A, B, B-like, and H monosaccharides confers a binding activity to tetrasaccharide (lacto-N-tetraose and lacto-N-ne**. *Biochemistry* (1994.0) **33** 6342-6349. DOI: 10.1021/bi00186a038
45. Hirabayashi J., Hashidate T., Arata Y., Nishi N., Nakamura T., Hirashima M., Urashima T., Oka T., Futai M., Muller W.E.. **Oligosaccharide specificity of galectins: A search by frontal affinity chromatography**. *Biochim. Biophys. Acta.* (2002.0) **1572** 232-254. DOI: 10.1016/S0304-4165(02)00311-2
46. Wu S.-C., Ho A.D., Kamili N.A., Wang J., Murdock K.L., Cummings R.D., Arthur C.M., Stowell S.R.. **Full-length galectin-3 is required for high affinity microbial interactions and antimicrobial activity**. *Front. Microbiol.* (2021.0) **12** 731026. DOI: 10.3389/fmicb.2021.731026
47. Stowell S.R., Arthur C.M., Dias-Baruffi M., Rodrigues L.C., Gourdine J.-P., Heimburg-Molinaro J., Ju T., Molinaro R.J., Rivera-Marrero C., Xia B.. **Innate immune lectins kill bacteria expressing blood group antigen**. *Nat. Med.* (2010.0) **16** 295-301. DOI: 10.1038/nm.2103
48. Paterson A.D., Lopes-Virella M.F., Waggott D., Boright A.P., Hosseini S.M., Carter R.E., Shen E., Mirea L., Bharaj B., Sun L.. **Genome-Wide Association Identifies the ABO Blood Group as a Major Locus Associated with Serum Levels of Soluble E-Selectin**. *Arterioscler. Thromb. Vasc. Biol.* (2009.0) **29** 1958. DOI: 10.1161/ATVBAHA.109.192971
49. Redondo-Bermejo B., Pascual-Figal D.A., Hurtado-Martínez J.A., Montserrat-Coll J., Peñafiel-Verdú P., Pastor-Pérez F., Giner-Caro J.A., Valdés-Chávar M.. **Clinical determinants and prognostic value of hemoglobin in hospitalized patients with systolic heart failure**. *Rev. Española Cardiol.* (2007.0) **60** 597-606. DOI: 10.1157/13107116
50. Meijers W.C., van der Velde A.R., Ruifrok W.P., Schroten N.F., Dokter M.M., Damman K., Assa S., Franssen C.F., Gansevoort R.T., van Gilst W.H.. **Renal handling of galectin-3 in the general population, chronic heart failure, and hemodialysis**. *J. Am. Heart Assoc.* (2014.0) **3** e000962. DOI: 10.1161/JAHA.114.000962
51. Mueller C., Laule-Kilian K., Scholer A., Nusbaumer C., Zeller T., Staub D., Perruchoud A.P.. **B-type natriuretic peptide for acute dyspnea in patients with kidney disease: Insights from a randomized comparison**. *Kidney Int.* (2005.0) **67** 278-284. DOI: 10.1111/j.1523-1755.2005.00079.x
52. Das S.R., Drazner M.H., Dries D.L., Vega G.L., Stanek H.G., Abdullah S.M., Canham R.M., Chung A.K., Leonard D., Wians F.H.. **Impact of body mass and body composition on circulating levels of natriuretic peptides: Results from the Dallas Heart Study**. *Circulation* (2005.0) **112** 2163-2168. DOI: 10.1161/CIRCULATIONAHA.105.555573
53. Winkelmann B.R., März W., Boehm B.O., Zotz R., Hager J., Hellstern P., Senges J.. **Rationale and design of the LURIC study--a resource for functional genomics, pharmacogenomics and long-term prognosis of cardiovascular disease**. *Pharmacogenomics* (2001.0) **2** S1-S73. DOI: 10.1517/14622416.2.1.S1
54. De Boer R.A., Schroten N.F., Bakker S.J.L., Mahmud H., Szymanski M.K., van der Harst P., Gansevoort R.T., van Veldhuisen D.J., van Gilst W.H., Hillege H.L.. **Plasma renin and outcome in the community: Data from PREVEND**. *Eur. Heart J.* (2012.0) **33** 2351-2359. DOI: 10.1093/eurheartj/ehs198
55. Diercks G.F., Janssen W.M., van Boven A.J., Bak A.A., de Jong P.E., Crijns H.J., van Gilst W.. **Rationale, design, and baseline characteristics of a trial of prevention of cardiovascular and renal disease with fosinopril and pravastatin in nonhypertensive, nonhypercholesterolemic subjects with microalbuminuria (the Prevention of REnal and Vascular E**. *Am. J. Cardiol.* (2000.0) **86** 635-638. DOI: 10.1016/S0002-9149(00)01042-0
56. Diercks G.F.H., van Boven A., Hillege H., Janssen W., Kors J., De Jong P., Grobbee D., Crijns H., van Gilst W.. **Microalbuminuria is independently associated with ischaemic electrocardiographic abnormalities in a large non-diabetic population. The PREVEND (Prevention of REnal and Vascular ENdstage Disease) study**. *Eur. Heart J.* (2000.0) **21** 1922-1927. DOI: 10.1053/euhj.2000.2248
57. Meijers W.C., Van Der Velde A.R., De Boer R.A.. **The ARCHITECT galectin-3 assay: Comparison with other automated and manual assays for the measurement of circulating galectin-3 levels in heart failure**. *Expert Rev. Mol. Diagn.* (2014.0) **14** 257-266. DOI: 10.1586/14737159.2014.892421
58. Christenson R.H., Duh S.-H., Wu A.H., Smith A., Abel G., Defilippi C.R., Wang S., Adourian A., Adiletto C., Gardiner P.. **Multi-center determination of galectin-3 assay performance characteristics: Anatomy of a novel assay for use in heart failure**. *Clin. Biochem.* (2010.0) **43** 683-690. DOI: 10.1016/j.clinbiochem.2010.02.001
59. Bedu-Addo G., Gai P.P., Meese S., Eggelte T.A., Thangaraj K., Mockenhaupt F.P.. **Reduced prevalence of placental malaria in primiparae with blood group O**. *Malar. J.* (2014.0) **13** 289. DOI: 10.1186/1475-2875-13-289
60. Pilz S., Tomaschitz A., Drechsler C., Ritz E., Boehm B.O., Grammer T.B., März W.. **Parathyroid hormone level is associated with mortality and cardiovascular events in patients undergoing coronary angiography**. *Eur. Heart J.* (2010.0) **31** 1591-1598. DOI: 10.1093/eurheartj/ehq109
|
---
title: 'The Effect of Different Physical Exercise Programs on Physical Fitness among
Preschool Children: A Cluster-Randomized Controlled Trial'
authors:
- Guangxu Wang
- Dan Zeng
- Shikun Zhang
- Yingying Hao
- Danqing Zhang
- Yang Liu
journal: International Journal of Environmental Research and Public Health
year: 2023
pmcid: PMC10002293
doi: 10.3390/ijerph20054254
license: CC BY 4.0
---
# The Effect of Different Physical Exercise Programs on Physical Fitness among Preschool Children: A Cluster-Randomized Controlled Trial
## Abstract
Background: Preschool children are in a period of rapid physical and psychological development, and improving their level of physical fitness is important for their health. To better develop the physical fitness of preschool children, it is very important to understand the behavioral attributes that promote the physical fitness of preschool children. This study aimed to determine the effectiveness of and the differences between different physical exercise programs in improving preschool children’s physical fitness. Methods: A total of 309 preschool children aged 4–5 years were recruited from 5 kindergartens to participate in the experiment. They were cluster-randomly allocated into five groups: basic movements (BM) group, rhythm activities (RA) group, ball games (BG) group, multiple activities (MA) group, and control (CG) group. The intervention groups received designed physical exercise programs with a duration of 30 min 3 times per week for 16 weeks. The CG group received unorganized physical activity (PA) with no interventions. The physical fitness of preschool children was measured using the PREFIT battery before and after the interventions. One-way analysis of variance, a nonparametric test; generalized linear models (GLM); and generalized linear mixed models (GLMM) were used to examine differences during the pre-experimental stage among groups and to assess the differential effects of the intervention conditions on all outcome indicators. The intervention condition models were adjusted for potential confounders (baseline test results, age, gender, height, weight, and body mass index) explaining the main outcome variance. Results: The final sample consisted of 253 participants (girls $46.3\%$) with an average age of 4.55 ± 0.28 years: the BG group ($$n = 55$$), the RA group ($$n = 52$$), the BM group ($$n = 45$$), the MA group ($$n = 44$$), and the CG group ($$n = 57$$). The results of the generalized linear mixed model and generalized linear model analyses indicated significant differences for all physical fitness tests between groups, except for the 20 m shuttle run test and the sit-and-reach test after the interventions. Grip strength was significantly higher in the BG and MA groups than in the BM group. The scores for standing long jump were significantly higher in the MA group than in the other groups. The scores for the 10 m shuttle run test were significantly lower in the BG and MA groups than in the CG, BM, and RA groups. The scores for skip jump were significantly lower in the BG and MA groups than in the RA group. The scores for balance beam were significantly lower in the BG and MA groups than in the RA group and significantly lower in the BG group than in the BM group. The scores for standing on one foot were significantly higher in the BG and MA groups than in the CG and RA groups and significantly higher in the BM group than in the CG group. Conclusions: Physical exercise programs designed for preschool physical education have positive effects on the physical fitness of preschool children. Compared with the exercise programs with a single project and action form, the comprehensive exercise programs with multiple action forms can better improve the physical fitness of preschool children.
## 1. Introduction
Physical fitness can be considered as the comprehensive performance of physical functions, such as muscular function, cardiovascular function, and metabolic function, effectively during daily physical activity (PA) or physical exercise [1]. Healthy levels of physical fitness guarantee that individuals participate in physical activity and work with vigor, and can promote resistance to fatigue [2]. Studies have indicated that a series of health problems in children are related to low levels of cardiorespiratory fitness and muscle strength, including skeletal dysplasia, cardiovascular metabolic diseases, and premature death in old age [3,4]. Additionally, physical fitness also plays an important role in the healthy life of preschool children, such as obesity prevention [5] and determining tibial bone mineral content, structure, and strength in 3–5-year-old children [6]. In addition to health concerns, physical fitness and intellectual maturity have been proven to be linked from an early age, even predicting intellectual maturity in 3–6-year-old children [7] and contributing to successful academic development in youth [8]. These findings highlight the need for promoting physical fitness among children and encouraging them to engage in regular physical activity.
Physical activity has been proven to be one of the important factors promoting physical fitness and is an essential factor of a healthy lifestyle [9,10]. Tan et al. [ 11] and Wick et al. [ 12] reported the advantages of physical activity programs over free play in improving the physical fitness of preschool children. The standardized physical activity lessons also exhibited significant advantages over the control group (unorganized physical activity) [13]. In addition, physical activity programs led by kindergartens and teachers have a positive effect on the physical fitness of preschool children [14]. A recent systematic review found that the physical exercise, whether on its own or combined with additional interventions, had beneficial effects on cardiorespiratory fitness, lower-body muscular strength, and speed agility in preschoolers [15]. The formulation of preschool education policy is inclined to using comprehensive exercise and encouraging kindergartens to build their own sports specialties, such as cheerleading, soccer, or basketball, to promote young children’s physical fitness [16]. In conclusion, physical activity plays a crucial role in promoting physical fitness in preschool children. The implementation of structured physical activity programs and the incorporation of exercise into preschool education policies can have a significant effect on the physical fitness and overall health of preschool children.
However, research has indicated that focusing on just one sport can lead to a series of problems in the growth and development of young athletes [17,18]. In addition, it has not been proven whether focusing on only one sport also can lead to problems in the growth and development of preschool children. Studies on the effect of different exercise plans on physical fitness have reported different results because of great differences in the quality and methods used. Moreover, the current evidence does not support a comparison of the effects of different exercise programs, which is not favorable to the selection of physical exercise programs for preschool children. In addition, most previous studies have employed professional coaches or physical educators as the implementers of intervention programs, which limits the generalizability of the findings to guiding physical education practices for preschool children. Studies on teacher-centered physical activity intervention have found no significant advantage over control groups in improving the physical fitness of preschool children [14]. To ensure positive physical fitness development in preschool children, it is important to understand the behavioral attributes and causative mechanisms that promote these outcomes [2].
On that basis, we designed a study to compare different physical exercise programs that have been proven to effectively improve the physical fitness of preschool children and are expected to respond to the evidence gap. Therefore, this study aimed to investigate the effectiveness and differences among these physical exercise programs in improving the physical fitness of preschool children.
## 2.1. Study Design and Participants
This study was a single-blind, cluster-RCT study, with the kindergarten class as the cluster for the intervention. The data were sourced from the Physical Exercise on Fundamental Movement Skills and Physical Fitness of preschoolers (PEFP) project [19]. The study population consisted of preschool children aged 4–5 years, who were physically capable of participating in sports and had obtained written consent from their parents or guardians. Participants with severe cognitive or motor impairments were accompanied by a support worker during physical exercise, but were not included in the data collection. Before the end of the interventions, the participants and teachers only participated in the physical exercise of the intervention groups and did not acquire the details of the intervention group allocation. The study was approved by the Ethics Committee of Shanghai Sport University and was registered under the ethical review number 102772019RT034.
In this study, a total of 309 preschool children aged 4 to 5 years were recruited from five kindergartens and cluster randomly assigned to 5 groups: basic movements (BM) group, rhythm activities (RA) group, ball games (BG) group, multiple activities (MA) group, and control (CG) group. The attendance rate of $30\%$ of the participants exceed $\frac{4}{5}$ of the total course, and all of the participants completed at least $\frac{2}{3}$ of the total course. After preschool children with missing pretest or posttest data were excluded, the final sample consisted of 253 participants (girls $46.3\%$) with an average age of 4.55 ± 0.28 years: the BG group ($$n = 55$$), the RA group ($$n = 52$$), the BM group ($$n = 45$$), the MA group ($$n = 44$$), and the CG group ($$n = 57$$). The flow diagram of the research process is shown in Figure 1.
## 2.2. Intervention Procedures
The present study comprised four intervention groups: the BM, RA, BG, and MA groups. Preschool children in the control group participated in unorganized PA, and the details of the interventions have been described elsewhere [19].
The intervention program consisted of structured lessons with a duration of 30 min performed three times a week for 16 weeks. Kindergarten teachers participated in the study and performed the physical exercise interventions after receiving 2 h of training at a local kindergarten. The structure of each lesson consisted of a warm-up period of 5 min, followed by a core exercise period of 20 min and a cool-down activity of 5 min. The study was performed in the winter, and precautions were taken to ensure the safety of the preschool children, such as starting with low-intensity physical activity (e.g., wrist rotations and leg swings), gradually increasing the intensity (e.g., arm rotations and knee-up walk to forceful swinging of arms and on-site running), and then slowly decreasing the intensity. To ensure comparability across the different programs, the core exercise period followed a consistent intensity control principle, whereby every 10 min of sports activities should include at least 5 min of moderate-to-high-intensity physical activity and 2 min of vigorous-intensity physical activity. The interventions were designed as games to increase the children’s interest, with the main differences being in the core exercise content. The interventions were performed within the existing physical activity plans of the kindergartens to avoid additional physical activity for the preschool children in the intervention groups. The intensity of PA was estimated by teachers on the basis of the active behavior of the preschool children and was determined using the “Compendium of Physical Activity” developed by Ainsworth et al. [ 20] and the Preschool-Age Children’s Physical Activity Questionnaire [21].
Preschool children in the control group participated in unorganized PA. The PA schedules were arranged by the kindergarten without the guidance of teachers, and the types and intensity of activities were determined by the preschool children.
## 2.3. Measurement Procedures
Physical fitness and descriptive data (e.g., age, sex, height, and weight) of preschool children were tested at baseline and at the end of the interventions, and each test was completed within a week. The physical fitness assessment was primarily based on the PREFIT battery, which has demonstrated satisfactory reliability and validity in evaluating the physical fitness of 4–6-year-old children [22].
The physical fitness of the preschool children was evaluated through a comprehensive test battery consisting of measures of cardiorespiratory fitness, musculoskeletal fitness, and motor fitness. The cardiorespiratory fitness of preschool children was assessed by testing the 20 m shuttle run. The musculoskeletal fitness of preschool children was assessed by testing grip strength and standing sit-and-reach. The motor fitness of preschool children was assessed by testing the 10 m shuttle run, balance beam walk, and standing on one foot and hoping. Additionally, anthropometric data, such as height and weight, were collected, and the body mass index (BMI) was calculated from these measurements. The standard testing procedures employed in this study have been described in detail elsewhere [19].
## 2.4. Statistical Analysis
The data were first tested for normality using standardized skewness and kurtosis values. Normally distributed data were presented as the mean and standard deviation, while non-normally distributed data were presented as the interquartile range. One-way analysis of variance (ANOVA) and the Kruskal–Wallis H test were used to examine differences during the pre-experimental stage among groups. The matched samples t-test and Wilcoxon rank-sum test were used to examine the differences of the physical fitness tests in groups before and after intervention. Generalized Linear models (GLMs) were used to assess the differential impacts of the intervention conditions on all outcome indicators for normally distributed data. Generalized Linear mixed models (GLMMs) were used to assess the differential effects of the intervention conditions on all outcome indicators for non-normally distributed data. The intervention condition (CG, BM, RA, BG, and MA) models were adjusted for potential confounders explaining main outcome variance (baseline test results, age, gender, height, weight, and BMI). Bonferroni adjusted pairwise comparisons were employed to analyze differences among conditions, and $p \leq 0.05$ indicated that the difference is statistically significant. All statistical analyses were performed using SPSS Statistics version 26.0 (IBM Corp, Chicago, IL, USA).
## 3.1. Participant Characteristics and Physical Fitness Test before Intervention
Table 1 presents participant characteristics and physical fitness tests during the pre-intervention stage. There were significant differences among the groups before the interventions with regard to the balance beam, grip, and 20 m shuttle run test ($p \leq 0.05$). The scores for balance beam in the BG group were significantly higher than in the other groups ($p \leq 0.05$). The grip strength of the CG and BG groups was significantly higher than that of the BM and RA groups ($p \leq 0.05$). The grip strength of the CG group was significantly higher than that of the MA group ($p \leq 0.05$). The scores for the 20 m shuttle run test in the CG group were significantly higher in the BM and RA groups ($p \leq 0.05$). The remaining indexes revealed no significant differences among the different groups (Table 1). On the basis of previous literature and results, age, gender, height, weight, and BMI were included as covariates in the subsequent analyses.
## 3.2. Physical Fitness Changes after Intervention
Table 2 presents the results of a matched samples t-test and Wilcoxon rank-sum test for the differences between the physical fitness tests before and after the interventions. The pre-post effect sizes exhibited a significant decrease in the sit-and-reach test in all groups after the interventions ($p \leq 0.01$). In the CG group, the 10 m shuttle run performance of preschool children decreased significantly after the experiment ($$p \leq 0.009$$). There were significant improvements in the 20 m shuttle run test ($$p \leq 0.001$$), grip ($$p \leq 0.000$$), standing on one foot ($$p \leq 0.027$$), and skip jump ($$p \leq 0.009$$) following the interventions in the BM group. The RA group had significant improvements in the 20 m shuttle run test ($$p \leq 0.042$$), grip ($$p \leq 0.000$$), and 10 m shuttle run test ($$p \leq 0.008$$) after the interventions. The BG group had significant improvements in grip ($$p \leq 0.000$$), standing on one foot ($$p \leq 0.009$$), 10 m shuttle run test ($$p \leq 0.000$$), and skip jump ($$p \leq 0.002$$) after the interventions. There was a significant improvement in grip ($$p \leq 0.000$$), standing long jump ($$p \leq 0.000$$), standing on one foot ($$p \leq 0.001$$), 10 m shuttle run test ($$p \leq 0.000$$), skip jump ($$p \leq 0.014$$), and balance beam ($$p \leq 0.047$$) following the interventions in the MA group. The remaining indexes revealed no significant differences before and after the interventions.
Figure 2 presents the results of the generalized linear mixed-model analyses and generalized linear models for each of the physical fitness tests after the interventions. Grip strength was significantly higher in the BG and MA groups than in the BM group ($p \leq 0.05$), indicating that the BG and MA groups had a significantly better improvement in the grip strength of preschool children than the BM group. The scores for standing long jump were significantly higher in the MA group than in the other groups ($p \leq 0.05$), indicating that the MA group had a significantly better improvement in the standing long jump of preschool children than the other groups. The scores for the 10 m shuttle run test were significantly lower in the BG and MA groups than in the CG, BM, and RA groups ($p \leq 0.05$), indicating that the BG and MA groups had a significantly better improvement in the 10 m shuttle run test of preschool children than the CG, BM, and RA groups. The scores for standing on one foot were significantly higher in the BG and MA groups than in the CG and RA groups ($p \leq 0.05$) and significantly higher in the BM group than in the CG group ($p \leq 0.05$), indicating that the BG and MA groups had a significantly better improvement in the standing on one foot of preschool children than the CG and RA groups. The scores for skip jump were significantly lower in the BG and MA groups than in the RA group ($p \leq 0.05$), indicating that the BG and MA groups had a significantly better improvement in the skip jump of preschool children than the RA group. The scores for balance beam were significantly lower in the BG and MA groups than in the RA group ($p \leq 0.05$) and significantly lower in the BG group than in the BM group ($p \leq 0.05$), indicating that the BG and MA groups had a significantly better improvement in the balance beam of preschool children than the RA group. However, the scores for the 20 m shuttle run test and the sit-and-reach test revealed no significant differences among the different groups.
## 4. Discussion
Preschool children undergo a period of rapid physical growth and maturation of the nervous system, requiring the development of corresponding physical fitness, such as agility, strength, and reaction speed [23,24]. Evidence from systematic reviews focuses on the strong association between cardiorespiratory fitness and musculoskeletal fitness and the development of motor competence throughout early years, childhood, and adolescence, with increasing strength with age [2,25]. Based on this evidence, it is rational to believe that the importance of physical fitness in preschool children should be the same as that of older children [26]. The present study aimed to identify more effective physical exercise programs to improve the physical fitness of preschool children and provide evidence for the implementation of preschool physical education. Following these 16-week interventions in preschool, children exhibited improvements in all physical fitness tests after intervention for all intervention groups, except for the sit-and-reach test, and the balance beam test in the RA group. In the CG group, the preschool children showed no significant increase in all physical fitness indicators of preschool children. The BG and MA groups had a certain advantage over the BM, RA, and CG groups in improving the physical fitness of preschool children.
In terms of cardiorespiratory fitness, pre-post effect sizes exhibited significant improvements in the 20 m shuttle run test in the BM and RA groups, which is consistent with previous studies [15]. However, the improvement in the 20 m shuttle run test before and after the interventions in the BG and MA groups was not as pronounced and not statistically significant. This may be because the baseline cardiorespiratory fitness levels of the preschool children in the BM and RA groups were lower than those of the children in the BG and MA groups. Previous research has indicated that the baseline level of physical fitness in preschool children can affect the effect size of interventions, with higher baseline scores leading to smaller improvements and lower baseline scores leading to larger changes [27,28]. In addition, after the baseline test value and other confounding factors were adjusted, the BG and MA groups demonstrated an advantage in terms of improving cardiorespiratory fitness when compared with the BM and RA groups. Systematic review and meta-analysis results from recent studies have indicated that all types of physical activity programs, including free play, can improve the cardiorespiratory fitness of preschool children to a certain extent [15,29], which is consistent with the findings of this study.
The muscle strength (grip and standing long jump) of preschool children in all intervention groups, including the control group, obviously improved after the interventions, which is consistent with previous research findings [15,27]. The BG and MA groups demonstrated advantages over the BM and RA groups in terms of grip strength improvement, whereas the MA group demonstrated significant improvements in standing long jump performance when compared with the other groups. However, the flexibility (sit-and-reach) of preschool children in all intervention groups and the control group decreased significantly, which contrasts with previous findings [15,27]. Long-term studies have indicated that preschool children’s physical fitness will gradually increase with age [30,31], except for flexibility, which may exhibit little change or even decrease without targeted practice [27,32]. In addition, the decline of the preschool children’s flexibility may be affected by the season (children’s clothing and temperature). The baseline test was in the autumn, when clothes and temperature had little effect on children’s motor performance. The interventions ended in the winter, when cold temperatures and heavy clothes have a great effects on children’s motor performance [33]. It is known that possessing adequate flexibility, range of motion, and muscle strength can mitigate the risk of injury in sports or everyday activities, particularly in later life, when the negative effect of decreased flexibility on health cannot be disregarded [34]. Therefore, the flexibility exercise of preschool children should be an important part of the physical exercise program formulation. The results of this study suggest that the MA intervention exhibited advantages in improving the muscle strength of preschool children when compared with other physical exercise programs and the control group. However, further research is warranted to better understand the effect of various physical exercise programs on the flexibility of preschool children.
The motor fitness of preschoolers was obviously improved in all intervention groups after the 16-week interventions, and the intervention groups had certain advantages over the CG group. Previous research, and systematic review and meta-analysis also, indicated that the designed physical activity programs had a positive effect on the motor fitness of preschool children [11,15,29], similar to the results of this study. The BG and MA groups displayed obvious advantages in improving the motor fitness of preschool children when compared with the BM, RA, and CG groups. This may be caused by a close relationship between the performance of children’s motor fitness and the level of motor skills [2]. A study on the effect of different exercise programs on the motor skills of preschool children has indicated that multilateral exercise has certain advantages over specific programs of rhythmic gymnastics and soccer [18]. The physical exercise of the MA group may better improve the motor performance of preschool children by better improving their motor skills. In this study, ball games have similar intervention effects on preschoolers’ motor fitness with multiple activities. In addition, research has indicated that the motor fitness of preschool children was significantly improved with small improvements in cardiovascular fitness. The BG and MA groups exhibited advantages in improving cardiorespiratory fitness when compared with the BM, RA, and CG groups. This may help explain why BG and MA can better improve the motor fitness of preschool children.
In summary, the MA group had advantages over the BM, RA, and CG groups in terms of the improvement of the physical fitness of preschool children. In addition, in this study, the BM and RA groups had no advantages over the CG group with regard to improvement of cardiorespiratory fitness, musculoskeletal fitness, and motor fitness. These results are similar to those of another study that found that teacher-centered intervention granted preschool children no advantage over the control group in terms of motor fitness [14]. There is an evidence gap with regard to the effect of different physical exercise programs on the physical fitness of preschool children, and there were no similar results for reference to verify whether the results of this study are reasonable. However, relevant studies have indicated that early specialized sports training or focusing on the development of just one sport may lead to a series of growth and development problems, such as physical and physiological imbalance, unilateral muscle development, risk of injury, coordinated development disorder, and limitations on differentiated skill acquisition, and even a negative effect on mental health, and can also reduce children’s enthusiasm for PA participation [17,35]. In addition, studies have indicated that the diversified sports activity module and the structured multisport program have significant advantages over free play or conventional sports activity in improving the physical fitness of preschool children [27,36]. According to the research of Stodden et al. [ 37] and Lubans et al. [ 38], PA, physical fitness, and motor skills all reinforce each other, and multilateral exercise has certain advantages over the single exercise mode in improving the motor skills of preschool children [17]. This evidence can help explain why multiple activity programs can better improve the physical fitness of preschool children.
There are several limitations that need to be addressed in this study. The first is in terms of sample representation; because of the scale and difficulty of the experiment, only 4–5-year-old preschoolers were included in this study. Therefore, the results of this study may not be applicable to all preschool children. Second, the baseline level of physical fitness in the experimental groups was not balanced. The improvement after the interventions will be greater if the baseline test level is low. In addition, the physical environments of the baseline test and post-intervention test were relatively different. Therefore, the significance of analyzing the improvement of physical fitness before and after the interventions is limited. However, we used a mixed-effects model to adjust the effect of the baseline test results, gender, and other factors in the intervention effect. In addition, all kindergartens participating in the experiment were in the same community, and the test environment was similar. Finally, the number of preschool children in each group included in the analysis was not balanced, but the minimum sample size that meets the statistical analysis was 30 children per group [19].
## 5. Conclusions
Physical exercise programs designed for preschool physical education have positive effects on the physical fitness of preschool children. Compared with the exercise programs with a single project and action form, comprehensive exercise programs with multiple action forms can better improve the physical fitness of preschool children.
## References
1. Ortega F.B., Ruiz J.R., Castillo M.J., Sjöström M.. **Physical fitness in childhood and adolescence: A powerful marker of health**. *Int. J. Obes.* (2008) **32** 1-11. DOI: 10.1038/sj.ijo.0803774
2. Cattuzzo M.T., Rafael D.S.H., Ré A.H.N., Oliveira I.S., Melo B.M., Moura M.S., Araújo R.C., Stodden D.. **Motor competence and health related physical fitness in youth: A systematic review**. *J. Sci. Med. Sport* (2016) **19** 123-129. DOI: 10.1016/j.jsams.2014.12.004
3. Smith J.J., Eather N., Morgan P.J., Plotnikoff R.C., Faigenbaum A.D., Lubans D.R.. **The health benefits of muscular fitness for children and adolescents: A systematic review and meta-analysis**. *Sports Med.* (2014) **44** 1209-1223. DOI: 10.1007/s40279-014-0196-4
4. Ortega F.B., Silventoinen K., Tynelius P., Rasmussen F.. **Muscular strength in male adolescents and premature death: Cohort study of one million participants**. *BMJ* (2012) **345** e7279. DOI: 10.1136/bmj.e7279
5. Latorre-Román P.Á., Castillo R.M., Lucena-Zurita M., Sánchez J.S., García-Pinillos F., Mora-López D.. **Physical fitness in preschool children: Association with sex, age and weight status**. *Child Care Health Dev.* (2016) **43** 267-273. DOI: 10.1111/cch.12404
6. Gómez-Bruton A., Marín-Puyalto J., Muñiz-Pardos B., Lozano-Berges G., Cadenas-Sanchez C., Matute-Llorente A., Gómez-Cabello A., Moreno L.A., Gonzalez-Agüero A., Casajus J.A.. **Association Between Physical Fitness and Bone Strength and Structure in 3- to 5-Year-Old Children**. *Sports Health* (2020) **12** 431-440. DOI: 10.1177/1941738120913645
7. Latorre-Román P., Mora-López D., García-Pinillos F.. **Intellectual maturity and physical fitness in preschool children**. *Pediatr. Int.* (2016) **58** 450-455. DOI: 10.1111/ped.12898
8. García-Hermoso A., Martinez-Gomez D., Fernández-Santos D.R.J., Ortega F.B., Castro-Piñero J., Hillman C.H., Veiga O.L., Esteban-Cornejo I.. **Longitudinal associations of physical fitness and body mass index with academic performance**. *Scand. J. Med. Sci. Sports* (2020) **31** 184-192. DOI: 10.1111/sms.13817
9. Bull F.C., Al-Ansari S.S., Biddle S., Borodulin K., Buman M.P., Cardon G., Carty C., Chaput J., Chastin S., Chou R.. **World Health Organization 2020 guidelines on physical activity and sedentary behaviour**. *Br. J. Sports Med.* (2020) **54** 1451-1462. DOI: 10.1136/bjsports-2020-102955
10. Tremblay M.S., Chaput J.P., Adamo K.B., Aubert S., Barnes J.D., Choquette L., Duggan M., Faulkner G., Goldfield G.S., Gray C.E.. **Canadian 24-Hour Movement Guidelines for the Early Years (0–4 years): An Integration of Physical Activity, Sedentary Behaviour, and Sleep**. *BMC Public Health* (2017) **17**. DOI: 10.1186/s12889-017-4859-6
11. Tan S., Cheng C., Mingyang S., Xue L., Wang J.. **Exercise training improved body composition, cardiovascular function, and physical fitness of 5-year-old children with obesity or normal body mass**. *Pediatr. Exerc. Sci.* (2017) **29** 245-253. DOI: 10.1123/pes.2016-0107
12. Wick K.K.S., Granacher U.. **Effects of a Strength-Dominated Exercise Program on Physical Fitness and Cognitive Performance in Preschool Children**. *J. Strength Cond. Res.* (2021) **35** 983-990. DOI: 10.1519/JSC.0000000000003942
13. Macak D., Popovic B., Babic N., Cadenas-Sanchez C., Madic D.M., Trajkovic N.. **The effects of daily physical activity intervention on physical fitness in preschool children**. *J. Sports Sci.* (2021) **40** 146-155. DOI: 10.1080/02640414.2021.1978250
14. Kobel S., Henle L., Laemmle C., Wartha O., Szagun B., Steinacker J.M.. **Intervention Effects of a Kindergarten-Based Health Promotion Programme on Motor Abilities in Early Childhood**. *Front. Public Health* (2020) **8** 219. DOI: 10.3389/fpubh.2020.00219
15. García-Hermoso A., Martínez A.M.A., Ramírez-Vélez R., Izquierdo M.. **Effects of Exercise Intervention on Health-Related Physical Fitness and Blood Pressure in Preschool Children: A Systematic Review and Meta-Analysis of Randomized Controlled Trials**. *Sports Med.* (2020) **50** 187-203. DOI: 10.1007/s40279-019-01191-w
16. Zhou Z., Ren H., Yin Z., Wang L., Wang K.. **A policy-driven multifaceted approach for early childhood physical fitness promotion: Impacts on body composition and physical fitness in young Chinese children**. *BMC Pediatr.* (2014) **14**. DOI: 10.1186/1471-2431-14-118
17. Myer G.D., Jayanthi N., Difiori J.P., Faigenbaum A.D., Kiefer A.W., Logerstedt D., Micheli L.J.. **Sport Specialization, Part I: Does Early Sports Specialization Increase Negative Outcomes and Reduce the Opportunity for Success in Young Athletes?**. *Sports Health* (2015) **7** 437-442. DOI: 10.1177/1941738115598747
18. Sanja S., Krmpotic M., Stamenković I.. **Are specific programs a threat to overall motor development of preschool children?**. *Kinesiol. Slov.* (2016) **22** 47-55
19. Wang G., Zi Y., Li B., Su S., Sun L., Wang F., Ren C., Liu Y.. **The Effect of Physical Exercise on Fundamental Movement Skills and Physical Fitness among Preschool Children: Study Protocol for a Cluster-Randomized Controlled Trial**. *Int. J. Environ. Res. Public Health* (2022) **19**. DOI: 10.3390/ijerph19106331
20. Ainsworth B.E., Haskell W.L., Herrmann S.D., Meckes N., Bassett J.D.R., Tudor-Locke C., Greer J.L., Vezina J., Whitt-Glover M.C., Leon A.S.. **2011 Compendium of physical activities: A second update of codes and MET values**. *Med. Sci. Sport Exerc.* (2011) **43** 1575-1581. DOI: 10.1249/MSS.0b013e31821ece12
21. Dwyer G.M., Hardy L.L., Peat J.K., Baur L.A.. **The validity and reliability of a home environment preschool-age physical activity questionnaire (Pre-PAQ)**. *Int. J. Behav. Nutr. Phys. Act.* (2011) **8** 86. DOI: 10.1186/1479-5868-8-86
22. Ortega F.B., Cadenas-Sánchez C., Sánchez-Delgado G., Mora-González J., Martínez-Téllez B., Artero E.G., Castro-Piñero J., Labayen I., Chillón P., Löf M.. **Systematic Review and Proposal of a Field-Based Physical Fitness-Test Battery in Preschool Children: The PREFIT Battery**. *Sports Med.* (2015) **45** 533-555. DOI: 10.1007/s40279-014-0281-8
23. Tanaka C., Hikihara Y., Ohkawara K., Tanaka S.. **Locomotive and non-locomotive activity as determined by triaxial accelerometry and physical fitness in Japanese preschool children**. *Pediatr. Exerc. Sci.* (2012) **24** 420-434. DOI: 10.1123/pes.24.3.420
24. Abbott G., Hnatiuk J., Timperio A., Salmon J., Best K., Hesketh K.D.. **Cross-sectional and longitudinal associations between parents’ and preschoolers’ physical activity and television viewing: The HAPPY study**. *J. Phys. Act. Health* (2016) **13** 269-274. DOI: 10.1123/jpah.2015-0136
25. Utesch T., Farid B., Dirk B., Strauss B.. **The Relationship Between Motor Competence and Physical Fitness from Early Childhood to Early Adulthood: A Meta-Analysis**. *Sports Med.* (2019) **49** 541-551. DOI: 10.1007/s40279-019-01068-y
26. Cadenas-Sánchez C., Martínez-Téllez B., Sánchez-Delgado G., Mora-Gonzalez J., Castro-Piñero J., Löf M., Ruiz J.R., Ortega F.B.. **Assessing physical fitness in preschool children: Feasibility, reliability and practical recommendations for the PREFIT battery**. *J. Sci. Med. Sport* (2016) **19** 910-915. DOI: 10.1016/j.jsams.2016.02.003
27. Popovic B., Cvetkovic M., Macak D., Šćepanović T., Čokorilo N., Belić A., Trajković N., Andrašić S., Bogataj Š.. **Nine Months of a Structured Multisport Program Improve Physical Fitness in Preschool Children: A Quasi-Experimental Study**. *Int. J. Environ. Res. Public Health* (2020) **17**. DOI: 10.3390/ijerph17144935
28. Trajkovi N., Radanovi D., Madi D., Andrašić S., Cadenas-Sanchez C., Mačak D., Popović B.. **Normative data for handgrip strength in Serbian children measured with a bulb dynamometer**. *J. Hand Ther.* (2020) **34** 479-487. DOI: 10.1016/j.jht.2020.03.001
29. Szeszulski J., Lorenzo E., Shaibi G.Q., Buman M.P., Vega-López S., Hooker S.P., Lee R.E.. **Effectiveness of early care and education center-based interventions for improving cardiovascular fitness in early childhood: A systematic review and meta-analysis**. *Prev. Med. Rep.* (2019) **15**. DOI: 10.1016/j.pmedr.2019.100915
30. Leppanen M.H., Henriksson P., Delisle N.C., Henriksson H., Ortega F.B., Pomeroy J., Ruiz J.R., Cadenas-Sanchez C., Löf M.. **Longitudinal Physical Activity, Body Composition, and Physical Fitness in Preschoolers**. *Med. Sci. Sports Exerc.* (2017) **49** 2078-2085. DOI: 10.1249/MSS.0000000000001313
31. Potter M., Spence J.C., Boule N., Stearns J.A., Carson V.. **Behavior Tracking and 3-Year Longitudinal Associations Between Physical Activity, Screen Time, and Fitness Among Young Children**. *Pediatr. Exerc. Sci.* (2018) **30** 134-143. DOI: 10.1123/pes.2016-0239
32. Kobel S., Wartha O., Lmmle C., Lämmle C., Steinacker J.M.. **Intervention effects of a kindergarten-based health promotion programme on obesity related behavioural outcomes and BMI percentiles**. *Prev. Med. Rep.* (2019) **15**. DOI: 10.1016/j.pmedr.2019.100931
33. Galloway S.D., Maughan R.J.. **Effects of ambient temperature on the capacity to perform prolonged cycle exercise in man**. *Med. Sci. Sports Exerc.* (1997) **29** 1240-1249. DOI: 10.1097/00005768-199709000-00018
34. Naka H., Demura S., Matsuzawa J.. **Change of physique and physical fitness with age and its sexual difference in the elderly**. *Taiikugaku Kenkyu* (1997) **42** 84-96
35. Myers J., McAuley P., Lavie C.J., Despres J., Arena R., Kokkinos P.. **Physical Activity and Cardiorespiratory Fitness as Major Markers of Cardiovascular Risk: Their Independent and Interwoven Importance to Health Status**. *Prog. Cardiovasc. Dis.* (2015) **57** 306-314. DOI: 10.1016/j.pcad.2014.09.011
36. Liu J., Gao F., Yuan L.. **Effects of Diversified Sports Activity Module on Physical Fitness and Mental Health of 4-5-Year-Old Preschoolers**. *Iran. J. Public Health* (2021) **50** 1233-1240. DOI: 10.18502/ijph.v50i6.6422
37. Stodden D.F., Goodway J.D., Langendorfer S.J., Mary A.R., Mary E.R., Clersida G., Luis E.G.. **A Developmental Perspective on the Role of Motor Skill Competence in Physical Activity: An Emergent Relationship**. *Quest* (2008) **60** 290-306. DOI: 10.1080/00336297.2008.10483582
38. Lubans D.R., Morgan P., Cliff D., Barnett L.M., Okely A.D.. **Fundamental Movement Skills in Children and Adolescents: Review of associated health benefits**. *Sports Med.* (2010) **40** 1019-1035. DOI: 10.2165/11536850-000000000-00000
|
---
title: Longitudinal Study on the Effect of Onboard Service on Seafarers’ Health Statuses
authors:
- Andrea Russo
- Rosanda Mulić
- Ivana Kolčić
- Matko Maleš
- Iris Jerončić Tomić
- Luka Pezelj
journal: International Journal of Environmental Research and Public Health
year: 2023
pmcid: PMC10002301
doi: 10.3390/ijerph20054497
license: CC BY 4.0
---
# Longitudinal Study on the Effect of Onboard Service on Seafarers’ Health Statuses
## Abstract
Seafaring is considered one of the most stressful professions. Stressors in seafaring lead to typical symptoms of stress, such as insomnia, loss of concentration, anxiety, lower tolerance of frustration, changes in eating habits, psychosomatic symptoms and diseases, and overall reduced productivity, with the possibility of burnout and chronic responsibility syndrome. It has been previously determined that seafarers belong to high-risk occupations in terms of developing metabolic syndrome, and according to their BMIs, almost $50\%$ of all seafarers belong to the overweight and obesity categories. This is the first longitudinal study conducted with the aim of using the BIA method to determine the anthropometrical changes that occur during several weeks of continuous onboard service. This study included an observed group consisting of 63 professional seafarers with 8 to 12 weeks of continuous onboard service and a control group of 36 respondents from unrelated occupations. It was determined that Croatian seafarers fit into the current world trends regarding overweight and obesity among the seafaring population, with the following percentages in the BMI categories: underweight, $0\%$; normal weight, $42.86\%$; overweight, $39.68\%$; and obesity, $17.46\%$. It was established that the anthropometric statuses of the seafarers significantly changed during several weeks of continuous onboard service. Seafarers who served on board for 11 weeks lost 0.41 kg of muscle mass, whereas their total fat mass increased by 1.93 kg. Changes in anthropometric parameters could indicate deterioration of seafarers’ health statuses.
## 1. Introduction
The atypicality and specificities of work and family life, i.e., social life, are the main characteristics and differences of seafarers’ lives in comparison to those of the rest of the working population [1]. The variety and speed of environmental changes and exposure to continuous noise and vibration entailed make it hard to maintain psychophysical homeostasis, not to mention other stressors that arise from the specifics of the maritime profession, which are still insufficiently taken into account [2]. At least one environmental factor, such as excessively cold or warm ambient temperatures, odors, noise, poor bedding conditions, or ambient light during sleeping in cabins disturbs $91.6\%$ of seafarers [3].
Stressors in seafaring lead to typical symptoms of stress, such as insomnia, loss of concentration, anxiety, lower tolerance of frustration, changes in eating habits, psychosomatic symptoms and diseases, and overall reduced productivity, with the possibility of burnout and chronic responsibility syndrome [4].
With circadian work rhythms such as the 6:6 and 4:8 shift systems, body recovery and sleep are interrupted and often insufficient. During night work on board in the 6:6 (midnight to 6:00 a.m.) and 4:8 (midnight to 4:00 a.m.) systems, seafarers experience increased sleepiness with shorter sleep episodes [4].
Metabolic health, cancer risk, cardiovascular health, and mental health are further compromised by shift work, especially night work. This is due to the problems caused by the shift-work lifestyle, which are mainly manifested in chronic sleep deprivation, sympathovagal and hormonal imbalance, inflammation, impaired glucose metabolism, and unregulated cell cycles. As a result, such long-term conditions lead to a number of health disorders such as obesity, metabolic syndrome, type II diabetes, gastrointestinal dysfunction, impaired immune function, cardiovascular disease, excessive sleepiness, mood and social disorders, and increased risk of cancer [5].
Compared with that of other transportation sectors, fatigue in the maritime sector has been much less researched. Fixed and rotating work schedules, along with cultural and commercial pressures, directly affect seafarers’ physical and mental health [4,6].
With knowledge that during their service on board, seafarers have a limited influence on quality and quantity of food [7], and that nutritional problems are even more pronounced in multiethnic crews with different eating habits, it is clear that the physical and psychological conditions of seafarers may imperceptibly deteriorate [4].
An individual’s ability to adequately cope with the demands of such a maritime occupation depends on that individual’s state of physical and mental health. An extremely demanding maritime occupation, which limits a person in maintaining the usual way of life on land in terms of food choices, regular sleep, and often the inability to exercise, can lead to a gradual loss of physical and mental fitness, which can ultimately lead to human error, illness, and disabilities related to seafarers’ work [8].
A diet that does not include enough fresh fruits and vegetables can contribute to fatigue and has an overall negative impact on seafarers’ health [9,10]. In addition, the circadian rhythm of work affects digestion, which is most productive during the day and much less so at night, even when a person is awake and in a working rhythm [11]. Gastrointestinal disorders are very common in people who eat outside of traditional mealtimes and tend to worsen with consumption of tea, coffee, alcohol, nicotine, and some medications and supplements. Night workers are five times more likely to contract peptic ulcers than are day workers [12].
Exercise and good physical fitness have beneficial effects on the body and psyche, help in coping with stress, and can help reduce a person’s susceptibility to certain diseases and infections [13]. Some of the anthropometric methods for assessing a person’s health status are analysis of body composition and evaluation of body structure. The most-used methods are bioelectrical impedance (BIA) and the body mass index (BMI) [14]. The BMI is widely accepted and used as a standard test, and BIA is a valid and precise method for determining the body compositions of normal, healthy people [15] and athletes [16]. Due to fast and noninvasive measurement, BIA is widely used within the athlete population, but it has never been used in the population of professional seafarers.
Therefore, the aim of this study was to determine the body compositions of Croatian seafarers and investigate changes in anthropometric parameters during continuous onboard service.
## 2.1. Subject and Variable Sample
The subject sample included 99 adults from Croatia (Caucasian), divided into a control group and an experimental group. The control group included 36 subjects with a mean chronological age of 33.56 ± 8.49 years, a mean body height of 183.22 ± 5.58 cm, and a mean body mass of 93.15 ± 15.36 kg. The sample in the control group was a convenience sample selected to resemble the experimental group in the initial testing. Furthermore, the test subjects in the control group were selected on the condition that they did not perform jobs that may be similar to those of the seafarers, or that involve long-term absence from home (e.g., drivers, pilots, soldiers, coaches, athletes, etc.). The experimental group included 63 subjects with a mean chronological age of 35.00 ± 8.08 years, a mean body height of 183.73 ± 5.94 cm, and a mean body mass of 89.43 ± 10.82 kg. The subject sample included professional seafarers who serve on merchant ships as officers aboard various types of ships and for various companies. To make the experimental group as homogenous as possible, subjects whose service aboard was shorter than 8 weeks or longer than 12 weeks were excluded from the sample. This period did not include the “idle” time between testing and departure, i.e., return from the ship. All subjects were measured on two occasions, the initial and final measurements, performed during morning hours. The subjects in the experimental group (professional seafarers) were measured within seven days before departure and within seven days after returning home. Subjects in the control group were measured with random selection in the final testing, 8 to 12 weeks after the initial testing.
Two anthropometric variables were measured, body height and body mass, which were then used to calculate the body mass index. All measurements were taken according to the International Society for the Advancement of Kinanthropometry—ISAK protocol [17]. Furthermore, the subjects were measured with a Tanita BC-418 (Tanita Corp., Tokyo, Japan) device following the recommendations of Kyle et al. [ 18], and the results of the following anthropometric measures were determined using the bioelectrical impedance method: the body fat percentage, fat mass, visceral fat, metabolic age, fat-free mass, total body water, extracellular water, intracellular water, muscle mass, the skeletal muscle index, bone mass, and the basal metabolic rate.
## 2.2. Description of Body Composition Measures
Body composition measures (the body fat percentage, fat mass, visceral fat, metabolic age, fat-free mass, total body water, extracellular water, intracellular water, muscle mass, the skeletal muscle index, bone mass, and the basal metabolic rate) were determined with the bioelectrical impedance method, using a Tanita BC-418 device (Tanita Corp., Tokyo, Japan). The subjects were measured barefoot and in dry underwear. The “body type” setting was set to “normal” for all subjects, whereas the “weight of clothes” was set to 0.0 kg.
Body Fat Percentage—the proportion of fat to the total body weight.
Fat Mass—the actual weight of the fat in the body.
Visceral fat—fat located deep in the core abdominal area, surrounding and protecting the vital organs.
Muscle Mass—the predicted weight of muscle in the body.
Total Body Water—the total amount of fluid in the body, expressed as a percentage of the total weight.
Extracellular Water—body fluid found outside of cells.
Intracellular Water—fluid found inside cells.
Bone Mass—the predicted weight of bone mineral in the body.
Basal Metabolic Rate—the daily minimum level of energy or calories the body requires when at rest (including sleeping) in order to function effectively.
Metabolic Age—a comparison of the basal metabolic rate (BMR) to the BMR average of a chronological age group. If the metabolic age is higher than the actual age, it is an indication that improving the metabolic rate is needed.
Skeletal Muscle Index—the ratio of the muscle in the arms and legs to height.
## 2.3. Methods of Data Analysis
For all the measured variables and for each subject sample separately, the following descriptive parameters were calculated: arithmetic mean (AM); standard deviation (SD); median (M), minimum (MIN) and maximum (MAX) results; and the coefficients of asymmetry (SKEW) and peakedness (KURT) of result distribution. Normality of distribution was tested with the Kolmogorov–Smirnov test (KS). The differences in initial testing in chronological age, anthropometric characteristics, and body composition measures between the control and experimental groups were determined using the independent samples t-test. The differences between the initial and final measurements of chronological age, anthropometric characteristics, and body composition measures between the control and experimental groups were determined using the t-test for dependent samples.
For each measured variable, the differences between the initial and final tests of the control and experimental groups were calculated and arithmetic means were determined. The differences between the initial and final tests of chronological age, anthropometric characteristics, and body composition measures in the control and experimental groups were determined using the independent samples t-test. The data were analyzed using Statistica Ver 11.0 (SoftStat, SAD, Tulsa, OK, USA).
## 3. Results
Table 1 presents the results of the Kolmogorov–Smirnov test of anthropometric variables indicate that no variable exceeded the cutoff value of the test, which was 0.23 for the observed sample. This indicates that there were no significant deviations of the variables from normal distribution, and all variables were suitable for further parametric statistical analysis.
Table 2 presents the results of the Kolmogorov–Smirnov test of anthropometric variables indicate that no variable exceeded the cutoff value of the test, which was 0.17 for the observed sample. This indicates that there were no significant deviations of the variables from normal distribution, and all variables were suitable for further parametric statistical analysis.
Table 3 presents that in the initial t-test measurement, no significant differences were found between the control and experimental groups in the arithmetic mean scores of the measured variables.
Table 4 presents that the t-test revealed significant differences between the initial and final measurements in the experimental group for the following variables: age, weight, the body mass index, the fat percentage, fat mass, visceral fat, metabolic age, fat-free mass, total body water, intracellular water, and muscle mass.
Table 5 presents that the t-test revealed significant differences between the control and experimental groups in the changes in the values of the measured variables between the initial and final measurements in the following variables: weight, the body mass index, fat percentage, fat mass, visceral fat, and metabolic age.
## 4. Discussion
The BMI was the most frequently measured/analyzed anthropometric variable in previous research on a sample of professional seafarers [19,20,21,22,23,24,25]. In this study, the following proportions of professional seafarers regarding the BMI categories to which they belong were determined: underweight, $0\%$; normal weight, $42.86\%$; overweight, $39.68\%$; and obesity, $17.46\%$. We should compare the obtained results with those of other authors with great caution because the BMI depends, among other things, on the cultural and ethnic characteristics of the population [26]. In a sample of 1155 subjects, Nittari [24] found an average BMI of 25.7 kg/m2, and the proportions were very similar to those found in this study: underweight, $0.8\%$; normal weight, $47.20\%$; overweight, $40.80\%$; and obesity, $11.20\%$. Similar results were found in a study conducted by Gamo Sagaro in 2021 [25], in which the mean BMI was 25.55 kg/m2 and the following percentages were determined in the BMI categories: underweight, $0\%$; normal weight, $51.90\%$; overweight, $39.30\%$; and obesity, $8.50\%$. The comparison with the 2021 study is even more significant because the average age of the subjects ($$n = 603$$) was 37.37 years: very similar to the sample in this study. The higher proportion of obesity and higher mean BMI values in these seafarers compared to the Nittari research can be explained with the fact that $51\%$ of subjects in that study were Filipinos and Indians, who by default have a lower tendency to be overweight and obese [27]. Results almost identical to the results of this study were determined in Hoeyer’s 2005 [19] study on seafarers aged 25–44 years ($$n = 613$$): underweight, $2.8\%$; normal weight, $40.0\%$; overweight, $38.8\%$; and obesity, $18.4\%$.
A higher proportion of overweight and obese seafarers compared to the observed sample was determined in a study conducted by Hansen in 2011 [20], which, among other things, indicated an increase in the frequency of overweight among seafarers. We can conclude that Croatian seafarers fit into the current world trends regarding overweight and obesity among the seafaring population, which is defined as one of the main health problems of today. However, in comparison of the BMIs of Croatian seafarers with WHO data for the *Croatian* general population, Croatian seafarers have lower mean BMI values and thus a lower proportion of overweight and obesity. The control group also had lower mean BMI values than the general population, according to the WHO [27].
BIA is a very fast, simple, and reliable method for body composition analysis [28,29,30,31]. Although BIA measurement is widely used among top athletes [32,33], it has not been used in the population of professional seafarers until now. Moreover, it has not even been used in the population of drivers, who, along with seafarers, belong to the group of highest-risk workers [34]. The observed sample of seafarers and the control group had lower %BF values than did maritime university students [22], even though the subjects in this study were significantly older and the percentage of fat tissue has a tendency to increase with age [35]. This can also be explained with the fact that the studies did not use body-composition-analysis instruments from the same manufacturer. In a study on a sample of professional firefighters, [36] the same analysis equipment was used as in this study, and the results indicated similar %BF values as in seafarers of the same age, i.e., slightly higher %BF values in older firefighters, as expected. In addition to determining anthropometrical characteristics of seafarers, this study aimed to analyze changes in body compositions of seafarers during service on board. To the authors’ knowledge, this is the first longitudinal study on the population of professional seafarers.
To ensure an unambiguous interpretation of results, this study also included a control group of subjects, which did not significantly differ statistically from the experimental group. During 10.97 weeks of onboard service, the total body mass of the professional seafarers increased by 1.50 kg. Although the change in total body mass compared to that of the control group was significant, it should not be a cause for concern in real life. However, analysis of body composition revealed fundamental problems that, at first glance, remained hidden in the relatively small change in total body mass. During their service on board, the seafarers on average, lost 0.41 kg of their total muscle mass, whereas their total fat masses increased by 1.93 kg. Of course, this “negative” transformation was also reflected in other indicators of body composition. Thus, an increase of 1.81 percentage points in the percentage of body fat and an increase of 0.73 in the visceral fat rating were determined.
Average, muscle mass loss of 0.41 kg and total body fat increase of 1.93 kg was recorded among the sample of subjects who served on board for 11 weeks. An increased proportion of fat mass in the body structure results in risk of metabolic syndrome, which is characterized with visceral obesity associated with insulin resistance, arterial hypertension, dyslipidemia, diabetes, and glucose intolerance. Possible causes of these rapid anthropometric changes are physical inactivity on board and circadian rhythm disorders with sleep disorders. Lack of sleep and circadian sleep disorders are symptoms of many conditions. Jepsen concluded that lack of sleep is associated with obesity [37], and it is debated whether circadian sleep disorders are the causes or the consequences of some neurodegenerative diseases [38,39].
Body composition is a much better indicator of the degree of nutritional status than the body mass index is because obesity is not defined as increased body mass but as an increased proportion of adipose tissue in body mass. Among the study subjects, an average increase of 1.81 percentage points % in the percentage of body fat and an average increase of 0.73 in the visceral fat rating was determined.
## 5. Conclusions
In this paper, anthropometrical characteristics of professional seafarers, which can certainly be a point of reference for future research, were determined. Furthermore, this is one of the rare studies in which the problem of the influence of onboard service on professional seafarers’ health was approached through a longitudinal study. It was established that the anthropometric statuses of the seafarers significantly changed during several weeks of continuous onboard service. These changes in anthropometric parameters could indicate deterioration of seafarers’ health statuses. We can only speculate about the causes of those anthropometric changes in a relatively short interval. The main shortcomings of this study are reflected in the fact that no external factors were measured. Therefore, it is recommended for future studies to include tests and methods aimed at detecting possible negative factors such as diet, sleep, psychological stress, etc. It is also recommended to repeat this research on seafarers who perform other types of work on merchant fleets (engineers, auxiliary staff, etc.), seafarers of other maritime occupations (fishermen, skippers, etc.), and other professionals who must leave home for a long time to perform their jobs (pilots, drivers, seasonal workers, etc.).
## References
1. Nenadić A., Jašić D., Krajnović A.. **Sociological Aspects of Seafarers’ Life and Work and Management Styles in Shipping**. *Tranzicija* (2015.0) **17** 25-40
2. Kim J.-H., Jang S.-N.. **Seafarers’ Quality of Life: Organizational Culture, Self-Efficacy, and Perceived Fatigue**. *Int. J. Environ. Res. Public Health* (2018.0) **15**. DOI: 10.3390/ijerph15102150
3. Kerkamm F., Dengler D., Eichler M., Materzok-Köppen D., Belz L., Neumann F.A., Zyriax B.C., Harth V., Oldenburg M.. **Measurement Methods of Fatigue, Sleepiness, and Sleep Behaviour Aboard Ships: A Systematic Review**. *Int. J. Environ. Res. Public Health* (2021.0) **19**. DOI: 10.3390/ijerph19010120
4. Slišković A., Penezić Z.. **Occupational stressors, risks and health in the seafaring population**. *Rev. Psychol.* (2015.0) **22** 29-39. DOI: 10.21465/rp0022.0004
5. James S.M., Honn K.A., Gaddameedhi S., Van Dongen H.P.. **Shift Work: Disrupted Circadian Rhythms and Sleep—Implications for Health and Well-being**. *Curr. Sleep Med. Rep.* (2017.0) **3** 104-112. DOI: 10.1007/s40675-017-0071-6
6. Allen P., Wadsworth E., Smith A.. **Seafarers’ fatigue: A review of the recent literature**. *Int. Marit. Health* (2008.0) **59** 81-92. PMID: 19227741
7. Oldenburg M., Harth V., Jensen H.-J.. **Overview and prospect: Food and nutrition of seafarers on merchant ships**. *Int. Marit. Health* (2013.0) **64** 191-194. DOI: 10.5603/IMH.2013.0003
8. Godinho M.R., Ferreira A.P., Greco R.M., Teixeira L.R., Teixeira M.T.B.. **Work ability and health of security guards at a public University: A cross-sectional study**. *Rev. Latino-Am. Enferm.* (2016.0) **24** e2725. DOI: 10.1590/1518-8345.0616.2725
9. Tardy A.-L., Pouteau E., Marquez D., Yilmaz C., Scholey A.. **Vitamins and Minerals for Energy, Fatigue and Cognition: A Narrative Review of the Biochemical and Clinical Evidence**. *Nutrients* (2020.0) **12**. DOI: 10.3390/nu12010228
10. Huskisson E., Maggini S., Ruf M.. **The Role of Vitamins and Minerals in Energy Metabolism and Well-Being**. *J. Int. Med Res.* (2007.0) **35** 277-289. DOI: 10.1177/147323000703500301
11. Grant C.L., Dorrian J., Coates A.M., Pajcin M., Kennaway D.J., Wittert G.A., Heilbronn L.K., Della Vedova C., Gupta C.C., Banks S.. **The impact of meal timing on performance, sleepiness, gastric upset, and hunger during simulated night shift**. *Ind. Health* (2017.0) **55** 423-436. DOI: 10.2486/indhealth.2017-0047
12. 12.
International Maritime Organization
Guidelines on FatigueInternational Maritime OrganizationLondon, UK2002. *Guidelines on Fatigue* (2002.0)
13. Anderson E., Shivakumar G.. **Effects of exercise and physical activity on anxiety**. *Front. Psychiatry* (2013.0) **4** 27. DOI: 10.3389/fpsyt.2013.00027
14. Buchholz A.C., Bartok C., Schoeller D.A.. **The Validity of Bioelectrical Impedance Models in Clinical Populations**. *Nutr. Clin. Pract.* (2004.0) **19** 433-446. DOI: 10.1177/0115426504019005433
15. Böhm A., Heitmann B.L.. **The use of bioelectrical impedance analysis for body composition in epidemiological studies**. *Eur. J. Clin. Nutr.* (2013.0) **67** S79-S85. DOI: 10.1038/ejcn.2012.168
16. Volpe S.L., Melanson E.L., Kline G.. **Validation of bioelectrical impedance analysis to hydrostatic weighing in male body builders**. *Acta Diabetol.* (2010.0) **47** 55-58. DOI: 10.1007/s00592-009-0098-3
17. Stewart A.D., Marfell-Jones M.J., Olds T., de Ridder J.H.. *International Standards for Anthropometric Assessment* (2011.0)
18. Kyle U.G., Bosaeus I., De Lorenzo A.D., Deurenberg P., Elia M., Gómez J.M., Heitmann B.L., Kent-Smith L., Melchior J.-C., Pirlich M.. **Bioelectrical impedance analysis—Part II: Utilization in clinical practice**. *Clin. Nutr.* (2004.0) **23** 1430-1453. DOI: 10.1016/j.clnu.2004.09.012
19. Hoeyer J.L., Hansen H.L.. **Obesity among Danish seafarers**. *Int. Marit. Health* (2005.0) **56** 48-55. PMID: 16532584
20. Hansen H.L., Hjarnoe L., Jepsen J.R.. **Obesity continues to be a major health risk for Danish seafarers and fishermen**. *Int. Marit. Health* (2011.0) **62** 98-103. PMID: 21910112
21. Pougnet R., Pougnet L., Loddé B.L., Canals-Pol M.L., Jegaden D., Lucas D., Dewitte J.-D.. **Cardiovascular risk factors in seamen and fishermen: Review of literature**. *Int. Marit. Health* (2013.0) **64** 107-113. PMID: 24072535
22. Rębacz-Maron E.. **Selected anthropometric indices of maritime university students**. *Int. Marit. Health* (2015.0) **66** 145-151. DOI: 10.5603/IMH.2015.0029
23. Jepsen J.R., Rasmussen H.B.. **The metabolic syndrome among Danish seafarers: A follow-up study**. *Int. Marit. Health* (2016.0) **67** 129-136. DOI: 10.5603/IMH.2016.0025
24. Nittari G., Tomassoni D., Di Canio M., Traini E., Pirillo I., Minciacchi A., Amenta F.. **Overweight among seafarers working on board merchant ships**. *BMC Public Health* (2019.0) **19**. DOI: 10.1186/s12889-018-6377-6
25. Sagaro G.G., Di Canio M., Amenta F.. **Correlation between body mass index and blood pressure in seafarers**. *Clin. Exp. Hypertens.* (2021.0) **43** 189-195. DOI: 10.1080/10641963.2020.1836193
26. Masood M., Aggarwal A., Reidpath D.D.. **Effect of national culture on BMI: A multilevel analysis of 53 countries**. *BMC Public Health* (2019.0) **19**. DOI: 10.1186/s12889-019-7536-0
27. **Worldwide Trends in Body-Mass Index, Underweight, Overweight, and Obesity**
28. Tannir H., Itani L., Kreidieh D., El Masri D., El Ghoch M.. **Can Intentional Weight Loss Ameliorate Sarcopenia in Individuals with Obesity? A Longitudinal Interventional Study**. *Clin. Pract.* (2022.0) **12** 106-112. DOI: 10.3390/clinpract12010014
29. Lee L.-C., Hsieh K.-C., Wu C.-S., Chen Y.-J., Chiang J.. **Validity of Standing Posture Eight-electrode Bioelectrical Impedance to Estimate Body Composition in Taiwanese Elderly**. *Int. J. Gerontol.* (2014.0) **8** 137-142. DOI: 10.1016/j.ijge.2013.08.010
30. Kelly J., Metcalfe J.. **Validity and reliability of body composition analysis using the tanita BC418-MA Heart Rate Variability in the Detection and Prevention of Overtraining View project Adventure experiences and resilience View project**. *J. Exerc. Physiol. Online* (2012.0) **15** 74-84
31. Verney J., Schwartz C., Amiche S., Pereira B., Thivel D.. **Comparisons of a Multi-Frequency Bioelectrical Impedance Analysis to the Dual-Energy X-Ray Absorptiometry Scan in Healthy Young Adults Depending on their Physical Activity Level**. *J. Hum. Kinet.* (2015.0) **47** 73-80. DOI: 10.1515/hukin-2015-0063
32. Bankovic V., Dopsaj M., Terzic Z., Nesic G.. **Descriptive Body Composition Profile in Female Olympic Volleyball Medalists Defined Using Multichannel Bioimpedance Measurement: Rio 2016 Team Case Study**. *Int. J. Morphol.* (2018.0) **36** 699-708. DOI: 10.4067/S0717-95022018000200699
33. Øvretveit K.. **Anthropometric and Physiological Characteristics of Brazilian Jiu-Jitsu Athletes**. *J. Strength Cond. Res.* (2018.0) **32** 997-1004. DOI: 10.1519/JSC.0000000000002471
34. Herttua K., Ahrenfeldt L.J., Paljarvi T.. **Risk of major chronic diseases in transport, rescue and security industries: A longitudinal register-based study**. *Occup. Environ. Med.* (2022.0) **79** 162-168. DOI: 10.1136/oemed-2021-107764
35. Mancuso P., Bouchard B.. **The Impact of Aging on Adipose Function and Adipokine Synthesis**. *Front. Endocrinol.* (2019.0) **10** 137. DOI: 10.3389/fendo.2019.00137
36. Smee D.J., Walker A., Rattray B., Cooke J.A., Serpell B.G., Pumpa K.L.. **Comparison of Body Composition Assessment Methods in Professional Urban Firefighters**. *Int. J. Sport Nutr. Exerc. Metab.* (2019.0) **29** 282-288. DOI: 10.1123/ijsnem.2018-0040
37. Jepsen J.R., Zhao Z., Van Leeuwen W.M.. **Seafarer fatigue: A review of risk factors, consequences for seafarers’ health and safety and options for mitigation**. *Int. Marit. Health* (2015.0) **66** 106-117. DOI: 10.5603/IMH.2015.0024
38. Mattis J., Sehgal A.. **Circadian Rhythms, Sleep, and Disorders of Aging**. *Trends Endocrinol. Metab.* (2016.0) **27** 192-203. DOI: 10.1016/j.tem.2016.02.003
39. Mander B.A., Winer J.R., Walker M.P.. **Sleep and Human Aging**. *Neuron* (2017.0) **94** 19-36. DOI: 10.1016/j.neuron.2017.02.004
|
---
title: Normative Values and Psychometric Properties of EQ-5D-Y-3L in Chilean Youth
Population among Different Weight Statuses
authors:
- Miguel Angel Perez-Sousa
- Pedro R. Olivares
- Rocio Carrasco-Zahinos
- Antonio Garcia-Hermoso
journal: International Journal of Environmental Research and Public Health
year: 2023
pmcid: PMC10002306
doi: 10.3390/ijerph20054096
license: CC BY 4.0
---
# Normative Values and Psychometric Properties of EQ-5D-Y-3L in Chilean Youth Population among Different Weight Statuses
## Abstract
Background: This study aimed to provide population norms among children and adolescents in Chile using the EQ-5D-Y-3L questionnaire and to examine its feasibility and validity among body weight statuses. Methods: This was a cross-sectional study in which 2204 children and adolescents (aged 8–18 years) from Chile completed a set of questionnaires providing sociodemographic, anthropometric and health-related quality of life (HRQoL) data using the five EQ-5D-Y-3L dimensions and its visual analogue scale (EQ-VAS). Descriptive statistics of the five dimensions and the EQ-VAS were categorized into body weight status groups for the EQ-5D-Y-3L population norms. The ceiling effect, feasibility and discriminant/convergent validity of the EQ-5D-Y-3L were tested. Results: The dimensions of the EQ-5D-Y-3L questionnaire presented more ceiling effects than the EQ-VAS. The validity showed that the EQ-VAS could discriminate among body weight statuses. However, the EQ-5D-Y-3L index (EQ-Index) demonstrated a non-acceptable discriminant validity. Furthermore, both the EQ-Index and the EQ-VAS presented an acceptable concurrent validity among weight statuses. Conclusions: The normative values of the EQ-5D-Y-3L indicated its potential use as a reference for future studies. However, the validity of the EQ-5D-Y-3L for comparing the HRQoL among weight statuses could be insufficient.
## 1. Introduction
Health-related quality of life (HRQoL) has been defined as a multidimensional concept beyond somatic indicators, including physical, psychological, social and functional aspects of self-assessment of the individual’s health [1]. The increases in chronic illness in children and adolescents [2] have framed HRQoL assessment as of significant interest to public health. This fact was indicated by the US Food and Drug Administration and the pharmaceutical industry, who recognize the need for assessing HRQoL in pediatric and adolescent patients to determine the effects of pharmacological treatments to complete the biomedical perspective [3].
HRQoL is measured via self-report or proxy report from a standardized questionnaire that includes different dimensions. The questionnaire provides a generic health perception allowing comparisons between different populations and conditions and also an econometric result that could be used in cost–utility analysis for economic evaluation [4].
The main HRQoL questionnaires for children and adolescents, including the PedsQL [5], Kidscreen [6], and EQ-5D-Y-3L [7], have culturally adapted their versions for most countries.
The EQ-5D-Y-3L is a widely used questionnaire with five dimensions of health (“mobility,” “looking after myself,” “doing usual activities,” “having pain or discomfort,” and “feeling worried, sad or unhappy”) and three levels of response indicating the severity of health problems in the participant, providing 243 possible health states [8]. The EQ-5D-Y-3L has been translated and adapted to Spanish and presents acceptable validity and reliability [7]. This questionnaire is also used in Latin American countries [9], but to our knowledge, there are scant normative data for this region.
Within the broad spectrum of childhood diseases, obesity takes up a prominent position due to its prevalence and effects on physical and psychological health [10,11]. One of the principal components of chronic illness in children and adolescents living in Latin American countries is overweight and obesity, which has grown continuously in the last decade [12]. In this respect, previous studies have shown an inverse relationship between body mass index (BMI) and HRQoL. For example, Perez-Sousa et al. [ 13] found that overweight and obese Spanish children showed a lower HRQoL than their normal-weight counterparts. Garcia-Rubio et al. [ 14] showed that overweight and obese children and adolescents had a reduced HRQoL compared to healthy children in a cross-sectional study carried out in Chile. However, several studies have presented a muddled relationship between excess body weight and HRQoL. For example, Petersen et al. [ 15] found a similar HRQoL in children with obesity and normal weight, and Liu et al. [ 16] only found a lower HRQoL for the social dimension in overweight/obese children compared with healthy-weight children after controlling for gender, age, school type, parental education and family income. *In* general terms, the studies emphasize that the lack of differences found may be due to cultural and/or socioeconomic characteristics. However, we hypothesize that the questionnaire cannot discern different health perceptions between weight status due to a lack of knowledge on performance regarding psychometric properties of the questionnaire in these subgroups.
Population norms are essential to characterize the study population, interpret research results, and compare studies. Furthermore, this action allows comparison of results from the general population or people with specific health characteristics in order to develop primary physician care standards [17]. However, Chile lacks studies on normative values of HRQoL in children and adolescents from general and specific populations using the EQ-5D-Y-3L questionnaire. Thus, based on the current evidence and the importance of screening for HRQoL within children and adolescents, we aimed to provide normative population values for HRQoL and examine the feasibility and convergent/discriminant validity among Chilean children and adolescents with different weight status using the EQ-5D-Y-3L.
## 2.1. Study Design and Participants
A cross-sectional analysis was conducted using data collected from 2204 Chilean children and adolescents aged 8–18 years from the general population. We recruited 3150 participants from primary and secondary schools in Chile and 2204 of these finally agreed to participate in the interviews. We requested the participation of eight schools (four primary and four secondary), with each providing access to four or five sections of different grades. According to the design, participants who met the following inclusion criteria formed our target group: children and adolescents aged 8–18 years; knowledge of the Spanish language; present on the day of the test; and gave their informed consent (subjects and parents or legal tutors).
Before data collection, the parents were informed of the methodology and objectives of the study via an official letter written by the researchers that included an informed consent form. The study was approved by University of Santiago Ethics Committee (code 938).
## 2.2. Procedure
The data were collected by two experienced research group members using direct administration in small groups (10–12 children per group). The survey duration varied from 5 min for children aged 8–12 years to 3 min for students aged 13–18 years. Each respondent was assigned a code for confidentiality and to facilitate data analysis. A phone number and email address were provided to respondents to address any concerns that may arise at any time.
## 2.3.1. Sociodemographic Information
A core set of questions on essential sociodemographic characteristics (age, gender and year of schooling) and HRQoL and subjective health measures were included. For anthropometric data, weight and height were assessed with the participants standing barefoot in minimal clothing. The instrument used was a Seca 769 (Seca, Hamburg, Germany) scale with a portable Seca 220 stadiometer (accuracy of 0.1 cm; Seca, Hamburg, Germany) placed on a rigid wall. BMI was calculated as the body weight divided by the squared height (kg/m2). Individuals were classified into four categories according to their BMI as follows: [0] underweight, [1] normal weight, [3] overweight and [4] obese, as indicated by Cole et al. [ 18].
## 2.3.2. Health-Related Quality of Life
The EuroQol group developed a tool with five dimensions (the EQ-5D) to quantify HRQoL. The dimensions are mobility, self-care, usual activities, pain or discomfort and anxiety or depression. The instrument also includes a visual analogue scale (EQ-VAS), which is anchored at 100 (best imaginable health) and 0 (worst imaginable health). Most recently, the EuroQol group implemented a version for children and adolescents between the ages of 8 and 18 years, called the EQ-5D-Y-3L [7]. The five questions are whether children have problems with walking, looking after themselves, doing their usual activities, have pain or discomfort and feel worried, sad or unhappy, to which they could respond with “no problems,” “some problems” and “a lot of problems.” The EQ-5D-Y-3L offers a state of health that can be converted into a unique index (EQ-Index) by applying a formula that attributes different weights to each dimension’s levels. The anchor points or references of the questionnaire are 0 (death) and 1 (perfect health). We used the formula to assess adult health status in Spain [19]. This procedure has already been applied in similar studies [20,21]. The reliability and validity of the Spanish version of the EQ-5D-Y-3L has been confirmed [7] and the EQ-VAS allows subjects to assess their health status from 0 (worst) to 100 (best).
## 2.3.3. Statistical Analysis
A descriptive analysis using the means ± standard deviation (SD) for continuous variables and frequency distribution for categorical variables was used to obtain the characteristics of the sample.
## Population Norms
The EQ-5D-Y-3L population norms were derived from the data given by the general population sample. Analysis of the EQ-5D-Y-3L population norms followed the standardized method recommended by the EuroQol group [22].
## Feasibility
We computed the proportion of children not answering to a few (i.e., partially incomplete questionnaire) or all dimensions (i.e., incomplete questionnaire) of the EQ-5D-Y-3L.
## Ceiling Effect
The proportions of children reporting “no problems” were calculated for each descriptive system dimension. We also computed the children reporting “no problems” ratio in all five dimensions [11111]. We hypothesized that normal-weight children would report a higher ceiling effect than their counterparts.
## Discriminant and Convergent Validity
The discriminant validity of the EQ-5D-Y-3L was examined by comparing the HRQoL profiles of the different weight status groups (underweight, normal weight, overweight and obesity). The level of problems reported in each EQ-5D-Y-3L dimension per group was compared using Fisher’s exact test rather than the chi-square test because some cells were sparsely populated. Post hoc analysis using the Kruskal–Wallis H test indicated which groups were significantly different from each other. Following studies in overweight and obese children [23,24], we assumed that complaints of health problems would be more common among underweight, overweight and obese children and that these individuals would therefore have lower scores on the EQ-5D-Y-3L dimensions and EQ-VAS than normal-weight children. The convergent validity of the EQ-5D-Y-3L was examined by correlating the EQ-Index with the EQ-VAS through Spearman’s rho correlation. The correlation coefficient (ρs) was interpreted as follows: small, 0.10–0.29; moderate, ≥0.30–0.49; strong, ≥0.50 [25].
Convergent validity is the ability of the scores to correlate with other measures that assess a similar construct. In contrast, discriminant validity examines the relationships of scores obtained from similar but different constructs [25].
## 3. Results
Table 1 shows the characteristics of the general population sample. Overall, a total of 2204 children and adolescents responded to the set of questions in the EQ-5D-Y-3L. The sample distribution was higher for females (1313 ± $59.6\%$) than males (891 ± $40.4\%$). The proportions among weight status groups were dissimilar, with the majority of respondents in the normal-weight group ($43.5\%$). The mean ± SD of the EQ-Index by gender, age group and weight status group are also presented.
The frequency of reported problems by weight status group is shown in Table 2. Fisher’s exact and Kruskal–*Wallis analysis* showed nonsignificant differences ($p \leq 0.05$) in the distribution of problems for each dimension of the EQ-5D-Y-3L; therefore, there were no differences in problems reported for HRQoL among underweight, normal-weight, overweight and obese children over the EQ-5D-Y-3L dimensions. Thus, the discriminant validity of the descriptive system appeared to be lower and was unable to discern problems among children and adolescents with different weight status. In contrast, there were statistically significant differences in HRQoL reported on the EQ-VAS among all weight status groups. The ceiling effect (no problems reported) was relatively higher in the physical dimensions (mobility; looking after myself; doing usual activities), whereas the psychological dimensions (having pain or discomfort; feeling worried, sad or unhappy) showed a lower ceiling effect in all groups.
Finally, convergent validity was examined (Table 3). Spearman’s rho test showed a significant correlation ($p \leq 0.001$) between all dimensions in all groups and for the EQ-VAS, with the exception of the “looking after myself” and “feeling worried, sad or unhappy” dimensions in the overweight group. The magnitude of the correlation was low in all dimensions and groups, except for “mobility,” “doing usual activities” and “feeling worried, sad or unhappy” in the underweight group.
## 4. Discussion
This study has provided population norms for the EQ-5D-Y-3L questionnaire by using a representative sample of Chilean children and adolescents ($$n = 2204$$) and has demonstrated the psychometric properties in terms of feasibility and discriminant/convergent validity to determine the instrument’s ability to discern health states among weight status groups.
A strength of this study’s EQ-5D-Y-3L population norms was the neutral context sample with the responses pooled across different weight statuses. To date, this is the first study to present normative data in Chilean children and adolescents using the EQ-5D-Y-3L questionnaire. Other studies in Europe [7] or North America [26] have been conducted in the general population.
The main findings of this study were that the Spanish version of the EQ-5D-Y-3L is a feasible instrument to assess HRQoL in the Chilean population because there were no missing values. The results are consistent with previous research, including a multinational study performed to analyze the validity and reliability of the EQ-5D-Y-3L [7]. Our study identified a higher ceiling effect on the physical dimensions (mobility; looking after myself; doing usual activities) and a lower effect on the psychological dimensions (having pain or discomfort; feeling worried, sad or unhappy) in all groups. This ceiling effect is similar to previous studies in the general population [7]. Furthermore, a previous study that used the EQ-5D-Y-3L with overweight and obese children reported few problems in the majority of dimensions, except for anxiety/depression [27].
Another finding in this study was the scarce discriminant validity of the descriptive system of the EQ-5D-Y-3L between health states across weight status. There were no significant differences in the distribution of problems in each dimension among underweight, normal weight, overweight and obese children. These results are similar to previous studies [7,27,28]. In contrast, we found several reviews that analyzed how overweight and obesity affect children’s HRQoL [29,30]. However, the questionnaires that assessed HRQoL were Kidscreen, PedsQL and KINDL-R. These questionnaires are based on 5–7 levels of response, whereas the EQ-5D-Y-3L only has three levels of response. Moreover, we found other studies where the score from PedsQL or Kidscreen discriminates significant health status among weight status groups. The score for these questionnaires is based on a scale of 0–100, whereas the EQ-5D-Y-3L dimensions are based on a score of 1–3. Nevertheless, our study found that the EQ-VAS was discerned among health states across weight groups. This finding suggests that a scale such as the EQ-VAS, based on 0–$100\%$, may be more accurate in identifying health states than a descriptive system based on three levels of response. This low discriminant validity of the descriptive dimension may be due, first, to the high ceiling effect of this instrument. Second, there is the effect of non-dimensionality of the EQ-VAS, with the descriptive dimension and the EQ-VAS starting from a different scale: the descriptive system is based on five dimensions of the state of health and the EQ-VAS as a percentage of the state of health compared to the best imaginable. Therefore, the EQ-VAS can cover as many different dimensions of health as the respondent interprets, all reduced to a single value. Third, several studies indicate that the response in the EQ-VAS is influenced not only by health status but also by personal characteristics such as age, gender, education and race [31,32,33]. Expanding the severity levels in the EQ-5D-Y-3L can reduce the instrument’s ceiling effects and enhance sensitivity, especially in milder health conditions [34]. Thus, it is probable that discriminant validity will be better using the new EQ-5D-Y-3L-5L instrument [35,36].
The convergent validity of the EQ-5D-Y-3L dimensions for each weight status group showed a significant association with the EQ-VAS, but the magnitude of correlation in general was low. Thus, these results should be considered with caution.
Our study has certain limitations. We did not collect information concerning comorbidities or include other populations, such as hospitalized children or those with chronic diseases. These factors need to be considered when applying normative data in other groups or individuals. Furthermore, this study was observational; thus, we might have missed some confounders. Additionally, the method was self-administration, whereas other studies apply proxy administration. Another limitation is the low prevalence of severity of health problems captured by the instruments used. Although children with overweight or obesity have a lower HRQoL than children with a healthy weight [24], the baseline state of their HRQoL using currently available instruments and assessments starts from a high level, which limits the capture of possible improvements. This fact is determined by the ceiling effect of the questionnaire, which can be observed in the proportion of individuals with severe or large HRQoL problems. Likewise, this ceiling effect has been reported in the EQ-5D-EL-Y with studies on individuals without severe health problems [1]. In fact, the EuroQol group is developing a version of the questionnaire with five response levels (EQ-5D-5L-Y) to obtain greater scaling in certain populations. Therefore, our results should be considered with caution.
The study strengths include the large sample: 2204 Chilean children and adolescents. Moreover, the results of this study provide better understanding and use of the EQ-5D-Y-3L questionnaire in children with obesity and help in deciding whether to use this questionnaire over another and how to interpret the results.
## 5. Conclusions
The Chilean population norms for the EQ-5D-Y-3L reported here can be used as reference values when comparing different weight status groups. Furthermore, the study confirmed its feasibility even though the convergent and discriminant validity of the EQ-5D-Y-3L was insufficient. Consequently, we recommend that the results of future studies using the EQ-5D-Y-3L on children with heterogeneous weight status should be interpreted with caution.
## References
1. Ravens-Sieberer U., Erhart M., Wille N., Wetzel R., Nickel J., Bullinger M.. **Generic health-related quality of life assessment in children and adolescents: Methodological considerations**. *Pharmacoeconomics* (2006.0) **24** 1199-1220. DOI: 10.2165/00019053-200624120-00005
2. Vos T., Barber R.M., Bell B., Bertozzi-Villa A., Biryukov S., Bolliger I., Charlson F., Davis A., Degenhardt L., Dicker D.. **Global Burden of Disease Study 2013 Collaborators, G.B. of D.S. 2013 Global, regional, and national incidence, prevalence, and years lived with disability for 301 acute and chronic diseases and injuries in 188 countries, 1990–2013: A systematic analysis for the Global Burden of Disease Study 2013**. *Lancet* (2015.0) **386** 743-800. DOI: 10.1016/S0140-6736(15)60692-4
3. Matza L.S., Swensen A.R., Flood E.M., Secnik K., Leidy N.K.. **Assessment of health-related quality of life in children: A review of conceptual, methodological, and regulatory issues**. *Value Health* (2004.0) **7** 79-92. DOI: 10.1111/j.1524-4733.2004.71273.x
4. Lizán T.L.. **La calidad de vida relacionada con la salud**. *Aten. Primaria* (2009.0) **41** 411. DOI: 10.1016/j.aprim.2008.10.019
5. Varni J.W., Burwinkle T.M., Seid M., Skarr D.. **The PedsQL 4.0 as a pediatric population health measure: Feasibility, reliability, and validity**. *Ambul. Pediatr.* (2003.0) **3** 329-341. DOI: 10.1367/1539-4409(2003)003<0329:TPAAPP>2.0.CO;2
6. Ravens-Sieberer U., Gosch A., Rajmil L., Erhart M., Bruil J., Power M., Duer W., Auquier P., Cloetta B., Czemy L.. **The KIDSCREEN-52 quality of life measure for children and adolescents: Psychometric results from a cross-cultural survey in 13 European countries**. *Value Health* (2008.0) **11** 645-658. DOI: 10.1111/j.1524-4733.2007.00291.x
7. Ravens-Sieberer U., Wille N., Badia X., Bonsel G., Burström K., Cavrini G., Devlin N., Egmar A.C., Gusi N., Herdman M.. **Feasibility, reliability, and validity of the EQ-5D-Y: Results from a multinational study**. *Qual. Life Res.* (2010.0) **19** 887-897. DOI: 10.1007/s11136-010-9649-x
8. Van Reenen M., Janssen B., Oppe M., Kreimeier S., Greiner W.. *EQ-5D-Y User Guide Basic Information on How to Use the EQ-5D-Y Instrument* (2014.0)
9. Gaitán-López D.F., Correa-Bautista J.E., Vinaccia S., Ramírez-Vélez R.. **Self-report health-related quality of life among children and adolescents from Bogotá, Colombia. The FUPRECOL study**. *Colomb. Med.* (2017.0) **48** 12-18. DOI: 10.25100/cm.v48i1.2071
10. Ogden C.L., Carroll M.D., Lawman H.G., Fryar C.D., Kruszon-Moran D., Kit B.K., Flegal K.M.. **Trends in Obesity Prevalence Among Children and Adolescents in the United States, 1988–1994 Through 2013–2014**. *JAMA* (2016.0) **315** 2292-2299. DOI: 10.1001/jama.2016.6361
11. Tsiros M.D., Coates A.M., Howe P.R.C., Grimshaw P.N., Buckley J.D.. **Obesity: The new childhood disability?**. *Obes. Rev.* (2011.0) **12** 26-36. DOI: 10.1111/j.1467-789X.2009.00706.x
12. Ng M., Fleming T., Robinson M., Thomson B., Graetz N., Margono C., Mullany E.C., Biryukov S., Abbafati C., Abera S.F.. **Global, regional, and national prevalence of overweight and obesity in children and adults during 1980–2013: A systematic analysis for the Global Burden of Disease Study 2013**. *Lancet* (2014.0) **384** 766-781. DOI: 10.1016/S0140-6736(14)60460-8
13. Perez-Sousa M.A., Olivares P.R., Escobar-Alvarez J.A., Parraça J.A., Gusi N.. **Fitness as mediator between weight status and dimensions of health-related quality of life**. *Health Qual. Life Outcomes* (2018.0) **16** 155. DOI: 10.1186/s12955-018-0981-0
14. García-Rubio J., Olivares P.R., Lopez-Legarrea P., Gómez-Campos R., Cossio-Bolaños M.A., Merellano-Navarro E.. **Association between health related quality of life, bodyweight status (bmi) and physical activity and fitness levels in chilean adolescents**. *Nutr. Hosp.* (2015.0) **32** 1695-1702. DOI: 10.3305/nh.2015.32.4.9182
15. Petersen S., Moodie M., Mavoa H., Waqa G., Goundar R., Swinburn B.. **Relationship between overweight and health-related quality of life in secondary school children in Fiji: Results from a cross-sectional population-based study**. *Int. J. Obes.* (2014.0) **38** 539-546. DOI: 10.1038/ijo.2013.212
16. Liu W., Lin R., Liu W., Guo Z., Xiong L., Li B., Cheng K.K., Adab P., Pallan M.. **Relationship between weight status and health-related quality of life in Chinese primary school children in Guangzhou: A cross-sectional study**. *Health Qual. Life Outcomes* (2016.0) **14** 166. DOI: 10.1186/s12955-016-0567-7
17. O’Connor P.J.. **Normative data: Their definition, interpretation, and importance for primary care physicians**. *Fam. Med.* (1990.0) **22** 307-311. PMID: 2200734
18. Cole T.J., Bellizzi M.C., Flegal K.M., Dietz W.H.. **Establishing a standard definition for child overweight and obesity worldwide: International survey**. *BMJ* (2000.0) **320** 1240-1243. DOI: 10.1136/bmj.320.7244.1240
19. Szende A., Oppe M., Devlin N.J., Nancy J.. *EQ-5D Value Sets: Inventory, Comparative Review, and User Guide* (2007.0)
20. Bouwmans C., van der Kolk A., Oppe M., Schawo S., Stolk E., van Agthoven M., Buitelaar J., van Roijen L.H.. **Validity and responsiveness of the EQ-5D and the KIDSCREEN-10 in children with ADHD**. *Eur. J. Health Econ.* (2013.0) **15** 967-977. DOI: 10.1007/s10198-013-0540-x
21. Cheung P.W.H., Wong C.K.H., Lau S.T., Cheung J.P.Y.. **Responsiveness of the EuroQoL 5-dimension (EQ-5D) in adolescent idiopathic scoliosis**. *Eur. Spine J.* (2018.0) **27** 278-285. DOI: 10.1007/s00586-017-5330-1
22. Szende A., Janssen B., Cabasés J.. *Self-Reported Population Health: An International Perspective Based on EQ-5D* (2014.0) **Volume 212** 1-196. DOI: 10.1007/978-94-007-7596-1
23. **SPSS Statistics Tutorials and Statistical Guides Laerd Statistics**
24. Pinhas-Hamiel O., Singer S., Pilpel N., Fradkin A., Modan D., Reichman B.. **Health-related quality of life among children and adolescents: Associations with obesity**. *Int. J. Obes.* (2006.0) **30** 267-272. DOI: 10.1038/sj.ijo.0803107
25. Tsiros M.D., Olds T., Buckley J.D., Grimshaw P., Brennan L., Walkley J., Hills A.P., Howe P.R.C., Coates A.M.. **Health-related quality of life in obese children and adolescents**. *Int. J. Obes.* (2009.0) **33** 387-400. DOI: 10.1038/ijo.2009.42
26. Wu X.Y., Ohinmaa A., Veugelers P.J.. **Sociodemographic and neighbourhood determinants of health-related quality of life among grade-five students in Canada**. *Qual. Life Res.* (2010.0) **19** 969-976. DOI: 10.1007/s11136-010-9663-z
27. Burströ K., Svartengren M., Egmar A.-C.. **Testing a Swedish child-friendly pilot version of the EQ-5D instrument-initial results**. *Eur. J. Public Health* (2011.0) **21** 178-183. DOI: 10.1093/eurpub/ckq042
28. Gusi N., Perez-Sousa M.A., Gozalo-Delgado M., Olivares P.R.. **Validity and reliability of the spanish EQ-5D-Y proxy version**. *An. Pediatr.* (2014.0) **81** 212-219. DOI: 10.1016/j.anpedi.2013.11.028
29. Buttitta M., Iliescu C., Rousseau A., Guerrien A.. **Quality of life in overweight and obese children and adolescents: A literature review**. *Qual. Life Res.* (2014.0) **23** 1117-1139. DOI: 10.1007/s11136-013-0568-5
30. Sanders R.H., Han A., Baker J.S., Cobley S.. **Childhood obesity and its physical and psychological co-morbidities: A systematic review of Australian children and adolescents**. *Eur. J. Pediatr.* (2015.0) **174** 715-746. DOI: 10.1007/s00431-015-2551-3
31. Whynes D.K.. **Correspondence between EQ-5D health state classifications and EQ VAS scores**. *Health Qual. Life Outcomes* (2008.0) **6** 94. DOI: 10.1186/1477-7525-6-94
32. Chapman B., Franks P., Duberstein P., Care A.J.-M.. **Differences between individual and societal health state valuations: Any link with personality?**. *Med. Care* (2009.0) **47** 902-907. DOI: 10.1097/MLR.0b013e3181a8112e
33. Perneger T.V., Combescure C., Courvoisier D.S.. **General Population Reference Values for the French Version of the EuroQol EQ-5D Health Utility Instrument**. *Value Health* (2010.0) **13** 631-635. DOI: 10.1111/j.1524-4733.2010.00727.x
34. Herdman M., Gudex C., Lloyd A., Janssen M., Kind P., Parkin D., Bonsel G., Badia X.. **Development and preliminary testing of the new five-level version of EQ-5D (EQ-5D-5L)**. *Qual. Life Res.* (2011.0) **20** 1727-1736. DOI: 10.1007/s11136-011-9903-x
35. Kreimeier S., Åström M., Burström K., Egmar A.C., Gusi N., Herdman M., Kind P., Perez M.A., Wolfgang S.. **EQ-5D-Y-5L: Developing a revised EQ-5D-Y with increased response categories**. *Qual. Life Res.* (2019.0) **28** 1951-1961. DOI: 10.1007/s11136-019-02115-x
36. Pérez-Sousa M.Á., Olivares P.R., Ramírez-Vélez R., Gusi N.. **Comparison of the Psychometric Properties of the EQ-5D-Y-3L and EQ-5D-5L-Y Instruments in Spanish Children and Adolescents**. *Value Health* (2021.0) **24** 1799-1806. DOI: 10.1016/j.jval.2021.07.006
|
---
title: 'A Digital Tool for Measuring Healing of Chronic Wounds Treated with an Antioxidant
Dressing: A Case Series'
authors:
- Inés María Comino-Sanz
- Rafael Cabello Jaime
- Josefina Arboledas Bellón
- Juan Francisco Jiménez-García
- Mercedes Muñoz-Conde
- María José Díez Requena
- Francisco Javier García Díaz
- Begoña Castro
- Pedro Luis Pancorbo-Hidalgo
journal: International Journal of Environmental Research and Public Health
year: 2023
pmcid: PMC10002323
doi: 10.3390/ijerph20054147
license: CC BY 4.0
---
# A Digital Tool for Measuring Healing of Chronic Wounds Treated with an Antioxidant Dressing: A Case Series
## Abstract
[1] Abstract: Wound monitoring is an essential aspect in the evaluation of wound healing. This can be carried out with the multidimensional tool HELCOS, which develops a quantitative analysis and graphic representation of wound healing evolution via imaging. It compares the area and tissues present in the wound bed. This instrument is used for chronic wounds in which the healing process is altered. This article describes the potential use of this tool to improve the monitoring and follow-up of wounds and presents a case series of various chronic wounds with diverse etiology treated with an antioxidant dressing. [ 2] Methods: A secondary analysis of data from a case series of wounds treated with an antioxidant dressing and monitored with the HELCOS tool. [ 3] Results: The HELCOS tool is useful for measuring changes in the wound area and identifying wound bed tissues. In the six cases described in this article, the tool was able to monitor the healing of the wounds treated with the antioxidant dressing. [ 4] Conclusions: the monitoring of wound healing with this multidimensional HELCOS tool offers new possibilities to facilitate treatment decisions by healthcare professionals.
## 1. Introduction
A chronic wound, also called a hard-to-heal wound, has been defined as any wound that has not healed by 40–$50\%$ after four weeks of appropriate treatment [1]. Several factors can delay the physiological process of healing, including oxygenation, infection, diabetes, medications, stress, nutrition, hormones, and age [2].
When assessing the wound healing process, clinicians often face the problem of reliably measuring wound size. Wound measurement is important for monitoring the healing process of chronic wounds and to evaluate the effect of treatments. This is a practical problem as most of the measures used are subjective and based on the clinical experience of the professional.
In the last few decades, technological advances have led to the development of several accurate methods and multidimensional tools for wound monitoring: manual or digital planimetry, simple ruler method, mathematical models, digital imaging, or more recently three-dimensional (3D) [3]. As a result, wound monitoring is more objective and allows the identification of different parameters and variables through a specific analysis.
One of the multidimensional assessment tools recently developed is the HELCOS software, a web-based integrated wound management system that allows the measurement of different wound parameters through digital analysis of images of the wounds [4]. HELCOS was designed and developed between 2015 and 2017 through a project funded by the Spanish Pressure Ulcer and Chronic Wound Advisory Group. This tool has been available free of charge since 2017 for clinicians working in clinical settings; only a short registration is required. There are no special hardware requirements to use this tool, only a computer or device connected to the Internet, so it can be used directly in any environment (hospital, wound clinic, primary care). All personal data security standards are guaranteed; each professional can only access his/her own cases. To perform a wound analysis, the clinician has to obtain an image of the wound with any type of camera or device. Good lighting conditions are highly recommended, taking the picture at 20 to 30 cm, perpendicular to the wound plane and placing a size test of known diameter close to the wound (such as a blue circle 2 cm in diameter). Photos can be uploaded directly from a smartphone or using a computer.
HELCOS allows clinicians to measure the wound area and the proportion of the wound bed covered with granulation or necrotic tissue. We have tested this tool in a series of wound cases treated with an antioxidant dressing.
It is known that wound healing is impaired when the wound remains in the inflammatory stage for too long [5]. Oxidative stress is among the factors that can delay the healing process [2]. Reactive oxygen species (ROS) are small oxygen-derived molecules that play a crucial role in the preparation of the normal wound healing response [6]. Therefore, a suitable balance between the levels of ROS is essential. A wound with a low level of ROS protects tissues against infection and stimulates effective wound healing by promoting cell survival [7,8], whereas if there is excess ROS in the wound, the cells are damaged with pro-inflammatory status and produce oxidative stress [9].
Therefore, the use of antioxidant compounds for wound treatment is increasing and has excellent potential for clinical use. Antioxidant dressings that regulate this balance are a target for new therapies [10,11]. Among these new advanced products is the antioxidant dressing Reoxcare® [12], developed by Histocell (Bizkaia, Spain). This product combines an absorbent matrix obtained from the locust bean gum galactomannan, of plant-based origin, with an antioxidant hydration solution with curcumin and N Acetylcysteine (NAC) [13].
Curcumin is a natural phenol extracted from the *Curcuma longa* rhizome. It has anti-inflammatory, antibacterial, and antioxidant properties, which improve wound healing [14]. NAC is an antioxidant molecule that plays an important role in regulating redox status [15]. The three components act synergistically, giving the product a potent antioxidant activity. Due to the innovative design, this antioxidant dressing combines the advantages of moist healing in exudate management and free radical neutralization, achieving wound reactivation.
This antioxidant dressing was tested in different studies. In vitro studies and animal wound healing models have shown that this product modulates the inflammatory phase of wound healing, controlling the excessive cell activation and allowing a more orderly transition between the inflammatory, proliferative, and remodeling phases of wound healing [13]. A multicenter, prospective-case study series revealed that this dressing can be applied to wounds independently of their level of recurrence or severity, effectively eliminating the biofilm and facilitating the progression of the wound out of the inflammatory phase [16,17]. These findings suggest that the dressing could be a new advanced alternative for managing hard-to heal wounds. In other words, the value of antioxidant dressing in the management has been reported and shown positive results.
The purpose of this article is to describe the potential use of a web-based wound measurement system (HELCOS) for monitoring the progress of wound healing in a case series of wounds.
## 2.1. Study Design
This consists of a secondary analysis of a case series from the intervention group of the main study. This is a descriptive design of healing monitoring using the HELCOS tool.
The main study is a prospective intervention study with two arms, intervention (antioxidant dressing) and comparison (usual care with moist dressing) [18]. Advanced practice wound nurses recruited patients with chronic wounds in three primary health care centers in the Andalusian Health Service in Spain between September 2019 and October 2021.
The main study included 54 patients (28 intervention group and 26 comparison group). Patients were eligible if they were aged 18 years or older with the following: leg ulcer (venous, ischemic, traumatic, or diabetic foot ulcer), dehisced surgical wound healing by second intention, or pressure ulcers. Wound area was between 1 and 250 cm2. Exclusion criteria were systemic inflammatory disease or oncological disease, wounds with clinical signs of infection, terminal situation (life expectancy less than 6 months), pregnancy or wounds treated with negative pressure therapy.
A cut-off of 8 weeks (or healing if this occurred before 8 weeks) was established. A clinical nurse assessed patients at baseline and at weeks 2, 4, 6, and 8 to determine their evolution. Data collected from each patient included demographic characteristics, patient’s clinical background (concomitant medical diagnosis, clinical antecedents, nutritional status, smoking habit), description of the wound (etiology, size, location, specific characteristics), healing measured by RESVECH 2.0 score and variation in wound are measured by HELCOS tool.
## 2.1.1. Wound Management
Patients were managed according to a good standard of care. A general protocol for wound management was established: cleaning the wound with sterile physiological saline solution, debridement to deep clean the nonviable tissues in the wound bed, antioxidant dressing application as a primary dressing, and cover with secondary dressing. The dressing is kept in place for 2 to 3 days, according to the manufacturer’s recommendations and depending on the level of wound exudates.
## 2.1.2. Statistical Analysis
Descriptive statistics were used (mean and standard deviation for quantitative variables; frequency and percentages for nominal variables).
## 2.1.3. Ethical Aspects
The study was approved by the Ethics Committee of Jaen (Andalusian Health System) with reference number 0645-N-19. The study was conducted in accordance with the ethical principles of the Declaration of Helsinki. The patients provided written informed consent, which ensured data confidentiality.
## 3.1. Description of HELCOS Wound Healing Software
HELCOS is an integrated wound management system that calculates wound area and the relative percentage of tissue types in the wound bed using an image of the lesion. This image is loaded into the system and assigned to a patient and a case. For each patient, different images of the lesion can be attached over time to evaluate its evolution using different methods. This version is free and accessible in Spanish [4].
First, wound area is checked by measuring length and width directly with a graduated ruler (Kundin method) [19]. Then, it is estimated using digital analysis of wound photography.
Second, the relative percentage of tissue types in the wound bed (granulation, slough and necrotic tissue) is estimated. This software identifies tissue types by using different colors in the wound bed: red for granulation tissue, yellow for slough, and black for necrotic tissue. It also creates a graph showing the evolution of the percentage of tissue present in the wound bed over the follow-up period.
In addition, the RESVECH 2.0 scale is integrated in this software for evaluation of the status of the wound [20]. It assesses six aspects (wound size, depth/affected tissues, wound edges, type of tissue in the wound bed, exudate, and infection/inflammation). The score of this scale ranges from 0 points (wound healed) to 35 points (the worst possible status of the wound). A lower score means an improvement in the healing process. This scale is an excellent tool for comparing the data grouped according to the type of wound, recurrence, or severity.
## 3.2. Description of Wounds
In reference to the etiology of the wounds, $28.6\%$ were venous, $7.1\%$ ischemic, $7.1\%$ diabetic, $25\%$ traumatic, $10.7\%$ surgical wound, and $21.4\%$ pressure injuries. The wound locations were $42.9\%$ leg, $39.3\%$ food, $10.7\%$ gluteus/coccyx, $3.6\%$ abdomen, and $3.6\%$ upper limb. *The* general wound characteristics are presented below (Table 1). Twelve wounds treated with the antioxidant dressing were healed at 8 weeks ($42.86\%$) and 16 had an increase of $50\%$ or more in granulation tissue ($57.14\%$).
## 3.3. Healing Monitoring
We present several significant cases of wounds treated with the antioxidant dressing over eight weeks, which were monitored with the HELCOS software and achieved complete wound healing, significantly reduced wound area, or showed an important change in the tissues present in the wound bed.
The data and graphs presented in each of the cases refer to the analysis of the percentage of tissues present in the wound bed (granulation tissue and devitalized tissue—sloughed or necrotic) and the area of the lesion as analyzed with the HELCOS system and demonstrates wound follow-up.
Case 1. Traumatic leg wound.
A 59-year-old male presented with a traumatic wound on the lower limb, which was not healing (Figure 1). The initial area of the wound was 5.86 cm2, with a depth affecting muscle, defined borders, tissue compatible with biofilm, and desquamation on the perilesional skin. At the week 6 assessment, we observed complete wound healing (Table 2).
The tissues present in the wound bed showed a favorable evolution toward healing throughout the 6 weeks of treatment, decreasing the percentage of sloughed tissue present in the wound bed and increasing granulation tissue (Figure 2).
Case 2. Incised leg wound.
A 71-year-old male presented with a traumatic injury to the internal tibial area. This wound had damaged edges and abundant exudate (Figure 3). The initial area was 4.73 cm2, with $89.98\%$ devitalized tissue (necrotic/sloughed), and only $10.02\%$ was granulation tissue. Over 8 weeks, the wound area was reduced and the wound bed was cleaned, until reaching complete healing (Table 3) (Figure 4).
Case 3. Wound with venous etiology.
This was a 72-year-old woman with a venous wound in the anterior tibial area. In the initial assessment, the wound area was 12.31 cm2, the edges were damaged, and there was a saturation of exudate (Figure 5). The percentage of tissues in the bed was $60.58\%$ granulation tissue and $39.43\%$ sloughed tissue. At 8 weeks, the antioxidant treatment achieved wound closure, contributed to the removal of sloughed tissues, and induced granulation tissue formation (Figure 6). It should be noted that this treatment also significantly reduced pain; at the initial assessment, the patient presented $\frac{10}{10}$ on the Visual Analog Scale (VAS), $\frac{4}{10}$ at week 2, $\frac{1}{10}$ at weeks 4 and 6, and no pain by week 8.
Case 4. Traumatic cavity wound.
This was a 67-year-old male with a cavity wound of traumatic etiology located in the lower extremity. This clinical case stands out for its rapid evolution. The initial area was 5.42 cm2, highlighting the depth of the cavitation, but in just four weeks he achieved complete healing and a favorable evolution of the tissues (Figure 7 and Figure 8). It should also be emphasized that initially he reported a $\frac{10}{10}$ on the VAS pain scale, which decreased to $\frac{5}{10}$ in week 2, and completely disappeared in week 4.
Case 5. Diabetic foot ulcer.
This was a 57-year-old man with a diabetic foot ulcer that had an initial area of 1.54 cm2, and closed at week 8 (Figure 9). The antioxidant treatment was able to clean the wound bed, completely eliminating the sloughed tissue and facilitating the production of granulation tissue. At baseline, the wound had $76.19\%$ granulation tissue and $23.81\%$ sloughed tissue; from week 2 to week 8 only granulation tissue was observed in the wound ($100\%$) (Figure 10).
Case 6. Dehiscence surgical wound.
This was a 75-year-old male presented with a surgical wound in the lower limb that was healing by second intention. The wound had muscle involvement, thickened borders, and exudate leakage. This injury stands out for two aspects, firstly, its initial surface; it was a large wound (27.41 cm2), which reduced in size by $50\%$ (13.88 cm2) at week 8 (Figure 11). Second was the favorable evolution in the percentage of tissues present in the wound bed. Table 4 shows how from week 4 and coinciding with the overcoming of the inflammatory phase of the wound, which is where the antioxidant dressing has its difference in effect with respect to other therapeutic strategies, it was possible to invert the percentage of tissue in bed, with granulation tissue predominating and devitalized tissue decreasing (Figure 12). This wound reached complete healing at week 13, outside the follow-up period established in the study.
## 4. Discussion
Wound monitoring is an essential action, providing baseline measurements, and guides us in assessing wound healing [21]. However, monitoring methods need to be accurate, reliable, and feasible in order to assess the healing process.
According to the available scientific evidence, the use of digital planimetry or digital images are highly recommended. This method provides high precision in measurements of the wound area and the tissues present in the lesion bed [3].
Based on the results of our study, the HELCOS software [22] is a complete multidimensional tool performing quantitative comparison both of the wound area and of the different types of tissues present in the wound bed throughout the follow-up period. Moreover, this information is provided through descriptive data and graphical representations. The graphs help to interpret the numerical data obtained and visually improve the interpretation of the evolution analysis performed. In addition, HELCOS [22] includes wound assessment with the validated RESVECH 2.0 scale [20]. Digital or web-based tools for wound measurement and monitoring can be a useful resource in clinical studies.
In addition, some of the data obtained in these cases align with two previously published observational studies with this antioxidant product. One was a multicenter case series developed by Castro et al. in 2017 [16] with 31 patients with acute and chronic wounds, with a follow-up period similar to ours. It describes that at the end of the follow-up period, $29\%$ of the wounds healed completely, while in our study this was $42.85\%$. Regarding the variation in the RESVECH scale, Castro et al. describes a decrease in the average score of 10.16 points; similarly, in our study it was 7.89 points [16].
The other observational study mentioned was developed by Jiménez-García et al. [ 17], in which 31 patients with chronic wounds were included with a follow-up period of 12 weeks. The results described the evolution of wound healing evaluated by RESVECH 2.0, with a $67.8\%$ reduction at week 12 after using the antioxidant dressing. Likewise, the percentage of wound healing increased significantly over time, and was $71\%$ at week 12. During the follow-up time, $50\%$ of the wounds healed completely.
One of the strengths of this study is the use of the HELCOS web-based tool, which can help clinicians differentiate between different types of tissue in the wound bed and monitor healing. This article is one of the first reports of the performance of this tool in a real context.
However, the use of this tool is not without limitations. Digital images can be affected by lighting, location, and variability when shooting, leading to an underestimation of the wound analysis [23], so it is recommended to standardize the lighting conditions for the picture.
## 5. Conclusions
The results obtained indicate that wounds monitoring helps improve healing, facilitating clinical decision-making in healthcare. For this reason, it is necessary that the measurement and monitoring methods are precise, reliable, and viable for their correct application in daily clinical practice. This is also reflected in how the use of digital applications in measuring and evaluating wounds is increasingly widespread. The HELCOS web-based system is a user-friendly and useful resource available to clinicians for wound analysis and wound healing monitoring. The antioxidant dressing used in these cases is an alternative for wound management that merits further research.
## References
1. Atkin L., Bućko Z., Conde Montero E., Cutting K., Moffatt C., Probst A., Romanelli M., Schultz G.S., Tettelbach W.. **Im-plementing TIMERS: The race against hard-to-heal wounds**. *J. Wound Care* (2019.0) **23** S1-S50. DOI: 10.12968/jowc.2019.28.Sup3a.S1
2. Guo S., Di Pietro L.A.. **Factors affecting wound healing**. *J. Dent. Res.* (2010.0) **89** 219-229. DOI: 10.1177/0022034509359125
3. Jørgensen L.B., Sørensen J.A., Jemec G.B., Yderstraede K.B.. **Methods to assess area and volume of wounds—A systematic review**. *Int. Wound J.* (2016.0) **13** 540-553. DOI: 10.1111/iwj.12472
4. Verdú Soriano J., López Casanova P., Rodríguez Palma M., García Fernández F.P., Pancorbo Hidalgo P.L., Soldevilla Ágreda J.J.. **HELCOS. Sistema integrado para el manejo de heridas**. *Rev. ROL Enferm.* (2018.0) **41** 778-783
5. Schilrreff P., Alexiev U.. **Chronic inflammation in non-healing skin wounds and promising natural bioactive compounds treatment**. *Int. J. Mol. Sci.* (2022.0) **23**. DOI: 10.3390/ijms23094928
6. Beckman K.B., Ames B.N.. **The free radical theory of aging matures**. *Physiol. Rev.* (1998.0) **78** 547-581. DOI: 10.1152/physrev.1998.78.2.547
7. Rodriguez P.G., Felix F.N., Woodley D.T., Shim E.K.. **The role of oxygen in wound healing: A review of the literature**. *Dermatol. Surg.* (2008.0) **34** 1159-1169. DOI: 10.1097/00042728-200809000-00001
8. Cano S.M., Lancel S., Boulanger E., Neviere R.. **Targeting Oxidative Stress and Mitochondrial Dysfunction in the Treat-ment of Impaired Wound Healing: A Systematic Review**. *Antioxidants* (2018.0) **7**. DOI: 10.3390/antiox7080098
9. Ponugoti B., Xu F., Zhang C., Tian C., Pacios S., Graves D.T.. **FOXO1 promotes wound healing through the up-regulation of TGF-beta1 and prevention of oxidative stress**. *J. Cell Biol.* (2013.0) **203** 327-343. DOI: 10.1083/jcb.201305074
10. Comino-Sanz I.M., López-Franco M.D., Castro B., Pancorbo-Hidalgo P.L.. **The Role of Antioxidants on Wound Healing: A Review of the Current Evidence**. *J. Clin. Med.* (2021.0) **10**. DOI: 10.3390/jcm10163558
11. Liu E., Gao H., Zhao Y., Pang Y., Yao Y., Yang Z., Zhang X., Wang Y., Yang S., Ma X.. **The potential application of natural products in cutaneous wound healing: A review of preclinical evidence**. *Front. Pharmacol.* (2022.0) **13** 9004339. DOI: 10.3389/fphar.2022.900439
12. **Reoxcare**
13. Castro B., Palomares T., Azcoitia I., Bastida F., del Olmo M., Soldevilla J.J., Alonso-Varona A.. **Development and preclin-ical evaluation of a new galactomannan-based dressing with antioxidant properties for wound healing**. *Histol. Histopathol.* (2015.0) **30** 1499-1512. DOI: 10.14670/HH-11-646
14. Alven S., Nqoro X., Aderibigbe B.A.. **Polymer-Based Materials Loaded with Curcumin for Wound Healing Applications**. *Polymers* (2020.0) **12**. DOI: 10.3390/polym12102286
15. Ozkaya H., Omma T., Bag Y.M., Uzunoglu K., Isildak M., Duymus M.E., Kismet K., Senes M., Fidanci V., Celepli P.. **Topical and Systemic Effects of N-acetyl Cysteine on Wound Healing in a Diabetic Rat Model**. *Wounds* (2019.0) **31** 91-96. PMID: 30802208
16. Castro B., Bastida F.D., Segovia T., López Casanova P., Soldevilla J.J., Verdú-Soriano J.. **The use of an antioxidant dressing on hard-to-heal wounds: A multicentre, prospective case series**. *J. Wound Care* (2017.0) **26** 742-750. DOI: 10.12968/jowc.2017.26.12.742
17. Jiménez-García J.F., Aguilera-Manrique G., Parra-Anguita L., Torra-Bou J.E., Villegas-Tripiana I., García-Fernández F.P.. **An antioxidant approach directed to non-healing wounds activation-A case report study**. *Clin. Case Rep. Rev.* (2021.0) **7** 1-6. DOI: 10.15761/CCRR.1000499
18. Comino-Sanz I.M., López-Franco M.D., Castro B., Pancorbo-Hidalgo P.L.. **Antioxidant dressing therapy versus standard wound care in chronic wounds (the REOX study): Study protocol for a randomized controlled trial**. *Trials* (2020.0) **8** 505. DOI: 10.1186/s13063-020-04445-5
19. Kundin J.. **A new way to size up a wound**. *Am. J. Nurs.* (1989.0) **89** 206-207. PMID: 2916597
20. Restrepo-Medrano J.C., Verdú Soriano J.. **Development of a wound healing index for chronic wounds**. *EWMA J.* (2012.0) **12** 39-46
21. Flanagan M.. **Wound measurement: Can it help us to monitor progression to healing?**. *J. Wound Care* (2003.0) **12** 189-194. DOI: 10.12968/jowc.2003.12.5.26493
22. **Integrated Wound Management System**. (2017.0)
23. Chang A.C., Dearman B., Greenwood J.E.. **A comparison of wound area measurement techniques: Visitrak versus photography**. *Eplasty* (2011.0) **11** e18. PMID: 21559060
|
---
title: Maternal Prepregnancy Obesity Affects Foetal Growth, Birth Outcome, Mode of
Delivery, and Miscarriage Rate in Austrian Women
authors:
- Katharina Syböck
- Beda Hartmann
- Sylvia Kirchengast
journal: International Journal of Environmental Research and Public Health
year: 2023
pmcid: PMC10002339
doi: 10.3390/ijerph20054139
license: CC BY 4.0
---
# Maternal Prepregnancy Obesity Affects Foetal Growth, Birth Outcome, Mode of Delivery, and Miscarriage Rate in Austrian Women
## Abstract
The increasing obesity rates among women of reproductive age create a major obstetrical problem as obesity during pregnancy is associated with many complications, such as a higher rate of caesarean sections. This medical record-based study investigates the effects of maternal prepregnancy obesity on newborn parameters, birth mode, and miscarriage rate. The data of 15,404 singleton births that had taken place between 2009 and 2019 at the public Danube Hospital in Vienna were enrolled in the study. Newborn parameters are birth weight, birth length, head circumference, APGAR scores, as well as pH values of the arterial and venous umbilical cord blood. In addition, maternal age, height, body weight at the beginning and the end of pregnancy, and prepregnancy body mass index (BMI) (kg/m2) have been documented. The gestational week of birth, the mode of delivery, as well as the number of previous pregnancies and births, are included in the analyses. Birth length, birth weight, and head circumference of the newborn increase with increasing maternal BMI. Furthermore, with increasing maternal weight class, there tends to be a decrease in the pH value of the umbilical cord blood. Additionally, obese women have a history of more miscarriages, a higher rate of preterm birth, and a higher rate of emergency caesarean section than their normal-weight counterparts. Consequently, maternal obesity before and during pregnancy has far-reaching consequences for the mother, the child, and thus for the health care system.
## 1. Introduction
The prevalence of overweight and obese people is increasing drastically all over the world [1]. Obesity is no longer only a typical problem in developed countries, it is also growing in developing and emerging nations [2,3]. For that reason, it is already being considered an “obesity pandemic”. With the increasing obesity rates worldwide, a rise in various other diseases associated with obesity is also occurring. Several studies show that the risk of cardiovascular diseases, metabolic diseases such as diabetes type 2, and various cancers increase significantly with obesity [4,5]. In addition, fertility and reproductive success are affected by obesity in both men and women [6,7]. In most countries, women have higher obesity rates than men, with less educated women at two to three times higher risk than those with more education [8]. These high obesity rates are especially serious in women of reproductive age, as overweight and obesity before and during pregnancy are associated with many complications, such as preeclampsia [9], higher caesarean section rates [10,11,12], macrosomia [13], and foetal acidosis [13,14], just to name a few. This is not only problematic for the women themselves, but also for the foetus or newborn, and the consequences can be far-reaching [15,16,17,18,19,20,21,22]. At first, obesity reduces the chance of successful fertilisation, especially in assisted fertilisation, such as in vitro fertilisation. During pregnancy, the risk of miscarriage is higher in obese women [23,24], this is especially true after assisted fertilisation [25]. Even after successful spontaneous conception, pregnancy in obese women is associated with many risks, not only for the mother but also for the newborn [26]. Obesity before and during pregnancy dramatically increases the risk of miscarriage, preterm birth, infant mortality, and stillbirth [27,28], but also of congenital defects, such as spina bifida, cleft lip and palate, hydrocephalus, and heart defects such as septal anomalies [29,30]. On the other hand, maternal obesity, as excessive gestational weight gain, enhances foetal growth and may result in larger head circumferences and macrosomia of the foetus [31,32,33].
A special problem is represented by the birth itself [34]. Some studies showed that the risk of oxygen deficiency during birth and related adverse outcomes was increased in newborns of overweight or obese women [14,35]. Oxygen deficiency during birth and perinatal asphyxia are associated with cerebral disorders, and thus increase the risk for impaired neurological development [36,37,38,39] and, in the worst case, for neonatal death [39]. There are also long-term consequences of oxygen deprivation such as cerebral palsy, a spastic dysfunction of locomotion [40]. Despite oxygen deficiency, maternal obesity is often associated with low APGAR scores, indicating severe neonatal stress and a poor adaptation to the postnatal environment of the newborn [41]. Furthermore, maternal obesity is associated with an increased caesarean section rate [11,12,15,34]. Caesarean sections in Class III obese women, however, are especially risky and technically difficult to perform. Accordingly, pregnancies in obese women are considered high-risk pregnancies, that pose special risks for mothers and foetuses.
The present study focuses on the association patterns between maternal weight status and pregnancy outcome in Vienna, Austria over the last 14. years. Austria is not only one of the richest countries of the European Union, the medical health care system is highly developed. All Austrian residents have social insurance that covers all medical costs in public hospitals. During the 1970s, the sophisticated system of the so-called “mother-child passport” was introduced, which guarantees at least three prenatal examinations starting at the 8th gestational week, and eight postnatal check-ups of the child by paediatricians between birth and the age of 4 years free of charge. The completion of all 11 medical examinations is rewarded with a financial premium by the government. The introduction of the mother-child passport helped to make pregnancies and birth much safer. Consequently, pregnant women and newborns are well cared for by the public health service. On the other hand, the prevalence of overweight and obese among young women of reproductive age is steadily increasing in Austria. In 2014, about $13\%$ of 15 to 29 years old females were overweight and about $6\%$ were obese; only 5 years later, however, the prevalence of overweight and obesity increased to $16.1\%$ and $6.7\%$, respectively, in this age range. A similar trend is found for women aged between 30 and 44 years. In 2014, $21\%$ of women in this age group were overweight and about $8\%$ were obese. In 2019, the numbers increased to $25\%$ and $12\%$, respectively [42]. Despite the increasing prevalence of overweight and obesity in Austria, the overweight and obesity rates, among women of reproductive age are still lower than in other countries such as the United States [43]. Overweight and obesity during reproductive age is a matter of concern in Austria because rising obesity rates among pregnant women not only place a short-term burden on the health care system, but can have long-term impacts, as the consequences of obesity during pregnancy can have long-lasting effects on both mother and child. In the present study, we examine association patterns between maternal weight status and parameters of pregnancy, delivery mode, foetal growth and birth outcome in Vienna, Austria. In detail we tested the following hypotheses: [1]Overweight or obese mothers are more likely to experience preterm birth (<37 gestational weeks), they have a history of more miscarriages, and a higher rate of caesarean section than normal-weight mothers.[2]Among term birth (≥37 gestational weeks), the newborns of overweight or obese mothers are larger and heavier, but show lower APGAR scores than newborns of normal-weight mothers.[3]Among term birth (≥37 gestational weeks), spontaneous delivered newborns of primiparous overweight or obese mothers have a higher risk of oxygen deficiency.
## 2.1. Dataset and Study Design
In this retrospective medical-record-based single-centre study, the data of 15,404 singleton births were included. Newborns with congenital anomalies were excluded. All births had taken place at the public Danube Hospital (Clinic Donaustadt) in Vienna, Austria between 2009 and 2019. This hospital is one of the largest public birth clinics in Vienna. In the first step, we included all singleton births ($$n = 15$$,404), independently of the duration of pregnancy and tested the associations between maternal prepregnancy weight status and the history of miscarriages, preterm birth, as well as the mode of delivery. In the second step, we included term birth exclusively ($$n = 14$$,444) and analysed the association between maternal prepregnancy weight status and pregnancy outcome, i.e., newborn size and APGAR scores among term birth (≥37 gestational weeks), only. In the third step, only spontaneously delivered term birth of primiparous mothers ($$n = 5260$$) were included and the associations between maternal prepregnancy weight status and cord blood pH values were analysed.
The study was conducted according to the guidelines of the Declaration of Helsinki and approved by the Ethics Committee of Vienna (responsible for Public Hospitals) (Protocol number: EK 19-274-VK 18 March 2020).
## 2.2. Newborn Parameters
Newborn parameters include birth weight (in grams) using a digital infant scale, birth length (in cm) using a standard measurement board for infants, and head circumference (in cm) using a standard tape. All measurements were taken immediately after birth by a trained midwife. In this study, macrosomia is defined as a birth weight greater than 4000 g [44]. APGAR scores have been recorded 1, 5, and 10 minutes after birth [45]. In addition, the foetal presentation at birth (cephalic, breech, transverse) is documented. The pH value of the arterial and venous umbilical cord blood, which represent an accurate, reproducible, and objective evaluation of the oxygen deficiency during birth, was measured to one decimal place. Using the pH value of the umbilical cord blood is therefore recommended to evaluate the newborn outcome [46,47]. For correct and adequate testing of the hypothesis that cord blood pH value is associated with maternal weight status, we defined some additional exclusion criteria. First, only pH values >6.4 were considered, since lower values are more likely to be erroneous. Secondly, caesarean sections and preterm births were excluded, since caesarean sections or preterm births could be the actual reasons for oxygen deficiency and thus for a low pH value of the umbilical cord blood. In addition, only primiparous women were included in order to obtain a sample that was as homogeneous as possible, as foetal acidosis depends on parity [14]. Therefore, only births that meet the following criteria are included in the cord blood analysis: primiparous women who experienced vaginal births after the 37th week of pregnancy (i.e., no premature births).
## 2.3. Maternal Parameters
The maternal parameters include maternal age (in years), height (in cm), and body weight at the beginning and at the end of pregnancy (in kg). Trained personnel measured height to the nearest 0.1 cm using a standard anthropometer. Prepregnancy weight was recorded by an interview using the retrospective method, this means that pregnant women were asked about their body weight before pregnancy at the first prenatal examination. In addition, body weight was measured to the nearest 0.1 kg on a balance beam scale, at the first prenatal visit around the 8th week of gestation. As pointed out above, maternal weight was measured again before delivery (at the end of pregnancy). The weight gain during pregnancy was calculated by subtracting prepregnancy weight from body weight before delivery. In addition, the prepregnancy BMI (kg/m2) is calculated and classified into the following categories according to the WHO criteria [48]: underweight < 18.50; normal weight 18.50–24.99 (kg/m2); overweight 25.00–29.99 (kg/m2); Class I and Class II obesity 30.00–39.99(kg/m2); Class III obesity > 40 (kg/m2). For a more detailed description of the anthropometric methods applied see Kirchengast and Hartmann [49].
## 2.4. Obstetrical Parameters
The following obstetrical characteristics were recorded: preterm birth (<37 gestational weeks), versus term birth (≥37 gestational weeks), mode of delivery (spontaneous vaginal delivery, vacuum extraction/forceps, planned caesarean section, unplanned or emergency caesarean section), foetal presentation at birth (cephalic presentation, breech presentation, transverse presentation), and number of miscarriages. At the first prenatal visit, all mothers were asked concerning previous miscarriages. In addition, the number of miscarriages was calculated as the difference of reported pregnancies and births.
## 2.5. Statistical Analysis
The statistical analysis was carried out with IBM SPSS version 27. The significance level was set to 0.05. After computing descriptive statistics, Kruskal–Wallis H tests were performed to test for differences in neonatal and maternal parameters between maternal weight classes. Bonferroni post hoc tests were then conducted to allow pairwise comparisons between all groups. To evaluate the risk of macrosomia, the odds ratios were calculated for each maternal weight class compared to the normal-weight women with the respective $95\%$ confidence interval. To test for an association between maternal weight class and miscarriages, as well as the mode of delivery, Pearson Chi2 tests were performed. In addition, the odds ratios for miscarriage were calculated for each maternal weight class compared to the normal-weight women with the respective $95\%$ confidence interval. The relative risk of an emergency caesarean section was calculated for each weight category compared to normal-weight women. A linear regression analyses was computed to test the association patterns between the pH value of the arterial cord blood and maternal body mass index. Binary logistic regression was used to calculate the effect of maternal weight class on the mode of delivery, independent of weight gain during pregnancy, maternal age, and body height.
## 3.1. Sample Characteristics
Table 1 presents maternal, newborn, and obstetrical characteristics. The average age of the mother was 30.0 years ranging from 15 to 45.5 years. Their average height was 165.4 cm. The average maternal weight before pregnancy was 66 kg, and it was 80 kg at the end of pregnancy. On average, the women in this sample gained $22.4\%$ of their prepregnancy body weight during pregnancy. A total of 952 women ($6.2\%$) corresponded to the definition of underweight, showing a BMI less than 18.50 kg/m2, $61.0\%$ were normal weight and corresponded to the recommended range of 18.50 to 24.99 kg/m2. A further $20.5\%$ were classified as overweight with a BMI between 25 and 29.99 kg/m2, $10.7\%$ were obese with a BMI between 30 to 39.99 kg/m2, and $1.2\%$ were Class III obese showing a BMI greater than 40 kg/m2. The average birth weight in this sample was 3385.6 g, the average birth length was 50.6 cm, and the average head circumference was 34.2 cm. A resulting $11.1\%$ of the newborns were classified as low-weight (<2500 g), and $11\%$ corresponded to the definition of macrosomia (≥4000 g).
As demonstrated in Table 1, the majority of births occurred spontaneously, the most common child presentation was cephalic. More than $45\%$ of the mothers were primiparous. The preterm birth rate was $6.2\%$. Furthermore, $31.3\%$ of the mothers had experienced at least one miscarriage, $9.9\%$ of the mothers two or more miscarriages. The mean number of miscarriages was 0.5, but one woman experienced 14 miscarriages.
## 3.2. Maternal and Obstetrical Characteristics According to Maternal Weight Status
The maternal age differs significantly between the weight status groups ($p \leq 0.001$) (Table 2). Subsequent post hoc tests show that only the underweight women differ significantly in age from the other groups, with the underweight mothers being on average two years younger than those in the other weight status groups. Significant differences between the groups can also be recognised with regard to height ($p \leq 0.001$). However, only the overweight and underweight women differ significantly in terms of height, with the underweight women being on average taller than the overweight ones. Gestational weight gain decreased significantly with increasing weight status ($p \leq 0.001.$) With the exception of underweight and normal-weight groups, all weight status groups differed significantly from every other group (Table 2).
Preterm birth rate differed significantly (λ = 22.840, d.f. = 4, $p \leq 0.001$) between maternal weight status groups. The highest rate of preterm birth occurred among obese women, the lowest among normal-weight women. Furthermore, a significant relationship (λ = 35.120, df = 4, $p \leq 0.001$) occurred between the history of miscarriages and maternal weight status. Underweight and normal-weight women do not differ significantly in miscarriage risk. Between overweight women and normal-weight women, however, significant differences in miscarriage risk occur. The risk of miscarriage is increased by about 1.2 times in overweight ($95\%$ CI 1.09–1.29) and obese women ($95\%$ CI 1.06–1.33) compared to normal-weight women. The effect is even stronger in Class III obese women, with a 1.8 times higher risk of miscarriage ($95\%$ CI 1.33–2.37) compared to normal-weight women (Table 2).
A significant relationship can be found between maternal weight classes and mode of delivery (λ = 81.777, d.f. = 12, $p \leq 0.001$). It can be seen that obese and Class III obese women in particular have more caesarean deliveries (planned and emergency) and fewer vaginal, spontaneous deliveries (Table 3). The risk of emergency caesarean section increases continuously from 1.2 times higher risk ($95\%$ CI 1.07–1.41) in overweight women to 2.2 times higher risk ($95\%$ CI 1.44–3.31) in Class III obese women compared to normal-weight women, while there is no significant difference between underweight and normal-weight women in terms of the risk of emergency caesarean deliveries.
A binary logistic regression model is calculated to show the independent effects of maternal BMI on the mode of delivery. It shows that maternal BMI has an independent significant effect ($p \leq 0.001$) on the occurrence of emergency caesarean sections. In addition to the BMI, weight gain during pregnancy, maternal age, and neonatal head circumference have a positive, significant, and independent effect on emergency caesarean section rates. Maternal height, gestational week, and birth weight are significantly negatively associated with the caesarean section rate (Table 3).
## 3.3. Newborn Characteristics According to Maternal Weight Status
With increasing maternal weight status, the newborns become significantly longer ($p \leq 0.001$). Post hoc tests show that only the birth lengths of the newborns of normal-weight and overweight women do not differ significantly, while there are significant differences between all other weight status groups. Birth weight also shows significant differences between the weight status groups ($p \leq 0.001$) with a tendency for birth weight to increase as the mother’s BMI increases. The highest birth weight was found among obese mothers. Pairwise comparisons show that the birth weights of the children of underweight and normal-weight women are each significantly different from every other group. Within the overweight, obese, and Class III obese women, however, no significant birth weight differences can be detected. Head circumference also differs significantly between the weight status groups ($p \leq 0.001$), in which the head circumference tends to increase as the mother’s BMI increases. More precisely, the head circumference of children of underweight women is significantly smaller than in all other groups and the head circumference of children of normal-weight women is significantly smaller than that of children of overweight and obese women (Table 4).
The risk of newborn macrosomia (>4000 g) increases significantly with increasing maternal weight status. While the newborns of underweight women have a significantly ($p \leq 0.001$) lower risk of macrosomia (OR = 0.45; CI $95\%$ 0.31–0.61) than those of normal-weight women, neonates of overweight women have a 1.37-fold (CI $95\%$ 1.21–1.55) higher risk of macrosomia compared to those of normal-weight women ($p \leq 0.001$). The effect is even more pronounced in the newborns of obese ($p \leq 0.001$) and Class III obese women ($p \leq 0.001$). Here, the risk of macrosomia increases by a factor 1.88 (CI $95\%$ 1.62–2.17) and 2.0 (CI $95\%$ 1.38–2.93), respectively, compared to newborns of normal-weight women.
Furthermore, we found a significant association between APGAR scores after 1, 5, and 10 min and maternal weight status. The APGAR scores decreased significantly with increasing maternal weight status.
The pH value of the arterial cord blood differed significantly across maternal weight status categories ($p \leq 0.001$, $H = 26.601$). Subsequent Bonferroni corrections show that obese women differ significantly from underweight and normal-weight women ($$p \leq 0.005$$ and $$p \leq 0.011$$, respectively). In addition, overweight women also differ significantly from underweight and normal-weight women ($$p \leq 0.012$$ and $$p \leq 0.014$$, respectively). Furthermore, according to the results of a linear regression analysis, the pH value of the arterial cord blood decreases with increasing maternal BMI ($B = 7.258$, β = −0.066, $p \leq 0.001$, R2 = 0.004). Although highly significant, the coefficient of determination is relatively low. Only $0.4\%$ of the variance of the pH value can be explained by the maternal BMI. The differences in venal blood pH values between the weight status groups are not significant; nevertheless, there is still a trend apparent that the pH value of the venous cord blood decreases with increasing maternal weight class (Table 4).
## 4. Discussion
The worldwide trend of rising obesity rates is particularly serious among women of reproductive age, as obesity before and during pregnancy is associated with various severe—sometimes long-term—complications for mother and child [19,22,26,50,51,52,53]. This study in particular focuses on newborn parameters, cord blood pH values, miscarriage rate, mode of delivery, and the respective effect of maternal obesity on each of them in an Austrian sample. Altogether the data of 15,404 singleton births taking place at the public Danube hospital in Vienna, Austria, were included in the analysis. A total of $20.5\%$ of the mothers corresponded to the definitions of overweight (BMI 25.00–29.99 kg/m2), and $10.7\%$ to the definition of obesity(BMI 30.00–39.99 kg/m2). Furthermore, $1.2\%$ of the included women showed a prepregnancy BMI above 40.00 kg/m2. These overweight and obesity rates are typical of Austria [42], although only an urban sample was analysed.
The first hypothesis that overweight or obese mothers are more likely to experience preterm birth (<37 gestational weeks), to have a history of more miscarriages, and a higher rate of caesarean section than normal-weight mothers could be verified. A significant association between weight status and history of miscarriage was detected ($p \leq 0.001$), with overweight and obese women having experienced more miscarriages than normal-weight women. The risk of ever having suffered a miscarriage is 1.2 times higher in overweight women and 1.8 times higher in Class III obese women than in normal-weight women. This finding is in accordance with previous studies which have shown that with higher maternal BMI, the risk of miscarriage increases [24]. A direct causality between obesity and increased miscarriage rate, however, cannot be concluded from our results due to the lack of data on weight status at the time of miscarriage. Furthermore, the prevalence of preterm birth (≤37 gestational weeks) was significantly highest ($8.5\%$) among mothers who were overweight before pregnancy, while the lowest rate ($5.7\%$) occurred among normal-weight ones. Concerning breech presentation, however, we found no significant association with maternal weight status.
Considering the association between mode of delivery and maternal prepregnancy obesity, our results are largely consistent with the previous literature [11,15]. The hypothesis that birth mode is related to maternal weight status was thus verified. A significant relationship between maternal weight class and mode of delivery could be shown ($p \leq 0.001$). It is particularly important to clarify to what extent maternal weight is related to emergency caesarean section rates. Emergency caesarean deliveries are those C-sections that are not planned and that require an acute change from a vaginal delivery to a caesarean section. Since emergency caesarean sections are acutely medically necessary interventions, those types of C-sections are in the focus of this study. As maternal BMI increases, the emergency caesarean section rate also increases. It is particularly noticeable that the risk for an emergency caesarean section very strongly relates to the severity of obesity. While overweight women have a 1.2-fold higher risk and obese women a 1.4-fold higher risk, Class III obese women have more than twice the risk (2.1-fold higher) of experiencing an emergency caesarean section compared to normal-weight women. This is a very abrupt increase in risk between obese and Class III obese women. These results are roughly in line with the findings of Chu et al. [ 27], although the calculated C-section risks of their study are even slightly higher.
The second hypothesis, that newborns of overweight or obese mothers are larger and heavier but show lower APGAR scores than newborns of normal-weight mothers, could be verified. Considering birth length, birth weight, and head circumference, it is evident that as maternal prepregnancy BMI weight class increases, the neonate has significantly larger dimensions. This is largely consistent with the previous literature [16,34,35]. Nevertheless, previous studies have shown that maternal obesity, and especially Class III obesity, can not only cause relatively large newborns and macrosomia, but can also be responsible for low birth weight and even foetal growth retardation. Maternal obesity may thus have multidirectional and opposing effects on birth weight [13]. In our sample, the risk of macrosomia increases with increasing weight status. While the risk for macrosomia is 1.4 times higher in overweight women than in normal-weight women, the risk in obese women is almost twice as high compared to normal-weight women.
The precise mechanisms of how maternal obesity affects foetal growth are not yet fully understood. Certainly, maternal obesity exposes the foetus to a different hormonal and external environment. The complex interaction of environment and genes can lead to enhanced growth in the foetus resulting in macrosomia [54], as found in the present study. Macrosomia, in turn, is associated with several complications, including increased risk of caesarean section, extended stay in hospital, chorioamnionitis [55], and many others. The results of the present study suggest, that maternal overweight and especially obesity before and during pregnancy, affect foetal growth and consequently newborn size.
The third hypothesis that maternal prepregnancy weight status is related to cord blood pH values, however, could only be partially verified. According to previous studies, overweight and obesity are associated with foetal acidosis [14]. A valid indicator of such oxygen deficiency during birth is the pH value of the umbilical cord blood [14]. In our study, the association of cord blood pH value and maternal weight status is only partially consistent with the previous literature. Although the regression line between (arterial) pH values and maternal BMI shows a negative slope, indicating a negative correlation between the two variables, the corresponding R2 is relatively small (0.004), maternal BMI can explain only a very small amount of variance in pH values. Dividing BMI again into the respective weight categories, there is a tendency for median values of pH to decrease with increasing maternal weight categories, with normal-weight women having a mean of 7.24 and Class III obese women having a mean of 7.21. Both obese and overweight women differ significantly from normal-weight and underweight women in arterial cord blood pH values. The question that arises is why, contrary to previous literature such as [14], the other groups do not differ significantly from each other with respect to this characteristic in this sample. This is likely due to the small sample sizes of some weight classes. The group of Class III obese women ($$n = 47$$), especially, is strongly underrepresented in our study. Looking now at the pH values of the venous cord blood, there are no significant differences at all in the pairwise comparisons between the groups. Although the same trend as in the analysis of the pH value of the arterial blood is observed, namely a decrease of the pH value with increasing maternal weight class, these results are not significant.
To sum it up, the results of the present study suggest, that maternal overweight, and in particular maternal obesity even before pregnancy, has a negative effect on birth outcome and on delivery, resulting in an increased risk of emergency caesarean sections. Considering the rising obesity rates among women of reproductive age worldwide, these findings are of particular concern. Increasing obesity rates are mostly the result of profound changes in lifestyle patterns, characterised by a reduction of physical activity and an increase of high calorie intake. Increasing obesity rates, and also the trend to postpone motherhood, have led to an increase in emergency caesarean section rates [12]. In our study, a binary logistic regression model shows an independent effect of BMI on the mode of delivery ($p \leq 0.001$), respectively, on the incidence of emergency caesarean section. Maternal obesity is clearly associated with an increased risk of emergency caesarean section, and poses in this way a special burden for health care systems, as they are associated with high costs and effort [56]. Surgeries including C-sections for severely obese individuals are a much greater expense for health care professionals and on average, obese individuals also require longer hospital stays [57,58,59]. In Class III obese women especially, a caesarean section presents great technical difficulties, some of which can have severe life-threatening consequences [60,61]. The problems range from difficulty in anaesthesia and finding the epidural space, to difficult conditions in transporting patients and increased numbers of staff required [11]. This causes another burden on the health care system. In order to achieve a sustainable reduction of the emergency caesarean section rate, i.e., the rate of those caesarean sections that are medically necessary, a reduction in the obesity rate among women of childbearing age would be of major importance.
Although the results of our study correspond to the finding of previous investigations, the limitations of our study should not go unmentioned. A major limitation is the lack of information regarding socioeconomic status and educational level of the mothers. It is well documented that maternal educational level and socioeconomic status are strongly related to reproductive performance and birth outcome, but also the prevalence of overweight and obesity [62,63,64]. Nevertheless, many previous studies show that the negative effects of maternal obesity during pregnancy on the newborn are independent of socioeconomic status [19]. Therefore, it still makes sense to examine the relationships between maternal obesity and neonatal parameters, even without information on social status. We are aware of this limitation; however, as pointed out in the methods section, this is a medical record-based study without access to socioeconomic data.
Another important limitation is the use of BMI as a measure of weight status. There is no doubt, that the BMI is a widely used and accepted indicator to evaluate weight status in adults. Nevertheless, the BMI has some limitations that cannot be ignored. Since BMI is only a weight to length measure and does not take body composition into account, no distinction is made between lean mass and fat mass. A person with a high muscle mass therefore also has a high BMI and might thus incorrectly be classified as overweight or even obese. Vice versa, the BMI of a person with high fat mass but low lean mass is also not sufficiently informative to a certain extent [65]. Nevertheless, the BMI is a practical measure to determine the weight status because, unlike DEXA (=dual-energy X-ray absorptiometry), air displacement plethysmography, and many other sophisticated methods, its use is cheap, easy to survey, non-invasive, and also safe.
Another limitation represents the documentation of previous miscarriages. The difference between the number of pregnancies and the number of actual births alone, of course, as an indicator of the miscarriage rate says nothing about intentional and wanted abortions. Since no information on wanted abortion rates is available, they cannot be considered in this analysis. Therefore, only the difference between the number of pregnancies and births is used as an indicator for miscarriage. Although induced abortions are legal in Austria, there is no official documentation of the total number of abortions.
The strength of this study is the huge sample size, i.e., more than 15,000 singleton births were included in the analysis.
## 5. Conclusions
The present study clearly demonstrates the various associations between maternal overweight or obesity and birth outcome as well as obstetrical parameters. Obesity during pregnancy is associated with many adverse effects, such as an increased miscarriage rate, a higher rate of emergency caesarean sections, lower pH values of the umbilical cord blood, and also greater newborn size and a higher risk of macrosomia. Besides acute birth complications, far-reaching long-term consequences for mother and child may arise from maternal obesity. It is therefore of extreme importance to address the issue at all levels of the health care system, but also at all levels of society. These activities should include prevention strategies as well as weight loss therapies that help obese women to sustainably achieve a healthy body weight. To do this, the problem must be considered in its entirety and not at the level of the individuals themselves. Here, it is up to decision-makers to act. Effective measures should be taken by politicians, such as stricter advertising rules, especially for products consumed by children, price regulation such as a sugar tax, and more awareness and education in schools. Bold actions are needed to get the complex phenomenon of obesity under control.
## References
1. **Obesity and Overweight**. (2021)
2. Ford N.D., Patel S.A., Narayan V.. **Obesity in Low-and middle-income countries; Burden, Drivers and Emerging Challenges**. *Ann. Rev. Public Health* (2017) **38** 145-164. DOI: 10.1146/annurev-publhealth-031816-044604
3. Poobalan A., Aucott L.. **Obesity Among Young Adults in Developing Countries: A Systematic Overview**. *Curr. Obes. Rep.* (2016) **5** 2-13. DOI: 10.1007/s13679-016-0187-x
4. Lavie C.J., Milani R.V.. **Obesity and cardiovascular disease: The hippocrates paradox**. *J. Am. Coll. Cardiol.* (2003) **42** 677-679. DOI: 10.1016/S0735-1097(03)00784-8
5. Lega I.C., Lipscombe L.L.. **Review: Diabetes, Obesity and Cancer—Pathophysiology and Clinical Implications**. *Endocr. Rev.* (2019) **41** 33-52. DOI: 10.1210/endrev/bnz014
6. Hammoud A.O., Gibson M., Peterson C.M., Meikle A.W., Carrell D.T.. **Impact of male obesity on infertility: A critical review of the current literature**. *Fertil. Steril.* (2008) **90** 897-904. DOI: 10.1016/j.fertnstert.2008.08.026
7. Harreiter J., Kautzky-Willer A.. **Gender Obesity Report—Einfluss von Adipositas auf Reproduktion und Schwangerschaft**. *Wiener. Med. Wochenschr.* (2016) **166** 129-138. DOI: 10.1007/s10354-015-0410-5
8. **Obesity Update—OECD**. (2017)
9. Bodnar L.M., Ness R.B., Markovic N., Roberts J.M.. **The risk of preeclampsia rises with increasing prepregnancy body mass index**. *Ann. Epidemiol.* (2005) **15** 475-482. DOI: 10.1016/j.annepidem.2004.12.008
10. Chu S.Y., Kim S.Y., Schmid C.H., Dietz P.M., Callaghan W.M., Lau J., Curtis K.M.. **Maternal obesity and risk of cesarean delivery: A meta-analysis**. *Obes. Rev.* (2007) **8** 385-394. DOI: 10.1111/j.1467-789X.2007.00397.x
11. O’Dwyer V., Turner M.J., Salim R.. **Caesarean Section and Maternal Obesity**. *Cesarean Delivery* (2012)
12. Kirchengast S., Hartmann B.. **Recent Lifestyle Parameters Are Associated with Increasi. ng Caesarean Section Rates among Singleton Term Births in Austria**. *Int. J. Environ. Res. Public Health* (2019) **16**. DOI: 10.3390/ijerph16010014
13. Lewandowska M.. **Maternal Obesity and Risk of Low Birth Weight, Fetal Growth Restriction, and Macrosomia: Multiple Analyses**. *Nutrients* (2021) **13**. DOI: 10.3390/nu13041213
14. Johansson S., Sandstrom A., Cnattingius S.. **Maternal overweight and obesity increase the risk of fetal acidosis during labor**. *J. Perinatol.* (2018) **38** 1144. DOI: 10.1038/s41372-018-0144-5
15. Masoud I., Al-Fadhil F.A., Padmakumar H., Al-Alawi A., Al-Bahri Z.A., Al-Rawahi N.A., Al-Balushi L.H., Al-Dhanki M.S., Kurup P.J., Al-Hakmani F.M.. **The Effect of Obesity on Pregnancy and Its Outcome in the Population of Oman, Seeb Province**. *Oman Med. J.* (2016) **31** 12-17. DOI: 10.5001/omj.2016.03
16. Athukorala C., Rumbold A.R., Willson K.J., Crowther C.A.. **The risk of adverse pregnancy outcomes in women who are overweight or obese**. *BMC Pregnancy Childbirth* (2010) **10**. DOI: 10.1186/1471-2393-10-56
17. Aviram A., Hod M., Yogev Y.. **Maternal obesity: Implications for pregnancy outcome and long-term risks–a link to maternal nutrition**. *Int. J. Gynecol. Obstet.* (2011) **115** S6-S10. DOI: 10.1016/S0020-7292(11)60004-0
18. Gaillard R., Durmuş B., Hofman A., Mackenbach J.P., Steegers E.A., Jaddoe V.W.. **Risk factors and outcomes of maternal obesity and excessive weight gain during pregnancy**. *Obesity* (2013) **21** 1046-1055. DOI: 10.1002/oby.20088
19. Gaillard R., Santos S., Duijts L., Felix J.F.. **Childhood Health Consequences of Maternal Obesity during Pregnancy: A Narrative Review**. *Ann. Nutr. Metabol.* (2017) **69** 171-181. DOI: 10.1159/000453077
20. Poston L.. **Maternal obesity, gestational weight gain and diet as determinants of offspring long term health**. *Best Prac. Res. Clin. Endocrinol. Metabol.* (2012) **26** 627-639. DOI: 10.1016/j.beem.2012.03.010
21. Rodriguez A., Miettunen J., Henriksen T.B., Olsen J., Obel C., Taanila A., Ebeling H., Linnet K.M., Moilanen I., Järvelin M.R.. **Maternal adiposity prior to pregnancy is associated with ADHD symptoms in offspring: Evidence from three prospective pregnancy cohorts**. *Int. J. Obes.* (2008) **32** 550-557. DOI: 10.1038/sj.ijo.0803741
22. Sanchez C.E., Barry C., Sabhlok A., Russell K., Majors A., Kollins S.H., Fuemmeler B.F.. **Maternal pre-pregnancy obesity and child neurodevelopmental outcomes: A meta-analysis**. *Obes. Rev.* (2018) **19** 464-484. DOI: 10.1111/obr.12643
23. Boots C.E., Stephenson M.D.. **Does obesity increase the rate of miscarriage in spontanous conception: A systematic review**. *Fertil. Steril.* (2011) **96** S284. DOI: 10.1016/j.fertnstert.2011.07.1083
24. Ghimire P.R., Akombi-Inyang B.J., Tannous C., Agho K.E.. **Association between obesity and miscarriage among women of reproductive age in Nepal**. *PLoS ONE* (2020) **15**. DOI: 10.1371/journal.pone.0236435
25. Brewer C.J., Balen A.H.. **The adverse effects of obesity on conception and implantation**. *Reproduction* (2010) **140** 347-364. DOI: 10.1530/REP-09-0568
26. Sebire N.J., Jolly M., Harris J.P., Wadsworth J., Joffe M., Beard R.W., Regan L., Robinson S.. **Maternal obesity and pregnancy outcome: A study of 287 213 pregnancies in London**. *Int. J. Obes.* (2001) **25** 1175-1182. DOI: 10.1038/sj.ijo.0801670
27. Chu S.Y., Kim S.Y., Lau J., Schmid C.H., Dietz P.M., Callaghan W.M., Curtis K.M.. **Maternal obesity and risk of stillbirth: A metaanalysis**. *Am. J. Obstet. Gynecol.* (2007) **197** 223-228. DOI: 10.1016/j.ajog.2007.03.027
28. Lindam A.. **High Maternal Body Mass Index in Early Pregnancy and Risks of Stillbirth and Infant Mortality-A Population-Based Sibling Study in Sweden**. *Am. J. Epidemiol.* (2016) **184** 98-105. DOI: 10.1093/aje/kww046
29. Marchi J., Berg M., Dencker A., Olander E.K., Begley C.. **Risks associated with obesity in pregnancy, for the mother and baby: A systematic review of reviews**. *Obesity Rev.* (2015) **16** 621-638. DOI: 10.1111/obr.12288
30. Stothard K.J., Tennant P.W., Bell R., Rankin J.. **Maternal overweight and obesity and the risk of congenital anomalies: A systematic review and meta-analysis**. *JAMA* (2009) **301** 636-650. DOI: 10.1001/jama.2009.113
31. Jolly M.C., Sebire N.J., Harris J.P., Regan L., Robinson S.. **Risk factors for macrosomia and its clinical consequences: A study of 350,311 pregnancies**. *Eur. J. Obstet. Gynecol. Reprod. Biol.* (2003) **111** 9-14. DOI: 10.1016/S0301-2115(03)00154-4
32. Kirchengast S., Das M., Bose K.. **Increasing incidence of macrosomia: The impact of maternal somatic and behavioral parameters on newborn weight status**. *Obesity and Syndrome X: A Global Public Health Burden* (2019) 143-166
33. Schummers L., Hutcheon J.A., Bodnar L.M., Lieberman E., Himes K.P.. **Risk of Adverse Pregnancy Outcomes by Prepregnancy Body Mass Index: A Population-Based Study to Inform Prepregnancy Weight Loss Counseling**. *Obstet. Gynecol.* (2015) **125** 133-143. DOI: 10.1097/AOG.0000000000000591
34. Smith G.C., Shah I., Pell J.P., Crossley J.A., Dobbie R.. **Maternal Obesity in Early Pregnancy and Risk of Spontaneous and Elective Preterm Deliveries: A Retrospective Cohort Study**. *Am. J. Public Health* (2007) **97** 157-162. DOI: 10.2105/AJPH.2005.074294
35. Persson M., Johansson S., Villamor E., Cnattingius S.. **Maternal overweight and obesity and risks of severe birth-asphyxia-related complications in term infants: A population-based cohort study in Sweden**. *PLoS Med.* (2014) **11** e1001648. DOI: 10.1371/journal.pmed.1001648
36. Aslam H.M., Saleem S., Afzal R., Iqbal U., Saleem S.M., Shaikh MW A., Shahid N.. **Risk factors of birth asphyxia**. *Ital. J. Pediatr.* (2014) **40** 94. DOI: 10.1186/s13052-014-0094-2
37. Alonso-Alconada D., Hassell K.J., Robertson N.J., Santos A.R.. **Birth Asphyxia: Prevalence, Causes, Clinical Progression and Neuroprotection**. *Asphyxia: Risk Factors, Prevalence and Neurological Impacts* (2015) 5-20
38. Goodwin T.M., Belai I., Hernandez P., Durand M., Paul R.H.. **Asphyxial complications in the term newborn with severe umbilical acidemia**. *Am. J. Obstet. Gynecol.* (1992) **167** 1506-1512. DOI: 10.1016/0002-9378(92)91728-S
39. Heller G., Schnell R., Misselwitz B., Schmidt S.. **Umbilical blood pH, Apgar scores, and early neonatal mortality**. *Zeitschr Geburtshilfe Neonatol.* (2003) **207** 84-89. DOI: 10.1055/s-2003-40975
40. Malin G.L., Morris R.K., Khan K.S.. **Strength of association between umbilical cord pH and perinatal and long term outcomes: Systematic review and meta-analysis**. *BMJ* (2010) **340** c1471. DOI: 10.1136/bmj.c1471
41. Zhu T., Tang J., Zhao F., Qu Y., Mu D.. **Association between maternal obesity and offspring Apgar score or cord pH: A systematic review and meta-analysis**. *Sci. Rep.* (2015) **5** 18386. DOI: 10.1038/srep18386
42. **Demographic Yearbook 2021**. (2021)
43. Hales C.M., Carroll M.D., Fryar C.D., Ogden C.L.. **Prevalence of Obesity and Severe Obesity Among Adults: United States, 2017–2018**. *NCHS Data Brief.* (2020) **360** 1-8
44. Araujo Júnior E., Peixoto A.B., Zamarian AC P., Elito Júnior J., Tonni G.. **Macrosomia**. *Best Prac. Res. Clinl. Obstet. Gynaecol.* (2017) **38** 83-96. DOI: 10.1016/j.bpobgyn.2016.08.003
45. Apgar V.. **A proposal for a new method of evaluation of the newborn infant**. *Curr. Res. Anesth. Analg.* (1953) **32** 260-267. DOI: 10.1213/00000539-195301000-00041
46. Mogos M., Herghelegiu C.G., Ioan R.G., Ionescu C.A., Neacsu A.. **Determining an Umbilical Cord pH Cutoff Value for Predicting Neonatal Morbidity Related to Intrapartum Hypoxia**. *Rev. Chim.* (2019) **70** 605-607. DOI: 10.37358/RC.19.2.6965
47. Uslu S., Bulbul A., Can E., Zubarioglu U., Salihoglu O., Nuhoglu A.. **Relationship Between Oxygen Saturation and Umbilical Cord pH Immediately After Birth**. *Pediatr. Neonatol.* (2012) **53** 340-345. DOI: 10.1016/j.pedneo.2012.08.005
48. 48.
World Health Organisation
Physical Status: The Use and Interpretation of AnthropometryWHO Technical Report SeriesWHOGeneva, Switzerland2000. *Physical Status: The Use and Interpretation of Anthropometry* (2000)
49. Kirchengast S., Hartmann B.. **Pregnancy Outcome during the First COVID 19 Lockdown in Vienna, Austria**. *Int. J. Environ. Res. Public Health* (2021) **18**. DOI: 10.3390/ijerph18073782
50. Motedayen M., Rafiei M., Rezaei Tavirani M., Sayehmiri K., Dousti M.. **The relationship between body mass index and preeclampsia: A systematic review and meta-analysis**. *Int. J. Reprod. Biomed.* (2019) **17** 463-472. DOI: 10.18502/ijrm.v17i7.4857
51. Khalil A., Syngelaki A., Maiz N., Zinevich Y., Nicolaides K.H.. **Maternal age and adverse pregnancy outcome: A cohort study**. *Ultrasound Obstet. Gynecol.* (2013) **42** 634-643. DOI: 10.1002/uog.12494
52. Deierlein A., Siega-Riz A.M., Gillman M., Poston L.. **Long-term consequences of obesity in pregnancy for the mother**. *Maternal Obesity* (2012) 81-86. DOI: 10.1017/CBO9780511782466.009
53. Forno E., Young O.M., Kumar R., Simhan H., Celedón J.C.. **Maternal obesity in pregnancy, gestational weight gain, and risk of childhood asthma**. *Pediatrics* (2014) **134** e535-e546. DOI: 10.1542/peds.2014-0439
54. Tenenbaum-Gavish K., Hod M.. **Impact of Maternal Obesity on Fetal Health**. *Fetal Diagn. Ther.* (2013) **34** 1-7. DOI: 10.1159/000350170
55. Stotland N.E., Caughey A.B., Breed E.M., Escobar G.J.. **Risk factors and obstetric complications associated with macrosomia**. *Int. J. Gynecol. Obstet.* (2004) **87** 220-226. DOI: 10.1016/j.ijgo.2004.08.010
56. Gibbons L., Belizán J.M., Lauer J.A., Betrán A.P., Merialdi M., Althabe F.. **The global numbers and costs of additionally needed and unnecessary caesarean sections performed per year: Overuse as a barrier to universal coverage**. *World Health Rep.* (2010) **30** 1-31
57. Heslehurst N., Simpson H., Ells L.J., Rankin J., Wilkinson J., Lang R., Brown T.J., Summerbell C.D.. **The impact of maternal BMI status on pregnancy outcomes with immediate short-term obstetric resource implications: A meta-analysis**. *Obes. Rev.* (2008) **9** 635-683. DOI: 10.1111/j.1467-789X.2008.00511.x
58. van Ham MA P.C., van Dongen PW J., Mulder J.. **Maternal consequences of caesarean section. A retrospective study of intra-operative and postoperative maternal complications of caesarean section during a 10-year period**. *Eur. J. Obstet. Gynecol. Reprod. Biol.* (1997) **74** 1-6. DOI: 10.1016/S0301-2115(97)02725-5
59. Weiss J.L., Malone F.D., Emig D., Ball R.H., Nyberg D.A., Comstock C.H., Saade G., Eddleman K., Carter S.M., Craigo S.D.. **Obesity, obstetric complications and cesarean delivery rate–a population-based screening study**. *Am. J. Obstet. Gynecol.* (2004) **190** 1091-1097. DOI: 10.1016/j.ajog.2003.09.058
60. Whitty R.J., Maxwell C.V., Carvalho JC A.. **Complications of neuraxial anesthesia in an extreme morbidly obese patient for cesarean section**. *Intl. J. Obstet. Anesth.* (2007) **16** 139-144. DOI: 10.1016/j.ijoa.2006.08.011
61. Ri M., Aikou S., Seto Y.. **Obesity as a surgical risk factor**. *Ann. Gastroenterol. Surg.* (2018) **2** 13-21. DOI: 10.1002/ags3.12049
62. Eskild A., Nesheim B.-I., Busund B., Vatten L., Vangen S.. **Childbearing or induced abortion: The impact of education and ethnic background. Population study of Norwegian and Pakistani women in Oslo, Norway**. *Acta Obstet. Gynecol. Scand.* (2007) **86** 298-303. DOI: 10.1080/00016340601133780
63. Pigeyre M., Rousseaux J., Trouiller P., Dumont J., Goumidi L., Bonte D., Dumont M.-P., Chmielewski A., Duhamel A., Amouyel P.. **How obesity relates to socio-economic status: Identification of eating behavior mediators**. *Int. J. Obes.* (2016) **40** 1794-1801. DOI: 10.1038/ijo.2016.109
64. Newton S., Braithwaite D., Akinyemiju T.F.. **Socio-economic status over the life course and obesity: Systematic review and meta-analysis**. *PLoS ONE* (2017) **12**. DOI: 10.1371/journal.pone.0177151
65. Daniels S.R.. **The Use of BMI in the Clinical Setting**. *Pediatrics* (2009) **124** S35-S41. DOI: 10.1542/peds.2008-3586F
|
---
title: Renoprotective Impacts of Inonotus obliquus Ethanol-Ethyl Acetate Extract on
Combined Streptozotocin and Unilateral Nephrectomy-Induced Diabetic Nephropathy
in Mice
authors:
- Kuang-Hsing Chiang
- Yi-Chun Chiu
- Noi Yar
- Yu-Chun Chen
- Chia-Hui Cheng
- Yi-Chien Liu
- Chia-Yu Chang
- Jiunn-Jye Chuu
journal: International Journal of Molecular Sciences
year: 2023
pmcid: PMC10002342
doi: 10.3390/ijms24054443
license: CC BY 4.0
---
# Renoprotective Impacts of Inonotus obliquus Ethanol-Ethyl Acetate Extract on Combined Streptozotocin and Unilateral Nephrectomy-Induced Diabetic Nephropathy in Mice
## Abstract
Diabetes nephropathy (DN) is one of the most common causes of end stage renal disease (ESRD) globally. Medication options to stop or slow the progression of chronic renal disease (CKD) are limited, and patients with DN remain at a high risk of developing renal failure. Inonotus obliquus extracts (IOEs) of Chaga mushroom have been shown to have anti-glycemic, anti-hyperlipidemia, antioxidant, and anti-inflammatory effects against diabetes. In this study, we examined the potential renal protective role of an ethyl acetate layer after water-ethyl acetate separation from *Inonotus obliquus* ethanol crude extract (EtCE-EA) from Chaga mushrooms in diabetic nephropathy mice after preparation with $\frac{1}{3}$ NT + STZ. Our data showed that treatment with EtCE-EA can effectively regulate blood glucose, albumin-creatinine ratio, serum creatinine, and blood urea nitrogen (BUN) levels, and it can improve the renal damage in $\frac{1}{3}$ NT + STZ-induced CRF mice with an increase in concentration (100, 300, and 500 mg/kg). In the immunohistochemical staining test, EtCE-EA can effectively reduce the expression of TGF-β and α-SMA after induction according to the increase in the concentration (100 mg/kg, 300 mg/kg), thereby slowing down the degree of kidney damage. Our findings demonstrate that EtCE-EA could provide renal protection in diabetes nephropathy, possibly due to the decreased expression of transforming growth factor-β1 and α-smooth muscle actin.
## 1. Introduction
Chronic kidney disease (CKD) is a global public health problem, and its prevalence and incidence have significantly increased in the past two decades [1]. The global burden of CKD is rapidly increasing, and it is expected to become the fifth most common cause of years of life lost globally by 2040 [2]. The prevalence of chronic kidney disease (CKD) has increased in recent decades alongside an increase in diabetes and hypertension, the main drivers of CKD [3]. Despite showing a decline in mortality due to the advancements in medical treatment in patients with end-stage kidney disease (ESKD), it has still remained one of the leading causes of death worldwide [4]. Globally, approximately 850 million people were reported to be affected by CKD in 2017 [3]. In 2010, 2.6 million people worldwide received renal replacement therapy, yet an estimated equivalent number died in the same year owing to a lack of access to dialysis and transplantation, particularly in low-income countries [5]. This contrasts with that of other major chronic illnesses, such as cardiovascular and respiratory disorders, whose effects on mortality are decreasing.
CKD can be defined as a persistent presence of kidney damage or decreased kidney function for more than three months, irrespective of the cause, and classified by cause, GFR category (G1–G5), and albuminuria category (A1–A3) [6]. Globally, diabetes and/or hypertension are the most prevalent causes of CKD [7]. The rising prevalence of type 2 diabetes is causing an increase in the number of patients with ESKD caused by diabetic nephropathy (DM) [8]. The presence of CKD is markedly higher in patients with diabetes. Diabetic nephropathy (DN) is the most common complication of diabetes mellitus, affecting approximately $40\%$ of patients with type II diabetes, and is a leading cause of end-stage renal disease (ESRD) worldwide in the last decade [9]. Therefore, the management of diabetes is a major component of CKD prevention. In type 2 DM, hyperglycemia leads to the elevation of key pathogeneses of renal damage, such as oxidative stress, insulin resistance, and pro-inflammatory cytokines, and glycemic control may delay the development and progression of CKD [10].
Moreover, DN is not the only cause of CKD in diabetes patients. The prevalence of nondiabetic kidney disease (NDKD) caused by factors irrelevant to DM, such as immunoglobulin A nephropathy (IgA N) and membranous nephropathy (MN), varies from 12 to $79\%$ in adults with DM [11]. In contrast to diabetic nephropathy, many kinds of non-diabetic kidney disease can be effectively treated (e.g., glomerulonephritis with immunosuppressive medication) [12]. New anti-diabetes agents (glucagon-like peptide-1 receptor (GLP-1R), agonists, dipeptidyl peptidase-4 (DPP-4) inhibitors, and sodium-glucose transporter-2 (SGLT-2) inhibitors) were found to have renal protective effects via anti-hypertensive, hemodynamic stabilization, anti-inflammatory, and anti-oxidative actions [13,14,15,16]. Modifiable risk factors for the development and progression of CKD in diabetes patients include systemic hypertension, proteinuria, and metabolic factors, such as insulin resistance, dyslipidemia, and hyperuricemia, etc. [ 17]. Thus, regardless of etiology, either DN or NDKD, control of glucose, hypertension, diet, and body weight is essential in the prevention of kidney disease in diabetes patients. In addition to glucose-lowering therapies, lifestyle interventions, including diet are associated with clinically significant improvements in diabetes control [18]. Studies have suggested that functional foods may improve hyperglycemia by modulating carbohydrate and lipid metabolism in adipose tissues and also by reducing oxidative stress and inflammatory processes, and subsequently, they could prevent the development of diabetes nephropathy [19].
Nowadays no definitive drug is available to stop or slow down the progression of chronic renal disease, and the medication options are influenced by the presence of comorbid diseases the patients have and their individual risk of complications. The *Inonotus obliquus* mushroom, also known as Chaga, mainly grows in cold areas (for example in northeast China, northern Europe, and Russia) and is used traditionally in the treatment of diabetes, cardiovascular disease, and gastrointestinal diseases [20]. To date, a few studies have suggested the significant therapeutic potential of *Inonotus obliquus* extracts (IOEs), which have been shown to have therapeutic effects against diabetes via multiple pathways including anti-glycemic, anti-hyperlipidemia, antioxidant, and anti-inflammatory effects in various studies [21]. The low molecular weight of IOEs has been shown to restore the integrity of the glomerular capsules, increase the number of glomerular mesangial cells, and protect renal tubular cells against STZ + AGEs-induced glucotoxicity in diabetic mice [22]. However, there has been no work so far presenting scientific findings on the renal protective effect of IOE in CKD patients. Our study aims to explore the potential possibility of using IOE as renal protective medication in CKD patients.
## 2. Results
Before evaluating the effect of *Inonotus obliquus* extracts (IOEs) on STZ-induced porcine proximal tubular (LLC-PK1) cells, a cell viability assay was performed to determine the appropriate concentration of IOEs for further STZ-induced improvement assays. The results showed better viability in the ACEI (1 mg/mL), EtCE (1 mg/mL), and HWCE (1 mg/mL) compared to the vehicle control ($p \leq 0.05$, $p \leq 0.05$ and $p \leq 0.05$, respectively) while more cytotoxicity and decreased viability were observed in EtCE-EA (1 mg/mL), EtCE-nB (1 mg/mL), and EtCE-W (1 mg/mL) compared to the vehicle control ($p \leq 0.05$, $p \leq 0.001$, and $p \leq 0.01$, respectively) at 72 h (Figure 1).
In order to further simulate the safe dose of induced animals, renal tubular epithelial cells (LLC-PK1) in vitro were treated with STZ (10 mM) at 24 h and 72 h to cause cellular injury. The proximal renal tubular cells of pigs were treated with IOEs for 24 h after STZ 10 mM injury. At either 24 h or 72 h, only the EtCE-EA (100 μg/mL) group, similar to the ACEI (100 μg/mL) and ARB (100 μg/mL), had a better survival rate when compared to the vehicle control ($p \leq 0.05$, $p \leq 0.05$ and $p \leq 0.05$, respectively). However, the remaining IOEs, including the EtCE (100 μg/mL), EtCE-nB (100 μg/mL), EtCE-W (100 μg/mL), and HWCE (100 μg/mL) groups, revealed low cell viability after being co-treated with STZ (10 mM) ($p \leq 0.001$, $p \leq 0.001$, and $p \leq 0.001$, respectively) at 72 h. Each value represents the mean ± SE of three replicated experiments, and the results are expressed as population growth (control as $100\%$) (Figure 2).
In order to achieve the ideal renal injury index value in this animal model, the combined induction of the chemical drug STZ at a medium dose (75 mg/kg) and high dose (100 mg/kg) was done at one week after the operation. Urinary albumin to creatinine ratio (ACR) has been used as the preferred indicator for quantifying albuminuria in terms of biochemical values and included in the indicators for assessing the risk of renal failure. The experimental results of $\frac{1}{3}$ NT + STZ 75 mg/kg compared with STZ 75 mg/kg reached the expected index Albumin-Creatinine Ratio of 200 mg/g or more while the result of $\frac{1}{3}$ NT + STZ 100 mg/kg compared with STZ 100 mg/kg has increased ACR to more than 300 mg/g ($p \leq 0.05$ and $p \leq 0.01$, respectively) (Figure 3A) with severe proteinuria, which destroyed glomerular and renal tubular cells in the kidney, making it from chronic renal failure to early renal failure. From blood Creatinine and Blood Urea Nitrogen values, the $\frac{1}{3}$ NT + STZ 100 mg/kg group showed the most severe damage, followed by $\frac{1}{3}$ NT + STZ 75 mg/kg ($p \leq 0.001$ and $p \leq 0.001$, respectively) (Figure 3B,C). In Figure 3D, the survival rate of the $\frac{1}{3}$ NT + STZ 100 mg/kg group was about $20\%$ in the third week, and that of the $\frac{5}{6}$ NT operation group in the first week was about $10\%$. Chronic renal failure models all lead to weight loss. Thus, the $\frac{1}{3}$ NT + STZ 75 mg/kg group with a survival rate of $80\%$ was selected as the animal model for the follow-up experiment.
Two weeks after dosing, we first measured fasting blood glucose (Figure 4A), mainly to observe the changes in chronic kidney disease. After induction of renal failure in experimental mice, due to the physical damage of the surgical side of the kidney, the other side will have compensatory hypertrophy, glomerular sclerosis, and functional decline. After administration of IOEs, we tested whether it can improve the oxidative damage of STZ to pancreatic β cells and cause hyperglycemia in vivo. We found that the treatment group with the IOE, EtCE-EA (300 and 500 mg/kg) can effectively utilize the glucose in the body. On the contrary, HWCE (500 mg/kg) group failed to effectively regulate blood sugar compared to the $\frac{1}{3}$ NT + STZ alone ($p \leq 0.01$, $p \leq 0.01$, and $p \leq 0.05$, respectively) (Figure 4A). The ratio of albumin to creatinine in urine was observed with Albumin-Creatinine Ratio (Figure 4B) in urine biochemical values. We also found that EtCE-EA (300 and 500 mg/kg) can effectively improve the discharge of proteinuria caused by renal damage compared to the $\frac{1}{3}$ NT + STZ alone ($p \leq 0.05$ and $p \leq 0.01$, respectively); however, the HWCE (500 mg/kg) group, still failed to effectively improve the renal damage caused by chronic renal failure and the damage degree is more serious than ACEI group ($p \leq 0.05$ and $p \leq 0.05$, respectively) (Figure 4B). At the same time, the blood creatinine (Figure 4C) and blood urea nitrogen (Figure 4D) were observed, and the results showed the EtCE-EA, according to the increase of its concentration (100 mg/kg, 300 mg/kg, 500 mg/kg), can effectively improve the abnormal metabolism caused by chronic renal failure in blood creatinine ($p \leq 0.05$, $p \leq 0.05$, and $p \leq 0.01$, respectively) and blood urea nitrogen ($p \leq 0.05$, $p \leq 0.01$, and $p \leq 0.01$, respectively) compared to the $\frac{1}{3}$ NT + STZ alone. On the other hand, the HWCE (500 mg/kg) group is not effective in improving the damage to the kidney after $\frac{1}{3}$ NT + STZ induction in blood creatinine ($p \leq 0.05$) and blood urea nitrogen ($p \leq 0.05$) (Figure 4C,D).
In the $\frac{1}{3}$ NT + STZ plus EtCE-EA (300 mg/kg) (Figure 5D), the focal glomerulus remained relatively intact and numerous with hematoxylin and eosin staining. In the $\frac{1}{3}$ NT + STZ plus ACEI (20 mg/kg) (Figure 5G), the glomerulus remained relatively intact with positive collagen staining; weak collagen staining was present in the tubules and the interstitium. In $\frac{1}{3}$ NT + STZ treated mice (Figure 5F), obvious mesangial matrix accumulation with diffuse collagen fibril deposition in different compartments was observed. In $\frac{1}{3}$ NT + STZ plus EtCE-EA (300 mg/kg) treated mice (Figure 5I), the matrix accumulation and collagen staining were less severe than those in the control group ($\frac{1}{3}$ NT + STZ alone). In $\frac{1}{3}$ NT + STZ + HWCE (500 mg/kg) treated mice (Figure 5J), obvious mesangial matrix accumulation and diffuse collagen staining within renal tissues were observed. It can be seen from the histopathological section of the renal corpus with H&E stain that after induction, the accumulation of renal interstitium was less obvious in the ACEI (20 mg/kg) group, and the morphology of the glomerulus was similar to that of the induction group ($\frac{1}{3}$ NT+ STZ) (Figure 5A) and almost complete (Figure 5B). The IOE, EtCE-EA group, according to the increase of its concentration (100 mg/kg and 300 mg/kg), can effectively improve the induced glomerular atrophy and interstitial accumulation (Figure 5C,D). We examined sections of CRF mouse renal cortex using Masson’s trichrome (MT) staining to detect the severity of overt nephropathy indicated by collagen fibril deposition in glomeruli, tubules, and interstitium (5F–J). As shown in Figure 5F, in the slices of the induction group without treatment ($\frac{1}{3}$NT + STZ alone), there was a large amount of collagen deposition in the atrophic interstitium of the glomerulus. While in EtCE-EA group (100 mg/kg, 300 mg/kg), the deposition of collagen showed a tendency to slow down due to the increase in concentration (Figure 5H,I). However, in the HWCE (500 mg/kg) group (Figure 5J), although there was no slowing and improvement, it was worse when compared with the pathological state of the ACEI group (Figure 5G).
In the immunohistochemical staining test, the expression levels of renal fibrosis factors TGF-β (Figure 6A–E) and α-SMA (Figure 6F–J) were analyzed. According to the increase in the concentration (100 mg/kg, 300 mg/kg), EtCE-EA can effectively reduce the expression of TGF-β and α-SMA after induction, thereby slowing down the degree of kidney damage. While in the EtCE-EA (300 mg/kg) group, the expression amount is close to the expression amount of the ACEI (20 mg/kg) group, which shows that it can effectively inhibit the expression amount and slow down the level of fibrosis. Finally, by quantifying the positive cells (%) marked by TGF-β and α-SMA, it can be determined that the EtCE-EA (300 mg/kg), rather than EtCE-EA (100 mg/kg) and HWCE (500 mg/kg), significantly inhibits the formation of α-SMA myofibroblasts ($p \leq 0.05$, $p \leq 0.05$, and $p \leq 0.05$, respectively) by reducing the expression of TGF-β ($p \leq 0.01$, $p \leq 0.05$, and $p \leq 0.05$, respectively); therefore, it can improve the phenomenon of renal deterioration in the animal model of chronic renal failure compared with the ACEI group ($p \leq 0.01$, $p \leq 0.001$, respectively) (Figure 6K), thus confirming the therapeutic potential of *Inonotus obliquus* fruit bodies extract (IOE), the EtCE-EA in chronic renal failure.
## 3. Discussion
In recent years, growing interest was seen in the use of *Inonotus obliquus* extracts (IOEs) for the treatment of diabetes and renal disease. Still, a limited number of studies have demonstrated the therapeutic effectiveness of IOEs in the treatment of diabetic nephropathy. In this study, treatment with the extraction of EtCE-EA (Ethyl acetate layer after water-ethyl acetate separation from *Inonotus obliquus* ethanol crude extract) can effectively improve the renal damage in $\frac{1}{3}$ NT + STZ-induced CRF mice with an increase in concentration (100, 300, and 500 mg/kg). An effective reduction in the expression of TGF-β and α-SMA after induction and subsequent slowing of the degree of kidney damage was observed.
Chaga fungus was proven to possess antioxidant, hypoglycemic, hypolipidemic, and anti-tumor properties, and the use of Chaga extracts, IOEs in the treatment of diabetes and kidney disease has been examined by several scientific studies. Chaga extracts contain several compounds such as polysaccharides, triterpenes, and polyphenols [23]. The exact mechanisms of action for the hypoglycemic effect of IOEs have not been reached conclusion. So far, it has been described that I. obliquus polysaccharides in streptozotocin (STZ)-induced diabetic rats reduced blood glucose levels and restored the structure of β-cells after diabetes-induced cellular damage [24]. Wang et al. reported that *Inonotus obliquus* polysaccharides enhanced the serum levels of insulin and alleviated the metabolic derangement of glucose enzymes in the STZ-induced diabetic mice model [21]. Another study has found that the ingestion of *Inonotus obliquus* polysaccharide had improved serum insulin levels, moderately expanded the pancreatic islets, and reduced pancreatic injuries in alloxan-induced diabetic mice [25]. One of the main ingredients of *Inonotus obliquus* extract, Trametenolic acid (TA), was also recently reported to have a renal protective effect in diabetic nephropathy by relieving oxidative stress and inflammation via Nrf2/HO-1 and NF-κB signaling pathways [26]. However, the role of *Inonotus obliquus* ethyl acetate extract in the protection of renal impairment caused by diabetes still remains uncertain. We proposed that the critical issue could be the preparation of IOE, and our method with ethyl acetate extract was evaluated. Further, transforming growth factor-β (TGF-β) is a potent stimulator that drives fibrosis, and the downregulation of TGF-β has been found to significantly limit the fibrotic process in chronic kidney disease [27,28,29]. The induction of alpha-smooth muscle actin (α-SMA), a smooth muscle cell marker protein, increases extracellular matrix deposition and glomerulosclerosis, and a high α-SMA expression in kidneys is a hallmark of tubular epithelial-myofibroblast trans-differentiation [30]. Our results demonstrated dose-related improvements in blood glucose and renal function results with the EtCE-EA. We also found that EtCE-EA had a renal protective function through the downregulation of both TGF-β1 and α-SMA. These findings suggest that EtCE-EA could provide renal protection in diabetic mice with severe CKD, which requires additional clinical validation.
Animal models are crucial for pathological and clinical research on disease treatment therapies to understand therapeutic outcomes and drug safety. The process of selection of the animal model is a very intricate part as many factors need to be considered to reproduce the disease and pathology at the same level as that of humans [31]. Streptozotocin is one of the most commonly used substances to induce diabetes in experimental mice [32]. Also known as subtotal nephrectomy, $\frac{5}{6}$ nephrectomy has been a widely used model for studying CKD [33]. However, this model causes a great risk of hemorrhage and infection during surgery and high animal mortality [34]. In the present study, the selected animal model was created with the combination of STZ and $\frac{1}{3}$ nephrectomy to closely mimic chronic and more severe renal injury and to signify the protective effect of EtCE-EA.
Fasting blood glucose, the albumin-creatinine ratio (ACR), serum creatinine levels, and serum blood urine nitrogen (BUN) levels are the most used biochemical parameters to estimate the progression of renal disease and diabetes control. In our study, treatment with EtCE-EA can effectively regulate blood sugar, ACR, serum creatinine, and BUN according to the increase of its concentration (100 mg/kg, 300 mg/kg, 500 mg/kg) while HWCE treatment has caused more severe damage. According to previous studies, different extraction methods have exhibited different drug components and properties [35]. It is preliminarily inferred that the hot water extraction of Chaga mushroom directly dissolved potential substances that accelerate the deterioration and failure of the kidneys. Cases of oxalate-induced nephropathy from long-term ingestion of Chaga mushroom powder were reported in recent studies [36]. Oxalate, an organic acid found in Chaga mushroom extracts, can cause nephropathy from excessive intake [37]. It is found in high concentrations especially in water extracts of Chaga mushroom than in ethanolic extracts [38]. Therefore, noting the oxalate concentration of Chaga mushroom extracts and methods of extraction may be important in order to avoid or lessen the risk of oxalate nephropathy. Xu and co. reported that an ethanol extract of the dry matter of a culture broth of I. obliquus has shown significant anti-hyperglycaemic, as well as anti-lipid peroxidative effects, against alloxan-induced diabetic mice [34].
In this study, we examined the potential renal protective role of EtCE-EA from Chaga mushroom in mice after preparation with $\frac{1}{3}$ NT + STZ [36]. To demonstrate the renal protective role of EtCE-EA beyond its anti-diabetic effect, an animal model was systematically established to mimic pathophysiology of severe renal impairment in diabetes patients. In natural medicine, unlike conventional medicine, methods of preparation have an impact on the function/chemical property of the extracts. Thus, we further investigated the efficacy of IOEs by different methods. We believe that this study has provided strong evidence that IOE plays a renal protective role in diabetes nephropathy, which may be at least partially attributed to the decreased expression of transforming growth factor-β1 and α-smooth muscle actin, deepening the understanding functions of IOE in diabetic nephropathy.
## 4.1. Chemicals and Reagents
Culture medium RPMI-1640, fetal bovine serum, sodium bicarbonate, l-glutamine, and $0.05\%$ trypsin-EDTA were from Gibco Ltd. Streptozotocin (STZ) was from Sigma (Saint Louis, MO, USA). Inonotus obliquus fruit body was produced by TCM Biotech International Corp. (Xizhi District, New Taipei, Taiwan). Selective ACE Inhibitor was produced by Taiwan Tanabe Seiyaku Co., Ltd. (Nangang, Taipei, Taiwan). Renal tubular cells, LLC-PK1 were purchased from Food Industry Research and Development Institute (Eastern Hsin Chu, Taiwan). Culture medium RPMI-1640 was produced by Thermo Fisher Scientific Inc. (Waltham, MA, USA). Thiazolyl Blue Tetrazolium Bromide was produced by Sigma–Aldrich Inc. (Burlington, MA, USA). The rabbit polyclonal antibodies-TGF-β, Rabbit α-SMA Polyclonal Antibody were from Santa Cruz Biotechnology, Inc. (Delaware Ave, Santa Cruz, CA, USA).
## 4.2. Preparation of Inonotus obliquus Body Extract
An amount of 10 g of *Inonotus obliquus* fruiting body was taken out again; 100 mL of $95\%$ ethanol was added to extract for 24 h, and the suspension was obtained by centrifugation. The suspension was placed in an oven at 60 °C for 6 h, and concentrated to 10 mL to obtain the EtCE, ethanol crude extract of *Inonotus obliquus* fruiting bodies. Then, the EtCE was partitioned between water and ethyl acetate solution (v/$v = 1$:1 ratio) and centrifuged to obtain the ethyl acetate layer (EtCE-EA) and water layer. The water layer was partitioned between water and n-butanol (v/$v = 1$:1 ratio) to finally gain the n-butanol layer (EtCE-nB) and the water layer (EtCE-W), respectively. Then, the water layer was mixed with n-butanol (v/$v = 1$:1 ratio) to finally obtain the n-butanol layer and the water layer. Additionally, the hot water crude extract of *Inonotus obliquus* fruit bodies (HWCE) was also prepared (w/$v = 1$:10 ratio). These fractions above were placed in an oven at 60 °C for 6 h, and after concentrating the suspension to 10 mL, freeze-drying was performed to obtain the extracts of *Inonotus obliquus* fruiting body (IOEs). The MWs of HWCE are closely correlated with their functional bioactivities, and HWCE active polysaccharide mostly ranged 780 kDa (Mw), identified with gel permeation chromatography analysis. Meanwhile, the HWCE extract contained 17.11 mg/mL of total polysaccharide by the phenol-sulfuric said method and dinitrosalicylic acid colorimetric method.
## 4.3. Cell Viability Assay
When the growth density of LLC-PK1 in 75 flask reached 80–$90\%$, trypsin was added to interact with the cells after washing with PBS. After the cells were dispersed, the number of cells was counted with a hemocytometer. RPMI-1640 containing $10\%$ (v/v) FBS and $1\%$ penicillin was put in 96-well of cell culture plates, and 10 μL of FBS-containing culture medium was added to each well. After culturing for 24 h and after the cells were adsorbed to the bottom, the culture medium was removed, and STZ was added to stimulate oxidative stress, and then, the culture was continued for 48 and 72 h. An amount of 20 uL MTT solution (dissolved in 5 mg/mL PBS) was added per well, and the wells were put into a carbon dioxide incubator with $5\%$ CO2 at 37 °C and a constant temperature of $90\%$ to react with cells for 4 h. Then, the solution in each well was poured out, followed by adding 100 uL DMSO to dissolve the blue-violet crystals in each well and protect them from light for about 10 min. The 96-well plates were shaken evenly to ensure that the blue-violet crystals are completely dissolved, and the absorbance was read at a wavelength of 570 nm with an enzyme immunoassay reader. Because only living cells have active mitochondrial dehydrogenase, the measured light absorbance was proportional to cell viability. The higher the reading, the greater the relative number of living cells. In the experimental framework of the above cell viability, the time points were set as the first day and the third day, and two concentrations of STZ (10 μm) were used to destroy renal tubular cells. The other drug concentrations were ACEI (1 mg/mL and 100 μg/mL), ARB (1 mg/mL and 100 μg/mL) and EtCE (1 mg/mL and 100 μg/mL), EtCE-EA (1 mg/mL and 100 μg/mL), EtCE-nB (1 mg/mL and 100 μg/mL), EtCE-W (1 mg/mL and 100 μg/mL), and HWCE (1 mg/mL and 100 μg/mL). The third day of cell culture from the 96-well plate was the zeroth day of the above experimental design. After removing the culture medium and drugs of various concentrations were added, the experiment started, and each culture time point was reached after collecting the data.
## 4.4. Animal Preparation
Female CRF mice, 6 weeks of age, were purchased from the BioLASCO Taiwan Co., Ltd. (Nangang, Taipei, Taiwan). All animals were maintained in laminar flow cabinets with free access to food and water under specific pathogen-free conditions in facilities approved by the Accreditation of Laboratory Animal Care and the Institutional Animal Care and Use Committee (IACUC) of the Animal Research Committee of the Southern Taiwan University of Science and Technology, Tainan, Taiwan (Approval No. STUT-IACUC-98-05). Five mice per cage were fed with mouse chow and water ad libitum. The mice were acclimatized to the $\frac{12}{12}$ h light-dark cycle conditions in the cages and were kept in the housing facility for a 1-week acclimation period before the surgical injury. After the experimental animals were stably raised for two weeks, the kidneys ($\frac{1}{3}$ NT and $\frac{5}{6}$ NT residual kidney) were removed in vivo. Each experimental mouse was anesthetized by intraperitoneal injection, and the dose of anesthetic was 0.01 c.c. Through back surgery, the unilateral kidney was divided into three equal parts, the upper and lower parts were sutured, and antibiotics were applied to the wound to avoid infection and death. The survival rate and postoperative recovery were recorded.
The experimental animals were stably raised for two weeks and underwent Sham-operated kidney excision surgery. Each mouse was anesthetized with an intraperitoneal injection with an anesthetic dose of 0.01 c.c. From the back operation, the unilateral kidney was taken out of the abdominal cavity and put back and sutured without harming the kidney, and antibiotics were applied to the wound to avoid infection and death; the mice were observed for 7 days, and the survival rate and postoperative recovery were recorded. After the experimental animals were stably reared for two weeks and fed a normal diet, they were given a high dose of STZ 100 mg/kg/7 days, and middle doses of STZ 75 mg/kg/7 days were injected intraperitoneally to induce renal lesions (chemically-induced chronic nephropathy). Their survival was recorded, and weight monitoring was done weekly. After the experimental animals were established with $\frac{1}{3}$ NT residual kidney animal type, they were given a high dose of STZ 100 mg/kg/7 days and a middle dose of STZ 75 mg/kg/7 days $\frac{1}{3}$ NT plus STZ-induced chronic renal failure model.
The study group consisted of normal control, Sham group, $\frac{1}{3}$ NT, STZ 75 mg/kg/7 days/i.p., STZ 100 mg/kg/7 days/i.p., $\frac{1}{3}$ NT + STZ 100 mg/kg/7 days/i.p., and $\frac{1}{3}$ NT+ STZ 100 mg/kg/7 days i.p. According to each time point before and after surgery and before and after administration of high-dose STZ, urine protein content was detected by the metabolic cage method. At each time point, the mice were treated with STZ before and after treatment. One week after induced STZ, the laboratory animals were stimulated to urinate, and the urine was collected and sent to the medical laboratory for testing. Normal control, Sham group, $\frac{1}{3}$ NT, STZ 75 mg/kg/7 days/i.p., STZ 100 mg/kg/7 days/i.p., $\frac{1}{3}$ NT + STZ 100 mg/kg/7 days/i.p., $\frac{1}{3}$ NT + STZ 100 mg/kg/7 days i.p. blood collection was performed at each time point before and after surgery and before and after administration of high-dose STZ for the relevant biochemical value detection. If the animal did not need anesthesia or restraint with a restraint device, blood collection from the eye socket was used to confirm. After there was blood flowing out, about 0.5 mL of whole blood was collected with a blood collection tube, and after centrifugation at 5000 rpm for 10 min, 0.2 mL of serum was taken for biochemical value experiments. Blood biochemical values measured were creatinine (Cre), blood urea nitrogen (BUN). During the experiments, we recorded the survival rate and weekly body weight measurement of each mouse.
## 4.5. Blood Biochemical Profile
All groups’ blood samples were collected from the tail vein. FBG was measured with glucose oxidase strips (Easytouch, Taipei, Taiwan). Sham group, $\frac{1}{3}$ NT + STZ group, $\frac{1}{3}$ NT + STZ + ACEI (20 mg/kg) group, $\frac{1}{3}$ NT + STZ + EtCE-W/EA group (100 mg/kg, 300 mg/kg and 500 mg/kg), and $\frac{1}{3}$ NT + STZ + HWCE group (500 mg/kg) were continuously administrated for 2 weeks. According to each time point before and after surgery and before and after administration of high-dose STZ, urine protein content was detected by the metabolic cage. At each time point, the mice were treated with STZ before and after treatment. One week after induced STZ, the laboratory animals were stimulated to urinate, and the urine was collected and sent to the medical laboratory for testing. Blood collection was performed at each time point before and after surgery and before and after administration of high-dose STZ for relevant biochemical value detection. If the animal did not need anesthesia or restraint with a restraint device, blood collection from the eye socket was used to confirm. After there was blood flowing out, about 0.5 mL of whole blood was collected by a blood collection tube, and after centrifugation at 5000 rpm for 10 min, 0.2 mL of serum is taken for biochemical value experiments. Blood biochemical values (Glucose, Albumin-Creatinine Ratio (%), blood urea nitrogen (BUN), Creatinine (Cre)) were measured during the experiments. Blood serum metabolic enzymes were quantified using an enzyme-linked immunosorbent assay (ELISA).
## 4.6. Hematoxylin and Eosin (HE)
After mice were sacrificed, the excised kidney samples were fixed in formalin. The samples were dehydrated through a gradient mixture of ethyl alcohol and water, then rinsed with xylene before being embedded in paraffin. The formalin fixed tissues were sliced in 5 μm sections using a Microtome RM2135 (Leica Microsystems Inc., Bannockburn, IL, USA) and prepared on silane-coated slides. The slides were immersed in Tris-buffered saline (TBS, pH 7.4) after being rehydrated in graded ethanol solutions, dried at 37 °C overnight, and then stored at room temperature. The 5 μm kidney sections were stained with hematoxylin (Shandon™ Gill™ III) and Shandon Eosin Y (Thermo Scientific™). Lastly, the slides underwent microscopic examination by means of a Motic BA 400 microscope with Motic Advance 3.0 software (Motic Co., Fujian, China).
## 4.7. Masson Trichrome Staining
Kidney samples were fixed in $10\%$ formal-saline for 48 h and then dehydrated by successively passing through a gradient of mixtures of ethyl alcohol and water. The samples were then rinsed with xylene and embedded in paraffin. Kidney sections (5 μm thick) were prepared and stained with Dietrich scarlet-acid fuchsin solution for 15 min, then transferred directly to aniline blue solution and stained for 5–10 min. Finally, the sections were mounted using neutral deparaffinated xylene (DPX) medium for microscopic examination on a Motic BA 400 microscope using Motic Advance 3.0 software.
## 4.8. Immunohistochemical Stain
The kidney samples were fixed in formalin and dehydrated with a gradient mixture of ethyl alcohol and water. The samples were then rinsed with xylene and embedded in paraffin. The formalin-fixed tissues were sliced by Microtome RM2135 (Leica Microsystems Inc., Bannockburn, IL, USA) into 5 μm sections and placed on silane-coated slides. The slides were immersed in Tris-buffered saline (TBS, pH 7.4) after being rehydrated in graded ethanol solutions, dried at 37 °C overnight, and stored at room temperature. After that, the sections were soaked in $0.3\%$ H2O2 to block the endogenous peroxidase activity. They were then placed in the 10 mM citrate buffer solution (pH = 6.0) and microwave boiled for 10 min for completing antigen retrieval. The sections were incubated with primary antibodies against TGF-β and α-SMA (1:250 dilution) in a humidified chamber at room temperature for 2 h. The LSAB2 detection and DAB substrate kits were used for staining processes according to the manufacturer DAKO’s instructions. Finally, the sections were counterstained with hematoxylin (Shandon™ Gill™ III) and the number of stained nuclei (dark blue color) per square millimeter was calculated using an eyepiece graticule. For the positive labeling index for TGF-β and α-SMA, each tissue slide was illustrated as an average percentage of dividing the numbers of a TGF-β and α-SMA-positive cell (visualized in brown) by the total numbers of nuclei (visualized in blue). With each staining run, both positive and negative controls were provided, and overexpression was considered positive if more than $10\%$ of the cells were showing.
## 4.9. Statistical Analysis
All the results were presented as the mean ± standard deviation (SD). Differences between groups were evaluated with an analysis of variance and post hoc comparisons with the Bonferroni step-down (Holm) correction. Statistical analysis was performed using SigmaPlot software (version 10.0; SPSS Inc., Chicago, IL, USA). Post hoc testing of behavioral data utilized a two-tailed Welch’s t-test. Post hoc testing of biochemical data utilized a regression analysis. Each value represents the mean ± SD of 8 mice; p values less than 0.05 were considered statistically significant. The values * $p \leq 0.05$, ** $p \leq 0.01$, and *** $p \leq 0.001$ represent significant differences between the vehicle control and normal group (drinking water alone), and # $p \leq 0.05$, ## $p \leq 0.01$, and ### $p \leq 0.001$ represent significant differences from the $\frac{1}{3}$ NT group and same dose of STZ.
## References
1. Xie Y., Bowe B., Mokdad A.H., Xian H., Yan Y., Li T., Maddukuri G., Tsai C.Y., Floyd T., Al-Aly Z.. **Analysis of the Global Burden of Disease study highlights the global, regional, and national trends of chronic kidney disease epidemiology from 1990 to 2016**. *Kidney Int.* (2018) **94** 567-581. DOI: 10.1016/j.kint.2018.04.011
2. Li P.K., Garcia-Garcia G., Lui S.F., Andreoli S., Fung W.W., Hradsky A., Kumaraswami L., Liakopoulos V., Rakhimova Z., Saadi G.. **Kidney Health for Everyone Everywhere: From Prevention to Detection and Equitable Access to Care**. *Can. J. Kidney Health Dis.* (2020) **93** 111-122. DOI: 10.1177/2054358120910569
3. Kovesdy C.P.. **Epidemiology of chronic kidney disease: An update 2022**. *Kidney Int. Suppl.* (2022) **12** 7-11. DOI: 10.1016/j.kisu.2021.11.003
4. Jager K.J., Kovesdy C., Langham R., Rosenberg M., Jha V., Zoccali C.. **A single number for advocacy and communication-worldwide more than 850 million individuals have kidney diseases**. *Kidney Int.* (2019) **96** 1048-1050. DOI: 10.1016/j.kint.2019.07.012
5. Liyanage T., Ninomiya T., Jha V., Neal B., Patrice H.M., Okpechi I., Zhao M.H., Lv J., Garg A.X., Knight J.. **Worldwide access to treatment for end-stage kidney disease: A systematic review**. *Lancet* (2015) **385** 1975-1982. DOI: 10.1016/S0140-6736(14)61601-9
6. Levey A.S., Eckardt K.U., Dorman N.M., Christiansen S.L., Hoorn E.J., Ingelfinger J.R., Inker L.A., Levin A., Mehrotra R., Palevsky P.M.. **Nomenclature for kidney function and disease: Report of a Kidney Disease: Improving Global Outcomes (KDIGO) Consensus Conference**. *Kidney Int.* (2020) **97** 1117-1129. DOI: 10.1016/j.kint.2020.02.010
7. Chen T.K., Knicely D.H., Grams M.E.. **Chronic Kidney Disease Diagnosis and Management: A Review**. *JAMA* (2019) **322** 1294-1304. DOI: 10.1001/jama.2019.14745
8. Grujicic M., Salapura A., Basta-Jovanovic G., Figurek A., Micic-Zrnic D., Grbic A.. **Non-Diabetic Kidney Disease in Patients with Type 2 Diabetes Mellitus-11-Year Experience from a Single Center**. *Med. Arch.* (2019) **73** 87-91. DOI: 10.5455/medarh.2019.73.87-91
9. Martínez-Castelao A., Navarro-González J.F., Górriz J.L., de Alvaro F.. **The Concept and the Epidemiology of Diabetic Nephropathy Have Changed in Recent Years**. *J. Clin. Med.* (2015) **4** 1207-1216. DOI: 10.3390/jcm4061207
10. Amorim R.G., Guedes G.D.S., Vasconcelos S.M.L., Santos J.C.F.. **Kidney Disease in Diabetes Mellitus: Cross-Linking between Hyperglycemia, Redox Imbalance and Inflammation**. *Arq. Bras. Cardiol.* (2019) **112** 577-587. DOI: 10.5935/abc.20190077
11. Kritmetapak K., Anutrakulchai S., Pongchaiyakul C., Puapairoj A.. **Clinical and pathological characteristics of non-diabetic renal disease in type 2 diabetes patients**. *Clin. Kidney J.* (2018) **11** 342-347. DOI: 10.1093/ckj/sfx111
12. Zeng Y.-Q., Yang Y.-X., Guan C.-J., Guo Z.-W., Li B., Yu H.-Y., Chen R.-X., Tang Y.-Q., Yan R.. **Clinical predictors for nondiabetic kidney diseases in patients with type 2 diabetes mellitus: A retrospective study from 2017 to 2021**. *BMC Endocr. Disord.* (2022) **22**. DOI: 10.1186/s12902-022-01082-8
13. Chung S., Kim G.H.. **Use of Anti-Diabetic Agents in Non-Diabetic Kidney Disease: From Bench to Bedside**. *Life* (2021) **11**. DOI: 10.3390/life11050389
14. Kim S.J., Kwon S.K., Kim H.Y., Kim S.M., Bae J.W., Choi J.K.. **DPP-4 inhibition enhanced renal tubular and myocardial GLP-1 receptor expression decreased in CKD with myocardial infarction**. *BMC Nephrol.* (2019) **20**. DOI: 10.1186/s12882-019-1243-z
15. Cappetta D., Ciuffreda L.P., Cozzolino A., Esposito G., Scavone C., Sapio L., Naviglio S., D’Amario D., Crea F., Rossi F.. **Dipeptidyl Peptidase 4 Inhibition Ameliorates Chronic Kidney Disease in a Model of Salt-Dependent Hypertension**. *Oxidative Med. Cell Longev.* (2019) **2019** 8912768. DOI: 10.1155/2019/8912768
16. Liu W.J., Xie S.H., Liu Y.N., Kim W., Jin H.Y., Park S.K., Shao Y.M., Park T.S.. **Dipeptidyl peptidase IV inhibitor attenuates kidney injury in streptozotocin-induced diabetic rats**. *J. Pharmacol. Exp. Ther.* (2012) **340** 248-255. DOI: 10.1124/jpet.111.186866
17. Vaidya S.R., Aeddula N.R.. **Chronic renal failure**. *StatPearls [Internet]* (2021)
18. O’Donoghue G., O’Sullivan C., Corridan I., Daly J., Finn R., Melvin K., Peiris C.. **Lifestyle interventions to improve glycemic control in adults with type 2 diabetes living in low-and-middle income countries: A systematic review and meta-analysis of randomized controlled trials (RCTs)**. *Int. J. Environ. Res. Public Health* (2021) **18**. DOI: 10.3390/ijerph18126273
19. Mirmiran P., Bahadoran Z., Azizi F.. **Functional foods-based diet as a novel dietary approach for management of type 2 diabetes and its complications: A review**. *World J. Diabetes* (2014) **5** 267. DOI: 10.4239/wjd.v5.i3.267
20. Wang J., Wang C., Li S., Li W., Yuan G., Pan Y., Chen H.. **Anti-diabetic effects of Inonotus obliquus polysaccharides in streptozotocin-induced type 2 diabetic mice and potential mechanism via PI3K-Akt signal pathway**. *Biomed. Pharmacother.* (2017) **95** 1669-1677. DOI: 10.1016/j.biopha.2017.09.104
21. Szychowski K.A., Skóra B., Pomianek T., Gmiński J.. **Inonotus obliquus—From folk medicine to clinical use**. *J. Tradit. Complement. Med.* (2021) **11** 293-302. DOI: 10.1016/j.jtcme.2020.08.003
22. Chou Y.-J., Kan W.-C., Chang C.-M., Peng Y.-J., Wang H.-Y., Yu W.-C., Cheng Y.-H., Jhang Y.-R., Liu H.-W., Chuu J.-J.. **Renal protective effects of low molecular weight of Inonotus obliquus polysaccharide (LIOP) on HFD/STZ-induced nephropathy in mice**. *Int. J. Mol. Sci.* (2016) **17**. DOI: 10.3390/ijms17091535
23. Zhang Y., Liao H., Shen D., Zhang X., Wang J., Zhang X., Wang X., Li R.. **Renal Protective Effects of**. *Front. Pharmacol.* (2022) **12** 743931. DOI: 10.3389/fphar.2021.743931
24. Géry A., Dubreule C., André V., Rioult J.P., Bouchart V., Heutte N., de Pécoulas P.E., Krivomaz T., Garon D.. **Chaga (inonotus obliquus), a future potential medicinal fungus in oncology? A chemical study and a comparison of the cytotoxicity against human lung adenocarcinoma cells (a549) and human bronchial epithelial cells (beas-2b)**. *Integr. Cancer Ther.* (2018) **17** 832-843. DOI: 10.1177/1534735418757912
25. Wang J., Hu W., Li L., Huang X., Liu Y., Wang D., Teng L.. **Antidiabetic activities of polysaccharides separated from inonotus obliquus via the modulation of oxidative stress in mice with streptozotocin-induced diabetes**. *PLoS ONE* (2017) **12**. DOI: 10.1371/journal.pone.0180476
26. Sun J.-E., Ao Z.-H., Lu Z.-M., Xu H.-Y., Zhang X.-M., Dou W.-F., Xu Z.-H.. **Antihyperglycemic and antilipidperoxidative effects of dry matter of culture broth of inonotus obliquus in submerged culture on normal and alloxan-diabetes mice**. *J. Ethnopharmacol.* (2008) **118** 7-13. DOI: 10.1016/j.jep.2008.02.030
27. Gifford C.C., Tang J., Costello A., Khakoo N.S., Nguyen T.Q., Goldschmeding R., Higgins P.J., Samarakoon R.. **Negative regulators of TGF-β1 signaling in renal fibrosis; pathological mechanisms and novel therapeutic opportunities**. *Clin. Sci.* (2021) **135** 275-303. DOI: 10.1042/CS20201213
28. Meng X.-M., Nikolic-Paterson D.J., Lan H.Y.. **Tgf-β: The master regulator of fibrosis**. *Nat. Rev. Nephrol.* (2016) **12** 325-338. DOI: 10.1038/nrneph.2016.48
29. Ma T.T., Meng X.M.. **TGF-β/Smad and Renal Fibrosis**. *Adv. Exp. Med. Biol.* (2019) **1165** 347-364. PMID: 31399973
30. Mukherjee P., Roy S., Ghosh D., Nandi S.K.. **Role of animal models in biomedical research: A review**. *Lab. Anim. Res.* (2022) **38** 18. DOI: 10.1186/s42826-022-00128-1
31. Tesch G.H., Allen T.J.. **Rodent models of streptozotocin-induced diabetic nephropathy (methods in renal research)**. *Nephrology* (2007) **12** 261-266. DOI: 10.1111/j.1440-1797.2007.00796.x
32. Yang H.C., Zuo Y., Fogo A.B.. **Models of chronic kidney disease**. *Drug Discov. Today Dis. Model.* (2010) **7** 13-19. DOI: 10.1016/j.ddmod.2010.08.002
33. Lim B.J., Yang H.C., Fogo A.B.. **Animal models of regression/progression of kidney disease**. *Drug Discov. Today Dis. Model.* (2014) **11** 45-51. DOI: 10.1016/j.ddmod.2014.06.003
34. Xu H.-Y., Sun J.-E., Lu Z.-M., Zhang X.-M., Dou W.-F., Xu Z.-H.. **Beneficial effects of the ethanol extract from the dry matter of a culture broth of inonotus obliquus in submerged culture on the antioxidant defence system and regeneration of pancreatic β-cells in experimental diabetes in mice**. *Nat. Prod. Res.* (2010) **24** 542-553. DOI: 10.1080/14786410902751009
35. Lee S., Lee H.Y., Park Y., Ko E.J., Ban T.H., Chung B.H., Lee H.S., Yang C.W.. **Development of end stage renal disease after long-termingestion of chaga mushroom: Case report and review of literature**. *J. Korean Med. Sci.* (2020) **35** e122. DOI: 10.3346/jkms.2020.35.e122
36. Kikuchi Y., Seta K., Ogawa Y., Takayama T., Nagata M., Taguchi T., Yahata K.. **Chaga mushroom-induced oxalate nephropathy**. *Clin. Nephrol.* (2014) **81** 440-444. DOI: 10.5414/CN107655
37. Glamočlija J., Ćirić A., Nikolić M., Fernandes Â., Barros L., Calhelha R.C., Ferreira I.C.F.R., Soković M., van Griensven L.J.L.D.. **Chemical characterization and biological activity of chaga (inonotus obliquus), a medicinal “mushroom”**. *J. Ethnopharmacol.* (2015) **162** 323-332. DOI: 10.1016/j.jep.2014.12.069
38. Uil M., Scantlebery A.M.L., Butter L.M., Larsen P.W.B., de Boer O.J., Leemans J.C., Florquin S., Roelofs J.J.T.H.. **Combining streptozotocin and unilateral nephrectomy is an effective method for inducing experimental diabetic nephropathy in the ‘resistant’ C57Bl/6J mouse strain**. *Sci. Rep.* (2018) **8** 5542. DOI: 10.1038/s41598-018-23839-9
|
---
title: 'Neuropsychological Performance and Cardiac Autonomic Function in Blue- and
White-Collar Workers: A Psychometric and Heart Rate Variability Evaluation'
authors:
- Ardalan Eslami
- Najah T. Nassif
- Sara Lal
journal: International Journal of Environmental Research and Public Health
year: 2023
pmcid: PMC10002343
doi: 10.3390/ijerph20054203
license: CC BY 4.0
---
# Neuropsychological Performance and Cardiac Autonomic Function in Blue- and White-Collar Workers: A Psychometric and Heart Rate Variability Evaluation
## Abstract
The 21st century has brought a growing and significant focus on performance and health within the workforce, with the aim of improving the health and performance of the blue- and white-collar workforce. The present research investigated heart rate variability (HRV) and psychological performance between blue and white-collar workers to determine if differences were evident. A total of 101 workers ($$n = 48$$ white-collar, $$n = 53$$ blue-collar, aged 19–61 years) underwent a three lead electrocardiogram to obtain HRV data during baseline (10 min) and active (working memory and attention) phases. The Cambridge Neuropsychological Test Automated Battery, specifically the spatial working memory, attention switching task, rapid visual processing and the spatial span, were used. Differences in neurocognitive performance measures indicated that white-collar workers were better able to detect sequences and make less errors than blue-collar workers. The heart rate variability differences showed that white-collar workers exhibit lower levels of cardiac vagal control during these neuropsychological tasks. These initial findings provide some novel insights into the relationship between occupation and psychophysiological processes and further highlight the interactions between cardiac autonomic variables and neurocognitive performance in blue and white-collar workers.
## 1. Introduction
In the 21st century, productivity is a crucial element in the strength and sustainability of a company’s gross business performance [1]. Both white-collar and blue-collar professions often require executive function to perform the tasks required for their work. However, compared to white-collar workers, blue-collar employees have been shown to have a higher prevalence of a large range of health complications, particularly cardiovascular disease (CVD) [2]. The workplace can often play a major role in the onset of cardiovascular disease and the current European guidelines on the prevention of CVD recommend an assessment of long-term stress, which includes occupational psychological stressors [3].
Executive cognitive function refers to a family of mental processes that are recruited for concentration and attention [4]. These executive functions have also been implicated in other aspects of health, such as obesity [5], occupational prosperity [6], and public safety [7]. Increasing evidence suggests an association between CVD and reduced psychological performance; however, few recent studies have delved into the inner workings which relate working memory (WM) to CVD. Additionally, many previous studies linking memory and working memory deficits to cardiac failures have mostly focused on patients with severe CVD [8].
Heart rate variability (HRV) has been extensively used to reflect the sympathetic and parasympathetic activity of the autonomic nervous system [9]. Furthermore, previous research has linked HRV to CVD [10,11], as well as various psychological processes [12,13]. Hansen et al. [ 12] established a relationship between HRV and performance tasks that taxed executive function in normal subjects ($$n = 53$$ male, average age = 23 years) and found that the qualitative differences between task demands could be predicted by the subject’s cardiac vagal tone. Other researchers have investigated this connection, but vagal tone relationships remain largely unexplored [14]. Furthermore, in order to predict cognitive performance by utilising cardiac vagal tone as an independent variable, Johnsen et al. [ 15] investigated attentional bias in 20 patients with anxiety in a dental setting using a modified Stroop-test [16] (14 male and 6 female, mean age = 36 years). Results showed that poor attentional performance was characterized by reduced HRV as compared to patients with higher HRV [15].
This indication of decreased HRV with increased working memory load and higher HRV in better performers supports the notion that, during working memory function, HRV may qualitatively predict cognitive differences among individuals [17]. This also implies that executive performance and autonomic functions, such as HRV, may be adaptively regulated by an interrelated neural network. Therefore, HRV may provide an index of an individual’s ability to function effectively in a dynamic environment [17].
Limited research has linked working memory and attentional deficits to cardiac deficits [18], with most studies focused on end stage patients [19]. Therefore, more research needs to be centred around healthy individuals, which may implicate HRV as a pre-emptive biomarker for working memory and attentional performance.
This study aims to investigate neuropsychological processes (working memory and attention) in two major working populations, white-collar ($$n = 48$$) and blue-collar ($$n = 53$$) workers, further identifying the fundamental associations between working memory, attention, and HRV. Heart rate variability and executive function are evaluated in a sample of healthy blue and white-collar workers to better understand the cardiac autonomic vagal influence during neuropsychological performance and risk factors that may contribute to cardiovascular complications. It was hypothesized that [1] attentional states will increase cardiac vagal input, HF and RMSSD HRV in white-collar workers while indicating a decrease in blue-collar workers, and [2] spatial neuropsychological stress will exhibit a decrease in cardiac vagal input, HF and RMSSD HRV in white-collar workers and an increase in blue-collar workers.
## 2.1. Participant Recruitment
Healthy participants between the ages of 18–68 years ($$n = 101$$) were recruited from the community. Participants were required to abstain from caffeine and nicotine for 4 h and alcohol for 12 h prior to the commencement of testing. These factors are known to influence physiological measures and their restrictions enhance the reliability of the data. Additionally, participants with pre-study blood pressure (BP) measures greater than 160 mmHg (systolic)/or 100 mmHg (diastolic) were excluded [20]. Testing was conducted between 8:30 am and 12:00 pm to minimize the effect of circadian rhythm fluctuation [21] on the data obtained. No volunteer was excluded from the current study and written informed consent was obtained prior to commencement of the study protocol. This study was approved by the Institutional Human Research Ethics Committee of the University of Technology Sydney (HREC: 2014000110 and HREC ETH19-3676).
## 2.2. Experimental Methodology
Participants were seated for 5 min prior to three BP recordings using an automated monitor (OMRON IA1B, Kyoto, Japan). Three blood pressure readings were obtained both before and after the study protocol with 2-min intervals between each measurement [22]. Participants were then asked to complete the General Health Questionnaire (GHQ60) [23], which obtained detailed health information. Participants then underwent a baseline electrocardiogram (ECG) for 10 min followed by an ECG recording during the neurocognitive tasks performed. The ECG was obtained using a FlexComp Infiniti encoder (Thought Technology Ltd., Montreal, QC, Canada) and an ECG-Flex/Pro amplifier sensor (Thought Technology Ltd., Canada) connected to three electrode leads. BioGraph Infiniti software (T7900) (Thought Technology Ltd., Canada) was used to record and display the ECG wave. Prior to placement of the electrodes, the skin was cleaned using Liv-Wipe (Livingstone International Pty Ltd., Sydney, Australia) $70\%$ alcohol swabs. Disposable electrodes were used in all cases (Ag/AgCl ECG electrodes (Red Dot TM) 2239, Tukwila, WA, USA).
The electrodes were placed in an inverted triangle to allow for positive deflections corresponding to the P, Q, R, S, and T waves [24]. The negative electrode was placed beneath the right clavicle, the ground electrode was placed beneath the left clavicle, and the positive electrode was placed 2 centimeters beneath the sternum and over the xyphoid process. Additionally, the ECG was sampled at 2048 samples per second for high precision detection of successive heart beats [25].
## 2.2.1. Neuropsychological Tasks
The tasks performed utilized the Cambridge Neuropsychological Test Automated Battery (CANTAB) and tests included were the spatial working memory (SWM) task, attention switching task (AST), rapid visual processing task (RVP), and the spatial span (SSP) task [26]. The SWM task requires the retention and manipulation of visuospatial information. Outcome measures include errors, strategy, and latency. The AST is a test of a participant’s ability to shift attention between tasks and to ignore irrelevant information during interfering and distracting events. This test measures top-down cognitive control and provides measures of latency and errors. The RVP task is a measure of sustained attention assessing latency, probability, and sensitivity to pattern recognition. Finally, the SSP task is an assessment of working memory capacity and provides outcome measures of span length, errors, attempts, and latency.
## 2.2.2. Heart Rate Variability
Prior to statistical analysis, ECG data was pre-processed to obtain time and frequency parameters of heart rate variability (HRV). The ECG data was imported into Kubios HRV software (Version 3.1, University of Kuopio, Kuopio, Finland). The R-waves were automatically detected by applying the built-in QRS detection algorithm [27]. Frequency bands obtained were low frequency (LF) (0.04–0.15 Hz), high frequency (HF) (0.15–0.4 Hz), total power HRV (TP), and the ratio of LF to HF (LF/HF). The inbuilt process within Kubios and the smoothness priors method was used to correct for artefacts and ectopic beats in the raw ECG data [27,28]. It should also be noted that the data were log-transformed prior to analysis, where relevant.
## 2.3. Statistical Analysis
Statistical analysis was performed using SPSS Version 22.0 (IBM Corp., 2013, New York, NY, USA) IBM Corp [29] with statistical significance reported at $p \leq 0.05.$ Independent sample t-tests were applied to establish significant differences in HRV parameters and neurocognitive performance measures between the blue and white-collar workers.
## 3.1. Demographic Data of Blue and White-Collar Workers
The demographic data of the blue and white-collar groups are shown below in Table 1. Compared to the blue-collar workers, the white-collar workers had spent significantly more time in education (3.4 ± 1.2 years and 4.33 ± 1.2 years, respectively) ($p \leq 0.001$).
## 3.2. Neuropsychological Performance of Blue and White-Collar Workers
Independent sample t-tests of neuropsychological performance showed significant differences in the tasks (SWM, AST, RVP, SSP) between white ($$n = 48$$) and blue-collar ($$n = 53$$) workers. The significant findings are presenting in Table 2.
Attention Switching Task: During the AST, the white-collar sample group were found to make fewer errors when incongruent cues were given compared to the blue-collar worker group (8 ± 3.14 and 9.4 ± 3.62, respectively) ($$p \leq 0.04$$). When the “side” cue was given, the white-collar worker group made more errors than the blue-collar worker group (4.33 ± 2.56 and, 3.19 ± 2.16, respectively) ($$p \leq 0.02$$). Moreover, the white-collar worker group made significantly more correct responses than the blue-collar worker group overall (144.20 ± 8.53 and, 139.43 ± 9.80, respectively) (t = $$p \leq 0.01$$).
Rapid Visual Processing Task: Throughout the RVP task, the ability to detect signals was significantly higher in the white-collar worker group as compared to the blue-collar worker group (0.90 ± 0.08 and, 0.85 ± 0.08, respectively) ($$p \leq 0.002$$).
Spatial Span Task: Finally, the SSP task saw the white-collar worker group make more total errors than the blue-collar worker group (13.48 ± 5.65 and, 9.74 ± 6.55, respectively) ($$p \leq 0.003$$).
## 3.3. HRV in Blue and White-Collar Workers
Independent sample t-tests were used to compare HRV parameters between the white and blue-collar worker groups. The significant findings are summarised in Table 3.
Spatial Working Memory Task: Throughout the SWM task, the white-collar worker group, compared to the blue-collar worker group, had significantly higher log LF (6.33 ± 0.60 and 6.01 ± 0.48, respectively) ($$p \leq 0.004$$), log LF/HF (1.61 ± 0.83 and 1.22 ± 0.84, respectively) ($$p \leq 0.02$$), and log TP (6.7 ± 0.52 and 6.48 ± 0.37, respectively) ($$p \leq 0.02$$).
Rapid Visual Processing Task: The RVP task highlighted significantly lower HRV parameters in the white-collar worker group compared to the blue-collar worker group, particularly log LF (6.16 ± 0.47 and 6.44 ± 0.33, respectively) ($p \leq 0.001$), log HF (5.04 ± 0.60 and 5.28 ± 0.42, respectively) ($$p \leq 0.03$$), log TP (6.61 ± 0.44 and 6.89 ± 0.27, respectively) ($p \leq 0.001$), and log SDNN (3.39 ± 0.22 and 3.53 ± 0.12, respectively) ($p \leq 0.001$).
Spatial Span Task: During the Spatial Span (SSP) task, it was found that log HF was significantly lower in the white-collar worker group as compared to the blue-collar worker group (4.81 ± 0.58 and, 5.07 ± 0.67, respectively) ($$p \leq 0.03$$).
## 4. Discussion
The present study aimed to investigate the differences in HRV and psychological performance between a sample of blue and white-collar workers. The analysis indicated higher vagal cardiac mediation in blue-collar workers, as indexed by RMSSD and HF HRV, in response to spatial working memory and attention based cognitive tasks. Additionally, these results show that blue-collar workers performed significantly better on spatial tasks while white-collar workers performed better on attentional process tasks.
The current literature comparing these two sub groups is very limited; however, early work by Myrtek [29] investigating the level of stress and strain and its relationship to heart rate, physical activity, emotional strain, and mental strain found no differences in variability of heart rate (HR) between the two groups. The authors did, however, find that white-collar workers were more stressed, subjectively [29]. Additionally, it is thought that blue-collar workers are subject to an increased physical workload while white-collar workers are thought to have a high mental workload, and although interviews and questionnaires supported this idea, the physiological measurements did not [29].
Early work in the literature highlights conflicting evidence regarding the predisposition of blue and white-collar workers to CVD with some studies suggesting blue-collar workers were more at risk [30] while others suggested white-collar workers were more at risk [2]. Moreover, there is very little research investigating HRV parameters, psychological performance measures, and their associations with CVD in these two cohorts, and the present research aimed to provide more information and data regarding the relationship between different occupational and physiological risk measures and CVD.
When comparing the two sample cohorts, the only statistically significant difference in demographics was the years spent in education, where the blue-collar workers had spent less time in education than the white collar workers. Interestingly, Prihartono et al. [ 2] found that the increased level of education of white-collar workers significantly increased the prevalence of CVD. Moreover, prevalence of CVD by diagnosis was higher in the white-collar worker population, while the prevalence by symptoms was higher among the blue-collar worker group [2]. Even though the blue-collar workers are inherently more physically active in their day to day work, their socio-economic status and lifestyle choices may have a significant impact, particularly in the available access to health care. Lower education and lower salaries are more likely to predispose to unhealthy lifestyle choices [31]. Moreover, a higher BMI increased the prevalence of CVD in both blue and white-collar workers [2].
## 4.1.1. Spatial Working Memory
During the SWM task, the LF, LF/HF, and TP parameters of HRV were all greater in the white-collar worker group compared to the blue-collar worker group. LF HRV was traditionally thought to reflect sympathetic activity, as previously mentioned, but recent research indicates it is influenced by both the sympathetic and parasympathetic branches of the ANS [32]. This increase in LF HRV activity may point to increased sympathetic activity and dominance during these tasks for the white-collar worker group. This finding has been previously associated with an increased risk of CVD [33]. This has also been contrasted by other literature reporting that low LF HRV was associated with certain risk factors which predispose to CVD, for example, hypertension [34]. Moreover, a review by Hillebrand et al. [ 35] highlighted that low HRV indices, including LF HRV, indicated a higher risk of CVD in populations without any prior CVD. Interestingly, much of the prior research indicates that vagal withdrawal, and therefore an increased sympathetic response, is responsible for the cardiovascular disease risks [36]. However, Hamaad et al. [ 33] provide a differing perspective, suggesting that it is sympathetic activation which may be associated to cardiac events, and not the former. The authors of [33] investigated the associations between indices of HRV (time and frequency) and inflammatory biomarkers in patients with acute coronary syndrome ($$n = 100$$, male = 77, average age = 63 ± 12 years) and healthy controls ($$n = 49$$, male = 32, average age = 60 ± 10 years). Though the correlations were modest, the authors reported an inverse relationship between LF HRV and inflammatory biomarkers and, therefore, implicate sympathetic tone in CVD [33]. This idea is further supported by several studies which further investigate the inflammatory biomarkers and associated HRV changes [37].
## 4.1.2. Rapid Visual Processing
The RVP task showed that the blue-collar workers had higher HRV parameters across the board, particularly LF, HF, TP and SDNN. This is an interesting finding as the RVP task is one of sustained attention, and it was therefore expected that the white-collar worker group would exhibit higher levels of cardiac vagal control, as indexed by HF HRV or RMSSD. The findings of the present research may reflect high levels of stress within the white-collar working population as shown by lower HF HRV. Previous research concluding which occupational group is more stressed is contentious, and the literature suggests a multitude of variables that may contribute. Dedele et al. [ 38] indicate that blue-collar workers are 1.5 times more likely to perceive higher levels of stress in general. However, the white-collar workers had a four times increased likelihood of perceiving greater stress when they had been sedentary for more than 3 h per day [38]. Contrastingly, Nydegger [39] found no significant differences in stress levels between blue and white-collar workers, nor any differences between genders. Given that these studies only assessed perceived stress by way of surveys, the results may be too subjective, with numerous factors potentially influencing the responses. The use of a more objective measure would have been of great benefit to support their findings. Notwithstanding, they do provide grounds to indicate intricate interrelationships between workplace stress, HRV, and CVD. Moreover, recommendations made to white-collar workers include making improvements in sedentary lifestyle and increasing physical activity during work hours, while blue-collar workers must avoid unhealthy lifestyle habits [39,40]. These practices will ultimately reduce stress, improve cardiac autonomic activity and parasympathetic input, and therefore may reduce the risk of a cardiovascular event.
## 4.1.3. Spatial Span
The final difference in HRV between the blue and white-collar worker group found in this study was related to the SSP task, whereby the blue-collar worker group showed higher vagal mediation than the white-collar worker group. This is indicative of better control and better performance. Moreover, it may indicate a more relaxed scenario, as the SSP task is designed to evaluate working memory capacity in the 3D space around them, an environment familiar to blue-collar workers.
## 4.2. Comparison of Neuropsychological Performance between Blue and White-Collar Workers
Occupation has been considered as an important predictor of cognitive ability and decline over time [41]. Furthermore, the executive function requirements in the workplace, as well as the complexities of the environment, seem to have a correlation to cognitive decline [42]. Prior research has tended to be focused on age-related decline in cognitive processing and few studies have focused on the occupational effects. However, given that people spend a substantial portion of life at work, the workplace environment may have a significant effect [43].
## 4.2.1. Attention Switching
The AST showed that the white-collar workers made less errors when the cues were changing and more errors when the “side” cue was given. However, in the task as a whole, the white-collar workers gave significantly more correct responses than the blue-collar workers. In a longitudinal study spanning 10 years, Kim et al. [ 44] assessed executive function in blue ($$n = 1216$$, $61\%$ Female, aged 70.7 ± 4.64 years) and white-collar workers ($$n = 242$$, $22\%$ Female, aged 69.98 ± 4.18 years). The authors gathered data using the Mini-Mental Sate Examination (MMSE) [45] and other potential covariates, including sociodemographic factors, health related factors and occupational factors [44]. Primary findings between the longest-held lifetime occupation and executive function decline showed that males had no significant risks, whilst females showed a 2.5-fold increased risk of cognitive impairment amongst blue-collar workers compared to white-collar workers [44].
## 4.2.2. Rapid Visual Processing/Spatial Span
The white-collar workers showed significantly better performance during the RVP task, where their ability to detect sequences was much better. However, the white-collar workers made more errors during the SSP task. The relationship between mental workload and cardiovascular parameters is further illustrated by Capuana et al. [ 46]. These authors assessed 22 young adults (17 women, 18–27 years, average age = 20.5 years (SD not specified)) and 18 older adults (11 women, 65–83 years, average age = 72.3 (SD not specified)) and indicated relationships between cardiac measures and performance, as well as an association between increased cardiac workload and more errors in the older adults but not the younger adults [46]. This further supports and adds to the age-related literature regarding neurocognitive performance with the added element of cardiac risk measures. The results of previous literature and the present findings suggest that the effects of occupation on executive functions are multifaceted [41]. Prior research has indicated that white-collar workers are more cognitively inclined in the later years [41]. Moreover, manual labor workers (including machine operators, assembly workers and plant operators) have been shown to have a significantly higher chance of reduced executive function as compared to non-manual laborers (including business executives, administrators, and managers) [47]. As a whole, the white-collar workers seem to have performed better on the executive function tasks. Notwithstanding the varying performance on different tasks, an in depth analysis must be conducted to supplement broader examinations in order to identify specific relationships between cardiac variables and neurocognitive performance measures.
Several factors may be considered when assessing the performance and risks between the blue and white-collar worker populations. Most people spend a large portion of their life at work, and so the inherent risks related to employment are something that must be further researched. These risks may be a result of the complexity in given occupations, which was first touched upon by Schooler [48] and further by Schooler et al. [ 49]. These authors suggested that complex environments at work, or during leisure time, allow for continued reinforcement of executive function. This greater intellectual stimulation increases neural growth and synaptic density, which protects against cognitive decline [50]. Therefore, lower intellectual demands for blue-collar workers may predispose them to executive function impairments. This is just one facet by which the literature suggests the enhanced ability of white-collar workers. Another theory indicates that, since blue-collar work is associated with a lower income, this translates to poor housing, nutrition, environment, and poor lifestyle habits and practices, which may be linked to cognitive decline [51,52]. Interestingly, white-collar workers are more educated in the traditional sense, but this does not necessarily reflect in overall intelligence. Given that white-collar workers are known to use cognitive abilities more often than blue-collar workers, it could be assumed that they have superior cognitive abilities. This may not be the case however, as a study showed that there was no evidence that regular use of computerized brain trainers improves general cognitive functioning [53].
## 4.3. Limitations and Future Directions
The present findings show perhaps that changes in HRV are in fact influenced by various tasks, spanning all professions. Increased sample numbers in each profession would allow for stratification and observations within the same job type. For example, one white-collar worker may perform more administrative tasks while another may perform more data analytics and these differences in neuropsychological load may further influence HRV. Moreover, this cross-sectional design provides a snapshot in time of the measures. Therefore, a longitudinal study would allow for a more in-depth analysis of how a particular profession may influence these physiological variables over the course of one’s life. It is also acknowledged that, even though only $18\%$ of the blue-collar worker group was made up of female workers, this is an accurate reflection of this population sample [54]. Though the present study identified numerous findings, it may only be predictive in nature and not causal. Therefore, future studies may be able to investigate the causal link between vagal tone, working memory, and attention through various techniques, such as transcutaneous vagal nerve stimulation or other neuroimaging techniques.
## 5. Conclusions
Overall, the present research identified multiple significant differences in HRV parameters and neurocognitive performance measures between the blue and the white-collar workforce. Blue-collar workers indicated higher vagally mediated cardiac control during neuropsychological tasks with better performance in spatial working memory exercises, whilst white-collar workers had superior performance on attention-based tasks.
Notably, reduced parasympathetic modulation of the heart, particularly in white-collar workers, was observed.
## References
1. Koopman C., Pelletier K.R., Murray J.F., Sharda C.E., Berger M.L., Turpin R.S., Hackleman P., Gibson P., Holmes D.M., Bendel T.. **Stanford presenteeism scale: Health status and employee productivity**. *J. Occup. Environ. Med.* (2002) **44** 14-20. DOI: 10.1097/00043764-200201000-00004
2. Prihartono N.A., Fitriyani F., Riyadina W.. **Cardiovascular disease risk factors among blue and white-collar workers in Indonesia**. *Acta Med. Indones.* (2018) **50** 96-103. PMID: 29950527
3. Graham I., Atar D., Borch-Johnsen K., Boysen G., Burell G., Cifkova R., Dallongeville J., De Backer G., Ebrahim S., Gjelsvik B.. **European guidelines on cardiovascular disease prevention in clinical practice: Executive summary: Fourth Joint Task Force of the European Society of Cardiology and Other Societies on Cardiovascular Disease Prevention in Clinical Practice (Constituted by representatives of nine societies and by invited experts)**. *Eur. Heart J.* (2007) **28** 2375-2414. DOI: 10.1093/eurheartj/ehm316
4. Diamond A.. **Executive functions**. *Annu. Rev. Psychol.* (2013) **64** 135-168. DOI: 10.1146/annurev-psych-113011-143750
5. Will Crescioni A., Ehrlinger J., Alquist J.L., Conlon K.E., Baumeister R.F., Schatschneider C., Dutton G.R.. **High trait self-control predicts positive health behaviors and success in weight loss**. *J. Health Psychol.* (2011) **16** 750-759. DOI: 10.1177/1359105310390247
6. Bailey C.E.. **Cognitive accuracy and intelligent executive function in the brain and in business**. *Ann. N. Y. Acad. Sci.* (2007) **1118** 122-141. DOI: 10.1196/annals.1412.011
7. Broidy L.M., Nagin D.S., Tremblay R.E., Bates J.E., Brame B., Dodge K.A., Fergusson D., Horwood J.L., Loeber R., Laird R.. **Developmental trajectories of childhood disruptive behaviors and adolescent delinquency: A six-site, cross-national study**. *Dev. Psychol.* (2003) **39** 222-245. DOI: 10.1037/0012-1649.39.2.222
8. Roman D.D., Kubo S.H., Ormaza S., Francis G.S., Bank A.J., Shumway S.J.. **Memory improvement following cardiac transplantation**. *J. Clin. Exp. Neuropsychol.* (1997) **19** 692-697. DOI: 10.1080/01688639708403754
9. Pumprla J., Howorka K., Groves D., Chester M., Nolan J.. **Functional assessment of heart rate variability: Physiological basis and practical applications**. *Int. J. Cardiol.* (2002) **84** 1-14. DOI: 10.1016/S0167-5273(02)00057-8
10. Olshansky B., Sabbah H.N., Hauptman P.J., Colucci W.S.. **Parasympathetic nervous system and heart failure: Pathophysiology and potential implications for therapy**. *Circulation* (2008) **118** 863-871. DOI: 10.1161/CIRCULATIONAHA.107.760405
11. Triposkiadis F., Karayannis G., Giamouzis G., Skoularigis J., Louridas G., Butler J.. **The sympathetic nervous system in heart failure physiology, pathophysiology, and clinical implications**. *J. Am. Coll. Cardiol.* (2009) **54** 1747-1762. DOI: 10.1016/j.jacc.2009.05.015
12. Hansen A.L., Johnsen B.H., Thayer J.F.. **Vagal influence on working memory and attention**. *Int. J. Psychophysiol.* (2003) **48** 263-274. DOI: 10.1016/S0167-8760(03)00073-4
13. Duschek S., Muckenthaler M., Werner N., del Paso G.A.. **Relationships between features of autonomic cardiovascular control and cognitive performance**. *Biol. Psychol.* (2009) **81** 110-117. DOI: 10.1016/j.biopsycho.2009.03.003
14. Gianaros P.J., Van Der Veen F.M., Jennings J.R.. **Regional cerebral blood flow correlates with heart period and high-frequency heart period variability during working-memory tasks: Implications for the cortical and subcortical regulation of cardiac autonomic activity**. *Psychophysiology* (2004) **41** 521-530. DOI: 10.1111/1469-8986.2004.00179.x
15. Johnsen B.H., Thayer J.F., Laberg J.C., Wormnes B., Raadal M., Skaret E., Kvale G., Berg E.. **Attentional and physiological characteristics of patients with dental anxiety**. *J. Anxiety Disord.* (2003) **17** 75-87. DOI: 10.1016/S0887-6185(02)00178-0
16. Stroop J.R.. **Studies of interference in serial verbal reactions**. *J. Exp. Psychol.* (1935) **18** 643-661. DOI: 10.1037/h0054651
17. Muthukrishnan S.-P., Gurja J.-P., Ratna S.. *Is Heart Rate Variability Related to Cognitive Performance in Visuospatial Working Memory?* (2015)
18. Haley A.P., Sweet L.H., Gunstad J., Forman D.E., Poppas A., Paul R.H., Tate D.F., Cohen R.A.. **Verbal working memory and atherosclerosis in patients with cardiovascular disease: An fMRI study**. *J. Neuroimaging* (2007) **17** 227-233. DOI: 10.1111/j.1552-6569.2007.00110.x
19. Rostamian S., van Buchem M.A., Westendorp R.G., Jukema J.W., Mooijaart S.P., Sabayan B., de Craen A.J.. **Executive function, but not memory, associates with incident coronary heart disease and stroke**. *Neurology* (2015) **85** 783-789. DOI: 10.1212/WNL.0000000000001895
20. 20.
National Heart Foundation of Australia (NHFA)
Guide to Management of Hypertension 2008-Assessing and Managing Raised Blood Pressure in Adults Updated 2010National Heart FoundationMelbourne, Australia2010. *Guide to Management of Hypertension 2008-Assessing and Managing Raised Blood Pressure in Adults Updated 2010* (2010)
21. Roeser K., Obergfell F., Meule A., Vogele C., Schlarb A.A., Kubler A.. **Of larks and hearts: Morningness/eveningness, heart rate variability and cardiovascular stress response at different times of day**. *Physiol. Behav.* (2012) **106** 151-157. DOI: 10.1016/j.physbeh.2012.01.023
22. Pickering T.G., Hall J.E., Appel L.J., Falkner B.E., Graves J., Hill M.N., Jones D.W., Kurtz T., Sheps S.G., Roccella E.J.. **Recommendations for blood pressure measurement in humans and experimental animals: Part 1: Blood pressure measurement in humans: A statement for professionals from the Subcommittee of Professional and Public Education of the American Heart Association Council on High Blood Pressure Research**. *Hypertension* (2005) **45** 142-161. DOI: 10.1161/01.HYP.0000150859.47929.8e
23. Goldberg D.P.. *The Detection of Psychiatric Illness by Questionnaire* (1972) **Volume 21**
24. Combatalade D.C.. *Basics of Heart Rate Variability Applied to Psychophysiology* (2010)
25. Berntson G.G., Bigger J.T., Eckberg D.L., Grossman P., Kaufmann P.G., Malik M., Nagaraja H.N., Porges S.W., Saul J.P., Stone P.H.. **Heart rate variability: Origins, methods, and interpretive caveats**. *Psychophysiology* (1997) **34** 623-648. DOI: 10.1111/j.1469-8986.1997.tb02140.x
26. 26.
Cambridge Cognition
CANTAB (Cognitive Assessment Software)Cambridge CognitionCambridge, UK2021. *CANTAB (Cognitive Assessment Software)* (2021)
27. Tarvainen M.P., Niskanen J.-P., Lipponen J.A., Ranta-aho P.O., Karjalainen P.A.. **Kubios HRV—Heart rate variability analysis software**. *Comput. Methods Programs Biomed.* (2014) **113** 210-220. DOI: 10.1016/j.cmpb.2013.07.024
28. Tarvainen M.P., Ranta-aho P.O., Karjalainen P.A.. **An advanced detrending method with application to HRV analysis**. *IEEE Trans. Biomed. Eng.* (2002) **49** 172-175. DOI: 10.1109/10.979357
29. Myrtek M., Fichtler A., Strittmatter M., Brügner G.. **Stress and strain of blue and white collar workers during work and leisure time: Results of psychophysiological and behavioral monitoring**. *Appl. Ergon.* (1999) **30** 341-351. DOI: 10.1016/S0003-6870(98)00031-3
30. Voelker R.. **Highest heart disease and stroke rates in service, blue collar jobs**. *J. Am. Med. Assoc.* (2014) **312** 1090. DOI: 10.1001/jama.2014.11164
31. Clougherty J.E., Souza K., Cullen M.R.. **Work and its role in shaping the social gradient in health**. *Ann. N. Y. Acad. Sci.* (2010) **1186** 102-124. DOI: 10.1111/j.1749-6632.2009.05338.x
32. Berntson G.G., Cacioppo J.T., Grossman P.. **Whither vagal tone**. *Biol. Psychol.* (2007) **74** 295-300. DOI: 10.1016/j.biopsycho.2006.08.006
33. Hamaad A., Sosin M., Blann A.D., Patel J., Lip G.Y.H., MacFadyen R.J.. **Markers of inflammation in acute coronary syndromes: Association with increased heart rate and reductions in heart rate variability**. *Clin. Cardiol.* (2005) **28** 570-576. DOI: 10.1002/clc.4960281207
34. Thayer J.F., Yamamoto S.S., Brosschot J.F.. **The relationship of autonomic imbalance, heart rate variability and cardiovascular disease risk factors**. *Int. J. Cardiol.* (2010) **141** 122-131. DOI: 10.1016/j.ijcard.2009.09.543
35. Hillebrand S., Gast K., de Mutsert R., Swenne C.A., Jukema J., Middeldorp S., Rosendaal F., Dekkers O.. **Heart rate variability and first cardiovascular event in populations without known cardiovascular disease: Meta-analysis and dose–response meta-regression**. *Europace* (2013) **15** 742-749. DOI: 10.1093/europace/eus341
36. Liu L., Zhao M., Yu X., Zang W.. **Pharmacological modulation of vagal nerve activity in cardiovascular diseases**. *Neurosci. Bull.* (2019) **35** 156-166. DOI: 10.1007/s12264-018-0286-7
37. Psychari S.N., Apostolou T.S., Iliodromitis E.K., Kourakos P., Liakos G., Kremastinos D.T.. **Inverse relation of C-reactive protein levels to heart rate variability in patients after acute myocardial infarction**. *Hell. J. Cardiol.* (2007) **48** 64-71
38. Dėdelė A., Miškinytė A., Andrušaitytė S., Bartkutė Ž.. **Perceived stress among different occupational groups and the interaction with sedentary behaviour**. *Int. J. Environ. Res. Public Health* (2019) **16**. DOI: 10.3390/ijerph16234595
39. Nydegger R.. **Occupational stress and job satisfaction in white-and blue-collar workers**. *Int. Bus. Econ. Res. J.* (2002) **1**. DOI: 10.19030/iber.v1i12.4006
40. Clays E., De Bacquer D., Crasset V., Kittel F., de Smet P., Kornitzer M., Karasek R., De Backer G.. **The perception of work stressors is related to reduced parasympathetic activity**. *Int. Arch. Occup. Environ. Health* (2011) **84** 185-191. DOI: 10.1007/s00420-010-0537-z
41. Nguyen C.T., Couture M.C., Alvarado B.E., Zunzunegui M.V.. **Life course socioeconomic disadvantage and cognitive function among the elderly population of seven capitals in Latin America and the Caribbean**. *J. Aging Health* (2008) **20** 347-362. DOI: 10.1177/0898264308315430
42. Finkel D., Andel R., Gatz M., Pedersen N.L.. **The role of occupational complexity in trajectories of cognitive aging before and after retirement**. *Psychol. Aging* (2009) **24** 563-573. DOI: 10.1037/a0015511
43. Then F.S., Luck T., Luppa M., Thinschmidt M., Deckert S., Nieuwenhuijsen K., Seidler A., Riedel-Heller S.G.. **Systematic review of the effect of the psychosocial working environment on cognition and dementia**. *Occup. Environ. Med.* (2014) **71** 358-365. DOI: 10.1136/oemed-2013-101760
44. Kim H.-J., Min J.-Y., Min K.-B.. **The association between longest-held lifetime occupation and late-life cognitive impairment: Korean longitudinal study of aging (2006-2016)**. *Int. J. Environ. Res. Public Health* (2020) **17**. DOI: 10.3390/ijerph17176270
45. Folstein M.F., Folstein S.E., McHugh P.R.. **Mini-mental state. A practical method for grading the cognitive state of patients for the clinician**. *J. Psychiatr. Res.* (1975) **12** 189-198. DOI: 10.1016/0022-3956(75)90026-6
46. Capuana L.J., Dywan J., Tays W.J., Segalowitz S.J.. **Cardiac workload and inhibitory control in younger and older adults**. *Biol. Psychol.* (2012) **90** 60-70. DOI: 10.1016/j.biopsycho.2012.02.018
47. Li C.Y., Wu S.C., Sung F.C.. **Lifetime principal occupation and risk of cognitive impairment among the elderly**. *Ind. Health* (2002) **40** 7-13. DOI: 10.2486/indhealth.40.7
48. Schooler C.. **Psychological effects of complex environments during the life span: A review and theory**. *Intelligence* (1984) **8** 259-281. DOI: 10.1016/0160-2896(84)90011-4
49. Schooler C., Mulatu M.S., Oates G.. **Occupational self-direction, intellectual functioning, and self-directed orientation in older workers: Findings and Iimplications for individuals and societies**. *Am. J. Sociol.* (2004) **110** 161-197. DOI: 10.1086/385430
50. Alvarado B.E., Zunzunegui M.V., Del Ser T., Béland F.. **Cognitive decline is related to education and occupation in a Spanish elderly cohort**. *Aging Clin. Exp. Res.* (2002) **14** 132-142. DOI: 10.1007/BF03324426
51. Zhang M., Gale S.D., Erickson L.D., Brown B.L., Woody P., Hedges D.W.. **Cognitive function in older adults according to current socioeconomic status**. *Neuropsychol. Dev. Cogn. Sect. B Aging Neuropsychol. Cogn.* (2015) **22** 534-543. DOI: 10.1080/13825585.2014.997663
52. Lee Y., Back J.H., Kim J., Byeon H.. **Multiple socioeconomic risks and cognitive impairment in older adults**. *Dement. Geriatr. Cogn. Disord.* (2010) **29** 523-529. DOI: 10.1159/000315507
53. Owen A.M., Hampshire A., Grahn J.A., Stenton R., Dajani S., Burns A.S., Howard R.J., Ballard C.G.. **Putting brain training to the test**. *Nature* (2010) **465** 775-778. DOI: 10.1038/nature09042
54. 54.
Workplace Gender Equality Agency (WGEA)
Gender Segregation in Australia’s WorkforceWGEASydney, Australia2019. *Gender Segregation in Australia’s Workforce* (2019)
|
---
title: 'Adapting and Developing A Diabetes Prevention Intervention Programme for South
Africa: Curriculum and Tools'
authors:
- Jillian Hill
- Mieke Faber
- Nasheeta Peer
- Cindy George
- Brian Oldenburg
- Andre P. Kengne
journal: International Journal of Environmental Research and Public Health
year: 2023
pmcid: PMC10002357
doi: 10.3390/ijerph20054463
license: CC BY 4.0
---
# Adapting and Developing A Diabetes Prevention Intervention Programme for South Africa: Curriculum and Tools
## Abstract
The South African Diabetes Prevention Programme (SA-DPP) is a lifestyle intervention targeting individuals at high risk of developing type 2 diabetes mellitus (T2DM). In this paper we describe the mixed-method staged approach that was used to develop and refine the SA-DPP intervention curriculum and the appropriate tools for local resource-poor communities. During the preparation phase, existing evidence on similar DPP interventions was reviewed, focus group discussions with individuals from the target population were conducted as part of a needs assessment, and experts were consulted. The curriculum booklet, a participant workbook and facilitator workbook were developed, and the content was evaluated by experts in the field. The design and layout of the booklet and workbooks needed to be culturally and contextually appropriate. The printed material was evaluated for readability and acceptability by participants of the target population; based on their feedback, the design and layout were refined and the printed material was translated. The suitability of the intervention was tested in a pilot study; based on feedback from the participants and facilitator, the curriculum was revised where needed and finalised. Through this process a context specific intervention and printed materials were developed. A complete evaluation of this culturally relevant model for T2DM prevention in South *Africa is* pending.
## 1. Introduction
It has been over two decades since the Finnish Diabetes Prevention Study (DPS) [1,2] and the Diabetes Prevention Program (DPP) in the United States [3] presented robust evidence on successful lifestyle intervention trials that reduced incident diabetes rates by $58\%$. Since then, several “real-world” (implemented in real-world conditions) diabetes prevention implementation trials and translational studies have been conducted, predominantly in high-income countries [4], including minority groups [5]. Only one (published) adaptation of a lifestyle intervention trial for diabetes prevention has been recorded in a low-income country, namely, India [6,7]. The original DPP included a high percentage of racial and ethnic minorities [3] and later adaptions thereof, for example, the YMCA model, included people of various ethnicities and socio-demographics [8]. However, a huge gap exists specifically in Africa.
The burden of type 2 diabetes (T2DM) in sub-Saharan Africa (SSA) is considerable and continues to rise rapidly. The number of people with diabetes in SSA is expected to increase from 24 million in 2021 to 55 million by 2045 [9]. In SSA, South Africa (SA) has the second largest number of people with diabetes. By 2017, there were 1,826,100 SA adults with diabetes [10], with the highest burden in socioeconomically disadvantaged populations in poorly resourced areas. A pooled analysis conducted as part of a systematic review and meta-analysis on the prevalence of diabetes in SA indicated that $11.3\%$ (6.97–16.52 %) of the black population and $23.7\%$ (13.93–15.73) of the mixed ancestry population have diabetes [11]. Global initiatives continually advocate for improved strategies in community-based approaches for T2DM prevention. Yet, little is known about such strategies in the African region.
The overall aim of the South African Diabetes Prevention Programme (SA-DPP) is to develop and evaluate a culturally relevant T2DM prevention approach for SA, based on the findings from successful diabetes prevention effectiveness and implementation programmes elsewhere [12,13,14,15]. The intention is to develop an exemplar that will inform lifestyle interventions aimed at preventing T2DM and other lifestyle-related conditions at the primary health care level in SA. This prototype may also serve as a model that can be adapted for other SSA countries facing similar challenges [16]. The SA-DPP lifestyle intervention proposes six group sessions based on empowerment ideology, emphasising the participant’s ability to make informed decisions, and his/her role as an independent decision maker who takes responsibility and regulates his/her own actions [17,18]. The group sessions are to be led by non-professional or community health workers (CHWs) (assisted by nurses and/or dieticians) and will be complemented by structured mobile phone messaging to augment adherence and retention. The lifestyle change objectives (diet and physical activity) of the SA-DPP are based on the original Finnish Diabetes Prevention Study, i.e., [1] < $30\%$ of total energy intake from fat; [2] < $10\%$ of total energy intake from saturated fat; [3] > 15 g of fibre/1000 kcal; [4] > 4 h/week moderate level of physical activity; and [5] > $5\%$ reduction in body mass index [2]. In this paper we describe the process followed in developing an evidence-based and context specific lifestyle intervention curriculum and tools for the SA-DPP suited to lower socioeconomic communities, for adults from 25–65 years old at high risk of developing diabetes.
## 2. Methods and Process
A qualitative mixed-method staged approach was followed in developing the SA-DPP intervention curriculum and tools (facilitator and participant workbooks). It included 11 steps, as follows: [1] reviewing and learning from existing evidence; [2] a needs assessment; [3] expert input; [4] development of the curriculum and tools; [5] expert content evaluation; [6] design and layout of the curriculum and tools; [7] participant readability and acceptability evaluation; [8] refining the design and layout of the curriculum and tools; [9] translating the curriculum and participant workbook; [10] suitability evaluation (pilot intervention); and [11] finalising the curriculum (see Figure 1). The research was approved by the Ethics Committee of the South African Medical Research Council (SAMRC) (approval no. EC018-$\frac{7}{2015}$).
## 2.1. Step 1: Reviewing and Learning from the Existing Evidence
Diabetes prevention implementation studies have gained prominence only over the past two decades. Implementation studies have been conducted primarily in Europe, North America, Australia, and more recently, India. These have been summarised from various perspectives in several systematic reviews over the last few years [19,20,21]. A consistent conclusion from these reviews is that there is sufficient evidence to support the reproducibility of the targeted lifestyle objectives and the efficacy of diabetes prevention trials in real-world settings, using less intensive interventions. However, the adaptation of DPPs to new settings has resulted in new programmes with different components, modes of delivery, duration, interventionists, target population, and outcomes [19,20]. This confirms that programmes must undertake the necessary processes for effective cultural adaptation in new settings. The review by Rawal et al., published in 2012, focused exclusively on developing countries and found only three efficacy or implementation trials on diabetes prevention, all conducted in India or China, with none in Africa [21]. Using the same search strategy as the latter review [21], we searched the literature published after 2012 to 2015 (in preparation for protocol submission) using PubMed/MEDLINE, and simplified and adapted the search terms to make them more specific to Africa by combining the following ‘diabetes’ AND ‘prevention’ AND [Africa OR “names for each African country”]; but found no ongoing or completed diabetes prevention study on the African continent. Subsequently, in preparation for intervention development we conducted a systematic review of the roles, responsibilities, and characteristics of the lay community health workers involved in diabetes prevention programmes [20], which rendered no papers from Africa. Evidence on the best strategies to implement diabetes prevention in *Africa is* therefore lacking.
The literature on adapted DPP interventions, similarly, is restricted to high-income countries and India [4,6,7,22]. Initial adaptions of the US DPP intervention included: [1] transforming the core curriculum from an intensive, individualised model to a group-based format; [2] removing costly toolbox incentives; [3] applying a formal exercise partner system (e.g., gym or sports club); and [4] delivering exercise components using local fitness club staff trained in behaviour change counselling rather than specialised life coaches [22]. The DPP lifestyle intervention was further adapted for the Group-Organized YMCA DPP (GO-YDPP), by merging the existing experiences and theory-driven adaptations of the intensive DPP lifestyle intervention to improve sustainability. The programme kept the physical activity and weight loss goals of the original DPP lifestyle intervention, but tailored these to meet individual needs with activities that were flexible, culturally sensitive, and acceptable to local communities. Emphasis was placed on self-esteem, empowerment, and social support. These principles were applied across the three key phases of their intervention, which comprised a core curriculum phase (16 lessons), a training and refinement phase (four weeks), and an ongoing maintenance phase [22].The Kerala Diabetes Prevention Program (K-DPP) was comprised of four key components as follows: [1] a group-based peer-support programme; [2] peer-leader training for lay people to lead groups; [3] resource materials (including a curriculum booklet and a participant workbook); and [4] strategies to inspire broader community engagement. The K-DPP adaptation employed a systematic approach, which was grounded in evidence-based behaviour change techniques [6].
## 2.2. Step 2: Needs Assessment
The needs assessment consisted of two rounds of focus group discussions (FGDs). Participants in the FGDs were recruited from amongst those who were part of the SA-DPP baseline screening pilot study that was conducted between August 2017 and March 2018 in eight resource-poor areas in Cape Town, Western Cape, South Africa [16]. The first four areas for which at least 20 individuals at risk of diabetes were identified were included in the needs assessment. Convenient sampling was employed whereby the first eight high-risk participants available per area were invited to participate in the FGDs. Two-to-three FGDs were conducted per area depending on the total number of participants. Two series of FGDs (17 in total) were conducted with 68 participants. Participants were mostly female ($$n = 50$$), between the ages of 45 and 65 years old ($$n = 58$$), and predominantly Xhosa speaking ($$n = 42$$) rather than English/Afrikaans speaking ($$n = 26$$) participants.
The aim of the first round of FGDs ($$n = 10$$) was to gauge participants’ knowledge and perceptions of diabetes and prevention, and to obtain inputs on the preferred intervention content, format, and delivery. The second round of FGDs ($$n = 7$$) focused specifically on lifestyle behaviours, i.e., diet and physical activity, and their related barriers and enablers. A focus group schedule was developed and used, which was informed using a combination of literature and consultation with SA-DPP project team members.
The FGDs were facilitated by a researcher (first author, trained in qualitative methodologies), while a second person (scribe) took notes. The participants were mainly English, Afrikaans, and Xhosa speakers. The FGDs were conducted predominantly in English as this was the language most suited to multiple language groups. The facilitator was fluent in both English and Afrikaans, and a Xhosa speaking scribe was present in the groups with Xhosa speaking participants. Participants were encouraged to respond in the language that they were most comfortable in. Thematic data analysis was employed to code and analyse the data, and have been described previously [23].
## 2.2.1. Diabetes Knowledge and Perceptions
The themes and relevant quotes on participants’ knowledge and perceptions on diabetes are summarised in Table 1. From the FGDs it was clear that the participants understood that T2DM is a serious disease. Knowledge around the causes of T2DM was however limited, and participants mixed misperceptions with factual information. All participants knew someone with T2DM, who was often a family member, a friend, or a neighbour. Several participants were aware of T2DM complications and knew people with these complications (e.g., gangrene, leg amputation, slow healing of sores). In terms of diabetes prevention, participants were aware that lifestyle plays a role in diabetes occurrence. While they were not all certain that diabetes is preventable, they were keen to know more. A few felt that regular medical check-ups are key to preventing diabetes.
Participants had a fair idea about what healthy eating entails, including healthy food preparation. They knew that exercise is important and alluded to the importance of living a balanced lifestyle. When asked how confident they were in their ability to live a healthy lifestyle, participants in three of the 10 FGDs provided no comments, while the responses in the other FGDs were mixed. Most participants felt that household circumstances, i.e., family, and limited finances, would make it difficult to eat healthily and live a healthy lifestyle, while some felt confident that they could because it is for their own well-being. Factors that enabled self-efficacy were a positive mindset, cooking for themselves, and the ability to purchase healthy foods on a small budget. Participants identified various barriers to living a healthy lifestyle. Limited finances were the biggest challenge; this included having to feed a (large) family on a small budget. The availability of water and space for cooking was a challenge for individuals living in squatter camps (informal settlements). A lack of knowledge and family support, and laziness were also highlighted as barriers. Most participants had safety concerns when it came to physical activity outside the home. In the South African context, this is related to the high crime rates in the country, with outdoor activities generally being avoided and discouraged unless conducted in secure environments.
## 2.2.2. Lifestyle Behaviours/Practices
The themes and relevant quotes on participants’ challenges, barriers, and enablers for healthy lifestyle behaviours are summarised in Table 2. Although some participants were attempting to live a healthy lifestyle, it seemed challenging for everyone. Again, the biggest barrier to eating healthy was financial constraints. The distance of the supermarkets from their home and knowledge on healthy eating were also mentioned. According to participants, factors that would enable them to eat healthier include greater financial spending power, their family’s buy-in and support, as well as bigger plots to plant their own vegetable gardens. Participants started to recognise the flaws in their perceptions on food and healthy eating and acknowledged that receiving the correct information and support would enable change. It was further evident that proper support from the community (for community vegetable gardens), the clinic (for useable information) and the ward councillor (for quality food parcels) is needed. The few participants who lived alone felt that eating healthy would be easy for them.
## 2.2.3. Intervention Format
The themes and relevant quotes for participants’ perceptions on the proposed intervention format are summarised in Table 3. Participants were satisfied that group sessions held in the community would be the appropriate intervention format. There were some mixed responses regarding the facilitation of the group sessions; this did not necessarily have to be a CHW or peer leader. Some felt that it could be anyone, provided they are well trained and knowledgeable on the subject matter. Most participants were fine with English as the language of the intervention. However, for the intervention materials, while some preferred reading in English, others felt that these should be translated into their home language. Xhosa speaking participants felt more strongly about materials being in their home language than Afrikaans speaking participants.
During the second half of the first round of FGDs, participants were presented with an overall idea of what the intervention would entail, i.e., it would cover topics such as healthy eating and physical activity; it would be delivered via a group session; and that it was envisaged that sessions would be facilitated by a lay community health worker. The feedback received from participants, the themes and relevant quotes are summarised in Table 4. Participants requested that basic information on healthy eating should be included in the intervention curriculum. This should include topics such as a description of healthy foods and healthy eating guidelines, e.g., portions sizes, healthier meat options, etc. When asked whether they would have their family’s support to follow healthy eating guidelines, the responses were mixed with some families being willing to comply, while others refused to eat healthier food options. The latter makes it difficult for those who want to eat healthier, as financial constraints prevent the preparation of different meals for different household members. Although participants had a positive attitude towards physical activity and were familiar with the benefits thereof, their motivation to engage in such activities was lacking, with laziness being the key barrier. Other barriers included the community culture with exercise deemed unimportant, safety concerns, and a lack of facilities for physical activity. Participants felt that increased confidence would come with observing the benefits of exercise and learning about different exercises for specific body parts and injuries (e.g., knees and backache).
## 2.3. Step 3: Expert Input
The SA-DPP team were able to draw on the lessons learnt from the K-DPP, with access to the curriculum provided by a co-investigator who had collaborated on various DPP interventions, including the K-DPP. The K-DPP curriculum, developed for Kerala, a poor, small, densely populated state in south India, was strongly theory driven in its approach, drawing from behaviour change techniques [6]. A local co-investigator modified the curriculum, which was circulated to the SA-DPP multi-disciplinary team who were all non-communicable disease research scientists. The team included three dieticians, a physician, and a counsellor (psychologist). After the consultations, the original themes/topics covered in previous DPP curriculums (i.e., diabetes knowledge, healthy eating, physical activity, smoking, alcohol, and stress) and the behaviour change theories/techniques were included. However, these were adapted to be more relevant for the South African population, using the South African food-based dietary guidelines (SAFBDG) [24] as the basis.
The core project team concluded that the SA-DPP curriculum should be framed around the processes/stages of change as encapsulated by the transtheoretical model [25]. With the aim of moving the participants through the stages of contemplation, preparation, action, and finally the maintenance phase (with the expectation that once participants have been identified as at risk of developing diabetes at screening, they would be expected to be in the precontemplation/contemplation stage once they enter the DPP) [25]. Also, it was appreciated that in lifestyle programmes participants would likely never reach the termination stage, which indicates no temptation and $100\%$ efficacy [25].
## 2.4. Step 4: Development of the Curriculum and Tools
A private practising dietitian was contracted to develop the SA-DPP curriculum and participant workbook, with guidance from one of the co-investigators, while the project manager developed the facilitator workbook.
Curriculum Development: Using the feedback from the FGDs, the topics were developed to provide basic knowledge together with practical tips to make lifestyle changes realistic and achievable. A section on setting SMART (specific, measurable, attainable, relevant, and timebound) goals was included. The curriculum content was kept as simple and relevant as possible for the target population considering their socioeconomic status (low income), educational level (matric or less), and environmental factors (safety and access).
Participant Workbook: The participant workbook was thereafter developed to encourage action-driven short- and long-term (realistic) goal attainment. The tasks in the workbook were goal oriented and subject specific (nutrition, physical activity, etc.), to enable participants to complete quick self-assessments and to set realistic goals, which they could then track with self-monitoring tools. The workbook also contained space for notes on self-reflections and in-session exercises.
Facilitator Workbook: The facilitator workbook, developed as a guideline/tool for facilitating the group sessions, included the roles and responsibilities of the facilitators, and a description and outline of each group session.
## 2.5. Step 5: Expert Content Evaluation
Curriculum: After finalisation, the curriculum was recirculated to the original expert group for confirmation that the content was valid and suitable for the study population.
Participant and Facilitator Workbook: The participant workbook and facilitator workbook were validated through an iterative process by two members of the multi-disciplinary team working independently and consultatively.
## 2.6. Step 6: Design and Layout of Curriculum and Tools
To ensure that the design and layout of the booklets, specifically the curriculum booklet and the participant workbook were culturally and contextually appropriate, the project manager worked closely with the South African Medical Research Council’s (SAMRC) Corporate and Marketing Division. The latter is the brand custodian of the SAMRC and is directed by the SAMRC’s broader organisational strategy and strategic goals. Important considerations for the design team were the target populations’ age, gender, ethnicity, and socioeconomic profile, including education level (for readability). Suitable photos and visuals were sourced and the layout, booklet size, letters, and spacing were mutually agreed upon. After a few rounds of engagement and amendments, the project team agreed that the format was acceptable.
## 2.7. Step 7: Participant Readability and Acceptability Evaluation
All participants who participated in the FGDs in the needs assessment phase were invited to participate in a workshop (conducted by members of the SA-DPP project team, with the first author as the lead facilitator) to test the readability and acceptability of the curriculum booklet and participant workbook. Only 16 participants were available to attend; four workshops (two Xhosa speaking and two Afrikaans/English speaking) were held with three to five participants each.
During these workshops, participants engaged with the printed materials, whereafter they completed a validated assessment tool [26] that was adapted for our setting and provided verbal feedback. The acceptability score was based on 38 questions pertaining to content [4], language [12], illustrations/photos [4], sufficiently specific/understandability of the information [6], legibility and printing characteristics [10], and the quality of information [2], using a 3-point Likert scale. An acceptability score was calculated as the sum of the scores for the 38 questions, expressed as a percentage. The mean acceptability score for the participant workbook was $94.3\%$ (SD 7.6) and that for the curriculum booklet was $89.3\%$ (SD 13.3).
From the verbal feedback received, participants felt that the booklets were very informative and easy to understand. Participants identified illustrations, figures, and photos that they were unhappy with, e.g., some photos showed vegetables not usually eaten in their communities, while others were of younger adults and children as opposed to people representing them, an older population. Most participants requested larger font sizes and a more simplistic front cover. Xhosa speaking participants expressed a strong need for the booklets to be translated into Xhosa.
## 2.8. Step 8: Refining the Design and Layout of the Curriculum and Tools
Based on the information provided by the participants (in Step 7), the design and layout of the curriculum and workbook were improved by redesigning the cover, increasing the font size, and reworking or replacing some illustrations and photographs.
## 2.9. Step 9: Translation of the Curriculum and Participant Workbook
The curriculum and participant workbook were translated into Xhosa by a language expert and verified by a native Xhosa speaker, via back translation. The translated version has the same design and layout as the English version.
## 2.10. Step 10: Suitability Evaluation (Pilot Study)
The overall aim of the pilot study was to test the suitability of the curriculum, tools, and group session format with a group of people at risk of T2DM. The planned intervention will comprise of bi-weekly sessions for the first five sessions with the sixth session at the end of month eight, with no plan for follow-up thereafter. The pilot study was an accelerated version of the latter with weekly sessions over a six week period (4–10 November 2020). The sessions were facilitated by a dietitian and co-facilitated by the first author onsite at the South African Medical Research Council (the COVID-19 pandemic precluded us from using community venues and recruiting a lay community health worker as a peer facilitator). The two smallest sites by the number of at-risk participants recruited (Du Noon (black) and Belhar (mixed ancestry)) were chosen as pilot sites. Out of the 20 participants (10 from each site) invited, only 10 (seven from Belhar and three from Du Noon) accepted the invitation to be part of the pilot study. The main reasons for declining the invitation related to the COVID-19 pandemic and their unavailability due to family commitments, i.e., taking care of grandchildren and other family members. The mean age of the participants in the pilot study was 54.6 (SD = 10.3) years.
## 2.10.1. Summary of the Sessions as Experienced by Participants [Sessions 1–5: Table 5 and Table 6]
The process monitoring and evaluation, via evaluation sheets completed at the end of each session and verbal feedback, revealed great satisfaction with the intervention sessions. Session 2 (Healthy Eating Part 1) was the best attended ($100\%$); all other sessions had about $70\%$ attendance, collectively. Overall, most participants felt that they learnt something new in all the sessions, and exposure to information they already knew was still helpful. Across the sessions, an equal number of participants felt that the sessions were easy enough to follow vs. struggling at times. Participants found the self-assessments easy initially but needed some assistance as the sessions progressed. Participants experienced all the sessions as fun and engaging and looked forward to learning more. Most participants were excited and confident about making lifestyle changes. A few participants were anxious but determined to implement the lifestyle changes (Table 5).
**Table 5**
| Session: | 1-Diabetes | 2-Healthy Eating, Part 1 | 3-Healthy Eating, Part 2 | 4-Physical Activity | 5-Smoking, Alcohol & Stress |
| --- | --- | --- | --- | --- | --- |
| No. of Participants | 7 | 10 | 8 | 6 | 5 |
| How did you find today’s session? | How did you find today’s session? | How did you find today’s session? | How did you find today’s session? | How did you find today’s session? | How did you find today’s session? |
| I learned something new | 6 | 6 | 8 | 6 | 5 |
| Not all new, but the information was helpful | 3 | 5 | 1 | 1 | 2 |
| I could follow the session easily | I could follow the session easily | I could follow the session easily | I could follow the session easily | I could follow the session easily | I could follow the session easily |
| Yes, it was easy to follow | 7 | 5 | 4 | 5 | 3 |
| It was okay, but I struggled at times | 1 | 5 | 4 | 1 | 4 |
| I found the self-assessment/goal setting to be… | I found the self-assessment/goal setting to be… | I found the self-assessment/goal setting to be… | I found the self-assessment/goal setting to be… | I found the self-assessment/goal setting to be… | I found the self-assessment/goal setting to be… |
| Easy | 6 | 6 | 1 | 4 | 0 |
| Not too difficult but needed some help | 3 | 4 | 7 | 2 | 5 |
| I found the session to be fun and engaging? | I found the session to be fun and engaging? | I found the session to be fun and engaging? | I found the session to be fun and engaging? | I found the session to be fun and engaging? | I found the session to be fun and engaging? |
| Yes, it was just right | 5 | 3 | 2 | 3 | 2 |
| I enjoyed it and look forward to the next session, I look forward to learning more | 7 | 10 | 8 | 6 | 5 |
| I am excited to make some changes to my lifestyle | I am excited to make some changes to my lifestyle | I am excited to make some changes to my lifestyle | I am excited to make some changes to my lifestyle | I am excited to make some changes to my lifestyle | I am excited to make some changes to my lifestyle |
| Yes, I am excited | 6 | 5 | 5 | 4 | 5 |
| I am excited, I know I can make lifestyle changes | 3 | 6 | 4 | 1 | 1 |
| I am anxious about it but determined | 3 | 1 | 2 | 2 | 1 |
| I am anxious and unsure whether I can do this | 1 | 0 | 0 | 0 | 0 |
The majority of participants indicated that the group sessions were sufficient to equip them with the knowledge and tools necessary to make lifestyle changes possible. Pertaining to the specific lifestyle changes adopted, most participants aimed to increase their fruit and vegetable intake, and physical activity levels, and maintain a healthy food plate. In rating the difficulty to introduce lifestyle changes, participants equally experienced it as easy, not too difficult but needing to figure out things, and challenging but determined to continue (Table 6).
**Table 6**
| Session: | 1-Diabetes | 2-Healthy Eating, Part 1 | 3-Healthy Eating, Part 2 | 4-Physical Activity | 5-Smoking, Alcohol & Stress |
| --- | --- | --- | --- | --- | --- |
| No. of Participants | 7 | 10 | 8 | 6 | 5 |
| What would make it possible to make lifestyle changes? | What would make it possible to make lifestyle changes? | What would make it possible to make lifestyle changes? | What would make it possible to make lifestyle changes? | What would make it possible to make lifestyle changes? | What would make it possible to make lifestyle changes? |
| These group sessions will equip me with the knowledge and tools that I need | 6 | 6 | 5 | 5 | 4 |
| In addition to the group sessions, I will need the support of my family | 1 | 4 | 3 | 0 | 0 |
| Tell us about the changes you started making… | Tell us about the changes you started making… | Tell us about the changes you started making… | Tell us about the changes you started making… | Tell us about the changes you started making… | Tell us about the changes you started making… |
| Fruit & vegetables | * | 2 | 2 | 3 | 5 |
| Sugar intake | * | 1 | 1 | 1 | 0 |
| Fat intake | * | 1 | 0 | 0 | 1 |
| Healthy food plate | * | 1 | 1 | 0 | 4 |
| Physical activity | * | 2 | 2 | 2 | 4 |
| How difficult is it to make changes to your lifestyle? | How difficult is it to make changes to your lifestyle? | How difficult is it to make changes to your lifestyle? | How difficult is it to make changes to your lifestyle? | How difficult is it to make changes to your lifestyle? | How difficult is it to make changes to your lifestyle? |
| It has actually been easy! | ** | ** | 3 | 3 | 2 |
| It’s not been too difficult; however, I have a few things I must still figure out. | ** | ** | 5 | 2 | 1 |
| Even though it has been challenging I am determined to continue. | ** | ** | 4 | 2 | 4 |
## 2.10.2. Session 6—Final Session: Check-in on Implemented Lifestyle Changes
Seven of the ten participants attended the final session, which consisted of qualitative feedback. The themes and relevant quotes are summarised in Table 7. All participants, irrespective of their perception of the interventions as easy or difficult, experienced the pilot study positively and received the information well. They applied the information not only to themselves, but also to their families, and the extended community, and became role models (and thus programme champions) for their community. Being able to share their knowledge with their friends, family, and community was a recurring theme throughout. Notably, one participant reported that her spouse suffered from uncontrolled diabetes and that by the end of the 6-week pilot study his diabetes was controlled, and he could move from insulin to oral medication only. The participant attributed this to her ability to apply the knowledge learnt in the group sessions to her household. The same participant’s son’s behaviour at school improved by simply replacing his juice intake with water. Those who smoked ($$n = 3$$) including the one person that consumed alcohol, reported struggling with cutting down on smoking and alcohol. However, they were aware of the need to reduce these behaviours and eventually plan to stop.
Participants found the curriculum and participant booklets extremely helpful, the curriculum booklet as a source of information and the participant workbook helped to keep them on track with their goals. Even though they found the goal setting (SMART goals) tough at first, they thought this was a good way to do it. Some of the participants did not keep track of their goals because they were busy or simply forgot. Participants found the intervention team to be easy going (non-judgmental) and supportive. Regarding the format of the programme, the content and the facilitators, the participants were satisfied and felt motivated. Their only suggestion was that friends and family could perhaps be included.
## 2.10.3. Facilitator’s Reflections on the Sessions
The facilitator of the six weekly sessions of the pilot study was a bilingual (Xhosa and English speaking) dietitian, while the project manager took on the role of peer facilitator instead of the CHWs because of the COVID-19 restrictions. The facilitator noted that a few of the sessions ran over time, and that participants became more interactive and comfortable as the sessions progressed. The facilitator felt that the participants seemed to understand the information and enjoy the sessions and activities. The participants struggled slightly with the setting of SMART goals but were fine after some assistance. It was further noted that the use of the participant booklet for goal tracking at home was not well utilised. One of the sessions included an exercise on portion sizes, using food models. This highlighted two issues. Firstly, it demonstrated that participants grasped the concept of the divisions of the healthy food plate ($\frac{1}{2}$ a plate of non-starchy vegetables, ¼ starch, and ¼ protein), however portions were heaped on the plate. Secondly, the food models made the participants expectant and hungry, as it looked so real and life like.
## 2.11. Finalisation of the Curriculum
The curriculum was well received by all participants in the pilot study. The food plate exercise, however, revealed that a section with a direct focus on portion size was needed, which was subsequently added. Although the exercise with the food models was very useful, it created an expectancy that cannot be fulfilled by the research programme. Considering that SA-DPP is aimed at people in low-income communities, it was decided to remove the food models from the SA-DPP toolkit, for ethical reasons.
## 3. Discussion
The existing literature on the real-world adaptation of DPPs in high-income countries, including low-intensity DPPs, has shown varying effectiveness [4]. DPPs have also expanded to include adaptions suitable for different environments, such as the workplace [27,28], at schools targeting mothers, and via digital platforms targeting different audiences [29,30,31,32]. The importance of culturally tailoring interventions to specific minority groups and low socioeconomic populations has been recognised more and more over the years [5,33,34]. Interventions that are context specific and considered acceptable by the target population may have better adherence and therefore be more effective [35]. To our knowledge, a DPP adaption for low- and middle-income African countries has not yet been developed. Building on the evidence in the literature, the SA-DPP project team embarked on developing a context specific and culturally appropriate healthy lifestyle curriculum for South Africa, to ultimately be implemented in low socioeconomic populations at high risk of developing TSDM.
Although the objectives, principles, and themes of the Finnish Diabetes Prevention Study [1,12] and the K-DPP [6] form the core of the SA-DPP (Table S1), the SA-DPP curriculum is aligned with the nutrition messages from the South African Department of Health (i.e., the SAFBDG). The SAFBDG are short encouraging food-based dietary recommendations aimed at empowering the South African population to make healthier food choices (based on existing eating patterns) that will contribute to a nutritionally adequate diet and lower the risk of non-communicable diseases [24]. Behaviour change is, however, complex. The context specific barriers and opportunities for healthy eating should therefore be considered when adapting lifestyle interventions to promote population-level uptake without widening socioeconomic inequalities [36]. Barriers and enablers for healthy eating were explored during two rounds of FGDs in our study, and financial constraints was identified as the major barrier. It was, therefore, important to tailor the nutrition messages within the boundaries of the financial constraints of our target population, without compromising the intervention.
Sekhon et al. [ 35] suggest that the intervention needs to be acceptable to both the recipients and the facilitators of the intervention. When assessing acceptability prior to an intervention, the recipient’s perspective on the content, context, and perceived quality of the intervention is considered. Should an intervention be deemed acceptable, recipients are more likely to adhere to the intervention recommendations and benefit from it [35]. If the facilitators perceive the delivery to have low acceptability, the intervention may not be delivered as intended, and thereby impact its overall effectiveness [35]. Sekhon et al. [ 35] suggest a theoretical framework of acceptability with methodology for the development phase, the pilot or feasibility phase, the evaluation phase, and implementation phase.
A systematic review of culturally targeted strategies for diabetes prevention in minority populations by Lagissetty et al. [ 5], showed that, to be effective, interventions need to tailored across the following four domains: facilitators, language, location, and messaging. The SA-DPP intervention is a group based, peer support programme that is designed to educate and to provide support for people at high risk of developing T2DM. The groups will be jointly led by SA-DPP personnel and peer leaders selected from each community and trained on the SA-DPP intervention (facilitators/deliverers), who will deliver the intervention in the spoken/preferred language of the group (language). The purpose of the SA-DPP peer support groups (messaging/mode of delivery), which will take place in a community venue (location), is to provide a safe place where people who are at high risk of T2DM can meet regularly to support each other, develop friendships, share experiences, and learn from the experiences of others to improve their lifestyle. During the SA-DPP curriculum development process described in this paper, the intervention developers (experts/programme deliverers) and the intended recipients found the final SA-DPP curriculum and tools acceptable (including language/readability/literacy competence and messaging, i.e., the educational content), and the pilot study deemed it feasible. In essence giving the SA-DPP an increased chance of being effective.
The next phase of the SA-DPP, is the intervention phase (trial context) that will be implemented in 16 resource-poor communities. During the intervention phase, process evaluation techniques will be used to evaluate the curriculum, tools, and intervention on an ongoing basis. Process evaluation will be both formative and summative, to aid the dual purpose of helping fine tune the intervention during the implementation phase, and to evaluate the extent to which the intervention was implemented as planned and reached the intended participants. This information will in turn be used to assist the interpretation and description of the SA-DPP outcomes, analysing how the programme worked, and providing input for the future scaling up of the SA-DPP. Targeted elements of the implementation will include the fidelity, the dose (delivered and received), the reach, the recruitment, the retention/maintenance, and the context.
The biggest limitation to our study can also be viewed as its biggest strength in the context of tailoring an intervention to be culturally appropriate. Our intervention tools and curriculum have been developed with a specific focus on individuals at risk of developing T2DM in resource-poor communities in Cape Town, South Africa. Our process has enabled us to be mindful of key barriers and facilitators and has been demonstrated as was the case in Goff et al. [ 33], factoring in the importance of the context in which the intervention is aiming to effect change. However, our findings may not be generalisable to other contexts. Engaging the target population in the development of the intervention, the curriculum, and the tools, will promote ownership and inherent programme retention.
## 4. Conclusions
This paper presents the mixed-method staged approach that was followed in developing a context specific, culturally tailored intervention for diabetes prevention in South Africa, which can potentially be adapted for other African countries. Engaging the target population in the developing process ensured that individual and societal barriers and facilitators for healthy living are considered in the intervention and curriculum. Furthermore, engaging the target population promotes ownership and inherent programme retention. The intervention and curriculum were deemed acceptable, which would make intervention adherence and retention more plausible. Evidence in the literature strongly supports the potential effectiveness of culturally tailored interventions. The next step for the SA-DPP is a complete evaluation of this culturally relevant model for T2DM prevention in South Africa, which will provide some of the first evidence on the applicability and transferability of the findings from the international diabetes prevention efficacy trials in an African country.
## References
1. Uusitupa M., Louheranta A., Lindström J., Valle T., Sundvall J., Eriksson J., Tuomilehto J.. **The Finnish Diabetes Prevention Study**. *Br. J. Nutr.* (2000) **83** S137-S142. DOI: 10.1017/S0007114500001070
2. Tuomilehto J., Lindstrom J., Eriksson J.G., Valle T.T., Hämäläinen H., Ilanne-Parikka P., Keinänen-Kiukaanniemi S., Laakso M., Louheranta A., Rastas A.. **Prevention of type 2 diabetes mellitus by changes in lifestyle among subjects with impaired glucose tolerance**. *N. Engl. J. Med.* (2001) **344** 1343-1350. DOI: 10.1056/NEJM200105033441801
3. **The Diabetes Prevention Program (DPP): Description of lifestyle intervention**. *Diabetes Care* (2002) **25** 2165-2171. DOI: 10.2337/diacare.25.12.2165
4. Aziz Z., Absetz P., Oldroyd J., Pronk N.P., Oldenburg B.A.. **systematic review of real-world diabetes prevention programs: Learnings from the last 15 years**. *Implement. Sci.* (2015) **10** 172. DOI: 10.1186/s13012-015-0354-6
5. Lagisetty P.A., Priyadarshini S., Terrell S., Hamati M., Landgraf J., Chopra V., Heisler M.. **Culturally Targeted Strategies for Diabetes Prevention in Minority Population**. *Diabetes Educ.* (2017) **43** 54-77. DOI: 10.1177/0145721716683811
6. Sathish T., Williams E.D., Pasricha N., Absetz P., Lorgelly P., Wolfe R., Mathews E., Aziz Z., Thankappan K.R., Zimmet P.. **Cluster randomised controlled trial of a peer-led lifestyle intervention program: Study protocol for the Kerala diabetes prevention program**. *BMC Public Health* (2013) **13**. DOI: 10.1186/1471-2458-13-1035
7. Mathews E., Thomas E., Absetz P., D’Esposito F., Aziz Z., Balachandran S., Daivadanam M., Thankappan K.R., Oldenburg B.. **Cultural adaptation of a peer-led lifestyle intervention program for diabetes prevention in India: The Kerala diabetes prevention program (K-DPP)**. *BMC Public Health* (2018) **17**. DOI: 10.1186/s12889-017-4986-0
8. Ackermann R.T.. **Working with the YMCA to implement the Diabetes Prevention Program**. *Am. J. Prev. Med.* (2013) **44** S352-S356. DOI: 10.1016/j.amepre.2012.12.010
9. 9.
International Diabetes Foundation
IDF Diabetes Atlas10th ed.IDFBrussels, Belgium2021Available online: https://diabetesatlas.org/atlas/tenth-edit(accessed on 12 January 2023). *IDF Diabetes Atlas* (2021)
10. 10.
International Diabetes Foundation
IDF Diabetes Atlas8th ed.IDFBrussels, Belgium2017Available online: https://diabetesatlas.org(accessed on 1 September 2021). *IDF Diabetes Atlas* (2017)
11. Pheiffer C., Pillay-van Wyk V., Turawa E., Levitt N., Kengne A.P., Bradshaw D.. **Prevalence of Type 2 Diabetes in South Africa: A Systematic Review and Meta-Analysis**. *Int. J. Environ. Res. Public Health* (2021) **18**. DOI: 10.3390/ijerph18115868
12. Lindstrom J., Louheranta A., Mannelin M., Rastas M., Salminen V., Eriksson J., Uusitupa M., Tuomilehto J.. **The Finnish Diabetes Prevention Study (DPS): Lifestyle intervention and 3-year results on diet and physical activity**. *Diabetes Care* (2003) **26** 3230-3236. DOI: 10.2337/diacare.26.12.3230
13. Absetz P., Valve R., Oldenburg B., Heinonen H., Nissinen A., Fogelholm M., Ilvesmäki V., Talja M., Uutela A.. **Type 2 diabetes prevention in the “real world”: One-year results of the GOAL Implementation Trial**. *Diabetes Care* (2007) **30** 2465-2470. DOI: 10.2337/dc07-0171
14. Knowler W.C., Barrett-Connor E., Fowler S.E., Hamman R.F., Lachin J.M., Walker E.A., Nathan D.M.. **Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin**. *N. Engl. J. Med.* (2002) **346** 393-403. DOI: 10.1056/NEJMoa012512
15. Thankappan K.R., Sathish T., Tapp R.J., Shaw J.E., Lotfaliany M., Wolfe R., Absetz P., Mathews E., Aziz Z., Williams E.D.. **A peer-support lifestyle intervention for preventing type 2 diabetes in India: A cluster-randomized controlled trial of the Kerala Diabetes Prevention Program**. *PLoS Med.* (2018) **15**. DOI: 10.1371/journal.pmed.1002575
16. Hill J., Peer N., Jonathan D., Mayige M., Sobngwi E., Kengne A.P.. **Findings from Community-Based Screenings for Type 2 Diabetes Mellitus in at Risk Communities in Cape Town, South Africa: A Pilot Study**. *Int. J. Environ. Res. Public Health* (2020) **17**. DOI: 10.3390/ijerph17082876
17. Rappaport J.. **Terms of empowerment/exemplars of prevention: Toward a theory for community psychology**. *Am. J. Community Psychol.* (1987) **15** 121-148. DOI: 10.1007/BF00919275
18. Perkins D.D.. **Speaking truth to power: Empowerment ideology as social intervention and policy**. *Am. J. Community Psychol.* (1995) **23** 765-794. DOI: 10.1007/BF02506991
19. Whittemore R.. **A systematic review of the translational research on the Diabetes Prevention Program**. *Transl. Behav. Med.* (2011) **1** 480-491. DOI: 10.1007/s13142-011-0062-y
20. Hill J., Peer N., Oldenburg B., Kengne A.P.. **Roles, responsibilities and characteristics of lay community health workers involved in diabetes prevention programmes: A systematic review**. *PLoS ONE* (2017) **12**. DOI: 10.1371/journal.pone.0189069
21. Rawal L.B., Tapp R.J., Williams E.D., Chan C., Yasin S., Oldenburg B.. **Prevention of type 2 diabetes and its complications in developing countries: A review**. *Int. J. Behav. Med.* (2012) **19** 121-133. DOI: 10.1007/s12529-011-9162-9
22. Ackermann R.T., Marrero D.G.. **Adapting the Diabetes Prevention Program lifestyle intervention for delivery in the community: The YMCA model**. *Diabetes Educ.* (2007) **33** 69-78. DOI: 10.1177/0145721706297743
23. Hill J., Lavigne Delville C., Auorousseau A.M., Jonathan D., Peer N., Oldenburg B., Kengne A.P.. **Development of a Tool to Increase Physical Activity among People at Risk for Diabetes in Low-Resourced Communities in Cape Town**. *Int. J. Environ. Res. Public Health* (2020) **17**. DOI: 10.3390/ijerph17030865
24. Vorster H.H., Badham J.B., Venter C.S.. **An introduction to the revised food-based dietary guidelines for South Africa**. *S. Afr. J. Clin. Nutr.* (2013) **26** S5-S12
25. Prochaska J.O., Velicer W.F.. **The transtheoretical model of health behavior change**. *Am. J. Health Promot.* (1997) **12** 38-48. DOI: 10.4278/0890-1171-12.1.38
26. Castro M.S., Pilger D., Fuchs F.D., Ferreira M.B.. **Development and validity of a method for the evaluation of printed education material**. *Pharm. Pract.* (2007) **5** 89-94. DOI: 10.4321/s1886-36552007000200007
27. DeJoy D.M., Padilla H.M., Wilson M.G., Vandenberg R.J., Davis M.A.. **Worksite translation of the Diabetes Prevention Program: Formative research and pilot study results from FUEL Your Life**. *Health Promot. Pract.* (2013) **14** 506-513. DOI: 10.1177/1524839912461014
28. Padilla H.M., Zuercher H., Robertson M., DeJoy D.M., Wilson M., Vandenberg R.J., Corso P.S.. **Reach, Uptake, and Satisfaction of Three Delivery Modes of FUEL Your Life**. *Health Promot. Pract.* (2021) **22** 415-422. DOI: 10.1177/1524839919869921
29. Kim S.E., Sweet C.M.C., Cho E., Tsai J., Cousineau M.R.. **Evaluation of a digital diabetes prevention program adapted for low-income patients, 2016–2018**. *Prev. Chronic Dis.* (2019) **16** E155. DOI: 10.5888/pcd16.190156
30. Alzeidan R., Shata Z., Hassounah M.M., Baghdadi L.R., Hersi A., Fayed A., Kashour T., Elmorshedy H.. **Effectiveness of digital health using the transtheoretical model to prevent or delay type 2 diabetes in impaired glucose tolerance patients: Protocol for a randomized control trial**. *BMC Public Health* (2019) **19**. DOI: 10.1186/s12889-019-7921-8
31. McGough B., Murray E., Brownlee L., Barron E., Smith J., Valabhji J.. **The Healthier You: NHS Diabetes Prevention Programme: Digital modes of delivery engage younger people. Letter**. *Diabetes Med.* (2019) **36** 1510-1511. DOI: 10.1111/dme.14083
32. Kim S.E., Castro Sweet C.M., Gibson E., Madero E.N., Rubino B., Morrison J., Rosen D., Imberg W., Cousineau M.R.. **Evaluation of a digital diabetes prevention program adapted for the Medicaid population: Study design and methods for a non-randomized, controlled trial**. *Contemp. Clin. Trials Commun.* (2018) **10** 161-168. DOI: 10.1016/j.conctc.2018.05.007
33. Goff L.M., Moore A.P., Harding S., Rivas C.. **Development of Healthy Eating and Active Lifestyles for Diabetes, a culturally tailored diabetes self-management education and support programme for Black-British adults: A participatory research approach**. *Diabetes Med.* (2021) **38** e14594. DOI: 10.1111/dme.14594
34. Neville L.M., O’Hara B., Milat A.J.. **Computer-tailored dietary behaviour change interventions: A systematic review**. *Health Educ. Res.* (2009) **24** 699-720. DOI: 10.1093/her/cyp006
35. Sekhon M., Cartwright M., Francis J.J.. **Acceptability of healthcare interventions: An overview of reviews and development of a theoretical framework**. *BMC Health Serv. Res.* (2017) **17**. DOI: 10.1186/s12913-017-2031-8
36. Woodside J., Young I.S., McKinley M.C.. **Culturally adapting the Mediterranean Diet pattern—A way of promoting more ‘sustainable’ dietary change?**. *Br. J. Nutr.* (2022) **128** 693-703. DOI: 10.1017/S0007114522001945
|
---
title: Strawberry, Blueberry, and Strawberry-Blueberry Blend Beverages Prevent Hepatic
Steatosis in Obese Rats by Modulating Key Genes Involved in Lipid Metabolism
authors:
- Ana María Sotelo-González
- Rosalía Reynoso-Camacho
- Ana Karina Hernández-Calvillo
- Ana Paola Castañón-Servín
- David Gustavo García-Gutiérrez
- Haiku Daniel de Jesús Gómez-Velázquez
- Miguel Ángel Martínez-Maldonado
- Ericka Alejandra de los Ríos
- Iza Fernanda Pérez-Ramírez
journal: International Journal of Environmental Research and Public Health
year: 2023
pmcid: PMC10002361
doi: 10.3390/ijerph20054418
license: CC BY 4.0
---
# Strawberry, Blueberry, and Strawberry-Blueberry Blend Beverages Prevent Hepatic Steatosis in Obese Rats by Modulating Key Genes Involved in Lipid Metabolism
## Abstract
There is an increasing interest in developing natural herb-infused functional beverages with health benefits; therefore, in this study, we aimed to evaluate the effect of strawberry, blueberry, and strawberry-blueberry blend decoction-based functional beverages on obesity-related metabolic alterations in high-fat and high-fructose diet-fed rats. The administration of the three berry-based beverages for eighteen weeks prevented the development of hypertriglyceridemia in obese rats (1.29–1.78-fold) and hepatic triglyceride accumulation (1.38–1.61-fold), preventing the development of hepatic steatosis. Furthermore, all beverages significantly down-regulated Fasn hepatic expression, whereas the strawberry beverage showed the greatest down-regulation of Acaca, involved in fatty acid de novo synthesis. Moreover, the strawberry beverage showed the most significant up-regulation of hepatic Cpt1 and Acadm (fatty acid β-oxidation). In contrast, the blueberry beverage showed the most significant down-regulation of hepatic Fatp5 and Cd36 (fatty acid intracellular transport). Nevertheless, no beneficial effect was observed on biometric measurements, adipose tissue composition, and insulin resistance. On the other hand, several urolithins and their derivatives, and other urinary polyphenol metabolites were identified after the strawberry-based beverages supplementation. In contrast, enterolactone was found significantly increase after the intake of blueberry-based beverages. These results demonstrate that functional beverages elaborated with berry fruits prevent diet-induced hypertriglyceridemia and hepatic steatosis by modulating critical genes involved in fatty acid hepatic metabolism.
## 1. Introduction
The global market of functional beverages was about $117 billion USD in 2021 and is expected to grow to $156 billion USD in 2026 at a compound annual growth rate of $6\%$. The key reasons associated with the growth in the functional beverages market include increased awareness of health-conscious consumers shifting from hypercaloric juices or carbonated beverages to healthier hydration products. Moreover, consumers are interested in ready-to-drink functional beverages with clean labels, including low-calorie natural sweeteners and natural pigments [1,2].
Interestingly, the natural botanical-infused drinks trend dominated the functional beverage market in 2021, including using plant extracts, such as roots, flowers, leaves, seeds, and fruits [2]. In this regard, berry fruits have been widely used to develop functional beverages due to their sensory attributes and bioactive compound content [3].
We previously demonstrated that strawberry and blueberry decoctions are rich in polyphenols, mainly anthocyanins; nevertheless, these decoctions showed contrasting polyphenol profiles. The strawberry decoction was rich in pelargonidin hexoside and included ellagitannins as minor components. Conversely, the blueberry decoction showed a high content of petunidin hexoside, malvidin hexoside, delphinidin hexoside, and cyanidin hexoside, as several flavonols as minor components [4].
Polyphenols can be absorbed in the small intestine or reach the colon, which can be metabolized by colonic microbiota and absorbed in the large intestine. Once absorbed, polyphenols suffer phase II metabolism (glucuronidation, sulphation, and methylation) in enterocytes and hepatocytes, which are then distributed to their target organs and are further excreted in urine [5]. Several anthocyanin metabolites, mainly glucuronide derivatives, have been reported associated with consuming blueberry and strawberry fruit and juice. In addition, ellagic acid and urolithins have been identified after strawberry fruit intake, produced during the metabolism of ellagitannins.
Interestingly, several studies have demonstrated that berry fruits exert beneficial effects mainly related to improving cardiovascular health. It has been previously demonstrated that strawberry and blueberry fruit prevent the development of hepatic steatosis hepatic by decreasing lipid accumulation, improving insulin sensitivity, and decreasing inflammation and oxidative stress based on in vitro, in vivo, and clinical studies [6]. Moreover, it has been demonstrated that a blended powder elaborated with freeze-dried strawberry and blueberry also exerts these insulin sensitizers and hypolipidemic effects, attributed to the modulation of inflammatory and lipogenic biomarkers [7]. Regarding berry-based beverages, the supplementation of a strawberry smoothie blended with other fruits ameliorated hepatic steatosis in high-fat diet-fed obese mice [8]. Similarly, blueberry juices ameliorate the development of prediabetes and hepatic steatosis in high-sucrose and high-fat diet-fed rats [9]. These health benefits have been attributed to their high content and diversity of polyphenols when different whole fruits, juices, or smoothie-based beverages were used.
Interestingly, we have previously demonstrated that strawberry and blueberry decoctions are rich in polyphenols showing potential complementary profiles, and thus can be proposed as ingredients for the elaboration of functional beverages; in addition, these beverages are low in calories [10]. Therefore, to contribute to developing natural botanical-infused functional beverages with health benefits, this study aimed to evaluate the preventive effect of polyphenol-rich strawberry and blueberry decoction-based beverages on developing obesity-related metabolic alterations in high-fat and high-fructose (HFF) diet-fed rats. Moreover, phase II and colonic polyphenol urinary metabolites were identified as potential contributors to the health-beneficial effects of berry-based beverages.
## 2.1. Chemical and Reagents
Optima LC/MS grade methanol and water; and JT *Baker potassium* sorbate, sodium carbonate, Invitrogen Trizol reagent; and ACS grade chloroform, methanol, and ethanol were purchased from Fisher Scientific (Waltham, MA, USA). Citric acid, 1 N Folin–Ciocalteu reagent, $99\%$ formic acid, (−)-epicatechin, naringenin, quercetin, cyanidin chloride, ellagic acid, gallic acid, 4-hydroxyphenylpropionic acid, 4-hydroxyphenylacetic acid, enterodiol, hippuric acid, 2,2-diphenyl-1-picrylhydrazyl (DPPH), 2,2′-azino-bis(3-ethylbenzo-thiazoline-6-sulfonic acid) (ABTS), 6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid (Trolox), potassium persulphate were purchased from Sigma-Aldrich (St. Louis, MO, USA). M-MLV RT, M-MLV 5× reaction buffer, oligo (dT)15 primer, dNTP mix, SybrGreen master mix, and RNAse-free water were purchased from Promega.
## 2.2. Elaboration of the Berry-Based Beverages
Strawberry and blueberry decoctions (aqueous extracts) were elaborated at $10\%$ (w/v) at 95 °C for 15 min. Then, beverages were prepared by mixing the berry decoctions ($100\%$ strawberry, $100\%$ blueberry, and $50\%$ strawberry/$50\%$ blueberry, respectively) with a preservative (potassium sorbate), non-caloric sweeteners (sucralose and acesulfame K), and a pH regulator (citric acid). Finally, the berry-based beverages were pasteurized at 73 °C for 15 s and immediately cooled in ice. All beverages were stored at 4 °C for 5 days until use. Microbiologic analyses (total aerobic mesophyll bacteria count, total coliform count, fecal coliform count, and mold and yeast count) were assessed to guarantee to be innocuous for the in vivo study.
## 2.3. Polyphenol Profile by UPLC-Q-ToF MS
Freshly prepared beverages were passed through syringe PVDF filters (0.45 μm, 13 mm), and were analyzed in an Ultra-Performance Liquid Chromatograph (UPLC) coupled to a Quadrupole/Time-of-Flight Mass Spectrometer (Q-ToF MS) with an electrospray ionization (ESI) interphase (Vion, Waters Co., Milford, MA, USA). Samples (1 μL) were injected into a BEH Acquity C18 column (2.1 × 100 mm2, 1.7 μm) at 35 °C. The mobile phase consisted of (A) water:formic acid (99:1 v/v), and (B) acetonitrile:formic acid (99:1 v/v). Gradient conditions and Q-Tof MS processing and acquisition conditions were previously reported by Reynoso-Camacho et al. [ 4]. Mass spectra were analyzed for the identification of polyphenols by comparison of their exact mass (mass error < 5 ppm) and fragmentation patterns. Calibration curves were constructed with (−)-epicatechin (flavanols), naringenin (flavanones), quercetin (flavonols), cyanidin chloride (anthocyanins), ellagic acid (hydroxycinnamic acids and ellagitannins), and gallic acid (hydroxybenzoic acids). High-resolution mass spectra at high and low collision energy of the significant polyphenols identified in this study are shown in Figure S1.
## 2.4. In Vivo Experimental Procedure
The animal experiment was performed following the guidelines of the National Research Council for using experimental animals. It was approved by the Bioethics Committee of the Chemistry School of the Autonomous University of Querétaro (Querétaro, México) (approval number: CBQ$\frac{16}{0831}$). Fifty male Wistar rats of 160–180 g weight were purchased from the Institute of Neurobiology of the Universidad Nacional Autónoma de México (Querétaro, México). Rats were maintained at 24 ± 1 °C and 50 ± $10\%$ RH under a $\frac{12}{12}$ h light/dark cycle.
After one week of acclimation, rats were randomly divided into five experimental groups of ten animals each: (i) standard diet-fed group (Rodent Lab Chow 5001); (ii) HFF diet-fed group (standard diet added with $20\%$ lard and $18\%$ fructose); (iii) HFF diet-fed group supplemented with the strawberry beverage (SB) group; (iv) HFF diet-fed group supplemented with the blueberry beverage (BB) group; and (v) HFF diet-fed group supplemented with the strawberry-blueberry blend beverage (SBB) group.
Berry-based beverages were prepared every five days and were administered for 12 h at night, followed by 12 h of tap water. In contrast, tap water was administered ad libitum to the control groups. Standard or HFF diets were administered ad libitum throughout the experiment, and food and beverage intake was recorded daily throughout the experiment.
After 18 weeks, rats were placed in metabolic cages to recollect of urine and feces. Then, rats were anesthetized with pentobarbital (0.4 mL/kg of body weight via intraperitoneal) and were euthanized by decapitation. Blood was collected into Vacutainer tubes (BD Co., Bergen, NJ, USA) and was centrifuged at 1000× g for 10 min for serum separation. Liver and mesenteric, epididymal, and perirenal adipose tissues were collected. A fraction of each liver was stored in $10\%$ neutral buffered formalin (pH 7.4) for histology analysis. The rest of the organs and the biological fluids were snap-frozen in liquid nitrogen and stored at −80 °C until analysis.
## 2.4.1. Biometric and Adipose Tissue Measurements
Body weight was measured weekly, and biometric measurements were assessed every two weeks. The biometric measurements included abdominal circumference, thoracic circumference, and naso-anal length. With these data, body mass index (BMI), Lee index, and rate of body mass gain were calculated with the following equations, which have been proposed to estimate obesity in rats [10]:[1]BMI=body weight (g)[length (cm)]2 [2]Lee index=bodyweight (g)×103length (mm) The relative weight of mesenteric, epididymal, perirenal, and total adipose tissue was determined. The adiposity index was estimated with the following equation:[3]Adiposity index (%)=Total adipose tissue (g)body weight (g)∗100
## 2.4.2. Determination of Serum Biochemical Analysis
Serum glucose (Glucose-LQ GOD-POD, Spinreact, Girone, Spain), triglycerides (Triglycerides-LQ GPO-POD, Spinreact), aspartate transaminase (GOT-AST-LQ, Spinreact), and alanine transaminase (GPT-ALT-LQ, Spinreact) were determined using commercial enzymatic-colorimetric kits following the manufacturer’s instructions. Following the manufacturers’ instructions, serum insulin was determined using an ELISA kit (EZRMI-13K, EMD Millipore Co., Burlington, MA, USA). Insulin resistance- and beta-Homeostatic model assessment (HOMA-IR and HOMA-beta), quantitative insulin sensitivity check index (QUICKI), fasting glucose-to-insulin ratio (FGIR), and fasting triglycerides and glucose (TyG) indexes were calculated with the following equations:[4]HOMA−IR=glucose (mgdL)×insulin(μMmL)2430 [5]HOMA−Beta=insulin (ngmL)×360glucose (mgdL)−63 [6]QUICKI=1log[insulin(μMmL)]+log[glucose(mgdL)] [7]FGIR=glucose (mgdL)insulin (μMmL) [8]TyG=ln[glucose(mgdL)∗triglycerides(mgdL)]2
## 2.4.3. Triglyceride Extraction and Quantification in Feces, Adipose Tissue, and Liver
Feces were dried at 45 °C for 24 h prior to triglyceride extraction. Dried feces (200 mg) were mixed with 2 mL of $0.9\%$ NaCl and with 2 mL of 2:1 (v/v) chloroform:methanol. Next, samples were centrifuged at 6000× g for 5 min at 25 °C, and 1 mL of the lower phase was collected in a new tube. Then, samples were vacuum dried at 35 °C and resuspended in 200 μL of ethanol. Finally, following the manufacturer’s instructions, triglycerides were determined using a commercial enzymatic-colorimetric kit (Triglycerides-LQ GPO-POD, Spinreact). Results were expressed as mg of triglycerides/g of feces [11].
Frozen livers (200 mg) were digested with 350 μL of $30\%$ potassium hydroxide in ethanol (2:1, v/v) at 55 °C overnight. Then, 650 μL water:ethanol (1:1, v/v) was added, and samples were centrifuged at 6000× g for 5 min. Supernatants (200 μL) were recovered and mixed with 215 μL of 1 M magnesium chloride. Samples were incubated in ice for 10 min and centrifuged at 6000× g. Finally, following the manufacturer’s instructions, triglycerides were determined using a commercial enzymatic-colorimetric kit (Triglycerides-LQ GPO-POD, Spinreact). Results were expressed as mg of triglycerides/g of the liver [12].
## 2.4.4. Fatty Acid Profile in Liver
Frozen livers (50 mg) were extracted with 400 μL of 1.25 M potassium hydroxide in methanol for 60 s and were sonicated for 5 min at 40 kHz at room temperature. Then, samples were mixed with 400 μL of 1.75 M sulfuric acid in methanol for 60 s and sonicated for 5 min at 40 kHz at room temperature. Finally, samples were mixed with 0.8 mL of n-hexane for 30 s, centrifuged at 10,000× g for 5 min at 25 °C, and supernatants were recovered [13].
Derivatized samples (1 μL) were injected into an HP-88 capillary column (30 m × 0.25 mm, 0.25 μm) in a Gas Chromatograph (Agilent 7890A, Agilent Technologies Inc., Santa Clara, CA, USA) coupled to a simple-Quadrupole Mass Spectrometer (Agilent 5976C) with an electron impact (EI) ionization source (GC/EI-Q MS). The injector temperature was set at 250 °C in split mode (1:50). Helium was used as carrier gas at 1 mL/min. The oven temperature gradient was set as follows: 50 °C held for 1 min, then raised to 175 °C at 15 °C/min, then raised to 240 °C at 1 °C/min and held for 5 min. The following MS conditions were used: Q MS at 150 °C, EI ionization source at 230 °C, MS electron energy at 70 eV with a mass range of 50–1100 m/z, and a solvent delay of 6.4 min. Results were expressed as mg/g of the liver.
## 2.4.5. Relative Expression of Genes Involved in Hepatic Lipid Metabolism
RNA was extracted from liver samples (30–50 mg) with Trizol reagent following the manufacturer’s instructions. Then, RNA integrity was confirmed by $0.5\%$ agarose gel electrophoresis, whereas RNA purity ($\frac{260}{280}$ and $\frac{260}{230}$ ratios) and concentration (260 nm) were determined in a microplate spectrophotometer (Multiskan GO, Thermo Fisher Scientific, Waltham, MA, USA).
cDNA synthesis was done by mixing 2 μg of total RNA with 2 μL of oligo (dT) at 2 μg/mL (Promega), and RNAse-free water up to a total volume of 15 μL. Samples were incubated at 70 °C for 5 min in a thermal cycler (C1000 Touch, Bio-Rad Laboratories, Hercules, CA, USA). Then, samples were cooled in ice. Afterward, samples were mixed with 5 μL of M-MLV 5× reaction buffer (Promega), 1.25 μL of 10 mM dNTP mix (Promega), 7 μL of RNAse inhibitors (Promega), 1 μL of M-MLV RT (Promega), and RNAse-free water up to a total volume of 25 μL. Finally, samples were incubated at 37 °C for 60 min.
mRNA expression was assessed by real-time PCR. Briefly, 1 μL of cDNA was mixed with 10 μL of SybrGreen master mix, 1 μL of each primer (10 μM), and 3 μL of RNAse-free water. Then, samples were incubated under the following conditions: pre-incubation, 95 °C for 10 min; denaturation, 40 cycles at 95 °C for 10 s; primer alignment, 56 °C for 10 s; elongation, 72 °C for 10 s. Melting curves were acquired with the following gradient: 95 °C for 10 s; 65 °C for 60 s; and 97 °C for 1 s. Amplification was assessed for the following transcripts Fasn, Acaca, Acadm, Cpt1, Cd36, and Fatp5 using the primers and annealing temperature described in the Supplementary Material (Table S1). mRNA relative expression was calculated by normalization against Actin according to the 2−ΔΔCt method [14].
## 2.4.6. Liver Histology Analysis
Formalin-fixed livers were cleared with xylene, then hydrated with gradient ethanol solutions. Then, samples were embedded in paraffin at 60 °C and sectioned at 5 μm. Finally, samples were stained with Hematoxylin and Eosin (H&E) solution, dewaxed, and dehydrated with gradient ethanol solutions. Samples were observed and photographed under a microscope at 40×, analyzing six sections per animal [15].
## 2.4.7. Extraction and Identification of Urinary Polyphenol Metabolites
Urine samples were subjected to solid phase extraction (SPE) to analyze polyphenol-derived colonic and phase II metabolites. Supel-Select HLB SPE (60 mg/3 mL) cartridges were activated with 3 mL of methanol and 3 mL of water. Then, 2 mL of urine samples were added, followed by a clean-up with 2 mL of 1.5 M formic acid and 2 mL of water:methanol 95:5 (v/v) to remove interferents. Finally, polyphenol-derived metabolites were eluted with 2 mL of methanol [16].
UPLC-Q-ToF MS was used to assess the polyphenol-derived metabolites profile by following the methodology described in Section 2.3. An aliquot of 1 mL of the samples eluted by SPE was evaporated to dryness (Speedvac Savant, Thermo Fisher Scientific) and resuspended in 200 μL of methanol. For quantification, calibration curves were constructed with 4-hydroxyphenyl propionic acid (phenyl propionic acids and valerolactones), 4-hydroxyphenyl acetic acid (phenylacetic acids), enterodiol (lignans), and hippuric acid (glycinate benzoic acids).
## 2.5. Statistical Analysis
All data are described as mean values ± standard deviation. For the berry-based characterization, three experimental replicates were carried out, and three technical replicates were analyzed. In contrast, ten biological replicates were used for the in vivo experimental design, and three technical replicates were analyzed in each assay. Outliers (>1.5 QR) and extreme (>3.0 QR) values were identified using box-and-whisker plots, and extreme values were excluded. Kolmogorov–Smirnov’s test was used for normality evaluation and Levene’s test for variance homogeneity assessment. Then, multiple comparisons were carried out with Tukey’s test for parametric variables and Kruskal–Wallis for non-parametric parameters. Differences with $p \leq 0.05$ were considered significant. All statistical analyses were carried out in JMP software v16. Principal Component Analysis (PCA), sparse Partial Least Square-Discriminant Analysis (sPLS-DA), and K-means plots were constructed with the urinary metabolites profile of each experimental group using the Metaboanalyst 5.0 online software. The preprocessing steps included data normalization by sum, square root transformation, and auto-scaling.
## 3.1. Polyphenol Characterization of the Strawberry, Blueberry, and Strawberry-Blueberry Blend Beverages
The polyphenol composition of the berry fruit-based beverages is shown in Table 1. Sixteen non-pigmented flavonoids, ten anthocyanins, twenty-five phenolic acids, and four ellagitannins were identified in the berry fruit-based beverages. Regarding non-pigmented flavonoids, the blueberry beverage and strawberry-blueberry beverage showed high concentrations of quercetin rhamnoside and quercetin hexoside, which were not identified in the strawberry beverage. Conversely, the strawberry beverage showed a high content of procyanidin dimer B2, eriodictyol, kaempferol hexoside, and (+)-catechin.
Nevertheless, the most significant flavonoids identified in all the berry-based beverages were anthocyanins. The strawberry beverage showed pelargonidin hexoside as the significant anthocyanin, followed by pelargonidin rutinoside, which was not identified in the blueberry beverage. Interestingly, the blueberry beverage showed a richer profile due to its wide variety of anthocyanins. This beverage showed malvidin hexoside as the significant component, followed by malvidin pentoside, peonidin hexoside, cyanidin hexoside, petunidin hexoside, and delphinidin hexoside. As expected, the strawberry-blueberry blend beverage showed a combination of the anthocyanin profile of the individual strawberry and blueberry beverages.
Regarding phenolic acids, the strawberry and strawberry-blueberry blend beverages showed high concentration of hydroxybenzoic acid hexoside and coumaric acid hexoside. In contrast, the blueberry beverage showed a poor profile of hydroxybenzoic acids but showed a high content of chlorogenic acid (caffeoylquinic acid isomer II). In addition, ellagic acid and ellagitannins, mainly peduncalagin and strictinin, were identified in the strawberry and strawberry-blueberry blend beverages, which were not detected in the blueberry beverage.
## 3.2. Effect of the Strawberry, Blueberry, and Strawberry-Blueberry Blend Beverages on Obesity in High-Fat and High-Fructose Diet-Fed Rats
HFF diet-fed rats were supplemented with berry-based beverages for 18 weeks to evaluate their effect on obesity-related metabolic alterations. The effect of berry-based beverages on the biometric and adipose tissue measurements in HFFD-fed rats is shown in Table 2. As expected, after 18 weeks of induction, the HFF diet-fed (HFFD) control group showed increased body weight gain as compared to the standard diet-fed (SD) control group (1.23-fold, $$p \leq 0.0002$$), leading to an augmented BMI (1.21-fold, $$p \leq 0.0006$$), thus indicating the development of diet-induced obesity.
Moreover, the HFFD control group showed increased triglyceride accumulation in adipose tissue (1.31-fold, $$p \leq 0.0026$$), and an increased mesenteric (2.83-fold, $$p \leq 0.0007$$), epididymal (2.17-fold, $p \leq 0.0001$), perirenal (2.64-fold, $p \leq 0.0001$), and total adipose tissue relative weight (2.48-fold, $p \leq 0.0001$), leading to an augmented adiposity index (2.00-fold, $p \leq 0.0001$) as compared to the SD control group. Rats supplemented with berry-based beverages showed a similar daily diet and beverage intake as compared to the HFFD control group (22.3–22.0 vs. 21.5 g/day/rat and 40.7–42.1 vs. 41.0 mL/day/rat, respectively). Nevertheless, none of the berry-based beverages significantly decreased obesity-related parameters compared to the HFFD control group.
## 3.3. Effect of the Strawberry, Blueberry, and Strawberry-Blueberry Blend Beverages on Insulin Resistance in High-Fat and Fructose Diet-Fed Rats
The effect of berry-based beverages on insulin resistance parameters in HFFD-fed rats is shown in Table 2. The HFFD control group showed similar serum glucose levels as compared to the SD control group. Nevertheless, the HFFD control group showed increased serum insulin values (2.03-fold, $p \leq 0.0001$), leading to the development of insulin resistance, as observed in high HOMA-IR and TyG index and low FGIR index values, and low pancreatic β-cell function, as observed in high HOMA-Beta index values as compared to the SD control group. Regarding the supplementation with berry-based beverages, no significant differences were observed in all the insulin resistance parameters compared to the HFFD control group.
## 3.4. Effect of the Strawberry, Blueberry, and Strawberry-Blueberry Blend Beverages on Triglyceride Metabolism in High-Fat and Fructose Diet-Fed Rats
The effect of berry-based beverages on serum, fecal, and hepatic triglycerides in HFFD-fed rats is shown in Table 3. The HFFD control group showed increased serum and hepatic triglyceride levels (2.45- and 2.63-fold, respectively; $$p \leq 0.0008$$ and $p \leq 0.0001$) compared to the SD control group. Moreover, the administration of the HFFD increased the fecal triglyceride excretion compared to the SD control group (1.43-fold, $$p \leq 0.0400$$). Interestingly, the three berry-based beverages decreased serum triglyceride levels compared to the HFFD control group (1.29–1.78-fold, $$p \leq 0.0048$$ and $$p \leq 0.0139$$). Similarly, all berry-based beverages reduced the accumulation of hepatic triglycerides in comparison with the HFFD control group (1.38–1.63-fold, $$p \leq 0.0009$$, $$p \leq 0.0012$$, and $$p \leq 0.0233$$). Moreover, the strawberry and blueberry beverage supplemented HFFD groups showed similar serum and hepatic triglyceride values as compared to the SD control group. On the other hand, the strawberry beverage slightly increased the fecal triglyceride excretion compared to the HFFD control group (1.45-fold, $$p \leq 0.0067$$); however, no significant differences were observed.
Figure 1 shows the liver histology analysis of the experimental groups. The HFFD control group developed hepatic steatosis observed in the presence of lipid vacuoles within the hepatocytes (Figure 1B), whereas the SD control group showed no lipid vacuoles (Figure 1A). Interestingly, the three berry-based beverages decreased the accumulation of lipid vacuoles (Figure 1C,D,E) compared to the HFFD control group (Figure 1B).
Regarding the hepatic lipid metabolism, Figure 2 shows the effect of berry-based beverages on crucial genes involved in fatty acid de novo synthesis, intracellular transport, and β-oxidation. The HFFD control group showed an increased expression of Acaca and Fasn (1.40- and 1.49-fold, respectively, $$p \leq 0.0022$$ and $p \leq 0.0001$, Figure 2A) and Fatp5 and Cd36 (1.53-fold, $$p \leq 0.0017$$ and $$p \leq 0.0002$$, Figure 2B), and a decreased expression of Cpt1 and Acadm (1.69- and 1.67-fold, respectively, $$p \leq 0.0451$$ and $$p \leq 0.0201$$, Figure 2C) as compared to the SD control group.
Interestingly, the three berry-based beverages down-regulated Fasn expression as compared to the HFFD control group (1.39–1.62-fold, $p \leq 0.0001$, $$p \leq 0.0002$$ and $$p \leq 0.0018$$), showing similar expression values to the SD control group, whereas the blueberry and strawberry-blueberry blend beverages down-regulated Acaca expression as compared to the HFFD control group (1.93- and 1.59-fold, respectively, $p \leq 0.0001$ and $$p \leq 0.0003$$).
The effect of berry-based beverages on the hepatic fatty acid profile in HFFD-fed rats is shown in Table 3. As expected, the HFFD control group showed a significant ($p \leq 0.05$) increased concentration of most hepatic fatty acids as compared to the SD control group, except for EPA (cis-5,8,11,14,17-eicosapentaenoic acid, and c205ω3), which was slightly decreased (1.50-fold). Interestingly, the strawberry and blueberry-based beverages slightly decreased the accumulation of saturated (1.37- and 1.25-fold, respectively), monounsaturated (1.36- and 1.30-fold, respectively), and polyunsaturated (1.35- and 1.15-fold, respectively) fatty acids as compared to the HFFD control group; however, no significant differences were observed. Even though the three berry-based beverages showed a similar beneficial effect on hepatic triglyceride accumulation, the strawberry beverage showed a more significant effect on the hepatic fatty acid profile.
## 3.5. Urinary Polyphenol Metabolites Associated with the Supplementation of Strawberry, Blueberry, and Strawberry-Blueberry Blend Beverages in High-Fat and High-Fructose Diet-Fed Rats
The effect of berry-based beverages on the urinary polyphenol metabolite profile in HFFD-fed rats is shown in Table S2. Twenty-three polyphenol metabolites were identified in SD- and HFFD-fed rats, mainly flavone and isoflavone metabolites, such as apigenin glucuronide, equol glucuronide, and hydroxydaidzein. Interestingly, the strawberry beverage excreted the greatest variety of polyphenol metabolites ($$n = 48$$), with urolithins A and B as significant compounds, which were not detected in the control groups. The blueberry and strawberry-blueberry blend beverages showed a lower variety of urinary metabolites ($$n = 33$$ and 30, respectively), where urinary enterolactone was significantly increased compared to the HFFD control group (3.02- and 2.55-fold, respectively, $$p \leq 0.008$$). Multivariate analyses of the urinary polyphenol metabolite profile are shown in Figure 3, Figure 4 and Figure 5.
The unsupervised PCA and the supervised sPLS-DA models showed a total explained variance of $54.8\%$ and $54.5\%$, respectively, showing the discrimination between some experimental groups (Figure 3A and Figure 4A, respectively). The greatest variance was explained by component 1 (x-axis), which discriminated between rats fed a SD, a HFFD, and a HFFD supplemented with the blueberry and blueberry-strawberry beverages and those rats fed a HFFD supplemented with the strawberry beverage. Similar results were observed in the K-means clustering analysis, which also confirmed the integration of three clear clusters: cluster 1: rats fed a HFFD supplemented with the blueberry and blueberry-strawberry beverages; cluster 2: rats fed a HFFD supplemented with the strawberry beverage; and cluster 3: rats fed with SD and HFFD (control groups), indicating that the urinary metabolite profile is similar between those rats included in each cluster.
The urinary metabolites responsible for the discrimination between the experimental groups are observed in the biplot (Figure 3B), VIP score plots (Figure 4C,D), and K-means features (Figure 5B). Interestingly, several polyphenol urinary metabolites were identified as descriptors of the rats fed a HFFD supplemented with the strawberry beverage, such as catechol sulfate, urolithin B glucuronide, urolithin C, dihydroferulic acid, methylurolithin A, methylpyrogallol sulfate, and hydroxyhippuric acid (loading score > 0.25; Table S3); whereas the primary discriminant metabolites of the rats fed a HFFD supplemented with the blueberry and strawberry-blueberry beverages were methyl-(epi)-catechin glucuronide, dihydroxyphenylvalerolactone, and enterolactone (loading score > 0.25; Table S4). On the other hand, both control groups showed methoxyhydroxyphenylvalerolactone, equol glucuronide, dimethyl quercetin, and hydroxyglicitein as discriminant urinary metabolites (loading score > 0.25; Table S3).
## 4. Discussion
Numerous experimental studies (in vitro, in vivo, and clinical trials), and a meta-analysis of epidemiological studies, have demonstrated that berry fruits can prevent or counteract diet-induced obesity and obesity-related complications due to their high content of bioactive compounds, mainly dietary fiber, vitamins, and polyphenols [17]. Therefore, in this study, we hypothesized that functional beverages could be developed with decoctions elaborated with polyphenol-rich berry fruits, such as strawberries and blueberries, which can exert some of the health beneficial effects associated with the consumption of whole-berry fruits since polyphenols are extracted during the decoction process used in the elaboration of functional beverages as previously demonstrated [4].
All the berry-based beverages developed in this study showed a high content of polyphenols, mainly anthocyanins. As expected, contrasting profiles were identified in strawberry and blueberry-based beverages since the strawberry-based beverage showed pelargonidin hexoside as a the main polyphenol, followed by several hydroxycinnamic acids and ellagitannins, such as coumaric acid hexoside, ellagic acid, peduncalagin, and strictinin, while the blueberry-based beverage was rich in several flavonoids, including malvidin hexoside, malvidin pentoside, peonidin hexoside, quercetin rhamnoside, quercetin hexoside, and in chlorogenic acid. These primary polyphenols agree with those identified in both strawberry and blueberry fruits [6,18].
In this study, we attempt to demonstrate the bioactivity of the berry fruits beverages using an in vivo model induced with a HFFD for eighteen weeks that led to the development of obesity accompanied by insulin resistance, impaired function of β-pancreatic cells, hypertriglyceridemia, and hepatic steatosis. The supplementation with all berry-based beverages did not prevent body weight gain nor the accumulation of triglycerides in adipose tissue induced by the HFFD. In this regard, Prior et al. [ 19] demonstrated that strawberry and blueberry fruits did not prevent weight gain in high-fat diet-fed rats. In contrast, an equivalent concentration of purified anthocyanins of these berry fruits exerted antiobesogenic effects. Moreover, a methanolic strawberry extract promoted adipocyte browning and inhibited adipogenesis in 3T3 L1 cells [18], whereas blueberry ethanolic extract increased energy expenditure in brown adipose tissue and adipocyte browning in the inguinal white adipose tissue [20].
Similarly, the berry-based beverages developed in this study did not exert a protective effect against developing HFFD-induced hyperglycemia, hyperinsulinemia, insulin resistance, and β-cell pancreatic damage. Conversely, Liu et al. [ 21] reported that the supplementation with freeze-dried whole blueberry powder increased insulin sensitivity and glucose tolerance in high-fat diet-induced mice, improving b-cell survival, and preventing β-cell hypertrophy. In contrast, Aranaz et al. [ 7] demonstrated that a freeze-dried strawberry and blueberry blend powder reduced hyperinsulinemia and insulin resistance in high-fat and high-sucrose diet-fed obese rats. Notably, the polyphenol content of the berry extracts or powders of these previous studies was higher than those found in our study, which could be partly related to the lack of an anti-obesogenic and insulin sensitizer effect of our berry-based beverages.
On the other hand, the three berry-based beverages developed in our study exerted a chronic anti-hypertriglyceridemic effect as observed in decreased serum triglyceride levels as compared to the HFFD control group, which was not associated with an increased fecal triglyceride excretion, but was accompanied by the prevention of the development of the fatty liver. Furthermore, all berry-based beverages decreased the hepatic accumulation of triglycerides, whereas the strawberry beverage decreased the hepatic accumulation of saturated fatty acids. Accordingly, Wang et al. [ 22] reported that anthocyanin-rich extracts of wild blueberry and strawberry fruits decreased triglyceride accumulation in HepG2 cells induced with oleic acid to simulate an in vitro fatty liver. Interestingly, when purified anthocyanins were evaluated, cyanidin 3-O-glucoside and delphinidin 3-O-glucoside exerted the greatest clearance of hepatic triglycerides, anthocyanins identified in both beverages elaborated with blueberries, but not in the $100\%$ strawberry-based beverage. Moreover, Liu et al. [ 23] demonstrated that the blueberry phenolic acid-rich fraction exerted a more significant inhibitory effect on triglyceride accumulation as compared to the anthocyanin-rich fraction in oleic acid-induced HepG2 cells, which was partly related to the improvement of triglycerides clearance by caffeic acid and chlorogenic acid, this latter was the major phenolic acid identified in the blueberry-based beverage.
The hepatic accumulation of triglycerides in a combination with obesity and insulin resistance is known as metabolic-associated fatty liver disease (MAFLD), which is associated with an imbalance of hepatic lipid metabolism pathways, such as the import/export of fatty acids, de novo fatty acid biosynthesis (lipogenesis), and fatty acid catabolism (β-oxidation) [24]. The HFFD model used for this study developed a mild stage of MAFLD as observed in the accumulation of lipid vacuoles within hepatocytes without the elevation of hepatic injury markers (ALT, AST, and ALP). Moreover, the HFFD control group showed an increased expression of Acaca and Fasn (lipogenesis), and Fatp5 and Cd36 (fatty acid uptake), and a decreased expression of Cpt1 and Acadm (β-oxidation).
Hepatic de novo lipogenesis is mainly regulated by acetyl-CoA carboxylase (ACC), which converts acetyl-CoA to malonyl-CoA. Then, FAS carries out the conversion to palmitate, which is further elongated and desaturated for the synthesis of multiple fatty acids, which can be further esterified for the synthesis of triglycerides. Interestingly, all berry-based beverages down-regulated Fasn expression, whereas both beverages elaborated with blueberries down-regulated Acaca expression, achieving a similar expression compared to the SD control group. However, only the supplementation with the strawberry-based beverage significantly reduced the hepatic accumulation of palmitic acid, a well-known lipotoxic compound that promotes the synthesis of ceramides and diglycerides that further promote the transition from MAFLD to the proinflammatory non-alcoholic steatohepatitis (NASH) [24].
Fatty acid β-oxidation is mainly regulated by the CPT1 enzyme, which is responsible for the transport of fatty acids into the mitochondria. In contrast, acyl-CoA dehydrogenases (ACAD) participate in the first step of mitochondrial fatty acid oxidation. All berry-based beverages up-regulated Cpt1a and Acadm hepatic expression; however, the strawberry-based beverage showed higher expression values than the SD control group. In addition, hepatocytes control the flux of fatty acids via fatty acid transport proteins (FATP5) and CD36, which are commonly increased in high-fat diet-induced MAFLD [24]. The supplementation with the $100\%$ strawberry- and $100\%$ blueberry-based beverages down-regulated Cd36 expression, whereas the blueberry- and the strawberry-blueberry blend-based beverages down-regulated Fatp5 expression.
Interestingly, even though different genes were up- or down-regulated by each berry-based beverage, the three beverages exerted a similar reduction effect on the accumulation of lipid vacuoles within the hepatocytes. Similar results were reported with the supplementation of wild blueberry fruits, which up-regulated the expression of the peroxisome proliferator-activated receptor-alpha (PPAR-α) in obese rats, which regulates the expression of Cpt1a and down-regulated the expression of the sterol regulatory element-binding protein 1 (SREBP-1), which regulates the expression of Fasn and *Acaca* genes [25]. On the other hand, the effect of strawberry fruit or derived products has not been reported on hepatic lipid metabolism. Nevertheless, Zhang et al. [ 26] demonstrated that ellagic acid, which was only detected in strawberry-based beverages, attenuates the development of hepatic steatosis by the inhibition of the activity and transcription of hepatic SREBP-1, FAS, and ACC. Moreover, punicalagin, the major ellagitannin identified in the strawberry-based beverage, and ellagic acid exert a protective effect against palmitate-induced mitochondrial dysfunction in HepG2 cells [27].
Finally, in this study, we conducted a targeted metabolomic analysis of urine samples obtained to identify the polyphenol metabolites associated with the chronic daily supplementation of berry-based beverages. Notably, several polyphenol urinary metabolites were identified in all the experimental groups, which is associated with the polyphenol intake of rodent diet that is elaborated with soybean as one of the major ingredients. All animals excreted a relatively high amount of apigenin glucuronide, equol glucuronide, and hydroxydaidzein. Accordingly, soybeans are uniquely rich in isoflavones, such as daidzin, genistin, and glycitin, which are hydrolyzed into their aglycones (daidzein, genistein, and glycitein, respectively) in the small intestine, and could be further absorbed intact or metabolized by colonic microbiota, which catalyzes their hydroxylation or glucuronidation and can promote the formation of equol (7,4′-isoflavandiol) [28,29]. Moreover, both control groups (SD and HFFD) showed a similar urinary polyphenol metabolite profile according to the chemometric analysis, since their polyphenol intake was exclusively provided by the standard rodent diet. The main metabolites identified as discriminants of the control groups were equol glucuronide (major urinary metabolite) and several minor components, such as methoxyhydroxyphenylvalerolactone, dimethyl quercetin, and hydroxyglycitein.
According to the multivariate analysis, the main urinary metabolites associated with the consumption of the strawberry-based beverage included catechol sulfate, dihydroferulic acid, methylpyrogallol sulfate, and hydroxyhippuric acid, and urolithin B glucuronide, urolithin C, and mehtylurolihin A. Truchado et al. [ 30] identified urolithin A glucuronide as the predominant urinary metabolite after the consumption of whole strawberry fruits, followed by urolithin A, urolithin B, and urolithin B glucuronide, which are produced during the colonic fermentation of ellagitannins and ellagic acid. It is noteworthy that ellagitannins are partially hydrolyzed into ellagic acid in the small intestine. However, these polyphenols have a low bioavailability; therefore, most of them reach the colon, where are subjected to gut microbiota metabolism, including the formation of urolithins and their further sulfatation, methylation, and glucuronidation [31]. Interestingly, the formation of urolithins after strawberry intake is not affected by the food processing process [31]; nevertheless, ellagitannins are not exclusive to strawberry since these polymeric polyphenols are also found in pomegranate and several nuts, and even though each source has a unique ellagitannin profile, the urinary metabolite profile is similar [32].
On the other hand, previous studies have identified anthocyanin metabolites after consuming different berry fruits and derived products [32]; however, these metabolites were not identified in our study. Nevertheless, catechol sulfate and hydroxyhippuric acid were identified as discriminant urinary metabolites of the strawberry-based beverage, which are metabolites derived from the colonic fermentation of several anthocyanins [33].
Interestingly, even though the strawberry-blueberry blend-based beverage showed a mixed polyphenol profile, its urinary polyphenol metabolite profile was similar to that found for the $100\%$ blueberry-based beverage, characterized by the excretion of methyl-(epi)-catechin glucuronide, dihydrophenylvalerolactone, and enterolactone. Accordingly, Ancillotti et al. [ 34] and Langer et al. [ 35] also identified phenylvalerolactone derivatives as blueberry fruit and juice urinary derivatives, respectively. However, these metabolites cannot be considered exclusive from blueberry consumption since valerolactones are produced during the colonic fermentation of flavanals and proanthocyanidins (polymeric flavanals), which are extensively distributed in fruits, vegetables, nuts, and seeds [36]. Interestingly, enterolactone is a colonic microbiota derivative from lignans [37]; however, lignans were not identified in the blueberry-based beverage. However, secoisolariciresinol and matairesinol are the main lignans identified in berry fruits [38]. Therefore, we hypothesized that lignans were lower than the detection limit in our berry-based beverages. However, their colonic metabolites could have been accumulated in the urinary bladder and thus were detected as urinary polyphenol metabolites. Even though the discriminant urinary polyphenol metabolites identified in this study are not exclusive of strawberry and blueberry consumption, they can be associated with the health-beneficial effects exerted by berry-based beverages. For instance, urolithin A and urolithin B down-regulate Srepb1a and up-regulate Ppara hepatic expression in high-fat diet-fed rats, decreasing the accumulation of hepatic triglycerides [39].
Finally, it is noteworthy that the main health-beneficial effect exerted by the berry fruit decoction-based beverages was the prevention of hypertriglyceridemia and hepatic steatosis. Even though we demonstrated that strawberry and blueberry blended beverage showed a greater diversity of polyphenols, this beverage exerted a slightly lower preventive effect on these metabolic alterations as compared to the single fruit-based beverage, which could be related to the complex synergetic, additive, and antagonistic interactions between polyphenols that affect their global health impact [40].
## 5. Conclusions
The results obtained in this study demonstrate that berry fruits, such as strawberries and blueberries, can be used to develop functional beverages to prevent high-fat and high-fructose diet-induced hypertriglyceridemia and hepatic steatosis through the modulation of key genes involved in fatty acid hepatic metabolism. Furthermore, these health-beneficial effects found in berry fruit decoction-based beverages were associated with their polyphenol composition, including urinary metabolites, such as urolithins. Nevertheless, further studies must be carried out to confirm the results obtained in this in vivo study through placebo-controlled clinical trials and understand the synergetic, additive, or antagonistic interactions between polyphenols.
## References
1. **North America Functional Beverage Market—Growth, Trends, COVID-19 Impact, and Forecasts (2022–2027)**
2. **Functional Beverages Global Market Report 2022—By Type (energy Drinks, Sport Drinks, Nutraceutical Drinks, Dairy-Based Beverages, Juices, Enhances Water, Other Types), by Function (Health & Wellness, Weight Management), by Distribution Channel (Brick & Mortar, Online)—Market Size, Trends, and Global Forecast 2022–2026**
3. Dey G., Sireswar S.. **Tailoring functional beverages from fruits and vegetables for specific disease conditions-are we there yet?**. *Crit. Rev. Food Sci. Nutr.* (2021.0) **61** 2034-2046. DOI: 10.1080/10408398.2020.1769021
4. Reynoso-Camacho R., Sotelo-González A.M., Patiño-Ortiz P., Rocha-Guzmán N.E., Pérez-Ramírez I.F.. **Berry by-products obtained from a decoction process are a rich source of low-and high-molecular weight extractable and non-extractable polyphenols**. *Food Bioprod. Process.* (2021.0) **127** 371-387. DOI: 10.1016/j.fbp.2021.03.014
5. Di Lorenzo C., Colombo F., Biella S., Stockley C., Restani P.. **Polyphenols and human health: The role of bioavailability**. *Nutrients* (2021.0) **13**. DOI: 10.3390/nu13010273
6. Di Pede G., Favari C., Bresciani L., Almutairi T.M., Del Rio D., Crozier A., Klimis-Zacas D., Roidriguez-Mateos A.. **Occurrence, bioavailability and metabolism of berry (poly)phenols**. *Berries and Berry Bioactive Compounds in Promoting Health* (2022.0) 41-82
7. Aranaz P., Romo-Hualde A., Zabala M., Navarro-Herrera D., de Galarreta M.R., Gil A.G., Martinez J.A., Milagro F.I., González-Navarro C.J.. **Freeze-dried strawberry and blueberry attenuates diet-induced obesity and insulin resistance in rats by inhibiting adipogenesis and lipogenesis**. *Food Funct.* (2017.0) **8** 3999-4013. DOI: 10.1039/C7FO00996H
8. Quitete F.T., Santos G.M.A., de Oliveira Ribeiro L., da Costa C.A., Freitas S.P., da Matta V.M., Daleprane J.B.. **Phenolic-rich smoothie consumption ameliorates non-alcoholic fatty liver disease in obesity mice by increasing antioxidant response**. *Chem. Biol. Interact.* (2021.0) **336** 109369. DOI: 10.1016/j.cbi.2021.109369
9. Gonçalves A.C., Nunes A.R., Flores-Félix J.D., Alves G., Silva L.R.. **Cherries and blueberries-based beverages: Functional foods with antidiabetic and immune booster properties**. *Molecules* (2022.0) **27**. DOI: 10.3390/molecules27103294
10. Novelli E.L.B., Diniz Y.S., Galhardi C.M., Ebaid G.M.X., Rodrigues H.G., Mani F., Fernandes A.A.H., Cicogna A.C., Novelli Filho J.L.V.B.. **Anthropometrical parameters and markers of obesity in rats**. *Lab. Anim.* (2007.0) **41** 111-119. DOI: 10.1258/002367707779399518
11. Kraus D., Yang Q., Kahn B.B.. **Lipid extraction from mouse feces**. *Bio-Protocol* (2015.0) **5** e1375. DOI: 10.21769/BioProtoc.1375
12. Norris A.W., Chen L., Fisher S.J., Szanto I., Ristow M., Jozsi A.C., Hirshman M.F., Rosen E.D., Goodyear L.J., Gonzalez F.J.. **Muscle-specific PPARγ-deficient mice develop increased adiposity and insulin resistance but respond to thiazolidinediones**. *J. Clin. Investig.* (2003.0) **112** 608-618. DOI: 10.1172/JCI17305
13. Sinosaki N.B.M., dos Santos P.D.D.S., Galuch M.B., da Silveira R., Bonafé E.G., Visentainer J.V., Santos Júnior O.D.O.. **Analytical method of direct derivatization of fatty acids in seeds**. *Chem. Pap.* (2019.0) **73** 2399-2407. DOI: 10.1007/s11696-019-00787-w
14. Livak K.J., Schmittgen T.D.. **Analysis of relative gene expression data using real-time quantitative PCR and the 2**. *Methods* (2001.0) **25** 402-408. DOI: 10.1006/meth.2001.1262
15. Sethunath D., Morusu S., Tuceryan M., Cummings O.W., Zhang H., Yin X.M., Vanderbeck S., Chalasani N., Gawrieh S.. **Automated assessment of steatosis in murine fatty liver**. *PLoS ONE* (2018.0) **13**. DOI: 10.1371/journal.pone.0197242
16. Roura E., Andrés-Lacueva C., Estruch R., Lamuela-Raventós R.M.. **Total polyphenol intake estimated by a modified Folin–Ciocalteu assay of urine**. *Clin. Chem.* (2006.0) **52** 749-752. DOI: 10.1373/clinchem.2005.063628
17. Jiang H., Zhang W., Li X., Xu Y., Cao J., Jiang W.. **The anti-obesogenic effects of dietary berry fruits: A review**. *Food Res. Int.* (2021.0) **147** 110539. DOI: 10.1016/j.foodres.2021.110539
18. Forbes-Hernández T.Y., Cianciosi D., Ansary J., Mezzetti B., Bompadre S., Quiles J.L., Giampieri F., Battino M.. **Strawberry (Fragaria × ananassa cv. Romina) methanolic extract promotes browning in 3T3-L1 cells**. *Food Funct.* (2020.0) **11** 297-304. DOI: 10.1039/C9FO02285F
19. Prior R.L., Wu X., Gu L., Hager T., Hager A., Wilkes S., Howard L.. **Purified berry anthocyanins but not whole berries normalize lipid parameters in mice fed an obesogenic high fat diet**. *Mol. Nutr. Food Res.* (2009.0) **53** 1406-1418. DOI: 10.1002/mnfr.200900026
20. Guo J., Han X., Tan H., Huang W., You Y., Zhan J.. **Blueberry extract improves obesity through regulation of the gut microbiota and bile acids via pathways involving FXR and TGR5**. *iScience* (2019.0) **19** 676-690. DOI: 10.1016/j.isci.2019.08.020
21. Liu W., Mao Y., Schoenborn J., Wang Z., Tang G., Tang X.. **Whole blueberry protects pancreatic beta-cells in diet-induced obese mouse**. *Nutr. Metab.* (2019.0) **16** 34. DOI: 10.1186/s12986-019-0363-6
22. Wang Y., Zhao L., Wang D., Huo Y., Ji B.. **Anthocyanin-rich extracts from blackberry, wild blueberry, strawberry, and chokeberry: Antioxidant activity and inhibitory effect on oleic acid-induced hepatic steatosis in vitro**. *J. Sci. Food. Agric.* (2016.0) **96** 2494-2503. DOI: 10.1002/jsfa.7370
23. Liu Y., Wang D., Zhang D., Lv Y., Wei Y., Wu W., Zhou F.M., Tang M., Mao T., Li M.. **Inhibitory effect of blueberry polyphenolic compounds on oleic acid-induced hepatic steatosis in vitro**. *J. Agric. Food Chem.* (2011.0) **59** 12254-12263. DOI: 10.1021/jf203136j
24. Badmus O.O., Hillhouse S.A., Anderson C.D., Hinds T.D., Stec D.E.. **Molecular mechanisms of metabolic associated fatty liver disease (MAFLD): Functional analysis of lipid metabolism pathways**. *Clin. Sci.* (2022.0) **136** 1347-1366. DOI: 10.1042/CS20220572
25. Vendrame S., Daugherty A., Kristo A.S., Klimis-Zacas D.. **Wild blueberry (**. *Br. J. Nutr.* (2014.0) **111** 194-200. DOI: 10.1017/S0007114513002390
26. Zhang C., Hu J., Sheng L., Yuan M., Wu Y., Chen L., Wang G., Qiu Z.. **Ellagic acid ameliorates AKT-driven hepatic steatosis in mice by suppressing de novo lipogenesis via the AKT/SREBP-1/FASN pathway**. *Food Funct.* (2019.0) **10** 3410-3420. DOI: 10.1039/C9FO00284G
27. Zou X., Yan C., Shi Y., Cao K., Xu J., Wang X., Chen C., Luo C., Li Y., Gao J.. **Mitochondrial dysfunction in obesity-associated nonalcoholic fatty liver disease: The protective effects of pomegranate with its active component punicalagin**. *Antioxid. Redox Signal.* (2014.0) **21** 1557-1570. DOI: 10.1089/ars.2013.5538
28. Chiou Y.S., Wu J.C., Huang Q., Shahidi F., Wang Y.J., Ho C.T., Pan M.H.. **Metabolic and colonic microbiota transformation may enhance the bioactivities of dietary polyphenols**. *J. Funct. Foods* (2014.0) **7** 3-25. DOI: 10.1016/j.jff.2013.08.006
29. Sri Harsha P.S., Wahab R.A., Garcia-Aloy M., Madrid-Gambin F., Estruel-Amades S., Watzl B., Andrés-Lacueva C., Brennan L.. **Biomarkers of legume intake in human intervention and observational studies: A systematic review**. *Genes Nutr.* (2018.0) **13** 25. DOI: 10.1186/s12263-018-0614-6
30. Truchado P., Larrosa M., García-Conesa M.T., Cerdá B., Vidal-Guevara M.L., Tomás-Barberán F.A., Espín J.C.. **Strawberry processing does not affect the production and urinary excretion of urolithins, ellagic acid metabolites, in humans**. *J. Agric. Food Chem.* (2012.0) **60** 5749-5754. DOI: 10.1021/jf203641r
31. Garcia-Munoz C., Vaillant F.. **Metabolic fate of ellagitannins: Implications for health, and research perspectives for innovative functional foods**. *Crit. Rev. Food Sci. Nutr.* (2014.0) **54** 1584-1598. DOI: 10.1080/10408398.2011.644643
32. Feliciano R.P., Boeres A., Massacessi L., Istas G., Ventura M.R., Dos Santos C.N., Heiss C., Rodriguez-Mateos A.. **Identification and quantification of novel cranberry-derived plasma and urinary (poly) phenols**. *Arch. Biochem. Biophys.* (2016.0) **599** 31-41. PMID: 26836705
33. Ulaszewska M., Garcia-Aloy M., Vázquez-Manjarrez N., Soria-Florido M.T., Llorach R., Mattivi F., Manach C.. **Food intake biomarkers for berries and grapes**. *Genes Nutr.* (2020.0) **15** 17. DOI: 10.1186/s12263-020-00675-z
34. Ancillotti C., Ulaszewska M., Mattivi F., Del Bubba M.. **Untargeted metabolomics analytical strategy based on liquid chromatography/electrospray ionization linear ion trap quadrupole/orbitrap mass spectrometry for discovering new polyphenol metabolites in human biofluids after acute ingestion of vaccinium myrtillus berry supplement**. *J. Am. Soc. Mass. Spectrom.* (2018.0) **30** 381-402. PMID: 30506347
35. Langer S., Kennel A., Lodge J.K.. **The influence of juicing on the appearance of blueberry metabolites 2 h after consumption: A metabolite profiling approach**. *Br. J. Nutr.* (2018.0) **119** 1233-1244. DOI: 10.1017/S0007114518000855
36. Clarke E.D., Collins C.E., Rollo M.E., Kroon P.A., Philo M., Haslam R.L.. **The relationship between urinary polyphenol metabolites and dietary polyphenol intakes in young adults**. *Br. J. Nutr.* (2022.0) **127** 589-598. DOI: 10.1017/S0007114521001343
37. Perez-Jimenez J., Hubert J., Hooper L., Cassidy A., Manach C., Williamson G., Scalbert A.. **Urinary metabolites as biomarkers of polyphenol intake in humans: A systematic review**. *Am. J. Clin. Nutr.* (2010.0) **92** 801-809. DOI: 10.3945/ajcn.2010.29924
38. Krogholm K.S., Bysted A., Brantsæter A.L., Jakobsen J., Rasmussen S.E., Kristoffersen L., Toft U.. **Evaluation of flavonoids and enterolactone in overnight urine as intake biomarkers of fruits, vegetables and beverages in the Inter99 cohort study using the method of triads**. *Br. J. Nutr.* (2012.0) **108** 1904-1912. DOI: 10.1017/S0007114512000104
39. Abdulrahman A.O., Kuerban A., Alshehri Z.A., Abdulaal W.H., Khan J.A., Khan M.I.. **Urolithins attenuate multiple symptoms of obesity in rats fed on a high-fat diet**. *Diabet. Metab. Synd. Obes. Targets Ther.* (2020.0) **13** 3337-3348. DOI: 10.2147/DMSO.S268146
40. Azzini E., Giacometti J., Russo G.L.. **Antiobesity effects of anthocyanins in preclinical and clinical studies**. *Oxid. Med. Cell. Longev.* (2017.0) **2017** 2740364. DOI: 10.1155/2017/2740364
|
---
title: Physical Impact of SARS-CoV-2 Infection in a Population of Italian Healthcare
Workers
authors:
- Lucrezia Ginevra Lulli
- Antonio Baldassarre
- Annarita Chiarelli
- Antonella Mariniello
- Diana Paolini
- Maddalena Grazzini
- Nicola Mucci
- Giulio Arcangeli
journal: International Journal of Environmental Research and Public Health
year: 2023
pmcid: PMC10002388
doi: 10.3390/ijerph20054506
license: CC BY 4.0
---
# Physical Impact of SARS-CoV-2 Infection in a Population of Italian Healthcare Workers
## Abstract
SARS-CoV-2 infection often causes symptoms and illness that can last for months after the acute phase, i.e., so-called “Long COVID” or Post-acute COVID-19. Due to the high prevalence of SARS-CoV-2 infection among Healthcare Workers (HCWs), post-COVID-19 symptoms can be common and threaten workers’ occupational health and healthcare systems’ functioning. The aim of this cross-sectional, observational study was to present data related to post-COVID-19 outcomes in a population of HCWs infected by COVID-19 from October 2020 to April 2021, and to identify possible factors associated with the persistence of illness, such as gender, age, previous medical conditions, and features of acute illness. A total of 318 HCWs who had become infected by COVID-19 were examined and interviewed approximately two months after their recovery from the infection. The clinical examinations were performed by Occupational Physicians in accordance with a specific protocol at the Occupational Medicine Unit of a tertiary hospital in Italy. The mean age of the participants was 45 years old, and $66.7\%$ of the workers were women while $33.3\%$ were men; the sample mainly consisted of nurses ($44.7\%$). During the medical examination, more than half of the workers mentioned that they had experienced multiple residual bouts of illness after the acute phase of infection. Men and women were similarly affected. The most reported symptom was fatigue ($32.1\%$), followed by musculoskeletal pain ($13.6\%$) and dyspnea ($13.2\%$). In the multivariate analysis, dyspnea ($p \leq 0.001$) and fatigue ($p \leq 0.001$) during the acute stage of illness and the presence of any limitation in working activities, in the context of fitness for a work evaluation performed while the occupational medicine surveillance program was being conducted ($$p \leq 0.025$$), were independently associated with any post-COVID-19 symptoms, which were considered final outcomes. The main post-COVID-19 symptoms—dyspnea, fatigue, and musculoskeletal pain—showed significant associations with dyspnea, fatigue, and musculoskeletal pain experienced during the acute stage of infection, with the presence of limitations in working activities, and pre-existing pneumological diseases. A normal weight according to body mass index was a protective factor. The identification of vulnerable workers as those with limitations in working activities, pneumological diseases, a high BMI, and of an older age and the implementation of preventive measures are key factors for preserving Occupational Health. Fitness-to-work evaluations performed by Occupational Physicians can be considered a complex index of overall health and functionality that can identify workers who may suffer from relevant post-COVID-19 symptoms.
## 1. Introduction
Three years after the outbreak of the COVID-19 pandemic in January 2020, SARS-CoV-2 virus remains a highly diffusive pathogen that is continuing to spread globally. On 30 January 2023, according to the World Health Organization (WHO) [1], 100,000 cases were recorded within 24 h, amounting to a total number of cumulative cases of over 750 million and constituting over 6.8 million deaths from the beginning of the pandemic. In Italy, in the last week of January 2023 from 27 January to 2 February 2023, 33,042 new cases were reported by the National Health Authorities along with 439 deaths [2]. New variants of interest and concern have emerged and spread [3], such as the Omicron variant and its sublineages [4]. Massive vaccination campaigns have reduced the burden of the virus on healthcare systems worldwide, consistently protecting people against severe forms of illness [5].
COVID-19 illness is highly variable, ranging from infection with no symptoms to pneumonia and life-threatening consequences. Common symptoms are fever, cough, musculoskeletal pain, myalgia, arthralgia, headache, fatigue, and dyspnea [6]. After acute infection symptoms emerge, $10\%$ to $20\%$ of infected patients (and up to $45\%$ according to a recent systematic review [7]) can experience a variety of mid- and long-term effects after they recover from their initial illness. Post COVID-19 condition, also known as “long COVID,” refers collectively to the constellation of long-term symptoms that some people experience after they have had COVID-19 [8]. The European Society of Clinical Microbiology and Infectious Disease (ESCMID) defines “Long COVID” symptoms as those persisting even 12 weeks post-infection and “Post-Acute COVID” symptoms as those persisting between 4 to 12 weeks [9]. Alternatively, according to NICE (National Institute for Health and Care Excellence, UK) guidelines, the term long COVID refers to signs and symptoms that continue after the acute stage of COVID-19 disease (4–12 weeks), and the term post-COVID-19 condition (PCC) refers to signs and symptoms that develop during or after infection with COVID-19 disease that continue for more than 12 weeks and cannot be explained by an alternative diagnosis [10,11]. *In* general, we can refer to the sequalae occurring after the acute stage of infection as “post-COVID”. Long COVID has been recognized as a clinical entity that may cause significant disability in not only hospitalized patients but also asymptomatic or mildly symptomatic ones [12,13,14,15]. The most commonly observed symptoms among patients with long COVID are fatigue, dyspnea, musculoskeletal pain, anosmia/dysgeusia, cognitive impairment (or brain fog), sleep disturbances, cough, and chest pain [9]. Older age, female sex, and pre-existing medical conditions (pulmonary diseases, psychiatric illness, obesity, and diabetes) have been found as possible risk factors for the development of Long COVID-19 syndrome [16,17,18]. In addition, the characteristics of the acute form of illness may have an influence on the development of post-COVID-19 symptoms [19]. The acquirement of further data is necessary to better define the clinical features of long COVID and identify strategies for its clinical management, as indicated by ESCMID and several other research authorities around the globe [20]. Omicron infection seems to present a lower prevalence of post-COVID-19 illness than previous variants, as it was prevalent in $4.5\%$ vs. $10.8\%$ of patients in a UK study [21], but other studies have reported higher percentages [22]. As the number of COVID-19 cases and survivors grows, the burden of post-COVID-19 illness will also increase, becoming a more relevant concern for the healthcare systems [10]. Vaccination could reduce the risk of long COVID, but further studies are needed to confirm this [23,24].
Post-COVID-19 disorders are relevant in the field of occupational medicine due to the fact that Healthcare Workers (HCWs) are continuously exposed to the risk of SARS-CoV-2 infection and its consequences. SARS-CoV-2 infection and its management by healthcare management systems remains a matter of significant concern. Omicron, the current SARS-CoV-2 variant, which includes BA.1, BA.2, BA.3, BA.4, BA.5, and descendent lineages, is much more contagious than its predecessors [25,26], which means that during surges, without an adequate level of preparedness, hospitals can become understaffed, thereby stressing and overburdening healthcare workers to a greater degree [27]. Healthcare workers are defined as workers who deliver care and services to the sick and ailing either directly (e.g., physicians, nurses, healthcare assistants, midwifes, etc.) or indirectly (e.g., aides, laboratory technicians, porters, etc.) [ 28]. In terms of occupational safety and health, healthcare workers are exposed to relevant biological risks, which include SARS-CoV-2 infection. In Italy, infection with SARS-CoV-2 is considered an occupational illness/accident for some categories of workers, primarily HCWs, and healthcare managers are required to adopt every necessary measure to reduce the spread of the virus to preserve workers’ and patients’ health. HCWs constitute a high-risk population that presents a 24-fold higher probability of contracting the infection than the general population [29]. HCWs’ exposure to occupational risks, including biological risks posed by SARS-CoV-2, can have different impacts on workers’ health according to their age, sex, and history of medical and psychological conditions, and occupational medicine is required to carefully evaluate such impacts to preserve occupational wellbeing. According to the literature, women and older patients may be more affected by post-COVID-19 symptoms, and some previous medical conditions can be risk factors, such as obesity and pulmonary diseases [23,30]. In a recent metanalysis analyzing 13,340 patients, female sex was associated with long COVID-19 affliction with an OR of 1.52 [31]. Post-COVID-19 sequelae can reduce functionality and work-related ability for some time, thus impacting work efficiency. This can be also a major issue in light of the highly physically and emotionally demanding tasks performed by HCWs [32]. The impact of persistent illness after being afflicted with the acute form of illness can also have negative effects on the broader healthcare delivery system, leading to a possible loss of skilled healthcare personnel due to post-COVID-19-related disabilities [33].
## Aims of the Study
The aim of this research is to provide observational data regarding post-COVID-19 outcomes in a population of HCWs infected by COVID-19 from October 2020 to March 2021. Through a cross-sectional design, herein, we present data regarding the physical health of workers two months after COVID-19 infection. We tested whether the baseline characteristics of the patients—age, sex, previous health conditions, and BMI—and the main symptoms of the acute form of the illness affected their health outcomes two months after COVID-19 infection. In addition, we hypothesized that fitness to work, which in the field of occupational medicine is the assessment of the possibility of performing a specific task, may be associated with post-COVID-19 symptoms.
## 2. Materials and Methods
Between 1 October 2020 and 30 April 2021 in a tertiary referral hospital employing approximately 5000 HCWs in Florence, Tuscany, Italy, 440 healthcare workers were infected with COVID-19 and tested positive for SARS-CoV-2 infection via a RT-PCR nasopharyngeal swab (Real-Time Polymerase Chain Reaction). After the acute phase of the illness, all the workers were contacted and invited to a post-COVID-19 return-to-work visit at the Occupational Medicine Unit. Of the 440 workers that tested positive in the considered period, 318 accepted the invitation for the medical examination, constituting a participation rate of $72.3\%$.
## 2.1. Definition of Healthcare Worker
We considered healthcare workers as all workers involved in the care of patients and divided them into physicians (structured doctors), nurses, healthcare assistants (workers in support of nurses and doctors, who assist patients with daily personal hygiene activities and carry out simple activities to aid nursing and technical healthcare activities), resident physicians (doctors in training), and others (radiology technicians, lab technicians, physiotherapists, midwifes, and patient transport workers).
## 2.2. Definition of COVID-19 Case
A COVID-19 case was defined via a positive RT- PCR nasopharyngeal swab, with or without the presence of COVID-19-related symptoms. The protocol of surveillance required a negative PCR swab to return to work according to the National Italian Laws for the management of the pandemic. In particular, at that time, the Health Ministry required infected individuals to undergo at-home isolation for at least 10 days after the onset of symptoms and to achieve a negative RT-PCR test at the end of the isolation period, in addition to attaining remission with respect to all COVID-19 symptoms except for alterations in smell and taste [34]. After the first positive swab, the following control swabs were programmed by the Occupational Medicine Unit after 10 days, and if they were still positive, subsequent swabs were scheduled once a week until the patient tested negative.
## 2.3. Pre-Existing Medical Conditions, Fitness to Work, and Vaccination
The patients’ pre-existing medical conditions were investigated by inquiring as to their medical histories during the medical examinations and were cross-checked with the occupational medical records available. The main pre-existing conditions uncovered were cardiovascular diseases (e.g., hypertension, cardiopathy of any kind, etc.), pneumological disease (e.g., asthma, COPD of any severity, etc.), neurological disease (e.g., epilepsy, neuropathy, etc.), psychiatric disorders (e.g., depression, anxiety requiring medical support, etc.), diabetes (type I or II), and endocrinological disorders (e.g., thyroid disorders). Diseases of any other form (e.g., breast cancer) were defined as “other disorder ”. Data regarding fitness to work, which were provided by occupational physicians, were collected retrospectively while analyzing the available medical records. The data referred to the last visit of routine medical surveillance wherein an Occupational Physician expressed the judgement for a patient’s suitability for a specific job (e.g., physician, nurse, healthcare assistant, etc.), as required by Italian national laws. A worker is judged as “completely fit to work” if his/her medical conditions allow him/her to completely perform an occupational task; thus, the clinical conditions of such a worker are completely suitable for said occupational task. Alternatively, the suitability to work can include specific limitations (e.g., the avoidance of nocturnal shifts, limitations on the amount of weight that can be handled, etc.) when required by the health status of the worker. This means that the clinical conditions of the worker allow him/her to perform the specific occupational task but with some limitations (e.g., for a healthcare assistant with back issues, a possible limitation could be that he/she cannot move patients alone; alternatively, for a nurse with decompensated type II diabetes, the limitation could be not working during night shifts). Therefore, the fitness-to-work evaluation is a comprehensive evaluation that accounts for all the clinical conditions of workers and relates them to a specific occupational task. This study took place before the rollout of the COVID-19 vaccine campaign in Italy and during the first few months when only mRNA vaccines were available. Full vaccination is defined as having received two shots of vaccine, according to the specific schedule, followed by a period of two weeks from the second dose.
## 2.4. Acute Symptoms and Post-COVID-19 Symptoms
In this study, post-COVID-19 symptoms are defined as symptoms that persist after having tested negative via RT-PCR nasopharyngeal swab and persisting at the time of a medical examination, i.e., between 4 to 8 weeks after recovery. Specific post-COVID-19 symptoms explored during the collection of data were cough, dyspnea, fatigue, headache, sleep disturbances, alteration in smell or taste, dizziness, and musculoskeletal pain. Other symptoms were defined as other forms of illness. Fatigue was defined as a state of physical asthenia that impacted typical activities to some extent. During the medical interview, workers were asked to report newly occurring or persistent symptoms after COVID-19 infection, excluding chronic issues. Workers were thus classified as suffering from any post-COVID-19 symptom when said symptom was described as having emerged during or immediately following COVID-19 infection. The information about acute and post-COVID-19 symptoms was obtained during the medical examination through an accurate recording of medical history. Notably, at the time the study was conducted, no Long/Post COVID-19 definition had officially been coined by Health Authorities or Medical Societies. Post-COVID-19 symptoms were the final outcome of the examination and concerned the presence (yes/no) of symptoms after the end of the acute stage of infection.
## 2.5. Medical Examination and Collection of Data
The medical examination occurred approximately 4 to 8 weeks following the patients’ recovery. The physicians of the Occupational Medicine Unit—who also followed the acute phase of infection through a telephonic surveillance—examined the workers and collected data about sociodemographic characteristics, previous contact with COVID-19-infected people, symptoms and length of infection, presence of post-COVID-19 symptoms, and previous medical history. In addition, a clinical examination was performed by updating the health history records, checking vital signs (e.g., blood pressure, heart rate, and respiratory rate), performing visual and physical exams, and prescribing, according to medical opinion, laboratory tests or specialist investigations. The examination was entirely carried out by specialized physicians, and the findings were reported on a specific file for each patient (see Supplementary Material for the complete form).
During the examination, a screening for post-traumatic stress disorder was performed using the IES-6 scale (Impact of Event Scale 6) [35], and questionnaires about COVID-19-related fear were administered; these psychological aspects will be addressed in a future paper. The data collection procedure was performed within a health surveillance program according to Italian Legislative Decree $\frac{81}{2008.}$ The sensitive data were collected anonymously, in accordance with the principles of the Declaration of Helsinki, and all participants gave their informed consent.
## 2.6. Statistical Analysis
IBM SPSS (Statistical Package for Social Sciences, version 29.0—IBM Corp. Armonk, NY, USA) was used to perform the statistical analysis. Univariate analysis was performed using a two-tailed chi-square test to compare the distribution of nominal data, while significative results for more than two groups were analyzed with standard residuals post hoc test. Student’s t test (or Mann Whitney U test if the distribution of the continuous variables was not normal) was used to compare the means between two groups.
The main outcomes were post-COVID-19 symptoms, which were analyzed as dichotomous dependent variables (e.g., fatigue: yes/no; dyspnea: yes/no; etc.). Firstly, the analysis was carried out for the whole sample; then, the sample was segregated by gender to analyze the differences between men and women. Afterwards, age was taken into account as a continuous variable, and the difference in the mean age between the two groups (presence/absence of a post-COVID-19 symptom) was analyzed through Student’s t test. Eventually, other interesting baseline variables (BMI, previous medical conditions, fitness to work, and symptoms and characteristics of the acute illness) were considered and accounted for as independent variables in the chi-square test. Correlation between continuous variables was tested with Pearson’s r or Kendall’s tau in cases of non-parametric distribution. Multiple logistic hierarchical regression including the statistically significative variables in the univariate analysis was used to perform multivariate analysis. Given the response rate, namely, 318 out of a total of 440 workers who had become infected during the study period, the sample obtained was judged to be adequate, and the significance level was set at 0.05.
## 3.1. Baseline Characteristics of the Sample and the Acute Form of Infection
The data were initially analyzed as a whole sample and according to age and sex. Table S1 of the Supplementary Materials contains the baseline characteristics of the sample and their segregation according to sex and age. The participants had a mean age of 45 years old (±11.9) and comprised 212 women ($66.7\%$) and 106 men ($33.3\%$). The participants presented generally good health, as limitations in work activities were present in only $13\%$ of the workers and were mainly connected to tasks involving manual labor. In addition, over two thirds of the sample were not on any medication at the time of the medical examination. The sample was primarily composed of nurses (142, $44.7\%$), followed by healthcare assistants (68, $21.4\%$), physicians (37, $11.6\%$), resident physicians (37, $11.6\%$), and others (34, $10.7\%$). Regarding the age distribution between the professional groups, the groups nurses and resident physicians show a lower mean age than the other groups. Generally, participants with a higher mean age show a higher burden of diseases, such as cardiovascular issues or diabetes, and take more drugs chronically. This decrease in general health through aging is also evidenced by the significative difference in the mean age between the completely fit-to-work participants and the workers with any kind of work limitation. Regarding the differences related to sex, women presented more limitations in work activities, although this is not reflected by a higher level of pre-existent medical conditions (apart from endocrinological problems). Among the male workers, there was a larger proportion of overweight workers compared to the female group, and when considering the whole sample, $11\%$ [35] of the workers were considered obese according to their BMI (Body Mass Index). A total of $9.7\%$ of the sample [31] was full vaccinated. The data regarding the main characteristics of the acute form of infection can be found in Table S2 of the Supplementary Material. Notably, only 10 workers were admitted to a hospital during the acute stage of illness.
## 3.2. Post-COVID-19 Symptoms
More than half of the workers ($56.3\%$) mentioned post-COVID-19 illness, with most of these cases including more than one symptom. The most reported symptom was fatigue ($32.1\%$), followed by musculoskeletal pain ($13.6\%$) and dyspnea ($13.2\%$). As determined through the univariate analysis, older workers are more likely to experience post-COVID-19 symptoms, especially in terms of musculoskeletal pain ($p \leq 0.001$), dyspnea ($$p \leq 0.030$$), and fatigue ($$p \leq 0.011$$). No difference was found between the male and female workers. Notably, in the acute phase of illness, there was no difference in the mean age of the group with musculoskeletal pain or fatigue and the mean age without these symptoms, while the difference in the mean age was significant when considering long-term illness. Table 1 contains the variables related to the analyzed post-COVID-19 symptoms according to sex and age.
## 3.3. Correlation between Most Common Post-COVID-19 Symptoms (Dyspnoea, Fatigue, and Musculoskeletal Pain) and Sample Characteristics
A univariate analysis was conducted to explore the relationships between the three most common post-COVID-19 symptoms (dyspnea, fatigue, and musculoskeletal pain) and the main baseline characteristics and features of the acute form of illness. The sample was divided according to the presence of any symptoms during the post-COVID-19 period and with respect to the three most common symptoms. The significative correlations are reported in Table 2. The only previous health conditions that showed a significative relationship with the post-COVID-19 symptoms were pneumological and psychiatric diseases, limitations in work activities, and BMI (considered as normal weight). Almost all symptoms and the use of medications during the acute phase of the disease show a significative relationship between the considered post-COVID-19 symptoms. The mean number of symptoms reported in the acute phase of infection was higher in the patients with the analyzed post-COVID-19 disorders. The complete results are reported in Table S3 of the Supplementary material.
## 3.4. Multivariate Analysis for Post-COVID-19 Symptoms
The variables that were found to be statistically significative in the univariate analysis were selected to be tested via multivariate analysis. Gender, as it was not found to be statistically significative, was excluded in the logistic regression. In order to analyze which variables were independently associated with post-COVID-19 disorders, four hierarchical multiple logistic regressions were performed, with each one having the following respective variables: the dependent variable, (a), the presence of any post-COVID-19 symptom; (b) post-COVID-19 dyspnea; (c) post-COVID-19 fatigue; and (d) post-COVID-19 musculoskeletal pain. Each regression consisted of two blocks. The first block included dyspnea, fatigue, and musculoskeletal pain during the acute phase of illness as predictors for (a), and age, number of acute symptoms, and the presence of dyspnea, fatigue, and musculoskeletal pain during the acute phase of illness for (b), (c), and (d), respectively. In block two, other predictors were included in the model, which were found to be statistically significative in the univariate analysis: for (a), (c), and (d), these predictors were normal weight according to BMI and limitations in work activities, while for (b), they included limitations in work activities and pre-existing pneumological diseases. The results of the logistic regression are reported in Table 3, and the details about the model and blocks tested are reported in the Supplementary Materials.
## 4. Discussion
After almost 3 years of the pandemic, COVID-19 infection and its long consequences are still relevant, especially in specific occupational contexts such as healthcare, whose workers are exposed to several occupational risks and are often overburdened. It is widely known that COVID-19 infection can result in several different ailments, denoted as post-COVID-19 symptoms or post-COVID-19 syndrome [8,36], which may last from several weeks to years [15]. These conditions challenge healthcare professionals’ return to work [37] and can transform an infection that usually lasts 1 to 3 weeks into a long, complex illness. Post-COVID-19 symptoms can affect patients with all levels of disease severity as well as young, healthy people [38,39].
This study aimed to describe the clinical post-COVID-19 outcomes of a very specific population, namely, healthcare workers who had become infected with COVID-19, to identify possible predictors of post-COVID-19 disorders. In our sample, the workers who presented several residual symptoms after having tested negative via a control swab yielded relevant results: over half of the participants mentioned at least one symptom persisting after having been declared cured and readmitted to work, which is consistent with previous works [7,30,40]. This should be also analyzed while considering the composition of our sample, that is, a working population with a mean age of 45.
The most reported disorder was fatigue, followed by dyspnea and musculoskeletal pain. This is consistent with several previous studies on COVID-19 patients’ outcomes at three months after infection [38,41,42]. For this reason, we conducted our main analysis on these three symptoms, judging them as the most relevant for the wellbeing of workers. Patients complaining of fatigue, musculoskeletal pain, and dyspnea showed a higher mean age, which is an interesting finding when considering that during the acute phase of illness, no significative difference in symptoms was reported according to either sex age. It is known that elderly patients may suffer more acutely from post-COVID-19 symptoms than younger people [43]. However, when a multivariate analysis was performed, age remained a significant factor only with respect to musculoskeletal pain. This can be explained by the fact that our sample has a relatively low average age, which may not be significative when other relevant factors are considered in the analysis.
Interestingly, no difference was found in the post-COVID-19 conditions regarding sex, and this is consistent with the findings of Petersen et al., who addressed post-COVID-19 in mild-course patients [44]. In the literature, however, female sex is considered one of the risk factors for developing post-COVID-19 disorders [45]. For example, in a review analyzing post-COVID-19 symptoms in mild-course COVID-19 patients, it was found that most common post-COVID-19 disorders occur among women (on average $60\%$) [46]. In a large population study of almost 400,000 patients, the risk factors for long COVID-19 included female sex [47]. In our sample, female and male workers had the same mean age, and when considering comorbidities, the two groups are very similar. This difference from the previous reported data may be related to the small size of our sample but also the different methodology with respect to the collection of data. In fact, for the most part, studies addressing post-COVID-19 disorders use self-administrated questionnaires or generic medical records. Instead, in this study, a specific post-COVID-19 medical evaluation was used to collect the data, and this could have led to the more precise reporting of symptoms, which was aided by medical personnel. In addition, this can be related to the relatively young age of our sample, which includes active, working people.
Although post-COVID-19 syndrome can also affect patients with a mild form of infection, the clinical characteristics of the acute form of infection and its severity are strictly related to the presence and severity of post-COVID-19 disorders; this relationship has already been demonstrated in previous large studies [19,36], and it was also found in our analysis. Dyspnea and fatigue during the acute phase of illness independently predicted the presence of post-COVID-19 disorders, while for each main post-COVID-19 symptom, having suffered from the same symptom during the acute phase of illness predicted its presence after recovery. For post-COVID-19 fatigue and musculoskeletal pain, the number of acute symptoms also showed a statistically significant relationship. Alternatively, for post-COVID-19 dyspnea, this relationship was not relevant. Instead, a history of pneumological diseases, such as asthma or COPD (chronic obstructive pulmonary disease), was independently related to post-COVID-19 dyspnea. In the literature, a wide range of comorbidities present at baseline assessment were associated with an increased risk of post-COVID-19 symptoms [38]; for example, associations with pre-existent medical conditions and post-COVID-19 syndrome were found for asthma and hypertension [36], and for COPD, for anxiety and depression [47]. In our sample, only post-COVID-19 dyspnea was associated with pre-existing pneumological diseases, while no other significant relationship was found with other comorbidities, which was probably due to the relatively good health of our population. Via multivariate analysis, normal weight according to BMI was found as an independent protective factor for post-COVID-19 fatigue, and this is consistent with previous literature data. High BMIs and obesity are recognized risk factors for the development of post-COVID-19 symptoms [38,47,48]. A possible explanation for this association may lie in the high levels of inflammation present in obese patients [49], which also constitutes a pathophysiological basis with which to explain the higher risk of mortality for obese patients during the acute stage of infection [50]. Remarkably, limitations in work activities constituted another factor independently linked to post-COVID-19 symptoms, particularly dyspnea and fatigue. To the best of our knowledge, this is the first study that provides a correlation between this kind of index and post-COVID-19 symptoms. According to Italian laws, occupational physicians can ascribe limitations in work activities to workers who cannot completely perform a given job for medical reasons. Therefore, fitness to work can be considered a complex index of overall health and functionality, explaining its correlation with post-COVID-19 symptoms. At the same time, post-COVID-19 symptoms can affect the fitness to work, as demonstrated by an Italian study conducted on hospitalized COVID-19-infected HCWs, where, during return-to-work examinations, fit-to-work judgements with restrictions increased from $31.4\%$ to $58.7\%$ [51].
## 4.1. Limitations and Strengths
This research has several limitations. This study’s cross-sectional design renders it incapable of inferring causal relations between the analyzed variables and determining a precise duration of post-COVID-19 symptoms. This study did not include controls or a longitudinal follow-up, nor an assessment of the participants’ health statuses before the infection; therefore, it serves an observational purpose and does not provide direct cause–effect relationships between COVID-19 infection and the symptoms that develop afterward. Moreover, this study did not provide a scale regarding the intensity of symptoms both during the acute illness and post-COVID-19 phases. Post-COVID-19 symptoms were explored through the recording of medical history by specialized medical personnel and based on information regarding disorders that was reported by the workers. Considering the complexity of post-COVID-19 syndrome and its associated factors, longitudinal studies are needed to explore the syndrome’s pathogenesis and clinical management. In this study, the potential bias conferred by non-respondents (those who refused to participate) should be considered; nevertheless, the response rate was acceptable. In addition, the sample was limited to a specific working population; thus, it may be not fully representative of the whole population. The relatively small sample may have affected the power of the statistical analysis. In addition, $90\%$ of the patients were unvaccinated at the time of the study; therefore, the study does not consider the effect of widespread vaccination on post-COVID-19 outcomes. Lastly, our study relates to physical symptoms and does not investigate the psychological aspects of the disease and its sequelae, thus allowing us to reach only partial conclusions as to how a SARS-CoV-2 infection can impact health, which is defined by the WHO as psycho-physical status and well-being.
However, this study has a major point of strength: all the data were collected by physicians during a medical examination. In our opinion, this rigorous methodology adds soundness to the findings. However, this method might have led to information collection bias due to the fact that the assessment was based on workers’ reports, and some symptoms, especially fatigue, can be influenced by several external situations. In addition, as stated above, this is the first study that correlates fitness to work and post-COVID-19 symptoms.
## 4.2. Practical Implications
The experience of dealing with COVID-19 in workplaces has provided a model with which to address future infectious disease emergencies, such as surges of monkeypox or seasonal flu [52]. First of all, the identification of vulnerable groups of workers has always been a goal of occupational medicine. Our study showed that the functional evaluations conducted by occupational physicians can identify workers who may suffer from relevant post-COVID-19 symptoms. This aspect should be considered when assigning workers to certain high-risk wards, not only with respect to COVID-19 risk but, more generally, biological risk as well. BMI remains one of the modifiable risk factors for post-COVID-19 syndrome, and workplace health promotion can play an important role in reducing obesity. For example, matching an occupational examination with a clinical nutrition check-up may educate workers and, in the long term, reduce the rates of obesity among this demographic. Lastly, as acute symptoms are associated with post-COVID-19 disorders, and in mild–moderate forms as well, monitoring such patients after their recovery and return to work may be useful for addressing and managing upcoming post-COVID-19 symptoms, both from occupational and general medicine perspectives.
## 5. Conclusions
The spread of SARS-CoV-2 and its variants will continue for years [53], and, in healthcare settings, the management of the biological risk related to SARS-CoV-2 has become and will continue to be a major issue for occupational health and safety services. Post-COVID-19 symptoms, which are mostly mild to moderate, are frequent among healthcare workers infected with SARS-CoV-2, especially with respect to fatigue, dyspnea, and musculoskeletal pain, as are the relevant psychological aspects, which are not of secondary importance. Acute symptoms, BMI, and pneumological comorbidities showed significant correlations with the main post-COVID-19 symptoms, thus leading to the identification of potentially vulnerable groups of workers. Fitness-to-work evaluations, performed in the context of occupational medicine through a comprehensive and functional evaluation, were related to several post-COVID-19 symptoms, mainly dyspnea and fatigue, and may predispose high-risk workers to post-COVID symptoms after the infection.
## References
1. **WHO Coronavirus (COVID-19) Dashboard**
2. Salute M.. **Della Monitoraggio Settimanale COVID-19, Report 23–29 Gennaio 2023**
3. **Update on Omicron**
4. Kamble P., Daulatabad V., Patil R., John N.A., John J.. **Omicron Variant in COVID-19 Current Pandemic: A Reason for Apprehension**. *Horm. Mol. Biol. Clin. Investig.* (2022.0). DOI: 10.1515/hmbci-2022-0010
5. Ghazvini K., Karbalaei M., Keikha M.. **Third Booster Vaccination and Stopping the Omicron, a New Variant of Concern**. *Vacunas* (2022.0) **23** S103-S110. DOI: 10.1016/j.vacun.2022.05.005
6. Struyf T., Deeks J.J., Dinnes J., Takwoingi Y., Davenport C., Leeflang M.M., Spijker R., Hooft L., Emperador D., Domen J.. **Signs and Symptoms to Determine If a Patient Presenting in Primary Care or Hospital Outpatient Settings Has COVID-19**. *Cochrane Database Syst. Rev.* (2022.0) **2022** CD013665. DOI: 10.1002/14651858.CD013665.pub3
7. O’Mahoney L.L., Routen A., Gillies C., Ekezie W., Welford A., Zhang A., Karamchandani U., Simms-Williams N., Cassambai S., Ardavani A.. **The Prevalence and Long-Term Health Effects of Long Covid among Hospitalised and Non-Hospitalised Populations: A Systematic Review and Meta-Analysis**. *eClinicalMedicine* (2023.0) **55** 101762. DOI: 10.1016/j.eclinm.2022.101762
8. **Coronavirus Disease (COVID-19): Post COVID-19 Condition**
9. Yelin D., Moschopoulos C.D., Margalit I., Gkrania-Klotsas E., Landi F., Stahl J.-P., Yahav D.. **ESCMID Rapid Guidelines for Assessment and Management of Long COVID**. *Clin. Microbiol. Infect. Off. Publ. Eur. Soc. Clin. Microbiol. Infect. Dis.* (2022.0) **28** 955-972. DOI: 10.1016/j.cmi.2022.02.018
10. Erinoso O.. **Post-COVID-19 Condition: Current Evidence and Unanswered Questions**. *Lancet Glob. Health* (2022.0) **10** e1210-e1211. DOI: 10.1016/S2214-109X(22)00323-0
11. **Overview | COVID-19 Rapid Guideline: Managing the Long-Term Effects of COVID-19 | Guidance | NICE**
12. Rao S., Amara V., Chaudhuri S., Rao B.K., Todur P.. **“Post-COVID-19 Syndrome:” The New Pandemic Affecting Healthcare Workers and How the Frontline Warriors Are Battling It**. *Indian J. Palliat. Care* (2021.0) **27** 313-318. DOI: 10.25259/IJPC_160_21
13. Vance H., Maslach A., Stoneman E., Harmes K., Ransom A., Seagly K., Furst W.. **Addressing Post-COVID Symptoms: A Guide for Primary Care Physicians**. *J. Am. Board Fam. Med. JABFM* (2021.0) **34** 1229-1242. DOI: 10.3122/jabfm.2021.06.210254
14. Goërtz Y.M.J., Van Herck M., Delbressine J.M., Vaes A.W., Meys R., Machado F.V.C., Houben-Wilke S., Burtin C., Posthuma R., Franssen F.M.E.. **Persistent Symptoms 3 | Months after a SARS-CoV-2 Infection: The Post-COVID-19 Syndrome?**. *ERJ Open Res.* (2020.0) **6** 00542-02020. DOI: 10.1183/23120541.00542-2020
15. **CDC Post-COVID Conditions**
16. Notarte K.I., de Oliveira M.H.S., Peligro P.J., Velasco J.V., Macaranas I., Ver A.T., Pangilinan F.C., Pastrana A., Goldrich N., Kavteladze D.. **Age, Sex and Previous Comorbidities as Risk Factors Not Associated with SARS-CoV-2 Infection for Long COVID-19: A Systematic Review and Meta-Analysis**. *J. Clin. Med.* (2022.0) **11**. DOI: 10.3390/jcm11247314
17. Nalbandian A., Desai A.D., Wan E.Y.. **Post-COVID-19 Condition**. *Annu. Rev. Med.* (2023.0) **74** 55-64. DOI: 10.1146/annurev-med-043021-030635
18. Thompson E.J., Williams D.M., Walker A.J., Mitchell R.E., Niedzwiedz C.L., Yang T.C., Huggins C.F., Kwong A.S.F., Silverwood R.J., Di Gessa G.. **Long COVID Burden and Risk Factors in 10 UK Longitudinal Studies and Electronic Health Records**. *Nat. Commun.* (2022.0) **13** 3528. DOI: 10.1038/s41467-022-30836-0
19. Dirican E., Bal T.. **COVID-19 Disease Severity to Predict Persistent Symptoms: A Systematic Review and Meta-Analysis**. *Prim. Health Care Res. Dev.* (2022.0) **23** e69. DOI: 10.1017/S1463423622000585
20. Powell M.. **Researching Long COVID: A New Global Health Challenge**
21. Antonelli M., Pujol J.C., Spector T.D., Ourselin S., Steves C.J.. **Risk of Long COVID Associated with Delta versus Omicron Variants of SARS-CoV-2**. *Lancet* (2022.0) **399** 2263-2264. DOI: 10.1016/S0140-6736(22)00941-2
22. Arjun M.C., Singh A.K., Roy P., Ravichandran M., Mandal S., Pal D., Das K., Alekhya G., Venkateshan M., Mishra B.. **Long COVID Following Omicron Wave in Eastern India—A Retrospective Cohort Study**. *J. Med. Virol.* (2022.0) **95** e28214. DOI: 10.1002/jmv.28214
23. Notarte K.I., Catahay J.A., Velasco J.V., Pastrana A., Ver A.T., Pangilinan F.C., Peligro P.J., Casimiro M., Guerrero J.J., Gellaco M.M.L.. **Impact of COVID-19 Vaccination on the Risk of Developing Long-COVID and on Existing Long-COVID Symptoms: A Systematic Review**. *EClinicalMedicine* (2022.0) **53** 101624. DOI: 10.1016/j.eclinm.2022.101624
24. Gao P., Liu J., Liu M.. **Effect of COVID-19 Vaccines on Reducing the Risk of Long COVID in the Real World: A Systematic Review and Meta-Analysis**. *Int. J. Environ. Res. Public. Health* (2022.0) **19**. DOI: 10.3390/ijerph191912422
25. Chen J., Wang R., Gilby N.B., Wei G.-W.. **Omicron Variant (B.1.1.529): Infectivity, Vaccine Breakthrough, and Antibody Resistance**. *J. Chem. Inf. Model.* (2022.0) **62** 412-422. DOI: 10.1021/acs.jcim.1c01451
26. Callaway E., Ledford H.. **How Bad Is Omicron? What Scientists Know so Far**. *Nature* (2021.0) **600** 197-199. DOI: 10.1038/d41586-021-03614-z
27. Pan S.-J., Qian W.-Y., Yang Y.-P., Zhang M.-X., Hu X.-M., Chen H.-X., Tung T.-H.. **Evaluation of Burnout among Stay-behind Healthcare Workers during the Current Omicron Wave of COVID-19 in Taizhou, China**. *Front. Psychiatry* (2022.0) **13** 1022881. DOI: 10.3389/fpsyt.2022.1022881
28. Joseph B., Joseph M.. **The Health of the Healthcare Workers**. *Indian J. Occup. Environ. Med.* (2016.0) **20** 71-72. DOI: 10.4103/0019-5278.197518
29. Nguyen L.H., Drew D.A., Graham M.S., Joshi A.D., Guo C.-G., Ma W., Mehta R.S., Warner E.T., Sikavi D.R., Lo C.-H.. **Risk of COVID-19 among Front-Line Health-Care Workers and the General Community: A Prospective Cohort Study**. *Lancet Public Health* (2020.0) **5** e475-e483. DOI: 10.1016/S2468-2667(20)30164-X
30. Nittas V., Gao M., West E.A., Ballouz T., Menges D., Wulf Hanson S., Puhan M.A.. **Long COVID through a Public Health Lens: An Umbrella Review**. *Public Health Rev.* (2022.0) **43** 1604501. DOI: 10.3389/phrs.2022.1604501
31. Maglietta G., Diodati F., Puntoni M., Lazzarelli S., Marcomini B., Patrizi L., Caminiti C.. **Prognostic Factors for Post-COVID-19 Syndrome: A Systematic Review and Meta-Analysis**. *J. Clin. Med.* (2022.0) **11**. DOI: 10.3390/jcm11061541
32. Lulli L.G., Giorgi G., Pandolfi C., Foti G., Finstad G.L., Arcangeli G., Mucci N.. **Identifying Psychosocial Risks and Protective Measures for Workers’ Mental Wellbeing at the Time of COVID-19: A Narrative Review**. *Sustainability* (2021.0) **13**. DOI: 10.3390/su132413869
33. Praschan N., Josephy-Hernandez S., Kim D.D., Kritzer M.D., Mukerji S., Newhouse A., Pasinski M., Chemali Z.. **Implications of COVID-19 Sequelae for Health-Care Personnel**. *Lancet Respir. Med.* (2021.0) **9** 230-231. DOI: 10.1016/S2213-2600(20)30575-0
34. Salute M.. **Della Norme, Circolari e Ordinanze**
35. Hosey M.M., Leoutsakos J.-M.S., Li X., Dinglas V.D., Bienvenu O.J., Parker A.M., Hopkins R.O., Needham D.M., Neufeld K.J.. **Screening for Posttraumatic Stress Disorder in ARDS Survivors: Validation of the Impact of Event Scale-6 (IES-6)**. *Crit. Care Lond. Engl.* (2019.0) **23** 276. DOI: 10.1186/s13054-019-2553-z
36. Shukla A.K., Atal S., Banerjee A., Jhaj R., Balakrishnan S., Chugh P.K., Xavier D., Faruqui A., Singh A., Raveendran R.. **An Observational Multi-Centric COVID-19 Sequelae Study among Health Care Workers**. *Lancet Reg. Health Southeast Asia* (2023.0) **10** 100129. DOI: 10.1016/j.lansea.2022.100129
37. Grazzini M., Lulli L.G., Mucci N., Paolini D., Baldassarre A., Gallinoro V., Chiarelli A., Niccolini F., Arcangeli G.. **Return to Work of Healthcare Workers after SARS-CoV-2 Infection: Determinants of Physical and Mental Health**. *Int. J. Environ. Res. Public. Health* (2022.0) **19**. DOI: 10.3390/ijerph19116811
38. Kisiel M.A., Janols H., Nordqvist T., Bergquist J., Hagfeldt S., Malinovschi A., Svartengren M.. **Predictors of Post-COVID-19 and the Impact of Persistent Symptoms in Non-Hospitalized Patients 12 Months after COVID-19, with a Focus on Work Ability**. *Ups. J. Med. Sci.* (2022.0) **127**. DOI: 10.48101/ujms.v127.8794
39. Tenforde M.W.. **Symptom Duration and Risk Factors for Delayed Return to Usual Health among Outpatients with COVID-19 in a Multistate Health Care Systems Network—United States, March–June 2020**. *MMWR Morb. Mortal. Wkly. Rep.* (2020.0) **69**. DOI: 10.15585/mmwr.mm6930e1
40. Carvalho-Schneider C., Laurent E., Lemaignen A., Beaufils E., Bourbao-Tournois C., Laribi S., Flament T., Ferreira-Maldent N., Bruyère F., Stefic K.. **Follow-up of Adults with Noncritical COVID-19 Two Months after Symptom Onset**. *Clin. Microbiol. Infect.* (2021.0) **27** 258-263. DOI: 10.1016/j.cmi.2020.09.052
41. Shanbehzadeh S., Tavahomi M., Zanjari N., Ebrahimi-Takamjani I., Amiri-Arimi S.. **Physical and Mental Health Complications Post-COVID-19: Scoping Review**. *J. Psychosom. Res.* (2021.0) **147** 110525. DOI: 10.1016/j.jpsychores.2021.110525
42. Sarda R., Kumar A., Chandra A., Bir M., Kumar S., Soneja M., Sinha S., Wig N.. **Prevalence of Long COVID-19 and Its Impact on Quality of Life Among Outpatients with Mild COVID-19 Disease at Tertiary Care Center in North India**. *J. Patient Exp.* (2022.0) **9** 23743735221117360. DOI: 10.1177/23743735221117358
43. Cabrera Martimbianco A.L., Pacheco R.L., Bagattini Â.M., Riera R.. **Frequency, Signs and Symptoms, and Criteria Adopted for Long COVID-19: A Systematic Review**. *Int. J. Clin. Pract.* (2021.0) **75** e14357. DOI: 10.1111/ijcp.14357
44. Petersen M.S., Kristiansen M.F., Hanusson K.D., Danielsen M.E., á Steig B., Gaini S., Strøm M., Weihe P.. **Long COVID in the Faroe Islands: A Longitudinal Study among Nonhospitalized Patients**. *Clin. Infect. Dis.* (2021.0) **73** e4058-e4063. DOI: 10.1093/cid/ciaa1792
45. Sylvester S.V., Rusu R., Chan B., Bellows M., O’Keefe C., Nicholson S.. **Sex Differences in Sequelae from COVID-19 Infection and in Long COVID Syndrome: A Review**. *Curr. Med. Res. Opin.* (2022.0) **38** 1391-1399. DOI: 10.1080/03007995.2022.2081454
46. Malkova A., Kudryavtsev I., Starshinova A., Kudlay D., Zinchenko Y., Glushkova A., Yablonskiy P., Shoenfeld Y.. **Post COVID-19 Syndrome in Patients with Asymptomatic/Mild Form**. *Pathogens* (2021.0) **10**. DOI: 10.3390/pathogens10111408
47. Subramanian A., Nirantharakumar K., Hughes S., Myles P., Williams T., Gokhale K.M., Taverner T., Chandan J.S., Brown K., Simms-Williams N.. **Symptoms and Risk Factors for Long COVID in Non-Hospitalized Adults**. *Nat. Med.* (2022.0) **28** 1706-1714. DOI: 10.1038/s41591-022-01909-w
48. Uniyal N., Sethi Y., Sharma P.C., Sayana A., Jeet N., Agarwal A., Rawat V.. **Post-COVID Syndrome and Severity of COVID-19: A Cross-Sectional Epidemiological Evaluation from North India**. *Cureus* (2022.0) **14** e27345. DOI: 10.7759/cureus.27345
49. Florencio L.L., Fernández-de-las-Peñas C.. **Long COVID: Systemic Inflammation and Obesity as Therapeutic Targets**. *Lancet Respir. Med.* (2022.0) **10** 726-727. DOI: 10.1016/S2213-2600(22)00159-X
50. Maurya R., Sebastian P., Namdeo M., Devender M., Gertler A.. **COVID-19 Severity in Obesity: Leptin and Inflammatory Cytokine Interplay in the Link between High Morbidity and Mortality**. *Front. Immunol.* (2021.0) **12** 649359. DOI: 10.3389/fimmu.2021.649359
51. Mendola M., Leoni M., Cozzi Y., Manzari A., Tonelli F., Metruccio F., Tosti L., Battini V., Cucchi I., Costa M.C.. **Long-Term COVID Symptoms, Work Ability and Fitness to Work in Healthcare Workers Hospitalized for Sars-CoV-2 Infection**. *Med. Lav. Work Environ. Health* (2022.0) **113** e2022040. DOI: 10.23749/mdl.v113i5.13377
52. Lulli L.G., Baldassarre A., Mucci N., Arcangeli G.. **Prevention, Risk Exposure, and Knowledge of Monkeypox in Occupational Settings: A Scoping Review**. *Trop. Med. Infect. Dis.* (2022.0) **7**. DOI: 10.3390/tropicalmed7100276
53. Murray C.J.L.. **COVID-19 Will Continue but the End of the Pandemic Is Near**. *Lancet* (2022.0) **399** 417-419. DOI: 10.1016/S0140-6736(22)00100-3
|
---
title: Metabolomics Analysis Reveals Novel Targets of Chemosensitizing Polyphenols
and Omega-3 Polyunsaturated Fatty Acids in Triple Negative Breast Cancer Cells
authors:
- Blake R. Rushing
- Alleigh Wiggs
- Sabrina Molina
- Madison Schroder
- Susan Sumner
journal: International Journal of Molecular Sciences
year: 2023
pmcid: PMC10002396
doi: 10.3390/ijms24054406
license: CC BY 4.0
---
# Metabolomics Analysis Reveals Novel Targets of Chemosensitizing Polyphenols and Omega-3 Polyunsaturated Fatty Acids in Triple Negative Breast Cancer Cells
## Abstract
Triple negative breast cancer (TNBC) is a subtype of breast cancer with typically poorer outcomes due to its aggressive clinical behavior and lack of targeted treatment options. Currently, treatment is limited to the administration of high-dose chemotherapeutics, which results in significant toxicities and drug resistance. As such, there is a need to de-escalate chemotherapeutic doses in TNBC while also retaining/improving treatment efficacy. Dietary polyphenols and omega-3 polyunsaturated fatty acids (PUFAs) have been demonstrated to have unique properties in experimental models of TNBC, improving the efficacy of doxorubicin and reversing multi-drug resistance. However, the pleiotropic nature of these compounds has caused their mechanisms to remain elusive, preventing the development of more potent mimetics to take advantage of their properties. Using untargeted metabolomics, we identify a diverse set of metabolites/metabolic pathways that are targeted by these compounds following treatment in MDA-MB-231 cells. Furthermore, we demonstrate that these chemosensitizers do not all target the same metabolic processes, but rather organize into distinct clusters based on similarities among metabolic targets. Common themes in metabolic targets included amino acid metabolism (particularly one-carbon and glutamine metabolism) and alterations in fatty acid oxidation. Moreover, doxorubicin treatment alone generally targeted different metabolites/pathways than chemosensitizers. This information provides novel insights into chemosensitization mechanisms in TNBC.
## 1. Introduction
Breast cancer is the most commonly diagnosed cancer in women and is the second leading cause of cancer-related deaths in women [1]. Triple negative breast cancer (TNBC) is a notoriously aggressive and highly metastatic classification of breast cancer characterized by a lack of expression of estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor 2 (HER2). TNBC accounts for approximately 15–$20\%$ of all breast cancer cases and has lower survival rates compared to hormone receptor-positive breast cancers due to a greater risk of recurrence and a more aggressive disease course [2,3]. Because TNBC does not respond to hormone therapy or HER2 directed therapeutics, treatment is limited to an aggressive course of cytotoxic chemotherapeutic drugs, which commonly includes high doses of anthracycline and taxane-based regimens [3]. This therapeutic strategy has significant health impacts, including issues with future fertility, premature menopause, cardiovascular toxicity, cognitive dysfunction, and poorer bone health [4]. As opposed to ER/PR-positive and HER2-positive breast cancers, few new treatment options have emerged for TNBC, resulting in little improvement in overall survival rates over the past 20–30 years compared to other cancers [5]. Recently, a panel of experts came up with a list of the top research needs in breast cancer, which was published in Annals of Oncology. This list included [1] improvement of care in young patients with breast cancer (due to higher rates of TNBC), [2] identification/validation of targets mediating chemotherapy resistance, and [3] identification of new targets in TNBC [6]. This highlights the necessity to explore new treatment options and/or methods to de-escalate chemotherapy in TNBC.
Certain dietary factors, such as polyphenols and omega-3 polyunsaturated fatty acids (PUFAs), have been shown to enhance the efficacy of chemotherapeutics or reverse multi-drug resistance (MDR) in vitro and in vivo against various cancers [7,8,9]. In particular, tannic acid [10], resveratrol [11,12], genistein [13], quercetin [14,15], curcumin [16], docosahexaenoic acid (DHA) [17], and eicosapentaenoic acid (EPA) [17] have been shown to increase the efficacy of doxorubicin—a first-line drug for TNBC—in MDA-MB-231 and MDA-MB-468 TNBC cell lines [10,11,12,13,14,15,16,18,19,20]. Moreover, these compounds have been shown to reverse MDR in cell lines of TNBC or other cancers [21,22,23,24,25,26,27,28]. A limitation of these compounds, particularly polyphenols, is their low bioavailability due to poor absorption or metabolic degradation by host/microbial enzymes [7,29,30], although major advancements have been made to overcome these issues using delivery systems such as nanoparticles or liposomes [31,32]. Nonetheless, these compounds still provide an excellent model to uncover novel therapeutic targets to enhance chemotherapeutic efficacy and/or reverse MDR. These dietary compounds are well-tolerated in humans and, therefore, therapeutics mimicking their actions (with improved pharmacokinetic properties) are likely to have favorable toxicity profiles. However, the mechanisms by which these compounds chemosensitize TNBC and reverse MDR remain unclear due to their pleiotropic nature. Indeed, many targets have been identified for these compounds, and it is ambiguous which targets are most crucial for their therapeutic effects [33,34,35,36].
Polyphenols and omega-3 PUFAs are highly studied molecules found in the diet, and much research has investigated their effects on cancer cells, including TNBC cells. In particular, these compounds and other anticancer nutrients/nutraceuticals have been shown to affect metabolic activity or processes regulating metabolism, which is associated with their anticancer activity [29,33,37,38]. Cancer cell metabolism greatly influences the response of cancer cells to therapeutics and the development of resistance, likely making metabolism a key target of these chemosensiziting compounds [39,40,41]. Moreover, co-administration of metabolic inhibitors has been shown to enhance the efficacy of chemotherapeutics [41], further strengthening the rationale that the metabolic targets of polyphenols/omega-3 PUFAs are critical for their chemosensitizing/MDR reversal effects. Research over many years has made it clear that these compounds affect multiple cellular targets, which has made it extremely difficult to identify the exact mechanisms of these nutrients using targeted methods. As a result, the literature is filled with various proposed mechanisms without a clear consensus on the critical targets of these compounds. Additionally, it is unknown whether the chemosensitizing effects of these compounds are due to targeting the same metabolic pathways as, or different ones from, TNBC chemotherapeutics (e.g., doxorubicin). Due to their pleiotropic properties, these nutrients/dietary compounds are well-positioned to be studied using omics techniques that can simultaneously measure many molecular species in a biological sample. In the current investigation, we present the use of metabolomics to elucidate metabolites/metabolic pathways targeted by a panel of polyphenols and PUFAs in the MDA-MB-231 TNBC cell line, providing insight towards the molecular mechanisms by which these compounds exert their chemosensitizing/MDR reversal effects (Figure 1).
## 2. Results
Following data preprocessing and filtering, 5097 peaks remained in the normalized metabolomics dataset, which were used for multivariate analyses. Quality control study pools (QCSPs) clustered tightly in the middle of the study samples, indicating the data collected was of sufficient quality (Figure S1). Principal component analysis (PCA) of all peaks showed moderate separation due to chemosensitizer treatment in MDA-MB-231 cells, with the largest separations seen with tannic acid, genistein, and EPA treatment compared to vehicle (Figure 2A). Orthogonal partial least squares-discriminant analysis (OPLS-DA), a supervised multivariate technique that uses class information, was able to produce very clear separation between treatment groups (Figure 2B). Importantly, this analysis showed treatments with similar metabolic profiles. For example, quercetin was closer in multivariate space to DHA compared to genistein, indicating that quercetin has a more similar metabotype with DHA. To further analyze overall similarities/differences in metabotypes between treatments, hierarchical clustering analysis (HCA) was performed on the OPLS-DA model to identify how treatments organized into clusters, identifying which treatments produced similar metabolic perturbations (Figure 2C). This clustering analysis revealed three distinct clusters: one cluster with resveratrol and curcumin, another cluster with DHA and quercetin, and a third cluster with tannic acid, EPA, and genistein. Notably, pairwise OPLS-DA comparisons between vehicle and each treatment group showed good model statistics with R2X, R2Y, and Q2 > 0.5, indicating that each treatment produced a robust effect on the metabolome of MDA-MB-231 cells, including treatment with doxorubicin (Supplementary Materials Table S1). These pairwise OPLS-DA models were used to calculate Variable Importance to Projection (VIP) scores for each peak, a multivariate score that indicates the contribution of a peak to the model. A full list of VIP scores, along with p-values and fold changes, for each vehicle–treatment combination is listed in Supplementary Materials Table S2. Furthermore, PCA of samples and QCSP replicates showed sufficient clustering and centering of QCSP samples, indicating good data quality.
To better understand the metabolic targets of each treatment compound, pairwise pathway analyses were performed for each vehicle–treatment combination. For each comparison, all peaks with their corresponding p-values and fold changes were input into MetaboAnalyst 5.0 for pathway analysis. Significant pathways for each treatment are listed in Table 1. This analysis showed that several pathways were found to be altered by multiple treatments. Notably, C21-steroid hormone biosynthesis and metabolism, histidine metabolism, aspartate and asparagine metabolism, linoleate metabolism, prostaglandin formation from arachidonate, and urea cycle/amino group metabolism were found to be perturbed by five out of the eight treatments. A graphical representation of these results can be found in Figure 3, which plots the −log (p-value) for each pathway for each treatment (only pathways significant in three or more treatments are displayed). This analysis highlights that for most treatments, there are one or two pathways that are noticeably more affected than the rest. For example, curcumin seems to primarily affect aspartate and asparagine metabolism, genistein primarily affects the carnitine shuttle, and resveratrol primarily affects histidine metabolism; however, each treatment has significant activity in many other pathways, which is in agreement with the pleiotropic properties that have been reported for these compounds.
While informative, MetaboAnalyst’s pathway analysis assigns metabolites to peaks based on accurate mass (MS) matches, which may lead to erroneous assignments due to lack of retention time (RT) and MS/MS matching. Because of this, we matched peaks to an in-house library of chemical reference standards that were run under identical instrument conditions, providing matches with increased evidence. From this, 169 peaks were matched to the in-house library at a level of OL1 (MS, RT, and MS/MS match), OL2a (RT and MS match), or OL2b (MS and MS/MS match). Displayed in Figure 4A is a heatmap of the in-house matched metabolites showing differences in abundance profiles across the different treatments. Clustering based on these metabolites leads to the similar grouping shown in Figure 2C using all of the metabolomics peaks, although EPA was shown to cluster with quercetin and DHA rather than tannic acid. ANOVA analysis of all in-house matched metabolites across all treatments was performed to identify compounds driving the observed clustering. Supplementary Materials Table S3 provides the p-values for all in-house metabolites using this analysis, revealing 58 in-house matched metabolites with an ANOVA of $p \leq 0.05.$ Notable significant metabolites identified from this analysis were creatine, glutamine, DHA, docosatetraenoic acid, sphinganine, spermine, and putrescine, which all had p-values < 1 × 10−5. Pathway analysis was performed on all 58 of these significant metabolites, which identified glutathione metabolism, aminoacyl-tRNA biosynthesis, and arginine and proline metabolism as major metabolic pathways driving the clustering of treatments (Supplementary Materials Table S4).
To gain a better understanding of each treatment on specific metabolite groups, metabolites were subdivided into categories based on Refmet classifications [42] and heatmaps were generated for each category, which included fatty acyls (Figure 4B), organic acids (Figure 4C), carbohydrates (Figure 4D), nucleic acids (Figure 4E), organoheterocyclics (Figure 4F), and acylcarnitines—a subgroup of fatty acyls (Figure 4G). Clustering of each treatment in these category heatmaps provide more insight into the metabolic targets of each chemosensitizer based on distance from the vehicle-treated MDA-MB-231 cells. EPA, DHA, curcumin, and quercetin had the largest effect on fatty acyls, generally increasing the long-chain forms and decreasing the short-chain forms (Figure 4B). Quercetin, DHA, resveratrol, and tannic acid had the largest effect on organic acids, particularly quercetin, which showed strong increases in glutamine, serine, asparagine, and betaine relative to the vehicle. Other amino acids were generally decreased with treatment by these four compounds; however, genistein, despite clustering closely with the vehicle, showed strong increases in some amino acids including histidine, methionine, isoleucine, phenylalanine, tryptophan, and cystine (Figure 4C). For carbohydrates, quercetin, DHA, and EPA had the largest effect, with increases in S-adenosylmethionine and decreases in mannose and lactose (Figure 4D). Quercetin, DHA, EPA, and curcumin had the largest effect on nucleic acids, with the former three decreasing and the latter increasing these metabolites (Figure 4E). For organoheterocyclics (a class that includes many B vitamin forms), quercetin and genistein had the largest effects, with the former leading to increases and the latter leading to decreases in these metabolites (Figure 4F). Finally, acylcarnitines were strongly increased in curcumin, resveratrol, genistein, and tannic acid, particularly medium- and long-chain acylcarnitines (Figure 4G).
To provide more robust pathway analysis, in-house matched metabolites with fold changes for each treatment were input into GeneGo Metacore for pathway analysis. Metabolites with a VIP > 1 for a given vehicle–treatment comparison were considered significant and used for pathway mapping. Figure 5 displays the top five metabolic pathways identified by this analysis (Supplementary Materials Table S5 contains the full list of pathways). This analysis identified several amino acid-related pathways as significantly altered by the treatments, including pathways related to glycine, serine, arginine, cysteine, glutathione, and aminoacyl tRNAs. This agrees with the MetaboAnalyst results in Figure 3, which also identified glycine, serine, arginine, and glutathione pathways as significantly altered by treatment. Notably, the GeneGo *Metacore analysis* identified fewer fatty acid/cholesterol-related pathways compared to the MetaboAnalyst results, which may be due to the GeneGo *Metacore analysis* only using the in-house matched metabolites, which had a higher representation of amino acids and their metabolites.
## 3. Discussion
Metabolic reprogramming is a hallmark of cancer and leads to cancer cells having distinct metabolic profiles compared to normal cells. This is due to cancer cells rewiring metabolic processes to overcome regulatory systems that would otherwise limit their growth and survival. This reprogramming also occurs in response to stressors, such as chemotherapy treatment, to promote the survival of cancer cells. In this way, alterations in cellular metabolism can modulate the response of cancer cells to drug treatment [43,44]. Certain nutrients/phytochemicals such as omega-3 PUFAs (often found in fish, nuts, and seeds) and polyphenols (commonly found in fruits, vegetables, nuts, and whole grains) have received significant interest in the research community for their observed health effects, such as their ability to prevent cancer and/or induce cancer cell death in experimental systems [34,45]. Included in these observations is the ability of these compounds to enhance the anticancer effect of chemotherapeutics. Because drug response is closely linked to cancer cell metabolism, we hypothesize that this chemosensitization effect is due to these compounds altering the metabotype of cancer cells, making them more responsive to the cytotoxic effect of chemotherapeutics. In the current investigation, we investigated a panel of polyphenols and omega-3 PUFAs that have previously been shown to increase the anticancer effect of doxorubicin in triple negative breast cancer cells. Using an untargeted metabolomics approach, we sought to identify the metabolites/metabolic pathways that are targeted by these chemosensitizing compounds (Figure 6).
Importantly, our findings indicated that the metabolic effects of these chemosensitizing compounds were broad, and often distinct from one another. This is in agreement with many studies that indicated that these compounds are pleiotropic. Additionally, this also suggests that there are multiple mechanisms by which metabolism can be altered to improve drug response. Even EPA and DHA, which are highly related metabolites that belong to the same metabolic pathway, showed different metabolic effects, although clustering analyses frequently placed these two treatments into the same cluster. This agrees with previous studies that have shown that EPA and DHA can have different anticancer effects, with DHA often shown to have greater anticancer effects than EPA [37]. Notably, doxorubicin-treated MDA-MB-231 cells generally showed very different metabolic profiles than chemosensitizer-treated cells. This suggests that these polyphenols and omega-3 PUFAs sensitize TNBC cells by targeting different, complementary metabolites to increase the drug’s cytotoxic effect.
Pathway analyses provide a means to more easily interpret overall biological effects in metabolomics data. Herein, we provided two pathway analyses: one using all peaks via MetaboAnalyst and another using only in-house matched metabolites via GeneGo. Both analyses indicated amino acid metabolism as a major target, with the GeneGo results (using only matches with the highest evidence basis) particularly identifying amino acids involved in one-carbon metabolism (glycine, serine, cysteine, and cystine)—a pathway that controls the flux of one-carbon units towards numerous pathways including nucleotide and lipid metabolism [46]. Cancer cells are particularly sensitive to deprivation of one-carbon units through nutrient restriction or pharmacological inhibition of the one-carbon metabolic pathway, as seen with the clinical success of folate inhibitors such as methotrexate and pemetrexed [47]. Indeed, cancer cells rely on this pathway to increase anabolic pathways (nucleotide/lipid synthesis), produce NADPH to adapt to the high levels of reactive oxygen species that are characteristic of cancer cells, produce energy in the form of adenosine triphosphate (ATP), and alter DNA methylation patterns [46,47]. Interestingly, the direction of change of metabolites in this pathway varied across treatments, suggesting that dysregulation of this pathway—by either increasing or decreasing activity—can lead to chemosensitization towards doxorubicin treatment.
Another amino acid that was heavily affected by chemosensitizer treatment was glutamine, which was decreased in all chemosensitizer treatments. Conversely, glutamine levels were strongly increased following doxorubicin treatment (VIP > 1, $$p \leq 8.61$$ × 10−5, fold change > 5) (Supplementary Materials Table S2). This was one of the few instances where an OL1 metabolite was consistently changed in the same direction by all chemosensitizers while also being significantly affected by doxorubicin treatment. Increases in glutamine uptake are commonly seen in cancers to support biosynthetic reactions and combat redox stress, and yield these results by replenishing TCA cycle intermediates, which are then shuttled to anabolic reactions [48]. The observation that glutamine was increased following doxorubicin treatment may be an indicator that increased glutamine uptake is a stress response that TNBC cells undergo to survive the cytotoxic effects of this drug. Consequently, this panel of polyphenols/omega-3 PUFAs may enhance doxorubicin’s cytotoxic effect by depleting glutamine levels, preventing this survival response. In addition to glutamine, three other in-house matched metabolites were altered in all treatment groups (VIP > 1): myristoylcarnitine, octadecanoylcarnitine, and 3-hydroxyhexadecanoylcarnitine (Supplementary Materials Table S2). This indicates that these acylcarnitines and glutamine are shared targets of doxorubicin and these chemosensitizers, and that the simultaneous disruption of these metabolic pathways during doxorubicin treatment may lead to increased drug efficacy. Future studies are needed to investigate the metabolic profiles of cells co-treated with doxorubicin and these chemosensitizers to determine if these effects are seen when both agents are administered simultaneously. In addition, while previous studies have shown an increase in cytotoxicity from these co-treatments in TNBC cells, we did not assess cell viability with these combinations, which should be confirmed in future studies. Future studies should also investigate if these metabolic processes are targeted by other polyphenols/PUFAs beyond those used in this study. Lastly, additional TNBC models should be studied to confirm our results, as the current investigation only assessed the effects of these chemosensitizers in the MDA-MB-231 cell line.
One of the observed effects of polyphenols on cancer cells is their ability to modulate cellular energetics. Polyphenols have been shown to activate AMPK, which alters many anabolic and catabolic processes, such as fatty acid oxidation, glycolysis, lipogenesis, and autophagy [49]. Cancer cells carefully balance energy consumption and generating pathways to sustain increased proliferation and manage reactive oxygen species (ROS) levels [50]. Disrupting this balance may be a mechanism by which polyphenols cause cancer cell death and/or increase chemotherapeutic efficacy. Our findings that a subset of polyphenols (curcumin, genistein, tannic acid, and resveratrol) greatly alters acylcarnitine levels in TNBC cells, typically increasing their levels. Acylcarnitines are intermediates in fatty acid oxidation that are formed from acyl-CoA and carnitine by the action of carnitine palmitoyltransferase 1 (CPT-1). Once formed, acylcarnitines are able to pass into the mitochondrial matrix, where they are re-converted into acyl-CoA by CPT-2 and then oxidized via β-oxidation to acetyl-CoA, which is then used in the TCA cycle for ATP production [51]. In the context of diabetes, disturbances in the acylcarnitine pool have been shown to be established markers of mitochondrial dysfunction and the uncoupling of fatty acid oxidation (FAO) from oxidative phosphorylation [52]. Indeed, elevation of acylcarnitines has been shown to occur when FAO activity outpaces the TCA cycle, leading to increased lipolysis and incomplete mitochondrial substrate oxidation [53,54]. Under these conditions, where substrate catabolism exceeds ATP demand, the increased reducing pressure of the cell (NADH, FADH2) on the electron transport chain leads to the generation of ROS (H2O2, •O2) [53]. Excessive mitochondrial ROS and accumulation of acyls in the mitochondria have been shown to open the mitochondrial permeability transition pore (PTP), causing cell death [52,55,56,57,58]. Our data suggest that a similar mechanism may occur in cancer cells following treatment with these polyphenols. Although polyphenols have historically been recognized as antioxidants, they are now recognized to have pro-oxidant effects in cancer cell environments [59,60,61,62], which may be due to the accumulation of acylcarnitines. Interestingly, the breast cancer cell response to doxorubicin is heavily influenced by mitochondrial activity, with factors such as mitochondrial oxidation state, depolarization, matrix calcium levels, and ROS production mediating its activity [63,64]. Although the exact mechanism of action of doxorubicin remains unclear, it has been well observed to lead to mitochondrial dysfunction, which is thought to play a critical role in the cardiotoxicity seen with this drug in in vitro, in vivo, and clinical studies [65]. Our observation that a subset of chemosensitizers heavily affects acylcarnitine levels suggests that these compounds increase doxorubicin efficacy in breast cancer cells by shifting the equilibrium of mitochondrial activity. In turn, this shifting of mitochondrial activity may lead to a cytoprotective effect of these compounds in cardiomyocytes against doxorubicin-mediated toxicity [66,67,68]. More research is needed to better understand if the metabolic effects of these compounds are also seen in cardiomyocytes and if they contribute to this observation of selective toxicity (cytotoxic in cancer cells, cytoprotective in healthy cells). *The* generally higher levels of ROS seen in cancer cells versus normal cells may play a role in determining this selective toxicity, making cancer cells more sensitive to imbalances in mitochondrial metabolism/ROS production [69]. Of note, acylcarnitine treatment has been shown to slow the development of certain cancers, such as colon cancer in vivo [70].
Additionally, polyphenols/omega-3 PUFAs have been shown to alter the activity of the PI3K-Akt/mTOR/AMPK signaling axis, which is well known to modulate amino acid metabolism and mitochondrial activity/fatty acid oxidation, providing a possible mechanism for how these chemosensitizing compounds affect the pathways seen in this study [37,71,72]. Indeed, the mTOR signaling pathway is a central metabolic regulator in the cell that senses the nutritional status of the cell, controlling growth and metabolic activity [73]. Previous studies have shown that modulation of mTOR activity in combination with doxorubicin shows synergistic activity in in vitro and in vivo models of various cancers [74,74,75,76]. This combination treatment has also been shown to be effective in a Phase I trial for the treatment of mesenchymal TNBC [77]. mTOR inhibition has also been shown to be effective in combination with other anticancer therapeutics, indicating that targeting of this pathway is a promising avenue for chemosensitization [78]. Because our panel of chemosensitizers shows different profiles of metabolic perturbations, it is possible that they target different locations on the mTOR pathway and/or have different off-target effects, which may contribute to their anticancer effects and/or favorable toxicity profiles.
Previous metabolic effects of these compounds in other disease contexts are in agreement with our results. Resveratrol, curcumin, genistein, and tannic acid have been shown to increase acylcarnitine profiles systemically and/or in skeletal muscle, alter the expression of mitochondrial β-oxidation enzymes, affect mitochondrial biogenesis, and change mitochondrial bioenergetics, which has been linked to the anti-obesogenic and anti-aging effects of these compounds [79,80,81,82,83,84,85]. Additionally, resveratrol, curcumin, genistein, quercetin, DHA, and EPA have all been shown to modulate glutaminolysis/glutamine levels in vitro or in vivo [86,87,88,89,90]. While these previous studies give additional validity to our results, this study was a screening approach to identify a list of metabolic targets of these compounds, and will need to be validated using targeted methods in additional model systems.
The pleiotropic effects of these polyphenols/omega-3 PUFAs highlight the promise of multi-targeted therapy in cancer. Indeed, the anticancer effect of these compounds may be attributed to a combination of effects on multiple metabolites/pathways rather than a single primary target. Our study uncovers novel metabolic targets of these compounds that aid in explaining their chemosensitizing effects, and these metabolic targets may provide an explanation for the health benefits of these compounds in other disease areas (aging, cardiovascular disease, neurodegeneration, etc). However, it should be noted that the concentrations, dose times, and cell model used in this study were specifically designed in the context of enhancing the effect of doxorubicin in TNBC, and therefore may not necessarily reflect the metabolic effects seen in these other contexts. Furthermore, our observation that each treatment generally produced unique metabolic profiles indicates that different polyphenols/omega-3 PUFAs target different sets of metabolites/metabolic pathways. This may provide a rationale for combining polyphenols/omega-3 PUFAs based on their targeting of similar/complementary metabolites to obtain an additive/synergistic effect—for example, resveratrol and genistein may be combined as a co-treatment since both increase acylcarnitines. Conversely, this may also be a way to predict compounds that could antagonize each other’s effects, leading to diminished therapeutic effects. Indeed, combinations of these compounds have been shown to be a promising area for improving their effects [7,30]; therefore, using omics technologies to make rationalized combinations is an area worthy of future study.
In conclusion, our study uncovered novel mechanisms by which polyphenols/omega-3 PUFAs target metabolism under doxorubicin-chemosensitizing conditions. While bioavailability issues continue to be a challenge in using these compounds clinically, mechanistic information, such as the data presented herein, may form a basis for developing mimetics with more favorable pharmacokinetic profiles. For TNBC specifically, understanding these chemosensitization mechanisms is clinically very valuable, as there is a great need to improve treatment outcomes and de-escalate drug doses to mitigate side effects. It is important to note that more information is needed concerning the mechanism of action of these polyphenols and omega-3 PUFAs, as well as more testing in experimental and clinical settings. These needs must be met to fully understand their anticancer properties and to use this information to improve outcomes for TNBC patients.
## 4.1. Chemical Reagents
Optima grade solvents (water with $0.1\%$ formic acid and methanol with $0.1\%$ formic acid) and fetal bovine serum (FBS) were purchased from Fisher Scientific (Waltham, MA, USA). Dubelcco’s Modified Eagle Medium (DMEM) with high glucose and phosphate buffered saline (PBS) was purchased from Gibco (Grand Island, NY, USA). Resveratrol, curcumin, quercetin, genistein, tannic acid, DHA, and EPA were purchased from Cayman Chemical (Ann Arbor, MI, USA). The MDA-MB-231 cell line was purchased from the American Type Culture Collection (ATCC) (Manassas, VA, USA).
## 4.2. Cell Culture
MDA-MB-231 cells were cultured according to manufacturer guidelines. Cells were cultured in DMEM supplemented with $10\%$ FBS, 2 mM glutamine, 50 U/mL penicillin, and 50 µg/mL streptomycin. Cells were plated in 10 cm culture dishes and grown to ~$70\%$ confluency. Cells were treated with individual test compounds at concentrations previously shown to chemosensitize cells to doxorubicin (resveratrol—50 µM, curcumin—40 µM, tannic acid—25 µM, genistein—50 µM, DHA—29 µM, EPA—32µM) [10,11,12,13,14,15,16,17]. Concentration of the vehicle (dimethyl sulfoxide, DMSO) was kept at $0.01\%$ for all treatments. Additional dishes were treated with vehicle alone and doxorubicin (0.2 µM). All treatments were performed in triplicate for 24 h.
## 4.3. Metabolite Extraction
After treatment, metabolites were extracted from cell samples as described previously [91,92,93]. Briefly, treatment media were aspirated, and cells were washed with 5 mL of ice-cold PBS. After aspirating off PBS, 2 mL of ice-cold $80\%$ methanol was added to culture dishes, and cells were detached using cell scrapers. Cell suspensions were added to MagNA lyser homogenization tubes with ceramic beads inside and were lysed using an Omni Bead Ruptor Elite (OMNI International) at 6.00 m/s for two cycles at 45 s each with 30 s dwell time between each cycle. Additional $80\%$ methanol was added to each tube to normalize for protein concentration. Samples were centrifuged at 16,000× g at 4 °C for 10 min and supernatants were transferred to autosampler vials for analysis by ultra-high-pressure liquid chromatography–high-resolution mass spectrometry (UHPLC-HRMS). Quality control study pools (QCSP) were created by combining 10 µL of each sample into a single mixture. Method blanks were created by adding 500 µL of $80\%$ methanol to empty MagNA lyser tubes and were processed in an identical manner as the study samples.
## 4.4. UHPLC-HRMS Metabolomics Data Acquisition, Preprocessing, and Multivariate Analysis
Metabolomics data were acquired via previously published UHPLC-HRMS methods using a Vanquish UHPLC system coupled to a Q Exactive™ HF-X Hybrid Quadrupole-Orbitrap Mass Spectrometer (Thermo Fisher Scientific, San Jose, CA, USA) equipped with an HSS T3 C18 column (2.1 × 100 mm, 1.7 µm, Waters Corporation) held at 50 °C [91,92,93,94,95,96,97,98]. A binary pump was used with water + $0.1\%$ formic acid (A) and methanol + $0.1\%$ formic acid (B) as mobile phases. The mobile phase gradient started from $2\%$ B, increased to $100\%$ B in 16 min, and was then held for 4 min with a flow rate of 400 µL/min. Mass spectral data were collected using a data-dependent acquisition mode in positive polarity at 70–1050 m/z. QCSP and blank injections were placed at a rate of $10\%$ throughout the study samples. An injection volume of 5 µL was used for analysis of each sample. Raw UHPLC-HRMS data were imported into Progenesis QI (version 2.1, Waters Corporation, MA, USA) for alignment, peak picking, and deconvolution. Background signals were removed by filtering out peaks with a higher average abundance in the blank injections as compared to the QCSP injections. Data were normalized using a QCSP reference sample using the “normalize to all” function in progenesis [99].
## 4.5. Multivariate and Univariate Statistical Analysis
The normalized, filtered data were imported into SIMCA 16 (Sartorius Stedim Data Analytics AB, Umeå, Sweden), scaled using Unit Variance (UV) scaling, and then used to generate principal component analysis (PCA) and orthogonal partial least squares-discriminant analysis (OPLS-DA). PCA plots were used to assess data quality and clustering of QCSP samples, and OPLS-DA plots were used to assess the separation of metabolomes between vehicle and treated cells as well as to calculate variable importance to projection (VIP) scores for each peak. Heatmaps were generated using MetaboAnalyst 5.0. Fold changes and p-values were calculated for each peak for each treatment as compared to the vehicle control. p-values were calculated using Student’s t-test. p-values were not adjusted for multiple testing due to the small sample size of this study and the exploratory, rather than confirmatory, nature of this study [100].
## 4.6. Compound Identification/Annotation
Peaks were matched to an in-house library of reference standards or public mass spectral databases from the National Institute of Standards and Technology (NIST) and METLIN. Peaks were matched to metabolites by retention time (RT, ±0.5 min, in-house library only), exact mass (MS, <5 ppm), and fragmentation pattern (MS/MS, similarity score > 30). An ontology system was given to denote the evidence basis for each metabolite assignment. OL1 refers to a match to the in-house library for RT, MS, and MS/MS; OL2a refers to an in-house match to the in-house library for RT and MS; OL2b refers to a match to the in-house library for MS and MS/MS; PDa refers to a match to public databases for MS and MS/MS; PDb refers to a public database match for MS and theoretical MS/MS (HMDB); PDc refers to a public database match for MS and isotopic similarity; PDd refers to a public database match for MS only.
## 4.7. Pathway Analysis
For each treatment–vehicle comparison, all untargeted peaks, along with their calculated p-values and fold changes, were imported into the “Functional Analysis” module in MetaboAnalyst 5.0 to identify significantly perturbed metabolic pathways. Both mummichog and gene set enrichment analysis (GSEA) algorithms were chosen for analysis. The default top $10\%$ of peaks by p-value was chosen for the mummichog algorithm. For more robust pathway analysis, only metabolites that matched to levels of OL1, OL2a, and OL2b along with their fold changes were imported into GeneGo Metacore (Clarivate Analytics, London, UK), and pathway analysis was performed for metabolites that had a VIP > 1 in the pairwise OPLS-DA comparisons between vehicle and each treatment.
## References
1. Siegel R.L., Miller K.D., Jemal A.. **Cancer statistics, 2019**. *CA Cancer J. Clin.* (2019) **69** 7-34. DOI: 10.3322/caac.21551
2. Dent R., Trudeau M., Pritchard K.I., Hanna W.M., Kahn H.K., Sawka C.A., Lickley L.A., Rawlinson E., Sun P., Narod S.A.. **Triple-Negative Breast Cancer: Clinical Features and Patterns of Recurrence**. *Clin. Cancer Res.* (2007) **13** 4429-4434. DOI: 10.1158/1078-0432.CCR-06-3045
3. Waks A.G., Winer E.P.. **Breast Cancer Treatment: A Review**. *JAMA* (2019) **321** 288-300. DOI: 10.1001/jama.2018.19323
4. Runowicz C.D., Leach C.R., Henry N.L., Henry K.S., Mackey H.T., Cowens-Alvarado R.L., Cannady R.S., Pratt-Chapman M., Edge S.B., Jacobs L.A.. **American Cancer Society/American Society of Clinical Oncology Breast Cancer Survivorship Care Guideline**. *CA Cancer J. Clin.* (2015) **66** 43-73. DOI: 10.3322/caac.21319
5. Zeichner S.B., Terawaki H., Gogineni K.. **A Review of Systemic Treatment in Metastatic Triple-Negative Breast Cancer**. *Breast Cancer: Basic Clin. Res.* (2016) **10** 25-36. DOI: 10.4137/BCBCR.S32783
6. Cardoso F., Harbeck N., Barrios C.H., Bergh J., Cortes J., El Saghir N., Francis P.A., Hudis C.A., Ohno S., Partridge A.H.. **Research needs in breast cancer**. *Ann. Oncol.* (2016) **28** 208-217. DOI: 10.1093/annonc/mdw571
7. Fantini M., Benvenuto M., Masuelli L., Frajese G.V., Tresoldi I., Modesti A., Bei R.. **In Vitro and in Vivo Antitumoral Effects of Combinations of Polyphenols, or Polyphenols and Anticancer Drugs: Perspectives on Cancer Treatment**. *Int. J. Mol. Sci.* (2015) **16** 9236-9282. DOI: 10.3390/ijms16059236
8. Lewandowska U., Gorlach S., Owczarek K., Hrabec E., Szewczyk K.. **Synergistic interactions between anticancer chemo-therapeutics and phenolic compounds and anticancer synergy between polyphenols**. *Adv. Hyg. Exp. Med.* (2014) **68** 528-540
9. Sarkar F.H., Li Y.. **Using Chemopreventive Agents to Enhance the Efficacy of Cancer Therapy**. *Cancer Res.* (2006) **66** 3347-3350. DOI: 10.1158/0008-5472.CAN-05-4526
10. Tikoo K., Sane M.S., Gupta C.. **Tannic acid ameliorates doxorubicin-induced cardiotoxicity and potentiates its anti-cancer activity: Potential role of tannins in cancer chemotherapy**. *Toxicol. Appl. Pharmacol.* (2011) **251** 191-200. DOI: 10.1016/j.taap.2010.12.012
11. Kim T.H., Shin Y.J., Won A.J., Lee B.M., Choi W.S., Jung J.H., Chung H.Y., Kim H.S.. **Resveratrol enhances chemosensitivity of doxorubicin in multidrug-resistant human breast cancer cells via increased cellular influx of doxorubicin**. *Biochim. Biophys. Acta (BBA)-Gen. Subj.* (2014) **1840** 615-625. DOI: 10.1016/j.bbagen.2013.10.023
12. Rai G., Mishra S., Suman S., Shukla Y.. **Resveratrol improves the anticancer effects of doxorubicin in vitro and in vivo models: A mechanistic insight**. *Phytomedicine* (2016) **23** 233-242. DOI: 10.1016/j.phymed.2015.12.020
13. Lim H.A., Kim J.H., Kim J.H., Sung M.-K., Kim M.K., Park J.H.Y., Kim J.-S.. **Genistein Induces Glucose-Regulated Protein 78 in Mammary Tumor Cells**. *J. Med. Food* (2006) **9** 28-32. DOI: 10.1089/jmf.2006.9.28
14. Staedler D., Idrizi E., Kenzaoui B.H., Juillerat-Jeanneret L.. **Drug combinations with quercetin: Doxorubicin plus quercetin in human breast cancer cells**. *Cancer Chemother. Pharmacol.* (2011) **68** 1161-1172. DOI: 10.1007/s00280-011-1596-x
15. Li S., Yuan S., Zhao Q., Wang B., Wang X., Li K.. **Quercetin enhances chemotherapeutic effect of doxorubicin against human breast cancer cells while reducing toxic side effects of it**. *Biomed. Pharmacother.* (2018) **100** 441-447. DOI: 10.1016/j.biopha.2018.02.055
16. Zhou Q., Ye M., Lu Y., Zhang H., Chen Q., Huang S., Su S.. **Curcumin Improves the Tumoricidal Effect of Mitomycin C by Suppressing ABCG2 Expression in Stem Cell-Like Breast Cancer Cells**. *PLoS ONE* (2015) **10**. DOI: 10.1371/journal.pone.0136694
17. Germain E., Chajès V., Cognault S., Lhcillery C., Bougnoux P.. **Enhancement of doxorubicin cytotoxicity by polyun-saturated fatty acids in the human breast tumor cell line MDA-MB-231: Relationship to lipid peroxidation**. *Int. J. Cancer* (1998) **75** 578-583. DOI: 10.1002/(SICI)1097-0215(19980209)75:4<578::AID-IJC14>3.0.CO;2-5
18. Qian J., Xia M., Liu W., Li L., Yang J., Mei Y., Meng Q., Xie Y.. **Glabridin resensitizes p-glycoprotein-overexpressing multidrug-resistant cancer cells to conventional chemo-therapeutic agents**. *Eur. J. Pharmacol.* (2019) **852** 231-243. DOI: 10.1016/j.ejphar.2019.04.002
19. Tyagi A.K., Agarwal C., Chan D.C., Agarwal R.. **Synergistic anti-cancer effects of silibinin with conventional cytotoxic agents doxorubicin, cisplatin and carboplatin against human breast carcinoma MCF-7 and MDA-MB468 cells**. *Oncol. Rep.* (2004) **11** 493-499. DOI: 10.3892/or.11.2.493
20. Kalyani C., Narasu M.L., Devi Y.P.. **Synergistic growth inhibitory effect of flavonol–kaempferol and conventional chemotherapeutic drugs on cancer cells**. *Int. J. Pharm. Pharm. Sci.* (2017) **9** 123. DOI: 10.22159/ijpps.2017v9i2.16021
21. Nabekura T., Yamaki T., Ueno K., Kitagawa S.. **Inhibition of P-glycoprotein and multidrug resistance protein 1 by dietary phytochemicals**. *Cancer Chemother. Pharmacol.* (2008) **62** 867-873. DOI: 10.1007/s00280-007-0676-4
22. Huang F., Wu X.-N., Chen J., Wang W.-X., Lu Z.-F.. **Resveratrol reverses multidrug resistance in human breast cancer doxorubicin-resistant cells**. *Exp. Ther. Med.* (2014) **7** 1611-1616. DOI: 10.3892/etm.2014.1662
23. Ye Q., Liu K., Shen Q., Li Q., Hao J., Han F., Jiang R.-W.. **Reversal of Multidrug Resistance in Cancer by Multi-Functional Flavonoids**. *Front. Oncol.* (2019) **9** 487. DOI: 10.3389/fonc.2019.00487
24. Imai Y., Tsukahara S., Asada S., Sugimoto Y.. **Phytoestrogens/flavonoids reverse breast cancer resistance pro-tein/ABCG2-mediated multidrug resistance**. *Cancer Res.* (2004) **64** 4346-4352. DOI: 10.1158/0008-5472.CAN-04-0078
25. Li H., Krstin S., Wink M.. **Modulation of multidrug resistant in cancer cells by EGCG, tannic acid and curcumin**. *Phytomedicine* (2018) **50** 213-222. DOI: 10.1016/j.phymed.2018.09.169
26. Sen G.S., Mohanty S., Hossain D.M.S., Bhattacharyya S., Banerjee S., Chakraborty J., Saha S., Ray P., Bhattacharjee P., Mandal D.. **Curcumin enhances the efficacy of chemotherapy by tailoring p65NFκB-p300 cross-talk in favor of p53-p300 in breast cancer**. *J. Biol. Chem.* (2011) **286** 42232-42247. DOI: 10.1074/jbc.M111.262295
27. Chen C., Zhou J., Ji C.. **Quercetin: A potential drug to reverse multidrug resistance**. *Life Sci.* (2010) **87** 333-338. DOI: 10.1016/j.lfs.2010.07.004
28. Wen C., Fu L., Huang J., Dai Y., Wang B., Xu G., Wu L., Zhou H.. **Curcumin reverses doxorubicin resistance via inhibition the efflux function of ABCB4 in doxorubicin-resistant breast cancer cells**. *Mol. Med. Rep.* (2019) **19** 5162-5168. DOI: 10.3892/mmr.2019.10180
29. Zhou Y., Zheng J., Li Y., Xu D.-P., Li S., Chen Y.-M., Li H.-B.. **Natural Polyphenols for Prevention and Treatment of Cancer**. *Nutrients* (2016) **8**. DOI: 10.3390/nu8080515
30. Niedzwiecki A., Roomi M.W., Kalinovsky T., Rath M.. **Anticancer Efficacy of Polyphenols and Their Combinations**. *Nutrients* (2016) **8**. DOI: 10.3390/nu8090552
31. Li Z., Jiang H., Xu C., Gu L.. **A review: Using nanoparticles to enhance absorption and bioavailability of phenolic phytochemicals**. *Food Hydrocoll.* (2015) **43** 153-164. DOI: 10.1016/j.foodhyd.2014.05.010
32. Mignet N., Seguin J., Chabot G.G.. **Bioavailability of Polyphenol Liposomes: A Challenge Ahead**. *Pharmaceutics* (2013) **5** 457-471. DOI: 10.3390/pharmaceutics5030457
33. Kim H.-S., Quon M.J., Kim J.-A.. **New insights into the mechanisms of polyphenols beyond antioxidant properties; lessons from the green tea polyphenol, epigallocatechin 3-gallate**. *Redox Biol.* (2014) **2** 187-195. DOI: 10.1016/j.redox.2013.12.022
34. Yashodhara B.M., Umakanth S., Pappachan J.M., Bhat S.K., Kamath R., Choo B.H.. **Omega-3 fatty acids: A comprehensive review of their role in health and disease**. *Postgrad. Med. J.* (2009) **85** 84-90. DOI: 10.1136/pgmj.2008.073338
35. Jing K., Wu T., Lim K.. **Omega-3 polyunsaturated fatty acids and cancer**. *Anti-Cancer Agents Med. Chem.* (2013) **13** 1162-1177. DOI: 10.2174/18715206113139990319
36. Kampa M., Nifli A.-P., Notas G., Castanas E.. **Polyphenols and cancer cell growth**. *Reviews of Physiology, Biochemistry and Pharmacology* (2007) **Volume 159** 79-113. DOI: 10.1007/112_2006_0702
37. D’Eliseo D., Velotti F.. **Omega-3 Fatty Acids and Cancer Cell Cytotoxicity: Implications for Multi-Targeted Cancer Therapy**. *J. Clin. Med.* (2016) **5**. DOI: 10.3390/jcm5020015
38. Wiggs A., Molina S., Sumner S.J., Rushing B.R.. **A Review of Metabolic Targets of Anticancer Nutrients and Nutraceuticals in Pre-Clinical Models of Triple-Negative Breast Cancer**. *Nutrients* (2022) **14**. DOI: 10.3390/nu14101990
39. Zaal E.A., Berkers C.R.. **The influence of metabolism on drug response in cancer**. *Front. Oncol.* (2018) **8** 500. DOI: 10.3389/fonc.2018.00500
40. Morandi A., Indraccolo S.. **Linking metabolic reprogramming to therapy resistance in cancer**. *Biochim. Biophys. Acta (BBA)-Rev. Cancer* (2017) **1868** 1-6. DOI: 10.1016/j.bbcan.2016.12.004
41. Zhao Y., Butler E.B., Tan M.. **Targeting cellular metabolism to improve cancer therapeutics**. *Cell Death Dis.* (2013) **4** e532. DOI: 10.1038/cddis.2013.60
42. Fahy E., Subramaniam S.. **RefMet: A reference nomenclature for metabolomics**. *Nat. Methods* (2020) **17** 1173-1174. DOI: 10.1038/s41592-020-01009-y
43. Boroughs L.K., DeBerardinis R.J.. **Metabolic pathways promoting cancer cell survival and growth**. *Nat. Cell Biol.* (2015) **17** 351-359. DOI: 10.1038/ncb3124
44. Rahman M., Hasan M.R.. **Cancer Metabolism and Drug Resistance**. *Metabolites* (2015) **5** 571-600. DOI: 10.3390/metabo5040571
45. Cory H., Passarelli S., Szeto J., Tamez M., Mattei J.. **The Role of Polyphenols in Human Health and Food Systems: A Mini-Review**. *Front. Nutr.* (2018) **5** 87. DOI: 10.3389/fnut.2018.00087
46. Pan S., Fan M., Liu Z., Li X., Wang H.. **Serine, glycine and one-carbon metabolism in cancer (Review)**. *Int. J. Oncol.* (2021) **58** 158-170. DOI: 10.3892/ijo.2020.5158
47. Newman A.C., Maddocks O.D.K.. **One-carbon metabolism in cancer**. *Br. J. Cancer* (2017) **116** 1499-1504. DOI: 10.1038/bjc.2017.118
48. Yoo H.C., Yu Y.C., Sung Y., Han J.M.. **Glutamine reliance in cell metabolism**. *Exp. Mol. Med.* (2020) **52** 1496-1516. DOI: 10.1038/s12276-020-00504-8
49. Kim J., Yang G., Kim Y., Kim J., Ha J.. **AMPK activators: Mechanisms of action and physiological activities**. *Exp. Mol. Med.* (2016) **48** e224. DOI: 10.1038/emm.2016.16
50. DeBerardinis R.J., Chandel N.S.. **Fundamentals of cancer metabolism**. *Sci. Adv.* (2016) **2** e1600200. DOI: 10.1126/sciadv.1600200
51. Houten S.M., Wanders R.J.A.. **A general introduction to the biochemistry of mitochondrial fatty acid β-oxidation**. *J. Inherit. Metab. Dis.* (2010) **33** 469-477. DOI: 10.1007/s10545-010-9061-2
52. Wajner M., Amaral A.U.. **Mitochondrial dysfunction in fatty acid oxidation disorders: Insights from human and animal studies**. *Biosci. Rep.* (2016) **36** e00281. DOI: 10.1042/BSR20150240
53. Muoio D.M., Neufer P.D.. **Lipid-Induced Mitochondrial Stress and Insulin Action in Muscle**. *Cell Metab.* (2012) **15** 595-605. DOI: 10.1016/j.cmet.2012.04.010
54. Koves T.R., Ussher J.R., Noland R.C., Slentz D., Mosedale M., Ilkayeva O., Bain J., Stevens R., Dyck J.R.B., Newgard C.B.. **Mitochondrial Overload and Incomplete Fatty Acid Oxidation Contribute to Skeletal Muscle Insulin Re-sistance**. *Cell Metab.* (2008) **7** 45-56. DOI: 10.1016/j.cmet.2007.10.013
55. Wieckowski M., Wojtczak L.. **Fatty acid-induced uncoupling of oxidative phosphorylation is partly due to opening of the mitochondrial permeability transition pore**. *FEBS Lett.* (1998) **423** 339-342. DOI: 10.1016/S0014-5793(98)00118-5
56. Rottenberg H., Hoek J.B.. **The path from mitochondrial ROS to aging runs through the mitochondrial permeability tran-sition pore**. *Aging Cell* (2017) **16** 943-955. DOI: 10.1111/acel.12650
57. Furuno T., Kanno T., Arita K., Asami M., Utsumi T., Doi Y., Inoue M., Utsumi K.. **Roles of long chain fatty acids and carnitine in mitochondrial membrane permeability transition**. *Biochem. Pharmacol.* (2001) **62** 1037-1046. DOI: 10.1016/S0006-2952(01)00745-6
58. Berezhnov A.V., Fedotova E.I., Nenov M.N., Kasymov V.A., Pimenov O.Y., Dynnik V.V.. **Dissecting Cellular Mechanisms of Long-Chain Acylcarnitines-Driven Cardiotoxicity: Disturbance of Calcium Homeostasis, Activation of Ca**. *Int. J. Mol. Sci.* (2020) **21**. DOI: 10.3390/ijms21207461
59. D'Angelo S., Martino E., Ilisso C.P., Bagarolo M.L., Porcelli M., Cacciapuoti G.. **Pro-oxidant and pro-apoptotic activity of polyphenol extract from Annurca apple and its underlying mechanisms in human breast cancer cells**. *Int. J. Oncol.* (2017) **51** 939-948. DOI: 10.3892/ijo.2017.4088
60. Babich H., Schuck A.G., Weisburg J.H., Zuckerbraun H.L.. **Research Strategies in the Study of the Pro-Oxidant Nature of Polyphenol Nutraceuticals**. *J. Toxicol.* (2011) **2011** 467305. DOI: 10.1155/2011/467305
61. Mileo A.M., Miccadei S.. **Polyphenols as Modulator of Oxidative Stress in Cancer Disease: New Therapeutic Strategies**. *Oxidative Med. Cell. Longev.* (2015) **2016** 1-17. DOI: 10.1155/2016/6475624
62. León-González A.J., Auger C., Schini-Kerth V.B.. **Pro-oxidant activity of polyphenols and its implication on cancer chemoprevention and chemotherapy**. *Biochem. Pharmacol.* (2015) **98** 371-380. DOI: 10.1016/j.bcp.2015.07.017
63. Kuznetsov A.V., Margreiter R., Amberger A., Saks V., Grimm M.. **Changes in mitochondrial redox state, membrane potential and calcium precede mitochondrial dysfunction in doxorubicin-induced cell death**. *Biochim. Biophys. Acta (BBA)-Mol. Cell Res.* (2011) **1813** 1144-1152. DOI: 10.1016/j.bbamcr.2011.03.002
64. Dornfeld K., Bjork J., Folkert G., Skildum A., Wallace K.B.. **Mitochondrial activities play a pivotal role in regulating cell cycle in response to doxorubicin**. *Cell Cycle* (2021) **20** 1067-1079. DOI: 10.1080/15384101.2021.1919839
65. Gorini S., De Angelis A., Berrino L., Malara N., Rosano G., Ferraro E.. **Chemotherapeutic Drugs and Mitochondrial Dysfunction: Focus on Doxorubicin, Trastuzumab, and Sunitinib**. *Oxidative Med. Cell. Longev.* (2018) **2018** 1-15. DOI: 10.1155/2018/7582730
66. Uygur R., Aktas C., Tulubas F., Alpsoy S., Topcu B., Ozen O.A.. **Cardioprotective effects of fish omega-3 fatty acids on doxorubicin-induced cardiotoxicity in rats**. *Hum. Exp. Toxicol.* (2013) **33** 435-445. DOI: 10.1177/0960327113493304
67. El Amrousy D., El-Afify D., Khedr R., Ibrahim A.M.. **Omega 3 fatty acids can reduce early doxorubicin-induced cardio-toxicity in children with acute lymphoblastic leukemia**. *Pediatr. Blood Cancer* (2022) **69** 1-8. DOI: 10.1002/pbc.29496
68. Razavi-Azarkhiavi K., Iranshahy M., Sahebkar A., Shirani K., Karimi G.. **The Protective Role of Phenolic Compounds Against Doxorubicin-induced Cardiotoxicity: A Comprehensive Review**. *Nutr. Cancer* (2016) **68** 892-917. DOI: 10.1080/01635581.2016.1187280
69. Nakamura H., Takada K.. **Reactive oxygen species in cancer: Current findings and future directions**. *Cancer Sci.* (2021) **112** 3945-3952. DOI: 10.1111/cas.15068
70. Roscilli G., Marra E., Mori F., Di Napoli A., Mancini R., Serlupi-Crescenzi O., Virmani A., Aurisicchio L., Ciliberto G.. **Carnitines slow down tumor development of colon cancer in the DMH-chemical carcinogenesis mouse model**. *J. Cell. Biochem.* (2013) **114** 1665-1673. DOI: 10.1002/jcb.24508
71. Szwed A., Kim E., Jacinto E.. **Regulation and metabolic functions of mTORC1 and mTORC2**. *Physiol. Rev.* (2021) **101** 1371-1426. DOI: 10.1152/physrev.00026.2020
72. Mirza-Aghazadeh-Attari M., Ekrami E.M., Aghdas S.A.M., Mihanfar A., Hallaj S., Yousefi B., Safa A., Majidinia M.. **Targeting PI3K/Akt/mTOR signaling pathway by polyphenols: Implication for cancer therapy**. *Life Sci.* (2020) **255** 117481. DOI: 10.1016/j.lfs.2020.117481
73. Saxton R.A., Sabatini D.M.. **mTOR Signaling in Growth, Metabolism, and Disease**. *Cell* (2017) **169** 361-371. DOI: 10.1016/j.cell.2017.03.035
74. Babichev Y., Kabaroff L., Datti A., Uehling D., Isaac M., Al-Awar R., Prakesch M., Sun R.X., Boutros P.C., Venier R.. **PI3K/AKT/mTOR inhibition in combination with doxorubicin is an effective therapy for leiomyosarcoma**. *J. Transl. Med.* (2016) **14** 67. DOI: 10.1186/s12967-016-0814-z
75. Iorio A.L., Da Ros M., Pisano C., de Martino M., Genitori L., Sardi I.. **Combined Treatment with Doxorubicin and Rapamycin Is Effective against In Vitro and In Vivo Models of Human Glioblastoma**. *J. Clin. Med.* (2019) **8**. DOI: 10.3390/jcm8030331
76. Li J., Liu W., Hao H., Wang Q., Xue L.. **Rapamycin enhanced the antitumor effects of doxorubicin in myelogenous leukemia K562 cells by downregulating the mTOR/p70S6K pathway**. *Oncol. Lett.* (2019) **18** 2694-2703. DOI: 10.3892/ol.2019.10589
77. Basho R.K., Gilcrease M., Murthy R.K., Helgason T., Karp D.D., Meric-Bernstam F., Hess K.R., Herbrich S.M., Valero V., Albarracin C.. **Targeting the PI3K/AKT/mTOR pathway for the treatment of mesenchymal triple-negative breast cancer: Evidence from a phase 1 trial of mTOR inhibition in combination with liposomal doxorubicin and bevacizumab**. *JAMA Oncol.* (2017) **3** 509-515. DOI: 10.1001/jamaoncol.2016.5281
78. Magaway C., Kim E., Jacinto E.. **Targeting mTOR and Metabolism in Cancer: Lessons and Innovations**. *Cells* (2019) **8**. DOI: 10.3390/cells8121584
79. Peron G., Sut S., Ben S.D., Voinovich D., Dall'Acqua S.. **Untargeted UPLC-MS metabolomics reveals multiple changes of urine composition in healthy adult volunteers after consumption of curcuma longa L. extract**. *Food Res. Int.* (2019) **127** 108730. DOI: 10.1016/j.foodres.2019.108730
80. Gimeno-Mallench L., Mas-Bargues C., Inglés M., Olaso G., Borras C., Gambini J., Vina J.. **Resveratrol shifts energy metabolism to increase lipid oxidation in healthy old mice**. *Biomed. Pharmacother.* (2019) **118** 109130. DOI: 10.1016/j.biopha.2019.109130
81. Guevara-Cruz M., Godinez-Salas E.T., Sanchez-Tapia M., Torres-Villalobos G., Pichardo-Ontiveros E., Guizar-Heredia R., Arteaga-Sanchez L., Gamba G., Mojica-Espinosa R., Schcolnik-Cabrera A.. **Genistein stimulates insulin sensitivity through gut microbiota reshaping and skeletal muscle AMPK activation in obese subjects**. *BMJ Open Diabetes Res. Care* (2020) **8** e000948. DOI: 10.1136/bmjdrc-2019-000948
82. De Oliveira M.R., Nabavi S.F., Manayi A., Daglia M., Hajheydari Z., Nabavi S.M.. **Resveratrol and the mitochondria: From triggering the intrinsic apoptotic pathway to inducing mi-tochondrial biogenesis, a mechanistic view**. *Biochim. Biophys. Acta.-Gen. Subj.* (2016) **1860** 727-745. DOI: 10.1016/j.bbagen.2016.01.017
83. De Oliveira M.R., Jardim F.R., Setzer W.N., Nabavi S.M., Nabavi S.F.. **Curcumin, mitochondrial biogenesis, and mitophagy: Exploring recent data and indicating future needs**. *Biotechnol. Adv.* (2016) **34** 813-826. DOI: 10.1016/j.biotechadv.2016.04.004
84. Do G.-M., Kwon E.-Y., Ha T.-Y., Park Y.B., Kim H.-J., Jeon S.-M., Lee M.-K., Choi M.-S.. **Tannic acid is more effective than clofibrate for the elevation of hepatic β-oxidation and the inhibition of 3-hydroxy-3-methyl-glutaryl-CoA reductase and aortic lesion formation in apo E-deficient mice**. *Br. J. Nutr.* (2011) **106** 1855-1863. DOI: 10.1017/S000711451100256X
85. De Oliveira M.R.. **Evidence for genistein as a mitochondriotropic molecule**. *Mitochondrion* (2016) **29** 35-44. DOI: 10.1016/j.mito.2016.05.005
86. Zhuang P., Li H., Jia W., Shou Q., Zhu Y., Mao L., Wang W., Wu F., Chen X., Wan X.. **Eicosapentaenoic and docosahexaenoic acids attenuate hyperglycemia through the microbiome-gut-organs axis in db/db mice**. *Microbiome* (2021) **9** 185. DOI: 10.1186/s40168-021-01126-6
87. Fan W.-H., Wang F.-C., Jin Z., Zhu L., Zhang J.-X.. **Curcumin Synergizes with Cisplatin to Inhibit Colon Cancer through Targeting the MicroRNA-137-Glutaminase Axis**. *Curr. Med Sci.* (2021) **42** 108-117. DOI: 10.1007/s11596-021-2469-0
88. Uifălean A., Schneider S., Gierok P., Ionescu C., Iuga C.A., Lalk M.. **The Impact of Soy Isoflavones on MCF-7 and MDA-MB-231 Breast Cancer Cells Using a Global Metabolomic Approach**. *Int. J. Mol. Sci.* (2016) **17**. DOI: 10.3390/ijms17091443
89. Freeman M.R., Kim J., Lisanti M.P., Di Vizio D.. **A metabolic perturbation by U0126 identifies a role for glutamine in resveratrol-induced cell death**. *Cancer Biol. Ther.* (2011) **12** 966-977. DOI: 10.4161/cbt.12.11.18136
90. Zhou Y., Zhang J., Wang K., Han W., Wang X., Gao M., Wang Z., Sun Y., Yan H., Zhang H.. **Quercetin overcomes colon cancer cells resistance to chemotherapy by inhibiting solute carrier family 1, member 5 transporter**. *Eur. J. Pharmacol.* (2020) **881** 173185. DOI: 10.1016/j.ejphar.2020.173185
91. Rushing B.R., Tilley S., Molina S., Schroder M., Sumner S.. **Commonalities in Metabolic Reprogramming between Tobacco Use and Oral Cancer**. *Int. J. Environ. Res. Public Health* (2022) **19**. DOI: 10.3390/ijerph191610261
92. Rushing B.R., Schroder M., Sumner S.C.J.. **Comparison of Lysis and Detachment Sample Preparation Methods for Cultured Triple-Negative Breast Cancer Cells Using UHPLC–HRMS-Based Metabolomics**. *Metabolites* (2022) **12**. DOI: 10.3390/metabo12020168
93. Rushing B.R., Fogle H.M., Sharma J., You M., McCormac J.P., Molina S., Sumner S., Krupenko N.I., Krupenko S.A.. **Exploratory Metabolomics Underscores the Folate Enzyme ALDH1L1 as a Regulator of Glycine and Methylation Reactions**. *Molecules* (2022) **27**. DOI: 10.3390/molecules27238394
94. Sharma J., Rushing B.R., Hall M.S., Helke K.L., McRitchie S.L., Krupenko N.I., Sumner S.J., Krupenko S.A.. **Sex-Specific Metabolic Effects of Dietary Folate Withdrawal in Wild-Type and**. *Metabolites* (2022) **12**. DOI: 10.3390/metabo12050454
95. Li Y., Rushing B., Schroder M., Sumner S., Kay C.D.. **Exploring the Contribution of (Poly)phenols to the Dietary Exposome Using High Resolution Mass Spectrometry Untargeted Metabolomics**. *Mol. Nutr. Food Res.* (2022) **66** 2100922. DOI: 10.1002/mnfr.202100922
96. Li S., Li Y., Rushing B.R., Harris S.E., McRitchie S.L., Dominguez D., Sumner S.J., Dohlman H.G.. **Multi-Omics Analysis of Multiple Glucose-Sensing Receptor Systems in Yeast**. *Biomolecules* (2022) **12**. DOI: 10.3390/biom12020175
97. Li S., Li Y., Rushing B.R., Harris S.E., McRitchie S.L., Jones J.C., Dominguez D., Sumner S.J., Dohlman H.G.. **Multi-omics analysis of glucose-mediated signaling by a moonlighting Gβ protein Asc1/RACK1**. *PLoS Genet.* (2021) **17** 1-30. DOI: 10.1371/journal.pgen.1009640
98. Rushing B.R., McRitchie S., Arbeeva L., Nelson A.E., Azcarate-Peril M.A., Li Y.-Y., Qian Y., Pathmasiri W., Sumner S.C., Loeser R.F.. **Fecal metabolomics reveals products of dysregulated proteolysis and altered microbial metabolism in obesity-related osteoarthritis**. *Osteoarthr. Cartil.* (2021) **30** 81-91. DOI: 10.1016/j.joca.2021.10.006
99. Välikangas T., Suomi T., Elo L.L.. **A systematic evaluation of normalization methods in quantitative label-free proteomics**. *Brief. Bioinform.* (2016) **19** 1-11. DOI: 10.1093/bib/bbw095
100. Bender R., Lange S.. **Adjusting for multiple testing—When and how?**. *J. Clin. Epidemiol.* (2001) **54** 343-349. DOI: 10.1016/S0895-4356(00)00314-0
|
---
title: Glucokinase Inactivation Ameliorates Lipid Accumulation and Exerts Favorable
Effects on Lipid Metabolism in Hepatocytes
authors:
- Ziyan Xie
- Ting Xie
- Jieying Liu
- Qian Zhang
- Xinhua Xiao
journal: International Journal of Molecular Sciences
year: 2023
pmcid: PMC10002408
doi: 10.3390/ijms24054315
license: CC BY 4.0
---
# Glucokinase Inactivation Ameliorates Lipid Accumulation and Exerts Favorable Effects on Lipid Metabolism in Hepatocytes
## Abstract
Glucokinase-maturity onset diabetes of the young (GCK-MODY) is a kind of rare diabetes with low incidence of vascular complications caused by GCK gene inactivation. This study aimed to investigate the effects of GCK inactivation on hepatic lipid metabolism and inflammation, providing evidence for the cardioprotective mechanism in GCK-MODY. We enrolled GCK-MODY, type 1 and 2 diabetes patients to analyze their lipid profiles, and found that GCK-MODY individuals exhibited cardioprotective lipid profile with lower triacylglycerol and elevated HDL-c. To further explore the effects of GCK inactivation on hepatic lipid metabolism, GCK knockdown HepG2 and AML-12 cell models were established, and in vitro studies showed that GCK knockdown alleviated lipid accumulation and decreased the expression of inflammation-related genes under fatty acid treatment. Lipidomic analysis indicated that the partial inhibition of GCK altered the levels of several lipid species with decreased saturated fatty acids and glycerolipids including triacylglycerol and diacylglycerol, and increased phosphatidylcholine in HepG2 cells. The hepatic lipid metabolism altered by GCK inactivation was regulated by the enzymes involved in de novo lipogenesis, lipolysis, fatty acid β-oxidation and the Kennedy pathway. Finally, we concluded that partial inactivation of GCK exhibited beneficial effects in hepatic lipid metabolism and inflammation, which potentially underlies the protective lipid profile and low cardiovascular risks in GCK-MODY patients.
## 1. Introduction
Glucokinase (GCK) catalyzes the phosphorylation of glucose to glucose 6-phosphate and is generally considered the initial glucose-sensing component and gatekeeper for glucose metabolism. *The* gene expression and protein level of GCK are enriched in the pancreas, liver, intestine, hypothalamus and pituitary [1]. In pancreatic β-cells, GCK participates in the regulation of glucose-induced insulin secretion. In the liver, GCK plays a leading role in glycogen synthesis and glycolysis [2].
Due to its important role in glucose homeostasis, the loss of function of GCK leads to diseases. Glucokinase-maturity-onset diabetes of the young (GCK-MODY) is caused by heterozygous inactivating mutations in GCK and impaired glucose sensing. However, unlike type 1 and 2 diabetes (T1D and T2D) or other MODYs, patients with GCK-MODY generally have a favorable prognosis without the requirement of antidiabetic treatment [3]. Additionally, GCK-MODY patients rarely suffer cardiovascular complications with the same risks as nondiabetic healthy individuals [4,5]. The low occurrence of vascular complications in GCK-MODY makes it a natural model for investigating the protective mechanisms of cardiovascular disorders under prolonged hyperglycemia. Research has shown that GCK-MODY individuals exhibit favorable serum lipid, with lower levels of triacylglycerols (TAGs) and higher high-density lipoproteins (HDLs), even compared with healthy subjects [6]. Our previous work [7] demonstrated that compared to T2D, several serum phosphatidylcholines (PCs) and plasmalogen PCs (PCps) were significantly increased in GCK-MODY, which contribute to the antiapoptotic and anti-inflammatory effects of HDL. Furthermore, evidence has suggested that the distinct lipid profile of GCK-MODY individuals exerts cardioprotective effects [8]. On the other hand, GCK activators (GKAs) have been reported to cause adverse effects including hyperlipidemia, hepatic fat accumulation and hepatic steatosis, in addition to hypoglycemic effects in both clinical trials [9,10,11] and animal studies [12,13], indicating the potential roles of GCK in maintaining lipid homeostasis.
Overall, current observations illustrated that in addition to glucose homeostasis, GCK also plays a crucial role in regulating lipid metabolism, and the inactivation of GCK may underlie the antiatherogenic profile associated with GCK-MODY. However, the association between GCK mutations and lipid profile and its underlying mechanism remains undefined. Given the critical role of the liver in lipid homeostasis of the body and the relatively high expression of GCK in the liver, we speculate that partial inactivation of GCK could exert favorable effects on hepatic lipid metabolism, probably through regulating key enzymes involved in metabolism pathways, thereby contributing to the cardioprotective lipid profile of GCK-MODY. The objective of the present study was to explore the protective lipid profile in GCK-MODY patients compared with T1D and T2D and the effects of GCK knockdown on hepatic lipid accumulation and inflammation in cell models.
## 2.1. GCK-MODY Patients Exert Favorable Lipid Profile
The characteristics and lipid profile of GCK-MODY, T1D and T2D patients and nondiabetic control subjects are shown in Table 1. In accordance with the type of diabetes, the glucose profile including FBG ($p \leq 0.0001$), HbA1c ($p \leq 0.001$) and GA ($p \leq 0.0001$) were increasingly elevated in three patient groups. The lipid metabolic profiles of GCK-MODY were significantly improved compared to T1D and T2D and were comparable to the normal control. The levels of TAG ($p \leq 0.0001$), TC ($p \leq 0.0001$) and LDL-c ($p \leq 0.0001$) were remarkably decreased in GCK-MODY compared with T2D. Additionally, a significant elevation in HDL-c was also shown in GCK-MODY compared with both T1D ($$p \leq 0.0060$$) and T2D ($p \leq 0.0001$). Furthermore, the level of CRP was also lower in GCK-MODY than T1D ($$p \leq 0.0256$$) and T2D ($$p \leq 0.0168$$), indicating a reduced cardiovascular risk.
## 2.2. GCK Knockdown Improved Lipid Accumulation in HFA-Treated HepG2 Cells
As the liver is the central organ of lipid metabolism in the body, we examined the effects of GCK inactivation on hepatic lipid metabolism via establishing in vitro liver cell models to explore the possible mechanism of the unique lipid profile in GCK-MODY individuals. Lentivirus transfection was applied to generate stable GCK knockdown in the human HpeG2 cell line. Glucokinase activity determination and Western blot were used to validate the transfection efficacy. The glucokinase activity was significantly reduced by $50\%$ in GCK knockdown HepG2 cells ($p \leq 0.0001$) (Figure 1A). Consistently, the level of GCK protein also displayed remarkable downregulation ($$p \leq 0.0017$$) (Figure 1B).
To investigate the impacts of GCK inactivation on hepatic lipid metabolism, the HepG2 cells were challenged with HFA to induce lipotoxicity. Oil Red O staining suggested that the knockdown of GCK significantly alleviated the HFA-treated lipid accumulation of HepG2 cells (Figure 1C). Meanwhile, the intracellular TAG content was also significantly reduced in the GCK knockdown group under HFA challenge ($p \leq 0.0001$) (Figure 1D). These results indicated that GCK knockdown reduced TAG content and prevented lipid accumulation in HFA-treated HepG2 cells.
## 2.3. Lipid Profile in GCK Knockdown HepG2 Cells
Lipidomic analysis was applied to reveal the entire lipid content variation caused by GCK knockdown (Figure 2). The score scatter plot of OPLS-DA (Figure 2A) showed that the samples from the control and GCK knockdown groups were independently grouped in both negative and positive ionization modes, which implied that the partial inactivation of GCK resulted in significant changes in the lipidome in HepG2 cells. Hierarchical cluster analysis was also performed on the screened differential metabolites, based on the threshold of variable importance values (VIP > 1.0) and p values (<0.05) (Figure S1).
The top 30 lipid metabolites were selected for further analysis based on the VIP score. The statistical data are depicted in a heatmap (Figure 2B). The selected metabolites could be classified as five lipid species, including glycerophospholipid, glycerolipid, sphingolipid, acylcarnitine and glycolipid. The bar chart shows the relative differences in the lipid species in the GCK knockdown group compared with the control (Figure 2C). The abundance of the identified lipids displayed a significant increase in PC, Cer, ACar and SQDG in the GCK knockdown group, as well as remarkable downregulation in PE, TAG, DAG, GM3 and GlcADG (lipids species detected in lipidomic analysis are shown in Table S1).The fold changes in the top 30 lipids are displayed in a matchstick plot (Figure S2).
## 2.4. GCK Knockdown Altered Hepatic Lipid Metabolism
Particularly, we analyzed the abundance of the differential metabolites involved in lipid metabolism pathways (Figure 3). Although fatty acids were not one of the top 30 lipids selected by VIP, due to their important role (the precursors of nearly all the kinds of lipids), the significantly changed fatty acids between groups were also brought into analysis. In fatty acid metabolism, palmitic acid (16:0) was drastically decreased in the GCK knockdown group; however, linoleic acid (18:2) was increased. Additionally, in ACars, the intermediates of fatty acids β-oxidation, the overall tendency was increased, despite some ACars being decreased, and correlation analysis (Figure 2D) showed that ACars were negatively correlated with TAG and DAG, suggesting active fatty acid utilization by β-oxidation in GCK-inactivating HepG2 cells. In glycerolipid metabolism, TAG and DAG were found significantly reduced, which may be due to inhibited lipogenesis. Additionally, in glycerophospholipid metabolism, most PCs showed an increasing trend, although some individual PCs were downregulated in the GCK knockdown group. The correlation analysis (Figure 2D) indicated that PCs were negatively correlated with TAG and DAG. Therefore, the biosynthesis of PC using DAG as precursors, referring to the Kennedy pathway, were promoted in HepG2 cells when GCK was inactivated. Moreover, PEs which could also be converted to PCs via PEMT pathways were found to be significantly reduced. In addition, sphingolipid metabolism was altered as well. GM3s were remarkably downregulated while the overall trend of Cers was upregulated in the GCK knockdown group. Taken together, these observations imply that the overall biosynthesis of PCs was preferentially enhanced, while palmitic acid, TAG and DAG were significantly reduced, probably due to inhibited synthesis or increased utilization in HepG2 cells with GCK inactivation.
## 2.5. Impact of GCK Knockdown on Lipid Metabolism-Related Enzymes and Inflammatory Genes in Human and Mouse Hepatic Cell Lines
To elucidate the mechanisms of GCK knockdown on hepatic lipid metabolism, the expression of proteins responsible for de novo lipogenesis (FASN and ACC), lipolysis (ATGL), β-oxidation of fatty acids (PPARα and CPT-1) and the Kennedy pathway for PC synthesis (CHPT-1) in the human HepG2 cells and mouse AML-12 cells was investigated using Western blot. The results showed that the levels of FASN and ACC were significantly downregulated in GCK knockdown HepG2 cells (Figure 4A). Conversely, the ratio of phosphorylated-ACC/ACC was increased (though not significant), indicating an inhibited state of ACC (Figure 4A). Additionally, the expressions of ATGL, PPARα, CPT-1 and CHPT-1 were significantly upregulated in GCK knockdown HepG2 cells (Figure 4B–D). Similar alterations in the protein levels were also found in siRNA-induced AML-12 cells, except for ATGL, which remained unaffected in both 100 nM and 200 nM dose groups (Figure 5). In the siRNA groups, the levels of GCK, FASN and ACC were reduced remarkably. Additionally, the decrease in FASN displayed a dose-dependent manner in 100 nM and 200 nM siRNA. The ratio of phosphorylated-ACC/ACC and expressions of PPARα, CPT-1 and CHPT-1 were found significantly increased. Among these, the ratio of phosphorylated-ACC/ACC was only elevated in the 200 nM siRNA-treated group. Altogether, these findings showed that GCK inactivation influenced the expression of target enzymes involved in de novo lipogenesis, lipolysis, FA oxidation and PC synthesis, resulting in reduced TAG accumulation and elevated PCs in the hepatic cells.
Additionally, we also measured the inflammatory cytokines (Figure 4E) and the expressions of NLRP3 and p-NF-kB (Figure S3) in GCK knockdown HepG2 cells under HFA challenges. The levels of IL-1β and MCP-1 were significantly decreased in GCK knockdown HepG2 cells under both normal and HFA conditions. IL-6 was significantly downregulated in GCK knockdown cells in normal conditions, but only showed a reduced trend without significance ($$p \leq 0.07$$). In the HFA group, the levels of cytokines in the control and GCK knockdown cells were both increased compared to normal groups, but the expressions of IL-1β and MCP-1 were still lower in GCK knockdown cells relative to the control. Consistently, the expressions of NLRP3 and p-NF-kB were also reduced significantly in GCK knockdown cells under normal and HFA conditions. Overall, the inflammatory markers were reduced in GCK knockdown HepG2 cells in both normal and HFA conditions, indicating the potential role of GCK knockdown in preventing inflammation induced by liptoxicity.
## 3. Discussion
Glucokinase is recognized as a glucose sensor. Recent evidence has suggested that inactivation of GCK may exert cardioprotective effects in GCK-MODY by regulating the lipid profile [7,8]. In the present study, we confirmed that GCK-MODY individuals exhibited metabolically normal and cardioprotective lipid profiles (i.e., lower TG, TC and LDLs, higher HDLs) compared to T1D and T2D. We also found that in hepatic cell models (HepG2 and AML-12), GCK inactivation improved lipid deposition and inflammation under HFA intervention and affected hepatic lipid profile by regulating key enzymes involved in lipid metabolism. These findings showed the beneficial effects of GCK inactivation in hepatic lipid metabolism and uncovered the potential mechanism of the protective lipid profile and low cardiovascular risks in GCK-MODY patients.
The liver is the major metabolic organ for lipid metabolism. We demonstrated that GCK inactivation alleviated the lipid accumulation under HFA intervention in HepG2 cells. To further investigate the molecular mechanism, lipidomic analysis was applied. A decreased level of palmitic acid as well as elevated PUFAs, including linoleic acid and docosahexaenoic acid (DHA), were detected in GCK knockdown liver cells. Evidence has shown that an excess of saturated FA palmitic acid results in lipotoxicity and inflammation in the liver, while some polyunsaturated fatty acids (PUFA), including linoleic acid, elicit opposite effects which improve insulin sensitivity and alleviate inflammation [14,15,16]. Additionally, ACars, the metric of mitochondrial β-oxidation [17], were increased in GCK knockdown HepG2 cells and were negatively correlated with DAG and TAG, suggesting an enhanced FA β-oxidation. Collectively, these results suggest that GCK inactivation may improve hepatic lipotoxicity and FA accumulation via decreasing the content of deleterious saturated FA and enhancing FA β-oxidation.
Moreover, the intracellular levels of glycerolipids TAG and DAG were both reduced in GCK knockdown HepG2 cells. Accumulating evidence has indicated that glycerolipids homeostasis is linked to glycerophospholipid [18]. When PC or PE synthesis was enhanced, the conversion of DAG to TAG would be inhibited. The main pathway for the biosynthesis of PC is the Kennedy pathway, which condenses CDP-choline with DAG to produce PC by the rate-determining enzyme cholinephosphotransferase (CHPT1) [19]. Hepatic PC synthesis has been considered metabolically beneficial in that the enhancement of PC synthesis facilitates the clearance of glycerolipids, including DAGs and TAGs, and induces the production and secretion of cardioprotective HDLs [20,21]. In this study, a correlation analysis showed that PCs were negatively correlated with TAGs and DAGs in GCK-inactivated HepG2 cells, indicating increased fluxes of lipids along the TAG-DAG-PC axis, which is in line with our previous works on serum lipidomics in GCK-MODY individuals [7]. Additionally, PEs which could be converted to PC by the PEMT pathway were found significantly reduced in the GCK knockdown group. Additionally, a reduced hepatic PC/PE ratio has been reported to be associated with hepatic steatosis, inflammation and fibrosis [22,23]. Subsequently, ELISA results showed that the levels of inflammatory cytokines (IL-1β, IL-6 and MCP-1) were significantly decreased in GCK knockdown groups under normal or HFA conditions, suggesting an anti-inflammatory state in GCK knockdown HepG2 cells. Taken together, GCK inactivation exerts favorable hepatic and serum lipid profiles, probably by promoting the biosynthesis of hepatic PC, thus inducing the clearance of TAG and DAG, increasing overall circulating HDLs and preventing liver inflammation and lipid accumulation.
Furthermore, two kinds of sphingolipids displayed significant but opposite changes in GCK knockdown HepG2 cells. The levels of several Cers were elevated, while the overall expressions of GM3 were reduced in the GCK knockdown group. It is notable that various sphingolipids were associated with lipotoxicity and inflammation, and were elevated in animal and human NAFLD and diabetes [24,25,26]. In this study, the elevation in Cers and the reduction in GM3 in hepatocytes may offset each other’s effects on lipotoxicity and inflammation, which explains the lower overall inflammation level in GCK inactivation cells.
In addition, several key enzymes involved in lipid metabolic pathways were examined in HepG2 and AML-12 cell lines. FASN and ACC are rate-limiting enzymes responsible for de novo lipogenesis in the liver [27,28]. Hepatic FASN and ACC proteins were decreased in the GCK knockdown group, which might inhibit de novo lipogenesis, thus contributing to reduced palmitic acids and TAG levels. PPARα and its downstream CPT-1 promoted FA β-oxidation in the liver [29,30]. Herein, GCK inactivation significantly upregulated the expression of PPARα and CPT-1. Additionally, ATGL and CHPT-1,which catalyzes the hydrolysis of TAG into DAG [31] and mediates PC synthesis from DAG precursors [32], were found both upregulated in GCK knockdown HepG2 cells, thus inducing lipolysis of TAG and facilitating the downstream biosynthesis of PCs. These enzymes were also examined in siRNA-induced mouse AML-12 cells, and similar alterations in the protein levels were found, except for ATGL, which remained unaffected in low- and high-dose groups, which probably suggested distinct metabolism regulations between human and mouse [33], and further investigation on animal models is needed.
This study also has several limitations. In this study, we used in vitro cell models to investigate the role of GCK inactivation in liver lipid metabolism; therefore, further in vivo experiments are needed. In addition to the liver, the effects of GCK inactivation on lipid metabolism should also be validated in other tissues, including serum and adipose tissue. Additionally, this study focuses on the partial inactivation of GCK; as there are more than 600 loss-of-function mutations of GCK [34], further studies on GCK point-mutation in cells or animal models are needed to determine whether the findings of the present study are common to different mutations of GCK.
In conclusion, partial inactivation of GCK ameliorated hepatic lipid accumulation and inflammation by altering the expressions of hepatic genes involved in lipogenesis, lipolysis and β-oxidation in HepG2 and AML-12 cell models. This finding proved that reduced GCK activity optimized hepatic lipid metabolism, providing a novel mechanism for the favorable lipid profile and low cardiovascular risks in GCK-MODY patients. Glucokinase inactivation could be a potential strategy for the prevention of diabetes-related vascular complications. Further investigations are required to explore the protective and curative effects of GCK inactivation on dyslipidemia and cardiovascular complications in diabetic and nondiabetic populations.
## 4.1. Study Population and Data Collection
This study cohort comprises GCK-MODY ($$n = 33$$), T1D ($$n = 34$$), T2D ($$n = 34$$) and healthy individuals ($$n = 30$$). All participants were recruited from the outpatient clinic and inpatient ward of the endocrinology department at the Peking Union Medical College Hospital (PUMCH), Beijing, China, between January 2017 and December 2021. Demographic information and laboratory tests were collected. The study protocol was approved by the ethical standards of the Peking Union Medical College Hospital Ethics Committee and written consent was provided from all participants.
## 4.2. Cell Culture and High Fatty Acid Treatment
Human hepatocellular carcinoma cell line (HepG2) and alpha mouse liver 12 cell line (AML-12) were obtained from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China). The HepG2 cells were cultured in DMEM medium with $10\%$ fetal bovine serum (FBS) and $1\%$ penicillin–streptomycin (100 U/mL penicillin 10 ug/mL streptomycin) and maintained at 37 °C with $5\%$ CO2. The AML-12 cells were grown in DMEM/F12 (1:1) supplemented with 0.005 mg/mL insulin, 0.005 mg/mL transferrin, 0.005 mg/mL selenium, $10\%$ FBS and 40 ng/mL dexamethasone (Gibco, San Diego, CA, USA) at 37 °C with $5\%$ CO2. For high fatty acid (HFA) treatment, cells were incubated with oleic and palmitic acid in the ratio 2:1 (500 μM/250 μM) or vehicle for 24 h or 48 h.
## 4.3. Lentivirus Transfection
Lentivirus-mediated GCK knockdown (hU6-GCK-CBh-gcGFP-IRES-puro, GV493) and control constructs were synthesized by Genechem (Shanghai, China). The GCK-KD group was transfected with GCK knockdown lentivirus, and the control group was transfected with empty lentivirus. HepG2 cells were cultured in 6-well plates (1 × 106/well). When the confluency reached about $60\%$ (24 h), HepG2 cells were transfected with the constructed human GCK knockdown lentivirus or GFP-expressing control vector (Genechem, Shanghai, China) at a multiplicity of infection (MOI) of 10, with 40 μL/mL infection enhancer HitransG P (Genechem, Shanghai, China) in the medium. The medium containing lentivirus was replaced with fresh medium after 12–16 h. Subsequently, the GCK-KD group were selected with 2 μg/mL puromycin for 72 h.
## 4.4. SiRNA Transfection
For the transient knockdown of GCK in AML-12 cells, mouse GCK-siRNA (GCTCAGAAGTTGGAGACTT) and negative control-siRNA were designed and synthesized by RIBOBIO Co., Ltd. (Guangzhou, China). The transfection of siRNA was facilitated by Lipofectamine 2000 reagent (Invitrogen, Waltham, MA, USA) with siRNA: Lipo2000 = 100 pmol:5 μL for each well of 6-well plates. Two concentration gradients of GCK-siRNA were used [100 nM (200 pmol siRNA) and 200 nM (400 pmol siRNA)] and treated for 24 h. Transfection efficiencies of GCK were confirmed with Western blot and enzyme activity examination.
## 4.5. Oil Red O Staining and Intracellular TAG Levels
The HepG2 cells were plated in 6-well plates. After the cells were fused to 40–$60\%$, the HepG2 cells were treated with FFA as described above. After 48 h, the cells were washed three times with PBS and fixed with $4\%$ paraformaldehyde for 30 min at room temperature. The fixed cells were washed gently with PBS and immersed in $60\%$ isopropanol for 5 min. Then, we removed the isopropanol and stained the cell with Oil Red O solution for 20 min and Mayer’s Hematoxylin Stain solution for 1 min. The excess dye was removed, and the cells were washed four times with distilled water before the microscopic observation under the bright field.
For intracellular TAG levels, after treatment with FFA for 48 h, the HepG2 cells were harvested and lysed to prepare cell lysates. Intracellular TG levels were measured using a triglyceride quantification kit (MICHY Biology, Suzhou, China) following the manufacturer’s instructions. The protein contents in the lysate were determined using the bicinchoninic acid kit (Invitrogen, Waltham, MA, USA). The output optical density was read immediately using a microplate reader at the wavelength of 505 nm. The TG content was measured as mg/mg protein.
## 4.6. Lipidomic Analysis
The lipid profiling in GCK knockdown HepG2 cells was further investigated using lipidomic analysis. As previously reported [35,36], for each sample, 480 μL of extracting solution (MTBE: MEOH = 5:1) was added for metabolites extraction. The samples were centrifuged, and the supernatants were analyzed by LC/MS. LC-MS/MS analyses were performed using an UHPLC system (Vanquish, Thermo Fisher Scientific) with a UPLC HSS T3 column (2.1 mm × 100 mm, 1.8 μm) coupled to Q Exactive HFX mass spectrometer (Orbitrap MS, Thermo) in both positive and negative electrospray ionization models. The QE mass spectrometer was used for its ability to acquire MS/MS spectra on data-dependent acquisition (DDA) mode in the control of the acquisition software (Xcalibur 4.0.27, Thermo).
## 4.7. Data Processing
The raw data files were converted to files in mzXML format using the ‘msconvert’ program from ProteoWizard. The CentWave algorithm in XCMS [37] was used for peak detection, extraction, alignment and integration, the minfrac for annotation was set at 0.5 and the cutoff for annotation was set at 0.3. Lipid identification was achieved through a spectral match using LipidBlast library, which was developed using R and based on XCMS. Orthogonal projections to latent structures discriminant analysis (OPLS-DA) were performed to identify the source of variation between groups. A permutation test repeated 200 times was conducted to ensure the model without overfitting. Variable importance for the projection (VIP) values exceeding 1.0 and p values of Kruskal–Wallis tests or Student’s t test ($p \leq 0.05$) were selected as discriminated metabolites. Correlations between lipids were analyzed by Pearson’s correlation. For the meta-analysis, the peak intensity data were converted using the Z-score transformation to represent metabolite abundance.
## 4.8. GCK Enzyme Activity Determination
The enzyme activities of GCK in cells were assessed by using commercial assay kits (Abcam, Cambridge, UK) according to the manufacturers’ instructions. GCK converts glucose into glucose-6-phosphate and produces a series of intermediates (NADPH) which could be detected by the probe, generating an intense fluorescence product (Ex/Em = $\frac{535}{587}$ nm). Briefly, the cell lysate was diluted by GCK assay buffer (Tris-HCl buffer, pH 8.0). The reaction medium included Tris-HCl, pH 7.4, MgCl2, dithiothreitol, $0.1\%$ bovine serum albumin, KCl, glucose, nicotinamide adenine dinucleotide phosphate, glucose-6-phosphate dehydrogenase and probe for NADPH. The definition of one unit of glucokinase activity is the amount of enzyme that catalyzes the release of 1.0 µmol of NADPH per minute at pH 8.0 and room temperature. The fluorescence was measured by a microplate reader (Biotek SynergyNeo2, BioTek, Vermont, VT, USA).
## 4.9. Enzyme-Linked Immunosorbent Assay (ELISA)
The supernatant of control and GCK knockdown cells was harvested after 48 h of HFA treatment and stored at −80 °C after centrifugation. The levels of Interleukin 1β (IL-1β), Interleukin 6 (IL-6) and monocyte chemotactic protein 1 (MCP-1) in supernatant were detected by ELISA kits (MULTI Science, Hangzhou, China) according to the manufacturers’ protocols.
## 4.10. Western Blotting
Cultured cells were harvested and lysed with RIPA containing $1\%$ PMSF and phosphatase inhibitors for 30 min. Total protein concentrations were determined by the bicinchoninic acid kit (Invitrogen, USA) according to the manufacturer’s instructions. Equal amounts of total protein lysates were separated by SDS-PAGE and transferred to a PVDF membrane and blocked with $5\%$ nonfat dry milk. The membranes were incubated overnight at 4 °C with the primary antibodies (Abcam, Cambridge, UK). The membranes were washed with TBST and incubated with an HRP-conjugated secondary antibody for 90 min. The blots were visualized using an enhanced chemiluminescence detection system.
## 4.11. Statistical Analysis
Statistical analysis was performed using the GraphPad Prism software (version 8.0.2, San Diego, CA, USA). Continuous variables were presented as mean ± SD or median (interquartile range), as appropriate. For two groups, an unpaired two-tailed t-test was performed for intergroup comparisons. For more than two groups, one-way analysis of variance (ANOVA) was used to assess statistically significant differences. Differences between groups were considered statistically significant when p values ≤ 0.05.
## References
1. Matschinsky F.M., Magnuson M.A., Zelent D., Jetton T.L., Doliba N., Han Y., Taub R., Grimsby J.. **The network of glucokinase-expressing cells in glucose homeostasis and the potential of glucokinase activators for diabetes therapy**. *Diabetes* (2006) **55** 1-12. DOI: 10.2337/diabetes.55.01.06.db05-0926
2. Iynedjian P.B.. **Molecular physiology of mammalian glucokinase**. *Cell Mol. Life Sci.* (2009) **66** 27-42. DOI: 10.1007/s00018-008-8322-9
3. Bishay R.H., Greenfield J.R.. **A review of maturity onset diabetes of the young (MODY) and challenges in the management of glucokinase-MODY**. *Med. J. Aust.* (2016) **205** 480-485. DOI: 10.5694/mja16.00458
4. Anik A., Catli G., Abaci A., Bober E.. **Maturity-onset diabetes of the young (MODY): An update**. *J. Pediatr. Endocrinol. Metab.* (2015) **28** 251-263. DOI: 10.1515/jpem-2014-0384
5. Steele A.M., Shields B.M., Wensley K.J., Colclough K., Ellard S., Hattersley A.T.. **Prevalence of vascular complications among patients with glucokinase mutations and prolonged, mild hyperglycemia**. *JAMA* (2014) **311** 279-286. DOI: 10.1001/jama.2013.283980
6. Spégel P., Ekholm E., Tuomi T., Groop L., Mulder H., Filipsson K.. **Metabolite profiling reveals normal metabolic control in carriers of mutations in the glucokinase gene (MODY2)**. *Diabetes* (2013) **62** 653-661. DOI: 10.2337/db12-0827
7. Wang X., Lam S.M., Cao M., Wang T., Wang Z., Yu M., Li B., Zhang H., Ping F., Song G.. **Localized increases in CEPT1 and ATGL elevate plasmalogen phosphatidylcholines in HDLs contributing to atheroprotective lipid profiles in hyperglycemic GCK-MODY**. *Redox Biol.* (2021) **40**. DOI: 10.1016/j.redox.2021.101855
8. Eid S., Sas K.M., Abcouwer S.F., Feldman E.L., Gardner T.W., Pennathur S., Fort P.E.. **New insights into the mechanisms of diabetic complications: Role of lipids and lipid metabolism**. *Diabetologia* (2019) **62** 1539-1549. DOI: 10.1007/s00125-019-4959-1
9. Agius L.. **Lessons from glucokinase activators: The problem of declining efficacy**. *Expert. Opin. Ther. Pat.* (2014) **24** 1155-1159. DOI: 10.1517/13543776.2014.965680
10. Meininger G.E., Scott R., Alba M., Shentu Y., Luo E., Amin H., Davies M.J., Kaufman K.D., Goldstein B.J.. **Effects of MK-0941, a novel glucokinase activator, on glycemic control in insulin-treated patients with type 2 diabetes**. *Diabetes Care* (2011) **34** 2560-2566. DOI: 10.2337/dc11-1200
11. Katz L., Manamley N., Snyder W.J., Dodds M., Agafonova N., Sierra-Johnson J., Cruz M., Kaur P., Mudaliar S., Raskin P.. **AMG 151 (ARRY-403), a novel glucokinase activator, decreases fasting and postprandial glycaemia in patients with type 2 diabetes**. *Diabetes Obes. Metab.* (2016) **18** 191-195. DOI: 10.1111/dom.12586
12. De Ceuninck F., Kargar C., Ilic C., Caliez A., Rolin J.O., Umbdenstock T., Vinson C., Combettes M., de Fanti B., Harley E.. **Small molecule glucokinase activators disturb lipid homeostasis and induce fatty liver in rodents: A warning for therapeutic applications in humans**. *Br. J. Pharmacol.* (2013) **168** 339-353. DOI: 10.1111/j.1476-5381.2012.02184.x
13. Kawata S., Nakamura A., Miyoshi H., Yang K., Shigesawa I., Yamauchi Y., Tsuchida K., Omori K., Takahashi K., Nomoto H.. **Glucokinase activation leads to an unsustained hypoglycaemic effect with hepatic triglyceride accumulation in db/db mice**. *Diabetes Obes. Metab* (2022) **24** 391-401. DOI: 10.1111/dom.14586
14. Palomer X., Pizarro-Delgado J., Barroso E., Vázquez-Carrera M.. **Palmitic and Oleic Acid: The Yin and Yang of Fatty Acids in Type 2 Diabetes Mellitus**. *Trends Endocrinol. Metab.* (2018) **29** 178-190. DOI: 10.1016/j.tem.2017.11.009
15. Joshi-Barve S., Barve S.S., Amancherla K., Gobejishvili L., Hill D., Cave M., Hote P., McClain C.J.. **Palmitic acid induces production of proinflammatory cytokine interleukin-8 from hepatocytes**. *Hepatology* (2007) **46** 823-830. DOI: 10.1002/hep.21752
16. Zhao N., Tan H., Wang L., Han L., Cheng Y., Feng Y., Li T., Liu X.. **Palmitate induces fat accumulation via repressing FoxO1-mediated ATGL-dependent lipolysis in HepG2 hepatocytes**. *PLoS ONE* (2021) **16**. DOI: 10.1371/journal.pone.0243938
17. Afshinnia F., Rajendiran T.M., Soni T., Byun J., Wernisch S., Sas K.M., Hawkins J., Bellovich K., Gipson D., Michailidis G.. **Impaired -Oxidation and Altered Complex Lipid Fatty Acid Partitioning with Advancing CKD**. *J. Am. Soc. Nephrol.* (2018) **29** 295-306. DOI: 10.1681/ASN.2017030350
18. Rodriguez-Cuenca S., Pellegrinelli V., Campbell M., Oresic M., Vidal-Puig A.. **Sphingolipids and glycerophospholipids—The “ying and yang” of lipotoxicity in metabolic diseases**. *Prog. Lipid. Res.* (2017) **66** 14-29. DOI: 10.1016/j.plipres.2017.01.002
19. McMaster C.R.. **From yeast to humans—Roles of the Kennedy pathway for phosphatidylcholine synthesis**. *FEBS Lett.* (2018) **592** 1256-1272. DOI: 10.1002/1873-3468.12919
20. Cole L.K., Vance J.E., Vance D.E.. **Phosphatidylcholine biosynthesis and lipoprotein metabolism**. *Biochim. Biophys. Acta* (2012) **1821** 754-761. DOI: 10.1016/j.bbalip.2011.09.009
21. Maeba R., Kojima K.-I., Nagura M., Komori A., Nishimukai M., Okazaki T., Uchida S.. **Association of cholesterol efflux capacity with plasmalogen levels of high-density lipoprotein: A cross-sectional study in chronic kidney disease patients**. *Atherosclerosis* (2018) **270** 102-109. DOI: 10.1016/j.atherosclerosis.2018.01.037
22. Cao P., Wu Y., Li Y., Xiang L., Cheng B., Hu Y., Jiang X., Wang Z., Wu S., Si L.. **The important role of glycerophospholipid metabolism in the protective effects of polyphenol-enriched Tartary buckwheat extract against alcoholic liver disease**. *Food Funct.* (2022) **13** 10415-10425. DOI: 10.1039/D2FO01518H
23. Piras I.S., Raju A., Don J., Schork N.J., Gerhard G.S., DiStefano J.K.. **Hepatic PEMT Expression Decreases with Increasing NAFLD Severity**. *Int. J. Mol. Sci.* (2022) **23**. DOI: 10.3390/ijms23169296
24. Luukkonen P.K., Zhou Y., Sädevirta S., Leivonen M., Arola J., Orešič M., Hyötyläinen T., Yki-Järvinen H.. **Hepatic ceramides dissociate steatosis and insulin resistance in patients with non-alcoholic fatty liver disease**. *J. Hepatol.* (2016) **64** 1167-1175. DOI: 10.1016/j.jhep.2016.01.002
25. Puri P., Baillie R.A., Wiest M.M., Mirshahi F., Choudhury J., Cheung O., Sargeant C., Contos M.J., Sanyal A.J.. **A lipidomic analysis of nonalcoholic fatty liver disease**. *Hepatology* (2007) **46** 1081-1090. DOI: 10.1002/hep.21763
26. Summers S.A.. **Editorial: The Role of Ceramides in Diabetes and Cardiovascular Disease**. *Front. Endocrinol.* (2021) **12**. DOI: 10.3389/fendo.2021.667885
27. Ou-Yang Q., Lin X.-M., Zhu Y.-J., Zheng B., Li L., Yang Y.-C., Hou G.-J., Chen X., Luo G.-J., Huo F.. **Distinct role of nuclear receptor corepressor 1 regulated de novo fatty acids synthesis in liver regeneration and hepatocarcinogenesis in mice**. *Hepatology* (2018) **67** 1071-1087. DOI: 10.1002/hep.29562
28. Pang Y., Xu X., Xiang X., Li Y., Zhao Z., Li J., Gao S., Liu Q., Mai K., Ai Q.. **High Fat Activates O-GlcNAcylation and Affects AMPK/ACC Pathway to Regulate Lipid Metabolism**. *Nutrients* (2021) **13**. DOI: 10.3390/nu13061740
29. Naiman S., Huynh F.K., Gil R., Glick Y., Shahar Y., Touitou N., Nahum L., Avivi M.Y., Roichman A., Kanfi Y.. **SIRT6 Promotes Hepatic Beta-Oxidation via Activation of PPARα**. *Cell Rep.* (2019) **29** 4127-4143e8. DOI: 10.1016/j.celrep.2019.11.067
30. Xu Y.-H., Luo Z., Wu K., Fan Y.-F., You W.-J., Zhang L.-H.. **Structure and Functional Analysis of Promoters from Two Liver Isoforms of CPT I in Grass Carp Ctenopharyngodon idella**. *Int. J. Mol. Sci.* (2017) **18**. DOI: 10.3390/ijms18112405
31. Kim S.-J., Tang T., Abbott M., Viscarra J.A., Wang Y., Sul H.S.. **AMPK Phosphorylates Desnutrin/ATGL and Hormone-Sensitive Lipase To Regulate Lipolysis and Fatty Acid Oxidation within Adipose Tissue**. *Mol Cell Biol.* (2016) **36** 1961-1976. DOI: 10.1128/MCB.00244-16
32. McMaster C.R., Bell R.M.. **CDP-choline:1,2-diacylglycerol cholinephosphotransferase**. *Biochim. Biophys. Acta* (1997) **1348** 100-110. DOI: 10.1016/S0005-2760(97)00097-0
33. Kawana M., Miyamoto M., Ohno Y., Kihara A.. **Comparative profiling and comprehensive quantification of stratum corneum ceramides in humans and mice by LC/MS/MS**. *J. Lipid. Res.* (2020) **61** 884-895. DOI: 10.1194/jlr.RA120000671
34. Osbak K.K., Colclough K., Saint-Martin C., Beer N.L., Bellanné-Chantelot C., Ellard S., Gloyn A.L.. **Update on mutations in glucokinase (GCK), which cause maturity-onset diabetes of the young, permanent neonatal diabetes, and hyperinsulinemic hypoglycemia**. *Hum. Mutat.* (2009) **30** 1512-1526. DOI: 10.1002/humu.21110
35. Dunn W.B., Broadhurst D., Begley P., Zelena E., Francis-McIntyre S., Anderson N., Brown M., Knowles J.D., Halsall A., Haselden J.N.. **Procedures for large-scale metabolic profiling of serum and plasma using gas chromatography and liquid chromatography coupled to mass spectrometry**. *Nat. Protoc.* (2011) **6** 1060-1083. DOI: 10.1038/nprot.2011.335
36. Want E.J., Wilson I.D., Gika H., Theodoridis G., Plumb R.S., Shockcor J., Holmes E., Nicholson J.K.. **Global metabolic profiling procedures for urine using UPLC-MS**. *Nat. Protoc.* (2010) **5** 1005-1018. DOI: 10.1038/nprot.2010.50
37. Smith C.A., Want E.J., O’Maille G., Abagyan R., Siuzdak G.. **XCMS: Processing mass spectrometry data for metabolite profiling using nonlinear peak alignment, matching, and identification**. *Anal. Chem.* (2006) **78** 779-787. DOI: 10.1021/ac051437y
|
---
title: Induction of Angiogenesis by Genetically Modified Human Umbilical Cord Blood
Mononuclear Cells
authors:
- Dilara Z. Gatina
- Ilnaz M. Gazizov
- Margarita N. Zhuravleva
- Svetlana S. Arkhipova
- Maria A. Golubenko
- Marina O. Gomzikova
- Ekaterina E. Garanina
- Rustem R. Islamov
- Albert A. Rizvanov
- Ilnur I. Salafutdinov
journal: International Journal of Molecular Sciences
year: 2023
pmcid: PMC10002409
doi: 10.3390/ijms24054396
license: CC BY 4.0
---
# Induction of Angiogenesis by Genetically Modified Human Umbilical Cord Blood Mononuclear Cells
## Abstract
Stimulating the process of angiogenesis in treating ischemia-related diseases is an urgent task for modern medicine, which can be achieved through the use of different cell types. Umbilical cord blood (UCB) continues to be one of the attractive cell sources for transplantation. The goal of this study was to investigate the role and therapeutic potential of gene-engineered umbilical cord blood mononuclear cells (UCB-MC) as a forward-looking strategy for the activation of angiogenesis. Adenovirus constructs Ad-VEGF, Ad-FGF2, Ad-SDF1α, and Ad-EGFP were synthesized and used for cell modification. UCB-MCs were isolated from UCB and transduced with adenoviral vectors. As part of our in vitro experiments, we evaluated the efficiency of transfection, the expression of recombinant genes, and the secretome profile. Later, we applied an in vivo Matrigel plug assay to assess engineered UCB-MC’s angiogenic potential. We conclude that hUCB-MCs can be efficiently modified simultaneously with several adenoviral vectors. Modified UCB-MCs overexpress recombinant genes and proteins. Genetic modification of cells with recombinant adenoviruses does not affect the profile of secreted pro- and anti-inflammatory cytokines, chemokines, and growth factors, except for an increase in the synthesis of recombinant proteins. hUCB-MCs genetically modified with therapeutic genes induced the formation of new vessels. An increase in the expression of endothelial cells marker (CD31) was revealed, which correlated with the data of visual examination and histological analysis. The present study demonstrates that gene-engineered UCB-MC can be used to stimulate angiogenesis and possibly treat cardiovascular disease and diabetic cardiomyopathy.
## 1. Introduction
Angiogenesis is the growth of new blood vessels from pre-existing vessels, an essential process for development, wound healing, and the restoration of blood flow and oxygen supply to tissues after injury. One of the main tasks of modern medicine is the stimulation of the processes of angiogenesis in the treatment of vascular diseases. To date, many approaches have been proposed for the induction of therapeutic angiogenesis. Among the proposed methods are surgical methods [1], the use of inducer proteins [2], recombinant DNA molecules [3], inducer genes [4], and the use of various cell types [5,6], including ex vivo genetically modified cells [7]. In this aspect, human umbilical cord blood mononuclear cells (UCB-MC) seem to be a promising “tool” for stimulating angiogenesis through the delivery of genetic engineering systems, expression of recombinant proteins, and possibly direct participation in new vessel formation. The choice of UCB-MC in studies for cell and gene-cell therapy looks promising because of some advantages of this cellular source. Umbilical cord blood contains many stem/progenitor cells and can be obtained easily [8]. The mononuclear fraction of UCB contains populations of different immature cells capable of differentiating into many cell types [9]. Cell populations that have been discovered in UCB are hematopoietic stem cells (HSCs), endothelial progenitor cells, mesenchymal stem cells (MSCs), unrestricted somatic stem cells (USSCs), and side population cells [10,11,12,13]. As cellular material for transplantation or carriers for genetic constructs, UCB-MCs have low immunogenicity because they do not express all the antigens on the cell membrane. This feature enhances the ability to cross donor-recipient HLA disparities. It allows for the usage of UCB-MC for transplantation in non-fully compatible HLA recipients with a much lower incidence of grade II-IV acute GVHD (graft versus host disease) cases after transplantation [14,15,16,17]. Furthermore, UCB-MCs can prevent the oncological transformation of recipient cells after transplantation [15]. Another appealing reason for using UCB cells for cell therapy is their ability to produce various biologically active molecules, such as proteins with antioxidant properties, angiogenic, neurotrophic, and growth factors [18,19,20,21,22], which make them suitable for effective stimulation of regenerative processes in non-compatible recipients for a short time before the immune system eliminates them. Overall, UCB cell transplantation can replace dead cells, prevent further death of surviving cells, and stimulate regeneration by secreting biologically active molecules. A genetic modification of UCB cells can enhance their ability to regenerate tissue [23,24]. This approach unites the advantages of cell- and gene therapy. Genetically modified UCB cells can provide targeted delivery of therapeutic genes and expression of recombinant molecules at the regeneration site. For example, our previous studies showed the positive effect of genetically modified umbilical cord blood mononuclear cells (UCB-MC) simultaneously produces three recombinant molecules (vascular endothelial growth factor (VEGF), glial cell-derived neurotrophic factor (GDNF), and neural cell adhesion molecule (NCAM) in animal models of amyotrophic lateral sclerosis [25], spinal cord injury [26] and stroke [7,27]. Many state-of-the-art methods and models for studying angiogenesis have been proposed, which are well analyzed in the review articles [28,29,30]. Among various models, the in vivo angiogenesis plug assay, which uses basement membrane extracts (BME) or Matrigel, is widely used for evaluating pro- or anti-angiogenic factors during in vivo angiogenesis. This assay is reliable, easy to perform without special equipment, reproducible, quantitative, and quick [31,32,33]. Matrigel predominantly contains laminin III, collagen IV, heparan sulfate proteoglycans, and various growth factors. The assay is performed by injecting the liquid Matrigel into the subcutaneous space of an animal at 4 °C, which solidifies to form a plug at body temperature. Over time, blood vessels sprout into the plug. The number of plug sites per animal can be several, allowing multiple test compounds or concentrations to be tested. Thus, drug screening can also be evaluated for effects on the activity of angiogenic or anti-angiogenic factors [34,35,36,37]. The drug can be placed in the plug with the test factor by mixing with the Matrigel matrix or given to the host animal. Cells or exosomes can also be examined when mixed into the gel to produce angiogenic factors. Furthermore, the assay is highly versatile. For example, the role of certain genes can be evaluated using genetically modified mice (overexpressing or ablating a protein gene) or animal models of diseases. This report aimed to study the effect of genetically modified umbilical cord blood mononuclear cells overexpressing recombinant proangiogenic proteins VEGF165, FGF2, and SDF1α on the induction of angiogenesis. Furthermore, we assessed the influence of all three factors on the tone of the secretory profile of modified UCB-MCs and tubule formation in the in vivo Matrigel plug assay. The present study shows that when combined with UCB, the three factors can enhance angiogenesis and be useful for developing new therapies.
## 2.1. Characterization of Isolated Human UCB-MCs
Isolated cells demonstrated high viability (>$97\%$) and included CD45+ lymphocytes ($58.9\%$). CD45+CD3+ lymphocytes constituted $59.2\%$, while CD14+ macrophages constituted $7.3\%$. This ratio of the central populations of blood cells (lymphocytes, T-lymphocytes, and monocytes) is believed to be typical for human UCB-MCs. We also examined the percentage of CD34+ blood cells among isolated UCB-MCs. According to the obtained data, CD34+ cells constituted $0.4\%$ of CD45+ cells. In addition, $91.8\%$ of CD45+CD34+ cells expressed CD38. Furthermore, $90\%$ of the CD45+CD34+ cells had the phenotype CD90+. The flow cytometry results are shown in Figure 1.
Immunophenotyping of a pool of CD34-positive cells showed that genetic modification and expression of recombinant factors by cells did not affect the viability and endothelial cell markers (Figure 2).
## 2.2. Transduction of UC-MCs with Recombinant Adenoviruses Increased Transgene Expression
It has been demonstrated that genetic modification of the UCB-MCs with recombinant adenoviruses (Ad-VEGF, Ad-FGF2, Ad-SDF1α, or Ad-EGFP) did not affect cell viability. Moreover, it has been shown that UCB-MCs transduced with Ad-EGFP exhibited green fluorescence, confirming the efficiency of transduction (Figure 3A). Furthermore, EGFP expression was sustained for 30 days after a genetic modification of UCB-MCs. According to the flow cytometry results, EGFP+ cells constituted 28 ± $2.7\%$ (Figure 3B). Analysis of the mRNA expression of VEGF165, FGF2, and SDF1α in genetically modified human UCB-MCs was carried out using qPCR. It has been established that genetic modification of hUCB-MCs with Ad-VEGF165 results in augmented VEGF expression (190.6 ± 8.9 fold). Simultaneous transduction with Ad-VEGF165, Ad-FGF2, and Ad-SDF1α resulted in the upregulation of VEGF, FGF2, and SDF1α expression (198.6 ± 0.45; 204.2 ± 0.36 and 140.9 ± 0.32 fold respectively) compared to non-transfected cells, and cells modified with Ad-EGFP (Figure 3C). The obtained results are evident for efficient modification of hUCB-MCs with developed genetic constructs which provide a synthesis of target genes in vitro.
## 2.3. Genetically Modified hUCB-MCs Produce a Broad Range of Cytokines, Chemokines, and Growth Factors
A complete analysis of all cytokines and chemokines measured in the Luminex assays demonstrated that gene modification and gene expression did not change levels of multiple anti and proinflammatory cytokines as well as chemokines. The results obtained from the eight donors in comparison to the untreated control are shown in Table 1 (Supplementary Table S1). We have not observed any statistically significant differences in cytokine and chemokine secretion between the groups of non-transfected cells and genetically modified ones except for upregulated levels of recombinant proteins in corresponding groups. Multiplex analysis revealed statistically significant ($p \leq 0.05$) upregulation of VEGF secretion (1087.12 ± 169.11 pg/mL) in UCB-MCs modified with Ad-VEGF compared to the UCB-MCs treated with Ad-EGFP (52.31 ± 10.36 pg/mL) and non-treated cells (51.75 ± 8.65 pg/mL). Simultaneous transduction with Ad-VEGF, Ad-FGF2, and Ad-SDF1 has resulted in the increased production of VEGF (701.94 ± 96.99 pg/mL), FGF2 (576.27 ± 57.83 pg/mL), and SDF1α (622.39 ± 113.07 pg/mL) (Figure 3D). Obtained results correlate with the data presented above of RT-qPCR and confirm the capacity of recombinant adenoviruses for infection of target cells.
It is also worth emphasizing that the UCB-MC-VEGF-FGF2-SDF1 and UCB-MC-VEGF did not differ from UCB-MC-EGFP and UCB-MC-NTC in vitro studies. What can be seen from the data of morphological, and phenotypic studies are the profiles of secreted factors. Therefore, UCB-MC-EGFP is the ideal control in our study in vivo.
## 2.4. Transplantation of Genetically Modified Cells Promotes Angiogenesis In Vivo
Matrigel mixtures were implanted into the subcutaneous space of the dorsal region in mice after seven days post-transplantation when implanted Matrigel samples containing genetically modified UCB-MCs were extracted from Balb/c nude mice. Embedded fragments represented discs with $d = 10$ mm and 2 mm height. The color of the implants correlated with vascularization density. The color of the implants varied from milky-white (Matrigel without cells and Matrigel with UCB-MC + Ad-EGFP) to red-brown (Matrigel with UCB-MC + Ad-VEGF165 + Ad-FGF2 + Ad-SDF1α) which is due to the vascular formation and presence of blood cells, particularly, erythrocytes (Figure 4A). Gross histological hematoxylin and eosin (H&E) staining of extracted plugs showed the absence of inflammatory sites. The skin and subcutaneous tissue in the area of implantation were not visually changed (Figure 4B). We have established that in isolated subcutaneous implants containing hUCB-MC, human-transduced Ad-VEGF165, or a combination of Ad-VEGF165, Ad-SDF1α, and Ad-FGF2, the hemoglobin concentration was significantly higher in comparison with Matrigel fragments without cell administration and implants with UCB-MCs transduced Ad-EGFP. Moreover, the significantly higher concentration of hemoglobin was determined in the samples containing UCB-MCs transduced with Ad-VEGF165, Ad-SDF1α, and Ad-FGF2 compared to the group with UCB-MCs transduced with single Ad-VEGF165 (Figure 4C). Moreover, we observed a two-fold increase of mCD31 mRNA expression in plugs containing hUCB-MC transduced Ad-VEGF165 or a combination of Ad-VEGF165, Ad-SDF1α, and Ad-FGF2 compared to controls. Moreover, we did not discover the difference between Ad-VEGF165 and the group with a mixture of Ad-VEGF165, Ad-SDF1α, and Ad-FGF2.
Analysis of the mRNA expression of VEGF165, FGF2, and SDF1α in genetically modified UCB-MCs in Matrigel implants was evaluated by RT-qPCR. Notably, obtained results confirmed the expression of target genes in genetically modified UCB-MCs implanted in Matrigel even at one-week post-transplantation. We discovered that the Matrigel complexes containing UCB-MC Ad-VEGF gave rise to more abundant VEGF mRNA than UCB-MC Ad-EGFP and PBS (Matrigel samples without UCB-MCs). Likewise, UCB-MCs contemporaneously transduced with Ad-VEGF, Ad-FGF2, and Ad-SDF1α exhibited upregulated levels of mRNA expression of VEGF, FGF2, and SDF1α. ( Figure 5A). During histological analysis of implants, it has been shown that control—PBS (Matrigel samples without UCB-MCs) contained small amounts of migrated fibroblast-like cells. Visually, the implants were surrounded by a thin connective tissue capsule, which contained rare capillaries in a density of 1.5 ± 0.5 units/mm2. In samples with implanted UCB-MCs transduced with a cocktail of adenoviruses (Ad-VEGF165, Ad-FGF2, and AdSDF1α), Matrigel mass contained a residual amount of VEGF+ cells. These vessels localized close to the capsule and migrated fibroblasts, some of which were positive for SDF1α and FGF2. In Matrigel samples with implanted UCB-MCs genetically modified with Ad-EGFP, we found single and small rounded clusters of EGFP-positive cells and rare migrated fibroblast-like cells. The implants were surrounded by a thin connective tissue capsule, from which strands of connective tissue grew into its depth with capillaries found in a density of 7.5 ± 3 units/mm2. Vessels were located close to the capsule. Fibroblasts that migrated into Matrigel expressed SDF1α and FGF2. Expression of VEGF, FGF2, and SDF1α in the implanted UCB-MCs were not confirmed. In the group with UCB-MCs modified with Ad-VEGF165, implant samples presented Matrigel mass with single small, rounded clusters of VEGF-positive UCB-MCs cells and rare migrated fibroblast-like cells. The implants were surrounded by a thin connective tissue capsule, from which strands of connective tissue grew more profound into the central regions of the implant with capillaries‘ density of 16 ± 5 units/mm2. In the group of UCB-MCs simultaneously transduced with a combination of Ad-VEGF165, Ad-SDF1α and Ad-FGF2, implant samples were represented by the mass of Matrigel with single and small rounded clusters of UCB-MCs, as well as rare migrated fibroblast-like cells. The implant was surrounded by a thin connective tissue capsule, from which the connective tissue and vessels of various calibers grew to the center of the implant with a capillary density of 23 ± 5 units/mm2. Implanted UCB-MCs expressed VEGF, FGF2, and SDF1α (Figure 5B).
## 3. Discussion
Adenoviruses mediate gene transfer into dividing and quiescent cells and can be produced with a significant titer. The high immunogenicity of adenoviruses as vehicles for the delivery of therapeutic genes represents one of the main disadvantages resulting in the activation of the immune response in immune-competent organisms and the absence of expression of the target therapeutic genes [38]. However, this negative effect is eliminated when using an ex vivo gene therapy approach. Moreover, adenoviral systems promote transient transgene expression due to their non-integration into the host cell genome [39]. However, this negative point might become beneficial for gene therapy based on growth factors: induction of angiogenesis does not require the prolonged expression of therapeutic proteins but, more importantly, their synergistic effect [40]. The absence of integration of adenoviruses eliminates the risk of insertional mutagenesis, which is a typical problem when using retroviral vectors [39]. Adenoviral vectors demonstrate comparatively low efficiency of genetic modification of hematopoietic cells, which might be increased with a higher concentration of virus [41] or its specific treatment, resulting in augmented tropism [42]. In the present study, we chose the adenovirus delivery vectors containing VEGF, FGF2, and SDF1α to investigate the angiogenic effect of UCB-MC in vitro and the Matrigel plug assay in Nude mice. In our investigation, cellular carriers expressed phenotype typical for UCB-MCs, and about $30\%$ of the cells were efficiently transduced with an MOI of 10. The transduction efficiency correlated with previous results and other research groups’ data [42,43]. After in vitro transduction, the UCB-MCs expressed the recombinant mRNA of proangiogenic factors in the cytoplasm and secreted those factors into the culture medium, which in our study we confirmed by RT-qPCR and immunological studies. The obtained data correlates with our previously published results [7].
Various approaches were proposed for stimulating therapeutic angiogenesis based on the delivery methods of genetically engineered systems expressing a broad range of proangiogenic factors. The therapeutic efficacy of proangiogenic factors has been proven in numerous experiments on animal models [44] and in several clinical studies [45,46]. The key inducers of angiogenesis, VEGF, FGF2, EGF, SDF1α, and PDGF-BB, are most often used as genetic components [2]. In particular, VEGF is perhaps the most characterized and frequently used mitogen in creating gene therapy systems and in the induction of therapeutic angiogenesis. VEGF is a crucial participant in forming new blood vessels and can induce the growth of pre-existing ones [47]. However, Zentilin et al. reported that overexpression of VEGF induced leaky neovessels that missed connecting correctly with existing vessels [48]. The FGF family includes vertebrates’ most versatile growth factors that play critical roles in many biological processes, including angiogenesis [49]. FGF, similar to VEGF, is a pleiotropic molecule capable of acting on various cell types, including endodermal, mesenchymal, and neuroectodermal origin cells. It has been shown that FGF2 induces the expression of VEGF and several other factors by endothelial cells through autocrine and paracrine mechanisms [50,51]. SDF1α is a constitutively expressed and inducible chemokine, associated with various physiological and pathological processes, including embryonic development, homeostasis maintenance, and angiogenesis activation [52]. There is evidence that the administration of SDF1α increases blood flow and perfusion via the recruitment of endothelial progenitor cells (EPCs). SDF1α binds exclusively to CXCR4 and has CXCR4 as its only receptor [53]. Compared with the effects of other angiogenic growth factors, SDF1α has unique properties. *The* generation of hyperpermeable vessels, a significant characteristic of VEGF-stimulated angiogenesis, may not be observed after injection of SDF1α contributes to the stabilization of neovessel formation by recruiting CXCR4 + PDGFR+ cKit+ smooth muscle progenitor cells during recovery from vascular injury [54]. Extensive evidence suggests that SDF1α up-regulates VEGF synthesis in several cell types, whereas VEGF and basic FGF induce SDF1α and its receptor CXCR4 in endothelial cells [55]. However, it should also be noted that in a wide range of studies using various models, the mutual synergistic role of VEGF, FGF2, SDF1α, and countless other factors responsible for the formation of vessels has been shown [56,57,58,59]. It is generally known that an optional cellular source for allogenic transplantation should meet the following criteria: it must be less immunogenic and contain a sufficient amount of immature cells capable of differentiation in various directions; it should have a prolonged period of storage and potency for expansion. *Most* gene-cell-mediated therapy protocols intend genetic modification of target cells with different vectors, providing stable expression of target proteins. Human UCB-MCs might be easily isolated and characterized; these cells exhibit low immunogenicity and are composed of unique populations of progenitor cells capable of differentiation into endothelial, muscular, and neural cells, etc. Mononuclear cells from umbilical cord blood are a well-characterized group of cells that are extensively used in pre-clinical and clinical trials of therapy for various human diseases and the induction of therapeutic angiogenesis as well [60]. However, relatively small amounts of UCB-MCs for achieving sufficient therapeutic effect remain the main limitation for its extensive introduction in the clinic [61]. To increase its biological activity, it was proposed to mix different cell pools with further genetic modification [62]. Contemporary cell-mediated approaches to gene therapy suggest UCB-MC as a cell carrier for the delivery of various therapeutic genes. This concept assumes either the differentiation of transplanted cells into different cell types or the realization of therapeutic effects due to the secretion of a broad range of bioactive molecules [63]. Furthermore, our previous study has demonstrated that UCB-MCs are capable of transferring therapeutic genes and promoting evident therapeutic effects using different models, such as ALS [64], SCI [25,26], and stroke [27]. Similar results were obtained in investigations dedicated to therapies for hematologic and non-hematologic disorders [65,66,67,68,69]. At the same time, there is no current data about the influence of the simultaneous transduction of several recombinant adenoviruses on the secretome profile and angiogenic properties of modified hUCB-MCs. A sustained balance of proangiogenic factors and their synergetic effect is essential for functional vascular formation. In the present study, we developed the UCB-MC application to simultaneously deliver many genes (VEGF, FGF2, and SDF1α) to stimulate angiogenesis. Our previous studies also showed the approach of preventive gene therapy with many genes to positively affect stroke. Adenoviral vectors carrying genes encoding vascular VEGF, glial cell-derived neurotrophic factor (GDNF), and NCAM or gene-engineered umbilical cord blood mononuclear cells (UCB-MC) overexpressing recombinant proteins were intrathecally injected before distal occlusion of the middle cerebral artery in the rat. Morphometric and immunofluorescence analysis revealed a reduction in infarction volume and a lower number of apoptotic cells. It decreased the expression of Hsp70 in the peri-infarct region in gene-treated animals [7].
The heterogeneous cell population from the mononuclear fractions UCB-MCs secretes different anti-inflammatory, pro-inflammatory cytokines, chemokines end grow factors [70]. Previously, it was shown that the duration of cultivation, cultivation medium, and the additives used in the culture are the main factors influencing the production of cytokines by UCB-MCs. Our study describes the profile of cytokines and chemokines released by UCB-MC following their in vitro gene modification by adenoviruses. Five groups of secreted factors were investigated: pro-inflammatory cytokines (IL-6, IL-1β, and TNF), an anti-inflammatory cytokine (IL-4 and IL-10), TH1-type cytokines (IL-12 and IFN-γ), chemokines (IL-8, MIP-1α, MIP-1β, and MCP-1) and growth factors (VEGF, FGF2, and SDF1α). Interestingly, the range of cytokine, chemokine, and growth factor concentrations detected in the supernatants of UBC-MC varied between donors, indicating major individual heterogeneity, comparable with previously published data [71].
The highest secretion level by modified and unmodified cells was shown for IL-8 and MCP-1. These factors are known to be produced more intensively than any other chemokines in the human body and are seen as the first line of defense in inflammatory responses [72]. In addition, the cells also secreted high concentrations of GROα, IL-6, MIF, MIP-1α, MIP-1β, and SCGF-β. Unfortunately, adenoviruses are potent activators of the innate and adaptive immune systems. The administration of high doses of Ad-based vectors to animals or patients, primarily through the intravascular pathway, leads to severe immunopathology manifested by cytokine storm syndrome, disseminated intravascular coagulation, thrombocytopenia, and hepatotoxicity, which can lead to morbidity and also death [69]. Research by Teigler et al. on peripheral blood mononuclear cells (PBMCs) showed that their stimulation with the Ad vector increases the secretion of IFN-γ, INF2α, IL-15, G-CSF, MIG, and IP-10. Supporting this perspective, it is worth emphasizing that the study’s authors used 103 vp/cell [73]. Previous studies have shown that treatment of myeloid dendritic cells and plasmacytoid dendritic cells with Ad5 does not lead to an increase in IFN production by them, even at the highest exposed rAd (100 vp/cell) [43]. Our previous examination has shown that genetic modification UCB-MC and expression of transgenes VEGF or EGFP did not influence the global transcriptome landscape [74]. In this study, we demonstrate that a gene-cell system with simultaneous delivery of genes based on UCB-MC can generate the expression of several transgenes both in vitro and in vivo. Furthermore, the UCB-MC-VEGF165 and UCB-MC-VEGF-FGF2-SDF1α Matrigel plugs in mice were filled with red blood cells and showed vessel-like structure formation. We did not find significant differences between the UCB-MC-VEGF and UCB-MC-VEGF-FGF2-SDF1α groups in the present study. Although in line with the RT-qPCR data and immunology tests, levels of expression of VEGF, SDF1α, and FGF varied. Perhaps this is because we used a small amount of cellular material and a short exposure period to Matrigel fragments.
Furthermore, the UCB-MC-VEGF165 and UCB-MC-VEGF-FGF2-SDF1α Matrigel plugs in mice were filled with red blood cells and showed vessel-like structure formation. We did not find significant differences between the UCB-MC-VEGF165 and UCB-MC-VEGF-FGF2-SDF1α groups in the present study. Although in line with the RT-qPCR data and immunology tests, levels of expression of VEGF, SDF1α, and FGF varied. Perhaps this is because we used a small amount of cellular material and a short exposure period to Matrigel fragments.
## 4.1. Obtaining Recombinant Adenovirus Ad-SDF1α
The creation of expression constructs based on adenovirus was carried out by using molecular cloning methods of Gateway-cloning technology (Invitrogen), as described previously [75]. Briefly, subcloning of SDF1α from the plasmid vector pBud-VEGF-SDF1α into the intermediate vector pDONR221 was performed [76].
## 4.2. The Production of Recombinant Adenoviruses
The HEK293A cells were infected with a coarse viral runoff to prepare the necessary amounts of Ad-VEGF, Ad-FGF2, Ad-SDF1α, and Ad-EGFP adenoviruses. To purify viral particles from cell debris, supernatants were filtered through 0.22 µm filters and centrifuged in a gradient of cesium chloride. Virus dialysis was performed using a membrane with a pore throughput of 3.5 kDa in two stages. After purification and concentration, the resulting recombinant adenoviruses were titrated by optical density, as well as by plaque formation. The titer of the recombinant adenoviruses we obtained was from 1 to 3.8 × 109 PFU/mL. The viral titer values were guided by the genetic modification of human UCB-MC.
## 4.3. UCB-MC Isolation and Characterization
All UCB-MC units were collected from healthy donors with a gestation period of 37–40 weeks in maternity public hospitals in Kazan. Blood collections were carried out into single blood bags of 250 mL, with the blood preservative CPDA-1 (GCMS, Republic of Korea). Exclusion criteria were maternal infections or viral diseases. Isolations of mononuclear cells were conducted within 16–18 h after blood collection. Nucleated blood cells were isolated using SepMate ™-50 tubes according to the manufacturer’s protocol (STEMCELL Technologies Inc., Vancouver, BC, Canada). The viability of the isolated cells was determined in a hemocytometer with a $0.4\%$ trypan blue solution. Cell viability, as measured by trypan blue exclusion, was >$97\%$. The immune phenotype of isolated cells was analyzed by staining with monoclonal antibodies CD45—PerCP (BioLegend, San Diego, CA, USA), CD3-FITC (BioLegend, San Diego, CA, USA) CD14-APC/Cy7 (BioLegend, San Diego, CA, USA), CD38-APC/Cy7 (BioLegend, San Diego, CA, USA) CD34-FITC (BioLegend, San Diego, CA, USA), CD90-PE/Cy5 (BioLegend, San Diego, CA, USA). Expression of CD markers were analyzed by flow cytometry using BD FACS Aria III (BD bioscience, San Jose, CA, USA)
## 4.4. Analysis of Adenoviral Transduction of hUCB-MCs
Genetic modification of hUCB-MCs with recombinant adenoviruses (MOI 10 for each virus) was performed according to a previously developed protocol [77].
The efficiency of genetic modification was assessed after 72 h by means of fluorescent microscopy on AxioObserverZ1 (Carl Zeiss, Oberkochen, Germany) and flow cytometry using BD FACS Aria III (BD Bioscience, San Jose, CA, USA).
## 4.5. Total RNA Extraction and RT-qPCR
Analysis of the mRNA expression of VEGF165, FGF2, and SDF1α in genetically modified cells and isolated Matrigel implants was carried out by qPCR with further statistical analysis. Isolation of total RNA was performed by using the TRIzol (Thermo Fisher Scientific, Waltham, MA, USA) reagent according to the manufacturer’s recommendations with further cDNA synthesis. Real-Time PCR was set up on the Real-Time CFX96 Touch (BioRad Laboratories, CA, USA). The nucleotide sequences of the primers and probes used in RT-qPCR are mentioned in Table 2. All reactions for each sample were performed in triplicate with a further calculation of ΔΔCt values and normalization to the housekeeping gene of β-actin rRNA. Standard curves for quantitative analysis were created using serial dilutions of plasmid DNA with corresponding inserts (VEGF, FGF2, and SDF1α). Expression of target genes in non-transduced UCB-MCs was considered $100\%$. The level of the murine target gene mCD31 was normalized to the mouse housekeeping gene of mGAPDH. The statistical analysis of the obtained results was carried out in MS Excel 2007 with further calculation using U criteria (Mann-Whitney).
## 4.6. Analysis of Cytokines and Chemokines
Supernatant cytokine profiles were analyzed using Bio-Plex Pro Human Cytokine 27-plex Panel and Bio-Plex Human Cytokine 21-plex Panel (Bio-Rad, Hercules, CA, USA) multiplex magnetic bead-based antibody detection kits, following the manufacturer’s instructions. Supernatant aliquots (50 µL) were used for analysis, with a minimum of 50 beads per analyte acquired. Median fluorescence intensities were measured using a Bioplex 200 (Bio-Rad, Hercules, CA, USA) analyzer. Data collected were analyzed with MasterPlex CT control software and MasterPlex QT analysis software (Hitachi Software, San Bruno, CA, USA). Standard curves for each analyte were generated using standards provided by the manufacturer.
## 4.7. In Vivo Experiments
In vivo experiments were performed using immune-deficient mice of Balb/c nude lineage of both sexes for 7–8 weeks. The animals were bred using the animal facilities in Puschino’s laboratory. After quarantine, animals were held in an SPF vivarium with HEPA filters according to GLP standards. In the area of the withers, mice were subcutaneously injected with 2 million human transduced or native UCB-MCs mixed with 300 µL of Matrigel matrix. Female and male Balb/c nude mice were randomly assigned to a few groups: 1. Matrigel without UCB-MCs; 2. UCB-MCs transduced with Ad-EGFP in Matrigel; 3. UCB-MCs transduced with Ad-VEGF165 in Matrigel; 4. UCB-MCs transduced with a combination of adenoviruses Ad-VEGF165; Ad-SDF1α and Ad-FGF2 in Matrigel. All experiments were performed in quadruplicates. After seven days post-transplantation mice were taken from the experiment. The status of subcutaneous Matrigel implants was evaluated visually, and concentrations of hemoglobin were evaluated. Levels of the expression of therapeutic genes were analyzed by RT-qPCR. Production of therapeutic proteins was assessed via immunohistochemistry.
## 4.8. Analysis of Hemoglobin Concentration
The analysis of hemoglobin concentration in subcutaneous implants was evaluated colorimetrically. Implants were balanced by weight and homogenized in DPBS using the Mini-Bead Beater-16 (BioSpec, Bartlesville, OK, USA) with zirconium beads ($d = 2$ mm for 100 mg), during 2 cycles for 20 sec each. The obtained homogenates were centrifuged at 15,000× g for 15 min. Supernatants containing hemoglobin were examined on a microplate reader Tecan Infinite Pro 2000 with an OD of 540 nm (Tecan Austria GmbH, Grödig, Austria).
## 4.9. Histological Analysis
For histological analysis, Matrigel implants were fixed in a $10\%$ buffered formalin solution for 48 h. After fixation, implants were dehydrated in increasing concentrations of ethanol and embedded in paraffin (Histomix, Biovitrum, Saint Petersburg, Russia). Paraffin slides with 5 µm thickness were prepared at the rotary microtome HM 355S (Thermo Fisher Scientific, Waltham, MA, USA). *For* general morphological characterization, slides were deparaffinized and stained with hematoxylin and eosin according to the standard protocol. For immunological studies, serial sections were deparaffinized and incubated in a citric buffer for 30 min to unmask epitopes. Cell membranes were permeabilized in a $0.1\%$ solution of Tween-20 in PBS. Non-specific binding was blocked by incubation in a $10\%$ solution of donkey serum for 30 min. Sections were stained with the antibodies to VEGF (mab293), FGF2 (sc-1390), and SFD1α (sc-28876), diluted 1:100 for 1 h at room temperature. After washing sections were stained for 1 h with secondary antibodies at room temperature followed by washing and DAPI staining (1:50,000 dilution) for 10 min. Primary and secondary antibodies are shown in Table 3. The microvessel density (MVD) was examined by counting the vessels in each implant, reported as the vessel number per square millimeter (vessel/mm2). Visualization of the results was performed on a scanning laser microscope LSM780 (Carl Zeiss, Oberkochen, Germany). Image analysis was performed using Image J software (https://fiji.sc/ (accessed on 18 December 2021)).
## 4.10. Statistical Analysis
GraphPad Prism® 7 software was used to show all data reports (GraphPad, Inc., La Jolla, CA, USA). The data are presented as the mean ± standard error (SE). p-values were analyzed using a one-way analysis of variance (ANOVA) followed by Tukey’s test. Statistical significance is denoted by $p \leq 0.05.$ All tests with animals, morphometric and statistical analyses were performed in a “blinded” manner with respect to the experimental groups.
## 5. Conclusions
The study suggests that UCB-MCs continue to be an essential source of stem cells for therapy and various gene-cell strategies for tissue regeneration. It is important to emphasize that the transplantation of genetically modified UCB-MCs is safer and more effective than direct gene therapy. Human UCB-MCs can be efficiently simultaneously modified with adenoviral vectors encoding VEGF, FGF2, and SDF1α. Modified UCB-MCs overexpress recombinant genes. Genetic modification of cells with recombinant adenoviruses (MOI 10) does not affect the profile of secreted pro- and anti-inflammatory cytokines, chemokines, or growth factors, except for an increase in the synthesis of recombinant proteins by cells. Modified cells can induce the formation of new vessels. Although many unresolved problems remain before modified UCB-MCs can be applied in the clinic, our results remain promising in terms of the induction of therapeutic angiogenesis in future clinical trials, including the treatment of decompensated forms.
## References
1. Rodriguez A.E., Pavlovsky H., Del Pozo J.F.. **Understanding the Outcome of Randomized Trials with Drug-Eluting Stents and Coronary Artery Bypass Graft in Patients with Multivessel Disease: A Review of a 25-Year Journey**. *Clin. Med. Insights Cardiol.* (2016) **10** 195-199. DOI: 10.4137/CMC.S40645
2. Kuwano M., Fukushi J., Okamoto M., Nishie A., Goto H., Ishibashi T., Ono M.. **Angiogenesis factors**. *Intern. Med.* (2001) **40** 565-572. DOI: 10.2169/internalmedicine.40.565
3. Ripa R.S., Wang Y., Jørgensen E., Johnsen H.E., Hesse B., Kastrup J.. **Intramyocardial injection of vascular endothelial growth factor-A165 plasmid followed by granulocyte-colony stimulating factor to induce angiogenesis in patients with severe chronic ischaemic heart disease**. *Eur. Heart J.* (2006) **27** 1785-1792. DOI: 10.1093/eurheartj/ehl117
4. Rissanen T.T., Ylä-Herttuala S.. **Current status of cardiovascular gene therapy**. *Mol. Ther.* (2007) **15** 1233-1247. DOI: 10.1038/sj.mt.6300175
5. Lee E.J., Park H.W., Jeon H.J., Kim H.S., Chang M.S.. *Potentiated Therapeutic Angiogenesis by Primed Human Mesenchymal Stem Cells in a Mouse Model of Hindlimb Ischemia* (2013) **Volume 8**. DOI: 10.2217/rme.13.17
6. Boyle A.J., McNiece I.K., Hare J.M.. **Mesenchymal stem cell therapy for cardiac repair**. *Stem Cells Myocard. Regen.* (2010) **660** 65-84
7. Markosyan V., Safiullov Z., Izmailov A., Fadeev F., Sokolov M., Kuznetsov M., Trofimov D., Kim E., Kundakchyan G., Gibadullin A.. **Preventive Triple Gene Therapy Reduces the Negative Consequences of Ischemia-Induced Brain Injury after Modelling Stroke in a Rat**. *Int. J. Mol. Sci.* (2020) **21**. DOI: 10.3390/ijms21186858
8. Urashima M., Hoshi Y., Shishikura A., Kamijo M., Kato Y., Akatsuka J., Maekawa K.. **Umbilical cord blood as a rich source of immature hematopoietic stem cells**. *Pediatr. Int.* (1994) **36** 649-655. DOI: 10.1111/j.1442-200X.1994.tb03262.x
9. Harris D.T., Rogers I.. **Umbilical cord blood: A unique source of pluripotent stem cells for regenerative medicine**. *Curr. Stem Cell Res. Ther.* (2007) **2** 301-309. DOI: 10.2174/157488807782793790
10. Divya M.S., Roshin G.E., Divya T.S., Rasheed V.A., Santhoshkumar T.R., Elizabeth K.E., James J., Pillai R.M.. **Umbilical cord blood-derived mesenchymal stem cells consist of a unique population of progenitors co-expressing mesenchymal stem cell and neuronal markers capable of instantaneous neuronal differentiation**. *Stem Cell Res. Ther.* (2012) **3** 57. DOI: 10.1186/scrt148
11. Kögler G., Sensken S., Airey J.A., Trapp T., Müschen M., Feldhahn N., Liedtke S., Sorg R.V., Fischer J., Rosenbaum C.. **A new human somatic stem cell from placental cord blood with intrinsic pluripotent differentiation potential**. *J. Exp. Med.* (2004) **200** 123-135. DOI: 10.1084/jem.20040440
12. Gluckman E.. **Ten years of cord blood transplantation: From bench to bedside**. *Br. J. Haematol.* (2009) **147** 192-199. DOI: 10.1111/j.1365-2141.2009.07780.x
13. Pimentel-Coelho P.M., Rosado-de Castro P.H., Da Fonseca L.M.B., Mendez-Otero R.. **Umbilical cord blood mononuclear cell transplantation for neonatal hypoxic-ischemic encephalopathy**. *Pediatr. Res.* (2012) **71** 464-473. DOI: 10.1038/pr.2011.59
14. Goldstein G., Toren A., Nagler A.. **Transplantation and other uses of human umbilical cord blood and stem cells**. *Curr. Pharm. Des.* (2007) **13** 1363-1373. DOI: 10.2174/138161207780618759
15. Balassa K., Rocha V.. **Anticancer cellular immunotherapies derived from umbilical cord blood**. *Expert Opin. Biol. Ther.* (2018) **18** 121-134. DOI: 10.1080/14712598.2018.1402002
16. Rocha V., Labopin M., Sanz G., Arcese W., Schwerdtfeger R., Bosi A., Jacobsen N., Ruutu T., de Lima M., Finke J.. **Transplants of umbilical-cord blood or bone marrow from unrelated donors in adults with acute leukemia**. *N. Engl. J. Med.* (2004) **351** 2276-2285. DOI: 10.1056/NEJMoa041469
17. Wagner J.E., Rosenthal J., Sweetman R., Shu X.O., Davies S.M., Ramsay N.K., McGlave P.B., Sender L., Cairo M.S.. **Successful transplantation of HLA-matched and HLA-mismatched umbilical cord blood from unrelated donors: Analysis of engraftment and acute graft-versus-host disease**. *Blood* (1996) **88** 795-802. DOI: 10.1182/blood.V88.3.795.795
18. Arien-Zakay H., Lecht S., Bercu M.M., Tabakman R., Kohen R., Galski H., Nagler A., Lazarovici P.. **Neuroprotection by cord blood neural progenitors involves antioxidants, neurotrophic and angiogenic factors**. *Exp. Neurol.* (2009) **216** 83-94. DOI: 10.1016/j.expneurol.2008.11.006
19. Bachstetter A.D., Pabon M.M., Cole M.J., Hudson C.E., Sanberg P.R., Willing A.E., Bickford P.C., Gemma C.. **Peripheral injection of human umbilical cord blood stimulates neurogenesis in the aged rat brain**. *BMC Neurosci.* (2008) **9**. DOI: 10.1186/1471-2202-9-22
20. Dasari V.R., Spomar D.G., Li L., Gujrati M., Rao J.S., Dinh D.H.. **Umbilical cord blood stem cell mediated downregulation of fas improves functional recovery of rats after spinal cord injury**. *Neurochem. Res.* (2008) **33** 134-149. DOI: 10.1007/s11064-007-9426-6
21. Schira J., Gasis M., Estrada V., Hendricks M., Schmitz C., Trapp T., Kruse F., Kögler G., Wernet P., Hartung H.P.. **Significant clinical, neuropathological and behavioural recovery from acute spinal cord trauma by transplantation of a welldefined somatic stem cell from human umbilical cord blood**. *Brain* (2012) **135** 431-446. DOI: 10.1093/brain/awr222
22. Xiao J., Nan Z., Motooka Y., Low W.C.. **Transplantation of a novel cell line population of umbilical cord blood stem cells ameliorates neurological deficits associated with ischemic brain injury**. *Stem Cells Dev.* (2005) **14** 722-733. DOI: 10.1089/scd.2005.14.722
23. Ikeda Y., Fukuda N., Wada M., Matsumoto T., Satomi A., Yokoyama S.I., Saito S., Matsumoto K., Kanmatsuse K., Mugishima H.. **Development of angiogenic cell and gene therapy by transplantation of umbilical cord blood with vascular endothelial growth factor gene**. *Hypertens. Res.* (2004) **27** 119-128. DOI: 10.1291/hypres.27.119
24. Chen H.K., Hung H.F., Shyu K.G., Wang B.W., Sheu J.R., Liang Y.J., Chang C.C., Kuan P.. **Combined cord blood stem cells and gene therapy enhances angiogenesis and improves cardiac performance in mouse after acute myocardial infarction**. *Eur. J. Clin. Investig.* (2005) **35** 677-686. DOI: 10.1111/j.1365-2362.2005.01565.x
25. Islamov R.R., Sokolov M.E., Bashirov F.V., Fadeev F.O., Shmarov M.M., Naroditskiy B.S., Povysheva T.V., Shaymardanova G.F., Yakupov R.A., Chelyshev Y.A.. **A pilot study of cell-mediated gene therapy for spinal cord injury in mini pigs**. *Neurosci. Lett.* (2017) **644** 67-75. DOI: 10.1016/j.neulet.2017.02.034
26. Izmailov A.A., Povysheva T.V., Bashirov F.V., Sokolov M.E., Fadeev F.O., Garifulin R.R., Naroditsky B.S., Logunov D.Y., Salafutdinov I.I., Chelyshev Y.A.. **Spinal Cord Molecular and Cellular Changes Induced by Adenoviral Vector- and Cell-Mediated Triple Gene Therapy after Severe Contusion**. *Front. Pharmacol.* (2017) **8** 813. DOI: 10.3389/fphar.2017.00813
27. Sokolov M.E., Bashirov F.V., Markosyan V.A., Povysheva T.V., Fadeev F.O., Izmailov A.A., Kuztetsov M.S., Safiullov Z.Z., Shmarov M.M., Naroditskyi B.S.. **Triple-Gene Therapy for Stroke: A Proof-of-Concept in Vivo Study in Rats**. *Front. Pharmacol.* (2018) **9** 111. DOI: 10.3389/fphar.2018.00111
28. Nowak-Sliwinska P., Alitalo K., Allen E., Anisimov A., Aplin A.C., Auerbach R., Augustin H.G., Bates D.O., van Beijnum J.R., Bender R.H.F.. **Consensus guidelines for the use and interpretation of angiogenesis assays**. *Angiogenesis* (2018) **21** 425-532. DOI: 10.1007/s10456-018-9613-x
29. Stryker Z.I., Rajabi M., Davis P.J., Mousa S.A.. **Evaluation of angiogenesis assays**. *Biomedicines* (2019) **7**. DOI: 10.3390/biomedicines7020037
30. Simons M., Alitalo K., Annex B.H., Augustin H.G., Beam C., Berk B.C., Dimmeler S.. **State-of-the-art methods for evaluation of angiogenesis and tissue vascularization: A scientific statement from the American Heart Association**. *Circ. Res.* (2015) **116** e99-e132. DOI: 10.1161/RES.0000000000000054
31. Passaniti A., Taylor R., Pili R., Guo Y., Long P., Haney J., Pauly R., Grant D., Martin G.. **A simple, quantitative method for assessing angiogenesis and antiangiogenic agents using reconstituted basement membrane, heparin, and fibroblast growth factor**. *Lab. Investig.* (1992) **67** 519-528. PMID: 1279270
32. Martin S., Murray J.C.. *Angiogenesis Protocols* (2009) **Volume 467**. DOI: 10.1007/978-1-59745-241-0
33. Norrby K.. **In vivo models of angiogenesis**. *J. Cell. Mol. Med.* (2006) **10** 588-612. DOI: 10.1111/j.1582-4934.2006.tb00423.x
34. Kibbey M.C., Corcoran M.L., Wahl L.M., Kleinman H.K.. **Laminin SIKVAV peptide-induced angiogenesis in vivo is potentiated by neutrophils**. *J. Cell. Physiol.* (1994) **160** 185-193. DOI: 10.1002/jcp.1041600121
35. Salcedo R., Ponce M.L., Young H.A., Wasserman K., Ward J.M., Kleinman H.K., Oppenheim J.J., Murphy W.J.. **Human endothelial cells express CCR2 and respond to MCP-1: Direct role of MCP-1 in angiogenesis and tumor progression**. *Blood* (2000) **96** 34-40. DOI: 10.1182/blood.V96.1.34
36. Johns A.. **Disruption of estrogen receptor gene prevents 17 beta estradiol-induced angiogenesis in transgenic mice**. *Endocrinology* (1996) **137** 4511-4513. DOI: 10.1210/endo.137.10.8828515
37. Ferber S., Tiram G., Satchi-Fainaro R.. **Monitoring functionality and morphology of vasculature recruited by factors secreted by fast-growing tumor-generating cells**. *J. Vis. Exp.* (2014) **93** 1-6. DOI: 10.3791/51525
38. Barkats M., Bilang-Bleuel A., Buc-Caron M.H., Castel-Barthe M.N., Corti O., Finiels F., Horellou P., Revah F., Sabate O., Mallet J.. **Adenovirus in the brain: Recent advances of gene therapy for neurodegenerative diseases**. *Prog. Neurobiol.* (1998) **55** 333-341. DOI: 10.1016/S0301-0082(98)00028-8
39. Lee C.S., Bishop E.S., Zhang R., Yu X., Farina E.M., Yan S., Zhao C., Zheng Z., Shu Y., Wu X.. **Adenovirus-Mediated Gene Delivery: Potential Applications for Gene and Cell-Based Therapies in the New Era of Personalized Medicine**. *Genes Dis.* (2017) **4** 43-63. DOI: 10.1016/j.gendis.2017.04.001
40. Seggern D.J., Nemerow G.R., Fernandez J., Hoeffler J.P.. **Adenoviral Vectors For Protein Expression**. *Gene Expr. Systems* (1998) 111-156. DOI: 10.1016/B978-012253840-7/50006-7
41. MacKenzie K.L., Hackett N.R., Crystal R.G., Moore M.A.S.. **Adenoviral vector–mediated gene transfer to primitive human hematopoietic progenitor cells: Assessment of transduction and toxicity in long-term culture**. *Blood* (2000) **96** 100-108. DOI: 10.1182/blood.V96.1.100.013k41_100_108
42. Adams W.C., Gujer C., McInerney G., Gall J.G.D., Petrovas C., Karlsson Hedestam G.B., Koup R.A., Loré K.. **Adenovirus type-35 vectors block human CD4+ T-cell activation via CD46 ligation**. *Proc. Natl. Acad. Sci. USA* (2011) **108** 7499-7504. DOI: 10.1073/pnas.1017146108
43. Loré K., Adams W.C., Havenga M.J.E., Precopio M.L., Holterman L., Goudsmit J., Koup R.A.. **Myeloid and plasmacytoid dendritic cells are susceptible to recombinant adenovirus vectors and stimulate polyfunctional memory T cell responses**. *J. Immunol.* (2007) **179** 1721-1729. DOI: 10.4049/jimmunol.179.3.1721
44. Liu M., Xie S., Zhou J.. **Use of animal models for the imaging and quantification of angiogenesis**. *Exp. Anim.* (2018) **67** 1-6. DOI: 10.1538/expanim.17-0054
45. Iyer S.R., Annex B.H.. **Therapeutic Angiogenesis for Peripheral Artery Disease: Lessons Learned in Translational Science**. *JACC Basic Transl. Sci.* (2017) **2** 503-512. DOI: 10.1016/j.jacbts.2017.07.012
46. Gupta R., Tongers J., Losordo D.W.. **Human studies of angiogenic gene therapy**. *Circ. Res.* (2009) **105** 724-736. DOI: 10.1161/CIRCRESAHA.109.200386
47. Duffy A.M., Bouchier-Hayes D.J., Harmey J.H.. **Vascular Endothelial Growth Factor (VEGF) and Its Role in Non-Endothelial Cells: Autocrine Signalling by VEGF.**. (2013)
48. Zentilin L., Tafuro S., Zacchigna S., Arsic N., Pattarini L., Sinigaglia M., Giacca M.. **Bone marrow mononuclear cells are recruited to the sites of VEGF-induced neovascularization but are not incorporated into the newly formed vessels**. *Blood* (2006) **107** 3546-3554. DOI: 10.1182/blood-2005-08-3215
49. Itoh N.. **The Fgf families in humans, mice, and zebrafish: Their evolutional processes and roles in development, metabolism, and disease**. *Biol. Pharm. Bull.* (2007) **30** 1819-1825. DOI: 10.1248/bpb.30.1819
50. Seghezzi G., Patel S., Ren C.J., Gualandris A., Pintucci G., Robbins E.S., Shapiro R.L., Galloway A.C., Rifkin D.B., Mignatti P.. **Fibroblast growth factor-2 (FGF-2) induces vascular endothelial growth factor (VEGF) expression in the endothelial cells of forming capillaries: An autocrine mechanism contributing to angiogenesis**. *J. Cell Biol.* (1998) **141** 1659-1673. DOI: 10.1083/jcb.141.7.1659
51. Jin S., Yang C., Huang J., Liu L., Zhang Y., Li S., Zhang L., Sun Q., Yang P.. **Conditioned medium derived from FGF-2-modified GMSCs enhances migration and angiogenesis of human umbilical vein endothelial cells**. *Stem Cell Res. Ther.* (2020) **11** 68. DOI: 10.1186/s13287-020-1584-3
52. Ho T.K., Shiwen X., Abraham D., Tsui J., Baker D.. **Stromal-Cell-Derived Factor-1 (SDF-1)/CXCL12 as Potential Target of Therapeutic Angiogenesis in Critical Leg Ischaemia**. *Cardiol. Res. Pract.* (2012) **2012** 143209. DOI: 10.1155/2012/143209
53. Horuk R.. **Chemokine receptors**. *Cytokine Growth Factor Rev.* (2001) **12** 313-335. DOI: 10.1016/S1359-6101(01)00014-4
54. Zernecke A., Schober A., Bot I., von Hundelshausen P., Liehn E.A., Möpps B., Mericskay M., Gierschik P., Biessen E.A., Weber C.. **SDF-1alpha/CXCR4 axis is instrumental in neointimal hyperplasia and recruitment of smooth muscle progenitor cells**. *Circ. Res.* (2005) **96** 784-791. DOI: 10.1161/01.RES.0000162100.52009.38
55. Salcedo R., Oppenheim J.J.. **Role of chemokines in angiogenesis: CXCL12/SDF-1 and CXCR4 interaction, a key regulator of endothelial cell responses**. *Microcirculation* (2003) **10** 359-370. DOI: 10.1080/mic.10.3-4.359.370
56. Kano M.R., Morishita Y., Iwata C., Iwasaka S., Watabe T., Ouchi Y., Miyazono K., Miyazawa K.. **VEGF-A and FGF-2 synergistically promote neoangiogenesis through enhancement of endogenous PDGF-B-PDGFRbeta signaling**. *J. Cell Sci.* (2005) **118** 3759-3768. DOI: 10.1242/jcs.02483
57. Hu G.j., Feng Y.g., Lu W.p., Li H.t., Xie H.w., Li S.f.. **Effect of combined VEGF165/ SDF-1 gene therapy on vascular remodeling and blood perfusion in cerebral ischemia**. *J. Neurosurg.* (2017) **127** 670-678. DOI: 10.3171/2016.9.JNS161234
58. Ljubimov A.V., Penn J.S.. **Growth Factor Synergy in Angiogenesis**. *Retinal and Choroidal Angiogenesis* (2008) 289-310. DOI: 10.1007/978-1-4020-6780-8{_}16
59. de Paula E.V., Flores-Nascimento M.C., Arruda V.R., Garcia R.A., Ramos C.D., Guillaumon A.T., Annichino-Bizzacchi J.M.. **Dual gene transfer of fibroblast growth factor-2 and platelet derived growth factor-BB using plasmid deoxyribonucleic acid promotes effective angiogenesis and arteriogenesis in a rodent model of hindlimb ischemia**. *Transl. Res.* (2009) **153** 232-239. DOI: 10.1016/j.trsl.2009.02.002
60. Hwang S., Choi J., Kim M.. **Combining Human Umbilical Cord Blood Cells With Erythropoietin Enhances Angiogenesis/Neurogenesis and Behavioral Recovery After Stroke**. *Front. Neurol.* (2019) **10** 357. DOI: 10.3389/fneur.2019.00357
61. Ballen K.K., Gluckman E., Broxmeyer H.E.. **Umbilical cord blood transplantation: The first 25 years and beyond**. *Blood* (2013) **122** 491-498. DOI: 10.1182/blood-2013-02-453175
62. Garbuzova-Davis S., Ehrhart J., Sanberg P.R.. **Cord blood as a potential therapeutic for amyotrophic lateral sclerosis**. *Expert Opin. Biol. Ther.* (2017) **17** 837-851. DOI: 10.1080/14712598.2017.1323862
63. Rizvanov A.A., Guseva D.S., Salafutdinov I.I., Kudryashova N.V., Bashirov F.V., Kiyasov A.P., Yalvaç M.E., Gazizov I.M., Kaligin M.S., Sahin F.. **Genetically modified human umbilical cord blood cells expressing vascular endothelial growth factor and fibroblast growth factor 2 differentiate into glial cells after transplantation into amyotrophic lateral sclerosis transgenic mice**. *Exp. Biol. Med.* (2011) **236** 91-98. DOI: 10.1258/ebm.2010.010172
64. Islamov R.R., Rizvanov A.A., Fedotova V.Y., Izmailov A.A., Safiullov Z.Z., Garanina E.E., Salafutdinov I.I., Sokolov M.E., Mukhamedyarov M.A., Palotás A.. **Tandem Delivery of Multiple Therapeutic Genes Using Umbilical Cord Blood Cells Improves Symptomatic Outcomes in ALS**. *Mol. Neurobiol.* (2017) **54** 4756-4763. DOI: 10.1007/s12035-016-0017-x
65. Zhu H., Poon W., Liu Y., Leung G.K.K., Wong Y., Feng Y., Ng S.C.P., Tsang K.S., Sun D.T.F., Yeung D.K.. **Phase I-II Clinical Trial Assessing Safety and Efficacy of Umbilical Cord Blood Mononuclear Cell Transplant Therapy of Chronic Complete Spinal Cord Injury**. *Cell Transplant.* (2016) **25** 1925-1943. DOI: 10.3727/096368916X691411
66. Liu J., Han D., Wang Z., Xue M., Zhu L., Yan H., Zheng X., Guo Z., Wang H.. **Clinical analysis of the treatment of spinal cord injury with umbilical cord mesenchymal stem cells**. *Cytotherapy* (2013) **15** 185-191. DOI: 10.1016/j.jcyt.2012.09.005
67. Ichim T.E., Solano F., Lara F., Paris E., Ugalde F., Rodriguez J.P., Minev B., Bogin V., Ramos F., Woods E.J.. **Feasibility of combination allogeneic stem cell therapy for spinal cord injury: A case report**. *Int. Arch. Med.* (2010) **3** 30. DOI: 10.1186/1755-7682-3-30
68. Yang W.Z., Zhang Y., Wu F., Min W.P., Minev B., Zhang M., Luo X.L., Ramos F., Ichim T.E., Riordan N.H.. **Safety evaluation of allogeneic umbilical cord blood mononuclear cell therapy for degenerative conditions**. *J. Transl. Med.* (2010) **8** 75. DOI: 10.1186/1479-5876-8-75
69. Atasheva S., Yao J., Shayakhmetov D.M.. **Innate immunity to adenovirus: Lessons from mice**. *FEBS Lett.* (2019) **593** 3461-3483. DOI: 10.1002/1873-3468.13696
70. Garanina E.E., Gatina D., Martynova E.V., Rizvanov A., Khaiboullina S., Salafutdinov I.. **Cytokine Profiling of Human Umbilical Cord Plasma and Human Umbilical Cord Blood Mononuclear Cells**. *Blood* (2017) **130** 4814. DOI: 10.1182/blood.V130.Suppl_1.4814.4814
71. Reuschel E., Toelge M., Entleutner K., Deml L., Seelbach-Goebel B.. **Cytokine profiles of umbilical cord blood mononuclear cells upon in vitro stimulation with lipopolysaccharides of different vaginal gram-negative bacteria**. *PLoS ONE* (2019) **14**. DOI: 10.1371/journal.pone.0222465
72. Newman M.B., Willing A.E., Manresa J.J., Sanberg C.D., Sanberg P.R.. **Cytokines produced by cultured human umbilical cord blood (HUCB) cells: Implications for brain repair**. *Exp. Neurol.* (2006) **199** 201-208. DOI: 10.1016/j.expneurol.2006.04.001
73. Teigler J.E., Iampietro M.J., Barouch D.H.. **Vaccination with adenovirus serotypes 35, 26, and 48 elicits higher levels of innate cytokine responses than adenovirus serotype 5 in rhesus monkeys**. *J. Virol.* (2012) **86** 9590-9598. DOI: 10.1128/JVI.00740-12
74. Salafutdinov I.I., Gatina D.Z., Markelova M.I., Garanina E.E., Malanin S.Y., Gazizov I.M., Khaiboullina S., Rizvanov A.A., Islamov R.I.. **Transcriptomic Landscape of Umbilical Cord Blood Mononuclear Cells after Genetic Modification**. *Blood* (2020) **136** 33-34. DOI: 10.1182/blood-2020-138664
75. Cherenkova E., Fedotova V., Borisov M., Islamov R., Rizvanov A.. **Generation of recombinant adenoviruses and lentiviruses expressing angiogenic and neuroprotective factors using Gateway cloning technology**. *Cell. Transplant. Tissue Eng.* (2012) **7** 164-168
76. Solovyeva V.V., Chulpanova D.S., Tazetdinova L.G., Salafutdinov I.I., Bozo I.Y., Isaev A.A., Deev R.V., Rizvanov A.A.. **In Vitro Angiogenic Properties of Plasmid DNA Encoding SDF-1α and VEGF165 Genes**. *Appl. Biochem. Biotechnol.* (2020) **190** 773-788. DOI: 10.1007/s12010-019-03128-5
77. Islamov R., Rizvanov A., Mukhamedyarov M., Salafutdinov I., Garanina E., Fedotova V., Solovyeva V., Mukhamedshina Y., Safiullov Z., Izmailov A.. **Symptomatic improvement, increased life-span and sustained cell homing in amyotrophic lateral sclerosis after transplantation of human umbilical cord blood cells genetically modified with adeno-viral vectors expressing a neuro-protective factor and a neural cell adhesion molecule**. *Curr. Gene Ther.* (2015) **15** 266-276. PMID: 25619885
|
---
title: 'Associations of Cooking Skill with Social Relationships and Social Capital
among Older Men and Women in Japan: Results from the JAGES'
authors:
- Yukako Tani
- Takeo Fujiwara
- Katsunori Kondo
journal: International Journal of Environmental Research and Public Health
year: 2023
pmcid: PMC10002414
doi: 10.3390/ijerph20054633
license: CC BY 4.0
---
# Associations of Cooking Skill with Social Relationships and Social Capital among Older Men and Women in Japan: Results from the JAGES
## Abstract
The health benefits of social relationships and social capital are well known. However, little research has examined the determinants of social relationships and social capital. We examined whether cooking skill was associated with social relationships and social capital in older Japanese people. We used 2016 Japan Gerontological Evaluation *Study data* on a population-based sample of men and women aged ≥ 65 years ($$n = 21$$,061). Cooking skill was assessed using a scale with good validity. Social relationships were evaluated by assessing neighborhood ties, frequency and number of meetings with friends, and frequent meals with friends. Individual-level social capital was evaluated by assessing civic participation, social cohesion, and reciprocity. Among women, high-level cooking skill was positively associated with all components of social relationships and social capital. Women with high-level cooking skill were 2.27 times ($95\%$ CI: 1.77–2.91) more likely to have high levels of neighborhood ties and 1.65 ($95\%$ CI: 1.20–2.27) times more likely to eat with friends, compared with those with middle/low-level cooking skill. Cooking skills explained $26.2\%$ of the gender difference in social relationships. Improving cooking skills may be key to boosting social relationships and social capital, which would prevent social isolation.
## 1. Introduction
Globally, there were 901 million people aged 60 years or older in 2015, and this number is projected to rise to 1.4 billion by 2030 [1]. In older age, social networks may decline because of retirement, adult children’s independence, and bereavement after the death of spouses or friends. Socially isolated older people are at increased risk of several detrimental health outcomes including mortality [2], dementia [3], and poor mental health [4]. Therefore, it is important to find modifiable factors that foster social relationships among older adults.
Social relationships are measured in a variety of ways, with three main aspects being used in research on health: social network, social activity, and social support [5]. Social networks and social activity represent structural aspects of social relationships, whereas social support represents functional aspects of social relationships [5]. Social network covers network size (number of members) and density (frequency of contact between members), social activities are represented by social participation and social engagement, and social support refers to a perception of the availability of support from members of the social network [5]. In addition to social relationships, social capital is another important health-promoting concept. Social capital is described as resources that people can receive through their social networks, although there is no universally agreed definition of social capital [6,7].
The health benefits of both Individual- and community-level social capital have been shown in many epidemiological studies [6,7,8,9]. However, little research has examined the determinants of social capital. Recently, gender inequality in social capital has been reported, with women having higher levels of some social capital components, such as reciprocity and bridging, compared with men [10,11]. Compared with men, women tend to invest more in social relationships and building intimate emotional relationships [12,13]. In a study of older adults in Japan and England, women more often met with friends than did men [14,15]. However, the reasons for differences in social relationships between men and women are still unknown. In addition to gender, ethnicity and socioeconomic status (SES) have been reported as possible determinants of social capital [10,11,16], but these factors (i.e., gender, ethnicity, and SES) are difficult or impossible to modify through intervention. To boost social capital, modifiable factors determining social relationships and social capital should be identified.
Activity-related food has been linked with social activity from an evolutionary perspective [17]. Meal preparation ability may contribute to fostering not only family relationships but also social relationships with neighbors and friends. A qualitative study among rural older adults in the United States reported that most older adults gave or received some kind of food, especially cooked foods and garden products, and women were more likely to receive food gifts than men [18]. This food sharing was valued as a way to maintain reciprocity in social relations and to create a feeling of community membership [18]. In Japan, there is a culture of osusowake, which refers to the mutual exchange of foodstuffs between neighbors. This culture may contribute to strengthening community networks through supporting cultural activities including local festivals and seasonal events [19]. A systematic report on the benefits of cooking interventions showed that community kitchen programs had a positive influence on socialization [20]. Higher levels of cooking skills have been found to increase the frequency of cooking and confidence in cooking [21,22,23,24,25]. Thus, cooking skill may increase opportunities to build better social relationships with others, such as sharing food with neighbors and attending local cultural activities.
Cooking skills represent a basic living ability that contributes to better diet quality. Several studies have shown the dietary benefits of cooking skills, such as higher consumption of vegetables and fruits and lower consumption of prepared meals, convenience foods, and ultra-processed foods [21,25,26,27]. However, little is known about the importance of cooking skills beyond dietary outcomes. Although one’s mother is the most common source for learning cooking skills, people also learn from partners, cookbooks, television shows, and cooking classes [23,28]. Thus, interventions are possible even in older age. In fact, because retirement allows more time to cook, it is reasonable for older people to newly start to learn cooking skills.
The aim of this study was to examine the associations of cooking skills with social relationships and social capital among older adults. First, to identify social relationships that can be modified through intervention, we examined the association of cooking skills with social relationships with neighbors and friends rather than with relatives. Specifically, the investigated social relationships included neighborhood ties, frequency of meetings with friends, number of meetings with friends, and shared meals with friends. Next, we examined the associations between cooking skills and individual-level social capital, which included civic participation, social cohesion, and reciprocity [29]. Finally, we examined gender differences in social relationships and social capital, as well as the mediating role of cooking skills in the associations of gender with social relationships and social capital.
## 2.1. Study Design and Participants
We used data from the Japan Gerontological Evaluation Study (JAGES), which was carried out in 39 municipalities across Japan in 2016. The study targeted community-dwelling older adults without functional disabilities, defined as not being certified as eligible to receive long-term public care insurance system services [30]. From October 2016 to January 2017, self-report questionnaires were mailed to 279,661 adults aged ≥ 65 years, and 196,438 individuals returned the questionnaire (response rate: $70.2\%$). The survey was conducted using random sampling in 22 large municipalities and was administered to all eligible residents in 17 small municipalities [25]. One-eighth of the target sample ($$n = 22$$,219) were randomly selected to receive the survey module inquiring about cooking skills. Of the 21,061 participants who had information on both gender and cooking skills and did not report any limitations in activities of daily living, those who had information on each outcome variable were included in the analysis; thus, the analytic sample differs depending on the outcome: $$n = 20$$,799 for neighborhood ties, $$n = 20$$,477 for frequency of meetings with friends, $$n = 20$$,445 for the number of meetings with friends, $$n = 21$$,061 for shared meals with friends, $$n = 15$$,631 for civic participation, $$n = 20$$,424 for social cohesion, and $$n = 20$$,224 for reciprocity. Participants were informed that participation in the study was voluntary and that completing and returning the questionnaire indicated their consent to participate in the study.
## 2.2. Social Relationships
Neighborhood ties, frequency of meetings with friends, number of meetings with friends, and frequent shared meals with friends were evaluated to assess social relationships. All components of social relationships were assessed using the self-report questionnaire. For neighborhood ties, participants were asked, “What kind of interactions do you have with people in your neighborhood?” The four response options were [1] mutual consultation, lending and borrowing daily commodities, and cooperation in daily life; [2] standing and chatting frequently; [3] no more than exchanging greetings; and [4] none, not even greetings [29,31]. We classified the participants as having high (response 1), middle (response 2), or low (response 3 or response 4) levels of ties, collapsing the two response categories because only $2.27\%$ of the participants reported having no interactions with people in their neighborhood (response 4). The frequency of meetings with friends was assessed using the following question: “How often do you see your friends?”. The six response options were [1] ≥4 times/week; [2] 2–3 times/week; [3] 1 time/week; [4] 1–3 times/month; [5] a few times/year; [6] never [29]. In this analysis, the scores of 4, 2.5, 1, 0.5, 0.125, and 0 (times/week) were assigned to response categories 1, 2, 3, 4, 5, and 6, respectively, and the resulting variable was treated as continuous. The number of meetings with friends was assessed using the following question: “How many friends/acquaintances have you seen over the past month?”. The five response options were [1] ≥10; [2] 6–9; [3] 3–5; [4] 1–2; [5] 0 [29]. In this analysis, the scores of 10, 7.5, 4, 1.5, and 0 (persons/month) were assigned to responses 1, 2, 3, 4, and 5, respectively, and the resulting variable was treated as continuous. Frequent shared meals with friends were assessed using the following question: “Who do you usually have meals with?”. The possible responses were no one, spouse, children, grandchildren, friends, and other [32]. Multiple responses were possible. We defined participants who selected “friends” as eating with friends.
## 2.3. Social Capital
Individual-level social capital was evaluated by assessing civic participation, social cohesion, and reciprocity using a validated scale to measure community-level social capital [29]. These variables were assessed using the self-report questionnaire, and details of this assessment have been described elsewhere [29]. For civic participation, we calculated the number of groups in which a respondent participated once or more [29]. Social cohesion was assessed using the following questions: “Do you think people living in your area can be trusted in general?” ( community trust), “Do you think most people in your community offer assistance to others?” ( norm of reciprocity), and “How strong is your residential place attachment?” ( community attachment). Responses were rated on a five-point scale ranging from strongly trusted, agree strongly, or strongly attached to not at all. We calculated the number of items on which the participant strongly or moderately agreed [29]. Reciprocity was assessed using the following questions: “Do you have someone who listens to your concerns and complaints?” ( received emotional support), “Do you listen to someone’s concerns and complaints?” ( provided emotional support), and “Do you have someone who looks after you when you are sick for a few days?” ( received instrumental support). The possible responses were no one, spouse, children, sibling/relative/parent/grandchildren, neighbors, friends, and other. Multiple responses were allowed. To explore the type of reciprocity that can be changed through intervention, we calculated the number of items for which the respondent selected neighbors, friends, or other.
## 2.4. Cooking Skills
Cooking skills were assessed using a cooking skills scale designed with consideration of basic Japanese cooking methods and typical meals; details of this assessment have been described elsewhere [25]. This scale had appropriate internal consistency (Cronbach’s α = 0.96) and notable discriminant validity, with women (experienced food preparers) scoring significantly better than men (food preparation novices) [25]. The scale consisted of seven items: [1] overall cooking skills; [2] able to peel fruits and vegetables; [3] able to boil eggs and vegetables; [4] able to grill fish; [5] able to make stir-fried meat and vegetables; [6] able to make miso soup; and [7] able to make stewed dishes. Participants were asked to evaluate their own cooking skills on a six-point scale ranging from unable (=0) to very well (=5). We calculated the mean of these seven items and divided the result into three categories: high (score of >4.0), middle (score of 2.1–4.0), and low (score of ≤2.0) [25]. For women, the middle group and the low group were combined into one category because the low group was quite small ($1.2\%$).
## 2.5. Covariates
Covariates were assessed using the self-report questionnaire (Table S1). We included education, current annual household income, and marital status as socio-demographic characteristics [25]. For health status, we asked whether the participants were currently under medical treatment for any of the following conditions: cancer, heart disease, stroke, hypertension, diabetes mellitus, and hyperlipidemia. Furthermore, depressive symptoms were assessed using the Geriatric Depression Scale [33]. To account for personality aspects such as curiosity regarding cooking, which may be directly associated with social relationships, as a sensitivity analysis, we controlled for whether the participants talked with young people [34] and the participants’ willingness to take on a leadership role in a community activity. Participants with missing data on covariates were included in the analysis as dummy variables.
## 2.6. Statistical Analysis
The analyses were stratified by gender because different associations between cooking skills and dietary behaviors have been reported for men and women [25]. First, after stratifying the sample by gender, we tested the differences using the chi-square test for categorical variables and the t-test or ANOVA for continuous variables. Next, participants were stratified by their level of cooking skills, and differences were tested using the chi-square test for categorical variables and the t-test or ANOVA for continuous variables. Second, for neighborhood ties, we used multinomial logistic regression to calculate adjusted relative risk ratios (RRRs) with $95\%$ CIs of high-level and middle-level ties, with low-level ties as the reference category. For the frequency and number of meetings with friends and social capital (civic participation, social cohesion, and reciprocity), we used multivariate linear regression models, adjusting for potential confounders. For frequent shared meals with friends, we used logistic regression to calculate adjusted odds ratios with $95\%$ CIs of eating meals with friends. The models were adjusted for the following potential confounding factors: age, socio-demographic characteristics (education, annual normalized household income, and marital status), and health status (medical treatment of cancer, heart disease, stroke, hypertension, diabetes mellitus, and hyperlipidemia, as well as depressive symptoms).
Additionally, we conducted structural equation modeling (SEM) analysis to explore the mediating role of cooking skills in the associations of gender with social relationships and social capital. In the SEM analysis, social relationships and social capital were treated as latent variables estimated from neighborhood ties, frequency of meetings with friends, number of meetings with friends, frequent shared meals with friends, civic participation, and reciprocity ($$n = 15$$,207 because of missing values on the variables used to estimate the latent variables). Cooking skill, operationalized as the mean value of the seven cooking skill items, was treated as a continuous variable. Overall model fit was tested using the comparative fit index, the root mean square error of approximation, and the standardized root mean square residual. All analyses were conducted using Stata, Version 15 (Stata Statistical Software: Release 15. College Station, TX, USA: StataCorp LP).
## 3. Results
The participants’ characteristics are summarized in Table S1. Women were about twice as likely as men to have a high level of neighborhood ties and to eat with their friends.
The associations between cooking skills and social relationships are shown in Table 1. The interaction effect between cooking skills and gender was significant: the relationships with all components of social relationships were higher among women than among men ($p \leq 0.05$ for the interaction). Women with a high level of cooking skills were 2.27 times ($95\%$ CI: 1.77–2.91) more likely to have a high level of neighborhood ties and 1.65 ($95\%$ CI: 1.20–2.27) times more likely to eat with friends, compared with women with middle/low-level cooking skills. High-level cooking skill was associated with a higher frequency and number of meetings with friends. Men with high-level cooking skills were 1.84 times ($95\%$ CI: 1.46–2.33) more likely to have a high level of neighborhood ties, compared with men with low-level cooking skills. For men, high-level cooking skill was associated with a higher frequency and number of meetings with friends. These associations remained significant after adjusting for prosocial behavior-related personality (Table S2).
The associations between cooking skills and social capital are shown in Table 2. The interaction effect between cooking skills and gender was significant ($p \leq 0.05$ for the interaction). For women, high-level cooking skill was positively associated with all components of social capital, whereas the relationship between high-level cooking skill and social cohesion was non-significant for men. These associations remained significant after adjusting for prosocial behavior-related personality (Table S3).
Compared with men, women had higher levels of social relationships and social capital except for social cohesion (Tables S4 and S5). Women were 3.01 times ($95\%$ CI: 2.76–3.29) more likely to have a high level of neighborhood ties and 2.47 times ($95\%$ CI: 2.20–2.78) more likely to eat with friends, compared with men. Women had a higher frequency and number of meetings with friends (coefficient = 0.34, $95\%$ CI: 0.30–0.38 and coefficient = 0.67, $95\%$ CI: 0.57–0.78), more civic participation (coefficient = 0.23, $95\%$ CI: 0.19–0.27), and higher reciprocity (coefficient = 0.11, $95\%$ CI: 0.10–0.13). However, women also had lower social cohesion compared with their male counterparts (coefficient = −0.04, $95\%$ CI: −0.07 to −0.01).
Figure 1 shows the result of the SEM analysis for the association between gender and social capital including social relationships except for social cohesion. This SEM analysis demonstrated good model fit (likelihood-ratio test of the model, chi-square = 208.4, $p \leq 0.001$; comparative fit index = 0.991; root mean square error of approximation = 0.03; standardized root mean square residual = 0.016). The association between gender and social capital was partially mediated by cooking skill (from gender to cooking skill: standardized coefficient = 0.570, $p \leq 0.001$; from cooking skill to social relationships including social capital: standardized coefficient = 0.152, $p \leq 0.001$). The indirect effect was $26.2\%$ of the total effect.
## 4. Discussion
To our knowledge, this is the first study to examine cooking skills as a modifiable determinant of social relationships and social capital. We found that, among older adults in Japan, a high level of cooking skill was positively associated with social relationships and social capital, and we identified significant interaction effects between cooking skill and gender on social relationships and social capital. We confirmed that women had higher levels of social relationships and social capital than men, and these associations were partially mediated by cooking skill.
Given that food plays a central role in connecting people in traditional Japanese culture [35], our results are plausible. Special meals for many rituals and celebrations throughout the year are handed down in various forms throughout Japan [36]. For events, people prepare special meals called gyoujisyoku and also hold “after parties” following the events [35]. Even outside of celebrations, many seasonal events connected with locally produced foods are held in communities [35]. For these events, people do not only eat together—they also make meals together, which strengthens friendships and cohesiveness [35]. Therefore, cooking skills are indispensable for these traditional and local events, and it is conceivable that people with higher levels of cooking skills will have more opportunities to play an important role in the community. We also found significant interaction effects between cooking skills and gender on social relationships and social capital: women are more likely to benefit from social relationships through a high level of cooking skills. This finding may be explained by women cooking more frequently than men [25], creating more opportunities for women to use their cooking skills.
In line with previous studies [10,15,18], we found that women were more likely than men to have strong social relationships and social capital. A nationally representative study in Ukraine showed that the gender difference in bonding social capital, which corresponds to the frequency/number of meetings with friends in our study, was explained by age and income [10]. Using SEM analysis, we found that cooking skill mediated $26\%$ of the association between gender and social capital. Therefore, we have newly identified cooking skills as a factor contributing to explaining the gender differences in social capital.
Among the components of social capital, social cohesion was found to be weakly associated with cooking skills for women but not for men. In contrast to the other social relationships and social capital, social cohesion was the only variable that was lower in women than in men (Table S5). Social cohesion, which is categorized as cognitive social capital rather than structural social capital, may have determinants that differ from those of other aspects of social relationships. A study conducted in the Netherlands showed that perceptions that one’s neighborhood is unsafe or unattractive and low SES were associated with low social cohesion but not with social networks (e.g., visiting neighbors, asking neighbors for advice) [37]. A study in the United States showed that neighborhood safety and SES were positively related to social cohesion [38]. Low-SES groups tend to be more pessimistic and express more feelings of unsafety and neighborhood problems compared with those with higher SES [37,39]. Therefore, neighborhood safety and SES may play key roles in cognitive social capital.
This study had several limitations. First, common method bias may have occurred because cooking skills, social relationships, and social capital were assessed via the self-report questionnaire. To address this common source bias, as a sensitivity analysis, we adjusted for mental health and prosocial behaviors, which are related to the tendency to respond to the questions. It would be useful to also collect information from a second person, such as a family member or experienced food preparer who could evaluate the participants’ cooking skills. Second, gender bias may have occurred because men tend to have higher self-esteem and more positive evaluations of their own abilities than women [40]. Men may overestimate their own cooking skills and women may underestimate theirs, in which case their relationship to social relationships and social capital may lead to underestimation. Third, there may be unmeasured confounding factors, such as regional characteristics. For example, in communities where cooking classes and events involving meal preparation are popular, residents will have more opportunities to build social relationships as well as improve their cooking skills. Regional characteristics may also influence participants’ evaluation of social capital. For example, participants may not feel as engaged in society as much as they should if they belong to an active community, and vice versa. In the future, indicators of community characteristics will need to be considered. Forth, because the JAGES survey study sites were not randomly selected, the generalizability of our findings to other populations in *Japan is* limited. Additionally, the cooking skill scale in this study is limited to Japanese culture. Therefore, the results of the study may be applicable only within Japan. In the future, cooking skill scales appropriate for each culture will need to be developed to evaluate aspects of health promotion in other countries. Finally, because this study was cross-sectional, causality could not be established: reverse causation is possible, and unmeasured factors may confound the examined associations. For example, having a low level of social relationships with neighbors/friends may reduce the chances of learning cooking skills, which may lead to poor cooking skills. However, more than half of the adult respondents learned most of their cooking skills from their mothers when they were teenagers [28].
## 5. Conclusions
Our study has produced novel findings regarding the associations of cooking skills with social relationships and social capital. Considering the health benefits of social relationships and social capital, our study is of great public health importance because it has demonstrated the importance of cooking skill, a factor that can be modified through intervention to improve social relationships and social capital.
## References
1. 1.
United Nations Department of Economic and Social Affairs Population Division
World Population Prospects: The 2015 RevisionUnited NationsNew York, NY, USA2014. *World Population Prospects: The 2015 Revision* (2014.0)
2. Steptoe A., Shankar A., Demakakos P., Wardle J.. **Social isolation, loneliness, and all-cause mortality in older men and women**. *Proc. Natl. Acad. Sci. USA* (2013.0) **110** 5797-5801. DOI: 10.1073/pnas.1219686110
3. Livingston G., Sommerlad A., Orgeta V., Costafreda S.G., Huntley J., Ames D., Ballard C., Banerjee S., Burns A., Cohen-Mansfield J.. **Dementia prevention, intervention, and care**. *Lancet* (2017.0) **390** 2673-2734. DOI: 10.1016/S0140-6736(17)31363-6
4. Blazer D.. **Social Isolation and Loneliness in Older Adults—A Mental Health/Public Health Challenge**. *JAMA Psychiatry* (2020.0) **77** 990-991. DOI: 10.1001/jamapsychiatry.2020.1054
5. Berkman L.F., Glass T., Brissette I., Seeman T.E.. **From social integration to health: Durkheim in the new millennium**. *Soc. Sci. Med.* (2000.0) **51** 843-857. DOI: 10.1016/S0277-9536(00)00065-4
6. Murayama H., Fujiwara Y., Kawachi I.. **Social Capital and Health: A Review of Prospective Multilevel Studies**. *J. Epidemiol.* (2012.0) **22** 179-187. DOI: 10.2188/jea.JE20110128
7. Moore S., Kawachi I.. **Twenty years of social capital and health research: A glossary**. *J. Epidemiol. Community Health* (2017.0) **71** 513-517. DOI: 10.1136/jech-2016-208313
8. Ehsan A., De Silva M.J.. **Social capital and common mental disorder: A systematic review**. *J. Epidemiol. Community Health* (2015.0) **69** 1021-1028. DOI: 10.1136/jech-2015-205868
9. Rodgers J., Valuev A.V., Hswen Y., Subramanian S.. **Social capital and physical health: An updated review of the literature for 2007–2018**. *Soc. Sci. Med.* (2019.0) **236** 112360. DOI: 10.1016/j.socscimed.2019.112360
10. Karhina K., Eriksson M., Ghazinour M., Ng N.. **What determines gender inequalities in social capital in Ukraine?**. *SSM Popul. Health* (2019.0) **8** 100383. DOI: 10.1016/j.ssmph.2019.100383
11. Eriksson M., Dahlgren L., Janlert U., Weinehall L., Emmelin M.. **Social Capital, Gender and Educational Level Impact on Self-Rated Health**. *Open Public Health J.* (2010.0) **3** 1-12. DOI: 10.2174/1874944501003010001
12. Kawachi I., Berkman L.F.. **Social ties and mental health**. *J. Urban Health* (2001.0) **78** 458-467. DOI: 10.1093/jurban/78.3.458
13. Leevesa G., Herbert R.. **Gender differences in social capital investment: Theory and evidence**. *Econ. Model.* (2014.0) **37** 377-385. DOI: 10.1016/j.econmod.2013.11.030
14. Noguchi T., Kondo K., Saito M., Nakagawa-Senda H., Suzuki S.. **Community social capital and the onset of functional disability among older adults in Japan: A multilevel longitudinal study using Japan Gerontological Evaluation Study (JAGES) data**. *BMJ Open* (2019.0) **9** e029279. DOI: 10.1136/bmjopen-2019-029279
15. Aida J., Cable N., Zaninotto P., Tsuboya T., Tsakos G., Matsuyama Y., Ito K., Osaka K., Kondo K., Marmot M.. **Social and Behavioural Determinants of the Difference in Survival among Older Adults in Japan and England**. *Gerontology* (2018.0) **64** 266-277. DOI: 10.1159/000485797
16. Barnes-Mauthe M., Gray S.A., Arita S., Lynham J., Leung P.. **What Determines Social Capital in a Social–Ecological System? Insights from a Network Perspective**. *Environ. Manag.* (2014.0) **55** 392-410. DOI: 10.1007/s00267-014-0395-7
17. Yamagiwa J., Tsubokawa K., Inoue E., Ando C.. **Sharing fruit of Treculia africana among western gorillas in the Moukalaba-Doudou National Park, Gabon: Preliminary report**. *Primates* (2014.0) **56** 3-10. DOI: 10.1007/s10329-014-0433-3
18. Quandt S., Arcury T., Bell R., McDonald J., Vitolins M.. **The social and nutritional meaning of food sharing among older rural adults**. *J. Aging Stud.* (2001.0) **15** 145-162. DOI: 10.1016/S0890-4065(00)00023-2
19. Kohsaka R., Matsuoka H., Uchiyama Y.. **Capturing the relationships between local foods and residents: A case in the Noto region, Japan**. *J. Ethn. Foods* (2016.0) **3** 86-92. DOI: 10.1016/j.jef.2016.05.001
20. Farmer N., Touchton-Leonard K., Ross A.. **Psychosocial Benefits of Cooking Interventions: A Systematic Review**. *Health Educ. Behav. Off. Publ. Soc. Public Health Educ.* (2018.0) **45** 167-180. DOI: 10.1177/1090198117736352
21. Hartmann C., Dohle S., Siegrist M.. **Importance of cooking skills for balanced food choices**. *Appetite* (2013.0) **65** 125-131. DOI: 10.1016/j.appet.2013.01.016
22. McGowan L., Pot G.K., Stephen A.M., Lavelle F., Spence M., Raats M., Hollywood L., McDowell D., McCloat A., Mooney E.. **The influence of socio-demographic, psychological and knowledge-related variables alongside perceived cooking and food skills abilities in the prediction of diet quality in adults: A nationally representative cross-sectional study**. *Int. J. Behav. Nutr. Phys. Act.* (2016.0) **13** 111. DOI: 10.1186/s12966-016-0440-4
23. Caraher M., Dixon P., Lang T., Carr-Hill R.. **The state of cooking in England: The relationship of cooking skills to food choice**. *Br. Food J.* (1999.0) **101** 590-609. DOI: 10.1108/00070709910288289
24. Adams J., Goffe L., Adamson A.J., Halligan J., O’Brien N., Purves R., Stead M., Stocken D., White M.. **Prevalence and socio-demographic correlates of cooking skills in UK adults: Cross-sectional analysis of data from the UK National Diet and Nutrition Survey**. *Int. J. Behav. Nutr. Phys. Act.* (2015.0) **12** 99. DOI: 10.1186/s12966-015-0261-x
25. Tani Y., Fujiwara T., Kondo K.. **Cooking skills related to potential benefits for dietary behaviors and weight status among older Japanese men and women: A cross-sectional study from the JAGES**. *Int. J. Behav. Nutr. Phys. Act.* (2020.0) **17** 82. DOI: 10.1186/s12966-020-00986-9
26. Van Der Horst K., Brunner T.A., Siegrist M.. **Ready-meal consumption: Associations with weight status and cooking skills**. *Public Health Nutr.* (2011.0) **14** 239-245. DOI: 10.1017/S1368980010002624
27. Lam M.C.L., Adams J.. **Association between home food preparation skills and behaviour, and consumption of ultra-processed foods: Cross-sectional analysis of the UK National Diet and nutrition survey (2008–2009)**. *Int. J. Behav. Nutr. Phys. Act.* (2017.0) **14** 68. DOI: 10.1186/s12966-017-0524-9
28. Lavelle F., Spence M., Hollywood L., McGowan L., Surgenor D., McCloat A., Mooney E., Caraher M., Raats M., Dean M.. **Learning cooking skills at different ages: A cross-sectional study**. *Int. J. Behav. Nutr. Phys. Act.* (2016.0) **13** 119. DOI: 10.1186/s12966-016-0446-y
29. Saito M., Kondo N., Aida J., Kawachi I., Koyama S., Ojima T., Kondo K.. **Development of an instrument for community-level health related social capital among Japanese older people: The JAGES Project**. *J. Epidemiol.* (2017.0) **27** 221-227. DOI: 10.1016/j.je.2016.06.005
30. Tamiya N., Noguchi H., Nishi A., Reich M.R., Ikegami N., Hashimoto H., Shibuya K., Kawachi I., Campbell J.C.. **Population ageing and wellbeing: Lessons from Japan’s long-term care insurance policy**. *Lancet* (2011.0) **378** 1183-1192. DOI: 10.1016/S0140-6736(11)61176-8
31. Sasaki Y., Tsuji T., Koyama S., Tani Y., Saito T., Kondo K., Kawachi I., Aida J.. **Neighborhood Ties Reduced Depressive Symptoms in Older Disaster Survivors: Iwanuma Study, a Natural Experiment**. *Int. J. Environ. Res. Public Health* (2020.0) **17**. DOI: 10.3390/ijerph17010337
32. Tani Y., Kondo N., Takagi D., Saito M., Hikichi H., Ojima T., Kondo K.. **Combined effects of eating alone and living alone on unhealthy dietary behaviors, obesity and underweight in older Japanese adults: Results of the JAGES**. *Appetite* (2015.0) **95** 1-8. DOI: 10.1016/j.appet.2015.06.005
33. Wada T., Ishine M., Kita T., Fujisawa M., Matsubayashi K.. **Depression screening of elderly community-dwelling Japanese**. *J. Am. Geriatr. Soc.* (2003.0) **51** 1328-1329. PMID: 12919256
34. Koyano W., Shibata H., Nakazato K., Haga H., Suyama Y.. **Measurement of competence: Reliability and validity of the TMIG Index of Competence**. *Arch. Gerontol. Geriatr.* (1991.0) **13** 103-116. DOI: 10.1016/0167-4943(91)90053-S
35. **WASHOKU**
36. Fuchigami M., Kuwada H., Ishii K., Kimura Y.. **Regionality of Cooking Culture and Cooking Science: Event Food / Ritual Food**. *J. Cook. Sci. Jpn.* (2011.0) **44** 436-441
37. Kamphuis C.B.M., MacKenbach J.P., Giskes K., Huisman M., Brug J., Van Lenthe F.J.. **Why do poor people perceive poor neighbourhoods? The role of objective neighbourhood features and psychosocial factors**. *Health Place* (2010.0) **16** 744-754. DOI: 10.1016/j.healthplace.2010.03.006
38. Yuma-Guerrero P.J., Cubbin C., von Sternberg K.. **Neighborhood Social Cohesion as a Mediator of Neighborhood Conditions on Mothers’ Engagement in Physical Activity: Results from the Geographic Research on Wellbeing Study**. *Health Educ. Behav.* (2017.0) **44** 845-856. DOI: 10.1177/1090198116687537
39. Gallo L.C., Matthews K.A.. **Understanding the association between socioeconomic status and physical health: Do negative emotions play a role?**. *Psychol. Bull.* (2003.0) **129** 10-51. DOI: 10.1037/0033-2909.129.1.10
40. Kearney-Cooke A.. **Gender differences and self-esteem**. *J. Gender-Specific Med. JGSM Off. J. Partnersh. Women’s Health Columbia* (1999.0) **2** 46-52
|
---
title: 'Systematic analysis of myocardial immune progression in septic cardiomyopathy:
Immune-related mechanisms in septic cardiomyopathy'
authors:
- Dunliang Ma
- Xianyu Qin
- Zhi-an Zhong
- Hongtao Liao
- Pengyuan Chen
- Bin Zhang
journal: Frontiers in Cardiovascular Medicine
year: 2023
pmcid: PMC10002421
doi: 10.3389/fcvm.2022.1036928
license: CC BY 4.0
---
# Systematic analysis of myocardial immune progression in septic cardiomyopathy: Immune-related mechanisms in septic cardiomyopathy
## Abstract
### Background
The immune infiltration and molecular mechanisms underlying septic cardiomyopathy (SC) have not been completely elucidated. This study aimed to identify key genes related to SC and elucidate the potential molecular mechanisms.
### Methods
The weighted correlation network analysis (WGCNA), linear models for microarray analysis (LIMMA), protein-protein interaction (PPI) network, CIBERSORT, Kyoto Encyclopedia of Genes and Genomes pathway (KEGG), and gene set enrichment analysis (GSEA) were applied to assess the key pathway and hub genes involved in SC.
### Results
We identified 10 hub genes, namely, LRG1, LCN2, PTX3, E LANE, TCN1, CLEC4D, FPR2, MCEMP1, CEACAM8, and CD177. Furthermore, we used GSEA for all genes and online tools to explore the function of the hub genes. Finally, we took the intersection between differential expression genes (DEGs) and hub genes to identify LCN2 and PTX3 as key genes. We found that immune-related pathways played vital roles in SC. LCN2 and PTX3 were key genes in SC progression, which mainly showed an anti-inflammatory effect. The significant immune cells in cardiomyocytes of SC were neutrophils and M2 macrophages.
### Conclusion
These cells may have the potential to be prognostic and therapeutic targets in the clinical management of SC. Excessive anti-inflammatory function and neutrophil infiltration are probably the primary causes of SC.
## 1. Introduction
Researchers recognize sepsis as a life-threatening condition that is caused by a dysregulated host response to infection [1]. It is the immune response of the organism to pathogens and immunogenic substances, causing autoimmune damage. Sepsis is common in severe health conditions, and the development of sepsis may lead to septic shock and multiple organ dysfunction syndromes (MODS), which once occurs, mortality can be up to 28–$56\%$. The heart is the main target organ of sepsis, and more than $50\%$ of patients with severe sepsis have myocardial dysfunction, which is called septic cardiomyopathy (SC) [2]. However, due to a lack of uniform diagnostic criteria, the prevalence of SC varies in different reports. Beesley et al. [ 3] reported the incidence of myocardial dysfunction in sepsis patients as ranging from 10 to $70\%$.
Inflammatory responses and immune cell infiltration widely exist in many types of cardiomyopathy. For example, in heart tissue with diabetic cardiomyopathy, inflammatory responses have been found to be significantly activated, as manifested by infiltration of multiple immune cells, increased cytokines, and multiple chemical factors [4]. Similar results have also been confirmed in animal models [5, 6]. In the state of diabetes mellitus, macrophages may induce tissue infiltration, transform into the proinflammatory phenotype of M1, and be associated with the activation of inflammatory signaling pathways in leukocytes [7]. Thus, inflammatory responses are closely related to cardiac function. Myocardial dysfunction can increase sepsis-induced mortality, but no reports have elucidated the underlying pathophysiological mechanisms of SC. The molecular mechanism that may be involved in the pathogenesis of SC remains to be studied, and there is a need to screen potential targets for the treatment of SC. Among the pathogenic factors contributing to SC, the imbalanced inflammatory responses caused by sepsis directly correlate with the dysfunction of myocardial cells. Previous studies have reported that sepsis begins with the host immune system’s response to invasive pathogens, eventually leading to activation of the innate immune response [8]. Bacterial products, including endotoxins and exotoxins, can directly or indirectly stimulate various target cells, including monocytes, polymorphonuclear neutrophils, or endothelial cells, thereby causing inflammation [9]. Endotoxins and exotoxins, through varied signal transduction pathways, activate both positive and negative feedback loops within the immune system. Sepsis-induced dysregulation of the normal immune response can lead to a variety of harmful effects, including SC. Therefore, a thorough understanding of the molecular immune mechanism involved in the pathogenesis of SC could be one of the breakthroughs that may help in the treatment of SC in the future.
In this study, we downloaded the gene expression profile (GSE79962) deposited by Matkovich et al. [ 10] from Gene Expression Omnibus (GEO) databases to uncover further the biomarkers associated with SC development and progression. We aimed to identify key genes related to SC, as well as to further elucidate the potential molecular mechanisms through bioinformatics analysis.
## 2.1. Data
We downloaded microarray data GSE79962 from the NCBI Gene Expression Omnibus database (GEO).1 *The data* involved ischemic heart disease (IHD, 11 samples), non-failing heart (control, 11 samples), dilated cardiomyopathy (DCM, 9 samples), and septic cardiomyopathy (SC, 20 samples). We chose all the samples in the study. We downloaded the annotation information of the microarray, GPL6244, Affymetrix Human Gene 1.0 ST Array [transcript (gene) version]. We preprocessed the raw data using R version 3.6.0. The analysis workflow is presented in Figure 1.
**FIGURE 1:** *Workflow used for bioinformatics analyses.*
## 2.2. WGCNA network construction
We constructed co-expression networks using the weighted correlation network analysis (WGCNA) package in R [11]. We did not filter genes. We imported gene expression values into WGCNA to create co-expression modules using the automatic network construction function blockwiseModules with default settings. We set the power value by the condition of scale independence as 0.9. The unsigned TOMType mergeCutHeight was 0.25, and the minModuleSize was 50.
## 2.3. Module and gene selection
To find biologically or clinically significant modules for SC, module eigengenes were used to calculate the correlation coefficient with samples. We calculated the intramodular connectivity (function softConnectivity) of each gene. We thought genes with high connectivity tended to be hub genes which might have essential functions. We imported the positive correlation modules into Cytoscape software (version 3.7.1), using the MCODE plugin, setting the degree cutoff at no less than 10, to screen the key sub-modules.
## 2.4. Functional analysis of module genes
Because the three modules were all positively correlated to SC, we imported all genes from these three sub-modules into the STRING database (version 11.0).2 We obtained the results of gene ontology (GO) enrichment analysis, Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis, and protein-protein interaction (PPI). Significantly enriched GO terms and pathways in genes in a module compared to the background were defined by hypergeometric tests and a threshold of a minimum required interaction score 0.700 (high confidence). After that, we imported a PPI network into Cytoscape software (version 3.7.1), using the cytoHubba plugin to screen the hub genes by 12 methods. We performed functional analysis and disease prediction of hub genes through the online tools Metascape database3 and ToppGene database.4 We used the “limma” package of the R language to identify the differential expression genes (DEGs) between the SC and control groups, according to the adjusted p-value < 0.05 and |logFC| > 0.7. We screened the shared genes between DEGs and hub genes as key genes.
## 2.5. GSEA and immune infiltration analysis using the CIBERSORT method
Meanwhile, we performed the gene set enrichment analysis (GSEA) with hallmark gene sets using the 11 control samples and 20 SC samples according to the default values. The criteria of significant results were set as normal enrichment score (|NES| > 1), normal p-value < 0.05, and FDR < 0.25. To characterize the immune cell subtypes in SC progression, we applied the CIBERSORT estimate software5 to quantify the immune cell fractions for the gene expression matrix derived from SC samples. Then, we identified the correlation between the hub genes and immune cell subtypes.
## 3.1. Screening the key modules in the network
Expression data of 18,818 genes in the 51 samples were screened. These samples included SC, control (non-failing donor heart), ischemic heart disease (IHD), and dilated cardiomyopathy (DCM). They were used to construct the co-expression modules with the WGCNA algorithm. Following the data preprocessing, we set the power value. The power value was four when the condition of scale independence was 0.9 (Supplementary Figure 1). We clustered genes into 26 correlated modules. We tried to identify sample-associated co-expression modules using WGCNA (Figures 2A, B). At last, we got 18 co-expression modules, which were illustrated in the branches of a dendrogram with different colors. We focused just on the SC group. Therefore, we chose the dark green module (Pearson cor = 0.69, $$p \leq 3$$e–8), blue module (Pearson cor = 0.83, $$p \leq 3$$e–14), and orange module (Pearson cor = 0.74, $$p \leq 4$$e–10), as moderately or more positively related with SC. The number of genes in the three modules was 2652 (blue), 73 (dark green), and 2,041 (orange). The information about the genes in the three modules is listed in Supplementary Table A1. The relationship of module membership to gene significance in the modules showed was cor = 0.9 and $p \leq 1$e-200 in the blue module (Figure 2C), cor = 0.9 and $$p \leq 3.3$$e-34 in the dark green module (Figure 2D), and cor = 0.79 and $p \leq 1$e-200 in the orange module (Figure 2E). We imported the three modules into the Cytoscape software and used the MCODE to screen the three sub-modules, setting the criteria of the MCODE score to more than 10. After screening, we got 58 genes from the orange module (average MCODE score = 33.98), 37 genes from the blue module (average MCODE score = 14.23), and 53 genes from the dark green module (average MCODE score = 41.16) (Supplementary Table A2).
**FIGURE 2:** *Overview of WGCNA network construction of all genes (A) Gene modules’ dendrogram plots of all genes; (B) module-trait relationships of four groups in 18 modules. (C–E) Module membership vs. gene significance between three significant modules, including blue module (Pearson cor = 0.9, p < 1e-200), dark green module (cor = 0.9, p = 3.3e-34), and orange module (Pearson cor = 0.79, p < 1e-200); (F) the bubble diagram showing the GO (biological process, BP) function enrichment of genes in sub-modules. The size represents the gene counts, and node colors show the gene expression negative Log10_FDR (false discovery rate).*
## 3.2. Functional enrichment analysis of genes in critical modules
We imported all of the screened genes (a total of 148 genes) from the three modules into the STRING database to construct a PPI network. Meanwhile, we got GO enrichment analysis results according to a false discovery rate (FDR) < 0.05.
We obtained the top 10 biological process (BP) terms, including neutrophil activation (FDR = 1.29E–22), neutrophil degranulation (FDR = 2.65E–22), regulated exocytosis (FDR = 6.34E–22), exocytosis (FDR = 2.02E–21), leukocyte mediated immunity (FDR = 2.20E–21), leukocyte activation involved in immune response (FDR = 9.78E–21), immune response (9.07E–20), leukocyte activation (FDR = 1.86E–18), secretion (FDR = 1.86E–18), and cell activation (FDR = 2.56E–18) (Figure 2F). We obtained GO terms of the top 10 cellular components (CC), consisting of secretory granule (FDR = 2.49E–19), cytoplasmic vesicle part (FDR = 1.34E–15), cytoplasmic vesicle (FDR = 4.74E–13), vesicle (FDR = 4.74E–13), tertiary granule (FDR = 4.74E–13), specific granule (FDR = 3.63E–11), secretory granule lumen (FDR = 6.73E–11), secretory granule membrane (FDR = 1.42E–10), ficolin-1 rich granule (FDR = 1.82E–10), and endomembrane system (FDR = 1.33E–09) (Figure 3A). Meanwhile, we harvested GO terms of 8 molecular functions (MFs), comprising protein binding (FDR = 0.00028), cytokine binding (FDR = 0.00028), enzyme binding (FDR = 0.0018), signaling receptor binding (FDR = 0.0124), CXCR chemokine receptor binding (FDR = 0.0297), protease binding (FDR = 0.0361), cytokine receptor activity (FDR = 0.0361), and pantetheine hydrolase activity (FDR = 0.0361; Figure 3B). Regarding the KEGG pathway enrichment, the genes were significantly enriched in pathways including neutrophil degranulation (FDR = 4.58E–23), innate immune system (FDR = 1.30E–16), immune system (FDR = 1.30E–16), signaling by interleukins (FDR = 8.52E–06), cytokine signaling in immune system (FDR = 0.00025), and others (Figure 3C and Supplementary Table A3).
**FIGURE 3:** *Analysis of gene ontology (GO) function, Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways, and protein-protein interaction (PPI) network of genes in septic cardiomyopathy (SC)-related modules. (A–C) The bubble map showing the GO function (cellular component, CC, and molecular function, MF) and KEGG pathways were constructed by the STRING database. The sizes represent negative Log10 (FDR). (D) The gene PPI network was also constructed based on the STRING database. (E) The plots showing the top 10 higher degree hub genes for SC.*
We used Cytoscape software to visualize the PPI network from the STRING database, which is shown in Figure 3D. Through the cytoHubba plugin, we exported the results of 12 algorithms and screened the top 10 genes as hub genes. These included LRG1, LCN2, PTX3, ELANE, TCN1, FPR2, CLEC4D, MCEMP1, CEACAM8, and CD177 (Figure 3E and Supplementary Table A4).
## 3.3. GSEA and immunocyte infiltration analysis
The results of GO and KEGG enrichment analysis also indicated that immune and inflammatory events played a vital role in cardiac tissue in SC. Furthermore, the results of GSEA with hallmark gene sets between the control and SC indicated significant differences in the myocardium of SC, such as TGF-beta signaling, TNF-alpha signaling through NF-kappa B, inflammatory response, and TNF and P53 pathways (we set the criteria |NES| > 1, NOM $p \leq 0.01$, FDR < 0.25; Figure 4A and Table 1). To characterize the immunocyte status of cardiac tissues in SC, we performed a tissue immune infiltration analysis. We found that there was no significant difference in 22 immune subtypes in the myocardium of SC and the control, except for M2 macrophages (Wilcoxon, $$p \leq 0.032$$) and neutrophils (Wilcoxon, $$p \leq 0.00064$$; Figures 4C, D). A heatmap of immune cell subtypes is illustrated in Figure 4B.
**FIGURE 4:** *Immunocyte infiltration analysis and potential immunocyte subtype detection. (A) This diagram shows the immune-related pathways by gene set enrichment analysis (GSEA) analysis. (B) This heatmap shows the immunocyte infiltration difference between SC and control heart samples. (C,D) Box plots presenting significantly infiltrated immunocyte subtypes, neutrophils, and M2 macrophages. (E) This correlated heatmap shows the relationship between immunocytes and hub genes. (F) Intersection between DEGs and hub genes, identifying the key genes, LCN2, and PTX3.* TABLE_PLACEHOLDER:TABLE 1
## 3.4. Identification of the relationship between hub genes and key immune cell subtypes in SC
We found that two subtypes of immune cells in the infiltration were significant. These were neutrophils and M2 macrophages. Importantly, the neutrophils had the highest relative infiltration value. We analyzed the relationship between the 10 hub genes in immune-related pathways and the two immune cell subtypes and found the neutrophils had positive correlations with all hub genes, especially with CLEC4D, CD177, LCN2, and TCN1, but M2 macrophages had negative correlations with all hub genes and neutrophils; Figure 4E). This suggests that to some extent, neutrophils can promote the progression of SC in the myocardium, but that M2 macrophages do the opposite.
## 3.5. Investigating the functional role of hub genes and identification of key genes
To further understand how the hub genes were correlated with SC, we applied the Metascape online database to explore their biological functions. The results of the top five GO term enrichments in hub genes included neutrophil degranulation, neutrophil activation involved in immune response, neutrophil activation, neutrophil mediated immunity, and granulocyte activation. The results showed all hub genes are involved in the neutrophil and immune response (Table 2). We explored enriched pathways of hub genes involved in neutrophil degranulation, the innate immune system, antimicrobial peptides, and similar topics (Supplementary Table A5). We also predicted diseases related to hub genes using the ToppGene database. These diseases included sepsis, immune neutropenia, septicemia, and others (Table 3 and Supplementary Table A5). We found 467 DEGs between the SC group and the control (Supplementary Table A6). Finally, we took the intersection between DEGs and hub genes to identify LCN2 and PTX3 as key genes (Figure 4F). As for the key genes, we compared the expression levels of LCN2 and PTX3 in different kinds of myocardial injury from these samples. We found LCN2 and PTX3 had significantly higher expression in the SC group (Supplementary Figure 2).
## 4. Discussion
In this study, we screened key modules in SC by analyzing a public dataset (GSE79962). Compared with the control group, IHD group, and DCM group, a total of three modules were positively correlated with SC. These included the orange module, blue module, and dark green module. We chose 148 genes from the three modules using the MCODE plugin of Cytoscape for functional enrichment analysis. Notably, we found most genes in the three modules were enriched in the immune response, leukocyte activation, neutrophil degranulation, and similar events.
Regarding the KEGG pathway enrichment, the genes were significantly enriched in immune-related pathways, including neutrophil degranulation, the innate immune system, the immune system, and cytokine signaling in the immune system. The results of GSEA with hallmark gene sets between control and SC indicated that significantly different pathways in the myocardium of SC were immune-related. Neutrophils degranulated during phagocytosis, releasing a series of lysosomal enzymes, which caused damage to blood vessels and surrounding tissues, leading to cardiac dysfunction [12]. Sepsis leads to an auto-amplifying cytokine production known as the cytokine storm. At the same time, activation of Toll-like receptors (TLRs) releases a large number of inflammatory cytokines, such as TNF, IL-1, interferon regulatory factor 3 (IRF3) [13]. The activation of these immune responses leads to damage of myocardial tissue and cardiac dysfunction. Previous studies have reported that TLRs can attenuate SC through activation of innate immune and inflammatory responses [14, 15]. TLR is a kind of pattern recognition receptor that can activate the innate immunity response, playing a critical role through activation of NFκB which is an important transcription factor controlling the expression of inflammatory cytokine genes [16]. TLRs play a major role in the pathophysiology of cardiac dysfunction during sepsis [14]. In an animal model, TLR2 was found that can influence cardiac function through deteriorating sarcomere shortening [17, 18]. TLRs deficiency attenuated cardiac dysfunction in a mouse model through inhibition of sepsis-induced activation of TLR4 mediated NF-κB signaling pathways, and prevention of the macrophage and neutrophil infiltration. In addition, lipopolysaccharide (LPS) has been demonstrated to induced macrophage inflammation through TLRs, leading to the release of proinflammatory cytokines [19]. In patients with sepsis, increased serum lactate levels increased mortality through activation of innate immune and inflammatory responses [20, 21]. Through the CIBERSORT method, we found that the infiltration value of neutrophils and M2 macrophages in the myocardium of SC is significant. Therefore, we found that immunity and inflammation play important roles in myocardial dysfunction in SC. In this condition, neutrophils were positively correlated with immune-related genes, and M2 macrophages were negatively correlated with immune-related genes.
Macrophages are the “frontier soldiers” of innate immunity. The function of macrophages is classified into two types, type M1 (classically activated) and type M2 (alternatively activated). Type M1 macrophages can secrete chemokines for a proinflammatory function, and type M2 macrophages mainly secrete chemokines in the late stage of inflammation to play an anti-inflammatory role [22, 23]. M2 macrophages mainly promote tissue remodeling and repair, and previous studies showed that an increase in M2 macrophage infiltration in myocardium promotes fibrosis (23–25). The decrease of M2 macrophages in SC leads to a reduction of anti-inflammatory chemokines and supports the progression of SC. The polarization of the M1 macrophage is mainly regulated by transcription factors IRF5 and STAT1, and the M2 macrophage is regulated by IRF4 and STAT6 [26]. Many immunomodulators can promote M1 macrophage polarization, such as IFN, TNF, IL-1, IL-6, LPS, B-cell activator (BAFF), and proliferation-inducing ligand (APRIL) (26–29). In addition, some metabolites, such as saturated fatty acids and oxidized lipoproteins, can also induce M1 macrophage polarization [27]. Similarly, inflammatory factors, such as IL-4, IL-13, IL-10, IL-33, and TGF, as well as metabolites such as unsaturated fatty acids and retinoic acids, can induce M2 macrophage polarization [30, 31]. Likewise, neutrophils are key factors in the immune response to sepsis. Under normal conditions, neutrophils control infection, but excessive stimulation or dysregulated neutrophil functions are believed to be responsible for sepsis pathogenesis [12]. In SC, we found significant neutrophil infiltration in cardiac tissues.
We screened 10 hub genes from the PPI network constructed from 148 genes. These 10 genes are also involved in several immune-related pathways directly, which include LRG1, LCN2, PTX3, ELANE, TCN1, CLEC4D, FPR2, MCEMP1, CEACAM8, and CD177. Next, we used online tools (the ToppGene and Metascape databases) to explore the function of the hub genes. The results showed that the hub genes were related to immune-related pathways and diseases. Among these genes, LRG1 is highly correlated with neutrophils and other genes composed of TCN1, FPR2, CLEC4D, and CD177. LRG1 is expressed during granulocyte differentiation. From the GeneCards database,6 we found that the super pathway for LRG1 is the innate immune system. BP terms included response to bacterium, positive regulation of transforming growth factor-beta receptor signaling pathway, neutrophil degranulation, and similar terms. Recombinant human LRG is used as a diagnostic aid in acute appendicitis [32]. Similarly, LRG1 may be used as a diagnostic marker for SC.
In our research, we found that LCN2 and PTX3 in 10 hub genes existed in DEGs, as key genes. LCN2, encoding the lipocalin-2 (also known as neutrophil gelatinase-associated lipocalin), is a critical iron regulatory protein during physiological and inflammatory conditions and exerts mostly a protective role in inflammatory bowel diseases and urinary tract infection by limiting bacterial growth [33, 34]. In the heart, some reports also indicated that LCN2 was significantly expressed during in vivo and in vitro experiments on cardiac hypertrophy and heart failure, and high plasma LCN2 was correlated with high mortality and myocardial dysfunction in severe sepsis [35, 36]. Nevertheless, Guo et al. [ 37] found that LCN2–/– mice displayed an up-regulation of M1 macrophages but down-regulation of M2 macrophages. These mice had profound up-regulation of proinflammatory cytokines, suggesting that LCN2 plays a role as an anti-inflammatory regulator in macrophage activation. Overexpression of LCN2 is consistent with down-regulation of M2 macrophages in SC. PTX3 plays a role in the regulation of innate resistance to pathogens and inflammatory reactions. Paeschke et al. [ 38] showed that inflammatory injury of heart tissue was aggravated in mice when PTX3 was knocked down. Yamazaki et al. [ 39] demonstrated that bacterial LPS, induced expression of anti-microbial glycoproteins-PTX3 and LCN2 in macrophages. Therefore, we concluded that LCN2 and PTX3 might lead to excessive anti-inflammatory effects for SC progression.
Our study has some limitations. First, the sample size of this study is relatively small and additional clinical samples are necessary. However, we barely obtain more, due to the difficulty in obtaining SC samples. Besides, despite that LCN2 and PTX2 are related to neutrophil function as reported, there is no direct evidence to validate that LCN2 and PTX3 are involved in the progression of SC. Simultaneously, the underlying mechanism by which LCN2 and PTX3 affect SC remains unclear. Last but not the least, although our conclusion is based on bioinformatics analysis, more experimental results will help to increase the reliability of the results. We expect further understanding of the regulatory functions of key genes on SC through other means.
## 5. Conclusion
To sum up, we found that genes in three modules played vital roles in the immune-related pathways. LCN2 and PTX3 were key genes in SC progression and mainly showed anti-inflammatory effects. The significant immune cells in cardiomyocytes of SC were neutrophils and M2 macrophages. Therefore, LCN2 and PTX3 may have the potential to perform as prognostic and therapeutic targets in the clinical management of SC. Excessive anti-inflammatory function and neutrophil infiltration were probably the primary causes of SC, but this needs further analysis.
## Data availability statement
The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found in the article/Supplementary material.
## Author contributions
DM, XQ, and Z-AZ analyzed the data and drafted the manuscript. HL and PC edited the manuscript. BZ supervised the project, gave advice regarding the project design, and edited the manuscript. All authors contributed to the article and approved the submitted version.
## Conflict of interest
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
## Publisher’s note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
## Supplementary material
The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fcvm.2022.1036928/full#supplementary-material
## References
1. Shankar-Hari M, Phillips G, Levy M, Seymour C, Liu V, Deutschman C. **Developing a new definition and assessing new clinical criteria for septic shock: for the third international consensus definitions for sepsis and septic shock (sepsis-3).**. (2016) **315** 775-87. DOI: 10.1001/jama.2016.0289
2. Sharawy N, Lehmann C. **New directions for sepsis and septic shock research.**. (2015) **194** 520-7. DOI: 10.1016/j.jss.2014.12.014
3. Beesley S, Weber G, Sarge T, Nikravan S, Grissom C, Lanspa M. **Septic cardiomyopathy.**. (2018) **46** 625-34. DOI: 10.1097/CCM.0000000000002851
4. Dinh W, Füth R, Nickl W, Krahn T, Ellinghaus P, Scheffold T. **Elevated plasma levels of TNF-alpha and interleukin-6 in patients with diastolic dysfunction and glucose metabolism disorders.**. (2009) **8**. DOI: 10.1186/1475-2840-8-58
5. Mariappan N, Elks C, Sriramula S, Guggilam A, Liu Z, Borkhsenious O. **NF-kappaB-induced oxidative stress contributes to mitochondrial and cardiac dysfunction in type II diabetes.**. (2010) **85** 473-83. DOI: 10.1093/cvr/cvp305
6. Tan X, Hu L, Shu Z, Chen L, Li X, Du M. **Role of CCR2 in the development of streptozotocin-treated diabetic cardiomyopathy.**. (2019) **68** 2063-73. DOI: 10.2337/db18-1231
7. Ururahy M, Loureiro M, Freire-Neto F, de Souza K, Zuhl I, Brandão-Neto J. **Increased TLR2 expression in patients with type 1 diabetes: evidenced risk of microalbuminuria.**. (2012) **13** 147-54. DOI: 10.1111/j.1399-5448.2011.00794.x
8. Russell J, Boyd J, Nakada T, Thair S, Walley K. **Molecular mechanisms of sepsis.**. (2011) **17** 48-85. DOI: 10.1159/000324009
9. Tonnesen E, Larsen K. **[Severe sepsis and septic shock].**. (2014) **176**
10. Matkovich S, Al Khiami B, Efimov I, Evans S, Vader J, Jain A. **Widespread down-regulation of cardiac mitochondrial and sarcomeric genes in patients with sepsis.**. (2017) **45** 407-14. DOI: 10.1097/CCM.0000000000002207
11. Langfelder P, Horvath S. **WGCNA: an R package for weighted correlation network analysis.**. (2008) **9**. DOI: 10.1186/1471-2105-9-559
12. Brown K, Brain S, Pearson J, Edgeworth J, Lewis S, Treacher D. **Neutrophils in development of multiple organ failure in sepsis.**. (2006) **368** 157-69. DOI: 10.1016/S0140-6736(06)69005-3
13. Kawai T, Akira S. **Toll-like receptors and their crosstalk with other innate receptors in infection and immunity.**. (2011) **34** 637-50. DOI: 10.1016/j.immuni.2011.05.006
14. Gao M, Ha T, Zhang X, Liu L, Wang X, Kelley J. **Toll-like receptor 3 plays a central role in cardiac dysfunction during polymicrobial sepsis.**. (2012) **40** 2390-9. DOI: 10.1097/CCM.0b013e3182535aeb
15. Williams D, Ha T, Li C, Kalbfleisch J, Schweitzer J, Vogt W. **Modulation of tissue Toll-like receptor 2 and 4 during the early phases of polymicrobial sepsis correlates with mortality.**. (2003) **31** 1808-18. DOI: 10.1097/01.CCM.0000069343.27691.F3
16. Chousterman B, Swirski F, Weber G. **Cytokine storm and sepsis disease pathogenesis.**. (2017) **39** 517-28. DOI: 10.1007/s00281-017-0639-8
17. Zhang X, Lu C, Gao M, Cao X, Ha T, Kalbfleisch J. **Toll-like receptor 4 plays a central role in cardiac dysfunction during trauma hemorrhage shock.**. (2014) **42** 31-7. DOI: 10.1097/SHK.0000000000000155
18. Zou L, Feng Y, Chen Y, Si R, Shen S, Zhou Q. **Toll-like receptor 2 plays a critical role in cardiac dysfunction during polymicrobial sepsis.**. (2010) **38** 1335-42. DOI: 10.1097/CCM.0b013e3181d99e67
19. Yang H, Ye X, Zhang X, Li X, Fu Q, Tang Z. **Intracellular osteopontin negatively regulates toll-like receptor 4-mediated inflammatory response via regulating GSK3β and 4EBP1 phosphorylation.**. (2018) **108** 89-95. DOI: 10.1016/j.cyto.2018.03.013
20. Chertoff J, Chisum M, Garcia B, Lascano J. **Lactate kinetics in sepsis and septic shock: a review of the literature and rationale for further research.**. (2015) **3**. DOI: 10.1186/s40560-015-0105-4
21. Samuvel D, Sundararaj K, Nareika A, Lopes-Virella M, Huang Y. **Lactate boosts TLR4 signaling and NF-kappaB pathway-mediated gene transcription in macrophages via monocarboxylate transporters and MD-2 up-regulation.**. (2009) **182** 2476-84. DOI: 10.4049/jimmunol.0802059
22. Derlindati E, Dei Cas A, Montanini B, Spigoni V, Curella V, Aldigeri R. **Transcriptomic analysis of human polarized macrophages: more than one role of alternative activation?**. (2015) **10**. DOI: 10.1371/journal.pone.0119751
23. Shivshankar P, Halade G, Calhoun C, Escobar G, Mehr A, Jimenez F. **Caveolin-1 deletion exacerbates cardiac interstitial fibrosis by promoting M2 macrophage activation in mice after myocardial infarction.**. (2014) **76** 84-93. DOI: 10.1016/j.yjmcc.2014.07.020
24. Hu D, Dong R, Yang Y, Chen Z, Tang Y, Fu M. **Human kallikrein overexpression alleviates cardiac aging by alternatively regulating macrophage polarization in aged rats.**. (2019) **33** 8436-52. DOI: 10.1096/fj.201802371RR
25. Carlson S, Helterline D, Asbe L, Dupras S, Minami E, Farris S. **Cardiac macrophages adopt profibrotic/M2 phenotype in infarcted hearts: role of urokinase plasminogen activator.**. (2017) **108** 42-9. DOI: 10.1016/j.yjmcc.2016.05.016
26. Mosser D, Edwards J. **Exploring the full spectrum of macrophage activation.**. (2008) **8** 958-69. DOI: 10.1038/nri2448
27. Chan K, Pillon N, Sivaloganathan D, Costford S, Liu Z, Theret M. **Palmitoleate reverses high fat-induced proinflammatory macrophage polarization via AMP-activated protein kinase (AMPK).**. (2015) **290** 16979-88. DOI: 10.1074/jbc.M115.646992
28. Martinez F, Gordon S. **The M1 and M2 paradigm of macrophage activation: time for reassessment.**. (2014) **6**. DOI: 10.12703/P6-13
29. Takahashi H, Hatta Y, Iriyama N, Hasegawa Y, Uchida H, Nakagawa M. **Induced differentiation of human myeloid leukemia cells into M2 macrophages by combined treatment with retinoic acid and 1α,25-dihydroxyvitamin D3.**. (2014) **9**. DOI: 10.1371/journal.pone.0113722
30. Liao X, Sharma N, Kapadia F, Zhou G, Lu Y, Hong H. **Krüppel-like factor 4 regulates macrophage polarization.**. (2011) **121** 2736-49. DOI: 10.1172/JCI45444
31. Liew F, Girard J, Turnquist H. **Interleukin-33 in health and disease.**. (2016) **16** 676-89. DOI: 10.1038/nri.2016.95
32. Kentsis A, Ahmed S, Kurek K, Brennan E, Bradwin G, Steen H. **Detection and diagnostic value of urine leucine-rich α-2-glycoprotein in children with suspected acute appendicitis.**. (2012) **60** 78-83.e1. DOI: 10.1016/j.annemergmed.2011.12.015
33. Xiao X, Yeoh B, Vijay-Kumar M. **Lipocalin 2: an emerging player in iron homeostasis and inflammation.**. (2017) **37** 103-30. DOI: 10.1146/annurev-nutr-071816-064559
34. Shields-Cutler R, Crowley J, Miller C, Stapleton A, Cui W, Henderson J. **Human metabolome-derived cofactors are required for the antibacterial activity of siderocalin in urine.**. (2016) **291** 25901-10. DOI: 10.1074/jbc.M116.759183
35. Marques F, Prestes P, Byars S, Ritchie S, Würtz P, Patel S. **Experimental and human evidence for lipocalin-2 (neutrophil gelatinase-associated lipocalin [NGAL]) in the development of cardiac hypertrophy and heart failure.**. (2017) **6**. DOI: 10.1161/JAHA.117.005971
36. Wang B, Chen G, Li J, Zeng Y, Wu Y, Yan X. **Neutrophil gelatinase-associated lipocalin predicts myocardial dysfunction and mortality in severe sepsis and septic shock.**. (2017) **227** 589-94. DOI: 10.1016/j.ijcard.2016.10.096
37. Guo H, Jin D, Chen X. **Lipocalin 2 is a regulator of macrophage polarization and NF-κB/STAT3 pathway activation.**. (2014) **28** 1616-28. DOI: 10.1210/me.2014-1092
38. Paeschke A, Possehl A, Klingel K, Voss M, Voss K, Kespohl M. **The immunoproteasome controls the availability of the cardioprotective pattern recognition molecule Pentraxin3.**. (2016) **46** 619-33. DOI: 10.1002/eji.201545892
39. Yamazaki S, Akira S, Sumimoto H. **Glucocorticoid augments lipopolysaccharide-induced activation of the IκBζ-dependent genes encoding the anti-microbial glycoproteins lipocalin 2 and pentraxin 3.**. (2015) **157** 399-410. DOI: 10.1093/jb/mvu086
|
---
title: Increased Expression of Autophagy-Related Genes in Alzheimer’s Disease—Type
2 Diabetes Mellitus Comorbidity Models in Cells
authors:
- Clara Vianello
- Marco Salluzzo
- Daniela Anni
- Diana Boriero
- Mario Buffelli
- Lucia Carboni
journal: International Journal of Environmental Research and Public Health
year: 2023
pmcid: PMC10002426
doi: 10.3390/ijerph20054540
license: CC BY 4.0
---
# Increased Expression of Autophagy-Related Genes in Alzheimer’s Disease—Type 2 Diabetes Mellitus Comorbidity Models in Cells
## Abstract
The association between Alzheimer’s disease (AD) and type 2 diabetes mellitus (T2DM) has been extensively demonstrated, but despite this, the pathophysiological mechanisms underlying it are still unknown. In previous work, we discovered a central role for the autophagy pathway in the common alterations observed between AD and T2DM. In this study, we further investigate the role of genes belonging to this pathway, measuring their mRNA expression and protein levels in 3xTg-AD transgenic mice, an animal model of AD. Moreover, primary mouse cortical neurons derived from this model and the human H4Swe cell line were used as cellular models of insulin resistance in AD brains. Hippocampal mRNA expression showed significantly different levels for Atg16L1, Atg16L2, GabarapL1, GabarapL2, and *Sqstm1* genes at different ages of 3xTg-AD mice. Significantly elevated expression of Atg16L1, Atg16L2, and GabarapL1 was also observed in H4Swe cell cultures, in the presence of insulin resistance. Gene expression analysis confirmed that Atg16L1 was significantly increased in cultures from transgenic mice when insulin resistance was induced. Taken together, these results emphasise the association of the autophagy pathway in AD-T2DM co-morbidity, providing new evidence about the pathophysiology of both diseases and their mutual interaction.
## 1. Introduction
Alzheimer’s disease (AD) is the most common cause for dementia, and 55 million people are estimated to live with this condition worldwide. Numbers are expected to increase to 113 million by 2050 [1], causing enormous impacts on global health and imposing a huge economic burden. Therapeutic approaches encompass cholinesterase inhibitors and memantine as symptomatic agents [2,3]. Great hopes have been raised by antibodies which target amyloid-beta aggregates in the brain as potential disease-modifying interventions [4,5]. However, whether meaningful clinical efficacy can be reached as well as cost-effectiveness are still questions, while safety concerns need further analyses and clarification [6,7,8]. Therefore, preventive actions directed at potentially modifiable risk factors are crucial to reduce AD severe disease burden [2].
*Both* genetic and environmental factors contribute to AD risk. Dominantly inherited mutations in APP, PSEN1, and PSEN2 genes account for rarer early-onset cases, whereas carrying at least one copy of the APOE ε4 allele is the strongest genetic risk factor for the common late-onset form [9]. Although age is the most relevant factor providing the largest impact, additional environmental components present important contributions that are potentially modifiable [2]. Among the latter, consistent evidence supports major roles for education, hypertension, obesity, hearing loss, traumatic brain injury, smoking, depression, physical inactivity, social isolation, type 2 diabetes mellitus (T2DM), and air pollution as potential targets of intervention in different life stages, particularly at midlife [2]. In addition, T2DM has been compellingly associated with significantly greater risk of dementia [10,11,12]. Moreover, metabolic syndrome and obesity, which are often associated with T2DM, represent dementia risk factors per se, thus further complicating the picture [13,14]. It has been suggested that since T2DM is modifiable, its reduction could constitute a possible strategy for reducing future AD incidence. Indeed, it has been estimated that if T2DM was removed as a risk factor, about $1.1\%$ of dementia cases could be prevented. Although this percentage is low, the number of impacted people is nonetheless high when considering global incidence rates [12].
Despite the demonstrated convincing association between AD and T2D, the pathophysiological mechanisms responsible are still unknown. As a result, the best approach to be adopted for prevention still needs to be elucidated [12]. Furthermore, whether antidiabetic treatments represent a useful way forward is uncertain at present, as available data are inconsistent [2,15,16]. Several hypotheses have been proposed to explain the mechanistic link between AD and T2DM [17,18,19]. Among them, insulin signalling is impaired in both AD and T2DM, and the definition of AD as type 3 diabetes is based on the observed insulin resistance [20,21,22,23]. In addition, defects in mitochondrial function are shared by both AD and T2DM, thus a common causative role has been proposed for this defect based on preclinical and clinical findings [24,25]. In a previous study, we adopted a system biology approach to address this important gap in knowledge about the common pathophysiological dysregulations contributing to AD and T2DM comorbidity. We compared molecular mechanistic networks underlying brain T2DM pathophysiology in AD and control subjects by analysing transcriptional datasets with a novel approach. We discovered a central role for the autophagy pathway in the mechanisms shared between AD and T2DM [26]. Autophagy is an intracellular degradation pathway that traffics organelles, dysfunctional proteins, and infectious agents to lysosomes via specific vesicles called autophagosomes [27]. In agreement with our findings, autophagy relevance in AD is supported by a wealth of data, and targeting this mechanism is proposed as a potential avenue for drug discovery [28,29,30]. Moreover, abnormal autophagic responses have been implicated in metabolic disorders [31].
The aim of this study was to further investigate the role of genes identified in our previous studies as relevant for the pathophysiology of Alzheimer’s disease and T2DM comorbidity, namely ATG16L1, ATG16L2, GABARAP, GABARAPL1, GABARAPL2, and SQSTM1. We thus investigated the modulation of these genes in an animal model of AD and in cellular models of insulin resistance in Alzheimer’s disease brains.
## 2.1. Antibodies
In immunofluorescence experiments, the following antibodies and dilutions were used: anti-Phospho SQSTM1/p62 (S349) (Abcam, Cambridge, UK, cat # ab211324) 1:100; anti-SQSTM1/p62 (Abcam, cat # ab56416) 1:50; anti-β-Tubulin III (Merck Millipore, Burlington, MA, USA cat # T2200) 1:500; anti-MAP2 (1:500, Merck Millipore, cat # M9942) 1:500; anti-GFAP (Thermo Fisher Scientific, Waltham, MA, USA, cat # 13-0300), 1:800, donkey anti-rabbit-IgG Alexa Fluor 488 (Thermo Fisher Scientific, cat # R37118) 1:1000; donkey anti-mouse-IgG Alexa Fluor 594 (Thermo Fisher Scientific, cat # A-21203, 1:1000); goat anti-mouse IgG1 CF 568 (Merck, cat # SAB4600313 1:1000); goat anti-rat Alexa Fluor 647 (Thermo Fisher Scientific, cat # A21247, 1:1000); and DAPI (4′,6-diamidino-2-phenylindole Merck Millipore, cat #D9542) 1:5000. In Western blotting experiments, the following antibodies and dilutions were used: anti-Phospho SQSTM1/p62 (S349) (Abcam cat # ab211324) 1:2000; anti-SQSTM1/p62 (Abcam cat # ab56416) 1:2000; Anti-GAPDH (Abcam cat # ab8245); 1:5000; anti-phospho-Akt (Ser473 D9E Cell Signaling, Danvers, MA, USA, cat #4060) 1:2000; anti-Akt (Cell Signaling, cat # 9272, 1:1000); goat anti-mouse IgG IRDye 800(Li-Cor, Lincoln, NE, USA, cat # 926-32210) 1:5000; and goat anti-rabbit-IgG Alexa 680 (Thermo Fisher Scientific cat # A21076) 1:5000.
## 2.2. Animals
A colony of triple-transgenic AD mice (3xTg-AD) expressing three mutant human transgenes—PS1M146V, APPSwe, and tauP301L—was established at the University of Verona by purchasing transgenic mice from The Jackson Laboratory (Sacramento, CA, USA). C57BL/6J mice were purchased from Charles River Italia (Calco, Italia). Mice were housed at 3/cage at a constant room temperature of 21 ± 1 °C and maintained on a 12:12h light/dark cycle with lights on at 7.30 a.m. with freely available food and water. All efforts to minimise animal suffering and number were made. This study is compliant with ARRIVE guidelines [32]. Procedures involving animals were conducted in conformity with the EU guidelines ($\frac{2010}{63}$/UE) and Italian law (decree $\frac{26}{14}$) and were approved by the University of Verona’s ethical committee and the local authority’s veterinary service. The Italian Health Ministry Ethical Committee for Protection of Animals approved the study (approval number: $\frac{283}{2019}$-PR). *For* gene expression studies, 18 (six/group aged 6, 12, and 18 months) female transgenic mice and 18 (six/group) female wild-type mice were used. For immunofluorescence on brain sections, 12-month-old female 3xTg-AD and wild-type mice were used ($$n = 3$$/group). Mice for gene or protein expression experiments were anesthetised using Tribromoethanol (Merck Millipore) and sacrificed. Brain dissections were performed in Petri dishes on ice; the hippocampi were collected, flash-frozen in liquid nitrogen, and stored at −80 °C until analysis. The whole procedure did not exceed 5 min to preserve brain integrity. Mice for immunofluorescence experiments were anesthetised using Tribromoethanol, perfused transcardially with 0.1 M phosphate buffered saline solution (PBS), followed by formaldehyde $10\%$ V/V, buffered $4\%$ w/v (Titolchimica, Rovigo, Italy), and brains were extracted and postfixed overnight. Seven dams were used (wild-type: $$n = 4$$; 3xTg-AD $$n = 3$$) and neuronal cultures were prepared from 5–6 pups/preparation for each genotype.
## 2.3. Neuronal Cultures
Human glioblastoma H4 cell lines stably expressing the βAPP-Swe mutation (K595N/M596L) were a kind gift from prof. M. Pizzi, University of Brescia, Italy. Cells were cultured in DMEM with $10\%$ foetal bovine serum (FBS), 100 Units/mL penicillin, 2 mM glutamine, and 100 μg/mL streptomycin (Thermo Fisher Scientific) [33]. After reaching $80\%$ confluence and twenty-four hours before starting the experiment, cells were trypsinised and seeded at a density of 4 × 106 cells in T75 cm2 flasks. Treatments were performed in DMEM medium without FBS.
## 2.4. Primary Cortical Neurons
Primary mouse cortical cultures were prepared as previously described [34] with modifications [35]. Briefly, newborn C57BL/6 and 3xTg-AD mice (P0-P1) brains were isolated and cortices were dissected in 1X ice-cold DBPS medium (cat # 14200075, Thermo Fisher Scientific). After removal of meninges, cortices were washed twice and enzymatically digested with DPBS solution containing $0.25\%$ (v/v) trypsin (Thermo Fisher Scientific), 1 mM sodium pyruvate, $0.1\%$ (w/v) glucose, and 10 mM HEPES pH 7.3 for 20 min at 37 °C. Following a 5 min incubation with 0.1mg/mL DNAse I (Merck Millipore) at room temperature, the enzymatic reaction was stopped with an MEM solution containing $10\%$ FBS, $0.45\%$ (w/v) glucose, 1 mM sodium pyruvate, 2 mM L-glutamine, 100 U/mL penicillin, and 100 μg/mL streptomycin (all reagents from Thermo Fisher Scientific). Next, the tissue was triturated through a P1000 pipette, and the cell suspension was passed through a 70 µm MACS SmartStrainer (Miltenyi Biotec, Bergisch Gladbach, Germany). Cells were then counted and diluted to 8 × 105 cells/mL in Neurobasal™-A Medium (NBA, Thermo Fisher Scientific) containing 1X B27 supplement (Thermo Fisher Scientific), 2 mM L-Glutamine (Thermo Fisher Scientific), 100 U/mL penicillin, and 100 μg/mL streptomycin (Thermo Fisher Scientific) and plated on 6-well plates pre-coated with 0.1 mg/mL poly-L lysine (Merck Millipore). Cells were maintained in a standard, humidified $5\%$ CO2 incubator until the day of the experiment (5–7 days in vitro, DIV).
## 2.5. Insulin Resistance
To monitor insulin response, cells were challenged with 100 nM insulin (Merck Millipore) for 30 min. To induce insulin resistance, cells were treated for 24 h with 40 mM glucose (Merck) or 20 nM insulin before receiving insulin challenge [36]. Controls were treated with vehicle. At the end of the experiments, both H4Swe cells and primary mouse neurons were washed with PBS and harvested by 5 min centrifugation at 2900× g, and the pellets were re-suspended in RNA later (Thermo Fisher Scientific), stored at 4 °C for 24 h, and transferred at −20 °C until RNA extraction. Treatments were repeated in 3–6 independent experiments.
## 2.6. Quantitative Real-Time PCR
Gene expression was assessed by qPCR as previously reported [37] with slight changes. RNA was extracted with the Aurum total RNA mini kit (Bio-Rad, Hercules, CA, USA) which includes a DNase I digestion step, following manufacturer’s instructions. RNA amount was assessed by UV absorbance in a NanoDrop 2000c spectrophotometer (Thermo Fisher Scientific). cDNA was synthesised using the iScript Advanced cDNA synthesis Kit (Bio-Rad). qPCR was performed by Sybr Green technology in a 7900HT Fast Real-Time PCR System (Thermo Fisher Scientific) with SSO Advanced Universal SYBR Green Supermix (Bio-Rad) in 20 µL according to this protocol: stage 1: 95 °C, 20 s; stage 2: 40 × (95 °C, 3 s; 60 °C, 30 s). Primers were selected with the NCBI Primer-BLAST tool and purchased from Eurofins Italia (Torino, Italy). Sequences are reported in Table 1. Data were analysed using the Delta-Delta-Ct method, converting to a relative ratio (2−DDCt) for statistical analysis [38] by normalising to the geometric average of two endogenous reference genes: Gapdh and Ywhaz, as previously reported [39,40]. The specificity of amplification products was evaluated by building a dissociation curve in the 60–95 °C range.
## 2.7. Western Blot
Hippocampi were homogenised with a micro-pestle in ice-cold lysis buffer ($10\%$ w/v) containing 50 mM Tris-HCl (pH 7.5), $2\%$ Igepal, 10 mM MgCl2, 0.5 M NaCl, 2 mM EDTA, 2 mM EGTA, 5 mM benzamidine, 0.5 mM phenyl-methylsulfonyl fluoride, 8 mg/mL pepstatin A, 20 mg/mL leupeptin, 50 mM β-glycerolphosphate, 100 mM sodium fluoride, 1 mM sodium vanadate, 20 mM sodium pyrophosphate, and 100 nM okadaic acid. Homogenates were clarified by 1 min centrifugation at 10,000× g at 4 °C and protein concentration was assessed by Precision Red Protein Quantification Assay (Cytoskeleton). H4Swe cells were seeded into 6-well plates at a density of 9.5 × 105 cells/well. Following treatments, cells were washed once in Tris-buffered saline (TBS), lysed, and assayed for protein with the Bradford method (Merck). In both instances, lysates were processed for Western blot as previously reported [41], with slight changes. Briefly, lysates were separated using 4–$12\%$ Bis-Tris gels (Novex pre-cast gel, Thermo Fisher Scientific) and transferred to 0.45 μm nitrocellulose membranes (Thermo Fisher Scientific). Blots were blocked for 1 h at room temperature in 1X Odyssey blocking buffer (TBS) and incubated with primary antibodies overnight in Odyssey blocking buffer (TBS) plus $0.1\%$ Tween-20 (Tween-20 TBS) at 4 °C. Membranes were washed 3 × 10 min in Tween-20 TBS at room temperature, followed by incubation with secondary antibody conjugated to IRDye diluted in Tween-20 TBS for 1 h at room temperature. Blots were washed 2 × 10 min in TBST, 1 × 10 min in TBS, and visualised with Odyssey Infrared Imaging System (Li-Cor) by quantifying fluorescent signals as Integrated Intensities (I.I. K Counts) using the Odyssey Infrared Imaging System. After background subtraction, protein levels were assessed as total protein to Gapdh loading control ratios or as phosphorylated to total protein ratios.
## 2.8. Immunofluorescence
In brain sections, immunofluorescence was carried out as previously reported [42]. Briefly, after post-fixing, brains were embedded in an OCT cryoembedding matrix and sectioned on the coronal plane at 30 mm thickness with a cryostat. Sections were treated with a blocking solution of $2\%$ bovine serum albumin, $2\%$ normal goat serum, and $0.2\%$ Triton X100 in PBS for 20 min at room temperature and incubated overnight at 4 °C in primary antibodies. Secondary antibodies were diluted 1:1000 in the above blocking solution, with the appropriate serum. After immunohistochemical processing, sections were counterstained with the fluorescent nuclear marker DAPI (100 ng/mL) for 10 min at room temperature and mounted on slides with $0.1\%$ paraphenylenediamine in glycerol-based medium ($90\%$ glycerol $10\%$ PBS). For H4Swe cell immunostaining, 5 × 105 cells/well were seeded onto 18 mm round coverslips in 24-well plates and left to attach overnight. The next day, cells were washed twice with PBS and fixed with $4\%$ paraformaldehyde for 20 min. Fixed cells were treated for 10 min with blocking and incubated overnight with primary antibodies in blocking solution. After three washes with PBS, samples were incubated with secondary antibodies diluted 1:2000 in blocking solution for 1 h. After final washes, coverslips were treated with DAPI solution. Coverslips were fixed onto glass slides with a drop of anti-fading mounting medium and sealed with nail polish. Primary cortical cells were fixed in $10\%$ (v/v) formalin solution (Titolchimica) for 15 min at room temperature, washed three times in PBS, and blocked in PBS containing $10\%$ (v:v) normal goat serum (Thermo Fisher Scientific) and permeabilised with $0.3\%$ (v:v) TritonX-100 (Merck Millipore) in PBS for 40 min. Next, cells were incubated with mouse anti-Map2 and rat anti-Gfap primary antibodies overnight at 4 °C, and after three PBS washing steps, with anti-secondary antibodies, anti-mouse IgG1 CF 568, and anti-rat Alexa Fluor 647 for 1 h at room temperature. Antibodies were diluted in PBS containing $5\%$ (v:v) normal goat serum. Nuclei were counterstained with DAPI 1:5000 and coverslips were mounted on slides using DAKO fluorescence mounting media (Agilent, Santa Clara, CA, USA). Images at different Z-planes were collected on a Leica tcs-sp5 confocal microscope. Images were processed with the software Imaris (Bitplane AG, Belfast, UK) or ImageJ.
## 2.9. Statistical Analysis
The data are presented as the observed mean values ± SEM. The data were analysed using a 1-way ANOVA with treatment (control, insulin, glucose + insulin, insulin + insulin) as the treatment factor or 2-way ANOVA with genotype (wild-type, 3xTg-AD) and age (6, 12, and 18 months) or treatment (control, insulin, glucose + insulin, insulin + insulin). When the samples were analysed in different plates using a complete block design, an additional blocking factor plate was also included in the statistical model in order to account for any plate-to-plate variability [43]. The analyses were followed by planned comparisons of the predicted means. The analysis was performed using the InVivoStat v4.4.0 software [44]. The data were log-transformed, where appropriate, in order to stabilise the variance and satisfy the parametric assumptions. A value of $p \leq 0.05$ was considered statistically significant.
## 3. Results
Since comorbidity is frequently observed between AD and T2DM, in our previous study [26], we applied a systems biology approach to investigate if common pathophysiological alterations could be identified at a molecular level. Similar approaches had previously highlighted the role of shared cellular signalling pathways contributing to both T2DM and AD. Among them, a prominent role was discovered for neurotrophin, PI3K/AKT, MTOR, and MAPK signalling, as well as for microglial-mediated immune responses, which can cross-talk to each other [45]. In addition, our previous data revealed a central role for autophagic mechanisms; in particular, a number of autophagy-related genes were indicated as important players, namely ATG16L1, ATG16L2, GABARAP, GABARAPL1, GABARAPL2, and SQSTM1. Therefore, we first aimed to investigate whether these genes were specifically modulated in association with neurobiological alterations characterising AD. We thus analysed the expression of the respective mouse orthologues (Atg16l1, Atg16l2, Gabarap, GabarapL1, GabarapL2, Sqstm1) in a transgenic mouse model of AD. 3xTg-AD mice harbour three mutant genes for the beta-amyloid precursor protein (βAPPSwe), presenilin-1 (PS1M146V), and tauP301L [46,47]; as a consequence, the mice progressively develop plaques and tangles, as well as cognitive impairments [47,48,49]. We thus compared hippocampal gene expression between 3xTg-AD mice and the respective wild-type controls at different ages. At 6 months, Atg16L1, Atg16L2, and GabarapL1 were expressed at significantly higher levels in 3xTg-AD mice (Figure 1A,B,D). In contrast, at 12 months, GabarapL2 expression was significantly reduced, whereas Sqstm1 levels were elevated (Figure 1E,F).
At the protein level, although increased Sqstm1 mRNA expression was observed qualitatively in the hippocampus (Figure 2A), the increase could not be confirmed in semi-quantitative Western blotting experiments, possibly because of the lower sensitivity of the technique (Figure 2B,C).
Next, we investigated whether these genes were modulated in the presence of AD-T2DM comorbidity. To model this condition, we first employed the human glioblastoma H4 cell line stably expressing the βAPP-Swe mutation [50,51,52] and applied treatments able to induce insulin resistance [53,54]. In this model, phospho-Akt/Akt levels were significantly increased by the insulin challenge (100 nM), whereas this response was abated after chronic treatment with high-concentration insulin, thus showing that insulin resistance was successfully achieved (Figure 3).
Similar to findings obtained in 3xTg-AD mice, in the presence of insulin resistance, Atg16L1, Atg16L2, and GabarapL1 expression levels were significantly increased (Figure 4A,B,D).
Subsequently, we examined whether Sqstm1 phosphorylation levels were affected by the onset of insulin resistance. No significant differences were revealed by Western blot or immunofluorescence analyses (Figure 5).
Next, we generated a second AD-T2DM cellular model by inducing insulin resistance in neuronal primary cultures obtained from 3xTg-AD mice and wild-type controls. In this model, we confirmed that primary cultures were enriched in neurons (Figure 6).
Gene expression analysis confirmed that Atg16L1 was significantly increased in cultures from transgenic mice when insulin resistance was induced (Table 2), whereas no other difference was detected in the other genes analysed. In addition, Gabarap showed a significant reduction by genotype (Table 2). However, the findings showed a very high level of variability within groups.
## 4. Discussion
In this study, we examined the modulation of genes recognised as relevant for the common cellular dysregulations sustaining the observed comorbidity between AD and T2 DM in our previous systems biology study [26]. Here, we explored their expression in 3xTg-AD mice, a transgenic mouse model of AD overexpressing mutated human genes associated with early-onset AD (PSEN1 and APP) or with the formation of neurofibrillary tangles (tau) [46]. In this mouse model, the neuropathological features of AD, amyloid plaques and neurofibrillary tangles, as well as neuroinflammation, developed progressively in an age-dependent fashion. In particular, extracellular amyloid beta deposition started at six months of age and progressively increased to reach its full extent at 15 months [47,49]. Tau pathology followed a similar age-related increase, although delayed with respect to amyloid beta pathology [46,47,49]. Likewise, cognitive impairments reproducing the human pathological feature of AD appeared at six months and became progressively more severe at 12 and 20 months [49]. We discovered that at 6 months of age, Atg16L1, Atg16L2, and GabarapL1 were expressed at higher levels in 3xTg-AD mice, whereas at later time points, this increase subsided. The alterations are in agreement with those obtained in the previous study, where pre-frontal cortex samples were analysed in two AD mice models [26]. These findings suggest that the increased expression may occur as an attempt to oppose the neuropathological alterations by activating a neuroprotective response.
A limitation of this experimental design is that 3xTg-AD mice were generated in a hybrid C57BL/6:129 genetic background; therefore, the control line we used, although similar, is not identical. However, the use of C57BL/6 as a control strain is well documented in previous studies [55,56,57,58].
To reproduce the molecular dysregulation characterising insulin resistance in AD brains, we used neuronal models of AD based either on a neuronal cell line generating amyloid beta deposits, H4Swe cells, or on the 3xTg-AD mouse primary neuron cultures. H4Swe cells are well established as tools to investigate AD-related cellular dysregulation [50,51,52]. However, a limitation is that they do not share all neuronal characteristics, being a neuroglioma-derived line. Therefore, primary neurons were also investigated. In both in vitro models, we established a condition of insulin resistance by prolonged treatment with high insulin concentrations. As a consequence, the normal response to insulin challenge is hampered by prolonged insulin exposure, and the normal Akt phosphorylation and activation responses characterising the insulin signal transduction pathway are not induced [53]. Similar to findings in 3xTg-AD mice, we found that Atg16L1, Atg16L2, and GabarapL1 were significantly elevated in insulin resistance conditions. The increased expression of these genes in the cell model of AD-T2DM comorbidity corroborates the hypothesis of a neuroprotective role of this response, as hyperglycaemia has been previously associated with the increased beta amyloid plaque production [59].
Atg16L1 was identified as the mammalian orthologue of the corresponding yeast gene, which was known to provide a crucial contribution to autophagic processes [60,61]. Autophagy was discovered as a process occurring in response to cellular stresses such as nutrient deprivation, infection, or hypoxia. Its chief function is providing nutrients for vital cellular activities during fasting by degrading cellular components and releasing them back to the cytoplasm to be used again. However, in addition to this non-selective approach, further studies demonstrated that autophagy can selectively eliminate potentially harmful damaged mitochondria or protein aggregates [61,62]. Consequently, autophagy dysfunction has been implicated in several diseases and its components generated interest as potential pharmacological targets [28,62]. In autophagy, starvation signals promote the recruitment of autophagy proteins to a specific subcellular location, where they assemble a structure called the phagophore. An isolation membrane is gradually formed to isolate a portion of the cytosol and is finally sealed into a vesicle, termed the autophagosome, which contains cytoplasmic material. The autophagosome then fuses with the lysosomal membrane, and the autophagic body together with its cargo are degraded [62,63]. In this process, the role of Atg16L1 is essential for autophagy initiation, as its recruitment in the Atg12-Atg5 complex is required to engage autophagic proteins in the phagophore assembly site and contribute to its scaffolding by Atg8/LC3 protein lipidation [60,64,65,66]. Therefore, the increase observed in the present study suggests an effort to trigger autophagic responses to counteract the increased production of abnormal proteins and rescue insulin response.
In addition to its well-demonstrated role in canonical autophagy, Atg16L1 was shown to exert different functions related to a structural component specifically observed in the C-terminal of the mammalian protein compared to the yeast counterpart. This specific component is necessary for the Atg16L1-mediated lipidation of single membranes, a non-canonical autophagy pathway, and specific cargo recruitment [66]. Furthermore, Atg16L1 contributes to modulating the extent of the innate immune response to injuries or infection, with an anti-inflammatory role [66,67]. Recent results showed that aged mice lacking this C-terminal domain of Atg16L1 develop beta amyloid plaques, excessive tau phosphorylation, reactive microgliosis, and memory impairments [68]. The proposed mechanism points to Atg16L1 involvement in a process defined as LANDO (LC3-associated endocytosis), which contributes to TREM2, CD36, and TLR4 recycling [68]. Therefore, the observed increased Atg16L1 levels may contribute to establishing a protective response that goes beyond the activation of autophagic responses, but also involves a rescue from neuronal damages through different mechanisms. Interestingly, we observed increased Atg16L1 expression in all investigated models. This result reinforces the notion of a primary role of this protein in the cellular response to both AD and T2DM pathophysiology, in a fashion independent from the in vivo or in vitro model which is well-conserved through evolution both in mice and in humans.
Atg16L2 is a second isoform of Atg16L1, sharing a similar domain structure and a similar ability to bind Atg12-Atg5 and form a complex. However, the Atg16L2 protein is not recruited to phagophores and does not contribute to autophagosome formation; thus, it is not essential to canonical autophagy [69]. However, data suggesting the possibility of a cell-specific involvement in canonical autophagy are also available [70]. In addition, a recent report on the generation of Atg16L2 knock-out mice demonstrated a contribution of this gene to the maturation of immune cells and suggested that distinct functions are associated with respect to Atg16L1 [71]. Data showing its relevance in serious diseases such as Crohn’s disease and various cancers notwithstanding, very incomplete information is available on the role of Atg16L2 [72]. Our findings also support the involvement of this widely expressed gene in the pathophysiology of insulin resistance in AD brains.
The GabarapL1 protein belongs to the Atg8/LC3 autophagy proteins, which include six members: LC3A, LC3B, LC3C, Gabarap, GabarapL1, and GabarapL2. The recruitment of Atg8 family proteins to the forming phagophore is mediated by the above-mentioned Atg12-Atg5–Atg16L1 complex and is essential for phagophore elongation and, ultimately, for autophagy [62,63,73]. GabarapL1 has also been implicated in autophagosome fusion with lysosome, and these functions are supposed to contribute to the degradation of oncogenic proteins and exert tumour-suppressive functions [73]. Interestingly, GabarapL1 has been specifically implicated in a newly discovered selective autophagy process termed glycophagy, which is involved in the transport and delivery of glycolytic fuel substrates [74]. Since these pathways regulate cellular energy demand, compelling evidence links glycophagy-mediated glucose availability to energy metabolism, in agreement with our findings.
With regard to Sqstm1 levels, contrasting findings have been previously reported. In agreement with the present results, no alterations were detected in the hippocampus or in mitochondria-enriched hippocampal fractions of young 3xTg-AD mice [75,76]. Conversely, a decrease was found in whole brain homogenates and in the mitochondria-enriched hippocampal fractions of old 3xTg-AD mice [76,77,78].
## 5. Conclusions
This study investigated the molecular underpinning of the comorbidity between AD and T2DM in cellular models of insulin resistance in the presence of AD-related neuropathological features. Our findings are in agreement with the hypothesis that impaired autophagic mechanisms are important in the pathophysiology of AD through nonstandard mechanisms. In particular, the autophagy-related genes Atg16L1, Atg16L2, and GabarapL1 were highlighted as having a more relevant function in this mechanism, in addition to GabarapL2 and Sqstm1.
## References
1. Scheltens P., De Strooper B., Kivipelto M., Holstege H., Chételat G., Teunissen C.E., Cummings J., van der Flier W.M.. **Alzheimer’s disease**. *Lancet* (2021) **397** 1577-1590. DOI: 10.1016/S0140-6736(20)32205-4
2. Livingston G., Huntley J., Sommerlad A., Ames D., Ballard C., Banerjee S., Brayne C., Burns A., Cohen-Mansfield J., Cooper C.. **Dementia prevention, intervention, and care: 2020 report of the Lancet Commission**. *Lancet* (2020) **396** 413-446. DOI: 10.1016/S0140-6736(20)30367-6
3. 3.Dementia: Assessment, Management and Support for People Living with Dementia and Their CarersNational Institute for Health and Care Excellence (NICE)London, UK2018. *Dementia: Assessment, Management and Support for People Living with Dementia and Their Carers* (2018)
4. van Dyck C.H., Swanson C.J., Aisen P., Bateman R.J., Chen C., Gee M., Kanekiyo M., Li D., Reyderman L., Cohen S.. **Lecanemab in Early Alzheimer’s Disease**. *N. Engl. J. Med.* (2022) **388** 9-21. DOI: 10.1056/NEJMoa2212948
5. Dunn B., Stein P., Cavazzoni P.. **Approval of Aducanumab for Alzheimer Disease-The FDA’s Perspective**. *JAMA Intern. Med.* (2021) **181** 1276-1278. DOI: 10.1001/jamainternmed.2021.4607
6. Alexander G.C., Knopman D.S., Emerson S.S., Ovbiagele B., Kryscio R.J., Perlmutter J.S., Kesselheim A.S.. **Revisiting FDA Approval of Aducanumab**. *N. Engl. J. Med.* (2021) **385** 769-771. DOI: 10.1056/NEJMp2110468
7. **The Lancet Lecanemab for Alzheimer’s disease: Tempering hype and hope**. *Lancet* (2022) **400** 1899. DOI: 10.1016/S0140-6736(22)02480-1
8. Ross E.L., Weinberg M.S., Arnold S.E.. **Cost-effectiveness of Aducanumab and Donanemab for Early Alzheimer Disease in the US**. *JAMA Neurol.* (2022) **79** 478-487. DOI: 10.1001/jamaneurol.2022.0315
9. Knopman D.S., Amieva H., Petersen R.C., Chételat G., Holtzman D.M., Hyman B.T., Nixon R.A., Jones D.T.. **Alzheimer disease**. *Nat. Rev. Dis. Prim.* (2021) **7** 33. DOI: 10.1038/s41572-021-00269-y
10. Chatterjee S., Peters S.A.E., Woodward M., Arango S.M., Batty G.D., Beckett N., Beiser A., Borenstein A.R., Crane P.K., Haan M.. **Type 2diabetes as a risk factor for dementia in women compared with men: A pooled analysis of 2.3 million people comprising more than 100,000 cases of dementia**. *Diabetes Care* (2016) **39** 300-307. DOI: 10.2337/dc15-1588
11. Gudala K., Bansal D., Schifano F., Bhansali A.. **Diabetes mellitus and risk of dementia: A meta-analysis of prospective observational studies**. *J. Diabetes Investig.* (2013) **4** 640-650. DOI: 10.1111/jdi.12087
12. Savelieff M.G., Chen K.S., Elzinga S.E., Feldman E.L.. **Diabetes and dementia: Clinical perspective, innovation, knowledge gaps**. *J. Diabetes Complicat.* (2022) **36** 108333. DOI: 10.1016/j.jdiacomp.2022.108333
13. Albanese E., Launer L.J., Egger M., Prince M.J., Giannakopoulos P., Wolters F.J., Egan K.. **Body mass index in midlife and dementia: Systematic review and meta-regression analysis of 589,649 men and women followed in longitudinal studies**. *Alzheimer’s Dement. Diagn. Assess. Dis. Monit.* (2017) **8** 165-178. DOI: 10.1016/j.dadm.2017.05.007
14. Campos-Peña V., Toral-Rios D., Becerril-Pérez F., Sánchez-Torres C., Delgado-Namorado Y., Torres-Ossorio E., Franco-Bocanegra D., Carvajal K.. **Metabolic Syndrome as a Risk Factor for Alzheimer’s Disease: Is Aβ a Crucial Factor in Both Pathologies?**. *Antioxid. Redox Signal.* (2017) **26** 542-560. DOI: 10.1089/ars.2016.6768
15. Nowell J., Blunt E., Edison P.. **Incretin and insulin signaling as novel therapeutic targets for Alzheimer’s and Parkinson’s disease**. *Mol. Psychiatry* (2022) **28** 217-229. DOI: 10.1038/s41380-022-01792-4
16. Areosa Sastre A., Vernooij R.W., González-Colaço Harmand M., Martínez G.. **Effect of the treatment of Type 2 diabetes mellitus on the development of cognitive impairment and dementia**. *Cochrane Database Syst. Rev.* (2017) **6** CD003804. DOI: 10.1002/14651858.CD003804.pub2
17. Patel V.N., Chorawala M.R., Shah M.B., Shah K.C., Dave B.P., Shah M.P., Patel T.M.. **Emerging Pathophysiological Mechanisms Linking Diabetes Mellitus and Alzheimer’s Disease: An Old Wine in a New Bottle**. *J. Alzheimer’s Dis. Rep.* (2022) **6** 349-357. DOI: 10.3233/ADR-220021
18. Hamzé R., Delangre E., Tolu S., Moreau M., Janel N., Bailbé D., Movassat J.. **Type 2 Diabetes Mellitus and Alzheimer’s Disease: Shared Molecular Mechanisms and Potential Common Therapeutic Targets**. *Int. J. Mol. Sci.* (2022) **23**. DOI: 10.3390/ijms232315287
19. Pugazhenthi S., Qin L., Reddy P.H.. **Common neurodegenerative pathways in obesity, diabetes, and Alzheimer’s disease**. *Biochim. Biophys. Acta—Mol. Basis Dis.* (2017) **1863** 1037-1045. DOI: 10.1016/j.bbadis.2016.04.017
20. González A., Calfío C., Churruca M., Maccioni R.B.. **Glucose metabolism and AD: Evidence for a potential diabetes type 3**. *Alzheimer’s Res. Ther.* (2022) **14** 56. DOI: 10.1186/s13195-022-00996-8
21. Frisardi V., Solfrizzi V., Seripa D., Capurso C., Santamato A., Sancarlo D., Vendemiale G., Pilotto A., Panza F.. **Metabolic-cognitive syndrome: A cross-talk between metabolic syndrome and Alzheimer’s disease**. *Ageing Res. Rev.* (2010) **9** 399-417. DOI: 10.1016/j.arr.2010.04.007
22. Steen E., Terry B.M., Rivera E.J., Cannon J.L., Neely T.R., Tavares R., Xu X.J., Wands J.R., de la Monte S.M.. **Impaired insulin and insulin-like growth factor expression and signaling mechanisms in Alzheimer’s disease--is this type 3 diabetes?**. *J. Alzheimers. Dis.* (2005) **7** 63-80. DOI: 10.3233/JAD-2005-7107
23. Craft S.. **Insulin resistance and Alzheimer’s disease pathogenesis: Potential mechanisms and implications for treatment**. *Curr. Alzheimer Res.* (2007) **4** 147-152. DOI: 10.2174/156720507780362137
24. Moreira P.I.. **Sweet Mitochondria: A Shortcut to Alzheimer’s Disease**. *J. Alzheimer’s Dis.* (2018) **62** 1391-1401. DOI: 10.3233/JAD-170931
25. Correia S.C., Santos R.X., Carvalho C., Cardoso S., Candeias E., Santos M.S., Oliveira C.R., Moreira P.I.. **Insulin signaling, glucose metabolism and mitochondria: Major players in Alzheimer’s disease and diabetes interrelation**. *Brain Res.* (2012) **1441** 64-78. DOI: 10.1016/j.brainres.2011.12.063
26. Caberlotto L., Nguyen T.-P., Lauria M., Priami C., Rimondini R., Maioli S., Cedazo-Minguez A., Sita G., Morroni F., Corsi M.. **Cross-disease analysis of Alzheimer’s disease and type-2 Diabetes highlights the role of autophagy in the pathophysiology of two highly comorbid diseases**. *Sci. Rep.* (2019) **9** 3965. DOI: 10.1038/s41598-019-39828-5
27. Bento C.F., Renna M., Ghislat G., Puri C., Ashkenazi A., Vicinanza M., Menzies F.M., Rubinsztein D.C.. **Mammalian Autophagy: How Does It Work?**. *Annu. Rev. Biochem.* (2016) **85** 685-713. DOI: 10.1146/annurev-biochem-060815-014556
28. Boland B., Yu W.H., Corti O., Mollereau B., Henriques A., Bezard E., Pastores G.M., Rubinsztein D.C., Nixon R.A., Duchen M.R.. **Promoting the clearance of neurotoxic proteins in neurodegenerative disorders of ageing**. *Nat. Rev. Drug Discov.* (2018) **17** 660-688. DOI: 10.1038/nrd.2018.109
29. Menzies F.M., Fleming A., Caricasole A., Bento C.F., Andrews S.P., Ashkenazi A., Füllgrabe J., Jackson A., Jimenez Sanchez M., Karabiyik C.. **Autophagy and Neurodegeneration: Pathogenic Mechanisms and Therapeutic Opportunities**. *Neuron* (2017) **93** 1015-1034. DOI: 10.1016/j.neuron.2017.01.022
30. Menzies F.M., Fleming A., Rubinsztein D.C.. **Compromised autophagy and neurodegenerative diseases**. *Nat. Rev. Neurosci.* (2015) **16** 345-357. DOI: 10.1038/nrn3961
31. Kitada M., Koya D.. **Autophagy in metabolic disease and ageing**. *Nat. Rev. Endocrinol.* (2021) **17** 647-661. DOI: 10.1038/s41574-021-00551-9
32. Percie du Sert N., Ahluwalia A., Alam S., Avey M.T., Baker M., Browne W.J., Clark A., Cuthill I.C., Dirnagl U., Emerson M.. **Reporting animal research: Explanation and elaboration for the ARRIVE guidelines 2.0**. *PLoS Biol.* (2020) **18**. DOI: 10.1371/journal.pbio.3000411
33. Capaldi S., Suku E., Antolini M., Di Giacobbe M., Giorgetti A., Buffelli M.. **Allosteric sodium binding cavity in GPR3: A novel player in modulation of Aβ production**. *Sci. Rep.* (2018) **8** 11102. DOI: 10.1038/s41598-018-29475-7
34. Beaudoin G.M.J., Lee S.-H., Singh D., Yuan Y., Ng Y.-G., Reichardt L.F., Arikkath J.. **Culturing pyramidal neurons from the early postnatal mouse hippocampus and cortex**. *Nat. Protoc.* (2012) **7** 1741-1754. DOI: 10.1038/nprot.2012.099
35. Bolognin S., Zatta P., Lorenzetto E., Valenti M.T., Buffelli M.. **β-Amyloid-aluminum complex alters cytoskeletal stability and increases ROS production in cortical neurons**. *Neurochem. Int.* (2013) **62** 566-574. DOI: 10.1016/j.neuint.2013.02.008
36. Peng Y., Liu J.J., Shi L., Tang Y., Gao D., Long J., Liu J.J.. **Mitochondrial dysfunction precedes depression of AMPK/AKT signaling in insulin resistance induced by high glucose in primary cortical neurons**. *J. Neurochem.* (2016) **137** 701-713. DOI: 10.1111/jnc.13563
37. Carboni L., Pischedda F., Piccoli G., Lauria M., Musazzi L., Popoli M., Mathé A.A., Domenici E.. **Depression-Associated Gene Negr1-Fgfr2 Pathway Is Altered by Antidepressant Treatment**. *Cells* (2020) **9**. DOI: 10.3390/cells9081818
38. Livak K.J., Schmittgen T.D.. **Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method**. *Methods* (2001) **25** 402-408. DOI: 10.1006/meth.2001.1262
39. Vandesompele J., De Preter K., Pattyn F., Poppe B., Van Roy N., De Paepe A., Speleman F.. **Accurate normalization of real-time quantitative RT-PCR data by geometric averaging of multiple internal control genes**. *Genome Biol.* (2002) **3** RESEARCH0034. DOI: 10.1186/gb-2002-3-7-research0034
40. Carboni L., Ponzoni L., Braida D., Sala M., Gotti C., Zoli M.. **Altered mrna levels of stress-related peptides in mouse hippocampus and caudate-putamen in withdrawal after long-term intermittent exposure to tobacco smoke or electronic cigarette vapour**. *Int. J. Mol. Sci.* (2021) **22**. DOI: 10.3390/ijms22020599
41. Borin M., Saraceno C., Catania M., Lorenzetto E., Pontelli V., Paterlini A., Fostinelli S., Avesani A., Di Fede G., Zanusso G.. **Rac1 activation links tau hyperphosphorylation and Aβ dysmetabolism in Alzheimer’s disease**. *Acta Neuropathol. Commun.* (2018) **6** 61. DOI: 10.1186/s40478-018-0567-4
42. Laperchia C., Allegra Mascaro A.L., Sacconi L., Andrioli A., Mattè A., De Franceschi L., Grassi-Zucconi G., Bentivoglio M., Buffelli M., Pavone F.S.. **Two-photon microscopy imaging of thy1GFP-M transgenic mice: A novel animal model to investigate brain dendritic cell subsets in vivo**. *PLoS ONE* (2013) **8**. DOI: 10.1371/journal.pone.0056144
43. Bate S.T., Clark R.A.. *The Design and Statistical Analysis of Animal Experiments* (2014)
44. Clark R.A., Shoaib M., Hewitt K.N., Stanford S.C., Bate S.T.. **A comparison of InVivoStat with other statistical software packages for analysis of data generated from animal experiments**. *J. Psychopharmacol.* (2012) **26** 1136-1142. DOI: 10.1177/0269881111420313
45. Karki R., Kodamullil A.T., Hofmann-Apitius M.. **Comorbidity Analysis between Alzheimer’s Disease and Type 2 Diabetes Mellitus (T2DM) Based on Shared Pathways and the Role of T2DM Drugs**. *J. Alzheimers. Dis.* (2017) **60** 721-731. DOI: 10.3233/JAD-170440
46. Oddo S., Caccamo A., Shepherd J.D., Murphy M.P., Golde T.E., Kayed R., Metherate R., Mattson M.P., Akbari Y., LaFerla F.M.. **Triple-transgenic model of Alzheimer’s Disease with plaques and tangles: Intracellular Aβ and synaptic dysfunction**. *Neuron* (2003) **39** 409-421. DOI: 10.1016/S0896-6273(03)00434-3
47. Oddo S., Caccamo A., Kitazawa M., Tseng B.P., LaFerla F.M.. **Amyloid deposition precedes tangle formation in a triple transgenic model of Alzheimer’s disease**. *Neurobiol. Aging* (2003) **24** 1063-1070. DOI: 10.1016/j.neurobiolaging.2003.08.012
48. Billings L.M., Oddo S., Green K.N., McGaugh J.L., LaFerla F.M.. **Intraneuronal Aβ causes the onset of early Alzheimer’s disease-related cognitive deficits in transgenic mice**. *Neuron* (2005) **45** 675-688. DOI: 10.1016/j.neuron.2005.01.040
49. Belfiore R., Rodin A., Ferreira E., Velazquez R., Branca C., Caccamo A., Oddo S.. **Temporal and regional progression of Alzheimer’s disease-like pathology in 3xTg-AD mice**. *Aging Cell* (2019) **18** e12873. DOI: 10.1111/acel.12873
50. Shin J., Yu S.-B., Yu U.Y., Jo S.A., Ahn J.-H.. **Swedish mutation within amyloid precursor protein modulates global gene expression towards the pathogenesis of Alzheimer’s disease**. *BMB Rep.* (2010) **43** 704-709. DOI: 10.5483/BMBRep.2010.43.10.704
51. Branca C., Sarnico I., Ruotolo R., Lanzillotta A., Viscomi A.R., Benarese M., Porrini V., Lorenzini L., Calzà L., Imbimbo B.P.. **Pharmacological targeting of the β-amyloid precursor protein intracellular domain**. *Sci. Rep.* (2014) **4** 4618. DOI: 10.1038/srep04618
52. Haugabook S.J., Yager D.M., Eckman E.A., Golde T.E., Younkin S.G., Eckman C.B.. **High throughput screens for the identification of compounds that alter the accumulation of the Alzheimer’s amyloid beta peptide (Abeta)**. *J. Neurosci. Methods* (2001) **108** 171-179. DOI: 10.1016/S0165-0270(01)00388-0
53. Kim B., McLean L.L., Philip S.S., Feldman E.L.. **Hyperinsulinemia induces insulin resistance in dorsal root ganglion neurons**. *Endocrinology* (2011) **152** 3638-3647. DOI: 10.1210/en.2011-0029
54. Son S.M., Song H., Byun J., Park K.S., Jang H.C., Park Y.J., Mook-Jung I.. **Altered APP processing in insulin-resistant conditions is mediated by autophagosome accumulation via the inhibition of mammalian target of rapamycin pathway**. *Diabetes* (2012) **61** 3126-3138. DOI: 10.2337/db11-1735
55. Magri C., Vitali E., Cocco S., Giacopuzzi E., Rinaudo M., Martini P., Barbon A., Grassi C., Gennarelli M.. **Whole Blood Transcriptome Characterization of 3xTg-AD Mouse and Its Modulation by Transcranial Direct Current Stimulation (tDCS)**. *Int. J. Mol. Sci.* (2021) **22**. DOI: 10.3390/ijms22147629
56. Wu M., Zhou F., Cao X., Yang J., Bai Y., Yan X., Cao J., Qi J.. **Abnormal circadian locomotor rhythms and Per gene expression in six-month-old triple transgenic mice model of Alzheimer’s disease**. *Neurosci. Lett.* (2018) **676** 13-18. DOI: 10.1016/j.neulet.2018.04.008
57. Piedrahita D., Hernández I., López-Tobón A., Fedorov D., Obara B., Manjunath B.S., Boudreau R.L., Davidson B., Laferla F., Gallego-Gómez J.C.. **Silencing of CDK5 reduces neurofibrillary tangles in transgenic alzheimer’s mice**. *J. Neurosci.* (2010) **30** 13966-13976. DOI: 10.1523/JNEUROSCI.3637-10.2010
58. Melkonyan M.M., Hunanyan L., Lourhmati A., Layer N., Beer-Hammer S., Yenkoyan K., Schwab M., Danielyan L.. **Neuroprotective, Neurogenic, and Amyloid Beta Reducing Effect of a Novel Alpha 2-Adrenoblocker, Mesedin, on Astroglia and Neuronal Progenitors upon Hypoxia and Glutamate Exposure**. *Int. J. Mol. Sci.* (2017) **19**. DOI: 10.3390/ijms19010009
59. Macauley S.L., Stanley M., Caesar E.E., Yamada S.A., Raichle M.E., Perez R., Mahan T.E., Sutphen C.L., Holtzman D.M.. **Hyperglycemia modulates extracellular amyloid-β concentrations and neuronal activity in vivo**. *J. Clin. Investig.* (2015) **125** 2463-2467. DOI: 10.1172/JCI79742
60. Mizushima N., Kuma A., Kobayashi Y., Yamamoto A., Matsubae M., Takao T., Natsume T., Ohsumi Y., Yoshimori T.. **Mouse Apg16L, a novel WD-repeat protein, targets to the autophagic isolation membrane with the Apg12-Apg5 conjugate**. *J. Cell Sci.* (2003) **116** 1679-1688. DOI: 10.1242/jcs.00381
61. Ohsumi Y.. **Historical landmarks of autophagy research**. *Cell Res.* (2014) **24** 9-23. DOI: 10.1038/cr.2013.169
62. Dikic I., Elazar Z.. **Mechanism and medical implications of mammalian autophagy**. *Nat. Rev. Mol. Cell Biol.* (2018) **19** 349-364. DOI: 10.1038/s41580-018-0003-4
63. Mercer T.J., Gubas A., Tooze S.A.. **A molecular perspective of mammalian autophagosome biogenesis**. *J. Biol. Chem.* (2018) **293** 5386-5395. DOI: 10.1074/jbc.R117.810366
64. Kaufmann A., Beier V., Franquelim H.G., Wollert T.. **Molecular mechanism of autophagic membrane-scaffold assembly and disassembly**. *Cell* (2014) **156** 469-481. DOI: 10.1016/j.cell.2013.12.022
65. Kuma A., Mizushima N., Ishihara N., Ohsumi Y.. **Formation of the ∼350-kDa Apg12-Apg5·Apg16 multimeric complex, mediated by Apg16 oligomerization, is essential for autophagy in yeast**. *J. Biol. Chem.* (2002) **277** 18619-18625. DOI: 10.1074/jbc.M111889200
66. Gammoh N.. **The multifaceted functions of ATG16L1 in autophagy and related processes**. *J. Cell Sci.* (2020) **133** jcs249227. DOI: 10.1242/jcs.249227
67. Hamaoui D., Subtil A.. **ATG16L1 functions in cell homeostasis beyond autophagy**. *FEBS J.* (2022) **289** 1779-1800. DOI: 10.1111/febs.15833
68. Heckmann B.L., Teubner B.J.W., Boada-Romero E., Tummers B., Guy C., Fitzgerald P., Mayer U., Carding S., Zakharenko S.S., Wileman T.. **Noncanonical function of an autophagy protein prevents spontaneous Alzheimer’s disease**. *Sci. Adv.* (2020) **6** eabb9036. DOI: 10.1126/sciadv.abb9036
69. Ishibashi K., Fujita N., Kanno E., Omori H., Yoshimori T., Itoh T., Fukuda M.. **Atg16L2, a novel isoform of mammalian Atg16L that is not essential for canonical autophagy despite forming an Atg12-5-16L2 complex**. *Autophagy* (2011) **7** 1500-1513. DOI: 10.4161/auto.7.12.18025
70. Li N., Wu X., Holzer R.G., Lee J.H., Todoric J., Park E.J., Ogata H., Gukovskaya A.S., Gukovsky I., Pizzo D.P.. **Loss of acinar cell IKKα triggers spontaneous pancreatitis in mice**. *J. Clin. Investig.* (2013) **123** 2231-2243. DOI: 10.1172/JCI64498
71. Khor B., Conway K.L., Omar A.S., Biton M., Haber A.L., Rogel N., Baxt L.A., Begun J., Kuballa P., Gagnon J.D.. **Distinct Tissue-Specific Roles for the Disease-Associated Autophagy Genes ATG16L2 and ATG16L1**. *J. Immunol.* (2019) **203** 1820-1829. DOI: 10.4049/jimmunol.1800419
72. Don Wai Luu L., Kaakoush N.O., Castaño-Rodríguez N.. **The role of ATG16L2 in autophagy and disease**. *Autophagy* (2022) **18** 2537-2546. DOI: 10.1080/15548627.2022.2042783
73. Jacquet M., Guittaut M., Fraichard A., Despouy G.. **The functions of Atg8-family proteins in autophagy and cancer: Linked or unrelated?**. *Autophagy* (2021) **17** 599-611. DOI: 10.1080/15548627.2020.1749367
74. Koutsifeli P., Varma U., Daniels L.J., Annandale M., Li X., Neale J.P.H., Hayes S., Weeks K.L., James S., Delbridge L.M.D.. **Glycogen-autophagy: Molecular machinery and cellular mechanisms of glycophagy**. *J. Biol. Chem.* (2022) **298** 102093. DOI: 10.1016/j.jbc.2022.102093
75. Carvalho C., Santos M.S., Oliveira C.R., Moreira P.I.. **Alzheimer’s disease and type 2 diabetes-related alterations in brain mitochondria, autophagy and synaptic markers**. *Biochim. Biophys. Acta* (2015) **1852** 1665-1675. DOI: 10.1016/j.bbadis.2015.05.001
76. Vaillant-Beuchot L., Mary A., Pardossi-Piquard R., Bourgeois A., Lauritzen I., Eysert F., Kinoshita P.F., Cazareth J., Badot C., Fragaki K.. **Accumulation of amyloid precursor protein C-terminal fragments triggers mitochondrial structure, function, and mitophagy defects in Alzheimer’s disease models and human brains**. *Acta Neuropathol.* (2021) **141** 39-65. DOI: 10.1007/s00401-020-02234-7
77. Du Y., Wooten M.C., Gearing M., Wooten M.W.. **Age-associated oxidative damage to the p62 promoter: Implications for Alzheimer disease**. *Free Radic. Biol. Med.* (2009) **46** 492-501. DOI: 10.1016/j.freeradbiomed.2008.11.003
78. Broderick T.L., Rasool S., Li R., Zhang Y., Anderson M., Al-Nakkash L., Plochocki J.H., Geetha T., Babu J.R.. **Neuroprotective Effects of Chronic Resveratrol Treatment and Exercise Training in the 3xTg-AD Mouse Model of Alzheimer’s Disease**. *Int. J. Mol. Sci.* (2020) **21**. DOI: 10.3390/ijms21197337
|
---
title: 'Sleep Disturbances in Generalized Anxiety Disorder: The Role of Calcium Homeostasis
Imbalance'
authors:
- Elvira Anna Carbone
- Giulia Menculini
- Renato de Filippis
- Martina D’Angelo
- Pasquale De Fazio
- Alfonso Tortorella
- Luca Steardo
journal: International Journal of Environmental Research and Public Health
year: 2023
pmcid: PMC10002427
doi: 10.3390/ijerph20054431
license: CC BY 4.0
---
# Sleep Disturbances in Generalized Anxiety Disorder: The Role of Calcium Homeostasis Imbalance
## Abstract
Patients with a generalized anxiety disorder (GAD) often report preeminent sleep disturbances. Recently, calcium homeostasis gained interest because of its role in the regulation of sleep–wake rhythms and anxiety symptoms. This cross-sectional study aimed at investigating the association between calcium homeostasis imbalance, anxiety, and quality of sleep in patients with GAD. A total of 211 patients were assessed using the Hamilton Rating Scale for Anxiety (HAM-A), Pittsburgh Sleep Quality Index questionnaire (PSQI) and Insomnia Severity Index (ISI) scales. Calcium, vitamin D, and parathyroid hormone (PTH) levels were evaluated in blood samples. A correlation and linear regression analysis were run to evaluate the association of HAM-A, PSQI, and ISI scores with peripheral markers of calcium homeostasis imbalance. Significant correlations emerged between HAM-A, PSQI, ISI, PTH, and vitamin D. The regression models showed that patients with GAD displaying low levels of vitamin D and higher levels of PTH exhibit a poor subjective quality of sleep and higher levels of anxiety, underpinning higher psychopathological burden. A strong relationship between peripheral biomarkers of calcium homeostasis imbalance, insomnia, poor sleep quality, and anxiety symptomatology was underlined. Future studies could shed light on the causal and temporal relationship between calcium metabolism imbalance, anxiety, and sleep.
## 1. Introduction
Sleep is a basic human need and is essential for good health, well-being, and good quality of life. We spend nearly a third of our life sleeping. However, people often experience difficulties in sleeping that may become disabling and result in daytime dysfunction [1,2,3]. According to the third edition of the International Classification of Sleep Disorders (ICSD-3), insomnia is characterized by difficulty in either initiating, maintaining, or continuing sleep, despite the adequate opportunity and condition for sleep. Nowadays, insomnia represents the most common sleep disorder [4,5] affecting especially women and older people, and it coexists very frequently with general health problems (e.g., cardiovascular diseases, chronic pain syndrome, diabetes, obesity, asthma) [6]. Sleep disturbances are commonly detected in the general population and individuals with psychiatric disorders [7]. Considering that sleep can affect mental health, having a psychiatric disorder, in turn, could impact on sleep quality. Studies indicate that insomnia very often coexists with psychiatric disorders [8]. Particularly, insomnia is most frequently associated with major depression or an anxiety disorder, mainly, generalized anxiety disorder (GAD) [9].
About 60–$70\%$ of patients with GAD and panic disorder reported prominent sleep disturbances [9], leading to a negative impact on functioning and quality of life [10] and the course and treatment of psychiatric illness [11].
Sleep–wake regulation is classically described as resulting from the interaction of circadian and homeostatic processes [12], which in turn influence the opposite activity of neurons stimulating wakefulness and neurons stimulating sleep [13]. The dysregulation of this process and consequent insomnia seems to be linked to the alteration of different hormones such as insulin, cortisol, leptin, orexin, ghrelin or growth factor, and vitamin D [14,15,16,17,18,19,20].
In recent years, calcium homeostasis has received increasing interest, with research supporting the role of parathyroid hormone (PTH), vitamin D (Vit D), and calcium (Ca++) in mental health conditions [21]. Vit D, together with PTH, regulates the homeostasis of Ca++, modulating calcium transportation in the gut, bone, and kidney and the immune modulation, the antioxidant defense system, and several inflammatory processes [22,23,24]. By appropriate actions of Vit D and PTH, Ca++ is maintained in the range or promptly corrected if necessary. An alteration or defect of any of this system results in the calcium homeostasis imbalance. It was already demonstrated in schizophrenia [25], depression [26], bipolar disorder [27], anxiety [28,29,30], and sleep disorders [31,32,33,34,35].
This could be explained considering different activities of Vit D, Ca++, and PTH. Vit D receptors are widely expressed in all human bodies and brain [36,37,38,39] and their increased expression was demonstrated in specific brain regions involved in anxiety and sleep regulation, such as the prefrontal cortex and the limbic system [40,41]. In these areas, particularly in the prefrontal cortex [42], Vit D can directly increase the biosynthesis of dopamine/noradrenaline and serotonin [43,44,45,46], and improve the expression of the growth factor hormone and the BDNF [47,48,49]. Ca++ is very important in the central nervous system (CNS) as a cofactor, second messenger, and signaling molecule, and for transmitters release [50]. Additionally, PTH contributes to neuronal homeostasis [51] regulating circulating and intracellular calcium levels in the CNS [52].
Vit D has gained prominence due to its antioxidant, anti-inflammatory, pro-neurogenic, and neuromodulator properties that appear to be fundamental to its anxiolytic properties [53,54,55,56]. Data are supported by studies demonstrating that supplementation of Vit D can improve anxiety symptoms, [57,58,59] as well as sleep disorders and sleep quality [60]. On the other hand, experimental evidence has shown that Ca++ signaling plays a crucial role in regulating sleep–wake rhythms [61]. There is also evidence suggesting that increased dietary Ca++ intake improves anxiety [62], quality of sleep, and reduces insomnia [63,64]. Interestingly, total Ca++ presents a diurnal variability during normal sleep [65], underlining the role in regulating sleep duration in mammals [66], possibly due to the involvement in producing melatonin from tryptophan in the brain [67].
Although several studies investigated the co-occurrence of sleep disturbances and anxiety disorders [68,69], showing that the relationship between these two conditions is particularly complex [70], few studies focused on calcium homeostasis imbalance and data are not conclusive. Therefore, such experimental evidence led clinicians to comprehensively investigate the effect of calcium metabolism imbalance on anxiety disorders.
Based on the above, the current study aimed at investigating the association between calcium imbalance through the determination of Ca++, Vit D, and PTH levels, anxiety psychopathology severity, and altered hypnic pattern in a sample of patients suffering from a generalized anxiety disorder. Thus, the current study tries to explore whether calcium metabolism imbalance could be associated with sleep quality and worsening of symptoms in patients with anxiety disorders. Therefore, the aims of the present study are [1] to identify the association between calcium imbalance and quality of sleep in patients suffering from generalized anxiety disorder (GAD) and [2] to evaluate how this association may impact illness severity in patients suffering from GAD.
## 2. Materials and Methods
Consecutive outpatients were screened for eligibility at the Psychiatric Unit of the University Hospital Mater Domini in Catanzaro from May 2020 to July 2022. Inclusion criteria were age between 18 and 75 years; primary diagnosis of GAD according to the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) [8]; and willingness to participate in the study. Participants were considered not eligible in cases of an inability to provide a written informed consent to participate in the study; presence of moderate or severe cognitive impairment as assessed at the first contact visit by Mini-Mental State Evaluation (MMSE) ≤22 [71]; comorbidity with neurologic diseases, endocrinological diseases (hypo/hyperparathyroidism), or substance and/or alcohol use disorders; pregnancy or post-partum period; current treatment with medications that can alter calcium metabolism, such as Vit D supplementation or calcium phosphonate or bisphosphonates.
Patients presenting comorbid depressive features were not excluded, considering the high prevalence of anxiety and depressive symptoms co-occurrence in clinical practice. However, we excluded patients with a severe or subthreshold depressive condition clinically evaluated at the moment of the enrollment.
All participants meeting the inclusion/exclusion criteria were recruited and included in the study after receiving a full description of the study aims and design and obtaining their written informed consent to participate in the study. The Structured Interview for DSM-5 Disorders, Clinician Version (SCID-5-CV) [72] was used for the diagnosis. All tests were performed by experienced psychiatrists who were trained in the administration of neuropsychiatric tests and used these tools in their daily clinical practice.
The study was carried out following the latest version of the Declaration of Helsinki and the protocol approval was obtained by the Ethics Committee of the University of Catanzaro ($\frac{307}{2020}$).
## 2.1. Procedures and Measures
Patients’ socio-demographic and clinical characteristics were collected using an ad hoc schedule evaluating sex, age, civil status, education, employment status, family history of psychiatric illnesses, and age at onset of the disorder.
## 2.1.1. Psychological Measures
Participant answered the following scales:Hamilton Rating Scale for Anxiety (HAM-A) [73], to assess the clinical severity of anxiety symptoms. The scale consists of 14 items scoring on a scale of 0 (not present) to 4 (severe). Each item is defined by a series of symptoms, and measures both psychic anxiety (mental agitation and psychological distress) and somatic anxiety (physical complaints related to anxiety). The total score ranges from 0 to 56, where <17 indicates mild severity, 18–24 mild to moderate severity, and 25–30 moderate to severe. Cronbach’s alpha was 0.934 in this study. Pittsburgh Sleep Quality Index Questionnaire (PSQI), to analyze sleep quality. The self-reported questionnaire is made up of 19 items, used to create seven components with a score ranging between 0 (no problem) and 3 (major problem), namely, subjective sleep quality (hereafter referred to as Quality), sleep latency (Latency), sleep duration (Duration), habitual sleep efficiency (Efficiency), sleep disturbances (Disturbances), use of sleeping medication (Medication), and daytime dysfunction (Dysfunction). The total score from these seven components varies between 0 (no problem) and 21 (major problem). A global score of ≥5 is used to identify people with poor sleep quality [74,75]. People with a score of 5 or higher, experienced poor sleep quality, and those with a score of less than 5 experienced good sleep quality. Cronbach’s alpha was 0.77 [76]. Cronbach’s alpha was 0.834 in this study. Insomnia Severity Index (ISI), to assess the nature, severity, and impact of sleep difficulties in the last 2 weeks. A 5-point Likert scale is used to rate the 7 items, with scores ranging 0–28 that yield four categories: absence of insomnia (0–7); subthreshold insomnia (8–14); moderate insomnia (15–21); and severe insomnia (22–28) [77]. Cronbach’s alpha was 0.784 in this study.
## 2.1.2. Biological Measures
Serum levels of calcium (mmol/L), 25-OH-vitamin D (ng/mL), and PTH (pg/mL) were assessed in the same laboratory to ensure standardized procedures. Blood samples were collected from all patients at recruitment after 12–14 h fasting.
Calcium was measured using standard laboratory methods. Blood was centrifuged, and serum was stocked at −30 °C for α,25 (OH)2 vitamin D and PTH and evaluated by chemiluminescence immunoassays using adequate kits (Diasorin Liaison; ADVIA Centaur). According to the Endocrine Society’s Clinical Practice Guideline, Vit D deficiency was considered when its values were <20 ng/mL; insufficiency between 21–29 ng/mL; and sufficiency between 30–100 ng/mL [78]. Levels of Ca++ between 8.9 and 10.01 mg/dL are considered normal, whilst the range 15–55 pg/mL is considered normal for the PTH.
Levels of Vit D < 20 ng/mL, Ca++ < 8.8 mg/dL or >10 mg/dL, and PTH < 15 pg/mL or >55 pg/mL were the cut-off considered for calcium homeostasis imbalance (Table 1).
## 2.2. Statistical Analysis
Descriptive statistics were calculated for socio-demographic and clinical characteristics, as well as for scores at relevant assessment instruments. The quantitative variables were expressed as mean and standard deviation (SD) and the qualitative variables as frequency and percentage (%).
A Spearman correlation analysis was used to assess the relationship between sleep quality, anxiety symptoms, and calcium homeostasis imbalance. Linear regression analysis was performed to further investigate the relationship between sleep quality, anxiety, and calcium homeostasis imbalance using PSQI, ISI, and HAM-A scores as dependents variables and PTH, calcium, and Vit D as independent variables. All tolerance values in the regression analyses were >0.1 and all variance inflation factors were <10, expressing that the assumption of multicollinearity was not violated. The p-value < 0.05 was considered significant in this study. Data were analyzed with the Statistical Package for Social Sciences Version 26 (SPSS, Chicago, IL, USA) [79].
## 3. Results
Overall, 211 participants suffering from GAD met the inclusion/exclusion criteria and were enrolled in the study. The average age (±standard deviation, SD) was 46.9 (±13.8). Most of the participants were female ($51\%$), married ($45.5\%$), graduated ($76\%$), employed ($63\%$), and with positive family history for psychiatric disorders ($64.5\%$). The mean age at onset was 27.8 ± 11.1. The mean of HAM-A total, PSQI total, ISI total was 25.6 ± 13.7, 10.96 ± 6.2 and 14.36 ± 8.2, respectively. Indices of calcium metabolism showed a normal calcium level 9.5 ± 0.4, higher PTH level (54.6 ± 20.5), and lower Vit D level (29.4 ± 25.1) (Table 2).
Table 3 includes the results of Spearman’s correlations between HAM total score, PSQI subscales and total score, ISI total score, calcium, PTH, and Vit D. Significant correlations emerged for all the variables, with the sole exception of Ca++.
A linear regression analysis was performed to assess the association between calcium imbalance, anxiety symptoms, and quality of sleep. In the three models, PSQI total, HAM-A total, and ISI total, respectively, were selected as dependent variables and PTH, Vit D, and Ca++ as independent variables. In the first model, higher PTH levels and lower Vit D levels (R2 = 0.603; $F = 80.752$; $p \leq 0.001$) predicted PSQI total; in the second model, higher PTH levels and lower Vit D levels predicted HAM-A total (R2 = 0.685; $F = 115.137$; $p \leq 0.001$), and in the last model, higher PTH levels and lower Vit D levels predicted ISI total (R2 = 0.672; $F = 105.516$; $p \leq 0.001$). Thus, an imbalance of PTH and Vit D levels predicted insomnia, higher levels of anxiety, and poor quality of sleep. See Table 4.
## 4. Discussion
This study found a strong relationship between calcium homeostasis imbalance, poor sleep quality, and anxiety symptomatology in patients suffering from GAD. To the best of our knowledge, this is the first study aimed at investigating the association between calcium homeostasis imbalance and quality of sleep in patients with GAD. The study findings suggest that patients with GAD and low levels of Vit D and higher levels of PTH exhibit insomnia, poor quality of sleep, and higher levels of anxiety, highlighting its impact on the psychopathological burden.
A growing body of literature focused on the calcium imbalance in psychiatric disorders [21,25,27,28,29,30,31,32,33,34,35] and our results are in line with them. In our sample, significant correlations emerged for PSQI, HAM-A, ISI, PTH, and Vit D. The association between poor sleep quality and high levels of PTH and low levels of Vit D may be read considering the sleep–wake dysregulation as a consequence of calcium imbalance [20]. Recently, a growing number of studies and a recent meta-analysis reported the link between Vit D and sleep [35]. Adequate levels of this hormone seem to be necessary for the maintenance of sleep, reducing the number of nocturnal awakenings [80] while low Vit D levels have been reported to be associated with shorter sleep duration [81,82]. Although the exact mechanism by which Vit D affects sleep regulation is still unclear, the key to this link seems to be the expression of Vit D receptors in the cortical and subcortical areas of the brainstem that are involved in sleep control [83] such as prefrontal cortex [84], cingulate gyrus [85], hippocampus [86], caudate nucleus [87], lateral geniculate nucleus [88], and substantia nigra [83,89].
Interestingly, Vit D is involved in regulating the conversion of tryptophan into 5-HTP and producing melatonin [90] from tryptophan in the brain [67,90]. Melatonin participates in the regulation of circadian rhythms [91] and adjusts the sleep–wake cycle with a consequent positive effect on the quality of sleep [92]. In fact, epidemiology studies found that dietary intake of Vit D was related to the midpoint of sleep, sleep duration, and maintaining sleep [93,94]. In this regard, it seems important to consider that in our sample some patients reported subthreshold depressive symptoms. The data is not surprising because it is well known that anxiety disorders, as well as sleep disturbances, often manifest in comorbidity with depressive symptoms [95]. In fact, other studies indicated that the serotoninergic pathway was implicated in the initiation and maintenance of sleep in different areas of the brain that have been associated with the sleep regulation and that Vit D plays a key function in the regulation of the serotonergic pathway [46] and melatonin production. Moreover, Vit D contributes to neuroplasticity [59] and in the synthesis of other neurotransmitters [96,97,98], confirming the importance of Vit D in sleep but also mood regulation [99].
Most studies evaluating anxiety-related symptoms in different populations indicate an association between low levels of Vit D and anxiety [28,100,101], and some reported that Vit D supplementation is associated with lower anxiety symptoms [102]. In our sample, the regression analysis confirmed the significant association between higher PTH and lower Vit D levels, poor quality of sleep, and anxiety symptomatology emphasizing the close relationship between calcium imbalance and psychopathology in patients with GAD. This finding can be explained by the role of calcium imbalance, especially Vit D, in many brain processes, including neuroimmunomodulation, neuroinflammation, oxidative stress, and neuroplasticity [59] and synthesis of neurotransmitters, all implicated in the pathogenesis of anxiety disorders [96,97,98]. In this regard, Vit D seems implicated in the synthesis of serotonin neurotransmitters through the tryptophan pathways [46]. The alteration of the serotonin synthesis is associated with the prefrontal cortex [103], hippocampal [104] and amygdala dysfunctions [105], brain regions important in regulating network activity, and neural oscillations in anxiety disorders [106,107].
On the other hand, many of the positive effects of Vit D on behaviors might be associated with its ability to regulate both peripheral and CNS immune responses. As noted, anxiety is frequently associated with a low-grade inflammatory status and peripheral increase of inflammatory cytokines [108,109]. As such, Vit D may help reduce anxiety symptoms because of its antioxidant and anti-inflammatory properties. More recently, the preclinical study described the anti-inflammatory and antioxidant effects of the pretreatment with Vit D3 underlying the ability of this vitamin to annul anxiety-like behaviors. Indeed, this effect was accompanied by a decrease in IL-6 levels [110]. Results were replicated in a clinical sample: Vit D supplementation in combination with standard of care improved the severity of anxiety in individuals diagnosed with GAD by increasing serotonin concentrations and decreasing the levels of the inflammatory biomarker neopterin [111].
The results of the present study should be read considering some limitations. First, the cross-sectional study design, the type of patients included (only outpatients), and the relatively small sample size does not allow to generalize to a large proportion of the psychiatric population and preclude establishing causal relationships. In this light, prospective studies are recommended. Second, the self-administered scale and the retrospective nature of the study were affected by the effect of recall bias and represent a structural limitation regarding the assembly and reliability of the data. Third, psychiatric medications are known to trigger symptoms of sleep disorders. Due to heterogeneity in our sample, patients were prescribed different psychotropic medications which would be difficult to control. Hence, it was not possible to examine the association between psychotropic medication and symptoms of sleep disorders. Lastly, the wide overlap of features and neurophysiological systems involved in anxiety and depressive symptoms, even if occurring only in a few patients of our sample, prevented us to examine the unique relationship between calcium imbalance and anxiety disorder. Further studies should assess the role that calcium imbalance plays in this relationship, distinguishing mood disorders from anxiety disorders and using major depressive disorder as a control group.
Despite these limitations, the major strengths of this study are represented by the focus on calcium imbalance and sleep quality in patients with GAD in a real-world setting with broad inclusion criteria. Furthermore, this was the first attempt to evaluate the role and implications of calcium homeostasis in GAD, considering its relationships to sleep and anxiety symptoms. Moreover, the study includes the concomitant assessment of Vit D, PTH, and Ca++ levels to assess and analyze the whole metabolism axis. Nevertheless, future large-scale prospective studies are needed to confirm the findings of this study and to better clarify the association between calcium imbalance, sleep quality, and psychopathology severity. Identifying and addressing sleep quality, insomnia, and calcium imbalance may have a positive impact on the prognosis and quality of life of patients with GAD.
## 5. Conclusions
In conclusion, the study found a strong association between levels of parathyroid hormone and Vit D, sleep quality, and anxiety symptomatology in patients suffering from GAD. The study results suggest that patients with GAD and low levels of Vit D and higher levels of PTH exhibit poor quality of sleep and higher levels of anxiety highlighting its impact on the psychopathological burden. Results should suggest that calcium homeostasis may be disrupted in this population but additional prospective studies in real-world settings with direct comparisons between these two conditions are needed. Therefore, it may represent an area of clinical research interest for the future, to reach more patients focused on clinical practice to anticipate a precise diagnosis, manage personalized treatment, and improve prognosis. Indeed, future studies could shed light on the causal and temporal relationship existing between calcium metabolism imbalance, anxiety, and sleep, opening new and interesting frontiers in both clinical and research fields.
## References
1. Bixler E.O., Kales A., Soldatos C.R., Kales J.D., Healey S.. **Prevalence of sleep disorders in the Los Angeles metropolitan area**. *Am. J. Psychiatry* (1979) **136** 1257-1262. DOI: 10.1176/ajp.136.10.1257
2. Mellinger G.D., Balter M.B., Uhlenhuth E.H.. **Insomnia and Its Treatment: Prevalence and Correlates**. *Arch. Gen. Psychiatry* (1985) **42** 225-232. DOI: 10.1001/archpsyc.1985.01790260019002
3. Ohayon M.. **Epidemiological Study on Insomnia in the General Population**. *Sleep* (1996) **19** S7-S15. DOI: 10.1093/sleep/19.suppl_3.S7
4. Morin C.M., Benca R.M.. **Insomnia Nature, Diagnosis, and Treatment**. *Handbook of Clinical Neurology* (2011) **Volume 99** 723-746. DOI: 10.1016/b978-0-444-52007-4.00004-7
5. Morin C.M., Benca R.. **Chronic insomnia**. *Lancet* (2012) **379** 1129-1141. DOI: 10.1016/S0140-6736(11)60750-2
6. Ohayon M.M.. **Epidemiology of insomnia: What we know and what we still need to learn**. *Sleep Med. Rev.* (2002) **6** 97-111. DOI: 10.1053/smrv.2002.0186
7. Benca R.M., Obermeyer W.H., Thisted R.A., Gillin J.C.. **Sleep and Psychiatric Disorders. A Meta-Analysis**. *Arch. Gen. Psychiatry* (1992) **49** 651-668. DOI: 10.1001/archpsyc.1992.01820080059010
8. 8.
American Psychiatric Association
Diagnostic and Statistical Manual of Mental Disorders. DSM-55th ed.American Psychiatric PublishingArlington, VA, USA2013. *Diagnostic and Statistical Manual of Mental Disorders. DSM-5* (2013)
9. Monti J.M., Monti D.. **Sleep disturbance in generalized anxiety disorder and its treatment**. *Sleep Med. Rev.* (2000) **4** 263-276. DOI: 10.1053/smrv.1999.0096
10. Szentkiralyi A., Madarász C.Z., Novak M.. **Sleep disorders: Impact on daytime functioning and quality of life**. *Expert Rev. Pharm. Outcomes Res.* (2009) **9** 49-64. DOI: 10.1586/14737167.9.1.49
11. Hartwig E.M., Rufino K.A., Palmer C.A., Shepard C., Alfano C.A., Schanzer B., Mathew S.J., Patriquin M.A.. **Trajectories of self-reported sleep disturbance across inpatient psychiatric treatment predict clinical outcome in comorbid major depressive disorder and generalized anxiety disorder**. *J. Affect. Disord.* (2019) **251** 248-255. DOI: 10.1016/j.jad.2019.03.069
12. Dijk D.-J., Czeisler C.A.. **Paradoxical timing of the circadian rhythm of sleep propensity serves to consolidate sleep and wakefulness in humans**. *Neurosci. Lett.* (1994) **166** 63-68. DOI: 10.1016/0304-3940(94)90841-9
13. Mignot E., Taheri S., Nishino S.. **Sleeping with the hypothalamus: Emerging therapeutic targets for sleep disorders**. *Nat. Neurosci.* (2002) **5** 1071-1075. DOI: 10.1038/nn944
14. Sakurai T., Mieda M., Tsujino N.. **The orexin system: Roles in sleep/wake regulation**. *Ann. N. Y. Acad. Sci.* (2010) **1200** 149-161. DOI: 10.1111/j.1749-6632.2010.05513.x
15. Öztürk Ö., Cebeci D., Şahin U., Tülüceoğlu E.E., Calapoğlu N.Ş., Gonca T., Calapoğlu M.. **Circulating levels of ghrelin, galanin, and orexin-A orexigenic neuropeptides in obstructive sleep apnea syndrome**. *Sleep Breath.* (2021) **26** 1209-1218. DOI: 10.1007/s11325-021-02514-w
16. Mieda M., Sakurai T.. **Overview of orexin/hypocretin system**. *Prog. Brain Res.* (2012) **198** 5-14. DOI: 10.1016/b978-0-444-59489-1.00002-1
17. El Mlili N., Ahabrach H., Cauli O.. **Hair Cortisol Concentration as a Biomarker of Sleep Quality and Related Disorders**. *Life* (2021) **11**. DOI: 10.3390/life11020081
18. Jain S.K., Kahlon G., Morehead L., Lieblong B., Stapleton T., Hoeldtke R., Bass P.F., Levine S.N.. **The Effect of Sleep Apnea and Insomnia on Blood Levels of Leptin, Insulin Resistance, IP-10, and Hydrogen Sulfide in Type 2 Diabetic Patients**. *Metab. Syndr. Relat. Disord.* (2012) **10** 331-336. DOI: 10.1089/met.2012.0045
19. Forouzanfar F., Negah S.S.. **Dual Role of Fibroblast Growth Factor Pathways in Sleep Regulation**. *Endocr. Metab. Immune Disord.-Drug Targets* (2023) **23** 63-69. DOI: 10.2174/1871530322666220802161031
20. Romano F., Muscogiuri G., Di Benedetto E., Zhukouskaya V.V., Barrea L., Savastano S., Colao A., Di Somma C.. **Vitamin D and Sleep Regulation: Is there a Role for Vitamin D?**. *Curr. Pharm. Des.* (2020) **26** 2492-2496. DOI: 10.2174/1381612826666200310145935
21. Głąbska D., Kołota A., Lachowicz K., Skolmowska D., Stachoń M., Guzek D.. **Vitamin D Supplementation and Mental Health in Multiple Sclerosis Patients: A Systematic Review**. *Nutrients* (2021) **13**. DOI: 10.3390/nu13124207
22. Sassi F., Tamone C., D’Amelio P.. **Vitamin D: Nutrient, Hormone, and Immunomodulator**. *Nutrients* (2018) **10**. DOI: 10.3390/nu10111656
23. Potts J.T., Gardella T.J.. **Progress, Paradox, and Potential: Parathyroid Hormone Research over Five Decades**. *Ann. N. Y. Acad. Sci.* (2007) **1117** 196-208. DOI: 10.1196/annals.1402.088
24. Brown E.. **The Calcium-Sensing Receptor: Physiology, Pathophysiology and Car-Based Therapeutics**. *Subcell. Biochem.* (2007) **45** 139-167. DOI: 10.1007/978-1-4020-6191-2_6
25. Cieslak K., Feingold J., Antonius D., Walsh-Messinger J., Dracxler R., Rosedale M., Aujero N., Keefe D., Goetz D., Goetz R.. **Low Vitamin D levels predict clinical features of schizophrenia**. *Schizophr. Res.* (2014) **159** 543-545. DOI: 10.1016/j.schres.2014.08.031
26. Maddock J., Berry D.J., Geoffroy M.-C., Power C., Hyppönen E.. **Vitamin D and common mental disorders in mid-life: Cross-sectional and prospective findings**. *Clin. Nutr.* (2013) **32** 758-764. DOI: 10.1016/j.clnu.2013.01.006
27. Steardo L., Luciano M., Sampogna G., Carbone E., Caivano V., Di Cerbo A., Giallonardo V., Palummo C., Vece A., Del Vecchio V.. **Clinical Severity and Calcium Metabolism in Patients with Bipolar Disorder**. *Brain Sci.* (2020) **10**. DOI: 10.3390/brainsci10070417
28. Bičíková M., Duskova M., Vítků J., Kalvachová B., Řípová D., Mohr P., Stárka L.. **Vitamin D in Anxiety and Affective Disorders**. *Physiol. Res.* (2015) **64** S101-S103. DOI: 10.33549/physiolres.933082
29. Han B., Zhu F.-X., Yu H.-F., Liu S., Zhou J.-L.. **Low serum levels of vitamin D are associated with anxiety in children and adolescents with dialysis**. *Sci. Rep.* (2018) **8** 1-6. DOI: 10.1038/s41598-018-24451-7
30. Wu C., Ren W., Cheng J., Zhu B., Jin Q., Wang L., Chen C., Zhu L., Chang Y., Gu Y.. **Association Between Serum Levels of Vitamin D and the Risk of Post-Stroke Anxiety**. *Medicine* (2016) **95**. DOI: 10.1097/MD.0000000000003566
31. Shi S., Ueda H.R.. **Ca**. *Bioessays* (2018) **40**. DOI: 10.1002/bies.201700105
32. Mansourian M., Rafie N., Khorvash F., Hadi A., Arab A.. **Are serum vitamin D, calcium and phosphorous associated with restless leg syndrome? A systematic review and meta-analysis**. *Sleep Med.* (2020) **75** 326-334. DOI: 10.1016/j.sleep.2020.08.022
33. Mosavat M., Smyth A., Arabiat D., Whitehead L.. **Vitamin D and sleep duration: Is there a bidirectional relationship?**. *Horm. Mol. Biol. Clin. Investig.* (2020) **41**. DOI: 10.1515/hmbci-2020-0025
34. Jeon Y.-S., Yu S., Kim C., Lee H.J., Yoon I.-Y., Kim T.. **Lower Serum Calcium Levels Associated with Disrupted Sleep and Rest–Activity Rhythm in Shift Workers**. *Nutrients* (2022) **14**. DOI: 10.3390/nu14153021
35. Gao Q., Kou T., Zhuang B., Ren Y., Dong X., Wang Q.. **The Association between Vitamin D Deficiency and Sleep Disorders: A Systematic Review and Meta-Analysis**. *Nutrients* (2018) **10**. DOI: 10.3390/nu10101395
36. Cui X., Gooch H., Groves N.J., Sah P., Burne T.H., Eyles D.W., McGrath J.J.. **Vitamin D and the brain: Key questions for future research**. *J. Steroid Biochem. Mol. Biol.* (2015) **148** 305-309. DOI: 10.1016/j.jsbmb.2014.11.004
37. DeLuca G.C., Kimball S.M., Kolasinski J., Ramagopalan S.V., Ebers G.C.. **Review: The role of vitamin D in nervous system health and disease**. *Neuropathol. Appl. Neurobiol.* (2013) **39** 458-484. DOI: 10.1111/nan.12020
38. Eyles D., Burne T., McGrath J.. **Vitamin D in fetal brain development**. *Semin. Cell Dev. Biol.* (2011) **22** 629-636. DOI: 10.1016/j.semcdb.2011.05.004
39. Harms L.R., Burne T.H.J., Eyles D.W., McGrath J.J.. **Vitamin D and the brain**. *Best Pract. Res. Clin. Endocrinol. Metab.* (2011) **25** 657-669. DOI: 10.1016/j.beem.2011.05.009
40. Eyles D.W., Smith S., Kinobe R., Hewison M., McGrath J.J.. **Distribution of the Vitamin D receptor and 1α-hydroxylase in human brain**. *J. Chem. Neuroanat.* (2005) **29** 21-30. DOI: 10.1016/j.jchemneu.2004.08.006
41. Berk M., Post R., Ratheesh A., Gliddon E., Singh A., Vieta E., Carvalho A.F., Ashton M.M., Berk L., Cotton S.M.. **Staging in bipolar disorder: From theoretical framework to clinical utility**. *World Psychiatry* (2017) **16** 236-244. DOI: 10.1002/wps.20441
42. Jiang P., Zhang L.-H., Cai H.-L., Li H.-D., Liu Y.-P., Tang M.-M., Dang R.-L., Zhu W.-Y., Xue Y., He X.. **Neurochemical Effects of Chronic Administration of Calcitriol in Rats**. *Nutrients* (2014) **6** 6048-6059. DOI: 10.3390/nu6126048
43. Kalueff A.V., Tuohimaa P.. **Neurosteroid hormone vitamin D and its utility in clinical nutrition**. *Curr. Opin. Clin. Nutr. Metab. Care* (2007) **10** 12-19. DOI: 10.1097/MCO.0b013e328010ca18
44. Kesby J.P., Turner K.M., Alexander S., Eyles D.W., McGrath J.J., Burne T.H.. **Developmental vitamin D deficiency alters multiple neurotransmitter systems in the neonatal rat brain**. *Int. J. Dev. Neurosci.* (2017) **62** 1-7. DOI: 10.1016/j.ijdevneu.2017.07.002
45. Lima L.A.R., Lopes M.J.P., Costa R.O., Lima F.A.V., Neves K.R.T., Calou I.B.F., Andrade G.M., Viana G.S.B.. **Vitamin D protects dopaminergic neurons against neuroinflammation and oxidative stress in hemiparkinsonian rats**. *J. Neuroinflammat.* (2018) **15** 249. DOI: 10.1186/s12974-018-1266-6
46. Patrick R.P., Ames B.N.. **Vitamin D and the omega-3 fatty acids control serotonin synthesis and action, part 2: Relevance for ADHD, bipolar disorder, schizophrenia, and impulsive behavior**. *FASEB J.* (2015) **29** 2207-2222. DOI: 10.1096/fj.14-268342
47. Gezen-Ak D., Dursun E., Yilmazer S.. **The Effect of Vitamin D Treatment on Nerve Growth Factor (NGF) Release from Hippocampal Neurons**. *Nöro Psikiyatr Arşivi* (2014) **51** 157. DOI: 10.4274/npa.y7076
48. Favre P., Pauling M., Stout J., Hozer F., Sarrazin S., Abé C., Alda M., Alloza C., Alonso-Lana S., Andreassen O.A.. **Widespread White Matter Microstructural Abnormalities Inbipolar Disorder: Evidence from Mega- and Meta-Analysesacross 3033 Individuals**. *Neuropsychopharmacology* (2019) **9** 2285-2293. DOI: 10.1038/s41386-019-0485-6
49. Khairy E.Y., Attia M.M.. **Protective effects of vitamin D on neurophysiologic alterations in brain aging: Role of brain-derived neurotrophic factor (BDNF)**. *Nutr. Neurosci.* (2019) **24** 650-659. DOI: 10.1080/1028415X.2019.1665854
50. Yarlagadda A., Kaushik S., Clayton A.H.. **Blood brain barrier: The role of calcium homeostasis**. *Psychiatry (Edgmont)* (2007) **4** 55. PMID: 20436765
51. Brown S.J., Ruppe M.D., Tabatabai L.S.. **The Parathyroid Gland and Heart Disease**. *Methodist DeBakey Cardiovasc. J.* (2017) **13** 49-54. DOI: 10.14797/mdcj-13-2-49
52. Shaheen M., Cheema Y., Shahbaz A.U., Bhattacharya S.K., Weber K.T.. **Intracellular Calcium Overloading and Oxidative Stress in Cardiomyocyte Necrosis via a Mitochondriocentric Signal-Transducer-Effector Pathway**. *Exp. Clin. Cardiol.* (2011) **16** 109. PMID: 22131852
53. Kumar R., Rathi H., Haq A., Wimalawansa S.J., Sharma A.. **Putative roles of vitamin D in modulating immune response and immunopathology associated with COVID-19**. *Virus Res.* (2021) **292**. DOI: 10.1016/j.virusres.2020.198235
54. Berk M., Sanders K.M., Pasco J.A., Jacka F.N., Williams L.J., Hayles A.L., Dodd S.. **Vitamin D deficiency may play a role in depression**. *Med. Hypotheses* (2007) **69** 1316-1319. DOI: 10.1016/j.mehy.2007.04.001
55. Fedotova J., Zarembo D., Dragasek J., Caprnda M., Kruzliak P., Dudnichenko T.. **Modulating Effects of Cholecalciferol Treatment on Estrogen Deficiency-Induced Anxiety-Like Behavior of Adult Female Rats**. *Folia Medica* (2017) **59** 139-158. DOI: 10.1515/folmed-2017-0022
56. D’Hellencourt C.L., Montero-Menei C.N., Bernard R., Couez D.. **Vitamin D3 inhibits proinflammatory cytokines and nitric oxide production by the EOC13 microglial cell line**. *J. Neurosci. Res.* (2003) **71** 575-582. DOI: 10.1002/jnr.10491
57. Zhu C., Zhang Y., Wang T., Lin Y., Yu J., Xia Q., Zhu P., Zhu D.. **Vitamin D supplementation improves anxiety but not depression symptoms in patients with vitamin D deficiency**. *Brain Behav.* (2020) **10** e01760. DOI: 10.1002/brb3.1760
58. Kouba B.R., Camargo A., Gil-Mohapel J., Rodrigues A.L.S.. **Molecular Basis Underlying the Therapeutic Potential of Vitamin D for the Treatment of Depression and Anxiety**. *Int. J. Mol. Sci.* (2022) **23**. DOI: 10.3390/ijms23137077
59. Casseb G.A.S., Kaster M.P., Rodrigues A.L.S.. **Potential Role of Vitamin D for the Management of Depression and Anxiety**. *CNS Drugs* (2019) **33** 619-637. DOI: 10.1007/s40263-019-00640-4
60. Abboud M.. **Vitamin D Supplementation and Sleep: A Systematic Review and Meta-Analysis of Intervention Studies**. *Nutrients* (2022) **14**. DOI: 10.3390/nu14051076
61. Herzog E.D., Hermanstyne T., Smyllie N.J., Hastings M.H.. **Regulating the Suprachiasmatic Nucleus (SCN) Circadian Clockwork: Interplay between Cell-Autonomous and Circuit-Level Mechanisms**. *Cold Spring Harb. Perspect. Biol.* (2017) **9**. DOI: 10.1101/cshperspect.a027706
62. Du C., Hsiao P.Y., Ludy M.-J., Tucker R.M.. **Relationships between Dairy and Calcium Intake and Mental Health Measures of Higher Education Students in the United States: Outcomes from Moderation Analyses**. *Nutrients* (2022) **14**. DOI: 10.3390/nu14040775
63. Grandner M.A., Jackson N., Gerstner J.R., Knutson K.L.. **Sleep symptoms associated with intake of specific dietary nutrients**. *J. Sleep Res.* (2014) **23** 22-34. DOI: 10.1111/jsr.12084
64. Nisar M., Mohammad R.M., Arshad A., Hashmi I., Yousuf S.M., Baig S.. **Influence of Dietary Intake on Sleeping Patterns of Medical Students**. *Cureus* (2019) **11**. DOI: 10.7759/cureus.4106
65. Ridefelt P., Axelsson J., Larsson A.. **Diurnal variability of total calcium during normal sleep and after an acute shift of sleep**. *Clin. Chem. Lab. Med.* (2012) **50** 147-151. DOI: 10.1515/cclm.2011.880
66. Tatsuki F., Sunagawa G.A., Shi S., Susaki E.A., Yukinaga H., Perrin D., Sumiyama K., Ukai-Tadenuma M., Fujishima H., Ohno R.-I.. **Involvement of Ca2+-Dependent Hyperpolarization in Sleep Duration in Mammals**. *Neuron* (2016) **90** 70-85. DOI: 10.1016/j.neuron.2016.02.032
67. Kitano N., Tsunoda K., Tsuji T., Osuka Y., Jindo T., Tanaka K., Okura T.. **Association between difficulty initiating sleep in older adults and the combination of leisure-time physical activity and consumption of milk and milk products: A cross-sectional study**. *BMC Geriatr.* (2014) **14** 1-7. DOI: 10.1186/1471-2318-14-118
68. Ohayon M.M., Roth T.. **Place of chronic insomnia in the course of depressive and anxiety disorders**. *J. Psychiatr. Res.* (2003) **37** 9-15. DOI: 10.1016/S0022-3956(02)00052-3
69. Alvaro P.K., Roberts R., Harris J.K.. **A Systematic Review Assessing Bidirectionality between Sleep Disturbances, Anxiety, and Depression**. *Sleep* (2013) **36** 1059-1068. DOI: 10.5665/sleep.2810
70. Khurshid K.A.. **Comorbid Insomnia and Psychiatric Disorders: An Update**. *Innov. Clin. Neurosci.* (2018) **15** 28-32
71. Tombaugh T.N., McIntyre N.J.. **The Mini-Mental State Examination: A Comprehensive Review**. *J. Am. Geriatr. Soc.* (1992) **40** 922-935. DOI: 10.1111/j.1532-5415.1992.tb01992.x
72. First M.B.. *SCID-5-CV: Structured Clinical Interview for DSM-5 Disorders: Clinician Version* (2016)
73. Maier W., Buller R., Philipp M., Heuser I.. **The Hamilton Anxiety Scale: Reliability, validity and sensitivity to change in anxiety and depressive disorders**. *J. Affect. Disord.* (1988) **14** 61-68. DOI: 10.1016/0165-0327(88)90072-9
74. Maheshwari G., Shaukat F.. **Impact of Poor Sleep Quality on the Academic Performance of Medical Students**. *Cureus* (2019) **11**. DOI: 10.7759/cureus.4357
75. Tsai P.-S., Wang S.-Y., Wang M.-Y., Su C.-T., Yang T.-T., Huang C.-J., Fang S.-C.. **Psychometric Evaluation of the Chinese Version of the Pittsburgh Sleep Quality Index (CPSQI) in Primary Insomnia and Control Subjects**. *Qual. Life Res.* (2005) **14** 1943-1952. DOI: 10.1007/s11136-005-4346-x
76. Curcio G.G., Tempesta D., Scarlata S., Marzano C., Moroni F., Rossini P.M., Ferrara M., De Gennaro L.. **Validity of the Italian Version of the Pittsburgh Sleep Quality Index (PSQI)**. *Neurol. Sci.* (2013) **34** 511-519. DOI: 10.1007/s10072-012-1085-y
77. Bastien C.H., Vallieres A., Morin C.M.. **Validation of the Insomnia Severity Index as an outcome measure for insomnia research**. *Sleep Med.* (2001) **2** 297-307. DOI: 10.1016/S1389-9457(00)00065-4
78. Hossein-Nezhad A., Holick M.F.. **Vitamin D for Health: A Global Perspective**. *Mayo Clin. Proc.* (2013) **88** 720-755. DOI: 10.1016/j.mayocp.2013.05.011
79. Buhr E.D., Yoo S.-H., Takahashi J.S.. **Temperature as a Universal Resetting Cue for Mammalian Circadian Oscillators**. *Science* (2010) **330** 379-385. DOI: 10.1126/science.1195262
80. Gominak S., Stumpf W.. **The world epidemic of sleep disorders is linked to vitamin D deficiency**. *Med. Hypotheses* (2012) **79** 132-135. DOI: 10.1016/j.mehy.2012.03.031
81. Evatt M.L.. **Vitamin D associations and sleep physiology-promising rays of information**. *Sleep* (2015) **38** 171-172. DOI: 10.5665/sleep.4386
82. Massa J., Stone K.L., Wei E.K., Harrison S.L., Barrett-Connor E., Lane N.E., Paudel M., Redline S., Ancoli-Israel S., Orwoll E.. **Vitamin D and Actigraphic Sleep Outcomes in Older Community-Dwelling Men: The MrOS Sleep Study**. *Sleep* (2015) **38** 251-257. DOI: 10.5665/sleep.4408
83. Muscogiuri G., Barrea L., Scannapieco M., Di Somma C., Scacchi M., Aimaretti G., Savastano S., Colao A., Marzullo P.. **The lullaby of the sun: The role of vitamin D in sleep disturbance**. *Sleep Med.* (2019) **54** 262-265. DOI: 10.1016/j.sleep.2018.10.033
84. Muzur A., Pace-Schott E.F., Hobson J.. **The prefrontal cortex in sleep**. *Trends Cogn. Sci.* (2002) **6** 475-481. DOI: 10.1016/S1364-6613(02)01992-7
85. Morrell M.J., McRobbie D.W., Quest R.A., Cummin A.R.C., Ghiassi R., Corfield D.R.. **Changes in brain morphology associated with obstructive sleep apnea**. *Sleep Med.* (2003) **4** 451-454. DOI: 10.1016/S1389-9457(03)00159-X
86. Cameron H.A., Mckay R.D.. **Adult neurogenesis produces a large pool of new granule cells in the dentate gyrus**. *J. Comp. Neurol.* (2001) **435** 406-417. DOI: 10.1002/cne.1040
87. Stoffers D., Altena E., van der Werf Y.D., Sanz-Arigita E.J., Voorn T.A., Astill R.G., Strijers R.L.M., Waterman D., Van Someren E.J.W.. **The caudate: A key node in the neuronal network imbalance of insomnia?**. *Brain* (2014) **137** 610-620. DOI: 10.1093/brain/awt329
88. A Marks G., Roffwarg H.P., Shaffery J.P.. **Neuronal activity in the lateral geniculate nucleus associated with ponto-geniculo-occipital waves lacks lamina specificity**. *Brain Res.* (1999) **815** 21-28. DOI: 10.1016/S0006-8993(98)01008-7
89. Lima M.M.S., Andersen M.L., Reksidler A.B., Vital M.A.B.F., Tufik S.. **The Role of the Substantia Nigra Pars Compacta in Regulating Sleep Patterns in Rats**. *PLoS ONE* (2007) **2**. DOI: 10.1371/journal.pone.0000513
90. Kaneko I., Sabir M.S., Dussik C.M., Whitfield G.K., Karrys A., Hsieh J.-C., Haussler M.R., Meyer M.B., Pike J.W., Jurutka P.W.. **1,25-Dihydroxyvitamin D regulates expression of the tryptophan hydroxylase 2 and leptin genes: Implication for behavioral influences of vitamin D**. *FASEB J.* (2015) **29** 4023-4035. DOI: 10.1096/fj.14-269811
91. Maria S., Witt-Enderby P.. **Melatonin effects on bone: Potential use for the prevention and treatment for osteopenia, osteoporosis, and periodontal disease and for use in bone-grafting procedures**. *J. Pineal Res.* (2014) **56** 115-125. DOI: 10.1111/jpi.12116
92. Slominski R.M., Reiter R.J., Schlabritz-Loutsevitch N., Ostrom R.S., Slominski A.T.. **Melatonin membrane receptors in peripheral tissues: Distribution and functions**. *Mol. Cell. Endocrinol.* (2012) **351** 152-166. DOI: 10.1016/j.mce.2012.01.004
93. Sato-Mito N., Shibata S., Sasaki S., Sato K.. **Dietary intake is associated with human chronotype as assessed by both morningness–eveningness score and preferred midpoint of sleep in young Japanese women**. *Int. J. Food Sci. Nutr.* (2011) **62** 525-532. DOI: 10.3109/09637486.2011.560563
94. Grandner M.A., Jackson N., Gerstner J.R., Knutson K.L.. **Dietary nutrients associated with short and long sleep duration. Data from a nationally representative sample**. *Appetite* (2013) **64** 71-80. DOI: 10.1016/j.appet.2013.01.004
95. Zbozinek T.D., Rose R.D., Wolitzky-Taylor K.B., Sherbourne C., Sullivan G., Stein M.B., Roy-Byrne P.P., Craske M.G.. **Diagnostic overlap of generalized anxiety disorder and major depressive disorder in a primary care sample**. *Depress. Anxiety* (2012) **29** 1065-1071. DOI: 10.1002/da.22026
96. Martin E.I., Ressler K.J., Binder E., Nemeroff C.B.. **The Neurobiology of Anxiety Disorders: Brain Imaging, Genetics, and Psychoneuroendocrinology**. *Clin. Lab. Med.* (2010) **30** 865-891. DOI: 10.1016/j.cll.2010.07.006
97. Bouayed J., Rammal H., Soulimani R.. **Oxidative Stress and Anxiety: Relationship and Cellular Pathways**. *Oxid. Med. Cell. Longev.* (2009) **2** 63-67. DOI: 10.4161/oxim.2.2.7944
98. Hassan W., Silva C.B., Mohammadzai I.U., da Rocha J.B.T., Landeira-Fernandez J.. **Association of Oxidative Stress to the Genesis of Anxiety: Implications for Possible Therapeutic Interventions**. *Curr. Neuropharmacol.* (2014) **12** 120-139. DOI: 10.2174/1570159X11666131120232135
99. Huiberts L.M., Smolders K.C.. **Effects of vitamin D on mood and sleep in the healthy population: Interpretations from the serotonergic pathway**. *Sleep Med. Rev.* (2020) **55**. DOI: 10.1016/j.smrv.2020.101379
100. Pu D., Luo J., Wang Y., Ju B., Lv X., Fan P., He L.. **Prevalence of depression and anxiety in rheumatoid arthritis patients and their associations with serum vitamin D level**. *Clin. Rheumatol.* (2018) **37** 179-184. DOI: 10.1007/s10067-017-3874-4
101. Karonova T.L., Andreeva A.T., Beljaeva O.D., Bazhenova E.A., Globa P.J., Vasil’eva E.J., Grineva E.N.. **Anxiety/depressive disorders and vitamin D status**. *Z. Nevrol. Psikhiatrii Im. SS Korsakova* (2015) **115** 55. DOI: 10.17116/jnevro201511510255-58
102. Borges-Vieira J.G., Cardoso C.K.S.. **Efficacy of B-vitamins and vitamin D therapy in improving depressive and anxiety disorders: A systematic review of randomized controlled trials**. *Nutr. Neurosci.* (2022) **26** 187-207. DOI: 10.1080/1028415X.2022.2031494
103. Chen Y., Hu N., Yang J., Gao T.. **Prefrontal cortical circuits in anxiety and fear: An overview**. *Front. Med.* (2022) **16** 518-539. DOI: 10.1007/s11684-022-0941-2
104. Bombardi C., Grandis A., Pivac N., Sagud M., Lucas G., Chagraoui A., Lemaire-Mayo V., De Deurwaerdère P., Di Giovanni G.. **Serotonin modulation of hippocampal functions: From anatomy to neurotherapeutics**. *Prog. Brain Res.* (2021) **261** 83-158. DOI: 10.1016/bs.pbr.2021.01.031
105. Bombardi C., Di Giovanni G.. **Functional anatomy of 5-HT2A receptors in the amygdala and hippocampal complex: Relevance to memory functions**. *Exp. Brain Res.* (2013) **230** 427-439. DOI: 10.1007/s00221-013-3512-6
106. Karayol R., Medrihan L., Warner-Schmidt J.L., Fait B.W., Rao M.N., Holzner E.B., Greengard P., Heintz N., Schmidt E.F.. **Serotonin receptor 4 in the hippocampus modulates mood and anxiety**. *Mol. Psychiatry* (2021) **26** 2334-2349. DOI: 10.1038/s41380-020-00994-y
107. Tovote P., Fadok J.P., Lüthi A.. **Neuronal circuits for fear and anxiety**. *Nat. Rev. Neurosci.* (2015) **16** 317-331. DOI: 10.1038/nrn3945
108. Vogelzangs N., Beekman A.T.F., de Jonge P., Penninx B.W.J.H.. **Anxiety disorders and inflammation in a large adult cohort**. *Transl. Psychiatry* (2013) **3**. DOI: 10.1038/tp.2013.27
109. Naudé P.J.W., Roest A., Stein D., De Jonge P., Doornbos B.. **Anxiety disorders and CRP in a population cohort study with 54,326 participants: The LifeLines study**. *World J. Biol. Psychiatry* (2018) **19** 461-470. DOI: 10.1080/15622975.2018.1433325
110. Assa A., Vong L., Pinnell L.J., Avitzur N., Johnson-Henry K.C., Sherman P.M.. **Vitamin D Deficiency Promotes Epithelial Barrier Dysfunction and Intestinal Inflammation**. *J. Infect. Dis.* (2014) **210** 1296-1305. DOI: 10.1093/infdis/jiu235
111. Eid A., Khoja S., AlGhamdi S., Alsufiani H., Alzeben F., Alhejaili N., Tayeb H.O., Tarazi F.I.. **Vitamin D supplementation ameliorates severity of generalized anxiety disorder (GAD)**. *Metab. Brain Dis.* (2019) **34** 1781-1786. DOI: 10.1007/s11011-019-00486-1
|
---
title: Characteristics of Stress Sensitivity in Heroin Use Disorder Patients during
Their Opioid Agonist Treatment
authors:
- Filippo Della Rocca
- Angelo G. I. Maremmani
- Silvia Bacciardi
- Matteo Pacini
- Francesco Lamanna
- Beniamino Tripodi
- Mario Miccoli
- Icro Maremmani
journal: International Journal of Environmental Research and Public Health
year: 2023
pmcid: PMC10002439
doi: 10.3390/ijerph20054566
license: CC BY 4.0
---
# Characteristics of Stress Sensitivity in Heroin Use Disorder Patients during Their Opioid Agonist Treatment
## Abstract
In the present study, performed on a sample of Heroin Use Disorder (HUD) patients undergoing Opioid Agonist Treatment (OAT), we attempted to explore the relationships between stress sensitivity and heroin addiction-related clinical aspects. HUD patients’ stress sensitivity was evaluated with the Heroin/PTSD-Spectrum questionnaire (H/PSTD-S). The Drug Addiction History Questionnaire (DAH-Q), the Symptomatological Check List-90 (SCL-90), and The Behavioural Covariate of Heroin Craving inventory (CRAV-HERO) were all used, as were the Deltito Subjective Wellness Scale (D-SWS), a self-report scale evaluating subjective well-being; the Cocaine Problem Severity Index (CPSI), a questionnaire determining the extent of a cocaine problem; and the Marijuana Craving Questionnaire (MC-Q), an instrument assessing craving for cannabinoids. We checked correlations between stress sensitivity and the extent of HUD clinical features and compared patients with and without problematic stress sensitivity. H/PTSD-S was positively correlated with patients’ income, altered mental status, legal problems, the lifetime different treatments index, the current treatment load index, and all SCL-90 indexes and factors. Regarding subjective well-being, stress sensitivity negatively correlated with the contrast best week (last five years) index. Patients with high-stress sensitivity were females with a low income. They exhibited a more severe mental status at treatment entry, greater difficulty in working adaptation, and legal problems during treatment. Additionally, these patients showed a higher level of psychopathology, more impairment in well-being, and more risky behaviours during treatment. Stress sensitivity, as H/PTSD-S, must be considered an outcome of HUD. HUD’s addiction history and clinical features are significant risk factors for H/PTSD-S. Therefore, social and behavioural impairment in HUD patients could be considered the clinical expression of the H/PTSD spectrum. In summary, the long-term outcome of HUD is not represented by drug-taking behaviours. Rather, the inability to cope with the contingent environmental conditions is the key feature of such a disorder. H/PTSD-S, therefore, should be seen as a syndrome caused by an acquired inability (increased salience) concerning regular (daily) life events.
## 1.1. Towards a Patient-Tailored Therapy in Opioid Use Disorder
Substantial compelling evidence demonstrates that substance use per se is still a worldwide issue threatening individuals and communities. In vulnerable people, continuous use of rewarding substances can enduringly modify brain functions. The reward circuit’s responsiveness to reward and motivation, which are not drug-related, is decreased; sensitivity of the emotional circuits to stress is enhanced; and self-regulation functions are impaired. Such modifications may, in turn, induce dysfunctional behavioural changes, such as dysfunctional loss of control in drug seeking and drug use. Moreover, these brain changes can persist throughout an individual’s life. As a result, craving—the intense, urgent, and spontaneous desire for the rewarding substance—and relapse can last decades after clinical remission and abstinence [1,2]. In this context, addiction must be considered the end stage of such a pathological process. Namely, addiction is a chronic and relapsing brain disease characterised by impairing drug-induced neural modifications [3,4], nowadays widely recognised as a chronic and relapsing brain disease. On a clinical background, it has been observed that Heroin Use Disorder (HUD) patients, during long-term treatment, tend to show an impaired capacity to experience pleasure—anhedonia—and a more severe stress reaction to life events that may interfere with the rehabilitative program [5,6]. From a therapeutic point of view, managing addiction—particularly (HUD)—might be challenging for clinicians, as it implies extensive knowledge in managing long-term opioid medications and rehabilitation programs [7]. Opioid Agonist Treatment (OAT) has proved to be an effective intervention for HUD patients, as it allows a more effective reduction in heroin use than treatments that do not involve opioid medications [8]. However, neither biological nor clinical correlates have been standardised for treatment monitoring and outcome during OAT, so clinicians still set treatment strategies according to ‘good clinical practice’ (considering, for instance, clinical presentation, comorbidities, and urinalyses).
Moreover, retention in treatment is still the primary goal of OAT. So far, harm reduction is the most applied treatment strategy in dealing with HUD in professional medical treatment services worldwide [9,10]. On this clinical background, information from patients’ specific psychopathology, craving behavioural covariates, stress sensitivity, and individuals’ subjective well-being pointed to novel clinical information monitoring patients under OAT. It might be the key to better personalising diagnostic and therapeutic interventions, setting the basis for moving from harm reduction to patient-tailored therapy in OAT.
## 1.2. Psychopathology Specific to Substance Use Disorder
The diagnosis of Substance Use Disorder (SUD) is nowadays based on the presence of specifically identified behavioural symptoms [11]. Unlike most mental disorders, psychopathological syndromes occurring in SUD patients do not find space within the diagnostic criteria of SUD [11]. In parallel, any psychopathological signs and symptoms occurring in such patients have historically been confined to the “comorbidity” framework, as if psychopathology belonged exclusively to comorbid psychiatric disorders or underlying personality traits and not SUD [12,13,14,15].
Nevertheless, increasing scientific evidence has highlighted the clinical inadequacy of the diagnostic model proposed by the current reference manuals. In particular, recent epidemiological, genetic, neurobiological, and neuropsychological findings do not appear to correctly support DSM categories and definitions [3,16,17,18]. Therefore, many investigators have called for greater integration between such findings and the different diagnostic criteria of psychiatric illness. Specifically, it has been proposed that the psychopathological syndromes observed in psychiatric patients refer directly to the neurobiological substrates of behavioural patterns, including the addictive behaviours of SUD individuals [17,18]. In this context, it has been widely observed that psychopathological signs and symptoms—such as novelty seeking, irritability, restlessness, impulsivity, diminished interest in activities, dysphoria, boredom, depression, and attention and concentration difficulties—usually accompany addictive behaviours, thus contributing to the complete picture of the psychopathological profile of SUD patients and the persistence of their substance use [5,6,19,20].
Under those circumstances, in recent years, some Italian studies—mainly performed by the V.P. Dole research group at the Santa Chiara University Hospital in Pisa, Italy—have advanced the main idea that all psychiatric symptoms and clusters displayed in individuals with SUD should not be evaluated merely as manifestations of a generic psychiatric “comorbidity” [20]. Moreover, they further suggested the idea that such psychopathological symptoms—which mainly belong to the domains of anxiety, mood, and impulse control [19]—should instead be attributed directly, and by their very nature, to the addiction process, thus being a core clinical manifestation of SUD itself. Consequently, by advancing the hypothesis that addiction may have a specific psychopathology, such investigations have also explored some possible clinical implications, clarifying how information on patients’ psychopathological symptoms could be helpful in treatment choice [21] or outcome [22].
In particular, using an exploratory principal component factor analysis of the Self-Report Symptom Inventory (SCL-90) questionnaire—which is a self-report questionnaire used to measure psychopathological symptoms [23], widely used in the field of substance use [24,25,26]—a five-factor solution was studied in a sample of more than 2500 HUD individuals at treatment entry. These analyses led to the discovery of the five main domains: [1] the “worthlessness/being trapped (W/BT)” dimension that assembles obsessive-compulsive, depressive, and psychotic symptoms; [2] “somatic symptoms (SS)”, which is characterised by several somatic and anxious features, and resembles opioid withdrawal; [3] “sensitivity-psychoticism (S/P)” features, such as psychoticism and sensitivity; [4] “panic-anxiety (PA)”, which can be described as a fear of travelling by train or bus, going around alone, sensations of dizziness or fear of feeling sick, and the experience of acute anxiety; and [5] “violence-suicide (V/S)”, comprising aggressiveness against others and self-directed aggressiveness, with anger, rage, and breaking things up being fundamental elements of this domain. These five psychopathological syndromes are stable, specific to SUD [27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48], and are useful clinical features during the OAT [21,22].
## 1.3. Stress Sensitivity
The role of stress in addiction processes has been studied for decades. Many of the leading theories of addiction converge in recognised stress as a key factor in increasing vulnerability to addiction. On the one hand, several theoretical approaches emerging from psychological research consider substance use behaviour as a coping strategy to deal with stress—i.e., blunting inner tension, self-medicating, and reducing withdrawal-related distress. Over time, many coping paradigms have attempted to describe how stress is involved in substance use initiation, ongoing addiction processes, and relapse [49,50,51]. To explain, according to Chaplin’s ‘two-pathway model’ of substance use development, adolescents with a history of dysregulated stress reactivity may cope with increased environmental and internal stressors by using substances to either down-regulate overly high-stress reactivity or to up-regulate blunted stress reactivity [52]. This fact is consistent with the self-medication hypothesis of addiction [53] and stress-coping theories of adolescent substance use [54].
On the other side, stress theories of addiction—based on neurobiological models—have been proposed to explain how neuroadaptations in reward, learning, and stress pathways may enhance the key features of the addiction process, such as craving, impaired executive functions, and the inability to stop recurrent substance seeking behaviour and use despite adverse consequences [55,56].
Regardless of the SCL-90 psychopathological dimensions, stress sensitivity appears to be a cross-sectional key element in the psychic structure of SUD patients [55,57,58,59,60,61,62,63,64,65,66,67]. *In* general, the relationships between stress sensitivity and mental disorders have mainly been studied in recent years, both on biological and clinical grounds. However, much less is known about the relationship between stress sensitivity and SUD.
In summary, research on the relationship between stress and addiction mainly focused on stress sensitivity and its role in the developmental psychopathology of adolescent substance use and addiction. This fact is further true considering the known clinical implications of stress-related disorders—such as Post-Traumatic Stress Disorder (PTSD)—on the development of SUD [55,68,69].
## 1.3.1. SUD as a Risk Factor in Increasing Individual Susceptibility to PTSD
Many neurobiological studies showed how substance use and its related acute and chronic behaviours could profoundly affect stress regulation. From a regular perspective, substance use increases stress response, even in individuals who do not exhibit comorbid conditions [60]. Furthermore, studies on the activity of the Hypothalamic-Pituitary-Adrenal (HPA) axis—which is widely considered a fundamental player in response to potential and actual stressors, controlling sympathetic, hormonal, and behavioural responses to stress [70]—revealed that baseline activity of the HPA-axis, specifically plasma ACTH and cortisol levels, is increased in individuals with SUD compared to healthy controls [25,71,72]. These findings suggest that SUD patients are even more vulnerable to the effects of stress, thus suggesting a higher stress sensitivity in such individuals than in non-addicted ones.
As stated above, much of the work exploring the relationship between substance use and stress has focused on the influence of chronic stress and trauma exposure on the likelihood of an individual developing SUD [55,68,69]. Nevertheless, we acknowledge that substance use itself may alter individuals’ stress sensitivity, changing how stress sensitivity contributes to the further development of SUDs over time [52,60]. Accordingly, a substance-abusing lifestyle might predispose substance users to experience stressful or traumatic events [68,73,74]. Moreover, data are available in the literature showing that drug-addicted subjects become more sensitive to stress the longer their history of drug addiction [55,62,75,76,77,78,79]. Along with the observation that stress sensitivity appears to be a cross-sectional key element in the psychic structure of SUD patients [55,57,58,59,60,61,62,63,64,65,66,67], in the present paper, we have attempted to address the topic of researching the relationship between stress and substance use from a different perspective. Specifically, the onset of substance use and alcohol misuse may precede the development of stress-related disorders—namely, the entire PTSD syndrome—in the clinical history, possibly contributing to increased individual susceptibility to them. In this case, a history of adverse and stressful experiences could have primarily induced substance use behaviours. In contrast, the neurobiological derangement caused by stress and psychotropic substances would be later responsible for the development of PTSD.
## 1.3.2. Assessment of PTSD Spectrum in HUD Patients: Developing an Instrument to Evaluate the PTSD Spectrum
Several epidemiological data show frequent associations between PTSD and SUD. With this regard, more than $30\%$ of patients suffering from SUD were found to meet the criteria for current PTSD, and half of them for lifetime PTSD [55,68,69]. Moreover, the co-occurrence of SUD-PTSD is associated with more severe psychopathology, more severe addiction history, worse executive functions, higher rates of overdoses, attempted suicide, and poorer treatment outcomes than those with PTSD or SUD alone [80,81,82]. Therefore, it appears clear that adequately screening for PTSD might be a crucial step in creating a patient-tailored therapy for SUD patients. Thus, assessing PTSD features must be considered clinically required during a daily routine. The need to develop a reliable tool to evaluate PTSD characteristics in SUD patients led to a first conceptual problem. In this regard, it is worth mentioning that an undefined area of the interactions between stress, traumatic experiences, and substance use was observed in many of these cases without any possibility of recognising a well-defined sequence in the clinical history and related neurobiological changes. Many patients also do not fulfil the Diagnostic and Statistical Manual of Mental Disorders (DSM) criteria for PTSD diagnosis, even though they show clear evidence of stressful and traumatic implications in developing substance use behaviours.
Moving from the role mentioned above of stress sensitivity in mediating the relationship between SUD and PTSD, the research group V.P. Dole of the Santa Chiara University Hospital of Pisa, Italy, has attempted, in recent years, to address this issue through the study of stress-reactivity indices in patients with HUD. They came to conceptualise a ‘PTSD-Spectrum’, based upon a concept that uses individual DSM criterion symptoms as a starting point and extends the DSM description to encompass the halo of surrounding clinical phenomena—following a similar conceptualisation for other psychiatric disorders, such as Bipolar Disorder [83]. These include associated features described in the DSM as well as symptoms, maladaptive behavioural traits, and temperamental characteristics that do not appear in the DSM. However, such features seemed to be based on real-world settings, providing the diagnostic categories with a useful and more complete characterisation of psychopathological dimensions.
To assess the PTSD-Spectrum in patients with HUD, the V.P. Dole research group developed the ‘Reaction to loss and traumatic life events—Heroin/PTSD-Spectrum’ (H/PTSD-S) inventory) [84,85], derived from the ‘Trauma and Loss Spectrum-Self-Report, Instrument Questionnaire, Lifetime version’ (TALS-SR) [86,87]. Such an instrument showed the ability to adequately explore the lifetime experience of a range of loss or traumatic events comprising lifetime symptoms, behaviours, and personal characteristics, which might represent manifestations and risk factors for the development of a stress response syndrome, thus allowing the study of patients’ stress sensitivity.
The aims were to continue the process of validation of the H/PTSD-S inventory [84,88], and, in the present study, we assessed the correlations between the H/PTSD-S inventory with heroin addiction-related clinical aspects (addiction history, comorbid substance use, severity of addictive behaviours, and subjective wellness) and the severity of the psychopathological symptoms. In this way, it is possible to define HUD patients’ stress sensitivity better.
## 2.1. Design of the Study
The data were collected using the HUDPyscho-study, an ongoing cross-sectional cohort study from Pisa, Italy’s Drug Addiction Service (SerD). This study is a naturalistic, single-centre, observational, and non-interventional cohort study applying the usual procedure consistent with daily clinical practice. This study allows researchers to examine clinical and psychopathological features, stress sensitivity, and the quality of life in HUD patients during the OAT through a single administration of questionnaires during treatment. The University of Pisa Ethical Committee has previously approved the HUDPyscho-study (ID 22656, CEAVNO 21.07.2022).
## 2.2. Sample
The study’s participants were recruited from the Drug Addiction Service (SerD) in Pisa, Italy. It is a local public unit for drug addiction located in the north-western part of central Italy (province of Pisa, Tuscany region). Persons referred to the service were mainly residing in the same area. We did not use specific criteria for eligibility other than the “wish to be treated” and “wanting to participate” in the survey.
Patients who were at least 18 years and diagnosed with Opioid Use Disorder (OUD)—according to the DSM-5 criteria [11]—and in treatment with agonists of opioid medications were included in the study. Patients who could not provide written informed consent to study participation were excluded. The clinical assessment of each participant has been performed through a single administration of questionnaires (around 2 h), both in a self-report and in a clinician-report modality. Both modalities were temporarily synchronised so that all clinical data could be considered a function of the patient’s time in OAT.
Data were obtained on all patients who voluntarily agreed to be enrolled in the present study at any time between June 2022 and October 2022; to avoid discrepancies, they were, without exception, evaluated by the same physician on duty (FDR).
In the Drug Addiction service of Pisa, an OAT, according to the Dole and Nyswander (D&N) treatment methodology, has been used since its foundation [89,90].
46 HUD patients in OAT—referring to the Drug Addiction service of Pisa—were asked to participate in the present study. Two patients declined to be enrolled: one was suspicious, possibly due to psychotic symptoms, and one did not explain any reason. The whole enrolled sample consisted of 44 patients, with a mean age of 42.14 ± 11.0 years (18–62 ranging), and 33 ($75.0\%$) patients were males while 11 ($25.0\%$) were females.
## 2.3. Instruments
The assessment of patients was achieved with the following questionnaires:Drug Addiction History Questionnaire (DAH-Q) for the standard demographic and drug history data collection. Symptomatological Check List-90 (SCL-90) is a list of psychopathology symptoms. Heroin Craving Behavioural covariate (CRAV–HERO) is an inventory of heroin addictive behaviour. Heroin/PTSD-Spectrum questionnaire (H/PSTD-S), a new instrument developed to evaluate stress-sensitivity in HUD patients. Deltito Subjective Wellness Scale (D-SWS) is a self-report measure evaluating the quality of life. Cocaine Problem Severity Index (CPSI) is a questionnaire determining the extent of a cocaine problem. The Marijuana Craving Questionnaire (MC-Q) assesses the craving for cannabinoids.
## 2.3.1. Drug Addiction History Questionnaire (DAH-Q)
The DAH-Q is a comprehensive questionnaire regarding heroin addiction history [91]. It takes the form of a semi-structured interview comprising a multidimensional questionnaire. Specifically, seven areas are enclosed: I–Somatic pathology (8 items, Cronbach’s alpha = 0.60); II–*Mental status* (12 items, Cronbach’s alpha = 0.78); III–Social adjustment (5 items, Cronbach’s alpha = 0.63); IV–Co-occurring substance use (7 items, Cronbach’s alpha = 0.72); V–Modality of use (5 items, Cronbach’s alpha = 0.53); VI–Past and Current Treatment History (8 items, Cronbach’s alpha = 0.61); and VII–Longitudinal addiction-history aspects (3 items, Cronbach’s alpha = 0.71). The III area is composed of (a) work, (b) family, (c) intimacy, (d) social/leisure, and (e) legal problems. The design of its related questionnaires is based on structures requiring dichotomous (presence/absence) answers. For each area, the score is reported as a ratio—based on the number of aspects present in that specific area and in relation to the maximum number of items listed for the same area. A ratio = 1 represents the presence of all investigated aspects referring to that area.
## 2.3.2. Symptomatological Check List-90 (SCL-90)
According to the methodology of Maremmani et al. [ 92], the Symptomatological Check List-90 (SCL-90) is a self-report rating scale evaluating outpatients’ psychiatric and symptomatic behaviours. It consists of 90 items, with five levels of severity (ranging from a minimum of ‘Not at all’ to ‘Extremely severe’), evaluating psychopathological severity across nine dimensions, including internalising and externalising symptoms. It was initially developed by Derogatis et al. [ 23]. By including HUD for the first time, the 90 items have been rearranged by the V.P. Dole research group into five main dimensions, which are viewed as the background of a large majority of symptomatic behaviours observed in individuals with such a disorder. The primary symptomatologic dimensions are [1] Worthlessness/Being Trapped (W/BT), [2] Somatic Symptoms (SS), [3] Sensitivity/Psychoticism (S/P), [4] Panic Anxiety (PA), and [5] Violence/Suicide (V/S). These five main domains have been primarily validated in over 2500 substance-use disorder individuals. According to the highest Z-score for each factor, participating subjects were distributed into five samples (called “dominant groups”). The largest group of patients was the dominant group distinguished by somatic symptoms ($24\%$). The second largest group was the ‘panic anxiety’ one ($22\%$), followed by the violence/suicide group ($20\%$), the sensitivity/psychoticism group ($20\%$), and, lastly, the worthlessness/being trapped group ($14\%$). Each of these five dimensions was independent of the others and showed no significant overlap [92]. Following this enquiry, several further studies were conducted to ascertain whether this psychopathological structure might be considered a stable trait or a variable state that potential confounding factors might condition [27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48]. Nowadays, the SCL-90 questionnaire might be regarded as a reliable instrument to assess psychopathology specific to SUD clinically.
## 2.3.3. Heroin Craving Behavioural Covariate (CRAV-HERO)
The Heroin Craving Behavioural Covariate chart (CRAV-Hero) allows researchers to record both the presence and severity of addictive behaviours, investigating the craving of HUD individuals. Using focus groups and brainstorming methodology, an expert panel belonging to the V.P. Dole Research Group at the Dual Disorder Unit of the Santa Chiara University Hospital in Pisa, Italy, selected items to be included in the inventory. Focus groups were open to rehabilitated HUD patients. In these ways, the possibility that specific craving behaviours accurately reflected what they were intended to reflect was stressed so that the results could be applied to real-world settings (content or construct validity of the inventory). Thirteen behaviours were selected. For each of these, the five available answers were related to five different levels of severity. Out of each set of five solutions proposed for every question, only one could be picked; the answers were presented in an order corresponding to a stepwise increase in the level of craving. In this way, a score (from 0 to 4) was attributed to each question, and by adding up the results, the total score was in line with craving intensity. Because the overall damage done by craving should be estimated in terms of the total costs (going beyond a merely economic level) that a subject is willing to pay to obtain the drug, the indirect questions are focused on four main themes: [1] exchange, [2] time, [3] risk, and [4] cue-induced/environmental stimuli to use. The CRAV-HERO has been validated in Italian HUD patients [93].
## 2.3.4. Deltito Subjective Wellness Scale (D-SWS)
The Deltito Subjective Wellness Scale (D-SWS) looks at three periods in assessing the patients’ well-being. Time one is defined as ‘now and characteristic of the past week’; time two is defined as ‘during the worst week of your current or most recent depressive episode—within the last year’; and time three is defined as ‘the best week you have had in the previous five years’. The D-SWS is intended to be directly filled out by the patient, following an explanation by the clinician to verify whether the patient understands. It explores patients’ self-rated quality of life, including, among others, hobbies, sexual life, and occupational features. Each HUD participant can choose among ‘0’ (No Way Whatsoever), ‘1’ (*To a* Minor Extent), ‘2’ (*To a* Major Extent), and ‘3’ (To the Highest Extent Possible). D-SWS evaluates the patient’s quality of life based on their subjective feelings, so it is free from clinical judgments. The assessment takes place through a temporal dimension, providing information about the disorder’s progression over time, the subjective functional impact on everyday life and the treatment outcome.
## 2.3.5. Cocaine Problem Severity Index (CPSI)
The Cocaine Problem Severity Index (CPSI) may help determine the extent of a cocaine problem. The CPSI was initially developed by Rawson et al. in 1989 [94] for the clinical assessment of cocaine use. Using the CPSI questionnaire, it is possible to assign each cocaine user to a four-subgroup set, namely ‘experimental/recreational use’, ‘cocaine abuse with a significant problem’, ‘cocaine dependence requiring assistance’, and ‘severe dependence’. Even though they might be using cocaine only in a social context, patients showing an ‘experimental/recreational use’ (E/R) are prone to increase their use if the stress increases or the substance becomes more accessible. In patients belonging to the ‘cocaine abuse with a significant problem’ (CASP) group, the effects of cocaine on their life tend to be substantial. These patients are recommended to obtain a professional evaluation to help them determine the most effective way to deal with problems related to cocaine use. In patients with ‘cocaine dependence requiring assistance’ (CDRA), cocaine use is a severe problem, and they need to seek assistance to learn what addiction is and how to deal with it. Lastly, it might be difficult for patients in the ‘severe dependence’ (SD) group to regain control of their life without hospitalisation. The CPSI questionnaire contained 18 items, each requiring multiple answers corresponding to various levels of severity from 0 to 8; thus, the maximum craving score was 86 [94].
## 2.3.6. Marijuana Craving Questionnaire (MC-Q)
The Marijuana Craving Questionnaire (MCQ) is a multidimensional questionnaire developed by Heishman et al. to assess marijuana cravings [95]. The development of the MCQ was based on the model of the ‘Questionnaire on Smoking Urges’ [96] and the ‘Cocaine Craving Questionnaire’ [97]. Items on the MCQ were drawn from five theoretical conceptualisations of craving. The proposed five categories were: [1] desire to use marijuana, [2] anticipation of positive outcomes from marijuana use, [3] anticipation of relief from withdrawal symptoms or negative mood, [4] intention and planning to use marijuana, and [5] lack of control over marijuana use.
## 2.3.7. Stress Sensitivity—Heroin/PTSD-Spectrum (H/PTSD-S)
The H/PTSD-S inventory has been previously proposed to explore the PTSD spectrum in HUD patients. It is a short self-report questionnaire comprising the 30 most significant items retained from the TALS-SR form. To assess the PTSD spectrum, the V.P. Dole research group used the ‘Trauma and Loss Spectrum-Self-Report, Instrument Questionnaire, Lifetime version’ (TALS-SR) [86,87]. TALS-SR includes 116 items exploring the lifetime experience of a range of loss or traumatic events comprising lifetime symptoms, behaviours, and personal characteristics that might represent manifestations and risk factors for developing a stress response syndrome. The validity of TALS-SR is well established [86,87,98], but some limitations in using it with HUD patients should be mentioned. Specifically, when addicted patients approach treatment settings, they typically make a spontaneous request for help, revealing a different motivational status and a constant ambivalence towards compliance with treatment, which they may be aware of to a certain degree. Their willingness to be clinically assessed before starting treatment is very low.
Moreover, their ability to adequately maintain their attention to fill in a rating scale is minimal at treatment entry—and during the various treatment phases. So, the time allocated to rating scale evaluation must be limited. The TALS-SR questionnaire needs a consistent amount of time to be completed. Thus, a shorter TALS-SR form was required to assess HUD patients. Items from TALS were selected to obtain a reduced form for HUD patients, making it possible to differentiate patients with and without a PTSD spectrum comparable with the one developed by the survivors of the L’Aquila (Italy) 2009 earthquake [98]. Eventually, the V.P. Dole research group developed the ‘Reaction to loss and traumatic life events—Heroin/PTSD-Spectrum’ (H/PTSD-S) inventory, which consists of 30 dichotomic items. The number of items in the TALS questionnaire was reduced using V.P. Dole’s TALS database, in which 235 questionnaires previously administered to a sample of ‘typical respondents’—evaluated on entering or during an OAT in previous research protocols were stored. Patients’ questionnaires were divided into two groups, indicating or not indicating the presence of H/PTSD-S, according to the methodology described in Dell’Osso et al. [ 57]. Logistic regression analysis for each TALS domain was then performed, using the H/PTSD-S presence as a criterion and domain items as predictors.
Only cases with $p \leq 0.05$ and OR above 3.0 were retained. ROC analysis determined a cut-off value and distinguished the H/PTSD-S presence/absence. The cut-off’s acceptable percentage of sensitivity and specificity was set to $80\%$. The confidence level of the Area Under the Curve (AUC) was $95\%$, and AUC was considered statistically significant with a p-value < 0.05. It was assumed that the 25–$75\%$ interval would be adequate in testing the discriminative effect of the selected items in our reduced TALS inventory, as it would minimise the floor effect. Internal consistency (reliability) was estimated by applying Cronbach’s alpha test. This type of spectrum was renamed Heroin/PTSD-Spectrum (H/PSTD-S) [84,85]. The cut-off value determined by the ROC analysis was 11. All the items demonstrated adequate variability. The internal consistency (reliability) estimated using Cronbach’s alpha was optimal (0.88). The proposed H/PTSD-S inventory, founded on achieving satisfactory internal consistency, measures the stress reactivity construct.
## 2.4. Data Analysis
The Shapiro–Wilk test was performed to verify the normality of distributions. Spearman’s correlation coefficient analysed correlations that concern quantitative and ordinal variables. Comparisons between groups were analysed using Fischer’s exact test for categorical variables and the Mann–Whitney U test for ordinal variables. Missing data were excluded from the analyses. As this is an exploratory study, statistical significance was considered for p ≤ 0.05.
## 3.1. Demographic and Clinical Characteristics
Most of the participants were single ($68.2\%$), highly educated—(>eight years of education) ($52.3\%$), unemployed ($55.3\%$), from a white-collar parental family ($62.2\%$), receiving adequate income ($75.0\%$), and living alone ($55.8\%$). The number of patients who reported no lifetime somatic pathology was $15\%$. On average, the sample presented $\frac{2.24}{8}$ lifetime somatic pathologies and $\frac{3.84}{12}$ aspects of altered mental status. Regarding the current social adjustment, most of the participants were satisfied with their household situation ($53.3\%$), intimacy ($57.7\%$), and social/leisure activity ($63.6\%$). Furthermore, most of them had no legal problems ($54.8\%$). Regarding lifetime cooccurring substance use—other than heroin—the lifetime use of CNS-Depressants was reported in $73.3\%$ of cases, the use of CNS-Stimulants in $82.9\%$, the use of Hallucinogens in $37.0\%$, the problematic use of alcohol in $59.4\%$, the use of cannabis in $88.6\%$, and, eventually, the use of inhalants in $4.0\%$. On average, the sample presented $\frac{4.20}{6}$ cooccurring substances. Regarding the modality of heroin use, the sample was characterised by a ‘multiday’ heroin intake modality ($90.9\%$) and an unstable modality of use—they used to lead an existence that is not likely to be accepted to social conventions (they are considered either ‘fanatic’, ‘loners’, or ‘violent’) ($74.2\%$). Moreover, most participants were in stage 3 of their heroin use disorder, the ‘revolving door’ stage ($96.8\%$), and reported a clinical typology related to psycho-social antecedents—that is, these patients are mostly highly responsive to personal and environmental stimuli or issues ($91.2\%$). On average, the sample presented $\frac{4.95}{12}$-lifetime different treatments and $\frac{2.76}{4}$ different current treatments. On the whole, the mean problematic areas load was $\frac{4.4}{10.}$
The age of first contact with the substance of primary use—heroin—was 20.71 ± 5.9 years; the age of continuous use was 28.78 ± 9.4 years; andthe dependence length was 152.75 ± 108.5 months. The age at the first treatment was 30.93 ± 8.6 years, and the current treatment length was 27.54 ± 44.0 months. The time between the age at first contact and continuous use was 6.54 ± 8.5 years; between first use and first treatment was 9.07 ± 8.5 years; and between age at ongoing use and first treatment was 2.48 ± 3.0 years.
Furthermore, participants were more likely to be diagnosed with Bipolar Spectrum ($29.5\%$) than with Recurrent Depression ($6.8\%$), Anxiety Disorders ($4.5\%$), and Chronic Psychosis ($2.3\%$). Most were not reported to display a Dual Disorder ($56.8\%$).
Eventually, $63.6\%$ of patients were undergoing a Methadone Maintenance Treatment (with a mean dosage of 81.96 ± 42.1 mg/day), $15.9\%$ were on a Buprenorphine Maintenance Treatment (11.14 ± 7.5 mg/day), and $20.5\%$ with other medications.
Most of the sample reported the presence of a Heroin PTSD Spectrum ($$n = 34$$, $77.3\%$).
## 3.2. Bivariate Correlation between the Severity of Stress Sensitivity, Length of Current Treatment, and Other Clinical Aspects
Table 1 shows significant correlations between stress sensitivity, current treatment duration (months), and other clinical aspects. The Total PTSD Spectrum Score represents stress sensitivity.
Regarding demographic and drug addiction history data, stress sensitivity positively correlated with participants’ income, altered mental status, legal problems, lifetime different treatments index, and current treatment load index.
Regarding the SCL-90 psychopathological indices, stress sensitivity showed a positive correlation with each of the variables, namely, with the global score index, the positive symptom index, the total SCL score, the positive symptom distress (severity), and the severity of each SCL-90 factor—W/BT, SS, S/P, PA and V/S dimension.
Regarding indices of quality of life retrieved from the D-SWS, the severity of stress sensitivity showed a high negative correlation with the contrast best week (last year) index.
No correlation was found between stress sensitivity and behavioural covariates of heroin (rho = 0.25; $$p \leq 103$$) and cocaine (rho = 0.26; $$p \leq 0.087$$) and between stress sensitivity and cannabis craving (rho = −0.04; $$p \leq 0.783$$).
## 3.3. Differences between Patients with and without Heroin/Post-TrauPosttraumaticisorder—Spectrum (H/PTSD-S)
Table 2 shows demographic and clinical differences between patients with ($$n = 34$$) and without ($$n = 10$$) Heroin PTSD-Spectrum.
Concerning demographic data, the two groups differed in gender and income. Specifically, patients without H/PTSD-S were all males and with adequate income. No significant differences were found regarding age ($U = 155.50$; z = −0.40; $$p \leq 0.689$$), marital status ($$p \leq 1.000$$), education ($$p \leq 1.000$$), occupation (χ2 = 1.88; $$p \leq 0.389$$), parental family working activity (χ2 = 1.85; $$p \leq 0.396$$), and living situation ($$p \leq 1.000$$).
Regarding the heroin addiction history, a significant difference was observed between the two groups regarding the problematic area ‘Mental Status’, with the group with H/PTSD-S reporting a higher score than the group without H/PTSD-S. Another difference was observed regarding the presence of an occupation, which was higher in patients without H/PTSD-S, and also in legal problems, which were higher in patients with H/PTSD-S.
No differences were observed regarding heroin intake ($$p \leq 0.523$$), modality of use ($$p \leq 0.335$$), HUD stages ($$p \leq 1.00$$), clinical typology ($$p \leq 1.00$$), diagnoses (χ2 = 8.65; $$p \leq 0.070$$), presence of dual disorder ($$p \leq 0.474$$), use of CNS depressants ($$p \leq 0.060$$), CNS stimulants ($$p \leq 0.117$$), hallucinogens ($$p \leq 0.666$$), age at heroin first contact ($U = 88.00$; $z = 0.19$; $$p \leq 0.872$$), age at heroin continuous use ($U = 69.50$; $z = 0.91$; $$p \leq 0.377$$), dependence length ($U = 43.50$; z = −0.84; $$p \leq 0.413$$), age at first treatment ($U = 82.00$; $z = 0.51$; $$p \leq 0.631$$), present treatment length ($U = 134.50$; $z = 1.04$; $$p \leq 0.304$$), the time between age at first contact and age at continuous use latency ($U = 55.00$; $z = 0.16$; $$p \leq 0.900$$), the time between age at first use and age at first treatment latency ($U = 81.50$; $z = 0.87$; $$p \leq 0.395$$), the time between age at continuous use and age at first treatment latency ($U = 38.00$; $z = 0.74$; $$p \leq 0.514$$), other problematic areas (somatic complications ($U = 87.00$; z = −0.39; $$p \leq 0.717$$), substance use ($U = 68.50$; z = −1.57; $$p \leq 0.122$$), lifetime different treatments ($U = 61.50$; z = −1.73; $$p \leq 0.084$$), current treatment load ($U = 72.00$; z = −1.59; $$p \leq 0.166$$), and the total problematic areas score ($U = 71.00$; z = −1.45; $$p \leq 0.154$$).
Regarding the SCL-90 psychopathological indices, the two groups showed significant differences concerning the global score index, the positive symptom index score, and the positive symptoms distress. The two groups also differed concerning W/BT, SS, SP, and V/S psychopathological dimensions. No difference was found in comparing the five prominent psychopathological typologies (χ2 = 4.65; $$p \leq 0.324$$). In all these variables, patients with H/PTSD-S reported higher scores.
Regarding the D-SWS indices, a highly significant difference between the two groups was found regarding the ‘contrast’ with the poor week in the last year and the ‘contrast’ with the best week in the previous five years. Specifically, patients without H/PTSD-S reported much improved subjective well-being compared to the previous year’s poor week and the ‘contrast’ with the best week in the last five years.
Eventually, no significant difference was observed between groups regarding the heroin addictive behaviour covariate—CRAV-ERO total score ($U = 122.50$; z = −1.61; $$p \leq 0.186$$)—cocaine problem severity index ($U = 113.50$; z = −1.60; $$p \leq 0.141$$), and cannabis craving severity ($U = 214.00$; $z = 1.52$; $$p \leq 0.227$$).
## 4.1. Demographic and Clinical Characteristics
The sample in the present study was predominantly male, and most participants reported being single, highly educated, and unemployed, yet overall they reported adequate income and living alone. These demographic characteristics are typical of the Italian drug addict population, where poverty is relatively rare. Furthermore, most participants in the present study did not report any legal problems. Regardless of the treatment duration and what might be expected for a sample retrieved from a real-world setting, this demographic characteristic is consistent with the literature regarding expectations of excellent clinical practice during OAT. Therefore, this high percentage of patients not reporting legal problems should be interpreted as a virtuous consequence of OAT. To explain, according to the current literature, the main objectives of long-term management of patients suffering from Opioid Use Disorder (OUD) include the reduction of the risk of death and disease—a basic goal, which is better known as ‘harm reduction’—improved mental health and outlook, and restoration of compromised social role due to issues such as unemployment, disrupted family relationships, and involvement with the criminal justice system [99,100].
Nevertheless, a very high percentage of participants reported an actual history of ‘addiction’ (i.e., having experienced a ‘revolving door’ life or repeated detoxifications). On the one hand, the high proportion of patients classified as stage 3 in their natural history of HUD at treatment entry should not be surprising, as HUD patients are more likely to get in treatment—and to retain in treatment—in a later stage in the course of their disorder than in any earlier stage (i.e., the ‘honeymoon’ stage and the intermediate or ‘dose-increasing’ stage [101]. A “revolving door” situation is characterised by a dramatic, unfolding sequence of events, such as being treated, dropping out of treatment, having an argument, being arrested, being admitted to the hospital, returning to treatment, and so on. This phase is of fundamental clinical importance, as the risk of death from “overdose” is higher than in the previous steps. This fact is due to detoxification, which should be viewed as a gradual decline in opioid tolerance.
Along with this, the onset of craving for the substance tends to push the subject to occasional heroin use. At this stage, taking a dose of heroin equal to the quantity taken during the’ tolerance period’ is at high risk of causing an “overdose” [102]. Furthermore, such a clinical characteristic should be considered a stable feature of the disorder, being more related to the patient’s drug addiction history than any current condition at their treatment entry, including ongoing treatment. Accordingly, in the present study, most participants had undergone several previous treatment programmes at the time of the evaluation—this also comes from data regarding participants’ mean age, age at the first treatment, and current treatment length.
On the other hand, the high rates of the less-than-daily modality of heroin intake among unstable users during treatment imply that most patients reduced but did not stop heroin use during treatment. This result does not conform to the Dole and Nyswander methodology, and its most likely interpretation is linked to a suboptimal result of the current treatment [103]. Regarding the rates of psychiatric diagnoses—Bipolar Spectrum, Recurrent Depression, Anxiety Disorders, Chronic Psychosis, and, on the whole, a Dual *Disorder diagnosis* ($43.2\%$)—we can state that our sample is consistent with the literature, confirming the high occurrence of a bipolar spectrum among dual disorder patients [104,105,106,107,108]. Furthermore, our sample reported a lifetime co-occurring substance use of CNS-Stimulants, CNS-Depressants, and Hallucinogens [109,110,111,112].
Interestingly, an overall improvement was found when comparing the least satisfying week over a longer period (past five years) with the current week’s satisfaction. Considering the mean duration of the current treatment (27.54 months), it is likely that the findings described above represent three different stages of drug addiction history in our sample. Specifically, the participants’ worst week over five years must be related to a subjective condition before entry into treatment, thus representing typically unstable and impaired life situations in HUD patients during stage 3 of the disorder. To the best of our knowledge, this is the first time that a degree of subjective well-being contrasts has been assessed in a sample of HUD patients according to the Deltito–Subjective Wellness Scale.
## 4.2. Bivariate Correlation between the Severity of Stress Sensitivity and Other Clinical Aspects
The present study found a high positive correlation between the severity of stress sensitivity and many clinical aspects. In particular, participants’ income, altered mental status areas, legal problems, lifetime load of different treatments, and current treatment load correlated with the severity of the PTSD-spectrum. In a previous study performed by the V.P. Dole research group on HUD patients, the severity of symptoms related to the PTSD-spectrum seemed to be positively correlated with both the duration and the intensity of the stressful condition. Higher levels of PTSD-Spectrum symptomatology in subjects with a long history of heroin abuse were found [57,113]. Similar results were again observed in a study on long-term survivors of Hodgkin’s and non-Hodgkin’s lymphoma: subjects whose disease began at an earlier age suffered significantly more intense intrusion and avoidance symptoms [114].
The positive correlation between the general index of psychopathology and specific psychopathological syndromes of addictions is not surprising. In a previous study [35], after three months of stay in a TC (Therapeutic Community), a general reduction of SCL-90 severity was accompanied by a reduction in the frequency of those dimensions which were most closely related to the intoxication/withdrawal state and with active substance use-related behaviour (SS and W/BT). The least frequent variation concerns the patients allocated in the dimensions most involved in the addiction processes (PA and V/S). The present study underlines the close link between changes in stress sensitivity and psychopathology. Likewise, a negative correlation was found between stress sensitivity and the ‘contrast best week of the last five years’ at the DSWS, thus suggesting a trend to report a long-term worsening rate of subjective well-being in patients with greater stress sensitivity.
## 4.3. Differences between Patients with and without Heroin/Post-TrauPosttraumaticisorder—Spectrum (H/PTSD-S)
Regarding demographics, significant differences between patients with and without H/PTSD-S were found in sex and income distribution. These data might suggest a depressive dimension of H/PTSD-S, as depression is more prevalent in women. In addition, women are known to be at particular risk of developing PTSD [115]. A significant difference was also found in the ‘Mental Status’ problematic area, which was more impaired in patients with H/PTSD-S. Additionally, occupational and legal problems were more complicated in the H/PTSD-S group. Consistently, previous findings regarding stress sensitivity in unemployed and shift workers reported higher overall impairment, higher perceived stress, and lower subjective wellness while exhibiting elevated hair cortisol concentration [116].
As regards psychopathological indices and syndromes, the severity of such symptomatology was more significant in the H/PTSD-S group. Namely, patients with H/PTSD-S reported higher levels of severity in each of the five psychopathological dimensions. This result should not be surprising since stress sensitivity and psychopathology appear closely related. To explain, women who had experienced lifetime intimate partner violence were twice as likely to report depressive disorders, four times more likely to meet the criteria for an anxiety disorder, and seven times more likely to be diagnosed with PTSD [117]. Mental disorders increase the perpetuation of partner violence [118]. Indeed, the study of Van Reekum et al. showed that the occurrence of psychological symptoms increases the likelihood of interpersonal conflicts in both men and women [119].
Moreover, Mills et al. [ 82] described the impact of current and lifelong PTSD on long-term recovery from heroin dependence among participants who took part in the 11-year follow-up of the Australian Treatment Outcome Study (ATOS), a prospective naturalistic longitudinal study involving 615 people with heroin dependence recruited from Sydney, Australia, in 2001–2002. Seventy-one per cent of the cohort ($$n = 431$$) were re-interviewed 11-year later the study enrolment. Outcomes reviewed included heroin and other substance use, addiction, general physical and mental health, depression, PTSD, occupation, the incidence of trauma exposure, overdose, incarceration, and attempted suicide over the 11-year follow up. Despite having a poorer profile at baseline, individuals with current PTSD or a history of PTSD at baseline demonstrated similar levels of improvement as those without a history of PTSD at baseline in all outcome domains during the 11-year follow up. Nevertheless, PTSD was associated with consistently higher levels of major depression, attempted suicide, subsequent trauma exposure, and poorer occupational functioning across the 11-year follow up. These findings highlight the importance of occupational rehabilitation interventions, reducing the likelihood of re-traumatisation. Eventually, in the present study, no difference was found between the two groups as regards the prominence of the five different SCL-90 dimensions.
Our findings were further confirmed by the differences observed between the two groups in terms of subjective well-being, according to the D-SWS. Notably, both groups (patients with H/PTSD-S and patients without H/PTSD-S) reported two levels of improvement over time regarding subjective well-being. In particular, both groups reported improved subjective well-being when comparing the current week’s subjective well-being to last year’s poor week and the best week over five years. However, compared to patients without H/PTSD-S, patients with H/PTSD-S showed a lower rate of improvement on both the ‘contrast’ measures.
This report seems to suggest a different specific outcome in the context of the OAT. Patients with HUD—especially in the third stage of the disease—before entering treatment, actively seek a better quality of life and therefore require entry into treatment. The present study suggests that improvement in subjective well-being may be a key factor for the retention in treatment for such individuals after initiation of treatment. Therefore, beyond the behavioural, clinical, and psychopathological aspects, the subjective perception of the patients’ high being should be an essential factor during the OAT.
The attention of researchers and health professionals to the reported well-being of opioid users is growing [120,121,122]. OAT is known to improve the quality of life of patients. Nonetheless, measuring and standardising these outcomes has proven challenging as there is no consensus on which outcome measures of functioning or quality of life are related to drug use, which still generates much debate. The present study uses a triple scale of subjective well-being to easily compare the patient’s perception in three different periods to assess patients’ quality of life. Moreover, to the best of our knowledge, this is the first time that a degree of subjective well-being—and its contrasts—has been assessed in a sample of HUD patients according to the Deltito–Subjective Wellness Scale.
The limitations of this study include the number of examined patients is the major limitation of this study. In addition, running many analyses, a Type 1 error is possible. The group without H/PTSD-S was made up exclusively of males, and a poor sample size prevented multivariate analyses from being performed. Furthermore, much data in the present study were collected through self-report instruments, particularly concerning psychopathological, stress sensitivity, and patients’ well-being characteristics, thus influencing PTSD symptoms collection and diagnosis. A self-assessment of PTSD symptoms may be considered less accurate than the reports of the physicians. Moreover, other theories have been posited to explain the associations between stress on the one hand and substance abuse and addiction on the other, including tension reduction theory, stress-dampening response, and common genetic risk factors. Moreover, D-SWS has not been standardised yet, and CPSI and MC-Q have not been standardised in Italian.
We acknowledge that other theories have been posited to explain the associations between stress, on the one hand, and substance use behaviours and addiction, on the other, including stress-coping theories and common genetic risk factors. Furthermore, it was not the purpose of this study to explore the aetiology of PTSD in individuals with HUD. Instead, we aimed to describe the characteristics of PTSD-Spectrum. Further studies would better clarify the relationship between addiction and PTSD in terms of mutual impact.
## 5. Clinical and Research Implications
There is still no consensus on which perspective should be prioritised to reflect better treatment benefits: adverse social outcomes to avoid, symptom reduction, drug-taking behaviour, or patient perspective.
On this clinical background, information from patient-specific psychopathology, behavioural covariates of craving, stress sensitivity, and individuals’ subjective well-being could indicate new clinical information for monitoring patients undergoing OAT. It could be the key to better tailoring diagnostic and therapeutic interventions, laying the foundations for moving from harm reduction to patient-tailored therapy in OAT.
Specifically, the present study’s findings highlight the importance of assessing patients’ stress sensitivity—thus screening for PTSD—and subjective well-being during OAT. H/PTSD-S patients might display a worse clinical picture and outcome. Interventions targeting occupational rehabilitation, reducing the likelihood of re-traumatisation, and addressing PTSD and associated comorbidities should be considered.
## 6. Conclusions
In conclusion, in the present study, we suggested the presence of a PTSD-Spectrum occurring in a sample of HUD patients (H/PTSD-S). This syndrome was observed in a subsample of patients with high sensitivity to stress during OAT and showed a close correlation with the psychopathological syndromes of patients with HUD. Compared with patients without H/PTSD-S, patients with H/PTSD-S were mainly females with low income. They exhibited a more severe mental status at treatment entry, greater difficulty in working adjustment and legal problems during treatment. Additionally, H/PTSD-S patients showed a higher level of psychopathology and a lower improvement in subjective well-being during OAT.
Eventually, we further suggest that H/PTSD-Spectrum should be considered an outcome of HUD. Both addiction history and clinical features of HUD, in turn, seem to become significant risk factors for the onset of H/PTSD-S. Therefore, in such a perspective, social and behavioural impairment in HUD patients could be considered the clinical expression of the H/PTSD-spectrum. According to this view, the long-term outcome of HUD is not represented by drug-taking behaviours. Rather, the inability to cope with the contingent environmental conditions is the key feature to such a disorder. The H/PTSD-Spectrum, therefore, should be seen as a syndrome caused by an acquired inability (increased salience) concerning everyday daily life events.
## References
1. Nestler E.J.. **Epigenetic mechanisms of drug addiction**. *Neuropharmacology* (2014.0) **76 Pt B** 259-268. DOI: 10.1016/j.neuropharm.2013.04.004
2. Potenza M.N., Sofuoglu M., Carroll K.M., Rounsaville B.J.. **Neuroscience of behavioral and pharmacological treatments for addictions**. *Neuron* (2011.0) **69** 695-712. DOI: 10.1016/j.neuron.2011.02.009
3. Koob G.F., Volkow N.D.. **Neurobiology of addiction: A neurocircuitry analysis**. *Lancet Psychiatry* (2016.0) **3** 760-773. DOI: 10.1016/S2215-0366(16)00104-8
4. **Addiction and Health**
5. Koob G.F.. **Neurobiology of Opioid Addiction: Opponent Process, Hyperkatifeia, and Negative Reinforcement**. *Biol. Psychiatry* (2020.0) **87** 44-53. DOI: 10.1016/j.biopsych.2019.05.023
6. Rosen D., Hunsaker A., Albert S.M., Cornelius J.R., Reynolds C.F.. **Characteristics and consequences of heroin use among older adults in the United States: A review of the literature, treatment implications, and recommendations for further research**. *Addict. Behav.* (2011.0) **36** 279-285. DOI: 10.1016/j.addbeh.2010.12.012
7. Maremmani I., Pacini M., Pani P.P.. **On Behalf of the ‘Basics on Addiction Group’, Basics on Addiction: A training package for medical practitioners or psychiatrists who treat opioid dependence**. *Heroin Addict. Relat. Clin. Probl.* (2011.0) **13** 5-40
8. Mattick R.P., Breen C., Kimber J., Davoli M.. **Buprenorphine maintenance versus placebo or methadone maintenance for opioid dependence**. *Cochrane Database Syst. Rev.* (2014.0) **2** CD002207. DOI: 10.1002/14651858.CD002207.pub4
9. Kalk N.J.. **Harm reduction in opioid treatment: An established idea under threat**. *Addiction* (2019.0) **114** 20-21. DOI: 10.1111/add.14468
10. Kapadia S.N., Griffin J.L., Waldman J., Ziebarth N.R., Schackman B.R., Behrends C.N.. **A Harm Reduction Approach to Treating Opioid Use Disorder in an Independent Primary Care Practice: A Qualitative Study**. *J. Gen. Intern. Med.* (2021.0) **36** 1898-1905. DOI: 10.1007/s11606-020-06409-6
11. 11.
American Psychiatric Association
DSM-5. Diagnostic and Statistical Manual of Mental DisordersDSM-5 Task ForceAmerican Psychiatric AssociationWashington, DC, USA2013. *DSM-5. Diagnostic and Statistical Manual of Mental Disorders* (2013.0)
12. Kessler R.C., McGonagle K.A., Zhao S., Nelson C.B., Hughes M., Eshleman S., Wittchen H.U., Kendler K.S.. **Lifetime and 12-month prevalence of DSM-III-R psychiatric disorders in the United States. Results from the National Comorbidity Survey**. *Arch. Gen. Psychiatry* (1994.0) **51** 8-19. DOI: 10.1001/archpsyc.1994.03950010008002
13. Ross J., Teesson M., Darke S., Lynskey M., Ali R., Ritter A., Cooke R.. **The characteristics of heroin users entering treatment: Findings from the Australian treatment outcome study (ATOS)**. *Drug Alcohol Rev.* (2005.0) **24** 411-418. DOI: 10.1080/09595230500286039
14. Franke B., Michelini G., Asherson P., Banaschewski T., Bilbow A., Buitelaar J.K., Cormand B., Faraone S.V., Ginsberg Y., Haavik J.. **Live fast, die young? A review on the developmental trajectories of ADHD across the lifespan**. *Eur. Neuropsychopharmacol.* (2018.0) **28** 1059-1088. DOI: 10.1016/j.euroneuro.2018.08.001
15. Ross H.E., Swinson R., Doumani S., Larkin E.J.. **Diagnosing comorbidity in substance abusers: A comparison of the test-retest reliability of two interviews**. *Am. J. Drug Alcohol Abuse* (1995.0) **21** 167-185. DOI: 10.3109/00952999509002686
16. Hiser J., Koenigs M.. **The Multifaceted Role of the Ventromedial Prefrontal Cortex in Emotion, Decision Making, Social Cognition, and Psychopathology**. *Biol. Psychiatry* (2018.0) **83** 638-647. DOI: 10.1016/j.biopsych.2017.10.030
17. Smoller J.W., Andreassen O.A., Edenberg H.J., Faraone S.V., Glatt S.J., Kendler K.S.. **Psychiatric genetics and the structure of psychopathology**. *Mol. Psychiatry* (2019.0) **24** 409-420. DOI: 10.1038/s41380-017-0010-4
18. Eaton N.R., Rodriguez-Seijas C., Carragher N., Krueger R.F.. **Transdiagnostic factors of psychopathology and substance use disorders: A review**. *Soc. Psychiatry Psychiatr. Epidemiol.* (2015.0) **50** 171-182. DOI: 10.1007/s00127-014-1001-2
19. Pani P.P., Maremmani I., Trogu E., Gessa G.L., Ruiz P., Akiskal H.S.. **Delineating the psychic structure of substance abuse and addictions: Should anxiety, mood and impulse-control dysregulation be included?**. *J. Affect. Disord.* (2010.0) **122** 185-197. DOI: 10.1016/j.jad.2009.06.012
20. Maremmani A.G.I., Pani P.P., Rovai L., Bacciardi S., Maremmani I.. **Toward the Identification of a Specific Psychopathology of Substance Use Disorders**. *Front. Psychiatry* (2017.0) **8** 68. DOI: 10.3389/fpsyt.2017.00068
21. Maremmani A.G., Rovai L., Pani P.P., Pacini M., Lamanna F., Rugani F., Schiavi E., Dell’osso L., Maremmani I.. **Do methadone and buprenorphine have the same impact on psychopathological symptoms of heroin addicts?**. *Ann. Gen. Psychiatry* (2011.0) **10** 17. DOI: 10.1186/1744-859X-10-17
22. Maremmani A.G., Pani P.P., Trogu E., Vigna-Taglianti F., Mathis F., Diecidue R., Kirchmayer U., Amato L., Ghibaudi J., Camposeragna A.. **The impact of psychopathological subtypes on retention rate of patients with substance use disorder entering residential therapeutic community treatment**. *Ann. Gen. Psychiatry* (2016.0) **15** 29. DOI: 10.1186/s12991-016-0119-x
23. Derogatis L.R., Lipman R.S., Rickels K., Uhlenhuth E.H., Covi L.. **The Hopkins Symptom Checklist (HSCL): A self-report symptom inventory**. *Behav. Sci.* (1974.0) **19** 1-15. DOI: 10.1002/bs.3830190102
24. Carpenter K.M., Hittner J.B.. **Dimensional characteristics of the SCL-90-R: Evaluation of gender differences in dually diagnosed inpatients**. *J. Clin. Psychol.* (1995.0) **51** 383-390. DOI: 10.1002/1097-4679(199505)51:3<383::AID-JCLP2270510310>3.0.CO;2-Q
25. Gerra G., Leonardi C., Cortese E., Zaimovic A., Dell’Agnello G., Manfredini M., Somaini L., Petracca F., Caretti V., Baroni C.. **Adrenocorticotropic hormone and cortisol plasma levels directly correlate with childhood neglect and depression measures in addicted patients**. *Addict. Biol.* (2008.0) **13** 95-104. DOI: 10.1111/j.1369-1600.2007.00086.x
26. Zack M., Toneatto T., Streiner D.L.. **The SCL-90 factor structure in comorbid substance abusers**. *J. Subst. Abuse* (1998.0) **10** 85-101. DOI: 10.1016/S0899-3289(99)80143-6
27. Pani P.P., Trogu E., Vigna-Taglianti F., Mathis F., Diecidue R., Kirchmayer U., Amato L., Davoli M., Ghibaudi J., Camposeragna A.. **Psychopathological symptoms of patients with heroin addiction entering opioid agonist or therapeutic community treatment**. *Ann. Gen. Psychiatry* (2014.0) **13** 35. DOI: 10.1186/s12991-014-0035-x
28. Pani P.P., Maremmani A.G.I., Trogu E., Vigna-Taglianti F., Mathis F., Diecidue R., Kirchmayer U., Amato L., Davoli M., Ghibaudi J.. **Psychic Structure of Opioid Addiction: Impact of Lifetime Psychiatric Problems on SCL-90-based Psychopathologic Dimensions in Heroin-dependent Patients**. *Addict. Disord. Their. Treat.* (2016.0) **15** 6-16. DOI: 10.1097/ADT.0000000000000072
29. Pani P.P., Maremmani A.G.I., Trogu E., Vigna-Taglianti F., Mathis F., Diecidue R., Kirchmayer U., Amato L., Ghibaudi J., Camposeragna A.. **Psychopathology of addiction: May a SCL-90 based five dimensions structure be applied irrespectively of the involved drug?**. *Ann. Gen. Psychiatry* (2016.0) **15** 13. DOI: 10.1186/s12991-016-0100-8
30. Carbone M.G., Maiello M., Spera V., Manni C., Pallucchini A., Maremmani A.G.I., Maremmani I.. **The SCL90-based psychopathological structure may be applied in Substance Use Disorder patients irrespectively of the involved drug also in heroin, alcohol and cocaine mono-drug users**. *Heroin Addict. Relat. Clin. Probl.* (2018.0) **20** 29-34
31. Maremmani A.G.I., Gazzarrini D., Fiorin A., Cingano V., Bellio G., Perugi G., Maremmani I.. **Psychopathology of addiction: Is the SCL90-based five dimensional structure applicable to a non-substance-related addictive disorder such as Gambling Disorder?**. *Ann. Gen. Psychiatry* (2018.0) **17** 3. DOI: 10.1186/s12991-018-0173-7
32. Maremmani A.G.I., Maiello M., Carbone M.G., Pallucchini A., Brizi F., Belcari I., Conversano C., Perugi G., Maremmani I.. **Towards a psychopathology specific to Substance Use Disorder: Should emotional responses to life events be included?**. *Compr. Psychiatry* (2018.0) **80** 132-139. DOI: 10.1016/j.comppsych.2017.10.001
33. Carbone M.G., Tagliarini C., Ricci M., Lupi A.M., Sarandrea L., Ceban A., Casella P., Maremmani I.. **Ethnicity and specific psychopathology of addiction. Comparison between migrant and Italian Heroin Use Disorder patients**. *Heroin Addict. Relat. Clin. Probl.* (2019.0) **21** 61-66
34. Maremmani A.G.I., Lovrecic M., Lovrecic B., Maremmani I.. **Ethnicity and specific psychopathology of addiction. Comparison between Slovenian and Italian Heroin Use Disorder patients**. *Heroin Addict. Relat. Clin. Probl.* (2019.0) **21** 35-39
35. Maremmani I., Iantomasi C., Pani P.P., Maremmani A.G.I., Mathis F.. **For the VOECT Group, A prospective study of psychopathology stability and changes after 3-month drug-abstinence in Italian Substance Use Disorder patients during a residential treatment**. *Heroin Addict. Relat. Clin. Probl.* (2020.0) **22** 35-43
36. Maremmani A.G.I., Rovai L., Bacciardi S., Massimetti E., Gazzarrini D., Pallucchini A., Rugani F., Pani P.P., Maremmani I.. **Psychopathological dimensions and addictive behaviours in Heroin Use Disorder patients**. *Heroin Addict. Relat. Clin. Probl.* (2016.0) **18** 43-50
37. Della Rocca F., Maremmani A., Rovai L., Bacciardi S., Lamanna F., Maremmani I.. **Further evidence of a specific psychopathology of Heroin Use Disorder. Relationships between psychopathological dimensions and addictive behaviors**. *Heroin Addict. Relat. Clin. Probl.* (2017.0) **19** 13-20
38. Pagni L., Conversano C., Miccoli M., Maremmani A.G.I., Maremmani I.. **Is the psychopathology specific of Substance Use Disorder able to differentiate problematic and non-problematic recreational drinkers?**. *Heroin Addict. Relat. Clin. Probl.* (2020.0) **22** 49-56
39. Miccoli M., Poli A., Pagni L., Conversano C., Pani P.P., Maremmani A.G.I., Maremmani I.. **Further evidence of a psychopathology specific to Substance Use Disorder. Relationships between psychopathological dimensions and alcohol craving in recreational drinkers**. *Heroin Addict. Relat. Clin. Probl.* (2021.0) **23** 69-77
40. Miccoli M., Della Rocca F., Maremmani A.G.I., Maremmani I.. **Psychopathological syndromes and risk of alcoholism in Heroin Use Disorder patients compared with Substance-Non-Use peers**. *Heroin Addict. Relat. Clin. Probl.* (2021.0) **23** 71-79
41. Maremmani A.G.I., Della Rocca F., Bacciardi S., Miccoli M., Maremmani I.. **Convergent validity of five-dimension psychopathology of Addiction: Relationships with aggressive behaviour**. *Heroin Addict. Relat. Clin. Probl.* (2021.0) **23** 75-80
42. Maremmani A.G.I., Della Rocca F., Bacciardi S., Miccoli M., Maremmani I.. **Convergent validity of five-dimension psychopathology of Addiction: Relationships with Tridimensional Personality Questionnaire**. *Heroin Addict. Relat. Clin. Probl.* (2021.0) **23** 59-64
43. Della Rocca F., Maremmani A.G.I., Bacciardi S., Miccoli M., Maremmani I.. **Can the Psychopathology Specific to Substance Use Disorder Distinguish between Recreational Cocaine Use and Cocaine Addiction?**. *Heroin Addict. Relat. Clin. Probl.* (2022.0) **24** 43-50
44. Maremmani A.G.I., Cerniglia L., Cimino S., Bacciardi S., Rovai L., Rugani F., Massimetti E., Gazzarrini D., Pallucchini A., Pani P.P.. **Towards a specific psychopathology of heroin addiction. Comparison between Heroin Use Disorder and Major Depression patients**. *Heroin Addict. Relat. Clin. Probl.* (2015.0) **17** 9-16
45. Rugani F., Paganin W., Maremmani A.G.I., Perugi G., Maremmani I.. **Towards a specific psychopathology of heroin addiction. Comparison between Heroin Use Disorder and Chronic Psychotic patients**. *Heroin Addict. Relat. Clin. Probl.* (2019.0) **21** 53-59
46. Maremmani A.G.I., Pallucchini A., Manni C., Cipollone G., Della Rocca F., Salarpi G., Perugi G., Maremmani I.. **Can the psychopathology specific to Substance Use Disorder distinguish between reward and relief craving?**. *Heroin Addict. Relat. Clin. Probl.* (2020.0) **22** 5-11
47. Maremmani A.G.I., Cerniglia L., Cimino S., Maiello M., Della Rocca F., Maremmani I.. **The psychopathology specific to Substance Use Disorder is able to discriminate between young people with and without Problematic Internet Use**. *Heroin Addict. Relat. Clin. Probl.* (2020.0) **22** 39-47
48. Maremmani A.G.I., Cerniglia L., Cimino S., Bacciardi S., Rovai L., Pallucchini A., Spera V., Perugi G., Maremmani I.. **Further evidence of a specific psychopathology of addiction. Differentiation from other psychiatric psychopathological dimensions (such as obesity)**. *Int. J. Environ. Res. Public Health* (2017.0) **14**. DOI: 10.3390/ijerph14080943
49. Baker T.B., Piper M.E., McCarthy D.E., Majeskie M.R., Fiore M.C.. **Addiction motivation reformulated: An affective processing model of negative reinforcement**. *Psychol. Rev.* (2004.0) **111** 33-51. DOI: 10.1037/0033-295X.111.1.33
50. Moos R.H., Finney J.W., Chan D.A.. **The process of recovery from alcoholism. I. Comparing alcoholic patients and matched community controls**. *J. Stud. Alcohol.* (1981.0) **42** 383-402. DOI: 10.15288/jsa.1981.42.383
51. Leventhal H., Cleary P.D.. **The smoking problem: A review of the research and theory in behavioral risk modification**. *Psychol. Bull.* (1980.0) **88** 370-405. DOI: 10.1037/0033-2909.88.2.370
52. Chaplin T.M., Niehaus C., Goncalves S.F.. **Stress reactivity and the developmental psychopathology of adolescent substance use**. *Neurobiol. Stress* (2018.0) **9** 133-139. DOI: 10.1016/j.ynstr.2018.09.002
53. Khantzian E.J.. **An ego/self theory of substance dependence: A contemporary psychoanalitic perspective**. *NIDA Res. Monogr.* (1980.0) **30** 29-33. PMID: 6779190
54. Wills T.A., Hirky A.E., Zeidner M., Endler N.S.. **Coping and substance abuse: A theoretical model and review of the evidence**. *Handbook of Coping: Theory, Research, Applications* (1966.0) 279-302
55. Sinha R.. **Chronic stress, drug use, and vulnerability to addiction**. *Ann. N. Y. Acad. Sci.* (2008.0) **1141** 105-130. DOI: 10.1196/annals.1441.030
56. Koob G.F., Buck C.L., Cohen A., Edwards S., Park P.E., Schlosburg J.E., Schmeichel B., Vendruscolo L.F., Wade C.L., Whitfield T.W.. **Addiction as a stress surfeit disorder**. *Neuropharmacology* (2014.0) **76 Pt B** 370-382. DOI: 10.1016/j.neuropharm.2013.05.024
57. Dell’Osso L., Rugani F., Maremmani A.G., Bertoni S., Pani P.P., Maremmani I.. **Towards a unitary perspective between post-traumatic stress disorder and substance use disorder. Heroin use disorder as case study**. *Compr. Psychiatry* (2014.0) **55** 1244-1251. DOI: 10.1016/j.comppsych.2014.03.012
58. Cacciola J.S., Koppenhaver J.M., Alterman A.I., McKay J.R.. **Post-traumatic stress disorder and other psychopathology in substance abusing patients**. *Drug Alcohol Depend.* (2009.0) **101** 27-33. DOI: 10.1016/j.drugalcdep.2008.10.018
59. Contoreggi C., Herning R.I., Na P., Gold P.W., Chrousos G., Negro P.J., Better W., Cadet J.L.. **Stress hormone responses to corticotropin-releasing hormone in substance abusers without severe comorbid psychiatric disease**. *Biol. Psychiatry* (2003.0) **54** 873-878. DOI: 10.1016/S0006-3223(03)00167-7
60. Fosnocht A.Q., Briand L.A.. **Substance use modulates stress reactivity: Behavioral and physiological outcomes**. *Physiol. Behav.* (2016.0) **166** 32-42. DOI: 10.1016/j.physbeh.2016.02.024
61. Goeders N.E.. **The impact of stress on addiction**. *Eur. Neuropsychopharmacol.* (2003.0) **13** 435-441. DOI: 10.1016/j.euroneuro.2003.08.004
62. Koob G., Kreek M.J.. **Stress, dysregulation of drug reward pathways, and the transition to drug dependence**. *Am. J. Psychiatry* (2007.0) **164** 1149-1159. DOI: 10.1176/appi.ajp.2007.05030503
63. Kosten T.R.. **Stress and addiction**. *Am. J. Psychiatry* (2011.0) **168** 566-568. DOI: 10.1176/appi.ajp.2011.11020180
64. McClintick J.N., Xuei X., Tischfield J.A., Goate A., Foroud T., Wetherill L., Ehringer M.A., Edenberg H.J.. **Stress-response pathways are altered in the hippocampus of chronic alcoholics**. *Alcohol* (2013.0) **47** 505-515. DOI: 10.1016/j.alcohol.2013.07.002
65. Preston K.L., Epstein D.H.. **Stress in the daily lives of cocaine and heroin users: Relationship to mood, craving, relapse triggers, and cocaine use**. *Psychopharmacology* (2011.0) **218** 29-37. DOI: 10.1007/s00213-011-2183-x
66. Stewart J.. **Pathways to relapse: The neurobiology of drug- and stress-induced relapse to drug-taking**. *J. Psychiatry Neurosci.* (2000.0) **25** 125-136. PMID: 10740986
67. Weiss F., Ciccocioppo R., Parsons L.H., Katner S., Liu X., Zorrilla E.P., Valdez G.R., Ben-Shahar O., Angeletti S., Richter R.R.. **Compulsive drug-seeking behavior and relapse. Neuroadaptation, stress, and conditioning factors**. *Ann. N. Y. Acad. Sci.* (2001.0) **937** 1-26. DOI: 10.1111/j.1749-6632.2001.tb03556.x
68. Reynolds M., Mezey G., Chapman M., Wheeler M., Drummond C., Baldacchino A.. **Comorbid post-traumatic stress disorder in a substance misusing clinical population**. *Drug Alcohol Depend.* (2005.0) **77** 251-258. DOI: 10.1016/j.drugalcdep.2004.08.017
69. Briand L.A., Blendy J.A.. **Not all stress is equal: CREB is not necessary for restraint stress reinstatement of cocaine-conditioned reward**. *Behav. Brain Res.* (2013.0) **246** 63-68. DOI: 10.1016/j.bbr.2013.02.026
70. Zorn J.V., Schur R.R., Boks M.P., Kahn R.S., Joels M., Vinkers C.H.. **Cortisol stress reactivity across psychiatric disorders: A systematic review and meta-analysis**. *Psychoneuroendocrinology* (2017.0) **77** 25-36. DOI: 10.1016/j.psyneuen.2016.11.036
71. Parrott A.C., Sands H.R., Jones L., Clow A., Evans P., Downey L.A., Stalder T.. **Increased cortisol levels in hair of recent Ecstasy/MDMA users**. *Eur. Neuropsychopharmacol.* (2014.0) **24** 369-374. DOI: 10.1016/j.euroneuro.2013.11.006
72. Xie P., Wang T.J., Yin G., Yan Y., Xiao L.H., Li Q., Bi K.S.. **Metabonomic Study of Biochemical Changes in Human Hair of Heroin Abusers by Liquid Chromatography Coupled with Ion Trap-Time of Flight Mass Spectrometry**. *J. Mol. Neurosci.* (2016.0) **58** 93-101. DOI: 10.1007/s12031-015-0655-x
73. Howard D.E., Debnam K.J., Wang M.Q.. **Ten-year trends in physical dating violence victimisation among US adolescent females**. *J. Sch. Health* (2013.0) **83** 389-399. DOI: 10.1111/josh.12042
74. Kingston S., Raghavan C.. **The relationship of sexual abuse, early initiation of substance use, and adolescent trauma to PTSD**. *J. Trauma. Stress* (2009.0) **22** 65-68. DOI: 10.1002/jts.20381
75. Macedo T., Ribeiro C.A., Cotrim D., Tavares P., Morgadinho M.T., Caramona M., Vicente M.T., Rodrigues L., Cardoso M.G., Keating M.L.. **Catecholamine and MHPG plasma levels, platelet MAO activity, and 3H-imipramine binding in heroin and cocaine addicts**. *Mol. Neurobiol.* (1995.0) **11** 21-29. DOI: 10.1007/BF02740681
76. Gerra G., Zaimovic A., Zambelli U., Timpano M., Reali N., Bernasconi S., Brambilla F.. **Neuroendocrine responses to psychological stress in adolescents with anxiety disorder**. *Neuropsychobiology* (2000.0) **42** 82-92. DOI: 10.1159/000026677
77. Gerra G., Somaini L., Manfredini M., Raggi M.A., Saracino M.A., Amore M., Leonardi C., Cortese E., Donnini C.. **Dysregulated responses to emotions among abstinent heroin users: Correlation with childhood neglect and addiction severity**. *Prog. Neuropsychopharmacol. Biol. Psychiatry* (2014.0) **48** 220-228. DOI: 10.1016/j.pnpbp.2013.10.011
78. Parsons L.H., Smith A.D., Justice J.B.. **Basal extracellular dopamine is decreased in the rat nucleus accumbens during abstinence from chronic cocaine**. *Synapse* (1991.0) **9** 60-65. DOI: 10.1002/syn.890090109
79. Sher L., Vilens A.. *Neurobiology of Post-Traumatic Stress Disorder* (2010.0)
80. Drapkin M.L., Yusko D., Yasinski C., Oslin D., Hembree E.A., Foa E.B.. **Baseline functioning among individuals with post-traumatic stress disorder and alcohol dependence**. *J. Subst. Abuse Treat.* (2011.0) **41** 186-192. DOI: 10.1016/j.jsat.2011.02.012
81. Read J.P., Brown P.J., Kahler C.W.. **Substance use and post-traumatic stress disorders: Symptom interplay and effects on outcome**. *Addict. Behav.* (2004.0) **29** 1665-1672. DOI: 10.1016/j.addbeh.2004.02.061
82. Mills K.L., Marel C., Darke S., Ross J., Slade T., Teesson M.. **The long-term impact of post traumatic stress disorder on recovery from heroin dependence**. *J. Subst. Abuse Treat.* (2018.0) **89** 60-66. DOI: 10.1016/j.jsat.2018.04.001
83. Malhi G.S., Boyce P.. **A spectrum of views**. *Aust. N. Z. J. Psychiatry* (2014.0) **48** 376-377. DOI: 10.1177/0004867414528056
84. Maremmani I., Miccoli M.. **The TALS—Reduced form: An inventory for assessing the Heroin Post-Traumatic Stress Disorder Spectrum (H/PTSD-S)**. *Heroin Addict. Relat. Clin. Probl.* (2020.0) **22** 29-39
85. Maremmani I., Miccoli M.. *An Inventory for Assessing the Heroin Post-Traumatic Stress Disorder Spectrum (H/PTSD-S)* (2022.0)
86. Dell’Osso L., Carmassi C., Rucci P., Conversano C., Shear M.K., Calugi S., Maser J.D., Endicott J., Fagiolini A., Cassano G.B.. **A multidimensional spectrum approach to post-traumatic stress disorder: Comparison between the Structured Clinical Interview for Trauma and Loss Spectrum (SCI-TALS) and the Self-Report instrument (TALS-SR)**. *Compr. Psychiatry* (2009.0) **50** 485-490. DOI: 10.1016/j.comppsych.2008.11.006
87. Dell’osso L., Shear M.K., Carmassi C., Rucci P., Maser J.D., Frank E., Endicott J., Lorettu L., Altamura C.A., Carpiniello B.. **Validity and reliability of the Structured Clinical Interview for the Trauma and Loss Spectrum (SCI-TALS)**. *Clin. Pract. Epidemiol. Ment. Health* (2008.0) **4** 2. DOI: 10.1186/1745-0179-4-2
88. Maremmani I., Cecchini L., Avella M.T., Novi M., Ciapparelli A., Maremmani A.G.I.. **Prominent psychopathological and clinical characteristics of Heroin Post-Traumatic Stress Disorder Spectrum patients during treatment**. *Heroin Addict. Relat. Clin. Probl.* (2020.0) **22** 37-48
89. Dole V.P., Nyswander M.E.. **A medical treatment for diacetylmorphine (heroin) addiction: A clinical trial with methadone hydrocloride**. *JAMA* (1965.0) **193** 80-84. DOI: 10.1001/jama.1965.03090080008002
90. Dole V.P., Nyswander M.E., Warner A.. **Successful treatment of 750 criminal addicts**. *JAMA* (1968.0) **206** 2708-2711. DOI: 10.1001/jama.1968.03150120042009
91. Maremmani I., Castrogiovanni P.. *DAH-Q: Drug Addiction History Questionnaire* (1989.0)
92. Maremmani I., Pani P.P., Pacini M., Bizzarri J.V., Trogu E., Maremmani A.G., Gerra G., Perugi G., Dell’Osso L.. **Subtyping patients with heroin addiction at treatment entry: Factor derived from the Self-Report Symptom Inventory (SCL-90)**. *Ann. Gen. Psychiatry* (2010.0) **9** 15. DOI: 10.1186/1744-859X-9-15
93. Maremmani A.G.I., Rovai L., Bacciardi S., Massimetti E., Gazzarrini D., Rugani F., Pallucchini A., Piz L., Maremmani I.. **An inventory for assessing the behavioural covariates of craving in heroin substance use disorder. Development, theoretical description, reliability, exploratory factor analysis and preliminary construct validity**. *Heroin Addict. Relat. Clin. Probl.* (2015.0) **17** 51-60
94. Rawson R.A., ObertJ L., McCann D.P., Smith D.P., Scheffey E.H.. *The Neurobehavioural Treatment Manual: A Therapist Manual for Outpatient Cocaine Addiction Treatment* (1989.0)
95. Heishman S.J., Singleton E.G., Liguori A.. **Marijuana Craving Questionnaire: Development and initial validation of a self-report instrument**. *Addiction* (2001.0) **96** 1023-1034. DOI: 10.1046/j.1360-0443.2001.967102312.x
96. Tiffany S.T., Drobes D.J.. **The development and initial validation of a questionnaire on smoking urges**. *Br. J. Addict.* (1991.0) **86** 1467-1476. DOI: 10.1111/j.1360-0443.1991.tb01732.x
97. Tiffany S.T., Singleton E., Haertzen C.A., Henningfield J.E.. **The development of a cocaine craving questionnaire**. *Drug Alcohol Depend.* (1993.0) **34** 19-28. DOI: 10.1016/0376-8716(93)90042-O
98. Dell’osso L., Carmassi C., Stratta P., Massimetti G., Akiskal K.K., Akiskal H.S., Maremmani I., Rossi A.. **Gender Differences in the Relationship between Maladaptive Behaviors and Post-Traumatic Stress Disorder. A Study on 900 L’ Aquila 2009 Earthquake Survivors**. *Front. Psychiatry* (2012.0) **3** 111. DOI: 10.3389/fpsyt.2012.00111
99. McLellan A.T., McKay J.R., Forman R., Cacciola J., Kemp J.. **Reconsidering the evaluation of addiction treatment: From retrospective follow-up to concurrent recovery monitoring**. *Addiction* (2005.0) **100** 447-458. DOI: 10.1111/j.1360-0443.2005.01012.x
100. Bell J., Strang J.. **Medication Treatment of Opioid Use Disorder**. *Biol. Psychiatry* (2020.0) **87** 82-88. DOI: 10.1016/j.biopsych.2019.06.020
101. Maremmani I., Hill D., Sherbaum N., Auriacombe M., Bacciardi S., Benyamina A., Casella P., D’Agnone O., Daulouede J.-P., Deruvo G.. **Early-readmission after Agonist Opioid Treatment in five European countries. A drug addiction health policy challenge?**. *Heroin Addict. Relat. Clin. Probl.* (2021.0) **23** 69-79
102. Newman R.G., Tagliamonte A., Maremmani I.. **The Pharmacological Rationale for Methadone Treatment of Narcotic Addiction**. *Drug Addiction and Related Clinical Problems* (1995.0) 109-118
103. Bizzarri I.V., Casetti V., Sanna L., Maremmani A.G.I., Rovai L., Bacciardi S., Piacentino D., Conca A., Maremmani I.. **Agonist Opioid Treatment as historical comprehensive treatment (‘Dole & Nyswander’ methodology) is associated with better toxicology outcome than Harm Reduction Treatment**. *Ann. Gen. Psychiatry* (2016.0) **15** 34. PMID: 27933094
104. Maremmani A.G.I., Pacini M., Pani P.P., Ceccanti M., Bacciardi S., Akiskal H.S., Maremmani I.. **Possible trajectories of addictions: The role of bipolar spectrum**. *Heroin Addict. Relat. Clin. Probl.* (2016.0) **18** 23-32
105. Maremmani A.G.I., Bacciardi S., Rovai L., Rugani F., Massimetti E., Gazzarrini D., Dell’Osso L., Maremmani I.. **Six-month outcome in bipolar spectrum alcoholics treated with acamprosate after detoxification: A retrospective study**. *Int. J. Environ. Res. Public Health* (2014.0) **11**. DOI: 10.3390/ijerph111212983
106. Maremmani A.G.I., Bacciardi S., Rovai L., Rugani F., Dell’Osso L., Akiskal H.S., Maremmani I.. **Do bipolar patients use street opioids to stabilise mood?**. *Heroin Addict. Relat. Clin. Probl.* (2013.0) **15** 25-32
107. Bacciardi S., Maremmani A.G.I., Rovai L., Rugani F., Pani P.P., Pacini M., Dell’Osso L., Akiskal H.S., Maremmani I.. **Drug (heroin) addiction, bipolar spectrum and impulse control disorders**. *Heroin Addict. Relat. Clin. Probl.* (2013.0) **15** 29-36
108. Maremmani I., Perugi G., Pacini M., Akiskal H.S.. **Toward a Unitary Perspective on the Bipolar Spectrum and Substance Abuse: Opiate Addiction as a Paradigm**. *J. Affect. Disord.* (2006.0) **93** 1-12. DOI: 10.1016/j.jad.2006.02.022
109. Maremmani I., Lamanna F., Pacini M., Maremmani I.. **Clinical Foundation for the Use of Methadone in Polyabuse Patients**. *The Principles and Practice of Methadone Treatment* (2009.0) 181-188
110. Maremmani I., Pacini M.. **Use of Sodium Gamma-Hydroxybutyrate (GHB) in Alcoholic Heroin Addicts and Polydrug-Abusers**. *Heroin Addict. Relat. Clin. Probl.* (2007.0) **9** 55-76
111. Bandettini Di Poggio A., Fornai F., Paparelli A., Pacini M., Perugi G., Maremmani I.. **Comparison between heroin and heroin-cocaine polyabusers: A psychopathological study**. *Ann. N Y Acad. Sci.* (2006.0) **1074** 438-445. DOI: 10.1196/annals.1369.044
112. Maremmani I., Shinderman M.S.. **Alcohol, benzodiazepines and other drugs use in heroin addicts treated with methadone. Polyabuse or undermedication?**. *Heroin Addict. Relat. Clin. Probl.* (1999.0) **1** 7-13
113. Maremmani A.G.I., Rovai L., Bacciardi S., Bertoni S., Massimetti E., Gazzarrini D., Rugani F., Pallucchini A., Maremmani I.. **Addictive behaviours in Heroin Use Disorder patients. Correlation with heroin-craving patients’ self-evaluation**. *Heroin Addict. Relat. Clin. Probl.* (2016.0) **18** 9-18
114. Geffen D.B., Blaustein A., Amir M.C., Cohen Y.. **Post-traumatic stress disorder and quality of life in long-term survivors of Hodgkin’s disease and non-Hodgkin’s lymphoma in Israel**. *Leuk Lymphoma* (2003.0) **44** 1925-1929. DOI: 10.1080/1042819031000123573
115. Christiansen D.M., Berke E.T.. **Gender- and Sex-Based Contributors to Sex Differences in PTSD**. *Curr. Psychiatry Rep.* (2020.0) **22** 19. DOI: 10.1007/s11920-020-1140-y
116. Stalder T., Kirschbaum C.. **Analysis of cortisol in hair—state of the art and future directions**. *Brain Behav. Immun.* (2012.0) **26** 1019-1029. DOI: 10.1016/j.bbi.2012.02.002
117. Trevillion K., Oram S., Feder G., Howard L.M.. **Experiences of domestic violence and mental disorders: A systematic review and meta-analysis**. *PLoS ONE* (2012.0) **7**. DOI: 10.1371/journal.pone.0051740
118. Oram S., Trevillion K., Khalifeh H., Feder G., Howard L.M.. **Systematic review and meta-analysis of psychiatric disorder and the perpetration of partner violence**. *Epidemiol. Psychiatr. Sci.* (2014.0) **23** 361-376. DOI: 10.1017/S2045796013000450
119. van Reekum E.A., Rosic T., Hudson J., Sanger N., Marsh D.C., Worster A., Thabane L., Samaan Z.. **Social functioning outcomes in men and women receiving medication-assisted treatment for opioid use disorder**. *Biol. Sex Differ.* (2020.0) **11** 20. DOI: 10.1186/s13293-020-00298-4
120. De Maeyer J., Vanderplasschen W., Camfield L., Vanheule S., Sabbe B., Broekaert E.. **A good quality of life under the influence of methadone: A qualitative study among opiate-dependent individuals**. *Int. J. Nurs. Stud.* (2011.0) **48** 1244-1257. DOI: 10.1016/j.ijnurstu.2011.03.009
121. De Maeyer J., Vanderplasschen W., Broekaert E.. **Quality of life among opiate-dependent individuals: A review of the literature**. *Int. J. Drug Policy* (2010.0) **21** 364-380. DOI: 10.1016/j.drugpo.2010.01.010
122. Feelemyer J.P., Jarlais D.C.D., Arasteh K., Phillips B.W., Hagan H.. **Changes in quality of life (WHOQOL-BREF) and addiction severity index (ASI) among participants in opioid substitution treatment (OST) in low and middle income countries: An international systematic review**. *Drug Alcohol Depend.* (2014.0) **134** 251-258. DOI: 10.1016/j.drugalcdep.2013.10.011
|
---
title: 'Effects of a Multi-Professional Intervention on Mental Health of Middle-Aged
Overweight Survivors of COVID-19: A Clinical Trial'
authors:
- Joed Jacinto Ryal
- Victor Augusto Santos Perli
- Déborah Cristina de Souza Marques
- Ana Flávia Sordi
- Marilene Ghiraldi de Souza Marques
- Maria Luiza Camilo
- Rute Grossi Milani
- Jorge Mota
- Pablo Valdés-Badilla
- Braulio Henrique Magnani Branco
journal: International Journal of Environmental Research and Public Health
year: 2023
pmcid: PMC10002443
doi: 10.3390/ijerph20054132
license: CC BY 4.0
---
# Effects of a Multi-Professional Intervention on Mental Health of Middle-Aged Overweight Survivors of COVID-19: A Clinical Trial
## Abstract
The present study aimed to investigate the effects of a multi-professional intervention model on the mental health of middle-aged, overweight survivors of COVID-19. A clinical trial study with parallel groups and repeated measures was conducted. For eight weeks, multi-professional interventions were conducted (psychoeducation, nutritional intervention, and physical exercises). One hundred and thirty-five overweight or obese patients aged 46.46 ± 12.77 years were distributed into four experimental groups: mild, moderate, severe COVID, and control group. The instruments were used: mental health continuum-MHC, revised impact scale–IES-r, generalized anxiety disorder-GAD-7, and Patient health questionnaire PHQ-9, before and after eight weeks. The main results indicated only a time effect, with a significant increase in global MHC scores, emotional well-being, social well-being, and psychological well-being, as well as detected a significant reduction in global IES-R scores, intrusion, avoidance, and hyperarousal, in addition to a reduction in GAD-7 and PHQ-9 scores ($p \leq 0.05$). In conclusion, it was possible to identify those psychoeducational interventions that effectively reduced anxiety, depression, and post-traumatic stress symptoms in post-COVID-19 patients, regardless of symptomatology, in addition to the control group. However, moderate and severe post-COVID-19 patients need to be monitored continuously since the results of these groups did not follow the response pattern of the mild and control groups.
## 1. Introduction
A high prevalence of people infected with COVID-19 had psychoemotional sequelae, such as anxiety ($42\%$), depression ($31\%$), post-traumatic stress disorder (PTSD; $28\%$), and insomnia ($40\%$) after one month of hospital discharge infection [1]. The persistent symptoms of COVID were defined as long COVID or post-COVID-19 syndrome, characterized by people with sequelae, even after medical discharge [2]. Because of this, Xiang et al. [ 3] point out that the COVID-19 pandemic is closely associated with a significant increase in anxiety and depressive symptoms. On the other hand, the meta-analysis by Deng et al. [ 4] found that the prevalence of depression in post-COVID-19 patients was $45\%$, anxiety was $47\%$, and sleep disturbances impacted $34\%$ of those evaluated.
A cross-sectional study aimed to verify possible factors associated with depression, anxiety, and PTSD in 898 people after COVID-19, with almost one-third of them having symptoms of depression ($43\%$), anxiety ($45\%$), and PTSD ($32\%$), and symptoms associated with loneliness and low-stress tolerance [5]. On the other hand, people with more resilience and greater family support showed more resistance to these symptoms [5]. Undoubtedly, looking at these people who still have psychological sequelae is essential since the deleterious conditions related to mental health negatively impact the population’s health and quality of life [6]. In addition, another epidemiological study that evaluated 1733 people pointed out that after six months of discharge from COVID-19, sequelae were still verified, such as fatigue ($63\%$), difficulty sleeping ($26\%$), depression ($23\%$), and anxiety ($23\%$) [7].
Patients with overweight and especially, obese classified by body mass index (BMI) present increased risks of moderate and severe physical symptoms of COVID-19 [8] and also, could initiate mental health problems due to a long time for rehabilitation. Given this, psychoeducation added to other intervention tools may be effective in mitigating the effects on mental health during the COVID-19 pandemic [9] through psychoeducation, mindfulness exercises, promoting social interactions, stimulating wellness and brain health, validating emotional responses, and exploring patients’ strengths and how to organize rehabilitation goals. The above-mentioned strategies are considered a tool to help reduce anxiety and depressive symptoms and promote mental health education [10]. Considering the listed aspects, this population requires greater assistance to recover mental health and quality of life after COVID-19. Therefore, the present study aimed to investigate the effects of a multi-professional intervention model on the mental health of middle-aged, overweight survivors of COVID-19. Based on previous studies [7,9], it is presumed that psychoeducation can improve COVID-19 survivors’ mental health, providing a better quality of life for these persons.
## 2.1. Study Design
This study presents an experimental design (controlled trial) of repeated measures and four parallel groups: three intervention groups (mild, moderate, and severe), and a control group (without a positive diagnosis of COVID-19). This study followed the guidelines of the Consolidated Standards of Reporting Trials (CONSORT) [11]. Multi-professional interventions conducted by exercise physiologists (physical exercise), nutritionists (nutritional intervention), and psychologists (psychoeducation) were carried out, all in groups, over eight weeks.
## 2.2. Participants
Participants were recruited via the Municipal Secretary of Health of Maringa and the Municipal Hospital of Maringa. Thus, 141 participants of both sexes were eligible for the study. It was accepted by people with the following characteristics: (i) male and female aged between 19 and 65 years old; (ii) present a positive diagnosis for COVID-19 by laboratory confirmation; (iii) having received the first dose of the COVID-19 vaccine; (iv) being overweight or obese according to the cut-off points established by the World Health Organization [12]; (v) having participated in at least $85\%$ of the interventions [11]; (vi) participate in the pre-participation assessment; and (vii) have contracted COVID-19 between 3 January 2021 and 1 July 2021. As exclusion criteria, the following were not accepted: (i) patients with debilitating neurological diseases (i.e., Alzheimer’s, Parkinson’s disease, and plegies); (ii) reduction in intellectual capacity via completion of the cognitive failures questionnaire [13]; (iii) use of corticosteroids and/or having a chronic or acute disease that would contraindicate physical exercise; (iv) pregnancy; and (v) not signing the informed consent form. Information was obtained via screening performed in the study by Lemos et al. [ 8]. Preliminarily, the sample calculation was performed via G*Power software version 3.1, using an analysis of the variance of repeated measures. An effect size of $F = 0.4$ was estimated considering an α = 0.05 and a correlation between repeated measures of 0.5; a sample of 56 individuals was estimated for β = $80\%$. Considering a possible loss of follow-up, it was decided to recruit more than 140 participants. One hundred and forty-one participants were recruited, of whom five were excluded for declining to participate in the study and one for refusing to take an anamnesis. The participants were divided into four groups, considering the symptoms of COVID-19 or the control group, namely: control group ($$n = 29$$); mild, with no hospitalization ($$n = 41$$); moderate, with hospitalization, but without the necessity for oxygen support ($$n = 37$$); and severe, with hospitalization and the necessity for oxygen support, i.e., mechanical or non-mechanical oxygen supply ($$n = 28$$). There was a sample loss of 80 participants due to lack of time, lack of motivation, financial issues, and transport issues, and those participants did not carry out the assessments. Dropouts occurred between the sixth and eighth weeks of the intervention. Finally, 56 participants in the four experimental groups were evaluated before and after the intervention. The current study was approved by the local Ethics and Research Committee under 4,546,726 and registered in the Brazilian Clinical Trial Registration Platform (ReBEC) under registration: RBR-4mxg57b, in full compliance with the Declaration of Helsinki. Figure 1 shows the experimental design of the present study.
## 2.3. Procedures
Participants went to the university laboratory for medical clearance, with the following measurements being taken: measurement of body weight and height (subsequent calculation of BMI [12]) and completing a detailed anamnesis to identify the clinical picture and symptoms of COVID-19. Subsequently, the study participants answered the applied questionnaires (see sections below). Participants self-completed the instruments before and after eight weeks of multi-professional interventions after all explanations about the instruments used.
## 2.3.1. Mental Health Continuum–Short Form (MHC-SF)
To assess the well-being of the participants, the MHC-SF questionnaire was used, consisting of a Likert scale (1 to 6) with questions that measure the following components of well-being: Emotional well-being (EWB), social well-being (SWB), and psychological well-being (PWB), in the experiences the last two weeks [14]. Higher scores represent better indices of well-being, and the instrument has validation for the Brazilian population [15].
## 2.3.2. Impact of Event Scale-Revised (IES-R)
The Event Impact Scale (IES-R) is a validated questionnaire for tracking post-traumatic symptoms. [ 16], being validated for the Brazilian population [17]. The instrument consists of 22 questions on a Likert-type scale (0 to 3) (considering the last 7 days), in which the total score is obtained by the sum of the questions based on the evaluation criteria for PTSD from the Diagnostic and Statistical Manual of Mental Disorders (DSM-4) [18], with questions related to intrusion (In), avoidance (Av), and hyperarousal (Hy). High scores represent more intense symptoms of post-traumatic stress disorder.
## 2.3.3. Generalized Anxiety Disorder–(GAD-7)
To track participants’ anxiety levels, the GAD-7 was used. The instrument consists of 7 items that assess how much the patient is bothered by feeling nervous, anxious, worried, restless, and irritated in the last two weeks. Questions were answered on a Likert-type scale (0–3), and scores ranged from 0 to 21, where higher scores refer to a higher degree of anxiety [19]. The GAD-7 was validated for the Brazilian population [20].
## 2.3.4. Patient Health Questionnaire-9–(PHQ-9)
The PHQ-9 is composed of nine questions that verify the presence of each of the symptoms of an episode of depression presented in the Diagnostic and Statistical Manual of Mental Disorders [18]. The nine symptoms are depressed mood, anhedonia (loss of interest in doing activities and/or things), problems with sleeping, tiredness or lack of energy, change in appetite or weight, feelings of guilt or worthlessness, problems concentrating, feeling sluggish or restless, and suicidal thoughts [21]. The questions were answered using a Likert scale (0 to 4), in which higher scores represent a symptomatology closer to the depression episode. The instrument was validated for the Brazilian population [22].
## 2.4. Compositions of Interventions
Psychological and nutritional interventions were carried out on the premises of the university where the research was conducted. The multi-professional team was duly instructed and prepared to meet the needs of the participants. Theoretical-practical activities started with nutritional interventions or psychoeducation, followed by physical exercises (use of concurrent training).
## 2.4.1. Physical Exercise
Physical exercises were performed twice a week, focused on improving cardiorespiratory and neuromuscular fitness (mainly to improve functional capacity), lasting approximately 60 min per session. The training plan consisted of resistance exercises focused on large muscle groups, and cardiorespiratory exercises (which were performed on a treadmill, vertical/horizontal bicycle, or rowing ergometer). Each training was assembled individually, according to the needs of the participants [23], and was conducted in groups.
## 2.4.2. Nutritional Intervention
Nutritional interventions were based on the Food Guide for the Brazilian Population [24] and were performed once a week in groups. The central objective of the interventions was to instruct participants about the benefits of healthy eating for health, quality of life, and the reduction of risks associated with chronic noncommunicable diseases (NCDs). Each intervention lasted an average of 40 min, with one section per week. The following topics were addressed: (i) food pyramid; (ii) nutritional density of foods; (iii) macro- and micronutrients; (iv) association of food with health and quality of life; (v) nutritional composition of foods; (vi) differences between diet and light foods; (vii) means for preparing healthy food; (viii) nutritional education for health and quality of life; (ix) differences between fresh, minimally processed, processed, and ultra-processed foods; and (x) food to combat sarcopenic obesity.
## 2.4.3. Psychoeducation
Psychoeducation was based on therapeutic interventions with the central objective of proposing a model for treating and preventing mental illnesses based on an educational character [25,26] and was conducted once a week in groups. This approach used concepts and information from psychology and other areas so that the individual could broadly understand their situation and other illnesses present in our society. In this sense, the following were discussed: (i) the importance of physical exercise for a better quality of life and mental health; (ii) anxiety in everyday life, how it can impact our daily lives and, therefore, how to face them; (iii) discussions about obesity today: demystification of beliefs, prejudices, and stereotypes associated with obesity; (iv) understanding of the role of food in the social, psychological, and physical spheres; (v) information on post-traumatic stress disorder; (vi) promotion of a healthy lifestyle; (vii) reflections on stress; (viii) reflections on depressive symptoms; (ix) reflections on insomnia and relaxation techniques; (x) reflections on denial; (xi) reflections on fear; (xii) reflections on binge eating; (xiii) reflections on a healthy lifestyle; and (xiv) reflections on behavior changes. In addition, information leaflets on the respective topics were distributed at each meeting to reinforce the interventions applied and promote support material for the participants and the community [27,28]. Digital resources were also used with expository classes dialoguing with multimedia resources. The objective of the interventions was to provide knowledge and the possibility of change about the psychological consequences of the COVID-19 pandemic, in addition to guiding the essential themes of our century and helping the participating individuals have greater knowledge in the face of mental disorders that were provoked by the process of infection and the COVID-19 pandemic. Each intervention lasted an average of 40 min. Figure 2 shows the methodological design of the present study.
## 2.5. Statistical Analysis
Asymmetry and kurtosis tests and visual inspection of histograms analyzed the distributions of numeric variables. After analyzing the distributions, numerical data were described by the mean and standard deviation (±) or median and 25–75 percentiles, depending on the data normality. Categorical data were described with absolute frequency and relative frequency. Differences in the scores of each instrument were evaluated via analysis of variance (ANOVA) of mixed measures (groups and time) to identify possible differences between groups, time, and/or interactions. If a significant difference was detected, Bonferroni’s post hoc was used. The homogeneity of the data was analyzed using the Levene test, and the residual distribution analysis was performed utilizing visual inspection of the residual graphs. When only the timing effect was found, paired Student’s t-tests were performed for each group, to verify the possible effects of each group intervention. Absolute deltas (∆) were also calculated by performing a one-way ANOVA between groups. The “eta square” ƞ2 effect size was calculated according to the classification established by Richardson [29], which is: 0.0099 [small], 0.0588 [moderate], and 0.1379 [large]. A significance level of $5\%$ was established for all analyses. Statistical analyses were performed using the Statistica 12.0 software (StatSoft, Tulsa, OK, USA).
## 3. Results
The final sample consisted of 55 individuals, 36 ($65.45\%$) male, with a mean age of 49.93 ± 13.08 years old, of whom 19 ($35.19\%$) had a graduate degree, 19 ($35.19\%$) had completed higher education, and 13 ($24.07\%$) had only completed high school. Of the 55 participants, 40 ($72.73\%$) had a spouse, 9 ($16.36\%$) were single, and 6 ($10.91\%$) were divorced or widowed. Table 1 presents the initial characteristics of the participants according to the assessed groups.
Figure 3 presents the Mental Health Continuum (MHC) questionnaire scores of the participants in this study before and after the multi-professional interventions.
At the beginning of the intervention, all the groups did not present significant differences among them for all questions of the MHC questionnaire ($p \leq 0.05$). As described in Figure 3, a timing effect was observed, with a significant increase in global MHC scores (F3,52 = 10.03; $$p \leq 0.002$$; ƞ2 = 0.04–small), EWB emotional well-being (F3, 52 = 6.69; $$p \leq 0.013$$; ƞ2 = 0.03–small), social welfare-SWB (F3,52 = 6.11; $$p \leq 0.017$$; ƞ2 = 0.03–small), and well-being psychological-PWB (F3,52 = 8.17; $$p \leq 0.006$$; ƞ2 = 0.03–small) after the interventions. An interaction effect between group and time for psychological well-being-PWB (F3,52 = 3.86; $$p \leq 0.014$$; ƞ2= 0.03–small) was also observed, with an increase in the scores of the control group after the interventions ($$p \leq 0.024$$). For the total MHC, there was no significant difference in the deltas (F3.52 = 0.74; $$p \leq 0.527$$; ƞ2 = 0.04–small) of the different experimental groups. For the MHC and EWB, there was no significant difference in the deltas (F3.52 = 1.25; $$p \leq 0.29$$; ƞ2 = 0.06–moderate) of the different experimental groups. For the MHC and EWB, there was no significant difference in the deltas (F3,52 = 0.49; $$p \leq 0.68$$; ƞ2= 0.02–small) of the different experimental groups. For the MHC and PWB, a significant difference was observed for the deltas (F3,52 = 4.20; $$p \leq 0.009$$; ƞ2= 0.19–large) of the different experimental groups, with the Bonferroni post hoc showing significantly higher values for the control group when compared to the moderate ($$p \leq 0.006$$). Paired t-tests showed only a significant difference in the PWB score with higher values after intervention in the control group ($p \leq 0.05$), and a significant difference in the MHC global score with higher values after intervention in the mild group ($p \leq 0.05$). Besides, there was a tendency for EWB and PWB ($$p \leq 0.07$$; for both comparisons) to have higher values after intervention in the mild group. There was only a significant difference in EWB and SWB scores, with higher values for the moderate group after intervention ($p \leq 0.05$). However, no significant differences were observed for the severe group ($p \leq 0.05$).
Figure 4 shows the impact event scale revised (IES-R) scores of the participants in this study before and after the multi-professional interventions.
At the beginning of the intervention, all the groups did not present significant differences among them for all questions of the IES-R questionnaire ($p \leq 0.05$). There was only a time effect for the IES-R (Figure 4), with a significant reduction in the global IES-R scores (F3,52 = 12.22; $p \leq 0.001$; ƞ2 = 0.05–small), intrusion (F3,52 = 10.75; $$p \leq 0.002$$; ƞ2 = 0.05–small), avoidance (F3.52 = 6.59; $$p \leq 0.013$$; ƞ2 = 0.03–small) and hyperarousal (F3, 52 = 13.72; $p \leq 0.001$; ƞ2= 0.07–moderate) after the interventions. For the total IES-R, there was no significant difference in the deltas (F3,52 = 0.74; $$p \leq 0.52$$; ƞ2= 0.04–small) of the different experimental groups. For intrusion, there was no significant difference in the deltas (F3,52 = 1.53; $$p \leq 0.21$$; ƞ2 = 0.08–moderate) of the different experimental groups. For avoidance, there was no significant difference in the deltas (F3,52 = 2.22; $$p \leq 0.09$$; ƞ2 = 0.11–moderate) of the different experimental groups. For hyperarousal, there was no significant difference in the deltas (F3.52 = 0.52; $$p \leq 0.66$$; ƞ2 = 0.02–small) of the different experimental groups. Paired t-tests showed only a tendency for IES-R global score ($$p \leq 0.08$$) and In score ($$p \leq 0.07$$) for the control group. For the mild group, there was a significant difference in the IES-R global score ($$p \leq 0.02$$), In score ($$p \leq 0.03$$), Av score ($$p \leq 0.05$$), and Hy score ($$p \leq 0.02$$), with lower values after the intervention. There was no significant difference in t-tests in all IES-R scores for the moderate group after intervention ($p \leq 0.05$). Finally, there was just a significant difference in the Av score ($$p \leq 0.01$$) with lower values after intervention in the severe group. There were no significant differences for other paired t-tests comparison among the groups ($p \leq 0.05$).
Figure 5 presents the GAD-7 and PHQ-9 scores of the participants in this study before and after the multi-professional interventions.
At the beginning of the intervention, all the groups did not present significant differences among them for all questions of GAD-7 and PHQ-9 questionnaires ($p \leq 0.05$). A time effect was observed, with a significant reduction in the scores of the GAD-7 (F3,52 = 31.96; $p \leq 0.001$; ƞ2 = 0.14–large) and PHQ-9 (F3,52 = 18.15; $p \leq 0.001$; ƞ2 = 0.07–moderate) after the interventions. However, no group effect or interaction was found between the responses of the GAD-7 and PHQ-9 ($p \leq 0.05$). For the GAD-7, there was no significant difference in the deltas (F3,52 = 1.11; $$p \leq 0.35$$; ƞ2 = 0.06–moderate) of the different experimental groups. For the PHQ-9, there was also no significant difference in the deltas (F3,52 = 1.84; $$p \leq 0.15$$; ƞ2 = 0.09–moderate) of the different experimental groups. Paired t-tests showed a significant reduction in GAD-7 scores for the control ($$p \leq 0.05$$), mild ($$p \leq 0.0002$$), moderate ($$p \leq 0.02$$), and severe groups ($$p \leq 0.03$$) after interventions. In addition, paired t-tests showed only a significant reduction in PHQ-9 in mild ($$p \leq 0.0005$$), and moderate groups ($$p \leq 0.01$$) after interventions.
## 4. Discussion
The present study aimed to investigate the effects of a multi-professional intervention model on the mental health of middle-aged, overweight survivors of COVID-19. The results of the present study confirmed that psychoeducation, added to multi-professional activities, was effective in significantly improving the psychological symptoms in different experimental groups with higher or lower emphasis depending on the disease severity of COVID-19, and even in the control group. Therefore, multi-professional interventions effectively improved the mental health and sleep quality of participants in the present study (regardless of the experimental group).
To date, to the authors’ knowledge, no studies have investigated multi-professional interventions using psychoeducation, nutritional intervention, and physical exercise (together—in multi-professional interventions) in individuals who survived COVID-19 with overweight or obesity. The excess fat could reduce physical fitness and extend the treatment to recover the COVID-19 survivors [8]. Therefore, these patients require special care, due to their physical condition, and with a long recovery period, they may develop mental health problems. The scientific literature indicates that psychoeducation added to physical activities and a multi-professional approach can positively influence psychological aspects, such as a well-being decrease in anxiety and depressive symptoms, and contribute to the treatment and prevention of depression, anxiety, and post-traumatic stress [23,30]. Some similar practices have already shown positive results, reducing the impacts caused by the COVID-19 pandemic and other psychiatric conditions [9,10,31]. Another recent study from our research group showed similar results in approaches with concurrent exercise and dietary reeducation in overweight or obese middle-aged females [32].
The COVID-19 pandemic has negatively influenced the well-being of the general population [33]. Social isolation, the COVID-19 pandemic, low social support, low family income, and other aspects directly influenced the population’s mental health [5,34,35]. Thus, multi-professional interventions are considered well-known tools for preventing and treating various physical and mental disorders [10,36]. As a result, there was a significant improvement in general and specific components of well-being: emotional, social, and psychological. These responses may result from the emotional, social, and psychological support that a multi-professional team provides since, in this type of study, all areas of mental health are worked on, thus promoting psychological support, development of self-esteem, and improved social interaction [37,38,39].
Multi-professional teams allow a holistic view of individuals, enabling personal, social, and psychological development [40], factors that may have directly influenced the three subscales assessed by the MHC-SF: psychological well-being, social well-being, and emotional well-being. Cacioppo et al. [ 41] showed that loneliness could be directly linked to cardiovascular disease, sleep deregulation, and high cortisol release. There is evidence that social isolation is directly associated with the inflammatory system [42]. The systematic review by Williams et al. [ 43] identified that multi-professional interventions have already been proven to reduce loneliness through physical exercises, cognitive behavioral therapy, and psychoeducation. Additionally, behavioral psychoeducation interventions were positive for combating post-COVID-19 sequelae in a previous study [8]. In the study, as mentioned above, an intervention model based on cognitive behavioral therapy was used with mindfulness and other tools to assist post-COVID-19 rehabilitation. However, the interventions took place in the online format, with a shorter time (1 to 2 weeks), compared to the present study, which was carried out for eight weeks in person. Finally, significantly higher values were verified for the delta of psychological well-being in the control group compared to hospitalized individuals. These significant differences may be related to the possible deleterious impacts of COVID-19 on the mental health of hospitalized individuals.
As verified in this study, the participants showed a significant reduction in the global score and the IES-r subscales, namely intrusion, hyperarousal, and avoidance. These scales are due to the diagnosis of PTSD [16]. The difference in the subscales between pre and post represents a significant improvement in the diagnosis of the syndrome since this questionnaire is based on the DSM-IV criteria [18]. Among the effects of the COVID-19 pandemic, the increase in depressive symptoms also stands out negatively [7,44]. An effect persisted even after the worst moments of the pandemic, a recurrent symptom of the post-COVID-19 syndrome [7,45]. Therefore, physical activities are a primordial tool in the fight against depressive symptoms [46]. Likewise, psychoeducational activities reduce these symptoms and are often used as a non-drug treatment [10,44]. With that, the present study presents similarities with the findings in the literature that observed improvements in the PHQ-9 and GAD-7 scores [34,47]. Considering the evidence that points to the need for care for these individuals who have post-COVID symptoms [9,47,48,49], a unique look at the population after the COVID-19 pandemic becomes relevant so that effective techniques can be developed for the treatment of all the symptoms of this syndrome. Prolonged symptoms affect individuals’ quality of life and well-being [50]. Given this, the indispensability of psychoeducation actions and interdisciplinary actions to improve the physical, nutritional, and psychosocial health of COVID-19 survivors is confirmed. Although significant improvements (time effect) were observed for all instruments applied in the different experimental groups, caution is recommended in interpreting the findings, since the moderate and severe groups alone did not follow, in some circumstances, the standards of the other groups (control and mild). Therefore, longer interventions are suggested for the groups that had more severe symptoms of the disease, and even follow-up analyses to identify the behavior of the groups over time and even possible relapses. Thus, early intervention strategies can again be incorporated, to recover the mental health of COVID-19 survivors.
Our study provides information regarding COVID-19 survivors, which is timely and informative data for the intervention and recovery of those patients. However, in our study, there was a significant sample loss over the eight weeks of intervention, with a possible lack of interest on the part of the population to continue with multi-professional care. This occurred due to a lack of motivation and time, and patients believed that they were already better and would not need to continue with multi-professional activities. Almost all of the patients who dropped out of the interventions were low-income people. Qualitative feedback on drop-outs was linked to financial and transport issues. During the most restrictive period of the COVID-19 pandemic, people received minimal financial assistance from the government and started treatment. When the resources were exhausted, people had to work or stay at home and save money to buy food and pay the essential expenses of their respective households. Unfortunately, in Brazil, the researchers and universities cannot pay the expenses of the patients. Thus, a big part of drop-out is linked to the Brazilian reality. In addition, it was not possible to perform an intention-to-treat analysis, as the participants did not return to the university to be reassessed. Finally, no studies have been found combining multi-professional interventions with psychoeducation in COVID-19 survivors. Thus, the present study’s findings can guide possible actions to recover the global health conditions of those who contracted COVID-19.
## 5. Conclusions
It was concluded that multi-professional interventions significantly improved general well-being, emotional well-being, social well-being, and psychological well-being indicators in middle-aged, overweight survivors of COVID-19. Complementarily, there was a significant reduction in the scores representing the symptoms of post-traumatic stress disorder, the general scale of intrusion, avoidance, and hyperarousal. Furthermore, it was also possible to conclude that the interventions effectively reduced anxiety and depressive symptoms due to the reduction in scores after the multi-professional interventions. A significant difference over time was also observed in all evaluated groups, suggesting that psychoeducation added to a multi-professional team has a great positive impact on the mental health of people with different symptoms of COVID-19 and even those who do not have the infection. Lastly, patients in the moderate and severe groups need to be monitored continuously since the results of these groups did not follow the response patterns of the mild and control groups. Thus, the monitoring process and follow-up analyses for moderate and severe groups are indispensable and urgent.
## References
1. Mazza M.G., De Lorenzo R., Conte C., Poletti S., Vai B., Bollettini I., Melloni E.M.T., Furlan R., Ciceri F., Rovere-Querini P.. **Anxiety and depression in COVID-19 survivors: Role of inflammatory and clinical predictors**. *Brain Behav. Immun.* (2020) **89** 594-600. DOI: 10.1016/j.bbi.2020.07.037
2. Yong S.J.. **Long COVID or post-COVID-19 syndrome: Putative pathophysiology, risk factors, and treatments**. *Infect. Dis.* (2021) **53** 737-754. DOI: 10.1080/23744235.2021.1924397
3. Xiong J., Lipsitz O., Nasri F., Lui L.M.W., Gill H., Phan L., Chen-Li D., Iacobucci M., Ho R., Majeed A.. **Impact of COVID-19 pandemic on mental health in the general population: A systematic review**. *J. Affect. Disord.* (2020) **277** 55-64. DOI: 10.1016/j.jad.2020.08.001
4. Deng J., Zhou F., Hou W., Silver Z., Wong C.Y., Chang O., Huang E., Zuo Q.K.. **The prevalence of depression, anxiety, and sleep disturbances in COVID-19 patients: A meta-analysis**. *Ann. N. Y. Acad. Sci.* (2021) **1486** 90-111. DOI: 10.1111/nyas.14506
5. Liu C.H., Zhang E., Ba W.G.T.F., Hyun S., Hahm H.C.. **Factors associated with depression, anxiety, and PTSD symptomatology during the COVID-19 pandemic: Clinical implications for U.S. young adult mental health**. *Psychiatry Res.* (2020) **290** 113172. DOI: 10.1016/j.psychres.2020.113172
6. Guzmán-Muñoz E., Concha-Cisternas Y., Oñate-Barahona A., Lira-Cea C., Cigarroa-Cuevas I., Méndez-Rebolledo G., Castillo-Retamal M., Valdés-Badilla P., Zapata-Lamana R.. **Factores asociados a una baja calidad de vida en adultos chilenos durante la cuarentena por COVID-19**. *Rev. Med. Chil.* (2020) **148** 1759-1766. DOI: 10.4067/S0034-98872020001201759
7. Huang C., Huang L., Wang Y., Li X., Ren L., Gu X., Kang L., Guo L., Liu M., Zhou X.. **6-month consequences of COVID-19 in patients discharged from hospital: A cohort study**. *Lancet* (2021) **397** 220-232. DOI: 10.1016/S0140-6736(20)32656-8
8. Lemos M.M., Cavalini G.R., Henrique C.R.P., Perli V.A.S., Marchiori G.D.M., Marchiori L.L.D.M., Sordi A.F., de Moraes S.M.F., Ramos S.D.P., Valdés-Badilla P.. **Body composition and cardiorespiratory fitness in overweight or obese people post COVID-19: A comparative study**. *Front. Physiol.* (2022) **13** 949351. DOI: 10.3389/fphys.2022.949351
9. Jaywant A., Bueno-Castellano C., Oberlin L.E., Vanderlind W.M., Wilkins V.M., Cherestal S., Boas S.J., Kanellopoulos D.. **Psychological interventions on the front lines: A roadmap for the development of a behavioral treatment program to mitigate the mental health burden faced by COVID-19 survivors**. *Prof. Psychol. Res. Pract.* (2022) **53** 80-89. DOI: 10.1037/pro0000417
10. Lemes C.B., Neto J.O.. **Aplicações da psicoeducação no contexto da saúde**. *Temas Psicol.* (2017) **25** 17-28. DOI: 10.9788/TP2017.1-02
11. Schulz K.F., Altman D.G., Moher D.. **CONSORT 2010 Statement: Updated Guidelines for Reporting Parallel Group Randomised Trials**. *PLoS Med.* (2010) **7**. DOI: 10.1371/journal.pmed.1000251
12. 12.
World Health Organization
Obesity: Preventing and Managing the Global Epidemic ReportWHO Consultation on ObesityGeneva, Switzerland2000. *Obesity: Preventing and Managing the Global Epidemic Report* (2000)
13. de Paula J.J., Costa D.S., de Miranda D.M., Romano-Silva M.A.. **Brazilian version of the Cognitive Failures Questionnaire (CFQ): Cross-cultural adaptation and evidence of validity and reliability**. *Rev. Bras. Psiquiatr.* (2018) **40** 312-315. DOI: 10.1590/1516-4446-2017-2227
14. Lamers S.M., Westerhof G.J., Bohlmeijer E.T., Klooster P.T., Keyes C.L.. **Evaluating the psychometric properties of the mental health Continuum-Short Form (MHC-SF)**. *J. Clin. Psychol.* (2011) **67** 99-110. DOI: 10.1002/jclp.20741
15. Machado W.D.L., Bandeira D.R.. **Positive Mental Health Scale: Validation of the Mental Health Continuum—Short Form**. *Psico-Usf* (2015) **20** 259-274. DOI: 10.1590/1413-82712015200207
16. Creamer M., Bell R., Failla S.. **Psychometric properties of the Impact of Event Scale—Revised**. *Behav. Res. Ther.* (2003) **41** 1489-1496. DOI: 10.1016/j.brat.2003.07.010
17. Caiuby A.V.S., Lacerda S.S., Quintana M.I., Torii T.S., Andreoli S.B.. **Adaptação transcultural da versão brasileira da Escala do Impacto do Evento—Revisada (IES-R)**. *Cad. Saude Publica* (2012) **28** 597-603. DOI: 10.1590/S0102-311X2012000300019
18. 18.
American Psychiatric Association
Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Text Revision4th ed.American Psychiatric AssociationWashington, DC, USA2000Volume 110.1002/9780470515167.ch51. *Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Text Revision* (2000) **Volume 1**. DOI: 10.1002/9780470515167.ch51
19. Spitzer R.L., Kroenke K., Williams J.B.W., Löwe B.. **A Brief Measure for Assessing Generalized Anxiety Disorder: The GAD-7**. *Arch. Intern. Med.* (2006) **166** 1092-1097. DOI: 10.1001/archinte.166.10.1092
20. Moreno A.L., DeSousa D.A., Souza A.M.F.L.P., Manfro G.G., Salum G.A., Koller S.H., Osório F.L., Crippa J.A.S.. **Factor Structure, Reliability, and Item Parameters of the Brazilian-Portuguese Version of the GAD-7 Questionnaire**. *Temas Psicol.* (2016) **24** 367-376. DOI: 10.9788/TP2016.1-25
21. Kroenke K., Spitzer R.L., Williams J.B.. **The PHQ-9: Validity of a brief depression severity measure**. *J. Gen. Intern. Med.* (2001) **16** 606-613. DOI: 10.1046/j.1525-1497.2001.016009606.x
22. Santos I.S., Tavares B.F., Munhoz T.N., De Almeida L.S.P., Da Silva N.T.B., Tams B.D., Patella A., Matijasevich A.. **Sensibilidade e especificidade do Patient Health Questionnaire-9 (PHQ-9) entre adultos da população geral**. *Cad. Saude Publica* (2013) **29** 1533-1543. DOI: 10.1590/S0102-311X2013001200006
23. Jimeno-Almazán A., Franco-López F., Buendía-Romero Á., Martínez-Cava A., Sánchez-Agar J.A., Martínez B.J.S., Courel-Ibáñez J., Pallarés J.G.. **Rehabilitation for post-COVID-19 condition through a supervised exercise intervention: A randomized controlled trial**. *Scand. J. Med. Sci. Sports* (2022) **32** 1791-1801. DOI: 10.1111/sms.14240
24. 24.
Brasil Ministério da Saude
Guia Alimentar para a População Brasileira Guia Alimentar para a População Brasileira, 2°Brasil Ministério da SaudeBrasília, Brazil2014. *Guia Alimentar para a População Brasileira Guia Alimentar para a População Brasileira, 2°* (2014)
25. Authier J.. **The Psychoeducation Model: Definition, Contemporary Roots and Content**. *Can. J. Couns. Psychother.* (1977) **12** 15-22
26. Beck J.S.. *Terapia Cognitivo-Componamental: Teoria e Prática, 2°* (2013)
27. McCrorie A., Donnelly C., McGlade K.. **Infographics: Healthcare Communication for the Digital Age**. *Ulst. Med. J.* (2016) **85** 71-75
28. De Oliveira D.A., Lessa R.S., Ribeiro S.C.S., De Vasconcelos P.F.. **The Visual Practice: The Infographic as a Facilitating Tool for Learning in Medical School**. *Rev. Bras. Educ. Med.* (2020) **44** 4. DOI: 10.1590/1981-5271v44.4-20200158.ing
29. Richardson J.T.E.. **Eta squared and partial eta squared as measures of effect size in educational research**. *Educ. Res. Rev.* (2011) **6** 135-147. DOI: 10.1016/j.edurev.2010.12.001
30. Arantes D.J., Picasso R., da Silva E.A.. **Grupos psicoeducativos com familiares dos usuários de um Centro de Atenção Psicossocial**. *Pesqui. Prát. Psicossociais* (2019) **14** 1-15
31. Bhattacharjee D., Rai A.K., Singh N.K., Kumar P., Munda S.K., Das B.. **Psychoeducation: A Measure to Strengthen Psychiatric Treatment**. *Delhi Psychiatry J.* (2011) **14** 33-39
32. Bolognese M.A., Franco C.B., Ferrari A., Bennemann R.M., Lopes S.M.A., Bertolini S.M.M.G., Júnior N.N., Branco B.H.M.. **Group Nutrition Counseling or Individualized Prescription for Women With Obesity? A Clinical Trial**. *Front. Public Health* (2020) **8** 127. DOI: 10.3389/fpubh.2020.00127
33. Golińska P., Cieślak M., Hubert O., Bidzan M.. **Mental Health and the Symptoms of PTSD in People with Depression and Anxiety Disorders during the COVID-19 Pandemic**. *Int. J. Environ. Res. Public Health* (2021) **18**. DOI: 10.3390/ijerph18115542
34. Matsumoto K., Hamatani S., Shimizu E., Käll A., Andersson G.. **Impact of post-COVID conditions on mental health: A cross-sectional study in Japan and Sweden**. *BMC Psychiatry* (2022) **22**. DOI: 10.1186/s12888-022-03874-7
35. Anjum S., Ullah R., Rana M.S., Khan H.A., Memon F.S., Ahmed Y., Jabeen S., Faryal R.. **COVID-19 Pandemic: A Serious Threat for Public Mental Health Globally**. *Psychiatr. Danub.* (2020) **32** 245-250. DOI: 10.24869/psyd.2020.245
36. Khoramipour K., Basereh A., Hekmatikar A.A., Castell L., Ruhee R.T., Suzuki K.. **Physical activity and nutrition guidelines to help with the fight against COVID-19**. *J. Sports Sci.* (2021) **39** 101-107. DOI: 10.1080/02640414.2020.1807089
37. Guo K., Zhang X., Bai S., Minhat H.S., Nazan A.I.N.M., Feng J., Li X., Luo G., Zhang X., Feng J.. **Assessing social support impact on depression, anxiety, and stress among undergraduate students in Shaanxi province during the COVID-19 pandemic of China**. *PLoS ONE* (2021) **16**. DOI: 10.1371/journal.pone.0253891
38. Styra R., Hawryluck L., Mc Geer A., Dimas M., Sheen J., Giacobbe P., Dattani N., Lorello G., Rac V.E., Francis T.. **Surviving SARS and living through COVID-19: Healthcare worker mental health outcomes and insights for coping**. *PLoS ONE* (2021) **16**. DOI: 10.1371/journal.pone.0258893
39. van Agteren J., Iasiello M., Lo L., Bartholomaeus J., Kopsaftis Z., Carey M., Kyrios M.. **A systematic review and meta-analysis of psychological interventions to improve mental wellbeing**. *Nat. Hum. Behav.* (2021) **5** 631-652. DOI: 10.1038/s41562-021-01093-w
40. Egan K.P.. **Supporting Mental Health and Well-being Among Student-Athletes**. *Clin. Sports Med.* (2019) **38** 537-544. DOI: 10.1016/j.csm.2019.05.003
41. Cacioppo J.T., Hawkley L.C., Crawford L.E., Ernst J.M., Burleson M.H., Kowalewski R.B., Malarkey W.B., Van Cauter E., Berntson G.G.. **Loneliness and Health: Potential Mechanisms**. *Psychosom. Med.* (2002) **64** 407-417. DOI: 10.1097/00006842-200205000-00005
42. Smith K.J., Gavey S., Riddell N.E., Kontari P., Victor C.. **The association between loneliness, social isolation and inflammation: A systematic review and meta-analysis**. *Neurosci. Biobehav. Rev.* (2020) **112** 519-541. DOI: 10.1016/j.neubiorev.2020.02.002
43. Williams C.Y.K., Townson A.T., Kapur M., Ferreira A.F., Nunn R., Galante J., Phillips V., Gentry S., Usher-Smith J.A.. **Interventions to reduce social isolation and loneliness during COVID-19 physical distancing measures: A rapid systematic review**. *PLoS ONE* (2021) **16**. DOI: 10.1371/journal.pone.0247139
44. Vindegaard N., Benros M.E.. **COVID-19 pandemic and mental health consequences: Systematic review of the current evidence**. *Brain Behav. Immun.* (2020) **89** 531-542. DOI: 10.1016/j.bbi.2020.05.048
45. Shanbehzadeh S., Tavahomi M., Zanjari N., Ebrahimi-Takamjani I., Amiri-Arimi S.. **Physical and mental health complications post-COVID-19: Scoping review**. *J. Psychosom. Res.* (2021) **147** 110525. DOI: 10.1016/j.jpsychores.2021.110525
46. Dinas P.C., Koutedakis Y., Flouris A.D.. **Effects of exercise and physical activity on depression**. *Ir. J. Med. Sci.* (2011) **180** 319-325. DOI: 10.1007/s11845-010-0633-9
47. Nicoletti M., Gonzaga A.P., Modesto S.E.F., Cobelo A.W.. **Grupo psicoeducativo multifamiliar no tratamento dos transtornos alimentares na adolescência**. *Psicol. Estud.* (2010) **15** 217-223. DOI: 10.1590/S1413-73722010000100023
48. Desai A.D., Lavelle M., Boursiquot B.C., Wan E.Y.. **Long-term complications of COVID-19**. *Am. J. Physiol. -Cell Physiol.* (2022) **322** C1-C11. DOI: 10.1152/ajpcell.00375.2021
49. Costa A.C.D.S., Menon V., Phadke R., Dapke K., Miranda A.V., Ahmad S., Essar M.Y., Hashim H.T.. **Mental health in the post COVID-19 era: Future perspectives**. *Einstein* (2022) **20**. DOI: 10.31744/einstein_journal/2022CE6760
50. Vlake J.H., Wesselius S., van Genderen M.E., van Bommel J., Klerk B.B.-D., Wils E.-J.. **Psychological distress and health-related quality of life in patients after hospitalization during the COVID-19 pandemic: A single-center, observational study**. *PLoS ONE* (2021) **16**. DOI: 10.1371/journal.pone.0255774
|
---
title: Changes in the Histological Structure of Adrenal Glands and Corticosterone
Level after Whey Protein or Bee Pollen Supplementation in Running and Non-Running
Rats
authors:
- Karolina Frankowska
- Michał Zarobkiewicz
- Mirosław A. Sławiński
- Ewelina Wawryk-Gawda
- Monika Abramiuk
- Barbara Jodłowska-Jędrych
journal: International Journal of Environmental Research and Public Health
year: 2023
pmcid: PMC10002451
doi: 10.3390/ijerph20054105
license: CC BY 4.0
---
# Changes in the Histological Structure of Adrenal Glands and Corticosterone Level after Whey Protein or Bee Pollen Supplementation in Running and Non-Running Rats
## Abstract
Due to the many health-promoting properties of bee pollen and whey protein, both products are widely used as dietary supplements. According to these reports on their health-promoting properties, the aim of our study is to assess whether these products can influence the structure and function of the adrenal glands in rats. Thirty male Wistar rats were divided into six equal groups. Among them, there were three groups which included non-running rats and three groups which included running rats. Both of these running ($$n = 3$$) and non-running ($$n = 3$$) groups included non-supplemented (control groups), bee-pollen-supplemented groups, and whey-protein-supplemented groups. After 8 weeks, the rats were decapitated, their adrenal glands were collected, and paraffin slides were prepared. Then, staining according to the standard H&E and Masson’s trichrome protocols was performed. Fecal and urine samples were collected prior to the end of the study to measure corticosterone levels. In the group of non-running rats, the consumption of bee pollen was noted to be significantly higher when compared to the group of running rats ($p \leq 0.05$). The thickness of the particular adrenal cortex layers was similar among all of the groups ($p \leq 0.05$). The statistically significant changes in the microscopic structure of the adrenal glands, especially regarding cell nuclei diameter and structure, as well as the architecture of sinusoids, were observed between the groups. Moreover, urine corticosterone concentrations were found to vary between all of the analyzed groups ($p \leq 0.05$). These results indicate that both bee pollen and whey protein have limited stress-reducing potential.
## 1. Introduction
Stress is defined as a physiological response of the body to a state of danger. Such modification of homeostasis is achieved as a result of the complex interactions within the elements of the hypothalamus–pituitary–adrenal axis (HPA axis). As a subsequent consequence of axis activation, adrenal glands—the main organs involved in stress-response—secrete stress hormones, finally regulating the organism functioning at the multi-organ level [1,2].
The occurrence of stress reactions is conditioned multifactorial—both by internal and external stimuli. For sure, any differences in physical activity—its abandonment or limitation—are a possible source of stress induction [3,4]. Moreover, physical activity is also significant from the perspective of counteracting subsequent negative stress effects [5]. It is especially important taking into consideration that exposure to stress has many health consequences and may affect among others the cardiovascular system, the nervous system, or the immune state. The current literature indicates a significant role of psychological stress in the pathogenesis of asthma, Alzheimer’s disease, and cancer development, although due to the complex relevant determinants of these diseases, it is difficult to assess the exact role of stress in their etiology [6,7,8,9]. In addition to triggering the onset of disease, stress modifies the amount and type of food intake, which has been observed both in humans and animal models [10]. The occurrence of some of such changes in food preferences is explained by the influence of corticosterone produced by the adrenal glands [11]. Although the effect of stress on food preferences and the amount and type of consumed food varies, stress can exacerbate the desire to eat palatable foods, rich in fats and sugars, referred to as comfort food [12,13]. Due to the high prevalence of negative stress-related effects in people, substances with beneficial effects on health are being sought to minimize the negative consequences of stress. Hence, it has been proven that different types of food, e.g., coconut oil and mung beans have potential anti-stress values [14,15].
Bee pollen, a product obtained from honeybees, is a very complex compound consisting of approximately two hundred different substances. Although bee pollen composition differs depending on the species from it which originates, all major components, such as proteins, lipids, carbohydrates, or vitamins, and bio-elements are present in bee pollen independently of the origin [16]. When analyzing the percentage share of each group of compounds, it is noticeable that the largest part of it is carbohydrates, the content of which reaches about 30 percent. The average protein content in bee pollen oscillates to and from approximately 20 percent, of which essential amino acids represent a significant proportion. Among the other components of bee pollen are nucleic acids, lipids, and crude fiber [16,17]. In addition, bee pollen is a source of numerous macro- and microelements with a particularly significant content of potassium, iron, phosphorus, magnesium, and zinc [17]. Moreover, its composition is characterized by the presence of practically all vitamins including provitamin A, vitamin E, thiamine, niacin, pantothenic, nicotinic, and folic acid. Among the bioactive substances contained in bee pollen, the presence of phenolic compounds is worthy of note, since flavonoids, which make up a significant proportion of that group of compounds, are responsible for the antioxidant properties of bee pollen [17,18]. Such a rich composition of bee pollen is responsible for a number of the known health benefits of this substance, including hypolipidemic and glucose-ameliorating activities, as well as detoxifying and anti-inflammatory action [17,19]. All of these nutritional properties make bee pollen a valuable functional food able to enrich the diet [19,20].
Whey protein is a substance representing a significant proportion of the proteins contained in cow milk. It is processed to produce preparations such as whey protein concentrate (WPC), whey protein isolate (WPI), or whey protein hydrolyzed (WPH) with varying protein contents [21]. Regardless of the processing route, whey protein is primarily a rich source of β-lactoglobulin and α-lactalbumin. In addition, it is characterized by a content of ingredients such as essential amino acids, branched-chain amino acids, immunoglobulins, and lactoferrin [22,23]. Currently, whey protein is widely used as a supplement among athletes due to its beneficial effects on muscles [23,24]. Its anti-inflammatory, cardioprotective, neuroprotective, and anti-cancer properties also support its role as a functional food [23,24,25].
Although to date, there are no studies on the anti-stress potential of bee pollen, there are reports on the effect on stress of similar products such as propolis and royal jelly, suggesting a potential anti-stress effect in different animals [26,27,28,29,30]. Furthermore, a protein-rich diet is also known for its potential anti-stress properties [31]. In view of these reports, we have investigated whether there is a possibility that bee pollen and whey protein supplementation may influence histological properties, the function of adrenal glands, and thus also the response to stress [32].
## 2.1. Study Protocol
Thirty eight-week-old male Wistar rats were divided into six equal groups (five rats per group). Non-supplemented groups (No. I and No. II), also referred to as the control groups, included a non-running group (No. I) and a running group (No. II). The experimental groups (No. III–VI), understood as the supplemented ones, were non-running (No. III–IV) supplemented with whey protein (No. III) or bee pollen (No. IV), as well as running (No. V–VI) supplemented with whey protein (No. V) or bee pollen (No. VI) (Figure 1). During the 8 weeks of the experimental phase, all of the animals received water and rodent food ad libitum; the bee-pollen-supplemented group also received bee pollen and the whey-protein-supplemented group also received enriched whey protein concentrate (Olimp Laboratories Sp. z.o.o., Dębica, Poland). The daily rodent food, bee pollen, whey protein, and water consumption were measured each day. During the experimental phase, the rats in the running groups ran five times per week, with the duration of a single run being 5 min, on a treadmill built by us before starting the experiment. The average velocity was 6 km/h and the rats were not assisted by electrical shock. The rats from the non-running groups did not use the treadmill. At the beginning of the experiment, the body mass of the rats was approximately 330 g, while at the end of the experiment, it increased to approximately 400 g, regardless of the group. At the end of the experimental phase, all of the rats were decapitated and their adrenal glands were collected. Immediately after collection, the adrenal gland mass was measured with a digital analytical balance with 0.1 mg readability AS 110.R2 (Radwag, Lublin, Poland). After fixation in formalin, paraffin blocks were prepared.
The study protocol was approved by the Bioethical Committee at the Medical University of Lublin (No. $\frac{24}{2015}$).
## 2.2. Supplements
The bee pollen was collected in the vicinity of Lublin, Poland. It contained approximately 31 g of carbohydrates, 23 g of protein, 5 g of lipids, and 0.8 g of various vitamins (A, E, D, B1, B2, B3, B5, B6, B7, and C) per 100 g [16]. The 100 g of enriched whey protein concentrate (further called either whey protein or WPC) contained 77 g of protein, 6 g of carbohydrates, and 7 g of lipids, as reported previously [33].
## 2.3. Histological Staining and Analysis
Five μm-thick slides were prepared and stained according to the standard H&E and Masson’s trichrome protocols. The slides were then analyzed under a light microscope. Olympus BX4 with a digital camera and CellSens software (Version 4.1. CS-EN-V4) were used for image capture. The measurement of vacuolization was performed in Fiji as previously described [34]. The vacuolization rate was calculated as a percentage of the area occupied by vacuoles to the total analyzed area in the particular cortex layer. The measurement of the extent of fibrosis was performed in Fiji, as reported previously [33].
## 2.4. ELISA
A corticosterone ELISA kit (Cayman Chemical, Ann Arbor, MI, USA) was used for the measurement of fecal corticosterone content. Fresh feces samples, collected during the second to last week of the experiment, were frozen at −80 °C immediately after collection. Prior to corticosterone measurement, the samples were dried in a heat cabinet at 30 °C for 2 h as proposed by L. Pihl and J. Hau [35]. Then, the samples were prepared according to the manufacturer’s instructions. Large particles were removed by shifting through a stainless steel mesh. Twenty mg of each sample was suspended in 1ml of methanol. Then, the samples were vortexed for 30 min and centrifuged for 20 min at 2500× g. The supernatant was transferred into clean a Eppendorf tube and diluted 1:50 in ELISA Buffer (Cayman Chemical, Ann Arbor, MI, USA). The samples were prepared in such a way were then used in ELISA according to the manufacturer’s protocol. The absorbance was measured with a plate reader Biotek Elx-800 (BioTek, Winooski, VT, USA).
A corticosterone ELISA Kit (R&D Systems, Minneapolis, MN, USA) was used for corticosterone measurement in urine. Each sample was run in duplicate. Fresh urine samples were collected from metabolic cages shortly prior to study termination and immediately frozen at −80 °C. Prior to analysis, the samples were first transferred to −20 °C and then completely thawed. They were centrifuged for 10 min at 18,000× g. The supernatant was collected and diluted 100-times with Calibrator Diluent RD5-43 supplied as part of the kit. An ELISA assay was prepared according to manufacturer’s instructions. The microplate was read with Biotek Elx-800. The raw data were analyzed with elisaanalysis.com (ElisaKit) in the case of urine and with a spreadsheet supplied by Cayman Chemicals in the case of feces. Each sample was run in duplicate.
## 2.5. Statistical Analysis
The collected data were statistically analyzed with Statistica 12 (StatSoft, St. Tulsa OK, USA). The distribution of the data was analyzed with the Shapiro–Wilk test. The statistical significance was calculated with the Kruskal–Wallis test, and the level of significance was set at $p \leq 0.05.$
## 3.1. Food Intake
The daily whey protein intake was similar in both the running (5.19 g per rat) and non-running (5.18 g per rat) groups, while the consumption of bee pollen varied with 13.45 g per rat in the non-running group and 11.96 g per rat in the running group. The detailed consumption and mass changes have been reported previously [33].
## 3.2. Adrenal Gland Mass
No statistically significant difference in adrenal gland mass was observed ($$p \leq 0.65$$). Specifically, no differences were noted between any of the experimental groups or their respective control groups. The two controls also did not differ from one another either.
## 3.3. Vacuoles and Structure
No significant changes in adrenal architecture were noted under the optical microscope after H&E staining at 400× magnification. In addition, no statistically significant changes in the contribution of lipid droplets to the total selected area within the different layers of the cortex were observed but the experimental groups supplemented with bee pollen exhibited a slight decrease in the vacuolization of both zona glomerulosa and zona fasciculata (Table 1).
The initial evaluation revealed possible differences in the diameter of the nuclei which was later confirmed by detailed measurements (Table 2). Next, the sinusoids were assessed. The differences in the sinusoid width between the groups were statistically significant (Table 2). Sinusoid epithelium thickness was decreased in both of the non-running experimental groups in comparison to the non-running control group. On the contrary, there was a tendency to increase the sinusoid epithelium thickness in the supplemented running groups compared to the running control (Table 2). The sinusoid epithelial cell nucleus diameter tended to increase in the bee-pollen-supplemented non-running group compared to the non-running control group. The capsule thickness increased in all four experimental groups in comparison to the control groups (Table 2).
## 3.4. Fibrosis
Neither visual evaluation nor computed analysis revealed significant fibrosis in any of the groups except for the bee-pollen-supplemented running group. In the latter, mild fibrosis was noted in all of the layers of the adrenal cortex. Additionally, the results of the computed analysis indicated a decrease in collagen fibers in the bee-pollen-supplemented non-running group in comparison to the non-running control group (Table 3).
## 3.5. Corticosterone Production
The urine corticosterone concentration significantly differed in all four experimental groups—both of the tested not-running groups (III—whey-protein-supplemented and IV—bee-pollen-supplemented) exhibited values lower than the non-running control, while both of the tested running groups (V—whey-protein-supplemented and VI—bee-pollen-supplemented) scored higher than the running control (Table 4). Moreover, in the whey-protein-supplemented groups, both in the running and non-running groups, lower values of urinary corticosterone level were observed in comparison to the bee-pollen-supplemented groups (Table 4). Furthermore, the non-running control group showed higher urinary corticosterone levels than the running control group (Table 4). No statistically significant changes in feces corticosterone concentration and total daily corticosterone excretion were noted (Table 4).
## 3.6. Pyknotic Nuclei
In all zones of the adrenal gland cortex, statistically significant changes in the percentage of pyknotic nuclei were observed in particular groups. In the non-running experimental groups, supplementation with both bee pollen and whey protein resulted in a lower mean percentage of the pyknotic nuclei in the cells in all of the adrenal cortex zones in comparison to the non-running control group (Table 5), (Figure 2).
## 4. Discussion
To the best of the authors’ knowledge, this is the first study on the effects of bee pollen and whey protein supplementation on adrenal function and structure. The impact of stress on eating behavior is a very complex phenomenon. Depending on the severity of stress and its duration, changes observed in dietary habits are substantially different. The efficient regulation of the whole process especially in the case of chronic stress is provided by the comprehensive action of the hypothalamic–pituitary–adrenal (HPA) axis. Stress affects both the amount and the tendency to eat specific foods as observed both in human and animal models [10,36,37]. While acute stress usually results in reducing food intake, exposure to chronic stressful stimuli leads to an increased desire to consume food. In addition, the regulation of food intake is influenced by the reward center, which in response to palatable food consumption decreases the activity of the HPA axis, resulting in the suppression of the stress response. It may explain why rats in stress are usually more likely to consume sugar-rich comfort products [38,39].
The nutritional trends in rats observed in the current study indicated no change in whey protein intake between the running and non-running groups of rats, whereas the difference in bee pollen intake between these groups was statistically significant. Rats in the non-running group tended to consume more bee pollen compared to the running rats. When analyzing the possible reasons for these observations, it is worth considering the differences in composition in these two groups of palatable food. Bee pollen has a much higher carbohydrate content than whey protein and it has been already reported that rats in stressful situations increase their intake of sugar-rich foods [39].
Consistent with the aforementioned findings, the body’s response to stress is inextricably linked to the functioning of the HPA axis. Especially adrenals, due to their high plasticity, are most significantly modified morphologically and functionally in response to stress [40]. Hence, this is why we decided to evaluate the adrenal glands for morphologic and functional aspects.
In the current study, no changes in the adrenal glands’ weight were observed between the groups. Our findings are consistent with several previous publications evaluating the effects of stress on adrenal mass. Marin et al., [ 2007] did not observe any changes in adrenal glands’ weight after exposure to stress or chronic restraint [41]. Similarly, it has previously been reported that exposure to chronic variable stress also did not induce changes in adrenal mass [42]. On the other hand, Díaz-Aguila et al., [ 2018] reported that the combined effect of stress and a high-sucrose diet on the rats caused an increase in the mass of the right adrenal gland. However, this study showed that these two variables separately did not affect adrenal weight which is partly consistent with our results [43]. Nonetheless, there are also reports contrasting with the above conclusions, indicating that stress can increase adrenal weight in rats [44,45,46]. Such an increase in adrenal weight may be due to stimulation of the adrenal glands by ACTH resulting in hypertrophy of the organ [46].
Exposure to stress in rats may lead to changes in the thickness of the adrenal cortex layers—both thinning and thickening are possible [43,47,48]. However, in the present study, microscopic evaluation of adrenal cortex layers in rats in particular groups did not show statistically significant differences.
We found out that the differences in the mean diameter of the cell nuclei in all of the cortex layers and in the medulla between all of the groups were statistically significant. Moreover, we observed a noticeable tendency to increase in the mean diameter of the cell nuclei in the zona glomerulosa and zona fasciculata both in the non-running supplemented groups compared to the non-running control as in the running supplemented groups compared to running control. Increased nucleus diameter may be a sign of increased protein synthesis and may also be related to increased hormonal secretion [49]. The findings previously o0btained from rat models have indicated that changes in nuclei structure are the result of experiencing stress. In a study conducted by Zaki et al., [ 2018], stress resulted in increased nuclear pyknosis [48] and our results are in line with these statements as we noticed that supplementation with our comfort foods (whey protein and bee pollen) lowered the number of pyknotic nuclei. Similarly, recent experiments have shown the protective effects of whey protein on hepatic cell nuclei [50] and propolis on olfactory bulb [51].
Although there are reports that acute stress induces adrenal fibrosis, in our experiment there is no evidence thereof [52]. Perhaps this is due to the fact that in our experiment, the stress situation was chronic. Nevertheless, we found mild fibrosis of all layers of the adrenal cortex in the group of running rats supplemented with bee pollen. Additionally, computed analysis showed that bee pollen supplementation in non-running groups resulted in a decrease in collagen fibers in comparison to non-running control group. Our results suggest a potential fibrosis-reducing effect exerted by bee pollen. This has been confirmed by previous reports suggesting that another bee product—bee bread—has the ability to reduce liver fibrosis induced by a high-fat diet [53].
Further microscopic analysis also provided observations on the degree of vacuolization of individual layers of the adrenal cortex in each group. The tendency to decrease the level of vacuolization accompanying reduced corticosterone levels according to Koko et al., [ 2004], can be the result of exposure to short-term stress [52]. The measurement of corticosterone levels in urine and feces is a sensitive marker reflecting the adrenal condition and the level of stress in the body of rats [54]. In the present study, significant differences in corticosterone concentrations were observed only in urine measurements, whereas no differences were found in measurements from stool samples. It has been proven previously that corticosterone concentrations in urine samples reflect the diurnal secretion profile of the hormone [55]. Thus, analysis of changes in urinary corticosterone levels captures a likely picture of stress levels in rats. The differences in the urinary corticosterone concentrations observed in the current study indicate that additional movement was effective in minimizing stress in rats. These results are in line with the reports that restriction of movement in rats caused an increase in blood corticosterone levels [56]. Similarly, our results corroborate those of a previous study in which immobilization and restriction to small space resulted in an increase of urinary corticosterone levels [57]. In addition, Premack et al., [ 1963], showed that depriving rats of their daily activities leads to an increase in food intake, confirming our previously analyzed model of stress as an inducer of changes in food intake [58].
Moreover, the results of the present study show that, compared to the non-running control group, the supplemented non-running rats had a reduction in urinary corticosterone excretion. This suggests that stress was decreased by both bee pollen and whey protein consumption by rats. The present findings corroborate previous research on the impact of high-protein high-carbohydrate comfort food on the level of stress in animals [31]. Previous studies have shown that royal jelly (another bee product) has the ability to lower plasma corticosterone levels [29,30]. Additionally, an experiment conducted by Teixeira et al., [ 2017] using royal jelly showed that this product has the ability to reduce corticosterone levels in rats also when they are not under stress [30]. Similarly, an experiment conducted on broilers proved that propolis, another product of bee origin, attenuates the endocrine component of the stress response by lowering corticosterone levels in broilers that were kept in stressful conditions [28]. Thus, the common anti-stress effect of bee products may be based on the activity of one of the proteins contained in them capable of inhibiting cholesterol synthesis and consequently inhibiting corticosterone synthesis [59]. Furthermore, based on the fact that bee pollen has a high carbohydrate content and that comfort food with a similar percentage carbohydrate content is able to lower serum corticosterone concentrations, we can speculate whether this contributed to our results [31,60,61]. On the other hand, the potential anti-stress effect achieved by carbohydrate-rich foods consumption is not supported by the results of another study conducted by Zeeni et al., [ 2015], in which two types of diets—high carbohydrate and high carbohydrate enriched with highly palatable products—were used in stressed rats. Indeed, rats supplemented with the latter one showed significantly lower serum corticosterone levels than rats eating only a carbohydrate-rich diet [44]. It can be concluded that not only the use of a high-carbohydrate diet but also its additional enrichment was responsible for the reduction of corticosterone concentrations. Since in the current study the intake of whey protein resulted in lower levels of hormone excreted in urine in comparison with bee pollen supplementation, we speculate that whey protein consumption is more effective in affecting adrenal function. Indeed, it has previously been reported that the production of corticosterone in rats is at least partially regulated by diet protein intake [62,63]. Additionally, Makkar et al., [ 2016] found that supplementation with $0.5\%$ whey protein caused a noticeable decrease in serum corticosterone levels in poultry [64] and similarly, Greco et al., [ 1982] observed that a high protein diet caused a decrease in serum corticosterone levels in rats [65]. Nonetheless, to understand the potential properties of whey protein on adrenal function, we need to look at its individual bioactive fractions such as lactoferrin and lactalbumin [22]. Maekawa et al., [ 2017] found that intraperitoneal administration of bovine lactoferrin to rats resulted in a decrease in serum corticosterone levels [66]. Furthermore, another prior experiment showed similar effects to lactoferrin administration [67].
Although our study is, to the best of our knowledge, the first one evaluating the impact of bee pollen and whey protein supplementation on adrenal histology and function, we are aware of its several limitations. First, we regard small sample sizes as a substantial restriction factor. Moreover, urine samples were collected while the rats were in metabolic cages. Although they were held in them for only 24 h, they might have been significantly stressed due to new housing and solitude. Additionally, the implementation of other stress sources in rats will allow us to draw more reliable results. What is equally important, the entire concept of the study topic followed the growing interest of functional foods consumption among people, as well as frequent experience of stress. Hence, in order to verify the effects of these substances on human organisms, future studies should be conducted on this target group.
## 5. Conclusions
The histological structure and functioning of the adrenal glands are a reflection of the impact of stress on the body. Application of exogenous factors including diet enrichment may modify some observed stress-derived changes. We noticed that bee pollen and whey protein have a significant effect on the reduction of urine corticosterone concentrations, which strongly supports their anti-stress value. Additionally, the observed decrease in the percentage of pyknotic nuclei in particular layers in both of the non-running supplemented groups in comparison to the non-running control may suggest the protective and beneficial effects of bee pollen and whey protein consumption on adrenal glands. Overall, the intake of bee pollen and whey protein seems to have limited but promising potential for stress reduction. Nevertheless, as we mentioned above, since the primary objective is to prove the effects of these substances on the human body, there is a need for further experiments.
## References
1. Nicolaides N.C., Kyratzi E., Lamprokostopoulou A., Chrousos G.P., Charmandari E.. **Stress, the Stress System and the Role of Glucocorticoids**. *Neuroimmunomodulation* (2015) **22** 6-19. DOI: 10.1159/000362736
2. Gjerstad J.K., Lightman S.L., Spiga F.. **Role of Glucocorticoid Negative Feedback in the Regulation of HPA Axis Pulsatility**. *Stress* (2018) **21** 403-416. DOI: 10.1080/10253890.2018.1470238
3. Smith S.M., Vale W.W.. **The Role of the Hypothalamic-Pituitary-Adrenal Axis in Neuroendocrine Responses to Stress**. *Dialogues Clin. Neurosci.* (2006) **8** 383-395. DOI: 10.31887/DCNS.2006.8.4/ssmith
4. Pacák K., Palkovits M.. **Stressor Specificity of Central Neuroendocrine Responses: Implications for Stress-Related Disorders**. *Endocr. Rev.* (2001) **22** 502-548. DOI: 10.1210/edrv.22.4.0436
5. Silva A.A., Perilhão M.S., Portes L.A., Serra A.J., Tucci P.J.F., Leopoldo A.S., dos Santos L., Bocalini D.S.. **Physical Exercise Attenuates Stress-Induced Hypertension in Rats but Not the Impairments on the Myocardial Mechanics**. *J. Hypertens.* (2021) **40** 528-535. DOI: 10.1097/HJH.0000000000003044
6. Yaribeygi H., Panahi Y., Sahraei H., Johnston T.P., Sahebkar A.. **The Impact of Stress on Body Function: A Review**. *EXCLI J.* (2017) **16** 1057. DOI: 10.17179/EXCLI2017-480
7. Barnthouse M., Jones B.L.. **The Impact of Environmental Chronic and Toxic Stress on Asthma**. *Clin. Rev. Allergy Immunol.* (2019) **57** 427-438. DOI: 10.1007/s12016-019-08736-x
8. Xu W., Tan L., Wang H.-F., Jiang T., Tan M.-S., Tan L., Zhao Q.-F., Li J.-Q., Wang J., Yu J.-T.. **Meta-Analysis of Modifiable Risk Factors for Alzheimer’s Disease**. *J. Neurol. Neurosurg. Psychiatry* (2015) **86** 1299-1306. DOI: 10.1136/jnnp-2015-310548
9. Yang T., Qiao Y., Xiang S., Li W., Gan Y., Chen Y.. **Work Stress and the Risk of Cancer: A Meta-analysis of Observational Studies**. *Int. J. Cancer* (2019) **144** 2390-2400. DOI: 10.1002/ijc.31955
10. Torres S.J., Nowson C.A.. **Relationship between Stress, Eating Behavior, and Obesity**. *Nutrition* (2007) **23** 887-894. DOI: 10.1016/j.nut.2007.08.008
11. Pecoraro N., Gomez F., Dallman M.F.. **Glucocorticoids Dose-Dependently Remodel Energy Stores and Amplify Incentive Relativity Effects**. *Psychoneuroendocrinology* (2005) **30** 815-825. DOI: 10.1016/j.psyneuen.2005.03.010
12. Groesz L.M., McCoy S., Carl J., Saslow L., Stewart J., Adler N., Laraia B., Epel E.. **What Is Eating You? Stress and the Drive to Eat**. *Appetite* (2012) **58** 717-721. DOI: 10.1016/j.appet.2011.11.028
13. Pecoraro N., Reyes F., Gomez F., Bhargava A., Dallman M.F.. **Chronic Stress Promotes Palatable Feeding, Which Reduces Signs of Stress: Feedforward and Feedback Effects of Chronic Stress**. *Endocrinology* (2004) **145** 3754-3762. DOI: 10.1210/en.2004-0305
14. Yeap S.K., Beh B.K., Ali N.M., Yusof H.M., Ho W.Y., Koh S.P., Alitheen N.B., Long K.. **Antistress and Antioxidant Effects of Virgin Coconut Oil in Vivo**. *Exp. Ther. Med.* (2015) **9** 39-42. DOI: 10.3892/etm.2014.2045
15. Yeap S.K., Beh B.K., Ali N.M., Mohd Yusof H., Ho W.Y., Koh S.P., Alitheen N.B., Long K.. **In Vivo Antistress and Antioxidant Effects of Fermented and Germinated Mung Bean**. *BioMed Res. Int.* (2014) **2014** 694842. DOI: 10.1155/2014/694842
16. Komosinska-Vassev K., Olczyk P., Kaźmierczak J., Mencner L., Olczyk K.. **Bee Pollen: Chemical Composition and Therapeutic Application**. *Evid. Based Complement. Alternat. Med.* (2015) **2015** 297425. DOI: 10.1155/2015/297425
17. Denisow B., Denisow-Pietrzyk M.. **Biological and Therapeutic Properties of Bee Pollen: A Review: Bee Pollen and Medicine**. *J. Sci. Food Agric.* (2016) **96** 4303-4309. DOI: 10.1002/jsfa.7729
18. Kocot J., Kiełczykowska M., Luchowska-Kocot D., Kurzepa J., Musik I.. **Antioxidant Potential of Propolis, Bee Pollen, and Royal Jelly: Possible Medical Application**. *Oxid. Med. Cell. Longev.* (2018) **2018** 7074209. DOI: 10.1155/2018/7074209
19. Khalifa S.A.M., Elashal M.H., Yosri N., Du M., Musharraf S.G., Nahar L., Sarker S.D., Guo Z., Cao W., Zou X.. **Bee Pollen: Current Status and Therapeutic Potential**. *Nutrients* (2021) **13**. DOI: 10.3390/nu13061876
20. Habryka C., Socha R., Juszczak L.. **Effect of Bee Pollen Addition on the Polyphenol Content, Antioxidant Activity, and Quality Parameters of Honey**. *Antioxidants* (2021) **10**. DOI: 10.3390/antiox10050810
21. Lagrange V., Whitsett D., Burris C.. **Global Market for Dairy Proteins: Global Market for Dairy Proteins**. *J. Food Sci.* (2015) **80** A16-A22. DOI: 10.1111/1750-3841.12801
22. Krissansen G.W.. **Emerging Health Properties of Whey Proteins and Their Clinical Implications**. *J. Am. Coll. Nutr.* (2007) **26** 713S-723S. DOI: 10.1080/07315724.2007.10719652
23. Patel S.. **Emerging Trends in Nutraceutical Applications of Whey Protein and Its Derivatives**. *J. Food Sci. Technol.* (2015) **52** 6847-6858. DOI: 10.1007/s13197-015-1894-0
24. Castro L.H.A., de Araújo F.H.S., Olimpio M.Y.M., Primo R.B.D.B., Pereira T.T., Lopes L.A.F., Trindade E.B.S.D.M., Fernandes R., Oesterreich S.A.. **Comparative Meta-Analysis of the Effect of Concentrated, Hydrolyzed, and Isolated Whey Protein Supplementation on Body Composition of Physical Activity Practitioners**. *Nutrients* (2019) **11**. DOI: 10.3390/nu11092047
25. Song X., Perez-Cueto F., Bredie W.. **Sensory-Driven Development of Protein-Enriched Rye Bread and Cream Cheese for the Nutritional Demands of Older Adults**. *Nutrients* (2018) **10**. DOI: 10.3390/nu10081006
26. Mahmoud U.T., Amen O.A., Applegate T.J., Cheng H.W.. **Brazilian Propolis Effects on Growth, Productivity Performance, Gut Characteristics and Physiological Changes in Broiler Chickens**. *Int. J. Poult. Sci.* (2017) **16** 169-179. DOI: 10.3923/ijps.2017.169.179
27. Mahmoud U.T., Abdel-Rahman M.A.M., Darwish M.H.A.. **Effects of Propolis, Ascorbic Acid and Vitamin E on Thyroid and Corticosterone Hormones in Heat Stressed Broilers**. *J. Adv. Vet. Res.* (2014) **4** 18-27
28. Chegini S., Kiani A., Rokni H.. **Alleviation of Thermal and Overcrowding Stress in Finishing Broilers by Dietary Propolis Supplementation**. *Ital. J. Anim. Sci.* (2018) **17** 377-385. DOI: 10.1080/1828051X.2017.1360753
29. Caixeta D.C., Teixeira R.R., Peixoto L.G., Machado H.L., Baptista N.B., de Souza A.V., Vilela D.D., Franci C.R., Salmen Espindola F.. **Adaptogenic Potential of Royal Jelly in Liver of Rats Exposed to Chronic Stress**. *PLoS ONE* (2018) **13**. DOI: 10.1371/journal.pone.0191889
30. Teixeira R.R., de Souza A.V., Peixoto L.G., Machado H.L., Caixeta D.C., Vilela D.D., Baptista N.B., Franci C.R., Espindola F.S.. **Royal Jelly Decreases Corticosterone Levels and Improves the Brain Antioxidant System in Restraint and Cold Stressed Rats**. *Neurosci. Lett.* (2017) **655** 179-185. DOI: 10.1016/j.neulet.2017.07.010
31. Ortolani D., Garcia M.C., Melo-Thomas L., Spadari-Bratfisch R.C.. **Stress-Induced Endocrine Response and Anxiety: The Effects of Comfort Food in Rats**. *Stress* (2014) **17** 211-218. DOI: 10.3109/10253890.2014.898059
32. Harvey S., Phillips J.G., Rees A., Hall T.R.. **Stress and Adrenal Function**. *J. Exp. Zool.* (1984) **232** 633-645. DOI: 10.1002/jez.1402320332
33. Zarobkiewicz M.K., Sławiński M.A., Wawryk-Gawda E., Woźniakowski M.M., Kulak-Janczy E., Korzeniowska S., Jodłowska-Jędrych B.. **Changes in Histological Structure and Nitric Oxide Synthase Expression in Aorta of Rats Supplemented with Bee Pollen or Whey Protein**. *Appl. Physiol. Nutr. Metab.* (2019) **44** 1150-1158. DOI: 10.1139/apnm-2018-0702
34. Zarobkiewicz M.K., Woźniakowski M.M., Wawryk-Gawda E., Sławiński M.A., Halczuk P., Korolczuk A., Jodłowska-Jędrych B.. **Decrease in Lipid Droplets in Adrenal Cortex of Male Wistar Rats after Chronic Exposure to Energy Drinks**. *Medicina* (2018) **54**. DOI: 10.3390/medicina54050090
35. Pihl L., Hau J.. **Faecal Corticosterone and Immunoglobulin A in Young Adult Rats**. *Lab. Anim.* (2003) **37** 166-171. DOI: 10.1258/00236770360563822
36. Maniam J., Morris M.J.. **The Link between Stress and Feeding Behaviour**. *Neuropharmacology* (2012) **63** 97-110. DOI: 10.1016/j.neuropharm.2012.04.017
37. Zellner D.A., Loaiza S., Gonzalez Z., Pita J., Morales J., Pecora D., Wolf A.. **Food Selection Changes under Stress**. *Physiol. Behav.* (2006) **87** 789-793. DOI: 10.1016/j.physbeh.2006.01.014
38. Adam T.C., Epel E.S.. **Stress, Eating and the Reward System**. *Physiol. Behav.* (2007) **91** 449-458. DOI: 10.1016/j.physbeh.2007.04.011
39. Ulrich-Lai Y.M., Fulton S., Wilson M., Petrovich G., Rinaman L.. **Stress Exposure, Food Intake and Emotional State**. *Stress Amst. Neth.* (2015) **18** 381-399. DOI: 10.3109/10253890.2015.1062981
40. Berger I., Werdermann M., Bornstein S.R., Steenblock C.. **The Adrenal Gland in Stress—Adaptation on a Cellular Level**. *J. Steroid Biochem. Mol. Biol.* (2019) **190** 198-206. DOI: 10.1016/j.jsbmb.2019.04.006
41. Marin M.T., Cruz F.C., Planeta C.S.. **Chronic Restraint or Variable Stresses Differently Affect the Behavior, Corticosterone Secretion and Body Weight in Rats**. *Physiol. Behav.* (2007) **90** 29-35. DOI: 10.1016/j.physbeh.2006.08.021
42. Kim Y.S., Lee M.Y., Choi C.S., Sohn Y.W., Park B.R., Choi M.-G., Nah Y.-H., Choi S.C.. **The Effect of Chronic Variable Stress on Bowel Habit and Adrenal Function in Rats**. *J. Gastroenterol. Hepatol.* (2008) **23** 1840-1846. DOI: 10.1111/j.1440-1746.2008.05524.x
43. Díaz-Aguila Y., Cuevas-Romero E., Castelán F., Martínez-Gómez M., Rodríguez-Antolín J., Nicolás-Toledo L.. **Chronic Stress and High Sucrose Intake Cause Distinctive Morphometric Effects in the Adrenal Glands of Post-Weaned Rats**. *Biotech. Histochem.* (2018) **93** 565-574. DOI: 10.1080/10520295.2018.1499961
44. Zeeni N., Bassil M., Fromentin G., Chaumontet C., Darcel N., Tome D., Daher C.F.. **Environmental Enrichment and Cafeteria Diet Attenuate the Response to Chronic Variable Stress in Rats**. *Physiol. Behav.* (2015) **139** 41-49. DOI: 10.1016/j.physbeh.2014.11.003
45. Martí O., Martí J., Armario A.. **Effects of Chronic Stress on Food Intake in Rats: Influence of Stressor Intensity and Duration of Daily Exposure**. *Physiol. Behav.* (1994) **55** 747-753. DOI: 10.1016/0031-9384(94)90055-8
46. Macedo I.C., Medeiros L.F., Oliveira C., Oliveira C.M., Rozisky J.R., Scarabelot V.L., Souza A., Silva F.R., Santos V.S., Cioato S.G.. **Cafeteria Diet-Induced Obesity plus Chronic Stress Alter Serum Leptin Levels**. *Peptides* (2012) **38** 189-196. DOI: 10.1016/j.peptides.2012.08.007
47. Oliveira M.J., Monteiro M., Ribeiro A., Pignatelli D., Aguas A.. **Chronic Exposure of Rats to Occupational Textile Noise Causes Cytological Changes in Adrenal Cortex**. *Noise Health* (2009) **11** 118. DOI: 10.4103/1463-1741.50697
48. Zaki S.M., Abdelgawad F.A., El-Shaarawy E.A.A., Radwan R.A.K., Aboul-Hoda B.E.. **Stress-Induced Changes in the Aged-Rat Adrenal Cortex. Histological and Histomorphometric Study**. *Folia Morphol.* (2018) **77** 629-641. DOI: 10.5603/FM.a2018.0035
49. Webster M., Witkin K.L., Cohen-Fix O.. **Sizing up the Nucleus: Nuclear Shape, Size and Nuclear-Envelope Assembly**. *J. Cell Sci.* (2009) **122** 1477-1486. DOI: 10.1242/jcs.037333
50. Saad Al-Dh I.. **Possible Protective Role of Whey Protein on the Rat’s Liver Tissues Treated with Nandrolone Decanoate**. *Pak. J. Biol. Sci.* (2018) **21** 262-274. DOI: 10.3923/pjbs.2018.262.274
51. Mahmoud O.M., Salem N.A., Al Badawi M.H.. **Protective Effect of Propolis on Manganese Chloride Neurotoxicity of Olfactory Bulb in Adult Male Albino Rat**. *Folia Morphol.* (2020) **79** 672-680. DOI: 10.5603/FM.a2019.0127
52. Koko V., Djordjeviæ J., Cvijiæ G., Davidoviæ V.. **Effect of Acute Heat Stress on Rat Adrenal Glands: A Morphological and Stereological Study**. *J. Exp. Biol.* (2004) **207** 4225-4230. DOI: 10.1242/jeb.01280
53. Zakaria Z., Othman Z.A., Suleiman J.B., Che Jalil N.A., Ghazali W.S.W., Nna V.U., Mohamed M.. **Hepatoprotective Effect of Bee Bread in Metabolic Dysfunction-Associated Fatty Liver Disease (MAFLD) Rats: Impact on Oxidative Stress and Inflammation**. *Antioxidants* (2021) **10**. DOI: 10.3390/antiox10122031
54. Thorpe J., Gould K., Borman E., deCatanzaro D.. **Circulating and Urinary Adrenal Corticosterone, Progesterone, and Estradiol in Response to Acute Stress in Female Mice (**. *Horm. Metab. Res.* (2014) **46** 211-218. DOI: 10.1055/s-0033-1363958
55. Thorpe J., Rajabi N., deCatanzaro D.. **Circadian Rhythm and Response to an Acute Stressor of Urinary Corticosterone, Testosterone, and Creatinine in Adult Male Mice**. *Horm. Metab. Res.* (2012) **44** 429-435. DOI: 10.1055/s-0032-1306307
56. Zylan K.D., Brown S.D.. **Effect of Stress and Food Variety on Food Intake in Male and Female Rats**. *Physiol. Behav.* (1996) **59** 165-169. DOI: 10.1016/0031-9384(95)02039-X
57. Sudo A., Miki K.. **Dissociation of Catecholamine and Corticosterone Responses to Different Types of Stress in Rats**. *Ind. Health* (1993) **31** 101-111. DOI: 10.2486/indhealth.31.101
58. Premack D., Premack A.J.. **Increased Eating in Rats Deprived of Running**. *J. Exp. Anal. Behav.* (1963) **6** 209-212. DOI: 10.1901/jeab.1963.6-209
59. Kashima Y., Kanematsu S., Asai S., Kusada M., Watanabe S., Kawashima T., Nakamura T., Shimada M., Goto T., Nagaoka S.. **Identification of a Novel Hypocholesterolemic Protein, Major Royal Jelly Protein 1, Derived from Royal Jelly**. *PLoS ONE* (2014) **9**. DOI: 10.1371/journal.pone.0105073
60. Ortolani D., Oyama L.M., Ferrari E.M., Melo L.L., Spadari-Bratfisch R.C.. **Effects of Comfort Food on Food Intake, Anxiety-like Behavior and the Stress Response in Rats**. *Physiol. Behav.* (2011) **103** 487-492. DOI: 10.1016/j.physbeh.2011.03.028
61. Corona-Pérez A., Díaz-Muñoz M., Rodríguez I.S., Cuevas E., Martínez-Gómez M., Castelán F., Rodríguez-Antolín J., Nicolás-Toledo L.. **High Sucrose Intake Ameliorates the Accumulation of Hepatic Triacylglycerol Promoted by Restraint Stress in Young Rats**. *Lipids* (2015) **50** 1103-1113. DOI: 10.1007/s11745-015-4066-0
62. Oi Y., Imafuku M., Shishido C., Kominato Y., Nishimura S., Iwai K.. **Garlic Supplementation Increases Testicular Testosterone and Decreases Plasma Corticosterone in Rats Fed a High Protein Diet**. *J. Nutr.* (2001) **131** 2150-2156. DOI: 10.1093/jn/131.8.2150
63. Oi-Kano Y., Kawada T., Watanabe T., Koyama F., Watanabe K., Senbongi R., Iwai K.. **Oleuropein Supplementation Increases Urinary Noradrenaline and Testicular Testosterone Levels and Decreases Plasma Corticosterone Level in Rats Fed High-Protein Diet**. *J. Nutr. Biochem.* (2013) **24** 887-893. DOI: 10.1016/j.jnutbio.2012.06.003
64. Makkar S.K., Rath N.C., Packialakshmi B., Zhou Z.Y., Huff G.R., Donoghue A.M.. **Nutritional Supplement of Hatchery Eggshell Membrane Improves Poultry Performance and Provides Resistance against Endotoxin Stress**. *PLoS ONE* (2016) **11**. DOI: 10.1371/journal.pone.0159433
65. Greco A.M., Boschi G., Valentino R.. **Evaluation of Aldosterone, Corticosterone, Glucose and Triglyceride Blood Levels in Pregnant Rats Fed a High Protein Diet**. *Boll. Soc. Ital. Biol. Sper.* (1982) **58** 115-120. PMID: 7066106
66. Maekawa Y., Sugiyama A., Takeuchi T.. **Lactoferrin Ameliorates Corticosterone-Related Acute Stress and Hyperglycemia in Rats**. *J. Vet. Med. Sci.* (2017) **79** 412-417. DOI: 10.1292/jvms.16-0498
67. Takeuchi T., Matsunaga K., Sugiyama A.. **Antidepressant-like Effect of Milk-Derived Lactoferrin in the Repeated Forced-Swim Stress Mouse Model**. *J. Vet. Med. Sci.* (2017) **79** 1803-1806. DOI: 10.1292/jvms.17-0200
|
---
title: Comparison of Nutri-Score and Health Star Rating Nutrient Profiling Models
Using Large Branded Foods Composition Database and Sales Data
authors:
- Edvina Hafner
- Igor Pravst
journal: International Journal of Environmental Research and Public Health
year: 2023
pmcid: PMC10002453
doi: 10.3390/ijerph20053980
license: CC BY 4.0
---
# Comparison of Nutri-Score and Health Star Rating Nutrient Profiling Models Using Large Branded Foods Composition Database and Sales Data
## Abstract
Front-of-package nutrition labelling (FOPNL) is known as an effective tool that can encourage healthier food choices and food reformulation. A very interesting type of FOPNL is grading schemes. Our objective was to compare two market-implemented grading schemes—European Nutri-Score (NS) and Australian Health Star Rating (HSR), using large Slovenian branded foods database. NS and HSR were used for profiling 17,226 pre-packed foods and drinks, available in Slovenian food supply dataset [2020]. Alignment between models was evaluated with agreement (% of agreement and Cohen’s Kappa) and correlation (Spearman rho). The 12-month nationwide sales-data were used for sale-weighing, to address market-share differences. Study results indicated that both models have good discriminatory ability between products based on their nutritional composition. NS and HSR ranked $22\%$ and $33\%$ of Slovenian food supply as healthy, respectively. Agreement between NS and HSR was strong ($70\%$, κ = 0.62) with a very strong correlation (rho = 0.87). Observed profiling models were most aligned within food categories Beverages and Bread and bakery products, while less aligned for Dairy and imitates and Edible oils and emulsions. Notable disagreements were particularly observed in subcategories of Cheese and processed cheeses ($8\%$, κ = 0.01, rho = 0.38) and Cooking oils ($27\%$, κ = 0.11, rho = 0.40). Further analysis showed that the main differences in Cooking oils were due to olive oil and walnut oil, which are favoured by NS and grapeseed, flaxseed and sunflower oil that are favoured by HSR. For Cheeses and cheese products, we observed that HSR graded products across the whole scale, with majority ($63\%$) being classified as healthy (≥3.5 *), while NS mostly graded lower scores. Sale-weighting analyses showed that offer in the food supply does not always reflect the sales. Sale-weighting increased overall agreement between profiles from $70\%$ to $81\%$, with notable differences between food categories. In conclusion, NS and HSR were shown as highly compliant FOPNLs with few divergences in some subcategories. Even these models do not always grade products equally high, very similar ranking trends were observed. However, the observed differences highlight the challenges of FOPNL ranking schemes, which are tailored to address somewhat different public health priorities in different countries. International harmonization can support further development of grading type nutrient profiling models for the use in FOPNL, and make those acceptable for more stake-holders, which will be crucial for their successful regulatory implementation.
## 1. Introduction
Strong evidences show that dietary factors are linked to the development of non-communicable chronic diseases [1,2]. Therefore, improving people’s diets is an important public health goal. In high-income countries, consumption of ultra-processed prepacked foods and drinks is increasing [3,4,5], while the nutritional quality of such products is often questionable [6].
Two key approaches for improving people’s diet are encouraging them to choose healthier foods (e.g., education, promotion) and the implementation of policies, which stimulate the development of foods with favourable nutritional composition (e.g., food reformulation policies, taxation, …) [7]. One way to address these approaches is through food labelling, representing the link between the consumer and the manufacturer. In recent years, different authorities emphasize the importance of front-of-package nutrition labelling (FOPNL) [8,9]. FOPNL should enable the consumer to distinguish quickly and clearly between foods according to their nutritional composition [10]. Such labelling can empower consumers to recognize differences between similar pre-packed foods and encourages manufacturers to reformulate their products [11], i.e., lower content of saturated fat, sugar, and sodium.
However, not all FOPNLs are equally effective. FOPNL schemes vary based on their graphic representation and parameters, which are included in the nutrient profiling algorithm. In recent times, emphasis has been placed on interpretive FOPNL schemes, which can subconsciously direct consumers towards healthier options (e.g., use of traffic light colours, rating, and heart symbol) [12,13]. Such labelling could support all consumers (regardless of their background nutritional knowledge) with making not only informed but also healthier food choices. The effectiveness of such schemes was supported by recent studies [14,15].
The European Commission (EC) is currently discussing the implementation of mandatory harmonized FOPNL across the EU region [16]. One type of interpretive FOPNL are overall grading schemes [10], which became interesting for public health policy makers as they provide simplified information (easier for interpretation), can be used on all products, and they avoid dichotomous thinking that foods can be either healthy or unhealthy [17]. Only two such schemes are currently used on the market: Nutri-Score (NS) in Europe, and Health Star Rating (HSR) in Australia. NS is an overall grading scheme developed in France, which grades foods on a scale from A-dark green (healthiest) to E-dark orange (least healthy) [18]. NS was developed with the considerations of the market; it is one of the most scientifically supported voluntarily FOPNLs, successfully implemented in several European countries, and therefore one of the candidates for the introduction of harmonized FOPNL in Europe. On the other hand, HSR is a voluntary scheme developed in Australia by the government in collaboration with the food industry and consumers groups. It grades foods in half-star increments from 0.5 * (least healthy) to 5 * (most healthy) and is commonly used in scientific research to evaluate the healthiness of foods and drinks [19]. Both schemes were built on the Ofcom Nutrient Profiling System, which was established to limit the marketing of unhealthy foods to children in the United Kingdom [20]. Even though both schemes have a very similar core system, adaptations made in the development process affected the way how food products are presented to consumers. For example, NS is 5-colour graded FOPNL, while HSR in the monochrome system with 10 possible star grades. Unlike NS, which only included minor changes to Ofcom profiling, HSR extended score scales for most of the attributes included in the profiling. The main differences between NS and HSR are described in Table 1.
Even though rating schemes can be easier for interpretation, some researchers are noting limited ability of FOPNLs for long-term changes into healthier dietary patterns, and emphasize the importance of consumer education. Some challenges were also highlighted for NS and HSR [21,22]. However, to our knowledge, no study so far did a comparison of these two rating systems across whole food supply. Such comparison would give us valuable data, highlighting the main challenges of grading schemes and opportunities for their improvement. Considering that FOPNLs have been particularly emphasized in the last few years, there is a lack of real-life data. Furthermore, the use of individual foods market-shares was shown to be useful for monitoring critical nutrients, but such an approach can be also used to examine what kind of products consumers are exposed to.
The aim of this study was to compare two grading FOPNL schemes—NS and HSR, which both aim to promote healthier foods and beverages to the consumer. We wanted to evaluate the alignment between these two systems and identify the main differences and related challenges, based on nationally representative branded food composition database. Market-share differences were addressed with the use of a sale-weighting approach, with consideration of 12-month sales data.
## 2.1. Data Sources
The study included cross-sectional data on available pre-packed foods from the Slovenian food supply (Slovenian branded foods dataset, 2020). The data source was the Composition and Labelling Information System (CLAS, Nutrition Institute, Ljubljana, Slovenia). The detailed methodology on data collection for CLAS is described elsewhere [23]. Briefly, CLAS includes the photo collection of labels of pre-packed foods from major retailers that cover the majority of the food market share in Slovenia. It includes all products with unique barcodes (most commonly in the form of the Global Trade Item Number—GTIN) that were available at the time of sampling. The 2020 dataset included 28.028 pre-packed foods from 7 stores all located in Ljubljana, Slovenia: two mega-markets (Interspar, Mercator Center), two supermarkets (Tuš, Spar), and three discount markets (Eurospin, Hofer, Lidl). Information on nutritional composition and ingredients were extracted from the photos.
Yearly product specific sales data for 2020 were obtained from retailers, which represent over $50\%$ of the national market share. With the help of GTIN barcodes, sales data were matched with the CLAS database. Using package size, we were able to calculate the quantity (kg/L) of product sold in a year, which was used for sale-weighting.
## 2.2. Product Categorization and Exclusion Process
Products in CLAS were categorized according to the international food categorization developed by the Global Food Monitoring group, to enable the monitoring of pre-packed foods and beverages [24]. The categorization was slightly modified according to European market specifics and study methodology. Product exclusion consisted of four steps. At first, we excluded foods from categories that are not covered by NS or HSR ($$n = 9$.250$). Then, we excluded products with missing data, which is a part of mandatory nutritional declaration ($$n = 881$$) and products that needed preparation, which would require the addition of other ingredients ($$n = 546$$). To avoid possible errors in the database, we also excluded products where calculated energy content (based on the content of nutrients) was differentiated from the labelled energy content for over ±$20\%$ ($$n = 125$$). These calculations were conducted with a protocol provided in the Regulation (EU) No. $\frac{1169}{2011}$ [25]. After the exclusion process, we ended up with 17,226 products from 12 main food categories and 53 subcategories.
## 2.3. Dealing with Missing Data
Since nutrient profiling requires some information, which is not part of mandatory food labelling, our dataset was supplemented with missing data for fibre, content of fruit, vegetables, nuts, and legumes (% FVNL), and calcium content.
The determination of missing dietary fibre and % FVNL content was performed as described elsewhere [26]. For dietary fibre, we defined food categories where fibre could be above 0.9 g/100 g, where the dietary fibre content becomes relevant for NS and HSR. The missing fibre content for relevant categories was then calculated for each product based on its energy value, content of other nutrients and their energy factors, using a previously described method [25]. fibre content was calculated for 2814 ($16\%$) products. The % FVNL was assessed using ingredient lists, and sometimes with consideration of the legislation [27,28]. We separately determined the content for specific oils important to NS (olive, walnut, and rapeseed oil), concentrated, dried and fresh FVNL, which was then converted into the total % FVNL by using the method described in the profiling protocol [29,30].
The calcium content was needed for the calculation of HSR for dairy products. When possible, the calcium content was extracted from the label; otherwise, it was assessed based on similar products (e.g., same type of cheese) available in national food composition database Open Platform for Clinical Nutrition (OPKP) [31]. Alternatively, the Fineli [32] database was used. For dairy imitates, the calcium content was only taken into account if the product had declared calcium content on the label.
Profiling exceptions were done with the use of food name and ingredient lists (i.e., plain water (for both models), water-based ice confections, minimally processed fruits and vegetables, and unsweetened flavoured waters (for HSR)).
## 2.4. Calculation of the NS Grade
Calculations of NS grades were performed in the alignment with Scientific and Technical instructions of French Public Health Agency [29]. At first, we assigned products to one of the four NS profiling categories (General, Beverages, Added fats, and Cheeses). Then, based on nutritional composition per 100 g or 100 mL, we assigned positive points (0–40) to negative attributes and negative points (0–15) to positive attributes. Negative attributes were energy (kJ), total sugars (g), saturated fatty acids (g), and sodium, while positive attributes include fibre (g), proteins (g), and % FVNL. The sum of positive and negative attributes gave the products a final score. Based on the final score, the product was graded from A (healthiest) to E (least healthy).
## 2.5. Calculation of the HSR Grade
Calculations of the HSR grades were performed according to the criteria provided by the Front-of-Pack Labelling Secretariat of the Australian Government [30]. Products were assigned to one of six HSR profiling categories: 1: Non-dairy beverages, 2: Foods, 3: Oils and spreads, 1D: Dairy beverages, 2D: Dairy foods, and 3D: Cheeses. Similar to NS, points were assigned to the same positive and negative attributes, but the amount of maximum points varied from 10 to 30 for each negative attribute and 8 to 15 for each positive attribute (depending on the profiling category). The sum of all negative and all positive points resulted in the product final score, based on which product was graded to one of ten HSR grades: from 5 stars (*) (healthiest) to 0.5 stars (*) (least healthy).
## 2.6. Data Analyses
Statistical analyses were performed using Microsoft Excel 2019 (Microsoft, Redmond, WA, USA) and R Studio (R Core Team, Vienna, Austria).
HSR has ten possible grades, while NS only has five. For easier comparison, we combined HSR grades into five, which were comparable to NS (HSR grades 5 * and 4.5 * were joined and compared to NS grade A, 4 * and 3.5 * with B, and so on). In line with previous studies, the cut-off value for healthier products was set at 3.5 * or more for HSR, and A/B for NS [33,34,35].
First, we assessed the distribution of different NS and HSR grades and overall healthiness of the Slovenian food supply. We compared both models based on available foods and based on sale-weighted distribution, and assessed their ability to discriminate products within a specific food (sub)category. The ability to discriminate between products was defined as good when there were at least three grades within a specific (sub)category [36].
The alignment between NS and HSR was assessed with the calculation of agreement (% of agreement and Cohen’s Kappa) and correlation (Spearman rho). At first, we calculated the % of agreement. If the model had sufficient discriminatory ability, we continued assessing agreement with Cohen’s Kappa. We have considered that models may not grade products equally high, but the way the products are ranked from the best to the worst can be similar. For this purpose, we also calculated the correlation with Spearman rho. Cut-off ranges for agreement and correlation were as follows (multiplied by 100 % for % of agreement): 0–0.20, negligible; 0.21–0.40, weak; 0.41–0.60, moderate; 0.61–0.80, strong; 0.81–1, very strong [33,37]. We then identified food categories that had the worst agreement and correlation by using all three approaches. These categories were subject to further categorization [38,39] and were then presented with descriptive statistics.
## 3. Results
Our final sample consisted of 17,226 pre-packed products. The largest share of the sample represented Dairy and imitates ($18.6\%$), followed by Bread and bakery products ($13.4\%$), Confectionery ($13.3\%$), Meat and meat products ($11.0\%$), Beverages ($10.6\%$), Fruits and vegetables ($8.4\%$), Sauces and spreads ($7.0\%$), Convenience foods ($4.4\%$), Snack foods ($3.5\%$), Edible oils and emulsions ($3.4\%$), Fish and fish products ($3.3\%$), and Cereal and cereal products ($3.2\%$) (Table 2).
## 3.1. Discriminatory Ability and Distribution of Grades across Food Supply
The overall distribution of NS and HSR grades of products in the Slovenian food supply is displayed in Figure 1. NS graded $10\%$ of products A, $12\%$ B, $20\%$ C, $32\%$ D, and $25\%$ E. HSR grades were slightly in favour of healthier foods: $12\%$ of products were graded 4.5–5 *; $21\%$, 3.5–4 *; $16\%$, 2.5–3 *; $22\%$, 1.5–2 *; and $29\%$, 0.5–1 *. It should be noted that, in HSR, products with a higher score (>20), which are otherwise associated with negative grades, can get a final grade of 4.5–5 *. This reflects the modifications of HSR for dairy and oils and spreads. Using criteria A/B for NS and 3.5 * for HSR [33,34,35], the NS model ranked $22\%$ of products in the Slovenian food supply as “healthy”, while HSR ranked $33\%$ of the products as healthy. The discriminating ability (3 or more different grades within a category) was good in all the main categories for both NS and HSR and in $91\%$ of subcategories for NS and $94\%$ for HSR. The discriminating ability was low in some smaller homogenous subcategories such as Water, Frozen fruit, and Frozen vegetables, for both NS and HSR, and for NS also in categories of Butter and Animal fat products.
The distribution of NS and HSR across main categories is displayed in Figure 2. In the main categories, NS was stricter in grading Convenience foods, Dairy and imitates, Fish and fish products, Meat and meat products, Edible oils and emulsions, and Snack foods. However, HSR was stricter than NS for the highest rated products (4.5–5 * or A) in categories of Cereal and cereal products, Convenience foods, Fish and fish products, Meat and meat products, and Sauces and spreads. HSR was also stricter for Confectionery, rating $69\%$ of the available foods with lowest grades 0.5–1 *, while NS rated $56\%$ with a grade E. The most notable difference in grading was observed in categories of Edible oils and emulsion and Dairy and imitates. For Edible oils and emulsions, HSR graded almost half ($48\%$) of the products as “healthy” while for NS, none of these products were assessed as healthy (highest grade was C). Similar results were observed for Dairy and imitates, where HSR assessed $56\%$ of products “healthy”, while NS only $35\%$. There were no notable differences in the rating of Beverages.
Differences become clearer, when results of modelling are compared for food subcategories (Supplementary Material S1), especially in Dairy and imitates. In Cottage cheese, HSR graded $73\%$ of products with 4.5–5 *, while NS rated less than half of the products ($47\%$) with A. In Cheese and processed cheese, NS in vast majority ($82\%$) rated products D, $12\%$ C, and a few products B ($1\%$) and E ($5\%$). HSR distributed such products across all grades: $35\%$ with grades 4.5–5 *; $28\%$, 3.5–4 *; $8\%$, 2.5–3 *; $6\%$, 1.5–2 *; and $22\%$, 0.5–1 *. NS showed good discriminating ability in the category of Plain yogurts, where full fat yogurts got a B ($38\%$), and skimmed yogurts got an A ($50\%$). HSR rated $83\%$ of products 4.5–5 *, with full fat yogurts mostly getting a 4.5 * and skimmed yogurts a 5 *. In the category of Flavoured yogurts, HSR rated products higher than NS, with $19\%$, 4.5–5 *; $54\%$, 3.5–4 *; $24\%$, 2.5–3 *; and $2\%$, 1.5–2 *. NS rated Flavoured yogurts mostly B ($46\%$) or C ($45\%$), and some products with A ($8\%$) and D ($1\%$).
In the main category of Convenience foods, NS was slightly stricter from HSR in rating Pizza; however, HSR was stricter when it comes to the highest grades (4.5–5 * or A) for Ready meals, Pre-prepared salads and sandwiches, and Side dishes.
For the subcategory of Cooking oils, NS was stricter and graded products with C ($51\%$), D ($40\%$), and E ($9\%$), while HSR distributed such products across all grades: $2\%$ with grades 4.5–5 *; $54\%$, 3.5–4 *; $33\%$, 2.5–3 *; $1\%$, 1.5–2 *; and $10\%$, 0.5–1 *. Another interesting category where we observed notable differences were Breakfast cereals, where HSR assessed $51\%$ of products as “healthy”, while for NS, this was $37\%$. In the category of Beverages, the results of profiling were most aligned, with some smaller differences. In the subcategory of Nectars, HSR graded some products better than NS due to the lower threshold for % FVNL. We also observed that HSR made notable differences in grading unsweetened flavoured waters (4.5 *), $100\%$ fruit juices with lower sugar content (<7 g per 100 mL) (4 *), and drinks with non-caloric sweeteners (3.5 *), while for NS, such drinks were mostly graded B.
## 3.2. Agreement and Correlation between Nutri-Score and Health Star Rating
The evaluation of alignment based on agreement and correlation between NS and HSR for subcategories is displayed in Figure 3, with exact values available in Supplementary Material S1. Overall, we determined strong agreement ($70\%$ and κ = 0.62) and very strong correlation (rho = 0.87) between both models. The percentage of agreement was perfect ($100\%$) for Waters, Frozen fruit, and Frozen vegetables, where both models rated all products with the highest possible grade. The percentage of agreement was also very strong (81–$100\%$) for 16 other subcategories, especially in the main categories of Beverages and Bread and bakery products and in other more homogenous subcategories such as Cereal flakes and bran ($94\%$), Butter ($93\%$), Jelly candy ($86\%$), and Milk and milk drinks ($82\%$). We observed that 13 subcategories had moderate to strong % of agreement and kappa, but very strong correlation, which indicates that models do not always agree on the grade, but still rank products similarly. This includes subcategories such as Breakfast cereals (rho = 0.83), Crispy bread (rho = 0.89), Cream imitates (rho = 0.85), Desserts (rho = 0.88), Yogurt imitates (rho = 0.90), Snack foods (rho = 0.82) and most subcategories in the main categories of Convenience foods and Sauces and spreads. We identified that NS and HSR have either very strong agreement or correlation in 32 ($60\%$) subcategories. A slightly lower agreement but still strong correlation was found in most subcategories of Fruit and vegetables and Meat and meat products. For Canned ($78\%$, κ = 0.53, rho = 0.71) and Dried fruit ($59\%$, κ = 0.43, rho = 0.79), HSR rated products were slightly stricter due to a higher penalization of sugars. For Nuts and fruit mixes ($48\%$, κ = 0.23, rho = 0.65) and Nuts and seeds ($50\%$, κ = 0.31, rho = 0.78), HSR graded products higher due to a higher contribution of positive components (fibre, % FVNL and protein), in comparison to NS. HSR was stricter in grading Unprocessed meat and Meat alternatives, and NS was stricter in grading Processed meat and meat products. Five categories with the lowest alignment between models were Cheese and processed cheese ($8\%$, κ = 0.01, rho = 0.38), Cream ($14\%$, κ = 0.03, rho = 0.26), Cooking oils ($27\%$, κ = 0.11, rho = 0.40), Margarine ($27\%$, κ = 0.13, rho = 0.72), and Unprocessed chilled fish ($54\%$, κ = 0.30, rho = 0.55). We observed that differences between models for Unprocessed chilled fish were due to extended HSR scale for protein. Most such products got maximum points for protein with NS, while for HSR protein, points differ substantially between products, making HSR a stricter profiling model for the highest grades (4.5–5 *). For Margarine, NS was much stricter (highest grade C) but the overall ranking/correlation was strong. The subcategory of Cream was a small and homogenous group, where most products were graded the same. NS graded most of these products ($86\%$) D and HSR ($81\%$) 0.5–1 *, which resulted in poor agreement. Two subcategories with the lowest alignment (and with enough diverse products for further analysis) were Cheese and cheese products and Cooking oils.
## 3.3. Main Differences in Grading with NS and HSR
Categories of Cheese and processed cheese and Cooking oils, where major differences were observed for profiling with NS and HSR, were subject to further descriptive analysis (Figure 4). After dividing products according to the type of oil/cheese, we could determine the type of products on which the models agree and disagree. In Cooking oils, the models were in alignment especially for coconut oil, pumpkin seed oil, mixed vegetable oil, and hemp seed oil. Smaller differences in grading were observed for special oils, sesame oils, and rapeseed oils, but the overall ranking stayed the same. The main differences were observed for olive oil and walnut oil, which are favoured by NS (mostly graded C) and grapeseed, flaxseed, and sunflower oil, which are favoured by HSR (mostly graded 3.5–4 *). For Cheeses and cheese products, we observed that HSR grades products across the whole scale, with the majority ($63\%$) being ranked as healthy (≥3.5 *). On average, products that were defined as healthy were soft, firm, and hard cheeses, while less healthy were extra hard cheeses, processed cheese, and cheese spreads. On the other hand, NS graded most of the products with lower grade D ($82\%$). The highest average grade had soft cheeses with some graded C ($$n = 69$$), and the lowest grade for extra hard cheeses. The most notable difference was observed for processed cheese and cheese spreads, with typical HSR grades 0.5–1 *, while NS graded them similar to other cheeses (mostly D).
## 3.4. Accounting for Market Share-Differences
To examine agreement between both profiling models with consideration of market-share differences of individual foods, we used 12-month national sales data for the year 2020. Sales data were available for $62\%$ ($$n = 10$$,724) of the products in our dataset, which were used for the assessment. Our results highlighted (Supplementary Material S1) that the food availability does not necessarily reflect the sales. This is not surprising in the food supply, where some market-leading foods can have magnitudes of higher sales in comparison to some niche and specialty products. The sale-weighing approach enabled us to account for these differences, prioritizing market-leading products. After sale-weighting, categories of Beverages, Bread and bakery products, Convenience foods, Dairy and imitates and Fruit and vegetables showed higher proportion of healthy products (NS A or B or HSR ≥ 3.5 *) in comparison to the product availability, while the contrary was observed for Confectionery, Cereal and cereal products, Fish and fish products, Meat and meat products, Snack foods and Sauces and spreads. The overall % of agreement between NS and HSR increased after sale-weighting from $70\%$ to $81\%$. The percentage of agreement between models was increased after sale-weighting for 8 out of 12 main categories and for 26 out of 53 subcategories. The subcategories where agreement notably decreased were Side dishes (from $67\%$ to $32\%$), Plain yogurt (from $53\%$ to $33\%$), Desserts (from $72\%$ to $49\%$), and Cooking oils (from $27\%$ to $6\%$).
## 4. Discussion
The current study showed that both the NS and HSR profiling models have good discriminatory abilities and are in very good alignment for most prepacked foods in the Slovenian food supply. However, there notable differences were observed in some subcategories, particularly in Cheese and processed cheese and Cooking oils.
Differences in some subcategories affected the overall grading of the food supply. Using both models, we saw that NS was the stricter model, grading $11\%$ less food supply as “healthy” in comparison to HSR. The evaluation of the healthiness of the products in the food supply is investigated in many studies, and is also an indicator of the food environment and efficiency of public health policies and interventions [40]. The diversity of nutrient profiling models used in research, and lack of a globally harmonized profiling model is limiting comparisons between countries and between different studies, because models were developed with different priorities [33]. Even for similar nutrient profiling models, such as NS and HSR, such challenges were noticeable, which emphasizes how the very choice of the profiling method affects the result. This should be considered when discussing the healthiness of the products in the food supply.
Overall, NS and HSR showed strong agreement and a very strong correlation. Similarities were observed, especially in categories of Beverages. In the year 2020, the HSR model was adapted for beverages, based on the NS algorithm, which explains very good alignment in this category. Alignment was also very good in categories of Bread and bakery products, Convenience foods, Sauces and spreads, Snack foods and for the majority of Dairy and imitates. For most categories, we observed that even though the grades were not always equally high, within the category ranking was comparable for both the models. Differences between NS and HSR mostly resulted from HSR extended scoring scales and adjustment of the profiling for specific food groups. The effect of extended scoring scales was particularly noticeable in products with high content of specific nutrients. For example, Nuts and seeds mostly have high protein and fibre content. Therefore, most products get maximum points for these attributes when profiling with NS. Using HSR, extended scales indicate such products can get more positive points, resulting in a greater impact of these attributes on the grade and overall better grading for this category using HSR ($85\%$ assigned as healthy). The same applies to negative points. Greater penalization of sugars and saturated fatty acids using HSR resulted in slightly stricter profiling, notable in Confectionery. Even though extended scales can be an advantage and address some of the issues that have already been highlighted for NS, caution is required when it comes to products, for which are very high in energy or some nutrient. Because of high fibre and protein, Nuts and seeds mostly get a HSR grade of 3.5 * or more (healthier products), even though some products have added salt or sugar. An extended scale for protein can also result in relatively high grades of Processed meat and meat products, a controversial category, for which limited consumption is recommended [41]. In Breakfast cereals, we observed that the inclusion of positive components, such as dietary fibre and protein, can “cancel out” the high amount of sugar to some extent. HSR, therefore, rated $51\%$ Breakfast cereals as healthy, while NS only $37\%$. While the higher number of star ratings allows HSR to still distinguish between products within a category, consumers might perceive higher (“healthier”) ratings differently. Pelly et al. [ 2020] explored consumers’ perception of HSR. They reported that even though consumers find HSR useful, some scepticism was highlighted towards the profiling algorithm, when it came to discretionary foods, which they felt were often rated too high [42]. Consumers’ trust and support is a key aspect that can determine efficiency and even survival of the specific FOPNL. Such an example is the Choices symbol, which was implemented in the Netherland and well known by more than $90\%$ of Dutch consumers. However, changes in the profiling algorithm, and particularly the use of modified symbol on discretionary foods, were not well understood among consumers and criticized extensively by consumer organizations. Loss of credibility with consumers and the government resulted in phasing out the logo in the Netherland [43]. This indicated that, in addition to discriminatory ability, it is also important to consider that grading of foods is aligned with the consumer’s logic.
Categories where grading is often exposed as questionable are dairy products and edible oils [21,44]. Our study also highlighted major differences in the nutrient profiling of Cheese and processed cheese and Cooking oils between NS and HSR. Differences in these categories were also found in a study, when a comparison was performed with the WHO Europe nutrient profiling system [26]. This reflects adjustments in the nutrient profiling, which were not established to the same extent in different models. For Cooking oils, NS kept the threshold for the final score based on the profiling of general foods, while adding positive points in the form of % FVNL for olive, walnut and rapeseed oils, which had the biggest impact on the overall grading. Meanwhile, HSR changed thresholds for the final grade, but grading is still based only on the nutritional composition, without consideration of the content of specific oil types. The changed thresholds resulted in HSR being a less strict model and the grading of oils being distributed over all grades, while the NS graded oils with C or lower, giving the preferences to olive, walnut and rapeseed oil (grade C) ahead of other oils (grades D and E). NS adjusted the system based on the French Agency for Food, Environmental and Occupational Health & Safety (Agence nationale de sécurité sanitaire de l’alimentation, de l’environnement et du travail—ANSES) nutritional recommendations, which highlighted the advantages of mentioned oils, such as higher content of omega 3 fats and polyphenols [45]. Many stakeholders suggested similar modifications for the HSR system. After a five-year review, the HSR remained the same with an argument that the HSR calculator does not consider components such as polyphenols, omega-3 fats, and vitamins, and it would be inappropriate to highlight these constituents only for the category of oils [46]. However, HSR does include calcium as an important component for profiling dairy products, but not in other categories. A customized profiling method for these products led to HSR having better discriminatory ability especially for Cheese and processed cheese, distributing products across all grades, while NS mostly graded such products with less favourable grade D. The inclusion of some micronutrients (i.e., potassium) and other components in nutrient profiling has been also encouraged recently by the European Food Safety Authority (EFSA) [47], but such modifications are challenging, when it comes to constituents that are not part of mandatory food labelling, and where estimations would not be reliable. Furthermore, nutrient profiling systems such as the recently introduced Food Compass [48], which includes 54 attributes, can be very complex for general use; a balance in including additional components is crucial for any model with the ambition of being implemented into food policies.
Using the sale-weighting approach, we observed that the availability of foods in the supply does not necessary reflect the sales. Moreover, adjusting for market-share differences by sale-weighting notably affected the agreement between models. Sale-weighting deepened the differences, especially in food categories with lower agreement between models. This was most notable in Cooking oils, where sunflower oils and mixed vegetable oils have the largest market share, but grades for these oils differ notably between models (mostly D for NS, and 3.5–4 * for HSR). Sales data included in our study only covered the number of products sold in a year, which did not involve information on what impacted the sales and purchasing decisions of consumers. Normalizing such data based on brand marketing, price of a product, product packaging, taste, etc. would not only give us further insight into consumers decision but would also be valuable for FOP research. It would be especially beneficial for examining the use of FOP as a marketing tool and how different FOP schemes agree in this context.
The European Scientific Committee recently issued a report, which suggested further updates of the NS algorithm [49]. Based on this report, NS will change the algorithm for solid foods more similarly to HSR. The scales will be extended for most nutrients, which should enable NS to be more aligned with nutritional recommendations. The update also addresses some challenges exposed in this study, for example for cooking oils, cheese, and nuts and seeds, which will be able to get a higher grade. The report also announced further changes in the category of beverages, which are still under revision and should be published soon. Even though we observed that NS and HSR are mostly aligned for beverages, differences were observed in the discriminatory ability of flavoured waters, some juices, and beverages with non-caloric sweeteners. Recent studies based on a French NutriNet-Santé cohort showed that the intake of non-caloric sweeteners could be potentially associated with increased cardiovascular risk and increased cancer risk [50,51]. Together with the increasing consumption of beverages with non-caloric sweeteners [52] and unfavourable WHO opinion in the draft guidelines [53], changes in the NS algorithm could be expected for these beverages.
The main advantages of this study are the use of a large representative dataset of prepacked products ($$n = 17$$,226) and the performance of a comparison of NS and HSR across the whole food supply, which was not examined before. Furthermore, access to 12-month nation-wide sales data on the product level enabled us to also account huge market-share differences in the food supply.
The study limitations also need to be noted. Nutritional composition data needed for nutrient profiling were not always available for all parameters. Therefore, data such as dietary fibre content, % FVNL, and calcium content were estimated with the use of previously established methods. Furthermore, while sales data were available for the majority of foods, this was not the case for some foods—particularly for those marketed in discount stores. To avoid methodological mistakes, we compared the distribution of the whole sample with the dataset for which sales data were available, and no notable differences were found. Such an approach was also used previously [26]. We should also mention that we combined ten HSR grades into five to enable more meaningful comparison between NS and HSR. This was done based on the previous methodology and was also considered in the interpretation of the results. Finally, we should mention that NS and HSR were developed for two different markets, in different continents, with somewhat different dietary patterns, food supply, and public health priorities. The selection of a tested branded food dataset (in our case, this was a branded food dataset, compiled in Slovenia as an European Union country) therefore affected the study results. Since both models are gaining interest in other markets (HSR in India [54], and NS in Latin America [55]), further studies in other food environments would be useful.
## 5. Conclusions
This study highlighted the high discriminatory ability of two grading FOPNL schemes (NS and HSR) in most food (sub)categories. Both FOPNL models are highly compliant, with a few divergences in some subcategories. We observed that even though these two models do not always grade products equally high, the trend of ranking is mostly similar. However, the observed differences highlighted some challenges in the rating systems. Major differences were observed in categories Cooking oils and Cheese and processed cheese. Sales data showed that sales do not always follow the offer, which sometimes resulted in deepened differences between FOPNL schemes after accounting market-share data. International harmonization can support the further development of grading type nutrient profiling models for use in FOPNL, and make those acceptable for more stake-holders, which will be crucial for their successful regulatory implementation as part of efficient and useful food labelling.
This study also highlighted challenges in monitoring pre-packed products and examining the use of FOPNL schemes on real-life data. Missing and non-sufficient data limit comparison between different FOPNLs, which is especially evident in more complex nutrient profiling schemes.
## References
1. Pagliai G., Dinu M., Madarena M.P., Bonaccio M., Iacoviello L., Sofi F.. **Consumption of ultra-processed foods and health status: A systematic review and meta-analysis**. *Br. J. Nutr.* (2020.0) **125** 308-318. DOI: 10.1017/S0007114520002688
2. **Global Strategy on Diet, Physical Activity and Health**
3. Juul F., Hemmingsson E.. **Trends in consumption of ultra-processed foods and obesity in Sweden between 1960 and 2010**. *Public Health Nutr.* (2015.0) **18** 3096-3107. DOI: 10.1017/S1368980015000506
4. Julia C., Martinez L., Allès B., Touvier M., Hercberg S., Méjean C., Kesse-Guyot E.. **Contribution of ultra-processed foods in the diet of adults from the French NutriNet-Santé study**. *Public Health Nutr.* (2017.0) **21** 27-37. DOI: 10.1017/S1368980017001367
5. Mertens E., Colizzi C., Peñalvo J.L.. **Ultra-processed food consumption in adults across Europe**. *Eur. J. Nutr.* (2022.0) **61** 1521-1539. DOI: 10.1007/s00394-021-02733-7
6. Martini D., Godos J., Bonaccio M., Vitaglione P., Grosso G.. **Ultra-Processed Foods and Nutritional Dietary Profile: A Meta-Analysis of Nationally Representative Samples**. *Nutrients* (2021.0) **13**. DOI: 10.3390/nu13103390
7. Temple N.J.. **A Comparison of Strategies to Improve Population Diets: Government Policy versus Education and Advice**. *J. Nutr. Metab.* (2020.0) **2020** 5932516. DOI: 10.1155/2020/5932516
8. **Guiding Principles and Framework Manual for Front-of-Pack Labelling for Promoting Healthy Diets**
9. **Farm to Fork Strategy**
10. Storcksdieck genannt Bonsmann S., Marandola G., Ciriolo E., van Bavel R., Wollgast J.. *Front-of-Pack Nutrition Labelling Schemes—A Comprehensive Review* (2020.0)
11. Roberto C.A., Ng S.W., Ganderats-Fuentes M., Hammond D., Barquera S., Jauregui A., Taillie L.S.. **The Influence of Front-of-Package Nutrition Labeling on Consumer Behavior and Product Reformulation**. *Annu. Rev. Nutr.* (2021.0) **41** 529-550. DOI: 10.1146/annurev-nutr-111120-094932
12. Vlaev I., King D., Dolan P., Darzi A.. **The Theory and Practice of “Nudging”: Changing Health Behaviors**. *Public Adm. Rev.* (2016.0) **76** 550-561. DOI: 10.1111/puar.12564
13. Talati Z., Pettigrew S., Kelly B., Ball K., Dixon H., Shilton T.. **Consumers’ responses to front-of-pack labels that vary by interpretive content**. *Appetite* (2016.0) **101** 205-213. DOI: 10.1016/j.appet.2016.03.009
14. Talati Z., Egnell M., Hercberg S., Julia C., Pettigrew S.. **Food Choice under Five Front-of-Package Nutrition Label Conditions: An Experimental Study across 12 Countries**. *Am. J. Public Health* (2019.0) **109** 1770-1775. DOI: 10.2105/AJPH.2019.305319
15. Egnell M., Talati Z., Galan P., Andreeva V.A., Vandevijvere S., Gombaud M., Dréano-Trécant L., Hercberg S., Pettigrew S., Julia C.. **Objective understanding of the Nutri-score front-of-pack label by European consumers and its effect on food choices: An online experimental study**. *Int. J. Behav. Nutr. Phys. Act.* (2020.0) **17** 146. DOI: 10.1186/s12966-020-01053-z
16. **Proposal for a Harmonized Mandatory Front-of-Pack Nutrition Labelling/before 2023-1**
17. Julia C., Fialon M., Galan P., Deschasaux M., Andreeva V.A., Kesse-Guyot E., Touvier M., Hercberg S.. **Are foods ‘healthy’ or ‘healthier’? Front of pack labelling and the concept of healthiness applied to foods**. *Br. J. Nutr.* (2021.0) **127** 948-952. DOI: 10.1017/S0007114521001458
18. Julia C., Hercberg S.. **Development of a new front-of-pack nutrition label in France: The five-colour Nutri-Score**. *Public Health Panor.* (2017.0) **3** 712-725
19. Dunford E.K., Ni Mhurchu C., Huang L., Vandevijvere S., Swinburn B., Pravst I., Tolentino-Mayo L., Reyes M., L’Abbe M., Neal B.C.. **A comparison of the healthiness of packaged foods and beverages from 12 countries using the Health Star Rating nutrient profiling system, 2013–2018**. *Obes. Rev.* (2019.0) **20** 107-115. DOI: 10.1111/obr.12879
20. Rayner M., Scarborough P., Lobstein T.. **The UK Ofcom Nutrient Profiling Model**
21. Katsouri E., Magriplis E., Zampelas A., Drosinos E.H., Nychas G.-J.. **Dietary Intake Assessment of Pre-Packed Graviera Cheese in Greece and Nutritional Characterization Using the Nutri-Score Front of Pack Label Scheme**. *Nutrients* (2021.0) **13**. DOI: 10.3390/nu13020295
22. Wellard L., Hughes C., Watson W.L.. **Investigating nutrient profiling and Health Star Ratings on core dairy products in Australia**. *Public Health Nutr.* (2016.0) **19** 2860-2865. DOI: 10.1017/S1368980016000975
23. Pravst I., Hribar M., Žmitek K., Blažica B., Koroušić Seljak B., Kušar A.. **Branded Foods Databases as a Tool to Support Nutrition Research and Monitoring of the Food Supply: Insights From the Slovenian Composition and Labeling Information System**. *Front. Nutr.* (2022.0) **8** 1077. DOI: 10.3389/fnut.2021.798576
24. Dunford E., Webster J., Metzler A.B., Czernichow S., Ni Mhurchu C., Wolmarans P., Snowdon W., L’Abbe M., Li N., Maulik P.K.. **International collaborative project to compare and monitor the nutritional composition of processed foods**. *Eur. J. Prev. Cardiol.* (2012.0) **19** 1326-1332. DOI: 10.1177/1741826711425777
25. **Regulation (EU) No 1169/2011 of the European Parliament and of the Council of 25 October 2011 on the Provision of Food Information to Consumers**
26. Hafner E., Pravst I.. **Evaluation of the Ability of Nutri-Score to Discriminate the Nutritional Quality of Prepacked Foods Using a Sale-Weighting Approach**. *Foods* (2021.0) **10**. DOI: 10.3390/foods10081689
27. **Council Directive 2001/112/EC of 20 December 2001 Relating to Fruit Juices and Certain Similar Products Intended for Human Consumption**
28. **Council Directive 2001/113/EC of 20 December 2001 Relating to Fruit Jams, Jellies and Marmalades and Sweetened Chestnut Purée Intended for Human Consumption**
29. **Nutri-Score Frequently Asked Questions—Scientific & Technical**
30. **Health Star Rating System: Calculator and Style Guide**
31. **OPKP (Open Platform for Clinical Nutrition)**
32. **Fineli—National Food Composition Database in Finland**
33. Pivk Kupirovič U., Hristov H., Hribar M., Lavriša Ž., Pravst I.. **Facilitating Consumers Choice of Healthier Foods: A Comparison of Different Front-of-Package Labelling Schemes Using Slovenian Food Supply Database**. *Foods* (2020.0) **9**. DOI: 10.3390/foods9040399
34. Dunford E.K., Huang L., Peters S.A.E., Crino M., Neal B.C., Ni Mhurchu C.. **Evaluation of Alignment between the Health Claims Nutrient Profiling Scoring Criterion (NPSC) and the Health Star Rating (HSR) Nutrient Profiling Models**. *Nutrients* (2018.0) **10**. DOI: 10.3390/nu10081065
35. Labonté M.-È., Poon T., Mulligan C., Bernstein J.T., Franco-Arellano B., L’Abbé M.R.. **Comparison of global nutrient profiling systems for restricting the commercial marketing of foods and beverages of low nutritional quality to children in Canada**. *Am. J. Clin. Nutr.* (2017.0) **106** 1471-1481. DOI: 10.3945/ajcn.117.161356
36. Dréano-Trécant L., Egnell M., Hercberg S., Galan P., Soudon J., Fialon M., Touvier M., Kesse-Guyot E., Julia C.. **Performance of the Front-of-Pack Nutrition Label Nutri-Score to Discriminate the Nutritional Quality of Foods Products: A Comparative Study across 8 European Countries**. *Nutrients* (2020.0) **12**. DOI: 10.3390/nu12051303
37. Prion S., Haerling K.A.. **Making Sense of Methods and Measurement: Spearman-Rho Ranked-Order Correlation Coefficient**. *Clin. Simul. Nurs.* (2014.0) **10** 535-536. DOI: 10.1016/j.ecns.2014.07.005
38. **General Standard for Cheese**
39. **Standard for Named Vegetable Oils**
40. Drewnowski A.. **Uses of nutrient profiling to address public health needs: From regulation to reformulation**. *Proc. Nutr. Soc.* (2017.0) **76** 220-229. DOI: 10.1017/S0029665117000416
41. 41.
International Agency for Research on Cancer
Red Meat and Processed MeatIARCLyon, France2018Volume 114. *Red Meat and Processed Meat* (2018.0) **Volume 114**
42. Pelly F.E., Swanepoel L., Rinella J., Cooper S.. **Consumers’ Perceptions of the Australian Health Star Rating Labelling Scheme**. *Nutrients* (2020.0) **12**. DOI: 10.3390/nu12030704
43. Roodenburg A.J.C.. **Nutrient profiling for front of pack labelling: How to align logical consumer choice with improvement of products?**. *Proc. Nutr. Soc.* (2017.0) **76** 247-254. DOI: 10.1017/S0029665117000337
44. Delhomme V.. **Front-of-pack nutrition labelling in the European Union: A behavioural, legal and political analysis**. *Eur. J. Risk Regul.* (2021.0) **12** 825-848. DOI: 10.1017/err.2021.5
45. **Recommendations Concerning Diet, Physical Activity and Sedentary Behaviour for Adults**
46. **Health Star Rating System Five Year Review Report**
47. **Scientific advice related to nutrient profiling for the development of harmonised mandatory front-of-packnutrition labelling and the setting of nutrient profiles for restricting nutrition and health claims on foods**. *EFSA J.* (2022.0) **20** 7259
48. Mozaffarian D., El-Abbadi N.H., O’Hearn M., Erndt-Marino J., Masters W.A., Jacques P., Shi P., Blumberg J.B., Micha R.. **Food Compass is a nutrient profiling system using expanded characteristics for assessing healthfulness of foods**. *Nat. Food* (2021.0) **2** 809-818. DOI: 10.1038/s43016-021-00381-y
49. **Update of the Nutri-Score Algorithm**
50. Debras C., Chazelas E., Sellem L., Porcher R., Druesne-Pecollo N., Esseddik Y., de Edelenyi F.S., Agaësse C., De Sa A., Lutchia R.. **Artificial sweeteners and risk of cardiovascular diseases: Results from the prospective NutriNet-Santé cohort**. *BMJ* (2022.0) **378** e071204. DOI: 10.1136/bmj-2022-071204
51. Debras C., Chazelas E., Srour B., Druesne-Pecollo N., Esseddik Y., de Edelenyi F.S., Agaësse C., De Sa A., Lutchia R., Gigandet S.. **Artificial sweeteners and cancer risk: Results from the NutriNet-Santé population-based cohort study**. *PLoS Med.* (2022.0) **19**. DOI: 10.1371/journal.pmed.1003950
52. Hafner E., Pravst I.. **The Sharp Rise in the Use of Low- and No-Calorie Sweeteners in Non-Alcoholic Beverages in Slovenia: An Update Based on 2020 Data**. *Front. Nutr.* (2021.0) **8** 778178. DOI: 10.3389/fnut.2021.778178
53. **WHO Draft Guideline: Use of Non-Sugar Sweeteners**
54. Bhatt N.. **Health experts cry foul as food industry targets India’s ripe market**. *BMJ* (2022.0) **377** o1382. DOI: 10.1136/bmj.o1382
55. Mora-García C.A., Tobar L.F., Young J.C.. **The Effect of Randomly Providing Nutri-Score Information on Actual Purchases in Colombia**. *Nutrients* (2019.0) **11**. DOI: 10.3390/nu11030491
|
---
title: 'Model Integration in Computational Biology: The Role of Reproducibility, Credibility
and Utility'
authors:
- Jonathan Karr
- Rahuman S. Malik-Sheriff
- James Osborne
- Gilberto Gonzalez-Parra
- Eric Forgoston
- Ruth Bowness
- Yaling Liu
- Robin Thompson
- Winston Garira
- Jacob Barhak
- John Rice
- Marcella Torres
- Hana M. Dobrovolny
- Tingting Tang
- William Waites
- James A. Glazier
- James R. Faeder
- Alexander Kulesza
journal: Frontiers in systems biology
year: 2022
pmcid: PMC10002468
doi: 10.3389/fsysb.2022.822606
license: CC BY 4.0
---
# Model Integration in Computational Biology: The Role of Reproducibility, Credibility and Utility
## Abstract
During the COVID-19 pandemic, mathematical modeling of disease transmission has become a cornerstone of key state decisions. To advance the state-of-the-art host viral modeling to handle future pandemics, many scientists working on related issues assembled to discuss the topics. These discussions exposed the reproducibility crisis that leads to inability to reuse and integrate models. This document summarizes these discussions, presents difficulties, and mentions existing efforts towards future solutions that will allow future model utility and integration. We argue that without addressing these challenges, scientists will have diminished ability to build, disseminate, and implement high-impact multi-scale modeling that is needed to understand the health crises we face.
## INTRODUCTION–THE PROMISE OF MODELING
Starting with the development of the SIR model in 1927 (Kermack et al., 1927), there has been a long history of population-level epidemiological modeling. The idea of being able to forecast biological phenomena computationally seemed very promising and as a result it has engaged researchers for a long time. Within time, many promises were made in the name of modeling, and enough effort was spent to warrant evaluating current capabilities, concepts, and hurdles.
Over the years, these studies have included the addition of population compartments, as well as the inclusion of age structure, vaccine, and quarantine, to name just a few advancements. Other studies have moved away from classical deterministic models to better understand the role of stochasticity. Moreover, this wide range of epidemic models has been used to study numerous diseases. Because of the vast amount of knowledge that has been gained through epidemic modeling over approximately the past 100 years, it was possible for researchers to quickly adapt their models to make predictions about the spread and control of the SARS-CoV-2 coronavirus.
However, unlike the population-level epidemic modeling effort, much less is known about how viral infections spread throughout the body, including its immune response and the response of different organ systems. Moreover, very little is known about the connection between infection at the individual scale and infection at the population scale.
Some examples like (Azmy et al., 2018) and (Ackleh et al., 2013) use bacteria or virus load as a feature parameter in partial differential equation compartment systems, where the progression of the disease among the population is linked to the virus load in infectious individuals.
Within-host models can explore how the variance of biology within the body can impact both a disease and its treatment (Zarnitsyna et al., 2021). Such mathematical models of viruses focusing on the in-host dynamics should be developed and utilized to a greater extent.
Multi-scale within-host modeling is common, with scales from the molecular and cellular levels integrated successfully with the larger whole-organ or whole-body scales (Powathil et al., 2012; Bowness et al., 2018), (Segovia-Juarez et al., 2004). There are fewer models, however, that successfully combine within-host models with population-level models. One barrier to the development of such models is the potential to lose the within-host granularity that is often seen when integrating to a higher scale. Many researchers may question if these multi-scale models truly offer new insight or if they are more informative when analyzed independently. Such questions emphasize the challenges of multi-scale model integration.
As we move from conceptual models (the diagrams and verbal models of biologists) to mathematical models to computer code, we gain in executability, but we lose shareability. Shareability requires multiple concepts: Reusability–someone else can run the same computer code, perhaps with different initial conditions or parameter values; Extensibility–someone can add modules or replace modules within the model without breaking it; Extractability–someone can select model components and use them, and they continue to function independently of their initial context; Portability–the model can be reused in a different computational instantiation from its original implementation. In particular, the knowledge embedded in computer code is generally stranded or lost, since you cannot easily infer the underlying conceptual model from the mathematical model or the mathematical model from its computer code. As a result, an essential aspect of model development is a formal process that begins with a detailed and complete specification of a fully sharable conceptual model, then develops a less sharable, but quantitative mathematical model, which is an interpretation of the biology and physics of the conceptual model, and finally a computer simulation, which implements the mathematical model in the form of specific algorithms and methodologies. At each step we need to define additional parameters and concepts.
One relatively new key concept enabling construction of models of complex phenomena is composition (Halter et al., 2020). Decomposition lets us break down a complex problem into simpler problems that can be solved or simulated and composition lets us systematically recombine these solutions into a solution of the original problem. To accomplish this, we need to think in terms of higher-order operations on models: what the models are is less important than what can be done with them. There are two kinds of composition: parallel and serial. Parallel composition means running models concurrently. This is useful for ensemble or consensus approaches that combine multiple models to arrive at a best estimate. Serial composition is when the output of one model becomes the input of another. It is important to think about the type of the model, what input it requires and what output it produces because compatibility is required for serial composition. Serial composition has been used to great effect, for example in whole cell models. Of course, it is possible to compose these combined models.
Combination of models can be from white box models or black box models. Black box models do not expose much information to the modeler, but even models that are composed of white-box models may suffer from transparency shortcomings due to the composition. In modular models, three different types of submodel couplings can be found: 1) black-box models with code-level coupling using information-hiding interfaces, 2) white-box models with code-level coupling and 3) white-box models with biological-level coupling. Compositionality and modeling has been extensively studied theoretically and the primitive operations, parallel and serial composition, explored in detail for certain classes of model (Baez and Master, 2020; Baez et al., 2021) that appear in fields as diverse as electronics engineering, chemistry, molecular biology (Danos and Laneve, 2004), plant biology (Honorato-Zimmer et al., 2018), infectious diseases (Halter et al., 2020; Waites et al., 2021) and economic game theory (Atkey et al., 2020). However, to address practical problems across scales, infrastructure is required. First, it is necessary to be able to discover models; models cannot be composed if they are unknown or unavailable. To do this, a catalog is needed with metadata about models and how to obtain them. We note the existence of mature software for data catalogs that is easily repurposed (CKAN, 2021). The models must be described sufficiently well to know if they can be composed, annotated with information about their input and output types. Annotations facilitate auxiliary tasks such as searching for appropriate models and ascertaining provenance. Finally, attention is needed to the detail of composition of a broad class of models, recognizing that errors introduced by (de)composition are only well-understood for some cases (Hairer et al., 2006; Blanes et al., 2008).
With so many multiscale modeling methods that are seemingly disjoint and mutually exclusive, recent efforts have sought to bring some order in the discussion of multiscale models of pandemics by providing a complete categorization of them (Garira, 2017; Garira, 2018). These publications identified five different categories of multiscale models of diseases that use different integration frameworks to integrate across scales. While this categorization cannot be claimed to be unique, it constitutes a good starting point, which may be found useful as a basis for further refinement in the discourse of multiscale modeling of pandemics. The paper (Garira, 2018) further identified ten of the most significant challenges that stand in the way of future advances in integration across scales in the development of multiscale modeling of disease dynamics. Collaborative research among scientists with different skills is needed to fully resolve these challenges.
The recent COVID-19 pandemic has highlighted the importance of modeling the disease and its potential vulnerabilities for interventions. Notable examples include (Russo et al., 2020) and (Russo et al., 2021) that develop models for vaccines. Between-host infection simulations helped researchers to make forecasts about the spread and potential control of the coronavirus. These models build from a century of work in population-level epidemiological modeling. While simple models have been attempting to assess potential antiviral drug combinations, very little is still known about the connection between infection at the individual scale and infection at the population scale (Dodds et al., 2021).
Indeed during the COVID-19 pandemic, many scientists working on related issues assembled under the umbrella of the Multiscale Modeling and Viral Pandemics Working Group (Multiscale Modeling and Viral Pandemics, 2021). This group is part of the MultiScale Modeling (MSM) Consortium hosted by the Interagency Modeling and Analysis Group (IMAG) (Interagency Modeling and Analysis Group, 2021). The discussions about modeling and its promise collected many ideas that were represented there, many times representing more than one opinion, creating a choir of voices. Among those voices, the group located many hurdles that might explain why the promise of modeling is not yet fully fulfilled–those are discussed below and organized considering the workflow of reproducing, gaining credibility, reuse, and integration.
## FROM REPRODUCIBILITY TO CREDIBILITY TO UTILITY TO INTEGRATION
Multi-scale models are intrinsically complex, and usually are modular, whereby the model is divided into units that interact with each other. Modularity facilitates component reuse and model integration, including the ability to exchange modules during, or between, simulations has many advantages (Petersen et al., 2014), but it depends both upon the validity of each individual module, as well as the ability to connect modules, so that they inter-operate appropriately. The promise of multiscale, modular modeling is that researchers can build from each other, using prior published models and building blocks for new, more accurate or more impactful models. For this vision to be achieved, 1) models must be reproducible, so that researchers are assured the module will perform as expected, 2) models must be credible, so that researchers are confident that reusing a module will be useful and appropriate, 3) models must be reusable, meaning not only can they reproduce published results, but also that they can be modified to fit new contexts, and 4) researchers must be able to integrate models with other models. Figure 1 shows these four steps schematically, including how each step depends on its predecessors.
The discussions in this group raised many issues that prevent model integration that start with inability to reproduce models, which leads to low credibility of those models, which reduces reuse, which leads to inability to combine and integrate larger more complex models. Therefore, we address many issues at lower levels that will help reach integration.
## The Reproducibility Crisis
Computational biomedical modeling involves mathematical representation of biological processes to study complex system behavior and was expected to be less affected by the reproducibility crisis. After all, computer software should be deterministic and therefore repeatable if designed well, compared to biological processes that have a random nature where experiments are not guaranteed to repeat themselves and many repetitions are required for the mean to converge.
However, models often fail to be reproducible and the reasons for the failure and prevalence are not fully understood. In a recent study (Tiwari et al., 2021), the BioModels group analyzed 455 kinetic models published in 152 peer-reviewed journals, a collective work of about 1,400 scientists from 49 countries. Most of these models were manually encoded from scratch to assess the reproducibility. Their investigation revealed that $49\%$ of the models could not be reproduced using the information provided in the manuscripts. With further effort, they managed to reproduce an additional $12\%$ either by empirical correction or support from authors. The other $37\%$ remained non-reproducible due to missing parameter values, missing initial concentrations, inconsistent model structure, or missing information. Among the corresponding authors they contacted less than $30\%$ responded. Models from many life science journals failed to be reproducible, revealing a common problem in the peer-review process. The group proposed an 8-point reproducibility scorecard to assess each model and address the reproducibility crisis. A similar study that reports similar deficiencies is reported in (Kirouac et al., 2019). The term crisis is not exaggerated and is well justified since the need to combine models together already exists and building blocks should be solid and match expectations. The “promise of modeling” should be fulfilled already, yet we find modelers consistently unable to reproduce basic steps—thus stagnating instead of innovating, or even worse—backtracking progress. The numbers quoted above cannot be ignored or claimed as being normal—thus the term crisis is used.
This shows only part of the larger crisis. The ideal we would like to reach is repeatability and reproducibility of models in publications and repositories. Repeatability is the ability to repeat the same experiment with the same system, and reproducibility is the ability to repeat the same experiment by another scientist without the same system. The goal here is to have all aspects of the simulation pipeline (Biological Model, Mathematical Model, Numerical Methods, Computational Implementation) be auditable, i.e., they should be fully described (as appropriate for each component) to enable results to be repeated and reproduced. However, the modeling community is far from this ideal. With the diagnosis of this problem, different groups aim for collaborative efforts to set up best practices.
It is important to acknowledge that the situation is much worse for other aspects of modeling. For any other aspect, it would be hard to even conduct such a study to evaluate reproducibility because the information often is not systematically cataloged let alone shared in a common format. Examples include how models were constructed, calibrated, or validated. Compared to the software industry or finite element modeling that have reached well established methods of exchanging information and repositories, such as file exchange formats/git, computational biological modeling has a long way to go and in this paper we will try to address some topics to be dealt with. If we cannot reproduce models, how can those be considered credible by other modelers, stakeholders, or even the public?
## Credibility of Models
In a larger context, model reproducibility is strongly tied to model credibility in light of a given purpose. A model without a purpose is a mere exercise suitable for the classroom and therefore the prerequisite for any realistic application is the design and testing of the model with that specific purpose in mind (often called the “Context of Use”, see below). Often, not only the model developer, but also others will need to confirm or challenge the credibility of a model and thus, a model built for a specific purpose must be at least repeatable and it is highly recommended it be reproducible. In turn, a model that is not repeatable, cannot be reproduced, and therefore cannot be deemed credible by others (who cannot understand the internals, especially if expert modelers cannot reproduce it). Therefore, the proof of model repeatability and reproducibility lies with the modeler who needs to prove the value of the model for a given application.
A modeler should consider the model’s purpose from the start of development and consider: for what; by who; what is the level of knowledge and skill of the user; and in what environment. If only the “for what” is specified, this implies the model can be used by anyone on any system, which broadens the scope and reduces the chance of reproducibility and hence reduces credibility by the potential user.
This proof of value becomes of utmost importance if critical decisions rely on a model. There are, in fact, many cases where human lives depend (directly or indirectly) on a model.
In most cases, users expect a system based on a model to be accredited somehow before use. In analogy, a physician will also not use a medical product, for example a medical device, that is not approved and tested for the anticipated use—because of the risk to harm the patient should the product fail or not work consistently.
This accreditation role many times falls on government agencies such as the FDA, or NASA. Those agencies have different approaches towards credibility of models. Guidelines issued by authorities regulating the use of such models give a “gold standard” recipe for how a modeler can ensure reproducibility and establish credibility of a model.
NASA takes modeling seriously. After the space shuttle Challenger disaster NASA rewrote a standard (NASA, 2016) and wrote guidelines (NASA, 2019). An interesting component in the NASA approach was a risk-adjusted approach that considered both the probability and consequences of a modeled systems failure, in which the level of risk raises or lowers the bar for the data needed to accredit a model. This approach also helps with cost.
FDA regulates the use of medical devices and drugs and also assesses computational models submitted as part of the market authorization dossiers. For many years, simulations have been part of these dossiers. If models can systematically shortcut and prevent issues related with long and costly clinical trials still needs proof, but the number of submissions to the FDA under the use of models has been constantly rising over the past years. The FDA has released and adopted guidance documents on the reporting and validation of computational models for regulatory submissions, (FDA, 2016; FDA, 2003), and actually considered the NASA standards when creating those (FDA, 2016). In 2018 the American Society of Mechanical Engineers (ASME) issued an important guidance ASME V&V 40 (ASME, 2018) of how to assess credibility of computational models of medical devices through verification and validation (V&V). The guideline is centered around the definition of the context of use (CoU) of the model, which is formulated based on the questions of interest the model will answer. The CoU is then analyzed in terms of the “model risk” - being the influence the model exerts on a decision and the potential consequences these decisions might incur. Commensurate with this model risk, the modeler establishes the credibility goals, performs verification validation and uncertainty quantification actions, and then assesses the outcome of this exercise in order to allow judging of the acceptability of the model CoU. Key to this guidance is its overarching nature that also allows adoption in other (e.g., drug development) fields irrespective of the model type (Viceconti et al., 2019; Kuemmel et al., 2020). Very recently, an FDA guidance draft has been updated taking into account this standard (FDA 2021b). In the paper by (Viceconti et al., 2019) the verification, validation and uncertainty quantification (VVUQ) pipeline is streamlined to different types of models. It is, perhaps, the closest to score credibility across model types from mechanistic physics-driven models to machine learning models. However, it is still short of including very recent developments such as ensemble models, although it touches upon the topic.
The FDA understood the potential value of models and modeled data to make developments of medical devices in an efficient manner. Early in the collaboration with the device industry about use of model data in trial processes, the FDA suggested having a “library of “reusable”, “regulatory grade” models. The FDA passed on the idea but is revisiting the library idea given models that meet the information guide for first accreditation. The idea is that the FDA understood the model and it proved useful, so they can accredit it much faster (cheaper) for reuse on a very similar application. Time will tell if this approach works.
In order to advance modeling and simulation fit for regulatory application, FDA and pharmaceutical companies engage in a model-informed drug development (MIDD) pilot program (FDA, 2021a). This pilot program was released in response to a performance goal agreed to under the sixth iteration of the Prescription Drug User Fee Act (PDUFA VI), included as part of the FDA Reauthorization Act of 2017 and advises how particular MIDD approaches can be used in a specific drug development program (Zineh, 2019), and how to report those complying with existing guidelines for regulatory submissions. While for MIDD, commonly data-driven or phenomenological models are used, more complex and multiscale models are coming of age and get submitted to regulators (Viceconti et al., 2021). Data from relevant MSM tools to refine, reduce or even replace trials, could provide additional economic incentives to sponsors (Galluppi et al., 2021). Likewise, however, especially complex models and MSMs models that do not meet the FDA requirements for credibility of model data, will fail to be considered. Guidelines, especially for complex multiscale models are still lacking and thus adoption of other guidelines e.g., ASME V&V40 is needed and discussion with regulatory agencies should be conducted before submission (Musuamba et al., 2021). Competitors focused on the regulatory side will out-compete those that do not, or cannot, comply. It will therefore be necessary to improve available guidance and standardization efforts with regards to repeatability, reproducibility, and reuse so that guidance can be adopted cross-community and entity, by academic and commercial ones alike.
Despite the importance of developing a model with a question of interest and the respective context of use (CoU) in mind, it is important to note that the past paradigm used towards model acceptance/credibility may change in the future. For example there could be multiple motivations for developing a model, motivations could change over time, and someone else could find a new use for a model that was intended for another purpose. Since some contributors to this manuscript have a less strict opinion regarding models being developed with a purpose in mind, we therefore recorded the range of opinions in this manuscript.
An example of a less strict approach to model credibility are new ensemble techniques, such as in (Barhak, 2016; Barhak, 2017), which allow judging a model by its performance in a group of models. This is similar to building teams in sports, where each individual contributes to a team and the value contributed to the team can be determined. Ensemble models allow assigning influence to single models and judging their performance by validation in different scenarios. Thus assigning a score to the model and its assumptions compared to others is possible. So the idea of credibility score may evolve through time and government agencies should consider this newer approach towards credibility. However, even if reproducibility and credibility are amenable there are many issues that prevent reuse of models.
While guidelines and concepts of how to establish credibility of models—even for critical applications—do exist, the field is still evolving and lots of work on completing, harmonizing, and adopting these guidelines still exist. One central question is what a minimal requirement might be for a model to be credibly re-usable.
## ONGOING DIFFICULTIES IMPEDING THE UTILITY AND INTEGRATION OF MODELS.
Since there is a large variety of known issues that prevent reuse and many solutions, we have divided them by topic. For simplicity Table 1 describes the difficulties and possible solutions: We also attempted to spread those difficulties as hurdles that relate to reproducibility, credibility, utility, and integration. Figure 2 depicts this analogy.
The paper continues elaborating on those topics and expands explanations hereafter.
## Built-In Barriers for Evaluating Model Credibility
If one considers models that exist currently, what impedes third parties in assessing their credibility according to “gold standard” guidelines (discussed before)? Most of them are not designed, too little or not transparently documented or supported with material allowing a third party for such credibility assessment.
Models include assumptions that need to be specified. Users need to know, under what conditions the model is appropriate? *This is* a question asked by any modeler. More provenance information is needed for reuse and composition. Another investigator who wants to expand a model may need to know what the assumptions or design decisions were so they know how to appropriately modify or expand a model. A regulatory body might want to be able to trace a model back to the data sources that informed it. Someone who wants to re-train a model for a different cell type or tissue might want to trace the data back to know what aspects of the training data need to be replaced. Enhancing model credibility can be achieved through enhancing documentation, establishing best practices, and tests.
Examples of information suggested to include in documentation are the design decisions that motivated a model, what the model is designed to explain/forecast, and explanations of data sources that contributed to a model. Ideally, this would include links to data repositories, indicate which assumptions were used to interpret the data, consider the methods/tools/users associated with model calibration, and evaluate if the model fulfills its intended purpose (Parker et al., 2002).
In addition, documentation should describe model limitations. It can be difficult to quickly determine which populations or scenarios a model can be reasonably applied to. This information can usually be teased out by carefully considering the data that has been used for fitting or validation, as well as digging through the discussion or conclusion. However, some doubt often remains because of the natural tendency to promote one’s work, and the, perhaps unrealistic, expectation that publishable work be as widely applicable as possible. If it were standard practice in model reporting to recommend specific model applications, this could provide clarity for those implementing or extending the model.
Beyond design and implementation, best practices should include reports of tests that describe what was simulated and the experimental or other data that was used to evaluate the test. Unlike software test reports which focus on failures, these reports must also focus on passes because they help establish the domain under which the model has been established to make trustworthy predictions. For example, this establishes the domain under which their clinical use would be supported. When considering test implementation, some suggestions emphasize the need for a structured approach with unit-test style tests (Sarma et al., 2016; Gerkin et al., 2019) (Lieven et al., 2020) and continuous evaluation of such tests similar to continuous integration of software (Meyer, 2014; Krafczyk et al., 2019; Zhao et al., 2017). To enhance model credibility further, the model description should include validation tests against independent data, uncertainty assessments, and peer reviews (Refsgaard et al., 2005; Jakeman et al., 2006).
## Models Are Written in Different Languages
When modelers do use a consistent, declarative language to describe their models, these models can then be stored and searched in readily-available repositories. The BioModels collection is a good example of such a repository for Systems Biology Markup Language (SBML) (Systems Biology Markup Language, 2021) models. As another example, the Physiome Model Repository (PMR) is a collection of CellML models. Although these repositories are a good step forward toward finding and reusing published models, by themselves, they are insufficient.
First, there are often significant differences between modeling languages–e.g., the CellML language and SBML are almost opposite in their approach to capturing the information in a model. Second, even within one modeling language it can be difficult for an outside user to understand the biological and mathematical content of a model written by someone else. As with software engineering, the key to enabling understandability and reuse of models is to provide unambiguous documentation about the intended semantics of the model.
One major problem we face for many kinds of models, which SBML and SBGN (Kitano, 2021) and projects like Biotapestry address partially for biological networks, is that we lack tools and formalism for consistently building, annotating, representing, displaying and manipulating conceptual models of complex biological phenomena with a spatial component. We lack standards for all of the key elements that need to be represented: the objects, the processes (behaviors and interactions) they participate in, the initial and boundary conditions and the dynamics and events that govern their evolution.
In many cases we also lack the scientific understanding of how to convert these conceptual models into mathematical models because we lack the “constitutive relations” that are the equivalent of the standard rate laws for chemical reactions. In this case we don’t have an agreed upon way to parametrize the submodels and to define their inputs and outputs.
Another big missing piece is a language to describe the possible experimental manipulations or perturbations of a biological system. We have concentrated on building mathematical and computational descriptions of biology, but not on the things we can do to them. Without such a description, classical techniques like perturbation and sensitivity analysis are much less useful. If we want to achieve a desired outcome by manipulating a given biological system, we need to know the constraints in our ability to manipulate that system. Knowing that we could achieve what we want by increasing the value of k_xx by $25\%$ is not actionable unless we can increase k_xx. The lack of orthogonality in biology (any perturbation of a biological system affects many aspects simultaneously, is what makes mathematical models so valuable for understanding (we have clean control parameters). But it also reduces their utility in designing experiments or clinical interventions. We need models that combine the model of the biological system with a model of the space of possible experiments. The sensitivity of this combined system is what tells us what is achievable in the lab or clinic.
Understanding the biological content of a model is critical to both reuse and reproducibility. If the model itself is incomprehensible, how can one know what its expected behavior and performance should be under different conditions? Semantic annotation is not necessary for simple repeatability, but if our goals include reproducibility and reusability, then we must make explicit and clear the biology and physics that underlie the model. An ideal modeling language should address this, yet until such a standard language is established we are faced with a need to integrate among different languages.
One simple integration example (Another example of integration and reproduction of a model, 2021) involving two popular languages, python and MATLAB, demonstrates the problem of transition between languages. There is no real translation between languages. *No* general compiler exists between multiple languages and human efforts are required. Fortunately, there are standardization efforts among languages.
The standardization problem is not new and was considered by modelers a long time ago, resulting in the Systems Biology Markup Language (SBML) (Systems Biology Markup Language, 2021) that is a very helpful format that can help transport models between systems. SBML has a track record of success and allows transporting models between hundreds of systems. However, despite its popularity, it is not an official standard and the community decided not to go in that direction (Recent SDO/COMBINE legal entity issues, 2021). Note that there are many similar community standardization efforts aggregated in the biosimulation modeling community known as COMBINE (Computational Modeling in Biology Network) (COMBINE, Online). COMBINE includes SBML as well as many other specifications, yet those communities are still in the process of standardization and need to organize legally.
Nevertheless, the lack of legal governance does not stop communities from developing even more tools for result handling and analysis like PETab (Schmiester et al., 2021), SED-ML (Waltemath et al., 2011), SESSL (Ewald and Uhrmacher, 2014), KiSAO (Courtot et al., 2011), SBRML (Dada et al., 2010), HDF5 (Folk et al., 2011), Vega (Satyanarayan et al., 2014; Satyanarayan et al., 2017), ggplot2 (Wickham, 2011), and others. Those tools show actual needs by the community, but these are much less mature and much less adopted. Their capabilities need to be expanded; they need to be adopted; software tools need to support them; and there needs to be infrastructure to share them, such as a repository. Another piece is that the software tools needed for the above are scattered, plus it is often unclear what subset of the above they support, and tools often become inaccessible. Tools need to be submitted to registries and the capabilities need to be annotated.
There is a need to coordinate the various standardization efforts that are needed for the different scales and biology that need to be involved in multi-scale models. The need for multiple standards may be recognized, yet the need to coordinate them to be able to compose multiscale models has received less attention.
## Models Are Hard to Locate
Many times model location is a difficult task since models are published in different sources. Despite many repositories available there are many ways models are published including journal papers, conferences, preprint services such as BioArxiv, web sites, and code repositories such as GitHub. In some good cases, there are model archive/linking web sites such as BioModels (BioModels, 2021), SimTK (SimTK, 2021), IMAGWiki (IMAGWiki, 2021), and in the future modeleXchange (Malik-Sheriff et al., 2020). However, currently there is no one aggregator that helps locate all models and many times community members cannot agree on location and attempt to create more repositories rather than centralize efforts.
Moreover, simulation workflows are even harder to find. For example, BioModels primarily focuses on models. There has been much less focus on publishing the construction/calibration of models, simulations, their results, analyses of their results, or entire workflows for the above. Sharing all of this needs embracing other repositories and developing some new ones.
We recommend that modelers use those repositories since we are at a stage in evolution of modeling where model composition is of interest and availability of modeling components is important. We ask that modelers consider from development to start permissive Intellectual Property for the versions published in those repositories to increase accessibility.
## Lack of Common Platforms for Executing Models and Simulations
Even if models can be located, their simulation is a different issue. Due to the existence of many partially supported standardization efforts in this field, it is often difficult to know what tool needs to be used with which model. It can also be difficult to find that tool, download it, install it, learn it, and to use it, especially for large simulations. These issues keep modelers in silos.
Even formats associated with standardization efforts have difficulties. It is not possible, for example, to load a MATLAB/SimBiology SBML L2V4 model into COPASI and someone adhering to latest standards implementing L3V2 SBML support for import will find difficulties in importing models. Moreover, it is difficult to find a common platform that supports all SBML versions. This version compatibility gap is not uncommon. However, since biological models take a long time to develop and represent phenomena that will persist for long terms, it is important to have long term stability and support with newer platforms and older models.
If the goal is for non-modelers to be able to interact with models (e.g., to analyze data, to contribute data toward a modeling project, or to apply a model for medicine), it needs to be much easier to find and use these tools. Two initiatives that are trying to address this are BioSimulators (BioSimulators, 2021) and runBioSimulations (Shaikh et al., 2021).
## Modeling Requires Adaptation Towards Integration
Many times the models as published need some level of manipulation to plug into another model. For example in (Castiglione et al., 2021) the survival function needs adaptation to transform it as can be seen from the public discussion in (About using a multi-scale mortality model in the ensemble, Online). Note that all those models need to be scaled to the same units and scales. Another example is in (Ke et al., 2020) where infectiousness is proportional to max infectiousness while the models in (Hart et al., 2021) are density models. In the model in (Castiglione et al., 2021) the time scale was originally 8 h and it needed to change to daily probability to merge into another model in (COVID19Models/COVID19_Mortality_Castiglione at main, 2021), which required scaling of the probability function. Those examples are relatively simple integrations and in more complex integrations the adaptation effort is more significant and many more obstacles exist.
One obstacle is lack of standards for describing composite models and software tools for merging models. One specification is SBML-comp, but it is cumbersome and few tools support it. Another tool is SemGen (Neal et al., 2019a), but it focuses on finding mappings between similar models. To the point here, SBML-comp is designed to compose models that were not intended to be composed. Instead, composition needs to be deeply ingrained into the entire community so that models are anticipating the needs of composition from the beginning.
Note that adaptation towards standardization also requires matching terminology, and especially matching of units of measure, as well as proper documentation which we will address in the next topics.
## Unit Standardization
Unfortunately, units of measure are not yet standardized, an open problem despite many attempts to resolve it by multiple standardization bodies such as IEEE, CDCIC, and NIST. One indication of the severity of the problem is that a Github search for “unit conversion” shows over a thousand results. Another good example of the severity of the problem is ClincialTrials. Gov that aggregates quantitative data from around the world and this database shows over 24 K different units of measure (Barhak and Schertz, 2019). One attempt at solving this standardization issue using machine learning is ClinicalUnitMapping.com, yet this project requires more effort.
Unit mismatches become particularly problematic when trying to integrate models across different spatial or time scales. For example, intracellular processes occur on micrometer spatial scales and seconds to minutes time scales. An in-host, tissue-level model of infection processes operates at millimeter to centimeter distances and hour to day time scales. When trying to integrate the two into a single multiscale model, care must be taken to ensure appropriate conversion of units when transferring output of one model as input to the other. This broad range of spatial and temporal scales can also cause computational problems requiring development of new algorithms to make computation more efficient across multiple scales (Jung and Sugita, 2017). While it is impossible to avoid having to convert units, we are advocating for clarity in the use of units. Sometimes models are used (and published) without specifying the units used for simulation parameters—this is a practice that needs to be corrected. Moreover, tools such as ClinicalUnitMapping.com can help overcome standardization difficulties in the future with more development. Once mapping to standardized units is easier, then simulation parameters can be converted appropriately when different scales or units are needed.
Lack of standard units for measurement of infectious virions is particularly problematic when trying to develop stochastic viral models. Stochastic models often require that we track individual infectious viral particles, yet it is not clear how the typical viral titer units of TCID50/ml and pfu/ml convert to individual virions. Two attempts have been made to estimate the conversion factor, both for influenza, resulting in estimates of 1 TCID50/ml of nasal wash corresponding to 102–105 (Handel et al., 2007) or 3 × 104-3 × 105 (Perelson et al., 2012) virions at the site of infection. Such order of magnitude uncertainty in unit conversion makes it difficult to develop accurate model representations of viral infections. We are not advocating for a standard conversion factor, since the conversion factor likely depends not only on the specific virus, but also conditions such as temperature and pH, which are known to affect viral infectivity (Rowell and Dobrovolny, 2020; Heumann et al., 2021). Rather, we are advocating for development of new viral measurement techniques that can more reliably quantify the number of infectious viruses present in a sample.
## Data Availability and Measurement Definitions
When attempting to integrate models, the phenomenon being reproduced by the models or the data they are based on might not be the same. This is especially true when model definitions evolve or can change in many ways. Examples include International Classification of disease (ICD) codes (Wikipedia, 2021a) that went through multiple versions through the years, or even a disease definition that has evolved for sepsis (Gary et al., 2021). Even outcomes of clinical trials change if counted using different definitions as seen in (clinicaltrials.gov/NCT00379769 - GlaxoSmithKline, 2017). Those definitions can hinder connecting different models together. Possible solutions are machine learning techniques that can transfer interpretation or modeling techniques that merge human interpretation from multiple experts into the modeling process (Barhak, 2020a).
Another issue specific to models dealing with viruses may seem like the lack of unit standardization for measurement of virus. However, it is a measurement definition issue. Infectious virus concentrations are measured using TCID50/ml ($50\%$ tissue culture infectious dose) or in pfu/ml (plaque forming units), both of which depend on specific experimental conditions such as temperature, humidity, and measurement time. Studies have shown that even a lab using identical experimental conditions cannot reproduce the same measured experimental values of virus leading to differences in estimated parameters for models (Paradis et al., 2015). There is also an underlying assumption for both units of measurement that an observed plaque was initiated by a single infectious virion, which has never been clearly proven to be true. More recently, non-infectious virus particle concentrations are being measured using PCR. In this technique, the number of segments of a particular piece of RNA are measured. While this unit is more tangible and consistent than the infectious viral titer units, viral kinetics models often consider only infectious virions. Although non-infectious viruses are starting to be incorporated into models, the relationship between infectious and non-infectious virions changes over the course of an infection (Petrie et al., 2013), making it difficult to use these measurements to get at the underlying infectious virus dynamics. New measurement techniques and strategies for more direct measurement of infectious virions are being developed (Cresta et al., 2021).
Data availability to rationalize calibration and validation of models is crucial but often not possible because of data sharing policy and privacy (especially for individual human data). Moreover, undisclosed data from industry sponsored clinical trials used in model building and validation generally excludes many useful models from any assessment by the scientific community. This is a difficult problem but there may exist some partial solutions. Synthetic data that is statistically similar to real-world data without containing information about any real individual can be shared. While the similarity can only be evaluated with access to the original data as in (Stack et al., 2013), because it can be shared, it can be used for calibration and validation of other models (Barhak, 2017; Ajelle et al., 2018; Liu et al., 2018; Wang et al., 2019). There is a risk of error using synthetic data in this way since, though it may have been similar in some respects to the original data, it might be different in some other respect that matters for a model different from that used for validation. For validation of model results against individual data that cannot be shared, it is conceivable that services could be deployed to query the data. Differential privacy (Dwork 2008; Garfinkel and Leclerc, 2020) to establish a privacy budget for such a service providing an information theoretic bound on how much information is allowed to be revealed in response to queries. This budget can be set to whatever is considered ethically and administratively acceptable. More research is needed to adapt this idea to suit model making needs.
Even data that are publicly available has limitations. In the case of within-host viral kinetics models, sampling of viral loads and immune responses is often not done frequently enough or long enough to ensure parameter identifiability (Miao et al., 2011). During the recent SARS-CoV-2 pandemic, several attempts were made to parameterize within-host viral kinetics models using viral loads measured from patients, but these measurements were often collected only after a patient was hospitalized, so the crucial viral growth phase is missing (Hernandez-Vargas and Velasco-Hernandez., 2020; Wang et al., 2020). Additionally, viral loads were measured via nasal swab, though it is not clear that the viral load in the nose is correlated to the viral load in the respiratory tract, which is the infection location simulated in the models. Other viruses can also infect internal organs that are difficult to access for frequent measurements without invasive procedures. Immune responses are often measured using levels in the blood, which is typically not the site of infection and often not the location of viral load measurements either. Since models are attempting to replicate virus and immune dynamics at the site of infection, these data collection limitations make it difficult to collect the data needed to accurately parameterize such models for humans.
A further methodological issue with some of the available in vitro and in vivo experimental data is that it often does not represent the infection time course in a single individual or single experimental well. Clinical trial data is often presented as medians or means taken over all patients. A recent study of influenza infections showed that parameter estimates based on fits of models to a single median viral titer curve do not match estimated parameter values based on fits to individual patients (Hooker and Ganusov, 2021). This also masks patient-to-patient variability in infection. Pre-clinical animal studies and in vitro studies can be even worse as animals are often sacrificed and infections in individual wells are stopped to make measurements at each time point. In this case, experimental data consist of an average of measurements from several animals/wells at each time point that differ from the set of animals/wells at other time points.
## Missing Annotations in Models
In biosimulation models, documentation about the intended semantics of the model is captured by annotations - additional information that describes the model, and the biological entities included in that model. Further, these annotations can leverage the rich resources of bio-ontologies–consistent nomenclatures and terminologies that describe the biological world in great detail.
COMBINE has recognized these challenges for understandability and reuse of models and is working hard to disseminate best practices around semantic annotation. COMBINE consensus around annotation is described in (Neal et al., 2019b). This paper describes some key tenants for improved semantic annotation: First, these annotations should be written using a standard format, and one that is independent of modeling languages. Thus, COMBINE recommends RDF as a simple triple-based representation to connect model elements to annotations and knowledge resources (e.g., ontologies). Next, COMBINE recommends that annotations should be stored externally from the source code of the model. Obviously, the annotations should be linked to elements within the model source code, but in order to be language independent, they should be stored separately. Finally, COMBINE recommends that modelers and model building communities provide policies and rationale for choosing which knowledge resources to use for which types of annotations. Otherwise, the same biological entity may look different if different modelers annotate the entity against different bioontologies.
Annotations can be useful for multiple tasks such as: annotation of semantic meaning where biology or real-world relevance is explained, annotation of provenance where the origins of the model and its creators are referenced, and annotation of verification indicating tests the model should undertake and pass. However, there is a lack of sufficient annotation about the components of models. This is particularly because modelers choose not to provide annotation and because tools for describing the semantic meaning of components are just starting to emerge. For example, for biochemical models there’s HELM and BpForms. The lack of such annotation makes it hard to determine the points of overlap between models. In addition, there is a lack of annotation about data sources and assumptions which makes it hard to determine whether models are compatible or what needs to be done to make them compatible. For example, do two models represent the same cell type, tissue, or gender? Hopefully policies will be adopted to resolve this issue.
## Models Are Not Consistently Licensed in an Easy Way That Allows Reuse
Different institutions have different approaches towards licensing, as can be seen from this discussion (Licensing issues, 2021). Therefore, model creators may not be aware of the implications of licensing many times when they publish their models. Moreover, some licenses are incompatible with each other or other forms of Intellectual Property (IP) such as patents (Wikipedia, 2021b). Even open source licenses are quite restricted since they are based on copyright laws, which give the owner rights to restrict usage (Barhak, 2020b). In this sense open source licenses resemble patents and in some cases are more restrictive since patents become public domain quicker. Moreover, community members take different sides with regards to licensing issues as can be seen in this discussion (Issues with regard to Call for transparency of COVID-19 models, 2021). Specifically, one license that will make reuse much easier is Creative Commons Zero (CC0) (CC0 Creative Commons, 2021). This license uses the term “No rights reserved” and makes it easier for models and text to be reused with less restrictions. In fact model repositories such as BioModels require releasing the models uploaded there under CC0 (Malik-Sheriff et al., 2020). However, CC0 license has not been adopted by some (Wikipedia, 2021c). To eliminate the licensing problem, modeling communities will have to abandon old school open source licenses that are based on copyright and create conflicts and recommend releasing models to the public domain using licenses such as CC0.
We therefore recommend that models and their associated data should be published under permissive terms. For maximizing reproducibility and integration, we suggest that the most permissive license possible should be chosen. In that regard the CC0 license would be a good choice, effectively waiving interests of the creator in their works and therefore emulating the public domain in jurisdictions where this is necessary.
## Different Scales and Modeling Paradigms
Models are operating on different spatial scales (population or individual) with different modeling paradigms (continuous vs discrete). A tissue could be modeled as a continuum leading to Partial Differential Equations (PDEs) or as a collection of individual interacting cells leading to an agent-based model. Specification of these two models would probably require different languages.
The fact that models capture different scales or that they don’t consistently capture any single scale creates challenges for composition. One opinion is that the challenge is that the scale of a model is not clearly annotated. To compose models, this forces the composing investigator to try to figure out the scales of each model and how to mesh them. Typically this is combined with lack of annotation of units. When developing a standard for specifying models, developers will probably need standards specific for each modelling approach. One possibility is that a family of specification standards may be created. The need for different formats for different domains and scales will probably create the need for a central place where, especially non-modelers, can find information about these various future standards and which tools support them. Ideally, there would also be a central place where these tools can be obtained and executed so that even non-modelers can easily explore models without having to figure out what software is needed, install it, etc. However, it is still unclear what common practices might facilitate composition across scales and how the various component standards should be architected to facilitate integration.
## Model Application and Implementation Barriers
Models are difficult to be used by a community or government. Scientific, regulation, and social communities have different sets of models and different understanding and standards in models. It is hard to convince and establish a common popular model widely acceptable by a wide range of communities and even adopted by the government. The long term validation and approval process may delay the cycle from model application to implementation.
Many models can increase their utility to the scientific community if their applicability and implementation is easy. These aspects can be improved by means of several features. One crucial feature is the reproducibility of the model since usually this is necessary before the model can be applied by the scientific community. The likelihood that a model is applied to different problems increases if their results are reproducible. The reproducibility is difficult to test if there are implementation barriers. Some of these barriers have been pointed out in this article. Thus, we can infer that implementation issues of the models affect the reproducibility of the models and therefore a broad utility of the models to the scientific community.
Currently many models are difficult to implement and therefore unable to make a real impact. Many of the existing models that are used by decision makers are used because those were implementable. More sophisticated models are many times not used due to a need for proper tools or proper expertise. Therefore, many good ideas remain unused due to implementation difficulties. The solution to this problem is long term and requires education of developers, users, and the public.
## Stochastic Modeling Difficulties
Biological systems are exceptionally complex, involving a multitude of interactions among a large number of components at different spatial and temporal scales. Over the years, much work has been performed wherein deterministic models have been developed to understand dynamics from the cellular level to the population level (Murray, 2002). Although these works have provided much insight, it is known that the mean-field dynamics of these deterministic models do not always capture important phenomena (Forgoston and Moore, 2018). For example, disease population models often have a stable endemic state for reproduction numbers greater than one, and therefore it is not possible for the disease to go extinct in the models. This is in direct contrast to the local extinctions of disease that occur all the time in the real-world (Assaf and Meerson, 2010; Bauver et al., 2016), (Doering et al., 2005; Dykman et al., 2008; Ovaskainen, and Meerson, 2010; Forgoston et al., 2011; Schwartz et al., 2011; Nieddu et al., 2017; Billings and Forgoston, 2018). Similarly, at the within-host level, new infections may or may not establish. This phenomenon can be captured by stochastic models, but is not realised by a deterministic model with a single set of model parameters. Furthermore, deterministic models do not account for the random interactions of cells or individuals, nor do they account for the changes in the model’s rates which are related to random events.
To properly model real-world multiscale dynamics, it is often necessary to use stochastic approaches that allow one to make quantitative, statistical predictions, while simultaneously providing qualitative descriptions of system dynamics. The ability to generate stochastic simulations that provide quantitative statistics for the emergence of new dynamics is increasing with advances in computational power. However, the inclusion of stochasticity leads to a variety of issues related to reproducibility. One concern is associated with noise-induced transitions (Assaf and Meerson, 2010; Forgoston and Moore, 2018) or stochastic resonance in which the deterministic system is qualitatively different from the stochastic system. The output of a stochastic model is a distribution or time-series of distributions. In particular, it is possible to have different outcomes for the same model or set of parameter values (e.g., a multimodal equilibrium distribution). In this case, one must use more sophisticated techniques such as Kullback-Leibler divergence or Wasserstein’s distance to make quantitative comparisons with data or other models. Moreover, while it may be possible to compare distributions generated by different stochastic models, it is often not possible to generate identical, individual realisations.
Stochastic models are critically important. Indeed, unlike deterministic models, stochastic models give rise to probabilistic predictions based on ensembles of realisations. However, in general, these types of models present difficulties that include: 1) how one validates a stochastic simulation; and 2) how one can ensure the repeatability of a stochastic simulation. It is worth noting that the latter issue becomes more problematic when software libraries that support modern high-performance computation (including standard parallel computation as well as GPU computation meant to accelerate simulation), cannot guarantee deterministic reproducibility (DOC, 2021). Potential solutions include the development of tools that guarantee repeatability such as ways to set and record pseudo-random number generator seeds as used by the MIcro Simulation Tool (MIST) (Barhak, 2013) and developing standards to address stochastic simulations.
## Open Discussion
During the work on the paper, a few other items were raised that were not resolved and we assembled those here so those can be addressed in a future version. Some of those issues are visible, yet out of reach, at least for this group of authors and the group invites correspondence on resolving the issues we list below.
How existing Multi-Scale frameworks could be made more transparent with respect to the models they encode and potentially more interoperable. How are most multiscale models encoded? Do they use general modeling frameworks that have been developed for this purpose or do they rely on custom-built code? *What* general purpose multiscale frameworks exist and what types of model integration and linking do they support? Are there opportunities to develop standards that would make such models more reproducible and understandable? What intermediate steps might be possible? For example, some of these frameworks support integration of SBML models (or models encoded in other somewhat standardized languages). Is it beneficial to leverage existing standardization efforts and to develop new ones for the description and implementation of MSM’s going forward? Right now, it is a bit of a wild west where these models are being developed with submodels as generic pieces of code in general purpose languages like C++ and Python that rely only on unstructured comments for documentation.
It is still unclear how various component standards should be architected to facilitate integration. Many of the solutions discussed will require dedicated tools and standards and there will be many of those. It is unclear how distributed and decentralized components will be governed. A resource such as modeleXchange (Honorato-Zimmer et al., 2018) may help investigators navigate this landscape for models, and similar tools may be needed to navigate tools and standards.
There is no common methodology on how to deal with the gap between model parameters, data collection, and standards. This topic should be discussed in the future in light of a potential solution of standardized model development explanation. However, standards have to evolve to handle this issue in proper standard organizations.
The COMBINE (COMBINE - Coordinating standards for modeling in biology, 2021) organization has been helpful to start the standardization process. However, it is not a Standards Development Organization (SDO) and its members rejected joining SISO, which was an established SDO. Therefore the products of this work may not be widely accepted unless the organization matures and adopts a legal entity standing with all regulations involved. However, will the community behind this organization mature enough to adopt legal bindings and regulations?
Common formats for results and visualizations still have not been established despite importance. When models produce results during simulation those should be archived and visualized somehow to help user interactions. A common format to represent results and how to generate graphics to represent will help with credibility and integration efforts.
For models to mature, there is a need to establish testing paradigms. Testing included all general software development good testing practices such as verification of calculation code, regression testing, plus model validation and usability testing. One of the authors recommended that a future model testing practices paper be developed. Another voice mentioned that the ensemble modeling approach includes tests within it. Yet the group reached no conclusion.
Barriers preventing validation reduce credibility and therefore have a negative impact on model credibility, which prevents reuse. One barrier is model validation at different scales: Models at different scales from molecular to population scale are usually validated at different standard and testing samples. Cross-scale validation is very difficult since there are multiple factors involved that influence the outcome of different scales. Another barrier is model validation for practical use cases: Real world prediction from the developed model is challenging because of the complexity of the pathogen spreading process. The real spreading process always has a lot of random social and physiological variables that are hard to be included in any model. With more advanced models and availability of more data, practical forecasts will get more accurate.
Another, difficult topic is spatial models. It seems we are pretty much at the beginning in terms of defining spatial models. We don’t have a quantitative language to specify cell shapes, cell behaviors, or tissue architecture. In many cases we don’t even have a qualitative language to do this.
Bioscience modelers are not alone dealing with utility and reuse related issues. A 2016 report on complex systems engineering challenges (Fujimoto et al., 2017) identified other non-technical barriers in the form of social, behavioral and programmatic barriers that were not addressed among the technical issues in this paper.
These and many other topics may be issues for the group to discuss in the future and readers are welcome to join the discussion.
## CONCLUSION
This manuscript discussed the reproducibility crisis in biological computational models. Many issues and difficulties and barriers have been presented. Nevertheless, some efforts towards solutions already are in progress and have been mentioned. We can categorize those challenges to scientific problems, and cultural and community based challenges.
Examples of scientific problems include the need to build good model integration environments, and to establish how to integrate models across paradigms. We also need to resolve many of the stochastic modeling challenges.
Examples of cultural and community barriers include education towards standardization of units, education towards proper annotation of models. Good tools may help with education by helping do those tasks semi automatically.
It is expected that many solutions will have both scientific and cultural aspects. The list of issues should not discourage modelers from developing models. Instead modelers should view this list as a reference of issues to be solved in the future and issues to avoid. The first step in solving the problem is admitting it exists. With this paper the authors recognize the challenges and admit the current state of modeling needs fixing. Hopefully fixing those issues, starting with reproducibility, will increase model credibility and will facilitate reuse and later integration of models. The long term goal of this group is improving models to achieve better human and machine comprehension of biological processes.
## FUNDING
JK was supported by Grant NIH-NIBIB P41EB023912 “Center for Reproducible Biomedical modeling”. WW is supported by UK Medical Research Council MRC grant MR/V$\frac{027956}{1.}$ RMS is supported by EMBL-core funding from themember states of the European Molecular Biology Laboratory. RB was supported by a fellowship funded by the Medical Research Council MR/P$\frac{014704}{1.}$ JG acknowledges funding support from grants NSF 188553, NSF 186890, NSF 1720625, NIH U24 EB028887 and NIH R01 GM122424.
## Conflict of Interest
AK was employed by the Novadiscovery SA. JB: Payment/services info: JB reports non-financial support and other from Rescale, and MIDAS Network, other from Amazon AWS, Microsoft Azure, MIDAS network, other from The COVID tracking project at the Atlantic, other from JR and Jered Hodges, other from Anaconda. Financial relationships: JB declare(s) employment from U.S. Bank/Apexon, MacroFab, United Solutions, B. Well Connected health and Anaconda. Intellectual property info: JB holds US Patent 9,858,390 - Reference model for disease progression issued to JB, and a US patent 10,923,234 - Analysis and Verification of Models Derived from Clinical Trials Data Extracted from a Database. Other relationships: During the conduct of the study; personal fees from U.S. Bank/Apexon, personal fees from MacroFab, personal fees from United Solutions, personal fees from B. Well Connected health, personal fees and non-financial support from Anaconda, outside the submitted work.
The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
## References
1. About using a multi-scale mortality model in the ensemble (Online) (2021). MSM Viral Pandemic Integration Subgroup Mailing List Archives. Available at: https://lists.simtk.org/pipermail/vp-integration-subgroup/2021-January/000011.html (Accessed
October 24, 2021).. *MSM Viral Pandemic Integration Subgroup Mailing List Archives* (2021)
2. Ackleh AS, Ma B, Thibodeaux JJ. **A Second-Order High Resolution Finite Difference Scheme for a Structured Erythropoiesis Model Subject to Malaria Infection**. *Math. Biosciences* (2013) **245** 2-11. DOI: 10.1016/j.mbs.2013.03.007
3. Ajelli M, Zhang Q, Sun K, Merler S, Fumanelli L, Chowell G. **The RAPIDD Ebola Forecasting challenge: Model Description and Synthetic Data Generation**. *Epidemics* (2018) **22** 3-12. DOI: 10.1016/j.epidem.2017.09.001
4. Another example of integration and reproduction of a model (2021). MSM Viral Pandemic Integration Subgroup Mailing List Archives
Available at: https://lists.simtk.org/pipermail/vp-integration-subgroup/2021-January/000012.html (Accessed
Nov 1, 2021).. *MSM Viral Pandemic Integration Subgroup Mailing List Archives* (2021)
5. ASME (2018). Assessing Credibility of Computational Modeling through Verification & Validation: Application to Medical Devices - ASME
Available at: https://www.asme.org/codes-standards/find-codes-standards/vv-40-assessing-credibility-computational-modeling-verification-validation-application-medical-devices (Accessed
October 24, 2021).. *Assessing Credibility of Computational Modeling through Verification & Validation: Application to Medical Devices - ASME* (2018)
6. Assaf M, Meerson B. **Extinction of Metastable Stochastic Populations**. *Phys. Rev. E* (2010) **81** 021116. DOI: 10.1103/PhysRevE.81.021116
7. Atkey R, Gavranović B, Ghani N, Kupke C, Ledent J, Forsberg FN. *Compositional Game Theory, Compositionally* (2020)
8. Ackleh Azmy S., Sutton Karyn, Zhao Lihong L.. **A Second Order Finite Difference Scheme for a Variable Infection-Structured Model of mycobacterium Marinum Dynamics in Aquatic Animals**. *J. Nonlinear Variational Anal* (2018) **2** 2. DOI: 10.23952/jnva.2.2018.2.06
9. Baez JC, Genovese F, Master J, Shulman M. **Categories of Nets**. *arXiv:2101.04238* (2021)
10. Baez JC, Master J. **Open Petri Nets**. *Math. Struct. Comp. Sci* (2020) **30** 314-341. DOI: 10.1017/S0960129520000043
11. Barhak J. *MIST: Micro-Simulation Tool to Support Disease Modeling; SciPy 2013 Presentation* (2013)
12. Barhak J. *The Reference Model for Disease Progression Combines Disease Models* (2016)
13. Barhak J. *The Reference Model: A Decade of Healthcare Predictive Analytics (PyTexas 2017)* (2017)
14. Barhak J. *The Reference Model for Disease Progression Handles Human Interpretation* (2020a)
15. Barhak J. *Open Source and Sustainability* (2020b)
16. Barhak J, Schertz J. *Clinical Unit Mapping with Multiple Standards* (2019)
17. Bauver M, Forgoston E, Billings L. **Computing the Optimal Path in Stochastic Dynamical Systems**. *Chaos* (2016) **26** 083101. DOI: 10.1063/1.4958926
18. Billings L, Forgoston E. **Seasonal Forcing in Stochastic Epidemiology Models**. *Ricerche mat* (2018) **67** 27-47. DOI: 10.1007/s11587-017-0346-8
19. BioModels (Online) (2021). Available at: https://www.ebi.ac.uk/biomodels/(Accessed
October 24, 2021).. (2021)
20. BioSimulators (Online) (2021). Available at: https://biosimulators.org/(Accessed
October 24, 2021).. (2021)
21. Blanes S, Casas F, Murua A. **Splitting and Composition Methods in the Numerical Integration of Differential Equations**. *arXiv:0812.0377* (2008)
22. Bowness R, Chaplain MAJ, Powathil GG, Gillespie SH. **Modelling the Effects of Bacterial Cell State and Spatial Location on Tuberculosis Treatment: Insights from a Hybrid Multiscale Cellular Automaton Model**. *J. Theor. Biol* (2018) **446** 87-100. DOI: 10.1016/j.jtbi.2018.03.006
23. Castiglione F, Deb D, Srivastava AP, Liò P, Liso A. *From Infection to Immunity: Understanding the Response to SARS-CoV2 through In-Silico Modeling* (2021)
24. CC0 Creative Commons (2021). Available at: https://creativecommons.org/share-your-work/public-domain/cc0/(Accessed
October 24, 2021).. (2021)
25. CKAN (Online) (2021). The Open Source Data Management System ckan.Org.
Available at: http://ckan.org/(Accessed
October 24, 2021).. *The Open Source Data Management System* (2021)
26. **A Long Term, Open Label, Randomised Study in Patients With Type 2 Diabetes, Comparing the Combination of Rosiglitazone and Either Metformin or Sulfonylurea With Metformin Plus Sulfonylurea on Cardiovascular Endpoints and Glycaemia**. (2017)
27. COMBINE (Online) - Coordinating standards for modeling in biology (2021). COMBINE.
Available at: http://co.mbine.org/(Accessed
October 24, 2021).. *COMBINE (Online) - Coordinating standards for modeling in biology* (2021)
28. Courtot M, Juty N, Knüpfer C, Waltemath D, Zhukova A, Dräger A. **Controlled Vocabularies and Semantics in Systems Biology**. *Mol. Syst. Biol* (2011) **7** 543. DOI: 10.1038/msb.2011.77
29. COVID19Models/COVID19_Mortality_Castiglione at main (Online) (2021). Jacob-Barhak/COVID19Models GitHub
Available at: https://github.com/Jacob-Barhak/COVID19Models (Accessed
October 24, 2021).. *Jacob-Barhak/COVID19Models GitHub* (2021)
30. Cresta D, Warren DC, Quirouette C, Smith AP, Lane LC, Smith AM. **Time to Revisit the Endpoint Dilution Assay and to Replace TCID**. *PLoS Comput. Biol* (2021) **17** e1009480. DOI: 10.1371/journal.pcbi.1009480
31. Dada JO, Spasić I, Paton NW, Mendes P. **SBRML: a Markup Language for Associating Systems Biology Data with Models**. *Bioinformatics* (2010) **26** 932-938. DOI: 10.1093/bioinformatics/btq069
32. Danos V, Laneve C. **Formal Molecular Biology**. *Theor. Computer Sci* (2004) **325** 69-110. DOI: 10.1016/j.tcs.2004.03.065
33. DOC (2021). Clearly Indicate Deterministic Algorithms for Reproducible Science Issue #2685 Rapidsai/cuml GitHub (Online). Available at: https://github.com/rapidsai/cuml/issues/2685 (Accessed
October 24, 2021). *Clearly Indicate Deterministic Algorithms for Reproducible Science Issue #2685 Rapidsai/cuml GitHub* (2021)
34. Dodds MG, Krishna R, Goncalves A, Rayner CR. **Model-informed Drug Repurposing: Viral Kinetic Modelling to Prioritize Rational Drug Combinations for COVID-19**. *Br. J. Clin. Pharmacol* (2021) **87** 3439-3450. DOI: 10.1111/bcp.14486
35. Doering CR, Sargsyan KV, Sander LM. **Extinction Times for Birth-Death Processes: Exact Results, Continuum Asymptotics, and the Failure of the Fokker--Planck Approximation**. *Multiscale Model. Simul* (2005) **3** 283-299. DOI: 10.1137/030602800
36. Dwork C, Agrawal M, Du D, Duan Z, Li A. *Theory and Applications of Models of Computation Lecture Notes in Computer Science* (2008) 1-19. DOI: 10.1007/978-3-540-79228-4
37. Dykman M, Schwartz I, Landsman A. **Disease Extinction in the Presence of Random Vaccination**. *Phys. Rev. Lett* (2008) **101** 078101. DOI: 10.1103/PhysRevLett.101.078101
38. Ewald R, Uhrmacher AM. **Sessl**. *ACM Trans. Model. Comput. Simul* (2014) **24** 1-25. DOI: 10.1145/2567895
39. FDA (2003). Guidance for Industry Exposure-Response Relationships — Study Design, Data Analysis, and Regulatory Applications
Available at: https://www.fda.gov/media/71277/download.. *Guidance for Industry Exposure-Response Relationships — Study Design, Data Analysis, and Regulatory Applications* (2003)
40. FDA (2016). Reporting of Computational Modeling Studies in Medical Device Submissions - Guidance for Industry and Food and Drug Administration Staff
Available at: https://www.fda.gov/media/87586/download.. *Reporting of Computational Modeling Studies in Medical Device Submissions - Guidance for Industry and Food and Drug Administration Staff* (2016)
41. FDA (2021a). Model-Informed Drug Development Pilot Program. FDA. Available at: https://www.fda.gov/drugs/development-resources/model-informed-drug-development-pilot-program (Accessed
October 24, 2021).. *Model-Informed Drug Development Pilot Program* (2021a)
42. FDA (2021b). Assessing the Credibility of Computational Modeling and Simulation in Medical Device Submissions. FDA. Available at: https://www.fda.gov/regulatory-information/search-fda-guidance-documents/assessing-credibility-computational-modeling-and-simulation-medical-device-submissions (Accessed
January 2, 2022).. *Assessing the Credibility of Computational Modeling and Simulation in Medical Device Submissions* (2021b)
43. Folk M, Heber G, Koziol Q, Pourmal E, Robinson D. *An Overview of the HDF5 Technology Suite and its Applications AD ‘11* (2011) 36-47. DOI: 10.1145/1966895.1966900
44. Forgoston E, Bianco S, Shaw LB, Schwartz IB. **Maximal Sensitive Dependence and the Optimal Path to Epidemic Extinction**. *Bull. Math. Biol* (2011) **73** 495-514. DOI: 10.1007/s11538-010-9537-0
45. Forgoston E, Moore RO. **A Primer on Noise-Induced Transitions in Applied Dynamical Systems**. *SIAM Rev* (2018) **60** 969-1009. DOI: 10.1137/17M1142028
46. Fujimoto R, Bock C, Chen W, Page E, Panchal JH. *Research Challenges in Modeling and Simulation for Engineering Complex Systems* (2017)
47. Galluppi GR, Brar S, Caro L, Chen Y, Frey N, Grimm HP. **Industrial Perspective on the Benefits Realized From the FDA’s Model-Informed Drug Development Paired Meeting Pilot Program**. *Clin. Pharmacol. Ther* (2021) **110** 1172-1175. DOI: 10.1002/cpt.2265
48. Garfinkel SL, Leclerc P. *Randomness Concerns When Deploying Differential Privacy* (2020) 73-86. DOI: 10.1145/3411497.3420211
49. Garira W. **A Complete Categorization of Multiscale Models of Infectious Disease Systems**. *J. Biol. Dyn* (2017) **11** 378-435. DOI: 10.1080/17513758.2017.1367849
50. Garira W. **A Primer on Multiscale Modelling of Infectious Disease Systems**. *Infect. Dis. Model* (2018) **3** 176-191. DOI: 10.1016/j.idm.2018.09.005
51. Gary T, Mingle D, Yenamandra A. *The Evolving Definition of Sepsis* (2021)
52. Gerkin RC, Birgiolas J, Jarvis RJ, Omar C, Crook SM. *NeuronUnit: A Package for Data-Driven Validation of Neuron Models Using SciUnit* (2019)
53. Hairer E, Wanner G, Lubich C. *Geometric Numerical Integration* (2006)
54. Halter M, Patterson E, Baas A, Fairbanks J. **Compositional Scientific Computing with Catlab and SemanticModels**. *arXiv:2005.04831* (2020)
55. Handel A, Longini IM, Antia R. **Neuraminidase Inhibitor Resistance in Influenza: Assessing the Danger of Its Generation and Spread**. *Plos Comput. Biol* (2007) **3** e240. DOI: 10.1371/journal.pcbi.0030240
56. Hart WS, Maini PK, Thompson RN. **High Infectiousness Immediately before COVID-19 Symptom Onset Highlights the Importance of Contact Tracing**. *Elife* (2021) **10** e65534. DOI: 10.7554/eLife.65534
57. Hernandez-Vargas EA, Velasco-Hernandez JX. **In-host Modelling of COVID-19 in Humans**. *medRxiv* (2020). DOI: 10.1101/2020.03.26.20044487
58. Honorato-Zimmer R, Millar AJ, Plotkin GD, Zardilis A. **Chromar, a Rule-Based Language of Parameterised Objects**. *Electron. Notes Theor. Computer Sci* (2018) **335** 49-66. DOI: 10.1016/j.entcs.2018.03.008
59. Hooker KL, Ganusov VV. **Impact of Oseltamivir Treatment on Influenza A and B Virus Dynamics in Human Volunteers**. *Front. Microbiol* (2021) **12** 631211. DOI: 10.3389/fmicb.2021.631211
60. IMAGWiki (2021). MODELS, TOOLS & DATABASES
Available at: https://www.imagwiki.nibib.nih.gov/resources/models-tools-databases.. *MODELS, TOOLS & DATABASES* (2021)
61. Interagency Modeling and Analysis Group (2021). Available at: https://www.imagwiki.nibib.nih.gov/(Accessed
October 24, 2021).. (2021)
62. Issues with regard to Call for transparency of COVID-19 models (2021). CoMSES Net Discourse Forums
Available at: https://forum.comses.net/t/issues-with-regard-to-call-for-transparency-of-covid-19-models/8433 (Accessed
October 24, 2021).. *CoMSES Net Discourse Forums* (2021)
63. Jakeman AJ, Letcher RA, Norton JP. **Ten Iterative Steps in Development and Evaluation of Environmental Models**. *Environ. Model. Softw* (2006) **21** 602-614. DOI: 10.1016/j.envsoft.2006.01.004
64. Jung J, Sugita Y. **Multiple Program/multiple Data Molecular Dynamics Method with Multiple Time Step Integrator for Large Biological Systems**. *J. Comput. Chem* (2017) **38** 1410-1418. DOI: 10.1002/jcc.24511
65. Ke R, Zitzmann C, Ribeiro RM, Perelson AS. **Kinetics of SARS-CoV-2 Infection in the Human Upper and Lower Respiratory Tracts and Their Relationship with Infectiousness**. *medRxiv* (2020). DOI: 10.1101/2020.09.25.20201772
66. Kermack WO, McKendrick AG, Walker GT. **A Contribution to the Mathematical Theory of Epidemics**. *Proc. R. Soc. Lond. A* (1927) **115** 700-721. DOI: 10.1098/rspa.1927.0118
67. Kirouac DC, Cicali B, Schmidt S. **Reproducibility of Quantitative Systems Pharmacology Models: Current Challenges and Future Opportunities**. *CPT Pharmacometrics Syst. Pharmacol* (2019) **8** 205-210. DOI: 10.1002/psp4.12390
68. Kitano H. *The Standard Graphical Notation for Biological Networks* (2021)
69. Krafczyk M, Shi A, Bhaskar A, Marinov D, Stodden V. *Scientific Tests and Continuous Integration Strategies to Enhance Reproducibility in the Scientific Software Context P-RECS ‘19* (2019) 23-28. DOI: 10.1145/3322790.3330595
70. Kuemmel C, Yang Y, Zhang X, Florian J, Zhu H, Tegenge M. **Consideration of a Credibility Assessment Framework in Model-Informed Drug Development: Potential Application to Physiologically-Based Pharmacokinetic Modeling and Simulation**. *CPT Pharmacometrics Syst. Pharmacol* (2020) **9** 21-28. DOI: 10.1002/psp4.12479
71. Licensing issues (2021). MSM Viral Pandemic Integration Subgroup Mailing List Archives
Available at: https://lists.simtk.org/pipermail/vp-integration-subgroup/2021-January/000022.html.. *MSM Viral Pandemic Integration Subgroup Mailing List Archives* (2021)
72. Lieven C, Beber ME, Olivier BG, Bergmann FT, Ataman M, Babaei P. **MEMOTE for Standardized Genome-Scale Metabolic Model Testing**. *Nat. Biotechnol* (2020) **38** 272-276. DOI: 10.1038/s41587-020-0446-y
73. Liu Q-H, Ajelli M, Aleta A, Merler S, Moreno Y, Vespignani A. **Measurability of the Epidemic Reproduction Number in Data-Driven Contact Networks**. *Proc. Natl. Acad. Sci. USA* (2018) **115** 12680-12685. DOI: 10.1073/pnas.1811115115
74. Malik-Sheriff RS, Glont M, Nguyen TVN, Tiwari K, Roberts MG, Xavier A. **BioModels-15 Years of Sharing Computational Models in Life Science**. *Nucleic Acids Res* (2020) **48** D407-D415. DOI: 10.1093/nar/gkz1055
75. Meyer M. **Continuous Integration and Its Tools**. *IEEE Softw* (2014) **31** 14-16. DOI: 10.1109/ms.2014.58
76. Miao H, Xia X, Perelson AS, Wu H. **On Identifiability of Nonlinear ODE Models and Applications in Viral Dynamics**. *SIAM Rev* (2011) **53** 3-39. DOI: 10.1137/090757009
77. Multiscale Modeling and Viral Pandemics (2021). Available at: https://www.imagwiki.nibib.nih.gov/working-groups/multiscale-modeling-and-viral-pandemics.. (2021)
78. Murray JD. *Mathematical Biology: I. An Introduction* (2002)
79. Musuamba FT, Skottheim Rusten I, Lesage R, Russo G, Bursi R, Emili L. **Scientific and Regulatory Evaluation of Mechanistic In Silico Drug and Disease Models in Drug Development: Building Model Credibility**. *CPT Pharmacometrics Syst. Pharmacol* (2021) **10** 804-825. DOI: 10.1002/psp4.12669
80. NASA (2016). NASA-STD-7009 | NASA Technical Standards System (NTSS)
Available at: https://standards.nasa.gov/standard/nasa/nasa-std-7009 (Accessed
October 24, 2021).. *NASA-STD-7009 | NASA Technical Standards System (NTSS)* (2016)
81. NASA (2019). Nasa Handbook For Models And Simulations: An Implementation Guide For Nasa-Std-7009a
Available at: https://standards.nasa.gov/sites/default/files/standards/NASA/PUBLISHED/A/nasa-hdbk-7009a.pdf.. *Nasa Handbook For Models And Simulations: An Implementation Guide For Nasa-Std-7009a* (2019)
82. Neal ML, Thompson CT, Kim KG, James RC, Cook DL, Carlson BE. **SemGen: a Tool for Semantics-Based Annotation and Composition of Biosimulation Models**. *Bioinformatics* (2019a) **35** 1600-1602. DOI: 10.1093/bioinformatics/bty829
83. Neal ML, König M, Nickerson D, Mısırlı G, Kalbasi R, Dräger A. **Harmonizing Semantic Annotations for Computational Models in Biology**. *Brief. Bioinform* (2019b) **20** 540-550. DOI: 10.1093/bib/bby087
84. Nieddu GT, Billings L, Kaufman JH, Forgoston E, Bianco S. **Extinction Pathways and Outbreak Vulnerability in a Stochastic Ebola Model**. *J. R. Soc. Interf* (2017) **14** 20160847. DOI: 10.1098/rsif.2016.0847
85. Ovaskainen O, Meerson B. **Stochastic Models of Population Extinction**. *Trends Ecol. Evol* (2010) **25** 643-652. DOI: 10.1016/j.tree.2010.07.009
86. Paradis EG, Pinilla LT, Holder BP, Abed Y, Boivin G, Beauchemin CAA. **Impact of the H275Y and I223V Mutations in the Neuraminidase of the 2009 Pandemic Influenza Virus**. *PLoS ONE* (2015) **10** e0126115. DOI: 10.1371/journal.pone.0126115
87. Parker P, Letcher R, Jakeman A, Beck MB, Harris G, Argent RM. *Progress in Integrated Assessment and modelling1A Summary of a Workshop on Integrated Assessment and Modelling, Held at EcoSummit 2000* (2002) **17** 209-217. DOI: 10.1016/S1364-8152(01)00059-7
88. Perelson AS, Rong L, Hayden FG. **Combination Antiviral Therapy for Influenza: Predictions From Modeling of Human Infections**. *J. Infect. Dis* (2012) **205** 1642-1645. DOI: 10.1093/infdis/jis265
89. Petersen BK, Ropella GE, Hunt CA. **Toward Modular Biological Models: Defining Analog Modules Based on Referent Physiological Mechanisms**. *BMC Syst. Biol* (2014) **8** 95. DOI: 10.1186/s12918-014-0095-1
90. Petrie SM, Guarnaccia T, Laurie KL, Hurt AC, McVernon J, McCaw JM. **Reducing Uncertainty in Within-Host Parameter Estimates of Influenza Infection by Measuring Both Infectious and Total Viral Load**. *PLOS ONE* (2013) **8** e64098. DOI: 10.1371/journal.pone.0064098
91. Powathil GG, Gordon KE, Hill LA, Chaplain MAJ. **Modelling the Effects of Cell-Cycle Heterogeneity on the Response of a Solid Tumour to Chemotherapy: Biological Insights from a Hybrid Multiscale Cellular Automaton Model**. *J. Theor. Biol* (2012) **308** 1-19. DOI: 10.1016/j.jtbi.2012.05.015
92. Recent SDO/COMBINE legal entity issues (2021). Available at: https://groups.google.com/g/sbml-discuss/c/8-ljMxJkMqA.. *Recent SDO/COMBINE legal entity issues* (2021)
93. Refsgaard JC, Henriksen HJ, Harrar WG, Scholten H, Kassahun A. **Quality Assurance in Model Based Water Management - Review of Existing Practice and Outline of New Approaches**. *Environ. Model. Softw* (2005) **20** 1201-1215. DOI: 10.1016/j.envsoft.2004.07.006
94. Russo G, Di Salvatore V, Sgroi G, Parasiliti Palumbo GA, Pappalardo F. **A Multi-step and Multi-Scale Bioinformatic Protocol to Investigate Potential SARS-CoV-2 Vaccine Targets**. *Brief Bioinform* (2021) **23**. DOI: 10.1093/bib/bbab403
95. Russo G, Pennisi M, Fichera E, Motta S, Raciti G, Viceconti M. **In Silico trial to Test COVID-19 Candidate Vaccines: a Case Study with UISS Platform**. *BMC Bioinformatics* (2020) **21** 527. DOI: 10.1186/s12859-020-03872-0
96. Sarma GP, Jacobs TW, Watts MD, Ghayoomie SV, Larson SD, Gerkin RC. **Unit Testing, Model Validation, and Biological Simulation [Version 1; Peer Review: 2 Approved, 1 Approved with Reservations]**. *F1000Res* (2016) **5** 1946. DOI: 10.12688/f1000research.9315.1
97. Satyanarayan A, Moritz D, Wongsuphasawat K, Heer J. **Vega-lite: A Grammar of Interactive Graphics**. *IEEE Trans. Vis. Comput. Graphics* (2017) **23** 341-350. DOI: 10.1109/tvcg.2016.2599030
98. Satyanarayan A, Wongsuphasawat K, Heer J. *Declarative Interaction Design for Data Visualization UIST ‘14* (2014)
99. Schmiester L, Schälte Y, Bergmann FT, Camba T, Dudkin E, Egert J. **PEtab-Interoperable Specification of Parameter Estimation Problems in Systems Biology**. *Plos Comput. Biol* (2021) **17** e1008646. DOI: 10.1371/journal.pcbi.1008646
100. Schwartz IB, Forgoston E, Bianco S, Shaw LB. **Converging towards the Optimal Path to Extinction**. *J. R. Soc. Interf* (2011) **8** 1699-1707. DOI: 10.1098/rsif.2011.0159
101. Segovia-Juarez JL, Ganguli S, Kirschner D. **Identifying Control Mechanisms of Granuloma Formation during**. *J. Theor. Biol* (2004) **231** 357-376. DOI: 10.1016/j.jtbi.2004.06.031
102. Shaikh B, Marupilla G, Wilson M, Blinov ML, Moraru II, Karr JR. *runBioSimulations: an Extensible Web Application that Simulates a Wide Range of Computational Modeling Frameworks, Algorithms, and Formats* (2021)
103. Sim TK. *Welcome* (2021)
104. Stack JC, Bansal S, Kumar VSA, Grenfell B. **Inferring Population-Level Contact Heterogeneity from Common Epidemic Data**. *J. R. Soc. Interf* (2013) **10** 20120578. DOI: 10.1098/rsif.2012.0578
105. Systems Biology Markup Language (SBML) (2021). Available at: http://sbml.org/.. (2021)
106. Tiwari K, Kananathan S, Roberts MG, Meyer JP, Sharif Shohan MU, Xavier A. **Reproducibility in Systems Biology Modelling**. *Mol. Syst. Biol* (2021) **17** e9982. DOI: 10.15252/msb.20209982
107. Viceconti M, Emili L, Afshari P, Courcelles E, Curreli C, Famaey N. **Possible Contexts of Use for In Silico Trials Methodologies: A Consensus-Based Review**. *IEEE J. Biomed. Health Inform* (2021) **25** 3977-3982. DOI: 10.1109/jbhi.2021.3090469
108. Viceconti M, Juárez MA, Curreli C, Pennisi M, Russo G, Pappalardo F. **POSITION PAPER: Credibility of In Silico Trial Technologies: A Theoretical Framing**. *arXiv:1909.04660* (2019)
109. Waites W, Cavaliere M, Manheim D, Panovska-Griffiths J, Danos V. **Rule-based Epidemic Models**. *J. Theor. Biol* (2021) **530** 110851. DOI: 10.1016/j.jtbi.2021.110851
110. Waltemath D, Adams R, Bergmann FT, Hucka M, Kolpakov F, Miller AK. **Reproducible Computational Biology Experiments with SED-ML -The Simulation Experiment Description Markup Language**. *BMC Syst. Biol* (2011) **5** 198. DOI: 10.1186/1752-0509-5-198
111. Wang L, Chen J, Marathe M. **). DEFSI: Deep Learning Based Epidemic Forecasting with Synthetic Information**. *Aaai* (2019) **33** 9607-9612. DOI: 10.1609/aaai.v33i01.33019607
112. Wang S, Pan Y, Wang Q, Miao H, Brown AN, Rong L. **Modeling the Viral Dynamics of SARS-CoV-2 Infection**. *Math. Biosciences* (2020) **328** 108438. DOI: 10.1016/j.mbs.2020.108438
113. Wickham H. **ggplot2**. *Wires Comp. Stat* (2011) **3** 180-185. DOI: 10.1002/wics.147
114. Wikipedia (2021a). International Classification of Diseases (Online). Available at: https://en.wikipedia.org/wiki/International_Classification_of_Diseases.. *International Classification of Diseases* (2021a)
115. Wikipedia (2021b). License Compatibility (Online). Available at: https://en.wikipedia.org/w/index.php?title=License_compatibility&oldid=1051142395 (Accessed
October 24, 2021).. *License Compatibility* (2021b)
116. Wikipedia (2021c). Public-domain-equivalent License (Online). Available at: https://en.wikipedia.org/w/index.php?title=Public-domain-equivalent_license&oldid=1051493543 (Accessed
October 24, 2021).. *Public-domain-equivalent License* (2021c)
117. Zarnitsyna VI, Gianlupi JF, Hagar A, Sego T, Glazier JA. **Advancing Therapies for Viral Infections Using Mechanistic Computational Models of the Dynamic Interplay between the Virus and Host Immune Response**. *Curr. Opin. Virol* (2021) **50** 103-109. DOI: 10.1016/j.coviro.2021.07.007
118. Zhao Y, Serebrenik A, Zhou Y, Filkov V, Vasilescu B. **The Impact of Continuous Integration on Other Software Development Practices: A Large-Scale Empirical Study**. (2017) 60-71
119. Zineh II. **Quantitative Systems Pharmacology: A Regulatory Perspective on Translation**. *CPT Pharmacometrics Syst. Pharmacol* (2019) **8** 336-339. DOI: 10.1002/psp4.12403
|
---
title: Gut Microbiota Dysbiosis Ameliorates in LNK-Deficient Mouse Models with Obesity-Induced
Insulin Resistance Improvement
authors:
- Jingbo Chen
- Jiawen Xu
- Yan Sun
- Yuhuan Xue
- Yang Zhao
- Dongzi Yang
- Shuijie Li
- Xiaomiao Zhao
journal: Journal of Clinical Medicine
year: 2023
pmcid: PMC10002478
doi: 10.3390/jcm12051767
license: CC BY 4.0
---
# Gut Microbiota Dysbiosis Ameliorates in LNK-Deficient Mouse Models with Obesity-Induced Insulin Resistance Improvement
## Abstract
Purpose: To investigate the potential role of gut microbiota in obesity-induced insulin resistance (IR). Methods: Four-week-old male C57BL/6 wild-type mice ($$n = 6$$) and whole-body SH2 domain-containing adaptor protein (LNK)-deficient in C57BL/6 genetic backgrounds mice ($$n = 7$$) were fed with a high-fat diet (HFD, $60\%$ calories from fat) for 16 weeks. The gut microbiota of 13 mice feces samples was analyzed by using a 16 s rRNA sequencing analysis. Results: The structure and composition of the gut microbiota community of WT mice were significantly different from those in the LNK-/- group. The abundance of the lipopolysaccharide (LPS)-producing genus Proteobacteria was increased in WT mice, while some short-chain fatty acid (SCFA)-producing genera in WT groups were significantly lower than in LNK-/- groups ($p \leq 0.05$). Conclusions: The structure and composition of the intestinal microbiota community of obese WT mice were significantly different from those in the LNK-/- group. The abnormality of the gut microbial structure and composition might interfere with glucolipid metabolism and exacerbate obesity-induced IR by increasing LPS-producing genera while reducing SCFA-producing probiotics.
## 1. Introduction
Obesity is becoming a worldwide health risk factor, and obesity-induced morbidity and complications account for huge costs for affected individuals, families, healthcare systems, and society at large. Obesity is a low-grade sustained inflammatory state that alters the whole-body metabolism that frequently leads to insulin resistance (IR) [1], which in turn plays a vital role in the pathogenesis of obesity-associated hyperlipidemia, non-alcoholic fatty liver disease, polycystic ovary syndrome, type 2 diabetes, and atherosclerotic cardiovascular disease [2]. Nutrients and substrates as well as systems involved in host–nutrient interactions, including gut microbiota, have been also identified as modulators of metabolic pathways controlling insulin action and obesity regulation [3]. However, the molecular mechanism of IR has not been exactly clarified.
Gut microbiota is the general term for the microbes that inhabit the gastrointestinal tract of the human body. Around 98–$99\%$ of the intestinal microbiomes can be classified into four groups: Bacteroidetes, Firmicutes, Proteobacteria, and Actinomycetes. The balance of intestinal microbe species is the key to keeping the intestinal immune function normal and maintaining the homeostasis of the body. Breaking the balance will lead to serious pathophysiological changes, which is called gut microbiota dysbiosis [4]. Increasing studies showed that Bacteroides are associated with high-fat and high-protein diets [5] and the imbalance of intestinal microecology might be involved in the occurrence of many diseases, such as irritable bowel syndrome, obesity, type 2 diabetes, metabolic syndrome (MetS), and cardiovascular diseases [6,7,8].
Metagenomic sequencing and 16S RNA sequencing were used to detect the changes in intestinal microbiota in patients with prediabetes, type 2 diabetes, and MetS. Two studies found that although the races and their diets were different, in type 2 diabetes patients, the proportion of Clostridium butyrate-producing *Roche fusobacterium* and *Clostridium leptum* decreased while the proportion of non-*Clostridium butyrate* increased [9,10]. The levels of Firmicutes and Clostridia in the gut microbiota of type 2 diabetes patients were significantly decreased as compared to normal controls, and the ratio of Bacteroidetes to Firmicutes was increased and positively correlated with blood glucose concentrations [11]. There are changes in the intestinal microbiota in people with abnormal glucose metabolism, and the changes in the intestinal microbiota also seem to be involved in the occurrence and remission of abnormal glucose metabolism. It was reported that feces from mice with abnormal glucose metabolism transplanted into healthy germ-free mice could cause abnormal glucose metabolism [12]. Furthermore, transplanting feces from lean donors into patients with MetS could increase their gut microbiota diversity and insulin sensitivity [13]. The results above suggested that the intestinal microbiota are closely related to the occurrence and development of abnormal glucose metabolism, while IR, as an important link in the occurrence and development of abnormal glucose metabolism, also seems to be related to the intestinal microbiota.
Our previous study discovered that ovarian tissues from PCOS patients with IR exhibited higher expression of the SH2 domain-containing adaptor protein (LNK) than ovaries from normal control subjects and PCOS patients without IR [14]. In addition, we found that there were more accumulated intrahepatic triglyceride, higher serum triglyceride (TG), and free fatty acid (FFA) in wild-type (WT) mice as compared to LNK-deficient (LNK-/-) mice fed with a high-fat diet (HFD). LNK deficiency improved glucose metabolism and IR in obese mice, suggesting the LNK might play a pivotal role in controlling glucolipid metabolism and obesity-induced IR by regulating IRS1/PI3K/Akt/AS160 signaling and the AKT/FOXO3 pathway [15,16]. Therefore, we chose LNK-/- mice as the IR-improved model and WT mice as the MetS/IR model. In this study, we compared intestinal microbiota of LNK-/- mice and WT mice that consumed HFD, with the aim to explore the potential influence of gut microbiomes on the glucolipid metabolic disorder and obesity-induced IR.
## 2.1. Animals
The study protocol was approved by the Research Ethics Board of Sun Yat-sen memorial hospital of Sun Yat-sen University and Guangdong Provincial People’s Hospital. All the experimental procedures were approved by the Committee for Animal Research of Sun Yat-sen University and the National Institutes of Health (NIH) Guide for the Care and Use of Laboratory Animals.
Four-week-old male C57BL/6 wild-type mice ($$n = 6$$) were purchased from the animal research center of Sun Yat-sen University. Whole-body LNK-deficient in C57BL/6 genetic backgrounds mice ($$n = 7$$) were created via CRISPR/Cas mediated genome engineering by Cyagen Biosciences Inc. The mouse Sh2b3 gene (GenBank accession number: NM_001306127.1; Ensembl: ENSMUSG00000042594) is located on mouse chromosome 5. Exon 1 to exon 3 were selected as target sites. Cas9 mRNA and gRNA generated using an in vitro transcription were then injected into fertilized eggs for knockout mouse production. All mice were randomly divided into different groups, housed 4 to 5 per cage, with standard laboratory conditions (12 h light:12 h darkness cycle) at a controlled temperature (23 ± 2 °C) and free access to rodent feed and water. All mice (4–5 weeks old) were fed a high-fat diet (HFD, $60\%$ calories from fat, D12492; Research Diets Inc., New Brunswick, NJ, USA) for 16 weeks.
## 2.2. Sample Collection
When mice were fed with a HFD for up to 16 weeks, fecal samples were collected and immediately kept frozen at −80 °C until processed for analysis. Total DNA was isolated from the fecal samples using the MasterPure Complete DNA&RNA Purification Kit (Epicenter) according to the manufacturer’s instructions with some modifications as described previously [17].
## 2.3. 16S rRNA Extraction and Sequencing
DNA was extracted using a DNA extraction kit for the corresponding sample. The concentration and purity were measured using the NanoDrop One (Thermo Fisher Scientific, Waltham, MA, USA). Next, 16S rRNA/18SrRNA/ITS genes of distinct regions (e.g., Bac 16S: V3-V4/V4/V4-V5; Fug 18S: V4/V5; ITS1/ITS2; Arc 16S: V4-V5 et al.) were amplified used specific primer (e.g., 16S: 338F and 806R/515F and 806R/515F and 907R; 18S: 528F and 706R/817F and 1196R; ITS5-1737F and ITS2-2043R/ITS3-F and ITS4R; Arc: Arch519F and Arch915R et al.) with a 12bp barcode. Primers were synthesized by Invitrogen (Invitrogen, Carlsbad, CA, USA). PCR reactions, containing 25 μL 2× Premix Taq (Takara Biotechnology, Dalian Co. Ltd., Dalian, China), 1 μL each primer (10 μM), and 3 μL DNA (20 ng/μL) template in a volume of 50 µL, were amplified via thermocycling: 5 min at 94 °C for initialization; 30 cycles of 30 s denaturation at 94 °C, 30 s annealing at 52 °C, and 30 s extension at 72 °C; followed by 10 min final elongation at 72 °C. The PCR instrument was BioRad S1000 (Bio-Rad Laboratory, Hercules, CA, USA). The length and concentration of the PCR product were detected via $1\%$ agarose gel electrophoresis. Samples with the bright main strip between (e.g., 16S V4: 290–310 bp/16S V4V5: 400–450 bp et al.) could be used for further experiments. PCR products were mixed in equidensity ratios according to the GeneTools Analysis Software (Version 4.03.05.0, SynGene, Cambridge, UK). Then, the mixture of PCR products was purified with E.Z.N.A. Gel Extraction Kit (Omega, Bellevue, WA, USA). Next, sequencing libraries were generated using NEBNext® Ultra™ II DNA Library Prep Kit for Illumina® (New England Biolabs, Ipswich, MA, USA) following the manufacturer’s recommendations, and index codes were added. The library quality was assessed on the Qubit@ 2.0 Fluorometer (Thermo Fisher Scientific, Waltham, MA, USA). At last, the library was sequenced on an Illumina Nova6000 platform and 250 bp paired-end reads were generated.
## 2.4. Data Analysis
Fastp (version 0.14.1) was used to control the quality of the raw data by sliding the window (-W 4 -M 20). The primers were removed by using cutadapt software according to the primer information at the beginning and end of the sequence to obtain the paired-end clean reads. Paired-end clean reads were merged using usearch -fastq_mergepairs (V10) according to the relationship of the overlap between the paired-end reads; when with at least a 16 bp overlap, the read generated from the opposite end of the same DNA fragment, the maximum mismatch allowed in the overlap region was 5 bp, and the spliced sequences were called Raw Tags. Fastp (version 0.14.1) was used to control the quality of the raw data by sliding the window (-W 4 -M 20) to obtain the paired-end clean tags.
R software was used to count the union (pan) and intersection (core) of the target classification level in different samples to evaluate whether the sample size was sufficient. R software was used to analyze the common and endemic species, the composition of the community, and the richness of species.
## 3.1. Diversity Difference of Intestinal Microbiota between LNK-/- and WT Mice
A total of 13 mice (7 LNK-/-mice and 6 WT mice) were included in this study. The average body weights of 0W LNK-/-mice and WT mice were 21 g ± 2.2 g and 21.1 g ± 2.1 g, respectively, with no significance ($p \leq 0.05$). During the process of the mice fed with HFD, we observed that LNK-/- mice had a loss of appetite compared with WT mice. The food intakes of LNK-/- mice and WT mice were 18.8 g ± 1.3 g and 19.4 g ± 1.8 g, respectively, with statistical significance ($p \leq 0.05$). After 16 weeks, there was a significant difference in body weight between LNK-/- mice (47.5 g ± 4.6 g) and WT mice (52.6 g ± 3.3 g) ($p \leq 0.05$). All thirteen feces samples from seven LNK-/-mice and six WT mice were analyzed. A majority of intestinal microbe species of LNK-/-mice and WT mice were similar, however, the diversity of gut microbiomes in the WT mice group was less than that of the LNK-/- mice group (Figure 1A). The α diversity of the gut microbiota calculated using the Shannon index showed that the LNK-/- group species diversity was higher than that of the WT group at the phylum level ($p \leq 0.05$, t test) (Figure 1B,C).
## 3.2. Composition and Abundance Difference of Intestinal Microbiota between LNK-/- and WT Mice
To compare the composition difference in the intestinal microbiota between LNK-/- and WT mice, we next performed a Bray–Curtis-based principal coordinates analysis (PCoA) (Figure 2A). It was shown that the degree of similarity between the two groups of microbial communities was significantly different (Bray–Curtis PERMANOVA, $$p \leq 0.016$$). In addition, the composition of the microbiota in the samples of the LNK-/- group was more heterogeneous and significantly different from that of the WT group. The heat map showed that gut microbiota compositions between the LNK-/- and WT groups were markedly different (Figure 2B).
In the phylum-level taxonomy classification, the WT group was dominated by Proteobacteria, Verrucomicrobia, and Bacteroidetes; the LNK-/- group was dominated by Bacteroidetes, Proteobacteria, and Firmicutes (Figure 2C). Although bacteria are similar at the phylum level between the two groups, Figure 2C showed that their proportion was different. The WT group was dominated by Proteobacteria and had a relative abundance of Verrucomicrobia, with the significance compared with LNK-/- mice ($p \leq 0.05$) (Figure 2D), while the LNK-/- group has a relatively large proportion of Firmicutes ($p \leq 0.05$) and Bacteroidetes (Figure 2D).
According to the results of the linear discriminant analysis effect size (LEfSe) (LDA ≥ 2.0), the abundances of Proteobacteria, Helicobacteraceae, Epsilonproteobacteria, and Campylobacterales were significantly increased in WT mice, while the abundance of Erysipelotrichales, Allobaculum, and Bacteroidales was significantly increased in LNK-/- mice (Figure 2E).
To explore the gut microbial differences between LNK-/- and WT mice further, we used STAMP software to analyze the genera with significant differences ($p \leq 0.05$). We found that the abundances of some short-chain fatty acid (SCFA)-producing genera in the WT groups were significantly lower than in the LNK-/- groups, such as Prevotella_9, Prevotellaceae_UCG-001, Clostridium_sensu_strict_1, Ruminococcaceae_UCG-010, and Stenotrophomonas (Figure 2F).
## 4. Discussion
Our previous study showed that upon the consumption of HFD, LNK-/- mice had a loss of appetite, and WT mice accumulated more intrahepatic triglyceride, TG, and FFA compared with LNK-/- mice. LNK plays a pivotal role in adipose glucose transport by regulating insulin-mediated IRS1/PI3K/Akt/AS160 signaling. In this study, we found that the abundance of Proteobacteria was significantly increased in the WT mice group, which was one of the main LPS-producing bacteria. Some SCFA-producing genera in WT groups were significantly lower than in the LNK-/- groups.
LPS is also called endotoxin. The complex of LPS and its receptor CD14 can be recognized by Toll-like receptor 4 (TLR4) on the surface of immune cells to induce an inflammatory response. When the change in diet or the use of antibiotics affects the balance of gut microbiota, the number of harmful bacteria such as G- bacteria increases, and the decomposed product LPS passes into the blood circulation through the intestinal epithelium to cause endotoxemia, which triggers a systemic inflammatory response [18]. This study revealed that inflammation and LPS levels were elevated in patients with type 2 diabetes. Both animal and human experiments have demonstrated that the direct injection of LPS can increase fasting blood glucose and insulin levels, resulting in hyperinsulinemia and insulin resistance. When the number of G- bacteria decreased with the use of antibiotics, the amount of LPS entering the circulation decreased, which could relieve the systemic inflammation and increase insulin sensitivity. LPS receptor CD14 knockout mice fed a high-fat diet or injected with LPS had decreased inflammatory factors in adipose tissue, increased insulin sensitivity in liver and adipose tissue, and had a delayed development of insulin resistance, and their weight gain slowed down [19,20,21,22]. The results suggest that LPS plays an important role in the induction of the inflammatory response and insulin resistance.
The intestinal microbiota may affect the content of circulating LPS in the following two ways to induce insulin resistance. For one thing, the structure of intestinal microbiota is unbalanced, the number of G + bacteria is decreased, the proportion of G- bacteria is increased, and the production of LPS is increased. Studies have shown that the number of G + bacteria such as *Clostridium decreased* and the number of LPS-containing bacteria such as Bacteroides and Proteobacteria increased in diabetic patients. Adding Lactobacillus and Bifidobacterium to the diet of high-fat-induced obese mice could help restore a balance between probiotics and pernicious bacteria in the gut and increase insulin sensitivity. The addition of prebiotic oligosaccharides to a high-fat diet-induced diabetic mouse model also increased the number of bifidobacteria, decreased the level of LPS, and improved insulin secretion and inflammation, which was significantly associated with the number of bifidobacteria [23]. Additionally, intestinal microbiota alter intestinal permeability. Studies have shown that a high-fat diet may interact with intestinal microbiota, alter intestinal permeability, promote the rise of LPS levels, and cause an inflammatory state and insulin resistance [24,25]. The intestinal microbiota selectively regulates the expression of colonic Cannabinoid receptor 1, which affects intestinal permeability by altering the distribution of Claudin-1 [26]. In addition, obesity itself affects intestinal permeability. A study of normal-weight and overweight healthy women showed a positive correlation between gut permeability and waist circumference and visceral fat content [27]. Increased visceral adipose promotes the secretion of the pro-inflammatory factors TNF α, IL-1, and IL-6 by infiltrating macrophages in adipose tissue and reducing the production of the anti-inflammatory factor adiponectin. With the action of multiple pro-inflammatory factors, intestinal mucus production was decreased, and intestinal permeability was increased. TNF-α can also act on tight junction proteins, resulting in the increased permeability of the tight junction of intestinal cells [28,29,30]. These proinflammatory factors can also promote insulin resistance and lipid storage in adipocytes, thereby forming a vicious cycle.
Probiotics such as Bifidobacterium, Lactobacillus, and Prevotella_9 can promote the release of SCFAs from the undigested soluble dietary fiber in the colon via fermentation, at the same time reducing the intestinal pH, inhibiting the growth of harmful bacteria, to reduce the production of LPS in the intestinal lumen [31]. SCFAs can also promote the secretion of insulin by pancreatic β cells by regulating the secretion of gut-derived hormones, such as glucagon-like Peptide 1 (Glp-1), Glucagon Peptide 2 (Glp-2), Peptide YY (PYY), and glucose-dependent insulinotropic Peptide (GIP), etc., to increase insulin sensitivity and suppress appetite and food intake, thereby improving insulin resistance. After 8-week oral medication of VSL#3 probiotics containing 8 kinds of viable bacteria, the diet-induced obesity mice had increased GLP-1 production, decreased food intake, reduced body weight, and improved glucose tolerance. Their intestinal microbiota composition also changed the number of probiotics of Firmicutes such as lactobacillus, and Bifidobacterium increased, which was related to the increase in butyrate in SCFAs [32]. Butyrate can improve the function of the intestine, promote the activity of the intestine, and has a better therapeutic effect on patients with a loss of appetite, diarrhea, dyspepsia, and so on. In addition, butyrate can promote the reduction of dietary intake and digestion and is also beneficial to obese or fatty liver patients [32]. In addition, another study showed that healthy volunteers ate inulin-containing foods that promoted probiotic growth and a regular diet, respectively. Moreover, GLP-2 was found to be increased in fasting serum and decreased in intestinal permeability after eating inulin-containing foods [33]. The results demonstrated that probiotics could promote the production of SCFAs and the secretion of GLP-1 and Glp-2 by regulating the balance of intestinal microbiota, further improving intestinal permeability and alleviating IR.
Our research explored the changes in the gut microbiota in LNK-/- and ET mice, which provided new ideas for the mechanism and treatment of MetS and IR. Although previous studies had shown that the disorder of intestine microbiota was related to MetS, the underlying mechanism remains unclear. Therefore, this study was a supplement to this research field. Nevertheless, this study still had some shortcomings. Firstly, as is known to all, sex hormones strongly influence body fat distribution and adipocyte differentiation. Estrogen and testosterone differentially affect adipocyte physiology and estrogens play a leading role in the causes and consequences of female obesity. Therefore, in this study, to avoid the influence of estrogen on the occurrence of obesity, we did not put male and female mice together to compare, and only collected fecal samples based on previous obesity-induced IR male mouse models. The sample size was not large enough, and there may be bias in the results for female mice. The results of female mice and the potential effects of sex hormones on gut microbiota need further research. Secondly, in the study, we focused on the difference in gut microbiota between LNK-/- and WT mice. We will continue relevant studies, and the indexes such as LPS, butyrate, gut permeability, and mucosal structural changes will be measured or observed in our next study. The relationship between changes in gut microbiomes and IR needs to be confirmed by further experiments. Finally, the 16S rRNA gene sequencing had some limitations, such as a short reading length, sequencing errors, and difficulty in evaluating and operating taxa. It would be important to combine signaling pathways and metabolomics analysis in the next step.
## 5. Conclusions
Our research described that the structure and composition of the gut microbiota community between LNK-/- and WT mice were significantly different. The change in the gut microbial structure and composition of obese WT mice might aggravate glucolipid metabolic disorder and IR by increasing the production of LPS while reducing the production of SCFAs.
## References
1. Boucher J., Kleinridders A., Kahn C.R.. **Insulin receptor signaling in normal and insulin-resistant states**. *Cold Spring Harb. Perspect. Biol.* (2014) **6** a009191. DOI: 10.1101/cshperspect.a009191
2. Nolan C.J., Prentki M.. **Insulin resistance and insulin hypersecretion in the metabolic syndrome and type 2 diabetes: Time for a conceptual framework shift**. *Diabetes Vasc. Dis. Res.* (2019) **16** 118-127. DOI: 10.1177/1479164119827611
3. He F.F., Li Y.M.. **Role of gut microbiota in the development of insulin resistance and the mechanism underlying polycystic ovary syndrome: A review**. *J. Ovarian Res.* (2020) **13** 73. DOI: 10.1186/s13048-020-00670-3
4. Xu W., Xu L., Xu C.. **Relationship between Helicobacter pylori infection and gastrointestinal microecology**. *Front. Cell. Infect. Microbiol.* (2022) **12** 938608. DOI: 10.3389/fcimb.2022.938608
5. Wu G.D., Chen J., Hoffmann C., Bittinger K., Chen Y.Y., Keilbaugh S.A., Bewtra M., Knights D., Walters W.A., Knight R.. **Linking long-term dietary patterns with gut microbial enterotypes**. *Science* (2011) **334** 105-108. DOI: 10.1126/science.1208344
6. Ye Y., Liu Y., Cheng K., Wu Z., Zhang P., Zhang X.. **Effects of Intestinal Flora on Irritable Bowel Syndrome and Therapeutic Significance of Polysaccharides**. *Front. Nutr.* (2022) **9** 810453. DOI: 10.3389/fnut.2022.810453
7. Liu J., Xiao L., Nie H., Pan Y., Liu Y., Zhang Z., Lin X., Zhang Y., Cai J., Yang M.. **Microecological preparation combined with a modified low-carbon diet improves glucolipid metabolism and cardiovascular complication in obese patients**. *Diabetol. Metab. Syndr.* (2021) **13** 77. DOI: 10.1186/s13098-021-00697-6
8. Zhang S., Cai Y., Meng C., Ding X., Huang J., Luo X., Cao Y., Gao F., Zou M.. **The role of the microbiome in diabetes mellitus**. *Diabetes Res. Clin. Pract.* (2021) **172** 108645. DOI: 10.1016/j.diabres.2020.108645
9. Xu J., Liu W., Wu J., Wang W., Wang Z., Yu X., Zhang H., Zhu L., Zhan X.. **Metabolic profiles of oligosaccharides derived from four microbial polysaccharides by faecal inocula from type 2 diabetes patients**. *Int. J. Food Sci. Nutr.* (2021) **72** 1083-1094. DOI: 10.1080/09637486.2021.1908964
10. McMurdie P.J., Stoeva M.K., Justice N., Nemchek M., Sieber C.M.K., Tyagi S., Gines J., Skennerton C.T., Souza M., Kolterman O.. **Increased circulating butyrate and ursodeoxycholate during probiotic intervention in humans with type 2 diabetes**. *BMC Microbiol.* (2022) **22**. DOI: 10.1186/s12866-021-02415-8
11. Zhang L., Wang Z., Zhang X., Zhao L., Chu J., Li H., Sun W., Yang C., Wang H., Dai W.. **Alterations of the Gut Microbiota in Patients with Diabetic Nephropathy**. *Microbiol. Spectr.* (2022) **10** e0032422. DOI: 10.1128/spectrum.00324-22
12. Liu Y., Qin S., Feng Y., Song Y., Lv N., Liu F., Zhang X., Wang S., Wei Y., Li S.. **Perturbations of gut microbiota in gestational diabetes mellitus patients induce hyperglycemia in germ-free mice**. *J. Dev. Orig. Health Dis.* (2020) **11** 580-588. DOI: 10.1017/S2040174420000768
13. Aron-Wisnewsky J., Clément K., Nieuwdorp M.. **Fecal Microbiota Transplantation: A Future Therapeutic Option for Obesity/Diabetes?**. *Curr. Diabetes Rep.* (2019) **19** 51. DOI: 10.1007/s11892-019-1180-z
14. Hao M., Yuan F., Jin C., Zhou Z., Cao Q., Xu L., Wang G., Huang H., Yang D., Xie M.. **Overexpression of Lnk in the Ovaries Is Involved in Insulin Resistance in Women with Polycystic Ovary Syndrome**. *Endocrinology* (2016) **157** 3709-3718. DOI: 10.1210/en.2016-1234
15. Tan M., Cheng Y., Zhong X., Yang D., Jiang S., Ye Y., Ding M., Guan G., Yang D., Zhao X.. **LNK promotes granulosa cell apoptosis in PCOS via negatively regulating insulin-stimulated AKT-FOXO3 pathway**. *Aging* (2021) **13** 4617-4633. DOI: 10.18632/aging.202421
16. Zhong X., Ke C., Cai Z., Wu H., Ye Y., Liang X., Yu L., Jiang S., Shen J., Wang L.. **LNK deficiency decreases obesity-induced insulin resistance by regulating GLUT4 through the PI3K-Akt-AS160 pathway in adipose tissue**. *Aging* (2020) **12** 17150-17166. DOI: 10.18632/aging.103658
17. Rodríguez-Díaz J., García-Mantrana I., Vila-Vicent S., Gozalbo-Rovira R., Buesa J., Monedero V., Collado M.C.. **Relevance of secretor status genotype and microbiota composition in susceptibility to rotavirus and norovirus infections in humans**. *Sci. Rep.* (2017) **7** 45559. DOI: 10.1038/srep45559
18. Adda-Rezig H., Carron C., Pais de Barros J.P., Choubley H., Charron É., Rérole A.L., Laheurte C., Louvat P., Gaiffe É., Simula-Faivre D.. **New Insights on End-Stage Renal Disease and Healthy Individual Gut Bacterial Translocation: Different Carbon Composition of Lipopolysaccharides and Different Impact on Monocyte Inflammatory Response**. *Front. Immunol.* (2021) **12** 658404. DOI: 10.3389/fimmu.2021.658404
19. Ghosh S.S., Righi S., Krieg R., Kang L., Carl D., Wang J., Massey H.D., Sica D.A., Gehr T.W., Ghosh S.. **High Fat High Cholesterol Diet (Western Diet) Aggravates Atherosclerosis, Hyperglycemia and Renal Failure in Nephrectomized LDL Receptor Knockout Mice: Role of Intestine Derived Lipopolysaccharide**. *PLoS ONE* (2015) **10**. DOI: 10.1371/journal.pone.0141109
20. Fernández-Real J.M., Pérez del Pulgar S., Luche E., Moreno-Navarrete J.M., Waget A., Serino M., Sorianello E., Sánchez-Pla A., Pontaque F.C., Vendrell J.. **CD14 modulates inflammation-driven insulin resistance**. *Diabetes* (2011) **60** 2179-2186. DOI: 10.2337/db10-1210
21. Imajo K., Fujita K., Yoneda M., Nozaki Y., Ogawa Y., Shinohara Y., Kato S., Mawatari H., Shibata W., Kitani H.. **Hyperresponsivity to low-dose endotoxin during progression to nonalcoholic steatohepatitis is regulated by leptin-mediated signaling**. *Cell Metab.* (2012) **16** 44-54. DOI: 10.1016/j.cmet.2012.05.012
22. Roncon-Albuquerque R., Moreira-Rodrigues M., Faria B., Ferreira A.P., Cerqueira C., Lourenço A.P., Pestana M., von Hafe P., Leite-Moreira A.F.. **Attenuation of the cardiovascular and metabolic complications of obesity in CD14 knockout mice**. *Life Sci.* (2008) **83** 502-510. DOI: 10.1016/j.lfs.2008.07.021
23. Moya-Pérez A., Neef A., Sanz Y.. **Bifidobacterium pseudocatenulatum CECT 7765 Reduces Obesity-Associated Inflammation by Restoring the Lymphocyte-Macrophage Balance and Gut Microbiota Structure in High-Fat Diet-Fed Mice**. *PLoS ONE* (2015) **10**. DOI: 10.1371/journal.pone.0126976
24. Malesza I.J., Malesza M., Walkowiak J., Mussin N., Walkowiak D., Aringazina R., Bartkowiak-Wieczorek J., Mądry E.. **High-Fat, Western-Style Diet, Systemic Inflammation, and Gut Microbiota: A Narrative Review**. *Cells* (2021) **10**. DOI: 10.3390/cells10113164
25. Balakumar M., Prabhu D., Sathishkumar C., Prabu P., Rokana N., Kumar R., Raghavan S., Soundarajan A., Grover S., Batish V.K.. **Improvement in glucose tolerance and insulin sensitivity by probiotic strains of Indian gut origin in high-fat diet-fed C57BL/6J mice**. *Eur. J. Nutr.* (2018) **57** 279-295. DOI: 10.1007/s00394-016-1317-7
26. Wang J., Zhang X., Yang C., Zhao S.. **Effect of monoacylglycerol lipase inhibition on intestinal permeability in chronic stress model**. *Biochem. Biophys. Res. Commun.* (2020) **525** 962-967. DOI: 10.1016/j.bbrc.2020.02.173
27. Gummesson A., Carlsson L.M., Storlien L.H., Bäckhed F., Lundin P., Löfgren L., Stenlöf K., Lam Y.Y., Fagerberg B., Carlsson B.. **Intestinal permeability is associated with visceral adiposity in healthy women**. *Obesity* (2011) **19** 2280-2282. DOI: 10.1038/oby.2011.251
28. Kaminsky L.W., Al-Sadi R., Ma T.Y.. **IL-1β and the Intestinal Epithelial Tight Junction Barrier**. *Front. Immunol.* (2021) **12** 767456. DOI: 10.3389/fimmu.2021.767456
29. Obermüller B., Frisina N., Meischel M., Singer G., Stanzl-Tschegg S., Lichtenegger H., Kolb D., Klymiuk I., Till H., Castellani C.. **Examination of intestinal ultrastructure, bowel wall apoptosis and tight junctions in the early phase of sepsis**. *Sci. Rep.* (2020) **10** 11507. DOI: 10.1038/s41598-020-68109-9
30. Fusco R., Cordaro M., Siracusa R., D’Amico R., Genovese T., Gugliandolo E., Peritore A.F., Crupi R., Impellizzeri D., Cuzzocrea S.. **Biochemical Evaluation of the Antioxidant Effects of Hydroxytyrosol on Pancreatitis-Associated Gut Injury**. *Antioxidants* (2020) **9**. DOI: 10.3390/antiox9090781
31. Schoeler M., Caesar R.. **Dietary lipids, gut microbiota and lipid metabolism**. *Rev. Endocr. Metab. Disord.* (2019) **20** 461-472. DOI: 10.1007/s11154-019-09512-0
32. Tom Dieck H., Schön C., Wagner T., Pankoke H.C., Fluegel M., Speckmann B.. **A Synbiotic Formulation Comprising Bacillus subtilis DSM 32315 and L-Alanyl-L-Glutamine Improves Intestinal Butyrate Levels and Lipid Metabolism in Healthy Humans**. *Nutrients* (2021) **14**. DOI: 10.3390/nu14010143
33. Dong C.X., Zhao W., Solomon C., Rowland K.J., Ackerley C., Robine S., Holzenberger M., Gonska T., Brubaker P.L.. **The intestinal epithelial insulin-like growth factor-1 receptor links glucagon-like peptide-2 action to gut barrier function**. *Endocrinology* (2014) **155** 370-379. DOI: 10.1210/en.2013-1871
|
---
title: The Longitudinal Changes in Subcutaneous Abdominal Tissue and Visceral Adipose
Tissue Volumetries Are Associated with Iron Status
authors:
- Alejandro Hinojosa-Moscoso
- Anna Motger-Albertí
- Elena De la Calle-Vargas
- Marian Martí-Navas
- Carles Biarnés
- María Arnoriaga-Rodríguez
- Gerard Blasco
- Josep Puig
- Diego Luque-Córdoba
- Feliciano Priego-Capote
- José María Moreno-Navarrete
- José Manuel Fernández-Real
journal: International Journal of Molecular Sciences
year: 2023
pmcid: PMC10002479
doi: 10.3390/ijms24054750
license: CC BY 4.0
---
# The Longitudinal Changes in Subcutaneous Abdominal Tissue and Visceral Adipose Tissue Volumetries Are Associated with Iron Status
## Abstract
Excess iron is known to trigger adipose tissue dysfunction and insulin resistance. Circulating markers of iron status have been associated with obesity and adipose tissue in cross-sectional studies. We aimed to evaluate whether iron status is linked to changes in abdominal adipose tissue longitudinally. Subcutaneous abdominal tissue (SAT) and visceral adipose tissue (VAT) and its quotient (pSAT) were assessed using magnetic resonance imaging (MRI), at baseline and after one year of follow-up, in 131 (79 in follow-up) apparently healthy subjects, with and without obesity. Insulin sensitivity (euglycemic– hyperinsulinemic clamp) and markers of iron status were also evaluated. Baseline serum hepcidin ($$p \leq 0.005$$ and $$p \leq 0.002$$) and ferritin ($$p \leq 0.02$$ and $$p \leq 0.01$$)) were associated with an increase in VAT and SAT over one year in all subjects, while serum transferrin ($$p \leq 0.01$$ and $$p \leq 0.03$$) and total iron-binding capacity ($$p \leq 0.02$$ and $$p \leq 0.04$$) were negatively associated. These associations were mainly observed in women and in subjects without obesity, and were independent of insulin sensitivity. After controlling for age and sex, serum hepcidin was significantly associated with changes in subcutaneous abdominal tissue index (iSAT) (β = 0.406, $$p \leq 0.007$$) and visceral adipose tissue index (iVAT) (β = 0.306, $$p \leq 0.04$$), while changes in insulin sensitivity (β = 0.287, $$p \leq 0.03$$) and fasting triglycerides (β = −0.285, $$p \leq 0.03$$) were associated with changes in pSAT. These data indicated that serum hepcidin are associated with longitudinal changes in SAT and VAT, independently of insulin sensitivity. This would be the first prospective study evaluating the redistribution of fat according to iron status and chronic inflammation.
## 1. Introduction
Iron is increasingly recognized as modulating the phenotype of metabolic diseases. Different parameters evaluating iron stores have been associated with obesity status. In an initial study, circulating ferritin was positively associated with visceral and subcutaneous fat areas and negatively associated with the hepatic fat content [1]. Serum ferritin concentration was later confirmed to be associated with several obesity indicators [2,3]. In addition to serum ferritin, serum hepcidin and hepatic iron content (HIC) were increased in subjects with obesity [4]. Out of the 25 studies included in a systematic review of the literature from 2012, only 10 performed a comparative analysis between subjects with and without obesity. Of these ten studies, seven showed that obesity groups had increased mean-hemoglobin concentrations; six had raised serum ferritin; and four had reduced transferrin saturation [5]. All these results are indicative of alterations in iron storage and chronic inflammation in obesity. Interestingly, increased serum ferritin and iron were also associated with decreased markers of adipocyte differentiation [6], and the reduction in iron by phlebotomy led to improved glucose tolerance [7]. Increased adipose-tissue markers of iron accumulation using magnetic resonance imaging were also associated with obesity, insulin resistance and markers of adipose tissue dysfunction [8]. Despite all this information, we found no longitudinal studies testing whether iron status is associated with spontaneous changes in fatness.
There is experimental evidence of the close links between adipose tissue function and iron metabolism. Transferrin, the main circulating protein that binds iron with high affinity, is highly expressed in adipocytes in association with systemic insulin sensitivity, being required for adipogenesis and the maintenance of adipocyte function [8]. In fact, iron chelation blunted adipocyte differentiation in the absence of iron overload, which was recovered after the administration of transferrin [9] or other iron donors such as lactoferrin [10]. Increased body-iron stores have also been found to impact the development of adiposity in animal studies, impairing both adipogenesis and insulin action [11]. Mice with diabetes and obesity exhibited raised accumulation of iron in adipose tissue [12], a situation also associated with impaired adipocyte differentiation [9]. Iron chelation by deferoxamine in this context led to improved adipogenesis and reduced oxidative stress, and improved inflammation and adipocyte hypertrophy, in parallel with improved insulin action [7,12]. Similar findings in both extremes of iron stores suggest that a precise and fine-tuned iron availability is required for adipogenesis. Strengthening the importance of adipose-tissue iron on whole-body metabolism, a recent study demonstrated that low adipocyte-iron levels can restrain intestinal lipid absorption and prevent caloric influx in high fat-fed mice [13].
The evaluation of certain parameters as serum ferritin needs to take into account the inflammatory counterpart of this protein. In fact, serum uric acid, a marker of oxidative stress, is positively associated with VAT and SAT areas [14], while C-reactive protein (CRP) in plasma is also well known for increasing with measurements of abdominal adiposity, including visceral fat [2], and has also been associated with weight gain in some longitudinal studies [15,16].
The findings linking iron metabolism and adiposity are limited by their cross-sectional approach in the majority of studies. Analyzing the associations between the iron-metabolism parameters and the temporal evolution of adipose tissue could establish those indicators as predictive for adipose tissue changes. Therefore, we aimed to evaluate whether different parameters affected by iron exposure were associated with changes in adiposity and insulin action in a longitudinal study, controlling for inflammatory- and oxidative-stress markers. Specifically, we evaluated the temporal variations in subcutaneous and visceral adipose-tissue volumetry and distribution using MRI over one year in subjects with and without obesity. Considering that adrenal- or gonadal-steroid hormones are important metabolism modulators that contribute to body fat distribution [17,18,19] and iron homeostasis [20,21], levels of cortisol, estradiol, dehydroepiandrosterone sulphate, testosterone, progesterone, corticosterone and aldosterone were also analyzed in relation to adipose-tissue volumetry and distribution and baseline levels of iron-metabolism-related parameters.
## 2. Results
From the initial cohort of 175 participants, we excluded those that had missing or defective MRI scans ($$n = 26$$) and those that were subjected to bariatric surgery ($$n = 18$$) between baseline and follow-up, to give a final population of 131 subjects at baseline and 79 subjects that agreed to be examined again approximately one year (1.05 ± 0.24 years) later (Figure 1).
Clinical characteristics of the subjects with and without obesity at baseline and the comparison of baseline against follow-up for both groups are shown in Table 1. Body mass index (BMI), SAT and VAT (both raw volumes and corrected indexes) were higher in subjects with obesity, in parallel with decreased glucose-infusion rate during the euglycemic clamp as an indicator of decreased insulin sensitivity. Subjects with obesity also had increased hepatic-iron concentration, circulating ultrasensitive C-reactive protein (CRP), fasting triglycerides, serum uric acid, serum transferrin, and total iron-binding capacity (TIBC).
After stratification by sex and obesity status, we observed the expected differences between men and women with obesity in circulating hepcidin (26.23 ± 16.38 vs. 17.1 ± 10.87 ng/mL, respectively, $$p \leq 0.046$$), and ferritin levels (123 (63.5, 251) vs. 63 [35, 94] ng/mL, respectively, $$p \leq 0.02$$). Similar differences were observed in subjects without obesity (23.77 ± 9.82 in men vs. 19.50 ± 16.71 ng/mL in women, $$p \leq 0.035$$ for serum hepcidin; 196 [133, 235] in men vs. 64 [32, 95] ng/mL in women, $p \leq 0.0001$ for ferritin).
## 2.1. Cross-Sectional Findings
iVAT was positively correlated with blood hemoglobin, hematocrit, hepatic iron, transferrin, TIBC, ultrasensitive CRP, and uric acid (Figure 2A). Similar associations were found with iSAT regarding hepatic iron, serum transferrin and TIBC, ultrasensitive CRP and uric acid. On the other hand, pSAT was negatively associated with blood hemoglobin, hematocrit, hepatic iron, serum ferritin, hepcidin and serum uric acid.
Baseline insulin sensitivity (M value) measured through euglycemic clamp was negatively correlated with iSAT (r = −0.68, $p \leq 0.0001$), iVAT (r = −0.74, $p \leq 0.0001$), blood hematocrit (r = −0.19, $$p \leq 0.036$$), hepatic iron (r = −0.34, $$p \leq 0.001$$), TIBC (r = −0.22, $$p \leq 0.017$$), serum transferrin (r = −0.2, $$p \leq 0.03$$), ultrasensitive CRP (r = −0.56, $p \leq 0.0001$), uric acid (r = −0.50, $p \leq 0.0001$), while a positive association was found between insulin sensitivity and pSAT ($r = 0.23$, $$p \leq 0.011$$).
Fasting glucose and HbA1c levels did not correlate with haemoglobin, haematocrit, ferritin, transferrin, TIBC, hepcidin, iron and hepatic iron content (all them with r < 0.12 and $p \leq 0.1$). Interestingly, plasma leptin (a circulating marker of adipose tissue function) was negatively correlated with haemoglobin (r = −0.21, $$p \leq 0.034$$), ferritin (r = −0.26, $$p \leq 0.006$$) and iron (r = −0.24, $$p \leq 0.011$$) and positively with transferrin ($r = 0.352$, $p \leq 0.001$) and TIBC ($r = 0.353$, $p \leq 0.001$).
## 2.2.1. Baseline Parameters of Iron Metabolism and Chronic Inflammation Are Associated with Changes in Adipose Tissue Volumetries
In all subjects, a positive association was found between baseline serum hepcidin and ferritin and changes in iSAT and iVAT (Figure 2B), and negative associations with serum transferrin and TIBC. Serum ferritin and iron were negatively associated with pSAT changes, pointing towards a reduction in pSAT with high baseline ferritin concentrations. Baseline ultrasensitive CRP was positively associated with pSAT changes.
When the findings were stratified by sex, similar associations were found among women (Figure 2C) with serum hepcidin and ferritin being positively associated with changes in iSAT and iVAT. Accordingly, changes in pSAT had a negative relationship with serum ferritin. The positive and negative trends of these associations can be appreciated in Figure 3. In men (Figure 2D), only serum transferrin and TIBC were negatively associated with changes in iSAT and iVAT. Additionally, white blood cell count was reciprocally associated with changes in iVAT (positively) and pSAT (negatively).
According to obesity status, serum hepcidin (positively) and transferrin (negatively) were associated with changes in iSAT and iVAT in subjects without obesity (Figure 2E). In subjects with obesity, iron and ferritin were negatively associated with changes in pSAT (Figure 2F).
## 2.2.2. Changes in the Chronic-Inflammation Marker, Ultrasensitive CRP, Are Associated with Changes in Adipose Tissue Volumetries
Additionally, after computing the linear correlations between changes in adipose tissue volumetries and changes in iron metabolism (haemoglobin, haematrocrit, iron, transferrin, TIBC and ferritin) or inflammatory parameters (total WBC and ultrasensitive CRP), the only significant associations were found between changes in ultrasensitive CRP and changes in iVAT and pSAT in all subjects (iVAT: $r = 0.31$, $$p \leq 0.0084$$; pSAT: r = −0.27, $$p \leq 0.018$$) and women (iVAT: $r = 0.35$, $$p \leq 0.011$$; pSAT: r = −0.28, $$p \leq 0.047$$). However, these associations were lost after adjusting for age, sex and BMI in multiple-linear-regression analysis.
## 2.2.3. Baseline Parameters of Iron Metabolism, Chronic Inflammation and Adipose Tissue Volumetries Are Associated with Changes in Insulin Sensitivity
In all subjects, baseline blood hemoglobin (r = −0.23, $$p \leq 0.037$$), haematocrit (r = −0.26, $$p \leq 0.021$$), iVAT (r = −0.23, $$p \leq 0.039$$) and uric acid (r = −0.31, $$p \leq 0.004$$) were negatively associated with changes in insulin sensitivity (M value).
After stratification by obesity status, blood hemoglobin (r = −0.36, $$p \leq 0.019$$), haematocrit (r = −0.44, $$p \leq 0.003$$) and ultrasensitive CRP (r = −0.37, $$p \leq 0.016$$) were significantly associated with the change in insulin sensitivity in subjects without obesity. No significant associations were found in subjects with obesity.
In men, only the associations of ultrasensitive CRP (r = −0.55, $$p \leq 0.016$$) and iSAT (r = −0.59, $$p \leq 0.006$$) with the change in insulin sensitivity was significant. No significant associations were observed among women.
## 2.2.4. Adrenal- and Gonadal-Steroid Hormones Are Not Associated with Changes in Adipose Tissue Volumetries or Baseline Parameters of Iron Metabolism
No significant associations between adrenal- and gonadal-steroid hormones and the percent change of iSAT, iVAT and pSAT or baseline levels of serum hepcidin and transferrin were found (Table 2). In addition, while baseline levels of serum ferritin were negatively correlated with estradiol (r = −0.20, $$p \leq 0.02$$) and progesterone (r = −0.22, $$p \leq 0.01$$), and positively with testosterone ($r = 0.38$, $p \leq 0.001$) (Table 2), these associations were lost after adjusting for age, sex and BMI in multiple-linear-regression analysis.
## 2.2.5. Multiple-Linear-Regression Models
Multiple-linear-regression models were built with the change in adipose tissue volumetries as dependent variables, and age, sex, BMI and baseline serum ferritin, hepcidin, transferrin, ultrasensitive CRP, insulin sensitivity, fasting triglycerides, testosterone and estradiol, as independent variables. The baseline and changes regarding insulin sensitivity after one year were used separately (Table 3). In models with baseline insulin sensitivity, hepcidin (β = 0.283 $$p \leq 0.05$$) and insulin sensitivity (β = 0.354, $$p \leq 0.04$$) were found to be significantly associated with changes in iVAT, while only hepcidin (β = 0.392, $$p \leq 0.008$$) and fasting triglycerides (β = −0.278, $$p \leq 0.04$$) were associated with changes in iSAT and pSAT, respectively. In models with changes in insulin sensitivity as the independent variable, only baseline hepcidin was significantly associated with changes in iSAT (β = 0.406, $$p \leq 0.007$$) and iVAT (β = 0.306, $$p \leq 0.04$$), while changes in insulin sensitivity (β = 0.287, $$p \leq 0.03$$) and fasting triglycerides (β = −0.285, $$p \leq 0.03$$) were associated with changes in pSAT. Models using changes in ferritin and ultrasensitive CRP instead of the baseline values were also computed, but the results remained essentially the same.
## 2.2.6. Serum Ferritin and Hepatic Iron
We also explored at baseline the associations with hepatic iron content, as an independent measure of iron stores. As expected, baseline hepatic iron positively correlated with serum ferritin concentration in all subjects ($r = 0.28$, $$p \leq 0.0037$$) and in subjects without obesity ($r = 0.42$, $$p \leq 0.0012$$).
To further examine the potential influence of baseline ferritin and hepatic iron on changes in adipose tissue volumetries, we divided our subjects into low ferritin ($$n = 66$$ in baseline, $$n = 37$$ in follow-up) and high ferritin ($$n = 63$$ in baseline, $$n = 40$$ in follow-up), taking the median value for men and women separately as the cut-off points between groups (men: 184 ng/mL; women: 64 ng/mL). A Student’s t-test between groups showed increases in iSAT and iVAT with a high baseline ferritin concentration to be statistically significant (iSAT: $$p \leq 0.008$$, iVAT: $$p \leq 0.017$$). The same analysis was performed separating low ($$n = 53$$ in baseline, $$n = 38$$ in follow-up) and high ($$n = 53$$ in baseline, $$n = 30$$ in follow-up) hepatic iron concentration with a cut-off point based on the overall median (11.90 μg/g), but no significant differences were found between those groups.
The sample was later divided into three groups depending on ferritin and hepatic iron: low ferritin and low hepatic iron ($$n = 28$$ in baseline, $$n = 20$$ in follow-up), high ferritin and high hepatic iron ($$n = 26$$ in baseline, $$n = 16$$ in follow-up) and low ferritin and high hepatic iron or high ferritin and low hepatic iron ($$n = 52$$ in baseline, $$n = 32$$ in follow-up). An ANOVA analysis, including a post-hoc multiple comparisons test, showed no significant differences in adiposity changes between these groups. However, comparison between low ferritin/low hepatic iron and high ferritin/high hepatic iron showed significant differences in baseline for the iVAT ($$p \leq 0.021$$) and in follow-up for iVAT ($$p \leq 0.016$$) and pSAT ($$p \leq 0.02$$), pointing towards subjects with higher concentrations of ferritin and hepatic iron having a higher amount and proportion of abdominal visceral adipose tissue. This section may be divided under subheadings. It should provide a concise and precise description of the experimental results and their interpretation, as well as the experimental conclusions that can be drawn.
## 3. Discussion
The results of our study show the association of iron-metabolism markers with adipose tissue volumetries both cross-sectionally and longitudinally. iVAT was positively correlated with circulating iron stores (blood hemoglobin, hematocrit) and markers of tissue iron (hepatic iron, serum hepcidin and ferritin, transferrin and TIBC). Accordingly, pSAT was negatively associated with blood hemoglobin, hematocrit, hepatic iron, serum ferritin and hepcidin, and positively with insulin sensitivity.
Further analysis through multiple-linear-regression models confirmed the independent association of baseline serum hepcidin with iSAT and iVAT changes and fasting triglycerides or changes in insulin sensitivity with pSAT changes.
We also analyzed adrenal- or gonadal-steroid hormones, which are important metabolism modulators that contribute to body-fat distribution [17,18,19] and iron homeostasis [20,21], and no significant association between these hormones and circulating iron metabolism-related parameters or changes in iSAT, iVAT and pSAT were found, suggesting a minor role for these hormones in the relationship between iron and adiposity. However, these data should be considered cautiously, because important aspects such as the reproductive cycle were not taken into account during sampling.
## 3.1. Iron or Chronic Low-Grade Inflammation: The Role of Iron
Through different mechanisms, high hepcidin and high serum-ferritin levels are a sign of an excess iron accumulation inside adipose tissue cells. Hepcidin functions as a regulator of intestinal iron absorption, iron recycling and mobilization from iron stores [22], while serum ferritin is used as a measurement of iron stores, and high values of ferritin are a sign of iron overload [23]. This excess iron has been related to adipocyte hypertrophy and lower adipogenesis [9], which leads to a dysfunction in adipose tissue and an unhealthy fat-mass expansion [24]. In the current study, the negative association between markers of body iron (haemoglobin, ferritin and iron) and circulating leptin, which was in agreement with previous studies [25,26], might reflect the negative impact of iron excess on healthy fat-mass expansion. Supporting this idea, circulating adiponectin, which is an optimal marker of adipose tissue function, was negatively correlated with circulating ferritin and adipocyte iron excess in humans and mice [6,7].
Further inquiry into the role of iron led us to explore the hepatic iron content, finding the expected positive association with adiposity. However, there were no significant differences in changes in adipose tissue between subjects with high or low hepatic iron, nor in subjects with high ferritin/high hepatic iron against low ferritin/low hepatic iron. In the latter case, differences between those groups were found when comparing baseline and follow-up values of adipose tissue separately. The key behind these results could be found in the group of high ferritin/low hepatic iron, as an increase in ferritin concentration without raised hepatic-iron stores might be a sign of inflammatory conditions [27], as discussed below.
Transferrin has also been a factor in the modulation of adipocyte differentiation, as the administration of transferrin improved the attenuation in adipogenesis caused by iron deficiency [9], and in the modulation of iron overload, as increasing serum transferrin reduced tissue iron accumulation [28]. Transferrin saturation, measured as the ratio between serum iron and TIBC, is an indicator of iron exposure. In iron-overload conditions, transferrin saturation is high, and consequently, serum iron is high or TIBC values are low [29]. These notions are in line with our findings. in which negative associations were observed in both transferrin and TIBC with respect to changes in iSAT and iVAT.
## 3.2. The Role of Chronic Low-Grade Inflammation
Higher hepcidin is also known to be associated with inflammatory markers such as, interleukin-6 (IL-6) and CRP [30], leading to lower iron in circulation and difficulties of the proliferation of infectious organisms in the bloodstream [31]. Serum-ferritin levels are also found to be increased during inflammation, and this is known as an inflammatory marker [32]; its involvement could have a protective role, by limiting the production of free radicals and additional pro-inflammatory effects [33].
It is well known that adipose tissue becomes an active metabolic tissue in obesity, leading to alterations in innate immunity and translating to a chronic low-grade inflammatory state [34]. Among those, IL-6 [35] and CRP [36] are up-regulated in parallel with adipose tissue enlargement in subjects with obesity. Current baseline results showed that CRP was very strongly associated with abdominal adipose tissue indexes. In addition to this, previous studies have found that elevated inflammatory markers are associated with weight gain, especially in the case of CRP [15,16].
The associations of hepcidin and ferritin with iVAT and iSAT, respectively, were also found within the group of women. Men typically show increased serum ferritin [37] in parallel with hepcidin [38]. In this context, increased concentrations of hepcidin or ferritin could be the result of expression of a facilitated adipose-tissue expansion in the case of excessive caloric intake, given the well-known requirements of iron for cell replication (in fact, DNA polymerase is an iron-requiring enzyme). Women and subjects without obesity would be more sensitive to these effects. This could be because men and subjects with obesity have already reached a plateau of iron stores, above which the relationships lack linearity. In a state of chronic low-grade inflammation, increases in hepcidin and ferritin would lead to increased iron uptake by adipose tissue cells, an increase in adipose tissue volumetry and to adipocyte dysfunction over time.
## 3.3. The Role of Insulin Sensitivity
Insulin resistance is known to be associated with body iron accumulation [8] and high serum ferritin [39], as well as having a strong relationship with obesity [40], even though some authors have suggested that insulin resistance is the primary driver of inflammation in adipose tissue [41]. Differences in insulin action are known to be associated with redistribution of adipose tissue compartments. We observed a negative association between baseline iVAT and changes in the glucose-infusion rate during the clamp, while the multiple-linear-regression models showed that changes in insulin sensibility were positively associated with changes in pSAT and that baseline insulin sensitivity was negatively associated with changes in iSAT and iVAT. In cross-sectional studies, VAT has been repeatedly identified as the main compartment associated with insulin resistance [42,43,44]. There are relatively few longitudinal studies in the literature. In one report, the subjects with obesity were classified into two groups according to baseline insulin resistance measured using the clamp. In the follow up after 5 years, visceral fat was significantly increased, with the steepest increase found in the obesity insulin-resistant group [45]. Our findings could be seen as confirmatory of these previous ones, but even a short follow-up of one year could point towards a bidirectionality between visceral adiposity and insulin resistance.
The study has its limitations. In some subjects with morbid obesity, subcutaneous tissue was not covered in full during image acquisition, and therefore these volumes were not segmented, which would lead to an underestimation of iSAT values. However, this limitation represents the fact that the findings could be even more significant if the whole SAT abdominal segment could be covered. Contrary to our results, in some studies [1,2,3] serum ferritin was significantly associated with SAT and VAT or increased in subjects with obesity. It is worth noting that we did not include subjects with type 2 diabetes mellitus, and this could influence current results.
Future studies should be designed to investigate whether the early correction of iron excess in humans might prevent the unhealthy expansion of abdominal adipose tissue.
## 4.1. Study Design
A total of 175 participants were recruited consecutively at the Dr. Josep Trueta University Hospital facilities in Girona, Spain from January 2016 to July 2018. From February 2017 to April 2019, 87 of the previously recruited subjects underwent the same protocol as a follow-up assessment. Subjects with obesity were referred from the Endocrinology Department of Dr. Josep Trueta Hospital and from registered research databases, and the subjects without obesity contacted us through e-mail contact or telephone number. This study was approved by the Scientific Research Ethics Committee in September 2015, and all subjects provided informed written consent prior to inclusion. Participants had a first visit where they underwent neuropsychological testing and demographic- and medical-data collection, while MRI acquisitions were performed at a later date. Inclusion criteria were age over 30 years and the ability to understand study procedures. Subjects with BMI ≥ 30 kg/m2 were considered for the obesity group and their matched counterparts (BMI from 18.5 to 30 kg/m2), according to age and sex, were included in the control group. Exclusion criteria were type 2 diabetes mellitus, NAFLD or liver cirrhosis, major eating or psychiatric-disorder antecedents, anemia or hemoglobinopathy, language disorders, neurological diseases, history of trauma or brain injury, alcohol intake >20 g per day, infection in the last month, serious chronic illness, pregnancy, lactation and MRI contraindications, such as claustrophobia or ferro-magnetic implants. No participants used iron supplements during the follow-up period. Diabetes was excluded, according to the American Diabetes Association criteria, which include fasting glucose equal to or higher than 126 mg/dL and HbA1c equal to or higher than $6.5\%$.
## 4.2. Clinical and Laboratory Features
Participants underwent anthropometric measurements of height and weight. The body mass index was calculated as weight/height squared (kg/m2). Measurements of whole-blood hemoglobin, hematocrit, total white blood cells, iron, ultrasensitive C-reactive protein, transferrin, transferrin saturation (total iron-binding capacity), uric acid, ferritin, glucose, total cholesterol, HDL and LDL cholesterol and triglycerides were performed in the clinical laboratory of the Hospital Dr Josep Trueta in Girona using a routine laboratory test, as detailed elsewhere [4,46]. Circulating hepcidin levels in serum were measured by a solid phase enzyme-linked immunosorbent assay (ELISA) (DRG® Hepcidin 25 (Bioactive) (EIA-5258, DRG International, Inc., Marburg, Germany). Detection limit was 0.35 ng/mL. Intra- and inter-assay coefficients of variation were between 5 and $15\%$. Serum insulin was measured using a Human Insulin ELISA kit (RIS006R, Biovendor—Laboratorni medicina, a.s., Brno, Czech Republic) with intra- and inter-assay coefficient of variation <7 and <$10\%$, respectively. Insulin resistance was estimated by using the formula of the homeostatic model assessment of insulin resistance (HOMA-IR) as: (Fasting glucose (mg/dL) × Fasting insulin (μIU/mL))/405. Plasma leptin levels were measured by Human Leptin ELISA kit (RAB0333-1KT, Merck Life Science S.L.U., Madrid, Spain). Longitudinal changes were calculated as follows: [(Follow up − baseline)/baseline] × 100.
The determination of adrenal- and gonadal-steroid hormones in serum was carried out using liquid-chromatography with tandem-mass-spectrometry (LC–MS/MS) detection, as previously described [47]. Briefly, sample preparation involved solid-phase extraction in an automated unit from Spark Holland (Emmen, The Netherlands), which was on-line coupled to a LC–MS/MS with a triple quadrupole mass detector from Agilent (Palo Alto, CA, USA). Chromatographic separation of steroids was carried out using a Kinetex C18 analytical column (particle size 2.6 μm, 10 cm length, and 3 mm inner diameter) from Phenomenex (Torrance, CA, USA). The MS/MS detection was carried out in multiple reaction monitoring (MRM) with electrospray ionization in fast-switching polarity mode. Information about the capillary voltage, the nebulizer pressure and parameters for MRM detection are detailed elsewhere [47]. Calibration models were prepared by analysis of aliquots of a serum pool spiked with variable concentrations of the target steroids. The endogenous content of each steroid in the sample loop was subtracted in the preparation of the calibration models. Isotopically labeled steroids were used as internal standards for the quantitative analysis of structurally similar analytes. Serum samples were collected under 8h fasting conditions, between 8.00 and 9.00 a.m., without taking into account the reproductive cycle affecting estradiol and progesterone levels.
Insulin action was determined by the hyperinsulinemic–euglycemic clamp. After an overnight fast, two catheters were inserted into an antecubital vein, one for each arm, used to administer constant infusions of glucose and insulin and to obtain arterialized-venous-blood samples. A 2 h hyperinsulinemic–euglycemic clamp was initiated by a two-step primed infusion of insulin (80 mU/m2/min for 5 min, 60 mU/m2/min for 5 min) immediately followed by a continuous infusion of insulin at a rate of 40 mU/m2/min (regular insulin [Actrapid; Novo Nordisk, Plainsboro, NJ, USA]). Glucose infusion began at min 4 at an initial perfusion rate of 2 mg/kg/min, and it was then adjusted to maintain plasma glucose concentration at 88.3–99.1 mg/dL. Blood samples were collected every 5 min, for the determination of plasma glucose and insulin. Insulin sensitivity was assessed as the mean glucose-infusion rate during the last 40 min. In the stationary equilibrium, the amount of glucose administered (M) equals the glucose taken in by the body tissues, and is a measure of overall insulin sensitivity.
## 4.3. Image Acquisition
The MRI study was performed using a 1.5 Tesla scanner (Ingenia Philips Medical Systems, Eindhoven, The Netherlands) using an 8-channel receiver-coil array. The imaging protocol included three-dimensional volumetric mDIXON gradient-recalled echo acquisitions in the axial plane covering the whole abdominal area, from which fat and water images and in-phase and out-of-phase images were reconstructed (matrix size = 180 × 154, field-of-view = 445 × 381 mm, repetition time (TR) = 5.9 ms, excitation time (TE) = 1.8 and 4 ms, flip angle = 15°, number of slices = 50, acquired voxel volume = 2.5 × 2.5 × 10 mm, reconstructed voxel volume = 2 × 2 × 5 mm). Two stacks were acquired, the first one from the diaphragm to the kidneys, and the second one from the kidneys to below the pubic symphysis. Each stack was acquired within an 11s breath-hold task. R2* values were obtained from the mDIXON sequence. Additional sequences from the protocol include T2* (TR = 120 ms, TE = 14 ms, flip angle = 20°) and proton density (TR = 120 ms, TE = 4 ms, flip angle = 20°).
## 4.4. Image Processing
Hepatic iron content (HIC) was assessed from the R2* values, T2* and proton-density sequences, as described elsewhere [4]. In brief, R2* was calculated as 1/T2* by fitting the monoexponential terms to the T2* signal-decay curve of the respective echo times. R2* values were obtained from the mean averages of signal intensity by drawing three regions of interest (ROI) at the liver. HIC was calculated following the recommendations established by the Spanish Society of Abdominal Diagnostic Imaging (SEDIA). All image analyses were performed by trained and experienced technicians, blinded to clinical information. Abdominal-adipose-tissue volumetric measurements were obtained from the two stacks of mDIXON fat-only images. Scans were converted from DICOM format to NIfTI using the dcm2nii conversion tool (https://people.cas.sc.edu/rorden/mricron/dcm2nii.html, accesses on 8 February 2022) and stitched together using the fslmerge command from the FMRIB Software Library (http://www.fmrib.ox.ac.uk/fsl, accesses on 8 February 2022). The segmentation of volumes of interest (VOI) of SAT and VAT was performed using the software Imfusion Labels (ImFusion GmbH, Munich, Germany). The coverage range was from the superior pole of the left kidney to the pubic symphysis. Percentage of SAT (pSAT) was computed as the ratio of SAT/(VAT+SAT) × 100. SAT and VAT values for each subject were divided by the square of its VOI height (computed as the number of slices of each VOI multiplied by the interslice space of the transverse plane) to minimize the variability due to body morphology. This correction was performed in a similar way as that in which BMI is computed, as the body mass divided by the square of the body height, and the resulting subcutaneous adipose tissue index (iSAT) and visceral adipose tissue index (iVAT) are measured in L/m2.
## 4.5. Statistical Analysis
Data was expressed as mean ± standard deviation for normally distributed parameters or median (interquartile range) for non-normally distributed parameters. Normality assumption was checked with the Shapiro–Wilk test. Comparison between two groups were performed with Student’s t-test if n > 30 in both groups, otherwise the Wilcoxon rank-sum test was used. ANOVA and post-hoc multiple comparisons tests were applied for comparison between three groups resulting from sex-stratified tertiles of changes in adipose-tissue parameters. Linear correlations between variables were analyzed with Spearman’s rank coefficient. We performed several sets of correlations between (a) baseline adipose-tissue volumetries against iron-metabolism and chronic-inflammation parameters; (b) changes in adipose tissue against baseline iron-metabolism and chronic-inflammation parameters; and (c) changes in insulin sensitivity against baseline adipose-tissue, iron-metabolism and chronic-inflammation parameters. Multiple-linear-regression analysis was performed to observe associations between changes in adipose tissue and other relevant indicators. All statistical analysis was performed with the maximum sample size according to the available data for the variables analyzed. Statistical significance was assumed when the p value was <0.05. All the statistical analysis was computed using R version 4.1.2.
## 5. Conclusions
In summary, increased parameters for iron metabolism and chronic inflammation were associated with the redistribution of adipose tissue both cross-sectionally and longitudinally. Baseline serum-hepcidin and serum-ferritin concentrations, as well as low transferrin and TIBC, were the main iron parameters associated with an increase in subcutaneous and visceral adipose tissue after one year of follow up. This outcome could be related to the low-grade inflammation state, characteristic of obesity, and a status of iron overload. These associations were independent of circulating major steroids known to be deregulated in obesity. Under these circumstances, adipocyte hypertrophy and lower adipogenesis triggers an unhealthy expansion of abdominal adipose tissue, while excess abdominal adiposity is associated with an increase in insulin resistance. Therefore, having an iron-overload influence on the expansion of visceral adipose tissue independently of insulin sensitivity suggests that correcting early iron excess could have a preventive effect on insulin-resistance progression associated with visceral adiposity.
## References
1. Iwasaki T., Nakajima A., Yoneda M., Yamada Y., Mukasa K., Fujita K., Fujisawa N., Wada K., Terauchi Y.. **Serum Ferritin Is Associated With Visceral Fat Area and Subcutaneous Fat Area**. *Diabetes Care* (2005) **28** 2486-2491. DOI: 10.2337/diacare.28.10.2486
2. Kerkadi A., Ali R.M., Shehada A.A.H., AbouHassanein E.A., Moawad J., Bawadi H., Shi Z.. **Association between central obesity indices and iron status indicators among Qatari adults**. *PLoS ONE* (2021) **16**. DOI: 10.1371/journal.pone.0250759
3. Wu H., Qi Q., Yu Z., Sun L., Li H., Lin X.. **Opposite associations of trunk and leg fat depots with plasma ferritin levels in middle-aged and older Chinese men and women**. *PLoS ONE* (2010) **5**. DOI: 10.1371/journal.pone.0013316
4. Moreno-Navarrete J.M., Moreno M., Puig J., Blasco G., Ortega F.J., Xifra G., Ricart W., Fernández-Real J.M.. **Hepatic iron content is independently associated with serum hepcidin levels in subjects with obesity**. *Clin. Nutr.* (2017) **36** 1434-1439. DOI: 10.1016/j.clnu.2016.09.022
5. Cheng H.L., Bryant C., Cook R., O’Connor H., Rooney K., Steinbeck K.. **The relationship between obesity and hypoferraemia in adults: A systematic review**. *Obes. Rev.* (2012) **13** 150-161. DOI: 10.1111/j.1467-789X.2011.00938.x
6. Wlazlo N., Van Greevenbroek M.M., Ferreira I., Jansen E.H., Feskens E.J., Van Der Kallen C.J., Schalkwijk C.G., Bravenboer B., Stehouwer C.D.. **Iron metabolism is associated with adipocyte insulin resistance and plasma adiponectin: The Cohort on Diabetes and Atherosclerosis Maastricht (CODAM) study**. *Diabetes Care* (2013) **36** 309-315. DOI: 10.2337/dc12-0505
7. Gabrielsen J.S., Gao Y., Simcox J.A., Huang J., Thorup D., Jones D., Cooksey R.C., Gabrielsen D., Adams T.D., Hunt S.C.. **Adipocyte iron regulates adiponectin and insulin sensitivity**. *J. Clin. Investig.* (2012) **122** 3529-3540. DOI: 10.1172/JCI44421
8. Moreno-Navarrete J.M., Novelle M.G., Catalán V., Ortega F., Moreno M., Gomez-Ambrosi J., Xifra G., Serrano M., Guerra E., Ricart W.. **Insulin resistance modulates iron-related proteins in adipose tissue**. *Diabetes Care* (2014) **37** 1092-1100. DOI: 10.2337/dc13-1602
9. Moreno-Navarrete J.M., Ortega F., Moreno M., Ricart W., Fernández-Real J.M.. **Fine-tuned iron availability is essential to achieve optimal adipocyte differentiation and mitochondrial biogenesis**. *Diabetologia* (2014) **57** 1957-1967. DOI: 10.1007/s00125-014-3298-5
10. Moreno-Navarrete J.M., Ortega F., Moreno M., Serrano M., Ricart W., Fernández-Real J.M.. **Lactoferrin gene knockdown leads to similar effects to iron chelation in human adipocytes**. *J. Cell. Mol. Med.* (2014) **18** 391-395. DOI: 10.1111/jcmm.12234
11. Dongiovanni P., Ruscica M., Rametta R., Recalcati S., Steffani L., Gatti S., Girelli D., Cairo G., Magni P., Fargion S.. **Dietary iron overload induces visceral adipose tissue insulin resistance**. *Am. J. Pathol.* (2013) **182** 2254-2263. DOI: 10.1016/j.ajpath.2013.02.019
12. Yan H.F., Liu Z.Y., Guan Z.A., Guo C.. **Deferoxamine ameliorates adipocyte dysfunction by modulating iron metabolism in ob/ob mice**. *Endocr. Connect.* (2018) **7** 604-616. DOI: 10.1530/EC-18-0054
13. Zhang Z., Funcke J.-B., Zi Z., Zhao S., Straub L.G., Zhu Y., Zhu Q., Crewe C., An Y.A., Chen S.. **Adipocyte iron levels impinge on a fat-gut crosstalk to regulate intestinal lipid absorption and mediate protection from obesity**. *Cell Metab.* (2021) **33** 1624-1639.e9. DOI: 10.1016/j.cmet.2021.06.001
14. Rospleszcz S., Dermyshi D., Müller-Peltzer K., Strauch K., Bamberg F., Peters A.. **Association of serum uric acid with visceral, subcutaneous and hepatic fat quantified by magnetic resonance imaging**. *Sci. Rep.* (2020) **10** 442. DOI: 10.1038/s41598-020-57459-z
15. Thompson A.L., Koehler E., Herring A.H., Paynter L., Du S., Zhang B., Popkin B., Gordon-Larsen P.. **Weight Gain Trajectories Associated With Elevated C-Reactive Protein Levels in Chinese Adults**. *J. Am. Heart Assoc.* (2016) **5** e003262. DOI: 10.1161/JAHA.116.003262
16. Holz T., Thorand B., Döring A., Schneider A., Meisinger C., Koenig W.. **Markers of Inflammation and Weight Change in Middle-Aged Adults: Results From the Prospective MONICA/KORA S3/F3 Study**. *Obesity* (2010) **18** 2347-2353. DOI: 10.1038/oby.2010.73
17. Purnell J.Q., Brandon D.D., Isabelle L.M., Loriaux D.L., Samuels M.H.. **Association of 24-hour cortisol production rates, cortisol-binding globulin, and plasma-free cortisol levels with body composition, leptin levels, and aging in adult men and women**. *J. Clin. Endocrinol. Metab.* (2004) **89** 281-287. DOI: 10.1210/jc.2003-030440
18. Bracht J.R., Vieira-Potter V.J., De Souza Santos R., Öz O.K., Palmer B.F., Clegg D.J.. **The role of estrogens in the adipose tissue milieu**. *Ann. N. Y. Acad. Sci.* (2020) **1461** 127-143. DOI: 10.1111/nyas.14281
19. De Maddalena C., Vodo S., Petroni A., Aloisi A.M.. **Impact of testosterone on body fat composition**. *J. Cell. Physiol.* (2012) **227** 3744-3748. DOI: 10.1002/jcp.24096
20. Hennigar S.R., Berryman C.E., Harris M.N., Karl J.P., Lieberman H.R., McClung J.P., Rood J.C., Pasiakos S.M.. **Testosterone Administration During Energy Deficit Suppresses Hepcidin and Increases Iron Availability for Erythropoiesis**. *J. Clin. Endocrinol. Metab.* (2020) **105** dgz316. DOI: 10.1210/clinem/dgz316
21. Beggs L.A., Yarrow J.F., Conover C.F., Meuleman J.R., Beck D., Morrow M., Zou B., Shuster J.J., Borst S.E.. **Testosterone alters iron metabolism and stimulates red blood cell production independently of dihydrotestosterone**. *Am. J. Physiol. Endocrinol. Metab.* (2014) **307** E456-E461. DOI: 10.1152/ajpendo.00184.2014
22. Ganz T.. **Hepcidin—A regulator of intestinal iron absorption and iron recycling by macrophages**. *Best Pract. Res. Clin. Haematol.* (2005) **18** 171-182. DOI: 10.1016/j.beha.2004.08.020
23. Jacobs A., Miller F., Worwood M., Beamish M.R., Wardrop C.A.J.. **Ferritin in the serum of normal subjects and patients with iron deficiency and iron overload**. *Br. Med. J.* (1972) **4** 206. DOI: 10.1136/bmj.4.5834.206
24. Longo M., Zatterale F., Naderi J., Parrillo L., Formisano P., Raciti G.A., Beguinot F., Miele C.. **Adipose Tissue Dysfunction as Determinant of Obesity-Associated Metabolic Complications**. *Int. J. Mol. Sci.* (2019) **20**. DOI: 10.3390/ijms20092358
25. Gao Y., Liu J., Bai Z., Sink S., Zhao C., Lorenzo F.R., McClain D.A.. **Iron down-regulates leptin by suppressing protein**. *J. Biol. Chem.* (2019) **294** 5487-5495. DOI: 10.1074/jbc.RA118.005183
26. Gao Y., Li Z., Gabrielsen J.S., Simcox J.A., Lee S.-H., Jones D., Cooksey B., Stoddard G., Cefalu W.T., McClain D.A.. **Adipocyte iron regulates leptin and food intake**. *J. Clin. Investig.* (2015) **125** 3681-3691. DOI: 10.1172/JCI81860
27. Koperdanova M., Cullis J.O.. **Interpreting raised serum ferritin levels**. *Br. Med. J.* (2015) **351** h3692-h3695. DOI: 10.1136/bmj.h3692
28. Tolosano E.. **Increasing serum transferrin to reduce tissue iron overload due to ineffective erythropoiesis**. *Haematologica* (2015) **100** 565. DOI: 10.3324/haematol.2015.124966
29. Faruqi A., Mukkamalla S.K.R.. **Iron Binding Capacity**. *StatPearls* (2022)
30. Bekri S., Gual P., Anty R., Luciani N., Dahman M., Ramesh B., Iannelli A., Staccini–Myx A., Casanova D., Ben Amor I.. **Increased Adipose Tissue Expression of Hepcidin in Severe Obesity Is Independent From Diabetes and NASH**. *Gastroenterology* (2006) **131** 788-796. DOI: 10.1053/j.gastro.2006.07.007
31. Fleming R.E.. **Hepcidin Activation During Inflammation: Make It STAT**. *Gastroenterology* (2007) **132** 447-449. DOI: 10.1053/j.gastro.2006.11.049
32. Kell D.B., Pretorius E.. **Serum ferritin is an important inflammatory disease marker, as it is mainly a leakage product from damaged cells**. *Metallomics* (2014) **6** 748-773. DOI: 10.1039/C3MT00347G
33. Kernan K.F., Carcillo J.A.. **Hyperferritinemia and inflammation**. *Int. Immunol.* (2017) **29** 401. DOI: 10.1093/intimm/dxx031
34. Lee H., Lee I.S., Choue R.. **Obesity, Inflammation and Diet**. *Pediatr. Gastroenterol. Hepatol. Nutr.* (2013) **16** 143. DOI: 10.5223/pghn.2013.16.3.143
35. Galic S., Oakhill J.S., Steinberg G.R.. **Adipose tissue as an endocrine organ**. *Mol. Cell. Endocrinol.* (2010) **316** 129-139. DOI: 10.1016/j.mce.2009.08.018
36. Anty R., Bekri S., Luciani N., Saint-Paul M.-C., Dahman M., Iannelli A., Ben Amor I., Staccini-Myx A., Huet P.-M., Gugenheim J.. **The inflammatory C-reactive protein is increased in both liver and adipose tissue in severely obese patients independently from metabolic syndrome, Type 2 diabetes, and NASH**. *Am. J. Gastroenterol.* (2006) **101** 1824-1833. DOI: 10.1111/j.1572-0241.2006.00724.x
37. Rushton D.H., Barth J.H.. **What is the evidence for gender differences in ferritin and haemoglobin?**. *Crit. Rev. Oncol. Hematol.* (2010) **73** 1-9. DOI: 10.1016/j.critrevonc.2009.03.010
38. Galesloot T.E., Vermeulen S.H., Geurts-Moespot A.J., Klaver S.M., Kroot J.J., Van Tienoven D., Wetzels J.F.M., Kiemeney L.A.L.M., Sweep F.C., Heijer M.D.. **Serum hepcidin: Reference ranges and biochemical correlates in the general population**. *Blood* (2011) **117** e218-e225. DOI: 10.1182/blood-2011-02-337907
39. Wrede C.E., Buettner R., Bollheimer L.C., Schölmerich J., Palitzsch K.D., Hellerbrand C.. **Association between serum ferritin and the insulin resistance syndrome in a representative population**. *Eur. J. Endocrinol.* (2006) **154** 333-340. DOI: 10.1530/eje.1.02083
40. Al-Goblan A.S., Al-Alfi M.A., Khan M.Z.. **Mechanism linking diabetes mellitus and obesity**. *Diabetes Metab. Syndr. Obes. Targets Ther.* (2014) **7** 587. DOI: 10.2147/DMSO.S67400
41. Shimobayashi M., Albert V., Wölnerhanssen B., Frei I.C., Weissenberger D., Meyer-Gerspach A.C., Clement N., Moes S., Colombi M., Meier J.A.. **Insulin resistance causes inflammation in adipose tissue**. *J. Clin. Investig.* (2018) **128** 1538-1550. DOI: 10.1172/JCI96139
42. Hayashi T., Boyko E.J., McNeely M.J., Leonetti D.L., Kahn S.E., Fujimoto W.Y.. **Visceral adiposity, not abdominal subcutaneous fat area, is associated with an increase in future insulin resistance in Japanese Americans**. *Diabetes* (2008) **57** 1269-1275. DOI: 10.2337/db07-1378
43. McLaughlin T., Lamendola C., Liu A., Abbasi F.. **Preferential Fat Deposition in Subcutaneous Versus Visceral Depots Is Associated with Insulin Sensitivity**. *J. Clin. Endocrinol. Metab.* (2011) **96** E1756-E1760. DOI: 10.1210/jc.2011-0615
44. Liu A.W., Song S.O., Hayashi T., Sato K.K., Kahn S.E., Leonetti D.L., Fujimoto W.Y., Boyko E.. **Change in CT-Measured Abdominal Subcutaneous and Visceral but Not Thigh Fat Areas Predict Future Insulin Sensitivity**. *Diabetes Res. Clin. Pract.* (2019) **154** 17. DOI: 10.1016/j.diabres.2019.06.008
45. Tang A., Coster A.C.F., Tonks K.T., Heilbronn L.K., Pocock N., Purtell L., Govendir M., Blythe J., Zhang J., Xu A.. **Longitudinal Changes in Insulin Resistance in Normal Weight, Overweight and Obese Individuals**. *J. Clin. Med.* (2019) **8**. DOI: 10.3390/jcm8050623
46. Moreno-Navarrete J.M., Blasco G., Xifra G., Karczewska-Kupczewska M., Stefanowicz M., Matulewicz N., Puig J., Ortega F.J., Ricart W., Straczkowski M.. **Obesity Is Associated With Gene Expression and Imaging Markers of Iron Accumulation in Skeletal Muscle**. *J. Clin. Endocrinol. Metab.* (2016) **101** 1282-1289. DOI: 10.1210/jc.2015-3303
47. Luque-Córdoba D., Priego-Capote F.. **Fully automated method for quantitative determination of steroids in serum: An approach to evaluate steroidogenesis**. *Talanta* (2021) **224** 121923. DOI: 10.1016/j.talanta.2020.121923
|
---
title: PPAR Pan Agonist MHY2013 Alleviates Renal Fibrosis in a Mouse Model by Reducing
Fibroblast Activation and Epithelial Inflammation
authors:
- Minjung Son
- Ga Young Kim
- Yejin Yang
- Sugyeong Ha
- Jeongwon Kim
- Doyeon Kim
- Hae Young Chung
- Hyung Ryong Moon
- Ki Wung Chung
journal: International Journal of Molecular Sciences
year: 2023
pmcid: PMC10002481
doi: 10.3390/ijms24054882
license: CC BY 4.0
---
# PPAR Pan Agonist MHY2013 Alleviates Renal Fibrosis in a Mouse Model by Reducing Fibroblast Activation and Epithelial Inflammation
## Abstract
The peroxisome proliferator-activated receptor (PPAR) nuclear receptor has been an interesting target for the treatment of chronic diseases. Although the efficacy of PPAR pan agonists in several metabolic diseases has been well studied, the effect of PPAR pan agonists on kidney fibrosis development has not been demonstrated. To evaluate the effect of the PPAR pan agonist MHY2013, a folic acid (FA)-induced in vivo kidney fibrosis model was used. MHY2013 treatment significantly controlled decline in kidney function, tubule dilation, and FA-induced kidney damage. The extent of fibrosis determined using biochemical and histological methods showed that MHY2013 effectively blocked the development of fibrosis. Pro-inflammatory responses, including cytokine and chemokine expression, inflammatory cell infiltration, and NF-κB activation, were all reduced with MHY2013 treatment. To demonstrate the anti-fibrotic and anti-inflammatory mechanisms of MHY2013, in vitro studies were conducted using NRK49F kidney fibroblasts and NRK52E kidney epithelial cells. In the NRK49F kidney fibroblasts, MHY2013 treatment significantly reduced TGF-β-induced fibroblast activation. *The* gene and protein expressions of collagen I and α-smooth muscle actin were significantly reduced with MHY2013 treatment. Using PPAR transfection, we found that PPARγ played a major role in blocking fibroblast activation. In addition, MHY2013 significantly reduced LPS-induced NF-κB activation and chemokine expression mainly through PPARβ activation. Taken together, our results suggest that administration of the PPAR pan agonist effectively prevented renal fibrosis in both in vitro and in vivo models of kidney fibrosis, implicating the therapeutic potential of PPAR agonists against chronic kidney diseases.
## 1. Introduction
Chronic kidney disease (CKD) affects approximately $10\%$ of the global population, with high mortality due to limited treatment options [1]. CKD often leads to end-stage renal disease, which is fatal without renal replacement therapy, such as dialysis or kidney transplantation. Kidney fibrosis is considered a major underlying pathological process that is commonly detected in CKD development [2]. Understanding the mechanisms of renal fibrosis is essential for developing therapies to prevent or slow CKD progression. Fibrosis is defined by the formation and accumulation of the extracellular matrix (ECM), mainly by tissue-resident fibroblast cells [3]. Under physiological conditions, minimal amounts of ECM support kidney structure and function. In response to tissue injury, wound-healing processes are activated to inhibit the inflammatory response with proper tissue regeneration. However, persistent inflammatory responses result in incomplete regeneration, with the formation of fibrotic scar tissue [4]. Exaggerated deposition of ECM during chronic and pathological fibrosis development disrupts the normal kidney architecture and interferes with kidney function. At a certain stage, unresolved kidney fibrosis becomes irreversible and contributes to renal failure.
The mechanisms underlying the development of kidney fibrosis have been studied extensively [5]. Regardless of the trigger, multiple cell types participate in fibrogenesis, including fibroblasts, pericytes, epithelial cells, endothelial cells, and inflammatory cells [6]. The main contributor to fibrosis progression is the accumulation of fibroblasts with a phenotypic appearance of myofibroblasts. During progressive fibrosis, the interstitium is filled with myofibroblasts, which produce large amounts of ECM proteins [6]. Although myofibroblasts are the executing cells of fibrosis, other cells also contribute to the development of fibrosis through both direct and indirect mechanisms. Pericytes, epithelial cells, and endothelial cells have been shown to directly contribute to fibrosis through the transition to mesenchymal-like cell types [7]. Epithelial cells also contribute to fibrosis through the secretion of pro-fibrogenic and pro-inflammatory factors, such as TGF-β, CTGF, and cytokines [8,9]. Considerable evidence suggests that inflammatory cells play a critical role in the initiation and progression of renal fibrosis [10,11]. The chemokine is mainly secreted from tubule epithelial cells during injury and recruits various inflammatory cell types, including monocytes, T cells, dendritic cells, and fibrocytes [9]. The infiltration of inflammatory cells is a major phenotype of kidney fibrosis that promotes fibrosis [12].
Peroxisome proliferator-activated receptors (PPARs), PPARα, PPARβ/δ, and PPARγ, play an essential role in the regulation of various physiological processes, including lipid and energy metabolism [13]. Fibrates (PPARα agonists) are used to treat dyslipidemia, and thiazolidinediones (PPARγ agonists) are used to increase insulin sensitivity in type 2 diabetics. In addition, PPAR dual agonists have been developed to treat type 2 diabetes with secondary cardiovascular complications [14,15]. Many synthetic ligands for PPARs are still under development to expand their therapeutic applications. In addition to their original roles in metabolism, PPAR agonists have been shown to exert various physiological effects. PPAR agonists have been reported to block the development of fibrosis in the liver, heart, kidneys, and lungs [16,17]. Furthermore, several studies have reported the anti-inflammatory action of peroxisome proliferator-activated receptor (PPAR) agonists [18].
Previously, we synthesized and evaluated the role of MHY2013, a potent PPAR pan-agonist, in several metabolic disease models [19,20]. In addition, MHY2013 showed anti-fibrotic effects in an age-related renal fibrosis model by regulating the lipid metabolism in epithelial cells [21]. However, the effects of MHY2013 on general aspects of renal fibrosis have not yet been investigated. In this study, we demonstrated the role and efficacy of MHY2013 in a general renal fibrosis model. Using mouse models of renal fibrosis induced by folic acid, we demonstrated the anti-fibrotic efficacy of PPAR pan agonism in renal fibrosis. MHY2013 treatment significantly reduced fibrosis and inflammation in a mouse model of renal fibrosis. In addition, using in vitro analysis, we found anti-fibrotic and anti-inflammatory effects of MHY2013 in renal fibroblasts and epithelial cells.
## 2.1. MHY2013 Reduces Folic-Acid-Induced Renal Damage and Tubule Dilation in Mice
To evaluate the anti-fibrotic effects of MHY2013, folic-acid-induced renal fibrosis models were used. MHY2013 was intraperitoneally administered at a low (0.5 mg/kg/day) or high dose (3 mg/kg/day) during the experimental period (Figure 1A). The MHY2013-treated group showed lower expression of kidney damage-related genes (Havcr1, Timp2, Igfbp7, and Spp1) than those of the FA-treated group (Figure 1B). Blood urea nitrogen (BUN) levels were increased in the folic acid (FA)-treated group, and high-dose MHY2013 treatment significantly blocked the FA-induced BUN increase (Figure 1C). Structural changes were analyzed with hematoxylin and eosin (H&E) staining. Tubule dilation and damage were detected in the cortex and medulla regions of FA-treated kidneys (Figure 1D). MHY2013-treated kidneys showed a smaller increase in tubule dilation (Figure 1D). These results indicate that MHY2013 has protective effects against folic-acid-induced kidney damage.
## 2.2. MHY2013 Suppresses FA-Induced Renal Fibrosis Development in Mice
We further analyzed the effects of MHY2013 on the development of renal fibrosis. The MHY2013-treated group showed lower expression of fibrosis-related genes (Col1a2, Col3a1, Vim) than that of the FA-treated group (Figure 2A). The increased expression of Col1a2 and Vim was confirmed with in situ hybridization (ISH) analysis. FA treatment significantly increased Col1a2 and Vim expression in the interstitial region of the kidney, and MHY2013-treated groups showed lower Col1a2 and Vim expression (Figure 2B,C). The protein levels of fibrosis markers were further checked. FA-induced α-SMA and collagen I levels were significantly decreased with MHY2013 treatment (Figure 3A). An immunohistochemical analysis confirmed that fewer αSMA-positive myofibroblasts were detected in the MHY2013-treated kidneys (Figure 3B). The extent of fibrosis was confirmed using Sirius Red (SR) staining. The FA treatment significantly increased SR-positive regions, whereas the MHY2013 treatment reduced SR-positive regions (Figure 3C,D). Finally, the activation of SMAD proteins was detected. Less SMAD2 and SMAD3 phosphorylation was detected in the MHY2013-treated groups than in the FA groups (Figure 3E). Collectively, these data indicate that MHY2013 effectively blocked FA-induced kidney fibrosis.
## 2.3. FA-Induced Inflammatory Responses Are Down-Regulated by MHY2013
The development of fibrosis is accompanied by pro-inflammatory responses. FA treatment also increases the inflammatory responses in the kidneys [22]. We further examined the inflammatory responses in animal models. MHY2013 treatment significantly reduced pro-inflammatory gene (Tnfa, Il1b, and Ccl2) expression and the macrophage marker Emr1 in the kidneys (Figure 4A). The activation of NF-κB, induced by FA, was effectively blocked with MHY2013 treatment (Figure 4B). Activated NF-κB was mainly detected in the epithelial cells of dilated tubules, and MHY2013 significantly reduced p-NF-κB expression in tubule cells (Figure 4C). Macrophage infiltration was confirmed using ISH analysis. Increased Emr1 expression was mainly detected in the interstitial region of FA-treated kidneys (Figure 4D). MHY2013-treated groups showed less macrophage infiltration in the kidneys (Figure 4D). We further detected the co-expression of Col1a2 and Emr1. In the FA group, Emr1- and Col1a2-positive cells were colocalized in the kidneys, indicating that inflammation is connected to fibrosis development (Figure 4D). In accordance with the qPCR results, the MHY2013-treated group showed lower Emr1 and Col1a2 expression in the kidney (Figure 4D). These results indicate that MHY2013 exerts anti-inflammatory effects against FA-induced kidney fibrosis.
## 2.4. MHY2013 Blocks TGF-β-Induced NRK49F Kidney Fibroblast Activation
To investigate the anti-fibrotic role of MHY2013 under in vitro conditions, we used kidney-derived fibroblast cells. First, we confirmed the activation of PPAR by MHY2013 in NRK49F kidney fibroblasts. MHY2013 significantly increased PPRE activity under PPARα, PPARβ, and PPARγ expression conditions, confirming MHY2013 as a PPAR pan agonist (Figure 5A–C). TGF-β treatment significantly increased Col1a2, Acta2, and Vim expression in NRK49F fibroblasts, and MHY2013 pre-treatment effectively blocked fibroblast activation (Figure 5D). The protein expression levels of α-SMA and Col1 were analyzed. MHY2013 treatment significantly reduced TGF-β-induced α-SMA and Col1 protein expression (Figure 5E). The increased expression of α-SMA was confirmed using immunofluorescence. TGF-β increased αSMA expression in cells, whereas MHY2013 reduced αSMA expression (Figure 5F). To examine which PPAR subtype influenced fibroblast activation, we overexpressed PPAR before TGF-β treatment. We found that PPARγ overexpression effectively blocked TGF-β-induced fibroblast activation (Figure 5G), whereas other PPAR subtypes did not show a significant reduction (data not shown). These results indicate that MHY2013 effectively blocks TGF-β-induced NRK49F kidney fibroblast activation, mainly through PPARγ activation.
## 2.5. MHY2013 Reduces LPS-Induced Chemokine Expression in NRK52E Kidney Epithelial Cells
To examine the anti-inflammatory effects of MHY2013 under in vitro conditions, kidney tubule epithelial cells were used. Stimulation of NRK52E cells with a lipopolysaccharide (LPS) significantly increased chemokine gene expression, and MHY2013 pretreatment effectively reduced their expression (Figure 6A). We further evaluated NF-κB activity using a luciferase assay. LPS treatment significantly increased NF-κB activity, whereas MHY2013 effectively blocked NF-κB activity (Figure 6B). Finally, to examine which PPAR subtype influences LPS-induced chemokine expression, we overexpressed PPAR before LPS treatment. We found that PPARβ overexpression effectively blocked LPS-induced chemokine expression (Figure 6C), whereas other PPAR subtypes did not show a significant reduction (data not shown). Collectively, these data show that MHY2013 reduces LPS-induced NF-κB activation and chemokine expression in renal epithelial cells, mainly through PPARβ activation.
## 3. Discussion
Renal fibrosis, which is generally accompanied by CKD progression, is defined by the loss of renal parenchymal cells and their substitution with ECM proteins. During fibrosis development, both the synthesis and degradation of ECM proteins occur via several intra- and extracellular events. When ECM protein synthesis exceeds degradation, excessive ECM accumulation results in fibrosis [23]. It is well established that various cell types directly and indirectly participate in fibrosis development. Resident fibroblasts are the main responsible cells for the synthesis of ECM proteins [3]. During fibrogenesis, fibroblasts receive signals from other cells and begin to proliferate and become myofibroblasts. Myofibroblasts produce large amounts of ECM proteins that primarily contribute to the pathogenesis of kidney fibrosis.
Transforming growth factor-β (TGF-β) is considered a key player of renal fibrosis by stimulating fibroblasts in the kidney, thus making it an interesting target for the treatment of fibrosis [24]. Indeed, anti-TGF-β treatments using neutralizing antibodies, inhibitors against the TGF-β receptor, or antisense oligonucleotides to TGF-β1 halt the progression of renal fibrosis development, suggesting its fibrotic role in CKD [25]. We found that MHY2013 significantly reduced TGF-β-induced fibroblast activation in vitro. MHY2013 effectively inhibits TGF-β-induced α-SMA and collagen I expression in fibroblasts. Several studies have reported that PPARγ activation blocks TGF-β-induced ECM production in fibroblasts. Wang et al. evaluated three PPARγ agonists (15d-PGJ2, troglitazone, and ciglitazone) and found that PPARγ activation directly inhibits TGF-β/SMAD signaling pathways and alleviates renal fibroblast activation, resulting in reduced ECM synthesis [26]. Another PPARγ agonist, pioglitazone, similarly prevents renal fibrosis by repressing the TGF-β signaling pathway [27]. MHY2013 also showed direct anti-fibrotic effects on fibroblasts. Using PPAR transfection, we found that PPARγ overexpression inhibits TGF-β-induced fibroblast activation. Based on these results, we concluded that MHY2013 directly reduces fibroblast activation through PPARγ activation.
Renal inflammation is a protective response induced during kidney injury, which eliminates the cause of injury and promotes tissue repair. However, unresolved inflammatory responses can promote abnormal fibrosis in the kidneys, leading to CKD [28]. During prolonged inflammation, bone-marrow-derived leukocytes, including neutrophils and macrophages, are the main players in kidney inflammation. The accumulation of these cells is a major feature of pro-inflammatory kidney disease. In addition to these cells, studies have also revealed the important role of locally activated kidney cells, such as tubular epithelial cells (TECs), mesangial cells, podocytes, and endothelial cells. During the development of interstitial fibrosis, TECs play an important role in initiating the inflammatory response [29]. Under damaged conditions, TECs actively participate in pro-inflammatory responses through chemokine production. Several lines of evidence suggest that chemokines produced from TECs are crucial for the recruitment of monocytes and macrophages [30]. Based on these observations, the regulation of epithelial inflammation has been an interesting target for modulating kidney inflammation and fibrosis.
Based on our finding that MHY2013 decreases inflammation in animal models, we further demonstrated its role in epithelial inflammation. MHY2013 significantly reduces NF-κB activation and chemokine production in epithelial cells. Furthermore, using PPAR subtype transfection, we found that PPARβ overexpression decreases chemokine production in epithelial cells. There is evidence that PPARβ exerts anti-inflammatory effects in kidney disease. PPARβ-null mice developed more severe ischemic renal injury with more severe tubule damage than wild-type mice [31]. A macrophage-specific PPARβ-deleted mouse model also showed impaired apoptotic cell clearance and reduced anti-inflammatory cytokine production [32]. These mice were much more likely to develop autoimmune kidney disease, a lupus-like autoimmune disease. In addition, several reports have demonstrated the anti-inflammatory role of PPARβ agonists in kidney disease. GW0742 has been shown to inhibit streptozotocin-induced diabetic nephropathy in mice by reducing inflammatory mediators, including MCP-1 and osteopontin [33]. Another study showed that PPARβ agonists reduced the incidence of hypertension, endothelial dysfunction, inflammation, and organ damage in lupus mice [34]. Collectively, the reduced inflammatory responses observed in our in vitro and in vivo experiments were associated with the PPARβ activation property of MHY2013.
## 4.1. Animal Studies
All animal experiments were approved by the Institutional Animal Care Committee of the Pusan National University (PNU-IACUC approval No. PNU-2022-3164) and performed according to the guidelines issued by Pusan National University. C57BL/6J mice were obtained from Hyochang Science (Daegu, Republic of Korea). To establish the renal fibrosis mouse model, male mice (7-week-old) were intraperitoneally injected with a single dose of folic acid (250 mg/kg dissolved in 0.3 M NaHCO3) or vehicle. For the MHY treatment groups, MHY2013 was intraperitoneally administered in low (0.5 mg/kg/day) or high doses (3 mg/kg/day) during the experimental period ($$n = 5$$~7). All mice were maintained at 23 ± 2 °C with a relative humidity of 60 ± $5\%$ and 12 h light/dark cycles. One week after the folic acid treatment, the mice were sacrificed using CO2 inhalation. Serum was collected for biochemical analyses. Kidneys were collected and then immediately frozen in liquid nitrogen. For long-term storage, kidney samples were moved to a −80 °C deep freezer. Part of kidneys was fixed in neutral-buffered formalin for histochemical experiments.
## 4.2. Cell Culture Experiments
NRK49F rat-kidney fibroblasts were purchased from ATCC (CRL-1570) and grown in Dulbecco’s modified Eagle’s medium (DMEM), supplemented with $10\%$ fetal bovine serum (FBS) and $1\%$ penicillin. All cells were incubated at $5\%$ CO2 and 37 °C in a water-saturated atmosphere. To determine the effect of MHY2013 on TGFβ-induced fibroblast activation and ECM production, a MHY2013 concentration with 10 μM was pre-treated 30 min before the TGFβ (10 ng/mL) treatment. Protein or RNA samples were collected 24 h after the TGF-β treatment to determine the effect of MHY2013. NRK52E rat-kidney epithelial cells were purchased from ATCC (CRL-1571) and grown in DMEM supplemented with $10\%$ FBS and $1\%$ penicillin. To determine the effect of MHY2013 on LPS-induced inflammation, a MHY2013 concentration of 10 μM was pre-treated 30 min before LPS (10 μg/mL) treatment. Protein and RNA samples were collected 1 h after LPS treatment to determine the effect of MHY2013. All cell culture experiments were performed at least 3 times per experiment.
## 4.3. Serum Biochemical Measurements
Serum samples were obtained using centrifugation at 3000 rpm for 20 min at 4 °C. Blood urea nitrogen (BUN) levels were measured using a commercial assay kit from Shinyang Diagnostics (SICDIA L-BUN, 1120171, Seoul, Republic of Korea) according to the manufacturer’s instructions.
## 4.4. Protein Extraction and Western Blot Analysis
Two different solutions were used to extract proteins: ProEXTM CETi protein extract solution (Translab, Daejeon, Republic of Korea) was used to extract protein from tissues, and RIPA buffer (#9806, Cell Signaling Technology, Danvers, MA, USA) was used to obtain the total protein from the cells. Both solutions contained protease inhibitor cocktails to prevent protein degradation and phosphate inhibitor to prevent dephosphorylation. Protein concentration was measured using a BCA reagent (Thermo Scientific, Waltham, MA, USA). Extracted proteins (5–20 μg of protein) were then mixed with 4× sample buffer (Cat#1610747, Bio-Rad, CA, USA) and boiled for 5 min. The proteins were then separated using sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transferred to polyvinylidene difluoride membranes (Millipore, Burlington, MA, USA). The membranes were blocked in $5\%$ nonfat milk and washed with Tris-buffered saline-Tween buffer for 30 min. Specific primary antibodies (1:500 to 1:2000 dilution, Supplementary Table S1) were added to the membranes and incubated overnight at 4 °C. After three washes with the TBS-Tween buffer, the membranes were incubated with a horseradish peroxidase-conjugated anti-mouse, anti-rabbit, or anti-goat antibody (diluted 1:10,000) for 1 h at 25 °C. The resulting immunoblots were visualized using Western Bright Peroxide solution (Advansta, San Jose, CA, USA) and a ChemiDoc imaging system (Bio-Rad) according to the manufacturer’s instructions. All western blot analyses were performed at least 3 times per experiment.
## 4.5. RNA Extraction and qRT-PCR
Total RNA was prepared using a TRIzol reagent (Invitrogen, Carlsbad, CA, USA). Briefly, kidney tissues ($$n = 5$$~7) or cells ($$n = 3$$) were homogenized in the TRIzol reagent. To isolate RNA, 0.2 mL chloroform was added to the 1 mL homogenate and shaken vigorously for 15 min. The aqueous phases were transferred to fresh tubes, and an equal volume of isopropanol was added. The samples were then incubated at 4 °C for 15 min and centrifuged at 12,000× g for 15 min at 4 °C. The supernatants were removed, and the resulting RNA pellets were washed once with $75\%$ ethanol and then dried, followed by dissolving in diethyl pyrocarbonate-treated water. Next, 1.0 μg of isolated RNA was reverse-transcribed using a cDNA synthesis kit from GenDEPOT (Katy, TX, USA). qPCR was performed using a SYBR Green Master Mix (BIOLINE, Taunton, MA, USA) and a CFX Connect System (Bio-Rad). Primers were designed using Primer3Plus [35], and the primer sequences used are listed in Supplementary Table S2. For qPCR data analysis, the 2−ΔΔCT method was used as a relative quantification strategy.
## 4.6. Histological Analysis
To visualize histological changes in the kidneys, the kidneys were fixed in $10\%$ neutral formalin, and paraffin-embedded sections were stained with H&E. To assess the degree of renal fibrosis and damage, SR staining was performed using a commercially available kit (VB-3017; Rockville, MD, USA). This staining method is commonly used to visualize collagen fibers, which are a hallmark of fibrosis. Immunohistochemical analysis was performed to visualize the protein expression regions in the kidneys. Briefly, paraffin-embedded sections were deparaffinized and rehydrated. The sections were then incubated with the primary antibodies and visualized using diaminobenzidine substrates. The sections were counterstained with hematoxylin, which allows for the visualization of cell nuclei. Images were obtained using a microscope (LS30; Leam Solution, Seoul, Republic of Korea).
## 4.7. In Situ Hybridization
ISH was performed using formalin-fixed paraffin-embedded tissue samples. RNAscope 2.5 HD Assay (322300, Biotechne, Minneapolis, MN, USA) or RNAscope 2.5 HD Duplex Detection Kit (322436, bio-techne, Minneapolis, MN, USA) was used to visualize RNA expression in the tissue, in accordance with the manufacturer’s instructions. The following probes were used to perform the RNAscope assay: Mm-Vim cat# 457961, Mm-Emr1 cat# 317969-C2, and Mm-Col1a1 cat# 319379. Images were obtained using a microscope (LS30; LEAM Solution, Seoul, Republic of Korea).
## 4.8. Measurement of Transcriptional Activity
Luciferase assays were performed to determine the transcriptional activity of PPAR transcription factors in the NRK49F cells. Briefly, NRK49F cells were transfected with the PPRE-X3-TK-LUC plasmid (0.1 µg) with PPARα, PPARβ/δ, or PPARγ expression vectors (0.01 µg) using Lipofectamine 3000 reagent (Invitrogen, Carlsbad, CA, USA.). The cells were further treated with MHY2013 or WY14643 (a known PPARα agonist), GW501516 (a known PPARβ/δ agonist), and rosiglitazone (a known PPARγ agonist). The luciferase activity was measured using a One-Glo Luciferase Assay System (Promega, Madison, WI, USA). After adding the luciferase substrate, the luminescence was measured using a luminescence plate reader (Berthold Technologies GmbH & Co., Bad Wildbad, Germany). Luciferase assays were performed to determine the transcriptional activity of NF-κB in the NRK52E cells. The cells were transfected with the NF-κB promoter-LUC plasmid, and the luciferase activity was measured using a One-Glo Luciferase Assay System and a luminescence plate reader.
## 4.9. Immunofluorescence
Immunofluorescence was performed to visualize protein expression in the cells. The cells were fixed in $4\%$ formaldehyde for 10 min, washed thrice with ice-cold PBS, and exposed to $0.25\%$ Triton-X 100 in PBS for 10 min for permeabilization. To prevent non-specific binding of antibodies, the cells were blocked using a solution containing $1\%$ BSA and $0.1\%$ Tween 20 in PBS at room temperature for 30 min. Next, the cells were incubated overnight with anti-αSMA antibody, which had been diluted in the blocking buffer at 4 °C. After washing off any unbound antibodies with PBS, the cells were incubated with a secondary antibody conjugated with a fluorescent tag for 1 h in the dark. The cells counterstained with Hoechst 33258 in PBS for 1 min to visualize the nuclei. The images were captured using a fluorescence microscope (LS30).
## 4.10. Quantification and Statistical Analysis
Student’s t-test was used to analyze the differences between the two groups, and an analysis of variance was used to analyze intergroup differences. The level of statistical significance was set at $p \leq 0.05.$ The software used for the analyses was GraphPad Prism version 5 (GraphPad Software Inc., San Diego, CA, USA). Image calculations were performed using the ImageJ software (National Institutes of Health, Bethesda, MD, USA).
## 5. Conclusions
In conclusion, we investigated the anti-fibrotic and anti-inflammatory roles of the PPAR pan agonist MHY2013 using in vitro and in vivo kidney fibrosis models. When administered to the FA-induced mouse kidney fibrosis model, MHY2013 effectively reduced fibrosis development and inflammatory responses in the kidney. The anti-fibrotic and anti-inflammatory mechanisms of MHY2013 were further demonstrated using NRK49F kidney fibroblasts and NRK52E kidney epithelial cells. MHY2013 directly reduced TGF-β-induced ECM production in fibroblasts mainly through PPARγ activation, whereas MHY2013 suppressed LPS-induced pro-inflammatory responses in TECs mainly through PPARβ activation. Taken together, our results suggest that the administration of the PPAR pan agonist effectively prevented renal fibrosis in both in vitro and in vivo models of kidney fibrosis, implicating the therapeutic potential of PPAR agonists against chronic kidney diseases (Figure 7).
## References
1. Kovesdy C.P.. **Epidemiology of chronic kidney disease: An update 2022**. *Kidney Int. Suppl.* (2022) **12** 7-11. DOI: 10.1016/j.kisu.2021.11.003
2. Hwang S., Chung K.W.. **Targeting fatty acid metabolism for fibrotic disorders**. *Arch. Pharm. Res.* (2021) **44** 839-856. DOI: 10.1007/s12272-021-01352-4
3. Kendall R.T., Feghali-Bostwick C.A.. **Fibroblasts in fibrosis: Novel roles and mediators**. *Front. Pharmacol.* (2014) **5** 123. DOI: 10.3389/fphar.2014.00123
4. Wynn T.A., Vannella K.M.. **Macrophages in Tissue Repair, Regeneration, and Fibrosis**. *Immunity* (2016) **44** 450-462. DOI: 10.1016/j.immuni.2016.02.015
5. Zhou D., Liu Y.. **Renal fibrosis in 2015: Understanding the mechanisms of kidney fibrosis**. *Nat. Rev. Nephrol.* (2016) **12** 68-70. DOI: 10.1038/nrneph.2015.215
6. Djudjaj S., Boor P.. **Cellular and molecular mechanisms of kidney fibrosis**. *Mol. Aspects Med.* (2019) **65** 16-36. DOI: 10.1016/j.mam.2018.06.002
7. Humphreys B.D.. **Mechanisms of Renal Fibrosis**. *Annu. Rev. Physiol.* (2018) **80** 309-326. DOI: 10.1146/annurev-physiol-022516-034227
8. Frangogiannis N.. **Transforming growth factor-beta in tissue fibrosis**. *J. Exp. Med.* (2020) **217** e20190103. DOI: 10.1084/jem.20190103
9. Qi R., Yang C.. **Renal tubular epithelial cells: The neglected mediator of tubulointerstitial fibrosis after injury**. *Cell Death Dis.* (2018) **9** 1126. DOI: 10.1038/s41419-018-1157-x
10. Tang P.M., Nikolic-Paterson D.J., Lan H.Y.. **Macrophages: Versatile players in renal inflammation and fibrosis**. *Nat. Rev. Nephrol.* (2019) **15** 144-158. DOI: 10.1038/s41581-019-0110-2
11. Ryu S., Shin J.W., Kwon S., Lee J., Kim Y.C., Bae Y.S., Bae Y.S., Kim D.K., Kim Y.S., Yang S.H.. **Siglec-F-expressing neutrophils are essential for creating a profibrotic microenvironment in renal fibrosis**. *J. Clin. Investig.* (2022) **132** e156876. DOI: 10.1172/JCI156876
12. Black L.M., Lever J.M., Agarwal A.. **Renal Inflammation and Fibrosis: A Double-edged Sword**. *J. Histochem. Cytochem.* (2019) **67** 663-681. DOI: 10.1369/0022155419852932
13. Grygiel-Gorniak B.. **Peroxisome proliferator-activated receptors and their ligands: Nutritional and clinical implications—A review**. *Nutr. J.* (2014) **13** 17. DOI: 10.1186/1475-2891-13-17
14. Charbonnel B.. **PPAR-alpha and PPAR-gamma agonists for type 2 diabetes**. *Lancet* (2009) **374** 96-98. DOI: 10.1016/S0140-6736(09)61040-0
15. Etgen G.J., Oldham B.A., Johnson W.T., Broderick C.L., Montrose C.R., Brozinick J.T., Misener E.A., Bean J.S., Bensch W.R., Brooks D.A.. **A tailored therapy for the metabolic syndrome: The dual peroxisome proliferator-activated receptor-alpha/gamma agonist LY465608 ameliorates insulin resistance and diabetic hyperglycemia while improving cardiovascular risk factors in preclinical models**. *Diabetes* (2002) **51** 1083-1087. DOI: 10.2337/diabetes.51.4.1083
16. Kang H.M., Ahn S.H., Choi P., Ko Y.A., Han S.H., Chinga F., Park A.S., Tao J., Sharma K., Pullman J.. **Defective fatty acid oxidation in renal tubular epithelial cells has a key role in kidney fibrosis development**. *Nat. Med.* (2015) **21** 37-46. DOI: 10.1038/nm.3762
17. Iglarz M., Touyz R.M., Viel E.C., Paradis P., Amiri F., Diep Q.N., Schiffrin E.L.. **Peroxisome proliferator-activated receptor-alpha and receptor-gamma activators prevent cardiac fibrosis in mineralocorticoid-dependent hypertension**. *Hypertension* (2003) **42** 737-743. DOI: 10.1161/01.HYP.0000083511.91817.B1
18. Daynes R.A., Jones D.C.. **Emerging roles of PPARs in inflammation and immunity**. *Nat. Rev. Immunol.* (2002) **2** 748-759. DOI: 10.1038/nri912
19. An H.J., Lee B., Kim S.M., Kim D.H., Chung K.W., Ha S.G., Park K.C., Park Y.J., Kim S.J., Yun H.Y.. **A PPAR Pan Agonist, MHY2013 Alleviates Age-Related Hepatic Lipid Accumulation by Promoting Fatty Acid Oxidation and Suppressing Inflammation**. *Biol. Pharm. Bull.* (2018) **41** 29-35. DOI: 10.1248/bpb.b17-00371
20. An H.J., Lee B., Kim D.H., Lee E.K., Chung K.W., Park M.H., Jeong H.O., Kim S.M., Moon K.M., Kim Y.R.. **Correction: Physiological characterization of a novel PPAR pan agonist, 2-(4-(5,6-methylenedioxybenzo[d]thiazol-2-yl)-2-methylphenoxy)-2-methylpropanoic acid (MHY2013)**. *Oncotarget* (2017) **8** 45030. DOI: 10.18632/oncotarget.18922
21. Chung K.W., Ha S., Kim S.M., Kim D.H., An H.J., Lee E.K., Moon H.R., Chung H.Y.. **PPARalpha/beta Activation Alleviates Age-Associated Renal Fibrosis in Sprague Dawley Rats**. *J. Gerontol.* (2020) **75** 452-458. DOI: 10.1093/gerona/glz083
22. Yang Y., Ha S., Jeong S., Jang C.W., Kim J., Im D.S., Chung H.Y., Chung K.W.. **Comparison of two different toxin-induced kidney fibrosis models in terms of inflammatory responses**. *Toxicology* (2021) **463** 152973. DOI: 10.1016/j.tox.2021.152973
23. Bonnans C., Chou J., Werb Z.. **Remodelling the extracellular matrix in development and disease**. *Nat. Rev. Mol. Cell Biol.* (2014) **15** 786-801. DOI: 10.1038/nrm3904
24. Sureshbabu A., Muhsin S.A., Choi M.E.. **TGF-beta signaling in the kidney: Profibrotic and protective effects**. *Am. J. Physiol. Renal Physiol.* (2016) **310** F596-F606. DOI: 10.1152/ajprenal.00365.2015
25. Gu Y.Y., Liu X.S., Huang X.R., Yu X.Q., Lan H.Y.. **Diverse Role of TGF-beta in Kidney Disease**. *Front. Cell Dev. Biol.* (2020) **8** 123. DOI: 10.3389/fcell.2020.00123
26. Wang W., Liu F., Chen N.. **Peroxisome proliferator-activated receptor-gamma (PPAR-gamma) agonists attenuate the profibrotic response induced by TGF-beta1 in renal interstitial fibroblasts**. *Mediators Inflamm.* (2007) **2007** 62641. DOI: 10.1155/2007/62641
27. Nemeth A., Mozes M.M., Calvier L., Hansmann G., Kokeny G.. **The PPARgamma agonist pioglitazone prevents TGF-beta induced renal fibrosis by repressing EGR-1 and STAT3**. *BMC Nephrol.* (2019) **20**. DOI: 10.1186/s12882-019-1431-x
28. Stenvinkel P., Chertow G.M., Devarajan P., Levin A., Andreoli S.P., Bangalore S., Warady B.A.. **Chronic Inflammation in Chronic Kidney Disease Progression: Role of Nrf2**. *Kidney Int. Rep.* (2021) **6** 1775-1787. DOI: 10.1016/j.ekir.2021.04.023
29. Liu B.C., Tang T.T., Lv L.L., Lan H.Y.. **Renal tubule injury: A driving force toward chronic kidney disease**. *Kidney Int.* (2018) **93** 568-579. DOI: 10.1016/j.kint.2017.09.033
30. Disteldorf E.M., Krebs C.F., Paust H.J., Turner J.E., Nouailles G., Tittel A., Meyer-Schwesinger C., Stege G., Brix S., Velden J.. **CXCL5 drives neutrophil recruitment in TH17-mediated GN**. *J. Am. Soc. Nephrol.* (2015) **26** 55-66. DOI: 10.1681/ASN.2013101061
31. Letavernier E., Perez J., Joye E., Bellocq A., Fouqueray B., Haymann J.P., Heudes D., Wahli W., Desvergne B., Baud L.. **Peroxisome proliferator-activated receptor beta/delta exerts a strong protection from ischemic acute renal failure**. *J. Am. Soc. Nephrol.* (2005) **16** 2395-2402. DOI: 10.1681/ASN.2004090802
32. Mukundan L., Odegaard J.I., Morel C.R., Heredia J.E., Mwangi J.W., Ricardo-Gonzalez R.R., Goh Y.P., Eagle A.R., Dunn S.E., Awakuni J.U.. **PPAR-delta senses and orchestrates clearance of apoptotic cells to promote tolerance**. *Nat. Med.* (2009) **15** 1266-1272. DOI: 10.1038/nm.2048
33. Matsushita Y., Ogawa D., Wada J., Yamamoto N., Shikata K., Sato C., Tachibana H., Toyota N., Makino H.. **Activation of peroxisome proliferator-activated receptor delta inhibits streptozotocin-induced diabetic nephropathy through anti-inflammatory mechanisms in mice**. *Diabetes* (2011) **60** 960-968. DOI: 10.2337/db10-1361
34. Romero M., Toral M., Robles-Vera I., Sanchez M., Jimenez R., O’Valle F., Rodriguez-Nogales A., Perez-Vizcaino F., Galvez J., Duarte J.. **Activation of Peroxisome Proliferator Activator Receptor beta/delta Improves Endothelial Dysfunction and Protects Kidney in Murine Lupus**. *Hypertension* (2017) **69** 641-650. DOI: 10.1161/HYPERTENSIONAHA.116.08655
35. Untergasser A., Nijveen H., Rao X., Bisseling T., Geurts R., Leunissen J.A.. **Primer3Plus, an enhanced web interface to Primer3**. *Nucleic. Acids Res.* (2007) **35** W71-W74. DOI: 10.1093/nar/gkm306
|
---
title: The Metagenomic Composition and Effects of Fecal-Microbe-Derived Extracellular
Vesicles on Intestinal Permeability Depend on the Patient’s Disease
authors:
- Cristina Rodríguez-Díaz
- Flores Martín-Reyes
- Bernard Taminiau
- Ailec Ho-Plágaro
- Raquel Camargo
- Felix Fernandez-Garcia
- José Pinazo-Bandera
- Juan Pedro Toro-Ortiz
- Montserrat Gonzalo
- Carlos López-Gómez
- Francisca Rodríguez-Pacheco
- Dámaris Rodríguez de los Ríos
- Georges Daube
- Guillermo Alcain-Martinez
- Eduardo García-Fuentes
journal: International Journal of Molecular Sciences
year: 2023
pmcid: PMC10002483
doi: 10.3390/ijms24054971
license: CC BY 4.0
---
# The Metagenomic Composition and Effects of Fecal-Microbe-Derived Extracellular Vesicles on Intestinal Permeability Depend on the Patient’s Disease
## Abstract
The composition and impact of fecal-microbe-derived extracellular vesicles (EVs) present in different diseases has not been analyzed. We determined the metagenomic profiling of feces and fecal-microbe-derived EVs from healthy subjects and patients with different diseases (diarrhea, morbid obesity and Crohn’s disease (CD)) and the effect of these fecal EVs on the cellular permeability of Caco-2 cells. The control group presented higher proportions of *Pseudomonas and* Rikenellaceae_RC9_gut_group and lower proportions of Phascolarctobacterium, Veillonella and Veillonellaceae_ge in EVs when compared with the feces from which these EVs were isolated. In contrast, there were significant differences in 20 genera between the feces and EV compositions in the disease groups. Bacteroidales and *Pseudomonas were* increased, and Faecalibacterium, Ruminococcus, *Clostridium and* Subdoligranum were decreased in EVs from control patients compared with the other three groups of patients. Tyzzerella, Verrucomicrobiaceae, Candidatus_Paracaedibacter and Akkermansia were increased in EVs from the CD group compared with the morbid obesity and diarrhea groups. Fecal EVs from the morbid obesity, CD and, mainly, diarrhea induced a significant increase in the permeability of Caco-2 cells. In conclusion, the metagenomic composition of fecal-microbe-derived EVs changes depending on the disease of the patients. The modification of the permeability of Caco-2 cells produced by fecal EVs depends on the disease of the patients.
## 1. Introduction
Gram-positive and Gram-negative bacteria release membrane vesicles with sizes ranging from 20 to 400 nm in different abundances, structures, and molecular cargo [1,2]. These microbial extracellular vesicles (EVs) represent a secretion and transport mechanism for carbohydrates, lipids and several cell wall components as well as proteins, DNA, RNA and signaling molecules, among others [3]. Therefore, EVs have been related to cell-to-cell communication, virulence, horizontal gene transfer or phage infection [4,5]. Although the outer membrane vesicles (OMVs) of Gram-negative bacteria were the first found and described, recent work has demonstrated the production of other types of EVs by both Gram-positive and Gram-negative bacteria and even mycobacteria and fungi [4]. The types and origins of these EVs were summarized in a previous review, including OMVs, outer-inner membrane vesicles (OIMVs), cytoplasmic membrane vesicles (CMVs) and tube-shaped membranous structures (TSMSs) [1].
EVs play an essential role in bacterial survival and host interactions due to inter-kingdom signaling and their potential properties in the fecal microbiota–eukaryote interaction [6]. For example, it has been shown that EVs of toxigenic *Bacteroides fragilis* (B. fragilis) contribute to bowel disease and colon cancer [7]. EVs of *Akkermansia muciniphila* (A. muciniphila) have been shown to play a role in controlling intestinal permeability and regulating intestinal barrier integrity, improving metabolic function and ameliorating obesity in mice [8]. A previous study also demonstrated how *Listeria monocytogenes* (L. monocytogenes) produces EVs that carry the majority of listerial virulence proteins, and it uses these EVs for toxin release and mammalian toxicity [9].
Recently, several studies have investigated the secretion of EVs by pure bacterial cultures, while less data are available regarding the secretion of EVs by complex microbial communities or environments. Lagos [10] isolated EVs secreted from fresh pig feces in vitro and observed modifications in their composition and abundance in function under the environmental conditions, especially with respect to carbohydrate availability. Tulkens [2] described the presence of bacterial EVs in human plasma and correlated their abundance with immune activation and barrier integrity in patients with Crohn’s disease (CD), human immunodeficiency viruses (HIVs) and cancer. Further research on the fecal microbiota composition and the derived EVs in feces demonstrated the role of bacterial EVs in the regulation of intestinal immunity and homeostasis and highlighted the protective effect of A. muciniphila EVs in the development of dextran-sulfate-sodium-induced colitis [11]. Recently, it was demonstrated how *Staphylococcus aureus* secretes EVs which can be delivered into macrophage cells, stimulating a potent IFN-β response in recipient cells [12]. In addition to feces, the presence of bacterial EVs has also been studied in human breast milk, suggesting a role in the vertical transfer of the fecal microbiota [13]. Furthermore, milk EVs have been demonstrated to be an important source of mRNA and therefore have important potential as a tool for monitoring the clinical stage of bovine leukemia virus infection [14]. However, less data are available in relation to the EVs present in human feces, their metagenomic profiling, their differences according to different diseases associated with an intestinal dysbiosis and their effects on intestinal permeability.
In this study, we first implemented a procedure to isolate fecal-microbe-derived EVs from human feces. To exclude the presence of free DNA in the EV samples, we also investigated the use of a PMA treatment. Second, we compared the fecal microbiota composition with the composition of the fecal-microbe-derived EVs using 16S ribosomal DNA sequencing in healthy subjects and in patients with diarrhea, morbid obesity and CD. Finally, we tested the effect of these EVs on the cellular permeability of Caco-2 cells in vitro.
## 2.1. PMA Treatment
The performance of the purified EVs with qEVoriginal size exclusion columns and PMA treatment was evaluated prior to sequencing and statistical analysis (Figure 1A). Fecal EV purification with qEV IZON columns normally removes free DNA from the samples; however, the objective of this assay was to verify if the final concentration of the EVs (and therefore the DNA concentration) was enough to perform a good-quality sequencing analysis. Therefore, this test allowed us to determine if the purified EVs from feces had too much free DNA, which might have interfered in sequencing results, and if additional PMA treatment was necessary for these samples.
The PMA treatment was applied as previously described, and sequencing and statistical analysis were performed. The α-diversity metrics, including the number of observed genera, Chao1, the reciprocal Simpson index and Simpson evenness, were used to assess community richness and diversity. Good’s coverage was >0.99 for all samples, indicating that although the number of generated sequence reads (on average, 7000) was limited, this sampling effort allowed for the production of an accurate caption of the fecal-microbe-derived EV communities. No significant differences in bacterial richness, diversity or evenness were observed at genus level, regardless of whether the PMA treatment was used or not (Supplementary Figure S1). Regarding the microbiota composition, the purified fecal-microbe-derived EVs presented a few significant differences at genus level compared with those treated with PMA. Post-hoc pairwise differences between the two groups (with and without PMA treatment) were detected only in genus Alistipes and Acidibacter, which were found to be increased in samples without PMA treatment (Supplementary Figure S2). These results demonstrated that this treatment is not essential for the characterization of human-fecal-microbe-derived EVs when following the protocol implemented in the present study.
## 2.2. Characterization of EVs
After isolation with qEV IZON columns and without PMA treatment, the fecal-microbe-derived EVs showed a typical particle shape and size when analyzed by NTA (Figure 1B) and TEM (Figure 1C). The Western blot analyses revealed the presence a band of bacterial peptidoglycan (Figure 1D).
## 2.3. Microbiome Profile of Fecal-Microbe-Derived EVs and Their Feces of Origin
We further compared the microbial profile of the feces from which the EVs were isolated with the microbial profile of their derived EVs. First, a metagenomics analysis of feces was performed to study the microbial profile. The bacterial EVs were purified using qEVoriginal size exclusion columns, and PMA treatment was not performed. Once the EVs were isolated, a metagenomics analysis was performed to sequence and identify the genera from which these EVs originated. The 16S amplicon sequencing yielded 10,000 cleaned reads per sample from which taxonomic identification was obtained. No significant differences were found in bacterial richness (Chao1 richness index), alpha diversity (inverse Simpson index) and evenness (derived from Simpson index) between the feces and fecal-microbe-derived EVs when all samples were compared together (Figure 2).
When the same analysis was performed by group of patients (feces vs. fecal-microbe-derived EVs from the control group and feces vs. fecal-microbe-derived EVs from a group of patients with disease (CD, diarrhea or morbid obesity)), no significant differences were observed between the fecal microbiota and the fecal-microbe-derived EVs after multiple comparisons using the Kruskal–Wallis test with Benjamini–Hochberg FDR corrections (Supplementary Figure S3). AMOVA and HOMOVA analyses showed that the genetic diversity in the fecal-microbe-derived EVs was significantly different from that from fecal bacteria ($$p \leq 0.037$$); however, the amount or variation of this genetic diversity in each group (fecal bacteria and EVs) was not significantly different ($p \leq 0.05$). Finally, the NMDS and dbRDA analyses are shown in Figure 3A and Figure 3B, respectively.
## 2.4. Global Comparison between Feces and Fecal-Microbe-Derived EVs
Twenty-one genera presented a relative abundance greater than $1\%$ in both types of samples: fecal bacteria and EVs (Figure 4A(i)).
Seven dominant genera with a relative abundance >$3\%$ in both groups were identified: namely, Bacteroides, Faecalibacterium, Prevotella_9, Romboutsia, Escherichia-Shigella, *Streptococcus and* Laschnospiraceae_ge. Significant differences were observed in 18 different genera (Figure 4A(ii)). Among them, only two taxa, identified as Oscillibacter and Saccharimonadaceae_ge, were increased in the EV samples, while the remaining 16 genera were significantly increased in the feces samples
## 2.5. Comparison between Feces and Fecal-Microbe-Derived EVs in the Control Group
The comparison between fecal bacteria and EVs obtained only from the control group revealed the presence of 17 genera with a relative abundance greater than $1\%$ and 6 dominant genera with a relative abundance of >$3\%$ in fecal bacteria and the EV samples, which were identified as Bacteroides, Prevotella_9, Prevotellaceae_NK3B31_group, Dialister, Alistipes and Parabacteroides (Figure 4B(i)). Few significant differences were observed between the composition of the fecal bacteria and EVs with only five genera implicated, including Phascolarctobacterium, Rikenellaceae_RC9_gut_group, Pseudomonas, Veillonella and Veillonellaceae_ge. Almost all of these genera were increased in the fecal bacteria and reduced in the EVs with the exception of Rikenellaceae_RC9_gut_group and Pseudomonas, which were found in higher proportions in the EVs (Figure 4B(ii)). The relative abundance of bacterial genera in the fecal microbiota and fecal-microbe-derived EVs for each patient is shown in Supplementary Figure S4.
## 2.6. Comparison between Feces and Fecal-Microbe-Derived EVs in the Group of Patients with Different Diseases
Finally, we compared the composition of the microbiota from fecal bacteria and EVs in the groups of patients with different diseases (morbid obesity, CD and diarrhea groups together). In the fecal microbiota samples, eleven taxa were identified as dominant, with a relative abundance greater than $3\%$ (Figure 4C(i)), while only five genera were observed in these proportions in the EVs (Faecalibacterium, Prevotella_9, Romboutsia, Bacteroides and Parabacteroides). Several significant differences between the fecal composition and the composition of the EVs were detected, with 20 different genera implicated (Figure 4C(ii)). Only Saccharimonadaceae_ge was found to be increased in EVs, while the remaining 19 genera were all increased in the fecal bacterial samples. The relative abundance of bacterial genera in the fecal microbiota and fecal-microbe-derived EVs per type of disease is shown in Figure 4D (Crohn’s disease), 4E (morbid obesity) and 4F (diarrhea group). The relative abundance of bacterial genera in the fecal microbiota and fecal-microbe-derived EVs for each patient is shown in Supplementary Figure S4.
## 2.7. Comparison between Fecal-Microbe-Derived EVs from Different Diseases
Table 1 shows the significant differences found in the composition of EVs between the four groups of patients. A group of genera was increased (Bacteroidales and Pseudomonas) or decreased (Faecalibacterium, Ruminococcus, *Clostridium and* Subdoligranum) in EVs from control patients with respect to the rest of the groups. *Other* genera were also found to be decreased in the control patients with respect to most of the groups (Table 1). There were also some genera that were exclusively increased or decreased, depending on the type of disease, when they were compared to the control group (marked with * in Table 1). Moreover, our findings showed that Tyzzerella, Verrucomicrobiaceae, Candidatus_Paracaedibacter and Akkermansia were increased in EVs from the CD group compared to the morbid obesity and diarrhea groups. In addition, Parabacteroides was increased in EVs from the morbid obesity group compared to the CD and diarrhea groups. No other significant differences were found.
## 2.8. Intestinal Permeability in Caco-2 Cells
We first tested whether fecal EVs induced an alteration of the intestinal permeability of Caco-2 cells by measuring TEER and FD4. Fecal EVs from the different groups of patients were used. We found that Caco-2 cells incubated with fecal EVs from the control patients presented an increase of 29.1 ± $4.0\%$ in the TEER value (Figure 5A).
However, Caco-2 cells incubated with fecal EVs from patients with morbid obesity, CD and diarrhea presented an increase of 12.1 ± $3.52\%$, 9.9 ± $1.7\%$ and 4.4 ± $1.6\%$, respectively, in TEER values. The change produced by EVs from patients with diarrhea was significantly lower than the change produced by fecal EVs from the control patients ($$p \leq 0.0$$ 45). Next, we measured paracellular permeability by monitoring the flux of FD4 through the Transwell. As shown in Figure 5B,C, the fecal EVs from control group did not exert a significant effect on the permeability. Fecal EVs from patients with morbid obesity and CD induced a slightly significant increase in the permeability of Caco-2 cells at 30 min. However, the fecal EVs from patients with diarrhea induced the highest increase at each time in the translocation of FD4 to the basolateral compartment when compared to the control group. Moreover, the increase found with the diarrhea fecal EVs was also significantly higher than with the fecal EVs from patients with morbid obesity. Therefore, the permeability of the Caco-2 cells was modified by fecal EVs according to their origin.
## 3. Discussion
Sequencing methods do not discriminate between live (dormant cells and non-growing or growing cells, which are metabolically active) or dead bacteria. In our study, the analysis of the PMA-treated EVs did not present enough differences in richness, alpha diversity or Good’s coverage to be statistically different from those that were not treated with PMA. This method is recognized as a valuable tool for the distinction of dead/viable cells since PMA treatment is a DNA-intercalating agent that acts on free DNA and penetrates cells with compromised membranes [14]. Regarding the composition of EVs, significant differences were only found for two genera, indicating that most populations can be found in the same proportions in EVs treated and not treated with PMA. Therefore, this treatment is not essential for the metagenomics analysis of fecal-microbe-derived EVs from human feces when the protocol implemented in the present study is followed. The use of qEV original size exclusion columns for the purification of fecal EVs appears to be sufficiently efficient to remove free DNA from the fecal EVs.
Microbe-derived EVs have been directly associated with disease development [15]. However, there are few studies on the composition of fecal-microbe-derived EVs compared with their feces of origin, and a large proportion of these studies focused on colorectal cancer and inflammatory bowel disease (IBD) patients [16,17,18]. In our study, the overall bacterial richness and diversity were not significantly different between the two types of samples studied (fecal bacteria and fecal-microbe-derived EVs) within the control group or within patients with disease. However, differences were detected in the microbial composition of EVs in relation to the fecal microbiota. *In* general, higher proportions of various genera were found in the fecal microbiota compared with the EVs. Previous studies have also demonstrated that the protein composition of fecal EVs differs from that of fecal samples in IBD patients [16]. These findings may suggest that the different bacterial genera secrete variable proportions of EVs which, in turn, may be influenced by the patient’s intestinal disease. These fecal-microbe-derived EVs may be used as novel biomarkers to detect various intestinal diseases, as proposed by Park for colorectal cancer [15]. We also observed more differences in the microbiota structure between the fecal bacteria and EVs in patients with disease than in the control group. However, the main limitation of this study is the low number of recruited patients in each group, which did not allow us to describe the fecal-microbiota-derived EVs that are candidates for predicting each disease. Nevertheless, our results provide preliminary data to further study how the composition of these fecal-microbe-derived EVs is modified in different diseases and to analyze whether these EVs may be involved in the microbiota–host interaction. Previously, significant compositional differences were demonstrated in obese and diabetic rats compared to normal rats in terms of the composition of microbial EVs [19]. Another study also showed that the composition of intestinal EVs was greatly altered after vertical sleeve gastrectomy in mice [20]. As bacteria proliferate, the secretion of EVs should increase in line with the increase in the relative abundance of taxa [10]. However, it is possible that bacteria, depending on the group to which they belong, are capable of producing a greater or lesser number of EVs. Furthermore, this production may be influenced by the presence of other bacterial communities and by the physiological conditions of the environment. This could be a hypothesis to explain the increase in EVs in certain bacterial groups with respect to their percentage in fecal microbiota.
In this study, the control patients presented high proportions of *Pseudomonas and* Rikenellaceae_RC9_gut_group in fecal-microbe-derived EVs but lower proportions of Phascolarctobacterium, Veillonella and Veillonellaceae_ge when compared with the fecal bacterial samples. The Pseudomonas genus, specifically Pseudomonas fragi, also commonly produced important levels of EVs during growth. These vesicles display considerable proteolytic activity but are not associated with bacteriocinogenicity. They most likely act in the physiological distribution of extracellular proteinases [21]. Regarding Rikenellaceae_RC9_gut_group, only one previous study described an increase in their EVs in patients with colorectal cancer [15]. In addition, a high proportion was found after fecal microbiota transplantation upon Salmonella Enteritidis infection in chicks [22]. Moreover, the supplementation with probiotics in broilers had a promoting effect on the growth performance and increased the colonization of beneficial bacteria in the cecum as Rikenellaceae_RC9_gut_group [23]. It is involved in degrading carbohydrates [24] and metabolizes lipids [25]. In addition, members of the Veillonellaceae family are often found in association with gut inflammation [26] and are more abundant in patients with IBD, fibrosis and other diseases [27,28]. Taken together, these data seem to suggest that the EVs derived from certain bacteria might have an important role in the maintenance of intestinal homeostasis.
In our disease patients, we observed differences between the composition of the fecal bacteria and EVs in a total of 20 genera, all of which showed decreased proportions in EVs except Saccharimonadaceae_ge, which was increased in the EVs. In the literature, there is no specific information about the presence of Saccharimonadacea-derived EVs in the feces of patient; therefore, its role in the gut requires further investigation. EVs belonging to other genera that were found to be decreased in our study have been previously studied due to their possible effects on intestinal diseases. An example of this includes Akkermansia (A. muciniphila)-derived EVs, which have been reported to act as a functional moiety for controlling gut permeability and regulating the intestinal barrier integrity in mice [8]. Other important bacterial groups that showed significant differences between the composition of the fecal microbiota and EVs are Lactobacillus and Bifidobacterium. The EVs of *Lactobacillus plantarum* Q7 have been demonstrated to alleviate induced colitis symptoms and histological damage in mice. They also reduced the levels of proinflammatory bacteria (Proteobacteria) and increased the levels of anti-inflammatory groups (Bifidobacterium and Muribaculaceae) [29]. The EVs of *Bifidobacterium longum* can export several cytoplasmic proteins that could be involved in bifidobacterial adhesion and survival in the gastrointestinal tract [30].
Our in vitro experiment demonstrated that fecal EVs act as regulators of epithelial barrier integrity with differences depending on the disease of the patients. This is in accordance with the different compositions of fecal-microbe-derived EVs that are dependent on the type of disease. In contrast to most studies, our findings describe the effects of EVs from a mixture of fecal bacteria, not from a specific bacterium. Several studies with different species of fecal microbiota have demonstrated the role of bacterial-derived EVs as modulators of epithelial barrier integrity [31]. In this context, the fecal microbiota-derived EVs, besides the host-derived EVs [31], could be involved in the regulation of gut homeostasis by enhancing the intestinal permeability, a condition that subsequently leads to inflammatory and metabolic diseases [32]. This increased gut permeability would allow for the passage of endotoxins and luminal antigens into the intestinal lamina propria, initiating a mucosal immune response that causes chronic, low-grade inflammation, prompting metabolic disorders such as insulin resistance and obesity [31]. Possible differences in the surface cargo molecules, such as microbe-associated molecular patterns (MAMPs), could be mediating the adhesion of these fecal-microbiota-derived EVs to host epithelial cells and, consequently, the downstream effects [33]. Moreover, in a later study, it would be interesting to analyze the metabolic and transcriptomic changes produced by these fecal EVs in different types of cells.
A limitation of this study was that we did not characterize the total composition of these EVs, i.e., we did not analyze their protein, RNA, DNA and lipid contents. These factors could be associated with the effects produced by these EVs. In this study, we only focused on analyzing the genera from which the EVs originated by metagenomic analysis. In addition, although the method used in this study to isolate the EVs has been previously described and used [34,35,36], it is possible that it could be improved by performing EV isolation prior to freezing in order to minimize the presence of intracellular artifacts/contaminants from microorganisms/cells derived from the freezing process. This point will require further study to analyze the differences between these two methodologies.
In summary, we conducted a metagenomic study to reveal associations between the fecal microbiota and the microbial composition of EVs in control subjects and in patients with disease. We found that fecal-microbiota-derived EVs from control subjects have a metagenomic profile closely similar to that of the fecal microbiota. However, we have shown that the presence of a dysbiotic fecal microbiota in different diseases is accompanied by an altered composition of fecal-microbe-derived EVs. Therefore, our findings demonstrate that diseases such as diarrhea, CD or morbid obesity alter the microbial composition of EVs in relation to the fecal microbiota. On the other hand, we found an increase in intestinal permeability with fecal EVs from patients with different diseases. We suggest that the fecal-microbiota-derived EVs from certain bacteria might cause increased intestinal permeability as part of their infectious mechanisms, while other bacterial strains attenuate inflammation and reinforce the gut barrier integrity [37]. We postulated the importance of controlling the balance between the different subsets of fecal microbiota and their EVs in the development of diseases associated with altered intestinal permeability. However, the cause-and-effect relationships and the role of these fecal-microbiota-derived EVs, as mediators of interspecies interactions and as novel biomarkers, in the course of a disease require future careful, experimental studies.
## 4.1. Patient Recruitment
Our cohort study included 32 patients: 9 healthy volunteers, 10 diarrheic patients, 9 patients with morbid obesity and 4 patients with CD. These diseases were chosen because they demonstrates a clear alteration of fecal microbiota [38,39,40]. In those patients with diarrhea, neither parasites nor Cryptosporidium were isolated in the feces, they had normal flora, no Salmonella, Shigella, Campylobacter, *Yersinia and* Aeromonas were isolated, and the presence of *Clostridium difficille* toxin and adenovirus and rotavirus antigens was negative. Fecal samples were collected from all patients ($$n = 32$$) and immediately stored at −80 °C in the Virgen de la Victoria University Hospital Biobank (Andalusian Public Health System BioBank) until analysis. All participants were of Caucasian origin. All participants gave their written informed consent, and the study protocol was carried out in accordance with the ethical guidelines of the Declaration of Helsinki. The study was approved by the Malaga Provincial Research Ethics Committee, Malaga, Spain (PI$\frac{18}{01652}$, PE-0098-2019).
## 4.2. Isolation of EVs from Human Feces
A total of 10 g of feces was inoculated into 40 mL of sterile, phosphate-buffered saline (PBS) and homogenized. The EVs were then isolated through centrifugation as previously described with some modifications [10]. Briefly, a first centrifugation of the homogenate was performed (40 min, 4000× g, and 4 °C. The supernatant was recovered and filtered using sterilized vacuum filtration units, Rapid-Flow™ filters MF 75, 1000 mL of Nalgene® and 0.2 μm of cold ice (Thermo Fisher Scientific, Waltham, MA, USA). The filtrate was transferred to 10 mL polycarbonate, open-top, thick-wall tubes and ultracentrifuged at 100,000× g for 3 h at 4 °C with a fixed-angle rotor (Type 70.1 Ti) in a Beckman Optima XL-100K ultracentrifuge (Beckman Coulter Life Sciences, Indianapolis, IN, USA). Pellets were resuspended in 200 µL of PBS and the EVs were purified using qEVoriginal size exclusion columns of 70 nnm (Izon Science Europe Ltd., Oxford, UK), following the manufacturer’s recommendations. Fractions 6–8 (enriched in EVs) were collected, mixed, concentrated with Vivaspin® 6 100K centrifugal concentrators (Sartorius AG, Göttingen, Germany), aliquoted and frozen at −80 °C until use. This protocol was used to separate bacteria and other contaminating soluble molecules, such as toxins and proteins, from the EVs. These aliquots of fecal EVs were used for treatment with propidium monoazide, metagenomic analysis, transmission electron microscopy, nanoparticle tracking analysis, Western blot and for the incubation of Caco-2 cells.
## 4.3. Transmission Electron Microscopy (TEM) of EVs
The isolated EVs ($$n = 4$$; one from a healthy control, one from a CD patient, one from a diarrheic patient and one from a patient with morbid obesity) were fixed in $2\%$ paraformaldehyde—0.1 M PBS for 30 min. A glow discharge technique (60 s, 7.2 V, using a Bal-Tec MED 020 Coating System) was applied over carbon-coated copper grids, and these grids were immediately placed on top of sample drops for 15 min. Then, the grids with adherent EVs were washed in a 0.1 M PBS drop. Additional fixation in $1\%$ glutaraldehyde was performed for 5 min. After washing the grids properly in distilled water, the grids were contrasted with $1\%$ uranyl acetate and embedded in methylcellulose. Excess fluid was removed and allowed to dry before examination with a transmission electron microscope FEI Tecnai G2 Spirit (ThermoFisher Scientific, Waltham, MA, USA). All images were acquired using a Morada digital camera (Olympus Soft Image Solutions GmbH, Münster, Germany). The magnification used for the TEM images was 49,000×.
## 4.4. Nanoparticle Tracking Analysis (NTA)
The EV size and concentration were assessed using the NanoSight NS300 system (Malvern Panalytical, Malvern, UK) ($$n = 3$$; one from a morbidly obese patient, one from a diarrheic patient and one from a healthy control). Particles were automatically tracked and sized-based on Brownian motion and the diffusion coefficient. The EVs were resuspended and diluted with 0.22 μm filtered PBS at a concentration range 109 particles/mL, and 1 mL was used for NanoSight analysis. Five replicates of 30 s videos were captured to analyze the concentration and size distribution of the EVs at the detection threshold of 5. A data analysis was performed using NanoSight analysis software.
## 4.5. Western Blot of EVs
Fecal EVs ($$n = 3$$; one from a CD patient, one from a morbidly obese patient and one from a healthy control) were lysed with 1× RIPA buffer (Thermo Fisher (Kandel) GmbH, Kandel, Germany) and supplemented with a protease inhibitor cocktail (Merck KGaA, Darmstadt, Germany). The protein lysate was incubated with the same volume of Laemmli Buffer 2× (Bio-Rad Laboratories, Inc., Hercules, CA, USA) and supplemented with 2-mercaptoethanol ($5\%$) at 95 °C for 5 min. The samples were subjected to 4–$20\%$ SDS-PAGE (NB12-420) (NuSep, Inc., Germantown, MD, USA) and transferred onto polyvinylidene fluoride membranes (Trans-Blot Turbo Midi 0.2 µm PVDF Transfer Packs) (Bio-Rad Laboratories, Inc., Hercules, CA, USA) at 13 V and 1.1 A for 20 min. The membranes were subsequently blocked in PBS–bovine serum albumin (BSA) $5\%$ for 1 h at room temperature. The membranes were then incubated for 48 h at 4 °C with a mouse monoclonal anti-bacterial peptidoglycan antibody, clone 3F6B3 (Merck KGaA, Darmstadt, Germany). This antibody is specific to the three-dimensional polymer complex structure of bacterial peptidoglycan. The membranes were washed three times with $0.05\%$ Tween-20 washing buffer in PBS and incubated with a horseradish-peroxidase-conjugated secondary antibody (VeriBlot for IP Detection Reagent (HRP), ab131366) (Abcam, Cambridge, UK) for 3 h at room temperature. Finally, after another three washes, the membranes were revealed with Clarity Western ECL substrate (Bio-Rad Laboratories, Inc., Hercules, CA, USA). The proteins were visualized by an ImageQuant LAS 4000 (GE Healthcare, Buckinghamshire, UK).
## 4.6. Propidium Monoazide (PMA) Treatment
The procedure used to isolate the microbe-derived EVs could result in the presence of a small percentage of free DNA in the sample. Therefore, we tested sample treatment with PMA [41] in order to detect the co-extraction and amplification of the nonprotected DNA of the membrane-compromised EVs. Furthermore, we wanted to evaluate the optimal separation of EVs from free DNA using qEVoriginal size exclusion columns of 70 nnm (Izon Science Europe Ltd., Oxford, UK). For this assay, seven samples from patients were evaluated (two from healthy volunteers, two from diarrheic patients, two from morbidly obese patients and 1 from a patient with CD). In total, 100 µL of each sample was centrifuged at 5000× g in duplicate from which one was left untreated and the other one was treated with PMA (PMAxx™ dye) (Biotium, Fremont, CA, USA) prior to DNA extraction (Figure 1A). The manufacturer’s protocol for PMA treatment was used and involved the use of the PMA-Lite™ LED Photolysis Device (Biotium, Fremont, CA, USA). The statistical analyses were performed with the seven samples tested.
## 4.7. DNA Extraction
The total DNA was extracted from the EVs that were treated and not treated with PMA (Izon Science Europe Ltd., Oxford, UK) and directly from the fecal samples using DNeasy blood and Tissue Kits (QIAGEN Science, Hilden, Germany), following the manufacturer’s recommendations. Briefly, after the isolation and purification of the EVs from feces, DNA extraction was performed. Once this DNA was obtained, seven of these DNA samples from the EVs were aliquoted in duplicate to treat one half with PMA. In parallel, a DNA extraction was also performed in all fecal samples used for the isolation of EVs. The DNA was eluted into DNase/RNase-free water and its concentration and purity were evaluated using a NanoDrop ND-1000 spectrophotometer (NanoDrop Technologies, Inc., Wilmington, DE, USA). Extracts were stored at −20 °C until use.
## 4.8. Libraries Preparation and Sequencing
Libraries and sequencing were performed as previously described [42]. Briefly, amplification of the V1-V3 regions of the 16S rRNA bacterial gene was performed using the primers 5′-GAGAGTTTGATYMTGGCTCAG-3′ forward and 5′-ACCGCGGCTGCTGGCAC-3′ reverse with overhand adapters. Amplicons were purified using Agencourt AMPure XP bead kit (Beckman Coulter, Pasadena, CA, USA), indexed using Nextera XT index primers 1 and 2 (Illumina, San Diego, CA, USA), quantified by Quant-IT PicoGreen (Thermo Fisher Scientific, Waltham, MA, USA) and diluted to a concentration of 10 ng/μL. DNA samples were quantified by qPCR with a KAPA 170 SYBR®® FAST qPCR Kit (Kapa Biosystems, Wilmington, MA, USA). Samples were normalized, pooled and sequenced using Illumina MiSeq technology with v3 reagents (Illumina, San Diego, CA, USA), using paired end reads by GIGA Genomics platform (Liège, Belgium). A bacterial community composed of known proportions of Carnobacterium maltaromaticum, *Lactococcus lactis* subsp. cremoris, Leuconostoc carnosum, *Pseudomonas aeruginosa* and *Streptococccus thermophilus* was used as a positive control. Negative controls were used in their entirety for DNA extraction, library preparation and sequencing.
## 4.9. Bioinformatics, Ordination and Statistical Analysis
Sequence reads were processed using Mothur v1.44.3 and VSearch for alignment, clustering and chimera detection, respectively [43,44]. The sequences were clustered into operational taxonomic units (OTUs) at an identity of $97\%$. The SILVA 138 database of full-length 16S rDNA gene sequences was used for the alignments of unique sequences and taxonomical assignations. For each sample, a subsampling dataset containing 10,000 representative, cleaned reads was retained (mean: 10,000, SD: 0) and used to generate OTUS (cut off: 0.03) as well as to evaluate several ecological indicators.
All statistical analyses were performed at the genus level. Regarding alpha diversity (reciprocal Simpson diversity index and Simpson evenness), Goods’s coverage and population richness (Chao1 estimator of richness) were calculated using Mothur v1.44.3 and compared between two groups using a Wilcoxon matched-pairs signed rank test (PRISM 8) (GraphPad Software, Boston, MA, USA) or between three or more groups using Kruskal–Wallis multiple testing with Benjamini–Hochberg FDR corrections (PRISM 8) (GraphPad Software, Boston, MA, USA). Bar plots were built using PRISM 8, including only genera with a relative abundance >$1\%$. The β-diversity was estimated with the Bray–Curtis dissimilarity index using Mothur (v1.44.3) and R for graphical analysis (v1.2.5033). Non-metric multidimensional scaling (NMDS) was performed using Mothur and was considered satisfying when the stress value was <0.20. An AMOVA (analysis of molecular variance) and a HOMOVA (homogeneity of molecular variance) were performed using Mothur in order to reveal eventual significant population structure differences and to determine if the genetic diversity within two or more populations was homogeneous [44]. A distance-based redundancy analysis (dbRDA) was constructed using RStudio. Post-hoc pairwise differences between groups were assessed with Deseq2 package in R, and differences were then identified with Kruskal–Wallis tests using Benjamini–Hochberg FDR correction [45].
## 4.10. In Vitro Cell Culture
Caco-2 (ECACC, Cat. No. 09042001) epithelial cell lines were maintained in complete medium (Dulbecco’s modified Eagle’s Medium (DMEM) of high glucose with L-glutamine (Biowest, Nuaillé, France) supplemented with $10\%$ heat-inactivated fetal bovine serum (FBS) (Biowest, Nuaillé, France), $1\%$ penicillin/streptomycin (Biowest, Nuaillé, France) and $1\%$ MEM non-essential amino acids (Sigma-Aldrich, St. Louis, MO, USA) under standard conditions inside a humidified cell culture incubator at 37 °C with $5\%$ CO2. Caco-2 cells were harvested by washing three times in sterile DPBS, followed by treatment with trypsin-EDTA. Harvested cells were counted and seeded in 12-well PET Transwell™ inserts of 0.4 μm pore size (Corning Inc., Corning, MA, USA) at 105 cells/insert by adding 0.5 mL of cell suspension. The apical and basal cell culture media, 0.5 mL and 1.5 mL respectively, were changed every two days. Cells were maintained for approximately 3 weeks in the same medium to allow for full cell differentiation. The culture medium was changed, and 1 μg of protein from the purified EVs suspension was added for 24 h of incubation [8]. The protein concentration of the purified fecal EV suspension was determined using the bicinchoninic acid (BCA) assay (Thermo Fisher Scientific, Waltham, MA, USA). After 24 h of incubation with fecal EVs ($$n = 4$$ for each group of patients), the trans-epithelial electrical resistance (TEER) and para-cellular permeability were measured to analyze the EV-induced changes.
## 4.11. Trans-Epithelial Electrical Resistance
TEER was measured using a Millicell®® ERS-2 Voltohmmeter (Merck Millipore, Burlington, MA, USA). Once Caco-2 cells reached a TEER > 1000 (Ω·cm2), experiments with the purified fecal EV suspension were performed as described above. TEER values were obtained by subtracting cell-free filter readings and correcting for the surface area (1.1 cm2). All readings of TEER were repeated across triplicate sample Transwells. TEER values were expressed as the percentage of change with respect to the TEER value obtained prior to the incubation with purified fecal EV suspension. Data were presented as means ± SEM ($$n = 4$$).
## 4.12. Paracellular Permeability
The Caco-2 monolayer paracellular permeability was assessed by measuring the unidirectional flux of fluorescein isothiocyanate (FITC)-dextran (FD4; 4000 Da, Sigma-Aldrich, Saint-Louis, MO, USA) from the apical to the basolateral compartments of the Transwell™. The complete DMEM medium was removed from the apical and basolateral compartments, replaced with Krebs Ringer Bicarbonate Buffer Hepes Albumin (KRBHA), and equilibrated for 1 h at pH 7.4. The KRBHA medium was replaced again, and 25 mg/mL stock solution of FD4 was added to the apical compartment at time zero to obtain a final concentration of 1 mg/mL. An aliquot of 100 μL from the basolateral compartment was removed every 30 min over 2 h, followed by replacement with fresh KRBHA. Samples were transferred onto Nunclon®® MicroWell plates (Thermo Scientific, MA, USA) and the fluorescence of FD4 was measured in a microplate fluorescence reader (FLx 800, Bio-tek Instruments Inc., Winooski, VT, USA) with an excitation of $\frac{485}{20}$ nm and an emission of $\frac{528}{20}$ nm. A negative control was performed with the Caco-2 cells without fecal EV treatment. A positive control was performed with the Caco-2 cells and 5 mM EGTA instead of fecal EVs. EGTA causes a breakdown of the tight junctions by sequestering bivalent ions independently of inflammatory stimuli [46]. Based on the relative fluorescence units, FD4 concentrations were expressed as the percentage of change from Caco-2 cells without fecal EV treatment. All the results were analyzed in triplicate. Data were presented as means ± SEM ($$n = 4$$).
## 4.13. Statistical Analysis
All data were analyzed with GraphPad Software (Prism 8.1.1) (GraphPad Software, San Diego, CA, USA). Differences between groups were compared using Kruskal–Wallis tests followed by post hoc analyses using Dunn’s test. Values were considered to be statistically significant when $p \leq 0.05.$
## References
1. Toyofuku M., Nomura N., Eberl L.. **Types and origins of bacterial membrane vesicles**. *Nat. Rev. Microbiol.* (2019) **17** 13-24. DOI: 10.1038/s41579-018-0112-2
2. Tulkens J., De Wever O., Hendrix A.. **Analyzing bacterial extracellular vesicles in human body fluids by orthogonal biophysical separation and biochemical characterization**. *Nat. Protoc.* (2020) **15** 40-67. DOI: 10.1038/s41596-019-0236-5
3. Tsatsaronis J.A., Franch-Arroyo S., Resch U., Charpentier E.. **Extracellular Vesicle RNA: A Universal Mediator of Microbial Communication?**. *Trends Microbiol.* (2018) **26** 401-410. DOI: 10.1016/j.tim.2018.02.009
4. Brown L., Wolf J.M., Prados-Rosales R., Casadevall A.. **Through the wall: Extracellular vesicles in Gram-positive bacteria, mycobacteria and fungi**. *Nat. Rev. Microbiol.* (2015) **13** 620-630. DOI: 10.1038/nrmicro3480
5. Schwechheimer C., Kuehn M.J.. **Outer-membrane vesicles from Gram-negative bacteria: Biogenesis and functions**. *Nat. Rev. Microbiol.* (2015) **13** 605-619. DOI: 10.1038/nrmicro3525
6. Ahmadi Badi S., Moshiri A., Fateh A., Rahimi Jamnani F., Sarshar M., Vaziri F., Siadat S.D.. **Microbiota-Derived Extracellular Vesicles as New Systemic Regulators**. *Front. Microbiol.* (2017) **8** 1610. DOI: 10.3389/fmicb.2017.01610
7. Zakharzhevskaya N.B., Vanyushkina A.A., Altukhov I.A., Shavarda A.L., Butenko I.O., Rakitina D.V., Nikitina A.S., Manolov A.I., Egorova A.N., Kulikov E.E.. **Outer membrane vesicles secreted by pathogenic and nonpathogenic Bacteroides fragilis represent different metabolic activities**. *Sci. Rep.* (2017) **7** 5008. DOI: 10.1038/s41598-017-05264-6
8. Ashrafian F., Shahriary A., Behrouzi A., Moradi H.R., Keshavarz Azizi Raftar S., Lari A., Hadifar S., Yaghoubfar R., Ahmadi Badi S., Khatami S.. **Akkermansia muciniphila-Derived Extracellular Vesicles as a Mucosal Delivery Vector for Amelioration of Obesity in Mice**. *Front. Microbiol.* (2019) **10** 2155. DOI: 10.3389/fmicb.2019.02155
9. Coelho C., Brown L., Maryam M., Vij R., Smith D.F.Q., Burnet M.C., Kyle J.E., Heyman H.M., Ramirez J., Prados-Rosales R.. **Listeria monocytogenes virulence factors, including listeriolysin O, are secreted in biologically active extracellular vesicles**. *J. Biol. Chem.* (2019) **294** 1202-1217. DOI: 10.1074/jbc.RA118.006472
10. Lagos L., Leanti La Rosa S., Arntzen M.Ø., Ånestad R., Terrapon N., Gaby J.C., Westereng B.. **Isolation and Characterization of Extracellular Vesicles Secreted In Vitro by Porcine Microbiota**. *Microorganisms* (2020) **8**. DOI: 10.3390/microorganisms8070983
11. Kang C.S., Ban M., Choi E.J., Moon H.G., Jeon J.S., Kim D.K., Park S.K., Jeon S.G., Roh T.Y., Myung S.J.. **Extracellular vesicles derived from gut microbiota, especially Akkermansia muciniphila, protect the progression of dextran sulfate sodium-induced colitis**. *PLoS ONE* (2013) **8**. DOI: 10.1371/journal.pone.0076520
12. Rodriguez B.V., Kuehn M.J.. **Staphylococcus aureus secretes immunomodulatory RNA and DNA via membrane vesicles**. *Sci. Rep.* (2020) **10** 18293. DOI: 10.1038/s41598-020-75108-3
13. Kim S.Y., Yi D.Y.. **Analysis of the human breast milk microbiome and bacterial extracellular vesicles in healthy mothers**. *Exp. Mol. Med.* (2020) **52** 1288-1297. DOI: 10.1038/s12276-020-0470-5
14. Fittipaldi M., Nocker A., Codony F.. **Progress in understanding preferential detection of live cells using viability dyes in combination with DNA amplification**. *J. Microbiol. Methods* (2012) **91** 276-289. DOI: 10.1016/j.mimet.2012.08.007
15. Park J., Kim N.E., Yoon H., Shin C.M., Kim N., Lee D.H., Park J.Y., Choi C.H., Kim J.G., Kim Y.K.. **Fecal Microbiota and Gut Microbe-Derived Extracellular Vesicles in Colorectal Cancer**. *Front. Oncol.* (2021) **11** 650026. DOI: 10.3389/fonc.2021.650026
16. Zhang X., Deeke S.A., Ning Z., Starr A.E., Butcher J., Li J., Mayne J., Cheng K., Liao B., Li L.. **Metaproteomics reveals associations between microbiome and intestinal extracellular vesicle proteins in pediatric inflammatory bowel disease**. *Nat. Commun.* (2018) **9** 2873. DOI: 10.1038/s41467-018-05357-4
17. Alberti G., Mazzola M., Gagliardo C., Pitruzzella A., Fucarini A., Giammanco M., Tomasello G., Carini F.. **Extracellular vesicles derived from gut microbiota in inflammatory bowel disease and colorectal cancer**. *Biomed. Pap. Med. Fac. Univ. Palacky Olomouc Czech Repub.* (2021) **165** 233-240. DOI: 10.5507/bp.2021.042
18. Kim D.J., Yang J., Seo H., Lee W.H., Ho Lee D., Kym S., Park Y.S., Kim J.G., Jang I.J., Kim Y.K.. **Colorectal cancer diagnostic model utilizing metagenomic and metabolomic data of stool microbial extracellular vesicles**. *Sci. Rep.* (2020) **10** 2860. DOI: 10.1038/s41598-020-59529-8
19. Huh Y.J., Seo J.Y., Nam J., Yang J., McDowell A., Kim Y.K., Lee J.H.. **Bariatric/Metabolic Surgery Induces Noticeable Changes of Microbiota and Their Secreting Extracellular Vesicle Composition in the Gut**. *Obes. Surg.* (2019) **29** 2470-2484. DOI: 10.1007/s11695-019-03852-1
20. Myronovych A., Peck B.C.E., An M., Zhu J., Warm A., Kupe A., Lubman D.M., Seeley R.J.. **Intestinal extracellular vesicles are altered by vertical sleeve gastrectomy**. *Am. J. Physiol. Gastrointest. Liver Physiol.* (2021) **320** G153-G165. DOI: 10.1152/ajpgi.00224.2020
21. Myhara R.M., Skura B.J., Patel T., Nakajima M.. **Functional Characteristics of Extracellular Vesicles Produced by Pseudomonas fragi ATCC 4973**. *J. Food Prot.* (1990) **53** 859-868. DOI: 10.4315/0362-028X-53.10.859
22. Wang X., Wu X., Cong X., Ren J., Li J., Zhu J., Dai M., Hrabchenko N., Du Y., Qi J.. **The functional role of fecal microbiota transplantation on Salmonella Enteritidis infection in chicks**. *Vet. Microbiol.* (2022) **269** 109449. DOI: 10.1016/j.vetmic.2022.109449
23. Wu Y., Yang F., Jiang W., Hu A., Xiong Z., Yang S., Cao P., Cao Z., Xiong Z., Cao H.. **Effects of compound probiotics on intestinal barrier function and caecum microbiota composition of broilers**. *Avian Pathol.* (2022) **51** 465-475. DOI: 10.1080/03079457.2022.2100740
24. Rowland I., Gibson G., Heinken A., Scott K., Swann J., Thiele I., Tuohy K.. **Gut microbiota functions: Metabolism of nutrients and other food components**. *Eur. J. Nutr.* (2018) **57** 1-24. DOI: 10.1007/s00394-017-1445-8
25. Zhou L., Xiao X., Zhang Q., Zheng J., Li M., Yu M., Wang X., Deng M., Zhai X., Li R.. **Improved Glucose and Lipid Metabolism in the Early Life of Female Offspring by Maternal Dietary Genistein Is Associated With Alterations in the Gut Microbiota**. *Front. Endocrinol.* (2018) **9** 516. DOI: 10.3389/fendo.2018.00516
26. Ling Z., Jin C., Xie T., Cheng Y., Li L., Wu N.. **Alterations in the Fecal Microbiota of Patients with HIV-1 Infection: An Observational Study in A Chinese Population**. *Sci. Rep.* (2016) **6** 30673. DOI: 10.1038/srep30673
27. Forbes J.D., Van Domselaar G., Bernstein C.N.. **The Gut Microbiota in Immune-Mediated Inflammatory Diseases**. *Front. Microbiol.* (2016) **7** 1081. DOI: 10.3389/fmicb.2016.01081
28. Lee G., You H.J., Bajaj J.S., Joo S.K., Yu J., Park S., Kang H., Park J.H., Kim J.H., Lee D.H.. **Distinct signatures of gut microbiome and metabolites associated with significant fibrosis in non-obese NAFLD**. *Nat. Commun.* (2020) **11** 4982. DOI: 10.1038/s41467-020-18754-5
29. Hao H., Zhang X., Tong L., Liu Q., Liang X., Bu Y., Gong P., Liu T., Zhang L., Xia Y.. **Effect of Extracellular Vesicles Derived From Lactobacillus plantarum Q7 on Gut Microbiota and Ulcerative Colitis in Mice**. *Front. Immunol.* (2021) **12** 777147. DOI: 10.3389/fimmu.2021.777147
30. Nishiyama K., Takaki T., Sugiyama M., Fukuda I., Aiso M., Mukai T., Odamaki T., Xiao J.Z., Osawa R., Okada N.. **Extracellular Vesicles Produced by Bifidobacterium longum Export Mucin-Binding Proteins**. *Appl. Environ. Microbiol.* (2020) **86** e01464-20. DOI: 10.1128/AEM.01464-20
31. Díaz-Garrido N., Badia J., Baldomà L.. **Microbiota-derived extracellular vesicles in interkingdom communication in the gut**. *J. Extracell. Vesicles* (2021) **10** e12161. DOI: 10.1002/jev2.12161
32. Suzuki T.. **Regulation of intestinal epithelial permeability by tight junctions**. *Cell. Mol. Life Sci.* (2013) **70** 631-659. DOI: 10.1007/s00018-012-1070-x
33. Kulp A., Kuehn M.J.. **Biological functions and biogenesis of secreted bacterial outer membrane vesicles**. *Annu. Rev. Microbiol.* (2010) **64** 163-184. DOI: 10.1146/annurev.micro.091208.073413
34. Park K.S., Lee J., Lee C., Park H.T., Kim J.W., Kim O.Y., Kim S.R., Rådinger M., Jung H.Y., Park J.. **Sepsis-Like Systemic Inflammation Induced by Nano-Sized Extracellular Vesicles From Feces**. *Front. Microbiol.* (2018) **9** 1735. DOI: 10.3389/fmicb.2018.01735
35. Kameli N., Becker H.E.F., Welbers T., Jonkers D.M.A.E., Penders J., Savelkoul P., Stassen F.R.. **Metagenomic Profiling of Fecal-Derived Bacterial Membrane Vesicles in Crohn’s Disease Patients**. *Cells* (2021) **10**. DOI: 10.3390/cells10102795
36. Kameli N., Borman R., Lpez-Iglesias C., Savelkoul P., Stassen F.R.M.. **Characterization of Feces-Derived Bacterial Membrane Vesicles and the Impact of Their Origin on the Inflammatory Response**. *Front. Cell. Infect. Microbiol.* (2021) **11** 667987. DOI: 10.3389/fcimb.2021.667987
37. Hiippala K., Jouhten H., Ronkainen A., Hartikainen A., Kainulainen V., Jalanka J., Satokari R.. **The Potential of Gut Commensals in Reinforcing Intestinal Barrier Function and Alleviating Inflammation**. *Nutrients* (2018) **10**. DOI: 10.3390/nu10080988
38. Koulas S.G., Stefanou C.K., Stefanou S.K., Tepelenis K., Zikos N., Tepetes K., Kapsoritakis A.. **Gut Microbiota in Patients with Morbid Obesity Before and After Bariatric Surgery: A Ten-Year Review Study (2009–2019)**. *Obes. Surg.* (2021) **31** 317-326. DOI: 10.1007/s11695-020-05074-2
39. Sankarasubramanian J., Ahmad R., Avuthu N., Singh A.B., Guda C.. **Gut Microbiota and Metabolic Specificity in Ulcerative Colitis and Crohn’s Disease**. *Front. Med.* (2020) **7** 606298. DOI: 10.3389/fmed.2020.606298
40. Mao J., Wang D., Long J., Yang X., Lin J., Song Y., Xie F., Xun Z., Wang Y., Wang Y.. **Gut microbiome is associated with the clinical response to anti-PD-1 based immunotherapy in hepatobiliary cancers**. *J. Immunother. Cancer* (2021) **9** e003334. DOI: 10.1136/jitc-2021-003334
41. Heise J., Nega M., Alawi M., Wagner D.. **Propidium monoazide treatment to distinguish between live and dead methanogens in pure cultures and environmental samples**. *J. Microbiol. Methods* (2016) **121** 11-23. DOI: 10.1016/j.mimet.2015.12.002
42. Gérard A., El-Hajjaji S., Burteau S., Fall P.A., Pirard B., Taminiau B., Daube G., Sindic M.. **Study of the microbial diversity of a panel of Belgian artisanal cheeses associated with challenge studies for Listeria monocytogenes**. *Food Microbiol.* (2021) **100** 103861. DOI: 10.1016/j.fm.2021.103861
43. Rognes T., Flouri T., Nichols B., Quince C., Mahé F.. **VSEARCH: A versatile open source tool for metagenomics**. *PeerJ* (2016) **4** e2584. DOI: 10.7717/peerj.2584
44. Schloss P.D., Westcott S.L., Ryabin T., Hall J.R., Hartmann M., Hollister E.B., Lesniewski R.A., Oakley B.B., Parks D.H., Robinson C.J.. **Introducing mothur: Open-source, platform-independent, community-supported software for describing and comparing microbial communities**. *Appl. Environ. Microbiol.* (2009) **75** 7537-7541. DOI: 10.1128/AEM.01541-09
45. Fernandes A.D., Reid J.N., Macklaim J.M., McMurrough T.A., Edgell D.R., Gloor G.B.. **Unifying the analysis of high-throughput sequencing datasets: Characterizing RNA-seq, 16S rRNA gene sequencing and selective growth experiments by compositional data analysis**. *Microbiome* (2014) **2** 15. DOI: 10.1186/2049-2618-2-15
46. Bardenbacher M., Ruder B., Britzen-Laurent N., Schmid B., Waldner M., Naschberger E., Scharl M., Müller W., Günther C., Becker C.. **Permeability analyses and three dimensional imaging of interferon gamma-induced barrier disintegration in intestinal organoids**. *Stem Cell Res.* (2019) **35** 101383. DOI: 10.1016/j.scr.2019.101383
|
---
title: Redox-Mediated Gold Nanoparticles with Glucose Oxidase and Egg White Proteins
for Printed Biosensors and Biofuel Cells
authors:
- Natcha Rasitanon
- Kornautchaya Veenuttranon
- Hnin Thandar Lwin
- Kanyawee Kaewpradub
- Tonghathai Phairatana
- Itthipon Jeerapan
journal: International Journal of Molecular Sciences
year: 2023
pmcid: PMC10002497
doi: 10.3390/ijms24054657
license: CC BY 4.0
---
# Redox-Mediated Gold Nanoparticles with Glucose Oxidase and Egg White Proteins for Printed Biosensors and Biofuel Cells
## Abstract
Glucose oxidase (GOx)-based electrodes are important for bioelectronics, such as glucose sensors. It is challenging to effectively link GOx with nanomaterial-modified electrodes while preserving enzyme activity in a biocompatible environment. To date, no reports have used biocompatible food-based materials, such as egg white proteins, combined with GOx, redox molecules, and nanoparticles to create the biorecognition layer for biosensors and biofuel cells. This article demonstrates the interface of GOx integrated with egg white proteins on a 5 nm gold nanoparticle (AuNP) functionalized with a 1,4-naphthoquinone (NQ) and conjugated with a screen-printed flexible conductive carbon nanotube (CNT)-modified electrode. Egg white proteins containing ovalbumin can form three-dimensional scaffolds to accommodate immobilized enzymes and adjust the analytical performance. The structure of this biointerface prevents the escape of enzymes and provides a suitable microenvironment for the effective reaction. The bioelectrode’s performance and kinetics were evaluated. Using redox-mediated molecules with the AuNPs and the three-dimensional matrix made of egg white proteins improves the transfer of electrons between the electrode and the redox center. By engineering the layer of egg white proteins on the GOx-NQ-AuNPs-mediated CNT-functionalized electrodes, we can modulate analytical performances such as sensitivity and linear range. The bioelectrodes demonstrate high sensitivity and can prolong the stability by more than $85\%$ after 6 h of continuous operation. The use of food-based proteins with redox molecule-modified AuNPs and printed electrodes demonstrates advantages for biosensors and energy devices due to their small size, large surface area, and ease of modification. This concept holds a promise for creating biocompatible electrodes for biosensors and self-sustaining energy devices.
## 1. Introduction
Glucose oxidase (GOx, Enzyme Commission number: 1.1.3.4) is a flavoenzyme that catalyzes the oxidation of glucose. It is widely employed for glucose biosensors to determine the amount of glucose in body fluids. GOx has a homodimeric protein structure with an active site located in a deep pocket where glucose binding takes place [1]. The structure of GOx creates one of the most challenging tasks in biosensor development when dealing with the immobilization of GOx on an electrode surface. A variety of nanomaterials introduced for surface modification in biosensor applications have been reported, such as carbon nanotubes, nanofibers, and particularly metallic nanoparticles [2]. Gold nanoparticles (AuNPs) serve as an electrical wire between the redox-active center of the enzyme and the electrode, allowing intimate interaction with GOx [3]. This example was carried out by reconstituting apo-glucose oxidase on a 1.4-nanometer gold nanocrystal functionalized with flavin adenine dinucleotide (FAD) and incorporated into a conductive film generates a bioelectrocatalytic system with electrical contact with the electrode support. The reconstitution of GOx on AuNPs via FAD provides a high electron transfer rate that is seven times higher than that of the natural cosubstrate of the enzyme, offering excellent analytical performances of electrochemical behaviors [3]. Although AuNPs have been explored as a conducting material to receive electrons from GOx, which can improve the efficiency of electron transfer, there are still limitations to the use of DET [4]. Researchers continue to face critical challenges in immobilizing GOx on solid electrodes.
Although proteins have been applied for creating bio-matrix to host enzymes, few reports have used highly biocompatible food-based biomolecules to hold enzymes on nanoparticle-based bioelectrodes. Previously, proteins, such as bovine serum albumin (BSA) and silk fibroin, could be used as an immobilization matrix for biosensor development due to their biocompatibility, biodegradability, presence of functional groups, as well as high stability. BSA is often used in the cross-linked matrix to protect the activity of oxidase-based electrodes while enhancing the sensor stability, such as the co-immobilization of methylene green and horseradish peroxidase in the montmorillonite-modified BSA-glutaraldehyde matrix for hydrogen peroxide detection [5]. Burmeister et al. reported the use of BSA and glutaraldehyde cross-linked glutamate oxidase for glutamate analysis [6]. Another example incorporated BSA with glycerol and poly(ethylene glycol) diglycidyl ether matrix to provide the stability of glucose and lactate sensors [7]. Similarly, silk fibroin is a natural protein that possesses environmentally friendly, excellent tensile strength, and many amino groups. The molecular structure provides multiple functional groups such as −OH, −NH2, and −COOH, allowing the entrapment of the biomolecules without the use of any chemical reagents [8]. Along with that, it has been reported that the use of silk fibroin as supporting materials offers long-term operational stability, maintaining GOx activities to minimize the leakage of the immobilized GOx from the matrix [9,10,11]. Kuzuhara et al. introduced a colorimetric study of GOx entrapped in silk fibroin. The results showed that the silk fibroin-based biosensor could preserve the enzyme activity ($98.7\%$) and avoid leakage of the enzymes ($0.05\%$) after 36 days [11]. Liu and colleagues recently reported on the coupling of an enzymatic silk fibroin nanofibrils membrane via glutaraldehyde with ultrathin platinum nanoparticles/graphene film for glucose and lactate sensing. The glutaraldehyde can be crosslinked with a hydroxyl group of silk fibroin and an oxidase enzyme, thus creating a porous enzymatic nanofiber membrane or three-dimensional (3D) scaffold structure. The results revealed that the fabricated sensors provided high sensitivity and long-term stability for several hours (up to 25 and 23.6 h for glucose and lactate sensors) when retaining the enzymes in the 3D space [9]. To date, there are no reports on the use of cheap and biocompatible food-based materials, such as egg white proteins, conjugated with GOx and nanoparticles to fabricate bioelectrodes for biosensors and biofuel cells (BFCs). Similar to previously reported proteins, it is expected that a 3D network can also be formed due to the availability of amino acids in egg white proteins for the glutaraldehyde crosslinking process.
The use of glucose biosensors, self-powered glucose sensors, and BFCs can revolutionize the technology for managing diabetes mellitus, a condition characterized by high blood glucose levels because the body fails to produce enough insulin. With diabetes’ increasing prevalence and complications, glucose monitoring devices are in demand. Even though glucose sensors have been developed for many years, strategies for detecting glucose still need to be developed. Electrochemical analysis is the most widely used method for measuring glucose, thanks to its simplicity, quantitative nature, and wide detection range. As soon as the sample has been analyzed, the electrochemical signals can be converted directly into glucose concentrations [12]. These days, with the proliferation of smartphones and the growing need for real-time and continuous monitoring, wearable devices are expected to become the mainstream for glucose monitoring. Interestingly, it is possible to connect glucose levels in perspiration, saliva, urine, tears, and interstitial fluid with blood glucose levels; thus, non-invasive or minimally invasive glucose monitoring is possible [13]. With the ability to continuously measure glucose, wearable sensing has opened promising avenues for delivering information to aid in diabetes diagnosis and enable early treatment intervention. Considerable efforts have been made in the development of wearable noninvasive glucose biosensors, including sweat-based devices [14,15,16,17], tear-based devices [18,19,20,21], oral cavity-based devices [22,23,24], and interstitial fluid-based devices [25,26]. Furthermore, glucose BFCs were discovered to be useful for powering sensors, as they harvest energy from the biofluid and convert it into electricity, providing self-sustaining power sources that can be utilized to monitor their environments [27]. Recently, numerous efforts have been made to glucose-based BFC in living organisms and humans with the expectation of possible future biomedical applications and self-sustaining power generation, such as a glucose-based BFC implanted in a snail to generate electrical power [28], a glucose-based BFC in a lobster capable of powering a pacemaker and an electronic watch [29,30], and a glucose-based BFC in a pigeon to generate power for intermittent neurostimulation [31].
For safety and to avoid negative reactions, bioelectrodes should be compatible with tissue, especially for flexible wearable devices. Additionally, enhancing the stability of enzymatic electrodes is necessary for consistent monitoring and to maintain the device’s performance over time, ensuring it can function for prolonged periods. Therefore, it is essential to exploit new biocompatible molecules to create an immobilization matrix for developing bioelectrodes. We aim to design glucose biosensors and self-powered sensors, while BFCs provide a sustainable energy source for biodevices.
This work describes the first example investigating bioelectrochemical interface of GOx and egg white proteins on 5 nm AuNPs functionalized with a redox mediator molecule (1,4-naphthoquinone, NQ) and integrated on a screen-printed conductive carbon nanotube (CNT)-modified material. The effective conjugation of redox-mediated AuNPs with GOx and egg white proteins was investigated by evaluating bioelectrochemical kinetics. Different electrode configurations were studied. This bioelectrode could be applied to create an energy-harvesting device that can convert glucose into electricity. *The* generated electricity could also drive the self-powered biosensing module. Applying biocompatible food-based proteins and a newly engineered bioelectrode interface could improve biosensing functions and bioenergy conversion systems by connecting the GOx with redox molecule-modified nanoparticles and maximizing electron flow from the redox center of GOx for advanced biosensing applications while maintaining the activity of the GOx.
## 2.1. The Concept of Redox-Mediated AuNPs with GOx and Egg White Proteins Conjugated with a Printed CNT-Modified Amperometric Biosensor and a BFC
We describe a new electrochemical biosensing interface relying on GOx with egg white proteins and 5 nm AuNPs functionalized with redox mediator molecules (NQ), coated on our lab-made screen-printed CNT-modified electrode (Figure 1). This enzyme-based electrode was based on glucose oxidation catalyzed by GOx with the help of NQ-mediated AuNPs in the matrix of egg white proteins.
We modified a screen-printable ink by adding CNTs to conductive ink due to their high conductive properties and ability to act as efficient current collectors. The addition of CNTs with a high aspect ratio (~700–6000) promoted the percolation of ink within its matrix, thereby facilitating the flow of electrons in electrochemical processes [32]. Then, the printed CNT-modified electrode was functionalized with a GOx enzyme, co-immobilized with egg white proteins and NQ-AuNPs, to enhance the enzyme stability, conductivity, and adsorption surface (Figure 1A).
GOx from Aspergillus niger is a flavoprotein that utilized molecular oxygen as an electron acceptor in GOx-catalyzed oxidation of β-D-glucose at its linked hydroxyl group to produce gluconolactone and hydrogen peroxide. It is recognized as an ideal enzyme for biosensor application due to its high specificity toward β-D-glucose, stability, high activity, and commercial availability [33]. Due to the advantages of GOx enzyme, biosensing has been shown to be a revolutionary approach in various fields in recent decades, ranging from environmental to biomedical applications, including diabetes control.
The first generations of biosensors that measure the concentration of the depletion of oxygen or the products of enzymatic processes were developed. This class of biosensors is oxygen-dependent, which relies on the use of natural oxygen. Glucose concentration is determined by following the consumption of oxygen or the generation of hydrogen peroxide [33,34]. However, the major obstacles of first-generation electrochemical biosensors in analyzing real samples are the interference of electroactive species (such as ascorbic acid) and the effect of oxygen. This difficulty has been tackled with a number of strategies.
One of the most useful strategies is to design an efficient mediated electrochemical biosensor, which is a second-generation biosensor. This class utilizes mediators as redox agents to act as electron carriers, thus replacing the oxygen in the reaction. For the GOx-based electrode, movement of electrons specifically occurs when the GOx enzyme catalyzes the oxidation of glucose to gluconolactone via the reduction of the FAD, the redox center of GOx, to FADH2 [35]. However, the 3D structure of GOx reveals that FAD (redox cofactor) is deeply buried within the protein shell. Thus, the electron transfer between the GOx active site and the electrode surface in GOx-based biosensors is limited by a thick protein layer of GOx surrounding a FAD center. This prevents the electron from communicating directly with the electrode surface. Since the electron transfer from the FAD active site of GOx to the electrode surface is challenging, the use of a mediator molecule can help to shuttle electrons, which is called mediated electron transfer (MET) [4]. Hence, we designed the mediator-based glucose biosensor in this study to enhance electron transfer, reduce interference effects, and mitigate the influence of oxygen. We utilized NQ as a redox mediator to transport electrons between the FAD center of GOx and the electrode surface. *In* general, quinones are a family of carbonyl compounds with two carbonyl groups in a six-member ring structure. Due to their excellent electrochemical reversibility, rapid redox kinetics, stable structure, and low molecular weight (158.16 g mol–1), redox-active quinones compounds have the potential to rapidly transfer electrons from the enzyme redox center to our CNT-modified working electrode surface, achieving high-rate capability and long-term cycle stability [36,37]. However, only the immobilization of NQ on the electrode by physical absorption or noncovalent interactions can lead to leaching from the electrode surface [38]. Thus, a conductive nano-matrix with CNTs and AuNPs is an important material for NQ immobilization. NQ can be strongly integrated with the bottom layer of the CNT-based electrode since NQ can interact with a CNT-modified electrode by means of π–π interactions between the aromatic groups of quinones and CNT surfaces [37]. The electrical wiring from the FAD center to CNTs with NQ as a linker is possible for enhancing the electron-transfer rate. In addition to using NQ, we additionally connected AuNPs with this mediator. The possible interaction between NQ and AuNPs could involve adsorption. The surface-to-volume ratio of a AuNP in this work was 1.2 nm2/nm3. The increase in active surface area induced by the presence of conductor nanomaterials was expected to improve electron transfer kinetics, resulting in improved glucose-sensing sensitivity and power density for BFCs.
However, a key barrier to achieving stable biosensors is the problems associated with enzyme immobilization. Thus, the stability of enzymes is important for enzyme-based glucose sensors. In conventional devices, substrates for immobilizing enzyme are often conducting polymers, semiconductors, metals, and carbon-based materials, which have some disadvantages, including high cost, potential environmental risks, and non-renewability [8].
In this work, raw egg white proteins (mainly ovalbumin) were employed as an immobilization matrix to entrap the GOx enzyme. Ovalbumin is a unique protein biopolymer, consisting of both hydrophobic domains (at both ends of the chain) and hydrophilic domains (at the middle of the chain) [39]. As shown in Supplementary Figure S1, the peptide chain of ovalbumin, with a molecular weight of about 45 kDa, consists of 386 amino acids, such as lysine residues, glycine, and proline [40], which offer a large number of functional groups such as −COOH and −NH2. Due to having both hydrophilic and hydrophobic regions, the protein in egg whites can readily undergo a conformational transition between a water-soluble structure and a water-insoluble structure in response to changes in temperature or chemicals [41]. Additionally, the availability of amino acids is important for the glutaraldehyde crosslinking process. The GOx enzymatic network formed by cross-linking egg white proteins serves as 3D skeletons to which GOx could be linked by glutaraldehyde, resulting in a large scaffold for biochemical reactions with immobilized enzymes (Supplementary Figure S2). It was shown that the amide bond of the NH2 molecule, when it came in contact with the carbonyl group of glutaraldehyde, could react to form the –N=C– bond by losing a molecule of water [42]. In this way, egg white proteins crosslinked with glutaraldehyde were formed as a 3D skeleton. The Fourier-transform infrared spectroscopy (FTIR) was further conducted to examine functional groups and the bonding information of egg white proteins with and without glutaraldehyde crosslinking (Supplementary Figure S3). Specifically, the FTIR spectra of egg white proteins without glutaraldehyde showed a peak at 1550 cm–1, indicating the presence of the N–H bond, which is a characteristic functional group of proteins. However, the intensity of this peak was reduced when egg white proteins were crosslinked with glutaraldehyde. The C–H stretching vibration (at around 2959 cm–1) and the peak of C–C bonding (at around 1004 cm–1) were more prominent when crosslinking with egg white proteins. This suggested the successful crosslinking reaction between egg white proteins and glutaraldehyde. Both the interaction between binding sites of egg white proteins and enzyme and a porous structure formed during the crosslinking process of protein can serve as a skeleton for the immobilization of enzyme, preventing the enzyme from leaking off the electrode surface and ensuring the sensor’s stability [9].
Importantly, biocompatible bioelectronic devices are of particularly great interest regarding biomedical applications. In comparison with other commercial proteins, egg white proteins are less expensive. For example, egg white proteins cost only 0.003 USD per 1 g, whereas BSA cost much more (25 USD per 1 g). Since egg white is safe, biocompatible, and cheap, these characteristics make it a preferable choice as a material for sensing applications, such as a stabilizer of nanoclusters [43], a hydrogel crosslinker [44], and a coating layer of nanoparticles [45], and also for modern applications such as wearable devices and soft bioelectronics that have non-inflammatory contact with human tissue. Therefore, with this strategy of enzyme immobilization, our design of an enzyme-based electrode has a stable biosensor structure.
For the kinetic of the redox mediator-modified nanoparticles–enzymatic proteins interaction, an electron mediator should be capable of efficiently facilitating electron transfer at a specific potential and must be able to transfer electrons rapidly. The study of electrochemical enzyme kinetics of redox-mediated GOx reactions is important since it can help determine the enzyme’s affinity to the substrate (glucose) and its maximum catalytic efficiency. The Michaelis–Menten constant (Km) is an important enzyme-substrate kinetic indicator of the corresponding enzyme [46]. In the heterogeneous system, we investigated the interaction between GOx and glucose (substrate) using an amperometric technique to observe the number of generated electrons per time unit (Figure 1B). In electrochemical kinetics, the current output can be correlated to the reaction rate with successive additions of the substrate. The dependence of the current responses on substrate concentration showed the characteristics of the Michaelis–Menten kinetic mechanism (inset of Figure 1B). For enzymes that conformed to the Michaelis–Menten mechanism, the early phase of increasing substrate concentrations results in a rapid increase in current, followed by a gradual increase as the enzyme approached its maximum activity. Due to enzyme saturation, the maximum current achieved at high substrate concentrations is fully in enzyme-substrate complex form. Thus, the curve depicts the kinetic parameters defining the curve’s high and low substrate concentration boundaries.
In this article, we also demonstrated the working operation for energy-harvesting and self-powered sensing modules by coupling these egg white proteins/GOx/NQ-AuNPs/CNT-modified bioelectrodes with a printed Pt-based cathode (Figure 1C). The reaction occurring on the bioanode was based on glucose oxidation catalyzed by GOx, resulting in electrons being released. These electrons went through the power load of the self-powered biosensing unit toward the Pt-based cathode; eventually, natural oxygen reached the cathode to gain the electrons, completing the power circuit. Our BFC (using redox-mediated AuNPs with GOx immobilized by egg white proteins on the anode and the Pt/CNT-based catalysts on the cathode) could convert chemical energy in glucose molecules into electrical energy in the form of electricity (Figure 1C, right-top). We also observed that this electrical energy produced by redox reactions could be proportional to the concentration of glucose. Consequently, the screen-printable device can also function as a self-powered biosensor capable of measuring glucose concentration without the need of an external energy supply to force the glucose oxidation (Figure 1C, right-bottom).
## 2.2. The Effect of AuNPs on Glucose Oxidation Kinetics
The effect of AuNPs on glucose oxidation kinetics at the four different modifications of the CNT-based screen-printed bioelectrodes using amperometry was studied in terms of sensitivity, Imax, and Km. Figure 2 illustrates the four different approaches of electrode modification based on the different number of AuNPs, including 0, 27, 137, and 274 G-units in a matrix of NQ (750 nmol), and the enzyme layer (25 units of GOx) with the egg protein layer were controlled (Figure 2A). The amperometric responses of the different modified screen-printed bioelectrodes were recorded at different glucose concentrations in a range of 0.0 to 40 mM by holding a potential of 0.40 V vs. Ag/AgCl (Supplementary Figure S4). Figure 2B shows the dependence of current response on glucose substrate concentrate ion, which agreed with the Michaelis–Menten equation (Equation [1]). This curve shows that currents increased rapidly after glucose addition, and slowly before reaching their limit. The maximal current at high glucose concentrations was due to saturation. Adding more glucose would no longer increase the current. The calibration plots in Figure 2B (a–d) of the CNT screen-printed electrodes with AuNPs incorporated in a NQ matrix revealed that the current response of the 274 G-unit of AuNPs-modified bioelectrode (trace d), which had twice and 10-times AuNPs higher than Electrode c and Electrode b showed the highest current signals, followed by 137 and 27 G-unit of AuNPs (c and b), respectively. [ 1]I=Imax[C]Km+ [C] where I is the steady-state current after the addition of glucose, *Imax is* the maximum current obtained from saturated glucose concentrations, C is the glucose concentration, and *Km is* the Michaelis–Menten constant.
Additionally, a supporting equation presented the double reciprocal relationship following the Lineweaver–Burk equation (Equation (S1)) which was algebraically transformed from the Michaelis–Menten equation. Moreover, the Hanes–Woolf equation (Equation (S2)), and the Eadie–Hofstee (Equation (S3)) were also used to calculate Km and Imax values (see details in Supporting Information). The Km and Imax from straight lines of the double reciprocal plot, the Hanes–Woolf plot, and the Eadie–Hofstee plot were also demonstrated (Supplementary Figure S5).
The parameters of enzymatic kinetics are shown in Figure 2C. When using NQ-AuNPs, the Km value could reach ~3 mM, indicating good apparent affinity of GOx immobilized on the bioelectrode to the glucose target. Regarding the ratio of AuNPs and NQ molecules, the results indicated that highly conductive AuNPs offered a high specific surface area for GOx. Hence, the number of AuNPs had a clear impact on the analytical performances of the glucose. The greater the number of AuNPs, the higher Imax and higher sensitivity were obtained. Regarding the effect of AuNPs on the Imax/Km ratio, it appears that the use of AuNPs on Electrode d leads to a significantly higher Imax/Km ratio compared to Electrode a. The Imax/Km ratio was an indicator of the catalytic efficiency of an enzyme. The 2.1 to 3.0 times increase in the Imax/Km ratio of Electrode d compared to Electrode a, as calculated from various kinetics models (Michaelis–Menten, Lineweaver–Burk, Hanes–Woolf, and Eadie–Hofstee plots), further confirms the advantages of using AuNPs in electrochemical sensing. Interestingly, the Imax response of the egg white proteins/GOx/NQ-based bioelectrodes without AuNPs (*Electrode a* in Figure 2) was higher than Electrode b (27 G-units of AuNPs). However, these current signals of *Electrode a* saturated at a lower concentration (at 10 mM glucose). The sensitivity of *Electrode a* was higher than Electrodes b and c. However, a maximum current of Electrode c was higher than electrode a. This might be because the small amount of AuNPs dispersed in 0.1 mg mL−1 sodium citrate could attribute the AuNPs-NQ surface to be a negative charge, while the GOx had a negative charge at pH 7.0 [47]. This could lead to repelling each other, resulting in the observation that Electrode b displayed a lower maximum current response than Electrode a.
## 2.3. The Effect of Layout Arrangements on Glucose Oxidation Kinetics
The effect of layout arrangements on glucose oxidation kinetics at the two different modifications of the layers was studied. Figure 3A illustrates the comparison of electrode modification based on the different layer arrangement while maintaining a constant total volume of 10 µL of $10\%$ v/v egg white proteins, 10 µL of 10 mg mL−1 GOx (in $10\%$ v/v egg white proteins), and 274 G-units of AuNPs with 750 nmol of NQ). The amperometric responses of the different modified screen-printed electrodes (Supplementary Figure S6) were performed at different glucose concentrations in the range of up to 40 mM. The calibration plots of two different layout arrangements were demonstrated in Figure 3B. Additionally, the double reciprocal plot of the calibration curve was obtained from screen-printed bioelectrodes (Supplementary Figure S7). It is clearly seen from the curve that Electrode b showed a lower Imax than Electrode a. Considering the sensitivity, Imax, and Km values, the effectiveness of the layout arrangement could be evaluated (Figure 3C). Because Electrode b had lower sensitivity and Imax than Electrode a, it confirmed that the layout arrangement of the bioelectrode is important. As discussed earlier, the GOx layer needed to be arranged close to the redox-mediated AuNPs to assist the electron transfer from the enzyme redox center to the electrode surface. This study confirmed that our design of the bioelectrode layering is suitable for further development of advanced biosensing.
## 2.4. The Effect of the Thickness of Egg White Proteins on Glucose Oxidation Kinetics
It is important to secure the enzyme immobilized on the electrode surface with the highest possible specific activity. Immediately adjacent to the NQ-AuNPs-mediated CNT-modified electrode was the GOx layer, which was formed by the entrapment of GOx within the protein matrix by covalent glutaraldehyde cross-linking with egg white proteins. The protein as an immobilization matrix could prevent the leakage of GOx into the analyte solution and act as a diffusional barrier for the substrate. A carefully determined protein layer thickness is also advantageous to tune biosensor performances. Therefore, the effect of the thickness of egg white proteins on glucose oxidation kinetics was investigated. Figure 4A shows the effect of the variation of egg white proteins solution (0, 10, 20, and 40 µL) on the performance of glucose sensing (amperometric current response to glucose addition). As a reference, the bioelectrode without the egg white proteins solution on the top layer was fabricated (Figure 4A(a)). The calibration plots are shown in Figure 4B. The increase in thickness of the egg white proteins layer resulted in a lower maximum current and a lower sensitivity, as shown in Figure 4C. This may have been due to the porosity of the outermost surface (the layer of egg white proteins); the thicker protein matrix layer increased the diffusion length, resulting in slower glucose molecule diffusion. However, this finding is also advantageous. Most enzymes conform to the Michaelis–Menten kinetic, in which the reaction is largely non-linear with substrate addition. The thicker protein layer could broaden the linear range of the substrate’s detection. In addition to extending the linear range, the thicker layer can also protect the catalytic enzyme. A porous surface of egg white proteins protects the GOx layer from leaching out while allowing glucose to diffuse to the enzymatic layer. Adjusting the thickness of the protein layer allows a flexible screen-printed CNT-based bioelectrode to detect glucose over an adjustable range, as the response is controlled by diffusion through the egg white proteins’ matrix and not solely enzymatic kinetics. This design allows biosensors to detect glucose efficiently since glucose could be easily diffused into the outermost layer. The good porosity of the egg-protein matrix could be evaluated by considering the sensitivity, Imax, and Km values. Electrode a (without the topmost layer of egg white proteins) and Electrode b (with the topmost layer of egg white proteins) had similar Imax values, but *Electrode a* had a lower Km value. This may have been because the presence of egg white proteins on the surface decreased the apparent affinity of GOx. However, both *Electrode a* and b showed higher current values than Electrode c and d, respectively. Since the small Km value confirmed the high affinity of the GOx immobilized on the bioelectrode, it was apparent from the *Electrode a* and b (i.e., 0 and 10 µL of egg white solution, respectively) that having a protein layer with only a small amount had no major effect on the accessibility of glucose to GOx. It could still maintain the small value of Km and a good sensitivity while offering a bio-friendly environment for GOx. The configuration of the protein layer on the NQ-AuNPs-mediated CNT-modified electrodes, which is a simple way to modulate analytical performances (sensitivity and linear range), enabled the bioelectrodes to be used to measure glucose concentration in a desirable range of analytical substances.
## 2.5. The Effect of Egg White Proteins on Operational Stability
The operational stability of the electrode was a particular concern in terms of its future applications. In Figure 5, the operational stability of the device was evaluated in a batch system for 6 h in 0.1 M phosphate buffer solution (PBS), pH 7.0, containing 10 mM glucose to demonstrate the feasibility of the bioelectrode in a continuous operation mode. The electrodes with and without egg white proteins in their layer configuration (Figure 5A(a) and Figure 5A(b), respectively) were compared to determine the influence of egg white proteins matrix on electrode stability. Under a constant potential of 0.40 V vs. Ag/AgCl, it is clearly seen from Figure 5B that the electrode with the egg white proteins matrix at the outermost layer possessed a stable current response throughout 6 h period. For the first three hours, the current response decreased to about $87\%$ and $47\%$ for the electrode with and without egg white proteins, respectively, while for the last three hours, the current response showed overall stability because at the 6th h, it went up and down to almost the same value as at the 3rd h. Overall, the electrode with egg white proteins showed only about a $12\%$ decrease from the initial stage, while the electrode without the egg white proteins matrix faced a sharp decline in the current response, ending with less than half of its initial ability after 6 h of measurement. The bioelectrode consisting of the NQ-AuNPs nanocomposite film and porous structure formed during the crosslinking process of egg white proteins with GOx could firmly hold enzyme molecules and facilitate electron flow between the redox center of the enzyme and the electrode surface, resulting in a more stable sensing performance than the electrode structure without a protein matrix layer. Moreover, as shown in Supplementary Figure S2, there were a large number of amino acids in egg white proteins; thus, a large scaffold for biochemical reactions can be formed by the crosslinking between egg white proteins’ amino acids and GOx, and also between GOx and glutaraldehyde. This 3D skeleton could prevent the enzyme from leaking off the electrode surface, resulting in the stability of the electrode.
Enzyme immobilization with other protein matrices was also reported to protect the enzyme’s activity while enhancing the electrode’s stability. Similarly, the use of silk fibroin nanofibrils, a natural protein in arthropods, was shown to improve operational stability since its structure is based on nanoporous enzymatic membranes formed by the cross-linking of silk fibroin nanofibrils with GOx [9]. To construct a cross-linked matrix to protect enzymes, BSA was also reported to be used as co-immobilization matrix with glutaraldehyde for the hydrogen peroxide sensor [5]. Unlike other immobilization materials, our bioelectrodes used low-cost, highly biocompatible food-based biomolecules to hold enzymes on nanoparticle-based bioelectrodes, which provide an advantage in further applications, such as wearable and flexible devices that should be compatible with human skin. Our result confirmed that the electrode with egg white proteins could function as an immobilization matrix to entrap enzymes within the electrode surface without leaking off into the solution, and the electrode was still functional after many hours of continuous operation.
## 2.6. Studies of a BFC and a Self-Powered Sensor Using a Screen-Printed Egg White Proteins/GOx-Egg White Proteins/NQ-AuNPs-Based Bioelectrode and a Pt-Based Cathode
With the concept of flexible bioelectronics, many useful and versatile applications are possible, including wearables [48]. However, the developed bioelectrodes should be compatible with human skin for safety reasons. Enzymatic BFCs, which transform the biological energy available in human biofluids into electricity, are one of the most potent energy generation alternatives, as a sustainable energy source for flexible bioelectronics. Because of their benefits for operation with enzymes that are active at room temperature and under mild physiological circumstances, enzymatic BFC is good for contact with human skin, enabling on-body applications [49]. BFCs, in addition to being energy-conversion devices, can be utilized as self-powered electrochemical biosensors to detect analytes without the need for external power. The oxidation reaction occurs at the bioanode when glucose interacts with the active site of GOx, whereas the reduction reaction occurs at the cathode. The bioanode and the cathode have different electrical potentials, resulting in the flow of electrons. Therefore, we aimed to evaluate our BFC to confirm its functions as an energy harvester and as a self-powered sensing device.
The energy conversion performance of the fully functionalized BFCs using the bioanode (10 µL of $10\%$ v/v egg white proteins, 10 µL of 10 mg mL−1 GOx in $10\%$ v/v egg white proteins, and 274 G-units of AuNPs with 750 nmol of NQ) and a printed Pt-based cathode is shown in Figure 6A,B. The BFC’s performance was tested in different concentrations of glucose under pH 7.0 and ambient conditions. Various potential differences were applied between the bioanode and cathode to evaluate the BFC performance at varying glucose concentrations. A lab-made screen-printed egg white proteins/GOx-egg white proteins/NQ-AuNPs/CNT-modified bioelectrode coupled with the Pt-based cathode provided an open circuit voltage of 220 mV, a maximum output density of 0.9 μW, and a maximum current density of 1.4 μA. The power output was linearly proportional to the analyte concentration in the low glucose concentration range and reached a plateau at high concentrations. Self-powered glucose biosensors can benefit from this characteristic for further development.
The screen-printed glucose BFC was investigated as a self-powered glucose biosensor. The short-circuit method was employed to obtain self-powered detection without the requirement of any externally applied potential. The current response generated by the BFC itself was observed with successive additions of glucose ranging from 0 to 5 mM (Figure 6C,D). The saturation attained at 5 mM could be because the cathode was a limiting factor for our BFC. In the BFC mode, the bioanode uses GOx to catalyze the oxidation of glucose to generate electrons. These electrons are then transferred to the cathode, where the oxygen reduction reaction (ORR) occurs to consume the electrons and protons to form water [49,50], resulting in the generation of the electric current. In this study, the glucose oxidase-based bioelectrode acted as the bioanode, while the Pt-based electrode acted as the cathode. On the cathode surface, the rate of ORR could be limited by several factors, including the concentration of oxygen and the availability of active sites on the cathode surface. At high glucose concentrations of 5 mM, the bioanode produced a large number of electrons, which needed to be consumed by the cathode. In our BFC system, the rate of ORR on the cathode surface may not be able to keep up with the rate of electron and proton production on the bioanode; thus, the overall current output was limited. However, without external power or potentiostat, glucose could be sensed using our self-powered BFC consisting of the egg white proteins/GOx-egg white proteins/NQ-AuNPs-based bioelectrode. Moreover, the device was expected to run in a stand-alone mode because the generated power was proportional to the analyte concentration.
## 3.1. Chemicals and Materials
Multiwalled CNTs ($95\%$ purity, diameter = 5–15 nm, length = 10–30 μm) were from Luoyang advanced material Co., Ltd., Shanghai, China. D-(+)-glucose anhydrous was from Fluka, Neu-Ulm, Germany. Gox (from Aspergillus niger, type VII, >100,000 units g–1), NQ, glutaraldehyde solution (grade II, $25\%$ in H2O), platinum black, and Nafion-117 solution were from Sigma-Aldrich, (Saint Louis, MO, USA). Potassium phosphate dibasic (K2HPO4) and potassium phosphate monobasic (KH2PO4) were from BDH Laboratory Supplies Poole, UK. Sodium carbonate (Na2CO3) was from Ajax Finechem, (New South Wales, Australia). Tetrahydrofuran (THF) was from Honeywell, International Inc., (Charlotte, NC, USA). Toluene was from Guangdong Guanghua Chemical Factory Co., Ltd., Guangdong, China. AuNPs (5 nm in 0.1 mg mL−1 sodium citrate with stabilizer, 5.47 × 1013 particles mL−1) were from Thermo Fisher Scientific Waltham, MA, USA. Ethanol was from RCl Labscan Ltd., (Bangkok, Thailand). Acetone was from VWR International Ltd., (Poole, UK. Chicken eggs (large size, weight 63–73 g) were brought from Charoen Pokphand Foods PCL, (Bangkok, Thailand). All chemical solutions were prepared using ultrapure deionized water (18.2 MΩ cm) from a Milli Q Merck system, (Darmstadt, Germany). 0.1 M PBS was produced with a pH value of 7.0 and used as the supporting electrolyte.
## 3.2. Instruments and Electrochemical Measurement
The electrochemical performance of a screen-printed bioelectrode was evaluated in 0.1 M PBS, pH 7.0. A potentiostat/galvanostat (Autolab Type III FRA 2, Metrohm) and µStat-I 400 (Dropsens, Metrohm) were used to examine electrochemical performances, including linear sweep voltammetry and amperometry. Working electrodes (such as screen-printed egg white proteins/GOx/NQ-AuNPs/CNT-modified electrodes) were evaluated with Pt counters and Ag/AgCl (3.0 M KCl) reference electrodes during the three-electrode-system analysis. The amperometric response was measured, after 30 s immersion in the test solution, by stepping the potential to 0.40 V (vs. Ag/AgCl) for 30 s. Calibration plots were obtained by increasing the glucose concentration in 0.1 M PBS, pH 7.0. We set a two-electrode system to determine the BFC performances as an energy harvester and as a self-powered sensing device. Open circuit voltage (OCV) was assessed prior to recording a polarization plot. We recorded the BFC signal’s polarization curve between OCV and 0 V. The scan rate for recording the curve was 1 mV s–1. The autonomous biosensing method was recorded using a digital multimeter (Keysight model 34465A) coupled with Keysight BenchVue Software. The short-circuit method was employed to obtain self-powered detection. Infrared spectra (IR) were measured by an FTS165 FTIR spectrometer.
## 3.3. Preparation of CNT-Modified Carbon Ink
To prepare a CNT-modified ink, 30 mg of CNTs were dispersed in 490 µL of toluene using a homogenizer probe (model AR-0975) level 1 for 3 min. Next, 7 g of the carbon conductive ink (Guangzhou Print Area Technology Co. Ltd., Guangzhou, China), was mixed with the mixture using a mixing machine (Shashin Kagaku, Kakuhunter, SK-300SII, Japan) for 30 min at 2000 rpm. The resulting CNT-modified carbon ink was printed on PET as a working electrode. In this study, the area of the working electrode was 0.3 cm × 0.5 cm.
## 3.4. Egg White Proteins/GOx/NQ-AuNPs-Based Immobilization
The printed CNT-modified carbon electrode was cleaned with 1 M Na2CO3 at a potential of 1.5 V for 60 s. The CNT-modified carbon electrode was functionalized by dropping 10 µL of NQ-AuNPs (prepared by mixing 5 µL of AuNPs solution with 5 µL of 150 mM NQ dissolved in ethanol/acetone mixture (a volume ratio of 4:1)). In other words, this resulting electrode was coated with 274 G-units of AuNPs and 750 nmol of NQ. The dropped layer was left to dry at room temperature. Next, 10 µL of 10 mg mL−1 GOx (dissolved in $10\%$ v/v of egg white solution in 0.05 M PBS, pH 7.0) was dropped on an electrode surface. After drying at room temperature, 10 µL of 10 % v/v of egg white in 0.05 M PBS, pH 7.0 was dropped above NQ-AuNPs layer. Then, 10 µL of 0.75 wt% glutaraldehyde was dropped on the bioelectrode and left to dry at room temperature. Additionally, different electrode configurations were prepared to investigate different enzyme kinetics of glucose oxidation. Effects of AuNPs dispersion were studied by varying numbers of particles at 0, 27, 137, and 274 G-units of AuNPs with a constant amount of NQ (750 nmol). The different molarities of AuNPs in the AuNPs/NQ mixture were controlled to be 0, 4.54, 22.7, and 45.4 nM, while the molarity of NQ in the AuNPs/NQ mixture was controlled to be 75 mM to achieve different numbers of AuNPs per one electrode (0, 27, 137, and 274 G-units, respectively). The effect of layer was studied by varying the arrangement of the component of egg white proteins/GOx/NQ-AuNPs on screen-printed bioelectrodes which maintained a total volume of 10 µL of egg white proteins, 10 µL of 10 mg mL−1 GOx in $10\%$ v/v egg white proteins, and 274 G-units of AuNPs with 750 nmol NQ. The effect of the thickness of egg white proteins was studied by varying volumes of egg white proteins at 10, 20, and 40 µL of $10\%$ v/v egg white solution.
## 3.5. GOx/NQ-AuNPs-Based Immobilization without Egg White Proteins
To set a control experiment, the modified bioelectrode without egg white proteins coated on the outermost layer was prepared. The control electrode was coated with 10 µL of NQ-AuNPs (prepared by mixing 5 µL of AuNPs solution with 5 µL of 150 mM NQ). Next, 10 µL of 10 mg mL−1 GOx (dissolved in 0.05 M PBS, pH 7.0) was dropped on NQ-AuNPs/CNT-modified bioelectrode. Then, 10 µL of 0.75 wt% glutaraldehyde was dropped on the bioelectrode and left to dry at room temperature.
## 3.6. Preparation of a Pt-Based Cathode
The cathode was covered with a layer of carbon ink modified with CNTs. The cathode was fabricated by drop casting solution of 5 μL of 5 mg mL−1 CNTs dispersed in 10 mg mL−1 platinum black in ethanol solution. Then, 3 µL of $0.5\%$ Nafion ® solution was dropped and dried overnight at room temperature.
## 4. Conclusions
We have presented a new approach to fabricating a glucose amperometric biosensor and a bioanode for a BFC and self-powered biosensor using redox-mediated AuNPs, GOx, and egg white proteins. This highly biocompatible and cost-effective strategy enhances the long-term stability and self-powering capabilities of the biosensor while maximizing electron flow from the redox center of GOx. The results of our study provide insight into the relationship between the immobilization of enzymes, nanoparticles, and food-based proteins and the subsequent bioelectrochemical behavior. Due to their versatility, egg white proteins are considered useful for in vitro themes as biomaterials because they are inexpensive, commercially available, and biocompatible. Vividly, AuNPs have a great impact on electrochemical performances including high sensitivity and Imax because of excellent glucose oxidation currents and high surface-to-volume ratio properties. Moreover, adjusting the thickness of the egg white proteins layer may allow a wider linear range due to the diffusion limiting. In addition, through the studies of bioelectrodes, we demonstrated the continuous monitoring system in the batch cell system for 6 h utilizing the bioelectrodes with egg white proteins. We ascertained that the egg white proteins can support the stability of the bioelectrodes compared to non-egg white proteins bioelectrodes. These findings have important implications for the development of advanced biosensing and bioenergy conversion systems, as we developed the self-powered biosensor and the outcome stated that the generated power was proportional to the analyte concentrations. Herein, this flexible platform could be a promising tool as a portable and stand-alone device eliminating the need for a potentiostat. Future work should focus on studying a surface charge investigation for an in-depth understanding of AuNPs charge transferring and enhancing the performance of the cathode by engineering catalytic materials to improve ORR electrocatalysis.
## References
1. Wilson R., Turner A.P.F.. **Glucose oxidase: An ideal enzyme**. *Biosens. Bioelectron.* (1992) **7** 165-185. DOI: 10.1016/0956-5663(92)87013-F
2. Hou S., Zhang A., Su M.. **Nanomaterials for Biosensing Applications**. *Nanomaterials* (2016) **6**. DOI: 10.3390/nano6040058
3. Xiao Y., Patolsky F., Katz E., Hainfeld J.F., Willner I.. **“Plugging into Enzymes”: Nanowiring of Redox Enzymes by a Gold Nanoparticle**. *Science* (2003) **299** 1877-1881. DOI: 10.1126/science.1080664
4. Bartlett P.N., Al-Lolage F.A.. **There is no evidence to support literature claims of direct electron transfer (DET) for native glucose oxidase (GOx) at carbon nanotubes or graphene**. *J. Electroanal. Chem.* (2018) **819** 26-37. DOI: 10.1016/j.jelechem.2017.06.021
5. Lei C., Deng J.. **Hydrogen Peroxide Sensor Based on Coimmobilized Methylene Green and Horseradish Peroxidase in the Same Montmorillonite-Modified Bovine Serum Albumin−Glutaraldehyde Matrix on a Glassy Carbon Electrode Surface**. *Anal. Chem.* (1996) **68** 3344-3349. DOI: 10.1021/ac960291n
6. Burmeister J.J., Davis V.A., Quintero J.E., Pomerleau F., Huettl P., Gerhardt G.A.. **Glutaraldehyde Cross-Linked Glutamate Oxidase Coated Microelectrode Arrays: Selectivity and Resting Levels of Glutamate in the CNS**. *ACS Chem. Neurosci.* (2013) **4** 721-728. DOI: 10.1021/cn4000555
7. Samper I.C., Gowers S.A.N., Booth M.A., Wang C., Watts T., Phairatana T., Vallant N., Sandhu B., Papalois V., Boutelle M.G.. **Portable Microfluidic Biosensing System for Real-Time Analysis of Microdialysate in Transplant Kidneys**. *Anal. Chem.* (2019) **91** 14631-14638. DOI: 10.1021/acs.analchem.9b03774
8. Prakash N.J., Mane P.P., George S.M., Kandasubramanian B.. **Silk Fibroin As an Immobilization Matrix for Sensing Applications**. *ACS Biomater. Sci. Eng.* (2021) **7** 2015-2042. DOI: 10.1021/acsbiomaterials.1c00080
9. Liu X., Zhang W., Lin Z., Meng Z., Shi C., Xu Z., Yang L., Liu X.Y.. **Coupling of Silk Fibroin Nanofibrils Enzymatic Membrane with Ultra-Thin PtNPs/Graphene Film to Acquire Long and Stable On-Skin Sweat Glucose and Lactate Sensing**. *Small Methods* (2021) **5** 2000926. DOI: 10.1002/smtd.202000926
10. Liang B., Fang L., Hu Y., Yang G., Zhu Q., Ye X.. **Fabrication and application of flexible graphene silk composite film electrodes decorated with spiky Pt nanospheres**. *Nanoscale* (2014) **6** 4264-4274. DOI: 10.1039/C3NR06057H
11. Kuzuhara A., Asakura T., Tomoda R., Matsunaga T.. **Use of silk fibroin for enzyme membrane**. *J. Biotechnol.* (1987) **5** 199-207. DOI: 10.1016/0168-1656(87)90016-2
12. Lee H., Hong Y.J., Baik S., Hyeon T., Kim D.-H.. **Enzyme-Based Glucose Sensor: From Invasive to Wearable Device**. *Adv. Healthc. Mater.* (2018) **7** 1701150. DOI: 10.1002/adhm.201701150
13. Johnston L., Wang G., Hu K., Qian C., Liu G.. **Advances in Biosensors for Continuous Glucose Monitoring towards Wearables**. *Front. Bioeng. Biotechnol.* (2021) **9** 733810. DOI: 10.3389/fbioe.2021.733810
14. Bhide A., Muthukumar S., Prasad S.. **CLASP (Continuous lifestyle awareness through sweat platform): A novel sensor for simultaneous detection of alcohol and glucose from passive perspired sweat**. *Biosens. Bioelectron.* (2018) **117** 537-545. DOI: 10.1016/j.bios.2018.06.065
15. Xuan X., Yoon H.S., Park J.Y.. **A wearable electrochemical glucose sensor based on simple and low-cost fabrication supported micro-patterned reduced graphene oxide nanocomposite electrode on flexible substrate**. *Biosens. Bioelectron.* (2018) **109** 75-82. DOI: 10.1016/j.bios.2018.02.054
16. Yuan Y., Wang Y., Wang H., Hou S.. **Gold nanoparticles decorated on single layer graphene applied for electrochemical ultrasensitive glucose biosensor**. *J. Electroanal. Chem.* (2019) **855** 113495. DOI: 10.1016/j.jelechem.2019.113495
17. Wiorek A., Parrilla M., Cuartero M., Crespo G.A.. **Epidermal Patch with Glucose Biosensor: pH and Temperature Correction toward More Accurate Sweat Analysis during Sport Practice**. *Anal. Chem.* (2020) **92** 10153-10161. DOI: 10.1021/acs.analchem.0c02211
18. Elsherif M., Hassan M.U., Yetisen A.K., Butt H.. **Wearable Contact Lens Biosensors for Continuous Glucose Monitoring Using Smartphones**. *ACS Nano* (2018) **12** 5452-5462. DOI: 10.1021/acsnano.8b00829
19. Kownacka A.E., Vegelyte D., Joosse M., Anton N., Toebes B.J., Lauko J., Buzzacchera I., Lipinska K., Wilson D.A., Geelhoed-Duijvestijn N.. **Clinical Evidence for Use of a Noninvasive Biosensor for Tear Glucose as an Alternative to Painful Finger-Prick for Diabetes Management Utilizing a Biopolymer Coating**. *Biomacromolecules* (2018) **19** 4504-4511. DOI: 10.1021/acs.biomac.8b01429
20. Zou R., Shan S., Huang L., Chen Z., Lawson T., Lin M., Yan L., Liu Y.. **High-Performance Intraocular Biosensors from Chitosan-Functionalized Nitrogen-Containing Graphene for the Detection of Glucose**. *ACS Biomater. Sci. Eng.* (2020) **6** 673-679. DOI: 10.1021/acsbiomaterials.9b01149
21. Sempionatto J.R., Brazaca L.C., García-Carmona L., Bolat G., Campbell A.S., Martin A., Tang G., Shah R., Mishra R.K., Kim J.. **Eyeglasses-based tear biosensing system: Non-invasive detection of alcohol, vitamins and glucose**. *Biosens. Bioelectron.* (2019) **137** 161-170. DOI: 10.1016/j.bios.2019.04.058
22. Liu J., Sun S., Shang H., Lai J., Zhang L.. **Electrochemical Biosensor Based on Bienzyme and Carbon Nanotubes Incorporated into an Os-complex Thin Film for Continuous Glucose Detection in Human Saliva**. *Electroanalysis* (2016) **28** 2016-2021. DOI: 10.1002/elan.201501179
23. Bollella P., Gorton L., Ludwig R., Antiochia R.. **A Third Generation Glucose Biosensor Based on Cellobiose Dehydrogenase Immobilized on a Glassy Carbon Electrode Decorated with Electrodeposited Gold Nanoparticles: Characterization and Application in Human Saliva**. *Sensors* (2017) **17**. DOI: 10.3390/s17081912
24. Arakawa T., Tomoto K., Nitta H., Toma K., Takeuchi S., Sekita T., Minakuchi S., Mitsubayashi K.. **A Wearable Cellulose Acetate-Coated Mouthguard Biosensor for In Vivo Salivary Glucose Measurement**. *Anal. Chem.* (2020) **92** 12201-12207. DOI: 10.1021/acs.analchem.0c01201
25. Yoon H., Xuan X., Jeong S., Park J.Y.. **Wearable, robust, non-enzymatic continuous glucose monitoring system and its in vivo investigation**. *Biosens. Bioelectron.* (2018) **117** 267-275. DOI: 10.1016/j.bios.2018.06.008
26. Chen Y., Lu S., Zhang S., Li Y., Qu Z., Chen Y., Lu B., Wang X., Feng X.. **Skin-like biosensor system via electrochemical channels for noninvasive blood glucose monitoring**. *Sci. Adv.* (2017) **3** e1701629. DOI: 10.1126/sciadv.1701629
27. Kim J., Campbell A.S., de Ávila B.E.-F., Wang J.. **Wearable biosensors for healthcare monitoring**. *Nat. Biotechnol.* (2019) **37** 389-406. DOI: 10.1038/s41587-019-0045-y
28. Halámková L., Halámek J., Bocharova V., Szczupak A., Alfonta L., Katz E.. **Implanted Biofuel Cell Operating in a Living Snail**. *J. Am. Chem. Soc.* (2012) **134** 5040-5043. DOI: 10.1021/ja211714w
29. Katz E., MacVittie K.. **Implanted biofuel cells operating in vivo—methods, applications and perspectives—feature article**. *Energy Environ. Sci.* (2013) **6** 2791-2803. DOI: 10.1039/c3ee42126k
30. MacVittie K., Halámek J., Halámková L., Southcott M., Jemison W.D., Lobel R., Katz E.. **From “cyborg” lobsters to a pacemaker powered by implantable biofuel cells**. *Energy Environ. Sci.* (2013) **6** 81-86. DOI: 10.1039/C2EE23209J
31. Lee D., Jeong S.H., Yun S., Kim S., Sung J., Seo J., Son S., Kim J.T., Susanti L., Jeong Y.. **Totally implantable enzymatic biofuel cell and brain stimulator operating in bird through wireless communication**. *Biosens. Bioelectron.* (2021) **171** 112746. DOI: 10.1016/j.bios.2020.112746
32. Jeerapan I., Ma N.. **Challenges and Opportunities of Carbon Nanomaterials for Biofuel Cells and Supercapacitors: Personalized Energy for Futuristic Self-Sustainable Devices**. *C* (2019) **5**. DOI: 10.3390/c5040062
33. Suzuki N., Lee J., Loew N., Takahashi-Inose Y., Okuda-Shimazaki J., Kojima K., Mori K., Tsugawa W., Sode K.. **Engineered Glucose Oxidase Capable of Quasi-Direct Electron Transfer after a Quick-and-Easy Modification with a Mediator**. *Int. J. Mol. Sci.* (2020) **21**. DOI: 10.3390/ijms21031137
34. Wang J.. **Electrochemical Glucose Biosensors**. *Chem. Rev.* (2008) **108** 814-825. DOI: 10.1021/cr068123a
35. Holland J.T., Lau C., Brozik S., Atanassov P., Banta S.. **Engineering of Glucose Oxidase for Direct Electron Transfer via Site-Specific Gold Nanoparticle Conjugation**. *J. Am. Chem. Soc.* (2011) **133** 19262-19265. DOI: 10.1021/ja2071237
36. delle Noci S., Frasconi M., Favero G., Tosi M., Ferri T., Mazzei F.. **Electrochemical Kinetic Characterization of Redox Mediated Glucose Oxidase Reactions: A Simplified Approach**. *Electroanalysis* (2008) **20** 163-169. DOI: 10.1002/elan.200704030
37. Han C., Li H., Shi R., Zhang T., Tong J., Li J., Li B.. **Organic quinones towards advanced electrochemical energy storage: Recent advances and challenges**. *J. Mater. Chem. A* (2019) **7** 23378-23415. DOI: 10.1039/C9TA05252F
38. Hou C., Lang Q., Liu A.. **Tailoring 1,4-naphthoquinone with electron-withdrawing group: Toward developing redox polymer and FAD-GDH based hydrogel bioanode for efficient electrocatalytic glucose oxidation**. *Electrochim. Acta* (2016) **211** 663-670. DOI: 10.1016/j.electacta.2016.06.078
39. Jin H., Jin Y., Pan J., Mi S., Zeng Q., Li Z., Wang Q., Sun Y., Sheng L.. **Comprehensive identification and hydrophobic analysis of key proteins affecting foam capacity and stability during the evolution of egg white foam**. *Food Hydrocoll.* (2023) **134** 108033. DOI: 10.1016/j.foodhyd.2022.108033
40. Abeyrathne E.D.N.S., Lee H.Y., Ahn D.U.. **Egg white proteins and their potential use in food processing or as nutraceutical and pharmaceutical agents—A review**. *Poult. Sci.* (2013) **92** 3292-3299. DOI: 10.3382/ps.2013-03391
41. Karimpil J.J., Melo J.S., D’Souza S.F.. **Hen egg white as a feeder protein for lipase immobilization**. *J. Mol. Catal. B Enzym.* (2011) **71** 113-118. DOI: 10.1016/j.molcatb.2011.04.006
42. Mahobia S., Bajpai J., Bajpai A.K.. **An in-vitro investigation of swelling controlled delivery of insulin from egg albumin nanocarriers**. *Iran. J. Pharm. Res. IJPR* (2016) **15** 695. PMID: 28243266
43. Yan X., Li H., Hu T., Su X.. **A novel fluorimetric sensing platform for highly sensitive detection of organophosphorus pesticides by using egg white-encapsulated gold nanoclusters**. *Biosens. Bioelectron.* (2017) **91** 232-237. DOI: 10.1016/j.bios.2016.11.058
44. Chang Q., Darabi M.A., Liu Y., He Y., Zhong W., Mequanin K., Li B., Lu F., Xing M.M.Q.. **Hydrogels from natural egg white with extraordinary stretchability, direct-writing 3D printability and self-healing for fabrication of electronic sensors and actuators**. *J. Mater. Chem. A* (2019) **7** 24626-24640. DOI: 10.1039/C9TA06233E
45. Mutee Ur Rehman H.M., Rehman M.M., Saqib M., Ali Khan S., Khan M., Yang Y., Kim S., Rahman S.A., Kim W.-Y.. **Highly Efficient and Wide Range Humidity Response of Biocompatible Egg White Thin Film**. *Nanomaterials* (2021) **11**. DOI: 10.3390/nano11071815
46. Cracknell J.A., Vincent K.A., Armstrong F.A.. **Enzymes as Working or Inspirational Electrocatalysts for Fuel Cells and Electrolysis**. *Chem. Rev.* (2008) **108** 2439-2461. DOI: 10.1021/cr0680639
47. Pazur J.H., Kleppe K.. **The Oxidation of Glucose and Related Compounds by Glucose Oxidase from Aspergillus niger***. *Biochemistry* (1964) **3** 578-583. DOI: 10.1021/bi00892a018
48. Wu H., Zhang Y., Kjøniksen A.-L., Zhou X., Zhou X.. **Wearable Biofuel Cells: Advances from Fabrication to Application**. *Adv. Funct. Mater.* (2021) **31** 2103976. DOI: 10.1002/adfm.202103976
49. Jeerapan I., Sempionatto J.R., Wang J.. **On-Body Bioelectronics: Wearable Biofuel Cells for Bioenergy Harvesting and Self-Powered Biosensing**. *Adv. Funct. Mater.* (2020) **30** 1906243. DOI: 10.1002/adfm.201906243
50. Wu J., Yang H.. **Platinum-based oxygen reduction electrocatalysts**. *Acc. Chem. Res.* (2013) **46** 1848-1857. DOI: 10.1021/ar300359w
|
---
title: Malonyl-CoA Accumulation as a Compensatory Cytoprotective Mechanism in Cardiac
Cells in Response to 7-Ketocholesterol-Induced Growth Retardation
authors:
- Mei-Ling Cheng
- Cheng-Hung Yang
- Pei-Ting Wu
- Yi-Chin Li
- Hao-Wei Sun
- Gigin Lin
- Hung-Yao Ho
journal: International Journal of Molecular Sciences
year: 2023
pmcid: PMC10002498
doi: 10.3390/ijms24054418
license: CC BY 4.0
---
# Malonyl-CoA Accumulation as a Compensatory Cytoprotective Mechanism in Cardiac Cells in Response to 7-Ketocholesterol-Induced Growth Retardation
## Abstract
The major oxidized product of cholesterol, 7-Ketocholesterol (7KCh), causes cellular oxidative damage. In the present study, we investigated the physiological responses of cardiomyocytes to 7KCh. A 7KCh treatment inhibited the growth of cardiac cells and their mitochondrial oxygen consumption. It was accompanied by a compensatory increase in mitochondrial mass and adaptive metabolic remodeling. The application of [U-13C] glucose labeling revealed an increased production of malonyl-CoA but a decreased formation of hydroxymethylglutaryl-coenzyme A (HMG-CoA) in the 7KCh-treated cells. The flux of the tricarboxylic acid (TCA) cycle decreased, while that of anaplerotic reaction increased, suggesting a net conversion of pyruvate to malonyl-CoA. The accumulation of malonyl-CoA inhibited the carnitine palmitoyltransferase-1 (CPT-1) activity, probably accounting for the 7-KCh-induced suppression of β-oxidation. We further examined the physiological roles of malonyl-CoA accumulation. Treatment with the inhibitor of malonyl-CoA decarboxylase, which increased the intracellular malonyl-CoA level, mitigated the growth inhibitory effect of 7KCh, whereas the treatment with the inhibitor of acetyl-CoA carboxylase, which reduced malonyl-CoA content, aggravated such a growth inhibitory effect. Knockout of malonyl-CoA decarboxylase gene (Mlycd−/−) alleviated the growth inhibitory effect of 7KCh. It was accompanied by improvement of the mitochondrial functions. These findings suggest that the formation of malonyl-CoA may represent a compensatory cytoprotective mechanism to sustain the growth of 7KCh-treated cells.
## 1. Introduction
High levels of 7-ketocholesterol (7KCh), an oxysterol derived from the oxidation of cholesterol (Chol), is detected in the vascular plaques of atherosclerosis patients and in the plasma of those at high risk of cardiovascular diseases [1]. It can be catabolized via the intra- and extrahepatic pathways [2,3,4]. The latter pathway involves sterol O-acyltransferase-mediated acylation and the reverse transport of the esterified form to high-density lipoprotein (HDL) [4]. A reduction of the sterol O-acyltransferase expression in heart leads to 7KCh accumulation and tissue damage. Highly elevated 7KCh levels were found in the red blood cells (RBCs) of heart failure patients [5]. It is envisaged that 7KCh can be an important risk factor for cardiovascular diseases and heart failure.
Heart tissue utilizes a number of energy sources for maintaining its continual contraction. The preference for fuel sources varies with developmental stages, shifting from glucose and lactate for fetal hearts to fatty acids for adult hearts. The remodeling of energy metabolism occurs in response to stressful conditions. Different physiological and pathophysiological conditions, including physical exercise, hypoxia, hypertrophy, and heart failure, causes the cardiac cells to switch to glucose for ATP production [6]. Altered energy metabolism in cardiac tissues is postulated to contribute to the pathogenesis of cardiovascular diseases such as cardiomyopathy and heart failure [7,8].
Exposure to 7KCh has an impact on redox homeostasis and cellular metabolism. It induces reactive oxygen species (ROS) generation in the endothelial cells and cardiomyocytes [5,9,10] and inflicts cellular damage. The other oxidized products of cholesterol cause ROS production, which is implicated in cognitive impairment and the development of cataracts [11,12]. Mitochondria represent an important endogenous source of ROS [13], which can damage the mitochondria in a reciprocal manner. Oxysterols and Chol accumulate in the mitochondria in the cardiac tissues of the animal models subject to ischemia reperfusion [14,15]. It is associated with anomalous changes in the mitochondria, such as the peroxidation of membrane lipids and the loss of the mitochondrial membrane potential [15]. Given the central roles of mitochondria in the metabolism, mitochondrial dysfunction is accompanied by the reprogramming of the metabolism in cardiomyocytes. We have recently reported that a 7KCh treatment induced changes in the cholesterol and lipid metabolism of HL-1 cells [16]. *The* genes involved in mevalonic acid biosynthesis and the metabolism of fatty acids, triacylglycerides and ketone bodies were down-regulated, while those involved in cholesterol transport and esterification were up-regulated [16]. Intriguingly, 7KCh enhanced the transcription of the genes regulated by ATF4, which has been recently identified as a key regulator of mitochondrial stress [17]. These findings have prompted us to study whether 7KCh induces mitochondrial stress and reprogramming of the cellular metabolism, in particular, energy metabolism.
In the present study, we demonstrate that 7KCh exposure induces the reprogramming of energy metabolism and mitochondrial dysfunction in cardiomyocytes and inhibits their growth. It is associated with a compensatory increase in the mitochondrial mass and malonyl-CoA accumulation. Malonyl-CoA suppresses carnitine palmitoyltransferase-1 (CPT-1) activity and β-oxidation. Studies involving the use of pharmaceutical inhibitors and Mlycd−/− knockout cells indicated that an increase in malonyl-CoA reverses the 7KCh-induced mitochondrial defects and growth inhibition.
## 2.1. The Growth of Cardiomyocytes Is Inhbited by 7KCh
Decreases in the plasma HDL levels and increases in 7KCh in RBCs correlated with the prevalence of cardiovascular diseases and heart failure [5], implying that 7KCh may affect the physiology of cardiomyocytes. The treatment of cardiomyocytes resulted in the cellular uptake of 7KCh. The immunofluorescence assay with an anti-7KCh antibody was used to detect the localization of 7KCh in the HL-1 cells. The fluorescence of the antibody-stained 7KCh was significantly elevated in the 7KCh-treated HL-1 cells (Figure S1; lower left panel) compared with that of untreated cells (Figure S1; upper left panel), suggesting the uptake and intracellular accumulation of 7KCh.
We examined the effect of 7KCh on the growth of cardiac cell lines HL-1 and AC16. The HL-1 and AC16 cells were treated without or with 10 or 20 μM 7KCh. The concentrations used were in the physiological range of 7KCh. The erythrocytic 7KCh levels in the healthy volunteers were between 1 and 2 μM [5,18], and the blood 7KCh levels in the heart failure patients were at least 10-to 20-fold higher than those of the normal controls [5]. 7KCh inhibited growth of HL-1 and AC16 cells. The numbers of HL-1 and AC16 cells that were treated with 20 μM 7KCh were about $15\%$ lower those of the untreated cells (Figure 1A,B).
## 2.2. Mitochondrial Dysfunction and ROS Formation in Cardiomyocytes Are Induced by 7KCh
Our previous study showed that 7KCh induces oxidative stress in cardiomyocytes [5]. Consistent with this, the HL-1 or AC16 cells were treated with 7KCh, stained with MitoSOX Red, and analyzed cytometrically. The fluorescence of MitoSOX Red was significantly elevated in the treated cells (Figure 2A,J), indicating that 7KCh may cause mitochondrial ROS generation. It is possible that 7KCh may induce the functional alteration of the mitochondria. The effect of 7KCh on the mitochondrial membrane potential ΔΨm was studied. The 7KCh-treated cells were stained with JC-1 dye and analyzed by flow cytometry. The mean fluorescence intensities (MFI) of channels FL2 and FL1 were measured, and their ratio (i.e., FL2 MFI/FL1 MFI), an indicator of ΔΨm, was calculated. The treatment of the HL-1 and AC16 cells with 20 μM 7KCh resulted in $40\%$ decreases in ΔΨm (Figure 2B,K). To further delineate how 7KCH affects the mitochondrial respiration of cardiomyocytes, we resorted to mitochondrial respirometry to study such changes. The oxygen consumption rate (OCR) and extracellular acidification rate (ECAR) of the 7KCh-treated cells were assessed in real-time using a Seahorse XF24 Extracellular Flux Analyzer. The HL-1 cells that were treated without or with 10 or 20 μM 7KCh for 24 h had a reduced basal OCR, but an increased ECAR (Figure 2C–E). Such a shift in energy metabolism toward glycolysis was also observed in the AC16 cells treated with 20 μM 7KCh (Figure 2L–N). The use of respiration inhibitors (oligomycin, FCCP, rotenone, and antimycin A) allowed us to define the OCRs due to maximum respiration, the spare respiratory capacity, and proton leak. Both the maximum respiration and spare respiratory capacity of the HL-1 and AC16 cells diminished in response to 7KCh (Figure 2F,G,O,P). The proton leak was substantially reduced in the 7KCh-treated HL-1 and AC16 cells (Figure 2H,Q). Consistent with the reduction of respiratory function, the ATP levels in the HL-1 and AC16 cells treated with 20 μM 7KCh were about $30\%$ and $35\%$ lower than that of the control cells, respectively (Figure 2I,R). These findings suggest that 7KCh inhibits electron transport and oxidative phosphorylation and induces ROS generation.
## 2.3. Compensatory Biogenesis of Mitochondria in 7KCh-Treated Cardiomyocytes
The deteriorative changes in respiratory function are accompanied by an increase in the mitochondrial mass. As shown in Figure 3A,B, the porin level increased with the 7KCh concentration. In addition, the mitochondrial mass was evaluated by Mitotracker Green staining and cytometric analysis. The mean fluorescence intensity (MFI) of the stained HL-1 cells that had been treated with 20 μM 7KCh was $65\%$ higher than that of the control cells (Figure 3C). These findings suggest that 7KCh causes mitochondrial biogenesis. Consistent with this, the levels of various mitochondrial respiratory chain proteins increased in the 7KCh-treated cells (Figure 3D). The nuclear gene-encoded complex I protein NADH:ubiquinone oxidoreductase subunit B8 (NDUFB8), complex II protein succinate dehydrogenase complex iron sulfur subunit B (SDHB), complex III protein ubiquinol-cytochrome C reductase core protein 2 (UQCRC2), and ATP Synthase F1 Subunit α (ATP5F1A), as well as mitochondrial gene-encoded complex IV protein cytochrome C oxidase I (MTCO1), increased in their expression, albeit to different extents, in the 7KCh-treated cells (Figure 3E). These findings suggest that 7KCh induces compensatory mitochondrial biogenesis in cardiac cells.
## 2.4. Energy Metabolic Profiling and the Changes in Metabolic Fluxes in the 7KCh-Treated Cardiomyocytes
We applied a metabolomic approach for studying the metabolic reprogramming associated with the functional changes in the mitochondria. The metabolites involved in pathways such as glycolysis, pentose phosphate, and the citric acid cycle were analyzed by the liquid-chromatography coupled with tandem mass spectrometry (LC-MS/MS). A twenty-four hour treatment of HL-1 cells with 20 μM 7KCH caused a concentration-dependent change in the metabolome (Figure 4A). Increased levels of glucose-6-phosphate, fructose-1,6-bisphosphate (FBP), and lactate and decreased levels of 2-phosphoglycerate and 2,3-bisphosphoglycerate indicate an increase in the glycolytic rate (Figure 4B,D). Notably, malonyl-CoA accumulated at a high level in these cells. In contrast, the 3-hydroxy-3-methyl-glutaryl-coenzyme A (HMG-CoA) level decreased in the 7KCh-treated cells compared with that of the untreated cells. The accumulation of malonyl-CoA and the reduction of HMG-CoA formation were also observed in the 7KCh-treated AC16 cells (Figure 4E). We further examined the metabolism of malonyl-CoA using [U-13C] glucose labeling and tracking. The HL-1 cells were treated with 20 μM 7KCh and labeled with [U-13C] glucose. The relative abundances of the isotopologues of various metabolites were analyzed. As shown in Figure 4D,E, the M and M+2 isotopologues of malonyl-CoA increased, while the M, M+2, M+4, and M+6 isotopologues of HMG-CoA decreased. Interestingly, the total levels of the TCA cycle intermediates, citrate, succinyl-CoA, and oxaloacetate (OAA), remained unchanged or slightly increased in the cells treated with 10 μM 7KCh. However, these metabolites declined substantially in abundance upon exposure to the treatment with 20 μM 7KCh. An analysis of the isotopologues of the metabolites revealed that in addition to the M+2 and M+4 isotopologues formed from TCA cycle, the M+3 isotopologue of oxaloacetate generated in the anaplerotic reactions (Figure 4D). The proportions of different isotopologues of citrate were nearly unchanged in the cells treated with 20 μM 7KCh. In contrast, the proportions of isotopologues other than that of the M+2 succinyl-CoA decreased substantially. These findings suggest that the flux of the conversion of citrate to the downstream TCA intermediates is inhibited by 7KCh.
## 2.5. Fatty Acid Oxidation in Cardiomyocytes Is Inhibited by 7KCh
The formation of malonyl-CoA has its functional implication. Malonyl-CoA is an inhibitor of CPT-1 involved in the fatty acid uptake into the mitochondria [19,20]. We isolated the mitochondria from the HL-1 cells treated without 7KCh and assayed the CPT-1 activity. The expression levels of CPT-1 were similar in the mitochondrial preparations to those of the control and treated cells (Figure 5A). The CPT-1 activity level was significantly lower in the mitochondria from the 7KCh-treated cells than it was in those from the control cells (Figure 5B). As control, malonyl-CoA directly inhibited the CPT-1 activity of the mitochondria (Figure 5C). These findings suggest that 7KCh inhibits the CPT-1 activity, and probably, β-oxidation in the cardiomyocytes.
To study the possibility that 7KCh inhibits the cardiomyocytic β-oxidation, we studied the fatty acid oxidation (FAO) in the HL-1 cells. The HL-1 cells were treated with or without 7KCh, and the palmitate (PA)-stimulated (or vehicle-stimulated) oxygen consumption was measured using the Seahorse XF24 Extracellular Flux Analyzer. The PA-stimulated oxygen consumption represents the reliance of the cells on β-oxidation for energy. 7KCh inhibited the basal respiration, maximum respiration, and spare respiratory capacity in the cells supplied with palmitate as an energy source (Figure 5D). In contrast, 7KCh did not significantly alter these respiration parameters in the cells the treated with the vehicle. The effect of 7KCh on PA-stimulated oxygen consumption is not attributable to expression of different CPT-1 isoforms. The transcripts of Cpt1a, Cpt1b, and Cpt1c genes, as quantified by RT-qPCR, did not differ between the 7KCh-treated and control cells (Figure 5E). These findings suggest that 7KCh may suppress the cardiomyocytic β-oxidation through the malonyl-CoA-mediated inhibition of CPT-1 activity (Figure 5A).
## 2.6. Inhibition of the Mevalonic Acid (MVA) Pathway Contributes to the Growth Retardation Effect of 7KCh
The reduction of the HMG-CoA level in the 7KCh-treated HL-1 cells suggests that the inhibition of the MVA pathways may retard the growth of cardiac cells. To study this possibility, we treated the HL-1 cells with 0.25, 1, and 10 μM lovastatin, in addition to 20 μM 7KCh, and determined the cell number. Lovastatin is an inhibitor of enzyme 3-hydroxy-3-methylglutaryl-coenzyme A reductase (HMGCR) (Figure 6A) [21]. The treatment with 1 μM lovastatin alone caused a three-fold increase in the HMG-CoA level, and a $40\%$ decrease in the cell number (Figure 6B). The co-treatment with 7KCh and lovastatin reduced the HMG-CoA level by $94\%$. It is consistent with our previous finding that 7KCh represses the expression of Acat2 and *Hmgcs1* genes [16], the products of which act upstream of HMGCR in the MVA pathway. Such a co-treatment caused a $54\%$ decrease in the cell number (Figure 6B). These findings suggest that the products of the MVA pathway, such as coenzyme Q (CoQ), may play essential roles in growth of cardiomyocytes. It is in agreement with a decline in the CoQ10 level in the 7KCh-treated cells (Figure S2).
## 2.7. Malonyl-CoA Production Is Cytoprotective
To study the physiological roles of malonyl-CoA accumulation in the 7KCh-treated cells, we studied the effect of the inhibitors of malonyl-CoA decarboxylase and acetyl-CoA carboxylase on cell growth (Figure 6A). Malonyl-CoA decarboxylase (MLYCD) catalyzes the degradation of malonyl-CoA to acetyl-CoA; acetyl-CoA carboxylase catalyzes the carboxylation of acetyl-CoA to form malonyl-CoA. The HL-1 cells were co-treated with 20 μM 7KCh and 0.5, 2.5, or 10 μM malonyl-CoA decarboxylase inhibitor CBM 301940, and the cell number was determined. The CBM 301940 treatment, which significantly increased the intracellular ratio of malonyl-CoA level to acetyl-CoA, alleviated the growth retardation effect of 7KCh (Figure 6C,F). For the study with acetyl-CoA carboxylase inhibitor, the HL-1 cells were co-treated with 25,100,250 nM ND 646, and the cell number was determined. The ND 646 treatment, which suppressed the intracellular formation of malonyl-CoA, acted synergistically with 7KCh to inhibit cell growth (Figure 6D,G). These findings suggest that the formation of malonyl-CoA is beneficial to the growth of cardiomyocytes and may represent a compensatory cytoprotective response to 7KCh.
To study if malonyl-Co-A accumulation in the cardiomyocytes offers a cytoprotective effect, we derived Mlycd−/− cells from AC16 cells using the CRISPR/Cas9 technology. The expression of the MLYCD protein was nullified in the Mlycd−/− cells (Figure 7A). Mlycd knockout mitigated the growth inhibitory effect of 7KCh (Figure 7B). It was associated with the restoration of ΔΨm and decreased mitochondrial ROS production (Figure 7C,D). The knockout of the *Mlycd* gene maintained the steady state level of malonyl-CoA, while it reduced that of acetyl-CoA (Figure 7F,G). This increased the malonyl-CoA/acetyl-CoA ratio by about $70\%$. The malonyl-CoA/acetyl-CoA ratio was elevated four-fold in the 7KCh-treated Mlycd−/− cells as compared to that of the treated AC16 cells (Figure 7F,G). These findings validate that an increase in malonyl-CoA production can relieve the cardiomyocytes from the 7KCh-induced growth inhibition and mitochondrial dysfunction.
## 2.8. The Expression of the Genes Involved in Malonyl-CoA Metabolism in Cardiomyocytes Is Differentially Modulated by 7KCh
We studied the effect of 7KCh on the expression of the genes encoding acetyl-CoA carboxylase (ACAC) and acyl-CoA synthetase family member 3 (ACSF3). The level of the acetyl-CoA carboxylase gene transcript, as measured by RT-qPCR using ACAC universal primers, was elevated in the 7KCh-treated HL-1 cells (Figure 8A). There are two ACAC isoforms, namely acetyl-CoA carboxylase α (ACACA) and β (ACACB), in mammals [22]. The Acacb transcript level increased, while that of the Acaca transcript declined (Figure 8B). In addition, there was a significant increase in the Acsf3 transcript level (Figure 8D). In contrast, the Mlycd transcript level remained nearly the same after the 7KCh treatment. These findings suggest that 7KCh enhances the expression of Acacb and *Acsf3* genes.
## 3. Discussion
It is possible that the 7KCh-induced disruption of normal energy metabolism and metabolic reprogramming in the cardiomyocytes may play important pathogenic roles in cardiovascular diseases. The way in which 7KCh induces metabolic changes in cardiomyocytes remains elusive. In the present work, we demonstrate that 7KCh causes mitochondrial dysfunction, ROS production, and growth retardation. Malonyl-CoA accrues in the cardiomyocytes in response to 7KCh and inhibits CPT-1 activity and β-oxidation. The studies involving pharmacological inhibitors and Mlycd-knockout cardiomyocytes indicate that an experimental rise in the malonyl-CoA level restores the mitochondrial ΔΨm and redox homeostasis and reverses the growth inhibitory effect of 7KCh.
It is speculative whether 7KCh directly affects the cardiomyocytic mitochondrial functions. The treatment of the translocator protein ligand 4′-chlorodiazepam suppressed the formation of oxysterols and partially restored the mitochondrial respiration [15], implying an adverse effect of 7KCh on the mitochondria. 7KCh damages the mitochondrial DNA and causes a loss of ΔΨm in retinal pigment cells [23]. We found that the 7KCh treatment resulted in reduction of ΔΨm in the HL-1 and AC16 cells, which were treated with 20 μM KCh (Figure 2B,K). It was associated with the decreases in basal respiration, maximum respiration, as well as spare respiratory capacity in these cells (Figure 2E,F,G,N,O,P). Both the flux of mitochondrial metabolism and the efficiency of respiratory processes, but not the expression of mitochondrial proteins, are adversely affected by 7KCh. The expression of the typical respiratory complex proteins actually increased after the 7KCh treatment (Figure 3D,E). The flux of the TCA cycle decreases in the treated cells (Figure 4), contributing to the decline in oxygen consumption. The efficiency of electron transport is lowered by 7KCh, which is indicated by an increase in ROS generation in the treated cells (Figure 2A,J). The proper assembly of the respiratory supercomplex is essential to efficient electron transport. It has been found that the basal and maximum respiration and the spare respiratory capacity are reduced in the PHB-deficient cells, with impaired organization of the respiratory supercomplexes [24,25]. Consistent with such a notion, our previous study has shown that the Phb1, Phb2, Higd1a, and Higd2a transcript levels declined in the 7KCh-treated cells [16]. Prohibin 1 (PHB1) and 2 (PHB2) participate in the assembly of respiratory supercomplexes and the maintenance of their stability [26]. HIGD1A and HIGD2A are involved in the dynamic assembly of complexes III and IV and in supercomplex formation [27]. Moreover, changes in the anabolic pathways may hamper the electron transport. The reduction of the flux of the MVA pathway dwindled the supply of HMG-CoA (Figure 4B,D,E), which is the precursor of CoQ biosynthesis. As CoQs serve as an important electron carrier, their decrease is likely to impair the normal functioning of the electron transport chain (Figure S2).
The 7KCh-induced changes in the mitochondrial functions correlate with the reprogramming of cardiomyocytic energy metabolism. Increases in the levels of glucose-6-phosphate, fructose-1,6-bisphosphate, and lactate (Figure 4A), together with an increase in ECAR (Figure 2), indicate an increase in the glycolytic rate in the 7KCh-treated cardiac cells. Substantial decreases in TCA intermediates, such as citrate, succinyl-CoA, and oxaloacetate, in the cells treated with 20 μM 7KCh are suggestive of a reduction of the TCA cycle flux in these cells (Figure 4D). The isotopologue analysis reveals that the proportions of the M+2 and M+4 isotopologues of succinyl-CoA were similar to those of the corresponding isotopologues of citrate in the control cells. However, the levels of the M+2 and M+4 isotopologues of succinyl-CoA were substantially reduced by the 7KCh treatment. It is likely that the flux of the onward reaction of citrate in TCA cycle decreased upon 7KCh treatment. It also implies that citrate is probably involved in to reactions such as that catalyzed by citrate lyase. Interestingly, most of succinyl-CoA molecules remained unlabeled. These molecules formed from the unlabeled α-ketoglutarate, which may be derived from glutamate in a glutamate dehydrogenase-catalyzed reaction. Additionally, anaplerotic reactions made significant contribution to the oxaloacetate pool in the control and 7KCh-treated cells. Overall, pyruvate derived from glycolysis is converted either to lactate or to oxaloacetate (through anaplerotic reactions). Citrate synthase converts oxaloacetate to citrate, which can be metabolized in the TCA cycle or converted to acetyl-CoA for malonyl-CoA synthesis. In the 7KCh-treated cells, glycolysis is enhanced, while the TCA cycle is inhibited. More citrate molecules contribute to malonyl-CoA synthesis.
Malonyl-CoA itself is an inhibitor of CPT-1, which is involved in the fatty acid uptake into the mitochondria and β-oxidation (Figure 5C) [19]. The accumulation of malonyl-CoA in the 7KCh-treated cells reduced the use of fatty acids as fuel molecules for β-oxidation. Indeed, 7KCh inhibited the CPT-1 activity (Figure 5B) and palmitate-stimulated respiration (Figure 5D). It is evident that malonyl-CoA is cardioprotective. The pharmacological inhibition of MLYCD activity and the knockdown of the *Mlycd* gene improve the biomechanical functions of the post-ischemic heart [28,29]. The 7KCh-induced malonyl-CoA accumulation probably represents a compensatory mechanism to reduce the toxic effect of 7KCh. The pharmacological treatment with CBM 301940, which increased the intracellular malonyl-CoA level, lessened the growth inhibitory effect of 7KCh (Figure 6C,F). The growth of Mlycd−/− cardiac cells were inhibited to a much lesser extent than the control cells were after 7KCh exposure (Figure 7B). Such a cytoprotective effect was associated with the restoration of ΔΨm and a decrease in mitochondrial ROS generation, suggesting an improvement of the mitochondrial functions. These findings advocate that metabolic reprogramming occurs in the 7KCh-treated cardiomyocytes to compensate for mitochondrial dysfunction and growth defects.
An additional point about the malonyl CoA-inhibited β-oxidation is noteworthy. Our previous study has revealed that fatty acids can be released from phosphatidylcholines by phospholipase A2, and are used for esterification [16]. The fatty acid molecules are spared from β-oxidation through the action of malonyl-CoA. Malonyl-CoA appears not to serve as a precursor of fatty acid synthesis. The expression of the genes involved in fatty acid synthesis, such as fatty acid synthase and desaturases, are down-regulated [16].
The shift in energy metabolism from oxidative phosphorylation to glycolysis is physiologically relevant. Fatty acids are the major fuel for healthy adult hearts, responsible for 60–$80\%$ of ATP generation [30]. The remaining ATP molecules are derived from the metabolism of glucose and lactate. A change in fuel preference and energy metabolism may occur under different physiological or pathophysiological conditions. For instance, an increased glycolysis level was observed in the cardiac tissues isolated from the animal models of left ventricular hypertrophy [31,32,33]. The uptake of 2-deoxy-2-[18F] fluoro-D-glucose increased in the right ventricles of pulmonary arterial hypertension patients, and the uptake value correlated with the disease severity scores [34]. Additionally, it has been recently shown that myocardial infarction or pressure loading induces the cycling of specialized cardiomyocytes that may be involved in cardiac repair [35]. The activation of glycolysis is apparently essential to cardiomyocytic proliferation after an injury [36,37]. It is possible that the remodeling of energy metabolism in cardiomyocytes may affect cellular proliferation and tissue regeneration under certain circumstances, for example, the presence of high plasma 7KCh levels [38].
It is not unprecedented that the compensatory mitochondrial biogenesis occurs in response to mitochondrial dysfunction. Knockout of the leucine-rich pentatricopeptide repeat containing protein (LRPPRC)-encoding gene, whose mutations have been identified in Leigh syndrome, causes the defective assembly of the electron transport chain and triggers compensatory mitochondrial biogenesis [39]. Compensatory mitochondrial biogenesis occurs in cells with mitochondrial DNA mutations [40]. The carriers of LHON mutations display increases in the mitochondrial mass [41]. The declines in mitochondrial functions and ATP supply probably induce the expression of respiratory complex proteins and mitochondrial biogenesis in the 7KCh-treated cells (Figure 3). This is consistent with increases in the transcripts of the Tfam, Tfb1m, Tfb2m, and *Pprc1* genes [16], which are involved in mitochondrial transcription, ribosome assembly, and biogenesis [42,43].
The transcription of the genes encoding the malonyl-CoA-metabolizing enzymes is differentially modulated by 7KCh. The Acacb transcription is up-regulated at the expense of the transcription of the *Acaca* gene (Figure 8B,C), resulting in an overall increase in the ACAC-encoding transcript (Figure 8A). The ACACB protein, located at the outer mitochondrial membrane, promotes the formation of malonyl-CoA, which allosterically inhibits CPT-1 activity [22,44]. Such a change in the expression of ACAC isoforms in the 7KCh-treated cells is congruous with the inhibition of β-oxidation. ACSF3 is involved in the detoxification of malonate, which is a competitive inhibitor of succinate dehydrogenase [45]. The ACSF3-derived malonyl-CoA is used in the malonylation of proteins, which may be involved in the post-translational regulation of mitochondrial proteins and metabolism [45]. The knockout of the *Acsf3* gene in HEK 293T cells was found to alter their metabolism [46]. The enhanced transcription of the *Acsf3* gene in the 7KCh-treated cells may imply the role of protein malonylation in the observed metabolic changes.
The present study has limitations, such as cardiac cell lines were used and more work with animal models is needed to fully understand the cytoprotective mechanism. The results might be subject to technical variation caused by the methods used (e.g., cell counting).
## 4.1. Materials
Unless otherwise stated, all the chemicals were purchased from Sigma-Aldrich (St. Louis, MO, USA). We dissolved 7-Ketocholesterol (7KCh; Sigma-Aldrich) in dimethyl sulfoxide (DMSO). In most of the experiments, 7KCh was used at a concentration of 10 or 20 μM. Lovastatin (Caymen Chemical, Ann Arbor, MI, USA), CBM 301940 (Torcis Bioscience, Bio-Techne, Minneapolis, MN, USA), and ND 646 (Caymen Chemical) were dissolved in DMSO. Lovastatin was used at the concentration range 0.25–10 μM; CBM 301940 was used at the concentration range 0.5–10 μM; ND 646 was used at the concentration range 25–250 nM.
## 4.2. Cell Culture and Viability Assay
HL-1 atrial myocytes (Research Resource Identifier (RRID): CVCL_0303) were cultured in the fibronectin-gelatin-coated flasks containing the *Claycomb medium* (51800C, Sigma-Aldrich), which was supplemented with $10\%$ HL-1 qualified fetal bovine serum (FBS; TMS-016, Sigma-Aldrich), 100 U/mL penicillin, 100 μg/mL streptomycin, 2 mM L-glutamine, and 0.1 mM norepinephrine in a humidified atmosphere of $5\%$ CO2 at 37 °C, as previously described [47]. The AC16 human cardiomyocyte cell line (RRID: CVCL_4U18; EMD Millipore Corp., Temecula, CA, USA) was cultured in Dulbecco’s modified Eagle medium/nutrient mixture F-12 medium (DMEM/F-12; Thermo Fisher Scientific, Waltham, MA, USA) containing $12.5\%$ FBS according to manufacturer’s instructions. The malonyl-CoA decarboxylase (Mlycd) gene knockout AC16 cells (i.e., Mlycd−/− cells) were generated using the CRISPR/Cas9 knockout service provided by the RNA Technology Platform and Gene Manipulation Core (National Core Facility for Biopharmaceuticals, Taipei, Taiwan).
For determination of the growth curves of the 7KCh-treated cells, 5 × 104 cells were seeded in 12-well culture plate and treated with the indicated 7KCh concentrations (i.e., the concentrations indicated in the legends of the respective figures) for various periods. Cardiac cells were fixed in $3.7\%$ formaldehyde for 10 min, and then stained with 5 μg/mL of Hoechst 33342 for 15 min. The cell number was determined using IN Cell Analyzer 1000 (GE Healthcare Life Sciences, Chicago, IL, USA) [48]. As a control for the 7KCh treatment, the cells were treated with the dimethyl sulfoxide (DMSO) vehicle. DMSO also served as the vehicle for lovastatin, CBM 301940, and ND 646.
## 4.3.1. Oxygen Consumption Rate (OCR) and Extracellular Acidification Rate (ECAR) Determination and Mitochondrial Function Tests
The measurement of the OCR and ECAR was performed using Seahorse XF24-3 Analyzer (Agilent Technologies, Santa Clara, CA, USA) as previously described [49]. In brief, 8 × 103 cells were seeded per well in an XF24 cell culture microplate and maintained in DMEM (Thermo Fisher Scientific). They were treated with the indicated concentrations of 7KCh. After 24 h, the medium was replaced with DMEM without sodium bicarbonate. The XF24 assay cartridge was prepared, loaded with 10 μM oligomycin, 3 μM FCCP, 10 μM antimycin A, and rotenone in the injection ports, and calibrated according to the manufacturer’s instruction. The oxygen consumption of the cells was measured using Seahorse XF-24 analyzer (Agilent Technologies) under basal condition and after the sequential injection of oligomycin (32 min), carbonyl cyanide 4-(trifluoromethoxy)phenylhydrazone (FCCP) (56 min), and antimycin A/rotenone (88 min). The proton flux in the medium was also monitored. The OCR and ECAR were automatically calculated.
Superoxide anion production was determined by MitoSOX Red staining and flow cytometry as previously described [50]. The mitochondrial mass and mitochondrial membrane potential were determined, respectively, by staining with MitoTracker Green and JC-1 and performing cytometric analyses as previously described [51,52]. BD LSR II flow cytometer (Becton Dickinson, Franklin Lakes, NJ, USA) was used for the analyses.
## 4.3.2. Fatty Acid Oxidation (FAO) Assay
At 24 h before the experiment, 8 × 103 cells were cultured per well in an XF-24 plate. The cells were treated with the indicated concentrations of 7KCh for 24 h. The medium was replaced with 375 μL of FAO Assay Medium (111 mM NaCl, 4.7 mM KCl, 1.25 mM CaCl2, 2 mM MgSO4, 1.2 mM NaH2PO4 supplemented with 2.5 mM glucose, 0.5 mM carnitine, and 5 mM HEPES, pH 7.4) and incubated at 37 °C for 45 min. Before the analysis, 87.5 μL of 1 mM palmitate–$1\%$ BSA conjugate or $1\%$ BSA were added to each well. During the experiment, respiration was measured under basal condition and after the sequential injection of oligomycin, FCCP, and antimycin A/rotenone as described in Section 4.3.1.
## 4.4. Metabolite Analysis Using Liquid-Chromatography Coupled with Tandem Mass Spectrometry (LC-MS/MS)
The metabolites were extracted as previously described [53,54]. Briefly, the cells were treated with the indicated concentrations of 7KCh for 24 h, and then washed twice with cold phosphate-buffered saline (PBS). The metabolites were extracted with $80\%$ (v/v) MeOH/H2O pre-equilibrated at 80 °C. The extract was collected into 1.5 mL tubes, vortexed for 5 min, and centrifuged at 12,000× g for 30 min. The resulting supernatant was dried using a centrifugal evaporator under reduced pressure. The samples were re-suspended in 200 μL of $1\%$ acetic acid for mass spectrometric analysis, and then were analyzed using the Xevo TQ-XS Triple Quadrupole Mass Spectrometry System (Waters Corp., Milford, MA, USA) [16]. The chromatographic separation was achieved on a BEH C18 (100 × 2.1 mm, particle size of 1.7 µm; Waters Corp.) at 45 °C. The mobile phase consisted of eluent A (10 mM tributylamine (TBA)/15 mM acetic acid) and eluent B (10 mM TBA/15 mM acetic acid/$50\%$ ACN). The flow rate was set at 0.3 mL/min. The elution profile was as follows: $4\%$ B, 6 min; linear gradient 4–$50\%$ B, 0.1 min; 50–$60\%$ B, 2.9 min; 60–$100\%$ B, 0.8 min, and $100\%$ B for an additional 2.2 min. The mass spectrometer was operated in negative ion mode at an ESI voltage of 3 kV. The metabolites were analyzed using MassLynx software (v4.1; Waters Corp., Milford, MA, USA).
## 4.5. [U-13C] Glucose Labeling and Isotopologue Analysis
Stable isotope-labeling was performed as previously described with slight modifications [55]. The cells were treated with or without 7KCh for 24 h, and then incubated in the medium containing 0.5 mM glucose (i.e., low-glucose medium) for 0.5 h. [U-13C] Glucose was added to a final concentration of 20 mM and the incubation continued for 1 h. Metabolites were extracted in the ice-cold $80\%$ methanol and analyzed using the Vion IMS QTOF Acquity UPLC system (Waters Corp.). Chromatographic separation was achieved on a BEH C18 (100 × 2.1 mm, particle size of 1.7 um; Waters Corp.) at 45 °C. The mobile phase consisted of eluent A (10 mM TBA/15 mM acetic acid) and eluent B (10 mM TBA/15 mM acetic acid/$50\%$ ACN). The flow rate was set at 0.3 mL/min. The elution profile was the same as that which is described in the preceding section. The mass spectrometer was operated in negative ion mode at an ESI voltage of 2.5 kV. Metabolites were analyzed using UNIFI software (v1.0.6171; Waters Corp.).
## 4.6. Isolation of Mitochondria and CPT-1 Activity Assay
Isolation of the mitochondria was performed by a modification of previously described method [56]. Around 1.5 × 106 HL-1 cells were harvested and resuspended in ice-cold mitochondria isolation buffer (20 mM HEPES, 5 mM KH2PO4, 50 μM MgCl2, 250 mM sucrose, and $0.2\%$ BSA, pH 7.5). The sample was homogenized with 20 passages using a 27 gauge needle. The homogenate was centrifuged at 500× g for 10 min, and the supernatant was retained. The pellet was washed with the mitochondria isolation buffer, and then subjected to centrifugation. The supernatant fractions were combined and centrifuged at 10,000× g for 30 min at 4 °C to obtain a crude mitochondrial pellet.
The CPT-1 activity was measured as described elsewhere with modifications [57]. The enriched mitochondria were suspended in 80 μL of mitochondria isolation buffer and incubated at 37 °C for 2 min. The enzymatic reaction was initiated by addition of 10 μL 1 mM palmitoyl-CoA and 10 μL 10 mM carnitine and incubated at 37 °C for 5 min. The product palmitoylcarnitine was analyzed using LC-MS under conditions that had been described previously [57].
## 4.7. Western Blotting and Immunofluorescence
The cells were rinsed with cold PBS, scraped, and collected for centrifugation. They were immediately lysed in a lysis buffer (20 mM Tris·HCl (pH 8), $1\%$ Triton X-100, 137 mM NaCl, 1.5 mM MgCl2, $10\%$ glycerol, 1 mM EGTA, 50 mM NaF, 1 mM Na3VO4, 10 mM β-glycerophosphate, 1 mM PMSF, 1 μg/mL leupeptin, and1 μg/mL aprotinin). The protein concentration of the lysate was determined using the Bradford method. The sample was analyzed by SDS-PAGE and immunoblotting with primary antibodies (including anti-VDAC1/porin antibody (ab14734; Abcam, Cambridge, UK), anti-actin (clone AC-40; Sigma-Aldrich), total OXPHOS rodent WB antibody cocktail (ab110413 (MS-604), Abcam), anti-CPT1A antibody (15184-1-AP; Proteintech group Inc., Rosemont, IL, USA), and anti-GAPDH antibody (GTX100118; GeneTex Inc., Irvine, CA, USA)), and appropriate secondary antibodies (including the horseradish-peroxidase (HRP)-conjugated goat anti-mouse antibody (sc-2005; Santa Cruz, Dallas, TX, USA), and the HRP-conjugated mouse anti-rabbit antibody (SC-2357; Santa Cruz Biotechnology, Dallas, TX, USA)).
For immunofluorescence staining, the cells were cultured in a 35 mm glass-bottomed culture dish (Mattek Life Sciences, Ashland, MA, USA), and after the 7KCh treatment, the cells were fixed in $4\%$ paraformaldehyde/PBS for 1 h. The fixed cells were rinsed with PBS and stained with 1 μg/mL anti-7-ketochosterol antibody (MKC-100n; Japan Institute for the Control of Aging (JaICA) Nikken Seil Co., Ltd., Shizuoka, Japan) in PBS/$0.1\%$ Triton X-100/$1\%$ bovine serum albumin (BSA) at 4 °C overnight. After rinsing it with PBS, the sample was stained with 2 μg/mL anti-mouse DyLight 488-conjugated secondary antibody (Thermo Fisher Scientific) and counterstained with Hoechst 33342 at room temperature for 2 h. The sample was examined under a Zeiss LSM780 fluorescence microscope (Carl Zeiss Microscopy, Oberkochen, Germany).
## 4.8. Reverse Transcription-Quantitative Polymerase Chain Reaction (RT-qPCR)
The HL-1 cells were treated with 0, 10, 20, or 50 μM 7KCh for 24 h. They were washed with PBS, and subsequently, lysed with TRIzol reagent (Life Technologies, Carlsbad, CA, USA) for total RNA isolation. The total RNA was quantified using NanoDrop (Implen, Munich, Germany). The cDNA was synthesized from the total RNA using a RevertAid first-strand cDNA synthesis kit (catalog no. K1622; Thermo Fisher Scientific) according to the manufacturer’s protocol. The cDNA sample was mixed with the SsoFast EvaGreen supermix (catalog no. 172-5201; Bio-Rad, Hercules, CA, USA) in a reaction volume of 10 μL. PCR was performed with the CFX96 Touch real-time PCR detection system (Bio-Rad, Hercules, CA, USA). The relative expression of mRNAs was calculated using the comparative CT method and normalized to GAPDH expression. The primer sequences are listed in Table S1.
## 4.9. Statistical Analyses
The number of independent experiments (each with a triplicate set of samples) is stated or given as the N value in the respective figure legend. Data are mean ± SD. All the statistical analyses were performed with IBM SPSS 20.0 (IBM, Armonk, NY, USA). Two-way analysis of variance (ANOVA) with Sidak’s multiple comparison test, Mann–Whitney test, Kruskal–Wallis test with Dunn’s multiple comparison test, and Student’s t test were used where appropriate. A p value of < 0.05 is considered to be significant.
## References
1. Brown A.J., Jessup W.. **Oxysterols and atherosclerosis**. *Atherosclerosis* (1999) **142** 1-28. DOI: 10.1016/S0021-9150(98)00196-8
2. Schweizer R.A., Zurcher M., Balazs Z., Dick B., Odermatt A.. **Rapid hepatic metabolism of 7-ketocholesterol by 11beta-hydroxysteroid dehydrogenase type 1: Species-specific differences between the rat, human, and hamster enzyme**. *J. Biol. Chem.* (2004) **279** 18415-18424. DOI: 10.1074/jbc.M313615200
3. Lyons M.A., Samman S., Gatto L., Brown A.J.. **Rapid hepatic metabolism of 7-ketocholesterol in vivo: Implications for dietary oxysterols**. *J. Lipid Res.* (1999) **40** 1846-1857. DOI: 10.1016/S0022-2275(20)34901-4
4. Lee J.W., Huang J.D., Rodriguez I.R.. **Extra-hepatic metabolism of 7-ketocholesterol occurs by esterification to fatty acids via cPLA2alpha and SOAT1 followed by selective efflux to HDL**. *Biochim. Biophys. Acta* (2015) **1851** 605-619. DOI: 10.1016/j.bbalip.2015.01.007
5. Tang H.Y., Wang C.H., Ho H.Y., Wu P.T., Hung C.L., Huang C.Y., Wu P.R., Yeh Y.H., Cheng M.L.. **Lipidomics reveals accumulation of the oxidized cholesterol in erythrocytes of heart failure patients**. *Redox Biol.* (2018) **14** 499-508. DOI: 10.1016/j.redox.2017.10.020
6. Kolwicz S.C., Tian R.. **Glucose metabolism and cardiac hypertrophy**. *Cardiovasc. Res.* (2011) **90** 194-201. DOI: 10.1093/cvr/cvr071
7. Lopaschuk G.D., Karwi Q.G., Tian R., Wende A.R., Abel E.D.. **Cardiac Energy Metabolism in Heart Failure**. *Circ. Res.* (2021) **128** 1487-1513. DOI: 10.1161/CIRCRESAHA.121.318241
8. Pulinilkunnil T., Kienesberger P., Nagendran J.. **Editorial: Novel Concepts in Cardiac Energy Metabolism: From Biology to Disease**. *Front. Cardiovasc. Med.* (2019) **6** 97. DOI: 10.3389/fcvm.2019.00097
9. Luchetti F., Canonico B., Cesarini E., Betti M., Galluzzi L., Galli L., Tippins J., Zerbinati C., Papa S., Iuliano L.. **7-Ketocholesterol and 5,6-secosterol induce human endothelial cell dysfunction by differential mechanisms**. *Steroids* (2015) **99** 204-211. DOI: 10.1016/j.steroids.2015.02.008
10. Kahn E., Baarine M., Pelloux S., Riedinger J.M., Frouin F., Tourneur Y., Lizard G.. **Iron nanoparticles increase 7-ketocholesterol-induced cell death, inflammation, and oxidation on murine cardiac HL1-NB cells**. *Int. J. Nanomed.* (2010) **5** 185-195. DOI: 10.2147/IJN.S8458
11. Lutjohann D., Papassotiropoulos A., Bjorkhem I., Locatelli S., Bagli M., Oehring R.D., Schlegel U., Jessen F., Rao M.L., von Bergmann K.. **Plasma 24S-hydroxycholesterol (cerebrosterol) is increased in Alzheimer and vascular demented patients**. *J. Lipid Res.* (2000) **41** 195-198. DOI: 10.1016/S0022-2275(20)32052-6
12. Girao H., Mota M.C., Ramalho J., Pereira P.. **Cholesterol oxides accumulate in human cataracts**. *Exp. Eye Res.* (1998) **66** 645-652. DOI: 10.1006/exer.1998.0465
13. Zorov D.B., Juhaszova M., Sollott S.J.. **Mitochondrial reactive oxygen species (ROS) and ROS-induced ROS release**. *Physiol. Rev.* (2014) **94** 909-950. DOI: 10.1152/physrev.00026.2013
14. Musman J., Pons S., Barau C., Caccia C., Leoni V., Berdeaux A., Ghaleh B., Morin D.. **Regular treadmill exercise inhibits mitochondrial accumulation of cholesterol and oxysterols during myocardial ischemia-reperfusion in wild-type and ob/ob mice**. *Free Radic. Biol. Med.* (2016) **101** 317-324. DOI: 10.1016/j.freeradbiomed.2016.10.496
15. Paradis S., Leoni V., Caccia C., Berdeaux A., Morin D.. **Cardioprotection by the TSPO ligand 4′-chlorodiazepam is associated with inhibition of mitochondrial accumulation of cholesterol at reperfusion**. *Cardiovasc. Res.* (2013) **98** 420-427. DOI: 10.1093/cvr/cvt079
16. Cheng M.L., Tang H.Y., Wu P.T., Yang C.H., Lo C.J., Lin J.F., Ho H.Y.. **7-Ketocholesterol Induces Lipid Metabolic Reprogramming and Enhances Cholesterol Ester Accumulation in Cardiac Cells**. *Cells* (2021) **10**. DOI: 10.3390/cells10123597
17. Quiros P.M., Prado M.A., Zamboni N., D’Amico D., Williams R.W., Finley D., Gygi S.P., Auwerx J.. **Multi-omics analysis identifies ATF4 as a key regulator of the mitochondrial stress response in mammals**. *J. Cell Biol.* (2017) **216** 2027-2045. DOI: 10.1083/jcb.201702058
18. Pataj Z., Liebisch G., Schmitz G., Matysik S.. **Quantification of oxysterols in human plasma and red blood cells by liquid chromatography high-resolution tandem mass spectrometry**. *J. Chromatogr. A* (2016) **1439** 82-88. DOI: 10.1016/j.chroma.2015.11.015
19. Folmes C.D., Lopaschuk G.D.. **Role of malonyl-CoA in heart disease and the hypothalamic control of obesity**. *Cardiovasc. Res.* (2007) **73** 278-287. DOI: 10.1016/j.cardiores.2006.10.008
20. McGarry J.D., Brown N.F.. **The mitochondrial carnitine palmitoyltransferase system. From concept to molecular analysis**. *Eur. J. Biochem.* (1997) **244** 1-14. DOI: 10.1111/j.1432-1033.1997.00001.x
21. Alberts A.W.. **Discovery, biochemistry and biology of lovastatin**. *Am. J. Cardiol.* (1988) **62** 10J-15J. DOI: 10.1016/0002-9149(88)90002-1
22. Wang Y., Yu W., Li S., Guo D., He J., Wang Y.. **Acetyl-CoA Carboxylases and Diseases**. *Front. Oncol.* (2022) **12** 836058. DOI: 10.3389/fonc.2022.836058
23. Gramajo A.L., Zacharias L.C., Neekhra A., Luthra S., Atilano S.R., Chwa M., Brown D.J., Kuppermann B.D., Kenney M.C.. **Mitochondrial DNA damage induced by 7-ketocholesterol in human retinal pigment epithelial cells in vitro**. *Investig. Ophthalmol. Vis. Sci.* (2010) **51** 1164-1170. DOI: 10.1167/iovs.09-3443
24. Anderson C.J., Kahl A., Qian L., Stepanova A., Starkov A., Manfredi G., Iadecola C., Zhou P.. **Prohibitin is a positive modulator of mitochondrial function in PC12 cells under oxidative stress**. *J. Neurochem.* (2018) **146** 235-250. DOI: 10.1111/jnc.14472
25. Jian C., Xu F., Hou T., Sun T., Li J., Cheng H., Wang X.. **Deficiency of PHB complex impairs respiratory supercomplex formation and activates mitochondrial flashes**. *J. Cell Sci.* (2017) **130** 2620-2630. DOI: 10.1242/jcs.198523
26. Vartak R., Porras C.A., Bai Y.. **Respiratory supercomplexes: Structure, function and assembly**. *Protein Cell* (2013) **4** 582-590. DOI: 10.1007/s13238-013-3032-y
27. Timon-Gomez A., Garlich J., Stuart R.A., Ugalde C., Barrientos A.. **Distinct Roles of Mitochondrial HIGD1A and HIGD2A in Respiratory Complex and Supercomplex Biogenesis**. *Cell Rep.* (2020) **31** 107607. DOI: 10.1016/j.celrep.2020.107607
28. Dyck J.R., Cheng J.F., Stanley W.C., Barr R., Chandler M.P., Brown S., Wallace D., Arrhenius T., Harmon C., Yang G.. **Malonyl coenzyme a decarboxylase inhibition protects the ischemic heart by inhibiting fatty acid oxidation and stimulating glucose oxidation**. *Circ. Res.* (2004) **94** e78-e84. DOI: 10.1161/01.RES.0000129255.19569.8f
29. Dyck J.R., Hopkins T.A., Bonnet S., Michelakis E.D., Young M.E., Watanabe M., Kawase Y., Jishage K., Lopaschuk G.D.. **Absence of malonyl coenzyme A decarboxylase in mice increases cardiac glucose oxidation and protects the heart from ischemic injury**. *Circulation* (2006) **114** 1721-1728. DOI: 10.1161/CIRCULATIONAHA.106.642009
30. Brown D.A., Perry J.B., Allen M.E., Sabbah H.N., Stauffer B.L., Shaikh S.R., Cleland J.G., Colucci W.S., Butler J., Voors A.A.. **Expert consensus document: Mitochondrial function as a therapeutic target in heart failure**. *Nat. Rev. Cardiol.* (2017) **14** 238-250. DOI: 10.1038/nrcardio.2016.203
31. Kato T., Niizuma S., Inuzuka Y., Kawashima T., Okuda J., Tamaki Y., Iwanaga Y., Narazaki M., Matsuda T., Soga T.. **Analysis of metabolic remodeling in compensated left ventricular hypertrophy and heart failure**. *Circ. Heart Fail.* (2010) **3** 420-430. DOI: 10.1161/CIRCHEARTFAILURE.109.888479
32. Nascimben L., Ingwall J.S., Lorell B.H., Pinz I., Schultz V., Tornheim K., Tian R.. **Mechanisms for increased glycolysis in the hypertrophied rat heart**. *Hypertension* (2004) **44** 662-667. DOI: 10.1161/01.HYP.0000144292.69599.0c
33. Fillmore N., Levasseur J.L., Fukushima A., Wagg C.S., Wang W., Dyck J.R.B., Lopaschuk G.D.. **Uncoupling of glycolysis from glucose oxidation accompanies the development of heart failure with preserved ejection fraction**. *Mol. Med.* (2018) **24** 3. DOI: 10.1186/s10020-018-0005-x
34. Lundgrin E.L., Park M.M., Sharp J., Tang W.H., Thomas J.D., Asosingh K., Comhair S.A., DiFilippo F.P., Neumann D.R., Davis L.. **Fasting 2-deoxy-2-[18F]fluoro-D-glucose positron emission tomography to detect metabolic changes in pulmonary arterial hypertension hearts over 1 year**. *Ann. Am. Thorac. Soc.* (2013) **10** 1-9. DOI: 10.1513/AnnalsATS.201206-029OC
35. Liu X., Pu W., He L., Li Y., Zhao H., Li Y., Liu K., Huang X., Weng W., Wang Q.D.. **Cell proliferation fate mapping reveals regional cardiomyocyte cell-cycle activity in subendocardial muscle of left ventricle**. *Nat. Commun.* (2021) **12** 5784. DOI: 10.1038/s41467-021-25933-5
36. Fukuda R., Marin-Juez R., El-Sammak H., Beisaw A., Ramadass R., Kuenne C., Guenther S., Konzer A., Bhagwat A.M., Graumann J.. **Stimulation of glycolysis promotes cardiomyocyte proliferation after injury in adult zebrafish**. *EMBO Rep.* (2020) **21** e49752. DOI: 10.15252/embr.201949752
37. Mohamed T.M.A., Abouleisa R., Hill B.G.. **Metabolic Determinants of Cardiomyocyte Proliferation**. *Stem Cells* (2022) **40** 458-467. DOI: 10.1093/stmcls/sxac016
38. Uchikawa T., Matoba T., Kawahara T., Baba I., Katsuki S., Koga J.I., Hashimoto Y., Yamasaki R., Ichi I., Akita H.. **Dietary 7-ketocholesterol exacerbates myocardial ischemia-reperfusion injury in mice through monocyte/macrophage-mediated inflammation**. *Sci. Rep.* (2022) **12** 14902. DOI: 10.1038/s41598-022-19065-z
39. Cuillerier A., Ruiz M., Daneault C., Forest A., Rossi J., Vasam G., Cairns G., Cadete V., Consortium L., Des Rosiers C.. **Adaptive optimization of the OXPHOS assembly line partially compensates lrpprc-dependent mitochondrial translation defects in mice**. *Commun. Biol.* (2021) **4** 989. DOI: 10.1038/s42003-021-02492-5
40. Wu Y.T., Wu S.B., Wei Y.H.. **Metabolic reprogramming of human cells in response to oxidative stress: Implications in the pathophysiology and therapy of mitochondrial diseases**. *Curr. Pharm. Des.* (2014) **20** 5510-5526. DOI: 10.2174/1381612820666140306103401
41. Giordano C., Iommarini L., Giordano L., Maresca A., Pisano A., Valentino M.L., Caporali L., Liguori R., Deceglie S., Roberti M.. **Efficient mitochondrial biogenesis drives incomplete penetrance in Leber’s hereditary optic neuropathy**. *Brain* (2014) **137** 335-353. DOI: 10.1093/brain/awt343
42. Scarpulla R.C., Vega R.B., Kelly D.P.. **Transcriptional integration of mitochondrial biogenesis**. *Trends Endocrinol. Metab.* (2012) **23** 459-466. DOI: 10.1016/j.tem.2012.06.006
43. Vercauteren K., Gleyzer N., Scarpulla R.C.. **Short hairpin RNA-mediated silencing of PRC (PGC-1-related coactivator) results in a severe respiratory chain deficiency associated with the proliferation of aberrant mitochondria**. *J. Biol. Chem.* (2009) **284** 2307-2319. DOI: 10.1074/jbc.M806434200
44. Abu-Elheiga L., Matzuk M.M., Abo-Hashema K.A., Wakil S.J.. **Continuous fatty acid oxidation and reduced fat storage in mice lacking acetyl-CoA carboxylase 2**. *Science* (2001) **291** 2613-2616. DOI: 10.1126/science.1056843
45. Bowman C.E., Wolfgang M.J.. **Role of the malonyl-CoA synthetase ACSF3 in mitochondrial metabolism**. *Adv. Biol. Regul.* (2019) **71** 34-40. DOI: 10.1016/j.jbior.2018.09.002
46. Bowman C.E., Rodriguez S., Selen Alpergin E.S., Acoba M.G., Zhao L., Hartung T., Claypool S.M., Watkins P.A., Wolfgang M.J.. **The Mammalian Malonyl-CoA Synthetase ACSF3 Is Required for Mitochondrial Protein Malonylation and Metabolic Efficiency**. *Cell Chem. Biol.* (2017) **24** 673-684. DOI: 10.1016/j.chembiol.2017.04.009
47. Claycomb W.C., Lanson N.A., Stallworth B.S., Egeland D.B., Delcarpio J.B., Bahinski A., Izzo N.J.. **HL-1 cells: A cardiac muscle cell line that contracts and retains phenotypic characteristics of the adult cardiomyocyte**. *Proc. Natl. Acad. Sci. USA* (1998) **95** 2979-2984. DOI: 10.1073/pnas.95.6.2979
48. Ho H.Y., Lin Y.T., Lin G., Wu P.R., Cheng M.L.. **Nicotinamide nucleotide transhydrogenase (NNT) deficiency dysregulates mitochondrial retrograde signaling and impedes proliferation**. *Redox Biol.* (2017) **12** 916-928. DOI: 10.1016/j.redox.2017.04.035
49. Zhang J., Nuebel E., Wisidagama D.R., Setoguchi K., Hong J.S., Van Horn C.M., Imam S.S., Vergnes L., Malone C.S., Koehler C.M.. **Measuring energy metabolism in cultured cells, including human pluripotent stem cells and differentiated cells**. *Nat. Protoc.* (2012) **7** 1068-1085. DOI: 10.1038/nprot.2012.048
50. Cheng M.L., Wu C.H., Chien K.Y., Lai C.H., Li G.J., Liu Y.Y., Lin G., Ho H.Y.. **Enteroviral 2B Interacts with VDAC3 to Regulate Reactive Oxygen Species Generation That Is Essential to Viral Replication**. *Viruses* (2022) **14**. DOI: 10.3390/v14081717
51. Cheng M.L., Weng S.F., Kuo C.H., Ho H.Y.. **Enterovirus 71 induces mitochondrial reactive oxygen species generation that is required for efficient replication**. *PLoS ONE* (2014) **9**. DOI: 10.1371/journal.pone.0113234
52. Ho H.Y., Cheng M.L., Cheng P.F., Chiu D.T.. **Low oxygen tension alleviates oxidative damage and delays cellular senescence in G6PD-deficient cells**. *Free Radic. Res.* (2007) **41** 571-579. DOI: 10.1080/10715760601184819
53. Ho H.Y., Cheng M.L., Shiao M.S., Chiu D.T.. **Characterization of global metabolic responses of glucose-6-phosphate dehydrogenase-deficient hepatoma cells to diamide-induced oxidative stress**. *Free Radic. Biol. Med.* (2013) **54** 71-84. DOI: 10.1016/j.freeradbiomed.2012.10.557
54. Cheng M.L., Shiao M.S., Chiu D.T., Weng S.F., Tang H.Y., Ho H.Y.. **Biochemical disorders associated with antiproliferative effect of dehydroepiandrosterone in hepatoma cells as revealed by LC-based metabolomics**. *Biochem. Pharmacol.* (2011) **82** 1549-1561. DOI: 10.1016/j.bcp.2011.07.104
55. Ma H., Zhang J., Zhou L., Wen S., Tang H.Y., Jiang B., Zhang F., Suleman M., Sun D., Chen A.. **c-Src Promotes Tumorigenesis and Tumor Progression by Activating PFKFB3**. *Cell Rep.* (2020) **30** 4235-4249. DOI: 10.1016/j.celrep.2020.03.005
56. Setoyama D., Fujimura Y., Miura D.. **Metabolomics reveals that carnitine palmitoyltransferase-1 is a novel target for oxidative inactivation in human cells**. *Genes Cells* (2013) **18** 1107-1119. DOI: 10.1111/gtc.12098
57. Dai J., Liang K., Zhao S., Jia W., Liu Y., Wu H., Lv J., Cao C., Chen T., Zhuang S.. **Chemoproteomics reveals baicalin activates hepatic CPT1 to ameliorate diet-induced obesity and hepatic steatosis**. *Proc. Natl. Acad. Sci. USA* (2018) **115** E5896-E5905. DOI: 10.1073/pnas.1801745115
|
---
title: Untargeted Lipidomics of Erythrocytes under Simulated Microgravity Conditions
authors:
- Cristina Manis
- Antonio Murgia
- Alessia Manca
- Antonella Pantaleo
- Giacomo Cao
- Pierluigi Caboni
journal: International Journal of Molecular Sciences
year: 2023
pmcid: PMC10002504
doi: 10.3390/ijms24054379
license: CC BY 4.0
---
# Untargeted Lipidomics of Erythrocytes under Simulated Microgravity Conditions
## Abstract
Lipidomics and metabolomics are nowadays widely used to provide promising insights into the pathophysiology of cellular stress disorders. Our study expands, with the use of a hyphenated ion mobility mass spectrometric platform, the understanding of the cellular processes and stress due to microgravity. By lipid profiling of human erythrocytes, we annotated complex lipids such as oxidized phosphocholines, phosphocholines bearing arachidonic in their moiety, as well as sphingomyelins and hexosyl ceramides associated with microgravity conditions. Overall, our findings give an insight into the molecular alterations and identify erythrocyte lipidomics signatures associated with microgravity conditions. If the present results are confirmed in future studies, they may help to develop suitable treatments for astronauts after return to Earth.
## 1. Introduction
The term microgravity in general refers to existing residual accelerations. When gravitation is the only force acting on an object, then it results is in free fall and hence it will experience microgravity [1]. Weightlessness is the state in which a body having a certain weight is balanced by another force or remains in free fall without feeling the effects of the atmosphere, equivalent to the situation faced by an astronaut aboard a spaceship. The effects of microgravity on human physiology have been studied extensively since the time of Yuri Gagarin (in 1961) who experienced the first man-on-board orbital flight, revealing profound implications for human health [2]. Despite the great interest and commitment of the scientific community, the mechanisms by which microgravity exerts its effects on the human body are not entirely clear.
Acute changes in normal physiology are typically seen in astronauts as a response and adaptation to abnormal environments. Such peculiar alterations require the attention of doctors and scientists [3]. In addition to alterations at the genetic level [4], microgravity experienced by space travellers also induces profound alterations at the cellular level. These alterations occurring at the cellular level are reflected in a series of pathological conditions such as a reduction of bone density, muscle atrophy, endocrine disorders, cognitive disorders and cardiovascular disfunctions, body fluid and electrolyte reduction, motion sickness, immune inhibition, and anaemia [5]. For these reasons, ground-based experiments simulating factors of spaceflight conditions are needed.
Microgravity is studied in several scientific and technological fields with the aim to highlight processes that on Earth are masked by the effects of the high gravitational field. Furthermore, the study of physiological processes in microgravity conditions allows the identification of the molecular mechanisms involved in different pathologies [6].
Roughly 350 people have experienced spaceflight in the past four decades, making it difficult to develop higher levels of clinical evidence to evaluate the effectiveness of space medicine interventions [7]. This great limitation, and the importance of studying the alterations at the cellular level that affect astronauts has led various research groups to study and build instruments capable of simulating space gravitational conditions on Earth.
The most employed methods to simulate microgravity are random positioning machines (RPM) and clinostats [8]. By controlled simultaneous rotating of the two axes, the clinostat cancels the cumulative gravity vector at the centre of the device, producing an environment with an average of 10−3 g. This is accomplished by the rotation of a chamber at the centre of the device to disperse the gravity vector uniformly within a spherical volume at a constant angular speed [9].
Bioactive lipid molecules known as signalling molecules, such as fatty acid, eicosanoids, diacylglycerol, phosphatidic acid, lysophosphatidic acid, ceramide, sphingosine, sphingosine-1-phosphate, phosphatidylinositol-3 phosphate, and cholesterol, are involved in the activation or regulation of different signalling pathways leading to apoptosis. Furthermore, alterations in the lipid composition determine membrane rigidity and fluidity, and play a crucial role in membrane organization, dynamics, and function [10]. Because of their biological role, lipids have been the subject of an intense area of research since the 1960s, which unfortunately was held back due to limited instrument platforms. Nowadays, lipidomics is considered an emerging science of fundamental importance for clarifying the biochemical pathways involved in several pathologies or cellular stress adaptations [11]. Advances in mass spectrometry (MS) and data processing, as well as the incorporation of soft ionization techniques, as ESI-MS2 method, has revolutionized the use of mass spectrometry, ushering this analytical tool in the field of lipidomics [12]. The lipidomics study can be applied as untargeted and targeted approaches, each with its own advantages and limitations [13]. Untargeted lipidomics focuses on the analysis of all detectable metabolites in a sample, including unknown chemicals, while targeted lipidomics is the measurement of defined groups of metabolites. While the strength of the targeted approach is validating one or more hypotheses, untargeted lipidomics allows for the discovery of new compounds that have led to a number of breakthroughs in understanding human disease risks [14]. Untargeted analyses can be performed with or without the addition of internal rules. When internal standards are added to samples, the method can provide pseudoconcentration results for particular metabolites or for metabolites with similar physicochemical properties (e.g., lipids). While these results are not truly quantitative, they may be accurate enough for case/control comparisons [15].
Despite the great relevance of the topic, currently few studies have been carried out to investigate the behaviour of lipids in erythrocyte samples cultured under simulated gravity conditions. In 2009, Ivanova et al. investigated blood samples from Russian cosmonauts by observing significant changes in the phospholipids class [16]. An increase in the percentage of phosphatidylcholine may be clearly associated with the increase in membrane rigidity. On the other hand, changes in the physicochemical properties of the plasma membrane of erythrocytes (microviscosity and permeability) can influence the efficiency of oxygen transfer, the state of the haemoglobin, and changes in the conformation of hematoporphyrin. Furthermore, changes in the in erythrocyte structure through an ultrastructural morphological analysis can be assessed by atomic force microscopy [17]. However, the study conducted by Ivanova’s team reported data deriving mainly from studies carried out after the end of a space flight, while only few data are related to changes that occur during a space flight. Moreover, lipid and phospholipid compositions of erythrocyte membranes were assayed by thin layer chromatography followed by densitometric measurement of stained dots. This technique provides information on the entire lipid class, but hardly allows the recognition of the specific lipid compounds. Since no data are reported on this subject, we decided to exploit the potential offered by chromatographic and mass spectrometry innovations to better understand the lipid modifications suffered by erythrocytes during simulated microgravity conditions. For this reason, with the aim to better understand which metabolic and/or structural changes occur in the erythrocytes subjected to low gravity, an experimental analysis of the erythrocytes’ lipid profile and their morphology under normal- and micro-g conditions was carried out following a recent investigation on the subject [18]. In detail, human erythrocytes were cultured in simulated gravity conditions, and they were collected at different times of clinorotation. For each sample, the organic phase was collected and analysed through ion mobility Q-TOF mass spectrometer (UHPLC-IM-QTOF-MS).
## 2. Results and Discussion
To investigate the erythrocytes’ lipid profile after clinorotation and to describe possible variations among the different lipid categories, samples were analysed by IM-QTOF-LC/MS and representative total ion chromatograms are shown in Figure 1.
Data processing yielded 215 and 160 features for the positive (PIA) and negative ionization analysis (NIA), respectively, which were subjected to multivariate statistical analysis (MVA). Chemical composition analysis indicated that the lipid fraction was composed of lipids from the following classes: free fatty acid (FA), lysophosphatidylcholines (LysoPC), phosphatidylcholines (PC), phosphatidylethanolamines (PE), sphingomyelins (SM), ceramides (Cer), and ether-linked oxidized phosphatidylcholine (EtherOxPC). Initially, to study sample distribution, to detect outliers, and to highlight differences or common features, a PCA was performed. The unsupervised analysis of both PIA and NIA features did not indicate any sample clustering correlated to clinorotation as shown in Figure 2.
However, the arrangement of the samples in the multivariate space appeared to be influenced by the time factor. Thus, to further limit the time factor influence, for each time point, we performed a PLS-DA. The validation parameters of the PIA and NIA models built for the samples collected at 6, 9, and 24 h are reported in the caption of the resulting plots (Figure 3).
To identify metabolites that can discriminate for the two classes of samples (clinorotated vs. control samples), an OPLS-DA model of the IM-QTOF-LC/MS data was performed for each time point and for both polarities of acquisition. The OPLS-DA score plots are reported in Figure 4. In Table 1, we reported the discriminant metabolites between two classes and selected based on VIP value.
Using MS/MS fragmentation data and consulting the Metlin and Lipidomics libraries, we were able to tentatively identify the most discriminant metabolites as reported in Table 1.
Astronauts, after their return from space missions, manifest significant haematological alterations. Since the earliest space missions, symptoms such as structural alterations of red blood cells [18], anaemia [19], thrombocytopenia, [20,21], 10–$17\%$ reduction in plasma volume, and haemolysis [22] were reported. For these reasons, concern about the effects of space flight on haematological processes has been increasing. Several scientific studies allowed different theories to be proposed that may explain the alterations in the size and number of erythrocytes [23,24]. Recently, Trudel et al. showed in astronauts a degradation and $54\%$ reduction in red blood cells [25]. Different factors can lead a human cell to programmed death, such as changes to lipid signal activity [26].
Human cells determine the characteristics of the plasma membrane bilayer by tightly controlling lipid composition and recruiting cytosolic proteins involved in structural functions or signal transduction [27]. The cell membrane is a lipid bilayer essentially formed by phospholipids, cholesterol, and glycolipids [28]. Small variations in percentage composition and molar ratio of the different classes of phospholipids and glycolipids might induce changes in the cell membrane’s fluidity and permeability. In particular, phospholipids are the main components of cell membranes and perform important biological functions.
From our results, it appears that after 6 h of clinorotation, levels of phosphocholines were increased in human erythrocytes. In particular, PC 18:1_20:4, PC 18:0_20:4, PC18:1_18:1, and PC 18:2_18:1 were found to be upregulated. Notably, PC with the arachidonic acid in their moiety were found discriminants. In particular, the proportion of sn-2-arachidonoyl-phosphatidylcholine (20:4-PC) has been shown to be inversely correlated with the activity of protein kinase B (Akt), an important kinase which promotes cell proliferation and survival. 20:4-PC reduces cell proliferation by interfering with the S-phase cell transition and by suppressing Akt downstream signalling and the expression of cyclin, such as LY294002, which is a specific inhibitor of the phosphatidylinositol-3-kinase/Akt [29]. At 9 and 24 h, erythrocytes showed other 20:4-PC upregulated: PC 18:2_20:4, PC 18:3_20:4 and PC 16:0_20:4, and PC 15:0_20:4, 15:1_20:4 and 16:0_20:4, respectively.
With the classical techniques of liquid chromatography coupled to mass spectrometry, the annotation of lipids and thus phosphocholine fatty acid composition with a good confidence interval is difficult due to the large variety of lipid species with different regiochemistry. In our study, the use of an analytical platform such as ion mobility coupled to mass spectrometry providing the collision cross section (CCS) value allows a better and more confident annotation of each metabolite. Each CCS was compared with an internal database and against the unified collision cross section compendium available on LipidMaps [30].
Additionally, a different fatty acid composition of membrane components can result in a greater sensitivity to peroxidative stress, with a consequent increase in membrane fragility. Phosphatidylcholine species containing polyunsaturated fatty acids in their moiety, particularly arachidonate, at the sn-2 position are susceptible to free radical oxidation [31]. An example is represented by 1-palmitoyl-2-arachidonoyl-sn-glycero-3-phosphatidylcholine (PC16.0_20:4), which is a common cell membrane constituent, and circulates within cholesterol particles. At 6 h of clinorotation, erythrocytes showed an upregulation of EtherOxPC 16:0_20:4, while the respective phosphocholine, PC 16:0_20:4, was found to be not discriminant. Simulated microgravity conditions increase reactive oxygen species (ROS) production in various cell types [32]. Generally, in microgravity conditions a different management of cellular resources was observed. In fact, in G0 conditions, there is a more rapid consumption of intracellular ATP, and an increase in ATP expulsion compared to cells cultured under terrestrial gravity conditions, coupled with a reducing power [8] resulting in a more oxidant environment.
Furthermore, inflammation and oxidative stress are associated with lipid peroxidation and the formation of bioactive lipids such as oxidized phosphocholines [33]. C-reactive protein (CRP), an acute-phase protein of hepatic origin that binds to specific structures expressed on the surface of dead or dying cells, promotes phagocytosis as macrophages may bind to these PC-oxidized species. Furthermore, recent studies demonstrate an enrichment of oxidized phosphatidylcholine in apoptotic cells [34]. Indeed, CRP can selectively bind on oxidized phosphatidylcholine but not on native phosphatidylcholine. In addition, oxidized phospholipids are recognized by macrophage scavenger, implying that these innate immune responses participate in cell clearance due to their proinflammatory properties [35]. Moreover, oxidized phosphatidylcholine, specifically oxidized-1-palmitoyl-2-arachidonoyl-sn-glycero-3-phosphatidylcholine (EtherOxPC 16:0_20:4), seems to be involved in ROS production. According to the study of Rouhanizadeh et al., EtherOxPC 16:0_20:4 was able to induce vascular endothelial superoxide production [36].
On the contrary, after 9 h of clinorotation, EtherOxPC(16:0_20:4) was downregulated, while PC 16:0_20:4 was upregulated, resulting non-discriminant after 24 h. These findings can lead us to hypothesize the complex adaptive response of cells.
Interestingly, several sphingomyelins were found to be downregulated for each experimental time point. This should not be surprising considering the mechanism of sphingomyelin synthesis. Indeed, sphingomyelinases (SMases) catalyse the hydrolysis of sphingomyelin to form ceramide and phosphocholine [37].
Taken together, these findings indicate that there are probably several mechanisms underlying spatial anaemia: inhibition of 20:4 PC-mediated cell proliferation and a simultaneous increase in pro-apoptotic signals.
## 3.1. Chemicals
Analytical LC-grade methanol, chloroform, acetonitrile, 2-propanol, and ammonium acetate and formiate were purchased from Sigma Aldrich (Milan, Italy). Bi-distilled water was obtained with a MilliQ purification system (Millipore, Milan, Italy). A SPLASH® LIPIDOMIX® standard component mixture was purchased from Sigma Aldrich (Milan, Italy): PC (15:0–18:1) (d7), PE (15:0–18:1) (d7), PS (15:0–18:1) (d7), PG (15:0–18:1) (d7), PI (15:0–18:1) (d7), PA (15:0–18:1) (d7), LPC (18:1) (d7), LPC 25, LPE (18:1) (d7), Chol Ester (18:1) (d7), MG (18:1) (d7), DG (15:0–18:1) (d7), TG ((15:0–18:1) (d7)-15:0)), SM (18:1) (d9), cholesterol (d7).
## 3.2. Cell Culture
Freshly drawn blood (Rh+) from 9 healthy adults of both sexes (men and women) was used, heparin was added and preserved in citrate-phosphate-dextrose with adenine (CPDA-1). Data are the average ± SD of three independent experiments. RBCs were separated from plasma and leukocytes by washing three times with phosphate-buffered saline (127 mM NaCl, 2.7 mM KCl, 8.1 mM Na2HPO4, 1.5 mM KH2PO4, 20 mM HEPES, 1 mM MgCl2, and pH 7.4) supplemented with 5 mM glucose (PBS glucose) to obtain packed cells. This study was conducted in accordance with Good Clinical Practice guidelines and the Declaration of Helsinki. No ethical approval has been requested as human blood samples were used only to sustain in vitro cultures and patients provided written, informed consent in ASL. 1-Sassari (Azienda Sanitaria Locale. 1-Sassari) centre before entering the study.
## 3.3. Microgravity Simulation
In order to study the effects caused by microgravity on human erythrocytes, the gravity simulator 3D Random Positioning Machine (RPM, Fokker Space, Netherlands) was used at the laboratory of the Department of Biomedical Sciences, University of Sassari, Sardinia, Italy. The 3D Random Positioning Machine (RPM) is a micro-weight (‘microgravity’) simulator based on the principle of ‘gravity-vector-averaging’, built by Dutch Space. The 3D RPM is constructed from two perpendicular frames that rotate independently. This setup was used to constantly change the mean value of the gravity vector to zero. In this way, the 3D RPM provides a simulated microgravity less than 10−3 g. The dimensions of the 3D RPM are limited to 1000 × 800 × 1000 mm (length × width × height). The 3D RPM is connected to a computer, and through a specific software the mode and speed of rotation were selected. Random Walk mode with an 80 degree/s (rpm) was chosen.
The red blood cell samples were carefully deposited in 2 mL tubes together with PBS-glucose ($30\%$ haematocrit, approximately 3.4 × 109 cells) in a dedicated room at 37 °C. The control group samples were placed in the static bar at 1 g to undergo the same vibrations as the samples placed in µg conditions. Both control (1 g) and case (0 g) samples were collected after different time points (0, 6, 9, 24 h). Subsequently, the red blood cells were centrifuged and resuspended in 1 mL of lysis buffer [5 mM Na2HPO4, 1 mM EDTA (pH 8.0)] and stored at −20 °C until use for lipidomic analysis or fixed for confocal microscopic analysis.
## 3.4. Sample Preparation for UHPLC-IM-QTOF-MS Analysis
In order to investigate changes in the lipidome, analysis by UHPLC- IM-QTOF-MS requires the extraction of lipid content from cells [38]. An amount of 50 µL of human erythrocyte solution was extracted following the Folch procedure using 0.700 mL of a methanol and chloroform mixture ($\frac{2}{1}$, v/v). Samples were vortexed every 15 min up to 1 h, when 0.350 mL of chloroform and 0.150 mL of water were subsequently added. The solution thus obtained was centrifuged at 17,700 rcf for 10 min, and 0.600 mL of the organic layer was transferred into a glass vial and dried under a nitrogen stream. The dried chloroform phase was reconstituted with 50 μL of a methanol and chloroform mixture ($\frac{1}{1}$, v/v) and 75 μL isopropanol:acetonitrile:water mixture (2:1:1 v/v/v). Quality control (QC) samples were prepared taking an aliquot of 10 μL of each sample. All samples thus prepared were injected in UHPLC-IM-QTOF-MS/MS and acquired in negative ionization mode, while for positive ionization mode they were diluted in ratio 1:10.
## 3.5. UHPLC-IM-QTOF-MS/MS Analysis
The chloroform phase was analysed with a 6560-drift tube ion mobility LC-QTOF-MS coupled with an Agilent 1290 Infinity II LC system. An aliquot of 4.0 μL from each sample was injected in a Luna Omega C18, 1.6 μm, 100 mm × 2.1 mm chromatographic column (Phenomenex, Castel Maggiore (BO), Italy). The column was maintained at 50 °C at a flow rate of 0.4 mL/min. The mobile phase for positive ionization mode consisted of (A) 10 mM ammonium formate solution in $60\%$ of milliQ water and $40\%$ of acetonitrile and (B) 10 mM ammonium formate solution containing $90\%$ of isopropanol and $10\%$ of acetonitrile. In positive ionization mode, the chromatographic separation was obtained with the following gradient: initially, $80\%$ of A, then a linear decrease from $80\%$ to $50\%$ of A in 2.1 min, then at $30\%$ in 10 min. Subsequently, the mobile phase A was again decreased from $30\%$ to $1\%$ and stayed at this percentage for 1.9 min, and then was brought back to the initial conditions in 1 min. The mobile phase for the chromatographic separation in the negative ionization mode differed only for the use of 10 mM ammonium acetate instead of ammonium formate.
An Agilent jet stream technology source was operated in both positive and negative ion modes with the following parameters: gas temperature, 200 °C; gas flow (nitrogen) 10 L/min; nebulizer gas (nitrogen), 50 psig; sheath gas temperature, 300 °C; sheath gas flow, 12 L/min; capillary voltage 3500 V for positive and 3000 V for negative; nozzle voltage 0 V; fragmentor 150 V; skimmer 65 V, octapole RF 7550 V; mass range, 50−1700 m/z; capillary voltage, 3.5 kV; collision energy 20 eV in positive and 25 eV in negative mode, mass precursor per cycle = 3. High-purity nitrogen ($99.999\%$) was used as a drift gas with a trap fill time and a trap release time of 2000 and 500 µs, respectively. Before the analysis, the instrument was calibrated using an Agilent tuning solution at the mass range of m/z 50–1700. Samples were evaporated with nitrogen at the pressure of 48 mTorr and at the temperature of 375 °C, while an Agilent reference mass mix for mass re-calibration was continuously injected during the run schedule.
The Agilent MassHunter LC/MS Acquisition console (revision B.09.00) from The MassHunter suite was used for data acquisition.
## 3.6. Data Analysis
Data acquired with the Agilent 6560 DTIM Q-TOF LC-MS were pre-processed with the software MassHunter Workstation suite (Agilent Technologies, Santa Clara, CA, USA). This software (Mass Profiler 10.0) allowed us to perform mass re-calibration, DTCCSN2 re-calibration, time alignment, and deconvolution of signals, yielding a matrix containing all features present across all samples. The removal of background noise and unrelated ions was performed by a recursive feature extraction tool, yielding a matrix containing all the features present across all samples. Furthermore, to eliminate non-specific information, data matrix quality assurance was performed. This filtered matrix was then subjected to multivariate statistical analysis using SIMCA software 15.0 (Umetrics, Umeå, Sweden).
First, a principal component analysis (PCA) was carried out. This unsupervised analysis allows an observation of samples and variables distribution in the multivariate space on the basis of their similarity and dissimilarity. This was followed by partial least square-discriminant analysis (PLS-DA) with its orthogonal extension (OPLS-DA), which was used as a classificatory model to visualize and evaluate the differences between sample classes.
## 4. Conclusions
Spatial anaemia in astronauts has been noted since the earliest space missions, while the contributing mechanisms during space flight remained unclear. To investigate the molecular mechanisms that induce a reduction in the number of erythrocytes during spaceflight, we decided to analyse the lipid profile of human erythrocytes under microgravity conditions. Thanks to the advancement of hyphenated techniques and mass analysers, we were able to identify biologically active complex lipids susceptible to microgravity, allowing new possible hypotheses that explain the anaemia experienced by astronauts.
In more detail, lipidomic analysis of erythrocytes revealed a double mechanism that generates the reduction in the number of red blood cells. On one hand, there is an increase in the levels of 20:4 PC, reducing cellular proliferation. On the other hand, the increase in the levels of EtherOxPC 16:0_20:4 stimulates the immune response by attracting the C-reactive protein and macrophages and induces an increase in ROS production [33,34]. ROS increase caused by microgravity inevitably induces mitochondrial damage and dysfunction as indicated by the accumulation of HexCer lipid species in clinorotated erythrocytes. This accumulation acts as a pro-apoptotic signal condemning the erythrocytes to death.
In this study, we reported a set of lipid discriminants or potential biomarkers linked to microgravity exposure with the aim to explore in the future specific lipid pathways and form the foundation in the development of novel therapeutics in hopes of reducing the effects of space flights. However, further studies are needed to accurately measure lipids in erythrocytes samples to better understand the clinical effects of microgravity.
## References
1. Riwaldt S., Corydon T.J., Pantalone D., Sahana J., Wise P., Wehland M., Krüger M., Melnik D., Kopp S., Infanger M.. **Role of Apoptosis in Wound Healing and Apoptosis Alterations in Microgravity**. *Front. Bioeng. Biotechnol.* (2021) **9** 1-22. DOI: 10.3389/fbioe.2021.679650
2. Bizzarri M., Fedeli V., Piombarolo A., Angeloni A.. **Space Biomedicine: A Unique Opportunity to Rethink the Relationships between Physics and Biology**. *Biomedicines* (2022) **10**. DOI: 10.3390/biomedicines10102633
3. Thirsk R., Kuipers A., Mukai C., Williams D.. **The space-flight environment: The International Space Station and beyond**. *CMAJ* (2009) **180** 1216-1220. DOI: 10.1503/cmaj.081125
4. Sheyn D., Pelled G., Netanely D., Domany E., Gazit D.. **The Effect of Simulated Microgravity on Human Mesenchymal Stem Cells Cultured in an Osteogenic Differentiation System: A Bioinformatics Study**. *Tissue Eng.-Part A* (2010) **16** 3403-3412. DOI: 10.1089/ten.tea.2009.0834
5. Hughes-Fulford M., Tjandrawinata R., Fitzgerald J., Gasuad K., Gilbertson V.. **Effects of microgravity on osteoblast growth**. *Gravit. Space Biol. Bull.* (1998) **11** 51-60. PMID: 11540639
6. Han X., Qiu L., Zhang Y., Kong Q., Wang H., Wang H., Li H., Duan C., Wang Y., Song Y.. **Transplantation of Sertoli-Islet Cell Aggregates Formed by Microgravity: Prolonged Survival in Diabetic Rats**. *Exp. Biol. Med.* (2009) **234** 595-603. DOI: 10.3181/0812-RM-359
7. **National Space Biomedical Research Institute Strategic Plan 2010**. *Biomed Res.* (2010)
8. Dinarelli S., Longo G., Dietler G., Francioso A., Mosca L., Pannitteri G., Boumis G., Bellelli A., Girasole M.. **Erythrocyte’s aging in microgravity highlights how environmental stimuli shape metabolism and morphology**. *Sci. Rep.* (2018) **8** 5277. DOI: 10.1038/s41598-018-22870-0
9. Poon C.. **Factors implicating the validity and interpretation of mechanobiology studies in simulated microgravity environments**. *Eng. Rep.* (2020) **2** 1-18. DOI: 10.1002/eng2.12242
10. Maxfield F.R., Tabas I.. **Role of cholesterol and lipid organization in disease**. *Nature* (2005) **438** 612-621. DOI: 10.1038/nature04399
11. Wenk M.R.. **The emerging field of lipidomics**. *Nat. Rev. Drug Discov.* (2005) **4** 594-610. DOI: 10.1038/nrd1776
12. Carrasco-Pancorbo A., Navas-Iglesias N., Cuadros-Rodríguez L.. **From lipid analysis towards lipidomics, a new challenge for the analytical chemistry of the 21st century. Part I: Modern lipid analysis**. *TrAC-Trends Anal. Chem.* (2009) **28** 263-278. DOI: 10.1016/j.trac.2008.12.005
13. Roberts L.D., Souza A.L., Gerszten R.E., Clish C.B.. **Targeted Metabolomics**. *Curr. Protoc. Mol. Biol.* (2012) **98** 30.2.1-30.2.24. DOI: 10.1002/0471142727.mb3002s98
14. Wang Z., Klipfell E., Bennett B.J., Koeth R., Levison B.S., DuGar B., Feldstein A.E., Britt E.B., Fu X., Chung Y.-M.. **Gut Flora Metabolism of Phosphatidylcholine Promotes Cardiovascular Disease**. *Nature* (2011) **472** 57-63. DOI: 10.1038/nature09922
15. Cajka T., Fiehn O.. **Toward Merging Untargeted and Targeted Methods in Mass Spectrometry-Based Metabolomics and Lipidomics**. *Anal. Chem.* (2016) **88** 524-545. DOI: 10.1021/acs.analchem.5b04491
16. Ivanova S.M., Morukov B.V., Labetskaia O.I., IuV I., Levina A.A., Kozinets G.I.. **Morphobiochemical assay of the red blood system in members of the prime crews of the International Space Station**. *Hum. Physiol.* (2009) **43** 43-47
17. Girasole M., Pompeo G., Cricenti A., Congiu-Castellano A., Andreola F., Serafino A., Frazer B., Boumis G., Amiconi G.. **Roughness of the plasma membrane as an independent morphological parameter to study RBCs: A quantitative atomic force microscopy investigation**. *Biochim. Biophys. Acta (BBA)-Biomembr.* (2007) **1768** 1268-1276. DOI: 10.1016/j.bbamem.2007.01.014
18. Manis C., Manca A., Murgia A., Uras G., Caboni P., Congiu T., Faa G., Pantaleo A., Cao G.. **Understanding the Behaviour of Human Cell Types under Simulated Microgravity Conditions: The Case of Erythrocytes**. *Int. J. Mol. Sci.* (2022) **23**. DOI: 10.3390/ijms23126876
19. Alfrey C.P., Udden M.M., Leach-Huntoon C., Driscoll T., Pickett M.H.. **Control of red blood cell mass in spaceflight**. *J. Appl. Physiol.* (1996) **81** 98-104. DOI: 10.1152/jappl.1996.81.1.98
20. Taylor G.R.. **Advances in experimental medicine**. *Advances in Experimental Medicine and Biology* (1997) **Volume 225** 269-271
21. Meehan R.T., Neale L.S., Kraus E.T., Stuart C.A., Smith M.L., Cintron N.M., Sams C.F.. **Alteration in human mononuclear leucocytes following space flight**. *Immunology* (1992) **76** 491-497. PMID: 1326479
22. Diedrich A., Paranjape S.Y., Robertson D.. **Plasma and Blood Volume in Space**. *Am. J. Med. Sci.* (2007) **334** 80-86. DOI: 10.1097/MAJ.0b013e318065b89b
23. Erslev A.J.. **Erythropoietin**. *N. Engl. J. Med.* (1991) **324** 1339-1344. DOI: 10.1056/NEJM199105093241907
24. Koury M.J., Bondurant M.C.. **Erythropoietin Retards DNA Breakdown and Prevents Programmed Death in Erythroid Progenitor Cells**. *Science* (1990) **248** 378-381. DOI: 10.1126/science.2326648
25. Trudel G., Shahin N., Ramsay T., Laneuville O., Louati H.. **Hemolysis contributes to anemia during long-duration space flight**. *Nat. Med.* (2022) **28** 59-62. DOI: 10.1038/s41591-021-01637-7
26. Crimi M., Degli Esposti M.. **Apoptosis-induced changes in mitochondrial lipids**. *Biochim. Biophys. Acta (BBA)-Mol. Cell Res.* (2011) **1813** 551-557. DOI: 10.1016/j.bbamcr.2010.09.014
27. Sprong H., Van Der Sluijs P., van Meer G.. **How proteins move lipids and lipids move proteins**. *Nat. Rev. Mol. Cell Biol.* (2001) **2** 504-513. DOI: 10.1038/35080071
28. Robertson J.L.. **The lipid bilayer membrane and its protein constituents**. *J. Gen. Physiol.* (2018) **150** 1472-1483. DOI: 10.1085/jgp.201812153
29. Koeberle A., Shindou H., Koeberle S.C., Laufer S.A., Shimizu T., Werz O.. **Arachidonoyl-phosphatidylcholine oscillates during the cell cycle and counteracts proliferation by suppressing Akt membrane binding**. *Proc. Natl. Acad. Sci. USA* (2013) **110** 2546-2551. DOI: 10.1073/pnas.1216182110
30. Picache J.A., Rose B.S., Balinski A., Leaptrot K.L., Sherrod S.D., May J.C., McLean J.A.. **Collision cross section compendium to annotate and predict multi-omic compound identities**. *Chem. Sci.* (2019) **10** 983-993. DOI: 10.1039/C8SC04396E
31. Subbanagounder G., Leitinger N., Schwenke D.C., Wong J.W., Lee H., Rizza C., Watson A.D., Faull K.F., Fogelman A.M., Berliner J.A.. **Determinants of bioactivity of oxidized phospholipids: Specific oxidized fatty acyl groups at the sn-2 position**. *Arterioscler. Thromb. Vasc. Biol.* (2000) **20** 2248-2254. DOI: 10.1161/01.ATV.20.10.2248
32. Ran F., An L., Fan Y., Hang H., Wang S.. **Simulated microgravity potentiates generation of reactive oxygen species in cells**. *Biophys. Rep.* (2016) **2** 100-105. DOI: 10.1007/s41048-016-0029-0
33. Ashraf M.Z., Kar N.S., Podrez E.A.. **Oxidized phospholipids: Biomarker for cardiovascular diseases**. *Int. J. Biochem. Cell Biol.* (2009) **41** 1241-1244. DOI: 10.1016/j.biocel.2008.11.002
34. Huber J., Vales A., Mitulović G., Blumer M., Schmid R., Witztum J.L., Binder B.R., Leitinger N.. **Oxidized Membrane Vesicles and Blebs from Apoptotic Cells Contain Biologically Active Oxidized Phospholipids That Induce Monocyte-Endothelial Interactions**. *Arterioscler. Thromb. Vasc. Biol.* (2002) **22** 101-107. DOI: 10.1161/hq0102.101525
35. Chang M.-K., Binder C.J., Torzewski M., Witztum J.L.. **C-reactive protein binds to both oxidized LDL and apoptotic cells through recognition of a common ligand: Phosphorylcholine of oxidized phospholipids**. *Proc. Natl. Acad. Sci. USA* (2002) **99** 13043-13048. DOI: 10.1073/pnas.192399699
36. Rouhanizadeh M., Hwang J., Clempus R.E., Marcu L., Lassègue B., Sevanian A., Hsiai T.K.. **Oxidized-1-palmitoyl-2-arachidonoyl-sn-glycero-3-phosphorylcholine induces vascular endothelial superoxide production: Implication of NADPH oxidase**. *Free Radic. Biol. Med.* (2005) **39** 1512-1522. DOI: 10.1016/j.freeradbiomed.2005.07.013
37. Patwardhan G.A., Beverly L.J., Siskind L.J.. **Sphingolipids and mitochondrial apoptosis**. *J. Bioenerg. Biomembr.* (2016) **48** 153-168. DOI: 10.1007/s10863-015-9602-3
38. Ulmer C.Z., Jones C.M., Yost R.A., Garrett T.J., Bowden J.A.. **Optimization of Folch, Bligh-Dyer, and Matyash sample-to-extraction solvent ratios for human plasma-based lipidomics studies**. *Anal. Chim. Acta* (2018) **1037** 351-357. DOI: 10.1016/j.aca.2018.08.004
|
---
title: Clostridium butyricum Prevents Dysbiosis and the Rise in Blood Pressure in
Spontaneously Hypertensive Rats
authors:
- Xianshu Luo
- Zhuoyu Han
- Qing Kong
- Yuming Wang
- Haijin Mou
- Xuefeng Duan
journal: International Journal of Molecular Sciences
year: 2023
pmcid: PMC10002514
doi: 10.3390/ijms24054955
license: CC BY 4.0
---
# Clostridium butyricum Prevents Dysbiosis and the Rise in Blood Pressure in Spontaneously Hypertensive Rats
## Abstract
Hypertension is accompanied by dysbiosis and a decrease in the relative abundance of short-chain fatty acid (SCFA)-producing bacteria. However, there is no report to examine the role of C. butyricum in blood pressure regulation. We hypothesized that a decrease in the relative abundance of SCFA-producing bacteria in the gut was the cause of spontaneously hypertensive rats (SHR)-induced hypertension. C. butyricum and captopril were used to treat adult SHR for six weeks. C. butyricum modulated SHR-induced dysbiosis and significantly reduced systolic blood pressure (SBP) in SHR ($p \leq 0.01$). A 16S rRNA analysis determined changes in the relative abundance of the mainly SCFA-producing bacteria Akkermansia muciniphila, Lactobacillus amylovorus, and Agthobacter rectalis, which increased significantly. Total SCFAs, and particularly butyrate concentrations, in the SHR cecum and plasma were reduced ($p \leq 0.05$), while C. butyricum prevented this effect. Likewise, we supplemented SHR with butyrate for six weeks. We analyzed the flora composition, cecum SCFA concentration, and inflammatory response. The results showed that butyrate prevented SHR-induced hypertension and inflammation, and the decline of cecum SCFA concentrations ($p \leq 0.05$). This research revealed that increasing cecum butyrate concentrations by probiotics, or direct butyrate supplementation, prevented the adverse effects of SHR on intestinal flora, vascular, and blood pressure.
## 1. Introduction
The intestinal flora’s potential role in influencing host health has attracted considerable attention in recent decades. Many metabolic diseases, inflammatory bowel diseases, and cardiovascular diseases have been reported to be linked to flora disorders [1,2,3]. In recent years, much seminal evidence has shown that abnormal intestinal flora is closely associated with changes in blood pressure in the host. Iñaki Robles-Vera and co-workers reported that fecal microbiota transplantation from adult spontaneously hypertensive rats (SHR), to adult Kyoto rats (WKY), resulted in a chronic rise in blood pressure (BP), vascular oxidative stress, and impaired endothelial function. Conversely, fecal microbiota transplantation from WKY to adult SHR induced a blood pressure reduction and improvement of endothelial dysfunction [4]. Mechanisms linking the gut microbiota to hypertension include dysbiosis, inflammation, gut permeability, and decreased production of short-chain fatty acids (SCFAs), particularly butyrate [5,6,7].
SCFAs are mainly produced through fermentation of indigestible carbohydrates by bacteria in the intestine [8], which can regulate body weight, energy metabolic balance, lipid metabolism, and other pathophysiological processes [9,10], including hypertension. SCFAs can be transported to the surface of the cell membrane, through mono-carboxylate transporter (MCT), to bind with G protein-coupled receptors (GPCRs), and modulate vascular tone and inflammation to regulate blood pressure through the interaction with GPCRs or inhibition of histone deacetylases [11,12]. Onyszkiewicz and co-workers demonstrated that butyrate could enter the circulation through the intestinal-vascular barrier, and act on GPR41/GPR43 to relax mesenteric arteries and decrease blood pressure [13]. In addition, treatment of Olfr78 knockout and non-knockout mice with propionate revealed that Olfr78 knockout mice showed lower blood pressure levels [14].
It has been reported that the probiotics Bifidobacterium breve CECT7263 and *Lactobacillus fermentum* CECT5716, could regulate the blood pressure of SHR through modulating short-chain fatty acid-producing genera in the intestine [15]. SHR is an established model of genetic hypertension characterized by elevated blood pressure, arterial remodeling, endothelial dysfunction, hyperlipidemia, vascular inflammation, and immune system dysregulation [16,17,18]. In addition, dysbiosis and a decrease in the relative abundance of SCFA-producing bacteria have been reported in SHR [19,20]. Clostridium butyricum (C. butyricum) is an anaerobic, gram-positive bacillus, known as a butyrate producer and a regulator of gut health [2]. It resides in the gastrointestinal tract and has a protective role against pathogenic bacteria and intestinal injury, via the modulation of gut microbial metabolites [21,22,23]. C. butyricum are capable of utilizing a range of carbohydrates, and produce several fermentation products, with butyrate as the major product. Butyrate plays a crucial role in promoting gut health and has been used in clinical practice to treat various chronic enteric diseases [24]. While probiotics and SCFAs have the potential to regulate blood pressure, we did not know if C. butyrcium, the main butyrate producer, could regulate blood pressure in SHR.
Based on the potential of probiotic and bacterial-derived SCFAs for blood pressure regulation, we hypothesized that C. butyricum could modulate intestinal flora, improve vascular inflammation, and lower blood pressure. We investigated the hypotensive effect of C. butyricum on SBP in SHR, using SHR animals as a model, and investigated the effects of SHR, WKY, and C. butyricum on intestinal immunity and vascular inflammation, a key component of SHR-induced hypertension. In addition, we analyzed the potential mechanistic link between dysbiosis and hypertension by 16S rRNA V1–V9 sequencing.
## 2.1. C. butyricum and Butyrate Prevented SHR-Induced Hypertension
In fed rats, the SBP of WKY remained around 100 mmHg, which was highly significantly different from that of SHR ($p \leq 0.01$). The SBP of SHR gradually increased between week 9 and week 13, and stabilized at 189 ± 7 mmHg by week 15. Given that multiple animal models of hypertension exhibit reduced SCFA-producing bacteria, we treated SHR with C. butyricum or butyrate (the main metabolite of C. butyricum). The SBP of SHR remained steadily lower throughout the treatment period of C. butyricum or butyrate. By week 15, the SBP of rats in the SHR-Cb group was 50 mmHg lower than that of the rats in the SHR group (Figure 1b,c). In contrast, there was no significant difference in SBP between rats in the WKY-C. butyricum group compared to those in the WKY group (Figure 1a). This suggests that C. butyricum was able to prevent elevated SBP in SHR but did not affect normotensive rats. Captopril lowers blood pressure in hypertension. The results of the repeated measures ANOVA showed that C. butyricum treatment significantly reduced the SBP in SHR after two weeks (Tables S1–S3). In addition, to clarify the effect of C. butyricum in preventing elevated blood pressure, we treated SHR with captopril for 6 weeks. The results showed that the SBP of SHR was significantly reduced after 2 weeks of treatment (Tables S1–S3), but C. butyricum or butyrate had a significant effect in preventing elevated SBP after 4 to 6 weeks of treatment (Figure 1d).
## 2.2. C. butyricum and Butyrate Altered the Colonic Microbiota Composition
To investigate the effect of C. butyricum and butyrate intervention on gut microbiology, we analyzed the full-length 16S rRNA sequencing of the colon contents. The richness and diversity of the intestinal flora were lower in SHR than in WKY ($p \leq 0.05$). C. butyricum increased the richness and diversity of the colonic flora by increasing the values of Chao and Shannon at the OUT level (Figure 2a), while butyrate did not alter the abundance of colonic flora. Similarly, C. butyricum and butyrate showed a clear separation from SHR alone in both unweighted and weighted UniFrac principal coordinate analyses (Figure 3). Firmicutes, Bacteroidetes, Actinobacteria, Proteobacteria, Fusobacteria, and Verrucomicrobia are the main phyla in the gut, with Firmicutes and Bacteroidetes accounting for $90\%$ of the intestinal flora [25,26]. As shown in Figure 2c, Firmicutes were significantly increased in the SHR colon compared to WKY (WKY: $67\%$, SHR: $80\%$), while the relative abundance of Proteobacteria in the SHR colon was $6\%$ higher than that of WKY. C. butyricum and butyrate significantly increased the relative abundance of Verrucomicrobia and Bacteroidetes in the SHR colon ($p \leq 0.01$). This indicates that C. butyricum and butyric acid altered the intestinal microbial community of SHR. On the genus level, the relative abundance of Akkermansia in the colon of SHR was significantly increased by C. butyricum and butyrate (SHR: $4\%$, SHR-C. butyricum: $33\%$, SHR-butyrate: $18\%$) (Figure 2c).
The Firmicutes/Bacteroides ratio is a marker of intestinal health. Some diseases such as inflammatory bowel disease and excessive obesity, including hypertension, have been reported to have significantly higher Firmicutes/Bacteroides ratios, and regulation of the ratio is effective in treating the disease [27,28]. We analyzed Firmicutes/Bacteroides in all groups of rats, and we demonstrated that Firmicutes/Bacteroides were significantly higher in SHR than WKY ($p \leq 0.01$). Treatment with C. butyricum and butyrate prevented Firmicutes/Bacteroides from being elevated ($p \leq 0.01$) (Figure 2b).
In addition, we analyzed changes in colonic microorganisms on the species level of SHR after 6 weeks of C. butyricum and butyrate treatment. Consistent with previous studies, short-chain fatty acid-producing bacteria were decreased in the SHR gut. Akkermansia muciniphila, Lactobacillus amylovorus, Muribaculum sp002492595, Agthobacter rectalis, Romboutsia ilealis, and lleibacterium valens had their relative abundance altered by C. butyricum and butyrate (Figure 2c–h). We demonstrated that C. butyricum and butyrate prevented the decrease in the relative abundance of Akkermansia muciniphila, Muribaculum sp002492595, and *Agthobacter rectalis* in the colonic contents caused by SHR (Figure 2c–e). C. butyricum and butyrate significantly increased *Akkermansia muciniphila* (SHR-C. butyricum: $34\%$, SHR-butyrate: $23\%$) (Figure 2c), interestingly, Roshanravan N. and co-workers reported that supplementation with butyrate promoted *Akkermansia muciniphila* levels in the intestine, improved vascular inflammation and oxidative stress, and lowered blood pressure [29,30], which is consistent with the results of our study. Both *Lactobacillus amylovorus* and lleibacterium valens have been reported to reduce obesity [31,32]. The results showed that colonic *Lactobacillus amylovorus* increased after 6 weeks of C. butyricum treatment, but that butyrate did not alter their relative abundance in the colon (Figure 2d), while treatment with C. butyricum and butyrate did not alter the colonic levels of lleibacterium valens (Figure 2h). In addition, treatment with both C. butyricum and butyrate reduced the levels of the SHR colonic pathogen *Romboutsia ilealis* (Figure 2g).
## 2.3. Inflammation of the Colon and Vascular Improved by C. butyricum and Butyrate
We examined the effects of SHR on the colon and vascular system with and without C. butyricum and butyrate. Figure 4b shows that SHR caused severe intestinal mucosal detachment in the colon, goblet cells in the colon were reduced, and there was a proliferation of inflammatory fibrous tissue; all of these were prevented by treatment with C. butyricum and butyrate. In addition, we observed that SHR caused endothelial cell damage, proliferation, and shedding of outer membrane cells in the aorta. Treatment with C. butyricum and butyrate improved the vascular damage caused by SHR (Figure 4b). Since SHR caused inflammation in the colon and vascular system, we next examined serum levels of inflammatory factors. ELISA demonstrated that C. butyricum and butyrate decreased tumor necrosis factor-α (TNF-α), interleukin-6 (IL-6), interleukin-17A (IL-17A), and lipopolysaccharide (LPS), and increased the inhibitory inflammatory factor interleukin-10 (IL-10) in the SHR circulation (Figure 4a,c–f). This suggests that the regulation of the SBP in SHR by C. butyricum and butyrate possibly correlates with reduced levels of inflammation.
## 2.4. SHR Reduced Cecal and Plasma SCFAs, Which Were Prevented by C. butyricum and Butyrate
To clarify the effect of dysbiosis on SHR, we next tested whether the SHR-induced changes to the microbiota led to changes in microbial metabolites that may contribute to the adverse effects of SHR on the gut and blood pressure. Microbiota analysis showed that some short-chain fatty acid producers were modified by SHR, which was prevented by C. butyricum and butyrate (Figure 2d–h). Therefore, we measured the SCFAs concentrations in colonic contents and plasma. SHR caused a significant decrease in colonic and plasma total SCFAs, particularly butyrate concentrations (Figure 5). C. butyricum and butyrate significantly prevented the SHR-induced decrease in colonic butyrate, acetate, and propionate concentrations, but the effects on acetate and propionate were not significant (Figure 5c and Figure 6). SHR caused a decrease in total SCFAs in plasma, but we found that SHR did not change the concentration of propionate in plasma (Figure 5b and Figure 6d).
## 2.5. Relative Expression Levels of SCFA Transporters in the Proximal Colon of WKY and SHR
We tested the relative expression levels of three monocarboxylate transporter proteins involved in transporting butyrate and other SCFAs into and through the intestinal epithelium [11,33]. Figure 7a,b shows that the relative expression levels of MCT1 and MCT4 were significantly reduced in the colon of SHR compared to WKY ($p \leq 0.05$), which was enhanced by treatment with C. butyricum or butyrate. Similarly, the relative expression level of Slc5a8 was also enhanced by C. butyricum or butyrate, but it was not significant ($$n = 4$$ per group, $$p \leq 0.229$$ for SHR vs. SHR-C. butyricum, $$p \leq 0.324$$ for SHR vs. SHR-butyrate). This result suggested that the decrease in circulating total SCFAs and butyrate concentration may be due to a decrease in SCFAs transport proteins caused by SHR.
## 2.6. Reduced Expression of SCFA-Sensing Receptors in the SHR Vascular
To evaluate the direct effects of SCFAs on the vasculature, we examined the relative expression levels of SCFA-sensing receptors in the aorta. Most notable among the SCFA targets is the mammalian G protein-coupled receptor pair of GPR41 and GPR43, that can be expressed in blood vessels [34]. We observed reduced relative expression levels of GPR41 and GPR43 in the SHR aorta (GPR41, $$p \leq 0.051$$; GPR43, $$p \leq 0.004$$ for WKY vs. SHR). After 6 weeks of treatment, the mRNA expression levels of GPR41 and GPR43 in the SHR- C. butyricum group and SHR-butyrate group were upregulated compared with those in the SHR group, meanwhile, compared with the C. butyricum group, the results showed that butyrate treatment group had a more significant regulatory effect ($p \leq 0.01$). Therefore, the mechanism for sensing SCFAs appears to be partially compromised in the SHR aorta, but was ameliorated by treatment with C. butyricum or butyrate.
## 2.7. SHR Increased Th17/Treg in the Spleen and Aorta and C. butyricum Restored Th17/Treg
The host inflammation is associated with a balance between the pro-inflammatory and anti-inflammatory actions of regulatory T cells. The Th17/Treg balance has been shown to normalize endotoxemia, prevent endothelium-dependent diastolic damage to acetylcholine, and reduce blood pressure in spontaneously hypertensive rats [30]. Therefore, we examined the distribution of Th17 and Treg in the aorta and spleen. Figure 8 shows that SHR leads to Th17 increases and Treg decreases in the spleen and aorta. These were ameliorated by treatment with C. butyricum or butyrate.
## 3. Discussion
In recent years, much seminal evidence has demonstrated for the first time that abnormal gut flora is closely associated with changes in blood pressure in the host [35]. Both animal models of hypertension and human hypertension are accompanied by dysbiosis [36]. Through the study of the effect of C. butyricum on the intestinal flora, we have demonstrated that C. butyricum could regulate the intestinal flora and increase the relative abundance of short-chain fatty acid-producing bacteria [37]. We hypothesized that a reduction in SCFAs was responsible for SHR hypertension.
Probiotics such as Bifidobacterium breve CECT7263 and *Lactobacillus fermentum* CECT5716, have been shown to restore SHR-induced dysbiosis and reduce SBP [15]. These probiotics were found to ferment dietary fiber in the gut to produce SCFAs, which protect against the vascular oxidative stress and endothelial dysfunction caused by hypertension. Several studies have shown that the relative abundance of many SCFA-producing bacteria is reduced in animal models of hypertension [19,20,38]. Bhanu P. Ganesh and co-workers have demonstrated that C. butyricum reduced the effect of hypertension in a model of obstructive sleep apnea on altered microbiota, and increased the relative abundance of many SCFA-producing genera such as Parabacteroides, Roseburia, Clostridium, Bifidobacterium, Ruminococcus, and Blauti [39]. We suggested that C. butyricum and butyrate reduced the effect of SHR on altering the microbiota. Unweighted and weighted UniFrac principal coordinate analyses were used to show that the flora of SHR and WKY were significantly separated, altered by C. butyricum and butyrate. Chao and Shannon showed that C. butyricum and butyrate restored the reduction in flora diversity and abundance caused by SHR. In addition, treatment with C. butyricum and butyrate significantly reduced the Firmicutes/Bacteroides ratio ($p \leq 0.01$). We concluded that SHR induced dysbiosis and elevation of Firmicutes/Bacteroides, which was prevented by C. butyricum or butyrate. On a species-level analysis of all rat colon flora, we found that C. butyricum and butyrate prevented SHR-induced decreases in colonic Akkermansia muciniphila, Muribaculum sp002492595, and *Agthobacter rectalis* levels, which is consistent with previous findings. Akkermansia muciniphila is a native bacterium in the gut that ferments carbohydrates in the intestine to produce acetate and propionate, reducing metabolic disorders and improving low levels of inflammation [40]. Levels of intestinal *Akkermansia muciniphila* are negatively correlated with diabetes, obesity, and other metabolic syndromes [41]. Interestingly, previous studies have reported that supplementation with butyrate promotes intestinal levels of Akkermansia muciniphila, reduces vascular inflammation and oxidative stress, and lowers blood pressure [30]. Thus, *Akkermansia muciniphila* might play a significant role in reducing the SBP in SHR. In conclusion, C. butyricum was demonstrated to modulate the intestinal flora of SHR, reduce the Firmicutes/Bacteroides ratio, and prevent the SHR-induced reduction of short-chain fatty acid-producing species, thereby potentially maintaining colonic butyrate levels.
SCFAs, a major source of energy for epithelial cells, have many beneficial effects, including maintaining the integrity of the intestinal barrier [42], reducing mucosal inflammation, and improving intestinal health [43,44]. We found that SHR caused a significant reduction in colon and plasma SCFAs, particularly butyrate concentrations. C. butyricum and butyrate prevented the SHR-induced decrease in colon butyrate concentrations, but had no significant effect on acetate and propionate. SHR caused a decrease in the total SCFAs of plasma, but it did not modify plasma propionate concentrations. SCFAs could affect the host by activating the G protein-coupled receptor (GPCR) or by inhibiting histone deacetylases, to stabilize the intestinal epithelial barrier, regulate cytokine secretion, alter the T lymphocyte population, increase the protective mucus layer, and regulate antibody secretion [45,46,47,48]. If SCFAs entered the circulatory system, it would also affect tissues and organs outside the intestinal tract [49]. We observed that SHR induced a reduction in plasma concentrations of SCFAs, which was not significantly improved by C. butyricum and butyrate; so we examined the relative expression of SCFAs transporters in the proximal colon and SCFA-sensing receptors. We demonstrated that SHR caused a reduction in colonic MCT1 and MCT4 expression, which was prevented by C. butyricum or butyrate. Similarly, the reduction in the SCFA-sensing receptors GPR-41 and GPR-43 mRNA expression in the aorta of SHR, was ameliorated by C. butyricum or butyrate.
Treatment with C. butyricum and butyrate increased the concentration of intestinal SCFAs, while having no significant effect on plasma SCFAs. Therefore, we examined the possibility of colonic and aortic effects. Our findings in this study showed that the number of mucus-producing goblet cells of the colon was reduced, and there was intestinal mucosal detachment in SHR, and this could be prevented by C. butyricum and butyrate treatment. Similarly, Santisteban and co-workers reported a reduction in goblet cells in the SHR colon [50]. Similarly, C. butyricum and butyrate improved SHR-induced vascular injury. LPS could increase the intestinal permeability, and an increased intestinal permeability would allow more LPS to enter the circulation and exacerbate the inflammatory state [51]. Iñaki Robles-Vera and co-workers reported that LPS levels in SHR serum were significantly higher than WKY, and after treatment with the probiotic *Bifidobacterium bifidum* and SCFAs, blood pressure and LPS were normalized in the treated group of rats. They hypothesized that a reduction in LPS in rat serum was a key factor in the modulatory effect of probiotics on hypertension [15]. Our results have shown that treatment with C. butyricum and butyrate reduced LPS levels in serum of SHR.
In addition, TNF-α is a pro-inflammatory factor produced by macrophages, and plays an important pathogenic role in inflammatory bowel disease. The downregulation of TNF-α can significantly downregulate the inflammation of Crohn’s disease, and the content is positively correlated with inflammation [52]. IL-10 is an anti-inflammatory cytokine produced by immature T cells. Previous studies have shown that 17 strains of clostridium from healthy human microbiota can induce the production of IL-10, thus inhibiting colitis. The probiotic C. Butyricum can promote the production of IL-10 by T cells and thus prevent the occurrence of colitis through an IL-10-dependent mechanism [53]. IL-6 and IL-17A are pro-inflammatory factors produced by Th17 cells. IL-6 is an important cytokine in the process of immune inflammatory reaction, its abnormal content will damage the vascular endothelium and lead to an increase in blood pressure in hypertensive patients [54]. We observed that serum levels of the inflammatory factors TNF-α, IL-6, and IL-17A were increased by SHR and decreased by C. butyricum or butyrate. Conversely, IL-10 was increased by C. butyricum or butyrate. These findings suggest that C. butyricum and butyrate ameliorate vascular inflammation and modulate the SBP of SHR, which may be linked to the modulation of inflammatory factor levels. The balance between Th17 and *Treg is* critical for maintaining immune homeostasis, with the over-activation of Th17 cells exacerbating intestinal inflammation, while the lack of Treg in intestine-associated lymphoid tissue, or its inability to circulate naturally to sites of inflammation, has been shown to cause an immune response in commensal flora, and to induce colitis [55]. In this study, to analyze the effect of SHR on Th17 and Treg, we performed immunofluorescence staining of the spleen and aorta. We concluded that SHR induced an increase in splenic and aortic Th17 and a decrease in Treg, which improved through treatment with C. butyricum or butyrate.
We have shown that [1] SHR caused a decrease in colonic SCFAs, especially butyrate concentrations, and increased intestinal and vascular inflammation, and hypertension. [ 2] in SHR rats, the probiotic C. butyricum, and SCFA butyrate, increased total colonic SCFAs and butyrate concentrations, reduced dysbiosis and colonic injury, and prevented vascular inflammation and hypertension. [ 3] C. butyricum and butyrate regulated the expression level of the major SCFAs transporters MCT1 and MCT4, and the receptors GPR41 and GPR43, in SHR rats. They may effectively regulate the colonic inflammation of SHR by regulating the concentration of SCFAs. [ 4] C. butyricum and butyrate modulated the signal pathway, to elevate the anti-inflammatory level by reducing the expression level of pro-inflammatory factor TNF-α, IL-6 and IL-17A, and improving the expression level of anti-inflammatory factor Il-10. [ 5] C. butyricum and butyrate decreased the level of intestinal LPS, and then ameliorated the intestinal barrier dysfunction caused by SHR. These findings demonstrate the critical role of impaired butyrate production in the development of SHR-induced hypertension, and suggest that treatment targeting increased C. butyricum and microbial butyrate production may prove effective in treating hypertension (Figure 9).
## 4.1. C. butyricum Cultivation
C. butyricum CGMCC 1.5205 (C. butyricum) was preserved at the China General Microbiological Culture Collection Center (CGMCC). A thermo-static incubator (DRP-9052, Senxin, Shanghai, China) was used to cultivate C. butyricum in reinforced *Clostridium medium* for 24 h at 37 °C. 16S rDNA sequencing (27F: AGAGTTTGATCCTGGCTCAG, 1492R: TACGGCTACCTTGTTACGACTT) was performed on the broth of C. butyricum, and the sequencing results were compared on the NCBI website (https://www.ncbi.nlm.nih.gov (accessed on 25 January 2022)) to confirm that the strain cultivated was C. butyricum. The C. butyricum fermentation broth was centrifuged at 10,000× g rpm for 10 min and $20\%$ trehalose was added to the bacterial pellets as a protective agent before freeze-drying. The freeze-dried bacteria were counted using the blood counting chamber, with a result of 1010 cfu/mL. The freeze-dried powder was stored until use.
## 4.2. Experimental Animals
All studies in animals should be conducted in accordance with the National Institutes of Health (NIH) Guide for the Care and Use of Laboratory Animals, or the equivalent. All experimental procedures were approved by the Animal Ethics Committee of Ocean University of China (Approved protocol no: SPXY2017050402).
Thirty-two male, 5-week-old spontaneously hypertensive rats (SHR) and 12 Kyoto rats (WKY) were purchased from Beijing Viton Lever Laboratory Animal Technology Co. (Beijing, China). After 4 weeks of acclimatization feeding, 32 SHR rats were randomly divided into 4 groups ($$n = 8$$) and 12 WKY rats were randomly divided into 2 groups ($$n = 6$$), according to body weight and blood pressure. The WKY group (gavaged with 3 mL of sterilized saline NaCl $0.8\%$), WKY C. butyricum group (WKY-Cb) (3 mL of C. butyricum freeze-dried powder dissolved in sterilized normal saline (108 CFU/mL) was administered to the rats), SHR group (gavaged with sterilized saline NaCl $0.8\%$), SHR *Clostridium butyricum* (SHR-C. butyricum) (3 mL of C. butyricum freeze-dried powder dissolved in sterilized normal saline (108 CFU/mL) was administered to the rats), SHR butyrate group (0.5 mg/kg), and SHR captopril group (SHR-CAP) (captopril 10 mg/kg, Changzhou Pharmaceutical factory). All groups were gavaged once a day. Rats were housed in individual ventilated cages, in a pathogen-free animal facility under temperature (20–22 °C) and humidity (50–$60\%$) control, and with a pre-set light-dark cycle (12 h:12 h). During the experimental period, the rats had free access to tap water and feed. All rats were executed after 6 weeks, and the intestinal contents, spleen, aorta, and serum were collected and stored at −80 °C until use.
## 4.3. Blood Pressure Measurements
In a consistent environment, to reduce disturbances for blood pressure measurement of the rats, 44 rats were acclimatized and fed for 4 weeks before being prepared for grouping.
The SBP was measured in unanesthetized rats using the CODA volume-pressure relationship tail-cuff system (Kent Scientific Corporation, Muscatine, IA, USA). Blood pressure was measured at regular intervals, every 2 weeks, during the treatment period, consecutive measurements were taken for each one, and the blood pressure data were averaged for each measurement.
## 4.4. Quantitative Real-Time PCR of Colonic Tissue
Gene expression in colonic tissue was measured by quantitative real-time PCR with SYBR Green. Quantitative real-time PCR was performed on a Thermo Lifetech ABI QuantStudio 3 from Applied Biosystems (Applied Biosystems, Thermo Fisher, Waltham, MA, USA). All amplification reactions were carried out in 96-well optical-grade PCR plates in triplicate (Applied Biosystems, Thermo Fisher, USA), each with 20 μL, sealed with optical sealing tape (Ruibiotech, Qingdao, China). The primers were designed with the GenBank database or DNAMAN for Windows, and synthesized commercially by Ruibiotech. All primers used are shown in Table S4.
## 4.5. DNA Extraction and 16S rRNA Gene Sequencing
The TIANGEN Bacterial Genomic DNA Extraction Kit (DP302-02, TIANGEN, Beijing, China) was used to extract colonic fecal genomic DNA. Colonic feces were collected under aseptic conditions and stored at −80 °C until subsequent analysis. Genomic DNA from the colonic fecal samples was extracted using the TIANGEN Fecal Genomic DNA Extraction Kit (DP328-02, TIANGEN, Beijing, China) according to the manufacturer’s instructions. A NanoDrop spectrophotometer ND 3.0 1000 (NanoDrop Technologies, Wilmington, DE, USA) was used to quantify the concentration of extracted DNA. The purity of the extracted DNA was checked by $1\%$ agarose gel electrophoresis.
PacBio 16S rRNA V1–V9, sequenced and analyzed on the Pacbio Sequel II sequencing platform, was performed by Biozeron (Lingen, Shanghai, China). The recovered purified PCR products were detected and quantified by a QuantiFluor™-ST Blue Fluorescence Quantification System (Promega, Madison, WI, USA), then mixed in the appropriate proportions according to the sequencing volume required for each sample, and analyzed in PacBio libraries.
## 4.6. Bioinformatic Analysis of Sequencing Data
The raw data from PacBio were processed using the SMRT analysis software, version 9.0, to obtain demultiplexed circular consensus sequence (CCS) reads. OUT clustering was performed using UPARSE (version 7.1), based on $98.65\%$ similarity (http://drive5.com/uparse/ (accessed on 12 February 2022)), and chimeric sequences were identified and removed by UCHIME. The phylogeny of each 16S rRNA gene sequence was analyzed by the RDP classifier (http://rdp.cme.msu.edu/ (accessed on 18 February 2022)) against the Silva (SSU132) 16S rRNA database, multiple diversity index analysis based on OUT data, and statistical analysis of community structure.
## 4.7. Enzyme-Linked Immunosorbent Assay (ELISA)
The levels of IL-6, TNF-α, IL-10, IL-17A, and LPS in serum were measured using ELISA kits (Lianke, Hangzhou, China). All procedures were performed according to the steps of the kits’ instructions. Plasma was centrifuged at 3500 rpm for 15 min and three duplicate samples were prepared for each group. An enzyme marker was used to collect the fluorescence intensity. After removing background and normalization, the concentration of each cytokine in the sample was calculated from the standard curve.
## 4.8. Histological Analysis of Colon, Spleen and Aortic Tissue
A suitably sized colonic, splenic, and aortic ring was fixed in $10\%$ (v/v%) neutral buffered formalin, then washed in running water, dehydrated in alcohol, cleaned in xylene, and treated with paraffin. Thin tissue sections (3–5 μm) were then rehydrated and stained with hematoxylin and eosin (H&E) [56]. After staining, the dried slides were photographed and preserved using a microscope (BX41, Olympus Corporation, Tokyo, Japan) to observe the state of the transverse colon, spleen, and aorta.
## 4.9. Immunofluorescence Staining of Aorta and Spleen
Immunofluorescence assays were performed by Servicebio (Wuhan, China). Aortic slides were dewaxed, hydrated, and then subjected to microwave antigen repair in ethylenediaminetetraacetic acid buffer (pH 9.0). After serum closure ($4\%$ goat serum for 40 min), sections were incubated with the following primary antibody combinations: anti-CD4 and anti-IL-10, anti-CD4, and anti-IL-17, diluted overnight at 4 °C at 1:100. After incubation with the primary antibodies, the slides were incubated with the secondary antibodies for 50 min, protected from light. In addition, the slides were stained with a 1:200 dilution of 4′-6-diamidino-2-phenylindole (DAPI) solution in the dark for 10 min. Finally, the stained cells were observed by fluorescence microscopy (BX41, Olympus Corporation, Tokyo, Japan) and images were collected.
## 4.10. Statistical Analysis
The Graphpad prism 8.0 software was used for graphing and the SPSS 22.0 software was used for data analysis. Student’s t-tests or one-way ANOVA was used to analyze the data between groups. Two-way repeated-measures ANOVA was used when analyzing blood pressure at multiple time points. $p \leq 0.05$ indicates a significant difference. Line and bar plot data are expressed as mean ± SEM.
## References
1. Sender R., Fuchs S., Milo R.. **Revised Estimates for the Number of Human and Bacteria Cells in the Body**. *PLoS Biol.* (2016) **14**. DOI: 10.1371/journal.pbio.1002533
2. Spychala M.S., Venna V.R., Jandzinski M., Doran S.J., Durgan D.J., Ganesh B.P., Ajami N.J., Putluri N., Graf J., Bryan R.M.. **Age-related changes in the gut microbiota influence systemic inflammation and stroke outcome**. *Ann. Neurol.* (2018) **84** 23-36. DOI: 10.1002/ana.25250
3. Tremaroli V., Bäckhed F.. **Functional interactions between the gut microbiota and host metabolism**. *Nature* (2012) **489** 242-249. DOI: 10.1038/nature11552
4. Robles-Vera I., Toral M., Romero M., Jiménez R., Sánchez M., Pérez-Vizcaíno F., Duarte J.. **Antihypertensive Effects of Probiotics**. *Curr. Hypertens. Rep.* (2017) **19** 26. DOI: 10.1007/s11906-017-0723-4
5. Gomez-Arango L.F., Barrett H.L., McIntyre H.D., Callaway L.K., Morrison M., Nitert M.D., Grp S.T.. **Increased Systolic and Diastolic Blood Pressure Is Associated with Altered Gut Microbiota Composition and Butyrate Production in Early Pregnancy**. *Hypertension* (2016) **68** 974-981. DOI: 10.1161/HYPERTENSIONAHA.116.07910
6. Kim S., Goel R., Kumar A., Qi Y.F., Lobaton G., Hosaka K., Mohammed M., Handberg E.M., Richards E.M., Pepine C.J.. **Imbalance of gut microbiome and intestinal epithelial barrier dysfunction in patients with high blood pressure**. *Clin. Sci.* (2018) **132** 701-718. DOI: 10.1042/CS20180087
7. Li J., Zhao F., Wang Y., Chen J., Tao J., Tian G., Wu S., Liu W., Cui Q., Geng B.. **Gut microbiota dysbiosis contributes to the development of hypertension**. *Microbiome* (2017) **5** 14. DOI: 10.1186/s40168-016-0222-x
8. Cummings J.H., Pomare E.W., Branch W.J., Naylor C.P., Macfarlane G.T.. **Short chain fatty acids in human large intestine, portal, hepatic and venous blood**. *Gut* (1987) **28** 1221-1227. DOI: 10.1136/gut.28.10.1221
9. den Besten G., Bleeker A., Gerding A., van Eunen K., Havinga R., van Dijk T.H., Oosterveer M.H., Jonker J.W., Groen A.K., Reijngoud D.-J.. **Short-Chain Fatty Acids Protect Against High-Fat Diet–Induced Obesity via a PPARγ-Dependent Switch from Lipogenesis to Fat Oxidation**. *Diabetes* (2015) **64** 2398-2408. DOI: 10.2337/db14-1213
10. Hu J., Lin S., Zheng B., Cheung P.C.K.. **Short-chain fatty acids in control of energy metabolism**. *Crit. Rev. Food Sci. Nutr.* (2018) **58** 1243-1249. DOI: 10.1080/10408398.2016.1245650
11. Sivaprakasam S., Bhutia Y.D., Yang S., Ganapathy V.. **Short-Chain Fatty Acid Transporters: Role in Colonic Homeostasis**. *Compr. Physiol.* (2017) **8** 299-314. DOI: 10.1002/cphy.c170014
12. Priyadarshini M., Kotlo K.U., Dudeja P.K., Layden B.T.. **Role of Short Chain Fatty Acid Receptors in Intestinal Physiology and Pathophysiology**. *Compr. Physiol.* (2018) **8** 1091-1115. DOI: 10.1002/cphy.c170050
13. Onyszkiewicz M., Gawrys-Kopczynska M., Konopelski P., Aleksandrowicz M., Sawicka A., Kozniewska E., Samborowska E., Ufnal M.. **Butyric acid, a gut bacteria metabolite, lowers arterial blood pressure via colon-vagus nerve signaling and GPR41/43 receptors**. *Pflug. Arch. Eur. J. Physiol.* (2019) **471** 1441-1453. DOI: 10.1007/s00424-019-02322-y
14. Pluznick J.L., Protzko R.J., Gevorgyan H., Peterlin Z., Sipos A., Han J., Brunet I., Wan L.X., Rey F., Wang T.. **Olfactory receptor responding to gut microbiota-derived signals plays a role in renin secretion and blood pressure regulation**. *Proc. Natl. Acad. Sci. USA* (2013) **110** 4410-4415. DOI: 10.1073/pnas.1215927110
15. Robles-Vera I., Toral M., de la Visitación N., Sánchez M., Gómez-Guzmán M., Romero M., Yang T., Izquierdo-Garcia J.L., Jiménez R., Ruiz-Cabello J.. **Probiotics Prevent Dysbiosis and the Rise in Blood Pressure in Genetic Hypertension: Role of Short-Chain Fatty Acids**. *Mol. Nutr. Food Res.* (2020) **64** 1900616. DOI: 10.1002/mnfr.201900616
16. Ba D., Takeichi N., Kobayashi H.. **Restoration of T-Cell Depression and Suppression of Blood Pressure in Spontaneously Hypertensive Rats (SHR) by Allogeneic Thymus Grafts or Injection of Thymic Hormone**. *Jpn. Heart J.* (1981) **22** 494. DOI: 10.1536/ihj.22.494
17. Pascual V.H., Oparil S., Eldridge J.H., Jin H., Bost K.L., Pascual D.W.. **Spontaneously hypertensive rat: Lymphoid depression is age dependent and mediated via a mononuclear cell subpopulation**. *Am. J. Physiol.* (1992) **262** R1-R7. DOI: 10.1152/ajpregu.1992.262.1.R1
18. Takeichi N., Suzuki K., Okayasu T., Kobayashi H.. **Immunological depression in spontaneously hypertensive rats**. *Clin. Exp. Immunol.* (1980) **40** 120-126. DOI: 10.1536/ihj.21.590
19. Adnan S., Nelson J.W., Ajami N.J., Venna V.R., Petrosino J.F., Bryan R.M.B., Durgan D.J.. **Alterations in the gut microbiota can elicit hypertension in rats**. *Physiol. Genom.* (2017) **49** 96-104. DOI: 10.1152/physiolgenomics.00081.2016
20. Yang T., Santisteban M.M., Rodriguez V., Li E., Ahmari N., Carvajal J.M., Zadeh M., Gong M., Qi Y., Zubcevic J.. **Gut dysbiosis is linked to hypertension**. *Hypertension* (2015) **65** 1331-1340. DOI: 10.1161/HYPERTENSIONAHA.115.05315
21. Duncan S.. **Roseburia intestinalis sp. nov., a novel saccharolytic, butyrate-producing bacterium from human faeces**. *Int. J. Syst. Evol. Microbiol.* (2002) **52** 1615-1620. DOI: 10.1099/00207713-52-5-1615
22. Seo M., Inoue I., Tanaka M., Matsuda N., Nakano T., Awata T., Katayama S., Alpers D.H., Komoda T.. *Dig. Dis. Sci.* (2013) **58** 3534-3544. DOI: 10.1007/s10620-013-2879-3
23. Zhang J., Chen X.Y., Liu P., Zhao J.B., Sun J., Guan W.Y., Johnston L.J., Leyesque C.L., Fan P.X., He T.. **Dietary**. *J. Agric. Food Chem.* (2018) **66** 5157-5166. DOI: 10.1021/acs.jafc.8b01253
24. Wang S., Zhang C., Yang J.C., Wang X., Wu K.T., Zhang B.Y., Zhang J.C., Yang A., Rajput S.A., Qi D.S.. **Sodium Butyrate Protects the Intestinal Barrier by Modulating Intestinal Host Defense Peptide Expression and Gut Microbiota after a Challenge with Deoxynivalenol in Weaned Piglets**. *J. Agric. Food Chem.* (2020) **68** 4515-4527. DOI: 10.1021/acs.jafc.0c00791
25. Arumugam M., Raes J., Pelletier E., Le Paslier D., Yamada T., Mende D.R., Fernandes G.R., Tap J., Bruls T., Batto J.-M.. **Enterotypes of the human gut microbiome**. *Nature* (2011) **473** 174-180. DOI: 10.1038/nature09944
26. Rinninella E., Raoul P., Cintoni M., Franceschi F., Miggiano G.A.D., Gasbarrini A., Mele M.C.. **What is the Healthy Gut Microbiota Composition? A Changing Ecosystem across Age, Environment, Diet, and Diseases**. *Microorganisms* (2019) **7**. DOI: 10.3390/microorganisms7010014
27. Magne F., Gotteland M., Gauthier L., Zazueta A., Pesoa S., Navarrete P., Balamurugan R.. **The**. *Nutrients* (2020) **12**. DOI: 10.3390/nu12051474
28. Sun J., Xu J., Yang B., Chen K., Kong Y., Fang N., Gong T., Wang F., Ling Z., Liu J.. **Effect of**. *Mol. Nutr. Food Res.* (2020) **64** 1900636. DOI: 10.1002/mnfr.201900636
29. Grimm V., Radulovic K., Riedel C.U.. **Colonization of C57BL/6 Mice by a Potential Probiotic**. *PLoS ONE* (2015) **10**. DOI: 10.1371/journal.pone.0139935
30. Roshanravan N., Mahdavi R., Alizadeh E., Ghavami A., Saadat Y.R., Alamdari N.M., Alipour S., Dastouri M.R., Ostadrahimi A.. **The effects of sodium butyrate and inulin supplementation on angiotensin signaling pathway via promotion of**. *J. Cardiovasc. Thorac. Res.* (2017) **9** 183-190. DOI: 10.15171/jcvtr.2017.32
31. den Hartigh L.J., Gao Z., Goodspeed L., Wang S., Das A.K., Burant C.F., Chait A., Blaser M.J.. **Obese Mice Losing Weight Due to trans-10,cis-12 Conjugated Linoleic Acid Supplementation or Food Restriction Harbor Distinct Gut Microbiota**. *J. Nutr.* (2018) **148** 562-572. DOI: 10.1093/jn/nxy011
32. Omar J.M., Chan Y.-M., Jones M.L., Prakash S., Jones P.J.H.. *J. Funct. Foods* (2013) **5** 116-123. DOI: 10.1016/j.jff.2012.09.001
33. Wright E.M.. **Glucose transport families SLC5 and SLC50**. *Mol. Asp. Med.* (2013) **34** 183-196. DOI: 10.1016/j.mam.2012.11.002
34. Ang Z.W., Ding J.L.. **GPR41 and GPR43 in Obesity and inflammation—Protective or Causative?**. *Front. Immunol.* (2016) **7** 28. DOI: 10.3389/fimmu.2016.00028
35. Huart J., Leenders J., Taminiau B., Descy J., Saint-Remy A., Daube G., Krzesinski J.M., Melin P., de Tullio P., Jouret F.. **Gut Microbiota and Fecal Levels of Short-Chain Fatty Acids Differ Upon 24-Hour Blood Pressure Levels in Men**. *Hypertension* (2019) **74** 1005-1013. DOI: 10.1161/HYPERTENSIONAHA.118.12588
36. Karlsson F.H., Fåk F., Nookaew I., Tremaroli V., Fagerberg B., Petranovic D., Bäckhed F., Nielsen J.. **Symptomatic atherosclerosis is associated with an altered gut metagenome**. *Nat. Commun.* (2012) **3** 1245. DOI: 10.1038/ncomms2266
37. Luo X., Kong Q., Wang Y.-m., Duan X., Wang P., Li C., Huan Y.. **Colonization of**. *Microorganisms* (2021) **9**. DOI: 10.3390/microorganisms9081573
38. Durgan D.J., Ganesh B.P., Cope J.L., Ajami N.J., Phillips S.C., Petrosino J.F., Hollister E.B., Bryan R.M.. **Role of the Gut Microbiome in Obstructive Sleep Apnea-Induced Hypertension**. *Hypertension* (2016) **67** 469-474. DOI: 10.1161/HYPERTENSIONAHA.115.06672
39. Ganesh B.P., Nelson J.W., Eskew J.R., Ganesan A., Ajami N.J., Petrosino J.F., Bryan R.M., Durgan D.J.. **Prebiotics, Probiotics, and Acetate Supplementation Prevent Hypertension in a Model of Obstructive Sleep Apnea**. *Hypertension* (2018) **72** 1141-1150. DOI: 10.1161/HYPERTENSIONAHA.118.11695
40. Huang K., Wang M.M., Kulinich A., Yao H.L., Ma H.Y., Martínez J.E., Duan X.C., Chen H., Cai Z.P., Flitsch S.L.. **Biochemical characterisation of the neuraminidase pool of the human gut symbiont**. *Carbohydr. Res.* (2015) **415** 60-65. DOI: 10.1016/j.carres.2015.08.001
41. Xu Y., Wang N., Tan H.-Y., Li S., Zhang C., Feng Y.. **Function of**. *Front. Microbiol.* (2020) **11** 219. DOI: 10.3389/fmicb.2020.00219
42. Peng L., Li Z.R., Green R.S., Holzman I.R., Lin J.. **Butyrate enhances the intestinal barrier by facilitating tight junction assembly via activation of AMP-activated protein kinase in Caco-2 cell monolayers**. *J. Nutr.* (2009) **139** 1619-1625. DOI: 10.3945/jn.109.104638
43. Gaudier E., Rival M., Buisine M.P., Robineau I., Hoebler C.. **Butyrate enemas upregulate Muc genes expression but decrease adherent mucus thickness in mice colon**. *Physiol. Res.* (2009) **58** 111-119. DOI: 10.33549/physiolres.931271
44. Lewis K., Lutgendorff F., Phan V., Söderholm J.D., Sherman P.M., McKay D.M.. **Enhanced translocation of bacteria across metabolically stressed epithelia is reduced by butyrate**. *Inflamm. Bowel Dis.* (2010) **16** 1138-1148. DOI: 10.1002/ibd.21177
45. Lee W.J., Hase K.. **Gut microbiota-generated metabolites in animal health and disease**. *Nat. Chem. Biol.* (2014) **10** 416-424. DOI: 10.1038/nchembio.1535
46. Peti-Peterdi J., Kishore B.K., Pluznick J.L.. **Regulation of Vascular and Renal Function by Metabolite Receptors**. *Annu. Rev. Physiol.* (2016) **78** 391-414. DOI: 10.1146/annurev-physiol-021115-105403
47. Pluznick J.L.. **Extra sensory perception: The role of Gpr receptors in the kidney**. *Curr. Opin. Nephrol. Hypertens.* (2014) **23** 507-512. DOI: 10.1097/MNH.0000000000000048
48. Tan J., McKenzie C., Potamitis M., Thorburn A.N., Mackay C.R., Macia L.. **The role of short-chain fatty acids in health and disease**. *Adv. Immunol.* (2014) **121** 91-119. DOI: 10.1016/b978-0-12-800100-4.00003-9
49. Nicholson J.K., Holmes E., Kinross J., Burcelin R., Gibson G., Jia W., Pettersson S.. **Host-gut microbiota metabolic interactions**. *Science* (2012) **336** 1262-1267. DOI: 10.1126/science.1223813
50. Santisteban M.M., Qi Y.F., Zubcevic J., Kim S., Yang T., Shenoy V., Cole-Jeffrey C.T., Lobaton G.O., Stewart D.C., Rubiano A.. **Hypertension-Linked Pathophysiological Alterations in the Gut**. *Circ. Res.* (2017) **120** 312-323. DOI: 10.1161/CIRCRESAHA.116.309006
51. Farhadi A., Gundlapalli S., Shaikh M., Frantzides C., Harrell L., Kwasny M.M., Keshavarzian A.. **Susceptibility to gut leakiness: A possible mechanism for endotoxaemia in non-alcoholic steatohepatitis**. *Liver Int. Off. J. Int. Assoc. Study Liver* (2008) **28** 1026-1033. DOI: 10.1111/j.1478-3231.2008.01723.x
52. Kanai T., Mikami Y., Hayashi A.. **A breakthrough in probiotics: Clostridium butyricum regulates gut homeostasis and anti-inflammatory response in inflammatory bowel disease**. *J. Gastroenterol.* (2015) **50** 928-939. DOI: 10.1007/s00535-015-1084-x
53. Hayashi A., Sato T., Kamada N., Mikami Y., Matsuoka K., Hisamatsu T., Hibi T., Roers A., Yagita H., Ohteki T.. **A Single Strain of Clostridium butyricum Induces Intestinal IL-10-Producing Macrophages to Suppress Acute Experimental Colitis in Mice**. *Cell Host Microbe* (2013) **13** 711-722. DOI: 10.1016/j.chom.2013.05.013
54. Sender V., Stamme C.. **Lung cell-specific modulation of LPS-induced TLR4 receptor and adaptor localization**. *Commun. Integr. Biol.* (2014) **7** e29053. DOI: 10.4161/cib.29053
55. Sakaguchi S., Miyara M., Costantino C.M., Hafler D.A.. **FOXP3+ regulatory T cells in the human immune system**. *Nat. Rev. Immunol.* (2010) **10** 490-500. DOI: 10.1038/nri2785
56. Feldman A.T., Wolfe D., Day C.E.. **Tissue Processing and Hematoxylin and Eosin Staining**. *Histopathology: Methods and Protocols* (2014) 31-43
|
---
title: Amorphous System of Hesperetin and Piperine—Improvement of Apparent Solubility,
Permeability, and Biological Activities
authors:
- Kamil Wdowiak
- Andrzej Miklaszewski
- Robert Pietrzak
- Judyta Cielecka-Piontek
journal: International Journal of Molecular Sciences
year: 2023
pmcid: PMC10002548
doi: 10.3390/ijms24054859
license: CC BY 4.0
---
# Amorphous System of Hesperetin and Piperine—Improvement of Apparent Solubility, Permeability, and Biological Activities
## Abstract
The low bioaccessibility of hesperetin and piperine hampers their application as therapeutic agents. Piperine has the ability to improve the bioavailability of many compounds when co-administered. The aim of this paper was to prepare and characterize the amorphous dispersions of hesperetin and piperine, which could help to improve solubility and boost the bioavailability of both plant-origin active compounds. The amorphous systems were successfully obtained by means of ball milling, as confirmed by XRPD and DSC studies. What’s more, the FT-IR-ATR study was used to investigate the presence of intermolecular interactions between the systems’ components. Amorphization enhanced the dissolution rate as a supersaturation state was reached, as well as improving the apparent solubility of both compounds by 245-fold and 183-fold, respectively, for hesperetin and piperine. In the in vitro permeability studies simulating gastrointestinal tract and blood-brain barrier permeabilities, these increased by 775-fold and 257-fold for hesperetin, whereas they were 68-fold and 66-fold for piperine in the GIT and BBB PAMPA models, respectively. Enhanced solubility had an advantageous impact on antioxidant as well as anti-butyrylcholinesterase activities—the best system inhibited 90.62 ± $0.58\%$ of DPPH radicals and 87.57 ± $1.02\%$ butyrylcholinesterase activity. To sum up, amorphization considerably improved the dissolution rate, apparent solubility, permeability, and biological activities of hesperetin and piperine.
## 1. Introduction
Poor bioaccessibility and solubility are important factors in limiting the overall bioavailability of many compounds, including active substances derived from plants. Furthermore, in the case of polyphenolic natural compounds, intestinal protein efflux and cytochrome P450 metabolism are crucial factors to consider [1]. Hesperetin is a plant-derived compound that shows significant potential for preventing and supporting the treatment of chronic diseases. Its antioxidant, anti-inflammatory, anti-diabetic, anti-cancer, and neuroprotective activities have been well documented [2,3,4,5,6,7]. Hesperetin is found in substantial amounts in citrus fruits such as sweet oranges (*Citrus sinensis* L.) Osbeck and lemons (Citrus limon L.) Burm [2]. However, its role as a preventive or therapeutic agent is limited by its poor solubility, which translates into limited bioavailability. To date, several methods to improve hesperetin solubility have been described. Stahr et al prepared nanocrystals of hesperetin, which resulted in an enhancement in the dissolution rate as well as apparent solubility [8]. They observed a rapid increase in apparent solubility just after the start of the dissolution process, the so-called “spring effect,” without the parachute phenomenon, meaning that there was a high degree of increase in solubility at the beginning and a decrease in the amount of dissolved hesperetin over time [8]. Next, Wang et al fabricated self-assembling rebaudioside A nanomicelles with hesperetin, which considerably increased solubility and provided a sustained release. The systems were also characterized by enhanced biological potential with regard to anticancer activity [9]. While Gu et al produced hesperetin micelles with D-α-tocopheryl polyethylene glycol succinate and phosphatidylcholine that increased solubility by 21.5-fold, boosted antioxidant activity, and enhanced bioavailability by about 16.2-fold, whereas phosphatidylcholine enhanced solubility, antioxidant potential as well as bioavailability by 20.7-fold, 3.9-fold and 18-fold, respectively [10]. Trendafilova et al developed systems of hesperetin with Mg- and Ag-modified SBA-16 carriers, which translated into a higher apparent solubility and dissolution rate [11].
Piperine is an alkaloid of natural origin found in black pepper (*Piper nigrum* L.). It exhibits a number of health-promoting properties, such as anti-inflammatory, anti-diabetic, anti-cancer, and neuroprotective [12,13,14,15]. It is characterized by poor solubility, so researchers have tried a number of approaches to fight this issue. Zafar et al fabricated a self-nanoemulsifying drug delivery system with piperine. This approach improved its dissolution profile and permeability. It also contributed to better bioavailability in vivo as well as therapeutic efficacy such as anti-hypertensive, antibacterial, and antioxidant activities [16]. In turn, Zaini et al prepared piperine-succinic acid cocrystals, which were characterized by improved solubility and dissolution rate [17]. Ren et al developed piperine-loaded nanoparticles that exhibited enhanced dissolution rate, oral bioavailability, and brain delivery. Moreover, these nanoparticles showed great anti-epileptic activity [18]. Imam et al used solvent evaporation and microwave irradiation to create binary and ternary complexes of piperine with hydroxypropyl-β-cyclodextrin and D-α-tocopheryl polyethylene glycol succinate. Ternary complexes considerably improved piperine solubility and dissolution rate [19].
Amorphization is one of the promising techniques for improving the solubility of active compounds. Disruption of the crystalline structure and obtaining an amorphous state significantly improve the solubility and dissolution rate of compounds [20,21]. Taking into account the potential benefits of the combination of hesperetin with piperine and the limitations in the application of both compounds resulting from their poor solubility, the subject of this study was the development of an amorphous system containing hesperetin and piperine. Combining hesperetin with piperine could be beneficial in terms of bioavailability since the co-administration of poorly soluble compounds and bioenhancing substances, like piperine, is supposed to boost the bioavailability. A similar idea was proposed by Liu et al.; however, their approach involved obtaining hesperetin-piperine co-crystals, and as a result of the improved dissolution of both tested compounds after their introduction into the crystals, their concentration in the blood was increased [22].
The goal of this study was to create amorphous hesperetin-piperine dispersions with improved solubility for both compounds. The following stages of the research included [1] amorphous system preparation via ball milling, [2] amorphous system identification and characterization of physicochemical properties of its components, such as dissolution rate and permeability, and [3] evaluation of biological activity—antioxidant and neuroprotective activities.
## 2. Results
The systems were prepared by ball milling at various mass ratios using a vinylpyrrolidone-vinyl acetate copolymer in a 6:4 ratio (known as Kollidon VA64 or PVP VA64) as the carrier to prevent crystallization during the dissolution process and to ensure supersaturation was maintained. After confirming the amorphization, tests were carried out to characterize the physicochemical properties such as dissolution rate, solubility, and permeability. Then, it was checked whether the amorphization had an impact on the biological potential of the systems.
## 2.1.1. X-ray Powder Diffraction (XRPD)
XRPD analysis gave some information on changes in solid-state form due to the milling process (Figure 1). In the case of unmodified compounds, the diffractograms are characterized by well-defined sharp peaks, thus indicating their crystalline nature. For hesperetin, peaks are seen at 7.33, 14.23, 14.64, 15.66, 17.10, 17.82, 20.22, 21.08, 21.99, 22.72, 23.16, 23.68, 25.07, 25.64, 26.39, 27.82, 28.65, 29.63° 2 Theta whereas for piperine they are visible at 13.01, 14.25, 14.84, 15.66, 16.02, 16.87, 19.35, 19.76, 20.65, 21.43, 22.37, 22.63, 24.25, 24.50, 25.34, 25.89, 28.34, 29.88, 31.96° 2 Theta. In the XRPD pattern of the obtained systems, a “halo effect” appeared, indicating that they are in an amorphous state. XRPD analysis of physical mixtures shows a pattern characteristic for amorphous materials and noticeable peaks suggesting the presence of crystalline material (Supplementary Materials Figure S1).
## 2.1.2. Differential Scanning Calorimetry (DSC)
To further confirm the XRPD study results as well as observe phase transitions, the DSC study was performed. Hesperetin in the first heating run shows a sharp endothermic peak at 234.1 °C, whereas piperine is at 133.0 °C. These thermal events correspond to the melting points of each compound. In the second heating of raw compounds, one can notice glass transitions of amorphous form, Tg = 76.8 °C and Tg = 53.9 °C for hesperetin and piperine, respectively. In the DSC patterns of amorphous systems, there is a lack of endothermic peaks corresponding to the melting points of individual components, which confirms the amorphization (Figure 2). Moreover, the single-glass transition is noticeable, suggesting good miscibility and molecular dispersion of components in amorphous systems. It’s worth noting that as the amount of polymer in a system increases, the glass transition shifts to higher values, which could indicate that the polymer provides better molecular mobility restriction.
## 2.1.3. Fourier Transform Infrared Spectroscopy with Attenuated Total Reflectance (FTIR-ATR)
To investigate the presence of intermolecular interactions in the systems, FT-IR analysis was performed. Changes in FT-IR spectra, such as shift, disappearance, and reduction in intensity, as well as the appearance of new peaks, are strong indicators of interaction formation.
Each component of the system has various possible sites for hydrogen bond formation. Hesperetin is rich in hydroxyl groups, meaning it will most likely participate in hydrogen bonding as a proton donor. Piperine possesses a potent hydrogen bond acceptor, which is the oxygen atom of the amide group, as well as a proton donor, i.e., hydrogen, in the methylendioxyphenyl group [23,24]. Kollidon VA 64 has two hydrogen bond acceptor groups, which are the carbonyl group and vinyl acetate [25].
When comparing raw compounds in crystalline and amorphous forms, clear changes in FT-IR spectra are evident (Figure 3a). In the case of hesperetin, the crystalline form shows a sharp peak at 3495 cm−1 and a broad peak in the range 2839–3190 cm−1, which, in the case of the amorphous form, merged into a single peak with a broad base and an absorption maximum at 3351 cm−1. In addition, the peaks at 1260 and 1240 cm−1 merged to give a peak at 1268 cm−1. A shift of the peak at 1576 to 1588 cm−1 and the peak at 1091 to 1084 cm−1 can also be seen. Moreover, the peaks at 1611, 1402, 1359, 1305, and 1203 cm−1 disappeared and are not visible in the spectrum of the amorphous form of hesperetin. In the 900–400 cm−1 range, a significant reduction in intensity and broadening of the peaks can be seen. As for piperine, the broad peak with several maxima at 2942, 2919, 2863, and 2849 cm−1 merged into one broad peak with two absorption maxima at 2934 and 2854 cm−1. The peak with three distinct maxima at 1634, 1609, and 1582 cm−1 merged into one, yielding a maximum at 1626 cm−1. The base of the peaks at 1512, 1490, and 1433 cm−1 broadened, causing these peaks to merge but retain their maxima. The peaks at 1249, 1227, and 1192 cm−1 merged into one, with the maximum at 1246 cm−1. The same is true of the peaks at 1029, 1018, and 994 cm−1; these merged into one with the maximum at 1035 cm−1. The peak at 845 shifted to 853 cm−1, the peak at 829 cm−1 disappeared, and the peak at 786 merged with the one at 802 cm−1. In the 750–400 cm−1 range, reductions in intensity and broadening of the bases of the peaks can be seen. The FT-IR/ATR spectra of raw compounds with main peaks description were placed in Supplementary Materials (Figure S2).
In the case of the spectra of the systems (Figure 3b), the individual bands can be assigned to the bands of the individual components of the systems, except for the bands at 1164 and 1187 cm−1, which may be hesperetin bands at 1150 and 1179 cm−1, respectively, which shifted in the systems. As the amount of polymer in the system increased, the bands originating from the polymer became dominant. In turn, the indicated shifts of the two peaks may suggest some interactions, but no changes are observed in the regions of special interest, i.e., at the bands of functional groups that may be involved in hydrogen bond formation, therefore they cannot be clearly confirmed with a high degree of certainty.
## 2.2.1. Dissolution Rate Studies
One of the advantages of amorphous dispersions is that they improve the dissolution rate profile. The dissolution profiles for both substances showed an increase in the level of apparent solubility compared to raw compounds, so the obtained amorphous dispersions ensured that a supersaturated state was obtained. In both cases, the amorphous systems reached the plateau state after about an hour, and it was maintained until the end of the test, i.e., 6 h. In the case of raw hesperetin, 23.75 ± $2.99\%$ dissolved after one hour, while raw piperine reached the level of 27.79 ± $3.38\%$ of the dose added to the dissolution medium (Figure 4). As for the amorphous systems, the best improvement in apparent solubility was provided by the hesperetin-piperine-VA64 system in a mass ratio of 1:1:16. It allowed the dissolution of 89.39 ± $1.48\%$ of hesperetin and 94.51 ± $1.97\%$ of piperine.
## 2.2.2. Solubility Study
Solubility studies were performed in phosphate buffers at pH 6.8. The solubility of raw compounds was determined to be as low as 0.005 ± 0.001 mg/mL and 0.006 ± 0.001 mg/mL for hesperetin and piperine, respectively. Solubility was significantly improved by using systems. The best amorphous systems turned out to be hesperetin-piperine-VA64 1:1:16. The compounds in this system reached concentrations above 1 mg/mL, giving 245-fold and 183-fold better solubility, with regards to raw compounds, for hesperetin and piperine, respectively. The results of the solubility studies are presented in Table 1.
## 2.2.3. Permeability Studies
In order to determine whether the obtained systems led to improved permeability, the parallel artificial membrane permeability assay (PAMPA) model was used to simulate passive diffusion through the cell walls of the gastrointestinal tract (GIT) and across the blood-brain barrier (BBB). To test if the model compounds themselves tend to passively permeate through the aforementioned barriers, an apparent permeability coefficient was determined. For PAMPA GIT, it was determined to be 2.59 × 10−6 ± 2.37 × 10−7 cm/s and 3.73 × 10−5 ± 3.33 × 10−6 cm/s for hesperetin and piperine, respectively. Based on this, it can be concluded that the tested compounds have good permeability across the intestinal barrier. In the case of PAMPA BBB, the calculated coefficients are 7.05 × 10−6 ± 3.02 × 10−6 cm/s for hesperetin and 4.00 × 10−5 ± 1.89 × 10−6 cm/s for piperine, respectively, which means that these compounds will also be able to cross the blood-brain barrier. To compare the permeability of the compounds in systems, the concentrations obtained in the acceptor parts were set side by side. Here, one can see a clear improvement in permeability in both PAMPA models. The concentration of raw hesperetin was determined to be as low as 2.58 × 10−5 ± 6.16 × 10−6 mg/mL in the PAMPA GIT model and 3.89 × 10−5 ± 1.66 × 10−5 mg/mL in the PAMPA BBB model. The best system enhanced permeability, enabling a concentration of 2.00 × 10−2 ± 2.00 × 10−3 mg/mL in the GIT model and 1.01 × 10−2 ± 5.02 × 10−4 mg/mL in the BBB model, meaning that these parameters increased by 775-fold and 257-fold, respectively. Similar observations can be reported for piperine. The concentration in the acceptor part increased from 2.01 × 10−3 ± 1.00 × 10−4 mg/mL to 1.36 × 10−1 ± 4.07 × 10−3 mg/mL in the GIT model and from 1.03 × 10−3 ± 1.04 × 10−4 mg/mL to 6.41 × 10−2 ± 2.00 × 10−3 mg/mL in the BBB model, which translated into an increase of 68- and 66-fold, respectively. The results of permeability studies (Figure 5) are related to what was observed in the solubility study. Both hesperetin and piperine are well permeable through biological barriers; therefore, their bioavailability is limited by poor solubility.
## 2.3. Biological Activity Studies
This biological activity research looked at the antioxidant (2, 2-diphenyl-1-picrylhydrazyl radical—DPPH radical) and butyrylcholinesterase (BChE) suppressing abilities. It is noticeable that the amorphization process and, as a consequence, improved solubility had an advantageous impact on biological activity. The best activity was shown by the Hes:Pip:VA 64 system at a mass ratio of 1:1:16, which reduced the DPPH radical at 90.62 ± $0.58\%$ and suppressed BChE activity at 87.53 ± $1.02\%$, whereas a physical mixture of raw compounds in a 1:1 mass ratio produced 2.51 ± $1.05\%$ and 2.12 ± $0.53\%$ of inhibition in the DPPH and BChE tests, respectively (Table 2).
## 3. Discussion
Hesperetin and piperine are plant origin active substances that have a vast pro-healthy potential when it comes to prophylactic as well as the treatment of various diseases. Both compounds possess well-documented neuroprotective activity involving antioxidant and anti-neuroinflammatory activity as well as inhibition of toxic protein aggregation [14,26]. However, their use in therapeutic strategies is limited due to their low bioavailability, which is connected to insufficient levels of solubility. This in turn causes inadequate blood concentrations, which results in poor pharmacological effects. This work aims to develop amorphous systems to address these issues. When designing the dispersions, the authors took advantage of the beneficial effect of compounds on boosting bioavailability.
In our research, an environmentally friendly approach to preparing amorphous systems was used. The applied ball milling technique enabled the formation of amorphous systems. This observation was supported by the X-ray powder diffraction (XRPD) study, which revealed an amorphous pattern, as well as the differential scanning calorimetry (DSC) study, which showed a lack of crystalline structure in the systems due to the disappearance of the endothermic peaks corresponding to melting points and the appearance of the glass transition. What’s more, Fourier transform infrared spectroscopy with attenuated total reflectance (FT-IR-ATR) analysis made it possible to investigate intermolecular interactions. No obvious suggestions of the formation of hydrogen-bond-like interactions can be seen.
The amorphization process improved bioaccessibility, which means there are more free compounds available for absorption. This may influence permeability, as these two factors are closely related [27]. It is assumed that flavanone aglycones—hesperetin—have limited bioaccessibility but high intestinal permeability [27], meaning that the only limiting factor, in this case, is solubility. The same observation applies to piperine, since it also possesses high absorption potential. Hesperetin and piperine may be considered II-class compounds in the Biopharmaceutical Classification System. For most compounds, passive diffusion is the primary mode of absorption [28]. As amorphous systems were produced and the supersaturated state was maintained, it can be concluded that the Kollidon VA64 used in this study prevented crystallization of the model compounds. Apparent solubility, in the dissolution rate study, increased by 3.9-fold and 3.4-fold for hesperetin and piperine, respectively. In the case of amorphous dispersions, this is not obvious, as there is a high risk of precipitation of active substances due to reaching concentrations above their crystalline solubility. In addition, the rate of release of active ingredients from amorphous dispersions was controlled by the carrier. The tendency is that the more polymer there is in the system and, thus, in the dissolution medium, the greater the apparent solubility. Similar observations were reported by Szafraniec-Szczęsny et al. [ 29]. In their study on ezetimibe in amorphous dispersion with Kollidon VA64 obtained by spray drying, they pointed out the dependence of the amount of polymer in the system on the amount of dissolved active substance. They also included amorphous ezetimibe without a polymer carrier in the study. Despite the amorphous state, the achieved amounts of dissolved active ingredient were close to those of the crystalline substance, which also highlights that the presence of polymer is significant in the overall performance of the amorphous form in the dissolution process.
The supersaturation state achieved by amorphous systems can directly translate into improved absorption of active substances, as only free, dissolved molecules are capable of penetrating biological barriers. Several studies show that the supersaturated state contributes to achieving higher drug concentrations in the blood [30,31,32,33,34]. Accordingly, it can be expected that the amorphous dispersions obtained would also enable increased absorption of hesperetin and piperine. What’s more, since the solubilities of the compounds increased markedly, i.e., 245-fold for hesperetin and 183-fold for piperine, the potential for bioavailability enhancement is great.
It is also important that the systems provide a state of supersaturation within a 6-h window, i.e., the approximate time that intestinal contents can reside in the duodenum, the main site of absorption in the body [35]. The use of amorphous dispersion does not necessarily immediately imply that the aforementioned supersaturation will be stable over time. The behavior of the active substances will depend on the carrier and the extent to which it prevents crystallization. Knopp et al., using their celecoxib study as an example, showed that the performance of amorphous solid dispersion is significantly affected by crystallization [36]. They showed that the crystallization that occurs is directly related to the decrease in AUC. Moreover, this is strongly influenced by the polymer used as a carrier. In their study, the authors used HPMC and PVP, with PVP being more effective in preventing crystallization.
The prepared amorphous systems were characterized by significantly improved passive permeability compared to raw compounds, which was proven by the PAMPA model both in terms of absorption in the gastrointestinal tract and overcoming the blood-brain barrier. For hesperetin, the enhancement was up to 775-fold in the intestinal model and 257-fold in blood-brain barrier model studies, whereas, in the case of piperine, it was 68-fold in intestinal model studies and 66-fold in blood-brain barrier model studies. However, the bioavailability of these compounds is also closely related to active transport, which may hinder reaching the desired blood concentrations. It is worth noting that the BCRP (breast cancer resistance protein) transporter may be considered one of the biggest factors hampering hesperetin bioavailability since it contributes to the efflux of hesperetin [37]. The inhibition of this protein by piperine could considerably improve bioavailability. In fact, Bi Xiaoli et al supplied some pieces of evidence supporting this statement. They demonstrated that co-administration of piperine with silybin increased the bioavailability of silybin via the inhibition of the efflux transporters, including multidrug resistance-associated protein 2 (MRP2) as well as BCRP [38]. Moreover, Denni Yu et al showed that the co-administration of piperine with another compound boosts its bioavailability. The fabricated co-amorphous system of ursolic acid and piperine enhanced dissolution by about 7-fold as well as AUC values for both components in the pharmacokinetic study [39]. We assume that in the case of our ternary amorphous systems, similar observations could be anticipated. What’s more, as in the obtained systems, piperine is in an amorphous state, so one can expect it to show even greater potential in suppressing the activity of efflux proteins, which might result in a better capability to enhance permeability by this mechanism.
Increased bioaccessibility can improve hesperetin’s overall bioavailability. Omidfar et al obtained nanophytosmoes of hesperetin, which enabled them to reach higher blood concentrations in vivo than those of raw hesperetin. The authors attributed it to the amorphous nature of the systems as well as reduced particle size. Additionally, the factors contributing to better bioavailability may also be increased permeability and protection from first-pass metabolism provided by phytosomes [40]. In this context, our systems may ensure similar benefits since amorphization leads to improved solubility and stimulates passive diffusion, whereas piperine is expected to inhibit efflux as well as first-pass metabolism. All in all, these factors may lead to improved bioavailability of both active components of the ternary amorphous system. Moreover, one cannot forget that piperine, an alkaloid compound, also possesses great biological potential, and its bioaccessibility is limited by its solubility. Obtained amorphous systems may improve the bioaccessibility of both plant origin active ingredients, therefore boosting their extraordinary biological potential, which is hampered by poor solubility.
Neurodegenerative diseases seem to be a growing issue, especially in aging populations [41]. It is well documented that oxidative stress plays a crucial role in the development of neurodegenerative diseases. Oxidative stress is involved in the progression of neuroinflammation as well as the stimulation of protein misfolding and aggregation. In addition, it has an important role in damaging intracellular structures. The above mechanisms are factors that trigger cell death processes leading to neuronal loss [42,43]. Cholinesterases are important molecular targets to consider. Both hesperetin and piperine have been shown to have esterase enzyme inhibitory activity [44,45]. These enzymes take part in the regulation of acetylcholine levels and are thus engaged in cholinergic transmission [46]. What’s more, the presence of BChE has been confirmed in amyloid plaques as well as neurofibril tangles, which suggests involvement in Alzheimer’s disease pathophysiology [47,48]. It was demonstrated that BChE might transform β-amyloid plaques from a benign to a malignant form [48]. Furthermore, BChE appears to play an important role in the development of multiple sclerosis. A rise in its activity is associated with an increase in inflammation [49]. The development of amorphous ternary systems of hesperetin and piperine resulted in suppression of the DPPH radical up to 90.62 ± $0.58\%$ as well as BChE activity to the extent of 87.53 ± $1.02\%$. These effects are probably related to the fact of improved solubility, since only free molecules could act on the DPPH radical or BChE enzyme and thus suppress their activity. In fact, Stahr et al prepared hesperetin nanocrystals, which showed potential in the treatment of Alzheimer’s disease [8]. In their study, higher apparent solubility was responsible for better anti-Alzheimer’s activity. This research supports the statement that enhanced solubility may contribute to increasing biological potential.
## 4.1. Materials
All materials including the tested compounds: hesperetin (purity > $95\%$) and piperine (purity > $95\%$, FG) were supplied by Sigma-Aldrich (Sigma-Aldrich, St. Louis, MO, USA), except for vinylpyrrolidone-vinyl acetate copolymer (Kollidon®VA64, PVP/VA, BASF, Ludwigshafen am Rhein, Germany), dimethyl sulfoxide (DMSO), sodium hydroxide (Avantor Performance Materials Poland S.A., Gliwice, Poland), acetic acid 98–$100\%$ (POCH, Gliwice, Poland), sodium dihydrogen phosphate (PanReac AppliChem ITW Reagents, Darmstadt, Germany), and methanol of an HPLC grade (J. T. Baker, Center Valley, PA, USA). High-quality pure water was prepared using a Direct-Q 3 UV purification system (Millipore, Molsheim, France; model Exil SA 67120). Prisma HT, GIT/BBB lipid solution, and acceptor sink buffer were supplied by Pion Inc. (Forest Row, East Sussex, UK).
## 4.2. Preparation of the Systems
The amorphous systems of Hes:Pip:VA 64 were obtained in different mass ratios, i.e., 1:1:4, 1:1:8, 1:1:12, and 1:1:16, by the means of ball milling (Retsch Mixer Mill MM 400). In short, 800 mg of the physical mixture was placed in a 25-mL stainless steel jar with 2 balls ø 10 mm, 3 balls ø 7 mm, and 3 balls ø 5 mm. The milling process was performed in six cycles, each lasting 20 min. The 5-min break was applied between cycles.
## 4.3.1. X-ray Powder Diffraction (XRPD)
The crystallographic structure of the samples were analyzed by an X-ray diffraction (XRD, Panalytical Empyrean, Almelo, Netherlands) equipment with the copper anode (CuKα—1.54 Å) at a Brag-Brentano reflection mode configuration with 45 kV and 40 mA parameters. The measurement parameters were set up for 3–60° with a 45-s step per 0.05° in all cases.
## 4.3.2. Differential Scanning Calorimetry (DSC)
Thermal analysis was performed using a DSC 214 Polyma differential scanning calorimeter (Netzsch, Selb, Germany). Samples of about 5–8 mg were placed in crimped aluminum pans with a small hole in the lid. First, the samples were heated up to 80 °C and kept at this temperature for 8 min to remove water from the samples, then they were cooled down to 25 °C and heated again to 280 °C. To measure the glass transition value of raw compounds, they were heated up to 280 °C, then cooled down and heated again to 280 °C. The measurements were performed at a constant heating rate of 20 °C/min under a nitrogen atmosphere with a flow rate of 30 mL/min. The glass transition value was taken as a midpoint between on-set and end-point temperatures.
## 4.3.3. Fourier Transform Infrared Spectroscopy with Attenuated Total Reflectance (FTIR-ATR)
The FTIR-ATR spectra were measured between 400 cm−1 and 4000 cm−1, with a resolution set to 1 cm−1, with a Shimadzu IRTracer-100 spectrometer equipped with a QATR-10 single bounce, a diamond extended range, and LabSolutions IR software (Warsaw, Poland). Amorphous forms of raw compounds were prepared by DSC by heating them to 280 °C and maintaining that temperature for 5 min.
## 4.4.1. HPLC Conditions
Concentrations of hesperetin and piperine during solubility, dissolution rate, and permeability studies were measured by high-performance liquid chromatography with the DAD detector (HPLC-DAD). In this study, a Shimadzu Nexera (Shimadzu Corp., Kyoto, Japan) equipped with an SCL-40 system controller, a DGU-403 degassing unit, a LC-40B XR solvent delivery module, a SIL-40C autosampler, a CTO-40C column oven, and a SPD-M40 photodiode array detector was used. For the stationary phase, a Dr. Maisch ReproSil-Pur Basic-C18 100 Å column with 5 µm particle size and 250 × 4.60 mm (Dr. Maisch, Ammerbuch-Entringen, Germany) was used. The mobile phase was methanol:$0.1\%$ acetic acid (80:20 v/v). The mobile phase was vacuum-filtered through a 0.45 µm nylon filter (Phenomenex, Torrance, CA, USA). The experimental conditions were as follows: 1.0 mL/min flow rate, wavelengths of 288 nm for hesperetin and 340 nm for piperine, and a column temperature of 30 °C. The injection volume differed depending on the assay. For the solubility study, it was 1 µL, whereas for the dissolution rate and permeability assays, it was 10 µL. The method’s duration was 10 min. The retention times were 4.11 min for hesperetin and 6.77 min for piperine. Chromatograms (Figure S3a,b) and validation parameters (Table S1) were placed in Supplementary Materials.
## 4.4.2. Media for Dissolution and Solubility Studies
Phosphate buffer at pH 6.8 was prepared according to the following description: in a 1000-mL volumetric flask, we placed 250 mL of 0.2 N potassium dihydrogen phosphate solution, then added 112 mL of 0.2 N sodium hydroxide solution, and filled the mixture up to 1000 mL with distilled water. High-quality pure water was prepared using a Direct-Q 3 UV purification system (Millipore, Molsheim, France, model Exil SA 67120).
## 4.4.3. Dissolution Studies
The dissolution study was performed on the paddle apparatus. The amount of the compound and systems corresponding to 7.0 mg of each plant-origin active ingredient was added to the dissolution medium. The vessels were filled with 500 mL of phosphate buffer, pH 6.8; the temperature was maintained at 37 °C, and the paddles were set at a stirring speed of 50 rotations per minute. The 2.0 mL samples were withdrawn at predetermined time points with the replacement of equal volumes of temperature-equilibrated media.
## 4.4.4. Solubility Studies
An excess amount of raw compounds and systems was placed in a 10 mL glass tube; then, 2.0 mL of phosphate buffer (pH 6.8) was added and left at room temperature for 3 h. The obtained solutions were diluted 1:1 with water, filtered through a 0.2 μm PTFE membrane filter (Sigma-Aldrich, St. Louis, MO, USA), and analyzed using the HPLC method.
## 4.4.5. Permeability Studies
In vitro gastrointestinal (GIT) and blood-brain barrier (BBB) permeability were studied using the PAMPA (Parallel Artificial Membrane Permeability Assay) models. The sandwich consists of two 96-well microfilter plates. The PAMPA systems contain two chambers: the donor chamber at the bottom and the acceptor chamber at the top. The chambers are separated by a 120 μm thick microfilter disc coated with a $20\%$ (w/v) dodecane solution of a lecithin mixture (Pion, Inc.). The donor solution was adjusted to pH ≈ 6.8 for GIT application and to pH ≈ 7.4 for BBB application using 0.5 M NaOH. The plates were combined and then incubated for 3 h for both models in a humidity-saturated atmosphere with the temperature set at 37 °C. To assess the apparent permeability coefficient factor (Papp), 5.0 mg of raw compounds were dissolved in 1.0 mL of DMSO. Then, we followed the manufacturer’s guidelines for further performance in the assay. The Papp factor was calculated according to the previously reported method [50]. In the case of the studied systems, the solutions were first prepared in the same manner as in the solubility study. Then the systems were diluted 1:1 with water, filtered through a 0.2 μm PTFE membrane filter, and further diluted 1:1 with DMSO. Next, the obtained solution was diluted 1:1 with donor solution for the GIT and BBB assays and placed in the donor compartment. The results were expressed as a concentration in the acceptor solution.
## 4.5.1. Antioxidant Activity Assay
Briefly, 25.0 μL of studied solutions (prepared for concentration determination in a solubility study) were mixed with 175.0 μL of DPPH radical solution. The rest of the analysis was performed according to the outlined procedure [50].
## 4.5.2. Determination of Butyrylcholinesterase (BuChE) Inhibition
The determination of BuChE inhibition was performed according to the previously reported method [50].
## 4.6. Statistical Analysis
Results are expressed as mean ± standard deviation. Statistical tests were performed using a one-way analysis of variance (ANOVA), and statistical differences using Duncan’s tests with a significance threshold of $p \leq 0.05$ were determined. All statistical analyses were performed using Statistica 13.1 software (TIBCO Software Inc., Palo Alto, CA, USA).
## 5. Conclusions
Obtaining amorphous systems of hesperetin and piperine seems to be the righteous approach to improving the overall bioavailability of both compounds.
We successfully obtained amorphous ternary dispersions of hesperetin and piperine. The systems were characterized by an improved dissolution rate, apparent solubility, permeability, and biological activities. Enhanced biological activity is strictly related to greater solubility, since only molecules that are freely dispersed in solution may act on biological targets. The administration of these systems might be beneficial in the context of various diseases due to the multidirectional mode of action of the studied plant-origin compounds.
## References
1. Nguyen J.T., Tian D., Tanna R.S., Hadi D.L., Bansal S., Calamia J.C., Arian C.M., Shireman L.M., Molnár B., Horváth M.. **Assessing transporter-mediated natural product-drug interactions via in vitro-in vivo extrapolation: Clinical evaluation with a probe cocktail**. *Clin. Pharmacol. Ther.* (2021) **109** 1342-1352. DOI: 10.1002/cpt.2107
2. Salehi B., Cruz-Martins N., Butnariu M., Sarac I., Bagiu I.-C., Ezzat S.M., Wang J., Koay A., Sheridan H., Adetunji C.O.. **Hesperetin’s health potential: Moving from preclinical to clinical evidence and bioavailability issues, to upcoming strategies to overcome current limitations**. *Crit. Rev. Food Sci. Nutr.* (2022) **62** 4449-4464. DOI: 10.1080/10408398.2021.1875979
3. Teng J., Li J., Zhao Y., Wang M.. **Hesperetin, a dietary flavonoid, inhibits AGEs-induced oxidative stress and inflammation in RAW264. 7 cells**. *J. Funct. Foods* (2021) **81** 104480. DOI: 10.1016/j.jff.2021.104480
4. Yang H., Wang Y., Xu S., Ren J., Tang L., Gong J., Lin Y., Fang H., Su D.. **Hesperetin, a Promising Treatment Option for Diabetes and Related Complications: A Literature Review**. *J. Agric. Food Chem.* (2022) **70** 8582-8592. DOI: 10.1021/acs.jafc.2c03257
5. Jayaraman R., Subramani S., Abdullah S.H.S., Udaiyar M.. **Antihyperglycemic effect of hesperetin, a citrus flavonoid, extenuates hyperglycemia and exploring the potential role in antioxidant and antihyperlipidemic in streptozotocin-induced diabetic rats**. *Biomed. Pharmacother.* (2018) **97** 98-106. DOI: 10.1016/j.biopha.2017.10.102
6. Sohel M., Sultana H., Sultana T., Al Amin M., Aktar S., Ali M.C., Rahim Z.B., Hossain M.A., Al Mamun A., Amin M.N.. **Chemotherapeutic potential of hesperetin for cancer treatment, with mechanistic insights: A comprehensive review**. *Heliyon* (2022) **8** e08815. DOI: 10.1016/j.heliyon.2022.e08815
7. Khan A., Ikram M., Hahm J.R., Kim M.O.. **Antioxidant and anti-inflammatory effects of citrus flavonoid hesperetin: Special focus on neurological disorders**. *Antioxidants* (2020) **9**. DOI: 10.3390/antiox9070609
8. Stahr P.-L., Grewal R., Eckert G.P., Keck C.M.. **Investigating hesperetin nanocrystals with tailor-made sizes for the prevention and treatment of Alzheimer’s disease**. *Drug Deliv. Transl. Res.* (2021) **11** 659-674. DOI: 10.1007/s13346-020-00888-0
9. Wang J., Li Q., Chen Z., Qi X., Wu X., Di G., Fan J., Guo C.. **Improved bioavailability and anticancer efficacy of Hesperetin on breast cancer via a self-assembled rebaudioside A nanomicelles system**. *Toxicol. Appl. Pharmacol.* (2021) **419** 115511. DOI: 10.1016/j.taap.2021.115511
10. Gu S.-F., Wang L.-Y., Tian Y.-J., Zhou Z.-X., Tang J.-B., Liu X.-R., Jiang H.-P., Shen Y.-Q.. **Enhanced water solubility, antioxidant activity, and oral absorption of hesperetin by D-α-tocopheryl polyethylene glycol 1000 succinate and phosphatidylcholine**. *J. Zhejiang Univ. B* (2019) **20** 273-281. DOI: 10.1631/jzus.B1800346
11. Trendafilova I., Mihály J., Momekova D., Chimshirova R., Lazarova H., Momekov G., Popova M.. **Antioxidant activity and modified release profiles of morin and hesperetin flavonoids loaded in Mg-or Ag-modified SBA-16 carriers**. *Mater. Today Commun.* (2020) **24** 101198. DOI: 10.1016/j.mtcomm.2020.101198
12. Haq I., Imran M., Nadeem M., Tufail T., Gondal T.A., Mubarak M.S.. **Piperine: A review of its biological effects**. *Phyther. Res.* (2021) **35** 680-700. DOI: 10.1002/ptr.6855
13. Shityakov S., Bigdelian E., Hussein A.A., Hussain M.B., Tripathi Y.C., Khan M.U., Shariati M.A.. **Phytochemical and pharmacological attributes of piperine: A bioactive ingredient of black pepper**. *Eur. J. Med. Chem.* (2019) **176** 149-161. DOI: 10.1016/j.ejmech.2019.04.002
14. Tripathi A.K., Ray A.K., Mishra S.K.. **Molecular and pharmacological aspects of piperine as a potential molecule for disease prevention and management: Evidence from clinical trials**. *Beni-Suef Univ. J. Basic Appl. Sci.* (2022) **11** 1-24. DOI: 10.1186/s43088-022-00196-1
15. Stojanović-Radić Z., Pejčić M., Dimitrijević M., Aleksić A., V. Anil Kumar N., Salehi B., C. Cho W., Sharifi-Rad J.. **Piperine-a major principle of black pepper: A review of its bioactivity and studies**. *Appl. Sci.* (2019) **9**. DOI: 10.3390/app9204270
16. Zafar A., Imam S.S., Alruwaili N.K., Alsaidan O.A., Elkomy M.H., Ghoneim M.M., Alshehri S., Ali A.M.A., Alharbi K.S., Yasir M.. **Development of piperine-loaded solid self-nanoemulsifying drug delivery system: Optimization, in-vitro, ex-vivo, and in-vivo evaluation**. *Nanomaterials* (2021) **11**. DOI: 10.3390/nano11112920
17. Zaini E., Afriyani A., Fitriani L., Ismed F., Horikawa A., Uekusa H.. **Improved Solubility and Dissolution Rates in Novel Multicomponent Crystals of Piperine with Succinic Acid**. *Sci. Pharm.* (2020) **88**. DOI: 10.3390/scipharm88020021
18. Ren T., Hu M., Cheng Y., Shek T.L., Xiao M., Ho N.J., Zhang C., Leung S.S.Y., Zuo Z.. **Piperine-loaded nanoparticles with enhanced dissolution and oral bioavailability for epilepsy control**. *Eur. J. Pharm. Sci.* (2019) **137** 104988. DOI: 10.1016/j.ejps.2019.104988
19. Imam S.S., Alshehri S., Alzahrani T.A., Hussain A., Altamimi M.A.. **Formulation and evaluation of supramolecular food-grade piperine HP β CD and TPGS complex: Dissolution, physicochemical characterization, molecular docking, in vitro antioxidant activity, and antimicrobial assessment**. *Molecules* (2020) **25**. DOI: 10.3390/molecules25204716
20. Kim D., Kim Y., Tin Y.-Y., Soe M.-T.-P., Ko B., Park S., Lee J.. **Recent technologies for amorphization of poorly water-soluble drugs**. *Pharmaceutics* (2021) **13**. DOI: 10.3390/pharmaceutics13081318
21. Schittny A., Huwyler J., Puchkov M.. **Mechanisms of increased bioavailability through amorphous solid dispersions: A review**. *Drug Deliv.* (2020) **27** 110-127. DOI: 10.1080/10717544.2019.1704940
22. Liu Y., Yang F., Zhao X., Wang S., Yang Q., Zhang X.. **Crystal Structure, Solubility, and Pharmacokinetic Study on a Hesperetin Cocrystal with Piperine as Coformer**. *Pharmaceutics* (2022) **14**. DOI: 10.3390/pharmaceutics14010094
23. Saravanan K., Sivanandam M., Hunday G., Pavan M.S., Kumaradhas P.. **Exploring the different environments effect of piperine via combined crystallographic, QM/MM and molecular dynamics simulation study**. *J. Mol. Graph. Model.* (2019) **92** 280-295. DOI: 10.1016/j.jmgm.2019.07.019
24. Garrido B., González S., Hermosilla J., Millao S., Quilaqueo M., Guineo J., Acevedo F., Pesenti H., Rolleri A., Shene C.. **Carbonate-β-cyclodextrin-based nanosponge as a nanoencapsulation system for piperine: Physicochemical characterization**. *J. Soil Sci. Plant Nutr.* (2019) **19** 620-630. DOI: 10.1007/s42729-019-00062-7
25. Akram A., Irfan M., Abualsunun W.A., Bukhary D.M., Alissa M.. **How to Improve Solubility and Dissolution of Irbesartan by Fabricating Ternary Solid Dispersions: Optimization and In-Vitro Characterization**. *Pharmaceutics* (2022) **14**. DOI: 10.3390/pharmaceutics14112264
26. Evans J.A., Mendonca P., Soliman K.F.A.. **Neuroprotective Effects and Therapeutic Potential of the Citrus Flavonoid Hesperetin in Neurodegenerative Diseases**. *Nutrients* (2022) **14**. DOI: 10.3390/nu14112228
27. Zhang M., Zhu S., Yang W., Huang Q., Ho C.-T.. **The biological fate and bioefficacy of citrus flavonoids: Bioavailability, biotransformation, and delivery systems**. *Food Funct.* (2021) **12** 3307-3323. DOI: 10.1039/D0FO03403G
28. Linnankoski J., Ranta V.-P., Yliperttula M., Urtti A.. **Passive oral drug absorption can be predicted more reliably by experimental than computational models—Fact or myth**. *Eur. J. Pharm. Sci.* (2008) **34** 129-139. DOI: 10.1016/j.ejps.2008.03.001
29. Szafraniec-Szczęsny J., Antosik-Rogóż A., Kurek M., Gawlak K., Górska A., Peralta S., Knapik-Kowalczuk J., Kramarczyk D., Paluch M., Jachowicz R.. **How does the addition of Kollidon**. *Pharmaceutics* (2021) **13**. DOI: 10.3390/pharmaceutics13020147
30. Strindberg S., Plum J., Stie M.B., Christiansen M.L., Hagner Nielsen L., Rades T., Müllertz A.. **Effect of supersaturation on absorption of indomethacin and tadalafil in a single pass intestinal perfusion rat model, in the absence and presence of a precipitation inhibitor**. *Eur. J. Pharm. Biopharm.* (2020) **151** 108-115. DOI: 10.1016/j.ejpb.2020.03.019
31. Suys E.J.A., Chalmers D.K., Pouton C.W., Porter C.J.H.. **Polymeric precipitation inhibitors promote fenofibrate supersaturation and enhance drug absorption from a type IV lipid-based formulation**. *Mol. Pharm.* (2018) **15** 2355-2371. DOI: 10.1021/acs.molpharmaceut.8b00206
32. Zhong Y., Jing G., Tian B., Huang H., Zhang Y., Gou J., Tang X., He H., Wang Y.. **Supersaturation induced by Itraconazole/Soluplus**. *Asian J. Pharm. Sci.* (2016) **11** 255-264. DOI: 10.1016/j.ajps.2015.07.001
33. Ozaki S., Minamisono T., Yamashita T., Kato T., Kushida I.. **Supersaturation–nucleation behavior of poorly soluble drugs and its impact on the oral absorption of drugs in thermodynamically high-energy forms**. *J. Pharm. Sci.* (2012) **101** 214-222. DOI: 10.1002/jps.22760
34. Higashino H., Hasegawa T., Yamamoto M., Matsui R., Masaoka Y., Kataoka M., Sakuma S., Yamashita S.. **In vitro-in vivo correlation of the effect of supersaturation on the intestinal absorption of BCS class 2 drugs**. *Mol. Pharm.* (2014) **11** 746-754. DOI: 10.1021/mp400465p
35. Hua S.. **Advances in oral drug delivery for regional targeting in the gastrointestinal tract-influence of physiological, pathophysiological and pharmaceutical factors**. *Front. Pharmacol.* (2020) **11** 524. DOI: 10.3389/fphar.2020.00524
36. Knopp M.M., Wendelboe J., Holm R., Rades T.. **Effect of amorphous phase separation and crystallization on the in vitro and in vivo performance of an amorphous solid dispersion**. *Eur. J. Pharm. Biopharm.* (2018) **130** 290-295. DOI: 10.1016/j.ejpb.2018.07.005
37. Brand W., Van Der Wel P.A.I., Rein M.J., Barron D., Williamson G., Van Bladeren P.J., Rietjens I.M.C.M.. **Metabolism and transport of the citrus flavonoid hesperetin in Caco-2 cell monolayers**. *Drug Metab. Dispos.* (2008) **36** 1794-1802. DOI: 10.1124/dmd.107.019943
38. Bi X., Yuan Z., Qu B., Zhou H., Liu Z., Xie Y.. **Piperine enhances the bioavailability of silybin via inhibition of efflux transporters BCRP and MRP2**. *Phytomedicine* (2019) **54** 98-108. DOI: 10.1016/j.phymed.2018.09.217
39. Yu D., Kan Z., Shan F., Zang J., Zhou J.. **Triple strategies to improve oral bioavailability by fabricating coamorphous forms of ursolic acid with piperine: Enhancing water-solubility, permeability, and inhibiting cytochrome p450 isozymes**. *Mol. Pharm.* (2020) **17** 4443-4462. DOI: 10.1021/acs.molpharmaceut.0c00443
40. Omidfar F., Gheybi F., Davoodi J., Amirinejad M., Badiee A.. **Nanophytosomes of Hesperidin and of Hesperetin: Preparation, characterization and in vivo evaluation**. *Biotechnol. Appl. Biochem.* (2022). DOI: 10.1002/bab.2404
41. Simpson D.S.A., Oliver P.L.. **ROS generation in microglia: Understanding oxidative stress and inflammation in neurodegenerative disease**. *Antioxidants* (2020) **9**. DOI: 10.3390/antiox9080743
42. Rekatsina M., Paladini A., Piroli A., Zis P., Pergolizzi J.V., Varrassi G.. **Pathophysiology and therapeutic perspectives of oxidative stress and neurodegenerative diseases: A narrative review**. *Adv. Ther.* (2020) **37** 113-139. DOI: 10.1007/s12325-019-01148-5
43. Singh A., Kukreti R., Saso L., Kukreti S.. **Oxidative stress: A key modulator in neurodegenerative diseases**. *Molecules* (2019) **24**. DOI: 10.3390/molecules24081583
44. Li B., Huang A.-L., Zhang Y.-L., Li Z., Ding H.-W., Huang C., Meng X.-M., Li J.. **Design, synthesis and evaluation of hesperetin derivatives as potential multifunctional anti-Alzheimer agents**. *Molecules* (2017) **22**. DOI: 10.3390/molecules22071067
45. Abdul Manap A.S., Wei Tan A.C., Leong W.H., Yin Chia A.Y., Vijayabalan S., Arya A., Wong E.H., Rizwan F., Bindal U., Koshy S.. **Synergistic effects of curcumin and piperine as potent acetylcholine and amyloidogenic inhibitors with significant neuroprotective activity in SH-SY5Y cells via computational molecular modeling and in vitro assay**. *Front. Aging Neurosci.* (2019) **11** 206. DOI: 10.3389/fnagi.2019.00206
46. Stanciu G.D., Luca A., Rusu R.N., Bild V., Beschea Chiriac S.I., Solcan C., Bild W., Ababei D.C.. **Alzheimer’s disease pharmacotherapy in relation to cholinergic system involvement**. *Biomolecules* (2019) **10**. DOI: 10.3390/biom10010040
47. Jasiecki J., Wasąg B.. **Butyrylcholinesterase protein ends in the pathogenesis of Alzheimer’s disease—Could BCHE genotyping be helpful in Alzheimer’s therapy?**. *Biomolecules* (2019) **9**. DOI: 10.3390/biom9100592
48. Jasiecki J., Targońska M., Wasąg B.. **The role of butyrylcholinesterase and iron in the regulation of cholinergic network and cognitive dysfunction in Alzheimer’s disease pathogenesis**. *Int. J. Mol. Sci.* (2021) **22**. DOI: 10.3390/ijms22042033
49. De la Rubia Ortí J.E., Platero J.L., Yang I.H., Ceron J.J., Tvarijonaviciute A., Sabater P.S., Benlloch M., Sancho-Cantus D., Sancho S.. **Possible Role of Butyrylcholinesterase in Fat Loss and Decreases in Inflammatory Levels in Patients with Multiple Sclerosis after Treatment with Epigallocatechin Gallate and Coconut Oil: A Pilot Study**. *Nutrients* (2021) **13**. DOI: 10.3390/nu13093230
50. Wdowiak K., Rosiak N., Tykarska E., Żarowski M., Płazińska A., Płaziński W., Cielecka-Piontek J.. **Amorphous Inclusion Complexes: Molecular Interactions of Hesperidin and Hesperetin with HP-Β-CD and Their Biological Effects**. *Int. J. Mol. Sci.* (2022) **23**. DOI: 10.3390/ijms23074000
51. Guo J., Tang W., Lu S., Fang Z., Tu K., Zheng M.. **Solubility improvement of hesperetin by using different octenyl succinic anhydride modified starches**. *LWT* (2018) **95** 255-261. DOI: 10.1016/j.lwt.2018.04.056
52. Krysa M., Szymańska-Chargot M., Zdunek A.. **FT-IR and FT-Raman fingerprints of flavonoids—A review**. *Food Chem.* (2022) **393** 133430. DOI: 10.1016/j.foodchem.2022.133430
53. Gorgani L., Mohammadi M., Najafpour G.D., Nikzad M.. **Sequential microwave-ultrasound-assisted extraction for isolation of piperine from black pepper (**. *Food Bioprocess Technol.* (2017) **10** 2199-2207. DOI: 10.1007/s11947-017-1994-0
54. Quilaqueo M., Millao S., Luzardo-Ocampo I., Campos-Vega R., Acevedo F., Shene C., Rubilar M.. **Inclusion of piperine in β-cyclodextrin complexes improves their bioaccessibility and in vitro antioxidant capacity**. *Food Hydrocoll.* (2019) **91** 143-152. DOI: 10.1016/j.foodhyd.2019.01.011
55. Školáková T., Slámová M., Školáková A., Kadeřábková A., Patera J., Zámostný P.. **Investigation of dissolution mechanism and release kinetics of poorly water-soluble tadalafil from amorphous solid dispersions prepared by various methods**. *Pharmaceutics* (2019) **11**. DOI: 10.3390/pharmaceutics11080383
56. Ijaz Q.A., Latif S., Rashid M., Arshad M.S., Hussain A., Bukhari N.I., Riaz S., Abbas N.. **Preparation and Characterization of pH-Independent Sustained-Release Tablets Containing Hot Melt Extruded Solid Dispersions of Clarithromycin**. *AAPS PharmSciTech* (2021) **22** 1-12. DOI: 10.1208/s12249-021-02115-6
57. Fu Q., Fang M., Hou Y., Yang W., Shao J., Guo M., Li M., Li J., Wang Y., He Z.. **A physically stabilized amorphous solid dispersion of nisoldipine obtained by hot melt extrusion**. *Powder Technol.* (2016) **301** 342-348. DOI: 10.1016/j.powtec.2016.06.032
58. Liu K., Liu H., Li Z., Li W., Li L.. **In vitro dissolution study on inclusion complex of piperine with ethylenediamine-β-cyclodextrin**. *J. Incl. Phenom. Macrocycl. Chem.* (2020) **96** 233-243. DOI: 10.1007/s10847-020-00980-5
|
---
title: Multi-Omics Profiling of Hypertrophic Cardiomyopathy Reveals Altered Mechanisms
in Mitochondrial Dynamics and Excitation–Contraction Coupling
authors:
- Jarrod Moore
- Jourdan Ewoldt
- Gabriela Venturini
- Alexandre C. Pereira
- Kallyandra Padilha
- Matthew Lawton
- Weiwei Lin
- Raghuveera Goel
- Ivan Luptak
- Valentina Perissi
- Christine E. Seidman
- Jonathan Seidman
- Michael T. Chin
- Christopher Chen
- Andrew Emili
journal: International Journal of Molecular Sciences
year: 2023
pmcid: PMC10002553
doi: 10.3390/ijms24054724
license: CC BY 4.0
---
# Multi-Omics Profiling of Hypertrophic Cardiomyopathy Reveals Altered Mechanisms in Mitochondrial Dynamics and Excitation–Contraction Coupling
## Abstract
Hypertrophic cardiomyopathy is one of the most common inherited cardiomyopathies and a leading cause of sudden cardiac death in young adults. Despite profound insights into the genetics, there is imperfect correlation between mutation and clinical prognosis, suggesting complex molecular cascades driving pathogenesis. To investigate this, we performed an integrated quantitative multi-omics (proteomic, phosphoproteomic, and metabolomic) analysis to illuminate the early and direct consequences of mutations in myosin heavy chain in engineered human induced pluripotent stem-cell-derived cardiomyocytes relative to late-stage disease using patient myectomies. We captured hundreds of differential features, which map to distinct molecular mechanisms modulating mitochondrial homeostasis at the earliest stages of pathobiology, as well as stage-specific metabolic and excitation-coupling maladaptation. Collectively, this study fills in gaps from previous studies by expanding knowledge of the initial responses to mutations that protect cells against the early stress prior to contractile dysfunction and overt disease.
## 1. Introduction
Hypertrophic cardiomyopathy (HCM) is among the most common inherited cardiomyopathies and, historically, a leading cause of sudden cardiac death in young adults [1,2]. Roughly $60\%$ of patients have a defined genetic disease, with the majority of mutations mapping to genes encoding thick and thin myofilament proteins [3,4]. Of these defined loci, mutations in myosin heavy chain 7 (MYH7) and myosin binding protein C3 (MYBC3) are most common and are clinically characterized by asymmetric left ventricular thickening, diastolic dysfunction, and fibrosis [3]. For instance, the Arg403Gln (R403Q) mutation in the MYH7 gene can result in a severe HCM phenotype with progressive myocardial dysfunction and increased incidence of sudden cardiac death [5,6]. Despite profound insights into the genetics of HCM, however, there is low correlation between mutation and clinical prognosis [7]. Moreover, there remains a need to define the earliest cell intrinsic responses to mutation prior to frank pathology.
Studies on isolated cardiac tissue have highlighted changes in contraction mechanics as potential drivers of the observed phenotype. HCM cardiomyocytes have increased tension costs (i.e., force generation per ATP utilized) due to changes at the biophysical level [8,9]. For example, both MYBC3 and MYH7 mutants exhibit a decreased number of super-relaxed-state myosin heads, a configuration with low ATPase cross-bridge utilization, and MYH7 mutants also show faster cross-bridge kinetics that directly increase tension costs [8,10,11]. Moreover, HCM exhibits marked Ca2+ handling dysfunction (e.g., increased Ca2+ sensitivity and intracellular diastolic Ca2+), further exacerbating tension costs and ATPase activity [4,12]. Although the exact mechanism remains unclear, this high energy demand is hypothesized to increase mitochondrial workload and oxidative stress, ultimately resulting in maladaptive cardiac remodeling [13].
Insight into cardiomyocyte remodeling has come from large-scale molecular (i.e., omics-type) profiling studies, whose findings imply that increased energy demand damages mitochondria via augmented oxidative stress [9]. Mutant cardiac cells exhibit increased mitochondrial ultrastructure damage and evidence of oxidative damage (presumably resulting from increased reactive oxygen species generation) [14,15]. Excessive oxidative stress can impair mitochondrial components, such as the electron transport apparatus and mitochondrial DNA, which can exacerbate ATP generation deficits [9,16]. Notably, patient myectomies exhibit marked metabolic deficits associated with mitochondrial dysfunction, such as significantly decreased levels of fatty acid metabolic enzymes and lower ratio of phosphocreatine-to-ATP levels, which are presumably secondary to mitochondrial dysfunction [9,17,18].
Precision mass spectrometry is a powerful tool for elucidating biomolecular networks and associated signaling cascades driving pathobiology [19]. Recently, we applied a quantitative phosphoproteomic profiling workflow to study the impact of cardiac fibrosis in a three-dimensional biomimetic in vitro co-culture model system (heart-on-a-chip) and in myectomy specimens from HCM patients, which revealed pathway-level alterations associated with altered energetics and calcium handling [20]. However, this model lacked associated HCM mutations and major outstanding questions remained regarding molecular cascades driving mitochondrial dysfunction and associated excitation–contraction coupling perturbations.
To address these gaps, we have now performed a comparative multi-omics (proteomic, phosphoproteomic, and metabolomic) analysis to illuminate the early and direct consequences of mutations in MYH7 in engineered human induced pluripotent stem-cell-derived cardiomyocytes (hiPSC-CM), in comparison to isogenic control cells, versus human HCM-patient-derived cardiac myectomy tissue, representing the spectrum of very early to clinically advanced stages of pathology. While some of the changes observed were co-ordinate, major differences were also detected. For instance, whereas we noted parallel increases in mitochondrial dynamics, favoring fission, in the early model and myectomy specimens, the myectomy specimens showed a complete abrogation of the major protective mechanisms against oxidative stress.
Integrated metabolic pathway analysis also highlighted increases in glutamine anaplerosis in the mutant hiPSC-CMs, which apparently replenished Krebs cycle intermediates and glutathione. Conversely, advanced-stage disease showed greatly diminished use of this pathway. We also characterized pronounced shifts in the post-translational (phosphorylation) states of the sarcoplasmic reticulum (SR) ATPase and functionally related excitation–contraction coupling proteins between the early and late disease stages, which correlated with alterations to Ca2+ handling, likely secondary to the metabolic defects.
In summary, our comprehensive assessment highlights metabolic, mitochondrial, and Ca2+ handling dysfunction across different stages of HCM pathology.
## 2.1. Differential Phosphoproteomics in Early HCM Specimens
To investigate metabolic status and mitochondria stress information in a model of the earliest pathobiological changes preceding overt disease, we first performed liquid chromatography–tandem mass spectrometry (LC/MS)-based global proteomics on 14 paired, isogenic hiPSC-CM samples (7 WT and 7 MYH7R403Q+/− replicate cultures) (Figure 1A). The hiPSC-CM cultures were differentiated from CRISPR/Cas9-edited MYH7R403Q+/− hiPSCs via small-molecule, monolayer manipulation of the Wnt signaling pathway (Figure 1B) [21,22]. MYH7R403Q+/− mutations led to hypertrophy, heightened metabolic activity, and increased force output at the single-cell level [23]. To ensure quantitative accuracy, we subjected each proteolytically digested sample to stable isotope labeling using isobaric tandem mass tag reagents and augmented the depth of proteome coverage by applying extensive peptide-level offline prefractionation prior to precision Orbitrap mass analysis (see Section 4, Figure 1C) [24].
From our initial proteomic analysis, we identified and quantified 7829 proteins between the mutant and control (WT) hiPSC-CM samples (Figure 2A). Differential proteome analysis revealed 648 statistically significant proteins (moderated Student t-test, false discovery rate (FDR) adjusted $p \leq 0.05$) (Figure 2B, Supplementary Table S1A, Supplementary Figure S1) [25]. From this set, we captured differential components linked to myofibrillary function and stability, including elevated levels of troponin I (TNNI3), filamin-C (FLNC), and four and a half LIM domains protein 1 (FHL-1) [26,27]. Moreover, we observed significantly altered expression of factors involved in Ca2+ handling, including downregulation of L-type calcium channel subunits and upregulation of calsequestrin (CASQ2).
Since phospho-signaling responses provide additional functional information complementary to differential protein expression, we next performed large-scale phosphopeptide affinity capture and a secondary LC/MS analysis, which allowed us to map 4362 phosphorylation sites on 1357 distinct phosphoproteins in our hiPSC-CM model system (Supplementary Table S1B). We observed 129 statistically significant (FDR < 0.05) differential phosphosites (Figure 2A), which included altered phosphorylation patterns on key sarcomeric proteins, such as titin (TTNSer1423), myosin-binding protein C (MYBPC3Ser427), filamin-C (FLNCSer2233), and leiomodin-2 (LMOD2Ser392) (Figure 2B).
Collectively, these twin data sets offer a rich resource for analyzing biochemical processes altered in the mutant (early-stage) samples.
## 2.2. Pathway-Level Changes in Early Model of HCM Reveal Altered Mitochondrial Dynamics in Early HCM Specimens
To systematically assess the pathway-level changes in the mutant cells, we performed gene set enrichment analysis (GSEA) with the proteomics data to parse annotated molecular pathways altered as a result of the MYH7R403Q+/− mutation (Supplementary Table S2) [25,28]. We noted significant enrichment (FDR < 0.05) of the mitochondrion organization and mitophagy gene sets in the mutated cells (Figure 3A), which pointed to potential modulation of morphology and function. Within the mitochondrion organization pathway, increased expression of key mitochondrial fusion and fission effectors was detected. Most notably, fusion proteins upregulated in MYH7R403Q+/− mutant cells included the primary GTPases driving fusion of the outer mitochondrial membranes, mitofusion-1 and -2 (MFN1 and −2) [29].
We also measured increased levels of Dynamin-1-like protein (DNM1L), the main GTPase conferring mitochondrial constriction during fission, as well as significant upregulation of clustered mitochondria protein homolog (CLUH) and mitochondrial Rho GTPase 2 (RHOT2), important regulators of mitochondrion trafficking that also mediate mitochondrial dynamics [29,30,31,32,33]. In particular, CLUH directly binds the mRNA transcripts encoding DNM1L receptor proteins (i.e., mitochondrial fission factor (MFF) and mitochondrial dynamics protein MID51 (MIEF1)) to facilitate fission [34,35]. Consistent with the upregulation of fission mediators, our phosphoproteomics data also showed an increasing trend (just short of statistical significance) in the phosphorylation levels of these DNM1L recruitment factors, MFFSer155 and MIEF1Ser55,59 (Supplementary Table S1B).
Mitochondrial fission primes cells for mitophagy, a selective form of autophagy during oxidative stress, which fragments the mitochondrial network and degrades damaged mitochondria via autophagosome engulfment [36,37]. Consistent with this, we observed overexpression of a number of key modulators of mitophagy and autophagosome pathways in the mutant cells (Figure 3A), suggesting increased engagement of mitochondrial quality control mechanisms. Notably, microtubule associated protein 1 light chain 3 (MAP1LC3B) and its ubiquitin binding adaptor protein sequestosome-1 (SQSTM1) were upregulated, as were the related ubiquitin-like proteins ATG5 and -12 [38,39,40]. MAP1LC3B is important for maturation of the autophagosome, and the MAP1LC3B-SQSTM1 interaction is crucial for targeting mitochondria to autophagosome and initiating degradation [41].
Taken together, these findings provide evidence of increased fission and mitophagy in mutant CMs, highlighting these as potential determinants of mitochondrial health in early disease.
## 2.3. Insufficient Oxidative Stress Response Associated with Mutant hiPSC-CMs
Oxidative stress and mitochondrial damage are well-known features of cardiac disease [15,42]. In HCM, mitochondria are especially vulnerable to oxidative damage due to augmented reactive oxygen species (ROS) production [9,18,43,44]. An important consequence of oxidative stress is a change in mitochondrial morphology and function, including key macromolecule modifications [9,15]. Accordingly, we found the ROS sensing response as significantly enriched in the mutant hiPSC-CM samples (Figure 3A). This pathway includes components of the proteasome complex and assembly, which function in clearance of accumulated oxidized proteins, such as proteasome activator complex subunit 4 (PSME4) [45,46].
Moreover, we noted differential expression of thiol-based antioxidants, which protect against excessive ROS (Figure 3B). These included upregulation of components of the glutaredoxin and thioredoxin systems, which assist in reversing protein glutathionylation, a consequence of increased oxidative stress (i.e., cysteine residue thiol oxidation) (Supplementary Table S1A) [47,48]. In particular, we noted significantly (log2FC = 0.791) elevated expression of glutaredoxin-2 (GLRX2). However, we also detected decreases in other major components of these pathways, such as thioredoxin reductase 2 (TXNRD2) and glutathione peroxidase 7 (GPX7). TXNRD2 and GLRX2 serve similar roles in maintaining redox homeostasis by reducing mitochondrial-specific redox proteins, suggesting moderate antioxidant protection in the mutant hiPSC-CM [47].
Mitochondrial DNA is a principal target of oxidative stress (i.e., oxidized bases and strand breaks) [49,50]. Consistent with this, we noted significant decreases in key components of mitochondrial base excision repair (BER) (Figure 2B), the main repair mechanism of oxidative damage in the mitochondria [51,52]. For example, the mutant hiPSC-CM samples significantly downregulated endonuclease-III-like protein 1 (NTHL1), a bifunctional glycosylase that excises oxidized DNA bases and generates abasic sites [53]. We also noted decreases in Poly (ADP-ribose) polymerase 1 and 2 (PARP$\frac{1}{2}$), which detect abasic sites and recruit the DNA repair protein XRCC1 (XRCC1), forming a scaffolding complex for other repair factors [54,55]. While PARP$\frac{1}{2}$ are predominately localized to the nucleus, PARP1 migrates to the mitochondria via interactions with Mitofilin (IMMT), where it plays a protective/repair role for mtDNA by interacting with DNA ligase 3 (LIG3) [56,57]. Notably, then, is the observation that LIG3 also downregulated in our MYH7R403Q+/− samples, which catalyzes the last ligation step of BER. Taken together, these data suggest that mtDNA is particularly vulnerable to oxidative damage, which could further facilitate mitochondrial dysfunction in mutant hiPSC-CMs.
## 2.4. Overlapping Mechanisms Driving Mitochondrial Dynamics and Downregulation of Oxidative Stress Response in Advanced HCM
To examine the persistence of the defects we observed in the mutant hiPSC-CM cells, we next measured the proteome and phosphoproteome of human donor biospecimens using the same workflow. We sampled explants from 10 sex-matched patients, 5 donor WT and 5 presenting with MYH7 mutant alleles (Supplementary Table S3, Table 1). As with the hiPSC-CM model, we subjected the peptide samples to stable isotope labeling, followed by extensive prefractionation before in-depth quantitative LC/MS analysis. In total, we identified and quantified 6463 proteins along with 9697 phosphorylation sites on 2766 phosphoproteins in the human donor samples (Figure 2A).
Comparative statistical analysis demonstrated hundreds of statistically significant (FDR < 0.05) differences in proteins and phosphosites between the HCM patient and case controls (Figure 2A). Notably, we measured significantly elevated phosphorylation-based regulation of key mitochondrial fission factors via RAS/MEK/MAPK1 pathway activation (Figure 3C). For example, the HCM donors exhibited increased phosphorylation of serine-616 on Dynamin-1-like protein (DNM1LSer616), which promotes activity and dimerization for constriction-based mitochondrial fission [58,59]. Further upstream, we noted evidence of activation of the mitogen-activated protein kinase (MAPK) cascade via hyperphosphorylation of the protein kinase domain on mitogen-activated protein kinase 1 (MAPK1Thr185), a conserved threonine/glutamate/tyrosine motif whose phosphorylation induces kinase activity [60]. MAPK1 increases mitochondrial dynamics by directly phosphorylating and activating DNM1LSer616 [61,62]. Consistent with this, we detected significant phosphorylation of the mitochondrial outer membrane docking receptors for DNM1L, including MIEFSer59, that are recruited during fission [34,63]. Hence, although not recapitulating the fusion and mitophagy alterations seen in the hiPSC-CM early model of pathology, these findings demonstrate a persistent upregulation of mitochondrial fission in advanced disease.
In contrast to the hiPSC-CM model, we observed profound decreases in factors linked to the oxidative stress response, such as thiol-based peroxidases (Figure 3D). The HCM patient samples exhibited significant decreases in the cell redox homeostasis pathway from GSEA, which included decreased expression of peroxiredoxin-6 (PRDX6) and mitochondrial thioredoxin (TXN2), important enzymes for reducing cellular peroxide levels (Supplementary Table S4, Supplementary Figure S2) [64]. From this pathway-level analysis, we detected decreased levels in thioredoxin reductase-1 and -2 (TXNRD$\frac{1}{2}$), which function to reduce thioredoxin, suggesting reduced thioredoxin antioxidant capacity [47,65]. Moreover, our analysis revealed concomitant alterations in key components of the double-strand DNA repair pathway (Supplementary Table S4), an important response that normally counters chronic oxidative stress [66].
Overall, in only partial accordance with the hiPSC-CM results, the advanced HCM specimens showed decreased expression in thiol-based oxidative response elements that are integral for maintaining mitochondrial integrity.
## 2.5. Integrated Metabolic Network Analysis Reveals Increased Dependence on Glutaminolysis in Early HCM
Metabolic remodeling is associated with increased mitochondrial dynamics in cardiomyocytes [67]. Thus, we applied an integrated metabolic pathway evaluation strategy on the hiPSC-CM samples to determine early metabolic alterations arising from the mutation.
First, we performed metabolic enrichment network analysis (MOMENTA) to discover differentially expressed metabolic pathways predicted from our proteomic data, which we then validated by direct metabolomic analysis [68]. Among the many metabolic pathways altered in the early-stage mutant model (Figure 4A), we noted significant increases in enzyme levels mediating glutamate degradation, as well as 2-oxoglutarate (α-ketoglutarate) decarboxylation to succinyl-CoA, suggesting upregulation of glutaminolysis in the early HCM samples (Supplementary Table S5).
To confirm these predictions further, we performed two independent phases (global and targeted) of small-molecule mass-spectrometry-based metabolomics analysis to directly confirm changes in metabolic profiles and flux.
Consistent with our proteomics predictions, our global metabolomics survey found direct evidence for a significant increase in glutamate levels in the MYH7R403Q+/− samples (Figure 4A, Supplementary Table S6). Among the hundreds of changes in metabolite levels that were detected, glutamate was among the top differential features (log2FC = 2.20, FDR < 0.001).
Given the evidence for glutaminolysis, we subsequently performed targeted metabolomic analyses to find further support of this pathway. For this analysis, we accurately quantified select glycolytic/citric acid cycle intermediates using 13C stable-isotope glucose tracer to determine whether unlabeled glutamate was supplementing the citric acid cycle (Figure 4B). As expected, the mutant cells showed elevated levels of labeled glucose-derived TCA metabolites, including 13C malate (Supplementary Table S7). In contrast, we observed a significant decrease in 13C succinate, the TCA produced following 2-oxoglutarate decarboxylation (Figure 4B). We additionally saw a significant increase in 13C alanine. Viewed from the vantage of current models of metabolism, these findings are consistent with the transamination of unlabeled glutamate (a downstream metabolite of glutaminolysis) with labeled pyruvate, resulting in increased labeled alanine and decreased succinate.
To verify that the enhanced glutamine is directly entering the citric acid cycle, we treated the WT and mutant hiPSC-CM cultures with isotopically (13C5) labeled glutamine (Supplementary Table S8). As expected, we observed a significant increase in labeled glutamine preferentially in the mutant cells (Figure 4C). Concomitantly, we detected increases in the labeled form of related TCA intermediate α-ketoglutarate in the HCM samples, further supporting the role of increased flux of glutamine into the TCA in the mutant cells.
To confirm the glutamine dependency of the MYH7R403Q+/− cells, we treated the hiPSC-CMs with the selective glutaminase inhibitor BPTES. Strikingly, we saw a rapid and significant decrease in contraction over the treatment period in the mutant cells (Figure 4B), leading to a complete loss of contraction in the MYH7R403Q+/− CMs by 90 min.
Intriguingly, we also detected a trend towards elevated glutathione-based synthesis from glutamine from our targeted metabolomics analysis (log2FC = 1.21), though slightly beyond statistical threshold (p-value = 0.0521) (Figure 4C, Supplementary Table S8). Our proteome data further supported this finding, as we observed significant decreases in glutathione catabolic components (Figure 3B). These included the putative catabolic factor gamma-glutamyl transpeptidase 3 (GGT3P). Moreover, MOMENTA revealed enrichment in the mutant cells of the pentose phosphate pathway, which produces NADPH (Figure 4A). NADPH is a cofactor for glutathione reductase that enables reduction of glutathione to alleviate oxidative stress [69]. We likewise detected increased levels of the fructose-2,6-bisphosphatase enzyme TIGAR, which increases the activity of the pentose phosphate pathway [70,71]. Glutathione is the primary intracellular thiol-based redox buffer, further supporting the protective engagement of the thiol-based oxidative stress response machinery in the early-stage disease model [48].
Taken together, our comprehensive metabolic analysis of the hiPSC-CM model strongly supports glutamine anaplerosis as an important mechanism for replenishing TCA metabolites as an early adaptive response to altered contractile function.
## 2.6. Metabolic Network Analysis Reveals Decompensated Metabolism in Advanced Specimens
In striking contrast to the results from our cell culture model, our proteomics analysis of the myectomy samples indicated decreased metabolic supplementation through the TCA/glutamine shunt.
Beyond significant reductions in the components of the glutamate degradation II pathway from MOMENTA (Supplementary Table S9), our GSEA of the clinical samples also revealed a significant (FDR < 0.05) decrease in mitochondrial fatty acid beta-oxidation and citric acid cycle/respiratory electron transport gene sets in the affected patient specimens (Figure 4A, Supplementary Table S4). Within the citric acid cycle/respiratory electron transport set, the HCM samples showed markedly decreased expression of electron transport complex proteins, such as NADH dehydrogenase (ubiquinone) 1 (NDUFV1).
Moreover, we noted decreases in several enzymes crucial in fatty acid metabolism, such as ACAT1 (acetyl-CoA acetyltransferase) and ACAA2 (3-ketoacyl-CoA thiolase) from the fatty acid beta-oxidation gene set (Supplementary Table S4). Given the importance of fatty acid oxidation in cardiac cells, these findings suggest poor metabolic capacity in the affected tissue [72].
Lastly, again in contrast to our early model, we noted significantly decreased enrichment of the pentose phosphate pathway in the advanced disease with respect to myectomy controls (Figure 4A). This pathway is crucial for its production of NADPH, which is used by glutaredoxin and thioredoxin systems to reduce key antioxidant enzymes.
Taken together with the decreases seen in mitochondrial-specific metabolic pathways, compounded with the decrease in glutamine anaplerosis, these findings demonstrate the potential for the persistence in impaired bioenergetic replenishment and antioxidant protection in advanced HCM.
## 2.7. Phosphorylation-Dependent Regulation of Excitation Contraction Coupling in hiPSC-CMs
Given the profound differences in mitochondrial function and metabolism noted in both the mutant model and advanced HCM specimens, we anticipated persistent dysregulation of calcium handling and contractility mechanisms in both contexts.
First, examining the hiPSC-CM model, we observed significantly altered phosphosites on key calcium handling factors localizing to the sarcoplasmic reticulum, mitochondria, sarcomere, and plasma membrane between the mutant and control samples in both the early model and advanced disease stages. For instance, in the hiPSC-CM mutant cells, we measured significantly increased phosphorylation events on a number of key SR proteins (Figure 5A, Supplementary Table S1B). These included hyperphosphorylation of cardiac phospholamban (PLNSer16,Thr17), ryanodine receptor 2 (RYR2Ser2814), sarcoplasmic/endoplasmic reticulum calcium ATPase 2 (ATP2A2Ser663), and junctophilin-2 (JPH2Ser241). The RYR2Ser2814 and PLNSer16,Thr17 events are linked to increased flux of calcium through the SR, with RYR2Ser2814 associated with channel opening and augmented outward flux, while PLNSer16,Thr17 has been shown to increase inward flux by releasing PLN inhibition of ATP2A2 (Supplementary Table S10A) [73,74].
We additionally measured persistent changes in the regulation of key sarcomeric proteins modulating contractility. These included increased phosphorylation of filamin-C (FLNCSer2233,2236) and leiomodin-2 (LMOD2Ser392), along with decreased levels of major phosphosites on titin (TTNSer1423,1418) and myosin-binding protein C (MYBPC3Ser424,427). While M-domain phosphorylation on MYBPC3 is crucial for modulating the ATPase activity of myosin via stability of its super relaxed state and its interaction with F-actin, we captured events in the adjacent C2 domain, an F-actin binding domain [75,76]. Given the decrease in super-relaxed-state myosin heads, this event could represent a novel mode of phosphorylation events increased in HCM regulation for this process and energetics. We also noted increased phosphorylation LMOD2, another actin binding protein that regulates contractility, and the protein phosphatase 1 regulatory subunit 12B (PPP1R12BSer842) and rho-associated protein kinase 1 (ROCK1Ser1105), upstream modulators of calcium sensitivity of the contractile machinery (Figure 5B) [77,78,79].
From our proteomics analysis of the early disease model, we observed decreases in the abundance of key plasma membrane calcium channels, as well as electrochemical gradient-forming ATPases (Supplementary Table S1A). These included decreased expression of the β-2/-3 and α2/δ3 subunits of the L-type calcium channel and plasma membrane calcium-transporting ATPase 1 (PMCA1), which directly modulate calcium flux at the plasma membrane [80,81]. Moreover, the MYH7R403Q+/− hiPSC-CM samples had decreased levels of the sodium/potassium-transporting ATPase subunit alpha-$\frac{1}{2}$ (catalytic) and beta-$\frac{1}{2}$, which are essential for cardiac excitability by assisting in calcium extrusion at the membrane [82]. Furthermore, they displayed reductions in CASQ2 and cardiac junction (ASPH), which play key roles in SR Ca2+ storage and the magnitude of release during excitation-contraction coupling buffering via interacting with ryanodine channels [83,84].
Overall, these findings suggested increased SR Ca2+ storage, with concurrent increases in Ca2+ flux throughout the SR in the MYH7R403Q+/− cells. To independently validate our prediction of defects in Ca2+ handling in the mutant cells, we directly measured Ca2+ flux in paced hiPSC-CMs via live-cell Ca2+ imaging. As summarized in Figure 5C, we observed a significant increase in SR Ca2+ storage in the HCM cells relative to WT following caffeine treatment, in line with our proteome measurements of CASQ2 expression. Surprisingly, however, we also detected a significant increase in time of Ca2+ release and re-uptake into the SR, despite the increased flux predicted from our phosphoproteome analysis. We note that this counterintuitive result is potentially explained by the increase in overall SR Ca2+, by which the additional Ca2+ must be pumped against a higher concentration gradient, thus increasing ATP2A2 (SERCA) energetic requirements. This would be expected to exacerbate ATP production demands on the mitochondria and to further thermodynamically limit SERCA function [85].
## 2.8. Phosphorylation-Dependent Regulation of Excitation Contraction Coupling in Myectomy Specimens
Focusing on pronounced changes in the HCM tissue samples, we again consistently measured differential phosphosites belonging to the excitation contraction coupling mechanism in advanced disease, as was seen in the early-stage model, though we noted a markedly different phosphorylation landscape (Figure 5D, Supplementary Table S3B). Top differential excitation contraction coupling phosphosites included downregulation of RYR2Ser2811 and JPH2Ser247 and increases in RYR2Ser2363 AKAP-12Ser283,286,T285/-13Ser2398,2728, ankyrin-1 (ANK1Ser1686). Yet, in contrast to the early HCM model, we detected decreases in PLNSer16,Thr17, RYR2Ser2814, and ATP2A2Ser663, which were increased in the early disease model samples (Figure 5E, Supplementary Table S10B).
We also noted increases in effectors that directly phosphorylate SR ATPases (e.g., RYR2), calcium channels, and sarcomere proteins to modulate Ca2+ homeostasis and sensitivity, which dictate overall ATP utilization. For example, protein phosphatase 1 and its regulatory subunits decrease PLNThr17 phosphorylation, allowing PLN to increase its inhibitory effects on ATP2A2 [74]. In the HCM tissue, we captured differential phosphorylation on regulatory subunits PPP1R12BSer711, PPP1R12CSer509, 453, PPP1R12AThr443, Ser445, PPP1R3ASer649, and PPP1R3DSer46. Other effectors seen in the clinical specimens included key upstream modulators of Na+ and Ca2+ homeostasis regulation, such as calcium/calmodulin-dependent protein kinase type II subunit beta (CAMK2BSer367,276).
Other important regulators of sarcomeric calcium sensitivity showing persistent changes in advanced disease included differential phosphorylation of TTN, FLNC, synaptopodin-2 (SYNPO2), MYBPC3, and myosin regulatory light chain 2 (MYL2) (Figure 5D). In particular, the MYL2Ser15,19,Thr24 decreased phosphorylation levels in advanced HCM, as well as decreased phosphorylation of its upstream kinase MYLK3Ser152,355,Thr359, having important regulatory implications, since these phosphosites function in cross-bridge kinetics [86]. For instance, the serine-15 modification increases lever arm stiffness and myosin attachment, enhancing contraction and protecting against hypertrophy-related stress [87,88].
Among the main plasma membrane channels detected in the clinical samples, we found decreased phosphorylation of voltage-dependent L-type calcium channel subunits (CACNB2Ser514,550 and CACNA1CSer1784) and potassium voltage-gated channels (KCNH2Ser255) in the HCM tissue. These components regulate the influx of calcium into the cytoplasm and the rectifying potassium current, respectively, and their phosphorylation provides precise regulation of open probability [89].
The overall phospho-pattern of SR proteins and modulators of contractility suggests differing signaling-based regulation in advanced disease compared to the early specimens.
## 3. Discussion
While previous molecular studies by our group and others have elucidated important mechanisms of HCM pathobiology, few have explored consequences associated with MYH7 mutations over different disease-state phenotypes [9,90,91]. Through an integrated quantitative multi-omics analysis, we captured hundreds of additional differential features which mapped to distinct molecular mechanisms dictating mitochondrial homeostasis in both early and advanced models of HCM, as well as stage-specific metabolic pathways and excitation-coupling mechanistic adaptations.
Although there are well-known limitations associated with hiPSC-based models, in particular the immature fetal-like maturation states, our mutant system provides a tractable means to explore the direct molecular and phenotypic consequences of defective MYH7 function at the early stages of pathology. Conversely, despite substantial clinical heterogeneity in HCM presentation, our comparative analysis of patient myectomy specimens, belonging to a single defined mutation subpopulation, allowed us to elucidate perturbations associated with both early and late stages of pathology. Moreover, while we analyzed an array of MYH7 mutations in this subpopulation, we were able to identify consistent findings amongst late-stage HCM specimens.
Oxidative stress and mitochondrial damage are commonly described features of HCM [9,18]. In physiological conditions, the deleterious effects of ROS (e.g., oxidative modification) are balanced by endogenous antioxidant systems [92]. However, excess production can impair the function of mitochondrial components [5,93,94]. We observed a mixed antioxidant response in the mutant hiPSC-CM, followed by a near total loss in this pathway signature in the advanced myectomy specimens. The early-stage in vitro model suggests increased protection against protein glutathionylation through thiol-based systems, the primary intracellular redox buffer for protecting protein cysteine residues from oxidative modification. We found evidence in both our proteomics and metabolomics data for increased antioxidant protection via expression of glutaredoxin and thioredoxin enzymes and upregulation of the pentose phosphate pathway. The latter provides NADPH for reducing oxidized glutathione, allowing another redox cycle [95]. From our targeted metabolomics experiments, we confirmed that glutamine moved faster towards GSH synthesis in the mutant cells, consistent with a crucial role of glutaminolysis for cardiomyocytes under oxidative stress [96]. Thus, there appears to be a relatively robust intracellular antioxidant protection early on, which likely serves to counteract the known overproduction of ROS reported before in in vitro HCM models [14,15].
Conversely, our myectomy data revealed a profound loss of these protective mechanisms in advanced stage disease, including TXN2 and its reductase TXRN2. These expression deficits are likely associated with the reduced ROS scavenging capabilities noted by previous studies of HCM myectomy samples, particularly the increased ratio of GSSG:GSH (i.e., reduced GSH antioxidant capacity) [15,97].
Interestingly, Wang et al. recently reported increased evidence of pentose phosphate pathway in advanced HCM, but our own patient cohort data suggest the opposite trend [90]. In contrast to their KEGG pathway analysis, our integrative MOMENTA framework detected significant downregulation of transaldolase and the NAD-dependent master regulatory deacetylase enzyme sirtuin-2. One potential explanation for this discrepancy is the clinical heterogeneity of HCM, which varies greatly from mutation to mutation. While we exclusively utilized MYH7 mutants, Wang et al. analyzed samples representing an array of mutations. Overall, our observation of poor thiol-based antioxidant protection in the clinical HCM specimens suggests exacerbation of the mitochondrial damage and dysfunction observed before in myectomy specimens [9].
Given the importance of double-strand break response in oxidative stress, the decreased expression of DNA repair mechanisms in our analysis of advanced HCM specimens further demonstrates poor oxidative stress adaptation [66]. Mitochondrial DNA is especially vulnerable to oxidative modification given its proximity to ROS production [98]. Hence, the decreased expression of BER enzymes in the mutant hiPSC-CM model implies the potential accumulation of damaged mtDNA at the early stages of pathology. Mitochondrial damage is known to result in reduced bioenergetics and could be the cause of the overall decreased ATP generation and mitochondrial metabolic pathways noted before in advanced HCM [9,99]. Further supporting this, a number of cardiomyopathies linked to mtDNA mutations exhibit pronounced mitochondrial dysfunction [100].
Mitochondrial dynamics are an important mechanism for maintaining mitochondrial homeostasis and function [23,101]. Mitochondrial fusion ensures optimal respiratory function by exchanging contents (e.g., proteins and DNA) after merging [102,103]. Conversely, fission is the division of a single mitochondrion into two daughter organelles, which assists in mitochondrial distribution but can also serve in quality control during oxidative stress via activation of mitophagy. Imbalances between these processes can result in malfunctioning mitochondria and metabolic disturbances [101]. There is substantial evidence that fission and fusion processes are active in HCM pathology, especially given the increases in mitochondrial number and cristae disorganization noted before in both clinical and in vitro culture models of HCM [9,23]. Accordingly, we noted an increase in mitochondrial dynamics in response to mutation in both the early and advanced pathology. In our advanced specimens, we found significant phosphorylation of DNM1LSer616 and its recruitment factors MFFSer157 and MIEFSer59, which are the principal components for fission [104]. This echoes the increased evidence from Ranjbarvaziri et al., in which they found this ultrastructural mitochondrial remodeling from imaging [9]. Perplexingly, despite increased oxidative stress (e.g., elevated 4-hydroxy-2-nonenal-modified proteins) and the mitochondrial damage noted in their study, they failed to find molecular activation of the fusion/fission mechanisms. In contrast, we provide direct phosphorylation level regulation of this pathway, which helps explain those ultrastructural changes in HCM.
Though we found that the regulators of fission and fusion were upregulated in both early and advanced stages, we only found direct evidence of mitophagy in the in vitro culture model. Mitophagy is important for clearing impaired mitochondria via the autophagosome–lysosome pathway following fission/fusion and would provide another means of combating oxidative stress [105]. We failed to find direct molecular evidence to support elevated mitophagy in advanced specimens. Mitochondrial dynamics and mitophagy are dependent on cardiolipin lipid species for recruiting fusion and fission mediators [106]. Consistent with this, our MOMENTA of the hiPSC-CM samples showed increased enrichment of cardiolipin biosynthesis in the mutant cells (Figure 4A), as well as increased expression of cardiolipin synthase (CRLS1), which catalyzes the last step in de novo cardiolipin synthesis (Supplementary Table S1A). In contrast, the advanced specimens showed decreased expression of CRLS1 and the lysocardiolipin acyltransferase 1 (LCLAT1) (Supplementary Table S3A), both crucial enzymes for cardiolipin production in vivo [107]. Given the previously reported decreases in cardiolipin species in advanced HCM, the failure to upregulate these mechanisms could be associated with failure to upregulate mitophagy in advanced specimens [9].
The metabolism of cardiomyocytes is dynamic and cardiac tissues are able to adapt their preferences for carbon sources depending on metabolic demand [96,101,108]. Our discovery of increased glutamine anaplerosis in early mutant samples is one such demonstration, noting that glutamine catabolism may serve as an important contributor to preservation of cardiac function in early-stage pathology. We demonstrated that increased glutaminolysis serves a biosynthetic role in the mutant CMs, replenishing the TCA intermediate 2-oxoglutarate via transamination of glutamate, as well as providing a major precursor for glutathione synthesis. In contrast, the advanced samples exhibited decreased glutamate degradation, a pathway downstream of glutamine anaplerosis. Recently, Watanabe et al. found that glutaminolysis improves cell viability in healthy cardiomyocytes under oxidative stress [96]. We provide compelling evidence that this mechanism provides a dual protective role in HCM, while its loss in advanced stage disease likely represents a further maladaptation to chronic oxidative stress.
Excitation contraction coupling, particularly ATPase-dependent components, rely on readily available ATP for contiguous function. Unsurprisingly, our global proteome and phosphorylation landscape captured distinct calcium handling and calcium sensitivity between HCM and normal controls in both the in vitro and in vivo contexts. Our analysis of the mutant cell cultures, which showed increased metabolic activity supplemented by glutaminolysis, revealed significant activation of SR calcium handling proteins in HCM. We noted increases in PLNSer16,Thr17, events that remove its reversible inhibition of ATP2A2 [74]. Importantly, these sites are regulated by distinct kinases (e.g., CAMK2A vs. PKA), demonstrating multiple levels of regulation. Since this latter enzyme accounts for the majority of Ca2+ reuptake, reducing the degree of inhibition is predicted to increase SR Ca2+ reuptake activity. This regulation is likely necessary given the increased SR Ca2+ storage in MYH7R403+/− cells, revealed from Ca2+ imaging, particularly when coupled to the decrease we observed in PMCA1 and supporting Na+/K+ ATPases, which assist ATP2A2 in decreasing cytosolic Ca2+ for relaxation. This high Ca2+ reuptake burden on SR ATPases is expected to enhance ATP demands over time, and may be a predominant initial source of mitochondrial ATP demand. Conversely, we noted an opposite trend on SR handling in the advanced disease samples, with decreased phosphorylation events at the same sites. This suggests a lower burden of ATPase-dependent activity, perhaps due to decompensated ATP production. Other novel findings included phospho-regulation of contractility, such as decreased MYBPC3 phosphorylation in both systems.
Collectively, this study fills in gaps from previous studies of HCM and expands knowledge of the initial responses to mutations that protect against early-stage cardiac pathology that eventually are overwhelmed, leading to irreversible advanced disease. Beyond the molecular insights highlighted, all the data are being made publicly available to serve as a community resource for future mechanistic studies.
## 4. Materials and Methods
Human induced pluripotent stem cell (hiPSC)-derived cardiomyocyte cell culture—Human iPSC-derived cardiomyocytes were generated from the Harvard Personal Genome Project line 1 (PGP1, GM23338), generously provided by the Seidman Lab at Harvard Medical School [23,109]. Heterozygous mutations were introduced into the MYH7 allele in hiPSC as described [23,110]. hiPSCs were cultured with mTESR1 media (STEMCELL Technologies, Vancouver, BC, Canada) on Matrigel (Thermo Fisher Scientific, Waltham, MA, USA)-coated plates and differentiated into cardiomyocytes (day 0) at 80–$100\%$ confluency via activation of the WNT pathway with 12 μM CHIR 99021 (Tocris, Bristol, UK) in RPMI + GlutaMAX media supplemented with B27 minus insulin (RPMI and B27 minus, Thermo Fisher Scientific). After 24 h, the cells were washed with PBS and given RPMI and B27 minus. On day 3, the WNT pathway was inhibited via 5 μM IWP-4 (Tocris) in fresh RPMI and B27 minus media for 48 h. On day 5, the media was replaced with fresh RPMI and B27 minus and changed every other day with the use of B27 plus insulin, starting on day 9. On day 11 and 13, the cardiomyocyte population was purified by metabolic selection via RPMI glucose-free media (Gibco, Waltham, MA, USA) supplemented with 4 mM of DL-lactate (MilliporeSigma, Burlington, MA, USA). On day 15, the media was replaced with RPMI + GlutaMAX media supplemented with B27 plus insulin. Following 48 h, cells were replated onto fibronectin-coated tissue culture plastic plates using $0.25\%$ Trypsin-EDTA (Thermo Fisher Scientific) and 10 μg/mL deoxyribonuclease I (STEMCELL Technologies). They were then placed in RPMI B27+ supplemented with 5 μM Y-27632 (Tocris) and $2\%$ fetal bovine serum (MilliporeSigma) for seeding. Cultures were maintained on RPMI + GlutaMAX media supplemented with B27 plus insulin until harvesting at 30+ days post-differentiation. For harvesting, cells were trypsinized, centrifuged, and snap frozen in liquid nitrogen for proteomics analysis.
Patient Sample Acquisition—Surgical myectomy tissue was obtained from HCM patients with symptomatic LVOT obstruction. Patients gave informed consent for their myectomy tissue to be used in research. MYH7 pathogenic variants were identified by whole exome sequencing of peripheral blood mononuclear cell DNA and confirmed by commercially available Gene Panels (GeneDx, Stamford, CT, USA). Myectomy sample processing was conducted as previously described [111,112,113,114]. A total of 100 mg of collected myectomy tissue was minced into 1 mm3 pieces, placed in 0.5 mL of CryoStor CS10 Freeze Media (STEMCELL Technologies), and stored in a MrFrosty (ThermoFisher) at 4 °C for 10 min, and then transferred to −80 °C overnight. Sample collection was approved by the Tufts University/Medical Center Health Sciences Institutional Review Board under IRB protocol # 9487. All subjects gave their informed consent for inclusion before they participated in the study. The study was conducted in accordance with the Declaration of Helsinki. Tissue from organ donor patients without underlying cardiac disease was obtained and processed as described previously [111,112,113,114].
Homogenization, Metabolite/Protein Extraction, and Trypsin Digestion—The workflow is summarized in Figure 1C. For hiPSC-CMs, frozen samples were resuspended in a 4 vol. mixture of ice-cold methanol/acetonitrile/water (MeOH/ACN/H2O, $\frac{40}{40}$/20, v/v; MS-grade, Fisher) in chemically resistant microcentrifuge tubes (Eppendorf, Hamburg, DEU). The samples were flash frozen in liquid nitrogen for 1 min, allowed to thaw, and sonicated on ice (Ultra Autosonic, Pune, IND) for 5 min three times. The extracts were then incubated at −20 °C for 1h and centrifuged at 12,000× g at 4 °C for 15 min to pellet the protein precipitate. The supernatants (with the metabolites) were transferred to new microtubes, dried under vacuum at 30 °C. The crude metabolite extracts were then subjected to solid-phase micro-extraction (SPME) for cleaning up. In brief, the coated SPME blades were preconditioned in MeOH/H2O (50:50, v/v) for 30 min. The samples were resuspended in $2\%$ MeOH and incubated for 1 h with the resins (blades). After the blades were rinsed for 20s using water, the bound metabolites were desorbed using ACN/H2O (50:50, v/v) for 1 h. The eluted metabolites were dried under speedvac (Eppendorf) and kept at −80 °C prior to LC-MS analysis. The protein precipitate was transferred to 200 µL of lysis buffer containing 6 M GuHCl, 100 mM Tris pH 8.5, 1 mM CaCl2, 10 mM TCEP, and 40 mM CAM, with EDTA-free protease inhibitor (Sigma) and phosphatase inhibitor (PhosSTOP Roche, MilliporeSigma), and heated for 6 min at 95 °C. Human samples were placed in 5 vol. lysis buffer containing 8M urea, 5mM dithiothreitol (DTT), 50 mM ammonium bicarbonate (NH4HCO3), with EDTA free protease inhibitor (Sigma), and phosphatase inhibitor (PhosSTOP Roche). Afterwards, they were mechanically homogenized. All samples were sonicated on ice (Branson, Brookfield, CT, USA). Protein quantity was assessed by Bradford protein assay (Bio-Rad, Hercules, CA, USA), followed by digestion overnight at 37 °C with sequencing-grade Trypsin (1:50 enzyme to protein ratio, w/w, ThermoFisher). After adding trifluoroacetic acid to $0.1\%$ v/v, peptide digests were desalted using a C18 Sep-Pak (Waters, Milford, MA, USA) according to the manufacturer’s instructions, resuspended in 100 mM TEAB, and quantified by Quantitative Colorimetric Peptide Assay (Pierce) prior to TMT labeling.
Stable-Isotope Labeling and Offline Reverse-Phase High-Performance Liquid Chromatography Fractionation—For each sample, 100 µg of peptide digest (adjusted to 100 µL with 100mM TEAB) was mixed with a unique amine-reactive isotope-coded isobaric tandem mass tag (TMT-16-plex) reagent (Thermo Fisher Scientific) prior to sample multiplexing and precise quantification by LC/MS. After pooling, labeled peptide was desalted, dried, and suspended in 300 µL buffer containing $0.1\%$ ammonium hydroxide and $2\%$ acetonitrile (ACN). The pooled sample mixture was pre-fractionated by high pH reverse-phase HPLC on a XBridge Peptide BEH C18 column (130Å, 3.5 μm, 4.6 mm × 250 mm, Waters) using an Agilent 1100 HPLC system. Peptides were eluted using a gradient of mobile phase A ($0.1\%$ NH4OH −$2\%$ ACN) to B ($0.1\%$ NH4OH −$98\%$ ACN) over 48 min and collected as 12 pooled fractions. For phosphoproteomics, the bulk ($95\%$) of each sample was subject to phospho-peptide enrichment using FeO2 metal-chelate resin (PureCube Fe-NTA MagBeads, Cube Biotech, Rhein, DEU) [24], while the remaining ($5\%$) portions were analyzed directly by nanoflow LC/MS as bulk proteome measurements (a total of 24 injections, 12 for proteomics and 12 for phosphoproteomics).
Mass Spectrometry Analysis of Peptides and Identification—Isotope-labeled peptides were reconstituted in mobile phase A ($0.1\%$ formic acid, $2\%$ ACN) prior to LC/MS analysis on a Thermo-Fisher Exploris 480 hybrid quadrupole-Orbitrap mass spectrometer interfaced with Thermo-Fisher FAIMS Pro with integrated Proxeon EASY-nLC 1200 system. After loading onto a C18 reverse-phase pre-column (75 μm i.d. × 2 cm, 3 μm, 100Å, Thermo Fisher Scientific), peptides were gradient separated on an EASY-Spray C18 nanocolumn (75 μm i.d. × 50 cm, 2 μm, 100Å; ES803A, Thermo Fisher Scientific) using 2–$35\%$ mobile phase B ($0.1\%$ formic acid, $80\%$ ACN) over 120 min (proteome) or 180 min (phosphoproteome), and electro-sprayed at ~250 nL/min into the Exploris instrument operated in positive ion mode (capillary temperature 275 °C, 2100 V potential). Data-dependent spectra were acquired automatically via high-resolution [60,000] precursor ion scan (350–1500 m/z range) to select the 12 most intense peptides for MS/MS fragmentation by high energy dissociation (normalized collision energy of 33 at 45,000 resolution).
The resulting RAW files were searched by MaxQuant (1.6.7.0) using default settings against the human proteome (SwissProt Taxonomy ID: 9606, downloaded September, 2021), allowing for two missed cleavage sites and variable modifications (Ser/Thr/Tyr phosphorylation, N-terminal acetylation, and Met oxidation) and carbamidomethylation of cysteine and TMT labels as a fixed modification. Peptide- and protein-level matches were filtered to high confidence ($1\%$ FDR), with a minimum phosphosite localization probability of 0.7. TMT quantification involved label correction (lot values provided by ThermoFisher).
Phosphoproteomics Statistical Analysis and Pathway Enrichment—*Bioinformatic analysis* was performed using R (language and environment for Statistical Computing; http://www.R-project.org, accessed on 12 May 2021.). Peptide feature intensities were log transformed and loess normalized. LIMMA R package was used for differential analysis (moderated Student t-tests), and to generate ranked lists for subsequent enrichment analysis using the Benjamini–Hochberg FDR correction [25,115]. Statistical enrichment analysis was performed using fgsea R package [28,68]. Volcano plots were created with the EnhancedVolcano R package [116]. All figures were created with BioRender.com.
Mass Spectrometry of Metabolites and Identification—Metabolites were reconstituted in mobile phase A ($2\%$ ACN) prior to metabolite nanoflow (nLC) LC/MS analysis on a Hybrid Quadrupole-Orbitrap Q-Exactive HF (Thermo Scientific). After loading onto a C18 reverse-phase pre-column (75 mm i.d. × 2 cm, 3μm, ThermoScientific), metabolites were separated on a capillary column (75 mm i.d. × 25 cm, 2 μm, 100 Å, ThermoFisher Scientific) using a gradient of $2\%$ to $60\%$ mobile phase B ($80\%$ ACN) for 20 min, increased to $95\%$ mobile phase B over 10 min, and maintained at $95\%$ mobile phase B for 15 min, and electro-sprayed at 300 nL/min into the Q-Exactive HF. Data-dependent spectra were acquired automatically (automated switching ESI mode) via high-resolution [60,000] precursor ion scan over a full mass scan range of m/z 67−1000. The source ionization parameters were optimized for a transfer temperature at 300 °C and a spray voltage set to 2.1 kV and −1.8 kV for the positive and negative modes, respectively. MS2 scans were performed at 15,000 resolution, with a maximum injection time of 64 ms, using stepped normalized collision energies (NCEs) of 10, 20, and 40. Dynamic exclusion was enabled using a time window of 10s. Raw data (switching mode) were split into positive and negative files and subject to OmicsNotebook (R script) for peak detection, deconvolution, retention time alignment, and metabolite identification against open database [25]. Intensities of features (putative metabolites) were normalized prior to differential analysis (moderated Student t-test).
Live-Cell Calcium Imaging—On day 30+ of differentiation, WT and mutant hiPSC-CMs were plated at 80k cells/well in a 24-well plate. Three days after plating the hiPSC-CMs, they were washed three times in Tyrode’s buffer (Thermo Fisher Scientific) and incubated in 10 μM Rhod-3 AM calcium indicator, PowerLoad, and Probenecid for 45 min covered from light, following the protocol from the Rhod-3 AM calcium imaging kit (ThermoFisher). Cells were washed in Tyrode’s buffer and incubated in Probenecid for an additional 45 min. Cells were then washed in Tyrode’s buffer three additional times and kept in Tyrode’s during live imaging. Calcium transients were acquired at 30 frames per second at 6X on a Nikon Eclipse Ti (Nikon Instruments, Tokyo, JP) with an Evolve EMCCD Camera (Photometrics, Tucson, AZ, USA), equipped with a temperature and CO2 equilibrated environmental chamber. hiPSC-CMs were electrically stimulated at 1 Hz using a C-Pace EP stimulator (IonOptix, Westwood, MA, USA). Videos were acquired at 15 locations per experimental group with a 560 nm laser illumination wavelength. Calcium transients of each cell were calculated using a custom Matlab script tracking intensity change over time within the cell. Calcium release was calculated as the time it took the calcium intensity in the cytoplasm to reach $50\%$ maximum intensity, while calcium reuptake was calculated as the time it took the calcium intensity in the cytoplasm to decrease to $50\%$ of its maximum intensity. This was repeated over three differentiations ($$n = 3$$), with n > 23 cells per differentiation. SR storage was determined from the ratio of the calcium transient immediately following and preceding incubation with the addition of 20 mM caffeine to Tryrode’s buffer (final concentration of 10 mM). All data were normalized to the average WT values for the matched differentiation. Data were assessed with a two-way ANOVA with a mixed-effect model that was corrected for multiple comparisons using a Tukey test with a $95\%$ confidence level.
Targeted C13 Flux and BPTES application—C13 metabolomics flux metabolomics analyses were performed according to Yuan and collaborators [117]. Briefly, hiPSCs-derived cardiomyocytes, at day 30, were cultured during 24 h in RPMI glucose free + B27+ and 13C6 glucose (CLM-1396) added at a final concentration of 11.1 mM for glucose flux or cultured in RPMI glutamine free + B27+ and 13C5 glutamine (CLM-1822) added at a final concentration of 2.05 mM for glutamine flux. Metabolites were extracted with −80 °C cold $80\%$ methanol (LC-MS grade), centrifuged, and supernatant was dried on speedvac over 18 h. Metabolite pellets were stored at −80 °C up to analysis (no more than 7 days) and protein pellets were used to normalize metabolites levels. Targeted metabolomics was performed using SRM LC-MS (13C glucose: $$n = 5$$ and 13C glutamate: $$n = 8$$, where N = number of differentiations). In cases of no analyte measurement (i.e., below detection limit) in a given sample, analyte values were excluded for that sample. All possible 13C labeled transitions were monitored. BPTES inhibition of glutamine oxidation was performed according to the manufacturer’s instructions (Agilent Seahorse XF Mito Fuel Flex Test Kit), 2 differentiations. Contractile parameters were assessed according to Toepfer and collaborators [118]. Data were assessed via unpaired, two-tailed t-test with Welch’s correction. The schematic in panels (B) and (C) was created in reference to Figure 1 of Fan et al. and Mcdonald et al. [ 119,120].
## References
1. Geske J.B., Ommen S.R., Gersh B.J.. **Hypertrophic Cardiomyopathy: Clinical Update**. *JACC Heart Fail* (2018) **6** 364-375. DOI: 10.1016/j.jchf.2018.02.010
2. Ommen S.R., Mital S., Burke M.A., Day S.M., Deswal A., Elliott P., Evanovich L.L., Hung J., Joglar J.A.. **2020 AHA/ACC Guideline for the Diagnosis and Treatment of Patients With Hypertrophic Cardiomyopathy**. *Circulation* (2020) **142** e558-e631. DOI: 10.1161/CIR.0000000000000937
3. Marian Ali J.. **Braunwald Eugene Hypertrophic Cardiomyopathy**. *Circ. Res.* (2017) **121** 749-770. DOI: 10.1161/CIRCRESAHA.117.311059
4. Akhtar M., Elliott P.. **The Genetics of Hypertrophic Cardiomyopathy**. *Glob. Cardiol. Sci. Pract.* (2018) **2018**. DOI: 10.21542/gcsp.2018.36
5. Viola H.M., Hool L.C.. **Impaired Calcium Handling and Mitochondrial Metabolic Dysfunction as Early Markers of Hypertrophic Cardiomyopathy**. *Arch. Biochem. Biophys.* (2019) **665** 166-174. DOI: 10.1016/j.abb.2019.03.006
6. Geisterfer-Lowrance A.A.T., Kass S., Tanigawa G., Vosberg H.-P., McKenna W., Seidman C.E., Seidman J.G.. **A Molecular Basis for Familial Hypertrophic Cardiomyopathy: A β Cardiac Myosin Heavy Chain Gene Missense Mutation**. *Cell* (1990) **62** 999-1006. DOI: 10.1016/0092-8674(90)90274-I
7. Landstrom A.P., Ackerman M.J.. **Mutation Type Is Not Clinically Useful in Predicting Prognosis in Hypertrophic Cardiomyopathy**. *Circulation* (2010) **122** 2441-2450. DOI: 10.1161/CIRCULATIONAHA.110.954446
8. Witjas-Paalberends E.R., Ferrara C., Scellini B., Piroddi N., Montag J., Tesi C., Stienen G.J.M., Michels M., Ho C.Y., Kraft T.. **Faster Cross-Bridge Detachment and Increased Tension Cost in Human Hypertrophic Cardiomyopathy with the R403Q MYH7 Mutation**. *J. Physiol.* (2014) **592** 3257-3272. DOI: 10.1113/jphysiol.2014.274571
9. Ranjbarvaziri S., Kooiker K.B., Ellenberger M., Fajardo G., Zhao M., Vander Roest A.S., Woldeyes R.A., Koyano T.T., Fong R., Ma N.. **Altered Cardiac Energetics and Mitochondrial Dysfunction in Hypertrophic Cardiomyopathy**. *Circulation* (2021) **144** 1714-1731. DOI: 10.1161/CIRCULATIONAHA.121.053575
10. Anderson R.L., Trivedi D.V., Sarkar S.S., Henze M., Ma W., Gong H., Rogers C.S., Gorham J.M., Wong F.L., Morck M.M.. **Deciphering the Super Relaxed State of Human β-Cardiac Myosin and the Mode of Action of Mavacamten from Myosin Molecules to Muscle Fibers**. *Proc. Natl. Acad. Sci. USA* (2018) **115** E8143-E8152. DOI: 10.1073/pnas.1809540115
11. McNamara J.W., Li A., Lal S., Bos J.M., Harris S.P., van der Velden J., Ackerman M.J., Cooke R., Remedios C.G.. **dos MYBPC3 Mutations Are Associated with a Reduced Super-Relaxed State in Patients with Hypertrophic Cardiomyopathy**. *PLoS ONE* (2017) **12**. DOI: 10.1371/journal.pone.0180064
12. Coppini R., Ferrantini C., Mugelli A., Poggesi C., Cerbai E.. **Altered Ca2+ and Na+ Homeostasis in Human Hypertrophic Cardiomyopathy: Implications for Arrhythmogenesis**. *Front. Physiol.* (2018) **9** 1391. DOI: 10.3389/fphys.2018.01391
13. Wijnker P.J.M., van der Velden J.. **Mutation-Specific Pathology and Treatment of Hypertrophic Cardiomyopathy in Patients, Mouse Models and Human Engineered Heart Tissue**. *Biochim. Biophys. Acta (BBA)-Mol. Basis Dis.* (2020) **1866** 165774. DOI: 10.1016/j.bbadis.2020.165774
14. Cohn R., Thakar K., Lowe A., Ladha F.A., Pettinato A.M., Romano R., Meredith E., Chen Y.-S., Atamanuk K., Huey B.D.. **A Contraction Stress Model of Hypertrophic Cardiomyopathy Due to Sarcomere Mutations**. *Stem Cell Rep.* (2019) **12** 71-83. DOI: 10.1016/j.stemcr.2018.11.015
15. Wijnker P.J.M., Sequeira V., Kuster D.W.D., van der Velden J.. **Hypertrophic Cardiomyopathy: A Vicious Cycle Triggered by Sarcomere Mutations and Secondary Disease Hits**. *Antioxid. Redox Signal.* (2019) **31** 318-358. DOI: 10.1089/ars.2017.7236
16. Sequeira V., Bertero E., Maack C.. **Energetic Drain Driving Hypertrophic Cardiomyopathy**. *FEBS Lett.* (2019) **593** 1616-1626. DOI: 10.1002/1873-3468.13496
17. Jung W.-I., Sieverding L., Breuer J., Hoess T., Widmaier S., Schmidt O., Bunse M., van Erckelens F., Apitz J., Lutz O.. **31P NMR Spectroscopy Detects Metabolic Abnormalities in Asymptomatic Patients With Hypertrophic Cardiomyopathy**. *Circulation* (1998) **97** 2536-2542. DOI: 10.1161/01.CIR.97.25.2536
18. Senthil V., Chen S.N., Tsybouleva N., Halder T., Nagueh S.F., Willerson J.T., Roberts R., Marian A.J.. **Prevention of Cardiac Hypertrophy by Atorvastatin in a Transgenic Rabbit Model of Human Hypertrophic Cardiomyopathy**. *Circ. Res.* (2005) **97** 285-292. DOI: 10.1161/01.RES.0000177090.07296.ac
19. Moore J., Emili A.. **Mass-Spectrometry-Based Functional Proteomic and Phosphoproteomic Technologies and Their Application for Analyzing Ex Vivo and In Vitro Models of Hypertrophic Cardiomyopathy**. *Int. J. Mol. Sci.* (2021) **22**. DOI: 10.3390/ijms222413644
20. Kuzmanov U., Wang E.Y., Vanderlaan R., Kim D.H., Lee S.-H., Hadipour-Lakmehsari S., Guo H., Zhao Y., McFadden M., Sharma P.. **Mapping signalling perturbations in myocardial fibrosis via the integrative phosphoproteomic profiling of tissue from diverse sources**. *Nat. Biomed. Eng.* (2020) **4** 889-900. DOI: 10.1038/s41551-020-0585-y
21. Lian X., Hsiao C., Wilson G., Zhu K., Hazeltine L.B., Azarin S.M., Raval K.K., Zhang J., Kamp T.J., Palecek S.P.. **Robust Cardiomyocyte Differentiation from Human Pluripotent Stem Cells via Temporal Modulation of Canonical Wnt Signaling**. *Proc. Natl. Acad. Sci. USA* (2012) **109** E1848-E1857. DOI: 10.1073/pnas.1200250109
22. Zhang K., Cloonan P.E., Sundaram S., Liu F., Das S.L., Ewoldt J.K., Bays J.L., Tomp S., Toepfer C.N., Marsiglia J.D.C.. **Plakophilin-2 Truncating Variants Impair Cardiac Contractility by Disrupting Sarcomere Stability and Organization**. *Sci. Adv.* (2021) **7** eabh3995. DOI: 10.1126/sciadv.abh3995
23. Toepfer C.N., Garfinkel A.C., Venturini G., Wakimoto H., Repetti G., Alamo L., Sharma A., Agarwal R., Ewoldt J.F., Cloonan P.. **Myosin Sequestration Regulates Sarcomere Function, Cardiomyocyte Energetics, and Metabolism, Informing the Pathogenesis of Hypertrophic Cardiomyopathy**. *Circulation* (2020) **141** 828-842. DOI: 10.1161/CIRCULATIONAHA.119.042339
24. Cantin G.T., Shock T.R., Park S.K., Madhani H.D., Yates J.R.. **Optimizing TiO2-Based Phosphopeptide Enrichment for Automated Multidimensional Liquid Chromatography Coupled to Tandem Mass Spectrometry**. *Anal. Chem.* (2007) **79** 4666-4673. DOI: 10.1021/ac0618730
25. Blum B.C., Emili A.. **Omics Notebook: Robust, Reproducible and Flexible Automated Multiomics Exploratory Analysis and Reporting**. *Bioinform. Adv.* (2021) **1** vbab024. DOI: 10.1093/bioadv/vbab024
26. Candasamy A.J., Haworth R.S., Cuello F., Ibrahim M., Aravamudhan S., Krüger M., Holt M.R., Terracciano C.M.N., Mayr M., Gautel M.. **Phosphoregulation of the Titin-Cap Protein Telethonin in Cardiac Myocytes**. *J. Biol. Chem.* (2014) **289** 1282-1293. DOI: 10.1074/jbc.M113.479030
27. Knöll R., Linke W.A., Zou P., Miočiċ S., Kostin S., Buyandelger B., Ku C.-H., Neef S., Bug M., Schäfer K.. **Telethonin Deficiency Is Associated With Maladaptation to Biomechanical Stress in the Mammalian Heart**. *Circ. Res.* (2011) **109** 758-769. DOI: 10.1161/CIRCRESAHA.111.245787
28. Korotkevich G., Sukhov V., Sergushichev A.. **Fast Gene Set Enrichment Analysis**. *Bioinformatics* (2016). DOI: 10.1101/060012
29. Giacomello M., Pyakurel A., Glytsou C., Scorrano L.. **The Cell Biology of Mitochondrial Membrane Dynamics**. *Nat. Rev. Mol. Cell Biol.* (2020) **21** 204-224. DOI: 10.1038/s41580-020-0210-7
30. Cho E., Jung W., Joo H.-Y., Park E.-R., Kim M.-Y., Kim S.-B., Kim K.S., Lim Y.B., Lee K.H., Shin H.J.. **Cluh Plays a Pivotal Role during Adipogenesis by Regulating the Activity of Mitochondria**. *Sci. Rep.* (2019) **9** 6820. DOI: 10.1038/s41598-019-43410-4
31. Gao J., Schatton D., Martinelli P., Hansen H., Pla-Martin D., Barth E., Becker C., Altmueller J., Frommolt P., Sardiello M.. **CLUH Regulates Mitochondrial Biogenesis by Binding MRNAs of Nuclear-Encoded Mitochondrial Proteins**. *J. Cell Biol.* (2014) **207** 213-223. DOI: 10.1083/jcb.201403129
32. Fransson Å., Ruusala A., Aspenström P.. **The Atypical Rho GTPases Miro-1 and Miro-2 Have Essential Roles in Mitochondrial Trafficking**. *Biochem. Biophys. Res. Commun.* (2006) **344** 500-510. DOI: 10.1016/j.bbrc.2006.03.163
33. Wakim J., Goudenege D., Perrot R., Gueguen N., Desquiret-Dumas V., Chao de la Barca J.M., Dalla Rosa I., Manero F., Le Mao M., Chupin S.. **CLUH Couples Mitochondrial Distribution to the Energetic and Metabolic Status**. *J. Cell Sci.* (2017) **130** 1940-1951. DOI: 10.1242/jcs.201616
34. Yang H., Sibilla C., Liu R., Yun J., Hay B.A., Blackstone C., Chan D.C., Harvey R.J., Guo M.. **Clueless/CLUH Regulates Mitochondrial Fission by Promoting Recruitment of Drp1 to Mitochondria**. *Nat Commun* (2022) **13** 1582. DOI: 10.1038/s41467-022-29071-4
35. Pla-Martín D., Schatton D., Wiederstein J.L., Marx M.-C., Khiati S., Krüger M., Rugarli E.I.. **CLUH Granules Coordinate Translation of Mitochondrial Proteins with MTORC1 Signaling and Mitophagy**. *EMBO J.* (2020) **39** e102731. DOI: 10.15252/embj.2019102731
36. Ashrafi G., Schwarz T.L.. **The Pathways of Mitophagy for Quality Control and Clearance of Mitochondria**. *Cell Death Differ.* (2013) **20** 31-42. DOI: 10.1038/cdd.2012.81
37. Ni H.-M., Williams J.A., Ding W.-X.. **Mitochondrial Dynamics and Mitochondrial Quality Control**. *Redox Biol.* (2015) **4** 6-13. DOI: 10.1016/j.redox.2014.11.006
38. Yamada T., Dawson T.M., Yanagawa T., Iijima M., Sesaki H.. **SQSTM1/P62 Promotes Mitochondrial Ubiquitination Independently of PINK1 and PRKN/Parkin in Mitophagy**. *Autophagy* (2019) **15** 2012-2018. DOI: 10.1080/15548627.2019.1643185
39. Bhujabal Z., Birgisdottir Å.B., Sjøttem E., Brenne H.B., Øvervatn A., Habisov S., Kirkin V., Lamark T., Johansen T.. **FKBP8 Recruits LC3A to Mediate Parkin-Independent Mitophagy**. *EMBO Rep.* (2017) **18** 947-961. DOI: 10.15252/embr.201643147
40. Hirota Y., Yamashita S., Kurihara Y., Jin X., Aihara M., Saigusa T., Kang D., Kanki T.. **Mitophagy Is Primarily Due to Alternative Autophagy and Requires the MAPK1 and MAPK14 Signaling Pathways**. *Autophagy* (2015) **11** 332-343. DOI: 10.1080/15548627.2015.1023047
41. Kraft L.J., Dowler J., Manral P., Kenworthy A.K.. **Size, Organization, and Dynamics of Soluble SQSTM1 and LC3-SQSTM1 Complexes in Living Cells**. *Autophagy* (2016) **12** 1660-1674. DOI: 10.1080/15548627.2016.1199299
42. Tsutsui H., Kinugawa S., Matsushima S.. **Oxidative Stress and Heart Failure**. *Am. J. Physiol. Heart Circ. Physiol.* (2011) **301** H2181-H2190. DOI: 10.1152/ajpheart.00554.2011
43. Szyguła-Jurkiewicz B., Szczurek-Wasilewicz W., Osadnik T., Frycz-Kurek A.M., Macioł-Skurk K., Małyszek-Tumidajewicz J., Skrzypek M., Romuk E., Gąsior M., Banach M.. **Oxidative Stress Markers in Hypertrophic Cardiomyopathy**. *Medicina* (2021) **58**. DOI: 10.3390/medicina58010031
44. Nakamura K., Kusano K.F., Matsubara H., Nakamura Y., Miura A., Nishii N., Banba K., Nagase S., Miyaji K., Morita H.. **Relationship Between Oxidative Stress and Systolic Dysfunction in Patients With Hypertrophic Cardiomyopathy**. *J. Card. Fail.* (2005) **11** 117-123. DOI: 10.1016/j.cardfail.2004.05.005
45. Scherz-Shouval R., Elazar Z.. **ROS, Mitochondria and the Regulation of Autophagy**. *Trends Cell Biol.* (2007) **17** 422-427. DOI: 10.1016/j.tcb.2007.07.009
46. Aiken C.T., Kaake R.M., Wang X., Huang L.. **Oxidative Stress-Mediated Regulation of Proteasome Complexes**. *Mol. Cell. Proteom.* (2011) **10** R110.006924. DOI: 10.1074/mcp.M110.006924
47. Seco-Cervera M., González-Cabo P., Pallardó F.V., Romá-Mateo C., García-Giménez J.L.. **Thioredoxin and Glutaredoxin Systems as Potential Targets for the Development of New Treatments in Friedreich’s Ataxia**. *Antioxidants* (2020) **9**. DOI: 10.3390/antiox9121257
48. Ribas V., García-Ruiz C., Fernández-Checa J.C.. **Glutathione and Mitochondria**. *Front. Pharmacol.* (2014) **5** 151. DOI: 10.3389/fphar.2014.00151
49. Van Houten B., Woshner V., Santos J.H.. **Role of Mitochondrial DNA in Toxic Responses to Oxidative Stress**. *DNA Repair* (2006) **5** 145-152. DOI: 10.1016/j.dnarep.2005.03.002
50. Guo C., Sun L., Chen X., Zhang D.. **Oxidative Stress, Mitochondrial Damage and Neurodegenerative Diseases**. *Neural Regen. Res.* (2013) **8** 2003-2014. DOI: 10.3969/j.issn.1673-5374.2013.21.009
51. Szczesny B., Tann A.W., Longley M.J., Copeland W.C., Mitra S.. **Long Patch Base Excision Repair in Mammalian Mitochondrial Genomes**. *J. Biol. Chem.* (2008) **283** 26349-26356. DOI: 10.1074/jbc.M803491200
52. Krokan H.E., Bjørås M.. **Base Excision Repair**. *Cold Spring Harb. Perspect. Biol.* (2013) **5** a012583. DOI: 10.1101/cshperspect.a012583
53. Prakash A., Doublié S.. **Base Excision Repair in the Mitochondria**. *J. Cell. Biochem.* (2015) **116** 1490-1499. DOI: 10.1002/jcb.25103
54. Campalans A., Kortulewski T., Amouroux R., Menoni H., Vermeulen W., Radicella J.P.. **Distinct Spatiotemporal Patterns and PARP Dependence of XRCC1 Recruitment to Single-Strand Break and Base Excision Repair**. *Nucleic Acids Res.* (2013) **41** 3115-3129. DOI: 10.1093/nar/gkt025
55. Ronson G.E., Piberger A.L., Higgs M.R., Olsen A.L., Stewart G.S., McHugh P.J., Petermann E., Lakin N.D.. **PARP1 and PARP2 Stabilise Replication Forks at Base Excision Repair Intermediates through Fbh1-Dependent Rad51 Regulation**. *Nat. Commun.* (2018) **9** 746. DOI: 10.1038/s41467-018-03159-2
56. Rossi M.N., Carbone M., Mostocotto C., Mancone C., Tripodi M., Maione R., Amati P.. **Mitochondrial Localization of PARP-1 Requires Interaction with Mitofilin and Is Involved in the Maintenance of Mitochondrial DNA Integrity**. *J. Biol. Chem.* (2009) **284** 31616-31624. DOI: 10.1074/jbc.M109.025882
57. Schreiber V., Amé J.-C., Dollé P., Schultz I., Rinaldi B., Fraulob V., Murcia J.M., de Murcia G.. **Poly(ADP-Ribose) Polymerase-2 (PARP-2) Is Required for Efficient Base Excision DNA Repair in Association with PARP-1 and XRCC1**. *J. Biol. Chem.* (2002) **277** 23028-23036. DOI: 10.1074/jbc.M202390200
58. Taguchi N., Ishihara N., Jofuku A., Oka T., Mihara K.. **Mitotic Phosphorylation of Dynamin-Related GTPase Drp1 Participates in Mitochondrial Fission ***. *J. Biol. Chem.* (2007) **282** 11521-11529. DOI: 10.1074/jbc.M607279200
59. Perdiz D., Lorin S., Leroy-Gori I., Poüs C.. **Stress-Induced Hyperacetylation of Microtubule Enhances Mitochondrial Fission and Modulates the Phosphorylation of Drp1 at 616Ser**. *Cell. Signal.* (2017) **39** 32-43. DOI: 10.1016/j.cellsig.2017.07.020
60. Kraus I., Besong Agbo D., Otto M., Wiltfang J., Klafki H.. **Detection and Differentiation of Threonine- and Tyrosine-Monophosphorylated Forms of ERK1/2 by Capillary Isoelectric Focusing-Immunoassay**. *Sci. Rep.* (2015) **5** 12767. DOI: 10.1038/srep12767
61. Roe A.J., Qi X.. **Drp1 Phosphorylation by MAPK1 Causes Mitochondrial Dysfunction in Cell Culture Model of Huntington’s Disease**. *Biochem. Biophys. Res. Commun.* (2018) **496** 706-711. DOI: 10.1016/j.bbrc.2018.01.114
62. Kashatus J.A., Nascimento A., Myers L.J., Sher A., Byrne F.L., Hoehn K.L., Counter C.M., Kashatus D.F.. **Erk2 Phosphorylation of Drp1 Promotes Mitochondrial Fission and MAPK-Driven Tumor Growth**. *Mol. Cell* (2015) **57** 537-551. DOI: 10.1016/j.molcel.2015.01.002
63. Yu R., Liu T., Jin S.-B., Ankarcrona M., Lendahl U., Nistér M., Zhao J.. **MIEF1/2 Orchestrate Mitochondrial Dynamics through Direct Engagement with Both the Fission and Fusion Machineries**. *BMC Biol.* (2021) **19**. DOI: 10.1186/s12915-021-01161-7
64. Ambruso D.R.. **Peroxiredoxin-6 and NADPH Oxidase Activity**. *Methods Enzymol.* (2013) **527** 145-167. DOI: 10.1016/B978-0-12-405882-8.00008-8
65. Kudin A.P., Augustynek B., Lehmann A.K., Kovács R., Kunz W.S.. **The Contribution of Thioredoxin-2 Reductase and Glutathione Peroxidase to H2O2 Detoxification of Rat Brain Mitochondria**. *Biochim. Biophys. Acta (BBA)-Bioenerget.* (2012) **1817** 1901-1906. DOI: 10.1016/j.bbabio.2012.02.023
66. Winn L.M., Kim P.M., Nickoloff J.A.. **Oxidative Stress-Induced Homologous Recombination As a Novel Mechanism for Phenytoin-Initiated Toxicity**. *J. Pharmacol. Exp. Ther.* (2003) **306** 523-527. DOI: 10.1124/jpet.103.052639
67. Chaanine A.H., LeJemtel T.H., Delafontaine P.. **Mitochondrial Pathobiology and Metabolic Remodeling in Progression to Overt Systolic Heart Failure**. *J. Clin. Med.* (2020) **9**. DOI: 10.3390/jcm9113582
68. Blum B.C., Lin W., Lawton M.L., Liu Q., Kwan J., Turcinovic I., Hekman R., Hu P., Emili A.. **Multiomic Metabolic Enrichment Network Analysis Reveals Metabolite–Protein Physical Interaction Subnetworks Altered in Cancer**. *Mol. Cell. Proteom.* (2022) **21** 100189. DOI: 10.1016/j.mcpro.2021.100189
69. Gansemer E.R., McCommis K.S., Martino M., King-McAlpin A.Q., Potthoff M.J., Finck B.N., Taylor E.B., Rutkowski D.T.. **NADPH and Glutathione Redox Link TCA Cycle Activity to Endoplasmic Reticulum Homeostasis**. *iScience* (2020) **23** 101116. DOI: 10.1016/j.isci.2020.101116
70. Ge T., Yang J., Zhou S., Wang Y., Li Y., Tong X.. **The Role of the Pentose Phosphate Pathway in Diabetes and Cancer**. *Front. Endocrinol.* (2020) **11** 365. DOI: 10.3389/fendo.2020.00365
71. Liu M., Zhou X., Li Y., Ma S., Pan L., Zhang X., Zheng W., Wu Z., Wang K., Ahsan A.. **TIGAR Alleviates Oxidative Stress in Brain with Extended Ischemia via a Pentose Phosphate Pathway-Independent Manner**. *Redox Biol.* (2022) **53** 102323. DOI: 10.1016/j.redox.2022.102323
72. Grynberg A., Demaison L.. **Fatty Acid Oxidation in the Heart**. *J. Cardiovasc. Pharmacol.* (1996) **28** S11-S17. DOI: 10.1097/00005344-199600003-00003
73. Baier M.J., Noack J., Seitz M.T., Maier L.S., Neef S.. **Phosphorylation of RyR2 Ser-2814 by CaMKII Mediates Β1-adrenergic Stress Induced Ca2+-leak from the Sarcoplasmic Reticulum**. *FEBS Open Bio* (2021) **11** 2756-2762. DOI: 10.1002/2211-5463.13274
74. Mattiazzi A., Kranias E.. **The Role of CaMKII Regulation of Phospholamban Activity in Heart Disease**. *Front. Pharmacol.* (2014) **5** 5. DOI: 10.3389/fphar.2014.00005
75. McNamara J.W., Singh R.R., Sadayappan S.. **Cardiac Myosin Binding Protein-C Phosphorylation Regulates the Super-Relaxed State of Myosin**. *Proc. Natl. Acad. Sci. USA* (2019) **116** 11731-11736. DOI: 10.1073/pnas.1821660116
76. Sadayappan S., Gulick J., Osinska H., Barefield D., Cuello F., Avkiran M., Lasko V.M., Lorenz J.N., Maillet M., Martin J.L.. **A Critical Function for Ser-282 in Cardiac Myosin Binding Protein-C Phosphorylation and Cardiac Function**. *Circ. Res.* (2011) **109** 141-150. DOI: 10.1161/CIRCRESAHA.111.242560
77. Chen X., Ni F., Kondrashkina E., Ma J., Wang Q.. **Mechanisms of Leiomodin 2-Mediated Regulation of Actin Filament in Muscle Cells**. *Proc. Natl. Acad. Sci. USA* (2015) **112** 12687-12692. DOI: 10.1073/pnas.1512464112
78. Arimura T., Suematsu N., Zhou Y.-B., Nishimura J., Satoh S., Takeshita A., Kanaide H., Kimura A.. **Identification, Characterization, and Functional Analysis of Heart-Specific Myosin Light Chain Phosphatase Small Subunit**. *J. Biol. Chem.* (2001) **276** 6073-6082. DOI: 10.1074/jbc.M008566200
79. Grassie M.E., Moffat L.D., Walsh M.P., MacDonald J.A.. **The Myosin Phosphatase Targeting Protein (MYPT) Family: A Regulated Mechanism for Achieving Substrate Specificity of the Catalytic Subunit of Protein Phosphatase Type 1δ**. *Arch. Biochem. Biophys.* (2011) **510** 147-159. DOI: 10.1016/j.abb.2011.01.018
80. Buraei Z., Yang J.. **Structure and Function of the β Subunit of Voltage-Gated Ca2+ Channels**. *Biochim. Biophys. Acta* (2013) **1828** 1530-1540. DOI: 10.1016/j.bbamem.2012.08.028
81. Lu F., Pu W.T.. **The Architecture and Function of Cardiac Dyads**. *Biophys. Rev.* (2020) **12** 1007-1017. DOI: 10.1007/s12551-020-00729-x
82. Schmidt T.A., Kjeldsen K.. **Human Myocardial Na,K-ATPase—Quantification, Regulation and Relation to Ca**. *Cardiovasc. Res.* (1998) **37** 335-345. DOI: 10.1016/S0008-6363(97)00261-7
83. Knollmann B.C.. **New Roles of Calsequestrin and Triadin in Cardiac Muscle**. *J. Physiol.* (2009) **587** 3081-3087. DOI: 10.1113/jphysiol.2009.172098
84. Terentyev D., Viatchenko-Karpinski S., Györke I., Volpe P., Williams S.C., Györke S.. **Calsequestrin Determines the Functional Size and Stability of Cardiac Intracellular Calcium Stores: Mechanism for Hereditary Arrhythmia**. *Proc. Natl. Acad. Sci. USA* (2003) **100** 11759-11764. DOI: 10.1073/pnas.1932318100
85. Spindler M., Saupe K.W., Christe M.E., Sweeney H.L., Seidman C.E., Seidman J.G., Ingwall J.S.. **Diastolic Dysfunction and Altered Energetics in the AlphaMHC403/+ Mouse Model of Familial Hypertrophic Cardiomyopathy**. *J. Clin. Investig.* (1998) **101** 1775-1783. DOI: 10.1172/JCI1940
86. Chang A.N., Kamm K.E., Stull J.T.. **Role of Myosin Light Chain Phosphatase in Cardiac Physiology and Pathophysiology**. *J. Mol. Cell. Cardiol.* (2016) **101** 35-43. DOI: 10.1016/j.yjmcc.2016.10.004
87. Sheikh F., Lyon R.C., Chen J.. **Functions of Myosin Light Chain-2 (MYL2) in Cardiac Muscle and Disease**. *Gene* (2015) **569** 14-20. DOI: 10.1016/j.gene.2015.06.027
88. Warren S.A., Briggs L.E., Zeng H., Chuang J., Chang E.I., Terada R., Li M., Swanson M.S., Lecker S.H., Willis M.S.. **Myosin Light Chain Phosphorylation Is Critical for Adaptation to Cardiac Stress**. *Circulation* (2012) **126** 2575-2588. DOI: 10.1161/CIRCULATIONAHA.112.116202
89. Sang L., Dick I.E., Yue D.T.. **Protein Kinase A Modulation of CaV1.4 Calcium Channels**. *Nat. Commun.* (2016) **7** 12239. DOI: 10.1038/ncomms12239
90. Wang W., Wang J., Yao K., Wang S., Nie M., Zhao Y., Wang B., Pang H., Xu J., Wu G.. **Metabolic Characterization of Hypertrophic Cardiomyopathy in Human Heart**. *Nat. Cardiovasc. Res.* (2022) **1** 445-461. DOI: 10.1038/s44161-022-00057-1
91. Shimada Y.J., Raita Y., Liang L.W., Maurer M.S., Hasegawa K., Fifer M.A., Reilly M.P.. **Comprehensive Proteomics Profiling Reveals Circulating Biomarkers of Hypertrophic Cardiomyopathy**. *Circ. Heart Fail.* (2021) **14** e007849. DOI: 10.1161/CIRCHEARTFAILURE.120.007849
92. Valko M., Rhodes C.J., Moncol J., Izakovic M., Mazur M.. **Free Radicals, Metals and Antioxidants in Oxidative Stress-Induced Cancer**. *Chem.-Biol. Interact.* (2006) **160** 1-40. DOI: 10.1016/j.cbi.2005.12.009
93. Garrido Ruiz D., Sandoval-Perez A., Rangarajan A.V., Gunderson E.L., Jacobson M.P.. **Cysteine Oxidation in Proteins: Structure, Biophysics, and Simulation**. *Biochemistry* (2022) **61** 2165-2176. DOI: 10.1021/acs.biochem.2c00349
94. Kwon S.H., Pimentel D.R., Remondino A., Sawyer D.B., Colucci W.S.. **H**. *J. Mol. Cell. Cardiol.* (2003) **35** 615-621. DOI: 10.1016/S0022-2828(03)00084-1
95. Lu S.C.. **Glutathione Synthesis**. *Biochim. Biophys. Acta* (2013) **1830** 3143-3153. DOI: 10.1016/j.bbagen.2012.09.008
96. Watanabe K., Nagao M., Toh R., Irino Y., Shinohara M., Iino T., Yoshikawa S., Tanaka H., Satomi-Kobayashi S., Ishida T.. **Critical Role of Glutamine Metabolism in Cardiomyocytes under Oxidative Stress**. *Biochem. Biophys. Res. Commun.* (2021) **534** 687-693. DOI: 10.1016/j.bbrc.2020.11.018
97. Lynch T.L., Sivaguru M., Velayutham M., Cardounel A.J., Michels M., Barefield D., Govindan S., dos Remedios C., van der Velden J., Sadayappan S.. **Oxidative Stress in Dilated Cardiomyopathy Caused by MYBPC3 Mutation**. *Oxid. Med. Cell. Longev.* (2015) **2015** 424751. DOI: 10.1155/2015/424751
98. Bhatti J.S., Bhatti G.K., Reddy P.H.. **Mitochondrial Dysfunction and Oxidative Stress in Metabolic Disorders—A Step towards Mitochondria Based Therapeutic Strategies**. *Biochim. Biophys. Acta (BBA)-Mol. Basis Dis.* (2017) **1863** 1066-1077. DOI: 10.1016/j.bbadis.2016.11.010
99. Singh G., Pachouri U.C., Khaidem D.C., Kundu A., Chopra C., Singh P.. **Mitochondrial DNA Damage and Diseases**. *F1000Research* (2015) **4** 176. DOI: 10.12688/f1000research.6665.1
100. Taylor R.W., Turnbull D.M.. **Mitochondrial DNA mutations in human disease**. *Nat. Rev. Genet.* (2005) **6** 389-402. DOI: 10.1038/nrg1606
101. Zhou B., Tian R.. **Mitochondrial Dysfunction in Pathophysiology of Heart Failure**. *J. Clin. Investig* (2018) **128** 3716-3726. DOI: 10.1172/JCI120849
102. Adebayo M., Singh S., Singh A.P., Dasgupta S.. **Mitochondrial Fusion and Fission: The Fine-tune Balance for Cellular Homeostasis**. *FASEB J.* (2021) **35** e21620. DOI: 10.1096/fj.202100067R
103. Chen H., Vermulst M., Wang Y.E., Chomyn A., Prolla T.A., McCaffery J.M., Chan D.C.. **Mitochondrial Fusion Is Required for MtDNA Stability in Skeletal Muscle and Tolerance of MtDNA Mutations**. *Cell* (2010) **141** 280-289. DOI: 10.1016/j.cell.2010.02.026
104. Pangou E., Bielska O., Guerber L., Schmucker S., Agote-Arán A., Ye T., Liao Y., Puig-Gamez M., Grandgirard E., Kleiss C.. **A PKD-MFF Signaling Axis Couples Mitochondrial Fission to Mitotic Progression**. *Cell Rep.* (2021) **35** 109129. DOI: 10.1016/j.celrep.2021.109129
105. Zhang C., Nie P., Zhou C., Hu Y., Duan S., Gu M., Jiang D., Wang Y., Deng Z., Chen J.. **Oxidative Stress-Induced Mitophagy Is Suppressed by the MiR-106b-93-25 Cluster in a Protective Manner**. *Cell Death Dis.* (2021) **12** 1-16. DOI: 10.1038/s41419-021-03484-3
106. Frohman M.A.. **Role of Mitochondrial Lipids in Guiding Fission and Fusion**. *J. Mol. Med.* (2015) **93** 263-269. DOI: 10.1007/s00109-014-1237-z
107. Paradies G., Paradies V., Ruggiero F.M., Petrosillo G.. **Role of Cardiolipin in Mitochondrial Function and Dynamics in Health and Disease: Molecular and Pharmacological Aspects**. *Cells* (2019) **8**. DOI: 10.3390/cells8070728
108. Van der Velden J., Tocchetti C.G., Varricchi G., Bianco A., Sequeira V., Hilfiker-Kleiner D., Hamdani N., Leite-Moreira A.F., Mayr M., Falcão-Pires I.. **Metabolic Changes in Hypertrophic Cardiomyopathies: Scientific Update from the Working Group of Myocardial Function of the European Society of Cardiology**. *Cardiovasc. Res.* (2018) **114** 1273-1280. DOI: 10.1093/cvr/cvy147
109. Michas C., Karakan M.Ç., Nautiyal P., Seidman J.G., Seidman C.E., Agarwal A., Ekinci K., Eyckmans J., White A.E., Chen C.S.. **Engineering a Living Cardiac Pump on a Chip Using High-Precision Fabrication**. *Sci. Adv.* (2022) **8** eabm3791. DOI: 10.1126/sciadv.abm3791
110. Sharma A., Toepfer C.N., Ward T., Wasson L., Agarwal R., Conner D.A., Hu J.H., Seidman C.E.. **CRISPR/Cas9-Mediated Fluorescent Tagging of Endogenous Proteins in Human Pluripotent Stem Cells**. *Curr. Protoc. Hum. Genet.* (2018) **96**. DOI: 10.1002/cphg.52
111. Larson A., Chin M.T.. **A Method for Cryopreservation and Single Nucleus RNA-Sequencing of Normal Adult Human Interventricular Septum Heart Tissue Reveals Cellular Diversity and Function**. *BMC Med. Genom.* (2021) **14**. DOI: 10.1186/s12920-021-01011-z
112. Codden C.J., Chin M.T.. **Common and Distinctive Intercellular Communication Patterns in Human Obstructive and Nonobstructive Hypertrophic Cardiomyopathy**. *Int. J. Mol. Sci.* (2022) **23**. DOI: 10.3390/ijms23020946
113. Codden C.J., Larson A., Awata J., Perera G., Chin M.T.. **Single Nucleus RNA-Sequencing Reveals Altered Intercellular Communication and Dendritic Cell Activation in Nonobstructive Hypertrophic Cardiomyopathy**. *Cardiol. Cardiovasc. Med.* (2022) **6** 398-415. DOI: 10.26502/fccm.92920277
114. Larson A., Codden C.J., Huggins G.S., Rastegar H., Chen F.Y., Maron B.J., Rowin E.J., Maron M.S., Chin M.T.. **Altered Intercellular Communication and Extracellular Matrix Signaling as a Potential Disease Mechanism in Human Hypertrophic Cardiomyopathy**. *Sci. Rep.* (2022) **12** 5211. DOI: 10.1038/s41598-022-08561-x
115. Ritchie M.E., Phipson B., Wu D., Hu Y., Law C.W., Shi W., Smyth G.K.. **Limma Powers Differential Expression Analyses for RNA-Sequencing and Microarray Studies**. *Nucleic Acids Res.* (2015) **43** e47. DOI: 10.1093/nar/gkv007
116. Blighe K., Rana S., Turkes E., Ostendorf B., Grioni A., Lewis M.. **EnhancedVolcano: Publication-Ready Volcano Plots with Enhanced Colouring and Labeling**. (2023)
117. Yuan M., Kremer D.M., Huang H., Breitkopf S.B., Ben-Sahra I., Manning B.D., Lyssiotis C.A., Asara J.M.. **Ex Vivo and in Vivo Stable Isotope Labelling of Central Carbon Metabolism and Related Pathways with Analysis by LC–MS/MS**. *Nat. Protoc.* (2019) **14** 313-330. DOI: 10.1038/s41596-018-0102-x
118. Toepfer C.N., Sharma A., Cicconet M., Garfinkel A.C., Mücke M., Neyazi M., Willcox J.A.L., Agarwal R., Schmid M., Rao J.. **SarcTrack**. *Circ. Res.* (2019) **124** 1172-1183. DOI: 10.1161/CIRCRESAHA.118.314505
119. Fan T., Kucia M., Jankowski K., Higashi R., Ratajczak J., Ratajczak M., Lane A.. **Rhabdomyosarcoma Cells Show an Energy Producing Anabolic Metabolic Phenotype Compared with Primary Myocytes**. *Mol. Cancer* (2008) **7** 79. DOI: 10.1186/1476-4598-7-79
120. Mcdonald T., Carrasco-Pozo C., Hodson M., Borges K.. **Alterations in Cytosolic and Mitochondrial [U- 13 C]-Glucose Metabolism in a Chronic Epilepsy Mouse Model**. *eNeuro* (2017) **4**. DOI: 10.1523/ENEURO.0341-16.2017
|
---
title: Heat-Killed Enterococcus faecalis Inhibit FL83B Hepatic Lipid Accumulation
and High Fat Diet-Induced Fatty Liver Damage in Rats by Activating Lipolysis through
the Regulation the AMPK Signaling Pathway
authors:
- Jin-Ho Lee
- Keun-Jung Woo
- Joonpyo Hong
- Kwon-Il Han
- Han Sung Kim
- Tack-Joong Kim
journal: International Journal of Molecular Sciences
year: 2023
pmcid: PMC10002555
doi: 10.3390/ijms24054486
license: CC BY 4.0
---
# Heat-Killed Enterococcus faecalis Inhibit FL83B Hepatic Lipid Accumulation and High Fat Diet-Induced Fatty Liver Damage in Rats by Activating Lipolysis through the Regulation the AMPK Signaling Pathway
## Abstract
Continuous consumption of high-calorie meals causes lipid accumulation in the liver and liver damage, leading to non-alcoholic fatty liver disease (NAFLD). A case study of the hepatic lipid accumulation model is needed to identify the mechanisms underlying lipid metabolism in the liver. In this study, the prevention mechanism of lipid accumulation in the liver of *Enterococcus faecalis* 2001 (EF-2001) was extended using FL83B cells (FL83Bs) and high-fat diet (HFD)-induced hepatic steatosis. EF-2001 treatment inhibited the oleic acid (OA) lipid accumulation in FL83B liver cells. Furthermore, we performed lipid reduction analysis to confirm the underlying mechanism of lipolysis. The results showed that EF-2001 downregulated proteins and upregulated AMP-activated protein kinase (AMPK) phosphorylation in the sterol regulatory element-binding protein 1c (SREBP-1c) and AMPK signaling pathways, respectively. The effect of EF-2001 on OA-induced hepatic lipid accumulation in FL83Bs enhanced the phosphorylation of acetyl-CoA carboxylase and reduced the levels of lipid accumulation proteins SREBP-1c and fatty acid synthase. EF-2001 treatment increased the levels of adipose triglyceride lipase and monoacylglycerol during lipase enzyme activation, which, when increased, contributed to increased liver lipolysis. In conclusion, EF-2001 inhibits OA-induced FL83B hepatic lipid accumulation and HFD-induced hepatic steatosis in rats through the AMPK signaling pathway.
## 1. Introduction
As the percentage of high-calorie diets increases and the diet composition of modern society changes [1], overweight and obesity population rates are increasing not only in Korea but also globally [2,3]. Continuous intake of a high-fat diet induces lipid accumulation rather than energy consumption in the body [4]. Excess energy is stored in the form of body fat, which eventually affects lipid metabolism [5]. Regulation of lipid metabolism homeostasis is essential for maintaining the lipid balance in the body [6]. Diabetes, obesity, fatty liver disease, and cardiovascular disease are caused by impaired lipid metabolism [7]. As a result, current research is primarily focused on studying the mechanism underlying lipid metabolism to effectively prevent and improve associated disorders [8,9].
Non-alcoholic fatty liver disease (NAFLD) is associated with obesity and type 2 diabetes dyslipidemia and is used to determine the degree of metabolic syndrome in the liver [10,11]. The lipid accumulation model of liver cells induced with oleic acid (OA) is widely used to obtain baseline data for fatty liver research, and includes FL83B, HepG2, and huH7 cells [12,13,14,15]. OA-induced intracellular lipid accumulation proceeds via the activation of lipid synthesis pathways, such as the sterol regulatory element-binding protein 1c (SREBP-1c) and peroxisome proliferator-activated receptor (PPAR)-γ pathways [16,17]. It also reduces lipolysis processes, such as the activity of AMP-activated protein kinase (AMPK) and lipase. In particular, studies on lipid accumulation in the liver through OA induction in FL83B cells (FL83Bs) and control of the lipid decomposition pathway have primarily focused on the AMPK signaling pathway and lipase enzyme activity [18,19].
AMPK is a vital sensor that uses AMP to generate energy when the energy in the body is depleted. It plays an important role in lipid and carbohydrate metabolism in the liver and is a major factor in the recovery from obesity and diabetes [20,21,22]. In contrast, AMPK decreases lipid accumulation by regulating PPAR-α expression [23].
AMPK activity can induce acetyl-CoA carboxylase (ACC) phosphorylation and decrease ACC activity to suppress lipid biosynthesis [24]. Thus, phosphorylation of AMPK not only maintains energy balance but also inhibits the formation of triglycerides (TGs) to reduce lipid accumulation in the liver. In addition, sirt 1 plays a role in regulating AMPK activity to enhance AMPK phosphorylation in adipocytes and hepatocytes [25].
Hepatic lipid synthesis is performed by the transcription and translation of genes including SREBP-1 and fatty acid synthase (FAS) [26,27]. Hepatocytes activate adipose triglyceride lipase (ATGL), hormone-sensitive lipase (HSL), and monoacylglycerol (MGL) to decompose TGs and form glycerol and free fatty acids during the citric acid cycle for energy production [28,29]. Decomposed free fatty acids stimulate macrophages in the liver to cause an inflammatory reaction and activated macrophages release inflammatory mediators to induce insulin resistance in liver cells [30,31].
Enterococcus faecalis promotes intestinal microbiota balance, alleviates metabolic syndrome, and modulates immunity, among other functions [32]. E. faecalis is also effective in treating hyperlipidemia, obesity, and fatty liver disease [33]. Probiotic strains of E. faecalis have been identified though isolation from fecal samples of healthy individuals [34]. It has been demonstrated that E. faecalis is not only beneficial when alive but is also beneficial when dead [35]. Recently, the genome sequence of EF-2001 was revealed, and it was found to significantly inhibit depression by enhancing pre-frontal local myelination [36]. EF-2001 has been shown to have beneficial effects on human health. These include radioprotective, antitumor, anti-inflammatory, anti-atopic dermatitis, and muscle atrophy prevention [37,38,39,40]. In animal models of prostatic hyperplasia, EF-2001 was also found to be effective [41]. Previous studies have reported that certain products utilizing bacteria, such as *Lactobacillus plantarum* NCU116, *Lactobacillus acidophilus* NX2-6, and other Lactobacillus strains that overexpress bile salt hydrolase, can inhibit hepatic accumulation of lipids [42]. Several other studies have reported the inhibitory effects of bacterial products on lipid accumulation, including products utilizing bacteria, such as Lactobacillus sakei ADM14, L. brevis OPK-3, and L. plantarum LMT1-48 [43,44,45,46]. Recently, we demonstrated that administrating the EF-2001 exhibits an anti-obesity effect in high-fat diet (HFD)-induced rats. Our results showed that the intake of EF-2001 significantly prevented HFD-induced obesity in rats by inhibiting the C/EBP-α and PPAR-γ in the insulin signaling pathway, thus reducing lipid accumulation [47]. Another study reported that heat treatment of E. faecalis FK-23 could ameliorate HFD-induced obesity in mice. The inhibitory effect of FK-23 on hepatic steatosis in HFD-fed mice was induced by the prevention of fat accumulation in the liver through modulation of the activities of genes involved in hepatic fatty acid oxidation [48]. Mishra and Ghosh reported on the synergistic effect of the probiotic E. faecalis AG5 on HFD-induced obesity and the role of propionic acid (PA) in the induction of apoptosis in 3T3-L1 pre-adipocytes [33]. AG5 was found to reduce adipocyte hypertrophy and fatty acid accumulation. This study revealed low PPARγ activity inhibiting 5-LOX, which may be related to adipose apoptosis, and that 5-LOX inhibition increased caspase activity. This is associated with the initiation of cell death [33]. Fan et al. reported that heat-killed E. faecalis improved the abnormal hepatic lipid mechanism in diet-induced obese (DIO) mice by reducing triglyceride (TG) accumulation [49]. This suggests that administrating the EF-2001 may be effective in attenuating hepatic steatosis, as atherogenic dyslipidaemia has been found to be associated with hepatic steatosis, after adjusting for obesity, physical activity, and hyperglycemia [50].
In this study, the effect of heat-killed E. faecalis, EF-2001 on liver lipid accumulation in HFD-induced rats was investigated, and the effects of lipase enzyme activity and AMPK signaling pathways were investigated to provide a new theoretical basis for the treatment of liver lipid metabolic disorders.
## 2.1. EF-2001 Intake Effectively Prevents Fatty Liver Tissue and Liver Damage in HFD-Induced Rats
To establish HFD-induced hepatic steatosis, male rats were divided into SD or HFD groups. Rats were orally administered refined water or EF-2001 in water at each dose per day, as scheduled. HFD groups were subcategorized into three groups (only refined water, 3 mg/kg, or 30 mg/kg EF-2001 in water) to evaluate the effects of EF-2001 on fatty liver-induced rats. We investigated the effects of EF-2001 intake on HFD-induced elevation in non-alcoholic fatty liver disease (NAFLD). Rats fed the HFD weighed significantly higher than rats fed the SD. In the HFD group, brightened and enlarged livers with fat accumulation were observed. Among the HFD groups, the appearance of the liver with accumulated bright-toned fat in the EF-2001 group rats was similar to that of the HFD group (Figure 1A), but the size of the liver was similar to that of the SD group rats. Both groups of HFD rats were administered 3 mg/kg or 30 mg/kg EF-2001 and demonstrated a reduction in liver weight (Figure 1B).
Both glutamic oxaloacetic transaminase (GOT) and glutamic pyruvic transaminase (GPT) levels were significantly increased by the HFD. EF-2001 recovered GOT and GPT levels in both the 3 mg/kg and 30 mg/kg groups, and the levels of GOT and GPT were significantly reduced in both EF-2001 administration groups due to liver damage caused by high-fat diet induction (Figure 1C,D). These results show that EF-2001 intake downregulated HFD-induced fatty liver damage.
## 2.2. Effect of EF-2001 on Oleic Acid-Induced Hepatic Lipid Accumulation in FL83Bs
We measured the hepatic lipid accumulation with or without EF-2001 in FL83Bs to investigate how EF-2001 contributes to lipid accumulation in oleic acid (OA)-induced FL83Bs. The effects of EF-2001 on OA-induced hepatic lipid accumulation in FL83Bs were examined by ORO staining. FL83Bs were pretreated with OA (0.5 mM) in serum-free medium for 48 h and then treated with EF-2001 (0, 25, 50, 100 or 250 μg/mL) for 24 h. OA-induced cells showed significantly increased lipid accumulation in the FL83Bs. However, treatment with EF-2001 (0, 25, 50, 100 or 250 μg/mL) significantly decreased OA-induced lipid accumulation in FL83Bs (Figure 2).
## 2.3. Effect of EF-2001 on Neutral Lipid Droplet of Oleic Acid-Induced FL83Bs Hepatic Lipid Accumulation
We conducted confocal microscopy in OA-induced FL83Bs to investigate how lipogenesis and lipolysis occur during lipid accumulation in EF-2001 (0, 25, 50, 100, or 250 μg/mL). After 48 h of OA induction, lipid synthesis increased in the control group. In the EF-2001-treated group, intracellular neutral lipid droplets decreased for up to 24 h (Figure 3). Therefore, we confirmed that EF-2001 inhibited intracellular lipid accumulation.
## 2.4. Effects of EF-2001 on the Expression of Lipase Enzyme Protein in FL83Bs
After OA induction, ATGL and MGL expressions were observed within 24 h of EF-2001 treatment in FL83Bs. ATGL, an early lipolytic enzyme protein, showed increased expression in EF-2001-treated FL83Bs in a dose-dependent manner. EF-2001 also increased the expression of MGL, a late signal of lipolysis, in a dose-dependent manner in FL83Bs (Figure 4). Additionally, it was found to EF-2001 contributed to lipase activation in OA-induced hepatic lipid accumulation in FL83Bs.
## 2.5. Effects of EF-2001 on the Expression of AMPK and SREBP Signaling Pathway
To identify the mechanism of lipolysis induced by EF-2001, the effects of EF-2001 on AMPK and SREBP signaling pathways were investigated. We compared AMPK signaling pathway-related proteins (AMPK and ACC) and lipid synthesis-related proteins (SREBP-1C and FAS) in OA-induced FL83B hepatocytes. The treatment of EF-2001 significantly increased the expression of phosphorylated AMPK and ACC in a dose-dependent manner (Figure 5B,C). EF-2001 treatment decreased SREBP-1C and FAS expressions in a dose-dependent manner (Figure 5D,E). Thus, these results showed that EF-2001 treatment of hepatic lipid decomposition in lipogenesis of OA-induced FL83Bs was due to activation of the AMPK signaling pathway.
## 2.6. Effects of EF-2001 on AMPK Targeted Signaling Pathway
We conducted experiments to determine the relationship between the expression of lipid-related biomarkers and treatment with EF-2001 due to AMPK signal changes using activators (AICAR) and inhibitors (compound C) of AMPK as AMPK targets. In addition, we observed that AMPK phosphorylation was increased in FL83Bs treated with AICAR. However, FL83Bs treated with AICAR and EF-2001 showed increased ACC phosphorylation and ATGL expression (Figure 6). Interestingly, co-treatment of the AMPK inhibitors, compound C and EF-2001, with FL83Bs resulted in the phosphorylation of AMPK and ACC and the expression of a lipase protein, ATGL. We observed a significant difference in the degree of inhibition of SREBP-1C expression in the EF-2001-treated group (Figure 7).
## 2.7. Effects of EF-2001 AMPK Signaling Pathway on HFD Induced Fatty Liver
Finally, we investigated the effect of EF-2001 on the protein expression level of the AMPK signaling pathway in the liver tissues of SD- and HFD-fed rats. During hepatic lipid accumulation, p-AMPK was up-regulated in the EF-2001 group. In addition, experiments were conducted during lipogenesis to investigate how EF-2001 treatment of the expression levels of lipolysis-related proteins such as p-AMPK, p-ACC, and ATGL (Figure 8A). In the HFD group, AMPK phosphorylation was increased by EF-2001 treatment (Figure 8B). However, ACC phosphorylation did not changed (Figure 8C). Moreover, oral administration of EF-2001 (30 mg/kg) to the HFD group effectively decreased the protein expression level of SREBP-1C to a level lower than that in the untreated EF-2001 HFD group (Figure 8D). In addition, ATGL protein expression increased in the 3 mg/kg and 30 mg/kg EF-2001-treated groups (Figure 8E). However, there was no significant difference in the MGL expression in the HFD group (Figure 8F).
## 3. Discussion
In our previous studies, we reported the effect of downregulation of total cholesterol (such as TG and low-density lipoprotein (LDL)-cholesterol levels), leading to an increased potential of NAFLD in HFD-induced rats by EF-2001 [47]. Downregulation of LDL levels may be an important strategy for the prevention of NAFLD. Since oral administration of EF-2001 suppressed LDL levels, we hypothesized that EF-2001 would have an effect on NAFLD. Therefore, follow-up experiments were conducted to determine the effects and molecular mechanisms underlying NAFLD.
We demonstrated that the oral administration of EF-2001 lowered GOT and GPT levels at both doses (3 mg/kg and 30 mg/kg EF-2001). In addition, both doses reduced the liver weight in HFD-fed rats (Figure 1). Our results indicate that EF-2001 administration decreases liver damage by reducing the physical size of the HFD-induced fatty liver and lowering the levels of enzymes, such as GOT and GPT, released in the blood following liver damage.
FL83Bs are a normal component of liver tissue, and lipid accumulation in FL83Bs plays an important role in the mechanism of lipogenesis, lipolysis, and the onset of NAFLD [51]. We examined the effects of EF-2001 on lipid accumulation in FL83B hepatocytes stained with ORO, and confirmed that treatment with EF-2001 inhibited hepatic lipid accumulation (Figure 2). Furthermore, we observed a reduction in lipid droplets following EF-2001 treatment in FL83B hepatocytes (Figure 3).
To confirm the anti-lipid accumulation effect of EF-2001, we analyzed the molecular mechanisms by which EF-2001 inhibits lipid accumulation in FL83Bs. Several studies have noted the importance of lipolytic enzymes such as ATGL and MGL in the regulation of hepatic lipid accumulation [52,53]. Our results also showed that ATGL and MGL levels increased in a dose-dependent manner upon EF-2001 treatment, proving that EF-2001 contributes to an increase in lipolytic enzyme expression in FL83B hepatocytes (Figure 4).
Hence, we hypothesized that EF-2001 could reduce lipid synthesis and increase lipolysis by activating the AMPK pathway to inhibit the development of fatty liver cells. EF-2001 inhibited the expression of SREBP-1C signaling pathway unit proteins, such as SREBP-1C and FAS, which mediate the lipid accumulation process (Figure 5). We found that EF-2001 can significantly enhance AMPK phosphorylation but can also enhance ACC phosphorylation to inhibit the synthesis of fatty acid chains (Figure 5).
AMPK is an energy regulator that assists in regulating glucose and lipid metabolism to maintain the cellular energy balance [54]. Recent studies have also reported a relationship between the AMPK signaling pathway and lipolysis [55]. AMPK activation induces lipolysis in hepatocytes. It also plays a crucial role in lipolysis progression [56,57]. The SREBP-1C pathway is downregulated by phosphorylated AMPK [55]. ACC-1 is responsible for synthesizing fatty acids and can be controlled by inhibitory phosphorylation by AMPK [58]. Confirming the relationship between AMPK and EF-2001, recovered or increased ACC and AMPK phosphorylation in hepatocytes treated with compound C, an AMPK inhibitor, or AICAR, an AMPK activator (Figure 6 and Figure 7). This is related to the inhibition of the AMPK signaling pathway, which is a key factor in lipogenesis. Consequently, EF-2001 induced the phosphorylation of AMPK and many kinds of lipases and inhibited the expression of the SREBP-1C signaling pathway in HFD-induced obese rats (Figure 8). Therefore, we suggest that EF-2001 inhibits fatty liver cell development, and it can cause the inhibition of NAFLD through AMPK phosphorylation.
In addition to the metabolites produced by specific members of the microbial community, there are also metabolites that are consumed or transformed by bacteria outside the microbial community, making gut microbial metabolism a highly complex process. Although the composition of the microbial community determines the metabolism of microorganisms in the intestine, the substrates available to the microbial community is the most important factor since the metabolites produced from specific substrates reflect gut microbial metabolism [59].
When food is consumed, certain components containing choline groups in the food are metabolized by GM in the intestine and produce GM-derived products such as trimethylamine (TMA), short-chain fatty acids (SCFAs), and trimethylamine N-oxide (TMAO). Metabolites, such as TMA and TMAO, have been identified as the causative agents of metabolic diseases in animal models and human clinical studies. Recent research has revealed that GM and its metabolites play an important role in the development and progression of cardiovascular and metabolic diseases [60,61,62,63,64]. Therefore, in the process of developing drugs for obesity and metabolic diseases, the metabolic pathways in which these GM-derived metabolites are synthesized can be considered as more important as they can become new therapeutic target sites. Currently, our results do not identify the bacteria capable of modulating the host TMA/TMAO and this aspect needs to be analyzed in future studies.
In some studies, the relationship between Enterococcus and liver injury has been described. Ray et al. reported that the ratio of E. faecalis associated with cytolysin was increased in the feces of patients with alcoholic fatty liver, while Lang et al. reported that the ratio of E. faecalis associated with cytolysin increased in non-alcoholic fatty liver patients, although this finding is under debate [65,66]. In a study by Tan, nonalcoholic fatty liver was inhibited by reducing Roseburia, Intestinibacter, and Enterococcus [67]. There are various other claims; therefore, more research is required. Since the gut microbiome changes dramatically even with a high-fat diet, we are currently conducting a clinical trial to analyze the effects of fecal E. faecalis on obesity and metabolic diseases.
Intestinal microbes have been implicated in the pathogenesis of gastrointestinal diseases and metabolic syndromes, such as obesity. Microbes can also produce metabolites, genetic products, and exhibit pathogenic potential that can negatively affect the host [68]. Therefore, it is important to investigate the potential negative effects of consuming heat-treated dead cells of Enterococcus in future studies.
In order to calculate the balance of benefits and harms caused by the gut microbiome from the host’s perspective, a comprehensive analysis of the distribution, diversity, species composition, and metabolites of the microbiome must be performed. For example, the production of SCFA and vitamins by the gut microbiome has positive effects on energy supply and nutrition, but if this process is out of the normal range, it can lead to disease. Although there are no studies on E. faecalis in the intestine with a high-fat diet, we are currently conducting clinical trials, and we plan to analyze the distribution of E. faecalis in the feces, changes in structural function, and altered function.
We recently discovered that E. faecalis EF-2001 inhibits toll-like receptor (TLR) signaling via anti-inflammatory effects [39]. This suggests the possibility of suppressing metabolic syndromes such as obesity and NAFLD. When the composition of the gut microbiome changes, the uptake of TLR4 ligands (e.g., LPS) and TLR9 ligands (e.g., bacterial DNA) is increased and delivered to the liver through the portal vein. Therefore, blocking TLR signaling in the liver could suppress the expression of metabolic syndromes such as obesity, NAFLD, and NASH [69]. Microbial-derived SCFA (acetate, butyrate, and propionate) may be beneficial to the host as sources of carbon and energy. In fact, many studies on the beneficial effects of SCFA on obesity, appetite, and inflammation in the colon have been published [70,71,72]. The structure of heat killed E. faecalis or the secreted substances that modulate the activity of the lipid pathway were not identified in this study. However, these findings provide a starting point for further research on the potential association between heat-killed E. faecalis and metabolic diseases such as obesity and fatty liver.
We concluded that EF-2001 directly lowered hepatic lipid accumulation through the regulation of AMPK signaling. In conclusion, our study suggests that EF-2001 may be a promising candidate for reducing various diseases, including liver damage in obese individuals, by decreasing liver lipid accumulation. However, further research is needed to identify the specific components of EF-2001 that modulate lipogenesis and lipolysis mechanisms to gain a better understanding of its potential therapeutic effects.
As next-generation sequencing became common after the 2000s, metagenomic research has become increasingly popular. This analysis method has shown that various factors, such as diet, race, age, antibiotics, stress, psychological factors, maternal health, birth methods such as natural childbirth, environmental factors, and exercise affect the distribution of intestinal microbes [73]. However, research on each of these factors is still incomplete, and more studies are needed to fully understand their impact on the gut microbiome.
## 4.1. Preparation of Heat-Killed Enterococcus faecalis (EF-2001)
EF-2001, originating from human feces, is a merchantable parabiotic purified from Bereum Co., Ltd. (Wonju, Republic of Korea) and supplied as a heat-killed, dried powder. Prior to being heat-killed, dried EF-2001 contained 7.5 × 1012 units per gram.
## 4.2. Chemical Reagent
5-aminoimidazole-4-carboxamide-1-β-D-ribofuranoside (AICAR, an AMPK activator) and compound C (an AMPK inhibitor) were purchased from Sigma-Aldrich (St. Louis, MO, USA).
## 4.3. Animal Experiments
Twenty-four male Sprague–Dawley rats (3-week-old) with an initial body weight of 30–40 g were purchased from Orient Bio Tech Laboratories (Gyeonggi, Republic of Korea), and, used in the experiments. The rats were acclimatized for one week, and food intake measurements were initiated. The rats were divided into four groups ($$n = 6$$/group) according to their diet. For experimental procedures in HFD-induced obese rats, the rats in the standard diet (SD) group were fed a commercial diet (5L79, Lab Diet Inc., St. Louis, MO, USA) and those in the HFD group were fed an HFD diet (D12492, Research Diets, Inc., New Brunswick, NJ, USA) for six weeks. Rats were subcategorized into three groups: water only, 3 mg/kg EF-2001, and 30 mg/kg EF-2001 in water. Water and food were provided ad libitum, all times. Rats in the HFD group were orally administered pure water or EF-2001 (3 or 30 mg/kg) in water once daily. Gavage was continued for 6 weeks. All experimental procedures were approved by the Institutional Animal Care and Use Committee of Yonsei University and were performed in accordance with approved guidelines (YWCI-202102-003-01).
## 4.4. Serological Analysis
Blood serum was sampled at six weeks by heart puncture under ether anesthesia using a sterilized vacutainer tube. Serum samples were analyzed to determine the activities of hepatoenzymes, including alanine aminotransferase (ALT) and aspartate transaminase (AST), using ALT and AST detection kits purchased from Asan Pharmaceutical (Seoul, Republic of Korea). The kits were used in accordance with the manufacturer’s instructions.
## 4.5. Cell Culture and Induced Fatty Liver Cells
FL83Bs, purchased from the American Type Culture Collection, were maintained in F12K medium, supplemented with $10\%$ fetal bovine serum, $1\%$ penicillin, and $1\%$ streptomycin (Sigma-Aldrich). Cells were cultured at 37 °C in an incubator with $5\%$ CO2. FL83Bs were seeded in a complete medium for 24 h and then incubated with OA (0.5 mM) for 48 h to induce lipid accumulation. The cells were treated with or without EF-2001 (0, 25, 50, 100 or 250 μg/mL) for 24 h to analyze the experimental results.
## 4.6. Oil Red O Staining of FL83B Hepatocyte
Lipid accumulation was determined by Oil red O (ORO) staining. EF-2001 was treated with differentiation induction medium at doses of 0, 25, 50, 100, and 250 μg/mL at $100\%$ cell confluence, followed by the induction of lipid accumulation in FL83Bs. FL83Bs were washed with phosphate-buffered saline (PBS), fixed with $3.7\%$ formaldehyde (Junsei Chemical, Tokyo, Japan) diluted in PBS, and stained with $60\%$ ORO diluted in distilled water. Once the stain was eluted with $100\%$ isopropanol, lipid accumulation was quantified at 490 nm wavelength using a microplate reader (Molecular Devices, San Jose, CA, USA). The results are presented in the graphs. The percentage of ORO staining was relative to that of the untreated control cells, representing the percentage of stained intracellular lipid droplets.
## 4.7. Western Blot Analysis
FL83Bs and liver tissues of HFD-induced rats were treated with EF-2001, and each protein was added to lysis buffer (iNtRON Biotechnology Inc., Sungnam, Republic of Korea) at the appropriate stage of hepatic lipid accumulation. After sonication, proteins were quantified and tested using the Bradford assay (Bio-Rad, Hercules, CA, USA) for Western blotting. The sodium dodecyl sulfate–polyacrylamide gel electrophoresis ratio was determined based on the molecular weight of the protein. Electrophoresis was performed at 100 V for approximately 2 h. Antibody treatment was performed with primary antibodies (SREBP-1C, P-AMPK, AMPK, FAS, P-ACC, ACC, ATGL, P-HSL, MGL, CD36, and β-actin) at a rate of 1:2500 overnight at 4 °C. The membrane was washed three times with tris-buffered saline solution containing Tween 20 for 10 min, and then, secondary antibodies were added at a ratio of 1:5000 for 2 h at room temperature (RT). The transferred protein band on the polyvinylidene difluoride membrane was visualized using an LAS 4000 system (GE Healthcare, Little Chalfont, UK) by inducing an enhanced chemiluminescence reaction. Antibody treatment was performed using primary antibodies, and the signal intensity was quantified using ImageJ software (NIH).
## 4.8. Confocal Microscopy
For confocal microscopy, FL83B hepatocytes were cultured in a 3 cm plate of cover glass (Mattek Corp, Lauda-Königshofen, Baden, Germany), differentiated, and treated with a dose of EF-2001. To facilitate observation of the nucleus, cells were cultured for 10 min by immobilizing paraformaldehyde at RT with fluorescent 4′,6-diamidino-2-phenylindole diluted in PBS. To visualize triglycerol, the cells were incubated with fluorescent BODIPY $\frac{493}{503}$ dye GFP (Thermo Fisher Scientific, Waltham, MA, USA) diluted in PBS for 30 min at RT with paraformaldehyde fixation. GFP expression was visualized using a LSM710 confocal microscope (Carl Zeiss, Oberkochen, Germany).
## 4.9. Statistical Analysis
The experimental results are expressed as mean ± standard error (SE). Analysis of variance and paired or unpaired t-tests were performed for statistical analysis, as appropriate. Statistical p-value < 0.05 was considered statistically significant. All experiments were performed at least three times.
## 5. Conclusions
In this study, we identified the mechanism of EF-2001 in hepatic lipid accumulation in OA-induced FL83Bs. EF-2001 inhibited hepatic lipid accumulation in OA-induced FL83Bs, notably regulating lipogenesis by activating the AMPK signaling pathway in FL83B lipid accumulation. EF-2001 reduced the phosphorylation of downstream signals, such as SREBP-1C and FAS, through the SREBP-1C signaling pathway. However, EF-2001 induced the activation of lipase enzymes, such as ATGL and MGL. In addition, AMPK phosphorylation was increased by EF-2001 in AICAR and compound C treatments that targeted AMPK. Based on our results, we found that EF-2001 targeted the AMPK signaling pathway during FL83B hepatic lipid accumulation (Figure 9). We suggest that EF-2001 regulate HFD-induced abnormal hepatic lipid accumulation in the liver. Therefore, EF-2001 may prevent NAFLD associated with hepatic lipid accumulation and could potentially be used as a basis for the treatment of these diseases.
## References
1. Chang H.-H., Moro A., Takakura K., Su H.-Y., Mo A., Nakanishi M., Waldron R.T., French S.W., Dawson D.W., Hines O.J.. **Incidence of pancreatic cancer is dramatically increased by a high fat, high calorie diet in KrasG12D mice**. *PLoS ONE* (2017) **12**. DOI: 10.1371/journal.pone.0184455
2. Kim J.H., Moon J.S.. **Secular trends in pediatric overweight and obesity in Korea**. *J. Obes. Metab. Syndr.* (2020) **29** 12. DOI: 10.7570/jomes20002
3. Yang H.K., Han K., Cho J.-H., Yoon K.-H., Cha B.-Y., Lee S.-H.. **Ambient temperature and prevalence of obesity: A nationwide population-based study in Korea**. *PLoS ONE* (2015) **10**. DOI: 10.1371/journal.pone.0141724
4. Qiao Y., Sun J., Xia S., Li L., Li Y., Wang P., Shi Y., Le G.. **Effects of different Lactobacillus reuteri on inflammatory and fat storage in high-fat diet-induced obesity mice model**. *J. Funct. Foods* (2015) **14** 424-434. DOI: 10.1016/j.jff.2015.02.013
5. Ducharme N.A., Bickel P.E.. **Minireview: Lipid droplets in lipogenesis and lipolysis**. *Endocrinology* (2008) **149** 942-949. DOI: 10.1210/en.2007-1713
6. Lehrskov L.L., Christensen R.H.. **The role of interleukin-6 in glucose homeostasis and lipid metabolism**. *Seminars in Immunopathology* (2019) 491-499
7. Bardini G., Rotella C.M., Giannini S.. **Dyslipidemia and diabetes: Reciprocal impact of impaired lipid metabolism and Beta-cell dysfunction on micro-and macrovascular complications**. *Rev. Diabet. Stud. RDS* (2012) **9** 82. DOI: 10.1900/RDS.2012.9.82
8. Shin M.-R., Shin S.H., Roh S.-S.. *Can. J. Gastroenterol. Hepatol.* (2020) **2020** 8812634. DOI: 10.1155/2020/8812634
9. Qiu X., Ye Q., Sun M., Wang L., Tan Y., Wu G.. **Saturated hydrogen improves lipid metabolism disorders and dysbacteriosis induced by a high-fat diet**. *Exp. Biol. Med.* (2020) **245** 512-521. DOI: 10.1177/1535370219898407
10. Targher G., Corey K.E., Byrne C.D., Roden M.. **The complex link between NAFLD and type 2 diabetes mellitus—Mechanisms and treatments**. *Nat. Rev. Gastroenterol. Hepatol.* (2021) **18** 599-612. DOI: 10.1038/s41575-021-00448-y
11. Rodriguez L.A., Kanaya A.M., Shiboski S.C., Fernandez A., Herrington D., Ding J., Bradshaw P.T.. **Does NAFLD mediate the relationship between obesity and type 2 diabetes risk? evidence from the multi-ethnic study of atherosclerosis (MESA)**. *Ann. Epidemiol.* (2021) **63** 15-21. DOI: 10.1016/j.annepidem.2021.07.005
12. Huang W.-C., Liao P.-C., Huang C.-H., Hu S., Huang S.-C., Wu S.-J.. **Osthole attenuates lipid accumulation, regulates the expression of inflammatory mediators, and increases antioxidants in FL83B cells**. *Biomed. Pharmacother.* (2017) **91** 78-87. DOI: 10.1016/j.biopha.2017.04.051
13. Lee H., Lee M., Kim H.S.. **Effects of Micro-current stimulation on lipid metabolism in oleic acid-induced non-alcoholic fatty liver disease in FL83B cells**. *J. Biomed. Eng. Res.* (2022) **43** 1-10
14. Huang W.-C., Chen Y.-L., Liu H.-C., Wu S.-J., Liou C.-J.. **Ginkgolide C reduced oleic acid-induced lipid accumulation in HepG2 cells**. *Saudi Pharm. J.* (2018) **26** 1178-1184. DOI: 10.1016/j.jsps.2018.07.006
15. Ricchi M., Odoardi M.R., Carulli L., Anzivino C., Ballestri S., Pinetti A., Fantoni L.I., Marra F., Bertolotti M., Banni S.. **Differential effect of oleic and palmitic acid on lipid accumulation and apoptosis in cultured hepatocytes**. *J. Gastroenterol. Hepatol.* (2009) **24** 830-840. DOI: 10.1111/j.1440-1746.2008.05733.x
16. Yeh Y.-T., Cho Y.-Y., Hsieh S.-C., Chiang A.-N.. **Chinese olive extract ameliorates hepatic lipid accumulation in vitro and in vivo by regulating lipid metabolism**. *Sci. Rep.* (2018) **8** 1057. DOI: 10.1038/s41598-018-19553-1
17. Liou C.-J., Wei C.-H., Chen Y.-L., Cheng C.-Y., Wang C.-L., Huang W.-C.. **Fisetin protects against hepatic steatosis through regulation of the Sirt1/AMPK and fatty acid β-oxidation signaling pathway in high-fat diet-induced obese mice**. *Cell. Physiol. Biochem.* (2018) **49** 1870-1884. DOI: 10.1159/000493650
18. Uen W.C., Shi Y.C., Choong C.Y., Tai C.J.. **Cordycepin suppressed lipid accumulation via regulating AMPK activity and mitochondrial fusion in hepatocytes**. *J. Food Biochem.* (2018) **42** e12569. DOI: 10.1111/jfbc.12569
19. Liou C.-J., Wu S.-J., Shen S.-C., Chen L.-C., Chen Y.-L., Huang W.-C.. **Acacetin Protects against Non-Alcoholic Fatty Liver Disease by Regulating Lipid Accumulation and Inflammation in Mice**. *Int. J. Mol. Sci.* (2022) **23**. DOI: 10.3390/ijms23094687
20. Kang J.I., Hong J.-Y., Lee H.-J., Bae S.Y., Jung C., Park H.J., Lee S.K.. **Anti-tumor activity of yuanhuacine by regulating AMPK/mTOR signaling pathway and actin cytoskeleton organization in non-small cell lung cancer cells**. *PLoS ONE* (2015) **10**. DOI: 10.1371/journal.pone.0144368
21. Chen L., Wang J., You Q., He S., Meng Q., Gao J., Wu X., Shen Y., Sun Y., Wu X.. **Activating AMPK to restore tight junction assembly in intestinal epithelium and to attenuate experimental colitis by metformin**. *Front. Pharmacol.* (2018) **9** 761. DOI: 10.3389/fphar.2018.00761
22. Yao H., Tao X., Xu L., Qi Y., Yin L., Han X., Xu Y., Zheng L., Peng J.. **Dioscin alleviates non-alcoholic fatty liver disease through adjusting lipid metabolism via SIRT1/AMPK signaling pathway**. *Pharmacol. Res.* (2018) **131** 51-60. DOI: 10.1016/j.phrs.2018.03.017
23. Wang L., Chen L., Tan Y., Wei J., Chang Y., Jin T., Zhu H.. **Betaine supplement alleviates hepatic triglyceride accumulation of apolipoprotein E deficient mice via reducing methylation of peroxisomal proliferator-activated receptor alpha promoter**. *Lipids Health Dis.* (2013) **12** 34. DOI: 10.1186/1476-511X-12-34
24. Xi Y., Wu M., Li H., Dong S., Luo E., Gu M., Shen X., Jiang Y., Liu Y., Liu H.. **Baicalin attenuates high fat diet-induced obesity and liver dysfunction: Dose-response and potential role of CaMKKβ/AMPK/ACC pathway**. *Cell. Physiol. Biochem.* (2015) **35** 2349-2359. DOI: 10.1159/000374037
25. Ding R.-B., Bao J., Deng C.-X.. **Emerging roles of SIRT1 in fatty liver diseases**. *Int. J. Biol. Sci.* (2017) **13** 852. DOI: 10.7150/ijbs.19370
26. Quan H.Y., Kim S.J., Jo H.K., Kim G.W., Chung S.H.. **Betulinic acid alleviates non-alcoholic fatty liver by inhibiting SREBP1 activity via the AMPK–mTOR–SREBP signaling pathway**. *Biochem. Pharmacol.* (2013) **85** 1330-1340. DOI: 10.1016/j.bcp.2013.02.007
27. Zhang H., Cook J., Nickel J., Yu R., Stecker K., Myers K., Dean N.M.. **Reduction of liver Fas expression by an antisense oligonucleotide protects mice from fulminant hepatitis**. *Nat. Biotechnol.* (2000) **18** 862-867. DOI: 10.1038/78475
28. Tardelli M., Bruschi F.V., Trauner M.. **The role of metabolic lipases in the pathogenesis and management of liver disease**. *Hepatology* (2020) **72** 1117-1126. DOI: 10.1002/hep.31250
29. Han C., Wen X., Zheng Q., Li H.. **Effect of starvation on activities and mRNA expression of lipoprotein lipase and hormone-sensitive lipase in tilapia (**. *Areus). Fish Physiol. Biochem.* (2011) **37** 113-122. DOI: 10.1007/s10695-010-9423-6
30. Koyama Y., Brenner D.A.. **Liver inflammation and fibrosis**. *J. Clin. Investig.* (2017) **127** 55-64. DOI: 10.1172/JCI88881
31. Kraegen E.W., Clark P.W., Jenkins A.B., Daley E.A., Chisholm D.J., Storlien L.H.. **Development of muscle insulin resistance after liver insulin resistance in high-fat–fed rats**. *Diabetes* (1991) **40** 1397-1403. DOI: 10.2337/diab.40.11.1397
32. Wang K., Cao G., Zhang H., Li Q., Yang C.. **Effects of**. *Food Funct.* (2019) **10** 7844-7854. DOI: 10.1039/C9FO01650C
33. Mishra A.K., Ghosh A.R.. **Probiotic**. *Life Sci.* (2020) **261** 118292. DOI: 10.1016/j.lfs.2020.118292
34. Takahashi K., Nakagawasai O., Nemoto W., Odaira T., Sakuma W., Onogi H., Nishijima H., Furihata R., Nemoto Y., Iwasa H.. **Effect of**. *J. Neuroinflammation* (2019) **16** 201. DOI: 10.1186/s12974-019-1580-7
35. Adams C.A.. **The probiotic paradox: Live and dead cells are biological response modifiers**. *Nutr. Res. Rev.* (2010) **23** 37-46. DOI: 10.1017/S0954422410000090
36. Takahashi K., Kurokawa K., Hong L., Miyagawa K., Mochida-Saito A., Iwasa M., Iwasa H., Nakagawasai O., Tadano T., Takeda H.. **Antidepressant effects of**. *J. Psychiatr. Res.* (2022) **148** 137-148. DOI: 10.1016/j.jpsychires.2022.01.047
37. Choi E.J., Iwasa M., Han K.I., Kim W.J., Tang Y., Hwang Y.J., Kim E.K.. **Heat-killed**. *Nutrients* (2016) **8**. DOI: 10.3390/nu8030146
38. Jeon D.B., Shin H.G., Lee B.W., Jeong S.H., Kim J.H., Ha J.H., Lee I.C.. **Effect of heat-killed**. *Hum. Exp. Toxicol.* (2020) **39** 721-733. DOI: 10.1177/0960327119899987
39. Choi M.-S., Chang S.-J., Chae Y., Lee M.-H., Kim W.-J., Iwasa M., Han K.-I., Kim W.-J., Kim T.-J.. **Anti-inflammatory effect of**. *J. Life Sci.* (2018) **28** 1361-1368
40. Chang S.-J., Lee M.-H., Kim W.-J., Chae Y., Iwasa M., Han K.-I., Kim W.-J., Kim T.-J.. **Effect of heat-killed**. *J. Life Sci.* (2019) **29** 215-222
41. Choi Y.J., Fan M., Tang Y., Iwasa M., Han K.I., Lee H., Hwang J.Y., Lee B., Kim E.K.. **Heat-Killed and Live**. *J. Microbiol. Biotechnol.* (2021) **31** 1134-1143. DOI: 10.4014/jmb.2102.02032
42. Huang W., Wang G., Xia Y., Xiong Z., Ai L.. **Bile salt hydrolase-overexpressing**. *Food Nutr. Res.* (2020) **64**. DOI: 10.29219/fnr.v64.3751
43. Kitadokoro K., Tanaka M., Hikima T., Okuno Y., Yamamoto M., Kamitani S.. **Crystal structure of pathogenic**. *Sci. Rep.* (2020) **10** 5469. DOI: 10.1038/s41598-020-62427-8
44. Choi W.J., Dong H.J., Jeong H.U., Ryu D.W., Song S.M., Kim Y.R., Jung H.H., Kim T.H., Kim Y.H.. *Sci. Rep.* (2020) **10** 869. DOI: 10.1038/s41598-020-57615-5
45. Won S.M., Chen S., Park K.W., Yoon J.H.. **Isolation of lactic acid bacteria from kimchi and screening of**. *LWT* (2020) **126** 109296. DOI: 10.1016/j.lwt.2020.109296
46. Park J.E., Oh S.H., Cha Y.S.. *Nutrients* (2020) **12**. DOI: 10.3390/nu12030604
47. Lee J.-H., Woo K.-J., Kim M.-A., Hong J., Kim J., Kim S.-H., Han K.-I., Iwasa M., Kim T.-J.. **Heat-killed**. *Nutrients* (2022) **14**. DOI: 10.3390/nu14061308
48. Kondoh M., Shimada T., Fukada K., Morita M., Katada K., Higashimura Y., Yoshikawa T.. **Beneficial effects of heat-treated**. *Br. J. Nutr.* (2014) **112** 868-875. DOI: 10.1017/S0007114514001792
49. Fan M., Choi Y.J., Wedamulla N.E., Tang Y., Han K.I., Hwang J.Y., Kim E.K.. **Heat-Killed**. *Foods* (2022) **11**. DOI: 10.3390/foods11040575
50. Makadia S.S., Blaha M., Keenan T., Ndumele C., Jones S., DeFilippis A., Martin S., Kohli P., Conceicao R., Carvalho J.. **Relation of hepatic steatosis to atherogenic dyslipidemia**. *Am. J. Cardiol.* (2013) **112** 1599-1604. DOI: 10.1016/j.amjcard.2013.08.001
51. Wu S.-J., Huang W.-C., Yu M.-C., Chen Y.-L., Shen S.-C., Yeh K.-W., Liou C.-J.. **Tomatidine ameliorates obesity-induced nonalcoholic fatty liver disease in mice**. *J. Nutr. Biochem.* (2021) **91** 108602. DOI: 10.1016/j.jnutbio.2021.108602
52. Sun J., Ji H., Li X.-X., Shi X.-C., Du Z.-Y., Chen L.-Q.. **Lipolytic enzymes involving lipolysis in Teleost: Synteny, structure, tissue distribution, and expression in grass carp (**. *Comp. Biochem. Physiol. Part B Biochem. Mol. Biol.* (2016) **198** 110-118. DOI: 10.1016/j.cbpb.2016.04.008
53. Nam Y.K., Park S.J., Kim M.H., Yang W.M.. **Pharmacopuncture of Taraxacum platycarpum extract reduces localized fat by regulating the lipolytic pathway**. *Biomed. Pharmacother.* (2021) **141** 111905. DOI: 10.1016/j.biopha.2021.111905
54. Hardie D.G., Ashford M.L.. **AMPK: Regulating energy balance at the cellular and whole body levels**. *Physiology* (2014) **29** 99-107. DOI: 10.1152/physiol.00050.2013
55. Li Y., Xu S., Mihaylova M.M., Zheng B., Hou X., Jiang B., Park O., Luo Z., Lefai E., Shyy J.Y.-J.. **AMPK phosphorylates and inhibits SREBP activity to attenuate hepatic steatosis and atherosclerosis in diet-induced insulin-resistant mice**. *Cell Metab.* (2011) **13** 376-388. DOI: 10.1016/j.cmet.2011.03.009
56. Seo M.S., Kim J.H., Kim H.J., Chang K.C., Park S.W.. **Honokiol activates the LKB1–AMPK signaling pathway and attenuates the lipid accumulation in hepatocytes**. *Toxicol. Appl. Pharmacol.* (2015) **284** 113-124. DOI: 10.1016/j.taap.2015.02.020
57. Kuo Y.-T., Lin T.-H., Chen W.-L., Lee H.-M.. **Alpha-lipoic acid induces adipose triglyceride lipase expression and decreases intracellular lipid accumulation in HepG2 cells**. *Eur. J. Pharmacol.* (2012) **692** 10-18. DOI: 10.1016/j.ejphar.2012.07.028
58. Zordoky B.N., Nagendran J., Pulinilkunnil T., Kienesberger P.C., Masson G., Waller T.J., Kemp B.E., Steinberg G.R., Dyck J.R.. **AMPK-dependent inhibitory phosphorylation of ACC is not essential for maintaining myocardial fatty acid oxidation**. *Circ. Res.* (2014) **115** 518-524. DOI: 10.1161/CIRCRESAHA.115.304538
59. Louis P., Scott K.P., Duncan S.H., Flint H.J.. **Understanding the effects of diet on bacterial metabolism in the large intestine**. *J. Appl. Microbiol.* (2007) **102** 1197-1208. DOI: 10.1111/j.1365-2672.2007.03322.x
60. Aron-Wisnewsky J., Clement K.. **The gut microbiome, diet, and links to cardiometabolic and chronic disorders**. *Nat. Rev. Nephrol.* (2016) **12** 169-181. DOI: 10.1038/nrneph.2015.191
61. Tang W.H.W., Wang Z., Levison B.S., Koeth R.A., Britt E.B., Fu X., Wu Y., Hazen S.L.. **Intestinal microbial metabolism of phosphatidylcholine and cardiovascular risk**. *N. Engl. J. Med.* (2013) **368** 1575-1584. DOI: 10.1056/NEJMoa1109400
62. Kitai T., Tang W.H.W.. **Gut microbiota in cardiovascular disease and heart failure**. *Clin. Sci.* (2018) **132** 85-91. DOI: 10.1042/CS20171090
63. Peng J., Xiao X., Zhang X.. **Interaction between gut microbiome and cardiovascular disease**. *Life Sci.* (2018) **214** 153-157. DOI: 10.1016/j.lfs.2018.10.063
64. Tang W.H.W., Hazen S.L.. **The Gut microbiome and its role in cardiovascular diseases**. *Circulation* (2017) **135** 1008-1010. DOI: 10.1161/CIRCULATIONAHA.116.024251
65. Ray K.. **Manipulating the gut microbiota to combat alcoholic hepatitis**. *Nat. Rev. Gastroenterol. Hepatol.* (2020) **17** 3. DOI: 10.1038/s41575-019-0246-3
66. Lang S., Demir M., Duan Y., Martin A., Schnabl B.. **Cytolysin-positive**. *Liver Int.* (2020) **40** 860-865. DOI: 10.1111/liv.14377
67. Tan J., Hu R., Li B., Li Y., Yang X., He Z., Wu S.. **Protocatechuic acid as a phenolic intermediate to ameliorate non-alcoholic fatty liver disease by inhibiting**. *Res. Sq.* (2020). DOI: 10.21203/rs.3.rs-498143/v1
68. Kim J.M.. **Roles of Enteric microbial composition and metabolism in health and diseases**. *Korean J. Gastroenterol.* (2013) **62** 191-205. DOI: 10.4166/kjg.2013.62.4.191
69. Henao-Mejia J., Elinav E., Jin C., Hao L., Mehal W.Z., Strowig T., Thaiss C.A., Kau A.L., Eisenbarth S.C., Jurczak M.J.. **Inflammasome-mediated dysbiosis regulates progression of NAFLD and obesity**. *Nature* (2012) **482** 179-185. DOI: 10.1038/nature10809
70. Furusawa Y., Obata Y., Fukuda S., Endo T.A., Nakato G., Takahashi D., Nakanishi Y., Uetake C., Kato K., Kato T.. **Commensal microbe-derived butyrate induces the differentiation of colonic regulatory T cells**. *Nature* (2013) **504** 446-450. DOI: 10.1038/nature12721
71. Frost G., Sleeth M.L., Sahuri-Arisoylu M., Lizarbe B., Cerdan S., Brody L., Anastasovska J., Ghourab S., Hankir M., Zhang S.. **The short-chain fatty acid acetate reduces appetite via a central homeostatic mechanism**. *Nat. Commun.* (2014) **5** 3611. DOI: 10.1038/ncomms4611
72. Den Besten G., Bleeker A., Gerding A., van Eunen K., Havinga R., van Dijk T.H., Oosterveer M.H., Jonker J.W., Groen A.K., Reijngoud D.J.. **Short-chain fatty acids protect against high-fat diet-induced obesity via a PPARgamma-dependent switch from lipogenesis to fat oxidation**. *Diabetes* (2015) **64** 2398. DOI: 10.2337/db14-1213
73. Osadchiy V., Martin C.R., Mayer E.A.. **The gut brain axis and the microbiome: Mechanisms and clinical implications**. *Clin. Gastroenterol. Hepatol.* (2019) **17** 322-332. DOI: 10.1016/j.cgh.2018.10.002
|
---
title: Compositional Alteration of Gut Microbiota in Psoriasis Treated with IL-23
and IL-17 Inhibitors
authors:
- Yu-Huei Huang
- Lun-Ching Chang
- Ya-Ching Chang
- Wen-Hung Chung
- Shun-Fa Yang
- Shih-Chi Su
journal: International Journal of Molecular Sciences
year: 2023
pmcid: PMC10002560
doi: 10.3390/ijms24054568
license: CC BY 4.0
---
# Compositional Alteration of Gut Microbiota in Psoriasis Treated with IL-23 and IL-17 Inhibitors
## Abstract
Alterations in the gut microbiota composition and their associated metabolic dysfunction exist in psoriasis. However, the impact of biologics on shaping gut microbiota is not well known. This study aimed to determine the association of gut microorganisms and microbiome-encoded metabolic pathways with the treatment in patients with psoriasis. A total of 48 patients with psoriasis, including 30 cases who received an IL-23 inhibitor (guselkumab) and 18 cases who received an IL-17 inhibitor (secukinumab or ixekizumab) were recruited. Longitudinal profiles of the gut microbiome were conducted by using 16S rRNA gene sequencing. The gut microbial compositions dynamically changed in psoriatic patients during a 24-week treatment. The relative abundance of individual taxa altered differently between patients receiving the IL-23 inhibitor and those receiving the IL-17 inhibitor. Functional prediction of the gut microbiome revealed microbial genes related to metabolism involving the biosynthesis of antibiotics and amino acids were differentially enriched between responders and non-responders receiving IL-17 inhibitors, as the abundance of the taurine and hypotaurine pathway was found to be augmented in responders treated with the IL-23 inhibitor. Our analyses showed a longitudinal shift in the gut microbiota in psoriatic patients after treatment. These taxonomic signatures and functional alterations of the gut microbiome could serve as potential biomarkers for the response to biologics treatment in psoriasis.
## 1. Introduction
Psoriasis is an inflammatory skin disease that is associated with many other medical conditions, and affects adults and children worldwide [1]. Overall prevalence ranges from $0.1\%$ in east Asia to $1.5\%$ in western Europe, and is highest in high-income countries [2,3]. Most patients with psoriasis have some detriment to their quality of life attributable to the disease, and many feel a substantial, negative effect on their psychosocial wellbeing. It has been regarded that psoriasis involves the interplay between predisposing genetic and environmental (e.g., infection and antibiotics treatment) factors [1,4,5,6]. Studies have shown that skin and the gut microbiome play a role in modulating the development of chronic plaque psoriasis [7]. Recent evidence revealed a combined increase in Corynebacterium, Propionibacterium, Staphylococcus, and *Streptococcus in* psoriatic plaque sites [7,8]. Gut microbiota is known to play a critical role in the regulation of metabolism, the immune system, and intestine permeability [9]. A disturbed intestinal microbiome was shown to be involved in a number of autoimmune diseases including type 1 diabetes, rheumatoid arthritis, multiple sclerosis, celiac disease, and inflammatory bowel disease (IBD) [10,11]. In psoriasis, similar evidence demonstrated gut dysbiosis with lower diversity and altered relative abundance for certain bacteria [9,12]. Several studies have found the relative abundance of Bacteroidetes was lower and that of Firmicutes was higher in patients with psoriasis compared to healthy controls [12,13,14]. However, an inconsistent result reported by Huang et al., revealed an increased abundance of Bacteroidetes and decreased Firmicutes in psoriasis [15]. These changes in gut microbiota are considered to be crucial causes for initiating or exacerbating psoriasis in humans and animal models [16,17].
Treatment for psoriasis may change the composition of the skin and gut microbiota [18,19,20]. A change in lesional skin microbiota has been associated with a clinical response after balneotherapy [18] and phototherapy [19]. A reduced mean rate of *Staphylococcus aureus* on psoriatic plaques, reaching a nadir at weeks 16–20 after treatment, was noted in our previous research [20]. Regarding the gut microbial change after psoriasis treatment, the relative abundance of Pseudomonadaceae and Enterobacteriaceae increased significantly following secukinumab therapy, while no significant change was noted in gut microbiome composition following ustekinumab treatment [21].
In the past 20 years, findings from immunological and genetic studies have highlighted causal immunological circuits of psoriasis that converge on adaptive immune pathways involving interleukin (IL)-17 and IL-23 [1,22,23]. The suppression of psoriasis-related, proinflammatory, and Th17-associated cytokines, such as tumor necrosis factor (TNF)-α, IL-17A, and IL-23, was observed in mice fed with *Lactobacillus pentosus* [24]. The clinical significance of the interaction between microbiota and the immune system is of importance. Although guselkumab, a selective IL-23 inhibitor, and secukinumab and ixekizumab, monoclonal antibodies targeting IL-17A, were highly effective in treating psoriasis, their treatment results in IBD were not consistent. Clinical trials for biologics blocking either IL-17A or its receptor have contributed to the exacerbation of IBD [25,26]. This raised the possibility that blockade of IL-17 could interfere with the microbiota composition and homeostasis in the intestine that might predispose susceptible individuals to develop IBD [27,28]. Moreover, in a phase 2 trial, guselkumab demonstrated a greater efficacy than a placebo in patients with Crohn’s disease [29]. These findings indicated a sophisticated interaction between gut microbiota composition and biologic therapies. Yet, how gut microbiota in psoriasis react to the IL-17 and IL-23 blockers has scarcely been investigated. Therefore, this study aimed to investigate the dynamic alteration of gut microbiota in psoriasis patients before and after receiving IL-17 and IL-23 antagonists.
## 2.1. Patient Demographic and Characteristics
A total of one hundred and ninety-two fecal samples were obtained from 48 patients with 30 cases receiving the IL-23 inhibitor (guselkumab) (mean age 45.2 years) and 18 cases receiving IL-17 inhibitors (secukinumab and ixekizumab) (mean age 52.8 years). There was no significant difference in gender, weight, psoriatic arthritis, baseline PASI score, and baseline CRP level between the two groups. Patients treated with an IL-17 inhibitor were older than patients treated with an IL-23 inhibitor (Table 1).
The mean PASI scores decreased at weeks 4, 12, and 24 after either IL-23 or IL-17 inhibitor therapy. All these changes from baseline were significant (Figure 1A). In addition, the CRP level was significantly reduced after 12 weeks and 24 weeks of treatment (Figure 1B). Moreover, we found recruited patients did not change their eating habits during the study.
## 2.2. Gut Microbial Diversity in Psoriasis after the Treatment with Il-23 and IL-17 Inhibitors
We studied the temporal alteration of microbial diversity in patients treated with IL-23 or IL-17 inhibitors. Calculation of the weighted-UniFrac distance matrix (β diversity) displayed a significantly altered distance in microbial community structures among samples from patients receiving an IL-23 or IL-17 inhibitor during 24-week treatment, while no significant difference in the α diversity was observed among the groups (Figure 2A). Moreover, Bray–Curtis distance was used to measure β diversity at week 0 and 24 among the responders (R) and non-responders (NR) (Figure 2B). The results showed that β diversity of gut microbiota in the responders to the IL-23 inhibitor was significantly higher than that in non-responders both at baseline and week 24 ($p \leq 0.05$), while there was no significant difference in β diversity between responders and non-responders treated with IL-17 inhibitors.
## 2.3. Altered Composition of Gut Microbiota in Psoriatic Patients after Treatment with IL-23 and IL-17 Inhibitors
We then sought the most relevant taxa whose abundance altered after the treatment (week 24) to explore the effect of biologics on the composition of gut microbiota. In patients treated with the IL-23 inhibitor, we identified five taxa whose levels were significantly different from the baseline (Figure 3). The relative abundance of Roseburia, Lachnoclostridium, Bacteroides vulgatus, Anaerostipes, and Escherichia–Shigella increased over the time course of the treatment. In patients treated with IL-17 inhibitors, levels of *Bacteroides stercoris* and *Parabacteroides merdae* were significantly increased at week 24, while those of Blautia and Roseburia were significantly reduced (Figure 3).
## 2.4. Changes in Relative Abundance of Gut Bacteria between Responders and Non-Responders
Furthermore, we assessed the association between the therapeutic outcome and changes in relative abundance of individual taxa from the baseline to 24 weeks post-treatment. We found that among patients treated with the IL-23 inhibitor for 24 weeks, the relative abundance of Lachnospiraceae and Romboutsia significantly decreased from the baseline in the responders compared to non-responders (Figure 4). Meanwhile, the relative abundance of Fusicatenibacter in patients treated with IL-17 inhibitors for 24 weeks significantly increased compared to non-responders, whereas that of Lachnospiraceae NK4A136 and Roseburia significantly decreased (Figure 4).
## 2.5. Functional Prediction of Gut Microbiome after Treatment with IL-23 and IL-17 Inhibitors
Considering the pathways related to metabolism, we found a number of pathway modules associated with lipid metabolism, inositol phosphate metabolism, and glutathione metabolism enriched in patients treated with the IL-23 inhibitor at week 24. In contrast, bacterial genes assigned to energy metabolism, arginine biosynthesis, cysteine and methionine metabolism, fructose and mannose metabolism, and carbapenem biosynthesis were less abundant. In patients treated with IL-17 inhibitors, the abundance of pathway modules associated with indole alkaloid biosynthesis increased, while that with lysine biosynthesis decreased (Table 2).
In addition, we investigated alterations in microbial functions at week 24 from the baseline between responders and non-responders. Among patients treated with the IL-23 inhibitor, the pathway of taurine and hypotaurine metabolism was enriched in the responders compared to non-responders (Table 3). Among patients treated with IL-17 inhibitors, 13 metabolism pathways were significantly enriched and 3 decreased in responders after 24 weeks of treatment compared with non-responders (Table 3). The pathways involved in amino acids metabolism, biosynthesis of antibiotics, and carbohydrate metabolism were differentially enriched from the baseline after the treatment with IL-17 inhibitors.
## 3. Discussion
In the present study, we analyzed the gut microbial diversities and taxonomies in patients with psoriasis at weeks 4, 12, and 24 after the treatment with IL-23 or IL-17 inhibitors. This is the first study to demonstrate a significant increase in β diversity of gut microbial communities and altered abundance of certain bacteria in patients receiving the IL-23 inhibitor for 24 weeks. In addition, we identified microbial taxa and functional pathways associated with the therapeutic options and treatment responses.
Changes in gut microbiota composition due to therapeutic agents and their influence on clinical response have been reported in patients with inflammatory bowel disease (IBD) [30]. Common types of gut microbiota change after biologics treatment encompassed an increased abundance of short-chain fatty acids (SCFAs)-producing bacteria, which are considered beneficial commensal bacteria [30]. An improvement in intestinal dysbiosis was reported with an increment in the abundance of SCFAs-producing bacteria such as Anaerostipes, Blautia, and Roseburia from IBD patients after receiving infliximab [31]. Moreover, similar findings were also demonstrated in IBD patients receiving ustekinumab [32]. In this study, we found that the relative abundance of Anaerostipes and Roseburia increased in patients after IL-23 inhibitor treatment, which may increase the production of SCFAs and consequently restore the immunomodulatory function and intestinal epithelial barrier [33,34]. Conversely, the abundance of Blautia and Roseburia was reduced in those receiving IL-17 inhibitors. One previous study investigating the impact of secukinumab on gut microbial composition [21] showed a reduction in the abundance of the SCFAs-producing bacteria Firmicutes, consistent with our findings.
The *Bacteroides genus* constitutes $30\%$ of the total colonic bacteria and *Bacteroides vulgatus* is one of the most commonly encountered Bacteroides species in the human gut [35]. The role of B. vulgatus in modulating the immune system has been investigated in animal experiments. Supplementation with B. vulgatus attenuated symptoms of colitis in mice and decreased the expression of TNF-α, IL-1β, and IL-6 in the colon [36]. Moreover, suppression of the systemic and intestinal immune response was observed in mice gavaged with *Bacteroides vulgatus* [37,38]. The present study demonstrated that the relative abundance of *Bacteroides vulgatus* increased after anti-IL-23 inhibitor treatment, which might further imply the beneficial effect of gut immunomodulation by the IL-23 inhibitor in psoriasis.
The gut is considered to be a major immune organ, with gut-associated lymphoid tissue (GALT) being the most complex immune compartment [39]. It is well known that changes in the gut microbial composition may promote both health and disease [40]. Strong evidence has indicated that intestinal dysbiosis is clinically relevant to psoriasis [41]. The importance of the gut–skin axis in the pathogenesis of psoriasis has recently been documented in humans, as well as in animal models. [ 42]. In imiquimod-induced psoriasis-like mice, gut microbiota promoted intestinal and cutaneous inflammations by enhancing the IL-23/IL-17 axis [42,43]. In addition, a gut microbial genus, Romboutsia, increased in mice with imiquimod-induced psoriasis [43], suggesting that IL-23/IL-17-axis-related psoriasis may be associated with levels of gut Romboutsia. Intriguingly, our study revealed that the abundance of Romboutsia significantly decreased at week 24 in the responders to the IL-23 inhibitor when compared with non-responders. However, there was no significant difference in the gut Romboutsia level between responders and non-responders treated with IL-17 inhibitors. Based on these findings, we speculate that blocking IL-23 may ameliorate Romboutsia-mediated psoriasis by improving IL-23/IL-17-axis-related skin inflammation.
At the genus level, an enriched Lachnospiraceae NK4A136 group was detected in patients with ankylosing spondylitis [44] and IBD [45]. Recently, a study on the gut microbiome demonstrated an increase in the abundance of gut Lactobacillaceae in psoriatic patients [13]. Our results further revealed that the abundance of Lachnospiraceae NK4A136 at week 24 significantly decreased in responders to IL-17 inhibitors compared to non-responders. It has been shown that the Lachnospiraceae NK4A136 group is correlated with elevated levels of intestinal IL-17 and IL-6 in mice with diabetes mellitus, resulting in intestinal inflammation [46]. Thus, we hypothesize that responders to IL-17 inhibitors might benefit from reduction in the gut Lachnospiraceae NK4A136 group, which likely contributes to declined skin inflammation. Further investigation should be conducted to address the causal relationship of these findings.
In our study, sixteen KEGG pathways were found to be significantly enriched in responders to IL-17 inhibitors, such as the biosynthesis of amino acids, energy metabolism, and biosynthesis of antibiotics including vancomycin, validamycin, and novobiocin. Previously, dramatic changes in glucose metabolism, amino acid metabolism, and energy metabolism have been shown in psoriasis [47,48]. Metabolic regulation of cell proliferation and apoptosis was thought to be critical for dysregulated keratinocyte hyperproliferation in psoriasis [49,50]. Altogether, these findings suggest that altered gut-microbiota-mediated biosynthesis of amino acids and energy metabolism may also contribute to specific phenotypes in patients with psoriasis, such as uncontrolled keratinocyte hyperproliferation. It was reported that treatment with broad-spectrum antibiotics in mice with imiquimod-induced psoriasis reduced proinflammatory IL-17-producing T cells and skin thickness [16,42]. Moreover, Actinobacteria, isolated from the gut of freshwater fish, exhibited antimicrobial activities by producing antibiotic compounds [51]. Our data showed that gut microbiome-encoded metabolic KEGG pathways enriched in the responders to IL-17 inhibitors were concentrated in the biosynthesis of antibiotics. According to these findings, we suggest that IL-17 inhibitors may partially improve psoriasis-related skin inflammation by enhancing gut-microbiota-mediated biosynthesis of antibiotics.
In addition, reduction in the abundance of the taurine and hypotaurine metabolism pathway in patients with severe psoriasis has been observed in one recent study [52]. Our results demonstrated that the abundance of the taurine and hypotaurine metabolic pathway was significantly enhanced in the responders to the IL-23 inhibitor, as compared with that in non-responders. Taurine, an abundant amino acid in leukocytes, is found in high concentrations in inflammatory lesions and tissues exposed to oxidative stress. [ 53]. Collectively, these findings and our data imply that a shift in gut bacterial composition due to the IL-23 inhibitor could lead to significant changes in taurine metabolism, which may correlate with an improvement in the inflammatory status in patients with psoriasis.
Our results should be considered in the context of several limitations. First, sample sizes were limited, and larger cohorts should be assessed in future studies. Second, due to the relatively limited resolution of the 16S rRNA sequencing technique [54], shotgun metagenomic sequencing methods are needed to identify specific bacterial strains in psoriasis. Third, based on the gut-microbiota-mediated metabolic pathways related to the response to the IL-23 inhibitor or IL-17 inhibitors identified in psoriatic patients, it is necessary to explore their key regulatory targets. Finally, we did not investigate inflammatory markers collected from the peripheral blood, gut, and stool so we could not explain the association of inflammatory changes with the microbial composition.
In summary, treatments with IL-23 and IL-17 inhibitors were associated with distinct shifts in gut microbial composition in patients with psoriasis. Significant differences in the relative abundance of bacteria taxa between the responders and non-responders suggested that IL-23 and IL-17 inhibitors may functionally interact with gut microbiota to reduce cutaneous inflammation. Moreover, we demonstrated the association between the treatment response and gut microbial function, which might serve as potential biomarkers in the treatment response.
## 4.1. Study Design and Patients
This prospective study enrolled forty-eight patients with psoriasis, including 30 cases treated with an IL-23 inhibitor (guselkumab) and 18 cases with IL-17 inhibitors (ixekizumab or secukinumab) in the Chang Gung Memorial Hospital (Taoyuan, Taiwan) from September 2020 to March 2022. None of the included cases had taken systemic antibiotics, systemic immunosuppressant agents, oral corticosteroids, and probiotics one month before each sample collection other than guselkumab, ixekizumab, or secukinumab. The anti-IL-23 medication group received guselkumab 100 mg at week 0, 4, and every 8 weeks thereafter. The anti-IL-17 medication group received either ixekizumab 160 mg at week 0, 80 mg at week 2, 4, 6, 8, 10, and 12, and 80 mg every 4 weeks thereafter, or secukinumab 300 mg at week 0, 1, 2, 3, and 4, and every 4 weeks thereafter. The demographics and clinical data of the patients, including their age, gender, weight, and psoriatic arthritis (PsA), were collected at baseline. Psoriasis Area and Severity Index (PASI) score and serum C-reactive protein (CRP) level were collected at week 0, 4, 12, and 24. Responders were defined as those having a PASI improvement of ≥$90\%$ after 24 weeks of treatment and non-responders as having a PASI improvement of <$90\%$. Information about food intake during the study was collected at week 0, 12, and 24 through a food-frequency questionnaire (FFQ) [55].
## 4.2. DNA Isolation and 16S rRNA Gene Sequencing
Stool specimens were collected using the Longsee Fecalpro Kit (Longsee Medical Technology Co., Guangzhou, China) at baseline and 4 weeks, 12 weeks, and 24 weeks after treatment. As described previously [56], DNA was isolated by using the QIAamp PowerFecal Pro DNA Kit for Feces (Qiagen, Germantown, MD, USA) following the manufacturer’s instructions. Around 0.25 g of the sample in the Bead Tube was added with 750 μL of PowerBead Solution and 60 μL of Solution C1, which was then heated at 65 °C for 10 min. The mixture was vortexed by using a PowerLyser Homogenizer at 1000 RPM for 10 min. After the steps of cell lysis, removal of contaminating matters, washing and eluting with DNA-free, PCR-grade water, DNA was extracted. The concentrations and qualities of the extracted DNA were measured by using Qubit 4 Fluorometer (Thermo Fisher Scientific, Waltham, MA, USA).
The variable regions 3 and 4 (V3–V4) of 16S rRNA gene were PCR (polymerase chain reaction)-amplified by using the primer set (the Illumina V3 forward 5′-CCTACGGGNGGCWGCAG-3′ and V4 reverse 5′-GACTACHVGGGTATCTAATCC-3′) [57]. The Illumina sequencing adapters ligated to the purified amplicons by a second-stage PCR using the TruSeq DNA LT Sample Preparation Kit (Illumina, San Diego, CA, USA) were performed to construct a library. Purified libraries were quantified, normalized, pooled, and applied for cluster generation and sequencing on a MiSeq instrument (Illumina).
## 4.3. Sequencing Data Processing and Species Annotation
Paired-end reads were processed by using DADA2 [58] to filter out noisy sequences, correct errors in marginal sequences, remove chimeric sequences, and eliminate singletons to infer amplicon sequence variants (ASVs). Bacterial taxonomy was assigned by applying a pre-fitted QIIME2 classifier built with the Scikit-lean package [59] based on the information collected from the SILVA database [60]. Arrangement of multiple sequences were performed by the PyNAST software v.1.2 [61] for assessment of the phylogenetic relationship of various ASVs, and a phylogenetic tree was constructed with the FastTree 2.1.0 [62].
## 4.4. Microbial Gene Function Prediction
Functional composition of metagenomes was predicted from 16S rRNA data by the Tax4Fun2 software [63]. To predict functional profile of the microbial community, the taxonomic abundance transformed from the SILVA-based 16S rRNA and normalized by the 16S rRNA copy number acquired from the NCBI annotations were applied to incorporate the precomputed functional profiles of KEGG pathways [63]. KEGG analysis was only focused on “Metabolism” pathways.
## 4.5. Statistical Analysis
Demographic and clinical characteristics were presented as n (%) for categorical variables and mean ± standard deviation (SD) or median with range for continuous variables. For estimating alpha diversity, species richness was evaluated by inverse Simpson’s index. Beta diversity was analyzed by Bray–Curtis or unweighted-UniFrac distance matrix. In order to investigate the association of treatment effect and bacteria in the fecal specimens, we further identified differentially abundant bacterial taxa among groups. Statistically significant biomarkers were analysed by the non-parametric Kruskal–Wallis test, Wilcoxon rank-sum test, and linear discriminant analysis (LDA) to identify differentially abundant taxa. Change in relative abundance after treatment from the baseline was compared between responders and non-responders by fitting a linear mixed model, measured on a continuous scale to identify longitudinal biomarkers. All statistical tests are two-tailed, and a p-value less than 0.05 was considered statistically significant.
## References
1. Griffiths C.E.M., Armstrong A.W., Gudjonsson J.E., Barker J.. **Psoriasis**. *Lancet* (2021) **397** 1301-1315. DOI: 10.1016/S0140-6736(20)32549-6
2. Eder L., Widdifield J., Rosen C.F., Cook R., Lee K.A., Alhusayen R., Paterson M.J., Cheng S.Y., Jabbari S., Campbell W.. **Trends in the prevalence and incidence of psoriasis and psoriatic arthritis in Ontario, Canada: A population-based study**. *Arthritis Care Res.* (2019) **71** 1084-1091. DOI: 10.1002/acr.23743
3. Parisi R., Iskandar I.Y.K., Kontopantelis E., Augustin M., Griffiths C.E.M., Ashcroft D.M., Global Psoriasis A.. **National, regional, and worldwide epidemiology of psoriasis: Systematic analysis and modelling study**. *BMJ* (2020) **369** m1590. DOI: 10.1136/bmj.m1590
4. Capon F.. **The genetic basis of psoriasis**. *Int. J. Mol. Sci.* (2017) **18**. DOI: 10.3390/ijms18122526
5. Horton D.B., Scott F.I., Haynes K., Putt M.E., Rose C.D., Lewis J.D., Strom B.L.. **Antibiotic exposure, infection, and the development of pediatric psoriasis: A nested case-control study**. *JAMA Dermatol.* (2016) **152** 191-199. DOI: 10.1001/jamadermatol.2015.3650
6. Chen Y.J., Ho H.J., Wu C.Y., Juan C.K., Wu C.Y.. **Infantile infection and antibiotic exposure in association with pediatric psoriasis development: A nationwide nested case-control study**. *J. Am. Acad. Dermatol.* (2021) **85** 626-635. DOI: 10.1016/j.jaad.2020.12.014
7. Alekseyenko A.V., Perez-Perez G.I., De Souza A., Strober B., Gao Z., Bihan M., Li K., Methe B.A., Blaser M.J.. **Community differentiation of the cutaneous microbiota in psoriasis**. *Microbiome* (2013) **1** 31. DOI: 10.1186/2049-2618-1-31
8. Fahlen A., Engstrand L., Baker B.S., Powles A., Fry L.. **Comparison of bacterial microbiota in skin biopsies from normal and psoriatic skin**. *Arch. Dermatol. Res.* (2012) **304** 15-22. DOI: 10.1007/s00403-011-1189-x
9. Sikora M., Stec A., Chrabaszcz M., Knot A., Waskiel-Burnat A., Rakowska A., Olszewska M., Rudnicka L.. **Gut microbiome in psoriasis: An updated review**. *Pathogens* (2020) **9**. DOI: 10.3390/pathogens9060463
10. Marietta E., Mangalam A.K., Taneja V., Murray J.A.. **Intestinal dysbiosis in, and enteral bacterial therapies for, systemic autoimmune Diseases**. *Front. Immunol.* (2020) **11** 573079. DOI: 10.3389/fimmu.2020.573079
11. Kinashi Y., Hase K.. **Partners in leaky gut syndrome: Intestinal dysbiosis and autoimmunity**. *Front. Immunol.* (2021) **12** 673708. DOI: 10.3389/fimmu.2021.673708
12. Hidalgo-Cantabrana C., Gómez J., Delgado S., Requena-López S., Queiro-Silva R., Margolles A., Coto E., Sánchez B., Coto-Segura P.. **Gut microbiota dysbiosis in a cohort of patients with psoriasis**. *Br. J. Dermatol.* (2019) **181** 1287-1295. DOI: 10.1111/bjd.17931
13. Shapiro J., Cohen N.A., Shalev V., Uzan A., Koren O., Maharshak N.. **Psoriatic patients have a distinct structural and functional fecal microbiota compared with controls**. *J. Dermatol.* (2019) **46** 595-603. DOI: 10.1111/1346-8138.14933
14. Chen Y.J., Ho H.J., Tseng C.H., Lai Z.L., Shieh J.J., Wu C.Y.. **Intestinal microbiota profiling and predicted metabolic dysregulation in psoriasis patients**. *Exp. Dermatol.* (2018) **27** 1336-1343. DOI: 10.1111/exd.13786
15. Huang L., Gao R., Yu N., Zhu Y., Ding Y., Qin H.. **Dysbiosis of gut microbiota was closely associated with psoriasis**. *Sci. China Life Sci.* (2019) **62** 807-815. DOI: 10.1007/s11427-018-9376-6
16. Zanvit P., Konkel J.E., Jiao X., Kasagi S., Zhang D., Wu R., Chia C., Ajami N.J., Smith D.P., Petrosino J.F.. **Antibiotics in neonatal life increase murine susceptibility to experimental psoriasis**. *Nat. Commun.* (2015) **6** 8424. DOI: 10.1038/ncomms9424
17. Drago F., Ciccarese G., Indemini E., Savarino V., Parodi A.. **Psoriasis and small intestine bacterial overgrowth**. *Int. J. Dermatol.* (2018) **57** 112-113. DOI: 10.1111/ijd.13797
18. Martin R., Henley J.B., Sarrazin P., Seite S.. **Skin microbiome in patients with psoriasis before and after balneotherapy at the thermal care center of La Roche-Posay**. *J. Drugs Dermatol.* (2015) **14** 1400-1405. PMID: 26659932
19. Assarsson M., Duvetorp A., Dienus O., Söderman J., Seifert O.. **Significant changes in the skin microbiome in patients with chronic plaque psoriasis after treatment with narrowband ultraviolet B**. *Acta Derm. Venereol.* (2018) **98** 428-436. DOI: 10.2340/00015555-2859
20. Liu S.H., Yu H.Y., Chang Y.C., Hui C.Y.R., Huang Y.C., Huang Y.H.. **Host characteristics and dynamics of Staphylococcus aureus colonization in patients with moderate-to-severe psoriasis before and after treatment: A prospective cohort study**. *J. Am. Acad. Dermatol.* (2019) **81** 605-607. DOI: 10.1016/j.jaad.2018.05.031
21. Yeh N.L., Hsu C.Y., Tsai T.F., Chiu H.Y.. **Gut Microbiome in psoriasis is perturbed differently during secukinumab and ustekinumab therapy and associated with response to treatment**. *Clin. Drug Investig.* (2019) **39** 1195-1203. DOI: 10.1007/s40261-019-00849-7
22. Rendon A., Schakel K.. **Psoriasis Pathogenesis and Treatment**. *Int. J. Mol. Sci.* (2019) **20**. DOI: 10.3390/ijms20061475
23. Yang C.C.. **The other side of biologics for psoriasis**. *Dermatol. Sin.* (2022) **40** 65-66. DOI: 10.4103/ds.ds_29_22
24. Nermes M., Kantele J.M., Atosuo T.J., Salminen S., Isolauri E.. **Interaction of orally administered Lactobacillus rhamnosus GG with skin and gut microbiota and humoral immunity in infants with atopic dermatitis**. *Clin. Exp. Allergy* (2011) **41** 370-377. DOI: 10.1111/j.1365-2222.2010.03657.x
25. Targan S.R., Feagan B., Vermeire S., Panaccione R., Melmed G.Y., Landers C., Li D., Russell C., Newmark R., Zhang N.. **A randomized, double-blind, placebo-controlled Phase 2 study of brodalumab in patients with moderate-to-severe Crohn’s disease**. *Am. J. Gastroenterol.* (2016) **111** 1599-1607. DOI: 10.1038/ajg.2016.298
26. Hueber W., Sands B.E., Lewitzky S., Vandemeulebroecke M., Reinisch W., Higgins P.D., Wehkamp J., Feagan B.G., Yao M.D., Karczewski M.. **Secukinumab, a human anti-IL-17A monoclonal antibody, for moderate to severe Crohn’s disease: Unexpected results of a randomised, double-blind placebo-controlled trial**. *Gut* (2012) **61** 1693-1700. DOI: 10.1136/gutjnl-2011-301668
27. Colombel J.F., Sendid B., Jouault T., Poulain D.. **Secukinumab failure in Crohn’s disease: The yeast connection?**. *Gut* (2013) **62** 800-801. DOI: 10.1136/gutjnl-2012-304154
28. Wang J., Bhatia A., Krugliak Cleveland N., Gupta N., Dalal S., Rubin D.T., Sakuraba A.. **Rapid Onset of Inflammatory Bowel Disease after receiving secukinumab infusion**. *ACG Case Rep. J.* (2018) **5** e56. DOI: 10.14309/crj.2018.56
29. Sandborn W.J., D’Haens G.R., Reinisch W., Panés J., Chan D., Gonzalez S., Weisel K., Germinaro M., Frustaci M.E., Yang Z.. **Guselkumab for the treatment of Crohn’s Disease: Induction results from the Phase 2 GALAXI-1 study**. *Gastroenterology* (2022) **162** 1650-1664.e8. DOI: 10.1053/j.gastro.2022.01.047
30. Franzin M., Stefančič K., Lucafò M., Decorti G., Stocco G.. **Microbiota and drug response in Inflammatory Bowel Disease**. *Pathogens* (2021) **10**. DOI: 10.3390/pathogens10020211
31. Seong G., Kim N., Joung J.G., Kim E.R., Chang D.K., Chun J., Hong S.N., Kim Y.H.. **Changes in the intestinal microbiota of patients with Inflammatory Bowel Disease with clinical remission during an 8-Week infliximab infusion cycle**. *Microorganisms* (2020) **8**. DOI: 10.3390/microorganisms8060874
32. Doherty Matthew K., Ding T., Koumpouras C., Telesco Shannon E., Monast C., Das A., Brodmerkel C., Schloss Patrick D.. **Fecal microbiota signatures are associated with response to Ustekinumab therapy among Crohn’s Disease patients**. *mBio* (2018) **9** e02120-17. DOI: 10.1128/mBio.02120-17
33. Ratajczak W., Rył A., Mizerski A., Walczakiewicz K., Sipak O., Laszczyńska M.. **Immunomodulatory potential of gut microbiome-derived short-chain fatty acids (SCFAs)**. *Acta Biochim. Pol.* (2019) **66** 1-12. DOI: 10.18388/abp.2018_2648
34. Yao Y., Cai X., Fei W., Ye Y., Zhao M., Zheng C.. **The role of short-chain fatty acids in immunity, inflammation and metabolism**. *Crit. Rev. Food Sci. Nutr.* (2022) **62** 1-12. DOI: 10.1080/10408398.2020.1854675
35. Sears C.L.. **A dynamic partnership: Celebrating our gut flora**. *Anaerobe* (2005) **11** 247-251. DOI: 10.1016/j.anaerobe.2005.05.001
36. Liu L., Xu M., Lan R., Hu D., Li X., Qiao L., Zhang S., Lin X., Yang J., Ren Z.. **Bacteroides vulgatus attenuates experimental mice colitis through modulating gut microbiota and immune responses**. *Front. Immunol.* (2022) **13** 1036196. DOI: 10.3389/fimmu.2022.1036196
37. Wexler A.G., Goodman A.L.. **An insider’s perspective: Bacteroides as a window into the microbiome**. *Nat. Microbiol.* (2017) **2** 17026. DOI: 10.1038/nmicrobiol.2017.26
38. Yoshida N., Emoto T., Yamashita T., Watanabe H., Hayashi T., Tabata T., Hoshi N., Hatano N., Ozawa G., Sasaki N.. **Bacteroides vulgatus and Bacteroides dorei Reduce Gut Microbial Lipopolysaccharide Production and Inhibit Atherosclerosis**. *Circulation* (2018) **138** 2486-2498. DOI: 10.1161/CIRCULATIONAHA.118.033714
39. Ahluwalia B., Magnusson M.K., Ohman L.. **Mucosal immune system of the gastrointestinal tract: Maintaining balance between the good and the bad**. *Scand. J. Gastroenterol.* (2017) **52** 1185-1193. DOI: 10.1080/00365521.2017.1349173
40. Blum H.E.. **The human microbiome**. *Adv. Med. Sci.* (2017) **62** 414-420. DOI: 10.1016/j.advms.2017.04.005
41. Egeberg A., Mallbris L., Warren R.B., Bachelez H., Gislason G.H., Hansen P.R., Skov L.. **Association between psoriasis and inflammatory bowel disease: A Danish nationwide cohort study**. *Br. J. Dermatol.* (2016) **175** 487-492. DOI: 10.1111/bjd.14528
42. Zakostelska Z., Malkova J., Klimesova K., Rossmann P., Hornova M., Novosadova I., Stehlikova Z., Kostovcik M., Hudcovic T., Stepankova R.. **Intestinal Microbiota Promotes Psoriasis-Like Skin inflammation by enhancing Th17 response**. *PLoS ONE* (2016) **11**. DOI: 10.1371/journal.pone.0159539
43. Pinget G.V., Tan J.K., Ni D., Taitz J., Daien C.I., Mielle J., Moore R.J., Stanley D., Simpson S., King N.J.C.. **Dysbiosis in imiquimod-induced psoriasis alters gut immunity and exacerbates colitis development**. *Cell Rep.* (2022) **40** 111191. DOI: 10.1016/j.celrep.2022.111191
44. Li M., Dai B., Tang Y., Lei L., Li N., Liu C., Ge T., Zhang L., Xu Y., Hu Y.. **Altered Bacterial-Fungal interkingdom networks in the guts of Ankylosing Spondylitis patients**. *mSystems* (2019) **4** e00176-18. DOI: 10.1128/mSystems.00176-18
45. Chen Y.J., Wu H., Wu S.D., Lu N., Wang Y.T., Liu H.N., Dong L., Liu T.T., Shen X.Z.. **Parasutterella, in association with irritable bowel syndrome and intestinal chronic inflammation**. *J. Gastroenterol. Hepatol.* (2018) **33** 1844-1852. DOI: 10.1111/jgh.14281
46. Liu L., Chen Y., Wu Q., Shu A., Sun J.. **Sodium butyrate attenuated diabetes-induced intestinal inflammation by modulating gut microbiota**. *Evid.-Based Complement. Alternat. Med.* (2022) **2022** 4646245. DOI: 10.1155/2022/4646245
47. Zeng C., Wen B., Hou G., Lei L., Mei Z., Jia X., Chen X., Zhu W., Li J., Kuang Y.. **Lipidomics profiling reveals the role of glycerophospholipid metabolism in psoriasis**. *Gigascience* (2017) **6** 1-11. DOI: 10.1093/gigascience/gix087
48. Zhu J., Thompson C.B.. **Metabolic regulation of cell growth and proliferation**. *Nat. Rev. Mol. Cell Biol.* (2019) **20** 436-450. DOI: 10.1038/s41580-019-0123-5
49. Luo Y., Hara T., Kawashima A., Ishido Y., Suzuki S., Ishii N., Kambara T., Suzuki K.. **Pathological role of excessive DNA as a trigger of keratinocyte proliferation in psoriasis**. *Clin. Exp. Immunol.* (2020) **202** 1-10. DOI: 10.1111/cei.13455
50. Pohla L., Ottas A., Kaldvee B., Abram K., Soomets U., Zilmer M., Reemann P., Jaks V., Kingo K.. **Hyperproliferation is the main driver of metabolomic changes in psoriasis lesional skin**. *Sci. Rep.* (2020) **10** 3081. DOI: 10.1038/s41598-020-59996-z
51. Jami M., Ghanbari M., Kneifel W., Domig K.J.. **Phylogenetic diversity and biological activity of culturable Actinobacteria isolated from freshwater fish gut microbiota**. *Microbiol. Res.* (2015) **175** 6-15. DOI: 10.1016/j.micres.2015.01.009
52. Chang C.J., Zhang J., Tsai Y.L., Chen C.B., Lu C.W., Huo Y.P., Liou H.M., Ji C., Chung W.H.. **Compositional features of distinct microbiota base on serum extracellular vesicle metagenomics analysis in moderate to severe psoriasis patients**. *Cells* (2021) **10**. DOI: 10.3390/cells10092349
53. Marcinkiewicz J., Kontny E.. **Taurine and inflammatory diseases**. *Amino Acids* (2014) **46** 7-20. DOI: 10.1007/s00726-012-1361-4
54. Hillmann B., Al-Ghalith G.A., Shields-Cutler R.R., Zhu Q., Gohl D.M., Beckman K.B., Knight R., Knights D.. **Evaluating the Information Content of Shallow Shotgun Metagenomics**. *mSystems* (2018) **3** e00069-18. DOI: 10.1128/mSystems.00069-18
55. Huang Y.C., Lee M.S., Pan W.H., Wahlqvist M.L.. **Validation of a simplified food frequency questionnaire as used in the Nutrition and Health Survey in Taiwan (NAHSIT) for the elderly**. *Asia Pac. J. Clin. Nutr.* (2011) **20** 134-140. PMID: 21393121
56. Lee C.C., Feng Y., Yeh Y.M., Lien R., Chen C.L., Zhou Y.L., Chiu C.H.. **Gut dysbiosis, bacterial colonization and translocation, and neonatal sepsis in very-low-birth-weight preterm infants**. *Front. Microbiol.* (2021) **12** 746111. DOI: 10.3389/fmicb.2021.746111
57. Caporaso J.G., Lauber C.L., Walters W.A., Berg-Lyons D., Lozupone C.A., Turnbaugh P.J., Fierer N., Knight R.. **Global patterns of 16S rRNA diversity at a depth of millions of sequences per sample**. *Proc. Natl. Acad. Sci. USA* (2011) **108** 4516-4522. DOI: 10.1073/pnas.1000080107
58. Callahan B.J., McMurdie P.J., Rosen M.J., Han A.W., Johnson A.J., Holmes S.P.. **DADA2: High-resolution sample inference from Illumina amplicon data**. *Nat. Methods* (2016) **13** 581-583. DOI: 10.1038/nmeth.3869
59. Pedregosa F., Varoquaux G., Gramfort A., Michel V., Thirion B., Grisel O., Blondel M., Prettenhofer P., Weiss R., Dubourg V.. **Scikit-learn: Machine learning in python**. *J. Mach. Learn. Res.* (2011) **12** 2825-2830
60. Quast C., Pruesse E., Yilmaz P., Gerken J., Schweer T., Yarza P., Peplies J., Glockner F.O.. **The SILVA ribosomal RNA gene database project: Improved data processing and web-based tools**. *Nucleic Acids Res.* (2013) **41** D590-D596. DOI: 10.1093/nar/gks1219
61. Caporaso J.G., Bittinger K., Bushman F.D., DeSantis T.Z., Andersen G.L., Knight R.. **PyNAST: A flexible tool for aligning sequences to a template alignment**. *Bioinformatics* (2010) **26** 266-267. DOI: 10.1093/bioinformatics/btp636
62. Price M.N., Dehal P.S., Arkin A.P.. **FastTree: Computing large minimum evolution trees with profiles instead of a distance matrix**. *Mol. Biol. Evol.* (2009) **26** 1641-1650. DOI: 10.1093/molbev/msp077
63. Aßhauer K.P., Wemheuer B., Daniel R., Meinicke P.. **Tax4Fun: Predicting functional profiles from metagenomic 16S rRNA data**. *Bioinformatics* (2015) **31** 2882-2884. DOI: 10.1093/bioinformatics/btv287
|
---
title: Exploiting Vitamin D Receptor and Its Ligands to Target Squamous Cell Carcinomas
of the Head and Neck
authors:
- Laura Koll
- Désirée Gül
- Manal I. Elnouaem
- Hanaa Raslan
- Omneya R. Ramadan
- Shirley K. Knauer
- Sebastian Strieth
- Jan Hagemann
- Roland H. Stauber
- Aya Khamis
journal: International Journal of Molecular Sciences
year: 2023
pmcid: PMC10002563
doi: 10.3390/ijms24054675
license: CC BY 4.0
---
# Exploiting Vitamin D Receptor and Its Ligands to Target Squamous Cell Carcinomas of the Head and Neck
## Abstract
Vitamin D (VitD) and its receptor (VDR) have been intensively investigated in many cancers. As knowledge for head and neck cancer (HNC) is limited, we investigated the (pre)clinical and therapeutic relevance of the VDR/VitD-axis. We found that VDR was differentially expressed in HNC tumors, correlating to the patients’ clinical parameters. Poorly differentiated tumors showed high VDR and Ki67 expression, whereas the VDR and Ki67 levels decreased from moderate to well-differentiated tumors. The VitD serum levels were lowest in patients with poorly differentiated cancers (4.1 ± 0.5 ng/mL), increasing from moderate (7.3 ± 4.3 ng/mL) to well-differentiated (13.2 ± 3.4 ng/mL) tumors. Notably, females showed higher VitD insufficiency compared to males, correlating with poor differentiation of the tumor. To mechanistically uncover VDR/VitD’s pathophysiological relevance, we demonstrated that VitD induced VDR nuclear-translocation (VitD < 100 nM) in HNC cells. RNA sequencing and heat map analysis showed that various nuclear receptors were differentially expressed in cisplatin-resistant versus sensitive HNC cells including VDR and the VDR interaction partner retinoic acid receptor (RXR). However, RXR expression was not significantly correlated with the clinical parameters, and cotreatment with its ligand, retinoic acid, did not enhance the killing by cisplatin. Moreover, the Chou–Talalay algorithm uncovered that VitD/cisplatin combinations synergistically killed tumor cells (VitD < 100 nM) and also inhibited the PI3K/Akt/mTOR pathway. Importantly, these findings were confirmed in 3D-tumor-spheroid models mimicking the patients’ tumor microarchitecture. Here, VitD already affected the 3D-tumor-spheroid formation, which was not seen in the 2D-cultures. We conclude that novel VDR/VitD-targeted drug combinations and nuclear receptors should also be intensely explored for HNC. Gender-specific VDR/VitD-effects may be correlated to socioeconomic differences and need to be considered during VitD (supplementation)-therapies.
## 1. Introduction
In the last three decades, there have been tremendous attempts to undercover the role of vitamin D (VitD) in the prevention, prognosis, and treatment of cancer. Unfortunately, the results have been contradictory, and until now, no general recommendations or standard treatment options considering VitD for cancer patients exist [1,2]. However, the majority of the observational studies supported a benefit of higher vitamin D intake concerning the reduction in cancer incidence (e.g., colon and breast cancer) [3,4]. Other studies showed a correlation between high serum VitD levels and lower cancer risk [5,6]. Nevertheless, for many entities, the clinical relevance of VitD as well as its molecular mechanisms of action requires further investigation.
Head and neck squamous cell carcinomas (HNC) are among the top ten most common cancers worldwide, frequently exhibiting limited treatment response [7,8,9,10]. Here, reasons for unsatisfactory treatment success and the long-term survival of HNC patients can be found in the development of resistance toward established treatments as well as the lack of novel therapeutic targets. A promising approach to potentially increase the success of established treatment options, which are surgery, chemo-, radiotherapy as well as targeted (Cetuximab) therapy, is the use of combinational treatments. Here, the use of functional foods such as VitD offers an alternative, cost-effective cancer care regimen harboring the potential to improve treatment success. Functional foods and food components affect the body, reaching beyond a basic nutritional effect. Among the group of functional foods, VitD is one of the most important members for which anti-tumoral effects have already been suggested [11,12]. Interestingly, little is known about the relevance of VitD for HNC patients in general and the potential clinical benefit of combinational VitD therapies in particular. Previous studies and meta-analyses have already demonstrated the need to determine and evaluate the VitD influence on cancer pathogenesis and patient prognosis [13,14].
VitD, which is rather a (steroid) hormone than a vitamin, has a variety of functions in health and disease [15]. Approximately $90\%$ of the VitD requirement is produced in the skin in response to ultraviolet-B (UV-B) light from sun exposure. The biologically active form of VitD, 1α,25-dihydroxyVitD3 (1,25(OH)2D3), also called calcitriol, is produced by enzymatic conversions of its precursor calcidiol (25-hydroxy VitD, 25(OH)D3) via hydroxylation in the liver and kidneys [16]. 25-Hydroxy VitD is the most single reliable marker of VitD concentration in the body due to its relatively long half-life time (three weeks) compared to the active form of 1,25-OH2D (approximately 4 h) [15,16,17]. It is also an indication of the availability of the substrate for tissue production and the auto/paracrine action of 1,25-OH2D. On the other hand, although 1,25-OH2D is the active form, it is regulated by several enzymatic and physiological inputs [16]. The fact that 1,25-OH2D concentrations in the blood may not decrease or decrease at a late stage even in presence of VitD deficiency make 25-(OH)D3 a better marker for the assessment of VitD supply [18].
Notably, VitD deficiency is widespread and can cause various diseases such as rickets in children and osteoporosis [19]. Therefore, VitD food fortification is practiced in some countries. Moreover, VitD deficiency has been correlated to multiple systemic and physiological conditions such as insulin resistance and diabetes, autoimmune disease, cardiovascular disease, and all-cause mortality [16,20]. Importantly, VitD deficiency seems to be (in)directly correlated with the occurrence of cancer [21,22]. Furthermore, previous studies have suggested that sufficient levels of VitD can reduce the risk of many cancer types such as colon and breast cancer [3,4]. Meta-analyses show that patients with serum levels of the VitD pre-cursor calcidiol (25-hydroxyVitD) ranging from 20–40 ng/mL show a significantly reduced risk of about $35\%$ for breast and colorectal cancer [23,24]. Furthermore, there is evidence for some entities such as breast, colorectal, lung, bladder cancer, and prostate cancer that higher serum levels of calcidiol at the time of diagnosis are correlated with improved survival rates [25,26,27,28,29].
VitD executes its biological functions via binding to the VitD receptor (VDR), a member of the nuclear receptor family [30,31]. Nuclear receptors (NRs) are key regulators of health and disease including cancer, and thus represent important targets for anti-cancer therapies [32,33,34]. NRs are transcription factors involved in a wide range and extremely complex spectrum of physiological and pathophysiological processes, hence they are interesting therapeutic targets [32,35]. However, despite intensive research, the detailed mechanisms of homo-/heterodimerized nuclear receptors including VDR are still not resolved (for more details, see also following reviews [3,4,32,34]. Upon the binding of its ligand VitD and nuclear translocation, VDR is able to form heterodimers with the retinoid X receptor (RXR). By binding to specific VDR-responsive elements on the DNA, the nuclear receptors are able to activate various transcriptional programs [3,4,30,31,32,34,36]. Importantly, for VDR, anti-tumoral effects have been already suggested [3,4,32,34,37,38].
Hence in this study, we investigated the (pre)clinical and potential therapeutic relevance of the VDR/VitD-axis to assess the association between the VitD level and VDR expression for HNC. Aside from analyzing the HNC dataset of The Cancer Genome Atlas (TCGA), a case-control study was analyzed. To mechanistically uncover VDR/VitD’s pathophysiological relevance, we further combined the evaluation of clinical data with comprehensive dry and wet lab systematic studies of innovative HNC cell models. Besides the use of the 2D tumor cell model, there is increasing evidence that advanced 3D tumor spheroids react differently compared to conventional 2D cultures when exposed to drugs, radiation, or signaling ligands [39,40,41,42]. Hence, we established a 3D cell culture model aiming to approach the tumor situation in vivo. In comparison to the 2D culture systems, 3D spheroids exhibit a number of advantages, for example, they mimic a more realistic 3D architecture of a tumor including the supply of nutrients, oxygen, and anti-cancer drug. Another advantage is the development of polarity in the spheroid culture due to neighboring cell-to-cell contacts [39,40]. Collectively, cells in 3D tumor spheroids seem to preserve key morphological and signaling patterns closely associated with tumor development and drug resistance in animal models and patients [39,40,41,42].
## 2.1. VDR Expression and VitD Levels Correlate with HNC Patients’ Clinical Parameters
As knowledge of the VDR/VitD-axis for HNC is limited, we first investigated the VDR expression and VitD serum levels in a cohort of newly diagnosed HNC patients ($$n = 40$$) compared to the healthy individuals ($$n = 40$$) (details see Table 1, Supplementary Tables S3 and S4). The most common site of occurrence was the tongue ($60\%$) and the least was the lip (Figure 1a). Notably, regarding gender, there was a significant difference in the male-to-female ratio (Figure 1b, Supplementary Tables S3 and S5) which is often observed in the Middle East and North Africa (MENA) region [43,44,45,46]. Histopathologically, the most common differentiation subtype was moderately differentiated HNC ($60\%$, $$n = 24$$; Figure 1c, Supplementary Table S5).
In order to correlate the VitD serum levels with VDR expression in the tumor tissues, peripheral blood samples were taken from the patients before or during surgery. Total serum VitD (25-hydroxyVitD3) was quantified by using fully validated, modified high-performance liquid chromatography (HPLC) [47]. The VitD serum levels were lowest in patients with poorly differentiated cancers (4.1 ± 0.5 ng/mL), increasing from moderate (7.3 ± 4.3 ng/mL) to well-differentiated (13.2 ± 3.4 ng/mL) (Figure 1d, Table 2). The mean serum VitD level was 7.4 ± 4.5 ng/mL in cancer patients in comparison to 28.7 ± 4.6 ng/mL in healthy individuals (Figure 1e, Table 2). Notably, females showed higher VitD insufficiency compared to males, correlating with poor tumor differentiation (Table 1).
Additionally, the VDR protein expression was analyzed by immunofluorescence and immunohistochemical staining in tumor biopsies classified as poorly, moderately, and well-differentiated (Figure 1g–i). Here, a significant inversely proportional correlation between the VitD levels and VDR expression was found (Figure 1f). As shown in Figure 1g–i, all studied cases showed immunofluorescence reactivity to the VDR antibody with varying intensities. Moreover, we found that the VDR levels correlated with the patients’ clinical and pathobiological tumor parameters. Particularly, poorly differentiated tumors showed high VDR and Ki67 expression (Figure 1g), whereas VDR and Ki67 levels decreased from moderate to well-differentiated tumors (Figure 1h,i).
## 2.2. Clinical Relevance of VitD Receptor (VDR) and Retinoid X Receptor Alpha (RXRα) Expression in HNC Patients
To independently confirm the relevance of VDR expression in the HNC patients, we bioinformatically analyzed the PANCAN dataset acquired from The Cancer Genome Atlas (TCGA), encompassing more than 12,000 samples of cancer patients of various entities and clinical backgrounds. Moreover, upon binding of its ligand VitD and nuclear translocation, VDR is also able to form heterodimers with the retinoid X receptor (RXR), thereby activating various cancer-relevant transcriptional programs (Supplementary Figure S1) [23,25,32,34,48,49,50]. As VDR/RXR expression has not been studied for HNC, we also studied the expression of RXR in the datasets.
We found VDR overexpressed in the primary tumors. Comparing the different entities, the highest expression of VDR was found in rectal and colon adenocarcinoma and kidney cancer, directly followed by HNC (Supplementary Figure S2), supporting our conclusions obtained from the analyses of our cohort (see also Figure 1).
Thus, in the second step, we focused on the analysis of the TCGA HNC cohort ($$n = 604$$) showing upregulation of VDR in tumor versus non-tumor tissues (Figure 2a, $$n = 564$$, $$p \leq 0.0059$$ **). Interestingly, RXRα expression showed no correlation with the disease markers (Figure 2b, $$n = 520$$ $$p \leq 0.4931$$). VDR expression highly correlated with the histological differentiation of the tumor, in contrast to the RXRα levels (Figure 2c,d, $$n = 540$$, $$p \leq 0.0002$$ ***/$$p \leq 0.0056$$ **). Since the HPV status affects the therapy outcome and prognosis of HNC patients, we analyzed HPV-negative versus HPV-positive patients. VDR expression was significantly increased in HPV-negative HNC patients (Figure 2e, $$n = 114$$, $p \leq 0.0001$ ****). Again, changes in the RXRα levels were less significant (Figure 2f; $$n = 114$$; $p \leq 0.0108$). Moreover, high VDR expression correlated with perineural invasion (Figure 2g, $$n = 393$$, $$p \leq 0.0006$$ ***) in contrast to the RXRα levels (Figure 2h, $$n = 393$$, $$p \leq 0.4154$$), underlining again the relevance of VDR but not of RXRα as a biomarker and/or therapeutic target for HNC.
## 2.3. Nuclear Receptor Profiling and Translocation Kinetics in HNC Cells
It is accepted by the field that the superfamily of nuclear receptors are key regulators in many pathologies including cancer [32,33,34]. Thus, we used ‘omics’ approaches to profile nuclear receptor expression and the potential pathobiological relevance in HNC tumor cell models. As HNC treatment is often complicated by recurrence due to resistance to cisplatin-based treatments, we analyzed the chemoresistant HNC cells. The cisplatin-resistant cell line, Picares, was established by selecting HNC Pica cells with sub-toxic concentrations of cisplatin (3–5 µM) for six months. Hence, Picawt and Picares allow for the comparison of cisplatin-sensitive and resistant HNC cells. Here, next-generation RNA sequencing transcriptomics was used to analyze the expression of various nuclear receptors (Figure 3a, Supplementary Table S6). As illustrated in the heat map analysis (Figure 3a; green: downregulated, red: upregulated), VDR and several other receptors such as Nuclear Receptor Subfamily 4 Group A Member 2 (NR4A2) or RXRα were differentially expressed in therapy-resistant (res) versus sensitive (wt) Pica cells. These data also suggest investigating the pathobiological relevance of other nuclear receptors for HNC in comprehensive follow-up studies.
When studying the impact of nuclear receptors, it is also the key to control if the respective receptor is expressed and indeed capable of cytoplasmic to nuclear trafficking upon ligand binding in the relevant cell model. Nuclear translocation is required to activate ligand-dependent transcriptional programs [3,4,30,31,32,34,36]. The activation of VDR by ligand binding typically involves VDR-RXRα dimerization and the initiation of downstream signaling (Supplementary Figure S1). The immunofluorescence staining of endogenous VDR and RXR receptors demonstrated that VitD triggered nuclear accumulation of the receptors, in contrast to retinoic acid (RA) treatment alone (Figure 3b). When referring to VitD, the active form calcitriol (1,25(OH)2D3) was used if not indicated otherwise. Hence, although both receptors are capable of cytoplasmic to nuclear trafficking, VitD and VDR seem more relevant in HNC cells. We also confirmed and quantified the VDR expression in different HNC cell lines (Figure 3c,d).
To further study the kinetics of VDR translocation in real-time, we established HNC cell lines stably expressing VDR fused to GFP. Therefore, the VDR reading frame was cloned from primary HNC tumor cells and stably expressed VDR-GFP in the HNCUM-02T or HNC FaDu cells (Figure 3e). An important question regarding VDR’s nuclear translocation is the determination of the most effective ligand dose and the time kinetics of the process. Using the high-content screening microscopy platform, Array Scan VTI, we automatically quantified VDR translocation. Here, cells were treated with different clinically relevant doses of VitD (0–100 nM) for 30 min (Figure 3f,g). Fluorescence microscopy showed dose-dependent VDR translocation into the nucleus by VitD, which was most effective at a VitD concentration of 100 nM. Importantly, RA alone did not trigger the nuclear translocation of VDR (Supplementary Figure S3).
## 2.4. VitD/VDR Targeting Synergistically Improves Cisplatin-Mediated Killing of HNC Tumor Cells
Chemoresistance is not only one of the main causes influencing cancer progression, but it is also strongly correlated to the cancer mortality rates. Hence, developing strategies for enhancing chemo sensitivity, potentially also by functional food supplementation with VitD, is expected to benefit patients. Indeed, such efforts have been made to correct VitD deficiency in cancer patients [11,12,51,52]. However, the success of VitD/VDR targeting therapies requires mechanistic knowledge and experimental investigation in vitro.
To examine the effect of combination therapy on HNC, we thus measured the cell viability after the VitD/cisplatin treatments. To also mimic the pathophysiological conditions of high and low VitD serum levels, cells were seeded in the presence or absence of 100 nM VitD, which we found to trigger efficient VDR nuclear translocation, and thus biological activation (see Figure 3f,g). After 24 h of VitD pre-treatment, cells were additionally treated with physiological concentrations of VitD (100 nM), 15–20 µM cisplatin, or a combination (Figure 4). As expected, VitD alone did not affect cell viability. However, the combination treatments significantly enhanced tumor cell death compared to cisplatin alone in the three HNC cell lines tested (Figure 4a; Supplementary Figure S4). To objectively uncover a potential synergistic effect of VitD/cisplatin combinational treatments, we performed the Chou–Talalay method. The calculation of the combination index (CI) using the Chou–Talalay algorithm allowed us to uncover additive (CI = 1), synergistic (CI < 1), or antagonistic effects (CI > 1) of the drug combinations [53]. As shown in Figure 4b–d, all calculated indices were less than 1, revealing a synergistic effect on tumor cell killing for the VitD/cisplatin combinations in the tested HNCUM 02T, FaDu, and Pica cell lines.
## 2.5. Impact of VitD/VDR Targeting on HNC 3D Tumor Spheroids
Conventional 2D tumor cell models are well-established tools to assess various aspects of tumor pathobiology. However, there is increasing evidence that advanced 3D tumor spheroids react differently compared to conventional 2D cultures when exposed to drugs, radiation, or signaling ligands [39,40]. The architecture of spheroids leads to a gradient of nutrition and oxygen from the outer surface to the core, and drug delivery to parts of the 3D cell cluster also seems to differ. Additionally, cells in 3D tumor spheroids seem to preserve certain distinct signaling patterns that are closely associated with drug resistance in animal models and patients.
In order to closely approach the tumor situation in vivo, we next established HNC 3D tumor spheroids to investigate the effects of VitD/VDR targeting in an experimental setting, more closely mimicking the patients’ tumor microenvironments. Here, cells were cultured in ultra-low adhesion cell culture vials that promoted the formation of 3D spheroid-shaped tumor cell clusters. As summarized in Figure 5a, various pathobiological relevant properties of the established 3D spheroids were subsequently analyzed by fully automated high-content microscopy, allowing for an objective assessment of the tumor spheroids’ growth, morphology, and vitality.
First, we found that the synergistic killing effect of the VitD/cisplatin combinations observed in the 2D cultures was also relevant for the 3D spheroids (Figure 5b). Cotreatment significantly reduced the mean objective area and viability (Figure 5b,c). Interestingly, although VitD alone did not affect the vitality of the 2D cultures, it already affected the 3D spheroid formation and induced morphological and architectural changes. As shown in Figure 5b–e and Supplementary Figure S5, automated high-content microscopy revealed that spheroid formation and growth were significantly impaired, suggesting that the expression of epithelial surface markers may be reduced. Notably, the effect was more prominent for the cisplatin-resistant cell line Picares (Figure 5d,e, Supplementary Figure S5), although the molecular details are not known. In conclusion, these data uncover a novel effect of VitD and also demonstrate that 3D tumor spheroids are a valuable experimental tool to uncover aspects of tumor pathobiology potentially occluded in conventional 2D tumor cell models.
## 2.6. VitD Enhances the Chemotherapeutic Effect via mTOR-PI3K/Akt Downregulation in HNC
To further investigate how VitD or VitD/cisplatin combinations inhibit the proliferation and clonogenic survival of HNC cells, we examined the cancer-relevant signaling pathways. First, bioinformatics analyses employing the Ingenuity Pathway Analysis software (Version v01-04) revealed multiple molecular mechanisms involved in cancer pathogenesis and treatment resistance (Supplementary Figure S6). Subsequently, we focused on potential VDR-RXR activation pathways (Supplementary Figure S1) and further explored the literature [54,55,56,57]. As summarized in Figure 6a,b, VitD has been suggested to regulate several pathways including the cancer-relevant mTOR/PI3K-Akt pathways. Here, key regulatory proteins are (in)directly affected by VitD overlap such as the Akt kinase (Figure 6a,b). Under ‘healthy’ conditions, the mTOR/PI3K-Akt pathways are important players in development, cellular homeostasis, and health control.
However, in cancer, abnormally activated mTOR/PI3K-Akt signaling stimulates tumor cells to grow, metastasize, and become resistant to treatment [54,55,56,57,58]. Notably, when we examined the impact of VitD and VitD/cisplatin treatment combinations in HNC cell models, we found that expression of the active, phosphorylated forms of mTOR and Akt (i.e., of pmTOR and pAkt) was particularly decreased upon VitD/cisplatin cotreatment (Figure 6c,d). No significant reduction in pmTOR and pAkt was detected upon cisplatin treatment alone. These findings not only provide a potential molecular explanation for the enhanced cisplatin-killing effect on the cancer cells by VitD, but also suggest the further experimental exploitation of additional cotreatment combinations such as using mTOR and Akt inhibitors in combination with VitD.
## 3. Discussion
The VDR/VitD-axis has been intensively investigated for more than a decade for the prevention and/or treatment of many cancers. Such (pre)clinical studies range from VitD food supplementation and cancer-prevention trials to different combination therapies [3,4,32,34,37]. Indeed, various anti-tumoral effects have been suggested for this member of the nuclear receptor superfamily, and VitD deficiency is often observed in cancer patients [21,22,59,60]. However, the underlying mechanisms of the VitD/VDR-mediated effects are not understood in detail and sometimes conflicting reports underline that its role, especially in specific cancer types, remains to be dissected [3,4,34,37].
Our clinical and experimental data support a significant role of the VDR/VitD-axis in the prognosis and clinical outcome of HNC patients. First, by analyzing our cohort of $$n = 40$$ HNC patients compared to healthy controls ($$n = 40$$), we demonstrated that both the VitD serum levels and VDR expression correlate with clinical parameters such as histopathological tumor classification. Although we could not provide specific data on patient prognoses such as survival curves, in general, the HNC patients’ overall survival correlates with histopathological differentiation of the tumor (see Supplementary Figure S7). Our finding that patients with poorly differentiated tumors and thus poor prognosis exhibited the lowest VitD levels is in line with previous studies of other entities. For example, Yao et al. found that low serum 25OHD levels at diagnosis were associated with poorer survival and worse prognosis in breast cancer patients [61]. Additionally, there have been studies observing an inverse relationship between cancer mortality and serum VitD level [59,62], suggesting that VitD supplementation therapy was most effective in patients with VitD deficiency at diagnosis [62]. However, in contrast to other clinical studies, we here paid attention to recruiting an age-sex-matched control group of non-cancer patients, allowing us to draw conclusions about a potential (gender-specific) correlation between the serum VitD level and HNC. This study’s confinement of cases to 40 patients due to the complexity of the subject matter could be seen as a potential limitation. It also has to be mentioned that the study cohort includes tumors of different sites such as the tongue and lip, which can differ in their prognosis. Nevertheless, the cohort is suitable to represent the commonly observed distribution of subsites and histopathological differentiation.
Of note, our study cohort was recruited in Egypt, exhibiting socio-economical characteristics, which we feel worth discussing. First, the study cohort differed in its gender composition from typical Northern European and American study groups because it consisted mainly of women (male–female ratio 1:4). This is often observed in the Middle East and North Africa (MENA) region [43,44,45,46], which among other factors such as increased smoking [63,64] could be explained by differences in VitD supply. A normal VitD supply is defined as when the 25(OH)D serum concentrations ranged between 30 and 50 ng/mL, whereas levels <20 ng/mL were classified as VitD deficiency [65,66]. The mean level of serum VitD (25(OH)D) in the healthy population differs depending on the geographical residence, whereas mean VitD levels in adults in North America, Asia Pacific, and Europe range between 20.4 and 28.9 ng/mL, and thus could be classified as insufficient, but not yet deficient. Interestingly, in the Middle East and North Africa region (MENA) the mean VitD levels seem to be significantly lower with 13.6–15.2 ng/mL (applies to the same age group, does not take differences in sunshine duration into account) [67]. Different reasons may explain lower VitD levels in the MENA region such as increased air pollution, reducing the amount of UVB rays available for VitD production in the skin [17,68,69]. Another explanation could be a physiological de-toxification mechanism of VitD, which is activated after longtime sunlight exposure to prevent the toxic effects of very high VitD levels in the human body [15,16]. While the analyzed healthy patients of our study cohort lay above the statistic MENA value with a mean VitD concentration of 28.7 ng/mL, the HNC patients exhibited very low VitD levels (mean 7.4 ng/mL), classified as severe VitD deficiency (<12 ng/mL). Here, especially the female patients exhibited very low VitD levels (5.3 ng/mL), which is supported by other studies describing the female gender as a risk factor for hypovitaminosis [67]. Aside from the general reasons for VitD deficiency in the MENA region described above, additional circumstances such as veiling and/or reserved clothing style, lower socio-economic standard, and predominant indoor activity may contribute to VitD deficiency in women [67,70,71]. These factors come along with the lack of awareness about the importance of VitD to the human body [67]. Assuming a significant role of VitD in the pathogenesis of HNC, this could partly explain the increased incidence of HNC in females. Such a correlation has also been suggested for colorectal cancer. Here, the VitD levels were inversely proportional to the risk of cancer in women, but not statistically significant in men [24]. Furthermore, it has been proposed that VitD supplementation could be protective against breast cancer in menopausal women, underlining its effect on tumorigenesis [72]. Again, for the gender-specific conclusions also drawn from our study, the restricted sample size of $$n = 34$$ females should be considered, suggesting further larger studies focusing on the gender-specific relevance of VitD in HNC.
Since VitD executes its biological functions via nuclear receptor binding, we analyzed the clinical relevance as well as expression and ligand-dependent activation of VDR and its heterodimerization partner RXRα. Here, we showed that VDR, but not RXRα, was significantly overexpressed in the primary HNC patients, which also correlated with clinically relevant disease markers such as HPV status, perineural invasion, and histopathological differentiation. However, there are some conflicting studies about the clinical relevance of VDR overexpression for tumorigenesis [73,74,75]. For example, Choi et al. correlated the VDR overexpression with negative prognosis in thyroid cancer [73], supporting our data showing that VDR overexpression in poorly differentiated, highly proliferative tumor tissue. Other studies have correlated the high VDR expression with an improved prognosis of patients [74,75,76]. RXRα expression and its clinical relevance in HNC have also been controversially discussed. RXR agonists such as bexarotene can benefit HPV-negative HNC patients [77]. Bexarotene combination therapy was also effective in a preclinical trial [78]. For breast cancer, there are studies demonstrating a concurrent overexpression of VDR and RXRα [79], partially also describing a worse disease-free survival when RXRα is overexpressed [80,81,82]. Hence, RXR might be worth investigating in future experimental and clinical VitD/VDR studies in general.
Chemoresistance is a major cause of cancer progression and impacts the mortality of cancer patients, particularly for HNC [9,10,39,83]. Hence, developing strategies for enhancing chemosensitivity, potentially also by food supplementation with VitD, is needed and may benefit patients. Indeed, such efforts have been made to correct VitD deficiency in cancer patients [11,12,51,52]. Through our comprehensive in vitro studies applying established 2D as well as 3D spheroid HNC cell models, we could show that VitD treatment improves chemotherapeutic killing, especially of therapy-resistant HNC tumor cells, suggesting VitD supplementation during the primary (radio)chemotherapy of HNC patients. Of course, the serum VitD levels of respective patients should be carefully monitored during therapy, and other clinically relevant factors also have to be considered. Previous observational studies and clinical trials have partially reported improved survival of cancer patients after VitD supplementation, but the findings are not conclusive yet, and further studies combining clinical with wet lab investigation are needed [84].
Our nuclear receptor profiling by next-generation RNA sequencing transcriptomics provides the first data suggesting that other nuclear receptors may also be relevant for cisplatin-chemoresistance in HNC. Furthermore, VDR or RXRα investigated here, differentially expressed receptors such as Nuclear Receptor Subfamily 4 Group A Member 2 (NR4A2) seem to be relevant for various aspects of HNC pathobiology including HPV status and mTOR/Akt signaling, underlining the value of our datasets [85,86,87]. Due to the complexity of this area, we did not explore other nuclear receptors in this study, which might be considered as a potential limitation. Hence, we refer the reader to the literature regarding the specific receptor of interest. We conclude that the data provided here may stimulate the field to further explore the relevance of the nuclear receptor superfamily for therapy resistance in HNC.
In cancer, abnormally (de)activated signaling pathways such as mTOR/PI3K-Akt and NFκB signaling stimulate tumor cells to proliferate aggressively, metastasize, and become even more resilient to therapy [54,55,56,57,58]. Here, we found that the VitD/VDR-axis enhances the chemotherapeutic effect via mTOR-PI3K/Akt downregulation in HNC. The potential relevance of the Akt- and mTOR pathways in VitD/VDR signaling is supported by reports in other tumor types [56,88]. It has to be mentioned that VitD executes its biological functions by various cellular pathways, and thus is likely that additional proapoptotic pathways may contribute to cancer-associated VitD effects. Of note, bioinformatic modeling and predictions, as performed in our study, will aid in hypothesis building, but detailed investigations are needed to confirm the candidates’ relevance. Our findings not only suggest an additional molecular mechanism for the observed beneficial effects of VitD supplementation, but also suggest further exploitation of additional cotreatment combinations such as using mTOR and Akt inhibitors (e.g., ICSN3250, LY3023414, AZD8055, or rapamycin) [58,89]. However, these preliminary results give the first molecular evidence for further co-treatment options, and detailed analyses have to be performed in future (pre-)clinical studies.
Collectively, we can conclude that novel VDR/VitD-aided drug combinations should be intensely investigated in (pre)clinical studies. Here, gender-specific VDR/VitD-effects impacted by country-specific socioeconomic differences may need additional attention. Moreover, nuclear receptors should be further explored not only for breast or colon cancer, but also for HNC.
## 4.1. Chemicals and Reagents
Unless stated otherwise, chemicals were purchased from Sigma Aldrich/Merck (Darmstadt/Munich, Germany) or MSC (MSC UG and CoKG, Mainz, Germany). Cell culture media and reagents were sourced from Gibco/Thermo Fisher Scientific (Dreieich, Germany). Disposables were purchased from Greiner Bio-One (Frickenhausen, Germany). Ab used: α-VDR (sc-13133; Santa Cruz Biotechnology, Heidelberg, Germany), α-VDR (ab3508, Abcam, Erlangen, Germany), α-RXRα (5388, Cell Signaling, Leiden, The Netherlands), α-phospho-mTOR (5536, Cell Signaling, Leiden, The Netherlands), α-phospho-Akt (3787, Cell Signaling, Leiden, The Netherlands), and α-actin (A2066; Sigma Aldrich, Munich, Germany). Appropriate HRP-, Cy3-, or FITC-conjugated secondary antibodies (Sigma Aldrich, Munich, Germany; Santa Cruz Biotechnology, Heidelberg, Germany) were used (Supplementary Table S1). Reagents such as cisplatin were from Sigma (Sigma Aldrich, Munich, Germany) or MSC (MSC UG & CoKG, Mainz, Germany). 1α,25-Dihydroxy VitD3 (Calcitriol) was purchased from Sigma and Santa-Cruz (D1530, Sigma Aldrich, Munich, Germany and CAS 32222-06-3, Santa Cruz Biotechnology, Heidelberg, Germany). Ki67 (IR626, DAKO Agilent, Santa Clara, CA, USA), and DakoEnVision Flex (Linker) (DM824, DAKO Agilent, Santa Clara, CA, USA) were also used.
## 4.2. Study Population
The investigation was conducted following the ethical standards according to the Declaration of Helsinki of 1975 and according to the local, national, and international guidelines. Tissue samples were obtained from patients undergoing surgical resection of HNC at the Department of Oral and Maxillofacial Surgery at the Faculty of Dentistry of Alexandria University from December 2017 to November 2018. In that period, the cases were consecutively enrolled in the study. The study protocol was approved by the local ethics committee (#0008839) after obtaining the patient’s informed consent to participate in the study and was processed anonymously. Patients undergoing simultaneous chemo- or radio-treatment before or during the surgery were excluded from the study. All cases were diagnosed histopathologically as HNC and staged according to the TNM classification of malignant tumors recommended by the ‘Union International Contre le Cancer UICC (8th edition). All experiments were performed in accordance with the relevant laws and the Alexandria University Guidelines and approved by the institutional ethics committee at the Faculty of Dentistry, Alexandria University. In this study, tumor specimens and corresponding non-malignant tissue were analyzed, different tumor sizes (T1–T4), lymph node status (N0-2), and grading G1–G3. Upon resection, samples were immediately fixed in formaldehyde. Histological analyses were performed to ensure that each specimen contained >$70\%$ tumor tissue and <$10\%$ necrotic debris. Samples not meeting these criteria were rejected. Specimens were handled as usual (i.e., paraffin-embedded, sectioned, and H&E staining). The H&E stain was implemented by staining the specimens with Harris’ hematoxylin as described [83,90]. The interpretation was performed by oral pathologists at Alexandria University. Peripheral blood samples were taken from patients before or during surgery. The total serum 25-hydroxyVitD concentration (sum of D3 & D2 forms) is regarded as the best single marker of VitD status in the human body. Total serum VitD (25-hydroxyVitD3) was quantified by using fully validated, modified high-performance liquid chromatography (HPLC) [47].
## 4.3. Clinical Data Analysis
HNC tissue samples were included from The Cancer Genome Atlas (TCGA) Research Network (http://cancergenome.nih.gov/, accessed on 1 October 2022). The TCGA Research Network included patients following the guidelines of the Declaration of Helsinki of 1975 and all patients provided signed informed consent.
Publicly available gene expression and survival datasets were obtained from The Cancer Genome Atlas (TCGA) Research Network (http://cancergenome.nih.gov/, accessed on 1 October 2022), filtering for patients with HNCs (TCGA HNC). Of note, the expression values were not detectable for all genes of interest for every patient in the TCGA database. Here, VDR and RXR expression was found in $$n = 604$$ patients and analyzed as described [39]. Data were assessed via the USCS Xena server and patients were grouped according to the indicated phenotypic or clinical characteristics as described [37].
## 4.4. Cell Culture
Authenticated and characterized cell lines FaDu and SCC-4 were purchased from the ATCC repository, expanded, stocks prepared at early passages, and frozen stocks kept in liquid nitrogen. SCC-4 cells were established from a tongue squamous cell carcinoma. HNCUM-01T and -HNCUM-02T were established from tongue squamous cell carcinoma as described by Welkoborsky et al. [ 91]. The Pica cell line was established from laryngeal squamous cell carcinoma and maintained as described [39]. The FaDu cell line was established from a hypopharyngeal squamous cell carcinoma [92]. Thawed cells were routinely monitored by visual inspection and growth-curve analyses to keep track of the cell-doubling times, and were used for a maximum of 20 passages for all experiments. Depending on the passage number from purchase, cell line authentication was further performed at reasonable intervals by short tandem repeat (STR) profiling. We cultured the HNCUM-01T, HNCUM-02T, and SCC-4 cells in Dulbecco’s modified Eagle’s F-12 medium. Pica and FaDu cells were cultured in Dulbecco’s modified Eagle’s medium. We added $10\%$ fetal bovine serum (FBS), and $1\%$ penicillin-streptomycin to all medium types. Cells were cultured under a $5\%$ CO2 atmosphere at 37 °C and subcultured every 3 days as described [39]. We checked the absence of mycoplasma regularly via the Venor GeM Advance Detection Kit (Minverva Biolabs, Berlin, Germany) according to the manufacturer’s instructions. The cell numbers were determined using Casy Cell Counter and Analyzer TT (OMNI Life Science GmbH & Co KG, Bremen, Germany). To treat the cells, Hy-clone fetal bovine serum (FBS) (Sigma Aldrich, Munich, Germany) was used instead of standard FBS to ensure the absence of VitD in the controls and the control VitD treatment doses in the treated samples.
## 4.5. Generation of Cisplatin Resistant Cell Model
*We* generated constantly selected cell lines by treatment with sub-toxic doses of cisplatin corresponding to IC90 (5 µM) and then constant treatment (3 µM). We used the resistant cell line for experiments 6 months after constant exposure to cisplatin and the re-establishment of relatively regular proliferation.
## 4.6. Cell Viability Assays
To probe cell viability, we seeded the cells in 96-well plates (5000 to 15,000 cells/well) depending on the cell line and the treatment duration and treated them with the indicated substances and concentrations ($$n = 3$$) starting 24 h after seeding. After $\frac{48}{72}$ h treatment, we performed a commercially available assay CellTiter-Glo® 2.0 (Promega, Walldorf, Germany) according to the manufacturer’s instructions and recorded the luminescent signals using a Tecan Spark® (Tecan Group Ltd., Männedorf, Switzerland). Later, we normalized the signals to the untreated control samples.
In order to objectively determine the pharmacological effect of the proposed drug combination, we used the combination index equation described by Chou–Talalay [53]. In this algorithm, synergy is defined as combination index values < 1.0, antagonism as values > 1.0, and additivity as a value = 1.0.
## 4.7. Fluorescence Microscopy
Fluorescence images were acquired, analyzed, and quantified using an Axiovert 200 M fluorescence microscope (Zeiss, Oberkochen, Germany) or an automated high-content screening microscope Array Scan VTI (Thermo Fisher, Dreieich, Germany) as described [39,93,94]. We seeded cells in microscopic dishes (35 mm, MatTek, Ashland, MA, USA) or clear-bottom 96-well plates (Greiner, Kremsmünster Austria) and fixed them with $4\%$ PFA (20 min, RT). For immunofluorescence staining, we additionally permeabilized the cells via incubation with Triton-X 100 ($0.1\%$, 10 min, RT). Antibodies were diluted in $10\%$ FBS/PBS and incubated with samples for 1 h at RT. We washed the cells ($$n = 3$$) in PBS and then incubated the samples with fluorophore-labeled antibodies for 1 h at RT. Finally, we stained the nuclei by adding Hoechst 33342 (50 ng/mL in PBS) for 30 min at RT. For automated high-content screening, regions of interest were created using the nucleus signal and each sample was acquired in triplicate, imaging at least 5000 events per sample according to [39].
## 4.8. RNA Sequencing and Visualization
RNA sequencing was then performed as described in [95] and the visualizations were achieved with the help of GraphPad Prism. Ingenuity Pathway Analysis (Qiagen, Hilden, Germany) was used to visualize the mTOR and PI3K-AKT signaling pathways.
## 4.9. Plasmids and Transfection
To construct a VDR expression plasmid, cDNA was isolated out of the HNC cancer cell lines, and the full open reading frame of human VDR cDNA was cloned into the pcDNA3.1 mammalian expression vector (Invitrogen, Karlsruhe, Germany) with the C-terminal GFP-tag (for primer sequences, please see Supplementary Table S2). Colony PCR was performed to check for positive clones [93,94,96].
For cellular transfection, plasmid DNA and Lipofectamine 3000 (Fisher Scientific, Schwerte, Germany) were mixed according to the manufacturer’s instructions and added to the cells, which were cultured in Opti-MEM medium as described [97].To mark VDR-expressing cells, plasmid pC3 coding for GFP expression was co-transfected. To exclude artifacts, a control transfection of empty plasmid pC-DNA3 and the GFP-coding plasmid was conducted in parallel. The medium was changed 5 h post-transfection to a normal cell culture medium. We confirmed the VDR overexpression of cell lines via Western blot analysis, and positively transfected cells were selected by the addition of puromycin (1 µg/mL; Sigma Aldrich, Munich, Germany). To establish a uniform expression of the VDR transfected cells, the cells were sorted into low, medium, and high fluorescence using FACS as previously described [96].
## 4.10. Protein Extraction, Immunoblot Analysis
Whole-cell lysates were prepared using low salt lysis RIPA buffer (50mM Tris pH8.0, 150 mM NaCl, 5 mM EDTA, $0.5\%$ NP-40, 1 mM DTT, 1 mM PMSF, Complete EDTA-free from Roche Diagnostics, Mannheim, Germany) and samples were separated on 8–$12\%$ SDS gels, as has previously described [96,98,99]. Blotting onto activated PVDF membranes was achieved with Trans-Blot Turbo (Bio-Rad, Munich, Germany) and blocking and antibody incubations (1 h/RT or 16 h/4 °C depending on antibody) were performed in $5\%$ milk powder or BSA in TBST or PBS. The detection of the luminescence signal of HRP-coupled secondary antibodies after the addition of Clarity Western ECL Substrate was performed using the ChemiDocTM imaging system (Bio-Rad). Equal loading of lysates was controlled by reprobing blots for housekeeping genes (Actin). At least $$n = 2$$ biological replicates were performed and representative results are shown. Results of the densitometric analyses of all Western blots can be found in the Supplementary Materials.
## 4.11. Statistical Analysis
Statistical analyses were performed using GraphPad Prism (version 9.3.1) as described [39]. Survival data were obtained from the USCS Xena server, visualized, and analyzed by GraphPad Prism (Log-rank/Mantel-Cox test; Hazard Ratio (Mantel-Haenszel)). For two groups, a paired or unpaired Student’s t-test, for more group analysis of variance (ANOVA) was performed. Unless stated otherwise, p values represent data obtained from two independent experiments conducted in triplicate. Statistical significance is represented in the figures as follows: * $p \leq 0.05$, ** $p \leq 0.01$, *** $p \leq 0.001$, **** $p \leq 0.0001$, and n.s. indicates not significant. A p-value that was less than 0.05 was considered statistically significant.
## References
1. Muñoz A., Grant W.B.. **Vitamin D and cancer: An historical overview of the epidemiology and mechanisms**. *Nutrients* (2022) **14**. DOI: 10.3390/nu14071448
2. Henn M., Martin-Gorgojo V., Martin-Moreno J.M.. **Vitamin D in Cancer Prevention: Gaps in Current Knowledge and Room for Hope**. *Nutrients* (2022) **14**. DOI: 10.3390/nu14214512
3. Gangwar S.K., Kumar A., Jose S., Alqahtani M.S., Abbas M., Sethi G., Kunnumakkara A.B.. **Nuclear receptors in oral cancer-Emerging players in tumorigenesis**. *Cancer Lett.* (2022) **536** 215666. DOI: 10.1016/j.canlet.2022.215666
4. De Bosscher K., Desmet S.J., Clarisse D., Estébanez-Perpiña E., Brunsveld L.. **Nuclear receptor crosstalk—Defining the mechanisms for therapeutic innovation**. *Nat. Rev. Endocrinol.* (2020) **16** 363-377. DOI: 10.1038/s41574-020-0349-5
5. Tuna S., Aydin M.A., Aydin M.F.. **The Four Horsemen of the Apocalypse: Cancer, Depression, Vitamin D Deficiency, and Obesity: An Observational Study**. *Dis. Mrk.* (2023) **2023** 9652491. DOI: 10.1155/2023/9652491
6. Cheung F.S., Lovicu F.J., Reichardt J.K.. **Current progress in using vitamin D and its analogs for cancer prevention and treatment**. *Expert Rev. Anticancer Ther.* (2012) **12** 811-837. DOI: 10.1586/era.12.53
7. Sun Z., Sun X., Chen Z., Du J., Wu Y.. **Head and Neck Squamous Cell Carcinoma: Risk Factors, Molecular Alterations, Immunology and Peptide Vaccines**. *Int. J. Pept. Res. Ther.* (2022) **28** 19. DOI: 10.1007/s10989-021-10334-5
8. Papalouka C., Adamaki M., Batsaki P., Zoumpourlis P., Tsintarakis A., Goulielmaki M., Fortis S.P., Baxevanis C.N., Zoumpourlis V.. **DNA Damage Response Mechanisms in Head and Neck Cancer: Significant Implications for Therapy and Survival**. *Int. J. Mol. Sci.* (2023) **24**. DOI: 10.3390/ijms24032760
9. Zhang X., Weeramange C.E., Hughes B.G.M., Vasani S., Liu Z.Y., Warkiani M.E., Hartel G., Ladwa R., Thiery J.P., Kenny L.. **Application of circulating tumour cells to predict response to treatment in head and neck cancer**. *Cell. Oncol.* (2022) **45** 543-555. DOI: 10.1007/s13402-022-00681-w
10. Howren M.B., Christensen A.J., Pagedar N.A.. **Problem alcohol and tobacco use in head and neck cancer patients at diagnosis: Associations with health-related quality of life**. *Support. Care Cancer* (2022) **30** 8111-8118. DOI: 10.1007/s00520-022-07248-3
11. Dunlop E., Kiely M., James A.P., Singh T., Black L.J.. **The efficacy of vitamin D food fortification and biofortification in children and adults: A systematic review protocol**. *JBI Évid. Synth.* (2020) **18** 2694-2703. DOI: 10.11124/JBISRIR-D-19-00373
12. Aghajanpour M., Nazer M.R., Obeidavi Z., Akbari M., Ezati P., Kor N.M.. **Functional foods and their role in cancer prevention and health promotion: A comprehensive review**. *Am. J. Cancer Res.* (2017) **7** 740-769. PMID: 28469951
13. Zhang R., Zhang Y., Liu Z., Pei Y., Xu P., Chong W., Hai Y., He L., He Y., Yu J.. **Association between Vitamin D Supplementation and Cancer Mortality: A Systematic Review and Meta-Analysis**. *Cancers* (2022) **14**. DOI: 10.3390/cancers14153717
14. Zhang X., Hofmann S., Rack B., Harbeck N., Jeschke U., Sixou S.. **Fluorescence Analysis of Vitamin D Receptor Status of Circulating Tumor Cells (CTCS) in Breast Cancer: From Cell Models to Metastatic Patients**. *Int. J. Mol. Sci.* (2017) **18**. DOI: 10.3390/ijms18061318
15. Knuschke P.. **Sun Exposure and Vitamin D**. *Chall. Sun Prot.* (2021) **55** 296-315
16. Farrell C.-J., Herrmann M.. **Determination of vitamin D and its metabolites**. *Best Pract. Res. Clin. Endocrinol. Metab.* (2013) **27** 675-688. DOI: 10.1016/j.beem.2013.06.001
17. Cheng H., Narzo A.D., Howell D., Yevdokimova K., Zhang J., Zhang X., Pan Q., Zhang Z., Rogers L., Hao K.. **Ambient Air Pollutants and Traffic Factors Were Associated with Blood and Urine Biomarkers and Asthma Risk**. *Environ. Sci. Technol.* (2022) **56** 7298-7307. DOI: 10.1021/acs.est.1c06916
18. Holick M.F., Binkley N.C., Bischoff-Ferrari H.A., Gordon C.M., Hanley D.A., Heaney R.P., Murad M.H., Weaver C.M.. **Evaluation, Treatment, and Prevention of Vitamin D Deficiency: An Endocrine Society Clinical Practice Guideline**. *J. Clin. Endocrinol. Metab.* (2011) **96** 1911-1930. DOI: 10.1210/jc.2011-0385
19. Janoušek J., Pilařová V., Macáková K., Nomura A., Veiga-Matos J., Silva D.D.d., Remião F., Saso L., Malá-Ládová K., Malý J.. **Vitamin D: Sources, physiological role, biokinetics, deficiency, therapeutic use, toxicity, and overview of analytical methods for detection of vitamin D and its metabolites**. *Crit. Rev. Clin. Lab. Sci.* (2022) **59** 517-554. DOI: 10.1080/10408363.2022.2070595
20. Heath A.K., Kim I.Y., Hodge A.M., English D.R., Muller D.C.. **Vitamin D Status and Mortality: A Systematic Review of Observational Studies**. *Int. J. Environ. Res. Public Health* (2019) **16**. DOI: 10.3390/ijerph16030383
21. Migliaccio S., Di Nisio A., Magno S., Romano F., Barrea L., Colao A.M., Muscogiuri G., Savastano S.. **Vitamin D deficiency: A potential risk factor for cancer in obesity?**. *Int. J. Obes.* (2022) **46** 707-717. DOI: 10.1038/s41366-021-01045-4
22. Niedermaier T., Gredner T., Kuznia S., Schöttker B., Mons U., Brenner H.. **Vitamin D supplementation to the older adult population in Germany has the cost-saving potential of preventing almost 30 000 cancer deaths per year**. *Mol. Oncol.* (2021) **15** 1986-1994. DOI: 10.1002/1878-0261.12924
23. Gandini S., Boniol M., Haukka J., Byrnes G., Cox B., Sneyd M.J., Mullie P., Autier P.. **Meta-analysis of observational studies of serum 25-hydroxyvitamin D levels and colorectal, breast and prostate cancer and colorectal adenoma**. *Int. J. Cancer* (2011) **128** 1414-1424. DOI: 10.1002/ijc.25439
24. Yin L., Grandi N., Raum E., Haug U., Arndt V., Brenner H.. **Meta-analysis: Longitudinal studies of serum vitamin D and colorectal cancer risk**. *Aliment. Pharmacol. Ther.* (2009) **30** 113-125. DOI: 10.1111/j.1365-2036.2009.04022.x
25. Shi Q., Han X.P., Yu J., Peng H., Chen Y.Z., Li F., Cui X.B.. **Decreased vitamin D receptor protein expression is associated with progression and poor prognosis of colorectal cancer patients**. *Int. J. Clin. Exp. Pathol.* (2020) **13** 746-755. PMID: 32355523
26. Abdelgawad A., Hashem A., Mosbah A., Eissa L.A.. **A prospective trial investigating the role of Serum 25-Hydroxyvitamin D in diagnosis and prognosis of bladder cancer**. *PLoS ONE* (2022) **17**. DOI: 10.1371/journal.pone.0266371
27. Tokunaga E., Masuda T., Ijichi H., Tajiri W., Koga C., Koi Y., Nakamura Y., Ohno S., Taguchi K., Okamoto M.. **Impact of serum vitamin D on the response and prognosis in breast cancer patients treated with neoadjuvant chemotherapy**. *Breast Cancer* (2022) **29** 156-163. DOI: 10.1007/s12282-021-01292-3
28. Zhao Z., Cai W., Xing J., Zhao C.. **Lower vitamin D levels and VDR variants are risk factors for breast cancer: An updated meta-analysis**. *Nucl. Nucl. Nucleic Acids* (2022) **42** 17-37. DOI: 10.1080/15257770.2022.2107217
29. Lim S.T., Jeon Y.W., Gwak H., Suh Y.J.. **Clinical Implications of Serum 25-Hydroxyvitamin D Status after 5-Year Adjuvant Endocrine Therapy for Late Recurrence of Hormone Receptor-positive Breast Cancer**. *J. Breast Cancer* (2020) **23** 498-508. DOI: 10.4048/jbc.2020.23.e58
30. Carlberg C.. **Vitamin D and Its Target Genes**. *Nutrients* (2022) **14**. DOI: 10.3390/nu14071354
31. Pludowski P., Takacs I., Boyanov M., Belaya Z., Diaconu C.C., Mokhort T., Zherdova N., Rasa I., Payer J., Pilz S.. **Clinical Practice in the Prevention, Diagnosis and Treatment of Vitamin D Deficiency: A Central and Eastern European Expert Consensus Statement**. *Nutrients* (2022) **14**. DOI: 10.3390/nu14071483
32. Schweitzer A., Knauer S.K., Stauber R.H.. **Nuclear receptors in head and neck cancer: Current knowledge and perspectives**. *Int. J. Cancer* (2010) **126** 801-809. DOI: 10.1002/ijc.24968
33. Font-Díaz J., Jiménez-Panizo A., Caelles C., Vivanco M.d., Pérez P., Aranda A., Estébanez-Perpiñá E., Castrillo A., Ricote M., Valledor A.F.. **Nuclear receptors: Lipid and hormone sensors with essential roles in the control of cancer development**. *Semin. Cancer Biol.* (2021) **73** 58-75. DOI: 10.1016/j.semcancer.2020.12.007
34. Stauber R.H., Wunsch D., Knauer S.K., Fetz V.. **An update on the pathobiological relevance of nuclear receptors for cancers of the head and neck**. *Histol. Histopathol.* (2010) **25** 1093-1104. DOI: 10.14670/HH-25.1093
35. Germain P., Staels B., Dacquet C., Spedding M., Laudet V.. **Overview of nomenclature of nuclear receptors**. *Pharmacol. Rev.* (2006) **58** 685-704. DOI: 10.1124/pr.58.4.2
36. Wu X., Hu W., Lu L., Zhao Y., Zhou Y., Xiao Z., Zhang L., Zhang H., Li X., Li W.. **Repurposing vitamin D for treatment of human malignancies via targeting tumor microenvironment**. *Acta Pharm. Sin. B* (2019) **9** 203-219. DOI: 10.1016/j.apsb.2018.09.002
37. Khamis A., Gül D., Wandrey M., Lu Q., Knauer S.K., Reinhardt C., Strieth S., Hagemann J., Stauber R.H.. **The Vitamin D Receptor–BIM Axis Overcomes Cisplatin Resistance in Head and Neck Cancer**. *Cancers* (2022) **14**. DOI: 10.3390/cancers14205131
38. Huang Z., Zhang Y., Li H., Zhou Y., Zhang Q., Chen R., Jin T., Hu K., Li S., Wang Y.. **Vitamin D promotes the cisplatin sensitivity of oral squamous cell carcinoma by inhibiting LCN2-modulated NF-kappaB pathway activation through RPS3**. *Cell Death Dis.* (2019) **10** 936. DOI: 10.1038/s41419-019-2177-x
39. Siemer S., Fauth T., Scholz P., Al-Zamel Y., Khamis A., Gül D., Freudelsperger L., Wollenberg B., Becker S., Stauber R.H.. **Profiling Cisplatin Resistance in Head and Neck Cancer: A Critical Role of the VRAC Ion Channel for Chemoresistance**. *Cancers* (2021) **13**. DOI: 10.3390/cancers13194831
40. Giusti I., Poppa G., D’Ascenzo S., Esposito L., Vitale A.R., Calvisi G., Dolo V.. **Cancer Three-Dimensional Spheroids Mimic In Vivo Tumor Features, Displaying “Inner” Extracellular Vesicles and Vasculogenic Mimicry**. *Int. J. Mol. Sci.* (2022) **23**. DOI: 10.3390/ijms231911782
41. Siemer S., Bauer T.A., Scholz P., Breder C., Fenaroli F., Harms G., Dietrich D., Dietrich J., Rosenauer C., Barz M.. **Targeting cancer chemotherapy resistance by precision medicine-driven nanoparticle-formulated cisplatin**. *ACS Nano* (2021) **15** 18541-18556. DOI: 10.1021/acsnano.1c08632
42. Hagemann J., Jacobi C., Gstoettner S., Welz C., Schwenk-Zieger S., Stauber R., Strieth S., Kuenzel J., Baumeister P., Becker S.. **Therapy testing in a spheroid-based 3D cell culture model for head and neck squamous cell carcinoma**. *JoVE J. Vis. Exp.* (2018) **134** e57012
43. Khamis A.K., Fouad H.A., Raslan H.S., Fata M.M., Fayad A.I.. **Diagnostic and prognostic value of cancer stem cell marker CD44 and soluble CD44 in the peripheral Blood of patients with oral Squamous cell carcinoma**. *Open Sci. J.* (2017) **2** 1-23. DOI: 10.23954/osj.v2i3.1007
44. Brown A., Ravichandran K., Warnakulasuriya S.. **The unequal burden related to the risk of oral cancer in the different regions of the Kingdom of Saudi Arabia**. *Community Dent. Health* (2006) **23** 101. PMID: 16800366
45. Sawair F.A., Al-Mutwakel A., Al-Eryani K., Al-Surhy A., Maruyama S., Cheng J., Al-Sharabi A., Saku T.. **High relative frequency of oral squamous cell carcinoma in Yemen: Qat and tobacco chewing as its aetiological background**. *Int. J. Environ. Health Res.* (2007) **17** 185-195. DOI: 10.1080/09603120701254813
46. Deban M.A., El Didi F.H., Ramadan O.R.. **Quantitative Assessment of Tumor Associated Tissue Eosinophilia in Oral Squamous Cell Carcinoma (Clinicopathological Study)**. *Alex. Dent. J.* (2021) **46** 72-77. DOI: 10.21608/adjalexu.2020.21700.1034
47. Heijboer A.C., Blankenstein M.A., Kema I.P., Buijs M.M.. **Accuracy of 6 routine 25-hydroxyvitamin D assays: Influence of vitamin D binding protein concentration**. *Clin. Chem.* (2012) **58** 543-548. DOI: 10.1373/clinchem.2011.176545
48. Ferrer-Mayorga G., Larriba M.J., Crespo P., Munoz A.. **Mechanisms of action of vitamin D in colon cancer**. *J. Steroid Biochem. Mol. Biol.* (2018) **185** 1-6. DOI: 10.1016/j.jsbmb.2018.07.002
49. Idris S., Refaat B., Almaimani R.A., Ahmed H.G., Ahmad J., Alhadrami M., El-Readi M.Z., Elzubier M.E., Alaufi H.A.A., Al-Amin B.. **Enhanced in vitro tumoricidal effects of 5-Fluorouracil, thymoquinone, and active vitamin D3 triple therapy against colon cancer cells by attenuating the PI3K/AKT/mTOR pathway**. *Life Sci.* (2022) **296** 120442. DOI: 10.1016/j.lfs.2022.120442
50. Wu Z., Niu T., Xiao W.. **Uev1A promotes breast cancer cell survival and chemoresistance through the AKT-FOXO1-BIM pathway**. *Cancer Cell Int.* (2019) **19** 331. DOI: 10.1186/s12935-019-1050-4
51. Keum N., Lee D., Greenwood D., Manson J., Giovannucci E.. **Vitamin D supplementation and total cancer incidence and mortality: A meta-analysis of randomized controlled trials**. *Ann. Oncol.* (2019) **30** 733-743. DOI: 10.1093/annonc/mdz059
52. Chandler P.D., Chen W.Y., Ajala O.N., Hazra A., Cook N., Bubes V., Lee I.-M., Giovannucci E.L., Willett W., Buring J.E.. **Effect of vitamin D3 supplements on development of advanced cancer: A secondary analysis of the VITAL randomized clinical trial**. *JAMA Netw. Open* (2020) **3** e2025850. DOI: 10.1001/jamanetworkopen.2020.25850
53. Chou T.-C.. **Drug Combination Studies and Their Synergy Quantification Using the Chou-Talalay Method**. *Cancer Res.* (2010) **70** 440-446. DOI: 10.1158/0008-5472.CAN-09-1947
54. de Kort W., Spelier S., Devriese L., van Es R., Willems S.. **Predictive Value of EGFR-PI3K-AKT-mTOR-Pathway Inhibitor Biomarkers for Head and Neck Squamous Cell Carcinoma: A Systematic Review**. *Mol. Diagn. Ther.* (2021) **25** 123-136. DOI: 10.1007/s40291-021-00518-6
55. Day T.A., Shirai K., O’Brien P.E., Matheus M.G., Godwin K., Sood A.J., Kompelli A., Vick J.A., Martin D., Vitale-Cross L.. **Inhibition of mTOR Signaling and Clinical Activity of Rapamycin in Head and Neck Cancer in a Window of Opportunity Trial**. *Clin. Cancer Res.* (2019) **25** 1156-1164. DOI: 10.1158/1078-0432.CCR-18-2024
56. Lisse T.S., Hewison M.. **Vitamin D: A new player in the world of mTOR signaling**. *Cell Cycle* (2011) **10** 1888-1889. DOI: 10.4161/cc.10.12.15620
57. Hassan Z., Schneeweis C., Wirth M., Veltkamp C., Dantes Z., Feuerecker B., Ceyhan G.O., Knauer S.K., Weichert W., Schmid R.M.. **MTOR inhibitor-based combination therapies for pancreatic cancer**. *Br. J. Cancer* (2018) **118** 366-377. DOI: 10.1038/bjc.2017.421
58. Zou Z., Tao T., Li H., Zhu X.. **mTOR signaling pathway and mTOR inhibitors in cancer: Progress and challenges**. *Cell Biosci.* (2020) **10** 31. DOI: 10.1186/s13578-020-00396-1
59. Bochen F., Balensiefer B., Körner S., Bittenbring J.T., Neumann F., Koch A., Bumm K., Marx A., Wemmert S., Papaspyrou G.. **Vitamin D deficiency in head and neck cancer patients—Prevalence, prognostic value and impact on immune function**. *Oncoimmunology* (2018) **7** e1476817. DOI: 10.1080/2162402X.2018.1476817
60. Diaz L., Diaz-Munoz M., Garcia-Gaytan A.C., Mendez I.. **Mechanistic Effects of Calcitriol in Cancer Biology**. *Nutrients* (2015) **7** 5020-5050. DOI: 10.3390/nu7065020
61. Yao S., Ambrosone C.B.. **Associations between vitamin D deficiency and risk of aggressive breast cancer in African-American women**. *J. Steroid Biochem. Mol. Biol.* (2013) **136** 337-341. DOI: 10.1016/j.jsbmb.2012.09.010
62. Brenner H., Holleczek B., Schöttker B.. **Vitamin D insufficiency and deficiency and mortality from respiratory diseases in a cohort of older adults: Potential for limiting the death toll during and beyond the COVID-19 pandemic?**. *Nutrients* (2020) **12**. DOI: 10.3390/nu12082488
63. Forden C.L., Carrillo A.M.. **Smoking and attitudes toward smoking policy at a University in Egypt**. *J. Ethn. Subst. Abus.* (2016) **15** 329-345. DOI: 10.1080/15332640.2015.1066288
64. Hamadeh R.R., Lee J., Abu-Rmeileh N.M., Darawad M., Mostafa A., Kheirallah K.A., Yusufali A., Thomas J., Salama M., Nakkash R.. **Gender differences in waterpipe tobacco smoking among university students in four Eastern Mediterranean countries**. *Tob. Induc. Dis.* (2020) **18** 100. DOI: 10.18332/tid/129266
65. Jiang W., Wu D.-B., Xiao G.-B., Ding B., Chen E.-Q.. **An epidemiology survey of vitamin D deficiency and its influencing factors**. *Med. Clín. Engl. Ed.* (2020) **154** 7-12. DOI: 10.1016/j.medcle.2019.03.023
66. Welsh P., Peters M.J., Sattar N.. **Vitamin D insufficiency**. *N. Engl. J. Med.* (2011) **364** 1378-1379
67. Chakhtoura M., Akl E., El Ghandour S., Shawwa K., Arabi A., Mahfoud Z., Habib R., Hoballah H., El Hajj Fuleihan G.. **Impact of vitamin D replacement in adults and elderly in the Middle East and North Africa: A systematic review and meta-analysis of randomized controlled trials**. *Osteoporos. Int.* (2017) **28** 35-46. DOI: 10.1007/s00198-016-3837-7
68. Singh S., Mishra A.K., Jose S., Lodhi N.K.. **Atmospheric pollution and solar ultraviolet radiation in Asia**. *Asian Atmospheric Pollution* (2022) 129-146
69. Barrea L., Savastano S., Di Somma C., Savanelli M.C., Nappi F., Albanese L., Orio F., Colao A.. **Low serum vitamin D-status, air pollution and obesity: A dangerous liaison**. *Rev. Endocr. Metab. Disord.* (2017) **18** 207-214. DOI: 10.1007/s11154-016-9388-6
70. Al-Yatama F.I., AlOtaibi F., Al-Bader M.D., Al-Shoumer K.A.. **The Effect of Clothing on Vitamin D Status, Bone Turnover Markers, and Bone Mineral Density in Young Kuwaiti Females**. *Int. J. Endocrinol.* (2019) **2019** 6794837. DOI: 10.1155/2019/6794837
71. Judistiani R.T.D., Nirmala S.A., Rahmawati M., Ghrahani R., Natalia Y.A., Sugianli A.K., Indrati A.R., Suwarsa O., Setiabudiawan B.. **Optimizing ultraviolet B radiation exposure to prevent vitamin D deficiency among pregnant women in the tropical zone: Report from cohort study on vitamin D status and its impact during pregnancy in Indonesia**. *BMC Pregnancy Childbirth* (2019) **19** 1-9. DOI: 10.1186/s12884-019-2306-7
72. Estébanez N., Gómez-Acebo I., Palazuelos C., Llorca J., Dierssen-Sotos T.. **Vitamin D exposure and Risk of Breast Cancer: A meta-analysis**. *Sci. Rep.* (2018) **8** 9039. DOI: 10.1038/s41598-018-27297-1
73. Choi E.S., Oh S., Jang B., Yu H.J., Shin J.A., Cho N.P., Yang I.H., Won D.H., Kwon H.J., Hong S.D.. **Silymarin and its active component silibinin act as novel therapeutic alternatives for salivary gland cancer by targeting the ERK1/2-Bim signaling cascade**. *Cell. Oncol.* (2017) **40** 235-246. DOI: 10.1007/s13402-017-0318-8
74. Hendrickson W.K., Flavin R., Kasperzyk J.L., Fiorentino M., Fang F., Lis R., Fiore C., Penney K.L., Ma J., Kantoff P.W.. **Vitamin D receptor protein expression in tumor tissue and prostate cancer progression**. *J. Clin. Oncol.* (2011) **29** 2378-2385. DOI: 10.1200/JCO.2010.30.9880
75. Momen-Heravi F., Masugi Y., Qian Z.R., Nishihara R., Liu L., Smith-Warner S.A., Keum N., Zhang L., Tchrakian N., Nowak J.A.. **Tumor expression of calcium sensing receptor and colorectal cancer survival: Results from the nurses’ health study and health professionals follow-up study**. *Int. J. Cancer* (2017) **141** 2471-2479. DOI: 10.1002/ijc.31021
76. Kim S.H., Chen G., King A.N., Jeon C.K., Christensen P.J., Zhao L., Simpson R.U., Thomas D.G., Giordano T.J., Brenner D.E.. **Characterization of vitamin D receptor (VDR) in lung adenocarcinoma**. *Lung Cancer* (2012) **77** 265-271. DOI: 10.1016/j.lungcan.2012.04.010
77. Wang Z., Liu T., Li G., Cao Z.. **The exploration of new therapeutic targets for HPV-negative head and neck squamous cell cancer through the construction of a ceRNA network and immune microenvironment analysis**. *J. Cell. Biochem.* (2020) **121** 3426-3437. DOI: 10.1002/jcb.29615
78. Gao L., Wu Z.-X., Assaraf Y.G., Chen Z.-S., Wang L.. **Overcoming anti-cancer drug resistance via restoration of tumor suppressor gene function**. *Drug Resist. Updat.* (2021) **57** 100770. DOI: 10.1016/j.drup.2021.100770
79. Friedrich M., Axt-Fliedner R., Villena-Heinsen C., Tilgen W., Schmidt W., Reichrath J.. **Analysis of vitamin D-receptor (VDR) and retinoid X-receptor α in breast cancer**. *Histochem. J.* (2002) **34** 35-40. DOI: 10.1023/A:1021343825552
80. Carlberg C.. **Vitamin D in the Context of Evolution**. *Nutrients* (2022) **14**. DOI: 10.3390/nu14153018
81. Zhu D., Wu S., Hu C., Chen Z., Wang H., Fan F., Qin Y., Wang C., Sun H., Leng X.. **Folate-targeted polymersomes loaded with both paclitaxel and doxorubicin for the combination chemotherapy of hepatocellular carcinoma**. *Acta Biomater.* (2017) **58** 399-412. DOI: 10.1016/j.actbio.2017.06.017
82. Zehni A.Z., Batz F., Vattai A., Kaltofen T., Schrader S., Jacob S.-N., Mumm J.-N., Heidegger H.H., Ditsch N., Mahner S.. **The prognostic impact of retinoid X receptor and thyroid hormone receptor alpha in unifocal vs. multifocal/multicentric breast cancer**. *Int. J. Mol. Sci.* (2021) **22**. DOI: 10.3390/ijms22020957
83. Gul D., Schweitzer A., Khamis A., Knauer S.K., Ding G.B., Freudelsperger L., Karampinis I., Strieth S., Hagemann J., Stauber R.H.. **Impact of Secretion-Active Osteoblast-Specific Factor 2 in Promoting Progression and Metastasis of Head and Neck Cancer**. *Cancers* (2022) **14**. DOI: 10.3390/cancers14092337
84. Gnagnarella P., Muzio V., Caini S., Raimondi S., Martinoli C., Chiocca S., Miccolo C., Bossi P., Cortinovis D., Chiaradonna F.. **Vitamin D supplementation and cancer mortality: Narrative review of observational studies and clinical trials**. *Nutrients* (2021) **13**. DOI: 10.3390/nu13093285
85. Wang Y., Li C., Zhang Y., Zha X., Zhang H., Hu Z., Wu C.. **Aberrant mTOR/autophagy/Nurr1 signaling is critical for TSC-associated tumor development**. *Biochem. Cell Biol.* (2021) **99** 570-577. DOI: 10.1139/bcb-2021-0017
86. Wan P.K.-T., Siu M.K.-Y., Leung T.H.-Y., Mo X.-T., Chan K.K.-L., Ngan H.Y.-S.. **Role of Nurr1 in carcinogenesis and tumor immunology: A state of the art review**. *Cancers* (2020) **12**. DOI: 10.3390/cancers12103044
87. Wan P.K.-T., Leung T.H.-Y., Siu M.K.-Y., Mo X.-T., Tang H.W.-M., Chan K.K.-L., Cheung A.N.-Y., Ngan H.Y.-S.. **HPV-induced Nurr1 promotes cancer aggressiveness, self-renewal, and radioresistance via ERK and AKT signaling in cervical cancer**. *Cancer Lett.* (2021) **497** 14-27. DOI: 10.1016/j.canlet.2020.09.025
88. Suares A., Tapia C., González-Pardo V.. **Antineoplastic effect of 1α,25(OH)(2)D(3) in spheroids from endothelial cells transformed by Kaposi’s sarcoma-associated herpesvirus G protein coupled receptor**. *J. Steroid Biochem. Mol. Biol.* (2019) **186** 122-129. DOI: 10.1016/j.jsbmb.2018.10.004
89. Dienstmann R., Rodon J., Serra V., Tabernero J.. **Picking the Point of Inhibition: A Comparative Review of PI3K/AKT/mTOR Pathway InhibitorsDifferentiating PI3K/AKT/mTOR Pathway Inhibitors**. *Mol. Cancer Ther.* (2014) **13** 1021-1031. DOI: 10.1158/1535-7163.MCT-13-0639
90. Gül D., Habtemichael N., Dietrich D., Dietrich J., Gößwein D., Khamis A., Deuss E., Künzel J., Schneider G., Strieth S.. **Identification of cytokeratin24 as a tumor suppressor for the management of head and neck cancer**. *Biol. Chem.* (2021) **403** 869-890. DOI: 10.1515/hsz-2021-0287
91. Welkoborsky H.J., Jacob R., Riazimand S.H., Bernauer H.S., Mann W.J.. **Molecular biologic characteristics of seven new cell lines of squamous cell carcinomas of the head and neck and comparison to fresh tumor tissue**. *Oncology* (2003) **65** 60-71. DOI: 10.1159/000071206
92. Rangan S.. **A new human cell line (FaDu) from a hypopharyngeal carcinoma**. *Cancer* (1972) **29** 117-121. DOI: 10.1002/1097-0142(197201)29:1<117::AID-CNCR2820290119>3.0.CO;2-R
93. Habtemichael N., Wunsch D., Bier C., Tillmann S., Unruhe B., Frauenknecht K., Heinrich U.R., Mann W.J., Stauber R.H., Knauer S.K.. **Cloning and functional characterization of the guinea pig apoptosis inhibitor protein Survivin**. *Gene* (2010) **469** 9-17. DOI: 10.1016/j.gene.2010.08.007
94. Bier C., Knauer S.K., Klapthor A., Schweitzer A., Rekik A., Kramer O.H., Marschalek R., Stauber R.H.. **Cell-based analysis of structure-function activity of threonine aspartase 1**. *J. Biol. Chem.* (2011) **286** 3007-3017. DOI: 10.1074/jbc.M110.161646
95. Trothe J., Ritzmann D., Lang V., Scholz P., Pul Ü., Kaufmann R., Buerger C., Ertongur-Fauth T.. **Hypotonic stress response of human keratinocytes involves LRRC8A as component of volume-regulated anion channels**. *Exp. Dermatol.* (2018) **27** 1352-1360. DOI: 10.1111/exd.13789
96. Fetz V., Knauer S.K., Bier C., Von Kries J.P., Stauber R.H.. **Translocation biosensors–cellular system integrators to dissect CRM1-dependent nuclear export by chemicogenomics**. *Sensors* (2009) **9** 5423-5445. DOI: 10.3390/s90705423
97. Stauber R.H., Knauer S.K., Habtemichael N., Bier C., Unruhe B., Weisheit S., Spange S., Nonnenmacher F., Fetz V., Ginter T.. **A combination of a ribonucleotide reductase inhibitor and histone deacetylase inhibitors downregulates EGFR and triggers BIM-dependent apoptosis in head and neck cancer**. *Oncotarget* (2012) **3** 31-43. DOI: 10.18632/oncotarget.430
98. Tenzer S., Docter D., Kuharev J., Musyanovych A., Fetz V., Hecht R., Schlenk F., Fischer D., Kiouptsi K., Reinhardt C.. **Rapid formation of plasma protein corona critically affects nanoparticle pathophysiology**. *Nat. Nanotechnol.* (2013) **8** 772-781. DOI: 10.1038/nnano.2013.181
99. Docter D., Distler U., Storck W., Kuharev J., Wünsch D., Hahlbrock A., Knauer S.K., Tenzer S., Stauber R.H.. **Quantitative profiling of the protein coronas that form around nanoparticles**. *Nat. Protoc.* (2014) **9** 2030-2044. DOI: 10.1038/nprot.2014.139
|
---
title: The CaSR Modulator NPS-2143 Reduced UV-Induced DNA Damage in Skh:hr1 Hairless
Mice but Minimally Inhibited Skin Tumours
authors:
- Chen Yang
- Mark Stephen Rybchyn
- Warusavithana Gunawardena Manori De Silva
- Jim Matthews
- Katie Marie Dixon
- Andrew J. A. Holland
- Arthur David Conigrave
- Rebecca Sara Mason
journal: International Journal of Molecular Sciences
year: 2023
pmcid: PMC10002576
doi: 10.3390/ijms24054921
license: CC BY 4.0
---
# The CaSR Modulator NPS-2143 Reduced UV-Induced DNA Damage in Skh:hr1 Hairless Mice but Minimally Inhibited Skin Tumours
## Abstract
The calcium-sensing receptor (CaSR) is an important regulator of epidermal function. We previously reported that knockdown of the CaSR or treatment with its negative allosteric modulator, NPS-2143, significantly reduced UV-induced DNA damage, a key factor in skin cancer development. We subsequently wanted to test whether topical NPS-2143 could also reduce UV-DNA damage, immune suppression, or skin tumour development in mice. In this study, topical application of NPS-2143 (228 or 2280 pmol/cm2) to Skh:hr1 female mice reduced UV-induced cyclobutane pyrimidine dimers (CPD) ($p \leq 0.05$) and oxidative DNA damage (8-OHdG) ($p \leq 0.05$) to a similar extent as the known photoprotective agent 1,25(OH)2 vitamin D3 (calcitriol, 1,25D). Topical NPS-2143 failed to rescue UV-induced immunosuppression in a contact hypersensitivity study. In a chronic UV photocarcinogenesis protocol, topical NPS-2143 reduced squamous cell carcinomas for only up to 24 weeks ($p \leq 0.02$) but had no other effect on skin tumour development. In human keratinocytes, 1,25D, which protected mice from UV-induced skin tumours, significantly reduced UV-upregulated p-CREB expression ($p \leq 0.01$), a potential early anti-tumour marker, while NPS-2143 had no effect. This result, together with the failure to reduce UV-induced immunosuppression, may explain why the reduction in UV-DNA damage in mice with NPS-2143 was not sufficient to inhibit skin tumour formation.
## 1. Introduction
Skin cancers can be categorized into three main types: (i) basal cell carcinoma (BCC); (ii) squamous cell carcinoma (SCC), both of which arise from keratinocytes; and (iii) melanoma. Pathological changes in skin, including ultraviolet radiation (UV)-induced DNA damage [1,2,3], mutagenesis [4], inflammation [5], and immunosuppression [6,7] can ultimately lead to photocarcinogenesis. UV not only directly induces DNA lesions such as cyclobutane pyrimidine dimers (CPDs) and (6–4) photoproducts [8], but also induces indirect biological damage targeting DNA, protein, and lipids via the production of reactive oxygen species (ROS) and nitric oxide products, forming 8-hydroxy-2′-deoxyguanosine (8-OHdG) as a marker of oxidative DNA damage [9]. UV has potent immunosuppressive effects that promote tumour development [7,10,11,12].
Repetitive UV-induced epidermal thickening and pigmentation production together protected mice [13] and humans from subsequent UV challenges, with $75\%$ less erythema and $60\%$ less DNA damage in skin [14,15]. This thickening of the skin as a result of keratinocyte differentiation and may be more protective than melanogenesis (pigmentation production) in response to UV, at least in some populations [16]. Calcium concentration is believed to act as a switcher between proliferation and differentiation of keratinocytes [17]. This is consistent with a well-defined gradient for total calcium that increases from the basal to the outermost layers of the epidermis [18]. The responses of keratinocytes to extracellular calcium ion concentrations (Ca2+o) and the maintenance of systemic calcium homeostasis are mainly controlled by the calcium-sensing receptor (CaSR), a member of family C of the G protein-coupled receptors (GPCR) [19,20]. There are commercially available small molecule allosteric agents, for example, NPS-2143 works as an antagonist that reduces CaSR activity to block the increase of Ca2+i [21,22,23,24,25]. NPS-2143 has been used in an attempt to promote a brief secretion of parathyroid hormone in plasma for treatment of osteoporosis [22]. Previously we reported that CaSR knockdown or exposure to the CaSR negative allosteric modulator NPS-2143 protected human keratinocytes in culture against UV-induced DNA damage at a similar level to the known photoprotective agent, 1,25-dihydroxyvitamin D3 (1,25D) [26]. This photoprotective activity of NPS-2143 was attributed at least in part to enhanced DNA repair and to reduction in ROS [26].
Immunosuppression, along with UV-induced DNA lesions, is a key factor leading to photocarcinogenesis [27]. Topical 1,25D has been shown to protect mice from UV-induced CPDs, apoptotic sunburn cells, and UV-induced immunosuppression, and to reduce UV-induced skin tumours [28,29,30,31] as well as chemically-induced skin tumours [32]. The use of albino hairless (Skh:hr1) mice exposed to chronic UV is accepted as a reliable model of photocarcinogenesis [33,34,35]. While cultured primary keratinocytes provide a powerful approach for studying epidermal biology, they imperfectly model the multi-cell types and structural order of living epidermis [36]. Thus we aimed to investigate, for the first time in a mouse model, whether manipulation of the CaSR by its negative allosteric modulator NPS-2143 would protect against DNA damage in mouse epidermis after acute UV exposure. We also wanted to examine if topical treatment of NPS-2143 would reduce UV-induced skin inflammation and immune suppression, as well as in response to a chronic UV-exposure, whether it would reduce UV-induced skin tumours in comparison with the positive control, 1,25D.
## 2.1. NPS-2143 Protects against DNA Damage and Apoptotic Keratinocytes in Skh:hr1 Mice
Acute UV irradiation generated CPDs, oxidative DNA damage 8-OHdG (Figure 1), and sunburn cells (Figure 2) in mouse skin. The photoprotective hormonal form of vitamin D, 1,25D [29,37,38,39,40,41], was used as the positive control in these experiments. All agents in all experiments were applied topically immediately after exposure to solar-simulated UV (ssUV). Minimal staining in SHAM skin, particularly of 8-OHdG, indicates basal damage of the nuclei. In female mice, topical NPS-2143 at 2 concentrations, 228 pmol/cm2 and 2280 pmol/cm2 effectively reduced both CPD ($p \leq 0.05$, F(2.062, 19.25) = 15.64) and 8-OHdG ($p \leq 0.05$, F(1.537, 11.78) = 34.58) (Figure 1a–d). Topical NPS-2143 also reduced UV-induced sunburn cells ($p \leq 0.01$, F(2.044, 15.67)= 32.04) which are apoptotic keratinocytes with characteristic pyknotic nuclei and eosinophilic cytoplasm [42] (Figure 2a,b).
In male Skh:hr1 mice, significant reduction in UV-induced CPD was only seen after treatment with a high dose of NPS-2143, 2280 pmol/cm2 ($p \leq 0.01$, F(2.190, 18.98) = 6.738) (Figure 1e–h). Both concentrations of this agent, however, as well as 1,25D, significantly protected against oxidative DNA damage, 8-OHdG ($p \leq 0.01$, F(1.284, 14.55) = 8.726), in males (Figure 1f,h), and against sunburn cells ($p \leq 0.01$, F(2.663, 33.74) = 53.07) (Figure 2c,d).
## 2.2. NPS-2143 Effects on Inflammatory Response after ssUV and on Contact Hypersensitivity in Female Mice
After exposure to 3 minimal erythemal doses (MED) of solar-simulated UV, where a MED is defined as the lowest dose of UV which produces a mild reddening of the skin at 24 h, the mice developed skin edema [31]. In this study, skinfold thickness increased daily, reaching a maximum at day 4 post-UV, then decreased gradually (Figure 3a). On the 4th day, UV-induced edema was significantly reduced in the presence of 1,25D (11.4 pmol/cm2) ($p \leq 0.05$) or NPS-2143 (2280 pmol/cm2) ($p \leq 0.05$), compared to the vehicle-treated control mice (F(1.531,9.187) = 8.857).
In order to study how UV exposure affects contact hypersensitivity to oxazalone, female mice were exposed to ssUV or SHAM, then treated topically with the various agents. One week later, all the mice were sensitized with $2\%$ oxazolone applied to the non-irradiated abdominal skin. The mice were then challenged one week after this, by topical application of oxazalone to the ears to trigger swelling. Ear thickness measurements were taken before the challenge and again 18 h later. The average ear swelling expressed as the difference between ear thickness measured before and after challenge (at 18 h) in the non-UV exposed (SHAM) vehicle-treated mice was 293 ± 45 microns, and there were no differences among all non-irradiated groups (Figure 3b). In the UV-irradiated vehicle-treated mice, the average ear swelling of vehicle-treated mice was 155 ± 30 microns, indicating significant suppression of the immune response. With topical treatment with 1,25D, average ear swelling after UV was 217 ± 52 microns (Figure 3b), consistent with partial restoration of the contact hypersensitivity response. Though this swelling in response to oxazolone was smaller than in the SHAM with 1,25D-treated mice, the response was significantly better than in the vehicle-treated UV-exposed mice ($p \leq 0.05$, F (3.172, 28.55) = 21.73). Mice treated with NPS-2143 and UV had a measured average ear swelling of 177 ± 63 microns, not significantly different from vehicle-treated, UV exposed mice (Figure 3b). When calculated as a percent immune suppression after UV [31], the values were $52\%$ immune suppression in the vehicle-treated group, $26\%$ in the 1,25D-treated group ($p \leq 0.05$ vs vehicle-treated), and $39\%$ in the mice treated with NPS-2143 (n.s. vs vehicle-treated mice) (F (1.582, 21.36) = 3.405) (Figure 3c).
## 2.3. Study of NPS-2143 in Photocarcinogenesis in Female SKh:hr1 Mice
Albino hairless Skh:hr1 mice develop papillomas and then SCC after 10 weeks of chronic ssUV exposure [31,34,43]. During the 40 weeks of study, tumours normally appeared as small papillomas which gradually increased in diameter (Figure 4a). Papillomas are a benign outgrowth of skin in mice, comparable to the onset of actinic keratoses (AK) in humans [44]. A proportion of these papillomas showed signs of progression towards malignancy. These may be identified grossly and verified histologically as squamous cell carcinomas in later weeks (Figure 4a) [34].
The onset of detectable tumour (papilloma) formation in mice varied between treatment groups. As shown in Figure 4b, the latency in the vehicle-treated group was 24.0 ± 1.0 weeks. A significantly increased latency of 33.6 ± 2.5 weeks ($p \leq 0.0001$, F[2, 35] = 2.560) was seen in 1,25D-treated mice (11.4 pmol/cm2). The average latency for NPS-2143-treated mice (2280 pmol/cm2) was 22.4 ± 1.0 weeks, which was not significantly different from the vehicle control (Figure 4b).
Tumour multiplicity including both papillomas and SCCs was calculated at each weekly time point, as the average number of tumours per tumour-bearing mouse. Figure 4c shows tumour multiplicity throughout the 40-week study. Vehicle- and NPS-2143 (2280 pmol/cm2)-treated mice showed a steady increase in tumour multiplicity from week 16 to week 40, while 1,25D-treated mice (11.4 pmol/cm2) had remarkably lower tumour multiplicity. Compared to vehicle-treated mice, tumour multiplicity was significantly reduced in the 1,25D-treated group at all week-points assessed ($p \leq 0.05$ at week 20, $p \leq 0.01$ at week 25, 30 and 35, $p \leq 0.005$ at week 40, F[2, 45] = 2.255). However, there was no significant difference between NPS-2143 treated and vehicle-treated mice.
Progressive total tumour incidence including both papillomas and SCCs was calculated each week as the percentage of mice in each group bearing at least one tumour, as shown in Figure 4d. The incidence data were analysed statistically using a Mantel–Haenszel log-rank test (Mantel–Cox test) [45], in which all treatments were compared to vehicle-treated mice at 27 weeks and after (Table 1a). This analysis reveals whether there was a difference in the risk of developing a tumour. Mice treated with 1,25D (11.4 pmol/cm2) had significantly reduced tumour incidence compared with the vehicle-treated group throughout the entire experiment (Figure 4d green dotted line, Table 1a, Mantel–Cox test Chi-square Value = 27.09, df = 1). NPS-2143-treated (2280 pmol/cm2) mice demonstrated a time point-dependent increase in total tumour incidence compared to the vehicle-treated group at 27 weeks after the first irradiation, but by 28 weeks and over the subsequent period until 40 weeks, there was no significant difference (Figure 4d blue dotted line, Table 1a, Mantel–Cox test Chi-square Value = 4.061, df = 1).
Mice developed squamous cell carcinomas (SCCs) throughout the study from 18 weeks. The SCC-only incidence is shown in Figure 4e. Mice treated with 1,25D (11.4 pmol/cm2) had significantly reduced SCC incidence compared with the vehicle control group throughout the experiment (Figure 4e green dotted line, Table 1b, Mantel–Cox test Chi-square Value = 5.355, df = 1). Only one mouse in the group of 18 ($5.5\%$) treated with 1,25D developed an SCC at week 32 and this was still present at the end of the study. NPS-2143-treated mice had a significantly lower risk of developing SCC (5 out of 18, $27.8\%$) compared to the vehicle-treated group (8 out of 18, $44.4\%$) at the 24th week post-irradiation (Figure 4e Blue dotted line, Table 1b, Mantel-Cox test Chi-square Value = 22.09,df = 1). However, from the 25th week until the end of the experiment, there was no significant difference between the risk of SCC in NPS-2143- and vehicle-treated mice (Table 1b).
After UV exposure, CREB phosphorylation in epidermal cells increases and this has been proposed as a marker of tumour promoting activity [46]. In this study, 1,25D significantly reduced the risk of developing papillomas and SCC compared with the vehicle-treated group, while NPS-2143 had no overall effect on tumour or SCC incidence (Figure 4e, Table 1b).
Phosphorylation of CREB after UV, 1,25D, or NPS-2143 was studied in normal human keratinocytes. Negligible basal phospho-CREB (p-CREB) was seen in non-irradiated keratinocytes (SHAM) (Figure 4f). In cultured human keratinocytes, exposure to ssUV increased p-CREB measured 90 min after exposure (Figure 4f). Treatment of the cells immediately after UV with 1,25D significantly reduced p-CREB while treatment with NPS-2143 had no effect whether expressed as a function of tubulin as loading control (Figure 4f, F [3, 8] = 0.8378, and Figure S1a,) or as a function of total CREB (Figure S1b).
A summary of the main differences between responses to the positive control 1,25D and NPS-2143 is shown below (Table 2).
## 3. Discussion
In this study, the CaSR negative allosteric modifier NPS-2143, like the positive control 1,25D, when applied topically immediately after ssUV, effectively reduced UV-induced DNA lesions of CPD and 8-OHdG in female Skh:hr1 mice. Both NPS-2143 and 1,25D reduced oxidative DNA damage in male mice and at the higher concentration, NPS-2143 also reduced CPD in male mice. These results are consistent with the findings from our study using keratinocytes in primary culture from male human donors [26]. This is a discovery of a photo-protective role for NPS-2143, entirely different from its better recognised role as a therapeutic agent for raising parathyroid hormone levels. NPS-2143 negatively modulates the affinity of the CaSR for extracellular Ca2+, thereby reducing its activity [21,22,23,24,25]. In order to better discriminate the role of the CaSR in this study, it would have been useful to examine a CaSR antagonist (NPS2390 or Calcium-Sensing Receptor Antagonists I), but these studies were beyond the resources available for this work.
Sunburn cells and apoptotic keratinocytes were observed as soon as 3 h after acute exposure to UVB [47], despite being cells that undergo programmed cell death as a result of extensive and irreparable DNA damage [42]. It is reasonable to propose that reduced DNA damage, along with increased DNA repair [21] in the presence of NPS-2143, led to fewer apoptotic keratinocytes in mouse skin. In the meantime, we also observed improved survival of human keratinocytes in culture after UV exposure in the presence of NPS-2143 (Figure S2). It is likely that reduced generation of ROS, as previously reported with NPS-2143 after UV [21], contributed to reduced apoptosis. While sunburn cells are an index of apoptosis, analysis of more specific markers of apoptosis such as caspase$\frac{3}{7}$ or cleaved PARP in the mouse tissue would indicate early stages of apoptotic events and could help to elucidate the mechanism.
The reductions in 8-OHdG in male mice with 1,25D or NPS-2143 were similar to those seen in female mice. However, only the higher dose of NPS-2143 reduced CPD in male mice. Resistance in male mice to protection against UV-induced CPD in the presence of 1,25D has been previously reported in a separate study [47]. In that study, we demonstrated that the estrogen receptor-β (ER-β), the only estrogen receptor present in female mouse skin, seemed likely to be involved in reductions in CPD with 1,25D, since treatment with an ER-β antagonist or the use of female ER-β knockout mice reduced the response to 1,25D [47]. The current results indicate less effective protection against UV-induced CPD by a negative allosteric modulator of the CaSR. Whether this is also related to the presence of ER-β in female mouse skin or some other sex-related difference is an interesting question but was beyond the scope of the current study. A simple explanation for the reduced effectiveness of the lower dose of NPS-2143 could be that male mice have approximately $20\%$ thicker skin than female mice, regardless of UV [48]. DNA damage is a major contributor to UV-induced immune suppression [49] and susceptibility of male mice or humans to UV-induced immune suppression is greater than in their female counterparts [50,51]. Male mice are more susceptible than females to photocarcinogenesis [52], whereas incidence and mortality of skin cancers is greater in men [53,54].
Given resource limitations, the increased potential for male mice to fight and scratch, producing skin damage which would interfere with observations [55], meant that longer studies of skin edema, immune suppression, and tumour development after UV were only undertaken in female mice.
Topical NPS-2143, like 1,25D, produced a significant decrease in skin edema of mice, reflecting reduced inflammation after ssUV. UVR induces immediate and sustained production in NO in the skin [56,57,58,59] promoting the secretion of inflammatory mediators such as IL-6 [60]. A major limitation of the study is that it was not possible under the circumstances to examine a cytokine profile of mouse skin tissue before and after UV with or without NPS-2143 or 1,25D. Nevertheless, from the literature, there is evidence that NPS-2143 reduces NLRP3 inflammasome activation [61,62,63], overproduction of NO [64,65], the pro-inflammatory cytokine IL-6, and more [66,67,68,69]. These observations could explain the ability of NPS-2143 to reduce inflammation in mouse skin on day four after UV.
Somewhat surprisingly, despite a reduction in skin inflammation after UV, treatment with NPS-2143 had no effect on UV-induced immune suppression. Both DNA damage and increased IL-6 are important promoters of UV-dependent immune suppression [70]. Yet NPS-2143, like 1,25D, reduced DNA damage and, from the literature, also reduces IL-6 [66,67,68,69]. UV can directly damage antigen-presenting cells and promote the production of immunosuppressive cytokines such as IL-10 and IL-4 [71,72,73]. IL-10 is an anti-inflammatory cytokine and a potent immunosuppressant [74]. Secreted by UV-irradiated keratinocytes [75,76] and regulatory T cells [77], IL-10 not only prevents T cell expansion and activation but can also suppress other antigen-presenting cells [74]. It has been shown that reductions in CPD using liposomes containing T4N5 endonuclease led to reduced UV-immune suppression due to decreases in UV-upregulated IL-10 and TNF-α at both the mRNA and protein levels [78]. Furthermore, IL-10−/− mice were protected against photocarcinogenesis [79]. Though not tested in this study, IL-10 was increased with NPS-2143 treatment in rats [66,69]. This may explain why NPS-2143 failed to prevent UV-induced immunosuppression, though it protected against CPD and inflammation.
Skin tumour development depends on a combination of DNA damage, inflammation, and immunosuppression [49,80]. CPD are a major contributor to UV-induced mutations, so reduced CPDs might lead to fewer UV-induced mutations [80] and thus fewer tumours. Furthermore, enhanced repair of CPDs has been shown to reduce skin cancer incidence in mice [81] and humans [82] and we previously found that NPS-2143 increased DNA repair in keratinocytes [26]. Based on these findings, it seemed possible that NPS-2143 would have some protective capacity at an early stage of photocarcinogenesis due to its ability to reduce DNA damage and inflammatory reactions in vivo. In the photocarcinogenesis study, however, a single concentration of NPS-2143 (2280 pmol/cm2) was not superior to vehicle, either in time to develop the first tumour (including benign papilloma) or in the total number of tumours per mouse. Although NPS-2143 significantly reduced SCC incidence at 24 weeks, the effect did not persist.
Although the failure of NPS-2143 to prevent UV-induced immune suppression may explain its failure to prevent tumours in the chronic UV study, other factors may be involved. Cyclic AMP Response Element Binding protein (CREB) is a transcription factor essential for basic cellular function and homeostasis [83]. CREB is activated by phosphorylation at Ser133 by various kinases [83,84]. CREB overexpression supports growth and progression in various cancers [85,86,87,88,89]. CREB activation promotes enhanced cell proliferation, dysregulation of differentiation and reduced sensitivity to apoptosis and metastasis, particularly in melanoma [87,90] and SCC [88,91]. Using human keratinocytes, we observed that UV exposure increased phosphorylation of CREB at Ser 133 (phospho-CREB Ser133). While it would have been useful to verify the CREB and p-CREB changes in mouse skin, this was beyond the scope of the study and is a limitation. Our results in human keratinocytes are consistent with a recent study using reverse phase protein microarray analysis, which reported that p-CREB Ser133 was significantly activated at 1 h, 5 h, and 24 h after a single acute dose of 2MED UV in human skin [92] and with the report of increased p-CREB in mouse skin [46].
It has been argued that p-CREB is important in the initiation of papilloma formation, while other transcription factors such as CCAAT/enhancer binding protein (C/EBP)–B [93,94] control later stages of tumour growth and Activator Protein 1 (AP1) [95,96,97] maintains tumour identity. In a study of SCC, shRNA-mediated knockdown of CREB resulted in a significant increase in G2 phase arrest and a reduction in tumorigenic activity [91]. These authors identified that a key transcription factor complex, CREB and RFX1, which binds in the nucleus and is stabilized by CCAR2, is required to maintain proliferation in SCC [91]. Overexpression of CREB in a human squamous carcinoma cell line SCC13 remarkably increased its colony forming ability via a β-catenin-dependent pathway [88]. These studies suggest critical functions of CREB not only in the initial stage of papilloma formation but also in the development of neoplastic characteristics of SCC. Treatment of keratinocytes with 1,25D reduced UV-induced expression of p-CREB, fitting with its ability to protect mouse skin from developing both papillomas and SCC in the photocarcinogenesis study. NPS-2143, on the other hand, did not reduce UV-upregulated p-CREB. This may be part of the explanation for its inability to reduce tumour incidence, apart from its failure to decrease UV-induced immunosuppression.
Bikle et.al reported that double knockout of the vitamin D receptor and CaSR in the epidermis leads to spontaneous SCC formation in mice without any induction by UV, which was not observed in mice with deletion of either gene alone [98,99]. Those studies did not involve UV exposure. This is the first study to investigate whether negative modulation of the CaSR in skin alters responses to UV. NPS-2143 reduced two types of DNA damage in epidermal cells as well as skin inflammation to a similar extent as 1,25D, a known photo-protective agent (Figure 1, Figure 2 and Figure 3). However, NPS-2143 did not ameliorate UV-induced immune suppression (Figure 3). This latter observation, together with the failure of NPS-2143 to reduce post-UV CREB phosphorylation, probably explain the limited effect of this compound on skin tumour formation after ssUV (Figure 4). It is possible that the reduction in UV-induced DNA damage including oxidative damage by NPS-2143 may indicate an anti-aging effect [100]. These novel findings may lead to new research directions on the relationship between UV and the CaSR.
## 4.1. Studies in Mice
The in vivo studies were approved by the Animal Ethics Committee of the University of Sydney (Approval number: $\frac{2015}{794}$) and conformed to ARRIVE criteria. Skh:hr1 hairless albino mice, originally from Charles River (Wilmington, MA, USA), were from an in-house colony maintained at the University of Sydney. All Skh:hr1 hairless mice were housed in groups in wire-topped plastic boxes at an ambient temperature of 23–25°C under gold lighting (F40GO tubes, General Electric Co., Hobart, TAS, Australia) that does not emit UV radiation, and fed with Gordon Rat and Mouse Pellets (Yanderra, NSW, Australia) and tap water ad libitum. Male and female Skh:hr1 mice that were aged-matched in groups were used for experiments [31]. Mice were not allowed to be housed singly for this study but were housed in groups. Female mice are less prone to fighting than male mice and the fighting produces skin damage and artefacts [55]. For this reason, it is possible to study both female and male mice for DNA damage within hours after a UV exposure; however, the use of female mice for studies of skin edema or contact hypersensitivity conducted over 7 days and 16 days, respectively, or photocarcinogenesis (over 40 weeks) is preferred (Figure 5).
As previously established, the minimum erythemal dose (MED) of UV with this source for Skh:hr1 mice was 1.33 kJ/m2 UVB and 23.7 kJ/m2 UVA [31,101]. UV-irradiated mice were subjected to a single dose of 3 MED of UV (UVB value at 3.99 kJ/m2) for acute and immunosuppression studies. In the chronic photocarcinogenesis study, mice were subjected to 5 days of 0.75 MED followed by 5 days/week of 1 MED, for a total of 10 weeks (Figure 5).
## 4.2. Topical Treatments, DNA Damage and Sunburn Cells
Mice were treated topically over approximately 7 cm2 on the irradiated dorsal skin with 100 μL of vehicle only, or vehicle containing 1,25D (Sapphire Bioscience Pty Ltd., Redfern, NSW, Australia), or NPS-2143 2143 (HY-1007 MCE®, Medchem Express, Monmouth Junction, NJ, USA) immediately after irradiation, as previously described [31]. The compounds (1,25D and NPS-2143) were freshly diluted in spectroscopic grade ethanol (Merck, Darmstadt, Germany), combined with propylene glycol (Sigma-Aldrich, St. Louis, MO, USA) and MilliQ water at a ratio of 2:1:1 (v/v/v). Vehicle (base lotion) was combined, ethanol:propylene glycol:water 2:1:1 v/v [31]. The dose of NPS-2143, equivalent to 20× (228 pmol/cm2) and 200× (2280 pmol/cm2) of an effective dose of 1,25D 11.4 pmol/cm2 was determined according to the same ratio of 1,25D doses as determined from in vitro experiments [26,31]. Biopsies of dorsal skin were taken in triplicate from each mouse, 3 h post-UV and paraffin-embedded for immunohistochemistry of DNA damage as previously described [31]. Quantification of positive nuclei as % total nuclei (the percentage of CPD or 8-OHd positive nuclei staining in the selected nuclei in an area of epidermis) was obtained using MetaMorph (Molecular Devices, San Jose, CA, USA) and normalized to SHAM.
Routine haematoxylin and eosin staining was carried out by Veterinary Pathology Diagnostic Service (University of Sydney) to visualize sunburn cells. The stained sections were examined under a Zeiss Axioscan light microscope (Oberkochen, Germany) at 20× magnification, and the number of sunburn cells per linear millimetre of skin section recorded, as previously described [31,47]. Non-irradiated samples as SHAM control were obtained from the abdomen. Three areas of each section were analysed.
## 4.3. Skin Edema and Induction of Contact Hypersensitivity in Mice
Changes in dorsal skin thickness, a measure of edema, were recorded daily from 24 h onward until the until levels returned close to pre-UV condition on the 7th day after irradiation.
The contact hypersensitivity response was tested to investigate the effects of NPS-2143 on UV-induced systemic immunosuppression, as previously described [31]. Briefly, female mice were sensitized 1 week after irradiation and treatments, with 100 μL of $2\%$ oxazolone (Sigma-Aldrich, USA) (w/v) in absolute alcohol applied to the non-irradiated abdominal skin. Sensitization was repeated on the subsequent day. The sensitized mice were challenged 2 weeks after irradiation by application of 5 μL $2\%$ oxazolone to both surfaces of each ear, so that each mouse received 20 μL in total. Ear thickness measurements, taken using a spring micrometre (Interapid, Zurich, Switzerland), were recorded before the challenge and at 18 h after challenge, as previously reported [31]. The difference between pre- and post-oxazolone challenge ear thickness measurements of each mouse was recorded as ear swelling and the means for each group of 5 mice was calculated. Ear Swelling = pre-challenge ear thickness–post-challenge ear thickness.
The immune response was then calculated for each mouse, as shown in the formula below:[1]Immune Response =Ear swelling of UV IRRADIATED mice Ear swelling of UV NON−IRRADIATED SHAM mice Immunosuppression was calculated as $100\%$ minus this value, ± SEM [31] as in the formula below:Immunosuppression % = (1 − Immune Response) × 100[2]
## 4.4. Photocarcinogenesis
For this study, groups of 18 mice were used. Immediately after ssUV irradiation, mice were treated topically with either base lotion [31], 1,25D (11.4 pmol/cm2), or NPS-2143 (2280 pmol/cm2). During the next 30 weeks, the time of appearance, location, and visual identification of tumours with a diameter of at least 1 mm were monitored and mapped for each mouse. As previously described [31], the term “tumour” includes papilloma and SCC. The photocarcinogenic outcomes were reported as tumour latency, tumour incidence, tumour multiplicity, and SCC incidence. At the end of the experiment, all tumours were harvested for histological examination to confirm the classification.
## 4.5. Culture of Primary Human Keratinocytes
Keratinocytes were harvested from skin samples under University of Sydney Human Research Ethics Committee protocol no. $\frac{2015}{063}$ and cultured, as previously described [26]. The concentration of NPS-2143 used in these in vitro studies was based on previous experiments where we performed serial dilutions of NPS-2143 to determine the concentration-dependent response in human keratinocytes [26].A total of 500 nM was in the effective concentration range (5 nM~500 nM) and was also 10× IC 50; thus, it was chosen for all the in vitro experiments.
## 4.6. Western Blot
Keratinocytes were irradiated with an Oriel 1000 W xenon arc lamp (Newport Corporation, USA) and subsequently treated with vehicle, 1,25D, or NPS-2143, as in our previous study [26]. Western blot was performed, as previously described [102], with α-tubulin as the loading control. Primary antibodies used in this study were anti-phospho-CREB (Ser133) at 1 in 1000 dilution (mouse monoclonal, #9196, Cell Signaling Technology, Trask Lane Danvers, MA, USA), anti-CREB(Total) at 1 in 1000 dilution (mouse monoclonal, #9197, Cell Signaling Technology, Trask Lane Danvers, MA, USA), or anti-tubulin at 1µg/mL (mouse monoclonal, SC-5286, Santa Cruz Biotechnology). The band was imaged with the ChemiDocTM imaging system (Bio-Rad Laboratories, Inc, Hercules, CA, USA) and densitometry was carried out using Image J. SHAM, showing negligible expression of p-CREB which served as a negative control, and the data was normalized to UV+ vehicle to pool experiments.
## 4.7. Statistical Analysis
Animals in this study were divided into treatment groups of three for acute study, groups of five for immunosuppression study, and groups of eighteen for chronic photocarcinogenesis [31,47,101,103]. These numbers were determined by power analysis manually calculated from data from previous studies to have an $80\%$ chance showing a $20\%$ difference between treatment group at a significance level of $5\%$ [104,105]. Results are expressed as either as mean + SEM or as indicated. All the data sets passed goodness of fit tests, which determine whether sample data exhibit skewness and kurtosis that matches a normal distribution. All statistical analyses were performed with GraphPad Prism statistical program 9.0 (GraphPad Software Inc.). Unless otherwise stated, analysis of comparisons between treatment groups were made by linear mixed-model analysis, appropriate for dealing with repeated measurement data. The Mantel–Haenszel log-rank test (also called Mantel–Cox test) was used to analyse incidence data in the photocarcinogenesis study [31,45].
## References
1. Douki T., Bretonniere Y., Cadet J.. **Protection against radiation-induced degradation of DNA bases by polyamines**. *Radiat. Res.* (2000) **153** 29-35. DOI: 10.1667/0033-7587(2000)153[0029:PARIDO]2.0.CO;2
2. Cooke M.S., Podmore I.D., Mistry N., Evans M.D., Herbert K.E., Griffiths H.R., Lunec J.. **Immunochemical detection of UV-induced DNA damage and repair**. *J. Immunol. Methods* (2003) **280** 125-133. DOI: 10.1016/S0022-1759(03)00269-2
3. Mouret S., Charveron M., Favier A., Cadet J., Douki T.. **Differential repair of UVB-induced cyclobutane pyrimidine dimers in cultured human skin cells and whole human skin**. *DNA Repair* (2008) **7** 704-712. DOI: 10.1016/j.dnarep.2008.01.005
4. Ikehata H., Ono T.. **Significance of CpG methylation for solar UV-induced mutagenesis and carcinogenesis in skin**. *Photochem. Photobiol.* (2007) **83** 196-204. DOI: 10.1562/2006-02-28-IR-822
5. Hruza L.L., Pentland A.P.. **Mechanisms of UV-induced inflammation**. *J. Investig. Dermatol.* (1993) **100** 35S-41S. DOI: 10.1038/jid.1993.21
6. Halliday G.M.. **Inflammation, gene mutation and photoimmunosuppression in response to UVR-induced oxidative damage contributes to photocarcinogenesis**. *Mutat. Res.* (2005) **571** 107-120. DOI: 10.1016/j.mrfmmm.2004.09.013
7. Hart P.H., Norval M.. **Ultraviolet radiation-induced immunosuppression and its relevance for skin carcinogenesis**. *Photochem. Photobiol. Sci.* (2018) **17** 1872-1884. DOI: 10.1039/c7pp00312a
8. Ananthaswamy H.N., Bowden G.T., Fischer S.M.. **Ultraviolet Light as a Carcinogen**. *Comprehensive Toxicology, Chemical Carcinogens and Anticarcinogens* (1997) **Volume 12**
9. Peak M.J., Peak J.G., Carnes B.A.. **Induction of direct and indirect single-strand breaks in human cell DNA by far- and near-ultraviolet radiations: Action spectrum and mechanisms**. *Photochem. Photobiol.* (1987) **45** 381-387. DOI: 10.1111/j.1751-1097.1987.tb05390.x
10. Kripke M.L., Fisher M.S.. **Immunologic parameters of ultraviolet carcinogenesis**. *J. Natl. Cancer Inst.* (1976) **57** 211-215. DOI: 10.1093/jnci/57.1.211
11. Rangwala S., Tsai K.Y.. **Roles of the immune system in skin cancer**. *Br. J. Dermatol.* (2011) **165** 953-965. DOI: 10.1111/j.1365-2133.2011.10507.x
12. Katiyar S.K.. **UV-induced immune suppression and photocarcinogenesis: Chemoprevention by dietary botanical agents**. *Cancer Lett.* (2007) **255** 1-11. DOI: 10.1016/j.canlet.2007.02.010
13. Scott T.L., Christian P.A., Kesler M.V., Donohue K.M., Shelton B., Wakamatsu K., Ito S., D’Orazio J.. **Pigment-independent cAMP-mediated epidermal thickening protects against cutaneous UV injury by keratinocyte proliferation**. *Exp. Dermatol.* (2012) **21** 771-777. DOI: 10.1111/exd.12012
14. de Winter S., Vink A.A., Roza L., Pavel S.. **Solar-simulated skin adaptation and its effect on subsequent UV-induced epidermal DNA damage**. *J. Investig. Dermatol.* (2001) **117** 678-682. DOI: 10.1046/j.0022-202x.2001.01478.x
15. Young A.R., Potten C.S., Chadwick C.A., Murphy G.M., Hawk J.L., Cohen A.J.. **Photoprotection and 5-MOP photochemoprotection from UVR-induced DNA damage in humans: The role of skin type**. *J. Investig. Dermatol.* (1991) **97** 942-948. DOI: 10.1111/1523-1747.ep12491807
16. Gniadecka M., Wulf H.C., Mortensen N.N., Poulsen T.. **Photoprotection in vitiligo and normal skin-A quantitative assessment of the role of stratum corneum, viable epidermis and pigmentation**. *Acta Derm.-Venereol.* (1996) **76** 429-432. PMID: 8982403
17. Bikle D.D., Xie Z., Tu C.L.. **Calcium regulation of keratinocyte differentiation**. *Expert Rev. Endocrinol. Metab.* (2012) **7** 461-472. DOI: 10.1586/eem.12.34
18. Tu C.L., Crumrine D.A., Man M.Q., Chang W., Elalieh H., You M., Elias P.M., Bikle D.D.. **Ablation of the calcium-sensing receptor in keratinocytes impairs epidermal differentiation and barrier function**. *J. Investig. Dermatol.* (2012) **132** 2350-2359. DOI: 10.1038/jid.2012.159
19. Bikle D.D., Ratnam A., Mauro T., Harris J., Pillai S.. **Changes in calcium responsiveness and handling during keratinocyte differentiation. Potential role of the calcium receptor**. *J. Clin. Investig.* (1996) **97** 1085-1093. DOI: 10.1172/JCI118501
20. Tu C.L., Chang W., Bikle D.D.. **The extracellular calcium-sensing receptor is required for calcium-induced differentiation in human keratinocytes**. *J. Biol. Chem.* (2001) **276** 41079-41085. DOI: 10.1074/jbc.M107122200
21. Nemeth E.F.. **The search for calcium receptor antagonists (calcilytics)**. *J. Mol. Endocrinol.* (2002) **29** 15-21. DOI: 10.1677/jme.0.0290015
22. Nemeth E.F., Delmar E.G., Heaton W.L., Miller M.A., Lambert L.D., Conklin R.L., Gowen M., Gleason J.G., Bhatnagar P.K., Fox J.. **Calcilytic compounds: Potent and selective Ca2+ receptor antagonists that stimulate secretion of parathyroid hormone**. *J. Pharmacol. Exp. Ther.* (2001) **299** 323-331. PMID: 11561095
23. Hannan F.M., Walls G.V., Babinsky V.N., Nesbit M.A., Kallay E., Hough T.A., Fraser W.D., Cox R.D., Hu J., Spiegel A.M.. **The Calcilytic Agent NPS 2143 Rectifies Hypocalcemia in a Mouse Model With an Activating Calcium-Sensing Receptor (CaSR) Mutation: Relevance to Autosomal Dominant Hypocalcemia Type 1 (ADH1)**. *Endocrinology* (2015) **156** 3114-3121. DOI: 10.1210/en.2015-1269
24. Leach K., Wen A., Cook A.E., Sexton P.M., Conigrave A.D., Christopoulos A.. **Impact of clinically relevant mutations on the pharmacoregulation and signaling bias of the calcium-sensing receptor by positive and negative allosteric modulators**. *Endocrinology* (2013) **154** 1105-1116. DOI: 10.1210/en.2012-1887
25. Gregory K.J., Kufareva I., Keller A.N., Khajehali E., Mun H.C., Goolam M.A., Mason R.S., Capuano B., Conigrave A.D., Christopoulos A.. **Dual Action Calcium-Sensing Receptor Modulator Unmasks Novel Mode-Switching Mechanism**. *ACS Pharmacol. Transl. Sci.* (2018) **1** 96-109. DOI: 10.1021/acsptsci.8b00021
26. Yang C., Rybchyn M.S., De Silva W.G.M., Matthews J., Holland A.J.A., Conigrave A.D., Mason R.S.. **UV-induced DNA Damage in Skin is Reduced by CaSR Inhibition**. *Photochem. Photobiol.* (2022) **98** 1157-1166. DOI: 10.1111/php.13615
27. Nishisgori C.. **Current concept of photocarcinogenesis**. *Photochem. Photobiol. Sci.* (2015) **14** 1713-1721. DOI: 10.1039/c5pp00185d
28. Dixon K.M., Deo S.S., Wong G., Slater M., Norman A.W., Bishop J.E., Posner G.H., Ishizuka S., Halliday G.M., Reeve V.E.. **Skin cancer prevention: A possible role of 1,25dihydroxyvitamin D3 and its analogs**. *J. Steroid Biochem.* (2005) **97** 137-143. DOI: 10.1016/j.jsbmb.2005.06.006
29. Wong G., Gupta R., Dixon K.M., Deo S.S., Choong S.M., Halliday G.M., Bishop J.E., Ishizuka S., Norman A.W., Posner G.H.. **1,25-Dihydroxyvitamin D and three low-calcemic analogs decrease UV-induced DNA damage via the rapid response pathway**. *J. Steroid Biochem. Mol. Biol.* (2004) **89–90** 567-570. DOI: 10.1016/j.jsbmb.2004.03.072
30. Dixon K.M., Deo S.S., Norman A.W., Bishop J.E., Halliday G.M., Reeve V.E., Mason R.S.. **In vivo relevance for photoprotection by the vitamin D rapid response pathway**. *J. Steroid Biochem. Mol. Biol.* (2007) **103** 451-456. DOI: 10.1016/j.jsbmb.2006.11.016
31. Dixon K.M., Norman A.W., Sequeira V.B., Mohan R., Rybchyn M.S., Reeve V.E., Halliday G.M., Mason R.S.. **1 alpha,25(OH)(2)-Vitamin D and a Nongenomic Vitamin D Analogue Inhibit Ultraviolet Radiation-Induced Skin Carcinogenesis**. *Cancer Prev. Res.* (2011) **4** 1485-1494. DOI: 10.1158/1940-6207.CAPR-11-0165
32. Chida K., Hashiba H., Fukushima M., Suda T., Kuroki T.. **Inhibition of tumor promotion in mouse skin by 1 alpha,25-dihydroxyvitamin D3**. *Cancer Res.* (1985) **45** 5426-5430. PMID: 3840412
33. Forbes P.D., Blum H.F., Davies R.E.. **Photocarcinogenesis in hairless mice: Dose-response and the influence of dose-delivery**. *Photochem. Photobiol.* (1981) **34** 361-365. DOI: 10.1111/j.1751-1097.1981.tb09371.x
34. Reeve V.E., Ley R.D., Hill D., Elwood J.M., English D.R.. **Animal models of ultraviolet radiation-induced skin cancer**. *Prevention of Skin Cancer* (2004) 177-194. DOI: 10.1007/978-94-017-0511-0_10
35. Brown K., Balmain A.. **Transgenic mice and squamous multistage skin carcinogenesis**. *Cancer Metastasis Rev.* (1995) **14** 113-124. DOI: 10.1007/BF00665795
36. Demetriou S.K., Ona-Vu K., Teichert A.E., Cleaver J.E., Bikle D.D., Oh D.H.. **Vitamin D receptor mediates DNA repair and is UV inducible in intact epidermis but not in cultured keratinocytes**. *J. Investig. Dermatol.* (2012) **132** 2097-2100. DOI: 10.1038/jid.2012.107
37. Mason R., Holliday C.J., Norman A.W., Bouillon R., Thomasset M.. **1,25-Dihydroxyvitamin D Contributes to Photoprotection in Skin Cells**. *Vitamin D Endocrine System: Structural, Biological, Genetic and Clinical Aspects* (2000) 605-608
38. Gupta R., Dixon K.M., Deo S.S., Holliday C.J., Slater M., Halliday G.M., Reeve V.E., Mason R.S.. **Photoprotection by 1,25 dihydroxyvitamin D3 is associated with an increase in p53 and a decrease in nitric oxide products**. *J. Investig. Dermatol.* (2007) **127** 707-715. DOI: 10.1038/sj.jid.5700597
39. De Haes P., Garmyn M., Verstuyf A., De Clercq P., Vandewalle M., Degreef H., Vantieghem K., Bouillon R., Segaert S.. **1,25-Dihydroxyvitamin D3 and analogues protect primary human keratinocytes against UVB-induced DNA damage**. *J. Photochem. Photobiol. B Biol.* (2005) **78** 141-148. DOI: 10.1016/j.jphotobiol.2004.09.010
40. Lee J., Youn J.I.. **The photoprotective effect of 1,25-dihydroxyvitamin D3 on ultraviolet light B-induced damage in keratinocyte and its mechanism of action**. *J. Dermatol. Sci.* (1998) **18** 11-18. DOI: 10.1016/S0923-1811(98)00015-2
41. Manggau M., Kim D.S., Ruwisch L., Vogler R., Korting H.C., Schafer-Korting M., Kleuser B.. **1Alpha,25-dihydroxyvitamin D3 protects human keratinocytes from apoptosis by the formation of sphingosine-1-phosphate**. *J. Investig. Dermatol.* (2001) **117** 1241-1249. DOI: 10.1046/j.0022-202x.2001.01496.x
42. Sheehan J.M., Young A.R.. **The sunburn cell revisited: An update on mechanistic aspects**. *Photochem. Photobiol. Sci.* (2002) **1** 365-377. DOI: 10.1039/b108291d
43. Saran A., Spinola M., Pazzaglia S., Peissel B., Tiveron C., Tatangelo L., Mancuso M., Covelli V., Giovannelli L., Pitozzi V.. **Loss of tyrosinase activity confers increased skin tumor susceptibility in mice**. *Oncogene* (2004) **23** 4130-4135. DOI: 10.1038/sj.onc.1207565
44. Ortonne J.P.. **From actinic keratosis to squamous cell carcinoma**. *Br. J. Dermatol.* (2002) **146** 20-23. DOI: 10.1046/j.1365-2133.146.s61.6.x
45. Mantel N., Haenszel W.. **Statistical Aspects of the Analysis of Data From Retrospective Studies of Disease**. *J. Natl. Cancer Inst.* (1959) **22** 719-748. PMID: 13655060
46. Rozenberg J., Rishi V., Orosz A., Moitra J., Glick A., Vinson C.. **Inhibition of CREB function in mouse epidermis reduces papilloma formation**. *Mol. Cancer Res.* (2009) **7** 654-664. DOI: 10.1158/1541-7786.MCR-08-0011
47. Tongkao-On W., Yang C., McCarthy B.Y., De Silva W.G.M., Rybchyn M.S., Gordon-Thomson C., Dixon K.M., Halliday G.M., Reeve V.E., Mason R.S.. **Sex Differences in Photoprotective Responses to 1,25-Dihydroxyvitamin D3 in Mice Are Modulated by the Estrogen Receptor-beta**. *Int. J. Mol. Sci.* (2021) **22**. DOI: 10.3390/ijms22041962
48. Cooperstone J.L., Tober K.L., Riedl K.M., Teegarden M.D., Cichon M.J., Francis D.M., Schwartz S.J., Oberyszyn T.M.. **Tomatoes protect against development of UV-induced keratinocyte carcinoma via metabolomic alterations**. *Sci. Rep.* (2017) **7** 5106. DOI: 10.1038/s41598-017-05568-7
49. Kripke M.L., Cox P.A., Alas L.G., Yarosh D.B.. **Pyrimidine dimers in DNA initiate systemic immunosuppression in UV-irradiated mice**. *Proc. Natl. Acad. Sci. USA* (1992) **89** 7516-7520. DOI: 10.1073/pnas.89.16.7516
50. Damian D.L., Patterson C.R., Stapelberg M., Park J., Barnetson R.S., Halliday G.M.. **UV radiation-induced immunosuppression is greater in men and prevented by topical nicotinamide**. *J. Investig. Dermatol.* (2008) **128** 447-454. DOI: 10.1038/sj.jid.5701058
51. Reeve V.E., Allanson M., Domanski D., Painter N.. **Gender differences in UV-induced inflammation and immunosuppression in mice reveal male unresponsiveness to UVA radiation**. *Photochem. Photobiol. Sci.* (2012) **11** 173-179. DOI: 10.1039/c1pp05224a
52. Thomas-Ahner J.M., Wulff B.C., Tober K.L., Kusewitt D.F., Riggenbach J.A., Oberyszyn T.M.. **Gender differences in UVB-induced skin carcinogenesis, inflammation, and DNA damage**. *Cancer Res.* (2007) **67** 3468-3474. DOI: 10.1158/0008-5472.CAN-06-3798
53. Foote J.A., Harris R.B., Giuliano A.R., Roe D.J., Moon T.E., Cartmel B., Alberts D.S.. **Predictors for cutaneous basal- and squamous-cell carcinoma among actinically damaged adults**. *Int. J. Cancer* (2001) **95** 7-11. DOI: 10.1002/1097-0215(20010120)95:1<7::AID-IJC1001>3.0.CO;2-X
54. Rees J.R., Zens M.S., Celaya M.O., Riddle B.L., Karagas M.R., Peacock J.L.. **Survival after squamous cell and basal cell carcinoma of the skin: A retrospective cohort analysis**. *Int. J. Cancer* (2015) **137** 878-884. DOI: 10.1002/ijc.29436
55. Theil J.H., Ahloy-Dallaire J., Weber E.M., Gaskill B.N., Pritchett-Corning K.R., Felt S.A., Garner J.P.. **The epidemiology of fighting in group-housed laboratory mice**. *Sci. Rep.* (2020) **10** 16649. DOI: 10.1038/s41598-020-73620-0
56. Deliconstantinos G., Villiotou V., Stavrides J.C.. **Increase of particulate nitric oxide synthase activity and peroxynitrite synthesis in UVB-irradiated keratinocyte membranes**. *Biochem. J.* (1996) **320** 997-1003. DOI: 10.1042/bj3200997
57. Villiotou V., Deliconstantinos G.. **Nitric oxide, peroxynitrite and nitroso-compounds formation by ultraviolet A (UVA) irradiated human squamous cell carcinoma: Potential role of nitric oxide in cancer prognosis**. *Anticancer. Res.* (1995) **15** 931-942. PMID: 7544092
58. Deliconstantinos G., Villiotou V., Stravrides J.C.. **Release by ultraviolet B (u.v.B) radiation of nitric oxide (NO) from human keratinocytes: A potential role for nitric oxide in erythema production**. *Br. J. Pharmacol.* (1995) **114** 1257-1265. DOI: 10.1111/j.1476-5381.1995.tb13341.x
59. Mowbray M., McLintock S., Weerakoon R., Lomatschinsky N., Jones S., Rossi A.G., Weller R.B.. **Enzyme-Independent NO Stores in Human Skin: Quantification and Influence of UV Radiation**. *J. Investig. Dermatol.* (2009) **129** 834-842. DOI: 10.1038/jid.2008.296
60. Urbanski A., Schwarz T., Neuner P., Krutmann J., Kirnbauer R., Kock A., Luger T.A.. **Ultraviolet light induces increased circulating interleukin-6 in humans**. *J. Investig. Dermatol.* (1990) **94** 808-811. DOI: 10.1111/1523-1747.ep12874666
61. Rossol M., Pierer M., Raulien N., Quandt D., Meusch U., Rothe K., Schubert K., Schoneberg T., Schaefer M., Krugel U.. **Extracellular Ca**. *Nat. Commun.* (2012) **3** 1329. DOI: 10.1038/ncomms2339
62. Gutierrez-Lopez T.Y., Orduna-Castillo L.B., Hernandez-Vasquez M.N., Vazquez-Prado J., Reyes-Cruz G.. **Calcium sensing receptor activates the NLRP3 inflammasome via a chaperone-assisted degradative pathway involving Hsp70 and LC3-II**. *Biochem. Biophys. Res. Commun.* (2018) **505** 1121-1127. DOI: 10.1016/j.bbrc.2018.10.028
63. Lee G.S., Subramanian N., Kim A.I., Aksentijevich I., Goldbach-Mansky R., Sacks D.B., Germain R.N., Kastner D.L., Chae J.J.. **The calcium-sensing receptor regulates the NLRP3 inflammasome through Ca2+ and cAMP**. *Nature* (2012) **492** 123-127. DOI: 10.1038/nature11588
64. Chiarini A., Armato U., Liu D., Dal Prà I.. **Calcium-Sensing Receptor Antagonist NPS 2143 Restores Amyloid Precursor Protein Physiological Non-Amyloidogenic Processing in Aβ-Exposed Adult Human Astrocytes**. *Sci. Rep.* (2017) **7** 1277. DOI: 10.1038/s41598-017-01215-3
65. Armato U., Chiarini A., Chakravarthy B., Chioffi F., Pacchiana R., Colarusso E., Whitfield J.F., Dal Prà I.. **Calcium-sensing receptor antagonist (calcilytic) NPS 2143 specifically blocks the increased secretion of endogenous Aβ42 prompted by exogenous fibrillary or soluble Aβ25–35 in human cortical astrocytes and neurons—Therapeutic relevance to Alzheimer’s disease**. *Biochim. Biophys. Acta (BBA)-Mol. Basis Dis.* (2013) **1832** 1634-1652
66. Zhai T.Y., Cui B.H., Zou L., Zeng J.Y., Gao S., Zhao Q., Wang Y., Xie W.L., Sun Y.H.. **Expression and Role of the Calcium-Sensing Receptor in Rat Peripheral Blood Polymorphonuclear Neutrophils**. *Oxid. Med. Cell. Longev.* (2017) **2017** 3869561. DOI: 10.1155/2017/3869561
67. Li T.T., Sun M.R., Yin X., Wu C.L., Wu Q.Y., Feng S.L., Li H., Luan Y., Wen J., Yan L.X.. **Expression of the calcium sensing receptor in human peripheral blood T lymphocyte and its contribution to cytokine secretion through MAPKs or NF-kappa B pathways (vol 53, pg 414, 2013)**. *Mol. Immunol.* (2013) **55** 429. DOI: 10.1016/j.molimm.2013.02.005
68. Wu C.-L., Wu Q.-Y., Du J.-J., Zeng J.-Y., Li T.-T., Xu C.-Q., Sun Y.-H.. **Calcium-sensing receptor in the T lymphocyte enhanced the apoptosis and cytokine secretion in sepsis**. *Mol. Immunol.* (2015) **63** 337-342. DOI: 10.1016/j.molimm.2014.08.007
69. Hu B., Tong F., Xu L., Shen Z., Yan L., Xu G., Shen R.. **Role of Calcium Sensing Receptor in Streptozotocin-Induced Diabetic Rats Exposed to Renal Ischemia Reperfusion Injury**. *Kidney Blood Press. Res.* (2018) **43** 276-286. DOI: 10.1159/000487685
70. Nishimura N., Tohyama C., Satoh M., Nishimura H., Reeve V.E.. **Defective immune response and severe skin damage following UVB irradiation in interleukin-6-deficient mice**. *Immunology* (1999) **97** 77-83. DOI: 10.1046/j.1365-2567.1999.00733.x
71. Murphy G.M., Norris P.G., Young A.R., Corbett M.F., Hawk J.L.M.. **Low-dose ultraviolet-B irradiation depletes human epidermal Langerhans cells**. *Br. J. Dermatol.* (1993) **129** 674-677. DOI: 10.1111/j.1365-2133.1993.tb03330.x
72. Aberer W., Schuler G., Stingl G., Honigsmann H., Wolff K.. **Ultraviolet light depletes surface markers of Langerhans cells**. *J. Investig. Dermatol.* (1981) **76** 202-210. DOI: 10.1111/1523-1747.ep12525745
73. Shreedhar V., Giese T., Sung V.W., Ullrich S.E.. **A cytokine cascade including prostaglandin E2, IL-4, and IL-10 is responsible for UV-induced systemic immune suppression**. *J. Immunol.* (1998) **160** 3783-3789. DOI: 10.4049/jimmunol.160.8.3783
74. de Vries J.E.. **Immunosuppressive and anti-inflammatory properties of interleukin 10**. *Ann. Med.* (1995) **27** 537-541. DOI: 10.3109/07853899509002465
75. Matsumura Y., Ananthaswamy H.N.. **Toxic effects of ultraviolet radiation on the skin**. *Toxicol. Appl. Pharmacol.* (2004) **195** 298-308. DOI: 10.1016/j.taap.2003.08.019
76. Rivas J.M., Ullrich S.E.. **Systemic suppression of delayed-type hypersensitivity by supernatants from UV-irradiated keratinocytes. An essential role for keratinocyte-derived IL-10**. *J. Immunol.* (1992) **149** 3865-3871. DOI: 10.4049/jimmunol.149.12.3865
77. Yagi H., Tokura Y., Wakita H., Furukawa F., Takigawa M.. **TCRV beta 7+ Th2 cells mediate UVB-induced suppression of murine contact photosensitivity by releasing IL-10**. *J. Immunol.* (1996) **156** 1824-1831. DOI: 10.4049/jimmunol.156.5.1824
78. Wolf P., Maier H., Mullegger R.R., Chadwick C.A., Hofmann-Wellenhof R., Soyer H.P., Hofer A., Smolle J., Horn M., Cerroni L.. **Topical treatment with liposomes containing T4 endonuclease V protects human skin in vivo from ultraviolet-induced upregulation of interleukin-10 and tumor necrosis factor-alpha**. *J. Investig. Dermatol.* (2000) **114** 149-156. DOI: 10.1046/j.1523-1747.2000.00839.x
79. Loser K., Apelt J., Voskort M., Mohaupt M., Balkow S., Schwarz T., Grabbe S., Beissert S.. **IL-10 controls ultraviolet-induced carcinogenesis in mice**. *J. Immunol.* (2007) **179** 365-371. DOI: 10.4049/jimmunol.179.1.365
80. You Y.H., Lee D.H., Yoon J.H., Nakajima S., Yasui A., Pfeifer G.P.. **Cyclobutane pyrimidine dimers are responsible for the vast majority of mutations induced by UVB irradiation in mammalian cells**. *J. Biol. Chem.* (2001) **276** 44688-44694. DOI: 10.1074/jbc.M107696200
81. Yarosh D., Alas L.G., Yee V., Oberyszyn A., Kibitel J.T., Mitchell D., Rosenstein R., Spinowitz A., Citron M.. **Pyrimidine Dimer Removal Enhanced by DNA Repair Liposomes Reduces the Incidence of UV Skin Cancer in Mice**. *Cancer Res.* (1992) **52** 4227-4231. PMID: 1638536
82. Yarosh D., Klein J., O’Connor A., Hawk J., Rafal E., Wolf P.. **Effect of topically applied T4 endonuclease V in liposomes on skin cancer in xeroderma pigmentosum: A randomised study. Xeroderma Pigmentosum Study Group**. *Lancet* (2001) **357** 926-929. DOI: 10.1016/S0140-6736(00)04214-8
83. Gonzalez G.A., Montminy M.R.. **Cyclic AMP stimulates somatostatin gene transcription by phosphorylation of CREB at serine 133**. *Cell* (1989) **59** 675-680. DOI: 10.1016/0092-8674(89)90013-5
84. Mayr B., Montminy M.. **Transcriptional regulation by the phosphorylation-dependent factor CREB**. *Nat. Rev. Mol. Cell Biol.* (2001) **2** 599-609. DOI: 10.1038/35085068
85. Conkright M.D., Montminy M.. **CREB: The unindicted cancer co-conspirator**. *Trends Cell Biol.* (2005) **15** 457-459. DOI: 10.1016/j.tcb.2005.07.007
86. Abramovitch R., Tavor E., Jacob-Hirsch J., Zeira E., Amariglio N., Pappo O., Rechavi G., Galun E., Honigman A.. **A pivotal role of cyclic AMP-responsive element binding protein in tumor progression**. *Cancer Res.* (2004) **64** 1338-1346. DOI: 10.1158/0008-5472.CAN-03-2089
87. Melnikova V.O., Dobroff A.S., Zigler M., Villares G.J., Braeuer R.R., Wang H., Huang L., Bar-Eli M.. **CREB Inhibits AP-2 alpha Expression to Regulate the Malignant Phenotype of Melanoma**. *PLoS ONE* (2010) **5**. DOI: 10.1371/journal.pone.0012452
88. Kim S.Y., Lee J.H., Sohn K.C., Im M., Lee Y., Seo Y.J., Lee J.H., Kim C.D.. **beta-Catenin Regulates the Expression of cAMP Response Element-Binding Protein 1 in Squamous Cell Carcinoma Cells**. *Ann. Dermatol.* (2018) **30** 119-122. DOI: 10.5021/ad.2018.30.1.119
89. Chhabra A., Fernando H., Watkins G., Mansel R.E., Jiang W.G.. **Expression of transcription factor CREB1 in human breast cancer and its correlation with prognosis**. *Oncol. Rep.* (2007) **18** 953-958. DOI: 10.3892/or.18.4.953
90. Jean D., Bar-Eli M.. **Regulation of tumor growth and metastasis of human melanoma by the CREB transcription factor family**. *Mol. Cell. Biochem.* (2000) **212** 19-28. DOI: 10.1023/A:1007128101751
91. Best S.A., Nwaobasi A.N., Schmults C.D., Ramsey M.R.. **CCAR2 Is Required for Proliferation and Tumor Maintenance in Human Squamous Cell Carcinoma**. *J. Investig. Dermatol.* (2017) **137** 506-512. DOI: 10.1016/j.jid.2016.09.027
92. Einspahr J.G., Curiel-Lewandrowski C., Calvert V.S., Stratton S.P., Alberts D.S., Warneke J., Hu C., Saboda K., Wagener E.L., Dickinson S.. **Protein activation mapping of human sun-protected epidermis after an acute dose of erythemic solar simulated light**. *NPJ Precis. Oncol.* (2017) **1** 34. DOI: 10.1038/s41698-017-0037-7
93. Oh W.J., Rishi V., Orosz A., Gerdes M.J., Vinson C.. **Inhibition of CCAAT/enhancer binding protein family DNA binding in mouse epidermis prevents and regresses papillomas**. *Cancer Res.* (2007) **67** 1867-1876. DOI: 10.1158/0008-5472.CAN-06-2746
94. Zhu S., Yoon K., Sterneck E., Johnson P.F., Smart R.C.. **CCAAT/enhancer binding protein-beta is a mediator of keratinocyte survival and skin tumorigenesis involving oncogenic Ras signaling**. *Proc. Natl. Acad. Sci. USA* (2002) **99** 207-212. DOI: 10.1073/pnas.012437299
95. Gerdes M.J., Myakishev M., Frost N.A., Rishi V., Moitra J., Acharya A., Levy M.R., Park S.W., Glick A., Yuspa S.H.. **Activator protein-1 activity regulates epithelial tumor cell identity**. *Cancer Res.* (2006) **66** 7578-7588. DOI: 10.1158/0008-5472.CAN-06-1247
96. Young M.R., Li J.J., Rincon M., Flavell R.A., Sathyanarayana B.K., Hunziker R., Colburn N.. **Transgenic mice demonstrate AP-1 (activator protein-1) transactivation is required for tumor promotion**. *Proc. Natl. Acad. Sci. USA* (1999) **96** 9827-9832. DOI: 10.1073/pnas.96.17.9827
97. Thompson E.J., MacGowan J., Young M.R., Colburn N., Bowden G.T.. **A dominant negative c-jun specifically blocks okadaic acid-induced skin tumor promotion**. *Cancer Res.* (2002) **62** 3044-3047. PMID: 12036911
98. Bikle D.D., Oda Y., Tu C.L., Jiang Y.. **Novel mechanisms for the vitamin D receptor (VDR) in the skin and in skin cancer**. *J. Steroid Biochem.* (2015) **148** 47-51. DOI: 10.1016/j.jsbmb.2014.10.017
99. Bikle D.D., Jiang Y., Nguyen T., Oda Y., Tu C.L.. **Disruption of Vitamin D and Calcium Signaling in Keratinocytes Predisposes to Skin Cancer**. *Front. Physiol.* (2016) **7** 296. DOI: 10.3389/fphys.2016.00296
100. Panich U., Sittithumcharee G., Rathviboon N., Jirawatnotai S.. **Ultraviolet Radiation-Induced Skin Aging: The Role of DNA Damage and Oxidative Stress in Epidermal Stem Cell Damage Mediated Skin Aging**. *Stem Cells Int.* (2016) **2016** 7370642. DOI: 10.1155/2016/7370642
101. Reeve V.E., Domanski D., Slater M.. **Radiation sources providing increased UVA/UVB ratios induce photoprotection dependent on the UVA dose in hairless mice**. *Photochem. Photobiol.* (2006) **82** 406-411. DOI: 10.1562/2005-09-29-RA-703
102. Rybchyn M.S., Slater M., Conigrave A.D., Mason R.S.. **An Akt-dependent increase in canonical Wnt signaling and a decrease in sclerostin protein levels are involved in strontium ranelate-induced osteogenic effects in human osteoblasts**. *J. Biol. Chem.* (2011) **286** 23771-23779. DOI: 10.1074/jbc.M111.251116
103. De Silva W.G.M., Han J.Z.R., Yang C., Tongkao-On W., McCarthy B.Y., Ince F.A., Holland A.J.A., Tuckey R.C., Slominski A.T., Abboud M.. **Evidence for Involvement of Nonclassical Pathways in the Protection From UV-Induced DNA Damage by Vitamin D-Related Compounds**. *JBMR Plus* (2021) **5** e10555. DOI: 10.1002/jbm4.10555
104. Dell R.B., Holleran S., Ramakrishnan R.. **Sample size determination**. *ILAR J.* (2002) **43** 207-213. DOI: 10.1093/ilar.43.4.207
105. Schmidt S.A.J., Lo S., Hollestein L.M.. **Research Techniques Made Simple: Sample Size Estimation and Power Calculation**. *J. Investig. Dermatol.* (2018) **138** 1678-1682. DOI: 10.1016/j.jid.2018.06.165
|
---
title: A Mouse Systems Genetics Approach Reveals Common and Uncommon Genetic Modifiers
of Hepatic Lysosomal Enzyme Activities and Glycosphingolipids
authors:
- Anyelo Durán
- David A. Priestman
- Macarena Las Heras
- Boris Rebolledo-Jaramillo
- Valeria Olguín
- Juan F. Calderón
- Silvana Zanlungo
- Jaime Gutiérrez
- Frances M. Platt
- Andrés D. Klein
journal: International Journal of Molecular Sciences
year: 2023
pmcid: PMC10002577
doi: 10.3390/ijms24054915
license: CC BY 4.0
---
# A Mouse Systems Genetics Approach Reveals Common and Uncommon Genetic Modifiers of Hepatic Lysosomal Enzyme Activities and Glycosphingolipids
## Abstract
Identification of genetic modulators of lysosomal enzyme activities and glycosphingolipids (GSLs) may facilitate the development of therapeutics for diseases in which they participate, including Lysosomal Storage Disorders (LSDs). To this end, we used a systems genetics approach: we measured 11 hepatic lysosomal enzymes and many of their natural substrates (GSLs), followed by modifier gene mapping by GWAS and transcriptomics associations in a panel of inbred strains. Unexpectedly, most GSLs showed no association between their levels and the enzyme activity that catabolizes them. Genomic mapping identified 30 shared predicted modifier genes between the enzymes and GSLs, which are clustered in three pathways and are associated with other diseases. Surprisingly, they are regulated by ten common transcription factors, and their majority by miRNA-340p. In conclusion, we have identified novel regulators of GSL metabolism, which may serve as therapeutic targets for LSDs and may suggest the involvement of GSL metabolism in other pathologies.
## 1. Introduction
Hydrolytic enzymes are abundant in lysosomes [1]. In a healthy cell, the biosynthesis and catabolism of macromolecules are subject to regulatory mechanisms that maintain cellular homeostasis [2]. The degradative processes in lysosomes are controlled by their own enzymes [3,4]. Lysosomes play a central role in several biological processes, including energy metabolism, signaling, plasma membrane repair, secretion, and others [3]. Loss-of-function variants in genes encoding lysosomal proteins cause lysosomal storage disorders (LSDs), a group of diseases characterized by intracellular buildup of partially degraded material [5]. Growing evidence suggests that variants in lysosomal genes increase the risk of developing Parkinson’s disease (PD) [6,7].
In the sphingolipidoses, a subset of LSDs, glycosphingolipids (GSLs) accumulate in late endocytic organelles (late endosomes/lysosomes) and participate in their pathological cascades [8]. Current treatments for LSDs include substrate reduction therapy (SRT), which aims to reduce the rate of biosynthesis of stored substrates [5,9,10], and enzyme replacement therapies (ERT) aimed at replacing a deficient enzyme [11,12]. Emerging treatments include gene and cell therapies [13,14,15] and chaperones for improving enzyme folding and trafficking [16]. Although there is a range of therapeutic options for LSDs, they have limitations, such as tissue accessibility [17], antibody-mediated reaction [18], cost [19], and others. So far, therapies aimed at increasing enzyme activity or reducing lipid levels by modulating a second (modifier) gene have not been studied. In this context, a deeper understanding of the regulatory mechanisms that govern GSLs metabolism must be uncovered to fully develop this approximation.
Genome-wide association studies (GWAS) in humans and systems genetics strategies, which include gene mapping in model organisms, have identified genetic regulators of physiological and pathophysiological processes [20,21,22]. The Hybrid Mouse Diversity Panel (HMDP) has been a useful tool because genomes and tissue transcriptomes are freely available, allowing the combination of modifier gene mapping by GWAS and pathway analysis [23,24]. In this study, we have analyzed the activities of 11 lysosomal enzymes and several of their natural substrates in 25 strains of the HMDP panel followed by gene mapping and transcript integration. We identified a lack of correlation between most enzyme activities and their mRNA levels. Similarly, most substrates had no association between their levels and the enzyme activity that catabolizes them. Finally, we mapped putative modifier genes of each lysosomal enzyme and GSL by GWAS. We found associations between the mRNA levels of many modifier genes and enzyme activities or GSL levels. We clustered the putative modifiers in pathways and identified common and uncommon genetic regulators between GSLs and lysosomal enzymes, including transcription factors that regulate them. Our discoveries may help develop novel therapeutics for diseases with altered lysosomal enzyme activities and GSLs.
## 2.1. High Variability in the Hepatic Activity of Lysosomal Enzymes across Mouse Strains
We measured hepatic enzyme activity of β-hexosaminidase A and B (defective in Tay-Sachs and Sandhoff disease, respectively), α-neuraminidase (defective in Sialidosis/Mucolipidosis Type I), α-galactosidase A and B (defective in Fabry and Schindler disease), β-D-galactosidase (defective in GM1 Gangliosidosis), α-glucosidase (defective in Pompe), chitotriosidase (elevated in Gaucher disease), α-L-fucosidase (defective in fucosidosis), lysosomal acid phosphatase (elevated in patients with Gaucher), and Tartrate-resistant acid phosphatase (TRAP; altered in Gaucher disease) by fluorimetry in liver samples derived from 25 inbred mice strains using 4-methylumbelliferone (4-MU) based artificial substrates. We observed significant variability in the average enzymatic activity between the different strains (ANOVA p ≤ 0.05) (Figure 1). We did not find changes in α-galactosidase A, lysosomal acid phosphatase, and TRAP activities across the tissues analyzed (Figure 1d,j,k). We observed unique activity distribution patterns across the strains for the other enzymes, suggesting specific modifiers for each enzyme.
## 2.2. Lack of Correlation between the Enzyme Activity and Its mRNA Levels
Advantages of using tissues derived from the HMDP panel of inbred mouse strains include the fact that their genomes are sequenced, and transcriptomic data are available. Thus, we analyzed potential correlations between the genes encoding lysosomal enzymes and their activities. Recently we described the natural variation of hepatic acid β-glucocerebrosidase levels across many different mouse strains and included them in this analysis [20]. We did not identify significant correlations between enzyme activity and its transcript levels (Figure 2), with the only exception being Glb1, the gene encoding for β-D-galactosidase ($r = 0.5775$; p ≤ 0.002) (Figure 2c). These results indicate that mRNA levels are a poor proxy for enzyme activities.
## 2.3. High Variability in the Hepatic Glycosphingolipid Levels across Mouse Strains
Next, we measured the levels of GSLs in livers of the inbred mice strains in which we had access to enough material for three biological replicates ($\frac{23}{25}$) by Normal Phase-High-Performance Liquid Chromatography (NP-HPLC). We observed significant variability in GSLs among the strains, especially in total GSLs, GM3-Gc, GM2-Gc, GM1agc, GM3, Gb3, GM1a, GM1b, GD1b, and GD1a (Figure 3). For example, the levels of GM3-Gc were significantly increased (ANOVA $p \leq 0.0001$) in NOD/ShiLtJ compared with the other samples (Figure 3b). These results indicate that GSLs levels vary across strains.
## 2.4. Correlations between the GSLs and the mRNA Levels of the Biosynthetic Genes
A possibility is that GSL levels could correlate with their biosynthesis rate. Since we started from frozen tissues, we could not test this directly. Instead, we utilized the transcriptomic data available from the repository GSE16780 UCLA Hybrid MDP Liver Affy HT M430A [24]. We found transcript probes for 21 mRNA of the 21 anabolic enzymes of the GSLs pathway and four GSL transfer proteins. The analyzed gene list of the biosynthetic pathway is presented in the Supplementary Table S1. The expression values were organized according to GSLs levels from lowest to highest and presented as a heatmap. The analysis showed significant correlations for Cgt (r = −0.4263; $$p \leq 0.042$$) with total GSLs (Figure 4a). For GM2-Gc with Cgt (r = −0.4582; $$p \leq 0.0279$$), Galgt1 ($r = 0.6078$; $$p \leq 0.0021$$), A4galt ($r = 0.4903$; $$p \leq 0.0176$$), Gltp (r = −0.454; $$p \leq 0.0296$$) (Figure 4b). GM3 levels correlated with Galgt1 (r = −0.579; $$p \leq 0.0038$$), Gltp ($r = 0.4151$; $$p \leq 0.0489$$) (Figure 4c). GM1a with Col4a3bp ($r = 0.4458$ $$p \leq 0.033$$) (Figure 4d). GM3-*Gc is* associated with Galgt1 (r = −0.9591, p ≤ 0.0001) and it was the most significant correlation (Figure 4e). GM1agc levels with Slc17a2 ($r = 0.4163$; $$p \leq 0.0482$$) (Figure 5f). Gb3 with A4galt ($r = 0.6011$, $$p \leq 0.0024$$) (Figure 4g) and GM1b with Galgt1 (r = −0.5764; $$p \leq 0.004$$) and St8sia5 (r = −0.4194, $$p \leq 0.0046$$) (Figure 4h). No significant correlations were found between the majority of GSLs and biosynthetic genes (Supplementary Table S2); thus, we analyzed potential correlations between GSL levels and the enzyme activity that catabolizes them across the mouse panel.
## 2.5. Lack of Correlation between Hepatic Lysosomal Enzyme Activity and Their Natural Substrates across Mouse Strains
It is possible to speculate that the strains that present high activity of a particular enzyme should have reduced levels of its natural substrate because the enzyme catabolizes it. Unexpectedly, for most enzymes, we did not find significant correlations between the GSL levels and the enzyme activity that degrades it (Figure 5), except for neuraminidase and GM3-Gc (r = −0.4706; $$p \leq 0.0234$$) (Figure 5g). These results suggest that for most strains, the rate of biosynthesis and/or uptake of GSLs varies along with the catabolic rates which most likely are genetically regulated.
## 2.6. Identification of Putative Modifier Genes of Lysosomal Enzyme Activity and Sphingolipids Levels
To identify genetic regulators, we conducted genome-wide association studies with a quality control analysis that considered the population structure of the HMDP panel strains to reduce false associations [25,26]. We used enzyme activity levels as a trait and included the β-glucosidase activity, which we reported previously in the same and a few other strains [20]. For all the enzymes together, we identified 211 significant Single Nucleotide Variants (SNVs) that passed the empiric threshold of significance p ≤ 4.1 × 10−6 (−log10P = 5.39), previously calculated by permutations [21,22,26], while the Bonferroni threshold was p ≤ 3.9 × 10−7 [26]. These SNVs were located in different genomic regions (exonic, intronic, UTR3, downstream, and intergenic) (Table 1, Supplementary Table S3) in a total of 137 non-redundant genes. Similarly, we identified 3215 SNVs associated with GSLs levels (1744 non-redundant genes) whose variants are located in different genomic regions (Table 1, Supplementary Table S3). These analyses indicated that our strategy has sufficient power to map putative modifier genes.
## 2.7. Correlations between the Traits and the mRNA Levels of Putative Modifiers
To prioritize the putative modifier genes that could regulate each enzyme, we searched for correlations between the transcript levels of putative modifier genes and their traits (enzyme activity and GSL levels, respectively) (Figure 6). We found transcript probes for 67 mRNA of the 137 putative modifiers of the enzymes. The expression values were organized according to enzyme activity from lowest to highest and presented as a heatmap. The analysis showed significant correlations in Fip1l1 (r = −0.4462; $$p \leq 0.0254$$) with α-L-fucosidase (Figure 6a). For β-D-galactosidase with Lyplal1 (r = −0.702; p = <0.0001), Arrdc4 ($r = 0.627$; $$p \leq 0.0008$$), Pde2a ($r = 0.5306$; $$p \leq 0.0064$$), Glb1 ($r = 0.5753$; $$p \leq 0.0026$$), Bptf ($r = 0.5135$; $$p \leq 0.0087$$), Oxr1 (r = −0.447; $$p \leq 0.0251$$) (Figure 6b). No significant correlations were found for the other enzymes analyzed. We used SIFT to explore the impact of genetic variants on the genes identified by GWAS (benign or deleterious changes) associated with changes in enzyme activity [27], because the full genomes of the strains are known [28]. This strategy identified 308 predicted deleterious variants (Supplementary Table S4) in 43 of the 67 genes whose functions are related to organelle biogenesis (Chchd6) [29], intracellular signaling (Pde4dip) [30], and tissue development (Fam181b) [31], among others. These results suggest that amino acid substitution could affect protein function and signaling pathways leading to changes in enzyme activity.
The same analysis was performed to identify putative modifiers of GSL levels (Figure 6c–f). For 1744 non-redundant SNVs, we found expression values for 994 genes. The analysis identified 45 significant correlations, of which 33 were correlated with GM3-Gc levels, 10 genes with LacCer, and one gene with GD1b and GA2 (Figure 6c–f). Overall, we recorded $4.9\%$ ($\frac{52}{1061}$) of significant correlations distributed between the two traits. We also explored the impact of genetic variants associated with changes in GSLs with SIFT [27]. This strategy identified 515 deleterious variants predicted to disrupt the protein structure (Supplementary Table S4) in 132 genes related to DNA methyltransferase activity (Setdb1) [32] and synapse (Slitrk1) [33], among others.
## 2.8. Enrichment Analysis and Common Modifier Genes between Glycosphingolipids Levels and Lysosomal Enzyme Activities
If there is an orchestrated regulation of GSL levels and the enzymes that degrade them, it would be expected to observe enrichment in common pathways [34]. We therefore utilized gProfiler [35] to perform enrichment analysis using the putative modifier genes lists. For the modifier of enzyme activities, we found significantly associated pathways such as cell periphery ($$p \leq 5.9$$ × 10−4), plasma membrane ($$p \leq 2.4$$ × 10−3), and integral components of the plasma membrane ($$p \leq 2.6$$ × 10−2) (Figure 7b), which could be related to endocytic processes necessary to deliver key molecules to the lysosome, including the lysosomal enzymes that can be recycled from the extracellular space. Significant biological processes analysis included regulation of cellular processes ($$p \leq 3.9$$ × 10−2) (Figure 7d) (Supplementary Table S5). We did not find significant enrichment for the molecular function category. For GSLs, we observed enrichment in terms like cytoplasm ($$p \leq 3.5$$ × 10−28), cell junction ($$p \leq 7$$ × 10−21), synapse ($$p \leq 4.6$$ × 10−19), and 70 other pathways related to cellular components (Figure 7a; Supplementary Table S5). Many of these pathways require cellular membranes, where GSLs play a structural role. Significantly enriched Gene Ontology (GO) terms included protein binding ($$p \leq 9.1$$ × 10−31), ion binding ($$p \leq 8.8$$ × 10−14), binding ($$p \leq 2.3$$ × 10−13), ATP binding ($$p \leq 9.4$$ × 10−13), carbohydrate derivate binding ($$p \leq 1.8$$ × 10−1), and 27 other pathways related to molecular functions (Supplementary Table S5). Biological processes terms revealed 328 pathways, including system development ($$p \leq 4.3$$ × 10−39), anatomical structure development (5.6 × 10−38), and multicellular organism development ($$p \leq 1.4$$ × 10−37). We searched for the overlap between the cellular component domains of modifiers of enzyme activity and GSLs, which resulted in three common pathways (GO:0071944—cell periphery, GO:0005886—plasma membrane, and GO:0005887—integral component of plasma membrane) (Figure 7c) and one pathway associated with biological processes (GO:0050794; regulation of cellular process) (Supplementary Table S5).
## 2.9. Common and Uncommon Modifiers between Hepatic Lysosomal Enzyme Activity and Sphingolipids Levels
Common regulators of GSLs and enzymatic activities are relevant for understanding GSL metabolism and may be attractive therapeutic targets for LSDs. Therefore, we examined the overlap between them. We found 30 common and 1821 uncommon genes (Figure 7e). We explored their functions and identified genes involved in mitochondrial biogenesis and dynamics (Tfb1m, Timen135, Chchd6) [29,36,37], cell proliferation (Fstl5, Fzd10, Arhgap18) [38,39,40], platelet function (Cdh6) [41], vesicular trafficking (Vps45) [42], gene expression (Tfb1m, Zfat) [36,43], and regulating levels of the proto-oncogene MYC (Pvt1) [44]. Many of the 30 genes have been linked to diseases, such as Pvt1, Tiam2, Fstl5, Fzd10, Cdh6, Pvt1, Chchd6 in liver, colorectal, nasopharyngeal, and gastric cancer [45,46,47,48,49,50]. Others participate in neurodegenerative conditions; PD, schizophrenia, and intellectual disability (Tenm4, Pde4dip, Grid2, Arhgap18) [51,52]. These results suggest that lysosomal enzymes and GSLs may play a role in their pathophysiology and should be explored further (Table 2).
To better understand the molecular regulation of these 30 genes, we analyzed the transcription factors that bind to their promoters and/or enhancers (Figure 7f). We found no information for three of the 30 genes since they are putative (Rik) genes. The following transcription factors can bind to the 27 genes for which we have information: REST, TBP, CEBPB, EP300, POLR2A, FOS, DPF2, CTCF, RAD21, and SP1. Some of these transcription factors are broad regulators of transcription, such as TBP and POLR2A, while others are selective for specific processes, such as CTCF and RAD21. Considering all the promoters/enhancers of the 27 shared genes, we identified a total of 533 transcription factors that can bind them, although some only bind a few genes (Supplementary Table S6). We also searched for potential shared microRNA (miRNA) regulators using miRTarBase, a curated microRNA database [53]. We identified that miR-340-5p can bind to 11 of the 27 known common genes (Tusc1, Fam91a1, Zc3h12c, Adamts5, Tmem135, Tenm4, Grid2, Csnk1g3, Cdh6, Fam181b, and Pde4dip; $$p \leq 2.2$$ × 10−2) (Figure 7g). This result suggests that miRNA-340-5p regulates GSLs metabolism and may be involved in the pathogenesis of LSDs and the disorders described in Table 2.
**Table 2**
| Gene | Description | Traits | Traits.1 | Traits.2 | Traits.3 | Related Functions | Associated Human Diseases | Previosly Associated with Traits | References |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| Gene | Description | Enzyme | p-Value GWAS | GSLs | p-Value GWAS | | | | |
| Tiam2 | T cell lymphoma invasion and metastasis 2 | α-Glucosidase | 1.89 × 10−6 | GM3-Gc | 1.51 × 10−31 | neuroplasticity | liver cancer | No | [47,54] |
| Tfb1m | Dimethyladenosine transferase 1, mitochondrial | α-Glucosidase | 1.89 × 10−7 | GM3-Gc | 1.51 × 10−31 | promotion of mitochondrial biogenesis | deafness | No | [36,55] |
| Dok5 | Insulin receptor substrate 6 | β-D-galactosidase | 1.43 × 10−7 | Lac | 1.38 × 10−12 | osteoblast differentiation, insulin and IGF-1 signaling | cancer, Alzheimer’s disease | No | [56,57,58,59] |
| 4930433b08Rik | RIKEN cDNA 4930433B08 gene | β-D-galactosidase | 2.32 × 10−8 | Lac | 1.38 × 10−12 | - | - | - | - |
| A830019l24Rik | RIKEN cDNA A830019L24 gene | β-D-galactosidase | 1.43 × 10−7 | Lac | 1.38 × 10−12 | - | - | - | - |
| Tmem135 | Transmembrane protein 135 | β-D-galactosidase | 1.43 × 10−7 | GM3-Gc | 1.51 × 10−31 | involved in mitochondrial dynamics | retinal diseases | No | [37,60] |
| Fam181b | Family with sequence similarity 181, member B | β-D-galactosidase | 1.43 × 10−7 | GM3-Gc | 1.51 × 10−31 | increased expression during mouse development | - | No | [31] |
| Tenm4 | Teneurin transmembrane protein 4 | β-D-galactosidase | 1.43 × 10−7 | GM3-Gc | 1.51 × 10−31 | cell maturation and myelination in SNC | neuropsychiatric disorders, Parkinson’s disease | No | [51,61,62,63] |
| Plk2 | Serine/Threonine-protein kinase PLK2 | α-L-Fucosidase | 1.43 × 10−7 | GM3-Gc | 1.51 × 10−31 | cell proliferation, alpha-synuclein phosphorylation | pulmonary fibrosis | No | [64,65] |
| Stk32a | Serine/Threonine kinase 32A | α-L-Fucosidase | 1.43 × 10−7 | GM3-Gc | 1.51 × 10−31 | kinase activity | lung cancer | No | [66,67] |
| Dpysl3 | Dihydropyrimidinase like 3 | α-L-Fucosidase | 1.43 × 10−7 | GM3-Gc | 1.51 × 10−31 | cell migration, cytoskeletal dynamics and inflammation | gastric cancer, amyotrophic lateral sclerosis | No | [68,69,70] |
| Prex1 | PIP3 Dependent rac exchange factor 1 | α-L-Fucosidase | 6.67 × 10−8 | GM3-Gc | 1.51 × 10−31 | contributes to the effector activity of mouse neutrophils | prostate cancer | No | [71,72] |
| Fstl5 | Follistatin-related protein 5 | α-L-Fucosidase | 6.67 × 10−8 | Lac | 1.38 × 10−12 | play a role in cell proliferation | hepatocellular carcinoma | No | [38,73] |
| Vps45 | Vacuolar protein sorting-associated protein 45 | α-L-Fucosidase | 1.13 × 10−1 | GM3-Gc | 1.51 × 10−31 | vesicle-mediated protein trafficking from the Golgi | neutrophil disorders | No | [42,74] |
| Hist2h2be | Histone cluster 2 H2B family member E | α-L-Fucosidase | 1.13 × 10−1 | GM3-Gc | 1.51 × 10−31 | is necessary for proliferation | breast cancer | No | [75] |
| Pde4dip | Phosphodiesterase 4D interacting protein | α-L-Fucosidase | 2.29 × 10−7 | GM3-Gc | 1.51 × 10−31 | cAMP-dependent pathway to Golgi and/or centrosomes | schizophrenia | No | [52] |
| Tusc1 | Tumor suppressor candidate 1 | α-L-Fucosidase | 6.67 × 10−8 | GM3-Gc | 1.51 × 10−31 | reduced cell proliferation in vitro e in vivo | glioblastoma | No | [76,77] |
| Fzd10 | Frizzled class receptor 10 | α-L-Fucosidase | 6.67 × 10−8 | GM3-Gc | 1.51 × 10−31 | promotes cell proliferation through Wnt1 | cancer | No | [39,48] |
| Grid2 | Glutamate ionotropic receptor delta type subunit 2 | α-L-Fucosidase | 6.67 × 10−8 | Lac | 1.38 × 10−12 | receptor for glutamate | neurodevelopmental syndrome/intellectual disability | No | [78] |
| Zc3h12c | Zinc finger CCCH-type containing 12C | α-L-Fucosidase | 1.64 × 10−6 | GM3-Gc | 1.51 × 10−31 | RNA stability associated with inflammatory genes | psoriasis | No | [79,80] |
| Arhgap18 | Rho GTPase activating protein 18 | α-L-Fucosidase | 6.67 × 10−8 | GM3-Gc | 1.51 × 10−31 | role in migration, spreading and controls stress fiber formation | schizophrenia in Chinese population | No | [40,81] |
| Cdh6 | Cadherin 6 | α-L-Fucosidase | 1.34 × 10−6 | Lac | 1.38 × 10−12 | inhibit platelet aggregation | cancer | No | [41,49] |
| Fam91a1 | Family with sequence similarity 91 member A1 | α-L-Fucosidase | 6.67 × 10−8 | GM3-Gc | 1.51 × 10−31 | WDR11 complex (vesicular trafficking) | adenocarcinoma | No | [82,83] |
| 4933412e24Rik | RIKEN cDNA 4933412E24 gene | α-L-Fucosidase | 6.67 × 10−8 | Lac | 1.38 × 10−12 | - | - | - | - |
| A1bg | Alpha−1B-Glycoprotein | α-L-Fucosidase | 6.67 × 10−8 | GM3-Gc | 1.51 × 10−31 | cell dynamics and acquired immune response | cervical and bladder carcinogenesis | No | [84,85,86] |
| Pvt1 | Pvt1 Oncogene | α-L-Fucosidase | 6.67 × 10−8 | GM3-Gc | 1.51 × 10−31 | promotes cell proliferation | cancer | No | [50,87] |
| Adamts5 | ADAM Metallopeptidase with thrombospondin type 1 motif 5 | α-L-Fucosidase | 6.67 × 10−8 | GM3-Gc | 1.51 × 10−31 | metalloproteinase that remoldels connective tissue | osteoarthritis | No | [88] |
| Csnk1g3 | Casein kinase 1 gamma 3 | α-L-Fucosidase | 3.51 × 10−7 | Lac | 1.38 × 10−12 | wnt signaling pathway | breast, brain and colon cancer | No | [89,90] |
| Chchd6 | Coiled-coil-helix-coiled-coil-helix domain containing 6 | Chitotriosidase | 1.37 × 10−6 | GA2 | 3.27 × 10−6 | mitochondrial membrane morphology | cancer | No | [29] |
| Zfat | Zinc finger protein ZFAT | TRAP | 7.72 × 10−7 | Lac | 1.38 × 10−12 | immune response | hashimoto’s disease | No | [43,91] |
## 3. Discussion
In this study we searched for genetic modulators involved in the regulation of the lysosomal enzyme activities and the levels of substrates related to GSLs, with the idea of finding novel therapeutics targets for disorders in which they participate. By GWASs, we identified common and uncommon genetic regulators, evaluated the associations between modifier gene mRNA levels and each trait, and also clustered them in pathways. We identified 30 shared putative modifiers and described the transcription factors that are predicted to regulate them, and we noted that the miRNA340-5p can bind to 11 of these genes.
Our first unexpected finding was that most lysosomal enzyme activities do not correlate with their mRNA levels, nor with most of their substrate levels. Although enzyme activity can decrease with age [92], we used sex and age-matched samples; thus, the variation observed across strains was shown not to be due to any of these factors.
Another unexpected finding was that GM2-Gc levels correlate with the mRNA levels of the *Cgt* gene, which encodes for the UDP-galactose ceramide galactosyltransferase (CGT). CGT is a key enzyme for the biosynthesis of galactocerebrosides. Gangliosides, including GM2 derivates, are built from glucosylceramide and not from the galacto series [93]. However, for most of the biosynthetic genes there were no associations between the amount of lipids and the transcript levels of their anabolic pathways. Altogether, our results suggest that the GSL biosynthesis rate and uptake differ across the mouse strains, suggesting the existence of specific modifier genes for each trait.
Our third unexpected finding was that TFEB, the master transcriptional regulator of lysosomal genes [94], did not appear in the list of modifiers of lysosomal enzymes. This may be due to the fact that we screened for enzymatic activity instead of mRNA levels, and we showed a lack of correlation between transcript levels and enzyme activity under physiological conditions, at least for most enzymes. One exception was β-D-galactosidase, for which we found a positive correlation between its transcript levels and activity. Furthermore, the GWAS for this enzyme identified Glb1, the gene encoding for β-D-galactosidase, as a putative modifier of its activity, validating the power of discovery of our population-based strategy [95].
Our study had some limitations: First, we quantified lysosomal traits from liver homogenates that were not in living or isolated organelles, which may have diluted enzyme activity or promoted molecular interactions that might not occur in vivo because of cellular compartmentalization. Second, we could not directly measure GSLs biosynthesis and uptake because we started with mouse liver samples. Third, we used SNV catalogs with imputation, which may lead to false associations, though with increased mapping resolution.
Most of the enzymes we assayed are associated with LSDs [5,8]. For many LSDs, no therapies are available, and the few currently available treatments have severe limitations [5]. In this context, targeting a modifier gene could be a novel therapeutic approach. For example, lack of β-D-galactosidase activity triggers GM1 gangliosidosis, a disease with no approved therapies [96]. Our study identified the druggable Lypla1 and *Pkm* genes as putative modifiers of β-D-galactosidase activity, which can be pharmacologically modulated [97,98]. We found other druggable genes as well for several traits, and with the current gene editing technologies virtually any gene can be targeted. The potential modifying effects of these genes and compounds can be tested in LSDs disease models.
A hallmark of the sphingolipidoses is the intracellular buildup of GSLs, so strategies aimed at reducing their levels could lead us to novel therapies [5,8]. GSLs comprise a ceramide moiety with one or more sugar residues linked to it [99]. An approved therapy for Gaucher and Niemann-Pick disease type C is Miglustat [100,101], a small molecule inhibitor of GSL biosynthesis, thus reducing their levels. Our GSLs GWAS identified more than 50 genes previously associated with sphingolipid metabolism, which served as a positive control, including B3gnt5, Cln8, Hexb, Pnpla1, St8sia1, and Cgt. B3gnt5 regulates GSLs metabolism and lung tumorigenesis [102]. Our study also identified Lipc as a modifier of GM3-Gc levels, which has been previously associated with elevated serum levels of liver enzymes (alkaline phosphatase and γ-glutamyl transferase) [103], suggesting a new connection between GM3-Gc and liver damage. Variants in LIPC, CPS1, PABPC4, CITED2, TRPS1, and MVK are associated with changes in plasma lipoprotein levels [104], connecting novel traits to GSLs metabolism.
Lysosomal leakage has been associated with Alzheimers’ [105], cancer, and inflammation among other conditions [106]. Recently, the phosphoinositide signaling pathway was implicated in lysosomal repair [107]. *Many* genes of this pathway appear in our discovery list (Osbpl9, Osbpl6, Pde4dip, Pde2a, Pde1a, Pde7a, Pde7b, Pde4d, Pde8b, Pld5, Pik3r1, Pip4k2a, Pip5k1a, Pip5k1b, Pi4kb, Pdpk1, Atg4c, Atg10), suggesting that integrity of the lysosomal compartment is key to the proper functioning of enzymes and/or that these enzymes and lipids participate in lysosomal repair. Furthermore, this novel lysosomal repair pathway may facilitate the development of novel therapeutics for these diseases with lysosomal leakage.
Defects in the 30 shared genes are related to several pathologies, such as vision abnormalities (TMEM135) [60], cancer (CDH6 [49], FZD10 [48], TIAM2 [47]), neuropsychiatric disorders (Tenm4 [51], Pde4dip [52], Grid2 [78]), deafness (TFB1M) [55], neutrophil disorders (VPS45) [74] and others. Lysosomal enzymes and GSLs have been widely studied in cancer and neurodegenerative diseases [46,108,109,110,111]; however, their role in the other identified conditions should be explored.
Although not binding the complete list of shared genes, we identified some transcription factors previously known to be involved in lipid metabolism and autophagy-lysosomal functions (PPARγ, SREBF1, HNF1A, YY1, EGR1, SP1 and TFE3, E2F1, CREB1, MYC) [112,113,114,115,116,117,118,119,120,121], and many more that have not been previously linked to GSL metabolism. We also identified miR-340-5p as a putative regulator of many common modifier genes. Changes in miR-340-5p are linked to preeclampsia, neuroinflammation [122,123,124,125,126], adipocyte differentiation [127], as well as obesity and diabetes [128]. GSL metabolism plays a crucial role in the two last-mentioned disorders, and inhibitors of their biosynthesis have shown promising results in animal models of these conditions, validating the relevance of our strategy [129,130].
In conclusion, we described putative regulators of hepatic lysosomal enzymes and GSLs, many of them druggable and associated with diseases where alterations in GSL metabolism have not been previously described and should be assessed. We expect our findings may facilitate the development of novel therapeutics for conditions with alterations in these traits.
## 4.1. Mouse Tissues
We used 8 weeks-old mice livers derived from 25 inbred mouse strains, which were kindly donated by Dr. Aldons Lusis (University of California, Los Angeles, CA, USA). ( i) 129X1/SvJ, (ii) A/J, (iii) AKR/J, (iv) BALB/cJ, (v) BTBR T<+> tf/J, (vi) BUB/BnJ, (vii) C57BL/6J, (viii) C58/J, (ix) CAST/EiJ, (x) CBA/J, (xi) CE/J, (xii) DBA/2J, (xiii) KK/HlJ, (xiv) LG/J, (xv) LP/J, (xvi) MA/MyJ, (xvii) NOD/ShiLtJ, (xviii) NON/ShiLtJ, (xix) NZB/BlNJ, (xx) NZW/LacJ, (xxi) PL/J, (xxii) RIIIS/J, (xxiii) SEA/GnJ, (xxiv) SM/J, (xxv) SWR/J. Tissues were homogenized and adjusted to 50 mg tissue/mL in deionized water with a Potter-Elvehjem tissue homogenizer (Omni International, Kennesaw, GA, USA). Three or more livers per mouse strain were used to quantify traits (Supplementary Table S7).
## 4.2. Enzyme Activity Assays
Lysosomal hydrolase activities were determined using an artificial fluorescent substrate based on 4-methylumbelliferone (4-MU) [131]. For α-glucosidase, 1.47 mM 4-MU α-D-glucopyranoside (Sigma, Dorset, UK) in 100 mM citric acid/100 mM sodium phosphate, $0.1\%$ TritonX-100, pH 4.0 was used as substrate [132]. The substrate for α-galactosidase A and B activities was 5 mM 4-MU α-D-galactopyranoside (Santa Cruz, CA, USA) with and without 250 mM N-acetyl-galactosamine (Sigma, Dorset, UK) in 100 mM citric acid/100 mM tri-sodium citrate, $0.1\%$ TritonX-100, pH 4.0 [133,134]. For measuring β-hexosaminidase A and B activity, 3 mM 4-MU N-acetyl-β-D-glucosaminide (BioChemika, Dorset, UK) in 100 mM citric acid/100 mM sodium phosphate, $0.1\%$ TritonX-100, pH 4.5 was used as substrate. Heat inactivation assay for β-hexosaminidase A was carried out at 50 °C for 3 h [135]. For β-galactosidase activity, 1 mM 4-MU β-D-galactose (Sigma, Dorset, UK) in 200 mM sodium acetate buffer, 100 mM NaCl, $0.1\%$ TritonX-100, pH 4.3 was used as substrate [136]. The substrate for neuraminidase activity was 0.4 mM 4-MU α-D-N-acetylneuraminic acid (Sigma, Dorset, UK) in 0.1 M acetate buffer, $0.1\%$ TritonX-100, pH 4.6 [137,138]. For chitotriosidase activity, 0.013 mM 4-MU chitotrioside (Sigma, Dorset, UK) in 100 mM citric acid/200 mM sodium phosphate, $0.1\%$ TritonX-100, pH 5.2 was used as substrate [139,140]. For total acid phosphatase activity, 5 mM 4-MU phosphate (Sigma, Dorset, UK) with 40 mM NaCl in 200 mM citric acid/200 mM sodium phosphate, $0.1\%$ TritonX-100, pH 4.5 was used as substrate. For tartrate-resistant acid phosphatase (TRAP) activity, 5 mM 4-MU phosphate (Sigma, Dorset, UK) with 40 mM Na Tartrate in 200 mM citric acid/200 mM sodium phosphate, $0.1\%$ TritonX-100, pH 4.5 was used as substrate. The difference between total acid phosphatase activity and TRAP corresponded to lysosomal acid phosphatase (Lys AP) activity [141,142]. The substrate for α-L-fucosidase activity was 60 nM 4-MU α-L-fucopyranoside (Sigma, Dorset, UK) in 200 mM citric acid/200 mM sodium citrate, $0.1\%$ TritonX-100, pH 5.0 [143,144]. We determined the acid-β-glucosidase activity in the same tissues in a previous publication [20], and further analyses were performed here based on the published activity. Liver homogenates were diluted with the buffer corresponding to each enzymatic determination. Three cycles of freezing (liquid nitrogen) and thawing were performed on the samples. Three biological replicates of the diluted liver extracts were incubated with the corresponding substrate at 37 °C for 30 min (or 1 h for α-neuraminidase, β-D-galactosidase, and chitotriosidase). Cold 0.5 M Na2CO3 (pH 10.7) was added to stop the reaction. Fluorescence intensity in samples was measured in a Synergy HT plate reader (BioTek, Winooski, VT, USA) at $\frac{360}{460}$ nm. Protein concentration was measured using a BCA protein assay kit (Thermo Fisher Scientific, New Jersey, NJ USA). Fluorescence values were normalized to protein concentration. A 4-MU standard curve was constructed to calculate specific activity, and the final value was adjusted to one hour of enzymatic reaction.
## 4.3. Glycosphingolipids Levels Quantification
The GSLs were extracted and measured by Normal Phase-High-Performance Liquid Chromatography (NP-HPLC) following published methods [145]. Briefly, the aqueous tissue extract was homogenized in chloroform/methanol (C:M) (1:2 v/v) and kept overnight at 4 °C. Then, the extracts mixture was centrifuged at 3000 rpm for 10 min at room temperature. We added 0.5 mL of PBS and 0.5 mL of chloroform to the supernatant followed by a 3000-rpm centrifugation for 10 min at room temperature. The lower phase was carefully removed and dried under a stream of nitrogen gas (N2) in a heating block (42 °C), resuspended in 40 μL C:M 1:3 v/v and mixed with the upper phase. Afterwards, glycosphingolipids-derived oligosaccharides were purified from the samples using C18 columns (Telos, Kinesis, UK) previously pre-equilibrated with 1.25 mL methanol (four times) and 1.25 mL deionized water (three times). We loaded the mixed phase (lower/upper) onto a column and rinsed the sample tube with 1 × 1 mL of deionized water. Then, the C18 column was washed with 4 × 1.25 mL deionized water and eluted it with 1 × 1 mL (C:M) (98:2 v/v), 2 × 1 mL (C:M) (1:3 v/v), 1 × 1 mL methanol. The eluates were dried under N2 current and digested with a recombinant Endoglycoceramidase I (rEGCaseI) (GenScript, Oxford, UK) in buffer 50 mM sodium acetate, pH 5.0, $0.6\%$ TritonX-100 (4 μL enzyme + 86 μL buffer) at 37 °C for 16 h. The released glycans were labeled with 310 μL of labelling mix (30 mg/mL anthranilic acid (2AA) and 45 mg/mL sodium cyanoborohydride) in $4\%$ sodium acetate, $2\%$ boric acid in methanol, and heated at 80 °C. Then, we cooled the samples and mixed them with 3 × 1 mL acetonitrile: deionized water (97:3) (v/v) and added them to a Discovery DPA-6S-SPE tube (Supelco, PA, USA), pre-equilibrated with 1 × 1 mL acetonitrile, 2 × 1 mL deionized water, and 3 × 1 mL acetonitrile. The columns were cleaned with 3 × 1 mL acetonitrile: deionized water (95:5) (v/v), and the tubes were washed with 2 × 1 mL acetonitrile: deionized water (95:5) (v/v) and eluted in 0.6 mL deionized water. We took 60 μL from 0.6 mL sample eluted, added 140 μL acetonitrile, and injected 50 μL of this mix (deionized water: acetonitrile) (30:70) (v/v) onto NP-HPLC (Waters Alliance 2695 separations module and multi-fluorescent detector set at Ex 360/Em 425 nm). To calculate molar quantities from peaks in the chromatogram, we included a calibration standard containing 2.5 pmol 2AA-labelled chitotriose (Ludger, Oxford, UK) for each NP-HPLC run [145]. The chromatographic data were processed using Waters Empower software 3 (Waters, Milford, MA, USA). Fluorescence values by sample were normalized to protein content using a BCA Assay kit (Merck KGaA, Darmstadt, Germany).
## 4.4. Genome-Wide Association Studies (GWAS)
We used the genotype of each strain, and the enzymatic activity or substrate as trait, and its kinship matrix to perform the GWAS using The Efficient Mixed Model Association (EMMA) v.1.1.230 in the R package [26,146]. We used PLINK to remove SNVs in linkage disequilibrium to avoid false associations [25], considering an R2 = 0.25, leaving 127,285 independent variants out of the initial four million variants downloaded from the mouse HapMap reference panel (http://mouse.cs.ucla.edu/mousehapmap/full.html, accessed on 28 September 2020) [147].
## 4.5. Gene Expression Array and Heat Maps
*For* gene expression correlations, we obtained inbred mouse hepatic transcript data from the repository GSE16780 UCLA Hybrid MDP Liver Affy HTM430A [24]. The mRNA levels in the repository were expressed as log2 transformed and were calculated from the Affimetrix chip with the robust multiarray average (RMA) method. To plot the heatmaps, we used Morpheus software (https://software.broadinstitute.org/morpheus, accessed on 15 February 2022).
## 4.6. Functional Impact of Genomic Variants
The functional impact of genomic variants was assessed using the Sorting Intolerant From Tolerant (SIFT) software (https://sift.bii.a-star.edu.sg/www/SIFT_dbSNP.html, accessed on 12 July 2022) [27].
## 4.7. Enrichment Analysis
We used gProfiler [35] with the default settings to perform the pathway enrichment analyses.
## 4.8. Identification of Transcription Factors
We consulted the GeneHancer (GH) database, a catalogue of genome-wide enhancer-to-gene and promoter-to-gene associations, through GeneCards® (https://www.genecards.org/Guide/GeneCard, accessed on 6 September 2022) [148]. Only transcription factors with a significative GH Score were considered.
## 4.9. Statistics
We used Student’s t-test, ANOVA with Bonferroni correction, and Pearson correlation. All tests were two-tailed. The significance was considered to be $p \leq 0.05.$ We used an R package [146] and Prism v9.1.0 (GraphPad software, San Diego, CA, USA) for these analyses.
## References
1. Mosen P., Sanner A., Singh J., Winter D.. **Targeted Quantification of the Lysosomal Proteome in Complex Samples**. *Proteomes* (2021) **9**. DOI: 10.3390/proteomes9010004
2. Brubaker P.L., Martchenko A.. **Metabolic Homeostasis: It’s All in the Timing**. *Endocrinology* (2021) **163** bqab199. DOI: 10.1210/endocr/bqab199
3. Settembre C., Fraldi A., Medina D.L., Ballabio A.. **Signals from the lysosome: A control centre for cellular clearance and energy metabolism**. *Nat. Rev. Mol. Cell Biol.* (2013) **14** 283-296. DOI: 10.1038/nrm3565
4. Ballabio A., Bonifacino J.S.. **Lysosomes as dynamic regulators of cell and organismal homeostasis**. *Nat. Rev. Mol. Cell Biol.* (2019) **21** 101-118. DOI: 10.1038/s41580-019-0185-4
5. Platt F.M., D’Azzo A., Davidson B., Neufeld E.F., Tifft C.J.. **Lysosomal storage diseases**. *Nat. Rev. Dis. Prim.* (2018) **4** 27. DOI: 10.1038/s41572-018-0025-4
6. Sidransky E., Nalls M., Aasly J., Aharon-Peretz J., Annesi G., Barbosa E., Bar-Shira A., Berg D., Bras J., Brice A.. **Multicenter Analysis of Glucocerebrosidase Mutations in Parkinson's Disease**. *N. Engl. J. Med.* (2009) **361** 1651-1661. DOI: 10.1056/NEJMoa0901281
7. Klein A.D., Mazzulli J.R.. **Is Parkinson’s disease a lysosomal disorder?**. *Brain* (2018) **141** 2255-2262. DOI: 10.1093/brain/awy147
8. Platt F.M.. **Sphingolipid lysosomal storage disorders**. *Nature* (2014) **510** 68-75. DOI: 10.1038/nature13476
9. Jeyakumar M., Butters T.D., Cortina-Borja M., Hunnam V., Proia R.L., Perry V.H., Dwek R.A., Platt F.M.. **Delayed symptom onset and increased life expectancy in Sandhoff disease mice treated with**. *Proc. Natl. Acad. Sci. USA* (1999) **96** 6388-6393. DOI: 10.1073/pnas.96.11.6388
10. Peterschmitt M.J., Crawford N.P.S., Gaemers S.J.M., Ji A.J., Sharma J., Pham T.T.. **Pharmacokinetics, Pharmacodynamics, Safety, and Tolerability of Oral Venglustat in Healthy Volunteers**. *Clin. Pharmacol. Drug Dev.* (2020) **10** 86-98. DOI: 10.1002/cpdd.865
11. Kan S.-H., Aoyagi-Scharber M., Le S.Q., Vincelette J., Ohmi K., Bullens S., Wendt D.J., Christianson T.M., Tiger P.M.N., Brown J.R.. **Delivery of an enzyme-IGFII fusion protein to the mouse brain is therapeutic for mucopolysaccharidosis type IIIB**. *Proc. Natl. Acad. Sci. USA* (2014) **111** 14870-14875. DOI: 10.1073/pnas.1416660111
12. Weesner J.A., Annunziata I., Yang T., Acosta W., Gomero E., Hu H., van de Vlekkert D., Ayala J., Qiu X., Fremuth L.E.. **Preclinical Enzyme Replacement Therapy with a Recombinant β-Galactosidase-Lectin Fusion for CNS Delivery and Treatment of GM1-Gangliosidosis**. *Cells* (2022) **11**. DOI: 10.3390/cells11162579
13. Sevin C., Deiva K.. **Clinical Trials for Gene Therapy in Lysosomal Diseases With CNS Involvement**. *Front. Mol. Biosci.* (2021) **8** 624988. DOI: 10.3389/fmolb.2021.624988
14. Hobbs J.. **reversal of clinical features of hurler’s disease and biochemical improvement after treatment by bone-marrow transplantation**. *Lancet* (1981) **318** 709-712. DOI: 10.1016/S0140-6736(81)91046-1
15. Dever D.P., Scharenberg S.G., Camarena J., Kildebeck E.J., Clark J.T., Martin R.M., Bak R.O., Tang Y., Dohse M., Birgmeier J.A.. **CRISPR/Cas9 Genome Engineering in Engraftable Human Brain-Derived Neural Stem Cells**. *Iscience* (2019) **15** 524-535. DOI: 10.1016/j.isci.2019.04.036
16. Kirkegaard T., Gray J., Priestman D.A., Wallom K.-L., Atkins J., Olsen O.D., Klein A., Drndarski S., Petersen N.H.T., Ingemann L.. **Heat shock protein–based therapy as a potential candidate for treating the sphingolipidoses**. *Sci. Transl. Med.* (2016) **8** 355ra118. DOI: 10.1126/scitranslmed.aad9823
17. Wraith J.E.. **Limitations of enzyme replacement therapy: Current and future**. *J. Inherit. Metab. Dis.* (2006) **29** 442-447. DOI: 10.1007/s10545-006-0239-6
18. Broomfield A., Jones S., Hughes S.M., Bigger B.. **The impact of the immune system on the safety and efficiency of enzyme replacement therapy in lysosomal storage disorders**. *J. Inherit. Metab. Dis.* (2016) **39** 499-512. DOI: 10.1007/s10545-016-9917-1
19. Richardson J.S., Kemper A.R., Grosse S.D., Lam W.K.K., Rose A.M., Ahmad A., Ms A.G., Prosser L.A.. **Health and economic outcomes of newborn screening for infantile-onset Pompe disease**. *Anesthesia Analg.* (2020) **23** 758-766. DOI: 10.1038/s41436-020-01038-0
20. Durán A., Rebolledo-Jaramillo B., Olguin V., Rojas-Herrera M., Heras M.L., Calderón J.F., Zanlungo S., Priestman D.A., Platt F.M., Klein A.D.. **Identification of genetic modifiers of murine hepatic β-glucocerebrosidase activity**. *Biochem. Biophys. Rep.* (2021) **28** 101105. DOI: 10.1016/j.bbrep.2021.101105
21. Ferland R.J., Smith J., Papandrea D., Gracias J., Hains L., Kadiyala S.B., O’Brien B., Kang E.Y., Beyer B.S., Herron B.J.. **Multidimensional Genetic Analysis of Repeated Seizures in the Hybrid Mouse Diversity Panel Reveals a Novel Epileptogenesis Susceptibility Locus**. *G3 Genes Genomes Genet.* (2017) **7** 2545-2558. DOI: 10.1534/g3.117.042234
22. Klein A.D., Ferreira N.-S., Ben-Dor S., Duan J., Hardy J., Cox T.M., Merril A.H., Futerman A.H.. **Identification of Modifier Genes in a Mouse Model of Gaucher Disease**. *Cell Rep.* (2016) **16** 2546-2553. DOI: 10.1016/j.celrep.2016.07.085
23. Ghazalpour A., Rau C.D., Farber C.R., Bennett B.J., Orozco L.D., Van Nas A., Pan C., Allayee H., Beaven S.W., Civelek M.. **Hybrid mouse diversity panel: A panel of inbred mouse strains suitable for analysis of complex genetic traits**. *Mamm. Genome* (2012) **23** 680-692. DOI: 10.1007/s00335-012-9411-5
24. Bennett B.J., Farber C.R., Orozco L., Kang H.M., Ghazalpour A., Siemers N., Neubauer M., Neuhaus I., Yordanova R., Guan B.. **A high-resolution association mapping panel for the dissection of complex traits in mice**. *Genome Res.* (2010) **20** 281-290. DOI: 10.1101/gr.099234.109
25. Purcell S., Neale B., Todd-Brown K., Thomas L., Ferreira M.A.R., Bender D., Maller J., Sklar P., de Bakker P.I.W., Daly M.J.. **PLINK: A Tool Set for Whole-Genome Association and Population-Based Linkage Analyses**. *Am. J. Hum. Genet.* (2007) **81** 559-575. DOI: 10.1086/519795
26. Kang H.M., Zaitlen N.A., Wade C.M., Kirby A., Heckerman D., Daly M.J., Eskin E.. **Efficient Control of Population Structure in Model Organism Association Mapping**. *Genetics* (2008) **178** 1709-1723. DOI: 10.1534/genetics.107.080101
27. Vaser R., Adusumalli S., Leng S.N., Sikic M., Ng P.C.. **SIFT missense predictions for genomes**. *Nat. Protoc.* (2015) **11** 1-9. DOI: 10.1038/nprot.2015.123
28. Keane T.M., Goodstadt L., Danecek P., White M.A., Wong K., Yalcin B., Heger A., Agam A., Slater G., Goodson M.. **Mouse genomic variation and its effect on phenotypes and gene regulation**. *Nature* (2011) **477** 289-294. DOI: 10.1038/nature10413
29. An J., Shi J., He Q., Lui K., Liu Y., Huang Y., Sheikh M.S.. **CHCM1/CHCHD6, Novel Mitochondrial Protein Linked to Regulation of Mitofilin and Mitochondrial Cristae Morphology**. *J. Biol. Chem.* (2012) **287** 7411-7426. DOI: 10.1074/jbc.M111.277103
30. Verde I., Pahlke G., Salanova M., Zhang G., Wang S., Coletti D., Onuffer J., Jin S.-L.C., Conti M.. **Myomegalin Is a Novel Protein of the Golgi/Centrosome That Interacts with a Cyclic Nucleotide Phosphodiesterase**. *J. Biol. Chem.* (2001) **276** 11189-11198. DOI: 10.1074/jbc.M006546200
31. Marks M., Pennimpede T., Lange L., Grote P., Herrmann B.G., Wittler L.. **Analysis of the Fam181 gene family during mouse development reveals distinct strain-specific expression patterns, suggesting a role in nervous system development and function**. *Gene* (2015) **575** 438-451. DOI: 10.1016/j.gene.2015.09.035
32. Li H., Rauch T., Chen Z.-X., Szabó P.E., Riggs A.D., Pfeifer G.P.. **The Histone Methyltransferase SETDB1 and the DNA Methyltransferase DNMT3A Interact Directly and Localize to Promoters Silenced in Cancer Cells**. *J. Biol. Chem.* (2006) **281** 19489-19500. DOI: 10.1074/jbc.M513249200
33. Beaubien F., Raja R., Kennedy T.E., Fournier A.E., Cloutier J.-F.. **Slitrk1 is localized to excitatory synapses and promotes their development**. *Sci. Rep.* (2016) **6** 27343. DOI: 10.1038/srep27343
34. Hannun Y.A., Obeid L.M.. **Sphingolipids and their metabolism in physiology and disease**. *Nat. Rev. Mol. Cell Biol.* (2017) **19** 175-191. DOI: 10.1038/nrm.2017.107
35. Raudvere U., Kolberg L., Kuzmin I., Arak T., Adler P., Peterson H., Vilo J.. **g: Profiler: A web server for functional enrichment analysis and conversions of gene lists (2019 update)**. *Nucleic Acids Res.* (2019) **47** W191-W198. DOI: 10.1093/nar/gkz369
36. Gleyzer N., Vercauteren K., Scarpulla R.C.. **Control of Mitochondrial Transcription Specificity Factors (TFB1M and TFB2M) by Nuclear Respiratory Factors (NRF-1 and NRF-2) and PGC-1 Family Coactivators**. *Mol. Cell. Biol.* (2005) **25** 1354-1366. DOI: 10.1128/MCB.25.4.1354-1366.2005
37. Lee W.-H., Higuchi H., Ikeda S., Macke E.L., Takimoto T., Pattnaik B.R., Liu C., Chu L.-F., Siepka S.M., Krentz K.J.. **Mouse Tmem135 mutation reveals a mechanism involving mitochondrial dynamics that leads to age-dependent retinal pathologies**. *eLife* (2016) **5** e19264. DOI: 10.7554/eLife.19264
38. Zhang D., Ma X., Sun W., Cui P., Lu Z.. **Down-regulated FSTL5 promotes cell proliferation and survival by affecting Wnt/β-catenin signaling in hepatocellular carcinoma**. *Int. J. Clin. Exp. Pathol.* (2015) **8** 3386-3394. PMID: 26045876
39. Alrefaei A.F., Münsterberg A.E., Wheeler G.N.. **FZD10 regulates cell proliferation and mediates Wnt1 induced neurogenesis in the developing spinal cord**. *PLoS ONE* (2020) **15**. DOI: 10.1371/journal.pone.0219721
40. Maeda M., Hasegawa H., Hyodo T., Ito S., Asano E., Yuang H., Funasaka K., Shimokata K., Hasegawa Y., Hamaguchi M.. **ARHGAP18, a GTPase-activating protein for RhoA, controls cell shape, spreading, and motility**. *Mol. Biol. Cell* (2011) **22** 3840-3852. DOI: 10.1091/mbc.e11-04-0364
41. Dunne E., Spring C.M., Reheman A., Jin W., Berndt M.C., Newman D.K., Newman P.J., Ni H., Kenny D.. **Cadherin 6 Has a Functional Role in Platelet Aggregation and Thrombus Formation**. *Arter. Thromb. Vasc. Biol.* (2012) **32** 1724-1731. DOI: 10.1161/ATVBAHA.112.250464
42. Frey L., Ziętara N., Łyszkiewicz M., Marquardt B., Mizoguchi Y., Linder M.I., Liu Y., Giesert F., Wurst W., Dahlhoff M.. **Mammalian VPS45 orchestrates trafficking through the endosomal system**. *Blood* (2021) **137** 1932-1944. DOI: 10.1182/blood.2020006871
43. Koyanagi M., Nakabayashi K., Fujimoto T., Gu N., Baba I., Takashima Y., Doi K., Harada H., Kato N., Sasazuki T.. **ZFAT expression in B and T lymphocytes and identification of ZFAT-regulated genes**. *Genomics* (2008) **91** 451-457. DOI: 10.1016/j.ygeno.2008.01.009
44. Jin K., Wang S., Zhang Y., Xia M., Mo Y., Li X., Li G., Zeng Z., Xiong W., He Y.. **Long non-coding RNA PVT1 interacts with MYC and its downstream molecules to synergistically promote tumorigenesis**. *Cell. Mol. Life Sci.* (2019) **76** 4275-4289. DOI: 10.1007/s00018-019-03222-1
45. Tang T., Yang Z.-Y., Wang D., Yang X.-Y., Wang J., Li L., Wen Q., Gao L., Bian X.-W., Yu S.-C.. **The role of lysosomes in cancer development and progression**. *Cell Biosci.* (2020) **10** 131. DOI: 10.1186/s13578-020-00489-x
46. Davidson S.M., Vander Heiden M.G.. **Critical Functions of the Lysosome in Cancer Biology**. *Annu. Rev. Pharmacol. Toxicol.* (2017) **57** 481-507. DOI: 10.1146/annurev-pharmtox-010715-103101
47. Chen J.-S., Su I.-J., Leu Y.-W., Young K.-C., Sun H.S.. **Expression of T-cell lymphoma invasion and metastasis 2 (TIAM2) promotes proliferation and invasion of liver cancer**. *Int. J. Cancer* (2011) **130** 1302-1313. DOI: 10.1002/ijc.26117
48. Terasaki H., Saitoh T., Shiokawa K., Katoh M.. **Frizzled-10, up-regulated in primary colorectal cancer, is a positive regulator of the WNT—Beta-catenin—TCF signaling pathway**. *Int. J. Mol. Med.* (2002) **9** 107-112. PMID: 11786918
49. Zhao Z., Li S., Li S., Wang J., Lin H., Fu W.. **High expression of oncogene cadherin-6 correlates with tumor progression and a poor prognosis in gastric cancer**. *Cancer Cell Int.* (2021) **21** 493. DOI: 10.1186/s12935-021-02071-y
50. Cui M., Chang Y., Fang Q.-G., Du W., Wu J.-F., Wang J.-H., Liu S.-T., Luo S.-X.. **Non-Coding RNA Pvt1 Promotes Cancer Stem Cell–Like Traits in Nasopharyngeal Cancer via Inhibiting miR-1207**. *Pathol. Oncol. Res.* (2018) **25** 1411-1422. DOI: 10.1007/s12253-018-0453-1
51. Elotan A., Efenckova M., Ebralten J., Ealttoa A., Edixson L., Williams R.W., van der Voet M.. **Neuroinformatic analyses of common and distinct genetic components associated with major neuropsychiatric disorders**. *Front. Neurosci.* (2014) **8** 331. DOI: 10.3389/fnins.2014.00331
52. Kim S., Cho H., Lee D., Webster M.J.. **Association between SNPs and gene expression in multiple regions of the human brain**. *Transl. Psychiatry* (2012) **2** e113. DOI: 10.1038/tp.2012.42
53. Hsu S.-D., Lin F.-M., Wu W.-Y., Liang C., Huang W.-C., Chan W.-L., Tsai W.-T., Chen G.-Z., Lee C.-J., Chiu C.-M.. **miRTarBase: A database curates experimentally validated microRNA–target interactions**. *Nucleic Acids Res.* (2010) **39** D163-D169. DOI: 10.1093/nar/gkq1107
54. Chu C., Chen J., Chuang P., Su C., Chan Y.-L., Yang Y., Chiang Y., Su Y., Gean P., Sun H.S.. **TIAM2S as a novel regulator for serotonin level enhances brain plasticity and locomotion behavior**. *FASEB J.* (2020) **34** 3267-3288. DOI: 10.1096/fj.201901323R
55. Bykhovskaya Y., Mengesha E., Wang D., Yang H., Estivill X., Shohat M., Fischel-Ghodsian N.. **Human mitochondrial transcription factor B1 as a modifier gene for hearing loss associated with the mitochondrial A1555G mutation**. *Mol. Genet. Metab.* (2004) **82** 27-32. DOI: 10.1016/j.ymgme.2004.01.020
56. Fulzele K., Riddle R.C., DiGirolamo D.J., Cao X., Wan C., Chen D., Faugere M.-C., Aja S., Hussain M.A., Brüning J.C.. **Insulin Receptor Signaling in Osteoblasts Regulates Postnatal Bone Acquisition and Body Composition**. *Cell* (2022) **185** 746. DOI: 10.1016/j.cell.2022.01.016
57. Dearth R.K., Cui X., Kim H.-J., Kuiatse I., Lawrence N.A., Zhang X., Divisova J., Britton O.L., Mohsin S., Allred D.C.. **Mammary Tumorigenesis and Metastasis Caused by Overexpression of Insulin Receptor Substrate 1 (IRS-1) or IRS-2**. *Mol. Cell. Biol.* (2006) **26** 9302-9314. DOI: 10.1128/MCB.00260-06
58. Ochiai T., Sano T., Nagayama T., Kubota N., Kadowaki T., Wakabayashi T., Iwatsubo T.. **Differential involvement of insulin receptor substrate (IRS)-1 and IRS-2 in brain insulin signaling is associated with the effects on amyloid pathology in a mouse model of Alzheimer's disease**. *Neurobiol. Dis.* (2021) **159** 105510. DOI: 10.1016/j.nbd.2021.105510
59. Xu L., Wu J., Yu Y., Li H., Sun S., Zhang T., Wang M.. **Dok5 regulates proliferation and differentiation of osteoblast via ca-nonical Wnt/β-catenin signaling**. *J. Musculoskelet. Neuronal Interact.* (2022) **22** 113-122. PMID: 35234166
60. Landowski M., Bhute V.J., Takimoto T., Grindel S., Shahi P.K., Pattnaik B.R., Ikeda S., Ikeda A.. **A mutation in transmembrane protein 135 impairs lipid metabolism in mouse eyecups**. *Sci. Rep.* (2022) **12** 756. DOI: 10.1038/s41598-021-04644-3
61. Suzuki N., Fukushi M., Kosaki K., Doyle A.D., De Vega S., Yoshizaki K., Akazawa C., Arikawa-Hirasawa E., Yamada Y.. **Teneurin-4 Is a Novel Regulator of Oligodendrocyte Differentiation and Myelination of Small-Diameter Axons in the CNS**. *J. Neurosci.* (2012) **32** 11586-11599. DOI: 10.1523/JNEUROSCI.2045-11.2012
62. Pu J.-L., Gao T., Si X.-L., Zheng R., Jin C.-Y., Ruan Y., Fang Y., Chen Y., Song Z., Yin X.-Z.. **Parkinson’s Disease in Teneurin Transmembrane Protein 4 (TENM4) Mutation Carriers**. *Front. Genet.* (2020) **11** 598064. DOI: 10.3389/fgene.2020.598064
63. Hor H., Francescatto L., Bartesaghi L., Ortega-Cubero S., Kousi M., Lorenzo-Betancor O., Jiménez-Jiménez F.J., Gironell A., Clarimón J., Drechsel O.. **Missense mutations inTENM4, a regulator of axon guidance and central myelination, cause essential tremor**. *Hum. Mol. Genet.* (2015) **24** 5677-5686. DOI: 10.1093/hmg/ddv281
64. Kant T., Newe M., Winter L., Hoffmann M., Kämmerer S., Klapproth E., Künzel K., Kühnel M., Neubert L., El-Armouche A.. **Genetic Deletion of Polo-Like Kinase 2 Induces a Pro-Fibrotic Pulmonary Phenotype**. *Cells* (2021) **10**. DOI: 10.3390/cells10030617
65. Inglis K.J., Chereau D., Brigham E.F., Chiou S.-S., Schöbel S., Frigon N.L., Yu M., Caccavello R.J., Nelson S., Motter R.. **Polo-like Kinase 2 (PLK2) Phosphorylates α-Synuclein at Serine 129 in Central Nervous System**. *J. Biol. Chem.* (2009) **284** 2598-2602. DOI: 10.1074/jbc.C800206200
66. Dong J., Hu Z., Wu C., Guo H., Zhou B., Lv J., Lu D., Chen K., Shi Y., Chu M.. **Association analyses identify multiple new lung cancer susceptibility loci and their interactions with smoking in the Chinese population**. *Nat. Genet.* (2012) **44** 895-899. DOI: 10.1038/ng.2351
67. Manning G., Whyte D.B., Martinez R., Hunter T., Sudarsanam S.. **The Protein Kinase Complement of the Human Genome**. *Science* (2002) **298** 1912-1934. DOI: 10.1126/science.1075762
68. Manivannan J., Tay S., Ling E.-A., Dheen S.. **Dihydropyrimidinase-like 3 regulates the inflammatory response of activated microglia**. *Neuroscience* (2013) **253** 40-54. DOI: 10.1016/j.neuroscience.2013.08.023
69. Kanda M., Nomoto S., Oya H., Shimizu D., Takami H., Hibino S., Hashimoto R., Kobayashi D., Tanaka C., Yamada S.. **Dihydropyrimidinase-like 3 facilitates malignant behavior of gastric cancer**. *J. Exp. Clin. Cancer Res.* (2014) **33** 66. DOI: 10.1186/s13046-014-0066-9
70. Blasco H., Bernard-Marissal N., Vourc'H P., Guettard Y.O., Sunyach C., Augereau O., Khederchah J., Mouzat K., Antar C., Gordon P.H.. **A Rare Motor Neuron Deleterious Missense Mutation in the**. *Hum. Mutat.* (2013) **34** 953-960. DOI: 10.1002/humu.22329
71. Damoulakis G., Gambardella L., Rossman K.L., Lawson C., Anderson K., Fukui Y., Welch H.C., Der C., Stephens L., Hawkins P.T.. **P-Rex1 directly activates RhoG to regulate GPCR-driven Rac signalling and actin polarity in neutrophils**. *J. Cell Sci.* (2014) **127** 2589-2600. DOI: 10.1242/jcs.153049
72. Qin J., Xie Y., Wang B., Hoshino M., Wolff D.W., Zhao J., Scofield M.A., Dowd F.J., Lin M.-F., Tu Y.. **Upregulation of PIP3-dependent Rac exchanger 1 (P-Rex1) promotes prostate cancer metastasis**. *Oncogene* (2009) **28** 1853-1863. DOI: 10.1038/onc.2009.30
73. Zhang D.-Y., Lei J.-S., Sun W.-L., Wang D.-D., Lu Z.. **Follistatin Like 5 (FSTL5) inhibits epithelial to mesenchymal transition in hepatocellular carcinoma**. *Chin. Med. J.* (2020) **133** 1798-1804. DOI: 10.1097/CM9.0000000000000847
74. Vilboux T., Lev A., Malicdan M.C.V., Simon A.J., Järvinen P., Racek T., Puchalka J., Sood R., Carrington B., Bishop K.. **A Congenital Neutrophil Defect Syndrome Associated with Mutations in**. *N. Engl. J. Med.* (2013) **369** 54-65. DOI: 10.1056/NEJMoa1301296
75. Nayak S.R., Harrington E., Boone D., Hartmaier R., Chen J., Pathiraja T.N., Cooper K.L., Fine J.L., Sanfilippo J., Davidson N.E.. **A Role for Histone H2B Variants in Endocrine-Resistant Breast Cancer**. *Horm. Cancer* (2015) **6** 214-224. DOI: 10.1007/s12672-015-0230-5
76. Shan Z., Shakoori A., Bodaghi S., Goldsmith P., Jin J., Wiest J.S.. **TUSC1, a Putative Tumor Suppressor Gene, Reduces Tumor Cell Growth In Vitro and Tumor Growth In Vivo**. *PLoS ONE* (2013) **8**. DOI: 10.1371/journal.pone.0066114
77. Zhang R., Yu W., Liang G., Jia Z., Chen Z., Zhao L., Yuan Y., Zhou X., Li D., Shen S.. **Tumor Suppressor Candidate 1 Suppresses Cell Growth and Predicts Better Survival in Glioblastoma**. *Cell. Mol. Neurobiol.* (2016) **37** 37-42. DOI: 10.1007/s10571-016-0339-2
78. Grigorenko A.P., Protasova M.S., Lisenkova A.A., Reshetov D.A., Andreeva T.V., Garcias G.D.L., Roth M.D.G.M., Papassotiropoulos A., Rogaev E.I.. **Neurodevelopmental Syndrome with Intellectual Disability, Speech Impairment, and Quadrupedia Is Associated with Glutamate Receptor Delta 2 Gene Defect**. *Cells* (2022) **11**. DOI: 10.3390/cells11030400
79. Matsushita K., Takeuchi O., Standley D.M., Kumagai Y., Kawagoe T., Miyake T., Satoh T., Kato H., Tsujimura T., Nakamura H.. **Zc3h12a is an RNase essential for controlling immune responses by regulating mRNA decay**. *Nature* (2009) **458** 1185-1190. DOI: 10.1038/nature07924
80. Tsoi L.C., Spain S.L., Knight J., Ellinghaus E., Stuart P.E., Capon F., Ding J., Li Y., Tejasvi T., Gudjonsson J.E.. **Identification of 15 new psoriasis susceptibility loci highlights the role of innate immunity**. *Nat. Genet.* (2012) **44** 1341-1348. DOI: 10.1038/ng.2467
81. Guo W., Cai Y., Zhang H., Yang Y., Yang G., Wang X., Zhao J., Lin J., Zhu J., Li W.. **Association of ARHGAP18 polymorphisms with schizophrenia in the Chinese-Han population**. *PLoS ONE* (2017) **12**. DOI: 10.1371/journal.pone.0175209
82. Negredo P.N., Edgar J.R., Manna P.T., Antrobus R., Robinson M.S.. **The WDR11 complex facilitates the tethering of AP-1-derived vesicles**. *Nat. Commun.* (2018) **9** 596. DOI: 10.1038/s41467-018-02919-4
83. Ouyang W.-W., Li Q.-Y., Yang W.-G., Su S.-F., Wu L.-J., Yang Y., Lu B.. **Genetic characteristics of a patient with multiple primary cancers: A case report**. *World J. Clin. Cases* (2021) **9** 8563-8570. DOI: 10.12998/wjcc.v9.i28.8563
84. Abdul-Rahman P.S., Lim B.-K., Hashim O.H.. **Expression of high-abundance proteins in sera of patients with endometrial and cervical cancers: Analysis using 2-DE with silver staining and lectin detection methods**. *Electrophoresis* (2007) **28** 1989-1996. DOI: 10.1002/elps.200600629
85. Sun D., Zhao Y.Y., Dai S.P., Fang K., Dong L.Y., Ding Q.. **Cloning and analysis of human alpha-1B glycoprotein precursor gene: A novel member of human immunoglobulin superfamily**. *Yi Chuan Xue Bao* (2002) **29** 299-302. PMID: 11985261
86. Kreunin P., Zhao J., Rosser C., Urquidi V., Lubman D.M., Goodison S.. **Bladder Cancer Associated Glycoprotein Signatures Revealed by Urinary Proteomic Profiling**. *J. Proteome Res.* (2007) **6** 2631-2639. DOI: 10.1021/pr0700807
87. Tang J., Li Y., Sang Y., Yu B., Lv D., Zhang W., Feng H.. **LncRNA PVT1 regulates triple-negative breast cancer through KLF5/beta-catenin signaling**. *Oncogene* (2018) **37** 4723-4734. DOI: 10.1038/s41388-018-0310-4
88. Glasson S.S., Askew R., Sheppard B., Carito B., Blanchet T., Ma H.-L., Flannery C.R., Peluso D., Kanki K., Yang Z.. **Deletion of active ADAMTS5 prevents cartilage degradation in a murine model of osteoarthritis**. *Nature* (2005) **434** 644-648. DOI: 10.1038/nature03369
89. Agajanian M.J., Potjewyd F.M., Bowman B.M., Solomon S., LaPak K.M., Bhatt D.P., Smith J.L., Goldfarb D., Axtman A.D., Major M.B.. **Protein proximity networks and functional evaluation of the casein kinase 1 gamma family reveal unique roles for CK1γ3 in WNT signaling**. *J. Biol. Chem.* (2022) **298** 101986. DOI: 10.1016/j.jbc.2022.101986
90. Schittek B., Sinnberg T.. **Biological functions of casein kinase 1 isoforms and putative roles in tumorigenesis**. *Mol. Cancer* (2014) **13** 231. DOI: 10.1186/1476-4598-13-231
91. Inoue N., Watanabe M., Yamada H., Takemura K., Hayashi F., Yamakawa N., Akahane M., Shimizuishi Y., Hidaka Y., Iwatani Y.. **Associations between autoimmune thyroid disease prognosis and functional polymorphisms of susceptibility genes, CTLA4, PTPN22, CD40, FCRL3, and ZFAT, previously revealed in genome-wide association studies**. *J. Clin. Immunol.* (2012) **32** 1243-1252. DOI: 10.1007/s10875-012-9721-0
92. Wallom K.-L., Fernández-Suárez M.E., Priestman D.A., Vruchte D.T., Huebecker M., Hallett P.J., Isacson O., Platt F.M.. **Glycosphingolipid metabolism and its role in ageing and Parkinson’s disease**. *Glycoconj. J.* (2021) **39** 39-53. DOI: 10.1007/s10719-021-10023-x
93. Ryckman A.E., Brockhausen I., Walia J.S.. **Metabolism of Glycosphingolipids and Their Role in the Pathophysiology of Lysosomal Storage Disorders**. *Int. J. Mol. Sci.* (2020) **21**. DOI: 10.3390/ijms21186881
94. Sardiello M., Palmieri M., Di Ronza A., Medina D.L., Valenza M., Gennarino V.A., Di Malta C., Donaudy F., Embrione V., Polishchuk R.S.. **A Gene Network Regulating Lysosomal Biogenesis and Function**. *Science* (2009) **325** 473-477. DOI: 10.1126/science.1174447
95. Olguín V., Durán A., Heras M.L., Rubilar J.C., Cubillos F.A., Olguín P., Klein A.D.. **Genetic Background Matters: Population-Based Studies in Model Organisms for Translational Research**. *Int. J. Mol. Sci.* (2022) **23**. DOI: 10.3390/ijms23147570
96. Rha A.K., Maguire A.S., Martin D.R.. **GM1 Gangliosidosis: Mechanisms and Management**. *Appl. Clin. Genet.* (2021) **14** 209-233. DOI: 10.2147/TACG.S206076
97. Adibekian A., Martin B., Chang J., Hsu K.-L., Tsuboi K., Bachovchin D., Speers A., Brown S., Spicer T., Fernandez-Vega V.. **Characterization of a Selective, Reversible Inhibitor of Lysophospholipase 1 (LYPLA1)**. *Probe Reports from the NIH Molecular Libraries Program [Internet]* (2010)
98. Wu S., Cao R., Tao B., Wu P., Peng C., Gao H., Liang J., Yang W.. **Pyruvate Facilitates FACT-Mediated**. *Adv. Sci.* (2022) **9** 2104055. DOI: 10.1002/advs.202104055
99. Merrill A.H.. **Sphingolipid and Glycosphingolipid Metabolic Pathways in the Era of Sphingolipidomics**. *Chem. Rev.* (2011) **111** 6387-6422. DOI: 10.1021/cr2002917
100. Lachmann R.H., Vruchte D.T., Lloyd-Evans E., Reinkensmeier G., Sillence D., Fernandez-Guillen L., Dwek R.A., Butters T.D., Cox T.M., Platt F.M.. **Treatment with miglustat reverses the lipid-trafficking defect in Niemann–Pick disease type C**. *Neurobiol. Dis.* (2004) **16** 654-658. DOI: 10.1016/j.nbd.2004.05.002
101. Elstein D., Hollak C., Aerts J.M.F.G., van Weely S., Maas M., Cox T.M., Lachmann R.H., Hrebicek M., Platt F.M., Butters T.D.. **Sustained therapeutic effects of oral miglustat (Zavesca,**. *J. Inherit. Metab. Dis.* (2004) **27** 757-766. DOI: 10.1023/B:BOLI.0000045756.54006.17
102. Meng Q., Hu X., Zhao X., Kong X., Meng Y.-M., Chen Y., Su L., Jiang X., Qiu X., Huang C.. **A circular network of coregulated sphingolipids dictates lung cancer growth and progression**. *Ebiomedicine* (2021) **66** 103301. DOI: 10.1016/j.ebiom.2021.103301
103. Pazoki R., Vujkovic M., Elliott J., Evangelou E., Gill D., Ghanbari M., van der Most P.J., Pinto R.C., Wielscher M., Farlik M.. **Genetic analysis in European ancestry individuals identifies 517 loci associated with liver enzymes**. *Nat. Commun.* (2021) **12** 2579. DOI: 10.1038/s41467-021-22338-2
104. Willer C.J., Schmidt E.M., Sengupta S., Peloso G.M., Gustafsson S., Kanoni S., Ganna A., Chen J., Buchkovich M.L., Mora S.. **Discovery and refinement of loci associated with lipid levels**. *Nat. Genet.* (2013) **45** 1274-1283. DOI: 10.1038/ng.2797
105. Lee J.-H., Yang D.-S., Goulbourne C.N., Im E., Stavrides P., Pensalfini A., Chan H., Bouchet-Marquis C., Bleiwas C., Berg M.J.. **Faulty autolysosome acidification in Alzheimer’s disease mouse models induces autophagic build-up of Aβ in neurons, yielding senile plaques**. *Nat. Neurosci.* (2022) **25** 688-701. DOI: 10.1038/s41593-022-01084-8
106. Stahl-Meyer J., Stahl-Meyer K., Jäättelä M.. **Control of mitosis, inflammation, and cell motility by limited leakage of lysosomes**. *Curr. Opin. Cell Biol.* (2021) **71** 29-37. DOI: 10.1016/j.ceb.2021.02.003
107. Tan J.X., Finkel T.. **A phosphoinositide signalling pathway mediates rapid lysosomal repair**. *Nature* (2022) **609** 815-821. DOI: 10.1038/s41586-022-05164-4
108. Alessenko A.V., Albi E.. **Exploring Sphingolipid Implications in Neurodegeneration**. *Front. Neurol.* (2020) **11** 437. DOI: 10.3389/fneur.2020.00437
109. Cumin C., Huang Y.-L., Rossdam C., Ruoff F., Céspedes S.P., Liang C.-Y., Lombardo F.C., Coelho R., Rimmer N., Konantz M.. **Glycosphingolipids are mediators of cancer plasticity through independent signaling pathways**. *Cell Rep.* (2022) **40** 111181. DOI: 10.1016/j.celrep.2022.111181
110. Snaebjornsson M.T., Janaki-Raman S., Schulze A.. **Greasing the Wheels of the Cancer Machine: The Role of Lipid Metabolism in Cancer**. *Cell Metab.* (2019) **31** 62-76. DOI: 10.1016/j.cmet.2019.11.010
111. Petersen N.H., Olsen O.D., Groth-Pedersen L., Ellegaard A.-M., Bilgin M., Redmer S., Ostenfeld M.S., Ulanet D., Dovmark T.H., Lønborg A.. **Transformation-Associated Changes in Sphingolipid Metabolism Sensitize Cells to Lysosomal Cell Death Induced by Inhibitors of Acid Sphingomyelinase**. *Cancer Cell* (2013) **24** 379-393. DOI: 10.1016/j.ccr.2013.08.003
112. Fatehi-Hassanabad Z., Chan C.B.. **Transcriptional regulation of lipid metabolism by fatty acids: A key determinant of pancreatic β-cell function**. *Nutr. Metab.* (2005) **2** 1. DOI: 10.1186/1743-7075-2-1
113. Liu F., Zhu X., Jiang X., Li S., Lv Y.. **Transcriptional control by HNF-1: Emerging evidence showing its role in lipid metabolism and lipid metabolism disorders**. *Genes Dis.* (2021) **9** 1248-1257. DOI: 10.1016/j.gendis.2021.06.010
114. Pan G., Diamanti K., Cavalli M., Gutiérrez A.L., Komorowski J., Wadelius C.. **Multifaceted regulation of hepatic lipid metabolism by YY1**. *Life Sci. Alliance* (2021) **4** e202000928. DOI: 10.26508/lsa.202000928
115. Magee N., Zhang Y.. **Role of early growth response 1 in liver metabolism and liver cancer**. *Hepatoma Res.* (2017) **3** 268-277. DOI: 10.20517/2394-5079.2017.36
116. Banerjee A., Mahata B., Dhir A., Mandal T.K., Biswas K.. **Elevated histone H3 acetylation and loss of the Sp1–HDAC1 complex de-repress the GM2-synthase gene in renal cell carcinoma**. *J. Biol. Chem.* (2019) **294** 1005-1018. DOI: 10.1074/jbc.RA118.004485
117. Martina J.A., Diab H.I., Lishu L., Jeong-A L., Patange S., Raben N., Puertollano R.. **The Nutrient-Responsive Transcription Factor TFE3 Promotes Autophagy, Lysosomal Biogenesis, and Clearance of Cellular Debris**. *Sci. Signal.* (2014) **7** ra9. DOI: 10.1126/scisignal.2004754
118. Di Malta C., Siciliano D., Calcagni A., Monfregola J., Punzi S., Pastore N., Eastes A.N., Davis O., De Cegli R., Zampelli A.. **Transcriptional activation of RagD GTPase controls mTORC1 and promotes cancer growth**. *Science* (2017) **356** 1188-1192. DOI: 10.1126/science.aag2553
119. Polager S., Ofir M., Ginsberg D.. **E2F1 regulates autophagy and the transcription of autophagy genes**. *Oncogene* (2008) **27** 4860-4864. DOI: 10.1038/onc.2008.117
120. Seok S., Fu T., Choi S.-E., Li Y., Zhu R., Kumar S., Sun X., Yoon G., Kang Y., Zhong W.. **Transcriptional regulation of autophagy by an FXR–CREB axis**. *Nature* (2014) **516** 108-111. DOI: 10.1038/nature13949
121. Annunziata I., van de Vlekkert D., Wolf E., Finkelstein D., Neale G., Machado E., Mosca R., Campos Y., Tillman H., Roussel M.F.. **MYC competes with MiT/TFE in regulating lysosomal biogenesis and autophagy through an epigenetic rheostat**. *Nat. Commun.* (2019) **10** 3623. DOI: 10.1038/s41467-019-11568-0
122. Song N., Song R., Ma P.. **MiR-340-5p alleviates neuroinflammation and neuronal injury via suppressing STING in subarachnoid hemorrhage**. *Brain Behav.* (2022) **12** e2687. DOI: 10.1002/brb3.2687
123. Guo Y., Gao Y., Liu S.. **lncRNA XIST is associated with preeclampsia and mediates trophoblast cell invasion via miR-340-5p/KCNJ16 signaling pathway**. *Transpl. Immunol.* (2022) **74** 101666. DOI: 10.1016/j.trim.2022.101666
124. Ou D., Wu Y., Zhang J., Liu J., Liu Z., Shao M., Guo X., Cui S.. **miR-340-5p affects oral squamous cell carcinoma (OSCC) cells proliferation and invasion by targeting endoplasmic reticulum stress proteins**. *Eur. J. Pharmacol.* (2022) **920** 174820. DOI: 10.1016/j.ejphar.2022.174820
125. Huang Z., Xu Y., Wan M., Zeng X., Wu J.. **miR-340: A multifunctional role in human malignant diseases**. *Int. J. Biol. Sci.* (2021) **17** 236-246. DOI: 10.7150/ijbs.51123
126. Liu Y., Li X., Zhang Y., Wang H., Rong X., Peng J., He L., Peng Y.. **An miR-340-5p-macrophage feedback loop modulates the progression and tumor microenvironment of glioblastoma multiforme**. *Oncogene* (2019) **38** 7399-7415. DOI: 10.1038/s41388-019-0952-x
127. Liu H., Li B., Qiao L., Liu J., Ren D., Liu W.. **miR-340-5p inhibits sheep adipocyte differentiation by targeting**. *Anim. Sci. J.* (2020) **91** e13462. DOI: 10.1111/asj.13462
128. Zhu Y., Yang X., Zhou J., Chen L., Zuo P., Chen L., Jiang L., Li T., Wang D., Xu Y.. **miR-340-5p Mediates Cardiomyocyte Oxidative Stress in Diabetes-Induced Cardiac Dysfunction by Targeting Mcl-1**. *Oxidative Med. Cell. Longev.* (2022) **2022** 3182931. DOI: 10.1155/2022/3182931
129. Priestman D.A., Van Der Spoel A.C., Butters T.D., Dwek R.A., Platt F.M.. **N-butyldeoxynojirimycin causes weight loss as a result of appetite suppression in lean and obese mice**. *Diabetes Obes. Metab.* (2007) **10** 159-166. DOI: 10.1111/j.1463-1326.2006.00701.x
130. Chatterjee S., Zheng L., Ma S., Bedja D., Bandaru V.V.R., Kim G., Rangecroft A., Iocco D., Campbell S.A.. **Management of metabolic syndrome and reduction in body weight in type II diabetic mice by inhibiting glycosphingolipid synthesis**. *Biochem. Biophys. Res. Commun.* (2020) **525** 455-461. DOI: 10.1016/j.bbrc.2020.02.104
131. Robinson D.S.. **The fluorimetric determination of β-glucosidase: Its occurrence in the tissues of animals, including insects**. *Biochem. J.* (1956) **63** 39-44. DOI: 10.1042/bj0630039
132. Shigeto S., Katafuchi T., Okada Y., Nakamura K., Endo F., Okuyama T., Takeuchi H., Kroos M.A., Verheijen F.W., Reuser A.J.. **Improved assay for differential diagnosis between Pompe disease and acid α-glucosidase pseudodeficiency on dried blood spots**. *Mol. Genet. Metab.* (2011) **103** 12-17. DOI: 10.1016/j.ymgme.2011.01.006
133. Beutler E., Kuhl W.. **Purification and properties of human alpha-galactosidases**. *J. Biol. Chem.* (1972) **247** 7195-7200. DOI: 10.1016/S0021-9258(19)44613-9
134. Mayes J.S., Scheerer J.B., Sifers R.N., Donaldson M.L.. **Differential assay for lysosomal alpha-galactosidases in human tissues and its application to Fabry’s disease**. *Clin. Chim. Acta Int. J. Clin. Chem.* (1981) **112** 247-251. DOI: 10.1016/0009-8981(81)90384-3
135. Padeh B., Navon R.. **Diagnosis of Tay-Sachs disease by hexosaminidase activity in leukocytes and amniotic fluid cells**. *Isr. J. Med. Sci.* (1971) **7** 259-263. PMID: 5560980
136. Hauser E.C., Kasperzyk J.L., D'Azzo A., Seyfried T.N.. **Inheritance of Lysosomal Acid β-Galactosidase Activity and Gangliosides in Crosses of DBA/2J and Knockout Mice**. *Biochem. Genet.* (2004) **42** 241-257. DOI: 10.1023/B:BIGI.0000034429.55418.71
137. Potier M., Mameli L., Bélisle M., Dallaire L., Melançon S.. **Fluorometric assay of neuraminidase with a sodium (4-methylumbelliferyl-α-d-N-acetylneuraminate) substrate**. *Anal. Biochem.* (1979) **94** 287-296. DOI: 10.1016/0003-2697(79)90362-2
138. Seyrantepe V., Landry K., Trudel S., Hassan J.A., Morales C.R., Pshezhetsky A.V.. **Neu4, a Novel Human Lysosomal Lumen Sialidase, Confers Normal Phenotype to Sialidosis and Galactosialidosis Cells**. *J. Biol. Chem.* (2004) **279** 37021-37029. DOI: 10.1074/jbc.M404531200
139. Bussink A.P., Verhoek M., Vreede J., der Vlugt K.G.-V., Donker-Koopman W.E., Sprenger R.R., Hollak C.E., Aerts J.M.F.G., Boot R.G.. **Common G102S polymorphism in chitotriosidase differentially affects activity towards 4-methylumbelliferyl substrates**. *FEBS J.* (2009) **276** 5678-5688. DOI: 10.1111/j.1742-4658.2009.07259.x
140. Schoonhoven A., Rudensky B., Elstein D., Zimran A., Hollak C.E., Groener J.E., Aerts J.M.. **Monitoring of Gaucher patients with a novel chitotriosidase assay**. *Clin. Chim. Acta* (2007) **381** 136-139. DOI: 10.1016/j.cca.2007.02.042
141. Lau K.H., Onishi T., Wergedal J.E., Singer F.R., Baylink D.J.. **Characterization and assay of tartrate-resistant acid phosphatase activity in serum: Potential use to assess bone resorption**. *Clin. Chem.* (1987) **33** 458-462. DOI: 10.1093/clinchem/33.4.458
142. Goodlad G.A.J., Mills G.T.. **The acid phosphatases of rat liver**. *Biochem. J.* (1957) **66** 346-354. DOI: 10.1042/bj0660346
143. di Matteo G., Orfeo M.A., Romeo G.. **Human α-fucosidase: Purification and properties**. *Biochim. Biophys. Acta (BBA)-Enzym.* (1976) **429** 527-537. DOI: 10.1016/0005-2744(76)90300-4
144. Li C., Qian J., Lin J.-S.. **Purification and characterization of α-L-fucosidase from human primary hepatocarcinoma tissue**. *World J. Gastroenterol.* (2006) **12** 3770-3775. DOI: 10.3748/wjg.v12.i23.3770
145. Neville D.C., Coquard V., Priestman D.A., te Vruchte D.J.M., Sillence D.J., Dwek R.A., Platt F.M., Butters T.D.. **Analysis of fluorescently labeled glycosphingolipid-derived oligosaccharides following ceramide glycanase digestion and anthranilic acid labeling**. *Anal. Biochem.* (2004) **331** 275-282. DOI: 10.1016/j.ab.2004.03.051
146. 146.
R Core Team
A Language and Environment for Statistical ComputingR Foundation for Statistical ComputingVienna, Austria2020. *A Language and Environment for Statistical Computing* (2020)
147. Kirby A., Kang H.M., Wade C.M., Cotsapas C., Kostem E., Han B., Furlotte N., Kang E.Y., Rivas M., Bogue M.A.. **Fine Mapping in 94 Inbred Mouse Strains Using a High-Density Haplotype Resource**. *Genetics* (2010) **185** 1081-1095. DOI: 10.1534/genetics.110.115014
148. Rebhan M.. **GeneCards: Integrating information about genes, proteins and diseases**. *Trends Genet.* (1997) **13** 163. DOI: 10.1016/S0168-9525(97)01103-7
|
---
title: Targeting Cellular Retinoic Acid Binding Protein 1 with Retinoic Acid-like
Compounds to Mitigate Motor Neuron Degeneration
authors:
- Jennifer Nhieu
- Liming Milbauer
- Thomas Lerdall
- Fatimah Najjar
- Chin-Wen Wei
- Ryosuke Ishida
- Yue Ma
- Hiroyuki Kagechika
- Li-Na Wei
journal: International Journal of Molecular Sciences
year: 2023
pmcid: PMC10002585
doi: 10.3390/ijms24054980
license: CC BY 4.0
---
# Targeting Cellular Retinoic Acid Binding Protein 1 with Retinoic Acid-like Compounds to Mitigate Motor Neuron Degeneration
## Abstract
All-trans-retinoic Acid (atRA) is the principal active metabolite of Vitamin A, essential for various biological processes. The activities of atRA are mediated by nuclear RA receptors (RARs) to alter gene expression (canonical activities) or by cellular retinoic acid binding protein 1 (CRABP1) to rapidly (minutes) modulate cytosolic kinase signaling, including calcium calmodulin-activated kinase 2 (CaMKII) (non-canonical activities). Clinically, atRA-like compounds have been extensively studied for therapeutic applications; however, RAR-mediated toxicity severely hindered the progress. It is highly desirable to identify CRABP1-binding ligands that lack RAR activity. Studies of CRABP1 knockout (CKO) mice revealed CRABP1 to be a new therapeutic target, especially for motor neuron (MN) degenerative diseases where CaMKII signaling in MN is critical. This study reports a P19-MN differentiation system, enabling studies of CRABP1 ligands in various stages of MN differentiation, and identifies a new CRABP1-binding ligand C32. Using the P19-MN differentiation system, the study establishes C32 and previously reported C4 as CRABP1 ligands that can modulate CaMKII activation in the P19-MN differentiation process. Further, in committed MN cells, elevating CRABP1 reduces excitotoxicity-triggered MN death, supporting a protective role for CRABP1 signaling in MN survival. C32 and C4 CRABP1 ligands were also protective against excitotoxicity-triggered MN death. The results provide insight into the potential of signaling pathway-selective, CRABP1-binding, atRA-like ligands in mitigating MN degenerative diseases.
## 1. Introduction
All-trans-retinoic acid (atRA) is the principal active metabolite of vitamin A with well-known biological activities in development, differentiation, apoptosis, and many other biological processes [1]. These activities of atRA are known to be primarily mediated by nuclear RA receptors (RARs) that act as transcriptional factors to alter gene expression [1,2] and are referred to as atRA’s canonical activities. Recently, it has been shown that atRA also elicits RAR-independent activities that can be detected rapidly (within minutes) in the cytoplasm [2,3,4], referred to as non-canonical activities [3,4]. Using a gene knockout approach, it has been established that the non-canonical activities of atRA are mediated by a specific high-affinity cytosolic atRA-binding protein named cellular retinoic acid binding protein 1 (CRABP1) [5,6].
CRABP1 is a highly conserved cytosolic protein and is expressed in multiple cell types, including embryonic stem cells (ESCs) [5], cardiomyocytes [7], adipocytes [8,9], and motor neurons (MNs) [10], etc. In ESC, CRABP1-RA modulates cell cycle progression, and deleting CRABP1 from ESC accelerates cell cycle progression [5], supporting the notion that CRABP1 can be a tumor suppressor [11,12]. In cardiomyocytes, CRABP1 protects cardiomyocytes from apoptosis triggered by adrenergic over-stimulation; therefore, Crabp1 knockout (CKO) mice are prone to isoproterenol-induced cardiomyopathy and heart failure [7]. In adipocytes, CRABP1 facilitates adiponectin secretion and mitochondria homeostasis; therefore, CKO mice have significantly reduced adiponectin levels and are prone to high fat diet-induced adipose inflammation [9]. In MNs, CRABP1 protects against neuronal stress/death, and CKO mice spontaneously develop adult-onset progressive motor deterioration, mimicking amyotrophic lateral sclerosis (ALS) due to progressive MN death and neuromuscular junction defects [10]. While CRABP1-RA appears to affect various cell types and deleting CRABP1 causes different pathological outcomes in various organ systems, it is interesting that pathologies caused by CRABP1 deletion are related to two conserved signaling pathways, i.e., extracellular signal-regulated kinase (Erk) and calcium (Ca2+) calmodulin-activated kinase 2 (CaMKII).
In Erk kinase activation, CRABP1 directly interacts with rapidly accelerated fibrosarcoma 1 (Raf-1), which is the first kinase component in the mitogen-activated protein kinase (MAPK) signaling pathway, to ultimately dampen mitogen or growth factor-stimulated Erk activation. This is crucial to numerous cellular processes, especially growth [11]. CRABP1 modulation of Erk signaling plays out in the context of stem cell proliferation (in ESCs and tumors) [5,12], and for adiponectin secretion (in adipocytes) [9]. In CaMKII activation, CRABP1 directly interacts with CaMKII to dampen its enzyme activation, thereby preventing over-stimulation of CaMKII in excitable cells (such as cardiomyocytes and MNs) and protecting them from cytotoxicity and cell death [7,10]. This is supported by the finding that the physiological ligand of CRABP1, atRA, can be used to protect against isoproterenol-induced cardiomyopathy in wild-type mice but not in CKO mice, and suggests a therapeutic potential of atRA in specifically targeting CRABP1 to reduce CaMKII over-activation related pathologies [13]. However, given the well-known toxicity of atRA [14,15], via RAR activation, in long-term applications, it is desirable to explore atRA-like compounds that can specifically target CRABP1 without activating RARs in order to avoid retinoid toxicity.
To further validate that CRABP1 can be a useful therapeutic target in managing pathological conditions caused by CaMKII over-activation, we recently carefully examined the ALS-like motor deterioration phenotype of CKO mice and the mechanism of CRABP1 action in the motor system [10]. It appears that CRABP1 is specifically expressed in spinal MNs, and elevating its expression in MNs to dampen CaMKII activation is beneficial to the neuromuscular junction (NMJ) health, partially attributable to enhanced MN agrin expression and axon extension. Specifically, CaMKII is aberrantly activated in the spinal MN population of adult CKO mice, and re-introducing CRABP1 to young CKO mice could significantly lower their CaMKII activity in spinal MNs and rescue their motor defects later. In a preliminary in vitro test using an immortalized MN cell line, MN1, we found that elevating CRABP1 levels in MN1 improved their axon extension. These experiments provided further evidence for the potential therapeutic application of targeting CRABP1, such as in dealing with diseases caused by defects in MNs [10]. Extended from these interesting findings in the CKO mouse model, this current study aims to search for CRABP1-binding (without activating RAR), atRA-like compounds that can modulate CaMKII and to establish an in vitro model for studying if and how these CRABP1-ligands may affect the process of MN differentiation and health.
While the literature has reported several MN model systems [16], there remains a need for a more reliable and reproducible in vitro system where MN differentiation can be more robustly induced for systemic studies, especially studies allowing dissection of intermediate events in various stages of MN differentiation and maturation processes. For this current study, we, therefore, exploited P19, a widely used embryonal carcinoma cell line that is embryonic stem cell (ESC)-like and much easier to manipulate; further, P19 does not require a feeder layer in culture [17]. Using this newly developed P19-MN differentiation system, we tested the feasibility of targeting CRABP1 to modulate CaMKII activation in the process of MN differentiation and in maintaining MN health.
## 2.1. Characterization of C32 as Novel atRA-like Compound That Binds CRABP1
Previously, using a rational screening approach, we have identified novel, atRA-like compounds, C3 and C4, as CRABP1 ligands that modulate Erk activation [12]. Using this approach, we define a hit CRABP1-binding compound as [1] binding to CRABP1 and [2] lacking RAR activation activity. We now report another CRAPB1 ligand, C32 (chemical name: 2-(3,5,5,8,8-Pentamethyl-5,6,7,8-tetrahydronaphthalene-2-carboxamido) thiazole-5-carboxylic acid). The chemical structures of atRA, C32, and C4 are shown in Figure 1A. Differential scanning fluorimetry (DSF) was utilized to determine CRABP1-binding activity. DSF is a thermal shift assay in which ligand binding to the protein of interest results in a thermal-stable, ligand-protein complex with a higher melting temperature compared to that of vehicle control [18]. The relative increase in melting temperature upon ligand binding is reported as delta Tm (ΔTm). In this assay, CRABP1 binding was defined as a ΔTm greater than or equal to 1 °C (ΔTm ≥ 1 °C). Typical high-throughput screening approaches utilize a cut-off of 3 standard deviations above the mean [19,20,21]. Given that CRABP1 has a highly consistent Tm (Supplementary Figure S1A,B), with minimal variance across biological and technical replicates, ΔTm ≥ 1 °C provides a highly stringent cut-off for the hits. This high stringency cut-off allows the identification of robust CRABP1-binding compounds and reduces the potential for false positives.
This criterium for CRABP1 binding is described in-depth in Section 4.2. In DSF, atRA (100 μM) was first used as a positive control (Figure 1B). When compared to the DMSO control (blue curve), RA was detected to bind CRABP1 (red curve), generating a ΔTm = 20 °C, validating this DSF assay in detecting CRABP1-binding compounds. Using this test, C32 (100 μM) appeared to bind CRABP1, generating a ΔTm = 3 °C (orange curve) compared to the DMSO control (blue curve) (Figure 1C). Previously identified, using a conventional ligand displacement assay [12], CRABP1 ligands C3 and C4 were also subjected to DSF (Supplementary Figure S1C,D). C3 and C4 at 100uM generated ΔTm = 0.26 and ΔTm = 0.09, respectively. Although these values are below the CRABP1 binding criterium of ΔTm ≥ 1 °C, the positive shift in Tm indicates a CRABP1-binding event for both C3 and C4, consistent with the previous positive result using conventional ligand displacement assay [12]. These results suggest that DSF with a stringent cut-off of ΔTm ≥ 1 °C is suitable for identifying ligands that robustly bind CRABP1.
To ensure that C32 met the second criterion, i.e., lacking RAR-activation activity, we utilized a classical luciferase-based RAR activation assay in the Cos-1 cell line. As expected, RA (0.25 μM) significantly activated RAR (298 ± 83.5-fold activation) compared to the control, whereas C32 exhibited no RAR activation (0.76 ± 0.47-fold-activation) compared to the control (Figure 1D). These data allowed us to identify C32 as a new atRA-like compound that binds CRABP1 without activating RAR.
## 2.2. AtRA and C32 in CRABB1-Mediated CaMKII Dampening
CaMKII activity is marked by changes in the phosphorylation status of key regulatory residue threonine 286 (or 287, depending on the isoform) [22]. The regulatory role for CRABP1 in CaMKII activation was first examined in a reconstituted HEK293T cell model transfected with CaMKII and CRABP1 (or empty vector control). The status of pCaMKII was then assessed via western blot using an antibody specific for phosphorylated threonine $\frac{286}{7.}$ *Using this* system, we then determined the CaMKII-modulating activities of the three CRABP1-binding ligands, C32 and two previously reported ligands, C3 and C4 that were shown to elicit CRABP1-dependent Erk-modulating activity. The results show that atRA, C32, and C4 treatment had no significant effect on pCaMKII activity in the control vector transfected cells (Figure 2A, left blot; Figure 2B left graph). However, treatment with atRA, C32, or C4 at 0.5–5 μM for 15 min dampened CaMKII activity in CRABP1-transfected cells (Figure 2A, right blot; Figure 2B right graph), whereas C3 (that could elicit CRABP1-mediated Erk modulation) had no effect on CaMKII activation. Therefore, C32 and C4, as well as the positive control atRA, are CRABP1 ligands that can dampen CaMKII activity. This is further supported by the fact that this CRABP1-dependent CaMKII dampening activity is detected rapidly (15–60 min), confirming that this CRABP1-dependent C32 and C4 activity is non-canonical in nature.
We then sought to determine if this non-canonical activity of atRA, C32, and C4 on CaMKII modulation can be detected in a more biologically relevant context in which CRABP1 and CaMKII are endogenously present. We thus exploited the widely used P19 embryonal carcinoma cell line, which was shown to endogenously express both CRABP1 [23] and CaMKII [24]. In contrast to the reconstituted HEK293T cell studies where only a single isoform CaMKII isoform (CaMKII beta) was introduced, P19 cells endogenously expressed two isoforms of CaMKII, indicated by the detection of two bands with the pan-pCaMKII antibody. Four major isoforms of CaMKII -alpha, beta, delta, and gamma are known to exist and vary in expression levels depending on cell and tissue types [25], stage of development [26], and other biological and disease contexts [27,28].
Treatment with atRA, C32, or C4 at 0.5–5 μM for 15 min dampened endogenous CaMKII in P19 cells (Figure 2C,D), validating this CaMKII-modulatory effect of CRABP1 ligands in a physiological context. Therefore, it is concluded that C32 and C4 elicit CRABP1-dependent CaMKII-modulatory (dampening) activity.
## 2.3. A New In Vitro Stem Cell-MN Differentiation System, P19, for Studying CaMKII Activation
We have previously reported that CKO mice exhibited dramatically elevated CaMKII activation in MNs, which contributed to MN death and motor deterioration in adult mice [10]. This prompted us to carefully examine how CRABP1 signaling might affect the MN differentiation process. We thus exploited the P19 cell line, which has been extensively utilized for its differentiation potential in studying various stages of cell differentiation, including neuronal differentiation [17]. Here we developed a P19-derived motor neuron (MN) differentiation culture system to probe the effects of atRA and CRABP1 ligands in MN differentiation with regard to CaMKII activation. This P19 culture system is superior to ESC-differentiation systems which generally are very sensitive to technical complications and clonal variation [16,29].
Figure 3A depicts the workflow for the new P19-MN differentiation system and the data demonstrating MN differentiation efficiency. First, P19 cells in a single-cell suspension were transferred into P19 Differentiation Medium containing 0.5 μM atRA (+RA Medium, see Section 4.6 for complete media formulations) in a T75 flask. The flask was stored up-right to prevent cells from attaching to the coated surface of the flask, allowing the cells to form embryoid bodies (EB) over a two-day period (Day −4 to Day −2). The EBs were then collected and resuspended in fresh P19 Differentiation Medium containing 0.5 μM atRA and 200 ng/mL mouse Sonic Hedgehog protein (Shh) (+RA, +Shh Medium) in a new T-75 flask for neurosphere (NS) formation over the next two-day period (Day −2 to Day 0). The flask was stored in the same manner to promote NS formation. Shh is a potent morphogen known to promote ESC differentiation into functional MNs [30]. On Day 0, the NSs were dissociated into single cells and then plated onto a Matrigel-coated 6-well plate for MN differentiation (Day 1 to Day 3). Cells were collected and analyzed at various time points for the expression of relevant markers of MNs.
Figure 3B shows brightfield images of undifferentiated P19 stem cells (Day −4, left) and P19-derived Day 3 MNs (right). Undifferentiated P19 cells typically grow in a clumped manner with an epithelial-like morphology. In contrast, Day 3 P19-MNs grow elongated and branched processes. In addition to MN-like morphological features, the induction of several MN-specific marker genes, HB9 [31], ChAT [32], Isl1, and Isl2 [33,34] (Figure 3C, Supplementary Figure S2A) and markers for other spinal neurons such as V2 interneurons, LHx3 [35] (Supplementary Figure S2A) were also monitored. The induction of Isl2 specifically marks the presence of somatic-type spinal MNs, which innervate and maintain muscle tissue tone [36]. MN markers and spinal neuron markers appear to be readily elevated on Day 1 (Figure 3C, Supplementary Figure S2A), begin to decline after Day 3, and continue to decline on Day 5 (Supplementary Figure S2A). Peak expression of MN markers on Day 1–3 suggests that this could be the optimal time window to specifically study the effect of CRABP1-ligands in the MN-differentiation process with regards to CaMKII activation.
Importantly, *Crabp1* gene expression appears to be elevated from the EB stage and steadily maintained until Day 3 (Figure 3D, Supplementary Figure S2C). On Day 5 of MN differentiation, a sharp decrease in Crabp1 expression was observed (Supplementary Figure S2B), further supporting that Day 1–3 is the optimal time window to study the effects of CRABP1 ligands on CaMKII activation in this P19-MN differentiation system.
## 2.4. AtRA, C32, C4 Dampening CaMKII Activity in P19-MN Differentiation Process
To determine if atRA, C32, or C4 affects CRABP1-mediated CaMKII modulation, we carried out a series of experiments focusing on Day 1 and Day 3 in the P19-MN differentiation system. First, on the relevant day of interest (Day 1 or Day 3), atRA and Shh were depleted by replacing the medium with fresh differentiation medium without atRA and Shh. Additionally, dextran charcoal-treated bovine serum was used to further deplete other factors, particularly retinoids, to remove any potential contribution of genomic activities of atRA. Following this depletion step, cells were treated with atRA, C32, or C4 and then immediately harvested for western blot analyses to monitor CaMKII activation (Figure 3A, open circles).
On Day 1 (Figure 4A,B) and Day 3 P19-MN (Figure 4C,D), atRA, C32, and C4 significantly dampened endogenous CaMKII activity. For Day 1 MNs, RA, C32, or C4 were added at 1 μM for 15 min. For Day 3 MNs, atRA, C32, or C4 were added at 1 μM for 30 min. Together, the data show that atRA and CRABP1-binding ligands, C32 and C4, can dampen CaMKII in both Day 1 and Day 3 P19-MN differentiating/differentiated cells, suggesting their effects in multiple stages of the MN differentiation process. Note that this CaMKII-dampening activity was also detected rapidly (15–60 min), confirming that they elicited the non-canonical activity of atRA.
Interestingly, additional isoforms of CaMKII are routinely detected in P19-MNs, compared to undifferentiated P19 cells. This suggests expression of additional, neuron-specific alpha or beta CaMKII isoforms in more differentiated cells [25]. While the intensity of isoforms differed between D1 and D3 cells, all CaMKII isoforms were dampened by these CRABP1 ligands in both D1 and D3 cells, suggesting that CRABP-modulation is effective for these various CaMKII isoforms.
## 2.5. CRABP1 in Neuroprotection against Calcium (Ca2+)-Induced Toxicity in MN Cells
Aberrant CaMKII activity is frequently implicated in neurodegeneration, especially in mediating the destructive downstream events of excitotoxicity, such as calcium (Ca2+) overload-mediated cell death [37]. We previously reported that MN cells in CKO mice had highly elevated CaMKII activity and appeared unhealthy, which also coincided with their augmented expression of agrin protein, a proteoglycan essential for MN development and health [10,38]. This suggests that CRABP1 could be a protective player in maintaining healthy MNs.
To determine if CRABP1 indeed plays a neuroprotective role in committed MN cells, such as during excitotoxic insult, we exploited an established MN-committed cell line (MN1). *We* generated a stable CRABP1-overexpressing MN1 clone (CRABP1-MN1) and determined if this could provide a protective effect against ionomycin assault. Ionomycin is a selective Ca2+ ionophore that causes rapid increases in intracellular Ca2+ concentration ([Ca2+]i) [39], triggering CaMKII activation [22], an event contributing to toxicity and subsequent cell death [40]. Wild-type (WT) MN1 and CRABP1-MN1 were exposed to ionomycin at 5 μM for 18 h, and cell viability was monitored immediately after the 18 h incubation period with MTT using 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) viability assay. MTT is metabolized by living cells to an insoluble, measurable form known as formazan [41], which directly measures live cell/viability. Indeed the results show that CRABP1-MN1 exhibits significantly greater cell viability, as compared to WT, after ionomycin exposure (Figure 5A), confirming that elevating CRABP1 levels can protect MN from excitotoxicity-induced cell death.
To validate the neuroprotective, biological activity of CRABP1 ligands, atRA, C32, and C4, we determined if these ligands could protect against ionomycin assault in WT MN1 cells, a more physiologically relevant experimental system. Because this experiment is to determine the potential protective effects of the compounds, endogenous levels of CRABP1 expression provide a more biologically relevant cell context to study the effects of these compounds. First, MN1 cells were pre-treated with atRA, C32, or C4 (0.5–5 μM) for 1.5 h. Immediately after pre-treatment, ionomycin (4 μM) or DMSO (as vehicle control) was added, and cells were incubated overnight, which typically induced cell death. Compounds were present during the duration of ionomycin exposure. The next day, treated cells were subjected to an MTT assay. As expected, compared to DMSO control, ionomycin (4 μM) significantly reduced cell viability. Interestingly, atRA exhibited a trend towards improved cell viability, while C32 and C4 significantly improved cell viability (Figure 5B).
To demonstrate the proposed mechanism via CRABP1-mediated CaMKII dampening, we compared Wild-type (WT) and CRABP1-MN1 (over-expressing CRABP1 to elevate CRABP1 level) treated with medium (basal), DMSO (vehicle control), or ionomycin (10 μM, 5–10 min) with regards to their CaMKII activation as reflected on threonine $\frac{286}{7}$ phospho-status. As shown in Figure 5C, elevating the CRABP1 level (right panel, CRABP1 Over-Expression) clearly dampened CaMKII activity. This result supports the observations made previously using the reconstituted HEK293T system [10]. This further strengthens our hypothesis that targeting CRABP1-signaling can be developed into a protective and/or therapeutic strategy.
Taking data collected from P19-MN differentiation and MN1 systems, it is concluded CRABP1 can provide a neuroprotective effect against cell death induced by pathological Ca2+ overload. Furthermore, this protective effect can be exploited by using CRABP1 ligands such as C32 and C4 to enhance the protective mechanism and improve cell viability. Mechanistically this neuroprotective effect can be attributed to CRABP1-signaling that dampens CaMKII over-activation in differentiating or differentiated MNs. In summary, as depicted in the proposed model (Figure 5D), when MNs experience cytotoxic stimulation (step 1), it can result in pathological increases in [Ca2+]i and subsequent CaMKII activation and phosphorylation of AMPAR [42] (step 2), CRABP1, as well as its ligand (such as atRA, C32, and C4), could provide an inhibitory effect to dampen this aberrant CaMKII activation (step 3), thereby preventing MN death (step 4).
## 3. Discussion
Here we report the screening of new atRA-like compounds for binding to CRABP1 without activating RAR, which allowed us to identify a new CRABP1 ligand, C32. By testing C32, as well as previously reported CRABP1-binding compounds [12] for their ability to modulate CaMKII activation, we have identified C32 and C4 (previously reported to modulate Erk activation) [12], both are CRABP1-binding ligands that can modulate CaMKII signaling in MNs. In our previous reports, C3 and C4 were identified based on their binding to CRABP1, detected using conventional ligand displacement assay, and their ability to modulate Erk signaling [12]. Together, these results show that C32 exhibits CaMKII-modulatory activity, C3 is an Erk-selective CRABP1 ligand, whereas C4 appears to be a pan-acting CRABP1 ligand. How C32 behaves in Erk signaling will require further intensive study. Nevertheless, these three compounds comprise the first series of useful CRABP1-binding ligands that may be worthy of further investigation. In order to describe the precise binding profiles of C32, C3, C4, and next-generation CRABP1-binding compounds, future studies are needed to rigorously compare their pharmacological properties. For instance, it would be of most interest to determine their binding characteristics, such as affinity and kinetics, in order to more precisely differentiate and categorize CRABP1 hit compounds. These studies will also be important for generating more rationale-designed, next-generation CRABP1 compounds that could display pathway-selective compounds. To this end, assays to confirm the various biological activities of CRABP1 ligands will be important in order to contextualize the therapeutic potential and physiological relevance of these novel CRABP1 ligands. An important feature of these compounds is their lack of RAR-activation ability (see the following section). Preclinical models would be important for further studies to determine their potential as therapeutics by targeting CRABP1 to mitigate diseases associated with the over-activation of Erk, CaMKII, or both signaling pathways. We previously hypothesized that it is possible to design CRABP1-specific and signaling pathway-selective atRA-like compounds for safer therapeutic applications [5,12]. This current study supports such an interesting possibility.
AtRA and atRA-like compounds targeting RARs have been extensively and enthusiastically studied for therapeutic applications; however, the widely documented toxicity (RAR-mediated retinoid toxicity) has hindered the progress in this field and greatly limited their potential in clinical applications. The recently established non-canonical activities of atRA, mediated by CRABP1 [43], and the demonstration of multiple human disease-mimicking phenotypes of CKO mice [7,10] prompted us to propose a new therapeutic strategy using CRABP1-binding, atRA-like compounds that lack RAR activity to modulate specific disease-related signaling pathways such as Erk and CaMKII. These compounds were designed based on, specifically, the CRABP1 binding pocket, and it is known that the binding pocket of CRABP1 has little structural or sequence relationship with the ligand binding domain of RAR [44,45,46,47]. According to the original design strategy, it is tempting to speculate that C3, C4, and C32 may not bind RAR. However, for truly “CRABP1-specific” ligands, RAR binding and potential antagonism must be ruled-out in future studies. Nevertheless, this current study provides the first support for the possibility of designing signaling pathway-selective CRABP1-binding ligands for therapeutic intervention.
With regards to CRABP1 modulating CaMKII signaling, we have reported two spontaneously developed human disease-mimicking phenotypes of CKO mice; both are associated with aberrant CaMKII activation and cytotoxicity, i.e., cardiomyopathy/heart failure caused by cardiomyocyte apoptosis and death [7] and ALS-like motor deterioration caused by MN death/loss [10]. Importantly, in human studies, drastically reduced *Crabp1* gene expression has been reported in neurodegenerative disease patients, including ALS and SMA patients [48,49]; therefore, we prioritized the studies of CRABP1-binding ligands in the context of MN degeneration. Extensive classical studies have reported aberrant CaMKII activation in neurodegeneration because over-activation of CaMKII and abruptly surged intracellular Ca2+ concentration is often a result of excitotoxicity, which subsequently leads to neuron death [50]. As introduced earlier, in this current study, we extended our previous findings of CKO mouse MN degenerative phenotype, aiming to develop a more reliable and feasible in vitro MN model for mechanistic studies and for screening CRABP1-specific compounds that can modulate CaMKII to improve MN health. To this end, we were able to exploit P19 and develop a P19-MN differentiation system, which allowed us to identify C32 and C4 as CRABP1 ligands that could modulate CaMKII in the context of MN differentiation. Our data also support the notion that increasing CRABP1 signaling can improve MN health, as demonstrated by the reduction of excitotoxic-induced death in an MN1 cell line with stably elevated CRABP1 expression. We also show that C32 and C4 are protective against ionomycin assault (mimicking Ca2+ overload as observed in excitotoxicity). Mechanistically, this protection may be attributed to their ability to bind CRABP1 to further dampen CaMKII over-activation.
The potential role of CRABP1 in modulating MN health and motor function was first detected in CKO mice; however, it was unclear when CRABP1 and its ligands could play a role along the course of MN differentiation or maturation. Because MN1 is a committed MN cell line [51], this system is not appropriate for dissecting intermediate events during the differentiation, especially in the early differentiation stages. On the contrary, the P19-MN differentiation system spans the entire course of neural progenitor (or stem cell) progressing to MN differentiation, allowing interrogation of intermediate steps in the entire process of MN differentiation. As shown here, both C32 and C4 are effective, in terms of modulating endogenous CaMKII activation, at both early and late differentiation stages, suggesting that CRABP1 signaling can be involved in multiple steps of MN differentiation. Targeting CRABP1 may be beneficial to multiple steps in the MN differentiation process; therefore, this strategy may also be useful in preventive applications. Along with our findings that C32 and C4 pre-treatment improves cell viability after ionomycin assault in MN1 cells, it is tempting to speculate a potential clinical application as a prophylactic therapeutic to combating various neurological pathologies associated with Ca2+ overload/excitotoxicity. Preventative approaches may also offer an attractive strategy that can preserve a healthy neuronal population, in contrast to interventions that are typically introduced at disease onset when significant neuronal damage has already occurred [52].
Since both CaMKII and Erk signaling pathways are important for normal physiological processes in numerous organ systems and multiple cell types, pharmacological intervention to target these signaling pathways non-discriminatorily is more likely to cause toxicity. However, by selectively targeting one of these signaling pathways modulated by CRABP1 with CRABP1-specific ligands, it is possible to deliver therapeutic effects that are more specific and safer because these drug effects would be limited to certain disease-relevant cell types that are CRABP1-positive. Therefore, it would be interesting in future studies to address whether compounds like C32 and C4 can be used in therapeutic/preventive applications for diseases associated with CaMKII over-activation in CRABP1-positive cells/tissues, such as progressive motor deterioration and cardiomyopathy as revealed in the CKO mouse model. A pan-acting ligand like C4 may be useful in dealing with disease conditions where both Erk and CaMKII signaling pathways are altered, such as Alzheimer’s disease (AD), amyotrophic lateral sclerosis (ALS), and Parkinson’s disease (PD) [53,54,55,56].
The list of pathological conditions associated with the non-canonical activities of atRA is growing. This study provides the first insight into the potential of designing signaling pathway-selective, CRABP1-binding atRA-like ligands in mitigating diseases. In the future, CRABP1-specific compounds that can elicit the non-canonical activities of atRA may constitute an attractive and novel group of compounds that have the potential for therapeutic/preventive application for a wide spectrum of diseases with minimized retinoid toxicities.
## 4.1. Reagents and Compound Library
All-trans retinoic acid (atRA, Sigma Cat# R2625) and ionomycin salt (Cat# I0634) were obtained from Sigma (St. Louis, MO, USA) and dissolved in DMSO. RA-like compounds C32 and C4 were synthesized by our collaborator Dr. Hiroyuki Kagechika at Tokyo Medical and Dental University. More details on the RA-like compound library can be found in [12]. For compound studies, C32 and C4 were dissolved in DMSO. All compounds were stored at −80 °C with limited freeze-thaw cycles.
## 4.2. Differential Scanning Fluorimetry (DSF) CRABP1 Binding Assay and Data Analysis
His-tagged CRABP1 was purified as described in [11]. 5 ug of CRABP1 (14.1 μM) was incubated with 100 μM of atRA, C32, C3, or C4 to yield a ligand to CRABP1 molar ratio of 7:1 in order to achieve saturation and preserve ligand solubility in the aqueous reaction buffer. For all experiments, DMSO was used as the vehicle control, and binding reactions were carried out in 1XPBS, pH 8.0 buffer for 1 h at room temperature on an orbital shaker. After incubation, 18 μL of the CRABP1-ligand mixture was transferred into a 96-well plate and 2 μL of 20X SYPRO Orange was added to a final concentration of 2× SYPRO Orange and a final reaction volume of 20 μL in a single well. SYPRO Orange (Invitrogen Cat. S6650, Waltham, MA, USA) was diluted from a 5000xX stock to a 20× stock in 1XPBS, pH 8.0.
Data were acquired on the QuantStudio™ 3 Real-Time PCR instrument (Applied Biosystems, Waltham, MA, USA). Data acquisition parameters were created using Design and Analysis Software (Ver 2.6.0) and are as follows: [1] starting temperature of 25 °C held for 2 min, [2] temperature was then incrementally increased from 25 to 99 °C at a ramp speed of 0.05 °C/s, [3] fluorescent readings were acquired using filter settings of 520 ± 10 nm excitation wavelength and 558 ± 11 nm emission wavelength. For each independent experiment, 6 technical replicates were included for each condition. Wells containing only ligand and SYPRO Orange were assayed as “ligand-only” controls to ensure that compounds alone did not contribute any fluorescent signal that may interfere with data analysis or generate false positives.
For data analysis, CRABP1 melt curves were generated by calculating the negative first derivative of fluorescence (RFU) over temperature (T) (−ΔRFU/ΔT) and plotting against temperature. −ΔRFU/ΔT and temperature values were calculated in the Design and Analysis software. This melt curve data was then exported to Microsoft Excel, which was used to extract the minimum (lowest) −ΔRFU/ΔT value of the melt curve. The corresponding temperature to the minimum −ΔRFU/ΔT value defines the CRABP1 melting temperature (Tm) [57]. The Tm’s from the 6 technical replicates were averaged and used to calculate the thermal shift values (ΔTm). ΔTm was calculated by taking the difference between DMSO control and ligand conditions (ΔTm= ligand-DMSO). A hit compound was defined as a ΔTm greater than or equal to 1 °C above the DMSO control Tm (ΔTm ≥ 1 °C of DMSO). This cut-off was selected as a more stringent hit threshold to identify hit ligands that generated a robust ΔTm. Experiments were performed three independent times.
## 4.3. Cell Culture
Cos-1 cells were maintained as described in [12]. HEK293T cells were maintained in DMEM medium (Gibco #11965, Billings, MT, USA) containing 4.5 g/L D-glucose, 4 mM L-glutamine, 44 mM Sodium Bicarbonate, 100 U/mL penicillin, 100 mg/mL streptomycin, and $10\%$ heat-inactivated FBS as described in [10]. HEK293T cells were co-transfected with GFP-CaMKII (Addgene #21227, Watertown, MA, USA) and either empty vector (EV) or Flag-CRABP1 (construct information described in [11]) and in a GFP-CaMKII to EV/CRABP1 to a ratio of 1:5. A total of 10 ug for a 10 cm cell culture dish was transfected using polyethylenimine (PEI, Polysciences Cat #23966, Warrington, PA, USA) in a DNA to PEI ratio of 1:5. P19 cells were purchased from ATCC, Manassas VA, USA (Cat# CRL-1825) and maintained in alpha minimum essential medium (MEM) with ribonucleosides and deoxyribonucleosides (Gibco, Cat# 12571-063) supplemented with $7.5\%$ Bovine Calf Serum, Ion Fortified (ATCC, Cat # 30-2030), $2.5\%$ Fetal bovine serum (R&D Systems, Cat# S11150, Minneapolis, MN, USA) and Pen Strep (Gibco, Cat #15140-122). All cells were maintained at 37 °C in a humidified $5\%$ CO2 cell culture incubator.
For compound studies, transfected HEK293T and P19 cells were exchanged into complete medium with dextran charcoal treated (DCC) bovine serum in place of normal bovine serum to deplete exogenous RA or any other non-specific hormones for 18 h before compound treatment and downstream experiments. For P19 compound experiments, the $7.5\%$ calf serum and $2.5\%$ fetal bovine serum (FBS) mixture was replaced with $10\%$ DCC fetal bovine serum.
## 4.4. RAR Luciferase Reporter Assay
Luciferase assay for RAR activation was performed as described in [58]. Briefly, Cos-1 cells were transfected with RARE-tK-Luc and pRL renilla control plasmid using lipofectamine 3000 (Invitrogen). Following transfection, cells were washed and exchanged into fresh maintenance medium and were treated with DMSO control, RA, or compound at 0.25 μM for 24 h. Luciferase assay was performed using the Dual-Luciferase Reporter Assay kit (Promega, Madison, WI, USA). Luciferase and renilla signal was detected on an Infinite M1000 Pro Tecan (San Jose, CA, USA) plate reader. Assay was performed at least three independent times with three replicates each time. Fold activation was determined by luciferase activity and readings were normalized to renilla internal control readings.
## 4.5. Compound Studies and Western Blot
Preliminary studies determined that 15–60 min was the optimal window to detect compound effects on CaMKII activity for HEK293T, P19, and P19-MN experiments. For each cell line, an optimal time point within 15–60 min was identified and consistently applied across all independent experiments. DMSO, atRA, C32, or C4 were then added at 0.5–5 μM for the optimal time point determined for that particular cell line, and cells were immediately harvested for western blot analyses. All compound experiments were repeated for at least three independent times.
Western blot was performed as described in [10] with the following modification: cells were immediately lysed and harvested by adding lysis buffer (9 parts: 128 mM Tris base, $10\%$ (v/v) glycerol, $4\%$ (w/v) SDS, $0.1\%$ (w/v) bromophenol blue, pH to 6.8 and 1 part: beta-mercaptoethanol) directly to the dish or plate containing treated cells. For primary antibodies, Anti-p-CaMKII (cat #: 127165, $\frac{1}{1000}$) was obtained from Cell Signaling, Danvers, MA, USA, Anti-CRABP1 from Sigma (cat #:HPA017203), Anti-CRABP1 from Invitrogen (Cat #: MA3-813), anti-β-Actin (cat #: SC-47778, $\frac{1}{2000}$) was obtained from Santa Cruz Biotechnology, Dallas, TX, USA. For secondary antibodies, anti-Rabbit-IgG (cat #: 11-035-144, $\frac{1}{2000}$) was obtained from Jackson ImmunoResearch, Ely, UK, and anti-Mouse-IgG-HRP (cat #: GTX26789, $\frac{1}{5000}$) was obtained from GeneTex, Irvine, CA, USA.
Cell lysates were separated on $9\%$ (v/v) SDS polyacrylamide gels and transferred onto 0.45 µm PVDF membrane. The membranes were cut according to molecular weight and probed with appropriate primary and secondary antibodies. Images were acquired using the Bio-Rad ChemiDoc Imager, Hercules, CA, USA (cat #: 17001402). Image analysis was performed using BioRad Image Lab software (Ver. 6.1)of ImageJ [59].
## 4.6. P19-Derived Motor Neuron (MN) Differentiation, Compound Studies, and qPCR Gene Studies
P19 cells were suspended in P19 differentiation medium ($50\%$ neurobasal medium (Gibco, Cat# 21103), $25\%$ alpha MEM and $25\%$ P19 maintenance medium) containing 0.5 μM retinoic acid (RA) (Sigma-Aldrich, Cat# R2625) in a T75 flask to form embryo body (EB). The flask was set up-right in the culture incubator to promote EB formation. After two days, the EBs were collected and then resuspended in P19 differentiation medium with 0.5 μM RA and 200ng/mL mouse Shh (STEMCELL Technologies, Vancouver, BC, Canada, Cat# 78066) to form neurosphere (NS). After two days, the neurospheres were dissociated into single cells by using Accumax (Millipore, Burlington, MA, USA, Cat# A7089), then suspended in P19 differentiation medium with 0.5 μM RA and 200 ng/mL Shh. The cells were plated in the 6-well plate coated with 360 μg/mL Matrigel Basement Membrane (Thermal Fisher Scientific, Cat# A1413301, Waltham, MA, USA) to differentiate into motor neurons. Brightfield images were acquired on a Leica DM IRB inverted microscope with a 10× objective lens using an Infinity 1 camera and INFINITY ANALYZE software (Ver. 6.5).
For compound studies, on the relevant day (Day 1 or Day 3), media was exchanged in depletion medium ($50\%$ neurobasal medium (Gibco, Cat# 21103), $25\%$ alpha MEM and $25\%$ P19 culture medium). The P19 culture medium was supplemented with $10\%$ DCC FBS instead of normal FBS to deplete RA or other non-specific hormones that can originate from FBS.
qPCR was performed as described in [10]. Briefly, TRIzol (Invitrogen, Carlsbad, CA, USA) was used to isolate total RNA. The Omniscript RT Kit (QIAGEN, Germantown, CA, USA) was used to synthesize cDNA. qPCR was performed with SYBR-Green master mix (Agilent, Santa Clara, CA, USA) and detected with Mx3005P (Agilent). *Target* genes for qPCR were: ChaT, Hb9, Isl1, Isl2, and Crabp1. qPCR experiments were performed two independent times. Primer sequences can be found in Supplementary Table S1.
## 4.7. Hybrid Motor Neuron (MN1) Cell Culture and Stable CRABP1 Over-Expression Clone Generation
Wild-type MN1 cells were cultured in complete DMEM medium (Gibco #11965) containing 4.5 g/L D-glucose, 4 mM L-glutamine, 44 mM Sodium Bicarbonate, 100 U/mL penicillin, 100 mg/mL streptomycin, and $10\%$ heat-inactivated FBS. *To* generate the stable over-expression MN1 cell line, first, mouse Crabp1 cDNA was cloned into pCDH-EF1α-MCS-IRES-Puro plasmid (SBI, #CD532A-2) as previously described [11], resulting in 3XFlag-HA-tagged CRABP1 as a protein product. All plasmid DNA was purified using the PureLink HiPure Plasmid Filter Midiprep Kit (Invitrogen #K210014). For lentivirus production, 2 × 106 HEK-293T cells were seeded in complete DMEM medium without antibiotics dish in a 10 cm dish overnight. 9.6 µg Crabp1-pCDH target plasmid, 7.2 µg psPAX2 packaging plasmid, 2.4 µg pMD2.G envelope plasmid were co-transfected into cells with Lipofectamine 2000 transfection reagent (Invitrogen) following the manufacturer’s protocol. Media was changed to 6 mL of fresh complete DMEM medium containing $1\%$ BSA after 6 h. Infectious lentiviruses were harvested at 24 h and 48 h post-transfection and filtered through 0.45 µM pore cellulose acetate filters. For transduction, 1 × 105 MN1 cells were seeded in complete DMEM medium in 6-well plates overnight. 2 mL of lentivirus with 8 µg/mL polybrene (Millpore TR-1003-G) were added into cells, and then cells were centrifuged at 800× g, 37 °C for 60 min. Lentivirus was removed, and the medium was changed after 24 h. Puromycin selection was started at 48hrs post-transfection. Cells were selected and maintained in the same MN1 medium as described above with the addition of 3 μg/mL puromycin. After puromycin selection, stable MN1 cells were collected and examined for Crabp1 expression by qPCR.
## 4.8. Ionomycin-Induced Cell Death and MTT Viability Assay
MTT reagent (Sigma Cat# M5655) was prepared by dissolving in 1XDPBS to a final concentration of 5 mg/mL and then sterile-filtered. Wild-type (WT) and the CRABP1 stable clone MN1 were then seeded into a 24-well plate at a density of 1 × 105 cells/well the night before. Puromycin selection was withdrawn from CRABP1-MN1 of these experiments. DMSO (vehicle) or ionomycin (5 μM) were treated with WT and CRABP1- MN1 for 18 h. The Final volume in each well was 1ml. Then 100 μL ($10\%$ of total well volume) of MTT reagent was added to MN1 cells and incubated at 37 °C in a humidified $5\%$ CO2 cell culture incubator for three hours to allow for formazan crystal formation. Then cell culture and MTT reagent were gently removed via suction, and the remaining formazan crystals dissolved in 600 μL of DMSO. Formazan absorbance was measured at 570 nm, and a background reading at 690 nm was acquired using an Infinite M1000 Pro Tecan plate reader. Values were exported and analyzed in Microsoft Excel. After performing background subtraction, percent cell viability was calculated by the formula: (1−–(DMSO−–Ionomycin)) × * 100. This assay was performed 5 independent times with 4–12 replicates for each condition.
For compound experiments, WT-MN1 cells were seeded in a 24-well plate as described above. MN1 cells were pre-treated with RA, C32, and C4 at 0.5–5 μM for 1.5 h prior to ionomycin exposure. After pre-treatment, ionomycin was added to a final concentration of 4 μM, overnight (18 h) to induce cell death. atRA, C32, and C4 were also present in the cell culture medium for the duration of ionomycin exposure. Immediately after the 18-h co-treatment, MN1 cells were subjected to an MTT viability assay as described above. This experiment was performed three independent times with 3–4 technical replicates for each condition.
## 4.9. Statistical Analysis
HEK293T, undifferentiated P19, and P19 MN compound experiments were analyzed using paired Student’s t-test (DMSO vs. RA, C32, or C4). RAR activity luciferase assay data were analyzed using paired Student’s t-test. MTT cell viability assay data were analyzed using paired Student’s t-test. Significance was defined as * p ≤ 0.05. “ N.S.” indicates not significant. Error bars for all data are presented as mean ± standard deviation (SD).
## References
1. Cunningham T.J., Duester G.. **Mechanisms of retinoic acid signalling and its roles in organ and limb development**. *Nat. Rev. Mol. Cell Biol.* (2015) **16** 110-123. DOI: 10.1038/nrm3932
2. Wei L.-N.. **Retinoid Receptors and Their Coregulators**. *Annu. Rev. Pharmacol. Toxicol.* (2003) **43** 47-72. DOI: 10.1146/annurev.pharmtox.43.100901.140301
3. Nagpal I., Wei L.N.. **All-trans retinoic acid as a versatile cytosolic signal modulator mediated by CRABP1**. *Int. J. Mol. Sci.* (2019) **20**. DOI: 10.3390/ijms20153610
4. Wei L.-N.. **Non-canonical activity of retinoic acid in epigenetic control of embryonic stem cell**. *Transcription* (2013) **4** 158-161. DOI: 10.4161/trns.25395
5. Persaud S.D., Lin Y.-W., Wu C.-Y., Kagechika H., Wei L.-N.. **Cellular retinoic acid binding protein I mediates rapid non-canonical activation of ERK1/2 by all-trans retinoic acid**. *Cell. Signal.* (2013) **25** 19-25. DOI: 10.1016/j.cellsig.2012.09.002
6. Ong D.E., Chytil F.. **Cellular retinoic acid-binding protein from rat testis. Purification and characterization**. *J. Biol. Chem.* (1978) **253** 4551-4554. DOI: 10.1016/S0021-9258(17)30423-4
7. Park S.W., Persaud S.D., Ogokeh S., Meyers T.A., Townsend D.W., Wei L.N.. **CRABP1 protects the heart from isoproterenol-induced acute and chronic remodeling**. *J. Endocrinol.* (2018) **236** 151-165. DOI: 10.1530/JOE-17-0613
8. Lin Y.W., Park S.W., Lin Y.L., Burton F.H., Wei L.N.. **Cellular retinoic acid binding protein 1 protects mice from high-fat diet-induced obesity by decreasing adipocyte hypertrophy**. *Int. J. Obes.* (2020) **44** 466-474. DOI: 10.1038/s41366-019-0379-z
9. Wei C.W., Nhieu J., Lin Y.L., Wei L.N.. **Modulation of adipose inflammation by cellular retinoic acid-binding protein 1**. *Int. J. Obes.* (2022) **46** 1759-1769. DOI: 10.1038/s41366-022-01175-3
10. Lin Y.-L.Y.-W., Nhieu J., Liu P.-Y., Le G., Lee D.J., Wei C.-W., Lin Y.-L.Y.-W., Oh S.-H., Lowe D., Wei L.-N.. **CRABP1-CaMKII-Agrn regulates the maintenance of neuromuscular junction in spinal motor neuron**. *Cell Death Differ.* (2022) **29** 1744-1756. DOI: 10.1038/s41418-022-00959-4
11. Wook Park S., Nhieu J., Persaud S.D., Miller M.C., Xia Y., Lin Y.W., Lin Y.L., Kagechika H., Mayo K.H., Wei L.N.. **A new regulatory mechanism for Raf kinase activation, retinoic acid-bound Crabp1**. *Sci. Rep.* (2019) **9** 10929. DOI: 10.1038/s41598-019-47354-7
12. Persaud S.D., Park S.W., Ishigami-Yuasa M., Koyano-Nakagawa N., Kagechika H., Wei L.N.. **All trans-retinoic acid analogs promote cancer cell apoptosis through non-genomic Crabp1 mediating ERK1/2 phosphorylation**. *Sci Rep* (2016) **6** 22396. DOI: 10.1038/srep22396
13. Park S.W., Nhieu J., Lin Y.W., Wei L.N.. **All-trans retinoic acid attenuates isoproterenol-induced cardiac dysfunction through Crabp1 to dampen CaMKII activation**. *Eur. J. Pharmacol.* (2019) **858** 172485. DOI: 10.1016/j.ejphar.2019.172485
14. Frankel S.R., Eardley A., Lauwers G., Weiss M., Warrell R.P.. **The “retinoic acid syndrome” in acute promyelocytic leukemia**. *Ann. Intern. Med.* (1992) **117** 292-296. DOI: 10.7326/0003-4819-117-4-292
15. Nau H.. **Embryotoxicity and teratogenicity of topical retinoic acid**. *Ski. Pharmacol.* (1993) **6** 35-44. DOI: 10.1159/000211162
16. Bucchia M., Merwin S.J., Re D.B., Kariya S.. **Limitations and challenges in modeling diseases involving spinal motor neuron degeneration in vitro**. *Front. Cell. Neurosci.* (2018) **12** 61. DOI: 10.3389/fncel.2018.00061
17. Bain G., Ray W.J., Yao M., Gottlieb D.I.. **From embryonal carcinoma cells to neurons: The P19 pathway**. *BioEssays* (1994) **16** 343-348. DOI: 10.1002/bies.950160509
18. Gao K., Oerlemans R., Groves M.R.. **Theory and applications of differential scanning fluorimetry in early-stage drug discovery**. *Biophys. Rev.* (2020) **12** 85-104. DOI: 10.1007/s12551-020-00619-2
19. Malo N., Hanley J.A., Cerquozzi S., Pelletier J., Nadon R.. **Statistical practice in high-throughput screening data analysis**. *Nat. Biotechnol.* (2006) **24** 167-175. DOI: 10.1038/nbt1186
20. Luan C.H., Light S.H., Dunne S.F., Anderson W.F.. **Ligand screening using fluorescence thermal shift analysis (FTS)**. *Methods Mol. Biol.* (2014) **1140** 263-289. DOI: 10.1007/978-1-4939-0354-2_20
21. Samuel E.L.G., Holmes S.L., Young D.W.. **Processing binding data using an open-source workflow**. *J. Cheminform.* (2021) **13** 1-11. DOI: 10.1186/s13321-021-00577-1
22. Stratton M.M., Chao L.H., Schulman H., Kuriyan J.. **Structural studies on the regulation of Ca2+/calmodulin dependent protein kinase II**. *Curr. Opin. Struct. Biol.* (2013) **23** 292-301. DOI: 10.1016/j.sbi.2013.04.002
23. Wei L.-N., Lee C.-H.. **Demethylation in the 5′-flanking region of mouse cellular retinoic acid binding protein-I gene is associated with its high level of expression in mouse embryos and facilitates its induction by retinoic acid in P19 embryonal carcinoma cells**. *Dev. Dyn.* (1994) **201** 1-10. DOI: 10.1002/aja.1002010102
24. Omar Faison M., Perozzi E.F., Caran N., Stewart J.K., Tombes R.M.. **Axonal localization of delta Ca2+/calmodulin-dependent protein kinase II in developing P19 neurons**. *Int. J. Dev. Neurosci.* (2002) **20** 585-592. DOI: 10.1016/S0736-5748(02)00107-7
25. Zalcman G., Federman N., Romano A.. **CaMKII isoforms in learning and memory: Localization and function**. *Front. Mol. Neurosci.* (2018) **11** 445. DOI: 10.3389/fnmol.2018.00445
26. Bayer K.U., Löhler J., Schulman H., Harbers K.. **Developmental expression of the CaM kinase II isoforms: Ubiquitous γ- and δ-CaM kinase II are the early isoforms and most abundant in the developing nervous system**. *Mol. Brain Res.* (1999) **70** 147-154. DOI: 10.1016/S0169-328X(99)00131-X
27. Bangaru M.L.Y., Meng J., Kaiser D.J., Yu H., Fischer G., Hogan Q.H., Hudmon A.. **Differential expression of CaMKII isoforms and overall kinase activity in rat dorsal root ganglia after injury**. *Neuroscience* (2015) **300** 116-127. DOI: 10.1016/j.neuroscience.2015.05.007
28. Hoch B., Meyer R., Hetzer R., Krause E.G., Karczewski P.. **Identification and expression of δ-isoforms of the multifunctional Ca2+/calmodulin-dependent protein kinase in failing and nonfailing human myocardium**. *Circ. Res.* (1999) **84** 713-721. DOI: 10.1161/01.RES.84.6.713
29. Elliott J.L.. **Experimental models of amyotrophic lateral sclerosis**. *Neurobiol. Dis.* (1999) **6** 310-320. DOI: 10.1006/nbdi.1999.0266
30. Soundararajan P., Lindsey B.W., Leopold C., Rafuse V.F.. **Easy and Rapid Differentiation of Embryonic Stem Cells into Functional Motoneurons Using Sonic Hedgehog-Producing Cells**. *Stem Cells* (2007) **25** 1697-1706. DOI: 10.1634/stemcells.2006-0654
31. Arber S., Han B., Mendelsohn M., Smith M., Jessell T.M., Sockanathan S.. **Requirement for the homeobox gene Hb9 in the consolidation of motor neuron identity**. *Neuron* (1999) **23** 659-674. DOI: 10.1016/S0896-6273(01)80026-X
32. Barber R.P., Phelps P.E., Houser C.R., Crawford G.D., Salvaterra P.M., Vaughn J.E.. **The morphology and distribution of neurons containing choline acetyltransferase in the adult rat spinal cord: An immunocytochemical study**. *J. Comp. Neurol.* (1984) **229** 329-346. DOI: 10.1002/cne.902290305
33. Tsuchida T., Ensini M., Morton S.B., Baldassare M., Edlund T., Jessell T.M., Pfaff S.L.. **Topographic organization of embryonic motor neurons defined by expression of LIM homeobox genes**. *Cell* (1994) **79** 957-970. DOI: 10.1016/0092-8674(94)90027-2
34. Varela-Echavarría A., Pfaff S.L., Guthrie S.. **Differential expression of LIM homeobox genes among motor neuron subpopulations in the developing chick brain stem**. *Mol. Cell. Neurosci.* (1996) **8** 242-257. DOI: 10.1006/mcne.1996.0061
35. Sharma K., Sheng H.Z., Lettieri K., Li H., Karavanov A., Potter S., Westphal H., Pfaff S.L.. **LIM homeodomain factors Lhx3 and Lhx4 assign subtype identities for motor neurons**. *Cell* (1998) **95** 817-828. DOI: 10.1016/S0092-8674(00)81704-3
36. Thaler J.P., Koo S.J., Kania A., Lettieri K., Andrews S., Cox C., Jessell T.M., Pfaff S.L.. **A Postmitotic Role for Isl-Class LIM Homeodomain Proteins in the Assignment of Visceral Spinal Motor Neuron Identity**. *Neuron* (2004) **41** 337-350. DOI: 10.1016/S0896-6273(04)00011-X
37. Ashpole N.M., Hudmon A.. **Excitotoxic neuroprotection and vulnerability with CaMKII inhibition**. *Mol. Cell. Neurosci.* (2011) **46** 720-730. DOI: 10.1016/j.mcn.2011.02.003
38. Li L., Xiong W.C., Mei L.. **Neuromuscular Junction Formation, Aging, and Disorders**. *Annu. Rev. Physiol.* (2018) **80** 159-188. DOI: 10.1146/annurev-physiol-022516-034255
39. Liu C.M., Hermann T.E.. **Characterization of ionomycin as a calcium ionophore**. *J. Biol. Chem.* (1978) **253** 5892-5894. DOI: 10.1016/S0021-9258(17)34550-7
40. Gwag B.J., Canzoniero L.M.T., Sensi S.L., Demaro J.A., Koh J.Y., Goldberg M.P., Jacquin M., Choi D.W.. **Calcium ionophores can induce either apoptosis or necrosis in cultured cortical neurons**. *Neuroscience* (1999) **90** 1339-1348. DOI: 10.1016/S0306-4522(98)00508-9
41. Gerlier D., Thomasset N.. **Use of MTT colorimetric assay to measure cell activation**. *J. Immunol. Methods* (1986) **94** 57-63. DOI: 10.1016/0022-1759(86)90215-2
42. Kristensen A.S., Jenkins M.A., Banke T.G., Schousboe A., Makino Y., Johnson R.C., Huganir R., Traynelis S.F.. **Mechanism of Ca2+/calmodulin-dependent kinase II regulation of AMPA receptor gating**. *Nat. Neurosci.* (2011) **14** 727-735. DOI: 10.1038/nn.2804
43. Nhieu J., Lin Y.L., Wei L.N.. **CRABP1 in Non-Canonical Activities of Retinoic Acid in Health and Diseases**. *Nutrients* (2022) **14**. DOI: 10.3390/nu14071528
44. Huang P., Chandra V., Rastinejad F.. **Retinoic acid actions through mammalian nuclear receptors**. *Chem. Rev.* (2014) **114** 233-254. DOI: 10.1021/cr400161b
45. Newcomer M.E.. **Structure of the epididymal retinoic acid binding protein at 2.1 Å resolution**. *Structure* (1993) **1** 7-18. DOI: 10.1016/0969-2126(93)90004-Z
46. Newcomer M.E., Pappas R.S., Ong D.E.. **X-ray crystallographic identification of a protein-binding site for both all-trans- and 9-cis-retinoic acid**. *Proc. Natl. Acad. Sci. USA* (1993) **90** 9223-9227. DOI: 10.1073/pnas.90.19.9223
47. Kleywegt G.J., Bergfors T., Senn H., le Motte P., Gsell B., Shud K., Jones T.A., le Motte P., Gsell B., Shudo K.. **Crystal structures of cellular retinoic acid binding proteins I and II in complex with all-trans-retinoic acid and a synthetic retinoid**. *Structure* (1994) **2** 1241-1258. DOI: 10.1016/S0969-2126(94)00125-1
48. Jiang Y.M., Yamamoto M., Kobayashi Y., Yoshihara T., Liang Y., Terao S., Takeuchi H., Ishigaki S., Katsuno M., Adachi H.. **Gene expression profile of spinal motor neurons in sporadic amyotrophic lateral sclerosis**. *Ann. Neurol.* (2005) **57** 236-251. DOI: 10.1002/ana.20379
49. Rizzo F., Nizzardo M., Vashisht S., Molteni E., Melzi V., Taiana M., Salani S., Santonicola P., di Schiavi E., Bucchia M.. **Key role of SMN/SYNCRIP and RNA-Motif 7 in spinal muscular atrophy: RNA-Seq and motif analysis of human motor neurons**. *Brain* (2019) **142** 276-294. DOI: 10.1093/brain/awy330
50. Arundine M., Tymianski M.. **Molecular mechanisms of calcium-dependent neurodegeneration in excitotoxicity**. *Cell Calcium* (2003) **34** 325-337. DOI: 10.1016/S0143-4160(03)00141-6
51. Salazar-Grueso E.F., Kim S., Kim H.. **Embryonic mouse spinal cord motor neuron hybrid cells**. *Neuroreport* (1991) **2** 505-508. DOI: 10.1097/00001756-199109000-00002
52. Golde T.E.. **The therapeutic importance of understanding mechanisms of neuronal cell death in neurodegenerative disease**. *Mol. Neurodegener.* (2009) **4** 1-7. DOI: 10.1186/1750-1326-4-8
53. Kim E.K., Choi E.J.. **Pathological roles of MAPK signaling pathways in human diseases**. *Biochim. Biophys. Acta Mol. Basis Dis.* (2010) **1802** 396-405. DOI: 10.1016/j.bbadis.2009.12.009
54. Ghosh A., Giese K.P.. **Calcium/calmodulin-dependent kinase II and Alzheimer’s disease**. *Mol. Brain* (2015) **8** 1-7. DOI: 10.1186/s13041-015-0166-2
55. Convertini P., Zhang J., de la Grange P., Hayward L.J., Zhu H., Stamm S.. **Genome wide array analysis indicates that an amyotrophic lateral sclerosis mutation of FUS causes an early increase of CAMK2N2 in vitro**. *Biochim. Biophys. Acta Mol. Basis Dis.* (2013) **1832** 1129-1135. DOI: 10.1016/j.bbadis.2013.03.015
56. Picconi B., Gardoni F., Centonze D., Mauceri D., Cenci M.A., Bernardi G., Calabresi P., di Luca M.. **Abnormal Ca2+-calmodulin-dependent protein kinase II function mediates synaptic and motor deficits in experimental parkinsonism**. *J. Neurosci.* (2004) **24** 5283-5291. DOI: 10.1523/JNEUROSCI.1224-04.2004
57. Bai N., Roder H., Dickson A., Karanicolas J.. **Isothermal Analysis of ThermoFluor Data can readily provide Quantitative Binding Affinities**. *Sci. Rep.* (2019) **9** 2650. DOI: 10.1038/s41598-018-37072-x
58. Lee C.H., Chang L., Wei L.N.. **Molecular cloning and characterization of a mouse nuclear orphan receptor expressed in embryos and testes**. *Mol. Reprod. Dev.* (1996) **44** 305-314. DOI: 10.1002/(SICI)1098-2795(199607)44:3<305::AID-MRD4>3.0.CO;2-Q
59. Schindelin J., Arganda-Carreras I., Frise E., Kaynig V., Longair M., Pietzsch T., Preibisch S., Rueden C., Saalfeld S., Schmid B.. **Fiji: An open-source platform for biological-image analysis**. *Nat. Methods* (2012) **9** 676-682. DOI: 10.1038/nmeth.2019
|
---
title: 'The Landscape of Lipid Metabolism in Lung Cancer: The Role of Structural Profiling'
authors:
- Chanchan Hu
- Luyang Chen
- Yi Fan
- Zhifeng Lin
- Xuwei Tang
- Yuan Xu
- Yiming Zeng
- Zhijian Hu
journal: Journal of Clinical Medicine
year: 2023
pmcid: PMC10002589
doi: 10.3390/jcm12051736
license: CC BY 4.0
---
# The Landscape of Lipid Metabolism in Lung Cancer: The Role of Structural Profiling
## Abstract
The aim of this study was to explore the relationship between lipids with different structural features and lung cancer (LC) risk and identify prospective biomarkers of LC. Univariate and multivariate analysis methods were used to screen for differential lipids, and two machine learning methods were used to define combined lipid biomarkers. A lipid score (LS) based on lipid biomarkers was calculated, and a mediation analysis was performed. A total of 605 lipid species spanning 20 individual lipid classes were identified in the plasma lipidome. Higher carbon atoms with dihydroceramide (DCER), phosphatidylethanolamine (PE), and phosphoinositols (PI) presented a significant negative correlation with LC. Point estimates revealed the inverse associated with LC for the n-3 PUFA score. Ten lipids were identified as markers with an area under the curve (AUC) value of 0.947 ($95\%$, CI: 0.879–0.989). In this study, we summarized the potential relationship between lipid molecules with different structural features and LC risk, identified a panel of LC biomarkers, and demonstrated that the n-3 PUFA of the acyl chain of lipids was a protective factor for LC.
## 1. Introduction
Lung cancer (LC) is the second most common cancer and the leading cause of cancer-related death worldwide [1]. Despite several initiatives to control tobacco, there has been no significant downward trend in incidences of LC. There are no or obvious specific symptoms in the early stages of LC, and most people are diagnosed in the mid- to late- stages. This creates a substantial burden for healthcare systems and severely compromises the quality of life of people with LC.
LC has various pathogenic factors and complex pathological mechanisms, but metabolic reprogramming is one of the most important hallmarks of tumor cells. It is commonly found in the process of glucose metabolism, amino acid metabolism, and lipid metabolism, while changes in lipid metabolism have received less attention compared with other topics. Lipids are essential components of the biological membranes and structural units of cells. A comprehensive classification system organizes lipids into these eight well-defined categories: fatty acyls (FA), glycerolipids (GL), glycerophospholipids (GP), sphingolipids (SP), sterol lipids (ST), prenol lipids (PR), saccharolipids (SL), and polyketides (PK) [2]. Lipid metabolism is critical in the development of tumorigenesis [3,4]. Recently, research interests have shifted toward using omics to explore the pathophysiological processes involved in the development of LC, where lipidomics captures changes in endogenous and exogenous molecules that confer further insights [5]. The dysregulation of lipid metabolism is one of the most prominent metabolic alterations in LC which could lead to abnormal gene expression and disorders of signaling pathways [6]. Research on major lung cell types isolated from human donors illustrated the significant role of lipids in lung functions and lung development, including phosphoglycerol (PG), diacylglyceride (DAG), and triacylglyceride (TAG) [7]. In addition, research on lipidomics for LC [3,8,9] has become more common, but it merely provides clues regarding the importance of lipid classes, including phosphatidylcholine (PC), PE, and lysophosphatidylcholine (LPC), in the pathology of non-small cell lung cancer (NSCLC). FAs are at the root of these complex lipids, and their functions and features are mainly determined by their structure, which depends on the number of carbons in the chain (short, medium, long, or extra-long fatty acids) and the number of double bonds (saturated, monounsaturated, and polyunsaturated fatty acid (PUFA)) [10]. Changes in saturated and unsaturated fatty acid levels can disrupt homeostasis in vivo, enhance cellular stress, alter cell membrane dynamics, and affect the uptake and efficacy of chemotherapeutic drugs [11,12]. Due to the structural and biosynthetic complexity of lipids, the results of previous studies are varied [3,13], and the contribution of lipid structural features in the pathogenesis of LC remains unexplored.
The identification of biomarkers or novel metabolic dysregulation pathways has long been a popular field. Thousands of candidate cancer biomarkers have been identified, but only a few are currently used in clinical practice. Lipid metabolism disorders have been found to have great potential for discovering biomarkers and understanding the pathogenesis of LC.
Chronic inflammation is a precondition for the progression of cancers [14]. Although numerous studies [15,16] have reported that inflammation-driven markers, including the neutrophil-lymphocyte ratio (NLR) and the platelet lymphocyte ratio (PLR), play a key role in tumorigenesis and progression, it is unknown whether inflammatory mediators play a role in the pathogenesis of LC caused by lipid metabolism disorders.
Herein, we aim to descriptively summarize the potential relationship between lipid molecules with different structural features and LC risk, identify prospective biomarkers of LC that can be applied in clinical diagnosis and treatment, and further discuss how inflammation mediates the relationship between lipids and LC risk.
## 2.1. Study Population
As part of an ongoing hospital-based case control study, patients were recruited from the Second Affiliated Hospital of Fujian Medical University. The inclusion criteria for the cases were as follows: [1] patients with diagnosed primary LC by fiberoptic bronchoscopy or histopathologic evaluation; [2] patients who have not received chemotherapy; and [3] patients without other lung diseases or systemic diseases, such as heart, liver, kidney, cranial, or brain. The exclusion criteria included patients with a pathologic diagnosis of lung inflammation, benign lesion, or secondary LC. During the same study period, 62 age-matched (±2 years) and sex-matched cancer-free healthy controls (median age: 54 years) were randomly selected from a health examination cohort. All subjects were Han Chinese people who had lived in Fujian for at least 10 years, did not suffer from coronary atherosclerotic heart disease, cerebrovascular disease, thyroid insufficiency, diabetes, or hyperlipidemia, and were able to answer the study questions. All participants provided written and informed consent. The present study was approved by the Second Affiliated Hospital of Fujian Medical University’s Institutional Review Board with the certificate number IRB No. 2021–452.
## 2.2. Chemicals and Reagents
The internal standards were purchased from AB SCIEX (refer to Supplementary Materials File S1 for details). HPLC grade methanol, isopropanol (IPA), acetonitrile (ACN), and water were purchased from Merck (Darmstadt, Germany).
## 2.3. Sample Collection and Preparation
Approximately 10 mL of peripheral venous blood was collected from each study subject under fasting conditions, fractioned by a trained researcher according to standard protocols, and the EDTA plasma was stored at −80 °C in deep freezers. Plasma samples were processed according to the following sequential steps: [1] 225 μL methanol was added to each of the 20 μL plasma samples and then vortexed for 10 s; [2] 13 μL internal standards and 750 μL MTBE were added, vortexed for 10 s, and then stood for 30 min; [3] an additional 188 μL water was added, vortexed for 20 s and then stood for 10 min before centrifugation at 15,000 rpm at 4 °C for 15 min; [4] 700 μL supernatant was collected from each sample separately and blow dried with a nitrogen blower; [5] 100 μL reconstitution reagent ($65\%$ ACN, $30\%$ IPA, and $5\%$ H2O, v:v:v) was added to each dried sample, then vortexed for 10 s, and centrifuged at 14,000 rpm at 4 °C for 10 min; and [6] 100 μL supernatant from each tube was then transferred into vials for HPLC-MS/MS analysis. Before sequence analysis, four consecutive quality control (QC) samples were injected to assess the reproducibility of the system. QC samples were inserted every 10 samples to evaluate the stability of the system in the sequence analysis.
## 2.4. HPLC-MS/MS Analysis
Seminal plasma lipidomics were detected using a HPLC coupled with 4500 QTRAP mass spectrometry (AB SCIEX Pte. Ltd., Framingham, MA, USA). Chromatographic separations were performed on an ACQUITY UPLC BEH HILIC column (1.7 μm, 100 mm × 2.1 mm, 186003461) (Waters). Its parameters are listed in Table S1.
Quantitation was performed on a 4500 QTRAP tandem mass spectrometer coupled with an electrospray ionization (ESI) source (the conditions of the ESI are shown in Table S2). Multiple reaction monitoring (MRM) was used to quantify the compounds in the positive and negative ion modes (Table S3 sheet 1–2 present detailed information about lipids in positive and negative modes). The ion adduct form of each lipid is shown in Table S3 sheet 3.
Throughout the analysis process, the samples were placed in an autosampler and analyzed continuously in a randomized order to avoid fluctuation in the instrument’s detection signal that would affect the experimental results.
## 2.5. Data Processing and Statistical Analysis
Peak identification, peak filtering, peak alignment, and lipid identification were performed using SCIEX OS software to obtain a two-dimensional data matrix, including the mass-to-nucleus ratio, retention time, peak area, and lipid class information. The relative abundance of each lipid was measured while considering the total area of all the transformations that were analyzed. The peak area data from the individual lipids were then transformed and normalized.
Since the relationship between lipids and disease risk may vary depending on the length and unsaturation of the acyl chain [17], lipid subclasses were grouped and further analyzed based on the carbon atom and double bond number. Twenty lipid subclasses were grouped by their total carbon number and total double bond number. The total concentrations of each group in the individual lipid subclasses were log-transformed and standardized into unit variance. Odds ratios (ORs) of LC risk per log-transformed SD increase in each lipid species, based on the number of carbon atoms and double bonds, were calculated using conditional logistic regression models and visualized by bubble plots. To further reveal and clarify the potential biological mechanisms of lipid species based on FA, the n-3 and n-6 PUFA scores among the subclass of lipid profiling were calculated, and the relationships were explored. Univariate and multivariate statistical analyses were used to screen the differential lipids. Univariate statistical analyses included the t-test, fold change (FC) analysis, and volcano plots based on the first two analyses. Multivariate statistical analysis included unsupervised principal component analysis (PCA), supervised partial least squares-discriminant analysis (PLS-DA), orthogonal partial least squares-discriminant analysis (OPLS-DA), and sparse partial least squares-discriminant analysis (sPLS-DA). Then, two machine learning approaches (including the random forest algorithm and support vector machine algorithm) were used to define a combinational lipid biomarker in the plasma samples to distinguish patients with LC from healthy controls. Lipid scores (LS) were calculated by multiplying the OR of the ten biomarkers by their corresponding concentration values. LS = 0.23 × DAG (32:0) + 0.28 × DAG (34:0) + 0.27 × FFA (16:2) + 0.20 × FFA (24:1) + 0.17 × PE (O-38:5) + 0.32 × PC (40:4) + 0.16 × PS (38:6) + 0.30 × TAG (55:2/FA 18:2) + 0.26 × TAG (54:7/FA 18:1) + 0.25 × DAG (40:8). A mediation analysis [18] was performed to test whether the observed associations between LS and LC could be explained by the blood mediator using the medflex package in R. All statistical analyses were performed using R 4.1.1 software. A two-sided p-value < 0.05 was used for this study and considered statistically significant.
## 3.1. Characteristics of the Participants
After matching the sex and age of the cases and controls, a total of 124 participants were enrolled (including 62 cases and 62 controls). The characteristics of the participants with LC and the controls are shown in Table 1. As expected, the participants showed similar characteristics. To broadly explore the distribution of blood indexes between those with LC and the controls, 10 blood indexes were detected, including white blood cell count (WBC), neutrophil count (NEUT), lymphocyte count (LYMPH), monocyte count (MONO), platelet count (PLT), eosinophil count (EOS), NLR, PLR, neutrophil monocyte ratio (NMR), and lymphocyte monocyte ratio (LMR). Compared with the control group, the LC group had higher WBC, MONO, PLT, NLR, PLR, and NMR values (all $p \leq 0.05$) and lower LYMPH and LMR values ($p \leq 0.05$).
## 3.2. Lipid Profiling Grouped by Lipid Structure and Risk of LC
Representative chromatograms of QC and case and control lipids in positive and negative ion mode are shown in Figures S1–S3. The results of unsupervised PCA are shown in Figure S4. A high experiment quality with a large degree of QC clustering was observed in unsupervised PCA. Apparent differences in grouping between the LC and control groups are shown in the PCA score scatter plots for the negative (Figure S4A) and positive (Figure S4B) ionization modes. A total of 605 lipid species spanning 20 individual lipid classes were identified in the plasma lipidome from 124 subjects, which is shown in Figure S4C. A volcano plot representing the levels of the lipids that were upregulated or downregulated in patients with LC compared with the control group is shown in Figure S4D.
An overview of the relationship between 20 lipid subclasses and LC risk is described in Figure S5. Cholesteryl esters (CE), ceramides (CER), DCER, free fatty acid (FFA), hexosylceramide (HCER), PC, LPC, and lysophosphatidylethanolamine (LPE) were positively correlated with LC, while phosphates (PA), PE (O), PE (P) and phosphoserines (PS) were negatively correlated. DAG, LCER, LPG, PE, PG, PI, SM, and TAG were not significantly associated with LC. To clarify the potential biological mechanisms of each lipid with LC, the effect of the lipids based on their chemical structure was obtained using a multivariate conditional logistic regression model which is visualized in Figure 1. The ORs for the individual lipids and their FDR values are plotted in a two-dimensional graph defined by the number of carbon atoms (x-axis) and the number of double bonds (y-axis) in the acyl chain (detailed Supplemental Table S3 Sheet 4 includes the exact OR and FDR-corrected p-values). CE, CER, DCER, FFA, HCER, LPC, LPE, and PC positively related to LC risk, whereas PA, PE, PI and PS were related to lower odds of LC. Higher carbon atoms with DCER, PE, and PI presented a significant, negative correlation with LC. However, we did not find any clear correlation between the parity of the number of double bonds and the carbon atoms with LC risk. Compared with the FFA of unsaturated double bonds, the risk of saturated double bonds positively correlated with LC risk.
Moreover, the n-3 PUFA score and n-6 PUFA scores in the acyl chain were calculated, and point estimates of all lipids revealed an inverse association between LC risk and n-3 PUFA scores and the n-6/n-3 ratios obtained from all the lipid species, while there was a positive association between n-6 PUFA scores and LC risk. These results are shown in Table 2. Significant and negative associations between n-3 PUFA scores and LC risk were observed in DAG, PA, PE, PS, and TAG classes, whereas significant and positive associations were observed between n-6 PUFA scores and LC risk in FFA, LPC, LPE and TAG classes (all $p \leq 0.05$).
## 3.3. Screening for Differential Lipids and the Risk of LC
PLS-DA, OPLS-DA, and sPLS-DA were used to identify differences in lipid profiles between the LC and control groups in positive and negative modes (Figure 2A,C and Figure S6A–H). Compared with OPLS-DA and sPLS-DA, there were remarkable separations with the performance of R2 = 0.816 and Q2 = 0.647 in the positive mode and R2 = 0.908 and Q2 = 0.738 in the negative mode of PLS-DA for the LC and control groups (Figure 2B,D). After combining the FC, t-tests, and VIP from PLS-DA, 36 differential lipid species were extracted with an FC > 2.0, an FDR-corrected p-value < 0.05, and a VIP value > 1.5. The association between the 36 differential lipids and LC risk are shown in Figure S7. When applying 36 differential lipids in KEGG, the pathways of GP metabolism, glycosylphosphatidylinositol-anchor biosynthesis, and GL metabolism were detected (Figure S6I). SVM and random forest algorithms were used to further obtain significant lipid biomarkers between the LC and control groups, and the random forest algorithm had a better predictive accuracy of the biomarker model (Figure 3A,B and Figure S8). A panel of 10 lipid biomarkers was identified by the random forest algorithm and included DAG (34:0), DAG (32:0), FFA (16:2), FFA (24:1), PE (O-38:5), PC (40:4), PS (38:6), TAG (55:2/FA 18:2), TAG (54:7/FA 18:1), and DAG (40:8). Lower odds were observed between DAG (34:0), DAG (32:0), FFA (16:2), FFA (24:1), PE (O-38:5), PC (40:4), PS (38:6), TAG (55:2/FA 18:2), TAG (54:7/FA 18:1), DAG (40:8) and LC risk in the multivariate conditional logistic regression (OR = 0.23, $95\%$ CI: 0.10–0.52; OR = 0.28, $95\%$ CI: 0.14–0.55; OR = 0.20, $95\%$ CI: 0.08–0.49; OR = 0.17, $95\%$ CI: 0.07–0.40; OR = 0.27, $95\%$ CI: 0.14–0.52; OR = 0.32, $95\%$ CI: 0.18–0.55; OR = 0.16, $95\%$ CI: 0.06–0.40; OR = 0.30, $95\%$ CI: 0.16–0.55; OR = 0.26, $95\%$ CI: 0.13–0.51; and OR = 0.25, $95\%$ CI: 0.12–0.49, respectively) (Figure 3D). Notably, there was a high predictive performance of lipid biomarkers with AUC = 0.909 ($95\%$ CI: 0.813–0.984) for 5 lipid species and AUC = 0.96 ($95\%$ CI: 0.909–0.993) for 10 lipid species (Figure 3C). Moreover, an LS based on 10 lipid biomarkers was calculated, and the association between LS and LC risk was investigated (OR = 0.11, $95\%$ CI: 0.06–0.67). As shown in Figure S9, a higher LS was observed in the control group compared with the LC group and showed close and high predictive efficiency (0.96, $95\%$ CI: 0.92–1.00). We also divided the 62 patients into two groups, including 28 in the early-stage group (stages 0 and I) and 34 in the intermediate- and advanced-stage groups (stages II–IV). Similarly, we screened for nine differential lipids between the early and intermediate to late groups and then showed the association between lipid signatures and lung cancer via forest plots (Figure S10).
## 3.4. Mediation Effects for Blood Indexes between Lipid Biomarkers and LC
Figure S9 shows LS and ROC curve analyses for the LC and control groups. Next, the association between blood indexes and lipid biomarkers was investigated (Figure S11). Generally, LMR and LYMPH showed a positive correlation with lipid biomarkers and LS, whereas PLR, NLR, and PLT were negatively associated with lipid biomarkers. In addition, MONO, NEUT, and WBC were negatively correlated with some markers, including LS, TAG (55:2/FA 18:2), TAG (54:7/FA 18:1), PE (O-38:5), PC (40:4), PS (38:6), and DAG (40:8).
A mediation analysis was performed to test whether the observed associations between LS and LC could be explained by a blood mediator. As shown in Table 3, LS had a partial, indirect effect on LS through LMR, LYMPH, MONO, NLR, and PLR with a matched $95\%$ CI that excluded zero. The LMR, LYMPH, MONO, NLR, and PLR mediated proportions were $2.87\%$, $1.89\%$, $2.03\%$, $5.04\%$, and $2.95\%$, respectively.
## 4. Discussion
In the current study, we used targeted HPLC-MS/MS lipidomics and multiple statistical analyses to identify and quantify potentially differential lipid molecules associated with LC and reveal the specific relationships between different lipid structures and LC. Collectively, these findings capture the characteristic metabolic fingerprints of LC patients and elucidate the role of inflammatory mediators between lipids and LC, which adds value to recent studies on lipid metabolic reprogramming. To our knowledge, this is the first relatively comprehensive lipidomics study to explore lipid structural profiles and LC risk. In addition, this study innovatively used inflammatory indicators in blood as mediators to explore the relationships between different lipid molecules and LC.
One ST (CE), one FA (FFA), three SPs (CER, DCER, and HCER), and three GPs (LPC, LPE, and PC) were directly associated with LC risk, while the other four GPs (PA, PE, PI, and PS) were inversely associated with LC risk. The dysregulation of CE metabolism has been demonstrated in many tumor biomarker studies [19,20,21], including studies on LC [22]. A pilot study identified two lipid markers that distinguish squamous cell LC from high-risk individuals with high sensitivity, specificity, and accuracy, including CE (C 18:2) [23]. A study in Germany showed a significant accumulation of free cholesterol and cholesteryl esters within lung tumor tissue, and based on reports of elevated cholesterol levels in cancer cells, strategies to reduce cholesterol synthesis have been suggested as an anti-tumor strategy [24]. SPs are bioactive lipids that are involved in the modulation of cell survival, proliferation, and inflammatory responses, and the SphK/S1P/S1PR (S1P) pathway drives many anti-apoptotic and proliferative processes [25]. The disruption of SP metabolism has been associated with the pathogenesis of LC [6,26]. From the perspective of conversion, lipid-based multi-biomarker panels may capture information on the common etiological mechanisms of LC.
We identified 10 lipids as a fingerprint of patients with LC, and the results showed that 10 promising lipid biomarkers had an AUC range of 0.960. Changes in phospholipid metabolism significantly impact membrane structure, thereby affecting its function, altering key cellular signaling pathways (e.g., cell proliferation and survival), and promoting tumorigenesis [27]. For instance, altered PC and PE membrane content, phospholipid metabolite levels, and FA profiles are commonly recognized as indicators of carcinoma development and progression [28]. PC and PE are phospholipids (crucial components of the cell membrane), and their quantities have been altered in various malignancies [29]. PC could govern cancer cell death, and the blood levels of various PCs were significantly lower in LC patients [30]. PE participates in cell signaling and may control cellular growth and death programs. PE serum levels were higher in participants with malignant LC and decreased after surgical excision of the malignant nodules [31]. In a nested case control study conducted in China, PC and PE-O showed significantly different levels between the LC and control groups and were negatively associated with LC risk [32]. DAG plays a lipid second messenger role used to transduce signals. Few studies have discovered an association between DAG and LC; however, the induction of LC by combining DAG kinase brings etiological implications [33,34]. FFA, which acts as a substrate for cell membrane structures, has been considered an independent risk factor for cancer [35]. The association between FA and LC has been extensively described [36,37]. A previous review discussed the role of fatty acids and their lipid mediators in the apoptosis of cancer cells [38]. Inconsistent study results may be partly due to differences in cancer type, study design, population, sample size, and confounding elimination.
There are findings in our study that deserve particular attention. The association of lipids with LC risk can differ based on acyl chain length and the unsaturation degree; therefore, the lipids were further analyzed based on their carbon atom and double bond numbers. Our results partially corroborated previous observations [39] that DCER, PE, and PI with longer acyl chains were associated with a lower risk of LC. Chen et al. demonstrated that short carbon chain C2-ceramide can effectively sensitize PTX-induced senescence of H1299 cells via p21waf1/cip1- and p16ink4-independent pathways [39]. A previous study [31] suggested that long-chain PE is a potential diagnostic marker for LC. The overexpression of phosphatidylethanolamine-binding protein 4 (PEBP4) in LC regulates tumor progression, invasion, and metastatic potential, which may be partly due to an increase in PEs that act as agonists of PEBP and mediate signal transduction [40]. The relationship between FFA and LC varied at differing degrees of unsaturation. Our results, as well as previous studies [41], support the hypothesis that SFA may be significant in the etiology of LC and deterministic in its development. Saturated lipids are less susceptible to lipid peroxidation, which may protect cancer cells from lipid peroxidation-mediated cell death, and it also alters membrane dynamics and affects the uptake and efficacy of chemotherapeutics [12]. PUFA was found to be important for maintaining cellular function and internal environmental homeostasis, including signal transduction, cell growth, differentiation, and viability [37]. In particular, the levels of n-3 and n-6 PUFAs play an important role in LC risk and progression. Also, n-3 PUFAs contain docosahexaenoic acid (DHA), eicosapentaenoic acid (EPA), and alpha-linolenic acid (ALA). A previous study reported that DHA-PC and EPA-PC significantly inhibited orthotopic tumor growth and lung metastasis, via the activation of PPARγ and the downregulation of the NF-κB pathway to control tumor growth and metastasis [42]. Another study [36] used genome-wide association study (GWAS) data from a Mendelian randomization (MR) approach to demonstrate that n-3 PUFA is a causal protective factor for LC, which is similar to our results. In addition, a mouse experiment [43] showed that mice that were fed a diet rich in n-6 PUFA had significantly increased proliferation, angiogenic and pro-inflammatory markers, and decreased expression of pro-apoptotic proteins in their tumors. Nevertheless, the association between n-6 PUFA, n6/n3 ratios, and LC risk was not found in our study.
In terms of the causal association between lipids and LC, it is plausible that the mediation effects of inflammatory mediators can be partially explained. Previous studies [44,45] have discovered an association between inflammatory mediators and lipid metabolites. Qian et al. [ 46] demonstrated that a key mechanism involved in inflammatory states is GP metabolism, which raises the possibility that phospholipids may act as inflammatory mediators. The polyunsaturated alkenyl-linked fatty acids found in PE(P) and PE(O), together known as plasmalogens, are essential for the storage of precursors as inflammatory mediators, the control of membrane fluidity, and anti-oxidation [47]. LPC plays an important role in mediating inflammation [48] and endothelial cell activation [49]. In addition, systemic inflammatory biomarkers relate to the occurrence and progression of cancer, including LC [50,51,52]. A prospective UK Biobank cohort recruited 440,000 participants and assessed the associations between systemic inflammation markers and risks for 17 cancer sites and revealed that inflammation markers could serve as biomarkers of cancer [53]. In the current study, it was revealed that LMR, LYMPH, MONO, NLR, and PLR mediated $2.87\%$, $1.89\%$, $2.03\%$, $5.04\%$, and $2.95\%$, respectively, of associations between LS and LC risk. Nevertheless, future research must still determine the precise processes underlying the detected connections.
The major strengths in our current study lie in its analytical approaches and well-characterized study design. Our targeted lipidomics approach was constructed upon HPLC-MS/MS, which allowed for the explicit identification and relative quantification of plasma lipids. Two univariate methods, three multivariate methods, and two machine learning methods were used to select differential lipid molecules, and their similarities and differences were compared. To our knowledge, the present study is the first to explore the relationship between lipids, blood inflammatory markers, and LC risk, and it provides a new perspective for future studies. However, the study has several limitations. Selection bias may be present in any hospital-based case control study, but all subjects were recruited according to strict criteria, which may minimize selection bias. We also did not further explore the classification and staging of LC. Therefore, we will consider expanding the study sample, refining the subgroups, and staging of the study population in future studies.
Finally, current immunotherapies have shown remarkable effects for controlling cancer, with the PD-1/PD-L1 axis being one of the most important and well-studied checkpoint pathways in cancer immunity [54]. Immune-related adverse events were found to be closely related to the mechanism by which PD-1/PD-L1 antibodies restarted anti-cancer immune attacks in a lung cancer study by Sun et al. [ 55]. The specific role of lipids in the regulation of the PD-1/PD-L1 axis was revealed by Yang et al. [ 54]. Lipids are key metabolic switches in the immune response [54,56]. The link between lipids and cancer immunity provides an opportunity for future studies to use lipids as biomarkers to evaluate cancer immune responses.
## 5. Conclusions
In summary, the current study provides a comprehensive analysis of the plasma levels of 605 lipids. The findings of this study deepen our knowledge of the pathophysiological mechanisms of LC, highlight the importance of detailed studies on structural differences between various species of lipids in LC research, reveal relationships between different lipid subclass characteristics and LC risk, identify 10 lipid metabolites as potential novel biomarkers of LC risk, and explore associations between lipids and blood inflammatory indicators. In addition, clearly altered lipids were noted to be related to GP metabolism and alpha-linolenic acid metabolism. Our results suggest that lipid profiling may provide novel tools for the research of LC and facilitate the advancement of disease diagnosis and treatment. Blood lipids are promising LC biomarkers that may lead to new treatment strategies.
## References
1. Sung H., Ferlay J., Siegel R.L., Laversanne M., Soerjomataram I., Jemal A., Bray F.. **Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries**. *CA A Cancer J. Clin.* (2021) **71** 209-249. DOI: 10.3322/caac.21660
2. Fahy E., Subramaniam S., Brown H.A., Glass C.K., Merrill A.H., Murphy R.C., Raetz C.R., Russell D.W., Seyama Y., Shaw W.. **A comprehensive classification system for lipids**. *J. Lipid Res.* (2005) **46** 839-861. DOI: 10.1194/jlr.E400004-JLR200
3. Wang G., Qiu M., Xing X., Zhou J., Yao H., Li M., Yin R., Hou Y., Li Y., Pan S.. **Lung cancer scRNA-seq and lipidomics reveal aberrant lipid metabolism for early-stage diagnosis**. *Sci. Transl. Med.* (2022) **14** eabk2756. DOI: 10.1126/scitranslmed.abk2756
4. Lesko J., Triebl A., Stacher-Priehse E., Fink-Neuböck N., Lindenmann J., Smolle-Jüttner F.M., Köfeler H.C., Hrzenjak A., Olschewski H., Leithner K.. **Phospholipid dynamics in ex vivo lung cancer and normal lung explants**. *Exp. Mol. Med.* (2021) **53** 81-90. DOI: 10.1038/s12276-020-00547-x
5. Zhang L., Zhu B., Zeng Y., Shen H., Zhang J., Wang X.. **Clinical lipidomics in understanding of lung cancer: Opportunity and challenge**. *Cancer Lett.* (2020) **470** 75-83. DOI: 10.1016/j.canlet.2019.08.014
6. Meng Q., Hu X., Zhao X., Kong X., Meng Y.M., Chen Y., Su L., Jiang X., Qiu X., Huang C.. **A circular network of coregulated sphingolipids dictates lung cancer growth and progression**. *EBioMedicine* (2021) **66** 103301. DOI: 10.1016/j.ebiom.2021.103301
7. Kyle J.E., Clair G., Bandyopadhyay G., Misra R.S., Zink E.M., Bloodsworth K.J., Shukla A.K., Du Y., Lillis J., Myers J.R.. **Cell type-resolved human lung lipidome reveals cellular cooperation in lung function**. *Sci. Rep.* (2018) **8** 13455. DOI: 10.1038/s41598-018-31640-x
8. Fan Y., Noreldeen H.A.A., You L., Liu X., Pan X., Hou Z., Li Q., Li X., Xu G.. **Lipid alterations and subtyping maker discovery of lung cancer based on nontargeted tissue lipidomics using liquid chromatography-mass spectrometry**. *J. Pharm. Biomed. Anal.* (2020) **190** 113520. DOI: 10.1016/j.jpba.2020.113520
9. Cang S., Liu R., Mu K., Tang Q., Cui H., Bi K., Zhang Y., Li Q.. **Assessment of plasma amino acids, purines, tricarboxylic acid cycle metabolites, and lipids levels in NSCLC patients based on LC-MS/MS quantification**. *J. Pharm. Biomed. Anal.* (2022) **221** 114990. DOI: 10.1016/j.jpba.2022.114990
10. Godoy-Corchuelo J.M., Fernández-Beltrán L.C., Ali Z., Gil-Moreno M.J., López-Carbonero J.I., Guerrero-Sola A., Larrad-Sainz A., Matias-Guiu J., Matias-Guiu J.A., Cunningham T.J.. **Lipid Metabolic Alterations in the ALS&ndash**. *FTD Spectr. Disord.* (2022) **10** 1105
11. Corn K.C., Windham M.A., Rafat M.. **Lipids in the tumor microenvironment: From cancer progression to treatment**. *Prog. Lipid Res.* (2020) **80** 101055. DOI: 10.1016/j.plipres.2020.101055
12. Juloski J., Popović T.B., Debeljak-Martačić J., Ćuk V.V., Milić Perović M.S., Stanković M.S., Trbovich A.M., De Luka S.R.. **Fatty acid in colorectal cancer in adult and aged patients of both sexes**. *J. BUON* (2021) **26** 1898-1907. PMID: 34761598
13. Noreldeen H.A.A., Du L., Li W., Liu X., Wang Y., Xu G.. **Serum lipidomic biomarkers for non-small cell lung cancer in nonsmoking female patients**. *J. Pharm. Biomed. Anal.* (2020) **185** 113220. DOI: 10.1016/j.jpba.2020.113220
14. Liu W., Deng Y., Li Z., Chen Y., Zhu X., Tan X., Cao G.. **Cancer Evo-Dev: A Theory of Inflammation-Induced Oncogenesis**. *Front. Immunol.* (2021) **12** 768098. DOI: 10.3389/fimmu.2021.768098
15. Zhu X., Song H., Chen Y., Han F., Wang Q., Cui Y.. **Neutrophil-to-Lymphocyte Ratio and Platelet-to-Lymphocyte Ratio in Blood to Distinguish Lung Cancer Patients from Healthy Subjects**. *Dis. Markers* (2020) **2020** 8844698. DOI: 10.1155/2020/8844698
16. Chen J.L., Wu J.N., Lv X.D., Yang Q.C., Chen J.R., Zhang D.M.. **The value of red blood cell distribution width, neutrophil-to-lymphocyte ratio, and hemoglobin-to-red blood cell distribution width ratio in the progression of non-small cell lung cancer**. *PLoS ONE* (2020) **15**. DOI: 10.1371/journal.pone.0237947
17. Lu J., Lam S., Wan Q., Shi L., Huo Y., Chen L., Tang X., Li B., Wu X., Peng K.. **High-Coverage Targeted Lipidomics Reveals Novel Serum Lipid Predictors and Lipid Pathway Dysregulation Antecedent to Type 2 Diabetes Onset in Normoglycemic Chinese Adults**. *Diabetes Care* (2019) **42** 2117-2126. DOI: 10.2337/dc19-0100
18. Steen J., Loeys T., Moerkerke B., Vansteelandt S.. **Medflex: An R Package for Flexible Mediation Analysis using Natural Effect Models**. *J. Stat. Softw.* (2017) **76** 1-46. DOI: 10.18637/jss.v076.i11
19. Liu Z., Gomez C.R., Espinoza I., Le T.P.T., Shenoy V., Zhou X.. **Correlation of cholesteryl ester metabolism to pathogenesis, progression and disparities in colorectal Cancer**. *Lipids Health Dis.* (2022) **21** 22. DOI: 10.1186/s12944-022-01629-7
20. Ueno G., Iwagami Y., Kobayashi S., Mitsufuji S., Yamada D., Tomimaru Y., Akita H., Asaoka T., Noda T., Gotoh K.. **ACAT-1-Regulated Cholesteryl Ester Accumulation Modulates Gemcitabine Resistance in Biliary Tract Cancer**. *Ann. Surg. Oncol.* (2022) **29** 2899-2909. DOI: 10.1245/s10434-021-11152-1
21. Wang X., Li H., Zou X., Yan X., Cong P., Li H., Wang H., Xue C., Xu J.. **Deep mining and quantification of oxidized cholesteryl esters discovers potential biomarkers involved in breast cancer by liquid chromatography-mass spectrometry**. *J. chromatogr. A* (2022) **1663** 462764. DOI: 10.1016/j.chroma.2021.462764
22. Hall Z., Wilson C.H., Burkhart D.L., Ashmore T., Evan G.I., Griffin J.L.. **Myc linked to dysregulation of cholesterol transport and storage in nonsmall cell lung cancer**. *J. Lipid Res.* (2020) **61** 1390-1399. DOI: 10.1194/jlr.RA120000899
23. Yu Z., Chen H., Zhu Y., Ai J., Li Y., Gu W., Borgia J.A., Zhang J., Jiang B., Chen W.. **Global lipidomics reveals two plasma lipids as novel biomarkers for the detection of squamous cell lung cancer: A pilot study**. *Oncol. Lett.* (2018) **16** 761-768. DOI: 10.3892/ol.2018.8740
24. Hoppstädter J., Dembek A., Höring M., Schymik H.S., Dahlem C., Sultan A., Wirth N., Al-Fityan S., Diesel B., Gasparoni G.. **Dysregulation of cholesterol homeostasis in human lung cancer tissue and tumour-associated macrophages**. *EBioMedicine* (2021) **72** 103578. DOI: 10.1016/j.ebiom.2021.103578
25. Khan S.A., Goliwas K.F., Deshane J.S.. **Sphingolipids in Lung Pathology in the Coronavirus Disease Era: A Review of Sphingolipid Involvement in the Pathogenesis of Lung Damage**. *Front. Physiol.* (2021) **12** 760638. DOI: 10.3389/fphys.2021.760638
26. Canals D.. **Peeking inside the sphingolipid network in lung cancer**. *EBioMedicine* (2021) **67** 103340. DOI: 10.1016/j.ebiom.2021.103340
27. Baxter A.A., Hulett M.D., Poon I.K.. **The phospholipid code: A key component of dying cell recognition, tumor progression and host-microbe interactions**. *Cell Death Differ.* (2015) **22** 1893-1905. DOI: 10.1038/cdd.2015.122
28. Stoica C., Ferreira A.K., Hannan K., Bakovic M.. **Bilayer Forming Phospholipids as Targets for Cancer Therapy**. *Int. J. Mol. Sci.* (2022) **23**. DOI: 10.3390/ijms23095266
29. Bandu R., Mok H.J., Kim K.P.. **Phospholipids as cancer biomarkers: Mass spectrometry-based analysis**. *Mass Spectrum. Rev.* (2018) **37** 107-138. DOI: 10.1002/mas.21510
30. Guo Y., Ren J., Li X., Liu X., Liu N., Wang Y., Li Z.. **Simultaneous Quantification of Serum Multi-Phospholipids as Potential Biomarkers for Differentiating Different Pathophysiological states of lung, stomach, intestine, and pancreas**. *J. Cancer* (2017) **8** 2191-2204. DOI: 10.7150/jca.19128
31. Fahrmann J.F., Grapov D., DeFelice B.C., Taylor S., Kim K., Kelly K., Wikoff W.R., Pass H., Rom W.N., Fiehn O.. **Serum phosphatidylethanolamine levels distinguish benign from malignant solitary pulmonary nodules and represent a potential diagnostic biomarker for lung cancer**. *Cancer Biomark. Sec. A Dis. Mark.* (2016) **16** 609-617. DOI: 10.3233/CBM-160602
32. Bai Y., Cao Q., Guan X., Meng H., Feng Y., Wang C., Fu M., Hong S., Zhou Y., Yuan F.. **Metabolic linkages between zinc exposure and lung cancer risk: A nested case-control study**. *Sci. Total Environ.* (2022) **837** 155796. DOI: 10.1016/j.scitotenv.2022.155796
33. Cooke M., Zhou X., Casado-Medrano V., Lopez-Haber C., Baker M.J., Garg R., Ann J., Lee J., Blumberg P.M., Kazanietz M.G.. **Characterization of AJH-836, a diacylglycerol-lactone with selectivity for novel PKC isozymes**. *J. Biol. Chem.* (2018) **293** 8330-8341. DOI: 10.1074/jbc.RA117.000235
34. Cooke M., Casado-Medrano V., Ann J., Lee J., Blumberg P.M., Abba M.C., Kazanietz M.G.. **Differential Regulation of Gene Expression in Lung Cancer Cells by Diacyglycerol-Lactones and a Phorbol Ester Via Selective Activation of Protein Kinase C Isozymes**. *Sci. Rep.* (2019) **9** 6041. DOI: 10.1038/s41598-019-42581-4
35. Zhang L., Han L., He J., Lv J., Pan R., Lv T.. **A high serum-free fatty acid level is associated with cancer**. *J. Cancer Res. Clin. Oncol.* (2020) **146** 705-710. DOI: 10.1007/s00432-019-03095-8
36. Liu X., Peng Y., Tao R., Meng L., Li X.. **Mendelian Randomization Study of Causal Relationship between Omega-3 Fatty Acids and Risk of Lung Cancer**. *BioMed Res. Int.* (2022) **2022** 2786567. DOI: 10.1155/2022/2786567
37. Luu H.N., Cai H., Murff H.J., Xiang Y.B., Cai Q., Li H., Gao J., Yang G., Lan Q., Gao Y.T.. **A prospective study of dietary polyunsaturated fatty acids intake and lung cancer risk**. *Int. J. Cancer* (2018) **143** 2225-2237. DOI: 10.1002/ijc.31608
38. Biswas P., Datta C., Rathi P., Bhattacharjee A.. **Fatty acids and their lipid mediators in the induction of cellular apoptosis in cancer cells**. *Prostaglandins Other Lipid Mediat.* (2022) **160** 106637. DOI: 10.1016/j.prostaglandins.2022.106637
39. Chen J.Y.-F., Hwang C.-C., Chen W.-Y., Lee J.-C., Fu T.-F., Fang K., Chu Y., Huang Y.-L., Lin J., Tsai W.H.. **Additive effects of C(2)-ceramide on paclitaxel-induced premature senescence of human lung cancer cells**. *Life Sci.* (2010) **87** 350-357. DOI: 10.1016/j.lfs.2010.06.017
40. Chen Y., Ma Z., Shen X., Li L., Zhong J., Min L.S., Xu L., Li H., Zhang J., Dai L.. **Serum Lipidomics Profiling to Identify Biomarkers for Non-Small Cell Lung Cancer**. *BioMed Res. Int.* (2018) **2018** 5276240. DOI: 10.1155/2018/5276240
41. Reczyńska K., Khanal D., Pielichowska K., Pamuła E., Chrzanowski W.. **Distinct Influence of Saturated Fatty Acids on Malignant and Nonmalignant Human Lung Epithelial Cells**. *Lipids* (2020) **55** 117-126. DOI: 10.1002/lipd.12216
42. Liu Y., Tian Y., Guo Y., Yan Z., Xue C., Wang J.. **DHA-enriched phosphatidylcholine suppressed angiogenesis by activating PPARγ and modulating the VEGFR2/Ras/ERK pathway in human umbilical vein endothelial cells**. *Food Sci. Biotechnol.* (2021) **30** 1543-1553. DOI: 10.1007/s10068-021-00990-0
43. Montecillo-Aguado M., Tirado-Rodriguez B., Antonio-Andres G., Morales-Martinez M., Tong Z., Yang J., Hammock B.D., Hernandez-Pando R., Huerta-Yepez S.. **Omega-6 Polyunsaturated Fatty Acids Enhance Tumor Aggressiveness in Experimental Lung Cancer Model: Important Role of Oxylipins**. *Int. J. Mol. Sci.* (2022) **23**. DOI: 10.3390/ijms23116179
44. Pietzner M., Kaul A., Henning A.K., Kastenmüller G., Artati A., Lerch M.M., Adamski J., Nauck M., Friedrich N.. **Comprehensive metabolic profiling of chronic low-grade inflammation among generally healthy individuals**. *BMC Med.* (2017) **15**. DOI: 10.1186/s12916-017-0974-6
45. Nguyen S., Li H., Yu D., Cai H., Gao J., Gao Y., Luu H., Tran H., Xiang Y., Zheng W.. **Dietary fatty acids and colorectal cancer risk in men: A report from the Shanghai Men’s Health Study and a meta-analysis**. *Int. J. Cancer* (2021) **148** 77-89. DOI: 10.1002/ijc.33196
46. Zhu Q., Wu Y., Mai J., Guo G., Meng J., Fang X., Chen X., Liu C., Zhong S.. **Comprehensive Metabolic Profiling of Inflammation Indicated Key Roles of Glycerophospholipid and Arginine Metabolism in Coronary Artery Disease**. *Front. Immunol.* (2022) **13** 829425. DOI: 10.3389/fimmu.2022.829425
47. Wallner S., Orsó E., Grandl M., Konovalova T., Liebisch G., Schmitz G.. **Phosphatidylcholine and phosphatidylethanolamine plasmalogens in lipid loaded human macrophages**. *PLoS ONE* (2018) **13**. DOI: 10.1371/journal.pone.0205706
48. Xing F., Liu J., Mo Y., Liu Z., Qin Q., Wang J., Fan Z., Long Y., Liu N., Zhao K.. **Lysophosphatidylcholine up-regulates human endothelial nitric oxide synthase gene transactivity by c-Jun N-terminal kinase signalling pathway**. *J. Cell. Mol. Med.* (2009) **13** 1136-1148. DOI: 10.1111/j.1582-4934.2008.00394.x
49. Zhang R., Bai N., So J., Laher I., MacLeod K.M., Rodrigues B.. **The ischemic metabolite lysophosphatidylcholine increases rat coronary arterial tone by endothelium-dependent mechanisms**. *J. Mol. Cell. Cardiol.* (2009) **47** 112-120. DOI: 10.1016/j.yjmcc.2009.03.026
50. Şahin F., Aslan A.F.. **Relationship between Inflammatory and Biological Markers and Lung Cancer**. *J. Clin. Med.* (2018) **7**. DOI: 10.3390/jcm7070160
51. Wang K., Zhao Q., Yan T., Guo D., Liu J., Wang G., Du J.. **The Prognostic Value of Multiple Systemic Inflammatory Biomarkers in Preoperative Patients With Non-small Cell Lung Cancer**. *Front. Surg.* (2022) **9** 830642. DOI: 10.3389/fsurg.2022.830642
52. Stares M., Ding T.E., Stratton C., Thomson F., Baxter M., Cagney H., Cumming K., Swan A., Ross F., Barrie C.. **Biomarkers of systemic inflammation predict survival with first-line immune checkpoint inhibitors in non-small-cell lung cancer**. *ESMO Open* (2022) **7** 100445. DOI: 10.1016/j.esmoop.2022.100445
53. Nøst T.H., Alcala K., Urbarova I., Byrne K.S., Guida F., Sandanger T.M., Johansson M.. **Systemic inflammation markers and cancer incidence in the UK Biobank**. *Eur. J. Epidemiol.* (2021) **36** 841-848. DOI: 10.1007/s10654-021-00752-6
54. Yu Y., Gao L., Wang Y., Xu B., Maswikiti E.P., Li H., Zheng P., Tao P., Xiang L., Gu B.. **A Forgotten Corner in Cancer Immunotherapy: The Role of Lipids**. *Front. Oncol.* (2021) **11**. DOI: 10.3389/fonc.2021.751086
55. Sun X., Roudi R., Dai T., Chen S., Fan B., Li H., Zhou Y., Zhou M., Zhu B., Yin C.. **Immune-related adverse events associated with programmed cell death protein-1 and programmed cell death ligand 1 inhibitors for non-small cell lung cancer: A PRISMA systematic review and meta-analysis**. *BMC Cancer* (2019) **19**. DOI: 10.1186/s12885-019-5701-6
56. Mabrouk N., Lecoeur B., Bettaieb A., Paul C., Végran F.. **Impact of Lipid Metabolism on Antitumor Immune Response**. *Cancers* (2022) **14**. DOI: 10.3390/cancers14071850
|
---
title: Gut-Microbiota Dysbiosis in Stroke-Prone Spontaneously Hypertensive Rats with
Diet-Induced Steatohepatitis
authors:
- Shini Kanezawa
- Mitsuhiko Moriyama
- Tatsuo Kanda
- Akiko Fukushima
- Ryota Masuzaki
- Reina Sasaki-Tanaka
- Akiko Tsunemi
- Takahiro Ueno
- Noboru Fukuda
- Hirofumi Kogure
journal: International Journal of Molecular Sciences
year: 2023
pmcid: PMC10002594
doi: 10.3390/ijms24054603
license: CC BY 4.0
---
# Gut-Microbiota Dysbiosis in Stroke-Prone Spontaneously Hypertensive Rats with Diet-Induced Steatohepatitis
## Abstract
Metabolic-dysfunction-associated fatty-liver disease (MAFLD) is the principal worldwide cause of liver disease. Individuals with nonalcoholic steatohepatitis (NASH) have a higher prevalence of small-intestinal bacterial overgrowth (SIBO). We examined gut-microbiota isolated from 12-week-old stroke-prone spontaneously hypertensive-5 rats (SHRSP5) fed on a normal diet (ND) or a high-fat- and high-cholesterol-containing diet (HFCD) and clarified the differences between their gut-microbiota. We observed that the Firmicute/Bacteroidetes (F/B) ratio in both the small intestines and the feces of the SHRSP5 rats fed HFCD increased compared to that of the SHRSP5 rats fed ND. Notably, the quantities of the 16S rRNA genes in small intestines of the SHRSP5 rats fed HFCD were significantly lower than those of the SHRSP5 rats fed ND. As in SIBO syndrome, the SHRSP5 rats fed HFCD presented with diarrhea and body-weight loss with abnormal types of bacteria in the small intestine, although the number of bacteria in the small intestine did not increase. The microbiota of the feces in the SHRSP5 rats fed HFCD was different from those in the SHRP5 rats fed ND. In conclusion, there is an association between MAFLD and gut-microbiota alteration. Gut-microbiota alteration may be a therapeutic target for MAFLD.
## 1. Introduction
The number of patients with nonalcoholic fatty liver disease (NAFLD), including nonalcoholic steatohepatitis (NASH), has increased over the years [1]. As NASH causes cirrhosis and hepatocellular carcinoma (HCC), NASH is one of the important health issues worldwide. However, an unknown mechanism is also present in the pathogenesis of the development of NASH [2,3].
Fatty liver associated with metabolic dysfunction is common [4,5,6]. Metabolic-dysfunction-associated fatty-liver disease, “MAFLD,” may be a more appropriate overarching term [4,5]. Metabolic-dysfunction-associated fatty-liver disease is the principal worldwide cause of liver disease and affects nearly a quarter of the global population [4,5]. Diagnosis of MAFLD is based on the detection of liver steatosis together with the presence of at least one of three criteria that includes overweight or obesity, type 2 diabetes mellitus, or clinical evidence of metabolic dysfunction, such as an increased waist circumference and an abnormal lipid or glycemic profile [5]. Patients with hepatic steatosis and lean/normal weight is diagnosed as MAFLD in the presence of more than two metabolic risk abnormalities of the following criteria: an increased waist circumference, hypertension, an abnormal lipid or glycemic profile [5]. Patients with NAFLD are at a substantially higher risk of fatal and non-fatal cardiovascular events [6]. NAFLD and cardiovascular disease share multiple common conditions, such as obesity, diabetes, dyslipidemia and hypertension. These diseases may also share multiple common mechanisms, such as dietary habits, smoking, lack of exercise, gut-microbial dysbiosis, and genetics [6].
It has been reported that there is an association between NAFLD/NASH and gut-microbiota [5]. Individuals with NASH have a higher prevalence of small-intestinal bacterial overgrowth (SIBO) [7]. Intestinal mucosa-barrier malfunction may also play a role in NASH [8]. Individuals with NASH have a lower percentage of Bacteroidetes (Bacteroidetes total bacteria counts) than those with simple steatosis or healthy controls [9]. Thus, intestinal bacteria and gut-microbiota dysbiosis may play an important role in the development of NAFLD and NASH [3]. It has also been reported that gut-microbiota dysbiosis is linked to hypertension [10]. The gut-microbiota influence stroke pathogenesis and treatment outcomes [11,12].
Spontaneously hypertensive rats (SHR) and stroke-prone spontaneously hypertensive rats (SHRSP) are well-established parallel lines from outbred Wistar–Kyoto (WKY) rats [13,14]. We previously demonstrated a NASH model using arteriolipidosis-prone rats (ALR; SHRSP5), which are sublines obtained by the feeding of high-fat- and high-cholesterol-containing diets (HFCD) to SHRSP rats [15]. SHRSP5 rats fed HFCD possessed NASH, abnormal lipid, lean body, hypertension, and stroke [13,14,15].
In the present study, we examined the gut-microbiota isolated from stroke-prone spontaneously hypertensive-5 rats (SHRSP5) that were fed a normal diet (ND) or HFCD at 12 weeks of age and clarified the difference between their gut-microbiota. We observed differences between the microbiota of the feces in the SHRSP5 rats fed HFCD and those in the SHRP5 rats fed ND, as well as an increase in the Firmicutes/Bacteroidetes (F/B) ratio, which is a signature of gut dysbiosis, in the microbiota from the small intestine in the SHRSP5 rats fed HFCD. Our observation partially supports the concept of “MAFLD” from the point of view of gut-microbiota dysbiosis.
## 2.1. Quantitative Analysis of 16S Ribosomal RNA Genes of Bacteria of Microbiota
Fecal-pellet DNA was isolated from the 12-week-old SHRSP5 rats fed ND or HFCD for 7 weeks [15]. As previously reported [15], in the HFCD group, pathological findings consistent with NASH were observed; however, in the ND group, only diffuse lipid droplets were seen in the hepatocytes at 12 weeks of age. As one rat died in the HFCD group, its fecal DNA could not be analyzed. At the same time, the DNA contents in the small intestines were also isolated from both groups of rats.
First, we performed real-time PCR to measure the 16S ribosomal (r)RNA genes of the bacteria in the small intestines and feces in both groups of rats (Table 1). We noticed that the quantities of the 16S rRNA genes in the small intestines of the SHRSP5 rats fed HFCD were significantly lower than those of the SHRSP5 rats fed ND ($p \leq 0.05$). However, the DNA from all samples were sufficient for the subsequent analysis.
Thus, HFCD reduced the 16S rRNA genes in the small intestines of the SHRSP5 rat, compared with ND. Interestingly, there may be an association between the reduction in bacteria and the fibrosis of the steatosis of the liver in the SHRSP5 rat fed HFCD. The effects of HFCD intake may be more important for the development of hepatic fibrosis in NASH than SIBO.
## 2.2. Next-Generation Sequencing of the V4–V5 Region of 16S rRNA Genes of Gut-Microbiota
Gut-microbiota dysbiosis is occasionally observed in patients with NASH [16]. The bacterial 16S rRNA gene has been used to define bacterial taxonomy and phylogeny. In order to understand the association between the gut-microbiota and the pathogenesis of NASH, we analyzed the V4–V5 region of the 16S rRNA from the bacteria in the small intestines and feces in the SHRSP5 rats fed ND or HFCD on the Illumina-MiSeq platform. The sequencing-read numbers are shown in Table 2. The sequence-read number ranged from 18,255 to 31,756. In the small intestines, the average sequence-read number of rats fed ND was similar to those of rats fed HFCD (28,819 ± 1944 vs. 29,567 ± 1956; no statistically significant difference). The coverage numbers were in a sequence around ~410 bp. These results indicate successful next-generation sequencing in the present study.
## 2.3. Microbiota of Small Intestine in SHRSP5 Rats Fed ND Are More Similar to Those of Small Intestine or Feces in SHRSP5 Rats Fed HFCD Than to Those of Feces in SHRSP5 Rats Fed ND
Next, we performed weighted UniFrac analyses to calculate the distances between the microbiota populations from the small intestines and feces in the SHRSP5 rats fed with a ND or HFCD [17] (Figure 1A). The microbiota of the small intestines in the SHRSP5 rats fed on ND were more similar to those of the small intestines or feces in the SHRSP5 rats fed on a HFCD than to those of the feces in the SHRSP5 rats fed with ND. The clustering analysis in the ß-diversity analysis of the microbiota populations also supported these results (Figure 1B,C).
A clear separation was observed in the principal-components analysis, clustering analysis, and ß-diversity analysis of the microbiota of the feces between the SHRSP5 rats fed on a HFCD and those fed on a ND (Figure 1A–C). Notably, the microbiota of the feces of the SHRSP5 rats fed an HFCD was different from those of the SHRSP5 rats fed an ND.
## 2.4. The Firmicutes/Bacteroidetes (F/B) Ratio Increased in the Small Intestines of SHRSP5 Rats Fed HFCD Compared to That in SHRSP5 Rats Fed ND
An increase in the Firmicutes/Bacteroidetes (F/B) ratio, caused by an expansion of Firmicutes and/or a contraction of Bacteroidetes, is considered a signature of gut dysbiosis [10]. The F/B ratio in the small intestines of the SHRSP5 rats fed an HFCD increased compared to that of the SHRSP5 rats fed an ND (Figure 2A). The F/B ratio in the feces of the SHRSP5 rats fed with the HFCD tended to increase compared to that of the SHRSP5 rats fed with the ND (Figure 2B).
The F/B ratio in the small intestines of the SHRSP5 rats fed with the HFCD was ~4.6-fold higher than that of the SHRSP5 rats fed the ND (Figure 2A). The F/B ratio in the feces of the SHRSP5 rats fed the HFCD tended to be ~1.7-fold higher than that of the SHRSP5 rats fed the ND (Figure 2B).
In both the small intestines and the feces of SHRSP5 rats fed on an HFCD, the number of both Firmicutes and Bacteroidetes decreased. In the feces of the SHRSP5 rats fed the HFCD, the number of Proteobacteria increased (Figure 3).
In the present study, among the Firmicutes, the Allobaculum decreased in the feces of the SHRSP5 rats fed with a HFCD. The Lactobacillus decreased and the *Streptococcus increased* in the small intestines of the SHRSP5 rats fed the HFCD. The *Clostridium increased* in both the small intestines and the feces of the SHRSP5 rats fed the HFCD. Of the Bacteroides, the Porphyromonadaceae decreased in feces of SHRSP5 rats fed the HFCD. Of the Proteobacteria, the Escerichia increased in both the small intestines and the feces of the SHRSP5 rats fed the HFCD.
## 3. Discussion
In the present study, we examined the gut-microbiota isolated from 12-week-old SHRSP5 rats fed a ND or a HFCD and clarified the differences between their gut-microbiota. We observed that the F/B ratio in both the small intestines and the feces of SHRSP5 rats fed the HFCD increased compared to that of the SHRSP5 rats fed the ND. Notably, the quantity of 16S rRNA genes in the small intestines of the SHRSP5 rats fed the HFCD were significantly lower than those of the SHRSP5 rats fed the ND. The microbiota of the feces of the SHRSP5 rats fed the HFCD was different from those of the SHRSP5 rats fed the ND.
Li et al. reported the ability of *Grifola frondosa* heteropolysaccharide to ameliorate NAFLD in rats fed a high-fat diet (HFD) and significantly increase the proportions of Allobaculum [18]. Increases in Allobaculum can help infant mice resist the development of obesity, according to an investigation of the intestinal microbiota in mice [19]. These reports partially support our observation that Firmicutes and Allobaculum decreased in the feces of the SHRSP5 rats fed the HFCD.
Panasevich et al. reported that soy protein is effective at preventing hepatic steatosis, and an analysis of fecal bacterial 16S rRNA revealed that soy-protein isolate intake elicited increases in Lactobacillus in obese Otsuka Long–Evans Tokushima fatty (OLETF) rats [20]. The rates of *Streptococcus belonging* to Bacilli were significantly increased in rats fed with a high-fat diet [21]. Compared with healthy subjects, NAFLD patients show an increase in the percentage of bacteria of pathogenic Streptococcus [22]. Previous studies [20,21,22] support our observations that the rates of Lactobacillus decreased and those of *Streptococcus increased* in the small intestines of the SHRSP5 rats fed the HFCD.
Individuals with NAFLD might be at increased risk of the development of Clostridioides difficile colitis [23]. Clostridioides difficile colitis can trigger changes associated with the development of NAFLD [24]. In our study, the *Clostridium also* increased in both small intestines and the feces of the SHRSP5 rats fed the HFCD.
High-fat diets result in quantitative alterations in the aerobes (Escherichia coli) in NASH rats [25]. Of the Proteobacteria, the *Escherichia increased* in both the small intestines and the feces of the SHRSP5 rats fed the HFCD. In $37.5\%$ ($\frac{12}{32}$) of the patients with NAFLD, SIBO was present, with *Escherichia coli* as the predominant bacterium [26]. A previous study also demonstrated an increase in the *Escherichia genus* among gut-microbiota in the development and progression of NASH [27,28].
The presence of SIBO decreases small-intestinal movement in NASH rats [25]. A high-fat diet did not increase the anaerobics (Lactobacilli) [25]. Bacteroides species are also anaerobic. In the present study, of the Bacteroides, Porphyromonadaceae decreased in the feces of the SHRSP5 rats fed the HFCD. The presence of SIBO and endotoxemia can result in changes in toll-like receptor (TLR)-signaling gene expression, leading to the development of NAFLD [26]. The abundance of *Bacteroidetes phylum* may be increased, decreased, or unaltered in NASH patients [28].
Thus, SIBO plays a role in the development of NASH pathogenesis [7]. Patients with NASH and those with significant liver fibrosis on liver biopsy had a significantly higher incidence of SIBO than patients without NASH and those without significant liver fibrosis, respectively [29,30]. The onset of NASH in childhood is also a significant health problem [31]. There is an association between NAFLD and SIBO in obese children [32]. SIBO has an effect on the structural and functional characteristics of the liver, resulting in higher insulin and glucose levels, higher neutrophil-to-lymphocyte ratios, and a greater prevalence of NAFLD. A meta-analysis showed a possible association between SIBO and NAFLD in children [33].
The higher the grade of liver steatosis, the higher were the circulating lipopolysaccharide (LPS)-binding protein levels and SIBO rates seen in patients with morbid obesity and NAFLD [34]. The presence of SIBO may enhance intestinal permeability and endotoxemia in NASH patients [35]. Increased endotoxemia may enhance the innate immune response, including TLR-signaling pathways, as well as leading to inflammation and fat deposition in the liver.
The symptoms related to SIBO are bloating, diarrhea, malabsorption, body-weight loss, and malnutrition [36]. SIBO is a heterogeneous syndrome characterized by an increased number and/or abnormal type of bacteria in the small intestine [36]. Notably, the SHRSP5 rats fed with the HFCD presented diarrhea and body-weight loss compared to those fed with the ND [15]; these symptoms were consistent with those of SIBO. In the SHRSP5 rats fed the HFCD, abnormal types of bacteria were observed in the small intestines, although the number of these bacteria did not increase (Table 1). We noticed that the HFCD is more important for the development of hepatic fibrosis in NASH than SIBO.
High-fat diet (HFD)-dependent differences at the phylum, class, and genus levels appear to lead to dysbiosis, characterized by an increase in the F/B ratio, and Firmicutes was the dominant class in a male Sprague-Dawley (SD) rat (7 weeks old) fed HFD with steatohepatitis [37], supporting our observation (Figure 2B). An eight-week treatment of Gegen Qinlian decoction (GGQLD), a well-known traditional Chinese herbal medicine, improved these HFD-induced change [37]. Hugan Qingzhi tablet (HQT), which is a lipid-lowering and anti-inflammatory medicinal formula, has been used to prevent and treat NAFLD and reduced the abundance of the F/B ratio in HFD-fed rats [21]. Curcumin and metformin, which have a therapeutic effect against NAFLD, reduced the F/B ratio and reverted the composition of the HFD-disrupted gut-microbiota in male Sprague–Dawley rats fed HFD [38]. Gut-microbiota can play a role in the pathogenesis of NAFLD, as dysbiosis is associated with reduced bacterial diversity, altered F/B ratio, a relative abundance of alcohol-producing bacteria, or other specific genera [39].
Major risk factors of MAFLD are overweight/obesity, central obesity, type 2 diabetes mellitus, dyslipidemia, arterial hypertension, metabolic syndrome, insulin resistance, dietary factors, lifestyle, and sarcopenia [5]. It is known that gut-microbiota, hyperuricemia, hypothyroidism, sleep apnea syndrome, polycystic ovary syndrome, polycythemia, hypopituitarism, genetic and epigenetic factors, and family history of metabolic syndrome including high blood pressure are common and uncommon risk factors of MAFLD [5].
An association between hypertension and gut-microbiota alteration has been reported [10], as has an association between stroke and gut-microbiota alteration [11,12]. An association between obese and gut-microbiota alteration has also been reported [40], although fecal microbiota transplantation did not reduce body mass index. Evidence for the role of gut-microbiota in metabolic diseases including type 2 diabetes was provided [41]. Human and animal studies indicate the association between diets and hepatic steatosis [42,43]. The association between MAFLD and gut-microbiota alteration should now be clearer given the results of the present study. Dietary factors, such as high-calorie diets with rich saturated fats and cholesterol, soft drinks high in fructose, and highly processed foods, are known to influence the severity of NAFLD. Changing gut-microbiota also does so, at least in part [44]. In the present study, HFCD had an impact on changing gut-microbiota.
We observed an association between NASH and gut-microbiota alteration in the SHRSP5 rats, which originated from the stroke-prone, spontaneously hypertensive rats (SHRSP) fed the HFCD. The recent concept of MAFLD highlights the association between fatty liver disease, hypertension, stroke, and other metabolic diseases. The results from the present study may partially support the association between MAFLD and gut-microbiota alteration. Gut-microbiota alteration may be a therapeutic target for MAFLD. The real interest is how and why the altered microbiota are related to the pathological phenotype. Studies of the associated mechanism should be performed.
The 16S rRNA gene is present in multiple copies in the genomes of bacterial pathogens [45,46]. Therefore, amplicon-sequencing of the bacterium-specific 16S rRNA gene is a useful method for investigating a broad range of bacterial species. However, it is unclear whether the amplicon-sequencing-based detection of the 16S rRNA gene is useful for determining the causative pathogen. A major problem is that the 16S rRNA gene can be amplified not only for meaningful bacteria but also for meaningless bacteria, which is one of the limitations of this study.
Another limitation of the present study is that the number of rats used was small. This was because the present study was an initial study; we will elucidate the mechanisms further in a future study. For example, further improvement of bioinformatics and their analysis, the use of the QIME2 software, which uses amplicon sequence variant (ASV) instead of operational taxonomic unit (out) [47,48,49,50,51], or a denoising step, which allows for obtaining microbial taxa with a higher confidence [52], will be needed.
## 4.1. Animals
This investigation conformed to the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH publication no. 85-23, 1996). The Ethics Committee of Nihon University School of Medicine examined all research protocols involving the use of animals and approved this study (no. 11-034). SHRSP5 rats were obtained from Disease Model Cooperative Research Association (Kyoto, Japan) [13,14]. The SHRSP5 rat is a subline obtained by feeding HFCD to SHRSP rats [53,54]. These SHRSP5 rats are characterized by fat deposition in their arteries, as well as fat deposition in and fibrosis of their livers, indicating the development of diet-induced NASH [15].
## 4.2. Dietary Intervention
The ND group was fed only a stroke-prone (SP) diet. SP diet was purchased as MF from Oriental Yeast Co., Ltd., Itabashi-ku, Tokyo, Japan. The HFCD consisted of $68\%$ (w/w) SP diet, $25\%$ (w/w) palm oil, $5\%$ (w/w) cholesterol, and $2\%$ (w/w) cholic acid [15]. In 100 g of ND, there was approximately 7.9 g water, 23.1 g protein, 5.1 g fat, 5.8 g ash, 2.8 g fiber, 55.3 g soluble without asphyxiation, and 359 kcal, according to the information from Oriental Yeast (https://www.oyc.co.jp/bio/LAD-equipment/LAD/ingredient.html (accessed on 13 February 2023)). The quantities of vitamins A, D3, E, K3, B1, B2, C, B6, B12, inositol, biotin, pantothenic acid, niacin, colin, and folic acid were 1283 IU, 137 IU, 9.1 mg, 0.04 mg, 2.05 mg, 1.1 mg, 4 mg, 0.87 mg, 5.5 mg, 439 mg, 27 μg, 2.45 mg, 10.61 mg, 0.18 g, and 0.17 mg, respectively, in 100 g of ND. We expected each rat to eat ~20 g of the diet daily. Experiments were conducted at least twice for consistent observations.
## 4.3. Sample Collection
Three rats from each group were examined [15]. Their feces were collected for 16S rRNA sequencing analysis. We only gathered the top layers of the feces and performed the isolation under sterile conditions to avoid bacterial contamination. Isoflurane was used as an anesthesia method for sampling the contents of small intestines. Heart blood was collected under general anesthesia; after abdominal median incision, heart blood was collected as described elsewhere [15]. After the incision of perianal, we collected the content of small intestine for further analysis. We performed animal experiments according to the Japanese animal welfare guidelines (https://www.maff.go.jp/j/chikusan/sinko/animal_welfare.html (accessed on 13 February 2023)) at that time.
## 4.4. Quantification of 16S rRNA Genes by Real-Time PCR
The total bacterial genomic DNA was extracted using the Extrap Soil DNA Kit Plus ver.2 (Nippon Steel Corporation, Tokyo, Japan) and stored at −20 °C prior to further analysis. The DNA was used in equal amounts for further PCR analysis.
The total number of bacterial 16S rRNA genes was estimated using a TaqMan-based qPCR approach with primers Bac1055YF, Bac1392R, and Q-probe Bac 1115Probe, which were described previously [55] (Table 3).
## 4.5. Next-Generation Sequencing 16S rRNA Genes
*In* general, 16S and/or internal transcribed spacer ribosomal RNA sequencing are performed for the amplicon sequencing methods to identify and compare the flora of bacteria or fungus of collected samples [50]. This method could identify them after concentrating the original materials using the next-generation sequencing. We performed sequencing analysis of 16S rRNA genes in the present study.
The PCR with high-fidelity-DNA polymerase was used to amplify the V4–V5 region of the 16S rRNA gene with primers U515F and 926R (Table 3). Agilent 2100 bioanalyzer (Agilent technologies, Santa Clara, CA, USA) and PicoGreen dsDNA Assay Kit (Invitrogen, Carlsbad, CA, USA) were used to purify and quantify the resulting PCR amplicons. The Illumina-MiSeq platform (Illumina, San Diego, CA, USA) was used to pool the amplicons in equal amounts and implement the paired-end 2 × 250-base-pair sequencing. Finally, base-pair sequences of ~410 bp were analyzed.
## 4.6. Data Analysis
Standard bioinformatics-alignment comparison was utilized for data analysis [56]. The Quantitative Insights Into Microbial Ecology (QIIME) pipeline was employed to process the sequencing data [16]. Paired-end reads were demultiplexed according to a combination of forward and reverse indices. Additional quality filtering included exact match to sequencing primers and an average quality score of 30 or higher on each read. Prior to further analysis, each paired-end read was stitched into one contiguous read using the fast length adjustment of short reads (FLASH) software tool. Reads that could not be joined were excluded from downstream analysis. All sequences passing filters were aligned against a Silva non-redundant 16S reference database (v108) and assigned taxonomic classifications using USEARCH at a $97\%$ identity threshold. Dereplication to unique reference-sequence-based operational taxonomic units (refOTU) was performed using UCLUST at a $97\%$ clustering threshold and summarized in a refOTU table. Additional alpha-diversity measures and normalized-per-level taxonomic abundances were created using custom scripts written in R [10]. Differentially significant features at each level were identified using linear discriminant analysis (LDA), along with effect-size measurements (LEfSe) [57]. Three-dimensional principal-coordinates analysis (PCoA) plots using the tree-based UniFrac distance metric were generated through custom scripts in R and scripts from the QIIME package [16]. The OUT taxonomic classification was conducted by BLAST, searching the representative sequences set against the database using the best hit, as in previous studies [58]. Classification of bacterial taxonomy based on the end product was performed as previously described [59]. Briefly, genera were classified into more than one group if they were defined as producers of multiple metabolites. Genera that were defined as producing equol, histamine, hydrogen, and propionate constituted only a minor portion of the population and were therefore excluded from this analysis. A representative sequence from each OTU was selected according to the default parameters.
## 5. Conclusions
As in SIBO syndrome, the SHRSP5 rats fed with a HFCD presented diarrhea and body-weight loss with abnormal types of bacteria in their small intestines, although the number of these bacteria did not increase. Our results strongly support the association between MAFLD and gut-microbiota alteration.
## References
1. Henry L., Paik J., Younossi Z.M.. **Review article: The epidemiologic burden of non-alcoholic fatty liver disease across the world**. *Aliment. Pharmacol. Ther.* (2022) **56** 942-956. DOI: 10.1111/apt.17158
2. Kanda T., Matsuoka S., Yamazaki M., Shibata T., Nirei K., Takahashi H., Kaneko T., Fujisawa M., Higuchi T., Nakamura H.. **Apoptosis and non-alcoholic fatty liver diseases**. *World J. Gastroenterol.* (2018) **24** 2661-2672. DOI: 10.3748/wjg.v24.i25.2661
3. Kanda T., Goto T., Hirotsu Y., Masuzaki R., Moriyama M., Omata M.. **Molecular Mechanisms: Connections between Nonalcoholic Fatty Liver Disease, Steatohepatitis and Hepatocellular Carcinoma**. *Int. J. Mol. Sci.* (2020) **21**. DOI: 10.3390/ijms21041525
4. Eslam M., Sanyal A.J., George J.. **MAFLD: A Consensus-Driven Proposed Nomenclature for Metabolic Associated Fatty Liver Disease**. *Gastroenterology* (2020) **158** 1999-2014. DOI: 10.1053/j.gastro.2019.11.312
5. Eslam M., Sarin S.K., Wong V.W., Fan J.G., Kawaguchi T., Ahn S.H., Zheng M.H., Shiha G., Yilmaz Y., Gani R.. **The Asian Pacific Association for the Study of the Liver clinical practice guidelines for the diagnosis and management of metabolic associated fatty liver disease**. *Hepatol. Int.* (2020) **14** 889-919. DOI: 10.1007/s12072-020-10094-2
6. Huang D., Downes M., Evans R., Witztum J., Glass C., Loomba R.. **Shared mechanisms between cardiovascular disease and NAFLD**. *Semin. Liver Dis.* (2022). DOI: 10.1055/a-1930-6658
7. Wigg A.J., Roberts-Thomson I.C., Dymock R.B., McCarthy P.J., Grose R.H., Cummins A.G.. **The role of small intestinal bacterial overgrowth, intestinal permeability, endotoxaemia, and tumour necrosis factor alpha in the pathogenesis of non-alcoholic steatohepatitis**. *Gut* (2001) **48** 206-211. DOI: 10.1136/gut.48.2.206
8. Miele L., Valenza V., La Torre G., Montalto M., Cammarota G., Ricci R., Mascianà R., Forgione A., Gabrieli M.L., Perotti G.. **Increased intestinal permeability and tight junction alterations in nonalcoholic fatty liver disease**. *Hepatology* (2009) **49** 1877-1887. DOI: 10.1002/hep.22848
9. Mouzaki M., Comelli E.M., Arendt B.M., Bonengel J., Fung S.K., Fischer S.E., McGilvray I.D., Allard J.P.. **Intestinal microbiota in patients with nonalcoholic fatty liver disease**. *Hepatology* (2013) **58** 120-127. DOI: 10.1002/hep.26319
10. Yang T., Santisteban M.M., Rodriguez V., Li E., Ahmari N., Carvajal J.M., Zadeh M., Gong M., Qi Y., Zubcevic J.. **Gut dysbiosis is linked to hypertension**. *Hypertension* (2015) **65** 1331-1340. DOI: 10.1161/HYPERTENSIONAHA.115.05315
11. Sharma V., Sharma V., Shahjouei S., Li J., Chaudhary D., Khan A., Wolk D.M., Zand R., Abedi V.. **At the Intersection of Gut Microbiome and Stroke: A Systematic Review of the Literature**. *Front Neurol.* (2021) **12** 729399. DOI: 10.3389/fneur.2021.729399
12. Chidambaram S.B., Rathipriya A.G., Mahalakshmi A.M., Sharma S., Hediyal T.A., Ray B., Sunanda T., Rungratanawanich W., Kashyap R.S., Qoronfleh M.W.. **The Influence of Gut Dysbiosis in the Pathogenesis and Management of Ischemic Stroke**. *Cells* (2022) **11**. DOI: 10.3390/cells11071239
13. Ueno T., Fukuda N., Nagase H., Tsunemi A., Tahira K., Matsumoto T., Hiraoka-Yamamoto J., Ikeda K., Mitsumata M., Sato Y.. **Atherogenic dyslipidemia and altered hepatic gene expression in SHRSP.Z-Leprfa/IzmDmcr rats**. *Int. J. Mol. Med.* (2009) **23** 313-320. DOI: 10.3892/ijmm_00000133
14. Yoshida Y., Fukuda N., Maeshima A., Yamamoto C., Matsumoto T., Ueno T., Nojima Y., Matsumoto K., Soma M.. **Treatment with valsartan stimulates endothelial progenitor cells and renal label-retaining cells in hypertensive rats**. *J. Hypertens.* (2011) **29** 91-101. DOI: 10.1097/HJH.0b013e32834000e2
15. Higuchi T., Moriyama M., Fukushima A., Matsumura H., Matsuoka S., Kanda T., Sugitani M., Tsunemi A., Ueno T., Fukuda N.. **Association of mRNA expression of iron metabolism-associated genes and progression of non-alcoholic steatohepatitis in rats**. *Oncotarget* (2018) **9** 26183-26194. DOI: 10.18632/oncotarget.25488
16. Caporaso J.G., Kuczynski J., Stombaugh J., Bittinger K., Bushman F.D., Costello E.K., Fierer N., Peña A.G., Goodrich J.K., Gordon J.I.. **QIIME allows analysis of high-throughput community sequencing data**. *Nat. Methods.* (2010) **7** 335-336. DOI: 10.1038/nmeth.f.303
17. Lozupone C., Knight R.. **UniFrac: A new phylogenetic method for comparing microbial communities**. *Appl. Environ. Microbiol.* (2005) **71** 8228-8235. DOI: 10.1128/AEM.71.12.8228-8235.2005
18. Li X., Zeng F., Huang Y., Liu B.. **The Positive Effects of Grifola frondosa Heteropolysaccharide on NAFLD and Regulation of the Gut Microbiota**. *Int. J. Mol. Sci.* (2019) **20**. DOI: 10.3390/ijms20215302
19. Cox L.M., Yamanishi S., Sohn J., Alekseyenko A.V., Leung J.M., Cho I., Kim S.G., Li H., Gao Z., Mahana D.. **Altering the intestinal microbiota during a critical developmental window has lasting metabolic consequences**. *Cell* (2014) **158** 705-721. DOI: 10.1016/j.cell.2014.05.052
20. Panasevich M.R., Schuster C.M., Phillips K.E., Meers G.M., Chintapalli S.V., Wankhade U.D., Shankar K., Butteiger D.N., Krul E.S., Thyfault J.P.. **Soy compared with milk protein in a Western diet changes fecal microbiota and decreases hepatic steatosis in obese OLETF rats**. *J. Nutr. Biochem.* (2017) **46** 125-136. DOI: 10.1016/j.jnutbio.2017.05.004
21. Tang W., Yao X., Xia F., Yang M., Chen Z., Zhou B., Liu Q.. **Modulation of the Gut Microbiota in Rats by Hugan Qingzhi Tablets during the Treatment of High-Fat-Diet-Induced Nonalcoholic Fatty Liver Disease**. *Oxid. Med. Cell Longev.* (2018) **2018** 7261619. DOI: 10.1155/2018/7261619
22. Prorok-Hamon M., Friswell M.K., Alswied A., Roberts C.L., Song F., Flanagan P.K., Knight P., Codling C., Marchesi J.R., Winstanley C.. **Colonic mucosa-associated diffusely adherent afaC+ Escherichia coli expressing lpfA and pks are increased in inflammatory bowel disease and colon cancer**. *Gut* (2014) **63** 761-770. DOI: 10.1136/gutjnl-2013-304739
23. Adenote A., Dumic I., Madrid C., Barusya C., Nordstrom C.W., Rueda Prada L.. **NAFLD and Infection, a Nuanced Relationship**. *Can. J. Gastroenterol. Hepatol.* (2021) **2021** 5556354. DOI: 10.1155/2021/5556354
24. Papić N., Jelovčić F., Karlović M., Marić L.S., Vince A.. **Nonalcoholic fatty liver disease as a risk factor for Clostridioides difficile infection**. *Eur. J. Clin. Microbiol. Infect. Dis.* (2020) **39** 569-574. DOI: 10.1007/s10096-019-03759-w
25. Wu W.C., Zhao W., Li S.. **Small intestinal bacteria overgrowth decreases small intestinal motility in the NASH rats**. *World J. Gastroenterol.* (2008) **14** 313-317. DOI: 10.3748/wjg.14.313
26. Kapil S., Duseja A., Sharma B.K., Singla B., Chakraborti A., Das A., Ray P., Dhiman R.K., Chawla Y.. **Small intestinal bacterial overgrowth and toll-like receptor signaling in patients with non-alcoholic fatty liver disease**. *J. Gastroenterol. Hepatol.* (2016) **31** 213-221. DOI: 10.1111/jgh.13058
27. Zhu L., Baker S.S., Gill C., Liu W., Alkhouri R., Baker R.D., Gill S.R.. **Characterization of gut microbiomes in nonalcoholic steatohepatitis (NASH) patients: A connection between endogenous alcohol and NASH**. *Hepatology* (2013) **57** 601-609. DOI: 10.1002/hep.26093
28. De Faria Ghetti F., Oliveira D.G., de Oliveira J.M., de Castro Ferreira L.E.V.V., Cesar D.E., Moreira A.P.B.. **Influence of gut microbiota on the development and progression of nonalcoholic steatohepatitis**. *Eur. J. Nutr.* (2018) **57** 861-876. DOI: 10.1007/s00394-017-1524-x
29. Lira M.M.P., de Medeiros Filho J.E.M., Baccin Martins V.J., da Silva G., de Oliveira Junior F.A., de Almeida Filho É.J.B., Silva A.S., Henrique da Costa-Silva J., de Brito Alves J.L.. **Association of worsening of nonalcoholic fatty liver disease with cardiometabolic function and intestinal bacterial overgrowth: A cross-sectional study**. *PLoS ONE* (2020) **15**. DOI: 10.1371/journal.pone.0237360
30. Mikolasevic I., Delija B., Mijic A., Stevanovic T., Skenderevic N., Sosa I., Krznaric-Zrnic I., Abram M., Krznaric Z., Domislovic V.. **Small intestinal bacterial overgrowth and non-alcoholic fatty liver disease diagnosed by transient elastography and liver biopsy**. *Int. J. Clin. Pract.* (2021) **75** e13947. DOI: 10.1111/ijcp.13947
31. Kanda T., Yokosuka O., Suzuki Y.. **Prolonged hepatitis caused by cytomegalovirus and non-alcoholic steatohepatitis in 16-year-old obese boy**. *Eur. J. Pediatr.* (2005) **164** 212-215. DOI: 10.1007/s00431-004-1584-9
32. Stepanov Y.M., Zavhorodnia N.Y., Yagmur V.B., Lukianenko O.Y., Zygalo E.V.. **Association of nonalcoholic fatty liver disease with small intestine bacterial overgrowth in obese children**. *Wiad. Lek.* (2019) **72** 350-356. DOI: 10.36740/WLek201903107
33. Kuang L., Zhou W., Jiang Y.. **Association of small intestinal bacterial overgrowth with nonalcoholic fatty liver disease in children: A meta-analysis**. *PLoS ONE* (2021) **16**. DOI: 10.1371/journal.pone.0260479
34. Domper Bardají F., Gil Rendo A., Illescas Fernández-Bermejo S., Patón Arenas R., Hernández Albújar A., Martín Dávila F., Murillo Lázaro C., Sánchez Alonso M., Serrano Dueñas M., Sobrino López A.. **An assessment of bacterial overgrowth and translocation in the non-alcoholic fatty liver of patients with morbid obesity**. *Rev. Esp. Enferm. Dig.* (2019) **111** 294-300. DOI: 10.17235/reed.2019.5942/2018
35. Fu X.S., Jiang F.. **Cisapride decreasing orocecal transit time in patients with nonalcoholic steatohepatitis**. *Hepatobiliary Pancreat. Dis. Int.* (2006) **5** 534-537. PMID: 17085338
36. Bures J., Cyrany J., Kohoutova D., Förstl M., Rejchrt S., Kvetina J., Vorisek V., Kopacova M.. **Small intestinal bacterial overgrowth syndrome**. *World J. Gastroenterol.* (2010) **16** 2978-2990. DOI: 10.3748/wjg.v16.i24.2978
37. Guo Y., Ding P.H., Liu L.J., Shi L., Mao T.Y., Li J.X., Wang Y.L.. **Gegen Qinlian Decoction Attenuates High-Fat Diet-Induced Steatohepatitis in Rats via Gut Microbiota**. *Evid. Based Complement Alternat. Med.* (2018) **2018** 7370891. DOI: 10.1155/2018/7370891
38. Li R., Yao Y., Gao P., Bu S.. **The Therapeutic Efficacy of Curcumin vs. Metformin in Modulating the Gut Microbiota in NAFLD Rats: A Comparative Study**. *Front Microbiol.* (2021) **11** 555293. DOI: 10.3389/fmicb.2020.555293
39. Di Ciaula A., Bonfrate L., Portincasa P.. **The role of microbiota in nonalcoholic fatty liver disease**. *Eur. J. Clin. Investig.* (2022) **52** e13768. DOI: 10.1111/eci.13768
40. Allegretti J.R., Kassam Z., Mullish B.H., Chiang A., Carrellas M., Hurtado J., Marchesi J.R., McDonald J.A.K., Pechlivanis A., Barker G.F.. **Effects of Fecal Microbiota Transplantation With Oral Capsules in Obese Patients**. *Clin. Gastroenterol. Hepatol.* (2020) **18** 855-863. DOI: 10.1016/j.cgh.2019.07.006
41. Gurung M., Li Z., You H., Rodrigues R., Jump D.B., Morgun A., Shulzhenko N.. **Role of gut microbiota in type 2 diabetes pathophysiology**. *EBioMedicine* (2020) **51** 102590. DOI: 10.1016/j.ebiom.2019.11.051
42. Cohen J.C., Horton J.D., Hobbs H.H.. **Human fatty liver disease: Old questions and new insights**. *Science* (2011) **332** 1519-1523. DOI: 10.1126/science.1204265
43. Vancells Lujan P., Viñas Esmel E., Sacanella Meseguer E.. **Overview of Non-Alcoholic Fatty Liver Disease (NAFLD) and the Role of Sugary Food Consumption and Other Dietary Components in Its Development**. *Nutrients* (2021) **13**. DOI: 10.3390/nu13051442
44. Leung C., Rivera L., Furness J.B., Angus P.W.. **The role of the gut microbiota in NAFLD**. *Nat. Rev. Gastroenterol. Hepatol.* (2016) **13** 412-425. DOI: 10.1038/nrgastro.2016.85
45. Brosius J., Dull T.J., Sleeter D.D., Noller H.F.. **Gene organization and primary structure of a ribosomal RNA operon from Escherichia coli**. *J. Mol. Biol.* (1981) **148** 107-127. DOI: 10.1016/0022-2836(81)90508-8
46. Enomoto H., Inoue S.I., Matsuhisa A., Iwata Y., Aizawa N., Sakai Y., Takata R., Ikeda N., Hasegawa K., Nakano C.. **Amplification of bacterial genomic DNA from all ascitic fluids with a highly sensitive polymerase chain reaction**. *Mol. Med. Rep.* (2018) **18** 2117-2123. DOI: 10.3892/mmr.2018.9159
47. Pataky Z., Genton L., Spahr L., Lazarevic V., Terraz S., Gaïa N., Rubbia-Brandt L., Golay A., Schrenzel J., Pichard C.. **Impact of Hypocaloric Hyperproteic Diet on Gut Microbiota in Overweight or Obese Patients with Nonalcoholic Fatty Liver Disease: A Pilot Study**. *Dig. Dis. Sci.* (2016) **61** 2721-2731. DOI: 10.1007/s10620-016-4179-1
48. Delik A., Dinçer S., Ülger Y., Akkız H., Karaoğullarından Ü.. **Metagenomic identification of gut microbiota distribution on the colonic mucosal biopsy samples in patients with non-alcoholic fatty liver disease**. *Gene* (2022) **833** 146587. DOI: 10.1016/j.gene.2022.146587
49. Schloss P.D.. **Amplicon Sequence Variants Artificially Split Bacterial Genomes into Separate Clusters**. *mSphere* (2021) **6** e0019121. DOI: 10.1128/mSphere.00191-21
50. Jin B.T., Xu F., Ng R.T., Hogg J.C.. **Mian: Interactive web-based microbiome data table visualization and machine learning platform**. *Bioinformatics* (2022) **38** 1176-1178. DOI: 10.1093/bioinformatics/btab754
51. Chiarello M., McCauley M., Villéger S., Jackson C.R.. **Ranking the biases: The choice of OTUs vs. ASVs in 16S rRNA amplicon data analysis has stronger effects on diversity measures than rarefaction and OTU identity threshold**. *PLoS ONE* (2022) **17**. DOI: 10.1371/journal.pone.0264443
52. Serrana J.M., Watanabe K.. **Sediment-associated microbial community profiling: Sample pre-processing through sequential membrane filtration for 16S rRNA amplicon sequencing**. *BMC Microbiol.* (2022) **22**. DOI: 10.1186/s12866-022-02441-0
53. Kitamori K., Naito H., Tamada H., Kobayashi M., Miyazawa D., Yasui Y., Sonoda K., Tsuchikura S., Yasui N., Ikeda K.. **Development of novel rat model for high-fat and high-cholesterol diet-induced steatohepatitis and severe fibrosis progression in SHRSP5/Dmcr**. *Environ. Health Prev. Med.* (2012) **17** 173-182. DOI: 10.1007/s12199-011-0235-9
54. Moriya T., Kitamori K., Naito H., Yanagiba Y., Ito Y., Yamagishi N., Tamada H., Jia X., Tsuchikura S., Ikeda K.. **Simultaneous changes in high-fat and high-cholesterol diet-induced steatohepatitis and severe fibrosis and those underlying molecular mechanisms in novel SHRSP5/Dmcr rat**. *Environ. Health Prev. Med.* (2012) **17** 444-456. DOI: 10.1007/s12199-012-0273-y
55. Ritalahti K.M., Amos B.K., Sung Y., Wu Q., Koenigsberg S.S., Löffler F.E.. **Quantitative PCR targeting 16S rRNA and reductive dehalogenase genes simultaneously monitors multiple Dehalococcoides strains**. *Appl. Environ. Microbiol.* (2006) **72** 2765-2774. DOI: 10.1128/AEM.72.4.2765-2774.2006
56. Lightfoot Y.L., Yang T., Sahay B., Zadeh M., Cheng S.X., Wang G.P., Owen J.L., Mohamadzadeh M.. **Colonic immune suppression, barrier dysfunction, and dysbiosis by gastrointestinal bacillus anthracis Infection**. *PLoS ONE* (2014) **9**. DOI: 10.1371/journal.pone.0100532
57. Segata N., Izard J., Waldron L., Gevers D., Miropolsky L., Garrett W.S., Huttenhower C.. **Metagenomic biomarker discovery and explanation**. *Genome Biol.* (2011) **12** R60. DOI: 10.1186/gb-2011-12-6-r60
58. DeSantis T.Z., Hugenholtz P., Larsen N., Rojas M., Brodie E.L., Keller K., Huber T., Dalevi D., Hu P., Andersen G.L.. **Greengenes, a chimera-checked 16S rRNA gene database and workbench compatible with ARB**. *Appl. Environ. Microbiol.* (2006) **72** 5069-5072. DOI: 10.1128/AEM.03006-05
59. Antharam V.C., Li E.C., Ishmael A., Sharma A., Mai V., Rand K.H., Wang G.P.. **Intestinal dysbiosis and depletion of butyrogenic bacteria in clostridium difficile infection and nos-ocomial diarrhea**. *J. Clin. Microbiol.* (2013) **51** 2884-2892. DOI: 10.1128/JCM.00845-13
|
---
title: The Impact of Semicarbazide Sensitive Amine Oxidase Activity on Rat Aortic
Vascular Smooth Muscle Cells
authors:
- Vesna Manasieva
- Shori Thakur
- Lisa A. Lione
- Anwar R. Baydoun
- John Skamarauskas
journal: International Journal of Molecular Sciences
year: 2023
pmcid: PMC10002598
doi: 10.3390/ijms24054946
license: CC BY 4.0
---
# The Impact of Semicarbazide Sensitive Amine Oxidase Activity on Rat Aortic Vascular Smooth Muscle Cells
## Abstract
Semicarbazide-sensitive amine oxidase (SSAO) is both a soluble- and membrane-bound transmembrane protein expressed in the vascular endothelial and in smooth muscle cells. In vascular endothelial cells, SSAO contributes to the development of atherosclerosis by mediating a leukocyte adhesion cascade; however, its contributory role in the development of atherosclerosis in VSMCs has not yet been fully explored. This study investigates SSAO enzymatic activity in VSMCs using methylamine and aminoacetone as model substrates. The study also addresses the mechanism by which SSAO catalytic activity causes vascular damage, and further evaluates the contribution of SSAO in oxidative stress formation in the vascular wall. SSAO demonstrated higher affinity for aminoacetone when compared to methylamine (Km = 12.08 µM vs. 65.35 µM). Aminoacetone- and methylamine-induced VSMCs death at concentrations of 50 & 1000 µM, and their cytotoxic effect, was reversed with 100 µM of the irreversible SSAO inhibitor MDL72527, which completely abolished cell death. Cytotoxic effects were also observed after 24 h of exposure to formaldehyde, methylglyoxal and H2O2. Enhanced cytotoxicity was detected after the simultaneous addition of formaldehyde and H2O2, as well as methylglyoxal and H2O2. The highest ROS production was observed in aminoacetone- and benzylamine-treated cells. MDL72527 abolished ROS in benzylamine-, methylamine- and aminoacetone-treated cells (**** $p \leq 0.0001$), while βAPN demonstrated inhibitory potential only in benzylamine-treated cells (* $p \leq 0.05$). Treatment with benzylamine, methylamine and aminoacetone reduced the total GSH levels (**** $p \leq 0.0001$); the addition of MDL72527 and βAPN failed to reverse this effect. Overall, a cytotoxic consequence of SSAO catalytic activity was observed in cultured VSMCs where SSAO was identified as a key mediator in ROS formation. These findings could potentially associate SSAO activity with the early developing stages of atherosclerosis through oxidative stress formation and vascular damage.
## 1. Introduction
Semicarbazide-sensitive amine oxidase (SSAO) is a copper-rich amine oxidase encoded by the amine oxidase copper containing 3 (Aoc3) gene, and it exists as both a soluble- and membrane-bound transmembrane protein, also known as vascular adhesion protein 1 (VAP-1). Soluble SSAO is a result of the proteolytic cleavage of the membrane-bound VAP-1. During this process, anchored molecules are released into the bloodstream by shedding from the membrane through a metalloproteinase-dependent activity [1,2]. SSAO converts primary amines into their corresponding aldehydes while also generating hydrogen peroxide and ammonia. Being a vascular enzyme SSAO is highly expressed in the vascular endothelial and in smooth muscle cells. In endothelial cells, SSAO is localized in the intracellular/cytoplasmic vesicles and its activity in these cells is associated with the development of atherosclerosis, as it induces a leukocyte adhesion cascade into damaged inflammatory sites [3,4]. In smooth muscle cells, SSAO is localized in the caveolae of the plasma membrane. Previously [5], we demonstrated SSAO to be closely associated with another vascular enzyme, lysyl oxidase (LOX), whose alterations in activity and expression have been linked with the early developing stages of atherosclerosis [6,7]. Moreover, we have demonstrated LOX as a regulator of SSAO activity, VAP-1 protein and Aoc3 mRNA expression in early passage rat aortic VSMCs, highlighting SSAO as an important novel therapeutic target for the treatment/prevention of atherosclerosis [5].
Atherosclerosis is defined as a progressive and complex inflammatory disease that develops because of disturbed vascular homeostasis caused by endothelial injury [8]. A lipid profile is an important pathological factor for the development of atherosclerosis. Elevated low-density lipoprotein (LDL) cholesterol and elevated triglyceride-rich lipoproteins (TGRL) and low high-density lipoprotein (HDL) now comprise a major pattern of lipid abnormality in atherosclerosis [9]. Since the development of atherosclerosis is based on metabolic changes in lipid metabolism, major sex-based differences in cholesterol metabolism have been shown to contribute to differences in the pathogenesis of this disease [10]. An SSAO-mediated increase in free radicals provoke an oxidative modification from low-density lipoprotein (LDL) to oxidized low-density lipoprotein (oxLDL) in the vascular wall, which is an important step in the early development stages of atherosclerosis [3,4].
SSAO substrates are aromatic and aliphatic monoamines. They are produced endogenously or absorbed as dietary or xenobiotic substances [11]. Methylamine and aminoacetone are SSAO specific substrates, endogenously produced as short-chain primary amines, and oxidatively deaminated by SSAO to formaldehyde and methylglyoxal. Methylamines have been shown to enhance atherosclerosis in animal models [12] and, in clinical studies, have been associated with cardiovascular risks [13,14,15]. Furthermore, the toxic effects of formaldehyde and methylglyoxal have been widely implicated in cardiovascular pathologies [16,17,18,19]. Both formaldehyde and methylglyoxal are highly reactive aldehydes capable of cross-linking with proteins following a pseudo-first order kinetic [20]. Unlike free radicals, SSAO-derived aldehydes are more stable. This high stability enables methylglyoxal and formaldehyde to diffuse easily and attack intracellular targets that are distant from the point of origin [21].
Early studies have shown that methylamine does not harm endothelial cells at concentrations of up to 100 mmol/L [22]; however, in the presence of SSAO, methylamine has been demonstrated to be cytotoxic due to formaldehyde formation [22]. Formaldehyde can induce cell death by interacting with macromolecular constituents, thus altering cellular structures. Additionally, it has been shown to be a main apoptotic inducer in vascular endothelial cells [23]. The formaldehyde-induced apoptosis in A7r5 cells was detected by chromatin condensation, caspase-3 activation, PARP cleavage and cytochrome c release [23]. In another study, a formaldehyde-driven expression of the pro-apoptotic protein p53 was shown to potentially be an additional mechanism through which formaldehyde induces apoptosis [24].
Methylglyoxal is a highly reactive aldehyde. It is also a powerful modifying agent of proteins and DNA and can act as a mediator in the synthesis of advanced glycation end products [25]. It has been previously established that by modifying proteins, as well as forming oxygen free radicals, methylglyoxal can act as a cytotoxic agent and induce apoptosis in cells [25]. Moreover, it contributes to the formation of advanced glycation end product (AGE) by modifying cell proteins non-enzymatically through the Maillard reaction, in which aldehydes and ketones react with ε-amino groups of lysine residues and guanidino groups of arginine residues [25].
H2O2 is another by-product of an SSAO-catalyzed reaction; if produced above normal physiological levels (1–100 nM), it acts as important contributor to oxidative stress [26]. Recent studies [18,19] have identified H2O2 as a source of reactive oxygen species (ROS) that can modify low-density lipoprotein (LDL) in the arterial wall and contribute to the development of atherosclerosis. Other studies have addressed H2O2 as a vasoactive agent with the ability to induce vasoconstriction of resistance vessels and increase vascular tone. Therefore, it can contribute to the development of hypertension [27,28].
Furthermore, an increased H2O2 production because of enhanced SSAO activity could initiate a signaling cascade which leads to an increased expression of inflammatory cytokines and adhesion molecules in the vascular wall and as such accelerate endothelial damage [29]. H2O2 is metabolized by catalase and glutathione peroxidase and when produced in large amounts, in the presence of transition metals (particularly iron) it can be converted to toxic hydrogen free hydroxyl radical (•OH) via the Fenton reaction (H2O2 + Fe2+ → •OH + OH– + Fe3+) [30]. Hydroxyl radicals pose a greater risk comparing to H2O2 and can directly injure cell membranes and nuclei [31].
SSAO is highly implicated in the pathophysiology of various cardiovascular diseases (CVD), including stroke, myocardial infarction and atherosclerosis, as well as health risks associated with CVD, such as obesity and diabetes [18,19,32,33,34]. Apoptotic VSMCs and increased ROS levels are important hallmarks in the early developing stages of atherosclerosis [9]. Being abundantly present in the vasculature, SSAO is a relatively novel enzymatic discovery associated with cytotoxicity and elevated ROS levels, through production of highly unstable and reactive aldehydes and H2O2. This study investigates the role of SSAO in the early developing stages of atherosclerosis by exploring its enzymatic activity, cytotoxic effects and contribution in oxidative stress formation in rat aortic VSMCs, using its respective substrates and inhibitors.
## 2.1. Active SSAO Induces VSMCs Death
To assess SSAO activity in rat aortic VSMCs, cells were treated with increasing concentrations of SSAO’s endogenous amines, aminoacetone (AA) and methylamine (M). Figure 1 shows the effect of AA and M on VSMC cell viability after 24 h treatment, and the suppressive effect of SSAO’s irreversible inhibitor MDL72527 on cytotoxicity induced by both amines. Moreover, 0 µM of aminoacetone and methylamine were considered as a vehicle only control. Aminoacetone at 50 µM caused $15\%$ and at 100 µM resulted with $30\%$ cell death comparing to control (Figure 1A). Methylamine at 1000 µM caused $40\%$ cell death comparing to the control (Figure 1B).
## 2.2. Enhanced Cytotoxic Effect Was Observed after Simultaneous Addition of Methylglyoxal and H2O2, and Formaldehyde and H2O2
To investigate direct cytotoxic effects from SSAO’s derived products and potential synergism between them, VSMCs were treated with methylglyoxal (MG), formaldehyde (F) and H2O2. Figure 2 shows the effect of MG, F and H2O2 on VSMC cell viability after 24 h treatment. MG (50 µM) and F (1000 µM) caused 30–$40\%$ cell death comparing to control. H2O2 at 50 µM caused $30\%$ and at 1000 µM $40\%$ cell death comparing to control (Figure 2A,B). MG and H2O2 combined caused $70\%$ cell death, with F and H2O2 causing $60\%$ cell death above control (Figure 2A,B).
## 2.3. SSAO Has Higher Affinity for Aminoacetone Comparing to Methylamine and Converts Both at a Fast Rate in Rat Aortic VSMCs
To understand the level of interaction between SSAO and its endogenous amines, SSAO kinetic parameters were assessed in the presence of aminoacetone and methylamine as substrates. Figure 3 shows SSAO’s reaction rate (nmol H2O2/h) as function of aminoacetone (A) or methylamine (B) concentration. SSAO demonstrates higher affinity for aminoacetone comparing to methylamine (12.08 µM vs. 65.35 µM), as observed in the Km values, and converts both amines at a fast rate (5 nmol/min for aminoacetone vs. 4 nmol/min for methylamine), as observed in the Vmax values (Figure 3A,B).
## 2.4. SSAO Activity Induces ROS Formation in Rat Aortic VSMCs
To observe the effect of SSAO on ROS formation in VSMCs and establish an optimal time to measure ROS cells were treated with SSAO substrate amines benzylamine, methylamine and aminoacetone over different time intervals. Further, 5 μM (20 μL/well) DMNQ was used like a positive control. Figure 4A shows gradual increase of ROS production after 5 μM DMNQ treatment over different time intervals. Figure 4B shows the detected ROS production (expressed as percentage of control) versus time (*) and versus different amine treatments (#).
ROS levels were measured once again after 30 min of incubation with benzylamine, methylamine and aminoacetone in the presence of SSAO’s irreversible inhibitor MDL72527 and SSAO’s competitive reversible inhibitor βAPN. Figure 5 shows the detected ROS formation (expressed as percentage of control) after 30 min treatment with benzylamine, methylamine and aminoacetone in the presence of MDL72527 or βAPN respectively.
ROS formation was also observed with cell imaging. Figure 6 shows cultured VSMCs stained with ROS red-staining solution to image ROS production. Figure 6A shows ROS production in VSMCs treated with PBS–negative control. Figure 6B shows ROS production in VSMCs treated with DMNQ–positive control. Figure 6C,D shows ROS production after benzylamine treatment (B) and benzylamine in the presence of MDL72527 (B + MDL72527). Figure 6E,F shows ROS production after methylamine treatment (M) and methylamine in the presence of MDL72527 (M + MDL72527). Figure 6G,H shows ROS production after aminoacetone treatment (A), and aminoacetone in the presence of MDL72527 (A + MDL72527).
## 2.5. SSAO Activity Reduces Total GSH Levels in Rat Aortic VSMCs
To assess whether SSAO driven ROS production reduces total glutathione (GSH) levels, GSH (nM/mg protein) was detected with a colorimetric recycling assay based on the glutathione recycling system by DTNB (Ellman’s reagent) and glutathione reductase. Figure 7 shows the detected GSH after benzylamine, methylamine and aminoacetone treatment, and after treatment with the amines in the presence of SSAO’s irreversible inhibitor MDL72527 and SSAO’s competitive reversible inhibitor βAPN. Cells in culture medium—without treatment, were considered as control.
## 3. Discussion
Apoptotic VSMCs and increased ROS levels are distinctive features in the early developing stages of atherosclerotic plaque formation. This study investigates the role of SSAO in the early developing stages of atherosclerosis by exploring its enzymatic activity, cytotoxic effects and contribution in oxidative stress formation in rat aortic VSMCs, using its respective substrates and inhibitors. Our findings show induced VSMCs death after 24 h exposure to 50 & 100 µM aminoacetone, and 1000 µM methylamine, and reversed cytotoxicity after addition of 100 µM irreversible SSAO inhibitor MDL72527, which completely abolished cell death (Figure 1A,B). This suggests that the cytotoxic effects observed here are a consequence of the deamination of methylamine and aminoacetone, a reaction catalyzed by SSAO. A similar effect was observed in previous studies in which another irreversible SSAO inhibitor, MDL-72974A, reversed formaldehyde- [22] and methylglyoxal-induced [35] cell death by inhibiting the deamination of their respective substrates, methylamine and aminoacetone.
Cellular concentrations of methylamine are estimated as <1 mM [36]. Aminoacetone has previously been suggested as cytotoxic at concentrations of ≥100 µM [37]. When the cellular levels of aminoacetone and methylamine reach higher than their physiological range, these amines have been reported to induce cell death in human aortic VSMCs and insulin-producing cells [23,37,38]. SSAO’s driven aminoacetone cytotoxic effect was observed in insulin-producing RINm5f cells where aminoacetone with concentrations 100 and 500 µM reduced cell viability [37]. Our data agrees with these findings; however, it also signifies SSAO driven cytotoxic effect at lower aminoacetone concentrations (50 µM) in rat aortic VSMCs (Figure 1A). This indicates higher VSMCs vulnerability to aminoacetone driven cytotoxic effects, which could be attributed to higher SSAO expression in VSMCs in comparison to pancreatic B cells. In another study, methylamine-induced toxicity was observed at 1 mM in human aortic smooth muscle cells because of SSAO mediated deamination [23]. This was confirmed by observing Caspase-3 activation, PARP cleavage and cytochrome c release [23]. Our data is congruent with this finding by demonstrating a methylamine-driven cytotoxicity at 1 mM in rat aortic VSMCs (Figure 1B).
Furthermore, our data shows direct cytotoxic effect induced by SSAO generated aldehydes (methylglyoxal and formaldehyde) and H2O2 (Figure 2A,B). Cytotoxic effects of methylglyoxal and formaldehyde have been previously indicated in endothelial but not in VSMCs [39,40]. Methylglyoxal activity in VSMCs has been associated with the production of advanced glycation end products, such as argpyrimidine [25].
Formaldehyde has been demonstrated as a main inducer of covalent binding between functional groups in lysine residues of protein, and DNA base in rat endothelial cells [39], and methylglyoxal has been shown to induce human umbilical vein endothelial cell death at concentrations of 400–800 µM by downregulating cell cycle associated genes and upregulating the heme-oxygenase 1 (HO-1) [40]. Cellular concentrations for methylglyoxal are between 1–5 µM [21], and for formaldehyde 200–500 µM [41]. Furthermore, physiological range of H2O2 in the cell is between 1 and 100 nM and high concentrations of 1 and 2 mM can induce cell death by increasing DNA protein crosslinks [39]. Interestingly, it has previously been postulated that the SSAO catalyzed reaction produces equal molar concentrations of cytotoxic aldehyde and H2O2 and that these by-products act in synergism in inducing cell damage and death [39]. Our study shows cellular toxicity at 50 µM methylglyoxal, 1000 µM formaldehyde, and 50 and 1000 µM H2O2, and enhanced cytotoxic effect after simultaneous addition of H2O2 and aldehydes, which indicates additive rather than synergistic relationship between the same (Figure 2A,B).
After identifying a safe concentration range at which SSAO endogenous amines failed to exert cytotoxic effect on rat aortic VSMCs (Figure 1A,B), we performed additional studies to understand the level of interaction between SSAO and these amines. Our data shows higher SSAO affinity for aminoacetone compared to methylamine, as observed in the Km values, and faster SSAO driven oxidative deamination of aminoacetone compared to methylamine, as observed in the Vmax values (Figure 3A,B). SSAO kinetic parameters have been previously investigated in rat aortic A7r5 cells after addition of methylamine, benzylamine and tyramine as substrates [23]. In contrast to our data, this study demonstrated higher SSAO Vmax (7.32 nmol/min) and smaller SSAO affinity for methylamine (1.04 mM) [23]. The reason for this could be that our study used primary cell line, with a focus on the membrane bound form of the enzyme and not soluble SSAO. Furthermore, our data shows fast generation of methylglyoxal and formaldehyde, because of SSAO catalyzed reaction (Figure 3A,B). The fast generation of methylglyoxal and formaldehyde could damage cell membranes due to auto-oxidation of lipids and fatty acids within the cell [21].
To assess the contribution of SSAO in oxidative stress formation, ROS levels were measured after treatment with SSAO’s respective substrates (benzylamine, methylamine and aminoacetone) and inhibitors (βAPN and MDL72527). This was further correlated with changes in total GSH content. The cytotoxic and ROS formation ability of SSAO derived by-products has been previously highlighted in other studies [42,43,44]. Methylglyoxal was previously shown to increases ROS through AGEs formation [43].
In another study, methylglyoxal-driven ROS was demonstrated as a crucial mechanism for methylglyoxal-induced cytotoxicity in brain endothelial cells, as it suppressed the Akt/hypoxia-inducible factor 1 alpha (HIF-1α) pathway [44]. Interestingly, previous studies have detected synergism between formaldehyde and free radicals in increasing oxidative stress levels and reducing cell viability [42]. Our data shows a significant difference in ROS production after 15-, 30-, 60- and 120-min incubation with the amines, in comparison to the control DMNQ (Figure 4B). Since ROS are defined as relatively short-lived molecules, 30 min was selected as an optimal time to measure ROS production. This is because the detected ROS after methylamine treatment at 30 min was higher in comparison to 15 min, and there was not a significant difference in benzylamine and aminoacetone driven ROS production between 15 and 30 min (Figure 4B).
Our data shows the highest ROS production after aminoacetone treatment (45 μM), followed by benzylamine (500 μM) and then methylamine (500 μM) (Figure 4B and Figure 5). Aminoacetone is catalytically deaminated to methylglyoxal through SSAO-driven enzymatic reaction. ROS formation has previously been associated with methylglyoxal in vascular endothelial cells [44] and pancreatic beta cells [45]. In vascular endothelial cells, methylglyoxal treatment was shown to increase mitochondrial and total cellular ROS formation [44]. In pancreatic beta cells, methylglyoxal treatment was shown to increase mitochondrial ROS and stimulate overproduction of advanced glycation end products (AGEs) [45]. Since methylglyoxal is an aldehyde produced through SSAO catalyzed reaction in which aminoacetone is oxidatively deaminated to aldehyde (methylglyoxal), our findings correlate with these studies and associate SSAO activity with mitochondrial ROS production in VSMCs.
Benzylamine is also a substrate for lysyl oxidase (LOX), another amine oxidase abundantly present in the VSMCs. While MDL72527 is a specific suicide inhibitor for SSAO, βAPN is a suicide inhibitor for LOX [5], and a competitive reversible inhibitor for SSAO. This explains why a significant reduction in ROS was detected between benzylamine, benzylamine- and βAPN-treated cells (Figure 5). However, the benzylamine-driven ROS reduction after βAPN treatment was smaller ($30\%$) in comparison to the inhibition induced by MDL72527 ($80\%$). Therefore, this data suggests that the ROS detected here is predominantly SSAO driven.
Furthermore, our data shows no significant reduction in ROS in methylamine- and aminoacetone-treated cells after βAPN treatment, and a significant reduction in ROS in methylamine-treated cells ($50\%$) and aminoacetone-treated cells ($90\%$) after MDL72527 treatment (Figure 5). Additionally, the comparison between the two different inhibitor treatments in reducing ROS distinguished MDL72527 as more potent ROS reducing agent in comparison to βAPN (Figure 5), which once again prioritize SSAO over other vascular enzymes, such as LOX in ROS formation. Furthermore, Figure 6 confirms the potency of MDL72527 in inhibiting ROS production in benzylamine-, methylamine- and aminoacetone-treated cells.
ROS levels were correlated with total GSH production (nM/mg protein) in VSMCs previously treated with benzylamine, methylamine, aminoacetone and the substrate amines in the presence of MDL72527 and βAPN. This is because GSH is the main antioxidant that reduces hydrogen peroxide through glutathione peroxidase (GPx) catalyzed reactions [46]. Exposure to ROS could reduce total GSH through its oxidation during which levels of oxidized GSH (GSSG) are increased as a defense mechanism of the cells to counteract ROS [47]. Moreover, hydroxyl radicals could lead to direct oxidation of GSH and consequently GSSG formation. Our data shows significant reduction in total GSH after benzylamine, methylamine and aminoacetone treatment (Figure 7). These findings complement the data from Figure 5, where significant increase in ROS was observed after aminoacetone, benzylamine and methylamine treatment. GSH is an important intracellular antioxidant and, thus, reduction in its levels are paralleled with the generation of different ROS including hydroxyl radicals, superoxide anions, hydrogen peroxide and lipid peroxide [48]. Since the amines used here are specific SSAO substrates (apart from benzylamine that is also deaminated by LOX), and hydrogen peroxide is a by-product of SSAO catalyzed reaction, our study suggests that active SSAO contributes to reduced GSH in rat aortic VSMCs, because of ROS formation.
In contrast to Figure 5, where MDL72527 significantly reduced ROS and its inhibitory potential over ROS was more potent than βAPN, the GSH data does not show significant restoration of total GSH after MDL72527 or βAPN treatment (Figure 7). Previous studies have dissociated the relationship between ROS and GSH by demonstrating that a reduction in GSH is a necessary contributing factor for ROS generation; however, inhibition of ROS by antioxidants does not necessarily restore GSH levels, which indicates independence from the generation of ROS [48].
## 4.1. Reagents
Cell culture reagents were purchased from Fisher Scientific (Loughborough, UK). Unless otherwise stated, chemicals and reagents were purchased from Sigma-Aldrich (Poole, UK).
## 4.2. Animals
The rat model was used due to being closely similar with humans in terms of aortic SSAO activity [49]. Indistinguishable levels of SSAO activity have been previously detected in human and rat arteries (human 2.56 nmol benzaldehyde/min/mg protein vs. rat 2.84 nmol benzaldehyde/min/mg protein) [49]. Male Wistar rats (180–220 g) were housed in pairs in standard cages (Tecniplast 2000P) with sawdust (dates and grade 7 substrate) and shredded paper wool bedding with water and food (5LF2 $10\%$ protein LabDiet) in the Biological Services Unit at the University of Hertfordshire. The housing environment was maintained at a constant temperature (21 ± 20 °C) and a light-dark cycle (12:12 h). All experiments were carried out in accordance with the University of Hertfordshire animal welfare ethical guidelines and European directive $\frac{2010}{63}$/EU and all tissues collected were naïve shared within teaching/research in accordance with the 3Rs.
## 4.3. Cells
The aortic VSMCs were selected due to expressing high levels of SSAO. This study used primary VSMCs because primary cell cultures most closely represent the tissue of origin [50].
## 4.4. Isolation and Characterisation of Rat Aortic VSMCs
VSMCs were isolated from the rat’s aorta, as per standard protocol, which consists of five steps: isolation of the aortic artery, removal of the fat tissue around the artery, cutting the artery into small tissue blocks, transferring the tissue blocks to cell culture flask and incubation until the cells reach confluency [50].
The rats were euthanized by exposure to carbon dioxide gas in a rising concentration. The aorta was removed and placed in a Dulbecco’s modified eagle medium (DMEM) solution supplemented with $10\%$ Fetal Bovine Serum (FBS (v/v)), $1\%$ penicillin (100 units mL−1), streptomycin (100 μg mL−1) and 2 mM L-Glutamine. The aorta was cleaned 3 times with 1× phosphate buffer saline (PBS) and the fat tissue around the artery was removed. The artery was then cut longitudinally, and the intima was softly scrapped to eliminate endothelial cells.
The artery was fixed by pressing it dorsally with a pair of ophthalmic curved tweezers and another pair of ophthalmic curved tweezers was used to separate the media from the artery by pressing and pushing the artery dorsally. The media was then cut into small tissue blocks and transferred to T25 cell culture flask containing Dulbecco’s Modified Eagle’s Medium (DMEM; Gibco®, Waltman, MA, USA), supplemented with $10\%$ Fetal Bovine Serum (FBS (v/v)), $1\%$ penicillin (100 units mL−1), streptomycin (100 μg mL−1) and 2 mM L-Glutamine. To characterize the cells, the isolated rat VSMCs were stained for the smooth muscle cell marker SM22α, as per standard protocol [51]. Please see Appendix A for isolated and characterized VSMCs images.
## 4.5. Cell Viability Assay
Cell viability was determined with the MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) tetrazolium reduction assay, as previously described [52]. Primary rat aortic VSMCs were plated at 5 × 104 in a 96-well plate and allowed to grow for 24–48 h to reach confluence. Confluent cells were pre-treated with different concentrations of aminoacetone or methylamine (dissolved in serum free DMEM), with and without the presence of 100 µM irreversible SSAO inhibitor MDL72527 (Sigma, St. Louis, MO, USA, M2949) for 24 h in a CO2 incubator ($5\%$ CO2 and $95\%$ humidified air) at 37 °C. After incubation with the amines, MTT solution (5 mg/mL) was added to each well, and the plate was incubated for additional 4 h in a CO2 incubator ($5\%$ CO2 and $95\%$ humidified air) at 37 °C. After incubation with MTT the media was removed, and the formazan crystals were dissolved by adding 200 µL isopropanol. The MTT assay was also utilized to determine cell viability after addition of aldehydes and H2O2. Confluent cells were pre-treated with equal concentrations of methylglyoxal (50 µM), H2O2, or, methylglyoxal + H2O2, as well as equal concentrations of formaldehyde (1000 µM), H2O2 or formaldehyde + H2O2, (dissolved in serum free DMEM) for 24 h in a CO2 incubator ($5\%$ CO2 and $95\%$ humidified air) at 37 °C before adding MTT solution (5 mg/mL) to each well, followed by further 4 h incubation in a CO2 incubator ($5\%$ CO2 and $95\%$ humidified air) at 37 °C. After incubation with MTT, the media was removed and the formazan crystals were dissolved by adding 200 µL isopropanol. The plates were then wrapped in a foil and placed on a shaker for 15 min. The quantity of formazan was directly proportional to the number of viable cells was measured by recording changes in absorbance at 570 nm, using a spectrophotometric Clario Star® Microplate Reader (BMG Labtech, Ortenberg, Germany).
## 4.6. Amplex Red Assay
After establishing non-toxic amine concentrations, SSAO kinetic parameters were assessed in the presence of methylamine and aminoacetone as substrates, with the Amplex® red assay previously optimized to detect SSAO activity in this cell type. Rat aortic VSMCs with confluency of ~80–$90\%$ were treated with reaction mixture containing 20 μL Amplex® Red, 10 μL horseradish peroxidase (HRPO) and 10 μL clorgyline, supplemented with 0.25 M sodium phosphate buffer at pH 7.4, and different concentrations of methylamine or aminoacetone as substrates. SSAO activity was measured 6 h from the addition of the reaction mixture using excitation 540 nm and emission 590 nm on a Clario Star® Microplate Reader (BMG Labtech). Resorufin was used to measure end-point fluorescence. Next, 2 mM of resorufin stock solution was diluted to a concentration of 1000 μM in a 1× reaction buffer (2 mL of 5× reaction buffer (0.25 M sodium phosphate at pH 7.4), 10 mL distilled water) to yield resorufin standards ranging from 0 to 20 μM. The data was transferred and analyzed on an Excel spreadsheet before preparing a standard curve of resorufin fluorescence (RFU) versus concentration (μM). To express SSAO activity in nmol H2O2/mL, the fluorescence readings from different time intervals were multiplied by the slope and added by the intercept (both calculated from the linear equation of the resorufin standard curve derived after 1 h incubation with resorufin standards). To express SSAO activity in nmol H2O2/h/mg protein, the nmol H2O2/mL values were divided over the protein concentration (mg/mL), which was previously calculated using the Bicinchoninic acid (BCA) assay.
The data for each methylamine and aminoacetone substrate concentration was transferred to an Excel spreadsheet and analyzed before plotting SSAO activity (nmol H2O2/h/mg protein) against time (h). The reaction velocity (V) expressed as (nmol H2O2/h) was derived from the slope of the linear part of the progress curve from the SSAO activity (nmol H2O2/h/mg protein) vs. time graph for each substrate concentration. SSAO’s kinetics (Km and Vmax) were determined by plotting reaction velocity (nmol H2O2/h) versus substrate concentration using the non-linear regression model of Michaelis–Menten Y = Vmax ∗ X/(Km + X) on Graph Pad Prism 7 software (version 7.05, San Diego, CA, USA).
## 4.7. Reactive Oxygen Species (ROS) Assay
Prior to measuring the ROS initial experiment was first performed to establish the optimal time for ROS measurement. Rat aortic vascular smooth muscle cells were plated at 3 × 104 cells/100 μL in a black 96-well plate and allowed to grow for 24–48 h to reach confluence. A ROS red-staining solution was prepared by adding 15 μL of ROS red dye to a 10 mL assay buffer. Confluent cells were washed with 1× PBS and ROS red-staining solution (80 μL) was added to each well before incubation at 37 °C/$5\%$ CO2 for 1 h.
Afterward, incubation cells were treated with different SSAO substrates, including benzylamine (500 μM), methylamine (500 μM) and aminoacetone (45 μM), all of which were previously diluted in 1× PBS. 1× PBS (10 μL/well) was used for untreated cells and 5 μM (20 μL/well) 2,3-dimethoxy-1,4-naphthguinone (DMNQ) was used like a positive control. DMNQ is a redox cycling agent that generates both superoxide and hydrogen peroxide intracellularly; it does not react with free thiol groups, is non alkylating and non-adduct forming in contrast to other quinones [53]. To induce ROS production, cells were incubated at 37 °C, and the reading was taken after 15, 30, 60 and 120 min using a Clario Star® Microplate Reader (BMG Labtech) with Ex/Em = $\frac{520}{605}$ nm. The plate was kept in incubator at 37 °C/$5\%$ CO2 between readings.
In the subsequent set of experiments, confluent cells (after previously been incubated for 1 h at 37 °C/$5\%$ CO2 with ROS red dye) were treated with benzylamine (500 μM), methylamine (500 μM),and aminoacetone (45 μM), with and without the presence of reversible competitive inhibitor of SSAO, β-aminopropionitrile (βAPN) (200 μM), or with and without the presence of mechanism-based, suicide inhibitor of SSAO, MDL72527 (100 μM). 1× PBS (10 μL/well) was used for untreated cells and 5 μM (20 μL/well) of 2,3-dimethoxy-1,4-naphthguinone (DMNQ) was used like a positive control. The cells were incubated for 30 min at 37 °C/$5\%$ CO2 and the readings were taken using Clario Star® Microplate Reader (BMG Labtech) with Ex/Em = $\frac{520}{605}$ nm.
## 4.8. Measurement of Total Glutathione (GSH)
Total GSH was assessed with a colorimetric recycling assay based on the glutathione recycling system by DTNB (Ellman’s reagent) and glutathione reductase [54]. Rat aortic VSMCs were plated at 5 × 105 cells/1 mL/well in a 24-well plate and allowed to grow for 24–48 h to reach confluence. Confluent cells were treated with benzylamine (500 μM), methylamine (500 μM) and aminoacetone (45 μM), with and without the presence of βAPN (200 μM), or with and without the presence of MDL72527 (100 μM) before incubation at 37 °C/$5\%$ CO2 for 30 min. Samples were prepared by washing the cells with sterile 1× PBS, scrapping, and centrifugation at 700× g for 5 min at 4 °C, after which the pellet was washed with 0.5 mL 1× PBS and centrifuged again at 700× g for 5 min at 4 °C. The pellet was then lysed with 80 μL ice-cold glutathione buffer and incubated on ice for 10 min, after which 20 μL of $5\%$ sulfosalicylic acid (SSA) was added, mixed well, and centrifuged again at 8000× g for 10 min. The supernatant was transferred to a fresh centrifuge tube and kept on ice ready to be used for the glutathione assay. Reaction mixture was prepared with: NADPH generating mix 20 μL/well, glutathione reductase 20 μL/well and glutathione reaction buffer 120 μL/well.
Furthermore, 1 mL of $1\%$ SSA was added to GSH standard to generate 1 μg/μL glutathione solution, which was then further diluted with $1\%$ SSA to generate 10 ng/μL stock. Next, 10 ng/μL GSH stock was used to prepare GSH standards. Moreover, 160 μL of the reaction mixture was added to each well and the plate was incubated at room temperature for 10 min to generate NADPH. After 10 min of incubation, 20 μL of either GSH standards or samples was added to each well containing reaction mixture and the plate was incubated at room temperature for another 10 min. Next, 20 μL of DTNB was added to each well containing GSH standards and samples and the plate was incubated at room temperature for another 10 min. The absorbance was read on a Clario Star® plate reader (BMG Labtech) and set at 412 nm. The data was transferred to an Excel spreadsheet and analyzed before plotting the absorbance ratio (412 nm) versus concentration of GSH standards (ng/µL). Total GSH was calculated as follows: Total GSH = (Abs sample − Abs blank)/slope STD curve. These values were then corrected for total protein concentration by subtracting them with the protein values obtained from the BCA assay (previously performed) and the total GSH content was expressed as nmol of GSH per mg of total cellular protein.
## 4.9. Statistical Analysis
The data was analyzed with the statistical software GraphPad Prism 7 (San Diego, CA, USA). Statistical comparisons were made using one or two-way ANOVA, followed by Dunnett’s or Tukey’s multiple comparison tests. Additionally, SSAO kinetics were analyzed with the non-linear regression model of Michaelis-Menten Y = Vmax ∗ X/(Km + X). Probability values < 0.05 were considered as being statistically significant.
## 5. Conclusions
These findings could potentially associate SSAO catalytic activity with the early developing stages of atherosclerosis and vascular damage through induced cellular toxicity, increased ROS levels and a reduction in total GSH. Furthermore, this data shows that methylglyoxal and formaldehyde generate quickly in rat aortic VSMCs because of the SSAO catalyzed reaction. We also noted a higher SSAO affinity for aminoacetone compared to methylamine, which indicates a greater production of methylglyoxal compared to formaldehyde in these cells.
Additional in vitro transcriptional and biochemical studies are needed to fully explore associated signaling pathways related to cellular toxicity and increase in ROS levels. This would provide insight into the potential mechanisms of these transduction pathways involved in the up- and downstream of SSAO’s catalytic activity. Furthermore, considering the different pathways that could lead to reduced cell viability, future in vitro studies are needed to investigate VSMCs phenotype, while focusing on cellular contraction capacity, inflammation and cellular senescence. In addition, in vivo studies using different gender ApoEd/d or LDLR−/− mice fed with a high fat diet and treated with MDL72527 would strongly corroborate the translational significance of the results presented here and shed more light on the sex-related issues of pathogenesis of atherosclerosis.
## References
1. Abella A., Garcia-Vicente S., Viguerie N., Ros-Baro A., Camps M., Palacin M., Zorzano A., Marti L.. **Adipocytes release a soluble form of VAP-1/SSAO by a metalloprotease-dependent process and in a regulated manner**. *Diabetologia* (2004) **47** 429-438. DOI: 10.1007/s00125-004-1346-2
2. Stolen C.M., Yegutkin G.G., Kurkijärvi R., Bono P., Alitalo K., Jalkanen S.. **Origins of Serum Semicarbazide-Sensitive Amine Oxidase**. *Circ. Res.* (2004) **95** 50-57. DOI: 10.1161/01.RES.0000134630.68877.2F
3. Boomsma F., Hut H., Bagghoe U., Van Der Houwen A., Meiracker A.V.D.. **Semicarbazide-sensitive amine oxidase (SSAO): From cell to circulation**. *Med. Sci. Monit.* (2005) **11** RA122-RA126. PMID: 15795708
4. Pannecoeck R., Serruys D., Benmeridja L., Delanghe J.R., van Geel N., Speeckaert R., Speeckaert M.M.. **Vascular adhesion protein 1: Role in human pathology and application as a biomarker**. *Crit. Rev. Clin. Lab. Sci.* (2015) **52** 284-300. DOI: 10.3109/10408363.2015.1050714
5. Manasieva V., Thakur S., Lione L.A., Patel J., Baydoun A., Skamarauskas J.. **Semicarbazide-sensitive amine oxidase (SSAO) and Lysyl oxi-dase (LOX) association in rat aortic vascular smooth muscle cells**. *Biomolecules* (2022) **11**. DOI: 10.3390/biom12111563
6. Rodríguez C., Alcudia J.F., Martínez-González J., Raposo B., Navarro M.A., Badimon L.. **Lysyl oxidase (LOX) down-regulation by TNFα: A new mechanism underlying TNFα-induced endothelial dysfunction**. *Atherosclerosis* (2004) **28** 558-564. DOI: 10.1016/j.atherosclerosis.2007.06.002
7. Rodríguez C., Martínez-González J., Raposo B., Alcudia J.F., Guadall A., Badimon L.. **Regulation of lysyl oxidase in vascular cells: Lysyl oxidase as a new player in cardiovascular diseases**. *Cardiovasc. Res.* (2008) **79** 7-13. DOI: 10.1093/cvr/cvn102
8. Gimbrone M.A., Garcia-Cardena G.. **Endothelial cell dysfunction and the pathobiology of atherosclerosis**. *Circ. Res.* (2016) **18** 620-636. DOI: 10.1161/CIRCRESAHA.115.306301
9. Libby P.. **The changing landscape of atherosclerosis**. *Nature* (2021) **52** 524-533. DOI: 10.1038/s41586-021-03392-8
10. Vakhtangadze T., Singh Tak R., Singh U., Baig M.S., Bezsonov E.. **Gender Differences in Atherosclerotic Vascular Disease: From Lipids to Clinical Outcomes**. *Front. Cardiovasc. Med.* (2021) **8** 223-225. DOI: 10.3389/fcvm.2021.707889
11. O’Sullivan J., Unzeta M., Healy J., O’Sullivan M.I., Davey G., Tipton K.F.. **Semicarbazide-Sensitive Amine Oxidases: Enzymes with Quite a Lot to Do**. *NeuroToxicology* (2004) **25** 303-315. DOI: 10.1016/S0161-813X(03)00117-7
12. Wang Z., Roberts A.B., Buffa J.A., Levison B.S., Zhu W., Org E., Gu X., Huang Y., Zamanian-Daryoush M., Culley M.K.. **Non-lethal Inhibition of Gut Microbial Trimethylamine Production for the Treatment of Atherosclerosis**. *Cell* (2015) **163** 1585-1595. DOI: 10.1016/j.cell.2015.11.055
13. Trøseid M., Ueland T., Hov J.R., Svardal A., Gregersen I., Dahl C.P., Aakhus S., Gude E., Bjørndal B., Halvorsen B.. **Microbiota-dependent metabolite trimethylamine-N-oxide is associated with disease severity and survival of patients with chronic heart failure**. *J. Intern. Med.* (2015) **277** 717-726. DOI: 10.1111/joim.12328
14. Mente A., Chalcraft K., Ak H., Davis A.D., Lonn E., Miller R., Potter M.A., Yusuf S., Anand S.S., McQueen M.J.. **The relationship between trimethylamine-N-oxide and prevalent cardiovascular disease in a multi-ethnic population living in Canada**. *Can. J. Cardiol.* (2015) **31** 1189-1194. DOI: 10.1016/j.cjca.2015.06.016
15. Lever M., George P.M., Slow S., Bellamy D., Young J.M., Ho M., McEntyre C.J., Elmslie J.L., Atkinson W., Molyneux S.L.. **Betaine and trimethylamine-N-oxide as predictors of cardiovascular outcomes show different patterns in diabetes mellitus: An observational study**. *PLoS ONE.* (2014) **9**. DOI: 10.1371/journal.pone.0114969
16. Uchida K.. **Role of reactive aldehyde in cardiovascular diseases**. *Free Radic. Biol. Med.* (2000) **28** 1685-1696. DOI: 10.1016/S0891-5849(00)00226-4
17. Maynard S., Fang E.F., Scheibye-Knudsen M., Croteau D.L., Bohr V.A.. **DNA damage, DNA repair, aging, and neurodegeneration**. *Cold Spring Harb. Perspect. Med.* (2015) **5** 25-30. DOI: 10.1101/cshperspect.a025130
18. Wang S.-H., Yu T.-Y., Hung C.-S., Yang C.-Y., Lin M.-S., Su C.-Y., Chen Y.-L., Kao H.-L., Chuang L.-M., Tsai F.-C.. **Inhibition of Semicarbazide- sensitive Amine Oxidase Reduces Atherosclerosis in Cholesterol-fed New Zealand White Rabbits**. *Sci. Rep.* (2018) **8** 9249. DOI: 10.1038/s41598-018-27551-6
19. Wang S.-H., Yu T.-Y., Tsai F.-C., Weston C.-J., Lin M.-S., Hung C.-S., Kao H.-L., Li Y.-I., Solé M., Unzeta M.. **Inhibition of Semicarbazide-Sensitive Amine Oxidase Reduces Atherosclerosis in Apolipoprotein E-Deficient Mice**. *Transl. Res.* (2018) **197** 12-31. DOI: 10.1016/j.trsl.2018.03.001
20. Gubisne-Haberle D., Hill W., Kazachkov M., Richardson J.S., Yu P.H.. **Protein cross-linkage induced by formaldehyde derived from semicarbazide-sensitive amine oxidase-mediated deamination of methylamine**. *J. Pharmacol. Exp. Ther.* (2004) **310** 1125-1132. DOI: 10.1124/jpet.104.068601
21. Dator R.P., Solivio M.J., Villalta P.W., Balbo S.. **Bioanalytical and Mass Spectrometric Methods for Aldehyde Profiling in Biological Fluids**. *Toxics* (2019) **7**. DOI: 10.3390/toxics7020032
22. Yu P.H., Mei-Zuo D.. **Oxidative Deamination of Methylamine by Semicarbazide-Sensitive Amine Oxidase Leads to Cytotoxic Damage in Endothelial Cells**. *Diabetes* (1993) **42** 87-93. DOI: 10.2337/diab.42.4.594
23. Hernandez M., Solé M., Boada M., Unzeta M.. **Soluble Semicarbazide Sensitive Amine Oxidase (SSAO) catalysis induces apoptosis in vascular smooth muscle cells**. *Biochim. Biophys. Acta (BBA) Mol. Cell Res.* (2006) **1763** 164-173. DOI: 10.1016/j.bbamcr.2005.11.017
24. Mercier N.. **The role of ‘‘semicarbazide-sensitive amine oxidase’’ in the arterial wall**. *Artery Res.* (2009) **3** 141-147. DOI: 10.1016/j.artres.2009.10.002
25. Mathys K.C., Ponnampalam S.N., Padival S., Nagaraj R.H.. **Semicarbazide-sensitive amine oxidase in aortic smooth muscle cells mediates synthesis of a methylglyoxal-AGE: Implications for vascular complications in diabetes**. *Biochem. Biophys. Res. Commun.* (2002) **297** 863-869. DOI: 10.1016/S0006-291X(02)02293-3
26. Lin Z., Li H., Luo H., Zhang Y., Luo W.. **Benzylamine and methylamine, substrates of semicarbazide-sensitive amine oxidase, attenuate inflammatory response induced by lipopolysaccharide**. *Int. Immunopharmacol.* (2011) **2** 1079-1089. DOI: 10.1016/j.intimp.2011.03.002
27. Vidrio H., Medina M., González-Romo P., Lorenzana-Jiménez M., Díaz-Arista P., Baeza A.. **Semicarbazide-sensitive amine oxidase substrates potentiate hydralazine hypotension: Possible role of hydrogen peroxide**. *J. Pharmacol. Exp. Ther.* (2003) **307** 497-504. DOI: 10.1124/jpet.103.055350
28. Zorzano A., Abella A., Marti L., Carpéné C., Palacín M., Testar X.. **Semicarbazide-sensitive amine oxidase activity exerts insulin-like effects on glucose metabolism and insulin-signalling pathways in adipose cells**. *Biochim. Biophys. Acta (BBA) Proteins Proteom.* (2003) **47** 3-9. DOI: 10.1016/S1570-9639(03)00039-6
29. Byon C.H., Heath J.M., Chen Y.. **Redox signalling in cardiovascular pathophysiology: A focus on hydrogen peroxide and vascular smooth muscle cells**. *Redox Biol.* (2016) **9** 244-253. DOI: 10.1016/j.redox.2016.08.015
30. Obata T.. **Diabetes, and semicarbazide-sensitive amine oxidase (SSAO) activity: A review**. *Life Sci.* (2006) **79** 417-422. DOI: 10.1016/j.lfs.2006.01.017
31. Xiao W., Loscalzo J., Quiles J.L., Sánchez-González C., Vera-Ramírez L., Giampieri F., Navarro-Hortal M.D., Llopis J., Battino M., Varela-López A.. **Metabolic Responses to Reductive Stress**. *Antioxid. Redox Signal.* (2020) **32** 1332-1347. DOI: 10.1089/ars.2019.7803
32. Yang W., Li H., Luo H., Luo W.. **Inhibition of Semicarbazide- Sensitive Amine Oxidase Attenuates Myocardial Ischemia-Reperfusion Injury in an In Vivo Rat Model**. *Life Sci.* (2011) **88** 302-306. DOI: 10.1016/j.lfs.2010.12.003
33. Zhang M., Liu L., Zhi F., Niu P., Yang M., Zhu X., Diao Y., Li Y., Wang J., Zhao Y.. **Inactivation of Semicarbazide-Sensitive Amine Oxidase Induces the Phenotypic Switch of Smooth Muscle Cells and Aggravates the Development of Atherosclerotic Lesions**. *Atherosclerosis* (2016) **249** 76-82. DOI: 10.1016/j.atherosclerosis.2016.03.039
34. Okin D., Medzhitov R.. **The Effect of Sustained Inflammation on Hepatic Mevalonate Pathway Results in Hyperglycemia**. *Cell* (2016) **165** 343-356. DOI: 10.1016/j.cell.2016.02.023
35. Deng Y., Yu P.H.. **Assessment of the Deamination of Aminoacetone, an Endogenous Substrate for Semicarbazide-Sensitive Amine Oxidase**. *Anal. Biochem.* (1999) **270** 97-102. DOI: 10.1006/abio.1999.4058
36. Nayak D.D., Agashe D., Lee M.-C., Marx C.J.. **Selection Maintains Apparently Degenerate Metabolic Pathways due to Tradeoffs in Using Methylamine for Carbon versus Nitrogen**. *Curr. Biol.* (2016) **26** 1416-1426. DOI: 10.1016/j.cub.2016.04.029
37. Sartori A., Garay-Malpartida H.M., Forni M.F., Schumacher R.I., Dutra F., Sogayar M.C., Bechara E.J.. **Aminoacetone, a putative endogenous source of methylglyoxal, causes oxidative stress and death to insulin-producing rinm5f cells**. *Chem. Res. Toxicol.* (2008) **21** 1841-1850. DOI: 10.1021/tx8001753
38. Sartori A., Mano C.M., Nascimento O.R., Dyszy F.H., Nantes I.L., Bechara E.J.H.. **Oxidative Damage to Cytochrome C Induced by Aminoacetone**. *Free Radic. Biol. Med.* (2010) **49** S171. DOI: 10.1016/j.freeradbiomed.2010.10.486
39. Lin Z., Luo W., Li H., Zhang Y.. **The effect of endogenous formaldehyde on the rat aorta endothelial cells**. *Toxicol. Lett.* (2005) **159** 134-143. DOI: 10.1016/j.toxlet.2005.05.003
40. Braun J.D., Pastene D.O., Breedijk A., Rodriguez A., Hofmann B.B., Sticht C., von Ochsenstein E., Allgayer H., Born J.V.D., Bakker S.. **Methylglyoxal down-regulates the expression of cell cycle associated genes and activates the p53 pathway in human umbilical vein endothelial cells**. *Sci. Rep.* (2019) **9** 1152. DOI: 10.1038/s41598-018-37937-1
41. Zhang Y., Yang Y., He X., Yang P., Zong T., Sun P., Sun R., Yu T., Jiang Z.. **The cellular function and molecular mechanism of formaldehyde in cardiovascular disease and heart development**. *J. Cell. Mol. Med.* (2021) **25** 5358-5371. DOI: 10.1111/jcmm.16602
42. Saito Y., Nishio K., Yoshida Y., Niki E.. **Cytotoxic effect of formaldehyde with free radicals via increment of cellular reactive oxygen species**. *Toxicology* (2005) **210** 235-245. DOI: 10.1016/j.tox.2005.02.006
43. Desai K.M., Chang T., Wang H., Banigesh A., Dhar A., Liu J., Untereiner A., Wu L.. **Oxidative stress and aging: Is methylglyoxal the hidden enemy?**. *Can. J. Physiol. Pharmacol.* (2010) **88** 10-21
44. Kim D., Kim K.-A., Kim J.-H., Kim E.-H., Bae O.-N.. **Methylglyoxal-Induced Dysfunction in Brain Endothelial Cells via the Suppression of Akt/HIF-1α Pathway and Activation of Mitophagy Associated with Increased Reactive Oxygen Species**. *Antioxidants* (2020) **9**. DOI: 10.3390/antiox9090820
45. Liu C., Cao B., Zhang Q., Zhang Y., Chen X., Kong X., Dong Y.. **Inhibition of thioredoxin 2 by intracellular methylglyoxal accumulation leads to mitochondrial dysfunction and apoptosis in INS-1 cells**. *Endocrine* (2020) **68** 103-115. DOI: 10.1007/s12020-020-02191-x
46. Aquilano K., Baldelli S., Ciriolo M.. **Glutathione: New roles in redox signalling for an old antioxidant**. *Front. Pharmacol.* (2014) **5** 196. DOI: 10.3389/fphar.2014.00196
47. Patel J.J., Bourne L.E., Thakur S., Farrington K., Gorog D.A., Orriss I.R., Baydoun A.R.. **2-Oxothiazolidine-4-carboxylic acid inhibits vascularcalcification via induction of glutathione synthesis**. *J. Cell. Physiol.* (2020) **236** 2696-2705. DOI: 10.1002/jcp.30036
48. Franco R., Panayiotidis M.I., Cidlowski J.A.. **Glutathione Depletion Is Necessary for Apoptosis in Lymphoid Cells Independent of Reactive Oxygen Species Formation**. *J. Biol. Chem.* (2007) **282** 30452-30465. DOI: 10.1074/jbc.M703091200
49. Boomsma F., van Dijk J., Bhaggoe U.M., Bouhuizen A.M., Meiracker A.H.V.D.. **Variation in semicarbazide-sensitive amine oxidase activity in plasma and tissues of mammals**. *Comp. Biochem. Physiol. Part C Pharmacol. Toxicol. Endocrinol.* (2000) **126** 69-78. DOI: 10.1016/S0742-8413(00)00101-8
50. Chi J., Meng L., Pan S., Lin H., Zhai X., Liu L., Zhou C., Jiang C., Guo H.. **Primary Culture of Rat Aortic Vascular Smooth Muscle Cells: A New Method**. *Med. Sci. Monit.* (2017) **23** 4014-4020. DOI: 10.12659/MSM.902816
51. Oosterhoff L.A., Kruitwagen H.S., Van Wolferen M.E., Van Balkom B.W., Mokry M., Lansu N., Dungen N.A.V.D., Penning L.C., Spanjersberg T.C., De Graaf J.W.. **Characterization of Endothelial and Smooth Muscle Cells from Different Canine Vessels**. *Front. Physiol.* (2019) **10** 101. DOI: 10.3389/fphys.2019.00101
52. Riss T.L., Moravec R.A., Niles A.L., Duellman S., Benink H.A., Worzella T.J., Minor L.. **Cell viability assays**. *Assay guidance manual* (2016)
53. Tchivilev I., Madamanchi N.R., Vendrov A.E., Niu X.L., Runge M.S.. **Identification of a protective role for protein phosphatase 1cγ1 against oxidative stress-induced vascular smooth muscle cell apoptosis**. *J. Biol. Chem.* (2008) **283** 22193-22205. DOI: 10.1074/jbc.M803452200
54. Rahman I., Kode A., Biswas S.K.. **Assay for quantitative determination of glutathione and glutathione disulphide levels using enzymatic recycling method**. *Nat. Protoc.* (2006) **27** 3150-3165
|
---
title: Targeted Delivery of Butyrate Improves Glucose Homeostasis, Reduces Hepatic
Lipid Accumulation and Inflammation in db/db Mice
authors:
- Signe Schultz Pedersen
- Michala Prause
- Christina Sørensen
- Joachim Størling
- Thomas Moritz
- Eliana Mariño
- Nils Billestrup
journal: International Journal of Molecular Sciences
year: 2023
pmcid: PMC10002599
doi: 10.3390/ijms24054533
license: CC BY 4.0
---
# Targeted Delivery of Butyrate Improves Glucose Homeostasis, Reduces Hepatic Lipid Accumulation and Inflammation in db/db Mice
## Abstract
Butyrate produced by the gut microbiota has beneficial effects on metabolism and inflammation. Butyrate-producing bacteria are supported by diets with a high fiber content, such as high-amylose maize starch (HAMS). We investigated the effects of HAMS- and butyrylated HAMS (HAMSB)-supplemented diets on glucose metabolism and inflammation in diabetic db/db mice. Mice fed HAMSB had 8-fold higher fecal butyrate concentration compared to control diet-fed mice. Weekly analysis of fasting blood glucose showed a significant reduction in HAMSB-fed mice when the area under the curve for all five weeks was analyzed. Following treatment, fasting glucose and insulin analysis showed increased homeostatic model assessment (HOMA) insulin sensitivity in the HAMSB-fed mice. Glucose-stimulated insulin release from isolated islets did not differ between the groups, while insulin content was increased by $36\%$ in islets of the HAMSB-fed mice. Expression of insulin 2 was also significantly increased in islets of the HAMSB-fed mice, while no difference in expression of insulin 1, pancreatic and duodenal homeobox 1, MAF bZIP transcription factor A and urocortin 3 between the groups was observed. Hepatic triglycerides in the livers of the HAMSB-fed mice were significantly reduced. Finally, mRNA markers of inflammation in liver and adipose tissue were reduced in mice fed HAMSB. These findings suggest that HAMSB-supplemented diet improves glucose metabolism in the db/db mice, and reduces inflammation in insulin-sensitive tissues.
## 1. Introduction
Type 2 diabetes (T2D) is a metabolic disease, characterized by hyperglycemia, that affects over 450 million people worldwide [1]. *Both* genetic and environmental factors play a role in the etiology, but the rapid rise in the incidence in the last decades is likely explained by lifestyle changes [2]. Physical inactivity and calorie-dense diets increase the risk of obesity associated with metabolic complications, such as insulin resistance and low-grade inflammation [3]. Insulin resistance increases the demand of pancreatic beta-cells to secrete insulin and, in susceptible individuals, the beta-cells eventually fail to produce and secrete sufficient amount of insulin to maintain normoglycemia [4]. Hyperglycemia causes serious life-threating, long-term complications, including cardiovascular disease and kidney failure, and therefore, strategies to prevent T2D are of great interest [2].
Growing evidence shows that the gut microbiota plays an important role in health and metabolic diseases. The gut microbiota produces short-chain fatty acids (SCFAs), such as acetate, propionate and butyrate, through fermentation of dietary fibers [5]. These SCFAs are immune-regulatory, promote gut health and play an important role in host homeostasis, including the maintenance of a healthy gut microbiota critical to control metabolic diseases [6]. SCFAs exert functions in the gut, but they also enter the circulation and reach peripheral tissues and organs [7,8,9]. Individuals with T2D have consistently shown a reduced abundance of butyrate-producing bacteria in the gut [10,11,12,13,14]. Furthermore, a direct correlation between beta-cell function and fecal butyrate concentrations has been found, indicating a role for butyrate in the regulation of glucose metabolism [12]. Therefore, targeting the gut microbiota and the production of SCFAs may have potential in the prevention or treatment of T2D.
Several studies have shown positive effects of butyrate on obesity, inflammation, insulin resistance, and liver dysfunction in rodent models, with various mechanisms identified [15,16,17,18,19,20,21,22]. However, since conventional routes of administrating butyrate do not model a sustained release and absorption of butyrate in the colon and into the circulation, clinical translation may be limited [23]. Moreover, butyrate has a short half-life (15–30 min) in circulation, and is found in low concentrations compared to acetate and propionate [24,25]. For example, rectal administration of butyrate in obese men only resulted in a transient increase in plasma butyrate [26]. This shows that frequent delivery is necessary for sustained release as only limited effects of oral butyrate in individuals with T2D [27,28] or type 1 diabetes (T1D) [29,30] have been reported.
An alternative approach for sustained delivery of SCFAs to the gut and into the circulation is to modulate the gut microbiota to promote a constant production of SCFAs. For example, a specially designed prebiotic starch (from high-amylose maize starch, HAMS) can be used as a carrier for butyrate (HAMSB) [31]. The starch carrier is highly resistant to digestion by host enzymes, and it specifically delivers butyrate to the colon, where it is liberated by bacterial enzymes. The starch residues themselves are degraded and fermented to SCFAs, thereby also promoting the growth of SCFA-producing bacteria. This is an efficient approach to enhance luminal concentrations of SCFAs in both animals [7,32,33,34] and humans [8,35,36]. In non-obese diabetic (NOD) mice, HAMS bound to both butyrate and acetate (HAMSAB) protected against T1D [7]. Furthermore, immune modulatory effects have been reported in humans with T1D [8] and hypertension [37]. We have previously shown that butyrate prevents cytokine-induced beta-cell dysfunction and inhibits proinflammatory gene expression in vitro [38,39]. Therefore, we hypothesized that butyrate has beneficial effects on beta-cell function and glucose homeostasis in mice prone to T2D. We chose the widely used leptin receptor knockout (db/db) mouse model, in which the lack of leptin signaling causes persistent hyperphagia, obesity, beta-cell dysfunction, and insulin resistance and diabetes [40]. We investigated the effect of butyrate, delivered in the form of HAMSB, on glucose metabolism, beta-cell function, hepatic lipid accumulation and inflammation.
## 2.1. Butyrate Improves the Function of Islets in db/db Mice In Vitro
We previously showed that butyrate protects pancreatic islets from cytokine-induced dysfunction, and reduces inflammatory NF-κB signaling [38,39]. To determine whether butyrate also exerts protective effects on islets from diabetic animals at different disease stages, islets were isolated from db/db mice at 9 and 13 weeks of age. Glucose-stimulated insulin secretion was measured and compared to islets isolated from 11-week-old control C57BL/6 mice (Figure 1A).
Glucose-stimulated insulin secretion was significantly reduced in the islets of db/db mice at both 9 and 13 weeks of age compared to controls (Figure 1A). Moreover, the islets of db/db mice contained significantly less insulin (Figure 1B). To investigate whether butyrate could improve db/db islet function, islets from the db/db mice were cultured for 10 days, with and without butyrate. Butyrate significantly increased glucose-stimulated insulin secretion compared to islets cultured without butyrate isolated from 13-week-old mice (Figure 1C). Although not statistically significant, we observed a similar increase in glucose-stimulated insulin secretion from islets of 9-week-old mice (Figure 1C). Butyrate had no effect on db/db islet insulin content (Figure 1D). Together, these results indicate that butyrate improves the impaired glucose-stimulated insulin secretion seen in the islets of db/db mice.
## 2.2. Effect of HAMSB on SCFA Production
We next investigated the effect of butyrate on glucose homeostasis in db/db mice. Db/db mice were fed either [1] a control diet (Ctr), [2] a diet supplemented with HAMS or [3] a diet supplemented with HAMSB for 5 weeks (Figure 2A). The HAMS diet was included to differentiate between the effects of butyrate and the increased SCFA production through fermentation of HAMS. After 5 weeks, butyrate concentrations in feces from HAMSB-fed mice were 8-fold higher compared to the levels found in feces from Ctr or HAMS-fed mice (Figure 2B). The HAMS diet increased fecal butyrate 2.5-fold compared to the control diet, although this effect was not statistically significant (Figure 2B). In addition, both HAMS and HAMSB increased fecal propionate (Figure 2C), while fecal acetate was only significantly increased in HAMSB-fed mice (Figure 2D). Total SCFA levels in feces were significantly higher in both HAMS- and HAMSB-fed mice compared to the Ctr group (Figure 2E). We did not observe any significant difference in plasma butyrate (Figure 2F) or acetate (Figure 2H) between the three groups. HAMS-fed mice had increased plasma propionate compared to the Ctr mice (Figure 2G). Total plasma SCFA levels were not different between the groups (Figure 2I). Together, this suggests that HAMSB effectively delivers butyrate, as well as propionate and acetate, to the large intestine, and that HAMS also increases SCFA production.
## 2.3. Effect of HAMSB on Body Weight and Glycemic Control
To follow disease progression, body weight (Figure 3) and fasting blood glucose levels (Figure 4) were assessed weekly. At baseline, total body weight and non-fasting blood glucose did not differ between the three groups (Figure S1). However, after 5 weeks HAMSB-fed mice had gained 15.8 ± 1.6 g of total body mass, whereas the Ctr mice only gained 13.7 ± 1.8 g (Figure 3B). Mice fed HAMS gained 14.9 ± 2.3 g (Figure 3B). The weight differences could not be explained by differences in food consumption, as the average food intake was similar among the groups (Figure 3C).
Over the course of 5 weeks, no major differences in fasting blood glucose were observed between the three groups (Figure 4A). Only in week 3 the HAMSB-fed mice had significantly lower blood glucose levels compared to the Ctr group (Figure 4B). As a measure of blood glucose control during the entire experimental period, we calculated the area under the blood glucose curve. The HAMSB-fed mice had significantly lower blood glucose compared to the Ctr mice (Figure 4B). Furthermore, we observed a non-significant ($$p \leq 0.059$$) decrease in plasma insulin in mice fed HAMSB for 5 weeks (Figure 4D). Analysis of fasting blood glucose (Figure 4C) and insulin levels (Figure 4D) together, in the homeostatic model assessment for insulin resistance (HOMA-IR), showed that HAMSB-fed mice had a significantly lower HOMA-IR compared to Ctr mice (Figure 4E). HAMS did not affect blood glucose, plasma insulin levels or HOMA-IR (Figure 4). Together, this indicates that although mice fed HAMSB gained more weight compared to Ctr mice, they developed less severe hyperglycemia and have improved insulin sensitivity.
## 2.4. Effect of HAMSB on Islet Function and Identity
To gain better insight into the mechanisms by which HAMSB improves glycemic control, we isolated islets and assessed glucose-stimulated insulin secretion. Basal and glucose-stimulated insulin release were similar in islets from Ctr and HAMSB-fed mice (Figure 5A), whereas islets from HAMS-fed mice secreted less insulin at 20 mM glucose (Figure 5B). When glucose-stimulated insulin secretion was augmented by forskolin, a non-significant increase was observed from islets isolated from HAMSB-fed mice compared to the other groups ($$p \leq 0.086$$) (Figure 5C). Total islet insulin content was significantly higher ($36\%$) in islets from mice fed HAMSB compared to islets from Ctr mice (Figure 5D). Moreover, the expression of the insulin gene Ins2 was $39\%$ higher in islets from HAMSB-fed mice compared to Ctr mice (Figure 5F). The expression of other key beta-cell identity genes, such as Ins1, MafA, Pdx1 and Ucn3, was not significantly different between the groups (Figure 5F). Total beta-cell area, determined as insulin-positive cells per pancreas area, did not differ between the groups (Figure 5E) and no obvious difference in the number of islets and islet size was found. Together, these results show that islets from the HAMSB-fed mice express higher levels of Ins2 mRNA and contain more insulin compared to islets from control mice.
## 2.5. Effect of HAMSB on Lipid and Glucose Metabolism in the Liver
Because dysregulation of glucose and lipid metabolism in the liver promotes systemic metabolic dysfunction [41], we measured hepatic lipid accumulation and expression of genes involved in lipid and glucose metabolism. The levels of hepatic triglycerides were significantly lower (~$9\%$) in HAMS- and HAMSB-fed mice compared to Ctr mice (Figure 6A). We analyzed the expression of genes involved in the regulation of lipid accumulation, such as uptake of fatty acids, de novo lipogenesis, fatty acid oxidation and/or export of fatty acids [42]. However, no differences were found in the expression of de novo lipogenesis genes Acaca, Fasn, Scd1 and Srebp-1c, the fatty acid transporter Cd36 and the rate-limiting enzyme in fatty acid oxidation Cpt1a (Figure 6B). The expression of Pck1, encoding phosphoenolpyruvate carboxykinase, and G6p, encoding glucose-6-phosphatase, was not statistically different between the groups (Figure 6C). Together, these results suggest that HAMS and HAMSB supplementation inhibit lipid accumulation in the liver, without affecting the expression of key genes involved in lipid metabolism.
## 2.6. Effect of HAMSB on Inflammation
Low-grade inflammation is associated with T2D and fatty liver disease [43]. To assess local inflammation, we measured the expression of genes associated with inflammation in the liver and adipose tissue. The expression of monocyte/macrophage marker genes Cd68 and F$\frac{4}{80}$ was significantly reduced in both the liver (Figure 7A) and adipose tissue of HAMSB-fed mice (Figure 7B). A non-significant reduction in the expression of the proinflammatory cytokine Tnf-α was observed in both liver and adipose tissue by HAMSB (Figure 7A,B). HAMS significantly reduced the expression of Tnf-α in the liver (Figure 7A), but not in the adipose tissue (Figure 7B). Systemic inflammation was assessed by measuring cytokine levels in plasma. However, no differences in neither proinflammatory cytokines (IL-1β, Tnf-α, IFN-ƴ, IL-2, IL-5, IL-6 and KC/Gro) nor the anti-inflammatory cytokine IL-10 were found between the groups (Figure S2). Together, these results indicate that HAMSB ameliorates inflammation locally in the liver and adipose tissue.
## 3. Discussion
In this study, we show that a diet containing butyrate in the form of HAMSB improves metabolic parameters associated with diabetes in db/db mice. HAMSB increased the insulin sensitivity, insulin content in pancreatic islets, and decreased the accumulation of hepatic triglycerides and inflammation in both the liver and adipose tissue. Together, these findings suggest anti-diabetic effects of HAMSB.
We used HAMS as a carrier to ensure targeted and sustained release of butyrate in the colon. Higher concentrations of butyrate were found in the feces of HAMSB-fed mice compared to HAMS-fed and control mice, indicating that this is an efficient way to increase the level of butyrate in the large intestine. Moreover, butyrate may decrease colonic pH, which can further boost the growth of butyrate-producing species [44,45]. As expected, the bacterial fermentation of HAMS promoted SCFA production. HAMSB also increased fecal acetate, suggesting that HAMSB modulates the overall gut microbiota community structure and promotes the growth of acetate-producing bacteria, or that butyrate is converted into acetate [46]. Therefore, we cannot rule out that the effects of HAMSB are solely caused by butyrate, as other metabolites could likely have an effect as well. However, it is important to note that fecal SCFAs only represent $5\%$ of total SCFAs in the colon and will always reflect a balance between excretion, cross-feeding interactions and absorption [25,47]. Because butyrate is the main energy source for the colonocytes and is further metabolized in the liver, only low concentrations of butyrate can be measured in the circulation [46,48]. HAMSB has been shown to be superior to HAMS in increasing butyrate levels in the circulation [7,34]. We found that HAMSB tended to increase plasma levels of butyrate, but it was not different from HAMS. Possible explanations for this discrepancy could be the model used, duration of intervention, diet composition or measurement under fasted condition. Analysis of butyrate concentration in the portal vein might provide a more accurate measure of butyrate uptake. Unexpectedly, we also observed that HAMSB-fed mice gained more weight compared to controls, despite similar energy intake. This finding contradicts previous studies showing that butyrate inhibits weight gain in high-fat diet, obesity-induced animal models [15,16,19]. This could possibly be explained by increased energy harvest from the higher levels of butyrate, or it could be a result of cecum enlargement. Several studies have shown that fiber fermentation and SCFA production increase cecum tissue and content weight [15,16,19].
Key characteristics of T2D are insulin resistance and insufficient insulin production from beta-cells, leading to hyperglycemia. We observed that islets of mice fed HAMSB contained more insulin, had higher insulin gene expression, and tended to secrete more insulin in response to stimulation with 20 mM glucose and forskolin compared to controls. Together, this suggests that islets of HAMSB-fed mice have a higher capacity to secrete insulin under increased metabolic demand. The db/db mice have a mutation in the leptin receptor, and display hyperphagia, and consequently develop obesity, hyperglycemia and insulin resistance, resembling key features of human T2D [40]. The beta-cells adapt to these metabolic challenges by producing and secreting more insulin to prevent hyperglycemia [49]. Although the mice fed HAMSB also developed hyperglycemia, these mice had decreased HOMA-IR, indicating improved insulin sensitivity compared to the control mice. This might explain the reduction in integrated blood glucose over the 5-week treatment period observed in HAMSB-fed mice. No effect of HAMSB feeding was observed on the beta-cell area as percent of total pancreas area, indicating no effect on beta-cell growth and apoptosis. However, these data were obtained using relatively few mice and total beta-cell mass, proliferation and apoptosis were not measured in this study. Whether the effects of butyrate on the islets are direct or secondary to its action on other tissues remains unknown. However, in the present study, we show that butyrate improves beta-cell function in islets of db/db mice in vitro. This, together with our previous findings that butyrate protects beta-cells from cytokine-induced dysfunction [38], indicates direct beneficial effects of butyrate on beta-cell function. It can be speculated whether the concentration of butyrate in the plasma of the db/db is sufficient to elicit a direct response in beta-cells. If not, the effects of the HAMSB on the beta-cells must be indirect. This could be explained by butyrate reducing blood glucose, local tissue inflammation, hepatic lipid accumulation and improved insulin sensitivity, which ameliorate stress from the beta-cells and could thereby, indirectly, maintain insulin production and reserve.
The liver plays a key role in lipid and glucose metabolism [50]. We observed that mice fed HAMSB had reduced lipid accumulation and inflammatory markers in the liver compared to controls. Similar beneficial effects of butyrate have previously been described in experimental models using butyrate mixed into the diet [16,51,52]. Hepatic steatosis impairs liver function and develops as a consequence of an imbalance between de novo lipogenesis, fatty acid oxidation and import and export of lipids [42]. We did not find any changes in the expression of selected genes involved in these processes, but other genes or posttranslational events could be important. Moreover, failure of insulin to inhibit lipolysis in the adipose tissue would release more fatty acids into the circulation, resulting in higher uptake in the liver [53]. Kosteli and coworkers showed that lipolysis in the adipose tissue recruits macrophages [54]. Notably, adipose tissue inflammation was reduced in mice fed HAMSB, suggesting that lipolysis is also reduced, and thus this could be a potential mechanism by which butyrate inhibits hepatic steatosis.
The gut is connected to the liver via the portal vein, and therefore the concentration of gut metabolites, such as butyrate, is higher in the liver compared to the periphery [55]. We have previously shown that HAMSB increases the levels of butyrate in portal blood [7]. Both HAMS and HAMSB had beneficial effects on the liver, indicating that these effects might result from the HAMS carrier. It is likely, that the concentration of butyrate or any of the other SCFAs delivered from HAMS or HAMSB are sufficient to directly affect the hepatocytes, e.g., via signaling through G protein-coupled rectors or inhibition of histone deacetylases. Direct effects of butyrate in the liver have been reported by others [56,57,58].
Since we deliver butyrate targeted to the large intestine, some of the effects of HAMSB (and HAMS) might also be directly mediated via the gut microbiota. Others have shown that butyrate in the gut stimulates the secretion of the gut hormones glucagon-like peptide (GLP-1) and peptide YY (PYY) from the enteroendocrine L-cells that play an important role in decreasing blood glucose by promoting insulin secretion and satiety [59,60,61]. Butyrate also improves the intestinal barrier function by increasing tight junction assembly [16] and mucus production [62]. Hyperglycemia and gut microbiota dysbiosis drive intestinal barrier dysfunction, allowing the passage of bacterial components and dietary antigens that can trigger an inflammatory response [63,64]. Improvement of the barrier function by butyrate could, thus, decrease inflammation locally and systemically. Accordingly, HAMSB reduced the expression of proinflammatory cytokine, Tnf-α, and macrophage markers, Cd68 and F$\frac{4}{80}$, in the liver and adipose tissue. Direct anti-inflammatory activity of butyrate via an inhibition of NF-κB signaling has also been reported by us and others [39,65,66,67]. As inflammation has been associated with metabolic dysfunction, the anti-inflammatory effect of HAMSB, whether through an improvement of the intestinal barrier and/or directly, suggests a potential mechanism by which butyrate improves insulin sensitivity, glycemic control and hepatic lipid accumulation. These findings are in line with previous studies showing the immunoregulatory properties of HAMS esterified to SCFAs [7,8].
We acknowledge that the db/db model of T2D has limited translational potential, as T2D in human is multifactorial and not caused by a single gene mutation. The db/db mouse model is a severe model and mice display hyperglycemia from a young age [40,68]. Recently, Li et al. showed anti-diabetic effects of different SCFAs bound to HAMS in a high-fat diet and streptozotocin (STZ)-induced rat model of T2D [34]. However, in this mode, l beta-cells were rapidly destroyed after STZ injection, and did not replicate the slow progression of beta-cell dysfunction as seen in human T2D [69]. On the other hand, the db/db model is a severe model of T2D and this could have prevented us from observing beneficial effects of HAMSB that could be relevant in an earlier pre-diabetic stage. Therefore, future studies need to address the potential of HAMSB in less severe models of T2D, such as high-fat, diet-induced models. Collectively, we demonstrate that HAMSB supplementation has beneficial effects on insulin sensitivity, insulin synthesis, and hepatic lipid accumulation, possibly via inhibition of local inflammation. Interestingly, a recent human study observed that enhanced SCFA levels and immunoregulatory effects of HAMSAB persisted 6 weeks after treatment intervention [8], showing that the effects are not transient. Together, this suggests that targeted delivery of butyrate, with HAMS as a carrier, may have potential in prevention and treatment of T2D.
## 4.1. Ex Vivo Stimulation of Islets of db/db Mice
Pancreatic islets from 9- and 13-week old male db/db mice (BKS.Cg -+Leprdb/+Leprdb/OlaHsd, Envigo, Horst, The Netherlands) fed a chow diet were isolated, as previously described [39]. The day after isolation, glucose-stimulated insulin secretion was performed on 25 islets in duplicate, and the rest of the islets were cultured in the RPMI 1640 medium (Gibco, Thermo Fisher Scientific, Roskilde, Denmark), supplemented with $2\%$ human serum (BioWhittaker, Basel, Switzerland), and $1\%$ penicillin (100 U/mL) and streptomycin (100 μg/mL) (P/S) for 10 days, with and without 0.2 mM butyrate (B5887, Sigma, Soeborg, Denmark). On day 5, the islets were transferred to new media with and without butyrate. On day 10, GSIS was performed, as described below. To compare diabetic vs. non-diabetic islets the day after isolation, islets from 11-week old mice (C57BL/6NRj, Janvier, Saint Berthewin Cedex, France) were included.
## 4.2. Experimental Design of Mouse Study
Five-week old male db/db mice (BKS(D)-Leprdb/JOrlRj) were purchased from Janvier. The mice were acclimatized to the animal facility at 20–22 °C with a 12 h light/dark cycle for one and a half weeks prior to the start of the study. The mice were then randomized into three dietary groups based on body weight and blood glucose measurements, to ensure similar baseline characteristics between the groups. The mice were fed either: [1] a control diet (AIN-93G, Ctr), [2] AIN-93G with $15\%$ corn starch replaced by HAMS (HAMS), or [3] AIN-93G with $15\%$ corn starch replaced by butyrylated HAMS (HAMSB) for 5 weeks. The diets were formulated by Envigo and the composition is found in Table S1. The mice were housed in groups of three to four per cage and had free access to food and water. Food intake, body weight and fasting blood glucose (4 h) were measured once a week. Blood was taken from the tail-tip and glucose levels measured using a glucometer (Bayer Contour, Leverkusen, Germany). After 5 weeks on the diet, the mice were fasted (4 h) before termination of the experiment. The mice were anesthetized using isoflurane and blood was collected by cardiac puncture, after which they were sacrificed by cervical dislocation. Liver and epididymal fat pads were collected and immediately snap-frozen and stored at −80 °C until further analyses. Pancreases from three mice per group were fixed in $4\%$ formaldehyde and pancreatic islets were isolated as previously described from remaining mice [39]. The experiments were approved by the local ethics committee, and animals were housed according to the Principles of Good Laboratory Care. Mice with a fasting blood glucose above 24 mM in week 1 were excluded from the analysis (three mice, from both the Ctr and HAMS group, and two from the HAMSB group).
## 4.3. Short-Chain Fatty Acid Analyses
SCFA concentrations in fecal and plasma samples were determined by liquid chromatography-tandem mass spectrometry. To prepare the fecal sample, they were mixed with 20 µL internal standard mix (1 mM D3-acetate, 5 mM D4-propionate, 0.1 mM D7-butyrate) and freeze-dried, weighted and homogenized by sonication in Milli-Q water. For both fecal and plasma samples, the derivatizing reagent was 200 mM 3-Nitrophenylhydrazine (NPH) and 120 mM N-(3-Dimethylaminopropyl)-N′-ethylcarbodiimide hydrochloride (EDC) in $50\%$ acetonitrile (AcN) (with $6\%$ pyridine). Freeze-dried fecal samples (20 µL) were mixed with derivatizing reagent (40 μL) and incubated for 1 h at room temperature. Plasma samples (10 µL) were mixed with 10 μL of internal standard mix in $50\%$ methanol (0.1 mM D3-acetate, 1 mM D4-propionate, 0.1 mM D7-butyrate) and derivatizing reagent (40 μL), and incubated for 1 h at room temperature. All the samples were centrifuged at 14,000 rpm for 10 min at 4 °C, and mixed 1:1 with $10\%$ AcN. The quantification of the SCFAs was achieved on the liquid chromatography-tandem mass spectrometer consisting of a Waters Acquity UPLC I-Class connected to a Waters Xevo TQ-XS tandem mass spectrometer (Waters, Manchester, UK). The separation was performed by injecting 2 µL of each sample to an Acquity UPLC HSS-T3 column (100 × 2.1 mm, 1.8µm, Waters, MA, USA). The mobile phase consisted of $0.1\%$ formic acid and AcN with $0.1\%$ formic acid, and was delivered on the column with a flow rate of 0.40 mL/min with the following gradient: $3\%$ B to $55\%$ B in 7 min, and then up to $100\%$ for 1 min, hold for 2 min, and thereafter back to initial condition, in 0.5 min and equilibrated for 2.5 min. Column and autosampler were thermostated at 55 °C and 4 °C, respectively. Analytes were ionized in an electrospray ion source operated in the negative mode. The source and gas parameters were set as follows: ion spray voltage 3.5 kV, desolvation temperature 300 °C, desolvation gas flow 800 L/h, nebulizer pressure 7 Bar, and cone gas flow 150 L/h. The instrument was operated in multiple reaction monitoring mode (MRM). Dwell time was set to 35 ms for all transitions. For quantification 6-point calibration curves were used, including different levels of non-labelled and constant levels of the labelled internal standards. The instrument was controlled by MassLynx 4.2, and data processing was performed with TargetLynx XS (Waters, Milford, MA, USA).
## 4.4. Blood Samples
Terminal blood was collected by cardiac puncture in EDTA tubes and placed immediately on ice, after which they were centrifuged at 5000× g for 10 min at 4 °C. Plasma was collected and stored at −80 °C until analysis. Plasma levels of insulin were measured using a mouse insulin ELISA kit (80-INSMS-E01, Alpco, Salem, NH, USA), according to the manufacturer’s instruction. As an index of insulin resistance, HOMA-IR was calculated (HOMA-IR = fasting glucose (mmol/L) × fasting insulin (μU/mL)/22.5). Proinflammatory cytokine levels were determined using the V-plex proinflammatory Panel 1 mouse kit (Meso Scale Discovery, # K15048D, Rockville, MD, USA), according to the manufacturer’s instructions, using a MESO QuickPlex SQ 120 instrument.
## 4.5. Glucose-Stimulated Insulin Secretion
After overnight culture in RPMI 1640 media with $1\%$ P/S and $10\%$ fetal bovine serum (FBS, Biosera, Herlev, Denmark), islets were collected in groups of 25 islets of similar size in duplicate from each mouse, and preincubated in 24-well plates with the Krebs–Ringer HEPES (KRBH) buffer, supplemented with 2 mM glucose for 1.5 h at 37 °C. The islets were then transferred to new wells with 2 mM glucose in the KRBH buffer for 30 min, half of the buffer was collected and the concentration of the glucose was increased to 20 mM. After 30 min, half of the buffer was collected and forskolin (20 µM) was added for 30 min. Then, the buffer was collected, and the islets were saved for the determination of cellular insulin and DNA content after sonication of the samples. Insulin was quantified using an in-house developed insulin ELISA. Insulin secretion was normalized to DNA content, determined by the Quant-IT PicoGreen dsDNA Reagent and Kit (Thermo Fisher Scientific, Roskilde, Denmark). The KRBH buffer contained: 115 mM NaCl, 10 mM HEPES, 5 mM NaHCO3, 4.7 mM KCl, 2.6 mM CaCl2, 1.2 mM KH2PO4, 1.2 mM MgSO4, $0.2\%$ BSA, 2 mM glutamine, and $1\%$ P/S, pH 7.4.
## 4.6. RNA Isolation and Gene Expression
Liver (20 mg) and adipose tissues (100 mg) were homogenized in TRIzol reagent with Qiagen Tissuelyser (2 × 2 min, 20 Hz). Adipose tissue samples were centrifuged at 12,000× g for 10 min at 4 °C, to allow a triglyceride phase to form and the aqueous phase was collected. Phase separation was performed using chloroform, followed by centrifugation at 12,000× g for 15 min at 4 °C. For RNA isolation from islets, pools of 300–500 islets from 1–5 mice were lyzed in TRIzol reagent. For all samples, total RNA was extracted using the Direct-zol RNA-mini prep kit (Zymo Research, Nordic Biosite), according to the manufacturer’s protocol. RNA quality and quantity was assed using Nanodrop, and cDNA was synthesized using qScript cDNA Super mix kit (Quantabio). Gene expression in islets was determined using TaqMan, whereas gene expression in the liver and adipose tissue was determined using SYBR. Probes and primers are found in Tables S2 and S3. RT-QPCR was performed on an ABI PRISM 7900HT Sequence Detection System (Applied Biosystems). Each sample was determined in triplicate and the expression of target genes was normalized to the expression of Ppia (encoding cyclophilin A).
## 4.7. Immunohistochemistry and Beta-Cell Area
The fixated pancreases were dehydrated in ethanol, cleared in xylene and embedded in paraffin. The paraffin blocks were cut in 4 µm sections, using a microtome. The sections were transferred to slides, dewaxed in Histolab-Clear (Histolab Products AB, #14250) and rehydrated in graded dilutions of alcohol (70–$99\%$) in tap water. The slides were stained for insulin using a primary in-house antibody (Insulin 2006-4). The signal was amplified by a biotinylated secondary antibody, goat anti-guinea pig (BA-7000, Vector Laboratories, Newark, CA, USA), followed by peroxidase (ABC) (PK-4000, Vector Laboratories). Finally, the reaction was developed by the use of 3,3–diaminobenzidine (SK-4100, Vector Laboratories), and counterstaining was performed with a Hematoxylin Mayer solution (RH-pharmacy, #854183). The slides were examined and photographed using the Zeiss Axio Scanner Z1. The islet area (percent insulin-positive tissue relative to total tissue area) and insulin intensity were quantified using the QuPath software (version0.3.2) Pancreases from three mice per group were analyzed. Three sections per mice were quantified and the average was calculated.
## 4.8. Hepatic Triglycerides
Triglycerides in the liver were quantified using a colorimetric triglyceride assay kit (Ab65336, Abcam, Cambridge, UK), according to the manufacturer’s instructions. Briefly, 60 mg liver tissue was homogenized in 600 µL $5\%$ NP-40 in water, using a pestle. The samples were heated to 90 °C for 5 min and cooled down to room temperature. This was repeated twice to solubilize all triglycerides, and the samples were centrifuged at maximum speed to remove insoluble material. The supernatant was collected, diluted 10-fold in water, and 4 µL was assayed. Absorbance was determined at 570 nm.
## 4.9. Statistics
Statistical analyses and graphs were performed using the GraphPad Prism software Version 9.3.1. For longitudinal measures of blood glucose and body weight, data were analyzed by a repeated measures ANOVA, followed by a Tukey’s test. For comparisons of the two groups, a two-tailed unpaired Student’s t-test was performed if data were normally distributed, otherwise a non-parametric Mann–Whitney u-test was performed. Outliers were defined as values above or below the mean +/− standard deviations, and were excluded from the data. Data are represented as the mean +/− SEM, and a p-value < 0.05 was considered statistically significant. Each symbol represents an individual mouse, unless otherwise specified. For blood glucose measurements over time, AUCs of blood glucose measurements were calculated using the incremental AUC.
## References
1. 1.
International Diabetes Federation
IDF Diabetes Atlas10th ed.International Diabetes FederationBrussels, Belgium2021. *IDF Diabetes Atlas* (2021)
2. Zheng Y., Ley S.H., Hu F.B.. **Global Aetiology and Epidemiology of Type 2 Diabetes Mellitus and Its Complications**. *Nat. Rev. Endocrinol.* (2018) **14** 88-98. DOI: 10.1038/nrendo.2017.151
3. Kahn S.E., Hull R.L., Utzschneider K.M.. **Mechanisms Linking Obesity to Insulin Resistance and Type 2 Diabetes**. *Nature* (2006) **444** 840-846. DOI: 10.1038/nature05482
4. Christensen A.A., Gannon M.. **The Beta-cell in Type 2 Diabetes**. *Curr. Diabetes Rep.* (2019) **19** 81. DOI: 10.1007/s11892-019-1196-4
5. McCarville J.L., Chen G.Y., Cuevas V.D., Troha K., Ayres J.S.. **Microbiota Metabolites in Health and Disease**. *Annu. Rev. Immunol.* (2020) **38** 147-170. DOI: 10.1146/annurev-immunol-071219-125715
6. Fan Y., Pedersen O.. **Gut Microbiota in Human Metabolic Health and Disease**. *Nat. Rev. Microbiol.* (2021) **19** 55-71. DOI: 10.1038/s41579-020-0433-9
7. Mariño E., Richards J.L., McLeod K.H., Stanley D., Yap Y.A., Knight J., McKenzie C., Kranich J., Oliveira A.C., Rossello F.J.. **Gut Microbial Metabolites Limit the Frequency of Autoimmune T Cells and Protect against Type 1 Diabetes**. *Nat. Immunol.* (2017) **18** 552-562. DOI: 10.1038/ni.3713
8. Bell K.J., Saad S., Tillett B.J., McGuire H.M., Bordbar S., Yap Y.A., Nguyen L.T., Wilkins M.R., Corley S., Brodie S.. **Metabolite-Based Dietary Supplementation in Human Type 1 Diabetes Is Associated with Microbiota and Immune Modulation**. *Microbiome* (2022) **10** 9. DOI: 10.1186/s40168-021-01193-9
9. Frost G., Sleeth M.L., Sahuri-Arisoylu M., Lizarbe B., Cerdan S., Brody L., Anastasovska J., Ghourab S., Hankir M., Zhang S.. **The Short-Chain Fatty Acid Acetate Reduces Appetite via a Central Homeostatic Mechanism**. *Nat. Commun.* (2014) **5** 3611. DOI: 10.1038/ncomms4611
10. Wu H., Tremaroli V., Schmidt C., Lundqvist A., Olsson L.M., Krämer M., Gummesson A., Perkins R., Bergström G., Bäckhed F.. **The Gut Microbiota in Prediabetes and Diabetes: A Population-Based Cross-Sectional Study**. *Cell Metab.* (2020) **32** 379-390.e3. DOI: 10.1016/j.cmet.2020.06.011
11. Karlsson F.H., Tremaroli V., Nookaew I., Bergström G., Behre C.J., Fagerberg B., Nielsen J., Bäckhed F.. **Gut Metagenome in European Women with Normal, Impaired and Diabetic Glucose Control**. *Nature* (2013) **498** 99-103. DOI: 10.1038/nature12198
12. Sanna S., van Zuydam N.R., Mahajan A., Kurilshikov A., Vich Vila A., Võsa U., Mujagic Z., Masclee A.A.M., Jonkers D.M.A.E., Oosting M.. **Causal Relationships among the Gut Microbiome, Short-Chain Fatty Acids and Metabolic Diseases**. *Nat. Genet.* (2019) **51** 600-605. DOI: 10.1038/s41588-019-0350-x
13. Allin K.H., Tremaroli V., Caesar R., Jensen B.A.H., Damgaard M.T.F., Bahl M.I., Licht T.R., Hansen T.H., Nielsen T., Dantoft T.M.. **Aberrant Intestinal Microbiota in Individuals with Prediabetes**. *Diabetologia* (2018) **61** 810-820. DOI: 10.1007/s00125-018-4550-1
14. Wang J., Qin J., Li Y., Cai Z., Li S., Zhu J., Zhang F., Liang S., Zhang W., Guan Y.. **A Metagenome-Wide Association Study of Gut Microbiota in Type 2 Diabetes**. *Nature* (2012) **490** 55-60. DOI: 10.1038/nature11450
15. Gao Z., Yin J., Zhang J., Ward R.E., Martin R.J., Lefevre M., Cefalu W.T., Ye J.. **Butyrate Improves Insulin Sensitivity and Increases Energy Expenditure in Mice**. *Diabetes* (2009) **58** 1509-1517. DOI: 10.2337/db08-1637
16. Matheus V.A., Monteiro L.C.S., Oliveira R.B., Maschio D.A., Collares-buzato C.B.. **Butyrate Reduces High-Fat Diet-Induced Metabolic Alterations, Hepatic Steatosis and Pancreatic Beta-cell and Intestinal Barrier Dysfunctions in Prediabetic Mice**. *Exp. Biol. Med.* (2017) **242** 1214-1226. DOI: 10.1177/1535370217708188
17. Xu Y.H., Gao C.L., Guo H.L., Zhang W.Q., Huang W., Tang S.S., Gan W.J., Xu Y., Zhou H., Zhu Q.. **Sodium Butyrate Supplementation Ameliorates Diabetic Inflammation in Db/Db Mice**. *J. Endocrinol.* (2018) **238** 231-244. DOI: 10.1530/JOE-18-0137
18. Zhang W.Q., Zhao T.T., Gui D.K., Gao C.L., Gu J.L., Gan W.J., Huang W., Xu Y., Zhou H., Chen W.N.. **Sodium Butyrate Improves Liver Glycogen Metabolism in Type 2 Diabetes Mellitus**. *J. Agric. Food Chem.* (2019) **67** 7694-7705. DOI: 10.1021/acs.jafc.9b02083
19. Mollica M.P., Raso G.M., Cavaliere G., Trinchese G., De Filippo C., Aceto S., Prisco M., Pirozzi C., Di Guida F., Lama A.. **Butyrate Regulates Liver Mitochondrial Function, Efficiency, and Dynamics in Insulin-Resistant Obese Mice**. *Diabetes* (2017) **66** 1405-1418. DOI: 10.2337/db16-0924
20. Mattace Raso G., Simeoli R., Russo R., Iacono A., Santoro A., Paciello O., Ferrante M.C., Canani R.B., Calignano A., Meli R.. **Effects of Sodium Butyrate and Its Synthetic Amide Derivative on Liver Inflammation and Glucose Tolerance in an Animal Model of Steatosis Induced by High Fat Diet**. *PLoS ONE* (2013) **8**. DOI: 10.1371/journal.pone.0068626
21. Henagan T.M., Stefanska B., Fang Z., Navard A.M., Ye J., Lenard N.R., Devarshi P.P.. **Sodium Butyrate Epigenetically Modulates High-Fat Diet-Induced Skeletal Muscle Mitochondrial Adaptation, Obesity and Insulin Resistance through Nucleosome Positioning**. *Br. J. Pharmacol.* (2015) **172** 2782-2798. DOI: 10.1111/bph.13058
22. Sato F.T., Yap Y.A., Crisma A.R., Portovedo M., Murata G.M., Hirabara S.M., Ribeiro W.R., Marcantonio Ferreira C., Cruz M.M., Pereira J.N.B.. **Tributyrin Attenuates Metabolic and Inflammatory Changes Associated with Obesity through a GPR109A-Dependent Mechanism**. *Cells* (2020) **9**. DOI: 10.3390/cells9092007
23. Gill P.A., van Zelm M.C., Muir J.G., Gibson P.R.. **Review Article: Short Chain Fatty Acids as Potential Therapeutic Agents in Human Gastrointestinal and Inflammatory Disorders**. *Aliment. Pharmacol. Ther.* (2018) **48** 15-34. DOI: 10.1111/apt.14689
24. Egorin M.J., Yuan Z.M., Sentz D.L., Plaisance K., Eiseman J.L.. **Plasma Pharmacokinetics of Butyrate after Intravenous Administration of Sodium Butyrate or Oral Administration of Tributyrin or Sodium Butyrate to Mice and Rats**. *Cancer Chemother. Pharmacol.* (1999) **43** 445-453. DOI: 10.1007/s002800050922
25. Cummings J.H., Pomare E.W., Branch H.W.J., Naylor C.P.E., MacFarlane G.T.. **Short Chain Fatty Acids in Human Large Intestine, Portal, Hepatic and Venous Blood**. *Gut* (1987) **28** 1221-1227. DOI: 10.1136/gut.28.10.1221
26. Canfora E.E., Van Der Beek C.M., Jocken J.W.E., Goossens G.H., Holst J.J., Olde Damink S.W.M., Lenaerts K., Dejong C.H.C., Blaak E.E.. **Colonic Infusions of Short-Chain Fatty Acid Mixtures Promote Energy Metabolism in Overweight/Obese Men: A Randomized Crossover Trial**. *Sci. Rep.* (2017) **7** 1-12. DOI: 10.1038/s41598-017-02546-x
27. Khosravi Z., Hadi A., Tutunchi H., Asghari-Jafarabadi M., Naeinie F., Roshanravan N., Ostadrahimi A., Fadel A.. **The Effects of Butyrate Supplementation on Glycemic Control, Lipid Profile, Blood Pressure, Nitric Oxide Level and Glutathione Peroxidase Activity in Type 2 Diabetic Patients: A Randomized Triple -Blind, Placebo-Controlled Trial**. *Clin. Nutr. ESPEN* (2022) **49** 79-85. DOI: 10.1016/j.clnesp.2022.03.008
28. Bouter K.E.C., Bakker G.J., Levin E., Hartstra A.V., Kootte R.S., Udayappan S.D., Katiraei S., Bahler L., Gilijamse P.W., Tremaroli V.. **Differential Metabolic Effects of Oral Butyrate Treatment in Lean versus Metabolic Syndrome Subjects Article**. *Clin. Transl. Gastroenterol.* (2018) **9** e155. DOI: 10.1038/s41424-018-0025-4
29. Tougaard N.H., Frimodt-Møller M., Salmenkari H., Stougaard E.B., Zawadzki A.D., Mattila I.M., Hansen T.W., Legido-Quigley C., Hörkkö S., Forsblom C.. **Effects of Butyrate Supplementation on Inflammation and Kidney Parameters in Type 1 Diabetes: A Randomized, Double-Blind, Placebo-Controlled Trial**. *J. Clin. Med.* (2022) **11**. DOI: 10.3390/jcm11133573
30. de Groot P.F., Nikolic T., Imangaliyev S., Bekkering S., Duinkerken G., Keij F.M., Herrema H., Winkelmeijer M., Kroon J., Levin E.. **Oral Butyrate Does Not Affect Innate Immunity and Islet Autoimmunity in Individuals with Longstanding Type 1 Diabetes: A Randomised Controlled Trial**. *Diabetologia* (2020) **63** 597-610. DOI: 10.1007/s00125-019-05073-8
31. Felizardo R.J.F., de Almeida D.C., Pereira R.L., Watanabe I.K.M., Doimo N.T.S., Ribeiro W.R., Cenedeze M.A., Hiyane M.I., Amano M.T., Braga T.T.. **Gut Microbial Metabolite Butyrate Protects against Proteinuric Kidney Disease through Epigenetic- and GPR109a-Mediated Mechanisms**. *FASEB J.* (2019) **33** 11894-11908. DOI: 10.1096/fj.201901080R
32. Bajka B.H., Topping D.L., Cobiac L., Clarke J.M.. **Butyrylated Starch Is Less Susceptible to Enzymic Hydrolysis and Increases Large-Bowel Butyrate More than High-Amylose Maize Starch in the Rat**. *Br. J. Nutr.* (2006) **96** 276-282. DOI: 10.1079/BJN20061807
33. Annison G., Illman R.J., Topping D.L.. **Acetylated, Propionylated or Butyrylated Starches Raise Large Bowel Short-Chain Fatty Acids Preferentially When Fed to Rats**. *J. Nutr.* (2003) **133** 3523-3528. DOI: 10.1093/jn/133.11.3523
34. Li M., Wang F., Wang J., Wang A., Yao X., Strappe P., Zhou Z., Wu Q., Guo T.. **Starch Acylation of Different Short-Chain Fatty Acids and Its Corresponding Influence on Gut Microbiome and Diabetic Indexes**. *Food Chem.* (2022) **389** 133089. DOI: 10.1016/j.foodchem.2022.133089
35. Clarke J.M., Topping D.L., Christophersen C.T., Bird A.R., Lange K., Saunders I., Cobiac L.. **Butyrate Esterified to Starch Is Released in the Human Gastrointestinal Tract**. *Am. J. Clin. Nutr.* (2011) **94** 1276-1283. DOI: 10.3945/ajcn.111.017228
36. Clarke J.M., Bird A.R., Topping D.L., Cobiac L.. **Excretion of Starch and Esterified Short-Chain Fatty Acids by Ileostomy Subjects after the Ingestion of Acylated Starches**. *Am. J. Clin. Nutr.* (2007) **86** 1146-1151. DOI: 10.1093/ajcn/86.4.1146
37. Jama H.A., Rhys-jones D., Nakai M., Yao C.K., Climie R.E., Sata Y., Anderson D., Creek D.J., Head G.A., Kaye D.M.. **Prebiotic Intervention with HAMSAB in Untreated Essential Hypertensive Patients Assessed in a Phase II Randomized Trial**. *Nat. Cardiovasc. Res.* (2023) **2** 35-43. DOI: 10.1038/s44161-022-00197-4
38. Prause M., Pedersen S.S., Tsonkova V., Qiao M., Billestrup N.. **Butyrate Protects Pancreatic Beta-cells from Cytokine-Induced Dysfunction**. *Int. J. Mol. Sci.* (2021) **22**. DOI: 10.3390/ijms221910427
39. Pedersen S.S., Prause M., Williams K., Barrès R., Billestrup N.. **Butyrate Inhibits IL-1β-Induced Inflammatory Gene Expression by Suppression of NF-ΚB Activity in Pancreatic Beta-cells**. *J. Biol. Chem.* (2022) **298** 102312. DOI: 10.1016/j.jbc.2022.102312
40. Burke S.J., Batdorf H.M., Burk D.H., Noland R.C., Eder A.E., Boulos M.S., Karlstad M.D., Jason Collier J.. **Db/Db Mice Exhibit Features of Human Type 2 Diabetes That Are Not Present in Weight-Matched C57BL/6J Mice Fed a Western Diet**. *J. Diabetes Res.* (2017) **2017** 8503754. DOI: 10.1155/2017/8503754
41. Geisler C.E., Renquist B.J.. **Hepatic Lipid Accumulation: Cause and Consequence of Dysregulated Glucoregulatory Hormones**. *J. Endocrinol.* (2017) **234** R1-R21. DOI: 10.1530/JOE-16-0513
42. Ipsen D.H., Lykkesfeldt J., Tveden-Nyborg P.. **Molecular Mechanisms of Hepatic Lipid Accumulation in Non-Alcoholic Fatty Liver Disease**. *Cell. Mol. Life Sci.* (2018) **75** 3313-3327. DOI: 10.1007/s00018-018-2860-6
43. Püschel G.P., Klauder J., Henkel J.. **Macrophages, Low-Grade Inflammation, Insulin Resistance and Hyperinsulinemia: A Mutual Ambiguous Relationship in the Development of Metabolic Diseases**. *J. Clin. Med.* (2022) **11**. DOI: 10.3390/jcm11154358
44. Walker A.W., Duncan S.H., Carol McWilliam Leitch E., Child M.W., Flint H.J.. **PH and Peptide Supply Can Radically Alter Bacterial Populations and Short-Chain Fatty Acid Ratios within Microbial Communities from the Human Colon**. *Appl. Environ. Microbiol.* (2005) **71** 3692-3700. DOI: 10.1128/AEM.71.7.3692-3700.2005
45. Firrman J., Liu L.S., Mahalak K., Tanes C., Bittinger K., Tu V., Bobokalonov J., Mattei L., Zhang H., Van Den Abbeele P.. **The Impact of Environmental PH on the Gut Microbiota Community Structure and Short Chain Fatty Acid Production**. *FEMS Microbiol. Ecol.* (2022) **98** fiac038. DOI: 10.1093/femsec/fiac038
46. Boets E., Gomand S.V., Deroover L., Preston T., Vermeulen K., De Preter V., Hamer H.M., Van den Mooter G., De Vuyst L., Courtin C.M.. **Systemic Availability and Metabolism of Colonic-Derived Short-Chain Fatty Acids in Healthy Subjects: A Stable Isotope Study**. *J. Physiol.* (2017) **595** 541-555. DOI: 10.1113/JP272613
47. Ruppin H., Bar-Meir S., Soergel K.H., Wood C.M., Schmitt M.G.. **Absorption of Short-Chain Fatty Acids by the Colon**. *Gastroenterology* (1980) **78** 1500-1507. DOI: 10.1016/S0016-5085(19)30508-6
48. Bloemen J.G., Venema K., van de Poll M.C., Olde Damink S.W., Buurman W.A., Dejong C.H.. **Short Chain Fatty Acids Exchange across the Gut and Liver in Humans Measured at Surgery**. *Clin. Nutr.* (2009) **28** 657-661. DOI: 10.1016/j.clnu.2009.05.011
49. Boland B.B., Rhodes C.J., Grimsby J.S.. **The Dynamic Plasticity of Insulin Production in β-Cells**. *Mol. Metab.* (2017) **6** 958-973. DOI: 10.1016/j.molmet.2017.04.010
50. Weickert M.O., Pfeiffer A.F.H.. **Signalling Mechanisms Linking Hepatic Glucose and Lipid Metabolism**. *Diabetologia* (2006) **49** 1732-1741. DOI: 10.1007/s00125-006-0295-3
51. Den Besten G., Bleeker A., Gerding A., Van Eunen K., Havinga R., Van Dijk T.H., Oosterveer M.H., Jonker J.W., Groen A.K., Reijngoud D.J.. **Short-Chain Fatty Acids Protect against High-Fat Diet-Induced Obesity via a Pparg-Dependent Switch from Lipogenesis to Fat Oxidation**. *Diabetes* (2015) **64** 2398-2408. DOI: 10.2337/db14-1213
52. Gart E., van Duyvenvoorde W., Toet K., Caspers M.P.M., Verschuren L., Nielsen M.J., Leeming D.J., Lima E.S., Menke A., Hanemaaijer R.. **Butyrate Protects against Diet-Induced NASH and Liver Fibrosis and Suppresses Specific Non-Canonical TGF-β Signaling Pathways in Human Hepatic Stellate Cells**. *Biomedicines* (2021) **9**. DOI: 10.3390/biomedicines9121954
53. Saponaro C., Gaggini M., Carli F., Gastaldelli A.. **The Subtle Balance between Lipolysis and Lipogenesis: A Critical Point in Metabolic Homeostasis**. *Nutrients* (2015) **7** 9453-9474. DOI: 10.3390/nu7115475
54. Kosteli A., Sugaru E., Haemmerle G., Martin J.F., Lei J., Zechner R., Ferrante A.W.. **Weight Loss and Lipolysis Promote a Dynamic Immune Response in Murine Adipose Tissue**. *J. Clin. Investig.* (2010) **120** 3466-3479. DOI: 10.1172/JCI42845
55. Anand S., Mande S.S.. **Host-Microbiome Interactions: Gut-Liver Axis and Its Connection with Other Organs**. *Npj Biofilms Microbiomes* (2022) **8** 89. DOI: 10.1038/s41522-022-00352-6
56. Xing X., Jiang Z., Tang X., Wang P., Li Y., Sun Y., Le G., Zou S.. **Sodium Butyrate Protects against Oxidative Stress in HepG2 Cells through Modulating Nrf2 Pathway and Mitochondrial Function**. *J. Physiol. Biochem.* (2016) **73** 405-414. DOI: 10.1007/s13105-017-0568-y
57. Zhao T., Gu J., Zhang H., Wang Z., Zhang W., Zhao Y., Zheng Y., Zhang W., Zhou H., Zhang G.. **Sodium Butyrate-Modulated Mitochondrial Function in High-Insulin Induced HepG2 Cell Dysfunction**. *Oxid. Med. Cell. Longev.* (2020) **2020** 1904609. DOI: 10.1155/2020/1904609
58. Yang T., Yang H., Heng C., Wang H., Chen S., Hu Y., Jiang Z., Yu Q., Wang Z., Qian S.. **Amelioration of Non-Alcoholic Fatty Liver Disease by Sodium Butyrate Is Linked to the Modulation of Intestinal Tight Junctions in Db/Db Mice**. *Food Funct.* (2020) **11** 10675-10689. DOI: 10.1039/D0FO01954B
59. Yadav H., Lee J.H., Lloyd J., Walter P., Rane S.G.. **Beneficial Metabolic Effects of a Probiotic via Butyrate-Induced GLP-1 Hormone Secretion**. *J. Biol. Chem.* (2013) **288** 25088-25097. DOI: 10.1074/jbc.M113.452516
60. Christiansen C.B., Gabe M.B.N., Svendsen B., Dragsted L.O., Rosenkilde M.M., Holst J.J.. **The Impact of Short-Chain Fatty Acids on Glp-1 and Pyy Secretion from the Isolated Perfused Rat Colon**. *Am. J. Physiol. Gastrointest. Liver Physiol.* (2018) **315** G53-G65. DOI: 10.1152/ajpgi.00346.2017
61. Tolhurst G., Heffron H., Lam Y.S., Parker H.E., Habib A.M., Diakogiannaki E., Cameron J., Grosse J., Reimann F., Gribble F.M.. **Short-Chain Fatty Acids Stimulate Glucagon-like Peptide-1 Secretion via the G-Protein-Coupled Receptor FFAR2**. *Diabetes* (2012) **61** 364-371. DOI: 10.2337/db11-1019
62. Liang L., Liu L., Zhou W., Yang C., Mai G., Li H., Chen Y.. **Gut Microbiota-Derived Butyrate Regulates Gut Mucus Barrier Repair by Activating the Macrophage/WNT/ERK Signaling Pathway**. *Clin. Sci.* (2022) **136** 291-307. DOI: 10.1042/CS20210778
63. Thaiss C.A., Levy M., Grosheva I., Zheng D., Soffer E., Blacher E., Braverman S., Tengeler A.C., Barak O., Elazar M.. **Hyperglycemia Drives Intestinal Barrier Dysfunction and Risk for Enteric Infection**. *Science* (2018) **1383** 1376-1383. DOI: 10.1126/science.aar3318
64. Stolfi C., Maresca C., Monteleone G., Laudisi F.. **Implication of Intestinal Barrier Dysfunction in Gut Dysbiosis and Diseases**. *Biomedicines* (2022) **10**. DOI: 10.3390/biomedicines10020289
65. Park J.W., Kim H.Y., Kim M.G., Jeong S., Yun C.H., Han S.H.. **Short-Chain Fatty Acids Inhibit Staphylococcal Lipoprotein-Induced Nitric Oxide Production in Murine Macrophages**. *Immune Netw.* (2019) **19** e9. DOI: 10.4110/in.2019.19.e9
66. Yin L., Laevsky G., Giardina C.. **Butyrate Suppression of Colonocyte NF-ΚB Activation and Cellular Proteasome Activity**. *J. Biol. Chem.* (2001) **276** 44641-44646. DOI: 10.1074/jbc.M105170200
67. Liu T., Li J., Liu Y., Xiao N., Suo H., Xie K., Yang C., Wu C.. **Short-Chain Fatty Acids Suppress Lipopolysaccharide-Induced Production of Nitric Oxide and Proinflammatory Cytokines through Inhibition of NF-KB Pathway in RAW264.7 Cells**. *Inflammation* (2012) **35** 1676-1684. DOI: 10.1007/s10753-012-9484-z
68. Yamazaki M., Kato A., Kato C., Fujii E., Adachi K., Miyoshi A., Suzuki M.. **Segmentation of the Pathophysiological Stages of Diabetic Changes in the Db/Db Mouse**. *J. Toxicol. Pathol.* (2009) **22** 133-137. DOI: 10.1293/tox.22.133
69. Skovsø S.. **Modeling Type 2 Diabetes in Rats Using High Fat Diet and Streptozotocin**. *J. Diabetes Investig.* (2014) **5** 349-358. DOI: 10.1111/jdi.12235
|
---
title: UBE3A and transsynaptic complex NRXN1-CBLN1-GluD1 in a hypothalamic VMHvl-arcuate
feedback circuit regulates aggression
authors:
- Yi Nong
- David C. Stoppel
- Mark A. Johnson
- Morgane Boillot
- Jelena Todorovic
- Jason Shen
- Xinyu Zhou
- Monica J.S. Nadler
- Carrie Rodriguez
- Yuda Huo
- Ikue Nagakura
- Ekkehard M. Kasper
- Matthew P. Anderson
journal: bioRxiv
year: 2023
pmcid: PMC10002692
doi: 10.1101/2023.02.28.530462
license: CC BY 4.0
---
# UBE3A and transsynaptic complex NRXN1-CBLN1-GluD1 in a hypothalamic VMHvl-arcuate feedback circuit regulates aggression
## Abstract
The circuit origins of aggression in autism spectrum disorder remain undefined. Here we report Tac1-expressing glutamatergic neurons in ventrolateral division of ventromedial hypothalamus (VMHvl) drive intermale aggression. Aggression is increased due to increases of Ube3a gene dosage in the VMHvl neurons when modeling autism due to maternal 15q11-13 triplication. Targeted deletion of increased Ube3a copies in VMHvl reverses the elevated aggression adult mice. VMHvl neurons form excitatory synapses onto hypothalamic arcuate nucleus AgRP/NPY neurons through a NRXN1-CBLN1-GluD1 transsynaptic complex and UBE3A impairs this synapse by decreasing *Cbln1* gene expression. Exciting AgRP/NPY arcuate neurons leads to feedback inhibition of VMHvl neurons and inhibits aggression. Asymptomatic increases of UBE3A synergize with a heterozygous deficiency of presynaptic Nrxn1 or postsynaptic Grid1 (both ASD genes) to increase aggression. Targeted deletions of Grid1 in arcuate AgRP neurons impairs the VMHvl to AgRP/NPY neuron excitatory synapses while increasing aggression. Chemogenetic/optogenetic activation of arcuate AgRP/NPY neurons inhibits VMHvl neurons and represses aggression. These data reveal that multiple autism genes converge to regulate the VMHvl-arcuate AgRP/NPY glutamatergic synapse. The hypothalamic circuitry implicated by these data suggest impaired excitation of AgRP/NPY feedback inhibitory neurons may explain the increased aggression behavior found in genetic forms of autism.
## One Sentence Summary:
A feedback circuit in the hypothalamus that inhibits aggression is impaired by converging autism genetic defects.
## Introduction
Autism spectrum disorders (ASDs) are early onset behavioral disorders defined by impaired social communication, increased repetitive behaviors and restrictive interests. The condition is associated with variable comorbidities such as heightened irritability including aggressive behaviors. Heightened irritability can include excessive tantrums and self-injurious aggressive behaviors that often require medical treatment. One effective medication is risperidone, an agent that acts as an atypical antipsychotic providing one of the few FDA approved treatments for irritability behavior in ASD [1]. However, the brain circuit mechanisms underlying elevated aggressive behaviors where medications might effectively modulate irritability in ASD remain undefined. Side effects of risperidone, obesity and dysregulated anterior pituitary function, correlate to major changes in gene expression that include up-regulated peptide hormone neurotransmitters such as NPY in the arcuate nucleus of the hypothalamus [2]. Activity of neurons in the adjacent ventral lateral subdivision of ventral medial hypothalamus (VMHvl) is necessary and sufficient to promote attack behaviour in mice (3–5). The glutamatergic neurons of VMHvl include subsets that express estrogen receptor alpha (Esr1), progesterone receptor (Pgr), and neuropeptide substance P (Tac1). VMHvl Esr1 and Pgr neurons were shown to promote aggressive behavior while a role for VMHvl Tac1 neurons in regulating aggression is so far only suggested by a recent single cell-seq study showing increased expression of immediate early genes in response to aggressive behavior [6].
## Results
To study the role of VMHvl Tac1 neurons in aggression, we applied the head mounted mini-microscope (Inscopix) technique to measure in vivo calcium dynamics in individual VMHvl Tac1 neurons in freely moving mice [7]. Calcium-sensitive fluorescent protein GCaMP7 was expressed selectively in VMHvl Tac1 neurons by unilateral stereotactic injection of AAV9-hSyn-DIO-GCaMP7f virus into VMHvl of Tac1-Cre male mice. A gradient index lens was implanted in VMHvl after GCaMP7 virus injection (Fig. 1A and fig. S1, A and B). Recordings during the resident-intruder aggression behavior test revealed an increase of calcium events in VMHvl Tac1 neurons during the introduction and attack of the intruder (Fig. 1A).
To test if VMHvl Tac1 neuron activity is sufficient to drive attack behavior, we expressed chemogenetic tools in VMHvl Tac1 neurons and activated these neurons during resident intruder testing [8]. AAV-hSyn-DIO-hM3D(Gq)-mCherry virus was injected bilaterally into VMHvl of Tac1-Cre male mice (fig. S1C) to express excitatory Gq-coupled designer receptor hM3D activated by clozapine-N-oxide (CNO). As a positive control, Pgr-Cre expressing neurons in VMHvl were also tested. Administering CNO (1 mg/kg i.p.) dramatically increased total attack time and attack numbers in both Tac1-Cre and Pgr-Cre male mice when compared to animals receiving saline (Fig. 1B). In wild type male mice, CNO had no effect on attack behavior (fig. S1D). Similarly, in Tac1-Cre male mice injected with virus expressing only GFP protein, CNO had no effect on attack behavior (fig. S1, E and F). These results indicate augmented activity of VMHvl Tac1 neurons is sufficient to magnify aggression.
There are now many genes implicated in ASD [9, 10] with recent studies suggesting the importance of polygenic interactions [11, 12]. One of the most common strongly penetrant forms of genetic ASD results from increased dosages of the imprinted gene UBE3A [9, 13, 14]. Maternal 15q11-13 triplications (due to a maternally-derived extranumerary isodicentric chromome 15q segment, idic15) triple the dosage of UBE3A gene expressed in mature neurons where the paternal allele is silenced. We studied Ube3a transgenic mice with additional copies of mouse full-length Ube3a gene designed to model this strongly penetrant genetic form of ASD (10–12, 15–18): Ube3a-1x mice (heterozygous for the genomic Ube3a transgene insert) modeling maternal 15q11-13 interstitial duplication and Ube3a-2x mice (homozygous for the genomic Ube3a transgene insert) modeling maternal extranumerary isodicentric chromosome 15 (idic15) (fig. S2, A and B). UBE3A encodes an E3 ubiquitin ligase and transcriptional co-regulator protein (14, 19–22). We previously showed that increasing the dosage of this non-imprinted full-length Ube3a gene in mice (Ube3a-2x mice) impairs social interactions and vocalizations while Ube3a-1x mice display more limited deficits [14]. Male Ube3a-2x mice were found to display hightened levels of aggression frequently producing severe tail bites and occasional ulcerative hindlimb wounds on their co-housed wild type littermates (fig. S2D). In the resident-intruder test, a common laboratory test for aggression, we observed increased attack of the intruder by Ube3a-2x male mice compared to their wild type male littermates (fig. S2E).
To determine the neuronal subtypes where increased UBE3A might elevate aggression we generated a conditional Ube3a-3xFLAG mouse line (LoxTB-Ube3a mouse) where the extra copies of full-length Ube3a are expressed only in the presence of Cre recombinase. By inserting a LoxP-flanked translational blocker cassette (LoxTB) into intron 1 of the mouse Ube3a gene with a C-terminal FLAG tag (Ube3a-3xFLAG), the LoxTB cassette disrupts expression of all Ube3a splice variants. When Cre recombinase is expressed in specific cell types (through cell-type specific promoters), it deletes the LoxTB cassette, permitting cell-type specific expression of the FLAG-tagged UBE3A (fig. S2F). Male mice with the Ube3a transgene selectively expressed in glutamatergic neurons (LoxTB-Ube3a-2x: VGluT2-Cre mice, Fig. 1C, lower panel), but not in GABAergic neurons (LoxTB-Ube3a-2x:VgatCre mice, Fig. 1D, lower panel), displayed increased attack towards the intruder when compared to littermates lacking Cre (Fig. 1, C and D, upper panel). No increase of attack behavior was observed in mice with Ube3a transgene selectively expressed in serotoninergic neurons (fig. S2G). The results indicate increasing Ube3a gene dosage in glutamatergic neurons is sufficient to heighten aggression.
To evaluate whether increases of Ube3a gene dosage in VMH might underlie this elevated aggression, we crossed LoxTB-Ube3a-2x and Sf1-Cre (steroidogenic factor 1 [Sf1,Nr5a1]) mice since Sf1-*Cre is* expressed predominantly in VMH [23]. LoxTB-Ube3a-2x:Sf1-Cre mice displayed increased attack behavior compared to control littermate mice (Fig. 1E). To permit a spatial as well as cell-type specific assessment of the sites where increased Ube3a heightens aggression, we generated an AAV viral vector that expresses Ube3a in a Cre-dependent manner (AAV-hSyn-DIO-Ube3a, Fig.1F and fig. S2H). Some investigators have been concerned that a C-terminal epitope tage might impair normal functions of UBE3A and so we performed a series of studies including the use of AAV-hSyn-DIO-Ube3a where not C-terminal tage is added. Stereotactic co-injections of AAV-hSyn-DIO-Ube3a and AAV-hSyn-DIO-GFP into VMHvl of VGluT2-Cre adult male mice recapitulated the heightened aggression observed in Ube3a-2x mice when compared to AAV-hSyn-DIO-GFP alone (Fig. 1F, with GFP expression in VMHvl). The results suggest increases of UBE3A in glutamatergic neurons of VMHvl elevate aggression in an ongoing manner not contingent on developmental effects.
To test if the increased UBE3A in VMHvl is necessary for the increased aggression and whether removing the elevated UBE3A in adult mice can rescue this behavioral phenotype, we generated mice in which LoxP sites flanked exons of the full-length untagged Ube3a transgene (Ube3a OFF#5 mouse, Fig. 1G, left and fig. S2I). Again, note that the Ube3a OFF#5 mouse carries a Ube3a transgene lacking an epitope tag and also increases aggression. Strikingly, injection of AAV-CMV-CreGFP in VMHvl rescued the heightened aggression behavior observed in these adult Ube3a-2x (untagged) mice (Fig. 1G). The results indicate increases of UBE3A in VMHvl neurons is both sufficient and necessary for the heighten aggression seen in this genetic ASD mouse model and suggest reversibility of aggression in adulthood.
To determine if the aggression-promoting Tac1 neuron in VMHvl might underlie this UBE3A-augmented aggression, AAV-hSyn-DIO-Ube3a (untagged) was stereotactically injected into VMHvl of Tac1-Cre male mice (Fig. 1, H and I). Progesterone-expressing neurons in VMHvl were also examined using Pgr-Cre male mice. Increasing UBE3A in either VMHvl Tac1 or Pgr neurons increases attack behavior (Fig. 1, H and I).
We previously established that UBE3A acts as a potent transcriptional regulator through its actions in the neuronal cell nucleus in brain in vivo and others have shown co-activation with nuclear hormone receptors and zinc-finger Sp1 transcription factors in vitro [13, 21]. The same gene regulatory effects were observed with a C-terminal tagged and an untagged version of Ube3a transgene and reciprocal changes in many of the same genes was observed in mice where the maternal copy of the Ube3a gene was deleted. To test whether increases of UBE3A heighten aggression through its actions in the nucleus we used our mice where the extra gene copies of Ube3a carry a C-terminal fused nuclear localization signal. A 3xFLAG tag follows two tandem copies of a nuclear localization signal (NLS) in exon 12 of the mouse Ube3a transgene (fig. S2C). Comparing mice of either nuclear-targeted or non-targeted Ube3a transgene with equivalent mRNA and protein levels [13, 14], we found that mice with non-nuclear-targeted Ube3a-1x lacked the elevated aggression whereas two different founder Ube3a-NLS-1x mouse lines (“Ube3a-NLS-3 and / Ube3a-NLS-7” mice) displayed heightened aggression (Fig. 1J). Therefore, the phenotype of nuclear-targeted Ube3a-NLS-1x mice is equivalent to the heightened aggression behavior observed in Ube3a-2x mice (non-nuclear-targeted). Targeting UBE3A to the nucleus selectively in VMHvl Tac1+ neurons, we observed that Tac1-Cre mice injected in VMHvl with AAV-DIO-Ube3a-NLS:AAV-DIO-GFP exhibited heightened aggression compared to littermate Tac1-Cre mice injected only with AAV-DIO-GFP alone (Fig. 1K and figs. S3, A and B). These mice also displayed pathologic aggression as the male mice expressing nuclear-targeted UBE3A not only attacked a male intruder, but also inflicted bite injuries on wild type female mice placed in their cages to prime them for aggression testing (fig. S3C). In contrast, when Sf1-Cre mice were injected with AAV-DIO-Ube3a-NLS:AAV-DIO-GFP in dorsal medial subdivision of ventromedial hypothalamus (VMHdm), no increase of aggression relative to littermate mice injected with AAV-DIO-GFP was observed (fig. S3, D to F), further supporting the role of VMHvl in aggression.
To test if nuclear-targeted UBE3A alters activity of VMHvl Tac1 neurons during aggression behavior, we injected AAV9-hSyn-DIO-GCaMP7f into VMHvl of Tac1-Cre male mice with or without AAV-DIO-Ube3a-NLS. VMHvl Tac1 neurons expressing Ube3a-NLS and AAV-DIO-GCaMP7f displayed increased calcium spiking activity when exposed to male intruder when compared to mice with AAV-DIO-GCaMP7f alone (Fig. 1L).
Nuclear-targeting of UBE3A magnifies its effects on gene regulation and we recently reported Cbln1 is one of the genes most strongly repressed in cerebral cortex of Ube3a-2x and Ube3a-NLS-1x mice by quantitative RT-PCR [13]. CBLN1 protein is a glutamate synapse organizer [24] and deleting Cbln1 in ventral tegmental area (VTA) reduces synaptic transmission from VTA glutamatergic neurons to medium spiny neurons of nucleus accumbens while impairing sociability [13]. Cbln1 is also strongly expressed in VMH neurons where increased Ube3a amplifies aggression (Allen Brain Atlas: Mouse Brain). We found that Cbln1 mRNA is repressed in VMH of Ube3a-NLS7-1x transgenic mice compared to wild type littermates (Fig. 2A). To further study how UBE3A regulates *Cbln1* gene expression, a 1.3 kb 5’ Cbln1 promoter sequence was introduced to drive luciferase expression. When this Cbln1 promoter-luciferase reporter was co-transfected with Ube3a-expressing or control vector, luciferase activity was reduced by as much as $54\%$ by UBE3A (Fig. 2B). The results indicate UBE3A can repress the Cbln1 promoter.
We next performed chromatin immunoprecipitation (ChIP) from wild type mouse cortex using an anti-UBE3A antibody, and quantified the levels of UBE3A-bound Cbln1 promoter sequences using quantitative PCR. The quantity of UBE3A-bound Cbln1 promoter DNA relative to input chromatin is strongly enriched (1.72 ± 0.29) when compared to mock antibody (0.39 ± 0.16, Fig. 2C). The results suggest UBE3A functions as a transcriptional co-regulator physically interacting with and repressing the 5’ Cbln1 promoter.
To determine if a loss of Cbln1 in VMH is sufficient to increase aggression behavior, we combined homozygous floxed Cbln1 (Cbln1flx/flx) mouse with Sf1-Cre to delete Cbln1 in VMH neurons (fig. S3G). Male Sf1-Cre:Cbln1flx/flx mice displayed increased attack behavior when compared to Cbln1flx/flx littermates (Fig. 2D). Injecting AAV-CMV-CreGFP (vs. AAV-CMV-GFP) in VMHvl of adult Cbln1flx/flx mice also increased attack behavior (Fig. 2E). Furthermore, deleting Cbln1 in Tac1 neurons (male mice with Tac1-Cre /Cbln1flx/flx vs. Cbln1flx/flx) increased attack behavior (Fig. 2F). The results indicate the same brain regions and cell-types underlying the increased aggression evoked by increased UBE3A, which represses Cbln1, are also the site where Cbln1 deletion increases aggression. To determine whether CBLN1 is the major target whereby UBE3A promotes aggression, we co-injected AAV-DIO-Ube3a-NLS in Tac1-Cre mice with or without AAV-hSyn-DIO-Cbln1. We found the increased aggression produced by nuclear-targeted UBE3A is reversed by an increase of CBLN1 in VMHvl Tac1 neurons (Fig. 2G). These results confirm that UBE3A increases aggression through its actions in the nucleus of VMHvl Tac1 neurons primarily by repressing *Cbln1* gene expression.
We previously used a bioinformatics strategy to examine the physical interactions between the proteins encoded by UBE3A-regulated and other ASD genes and discovered that CBLN1 physically interacts with two other gene products frequently deleted in ASD, presynaptic NRXN1 (encoded by NRXN1 gene,) and postsynaptic GluD1 (encoded by gene GRID1, Fig. 2H). CBLN1 organizes glutamate synapses by binding to NRXN1 and GluD1. This NRXN1-CBLN1-GluD1 transsynaptic trimolecular complex regulates glutamate synapse formation and maintenance [24]. Amongst NRXN and GRID gene family members, heterozygous deletions encompassing NRXN1 were found in 321 cases (second most frequent CNV after UBE3A in some case series) and GRID1 in 89 cases (Fig. 2I, AutDB [25]). UBE3A gene copy numbers are likely underestimated as autism cases diagnosed with extranumerary isodicentric chromosome (idic15) by cytogenetics are often excluded.
We found that homozygous deletion of CBLN1 binding partner genes Nrxn1 or Grid1 heightens aggression in mice as previously reported [26, 27], whereas heterozygous Nrxn1 or Grid1 deletions does not (Fig. 2, J to M), similar to the behavioral effects of Ube3a-1x (Fig. 1J). Since Ube3a-1x mice does moderately repress Cbln1 expression [13], we reasoned that haploinsufficiency of Grid1 or Nrxn1 might mimic the effects of homozygous deletion by enhancing aggression when added to the partial Cbln1-repressed state of the Ube3a-1x mice. Remarkably, heterozygous deletion of Nrxn1 or Grid1 increased aggression in an Ube3a-1x “genetic background” (Figs. 2, K and M) providing a biological proof-of-concept example whereby ASD-linked genetic defects could display polygenic interactions to effect behavior (10–12, 15–18).
The decreased expression of glutamatergic synapse organizer Cbln1 that results from increased UBE3A suggested Ube3a-2x mice might display impaired activation of a neuron population targeted by VMHvl neuron glutamate synapses during aggression behavior. To study the neuronal activity changes during aggression behavior in Ube3a-2x and littermate mice, we counted c-fos positive neurons in select brain regions immediately after aggression behavior testing (fig. S4A). We focused on brain regions which synaptically connect to VMHvl according to a previous report [28] and found changes in c-fos in several of these brain regions in Ube3a-2x mice (Fig. 3, A to F and fig. S4, B to G). The number of c-fos positive neurons was increased in brainstem outflow pathways of VMHvl including periaqueductal grey (PAG, mostly in the caudal region, Figs. 3, A and C) and dorsal raphe (fig. S4, D and E). Increased numbers of c-fos positive neurons were also observed in a specific medial ventral subregion of VMHvl (Fig. 3, B and F).
There were no changes in c-fos expression observed in the bed nucleus of the stria terminalis (BNST) (Fig. 3D), ventral premamillary nucleus (PMv), or the full VMHvl region (fig. S4, B, C). Medial amygdala (MEA), by contrast, showed a decrease in the number of c-fos positive neurons in the caudal region (fig. S4, F and G). Interestingly, the number of c-fos positive neurons was also decreased in arcuate nucleus in Ube3a-2x mice (Fig. 3, B and E).
VMHvl neurons receive multiple inputs, and project in turn to multiple brain regions. The arcuate nucleus is one of six brain regions harboring the most input-output connections with VMHvl neurons [28]. Cbln1 mRNA and Nrxn1 mRNA are highly expressed in VMHvl neurons, while, in hypothalamus, Grid1 mRNA is enriched in arcuate nucleus, in addition to its expression in multiple other brain regions (Allen Brain Atlas-mouse, Fig. 3G). These observations led us to reason that the increased aggression resulting from VMHvl-targeted loss of CLBN1 might disrupt NRXN1-CBLN1-GRID1 transsynaptic complexes to reduce activation of arcuate nucleus during aggression in Ube3a-2x mice. Consistent with this idea, a previous study found silencing neuropeptide F-expressing neurons (identified as the fly analog of NPY) increases aggression in fly [29], and chemogenetic activation of arcuate NPY/AgRP neurons decreases aggression in mice [30]. Based on these observations, we inferred that a loss of Grid1 specifically in arcuate nucleus might impair its excitatory synaptic inputs to decrease its activity leading to increased aggression. To test this hypothesis, we generated Grid1flx/flx mice and combined them with several mouse promoter-Cre lines. Deleting Grid1 in glutamatergic (Vglut2Cre:Grid1flx/flx) or GABAergic (Vgat-Cre:Grid1flx/flx) neurons failed to increase aggression (Figs. 3, H and I). Similarly, deleting Grid1 in POMC neurons (Pomc-Cre:Grid1flx/flx), a major neuron subtype in arcuate nucleus, failed to increase aggression (Fig. 3J). In contrast, deleting Grid1 in AgRP neurons (Agrp-Cre:Grid1flx/flx) increased aggression (Fig. 3K). To test if reducing Grid1 specifically in AgRP neurons of the hypothalamic arcuate nucleus increases aggression, we generated an AAV viral vector that expresses Grid1-shRNA in a Cre-dependent manner. Expressing Grid1-shRNA in arcuate AgRP neurons (Agrp-Cre) but not in arcuate POMC neurons (Pomc-Cre) increased aggression when injecting AAV-hSyn-TATAlox-Grid1-shRNA (vs. AAV-hSyn-TATAlox-scrambled-shRNA) (Fig. 3, L to N). Thus, a loss of CBLN1’s postsynaptic binding partner gene Grid1 in arcuate AgRP/NPY neurons, but not in other major neuronal cell-types, recapitulates the aggression-promoting effects of germline Grid1 deletion.
Cbln1-expressing glutamatergic cerebellar granule cell neurons form excitatory synapses onto local inhibitory neurons in a Grid1-dependent manner [31]. Caged glutamate photostimulation of the VMHvl region elicits excitatory synaptic responses in arcuate NPY/AgRP neurons [32]. Therefore, we tested whether a loss of Grid1 in arcuate NPY/AgRP neurons impairs glutamate synaptic transmission onto NPY/AgRP neurons. Whole–cell patch clamp recordings of arcuate NPY/AgRP neurons were guided using the Npy-hrGFP transgene (fig. S5, A to C). Miniature excitatory postsynaptic currents (mEPSCs) were recorded in the presence of bath voltage-gated sodium channel blocker tetrodotoxin (TTX) and GABAA receptor antagonist bicuculline. Loss of Grid1 reduced mEPSC frequency in arcuate NPY/AgRP neurons in homozygous Grid1 knockout mice with Npy-hrGFP (Fig. 3O) with no change in mEPSC amplitude (fig. S5D). Deletion of Grid1 in AgRP neurons (Agrp-Cre:Grid1flx/flx) or knockdown of Grid1 in arcuate AgRP neurons (AAV-hSyn-TATAlox-Grid1-shRNA vs. scrambled) also reduced mEPSC frequency (Fig. 3, P and Q) with no change in mEPSC amplitude (fig. S5, E). Thus, three genetic interventions that increase aggression behavior concurrently impair glutamate synapses onto arcuate NPY/AgRP neurons.
To directly examine whether decreasing arcuate AgRP/NPY neuron activity increases aggression, Cre-inducible inhibitory AAV-hSyn-DIO-hM4D(Gi) was injected into the arcuate nucleus of Agrp-Cre male mice (Fig. 4A). Applying the hM4D ligand CNO (vs. saline, i.p.) to inhibit these neurons increased attack behavior in these mice (Fig. 4B). In contrast, activating arcuate AgRP/NPY neurons decreased attack behavior using excitatory AAV-hSyn-DIO-hM3D (Gq) injected into the arcuate nucleus of Agrp-Cre male mice and applying the hM3D ligand CNO (vs. saline, i.p.) ( Fig. 4C). Injecting AAV-hSyn-DIO-GFP into these mice and applying CNO failed to change aggression behaviors (fig. S5, G and H). The results indicate that arcuate NPY/AgRP neuron activity is necessary and sufficient to inhibit aggression.
Activity of the inhibitory arcuate NPY/AgRP neurons could inhibit target neurons such as those in VMHvl to decreased aggression. To study the effects of stimulating Agrp/NPY neuron axons on activity of VMHvl neurons, we applied the optogenetics technique to specifically excite NPY/AgRP neuron axon terminals expressing light-activated excitatory opsin in Agrp-Cre crossed to Ai32 mice [33]. Ai32 mice contain a Cre-inducible allele encoding the Channelorhodopsin-2 (ChR2) [34]. We found the membrane potential of VMHvl neurons was hyperpolarized by −4.40 ± 0.50 mV when ChR2-expressing NPY/AgRP neuron axon terminals were stimulated by light (Fig. 4, D to G). This result establishes that arcuate NPY/AgRP neurons inhibit VMHvl neurons. Strikingly, chronic risperidone treatment (1 mg/kg twice daily i.p., for 15 consecutive days) magnified the NPY/AgRP axon terminal hyperpolarization of VMHvl neurons to −8.29 ± 0.98 mV (Fig. 4G). This same risperidone treatment regimen reduced the heightened aggression in Ube3a-2x male mice (Fig. 4H). These results reveal direct evidence that arcuate NPY/AgRP axon terminals provide synaptic inhibition to VMHvl neurons, and suggest that augmenting such arcuate NPY/AgRP inhibition might be a novel therapeutic approach to alleviating increased aggressive behaviors.
Loss of Grid1 impaired glutamatergic synaptic transmission onto arcuate NPY/AgRP neurons. We examine whether a loss of other components of the NRXN1-CBLN1-GRID1 transsynaptic complex or an increase of nuclear UBE3A also impairs such synapses. mEPSC frequency of arcuate NPY neurons was reduced in mice with homozygous Nrxn1 deletion, Cbln1 deletion in VMH (Sf1-Cre:Cbln1flx/flx), or increased nuclear UBE3A (Ube3a-NLS7 mice) with no change in mEPSC amplitude (fig. S6, A to F).
The mEPSCs recorded in arcuate NPY/AgRP neurons could come from a variety of afferent glutamatergic neurons. To measure synaptic transmission selectively from VMHvl glutamate neurons to arcuate NPY/AgRP neurons we deployed optogenetics to selectively photo-excite ChR2-expressing VMHvl glutamate neuron axon terminals when AAV-hSyn-DIO-ChR2 with AAV-VGlut2-mCherry-2A-Cre were co-injected into VMHvl (Fig. 4I and Fig. S6, G to H). Whole–cell patch-clamp recordings were performed in GFP-labeled arcuate NPY/AgRP neurons in Npy-hrGFP transgene mice 4 week after virus injection. The light-evoked excitatory post-synaptic currents (eEPSCs) recorded in the presence of antagonists of GABAA (bicuculline) and NMDA (D-2-amino-5-phosphonovaleric acid, D-APV) receptors were blocked by AMPA ionotropic glutamate receptor antagonist CNQX (6-cyano-7-nitroquinoxaline-2,3-dione, 30 μM, Fig. 4J). Compared to controls, the light-evoked AMPA EPSC amplitude was strongly reduced in mice with homozygous Nrxn1 or Grid1 deletion (Fig. 4K). Increasing nuclear UBE3A in VMHvl (via co-injection of AAV-Ube3a-NLS with VGluT2-Cre and AAV-hSyn-DIO-ChR2), or deleting CBLN in VMHvl (via co-injection of AAV-VGluT2-Cre and AAV-hSyn-DIO-ChR2 in VMHvl of Cbln1flx/flx mice) also markedly reduced light-evoked AMPA EPSC amplitude (Fig. 4K). Strikingly, adding CBLN1 (via co-injection of AAV-hSyn-DIO-Cbln1 with AAV-Ube3a-NLS) recovered the evoked EPSC amplitude repressed by the increased nuclear UBE3A (Fig. 4L). The results indicate glutamatergic synapses from VMHvl to arcuate NPY/AgRP neurons depend on each component of the NRXN1-CBLN1-GRID1 transsynaptic complex. Increased UBE3A promotes aggression by acting in the nucleus to impair this synapse largely by decreasing CBLN1.
## Discussion
Taken together, these results led us to propose a local circuit model where VMHvl glutamatergic neurons that express Ube3a, Nrxn1 and Cbln1 form excitatory glutamatergic synapses onto adjacent arcuate NPY/AgRP neurons that express Grid1 (Fig. 4M). In this model, VMHvl neurons form glutamatergic synapses organized by NRXN1-CBLN1-GluD1 transsynaptic complex onto arcuate NPY/AgRP neurons which then project back to inhibit VMHvl neurons. This feedback loop acts as a gate to limit aggressive behavior. The findings reveal a distinct microcircuitry defect as a basis for elevated aggressive behavior across multiple genetic forms of ASD. *The* genetic defects exert their effects by disrupting glutamatergic synapses from VMHvl to arcuate NPY/AgRP which in turn are needed to activate these feedback inhibitory neurons (Fig. 4M).
Irritability/aggressive behaviors seen in autism are a frequent reason for pharmacological treatment with antipsychotics such as risperidone, yet these medications have a range of behavioral and metabolic side effects [35, 36]. Here we show that risperidone represses the elevated aggression of Ube3a-2x mice while increasing the magnitude of arcuate AgRP/NPY mediated hyperpolarization of VMHvl neurons. This result suggests that by augmenting AgRP/NPY neuron synaptic transmission risperidone might both repress aggression while potentially also increasing feeding to cause weight gain (a known side effects). Other more complex pathways should also be considered. A study reported arcuate NPY/AgRP neurons may repress aggression through their projections to medial amygdala neurons with secondary projections to the bed nucleus of stria terminalis [30]. Other neuronal pathways have also been identified to inhibit aggression behavior including the subparaventricular zone GABAergic neurons that transduce circadian signals [37] and the lateral septum GABAergic neurons [38]. Future studies could examine if risperidone also augments arcuate NPY/AgRP neuron synaptic transmission to other targets such as the medial amygdala.
We establish that an array of converging genetic defects including increases of UBE3A or a loss of NRXN1-CBLN1-GRID1 transsynaptic complex members that are impacted by ASD-associated copy number variations impair excitatory glutamatergic synapses from VMHvl to arcuate NPY/AgRP neurons while increasing aggressive behaviors. Heterozygous NRXN1 or GRID1 deficiencies as found in some cases of ASD increase aggression in mice only when combined with moderate increases of Ube3a dosage (Ube3a-1x mice), providing a biological proof-of-concept example for how genetic interactions could give rise to behavioral deficits in neuropsychiatric diseases by acting on multiple members of a molecular pathways such as a transsynaptic complex. Importantly, the elevated aggression caused by increases of UBE3A can be rescued by viral vector-based Cbln1 expression in VMHvl neurons. Also removing the elevated UBE3A expression in VMHvl during adulthood rescues the heightened aggression. Behavioral reversibility in adulthood highlights the plasticity of this circuits and implicates its potential as a future therapeutic target to treat irritability and aggressive behaviors in ASD. Our findings, that multiple genetic forms of ASD disrupt VMHvl-arcuate AgRP/NPY neuron glutamate synapses and that augmenting the aggression-limiting feedback inhibition from arcuate to VMHvl reduces aggression pave the way for precision medicine approaches specifically directed at this neuronal circuitry.
## Animals
The Harvard Medical Area Standing Committee on Animals and the Institutional Animal Care and Use Committee of Beth Israel Deaconess Medical Center approved all mouse protocols. Mice were housed at the Center for Life Sciences barrier animal facility in sex-matched groups of 3–5 with ad libitum food and water access. Unless otherwise specified, littermate controls were used and cohorts were male only. Ube3a transgenic mice, Ube3aNLS mice and LoxTB-Ube3a mice were generated using bacterial artificial chromosome (BAC) recombineering techniques as described previously [13]. Male and female Ube3a-1x mice were crossed to result in litters containing WT, Ube3a-1x and Ube3a-2x mice. Ube3a-1x mice express similar quantities of Ube3a mRNA but encod an untargeted UBE3A protein, comparing with Ube3aNLS mice which increases of UBE3A targeted to the nucleus through a C-terminal nuclear localization signal fusion engineered into the full length Ube3a gene. Ube3a-2x mice are homozygous for the C-terminal FLAG-tagged, full-length Ube3a transgene. To determine transgenic Ube3a copy number, genomic copies of Ube3a and Lgi1 were quantified using real time PCR (Ube3a primers F: TACTGCTGAAGGTTTTCTTGGG, R: CTGCGAAATGCCTTGAATTGTT, Lgi1 primers F: ACCTAAGAGGGAACGCATTT, R: AATGATACAGTCAAAATCCT). KAPA SYBR FAST qPCR Master Mix (Kapa Biosystems) was used in 6 μ l triplicate reactions containing 2.5 ng genomic DNA and 200 nM primers. Absolute Ube3a copy numbers were calculated using the 2^−(ΔΔCT) method, multiplying by a factor of two (two Lgi1 alleles). The following lines were bred into an FVB/NJ genetic background for 6 or more generations. Grid1KO mice were generously provided by Dr. Jian Zuo (St. Judes Childrens Research Hospital, Memphis, USA). Cbln1flx/flx mice were a generous gift from Dr. Masayoshi Mishina [24] (Nigata University, Tokyo, Japan). VGluT2-Cre (Slc17a6tm2(cre)Lowl), VGaT-Cre (Slc32a1tm2(cre)Lowl), Agrp-Cre (Agrptm1(cre)Lowl) mice and Npy-hrGFP (B6.FVB-Tg(Npy-hrGFP)1Lowl/J) mice were provided as a generous gift from Dr. Bradford Lowell (BIDMC). The Sf1-Cre (Tg(Nr5a1-cre)7Lowl), Pomc-Cre (Tg(Pomc1-cre)16Lowl), Tac1-Cre (Tac1tm1.1(cre)Hze), Esr1-Cre (Esr1tm1.1(cre)And), Pgr-Cre (Pgrtm1.1(cre)Shah), and ePet-Cre (Tg(Fev-cre)1Esd) lines of mice, and Nrxn1α KO, Ai9 (RCL-tdT), Ai32/ChR2-EYFP [34] were purchased from Jackson Labs. *To* generate the conditional Grid1 knockout FVB/NJ mice, an Easi-CRISPR strategy [39]was employed to insert a loxP recombination site within each of the introns flanking exon 4 of Grid1 (NCBI ID: 14803). Chemically-modified sgRNAs were obtained from Synthego; gRNA target 1: 5’-TCCAGCCCCAGGATATAAGG-3’, gRNA target 2: 5’-CGGTTCCTTCACAGACCACG-3.’ The single-stranded DNA homology-directed repair template containing the loxP-flanked Grid1 exon 4 and homology arms was synthesized by Integrated DNA Technologies. CRISPR/Cas9-edited founder mice were generated at the BIDMC Transgenic Core Facility by microinjection of FVB/NJ zygotes with a mix of 200 ng/μl SpCas9-NLS, 50 ng/μl sgRNA1, 50 ng/μl sgRNA2 and 15 ng/μl ssDNA. Candidate founder mice were crossed to FVB/NJ to obtain germline-transmitted F1 mice. F1 mice with correct editing at the target locus were identified by sequence validation and used to establish the floxed Grid1 mouse line.
## Viral vectors
AAV2-CMV-CreGFP, AAV2-CMV-GFP, AAV2-hSyn-DIO-GFP, AAV2-hSyn-DIO-hM3D(Gq)-mCherry, AAV2-hSyn-DIO-hM4D(Gi)-mCherry and AAV2-EF1α-DIO-ChR2(E123T/T159C)-EYFP were purchased from the University of North Carolina Viral Vector Core. AAV2-hSyn-DIO-Cbln1, AAV$\frac{2}{9}$-hSyn-DIO-Ube3a, AAV$\frac{2}{9}$-hSyn-mVglut2-mCherry and AAV$\frac{2}{9}$-hSyn-mVglut2-mCherry-2A-Cre were generated as previously described [13]. AAV$\frac{2}{9}$-hSyn-DIO-Ube3a-NLS was generated by amplifying a 3xFLAG-2xNLS cassette from 1xUbe3aNLS genomic DNA using primers: 5’-TCTTCCGCATGCTGGACTATAAAGACCATGACGGTGAT-3’ 5’-TCTTCCGGATCCGGCGCGCCTTAGACCTTACGCTTCTTCTTAGGAC-3’ and subcloning into SphI and BamHI sites of the pLVX-Ube3a-IRES-mCherry construct. The resulting Ube3a-NLS cassette was then digested with SpeI and AscI and subcloned into NheI and AscI sites of the pAAV-hSyn-DIO-EGFP construct (Addgene, Plasmid #50457). pAAV-hSyn-DIO-Ube3a-NLS was then sent to the University of North Carolina Vector Core for AAV particle production with serotype 9. *To* generate the pAAV-mMeCP2-DIO-EGFP-W3SL vector, first, the W3SL enhancer sequence was amplified from Addgene Plasmid #61463 with a 5’ XhoI site addition and subcloned into EcoRI and KpnI sites of Addgene Plasmid #61591 to replace bGHpA. Next, the mMeCP2 promoter sequence was amplified from Addgene Plasmid #60957 with a 3’ SalI site addition and subcloned into XbaI and SpeI sites. Then, the DIO-EGFP cassette was amplified from Addgene Plasmid #50457 and subcloned into SalI and EcoRI sites: pAAV-mMeCP2-DIO-EGFP-W3SL. *To* generate Cre-conditional shRNA AAV constructs, the U6-TATAlox-CMV-EGFP-TATAlox cassette was amplified from pSico (Addgene Plasmid #11578) and subcloned into XbaI and XhoI sites of pAAV-mMeCP2-DIO-EGFP-W3SL to replace mMeCP2-DIO-EGFP. An HpaI site within W3SL was destroyed using the Q5® Site-Directed Mutagenesis Kit (NEB): pAAV-U6-TATAlox-CMV-EGFP-TATAlox-W3SL. For Grid1 knockdown, three Grid1 and one scrambled control shRNA hairpin sequences using TCTGCTT loop sequence were subcloned individually into HpaI and XhoI sites of pAAV-U6-TATAlox-CMV-EGFP-TATAlox-W3SL. shRNA target sequences were selected from the Broad Institute RNAi Consortium database: Grid1-A (GCTGAGAATATCCTTGGACAA) Grid1-B (GTGCTCATATTCGTGTTGAAT) Grid1-C (CGTTACAAAGGGTTCTCCATA) and
Scrambled (CCTAAGGTTAAGTCGCCCTCG).
AAV-mMeCP2-DIO-EGFP-W3SL and AAV-U6-TATAlox-CMV-EGFP-TATAlox- Scrambled - shRNA W3SL vectors were packaged using AAV9 serotype at Boston Children’s Hospital Viral Core. For the AAV$\frac{2}{9}$-U6-TATAlox-CMV-EGFP-TATAlox-Grid1-shRNA-W3SL virus, the three different Grid1 shRNA target plasmids were combined in equimolar ratio and packaged together in a single preparation. The AAV-hSyn-DIO-GCaMP7f-WPRE (AAV9 virus) are from Addgene.
## Stereotactic surgery
Anesthesia was induced in a chamber with isofluorane/oxygen following which mice were placed into a stereotaxic frame fitted with a continuous isofluorane delivery system. A single midline vertical scalp incision revealed skull landmarks. Stereotactic measurements were used to make 0.7mm wide burr holes over the entry point for bilateral viral injections directed at the VMHvl (0 deg, AP −1.5 mm, ML +/− 0.78 mm, DV −5.76 mm), VMHdm (0 deg, AP −1.46 mm, ML +/− 0.70 mm, DV −5.48 mm) and arcuate nucleus (0 deg, AP −1.7 mm, ML +/− 0.25 mm, DV −5.85 mm), all relative to bregma. 1 μl (or 0.5 μl for AAV2-EF1α-DIO-ChR2(E123T/T159C)-EYFP) of virus was infused at 0.2 μl/min through a 33 gauge Hamilton needle connected to an automated infusion pump. Following each infusion, the needle remained in place for 5 minutes. The incision was sutured closed and a single injection of 10 mg/kg meloxicam dissolved in saline was administered intraperitoneally for perioperative analgesia. All behavioral measurements were conducted approximately 28 days following surgery, and accurate targeting to the VMHvl and arcuate was confirmed by immunohistochemistry. Electrophysiological recordings were conducted at least 30 days after the surgery. The accuracy of injection sites at VMHvl and arcuate were also confirmed under microscopy when patch clamp recordings were performed.
## In vivo Ca2+ imaging with inscopix system
To prepare animals for in vivo Ca2+ imaging experiments in Tac1-Cre male mouse, a craniotomy was performed as described in stereotactic surgery above, and saline was repeatedly applied to the exposed tissue to prevent drying. The dura was then carefully removed with a 30-gauge beveled syringe needle. AAV-hSyn-DIO-GCaMP7f or AAV-hSyn-DIO-GCaMP7f + AAV-hSyn-DIO-Ube3a-NLS (300 nL) virus were injected unilaterally into VMHvl of Tac1-Cre male mice as described in stereotactic surgery procedures at the coordinates as (0 deg, AP −1.5 mm, ML +/− 0.78 mm, DV −5.76 mm). After finishing the virus injection, a ProView™ Lens Probe (0.5 mm diameter, 8.4 mm length, Inscopix, Palo Alto, CA) was implanted above the virus injection site with coordinates (−1.5 mm posterior to bregma, 0.6 mm lateral to midline, and −5.5 mm ventral to skull surface). The bottom of the lens was situated approximately 200 to 300 μm directly above the VMHvl. The portion of the lens extending ~2 to 3 mm above the skull surface was fixed with dental cement (C&B Metabond® Quick Adhesive Cement System, Parkell Inc). A silicone elastomer (Kwik-Cast; World Precision Instruments, Sarasota, FL) was applied to the top of the lens to protect the imaging surface from external damage. Four to six week later after virus injection and lens implantation, baseplate (Inscopix, Palo Alto, CA) was installed to support the miniaturized microscope. For this procedure, mice were anesthetized with isoflurane, and the silicone mold around the lens was carefully detached. Debris was removed from the exposed lens with compressed air canisters, and lens paper with ddH2O was used to clean the top of the lens. Next, the miniature microscope (nVista 3, 475-nm blue LED; Inscopix, Palo Alto, CA) with the baseplate attached was positioned above the implanted lens with an adjustable gripper (Inscopix, Palo Alto, CA). The microscope was lowered toward the top of the lens by micromanipulator until the field of view was in focus. To ensure that the objective lens was completely parallel and aligned with the implanted lens, the angle of the microscope’s position was adjusted by manually tilting the scope within the adjustable gripper. Once the field of view was in focus and the scope was parallel with the lens, the magnetic baseplate was cemented around the implanted lens. Finally, a baseplate cover (Inscopix, Palo Alto, CA) was secured into the baseplate with a set screw to protect the lens until imaging. One week later after installation of baseplate, a female mouse was added to the cage of the Tac1-Cre male mouse for one week for preparing resident intruder (R-I) test. Then the in vivo Ca2+ imaging was performed in freely moving animal when R-I tests were conducted. For each imaging session, the nVista3 microscope was connected to the baseplate on the cranium and fixed in place by the baseplate set screw. Animals acclimate in their home cage for 20 min prior to the starting imaging sessions. Calcium images were acquired using Inscopix nVista3 Data Acquisition Software v1.3.1 with collecting frame rate 10 Hz with an average exposure time of 65 ms. The analog gain on the image sensor was set to 6, while the LED power was set as 1 mW. The acquired data by the Inscopix microscope were analyzed using Inscopix Data Processing Software (IDPS version 1.6) to visualize calcium spike movies, extract calcium traces, threshold events and calculate calcium spike frequencies. The ΔF/F0 = [F(t) − F0]/F0, where the baseline image (F0) was calculated using mean frame of the input movie, and calcium imaging data were normalized as relative changes in fluorescence. The cell images and traces were identified by running PCA/ICA algorithm or manual ROI. The calcium spikes were calculated after deconvolution of cell traces. The amplitude value of the calcium spikes was export as excel file from the IDPS. The calcium spike score (sum of amplitude values of calcium spikes during the first 5 minutes male exposure epoch) was used to comparing calcium events in resident Tac1-Cre male mice expressing AAV-DIO-GCaMP7f + AAV-DIO-Ube3a-NLS to mice only injected AAV-DIO-GCaMP7f.
## Behavioral measurements
Resident intruder paradigm [40]: mature male mice (8–12 weeks old) were housed with a virgin FVB female (6–8 weeks old) for seven days in a room with an inverted light-dark cycle (The dark cycle began at 1:00 PM). Mice were transferred to the light/dark shifted room after weaning and acclimated for a minimum of two weeks. On the testing day the female mouse was removed at 10:00 AM and the first test was conducted at 1:30 PM. WT C57BL/6 male mice were used as intruder mice that were group-housed (five per cage) and matched with resident mice for approximate age (8 weeks old) and body weight (weighing less than the resident mice). The resident’s home cage was placed into a fume hood, and an intruder mouse was introduced into the resident’s cage, then their interactions were video recorded for 10 minutes. Aggressive behavior was video recorded and also scored live by a trained viewer blinded to genotypes. For studies involving DREADDs (designer receptors exclusively activated by designer drugs) [41] (i.e. AAV2-DIO-hM3D(Gq)-mCherry and AAV2-DIO-hM4D(Gi)-mCherry), approximately 28 days following AAV injections, mice received intraperitoneal injections of clozapine-N-oxide (CNO, Sigma, dissolved in saline) 10 minutes prior to the resident-intruder test.
For in vivo Ca2+ imaging with inscopix during resident-intruder test, aggression behavior was recorded with a Basler color camera, which was controlled by the Ethovision XT 15 software (Noldus). Inscopix Ca2+ imaging was synchronized with the behavior recordings through a USB I/O box which also controlled by Ethovision XT 15. Inscopix Data Processing Software was used to analyze the Ca2+ imaging data. Synchronous video recordings were used to manually score aggression behaviors.
## Immunofluorescence and c-fos quantifications
Mice were anesthetized with an intraperitoneal injection of tribromoethanol and then perfused with ice-cold PBS followed by cold $4\%$ paraformaldehyde. Brains were frozen in optimal cutting temperature (OCT) compound (Fisher Health Care) after cryoprotected in $15\%$ sucrose followed by $30\%$ sucrose (each for 24h). 16μm sections were cut on a cryostat and slide-mounted. When an anti-mouse secondary antibody was used, sections were blocked with MOM reagent (Vector) and then sections were permeabilized with $0.1\%$ Triton X100 and blocked with $10\%$ normal goat serum with $1\%$ BSA in PBS. The FLAG tag, GFP and mCherry were probed respectively with FITC-conjugated anti-FLAG (1:200, Sigma), Green Fluorescent Protein (GFP) antibody (JL-8, 1:500, Clontech), Red Fluorescent Protein Antibody (DsRed, 1:500, Clontech). The brain sections were incubated with antibodies in blocking solution at 4°C temperature overnight. Sections were washed and when required, incubated with Alexa-conjugated secondary antibodies (1:500, Invitrogen) for 3 hours at room temperature, and then mounted in Vectashield with DAPI (Vector). Fluorescent images were taken using a LSM510 confocal microscope (Zeiss), FV1000 Confocal microscope (Olympus) or the VS120-SL 5 Slide Scanner (Olympus). Images were processed using Fiji image J software.
For c-fos quantifications (Fig. 3, Fig S4), animals were sacrificed 75 minutes after resident-intruder test, then perfused with $4\%$ paraformaldehyde. Brains were frozen in OCT compound (Fisher Health Care) after cryoprotected in $15\%$ sucrose followed by $30\%$ sucrose (each for 24h). Coronal brain sections (50 μM, from bregma +3mm to −6mm) were cut on a cryostat. Sections were processed as described above for c-Fos antibody immunohistochemical staining (c-Fos Rabbit mAb #2250, 1:2000, Cell Signaling Technologies). Brain sections mounted on the slides were scanned using Zeiss Axio Scan. Z1 with 5x magnification. The whole slide images were sorted sequentially from rostral to caudal using image processing software QuPath [42] referenced to The Mouse Brain in Stereotaxic Coordinates Atlas (Paxinos and Franklin, Compact 3rd Edition). The GFP channel raw images were exported with the brain structure of interest and surrounding landmarks. Regions of Interest (ROIs) was selected on the exported GFP channel images using Fiji ImageJ software. ROIs were saved and applied for same-bregma slides in different mouse sections with minor adjustment for individual differences. A Macro including ImageJ functions of Brightness/Contrast adjust, Background Subtract, Median Filter and Auto Threshold (Yen module) was used to threshold images in each ROI consistently. The c-fos positive neurons were detected with StarDist 2D plugin after threshold, then converted to binary images for c-fos counting with Analyze Particle function using ImageJ. The masked images highlighting counted c-fos positive neurons within each ROI then be shown on raw images (fig. S4, D and F).
## Tissue preparation, RNA isolation and quantitative real time PCR (qRT-PCR)
Brains were rapidly removed and submerged in ice-cold phosphate buffered saline (PBS) for 20–30s. For the isolation of cortex (ChIP studies), two “slabs” of total cortex were dissected using curved forceps. For VMH punches, the PBS-cooled brains from wild type and Ube3aNLS7-1x mice ($$n = 6$$ each) were placed in a brain matrix (ASI Instruments) and razor blades were placed at 1mm intervals to obtain brain slices containing VMH. Punches were obtained with a 0.5mm punch core needle (Harris Unicore) and immediately frozen to −20C and stored at −80C. RNA was isolated using Trizol Reagent (ThermoFisher) and column-purified using the RNAeasy Protect Mini Kit (Qiagen). First strand cDNA synthesis was carried out with M-MLV Reverse Transcriptase (ThermoFisher) and Oligo (dT) 20. Expression of Cbln1 was quantified using RT-qPCR with Power SYBR Green Master Mix (ThermoFisher) and a Bio-Rad CFX 384 Real-Time System. Primer pairs were selected through Primer3 with specificity confirmed with Primer-BLAST. Primer sequences are listed in Supplementary Table2. Cbln1 expression between genotypes was determined using the ΔΔCT method with Syn1 as a reference gene. Melt-curve analysis and agarose gel electrophoresis were used to confirm a single PCR product of the appropriate amplicon size.
## Luciferase Assay
The Cbln1 promoter luciferase construct was purchased from GeneCopoeia, Inc. (Cat# MPRM22597-PG02), which encodes Gaussia luciferase driven by a 1331 bp genomic fragment upstream of the coding sequence of the murine *Cbln1* gene (Accession: NC_000074.6). The Ube3a expression construct was generated by amplifying the coding sequence of human Ube3a isoform III from Plasmid #37605 (Addgene) with primers 5’-TCTTCCACTAGTGCCACCATGGCCACAGCTTGTAAAAGATC-3’ and 5’-TCTTCCGGATCCTTACAGCATGCCAAATCCTTTGG-3’ and subcloning into the SpeI and BamHI sites of the pLVX-IRES-mCherry vector (Clontech Cat#631237). HEK293T cells were transfected in 96-well using Lipofectamine 3000 Reagent (ThermoFisher) with 50 ng Cbln1 promoter-Gluc plasmid and 50 ng of either pLVX-IRES-mCherry or pLVX-Ube3a-IRES-mCherry per well ($$n = 6$$ each condition). 48 hours after transfection, Gaussia luciferase activity was developed using the Secrete-Pair Gaussia Luciferase Assay Kit (Genecopoeia, Inc., Cat#LF061) and measured with the BioTek Synergy 2 luminescence plate reader.
## Chromatin immunoprecipitation
25 mg cortical tissues were dissected from wild type mice ($$n = 3$$) and chromatin was prepared using the SimpleChIP plus Enzymatic Chromatin IP Kit (Cell Signaling). UBE3A-bound chromatin was immunoprecipitated using rabbit anti-UBE3A antibody (Bethyl Laboratories). Rabbit IgG was used as a mock control. The ratio of UBE3A-bound Cbln1 promoter DNA relative to input chromatin was quantified using qRT-PCR with primers 5’-CTCGCCGCTCCTAATAACAA-3’, 5’- CCACCCTCCAGCCAATC-3’ and FAM/ZEN-labeled probe 5’-ACAGGGCAACCATTGGCTCG-3’ with Taqman Gene Expression Master Mix (ThermoFisher).
## Brain slice patch clamp electrophysiology and in vitro optogenetics
Mice were deeply anesthetized with isofluorane, following which mouse brains were rapidly removed and submerged in an ice-cold artificial cerebrospinal fluid (ACSF) containing (in mM): 126 choline chloride, 2.5 KCl, 1.2 NaH2PO4, 1.3 MgCl2, 8 MgSO4, 0.2 CaCl2, 20 glucose, and 46 NaHCO3, equilibrated with $95\%$ O2/$5\%$ CO2. Coronal brain slices containing the VMHvl or arcuate nucleus (300 μm thick) were made using a Leica VT1200S microtome (Leica Microsystems). The slices were transferred into a holding chamber containing ACSF comprised of (in mM): 124 NaCl, 3 KCl, 1 MgSO4, 1.25 NaH2PO4, 2 CaCl2, 25 Glucose, and 26 NaHCO3, equilibrated with $95\%$ O2/$5\%$ CO2. The slices were maintained at 33°C temperature for 30 minutes and then kept at room temperature until recorded. To increase viability of neurons, 5 μM glutathione, 500 μM pyruvate, and 250 μM kynurenic acid were added to the choline chloride replacement ACSF and ACSF in the holding chamber. During recordings, slices were perfused (2 ml/min) with ACSF without glutathione, pyruvate and kynurenic acid. Whole cell recordings were performed from neurons located in the VMHvl or arcuate nucleus under visual control on an upright microscope with differential interference contrast and infrared illumination. Arcuate NPY neurons were identified by GFP fluorescence labeled by NPY-hrGFP allele. Patch pipettes were pulled using a P-97 puller (Sutter Instruments) and filled with (mM): 130 K gluconate, 10 HEPES, 5 NaCl, 1 MgCl2, 0.02 EGTA, 2 MgATP, 0.5 NaGTP and 10 mM sodium phosphocreatine at a pH of 7.3 and osmolality of 275 to 290 mOsm. Filled patch pipettes had resistances of 3–5 MΩ. Synaptic currents were recorded in voltage clamp mode by an Axon MultiClamp 700B amplifier (Molecular Devices). Data were filtered at 1 kHz, digitized at 10 kHz with DigiData 1440A interface (Molecular Devices) and acquired by Clampex 10.5. Series resistance was monitored with a 5 mV hyperpolarizing step (5 ms). Resting membrane potential and action potentials (APs) were recorded in current clamp mode with APs induced by injecting 200 ms square wave pulse of positive currents ranging from 40 to 400 pA. An optical fiber (200 μm core diameter, 0.2 numerical apertures) was used to deliver optical stimulation. The fiber was placed 200 μm from the site of recording and coupled to a diode-pumped 473 nm laser (CrystaLaser Company) to control stimulus intensity light pulse duration (5ms) was controlled by a Master-8 stimulator and light pulse frequency (0.1 Hz) was determined and triggered by the Clampex 10.5. The evoked EPSCs in neurons-expressing channelrhodpsin-2 (ChR2) were induced by the laser light (30 mW)[33]. 12 consecutive sweeps were averaged with Clampfit 10.5 to determine the light-evoked EPSC amplitude. GABAA antagonist bicuculline (10 μM) and NMDA receptor antagonist D-APV (30 μM) were present to isolated the AMPA receptor currents. 30 μM 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX) was used in some studies to block and confirm AMPA receptor currents. Miniature EPSCs (mEPSCs) were recorded at a holding potential of −70 mV in the presence of GABAA antagonist bicuculline (10 μM) and tetrodotoxin (TTX, 300 nM). mEPSCs were detected and analyzed using Mini Analysis Program (Synaptosoft, Decatur, GA). Amplitude and area thresholds were set to 5 pA and 20 fC, respectively. The peak amplitude and inter-event interval of mEPSCs from 60 second episodes were used to generate cumulative probability plots, and the statistical significance was determined by Kolmogorov-Smirnov test.
## Single neuron RT-qPCR in VMHvl
The brain slices were prepared as described in electrophysiological recordings. Then single tdTomato fluorescent labeled neuron (Ai9 mice) were aspirated from the VMHvl and collected into 5 μl of 2x CellsDirect Reaction Mix (ThermoFisher, Cat# 11753-100). cDNA was generated and preamplifed 20 cycles for target sequences and treated with ExoSAP-IT (Affymetrix, Cat# 78201) as previously described [43]. Expression of Cbln1 and other target genes was quantified using RT-qPCR with Power SYBR Green Master Mix (ThermoFisher) and a Bio-Rad CFX 384 Real-Time System. Primer sequences are listed in Supplementary Table 2. Fold change in Cbln1 expression between genotypes was determined using the ΔΔCT method with Syn1 as a reference gene.
## Statistics
Sample sizes are shown in figure legends in parentheses and were chosen based on a power analysis designed to exceed a statistical power (1-beta) of $80\%$ and to meet or exceed sample sizes typically employed in similar mouse behavioral experiments. Behavioral studies performed in a blinded fashion and the code broken after the testing when data was analyzed. Where possible, animals were placed into groups through a randomization process. No animal are excluded from the data analysis. For resident–intruder aggression tests, an unpaired two-tailed Student’s t-test was used to determine statistical significance when comparing two groups. Comparisons across two groups before and after CNO treatment were analyzed by paired two-tailed Student’s t-test. Multiple groups were analyzed using 1-way ANOVA followed by Bonferroni’s Multiple Comparison Correction. 2-way Repeated-Measures ANOVA followed by Multiple Comparison was used to analyze c-fos quantification data (Fig. 3C–F and fig. S4, B, C, E and F). One-tailed Fisher’s exact test was used to analyze the data of male attacking female in (fig. S3C). The mEPSC amplitudes and inter-event intervals from 60 second episodes were used to generate cumulative probability plots, and the statistical significance was determined by Kolmogorov-Smirnov test (KS). Data are presented as mean ± standard error of the mean (SEM). $P \leq 0.05$ was considered statistically significant with ns indicating non-significant, *$P \leq 0.05$, **$P \leq 0.01$, ***$P \leq 0.001$ and ****$P \leq 0.0001.$
## Materials Availability
Mice created for this manuscript have been deposited at JAXS.
## References
1. McCracken J. T.. **Risperidone in children with autism and serious behavioral problems**. *N Engl J Med* (2002) **347** 314. PMID: 12151468
2. Wan X. Q.. **Risperidone stimulates food intake and induces body weight gain via the hypothalamic arcuate nucleus 5-HT2c receptor-NPY pathway**. *CNS Neurosci Ther* (2020) **26** 558. PMID: 31880085
3. Lee H.. **Scalable control of mounting and attack by Esr1+ neurons in the ventromedial hypothalamus**. *Nature* (2014) **509** 627. PMID: 24739975
4. Lin D.. **Functional identification of an aggression locus in the mouse hypothalamus**. *Nature* (2011) **470** 221. PMID: 21307935
5. Yang C. F.. **Sexually dimorphic neurons in the ventromedial hypothalamus govern mating in both sexes and aggression in males**. *Cell* (2013) **153** 896. PMID: 23663785
6. Kim D. W.. **Multimodal Analysis of Cell Types in a Hypothalamic Node Controlling Social Behavior**. *Cell* (2019) **179** 713. PMID: 31626771
7. Remedios R.. **Social behaviour shapes hypothalamic neural ensemble representations of conspecific sex**. *Nature* (2017) **550** 388. PMID: 29052632
8. Rogan S. C., Roth B. L.. **Remote control of neuronal signaling**. *Pharmacol Rev* (2011) **63** 291. PMID: 21415127
9. Glessner J. T.. **Autism genome-wide copy number variation reveals ubiquitin and neuronal genes**. *Nature* (2009) **459** 569. PMID: 19404257
10. Niemi M. E. K.. **Common genetic variants contribute to risk of rare severe neurodevelopmental disorders**. *Nature* (2018) **562** 268. PMID: 30258228
11. Bulik-Sullivan B. K.. **LD Score regression distinguishes confounding from polygenicity in genome-wide association studies**. *Nat Genet* (2015) **47** 291. PMID: 25642630
12. Weiner D. J.. **Polygenic transmission disequilibrium confirms that common and rare variation act additively to create risk for autism spectrum disorders**. *Nat Genet* (2017)
13. Krishnan V.. **Autism gene Ube3a and seizures impair sociability by repressing VTA Cbln1**. *Nature* (2017) **543** 507. PMID: 28297715
14. Smith S. E.. **Increased gene dosage of Ube3a results in autism traits and decreased glutamate synaptic transmission in mice**. *Sci Transl Med* (2011) **3** 103ra97
15. Anney R.. **Individual common variants exert weak effects on the risk for autism spectrum disorders**. *Hum Mol Genet* (2012) **21** 4781. PMID: 22843504
16. Gaugler T.. **Most genetic risk for autism resides with common variation**. *Nat Genet* (2014) **46** 881. PMID: 25038753
17. Grove J.. **Identification of common genetic risk variants for autism spectrum disorder**. *Nat Genet* (2019) **51** 431. PMID: 30804558
18. Klei L.. **Common genetic variants, acting additively, are a major source of risk for autism**. *Mol Autism* (2012) **3** 9. PMID: 23067556
19. Jiang Y. H.. **Mutation of the Angelman ubiquitin ligase in mice causes increased cytoplasmic p53 and deficits of contextual learning and long-term potentiation**. *Neuron* (1998) **21** 799. PMID: 9808466
20. Matsuura T.. **De novo truncating mutations in E6-AP ubiquitin-protein ligase gene (UBE3A) in Angelman syndrome**. *Nat Genet* (1997) **15** 74. PMID: 8988172
21. Nawaz Z.. **The Angelman syndrome-associated protein, E6-AP, is a coactivator for the nuclear hormone receptor superfamily**. *Mol Cell Biol* (1999) **19** 1182. PMID: 9891052
22. Scheffner M., Huibregtse J. M., Vierstra R. D., Howley P. M.. **The HPV-16 E6 and E6-AP complex functions as a ubiquitin-protein ligase in the ubiquitination of p53**. *Cell* (1993) **75** 495. PMID: 8221889
23. Dhillon H.. **Leptin directly activates SF1 neurons in the VMH, and this action by leptin is required for normal body-weight homeostasis**. *Neuron* (2006) **49** 191. PMID: 16423694
24. Uemura T.. **Trans-synaptic interaction of GluRdelta2 and Neurexin through Cbln1 mediates synapse formation in the cerebellum**. *Cell* (2010) **141** 1068. PMID: 20537373
25. Pereanu W.. **AutDB: a platform to decode the genetic architecture of autism**. *Nucleic Acids Res* (2018) **46** D1049. PMID: 29186576
26. **Grayton, M. Missler, D. A. Collier, C. Fernandes, Altered social behaviours in neurexin 1alpha knockout mice resemble core symptoms in neurodevelopmental disorders**. *PLoS One* (2013) **8** e67114. PMID: 23840597
27. Yadav R.. **Deletion of glutamate delta-1 receptor in mouse leads to aberrant emotional and social behaviors**. *PLoS One* (2012) **7** e32969. PMID: 22412961
28. Lo L.. **Connectional architecture of a mouse hypothalamic circuit node controlling social behavior**. *Proc Natl Acad Sci U S A* (2019) **116** 7503. PMID: 30898882
29. Dierick H. A., Greenspan R. J.. **Serotonin and neuropeptide F have opposite modulatory effects on fly aggression**. *Nat Genet* (2007) **39** 678. PMID: 17450142
30. Padilla S. L.. **Agouti-related peptide neural circuits mediate adaptive behaviors in the starved state**. *Nat Neurosci* (2016) **19** 734. PMID: 27019015
31. Konno K.. **Enriched expression of GluD1 in higher brain regions and its involvement in parallel fiber-interneuron synapse formation in the cerebellum**. *J Neurosci* (2014) **34** 7412. PMID: 24872547
32. Sternson S. M., Shepherd G. M., Friedman J. M.. **Topographic mapping of VMH --> arcuate nucleus microcircuits and their reorganization by fasting**. *Nat Neurosci* (2005) **8** 1356. PMID: 16172601
33. Deisseroth K.. **Optogenetics: 10 years of microbial opsins in neuroscience**. *Nat Neurosci* (2015) **18** 1213. PMID: 26308982
34. Madisen L.. **A toolbox of Cre-dependent optogenetic transgenic mice for light-induced activation and silencing**. *Nat Neurosci* (2012) **15** 793. PMID: 22446880
35. Correll C. U., Lencz T., Malhotra A. K.. **Antipsychotic drugs and obesity**. *Trends in Molecular Medicine* (2011) **17** 97. PMID: 21185230
36. Doyle C. A, McDougle C. J.. **Pharmacologic treatments for the behavioral symptoms associated with autism spectrum disorders across the lifespan**. *Dialogues Clin Neurosci* (2012) **14** 263. PMID: 23226952
37. Todd W. D.. **A hypothalamic circuit for the circadian control of aggression**. *Nat Neurosci* (2018) **21** 717. PMID: 29632359
38. Wong L. C.. **Effective Modulation of Male Aggression through Lateral Septum to Medial Hypothalamus Projection**. *Curr Biol* (2016) **26** 593. PMID: 26877081
39. Quadros R. M.. **Easi-CRISPR: a robust method for one-step generation of mice carrying conditional and insertion alleles using long ssDNA donors and CRISPR ribonucleoproteins**. *Genome Biol* (2017) **18** 92. PMID: 28511701
40. Koolhaas J. M.. **The resident-intruder paradigm: a standardized test for aggression, violence and social stress**. *J Vis Exp* (2013) e4367. PMID: 23852258
41. Urban D. J., Roth B. L.. **DREADDs (designer receptors exclusively activated by designer drugs): chemogenetic tools with therapeutic utility**. *Annu Rev Pharmacol Toxicol* (2015) **55** 399. PMID: 25292433
42. Bankhead P.. **QuPath: Open source software for digital pathology image analysis**. *Sci Rep* (2017) **7** 16878. PMID: 29203879
43. Citri A., Pang Z. P., Sudhof T. C., Wernig M., Malenka R. C.. **Comprehensive qPCR profiling of gene expression in single neuronal cells**. *Nat Protoc* (2012) **7** 118
|
---
title: A data-fusion approach to identifying developmental dyslexia from multi-omics
datasets
authors:
- Jackson Carrion
- Rohit Nandakumar
- Xiaojian Shi
- Haiwei Gu
- Yookyung Kim
- Wendy H. Raskind
- Beate Peter
- Valentin Dinu
journal: bioRxiv
year: 2023
pmcid: PMC10002702
doi: 10.1101/2023.02.27.530280
license: CC BY 4.0
---
# A data-fusion approach to identifying developmental dyslexia from multi-omics datasets
## Abstract
This exploratory study tested and validated the use of data fusion and machine learning techniques to probe high-throughput omics and clinical data with a goal of exploring the etiology of developmental dyslexia. Developmental dyslexia is the leading learning disability in school aged children affecting roughly 5–$10\%$ of the US population. The complex biological and neurological phenotype of this life altering disability complicates its diagnosis. Phenome, exome, and metabolome data was collected allowing us to fully explore this system from a behavioral, cellular, and molecular point of view. This study provides a proof of concept showing that data fusion and ensemble learning techniques can outperform traditional machine learning techniques when provided small and complex multi-omics and clinical datasets. Heterogenous stacking classifiers consisting of single-omic experts/models achieved an accuracy of $86\%$, F1 score of 0.89, and AUC value of 0.83. Ensemble methods also provided a ranked list of important features that suggests exome single nucleotide polymorphisms found in the thalamus and cerebellum could be potential biomarkers for developmental dyslexia and heavily influenced the classification of DD within our machine learning models.
## Molecular and Analytical Advances. Data Fusion
High-throughput technologies are continuously advancing, enabling researchers across various fields to utilize genomics, transcriptomics, and metabolomics for instance, to explore complex diseases and disabilities from a comprehensive perspective (1–21). When clinical data is not sufficient or practical enough to make a diagnosis, it is important to have the sufficient methodology and knowledge to utilize these multi-omics tools in making better informed decisions. Viewing complex neurodevelopmental disorders such as developmental dyslexia (DD), autism spectrum disorder (ASD), attention deficit hyperactivity disorder (ADHD), and Alzheimer’s disease through this multi-omics perspective can provide a much deeper understanding of not only the etiology of the disease or disorder but also its pathogenesis and pathophysiology [1, 2]. Machine learning (ML) and deep learning (DL) techniques can be very effective at processing single domain datasets, but these high-dimensional and heterogenous datasets seen in multi-omics research can be very complex and noisy for traditional ML methods [4]. To address these challenges, data fusion techniques have begun to provide insight on systems biology research by efficiently representing multi-omics data while also extracting the key features.
Yuan et al. [ 22] tested and validated various data fusion models to assist in the prognosis and clustering of various cancers utilizing not only traditional clinical data but also molecular data like DNA methylation, mRNA, microRNA, and protein expression. Minoura et al. [ 23] developed a mixture-of-experts model to extract biologically meaningful variables and predict surface protein expression for various cells using single-cell transcriptome, epigenome, and proteome data. Using new deep learning methodologies to explore systems biology data is still new but has shown potential in many disorders [4, 22, 23].
The two most common data fusion techniques are data-level and decision-level fusion [24]. Early fusion (data-level fusion) is designed to represent the relationship between modalities in a meaningful way and joins the data while keeping the semantics of each dataset [24]. When dealing with heterogeneous datasets, the links between the core data contain fundamental information that can be used to deep dive into the system being explored [24]. In patient specific data like clinical and bioinformatics data, these hidden links between different data types can provide researchers the insights into the entire system or pathway instead of specific time points [25]. The second form of data fusion is known as late fusion, (decision-level fusion) (2, 24–26). This can be best represented by mixture-of-experts (MoE) models, voting classifiers, stacking classifiers, deep neural networks, etc. [ 26]. Obtaining a consensus from heterogenous classifiers provides a more robust process by leveraging the strengths of the models while also mitigating their weaknesses. Assigning various models to respective modalities can improve generalization due to the division of labor among experts/models.
## Developmental Dyslexia
Developmental dyslexia (DD) is a specific reading disorder that is characterized by difficulties in word recognition, spelling, and phonological processing [27]. Dyslexia can be categorized into two main branches, developmental and acquired dyslexia. Reading difficulty resulting from a brain injury is referred to as acquired dyslexia [28]. Developmental dyslexia can have genetic components or arise from environmental causes before or during birth [29]. The prevalence of developmental dyslexia ranges from 5–$10\%$ in school aged children, depending on diagnostic criteria, making this the most common learning disability in US school children (29–31). In various studies, several traits have been reported as comorbid with dyslexia, such as attention deficit hyperactivity disorder (ADHD) (32–35), diminished general processing speed (36–38), and slowed performance during tests of rapid naming [39].
It has been estimated that the sibling recurrence risk of DD generally is 43–$60\%$, while the sibling recurrence risk of DD when both parents have been diagnosed with DD is $77\%$ [31, 40]. This suggests a heritable component of DD. Previous studies have nominated potential susceptibility genes, of which four are consistently being reported, DNAAF4, KIAA0319, DCDC2, and ROBO1 [31, 40, 41]. ROBO1 and DCDC2 have direct involvement in many neurodevelopmental processes like axon growth and neuronal migration, and they may even have a role in the proper development of the corpus callosum, while KIAA0319 is also assumed to play a direct role in developmental processes like cell adhesion [41, 42]. Although some of these genes are not fully understood on a biochemical level, they are all highly expressed in the brain [41, 42]. Various structural MRI studies on participants with and without DD have also shown significant differences between the two groups in areas of the brain like the bilateral insular cortex, cerebellum, cerebral cortex, and the thalamus (43–46). Functional MRI studies have also shown that areas like the thalamus respond abnormally in dyslexic individuals, with signs of reduced connectivity from regions like the left visual thalamus to specific cerebral cortex regions in dyslexic individuals [44, 45].
It is also thought the thalamus plays a role in the neural noise hypothesis, that individuals with dyslexia are more susceptible to neural hyperexcitability, leading to difficulty filtering out irrelevant stimuli [33]. Since the thalamus plays a role in filtering out irrelevant information supporting various gating process in the brain, it is possible that individuals with functional disruptions in the thalamus could experience heighted “noise” and difficulty with extracting the essential visual and phonological information that is needed for learning to associate written word forms with word meanings [33, 47]. Another DD hypothesis is the cerebellar deficit hypothesis, which suggest individuals with DD may have difficulties reading due to developmental abnormalities of cerebellar function [48, 49]. The cerebellum supports cognitive functions like the brain’s timing mechanism that helps coordinate the flow of information between different brain segments [48]. It has been observed that multiple cerebellar areas play a role in reading and can even provide enough insight to make predictions on an individual’s reading abilities [50]. DD is a very complex learning disability that incorporates developmental issues in various parts of the brain. Finding novel genes highly expressed in these regions of the brain may give researchers more insight and new targets to explore.
In this study, we developed and validated various ML/data-fusion models at the data and decision level to find the best way to incorporate heterogenous multi-omics datasets. Classifying dyslexic and non-dyslexic individuals can be difficult due to the complexity of the disorder, and with only clinical diagnostic approaches currently available, this exploratory study aims to prove a bioinformatics data-fusion approach can also provide accurate results. With ML techniques identifying the most influential features, feature importance was used to validate the biological importance of the most influential features.
## Participants and Phenotypic Assessments
This project was completed with the oversight and approval of the Institutional Review Boards at the University of Washington and Arizona State University as previously mentioned [51, 52]. Adults gave written consent and parents gave written permission for their minor children to participate. Altogether, 89 participants were included in the study, 44 males and 45 females, with an average age of 32.45 and a standard deviation of 19.74. Thirty-six participants were in, 2- or 3-generational families with dyslexia, with the average family size being 5.14 ranging from 4 to 7 [53].
To be included in the dyslexia group, participants were required to score below −1 SD in at least one of five measures of reading and/or spelling, described below. To be included in the typical control group, participants were required to score above −1 SD in all five measures. All participants were required to be free from any developmental or sensory condition that could introduce confounding.
Sight word recognition and nonword decoding were assessed under untimed conditions with the Word Identification (WID) and Word Attack (WATT) subtests of the Woodcock Johnson Reading Mastery Test, Third Edition [54] and under timed conditions using the Sight Word Efficiency (SWE) and Phonemic Decoding Efficiency (PDE) subtests of the Test of Word Reading Efficiency (TOWRE) [55]. Spelling ability was assessed with the Spelling subtest of the Wechsler Individual Achievement Test – Second Edition (WIAT-2) [56].
In addition to assessments of reading and spelling, participants were asked to produce rapid sequences of syllables (“papapa …,” “tatata …,” “kakaka …”) to measure motor speeds in a motor speech task. Average syllable durations were computed and transformed into z scores based on published norms [57] To measure rapid naming speeds, the RAN/RAS: Rapid Automatized Naming and Rapid Alternating Stimulus Test was administered [58]. Participants named the items in an array of 50 as rapidly as possible. This test includes the following subtests: Letters, Numbers, Colors, Objects, Letters alternating with Numbers, and Letters alternating with Numbers and Colors. For each subtest, completion time is converted to a standard score based on participant age.
## Phenotypic/Clinical Data
Due to ranges in age and other demographic features, only the standardized scores of each clinical test mentioned above were used in the analysis and the raw scores were ignored. Any participant or feature with more than $20\%$ missing values was removed from the analysis. As mentioned above, the resulting missing values were imputed with the average score for the respective phenotypic trait. These results were then scaled using the MinMaxScaler from Sklearn to transform the values to a range between 0 and 1 while keeping the original shape and distribution. This resulted in 58 participants with 47 phenotypic/clinical measurements, of whom 25 were unaffected and 33 were dyslexic.
## Genomics Data
Participants provided blood samples and exome sequencing was completed at the University of Washington Center for Mendelian Genomics (UW CMG) using the Illumina HiSeq 4000 sequencer and the CMG processing pipeline. The CMG processing pipeline consists of RTA (v2.7.6) for base calls generated in real-time, unaligned BAM files produced by Picard ExtractIlluminaBarcodes and IlluminaBasecallsToSam, and finally BWA (Burrows-Wheeler Aligner; v0.7.10) to assist in aligning BAM files to the human reference hg19hs37d5.
Read-pairs mapping within ± 2 standard deviations of the average library size (~150 ± 15 bp) were included while any read-pairs outside of this range (135 bp ~ 165 bp) were removed from further analysis. Picard MarkDuplicates (v1.111) was then used on all aligned data to remove duplicate reads that happen to have similar start positions. To adjust for indels and poor alignment around indels, GATK IndelRealigner (v3.2.2) was used for indel realignment. Finally, GATK BaseRecalibrator (v3.2.2) was used to recalibrate base qualities.
GATK HaplotypeCaller (v3.7) was used for variant detection and genotyping, resulting in the variant data containing genotype data for each individual sample. Further analysis included using the GATK VariantFiltration (v3.7) tool to mark sites of lower quality/confidence when aligned to the hg19 reference genome. Variants were annotated and allele frequency was reported using the (VEP) [59] and Combined Annotation Dependent Depletion (CADD) [60] respectively.
After sequencing the exomes of the participants and compiling a VCF file, an allele frequency PLINK [61] analysis was conducted and only significant ($p \leq 0.02$) single nucleotide polymorphisms (SNPs) were collected for future analysis. The P-value cutoff of 0.02 was used to select the 332 top scoring SNPs for further analysis. One-hot encoding was then used to better represent the participants DNA where 0, 1, and 2 represent homozygous for the reference allele, heterozygous, and homozygous for the alternative allele respectively. Of the 89 individuals that participated in this study, 54 had quality sequencing data sufficient for this study and 332 distinct SNPs were analyzed for every participant. Five of the qualified individuals were reported as ‘undetermined’ regarding their dyslexia diagnosis so they were removed from the analysis. The remaining 49 participants included 29 with dyslexia and 20 without dyslexia.
## Metabolomics Data
Participants provided saliva samples (2 ml), which were collected and stored at −80°C. Frozen saliva samples were thawed overnight at 4°C and 100 μL of each sample was used for further analysis. To isolate proteins and metabolites from the saliva, 500 μL of methanol and 50 μL standard solution (1,810.5 μM 13C3-lactate and 142 μM 13C5-glutamic acid in PBS; Sigma-Aldrich) were added to each sample. The mixture was vortexed (10 sec), stored at −20°C for 30 min, and was followed by centrifugation (14,000 RPM, 10 min, 4° C). A CentriVap Concentrator (Labconco) was used to transfer and dry the supernatants. The dried samples were then restored in a fixed volume (150 μL) of reconstitution buffer ($40\%$ (v/v) PBS (GE Healthcare), $60\%$ (v/v) acetonitrile (Fisher Scientific)).
All LC-MS/MS procedures were performed on an Agilent 1290 UPLC-6490 QQQ-MS (Santa Clara) system. Both negative and positive ionization modes were used and required 10 μL and 4 μL respectively for analysis. Liquid chromatography purification was performed using hydrophilic interaction chromatography with a Waters XBridge BEH Amide column (Waters Corporation). A fixed flow rate (0.3 mL/min) was selected, auto-sampler temperature was kept at 4 °C, and the column was set at 40 °C. The mobile phase was composed of Solvents A (10 mM ammonium acetate, 10 mM ammonium hydroxide in $95\%$ H2O/ $5\%$ acetonitrile) and B (10 mM ammonium acetate, 10 mM ammonium hydroxide in $95\%$ acetonitrile/ $5\%$ H2O). After the initial isocratic elution with $90\%$ B, the UPLC decreased solvent B to $40\%$, held the gradient for 4 minutes, and then return to $90\%$. This process was repeated for the following samples. Mass spectrometry was performed using an electrospray ionization source while multiple-reaction-monitoring (MRM) mode was used for more sensitive data collection. Agilent Masshunter Workstation (Santa Clara, CA) software was used to control the LC-MS system and extract MRM peaks. Statistical analysis of the metabolites was performed using MetaboAnalyst [62] and the metabolite levels were transformed using a log10 transformation. Linking these metabolite levels to the respective participant and class label were conducted using Pandas software in Python (v3.8.10).
Of these 296 metabolites, 28 contained over $20\%$ missing values and were subsequently removed from the study. For any remaining participants with missing specific metabolite levels, the missing values were imputed with the average metabolite level according to the other individuals in the study. These results were then scaled using the MinMaxScaler from Sklearn to transform the values to a range between 0 and 1 while keeping the original shape and distribution. From the original study, 26 of the 89 participants, 9 unaffected and 17 with dyslexia, had sufficient metabolomic data collected.
## Joint Data
Two datasets were then created by joining exome data with metabolite data and another dataset that contains the exome, metabolomic, and phenotypic datasets. Only a small subset of participants had quality data in both datasets, and this resulted in a small sample size ($$n = 19$$) in both datasets. Of the 19 participants in the analysis, only 4 were unaffected and 15 were individuals with dyslexia. Due to the imbalanced data, SMOTE [63] was used to replicate and create new data points for the minority class (random state 18 with k-nearest neighbors = 3). After utilizing SMOTE, the new balanced datasets included 30 instances where 15 were unaffected and the other 15 had dyslexia. These 30 participants provided 671 attributes including exome, metabolomic, and phenotypic features.
## Data-level fusion (Figure 1a)
Both multi-omics datasets generated by SMOTE, Exome-Metabolite and Exome-Metabolite-Phenotypic, were split $\frac{80}{20}$ for a training and test dataset respectively and analyzed using various machine learning techniques including neural network (multi-layer perceptron, MLP), random forest (RF), K-nearest neighbor (KNN), and ensemble learning techniques including AdaBoost (ADA). To optimize the various models for each dataset, grid search was implemented to tune the model and various parameters with respect to ROC AUC and paired with 10-fold cross validation. The grid search would return the most optimal model for that respective classifier. The testing dataset was then used as input for the resulting optimized models to give more accurate metrics like F1 score, accuracy, AUC. Interpretable tree-based models like AdaBoost and random forest were further examined, and features were ranked and plotted to visualize the importance of each feature. Feature importance is calculated with Gini importance using Sci-kit Learn in Python, allowing the impurity of nodes/features to be considered and reward/increase the importance of features that have purer splits. The feature importance algorithm is detailed in the Appendix. The minor allele frequencies for the important exome SNPs were determined via GnomAD v2.1 [64]
## Decision-level fusion (Figure 1b)
Several machine learning models were trained on the single-omic data sets individually. MLP, RF, KNN, ADA, and gaussian naive bayes (NB) models were trained on $80\%$ of their respective single-omic dataset and the remaining $20\%$ of samples were used for a testing set as mentioned before. Each of these models listed above were tested thoroughly with grid search to find the most optimal parameters for their respective dataset. A decision level classification model was then implemented by stacking the most optimal single-omic models and classified by a final predictor. The best classification model for exome data, metabolome data, and phenome data were implemented in the stacking classifier based upon F1 score and the final estimator was tested between Logistic Regression and Gaussian Naive Bayes. These ensemble models were then trained on $80\%$ of the joint multi-omics datasets and tested with the other $20\%$ as mentioned above.
## Data-level fusion
Table 1 summarizes the performance results of the top preforming models listed in the methods for their respective dataset. When looking at early data-fusion for exome and metabolome data, ensemble techniques like AdaBoost (learning rate of 0.0005 and 25 estimators/stumps) paired with decision tree (max depth of 4 and max leaves of 4) resulted in a model that had $86\%$ accuracy, an F1 score of 0.86, and an AUC value of 0.92 (Figure 2). Other models like deep neural networks (3 hidden layers with 300, 100, and 10 nodes in each layer) were also able to achieve $86\%$ accuracy with an F1 score of 0.89. Random forest and KNN performed the same achieving $71\%$ accuracy, an F1 score of 0.8, and an AUC value of 0.67.
When performing classification on the exome + metabolome + phenome data, MLP and KNN were the most optimal models. A MLP model with 3 hidden layers consisting of 10, 25, and 10 nodes in each layer was able to achieve $83\%$ accuracy, a F1 score of 0.89, and an AUC value of 0.83. KNN models with K equal to 3 resulted in matching accuracy and F1 score, $83\%$ and 0.89, but had an AUC value of 0.88 (Figure 2). AdaBoost (learning rate of 0.001 and 7 estimators/stumps) was able to achieve $83\%$ accuracy and an AUC value of 0.75 but had a F1 score of 0.67.
Figure 3 depicts feature importance in the interpretable AdaBoost models for the respective datasets. When exploring exome and metabolome merged data, Deoxyuridine monophosphate (dUMP) and the variant rs10988589 were identified as the most important. When using hg19 for alignment of the SNPs, rs10988589 is located at chr9:130092203 in the GPR107 gene and has a minor allele frequency of 0.227. rs62128466 (a SNP found at chr19:1487562 in the PCSK4 gene with a minor allele frequency of 0.00003) was the third most important feature, while metabolites lenalidomide and pyridoxine were fourth and fifth (equally) most important features. 6-Phosphogluconic acid, rs10215048 (a SNP found at chr7:150793348 in the TMEM176B gene with a minor allele frequency of 0.137), acetylglucosamine acid, muconic acid, and cystenine also showed minor importance when classifying individuals based soley on exome and metabolome data.
For the dataset containing phenotypic data, the top six features were also plotted and ranked based on importance. The most important feature was rs62128466 (a SNP found at chr19:1487562 in the PCSK4 gene with a minor allele frequency of 0.00003), followed by 5 features that are all equally important. Citraconic acid, the standardized score of a syllable repetition speed test (papapa), the standardized score of a rapid automatic naming of numbers test (RAN), the standardized score of the Wechsler Individual Achievement Spelling Test (WIAT), and the standardized score of the Word Identification (WID) subtest of the Woodcock Reading Mastery Test - 3.
Figure 4b shows brain-wide gene expression levels for PCSK4, the gene that contains the important SNP rs62128466 seen in both feature importance graphs, using the Allen Human Brain Atlas and Neurosynth. Neurosynth shows various gene expression levels in the brain including an increased gene expression in areas correlating to the thalamus and this can be confirmed by the standard MRI images as well in Figure 3a. The other figures, 3c and 3d, also show brain-wide gene expression levels for TMEM176B and GPR107, which included important SNPs in the genome+exome analysis.
## Decision-level fusion
Table 2 summarizes the results from various machine learning techniques on the respective single-omic datasets. Neural networks/multi-layered perceptron (MLP), random forest (RF), K-nearest neighbor (KNN), and AdaBoost (ADA), and gaussian naive bayes (NB) were all trained and tested on separate data subsets. ROC AUC values ranged from 0.5 to 0.75 for exome datasets, 0.4 to 0.7 for metabolite datasets, and 0.75 to 0.97 for phenotypic datasets. For each dataset, (exome, metabolite, and phenotypic) the following models were evaluated and optimized. MLP models optimized with gridsearch returned a model that has 3 different hidden layers with 10 nodes in each hidden level for exome data and a model that has 2 hidden layers with 10 nodes in each for metabolomic and phenotypic data. Optimized random forest models on the single-omic data resulted in trees with a maxDepth of 3 and 25 estimators. K-nearest neighbor algorithms where $$n = 3$$ were the most accurate models for the single-omic datasets. The most optimal gaussian naive bayes returned a model with smoothing parameters of 1e-8. ROC curves for each dataset can be seen in Figure 5 and the accuracy, F1 score, and AUC for each model can be found in Table 2.
The best predictors for each individual datatype, based on F1 score, were then used to construct a stacking classifier which resulted in a model that can achieve $86\%$ accuracy with a F1 score of 0.89 for Exome-Metabolite datasets and $92\%$ accuracy with an F1 score of 0.89 for Exome-Metabolite-Phenotypic datasets. Performance results for the modes mentioned above can be found in Table 3.
## Discussion
In this study, we aimed to create and compare ML models and data-fusion models, at the data (early) and decision (late) level, for their effectiveness in performing disorder classification using a combination of -omics (genomics and metabolomics) and clinical features. We applied these approaches in the context of developmental dyslexia, a disorder believed to have a complex etiology involving multiple genetic and environmental factors. We also explored the use of explainable AI techniques to identify the most influential features in making accurate predictions.
When exploring the single-omic datasets, for exome and metabolome data, we were able to consistently achieve ~$75\%$ accuracy using various traditional machine learning models. The phenome/clinical models achieved accuracy upwards of $92\%$ accuracy, which is most probably due to the ground truth diagnosis of DD being based on the individual’s performance on the clinical tests. When exploring the combination of these dataset, the biggest difference was the increase in performance seen in exome+metabolome models. Going from $75\%$ accuracy to $86\%$ accuracy is a significant boost in performance when using only exome and metabolome data. Both deep neural networks and ensemble methods like Adaboost provided high performance metrics at the data-level while at the decision-level, a stacking classifier consisting of the exome KNN model, and the metabolome RF model was able to achieve slightly higher performance metrics as seen above.
The biggest discrepancy seen in the results was the surprising decrease in performance in the models containing clinical (phenome) data when using the data-level fusion approach. Due to the clinical models achieving 0.94 – 0.96 AUC values, we assumed the addition of this datatype would have aided the models in making more educated predictions. Early fusion, data-level fusion, models that were trained only on the molecular data, exome + metabolome, received slightly higher F1 score and accuracy performance metrics compared to the heterogenous data containing both molecular and clinical data. At the data-level, we see a $3.5\%$ decrease in accuracy, 0.03 decrease in F1 score, and a 0.04 decrease in ROC AUC values when the models were exposed to the entire dataset containing the participant’s phenome. The ability to predict DD status in an individual with just the exome and metabolome at $86\%$ accuracy could provide more insight on the etiology and development of dyslexia as well as provide new early detection methods for younger and mentally challenged individuals who cannot take these clinical tests.
At the decision-level we see the expected increase in performance after the clinical data were incorporated resulting in the overall best model over both fusion techniques. The best model was a stacking classifier consisting of the exome KNN model, metabolome RF model, and the phenome/clinical RF model with logistic regression model as the final estimator. Although the F1 score is identical, 0.89, the ROC AUC increased from 0.83 to 0.94 and the accuracy changed slightly from 0.86 to 0.83 when comparing the decision-level exome+metabolome model and the decision-level exome+metabolome+phenome model respectively. This mixture-of-experts method performed better than the individual omic models when looking at exome and metabolome data. The initial $75\%$ accuracy that exome and metabolome data can achieve individually is a fair performance but when you combine the decisions of these two classifiers, the accuracy increases to $86\%$.
With the feature importance mapping we also determined that the mixture-of-experts methods do not simply weigh certain modalities higher than others but encompasses a true mixture of features from all datatypes. We can also see that the model also chose to weigh SNPs in the GPR107, PCSK4, and TMEM176B genes, which are expressed in various regions of the brain. *These* genes have not been reported yet as associated with DD, however, their respective region of high gene expression can be seen in areas that are known to affect components in the brain related to DD hypotheses. Affected regions of the brain like the cerebellum, nucleus accumbens, and thalamus can cause drastic effects in the development of one’s ability to effectively read, write, etc.
The rs62128466 SNP from the PCSK4 gene was also deemed important throughout both datasets. When analyzing the exome+metabolome data, rs62128466 was the third most important feature, and when analyzing the exome+metabolome+phenome data, rs62128466 was the most first important feature. This very rare SNP (minor allele frequency of 0.00003) is found in the PCSK4 gene, proprotein convertase subtilisn/kexin type 4, which is shown to have a high expression level in the thalamus. The thalamus has been reported to have direct influence in the diagnosis and development of DD, and looking at Figure 3b, we can see the highest levels of PCSK4 gene expression are around the center of the left and right thalamus. This potential genetic biomarker could play an influential role on the development and overall function of the thalamus, which may play a crucial role in DD according to the neural noise hypothesis.
The most important SNP in the exome+metabolome data is located in the GPR107 gene which can be seen in Figure 3d. GPR107 is a gene that codes for the G protein-coupled receptor 107 and not only does this gene show the highest gene expression levels in the brain out of the other high-importance SNPs, but it also has a high correlation to the cerebellum. G protein-coupled receptors, GPCRs, are major factors in cell communication and recognition pathways and they have been implicated in a variety of diseases as well [65]. GPCRs are expressed throughout the brain and can be influential in major senses like taste, smell, and, most importantly, vision [65]. Studies have shown that GPR107 is specifically involved with retrograde protein transport which is a very important factor in intracellular regulation [66]. The ability to use data-fusion techniques as a process to explore biologically important features from multi-omics datasets could play an important role in identifying new biomarkers or drug/therapy targets.
## Conclusion and Future studies
With high throughput technologies and as the use of systems biology knowledge grows, developing and validating new methods to analyze these heterogenous datasets and incorporate a systems biology point of view can provide more insight and make better predictions as seen above. Due to the diagnosis of DD being reliant on various clinical tests, we can see the clinical/phenome models were very accurate, showing a clinical diagnosis is still the gold standard. Although, for children under the age of 6, or individuals with hearing, slight, and other sensory/developmental abnormalities, they cannot confidently be diagnosed via the clinical tests mentioned in this study. This leads to a need for a more bioinformatics approach compared to the traditional clinical diagnosis. Both data-level and decision-level fusion methods provided better results when analyzing exome and metabolome data, compared to the traditional single-omics ML methods. This pure bioinformatics data-fusion approach has not been reported to our knowledge, in the space of DD, and compared to the $92\%$ accuracy we see in the clinical data, the ability to classify individuals with $86\%$ accuracy at both the data and decision level means exome and metabolome data can also be a good option for making accurate predictions. As states begin to require DD screening in young school children, results are showing these teacher provided screenings, that incorporate some of the clinical tests mentioned in this study, are 69–$72\%$ accurate [67]. The data-fusion method will not compete with a clinician but can perform better than teacher provided screening.
When adding an additional data type into the dataset, clinical tests, decision-level fusion methods were able to achieve the best predictions and highest performance metrics. The mixture-of-experts/ stacking classifiers models consistently outperformed any other decision-fusion model with our given dataset. Although the data-level fusion methods reacted poorly to adding another modality, the decision-level fusion was able to adjust and better its performance when accounting for the new modality. Due to traditional ML techniques resulting in $92\%$ accuracy when trained on clinical data, the use of data-fusion including clinical data is redundant and only provides more noise for the model to filter through. But we did notice how data-level and decision-level fusion methods react differently when supplied with an additional data type, suggesting decision-level fusion can be more robust and accurate when dealing with various heterogenous datasets. This knowledge can be applied to various fields of research that are incorporating multiple modalities from self-driving cars with multiple sensors to systems biology.
This is the first developmental dyslexia multi-omics study conducted with ML to our knowledge and has been able to not only allow for better screening of DD using a bioinformatics approach but also a better understanding from a biological aspect on why and how the ML models reached their decisions. These methods can be applied to an array of disorders and is not limited to just exome, metabolome, and clinical data. Mixture-of-Expert models provide a versatile approach when dealing with multi-omics datasets and other heterogeneous data. Designing and validating novel data-fusion techniques can give researchers in many fields a new way to explore multiple modalities/datatypes simultaneously. Allowing them to not only achieve higher performance metrics but also get a better understanding of the system.
As mentioned in the methods section, our biggest limitation is the fact that the multi-omics datasets are extremely small ($$n = 30$$) in this study. Future studies should include a larger sample size and should collect data evenly across both dyslexic and unaffected individuals to eliminate the need of creating synthetic data with SMOTE. This will also allow us to further validate these exploratory methods as well as find new patterns and potential biomarkers for DD. Further, different phenotypic profiles of dyslexia, e.g., neural noise vs. cerebellar underpinnings, should be added to the analysis. This could be done by including more modalities like structural and functional MRI data. As mentioned in the background, there are significant differences between a dyslexic and neurotypical brains and adding another data layer for the algorithms to differentiate against could be beneficial. Adding more diverse modalities may allow data-fusion models to gain even more insight and potentially identify DD and other disorders faster and more accurately than providers and physicians given the right data. This could lead to earlier diagnosis, new treatment approaches that are proactive and personalized, and better health, a translation of principles of precision medicine that is already in progress regarding disorders of spoken language [68]. Precision medicine is just one of many fields that can benefit greatly due to the advances of data-fusion and other advanced ML techniques.
## Funding Sources
Grant Number: R01HD088431, Arizona State University JumpStart Fund, University of Washington Royalty Research Fund
## Data and code availability
Input data can be found in the supplementary methods and the code can be found at https://github.com/jtcarrion/multiOmics_DataFusion
## References
1. Horgusluoglu E, Neff R, Song WM, Wang M, Wang Q, Arnold M. **Integrative metabolomics-genomics approach reveals key metabolic pathways and regulators of Alzheimer’s disease**. *Alzheimers Dement* (2022.0) **18** 1260-78. PMID: 34757660
2. Badhwar A, McFall GP, Sapkota S, Black SE, Chertkow H, Duchesne S. **A multiomics approach to heterogeneity in Alzheimer’s disease: focused review and roadmap**. *Brain* (2020.0) **143** 1315-31. PMID: 31891371
3. Leonavicius K, Nainys J, Kuciauskas D, Mazutis L. **Multi-omics at single-cell resolution: comparison of experimental and data fusion approaches**. *Curr Opin Biotechnol* (2019.0) **55** 159-66. PMID: 30368064
4. Leng D, Zheng L, Wen Y, Zhang Y, Wu L, Wang J. **A benchmark study of deep learning-based multi-omics data fusion methods for cancer**. *Genome Biol* (2022.0) **23** 171. PMID: 35945544
5. Dinu V, Miller PL, Zhao H. **Evidence for association between multiple complement pathway genes and AMD**. *Genetic Epidemiology* (2007.0) **31** 224-37. PMID: 17266113
6. Dinu V, Zhao H, Miller PL. **Integrating domain knowledge with statistical and data mining methods for high-density genomic SNP disease association analysis**. *Journal of Biomedical Informatics* (2007.0) **40** 750-60. PMID: 17625973
7. Xiang B, Li A, Dinu V, Nowak NJ, Zhao H, Li P. **Analytical and clinical validity of whole-genome oligonucleotide array comparative genomic hybridization for pediatric patients with mental retardation and developmental delay**. *American Journal of Medical Genetics* (2008.0) **146A** 1942-54. PMID: 18627053
8. Brown JR, Stafford P, Johnston SA, Dinu V. **Statistical methods for analyzing immunosignatures**. *BMC Bioinformatics* (2011.0) **12** 349. PMID: 21854615
9. Briones N, Dinu V. **Data mining of high density genomic variant data for prediction of Alzheimer’s disease risk**. *BMC Medical Genetics* (2012.0) **13** 7. PMID: 22273362
10. Gallitano AL, Tillman R, Dinu V, Geller B. **Family-based association study of early growth response gene 3 with child bipolar I disorder**. *Journal of Affective Disorders* (2012.0) **138** 387-96. PMID: 22370066
11. Huang K, Yellapantula V, Baier L, Dinu V. **NGSPE: A pipeline for end-to-end analysis of DNA sequencing data and comparison between different platforms**. *Computers in Biology and Medicine* (2013.0) **43** 1171-6. PMID: 23930810
12. Bradley BS, Loftus JC, Mielke CJ, Dinu V. **Differential expression of microRNAs as predictors of glioblastoma phenotypes**. *BMC Bioinformatics* (2014.0) **15** 21. PMID: 24438171
13. Huentelman MJ, Muppana L, Corneveaux JJ, Dinu V, Pruzin JJ, Reiman R. **Association of SNPs in EGR3 and ARC with Schizophrenia Supports a Biological Pathway for Schizophrenia Risk**. *PLoS One* (2015.0) **10** e0135076. PMID: 26474411
14. Day SE, Coletta RL, Kim JY, Campbell LE, Benjamin TR, Roust LR. **Next-generation sequencing methylation profiling of subjects with obesity identifies novel gene changes**. *Clinical Epigenetics* (2016.0) **8** 77. PMID: 27437034
15. Li C, Dinu V. **miR2Pathway: A novel analytical method to discover MicroRNA-mediated dysregulated pathways involved in hepatocellular carcinoma**. *Journal of Biomedical Informatics* (2018.0) **81** 31-40. PMID: 29578099
16. Li C, Liu L, Dinu V. **Pathways of topological rank analysis (PoTRA): a novel method to detect pathways involved in hepatocellular carcinoma**. *PeerJ* (2018.0) **6** e4571. PMID: 29666752
17. Peter B, Dinu V, Liu L, Huentelman M, Naymik M, Lancaster H. **Exome Sequencing of Two Siblings with Sporadic Autism Spectrum Disorder and Severe Speech Sound Disorder Suggests Pleiotropic and Complex Effects**. *Behav Genet* (2019.0) **49** 399-414. PMID: 30949922
18. Hingorani P, Dinu V, Zhang X, Lei H, Shern JF, Park J. **Transcriptome analysis of desmoplastic small round cell tumors identifies actionable therapeutic targets: a report from the Children’s Oncology Group**. *Scientific Reports* (2020.0) **10** 12318. PMID: 32703985
19. Lancaster HS, Liu X, Dinu V, Li J. **Identifying interactive biological pathways associated with reading disability**. *Brain and Behavior* (2020.0) **10** e01735. PMID: 32596987
20. Saul M, Dinu V. **Family Rank: A graphical domain knowledge informed feature ranking algorithm**. *Bioinformatics* (2021.0)
21. Ahmadinejad N, Troftgruben S, Wang J, Chandrashekar PB, Dinu V, Maley C. **Accurate Identification of Subclones in Tumor Genomes**. *Molecular Biology and Evolution* (2022.0) 39
22. Yuan Y, Van Allen EM, Omberg L, Wagle N, Amin-Mansour A, Sokolov A. **Assessing the clinical utility of cancer genomic and proteomic data across tumor types**. *Nat Biotechnol* (2014.0) **32** 644-52. PMID: 24952901
23. Minoura K, Abe K, Nam H, Nishikawa H, Shimamura T. **A mixture-of-experts deep generative model for integrated analysis of single-cell multiomics data**. *Cell Rep Methods* (2021.0) **1** 100071. PMID: 35474667
24. Lahat D, Adali T., Jutten C.. **Data Fusion: An Overview of Methods, Challenges, and Prospects**. *Proceedings of the IEEE* (2015.0) **103** 1449-77
25. Yuan Y, Savage RS, Markowetz F. **Patient-specific data fusion defines prognostic cancer subtypes**. *PLoS Comput Biol* (2011.0) **7** e1002227. PMID: 22028636
26. Ghani MKA. **Decision-level fusion scheme for nasopharyngeal carcinoma identification using machine learning techniques**. *Neural Computing and Applications* (2020.0) **32** 625-38
27. Peterson RL, Pennington BF. **Developmental dyslexia**. *Annu Rev Clin Psychol* (2015.0) **11** 283-307. PMID: 25594880
28. Castles A, Coltheart M. **Varieties of developmental dyslexia**. *Cognition* (1993.0) **47** 149-80. PMID: 8324999
29. Shastry BS. **Developmental dyslexia: an update**. *J Hum Genet* (2007.0) **52** 104-9. PMID: 17111266
30. Handler SM, Fierson WM. **Learning disabilities, dyslexia, and vision**. *Pediatrics* (2011.0) **127** e818-56. PMID: 21357342
31. Williams J, O’Donovan MC. **The genetics of developmental dyslexia**. *Eur J Hum Genet* (2006.0) **14** 681-9. PMID: 16721404
32. Germano E, Gagliano A, Curatolo P. **Comorbidity of ADHD and dyslexia**. *Dev Neuropsychol* (2010.0) **35** 475-93. PMID: 20721770
33. Hancock R, Pugh KR, Hoeft F. **Neural Noise Hypothesis of Developmental Dyslexia**. *Trends Cogn Sci* (2017.0) **21** 434-48. PMID: 28400089
34. Peter B, McCollum H, Daliri A, Panagiotides H. **Auditory gating in adults with dyslexia: An ERP account of diminished rapid neural adaptation**. *Clin Neurophysiol* (2019.0) **130** 2182-92. PMID: 31451333
35. Perrachione TK, Del Tufo SN, Winter R, Murtagh J, Cyr A, Chang P. **Dysfunction of Rapid Neural Adaptation in Dyslexia**. *Neuron* (2016.0) **92** 1383-97. PMID: 28009278
36. Peter B, Matsushita M, Raskind WH. **Global processing speed in children with low reading ability and in children and adults with typical reading ability: exploratory factor analytic models**. *Journal of speech, language, and hearing research : JSLHR* (2011.0) **54** 885-99. PMID: 21081672
37. Shanahan MA, Pennington BF, Yerys BE, Scott A, Boada R, Willcutt EG. **Processing speed deficits in attention deficit/hyperactivity disorder and reading disability**. *J Abnorm Child Psychol* (2006.0) **34** 585-602. PMID: 16850284
38. Stoodley CJ, Stein JF. **A processing speed deficit in dyslexic adults? Evidence from a peg-moving task**. *Neurosci Lett* (2006.0) **399** 264-7. PMID: 16500025
39. Catts HW, Gillispie M, Leonard LB, Kail RV, Miller CA. **The role of speed of processing, rapid naming, and phonological awareness in reading achievement**. *J Learn Disabil* (2002.0) **35** 509-24. PMID: 15493249
40. Scerri TS, Schulte-Korne G. **Genetics of developmental dyslexia**. *Eur Child Adolesc Psychiatry* (2010.0) **19** 179-97. PMID: 20091194
41. Galaburda AM, LoTurco J, Ramus F, Fitch RH, Rosen GD. **From genes to behavior in developmental dyslexia**. *Nat Neurosci* (2006.0) **9** 1213-7. PMID: 17001339
42. Cope N, Harold D, Hill G, Moskvina V, Stevenson J, Holmans P. **Strong evidence that KIAA0319 on chromosome 6p is a susceptibility gene for developmental dyslexia**. *Am J Hum Genet* (2005.0) **76** 581-91. PMID: 15717286
43. Ozernov-Palchik O, Gaab N. **Tackling the ‘dyslexia paradox’: reading brain and behavior for early markers of developmental dyslexia**. *Wiley Interdiscip Rev Cogn Sci* (2016.0) **7** 156-76. PMID: 26836227
44. Tschentscher N.. **Reduced Structural Connectivity Between Left AuditoryThalamus and the Motion-Sensitive Planum Temporale inDevelopmental Dyslexia**. *The Journal of Neuroscience* (2019.0) **39** 13
45. Diaz B, Hintz F, Kiebel SJ, von Kriegstein K. **Dysfunction of the auditory thalamus in developmental dyslexia**. *Proc Natl Acad Sci U S A* (2012.0) **109** 13841-6. PMID: 22869724
46. Stoodley CJ. **The Cerebellum and Neurodevelopmental Disorders**. *Cerebellum* (2016.0) **15** 34-7. PMID: 26298473
47. McCormick DA, Bal T. **Sensory gating mechanisms of the thalamus**. *Curr Opin Neurobiol* (1994.0) **4** 550-6. PMID: 7812144
48. Barth AE, Denton CA, Stuebing KK, Fletcher JM, Cirino PT, Francis DJ. **A test of the cerebellar hypothesis of dyslexia in adequate and inadequate responders to reading intervention**. *J Int Neuropsychol Soc* (2010.0) **16** 526-36. PMID: 20298639
49. Baillieux H, Vandervliet EJ, Manto M, Parizel PM, De Deyn PP, Marien P. **Developmental dyslexia and widespread activation across the cerebellar hemispheres**. *Brain Lang* (2009.0) **108** 122-32. PMID: 18986695
50. Li H, Yuan Q, Luo YJ, Tao W. **A new perspective for understanding the contributions of the cerebellum to reading: The cerebro-cerebellar mapping hypothesis**. *Neuropsychologia* (2022.0) **170** 108231. PMID: 35378104
51. Peter B, Lancaster H, Vose C, Middleton K, Stoel-Gammon C. **Sequential processing deficit as a shared persisting biomarker in dyslexia and childhood apraxia of speech**. *Clin Linguist Phon* (2018.0) **32** 316-46. PMID: 28933620
52. Peter B, Albert A, Gray S. **Spelling errors reveal underlying sequential and spatial processing deficits in adults with dyslexia**. *Clin Linguist Phon* (2021.0) **35** 310-39. PMID: 32552235
53. Nandakumar R.. **Joint exome and metabolome analysis in individuals with dyslexia: Evidence for associated dysregulations of olfactory perception and autoimmune functions**. *American Journal of Medical Genetics Under Review*
54. Woodcock S, Vialle W. **Are we exacerbating students’ learning disabilities? an investigation of preservice teachers’ attributions of the educational outcomes of students with learning disabilities**. *Ann Dyslexia* (2011.0) **61** 223-41. PMID: 21887592
55. Torgesen JK, Wagner RK, Rashotte CA. *Test of Word Reading Efficiency* (1999.0)
56. Wechsler D.. *Wechsler Individual Achievement Test* (2005.0)
57. Fletcher SG. **Time-by-count measurement of diadochokinetic syllable rate**. *Journal of speech and hearing research* (1972.0) **15** 763-70. PMID: 4652397
58. Wolf M, Denckla M. *RAN/RAS: Rapid automatized naming and rapid alternating stimulus test* (2005.0)
59. McLaren W, Gil L, Hunt SE, Riat HS, Ritchie GR, Thormann A. **The Ensembl Variant Effect Predictor**. *Genome Biol* (2016.0) **17** 122. PMID: 27268795
60. Kircher M, Witten DM, Jain P, O’Roak BJ, Cooper GM, Shendure J. **A general framework for estimating the relative pathogenicity of human genetic variants**. *Nature genetics* (2014.0) **46** 310-5. PMID: 24487276
61. Purcell S, Neale B, Todd-Brown K, Thomas L, Ferreira MA, Bender D. **PLINK: a tool set for whole-genome association and population-based linkage analyses**. *Am J Hum Genet* (2007.0) **81** 559-75. PMID: 17701901
62. Xia J, Psychogios N, Young N, Wishart DS. **MetaboAnalyst: a web server for metabolomic data analysis and interpretation**. *Nucleic Acids Res* (2009.0) **37** W652-60. PMID: 19429898
63. Chawla NV, Bowyer K. W., Hall L. O., Kegelmeyer W. P.. **SMOTE: Synthetic Minority Over-sampling Technique**. *Journal of Artificial Intelligence Research* (2002.0) **16** 321-57
64. Karczewski KJ, Francioli LC, Tiao G, Cummings BB, Alfoldi J, Wang Q. **The mutational constraint spectrum quantified from variation in 141,456 humans**. *Nature* (2020.0) **581** 434-43. PMID: 32461654
65. Vaidehi N, Floriano WB, Trabanino R, Hall SE, Freddolino P, Choi EJ. **Prediction of structure and function of G protein-coupled receptors**. *Proc Natl Acad Sci U S A* (2002.0) **99** 12622-7. PMID: 12351677
66. Tafesse FG, Guimaraes CP, Maruyama T, Carette JE, Lory S, Brummelkamp TR. **GPR107, a G-protein-coupled receptor essential for intoxication by Pseudomonas aeruginosa exotoxin A, localizes to the Golgi and is cleaved by furin**. *J Biol Chem* (2014.0) **289** 24005-18. PMID: 25031321
67. Fletcher JM, Francis DJ, Foorman BR, Schatschneider C. **Early Detection of Dyslexia Risk: Development of Brief, Teacher-Administered Screens**. *Learn Disabil Q* (2021.0) **44** 145-57. PMID: 34584341
68. Peter B.. **Translating principles of precision medicine into speech-language pathology: Clinical trial of a proactive speech and language intervention for infants with classic galactosemia**. *HGG Adv* (2022.0) **3**
|
---
title: Interpreting population and family-based genome-wide association studies in
the presence of confounding
authors:
- Carl Veller
- Graham Coop
journal: bioRxiv
year: 2023
pmcid: PMC10002712
doi: 10.1101/2023.02.26.530052
license: CC BY 4.0
---
# Interpreting population and family-based genome-wide association studies in the presence of confounding
## Abstract
A central aim of genome-wide association studies (GWASs) is to estimate direct genetic effects: the causal effects on an individual’s phenotype of the alleles that they carry. However, estimates of direct effects can be subject to genetic and environmental confounding, and can also absorb the ‘indirect’ genetic effects of relatives’ genotypes. Recently, an important development in controlling for these confounds has been the use of within-family GWASs, which, because of the randomness of Mendelian segregation within pedigrees, are often interpreted as producing unbiased estimates of direct effects. Here, we present a general theoretical analysis of the influence of confounding in standard population-based and within-family GWASs. We show that, contrary to common interpretation, family-based estimates of direct effects can be biased by genetic confounding. In humans, such biases will often be small per-locus, but can be compounded when effect size estimates are used in polygenic scores. We illustrate the influence of genetic confounding on population- and family-based estimates of direct effects using models of assortative mating, population stratification, and stabilizing selection on GWAS traits. We further show how family-based estimates of indirect genetic effects, based on comparisons of parentally transmitted and untransmitted alleles, can suffer substantial genetic confounding. In addition to known biases that can arise in family-based GWASs when interactions between family members are ignored, we show that biases can also arise from gene-by-environment (G×E) interactions when parental genotypes are not distributed identically across interacting environmental and genetic backgrounds. We conclude that, while family-based studies have placed GWAS estimation on a more rigorous footing, they carry subtle issues of interpretation that arise from confounding and interactions.
## Introduction
Genome-wide association studies (GWASs) have identified thousands of genetic variants that are associated with a wide variety of traits in humans. In the standard ‘population-based’ approach, the GWAS is conducted on a set of ‘unrelated’ individuals. The associations that are detected can arise when a variant causally affects the trait or when it is in tight physical linkage with causal variants nearby.
Central to the aims of GWASs is the estimation of variants’ effect sizes on traits of interest. These effect size estimates are important for identifying and prioritizing variants and implicated genes for functional followup, and may be used to form statistical predictors of trait values or to understand the causal or mechanistic role of genetic variation in traits. Understanding sources of error and bias in GWAS effect size estimates is therefore crucial.
The interpretation of GWAS effect size estimates is complicated by four broad factors (Vilhjálmsson and Nordborg 2013; Young et al. 2019). First, the causal pathways from an allele to phenotypic variation need not reside in the individuals who enrolled in the GWAS, but can also reflect causal effects on the individual’s environment of the genotypes of their siblings, parents, other ancestors, and neighbors (indirect genetic effects, or dynastic effects; Wolf et al. 1998). Second, a phenotypic association can result from correlations between the allele and environmental causes of trait variation (environmental confounding; Lander and Schork 1994). Third, a phenotypic association can be generated at a locus if it is genetically correlated with causal loci outside of its immediate genomic region (genetic confounding; Vilhjálmsson and Nordborg 2013). Fourth, an allele’s effect on a trait might depend on the environment and the allele’s genetic background (gene-by-environment and gene-by-gene interactions, or G×E and G×G; Freeman 1973; Marchini et al. 2005; Gauderman et al. 2017).
Since our primary interest here will be genetic confounding, we briefly describe some potential sources of the long-range allelic associations that drive it: population structure, assortative mating, and selection on the GWAS trait.
Population structure leads to genetic correlations across the genome when allele frequencies differ across populations or geographic regions: sampled individuals from particular populations are likely to carry, across their genomes, alleles that are common in those populations, which induces correlations among these alleles, potentially across large genomic distances. *Such* genetic correlations persist even after the populations mix, as alleles that were more common in a particular source population retain their association until uncoupled by recombination.
Assortative mating brings alleles with the same directional effect on a trait (or on multiple traits, in the case of cross-trait assortative mating) together in mates, and therefore bundles these alleles in offspring and subsequent generations. This bundling manifests as positive genetic correlations among alleles with the same directional effect (Wright 1921; Crow and Felsenstein 1968), which can confound effect size estimates in a GWAS on the trait.
Finally, natural selection on a GWAS trait can result in genetic correlations by favoring certain combinations of trait-increasing and trait-decreasing alleles. A form of selection that is expected to be common for many traits of interest is stabilizing selection, which penalizes deviations from an optimal trait value. By favoring compensating combinations of trait-increasing and trait-decreasing alleles, stabilizing selection generates negative correlations among alleles with the same directional effect (Bulmer 1971, 1974), and therefore can confound effect size estimates in a GWAS performed on the trait under selection or on a genetically correlated trait.
The potential for dynastic, environmental, and genetic confounds to bias GWAS effect size estimates has long been recognized (Lander and Schork 1994; Ewens and Spielman 1995), and so a major focus of the literature has been to develop methods to control for these confounds (Pritchard and Rosenberg 1999; Price et al. 2010). Standard approaches include using estimates of genetic relatedness as covariates in GWAS regressions (Price et al. 2006; Yang et al. 2014) or downstream analyses such as LD-Score regression (Bulik-Sullivan et al. 2015a,b; Bulik-Sullivan 2015). Such methods aim to control for both environmental and genetic confounding, but do so imperfectly (e.g., Berg et al. 2019; Sohail et al. 2019). Further, it is often unclear what features of genetic stratification are being addressed (Vilhjálmsson and Nordborg 2013; Young et al. 2019): assortative mating in particular may not be well accounted for by these methods (Border et al. 2022b). Moreover, in reality, there is no bright line separating dynastic, environmental, and genetic confounding.
One promising way forward is to estimate allelic effects within families, either by comparing the separate associations of parentally transmitted and untransmitted alleles with trait values in the offspring (Spielman et al. 1993; Allison 1997; Eaves et al. 2014; Weiner et al. 2017; Kong et al. 2018), or by associating differences in siblings’ trait values with differences in the alleles they inherited from their parents (Abecasis et al. 2000; Visscher et al. 2006; Lee et al. 2018). The idea is that, by controlling for parental genotypes, within-family association studies control for both environmental stratification and indirect/dynastic effects, while Mendelian segregation randomizes alleles across genetic backgrounds. In principle, this allows the ‘direct genetic effect’ of an allele—the causal effect of an allele carried by an individual on their trait value—to be estimated. Recognizing that a variant detected in a GWAS will usually not itself be causal for the trait variation but instead will only be correlated with true causal variants, the direct effect of a genotyped variant is usually interpreted as reflecting the direct causal effects of nearby loci that are genetically correlated with the focal locus (Young et al. 2019)—but not the effects of more distant loci that might also be genetically correlated with the focal genotyped locus (e.g., because of population structure or assortative mating).
Consistent with both the presence of substantial confounds in some population-based GWASs and the mitigation of these confounds in within-family GWASs, family-based estimates of direct effect sizes and aggregate quantities based on these estimates (e.g., SNP-based heritabilities) are substantially smaller than population GWAS estimates for a number of traits, most notably social and behavioural traits (Lee et al. 2018; Selzam et al. 2019; Mostafavi et al. 2020; Howe et al. 2022; Young et al. 2022). Likewise, estimates of genetic correlations between traits are sometimes substantially reduced when calculated using direct effect estimates from within-family GWASs (e.g. Howe et al. 2022). While some of these findings could reflect the contribution of indirect genetic effects to population GWASs, it is also likely that, at least for some traits, standard controls for population stratification in population GWASs have been insufficient (Berg et al. 2019; Sohail et al. 2019; Young et al. 2022; Okbay et al. 2022; Nivard et al. 2022; Border et al. 2022a).
Our aim in this paper is to study a general model of confounding in GWASs, to generate clear intuition for its influence on estimates of effect sizes in both population- and family-based designs. A number of the issues that we analyze have previously been raised, particularly in the context of population-based GWASs (e.g., Rosenberg and Nordborg 2006; Platt et al. 2010; Vilhjálmsson and Nordborg 2013; Young et al. 2019); here, we analyze them in a common framework that allows for comparison of multiple sources of confounding in both population and family-based GWASs. There is a large literature on GWASs in non-human organisms (e.g., Atwell et al. 2010; Hayes and Goddard 2010; Peiffer et al. 2014; Josephs et al. 2017). However, although the results and intuition that we derive here apply equally well to human and non-human GWASs, we shall interpret them primarily from the perspective of human GWASs, in which the inability to experimentally randomize environments, together with the small effects that investigators hope to detect, makes confounding a particular concern.
Our first focus is on confounding—and genetic confounding in particular—in the absence of G×E and G×G interactions. To better understand the differences between population and within-family GWASs, we first study a general model of genetic confounding in the absence of G×E and G×G interactions. We derive expressions for estimators of direct effects in both population and within-family GWASs, as functions of the true direct and indirect effects at a locus and the genetic confounds induced by other loci. In doing so, we find that family-based estimates of direct effects are in fact susceptible to genetic confounding, contrary to standard interpretation. Reassuringly, in many of the models we consider, the resulting biases are likely to be small in humans. We also address a related case: family-based GWAS designs that consider transmitted and untransmitted parental alleles and in which the indirect (or ‘dynastic’) effect of an allele is estimated from its association with the offspring’s phenotype when carried by the parent but not transmitted to the offspring. We show that this estimator of indirect effects can be substantially biased by genetic and environmental confounds, in a similar way to population estimates of direct effects. Next, we consider various sources of genetic confounding—assortative mating, population structure, and stabilizing selection on GWAS traits— and how they influence estimates of direct effects in both population and within-family GWASs.
We then turn to sibling indirect effects, which are known to bias estimates of direct effects in sibling-based GWASs (Young et al. 2019, 2022). We characterize this bias in a simple model, and contrast it to the bias caused by sibling indirect effects in a population GWAS.
Finally, we consider G×E and G×G interactions, showing how their presence can bias population and family-based estimates of direct genetic effects in contrasting ways, complicating the interpretation of family-based estimates.
## Effect size estimates in association study designs
Our primary focus will be on how genetic confounding can bias the estimation of direct genetic effects. *These* genetic confounds are due to associations between a genotyped variant at a GWAS locus and causal variants at other loci. As we will see, two kinds of association must be distinguished: cis-linkage disequilibrium (cis-LD) and trans-linkage disequilibrium (trans-LD). Genetic variants A and B are in positive cis-LD if, when an individual inherits A from a given parent, the individual is disproportionately likely to inherit B from that parent (Fig. 1A). A and B are in positive trans-LD if, when an individual inherits A from one parent, the individual is disproportionately likely to inherit B from the other parent (Fig. 1B). These covariances have also been called gametic and non-gametic LD, respectively (e.g. Weir 2008). To quantify the degrees of cis-LD and trans-LD, we denote by Dij and D˜ij the allelic covariances between focal variants at loci i and j in cis and in trans, and we denote by rij and r˜ij the analogous allelic correlation coefficients. For some of our results, it will be important to distinguish the LD present in the sample on which the association study is performed and the LD present among the parents of the sample.
Consider a trait Y influenced by genetic variants at a set of polymorphic loci L, each of which segregates for two alleles. For ease of interpretation, and without loss of generality, we designate the ‘focal’ allele at locus l ∈ L to be the allele that directly increases the trait value, and we denote by pl the frequency of this allele. Allelic effects are assumed to be additive within and across loci, such that the trait value of an individual can be written [1] Y=Y*+∑l∈Lglαld︸directeffects+∑l∈L(glm+glf)αli︸indirecteffects+ϵ. Here, gl, glm, and glf are the numbers (0, 1, or 2) of focal alleles carried at locus l by the individual, their mother, and their father respectively, αld>0 is the direct effect of the focal allele at l, and αli is its indirect effect via the maternal and paternal genotypes. ( For simplicity, we assume that indirect effects via the maternal and paternal genotypes are equal; this assumption is relaxed in Appendix A1.) ϵ is the environmental noise, with E[ϵ]=0, and Y* is the expected trait value of the offspring of parents who carry only trait-decreasing alleles.
## Population-based association studies
The variants at a genotyped locus will usually not themselves have causal effects on the trait, but will instead be in cis-LD with—and thus ‘tag’—causal variants at nearby loci. Thus, we typically think of the association at a focal genotyped locus as reflecting the direct contributions of a relatively small number of tightly linked loci, Llocal, found within tens or perhaps hundreds of kb from the focal locus (Pritchard and Przeworski 2001). Under the additive model, therefore, the standard interpretation is that a population association study performed at a focal genotyped locus λ provides an estimate of the quantity [2] αλ=1pλ(1−pλ)∑l∈LlocalDλlαld, where pλ is the frequency of the focal allele at λ, and Dλl is the degree of cis-LD between the focal allele at λ and a causal allele at a nearby locus l∈Llocal. It is reasonable to think of this quantity as the ‘direct effect’ tagged by the focal variant at the genotyped locus λ: in the absence of confounding, it can be interpreted as the average phenotypic effect of randomly choosing a non-focal allele in the population and swapping it for a focal allele, where in this hypothetical swap, the causal alleles near the locus are included. For concreteness, we assume some fixed Llocal in our analyses, but in practice researchers seldom have a pre-defined number of ‘local’ SNPs in mind.
Effect size estimation in a population GWAS is complicated by the presence of environmental and genetic stratification. Under the model in Eq. [ 1], if we perform a standard population association study at locus λ, the estimated effect of the focal allele on the trait Y is [3] α^λpop=2Vλ(∑l∈LlocalDλlαld+∑l∈L\LlocalDλlDλlαld+∑l∈LD˜λlαld︸geneticconfounds,direct+∑l∈L[Dλl′+D˜λl′+2D˜λl]αli︸geneticconfounds,indirect+12Cov(gλ,ϵ)︸environmentalconfound), where, of the cis- and trans-LD terms, Dλl and D˜λl are defined in the GWAS sample while Dλl' and D˜λl are defined in their parents (Appendix A1.2). Vλ is the genotypic variance at λ, equal to 2pλ1−pλ1+Fλ where Fλ is Wright’s coefficient of inbreeding at λ.
The environmental confound is Covgλ,ϵ/Vλ; all non-local cis- and trans-LD terms in the study sample Dλl and D˜λl,l∉Llocal) are direct genetic confounds (Fig. 1C,D); and all cis- and trans-LD terms among parents of sampled individuals Dλl' and D˜λl), together with all trans-LD terms in the study sample D˜λl, are indirect genetic confounds.
The direct genetic confounds arise because an allele carried by an offspring at λ is correlated with the alleles that they carry at other loci l∈L (via Dλl and D˜λl) that directly affect the trait value. The indirect genetic confounds arise because an allele carried by the offspring at λ—say, the maternal allele—is correlated with alleles carried by the offspring’s mother at other loci Dλl' and (D˜λl') and alleles carried by their father (as reflected by the trans-LD in the offspring, D˜λl). These alleles in the parents can indirectly affect the offspring’s trait value.
Thus, as is now well appreciated, population-based GWASs potentially suffer from many types of confounds (Vilhjálmsson and Nordborg 2013; Young et al. 2019). In practice, they can be reduced by including principal components—which capture genome-wide relatedness among GWAS participants—as regressors in a GWAS, or by using relatedness matrices in mixed models (Price et al. 2006; Yang et al. 2014). However, it is often unclear exactly what these methods control for in a given application (Vilhjálmsson and Nordborg 2013; Young et al. 2019), and they have been shown to be inadequate in important cases (e.g., Berg et al. 2019; Sohail et al. 2019). When principal components (or other controls) fail to account fully for stratification, then Eq. [ 3] can be interpreted as a decomposition of the remaining, uncontrolled-for confounding in the GWAS.1
## Within-family association studies
The two within-family association study designs that we consider are parent-offspring GWASs and sibling GWASs. Other designs have been proposed to control for genetic and environmental confounding in the estimation of aggregate quantities such as heritability (e.g., Young et al. 2018a), but our primary focus is on the estimation of single-marker effect sizes. We do later turn to the interpretation of polygenic score regressions within families.
## Estimates of direct genetic effects.
Parent-offspring studies can be used to estimate trait associations separately for parentally transmitted and untransmitted variants at a locus λ, αˆλ(T) and αˆλ(U), by regressing the trait value Y on the transmitted and untransmitted genotypes, gλT and gλU (Kong et al. 2018). The aim is often to estimate the direct effect of a variant, αˆλd, as the difference between these two estimates: [4] αˆλd,T−U=αˆλ(T)−αˆλU. A second aim is to treat αˆλ(U) as an estimate of the indirect, or family, effect of the variant. We return to this second aim later.
In Appendix A1.4, we show that, in the absence of interactions between parental and offspring genotypes, the estimate of the direct effect of a variant at locus λ in a parent-offspring study is [5] α^λd,T−U=α^λ(T)−α^λ(U)=2Vλ∑l∈L(1−2cλl)(Dλl′−D˜λl′)αld [6] ≈2Vλ(∑l∈LlocalDλl′αld+∑l∈L\Llocal(1−2cλl)(Dλl′−D˜λl′)αld)︸geneticconfounds,direct, where cλl is the sex-averaged recombination rate between λ and l. The cis- and trans-LD terms Dλl' and D˜λl′ are measured in the parents.
Similarly, an estimate of the direct effect can be obtained from pairs of siblings by regressing the differences in their phenotypes on the differences in their genotypes at the focal locus λ. In the presence of genetic confounds, this procedure yields a similar estimate to Eq. [ 6]: [7] α^λd,sib≈2Hλ(∑l∈LlocalDλl′αld+∑l∈L\Llocal(1−2cλl)(Dλl′−D˜λl′)αld︸geneticconfounds,direct), where Hλ is the fraction of parents who are heterozygous at locus λ (Appendix A1.3). An assumption in sibling GWASs is that an offspring’s phenotype is not influenced by the genotypes of their siblings — i.e., that there are no sibling indirect genetic effects. We consider violations of this assumption later.
In Eqs. [ 6] and [7], there is no environmental confound, because family-based GWASs successfully randomize the environments of family members with respect to within-family genetic transmission.
The derivations above further show that, while population association studies are biased by sums of trans- and cis-LD between the focal locus and all causal loci (Eq. 3), within-family association studies are instead biased by differences between trans- and cis-LD, and moreover, that the biases in within-family studies are driven only by LD between the focal locus and causal loci on the same chromosome cλl<$\frac{1}{2.}$ To provide an intuition for this result, we focus our discussion on a sibling association study performed at λ; the intuition is identical for the analogous parent-offspring study.
Because the difference between two siblings in their maternally inherited genotypes is independent of the difference in their paternally inherited genotypes, we may consider maternal and paternal transmissions separately in studying how a locus l∈L can confound effect size estimation at λ in a sibling association study. We will phrase our discussion in terms of maternal transmission.
For effect size estimation at λ to be genetically confounded by maternal transmission at a distant locus l, the mother must be heterozygous at both loci. For if she were homozygous at l, then maternal transmission at l could not contribute to any trait differences between her offspring, while if she were homozygous at λ, maternal transmission would not result in genetic variation among her offspring at λ with which trait variation could be associated. Therefore, we restrict our focus to mothers who are heterozygous at both λ and l, or ‘double heterozygotes’. Two kinds exist (Fig. 1E,F): coupling double heterozygotes who carry the focal alleles at λ and l on the same haploid genome (‘in cis’), and repulsion double heterozygotes who carry them on separate haploid genomes (‘in trans’).
We first consider the case where the recombination rate between λ and l is small cλl≪$\frac{1}{2.}$ *In this* case, if the mother is a coupling double heterozygote, then her offspring will tend to inherit either both or neither of the focal alleles at λ and l (Fig. 1E). Therefore, if one sibling inherits the focal allele at λ and another does not, the first sibling will tend to inherit the focal (trait-increasing, as we have defined it) allele at l and the second sibling will not, so that the effect of locus l positively confounds the association between λ and the trait (Fig. 1E). If the mother is instead a repulsion double heterozygote, then her offspring will tend to inherit either the focal allele at λ or the focal allele at l, but not both (Fig. 1F). In this case, if one sibling inherits the focal allele at λ and another does not, the second sibling will tend to inherit the focal (trait-increasing) allele at l and the first sibling will not, so that the effect of locus l negatively confounds the association between λ and the trait (Fig. 1F). When λ and l are linked, therefore, the way in which l genetically confounds the effect size estimate at λ depends, positively or negatively, on whether the fraction of coupling double heterozygotes among parents is greater or smaller, respectively, than the fraction of repulsion double heterozygotes.
In contrast, if λ and l are unlinked cλl=$\frac{1}{2}$, then transmissions from coupling and repulsion double heterozygote parents are equal, and so l cannot confound estimates at λ (Fig. 1E,F). Put differently, meiosis in double heterozygotes fully randomizes joint allelic transmissions at λ and l, with offspring equally likely to inherit any possible combination of alleles at the two loci.
Therefore, only linked loci l can confound a family-based association study at λ, and they do so in proportion to (i) how small the recombination rate between λ and l is, and (ii) the difference between the fractions of parents who are coupling and repulsion double heterozygotes at λ and l. Accordingly, if we write these fractions of parents as Hλlcoup and Hλlrep, then Dλl′−D˜λl′=(Hλlcoup−Hλlrep)/2, and so Eq. [ 7] (and Eq. 6) can be rewritten in terms of the relative frequencies of the two kinds of double-heterozygotes: αˆλd,sib≈2Hλ∑l∈Llocal Dλl′αld+∑l∈L\Llacal 12−cλlHλlcoup−Hλlrepαld.
In a species with many chromosomes, such as humans, for a given locus, there will be many more unlinked loci than linked loci. Therefore, the set of loci that can confound a family-based association study at a given locus will be much smaller than the set of loci that can confound a population association study at the locus. It will often be the case, therefore, that biases in the estimation of direct genetic effects will be smaller in family-based studies than in population studies, a point that we explore below when we consider sources of genetic confounding.
## Estimates of indirect genetic effects.
We now return to the regression of the trait on the untransmitted genotype in parent-offspring GWASs, αˆλ(U), which has sometimes been treated as an estimate of the indirect effect αˆλi. Assuming equal indirect effects via maternal and paternal genotypes (an assumption that we relax in Appendix A1.4), [8] α^λi=α^λ(U)=2Vλ(∑l∈LlocalDλl′αli︸localindirecteffect+∑l∈L(Dλl′cλl+D˜λl′(1−cλl)+D˜λl)αld︸geneticconfounds,direct+∑l∈L\LlocalDλl′αli+∑l∈L(D˜λl′+2D˜λl)αli︸geneticconfounds,indirect+12Cov(gλU,ϵ)︸environmentalconfound). The direct genetic confound reflects associations of the untransmitted alleles at the focal locus with alleles that are transmitted to the offspring at causal loci l∈L and which directly affect the offspring’s trait value (via αld. These associations are due to covariances among alleles in each parental genome Dλl′ and D˜λl′) and across the parental genomes (reflected as trans-LD in the offspring, D˜λl). The indirect genetic confound reflects associations of the untransmitted alleles to alleles at other loci in the parents, which can indirectly affect the offspring trait value (via αli). Finally, unlike in family-based estimates of direct genetic effects (Eqs. 6 and 7), family-based estimates of indirect effects suffer from environmental confounding, in the same way that population GWASs do (Eq. 3).
Therefore, estimating the indirect effect by regressing the trait value on the untransmitted genotype is highly susceptible to environmental confounding as well as both direct and indirect genetic confounding, in a similar way to estimating the direct effect via a population-based association study (Shen and Feldman 2020). Adjustments for assortative mating in particular have been included in some PGS-based analyses of indirect effects (e.g., Kong et al. 2018; Young et al. 2022). However, it is not clear how robust these adjustments are in the presence of multiple forms of confounding.
## Polygenic scores and their phenotypic associations
A current drawback to family-based GWASs is that sample sizes are often small, limiting power to estimate direct genetic effects. Because of this limitation, instead of estimating per-locus effect sizes in family designs, investigators often measure the within-family phenotypic association of a combined linear predictor, a polygenic score (PGS), constructed using effect size estimates across many loci from a population GWAS. In the sibling-based version of this study design, the difference in siblings’ population-based PGSs is regressed on their difference in phenotypes (e.g., Lee et al. 2018; Selzam et al. 2019). In parent-offspring designs, the population-based PGSs constructed separately for transmitted and untransmitted alleles are used as linear predictors of the offspring’s phenotype, and the difference in their slopes in this regression is estimated (e.g., Kong et al. 2018; Okbay et al. 2022).
When such PGS regressions are used within families for the same phenotype as the population GWAS, a non-zero slope of the PGS is usually interpreted as reflecting the fact that the PGS—despite having been calculated from a population GWAS and therefore subject to many potential confounds—nevertheless does capture the direct genetic effects of alleles. When the PGS for one phenotype is regressed within families on the value of another phenotype, non-zero slopes are often interpreted as evidence that direct genetic effects on the two phenotypes are causally related, for example through pleiotropic effects of the alleles involved.
Suppose that we have performed a population GWAS for trait 1, generating effect size estimates αˆλ at a set of genotyped loci λ∈Λ. To construct a PGS for trait 1, these effect size estimates are used as weights in a linear sum across an individual’s genotype: [9] PGS1=∑λ∈Λgλαˆλpop. In a sibling-based study (the results and intuition below will be the same for a parent-offspring study), the difference between siblings’ trait-1 PGSs, △PGS1, is regressed on the difference in their values for trait 2, ΔY2 (note that trait 2 could be the same as trait 1). If L is the set of loci that causally underlie variation in trait 2, and βl are the true effects of variants at these loci on trait 2, then the numerator of the slope in this regression can be written as [10] Cov(ΔPGS1,ΔY2)=2∑λ∈Λ∑l∈L(1−2cλl)(Dλl′−D˜λl′)α^λpopβl (see Appendix A2). Note that, while the population-based effect size estimates αˆλ depend on cis- and trans-LD, as detailed by Eq. [ 3], the patterns of LD may differ from those in the family study (the Dλl′−D˜λl′ term in Eq. 10) if the population- and family-based studies differ in relevant aspects of sample composition.
The intuition for Eq. [ 10] is similar to that for the single-locus effect size estimate in a sibling GWAS (Eq. 7). The numerator of the difference in slopes of transmitted and untransmitted PGSs in a parent-offspring design takes a similar form to Eq. [ 10].
In the absence of confounding and under some simplifying assumptions, the sibling PGS covariance measures the contribution of each locus included in the PGS to the additive genic covariance between traits 1 and 2 that is tagged by the genotyped variants included in the PGS (see Eq. A.23 in Appendix A2). Under these assumptions, the sibling PGS slope therefore does provide a measure of the underlying pleiotropy between the traits.
Interpretation of the sibling PGS slopes is more complicated in the presence of genetic confounding (see Eq. A.22 in Appendix A2), which is absorbed into the effect size estimates αˆλpop (Eq. 3) so that the PGS applies a potentially strange set of weights to the genotyped loci it includes. ( A related problem occurs when indirect genetic effects absorbed by the population-based PGS change the interpretation of within-family PGS slopes—see Trejo and Domingue [2018]; Fletcher et al. [ 2021].) A non-zero sibling PGS slope still establishes that the trait-1 PGS loci are in systematic signed intra-chromosomal LD with loci that causally affect trait 2. However, it no longer necessarily implies that traits 1 and 2 are causally related via pleiotropy, for two reasons. To understand these reasons, suppose that the causal loci for traits 1 and 2 are distinct, i.e., that there is in fact no pleiotropy. First, a SNP included in the trait-1 PGS could tag local variants that causally affect trait 1 but which are also, via sources of confounding such as cross-trait assortative mating, in systematic long-range LD with variants on the same chromosome that causally affect trait 2. Such SNPs will be predictive of sibling differences in trait 2, even though they locally tag only trait-1 causal variants. Second, LD between variants on the same or distinct chromosomes that are causal for trait 1 and trait 2 will cause some SNPs that locally tag trait-2 causal variants to be significantly associated with trait 1 in a population GWAS, and therefore to be included in the trait-1 PGS. These SNPs, since they tag trait-2 causal variants, will be predictive of sibling differences in trait 2.
In summary, in the presence of confounding, non-zero sibling PGS slopes cannot be viewed as de facto evidence for causal relationships between traits.
## Sources of genetic confounding in association studies
As we have seen, genetic confounding of association studies depends, in ways that vary across study designs, on levels of non-local cis- and trans-LD between the study locus and loci that influence the study trait. Below, we consider various processes that give rise to non-local cis- and trans-LD, and their likely impact on the different association study designs. We focus our attention on the potential for these sources of LD to confound measurement of several key metrics. First, the average deviation of the estimated effect size from its true value, Eαˆλ−αλ. This measure indicates if effect sizes are systematically overestimated or underestimated because of genetic confounding. Second, the average squared effect size estimate, weighted by heterozygosity: E2pλ1−pλαˆλ2. This quantity is related to important measures such as the genetic variance and SNP-based heritability (Bulik-Sullivan et al. 2015b b). It is also directly related to the variance of effect size estimates, and therefore captures the additional noise that genetic confounding creates in effect size estimation at a given locus. Finally, if GWASs have been performed on more than one trait, the covariance across loci of the effect size estimates for two traits may be of interest. This covariance is determined by the average heterozygosity-weighted product E2pλ1−pλαˆλβˆλ, where αˆλ and βˆλ are the effect size estimates at locus λ for traits 1 and 2.
In what follows, for simplicity, we ignore indirect effects and assume that there is no environmental confounding (i.e., no correlation between genotypes and the environmental effects ϵ). For each of the sources of genetic confounding that we consider, we calculate the three measures listed above both analytically and in whole-genome simulations carried out in SLiM 4.0 (Haller and Messer 2019). In our simulations, we use two recombination maps: (i) for illustrative purposes, a simple hypothetical map where the genome lies along a single chromosome of length 1 Morgan, and (ii) the human linkage map generated by Kong et al. [ 2010]. A more detailed description of the simulations can be found in the Methods, and code is available at github.com/cveller/confoundedGWAS.
## Assortative mating
Assortative mating is the tendency for mating pairs to be correlated for particular traits—either the same trait (same-trait assortative mating) or distinct traits (cross-trait assortative mating). For example, humans are known to exhibit same-trait assortative mating for height and cross-trait assortative mating for educational attainment and height (amongst many other examples, reviewed in Horwitz and Keller 2022; Border et al. 2022a). Assortative mating generates both cis- and trans-LD: *It* generates positive trans-LD among trait-increasing alleles because genetic correlations between mates translate to genetic correlations between maternally and paternally inherited genomes, and it generates positive cis-LD among trait-increasing alleles because, over generations, recombination converts trans-LD into cis-LD (Crow and Felsenstein 1968). ( In some cases, assortative mating can generate cis-LD by mechanisms additional to recombination—see Veller et al. 2020.) If there is a constant correlation among mates for their values of two traits, then a genetic equilibrium will eventually be achieved. In this equilibrium, for any pair of loci l and l′, the trans-LD D˜ll′ will be constant. Call this constant value Dll′*, and suppose that the recombination fraction between the loci is cll′. With D˜ll′, constant across generations, the balance of its conversion into cis-LD (at rate c cur per generation) and the destruction of cis-LD by recombination (at rate cll′ per generation) will result in an equilibrium level of cis-LD equal to the degree of trans-LD: Dll′=D˜ll′=Dll′* (e.g., Crow and Felsenstein 1968).
The value of Dll′* will, in general, depend in a complicated way on the strength of effects of l and l′ on the traits upon which assortative mating is based and on the linkage relations of these loci to one another and to other causal loci. However, while it is therefore difficult to calculate the individual equilibrium LD terms Dll′*, we can in some cases calculate weighted sums of these terms across locus pairs.
Let the set of loci that influence one or both traits be L, and let αl be the effect size of the focal variant at locus l on trait 1 and βl its effect on trait 2 (the analyses below also apply to same-trait assortative mating, setting αl=βl). Recall the notation glm,mat and glm,pat for a mother’s maternally and paternally inherited genotype at locus l, with glf,mat and glf,pat a father’s analogs. The mother’s breeding value for trait 1 is G1mn=∑l∈L glmαl=∑l∈L glm,mat+glm,patαl=∑l∈L glm,matαl+∑l∈L glm,patαl=G1m,mat+G1m,pat, and, similarly, her breeding value for trait 2 is G2m=∑l∈L glm,matβl+∑l∈L glm,patβl=G2m,mat+G2m,pat. The father’s breeding values for the two traits are G1f=∑l∈L glf,matαl+∑l∈L glf,patαl=G1f,mat+G1f,pat and G2f=∑l∈L glf,matβl+∑l∈L glf,patβl=G2f,mat+G2f,pat.
We assume that individual trait values equal the breeding values plus environmental disturbances that are uncorrelated with the breeding values: Y1mn=G1m+ϵ1m;Y2m=G2m+ϵ2m;Y1f=G1f+ϵ1f;Y2f=G2f+ϵ2f; where Varϵ1m=Varϵ1f=VE1,Varϵ2m=Varϵ2f=VE2, and Covϵim,Gim=Cov(ϵif,Gif)=0fori∈1,2.
## Constant-strength assortative mating.
If the strength of assortative mating (measured by the phenotypic correlation among mates ρ) is constant over time, and there are no other sources of genetic confounding such as population structure, then, for a given pair of loci l,l′∈L, the positive cis-LD Dll′ will initially be smaller than the positive trans-LD D˜ll′, but will gradually grow towards an equilibrium value equal to the trans-LD (Dll′*=D˜ll′*); in this equilibrium, assortative mating generates new cis-LD at the same rate as old cis-LD is destroyed by recombination (Crow and Felsenstein 1968, Appendix A3.1).
Therefore, in a population GWAS, effect size estimates will initially be biased upwards because of positive trans-LD, and the magnitude of the bias will grow over time as positive cis-LD too is generated from this trans-LD (Eq. 3; Fig. 2). In contrast, in a family-based GWAS, effect size estimates will initially be biased downwards because the positive trans-LD exceeds the positive cis-LD (Eqs. 6 and 7; Fig. 2). However, as the cis-LD grows over time towards the value of the trans-LD, the magnitude of the downward bias will shrink, and, in equilibrium, the family-based GWAS will not be confounded by assortative mating (Fig. 2).
Under certain simplifying assumptions, we can calculate the average bias that assortative mating induces in a population GWAS in equilibrium, in the absence of other sources of genetic confounding such as population structure (Appendix A3.1). In the case of same-trait assortative mating, effect size estimates are inflated by an average factor of approximately h2ρ/1−h2ρ, where ρ is the phenotypic correlation among mates and h2 is the trait heritability (for similar calculations, see Yengo et al. 2018; Border et al. 2022b). In the case of cross-trait assortative mating, if assortative mating is directional/asymmetric with respect to sex—i.e., the correlation ρ is between female trait 1 and male trait 2—then assortative mating generates spurious associations between trait 1 and alleles that affect trait 2 (and vice versa). If the loci underlying the two traits are distinct, then, in equilibrium, the spurious effect size estimate at non-causal loci is approximately h2ρ/2 times the effect at causal loci, assuming the traits to have the same heritabilities and genetic architectures (horizontal dahsed line in Fig. 2). If cross-trait assortative mating is bi-directional/symmetric with respect to sex, then, in equilibrium, the average spurious effect size estimate at non-causal loci is approximately h2ρ times the effect at causal loci. Upward biases in effect size estimates at causal loci are also expected under cross-trait assortative mating, but these are second-order relative to the biases at non-causal loci (Fig. S1).
The systematic over- and under-estimation of effect sizes that assortative mating induces in population and family-based GWASs, respectively, will also affect our second measure of interest, the heterozygosity-weighted average squared effect size estimate E2pλ1−pλαˆλ2 (and therefore also downstream quantities such as SNP heritabilities). In a population GWAS, the presence of trans-LD and the gradual creation of cis-LD under assortative mating will increase the biases in effect size estimates over time (Fig. 2), which will concomitantly increase the average value of αˆλ2 (Fig. 3; also see Border et al. 2022b). Moreover, cross-trait assortative mating will generate signals of genetic correlations among traits even in the absence of any pleiotropic effects of underlying variants (Border et al. 2022a). In a family-based GWAS, the temporary attenuation of effect size estimates owing to a transient excess of trans-LD over cis-LD under assortative mating will lead to a similar attenuation in the average squared effect size estimate (Fig. 3), although, like the bias in effect size estimates themselves, this attenuation is expected to be small in humans (Fig. 3B).
As shown by Border et al. ( 2022a,b), the effects of assortative mating on estimates of heritability and genetic correlations described above are not well controlled for by LD Score regression (Bulik-Sullivan et al. 2015a,b). The LD score of a variant proxies the amount of local causal variation the SNP tags, but because assortative mating generates long-range signed LD among causal variants, it causes local causal variants to be in long-range signed LD with other causal variants throughout the genome. Therefore, the slope of the LD score regression absorbs the effects of assortative mating, causing its estimates of heritability and of the degree of pleiotropy to be inflated.
## Historical assortative mating.
If, at some point in time, assortative mating for traits ceases and mating becomes random with respect to those traits, the positive trans-LD that was present under assortative mating will immediately disappear, leaving only the positive cis-LD that had built up; this cis-LD will then be gradually eroded by recombination. If equilibrium had been attained under assortative mating, the cis-LD would have grown to match the per-generation trans-LD. Therefore, in the first generation after assortative mating ceases, the upward bias in population GWAS effect size estimates would halve as the trans-LD disappears (Eq. 3); the bias would then shrink gradually to zero as the cis-LD erodes (Fig. 2). A similar pattern will be observed for the heterozygosity-weighted average value of αˆλ2 in the population GWAS, which eventually returns to its equilibrium level under random mating (Fig. 3).
In contrast, with the disappearance of the positive trans-LD but the persistence of positive cis-LD, the bias in family-based effect size estimates will suddenly become positive once assortative mating ceases (having temporarily been negative under assortative mating before equilibrium was attained); this bias too will then gradually shrink to zero as recombination erodes the remaining cis-LD (Fig. 2). Concomitantly, the average squared effect size estimate in the family GWAS will suddenly increase when assortative mating ceases, after which it too will gradually return to its equilibrium value under random mating (Fig. 3).
## Assortative mating between traits with different genetic architectures.
An important practical question is how genetic confounding affects the GWAS loci we prioritize for functional follow-up and for use in the construction of polygenic scores. SNPs are usually prioritized on the basis of their GWAS p-value, which is proportional to the estimated variance explained by a SNP, 2pλ1−pλαˆλ2 (where pλ is the minor allele frequency). The results above assume, in the case of cross-trait assortative mating, that the traits involved have similar genetic architectures (distribution of pl and αl at causal loci, and the total number of causal loci). In that case, if there is no pleiotropy between the traits, then while SNPs that tag trait-1 causal loci are predictive of the value of trait 2 owing to LD between trait-1 and trait-2 causal loci, we nonetheless expect the SNPs that tag trait-2 causal loci to be better predictors of trait 2, such that GWAS investigators would primarily pick out SNPs tagging trait-2 causal loci for prioritization and use in polygenic scores.
However, analysis of human GWASs suggests that quantitative traits can have widely different genetic architectures, with, in particular, substantial differences in the effective numbers of causal loci involved and in the distribution of minor allele frequencies (Simons et al. 2022, and references therein). If two traits with distinct genetic bases show cross-trait assortative mating, but trait 1 has a denser genetic architecture (fewer causal loci) than trait 2, then the genetic signal of assortative mating—systematic LD between trait-1 and trait-2 causal loci—will be more heavily loaded per-locus onto trait-1 loci than onto trait-2 loci. In a GWAS on trait 2, this will inflate the magnitude of spurious effect size estimates at SNPs that tag trait-1 loci relative to effect size estimates at SNPs that tag causal trait-2 loci. In Appendix A3.1, we quantify this effect, showing that, in a population GWAS for trait 2, the average magnitude of spurious effect size estimates at trait-1 loci is proportional to L2/L1, where L1 and L2 are the numbers of loci underlying variation in traits 1 and 2 respectively. Thus, when trait 1 has a denser genetic architecture than trait 2 (| L2/|L1 is large), the magnitudes of effect size estimates at non-causal trait-1 loci could substantially overlap with those at causal trait-2 loci (as illustrated in Fig. 4), potentially causing part of the apparent, mappable genetic architecture of the trait-2 GWAS to actually tag trait-1 loci.
## Population structure
When a population GWAS draws samples from individuals of dissimilar ancestries, differences in the distribution of causal genotypes, and potentially of environmental exposures, can confound the association study (Lander and Schork 1994; Vilhjálmsson and Nordborg 2013). Correcting for confounds due to population structure has therefore been an important pursuit in the GWAS literature (Spielman et al. 1993; Pritchard et al. 2000; Price et al. 2010).
For concreteness, consider a simple model where two populations diverged recently, with no subsequent gene flow between them. Genetic drift—and possibly selection—in the two populations will have led to allele frequency differences between them at individual loci. If allele frequencies have diverged at both a genotyped study locus and at loci that cansally influence the study trait, these frequency differences will manifest as linkage disequilibria between the study locus and the causal loci in a sample taken across both populations, even if the loci are not in LD within either population. Specifically, if the frequencies of the focal allele at a given locus k are pk[1] and pk[2] in populations 1 and 2, then the cis-LD between the focal alleles at the association study locus λ and a causal locus l is [11] Dλl(S)=14pλ[1]−pλ[2]pl[1]−pl[2] in a sample that weights the two populations equally, with the superscript (S) denoting that this LD is due to stratification. The trans-LD takes exactly the same form: D˜λl(S)=Dλl(S). From Eq. [ 3], locus l therefore confounds estimation of the direct effect at λ in a population GWAS, by an amount proportional to [12] 2(Dλl(S)+D˜λl(S))αld=(pλ[1]−pλ[2])(pl[1]−pl[2])αld. *These* genetic confounds are in addition to environmental confounding that would arise if the environments of the two populations alter their average trait values by different amounts.
In contrast, estimates of direct effects obtained from within-family association studies are not genetically confounded, because cis- and trans-LD are equal (Eqs. 6 and 7). Another way of seeing this is to consider that, by controlling for family, within-family GWASs control for the population, and in the scenario considered, by construction, there are no within-population LDs to confound effect size estimation.
In the model we have considered, with results displayed in Fig. 5, there are initially two isolated populations of equal size. The frequency of the focal variant at locus l is pl[1] in population 1 and pl[2] in population 2, so that its overall frequency is pl=pl[1]+pl[2]/2. A population GWAS at locus λ returns an effect size estimate αˆλpop=2Vλ∑l∈L Dλl+D˜λlαl, where Dλl and D˜λl are calculated across both populations and are generally nonzero because of allele frequency differences between the two populations at loci λ and l (Nei and Li 1973). In our case, Vλ=2pλ1−pλ1+Fλ, and Dλl=D˜λl=14pλ[1]−pλ[2]pl[1]−pl[2], so αˆλpop=pλ[1]−pλ[2]2pλ1−pλ1+Fλ∑l∈L pl[1]−pl[2]αl. Squaring this and multiplying by 2pλ1−pλ, (A.62) 2pλ(1−pλ)(αˆλpop)2=(pλ[1]−pλ[2])22pλ(1−pλ)(1+Fλ)2[∑l∈L(pl[1]−pl[2])2αl2+∑l≠l′(pl[1]−pl[2])(pl′[1]−pl′[2])αlαl′].
## Allele frequency divergence due to drift.
How do the confounds introduced by population structure affect the first of our measures of interest, the average deviation of effect size estimates from their true values? The answer depends on the source of allele frequency differences between the two populations. If the differences are due to neutral genetic drift, they will be independent of each other (assuming causal loci are sufficiently widely spaced) and independent of the direction and size of effects at individual loci. Therefore, the LD induced by these allele frequency differences will, on average, not bias effect size estimates in a population GWAS: [13] Epλ[1]−pλ[2]pl[1]−pl[2]αld=Epλ[1]−pλ[2]Epl[1]−pl[2]Eαld=0, since Epk[1]−pk[2]=0 at any locus k.
However, the LD induced by population structure will inflate the average squared effect size estimate, and by extension the variance of effect size estimates (Fig. 5). In Appendix A3.2, we quantify this effect for the same simple case of two separate populations. We find that the average squared effect size estimate in a population GWAS is an increasing function of the divergence between the two populations (as measured by FST, the number of loci contributing variation to the study trait, and the true average squared effect size per locus (see also Rosenbery and Nordborg 2006; Lee and Lee 2023a).
In contrast, because effect size estimates from within-family GWASs are not confounded in this model of isolated populations, the average squared effect size estimate will not differ substantially from its expectation in an unstructured population (Fig. 5).
While we have focused on a simple model of two isolated populations, the result that within-family association studies are not confounded holds for other kinds of population structure as well. Specifically, we may be concerned that a population GWAS suffers from genetic confounding along some given axis of population stratification. However, the family-based estimates will be unbiased by confounding along such an axis if the maternal and paternal genotypes at each locus are exchangeable with respect to each other along this axis (Appendix A3.2). This requirement will be met in expectation under many models of local genetic drift in discrete populations or along geographic gradients. However, as we will shortly argue, migration and admixture introduce further complications.
How do these patterns of LD affect a population GWAS? If allele frequency differences between populations arose from neutral drift, they will be independent of effect sizes at causal loci and across loci, and therefore will not contribute, on average, a systematic directional bias to effect size estimates. However, they will inflate the average squared effect size estimate, by a smaller amount than for a population GWAS performed when the populations were still separated (because of the elimination of trans-LD under random mating in the admixed population). Moreover, this amount will decline in the generations after admixture as the remaining cis-LD is eroded by recombination (Eq. 16; Fig. 5). We quantify these effects in Appendix A3.3 (see also Pfaff et al. 2001; Rosenberg and Nordborg 2006; Zaitlen et al. 2014; Lee and Lee 2023b).
Although within-family GWASs were not genetically confounded when the populations were separate (because cis- and trans-LDs were equal, as discussed above), they become genetically confounded in the admixed population, as all trans-LD is eliminated by random mating in the admixed population, leaving an excess of cis-LD relative to trans-LD that biases effect size estimates (Eqs. 6 and 7). As in the case of the population GWAS, these biases will be zero on average if allele frequency differences between the ancestral populations were due to drift. However, after admixture, they will still inflate the average squared effect size estimate (and thus the variance of effect size estimates), which will thereafter decline in subsequent generations as the cis-LD is gradually broken down by recombination (Eq. 16; Fig. 5).
In comparing the average squared effect size estimate in a population and a family-based GWAS, we observe that the value in the population GWAS rapidly declines to approximately the same level as the value in the within-family GWAS, despite the former having started at a much higher level in the initial admixed population (Fig. 5). The explanation is that LD between unlinked loci confounds effect size estimation in the population GWAS but not the within-family GWAS, such that (i) the average squared effect size estimate from the population GWAS is initially much higher than that from a within-family GWAS, because it is inflated by LD between many more pairs of loci, and (ii) the average squared effect size estimate from the population GWAS declines more rapidly, because LD between unlinked loci is broken down more rapidly than LD between linked loci.
## Allele frequency divergence due to selection or phenotype-biased migration.
Selection and phenotype-biased migration can also generate allele frequency differences among populations (for a review of phenotype-biased migration, see Edelaar and Bolnick 2012). *Unlike* genetic drift, both of these forces can lead to systematic directional associations between effect sizes and changes in allele frequencies between populations. For example, if selection has favored alleles that increase the trait in population 1 but not in population 2, then [14] Epl[1]−pl[2]αld>0. as directional selection causes systematic changes in allele frequencies across the loci l underlying variation in the trait under selection (e.g., Hayward and Sella 2022). Importantly, this form of selection can occur even if the mean phenotype of the two populations does not change (Harpak and Przeworski 2021; Yair and Coop 2022). Similarly, phenotype-biased migration, where, say, individuals with a higher value of the phenotype tend to migrate from population 2 to population 1, can also create a positive association between effect sizes and allele frequency differences (Eq. 14).
Unlike the case of neutral genetic drift in the two populations, where the sign of the LD between two alleles is independent of their effect sizes, the effect-size-correlated associations driven by selection or phenotype-biased migration can add up across loci, and thus lead to substantial, systematic biases in estimates of allelic effect sizes. This systematic genetic confounding would also substantially inflate the average squared effect size estimate and thus measures of the genetic variance tagged by SNPs.
In addition, these systematic sources of genetic confounding can generate genetic correlations between traits with no overlap in their sets of casual loci—i.e., with no pleiotropic relationship. This will occur if two traits have both experienced selection or biased migration along the same axis. To take a concrete example, if people tend to migrate to cities in part based on traits 1 and 2, then these traits will become genetically correlated. If this axis is explicitly included as a covariate in the GWAS, then its influence on estimates of heritability and genetic correlations will be removed. However, its influence will not be removed by inclusion of genetic principal components or the relatedness matrix, if this axis (here, city vs. non-city) is not a major determinant of genome-wide relatedness at non-causal loci (Vilhjálmsson and Nordborg 2013). Nor will LD score regression control for this influence, as the selection- or migration-driven differentiation of a variant along the axis will be correlated with the extent to which it tags long-range causal variants involved in either trait. This effect on LD score regression is similar to that discussed above for assortative mating (Border et al. 2022a,b). Thus, like assortative mating, selection and phenotype-biased migration along unaccounted-for axes of population stratification can generate genetic correlations between traits. These selection- and migration-driven correlations should not necessarily be viewed as spurious, since genetic correlations should include those that arise from systematic long-range LD, but they complicate the interpretation of population-level genetic correlations as evidence for pleiotropy.
Again, these issues largely vanish in family-based studies, although phenotype-biased migration can cause transient differences in cis- and trans-LD that lead to biases in family-based estimates of direct effects (Eqs. 6 and 7).
In addition to drift, and as discussed above, selection and phenotype-biased migration can generate systematic, signed (effect-size correlated) LD, which would lead to systematic cis-LD in the descendent admixed population. These would lead to larger inflations of genetic variance and genetic correlations than would be expected had allele frequency divergence between the ancestral populations been due to drift alone, and would complicate interpretations of genetic correlations as being due to pleiotropy. Moreoever, if the admixed population is more than a few generations old such that LD between unlinked loci but not linked loci has largely been broken down, then population- and family-based estimates of these quantities might be similar.
## Admixture
When populations that have previously been separated come into contact, alleles from the same ancestral population remain associated with each other in the admixed population until they are dissociated by recombination. If allele frequencies had diverged between the ancestral populations, this ‘ancestry disequilibrium’ can translate to cis-LD between loci affecting a trait (Nei and Li 1973), potentially confounding GWASs performed in the admixed population. *More* generally, long range LD will be an issue when there is genetic stratification and ongoing migration between somewhat genetically distinct groups.
For concreteness, we again consider a simple model where two populations have been separated for some time, allowing allele frequencies to diverge between them. The populations then come into contact and admix in the proportions A and 1−A. We assume that mating is random with respect to ancestry in the admixed population.
Suppose that, just before admixture, the frequencies of the focal allele at a given locus k were pk[1] and pk[2] in the two populations. Then the initial degree of cis-LD between loci λ and l in the admixed population is given by Eq. [ 11], weighted by the proportions in which the populations admix: [15] Dλl,0(A)=A1−Apλ1−pλ2pl1−pl2; see, e.g., Pfaff et al. [ 2001]. This cis-LD subsequently decays at a rate cλl per generation, so that, t generations after admixture, [16] Dλl,t(A)=Dλl,0(A)1−cλlt=A1−Apλ1−pλ2pl1−pl21−cλt. Because we assume that mating is random in the admixed population, the trans-LD is zero in every generation after admixture: D˜λl,t(A)=0. Note that the decay of cis-LD in an admixed population will be slowed if individuals mate assortatively by ancestry, because the trans-LD generated by assortative mating is contimuslly converted by recombination to new cis-LD (as in our assortative mating model above; see Zaitlen et al. [ 2017] for more discussion of this point in the context of population admixture).
Suppose that two previously isolated populations admix in proportions A and 1−A, with subsequent random mating in the admixed population. Following the notation in the Section A3.2 above, before admixture, the frequency of the focal variant at locus l was pl[1] in population 1 and pl[2] in population 2, so that its overall frequency in the admixed population is pl=Apl[1]+(1−A)pl[2].
When the two populations admix, trans-LD between all pairs of loci disappears in expectation, owing to random mating in the admixed population: D˜λlt=0 for any pairs of loci λ and l and for any number of generations t after admixture. However, cis-associations between alleles that were more prevalent in one ancestral population than in the other will be retained as cis-LD in the admixed population until these associations are eroded by recombination. The initial degree of cis-LD between loci λ and l in the admixed population is Dλl0=A1−Apλ1−pλ2pl1−pl2. When t generations have elapsed since admixture, this cis-LD will have been eroded by recombination to Dλlt=Dλl01−cλlt=A(1−A)pλ[1]−pλ[2]pl[1]−pl[2]1−cλlt, where cλl is the sex-averaged recombination rate between λ and l. Therefore, t generations after admixture, a population association study at λ returns an effect size estimate αˆλpop,$t = 2$Vλ∑l∈L Dλltαl=A(1−A)pλ[1]−pλ[2]pλ1−pλ∑l∈L pl[1]−pl[2]1−cλltαl, while a sibling-based association study at λ returns αˆλsib,$t = 2$Hλ∑l∈L 1−2cλlDλltαl=A(1−A)pλ[1]−pλ[2]pλ1−pλ∑l∈L pl[1]−pl[2]1−cλlt1−2cλlαl, where we have substituted Vλ=Hλ=2pλ1−pλ owing to random mating in the admixed population.
Squaring the population estimate and multiplying by 2pλ1−pλ, (A.68) 2pλ(1−pλ)(α^λpop,t)2=2A2(1−A)2(pλ[1]−pλ[2])2pλ(1−pλ)[∑l∈L(pl[1]−pl[2])2(1−cλl)2tαl2+∑l≠l′(pl[1]−pl[2])(pl′[1]−pl′[2])(1−cλl)t(1−cλl′)tαlαl′], while the heterozygosity-weighted squared sibling effect size is (A.69) 2pλ(1−pλ)(α^λsib,t)2=2A2(1−A)2(pλ[1]−pλ[2])2pλ(1−pλ)[∑l∈L(pl[1]−pl[2])2(1−cλl)2t(1−2cλl)2αl2+∑l≠l′(pl[1]−pl[2])(pl′[1]−pl′[2])(1−cλl)t(1−cλl′)t(1−2cλl)(1−2cλl′)αlαl′].
## Spurious genetic correlations due to confounding in population-based PGSs.
Factors other than selection and phenotype-biased migration can also generate non-pleiotropic genetic correlation signals in family-based studies of admixed populations. In fact, the use of confounded population GWAS effect sizes can be sufficient. As an example of the confounding of genetic correlations in admixed populations due to a confounded GWAS for one trait, consider the GIANT-GWAS height polygenic score. Owing to confounding within Europe (Berg et al. 2019; Sohail et al. 2019), the height PGS showed large differences between Northern Europeans and sets of individuals sampled in other locations, such as the African 1000 genomes samples (Martin et al. 2017). This confounding generated a spurious, systematic correlation between height effect sizes and allele frequency differences across populations, with height-increasing alleles that are more common among Northern Europeans being assigned larger effects (Berg et al. 2019). As a result, in a PGS constructed from these effect size estimates, larger PGS values are predictive of greater North European ancestry. Now imagine a sibling-based study performed in a sample with recently admixed ‘European’ and ‘non-European’ ancestry—African Americans, for example. An individual with a larger value than their sibling for the GIANT height PGS will, on average, carry more ‘European’ ancestry. In African Americans, there will also be a systematic association of lighter skin pigmentation with recent ‘European’ ancestry, and selection on skin pigmentation will have driven a signed difference in allele frequencies between European and West African ancestors. Putting these observations together, the GIANT height PGS, being predictive of the degree of European ancestry, may well be predictive of skin pigmentation differences between African American sibling pairs (Eq. 10), leading to the naive and incorrect conclusion that height and skin colour are causally linked. In reality, this result would reflect the fact that alleles predicted to increase height and alleles that affect skin color are in systematic effect-signed admixture LD, as in Eq. [ 15], as a consequence of stratification-biased effect size estimates from the GIANT European GWAS.
## Stabilizing selection
Stabilizing selection—selection against deviations from an optimal phenotypic value—is thought to be common (Sella and Barton 2019), and has recently been argued to be consistent with the genetic architectures of many human traits (Simons et al. 2022). By disfavoring individuals with too many or too few trait-increasing alleles, stabilizing selection generates negative cis-LD among alleles with the same directional effect on the trait (Bulmer 1971). Thus, stabilizing selection will attenuate GWAS effect size estimates at genotyped loci that tag these causal loci.
To quantify these biases, we consider the model of Bulmer [1971, 1974], in which a large number of loci contribute to variation in a trait under stabilizing selection, with the population having adapted such that the mean trait value is equal to the optimum. Under this model, stabilizing selection rapidly reduces variance in the trait by generating negative cis-LD among trait-increasing alleles. If we make the simplifying assumption that all loci have equal effect sizes, then the equilibrium reduction in trait variance, −d* (where d*<0), can be calculated as a function of the genic variance Vg, the environmental noise VE, the strength of stabilizing selection VS/VP (scaled according to the phenotypic variance VP), and the harmonic mean recombination rate, c‾h, among loci underlying variation in the trait (Bulmer 1974; Appendix A.3.4).
Under these same assumptions, we calculate in Appendix A3.4 the average per-locus attenuation bias in effect size estimates induced by stabilizing selection, αl−αˆl/αl. In a population GWAS, this attenuation bias is approximately αl−αˆlpopαl=−d*Vg. In a within-family GWAS, the average proportionate bias is approximately αl−αˆlfarmαl=−d*1−2c‾hVg; i.e., smaller than in a population GWAS by a factor of 1−2c¯h Thus, the bias in effect size estimation can be calculated given estimates of the phenotypic variance and heritability of the trait, the harmonic mean recombination rate, and the strength of stabilizing selection (Appendix A3.4). In the Methods, making some simplifying assumptions about the genetic architecture of the trait in question, we calculate an approximate value c‾h≈0.464 for humsns. Using this value, Fig. 6 shows the average proportionate reduction in GWAS effect size estimates for various strengths of stabilizing selection and heritabilities of the trait. The range of selection strengths was chosen to match that inferred for human traits by Sanjak et al. [ 2018].
Attenuation of effect size estimates is larger if stabilizing selection is stronger or if the trait is more heritable. Taking height as an example, heritability is ~0.8, VP≈7cm2, and Sanjak et al. [ 2018] estimate a sex-averaged strength of stabilizing selection of VS/VP≈30. From these values, we calculate that a population GWAS would systematically underestimate effect sizes at loci that causally influence height by about $3\%$ on average, in the absence of other sources of LD (Fig. 6A). *More* generally, within the range of reasonable strengths of stabilizing selection inferred by Sanjak et al. [ 2018], we calculate average attenutations of population-based effect size estimates of up to $5\%$ for highly heritable traits (h2≈1) under strong stabilizing selection VS/VP≈20, down to $0.25\%$ for less heritable traits h2≈0.4 under weak stabilizing selection VS/VP≈170 (Fig. 6A).
Given the estimate c‾h~0.464, the proportionate bias that stabilizing selection induces in within-family GWASs is expected to be a fraction 1−2c‾h≈$7\%$ that in population-based GWASs. Thus, for height, a within-family GWAS would underestimate effect sizes by only about $0.2\%$ on average (Fig. 6B).
The quantitative importance of these biases will vary by application. In situations where the goal is gene discovery, for example, $5\%$ reductions in effect size estimates are unlikely to flip the statistical significance of variants with large effects on a trait. However, the attenuations in effect size estimates caused by stabilizing selection are systematic across loci, and therefore could substantially affect aggregate quantities based on these estimates. For example, the range of average reductions in population effect size estimates calculated above for human traits would translate to reductions in naive estimates of SNP-based heritabilities of between $0.5\%$ and $10\%$ (~$6\%$ in the case of height). If effect sizes are estimated by within-family GWAS, on the other hand, the reductions in these SNP-based heritability estimates would be much smaller.
As a further example, by generating negative LD between alleles with the same directional effect on the trait, the impact of stabilizing selection opposes, and therefore masks, the genetic impact of assortative mating (Brown et al. 2016). A practical consequence is that stabilizing selection will tend to attenuate estimates of the strength of assortative mating based on GWAS effect sizes, which often use cross-chromosome correlations of polygenic scores (e.g., Yengo et al. 2018; Yamamoto et al. 2023). In humans, the phenotypic correlation among mates for height has been measured at about ~0.25 (Stulp et al. 2017). In Appendix A3.4, we calculate that estimates of this correlation based on cross-chromosome correlations in PGSs will be biased downwards by about $20\%$, to ~0.20, because of stabilizing selection on height. Were assortative mating weaker, or stabilizing selection stronger, the genetic impact of assortative mating would be masked to an even greater extent (Appendix A3.4).
As in our analysis of assortative mating above, if stabilizing selection ceases in some generation, the negative LD that built up during the period of stabilizing selection will decay over subsequent generations, rapidly for pairs of loci on different chromosomes and more slowly for linked pairs of loci. Patterns of selection on human traits have changed over time—for example, the strength of stabilizing selection on birth weight has relaxed (Ulizzi and Terrenato 1987). *In* general, therefore, patterns of confounding reflect a composite of contemporary and historic processes.
We consider the model of Bulmer [1971, 1974], in which a very large number of loci contribute variation to a trait under stabilizing selection. We assume that the distribution of trait values is centered on the optimal value Y*, and that the relative fitness of an individual with trait value Y is exp(−(Y−Y*)$\frac{2}{2}$VS), where VS, the width or ‘variance’ of this gaussian selection function, governs the strength of stabilizing selection, with larger VS values implying weaker selection. Under this model, selection acts to reduce the phenotypic variation each generation; if the trait value is normally distributed with variance VP, then selection reduces the within-generation phenotypic variance by an amount (A.70) ΔVP=−Vp2VS+Vp. How much of this reduction carries over to the offspring generation then depends on the heritability of the trait.
Owing to the large number of loci in this model, the buildup of LD among them occurs on a faster timescale than the change in allele frequencies at individual loci. Assuming the loci to have equal effect sizes, Bulmer [1974] showed that the overall reduction in the phenotypic variance due to stabilizing selection, d, rapidly approaches a quasi-equilibrium value that approximately satisifes (A.71) d*=12h*4ΔVp*/c‾h, where h*2 is the heritability of the trait in this equilibrium and c‾h is the harmonic mean of the recombination rates amongst all pairs of loci. On this rapid timescale, the reduction in variance is due to LD among the loci underlying the trait; in fact, (A.72) $d = 2$α2∑l∈L∑l′∈LDll′, where α is the common per-locus effect size and Dll′ is defined with respect to the trait-increasing alleles at l and l′. The individual linkage disequilibria Dll′, in expectation, are proportional to the inverse recombination rates 1/cll′. Writing (A.73) 2α2∑l∈L∑l′∈LDll′*=d*=12h*4ΔVP*/c¯$h = 12$h*4ΔVP*∑l∑l′≠l1/cll′(|L|2), where |L|2=|L|(|L|−1)/2 is the number of pairs of distinct loci in L, it is apparent that (A.74) E[Dll′*]=14α2h*4ΔVp1/cll′L2.
Henceforth we deal only with equilibriuj quantities and therefore drop the star superscript for neatness. The phenotypic variance Vp can be written Vp=VG+VE=Vg+d+VE, where VG is the additive genetic variance, *Vg is* the genic variance, and VE is the variance due to the environment. Eqs. ( A.70) and (A.71), together with the definition of heritability h2=VG/VP, define a quadratic equation in d: (A.75) (1+2H)d2+2VS+Vg+VEc‾h+Vgd+Vg2=0. Eq. ( A.75) matches Eq. [ 10] in Bulmer [1974], with Bulmer’s parameter c replaced by $\frac{1}{2}$VS. For ease of reference in what follows, we write Eq. ( A.75) in the standard form ad2+bd+$c = 0.$ The roots are (A.76) d+,−=−b±b2−4ac2a=−VS+Vg+VEc‾h+Vg±VS+Vg+VEc‾h+Vg2−(1+2H)Vg21+2c‾h. To see which of these roots is the relevant one, we first note that the roots are both real, since the requirement for this is VS+Vg+VEc‾h+Vg2≥1+2c‾hVg2⇔VS+Vg+VEc‾h+Vg≥1+2c‾hVg⇔VS+VE≥1+2c‾h−1−c‾hc‾hVg, and 1+2c‾h<1+c‾h for c‾h>0, while VS+VE>0. Furthermore, since b>0 and 4ac>0, both roots are in fact negative, with d−<d+<0. Now note that 2d−<d++d−=−ba=−2VS+Vg+VEc‾h+Vg1+2c‾h<−2Vg+Vg+VEc‾h+Vg1+2ch(sinceVg<VS)<−2Vg+Vgc‾h+Vg1+2c‾h(sinceVE>0=−2Vg, i.e., Vg+d−<0. But then if the relevant root were d=d−,0≤VG=Vg+d−<0, a contradiction. So the relevant root is in fact (A.77) d=d+=−VS+Vg+VEc‾h+Vg+VS+Vg+VEc‾h+Vg2−1+2c‾hVg21+2c‾h, from which (A.78) −dVg=1−c‾h1+21+1c‾hX+X2−1+X1+2c‾h, where X=VS+VEVg. Since, in the absence of selection, VG=Vg, Eq. ( A.78) gives the proportionate reduction in the additive genetic variance due to selection.
From Eq. ( A.72), $d = 2$α2∑l ∑l′≠l Dll′, and, since Vg=∑l 2pl1−plα2=α2H‾|L|, with |L| the number of loci and H the average heterozygosity across them, we have (A.79) dVg=2∑l ∑l′≠l Dll′H‾L.
In a population association study performed at locus l, the effect size estimate is (A.80) α^lpop=αl+22pl(1−pl)∑l′≠lDll′αl′=α(1+22pl(1−pl)∑l′≠lDll′), so that the proportionate error is (A.81) 22pl(1−pl)∑l′≠lDll′. The mean proportionate error across loci is therefore (A.82) 1|L|∑l∈L(22pl(1−pl)∑l′≠lDll′)≈2∑l∑l′≠lDll′H¯|L|=dVg, from Eq. ( A.79), and assuming that the heterozygosities do not vary much across loci. That is, the average proportionate bias to effect size estimation that stabilizing selection induces is approximately equal to the proportionate reduction in the additive genetic variance, which is given in general form by Eq. ( A.78).
In a within-family association study performed at locus l, the effect size estimate is (A.83) α^lfam=αl+22pl(1−pl)∑l′≠l(1−2cll′)Dll′αl′=α(1+22pl(1−pl)∑l′≠l(1−2cll′)Dll′), so that the proportionate error is (A.84) 22pl(1−pl)∑l′≠l(1−2cll′)Dll′. The mean proportionate error across loci is therefore (A.85) 1|L|∑l∈L(22pl(1−pl)∑l′≠l(1−2cll′)Dll′)≈2∑l∑l′≠l(1−2cll′)Dll′H¯|L|≈2∑l∑l′≠l(1−2cll′)dc¯h2α2(L∣2)cll′H¯|L|=dc¯hα2H¯|L|(|L|2)∑l∑l′(1cll′−2)=dc¯hVg(|L|2)((|L|2)c¯h−2(|L|2))=dVg(1−2c¯h), where we have used Eq. ( A.74) in the second line. Therefore, the mean error in the within-family GWAS is smaller in magnitude than that in a population GWAS by a factor 1−2c‾h.
If ~1,000 loci underlie variation in the trait (and all contribute approximately the same variation), c‾h≈0.4640 in humans (see Methods), and so the average bias that stabilizing selection induces in within-family GWASs will be about 1−2c‾h≈$7\%$ that in population GWASs. If ~10,000 loci underlie variation in the trait, c‾h≈0.4346, and so the bias in within-family GWASs will be about $13\%$ that in population GWASs.
The calculations above give the average proportionate bias to GWAS estimates in terms of the basic parameters of the model, Vg,VE,VS, and c‾h. Often, however, not all of these parameters will be measurable. For example, human height appears to be under stabilizing selection (Sanjak et al. 2018), is highly heritable, and this heritability is believed to be underlain largely by direct genetic effects (Lee et al. 2018). However, it is difficult to directly measure the genic variance in height Vg because not all causal loci will be assayed in association studies—and, moreover, even if they were, effect size estimation at these causal loci would be biased by the genetic confounds that we have studied in this paper. However, the phenotypic variance in height Vp can obviously be measured, and the heritability of height h2 can also be measured using classical methods rather than effect size estimation in association studies. The strength of stabilizing selection on height can also be measured (Sanjak et al. 2018). From Vp and h2, the additive genetic variance VG can be estimated VG=h2Vp.
This example suggests that, in many applications, it might be useful to be able to estimate the equilibrium value of d using VG (or VP), VE,VS, and c‾h, even though VG (and VP), in the model we have considered, is a state variable influenced by the state variable of primary interest, d. This is straightforward: returing to our use of a star superscript to denote equilibrium values, if we treat VG and VP as their equilibrium values VG* and VP*, Eq. ( A.71) can be estimated directly, and also simplifies to (A.86) d*=−12c‾h⋅VG*2VS+VG*+VE=−12c‾h⋅VG*2VS+Vp*=12c‾h⋅h*4Vp*2VS+VP*. The proportionate bias in a population GWAS, given by Eq. ( A.82), can similarly be estimated from h2, Vp,VS, and c‾h, by first observing that Vg=VG*−d*=VG*+12c‾hVG*2VP*=VG*1+12c‾h⋅VG*VS+VP*, so that Eq. ( A.82) can be written (A.87) d*Vg=−12c¯h⋅VG*2VS+VP*VG*(1+12c¯h⋅VG*VS+VP*)=−12c¯h⋅VG*VS+VP*1+12c¯h⋅VG*VS+VP*=−12c¯h⋅h*2VP*VS+VP*1+12c¯h⋅h*2V*VS+VP*=−12c¯h(1+VS/VP*h*2)+1, which reveals that the proportionate bias depends only on ch,h*2 and the scaled inverse strength of selection, VS/Vp.
From Eq. ( A.85), the proportionate bias in a within-family GWAS is then approximately (A.88) d*Vg1−2c‾h=−1−2c‾h2c‾h1+VS/VP*h*+1.
## Sibling indirect effects
Indirect effects of siblings’ genotypes on each other’s phenotypes are known to be a potential source of bias in sibling-based GWASs (Fletcher et al. 2021; Young et al. 2022), and can be measured and corrected only if, in addition to sibling genotypes, parental genotypes are also available (Kong et al. 2018; Young et al. 2022). *To* generate intuition for their impact on GWASs, we consider a simple model of indirect sibling effects in the absence of G×E interactions and other confounding effects, focusing on a single-locus model for simplicity. We suppose that the indirect effect of an individual’s phenotype on their sibling’s phenotype is β, so that the phenotypes of two siblings i and j can be written [17] Yi=Y*+αgi+βYj+ϵi,Yj=Y*+αgj+βYi+ϵj. Taking their difference and rearranging, we find that [18] ΔY=α1+βΔg+11+βΔϵ. Therefore, in the absence of genetic confounding and G×E interactions, a sibling-based association study would return an effect size estimate of [19] αˆsib=α1+β on average. Thus, if sibling indirect effects are synergistic (β>0), they lead to underestimation of the direct genetic effect at the locus. In contrast, if sibling indirect effects are antagonistic (β<0), they lead to overestimation of the direct genetic effect.
How would a population GWAS be affected by the same sibling indirect effects? Sibling i’s phenotype can be written [20] Yi=Y*+αgi+βYj+ϵi,=Y*+αgi+β(Y*+αgj+βYi+ϵj)+ϵi⇒Yi=11−β2(Y**+αgi+αβgj+ϵi+βϵj), where Y**=(1+β)Y*. Therefore, if we were to randomly choose one sibling from each sibship and estimate the effect size at the locus using a population association study across families, we would obtain [21] αˆpop=Covgi,YiVargi=11−β2α+αβrgsibs, where rgsibs=Covgi,gj/Vargi is the genotypic correlation between sibs at the locus. Sibling indirect effects alter the effect size estimate in a population GWAS via two channels. The first is through the factor $\frac{1}{1}$−β2 in Eq. [ 21], which reflects second-order feedbacks of an individual’s phenotype on itself, via the sibling. Since $\frac{1}{1}$−β2>1, these feedbacks act to exacerbate the effects of causal alleles. For example, if sibling indirect effects are antagonistic (β<0), then a sibling with a large trait value will tend to indirectly reduce the trait value of their sibling, which in turn will indirectly further increase the trait value of the focal individual. This channel therefore pushes population GWASs towards overestimating the magnitude of direct genetic effects.
The other channel by which sibling indirect effects can influence a population GWAS is driven by the genotypic correlation among siblings, and is easiest to understand if we assume that sibling indirect effects are weak β2≪1. In this case, αˆPop≈α+αβrgsibs. Since the genotypic correlation rgsibs>0, this channel of sibling indirect effects has the opposite effect to the one it has on a sibling GWAS: if sibling indirect effects are synergistic (β>0), the population GWAS overestimates the direct genetic effect at the locus, while if sibling indirect effects are antagonistic (β<0), the population GWAS underestimates the direct genetic effect. The reason for this difference is that a sibling GWAS is based on siblings whose genotypes differ at the focal locus, and whose genotypic values are therefore anticorrelated. If sibling indirect effects are synergistic (β>0), they will tend to attenuate the phenotypic differences between such siblings, and therefore attenuate effect size estimates. In contrast, because siblings’ genotypes are positively correlated across the entire population, synergistic sibling indirect effects (β>0) will tend to exacerbate phenotypic differences across families, leading a population GWAS to overestimate effect sizes.
## Gene-environment (GxE) and gene-gene (GxG) interactions
Up to this point, we have assumed that alleles’ direct effects do not vary across environments or genetic backgrounds. *To* generate intuition for the influence of G×E (and G×G) interactions on population and family-based GWAS designs, we restrict our focus to a single causal locus, assuming no genetic confounding and no indirect effects of siblings. To incorporate G×E interactions, we allow the effect size of the alleles at the locus to depend on the family environment. The phenotype of individual i in family f is [22] Yi=Y*+α+αf+αigi+ϵf+ϵi, where we arbitrarily define αf and αi such that their population means are zero: Eαf=Eαi=0. α is then the average causal effect of the allele were it randomized across individuals from different families in our sample. αf is the deviation of this effect in family f due to their environment, and αi is an individual deviation which we assume to be independent of i ‘s genotype; both αf and αi can be thought of as random slopes in a mixed model. Note that αf can reflect the interaction of alleles with the family’s external environment (as we have framed it here) or with the family’s genetic background (a G×G interaction).
If we perform a sibling GWAS by taking pairs of full siblings i and j in family f and regressing the difference in their phenotypes ΔYf=Yi−Yj on the difference in their genotypes at the focal locus Δgf=gi−gj, we obtain an effect size estimate [23] αˆsib=α+Eαf∣parentheterozygous, where the second term—the deviation of the family-based estimate αˆsib from α—is the average family deviation conditional on a parent being heterozygous at the focal locus (Appendix A4). The intuition is that, because only heterozygous parents contribute the genetic variation among siblings on which our effect size estimate is based, if these heterozygous parents are non-randomly distributed across environments, then the family-based GWAS samples values from a distribution of family effects αf that is different to the overall population distribution.
We can compare this estimate from a sibling GWAS to one from a population GWAS, again under the assumption of no genetic confounding or indirect effects from siblings: [24] αˆpop≈α+(1−2p)(1−2F)Eαf∣gi=1+2(p+(1−2p)F)Eαf∣gi=2 where p is the frequency of the focal variant and F is the inbreeding coefficient at the locus (Appendix A4). The approximation holds if F is small. Note that Eq. [ 24] conditions on the number of focal alleles carried by the sampled individual, whereas Eq. [ 23] conditions instead on the parental genotype.
Like the family-based estimate, the population-based effect size estimate is distorted when heterozygotes are not randomly distributed over family backgrounds Eαf∣gi=1≠0 as well as when homozygotes are not randomly distributed across family backgrounds (when Eαf∣gi=2≠0). Thus, effect size estimates from both family- and population-GWAS can differ from the genetic effects that would be estimated if genotypes were randomly distributed across interacting family backgrounds, and these distortions will in general not be the same aross population and family-based study designs.
As noted above, because of current sample size constraints in family-based studies, a common strategy is to calculate the association of population-based PGSs and phenotypic differences among family members. In the absence of confounding, it is clear from Eq. [ 10] that the influence of G×E interactions on the covariance of sibling differences in PGSs and trait values would depend on the average value of αˆsibαˆpop across loci. Thus the slope of the PGS in this regression could be affected if, on average, the alleles at casual loci tagged by genotyped variants in the PGS are more often found in environments that suppress (or enhance) their effects. G×E interactions across many loci have been suggested by some recent studies (Mostafavi et al. 2020; Zhu et al. 2022), but their quantitative impact on differences between population and family-based GWASs remains unknown.
An allele’s effect could also systematically differ across families αf if it is involved in epistatic interactions with alleles at other loci in the genome (G×G). By analogy to our G×E model above, epistatic interactions would lead to biases in family-based GWASs if parents who are heterozygous at the focal study locus tend to have systematically different genotypes at loci that interact epistatically with the focal locus, relative to the population distribution of such genetic backgrounds.
Up to this point, we have also ignored parent-offspring interactions as a possible source of bias in family-based studies. Following the the same logic above, interactions between parents’ and offsprings’ alleles will result in family GWAS estimates that are the average effect of the focal allele in an offspring conditional on the genetic background of a heterozygous parent. Thus, again, a non-random distribution of genetic backgrounds in heterozygous parents is a potential source of bias.
One way that heterozygous parents might exhibit a non-random distribution of genetic backgrounds is via trait-based assortative mating, which could therefore modify the way that epistasis and parent-offspring interactions influence effect size estimation in a family-based GWAS relative to a population GWAS and relative to the true average population effect.
A final, overarching complication is that the individuals participating in a population GWAS are not a random subset of the population(s) from which they are drawn (Fry et al. 2017; Pirastu et al. 2021; Tyrrell et al. 2021), and families enrolled in GWASs can be even less representative of the population as a whole (Mostafavi et al. 2020; Benonisdottir and Kong 2022). These participation biases can potentially lead to systematic differences between the distributions of genotypes and interacting environments experienced by the population, the GWAS sample, and participants in a family-based study.
## Discussion
It has long been recognized that population GWASs in humans can be biased by environmental and genetic confounding (Lander and Schork 1994; Vilhjálmsson and Nordborg 2013). Currently, population GWASs attempt to control for these confounds by focusing on sets of individuals that are genetically more similar and by controlling for population stratification. However, these controls are imperfect and are not always well defined. For example, controlling for genome-wide patterns of population stratification based on common alleles does not control for the genetic and environmental confounding of rare variants (Zaidi and Mathieson 2020). Work on genetic confounding has uncovered increasing evidence that assortative mating may be leading to large biases in estimates of direct genetic effects and to large genetic correlations for a number of traits (Yengo et al. 2018; Border et al. 2022b,a); moreover, it can often be unclear whether genetic signals of assortative mating are due to trait-based mate choice or some other more general form of genetic confounding (e.g., Haworth et al. 2019). Additionally, while we have focused primarily on genetic confounding, for a number of traits there are also signals of residual environmental confounding in GWAS signals (Selzam et al. 2019; Mostafavi et al. 2020; Okbay et al. 2022; Abdellaoui et al. 2022). Thus, subtle and often interwoven forms of genetic and environmental confounding remain a major issue in many GWASs (Young et al. 2022), compromising the interpretation of GWAS effect size estimates and downstream quantities such as SNP heritabilities and genetic correlations.
Effect size estimates from within-family GWASs are less affected by these various confounds. In the absence of G×E interactions, they are not subject to environmental confounding across families, because the environments of family members are effectively randomized with respect to within-family genetic transmission. As we have shown, family-based estimates should also suffer substantially less from genetic confounding, because genetic transmission at unlinked loci (but not linked loci) is randomized by independent assortment of chromosomes in meiosis. Nonetheless, family-based GWASs can suffer from residual genetic confounding as well as sibling indirect effects and G×E/G×G interactions; they also raise a number of conceptual problems that we discuss below.
## Sources of genetic confounding.
Genetic confounding is caused by long-range LD between loci that affect the trait or traits under study. To illustrate the potential for genetic confounds to bias GWAS effect size estimates, we have considered several sources of long-range LD. Some of these—assortative mating, selection on GWAS traits, and phenotype-biased migration—can cause systematic directional biases in GWAS effect size estimates. Others, such as neutral population structure, cause random biases that influence the variance of effect size estimates and related quantities. Assortative mating and neutral population structure have received considerable theoretical attention in the GWAS literature (e.g., Rosenberg and Nordborg 2006; Yengo et al. 2018; Border et al. 2022a,b). Here, we have further outlined how both selection and phenotyped-biased migration can drive systematic genetic confounding that may not be well accounted for by current methods of controlling for stratification.
We wish to emphasize stabilizing selection in particular as a potential source of systematic confounding in GWASs. Stabilizing selection has been well studied in the quantitative genetics literature but less so in the context of GWASs, despite its expected ubiquity. By selecting for compensating combinations of trait-increasing and trait-decreasing alleles, stabilizing selection generates negative LD between alleles with the same directional effect on the trait (Bulmer 1971, 1974), and can therefore bias GWAS effect size estimates downwards. While the potential for stabilizing selection to confound effect size estimation has been noted (e.g., Brown et al. 2016; Yair and Coop 2022; Li et al. 2023), the resulting biases have not, to our knowledge, been quantified. Our calculations suggest that these downward biases could, for some human traits, be as large as $5\%$ systematically across all causal loci in population GWASs. While biases of this magnitude are unlikely to compromise some goals of GWASs, such as gene discovery, they could be quantitatively problematic for other GWAS aims, such as estimation of SNP heritabilities and the strength of assortative mating. Moreover, while our results pertain to (a particular model of) stabilizing selection, many kinds of selection generate LD between genetically distant loci—in fact, only multiplicative selection among loci does not (Bürger 2000, pgs. 50 and 177). Therefore, the general result that selection can generate genetic confounding will hold more broadly.
For a given genotyped locus in a GWAS, there is no bright line between local ‘tagged’ LD and long-range confounding LD, and one reasonable objection to the approach taken here is that that we have used an arbitrary definition of the causal loci that are locally tagged by a genotyped locus (Llocal in Eq. 2). All of the sources of genetic confounding that we have considered generate LD among causal loci both within and across chromosomes. Under these models, the within-chromosome LD that is generated is, in a sense, a continuation of the LD generated across chromsomes (moving from a recombination rate = 0.5 to ≤ 0.5). Thus, while investigators may prefer some looser definition of ‘local’ when thinking about genotyped GWAS loci as tag SNPs, to extend that definition to include all loci on the same chromosome as the SNP would, by reasonable interpretation, be to include confounding into the desired estimator.
The extent to which the absorption of genetic confounding in estimated effect sizes is a problem depends on the application. In the case of polygenic prediction, absorbing environmental effects, indirect effects, the effects of untyped loci throughout the genome can help to improve prediction accuracy, although this does come at a cost to interpretability. For GWAS applications focused on understanding genetic causes and mechanisms, the biases in effect size estimates and spurious signals of pleiotropy among traits generated by genetic confounding will be more problematic.
## Indirect genetic effects.
Family GWASs are often interpreted as providing the opportunity to ask to what extent parental genotypes (or other family genotypes) causally affect a child’s phenotype (‘genetic nurture’; Kong et al. 2018). Viewed in this way, the association between untransmitted parental alleles and the child’s phenotype would seem, at first, a natural estimate of indirect genetic effects.
In practice, however, if the population GWAS suffers from genetic and environmental confounds, then the estimated effects of untransmitted alleles will absorb that confounding in much the same way that estimates of direct genetic effects from a population GWAS do (Eq. 8; Shen and Feldman 2020). For example, in the case of assortative mating, a given untransmitted allele is correlated with alleles that were transmitted both by this parent and by their mate, and these transmitted alleles can directly affect the offspring’s phenotype. Thus, while family-based estimates of direct genetic effects benefit from the randomization of meiosis and from controlling for the environment, family-based estimates of indirect genetic effects lack both of these features and should be interpreted with caution. Indeed, recent work using parental siblings to control for grandparental genotypes has shown that little of the estimated ‘indirect genetic effect’ may be causally situated in parents (Nivard et al. 2022). With empirical estimates of indirect genetic effects potentially absorbing a broad set of confounds (Demange et al. 2022; Young et al. 2022), and few current studies of indirect effects having designs that allow such confounding to be disentangled, it is premature—and potentially invalid—to interpret associations of untransmitted alleles causally in terms of indirect genetic effects (Wolf et al. 1998). Rather, they should be treated agnostically in terms of ‘non-direct’ effects.
## Direct genetic effects.
Mendelian segregation provides a natural randomization experiment within families (Fisher 1952), and so crosses in experimental organisms and family designs have long been an indispensable tool to geneticists in exploring genetic effects and causation. Growing concerns about GWAS confounding and the increasing availability of genotyped family members have led to a return of family-based studies to the association study toolkit (Young et al. 2019). Family-based estimates of direct genetic effects are often interpreted as being unbiased and discussed in terms of the counterfactual effect of experimentally substituting one allele for another (Morris et al. 2020; Brumpton et al. 2020; Young et al. 2022).
As we have shown, family-based GWASs are indeed less subject to confounding than population-based GWASs: in the presence of genetic and environmental confounding, the family-based estimate of the effect size at a given locus provides a much closer approximation to the true effects of tightly linked causal loci than a population-based estimate does. The family-based estimate is not biased by environmental variation across families and avoids the correlated effects of the many causal loci that lie on other chromosomes. Still, the family-based estimate does absorb the effects of non-local causal loci on the same chromosome, and so cannot truly be said to be free of genetic confounding. Rather than considering a single allele being substituted between individuals, a better experimental analogy for the effect size estimate would be to say that we are measuring the mean effect of transmission of a large chunk of chromosome surrounding the focal locus, potentially carrying many causal loci.
In addition, while within-family GWASs offer these advantages, in other ways, they move us further away from the questions about the sources and causes of variation among unrelated individuals that motivate population GWASs in the first place. Indeed, the presence of confounding introduces a number of conceptual issues in moving from within-family GWAS to the interpretation of differences among individuals from different families (Coop and Przeworski 2022a,b). For example, in the presence of genetic confounding, the effect of a causal allele of interest will depend on a set of weights: its LD to many other causal alleles. In estimating the direct effect of the allele, family-based approaches weight these LD terms differently to population-based approaches, which, we argue, can complicate the interpretation of these estimates. For example, when previously isolated populations admix, same-ancestry alleles will be held together in long genomic blocks until these are broken up by recombination, which will happen very quickly for alleles on different chromosomes but more slowly for alleles on the same chromosome. A few generations after admixture, therefore, cross-chromosome ancestry LD will largely have dissipated, but contiguous ancestry tracts will still span substantial portions of chromosome lengths. Since both population and within-family GWASs are similarly confounded by the same-chromosome LD, their mean squared effect sizes will be similar in this case (Fig. 5). Bearing in mind that the LD resulting from admixture is not present in the source populations, it becomes unclear which weighting of ancestry LD is appropriate if we want to interpret the resulting effect size estimates as direct effects. As this example illustrates, while family-based GWASs are a useful device for dealing with confounding, it is not always obvious how to interpret the quantities that they measure.
A number of additional complications arise when, to compensate for the small effect sizes of individual loci, researchers combine many SNPs into a polygenic score (PGS) and study the effects of PGSs within families (or use them as instruments in Mendelian randomization analyses). For one, SNPs are usually chosen for inclusion in the PGS on the basis of their statistical significance in a population GWAS. This approach prioritizes SNPs whose effect size estimates are amplified (or even wholly generated) by confounding (for an example of how this leads to residual environmental confounding in applications of sibling-based effect size estimates, see Zaidi and Mathieson 2020). Second, the weights given to SNPs that are included in the PGS absorb the effects of confounding, and this confounding is heterogeneous across SNPs. Thus, when we study the correlates of trait-A PGS differences between siblings in the presence of GWAS confounding, we are not observing the average phenotypic outcomes of varying the genetic component of trait A between siblings. Rather, we are varying a potentially strangely-weighted set of genetic correlates of trait A.
An observation that a population GWAS PGS is predictive of phenotypic differences among siblings demonstrates that the PGS SNPs tag nearby causal loci, but beyond that, interpretation is difficult. Notably, if there is cross-trait assortative mating for traits A and B, but no pleiotropic link between the traits, then some of the SNPs identified as significant in a GWAS on trait A may be tightly linked to loci that causally affect trait B but not trait A. If these loci are included in the trait-A PGS, then when we study the effect of variation in the trait-A PGS on sibling differences, we are accidentally absorbing some components of the variation in trait B across siblings. Thus, we might observe a correlation between the trait-A PGS and differences in trait B between siblings, and this correlation may be lower than is observed at the population level, without there existing any pleiotropic (or causal) link between A and B. These effects can be exacerbated if the two traits have different genetic architectures (Figure 4). Instead of using a set of SNPs and weights from a population GWAS, genetic correlations between traits due to pleiotropy could be estimated from the correlation of effect sizes estimated within families (Howe et al. 2022). Given current sample size constraints in family-based studies, the confidence intervals on these estimates are large. Moreover, significant family-based correlations need not reflect pure pleiotropy, since, as we have shown, they are not completely free of genetic confounding due to intra-chromosomal LD.
Also complicating the interpretation of family-based effect size estimates are various types of interactions. Indirect effects between siblings can bias family estimates of direct genetic effects (Eq. 19; Young et al. 2019; Fletcher et al. 2021; Young et al. 2022) in ways that are conceptually different from the biases they introduce to population-based estimates (Eq. 21). These sibling effects can potentially be addressed with fuller family information (e.g., parental genotypes in addition to sibling genotypes; Kong et al. 2018; Young et al. 2022).
As we have further shown here, G×E (and G×G) interactions can also complicate the interpretation of family-based effect size estimates. The reason is that, even if we were to know the causal alleles for a trait of interest, what we estimate by measuring their associations with phenotypic differences within families is not analogous to the counterfactual effects of experimentally substituting alleles in random individuals. Instead, we are necessarily restricting our focus to the effect of their transmission from heterozygous parents. If heterozygous parents tend to experience different environments or carry different genetic backgrounds than homozygotes do, within-family designs will tell us about direct effects in these particular environments or genetic backgrounds, rather than in the population as a whole. Thus, although the ongoing shift towards family-based studies is motivated by concerns about confounding, with different alleles experiencing different environmental and genetic backgrounds, family-based studies can be influenced by conceptually similar issues of confounding in the presence of G×E and G×G interactions. Such interactions are difficult to reliably identify and measure, but there are a growing number of potential examples from GWASs (Tropf et al. 2017; Barcellos et al. 2018; Young et al. 2018b; Mostafavi et al. 2020; Patel et al. 2022). The interaction issues raised here echo a set of conceptually distinct concerns about the interpretation of average treatment effects in other contexts (Słoczyński 2022), reinforcing the need for care in interpreting such estimates as informative about causes across heterogeneous groups.
In summary, family-based studies are a clear step forward towards quantifying genetic effects, with large-scale family studies carrying the potential to resolve long-standing issues in human genetics. However, these designs come with their own sets of caveats, which will be important to understand and acknowledge as family-based genetic studies become a key tool in the exploration of causal effects across disparate fields of study.
## Methods
All simulations were carried out in SLiM 4.0 (Haller and Messer 2019). Code is available at github.com/cveller/confoundedGWAS.
For the purpose of carrying out sibling association studies in our simulations, we assumed a simple, monogamous mating structure: each generation, each female and each male is involved in a single mating pair, and each mating pair produces exactly two offspring (who are therefore full siblings). To maintain the precisely even sex ratio required by this scheme, we assumed that a quarter of mating pairs produce two daughters, a quarter produce two sons, and half produce a son and a daughter. Population sizes were chosen to ensure that these numbers of mating pairs were whole numbers, and mating pairs were permuted randomly each generation before assigning brood sex ratios (to ensure that no artifact was introduced by SLiM’s indexing of individuals).
*Each* generation, per-locus effect size estimates were calculated for both population-wide and sibling GWASs. The former were calculated as the regression of trait values on per-locus genotypes, while the latter were calculated as the regression of sibling differences in trait values on sibling differences in per-locus genotypes.
In all simulations, the total population size was $$n = 40$$,000.
## Assortative mating.
For our general cross-trait assortative mating setup, traits 1 and 2 are influenced by variation at sets of bi-allelic loci L1 and L2 respectively. The effect sizes of the reference allele at locus l on traits 1 and 2 are αl and βl respectively. An individual’s polygenic score (PGS) is then P1=∑l∈L1 glαl for trait 1 and P2=∑l∈L2 glβl for trait 2. In all the scenarios we simulated, traits had heritability 1, so that individuals’ trait values are the same as their PGSs.
Our aim is to simulate a scenario where assortative mating is based on females’ values for trait 1 and males’ values for trait 2, such that, across mating pairs, the correlation of the mother’s PGS for trait 1, P1m, and the father’s PGS for trait 2, P2f, is a constant value ρ (in all of our simulations, ρ=0.2). To achieve this, we use an algorithm suggested by Zaitlen et al. [ 2017]: At the outset, we choose an accuracy tolerance ε such that, if by some assignment of mates the correlation of their PGSs falls within ε of the target value ρ, we accept that assigment. *Each* generation in which assortative mating occurs, we rank females in order of their PGSs for trait 1, and males in order of their PGSs for trait 2. We then calculate the PGS correlation across mating pairs, ρ0, if females and males were matched according to this ranking. If this (maximal) correlation is smaller than the upper bound of our target window ρ0<ρ+ε, which very seldom occurred in our simulations), then females and males mate precisely according to their PGS rankings and we move on to the next generation. If, instead, ρ0 exceeds ρ+ε, then we follow the following iterative procedure until we have found a mating structure under which the correlation of PGSs falls within ε of the target value ρ.
First, we choose initial ‘perturbation sizes’ ξ0 and ξ1=2ξ0. Suppose that, in iteration k of the procedure, the perturbation size is ξk and the chosen mating structure leads to a correlation among mates of ρk. If ρk−ρ<ε, we accept the mating structure and move on to the next generation. Otherwise, we choose a new perturbation size ξk+1: (i) if ρk−1,ρk>ρ, then ξk+1=2ξk; (ii) if ρk−1>ρ>ρk or ρk−1<ρ<ρk, then ξk+1=ξk−1+ξk/2; (iii) if ρk−1,ρk<ρ, then ξk+1=ξk/2. Once we have chosen ξk+1, for each individual we perturb their PGS (trait 1 for females; trait 2 for males) by a value chosen from a normal distribution with mean 0 and standard deviation ξk+1, independently across individuals. We then rank females and males according to their perturbed PGSs, and calculate the correlation ρk+1 of their true PGSs if they mate according to this ranking. ( Since, in our experience, there can be substantial variance in the ρk+1 values that result from this procedure, we repeat it 5 times and choose the mating structure that produces the value of ρk+1 closest to the target value ρ.) We then decide if another iteration—i.e., another perturbation size ξk+2—is required.
## The model
Under the general additive model we have studied, an individual’s value for trait Y is (A.1) Y=Y*+∑l∈Lαldgl+∑l∈Lαli,mglm+∑l∈Lαli,fglf+ϵ, where gl is the number of focal alleles at locus l carried by the individual, αld is the direct genetic effect on the trait value of the focal allele at l (which we assume to be positive, without loss of generality), glm and glf are the numbers of copies of the focal allele at locus l carried by the individual’s mother and father respectively, and αli,m and αli,f are the indirect genetic effects of the focal allele at l vis the mother’s and father’s genotype respectively. ϵ is the environmental disturbance, with mean zero, and Y* is the expected trait value of the offspring of parents who carry only trait-decreasing alleles.
It will be useful to expand Eq. ( A.1) in terms of the individual’s and the individual’s parents’ maternally and paternally inherited genotypes: (A.2) Y=Y*+∑l∈Lαld(glmat+glpat)+∑l∈Lαli,m(glm,mat+glm,pat)+∑l∈Lαli,f(glf,mat+glf,pat)+ϵ, where glmat is the number of focal alleles at locus l that the individual inherited maternally, glm,mat is the number of focal alleles at l that the individual’s mother inherited maternally, etc.
## Population GWAS
If we perform a standard population GWAS at a genotyped locus λ, the estimated effect of the focal allele at λ on the trait Y is (A.3) αˆλpop=Covgλ,YVargλ. Here, Vargλ is the genotypic variance at λ among sampled individuals, equal to 2pλ1−pλ1+Fλ, where pλ is the frequency of the focal allele at λ and Fλ is the coefficient of inbreeding at λ. For example, if λ is at Hardy-Weinberg equilibrium, then Vargλ=2pλ1−pλ; if, instead, the population is divided into several populations, in each of which Hardy-*Weinberg equilibrium* obtains at λ but between which the frequency of the focal variant differs, then Vargλ=2pλ1−pλ1+FST,λ, where FST,λ is the value of FST at locus λ.
The covariance term in Eq. ( A.3) expands out to (A.4) Cov(gλ,Y)=Cov(gλmat+gλpat,Y*+∑l∈Lαld(glmat+glpat)+∑l∈Lαli,m(glm,mat+glm,pat)+∑l∈Lαli,f(glf,mat+glf,pat)+ϵ)=Cov(gλmat+gλpat,∑l∈Lαld(glmat+glpat))+Cov(gλmas,∑l∈Lαli,m(glm,mat+glm,pat)+∑l∈Lαli,f(glf,mat+glf,pat))+Cov(gλpat,∑l∈Lαli,m(glm,mat+glm,pat)+∑l∈Lαli,f(glf,mant+glf,pat))+Cov(gλ,ϵ)=∑l∈L([Cov(gλmat,glmat)+Cov(gλmat,glpat)+Cov(gλpat,glmat)+Cov(gλpat,glpat)]αld+[Cov(gλmat,glm,mat+glm,pat)]αli,m+[Cov(gλpat,glf,mat+glf,pat)]αli,f+[Cov(gλmat,glf)]αli,f+[Cov(gλpat,glm)]αli,m)+Cov(gλ,ϵ)=∑l∈L(2(DM+D˜M)αld+[Cov(gλmat,glmm,mat+glmm,pat)]αli,m+[Cov(gλpat,glf,mmat+glf,pat)]αli,f+[Cov(gλmat,glf)]αli,f+[Cov(gλpat,glm)]αli,m)+Cov(gλ,ϵ), where Dλl and D˜λl are the degrees of cis- and trans-linkage disequilibrium between the focal alleles at loci λ and l in the GWAS sample. Since gλmat equals gλm,mat or gλm,pat with equal probability, Cov(gλmat,glm,mat+glm,pat)=Dλl′+D˜λl′, and similarly, Cov(gλpat,glf,mat+glf,pat)=Dλl′+D˜λl′ (here, Dλl′ and D˜λl′ are the LDs in the parents of the sample, assumed to be equal across mothers and fathers). Since maternal transmission is independent of paternal genotype, and vice versa, Cow gλmat,glf=Covgλm,glf/2 and Covgλpat,glm=Covgλf,glm/2. So (A.5) Cov(gλ,Y)=∑l∈L(2(Dλl+D˜λl)αld+(Dλl′+D˜λl′)(αli,m+αli,f)+12[Cov(gλm,glf)]αli,f+12[Cov(gλf,glm)]αli,m)+Cov(gλ,ϵ).
If αli,m=αli,f=αli, then (A.6) Cov(gλ,Y)=∑l∈L(2(Dλl+D˜λl)αld+(2(Dλl′+D˜λl′)+12[Cov(gλm,glf)+Cov(gλf,glm)])αli)+Cov(gλ,ϵ)=∑l∈L(2(Dλl+D˜λl)αld+(2(Dλl′+D˜λl′)+12[8D˜λl])αli)+Cov(gλ,ϵ)=2∑l∈L((Dλl+D˜λl)αld+(∣Dλl′+D˜λl′+2D˜λl)αli)+Cov(gλ,ϵ). In the second line of Eq. ( A.6), we have used the fact that covariances across parents translate to covariances across maternal and paternal genomes in the offspring. Note, however, that Covgλm,gl and Covgλf,glm need not, in general be equal—e.g., they will not be so under sex-besed cross-trait assortative mating—which is why we could not apply a similar simplification to Eq. ( A.5).
Dividing Eq. ( A.6) by Vargλ, and recognizing that, for l∈Llocal, cλ≈0, we recover Eq. [ 3] in the Main Text.
## Sibling GWAS
Consider two full siblings. Let glmat,1 and glmat,2 indicate whether sib 1 and sib 2 respectively inherited the focal (trait-increasing) allele from their mother at locus l. Let glpat,1 and glpat,2 be analogous indicators for paternal transmission. Write Δglmat=glmat,1−glmat,2 and Δglpat=glpat,1−glpat,2. Since maternal and paternal transmission are independent, Δglmat and Δglpat are independent for all pairs of loci l and l′ (including l=l′). The difference in the two siblings’ genotypic values at locus l is Δgl=Δglmat+Δglpat. From Eq. ( A.1), the difference in their trait values is (A.7) ΔY=∑l∈LΔglαld+Δϵ, where Δϵ is the difference in the environmental disturbances experienced by the two siblings. Notice that the indirect effects cancel out of Eq. ( A.7), since the parental genotypes are the same for the two siblings. So, in a sib-GWAS for trait Y, the estimated effect size at λ is αˆλsib=CovΔgλ,ΔYVarΔgλ=CovΔgλ,∑l∈L Δglαld+ΔϵVarΔgλ=CovΔgλmat+Δgλpat,∑l∈L Δglmat+Δgipatαld+CovΔgλ,ΔϵVarΔgλmat+Δgλpat=∑l∈L CovΔgλman,Δglmat+CovΔgλpat,Δglpatαld+CovΔgλ,ΔϵVarΔgλmat+VarΔgλpat=∑l∈L EΔgλmatΔglmat+EΔgλpatΔglpatαld+CovΔgλ,ΔϵEΔgλmat2+EΔgλpat2, since CovΔgλmat,Δglpat=CovΔglmat,Δgλpat=0 (line 3) and EΔgkmat=EΔgkpat=0 for all loci k (line 4). The denominator EΔgλmat2+EΔgλpat2=Hλ, the fraction of parents in the family GWAS sample who are heterozygous at locus λ. The only non-zero contributions to EΔgλmatΔglmat and EΔgλpatΔglpat come from parents who are heterozygous at both λ and l. Such parents are either ‘coupling’ double-heterozygotes carrying the focal alleles at λ and l in coupling phase (i.e., inherited from the same parent), or ‘repulsion’ double-heterozygotes carrying the focal alleles at λ and l in repulsion phase (inherited from different parents). Among parents, let the fractions of coupling and repulsion double-hets for loci λ and l be Hλlcoup and Hλlr respectively. If the recombination rate between the loci is cλl♀ in females and cλl♂ in males, then EΔgλmatΔglmat=EΔgλmatΔglmat∣motheriscouplingdouble-hetHλlcoup+EΔgλmatΔglmat∣motherisrepulsiondouble-hetHλlrep=12−cλl♀Hλlcoup−Hλlr=1−2cλl♀(Dλl′−D˜λl′), since Hλlcoup−Hλlr=2(Dλl′−D˜λl′), where Dλl′ and D˜λl′ are the cis- and trans-LD between the focal/traitincreasing alleles at λ and l among parents.. Similarly, EΔgλpatΔglpat=1−2cλl♂(Dλl′−D˜λl′) So (A.8) αˆλd,sib=2∑l∈L 1−2cλl(Dλl′−D˜λl′)αld+CovΔgl,ΔϵHλ, where cλl is the sex-averaged recombination fraction between λ and l. SinceCovΔgl,Δϵ=0, and recognizing that, for l∈Llocal, cλl≈0 and D˜λl≪Dλl in expectation, we recower Eq. [ 7] in the Main Text.
## Indirect effects: transmitted vs. untransmitted alleles
In Eq. ( A.2), glmat represents the allele that was transmitted maternally from among the set of maternal alleles glm,mat,glm,pat. Thus, if the maternally transmitted allele was the grandmaternal allele (with probability $\frac{1}{2}$, and in which case glmat=glm,mat), then the untransmitted allele at locus l is the grandpaternal allele, with genotypic value glm,pat. To make this distinction clear, we write glmatT for the genotypic value of the maternally transmitted allele at locus l, and glmatU for the maternally untransmitted allele at locus l. Similarly, glpatT and glpatU represent the paternally transmitted and untransmitted alleles at l. The transmitted and untransmitted genotypes are glT=glmanT+glpatT and glU=glmatU+glpatU respectively.
## Estimating direct effects
The regressions of the trait value on the transmitted and untransmitted genotypes are αˆλT=CovgλT,YVargλT=CovgλT,YVargλandαˆλU=CovgλU,YVargλU=CovgλU,YVargλ, where we have used the fact that, since transmission at λ is random, VargλT=VargλU=Vargλ.
The estimate of the direct effect of the focal variant at λ is then αˆλd=αˆλT−αˆλU=CovgλT,Y−CovgλU,YVargλ.
We have (A.9) Cov(gλmast,Y)=Cov(gλmat,Y*+∑l∈L(glmatT+glpatT)αld+∑l∈L(glm,mat+glm,pat)αli,mm+∑l∈L(glf,mat+glf,pat)αli,f+ϵ)=∑l∈L[Cov(gλmatT,glmatT)+Cov(gλmatT,glpatT)]αld+∑l∈L[Cov(gλmatT,glm,mat)+Cov(gλmatT,glm,pat)]αli,m+∑l∈L[Cov(gλmatT,glf,mat+glf,pat)]αli,f+Cov(gλmatT,ϵ)=∑l∈L[Dλl′(1−cλl♀)+D˜λl′cλl♀+Cov(gλmatT,glpatT)]αld+∑l∈L(Dλl′+D˜λl′)αli,m+∑l∈L[Cov(gλmatT,glf,mat+glf,pat)]αli,f+Cov(gλmanT,ϵ), and (A.10) Cov(gλmatU,Y)=Cov(gλmatU,Y*+∑l∈L(glmatT+glpatT)αld+∑l∈L(glm,mat+glm,pat)αli,m+∑l∈L(glf,mat+glf,pat)αli,f+ϵ)=∑l∈L[Cov(gλmatU,glmatT)+Cov(gλmatU,glpatT)]αld+∑l∈L[Cov(gλmatU,glm,mat)+Cov(gλmatU,glm,pat)]αli,m+∑l∈L[Cov(gλmatU,glf,mat+glf,pat)]αli,f+Cov(gλmatU,ϵ)=∑l∈L[Dλl′cλl♀+D˜λl′(1−∣cλl♀)+Cov(gλmatU,glpatT)]αld+∑l∈L(Dλl′+D˜λl′)αli,m+∑l∈L[Cov(gλmatU,glf,mat+glf,pat)]αli,f+Cov(gλmatU,ϵ). Since CovgλmatT,glpatT=CovgλmatU,glpatT and CovgλmatT,gif,mat+glf,pat=CovgλmatU,glf,mat+glf,pat, CovgλmatT,Y−CovgλmatU,Y=∑l∈L Dλ′(1−cλl♀)+D˜λl′cλl♀αld−∑l∈L Dλl′cλl♀+D˜λl′(1−cλl♀)αld+CovgλmatT−gλmatU,ϵ=∑l∈L (1−2cλl♀)Dλl′−D˜λl′αld+CovgλmatT−gλmat,ϵ. Similarly, CovgλpatT,Y−CovgλpatU,Y=∑l∈L 1−2cλ♂Dλl′−D˜λlrαld+CovgλpatT−gλpatU,ϵ. Since gλT=gλmatT+gλpatT and gλU=gλmanU+gλpatU, cov(gλT,Y)−cov(gλU,Y)=2∑l∈L(1−2cλl)(Dλl′−D˜λl′)αld+cov(gλT−gλU,ϵ), where cλl is the sex-averaged recombination fraction between λ and l. Therefore, the transmitted-untransmitted regression coefficient at locus λ is αˆλd,T−U=CovgλT,Y−CovgλU,YVargλ=2∑l∈L (1−2cλl)Dλl′−D˜λl′αld+CovgλT−gλU,ϵVλ.
## Estimating indirect effects
The coefficient in the regression of the trait value Y on the untransmitted genotype gλU at locus, αˆλU, has sometimes been considered to provide an estimate of the indirect ‘family’ effect of the focal variant at λ:αˆλi=αˆλU. From Eq. ( A.10) and its analog for the paternally untransmitted allele, CovgλmatU+gλpatU,Y=∑l∈L 2Dλl′cλ′+2D˜λl′1−cλl+CovgλmatU,glpatT+CovgλpatU,glmatTαld+Dλl′+D˜λl′αli,m+∑l∈L CovgλmatU,glf,mat+glf,patαli,f+Dλl′+D˜λl′αli,f+∑l∈L CovgλpatU,glmm,mat+glmm,patαli,m+CovgλmatU,ϵ+CovgλpatU,ϵ, where cλl is the sex-averaged recombination fraction. In this expression, CovgλmatU,glpatT+CovgλpatU,glmatT=CovgλmatT,glpatT+CovgλpatT,glmatT=2D˜λl, while CovgλmatU,glf,mat+glf,pat=CovgλmatU,glpatU+glpatT=CovgλmatU,glpatU+CovgλmatU,glpatT=2D˜λl, and, similarly, CovgλpatU,glm,mat+glm,pat=2D˜λl. So (A.12) Cov(gλU,Y)=Cov(gλmatU+gλpatU,Y)=∑l∈L[2(Dλl′cλl+D˜λl′(1−cλl)+D˜λl)αld+(Dλl′+D˜λl′+2D˜λl)(αli,m+αli,f)]+Cov(gλU,ϵ).
If we assume that indirect effects via the maternal and paternal families are equal (αli,m=αli,f=αli), then Eq. ( A.12) simplifies further to (A.13) Cov(gλU,Y)=2∑l∈L[(Dλl′′cλl+D˜λl′(1−cλl)+D˜λl)αld+(DM′+D˜Ml′+2D˜λl)αli]+Cov(gλU,ϵ).
In this case, the estimate of the indirect effect of the focal allele at λ is (A.14) αˆλi=CovgλU,YVargλ=2∑l∈L Dλl′cλl+D˜λl′1−cλl+D˜λαld+Dλl′+D˜λl′+2D˜λlαli+CovgλU,ϵVargλ.
## Polygenic scores and their phenotypic correlations
Suppose that we have estimated effect sizes αˆλ at a set of genotyped loci λ∈Λ using a population GWAS for trait 1. For each individual, we can then compute a polygenic score: (A.15) PGS1=∑λ∈Λgλαˆλpop. PGSs are often treated as predictions of individuals’ genetic values for traits. In this regard, we might therefore be interested in the covariance across the population between the PGS for a trait and individuals’ values for that trait: CovPGS1,Y1. Additionally, if PGSs are treated as predictions of genetic values of traits, then we might be interested in how the PGS calculated for one trait covaries with the value of another trait: CovPGS1,Y2. Such covariances might be informative of genetic correlations between traits, or pleiotropy of the alleles underlying genetic variation in the traits. We focus on the two-trait covariance, since it nests the single-trait covariance as a special case. If the total set of loci causally underlying variation in traits 1 and 2 is L, then the population covariance between the PGS for trait 1 and the value of trait 2 is (A.16) Cov(PGS1,Y2)=Cov(∑λ∈Λgλαˆλpop,∑l∈Lglβl)=Cov(∑λ∈Λ(gλm+gλp)αˆλpop,∑l∈L(glm+glP)βl)=2∑λ∈Λ∑l∈L(Dλl+D˜λl)αˆλpopβl. The effect size estimates from the population GWAS for trait 1 are αˆλpop=2Vλ∑l′∈L (Dλ′+D˜λl′)αl′≈αλ+2Vλ∑l′∈Ll′≠λ (Dλ′+D˜λl′)αl′, and so Eq. ( A.16) is, in general, (A.17) Cov(PGS1,Y2)=∑λ∈Λ2pλ(1−pλ)αλβλ+2∑λ∈Λ∑l∈Ll≠λ(Dλl+D˜λl)αλβl (A.18) +4∑λ∈Λ∑l′∈Ll′≠λ∑l∈Ll≠λ1Vλ(Dλl′+D˜λl′)(Dλl+D˜λl)αl′βl.
In a family-based study, we might instead be interested in the covariance between siblings’ differences in the trait-1 population PGS and their differences in trait 2. We can write this covariance in our model as (A.19) Cov(ΔPGS1,ΔY2)=Cov(∑λ∈Λ(Δgλm+Δgλp)αˆλpop,∑l∈L(Δglm+Δglp)βl)=E[(∑l∈Λ(Δgλm+Δgλp)αˆλpop)(∑l∈L(Δglm+Δglp)βl)]=∑λ∈Λ∑l∈LE[(Δgλm+Δgλp)(Δglm+Δglp)αˆλpopβl]=∑λ∈Λ∑l∈L(E[ΔgλmΔglmαˆλpopβl]+E[ΔgλpΔlpαˆλpogβl]), since maternal and paternal transmission are conditionally independent. Focusing on maternal transmission, and writing hλlc,m and hλlr,m for the events that the mother is respectively a coupling and a repulsion heterozygote at loci λ and l, with Hλlcoup and Hλlrep their associated probabilities (which are assumed to be the same for mothers and fathers), E[ΔgλmΔglmα^λpopβl]=E[ΔgλmΔglmα^λpopβl∣hλlc,m]Hλlcoup+E[ΔgλmΔglmα^λpopβl∣hλlr,m]Hλlrep=(E[ΔgλmΔglm∣hλlc,m]Hλlcoup+E[ΔgλmΔglm∣hλlr,m]Hλlrep)α^λpopβl=(12−cλl♀)(Hλlcoup−Hλlrep)α^λpopβl=(1−2cλl♀)(Dλl′−D˜λl′)α^λpopβl, with Dλl′ and D˜λl′ measured in the parents. Similarly, EΔgλPΔglPαˆλpogβl=1−2cλl♂Dλl′−D˜λlααˆλpopβl, and so Eq. ( A.19) becomes (A.20) Cov(ΔPGS1,ΔY2)=2∑λ∈Λ∑l∈L(1−2cλl)(Dλl′−D˜λl′)αˆλPopβl, where cλl is the sex-averaged recombination fraction between λ and l.
Before we substitute the population GWAS estimates αˆλpop into Eq. ( A.20), it is worth considering what value this expression would take if effect sizes were correctly estimated at every study locus, αˆλpop=αλ. In this case, Eq. ( A.20) becomes (A.21) Cov(ΔPGS1,ΔY2)=2∑λ∈Λ∑l∈L(1−2cλl)(Dλl−D˜λl)αλβl=∑λ∈Λ2pλ(1−pλ)αλβλ+2∑λ∈Λ∑l∈Ll≠λ(1−2cλl)(Dλl′−D˜λl′)αλβl. If the two traits are distinct, then the first term in Eq. ( A.21) is the genic covariance of traits 1 and 2 across the set of study loci (more precisely, tagged locally by the study loci), and reflects systematic pleiotropy at these loci; this term would, for example, be positive if alleles tend to have same-direction effects on traits 1 and 2. If we were studying only one trait, then αλ=βλ, and the first term would be the genic variance of the trait across study loci, ∑λ∈Λ 2pλ1−pλαλ2. The socond term in Eq. ( A.21) is an effect of linkage disequilibria between study loci and the loci that are causal for trait 2; these LDs are absorbed by the PGS because the PGS is a sum across loci. In the absence of such LDs, or in cases where the cis- and trans-LDs are equal so that Dλl′−D˜λl′=0, Eq. ( A.21) would equal the genic variance in the single-trait case and the genic covariance in the two-trait case.
The effect size estimates from a population GWAS are in fact αˆλpop=2Vλ∑l∈L (Dλl′+D˜λl′)αl′≈αλ+2Vλ∑l′∈Ll′≠λ (Dλ′+D˜λl′)αl′, Dλ′ and D˜λl′ are measured in the sample. We assume these to be equal to the values in parents in the family-based GWAS, D˜λl′ and D˜λl′, and so the value taken by Eq. ( A.20) is (A.22) Cov(ΔPGS1,ΔY2)=2∑λ∈Λ∑l∈L(1−2cλl)(Dλl−D˜λl)α^λpopβl=2∑λ∈Λ∑l∈L(1−2cλl)(Dλl−D˜λl)(αλ+2Vλ∑l′∈Ll′≠λ(Dλl′+D˜λl′)αl′)βl=∑λ∈Λ2pλ(1−pλ)αλβλ︸pleiotropy+2∑λ∈Λ∑l∈Ll≠λ(1−2cλl)(Dλl−D˜λl)αλβl︸covariancefromLDabsorbedbyPGSbecauseitisasumacrossloci+4∑λ∈Λ∑l∈Ll≠λ(1−2cλl)(Dλl2−D˜λl2)αlβl/Vλ︸covariancefromLDabsorbedbyPGSbecauseeffectsizeestimatesabsorbLD+4∑λ∈Λ∑l∈Ll≠λ∑l′∈Ll′≠λ,l(1−2cλl)(Dλl′+D˜λl′)αl′(Dλl−D˜λl)βl/Vλ. ︸covariancefromsystematicLDbetweenvariantswithsamedirectionaleffectontrait In the absence of genetic confounding Dλl=D˜λl=0 or, more generally, if genetic stratification is such that the cis- and trans-LDs are equal Dλl−D˜λl=0), then Eq- (A.22) simplifies to the SNP-tagged genic covariance between traits 1 and 2: (A.23) cov(ΔPGS1,ΔY2)=∑λ∈Λ2pλ(1−pλ)αλβλ. If traits 1 and 2 are the same, then this is simply the SNP-tagged genic variance of the trait: Cov(△PGS,△Y)= ∑λ∈Λ 2pλ1−pλαλ2.
Eq. ( A.22) simplifies somewhat if we focus on a single trait αl=βl and assume that there is no trans-LD D˜$X = 0$; in this case, (A.24) Cov(ΔPGS,ΔY)=∑λ∈Λ2pλ(1−pλ)αλ2︸SNP-taggedgenicvariance+2∑λ∈Λl≠L∑l∈L(1−2cλl)Dλlαλαl︸variancefromLDabsorbedbyPGSbecauseitisasumacrossloci+4∑λ∈Λ∑l∈Ll≠λ(1−2cλl)Dλl2αl2/Vλ︸variancefromLDabsorbedbyPGSbecauseeffectsizeestimatesabsorbLD+4∑λ∈Λ∑l∈Ll≠λ∑l′∈Ll′≠λ,l(1−2cλl)DλlαlDλl′αl′/Vλ︸variancefromsystematicLDbetweenvariantswithsamedirectionaleffectontrait.
## Sources of genetic confounding
The calculations above reveal that genetic confounds in GWAS designs can depend on long-range LD in the sample and among parents of the sample. Here, we consider several possible sources of long-range LD.
## Same-trait assortative mating, or cross-trait assortative mating that is symmetric with respect to sex
We first consider the case where the strength of assortative mating between two traits, ass measured by their correlation coefficient across mating pairs, is equal in the female-male and male-female directions. Notice that this scenario covers same-trait assortative mating. In the case of cross-trait assortative mating, it could occur if assortative mating arises by mechanisms other than direct female (or male) mating preferences.
We assume that there is a constant correlation ρ among mating pairs for their phenotypic values of traits 1 and 2. In equilibrium, this will translate to a constant correlation ρG between their breeding values as well (e.g., Felsenstein 1981). To calculate ρG, we first note that, because assortative mating is bssed on phenotypic values and not breeding values per se, if we know the phenotypes of a pair of mates, we obtain no further information about the similarity of their breeding values; that is, (A.25) CovG1m,G2f∣Y1m,Y2f=CovG2m,G1f∣Y2m,Y1f=0. For the same reason, if we know the phenotypic values of two mates, then the trait-2 value of the male does not offer any information on the female’s trait-1 breeding value beyond that already offered by the female’s trait-1 phenotype, and vice versa; that is, EG1m∣Y1m,Y2f=EG1m∣Y1m;EG2f∣Y1m,Y2f=EG2f∣Y2f;EG2m∣Y2m,Y1f=EG2m∣Y2m;EG1f∣Y2m,Y1f=EG1f∣Y1f. If Y1 and G1, and similarly Y2 and G2, are bivariate normal, then (A.27) EG1∣Y1=EG1+h12Y1−EY1andEG2∣Y2=EG2+h22Y2−EY2 where h12 and h22 are the heritabilities of traits 1 and 2, respectively.
From the law of total covariance, (A.28) cov(G1m,G2f)=Cov{Y1m,Y2f}(E[G1m∣{Y1m,Y2f}],E[G2f∣{Y1m,Y2f}])+E{Y1m,Y2f}[Cov(G1m,G2f∣{Y1m,Y2f})]=Cov{Y1m,Y2f}(E[G1m∣Y1m],E[G2f∣Y2f])[fromEqs. A.25andA.26]=Cov(h12Y1m,h22Y2f)[fromEq.(A.27)]=h12h22Cov(Y1m,Y2f). Similarly, Cov(G2m,G1f)=h12h22CovY2m,Y1f.
Let V1 and V2 be the phenotypic variances of traits 1 and 2, and VG1 and VG2 their additive genetic variances, assumed to be the same across the sexes. Given the calculations above, the correlation among mates for their breeding values of traits 1 and 2, ρG, can be written (A.29) ρG=12[Cov(G1m,G2f)+Cov(G2m,G1f)]VG1VG2=h12h222[Cov(Y1m,Y2f)+Cov(Y2m,Y1f)]h12V1h22V2 (A.30) =h1h212[Cov(Y1m,Y2f)+Cov(Y2m,Y1f)]V1V2=h1h2ρ. When traits 1 and 2 are the same, we have ρG=h2ρ, a standard result (e.g., Wright 1921; Felsenstein 1981).
Expanding the numerator of Eq. ( A.29), (A.31) 12CovG1m,G2f+CovG2m,G1f=12CovG1m,mat+G1m,pat,G2f,mat+G2f,pat+CovG2m,mat+G2m,pat,G1f,mat+G1f,pat=12CovG1m,mat,G2f,mat+CovG1m,mat,G2f,pat+12CovG1m,pat,G2f,mat+CovG1m,pat,G2f,pat+12CovG2m,mat,G1f,mat+CovG2m,mat,G1f,pat+12CovG2m,pat,G1f,mat+CovG2m,pat,G1f,pat=12CovG1m,mat,G2f,mat+CovG2m,mat,G1f,mat+12CovG1m,mat,G2f,pat+CovG2m,mat,G1f,pat+12CovG1m,pat,G2f,mat+CovG2m,pat,G1f,mat+12CovG1m,pat,G2f,pat+CovG2m,pat,G1f,pat. But 12CovG1m,mat,G2f,mat+CovG2m,mat,G1f,mat=12Cov∑l∈L glm,matαl,∑l′∈L gl′f,matβl′+Cov∑l∈L glm,matβl,∑l,∈L gVf,matαl′=12∑l∈L ∑l′∈L Covglm,mat,gl′f,matαlβl′+∑l∈L ∑l′∈L Covglm,mat,gl′f,matαl′βl=12∑l∈L ∑l′∈L Covglm,mat,gl′f,matαlβl′+∑l∈L ∑l′∈L Covglm,mat,glf,matαl′βl=∑l∈L ∑l′∈L 12Covglm,mat,gl′f,mat+Covgl′mm,mat,glf,matαlβl′=∑l∈L ∑l′∈L D˜ll′αlβl′, since grandmaternal and grandpaternal alleles are transmitted to the offspring with equal probability, independently across maternal and paternal transmission. The three additional terms in Eq. ( A.31) likewise each amount to ∑l∈L ∑l′∈L D˜ll′αlβl′, and so (A.32) 12[Cov(G1m,G2f)+Cov(G2m,G1f)]=4∑l∈L∑l′∈LD˜ll′αlβl′. Noting that the trans-covariance at a given locus D˜l=pl1−plr˜ll, where r˜ll is the within-locus correlation (equal to the inbreeding coefficient at the locus), we can split Eq. ( A.32) into within- and between-locus terms: (A.33) 12[Cov(G1m,G2f)+Cov(G2m,G1f)]=4∑l∈Lpl(1−pl)r˜llαlβl+4∑l∈L∑l′∈Ll′≠lD˜ll′αl∣βl′. In the denominator of Eq. ( A.29), (A.34) VG1=VarG1mm=VarG1m,mat+G1mm,pat=VarG1m,mat+VarG1m,pat+2CovG1m,mat,G1m,pat, Expanding the first term, VarG1m,mat=Var∑l∈L glm,matαl=∑l∈L Varglm,matαl2+∑l∈L ∑l′∈Ll′≠l Covglm,mat,gl′m,matαlαl′=∑l∈L pl1−plαl2+∑l∈L ∑l′∈Ll′≠l Dll′′αlαl′. Similarly, the second term is Var(G1m,pat)=∑l∈Lpl(1−pl)αl2+∑l∈L∑l′∈Ll′≠lDll′′αlαl′. The third, covariance term in Eq. ( A.34) is Cov(G1m,mat,G1m,pat)=Cov∑l∈L glm,matαl,∑l′∈L gl′mpatαl′=∑l∈L ∑l′∈L Covglm,mat,gl′m,patαlαl′=∑l∈L ∑l′∈L 12Covglmm,mat,gl′m,pat+Covgl′m,mat,glm,patαlαl′=∑l∈L ∑l′∈L D˜ll′′,αlαl′=∑l∈L pl1−plr˜ll′αl2+∑l∈L ∑l′∈L D˜ll′′D˜′αlαl′. Putting these together in Eq. ( A.34), VG1=Var(G1m)=2∑l∈Lpl(1−pl)(1+r˜ll′)αl2+2∑l∈L∑l′∈Ll′≠l(Dll′′+D˜ll′′)αlαl′. Similarly, VG2=Var(G2m)=2∑l∈Lpl(1−pl)(1+r˜ll′)βl2+2∑l∈L∑l′∈Ll′≠l(Dll′′+D˜ll′′)βlβl′. In equilibrium, Dll′′=D˜ll′′=D˜ll′=Dll′*, for l≠l′, and r‾ll′=r˜ll=r˜ll*, so (A.35) 12[Cov(G1m,G2f)+Cov(G2m,G1f)]=4∑l∈Lpl(1−pl)rll*αlβl+4∑l∈L∑l′∈Ll′≠lDll′*αlβl′, (A.36) VG1=2∑l∈Lpl(1−pl)(1+r˜ll*)αl2+4∑l∈L∑l′∈Ll′≠lDll′*αlαl′+VE1≈Vg1++4∑l∈L∑l′∈Ll′≠lDll′*αlαl′, (A.37) VG2=2∑l∈Lpl(1−pl)(1+r˜ll*)βl2+4∑l∈L∑l′∈Ll′≠lDll′*βlβl′+VE2≈Vg2+4∑l∈L∑l′∈Ll′≠l*Dll′*βlβl′, where Vg1 and Vg2 are the genic variances of traits 1 and 2, and the approximations come from the fact that, under assortative mating for a polygenic trait, the sum of the ~|L|2 cross-locus trans-LD terms D˜ll′* dominates the sum of the |L| within-locus trans-LD terms D˜ll*=pl1−plr˜ll* (Crow and Kimura 1970, Ch. 4). Eq. ( A.29) in equilibrium is therefore (A.38) ρG=4∑l∈Lpl(1−pl)r˜ll*αlβl+4∑l∈L∑l′∈Ll′≠lDl′*αlβl′VG1VG2≈4∑l∈L∑l′∈Ll′≠lDll′*αlβl′(Vg1+4∑l∈L∑l′∈Ll′≠lDll′*αlαl′)(Vg2+4∑l∈L∑l′∈Ll′≠lDll′*βlβl′). We now consider some special cases.
## Same-trait assortative mating with equal effect sizes.
In the case of same-trait assortative mating, αl=βl, so Eq. ( A.38) simplifies to (A.39) ρG=4∑l∈L∑l′∈Ll′≠lDll′*αlαl′Vg+4∑l∈L∑l′∈Ll′≠lDll′*αlαl′, from which (A.40) 4∑l∈L∑l′∈Ll′≠lDl′*αlαl′≈ρG1−ρGVg(=h2ρ1−h2ρVg). Since, in equilibrium, Dll′=D˜ll′, this expression can also be written (A.41) 2∑l∈L∑l′∈Ll′≠l(Dll′*+D˜ll′*)αlαl′≈ρG1−ρGVg. Because the additive genetic variance VG=Vg+2∑l∈L ∑l′∈Ll′≠l (Dll′*+D˜ll′*)αlαl′, Eq. ( A.41) can also be written (A.42) VG=Vg/1−ρG, which is a classic result (e.g., Wright 1921; Crow and Kimura 1970, Ch. 4).
If we make the further assumption that effect sizes are the same across loci (αl=α for all l∈L), then Eq. ( A.41) becomes (A.43) 2∑l∈L∑l′∈Ll′≠l(Dll′*+D˜ll′*)≈1α2ρG1−ρGVg. In a population association study at locus l, assuming no indirect effects and no sources of genetic confounding other than assortative mating, the effect size estimate is αˆl=α+2Vl∑l′∈Ll′≠l Dll′*+D˜ll′*αl′, so that the proportionate bias in the effect size estimate at l is (A.44) αˆl−αlαl=2Vl∑l′∈Ll′≠l(Dll'*+D˜ll′*)αl′αl=2Hl∑l′∈Ll′≠l(Dll'*+D˜ll′*), since αl=αl by assumption and Vl≈Hl=2pl1−pl because assortative mating does not substantially increase within-locus homozygosity (Crow and Kimura 1970, Ch. 4). The average proportionate bias across loci is then (A.45) 1|L|∑l∈Lαˆl−αlαl=1|L|∑l∈L2Hl∑l′∈Ll′≠l(Dll′*+D˜ll′*)≈2|L|H¯∑l∈L∑l′∈Ll′≠l(Dll′*+D˜ll′*)≈1|L|H¯α2ρG1−ρGVg=1VgρG1−ρGVg=ρG1−ρG, where we have used Eq. ( A.43) and have assumed that minor allele frequencies do not differ widely across loci. Since ρG=h2ρ, where ρ is the phenotypic correlation among mates and h2=VG/*Vp is* the heritability of the trait, Eq. ( A.45) can also be written (A.46) 1|L|∑l∈Lα^l−αlαl=h2ρ1−h2ρ.
## Sex-symmetric cross-trait assortative mating with distinct genetic bases and equal effect sizes.
In the case of cross-trait assortative mating, if the sets of loci underlying the two traits, L1 and L2, are distinct, then αl≠0⇒βl=0 and βl≠0⇒αl=0. In this case, Eq. ( A.38) becomes (A.47) ρG=4∑l∈L1∑l′∈L2Dll′*αlβl′VG1VG2, from which (A.48) ρGVG1VG2=4∑l∈L1∑l′∈L2Dll′*αlβl′=2∑l∈L1∑l′∈L2(Dll′*+D˜ll′*)αlβl′.
Because assortative mating is cross-trait, the LDs that assortative mating induces across L1 and L2 will dominate the second-order LDs induced within L1 and within L2. Therefore, VG1≈Vg1 and VG2≈Vg2.
The effect size estimate at a locus l∈L1 in a population GWAS on trait 2 is (A.49) β^l≈2Vl∑l′∈L2(Dll′+D˜ll′)βl′≈2Hl∑l′∈L2(Dll′+D˜ll′)βl′, while the true effect size βl is zero, since l∉L2. In equilibrium, the average effect size estimate, and thus the average deviation of these estimates from the true values, is therefore (A.50) 1|L1|∑l∈L1β^l≈1|L1|∑l∈L12Hl∑l′∈L2(Dll′+D˜ll′)βl′≈2|L1|H¯1∑l∈L1∑l′∈L2(Dll′+D˜ll′)βl′, where we have assumed that minor allele frequencies are not very different across L1 H‾1 is the average heterozygosity in L1). If we further assume that effect sizes at causal loci are equal for each trait (αl=α for all l∈L1 and βl′=β for all l′∈L2), then Eq. ( A.50) can be written (A.51) 1|L1|∑l∈L1β^l≈2|L1|H¯1∑l∈L1∑l′∈L2(Dll′+D˜ll′)β=α|L1|H¯1α2×2∑l∈L1∑l′∈L2(Dll′+D˜ll′)αβ=αVg1×ρGVG1VG2[fromEq. A.48]≈ρGVG1VG2α=VG1VP1⋅VG2VP2ρVG1VG2α=ρVG1VP1VP2α, recalling from Eq. ( A.30) that ρG=h1h2ρ.
In the further special case where both the genetic and the phenotypic variances of the two traits are equal, then so are the heritabilities of the two traits. In this case, Eq. ( A.51) simplifies to (A.52) 1|L1|∑l∈L1β^l≈VGVPρα=h2ρα, where h2 is the common heritability of the two traits.
## Sex-symmetric cross-trait assortative mating for traits with different genetic architectures.
Eq. ( A.52) reveals an interesting role for genetic architecture in the bias that cross-trait assortative mating can generate in population association studies performed at non-causal loci. Suppose, as we did in deriving Eq. ( A.52), that the two traits on which assortative mating is based have the same genetic and phenotypic variances, VG and V, and therefore also the same heritabilities, h2. We shall make the further assumption that the traits have the same genic variance, Vg. Assume further that the sets of loci underlying traits 1 and 2, L1 and L2, have similar mean heterozygocities ≈H‾. Normalize the effect size sizes at loci causal for trait 2 to β=1, so that the traits’ common genic variance is Vg=L2H‾.
Suppose that we now perform a population GWAS for trait 2. At loci that are causal for trait 2 l∈L2), we will estimate effect sizes accurately: βˆl≈1 (there will be a small positive second-order bias, of order ρ2, since the locus l∈L2 comes into positive LD with loci l′∈L1, which in turn have come into positive LD with loci l″∈L2).
At loci that are causal for trait 1l∈L1, and which therefore have no effect on trait 2, we will estimate effect sizes on average as given by Eq. ( A.52): βˆl=h2ρα.
How does the number of loci underlying variation in trait 1, L1, affect this biased estimate of their effect on trait 2? For the genic variance of trait 1 to be the same as that of trait 2, Vg=L1H‾α2=L2H‾β2=L2H‾, and so we must have α2=L2/L1. Substituting this into the average effect size estimate at non-causal loci, βˆl=h2ρL2/L1.
So, the average effect size estimate at causal loci l∈L2 is βˆl≈1, while the average effect size estimate at non-causal loci l∈L1 is βˆl=h2ρL2/L1. How do these two quantities compare? If the number of loci underlying the two traits is the same, L1=L2, and effect size estimates at non-causal loci are smaller than those at causal loci by a factor of about h2ρ. However, if there are more loci underlying trait 2 than underlying trait 1—i.e., if trait 1 has a more concentrated genetic architecture L1<L2—then the effect size estimates at non-causal loci will be closer to those at causal loci. Indeed, if trait 1 has a sufficiently concentrated architecture relative to trait 2, specifically, if L1<h4ρ2L2, then the effect size estimates at non-causal loci will, on average, be larger in magnitude than effect size estimates at causal loci.
*More* generally, the calculations above suggest that, in a more realistic scenario where effect sizes vary across loci, the trait-2 GWAS distribution of magnitudes of effect size estimates at trait-1 loci (non-causal) will overlap more with the distribution of magnitudes of effect size estimates at trait-2 loci (causal) if the genetic architecture of trait 1 is more concentrated (Fig. 4). This will lead to a greater number of trait 1 loci being identified as statistically significantly associated with trait 2 in the trait-2 GWAS.
## Cross-trait assortative mating that is asymmetric with respect to sex
We now consider the case where the strength of assortative mating between two traits, as measured by their correlation coefficient across mating pairs, is not equal in the female-male and male-female directions. This is clearest in the case of an active mate preference exhibited by one sex for some phenotype exhibited by the other sex.
To study this case, we make several simplifying assumptions. First, we assume that the genetic bases of variation in the two traits are distinct: αl≠0⇔βl=0. Second we assume that there is only one active direction of assortative mating: female trait 1 and male trait 2. That is, conditional on the mother’s breeding value for trait 1 and the father’s breeding value for trait 2, there is no correlation between the mother’s breeding value for trait 2 and the father’s breeding value for trait 1: CovG2m,G1f∣{G1m,G2f}=0.
Suppose that there is a constant correlation ρG between mothers’ breeding values for trait 1 and fathers’ breeding values for trait 2: (A.53) ρG=CovG1m,G2fVG1VG2. To study the genetic consequences of this assortment, we need to know the average bi-directional correlation among mates for traits 1 and 2 (Eq. A.29). Since traits 1 and 2 will come into a positive genetic correlation via assortative mating of female trait 1 and male trait 2, there will also be a positive covariance between mothers’ breeding values for trait 2 and fathers’ breeding values for trait 1, which we can express using the law of total covariance: (A.54) CovG2m,G1f=Cov{G1m,G2f}EG2m∣{G1m,G2f},EG1f∣{G1m,G2f}+E}G1m,G2f}CovG2m,G1f∣{G1m,G2f}=Cov{G1m,G2f}EG2m∣G1m,EG1f∣G2f. If G1m and G2m are bivariate normal (more generally, if G2m=a+bG1m+ε, with E[ε]=EεG1m=0), then EG2m∣G1m=EG2m+ρm1,m2VG2/VG1G1m−EG1m=EG2mm+ρm1,m2G1m−EG1mm, where ρm1,m2=CorrG1m,G2m is the genetic correlation between traits 1 and 2 in mothers, and where we have assumed that the two traits have equal variance. Similarly, if G1f and G2f are bivariate normal, then EG1f∣G2f=EG1f+ρfi,f2G2f−EG2f. Substituting these expressions into Eq. ( A.54), (A.55) CovG2m,G1f=ρm1,m2ρf1,f2CovG1m,G2f. But, in our case, ρm1,m2=ρf1,f2, the common value of which we shall call ρ12, and so the average bi-directional correlation is (A.56) 12CovG1m,G2f+CovG2m,G1fVG1VG2=121+ρ122CovG1m,G2fVG1VG2=ρG21+ρ122. Given this value, the calculations of the effect of assortative mating on the weighted sums of cis- and trans-covariances, and thus on the additive genetic variance, proceed as for the case of symmetric assortative mating above.
Assuming the genetic bases of the two traits to be distinct, we may substitute the average bi-directional correlation, ρG1+ρ$\frac{122}{2}$, into Eq. ( A.48) to find (A.57) ρG(1+ρ122)=4∑l∈L1∑l′∈L2(Dll′+D˜ll′)αlβl′VG1VG2. But ρ12=2∑l∈L1∑l′∈L2(Dll′+D˜ll′)αlβl′VG1VG2, and so Eq. ( A.57) can be written as the quadratic equation ρG1+ρ122=2ρ12, the relevant solution to which is ρ12=1−1−ρG2/ρG. If ρG is small, we use the first-order Taylor approximation 1−ρG2≈1−ρG$\frac{2}{2}$ to find (A.58) ρG2≈ρ12=2∑l∈L1∑l′∈L2(Dll′+D˜ll′)αlβl′VG1VG2≈2∑l∈L1∑l′∈L2(Dll′+D˜ll′)αlβl′Vg1Vg2.
In the particular scenario we have simulated in Fig. 2, Vg1=Vg2,αl=1 for all l∈L1, and βl=1 for all l∈L2, so Eq. ( A.58) further simplifies to (A.59) 4∑l∈L1∑l′∈L2(Dll′+D˜ll′)=ρVg1 *In a* population association study for trait 2 performed at a locus l∈L1 (so that βl=0), (A.60) β^l=βl+2Vl∑l′∈L2(Dll′+D˜ll′)βl′=2Vl∑l′∈L2(Dll′+D˜ll′). Across loci in L1, the average estimate is (A.61) β^l¯=1|L1|∑l∈L12Vl∑l′∈L2(Dll′+D˜ll′). In our simulations, pl≈$\frac{1}{2}$ for all l so that Vl≈2pl(1−pl)=$\frac{1}{2}$, and L1=L2=500, so Vg1=Vg2=250. Under this configuration, β^¯$l = 1$|L1|∑l∈L12Vl∑l′∈L2(Dll′+D˜ll′)=1500∑l∈L$\frac{121}{2}$∑l′∈L2(Dll′+D˜ll′)=4500∑l∈L1∑l′∈L2(Dll′+D˜ll′)=ρGVg1500=ρG/2. The trait we simulated is genetic, with heritability 1, and so ρG=ρ, the phenotypic correlation among mates. We chose a strength of assortative mating of ρ=0.2, and so, in equilibrium, the average effect size estimate at non-causal loci should be approximately 0.1, which is indeed the case in Fig. 2.
## Sex-asymmetric cross-trait assortative mating for traits with different genetic architectures.
For the case where the numbers of loci underlying traits 1 and 2 differ, and noting that the ‘effective’ correlation among mates in the sex-asymmetric case is approximately half that in the sex-symmetric case (Eq. A.58), we can perform a similar back-of-the-envelope calculation as in the sex-symmetric cross-trait assortative mating case above to find that, when effect sizes are constant across trait-1 loci and constant across trait-2 loci (though differing across traits 1 and 2), the effect size estimates at trait-1 (non-causal) loci in a trait-2 population GWAS is, on average, a fraction h2ρ2L2/L1 of the estimates at trait-2 (causal) loci.
Thus, more generally, when the number of loci underlying trait 1 is small relative to the number of loci underlying trait 2, the distribution of magnitudes of effect size estimates at trait-1 loci in a trait-2 GWAS can overlap substantially with the distribution of magnitudes of effect size estimates at trait-2 loci (Fig. 4), causing variants at these non-causal trait-1 loci to show up as significant in the trait-2 GWAS.
## Neutral allele frequency divergence.
If allele frequency divergence between the two populations is neutral, frequency changes at different loci are independent of one another and of effect sizes, so the second term in square brackets above is zero in expectation. In addition, because Hardy-*Weinberg equilibrium* obtains within each population, non-zero expected values of Fλ derive only from allele frequency differences between the populations, so that Fλ=FST,λ in expectation. Therefore, E[2pλ(1−pλ)(α^λpop)2]=1(1+FST)2E[(pλ[1]−pλ[2])22pλ(1−pλ)]|L|E[(pl[1]−pl[2])2]E[αl2]=1(1+FST)2E[2FST,λ]|L|E[2FST,lHl]E[αl2]≈4|L|(1+FST)2(E[FST,l])2E[Hl]E[αl2]=4|L|(FST1+FST)2E[Hl]E[αl2], where Hl=2pl1−pl. If the ancestral allele frequency at l was pla, then EHl∣pla=2pla1−pla1−FST,l, and so EHl is calculated using the law of iterated expectations by averaging this quantity over the ancestral distribution of allele frequencies: EHl≈EHla1−FST, where Hla=2pla1−pla. So (A.63) E[2pλ1−pλαˆλpop2]≈4|L|FST1+FST21−FSTEHlaEαl2.
If allele frequency divergence between the two populations was neutral, then frequency changes at different loci are independent of one another, of effect sizes, and of recombination rates (assuming the loci are sufficiently far apart), so the second terms in square brackets in Eqs. ( A.68) above is zero in expectation, so that E2pλ1−pλ(αˆλpop,t)2=4A2(1−A)2Epλ[1]−pλ[2]22pλ1−pλ|L|Epl[1]−pl[2]2(1−c)2tEαl2=4A2(1−A)2(1−c)2t¯E2FST,λ|L|E2FST,lHlEαl2≈16A2(1−A)2(1−c)2t¯|L|FST2EHlEαl2, where (1−c)2t¯ is the average value of 1−cl2t taken across all pairs of loci l,l′.
Similarly, under drift in the ancestral populations, the average squared sibling-based effect size estimate can be simplified to E2pλ1−pλαˆλsib,t2≈16A2(1−A)2(1−c)2t(1−2c)2¯|L|FST2EHlEαl2, where (1−c)2t(1−2c)2¯ is the average value of 1−cl′2t1−2cll′ taken across all pairs of loci l,l′.
## Selection and phenotype-biased migration.
Above, in calculating the mean heterozygosity-weighted value of αˆλ2 under neutral frequency divergence between populations, we assumed that in Eq. ( A.62) the second term in the square brackets was zero, i.e., that the effect-size-signed population allele frequency difference was uncorrelated across loci. Howevever, when selection or phenotype-biased migration acts, this will no longer be true. For example, if higher genetic values of the trait were favoured in population 1 relative to population 2, then selection will on average have driven a mean shift such that Epl[1]−pl[2]αl>0. This in turn will drive systematic positive covariances between terms pl[1]−pl[2]αl and pl′[1]−pl′[2]αl′, and as these covariances are summed over all pairs of loci in Eq. ( A.62), the resulting inflation of the average squared effect size estimate (and other genome-wide summaries) could be quantitatively substantial.
As in the case of population structure, selection and phenotype-biased migration in the ancestral populations can drive systematic positive covariances between the terms pl[1]−pl[2]αl and pl′[1]−pl′[2]αl′ in Eqs. ( A.68) and (A.69) above, so that the second terms in square brackets in these equations do not cancel in expectation as they did under neutral divergence between the ancestral populations. Again, as these covariances are summed over all pairs of loci in Eqs. ( A.68) and (A.69), the resulting inflation of the average squared effect size estimate and other genome-wide summaries could be substantial.
## More general population stratification.
Given a sample of N individuals, the sample cis-LD between two markers λ and l can be written generally as (A.64) Dλl=1N−1∑$i = 1$N(Δgi,λmΔgi,lm+Δgi,λpΔgi,lp), where Δgi,km and Δgi,kP are the deviations of individual i ‘s maternal and paternal focal allele count at locus k from their mean frequencies. The trans-LD between λ and l is (A.65) D˜λl=1N−1∑$i = 1$N(Δgi,λmΔgi,lp+Δgi,lmΔgi,λp).
These cis- and trans-LD terms are equal only if (A.66) Dλl−D˜λl=1N−1∑$i = 1$N(Δgi,λm−Δgi,λp)(Δgi,lm−Δgi,lp)=0, i.e., if the maternal and paternal alleles at the one locus are exchangeable with respect to deviations of the allelic state at the other locus.
We might often be concerned with stratification along some specific axis of variation in our sample. Call this axis v, with every individual having a value along v, with mean zero across individuals (for example, in our two population case above, the vector v could be 1 for population 1 and −1 for population 2). The covariance of the maternal allele at locus l with the vector v is proportional to alm⋅v=∑i ai,lmvi. So the contribution of LD along this axis to the difference in cis- and trans-LD is (A.67) Dλl(v)−D˜λlv=Δgλm−ΔgλP⋅vΔglm−ΔglP⋅v, which is zero only if the maternal and paternal genotypes at the two loci are exchangeable with respect to each other along the axis v.
## Stabilizing selection attenuates estimates of the strength of assortative mating based on cross-chromosome PGS correlations
Recently, the strength of assortative mating has been estimated based on measurement of the correlation of polygenic scores across distinct sets of chromosomes (e.g., Yengo et al. 2018; Yamamoto et al. 2023). Were assortative mating acting in isolation, such correlations would be due entirely to the positive cis- and trans-LDs among same-effect alleles created by assortative mating. Since stabilizing selection, acting in isolation, generates negative cis-LDs among same-effect alleles, it will attenuate the positive cis-LDs generated by assortative mating, and therefore reduce the correlation in PGSs among distinct sets of chromosomes, leading to underestimates of the strength of assortative mating if this effect is not taken into account.
To quantify this attenuation, we first calculate the strength of (positive) cross-chromosome LDs expected under assortative mating alone; then we calculate the strength of (negative) cross-chromosome LDs expected under stabilizing selection alone; then, assuming these LDs to be generated independently of one another—so that the LDs generated under the joint action of assortative mating and stabilizing selection are the sums of the LDs expected under these forces alone—we calculate how much stabilizing selection attenuates the correlation in PGSs across distinct sets of chromosomes.
## Cross-chromosome correlations in PGSs.
The number of autosomes in the haploid set is n (=22 in humans). Label the set of loci on chromosome k that contribute variation to our trait of interest Lk; the overall set of loci underlying variation in the trait is L=L1,L2,…,Lk. We divide the chromosomes into distinct sets K1 and K2 (e.g., K1 could be the set of odd numbered chromosomes and K2 the even). Let L[1] and L[2] be the sets of causal loci on the chromosomes in K1 and K2 respectively (i.e., L(i)=∪k∈K1Lk).
Suppose that we have accurately estimated effect sizes at all loci l∈L. For each individual, we then calculate a polygenic score for K1 and for K2: P1=∑l∈L[1] glαl;P2=∑l′∈L[2] gl′αl′. We are interested in the correlation in the population between P1 and P2, and in particular, how this correlation is affected by assortative mating and stabilizing selection for the focal trait. The correlation can be written CorrP1,P2=CovP1,P2VarP1VarP2, with (A.89) Cov(P1,P2)=Cov(∑l∈L[1]glαl,∑l′∈L[2]gl′αl′)=∑l∈L[1]∑l′∈L[2]Cov(gl,gl′)αlαl′=2∑l∈L[1]∑l′∈L[2](Dll′+D˜ll′)αlαl′. Since, to make progress in the case of stabilizing selection, we will assume effect sizes to be equal across loci, we make that assumption now, so that (A.90) Cov(P1,P2)=2α2∑l∈L[1]∑l′∈L[2](Dll′+D˜ll′). Since every pair of loci l,l′ across L[1] and L[2] are by definition unlinked, under many processes (including assortative mating and stabilizing selection), the values of Dll′, and D˜ll′ will not differ much in expectation across locus pairs, in equilibrium. Therefore, we may approximate Dll′=D* and D˜ll′=D˜* for all l∈L[1] and l′∈L(2, so that Eq. ( A.90) simplifies further: (A.91) CovP1,P2=2L1L2D*+D˜*α2.
## Assortative mating alone.
Under assortative mating with equal effect sizes across loci, in equilibrium, LDs are approximately equal across locus pairs, regardless of the recombination rate between them; moreover, cis- and trans-LDs are equal (see above). Therefore, to calculate D*(=D˜*), we simply apportion the total LD given by Eq. ( A.40) among individual locus pairs: (A.92) h2ρ1−h2ρVg≈4∑l∈L∑l′∈Ll′≠lDl′*αlαl′=4|L|(|L|−1)α2D*⇒D*≈h2ρ1−h2ρVg4|L|(|L|−1)α2=h2ρ1∣−h2ρ|L|H¯α24|L|(|L|−1)α2=h2ρ1−h2ρH¯4(|L|−1)≈14⋅h2ρ1−h2ρ⋅H¯|L|, when |L| is large. Similarly, (A.93) D˜*≈14⋅h2ρ1−h2ρ⋅H¯|L|, so that the overall contribution of assortative mating to the covariance in Eq. ( A.91) is proportional to (A.94) D*+D˜*≈12⋅h2ρ1−h2ρ⋅H¯|L|.
## Stabilizing selection alone.
Under stabilizing selection, the total amount of negative cis-LD is given by Eq. ( A.87): (A.95) 2α2∑l∈L∑l′∈ll′≠lDll′=d=−Vg2c¯h(1+VS/VPh2)+1, where we have dropped the equilibrium ‘*’ markers. This expression does not easily decompose into terms from individual locus pairs. However, if we assume that stabilizing selection is relatively weak (VS/VP*≫1) and that the recombination process is such that the harmonic mean recombination rate c‾h~$\frac{1}{2}$ (as is the case in humans), Eq. ( A.95) can be approximated by 2α2∑l∈L∑l′∈ll′≠lDll′=d≈−Vg2c¯h(1+VS/VPh2)=−12⋅h2Vg1+VS/VP⋅1c¯h=−12⋅h2Vg1+VS/VP⋅2∑l,l′1/cll′|L|(|L|−1), from which we infer that, in expectation, 2α2Dll′≈−h2Vg1+VS/VP⋅1/cll′|L|(|L|−1). Therefore, for unlinked l and l′cll′=$\frac{1}{2}$, in expectation, (A.96) Dll′≈−1α2|L|(|L|−1)⋅h2Vg1+VS/VP=−H¯α2H¯|L|(|L|−1)⋅h2Vg1+VS/VP=−H¯(|L|−1)Vg⋅h2Vg1+VS/VP=−H¯|L|−1⋅h21+VS/VP≈−H¯|L|⋅h21+VS/VP. Stabilizing selection does not systematically generate trans-LD, so, in expectation, D˜ll′=0. Therefore, under stabilizing selection alone, the contribution of an unlinked locus pair to the covariance in Eq. ( A.91) is (A.97) D*+D˜*=D*≈−H¯|L|⋅h21+VS/VP.
## How much does stabilizing selection attenuate the signal of assortative mating?
Comparing Eqs. ( A.94) and (A.97), we find that the proportionate attenuation of assortative mating’s effect (in isolation) by the action of stabilizing selection is (A.98) −H¯|L|⋅h21+VS/VP12⋅h2ρ1−h2ρ⋅H¯|L|=−21+VS/VP⋅1−h2ρρ. For example, in the case of human height h2~0.8, the signal of assortative mating (strength ρ~0.25 is attenuated by stabilizing selection (strength VS/VP~30) by a proportionate amount of approximately $20\%$. That is, one might measure by other means (e.g., the phenotypic correlation among mates, together with an estimate of the heritability of height) that the strength of assortative mating is ρ=0.25, but estimating this strength from cross-chromosome PGS correlations without accounting or correcting for stabilizing selection on height would yield ρˆ≈0.2,$20\%$ smaller than the true value.
## One-locus GxE
We study the phenotypic model in Eq. [ 22], with the phenotype of individual i in family f given by (A.99) Yi=Y*+α+αf+αigi+ϵf+ϵi, where, across the population, Eαf=Eαi=Eϵf=Eϵi=0, and αi,ϵf, and ϵi are all independent of gi
## Sibling GWAS.
Let i and j be siblings in family f, and define ΔYf=Yi−Yj,Δgf=gi−gj, and Δϵf=ϵi−ϵj. A sibling association study returns an effect size estimate αˆsib=CovΔYf,ΔgfVarΔgf=Covα+αfΔgf+αigi−αjgj+Δϵf,ΔgfVarΔgf=Eα+αfΔgf2+Eαigi−αjgjΔgf+EΔϵfΔgfH, where H is the fraction of parents who are heterozygous at the focal locus. Since αi,αj,ϵi, and ϵj are genotype-independent perturbations, Eαigi−αjgjΔgf=EΔϵfΔgf=0, and so (A.100) αˆ=EαΔgf2+EαfΔgf2H=α+EαfΔgf2H, which deviates from α by an amount EαfΔgf2/H.
Let Δgfmat and Δgfpat be the difference in the genotypes of the siblings in family f due to maternal and paternal transmission. Because of the independence of maternal and paternal transmission in a given family, the term additional to α in Eq. ( A.100) can be split into EαfΔgfmat2/H and EαfΔgfpat2/H, which we can analyze separately.
If the mother is heterozygous, then Δgfmat2 equals 1 with probability $\frac{1}{2}$ and 0 with probability $\frac{1}{2}$; if the mother is homozygous, then Δgfmat2 is 0. Therefore, denoting by hm the event that the mother is heterozygous, EαfΔgfmat2H=12Eαf∣hmProbhmH=12Eαf∣hm. The same holds for paternal transmission, and so the deviation of the family-based estimate αˆ from α is (A.101) αˆ−α=EαfΔgf2H=Eαf∣h. That is, quite intuitively, if the average G×E effect αf is different in the families of heterozygous parents than in the population as a whole, then limiting estimation to the offspring of heterozygous parents will be problematic.
## Population GWAS.
Under the same one-locus model, a population association study returns an effect size estimate of (A.102) αˆpop=CovYi,giVargi=Covα+αf+αigi+ϵf+ϵi,giVargi=α+Covαfgi,giVargi. We can immediately see from Eq. ( A.102) that if the family environments are randomized across genotypes, such that αf and gi are independent (implying Covαfgi,gi=0), then the population estimate will coincide with α.
To calculate the deviation of the population estimate from α in the general case, let F be the inbreeding coefficient at the locus. Then Vargi=2p(1−p)(1+F), where p is the frequency of the focal variant, and the frequency of heterozygotes is f1=2p(1−p)(1−F) while the frequencies of the two homozygotes are f0=(1−p)2+p(1−p)F (zero focal alleles) and f2=p2+p(1−p)F (two focal alleles). The covariance term in Eq. ( A.102) can then be written Covαfgi,gi=Eαfgi2−EαfgiEgi=Eαfgi2−2pEαfgi=0×Eαf∣gi=0f0+1×Eαf∣gi=1f1+4×Eαf∣gi=2f2−2p0×Eαf∣gi=0f0+1×Eαf∣gi=1f1+2×Eαf∣gi=2f2=Eαf∣gi=1f1(1−2p)+4Eαf∣gi=2f2(1−p)=2Eαf∣gi=1p(1−p)(1−2p)(1−F)+4Eαf∣gi=2p2(1−p)+p(1−p)2F. The deviation of the population-based estimate from α is therefore (A.103) αˆpop−α=Covαfgi,giVargi=2Eαf∣gi=1p(1−p)(1−2p)(1−F)+4Eαf∣gi=2p2(1−p)+p(1−p)2F2p(1−p)(1+F)=Eαf∣gi=1(1−2p)1−F1+F+2Eαf∣gi=2(p+(1−p)F)11+F (A.104) ≈Eαf∣gi=1(1−2p)(1−2F)+2Eαf∣gi=2(p+(1−2p)F). The approximation holds when F is small.
An interesting special case is where homozygotes for the focal allele and heterozygotes have the same distribution of environments, so that Eαf∣gi=1=Eαf∣gi=2=Eαf∣gi>0. In this case, Eq. ( A.103) simplifies to (A.105) αˆpop−α=Eαf∣gi>0, which reveals that, if individuals who carry the focal allele tend to experience different environments to individuals who do not carry the focal allele, then the population GWAS estimate will deviate from the average effect under true randomization, α. Moreover, in this case, if Eαf∣gi=1 and Eαf∣h are the same—that is, if the mean environment of heterozygous offspring is the same as that for heterozygous parents—then the sibling and population-based effect size estimates are the same.
## References
1. Abdellaoui A., Dolan C. V., Verweij K. J., Nivard M. G.. **Gene–environment correlations across geographic regions affect genome-wide association studies**. *Nature Genetics* (2022) **54** 1345-1354. PMID: 35995948
2. Abecasis G. R., Cardon L. R., Cookson W. O. C.. **A general test of association for quantitative traits in nuclear families**. *American Journal of Human Genetics* (2000) **66** 279-292. PMID: 10631157
3. Allison D. B.. **Transmission-disequilibrium tests for quantitative traits**. *American Journal of Human Genetics* (1997) **60** 676-690. PMID: 9042929
4. Atwell S., Huang Y. S., Vilhjálmsson B. J., Willems G., Horton M., Li Y., Meng D., Platt A., Tarone A. M., Hu T. T.. **Genome-wide association study of 107 phenotypes in**. *Nature* (2010) **465** 627-631. PMID: 20336072
5. Barcellos S. H., Carvalho L. S., Turley P.. **Education can reduce health differences related to genetic risk of obesity**. *Proceedings of the National Academy of Sciences* (2018) **115** E9765-E9772
6. Benonisdottir S., Kong A.. **The genetics of participation: method and analysis**. *bioRxiv* (2022). DOI: 10.1101/2022.02.11.480067
7. Berg J. J., Harpak A., Sinnott-Armstrong N., Joergensen A. M., Mostafavi H., Field Y., Boyle E. A., Zhang X., Racimo F., Pritchard J. K.. **Reduced signal for polygenic adaptation of height in UK Biobank**. *eLife* (2019) **8** e39725. PMID: 30895923
8. Border R., Athanasiadis G., Buil A., Schork A., Cai N., Young A., Werge T., Flint J., Kendler K., Sankararaman S., W D. A., A Z. N.. **Cross-trait assortative mating is widespread and inflates genetic correlation estimates**. *Science* (2022a) **378** 754-761. PMID: 36395242
9. Border R., O’Rourke S., de Candia T., Goddard M. E., Visscher P. M., Yengo L., Jones M., Keller M. C.. **Assortative mating biases marker-based heritability estimators**. *Nature Communications* (2022b) **13** 1-10
10. Brown B. C., Price A. L., Patsopoulos N. A., Zaitlen N.. **Local joint testing improves power and identifies hidden heritability in association studies**. *Genetics* (2016) **203** 1105-1116. PMID: 27182951
11. Brumpton B., Sanderson E., Heilbron K., Hartwig F. P., Harrison S., Vie G. Å., Cho Y., Howe L. D., Hughes A., Boomsma D. I.. **Avoiding dynastic, assortative mating, and population stratification biases in Mendelian randomization through within-family analyses**. *Nature Communications* (2020) **11** 3519
12. Bulik-Sullivan B.. **Relationship between LD score and Haseman-Elston regression**. *BioRxiv* (2015). DOI: 10.1101/018283
13. Bulik-Sullivan B. K., Finucane H. K., Anttila V., Gusev A., Day F. R., Loh P.-R., Duncan L., Perry J. R., Patterson N., Robinson E. B.. **An atlas of genetic correlations across human diseases and traits**. *Nature Genetics* (2015a) **47** 1236-1241. PMID: 26414676
14. Bulik-Sullivan B. K., Loh P.-R., Finucane H. K., Ripke S., Yang J., Patterson N., Daly M. J., Price A. L., Neale B. M.. **LD Score regression distinguishes confounding from polygenicity in genome-wide association studies**. *Nature Genetics* (2015b) **47** 291-295. PMID: 25642630
15. Bulmer M. G.. **The effect of selection on genetic variability**. *American Naturalist* (1971) **105** 201-211
16. Bulmer M. G.. **Linkage disequilibrium and genetic variability**. *Genetics Research* (1974) **23** 281-289
17. Bürger R.. *The Mathematical Theory of Selection, Recombination, and Mutation* (2000)
18. Coop G., Przeworski M.. **Lottery, luck, or legacy. A review of “The Genetic Lottery: Why DNA matters for social equality”**. *Evolution* (2022a) **76** 846-853. PMID: 35225362
19. Coop G., Przeworski M.. **Luck, lottery, or legacy? The problem of confounding. A reply to Harden**. *Evolution* (2022b) **76** 2464-2468. PMID: 35915930
20. Crow J. F.. **Mapping functions**. *Genetics* (1990) **125** 669-671. PMID: 2204577
21. Crow J. F., Felsenstein J.. **The effect of assortative mating on the genetic composition of a population**. *Eugenics Quarterly* (1968) **15** 85-97. PMID: 5702332
22. Crow J. F., Kimura M.. *An Introduction in Population Genetics Theory* (1970)
23. Demange P. A., Hottenga J. J., Abdellaoui A., Eilertsen E. M., Malanchini M., Domingue B. W., Armstrong-Carter E., De Zeeuw E. L., Rimfeld K., Boomsma D. I.. **Estimating effects of parents’ cognitive and non-cognitive skills on offspring education using polygenic scores**. *Nature Communications* (2022) **13** 4801
24. Eaves L. J., Pourcain B. S., Smith G. D., York T. P., Evans D. M.. **Resolving the effects of maternal and offspring genotype on dyadic outcomes in genome wide complex trait analysis (“M-GCTA”)**. *Behavior Genetics* (2014) **44** 445-455. PMID: 25060210
25. Edelaar P., Bolnick D. I.. **Non-random gene flow: an underappreciated force in evolution and ecology**. *Trends in Ecology & Evolution* (2012) **27** 659-665. PMID: 22884295
26. Ewens W. J., Spielman R. S.. **The transmission/disequilibrium test: history, subdivision, and admixture**. *American Journal of Human Genetics* (1995) **57** 455-464. PMID: 7668272
27. Felsenstein J.. **Continuous-genotype models and assortative mating**. *Theoretical Population Biology* (1981) **19** 341-357
28. Fisher R. A.. **Statistical methods in genetics**. *Heredity* (1952) **6** 1-12
29. Fletcher J., Wu Y., Li T., Lu Q.. **Interpreting polygenic score effects in sibling analysis**. *BioRxiv* (2021). DOI: 10.1101/2021.07.16.452740
30. Freeman G.. **Statistical methods for the analysis of genotype-environment interactions**. *Heredity* (1973) **31** 339-354. PMID: 4589174
31. Fry A., Littlejohns T. J., Sudlow C., Doherty N., Adamska L., Sprosen T., Collins R., Allen N. E.. **Comparison of sociodemographic and health-related characteristics of UK Biobank participants with those of the general population**. *American Journal of Epidemiology* (2017) **186** 1026-1034. PMID: 28641372
32. Gauderman W. J., Mukherjee B., Aschard H., Hsu L., Lewinger J. P., Patel C. J., Witte J. S., Amos C., Tai C. G., Conti D.. **Update on the state of the science for analytical methods for gene-environment interactions**. *American Journal of Epidemiology* (2017) **186** 762-770. PMID: 28978192
33. Greene W. H.. (2018)
34. Haller B. C., Messer P. W.. **SLiM 3: forward genetic simulations beyond the Wright–Fisher model**. *Molecular Biology and Evolution* (2019) **36** 632-637. PMID: 30517680
35. Harpak A., Przeworski M.. **The evolution of group differences in changing environments**. *PLoS Biology* (2021) **19** e3001072. PMID: 33493148
36. Haworth S., Mitchell R., Corbin L., Wade K. H., Dudding T., Budu-Aggrey A., Carslake D., Hemani G., Paternoster L., Smith G. D.. **Apparent latent structure within the UK Biobank sample has implications for epidemiological analysis**. *Nature Communications* (2019) **10** 1-9
37. Hayes B., Goddard M.. **Genome-wide association and genomic selection in animal breeding**. *Genome* (2010) **53** 876-883. PMID: 21076503
38. Hayward L. K., Sella G.. **Polygenic adaptation after a sudden change in environment**. *eLife* (2022) **11** e66697. PMID: 36155653
39. Horwitz T. B., Keller M. C.. **A comprehensive meta-analysis of human assortative mating in 22 complex traits**. *bioRxiv* (2022). DOI: 10.1101/2022.03.19.484997
40. Howe L. J., Nivard M. G., Morris T. T., Hansen A. F., Rasheed H., Cho Y., Chittoor G., Ahlskog R., Lind P. A., Palviainen T.. **Within-sibship genome-wide association analyses decrease bias in estimates of direct genetic effects**. *Nature Genetics* (2022) **54** 581-592. PMID: 35534559
41. Josephs E. B., Stinchcombe J. R., Wright S. I.. **What can genome-wide association studies tell us about the evolutionary forces maintaining genetic variation for quantitative traits?**. *New Phytologist* (2017) **214** 21-33. PMID: 28211582
42. Kong A., Thorleifsson G., Frigge M. L., Vilhjalmsson B. J., Young A. I., Thorgeirsson T. E., Benonisdottir S., Oddsson A., Halldorsson B. V., Masson G.. **The nature of nurture: Effects of parental genotypes**. *Science* (2018) **359** 424-428. PMID: 29371463
43. Kong A., Thorleifsson G., Gudbjartsson D. F., Masson G., Sigurdsson A., Jonasdottir A., Walters G. B., Jonasdottir A., Gylfason A., Kristinsson K. T.. **Fine-scale recombination rate differences between sexes, populations and individuals**. *Nature* (2010) **467** 1099-1103. PMID: 20981099
44. Lander E. S., Schork N. J.. **Genetic dissection of complex traits**. *Science* (1994) **265** 2037-2048. PMID: 8091226
45. Lee H., Lee M. H.. *Disentangling linkage and population structure in association mapping* (2023a)
46. Lee H., Lee M. H.. *Theoretical interpretation of genetic studies in admixed populations* (2023b)
47. Lee J. J., Wedow R., Okbay A., Kong E., Maghzian O., Zacher M., Nguyen-Viet T. A., Bowers P., Sidorenko J., Karlsson Linnér R.. **Gene discovery and polygenic prediction from a genome-wide association study of educational attainment in 1.1 million individuals**. *Nature Genetics* (2018) **50** 1112-1121. PMID: 30038396
48. Li A., Liu S., Bakshi A., Jiang L., Chen W., Zheng Z., Sullivan P. F., Visscher P. M., Wray N. R., Yang J.. **mBAT-combo: a more powerful test to detect gene-trait associations from GWAS data**. *American Journal of Human Genetics* (2023) **110** 30-43. PMID: 36608683
49. Marchini J., Donnelly P., Cardon L. R.. **Genome-wide strategies for detecting multiple loci that influence complex diseases**. *Nature Genetics* (2005) **37** 413-417. PMID: 15793588
50. Martin A. R., Gignoux C. R., Walters R. K., Wojcik G. L., Neale B. M., Gravel S., Daly M. J., Bustamante C. D., Kenny E. E.. **Human demographic history impacts genetic risk prediction across diverse populations**. *American Journal of Human Genetics* (2017) **100** 635-649. PMID: 28366442
51. Morris T. T., Davies N. M., Hemani G., Smith G. D.. **Population phenomena inflate genetic associations of complex social traits**. *Science Advances* (2020) **6** eaay0328. PMID: 32426451
52. Mostafavi H., Harpak A., Agarwal I., Conley D., Pritchard J. K., Przeworski M.. **Variable prediction accuracy of polygenic scores within an ancestry group**. *eLife* (2020) **9** e48376. PMID: 31999256
53. Nei M., Li W.-H.. **Linkage disequilibrium in subdivided populations**. *Genetics* (1973) **75** 213-219. PMID: 4762877
54. Nivard M., Belsky D., Harden K. P., Baier T., Ystrom E., Lyngstad T. H.. **Neither nature nor nurture: Using extended pedigree data to elucidate the origins of indirect genetic effects on offspring educational outcomes**. *PsyArXiv* (2022). DOI: 10.31234/osf.io/bhpm5
55. Okbay A., Wu Y., Wang N., Jayashankar H., Bennett M., Nehzati S. M., Sidorenko J., Kweon H., Goldman G., Gjorgjieva T.. **Polygenic prediction of educational attainment within and between families from genome-wide association analyses in 3 million individuals**. *Nature Genetics* (2022) **54** 437-449. PMID: 35361970
56. Patel R. A., Musharoff S. A., Spence J. P., Pimentel H., Tcheandjieu C., Mostafavi H., Sinnott-Armstrong N., Clarke S. L., Smith C. J., Durda P. P.. **Genetic interactions drive heterogeneity in causal variant effect sizes for gene expression and complex traits**. *American Journal of Human Genetics* (2022) **109** 1286-1297. PMID: 35716666
57. Peiffer J. A., Romay M. C., Gore M. A., Flint-Garcia S. A., Zhang Z., Millard M. J., Gardner C. A., McMullen M. D., Holland J. B., Bradbury P. J.. **The genetic architecture of maize height**. *Genetics* (2014) **196** 1337-1356. PMID: 24514905
58. Pfaff C. L., Parra E. J., Bonilla C., Hiester K., McKeigue P. M., Kamboh M. I., Hutchinson R. G., Ferrell R. E., Boerwinkle E., Shriver M. D.. **Population structure in admixed populations: effect of admixture dynamics on the pattern of linkage disequilibrium**. *American Journal of Human Genetics* (2001) **68** 198-207. PMID: 11112661
59. Pirastu N., Cordioli M., Nandakumar P., Mignogna G., Abdellaoui A., Hollis B., Kanai M., Rajagopal V. M., Parolo P. D. B., Baya N.. **Genetic analyses identify widespread sex-differential participation bias**. *Nature Genetics* (2021) **53** 663-671. PMID: 33888908
60. Platt A., Vilhjálmsson B. J., Nordborg M.. **Conditions under which genome-wide association studies will be positively misleading**. *Genetics* (2010) **186** 1045-1052. PMID: 20813880
61. Price A. L., Patterson N. J., Plenge R. M., Weinblatt M. E., Shadick N. A., Reich D.. **Principal components analysis corrects for stratification in genome-wide association studies**. *Nature Genetics* (2006) **38** 904-909. PMID: 16862161
62. Price A. L., Zaitlen N. A., Reich D., Patterson N.. **New approaches to population stratification in genome-wide association studies**. *Nature Reviews Genetics* (2010) **11** 459-463
63. Pritchard J. K., Przeworski M.. **Linkage disequilibrium in humans: models and data**. *American Journal of Human Genetics* (2001) **69** 1-14. PMID: 11410837
64. Pritchard J. K., Rosenberg N. A.. **Use of unlinked genetic markers to detect population stratification in association studies**. *American Journal of Human Genetics* (1999) **65** 220-228. PMID: 10364535
65. Pritchard J. K., Stephens M., Rosenberg N. A., Donnelly P.. **Association mapping in structured populations**. *American Journal of Human Genetics* (2000) **67** 170-181. PMID: 10827107
66. Rosenberg N. A., Nordborg M.. **A general population-genetic model for the production by population structure of spurious genotype–phenotype associations in discrete, admixed or spatially distributed populations**. *Genetics* (2006) **173** 1665-1678. PMID: 16582435
67. Sanjak J. S., Sidorenko J., Robinson M. R., Thornton K. R., Visscher P. M.. **Evidence of directional and stabilizing selection in contemporary humans**. *Proceedings of the National Academy of Sciences* (2018) **115** 151-156
68. Sella G., Barton N. H.. **Thinking about the evolution of complex traits in the era of genome-wide association studies**. *Annual Review of Genomics and Human Genetics* (2019) **20** 461-493
69. Selzam S., Ritchie S. J., Pingault J.-B., Reynolds C. A., O’Reilly P. F., Plomin R.. **Comparing within-and between-family polygenic score prediction**. *American Journal of Human Genetics* (2019) **105** 351-363. PMID: 31303263
70. Shen H., Feldman M. W.. **Genetic nurturing, missing heritability, and causal analysis in genetic statistics**. *Proceedings of the National Academy of Sciences* (2020) **117** 25646-25654
71. Simons Y. B., Mostafavi H., Smith C. J., Pritchard J. K., Sella G.. **Simple scaling laws control the genetic architectures of human complex traits**. *bioRxiv* (2022). DOI: 10.1101/2022.10.04.509926
72. Sl oczyński T.. **Interpreting OLS estimands when treatment effects are heterogeneous: Smaller groups get larger weights**. *Review of Economics and Statistics* (2022) **104** 501-509
73. Sohail M., Maier R. M., Ganna A., Bloemendal A., Martin A. R., Turchin M. C., Chiang C. W., Hirschhorn J., Daly M. J., Patterson N.. **Polygenic adaptation on height is overestimated due to uncorrected stratification in genome-wide association studies**. *eLife* (2019) **8** e39702. PMID: 30895926
74. Spielman R. S., McGinnis R. E., Ewens W. J.. **Transmission test for linkage disequilibrium: the insulin gene region and insulin-dependent diabetes mellitus (IDDM)**. *American Journal of Human Genetics* (1993) **52** 506-516. PMID: 8447318
75. Stulp G., Simons M. J. P., Grasman S., Pollet T. V.. **Assortative mating for human height: A meta-analysis**. *American Journal of Human Biology* (2017) **29** e22917. PMID: 27637175
76. Trejo S., Domingue B. W.. **Genetic nature or genetic nurture? Introducing social genetic parameters to quantify bias in polygenic score analyses**. *Biodemography and Social Biology* (2018) **64** 187-215. PMID: 31852332
77. Tropf F. C., Lee S. H., Verweij R. M., Stulp G., Van Der Most P. J., De Vlaming R., Bakshi A., Briley D. A., Rahal C., Hellpap R.. **Hidden heritability due to heterogeneity across seven populations**. *Nature Human Behaviour* (2017) **1** 757-765
78. Tyrrell J., Zheng J., Beaumont R., Hinton K., Richardson T. G., Wood A. R., Davey Smith G., Frayling T. M., Tilling K.. **Genetic predictors of participation in optional components of UK Biobank**. *Nature Communications* (2021) **12** 886
79. Ulizzi L., Terrenato L.. **Natural selection associated with birth weight v. the secular relaxation of the stabilizing component**. *Annals of Human Genetics* (1987) **51** 205-210. PMID: 3688835
80. Veller C., Muralidhar P., Haig D.. **On the logic of Fisherian sexual selection**. *Evolution* (2020) **74** 1234-1245. PMID: 32128812
81. Vilhjálmsson B. J., Nordborg M.. **The nature of confounding in genome-wide association studies**. *Nature Reviews Genetics* (2013) **14** 1-2
82. Visscher P. M., Medland S. E., Ferreira M. A. R., Morley K. I., Zhu G., Cornes B. K., Montgomery G. W., Martin N. G.. **Assumption-free estimation of heritability from genome-wide identity-by-descent sharing between full siblings**. *PLoS Genetics* (2006) **2** e41. PMID: 16565746
83. Weiner D. J., Wigdor E. M., Ripke S., Walters R. K., Kosmicki J. A., Grove J., Samocha K. E., Goldstein J. I., Okbay A., Bybjerg-Grauholm J.. **Polygenic transmission disequilibrium confirms that common and rare variation act additively to create risk for autism spectrum disorders**. *Nature Genetics* (2017) **49** 978-985. PMID: 28504703
84. Weir B. S.. **Linkage disequilibrium and association mapping**. *Annual Review of Genomics and Human Genetics* (2008) **9** 129-142
85. Wolf J. B., Brodie III E. D., Cheverud J. M., Moore A. J., Wade M. J.. **Evolutionary consequences of indirect genetic effects**. *Trends in Ecology & Evolution* (1998) **13** 64-69. PMID: 21238202
86. Wright S.. **Systems of mating. III. Assortative mating based on somatic resemblance**. *Genetics* (1921) **6** 144-161. PMID: 17245960
87. Yair S., Coop G.. **Population differentiation of polygenic score predictions under stabilizing selection**. *Philosophical Transactions of the Royal Society B* (2022) **377** 20200416
88. Yamamoto K., Sonehara K., Namba S., Konuma T., Masuko H., Miyawaki S., Kamatani Y., Hizawa N., Ozono K., Yengo L.. **Genetic footprints of assortative mating in the Japanese population**. *Nature Human Behaviour* (2023) **7** 65-73
89. Yang J., Zaitlen N. A., Goddard M. E., Visscher P. M., Price A. L.. **Advantages and pitfalls in the application of mixed-model association methods**. *Nature Genetics* (2014) **46** 100-106. PMID: 24473328
90. Yengo L., Robinson M. R., Keller M. C., Kemper K. E., Yang Y., Trzaskowski M., Gratten J., Turley P., Cesarini D., Benjamin D. J.. **Imprint of assortative mating on the human genome**. *Nature Human Behaviour* (2018) **2** 948-954
91. Young A. I., Benonisdottir S., Przeworski M., Kong A.. **Deconstructing the sources of genotype-phenotype associations in humans**. *Science* (2019) **365** 1396-1400. PMID: 31604265
92. Young A. I., Frigge M. L., Gudbjartsson D. F., Thorleifsson G., Bjornsdottir G., Sulem P., Masson G., Thorsteinsdottir U., Stefansson K., Kong A.. **Relatedness disequilibrium regression estimates heritability without environmental bias**. *Nature Genetics* (2018a) **50** 1304-1310. PMID: 30104764
93. Young A. I., Nehzati S. M., Benonisdottir S., Okbay A., Jayashankar H., Chanwook L., Cesarini D., Benjamin D. J., Turley P., Kong A.. **Mendelian imputation of parental genotypes improves estimates of direct genetic effects**. *Nature Genetics* (2022) **54** 897-905. PMID: 35681053
94. Young A. I., Wauthier F. L., Donnelly P.. **Identifying loci affecting trait variability and detecting interactions in genome-wide association studies**. *Nature Genetics* (2018b) **50** 1608-1614. PMID: 30323177
95. Zaidi A. A., Mathieson I.. **Demographic history mediates the effect of stratification on polygenic scores**. *eLife* (2020) **9** e61548. PMID: 33200985
96. Zaitlen N., Huntsman S., Hu D., Spear M., Eng C., Oh S. S., White M. J., Mak A., Davis A., Meade K.. **The effects of migration and assortative mating on admixture linkage disequilibrium**. *Genetics* (2017) **205** 375-383. PMID: 27879348
97. Zaitlen N., Pasaniuc B., Sankararaman S., Bhatia G., Zhang J., Gusev A., Young T., Tandon A., Pollack S., Vilhjálmsson B. J.. **Leveraging population admixture to characterize the heritability of complex traits**. *Nature Genetics* (2014) **46** 1356-1362. PMID: 25383972
98. Zhu C., Ming M. J., Cole J. M., Kirkpatrick M., Harpak A.. **Amplification is the primary mode of gene-by-sex interaction in complex human traits**. *bioRxiv* (2022). DOI: 10.1101/2022.05.06.490973
|
---
title: Microbial community-scale metabolic modeling predicts personalized short chain
fatty acid production profiles in the human gut
authors:
- Nick Quinn-Bohmann
- Tomasz Wilmanski
- Katherine Ramos Sarmiento
- Lisa Levy
- Johanna W. Lampe
- Thomas Gurry
- Noa Rappaport
- Erin M. Ostrem
- Ophelia S. Venturelli
- Christian Diener
- Sean M. Gibbons
journal: bioRxiv
year: 2023
pmcid: PMC10002715
doi: 10.1101/2023.02.28.530516
license: CC BY 4.0
---
# Microbial community-scale metabolic modeling predicts personalized short chain fatty acid production profiles in the human gut
## Abstract
Microbially-derived short chain fatty acids (SCFAs) in the human gut are tightly coupled to host metabolism, immune regulation, and integrity of the intestinal epithelium. However, the production of SCFAs can vary widely between individuals consuming the same diet, with lower levels often associated with disease. A systems-scale mechanistic understanding of this heterogeneity is lacking. We present a microbial community-scale metabolic modeling (MCMM) approach to predict individual-specific SCFA production profiles. We assess the quantitative accuracy of our MCMMs using in vitro, ex vivo, and in vivo data. Next, we show how MCMM SCFA predictions are significantly associated with blood-derived clinical chemistries, including cardiometabolic and immunological health markers, across a large human cohort. Finally, we demonstrate how MCMMs can be leveraged to design personalized dietary, prebiotic, and probiotic interventions that optimize SCFA production in the gut. Our results represent an important advance in engineering gut microbiome functional outputs for precision health and nutrition.
## Introduction
The human gut microbiota serves many functions: maintaining intestinal barrier function, regulating peripheral and systemic inflammation, and breaking down indigestible dietary components and host substrates into a wide range of bioactive compounds1,2. One of the primary mechanisms by which the gut microbiota impacts human health is through the production of small molecules that enter the circulation and are absorbed and transformed by host tissues3–5. Approximately half of the metabolites detected in human blood are significantly associated with cross-sectional variation in gut microbiome composition6.
Short chain fatty acids (SCFAs) are among the most abundant metabolic byproducts produced by the gut microbiota, largely through the fermentation of indigestible dietary fibers and resistant starches, with acetate, propionate and butyrate being the most abundant SCFAs7–9. Deficits in SCFA production, specifically butyrate and propionate, have been repeatedly associated with disease, including inflammatory bowel disease and colorectal cancer10–15. Therefore, SCFA production is a crucial ecosystem service that the gut microbiota provides to its host, with extensive impacts on health1,11,16,17. However, different human gut microbiota provided with identical dietary substrates can show variable SCFA production profiles18,19, and predicting this heterogeneity remains a fundamental challenge to the microbiome field. Measuring SCFA abundances in blood or feces is rarely informative of in situ production rates, due to the volatility of SCFAs, cross-feeding among microbes, and the rapid consumption and transformation of these metabolites by the colonic epithelium10,20,21. Furthermore, SCFA production fluxes (i.e., the amount of a metabolite produced over a given period of time) within an individual can vary longitudinally, depending upon dietary inputs and the availability of host substrates22. In order to account for this inter- and intra-individual heterogeneity, we propose the use of microbial community-scale metabolic models (MCMMs), which mechanistically account for metabolic interactions between gut microbes, host substrates, and dietary inputs, to estimate personalized, context-specific SCFA production profiles.
Statistical modeling and machine-learning approaches for predicting metabolic output from the microbiome have shown promising results in recent years. For example, postprandial blood glucose responses can be predicted by machine-learning algorithms trained on large human cohorts23,24. Nevertheless, machine-learning methods are limited by the measurements and interventions represented within the training data25. Mechanistic models like MCMMs, on the other hand, do not rely on training data and can provide causal insights21. Metabolic modeling of individual commensal taxa has been used to predict plasma concentrations of microbially derived metabolites26, but these methods have not been extended to diverse, real-world microbiomes. MCMMs can be constructed using existing knowledge bases, including curated genome-scale metabolic models (GEMs) of individual taxa27. MCMMs are limited by the availability of well-curated GEMs for abundant taxa present within every individual in a population and by information on individual-specific dietary variation. These limitations are further exacerbated in human populations that are generally underrepresented in microbiome research, where our databases are also less representative28. However, as our knowledge bases grow, so too will the power and scope of MCMMs. Overall, MCMMs have the potential to serve as powerful, transparent, knowledge-driven tools for predicting community-specific responses to a wide array of interventions or perturbations.
Here, we demonstrate the utility of MCMMs for the prediction of personalized SCFA production profiles in the context of different dietary, prebiotic, and probiotic inputs. We first validate our modeling platform using diverse synthetic in vitro gut microbial communities ($$n = 1$$,387) and ex vivo stool incubation assays ($$n = 29$$). Next, we investigate the relevance of this modeling strategy in vivo using data from a 10-week high-fiber dietary intervention cohort ($$n = 18$$), where individuals showed a variety of immune responses. We assess the clinical significance of these precision SCFA predictions by looking at associations between predicted SCFA production on an average European diet and a panel of blood-based clinical lab tests in a large human cohort ($$n = 2$$,687). Finally, we demonstrate the potential power of MCMMs in designing personalized prebiotic, probiotic, and dietary interventions that optimize predictions for individual-specific butyrate production rates.
## MCMMs capture SCFA production rates in vitro
Details on the origin and composition of each dataset used in these analyses can be found in the supplement (Table S1).
We sought to investigate whether MCMMs can predict production rates of the major SCFAs (i.e., acetate, propionate, and butyrate) under controlled experimental conditions (Fig. 1). Growth media, matching the environmental context of each experiment, were constructed and applied as bounds on metabolic import to MCMMs (Fig. 1A), which were concurrently constructed by combining manually-curated GEMs from the AGORA database29 using MICOM21, constraining taxon abundances using 16S amplicon or shotgun metagenomic sequencing relative abundance estimates (Fig. 1B). Sample-specific metabolic models were then solved using cooperative tradeoff flux balance analysis (ctFBA), a previously-reported two-step quadratic optimization strategy that yields empirically-validated estimates of the steady state growth rates and metabolic uptake and secretion fluxes for each taxon in the model21 (Fig. 1C, see Materials and Methods). Models constructed from 16S amplicon sequencing data were summarized at the genus level, which was the finest level of phylogenetic resolution that the data allowed for. When shotgun metagenomic sequencing data were available, models were constructed at the species level. Models constructed from both 16S and shotgun metagenomic data at the species and genus levels showed highly consistent results (Fig. S1). Measured SCFA production profiles from synthetic in vitro community and stool ex vivo experiments (Fig. 1D) were compared to paired SCFA flux predictions from MCMMs to validate the accuracy of the models.
First, we looked at published data from synthetically constructed communities of bacterial commensals isolated from the human gut30. This data set included endpoint measurements of relative microbial abundances, derived from 16S amplicon sequencing, measured endpoint butyrate concentrations, and the overall optical density for each of 1,387 independent co-cultures (Fig. 2A). Cultures varied in richness from 1–25 strains. MCMMs were constructed for each co-culture as described above, simulating growth of each of the models using a defined medium mapped to a database of metabolic constituents, matching the composition of the medium used in the in vitro experiments (see Materials and Methods). Model-predicted butyrate fluxes were compared with calculated butyrate production rates (endpoint butyrate divided by culturing time, assuming no butyrate at the start of growth, normalized to total biomass using OD600), stratifying results into low richness (1–5 genera) and high richness (10–25 genera) communities. Model predictions for butyrate production fluxes were significantly correlated with measured butyrate production fluxes across all communities (Pearson’s correlation; Low Richness: $r = 0.17$, $p \leq 0.001$; High Richness: $r = 0.53$, $p \leq 0.001$), but the predictions were more accurate in the higher richness communities (Fig. 2B–C).
Next, we compared MCMM predictions to anaerobic ex vivo incubations of human stool samples from a small number of individuals ($$n = 29$$), cultured after supplementation with sterile PBS buffer or with different dietary fibers across four independent studies. Study A contained samples from two donors cultured for 7 hours with a final dilution of 1:5, Study B18 contained samples from 10 donors cultured for 24 hours diluted 1:19, Study C contained samples from 8 donors cultured for 4 hours diluted 1:5, and Study D contained samples from 9 donors cultured for 6 hours diluted 1:3. Fecal ex vivo assays allow for the direct measurement of bacterial SCFA production fluxes without interference from the host. For all three studies, ex vivo incubations were performed by homogenizing fecal material in sterile buffer under anaerobic conditions, adding control or fiber interventions to replicate fecal slurries, and measuring the resulting SCFA production rates in vitro at 37°C (see Materials and Methods). Metagenomic (Studies A, C and D) or 16S amplicon (Study B) sequencing data from these ex vivo cultures were used to construct MCMMs, using relative abundances as a proxy for relative biomass for each bacterial taxon (see Materials and Methods). MCMMs were simulated using a diluted standardized European diet (i.e., to approximate residual dietary substrates still present in the stool slurry), with or without specific fiber amendments, matching the experimental treatments (see Material and Methods). Within studies, the divergence in measured SCFA production between control samples and fiber-treated samples seemed to be highly dependent upon the final dilution of the ex vivo cultures (Fig. S2). This was accounted for by matching the dilution of residual fiber (starch, cellulose and dextrin) in the medium used for growth simulation to the corresponding study. For instance, Study A was diluted 1:5, so the residual fiber in the medium used to simulate growth for these samples was diluted by a factor of 5. The resulting SCFA flux predictions were then compared to the measured fluxes. MCMM fluxes are given in units of mmol/gDW/h, while measured production fluxes are given in mmol/L/h. Without knowledge of the live-cell biomass within the fecal homogenates, it was not possible to normalize the units across the two axes, but the predicted and measured values were expected to be proportional. To overcome study-specific differences in protocols and allow for comparison of results across studies, we Z-scored both measured and predicted SCFA production fluxes within each data set (Fig. 2D–F). We observed a similar degree of agreement between MCMM-predicted and measured production fluxes for butyrate and propionate across all four ex vivo data sets (Fig. 2E–F). The model was notably less capable of accurately predicting differences in acetate production between individuals, with no significant association seen (Fig. 2–3). Significant agreement was observed between measured and predicted production fluxes of butyrate and propionate within each individual data set ($r = 0.41$–0.97, Pearson test, $p \leq 0.05$) with the exception of propionate in Study A, which had a very limited sample size ($$n = 2$$) (Fig. 3E–L). Notably, the correlation coefficient (Pearson r) for these associations was similar to that seen in the high-richness in vitro cultures (Fig. 2C). As previously seen, the prediction of acetate was worse, most notably in studies C and D, where no significant prediction was observed. In studies A and B, acetate production was more readily predicted, likely due to a strong treatment-effect (Fig. 3A–D). Within treatment groups, similar correlations were observed, though statistical power was severely limited by the smaller sample sizes (Table S2). Predictions from models built with shotgun metagenomic sequencing data showed slightly better results when constructed at the species level, as compared to building at the genus level (Fig. S3). To test whether SCFA production was impacted by sample diversity, we compared measured butyrate and propionate against Shannon index for each sample in each study (Fig. S4). A weak significant signal was seen in only one of the four studies (Study D). In summary, we observed agreement between MCMM predicted and measured in vitro production rates of butyrate and propionate in the presence or absence of fiber supplementation, with better agreement in more diverse communities and over longer experimental incubation times (Fig. 2–3). As acetate was not well predicted by the MCMMs (i.e., acetate was not strongly coupled to biomass production, and predictions could vary widely for the same biomass optimum), we focused our downstream predictions and analyses on the SCFAs butyrate and propionate.
## MCMM predictions correspond with variable immunological responses to a 10-week high-fiber dietary intervention
We next investigated whether MCMM-predicted SCFA production rates could be leveraged to help explain inter-individual differences in phenotypic response following a dietary intervention. Specifically, we looked at data from 18 individuals who were placed on a high-fiber diet for ten weeks31. These individuals fell into three distinct immunological response groups: one in which high inflammation was observed over the course of the intervention (high-inflammation group), and two other distinct response groups that both exhibited lower levels of inflammation (low-inflammation groups I and II; Fig. 4A). We hypothesized that these immune response groups could be explained, in part, by differences in MCMM-predicted production rates of anti-inflammatory SCFAs. Using 16S amplicon sequencing data from seven time points collected from each of these 18 individuals over the 10-week intervention, we built MCMMs for each study participant at each time point. Growth was then simulated for each model using a standardized high-fiber diet, rich in resistant starch (see Material and Methods). Throughout the study, a trend of decreasing propionate production was observed in high-inflammation individuals ($r = 0.39$, Pearson test, $$p \leq 0.019$$), showing less production as the intervention went on, despite the high fiber content of the diets consumed by participants (Fig. 4B). Individuals in the high-inflammation group showed significantly lower predicted propionate production, on average, compared to the individuals in each of the low-inflammation groups (High vs. Low I: 131.9 ± 5.8 vs 158.1 ± 5.7 mmol/(gDW h), Mann-Whitney $$p \leq 0.0053$$; High vs. Low II: 131.9 ± 5.8 vs 163.08.3 ± 6.5 mmol/(gDW h), Mann-Whitney $$p \leq 0.0017$$; Fig. 4C). Butyrate showed no such significant effects across immune response groups (Fig. 4D, 4E). To investigate whether sample alpha-diversity was sufficient to explain the differences between the immune response groups, we calculated the alpha diversity for each sample at each timepoint during the study. Across all seven time points tested, only one significant difference in alpha diversity was seen, between the two low inflammation groups at time point 2 (Mann-Whitney U-test, $p \leq 0.05$), leading us to determine that differences in SCFA production throughout the intervention were not the result of differences in diversity. ( Fig. S4).
## MCMM-predicted SCFA profiles are associated with a wide range of blood-based clinical markers
To further evaluate the clinical relevance of personalized MCMMs, we generated SCFA production rate predictions from stool 16S amplicon sequencing data for 2,687 individuals in a deeply phenotyped, generally-healthy cohort from the West Coast of the United States (i.e., the Arivale cohort)32. Baseline MCMMs were built for each individual assuming the same dietary input (i.e., an average European diet) in order to compare SCFA production rate differences, independent of background dietary variation. MCMM-predicted SCFA fluxes were then regressed against a panel of 128 clinical chemistries and health metrics collected from each individual, adjusting for a standard set of common covariates (i.e., age, sex, and microbiome sequencing vendor; Fig. 5A). After FDR correction, 20 markers were significantly associated with the predicted production rate of butyrate (Fig. 5B). Predicted butyrate production showed significant positive associations with only 3 markers, including the health-associated hormone adiponectin, and significant negative associations with 17 markers linked to disease, including C-reactive protein (CRP), low-density lipoprotein (LDL), and blood pressure (mean arterial pressure; $P \leq 0.05$, FDR-corrected t-test). Propionate showed no significant associations after covariate adjustment and multiple comparison correction (Fig. 5B). Total combined propionate and butyrate production was significantly associated with 16 clinical markers, all overlapping with those associated with butyrate. Predicted butyrate production was significantly negatively associated with BMI (β= −0.10, t-test, $p \leq 0.001$), while propionate was not (Fig. 5 C–D). Covariate-adjusted p-values and beta coefficients for all clinical markers included in the analysis can be found in the supplementary material (Table S3).
## Leveraging MCMMs to design precision dietary, prebiotic, and probiotic interventions
As a proof-of-concept for in silico engineering of the metabolic outputs of the human gut microbiome, we screened a set of potential interventions designed to increase SCFA production for individuals from the Arivale cohort (Fig. 6A). MCMMs were built using two different dietary contexts: an average European diet, and a vegan, high-fiber diet rich in resistant starch (see Material and Methods). As expected, models grown on a high-fiber diet showed higher average predicted butyrate production: 87.78 ± 0.67 mmol/(gDW h) vs 16.29 ± 0.13 mmol/(gDW h), t-test, $t = 104.3$, $p \leq 0.001$ (Fig. 6B). However, this increase in butyrate production between the European and high-fiber diets was not uniform across individuals. On the high-fiber diet, some individual gut microbiota compositions showed very large increases in butyrate production, some showed little-to-no change, and a small subset of samples actually showed a decrease in butyrate production, relative to the European diet. We identified a set of ‘non-responders’ ($$n = 9$$) who produced less than 15 of butyrate on the European diet and showed an increase in butyrate production of less than $20\%$ on the high-fiber diet (Fig. 6C). We also identified a set of ‘regressors’ ($$n = 7$$) who showed decreased butyrate production on the high-fiber diet when compared to the European diet (Fig. 6D). We then simulated a handful of simple prebiotic and probiotic interventions across these individuals, to identify optimal combinatorial interventions for each individual (Fig. 6C–E). MCMMs for each subset of individuals were simulated with prebiotic and probiotic interventions in the context of either the European or the high-fiber diet. Specifically, diets were supplemented with the dietary fiber inulin, with the dietary fiber pectin, or with a simulated probiotic intervention that consisted of introducing $10\%$ relative abundance of the butyrate-producing genus Faecalibacterium to the MCMM. *In* general, optimal combinatorial interventions significantly increased the population-level butyrate production well above either dietary intervention alone (Fig. 6C–D).
For $\frac{15}{16}$ individuals in the regressors or non-responders groups, supplementation of the background diet with a specific prebiotic or probiotic increased the butyrate production rate (Fig. 6C–E). For both regressors and non-responders, the optimal intervention showed substantial increases over the standard European diet (+290±$80\%$ for non-responders; +239±$102\%$ for regressors). The exact intervention that yielded the highest butyrate production for any given individual across both populations varied widely (Fig. 6E). For example, the probiotic intervention was more successful in raising predictions for butyrate production in non-responders than it was in regressors (Fig. 6E). Overall, no single combinatorial intervention was optimal for every individual in the population.
## Discussion
The objective of this study was to experimentally validate personalized MCMM SCFA predictions. Predictions of butyrate production in synthetically constructed in vitro co-cultures showed significant agreement between measured and predicted butyrate fluxes (Fig. 2), a first step toward validation. Interestingly, better agreement was observed in richer communities, indicating increased model complexity benefitted predictions. Decreasing accuracy of butyrate predictions as community richness declined may reflect a limitation of building models at the genus-level, as reconstructions contain a summarized aggregation of the metabolic capability of the genus as a whole, without species- or strain-level resolution. Furthermore, we are leveraging database models, which do not reflect the exact strains present in a given sample. Consequently, pathways included in the metabolic models are not a perfect match to the reality of what is present in a sample. In high richness models, predictions of SCFAs became more accurate, suggesting this mismatch gets averaged out as species richness increases, likely due to functional redundancies across organisms that can mask the inaccuracies of any single taxon model. Alternatively, there could be some unknown biological reason for why SCFA production is less variable in higher richness communities, which would affect our ability to make accurate MCMM predictions. Overall, the observed increase in accuracy with community diversity benefits modeling of real-world microbiomes, which are often more species-rich than synthetic in vitro communities33,34. As our model databases grow to better-reflect the metabolic diversity of real-world ecosystems, we expect MCMMs to become more and more accurate, independent of community diversity.
Further validation of MCMM predictions was observed from ex vivo anaerobic fecal incubations. We saw good agreement between SCFA flux predictions and measurements, especially for butyrate and propionate, across four independent studies (Fig. 3). Acetate is known to act as an overflow metabolite35,36, with a wide range of possible fluxes for a given biomass optimum, so it is perhaps not surprising that the predictions for this metabolite tended to be less accurate across studies and within treatment groups. Butyrate and propionate, however, showed a narrower range of possible fluxes for a given biomass optimum, suggesting that the production of these molecules is more strongly coupled to biomass production. The dilution level of the ex vivo stool incubations had a sizable effect on the results, where the in vitro prebiotic treatment effect was dampened in less dilute fecal homogenates, likely due to the presence of residual dietary fibers in stool. The more accurate predictions of acetate production in the more dilute fecal homogenates is likely due to the fact that total SCFA production was more strongly coupled to in vitro prebiotic treatment in these samples. Accounting for this dilution factor in the construction of the in silico media improves predictions and returns more accurate results for butyrate and propionate production.
We were interested in seeing how 16S- and metagenomic-based models compared at a similar taxonomic level, and how genus and species level predictions compared, in order to assess how applicable our modeling strategy could be to different data types. Using paired 16S and shotgun metagenomic sequencing data from Study C, we saw strong agreement between models constructed at the genus level for both 16S and metagenomic data (Fig. S1). Furthermore, we saw robust agreement between predictions at the genus and species levels across metagenomic data sets (Fig. S5). Interestingly, predictions from Studies A, C and D showed marginally better agreement with measured values when constructed at the species level vs. the genus level, indicating that higher specificity in model construction is desirable when possible (Fig. S5). Across the in vitro and ex vivo studies, our results strongly support the use of MCMMs for predicting personalized butyrate and propionate production rates in response to prebiotic, probiotic, and dietary interventions.
In vivo validation via direct measurement of SCFA production is not easily accomplished, due to the rapid consumption of these metabolites by the colonic epithelium and noisy measurements in either stool or serum37 38. However, higher SCFA production rates are known to influence the phenotype of the host in a number of ways, including a reduction in systemic inflammation and improvements in cardiometabolic health17,22,39,40. Wastyk et al. found that among 18 individuals given a 10-week high fiber dietary intervention, one third showed an increase in inflammation over the course of the intervention and two thirds showed a decline in systemic markers of inflammation31. In the original paper, there was no clear mechanism for explaining these variable immune response groups31. We found that propionate production, a strong inhibitor of inflammation through activation of FFA2 and FFA341,42, was predicted to be significantly lower in individuals who showed the high inflammation response (Fig. 4B–C)31. While it is impossible to say whether or not our propionate flux predictions are underlying these dietary response phenotypes, the observed immune response groups and propionate production fluxes could not be explained by differences in alpha-diversity between groups (Fig. S4). We also had access to blood-based clinical labs and microbiome data for a cohort of 2,687 Americans. We built MCMMs for this cohort, assuming a standard European diet, and predicted butyrate and propionate production. We found that butyrate was negatively associated with systemic inflammation, LDL cholesterol, and insulin resistance, blood pressure, and BMI (Fig. 5). These results are consistent with what is known about how butyrate is protective against inflammation, cardiovascular disease, obesity, and metabolic syndrome17,22,39,40,43 (Fig. 5B), and they provide us with further confidence in the predictive power of our MCMM approach. Dietary interventions have long been known to elicit variable responses, but a mechanistic framework for predicting this microbiome-mediated heterogeneity has not been available until now.
Given this set of promising associations between SCFA predictions and host phenotypic variation, we sought to demonstrate the potential for leveraging MCMMs for designing precision prebiotic, probiotic, and dietary interventions. Using the Arivale cohort, we identified two classes of individuals that responded differently to an in silico high-fiber dietary intervention: non-responders and regressors (Fig. 6). We designed combinatorial interventions that added either a prebiotic or a probiotic to the background diets, to see if we could rescue these non-responder and regressor phenotypes. We found significant heterogeneity in which combinatorial intervention was optimal across individuals from each of these response groups (Fig. 6E). Given that the non-responders had low baseline levels of butyrate production to begin with and did not respond to a high-fiber diet, this result underscores the importance of personalized predictions for those who tend not to respond well to population-scale interventions. These results also suggest that butyrate production in some individuals is limited by composition of the microbiota, indicating that probiotic interventions would be necessary to induce meaningful increases in production.
This study had several limitations that should be considered. First, we were limited by the availability of high-quality fluxomic data sets for model validation. For example, we had limited sample sizes in the ex vivo fecal studies presented above, due to the cost and difficulty of generating these kinds of data for larger cohorts. Additionally, the human cohort data presented here only provided indirect support for our MCMM predictions (Figs. 4–5). Second, predictions are dependent on the availability of GEMs. Obtaining large numbers of GEMs that faithfully recapitulate the full metabolic capacities of each organism in a sample is a challenging task. We used the publicly available AGORA model database29. While AGORA models have gone through some degree of manual curation, many of these models are not fully validated and have been shown to include infeasible and missing reactions44. Nevertheless, these GEMs appear to work well in the context of butyrate and propionate flux predictions. SCFA production pathways are fairly phylogenetically conserved and adjacent to central metabolism, so we might expect these reactions to be robust to strain- or species-level variation and variation in model quality. However, predictions for metabolites that are peripheral to central metabolism will likely be much noisier in the absence of well-curated models that closely match the organisms within a given sample. Third, model building is dependent on accurate taxonomic assignment of sequencing reads. For 16S amplicon sequencing, reads can only be confidently assigned at the genus level, limiting the specificity of a model to the genera present in the original samples. However, as model databases grow and shotgun metagenomic sequencing becomes more common, we anticipate this limitation will be resolved. Finally, the lack of individual-specific dietary constraints limits the accuracy of our predictions. For ex vivo fecal fermentations, as well as in vivo analysis, participant dietary information was not available, and so a standard European diet was used across all models. Detailed knowledge of dietary intake should increase the accuracy of MCMM predictions. Despite these limitations, MCMMs were able to explain 25–$35\%$ of the variance in butyrate and propionate production across individuals, and we expect that advances in model curation, pathway annotation, and personalized dietary constraints will only improve upon the accuracy of this approach over time.
## Conclusion
Here we present an approach for the rational prediction of personalized SCFA production rates from the human gut microbiome, validated using in vitro, ex vivo and in vivo experimental data. Additional analysis demonstrated a clear relationship between SCFA predictions and physiological responses in the host, including lower inflammation and improved cardiometabolic health. SCFA predictions were also significantly associated with variable immune responses to a high fiber dietary intervention. Finally, we showed how MCMMs could be used to rapidly design and test combinatorial prebiotic, probiotic and dietary interventions in silico for a large human population. Personalized prediction of SCFA production profiles from human gut MCMMs represents an important technological step forward in leveraging computational systems biology for precision nutrition. Mechanistic modeling allowed us to translate the ecological composition of the gut microbiome into concrete, individual-specific metabolic outputs, in response to particular interventions45,46. MCMMs are transparent models that do not require training data, with clear causal and mechanistic explanations behind each prediction. The clinical relevance of these predictions is evident, due to the widespread physiological effects of SCFAs on the human body47,48. A rational framework for engineering the production or consumption rates of these metabolites has broad potential applications in precision nutrition and personalized healthcare.
## In vitro culturing
Culturing of the synthetically assembled gut microbial communities is described in Clark et al., 202130. Culturing of ex vivo samples in Study A was done using the methodology described below. Culturing of ex vivo samples in Study B is described in Cantu-Jungles et al., 202118. Culturing of ex vivo samples in Study C was conducted by co-author Dr. Thomas Gurry, using the methodology described below.
## In vitro culturing of fecal-derived microbial communities (Study A)
Fecal samples were collected in 1200 mL 2-piece specimen collectors (Medline, USA) in the Public Health Science Division of the Fred Hutchinson Cancer Center (IRB Protocol number 5722) and transferred into an large vinyl anaerobic chamber (Coy, USA, 37°C, $5\%$ hydrogen, $20\%$ carbon dioxide, balanced with nitrogen) at the Institute for Systems Biology within 20 minutes of defecation. All further processing and incubation was then run inside the anaerobic chamber. 50 g of fecal material was transferred into sterile 50 oz Filter Whirl-Paks (Nasco, USA) with sterile PBS + $0.1\%$ L-cysteine at a 1:2.5 w/v ratio and homogenized with a Stomacher Biomaster (Seward, USA) for 15 minutes. After homogenization, each sample was transferred into three sterile 250 mL serum bottles and another 2.5 parts of PBS + $0.1\%$ L-cysteine was added to bring the final dilution to 1:5 in PBS. 87 ug/mL inulin or an equal volume of sterile PBS buffer were added to treatment or control bottles, respectively. Samples were immediately pipetted onto sterile round-bottom 2 mL 96-well plates in triplicates. Baseline samples were aliquoted into sterile 1.5 mL Eppendorf tubes and the plates were covered with Breathe-Easy films (USA Scientific Inc., USA). Plates were incubated for 7 h at 37°C and gently vortexed every hour within the chamber. Final samples were aliquoted into 1.5 mL Eppendorf tubes at the end of incubation. Baseline and 7 h samples were kept on ice and immediately processed after sampling. 500 uL of each sample were aliquoted for metagenomics and kept frozen at −80°C before and during transfer to the commercial sequencing service (Diversigen, Inc). The remaining sample was transferred to a table-top centrifuge (Fisher Scientific accuSpin, USA) and spun at 1,500 rpm for 10 minutes. The supernatant was then transferred to collection tubes kept on dry ice and transferred to the commercial metabolomics provider Metabolon, USA, for targeted SCFA quantification.
## In vitro culturing of fecal-derived microbial communities (Study C)
Homogenized fecal samples in this study again underwent anaerobic culturing at 37°C, as described above, but with a shorter culturing time of 4 hours. The slurry was diluted 2.5x in $0.1\%$ L-cysteine PBS buffer solution. Cultures were supplemented with the dietary fibers pectin or inulin to a final concentration of 10g/L, or a sterile PBS buffer control treatment. Aliquots were taken at 0h and 4h and further processed for measurement of SCFA concentrations, which were used to estimate experimental production flux (concentration[4h] - concentration[0h]/4h). SCFA concentrations were measured using GC-FID. Briefly, the pH of the aliquots was adjusted to 2–3 with $1\%$ aqueous sulfuric acid solution, after which they were vortexed for 10 minutes and centrifuged for 10 minutes at 10,000 rpm. 200 uL aliquots of clear supernatant were transferred to vials containing 200 uL of MeCN and 100 uL of a $0.1\%$ v/v 2-methyl pentanoic acid solution. The resulting solutions were analyzed by GC-FID on a Perkin Elmer Clarus 500 equipped with a DB-FFAP column (30m, 0.250mm diameter, 0.25um film) and a flame ionization detector.
## In vitro culturing of fecal-derived microbial communities (Study D)
Fecal samples were collected in 1200 mL 2-piece specimen collectors (Medline, USA) in the Public Health Science Division of the Fred Hutchinson Cancer Center (IRB Protocol number 10961) and transferred into a large vinyl anaerobic chamber (Coy, USA, 37°C, $5\%$ hydrogen, $20\%$ carbon dioxide, balanced with nitrogen) at the Institute for Systems Biology within 30 minutes of sample receipt. All further processing and incubation was then run inside the anaerobic chamber. 30 g of fecal material was transferred into sterile 50 oz Filter Whirl-Paks (Nasco, USA) with 90 mL sterile PBS + $0.1\%$ L-cysteine + $0.0001\%$ resazurin and homogenized with a Stomacher Biomaster (Seward, USA) for 5 minutes. For each individual fecal sample, triplicate baseline samples of 1500uL slurry were transferred to a deep 96-well place (Fisher Scientific, USA), sealed and centrifuged at 4000rpm for 10 minutes. 300uL of the supernatant were transferred to collection tubes and immediately frozen at −80°C. An additional 1800uL of fecal slurry was transferred into a 2mL Eppendorf tube and frozen at −80°C for metagenomic shotgun sequencing. Interventions of 100uL inulin at 625mg/L, pectin at 750mg/L or PBS were transferred to in duplicate to a new deep 96-well plate, topped with 1500uL fecal slurry each, and immediately sealed with Breathe-Easy films (USA Scientific Inc., USA). Plates were incubated for 6 h at 37°C and gently vortexed every 2 hours within the chamber. After incubation, plates were immediately centrifuged at 4000rpm for 10 minutes at room temperature and 300uL of the supernatant was again transferred to collection tubes and kept at −80°C. The frozen slurry sample for metagenomic shotgun sequencing was transferred to a commercial sequencing service (Diversigen, Inc) on dry ice. The remaining supernatant samples were kept on dry ice and transferred to the commercial metabolomics provider (Metabolon, USA) for targeted SCFA quantification.
## Metagenomic sequencing and analysis
For Study A, shallow metagenomic sequencing was performed by the sequencing vendor Diversigen, USA (i.e., their BoosterShot service). In brief, DNA was extracted from the fecal slurries with the DNeasy PowerSoil Pro Kit on a QiaCube HT (Qiagen, Germany) and quantified using the Qiant-iT Picogreen dsDNA Assay (Invitrogen, USA). Library preparation was performed with a proprietary protocol based on the Nextera Library Prep kit (Illumina, USA) and the generated libraries were sequenced on a NovaSeq (Illumina, USA) with a single-end 100bp protocol. Demultiplexing was performed using Illumina BaseSpace to generate the final FASTQ files used during analysis. For Study D, DNA extraction was performed under the same protocol as Study A. Libraries for Study D were prepared with the Nextera XT Library Prep kit (Illumina, USA) and sequenced with a paired-end 2×150bp protocol on a NovaSeq 6000 (Illumina, USA) yielding at least 70M reads/sample.
Preprocessing of raw sequencing reads from Study A and D was performed using FASTP49. The first 5bp on the 5’ end of each read were trimmed, and the 3’ end was trimmed using a sliding window quality filter that would trim the read as soon as the average window quality fell below 20. Reads containing ambiguous base calls or with a length of less than 15bp after trimming were removed from the analysis.
Bacterial species abundances were quantified using Kraken2 v2.0.8 and Bracken v2.2 using the Kraken2 default database which was based on Refseq release 94, retaining only those species with at least 10 assigned reads50,51. The analysis pipeline can be found at https://github.com/Gibbons-Lab/pipelines/tree/master/shallow_shotgun.
## Metabolomics
Targeted metabolomics were performed using Metabolon’s high-performance liquid chromatography (HPLC)–mass spectrometry (MS) platform, as described before52. In brief, fecal supernatants were thawed on ice, proteins were removed using aqueous methanol extraction, and organic solvents were removed with a TurboVap (Zymark, USA). Mass spectroscopy was performed using a Waters ACQUITY ultra-performance liquid chromatography (UPLC) and Thermo Scientific Q-Exactive high resolution/accuracy mass spectrometer interfaced with a heated electrospray ionization (HESI-II) source and an Orbitrap mass analyzer operated at 35,000 mass resolution. For targeted metabolomics ultra-pure standards of the desired short-chain fatty acids were used for absolute quantification. Fluxes for individual metabolites were estimated as the rate of change of individual metabolites during the incubation period (concentration[7h] - concentration[0h]/7h).
## Model Construction
Taxonomic abundance data inferred from 16S amplicon sequencing was summarized to the genus level (as in in vitro cultures, ex vivo study B, fiber intervention samples, and samples from the Arivale cohort), or to the species level when shotgun metagenomic sequencing was available (as in ex vivo studies A, C and D). Abundances were used to construct all MCMMs in this analysis using the community-scale metabolic modeling platform MICOM v0.32.521. Models were built using the MICOM build() function with a relative abundance threshold of 0.001, omitting taxa that made up less than $0.1\%$ relative abundance. The AGORA database (v1.03) of taxonomic reconstructions summarized to the genus level for 16S data or the species level for metagenomic sequencing data was used to collect genome-scale metabolic models for taxa present in each model. Building models at the genus level for metagenomic sequencing data was explored, but was outperformed by species level models. In silico media were applied to the grow() function, defining the bounds for metabolic imports by the MCMM. Medium composition varied between analyses (see Media Construction). Steady state growth rates and metabolic fluxes for all samples were then inferred using cooperative tradeoff flux balance analysis (ctFBA). In brief, this is a two-step optimization scheme, where the first step finds the maximal biomass production rate for the full microbial community and the second step infers taxon-specific growth rates and fluxes, while maintaining community growth within a suboptimal fraction of the theoretical maximum (i.e., the tradeoff parameter), thus balancing individual growth rates and the community-wide growth rate21. All models in the manuscript used a tradeoff parameter of 0.7. This parameter value was chosen through cooperative tradeoff analysis in MICOM. Multiple tradeoff parameters were tested, and the highest parameter value (i.e. the value closest to the maximal community growth rate at 1.0) that allowed most (>$90\%$) of taxa to grow was chosen (i.e., 0.7). Predicted growth rates from the simulation were analyzed to validate correct behavior of the models. All models were found to grow with minimum community growth rate of 0.3 h−1. Predicted values for export fluxes of SCFAs were collected from each MCMM using the production_rates() function, which calculates the overall production from the community that would be accessible to the colonic epithelium.
## Media Construction
Individual media were constructed based on the context of each individual analysis. For the synthetic in vitro cultures conducted by Clark et al. [ 2021], a defined medium (DM38) was used that supported growth of all taxa used in the experiments, excluding Faecalibacterium prausnitzii. Manually mapping each component to the Virtual Metabolic Human database, we constructed an in silico medium with flux bounds scaled to component concentration. All metabolites were found in the database. Using the MICOM fix_medium() function, a minimal set of metabolites necessary for all models to grow to a minimum community growth rate of 0.3 h−1 was added to the medium - here, only iron(III) was added (in silico medium available here: https://github.com/Gibbons-Lab/scfa_predictions/tree/main/media).
To mimic the medium used in ex vivo cultures of fecally-derived microbial communities, a carbon-stripped version of a standard European diet was used. First, a standard European diet was collected from the Virtual Metabolic Human database (www.vmh.life/#nutrition)53. Components in the medium which could be imported by the host, as defined by an existing uptake reaction in the Recon3D model54, were diluted to $20\%$ of their original flux, to adjust for absorption in the small intestine54. Additionally, host-supplied metabolites such as mucins and bile acids were added to the medium. The medium was augmented with a minimal set of metabolites required for growth of all taxa in the model database using the complete_db_medium() function within MICOM. As most carbon sources are consumed in the body and are likely not present in high concentrations in stool, this diet was then manually stripped of carbon sources by removing metabolites identified to be carbon sources for microbes. All components in the media were then diluted to $10\%$ of their original flux to mimic the fecal microenvironment. Residual dietary fiber not easily digested including resistant starch, dextrin and cellulose, was not removed from the medium during carbon removal. The amount of this residual fiber was scaled to the dilution factor of samples in each study prior to culturing. To simulate fiber supplementation, single fiber additions were made to the medium, either pectin, inulin or fructo-oligosaccharide (1.0 mmol/gDW*h for pectin, 10.0 mmol/gDW*h for inulin, 100 mmol/gDW*h for FOS, based on carbon content reported for each).
For in vivo modeling, two diets were used: a high-fiber diet containing high levels of resistant starch, and a standard European diet53,55. Again, both diets were collected from the Virtual Metabolic Human database (www.vmh.life/#nutrition). Each medium was subsequently adjusted to account for absorption in the small intestine by diluting metabolite flux as described previously. Additionally, host-supplied metabolites such as mucins and bile acids were added to the medium, to match the composition of the medium in vivo. Finally, the complete_db_medium() function was again used to augment the medium, as described above.
Prebiotic interventions were designed by supplementing the high-fiber or average European diet with single fiber additions, either pectin or inulin, as described previously.
## Probiotic Intervention
To model a probiotic intervention, $5\%$ relative abundance of the genus Faecalibacterium, a known butyrate-producing taxon56, was added to the MCMMs by adding a pan-genus model of the taxon derived from the AGORA database (v1.03). Measured taxonomic abundances were scaled to $95\%$ of their initial values, after which Faecalibacterium was artificially added to the model.
## External Data Collection
Data containing taxonomic abundance, optical density, and endpoint butyrate concentration for synthetically-constructed in vitro microbial cultures were collected from Clark et al. [ 2021]30. Endpoint taxonomic abundance data, calculated from fractional read counts collected via 16S amplicon sequencing, was used to construct individual MCMMs for each co-culture (see Model Construction). Resulting models ranged in taxonomic richness from 1 to 25 taxa.
Data from ex vivo studies A and D, containing shotgun metagenomic sequencing and SCFA metabolomics, was collected and processed as described previously. Taxonomic abundance data was used to construct MCMMs for each individual (see Model Construction).
From a study by Cantu-Jungles et al. [ 2021]18 (ex vivo Study B), preprocessed taxonomic abundance and SCFA metabolomics data was collected. Homogenized fecal samples in this study underwent a similar culturing process, with a culturing time of 24 hours. Cultures were supplemented with the dietary fiber pectin, or a PBS control. Initial and endpoint metabolomic SCFA measurements were used to estimate experimental production flux (concentration[24h] - concentration[0h]/24h). Taxonomic abundance data was used to construct MCMMs for each individual.
Data from a third (Study C) was collected from the Pharmaceutical Biochemistry Group at the University of Geneva, Switzerland, under study protocol 2019–00632, containing sequencing data in FASTQ format and targeted metabolomics SCFA measurements.
Data was collected from Wastyk, et al 202131, which provided 16S amplicon sequencing data at 9 timepoints spanning 14 weeks, along with immunological phenotyping, for 18 participants undergoing a high-fiber dietary intervention. Only 7 timepoints spanning 10 weeks were included in subsequent analysis, as the last 2 timepoints were taken after the conclusion of the dietary intervention. MCMMs were constructed for each participant at each timepoint at the genus level (see Model Construction). Mean total butyrate and propionate production were compared between immune response groups.
De-identified data was obtained from a former scientific wellness program run by Arivale, Inc. (Seattle, WA)32. Arivale closed its operations in 2019. Taxonomic abundances, inferred from 16S amplicon sequencing data, for 2,687 research-consenting individuals were collected and used to construct MCMMs. 128 paired blood-based clinical chemistries taken within 30 days of fecal sampling were also collected and used to find associations between MCMM SCFA predictions on a standard European diet and clinical markers. Blood pressure and BMI for each individual were also collected. Metadata for each sample including age, sex, and microbiome sequencing vendor were included in the analysis as confounders.
## Statistical analysis
Statistical analysis was performed using SciPy (v1.9.1) and statsmodels (v0.14.0) in Python (v3.8.13). Pearson correlation coefficients and p-values were calculated between measured and predicted SCFA production fluxes in in vitro and ex vivo cultures, as well as for predicted SCFA production fluxes across timepoints for an in vivo high-fiber intervention. Significance in SCFA production between immune response groups in the high-fiber dietary intervention was determined by non-parametric pairwise Mann-Whitney U test for butyrate, propionate, and combined butyrate and propionate production. Association of MCMM-predicted SCFA production flux with paired blood-based clinical labs was tested using OLS regression, adjusting for age, sex, microbiome sequencing vendor, and tested for significance by two-sided t-test. BMI was not included as a confounder in the analysis because it was itself negatively correlated with butyrate production43. Multiple comparison correction for p-values was done using the Benjamini–Yekutieli method for adjusting the False Discovery Rate (FDR)57. Comparison of butyrate production between dietary interventions was tested using paired Student’s t-tests. In all analyses, significance was considered at the $p \leq 0.05$ threshold.
## Funding
This research was funded by Washington Research Foundation Distinguished Investigator Award and by startup funds from the Institute for Systems Biology (to SMG). Fecal sample collection at Fred Hutchinson Cancer Center was supported by P30 CA015704. Research reported in this publication was supported by the National Institute of Diabetes and Digestive and Kidney Diseases of the National Institutes of Health (NIH) under award no. R01DK133468 (to SMG), by the Global Grants for Gut Health from Yakult and Nature Portfolio (to SMG), and by the National Institute on Aging of the National Institutes of Health (NIH) under award no. U19AG023122 (to NR).
## Data, Software, and Code Availability
Code used to run analysis and create figures for this manuscript can be found at https://github.com/Gibbons-Lab/scfa_predictions.
Processed data for synthetically constructed cultures can be found at https://github.com/RyanLincolnClark/DesignSyntheticGutMicrobiomeAssemblyFunction. Raw sequencing data can be found at https://doi.org/10.5281/zenodo.4642238.
Raw sequencing data for Study A can be found in the NCBI SRA under accession number PRJNA937304.
Processed data for ex vivo Study B can be found at https://github.com/ThaisaJungles/fiber_specificity. Raw sequencing data can be found in the NCBI SRA under accession number PRJNA640404.
Raw sequencing data for ex vivo Study C can be found in the NCBI SRA under accession number PRJNA939256.
Raw sequencing data for ex vivo Study D can be found in the NCBI SRA under accession number PRJNA1033794.
Processed data for the longitudinal high-fiber intervention study can be found at https://github.com/SonnenburgLab/fiber-fermented-study/.
Qualified researchers can access the full Arivale deidentified dataset supporting the findings in this study for research purposes through signing a Data Use Agreement (DUA). Inquiries to access the data can be made at [email protected] and will be responded to within 7 business days.
Illustrations were created with BioRender.com.
## References
1. Oliphant K., Allen-Vercoe E.. **Macronutrient metabolism by the human gut microbiome: major fermentation by-products and their impact on host health**. *Microbiome* (2019) **7** 91. PMID: 31196177
2. Rackerby B., Van De Grift D., Kim J. H., Park S. H.. **Effects of Diet on Human Gut Microbiome and Subsequent Influence on Host Physiology and Metabolism**. *Gut Microbiome and Its Impact on Health and Diseases* (2020) 63-84. DOI: 10.1007/978-3-030-47384-6_3
3. Tomasova L., Grman M., Ondrias K., Ufnal M.. **The impact of gut microbiota metabolites on cellular bioenergetics and cardiometabolic health**. *Nutr. Metab.* (2021) **18** 72
4. Glotfelty L. G., Wong A. C., Levy M.. **Small molecules, big effects: microbial metabolites in intestinal immunity**. *Am. J. Physiol. Gastrointest. Liver Physiol.* (2020) **318** G907-G911. PMID: 32249590
5. Donia M. S., Fischbach M. A.. **HUMAN MICROBIOTA. Small molecules from the human microbiota**. *Science* (2015) **349** 1254766. PMID: 26206939
6. Diener C.. **Genome-microbiome interplay provides insight into the determinants of the human blood metabolome**. *Nat Metab* (2022) **4** 1560-1572. PMID: 36357685
7. Ríos-Covián D.. **Intestinal Short Chain Fatty Acids and their Link with Diet and Human Health**. *Front. Microbiol.* (2016) **7** 185. PMID: 26925050
8. Nogal A., Valdes A. M., Menni C.. **The role of short-chain fatty acids in the interplay between gut microbiota and diet in cardio-metabolic health**. *Gut Microbes* (2021) **13** 1-24
9. Silva Y. P., Bernardi A., Frozza R. L.. **The Role of Short-Chain Fatty Acids From Gut Microbiota in Gut-Brain Communication**. *Frontiers in Endocrinology* (2020) **11**. DOI: 10.3389/fendo.2020.00025
10. Morrison D. J., Preston T.. **Formation of short chain fatty acids by the gut microbiota and their impact on human metabolism**. *Gut Microbes* (2016) **7** 189-200. PMID: 26963409
11. Cong J., Zhou P., Zhang R.. **Intestinal Microbiota-Derived Short Chain Fatty Acids in Host Health and Disease**. *Nutrients* (2022) **14**
12. Yang W.. **Intestinal microbiota-derived short-chain fatty acids regulation of immune cell IL-22 production and gut immunity**. *Nat. Commun.* (2020) **11** 4457. PMID: 32901017
13. Scheppach W.. **Effect of butyrate enemas on the colonic mucosa in distal ulcerative colitis**. *Gastroenterology* (1992) **103** 51-56. PMID: 1612357
14. Tang Y., Chen Y., Jiang H., Robbins G. T., Nie D.. **G-protein-coupled receptor for short-chain fatty acids suppresses colon cancer**. *Int. J. Cancer* (2011) **128** 847-856. PMID: 20979106
15. Singh N.. **Activation of Gpr109a, receptor for niacin and the commensal metabolite butyrate, suppresses colonic inflammation and carcinogenesis**. *Immunity* (2014) **40** 128-139. PMID: 24412617
16. Tan J.. **The role of short-chain fatty acids in health and disease**. *Adv. Immunol.* (2014) **121** 91-119. PMID: 24388214
17. Mortensen P. B., Clausen M. R.. **Short-chain fatty acids in the human colon: relation to gastrointestinal health and disease**. *Scand. J. Gastroenterol. Suppl.* (1996) **216** 132-148. PMID: 8726286
18. Cantu-Jungles T. M.. **Dietary Fiber Hierarchical Specificity: the Missing Link for Predictable and Strong Shifts in Gut Bacterial Communities**. *MBio* (2021) **12** e0102821. PMID: 34182773
19. Healey G. R., Murphy R., Brough L., Butts C. A., Coad J.. **Interindividual variability in gut microbiota and host response to dietary interventions**. *Nutr. Rev.* (2017) **75** 1059-1080. PMID: 29190368
20. Boets E.. **Quantification of in Vivo Colonic Short Chain Fatty Acid Production from Inulin**. *Nutrients* (2015) **7** 8916-8929. PMID: 26516911
21. Diener C., Gibbons S. M., Resendis-Antonio O.. **MICOM: Metagenome-Scale Modeling To Infer Metabolic Interactions in the Gut Microbiota**. *mSystems* (2020) **5**
22. van Deuren T., Blaak E. E., Canfora E. E.. **Butyrate to combat obesity and obesity-associated metabolic disorders: Current status and future implications for therapeutic use**. *Obes. Rev.* (2022) **23** e13498. PMID: 35856338
23. Zeevi D.. **Personalized Nutrition by Prediction of Glycemic Responses**. *Cell* (2015) **163** 1079-1094. PMID: 26590418
24. Rein M.. **Effects of personalized diets by prediction of glycemic responses on glycemic control and metabolic health in newly diagnosed T2DM: a randomized dietary intervention pilot trial**. *BMC Med.* (2022) **20** 56. PMID: 35135549
25. Gibbons S. M.. **Perspective: Leveraging the Gut Microbiota to Predict Personalized Responses to Dietary, Prebiotic, and Probiotic Interventions**. *Adv. Nutr.* (2022) **13** 1450-1461. PMID: 35776947
26. Shoaie S.. **Quantifying Diet-Induced Metabolic Changes of the Human Gut Microbiome**. *Cell Metab.* (2015) **22** 320-331. PMID: 26244934
27. Heinken A.. **Genome-scale metabolic reconstruction of 7,302 human microorganisms for personalized medicine**. *Nat. Biotechnol.* (2023). DOI: 10.1038/s41587-022-01628-0
28. Abdill R. J., Adamowicz E. M., Blekhman R.. **Public human microbiome data are dominated by highly developed countries**. *PLoS Biol.* (2022) **20** e3001536. PMID: 35167588
29. Magnúsdóttir S.. **Generation of genome-scale metabolic reconstructions for 773 members of the human gut microbiota**. *Nat. Biotechnol.* (2017) **35** 81-89. PMID: 27893703
30. Clark R. L.. **Design of synthetic human gut microbiome assembly and butyrate production**. *Nat. Commun.* (2021) **12** 3254. PMID: 34059668
31. Wastyk H. C.. **Gut-microbiota-targeted diets modulate human immune status**. *Cell* (2021) **184** 4137-4153.e14. PMID: 34256014
32. Manor O.. **Health and disease markers correlate with gut microbiome composition across thousands of people**. *Nat. Commun.* (2020) **11** 5206. PMID: 33060586
33. Quigley E. M. M.. **Gut bacteria in health and disease**. *Gastroenterol. Hepatol.* (2013) **9** 560-569
34. Guinane C. M., Cotter P. D.. **Role of the gut microbiota in health and chronic gastrointestinal disease: understanding a hidden metabolic organ**. *Therap. Adv. Gastroenterol.* (2013) **6** 295-308
35. Valgepea K.. **Systems biology approach reveals that overflow metabolism of acetate in Escherichia coli is triggered by carbon catabolite repression of acetyl-CoA synthetase**. *BMC Syst. Biol.* (2010) **4** 166. PMID: 21122111
36. Wolfe A. J.. **The acetate switch**. *Microbiol. Mol. Biol. Rev.* (2005) **69** 12-50. PMID: 15755952
37. Liu H.. **Butyrate: A Double-Edged Sword for Health?**. *Adv. Nutr.* (2018) **9** 21-29. PMID: 29438462
38. Sze M. A., Topçuoğlu B. D., Lesniak N. A., Ruffin M. T., Schloss P. D.. **Fecal Short-Chain Fatty Acids Are Not Predictive of Colonic Tumor Status and Cannot Be Predicted Based on Bacterial Community Structure**. *MBio* (2019) **10**
39. **Gut microbial metabolites lower blood pressure in patients with hypertension**. *Nat Cardiovasc Res* (2023) **2** 18-19. PMID: 36688212
40. Amiri P.. **Role of Butyrate, a Gut Microbiota Derived Metabolite, in Cardiovascular Diseases: A comprehensive narrative review**. *Front. Pharmacol.* (2021) **12** 837509. PMID: 35185553
41. Tough I. R., Forbes S., Cox H. M.. **Signaling of free fatty acid receptors 2 and 3 differs in colonic mucosa following selective agonism or coagonism by luminal propionate**. *Neurogastroenterol. Motil.* (2018) **30** e13454. PMID: 30136343
42. Ulven T.. **Short-chain free fatty acid receptors FFA2/GPR43 and FFA3/GPR41 as new potential therapeutic targets**. *Front. Endocrinol.* (2012) **3** 111
43. Coppola S., Avagliano C., Calignano A., Berni Canani R.. **The Protective Role of Butyrate against Obesity and Obesity-Related Diseases**. *Molecules* (2021) **26**
44. Babaei P., Shoaie S., Ji B., Nielsen J.. **Challenges in modeling the human gut microbiome**. *Nat. Biotechnol.* (2018) **36** 682-686. PMID: 30080831
45. Gurry T., Nguyen L. T. T., Yu X., Alm E. J.. **Functional heterogeneity in the fermentation capabilities of the healthy human gut microbiota**. *PLoS One* (2021) **16** e0254004. PMID: 34288919
46. Passi A.. **Genome-Scale Metabolic Modeling Enables In-Depth Understanding of Big Data**. *Metabolites* (2021) **12**
47. Gasaly N., de Vos P., Hermoso M. A.. **Impact of Bacterial Metabolites on Gut Barrier Function and Host Immunity: A Focus on Bacterial Metabolism and Its Relevance for Intestinal Inflammation**. *Front. Immunol.* (2021) **12** 658354. PMID: 34122415
48. Agus A., Clément K., Sokol H.. **Gut microbiota-derived metabolites as central regulators in metabolic disorders**. *Gut* (2021) **70** 1174-1182. PMID: 33272977
49. Chen S., Zhou Y., Chen Y., Gu J.. **fastp: an ultra-fast all-in-one FASTQ preprocessor**. *Bioinformatics* (2018) **34** i884-i890. PMID: 30423086
50. Wood D. E., Lu J., Langmead B.. **Improved metagenomic analysis with Kraken 2**. *Genome Biol.* (2019) **20** 257. PMID: 31779668
51. Lu J., Breitwieser F. P., Thielen P., Salzberg S. L.. **Bracken: estimating species abundance in metagenomics data**. *PeerJ Comput. Sci.* (2017) **3** e104
52. Gauglitz J. M.. **Enhancing untargeted metabolomics using metadata-based source annotation**. *Nat. Biotechnol.* (2022) **40** 1774-1779. PMID: 35798960
53. Elmadfa I.. *Österreichischer Ernährungsbericht 2012.* (2012)
54. Brunk E.. **Recon3D enables a three-dimensional view of gene variation in human metabolism**. *Nat. Biotechnol.* (2018) **36** 272-281. PMID: 29457794
55. Waldmann A., Koschizke J. W., Leitzmann C., Hahn A.. **Dietary intakes and lifestyle factors of a vegan population in Germany: results from the German Vegan Study**. *Eur. J. Clin. Nutr.* (2003) **57** 947-955. PMID: 12879089
56. Zhou L.. **Faecalibacterium prausnitzii Produces Butyrate to Maintain Th17/Treg Balance and to Ameliorate Colorectal Colitis by Inhibiting Histone Deacetylase 1**. *Inflamm. Bowel Dis.* (2018) **24** 1926-1940. PMID: 29796620
57. Benjamini Y., Hochberg Y.. **Controlling the false discovery rate: A practical and powerful approach to multiple testing**. *J. R. Stat. Soc.* (1995) **57** 289-300
|
---
title: CTLA-4 antibody-drug conjugate reveals autologous destruction of B-lymphocytes
associated with regulatory T cell impairment
authors:
- Musleh M. Muthana
- Xuexiang Du
- Mingyue Liu
- Xu Wang
- Wei Wu
- Chunxia Ai
- Lishan Su
- Pan Zheng
- Yang Liu
journal: bioRxiv
year: 2023
pmcid: PMC10002750
doi: 10.1101/2023.03.01.530608
license: CC BY 4.0
---
# CTLA-4 antibody-drug conjugate reveals autologous destruction of B-lymphocytes associated with regulatory T cell impairment
## Abstract
Germline CTLA-4 deficiency causes severe autoimmune diseases characterized by dysregulation of Foxp3+ Tregs, hyper-activation of effector memory T cells, and variable forms autoimmune cytopenia including gradual loss of B cells. Cancer patients with severe immune-related adverse events (irAE) after receiving anti-CTLA-4/PD-1 combination immunotherapy also have markedly reduced peripheral B cells. The immunological basis for B cell loss remains unexplained. Here we probe the decline of B cells in human CTLA-4 knock-in mice by using antihuman CTLA-4 antibody Ipilimumab conjugated to a drug payload emtansine (Anti-CTLA-4 ADC). The anti-CTLA-4 ADC-treated mice have T cell hyper-proliferation and their differentiation into effector cells which results in B cell depletion. B cell depletion is mediated by both CD4 and CD8 T cells and at least partially rescued by anti-TNF-alpha antibody. These data revealed an unexpected antagonism between T and B cells and the importance of regulatory T cells in preserving B cells.
## INTRODUCTION
Foxp3 is a master regulator of regulatory T cells (Tregs) and its mutations result in fatal autoimmune disease in mice and human.1–5 Among many known functions, Foxp3 is a transcriptional factor for expression of CTLA-4, which is constitutively expressed in Tregs.6,7 Ctla4 deletion in mice phenocopies that of Foxp3. In humans, CTLA-4 deficiency caused by autosomal heterozygous mutation of the CTLA-4 gene8,9 or mutations in recycling partner LPS-Responsive beige-like anchor (LRBA) protein10 are associated with severe autoimmune diseases. Although CTLA-4 can be expressed at lower levels in other cell types and has been suggested as a negative regulator for naïve T cell activation, lineage-specific deletion of the *Ctla4* gene in Foxp3+ Tregs results in development of systemic lymphoproliferation, fatal autoimmune disease, and potent tumor immunity.11 *These data* are consistent with the notion that the predominant function of CTLA-4 is Treg-intrinsic.
A largely overlooked area is the cross-regulation between B cells and regulatory T cells. CTLA-4 conditional null mice and those with depletion of Tregs showed increase in germinal center B cells and heightened antibody responses with some reports showing a decrease B cell percentage.12–15 B cell loss is a common feature in genetic mutations of Foxp3, scurfy mice, Foxp3 knockout mice or Treg depletion.16–21 Among cancer patients who received immunotherapy, early change of circulating B cells of patients who received combination immunotherapy anti-CTLA-4/PD-1 correlated to irAE.22 The change in B cells included decline of circulating B cells, and increase in CD21 low B cells and plasmablasts.22 The correlation between B cell loss and immune activation associated with defective Foxp3-CTLA-4 function remains unexplained. Since both CTLA-4 and Foxp3 are largely expressed outside of B cell compartment, and since this pathway is the master regulatory of Treg function, we hypothesized that B cell loss maybe associated with activation of T cells that are autodestructive of B cells. This hypothesis is noteworthy as no autodestructive T cells for B cells have been described. To test this hypothesis, we generated anti-CTLA-4 ADC and showed that the ADC caused selective depletion of Tregs in mice and a marked reduction of B cells. Remarkably, activation of CD4 and CD8 T cells is the underlying cause of B cell depletion. Our data explains the B cell loss associated with Foxp3 and CTLA-4 dysfunction and suggested an unexpected antagonism between T cells and B cells.
## CTLA-4 antibody-drug conjugate depleted regulatory T cells and B cells
Anti-CTLA-4 antibody Ipilimumab (Ipi) or human IgGFc (hIgGFc) control were conjugated to a well-known drug payload DM1, emtansine, to furnish the corresponding ADCs: Ipilimumab-DM1 (Ipi-DM1) and hIgGFc-DM1 with corresponding size shifts in the SDS-gel (Figure 1A). The drug to antibody ratio (DAR) was calculated based on experimentally determined extinction coefficient A280 & A252 for each antibody and reported values for DM1 (Table S1). The DAR from various conjugation is ~3.2 for Ipi-DM1 and ~ 1.7 for hIgGFc-DM1 control (Table S1), as expected based on the sizes of the proteins. The binding for Ipi-DM1 ADC was evaluated by ELISA binding to immobilized His-hCTLA-4 and by flow cytometry using hCTLA-4 expressing CHO cells (CHO-hCTLA-4). Ipi-DM1 binding was found comparable to the parent antibody Ipi (Figures 1B, 1C). Specific killing of CTLA-4-expressing cells by Ipi-DM1 was assessed in human CTLA-4 expressing CHO (CHO-hCTLA-4) and wild type CHO (CHO-WT) cells in vitro (Figure 1-figure supplement 1), which showed Ipi-DM1 reduced viability of CLTA-4 expressing CHO but not wild type cells. As expected, there was no change in viability for either CHO cell lines when treated with Ipi (Figure 1-figure supplement 1).
In order to evaluate the in vivo effects of CTLA-4 ADC on Tregs, we treated human CTLA-4 knock-in (Ctla4h/h) mice with control hIgGFc or Ipi-DM1 ADC (Figure 1D). Peripheral blood was stained for flow cytometry and gated on CD45 and T-cell subset, including CD8, CD4, CD4 Foxp3+ and CD4 Foxp3− subsets (Figure 1-figure supplement 2A). We found that Ipi-DM1 ADC significantly depleted Foxp3+ Tregs as indicated by the reduction in percentage and cell number of Foxp3+ CD4 T cells when compared to hIgGFc control (Figure 1E). In addition, total CTLA-4 levels and Ki67 staining in Tregs were decreased compared to control group (Figures 1F, 1G). In contrast, Ipi-DM1 ADC did not alter either % of CD4 Foxp3− cells among CD45+ leukocytes or CTLA-4 levels on the subset, but did slightly decrease the cell number of CD4 Foxp3− cells when compared hIgGFc control (Figure 1-figure supplement 3A, 3B). Unexpectedly, Ipi-DM1 ADC caused a dramatic decline of B cells as percentages of CD45 and cell number in the periphery (Figure 1H). B cells from mice treated with Ipi-DM1 exhibited higher proliferation than those from hIgGFc treated control mice (Figure 1I).
Since a dramatic loss of B cells and a slight decrease in CD4-nonTregs in Ipi-DM1 treated mice were observed, we sought ensure that this phenomenon was not a result of downstream release of payload drug DM1. As shown in Figure 1-figure supplement 4A, hIgGFc-DM1 ADC (payload control) did not alter Treg percentage and cell number. A slightly higher level of CTLA-4 was noted in CD4 subset, while Treg proliferation did not change significantly compared to control (Figure 1-figure supplement 4B, 4C). No other effects on T cell subsets were noted (Figure 1-figure supplement 4D, 4E). Importantly, B cell number and proliferation were unaffected by hIgGFc-DM1 ADC treatment (Figure 1-figure supplement 4F, 4G).
Additionally, to ensure B cell depletion was not directly caused by Ipi-DM1 ADC, we stained B220+ B cells and Foxp3+ Tregs for human CTLA-4 in both human CTLA-4 knock-in (Ctla4h/h) and WT mice to evaluate expression of CTLA-4 in B cells. As expected, CTLA-4 was detected in Foxp3+ Tregs of Ctla4h/h but not B cells (Figure 1-figure supplement 5).
We then investigated the impact of anti-CTLA-4 ADC in lymphoid organs. Treg percentage and absolute numbers were not impacted by Ipi-DM1 treatment compared to control in the spleen at day 9. The percent of B cells decreased significantly, although the reduction in absolute number of B cells was not statistically significant (Figure 1-figure supplement 6A, 6B). Flow analysis of lymph nodes after Ipi-DM1 treatment resulted in a significant increase in absolute cell numbers of Tregs and B cells and B cell percentage. ( Figure 1-figure supplement 6C, 6D).
We then asked how Ipi-DM1 ADC induced Treg impairment impacted B cell lymphopoiesis. Bone marrow cells were stained for flow cytometry and gated on CD45 and T-cell subsets were defined as CD8 (CD8+B220−), CD4 (CD4+B220−), Tregs (CD4+B220− Foxp3+) and CD4-nonTregs (CD4+B220− Foxp3−) (Figure 1-figure supplement 2B). Ipi-DM1 depletion of Tregs was confirmed in the bone marrow (Figure 1J). Additionally, total CTLA-4 level in bone marrow Tregs decreased, however Ki67 staining did not change significantly compared to control group (Figures 1K, 1L). In contrast, Ipi-DM1 ADC did not alter bone marrow CD4-nonTreg cell percentage among CD45+, absolute cell number, or CTLA-4 level compared hIgGFc control (Figure 1-figure supplement 3C, 3D). FACS analysis of bone marrow B cells revealed the loss of mature B cells but not immature transitional type1 or Pre-pro/Pro/Pre B cells from Ipi-DM1 treated mice (Figure 1M–1Q, Figure 1-figure supplement 7). Corresponding to mature B cell loss, the Pre-Pro/Pro/Pre B cell subtype percentage in B220 are enriched (Figure 1P, 1Q, Figure 1-figure supplement 7). Additionally, CD$\frac{21}{35}$ expression level in the remaining mature B cells in Ipi-DM1 treated mice was lower than those from hIgGFc treated control mice (Figure 1R).
We then investigated B cell apoptosis in peripheral blood and lymphoid organs (Figure 2, Figure 2-figure supplement 1). Only mature bone marrow B220hi B cells showed increase in apoptosis from Ipi-DM1 treatment, while progenitor/immature bone marrow B220lo B cells were similar to control group. Blood and spleen B cell apoptosis were similar to hIgGFC control group, while lymph node B cell apoptosis was lower than the control group. Thus, the loss of B cells in the peripheral blood is likely due to loss of mature B cells in the bone marrow, and that such loss correlates with Treg depletion.
Taken together, data in Figure 1 & 2 showed that Ipi-DM1 can impair Treg function by depleting Tregs, preferentially the proliferating Tregs with higher CTLA-4 levels, and loss of B cells in the blood and bone marrow while B cell progenitors are not impacted. B cell death increase is observed in only B220hi bone marrow B cells, which are predominantly mature B cells.
## B-cell depletion correlates with Treg impairment by Ipi-DM1 and increases immunoglobulins
To investigate the kinetic relationship between the decline of B cells and Treg impairment, human CTLA-4 knock-in (Ctla4h/h) mice were treated with control hIgGFc or Ipi-DM1 and bled according to schedule diagram in Figure 3A. We observed that peripheral blood samples from mice treated with Ipi-DM1 had a gradual decrease in total leukocytes compared hIgGFc control group followed by a full recovery by Day 25 (Figure 3B). The large decrease of CD45 cells in Ipi-DM1 group is predominantly attributed to the decline of B cells, which plateaus on days 9 and 13 and recovers by day 25 (Figures 3C, 3D) in correlation with the kinetics of Tregs (Figure 3E, 3F). Similar to the bone marrow data in Figure 1, peripheral blood mature B cells decreased and transitional type1 B cells were enriched in Ipi-DM1 group compared hIgGFc control (Figure 3G–3I). Additionally, Treg impairment by Ipi-DM1 increases immunoglobulins especially IgE (Figure 3-figure supplement 1). The lack of Treg depletion in the spleen and lymph nodes at these time points remains to be explained, although significant immune activation, as suggested by splenomegaly and adenopathy (Figure 3-figure supplement 2), may result in production of cytokines that drive Treg proliferation.
## Depletion of T-cells but not macrophage rescues B-cells from ablation by Ipi-DM1
We then evaluated how Ipi-DM1-mediated impairment of Treg function affects total CD4 and CD8 T cells in peripheral blood. Mice treated with Ipi-DM1 ADC did not affect the total percentage of CD4 or CD8 T cells among CD45+ leukocytes, but the cell number of both cell types decreased (Figure 4A, 4B). Analysis of lymphoid organ T cells showed differing T cell absolute number phenotype where bone marrow, thymus and spleen T cells were similar to control hIgGFc while lymph node T cells increased (Figure 4-figure supplement 1). Staining of Ki67 in CD4 and CD8 revealed a greater percentage of T cells in hyper-proliferative state in blood and lymphoid organs (Figure 4C, Figure 4-figure supplement 2). Alternatively, mice treated with hIgGFc-DM1 payload control did not change CD4 or CD8 percentage of CD45, cell numbers and Ki67 staining compared to hIgGFc control in the periphery (Figure 4-figure supplement 3).
We then analyzed the functional subsets of T cells in mice that received control hIgGFc or Ipi-DM1 at day 9. Using CD44 and CD62L markers, we observed an expansion of effector memory T cells (CD44hiCD62Llow) in Ipi-DM1 group for both CD4 and CD8 T cells (Figures 4D, 4E). Correspondingly, the frequency of naive T cells was reduced for Ipi-DM1 while central memory T cells were unchanged (Figures 4D, 4E). These results show that CTLA-4 ADC can impair Treg function thereby resulting an increase in effector memory T cells and a hyper-proliferative state similar to of CTLA-4 or Foxp3.
Since B cells are devoid of CTLA-4, Ipi-DM1 must have induced other effector cells that are directly responsible for B cell depletion. To understand which cell types are responsible for the destruction of B cells, we depleted either T cells or macrophages using either anti-Thy1.2 mAb or chlondrosome. Ctla4h/h mice were treated with control hIgGFc, Ipi-DM1, Ipi-DM1 in combination with anti-Thy1.2, or Ipi-DM1 in combination with chlondrosome and bled on day 9 (Figure 5A). As shown in the Figure 5B, top panel, anti-Thy1.2 mAb efficiently depleted total T cells while macrophage depletion by chlondrosome had no effect (Figure 5B top panel, 5C, 5D). Additionally, combination of Ipi-DM1 with T cell depleting antibody resulted in a remarkable rescue of B cells as indicated by percentage B220 in CD45 and cell number while combination with macrophage depletion did not rescue the B cells (Figures 5B bottom panel, 5E).
In order to understand which subsets of T cells played a critical role in the T cell mediated destruction of B cells, we treated Ctla4h/h mice with hIgGFc control, Ipi-DM1, IpiDM1 with CD4 depleting antibody, or Ipi-DM1 with CD8 depleting antibody (Figure 5A). Both CD4 and CD8 depleting antibodies provide efficient depletion (Figure 5F, top panel) of their respective targeted T cell. Depletion of either CD4 or CD8 T cells results in B cell increase (Figures 5F bottom panel, 5I). Interestingly, the cell number of CD4 and CD8 T cells increase reciprocally by depletion of the other subsets (Figures 5G, 5H).
Taken together our CTLA-4 antibody-drug conjugate Ipi-DM1 can impair Treg function, which results in a T cell mediated destruction of B cells. Whole T cell depletion with anti-Thy1.2, depletion of CD4 or CD8 T cells rescued B cells from abrogation by Ipi-DM1.
## B-cell rescue by Belatacept suggest a role for B7-CD28 interaction in B-cell depletion
The process for T cell activation requires two signals. The first signal is the binding of T-cell receptor (TCR) to antigen-bound major histocompatibility complex (MHC) and the second signal is a costimulatory molecule CD28 binding with B7−$\frac{1}{2}$ (CD$\frac{80}{86}$) on the antigen-presenting cell. Having shown that the destruction of B cell can be prevented by total depletion of T cell, we sought to rescue the B cells by breaking the CD28-B7 signal with a soluble CTLA-4 that can bind to B7 but not Ipilimumab or Ipi-DM1 ADC. As shown in Figure 6-figure supplement 1, Abatacept can neutralize Ipilimumab/ADC while Belatacept does not. Furthermore, we previously showed that human CTLA4-Ig or mutants can bind to murine B7−$\frac{1}{2}$ (CD$\frac{80}{86}$) and block their binding to murine CD28 efficiently.23,24 Therefore, Belatacept can be used to test if activation of T cells is required for B cell depletion.
We treated mice with Ipi-DM1 to impair Treg function and used Belatacept to break CD28/B7 signal to block T cell activation. Briefly, Ctla4h/h mice were treated with hIgGFc or Ipi-DM1 with/out Belatacept and bled on day 9 for flow analysis. Foxp3+ Tregs percentage in CD4 and cell number decreased similarly for Ipi-DM1 with/out Belatacept compared hIgGFc control (Figures 6A, 6B). Additionally, Belatacept did not impact Ipi-DM1 to target CTLA-4 expressing Tregs, as Ipi-DM1 with/out Belactacept had similar total CTLA-4 level reduction compared to control group (Figure 6C). However, Belatacept rescued B cells from Ipi-DM1-mediated depletion (Figures 6D, 6E). Correspondingly, Belatacept reduced effector memory T cells (Figures 6F, 6G) as well as granzyme B and IFN-γ expression in CD4 and CD8 T cells compared to Ipi-DM1 treated group (Figures 6H–6K, Figure 6-figure supplement 2).
Collectively, the data presented here show that Ipi-DM1 can impair Treg function resulting in T cell activation as indicated by increase in effector memory T cell markers, GranzymB, and cytokine production thereby resulting in the loss of B cells. However, under Ipi-DM1 induced Treg impairment the addition of mutant soluble CTLA-4-Ig, Belatacept, can reduce T cell activation and thereby preserving B cells.
## B-cell depletion is partially rescued by anti-TNF-alpha
Since T cells are the effector cells responsible for B cell abrogation, it is of great interest to evaluate the molecular mechanism by which B cells are eliminated by T cells. To address this issue, we tested if either FasL or TNF-alpha, which are produced by activated T cells are responsible. To answer this question, Ctla4h/h mice were treated with hIgGFc control or Ipi-DM1 with/out *Adalimumab a* human anti-TNF-alpha that also binds to mouse TNF-alpha (Figure 7A, 7B). Flow analysis showed Tregs were decreased similarly between Ipi-DM1 with/out Adalimumab compared hIgGFc control group (Figures 7C top panel, 7D). Remarkably, the anti-TNF-alpha partially rescued B cells as shown by percentage difference of B cell marker B220 between Ipi-DM1 and in combination with Adalimumab as well as cell number (Figures 7C bottom panel, 7E). Peripheral blood samples stimulated with Iononmycin/PMA increased intracellular cytokines TNF-alpha and IFN-gamma for T cells from Ipi-DM1 treated mice compared to hIgGFc control, while that with Ipi-DM1 in combination with Adalimumab had slight increase but the change was insignificant except for IFN-gamma in CD8 T cells (Figure 7F, 7G). In contrast, blocking FAS-L with antibody did not result in B cell rescue (Figure 7- figure supplement 1).
## DISCUSSION
We have reported that pH-insensitive anti-CTLA-4 antibodies trafficked to, and were degraded in the lysosomes.25 Since lysosomal degradation is needed to release DM1 from the antibody for cytotoxicity, the pH-insensitive Ipilimumab was chosen for preparation of antibody-drug conjugate to be used to impact Treg function. We found that in Ctla4h/h mice anti-CTLA-4 ADC, Ipi-DM1, can recapitulate the phenotype of loss of B cells associated clinical deficiency of CTLA-48,9 /recycling partner LBRA10. Our analysis revealed B cell loss in both bone marrow and peripheral blood, specifically mature B cells. Loss of mature peripheral B cells is consistent with clinical data for some patients with CTLA-4 haploinsufficiency.8,9 B cell loss was previously reported in mice with either naturally occurred (in Scurfy mice) or targeted mutation of Foxp3 and Treg depletion.16–21 The kinetics of B cell loss and Treg depletion in the Ipi-DM1 treated mice revealed that B cell loss was transient and correlated to Treg impairment (proliferation, CTLA-4 level, cell number). Our data further showed that B cell destruction under Treg impairment conditions is T cell-mediated and required T cell activation.
Scurfy mice have poor central and peripheral B lymphopoiesis, however neonatal WT Treg adoptive transfer in scurfy mice resulted in robust population of mature B cells in the spleen.19 Genetic ablation of TCRα gene by crossing scurfy with TCRAα−/− mice efficiently support B cell lymphopoiesis and was sufficient to restore B cells in the bone marrow and peripheral compartments.19 Additionally, reconstitution of bone marrow chimera from scurfy or wild type mixed with that from μMT mice that is deficient in B cells results in normal B cells after 8 weeks from reconstitution.18 *This is* in line with our data that Treg impairment results in activated T-cell-mediated apoptosis of mature B cells. Mature B cell loss in bone marrow explains the loss of B cell in the PBL and rules out the possibility that the loss of B cell in the periphery is merely a consequence of defective circulation of B cells normally produced in bone marrow. Others have implied that under Treg depletion, activated T cells are targeting interleukin 7 (IL-7) secreting ICAM1+ perivascular stromal cells needed to progress B cell progenitors.21 IL-7 is essential for B lymphopoiesis transition from Pro-B cell into Pre-B cell.26 However, our model system is distinct where only mature B cells are depleted, while B cell progenitor precursors, and specifically Pre-B cells are not impacted. The involvement of T cells is demonstrated by T cell depletion and requirement for T cell activation is demonstrated by requirement for B7-CD28 interaction for B cell loss. Furthermore, we found that anti-TNF-α can partially rescue B-cells from loss imposed by Treg impairment. These data showed that TNF- α is one of the mediators of B cell loss. Additional studies are needed to reveal other mechanism for B cells loss in Treg-defective environment.
While in vitro studies by others showed B cell killing by activated CD25+CD4 Tregs, but not by CD25− CD4 T cells,27 this is not the case in vivo where B cell loss is associated with Foxp3 Treg loss or impairment.16–21 Furthermore, our current study shows that both CD4 and CD8 T cells contributed to B cell loss. The concept that B cells are actively eliminated by T cells in vivo expands our understanding of T-B cell interaction by showing antagonism rather than just “immunological help” from T cells to B cells.
## Experimental animals
C57BL/6 mice that express the CTLA-4 protein with $100\%$ identity to human CTLA-4 protein under the control of endogenous mouse Ctla4 locus, Ctla4h/h, have been previously described.28 All animals used in experiments were 7–9 weeks age (female mice were used unless male mice are indicated). No blinding or randomization was used and mice were fairly distributed into different treatment groups so that initial average weight of each group were similar. All mice were maintained at the Research Animal Facility of the Institute of Human Virology at the University of Maryland Baltimore School of Medicine. All animal studies were approved by the Institutional Animal Care and Use Committee.
## Cell culture and treatment
CHO cells that were stably transfected with human CTLA-4 have been reported.23 CHO cells were grown in DMEM (Dulbecco’s Modified Eagle Medium, Gibco) supplemented with $10\%$ FBS (Hyclone), 100 units/mL of penicillin and 100 μg/mL of streptomycin (Gibco). Mice Peripheral blood leukocytes were cultured in RPMI-1640 medium (containing $10\%$ FBS and $2\%$ penicillin/streptomycin). All cell lines were incubated at 37 °C and were maintained in an atmosphere containing $5\%$ CO2.
## Viability assay
Wild type CHO (CHO-WT) or human CTLA-4 expressing CHO (CHO-hCTLA-4) cells were seeded at 1,000 cell/well in a flat 96-well plate at 37 °C for 24 hrs in cell culture incubator. The medium was then replaced with fresh medium containing $\frac{1}{4}$ serially diluted vehicle (PBS), Ipilimumab or Ipilimumab-DM1 ADC and cell were incubated at 37 °C for additional 72 hrs. Each treatment group was in duplicate or triplicates. CCK-8 viability dye was then added to each well according to manufacturer’s protcol and incubated for another (2–2.5 hrs) at 37 °C. Wells were subsequently read for absorbance at 450nM on Spectramax ID3 Molecular Devices plate reader. In case of testing whether soluble human CTLA-4-Ig or mutant can neutralize Ipilimumab or Ipilimumab-DM1, the same conditions above were used except that Abetacept or Belatacept concentrations were kept constant at 6 μg/mL. Data is normalized according to the following equation (treatmentOD450 - backgroundaverageOD450 /vehicleaverageOD450 - backgroundaverage) x 100.
## Cell surface CTLA-4 binding
Freshly trypsinized human CTLA-4 expressing CHO (CHO-hCTLA-4) cells were stained with $\frac{1}{4}$ serially dilute anti-CTLA-4 Iplimiumab or Iplimiumab-DM1 ADC in FACS buffer ($2\%$ FBS with 2 mM EDTA) for 30 minutes on ice. Cells were then washed twice with FACS buffer and incubated with anti-human IgG AF488 secondary antibody for 20 minutes on ice. Cells were washed twice with FACS buffer and processed on BD Canto II flow cytometer. In the case of whether soluble human CTLA-4-Ig or mutant can neutralize Ipilimumab or Ipilimumab-DM1 and prevent them from binding cell to surface CTLA-4 the same conditions above were used except that Abetacept or Belatacept concentrations were kept constant at 6 μg/mL.
## Peripheral blood T cell stimulation
Peripheral blood samples 50 μL were treated with ACK Lysis buffer and washed with RPMI-1640 medium. Leukocytes were then stimulated with 1 μg/ml each of phorbol 12-myristate 13-acetate (PMA) (Sigma-Aldrich, St. Louis, MO), ionomycin (Sigma Aldrich, St. Louis, MO) and BD GolgiStop™ (BD Biosciences, cat. 51–2092KZ) and cultured in RPMI-1640 medium (containing $10\%$ FBS and $2\%$ penicillin/streptomycin) for 4 hours at 37°C in 96-well plate. Medium was removed and cells were washed twice with FACS buffer ($2\%$ FBS with 2 mM EDTA) followed by surface staining and fix/perm and intracellular staining.
## Flow Cytometry
Leukocytes from blood or bone marrow were FACS stained directly or after red blood cell lysis with ACK buffer. Fc Receptor was blocked with anti-FCR clone 2.4G2 at 10 μg/mL in FACS buffer for 10 minutes at room temperature and respective surface staining antibodies cocktails were added to each sample and incubated on ice for additional 20 minutes. Cells were then washed twice with 1xPBS and stained with 1x live/Dead Fixable dye Aqua in 1x PBS for 7 minutes at room temperature. Cells were then washed twice with FACS buffer and fixed with eBioscience™ Foxp3/Transcription Factor Staining Buffer Set for 40 minutes. Samples were either washed twice and resuspended in FACS buffer and processed for flow acquisition or further permeabilized for intracellular staining using perm buffer from same kit (eBioscience™ Foxp3/Transcription Factor Staining Buffer Set). In General, all intracellular staining of Foxp3, Ki67, hCTLA-4, GranzymeB, or cytokines were done overnight at 4 degree. In the case of detecting intracellular cytokines, samples were cultured and simulated according to the above protocol followed by surface then intracellular cytokine staining. Apoptosis staining with Annexin V/Pi was performed according manufacturer protocol after surface staining. Samples were acquired by the BD Canto II Flow cytometer and data were analyzed by Flowjo software.
## ELISAs
96-well high-binding polystyrene plates were pre-coated with 50 μL of 1 μg/mL of His-hCTLA-4, or mouse TNF-alpha in coating buffer (0.1M bicarbonate) at 4°C overnight. After washing away the unbound protein/antibody thrice with $0.05\%$ PBST, the plates were blocked with blocking buffer ($1\%$ BSA in PBST) for 1hr at room temperature. All primary antibody incubation was done in blocking buffer at 4°C overnight or room temperature for 1 hour. For coated His-hCTLA-4 a given concentration of either Ipilimumab/DM1, or hIgGFc/DM1; For coated mouse TNF-alpha a given concentration of Adalimumab or Ipilimumab negative control were used. Following primary incubation plates were then washed with PBST for four times and incubated with a goat anti-human IgGFc HRP conjugate secondary antibody at $\frac{1}{20}$,000 dilution for detection in blocking buffer for 1 hr at room temperature. Plates were then washed 4 times with PBST followed by development with 1-Step™ Ultra TMB-ELISA Substrate Solution for 10 minutes and stopped with 2N sulfuric acid. For Immunoglobulin quantification, an ELISA kit was used following manufacturer protocol. Wells were read at 450nM on Spectramax ID3 Molecular Devices plate reader.
## Antibody-drug conjugate preparation
Ipilimumab or hIgGFc were buffer exchanged using prepacked column PD-10 into 1X PBS. Ipilimumab or hIgGFc 5 mg each at concentration 1.5 mg/mL were conjugated with SMCC-DM1(15 eq) in 1x PBS in the presence of $10\%$ DMSO at 37°C under mild shaking conditions for 50 minutes to furnish Ipilimumab-DM1 or hIgGFc-DM1 ADC respectively. Reaction mixture was cleaned up from excess SMCC-DM1 by buffer exchange with a in house ADC buffer at pH 6.5 containing (20 mM Histidine, 8 % sucrose, and Polysorbate 80 $0.02\%$) by Econo-Pac 10DG prepacked column according to manufacturer protocol. Confirmed fraction containing ADC by nano drop were combined and concentrated down to 1 mL using Pierce™ Protein Concentrator PES, (30K for hIgGFc-DM1) or (50K for Ipilimumab-DM1) MWCO according manufacture protocol followed by sterile filtration. ADC concentration was determined using BCA assay according to manufacturer protocol. ADC were diluted to 0.4 – 0.5 mg/mL in 1xPBS and A280 &A252 were recorded on Nanodrop One. The drug to antibody ratio (DAR) was calculated following previous literature.29
## Antibodies and fusion proteins used for in vivo studies
CTLA-4-Ig fusion proteins were synthesized by Sydlabs, Inc (Boston, MA). Recombinant Ipilimumab with amino acid sequence disclosed in WC500109302 was produced by Sydlabs Inc. (Boston, MA). Azide-free human IgG-Fc was purchased from Athens Research and Technology (Athens, GA, USA). Antibody-drug conjugates Ipilimumab-DM1 and hIgGFc-DM1 were prepared from parent antibodies in lab. Depletion antibodies anti-mouse Thy1.2, clone 30H12(BE0066); anti-mouse CD4, clone GK1.5 (BE0003–1); and anti-mouse CD8α, clone 2.43 (BE00–61) were purchased by Bioxcell Inc. (West Lebanon, NH, USA). Anti-mouse FASL, clone MFL3 (BE00319) was purchased by Bioxcell Inc. (West Lebanon, NH, USA). Anti-TNF-α, Adalimumab, clinical grade Humira™ was purchased from Premium Health Services (Columbia, MD, USA).
## Other reagents and material
SMCC-DM1 drug payload with cross linker (Cederlanelabs/Cayman Chemical Co, 23926–10) Anti-mouse FcγR, clone 2.4G2 (Bioxcell Inc, BE0307). Clodrosome® (Encapsula Nano Sciences, SKU# CLD-8909). Mouse anti-human IgGFc secondary antibody (Invitrogen/Thermo Fisher Scientific, 05–42-00). Goat anti-human IgGFc (HRP) preadsorbed (Abcam, ab98624). Polyhistidine-tagged human CTLA-4 (HIS-hCTLA-4) (Sino Biological Inc, 11159-H08H). Mouse TNF-alpha (Sino Biological Inc, 50349-MNAE). IgE ELISA Kit (Thermo Fisher Scientific/Invitrogen, 88–50460-86). IgG ELISA Kit (Thermo Fisher Scientific/Invitrogen, 88–50400-86). IgM ELISA Kit (Thermo Fisher Scientific/Invitrogen, 88–50470-86). IgA ELISA Kit (Thermo Fisher Scientific/Invitrogen, 88–50450-86). 123 eBeads™ counting beads flow (Thermo Fisher Scientific/Invitrogen, 01–1234-42). NuPage™ BIS-TRIS gels 4–$12\%$ (Thermo Fisher Scientific/Invitrogen, NP0335BOX). NuPAGE™ MOPS SDS Running Buffer (20X) (Thermo Fisher Scientific/Invitrogen, NP0001). NuPage™ LDS sample buffer (4x) (Thermo Fisher Scientific/Invitrogen, NP0007). Protein Ladder (Fisher Scientific, BP3603500). Sucrose (Sigma Aldrich, S0389–1KG). L-Histidine (Sigma Aldrich, H8000–10G). Dimethyl sulfoxide (DMSO) (Santa Cruz Biotechnology, Sc-258801). Polysorbate 80 (Fisher Scientific, L13315). Buffer exchange prepacked column PD-10 (GE health care/ Cytiva Life Sciences, 17085101). Buffer exchange prepacked column Econo-Pac 10DG (Bio-Rad, 7322010). Cell Counting Kit-8 (CCK-8) (Bimake, B34304). 1-Step™ Ultra TMB-ELISA Substrate Solution (Thermo Fisher Scientific, 34028). Gibco™ ACK Lysing buffer (Thermo Fisher scientific/Gibco, A1049201). Live/Dead™ Fixable Aqua Dead Cell Stain (Thermo Fisher Scientific/Life Technologies, L34966). eBioscience™ Foxp3 / Transcription Factor Staining Buffer Set (Thermo Fisher Scientific/Invitrogen, 00–5523-00). Pierce™ Protein Concentrator PES, 30K MWCO, 2–6 mL (Thermo Fisher Scientific/ Pierce, 88521) Pierce™ Protein Concentrators PES, 50K MWCO, 2–6 mL (Thermo Fisher Scientific/ Pierce, 88538). Pierce™ BCA Protein Assay Kit (Thermo Fisher Scientific/Pierce, 23227).
## Flow antibodies
eBioscience/ Thermo Fisher Scientific (San Diego, CA): APC-eFlour780 anti- mouse CD45, clone 30-F11 (47–0451-82);eFlour450 anti-mouse CD4, clone GK1.5 (48–0041-82); Pacific Blue anti-mouse CD4, clone RM-4–4 [116008]; PE-Cyanine7 anti-mouse CD8α, clone 53−6.7 (250081–82); PE-Cyanine7 anti-mouse CD8β, clone H35–17.2 (25–0083-82);PerCP-Cy5.5 anti-mouse B220, clone RA3–6B2 (45–0452-82); APC anti-mouse Foxp3, clone FJK-16s (17–5773-82);FITC anti-mouse Ki67, clone SolA15 (11–5698-82);FITC Isotype rat IgG2ak, clone eBR2a (11–4321-82);FITC anti- mouse CD44, clone IM7 (11–0441-85);PerCP-Cy5.5 anti-mouse CD62L, clone MEL-14 (45–0621-82);PE-Cyanine7 anti-mouse B220, clone RA3–6B2 (25–0452-82); APC anti-mouse IgM, clone II/41 (17–5790-82); eFlour 450 anti-mouse CD21/CD35, clone 4E3 (48–0212-82); AF488 anti-mouse TNF-α, clone MP6-XT22 (53–7321-82);AF488 Isotype rat IgG1k, clone eBRG1 (53–4301-80); APC anti-mouse IFNγ, clone XMG1.2 (17–7311-82);APC Isotype rat IgG1k, clone eBRG1 (17–4301-82);FITC anti- mouse GranzymeB, clone NGZB (118898–82); Alexa Fluor 488-conjugated goat anti-human IgG (H+L) cross-adsorbed secondary antibody (A-11013). BioLegend (San Diego, CA): PE anti-human CTLA-4, clone BNI3 [369604];PE Isotype mIgG2ak, clone MPC-173 [400212]; PerCP-Cy5.5 anti-mouse IgD, clone 11–26c.21 [405710]; BV421 anti-mouse CD21/CD35, clone 7E9 [12342]; APC Annexin V Apoptosis Detection Kit with PI [640932];
## Statistical analysis
The specific tests used to analyze each set of experiments are indicated in the figure legends. Data were analyzed using a two-tailed t-test to compare between two groups, one-way or two way ANOVA (analysis of variance) for multiple comparisons. In the graphs, y-axis error bars represent S.E.M. or S.D. as indicated. Statistical calculations were performed using GraphPad Prism software (GraphPad Software, San Diego, California). Non-significant [ns], *$p \leq 0.05$, **$p \leq 0.01$, ***$p \leq 0.001$, ****$p \leq 0.0001.$ Reported CTLA-4 level and absolute cell numbers are normalized by the average value of hIgGFc control group.
## Funding:
This study is supported by grants from the National Institutes of Health (R01AI154722) and OncoC4, Inc.
## References
1. Bennett CL, Christie J, Ramsdell F. **The immune dysregulation, polyendocrinopathy, enteropathy, X-linked syndrome (IPEX) is caused by mutations of FOXP3**. *Nature Genetics* (2001) **27** 20-21. PMID: 11137993
2. Brunkow ME, Jeffery EW, Hjerrild KA. **Disruption of a new forkhead/winged-helix protein, scurfin, results in the fatal lymphoproliferative disorder of the scurfy mouse**. *Nature Genetics* (2001) **27** 68-73. PMID: 11138001
3. Fontenot JD, Gavin MA, Rudensky AY. **Foxp3 programs the development and function of CD4+CD25+ regulatory T cells**. *Nature Immunology* (2003) **4** 330-336. PMID: 12612578
4. Wildin RS, Ramsdell F, Peake J. **X-linked neonatal diabetes mellitus, enteropathy and endocrinopathy syndrome is the human equivalent of mouse scurfy**. *Nature Genetics* (2001) **27** 18-20. PMID: 11137992
5. Wildin RS, Smyk-Pearson S, Filipovich AH. **Clinical and molecular features of the immunodysregulation, polyendocrinopathy, enteropathy, X linked (IPEX) syndrome**. *Journal of Medical Genetics* (2002) **39** 537-545. PMID: 12161590
6. Tivol EA, Borriello F, Schweitzer AN, Lynch WP, Bluestone JA, Sharpe AH. **Loss of CTLA-4 leads to massive lymphoproliferation and fatal multiorgan tissue destruction, revealing a critical negative regulatory role of CTLA-4**. *Immunity* (1995) **3** 541-547. PMID: 7584144
7. Waterhouse P, Penninger JM, Timms E. **Lymphoproliferative Disorders with Early Lethality in Mice Deficient in Ctla-4**. *Science* (1995) **270** 985-988. PMID: 7481803
8. Kuehn HS, Ouyang W, Lo B. **Immune dysregulation in human subjects with heterozygous germline mutations in CTLA4**. *Science* (2014) **345** 1623-1627. PMID: 25213377
9. Schubert D, Bode C, Kenefeck R. **Autosomal dominant immune dysregulation syndrome in humans with CTLA4 mutations**. *Nature Medicine* (2014) **20** 1410-1416
10. Lo B, Zhang K, Lu W. **Patients with LRBA deficiency show CTLA4 loss and immune dysregulation responsive to abatacept therapy**. *Science* (2015) **349** 436-440. PMID: 26206937
11. Wing K, Onishi Y, Prieto-Martin P. **CTLA-4 Control over Foxp3+ Regulatory T Cell Function**. *Science* (2008) **322** 271-275. PMID: 18845758
12. Wing James B, Ise W, Kurosaki T, Sakaguchi S. **Regulatory T Cells Control Antigen-Specific Expansion of Tfh Cell Number and Humoral Immune Responses via the Coreceptor CTLA-4**. *Immunity* (2014) **41** 1013-1025. PMID: 25526312
13. Sage Peter T, Paterson Alison M, Lovitch Scott B, Sharpe Arlene H. **The Coinhibitory Receptor CTLA-4 Controls B Cell Responses by Modulating T Follicular Helper, T Follicular Regulatory, and T Regulatory Cells**. *Immunity* (2014) **41** 1026-1039. PMID: 25526313
14. Klocke K, Sakaguchi S, Holmdahl R, Wing K. **Induction of autoimmune disease by deletion of CTLA-4 in mice in adulthood**. *Proceedings of the National Academy of Sciences* (2016) **113** E2383-E2392
15. Leonardo SM, De Santis JL, Gehrand A, Malherbe LP, Gauld SB. **Expansion of follicular helper T cells in the absence of Treg cells: Implications for loss of B-cell anergy**. *European Journal of Immunology* (2012) **42** 2597-2607. PMID: 22777796
16. Clark LB, Appleby MW, Brunkow ME, Wilkinson JE, Ziegler SF, Ramsdell F. **Cellular and Molecular Characterization of the scurfy Mouse Mutant**. *The Journal of Immunology* (1999) **162** 2546-2554. PMID: 10072494
17. Chen C, Liu Y, Liu Y, Zheng P. **Mammalian target of rapamycin activation underlies HSC defects in autoimmune disease and inflammation in mice**. *The Journal of Clinical Investigation* (2010) **120** 4091-4101. PMID: 20972332
18. Chang S-E, Guo L, Tian J. **Autoimmune bone marrow environment severely inhibits B cell development by inducing extensive cell death and inhibiting proliferation**. *Autoimmunity* (2012) **45** 210-217. PMID: 22053866
19. Riewaldt J, Düber S, Boernert M. **Severe Developmental B Lymphopoietic Defects in Foxp3Deficient Mice are Refractory to Adoptive Regulatory T Cell Therapy**. *Frontiers in Immunology* (2012) **3**
20. Leonardo SM, Josephson JA, Hartog NL, Gauld SB. **Altered B Cell Development and Anergy in the Absence of Foxp3**. *The Journal of Immunology* (2010) **185** 2147-2156. PMID: 20639490
21. Pierini A, Nishikii H, Baker J. **Foxp3+ regulatory T cells maintain the bone marrow microenvironment for B cell lymphopoiesis**. *Nature Communications* (2017) **8** 15068
22. Das R, Bar N, Ferreira M. **Early B cell changes predict autoimmunity following combination immune checkpoint blockade**. *The Journal of Clinical Investigation* (2018) **128** 715-720. PMID: 29309048
23. Du X, Tang F, Liu M. **A reappraisal of CTLA-4 checkpoint blockade in cancer immunotherapy**. *Cell Research* (2018) **28** 416-432. PMID: 29472691
24. Liu M, Wang X, Du X. **Soluble CTLA-4 mutants ameliorate immune-related adverse events but preserve efficacy of CTLA-4– and PD-1–targeted immunotherapy**. *Science Translational Medicine* (2023) **15** eabm5663. PMID: 36857433
25. Zhang Y, Du X, Liu M. **Hijacking antibody-induced CTLA-4 lysosomal degradation for safer and more effective cancer immunotherapy**. *Cell Research* (2019) **29** 609-627. PMID: 31267017
26. von Freeden-Jeffry U, Vieira P, Lucian LA, McNeil T, Burdach SE, Murray R. **Lymphopenia in interleukin (IL)-7 gene-deleted mice identifies IL-7 as a nonredundant cytokine**. *Journal of Experimental Medicine* (1995) **181** 1519-1526. PMID: 7699333
27. Zhao D-M, Thornton AM, DiPaolo RJ, Shevach EM. **Activated CD4+CD25+ T cells selectively kill B lymphocytes**. *Blood* (2006) **107** 3925-3932. PMID: 16418326
28. Lute KD, May KF, Lu P. **Human CTLA4 knock-in mice unravel the quantitative link between tumor immunity and autoimmunity induced by anti-CTLA-4 antibodies**. *Blood* (2005) **106** 3127-3133. PMID: 16037385
29. Chen Y., Ducry L. *Antibody-Drug Conjugates* (2013) 267-273
|
---
title: Analysis of the Role of Stellate Cell VCAM-1 in NASH Models in Mice
authors:
- Kyoung-Jin Chung
- Aigli-Ioanna Legaki
- Grigorios Papadopoulos
- Bettina Gercken
- Janine Gebler
- Robert F. Schwabe
- Triantafyllos Chavakis
- Antonios Chatzigeorgiou
journal: International Journal of Molecular Sciences
year: 2023
pmcid: PMC10002755
doi: 10.3390/ijms24054813
license: CC BY 4.0
---
# Analysis of the Role of Stellate Cell VCAM-1 in NASH Models in Mice
## Abstract
Non-alcoholic fatty liver disease (NAFLD) can progress to non-alcoholic steatohepatitis (NASH), characterized by inflammation and fibrosis. Fibrosis is mediated by hepatic stellate cells (HSC) and their differentiation into activated myofibroblasts; the latter process is also promoted by inflammation. Here we studied the role of the pro-inflammatory adhesion molecule vascular cell adhesion molecule-1 (VCAM-1) in HSCs in NASH. VCAM-1 expression was upregulated in the liver upon NASH induction, and VCAM-1 was found to be present on activated HSCs. We therefore utilized HSC-specific VCAM-1-deficient and appropriate control mice to explore the role of VCAM-1 on HSCs in NASH. However, HSC-specific VCAM-1-deficient mice, as compared to control mice, did not show a difference with regards to steatosis, inflammation and fibrosis in two different models of NASH. Hence, VCAM-1 on HSCs is dispensable for NASH development and progression in mice.
## 1. Introduction
With the continuously expanding obesity pandemic, the prevalence of nonalcoholic fatty liver disease (NAFLD) is constantly increasing [1]. NAFLD is highly associated with insulin resistance, metabolic syndrome and type 2 diabetes and comprises a spectrum of liver pathologies [2,3,4]. Specifically, apart from benign hepatic steatosis, which is characterized by elevated lipid accumulation in hepatocytes, the disease can progress to non-alcoholic steatohepatitis (NASH), characterized by hepatocyte damage, inflammation and fibrosis. NASH affects approximately 1 in 5 NAFLD patients and poses a significantly higher risk for development of cirrhosis and hepatocellular carcinoma (HCC) [5,6]. Since FDA approved treatments for NASH are missing, novel therapeutic strategies are of urgent need [7].
Inflammation is considered a major instigator for the progression of simple steatosis to NASH, with infiltrating monocyte-derived macrophages and activated Kupffer cells playing a cardinal role in this process, via the secretion of inflammatory cytokines and chemokines, such as IL-1b, TNF and CCL-2, as well as major pro-fibrotic mediators, such as TGF-β [8,9,10]. Importantly, these mediators lead to activation of hepatic stellate cells (HSCs), which constitute the principal fibrogenic cell type of the liver. Indeed, upon hepatic damage, HSCs become activated and transdifferentiate into an elongated population of myofibroblasts that produce large amounts of extracellular matrix (ECM) [11]. Continuous HSC activation in response to sustained hepatic damage results to excessive ECM accumulation causing liver fibrosis and scarring, a feature of chronic hepatic disorders including NASH [11].
Despite the multiple soluble mediators, which have been described to activate HSCs in a paracrine fashion provoking their differentiation into myofibroblasts, previous studies have shown that HSCs may also interact with other cells in a direct manner. For instance, HSCs express major histocompatibility molecules (MHC) of both class I and class II, as well as costimulatory molecules, such as CD86 [12,13]. Moreover, pro-inflammatory adhesion molecules, such as VCAM-1, are upregulated in HSCs under inflammatory conditions [14,15,16]. VCAM-1 represents a major counter-receptor for α4β1 integrin in different leukocytes [17,18,19]. In the liver, VCAM-1 in sinusoidal endothelial cells plays a role for leukocyte adhesion during NASH and contributes to fibrosis [20,21].
Interestingly, Lefere et al. reported that serum VCAM-1 levels predicted hepatic fibrosis in patients with NAFLD, indicating a potential role of VCAM-1 in the fibrotic pathogenesis of NASH [22]. Considering the special position of HSCs, which line the space of Disse, and previous findings that VCAM-1 is upregulated in these cells by different inflammatory triggers [14,15,16], we aimed here to investigate the role of VCAM-1 in HSCs for NASH development and progression. To this end, we utilized mice deficient for VCAM-1 in HSCs and appropriate control mice that were subjected to two different established models of diet-induced NASH. Our findings demonstrate that VCAM-1 in HSCs is dispensable for inflammation and fibrosis during NASH.
## 2.1. VCAM-1 Is Upregulated in the Liver during NASH and Expressed by Activated HSC
Leukocyte integrins have been implicated in fibrotic liver diseases [23]. Previous studies investigating the integrin ligand VCAM-1 focused on the hepatic endothelium [20,21], while only a few studies have mentioned the expression of VCAM-1 in HSCs, without providing any mechanistic evidence on its possible role in HSC function and HSC-related pathophysiology during NAFLD and NASH [14,15]. Therefore, we first fed wild-type mice with a control diet (ND) or a methionine-low, choline-deficient high-fat diet (HCD) for 6 weeks, as described in the Materials and Methods, to induce NASH, and we assessed the expression of α4 integrin, the receptor of VCAM-1, on different leukocyte subpopulations, utilizing flow cytometry. Expression of α4-integrin was upregulated upon NASH induction on monocytes, Kupffer cells and monocyte-derived macrophages (Figure 1A). Moreover, the mRNA expression of VCAM-1 was upregulated in the livers of NASH mice (Figure 1B).
As HSCs have been previously reported to express VCAM-1, and given its upregulation in the NASH liver, we next sought to investigate the expression and function of VCAM-1 in HSCs during NASH. To study VCAM-1 expression in activated HSCs during NASH, we first performed immunofluorescence stainings for VCAM-1 together with desmin, a marker of activated HSC, in liver sections from wild-type mice subjected to HCD-induced NASH (Figure 2A). Indeed, VCAM-1 showed substantial co-localization with desmin, thus confirming the presence of VCAM-1 in activated HSCs (Figure 2A). In order to further strengthen this finding, we applied the aforementioned staining strategy in liver sections from HSC-specific VCAM-1-deficient mice and control mice with floxed Vcam-1 (Cre+Vcam1f/f and Cre-Vcam1f/f, respectively) that received the HCD (Figure 2B). Quantification of the immunofluorescence analysis in the stained liver sections revealed that VCAM-1 expression in HSCs was significantly reduced in Cre+Vcam1f/f mice as compared to the Cre-Vcam1f/f mice (Figure 2B). Thus, this staining corroborated that VCAM-1 is expressed by activated HSCs and confirmed the sufficient deletion of VCAM-1 in HSCs in Cre+Vcam1f/f mice (Figure 2B). In addition, both mRNA and protein expression of VCAM-1, as assessed by qPCR and Western Blot analysis, respectively, were significantly reduced in livers of HCD-fed Cre+Vcam1f/f mice, as compared to Cre-Vcam1f/f mice (Figure 2C,D).
## 2.2. VCAM-1 in HSCs Is Dispensable for NASH Development
Next, in order to study the potential role of VCAM-1 in HSCs for the development and progression of NASH, a comprehensive analysis of the livers of Cre-Vcam1f/f and Cre+Vcam1f/f mice was performed. Despite the expression of VCAM-1 in activated HSCs in NASH, HSC-specific VCAM-1 deficiency did not affect the grade of steatosis and fibrosis upon HCD feeding (Figure 3A,B).
To assess NASH-related inflammation, we analyzed leukocyte populations by flow cytometry. No differences between HCD-fed Cre-Vcam1f/f and Cre+Vcam1f/f mice were observed with regards to the numbers of hepatic total leukocytes, neutrophils, Kupffer cells, monocyte-derived macrophages and infiltrating monocytes, as evaluated by flow cytometry analysis (Figure 4A). Moreover, quantitative PCR analysis of the expression of genes related to inflammation (Tnf, Il6, Il1b) and fibrosis (Tgfb1, Acta2, Col1a1, Desmin, Timp1) did not reveal any differences due to HSC-specific VCAM-1 deficiency (Figure 4B).
As, in the HCD-NASH model, liver pathology develops owing to choline deficiency in the diet, we next engaged a second model of NASH, in which pathology develops in a different fashion. In particular, we used a 12-week western diet with high sugar supplementation in the water in conjunction with CCl4 administration to accelerate fibrosis development (Figure 5A); this model was recently shown to mimic histological and transcriptomic characteristics of human NASH [24]. There was no difference in liver weight between Cre-Vcam1f/f and Cre+Vcam1f/f at the end of the feeding period. Systemic metabolism, as assessed by the levels of fasting glucose and triglycerides, was also not different between the two strains (Figure 5B,C). Consistent with the findings from the HCD-model, neither steatosis nor fibrosis was altered in HSC-specific VCAM-1 deficient mice, as compared to the control mice (Figure 5D,E).
Moreover, analysis of the inflammatory milieu of the liver of Cre-Vcam1f/f and Cre+Vcam1f/f mice by flow cytometry displayed no differences in the numbers of hepatic total leukocytes, neutrophils, Kupffer cells, monocyte-derived macrophages and infiltrating monocytes (Figure 6A). Furthermore, expression of genes related to inflammation (Tnf, Il6, Il1b) and fibrosis (Tgfb1, Acta2, Col1a1, Desmin, Timp1) was also not affected by HSC-specific VCAM-1 deficiency (Figure 6B). Together, VCAM-1 in HSCs does not contribute to liver steatosis, inflammation or fibrosis development in the course of NAFLD/NASH, as assessed in two different experimental models.
## 3. Discussion
HSCs are the cellular mediators of fibrosis during NASH via their differentiation from their quiescent state to activated HSCs and myofibroblasts [9,25]. Their activation is mediated by soluble mediators such as IL-1 and TNF, secreted by hepatocytes and several populations of immune cells, as well as major fibrosis-promoting factors such as TGF-β, expressed mainly by activated macrophages, both Kupffer cells and infiltrating monocyte-derived macrophages [9,26,27]. In contrast, less information exists about the role adhesion receptors, such as VCAM-1, may play in the context of HSC activation and transdifferentiation into myofibroblasts, although previous studies have reported upregulation of VCAM-1 in activated HSCs [14,15,16].
This prompted us to study the role of VCAM-1 in HSCs during NASH. We hypothesized that VCAM-1 in HSCs could regulate the accumulation of leukocyte populations in the liver microenvironment during NASH, or regulate intracellular signaling processes involved in HSC transdifferentiation into myofibroblasts. In line with this hypothesis, we have previously shown that adhesion of macrophages onto adipocytes, which are also cells of mesenchymal origin, in a manner that involved adipocyte VCAM-1 expression, can modulate their functional properties [19]. Herein, we first observed an upregulation of VCAM-1 expression in livers from NASH mice as compared to control mice, accompanied by upregulation of α4 integrin, the receptor of VCAM-1, on monocytes, Kupffer cells and monocyte-derived macrophages. By immunofluorescence analysis of liver sections we confirmed the presence of VCAM-1 in HSCs, utilizing desmin as a pan-HSC marker. It should be noted that other markers, such as a-SMA, which is specific for activated-HSCs, were not used in the present study. However, as the model of HCD-induced NASH in mice displays extensive liver fibrosis [21,28,29,30], the majority of HSCs have likely acquired an activated state; hence, our co-staining of liver sections for VCAM-1 and the pan-HSC marker desmin suggests the presence of VCAM-1 on activated HSCs. Previous studies have reported an upregulation of VCAM-1 in the liver and specifically in HSCs under inflammatory conditions, e.g., upon LPS administration or CCl4-induced fibrosis [14,15,16]. Importantly, TLR-4 activation of HSCs led to VCAM-1 upregulation [16]. However, the function of VCAM-1 in HSCs was not studied under NASH conditions so far.
Despite the interesting finding that VCAM-1 expression in HSCs was enhanced during NASH, HSC-specific VCAM-1 deficient mice did not display any differences in steatosis, inflammation and fibrosis, compared to the control mice, as assessed by histology, flow cytometry and gene expression studies in the HCD-induced model. The HCD model is nowadays widely utilized for NASH studies [21,28,29,30]. We further confirmed our findings by subjecting HSC-specific VCAM-1 deficient and control mice to a second model of NASH induction, which is of longer duration as compared to the HCD, while mimicking several aspects of human NASH [24]. The absence of difference in the phenotype of Cre+Vcam1f/f mice as compared to the Cre-Vcam1f/f ones upon NASH induction in the latter model confirmed that VCAM-1 in HSC is dispensable for the progression of the disease. It is possible that other adhesion molecules expressed in HSCs may have compensated for the absence of HSC VCAM-1 in the Cre+Vcam1f/f mice; thus a potential function of HSC VCAM-1 in NASH cannot be entirely excluded. Additionally, we can conclude that HSC VCAM-1 is dispensable for disease development and progression only in the two NASH models used. We cannot exclude that HSC VCAM-1 could play a role in a NASH or liver fibrosis model different from the two models used here. On the contrary, VCAM-1 on LSEC has a role for the accumulation of leukocytes during NASH, thereby accelerating hepatic inflammation and the progression of the disease [20,21].
Although serum VCAM-1 levels correlate with hepatic fibrosis in patients with NAFLD [22], a finding that may be linked with the upregulation of VCAM-1 in activated HSCs, as identified here, our functional results suggest that VCAM-1 in HSCs does not play a pathophysiological role in fibrosis progression. Hence, future studies should interrogate the utilization of VCAM-1 as a biomarker for NASH progression. Moreover, while no function of HSC VCAM-1 in liver fibrosis was found here, we cannot exclude that VCAM-1 in other cells could be a therapeutic target in NASH. These aspects should be addressed in future studies.
## 4.1. Animal Studies
Wild-type mice (C57BL/6) were from Charles River (Sulzfeld, Germany). Hepatic stellate cell-specific deletion of Vcam1 was achieved by crossing mice carrying a floxed Vcam1 allele (Jackson Laboratories, Bar Harbor, ME, USA) with mice in which Cre recombinase expression is driven by the mouse Lecithin:retinol acyltransferase (LRAT) promoter [31]. Wild-type mice were fed a normal chow diet (ND) as control or fed a methionine-low, choline-deficient high-fat diet (HCD, $60\%$ kcal from fat, $0.1\%$ methionine, A06071302, Research Diets) [28,29,32]. Lrat-Cre negative Vcam1 floxed/floxed and Lrat-Cre positive Vcam1 floxed/floxed mice (designated Cre-Vcam1f/f and Cre+Vcam1f/f, respectively) were fed the HCD.
In other experiments, Cre-Vcam1f/f and Cre+Vcam1f/f mice were fed a western diet, specifically a high fat, high fructose, and high cholesterol diet ($21.1\%$ fat, $41\%$ sucrose, and $1.25\%$ cholesterol, Teklad diets, TD. 120528) together with water including high sugar concentrations [23.1 g/L D-Fructose (SERVA, Heidelberg, Germany, 21830) and 18.9 g/L D-Glucose (Sigma-Aldrich, Taufkirchen, Germany, G8270)] for 12 weeks. In addition, the mice received weekly an intraperitoneal low dose of carbon tetrachloride (CCl4, Sigma-Aldrich, 289116, 0.32 µg/g of body weight) as an accelerator of liver fibrosis [24]. After 11 weeks of feeding and upon overnight fasting, blood glucose levels were measured in tail vein blood samples with a glucose meter device (Accu-Chek, Roche, Vienna, Austria), while the levels of triglycerides were measured with an Accutrend Plus system (Roche).
Mice were housed on a standard 12 h light/12 h dark cycle under specific pathogen-free conditions. Eight to ten week old male mice were used in experiments. At the end of the feeding period, mice were euthanized, undergoing also systemic perfusion with phosphate-buffered saline (PBS), and tissues were collected for further analysis. Animal experiments were approved by the Landesdirektion Sachsen, Germany and by the Region of Attica, Greece.
## 4.2. Histological Analysis
Mouse livers were isolated from euthanized mice and fixed with $4\%$ PFA for 24 h. For Hematoxylin and Eosin (H&E) staining, liver samples were embedded in paraffin, and cut liver sections were deparaffinized and rehydrated. The sections were stained with Mayers Haematoxylin (SAV, Liquid Production GmbH, Flintsbach am Inn, Germany, 10231.02500) and Eosin (Klinikapotheke Universitätsklinikum, Dresden, Germany) and mounted with VectaMount (Vector Laboratories, Newark, CA, USA, H-5000-60) after a series of ethanol washings ($80\%$, $95\%$, $100\%$). For Picrosirius red staining, deparaffinized and rehydrated liver sections were stained with Picrosirius red solution (Sigma-Aldrich, 365548) for 1 h and then washed with $1\%$ acetic acid. The liver sections were mounted with VectaMount after a series of ethanol washing as before. Images were acquired utilizing a ZEISS Axio Observer Z1-computerized microscope and Picrosirius red positive areas per field of vision were quantified from at least 12 images per mouse using the Fiji software (ImageJ $\frac{2.1.0}{1.53}$c).
For immunofluorescence staining, fixed liver samples were embedded in OCT upon incubation with a series of sucrose solutions ($10\%$, $20\%$, $30\%$) to achieve tissue cryoprotection. Liver sections were dried and permeabilized with $0.1\%$ Triton X-100 and then blocked using a serum-free protein block solution (Dako, Waldbronn, Germany, X090930-2). Liver sections were then incubated with primary antibodies against VCAM-1 (1:10, eBioscience, Darmstadt, Germany, # 14-1061-85) and desmin (1:100, Abcam, Berlin, Germany, ab32362) overnight at 4 °C. After washing with PBS, sections were incubated with secondary antibodies, namely Donkey anti-Rat (H + L) Alexa Fluor 555 (Abcam, ab150150) and Donkey anti-rabbit (H + L) Alexa Fluor 647 (Invitrogen, Darmstadt, Germany, A-31573) for 90 min at RT and counterstained with DAPI (Sigma-Aldrich, D9542). To reduce tissue autofluorescence, sections were incubated with TrueBlack® Lipofuscin Autofluorescence Quencher (Biotium, Fremont, CA, USA, #23007) for 30 s and mounted. Images were acquired with a ZEISS Axio Observer Z1 computerized microscope equipped with the Zen 3.2 (Blue edition) software. Images are shown in pseudocolor; the display color of the channels was set as to optimize clarity of merged images.
## 4.3. Flow Cytometry Analysis
The left lobe of the liver was isolated, minced and digested in high glucose DMEM containing $0.5\%$ BSA, collagenase D (1.5 mg/mL, COLLD-RO, Roche), and DNaseI (5U/mL, 04716728001, Roche) for 1 h at 37 °C with shaking. The cell suspension was filtered through a 100 µm cell strainer and centrifuged at 600× g for 7 min at 4 °C. Afterwards the red blood cells were lysed using RBC Lysis Buffer (eBioscience, 00-4300-54) for 5 min at room temperature. Additionally, cell debris were removed by utilizing a debris removal solution (Miltenyi Biotec, Bergisch Gladbach, Germany, 130-109-398).
For analysis of α4 integrin expression in hepatic immune cells, upon debris removal the cells were incubated with mouse CD45 Microbeads (Miltenyi Biotec, 130-052-301) for 15 min at 4 °C and CD45+ cells were collected by LS column (Miltenyi Biotec, 130-042-401). Then, they were stained with antibodies against CD11b (M$\frac{1}{70}$, Biolegend, San Diego, CA, USA, 101230), SiglecF (E50-2440, BD Biosciences, Heidelberg, Germany, 562680), Ly6G (1A8, Biolegend, 127624), F$\frac{4}{80}$ (BM8, eBioscience, 25-4801-82), Ly6C (AL-21, BD Biosciences, 553104), α4 integrin/CD49d (9C10, Biolegend, 103706), and Hoechst 33258 (Invitrogen, H1398). For the analysis of hepatic innate immune cells derived from livers of HCD-fed Cre-Vcam1f/f and Cre+Vcam1f/f mice, CD45+ cells, isolated as described above, were stained with antibodies against CD11b (M$\frac{1}{70}$, Invitrogen, 12-0112-82), SiglecF (E50-2440, BD Biosciences, 562680), Ly6G (1A8, Biolegend, 127624), F$\frac{4}{80}$ (BM8, eBioscience, 25-4801-82), Ly6C (AL-21, BD Biosciences, 553104), CD45 (30-F11, Biolegend, 103130), and Hoechst 33258 (Invitrogen, H1398). Stained cells were analyzed on a BD FACSCanto™ II cytometer (BD Biosciences) and analyzed by FlowJo software (v10.1r7).
For analyzing the hepatic innate immune cells acquired from Cre-Vcam1f/f and Cre+Vcam1f/f mice, which were fed a western diet combined with CCl4 treatment, upon debris removal, the cells were stained with antibodies against CD45 (30-F11, Biolegend, 103133), CD11b (M$\frac{1}{70.15}$, Invitrogen, RM2804), Ly6G (1A8, BD Biosciences, 560599), F$\frac{4}{80}$ (BM8, eBioscience, 25-4801-82), Ly6C (AL-21, BD Biosciences, 553104). Stained cells were run on an ARIA III cytometer (BD Biosciences) and analyzed by FlowJo software.
## 4.4. Gene Expression Analysis
Liver tissues were snap frozen in liquid nitrogen or kept in RNAlater (Invitrogen, AM7020). The liver samples were homogenized in TriReagent (MRC, Cincinnati, OH, USA, TR 118) by using the Precellys 24 tissue homogenizer and after phase separation, the RNA was precipitated using $75\%$ ethanol. Finally, RNA was isolated by NucleoSpin® RNA kit (Macherey-Nagel, Dueren, Germany, 740955.250) and reverse-transcribed with the iScript cDNA Synthesis Kit (Bio-Rad, Feldkirchen, Germany, 1708890). The qPCR was performed utilizing the SsoFast™ EvaGreen® Supermix (Bio-Rad, 1725204) and gene-specific primers on a CFX384 Real-time PCR detection system (Bio-Rad). Relative mRNA expression levels were calculated according to the ΔΔCt method upon normalization to 18S [33]. The primers used in this study are:Vcam1 (F:CTTCCCAGAACCCTTCTCAG, R:GGGACCATTCCAGTCACTTC)Tnf (F:AGCCCCCAGTCTGTATCCTTCT, R:AAGCCCATTTGAGTCCTTGATG),Il1b (F:ATCCCAAGCAATACCCAAAG, R:GTGCTGATGTACCAGTTGGG),Il6 (F:CCTTCCTACCCCAATTTCCAAT, R:AACGCACTAGGTTTGCCGAGTA),Tgfb1 (F:CACAATCATGTTGGACAACTGCTCC, R:CTTCAGCTCCACAGAGAAGAACTGC),Col1a1 (F:GAGCGGAGAGTACTGGATCG, R:GCTTCTTTTCCTTGGGGTTC),Desmin (F:GTGGATGCAGCCACTCTAGC, R:TTAGCCGCGATGGTCTCATAC),Acta2 (F:ACTGGGACGACATGGAAAAG, R:GTTCAGTGGTGCCTCTGTCA)Timp1 (F:TACACCCCAGTCATGGAAAGC, R:CGGCCCGTGATGAGAAACT)18S (F:GTTCCGACCATAAACGATGCC, R:TGGTGGTGCCCTTCCGTCAAT)
## 4.5. Immunoblot Analysis
Liver tissues were snap frozen in liquid nitrogen and homogenized in RIPA lysis buffer (Santa Cruz, Heidelberg, Germany, sc-24948A) containing a protease and phosphatase inhibitor cocktail (Roche, 04693159001, CO-RO) by using the Precellys evolution homogenizer (Bertin Technologies, Frankfurt am Main, Germany) and then centrifuged at 14,000× g for 20 min at 4 °C. The supernatant was collected and protein concentrations were determined by using a BCA protein assay kit (Thermo Scientific, Darmstadt, Germany, 23227). The protein samples (30 µg) were separated on a NuPAGE™ 4–$12\%$ gel (Thermo Scientific, NP0323BOX) and transferred to a PVDF membrane (Bio-Rad, 1620177). The membrane was blocked with $5\%$ skim milk for 1 h at RT and then incubated with primary antibody against VCAM-1 (Abcam, ab134047) overnight at 4 °C, followed by incubation with appropriate secondary antibody. After membrane stripping using a Restore Western Blot Stripping-Buffer (Thermo Scientific, 21059), the membrane was blocked again with $5\%$ skim milk for 1 h at RT and then incubated with antibody against Vinculin (Cell signalling, Leiden, The Netherlands, 4650) overnight at 4 °C, followed by incubation with appropriate secondary antibody. Detection in each case was performed with The Super Signal West Pico PLUS Chemiluminescent substrate (Thermo Scientific, 34579) in a VILBER imaging system (FUSION FX, Eberhardzell, Germany). Densitometry was performed by using the Fiji software.
## 4.6. Statistical Analysis
A two-tailed Student’s t-test or a Mann–Whitney U test was used. The Graph Pad Prism 8 software was used and significance was set at $p \leq 0.05.$
## References
1. Younossi Z.M.. **Non-alcoholic fatty liver disease—A global public health perspective**. *J. Hepatol.* (2019) **70** 531-544. DOI: 10.1016/j.jhep.2018.10.033
2. Chalasani N., Younossi Z., Lavine J.E., Diehl A.M., Brunt E.M., Cusi K., Charlton M., Sanyal A.J.. **The diagnosis and management of non-alcoholic fatty liver disease: Practice guideline by the American Association for the Study of Liver Diseases, American College of Gastroenterology, and the American Gastroenterological Association**. *Am. J. Gastroenterol.* (2012) **107** 811-826. DOI: 10.1038/ajg.2012.128
3. Chatzigeorgiou A., Kandaraki E., Papavassiliou A.G., Koutsilieris M.. **Peripheral targets in obesity treatment: A comprehensive update**. *Obes. Rev.* (2014) **15** 487-503. DOI: 10.1111/obr.12163
4. Legaki A.I., Moustakas I.I., Sikorska M., Papadopoulos G., Velliou R.I., Chatzigeorgiou A.. **Hepatocyte Mitochondrial Dynamics and Bioenergetics in Obesity-Related Non-Alcoholic Fatty Liver Disease**. *Curr. Obes. Rep.* (2022) **11** 126-143. DOI: 10.1007/s13679-022-00473-1
5. Loomba R., Friedman S.L., Shulman G.I.. **Mechanisms and disease consequences of nonalcoholic fatty liver disease**. *Cell* (2021) **184** 2537-2564. DOI: 10.1016/j.cell.2021.04.015
6. Wong R.J., Aguilar M., Cheung R., Perumpail R.B., Harrison S.A., Younossi Z.M., Ahmed A.. **Nonalcoholic steatohepatitis is the second leading etiology of liver disease among adults awaiting liver transplantation in the United States**. *Gastroenterology* (2015) **148** 547-555. DOI: 10.1053/j.gastro.2014.11.039
7. Friedman S.L., Neuschwander-Tetri B.A., Rinella M., Sanyal A.J.. **Mechanisms of NAFLD development and therapeutic strategies**. *Nat. Med.* (2018) **24** 908-922. DOI: 10.1038/s41591-018-0104-9
8. Huby T., Gautier E.L.. **Immune cell-mediated features of non-alcoholic steatohepatitis**. *Nat. Rev. Immunol.* (2022) **22** 429-443. DOI: 10.1038/s41577-021-00639-3
9. Katsarou A., Moustakas I.I., Pyrina I., Lembessis P., Koutsilieris M., Chatzigeorgiou A.. **Metabolic inflammation as an instigator of fibrosis during non-alcoholic fatty liver disease**. *World J. Gastroenterol.* (2020) **26** 1993-2011. DOI: 10.3748/wjg.v26.i17.1993
10. Chatzigeorgiou A., Chavakis T.. **Immune Cells and Metabolism**. *Handb. Exp. Pharmacol.* (2016) **233** 221-249. PMID: 25903417
11. Tsuchida T., Friedman S.L.. **Mechanisms of hepatic stellate cell activation**. *Nat. Rev. Gastroenterol. Hepatol.* (2017) **14** 397-411. DOI: 10.1038/nrgastro.2017.38
12. Winau F., Hegasy G., Weiskirchen R., Weber S., Cassan C., Sieling P.A., Modlin R.L., Liblau R.S., Gressner A.M., Kaufmann S.H.. **Ito cells are liver-resident antigen-presenting cells for activating T cell responses**. *Immunity* (2007) **26** 117-129. DOI: 10.1016/j.immuni.2006.11.011
13. Ichikawa S., Mucida D., Tyznik A.J., Kronenberg M., Cheroutre H.. **Hepatic stellate cells function as regulatory bystanders**. *J. Immunol.* (2011) **186** 5549-5555. DOI: 10.4049/jimmunol.1003917
14. Knittel T., Dinter C., Kobold D., Neubauer K., Mehde M., Eichhorst S., Ramadori G.. **Expression and regulation of cell adhesion molecules by hepatic stellate cells (HSC) of rat liver: Involvement of HSC in recruitment of inflammatory cells during hepatic tissue repair**. *Am. J. Pathol.* (1999) **154** 153-167. DOI: 10.1016/S0002-9440(10)65262-5
15. Paik Y.H., Schwabe R.F., Bataller R., Russo M.P., Jobin C., Brenner D.A.. **Toll-like receptor 4 mediates inflammatory signaling by bacterial lipopolysaccharide in human hepatic stellate cells**. *Hepatology* (2003) **37** 1043-1055. DOI: 10.1053/jhep.2003.50182
16. Seki E., De Minicis S., Osterreicher C.H., Kluwe J., Osawa Y., Brenner D.A., Schwabe R.F.. **TLR4 enhances TGF-beta signaling and hepatic fibrosis**. *Nat. Med.* (2007) **13** 1324-1332. DOI: 10.1038/nm1663
17. Harjunpaa H., Llort Asens M., Guenther C., Fagerholm S.C.. **Cell Adhesion Molecules and Their Roles and Regulation in the Immune and Tumor Microenvironment**. *Front. Immunol.* (2019) **10** 1078. DOI: 10.3389/fimmu.2019.01078
18. Hintermann E., Christen U.. **The Many Roles of Cell Adhesion Molecules in Hepatic Fibrosis**. *Cells* (2019) **8**. DOI: 10.3390/cells8121503
19. Chung K.J., Chatzigeorgiou A., Economopoulou M., Garcia-Martin R., Alexaki V.I., Mitroulis I., Nati M., Gebler J., Ziemssen T., Goelz S.E.. **A self-sustained loop of inflammation-driven inhibition of beige adipogenesis in obesity**. *Nat. Immunol.* (2017) **18** 654-664. DOI: 10.1038/ni.3728
20. Furuta K., Guo Q., Pavelko K.D., Lee J.H., Robertson K.D., Nakao Y., Melek J., Shah V.H., Hirsova P., Ibrahim S.H.. **Lipid-induced endothelial vascular cell adhesion molecule 1 promotes nonalcoholic steatohepatitis pathogenesis**. *J. Clin. Investig.* (2021) **131** e143690. DOI: 10.1172/JCI143690
21. Guo Q., Furuta K., Islam S., Caporarello N., Kostallari E., Dielis K., Tschumperlin D.J., Hirsova P., Ibrahim S.H.. **Liver sinusoidal endothelial cell expressed vascular cell adhesion molecule 1 promotes liver fibrosis**. *Front. Immunol.* (2022) **13** 983255. DOI: 10.3389/fimmu.2022.983255
22. Lefere S., Van de Velde F., Devisscher L., Bekaert M., Raevens S., Verhelst X., Van Nieuwenhove Y., Praet M., Hoorens A., Van Steenkiste C.. **Serum vascular cell adhesion molecule-1 predicts significant liver fibrosis in non-alcoholic fatty liver disease**. *Int. J. Obes.* (2017) **41** 1207-1213. DOI: 10.1038/ijo.2017.102
23. Patsenker E., Stickel F.. **Role of integrins in fibrosing liver diseases**. *Am. J. Physiol. Gastrointest. Liver Physiol.* (2011) **301** G425-G434. DOI: 10.1152/ajpgi.00050.2011
24. Tsuchida T., Lee Y.A., Fujiwara N., Ybanez M., Allen B., Martins S., Fiel M.I., Goossens N., Chou H.I., Hoshida Y.. **A simple diet- and chemical-induced murine NASH model with rapid progression of steatohepatitis, fibrosis and liver cancer**. *J. Hepatol.* (2018) **69** 385-395. DOI: 10.1016/j.jhep.2018.03.011
25. Subramanian P., Hampe J., Tacke F., Chavakis T.. **Fibrogenic Pathways in Metabolic Dysfunction Associated Fatty Liver Disease (MAFLD)**. *Int. J. Mol. Sci.* (2022) **23**. DOI: 10.3390/ijms23136996
26. Nati M., Chung K.J., Chavakis T.. **The Role of Innate Immune Cells in Nonalcoholic Fatty Liver Disease**. *J. Innate Immun.* (2022) **14** 31-41. DOI: 10.1159/000518407
27. Peiseler M., Schwabe R., Hampe J., Kubes P., Heikenwalder M., Tacke F.. **Immune mechanisms linking metabolic injury to inflammation and fibrosis in fatty liver disease-novel insights into cellular communication circuits**. *J. Hepatol.* (2022) **77** 1136-1160. DOI: 10.1016/j.jhep.2022.06.012
28. Matsumoto M., Hada N., Sakamaki Y., Uno A., Shiga T., Tanaka C., Ito T., Katsume A., Sudoh M.. **An improved mouse model that rapidly develops fibrosis in non-alcoholic steatohepatitis**. *Int. J. Exp. Pathol.* (2013) **94** 93-103. DOI: 10.1111/iep.12008
29. Moustakas I.I., Katsarou A., Legaki A.I., Pyrina I., Ntostoglou K., Papatheodoridi A.M., Gercken B., Pateras I.S., Gorgoulis V.G., Koutsilieris M.. **Hepatic Senescence Accompanies the Development of NAFLD in Non-Aged Mice Independently of Obesity**. *Int. J. Mol. Sci.* (2021) **22**. DOI: 10.3390/ijms22073446
30. Thangapandi V.R., Knittelfelder O., Brosch M., Patsenker E., Vvedenskaya O., Buch S., Hinz S., Hendricks A., Nati M., Herrmann A.. **Loss of hepatic Mboat7 leads to liver fibrosis**. *Gut* (2021) **70** 940-950. DOI: 10.1136/gutjnl-2020-320853
31. Mederacke I., Hsu C.C., Troeger J.S., Huebener P., Mu X., Dapito D.H., Pradere J.P., Schwabe R.F.. **Fate tracing reveals hepatic stellate cells as dominant contributors to liver fibrosis independent of its aetiology**. *Nat. Commun.* (2013) **4** 2823. DOI: 10.1038/ncomms3823
32. Subramanian P., Gargani S., Palladini A., Chatzimike M., Grzybek M., Peitzsch M., Papanastasiou A.D., Pyrina I., Ntafis V., Gercken B.. **The RNA binding protein human antigen R is a gatekeeper of liver homeostasis**. *Hepatology* (2022) **75** 881-897. DOI: 10.1002/hep.32153
33. Livak K.J., Schmittgen T.D.. **Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method**. *Methods* (2001) **25** 402-408. DOI: 10.1006/meth.2001.1262
|
---
title: Aptamers as Novel Binding Molecules on an Antimicrobial Peptide-Armored Composite
Hydrogel Wound Dressing for Specific Removal and Efficient Eradication of Pseudomonas
aeruginosa
authors:
- Markus Kraemer
- Magali Bellion
- Ann-Kathrin Kissmann
- Tilmann Herberger
- Christopher V. Synatschke
- Anil Bozdogan
- Jakob Andersson
- Armando Rodriguez
- Ludger Ständker
- Sebastien Wiese
- Steffen Stenger
- Barbara Spellerberg
- Kay-Eberhard Gottschalk
- Ahmet Cetinkaya
- Joanna Pietrasik
- Tanja Weil
- Frank Rosenau
journal: International Journal of Molecular Sciences
year: 2023
pmcid: PMC10002764
doi: 10.3390/ijms24054800
license: CC BY 4.0
---
# Aptamers as Novel Binding Molecules on an Antimicrobial Peptide-Armored Composite Hydrogel Wound Dressing for Specific Removal and Efficient Eradication of Pseudomonas aeruginosa
## Abstract
Here we present for the first time a potential wound dressing material implementing aptamers as binding entities to remove pathogenic cells from newly contaminated surfaces of wound matrix-mimicking collagen gels. The model pathogen in this study was the Gram-negative opportunistic bacterium Pseudomonas aeruginosa, which represents a considerable health threat in hospital environments as a cause of severe infections of burn or post-surgery wounds. A two-layered hydrogel composite material was constructed based on an established eight-membered focused anti-P. aeruginosa polyclonal aptamer library, which was chemically crosslinked to the material surface to form a trapping zone for efficient binding of the pathogen. A drug-loaded zone of the composite released the C14R antimicrobial peptide to deliver it directly to the bound pathogenic cells. We demonstrate that this material combining aptamer-mediated affinity and peptide-dependent pathogen eradication can quantitatively remove bacterial cells from the “wound” surface, and we show that the surface-trapped bacteria are completely killed. The drug delivery function of the composite thus represents an extra safeguarding property and thus probably one of the most important additional advances of a next-generation or smart wound dressing ensuring the complete removal and/or eradication of the pathogen of a freshly infected wound.
## 1. Introduction
In addition to potential global threats such as deadly virus infections, as exemplified by the current SARS-CoV-2 pandemic, infections with pathogenic microorganisms have become one of the most threatening health problems of our time [1,2]. Multi-resistant bacteria or fungi urge mankind to develop novel antimicrobial drugs to compensate for the expiration of currently used antimicrobials, which will otherwise lead to a drastic increase in death cases in the near future [3]. Of special concern are nosocomial, i.e. hospital-acquired, infections that can lead to severe complications in the course of their hospital therapies and can lead to significantly elevated mortality in patients [4]. Immunosuppressed patients are most at risk for these types of infections. According to the World Health Organization (WHO), nosocomial infections are the second leading cause of death in the world and more than four million patients are affected by nosocomial infections every year in Europe alone according to the European Center for Disease Control (ECDC)) [5,6]. These statistics can only be interpreted as a truly alarming wake-up call for medical science. Besides the lung in the case of all types of pneumonia, larger wounds, including burn wounds and those originating as consequences of surgery, represent major gateways for microbes to enter the human body, with an intrinsic risk of developing systemic infections in later stages [7,8]. Consequently, prevention of sepsis (also designated “septicemia”) as a still growing and menacing general problem requires efficient control of infections as early as possible, which may involve the precautionary application of antimicrobial drugs and the meticulous, anti-infective treatment of wounds [9,10]. Wound dressings, in general, may represent the first line of defense against infections of larger lesions and injuries and can play an important role in the immediate management of freshly infected tissue [11,12]. Next-generation wound dressing materials often contain hydrogel constituents responsible for moisture control on the wound surface and additional beneficial functions, including sensors to monitor infections, or equipped with an affinity toward dedicated pathogens allowing their specific immobilization and removal from the wound surface during wound care [13,14,15,16,17].
We have previously introduced composite wound dressing materials based on protein hydrogels functionalized with lectins for the efficient binding of pathogens such as the pathogenic yeast *Candida auris* and the opportunistic Gram-negative bacterium P. aeruginosa, which is one of the most important causative agents of severe infections and, as a carbapenem-resistant mutant strain, it is part of an alarming global health threat identified by the WHO [16,17]. Treatment of strains already resistant against last resort antibiotics requires novel antimicrobial drugs, which can be acquired from the heterogenic molecule class of antimicrobial peptides (AMPs) including those from marine animals (e.g., snails) [18,19,20,21,22,23]. These AMPs are expected to require proper delivery in these next-generation wound dressing concepts to successfully combat infections. The 16 amino acid-long antimicrobial peptide C14R (amino acid sequence: CSSGSLWRLIRRFLRR) has proven its antimicrobial activity against carbapenem-resistant P. aeruginosa as an antimicrobial drug molecule itself, and also upon release from a two-layered composite of a protein-based affinity hydrogel and a drug carrier compartment fabricated from fibril forming-protected amino acid monomers [17]. Such “capture-and-kill” concepts exemplified with the target pathogen P. aeruginosa used lectins as affinity molecules, which can bind specifically to glycosylated cell surface structures of bacteria and higher cells, including yeasts [16,17,24,25]. By contrast, DNA aptamers represent a class of ligand molecules that are evolved in the laboratory directedly by SELEX processes and can have truly remarkable specificities toward dedicated targets [26,27,28]. Polyclonal aptamer libraries, resulting directly from the SELEX process, as well as single aptamers isolated by bioinformatic analyses from these libraries can bind P. aeruginosa strains with high specificity, and have been used as binding entities on protein beads for the specific capturing of P. aeruginosa from serum and blood [29,30]. The immediate use of such polyclonal libraries originating directly from SELEX processes without the need for subsequent analyses and selection of their individual members is inspired by the concept of using polyclonal antibody preparations instead of monoclonal antibodies. We have previously shown that these libraries can outperform individual aptamers, because multiple epitopes on the target structure are recognized allowing a robust binding of the target regardless of possible alterations in the composition through mutation or the growth phase [29,30].
Here we present a two-layered hydrogel composite material, which makes use of the pool of P. aeruginosa-specific aptamers (C1R1, C2R1, C2R2, C2R10, C4R2, C6R3, C10R5 and C10R6) chemically crosslinked to the material surface to form a trapping zone for efficient binding of the pathogen from a minimalized wound model consisting of a collagen-based gel resembling the extracellular matrix in a fresh wound. A drug-loaded zone of the composite released the C14R antimicrobial peptide to deliver it directly to the bound pathogenic cells (Figure 1). We demonstrate that this material combining aptamer-mediated affinity and peptide-dependent pathogen eradication (AA-EraGel) can quantitatively remove bacterial cells from the “wound” surface, and we show that the surface-trapped bacteria are completely killed. The drug delivery function of the composite thus represents an extra safeguarding property and thus probably one of the most important additional advances of a next-generation or smart wound dressing ensuring the complete removal and/or eradication of the pathogen of a freshly infected wound.
## 2. Results
We have recently introduced BSA crosslinked with EDC as a promising hydrogel material for biomedical applications, including functionalization with aptamers; however, this was in the form of hydrogel spheres in the range of approximately 1 mm [30,31]. Similar to our previous wound dressing publications, the intention here was to prepare larger-sized patches of composites to be subsequently not only loaded with drugs but also equipped with a specific affinity layer. Thus, we first demonstrated the successful functionalization of the BSA hydrogel layer with a focused library consisting of eight different individual aptamers against the surface structures of P. aeruginosa [29]. The synthesized aptamer library was used to functionalize the BSA hydrogel material with the help of the crosslinker PEG4-SPDP. In a two-step reaction, PEG4-SPDP first reacts with the thiol group of the BSA in a displacement reaction of the sulfur atom. In the second step, the NHS ester attacks the amino group of the modified aptamers. We tentatively started using 1.27 pmol per well, which reflects the calculated amount of BSA molecules being present on the surface of the hydrogel (theoretical 1:1 ratio). The successful binding of the aptamers to the surface was then verified by incubation with the complementary strand oligonucleotide fluorescently labeled with Cyanine-5 (Cy-5 reverse Primer), which can hybridize to the 3′- primer binding site of the aptamer molecules. Fluorescence labeling of the material was only achieved when both crosslinkers and aptamers were applied in the labeling reaction (Figure 2A). By contrast, when only one of the modifying agents was absent, the material was not stained, as was the case for the negative control represented by the pure material (Figure 2A).
To demonstrate the functionality of our concept, we then needed to show the binding of *Pseudomonas cells* to the material. We thus used a P. aeruginosa PAO1 pVLT-31 eGFP derivative, which expressed a recombinant GFP under transcriptional control of the tac-promoter of pVLT-31 as a fluorescence label upon induction with isopropyl-ß-D-thiogalactopyranosid (IPTG). The amount of aptamers was systematically increased from 0.1 to 10 pmol to functionalize the hydrogel surface in a well of a 96-well plate (0.29 cm2) to estimate the influence of the aptamer concentration on the resulting binding capacity. The observed obvious development of cell numbers bound along the increasing aptamer concentrations below 1 pmol did not, however, progress with a significant gain in the binding capacity at higher concentrations of aptamers used for functionalization. Interestingly, the theoretical 1:1 ratio of aptamer to BSA molecules on the material surface of a well (1.27 pmol) used in the initial experiment is perfectly close to the fitted non-linear regression of the results from the experiment shown in Figure 2B, suggesting that the 1:1 ratio does not generate a significant surplus of binding capacity compared to 1 pmol. We thus decided to use 1 pmol per well (3.45 pmol·cm−2 or 0.0345 pmol·mm−2) for the aptamer functionalization in the subsequent experiments.
Specificity testing revealed that the material perfectly bound P. aeruginosa cells in its completely functionalized form, whereas the GFP expressing the probiotic gut bacterium *Escherichia coli* Nissle 1917 pVLT31-eGFP, which served as a Gram-negative control bacterium, completely failed to bind to the material (Figure S1). This specificity was not limited when the cell number of the target bacterium was increased by $100\%$, even when the same high amount of the “contaminating” control bacteria was additionally present (Figure 2C).
The binding capacity of the aptamer-functionalized hydrogel was determined by depositing 200 μL of bacterial cell cultures adjusted to different and increasing numbers of cells in a fresh culture medium. The maximum of cells binding to the material was reached when 100,000 cells were applied and did not increase further with higher cell numbers. Under these conditions, i.e., with a liquid column of approximately 0.7 cm above the affinity material, 20,000 cells could be bound, which is $\frac{1}{3}$ higher than the typical amount of bacteria in post-surgery wounds (calculated 14,500 cells per well area) (Figure 2D) [32].
After the characterization of the trapping zone of our intended composite material, we then had to demonstrate the functionality of the drug zone as a reservoir for antimicrobial peptides, their release from this reservoir and finally the complete functionality of the resulting composite material. First, the pure reservoir gel consisting of fibrillar Fmoc-protected phenylalanine was loaded with C14R to a final concentration of 40 μM and then covered with a 3x volume of assay buffer and incubated for 24 h. Samples were taken after 1, 2, 4 and 24 h and the peptide concentration in the assay buffer was determined to calculate the cumulative release of the peptide from the hydrogel reservoir. This gel showed the expected high release rate with the complete release already after two hours, whereas the composite showed a complete release after 24 h (Figure 2E) [33]. Fibrillary hydrogels of this type have been shown to be typically soft and have therefore been combined with mechanically supporting layers consisting of crosslinked BSA, which had the additional function of serving as attachment layers for binding molecules, equipping the resulting composite material with specificity toward selected pathogens [16,17]. Another important feature is that in combination this supporting layer does not only represent a not entirely impermeable barrier but allows an efficient diffusion to ensure a fast and even distribution of the drug molecules to be released. This is not only a relevant property with respect to later industrial processes of such materials but a homogeneously drug-preloaded affinity layer also results in instant pharmacological effects for a convenient topical application of the final wound dressing material.
The ultimate goal of the AA-EraGel affinity trap concept is the application of the hydrogel-based composite for wound care applications for high-risk patients or in high-risk hospital environments such as intensive care units, in which immediate post-surgery infections with severe microbial pathogens must be prohibited to protect patients from the establishment of persisting and life-threatening infections. This involves the removal of P. aeruginosa cells from the wound surface immediately after the onset of infection during or after the completion of surgery and, as the second key step, the subsequent killing of the pathogen safeguarding the total eradication of the bacteria. This dual mode of action represents a combined first defense line unifying simple binding and subsequent mechanical removal from the wound surface and early pharmaceutical inhibition of the establishment of an infection by cells remaining on the wound surface. As a first and simple experimental step to approaching this goal, we decided to use a collagen hydrogel as a model surface mimicking a naturally occurring extracellular matrix for the binding of pathogenic cells to which the affinity trap composite could be directly applied afterward. In this idealized early post-infection wound care situation, four regions (Figure 3A) of the large-area wound and the contact areas between the wound surface and wound dressing are of relevance for analyzing the success of pathogen removal and killing. Whereas in close proximity to the application site (Position A) an antimicrobial effect of the AMP can be expected due to its diffusion, this declines with the distance to the application site resulting in an undisturbed development of the pathogen far remote from the application site (Position B). Most important in this context is the area of the wound being in direct contact with a wound dressing, consisting of the respective wound surface (Position C) and the opposite surface of the wound dressing material itself, which after removal will carry attached cells in the case of successful high-affinity binding to the material (Position D). To reflect this situation, a simulated extracellular wound matrix was created on the bottom of a 12-well microtiter plate by coating it with collagen from rat tails, which was then “infected” with an excess of P. aeruginosa cells in our experimental setup. To simulate an immediate post-surgery infection, cells were forced toward the collagen matrix by gentle centrifugation and allowed to establish physical contact for 30 min. The AA-EraGel model wound-dressing patch consisting of the peptide-loaded affinity composite hydrogel trap, or the empty version serving as a negative control focusing only on the affinity aspect, were then applied to the “infected” wound surface for 24 h to exhibit their respective impacts on the pathogenic cells. The wound dressing patches were then removed and the number of cells present on the four relevant regions after this procedure was quantified by measuring the integrated density of fluorescence and, in the case of positions B and D (i.e., far away from the contact area as a control and directly on the surface of the wound-dressing patch, respectively), by the resazurin-based viability assay. Meeting the expectation, the killing effect of C14R on P. aeruginosa cells residing at positions A and B was more pronounced for the loaded material close to the patch but reduced at a larger distance, where the difference between the peptide-loaded material and the empty control turned out to be not significant (Figure 3B). Directly under the wound dressing (Position C), no distinct cells were visible in fluorescence microscopy independent from the material variant used. This is an indication that contact with the material surface without applying the antimicrobial peptide is sufficient to quantitatively remove bacterial cells from the wound surface model. Interestingly, both position C regions showed residual fluorescence, which may be caused by cell lyses upon contact with the wound dressing and appears to be independent of the presence of the peptide (Figure 3B). By contrast, on the former wound-oriented surface of the removed patch (Position D) cells were present in numbers similar to the unaffected regions B, which were significantly reduced in the presence of the peptide (Figure 3B). The fluorescence microscopy analysis was then complemented by the resazurin-based live/dead assay on the residual cells of Position D, which confirmed that the peptide effectively killed P. aeruginosa on the wound dressing surface (Figure 3C).
## 3. Discussion
“Smart wound dressings” as a concept to treat severe and predominantly chronic wounds (e.g., in the context of diabetes) in the future were introduced in the early 2000s, with a logarithmic rise in publication numbers leading to several hundred per year at present [34]. By definition, the development of such materials or devices involves different scientific disciplines including engineering, chemistry, biology, physics, or medical science, explaining why the main focus on what is interesting and “smart” can be extremely diverse. Materials have been developed which allow the monitoring of the wound status to evaluate wound healing and the presence or onset of potentially harmful infections, through different technologies including fluorescence-based techniques and the implementation of electronic components for signal generation and transduction to monitoring devices [35,36,37,38]. Another main aspect is to promote wound healing or protect wounds from being negatively affected by infections with pathogenic microbes. Wound dressing materials have been invented which can release growth factors or antimicrobial drugs including antimicrobial peptides [16,17,39,40]. We have recently enlarged the portfolio of options by introducing the so far underrepresented aspect of manipulating early phases of wound infections by providing affinity functions against pathogens to the wound dressing material by decorating the surface of hydrogel composites with microbial multivalent lectins [16,17,24,41,42]. However, the lectin-mediated binding is not limited to pathogenic yeasts or bacteria but can also immobilize human cells in vitro and exert potentially considerable adverse biological effects on wound healing in vivo [43,44]. In contrast to relatively unspecific lectins, aptamers as a versatile class of binding molecules are gaining a remarkable degree of scientific prominence for their broad spectrum of dedicated targets, which is not limited to small molecules or proteins but can include with high specificity and robustness the binding and recognition of whole cells, tissues and organs [45,46,47,48]. The group of whole cells as potential aptamer targets consists of both pathogenic or probiotic Gram-positive and Gram-negative bacteria, as well as other prominent pathogens such as yeasts from the *Candida genus* [29,49,50,51,52,53,54,55]. The aptamer technology has been used to develop various aptamer-based biosensors for clinical diagnostics, food, and environmental monitoring of P. aeruginosa in recent years (nicely reviewed in Zheng et al., 2020) [56]. We have previously shown that in contrast to individually selected aptamer molecules, it may be advantageous to use polyclonal aptamer libraries for robust and secure binding of P. aeruginosa cells [29]. These libraries were further developed into a focused library consisting of eight different aptamers and allowed specific and reliable binding of the pathogen on hydrogel-based surfaces in serum and blood [30]. Our previous work showed that composite hydrogels consisting of BSA-derived materials and fibrillar hydrogels composed of the Fmoc-protected amino acids phenylalanine and methionine were suited to developing possible next-generation wound dressing materials with respect to their mechanical, rheological and diffusion properties [16,17,25,33]. The functionality of such materials was demonstrated by efficiently delivering the antimicrobial peptides Cm-p5 and C14R to pathogenic Candida spec. or P. aeruginosa cells residing on their surfaces [16,17,23,57]. To our knowledge, so far, aptamers have not been implemented as affinity-mediating molecules on biomaterials intended for applications in wound care. We thus wanted to show that the specificity of the focused anti-P. aeruginosa library can be used to provide capturing functionality on the AA-EraGel basic wound care hydrogel composite. Functionalization of the BSA-based trapping zone top layer resulted in specific and efficient immobilization of the target cells. This was also the case for cells residing on the surface of a collagen-based gel mimicking, as an extremely simplified model, the matrix of a fresh wound. In this model, P. aeruginosa cell numbers could be removed with a capacity higher than the typical count of bacteria in post-surgery wounds [32]. The viability assays performed with the cells sticking to the removed model AA-EraGel wound dressing patch showed that as expected the residual cells were killed by the C14R peptide. Thus, we reasoned that the application of AA-EraGel to the model wound allows not only the sufficient quantitative removal of P. aeruginosa from freshly infected surfaces in a significant amount but can also be strengthened by delivering antimicrobial drugs to the trapped pathogens. In this respect, the drug delivery represents a safeguarding measure, which may have additional importance, especially if intermediate or long-term dwell times of the final wound care materials in hospital environments or at patient bedsides at home are intended. We believe that the use of highly specific binding molecules such as the anti-P. aeruginosa aptamers used in the first example presented here may open new avenues toward next-generation wound dressing materials with respect to specificity and efficiency to control or prevent the establishment of harmful (hospital-acquired) infections by removing the infective agent from the treated wound. We also believe that functionalization with aptamers is not only not limited to our model hydrogel materials but can also be attractive in combination with novel biomaterials and innovative wound care applications. The implementation of aptamer-based optical or electronic sensing platforms into wound care materials, or the analysis of binding events of pathogens to affinity materials with such sensors alone, represent logical and attractive further opportunities [58,59,60]. Examples of interest on the alternative material side may be polydopamine films, which have been successfully introduced as wound dressing constituents (nicely reviewed in Alfieri et al., 2022 and Yazdi et al., 2022) [61,62]. Pathogen-specific aptamers may be further developed as attractive molecular binding entities and important affinity mediators in the toolbox of materials, techniques and drugs which in integrated novel material design approaches can help to deliver new gentle and efficient wound care and skin regeneration options in healthcare environments.
## 4.1. Specific Anti-P. aeruginosa PAO1 Aptamer PCR
The eight specific aptamers against P. aeruginosa PAO1 (C1R1, C2R1, C2R2, C2R10, C4R2, C6R3, C10R5 and C10R6), which were already characterized in Kubiczek et al., 2020, were produced by PCR. An NH2-labeled primer ((5′[NH2]-TAG GGA AGA GAA GGA CAT ATG AT-3′), Eurofins Genomics Germany GmbH) and a biotin-labeled reverse primer ((5′[Biotin]-TCA AGT GGT CAT GTA CTA GTC AA-3′), Eurofins Genomics Germany GmbH, Ebersberg, Germany) were used [29]. The following was utilized for each PCR: 1x PCR buffer (1.5 mM MgCl2), 250 µM of each dNTP, 0.25 µM of the labeled primers and 0.2 µM aptamer ssDNA, Herculase II Fusion DNA Polymerase (Agilent Technologies, Inc., Santa Clara, CA, USA). Amplification was performed in the thermocycler SensoQuest Labcycler (SensoQuest GmbH, Göttingen, Germany) with an initial denaturation step at 94 °C for 3 min, followed by 25 cycles at 94 °C for 30 s for denaturation, 49.1 °C for 30 s for annealing, 72 °C for 30 s for elongation and a final extension at 72 °C for 2 min. Subsequently, the respective PCR products for each aptamer were pooled and gel electrophoresis (gel with $2\%$ agarose in $0.5\%$ TBE buffer) was performed. After loading the gel with 5 µL of dsDNA from each sample and 1 µL of 6x TriTrack DNA Loading Dye (Thermo Fisher Scientific, Inc., Waltham, MA, USA), it was performed at 150 V for 35 min and then stained in a $0.007\%$ ethidium bromide bath and viewed under UV light (E-Gel® Imager, Thermo Fisher Scientific, Inc., Waltham, MA, USA).
## 4.2. Preparation of Aptamer ssDNA
The preparation of the ssDNA aptamers was performed as in Krämer et al., 2021, already described to be further used on the hydrogel construct [30]. Here, 50 µL of streptavidin-coated magnetic beads were washed three times with 1 mL of 1× DPBS using a magnetic separator. Amplified aptamer dsDNA was then added, covered and incubated at 22 °C and 50 rpm for 16 h. To remove unbound dsDNA, the supernatant was removed, and the magnetic streptavidin beads were washed with 1 mL of 1× DPBS. After that, 50 µL of NaOH (100 mM) was added to the magnetic streptavidin beads and incubated for 2 min without magnetic separation. Then, 45 µL of the NaOH solution was transferred to a new reaction tube containing 126 µL of 1× DPBS and 34.4 µL of NaH2PO4 buffer (100 mM). The remaining 5 µL of ssDNA was used for separation by gel electrophoresis. After gel electrophoresis, the ssDNA concentration was measured in ng µL−1 using an Implen NP80 nanophotometer (Implen GmBH, Munich, Germany).
## 4.3. Synthesis of AMP
A derivative of BP100, named C14R, with the amino acid sequence CSSGSLWRLIRRFLRR was synthesized on a 0.10 mmol scale via standard Fmoc solid phase peptide synthesis techniques. This was achieved with a preloaded serine resin and washed with dimethylformamide (DMF). To remove the protecting group of Fmoc, $20\%$ (v/v) piperidine in DMF which was initialized with microwaves was used. After that, an additional washing step with DMF was performed. Subsequently, the addition of amino acids in 0.2 mol equiv. to the reactor was carried out, followed by HBTU 2-(1H-benzotriazol-1-yl)-1,1,3,3-tetramethyluronium-hexafluorophosphate) in a 0 and 5 mol equiv. into the amino acid solution. Then, 2 mol equiv. of N,Ndiisopropylethylamine (DIEA) was added. The coupling reaction was achieved via microwaves within minutes. The resin was again washed in DMF. This procedure was repeated for all amino acids in the sequence. After completion of the amino acid synthesis, the peptide was cleaved in $95\%$ (v/v) trifluoracetic acid (TFA), $2.5\%$ (v/v) triisopropylsilane (TIS) and $2.5\%$ (v/v) H2O for 1 h. A precipitation of the peptide residue and washing with cold diethyl ether (DEE) by centrifugation was carried out which was then under vacuum to remove the residual ether. To purify the peptide, reversed phase preparative high-performance liquid chromatography (Waters GmbH, Eschborn, Germany) was used in an acetonitrile/water gradient under acidic conditions on a Phenomenex C18 Luna column (5 mm pore size, 100 Å particle size, 250–21.2 mm). Liquid chromatography–mass spectroscopy (Waters GmbH, Eschborn, Germany) was used to measure the peptide mass.
## 4.4. Bacteria Cultivation
Two strains were used in the experiments: P. aeruginosa PAO1 pVLT31-eGFP and E. coli Nissle 1917 pVLT31-eGFP. For the cultivation of each bacterial strain, precultures were prepared in 5 mL of LB medium (Carl Roth GmbH + Co. KG, Karlsruhe, Germany) containing 10 µg mL−1 of tetracycline (Carl Roth GmbH + Co. KG, Karlsruhe, Germany). These precultures were incubated at 37 °C for 16 h while shaking at 150 rpm. The next day, the OD600 values of the precultures were determined, and a flask containing 25 mL LB medium and 10 µg mL−1 tetracycline was subsequently inoculated with an initial OD600 of 0.05. Cultivation of the flasks was performed at 37 °C with shaking (150 rpm). Upon reaching an OD600 of 0.6, cultures were induced with 0.4 mM isopropyl-ß-d-1-thiogalactopyranoside (IPTG, Carl Roth GmbH + Co. KG, Karlsruhe, Germany). The bacterial cultures were then incubated for up to three more hours until the stationary phase was reached. In order to achieve the desired cell numbers per mL in the subsequent experiments, the cells were counted under the microscope (Leica Microsystems CMS GmbH, Wetzlar, Germany) using a Thoma cell counting chamber (LO-Laboroptik GmbH, Friedrichsdorf, Germany).
## 4.5. BSA Hydrogel and Functionalization with Aptamer ssDNA
The BSA hydrogels were prepared as reported previously; in brief, two stock solutions were prepared, a $20\%$ (w/v) BSA- (neoFroxx GmbH, Einhausen, Germany) and a $10\%$ (w/v) EDC-solution (Carl Roth GmbH + Co. KG, Karlsruhe, Germany), both diluted in MES (2-(N-morpholino)ethanesulfonic acid) buffer (100 mM) [16,17,30]. These were then mixed in a one-to-one ratio (40 µL) in a 96-well plate. In the next step, PEG4 SPDP linker (4-unit polyethylene glycol spacer arm, with an amine-reactive N-hydroxysuccinimide (NHS) ester at one side and a sulfhydryl-reactive 2-pyridyldithiol group at the other end) (Thermo Fisher Scientific, Waltham, MA, USA) (6.5 μL, 0.2 μM in DMSO) was added and filled up with 200 µL PBS-EDTA and incubated for 16 h at room temperature, followed by extensive washing twice with 200 µL of PBS-EDTA. The eight aptamers were prepared in the desired concentrations and activated to ensure the correct folding in PBS-EDTA. They were heated to 95 °C for 5 min, cooled for 5 min on ice, and then the aptamer solution was covered and stored at RT for 30 min to facilitate the correct folding of the aptamers. The native 3D structure of the aptamers can be ensured after complete denaturation and direct refolding. The aptamer solution was covered and stored at RT for 30 min. After that, the aptamer solution was added to the well and filled up to 200 µL with PBS-EDTA buffer, and then incubated while covered for 1 h at RT. After incubation, the aptamer-crosslinked BSA hydrogels were washed 3 times with 200 µL PBS-EDTA buffer.
## 4.6. Peptide Hydrogel and AMP Loading
To prepare the peptide hydrogel, Fmoc-phe-OH was diluted in DMSO (100 mg mL−1) and then mixed with phosphate buffer (10 mM Na2HPO4; 42 mM NaH2PO4 in demin water, pH 7.4) in a ratio of 12.5:1, as described previously [17]. They were also prepared in a 96-well plate (50 µL). To load the hydrogels with AMP, the appropriate volume of AMP was added to the phosphate buffer to obtain a concentration of 10 μg mL−1. The hydrogels were stored at 4 °C overnight.
## 4.7. Composite Hydrogel
BSA and peptide hydrogels were prepared as described above. The peptide hydrogel, which was loaded with AMP, was cast first and polymerized overnight at 4 °C. The BSA hydrogel, which polymerizes very fast, was placed on top of the peptide hydrogel the next day and functionalized with aptamers.
## 4.8. Verification of the Functionalized BSA Hydrogels
To visualize the crosslinked aptamers on the BSA hydrogel, a Cy-5 labeled probe (5′[Cy5]-TCA AGT GGT CAT GTA CTA GTC AA-3′] was hybridized to the aptamers crosslinked to the hydrogels. Various approaches have been taken for this purpose. The complete construct (functionalized with crosslinker PEG4-SPDP and aptamer as described in Section 4.5) was used as well as hydrogels without aptamer or without crosslinker, as well as without both. Subsequently, after the preparation of the various constructs, the probe was activated as described for the aptamers in 4.5. Then, 1.27 µL of the 1 pmol µL−1 activated Cy-5 labeled probe was pipetted onto the gels. For hybridization, the hydrogel was covered with 200 µL PBS-EDTA buffer and incubated under light exclusion overnight at 4 °C. The next day, unattached probe remnants were removed by washing three times with 200 µL PBS-EDTA buffer. Afterward, the hydrogels were examined by fluorescent microscopy at 400× magnification with the Leica DMi8 inverted fluorescent microscope (Leica Microsystems CMS GmbH, Wetzlar, Germany). Images of the different conditions were taken and analyzed with FIJI image analysis software [63].
## 4.9. Binding Specificity Analysis of P. aeruginosa PAO1 pVLT31-eGFP with Different Aptamer Concentrations
The influence of the aptamer concentration used for BSA hydrogel functionalization on its binding capacity was investigated in the following experiment. BSA hydrogels were prepared as described in Section 4.5. After the first functionalization with the crosslinker PEG4-SPDP and washing 3 times with 200 µL PBS-EDTA, the desired amounts of 0.1, 0.25, 0.5, 0.75, 1, 2, 5 and 10 pmol of NH2-labeled aptamer were added to the gels for complete functionalization. Then, the samples were incubated while covered for 1 h at RT. After incubation, the BSA hydrogels were washed 3 times with 200 µL PBS-EDTA. In the next step, P. aeruginosa cells/200 µL (106 cells mL−1) were pipetted to each sample and incubated for 30 min while covered at RT. Then, the gels were washed 3 times with 200 µL PBS-EDTA and filled up to 200 µL with PBS-EDTA. In the next step, all samples were investigated with fluorescence microscopy at 100× magnification. After that, the microscopy images were analyzed with FIJI [63].
## 4.10. Binding Specificity Analysis of P. aeruginosa PAO1 pVLT31-eGFP to Incomplete and Fully Functionalized BSA Hydrogel
To determine the specific binding of P. aeruginosa PAO1 pVLT31-eGFP against the crosslinked aptamers to the BSA hydrogels, various constructs as described in Section 4.8 were used. Therefore, BSA hydrogels from Section 4.5, and P. aeruginosa PAO1 pVLT31-eGFP bacteria and E. coli Nissle 1917 pVLT31-eGFP bacteria, as a control strain, from Section 4.4, were used. The first experiment had the fully functionalized hydrogel construct. The second control experiment had an incomplete functionalization construct without NH2 aptamers. The third construct was crosslinked for 16 h with PEG4-SPDP but was not functionalized with NH2-labeled aptamers, and the last construct was without both crosslinker and aptamer. After washing 3 times with 200 µL PBS-EDTA, 25,000 bacterial cells/200 µL of P. aeruginosa PAO1 pVLT31-eGFP or E. coli Nissle 1917 pVLT31-eGFP were pipetted into the wells and incubated while covered for 30 min at RT. Afterward, all constructs were washed 3 times with 200 µL PBS-EDTA and then filled up to 200 µL with PBS-EDTA. The samples were then examined by fluorescence microscopy at 100× magnification with the Leica DMi8 inverted fluorescent microscope. Fluorescence microscopy images were analyzed with FIJI [63].
## 4.11. Binding Specificity Analysis of P. aeruginosa PAO1 pVLT31-eGFP in Presence of a “Contaminating” Control Bacteria
To show that other bacteria have no influence on the binding of P. aeruginosa to the hydrogel trap, E. coli Nissle 1917 pVLT31-eGFP bacteria were used to produce bacterial mixtures with P. aeruginosa PAO1 pVLT31-eGFP. For this purpose, the hydrogels were first functionalized as in Section 4.5. After functionalization, precultures of P. aeruginosa and E. coli were prepared as in Section 4.4, the bacteria cells were counted, and different ratios of 12,500 cells/200 µL and 25,000 cells/200 µL per bacterium were used. In the next step, the samples were covered and incubated for 30 min at RT, washed 3 times with 200 µL PBS-EDTA and then filled up to 200 µL with PBS-EDTA. All samples were examined with fluorescence microscopy at 100× magnification, and images were analyzed with FIJI [63].
## 4.12. Binding Capacity of P. aeruginosa PAO1 pVLT31-eGFP
For the subsequent experiments, BSA hydrogels were prepared as described in Section 4.5 in a 96-well plate. After full functionalization with PEG4- SPDP and 1 pmol NH2-labeled aptamer, the gels were washed 3 times with 200 µL PBS-EDTA. In the following step, 6,250, 12,500, 25,000, 50,000, 75,000, 100,000 and 125,000 cells/200 µL of P. aeruginosa PAO1 pVLT31-eGFP were pipetted to the fully aptamer-functionalized hydrogels. After covered incubation for 30 min at RT, the hydrogels were washed 3 times with 200 µL PBS-EDTA and then filled up to 200 µL with PBS-EDTA. In the next step, all samples were examined with fluorescence microscopy at 100× magnification, and all images were analyzed by FIJI [63].
## 4.13. Protein Release of AMP C14R
For the subsequent experiments, AMP-loaded hydrogels were prepared as described in Section 4.6 in a 96-well plate. AMP-loaded hydrogels were covered with 200 µL PBS. Samples were taken at regular intervals (1 h, 2 h, 4 h and 24 h) and then recovered with 200 µL PBS. The absorption of the samples was measured at 280 nm with a Tecan Spark microplate reader (Tecan Group Ltd., Männedorf, Switzerland). All measurements were performed in triplicate. The error bars represent the standard deviation.
## 4.14. Collagen Matrix Model
A collagen model was prepared for wound care. For this, 150 µL of a stock solution of 1 mg mL−1 collagen from rat tail in acetic acid and 10x PBS with 1.5 μL NaOH (1M) was poured into a 24-well plate. Gelation was performed for 1 h at 30 °C. Subsequently, 106 cell mL−1 of P. aeruginosa bacteria were centrifuged onto the collagen matrix and incubated while covered for 30 min at RT. The composite hydrogel loaded with C14R (10 μg mL−1) was placed on the collagen layer by removing it from the mold with tweezers, facing down with the affinity layer to the cells. After 24 h of incubation, the composite hydrogel was taken off and the affinity layer as well as the collagen matrix, in close proximity as well as far away and under the wound dressing, were examined with fluorescence microscopy at 10× magnification, and all images were analyzed with FIJI [63].
## 4.15. Viability Test: Resazurin Assay
A resazurin assay was performed on the wound patch to test cell viability. The composite hydrogel was covered after incubation on the collagen matrix with 200 μL resazurin (0.15 mg mL−1). After two hours of incubation, the amount of metabolized fluorescent resorufin was measured with a Tecan200 M fluorescence reader (Tecan Group Ltd., Mannedorf, Switzerland) at an excitation wavelength of 535 nm and an emission wavelength of 595 nm.
## 4.16. Statistical Analysis
Statistical analysis was performed with a Mann–Whitney U-test [64].
## References
1. Gorbalenya A.E., Baker S.C., Baric R.S., de Groot R.J., Drosten C., Gulyaeva A.A., Haagmans B.L., Lauber C., Leontovich A.M., Neuman B.W.. **The Species Severe Acute Respiratory Syndrome-Related Coronavirus: Classifying 2019-NCoV and Naming It SARS-CoV-2**. *Nat. Microbiol.* (2020) **5** 536-544. PMID: 32123347
2. Li F.. **Receptor Recognition and Cross-Species Infections of SARS Coronavirus**. *Antiviral Res.* (2013) **100** 246-254. DOI: 10.1016/j.antiviral.2013.08.014
3. Tanwar J., Das S., Fatima Z., Hameed S.. **Multidrug Resistance: An Emerging Crisis**. *Interdiscip. Perspect. Infect. Dis.* (2014) **2014** 541340. DOI: 10.1155/2014/541340
4. Liu J.-Y., Dickter J.K.. **Nosocomial Infections: A History of Hospital-Aquired Infections**. *Gastrointest. Endosc. Clin. N. Am.* (2020) **30** 637-652. DOI: 10.1016/j.giec.2020.06.001
5. Allegranzi B., Nejad S.B., Combescure C., Graafmans W., Attar H., Donaldson L., Pittet D.. **Burden of Endemic Health-Care-Associated Infection in Developing Countries: Systematic Review and Meta-Analysis**. *Lancet* (2011) **377** 228-241. DOI: 10.1016/S0140-6736(10)61458-4
6. Tacconelli E., Carrara E., Savoldi A., Harbarth S., Mendelson M., Monnet D.L., Pulcini C., Kahlmeter G., Kluytmans J., Carmeli Y.. **Discovery, Research, and Development of New Antibiotics: The WHO Priority List of Antibiotic-Resistant Bacteria and Tuberculosis**. *Lancet Infect. Dis.* (2018) **18** 318-327. DOI: 10.1016/S1473-3099(17)30753-3
7. Dhaliwal K., Lopez N.. **Hydrogel Dressings and Their Application in Burn Wound Care**. *Br. J. Community Nurs.* (2018) **23** S24-S27. DOI: 10.12968/bjcn.2018.23.Sup9.S24
8. Lanks C.W., Musani A.I., Hsia D.W.. **Community-Acquired Pneumonia and Hospital-Acquired Pneumonia**. *Med. Clin. N. Am.* (2019) **103** 487-501. DOI: 10.1016/j.mcna.2018.12.008
9. Flanagan J.M., Read C., Shindul-Rothschild J.. **Factors Associated with the Rate of Sepsis after Surgery**. *Crit. Care Nurse* (2020) **40** e1-e9. DOI: 10.4037/ccn2020171
10. Fan S., Miller N.S., Lee J., Remick D.G.. **Diagnosing Sepsis—The Role of Laboratory Medicine**. *Clin. Chim. Acta* (2017) **460** 203-210. DOI: 10.1016/j.cca.2016.07.002
11. Daeschlein G.. **Antimicrobial and Antiseptic Strategies in Wound Management**. *Int. Wound J.* (2013) **10** 9-14. DOI: 10.1111/iwj.12175
12. Kamoun E.A., Kenawy E.-R.S., Chen X.. **A Review on Polymeric Hydrogel Membranes for Wound Dressing Applications: PVA-Based Hydrogel Dressings**. *J. Adv. Res.* (2017) **8** 217-233. DOI: 10.1016/j.jare.2017.01.005
13. Farahani M., Shafiee A.. **Wound Healing: From Passive to Smart Dressings**. *Adv. Healthc. Mater.* (2021) **10** 2100477. DOI: 10.1002/adhm.202100477
14. Liang Y., He J., Guo B.. **Functional Hydrogels as Wound Dressing to Enhance Wound Healing**. *ACS Nano* (2021) **15** 12687-12722. DOI: 10.1021/acsnano.1c04206
15. Francesko A., Petkova P., Tzanov T.. **Hydrogel Dressings for Advanced Wound Management**. *Curr. Med. Chem.* (2019) **25** 5782-5797. DOI: 10.2174/0929867324666170920161246
16. Kubiczek D., Flaig C., Raber H., Dietz S., Kissmann A.K., Heerde T., Bodenberger N., Wittgens A., González-Garcia M., Kang F.. **A Cerberus-Inspired Anti-Infective Multicomponent Gatekeeper Hydrogel against Infections with the Emerging “Superbug” Yeast**. *Macromol. Biosci.* (2020) **20** 2000005. DOI: 10.1002/mabi.202000005
17. Bodenberger N., Kubiczek D., Halbgebauer D., Rimola V., Wiese S., Mayer D., Rodriguez Alfonso A.A., Ständker L., Stenger S., Rosenau F.. **Lectin-Functionalized Composite Hydrogels for “Capture-and-Killing” of Carbapenem-Resistant**. *Biomacromolecules* (2018) **19** 2472-2482. DOI: 10.1021/acs.biomac.8b00089
18. Raber H.F., Sejfijaj J., Kissmann A.K., Wittgens A., Gonzalez-Garcia M., Alba A., Vázquez A.A., Vicente F.E.M., Erviti J.P., Kubiczek D.. **Antimicrobial Peptides Pom-1 and Pom-2 from**. *Pathogens* (2021) **10**. DOI: 10.3390/pathogens10040496
19. Amann V., Kissmann A.K., Krämer M., Krebs I., Perez-Erviti J.A., Otero-Gonzalez A.J., Morales-Vicente F., Rodríguez A., Ständker L., Weil T.. **Increased Activities against Biofilms of the Pathogenic Yeast**. *Pharmaceutics* (2022) **14**. DOI: 10.3390/pharmaceutics14020318
20. García M.G., Rodríguez A., Alba A., Vázquez A.A., Vicente F.E.M., Pérez-Erviti J., Spellerberg B., Stenger S., Grieshober M., Conzelmann C.. **New Antibacterial Peptides from the Freshwater Mollusk**. *Biomolecules* (2020) **10**. DOI: 10.3390/biom10111473
21. Vicente F.E.M., González-Garcia M., Diaz Pico E., Moreno-Castillo E., Garay H.E., Rosi P.E., Jimenez A.M., Campos-Delgado J.A., Rivera D.G., Chinea G.. **Design of a Helical-Stabilized, Cyclic, and Nontoxic Analogue of the Peptide Cm-P5 with Improved Antifungal Activity**. *ACS Omega* (2019) **4** 19081-19095. DOI: 10.1021/acsomega.9b02201
22. González-García M., Morales-Vicente F., Pico E.D., Garay H., Rivera D.G., Grieshober M., Olari L.R., Groß R., Conzelmann C., Krüger F.. **Antimicrobial Activity of Cyclic-Monomeric and Dimeric Derivatives of the Snail-Derived Peptide Cm-P5 against Viral and Multidrug-Resistant Bacterial Strains**. *Biomolecules* (2021) **11**. DOI: 10.3390/biom11050745
23. Kubiczek D., Raber H., Gonzalez-García M., Morales-Vicente F., Staendker L., Otero-Gonzalez A.J., Rosenau F.. **Derivates of the Antifungal Peptide CM-P5 Inhibit Development of**. *Antibiotics* (2020) **9**. DOI: 10.3390/antibiotics9070363
24. Funken H., Bartels K.-M., Wilhelm S., Brocker M., Bott M., Bains M., Hancock R.E.W., Rosenau F., Jaeger K.-E.. **Specific Association of Lectin LecB with the Surface of**. *PLoS ONE* (2012) **7**. DOI: 10.1371/journal.pone.0046857
25. Bodenberger N., Kubiczek D., Rosenau F.. **Easy Manipulation of Architectures in Protein-Based Hydrogels for Cell Culture Applications**. *J. Vis. Exp.* (2017) **126** e55813. DOI: 10.3791/55813
26. Tuerk C., Gold L.. **Systematic Evolution of Ligands by Exponential Enrichment: RNA Ligands to Bacteriophage T4 DNA Polymerase**. *Science* (1990) **249** 505-510. DOI: 10.1126/science.2200121
27. Ellington A.D., Szostak J.W.. **In Vitro Selection of RNA Molecules That Bind Specific Ligands**. *Nature* (1990) **346** 818-822. DOI: 10.1038/346818a0
28. Ku T.-H., Zhang T., Luo H., Yen T.M., Chen P.-W., Han Y., Lo Y.-H.. **Nucleic Acid Aptamers: An Emerging Tool for Biotechnology and Biomedical Sensing**. *Sensors* (2015) **15** 16281-16313. DOI: 10.3390/s150716281
29. Kubiczek D., Raber H., Bodenberger N., Oswald T., Sahan M., Mayer D., Wiese S., Stenger S., Weil T., Rosenau F.. **The Diversity of a Polyclonal FluCell-SELEX Library Outperforms Individual Aptamers as Emerging Diagnostic Tools for the Identification of Carbapenem Resistant**. *Chem.—Eur. J.* (2020) **26** 14536-14545. DOI: 10.1002/chem.202000213
30. Krämer M., Kissmann A.K., Raber H.F., Xing H., Favella P., Müller I., Spellerberg B., Weil T., Kubiczek D., Sihler S.. **Bsa Hydrogel Beads Functionalized with a Specific Aptamer Library for Capturing**. *Int. J. Mol. Sci.* (2021) **22**. DOI: 10.3390/ijms222011118
31. Favella P., Sihler S., Raber H., Kissmann A.K., Krämer M., Amann V., Kubiczek D., Baatz J., Lang F., Port F.. **Albumin Microspheres as “Trans-Ferry-Beads” for Easy Cell Passaging in Cell Culture Technology**. *Gels* (2021) **7**. DOI: 10.3390/gels7040176
32. Bowler P.G., Duerden B.I., Armstrong D.G.. **Wound Microbiology and Associated Approaches to Wound Management**. *Clin. Microbiol. Rev.* (2001) **14** 244-269. DOI: 10.1128/CMR.14.2.244-269.2001
33. Favella P., Kissmann A.K., Raber H.F., Kubiczek D.H., Bodenberger P., Bodenberger N.E., Rosenau F.. **Diffusion-Controlled Release of the Theranostic Protein-Photosensitizer Azulitox from Composite of Fmoc-Phenylalanine Fibrils Encapsulated with BSA Hydrogels**. *J. Biotechnol.* (2021) **341** 51-62. DOI: 10.1016/j.jbiotec.2021.08.014
34. Sweitzer S.M., Fann S.A., Borg T.K., Baynes J.W., Yost M.J.. **What Is the Future of Diabetic Wound Care?**. *Diabetes Educ.* (2006) **32** 197-210. DOI: 10.1177/0145721706286897
35. Mariani F., Serafini M., Gualandi I., Arcangeli D., Decataldo F., Possanzini L., Tessarolo M., Tonelli D., Fraboni B., Scavetta E.. **Advanced Wound Dressing for Real-Time PH Monitoring**. *ACS Sens.* (2021) **6** 2366-2377. DOI: 10.1021/acssensors.1c00552
36. Pang Q., Lou D., Li S., Wang G., Qiao B., Dong S., Ma L., Gao C., Wu Z.. **Smart Flexible Electronics-Integrated Wound Dressing for Real-Time Monitoring and On-Demand Treatment of Infected Wounds**. *Adv. Sci.* (2020) **7** 1902673. DOI: 10.1002/advs.201902673
37. Zhang Y., Li T., Zhao C., Li J., Huang R., Zhang Q., Li Y., Li X.. **An Integrated Smart Sensor Dressing for Real-Time Wound Microenvironment Monitoring and Promoting Angiogenesis and Wound Healing**. *Front. Cell Dev. Biol.* (2021) **9** 701525. DOI: 10.3389/fcell.2021.701525
38. Qiao B., Pang Q., Yuan P., Luo Y., Ma L.. **Smart Wound Dressing for Infection Monitoring and NIR-Triggered Antibacterial Treatment**. *Biomater. Sci.* (2020) **8** 1649-1657. DOI: 10.1039/C9BM02060H
39. Khan M.U.A., Haider S., Raza M.A., Shah S.A., Razak S.I.A., Kadir M.R.A., Subhan F., Haider A.. **Smart and PH-Sensitive RGO/Arabinoxylan/Chitosan Composite for Wound Dressing: In-Vitro Drug Delivery, Antibacterial Activity, and Biological Activities**. *Int. J. Biol. Macromol.* (2021) **192** 820-831. DOI: 10.1016/j.ijbiomac.2021.10.033
40. Zhao Y., Huang L., Lin G., Tong M., Xie Y., Pan H., Shangguan J., Yao Q., Xu S., Xu H.. **Skin-Adaptive Film Dressing with Smart-Release of Growth Factors Accelerated Diabetic Wound Healing**. *Int. J. Biol. Macromol.* (2022) **222** 2729-2743. DOI: 10.1016/j.ijbiomac.2022.10.054
41. Loris R., Tielker D., Jaeger K.-E., Wyns L.. **Structural Basis of Carbohydrate Recognition by the Lectin LecB from**. *J. Mol. Biol.* (2003) **331** 861-870. DOI: 10.1016/S0022-2836(03)00754-X
42. Tielker D., Hacker S., Loris R., Strathmann M., Wingender J., Wilhelm S., Rosenau F., Jaeger K.-E.. *Microbiology* (2005) **151** 1313-1323. DOI: 10.1099/mic.0.27701-0
43. Bodenberger N., Kubiczek D., Trösch L., Gawanbacht A., Wilhelm S., Tielker D., Rosenau F.. **Lectin-Mediated Reversible Immobilization of Human Cells into a Glycosylated Macroporous Protein Hydrogel as a Cell Culture Matrix**. *Sci. Rep.* (2017) **7** 6151. DOI: 10.1038/s41598-017-06240-w
44. Thuenauer R., Landi A., Trefzer A., Altmann S., Wehrum S., Eierhoff T., Diedrich B., Dengjel J., Nyström A., Imberty A.. **The**. *mBio* (2020) **11** e03260-19. DOI: 10.1128/mBio.03260-19
45. Zhong W., Pu Y., Tan W., Liu J., Liao J., Liu B., Chen K., Yu B., Hu Y., Deng Y.. **Identification and Application of an Aptamer Targeting Papillary Thyroid Carcinoma Using Tissue-SELEX**. *Anal. Chem.* (2019) **91** 8289-8297. DOI: 10.1021/acs.analchem.9b01000
46. Li L., Wan J., Wen X., Guo Q., Jiang H., Wang J., Ren Y., Wang K.. **Identification of a New DNA Aptamer by Tissue-SELEX for Cancer Recognition and Imaging**. *Anal. Chem.* (2021) **93** 7369-7377. DOI: 10.1021/acs.analchem.1c01445
47. Kissmann A.K., Wolf D., Krämer M., Müller F., Amann V., Xing H., Gottschalk K.E., Weil T., Eichmann R., Schäfer P.. **Polyclonal Aptamer Libraries from a FluRoot-SELEX for the Specific Labeling of the Apical and Elongation/Differentiation Zones of**. *Int. J. Mol. Sci.* (2022) **23**. DOI: 10.3390/ijms232012220
48. Lou B., Liu Y., Shi M., Chen J., Li K., Tan Y., Chen L., Wu Y., Wang T., Liu X.. **Aptamer-Based Biosensors for Virus Protein Detection**. *TrAC Trends Anal. Chem.* (2022) **157** 116738. DOI: 10.1016/j.trac.2022.116738
49. **Paniel; Noguer Detection of Salmonella in Food Matrices, from Conventional Methods to Recent Aptamer-Sensing Technologies**. *Foods* (2019) **8**. DOI: 10.3390/foods8090371
50. Kusumawati A., Mustopa A.Z., Wibawan I.W.T., Setiyono A., Sudarwanto M.B.. **A Sequential Toggle Cell-SELEX DNA Aptamer for Targeting**. *J. Genet. Eng. Biotechnol.* (2022) **20** 95. DOI: 10.1186/s43141-022-00374-9
51. Shin H.-S., Gedi V., Kim J.-K., Lee D.. **Detection of Gram-Negative Bacterial Outer Membrane Vesicles Using DNA Aptamers**. *Sci. Rep.* (2019) **9** 13167. DOI: 10.1038/s41598-019-49755-0
52. Xing H., Zhang Y., Krämer M., Kissmann A.K., Henkel M., Weil T., Knippschild U., Rosenau F.. **A Polyclonal Selex Aptamer Library Directly Allows Specific Labelling of the Human Gut Bacterium**. *Molecules* (2022) **27**. DOI: 10.3390/molecules27175693
53. Raber H.F., Kubiczek D.H., Bodenberger N., Kissmann A.K., D’souza D., Hu X., Mayer D., Xu P., Knippschild U., Spellerberg B.. **Flucell-selex Aptamers as Specific Binding Molecules for Diagnostics of the Health Relevant Gut Bacterium**. *Int. J. Mol. Sci.* (2021) **22**. DOI: 10.3390/ijms221910425
54. Xing H., Kissmann A.K., Raber H.F., Krämer M., Amann V., Kohn K., Weil T., Rosenau F.. **Polyclonal Aptamers for Specific Fluorescence Labeling and Quantification of the Health Relevant Human Gut Bacterium**. *Microorganisms* (2021) **9**. DOI: 10.3390/microorganisms9112284
55. Kneißle K., Krämer M., Kissmann A.-K., Xing H., Müller F., Amann V., Noschka R., Gottschalk K.-E., Bozdogan A., Andersson J.. **A Polyclonal SELEX Aptamer Library Allows Differentiation of**. *J. Fungi* (2022) **8**. DOI: 10.3390/jof8080856
56. Zheng X., Gao S., Wu J., Hu X.. **Recent Advances in Aptamer-Based Biosensors for Detection of**. *Front. Microbiol.* (2020) **11** 605229. DOI: 10.3389/fmicb.2020.605229
57. López-Abarrategui C., McBeth C., Mandal S.M., Sun Z.J., Heffron G., Alba-Menéndez A., Migliolo L., Reyes-Acosta O., García-Villarino M., Nolasco D.O.. **Cm-P5: An Antifungal Hydrophilic Peptide Derived from the Coastal Mollusk**. *FASEB J.* (2015) **29** 3315-3325. DOI: 10.1096/fj.14-269860
58. Kissmann A.-K., Andersson J., Bozdogan A., Amann V., Kraemer M., Xing H., Raber H., Kubiczek D.H., Aspermair P., Knoll W.. **Polyclonal Aptamer Libraries as Binding Entities on a Graphene FET Based Biosensor for the Discrimination of Apo- and Holo- Retinol Binding Protein 4**. *Nanoscale Horiz.* (2022) **7** 770-778. DOI: 10.1039/D1NH00605C
59. Aspermair P., Mishyn V., Bintinger J., Happy H., Bagga K., Subramanian P., Knoll W., Boukherroub R., Szunerits S.. **Reduced Graphene Oxide—Based Field Effect Transistors for the Detection of E7 Protein of Human Papillomavirus in Saliva**. *Anal. Bioanal. Chem.* (2020) **413** 7-11. DOI: 10.1007/s00216-020-02879-z
60. Aspermair P., Ramach U., Reiner-Rozman C., Fossati S., Lechner B., Moya S.E., Azzaroni O., Dostalek J., Szunerits S., Knoll W.. **Dual Monitoring of Surface Reactions in Real Time by Combined Surface-Plasmon Resonance and Field-Effect Transistor Interrogation**. *J. Am. Chem. Soc.* (2020) **142** 11709-11716. DOI: 10.1021/jacs.9b11835
61. Alfieri M.L., Weil T., Ng D.Y.W., Ball V.. **Polydopamine at Biological Interfaces**. *Adv. Colloid Interface Sci.* (2022) **305** 102689. DOI: 10.1016/j.cis.2022.102689
62. Yazdi M.K., Zare M., Khodadadi A., Seidi F., Sajadi S.M., Zarrintaj P., Arefi A., Saeb M.R., Mozafari M.. **Polydopamine Biomaterials for Skin Regeneration**. *ACS Biomater. Sci. Eng.* (2022) **8** 2196-2219. DOI: 10.1021/acsbiomaterials.1c01436
63. Schindelin J., Arganda-Carreras I., Frise E., Kaynig V., Longair M., Pietzsch T., Preibisch S., Rueden C., Saalfeld S., Schmid B.. **Fiji: An Open-Source Platform for Biological-Image Analysis**. *Nat. Methods* (2012) **9** 676-682. DOI: 10.1038/nmeth.2019
64. Mann H.B., Whitney D.R.. **On a Test of Whether One of Two Random Variables Is Stochastically Larger than the Other**. *Ann. Math. Stat.* (1947) **18** 50-60. DOI: 10.1214/aoms/1177730491
|
---
title: 'Association between spironolactone use and COVID-19 outcomes in population-scale
claims data: a retrospective cohort study'
authors:
- Henry C. Cousins
- Russ B. Altman
journal: medRxiv
year: 2023
pmcid: PMC10002773
doi: 10.1101/2023.02.28.23286515
license: CC BY 4.0
---
# Association between spironolactone use and COVID-19 outcomes in population-scale claims data: a retrospective cohort study
## Abstract
### Background:
Spironolactone has been proposed as a potential modulator of SARS-CoV-2 cellular entry. We aimed to measure the effect of spironolactone use on the risk of adverse outcomes following COVID-19 hospitalization.
### Methods:
We performed a retrospective cohort study of COVID-19 outcomes for patients with or without exposure to spironolactone, using population-scale claims data from the Komodo Healthcare Map. We identified all patients with a hospital admission for COVID-19 in the study window, defining treatment status based on spironolactone prescription orders. The primary outcomes were progression to respiratory ventilation or mortality during the hospitalization. Odds ratios (OR) were estimated following either 1:1 propensity score matching (PSM) or multivariable regression. Subgroup analysis was performed based on age, gender, body mass index (BMI), and dominant SARS-CoV-2 variant.
### Findings:
Among 898,303 eligible patients with a COVID-19-related hospitalization, 16,324 patients ($1.8\%$) had a spironolactone prescription prior to hospitalization. 59,937 patients ($6.7\%$) met the ventilation endpoint, and 26,515 patients ($3.0\%$) met the mortality endpoint. Spironolactone use was associated with a significant reduction in odds of both ventilation (OR 0.82; $95\%$ CI: 0.75-0.88; $p \leq 0.001$) and mortality (OR 0.88; $95\%$ CI: 0.78-0.99; $$p \leq 0.033$$) in the PSM analysis, supported by the regression analysis. Spironolactone use was associated with significantly reduced odds of ventilation for all age groups, men, women, and non-obese patients, with the greatest protective effects in younger patients, men, and non-obese patients.
### Interpretation:
Spironolactone use was associated with a protective effect against ventilation and mortality following COVID-19 infection, amounting to up to $64\%$ of the protective effect of vaccination against ventilation and consistent with an androgen-dependent mechanism. The findings warrant initiation of large-scale randomized controlled trials to establish a potential therapeutic role for spironolactone in COVID-19 patients.
## INTRODUCTION
The continued proliferation of vaccine-evading severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) strains has reinforced the need for outpatient treatments to mitigate the clinical course of coronavirus disease 2019 (COVID-19).1 *While a* small number of antiviral therapies have received Food and Drug Administration approval in COVID-19, such treatments remain limited in both adoption and efficacy, owing to concerns about adverse reactions, drug-drug interactions, and cost.2,3 Consequently, it remains critical to identify any existing medications that may modulate the course of infection.
The potassium-sparing diuretic spironolactone has been proposed as a potential modulator of SARS-CoV-2 infection due to its interactions with multiple COVID-19-associated signaling pathways.4 Spironolactone functions chiefly as a mineralocorticoid receptor blocker, antagonizing the final stage of the renin-angiotensin-aldosterone system (RAAS).5 Given the involvement of angiotensin-converting enzyme 2 (ACE2), the canonical host receptor for SARS-CoV-2, in RAAS activity, mineralocorticoid antagonists have been hypothesized to alter ACE2 expression, which has been observed in vitro.6,7 In addition to its anti-mineralocorticoid effects, spironolactone is a strong inhibitor of the androgen receptor.8 The critical role of androgen signaling in upregulating TMPRSS2, which facilitates Spike processing during membrane fusion, suggests that spironolactone’s anti-androgenic activity could likewise impede viral entry.9 Existing clinical evidence for a protective role of spironolactone in COVID-19 is encouraging but inconclusive. One case-control study of 6,462 patients with liver cirrhosis in South Korea revealed a significant negative association between spironolactone use and COVID-19 diagnosis.10 A non-randomized, comparative study of bromhexine-spironolactone combination therapy in 103 patients identified a statistically significant $13\%$ reduction in hospitalization time for the intervention group.11 The only published randomized, controlled clinical trial of spironolactone in COVID-19, to our knowledge, was a trial of sitagliptin-spironolactone combination therapy in 263 patients, which suggested a potentially beneficial effect for the intervention group with respect to clinical progression score.12 To determine whether spironolactone use is associated with COVID-19 severity, we conducted the largest clinical investigation of spironolactone in COVID-19 to date. Using health insurance claims data from public and private payers covering over 325 million unique patients, we performed a retrospective cohort study of COVID-19 outcomes for spironolactone users.
## Study design and population
We conducted a retrospective cohort study based on deidentified medical and pharmaceutical data from the Komodo Healthcare Map, a collection of health insurance claims from public and private payers nationwide. The database contains claims data for approximately 325 million unique patients in the United States since October 1, 2015 and is closely aligned with the National Health Interview Survey population in terms of geography and demographics. The dataset encompassed medical claims, pharmacy claims, enrollment records, and mortality records.
We identified 909,531 patients in the database who experienced a hospitalization due to COVID-19 within the study window, spanning March 1, 2020 to June 3, 2022. In the event that a patient experienced multiple COVID-19-related hospitalizations, only the first encounter was considered. Patients under the age of 15 years were also excluded. Medical variables were defined using International Classification of Diseases, 10th Revision, Clinical Modification (ICD-10-CM) codes, procedures were defined using ICD-10 Procedure Coding System (ICD-10-PCS), Current Procedural Technology (CPT), and Healthcare Common Procedure Coding System (HCPCS) codes, and drug prescriptions were defined using National Drug Code (NDC) identifiers. The study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines.
## Ethics committee oversight
The study was declared exempt from institutional review board (IRB) review by the Stanford University IRB.
## Exposures and outcomes
Exposure was defined as a prescription for spironolactone within a 180-day window prior to the COVID-19-related hospitalization claim date.13,14 Only paid prescriptions, implying patient receipt of the medication, were considered, and multi-ingredient drugs were not included. The primary study outcome was progression to ventilation, defined as a claim for a respiratory ventilation procedure during the COVID-19-related hospitalization. We also considered mortality as an additional endpoint, which was defined as death recorded within the period covered by the COVID-19-related hospitalization claim. Time-stationarity of outcome variables was measured by Pearson correlation between endpoint probability and month of admission, beginning April 2020.
## Study variables
The study design controlled for demographic, medical, and pharmaceutical covariates. Demographic information included age as a continuous variable and gender. Medical covariates included body mass index (BMI ≥30 kg/m2 or <30 kg/m2), myocardial infarction, congestive heart failure, peripheral vascular disease, dementia, chronic pulmonary disease, rheumatic disease, peptic ulcer disease, mild liver disease, moderate or severe liver disease, diabetes without chronic complications, diabetes with chronic complications, hemiplegia or paraplegia, renal disease, malignancy, metastatic solid tumor, and human immunodeficiency virus (HIV) or acquired immunodeficiency syndrome (AIDS). For each condition, corresponding ICD-10-CM codes were obtained from the updated Charlson Comorbidity Index definitions (Supplementary Table S1).15 Pharmaceutical covariates included a paid prescription within a 180-day window prior to the COVID-19-related hospitalization for atorvastatin, levothyroxine, metformin, lisinopril, amlodipine, metoprolol, albuterol, omeprazole, losartan, or gabapentin. COVID-19 vaccination status was an additional covariate, defined as receipt of at least one dose of any COVID-19 vaccine prior to the COVID-19-related hospitalization. Values are reported as median (interquartile range [IQR]) for continuous variables and frequency (percent) for categorical variables.
## Propensity score matching
We performed propensity score matching (PSM) to obtain matched pairs of drug-exposed and non-drug-exposed patients.16 Propensity scores were derived by fitting a logistic regression model with L2 regularization to predict drug exposure status using all study covariates, normalized to unity. Nearest-neighbor matching on propensity scores was performed without a caliper to generate 1:1 matched pairs. Covariate balance between drug-exposed and non-drug-exposed groups was assessed by calculating the standardized bias for each covariate, with a standardized bias of less than 0.1 considered balanced (Supplementary Table S2).17 Odds ratios (OR) between drug exposure and each outcome of interest, as well as corresponding $95\%$ confidence intervals ($95\%$ CI) and p-values, were calculated using McNemar’s exact test.
## Regression model
We also fit a second model to control for covariates without matching. We fit an L1-regularized logistic regression (LR) model using the same explanatory variables as in the propensity score derivation, with the addition of drug exposure status. OR were calculated from coefficients of the fitted model, and confidence intervals and p-values for each OR were calculated from the corresponding t statistics.
## Subgroup and sensitivity analysis
We measured treatment effects in patient subgroups, grouping by male gender, female gender, obesity (BMI≥30 kg/m2), non-obesity (BMI<30 kg/m2), and age brackets (<60, 60-74, and ≥75 years). Additionally, we analyzed cases in time periods with predominance of specific variants in the US, including the Delta (July 1, 2021 to December 20, 2021) and Omicron (December 20, 2021 to June 3, 2022) strains.18 We ran additional sensitivity analyses considering alternate windows of drug exposure (90 days and 360 days).
## Computational resources
Bulk data queries were performed using Structured Query Language (SQL) in a Snowflake workspace (Snowflake Inc., Bozeman, MT). All statistical analyses were performed in a Python 3.10 environment using the scikit-learn (version 1.1.2), statsmodels (version 0.13.2), psmpy (version 0.3.5), NumPy (version 1.23.2), and pandas (version 1.4.3) packages.
## Role of the funding source
No study sponsor had any role in the design of the study; in the collection, analysis, or interpretation of the data; in the composition of the manuscript; or in the decision to submit the manuscript for publication.
## RESULTS
From the database, we identified 909,531 patients with a COVID-19-related hospitalization, of whom 11,228 patients were excluded due to age below the study minimum ($$n = 11$$,206) or missing information for gender ($$n = 22$$; Figure 1). Within the final study population of 898,303, the treatment group comprised 16,324 patients with a fulfilled prescription for spironolactone prior to hospitalization.
The study population had a median age of 64.9 (IQR 56.5-78.6) years, and 465,124 ($51.8\%$) were women. 59,937 patients ($6.7\%$) met the ventilation endpoint, and 26,515 patients ($3.0\%$) met the mortality endpoint. Endpoints were time-stationary, with no significant correlation between event frequency and claim month over the study window ($$p \leq 0.338$$ for ventilation; $$p \leq 0.248$$ for mortality). Following propensity score matching, all covariates were well balanced between treatment and control groups, with standardized biases of less than 0.1 in all cases (Table I).
Following 1:1 propensity score matching, 1,212 of 16,324 patients ($7.4\%$) in the spironolactone treatment group met the ventilation endpoint in aggregate, while 1,459 of 16,324 patients ($8.9\%$) in the control group met the endpoint (Table II). In the paired analysis, the OR for ventilation between treatment and controls was 0.82 ($95\%$ CI: 0.75-0.88; $p \leq 0.001$). The unmatched logistic regression analysis supported this protective effect, finding an OR of 0.78 ($95\%$ CI: 0.73-0.83; $p \leq 0.001$).
Spironolactone treatment was also associated with a protective effect for mortality, with 521 of 16,324 patients ($3.2\%$) in the treatment group and 592 of 16,324 patients ($3.6\%$) in the matched control group. In the paired analysis, this corresponded to an OR of 0.88 ($95\%$ CI: 0.78-0.99; $$p \leq 0.033$$), which was supported by similar findings in the regression analysis (OR 0.85; $95\%$ CI: 0.78-0.93; $p \leq 0.001$).
As a study-level control, we also measured the effect of vaccination on both endpoints, finding strongly protective effects in all cases. For ventilation, the OR for vaccination was 0.72 ($95\%$ CI: 0.69-0.76; $p \leq 0.001$) in the paired analysis and 0.68 ($95\%$ CI: 0.66-0.71; $p \leq 0.001$) in the regression analysis. For mortality, vaccination was associated with OR values of 0.62 ($95\%$ CI: 0.58-0.67; $p \leq 0.001$) and 0.61 ($95\%$ CI: 0.57-0.64; $p \leq 0.001$) in the paired and regression analyses, respectively. These findings are consistent with previous estimates of the protective effect of vaccination in hospitalized COVID-19 patients.19,20 We next analyzed treatment effects in predefined patient subgroups (Table III). For the ventilation endpoint, we observed a stronger protective treatment effect in men than in the general population, with an OR of 0.76 ($95\%$ CI: 0.67-0.85; $p \leq 0.001$) in the matched analysis and 0.72 ($95\%$ CI: 0.66-0.79; $p \leq 0.001$) in the regression analysis. The effect in women was weaker than in the general population, with estimated OR values of 0.87 ($95\%$ CI: 0.77-0.97; $$p \leq 0.012$$) and 0.84 ($95\%$ CI: 0.77-0.92; $p \leq 0.001$) in the matched and regression analyses, respectively. In non-high-BMI patients, treatment was associated with a more protective effect than in the general population, with an OR of 0.81 ($95\%$ CI: 0.73-0.89; $p \leq 0.001$) in the matched analysis and 0.74 ($95\%$ CI: 0.69-0.80) in the regression analysis. The estimated treatment effect was greatly reduced in high-BMI patients and did not meet significance in the matched analysis (OR 0.93; $95\%$ CI: 0.81-1.07; $$p \leq 0.352$$), although nominally significant in the regression analysis (OR 0.89; $95\%$ CI: 0.80-0.99; $$p \leq 0.026$$). Treatment was associated with a protective effect in all age brackets. However, it was most protective in the youngest bracket (<60 years old) and least protective in the oldest bracket (≥75 years old), with estimated OR of 0.78 ($95\%$ CI: 0.67-0.91; $$p \leq 0.002$$) and 0.81 ($95\%$ CI: 0.67-0.97; $$p \leq 0.028$$), respectively, in the paired analysis. The protective effect was also diminished for hospitalizations during the Omicron wave, with an OR of 0.91 ($95\%$ CI: 0.74-1.11; $$p \leq 0.37$$) in the paired analysis and 0.80 ($95\%$ CI: 0.69-0.93; $$p \leq 0.004$$) in the regression analysis.
The lower frequency of events for the mortality endpoint limited our ability to detect significant effects in subgroups for this endpoint. In the subgroup analysis for mortality (Supplementary Table S3), only males (OR 0.83; $95\%$ CI: 0.70-0.98; $$p \leq 0.029$$) and the 60-74 age bracket (OR 0.80; $95\%$ CI: 0.67-0.96; $$p \leq 0.017$$) met significance in the PSM analysis.
We also conducted sensitivity analyses to assess the effect of parameter selection in our analysis (Table IV). For the ventilation endpoint, treatment remained significantly associated with improved outcomes in all sensitivity analyses, including using a 90-day window (OR 0.82; $95\%$ CI: 0.75-0.90; $p \leq 0.001$) and a 360-day window (OR 0.89; $95\%$ CI: 0.83-0.95; $$p \leq 0.001$$) for drug exposure. For the mortality endpoint, significantly protective treatment effects were likewise observed for both the 90-day (OR 0.81; $95\%$ CI: 0.70-0.93; $$p \leq 0.002$$) and 360-day (OR 0.85; $95\%$ CI: 0.76-0.94; $$p \leq 0.003$$) windows.
## DISCUSSION
Our results, supported by the largest cohort study of spironolactone in COVID-19 to date, suggest that spironolactone may improve outcomes in patients hospitalized with COVID-19. In our study, spironolactone use was associated with an $18\%$ reduction in odds of ventilation following admission for COVID-19. This effect was more pronounced in men and in younger patients (15-59 years old), where the effects corresponded to a $22\%$ and $24\%$ reduction in ventilation odds, respectively. In contrast, high BMI (≥30 kg/m2) diminished the observed treatment effect. Spironolactone use was also associated with a significant $12\%$ reduction in odds of mortality. In our analysis, the protective effect of spironolactone amounted to $64\%$ and $32\%$ of the protective effect of COVID-19 vaccination against ventilation and mortality, respectively.
Several previous studies have investigated a possible protective effect for spironolactone in COVID-19, with encouraging but underpowered results. The strongest evidence, prior to our study, was a randomized controlled trial of spironolactone-sitagliptin combination therapy in 263 patients, which demonstrated a significant improvement in a subjective clinical progression score.12 While rates of mortality, intensive care unit admission, intubation, and end-organ damage were reduced in this trial, the effect did not meet statistical significance. A non-randomized trial of another combination therapy, spironolactone with bromhexine, reported faster time to temperature normalization and hospital discharge in the treatment group.11 The largest study prior to this work was a case-control study of 6,462 patients that identified an $80\%$ reduction in odds of spironolactone exposure in COVID-19 patients compared to matched controls, although this study was restricted to patients with liver cirrhosis.10 Spironolactone’s dual role as a RAAS modulator and androgen antagonist provides several plausible mechanisms for inhibition of viral entry. While clinical investigations have not demonstrated a clear relationship between RAAS modulators, such as ACE inhibitors and angiotensin II receptor blockers (ARBs), and clinical outcomes in COVID-19, there is extensive evidence that androgen signaling plays a meaningful role in viral entry.9,21-23 For instance, anti-androgenic drugs have been associated with protective effects in observational studies, and androgen antagonism inhibits SARS-CoV-2 cellular entry in vitro.22,24,25 Intriguingly, our study noted a consistent relationship between effect size and androgen levels in patient subgroups, consistent with a protective effect of spironolactone mediated by inhibition of androgen signaling. For instance, the observed effects were stronger in male, non-obese, and younger patients, all of whom tend to have higher androgen levels than their demographic counterparts.26-28 We also observed a smaller protective effect in hospitalizations during the predominant time period of the Omicron variant, which has been observed to rely less on androgen-dependent pathways for cellular entry.29 While we did not have access to laboratory data to measure a potential relationship between spironolactone effect and androgen levels directly, our results are consistent with such an association.
Our study has several limitations. Although we employ well-characterized causal inference methods like propensity score matching, the observational nature of the study precludes direct causal reasoning. Furthermore, while we control for a large collection of medical and pharmacologic covariates, unmodeled confounders may still exist that could limit the generalizability of our results. Furthermore, claims data only captures information regarding prescription issue and fulfillment and does not guarantee adherence to treatment in our control group. Claims data may also be more vulnerable to incomplete or changing coding practices than institutional medical records data. We were also unable to measure dose response in our dataset, although spironolactone dose variability is generally low.30 In conclusion, we show that spironolactone use is associated with improved outcomes following COVID-19 hospitalization in a nationwide cohort of nearly a million hospitalized patients. Treatment was associated with lower odds of ventilation and mortality compared to matched and unmatched controls. Furthermore, the protective effect on ventilation in patient subgroups was consistent with an androgen-dependent mechanism. Our findings support the initiation of well-powered randomized controlled trials to determine the clinical value of spironolactone in the treatment of COVID-19.
## References
1. Murakami N, Hayden R, Hills T. **Therapeutic advances in COVID-19**. *Nat Rev Nephrol 2022 191* (2022) **19** 38-52
2. Ranganath N, O’Horo JC, Challener DW. **Rebound Phenomenon after Nirmatrelvir/Ritonavir Treatment of Coronavirus Disease-2019 in High-Risk Persons**. *Clin Infect Dis* (2022) **76** e537-9
3. Marzolini C, Kuritzkes DR, Marra F. **Prescribing Nirmatrelvir–Ritonavir: How to Recognize and Manage Drug–Drug Interactions**. *Ann Intern Med* (2022) **175** 744-6. PMID: 35226530
4. Cadegiani FA, Goren A, Wambier CG. **Spironolactone may provide protection from SARS-CoV-2: Targeting androgens, angiotensin converting enzyme 2 (ACE2), and renin-angiotensin-aldosterone system (RAAS)**. *Med Hypotheses* (2020) **143**. DOI: 10.1016/j.mehy.2020.110112
5. Sica DA. **Pharmacokinetics and Pharmacodynamics of Mineralocorticoid Blocking Agents and their Effects on Potassium Homeostasis**. *Heart Fail Rev* (2005) **10** 23-9. PMID: 15947888
6. Kotfis K, Lechowicz K, Drożdżal S. **COVID-19-The Potential Beneficial Therapeutic Effects of Spironolactone during SARS-CoV-2 Infection**. *Pharmaceuticals (Basel)* (2021) **14** 1-14
7. Dong D, Fan T, Ji Y, Yu J, Wu S, Zhang L. **Spironolactone alleviates diabetic nephropathy through promoting autophagy in podocytes**. *Int Urol Nephrol* (2019) **51** 755-64. PMID: 30734886
8. Cadegiani FA, Wambier CG, Goren A. **Spironolactone: An Anti-androgenic and Anti-hypertensive Drug That May Provide Protection Against the Novel Coronavirus (SARS-CoV-2) Induced Acute Respiratory Distress Syndrome (ARDS) in COVID-19**. *Front Med* (2020) **7** 453
9. Samuel RM, Majd H, Richter MN. **Androgen Signaling Regulates SARS-CoV-2 Receptor Levels and Is Associated with Severe COVID-19 Symptoms in Men**. *Cell Stem Cell* (2020) **27** 876-889.e12. PMID: 33232663
10. Jeon D, Son M, Choi J. **Effect of Spironolactone on COVID-19 in Patients With Underlying Liver Cirrhosis: A Nationwide Case-Control Study in South Korea**. *Front Med* (2021) **8** 629176
11. Mareev VY, Orlova YA, Plisyk AG. **Results of open-label non-randomized comparative clinical trial: “Bromhexine and spironolactone for coronavirus infection requiring hospitalization (BISCUIT)**. *Kardiologiya* (2020) **60** 4-15
12. Abbasi F, Adatorwovor R, Davarpanah MA. **A Randomized Trial of Sitagliptin and Spironolactone With Combination Therapy in Hospitalized Adults With COVID-19**. *J Endocr Soc* (2022) **6** 1-11
13. Pottegård A, Hallas J. **Assigning exposure duration to single prescriptions by use of the waiting time distribution**. *Pharmacoepidemiol Drug Saf* (2013) **22** 803-9. PMID: 23703741
14. Tanskanen A, Taipale H, Koponen M. **Drug exposure in register-based research—An expert-opinion based evaluation of methods**. *PLoS One* (2017) **12** e0184070. PMID: 28886089
15. Quan H, Sundararajan V, Halfon P. **Coding algorithms for defining comorbidities in ICD-9-CM and ICD-10 administrative data**. *Med Care* (2005) **43** 1130-9. PMID: 16224307
16. Austin PC. **A critical appraisal of propensity-score matching in the medical literature between 1996 and 2003**. *Stat Med* (2008) **27** 2037-49. PMID: 18038446
17. Harder VS, Stuart EA, Anthony JC. **Propensity score techniques and the assessment of measured covariate balance to test causal associations in psychological research**. *Psychol Methods* (2010) **15** 234. PMID: 20822250
18. Lambrou AS, Shirk P, Steele MK. **Genomic Surveillance for SARS-CoV-2 Variants: Predominance of the Delta (B.1.617.2) and Omicron (B.1.1.529) Variants — United States, June 2021–January 2022**. *MMWR Morb Mortal Wkly Rep* (2022) **71** 206-11. PMID: 35143464
19. Baker TB, Bolt DM, Smith SS. **The Relationship of COVID-19 Vaccination with Mortality Among 86,732 Hospitalized Patients: Subpopulations, Patient Factors, and Changes over Time**. *J Gen Intern Med* (2023) 1
20. Balachandran S, Moni M, Sathyapalan DT. **A comparison of clinical outcomes between vaccinated and vaccine-naive patients of COVID-19, in four tertiary care hospitals of Kerala, South India**. *Clin Epidemiol Glob Heal* (2022) **13** 100971
21. Mancia G, Rea F, Ludergnani M, Apolone G, Corrao G. **Renin–Angiotensin–Aldosterone System Blockers and the Risk of Covid-19**. *N Engl J Med* (2020) **382** 2431-40. PMID: 32356627
22. Montopoli M, Zumerle S, Vettor R. **Androgen-deprivation therapies for prostate cancer and risk of infection by SARS-CoV-2: a population-based study (N = 4532)**. *Ann Oncol* (2020) **31** 1040-5. PMID: 32387456
23. Qiao Y, Wang XM, Mannan R. **Targeting transcriptional regulation of SARS-CoV-2 entry factors ACE2 and TMPRSS2**. *Proc Natl Acad Sci U S A* (2020) **118** e2021450118. PMID: 33310900
24. Meng J, Ge Q, Li J. **Protective trend of anti-androgen therapy during the COVID-19 pandemic: A meta-analysis**. *J Infect* (2022) **84** 834
25. Leach DA, Mohr A, Giotis ES. **The antiandrogen enzalutamide downregulates TMPRSS2 and reduces cellular entry of SARS-CoV-2 in human lung cells**. *Nat Commun* (2021) **12** 1-12. PMID: 33397941
26. Southren AL, Tochimoto S, Carmody NC, Isurugi K. **Plasma production rates of testosterone in normal adult men and women and in patients with the syndrome of feminizing testes**. *J Clin Endocrinol Metab* (1965) **25** 1441-50. PMID: 5843701
27. Giagulli VA, Kaufman JM, Vermeulen A. **Pathogenesis of the decreased androgen levels in obese men**. *J Clin Endocrinol Metab* (1994) **79** 997-1000. PMID: 7962311
28. Travison TG, Vesper HW, Orwoll E. **Harmonized Reference Ranges for Circulating Testosterone Levels in Men of Four Cohort Studies in the United States and Europe**. *J Clin Endocrinol Metab* (2017) **102** 1161-73. PMID: 28324103
29. Meng B, Abdullahi A, Ferreira IATM. **Altered TMPRSS2 usage by SARS-CoV-2 Omicron impacts infectivity and fusogenicity**. *Nature* (2022) **603** 706-14. PMID: 35104837
30. Ferreira JP, Rossello X, Pocock SJ. **Spironolactone dose in heart failure with preserved ejection fraction: findings from TOPCAT**. *Eur J Heart Fail* (2020) **22** 1615-24. PMID: 32452128
|
---
title: Apelin Enhances the Effects of Fusobacterium nucleatum on Periodontal Ligament
Cells In Vitro
authors:
- Pablo Cores Ziskoven
- Andressa V. B. Nogueira
- Lorena S. Gutierrez
- Jens Weusmann
- Sigrun Eick
- Nurcan Buduneli
- James Deschner
journal: International Journal of Molecular Sciences
year: 2023
pmcid: PMC10002786
doi: 10.3390/ijms24054733
license: CC BY 4.0
---
# Apelin Enhances the Effects of Fusobacterium nucleatum on Periodontal Ligament Cells In Vitro
## Abstract
This study aimed to explore effects of *Fusobacterium nucleatum* with or without apelin on periodontal ligament (PDL) cells to better understand pathomechanistic links between periodontitis and obesity. First, the actions of F. nucleatum on COX2, CCL2, and MMP1 expressions were assessed. Subsequently, PDL cells were incubated with F. nucleatum in the presence and absence of apelin to study the modulatory effects of this adipokine on molecules related to inflammation and hard and soft tissue turnover. Regulation of apelin and its receptor (APJ) by F. nucleatum was also studied. F. nucleatum resulted in elevated COX2, CCL2, and MMP1 expressions in a dose- and time-dependent manner. Combination of F. nucleatum and apelin led to the highest ($p \leq 0.05$) expression levels of COX2, CCL2, CXCL8, TNF-α, and MMP1 at 48 h. The effects of F. nucleatum and/or apelin on CCL2 and MMP1 were MEK$\frac{1}{2}$- and partially NF-κB-dependent. The combined effects of F. nucleatum and apelin on CCL2 and MMP1 were also observed at protein level. Moreover, F. nucleatum downregulated ($p \leq 0.05$) the apelin and APJ expressions. In conclusion, obesity could contribute to periodontitis through apelin. The local production of apelin/APJ in PDL cells also suggests a role of these molecules in the pathogenesis of periodontitis.
## 1. Introduction
Periodontitis is a chronic inflammatory disease mainly caused by a subgingival dysbiotic microbiota whose balance is shifted by several factors [1]. Additionally, there is also a dysbiotic status between the host and the subgingival microbiota in periodontitis. Hyperinflammatory immune responses of the host to this microbiota can lead to alveolar bone resorption and eventually tooth loss [1]. Risk factors such as smoking or genetic predisposition can contribute to the initiation and progression of periodontitis [2]. There is strong evidence that periodontitis is associated with systemic diseases and conditions, such as diabetes mellitus, cardiovascular disease, hypertension, obesity, and metabolic syndrome. It is thought that the oral microorganisms, their components, or metabolites as well as inflammatory mediators get into the systemic circulation and therefore to other parts of the human body [3,4,5,6,7,8]. Obesity is defined as abnormal or excessive fat accumulation that presents a risk to health [9]. Because adipose tissue is not only an energy reservoir but also a metabolic organ, dysregulation of cytokines, hormones, and metabolites occurs when this tissue increases [10]. There is evidence that obese individuals have systemically high levels of CRP, TNF-α, and IL-6 in comparison to normal-weight subjects and, therefore, are in a chronic subclinical inflammatory state [11]. A lot of possible pathomechanisms have been suggested to be responsible for the link between periodontitis and obesity, such as adipokines [12]. Adipokines are cytokines produced by adipocytes, but also by other cell types, such as periodontal cells [13,14,15,16,17,18]. Various adipokines such as leptin, visfatin, adiponectin, and resistin have been identified and studied in regard to systemic diseases. It is suggested that these adipokines have a wide range of functions, which include regulation of insulin metabolism, thirst and hunger sensation, angiogenesis, energy balance, bone metabolism, coagulation, and hematopoiesis, as well as inflammation and its resolution [13,19]. Adiponectin has mainly anti-inflammatory effects, whereas resistin, visfatin, and leptin are more pro-inflammatory [20,21]. Another adipokine, which has been rather less studied so far, is apelin. Apelin was first isolated and described in 1998 [22]. As early as 1993, the apelin receptor (angiotensin II protein J receptor, APJ) had been discovered in humans as a G protein-coupled receptor whose gene locus is located on chromosome 11 [23]. Apelin has a wide range of effects, which differ depending on cell types and tissues. Originally, apelin was isolated from tissues of the central nervous system. Accordingly, the molecule was found to be important in central signal transduction [24]. As research progressed, the apelin-APJ system was discovered in other tissues as well. For example, the molecule interferes with the regulation of bone turnover by modulating apoptosis, proliferation, and differentiation of osteoblasts [25,26]. It has been shown that apelin levels are increased in systemic diseases and conditions such as obesity and diabetes [27,28]. A recent study looked at serum levels of apelin in diabetes and/or periodontitis patients [29]. Those patients who suffered from both diabetes and periodontitis exhibited the highest serum levels of apelin as compared to healthy individuals. Another study could show that the salivary apelin levels of diabetic patients with periodontitis were increased as compared to healthy individuals [30]. This adipokine also has modulatory properties regarding inflammation. For example, apelin can increase the expression of TNF-α and IL-1β in glial cells, but at the same time downregulate inflammatory mediators in lung and heart cells [31,32]. Therefore, apelin could be a critical molecule, which may mediate the harmful effects of obesity on periodontal tissues. The aim of this in vitro study was to explore the regulatory effects of *Fusobacterium nucleatum* in the presence or absence of apelin on periodontal ligament (PDL) cells in order to test the hypothesis that apelin might be one of the pathomechanistic links between periodontal disease and obesity.
## 2.1. Regulation of COX2, CCL2, and MMP1 Expressions by F. nucleatum
First, we wanted to verify whether F. nucleatum would regulate the expression of COX2, CCL2, and MMP1 in PDL cells. F. nucleatum caused a significant ($p \leq 0.05$) and dose-dependent (O.D.660: 0.000, 0.025, 0.050, and 0.100) upregulation of the pro-inflammatory and proteolytic molecules COX2, CCL2, and MMP1 with the highest expression for the highest bacterial concentration (O.D.660 = 0.100) at 24 h (Figure 1a). In addition, the stimulatory effect of F. nucleatum (O.D.660 = 0.025) on these molecules was also time-dependent ($p \leq 0.05$), as shown in Figure 1b.
## 2.2. Modulatory Effects of Apelin on Pro-Inflammatory Actions by F. nucleatum
Next, we studied whether apelin (1 ng/mL) could modulate the stimulatory actions of F. nucleatum (O.D.660 = 0.025) on the expression of pro-inflammatory markers in PDL cells. Apelin was used at a concentration corresponding to physiological plasma levels and consistent with previous in vitro studies. For F. nucleatum, O.D.660 = 0.025 was chosen because even this minimal dose had a proinflammatory effect on PDL cells, as evidenced by a significant increase in the expression of COX2, CCL2, and MMP1. As shown by real-time PCR analysis, apelin significantly ($p \leq 0.05$) increased the F. nucleatum-stimulated expression of CCL2 at 24 h (Figure 2a). For COX2, CXCL-8, and TNF-α, no significant modulatory effect of apelin on the F. nucleatum-triggered expression was observed at this time point (Figure 2a). Moreover, apelin caused a further significant ($p \leq 0.05$) elevation of the F. nucleatum-induced expressions of COX2, CCL2, CXCL-8, and TNF-α at 48 h (Figure 2b). This shows that the stimulatory influence of apelin on the effects of F. nucleatum was stronger at 48 h as compared to 24 h.
## 2.3. Modulatory Effects of Apelin on Markers Involved in Soft and Hard Tissue Turnover
We then examined the effect of apelin (1 ng/mL) on the regulation of MMP1, TGF-β1, and RUNX2 by F. nucleatum (O.D.660 = 0.025) in PDL cells (Figure 3). F. nucleatum increased the expression of MMP1 at 24 h (Figure 3a) and 48 h (Figure 3b), and this upregulation was significantly ($p \leq 0.05$) enhanced by apelin at both time points. No upregulation by F. nucleatum was observed for TGF-β1 and RUNX2 at 24 h (Figure 3a) and 48 h (Figure 3b). Apelin had no significant effect on the actions of F. nucleatum on TGF-β1 at 24 h (Figure 3a) and 48 h (Figure 3b) and RUNX2 at 48 h (Figure 3b). Interestingly, apelin significantly ($p \leq 0.05$) counteracted the inhibitory effect of F. nucleatum on RUNX2 expression at 24 h (Figure 3a).
## 2.4. Involvement of Signaling Pathways in the Modulatory Effects of F. nucleatum and/or Apelin on CCL2 and MMP1 Expressions
We next sought to identify intracellular signaling pathways potentially involved in the actions of F. nucleatum on CCL2 and MMP1 in PDL cells. For this purpose, cells were pre-incubated with specific inhibitors for NF-κB or MEK$\frac{1}{2}$ signaling and subsequently stimulated with F. nucleatum (O.D.660 = 0.025) and/or apelin (1 ng/mL). Pre-incubation of cells with an NF-κB inhibitor resulted in a significant ($p \leq 0.05$) downregulation of the CCL2 expression in cells treated with either F. nucleatum alone or in combination with apelin at 24 h (Figure 4a). In contrast, the expressions of CCL2 and MMP1 induced by F. nucleatum and/or apelin were always significantly ($p \leq 0.05$) inhibited by the MEK$\frac{1}{2}$ inhibitor after 24 h (Figure 4a,b).
## 2.5. Effects of F. nucleatum on Apelin and Its Receptor
We also investigated whether apelin is expressed in PDL cells and, if so, whether this adipokine as well as its receptor are regulated by F. nucleatum (O.D.660 = 0.025). The periodontopathogen downregulated ($p \leq 0.05$) the expression of apelin and APJ over a variety of doses (Figure 5a). A slight time dependence was observed (Figure 5b).
## 2.6. Modulatory Effects of Apelin on CCL2 and MMP1 Protein Induced by F. nucleatum
Finally, we investigated whether apelin (1 ng/mL) can modulate the stimulatory effect of F. nucleatum (O.D.660 = 0.025) on pro-inflammatory markers also at protein level in PDL cells. As detected by ELISA, F. nucleatum resulted in increased protein levels of CCL2 and MMP1 in cell supernatants at 24 h and 48 h (Figure 6a,b). Incubation of F. nucleatum-stimulated cells with apelin resulted in a further significant ($p \leq 0.05$) increase in protein levels of CCL2 at 48 h (Figure 6a) and of MMP1 at 24 h and 48 h (Figure 6b).
## 3. Discussion
This study aimed to investigate the modulatory effect of the adipokine apelin on the action of the periodontopathogen F. nucleatum on PDL cells to better understand the relationship between periodontitis and obesity. Interestingly, apelin was able to modify bacterial regulation of molecules related to inflammation and hard and soft tissue turnover. The combination of F. nucleatum and apelin resulted in the highest expression levels of pro-inflammatory and proteolytic molecules, suggesting that apelin may be a pathomechanistic link mediating deleterious effects of obesity on periodontal tissues. In addition, F. nucleatum caused downregulation of the expression of apelin and its receptor, suggesting a role of these molecules in the pathogenesis of periodontitis.
There is strong evidence for an association between periodontitis and obesity [12,33]. It has been shown in several studies of our research group that adipokines represent a possible pathomechanistic link underlying the association between periodontitis and obesity [15,16,17,18,34,35,36,37]. Leptin, visfatin, and resistin exert pro-inflammatory effects on periodontal cells and tissues, whereas adiponectin has rather protective effects on periodontal cells [33]. However, with respect to the periodontium, almost nothing is known about the production, regulation, and action of apelin, another adipokine whose serum levels are altered in obesity [28]. Recently, Hirani et al. investigated the serum level of apelin in periodontally and systemically healthy individuals and periodontitis patients with and without type 2 diabetes [29]. The study showed that apelin levels were higher in the periodontitis group compared with the healthy control. When patients had concomitant periodontitis and obesity, apelin levels were highest. The authors concluded that the increased expression of apelin in patients with periodontitis and type 2 diabetes might indicate a possible role of this adipokine in inflammation and glucose regulation. Sarhat et al. examined the salivary apelin levels of periodontally diseased diabetic patients and of periodontally and systemically healthy individuals [30]. They also found the highest apelin levels in periodontitis patients with diabetes. In our study, the periodontopathogen F. nucleatum led to a dose- and time-dependent upregulation of pro-inflammatory and proteolytic molecules. Interestingly, apelin caused an increase in the F. nucleatum-stimulated expression of these pro-inflammatory and proteolytic molecules. In this respect, our in vitro data confirm that apelin may be associated with inflammation. Lee et al. also investigated the relationship between apelin and periodontitis and found a decrease in apelin expression in gingival tissues from periodontitis patients, which is in contrast to the aforementioned studies [38]. Moreover, overexpression of apelin or treatment with exogenous apelin suppressed TNF-α-stimulated gene expressions of MMP1, IL-6, and COX2 in PDL cells [38]. Further studies are needed to clarify whether apelin levels in gingiva, sulcus fluid, saliva, and serum are increased or decreased in gingivitis and periodontitis, and whether apelin exerts pro- or anti-inflammatory effects. In addition, it should be investigated whether periodontal therapy results in a change in these apelin levels.
Furthermore, we were interested in whether apelin and its receptor are expressed in periodontal cells, and if so, whether this expression can be regulated by F. nucleatum. Our in vitro experiments with PDL cells showed that both apelin and its receptor are constitutively produced in these cells. Moreover, our experiments revealed that the periodontopathogen F. nucleatum inhibited the expression of apelin and its receptor. In the study by Lee et al., incubation of PDL cells and gingival fibroblasts with the inflammatory mediator TNF-α also resulted in downregulation of apelin [38]. Therefore, this and our study suggest that the apelin-APJ system is downregulated during periodontal infection and inflammation, at least initially. Because our results suggest that apelin exerts rather pro-inflammatory effects, the initial downregulation of apelin and its receptor may represent the host tissues’ attempt to limit inflammation and associated tissue destruction. However, our experiments also showed that this possibly tissue-protective downregulation of apelin and its receptor was no longer observed after 48 h, which may suggest that in persistent periodontal infection, the apelin-APJ system may be of critical importance in the pathogenesis of periodontitis.
F. nucleatum is an obligate anaerobic gram-negative bacterium very prevalent in the subgingival biofilm and associated with the etiopathogenesis of periodontitis [39,40]. Infection with F. nucleatum alone has shown to cause alveolar bone loss in a murine experimental periodontitis [41]. When in combination with T. forsythia or P. gingivalis, F. nucleatum synergistically stimulated the host immune response and induced alveolar bone loss in this experimental periodontitis model [42,43]. F. nucleatum, such as other red complex bacteria, is associated with periodontitis [44]. As expected according to our previous studies [45,46,47], F. nucleatum led to increased expressions of pro-inflammatory and proteolytic molecules, underlining the special role of this bacterium in periodontal inflammation and destruction. As in our previous studies, F. nucleatum was used as lysate, so several factors may have been responsible for the observed stimulatory effects of F. nucleatum. Studies using live F. nucleatum or even biofilms consisting of a variety of different bacteria should be performed in the future to confirm the results of this study.
Our study clearly demonstrates that apelin can exert pro-inflammatory effects and thus enhance periodontal inflammatory processes. Although there are numerous publications on anti-inflammatory and thus protective effects of apelin [48,49], there are also studies that have demonstrated pro-inflammatory effects of apelin [32,50].
Our analyses regarding intracellular signal transductions suggest that pro-inflammatory effects of F. nucleatum and/or apelin are realized at least partially through MAPK and NF-kB. Further studies should clarify which other intracellular signaling pathways apelin uses for its modulatory effects. Our results are thus in agreement with other studies that have also shown that apelin uses the MAPK and NF-kB signaling pathways, among others, for its effects [27,48,51,52].
In the present study, apelin and APJ were also shown to be produced in periodontal cells and regulated by periodontal pathogenic bacteria, suggesting that apelin and APJ may play an important role in the pathogenesis of periodontitis. Interestingly, F. nucleatum led to downregulation of apelin and its receptor in PDL cells, which would imply an anti-inflammatory effect in accordance with the other results of this study. However, because the inhibitory effect of F. nucleatum was lost with increasing duration of bacterial incubation, this protective effect might also be lacking in persistent periodontal infection. Future studies should also address the apelin-APJ system in other cells of the periodontium, e.g., gingival epithelial cells, and fibroblasts.
The increased production of apelin by periodontal cells after bacterial stimulation suggests that this adipokine is increased in saliva, sulcus fluid, gingiva, and serum during periodontal inflammation. Clinical studies of experimental gingivitis and periodontitis as well as periodontal therapy, i.e., intervention, should further clarify the role of apelin locally in the periodontium but also systemically for the whole organism.
In summary, within its limitations, our in vitro study demonstrated that the adipokine apelin is able to modulate the effects of F. nucleatum on molecules associated with inflammation and hard and soft tissue turnover. Apelin was able to further increase the expression of pro-inflammatory and proteolytic molecules induced by F. nucleatum, which may suggest that apelin may be a pathomechanistic link mediating the deleterious effects of obesity on periodontal tissues. In addition, our study revealed that PDL cells express apelin and APJ and that these expressions are inhibited by F. nucleatum, suggesting a possible role for this adipokine and its receptor in the pathogenesis of periodontitis.
## 4.1. Cell Culture
A human PDL cell line PDL26 was used for cell culture. As described previously, this cell line was obtained from a third molar tooth of a healthy, 26-year-old non-smoking patient [47]. Cells were first cultured in cell culture flasks provided with nutrient medium. The culture medium was Dulbecco’s Modified Eagle Medium (DMEM) GlutaMAX (Invitrogen, Karlsruhe, Germany) supplemented with $10\%$ fetal bovine serum (FBS, Invitrogen). Furthermore, 100 units of penicillin and 100 μg/mL streptomycin (Invitrogen) were added to the medium. Cells were maintained in the incubator at 37 °C and with a humidified atmosphere of $5\%$ CO2. Cells were cultured (1 × 105 cells/well) on 6-well culture plates and grown until 70–$80\%$ confluence. The medium was changed every other day and 24 h before stimulation; the FBS concentration was reduced to $1\%$. The periodontopathogenic bacterium F. nucleatum ATCC 25586 was used at different concentrations (optical density, O.D.660 = 0.025, 0.050, and 0.100) to simulate microbial infection in vitro. The bacterial strain was pre-cultivated on Schaedler agar plates (Oxoid, Basingstoke, UK) in an anaerobic atmosphere for 48 h. Successively, bacteria were suspended in phosphate-buffered saline (O.D.660 = 1, corresponding to 1.2 × 109 bacterial cells/mL) and submitted twice to ultrasonication (160 W for 15 min) leading to total killing. Furthermore, apelin (recombinant human apelin protein, Abcam, Cambridge, United Kingdom) was used for in vitro stimulation at a concentration corresponding to physiological plasma levels (1 ng/mL) and consistent with previous in vitro studies [53,54,55]. In addition, cells were pre-incubated with PDTC (10 µM, Cell Signaling Technology, Danvers, MA, USA), a specific inhibitor of NF-κB, and U0126 (10 µM, Calbiochem, San Diego, CA, USA), a specific inhibitor of MEK$\frac{1}{2}$ signaling. Untreated cells served as control.
## 4.2. Real-Time PCR
RNA isolation was performed using RNeasy Mini Kit (Qiagen, Hilden, Germany) according to the manufacturer’s instructions. To determine the RNA concentration, the spectrophotometer NanoDrop ND-2000 (Thermo Fischer Scientific, Waltham, MA, USA) was used. Five hundred ng of total RNA was reverse transcribed using iScrip Select cDNA Synthesis Kit (Bio-Rad Laboratories, Munich, Germany) according to manufacturer’s protocol. Gene expression analysis of apelin and its receptor (APJ), C-C motif chemokine ligand 2 (CCL2), cyclooxygenase-2 (COX-2), C-X-C motif chemokine ligand 8 (CXCL8), glyceraldehyde-3-phosphate dehydrogenase (GAPDH), matrix metalloproteinase 1 (MMP1), runt-related transcription factor 2 (RUNX2), transforming growth factor-beta 1 (TGF-β1), and tumor necrosis factor alpha (TNF-α), was performed by real-time PCR using the PCR thermal cycler CFX96 (Bio-Rad Laboratories), SYBR green PCR master mix (QuantiFast SYBR Green PCR Kit, Qiagen), and specific primers (QuantiTect Primer Assay, Qiagen). One µL of cDNA was mixed with 12.5 µL master mix, 2.5 µL primer, and 9 µL nuclease-free water. The mix was heated at 95 °C for 5 min, followed by 40 cycles of denaturation at 95 °C for 10 s, and a combined annealing/extension step at 60 °C for 30 s. Data were analyzed by comparative threshold cycle method.
## 4.3. ELISA
The protein levels of CCL2 and MMP1 in the cell supernatants were measured using commercially available ELISA kits (DuoSet, R&D Systems, Minneapolis, MN, USA) according to the manufacturer’s instructions. The optical density was determined using a microplate reader (BioTek Instruments, Winooski, VT, USA) set to 450 nm. The readings at 450 nm were subtracted from the readings at 540 nm for optical correction as per manufacturer’s recommendation. Cell numbers were checked and there was no significant difference between groups.
## 4.4. Statistical Analysis
The statistical analysis was performed using the software GraphPad Prism (version 9.2.0, GraphPad Software, San Diego, CA, USA). For data analysis, mean values and standard errors of the mean (SEM) were calculated. Data were checked for normal distribution and, subsequently, analyzed with the t-test (parametric) or Mann–Whitney-U test (non-parametric). For multiple comparisons, ANOVA or the Kruskall–Wallis test was applied, depending on normal distribution. The Dunnet’s (parametric) or Dunn’s test (non-parametric) served as post hoc tests. The significance level was set at $p \leq 0.05$ for all experiments.
## References
1. Papapanou P.N., Sanz M., Buduneli N., Dietrich T., Feres M., Fine D.H., Flemmig T.F., Garcia R., Giannobile W.V., Graziani F.. **Periodontitis: Consensus Report of Workgroup 2 of the 2017 World Workshop on the Classification of Periodontal and Peri-Implant Diseases and Conditions**. *J. Clin. Periodontol.* (2018.0) **45** S162-S170. DOI: 10.1111/jcpe.12946
2. Tatakis D.N., Kumar P.S.. **Etiology and Pathogenesis of Periodontal Diseases**. *Dent. Clin. N. Am.* (2005.0) **49** 491-516. DOI: 10.1016/j.cden.2005.03.001
3. Stöhr J., Barbaresko J., Neuenschwander M., Schlesinger S.. **Bidirectional Association between Periodontal Disease and Diabetes Mellitus: A Systematic Review and Meta-Analysis of Cohort Studies**. *Sci. Rep.* (2021.0) **11** 13686. DOI: 10.1038/s41598-021-93062-6
4. Sanz M., Marco Del Castillo A., Jepsen S., Gonzalez-Juanatey J.R., D’Aiuto F., Bouchard P., Chapple I., Dietrich T., Gotsman I., Graziani F.. **Periodontitis and Cardiovascular Diseases: Consensus Report**. *J. Clin. Periodontol.* (2020.0) **47** 268-288. DOI: 10.1111/jcpe.13189
5. Muñoz Aguilera E., Suvan J., Buti J., Czesnikiewicz-Guzik M., Barbosa Ribeiro A., Orlandi M., Guzik T.J., Hingorani A.D., Nart J., D’Aiuto F.. **Periodontitis Is Associated with Hypertension: A Systematic Review and Meta-Analysis**. *Cardiovasc. Res.* (2020.0) **116** 28-39. DOI: 10.1093/cvr/cvz201
6. Suvan J., D’Aiuto F., Moles D.R., Petrie A., Donos N.. **Association between Overweight/Obesity and Periodontitis in Adults. A Systematic Review**. *Obes. Rev.* (2011.0) **12** e381-e404. DOI: 10.1111/j.1467-789X.2010.00808.x
7. Gobin R., Tian D., Liu Q., Wang J.. **Periodontal Diseases and the Risk of Metabolic Syndrome: An Updated Systematic Review and Meta-Analysis**. *Front. Endocrinol.* (2020.0) **11** 336. DOI: 10.3389/fendo.2020.00336
8. Monsarrat P., Blaizot A., Kémoun P., Ravaud P., Nabet C., Sixou M., Vergnes J.-N.. **Clinical Research Activity in Periodontal Medicine: A Systematic Mapping of Trial Registers**. *J. Clin. Periodontol.* (2016.0) **43** 390-400. DOI: 10.1111/jcpe.12534
9. **Obesity and Overweight**
10. Ouchi N., Parker J.L., Lugus J.J., Walsh K.. **Adipokines in Inflammation and Metabolic Disease**. *Nat. Rev. Immunol.* (2011.0) **11** 85-97. DOI: 10.1038/nri2921
11. Park H.S., Park J.Y., Yu R.. **Relationship of Obesity and Visceral Adiposity with Serum Concentrations of CRP, TNF-Alpha and IL-6**. *Diabetes Res. Clin. Pract.* (2005.0) **69** 29-35. DOI: 10.1016/j.diabres.2004.11.007
12. Jepsen S., Suvan J., Deschner J.. **The Association of Periodontal Diseases with Metabolic Syndrome and Obesity**. *Periodontology* (2020.0) **83** 125-153. DOI: 10.1111/prd.12326
13. Fasshauer M., Blüher M.. **Adipokines in Health and Disease**. *Trends Pharm. Sci.* (2015.0) **36** 461-470. DOI: 10.1016/j.tips.2015.04.014
14. de Leal V.O., Mafra D.. **Adipokines in Obesity**. *Clin. Chim. Acta* (2013.0) **419** 87-94. DOI: 10.1016/j.cca.2013.02.003
15. Nogueira A.V.B., Nokhbehsaim M., Tekin S., de Molon R.S., Spolidorio L.C., Memmert S., Damanaki A., Jäger A., Eick S., Deschner J.. **Resistin Is Increased in Periodontal Cells and Tissues: In Vitro and In Vivo Studies**. *Mediat. Inflamm.* (2020.0) **2020** 9817095. DOI: 10.1155/2020/9817095
16. Nogueira A.V.B., Nokhbehsaim M., Eick S., Bourauel C., Jäger A., Jepsen S., Cirelli J.A., Deschner J.. **Regulation of Visfatin by Microbial and Biomechanical Signals in PDL Cells**. *Clin. Oral Investig.* (2014.0) **18** 171-178. DOI: 10.1007/s00784-013-0935-1
17. Nokhbehsaim M., Keser S., Nogueira A.V.B., Jäger A., Jepsen S., Cirelli J.A., Bourauel C., Eick S., Deschner J.. **Leptin Effects on the Regenerative Capacity of Human Periodontal Cells**. *Int. J. Endocrinol.* (2014.0) **2014** 180304. DOI: 10.1155/2014/180304
18. Nokhbehsaim M., Keser S., Nogueira A.V.B., Cirelli J.A., Jepsen S., Jäger A., Eick S., Deschner J.. **Beneficial Effects of Adiponectin on Periodontal Ligament Cells under Normal and Regenerative Conditions**. *J. Diabetes Res.* (2014.0) **2014** 796565. DOI: 10.1155/2014/796565
19. Blüher M., Mantzoros C.S.. **From Leptin to Other Adipokines in Health and Disease: Facts and Expectations at the Beginning of the 21st Century**. *Metabolism* (2015.0) **64** 131-145. DOI: 10.1016/j.metabol.2014.10.016
20. Villarreal-Molina M.T., Antuna-Puente B.. **Adiponectin: Anti-Inflammatory and Cardioprotective Effects**. *Biochimie* (2012.0) **94** 2143-2149. DOI: 10.1016/j.biochi.2012.06.030
21. Al-Suhaimi E.A., Shehzad A.. **Leptin, Resistin and Visfatin: The Missing Link between Endocrine Metabolic Disorders and Immunity**. *Eur. J. Med. Res.* (2013.0) **18** 12. DOI: 10.1186/2047-783X-18-12
22. Tatemoto K., Hosoya M., Habata Y., Fujii R., Kakegawa T., Zou M.X., Kawamata Y., Fukusumi S., Hinuma S., Kitada C.. **Isolation and Characterization of a Novel Endogenous Peptide Ligand for the Human APJ Receptor**. *Biochem. Biophys. Res. Commun.* (1998.0) **251** 471-476. DOI: 10.1006/bbrc.1998.9489
23. O’Dowd B.F., Heiber M., Chan A., Heng H.H., Tsui L.C., Kennedy J.L., Shi X., Petronis A., George S.R., Nguyen T.. **A Human Gene That Shows Identity with the Gene Encoding the Angiotensin Receptor Is Located on Chromosome 11**. *Gene* (1993.0) **136** 355-360. DOI: 10.1016/0378-1119(93)90495-O
24. Castan-Laurell I., Dray C., Attané C., Duparc T., Knauf C., Valet P.. **Apelin, Diabetes, and Obesity**. *Endocrine* (2011.0) **40** 1-9. DOI: 10.1007/s12020-011-9507-9
25. Wattanachanya L., Lu W.-D., Kundu R.K., Wang L., Abbott M.J., O’Carroll D., Quertermous T., Nissenson R.A.. **Increased Bone Mass in Mice Lacking the Adipokine Apelin**. *Endocrinology* (2013.0) **154** 2069-2080. DOI: 10.1210/en.2012-2034
26. Xie H., Tang S., Cui R., Huang J., Ren X., Yuan L., Lu Y., Yang M., Zhou H., Wu X.. **Apelin and Its Receptor Are Expressed in Human Osteoblasts**. *Regul. Pept.* (2006.0) **134** 118-125. DOI: 10.1016/j.regpep.2006.02.004
27. Li C., Cheng H., Adhikari B.K., Wang S., Yang N., Liu W., Sun J., Wang Y.. **The Role of Apelin-APJ System in Diabetes and Obesity**. *Front. Endocrinol.* (2022.0) **13** 820002. DOI: 10.3389/fendo.2022.820002
28. Boucher J., Masri B., Daviaud D., Gesta S., Guigné C., Mazzucotelli A., Castan-Laurell I., Tack I., Knibiehler B., Carpéné C.. **Apelin, a Newly Identified Adipokine Up-Regulated by Insulin and Obesity**. *Endocrinology* (2005.0) **146** 1764-1771. DOI: 10.1210/en.2004-1427
29. Hirani T., Kumar S., Patel V., Hirani S., Mohammed I., Shishoo D.. **Expression of Apelin among the Individuals of Chronic Periodontitis, with and without Type Ii Diabetes Mellitus: A Study Using Enzyme-Linked Immunosorbent Assay**. *Adv. Hum. Biol.* (2020.0) **10** 182. DOI: 10.4103/AIHB.AIHB_84_20
30. Sarhat E.R., Rmaid Z.J., Jabir T.H.. **Changes of Salivary Interleukine-17, Apelin, Omentin and Vaspin Levels in Normal Subjects and Diabetic Patients with Chronic Periodontitis**. *Ann. Trop. Med. Public Health* (2020.0) **23** 135-141. DOI: 10.36295/ASRO.2020.23118
31. de Visser Y.P., Walther F.J., Laghmani E.H., van der Laarse A., Wagenaar G.T.M.. **Apelin Attenuates Hyperoxic Lung and Heart Injury in Neonatal Rats**. *Am. J. Respir. Crit. Care Med.* (2010.0) **182** 1239-1250. DOI: 10.1164/rccm.200909-1361OC
32. Chen L., Tao Y., Jiang Y.. **Apelin Activates the Expression of Inflammatory Cytokines in Microglial BV2 Cells via PI-3K/Akt and MEK/Erk Pathways**. *Sci. China Life Sci.* (2015.0) **58** 531-540. DOI: 10.1007/s11427-015-4861-0
33. Deschner J., Eick S., Damanaki A., Nokhbehsaim M.. **The Role of Adipokines in Periodontal Infection and Healing**. *Mol. Oral Microbiol.* (2014.0) **29** 258-269. DOI: 10.1111/omi.12070
34. Damanaki A., Nokhbehsaim M., Eick S., Götz W., Winter J., Wahl G., Jäger A., Jepsen S., Deschner J.. **Regulation of NAMPT in Human Gingival Fibroblasts and Biopsies**. *Mediat. Inflamm.* (2014.0) **2014** 912821. DOI: 10.1155/2014/912821
35. Nokhbehsaim M., Eick S., Nogueira A.V.B., Hoffmann P., Herms S., Fröhlich H., Jepsen S., Jäger A., Cirelli J.A., Deschner J.. **Stimulation of MMP-1 and CCL2 by NAMPT in PDL Cells**. *Mediat. Inflamm.* (2013.0) **2013** 437123. DOI: 10.1155/2013/437123
36. Nogueira A.V.B., Nokhbehsaim M., Damanaki A., Eick S., Kirschneck C., Schröder A., Jantsch J., Deschner J.. **Filifactor Alocis and Tumor Necrosis Factor-Alpha Stimulate Synthesis of Visfatin by Human Macrophages**. *Int. J. Mol. Sci.* (2021.0) **22**. DOI: 10.3390/ijms22031235
37. Kraus D., Winter J., Jepsen S., Jäger A., Meyer R., Deschner J.. **Interactions of Adiponectin and Lipopolysaccharide from Porphyromonas Gingivalis on Human Oral Epithelial Cells**. *PLoS ONE* (2012.0) **7**. DOI: 10.1371/journal.pone.0030716
38. Lee G., Song W.-H., Kim S.-J., Kim Y.-G., Ryu J.-H.. **Apelin-APJ Axis Inhibits TNF-Alpha-Mediated Expression of Genes Involved in the Inflammatory Response in Periodontal Ligament Cells**. *Int. J. Oral Biol.* (2019.0) **44** 182-190. DOI: 10.11620/IJOB.2019.44.4.182
39. Llama-Palacios A., Potupa O., Sánchez M.C., Figuero E., Herrera D., Sanz M.. **Proteomic Analysis of Fusobacterium Nucleatum Growth in Biofilm versus Planktonic State**. *Mol. Oral Microbiol.* (2020.0) **35** 168-180. DOI: 10.1111/omi.12303
40. Patini R., Staderini E., Lajolo C., Lopetuso L., Mohammed H., Rimondini L., Rocchetti V., Franceschi F., Cordaro M., Gallenzi P.. **Relationship between Oral Microbiota and Periodontal Disease: A Systematic Review**. *Eur. Rev. Med. Pharm. Sci.* (2018.0) **22** 5775-5788. DOI: 10.26355/eurrev_201809_15903
41. Chaushu S., Wilensky A., Gur C., Shapira L., Elboim M., Halftek G., Polak D., Achdout H., Bachrach G., Mandelboim O.. **Direct Recognition of Fusobacterium Nucleatum by the NK Cell Natural Cytotoxicity Receptor NKp46 Aggravates Periodontal Disease**. *PLoS Pathog.* (2012.0) **8**. DOI: 10.1371/journal.ppat.1002601
42. Settem R.P., El-Hassan A.T., Honma K., Stafford G.P., Sharma A.. **Fusobacterium Nucleatum and Tannerella Forsythia Induce Synergistic Alveolar Bone Loss in a Mouse Periodontitis Model**. *Infect. Immun.* (2012.0) **80** 2436-2443. DOI: 10.1128/IAI.06276-11
43. Polak D., Wilensky A., Shapira L., Halabi A., Goldstein D., Weiss E.I., Houri-Haddad Y.. **Mouse Model of Experimental Periodontitis Induced by Porphyromonas Gingivalis / Fusobacterium Nucleatum Infection: Bone Loss and Host Response**. *J. Clin. Periodontol.* (2009.0) **36** 406-410. DOI: 10.1111/j.1600-051X.2009.01393.x
44. Chen T., Marsh P.D., Al-Hebshi N.N.. **SMDI: An Index for Measuring Subgingival Microbial Dysbiosis**. *J. Dent. Res.* (2022.0) **101** 331-338. DOI: 10.1177/00220345211035775
45. Rath-Deschner B., Nogueira A.V.B., Beisel-Memmert S., Nokhbehsaim M., Eick S., Cirelli J.A., Deschner J., Jäger A., Damanaki A.. **Interaction of Periodontitis and Orthodontic Tooth Movement-an in Vitro and in Vivo Study**. *Clin. Oral Investig.* (2022.0) **26** 171-181. DOI: 10.1007/s00784-021-03988-4
46. Rath-Deschner B., Memmert S., Damanaki A., de Molon R.S., Nokhbehsaim M., Eick S., Kirschneck C., Cirelli J.A., Deschner J., Jäger A.. **CXCL5, CXCL8, and CXCL10 Regulation by Bacteria and Mechanical Forces in Periodontium**. *Ann. Anat.* (2021.0) **234** 151648. DOI: 10.1016/j.aanat.2020.151648
47. Memmert S., Damanaki A., Nogueira A.V.B., Eick S., Nokhbehsaim M., Papadopoulou A.K., Till A., Rath B., Jepsen S., Götz W.. **Role of Cathepsin S in Periodontal Inflammation and Infection**. *Mediat. Inflamm.* (2017.0) **2017** 4786170. DOI: 10.1155/2017/4786170
48. Wang X., Zhang L., Li P., Zheng Y., Yang Y., Ji S.. **Apelin/APJ System in Inflammation**. *Int. Immunopharmacol.* (2022.0) **109** 108822. DOI: 10.1016/j.intimp.2022.108822
49. Antushevich H., Wójcik M.. **Review: Apelin in Disease**. *Clin. Chim. Acta* (2018.0) **483** 241-248. DOI: 10.1016/j.cca.2018.05.012
50. Lu Y., Zhu X., Liang G.-X., Cui R.-R., Liu Y., Wu S.-S., Liang Q.-H., Liu G.-Y., Jiang Y., Liao X.-B.. **Apelin–APJ Induces ICAM-1, VCAM-1 and MCP-1 Expression via NF-ΚB/JNK Signal Pathway in Human Umbilical Vein Endothelial Cells**. *Amino Acids* (2012.0) **43** 2125-2136. DOI: 10.1007/s00726-012-1298-7
51. Liu J., Liu M., Chen L.. **Novel Pathogenesis: Regulation of Apoptosis by Apelin/APJ System**. *Acta Biochim. Et Biophys. Sin. (Shanghai)* (2017.0) **49** 471-478. DOI: 10.1093/abbs/gmx035
52. Zhou Q., Cao J., Chen L.. **Apelin/APJ System: A Novel Therapeutic Target for Oxidative Stress-Related Inflammatory Diseases (Review)**. *Int. J. Mol. Med.* (2016.0) **37** 1159-1169. DOI: 10.3892/ijmm.2016.2544
53. Chen L., Tao Y., Feng J., Jiang Y.R.. **Apelin Protects Primary Rat Retinal Pericytes from Chemical Hypoxia-Induced Apoptosis**. *J. Ophthalmol.* (2015.0) **2015** 186946. DOI: 10.1155/2015/186946
54. Rayalam S., Della-Fera M.A., Krieg P.A., Cox C.M., Robins A., Baile C.A.. **A Putative Role for Apelin in the Etiology of Obesity**. *Biochem. Biophys. Res. Commun.* (2008.0) **368** 815-819. DOI: 10.1016/j.bbrc.2008.02.008
55. Ma W.-Y., Yu T.-Y., Wei J.-N., Hung C.-S., Lin M.-S., Liao Y.-J., Pei D., Su C.-C., Lu K.-C., Liu P.-H.. **Plasma Apelin: A Novel Biomarker for Predicting Diabetes**. *Clin. Chim. Acta* (2014.0) **435** 18-23. DOI: 10.1016/j.cca.2014.03.030
|
---
title: Predictors of Cognitive Decline in Healthy Middle-Aged Individuals with Asymptomatic
Alzheimer’s Disease
authors:
- Raghav Tandon
- Liping Zhao
- Caroline M. Watson
- Morgan Elmor
- Craig Heilman
- Katherine Sanders
- Chadwick M. Hales
- Huiying Yang
- David W. Loring
- Felicia C. Goldstein
- John J. Hanfelt
- Duc M. Duong
- Erik C.B. Johnson
- Aliza P. Wingo
- Thomas S. Wingo
- Blaine R. Roberts
- Nicholas T. Seyfried
- Allan I. Levey
- Cassie S. Mitchell
- James J. Lah
journal: Research Square
year: 2023
pmcid: PMC10002814
doi: 10.21203/rs.3.rs-2577025/v1
license: CC BY 4.0
---
# Predictors of Cognitive Decline in Healthy Middle-Aged Individuals with Asymptomatic Alzheimer’s Disease
## Abstract
Alzheimer’s disease (AD) progresses through a lengthy asymptomatic period during which pathological changes accumulate prior to development of clinical symptoms. As disease-modifying treatments are developed, tools to stratify risk of clinical disease will be required to guide their use. In this study, we examine the relationship of AD biomarkers in healthy middle-aged individuals to health history, family history, and neuropsychological measures and identify cerebrospinal fluid (CSF) biomarkers to stratify risk of progression from asymptomatic to symptomatic AD. CSF from cognitively normal (CN) individuals ($$n = 1149$$) in the Emory Healthy Brain Study were assayed for Aβ42, total Tau (tTau), and phospho181-Tau (pTau), and a subset of 134 cognitively normal, but biomarker-positive, individuals were identified with asymptomatic AD (AsymAD) based on a locally-determined cutoff value for ratio of tTau to Aβ42. These AsymAD cases were matched for demographic features with 134 biomarker-negative controls (CN/BM-) and compared for differences in medical comorbidities and family history. Dyslipidemia emerged as a distinguishing feature between AsymAD and CN/BM-groups with significant association with personal and family history of dyslipidemia. A weaker relationship was seen with diabetes, but there was no association with hypertension. Examination of the full cohort by median regression revealed a significant relationship of CSF Aβ42 (but not tTau or pTau) with dyslipidemia and diabetes. On neuropsychological tests, CSF Aβ42 was not correlated with performance on any measures, but tTau and pTau were strongly correlated with visuospatial perception and visual episodic memory. In addition to traditional CSF AD biomarkers, a panel of AD biomarker peptides derived from integrating brain and CSF proteomes were evaluated using machine learning strategies to identify a set of 8 peptides that accurately classified CN/BM- and symptomatic AD CSF samples with AUC of 0.982. Using these 8 peptides in a low dimensional t-distributed Stochastic Neighbor *Embedding analysis* and k-Nearest Neighbor ($k = 5$) algorithm, AsymAD cases were stratified into “Control-like” and “AD-like” subgroups based on their proximity to CN/BM- or AD CSF profiles. Independent analysis of these cases using a Joint Mutual Information algorithm selected a set of 5 peptides with $81\%$ accuracy in stratifying cases into AD-like and Control-like subgroups. Performance of both sets of peptides was evaluated and validated in an independent data set from the Alzheimer’s Disease Neuroimaging Initiative. Based on our findings, we conclude that there is an important role of lipid metabolism in asymptomatic stages of AD. Visuospatial perception and visual episodic memory may be more sensitive than language-based abilities to earliest stages of cognitive decline in AD. Finally, candidate CSF peptides show promise as next generation biomarkers for predicting progression from asymptomatic to symptomatic stages of AD.
## Introduction
Pathophysiological changes of Alzheimer’s disease (AD) begin many years before the functional or cognitive decline associated with disease. For example, in individuals with dominantly inherited AD, cerebrospinal fluid (CSF) Tau begins to increase 15 years and Aβ42 begins to decline over 20 years prior to symptom onset1,2. Until a recent report of lecanemab3, clinical trials of anti-amyloid monoclonal antibodies4-6, secretase inhibitors7,8, and anti-tau monoclonal antibodies9,10 have had limited success in symptomatic AD patients. Given the long evolution of these pathologies before clinical symptoms, identifying and treating at-risk individuals during asymptomatic stages may be a more effective strategy to delay or prevent dementia onset11. Thus, a key to successful implementation of secondary prevention trials may lie in the ability to identify those at greatest risk for AD prior to symptom onset. It is also important to recognize that many cognitively normal (CN) individuals may have evidence of AD neuropathology at death12,13. Among CN controls in the National Alzheimer’s Coordinating Center (NACC) database (May 2022), $\frac{226}{787}$ ($28.7\%$) had moderate or frequent amyloid plaques (CERAD ≥2), $\frac{386}{787}$ ($49.2\%$) had neocortical neurofibrillary tangles (Braak ≥3), and $\frac{163}{787}$ ($20.7\%$) had both CERAD ≥2 and Braak ≥3. Therefore, simply identifying the presence of AD pathology does not imply a need for intervention. For effective deployment of preventative therapies, it is imperative to both identify the presence of silent pathology and determine those at greatest risk of developing symptomatic disease. As AD is a multifactorial neurodegenerative disorder with numerous etiopathogenic mechanisms, a number of factors may influence early disease evolution, including genetics, lifetime exposures, and medical comorbidities. Additionally, AD typically manifests as mixed pathologies which evolve and change over time14-16, and multiple biomarkers are likely to be required to predict underlying pathology, disease stage, and risk of clinical progression. Recent application of systems biology approaches has led to development of proteomics-based CSF biomarker panels that link to diverse brain pathologies and may refine disease-staging and support tailored therapeutic strategies17-21. This early work indicates that these additional biomarkers may have the ability to stratify risk of clinically symptomatic AD.
To better understand the evolution of AD in its earliest stages, we explored CSF characteristics in patients with symptomatic AD and a cohort of 1149 CN middle-aged individuals (50-75 years) in the Emory Healthy Brain Study (EHBS)22, including a subset of 134 individuals with CSF levels of Aβ42, total Tau (tTau), and phospho181-Tau (pTau) indicative of underlying AD pathology based on a locally-determined cutoff ratio of tTau to Aβ42. We compared this group of asymptomatic AD (AsymAD) individuals and a demographically matched group of CN and CSF biomarker-negative (CN/BM-) controls and examined the relationship of CSF AD biomarkers to comorbidities, family history, and performance on cognitive measures. Regression analyses in the full cohort of EHBS participants were performed to look at the correlation of CSF Aβ42, tTau, and pTau with comorbidities, family history, and neuropsychological measures. Machine learning algorithms were used to identify a set of CSF peptides that effectively discriminate CN/BM- controls from symptomatic AD cases and another set of peptides that sub-categorized AsymAD individuals into “AD-like” and “Control-like” groups. Our results reveal an association of specific risk factors and cognitive changes with asymptomatic stages of AD and identify a set of CSF biomarkers that may serve to stratify risk of conversion from asymptomatic to symptomatic stages of AD.
## Emory Healthy Brain Study
The EHBS22 is a longitudinal cohort study of cognitively normal adults (50-75 years) established in 2016. It is a research study specifically focused on discovering biomarkers that predict AD and other dementias. EHBS participants are self-reported cognitively and functionally intact and free of pre-existing diagnosis of mild cognitive impairment (MCI) or any dementia. All participants complete biennial study visits which include neuropsychological testing, cardiovascular measures, brain imaging, and biospecimen collection (blood, CSF). A total of 1149 EHBS participants who had completed baseline visits and CSF collection through August 2021 were included in the current report. From this cohort, we identified 134 cognitively normal, biomarker-positive individuals with AsymAD based on measurements of Aβ42, tTau, and pTau using a locally defined cutoff value for tTaû42 ratio (>0.24) identified by Gaussian mixture models23. These individuals were matched for age, sex, and race with 134 biomarker-negative CN/BM- controls and 134 biomarker-confirmed symptomatic AD patients seen in the Emory Cognitive Neurology Clinic. AsymAD and CN/BM- controls were additionally matched for education. All individuals included in our analyses provided informed consent to participate in research protocols approved by the Emory University Institutional Review Board. Table 1 shows the descriptive statistics for the matched groups of clinical AD, AsymAD, and CN/BM- control groups ($$n = 134$$ each). Statistical differences between the AsymAD and CN/BM- groups were evaluated using the McNemar-Bowker’s test for categorical variables and by paired t-test or Wilcoxon signed rank test for continuous variables depending on the distribution.
## Neuropsychological assessment and analysis
Neuropsychological measures collected in the EHBS cohort include the Montreal Cognitive Assessment (MoCA)24, Number Span forward and backward25, Trail Making Test A&B26, Multilingual Naming Test (MINT)27, Phonemic Fluency test28, Category Fluency test29, Rey-Osterreith Complex Figure Test (RCFT)30, Rey Auditory Verbal Learning Test (AVLT)31, and Judgment of Line Orientation (JoLO)32. Neuropsychological test scores were not normally distributed and therefore compared using the Wilcoxon signed rank test between matched AsymAD and CN/BM- groups. Because the residual distributions of neuropsychological assessments were highly skewed and medians were better measures of central tendency, neuropsychological assessments for the entire cohort were regressed upon standardized CSF values using non-parametric median regression, while adjusting for age, sex, race and education. Median regression is a better alternative to ordinary least-squares regression because it is more robust to outliers and makes no assumptions about the distribution of the residuals. This resulted in estimation of median of the cognitive scores conditional on CSF analytes, adjusted for sociodemographics in the EHBS population. Due to sparse data, Number Span Forward, Number Span Backward and MINT were dichotomized by its median and analyzed using logistic regression to assess their relationship with the standardized CSF values, controlling for sociodemographics. To correct for multiple testing, an FDR correction was applied by the Benjamini-Hochberg method (FDR<$20\%$). Goodness of fit statistics “R1” was calculated to assess model fit. Statistical analyses were performed using SAS® 9.4 (SAS Institute Inc) or R software (www.R-proiect.org).
## CSF collection and analysis
CSF samples from all participants were collected in a standardized fashion applying common preanalytical methods. EHBS participants were asked to fast for at least 6 hours prior to study visits. Patients donating CSF samples in the course of clinical evaluations were asked to fast prior to their lumbar puncture (LP) procedure, but failure to do so did not preclude LP and CSF collection. Most, but not all procedures, were conducted before noon. All clinicians performing LPs in the Cognitive Neurology Clinic are also active investigators in the EHBS and apply shared standard work in both settings. LPs are performed using a 24g atraumatic Sprotte spinal needle (Pajunk Medical Systems, Norcross, GA) with aspiration and, after clearing any blood contamination, CSF is transferred from syringe to 15 ml polypropylene tubes (Corning, Glendale, AZ), which are inverted several times. The CSF is aliquoted without further handling into 0.5 ml volume in 0.9 ml FluidX tubes (Azenta, Chemsford, MA) and placed into dry ice/methanol bath prior to transfer to −80°C freezers. Time from initial collection to storage at −80°C is less than 60 minutes. Aβ42, tTau, and pTau assays were performed on CSF samples following a single freeze-thaw cycle on a Roche Cobas e601 analyzer using the Elecsys assay platform33. All assays were performed in a single laboratory in the Emory Goizueta Alzheimer’s Clinical Research Unit following manufacturer’s recommended protocols. Due to skewness and outliers, median regression was also utilized to assess the effect of personal or family history of health conditions on CSF biomarkers with adjustment of age, sex, race and education. The CSF analytes were standardized to facilitate comparability of the relative importance in all regression analyses. When the CSF analytes are dependent variables, the β coefficients represent standard deviation (SD) changes of CSF level per 1-unit change in the independent variables; when CSF biomarkers are independent variables, the β coefficients indicate changes in the outcomes per one SD change in the CSF measurements. A more positive standardized β coefficient suggested a stronger association while a more negative standardized β coefficient showed a stronger inverse association.
## Peptide selection to discriminate healthy controls and AD cases
We recently reported CSF protein changes associated with AD from integrated discovery proteomics of brain and CSF34. In our current analyses, we examined expression of 75 peptides mapped to 58 unique proteins quantified by selected reaction monitoring mass spectrometry methods as detailed elsewhere21. To identify peptides differentiating CN/BM- controls from AD patients, a machine learning strategy of backward selection was employed using $80\%$ of all CN/BM- and AD individuals. A linear classifier, Support Vector Machine (SVM), first used all available peptides to distinguish AD cases from CN/BM- controls. We then applied Recursive Feature Elimination (RFE)35 to eliminate the weakest peptides in a stepwise fashion to arrive at a smaller subset deemed important for the classification task. Recent work has shown that RFE-based biomarker selection outperforms other biomarker selection methods from proteomic datasets in supervised settings36. The size of the subset at which to stop the recursive process is a user defined parameter (set to 14). The choice of the classifier model has some influence on selection of peptides as the set of peptides resulting from RFE is not invariant to the choice of the classifier model. To address this, we recombined RFE with a different linear classifier (logistic regression) resulting in a second peptide set, deemed useful for classifying CN/BM- controls and AD cases. The final set of selected peptides is the intersection of these two sets and provides a more compact set of peptides which are useful for classification. A schematic of the peptide selection process is shown in Figure 2a. Eighty percent of data (CN/BM- and AD cases) were used to identify peptides, and 8 peptides were chosen and validated on the held-out set (remaining $20\%$ data). This held-out data set played no role in peptide identification or classifier training. These peptides were also tested in a permutation test setting where the performance of the chosen peptides was compared to the performance of 100,000 randomly chosen peptide sets of the same size ($$n = 8$$). Correlation analyses between all measured peptides and MoCA score was performed using Kendall-Tau correlation, which assesses the strength of monotonic association between the peptide and MoCA. MoCA scores were available for all 3 groups (i.e., CN/BM-, AsymAD, and AD). The correlation coefficients are sorted in a descending order and peptides which were chosen to discriminate controls from AD are shown in bold (Fig. 2h).
## Stratifying AsymAD cases
To determine if the 8 peptides chosen to discriminate between CN/BM- controls and AD can be used to classify AsymAD with more resolution, we used a low-dimensional representation to stratify AsymAD cases. This involved two successive steps of dimensionality reduction. The first was through the peptide selection (going from 75 to 8 peptides), and the second step used the t-distributed Stochastic Neighbor Embedding (t-SNE) algorithm to reduce these 8 peptides into 2 features. This analysis enables 2-dimensional visualization of how high-dimensional peptide data varies across different subjects. Lastly, the AsymAD cases were categorized as “Control-like” or “AD-like”, depending on which class (CN/BM- or AD) shares greater proximity with a given AsymAD case. This proximity is calculated using the K-Nearest Neighbor (KNN) algorithm ($k = 5$). An AsymAD case is called “Control-like” if the majority of its 5 nearest neighbors are CN/BM- and “AD- like” otherwise. This is shown in Figure 3b-3e. The APOE genotypes of the resulting AsymAD sub-categories were analyzed for differences using the Fisher’s exact test (Fig. 3g). Peptides which are able to differentiate between these sub-categories were evaluated using an information theoretic algorithm, Joint Mutual Information (JMI)37, to provide an independent view for comparing sub-categories (Fig. 3h-3i).
## Evaluation of peptides in Alzheimer’s Disease Neuroimaging Initiative (ADNI)
To assess the ability of peptides identified in the EHBS cohort to discriminate between CSF from CN/BM- controls and AD and to sub-categorize individuals with AsymAD, these peptide panels were evaluated in the Alzheimer’s Disease Neuroimaging Initiative (ADNI) cohort (adni.loni.usc.edu). The ADNI was launched in 2003 as a public-private partnership, led by Principal Investigator Michael W. Weiner, MD. The primary goal of ADNI has been to test whether serial magnetic resonance imaging (MRI), positron emission tomography (PET), other biological markers, and clinical and neuropsychological assessment can be combined to measure the progression of mild cognitive impairment (MCI) and early AD. We recently performed targeted proteomics on 706 baseline CSF samples to quantify the same set of target proteins evaluated in the EHBS cohort. Baseline amyloid PET was used to ascertain presence or absence of underlying AD pathology with CN individuals with positive amyloid PET identified as AsymAD, and hypometabolism on baseline fluorodeoxyglucose (FDG) PET was used to identify AsymAD individuals who may be closer to symptomatic disease. Standardized uptake value ratios (SUVR) for florbetapir (AV45) and FDG PET were determined by ADNI investigators as described (https://adni.loni.usc.edu/methods/pet-analysis-method/pet-analysis/). Cutoff SUVR values were determined based on Youden index in ROC analyses for AV45 (>1.226) and FDG (<1.191) using results from individuals classified as CN and Dementia at baseline ADNI visit. Individuals with baseline classification of EMCI, LMCI, or SMC were not included in the ROC plot. The 8 RFE-selected peptides were analyzed in the ADNI cohort in three groups: CN/BM- (AV45≤1.226; $$n = 203$$), AsymAD (CN; AV45>1.226; $$n = 52$$), and AD (Dementia or MCI; AV45>1.226; $$n = 250$$). Individuals without AV45 or FDG data and individuals with Dementia or MCI with AV45 SUVR≤1.226 ($$n = 192$$) were not included in the analysis. The JMI-selected panel of 5 peptides was assessed for ability to discriminate between AsymAD individuals with positive (SUVR <1.191; $$n = 10$$) or negative (SUVR ≥1.191; $$n = 42$$) baseline FDG PET scans.
## Relationship of dyslipidemia and diabetes to asymptomatic AD and CSF Aβ42
Table 1 shows characteristics of the matched set of samples consisting of individuals with symptomatic AD, AsymAD, and CN/BM- controls with 134 in each group. All groups were matched for age, sex, and race, and the CN/BM- controls and AsymAD cases were also matched for education. As expected, the AD group was substantially different from both AsymAD and CN/BM- control groups in education, MoCA score, APOE ε4 allele frequency, and levels of Aβ42, tTau, and pTau ($p \leq 0.0001$ for all). All p-values listed in Table 1 are for comparisons between AsymAD and CN/BM- control groups only and show significantly higher APOE ε4 allele frequency, lower levels of Aβ42 and higher levels of tTau and pTau in AsymAD compared to CN/BM- controls ($p \leq 0.0001$ for all). As medical comorbidities and family history can influence AD risk, we also compared prevalence of comorbidities and frequency of positive family history in these two groups. The prevalence of hypertension, coronary artery disease (CAD), diabetes, and dyslipidemia and frequency of family history of MCI, AD, memory loss, CAD, stroke, hypertension, diabetes, and dyslipidemia are shown for the matched AsymAD and CN/BM- groups. Comparable data were not consistently collected for the clinical AD group and could not be compared. Relative to the CN/BM- control group, the AsymAD group had significantly higher prevalence of dyslipidemia ($$p \leq 0.015$$) and diabetes ($$p \leq 0.033$$), and a trend for higher prevalence of hypertension ($$p \leq 0.11$$) and CAD ($$p \leq 0.10$$). The AsymAD group also had significantly more family history of AD ($$p \leq 0.0004$$) and dyslipidemia ($$p \leq 0.030$$) in first degree relatives.
The impact of comorbidities and family history on CSF AD biomarkers for the entire EHBS cohort based on median regression analysis adjusted for age, sex, race, and education is presented in Table 2 (significant effects only) and Supplemental Table S1 (all regression results of the health conditions and CSF biomarkers). Compared to participants without dyslipidemia, diabetes, and family history of AD, participants with these conditions had significantly lower Aβ42 (standardized parameter estimates, −0.23, −0.33, and −0.42, respectively), higher tTau:Aβ42 ratio (0.07, 0.19, and 0.16, respectively) and higher pTau:Aβ42 ratio (0.04, 0.10, and 0.08, respectively). After FDR correction, the association of dyslipidemia and family history of AD with Aβ42 and tTau:Aβ42 remained significant. The association of diabetes with tTau:Aβ42 (but not Aβ42) remained significant after FDR correction. No significant association of dyslipidemia, diabetes, or family history of AD was seen with either tTau or pTau. These results from matched AsymAD and CN/BM- cases and the median regression analyses in the entire cohort suggest that dyslipidemia and diabetes play an important role in early evolution of AD, specifically mediated through effects on Aβ42 rather than tTau or pTau.
## Neuropsychological measures of visuospatial function are correlated with CSF AD biomarkers in CN individuals
Table 3 shows comparisons of cognitive performance in matched AsymAD and CN/BM- groups. The AsymAD group had significantly poorer performance on JoLO ($$p \leq 0.0022$$) and tended to perform worse on RCFT delayed recall ($$p \leq 0.09$$). No other test showed a significant difference between the AsymAD and CN/BM- groups. To further explore cognitive features that may be most sensitive to early stages of AD, we analyzed the relationship of standardized CSF AD biomarkers to neuropsychological measures in the entire EHBS cohort. A summary of significant correlations between cognitive performance and standardized CSF biomarkers are shown in Table 4, and correlations for all cognitive tests with standardized CSF analytes are provided in Supplemental Table S2. Levels of Aβ42 were not significantly correlated with any cognitive measures, but increases in tTau and pTau (as well as tTau:Aβ42 and pTau:Aβ42 ratios) were associated with substantially lower scores on immediate recall of the RCFT (Fig. 1). A similar, but weaker association was also seen with RCFT delayed recall. The only other tests that were significantly correlated with CSF analytes were MoCA and JoLO. The magnitude of the association with RCFT immediate recall was the strongest among all the cognitive measures. CSF biomarkers were inversely associated with RCFT immediate recall with the β coefficients per one SD increase of −0.91, −0.92, −0.79 and −0.79, respectively, for tTau, pTau, tTau:Aβ42 and pTau:Aβ42. Notably, no significant association was seen with any CSF analytes (or ratios) with measures of verbal episodic memory (Fig. 1 and Supplemental Table S2; AVLT).
## Identification of CSF peptides associated with AD and stratification of risk among AsymAD cases
In previous work we identified changes in networks of brain-derived proteins in the CSF that discriminate between CN controls and patients with AD20,34. Multidimensional scaling analysis of a small set of CSF samples revealed differences that segregated CSF samples into AD-like and Control-like groups34. These results suggest that changes in specific proteins may allow stratification of AsymAD individuals into groups at higher or lower risk of transitioning to symptomatic AD. Using a targeted panel of 65 proteins that discriminate AD and Control CSF21, we applied machine learning-based feature selection algorithms to identify a set of peptides which distinguish CN/BM- controls from symptomatic AD cases. Levels of these peptides in AsymAD CSF were evaluated by a series of unsupervised and supervised learning algorithms to determine their proximity to CN/BM- controls or AD cases and to stratify AsymAD individuals into those who may be at lower or higher risk of progression to AD.
Figure 2a shows a schematic for peptide biomarker selection using the machine learning strategy of Recursive Feature Elimination (RFE)35. The peptide biomarkers were identified by using RFE with two different linear classifiers (SVM and logistic regression), and then choosing only those peptides which appear in both selections. The training set for peptide selection used $80\%$ of CN/BM- control and AD cases and the selected peptides were validated on the held-out $20\%$ of the data. The selected peptides are shown in figure 2b. These peptides ($$n = 8$$) distinguished CN/BM- from AD with over $98\%$ classification accuracy (57 of 58 samples) on the held-out set using a logistic regression model38 (Fig. 2i). Such a high classification accuracy indicates that the chosen peptides are generalizable to unseen data and hence possess predictive value. These peptides also performed well on random permutation tests in which they were compared to randomly chosen sets of peptides for their classification ROC-AUC (Fig. 2d-g). Figure 2h shows Kendall-Tau correlation between all peptides measured across all subjects (CN/BM-, AsymAD, AD) and the MoCA score. The Kendall-Tau correlation shows the strength of monotonic association between the peptides and the MoCA score, and the coefficients are sorted in a decreasing order. The peptides that differentiate CN/BM- and AD cases (bolded in Fig. 2h) tend to appear on the extremes of the sorted correlation coefficients. These results suggest that the peptides chosen using the RFE approach classify CN/BM- and AD cases with very high accuracy and are also strongly associated to cognitive ability.
Figure 3 shows the low dimensional t-SNE analysis of the peptide data using the 8 RFE-selected peptides. Figure 3a shows the schematic of how AsymAD cases are sub-categorized into “Control-like” and “AD-like”, based on their proximity to CN/BM- controls and AD cases, respectively. Figure 3b-d shows 2-dimensional representation of the peptide data derived using t-SNE algorithm. While the CN/BM- and AD cases occur in separable clusters (Fig. 3b), the AsymAD cases extend between them (Fig. 3c). Such a result is expected given that these individuals are hypothesized to be in a transitional stage between CN/BM- controls and symptomatic AD. Figure 3d-e shows stratification of AsymAD into Control-like and AD-like groups by using a KNN ($k = 5$) algorithm. AD-like AsymAD cases are those with ≥3 of 5 nearest neighbors among AD cases, and Control-like AsymAD cases are those with the majority of nearest neighbors among CN/BM- controls. The low dimensional t-SNE representations were computed from only those 8 peptide features which were chosen for distinguishing CN/BM- from AD cases. The sub-categories were compared for age, sex, race, education, cognitive performance, and levels of CSF Aβ42, tTau, and pTau (Supplemental Fig. S1). No significant difference was seen between the two AsymAD sub-categories for any of these features. In contrast, APOE profiles are significantly different ($$p \leq 0.0011$$ by Fisher’s exact test) with higher ε4 allele frequency in the AD-like AsymAD cases (Fig. 3g), indicating a higher genetic risk for AD in these individuals.
We next applied Joint Mutual Information (JMI) algorithm to determine if a subset of peptides can be predictive of the Control-like and AD-like AsymAD sub-categories. Unlike RFE, JMI is an information theoretic approach free from the choice of a classifier. A set of 5 peptides was chosen by the JMI algorithm to stratify AsymAD sub-categories. Of these 5 peptides (Fig. 3h-i), two are directly linked to APOE genotype. LGADMEDVR is specific for the ApoE ε4 isoform, while LGADMEDVCGR is shared by ε2 and ε3 (muted for ε4/ε4 genotype). Note that 3 of the 5 peptides selected by the JMI algorithm to sub-categorize AsymAD cases were among the 8 peptides selected by RFE to discriminate between AD cases and CN/BM- controls (Fig. 2b). The ApoE peptide shared by ε2 and ε3 isoforms was also among those selected by RFE for discriminating AD cases and CN/BM- controls but the JMI algorithm selected a new peptide specific for ApoE ε4 for stratifying AsymAD cases. This result further supports an associative link between AsymAD sub-categories and APOE genotype and contrasts with the lack of difference between AD-like and Control-like subgroups in demographic features, cognitive performance, or levels of CSF Aβ42, tTau, or pTau (Supplemental Fig. S1). GLQEAAEER is associated to the VGF protein but is distinct from the VGF peptide selected by RFE. AQALEQAK to the SMOC1 protein and YDSLK to the housekeeping protein GAPDH are the same peptides selected by RFE. Together, these 5 peptides show $81\%$ success in classifying sub-categories (using a linear logistic regression model) on $20\%$ held-out AsymAD cases which had no role in JMI peptide selection or classifier training (Fig. 3j).
## Validation of CSF peptides in ADNI data
In order to assess the sets of peptides selected by RFE for discriminating AD from CN/BM- controls and by JMI to stratify AD-like and Control-like AsymAD cases, we sought independent validation using available data from the ADNI cohort. Amyloid (AV45) PET results were used to determine the presence or absence of AD pathology. AV45 SUVR cutoff (>1.226) was determined based on ROC analysis of results from individuals classified as CN or Dementia at their baseline ADNI visit. Figure 4a shows the performance of ($$n = 8$$) peptides which were previously identified for discriminating CN/BM- and AD cases in the EHBS dataset (Fig. 2b). In the ADNI dataset, when these peptides were used to classify CN/BM- controls (AV45 SUVR <1.226) and individuals with symptomatic AD (MCI or Dementia with AV45 SUVR >1.226), a 6-fold cross-validation approach using a linear logistic regression model gave a mean ROC-AUC of 0.89. For AsymAD cases (CN with AV45 SUVR>1.226), 52 cases were identified in ADNI. To test the ability of JMI-selected peptides ($$n = 5$$; Fig. 3h) to stratify cognitively normal individuals, we used these peptides to classify the 52 ADNI AsymAD individuals with a demographically matched set of 52 CN/BM- (AV45 SUVR ≤1.226) individuals. The 5 JMI-selected peptides were able to classify these two groups with a mean AUC of 0.75 (Fig. 4b). Lastly, we took advantage of baseline FDG PET results to identify individuals with hypometabolism as a means of stratifying ADNI AsymAD individuals who might be closer to developing clinical symptoms. As was done with AV45 results, FDG PET SUVR cutoff (<1.191) was determined based on ROC analysis of results from individuals classified as CN or Dementia at their baseline ADNI visit. This cutoff identified 10 AsymAD cases with evidence of hypometabolism and 42 with normal FDG PET scans. Figure 4c shows the performance of ($$n = 5$$) peptides previously identified by JMI algorithm for discriminating Control-like vs AD-like AsymAD cases in the EHBS cohort (Fig. 3). Despite small sample sizes, the mean ROC-AUC (6-fold cross-validation with a linear logistic regression model) for the ADNI dataset was 0.75 (0.81 with EHBS data). As in our sub-categorization of AsymAD cases in the EHBS cohort, demographic features (except gender), CSF AD biomarkers, and MoCA score were not different in the FDG PET-positive and -negative AsymAD subgroups (Supplemental Fig. S2). These results support the predictive ability of RFE- and JMI-selected peptide panels to discriminate CN/BM- controls from AD cases and to sub-categorize AsymAD cases, respectively, in an independent dataset.
## Discussion
Our findings demonstrate that cognitively normal individuals with CSF biomarkers indicating underlying AD pathology (AsymAD) have distinct patterns of medical comorbidities, family history, neuropsychological measures, and CSF peptide levels compared to AD biomarker-negative controls. Machine learning approaches successfully stratified AsymAD cases to identify a sub-category whose CSF peptide profiles are more “AD-like” and another that is more “Control-like”. These results identify key features predicting progression to symptomatic AD that may serve to prioritize individuals for secondary prevention trials or for treatment with emerging disease-modifying therapies.
Numerous studies have established links between hypertension, diabetes, dyslipidemia, and peripheral artery disease and risk of AD and dementia39,40. In our study, we found evidence supporting a relationship between dyslipidemia and diabetes with AD biomarkers, specifically with CSF Aβ42 (but not Tau), among cognitively normal individuals (Table 2). We also found higher prevalence of dyslipidemia and diabetes as well as higher frequency of family history of dyslipidemia in biomarker-positive AsymAD individuals compared to a matched group of biomarker-negative controls (Table 1). Previous studies have shown increased levels of serum cholesterol in mid-life associated with increased risk of developing AD later in life41-43. In addition to its potent risk-modifying effect on AD44, ApoE plays a key role in cholesterol metabolism in the periphery, with the APOE ε4 allele associated with dyslipidemia and coronary heart disease45-47. In our study, APOE ε4 allele was enriched among individuals with AsymAD and was also a distinguishing feature between AD-like and Control-like AsymAD subgroups. Diabetes and metabolic syndrome have been strongly linked to AD risk48, including a strong association of mid-life diabetes with AD risk49. While hypertension in mid-life has also been reported to significantly increase the risk for late-life cognitive decline and AD50,51, results of epidemiological studies have been mixed52. In our study, we did not find significant differences in prevalence, family history, or relationship of hypertension to CSF AD biomarkers among cognitively normal individuals. As we stratified groups solely based on objective measures of CSF AD biomarkers, the relationships that we identified are not dependent on clinical classifications, and these findings support a relationship between dyslipidemia and diabetes with early evolution of AD pathology, specifically mediated through effects on Aβ42 during clinically silent asymptomatic stages.
In contrast to the association of medical comorbidities and family history with levels of CSF Aβ42, we found significant relationships between CSF tTau and pTau, but not Aβ42, with cognitive test results. In our comparison of 134 matched biomarker-positive and -negative groups in the EHBS cohort, only the score on a visuospatial task (Judgment of Line Orientation; JoLO) differed significantly between groups. It should be noted that since all individuals in the EHBS cohort are cognitively normal at enrollment, these comparisons are likely limited by a strong ceiling effect. When we examined the relationship of these scores to CSF AD biomarkers in the larger cohort ($$n = 1149$$) with more robust statistical methods, we identified several additional relationships. Episodic memory, which is typically the earliest domain affected in AD53, was measured by two primary measures, the Rey Auditory Verbal Learning Test (AVLT) and the Rey-Osterreith Complex Figure Test (RCFT). Interestingly, visual memory (RCFT) but not verbal memory (AVLT) was associated with CSF levels of tTau and pTau ($p \leq 0.0001$; Fig. 1), and the parameter estimates for RCFT immediate recall (−0.91 tTau and −0.92 pTau) were high. JoLO was not significantly correlated with individual analytes, but was significantly associated with the ratios of tTau and pTau to Aβ42 (Table 4). These findings suggest that visual memory and visuospatial abilities may be more sensitive to very early changes in AD than language-based abilities. While this is a fairly novel conclusion, there are some precedents supporting this possibility54-56, and our results highlight the importance of assessing visuospatial abilities to identify the earliest cognitive changes associated with AD.
The present study showed that a small set of 8 differentially expressed peptides can effectively distinguish AD cases from cognitively healthy controls. These peptides were identified using machine learning algorithms (Recursive Feature Elimination – RFE, combined with linear classifiers) and show good generalizability on unseen data which had no role in peptide identification. Importantly, the set of predictive peptides have been shown in recent studies to be important in tracking disease status and progression. Neuronal Pentraxin Receptor (NPTXR) isoform 1 (protein for the ADQDTIR peptide) has been shown to be a CSF biomarker of AD progression57 with levels differing between MCI and more advanced AD stages. YWHAZ (protein for the VVSSIEQK peptide) has recently emerged as an important biomarker to discriminate AD from non-AD cases with cognitive impairment and also predicts individuals with high Tau and low Aβ42 levels58. CHI3L1 (protein for the IASNTQSR peptide; also known as YKL-40) has been reported in other studies as a potential prognostic fluid biomarker, and its ratio to Aβ42 is predictive for developing cognitive impairment59. CHI3L1 is also a glial/inflammation related biomarker17,58,60. VGF (protein for the EPVAGDAVPGPK peptide) has been strongly associated with cognitive trajectory and suggested to act through mechanisms independent of amyloid plaques and neurofibrillary tangles in contributing to cognitive decline61. Further, VGF has also been identified as a key regulator playing a causal role in protecting against AD pathogenesis and progression62. SMOC1 (protein for AQALEQAK peptide), which is related to the extracellular matrix and strongly correlated with global AD pathology in brain63, has shown the ability to discriminate between AD and non-AD cognitive impairment (specificity for AD) and to predict levels of CSF Aβ42, tTau, and pTau58. GAPDH (protein for the YDNSLK peptide) is known to form stable aggregates with extracellular Aβ, and these aggregates have been found to be proportional to the progressive stage of AD64,65. These peptides from 8 proteins, each with plausible biological connection to AD pathophysiology, were found to be among the most strongly associated with cognition and were able to discriminate CSF samples from AD patients and Controls with $98\%$ accuracy (Fig. 2).
The 15-20 year period during which AD neuropathology evolves silently prior to cognitive decline offers a window of opportunity to slow or prevent clinical disease. However, as many individuals with AD neuropathology never develop symptoms during life, it will be critical that we develop tools to identify those individuals at greater risk of cognitive decline. Toward this goal, we applied a stratification strategy to sub-categorize 134 cognitively normal individuals with asymptomatic AD into AD-like and Control-like groups using expression levels of the 8 peptides which were selected for differentiating AD cases from healthy, biomarker-negative controls. The AD-like AsymAD cases show a higher frequency of the APOE ε4 allele but are otherwise indistinguishable from the Control-like AsymAD cases based on demographics, cognitive performance, or level of CSF Aβ42, tTau, or pTau. Applying a Joint Mutual Information (JMI) algorithm, we identified an independent set of 5 peptides (from SMOC1, VGF, APOE ε2/ε3, APOE ε4, and GAPDH) which were able to classify AD-like and Control-like AsymAD cases with $81\%$ accuracy. The two ApoE-associated peptides selected by JMI specifically discriminate presence or absence of the ε4 isofom. While there have been conflicting reports regarding the impact of APOE genotype on clinical progression among symptomatic AD patients66-70, the selection of these peptides and higher frequency of APOE ε4 in the AD-like AsymAD subgroup suggests that ε4 carriers are predisposed to more rapid decline and a shortened asymptomatic phase of AD.
To test the peptides identified in the EHBS cohort, we took advantage of targeted proteomics analysis which our group recently performed on 706 baseline ADNI CSF samples. Since amyloid PET scans were available for the ADNI cohort, we used AV45 PET positivity as a means of defining individuals with underlying AD pathology. The 8 RFE-selected peptides were effective in discriminating between CN/BM- (AV45 PET negative) controls and amyloid PET-confirmed AD with mean AUC of 0.89. Among all CN individuals in ADNI, there were 52 with asymptomatic AD based on positive amyloid PET. The 5 JMI-selected peptides were able to discriminate these AsymAD cases from a demographically-match cohort of 52 CN individuals with negative AV45 PET scans with a mean AUC of 0.74. Only a very small number of individuals in ADNI have transitioned from CN to MCI or Dementia during longitudinal follow up, and, in addition to being a rare event, clinical progression is complicated by frequent reversions from MCI to CN71-73. To avoid these limitations, we used FDG PET to identify AsymAD individuals with evidence of hypometabolism and presumably at greater risk of symptomatic progression. Despite small sample size ($$n = 10$$), the 5 JMI-selected peptides classified FDG-positive AsymAD cases with mean AUC of 0.75. Unlike our previous studies with deep proteomics and network analyses20,34, the purpose of the current work was to evaluate CSF peptides that might serve as effective biomarkers to predict cognitive decline in cognitively normal individuals harboring AD pathology. Deep proteomics comparing AD-like and Control-like AsymAD cases should produce better understanding of changes occurring during the transition from asymptomatic to symptomatic stages of AD, and additional work will refine peptide panels to improve their predictive ability. Longitudinal follow up of individuals with asymptomatic AD will be required for ultimate validation of predictive biomarkers, and this will be possible in the EHBS cohort over time.
In sum, our current findings provide evidence linking dyslipidemia and diabetes (but not hypertension) to early evolution of AD pathology mediated through effects on Aβ42 in healthy middle-aged individuals. On cognitive measures, we identified a strong relationship of CSF tTau and pTau levels (but not Aβ42) with visual episodic memory and visuospatial perception, which was not observed with language-based tests, suggesting earlier impact on visual networks in the evolution of AD pathology. Lastly, using machine-learning algorithms we identified peptide panels that can stratify individuals at greater risk of progression to symptomatic disease who can be prioritized for prevention trials and for treatment with emerging disease-modifying therapies for AD.
## References
1. Bateman R.J.. **Clinical and biomarker changes in dominantly inherited Alzheimer's disease**. *The New England journal of medicine* (2012) **367** 795-804. PMID: 22784036
2. Fagan A.M.. **Longitudinal change in CSF biomarkers in autosomal-dominant Alzheimer's disease**. *Science translational medicine* (2014) **6** 226ra230
3. van Dyck C.H.. **Lecanemab in Early Alzheimer's Disease**. *The New England journal of medicine* (2022)
4. Vandenberghe R.. **Bapineuzumab for mild to moderate Alzheimer's disease in two global, randomized, phase 3 trials**. *Alzheimers Res Ther* (2016) **8** 18. PMID: 27176461
5. Doody R.S.. **Phase 3 trials of solanezumab for mild-to-moderate Alzheimer's disease**. *New England Journal of Medicine* (2014) **370** 311-321. PMID: 24450890
6. Scheltens Pv. **DT-01-02: Biomarker data from scarlet road: A global phase 3 study of gantenerumab in patients with prodromal Alzheimer's disease**. *Alzheimer's & Dementia* (2015) **11** P331-P331
7. Coric V.. **Safety and tolerability of the gamma-secretase inhibitor avagacestat in a phase 2 study of mild to moderate Alzheimer disease**. *Arch Neurol* (2012) **69** 1430-1440. PMID: 22892585
8. Doody R.S.. **A phase 3 trial of semagacestat for treatment of Alzheimer's disease**. *N Engl J Med* (2013) **369** 341-350. PMID: 23883379
9. Teng E.. **Safety and Efficacy of Semorinemab in Individuals With Prodromal to Mild Alzheimer Disease: A Randomized Clinical Trial**. *JAMA Neurol* (2022) **79** 758-767. PMID: 35696185
10. Panza F., Lozupone M.. **The challenges of anti-tau therapeutics in Alzheimer disease**. *Nat Rev Neurol* (2022) **18** 577-578. PMID: 35941199
11. Mehta D., Jackson R., Paul G., Shi J., Sabbagh M.. **Why do trials for Alzheimer's disease drugs keep failing? A discontinued drug perspective for 2010-2015**. *Expert Opin Investig Drugs* (2017) **26** 735-739
12. Serrano-Pozo A.. **Examination of the clinicopathologic continuum of Alzheimer disease in the autopsy cohort of the National Alzheimer Coordinating Center**. *J Neuropathol Exp Neurol* (2013) **72** 1182-1192. PMID: 24226270
13. Bennett D.A.. **Neuropathology of older persons without cognitive impairment from two community-based studies**. *Neurology* (2006) **66** 1837-1844. PMID: 16801647
14. Schneider J.A., Arvanitakis Z., Leurgans S.E., Bennett D.A.. **The neuropathology of probable Alzheimer disease and mild cognitive impairment**. *Ann Neurol* (2009) **66** 200-208. PMID: 19743450
15. Gross A.L., Mungas D.M., Leoutsakos J.S., Albert M.S., Jones R.N.. **Alzheimer's disease severity, objectively determined and measured**. *Alzheimers Dement (Amst)* (2016) **4** 159-168. PMID: 27830173
16. DeTure M.A., Dickson D.W.. **The neuropathological diagnosis of Alzheimer's disease**. *Mol Neurodegener* (2019) **14** 32. PMID: 31375134
17. Higginbotham L.. **Integrated proteomics reveals brain-based cerebrospinal fluid biomarkers in asymptomatic and symptomatic Alzheimer’s disease**. *Science advances* (2020) **6** eaaz9360. PMID: 33087358
18. Castrillo J.I., Lista S., Hampel H., Ritchie C.W.. **Systems Biology Methods for Alzheimer's Disease Research Toward Molecular Signatures, Subtypes, and Stages and Precision Medicine: Application in Cohort Studies and Trials**. *Methods Mol Biol* (2018) **1750** 31-66. PMID: 29512064
19. Lista S., Zetterberg H., O'Bryant S.E., Blennow K., Hampel H.. **Evolving Relevance of Neuroproteomics in Alzheimer's Disease**. *Methods Mol Biol* (2017) **1598** 101-115. PMID: 28508359
20. Johnson E.C.B.. **Large-scale proteomic analysis of Alzheimer's disease brain and cerebrospinal fluid reveals early changes in energy metabolism associated with microglia and astrocyte activation**. *Nat Med* (2020) **26** 769-780. PMID: 32284590
21. Watson C.M.. **Quantitative Mass Spectrometry Analysis of Cerebrospinal Fluid Biomarker Proteins Reveals Stage-Specific Changes in Alzheimer’s Disease**. *medRxiv* (2022)
22. Goetz M.E.. **Rationale and Design of the Emory Healthy Aging and Emory Healthy Brain Studies**. *Neuroepidemiology* (2019) **53** 187-200. PMID: 31454799
23. Shaw L.M.. **Derivation of cutoffs for the Elecsys((R)) amyloid beta (1-42) assay in Alzheimer's disease**. *Alzheimers Dement (Amst)* (2018) **10** 698-705. PMID: 30426066
24. Nasreddine Z.S.. **The Montreal Cognitive Assessment, MoCA: a brief screening tool for mild cognitive impairment**. *J Am Geriatr Soc* (2005) **53** 695-699. PMID: 15817019
25. Weintraub Sv. **Version 3 of the Alzheimer Disease Centers' Neuropsychological Test Battery in the Uniform Data Set (UDS)**. *Alzheimer Dis Assoc Disord* (2018) **32** 10-17. PMID: 29240561
26. 26.Army Individual Test Battery. Manual of directions and scoring., (War Department, Adjutant General, Washington (DC), 1944).. *Manual of directions and scoring* (1944)
27. Ivanova I., Salmon D.P., Gollan T.H.. **The multilingual naming test in Alzheimer's disease: clues to the origin of naming impairments**. *Journal of the International Neuropsychological Society: JINS* (2013) **19** 272-283. PMID: 23298442
28. Benton A., Hamsher K., Sivan A.. *Multilingual aphasia examination* (1983)
29. Rosen W.G.. **Verbal fluency in aging and dementia**. *Journal of Clinical Neuropsychology* (1980) **2** 135-146
30. Shin M.S., Park S.Y., Park S.R., Seol S.H., Kwon J.S.. **Clinical and empirical applications of the Rey-Osterrieth Complex Figure Test**. *Nat Protoc* (2006) **1** 892-899. PMID: 17406322
31. Schmidt M.. *Rey auditory verbal learning test: A handbook* (1996)
32. Benton A.L., Varney N.R., Hamsher K.D.. **Visuospatial judgment. A clinical test**. *Arch Neurol* (1978) **35** 364-367. PMID: 655909
33. Bittner Tv. **Technical performance of a novel, fully automated electrochemiluminescence immunoassay for the quantitation of beta-amyloid (1-42) in human cerebrospinal fluid**. *Alzheimers Dement* (2016) **12** 517-526. PMID: 26555316
34. Higginbotham L.. **Integrated proteomics reveals brain-based cerebrospinal fluid biomarkers in asymptomatic and symptomatic Alzheimer's disease**. *Sci Adv* (2020) **6**
35. Guyon I., Weston J., Barnhill S., Vapnik V.. **Gene Selection for Cancer Classification using Support Vector Machines**. *Machine Learning* (2002) **46** 389-422
36. Tandon Raghav, L. A.I., Lah James J., Seyfried Nicholas T., Mitchell Cassie S.. **Machine learning selection of most predictive brain proteins suggests role of sugar metabolism in Alzheimer’s Disease**. *Journal of Alzheimer’s Disease* (2023)
37. Yang H., Moody J.. **Data visualization and feature selection: New algorithms for nongaussian data**. *Advances in neural information processing systems* (1999) **12**
38. Pedregosa F.. **Scikit-learn: Machine learning in Python**. *the Journal of machine Learning research* (2011) **12** 2825-2830
39. Newman A.B.. **Dementia and Alzheimer's disease incidence in relationship to cardiovascular disease in the Cardiovascular Health Study cohort**. *J Am Geriatr Soc* (2005) **53** 1101-1107. PMID: 16108925
40. Stampfer M.J.. **Cardiovascular disease and Alzheimer's disease: common links**. *J Intern Med* (2006) **260** 211-223. PMID: 16918818
41. Solomon A., Kivipelto M., Wolozin B., Zhou J., Whitmer R.A.. **Midlife serum cholesterol and increased risk of Alzheimer's and vascular dementia three decades later**. *Dementia and geriatric cognitive disorders* (2009) **28** 75-80. PMID: 19648749
42. Whitmer R.A., Sidney S., Selby J., Johnston S.C., Yaffe K.. **Midlife cardiovascular risk factors and risk of dementia in late life**. *Neurology* (2005) **64** 277-281. PMID: 15668425
43. Oliveira B.C.d.L., Bellozi P.M.Q., Reis H.J., de Oliveira A.C.P.. **Inflammation as a Possible Link Between Dyslipidemia and Alzheimer’s Disease**. *Neuroscience* (2018) **376** 127-141. PMID: 29454102
44. Strittmatter W.J., Roses A.D.. **Apolipoprotein E and Alzheimer's disease**. *Annual Review of Neuroscience* (1996) **19** 53-77
45. DallongeviNe J., Lussier-Cacan S., Davignon J.. **Modulation of plasma triglyceride levels by apoE phenotype: a meta-analysis**. *Journal of lipid research* (1992) **33** 447-454. PMID: 1388198
46. Hanis C.L., Hewett-Emmett D., Douglas T.C., Bertin T.K., Schull W.J.. **Effects of the apolipoprotein E polymorphism on levels of lipids, lipoproteins, and apolipoproteins among Mexican-Americans in Starr County, Texas**. *Arteriosclerosis and Thrombosis: A Journal of Vascular Biology* (1991) **11** 362-370. PMID: 1998654
47. Wilson P.W.. **Apolipoprotein E alleles, dyslipidemia, and coronary heart disease. The Framingham Offspring Study**. *JAMA* (1994) **272** 1666-1671. PMID: 7966894
48. Profenno L.A., Porsteinsson A.P., Faraone S.V.. **Meta-analysis of Alzheimer's disease risk with obesity, diabetes, and related disorders**. *Biol Psychiatry* (2010) **61** 505-512
49. Xu W.. **Mid- and late-life diabetes in relation to the risk of dementia: a population-based twin study**. *Diabetes* (2009) **58** 71-77. PMID: 18952836
50. Launer L.J.. **Midlife blood pressure and dementia: the Honolulu-Asia aging study**. *Neurobiology of aging* (2000) **21** 49-55. PMID: 10794848
51. Launer L.J.. **The epidemiologic study of dementia: a life-long quest?**. *Neurobiology of aging* (2005) **26** 335-340. PMID: 15639311
52. Power M.C.. **The association between blood pressure and incident Alzheimer disease: a systematic review and meta-analysis**. *Epidemiology* (2011) **22** 646-659. PMID: 21705906
53. Wilson R.S., Leurgans S.E., Boyle P.A., Bennett D.A.. **Cognitive decline in prodromal Alzheimer disease and mild cognitive impairment**. *Arch Neurol* (2011) **68** 351-356. PMID: 21403020
54. Coughlan G., Laczo J., Hort J., Minihane A.M., Hornberger M.. **Spatial navigation deficits - overlooked cognitive marker for preclinical Alzheimer disease?**. *Nat Rev Neurol* (2018) **14** 496-506. PMID: 29980763
55. Johnson D.K., Storandt M., Morris J.C., Galvin J.E.. **Longitudinal study of the transition from healthy aging to Alzheimer disease**. *Arch Neurol* (2009) **66** 1254-1259. PMID: 19822781
56. Kawas C.H.. **Visual memory predicts Alzheimer's disease more than a decade before diagnosis**. *Neurology* (2003) **60** 1089-1093. PMID: 12682311
57. Begcevic I.. **Neuronal pentraxin receptor-1 is a new cerebrospinal fluid biomarker of Alzheimer's disease progression**. *F1000Res* (2018) **1** 1012
58. Zhou M.. **Targeted mass spectrometry to quantify brain-derived cerebrospinal fluid biomarkers in Alzheimer’s disease**. *Clinical proteomics* (2020) **11** 1-14
59. Craig-Schapiro R.. **YKL-40: a novel prognostic fluid biomarker for preclinical Alzheimer's disease**. *Biol Psychiatry* (2010) **68** 903-912. PMID: 21035623
60. Zhao T., Su Z., Li Y., Zhang X., You Q.. **Chitinase-3 like-protein-1 function and its role in diseases**. *Signal Transduction and Targeted Therapy* (2020) **5** 201. PMID: 32929074
61. Wingo A.P.. **Large-scale proteomic analysis of human brain identifies proteins associated with cognitive trajectory in advanced age**. *Nature Communications* (2019) **10** 1619
62. Beckmann N.D.. **Multiscale causal networks identify VGF as a key regulator of Alzheimer’s disease**. *Nature Communications* (2020) **11** 3942
63. Johnson E.C.B.. **Large-scale deep multi-layer analysis of Alzheimer's disease brain reveals strong proteomic disease-related changes not observed at the RNA level**. *Nature neuroscience* (2022) **25** 213-225. PMID: 35115731
64. Cumming R.C., Schubert D.. **Amyloid-beta induces disulfide bonding and aggregation of GAPDH in Alzheimer's disease**. *FASEB J* (2005) **19** 2060-2062. PMID: 16186172
65. Lazarev V.F.. **Extracellular GAPDH Promotes Alzheimer Disease Progression by Enhancing Amyloid-beta Aggregation and Cytotoxicity**. *Aging Dis* (2021) **12** 1223-1237. PMID: 34341704
66. Frisoni G.B.. **Gene dose of the epsilon 4 allele of apolipoprotein E and disease progression in sporadic late-onset Alzheimer's disease**. *Ann Neurol* (1995) **31** 596-604
67. Dal Forno Gv. **Apolipoprotein E genotype and rate of decline in probable Alzheimer's disease**. *Arch Neurol* (1996) **53** 345-350. PMID: 8929157
68. Craft Sv. **Accelerated decline in apolipoprotein E-epsilon4 homozygotes with Alzheimer's disease**. *Neurology* (1998) **51** 149-153. PMID: 9674794
69. Hoyt B.D., Massman P.J., Schatschneider C., Cooke N., Doody R.S.. **Individual growth curve analysis of APOE epsilon 4-associated cognitive decline in Alzheimer disease**. *Arch Neurol* (2005) **62** 454-459. PMID: 15767511
70. Suzuki K.. **Effect of apolipoprotein E epsilon4 allele on the progression of cognitive decline in the early stage of Alzheimer's disease**. *Alzheimers Dement (N Y)* (2020) **6** e12007. PMID: 32211510
71. Canevelli Mv. **Spontaneous Reversion of Mild Cognitive Impairment to Normal Cognition: A Systematic Review of Literature and Meta-Analysis**. *J Am Med Dir Assoc* (2016) **17** 943-948. PMID: 27502450
72. Koepsell T.D., Monsell S.E.. **Reversion from mild cognitive impairment to normal or near-normal cognition: risk factors and prognosis**. *Neurology* (2012) **79** 1591-1598. PMID: 23019264
73. Ganguli Mv. **Outcomes of mild cognitive impairment by definition: a population study**. *Arch Neurol* (2011) **68** 761-767. PMID: 21670400
|
---
title: 'Fibrinogen; a predictor of injury severity and mortality among patients with
traumatic brain injury in Sub-Saharan Africa: a prospective study'
authors:
- John Baptist Ssenyondwa
- Joel Kiryabwire
- Martin Kaddumukasa
- Devereaux Michael
- Larrey Kasereka Kamabu
- Moses Galukande
- Mark Kaddumukasa
- Martha Sajatovic
- Timothy Kabanda Makumbi
journal: Research Square
year: 2023
pmcid: PMC10002822
doi: 10.21203/rs.3.rs-2596161/v1
license: CC BY 4.0
---
# Fibrinogen; a predictor of injury severity and mortality among patients with traumatic brain injury in Sub-Saharan Africa: a prospective study
## Abstract
### Introduction
Fibrinogen levels drop quicker than any other factors in severe trauma such as Traumatic Brain Injury (TBI). Contemporaneous studies show that fibrinogen concentrations < 2 g/L are strongly related to mortality. However, little is known regarding fibrinogen levels and TBI severity as well as mortality in sub-Saharan Africa. We therefore set out to determine whether fibrinogen levels are associated with TBI severity and seven days outcomes.
### Objectives
To determine the sensitivity and specificity of fibrinogen levels and the association with severity and mortality among TBI patients at Mulago Hospital.
### Methods
We prospectively enrolled 213 patients with TBI aged between 13 and 60 years of age and presenting within 24hrs of injury. Patients with pre-existing coagulopathy, concurrent use of anticoagulant or antiplatelet agents, pre-existing hepatic insufficiency, diabetes mellitus and who were pregnant were excluded. Fibrinogen levels were determined using the Clauss fibrinogen assay.
### Results
Majority of the patients were male ($88.7\%$) and nearly half were aged 30 or less ($48.8\%$). Fibrinogen levels less than 2g/L were observed in 74 ($35.1\%$) of the patients while levels above 4.5 g/L were observed in 30($14.2\%$) of the patients. The average time spent in the study was 3.7 ± 2.4 days. The sensitivity and specificity using fibrinogen < 2g/L was $56.5\%$ and $72.9\%$ respectively. Fibrinogen levels predict TBI severity with an AUC = 0.656 ($95\%$ CI 0.58–0.73: $$p \leq 0.000$$) Fibrinogen levels < 2g/L (hypofibrinogenemia) were independently associated with severe TBI. ( AOR 2.87 CI,1.34–6.14: $$p \leq 0.007$$). Levels above 4.5g/L were also independently associated with injury severity (AOR 2.89, CI 1.12–7.48: $p \leq 0.05$) Fibrinogen levels more than 4.5g/L were independently associated with mortality (OR 4.5, CI;1.47–13.61, $p \leq 0.05$).
### Conclusions
The fibrinogen level is a useful tool in predicting severity including mortality of TBI in our settings. We recommend the routine use of fibrinogen levels in TBI patient evaluations as levels below 2g/L and levels above 4.5g/L are associated with severe injuries and mortality
## Introduction
Trauma accounts for $11\%$ of the world’s disability adjusted life years (DALYs) with $90\%$ of these occurring in Low and Middle Income Countries.[1] Traumatic Brain Injury (TBI) per se is a major cause of disability globally with an incidence rate of 200 per 100 000 people per year[2] In Uganda, head injuries with TBI are the commonest type of injuries accounting for $44\%$ of trauma admissions at hospitals in Kampala[3] and mortality rate of $\frac{220}{100}$,000.[4] The morbidity and mortality due to TBI is higher in low-income and middle-income countries[5, 6] despite advancements in the clinical evaluation of patients with TBI using standardized protocols such as the Advanced Trauma Life Support (ATLS) protocols.[7, 8] Evidence from prior studies shows that deaths from trauma can be prevented if adequate and timely identification of the problem is done and the appropriate line of management is decided early.[9] In low and middle income settings where there is limited access to prompt investigation modalities for TBI victims, clinicians often find themselves relying on trauma algorithms, trauma assessment tools and clinical examination findings to diagnose and direct TBI management. Some of the trauma assessment tools employed in the evaluation of TBI patients include; Abbreviated Injury Score(AIS), Trauma Injury Severity Score (TRISS) and the Glasgow Coma Scale (GCS) specifically for TBI.[10] These assessment tools have been found to have considerable limitations and that they may not correlate well with severity of injury.[11] Among these, the GCS remains the commonest tool used to assess TBI severity in Sub-Saharan Africa.[12, 13] Despite its wide utility, the GCS does not provide specific parametric clinical information about the pathophysiologic abnormalities in TBI which are the targets of our interventions.[14] One example of a pathophysiologic event is intracranial bleeding which is also associated with poor clinical outcomes such as mortality and disability.[15] Fibrinogen, which is a positive acute phase protein[16] as well as a haemostatic protein [17] has been retrospectively studied as a prognostic indicator among TBI patients as well as a predictor of in hospital mortality(18–20). Following trauma, fibrinogen levels deteriorate more frequently and earlier than other routine coagulation parameters.[21, 22] *Additionally hypofibrinogenemia* has been described as a common occurrence in TBI possibly due to trauma induced coagulopathy.[19, 21, 23] A recent study showed that fibrinogen concentrations less than 2g. L were associated with poor outcomes including mortality in contrast to concentrations above 2.5g/L that are associated with favourable outcomes.[18, 19] Current guidelines also emphasize that fibrinogen concentrations be maintained over 1.5–2.0 g/L in severe trauma patients[22] Fibrinogen, therefore has a pivotal role in TBI; however, little is known regarding its sensitivity and specificity in diagnosis of severe TBI in sub-Saharan Africa.
We therefore set out to study the predictive ability of fibrinogen levels in determining TBI severity and predicting clinical outcomes in TBI as a step in improving prompt diagnosis and management of TBI victims. The aims of the study were to determine the sensitivity and specificity of low fibrinogen levels in predicting severity of traumatic brain injuries, to describe the association of fibrinogen levels with TBI severity and 7-day outcomes among TBI patients at Mulago Hospital. We hypothesized that plasma fibrinogen levels are associated with severity and short-term clinical outcomes in TBI patients.
## Study design and setting.
We prospectively studied 213 randomly selected TBI patients admitted to the Casualty unit at Mulago National Referral Hospital (MNRH) between December 2021 and May 2022. MNRH is the biggest public hospital in Uganda at approximately 5 kilometres from the city centre and it receives $75\%$ of injured victims in Kampala.[24] The Casualty unit of the hospital is the entry point for all trauma cases presenting to the hospital.
## Study population and sampling.
The inclusion criteria were as follows: patients aged 13 to 60 years with a clinical diagnosis of TBI documented using Computed Tomography (CT) or Glasgow Coma Scale (GCS) score by clinician and admitted within 24Hrs of TBI occurrence. The age range of 13 to 60 years was used in consideration of the altered metabolism of fibrinogen that occurs at the young and elderly extremes of age.(25–27) TBI in this study was defined as any alteration of brain function or presence of other evidence of brain pathology based on the GCS or head CT scan in a patient, caused by an external force such as accidents, assault, falls and burns.[28] Patients on concurrent use of anticoagulant or antiplatelet agents, medical diagnosis of liver disease, hypertension, and Diabetes mellitus, patients admitted after 24hrs of the injury occurrence and pregnant women were excluded.
To achieve our first objectives, we used the proportion of patients with low fibrinogen from a prior study [19] and level of precision of $7\%$ at $95\%$ confidence interval to determine a sample size of 186 patients using the Kish and Leslie formula.[29] To study the relationship of fibrinogen with outcomes, we calculated the sample size using formula for cohort studies based on comparison of two proportions representing the event rates in both the exposed and the non-exposed groups.[30] Using proportions from the study by Lv et al.[31], with $95\%$ CI and power of $80\%$, we determined a sample size of 140 patients for the cohort. Adjusting upwards for losses to follow up, we estimated the sample size to be 200 patients. We therefore enrolled a total of 213 patients using systematic random sampling to answer our objectives. All patients were evaluated and treated according to the local protocol. Informed consent was obtained from the patients included in the study and for the unconscious patients, waiver of informed consent was obtained from the Research Ethics Committee of Makerere University and Mulago Hospital Ethics committee.
## Study procedure and data Collection.
Data obtained included demographic information such as age, sex, occupation, time of injury, level of education, mechanism of injury and type of head injury. Clinical data including, blood pressure, pulse oximetry, temperature, pupillary reaction, CT scan results, GCS score and fibrinogen levels taken at time of admission were also obtained. The severity of injury was determined by the GCS score obtained by the neurosurgical team. The GCS score of ≤ 8 was categorized as severe TBI and scores 9–15 as non-severe TBI. ( Fig. 1) Fibrinogen levels were measured by the Clauss fibrinogen assay using the “Yumizen G FIB 5” reagent. The test was carried out on fresh decalcified venous blood obtained from the participants. On admission, 5 milliliters of venous blood were drawn from each of the participants into $3.2\%$ sodium citrate vacutainers and transported to the laboratory within 60 minutes of collection for analysis. Samples were centrifuged to obtain plasma that was prepared for analysis as a 1:10 dilution with Yumizen G IMIDAZOL buffer. The prepared sample was then analysed using an automated analyser and results recorded in g/L. We obtained levels < 1g/L as well as those above 5g/L that were retested at 1:5 dilution and 1:20 dilution respectively to obtain final results. The fibrinogen levels were categorized as: normal fibrinogen levels between 2 and 4.5g/L.[32] A fibrinogen level of < 2 g/L was considered as being low and a level of > 4.5 g/L as high according to standard laboratory reference values. The patients were followed up daily for 7 days and outcomes documented. The clinical outcome studied was in hospital mortality within 7 days of admission. The 11 patients lost to follow up were not analysed for outcomes. ( Fig. 1) They were however included in the analysesis for the association between fibrinogen and injury severity on admission to the hospital.
## Statistical analysis.
All study data collected was entered in Epidata version 4.6 software, cleaned and exported to STATA version 14 for analysis. Continuous variables were summarised as means with standard deviation. Categorical variables are expressed as percentages. Bivariate analyses of categorical variables were performed using Pearson’s chi test and presented as p values. A Receiver Operating Characteristic (ROC) curve is used to describe the predictive ability of fibrinogen levels in TBI. The sensitivity ± positive predictive value and specificity ± negative predictive value of fibrinogen levels were calculated using a 2 2 table. Binary logistic regression models were used to describe the relationship between categorical variables between the patient groups with GCS ≤ 8 and that with GCS ≥ 9. Following bivariate analyses, logistic regression multivariate models were used to evaluate the association between fibrinogen and TBI severity as well as in-hospital mortality. All patients with missing data were excluded from the analyses. The models were tested for multiple collinearities, goodness of fit and all independent variables with correlation coefficient above ± 0.4 were excluded from the logistic regression model. The relationship is presented as odds ratio with $95\%$ confidence intervals. All statistical analyses were performed using Stata statistical software, StataCorp. 2015. Stata Statistical Software: Release 14. College Station, TX: StataCorp LP For all analysis, the level of statistical significance was considered when $p \leq 0.05.$
## Results
A total of 213 TBI patients were included in the analysis. Road Traffic Crashes (RTCs) were responsible for TBI in $72.3\%$ ($$n = 154$$) of the patients followed by assault ($$n = 45$$, $21.13\%$). 101 ($47.42\%$) of the participants had severe TBI while 112($52.58\%$) had non severe TBI. The majority of the participants were male ($$n = 189$$, $88.73\%$, M: $F = 189$:24) and aged 30 or less ($$n = 104$$, $48.83\%$). Most of the participants were casual labourers ($$n = 65$$, 30.52) and educated to primary school level ($$n = 99$$, $46.48\%$). The average age of the study population was 32.42 ±11.98 years with the minimum age of 13 and maximum age of 60. The majority of patients had closed head injuries ($$n = 115$$, $53.99\%$). The peak time of injury among the patients was during the evening hours between 1700hrs and 2300hrs with $51.5\%$ ($$n = 103$$) of injuries during this time of the day. The average time spent in the study by the participants was 3.7 ± 2.4 days. ( Table 1) The average length of time spent in hospital before discharge was 4 (± 1.79) days.
The majority 66($31\%$) of the patients were discharged within 4 days of stay while 21 ($10\%$) of the participants were discharged after 4 days. The average time spent in study before death occurred was 2.1 (± 1.98) days. In the majority (107, $50.71\%$) of the patients, the fibrinogen levels were between 2–4.5g/L. The maximum level observed was 7.81g/L. The minimum values were as low as < 1g/L. Levels < 2g/L were observed in 74 ($35.07\%$) of the patients. Levels of fibrinogen > 4.5 g/L were observed in 30($14.22\%$) of the patients. The 7-day mortality rate was $34.3\%$. Forty-seven patients (47,$64.38\%$) died at Casualty within 24hrs of admission, 14 ($19.18\%$) died in the Neurosurgery unit and 12($16.44\%$) in the ICU.
## Discussion
This study set out to determine the specificity and sensitivity of fibrinogen levels and the association with severity and mortality among TBI patients at Mulago Hospital. There are no studies addressing this topic from sub-Saharan Africa and this is the first study in Uganda to describe this relationship. Despite numerous evidence available regarding the predictability of fibrinogen in the prognosis of TBI outcomes from prior studies[19, 33], little is known concerning its predictive ability in the diagnosis of severe TBI.
The sensitivity using fibrinogen < 2g/L(hypofibrinogenemia) was $56.5\%$ with a positive predictive value of $64.9\%$. The specificity was $72.9\%$ with a negative predictive value of $61.7\%$. In addition, hypofibrinogenemia (fibrinogen levels < 2g/L) was common in TBI patients occurring in $35.07\%$ of TBI patients on admission. This is similar to $38.6\%$ found in a previous study done by Lv, et al.[19] Our study also found that $20.4\%$ of patients with TBI had high levels of fibrinogen(> 4.5g/L).
Recent research showed that for patients admitted with severe TBI, fibrinogen levels < 2g/L on admission are strongly related to increased mortality.[19] In addition, studies have demonstrated that in severe trauma, fibrinogen is reduced to critical levels. [ 21, 34] The debate about the critical value of fibrinogen in trauma is an ongoing matter of contention.[35] Floccard et’al defined critical levels as being ≤ 1.0g/L and abnormal levels being 1.0–1.8 g/L all of which have been reported in patients with severe Trauma. [ 36]. By contrast, high fibrinogen levels have been described as being protective in patients with multiple trauma.[35] Furthermore, previous studies have demonstrated that TBI is associated with abnormalities in clot formation due to differences in fibrinogen levels among victims.[37] Fibrinogen could therefore be used as a marker or predictor of TBI severity.
The study found that low fibrinogen levels (< 2 g/L) were fairly predictive of TBI severity with an AUC = 0.656, sensitivity of $56.5\%$ and specificity of $72.9\%$. Therefore, absence of hypofibrinogenemia in TBI patients found in this study was associated with milder forms of TBI. This is consistent with what previous studies have described in severe trauma.
This study shows that the likelihood of having a severe form of TBI increases with low levels of fibrinogen(< 2g/L). Possible explanations for the above relationship stem from the presence of intracranial bleeding which is a common occurrence in TBI as noted in the CRASH trial.[38] Trauma induced coagulopathy is a crucial element in severe TBI especially when compounded with intracranial bleeding. The consumption of clotting factors and platelets in response to intracranial bleeding further lowers fibrinogen levels.[39] TBI is also associated with systemic hyperfibrinolysis which occurs in as much as $20\%$ of critical trauma patients hence lowering fibrinogen levels further. [ 40, 41] In addition, prior studies have shown that fibrinogen levels drop drastically and most rapidly during haemorrhage.[34] Therefore, the association of low fibrinogen levels with severe TBI is possibly due to a combination of haemorrhage and trauma induced coagulopathy that occurs in severe TBI.[42] Much as there is a stronger association of severity with low fibrinogen levels, high fibrinogen levels (> 4.5g/L) were also associated with severe TBI. This could be due to the inflammation that accompanies major trauma and disruption of the blood brain barrier with release of procoagulant molecules.[37, 43] A study done by Samuels et al found that TBI patients commonly presented with a spectrum ranging from hypocoagulability to hypercoagulability.[37] *It is* therefore likely that severe TBI without intracranial haemorrhage leads to high fibrinogen levels while severe TBI with haemorrhage lowers the fibrinogen levels.
Unlike prior research findings, [19] this study showed that fibrinogen levels > 4.5g/L were strong predictors of mortality. A possible explanation of this observation starts with trauma induced disruption of the Blood Brain Barrier that incites an extensive inflammatory response.[16, 44] Such extensive inflammation can be caused by neural cell death with resultant secondary brain oedema.[43] Diffuse Axonal *Injury is* of utmost importance here since it is characterized by an intense inflammatory response. [ 45] *It is* this inflammatory process that is responsible for the elevated fibrinogen levels. Organ dysfunction then ensues from this trauma induced systemic inflammatory state.[46] The development of organ dysfunction is related to the intensity of the trauma induced inflammatory response.[47] Hence, a severe systemic inflammatory response due to a disrupted blood brain barrier causes early organ dysfunction and later multiple organ failure which leads to death.[47, 48]. This possibly explains the high fibrinogen levels found to be associated with mortality.
While one of the strengths of our study is that it was carried out prospectively in a high-volume trauma center, some limitations need to be acknowledged. This was a single centre study with a rather small sample size and with limited duration allocated to conduct the study due to specified time frames for research activities. Secondly, fibrinogen is not a routine test in Mulago Hospital for TBI patients and replacement therapy is not currently part of the management protocols in patients with TBI; hence, despite the identification of abnormalities among the participants, correction therapy with concentrate was not possible for the participants. Patients with coagulopathies however received other supplements such as tranexamic acid and Fresh frozen plasma. Additional prospective studies with larger sample size and longer study duration are needed to confirm the predictability of TBI severity and clinical outcomes using fibrinogen levels.
## Conclusions
In conclusion, we established that fibrinogen is a useful tool in predicting severity of TBI and mortality. The study reveals that the sensitivity of fibrinogen levels < 2g/L is $56.5\%$ and the specificity is $72.9\%$. Fibrinogen fairly predicts TBI severity with an AUC of 0.656.
Fibrinogen levels may be used as an additional tool to screen TBI patients for injury severity. Low fibrinogen levels (< 2g/L) are predictors of TBI severity. High fibrinogen levels > 4.5g/L are also predictors of TBI severity. A fibrinogen level of > 4.5g/L is a strong predictor of mortality in TBI patients. Integrating fibrinogen as a biomarker in TBI management could therefore provide critical information about trauma physiology and ultimately influence clinical decisions. Additional larger prospective studies are needed to confirm these findings.
## Funding
This research was supported by the National Institute of Neurological Disorders and Stroke and Stroke of the under-award Number D43NS118560. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institute of Health.
## Availability of data and materials
The datasets generated and/or analyzed during the current study are not publicly available due to confidentiality agreements but are available from the corresponding author on reasonable request.
## References
1. Bickler SN, Weiser TG, Kassebaum N, Higashi H, Chang DC, Barendregt JJ, Debas HT, Donkor P, Gawande A, Jamison DT, Kruk ME, Mock CN. *Essential Surgery: Disease Control Priorities* (2015) **1**
2. Bryan-Hancock C, Harrison J. **The global burden of traumatic brain injury: preliminary results from the Global Burden of Disease Project**. *Inj Prev* (2010) **16** A17-A
3. Kobusingye OC, Guwatudde D, Owor G, Lett RR. **Citywide trauma experience in Kampala, Uganda: a call for intervention**. *Inj Prev* (2002) **8** 133-6. PMID: 12120832
4. Mehmood A, Zia N, Hoe C, Kobusingye O, Ssenyojo H, Hyder AA. **Traumatic brain injury in Uganda: exploring the use of a hospital based registry for measuring burden and outcomes**. *BMC Res Notes* (2018) **11** 299. PMID: 29764476
5. Khellaf A, Khan DZ, Helmy A. **Recent advances in traumatic brain injury**. *J Neurol* (2019) **266** 2878-89. PMID: 31563989
6. Frost RB, Farrer TJ, Primosch M, Hedges DW. **Prevalence of traumatic brain injury in the general adult population: a meta-analysis**. *Neuroepidemiology* (2013) **40** 154-9. PMID: 23257914
7. de Luna W, Aglio LS, Aglio LS, Urman RD, Lekowski RW. *Essential Clinical Anesthesia Review: Keywords, Questions and Answers for the Boards* (2015) 510-2
8. Vella MA, Crandall ML, Patel MB. **Acute Management of Traumatic Brain Injury**. *Surg Clin North Am* (2017) **97** 1015-30. PMID: 28958355
9. Wärnberg Gerdin L, Khajanchi M, Kumar V, Roy N, Saha ML, Soni KD. **Comparison of emergency department trauma triage performance of clinicians and clinical prediction models: a cohort study in India**. *BMJ Open* (2020) **10** e032900
10. Raux M, Vivien B, Tourtier JP, Langeron O. **Severity assessment in trauma patient**. *Ann Fr Anesth Reanim* (2013) **32** 472-6. PMID: 23906735
11. Lavery RF, Livingston DH, Tortella BJ, Sambol JT, Slomovitz BM, Siegel JH. **The utility of venous lactate to triage injured patients in the trauma center**. *J Am Coll Surg* (2000) **190** 656-64. PMID: 10873000
12. Tenovuo O, Diaz-Arrastia R, Goldstein LE, Sharp DJ, van der Naalt J, Zasler ND. **Assessing the Severity of Traumatic Brain Injury-Time for a Change?**. *J Clin Med* (2021) **10**
13. Ackah M, Gazali Salifu M, Osei Yeboah C. **Estimated incidence and case fatality rate of traumatic brain injury among children (0–18 years) in Sub-Saharan Africa. A systematic review and meta-analysis**. *PLoS ONE* (2021) **16** e0261831. PMID: 34968399
14. Saatman KE, Duhaime AC, Bullock R, Maas AI, Valadka A, Manley GT. **Classification of traumatic brain injury for targeted therapies**. *J Neurotrauma* (2008) **25** 719-38. PMID: 18627252
15. Perel P, Roberts I, Shakur H, Thinkhamrop B, Phuenpathom N, Yutthakasemsunt S. **Haemostatic drugs for traumatic brain injury**. *Cochrane Database Syst Rev* (2010) Cd007877. PMID: 20091656
16. Schultz DR, Arnold PI. **Properties of four acute phase proteins: C-reactive protein, serum amyloid A protein, alpha 1-acid glycoprotein, and fibrinogen**. *Semin Arthritis Rheum* (1990) **20** 129-47. PMID: 1705051
17. van den Herik EG, Cheung EY, de Lau LM, den Hertog HM, Leebeek FW, Dippel DW. **Fibrinogen γ’ levels in patients with intracerebral hemorrhage**. *Thromb Res* (2012) **129** 807-9. PMID: 21917300
18. Sawamura A, Hayakawa M, Gando S, Kubota N, Sugano M, Wada T. **Disseminated intravascular coagulation with a fibrinolytic phenotype at an early phase of trauma predicts mortality**. *Thromb Res* (2009) **124** 608-13. PMID: 19660788
19. Lv K, Yuan Q, Fu P, Wu G, Wu X, Du Z. **Impact of fibrinogen level on the prognosis of patients with traumatic brain injury: a single-center analysis of 2570 patients**. *World J Emerg Surg* (2020) **15** 54. PMID: 32977824
20. McCully SP, Schreiber MA. **Traumatic brain injury and its effect on coagulopathy**. *Semin Thromb Hemost* (2013) **39** 896-901. PMID: 24114009
21. Hayakawa M, Gando S, Ono Y, Wada T, Yanagida Y, Sawamura A. **Fibrinogen level deteriorates before other routine coagulation parameters and massive transfusion in the early phase of severe trauma: a retrospective observational study**. *Semin Thromb Hemost* (2015) **41** 35-42. PMID: 25590522
22. Hayakawa M. **Dynamics of fibrinogen in acute phases of trauma**. *J Intensive Care* (2017) **5** 1-6. PMID: 28066556
23. Sørensen B, Larsen OH, Rea CJ, Tang M, Foley JH, Fenger-Eriksen C. **Fibrinogen as a hemostatic agent**. *Semin Thromb Hemost* (2012) **38** 268-73. PMID: 22460909
24. Kobusingye OC, Lett RR. **Hospital-based trauma registries in Uganda**. *J Trauma* (2000) **48** 498-502. PMID: 10744292
25. Hager K, Felicetti M, Seefried G, Platt D. **Fibrinogen and aging**. *Aging (Milano)* (1994) **6** 133-8. PMID: 7918730
26. Ignjatovic V, Ilhan A, Monagle P. **Evidence for age-related differences in human fibrinogen**. *Blood Coagul Fibrinolysis* (2011) **22** 110-7. PMID: 21192254
27. Fu A, Nair KS. **Age effect on fibrinogen and albumin synthesis in humans**. *Am J Physiol* (1998) **275** E1023-30. PMID: 9843745
28. McKee AC, Daneshvar DH. **The neuropathology of traumatic brain injury**. *Handb Clin Neurol* (2015) **127** 45-66. PMID: 25702209
29. Kish L. **SAMPLING ORGANIZATIONS AND, GROUPS OF UNEQUAL SIZES**. *Am Sociol Rev* (1965) **30** 564-72. PMID: 14325826
30. Khaled Fahim N, Negida A. **Sample Size Calculation Guide - Part 2: How to Calculate the Sample Size for an Independent Cohort Study**. *Adv J Emerg Med* (2019) **3** e12. PMID: 31172123
31. Lv K, Yuan Q, Fu P, Wu G, Wu X, Du Z. **Impact of fibrinogen level on the prognosis of patients with traumatic brain injury: a single-center analysis of 2570 patients**. *World J Emerg Surg* (2020) **15** 1-9. PMID: 31911813
32. Fenger-Eriksen C, Ingerslev J, Sørensen B. **Fibrinogen concentrate-a potential universal hemostatic agent**. *Expert Opin Biol Ther* (2009) **9** 1325-33. PMID: 19645632
33. Spahn DR, Bouillon B, Cerny V, Duranteau J, Filipescu D, Hunt BJ. **The European guideline on management of major bleeding and coagulopathy following trauma: fifth edition**. *Crit Care* (2019) **23** 98. PMID: 30917843
34. Levy JH, Welsby I, Goodnough LT. **Fibrinogen as a therapeutic target for bleeding: a review of critical levels and replacement therapy**. *Transfusion* (2014) **54** 1389-405. PMID: 24117955
35. Fries D, Martini WZ. **Role of fibrinogen in trauma-induced coagulopathy**. *Br J Anaesth* (2010) **105** 116-21. PMID: 20627882
36. Floccard B, Rugeri L, Faure A, Saint Denis M, Boyle EM, Peguet O. **Early coagulopathy in trauma patients: an on-scene and hospital admission study**. *Injury* (2012) **43** 26-32. PMID: 21112053
37. Samuels JM, Moore EE, Silliman CC, Banerjee A, Cohen MJ, Ghasabyan A. **Severe traumatic brain injury is associated with a unique coagulopathy phenotype**. *J Trauma Acute Care Surg* (2019) **86** 686-93. PMID: 30601456
38. Roberts I, Yates D, Sandercock P, Farrell B, Wasserberg J, Lomas G. **Effect of intravenous corticosteroids on death within 14 days in 10008 adults with clinically significant head injury (MRC CRASH trial): randomised placebo-controlled trial**. *Lancet* (2004) **364** 1321-8. PMID: 15474134
39. Chhabra G, Rangarajan K, Subramanian A, Agrawal D, Sharma S, Mukhopadhayay AK. **Hypofibrinogenemia in isolated traumatic brain injury in Indian patients**. *Neurol India* (2010) **58** 756-7. PMID: 21045504
40. Kushimoto S, Shibata Y, Yamamoto Y. **Implications of fibrinogenolysis in patients with closed head injury**. *J Neurotrauma* (2003) **20** 357-63. PMID: 12866815
41. Stettler GR, Moore EE, Nunns GR, Chandler J, Peltz E, Silliman CC. **Rotational thromboelastometry thresholds for patients at risk for massive transfusion**. *J Surg Res* (2018) **228** 154-9. PMID: 29907206
42. Hiippala ST, Myllylä GJ, Vahtera EM. **Hemostatic factors and replacement of major blood loss with plasma-poor red cell concentrates**. *Anesth Analg* (1995) **81** 360-5. PMID: 7542432
43. Shi K, Zhang J, Dong JF, Shi FD. **Dissemination of brain inflammation in traumatic brain injury**. *Cell Mol Immunol* (2019) **16** 523-30. PMID: 30846842
44. Engler R. **Acute-phase proteins in inflammation**. *C R Seances Soc Biol Fil* (1995) **189** 563-78. PMID: 8564570
45. Humble SS, Wilson LD, Wang L, Long DA, Smith MA, Siktberg JC. **Prognosis of diffuse axonal injury with traumatic brain injury**. *J Trauma Acute Care Surg* (2018) **85** 155-9. PMID: 29462087
46. Chakraborty RK, Burns B. *Systemic Inflammatory Response Syndrome* (2022)
47. Sauaia A, Moore FA, Moore EE. **Postinjury Inflammation and Organ Dysfunction**. *Crit Care Clin* (2017) **33** 167-91. PMID: 27894496
48. Baker CC, Oppenheimer L, Stephens B, Lewis FR, Trunkey DD. **Epidemiology of trauma deaths**. *Am J Surg* (1980) **140** 144-50. PMID: 7396078
|
---
title: 'Brain Ventricle and Choroid Plexus Morphology as Predictor of Treatment Response:
Findings from the EMBARC Study'
authors:
- Harald Murck
- Maurizio Fava
- Cristina Cusin
- Cherise Chin Fatt
- Madhukar Trivedi
journal: Research Square
year: 2023
pmcid: PMC10002825
doi: 10.21203/rs.3.rs-2618151/v1
license: CC BY 4.0
---
# Brain Ventricle and Choroid Plexus Morphology as Predictor of Treatment Response: Findings from the EMBARC Study
## Abstract
Recent observations suggest a role of the choroid plexus (CP) and cerebral ventricle volume (CV), to identify treatment resistance of major depressive disorder (MDD). We tested the hypothesis that these markers are associated with clinical improvement in subjects from the EMBARC study, as implied by a recent pilot study. The EMBARC study characterized biological markers in a randomized placebo-controlled trial of sertraline vs. placebo in patients with MDD. Association of baseline volumes of CV, CP and of the corpus callosum (CC) with treatment response after 4 weeks treatment were evaluated. 171 subjects (61 male, 110 female) completed the 4 week assessments; gender, site and age were taken into account for this analyses. As previously reported, no treatment effect of sertraline was observed, but prognostic markers for clinical improvement were identified. Responders ($$n = 54$$) had significantly smaller volumes of the CP and lateral ventricles, whereas the volume of mid-anterior and mid-posterior CC was significantly larger compared to non-responders ($$n = 117$$). A positive correlation between CV volume and CP volume was observed, whereas a negative correlation between CV volume and both central-anterior and central-posterior parts of the CC emerged. In an exploratory way correlations between enlarged VV and CP volume on the one hand and signs of metabolic syndrome, in particular triglyceride plasma concentrations, were observed. A primary abnormality of CP function in MDD may be associated with increased ventricles, compression of white matter volume, which may affect treatment response speed or outcome. Metabolic markers may mediate this relationship.
## Introduction
The pathophysiology of major depressive disorder (MDD) is heterogeneous. The identification of effective compounds on the basis of a specific underlying neurobiology is hampered by the currently accepted definition of MDD in relevant classifications, including the DSM-5, which does not take biological differentiation into account. Importantly, this variability may not only affect the response to a given pharmacotherapy, but also the natural course of clinical change. This situation has negative implications in the context of clinical trials, in which treatment arms are compared, which may show neurobiological heterogeneity at baseline. To stratify a population on the basis of biological variables would confirm a biologically defined subtype, which is suitable for the treatment of a specific nature.
An argument, which is often brought up as a challenge is the operational complexity of such an approach. However, broadly available and easily accessible biological markers are available. Markers, which are available and have been shown to differentiate patients with depression include inflammatory markers1–5, metabolic markers, in particular those related to metabolic syndrome5, 6, and neuroendocrine1, 7, 8 characteristics. More recently, markers of autonomic regulation, including blood pressure and heart rate variability received renewed attention9–12.
Furthermore, imaging biomarkers have been characterized to differentiate the subjects with presumed different clinical response. Many of these, including volumetry of gray- or white matter segments are of high importance from a research perspective, but are difficult to assess in standard practice13, 14. A more easily accessible imaging marker, which is unfortunately frequently not reported in recent imaging studies, is cerebral ventricular volume (VV), partially by the argument that changes in ventricular volume are biologically unspecific, as many different brain areas may contribute to this phenomenon. Here we explore the alternative hypothesis that choroid plexus driven ventricular expansion results in the compression surrounding anatomical areas, making ventricular volume changes the potential primary factor. In the context of depression, VV is increased in patients with depression in comparison to healthy controls15–17 and may be related to treatment outcome18. We recently demonstrated an association between an increased choroid plexus and ventricular volume and worse treatment outcome in hospitalized patients with depression and identified moderators of this relationship19, i.e. body mass index (BMI) and the salivary aldosterone/cortisol ratio. The effect may be mediated by a compression of of corpus callosum segments, which will affect anatomical projection areas.
In this context it is important to consider that VV and the volume of the corpus callosum show short term structural plasticity. Both underly sleep-related changes20 and VV is sensitive to stress, at least in animals21. A plausible mediator of these phenomena is again the change in activity of the choroid plexus (CP). The volumetric determination of the CP is a relatively new area of investigation, but is feasible with current MRI techniques. Changes have been described in complex pain syndrome22, anorexia nervosa23, multiple sclerosis24 and most recently in major depression25 and psychosis26. Mechanistically, stress leads in an animal model to changes in gene expression of the CP of receptors, which have been linked to MDD, including 5-HT2a, 5-HT2c, glucocorticoid, TNFα, IL1β, BDNF27 as well as IL1 receptor28 and the CRH-receptor29. The choroid plexus may play a role in inducing inflammatory changes in depression and may be involved in sickness behavior2. Downstream mechanisms of the involvement of the CP are therefore at least twofold: an increased CSF release may lead to a mechanical compression of anatomical areas, which are adjacent to the ventricles30. Secondly, molecular moderators may spread into brain tissue via volume transmission31, 32. Those moderators may be produced by the CP itself or stem from the circulation.
We want to replicate our earlier findings of the relationship between clinical outcome of patients with depression on the one hand and ventricular volume, choroid plexus function and corpus callosum volume in a larger sample in this retrospective analysis from data from the EMBARC study. In an exploratory way we also correlate metabolic and autonomic markers with the volume of these anatomical areas in order to generate hypothesis of the causality of the observed relationships.
## Methods
The EMBARC study characterized biological markers in a randomized placebo-controlled trial of sertraline vs. placebo in patients with MDD for 8 weeks, followed by an additional treatment section, based on the outcome of the first 8 weeks of treatment. For consort statement see33. This trial is conducted according to the Declaration of Helsinki. It was approved by the Institutional Review Board at each clinical site. Signed informed consent was obtained from subjects in order to participate in the trial. The main objective was to identify clinical and biological moderators of treatment response34. Patients with early onset (before age 30), chronicity (episode duration > 2 years) or recurrent MDD (two or more recurrences including current episode) were enrolled.
The clinical parameter of interest was the Hamilton-depression rating scale (17 item; HAMD-17). For correlational analysis of clinical improvement we used the ratio between the HAMD-17 at outcome divided to the HAMD-17 at baseline (HAMD-17 ratio). A value of 1 means no change from baseline, a value of 0.7 means a reduction to $70\%$ of the baseline value. Response was defined as a HAMD-17 ratio ≤ 0.5.
For these primary analyses we focused on subjects, who completed the first 4 weeks of the placebo-controlled treatment phase. In the current dataset, 207 subjects had assessments with the Hamilton depression rating scale (HAMD) at baseline, of which 171 (; age 37.5 ± 13.4; HAMD-17: 18.8 ±4.7) had an assessment at week 4. We a priori chose the 4-week treatment interval in order to optimize the time for clinical improvement with the number of drop-outs. For a time course of the correlation of the HAMD-17 value with imaging parameters, which we generated as a sensitivity analysis and to show consistency, please see Table S1.
Imaging was processed as described before34, 35. Of the subjects, who completed 4 weeks of treatment, 171 also had imaging data at baseline. Association of volumes of CV, CP and of the corpus callosum (CC), with treatment response were evaluated. The relationship of the volume of choroid plexus, cerebral ventricular volumes and the corpus callosum with response after 4 weeks from baseline (≤50 % reduction of the HAMD) was assessed; gender, age, and total brain volume were taken into account for the primary MANCOVA analysis. For the analysis of correlations Pearson correlation coefficients and p-values are provided. The relationship between the volumes of the choid plexus- and ventricular volumes should be regarded as primary analysis, as this analysis serves to replicate our earlier findings. As the anatomical parameters of interest are considered to be highly correlated and therefore not independent correction for multiple testing was not performed. The correlations with metabolic and autonomic parameters have to be regarded as exploratory.
## Results
A correlation between baseline HAMD-17 and the volumes of interest was performed in order to determine potential state related effects. Choroid plexus volumes were significantly correlated with the HAMD-17 score at baseline ($$n = 217$$; right: Pearson R: 0.22, $$p \leq 0.002$$; left: Pearson $R = 0.17$, $$p \leq 0.017$$), whereas no correlation between ventricular volumes or corpus callosum sections and baseline depression severity could be detected (all $p \leq 0.20$ with the exception of the right lateral ventricle, which showed a trend toward a significant correlation (Pearson $R = 0.13$; $$p \leq 0.06$$).
Regarding the analysis of factors related to treatment outcome: No statistically significant treatment effect of sertraline was observed, as reported earlier36, but prognostic markers for therapy response were identified. Therefore, treatment was not a factor of the analyses. Comparing responders and non-responders, we adjusted for gender and age. An overall global significant difference between responders and non-responders was observed for volumetric parameters ($$p \leq 0.007$$, see table 2). Univariate analyses revealed that responders at week 4 had significantly smaller volumes of the choroid plexi and lateral ventricles, whereas the volume of mid-anterior and mid-posterior CC was significantly larger compared to non-responders (Table 2).
Vice versa, splitting the population at the median for the ventricular volume demonstrates the difference of the course of depressive symptoms between the two VV groups clearly (Fig. 1): A significant difference between HAMD-17 scores for the high vs. low VV-volume groups were observed at week 2 and week 4.
As a sensitivity analysis we also compared the anatomical structures split into responders vs. non-responders for each timepoint of the study, up to 8 weeks. Choroid plexus volumes at baseline differentiated these groups starting at week 4 up to week 8 ($p \leq 0.05$), however, other parameters did not reach statistical significance past week 4. Please see suppl. Table S1 for the stability of the correlation between volume of anatomical structures and treatment effect over time.
In addition to the comparisons between responders and non-responders correlations between baseline parameters and the HAMD-17 ratio were performed, which is independent of a chosen cut off. These correlational analyses confirmed the relationship between clinical change on one hand and ventricular volume, choroid plexus volume and CC segment volumes at baseline on the other hand (Tab. 3, Fig. 2). These data as well as the previous ones confirm the difference between responders and non-responders regarding anatomical structures and therefore the results from our pilot study. In addition, we explored other factors, which may affect the volume of the anatomical arias in an exploratory fashion. These analyses, are part of Tab.3 and need replication. We found that the volumes of both lateral ventricles were positively correlated to LDL-cholesterol and triglyceride levels. The volume of the left VV was significantly correlated and the right VV showed a trend towards a significant correlation to systolic blood pressure. A similar pattern was observed for the CP volumes. All these parameters are also positively correlated to age, which we corrected for in the primary analysis.
In order to determine the relationship between the anatomical areas of interest, ventricular volumes were correlated with CC segments and CP volumes. A significant positive correlation between CP volumes and lateral ventricle volumes was established. More importantly, a significant negative correlation between third ventricular volumes and the mid-anterior and mid-posterior CC segments, as well as a significant negative correlation between the lateral ventricles and the mid-posterior CC volume were observed (Table S2).
DTI parameters as assessed for the corpus callosum did not predict outcome. However, the volume of the mid-anterior and mid posterior CC segments, adjusted for total brain volume, correlated negatively with the axial diffusivity of these segments (mid-anterior: R = −0.42, $p \leq 0.001$, $$n = 191$$; mid-posterior: R=−0.15; $$p \leq 0.036$$, $$n = 196$$), whereas the CC-segment volumes were not associated with fractional anisotropy (for all, $p \leq 0.1$).
## Discussion
The primary outcome of this study is that an easily accessible imaging marker, i.e. lateral ventricular volumes, show a strong predictive value for the improvement of depressive symptoms in MDD patients treated with either sertraline or placebo. Mechanistically, this appears to be related to an alteration in choroid plexus function, both of which may affect corpus callosum integrity.
The strong relationship of ventricular volume to choroid plexus volume on one hand and the volume of CC segments on the other hand could be of theoretical interest for the pathophysiology of some forms of MDD. A working hypothesis could be that changes in choroid plexus function, i.e. an increased release of CSF volume19, 37, or an increased release of specific bioactive molecules, including inflammation mediators24, 31, may lead to a change in white matter volume and/or integrity. The increased ventricular volume or, alternatively, such bioactive molecules may affect white matter function either by mechanical compression or an effect on white matter integrity via alternations of oligodendrocyte function. This could be related to changes in myelination or changes in the volume regulation of axons within the CC. Disturbance of white matter integrity has indeed frequently been described in patients with depressive disorders, mainly by using diffusion tensor imaging (DTI) methods38–43. In support of the hypothesis of the choroid plexus involvement in this pathway: the activity of the choroid plexus is affected by neuroendocrine influences, which have been linked to MDD, in particular vasopressin and aldosterone19, 37, as well as metabolic markers related to an increased BMI19, 44. These findings are also in line with the role of inflammation as both aldosterone45–47 and high BMI48–51 show a close association to increased inflammation. Finally, inflammation has recently been associated with increased choroid plexus volume in patients with depression25 and multiple sclerosis24.
Our observation that the HDRS-17 score correlates significantly to choroid plexus volumes at baseline, only by trend to ventricular volumes and not to CC segment volumes implies that choroid plexus volume shows a state characteristic, and that ventricular volume shows somewhat lesser plasticity in relationship to mood and may have a more trait/chronicity related characteristic. Corpus callosum segments volume furthermore appear mainly to be trait- or risk markers. As mentioned in the introduction, stress leads to an increase in ventricular volume in animals21. Childhood abuse has been related to later life increase ventricular volumes and reduced white matter volume52, 53 and also the therapy refractoriness in depression54, 55. This could imply that early life stress affects ventricular and white matter structure via a prolonged choroid plexus activation. However, a recent analysis, based on the self-report depression scale QIDS-SR did not confirm a difference regarding subjects with and without childhood adversity regarding clinical response56.
Other factors, which determine the size of the ventricles, which are potentially mediated via choroid plexus alterations are related to metabolic disorders. In particular, high fat diet is related to an increased ventricular volume in animals in the context of traumatic stress57. The finding of the correlation between triglyceride levels and systolic blood pressure at baseline on one hand and choroid plexus volumes and ventricular volumes on the other hand reported here confirm the influence of metabolic parameters to differentiate patients with depression1. Interestingly, and a link between increased ventricular volume and metabolic dysfunction, in particular hyperlipidemia, in subjects with normal pressure hydrocephalus58 has also been observed. Similarly, in our pilot study we previously described a strong correlation between BMI and both choroid plexus- and ventricular volume19.
As mentioned, markers of inflammation and metabolic disturbances are preferentially present in subjects with atypical depression, in comparison both to healthy subjects and patients with melancholic depression1, 59, 60. This is in line with the current findings, as atypical depression appears to be less responsive to standard antidepressant treatment61. Of importance, patients with atypical depression show in general an earlier age of onset62. The current study only enrolled patients with an age of onset ≤ 30 years, which means that there is probably an enrichment of this subtype in comparison to the general population. Age of onset appears to be associated with specific neurobiological differences in depression63, which may be related to alterations in autonomic function and endocrine characteristics. Whether age of onset also differentiates brain morphology needs further confirmation.
Regarding the relationship of DTI parameters, no relationship with clinical change was observed. This is in contrast to studies, which reported DTI parameters as predictive for response, for example to ketamine64, 65. Nevertheless, we observed that the volume of CC segments correlated inversely with axial diffusivity (AD), i.e. a smaller CC segment volume was correlated to an increased AD. An earlier DTI report from the EMBARC study, which focused on the structural connectivity in specific anatomical areas did find an increase in fractional anisotropy (FA) in non-remitters35. As AD and FA are correlated, this outcome appears consistent, but is nevertheless in contrast to a number of earlier cited findings39, 40, 42, 43, 66. This shows the importance to take into consideration that FA and AD and other DTI markers can be influenced by varying mechanisms, which depend on one hand on the structural integrity of an axon, but also an axonal density67.
Limitations of the study are the post hoc nature of the current analyses, however, they were motivated by the attempt to replicate data from an earlier study19 and the primary variables of interest are identical. Therefore, with all caution, the current analysis overall confirms the previous pilot study. It has, however, to be considered that inclusion/exclusion criteria differ between the studies.
In conclusion, we (re-)identified an easily accessible imaging marker which appears to be related to the clinical course of depression. Ventricular volume may affect other imaging parameters and should therefore be taken into account in future imaging studies, at least in studies in MDD. In addition, the current findings go beyond a strictly descriptive association. With the additional observation of the relationship of increased ventricular volumes and increased choroid plexus volumes, our findings provide a plausible hypothesis, how neuroendocrine and metabolic parameters mechanistically influence depressive symptoms. A new focus on choroid plexus function in stress-related disorders appears to be supported.
## Conflict of interest:
HM: full time employee at Reviva Pharmaceuticals. He also is the owner of Murck-Neuroscience LLC, which develops a patent in the area of major depression.
MF: lifetime disclosures: Research Support: Abbott Laboratories; Acadia Pharmaceuticals; Alkermes, Inc.; American Cyanamid; Aspect Medical Systems; AstraZeneca; Avanir Pharmaceuticals; AXSOME Therapeutics; BioClinica, Inc; Biohaven; BioResearch; BrainCells Inc.; Bristol-Myers Squibb; CeNeRx BioPharma; Centrexion Therapeutics Corporation; Cephalon; Cerecor; Clarus Funds; Clexio Biosciences; Clintara, LLC; Covance; Covidien; Eli Lilly and Company;EnVivo Pharmaceuticals, Inc.; Euthymics Bioscience, Inc.; Forest Pharmaceuticals, Inc.; FORUM Pharmaceuticals; Ganeden Biotech, Inc.; Gentelon, LLC; GlaxoSmithKline; Harvard Clinical Research Institute; Hoffman-LaRoche; Icon Clinical Research; Indivior; i3 Innovus/Ingenix; Janssen R&D, LLC; Jed Foundation; Johnson & Johnson Pharmaceutical Research & Development; Lichtwer Pharma GmbH; Lorex Pharmaceuticals; Lundbeck Inc.; Marinus Pharmaceuticals; MedAvante; Methylation Sciences Inc; National Alliance for Research on Schizophrenia & Depression (NARSAD); National Center for Complementary and Alternative Medicine (NCCAM); National Coordinating Center for Integrated Medicine (NiiCM); National Institute of Drug Abuse (NIDA); National Institutes of Health; National Institute of Mental Health (NIMH); Neuralstem, Inc.; NeuroRx; Novartis AG; Novaremed; Organon Pharmaceuticals; Otsuka Pharmaceutical Development, Inc.; PamLab, LLC.; Pfizer Inc.; Pharmacia-Upjohn; Pharmaceutical Research Associates., Inc.; Pharmavite® LLC; PharmoRx Therapeutics; Photothera; Praxis Precision Medicines; Premiere Research International; Protagenic Therapeutics, Inc.; Reckitt Benckiser; Relmada Therapeutics Inc.; Roche Pharmaceuticals; RCT Logic, LLC (formerly Clinical Trials Solutions, LLC); Sanofi-Aventis US LLC; Shenox Pharmaceuticals, LLC; Shire; Solvay Pharmaceuticals, Inc.; Stanley Medical Research Institute (SMRI); Synthelabo; Taisho Pharmaceuticals; Takeda Pharmaceuticals; Tal Medical; VistaGen; WinSanTor, Inc.; Wyeth- Ayerst Laboratories; Advisory Board/Consultant: Abbott Laboratories; Acadia; Aditum Bio Management Company, LLC; Affectis Pharmaceuticals AG; Alfasigma USA, Inc.; Alkermes, Inc.; Altimate Health Corporation; Amarin Pharma Inc.; Amorsa Therapeutics, Inc.; Ancora Bio, Inc.; Angelini S.p. A; Aptinyx Inc.; Arbor Pharmaceuticals, LLC; Aspect Medical Systems; Astella Pharma Global Development, Inc.; AstraZeneca; Auspex Pharmaceuticals; Avanir Pharmaceuticals; AXSOME Therapeutics; Bayer AG; Best Practice Project Management, Inc.; Biogen; BioMarin Pharmaceuticals, Inc.; BioXcel Therapeutics; Biovail Corporation; Boehringer Ingelheim; Boston Pharmaceuticals; BrainCells Inc; Bristol-Myers Squibb; Cambridge Science Corporation; CeNeRx BioPharma; Cephalon, Inc.; Cerecor; Clexio Biosciences; Click Therapeutics, Inc; CNS Response, Inc.; Compellis Pharmaceuticals; Cybin Corporation; Cypress Pharmaceutical, Inc.; DiagnoSearch Life Sciences (P) Ltd.; Dainippon Sumitomo Pharma Co. Inc.; Dr. Katz, Inc.; Dov Pharmaceuticals, Inc.; Edgemont Pharmaceuticals, Inc.; Eisai Inc.; Eli Lilly and Company; ElMindA; EnVivo Pharmaceuticals, Inc.; Enzymotec LTD; ePharmaSolutions; EPIX Pharmaceuticals, Inc.; Esthismos Research, Inc.; Euthymics Bioscience, Inc.; Evecxia Therapeutics, Inc.; ExpertConnect, LLC; FAAH Research Inc.; Fabre-Kramer Pharmaceuticals, Inc.; Forest Pharmaceuticals, Inc.; Forum Pharmaceuticals; Gate Neurosciences, Inc.; GenetikaPlus Ltd.; GenOmind, LLC; GlaxoSmithKline; Grunenthal GmbH; Happify; H. Lundbeck A/S; Indivior; i3 Innovus/Ingenis; Intracellular; Janssen Pharmaceutica; Jazz Pharmaceuticals, Inc.; JDS Therapeutics, LLC; Johnson & Johnson Pharmaceutical Research & Development, LLC; Knoll Pharmaceuticals Corp.; Labopharm Inc.; Lorex Pharmaceuticals; Lundbeck Inc.; Marinus Pharmaceuticals; MedAvante, Inc.; Merck & Co., Inc.; Mind Medicine Inc.; MSI Methylation Sciences, Inc.; Naurex, Inc.; Navitor Pharmaceuticals, Inc.; Nestle Health Sciences; Neuralstem, Inc.; Neurocrine Biosciences, Inc.; Neuronetics, Inc.; NextWave Pharmaceuticals; Niraxx Light Therapeutics, Inc; Northwestern University; Novartis AG; Nutrition 21; Opiant Pharmecuticals; Orexigen Therapeutics, Inc.; Organon Pharmaceuticals; Osmotica; Otsuka Pharmaceuticals; Ovid Therapeutics, Inc.; Pamlab, LLC.; Perception Neuroscience; Pfizer Inc.; PharmaStar; PharmaTher Inc.; Pharmavite® LLC.; PharmoRx Therapeutics; Polaris Partners; Praxis Precision Medicines; Precision Human Biolaboratory; Prexa Pharmaceuticals, Inc.; Protagenic Therapeutics, Inc; PPD; PThera, LLC; Purdue Pharma; Puretech Ventures; Pure Tech LYT, Inc.; PsychoGenics; Psylin Neurosciences, Inc.; RCT Logic, LLC (formerly Clinical Trials Solutions, LLC); Relmada Therapeutics, Inc.; Rexahn Pharmaceuticals, Inc.; Ridge Diagnostics, Inc.; Roche; Sanofi-Aventis US LLC.; Sensorium Therapeutics; Sentier Therapeutics; Sepracor Inc.; Servier Laboratories; Schering-Plough Corporation; Shenox Pharmaceuticals, LLC; Solvay Pharmaceuticals, Inc.; Somaxon Pharmaceuticals, Inc.; Somerset Pharmaceuticals, Inc.; Sonde Health; Sunovion Pharmaceuticals; Supernus Pharmaceuticals, Inc.; Synthelabo; Taisho Pharmaceuticals; Takeda Pharmaceutical Company Limited; Tal Medical, Inc.; Tetragenex; Teva Pharmaceuticals; TransForm Pharmaceuticals, Inc.; Transcept Pharmaceuticals, Inc.; University of Michigan, Department of Psychiatry; Usona Institute, Inc.; Vanda Pharmaceuticals, Inc.; Versant Venture Management, LLC; VistaGen; Xenon Pharmaceuticals Inc.; Speaking/Publishing: Adamed, Co; Advanced Meeting Partners; American Psychiatric Association; American Society of Clinical Psychopharmacology; AstraZeneca; Belvoir Media Group; Boehringer Ingelheim GmbH; Bristol-Myers Squibb; Cephalon, Inc.; CME Institute/Physicians Postgraduate Press, Inc.; Eli Lilly and Company; Forest Pharmaceuticals, Inc.; GlaxoSmithKline; Global Medical Education, Inc.; Imedex, LLC; MGH Psychiatry Academy/Primedia; MGH Psychiatry Academy/Reed Elsevier; Novartis AG; Organon Pharmaceuticals; Pfizer Inc.; PharmaStar; United BioSource, Corp.; Wyeth-Ayerst Laboratories; Equity Holdings: Compellis; Neuromity; Psy Therapeutics; Sensorium Therapeutics; Royalty/patent, other income: Patents for Sequential Parallel Comparison Design (SPCD), licensed by MGH to Pharmaceutical Product Development, LLC (PPD) (US_7840419, US_7647235, US_7983936, US_8145504, US_8145505); and patent application for a combination of *Ketamine plus* Scopolamine in Major Depressive Disorder (MDD), licensed by MGH to Biohaven. Patents for pharmacogenomics of Depression Treatment with Folate (US_9546401, US_9540691). Copyright: for the MGH Cognitive & Physical Functioning Questionnaire (CPFQ), Sexual Functioning Inventory (SFI), Antidepressant Treatment Response Questionnaire (ATRQ), Discontinuation-Emergent Signs & Symptoms (DESS), Symptoms of Depression Questionnaire (SDQ), and SAFER; Belvoir; Lippincott, Williams & Wilkins; Wolkers Kluwer; World Scientific Publishing Co. Pte. Ltd.
CCF: nothing to disclose CC: personal fees from Janssen, Perception, and Takeda; and grants from Clexio, Livanova, AFSP, and the National Institute of Mental Health.
MHT: research support from the Agency for Healthcare Research and Quality, Cyberonics Inc., National Alliance for Research in Schizophrenia and Depression, NIMH, National Institute on Drug Abuse, National Institute of Diabetes and Digestive and Kidney Diseases, and Johnson & Johnson; consulting and speaker fees from Abbott Laboratories Inc., Akzo (Organon Pharmaceuticals Inc.), Allergan Sales LLC, Alkermes, Astra Zeneca, Axon Advisors, Brintellix, Bristol-Myers Squibb Company, Cephalon Inc., Cerecor, Eli Lilly & Company, Evotec, Fabre Kramer Pharmaceuticals Inc., Forest Pharmaceuticals, GlaxoSmithKline, Health Research Associates, Johnson & Johnson, Lundbeck, MedAvante Medscape, Medtronic, Merck, Mitsubishi Tanabe Pharma Development America Inc., MSI Methylation Sciences Inc., Nestle Health Science-PamLab Inc., Naurex, Neuronetics, One Carbon Therapeutics Ltd, Otsuka Pharmaceuticals, Pamlab, Parke-Davis Pharmaceuticals Inc., Pfizer Inc., PgxHealth, Phoenix Marketing Solutions, Rexahn Pharmaceuticals, Ridge Diagnostics, Roche Products Ltd, Sepracor, SHIRE Development, Sierra, SK Life and Science, Sunovion, Takeda, Tal Medical/Puretech Venture, Targacept, Transcept, VantagePoint, Vivus, and Wyeth- Ayerst Laboratories.
## References
1. Lamers F, Vogelzangs N, Merikangas KR, de Jonge P, Beekman AT, Penninx BW. **Evidence for a differential role of HPA-axis function, inflammation and metabolic syndrome in melancholic versus atypical depression**. *Mol Psychiatry* (2013) **18** 692-699. PMID: 23089630
2. Dantzer R, O’Connor JC, Freund GG, Johnson RW, Kelley KW. **From inflammation to sickness and depression: when the immune system subjugates the brain**. *Nat Rev Neurosci* (2008) **9** 46-56. PMID: 18073775
3. Miller AH, Maletic V, Raison CL. **Inflammation and its discontents: the role of cytokines in the pathophysiology of major depression**. *Biol Psychiatry* (2009) **65** 732-741. PMID: 19150053
4. Raison CL, Rutherford RE, Woolwine BJ, Shuo C, Schettler P, Drake DF. **A randomized controlled trial of the tumor necrosis factor antagonist infliximab for treatment-resistant depression: the role of baseline inflammatory biomarkers**. *JAMA psychiatry* (2013) **70** 31-41. PMID: 22945416
5. Richter N, Juckel G, Assion HJ. **Metabolic syndrome: a follow-up study of acute depressive inpatients**. *Eur Arch Psychiatry Clin Neurosci* (2010) **260** 41-49. PMID: 19399357
6. Lamers F, de Jonge P, Nolen WA, Smit JH, Zitman FG, Beekman AT. **Identifying depressive subtypes in a large cohort study: results from the Netherlands Study of Depression and Anxiety (NESDA)**. *J Clin Psychiatry* (2010) **71** 1582-1589. PMID: 20673552
7. Juruena MF, Cleare AJ, Papadopoulos AS, Poon L, Lightman S, Pariante CM. **The prednisolone suppression test in depression: dose-response and changes with antidepressant treatment**. *Psychoneuroendocrinology* (2010) **35** 1486-1491. PMID: 20558006
8. Murck H, Schussler P, Steiger A. **Renin-angiotensin-aldosterone system: the forgotten stress hormone system: relationship to depression and sleep**. *Pharmacopsychiatry* (2012) **45** 83-95. PMID: 22290200
9. Buttner M, Jezova D, Greene B, Konrad C, Kircher T, Murck H. **Target-based biomarker selection - Mineralocorticoid receptor-related biomarkers and treatment outcome in major depression**. *J Psychiatr Res* (2015) **66–67** 24-37
10. Kemp AH, Quintana DS, Gray MA, Felmingham KL, Brown K, Gatt JM. **Impact of depression and antidepressant treatment on heart rate variability: a review and meta-analysis**. *Biol Psychiatry* (2010) **67** 1067-1074. PMID: 20138254
11. Licht CM, de Geus EJ, Zitman FG, Hoogendijk WJ, van Dyck R, Penninx BW. **Association between major depressive disorder and heart rate variability in the Netherlands Study of Depression and Anxiety (NESDA)**. *Arch Gen Psychiatry* (2008) **65** 1358-1367. PMID: 19047522
12. Engelmann J, Murck H, Wagner S, Zillich L, Streit F, Herzog DP. **Routinely accessible parameters of mineralocorticoid receptor function, depression subtypes and response prediction: a post-hoc analysis from the early medication change trial in major depressive disorder**. *World J Biol Psychiatry* (2022) 1-12
13. Drevets WC, Ongur D, Price JL. **Neuroimaging abnormalities in the subgenual prefrontal cortex: implications for the pathophysiology of familial mood disorders**. *Mol Psychiatry* (1998) **3** 220-226. PMID: 9672897
14. Samann PG, Hohn D, Chechko N, Kloiber S, Lucae S, Ising M. **Prediction of antidepressant treatment response from gray matter volume across diagnostic categories**. *Eur Neuropsychopharmacol* (2013) **23** 1503-1515. PMID: 23920122
15. Schlegel S, Maier W, Philipp M, Aldenhoff JB, Heuser I, Kretzschmar K. **Computed tomography in depression: association between ventricular size and psychopathology**. *Psychiatry Res* (1989) **29** 221-230. PMID: 2798601
16. Kempton MJ, Salvador Z, Munafo MR, Geddes JR, Simmons A, Frangou S. **Structural neuroimaging studies in major depressive disorder. Meta-analysis and comparison with bipolar disorder**. *Arch Gen Psychiatry* (2011) **68** 675-690. PMID: 21727252
17. Via E, Cardoner N, Pujol J, Martinez-Zalacain I, Hernandez-Ribas R, Urretavizacaya M. **Cerebrospinal fluid space alterations in melancholic depression**. *PLoS One* (2012) **7** e38299. PMID: 22761673
18. Cardoner N, Pujol J, Vallejo J, Urretavizcaya M, Deus J, Lopez-Sala A. **Enlargement of brain cerebrospinal fluid spaces as a predictor of poor clinical outcome in melancholia**. *J Clin Psychiatry* (2003) **64** 691-697. PMID: 12823084
19. Murck H, Luerweg B, Hahn J, Braunisch M, Jezova D, Zavorotnyy M. **Ventricular volume, white matter alterations and outcome of major depression and their relationship to endocrine parameters - A pilot study**. *World J Biol Psychiatry* (2020) 1-15
20. Bernardi G, Cecchetti L, Siclari F, Buchmann A, Yu X, Handjaras G. **Sleep reverts changes in human gray and white matter caused by wake-dependent training**. *Neuroimage* (2016) **129** 367-377. PMID: 26812659
21. Henckens MJ, van der Marel K, van der Toorn A, Pillai AG, Fernandez G, Dijkhuizen RM. **Stress-induced alterations in large-scale functional networks of the rodent brain**. *Neuroimage* (2015) **105** 312-322. PMID: 25462693
22. Zhou G, Hotta J, Lehtinen MK, Forss N, Hari R. **Enlargement of choroid plexus in complex regional pain syndrome**. *Scientific reports* (2015) **5** 14329. PMID: 26388497
23. Lavagnino L, Amianto F, Mwangi B, D’Agata F, Spalatro A, Zunta-Soares GB. **Identifying neuroanatomical signatures of anorexia nervosa: a multivariate machine learning approach**. *Psychol Med* (2015) **45** 2805-2812. PMID: 25990697
24. Fleischer V, Gonzalez-Escamilla G, Ciolac D, Albrecht P, Kury P, Gruchot J. **Translational value of choroid plexus imaging for tracking neuroinflammation in mice and humans**. *Proc Natl Acad Sci U S A* (2021) **118**
25. Althubaity N, Schubert J, Martins D, Yousaf T, Nettis MA, Mondelli V. **Choroid plexus enlargement is associated with neuroinflammation and reduction of blood brain barrier permeability in depression**. *Neuroimage Clin* (2022) **33** 102926. PMID: 34972034
26. Senay O, Seethaler M, Makris N, Yeterian E, Rushmore J, Cho KIK. **A preliminary choroid plexus volumetric study in individuals with psychosis**. *Hum Brain Mapp* (2023)
27. Sathyanesan M, Girgenti MJ, Banasr M, Stone K, Bruce C, Guilchicek E. **A molecular characterization of the choroid plexus and stress-induced gene regulation**. *Transl Psychiatry* (2012) **2** e139. PMID: 22781172
28. Wong ML, Licinio J. **Localization of interleukin 1 type I receptor mRNA in rat brain**. *Neuroimmunomodulation* (1994) **1** 110-115. PMID: 7489320
29. Wong ML, Licinio J, Pasternak KI, Gold PW. **Localization of corticotropin-releasing hormone (CRH) receptor mRNA in adult rat brain by in situ hybridization histochemistry**. *Endocrinology* (1994) **135** 2275-2278. PMID: 7956950
30. Murck H, Lehr L, Jezova D. **A viewpoint on aldosterone and BMI related brain morphology in relation to treatment outcome in patients with major depression**. *J Neuroendocrinol* (2022) e13219. PMID: 36539978
31. Dantzer R.. **Cytokine-induced sickness behavior: mechanisms and implications**. *Ann N Y Acad Sci* (2001) **933** 222-234. PMID: 12000023
32. Skipor J, Thiery JC. **The choroid plexus--cerebrospinal fluid system: undervaluated pathway of neuroendocrine signaling into the brain**. *Acta Neurobiol Exp (Wars)* (2008) **68** 414-428. PMID: 18668165
33. Cooper CM, Chin Fatt CR, Jha M, Fonzo GA, Grannemann BD, Carmody T. **Cerebral Blood Perfusion Predicts Response to Sertraline versus Placebo for Major Depressive Disorder in the EMBARC Trial**. *EClinicalMedicine* (2019) **10** 32-41. PMID: 31193824
34. Trivedi MH, McGrath PJ, Fava M, Parsey RV, Kurian BT, Phillips ML. **Establishing moderators and biosignatures of antidepressant response in clinical care (EMBARC): Rationale and design**. *J Psychiatr Res* (2016) **78** 11-23. PMID: 27038550
35. Pillai RLI, Huang C, LaBella A, Zhang M, Yang J, Trivedi M. **Examining raphe-amygdala structural connectivity as a biological predictor of SSRI response**. *J Affect Disord* (2019) **256** 8-16. PMID: 31158720
36. Webb CA, Trivedi MH, Cohen ZD, Dillon DG, Fournier JC, Goer F. **Personalized prediction of antidepressant v. placebo response: evidence from the EMBARC study**. *Psychol Med* (2019) **49** 1118-1127. PMID: 29962359
37. Sheldon CA, Kwon YJ, Liu GT, McCormack SE. **An integrated mechanism of pediatric pseudotumor cerebri syndrome: evidence of bioenergetic and hormonal regulation of cerebrospinal fluid dynamics**. *Pediatr Res* (2015) **77** 282-289. PMID: 25420176
38. Benedetti F, Yeh PH, Bellani M, Radaelli D, Nicoletti MA, Poletti S. **Disruption of white matter integrity in bipolar depression as a possible structural marker of illness**. *Biol Psychiatry* (2011) **69** 309-317. PMID: 20926068
39. Chen G, Guo Y, Zhu H, Kuang W, Bi F, Ai H. **Intrinsic disruption of white matter microarchitecture in first-episode, drug-naive major depressive disorder: A voxel-based meta-analysis of diffusion tensor imaging**. *Prog Neuropsychopharmacol Biol Psychiatry* (2017) **76** 179-187. PMID: 28336497
40. de Diego-Adelino J, Pires P, Gomez-Anson B, Serra-Blasco M, Vives-Gilabert Y, Puigdemont D. **Microstructural white-matter abnormalities associated with treatment resistance, severity and duration of illness in major depression**. *Psychol Med* (2014) **44** 1171-1182. PMID: 23962469
41. Guo WB, Liu F, Xue ZM, Gao K, Wu RR, Ma CQ. **Altered white matter integrity in young adults with first-episode, treatment-naive, and treatment-responsive depression**. *Neurosci Lett* (2012) **522** 139-144. PMID: 22721700
42. Repple J, Meinert S, Grotegerd D, Kugel H, Redlich R, Dohm K. **A voxel-based diffusion tensor imaging study in unipolar and bipolar depression**. *Bipolar Disord* (2017) **19** 23-31. PMID: 28239946
43. Wise T, Radua J, Nortje G, Cleare AJ, Young AH, Arnone D. **Voxel-Based Meta-Analytical Evidence of Structural Disconnectivity in Major Depression and Bipolar Disorder**. *Biol Psychiatry* (2016) **79** 293-302. PMID: 25891219
44. Nilsson C, Lindvall-Axelsson M, Owman C. **Neuroendocrine regulatory mechanisms in the choroid plexus-cerebrospinal fluid system**. *Brain Res Brain Res Rev* (1992) **17** 109-138. PMID: 1393190
45. Rocha R, Rudolph AE, Frierdich GE, Nachowiak DA, Kekec BK, Blomme EA. **Aldosterone induces a vascular inflammatory phenotype in the rat heart**. *Am J Physiol Heart Circ Physiol* (2002) **283** H1802-1810. PMID: 12384457
46. Hlavacova N, Wes PD, Ondrejcakova M, Flynn ME, Poundstone PK, Babic S. **Subchronic treatment with aldosterone induces depression-like behaviours and gene expression changes relevant to major depressive disorder**. *Int J Neuropsychopharmacol* (2012) **15** 247-265. PMID: 21375792
47. Bay-Richter C, Hallberg L, Ventorp F, Janelidze S, Brundin L. **Aldosterone synergizes with peripheral inflammation to induce brain IL-1beta expression and depressive-like effects**. *Cytokine* (2012) **60** 749-754. PMID: 22981168
48. Miller GE, Freedland KE, Carney RM, Stetler CA, Banks WA. **Pathways linking depression, adiposity, and inflammatory markers in healthy young adults**. *Brain Behav Immun* (2003) **17** 276-285. PMID: 12831830
49. Cooper JN, Tepper P, Barinas-Mitchell E, Woodard GA, Sutton-Tyrrell K. **Serum aldosterone is associated with inflammation and aortic stiffness in normotensive overweight and obese young adults**. *Clin Exp Hypertens* (2012) **34** 63-70. PMID: 22007645
50. de Kloet AD, Pioquinto DJ, Nguyen D, Wang L, Smith JA, Hiller H. **Obesity induces neuroinflammation mediated by altered expression of the renin-angiotensin system in mouse forebrain nuclei**. *Physiol Behav* (2014)
51. Pires PW, McClain JL, Hayoz SF, Dorrance AM. **Mineralocorticoid receptor antagonism prevents obesity-induced cerebral artery remodeling and reduces white matter injury in rats**. *Microcirculation* (2018) **25** e12460. PMID: 29758591
52. De Bellis MD, Keshavan MS, Clark DB, Casey BJ, Giedd JN, Boring AM. **A.E. Bennett Research Award. Developmental traumatology. Part II: Brain development**. *Biol Psychiatry* (1999) **45** 1271-1284. PMID: 10349033
53. De Bellis MD, Zisk A. **The biological effects of childhood trauma**. *Child Adolesc Psychiatr Clin N Am* (2014) **23** 185-222. PMID: 24656576
54. Heim C, Nemeroff CB. **The role of childhood trauma in the neurobiology of mood and anxiety disorders: preclinical and clinical studies**. *Biol Psychiatry JID - 0213264* (2001) **49** 1023-1039
55. Nelson J, Klumparendt A, Doebler P, Ehring T. **Childhood maltreatment and characteristics of adult depression: meta-analysis**. *Br J Psychiatry* (2017) **210** 96-104. PMID: 27908895
56. Medeiros GC, Prueitt WL, Rush AJ, Minhajuddin A, Czysz AH, Patel SS. **Impact of childhood maltreatment on outcomes of antidepressant medication in chronic and/or recurrent depression**. *J Affect Disord* (2021) **291** 39-45. PMID: 34023746
57. Kalyan-Masih P, Vega-Torres JD, Miles C, Haddad E, Rainsbury S, Baghchechi M. **Western High-Fat Diet Consumption during Adolescence Increases Susceptibility to Traumatic Stress while Selectively Disrupting Hippocampal and Ventricular Volumes**. *eNeuro* (2016) **3**
58. Israelsson H, Carlberg B, Wikkelso C, Laurell K, Kahlon B, Leijon G. **Vascular risk factors in INPH: A prospective case-control study (the INPH-CRasH study)**. *Neurology* (2017) **88** 577-585. PMID: 28062721
59. Lamers F, Bot M, Jansen R, Chan MK, Cooper JD, Bahn S. **Serum proteomic profiles of depressive subtypes**. *Transl Psychiatry* (2016) **6** e851. PMID: 27404283
60. Lasserre AM, Glaus J, Vandeleur CL, Marques-Vidal P, Vaucher J, Bastardot F. **Depression with atypical features and increase in obesity, body mass index, waist circumference, and fat mass: a prospective, population-based study**. *JAMA psychiatry* (2014) **71** 880-888. PMID: 24898270
61. Stewart JW, McGrath PJ, Fava M, Wisniewski SR, Zisook S, Cook I. **Do atypical features affect outcome in depressed outpatients treated with citalopram?**. *Int J Neuropsychopharmacol* (2010) **13** 15-30. PMID: 19341509
62. Novick JS, Stewart JW, Wisniewski SR, Cook IA, Manev R, Nierenberg AA. **Clinical and demographic features of atypical depression in outpatients with major depressive disorder: preliminary findings from STAR*D**. *J Clin Psychiatry* (2005) **66** 1002-1011. PMID: 16086615
63. Herzog DP, Wagner S, Engelmann J, Treccani G, Dreimuller N, Muller MB. **Early onset of depression and treatment outcome in patients with major depressive disorder**. *J Psychiatr Res* (2021) **139** 150-158. PMID: 34058654
64. Sydnor VJ, Lyall AE, Cetin-Karayumak S, Cheung JC, Felicione JM, Akeju O. **Studying pre-treatment and ketamine-induced changes in white matter microstructure in the context of ketamine’s antidepressant effects**. *Transl Psychiatry* (2020) **10** 432. PMID: 33319774
65. Vasavada MM, Leaver AM, Espinoza RT, Joshi SH, Njau SN, Woods RP. **Structural connectivity and response to ketamine therapy in major depression: A preliminary study**. *J Affect Disord* (2016) **190** 836-841. PMID: 26630613
66. Guo WB, Liu F, Chen JD, Xu XJ, Wu RR, Ma CQ. **Altered white matter integrity of forebrain in treatment-resistant depression: a diffusion tensor imaging study with tract-based spatial statistics**. *Prog Neuropsychopharmacol Biol Psychiatry* (2012) **38** 201-206. PMID: 22504778
67. Bishop JH, Shpaner M, Kubicki A, Clements S, Watts R, Naylor MR. **Structural network differences in chronic muskuloskeletal pain: Beyond fractional anisotropy**. *Neuroimage* (2018) **182** 441-455. PMID: 29242104
68. Murck H, Fava M, Cusin C, Chin Fatt C, trivedi M. **Brain Ventricle Morphology as Predictor of Treatment-Response-Findings From the EMBARC-Study**. *Biological Psychiatry* (2021) **89** S367-S368
|
---
title: Two concurrent randomized controlled trials of CommunityRx, a social care intervention
for family and friend caregivers delivered at the point of care
authors:
- Emily Marie Abramsohn
- MariaDelSol De Ornelas
- Soo Borson
- Cristianne RM Frazier
- Charles M Fuller
- Mellissa Grana
- Elbert S Huang
- Jyotsna S Jagai
- Jennifer A Makelarski
- Doriane Miller
- Dena Schulman-Green
- Eva Shiu
- Katherine Thompson
- Victoria Winslow
- Kristen Wroblewski
- Stacy Tessler Lindau
journal: Research Square
year: 2023
pmcid: PMC10002827
doi: 10.21203/rs.3.rs-2464681/v1
license: CC BY 4.0
---
# Two concurrent randomized controlled trials of CommunityRx, a social care intervention for family and friend caregivers delivered at the point of care
## Abstract
### Background
CommunityRx is an evidence-based social care intervention delivered to family and friend caregivers (“caregivers”) at the point of healthcare to address health-related social risks (HRSRs). CommunityRx-*Hunger is* a double-blind randomized controlled trial (RCT) that enrolls caregivers of hospitalized children. CommunityRx-*Dementia is* a single-blind RCT that enrolls caregivers of community-residing people with dementia. Clinical trials that enroll caregivers face recruitment barriers, including caregiver burden and lack of systematic strategies to identify and track caregivers. COVID-19 pandemic-related visitor restrictions exacerbated these barriers and prompted the need for iteration of the CommunityRx protocols from in-person to remote operations. This study describes the novel methods used to iterate existing RCT protocols and factors contributing to their successful iteration.
### Methods
CommunityRx uses individual-level data to generate personalized community resource referrals for basic, health and caregiving needs. Our research program uses an asset-based, community-engaged approach including study-specific community advisory boards (CABs). In early 2020, both RCT protocols were pre-tested in-person. In March 2020, when pandemic conditions prohibited enrollment during clinical encounters, both protocols were iterated to efficient, caregiver-centered remote operations. Iterations were enabled in part by the Automated Randomized Controlled Trial Information-Communication System (ARCTICS), a trial management system innovation engineered to integrate the data collection database (REDCap) with community resource referral (NowPow) and SMS texting (Mosio) platforms.
### Results
Enabled by engaged CABs and ARCTICS, both RCTs quickly adapted to remote operations. Designed before the pandemic, we had planned to launch both trials by March 2020 and complete enrollment by December 2021. The pandemic postponed launch until November (CommunityRx-Hunger) and December (CommunityRx-Dementia) 2020. Despite the delay, $65\%$ of all planned participants (CommunityRx-Hunger $$n = 417$$/640; CommunityRx-Dementia $$n = 222$$/344) were enrolled by December 2021, halfway through our projected enrollment timeline. Both trials enrolled $13\%$ more participants in 12 months than originally projected in-person.
### Conclusions
Our asset-based, community-engaged approach combined with widely accessible institutional and commercial information technologies facilitated rapid migration to remote trial operations. Remote or hybrid RCT designs for social care interventions may be a viable, scalable alternative to in-person recruitment and intervention delivery protocols, particularly for caregivers and other groups that are under-represented in traditional health services research.
### Trial Status
Both studies are registered on ClinicalTrials.gov: CommunityRx-Hunger (NCT04171999); CommunityRx for Caregivers (NCT04146545); My Diabetes My Community (NCT04970810)
## Introduction
Family and friend caregivers (“caregivers”) of people with severe or chronic illness are vulnerable to health-related socioeconomic risk factors (HRSRs) like food and housing insecurity and transportation difficulties.(1 –3) Clinical trials enrolling caregivers face particular recruitment and retention challenges including caregiver burden,(4–6) lack of caregiver identification in medical records,[5, 7] and intermittent caregiver presence at care recipients’ clinical visits.[8] The COVID-19 pandemic exacerbated HRSRs among caregivers[9] and, due to restrictions to caregiver attendance at visits, imperiled caregiver clinical trial enrollment.[10] CommunityRx is an evidence-based social care intervention, informed by self- and family management theory,[11] that systematically matches people at the point of healthcare to nearby resources for basic or health-related social needs, wellness, disease self-management and caregiving needs.(12–14) CommunityRx has been developed and iterated over more than a decade using an asset-based community-engaged approach.[13, 15] *Using this* approach - which involves solving population health problems by leveraging existing community assets and expertise - CommunityRx was designed for applicability in a wide range of contexts and for a broad spectrum of health and social conditions.[13, 15, 16] CommunityRx-Hunger[17] (Nintervention = 320, Ncontrols= 320) and CommunityRx-Dementia[18] (Nintervention = 172, Ncontrols = 172) are caregiver-centered adaptations of CommunityRx being tested in concurrent randomized controlled trials (RCTs). Both studies aim to improve caregiver self-efficacy and address HRSRs while minimizing stigma and maintaining satisfaction with care. These RCTs are unique in several ways. Both focus on African American/Black caregivers, a population under-represented, especially in dementia intervention studies. In addition, both trials are among very few in the social care field to assess outcomes over 12 months[19, 20] (most trials have 3 or 6 month follow-up). Last, few social care trials are blinded. To our knowledge, CommunityRx-*Hunger is* the first double-blind RCT of a social care intervention. CommunityRx-*Dementia is* a single-blind trial and is also unique in that it is the first social care intervention study - and one of exceedingly few dementia caregiver studies - to attempt to enroll caregivers at their own point of healthcare.
Before the pandemic, these CommunityRx trials were designed and pre-tested for in-person enrollment and intervention delivery during a child’s hospitalization (CommunityRx-Hunger) or an outpatient visit for a dementia caregiver or their care recipient (CommunityRx-Dementia). To overcome pandemic-related barriers, we rapidly pivoted to remote operations. The purpose of this study is to describe these two unique protocols and replicable strategies we implemented to sustain this research in a predominantly African American/Black community during the COVID-19 pandemic.
## The CommunityRx Intervention
The conceptual model underlying the CommunityRx interventions was drawn from Grey and colleagues’ Self- and Family Management Framework, an evidence-based framework widely used to study interventions that promote chronic condition management.[11,21] The Framework has been adapted for the CommunityRx-Hunger and -Dementia trials (Fig. 1) to include factors identified by Fundamental Cause Theory (e.g., socioeconomic status and stigma) among other known facilitators and barriers of self- and family management.(22–24) The framework also identifies evidence-based processes associated in prior studies with patient and family outcomes.[25] These processes underlie tasks, or the essential work of self- and family management, including learning about health needs and activating resources to address health and other needs.[26] The CommunityRx interventions assist patients and caregivers in learning the skills and resources needed to engage in these processes with confidence (self-efficacy).[27] Accordingly, the CommunityRx-Hunger and -Dementia interventions are comprised of three, evidence-based components that target key self- and family management processes: (a) education about the prevalence of HRSRs among caregivers; (b) activation of resources through delivery of and coaching on how to use a personalized resource “prescription” (HealtheRx, Fig. 2A); and (c) boosting of the intervention through a series of proactive text messages and ongoing navigator support. The CommunityRx intervention is delivered by a navigator (a research assistant) at the index clinical encounter - either a child’s hospital discharge (CommunityRx-Hunger) or following an outpatient visit (CommunityRx-Dementia). Initial intervention delivery includes the education and activation components of the intervention. Boosters are delivered over three months and caregiver outcomes are assessed over 12 months (Fig. 3).
## Education.
In both trials, the navigator, using a brief semi-structured script, provides caregivers with information about the prevalence of HRSRs among caregivers (to reduce stigma) and the availability of nearby community resources for assistance. The scripts for each study were developed with advisory board input and convey messages such as “caregivers can benefit from resources,” and “caring for yourself allows you to better care for your loved ones.” The scripts were designed to allow for personalization. An abridged version of the script is included on the HealtheRx (Fig. 2A).
## Resource Activation.
The activation component of the interventions aims to promote caregiver resource use through navigator-led coaching. The navigator reviews the HealtheRx with the caregiver and explains how to access resources by pointing out key resources and features (e.g., insurance accepted, hours). Caregivers are instructed to contact the navigator by text, email or phone to find additional resources. When contacted, the navigator then searches the resource referral platform to identify and share resources using the caregiver’s preferred delivery mode (email or text message).
In CommunityRx-Dementia, the navigator also uses a web-enabled tablet to demonstrate “FindRx,” a client-facing resource finder (Fig. 2B). ( FindRx was not yet available when CommunityRx-Hunger was designed.) The navigator coaches caregivers on how to use FindRx to search for and share community resources, request additional resource information and give feedback about resources.
## Boosting.
The interventions are “boosted” over 3 months with a series of automated text messages from the navigator offering caregivers ongoing support and resource information. The timing and frequency of these messages draws on evidence from the Critical Time Intervention model, which recognizes the transition from institution to community as a highly influential or teachable moment.[28] The Critical Time Intervention model uses a phased intervention approach with more frequent touchpoints early in the intervention that become less frequent over time (Fig. 3). The content of these text messages was designed to promote engagement with the navigator[29] and provide caregivers ongoing community resource information and navigational support. A prior observational study found that adding a text message component to the CommunityRx intervention increased participant engagement with the navigator by 70 fold (0.2–$14\%$).[13]
## Clinical Trial Design: Community Engagement and Innovation
The design characteristics, scientific aims and outcomes of both CommunityRx RCTs are outlined in Table 1. Each study uses stratified randomization to enroll potential participants at the point of care. Important caregiver and patient outcomes are assessed at multiple time points over 12 months following enrollment (Table 2). Participants in each study provide documentation of informed consent; both trials were approved by the University of Chicago Institutional Review Board (IRB).
CommunityRx-Hunger enrolls parents/caregivers during a child’s hospitalization at the University of Chicago Comer Children’s Hospital (Chicago, IL). CommunityRx-Dementia enrolls family/friend caregivers at the point of their own outpatient healthcare or during outpatient healthcare visits of their care recipient at The University of Chicago Medicine (UCM), a large, urban academic medical center. Both trials assess caregiver and patient/care recipient outcomes over 12 months. The study site serves a densely populated 110mi2 urban area and one of the largest contiguous African American/Black urban communities in the U.S. Here, $49\%$ of people have an annual household income < $200\%$ of the federal poverty level. Seventy-six percent of residents in the hospital’s Primary Service Area are African American/Black and $13\%$ are Hispanic.[30]
## Community Engagement.
CommunityRx uses an asset-based, community-engaged approach to research, which involves working with community members and organizations to achieve locally relevant scientific objectives.[15] Both RCTs involve community advisory boards (CABs) composed of community and clinical stakeholders, including caregivers, patients, clinicians, hospital staff, community advocacy organizations and volunteers. CABs convene regularly to review and advise on study protocols, intervention design and dissemination efforts.
CommunityRx-*Hunger is* advised by the Feed1st CAB, a group originally formed to work with researchers to combat high rates of food insecurity among people seeking healthcare, including parents and other caregivers with children admitted to our children’s hospital.[31, 32] Feed1st, in partnership with a regional food depository, hospital and medical student volunteers and others, operates 11 open-access, self-serve food pantries at the same site as the CommunityRx trials. The pantries are open $\frac{24}{7}$/365 and food is free for anyone in the hospital with no questions asked. During the pandemic, Feed1st launched 5 new pantries at UCM, including one new pantry in Comer Children’s Hospital (CommunityRx-Hunger trial site) and two new pantries in outpatient settings where we are enrolling for the CommunityRx-Dementia trial. As these two trials frequently identify people with food insecurity, Feed1st is essential to the ethical conduct of research in our setting.
The CommunityRx-Dementia CAB grew from a group originally established to advise the Supporting Healthy Aging Resources and Education (SHARE) Network, a Health Resources and Services Administration-funded program with a large network of community-based organizations serving older adults with and without dementia in the CommunityRx study area.[33] During the pandemic, to adjust to remote operations, CAB members recommended steps to ensure trial accessibility for caregivers with low technology literacy, connecting us to Tech Savvy Friends,[34] a medical student-led organization that provided technical support to caregivers who needed help with enrollment tasks, such as creating and accessing a personal email address and opening and navigating web links. Both CABs played an essential role in ensuring that changes to remote operations were caregiver-centered.
## ARCTICS: Automated Randomized Controlled Trial Intervention-Communication System.
The two trials are funded by different institutes at the National Institutes of Health (CommunityRx-Hunger by the National Institute on Minority Health and Health Disparities and CommunityRx-Dementia by the National Institute on Aging). Funding was awarded around the same time. Although not proposed in either application, we saw an innovation opportunity that would enable us to realize operational efficiency and minimize burden on participants. Drawing on experience developing the CommunityRx information technology (IT) platform and integrating it with EMR systems,[13] we created the Automated Randomized Controlled Trial Intervention-Communication System (ARCTICS), a novel application programming interface (API) and a custom middleware to enable interoperability of the survey database (REDCap[35, 36]) with the community resource referral (NowPow[37]) and text messaging (Mosio[38]) platforms (Fig. 4).[39] ARCTICS, developed in collaboration with the University of Chicago Center for Research Informatics, draws on individual-level demographic, health and social risk data captured via REDCap-administered surveys to facilitate generation and delivery of (a) personalized resource referrals (HealtheRxs) and (b) text messages to participants for intervention information, survey reminders, retention and scheduling.
## Iteration of the Trial Protocols for Remote Operation
Research protocols for both trials were initially planned for in-person administration. Both trials were pre-tested in-person in January 2020. Following the declaration of the COVID-19 pandemic on March 11, 2020, clinical trials involving in-person activities outside of routine care at UCM were suspended.[40] Furthermore, adult outpatients could not be accompanied by a caregiver for visits and hospitalized children were restricted to one parent/caregiver. Accordingly, we revised our protocols to enable remote recruitment, enrollment and intervention. These efforts were enabled by ARCTICS and an engaged clinical staff and were carried out with input from each study’s CAB.
## Recruitment.
Because we could no longer recruit caregivers in person, protocols were iterated to allow for contact via phone and text message. Researchers used demographic and emergency contact information in patients’ electronic medical records (EMR) to identify potential caregivers and facilitate recruitment. Compared to in-person pre-test data, remote pre-test data showed improvements in approach and enrollment rates in both RCTs.
For CommunityRx-Hunger, the pre-test approach rate (the number of caregivers we were able to contact to assess interest in the study) increased from $33\%$ (at hospital bedside) to $54\%$ using remote protocols. With input from the Feed1st CAB, the recruitment protocol for CommunityRx-Hunger was further modified for the full RCT to a three-pronged approach: 1) call to phone at hospital bedside; 2) text message to parent’s cell phone listed in their child’s EMR; and 3) follow-up phone call to caregiver’s cell phone.
For CommunityRx-Dementia, researchers attempted to approach all patients awaiting care in the target clinics during the in-person pre-test period (pre-pandemic). More than 1,000 patients were approached in person to assess their dementia caregiver status, ultimately enrolling 10 caregivers for this pre-test. To adjust to pandemic conditions, this recruitment protocol was iterated for remote recruitment by leveraging our EMR data warehouse and informatics innovations to help identify caregivers at their own point of care. Remote recruitment protocols included sending an introductory text message to the cell phone of the patient’s emergency contact person (listed in the EMR) and then proceeding to call and leaving a voicemail as needed.
## Screening.
In the original, in-person protocols, screening for food insecurity and other HRSRs was self-administered on a tablet at the point of care. To maintain privacy in the remote protocol, screening was conducted by phone. The percentage of caregivers screened for inclusion among those approached was $51\%$ for CommunityRx-Hunger (versus $56\%$ during in-person recruitment) and $68\%$ for CommunityRx-Dementia (versus $57\%$).
## Enrollment.
Following remote screening, caregivers were emailed or texted a link to the informed consent document. Researchers conducted the informed consent process by phone while the caregiver reviewed the form on their own device. Informed consent was documented electronically using REDCap’s e-consent Framework in accordance with FDA rule 21 CFR.[36, 37, 41] The percentage of caregivers who consented among those who were eligible was lower using remote compared to in-person protocols: remote $69\%$ ($\frac{22}{32}$) vs. in-person $77\%$ ($\frac{10}{13}$) for CommunityRx-Hunger and remote $71\%$ ($\frac{10}{14}$) vs. in-person $77\%$ ($\frac{10}{13}$) for CommunityRx-Dementia. Following consent, baseline data were collected by phone or videoconference.
Pre-pandemic, both studies were projected to launch by March 2020 and complete by December 2021 (22 months). The pandemic caused a ~ 9-month delay: CommunityRx-Hunger launched in November and CommunityRx-Dementia in December 2020. With no additional funding or time allotted for recruitment, both studies enrolled $65\%$ of their prepandemic targets in the first 9 months of a shortened (18 month) enrollment timeline and $13\%$ more participants than projected over the first 12 months of enrollment. Additionally, shifting to remote protocols did not jeopardize the diversity of our projected sample of caregivers. In fact, in both trials, we enrolled more African American/Black caregivers using remote protocols than originally estimated using in-person protocols (CommunityRx-Hunger: $58\%$ in-person versus $78\%$ currently in the RCT; CommunityRx-Dementia: $75\%$ in-person versus $86\%$ currently in the RCT).
## Intervention delivery.
Following enrollment and baseline data collection, participants were stratified by HRSR status (food secure versus food insecure for CommunityRx-Hunger and 0 HRSRs versus ≥ 1 HRSR for CommunityRx-Dementia) and randomized to either usual care or the CommunityRx intervention. The pivot to remote operations required major changes to the intervention delivery protocol that were facilitated by ARCTICS and rapid, pandemic-related uptake of videoconferencing by healthcare professionals and lay caregivers alike.
To simulate the brief in-person, face-to-face encounter that was originally planned for intervention delivery, we implemented videoconferencing. Using data that flowed through ARCTICS, navigators quickly generated a personalized HealtheRx for each caregiver. The navigator used the videoconferencing screen-sharing feature to coach the caregiver on how to use the HealtheRx (and, for CommunityRx-Dementia, how to use the online FindRx tool). When videoconferencing was infeasible (for example, the caregiver was not in the child’s hospital room to receive a tablet or did not have videoconferencing capabilities on their own device), we used phone. All caregivers, regardless of how the initial intervention was delivered, received the HealtheRx by email and text message to their mobile phone. For CommunityRx-Dementia participants, FindRx information was sent via email within a week of the index outpatient visit. This email included the caregiver’s unique login information for FindRx, a brief visual user guide with instructions on how to use the FindRx tool, a 6-minute video tutorial link and contact information for the navigator. The text message protocol remained the same as described above.
For CommunityRx-Hunger, we engaged specialists from hospital Child Life Services (CLS) to support remote intervention delivery. CLS specialists are trained professionals who routinely interact in person with patients and families before hospital discharge to help them understand their illnesses and procedures through expressive therapies, medical education and other support, often using tablets and other information technologies.[42] Given limits on family support at the bedside, CLS played a critical role in supporting hospitalized children during the pandemic.[43, 44] Following randomization, CLS specialists were dispatched by the research team to deliver a web-enabled tablet to the caregiver at a scheduled date and time prior to the patient’s discharge. They also provided technical support to the caregiver for videoconferencing. For CommunityRx-Dementia, caregivers who had trouble accessing a personal email or opening web links on their cell phone were referred to Tech Savvy Friends[34] before consenting to the study. After the caregiver received support from Tech Savvy Friends, they were re-contacted by the data collector to complete enrollment.
## Data Collection and Retention.
Item missingness in the in-person pre-test ranged from 0–$13\%$ at baseline and one week and was $0\%$ for the remote pre-test. Most retention strategies remained the same when moving from in-person to remote protocols, including the use of text messages and scheduled calls to facilitate follow-up survey completion. However, new strategies were implemented to promote retention over 12 months of follow-up. For CommunityRx-Hunger, a text message to participants between their 6- and 12-month surveys reminded study participants of their upcoming survey and confirmed their contact information. CommunityRx-Dementia implemented a 6- and 9-month check-in to facilitate retention and confirm contact information. All participants who verified their contact information were entered into a quarterly raffle in which 2 winners received a $50 gift card. CommunityRx-Dementia also implemented a graduated incentive structure wherein the compensation for each completed survey increased over time.
## Discussion
Due to the COVID-19 pandemic, many clinical trials were immediately halted or encountered long delays.[45] Published estimates show that only $40\%$ of halted non-oncology trials were reactivated as of March 2021.[46] Despite documented challenges associated with caregiver-centered research, pandemic-related delays and a clinical trial team working fully remotely, both of our trials were not only re-activated by late 2020, but yielded faster enrollment rates than projected pre-pandemic. The asset-based, community-engaged approach, combined with our innovation skills, enabled us to leverage strong, timely community advising and widely accessible institutional and commercial information technologies to facilitate rapid migration to remote trial operations. Our strategies and learnings can inform future social care interventional studies involving caregivers and other groups with limited access to traditional health services research participation.
For these two trials, rapid translation of in-person clinical trial protocols to a remote design was feasible due, in part, to a high-functioning network of stakeholders, community advisors and environmental supports in place well before the COVID-19 crisis. During the pandemic, CABs for both trials met remotely to provide ongoing support and advocacy for continuing the research. In the case of CommunityRx-Hunger, CLS specialists on the CAB were especially critical to the successful iteration of the intervention delivery protocols. Of note for pediatric trialists, CLS programs operate in more than 400 pediatric hospitals, emergency departments and community clinics in the U.S.,[47] and engagement in research on the psychosocial needs of children and families are standards of their clinical practice.[48] In the case of CommunityRx-Dementia, Tech Savvy Friends was recommended by a CAB member during a tele-convening where remote protocols were being discussed. An introduction to Tech Savvy Friends enabled us to quickly incorporate this community resource into our enrollment protocols. Additionally, because these trials were identifying people with food insecurity - and rates were rising as a result of the pandemic[49,50] - our ability to sustain and rapidly expand the Feed1st pantries[32] was important to preserving the ethical conduct of research.
With contemporaneous funding for two large RCTs, we created the ARCTICS innovation before the pandemic to realize economies of scale and minimize caregiver burden. This innovative technology, along with rapid adoption of videoconferencing, became essential to sustaining the trials remotely.[51] In addition, based on guidance from CAB members, we added an ARCTICS-driven text message to our recruitment outreach strategy to increase the likelihood that our calls would be answered. This message let each caregiver know who we were and from which number we were calling before we initiated phone outreach. While consent rates during remote pre-testing were slightly lower than in-person for both studies, our remote recruitment strategies yielded higher approach rates than in-person protocols for both trials, allowing us to approach more people at a faster rate than in-person. Our remote recruitment strategies, informed by each study’s CAB, enabled us to accommodate caregivers’ schedules and recruit them at times and in ways that were most convenient for them.
For comparison, a prior cross-sectional study of household food insecurity among parents of children admitted to the same hospital consented $85\%$ of eligible parents (versus $77\%$ for the in-person pretest and $69\%$ of the remote pretest for CommunityRx-Hunger).[31] An intervention development study of decision-making experiences of people with dementia implemented similar videoconferencing and virtual consent procedures as described hereto adapt to pandemic conditions.[52] Consent rate data for that study are not yet published. The demonstrated resilience of the ARCTICS innovation to pandemic conditions led to its adoption in other remotely-operated clinical trials, including the My Diabetes My Community (MDMC) Trial that launched in September 2021 with funding from the National Institutes of Diabetes, Digestive and Kidney Diseases.[53] Using ARCTICS, MDMC had remotely enrolled 74 of a planned 600 older patients with type 2 diabetes by December 2021.
Remote implementation of the CommunityRx studies also required major changes to the mode of intervention delivery that were not anticipated by the in-person design. In the remote scenario, the intervention could be delivered synchronously, meaning via videoconference during the child’s hospitalization (CommunityRx-Hunger) or soon after an outpatient clinical encounter (CommunityRx-Dementia), or asynchronously using text and email, as detailed above. Using implementation science methods, adaptations to the intervention were documented[54] and the provisional essential elements of each intervention are being systematically tracked to allow for a robust intervention fidelity assessment.[55] While not originally designed as pragmatic trials, iteration of the trials to adapt to external, real-world challenges is a hallmark of pragmatic trial design.[56] Our approach could be emulated by other trialists.
The CommunityRx trials introduce innovation to interventions informed by the Self- and Family Management Framework by expanding the ways and among whom the *Framework is* being used.[11] RCTs using the Framework that have included an e-health component and have focused on dementia or health disparities are limited, and few quantitative studies that have used the Framework have enrolled family caregivers.[21] The CommunityRx trials also highlight the relevance of the Self- and Family Management Framework in times of crisis. During the COVID-19 pandemic, the importance of the patient-family relationship came to the fore. Family members were largely unable to be present at the bedside where they would normally support patients, provide information to clinicians, and generally co-manage illness. The CommunityRx trials demonstrate a core concept of the Self- and Family Management Framework, which is the need to support caregiver needs related to family management (including self-care) so that they can sustainably support patient self-management.[57]
## Conclusions
In-person clinical trials enrolling caregivers of patients with severe or chronic illness face particular challenges, made only more apparent by the COVID-19 pandemic. Rapid iteration to a remote design of two social care RCTs was facilitated by longstanding community engagement and innovation to optimize trial efficiency using widely accessible institutional and commercial information technology tools. Beyond the pandemic, fully remote or hybrid RCT protocols for social care interventions may be a viable, scalable alternative to bedside protocols, including in studies of caregivers. These innovative design elements have implications for wider applicability and scalability for multi-site or adaptive trials, or trials enrolling people with limited mobility or living in rural or other remote areas.
## Competing interests:
Under the terms of Grant Number 1C1CMS330997-01-00 (ST Lindau, PI) from the Department of Health and Human Services, Centers for Medicare & Medicaid Services, we were expected to develop a sustainable business model which will continue and support the model that we tested after award funding ends. Dr. Stacy Lindau was the founder and owner of a social impact company, NowPow, LLC, which was acquired by Unite Us, LLC in 2021. Dr. Lindau is a paid advisor to and holds stock in Unite Us, LLC. Dr. Lindau is an editor on Female Sexual Dysfunction for UpToDate and received royalties <$100/year in 2019, 2020 for this work. Subsequent royalties have been paid to the University of Chicago. Neither the University of Chicago nor UChicago *Medicine is* endorsing or promoting Unite Us or its business, products, or services. All other authors have no competing interests to disclose.
## Availability of data and materials:
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request
## References
1. 1.AARR, National Alliance for Caregiving. Caregiving in the United States 2020 [Internet]. Washington, D.C.: AARP; 2020 May [cited 2021 Sep 7]. Available from: 10.26419/ppi.00103.001&data=02|01|[email protected]|5f009842880a4001f34f08d7ec455df7|a395e38b4b754e4493499a37de460a33|0|0|637237654979717425&sdata=0lpX5wicapPe54racciUyV67robinlMxSTEfqW90XkA=&reserved=0. *Caregiving in the United States 2020 [Internet]*. DOI: 10.26419/ppi.00103.001&data=02|01|[email protected]|5f009842880a4001f34f08d7ec455df7|a395e38b4b754e4493499a37de460a33|0|0|637237654979717425&sdata=0lpX5wicapPe54racciUyV67robinlMxSTEfqW90XkA=&reserved=0
2. Gottlieb LM, Hessler D, Long D, Laves E, Burns AR, Amaya A. **Effects of social needs screening and in-person service navigation on child health: A randomized clinical trial**. *JAMA Pediatr* (2016.0) **170** e162521. PMID: 27599265
3. De Marchis EH, Hessler D, Fichtenberg C, Fleegler EW, Huebschmann AG, Clark CR. **Assessment of social risk factors and interest in receiving health care-based social assistance among adult patients and adult caregivers of pediatric patients**. *JAMA Netw Open* (2020.0) **3** e2021201. PMID: 33064137
4. Heller C, Balls-Berry JE, Nery JD, Erwin PJ, Littleton D, Kim M. **Strategies addressing barriers to clinical trial enrollment of underrepresented populations: a systematic review**. *Contemp Clin Trials* (2014.0) **39** 169-82. PMID: 25131812
5. Indorewalla KK, O’Connor MK, Budson AE, Guess DiTerlizzi C, Jackson J. **Modifiable barriers for recruitment and retention of older adults participants from underrepresented minorities in Alzheimer’s Disease research**. *J Alzheimers Dis JAD* (2021.0) **80** 927-40. PMID: 33612540
6. Elliott LK, Bami H, Gelkopf MJ, Yee RC, Feldman BM, Goh YI. **Patient and caregiver engagement in research: factors that influence co-enrollment in research**. *Pediatr Rheumatol Online J* (2019.0) **17** 85. PMID: 31864404
7. 7.US Department of Health and Human Services. National Plan to Address Alzheimer’s Disease: 2021 Update [Internet]. 2021 [cited 2022 Aug 19]. Available from: https://aspe.hhs.gov/sites/default/files/documents/66904c18bb1f0843c3c113d7099e98c1/napa-national-plan-2021-update.pdf. *National Plan to Address Alzheimer’s Disease: 2021 Update [Internet]* (2021.0)
8. Foster JR, AlOthmani FI, Seabrook JA, AlOfisan T, AlGarni YM, Sarpal A. **Parental presence at the bedside of critically ill children in a unit with unrestricted visitation**. *Pediatr Crit Care Med J Soc Crit Care Med World Fed Pediatr Intensive Crit Care Soc* (2018.0) **19** e387-93
9. Boyd K, Winslow V, Borson S, Lindau ST, Makelarski JM. **Caregiving in a pandemic: health-related socioeconomic vulnerabilities among female caregivers early in the COVID-19 pandemic**. *Ann Fam Med* (2022.0) **20** 406-413. PMID: 36228073
10. Van Driest SL, Madell SM, Crum K, Smith AH, Bichell DR, Kannankeril PJ. **Research consent rates before and during a COVID-19 one-visitor policy in a children’s hospital**. *Pediatr Res* (2021.0) **89** 1386-8. PMID: 33446915
11. Grey M, Schulman-Green D, Knafl K, Reynolds NR. **A revised self- and family management framework**. *Nurs Outlook* (2015.0) **63** 162-70. PMID: 25771190
12. Lindau ST, Makelarski JA, Abramsohn EM, Beiser DG, Boyd K, Chou C. **CommunityRx: A real-world controlled clinical trial of a scalable, low-intensity community resource referral intervention**. *Am J Public Health* (2019.0) **109** 600-606. PMID: 30789775
13. Lindau ST, Makelarski J, Abramsohn E, Beiser DG, Escamilla V, Jerome J. **CommunityRx: A population health improvement innovation that connects clinics to communities**. *Health Aff Proj Hope* (2016.0) **35** 2020-9
14. Lindau ST, Makelarski JA, Abramsohn EM, Beiser DG, Boyd K, Huang ES. **Sharing information about health-related resources: Observations from a community resource referral intervention trial in a predominantly African American/Black community**. *J Assoc Inf Sci Technol* (2021.0) 1-11
15. Lindau ST, Makelarski JA, Chin MH, Desautels S, Johnson D, Johnson WE. **Building community-engaged health research and discovery infrastructure on the South Side of Chicago: science in service to community priorities**. *Prev Med* (2011.0) **52** 200-7. PMID: 21236295
16. Lindau ST, Makelarski J, Abramsohn E, Beiser DG, Boyd K, Chou C. **CommunityRx: A real-world controlled clinical trial of a scalable, low-intensity community resource referral intervention**. *Am J Public Health* (2019.0) **109** 600-6. PMID: 30789775
17. Lindau ST. **CommunityRx for Hunger [NIH RePORTER] [Internet]**. *NIH RePORTER* (2019.0)
18. Lindau ST, Huang ES. *CommunityRx-Dementia [NIH RePORTER] [Internet]* (2019.0)
19. Sege R, Preer G, Morton SJ, Cabral H, Morakinyo O, Lee V. **Medical-legal strategies to improve infant health care: A randomized trial**. *Pediatrics* (2015.0) **136** 97-106. PMID: 26034248
20. Samus QM, Black BS, Reuland M, Leoutsakos JMS, Pizzi L, Frick KD. **MIND at Home-Streamlined: Study protocol for a randomized trial of home-based care coordination for persons with dementia and their caregivers**. *Contemp Clin Trials* (2018.0) **71** 103-12. PMID: 29783091
21. Schulman-Green D, Feder SL, Montano AR, Batten J, Tan H, Hoang K. **Use of the self- and family management framework and implications for further development**. *Nurs Outlook* (2021.0) **S0029-6554** 00136-6
22. Hatzenbuehler ML, Phelan JC, Link BG. **Stigma as a fundamental cause of population health inequalities**. *Am J Public Health* (2013.0) **103** 813-21. PMID: 23488505
23. Link BG, Phelan J. **Social conditions as fundamental causes of disease**. *J Health Soc Behav* (1995.0) 80-94. PMID: 7560851
24. Phelan JC, Link BG, Diez-Roux A, Kawachi I, Levin B. **“Fundamental causes” of social inequalities in mortality: a test of the theory**. *J Health Soc Behav* (2004.0) **45** 265-85. PMID: 15595507
25. Schulman-Green D, Jaser S, Martin F, Alonzo A, Grey M, McCorkle R. **Processes of self-management in chronic illness**. *J Nurs Scholarsh Off Publ Sigma Theta Tau Int Honor Soc Nurs* (2012.0) **44** 136-44
26. Samson A, Siam H. **Adapting to major chronic illness: a proposal fora comprehensive task-model approach**. *Patient Educ Couns* (2008.0) **70** 426-9. PMID: 18096353
27. Bandura A.. **Self-efficacy: toward a unifying theory of behavioral change**. *Psychol Rev* (1977.0) **84** 191. PMID: 847061
28. Herman DB, Mandiberg JM. **Critical time intervention: Model description and implications for the significance of timing in social work interventions**. *Res Soc Work Pract* (2010.0) **20** 502-8
29. Psihogios AM, Li Y, Butler E, Hamilton J, Daniel LC, Barakat LR. **Text message responsivity in a 2-way short message service pilot intervention with adolescent and young adult survivors of cancer**. *JMIR MHealth UHealth* (2019.0) **7** e12547. PMID: 30998225
30. 30.US Census Bureau. American Community Survey Data [Internet]. 2019 [cited 2021 Dec 20]. Available from: https://www.census.gov/data.html. *American Community Survey Data [Internet]* (2019.0)
31. Makelarski JA, Thorngren D, Lindau ST. **Feed First, Ask Questions Later: Alleviating and understanding caregiver food insecurity in an urban children’s hospital**. *Am J Public Health* (2015.0) **105** e98-104
32. Frazier CRM, Pinkerton EA, Grana M, Davis M, Asay S, Makelarski JA. **Feed1st, No Cuestions Asked: How a hospital-based food pantry program grew Its impact during the COVID-19 pandemic**. *Am J Public Health* (2022.0) e1-5
33. 33.SHARE Network [Internet]. SHARE Network. [cited 2022 Oct 6]. Available from: https://sharenetworkchicago.org/. *SHARE Network*
34. 34.Tech Savvy Friends - Blueprint [Internet]. [cited 2021 Dec 2]. Available from: https://uchicago.campuslabs.com/engage/organization/techsavvyfriends. *Tech Savvy Friends - Blueprint [Internet]*
35. Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, Conde JG. **Research electronic data capture (REDCap)--a metadata-driven methodology and workflow process for providing translational research informatics support**. *J Biomed Inform* (2009.0) **42** 377-81. PMID: 18929686
36. Harris PA, Taylor R, Minor BL, Elliott V, Fernandez M, O’Neal L. **The REDCap consortium: Building an international community of software platform partners**. *J Biomed Inform* (2019.0) **95** 103208. PMID: 31078660
37. 37.NowPow [Internet]. Available from: https://www.nowpow.com/. *NowPow [Internet]*
38. 38.Mosio [Internet]. Available from: https://www.mosio.com/. *Mosio [Internet]*
39. Doyle K, Boyd K, Furner B, Choi S, Acevedo J, Lindau ST. *Novel Integration of Clinical Research Data Collection with Intervention Softwares: Implementation in Two Community Resource Referral Trials* (2020.0)
40. Stadler W, Martell B.. *Clinical research activities and COVID-19 [Internet]* (2020.0)
41. 41.U.S. Food and Drug Administration. Part 11, Electronic Records; Electronic Signatures - Scope and Application [Internet]. U.S. Food and Drug Administration. FDA; 2020 [cited 2021 Dec 20]. Available from: https://www.fda.gov/regulatory-information/search-fda-guidance-documents/part-11-electronic-records-electronic-signatures-scope-and-application. *Part 11, Electronic Records; Electronic Signatures - Scope and Application [Internet]* (2020.0)
42. **Child Life Services**. *Pediatrics* (2014.0) **133** e1471. PMID: 24777212
43. Winkie A, Kinnebrew S. **Reintegrating essential child life programs during the COVID-19 pandemic to optimize development and normalization of hospital experiences for pediatric patients**. *Am J Infect Control* (2021.0) **49** S8
44. Weaver MS, Rosenberg AR, Fry A, Shostrom V, Wiener L. **Impact of the Coronavirus pandemic on pediatric palliative care team structures, services, and care eelivery**. *J Palliat Med* (2021.0) **24** 1213-20. PMID: 33350874
45. McDermott MM, Newman AB. **Remote research and clinical trial integrity during and after the Coronavirus pandemic**. *JAMA* (2021.0) **325** 1935-6. PMID: 33885728
46. Upadhaya S, Yu JX, Hodge J, Campbell J. **COVID-19 impact on oncology clinical trials: a 1-year analysis**. *Nat Rev Drug Discov* (2021.0) **20** 415-415. PMID: 34002059
47. Romito B, Jewell J, Jackson M. **Child Life Services**. *Pediatrics* (2021.0) **147** e2020040261. PMID: 33372119
48. 48.Association of Child Life Professionals. Standards of Clinical Practice [Internet]. 2020
Jan [cited 2021 Jan 6]. Available from: https://www.childlife.org/docs/default-source/aclp-official-documents/4-standards-of-clinical-practice.pdf?sfvrsn=893e8c4d_2. *Standards of Clinical Practice [Internet]* (2020.0)
49. 49.Illinois Commission to End Hunger. From Food Insecurity to Food Equity: A Roadmap to End Hunger in Illinois [Internet]. Illinois Department of Human Services; 2021
Mar [cited 2022 Sep 23]. Available from: https://www.dhs.state.il.us/page.aspx?item=135774. *From Food Insecurity to Food Equity: A Roadmap to End Hunger in Illinois [Internet]* (2021.0)
50. Lindau ST, Makelarski JA, Boyd K, Doyle KE, Haider S, Kumar S. **Change in health-related socioeconomic risk factors and mental health during the early phase of the COVID-19 pandemic: A national survey of U.S. women**. *J Womens Health 2002* (2021.0) **30** 502-513
51. Koonin LM. **Trends in the Use of Telehealth During the Emergence of the COVID-19 Pandemic — United States, January–March 2020**. *MMWR Morb Mortal Wkly Rep [Internet]* (2020.0) **69**
52. Sharma RK, Teng A, Asirot MG, Taylor JO, Borson S, Turner AM. **Challenges and opportunities in conducting research with older adults with dementia during COVID-19 and beyond**. *J Am Geriatr Soc* (2022.0) **70** 1306-13. PMID: 35285942
53. Huang ES, Lindau ST. **My Diabetes My Community [NIH RePORTER] [Internet]**. *NIH RePORTER* (2020.0)
54. Baumann AA, Cabassa LJ, Stirman SW, Brownson RC, Colditz GA, Proctor EK. *Dissemination and Implementation Research in Health: Translating Science to Practice [Internet]* (2017.0) 0. DOI: 10.1093/OSO/9780190683214.003.0017
55. Haynes A, Brennan S, Redman S, Williamson A, Gallego G, Butow P. **Figuring out fidelity: a worked example of the methods used to identify, critique and revise the essential elements of a contextualised intervention in health policy agencies**. *Implement Sci* (2016.0) **11**
56. Ford I, Norrie J. **Pragmatic trials**. *N Engl J Med* (2016.0) **375** 454-63. PMID: 27518663
57. Schulman-Green D, Feder SL, Dionne-Odom JN, Batten J, En Long VJ, Harris Y. **Family caregiver support of patient self-management during chronic, life-limiting illness: A qualitative metasynthesis**. *J Fam Nurs* (2021.0) **27** 55-72. PMID: 33334232
|
---
title: FRET assay for live-cell high-throughput screening of the cardiac SERCA pump
yields multiple classes of small-molecule allosteric modulators
authors:
- Osha Roopnarine
- Samantha L. Yuen
- Andrew R. Thompson
- Lauren N. Roelike
- Robyn T. Rebbeck
- Phillip A. Bidwell
- Courtney C. Aldrich
- Razvan L. Cornea
- David D. Thomas
journal: Research Square
year: 2023
pmcid: PMC10002828
doi: 10.21203/rs.3.rs-2596384/v1
license: CC BY 4.0
---
# FRET assay for live-cell high-throughput screening of the cardiac SERCA pump yields multiple classes of small-molecule allosteric modulators
## Abstract
We have used FRET-based biosensors in live cells, in a robust high-throughput screening (HTS) platform, to identify small-molecules that alter the structure and activity of the cardiac sarco/endoplasmic reticulum calcium ATPase (SERCA2a). Our primary aim is to discover drug-like small-molecule activators that improve SERCA’s function for the treatment of heart failure. We have previously demonstrated the use of an intramolecular FRET biosensor, based on human SERCA2a, by screening a small validation library using novel microplate readers that can detect the fluorescence lifetime or emission spectrum with high speed, precision, and resolution. Here we report results from a 50,000-compound screen using the same biosensor, with hit compounds functionally evaluated using Ca2+-ATPase and Ca2+-transport assays. We focused on 18 hit compounds, from which we identified eight structurally unique compounds and four compound classes as SERCA modulators, approximately half of which are activators and half are inhibitors. While both activators and inhibitors have therapeutic potential, the activators establish the basis for future testing in heart disease models and lead development, toward pharmaceutical therapy for heart failure.
## Introduction
Sarco/endoplasmic reticulum calcium ATPase (SERCA), integrated in the sarcoplasmic reticulum (SR, muscle cells) or endoplasmic reticulum (ER, non-muscle cells) membrane in most mammalian cells, is integral for using Ca2+-dependent hydrolysis of ATP to fuel active transport of cytosolic Ca2+ into the SR or ER. In muscle, the activity of SERCA1a (skeletal isoform) or SERCA2a (cardiac isoform) is essential for relaxation (diastole), restoring SR Ca2+ following its release via Ca2+ channels (ryanodine receptors, RyR) for muscle contraction (systole). Decreased SERCA activity and excessive RyR leak results in failure to maintain the high gradient of [Ca2+] between the cytoplasm (mM) and the SR (sub-μM) during diastole (muscle relaxation) and are associated with heart failure (HF) in human and animal models1. Decreased SERCA activity has been attributed to multiple factors, including reduced SERCA gene expression, increased post-translational modifications, and altered interaction with regulatory proteins1. Overall, decreased SERCA activity and increased Ca2+-leak lead to a pathophysiological state of the cardiac myocyte2 (HF, cardiac hypertrophy, diabetic hypertrophy), skeletal myocyte (Brody’s disease and myotonic dystrophy) and non-muscle cells (Darier’s disease, diabetes, Alzheimer’s disease)3. Altered SERCA interaction with regulatory proteins (regulins), such as phospholamban (PLB), have been linked to ventricular dysfunction and maladaptive remodeling in failing hearts4. Of the seven regulins discovered5, the dwarf open reading frame (DWORF) regulatory peptide is the only one known to activate muscle-specific SERCA activity, both by direct activation of SERCA6,7 and by competing with PLB binding8,9, and to prevent HF in a mouse model of dilated cardiomyopathy10. Current therapeutic measures include beta-blockers, angiotensin-converting enzyme (ACE) inhibitors, angiotensin-receptor blockers (ARB), and mineralocorticoid receptor antagonists. However, these do not directly target proteins responsible for dysfunctional Ca2+ cycling. Discovery of small molecules (potential drugs) that target specific proteins is needed to exert improved control measures for positive therapeutic outcomes.
In the present study, we seek primarily SERCA2a activators to alleviate heart failure11,12. However, SERCA uncouplers are also of potential interest in other indications such as in nonshivering thermogenesis, enhancing metabolism and thus reducing obesity13,14. Targeted SERCA uncouplers or inhibitors may be useful for treatment of cancer or malaria15.
SERCA2a is a large transmembrane protein, with the phosphorylation (P) and nucleotide-binding (N) domains forming the catalytic site, coupled by the actuator (A) domain (Fig. 1A). Large (5–10 nm) relative movements of these domains are coupled to Ca2+ transport, as detected in living cells by an intramolecular FRET biosensor (Fig. 1A)16. The interaction of small molecules with SERCA can induce measurable structural changes, detectable by this biosensor, that often correlate with function, making this biosensor a powerful tool for HTS discovery of SERCA-binding compounds16,17.
In several previous early-stage drug discovery campaigns, we have focused on the SERCA regulator phospholamban (PLB)18,19, but we have also tested and validated FRET biosensor constructs of SERCA alone16,17,20, to detect binding of drug-like small-molecule compounds directly to SERCA. For this, we engineered a “two-color” SERCA (2CS, Fig. 1A) construct with fused eGFP and tagRFP fluorescent protein tags to the cytoplasmic N- and A-domains of SERCA, which detect relative motions of these domains during the enzymatic cycle responsible for Ca2+-transport17,20,21. Using the intramolecular FRET measurement of 2CS, stably expressed in HEK293 cells, we previously validated this biosensor using the NIH Clinical Collection (NCC, 727 compounds)17 and the LOPAC library (1280 compounds)20. Along with the fluorescent biosensors, high-throughput is enabled by a FLT-PR (fluorescence lifetime plate reader) that scans a 1536-well plate with unprecedented precision and speed, determining FRET with $0.3\%$ CV (30 times better than conventional intensity detection) in 2.5 min18,22,23, enabling a high-precision screen of 50,000-compounds in two days. This instrument is equipped with simultaneous FLT detection at two emission wavelengths (two-channel lifetime detection)21, a function used to filter out fluorescent compounds that would be falsely identified as hit compounds. We have demonstrated the additional high throughput acquisition of fluorescence emission spectra using a spectral unmixing plate reader (SUPR), which helps detect (and rule out) interfering compounds, based on changes in the line shape in the donor-only region of the spectrum21. As in our previous projects involving other drug targets, the next logical step is to screen a 50,000-compound DIVERSet library, a diverse collection of drug-like small molecules that has yielded effective hit compounds in previous drug discovery projects22–24.
Here we apply our HTS platform using FRET lifetime measurements of two-color human cardiac SERCA (2CS) in living cells (Fig. 1B) to identify hit compounds. To validate selected hit compounds, we then acquire concentration response curves (CRCs) using both FRET and functional assays (Ca2+-ATPase activity and Ca2+-transport). We hypothesize that the combination of using improved fluorescence technology and screening a larger compound library (50K DIVERSet) will result in a larger and more diverse collection of hit compounds that more effectively regulate cardiac SERCA function, thus increasing the potential for discovering lead compounds for new heart failure therapeutics.
## FLT HTS of 50K DIVERSet library
2CS expressed in live HEK-293 cells was incubated with 5 nL of compound at a final [compound] of 10μM (from a DIVERSet library of 50,000 compounds) or DMSO (as a control), preloaded on 40 assay plates for 20 min prior to being read on the FLT-PR. FLT measurements were observed to be normally distributed, with a coefficient of variation (CV) of $0.4\%$ across all 40 plates (Fig. 2A). Plate-by-plate CV varied by < 1 % (Fig. 2A).
Compounds that significantly altered the structure of 2CS were determined by computing the change in lifetime (Δτ) for each compound compared to DMSO control sample (2CS plus DMSO), and the magnitude of this change was compared to the normal statistical fluctuation of the biosensor by computing the robust z-score (Methods). The distinct FLT changes induced by the potential hit compounds (Fig. 2B, red) are illustrated by the normal distributions of compounds not affecting SERCA (Fig. 2C, blue) and the control (Fig. 2C, dark blue). A hit threshold was set at a robust z-score of ± 3, resulting in 2960 FLT hit compounds (Fig. 1B, step 1). Interference from fluorescent compounds was removed (Fig. 1B, step 2) using both the similarity index (SI, detected by SUPR)17 and two-channel lifetime detection21. We also eliminated compounds that affected the lifetime detected from a donor-only (1CS) sample; 295 compounds remained.
More FLT decreasers than increasers were found to fail these tests, in alignment with previous studies18,20,21,25. FLT increasers are more advantageous for two additional reasons: (a) They offer greater reproducibility between repeats of a screen18,20,21. ( b) Most previously identified SERCA modulators have been shown to be FLT increasers16,17,20. Therefore, we prioritized the 158 FLT increasers (termed “hit compounds” (Fig. 1B, step 3) for follow-up retesting and CRC evaluation.
## FLT retests of select hits compounds with 2CS and null biosensor
158 hit compounds were retested using 2CS (Step 4 in Fig. 1B; see Fig. 3A and C) and a null biosensor construct (Step 4 in Fig. 1B; see Fig. 3B and D), which consisted of GFP and RFP connected by a 32-residue unstructured flexible linker peptide (G32R)20. The null biosensor was used to rule out compounds that alter FLT by directly binding to the fluorescent proteins. A plot of the change in lifetime (Δτ) vs. the ΔR/G ratio (Fig. 3C and D) shows that the 2CS hits had little to no effects on the null biosensor.
Hit compounds that produced > 75 ps Δτ (76 compounds, Fig. 1B, step 4) (Fig. 3A) were targeted for further functional testing. We focused on 18 of these for the CRC testing, after the FLT data were subjected to the first four steps of the screening funnel (Fig. 1B) and compound repurchasing availability was determined. None of these compounds were in the PAINS (Pan-Assay INterference compoundS) category26, nor were they redox agents or metal chelators.
## Validation of hit compounds using FLT CRC
To further evaluate the 18 hit compounds, we determined the FLT response to compound concentrations ranging 0.78–100.μM (Step 5 in Fig. 1B). All 18 hit compounds (Table 1) decreased FRET (increased FLT) of the 2CS biosensor, suggesting that the compounds induced a structural change (Fig. 1, top) in the cytosolic headpiece region of SERCA2a. Compounds 1 and 4 showed a significant decrease in FRET at the lowest concentration, but no further effect at higher concentrations. The remaining 16 compounds decreased FRET with detectable EC50 (Fig. 4–7B, Table 1).
## Effects of hit compounds on SERCA2a activity using Ca2+-ATPase and Ca2+-transport assays
To assess the impacts of FRET hit compounds on SERCA2a function, we used an absorbance-based enzyme-coupled NADH-linked Ca2+-dependent ATPase assay (referred to below as ATPase) and a fluorescence-based Ca2+-transport assay (referred to below as transport), using pig cardiac SR vesicles enriched for SERCA2a18 (Step 5 in Fig. 1B). Ca2+-ATPase and Ca2+-transport activities were measured at VMAX (saturating, pCa 5.4) and VMID (subsaturating, midpoint, pCa 6.2) [Ca2+], revealing activators (Compounds 1–9) and inhibitors (Compounds 10–18) (Table 1 and Supplementary Fig. S1), identified on the basis of potency (1/EC50) and efficacy (amplitude of the effect, Δ) (Table 1). Activators were compounds that increased Ca2+-ATPase and/or Ca2+-transport activities (at one or both [Ca2+]) (Table 1, Fig. 4–6). We grouped the activators in two subcategories based on the functional effects at the two Ca2+ concentrations: [1] activates both Ca2+-ATPase and Ca2+-transport (Compounds 2, 4, 7, 8, and 9), and [2] activates Ca2+-ATPase with divergent effects on Ca2+-transport (Compounds 1, 3, 5, and 6). We define “divergent” to indicate that the compound induces opposing effects at two different [Ca] (an increase at one [Ca] and a decrease at the other) in one assay. These effects indicate induced changes in the coupling ratio (CR), which is optimally two Ca2+ ions transported per molecule of ATP hydrolyzed27–30. This change was determined from the ratio of VMAX (Ca2+-transport) to VMAX (Ca2+-ATPase) (Table 1), as discussed below in SERCA Activators.
We identified inhibitors that induced strong (≥ $68\%$), moderate (34 to $67\%$), and mild (≤ $33\%$) inhibition of SERCA function. Therefore, we define four subcategories of compound inhibition: 1) strong inhibition of Ca2+-ATPase and Ca2+-transport (Compounds 11, 12, 13), 2) moderate inhibition of Ca2+-ATPase and strong inhibition of Ca2+-transport (Compounds 15 and 17), 3) mild inhibition of Ca2+-ATPase and moderate-to-strong inhibition of Ca2+-transport (Compounds 14 and 16), and 4) mild inhibition on Ca2+-ATPase and mild-to-moderate inhibition on Ca2+-transport (Compounds 10 and 18) (Table 1, Fig. 7, and discussed below under SERCA Inhibitors).
## Classification of compounds by physicochemical characteristics
The 18 hit compounds were subjected to cheminformatic analysis, to determine whether any of the compounds shared a common chemical scaffold. Compounds with a Tanimoto coefficient and maximum common substructure (MCS)31 scores above 0.4 were binned as clusters, while those with scores below 0.4 were classified as singletons. The analysis yielded diverse scaffolds31,32 of the hit compounds (Supplementary Fig. S1 and Table S1).
Four clusters of compounds with multiple examples (A-D in Table 1) were found, and the remaining eight were unique compounds (singletons) (E-L in Table 1 and Supplementary Fig. S1 and Table S1). The three compounds in cluster A (Compounds 1, 2, and 3 in Table 1) have a common 5-(aryloxymethyl)oxazole-3-carboxamide)33, while those in cluster B (Compounds 4 and 5) share a N-heteroaryl-N-alkylpiperazine. Cluster C (Compounds 9 and 10) is defined by an amide linkage and Cluster D (Compounds 11, 12, and 13) by a piperidine scaffold. Clusters E-L (Compounds 6, 7, 8, 14, 15, 16, 17, and 18) contain a single compound (singleton) with no common scaffold with any other hit compound in this study. All of the hit compounds have physicochemical properties34, conducive of favorable drug disposition in vivo, including a low molecular weight (< 500), low cLogP (calculated partition coefficient for lipophilicity) values (< 5), low non-H rotatable bonds that describe the molecular flexibility (< 10), low degree of possible hydrogen bond formation (total number of hydrogen bond acceptors and donors should be less than 8), and low total polar surface area (tPSA < 140 Å) (Supplementary Table S1).
Next we describe in more detail the nine activators (Fig. 4–6) that are grouped into two subcategories, and the nine inhibitors (Fig. 7) that are grouped into three classes and four subcategories.
## SERCA Activators
In the first category of activators, Compounds 2, 4, 7, 8, and 9 activated both Ca2+-ATPase and Ca2+-transport. Compound 7 (Fig. 4A) (singleton F) decreased FRET of 2CS in live cells so that EC50 = 0.3 μM, indicating stabilization of the open conformation of SERCA, and accelerated Ca2+-ATPase to induce ΔVmax = $14\%$ and ΔVMID = $7\%$ (Fig. 4C and Table 1). Compound 7 induced the highest increase in Ca2+-transport of all the compounds at both VMAX ($24\%$) and VMID ($19\%$), (Fig. 4D), which was greater than that of Ca2+-ATPase activity (Fig. 4D). The CR increased to 0.74 compared to control (0.66, Table 1). Saturation of CRC was not reached at the highest [compound] measured, so the functional EC50 was not determined, therefore we determined C10, the compound concentration that increases function by $10\%$. At VMAX and VMID, C10 was 25 μM for Ca2+-ATPase, 14 μM and 21 μM for Ca2+-transport (Table 1). This compound will be placed at high priority for future optimization by medicinal chemistry and testing in animal models.
Compound 8 (singleton G) (Fig. 5A) decreased FRET (EC50 = 4.9 μM, Table 1) and increased Ca2+-ATPase at both VMAX ($49\%$, the largest increase observed in the screen) and VMID (31 %) (Table 1 and Fig. 5C). For Ca2+-transport, effects (Δ values) were lower ($10\%$ for VMAX and $7\%$ for VMID, Fig. 5D), decreasing CR to 0.4 (Table 1). EC50 values for Ca2+-ATPase were not significantly different at VMAX and VMID, and were ~ 2x greater than the values observed by FRET. C10 was 4.3 ^M (VMAX) and 7.8 ^M (VMID), indicating significant ATPase activation at low dosage. C10 for Ca-transport was 25 μM at VMAX, and was not determined at VMID.
Compound 2 (cluster A) induced similar Ca2+-ATPase activation ($16\%$ at VMAX, $7\%$ at VMID) as observed with Compound 7, but induced smaller increases on Ca2+-transport functions (Table 1), decreasing CR to 0.53. Compounds 4 (cluster B) and 9 (cluster C) also showed similar effects as Compound 7, increasing both activities. Compound 4 increased Ca2+-ATPase at both VMAX and VMID by ~ $10\%$ and increased Ca2+-transport at both VMAX ($0.5\%$) and VMID ($11\%$). Compound 9 increased both VMAX (21 %) and VMID ($10\%$) for Ca2+-ATPase, with smaller increases in Ca2+-transport (3–$5\%$). Compounds 4 and 9 decreased CR to similar extents (0.51) (Table 1).
In the second category of activators, Compounds 1 and 3 (cluster A), 5 (cluster B), and 6 (singleton E) activated Ca2+-ATPase at both VMAX and VMID, but induced divergent effects on Ca2+-transport. Compound 6 (Fig. 6A) decreased FRET with EC50 = 7.1 μM (Fig. 6B, Table 1), while moderately activating Ca2+-ATPase activity by $25\%$ at VMAX and by $30\%$ at VMID (Fig. 6C), with EC50 = 11 μM for both VMAX and VMID. It induced divergent effects during Ca2+-transport, inhibiting VMAX by $21\%$ and activating VMID by $17\%$ (Table 1 and Fig. 6D). C10 was similar at the two Ca2+ concentrations (9.7 μM and 8.7 μM), but was significantly different for Ca2+-transport (15 μM at VMAX, 4 μM at VMID). CR was decreased to 0.48 by Compound 6.
Compound 5 activated Ca2+-ATPase moderately at VMAX ($25\%$) and VMID ($35\%$). Ca2+-transport was inhibited slightly at VMAX ($2\%$), but activated at VMID ($20\%$) (Table 1). Compounds 1, and 3 induced low activating effects at VMID for Ca2+-transport ($2\%$ and $6\%$), but they inhibited Ca2+-transport at VMAX (−$10\%$ and − $20\%$) (Table 1). Compounds 1 and 5 induced similar decreases in the CR at VMAX (0.0.49 and 0.47), while Compound 3 induced a slightly smaller CR of 0.39 (Table 1). These effects are similar to that of unphosphorylated phospholamban (PLB) in cardiac SR35.
## SERCA Inhibitors
Compounds 10–18 all decreased Ca2+-ATPase and Ca2+-transport activities at both VMAX and VMID. Compared with FRET EC50 of the activators (0.3–7 μM), most of the inhibitors (Compounds 10–18) showed weaker affinity, with FRET EC50 values in the range of 5–32 μM, but the maximum functional effects (efficacies) of the inhibitors tended to be greater (Table 1).
Compound 12 (cluster D) strongly inhibited both the Ca2+-ATPase and Ca2+-transport activities (Fig. 7C and D) to levels similar to the well-known SERCA inhibitor, thapsigargin, although thapsigargin acts with much greater affinity (EC50 ≈ 7.5 nM18) than compound 12 (EC50 = 3.8 μM) (Table 1).
Compounds 11, 12, and 13 (cluster D) showed similar inhibition of both SERCA2a functions: Ca2+-ATPase was inhibited by 61 %, $93\%$, and 81 % at VMAX; by $59\%$, $90\%$, and $72\%$ at VMID. Ca2+-transport was completely inhibited. Compared with Ca2+-ATPase, Ca2+-transport inhibition at VMID required slightly higher compound concentration as shown by the shift to the right of the red curve (Fig. 7D).
Compounds 15 (singleton I) and 17 (singleton K) induced moderate inhibition of both activities, decreasing VMAX and VMID by ~ $50\%$ for Ca2+-ATPase and slightly more for Ca2+-transport (70–$85\%$). C10 values for Compound 15 were ~ 1 μM or less, slightly higher for Compound 17 (5–$6\%$) (Table 1) for Ca2+-ATPase, and similar for Ca2+-uptake at both VMAX and VMID ranging from (0.9–$1.7\%$).
Compounds 14 (singleton H) and 16 (singleton J) induced mild inhibition of Ca2+-ATPase, but a considerably larger effect of moderate-to-strong inhibition of the Ca2+-transport. Ca2+-ATPase decreased by $13\%$ and $8\%$ at VMAX, $26\%$ and $16\%$ at VMID. Ca2+-transport was inhibited by $83\%$ and $66\%$ at VMAX, $62\%$ and $48\%$ at VMID. C10 values ranged from 0.5 to 7μM, for Ca2+-ATPase at VMAX and VMID.
Compounds 10 (cluster C) and 18 (singleton L) induced mild inhibition of Ca2+-ATPase and moderate inhibition of Ca2+-transport. At VMAX, Ca2+-ATPase was inhibited by $31\%$ and $24\%$, respectively; while at VMID it was inhibited by 24 and $16\%$, respectively. At VMAX Ca-transport was inhibited by $57\%$ and 41 %, and at VMID by $34\%$ and $16\%$. C10 values were 1 μM and 7 μM for Ca2+-ATPase, and 4.1 μM and 22.7 μM for Ca2+-transport.
## Discussion
We identified new compounds based on an increase in donor FLT, within a human cardiac 2CS biosensor expressed in live mammalian cells, indicating a decrease in FRET, implying that the actuator (A) and nucleotide-binding domains (N) of SERCAC2a moved further apart. We confirmed that these compounds affect SERCA activity using Ca2+-ATPase and Ca2+-transport assays with SERCA2a in native SR preparations, where we further categorized them as activators or inhibitors. We identified two subcategories of activators, whereby the compound either [1] activates both Ca2+-ATPase and Ca2+-transport activities (Compounds 2, 4, 7, 8, and 9) (Figs. 4 and 5) and [2] activates Ca2+-ATPase with divergent effects on Ca-transport (Compounds 1, 3, 5, and 6) (Fig. 6). We identified four subcategories of inhibitors based on the extent of Ca2+ATPase and Ca2+-transport decrease for cardiac SERCA2a (Table 1 and Fig. 7).
*In* general, the FRET EC50 values were smaller (indicating higher potencies) for activators (Compounds 1–9; 0.3–7μM) than for inhibitors (Compounds 10–18; (3–32μM) (Table 1). The potencies observed by FRET and function are not precisely correlated, probably because the assays were performed on different types of samples (live cells vs purified proteins), measuring different properties (structure vs function).
Functional CRC assays showed that inhibitors tended to induce larger changes (indicating higher efficacies) than activators, in both Ca2+-ATPase activity and Ca2+-uptake. Also, most inhibitors induce a larger change in Ca2+-uptake than in Ca2+-ATPase, decreasing the coupling ratio. *In* general, the C10 and EC50 values were smaller, indicating greater potency, for inhibitors than for activators.
Effects of most activators were to reduce the CR, as they induced larger changes in the Ca2+-ATPase than in corresponding the Ca2+-uptake assay. The most notable exception is Compound 7, which activates Ca2+-transport even more than it activates Ca2+-ATPase, increasing CR. This compound will be a high priority as a lead compound for future efforts in medicinal chemistry and assays of physiological function. Ten compounds were binned into four clusters (A-D), while eight compounds were classified as singleton (E-L) (Table 1). Many compounds showed similar functional traits, suggesting that there are ligand-sensing sites in SERCA2a that recognize a range of scaffolds, or that the ligand-binding sites are close to each other, providing potentially powerful tools in the design of future compounds36,37,38. Compounds 2 (cluster A), 4 (cluster B), 7 (singleton F), 8 (singleton G), and 9 (cluster C) induced similar effects of moderate activation on VMAX in the Ca2+-ATPase assay with smaller activating effects on Ca2+-transport. Compounds in activator clusters A (Compounds 1 and 3) and B (Compound 5) along with Compound 6 from singleton E, showed similar functional effects: moderate activation of VMAX, with smaller activation of Ca2+-transport (decreased coupling) at both [Ca2+]. The compounds in clusters D (11, 12, and 13), C [10], and H-L (14–18) of inhibitors also induced similar functional effects.
There was little or no overlap in the hit compounds identified in our previous FRET HTS screen of the same (DIVERSet) library with another biosensor for tumor necrosis factor receptor 1 (TNFR122. There was $81\%$ overlap in the fluorescent compounds detected in these two HTS screens, indicating that our FRET HTS screening methodologies independently and successfully removed the fluorescently interfering compounds21,24. In another HTS study of the DIVERSet library, using a SERCA functional (Ca2+-ATPase) assay in the primary screen, we discovered several activators, several of which showed isoform specificity for either SERCA1a or SERCA2a28. However, there was negligible overlap between hit compounds identified in that ATPase HTS study and in the current study that used FRET in the primary screen. This observation highlights the value of complementary HTS assays for the same target.
SERCA activators are needed when cardiac relaxation is impaired, as in early-onset diastolic dysfunction that precedes systolic impairment in HF1, diabetic cardiomyopathy39, Alzheimer’s disease40, or Duchenne muscular dystrophy (DMD)41. The stimulation of SERCA2a activity, as a novel therapeutic measure to relieve cardiac dysfunction in heart failure without arrhythmogenic effects, is a promising strategy to be used in combination with other first-line therapeutic agents such as β-blockers and ACE inhibitors42. Until recently, only a few compounds were known to stimulate SERCA: CDN1163 (stimulates Ca2+ transport)43,19, CP-154526 (increases the apparent Ca2+ affinity of SERC2a)44, Ro 41–0960 (increases SERCA maximal activity in high Ca2+)44, and istaroxime (stimulates SERCA activity)45. However, our recent HTS using Ca2+-ATPase activity as the target HTS assay identified ~ 19 new activators of SERCA28, and we identified nine activator compounds in the present study. A SERCA activator from our previous work shows promise as a therapeutic target for Alzheimer’s disease, as it rescued memory function in a mouse model for the disease40 as well as for DMD as it was shown to ameliorate dystrophic phenotypes in dystrophin-deficient mdx mice41. Of all these SERCA2a activators only istaroxime, a known Na+/K+ transporting ATPase inhibitor as well as an inotropic/lusitropic agent acting to enhance SERCA2a activity, has been in phase IIb clinical trials for treatment of heart failure45,46. However, because of its unsuitability for human usage (poor gastrointestinal absorption, high clearance rate, and extensive metabolic transformation)46, istaroxime derivatives were designed from QSAR studies and a new promising class of SERCA2a activators has been identified47,48,49.
Compounds 1, 3, 5, and 6 induced small effects on the VMAX of Ca2+-ATPase (~ 10–$25\%$ increase) and induced a negative effect on the VMAX of Ca2+-transport (Fig. 6C and D), thus decreasing the CR, which is likely to increase heat output13,52,54. These effects are similar to that of SLN on SERCA1a (skeletal muscle), where SLN induces no observable effect on the VMAX of Ca-ATPase but reduces the VMAX of Ca2+-transport (SERCA1a uncoupling), thus reducing CR50. This results in futile cycling of SERCA1a and higher usage of ATP resulting in increased non-shivering thermogenesis (NST)50. Another contributor to NST is Ca2+ leak from SR to sarcoplasm through RyR channels, stimulating SERCA to re-sequester Ca2+ into the SR, thus using more ATP and generating heat51. This increase in energy expenditure in muscle has been suggested as a potential therapeutic strategy for weight loss50,52. Thus, the SERCA uncouplers in this study may serve as the basis for further drug development targeting weight loss.
Over the past several decades (~ 60 years), research on the potential for small-molecule SERCA inhibitors as oncology therapeutics has yielded hundreds of SERCA inhibitors with varying potencies and efficacies15. Similarly, our discovery of new SERCA inhibitors with a range of potencies and efficacies is likely to be advantageous for treatment of non-cardiac applications15,53.
In the present study, the 2CS biosensor has been used to identify novel small-molecule effectors of SERCA that have diverse chemical scaffolds for binding to SERCA, resulting in an array of hit compounds that are activators and inhibitors. Most importantly, we discovered a potential lead compound (Compound 7) that activates Ca2+-uptake more than the Ca2+-ATPase, increasing the CR, so this will be a high riority for future efforts in medicinal chemistry and assays of physiological function. The enabling technology included three novel plate-readers, the FLT instrument used in the primary screen, and two spectral instruments that were used to remove interference of fluorescent compounds, allowing us to focus on valid SERCA activators and inhibitors. In the future, hits from the present study will be evaluated in more functional detail, including studies on multiple SERCA isoforms and on intact muscles and animals. Medicinal chemistry will be applied to elucidate structure-activity relationships, with the goal of designing analogs with greater potency and specificity22,54. We will also expand our approach to much larger compound libraries, since our primary screening technology is capable of evaluating thousands of compounds per hour.
## Molecular biology
A two-color intramolecular human SERCA2a (2CS) biosensor, based on human cardiac SERCA2a fused to green fluorescent protein (eGFP) and red fluorescent protein (tagRFP) was developed to detect structural changes that are related to the functional changes of SERCA20. Briefly, tagRFP was genetically fused to the N-terminus of SERCA and eGFP was inserted as an intrasequence tag before residue 509 in the nucleotide-binding domain (N-domain)55,56. A donor-only (1CS) biosensor was created in a similar manner as the 2CS biosensor with the exception of the construct containing only eGFP The fluorescent proteins fused to SERCA in 2CS and 1CS do not significantly affect SERCA activity, in membranes purified from HEK cells18,20. A null biosensor construct consisting of eGFP and tagRFP connected by a 32-residue unstructured flexible linker peptide (G32R) was created as described previously18,20. All constructs were cloned into expression vectors containing the genes for antibiotic resistance to G418, puromycin, or blasticidin.
## Cell culture
Stable cell lines were generated using either HEK293 (ATCC, Manassas, VA) or HEK293–6E (National Research Council, Canada) cells20. Briefly, cells were transiently transfected with 2CS, 1CS, or G32R null biosensor plasmids using Lipofecatime 3000 or 293fectin (Thermo Fisher Scientific). Flow cytometry was used to select and enrich for the population of cells expressing respective biosensors. Stable HEK293 cell lines were maintained in phenol red-free DMEM media (Gibco, Waltham, MA) supplemented with 2 mM GlutaMAX (Gibco, Waltham, MA), $10\%$ fetal bovine serum (FBS) (Atlanta Biologicals, Lawrenceville, GA), 1 IU/mL penicillin/streptomycin (Gibco, Waltham, MA), and 250μg/mL G418 (Fisher Scientific). Stable HEK293–6E cell lines were maintained in F17 media (Sigma Aldrich) supplemented with Kolliphor p188 (Sigma Aldrich, St. Louis, MO), 200 nM/mL GlutaMAX, and either 1 μg/mL puromycin (Invitrogen, Carlsbad, CA) or 2μg/mL blasticidin (Goldbio) as a selection antibiotic. All cell lines were grown at 37°C with $5\%$ CO2.
## Compound handling
A DIVERSet 50,000 compound library was purchased from ChemBridge Corporation (San Diego, CA) at a 10 mM stock concentration for each compound. All compounds met the high quality standard of $100\%$ identification by NMR and/or LC-MS and have a minimum purity of $85\%$ and their identity verified using LC-MS/ELSD as confirmed by the ChemBridge Corporation. For the FRET HTS initial screens, the compound library was reformatted into 384 well Echo compatible plates using the Biomek FX (Beckman Coulter, Miami, FL) and 5 nL of either compound (columns 3–22 and 27–46) or DMSO (columns 1–2, 23–26, and 47–48) was dispensed into 1536 well black polystyrene assay plates (Greiner, Kremsmunste, Austria) using an Echo 550 liquid dispenser (Beckman Coulter) to yield a final assay screening concentration of 10μM. The low autofluorescence and low interwell cross-talk of these plates made them advantageous for FRET measurements. Plates were heat sealed with a PlateLoc Thermal Microplate Sealer (Agilent, Santa Clara, CA) and stored at −20°C prior to use. The same methods were applied for subsequent FRET retesting of the reproducible hit compounds identified in the pilot screen, except that the [compound] was tested at 10μM and 30 μM in triplicate.
FRET CRC assay plates (0.78–100 μM compound range) with at least ten different compound concentrations were made by adding the appropriate volume of compound or DMSO into black 384 well plates (Greiner Bio-One) using a Mosquito HV (SPTLabTech, United Kingdom). Subsequent ATPase and Ca2+-transport CRC assay plates (0–50 μM compound range) with repurchased compounds were made in a similar manner using with the Echo 550 (Beckman Coulter) using either 384 well transparent plates (Greiner Bio-One) or black-walled plates with transparent bottoms (Greiner Bio-One), respectively.
## HTS sample preparation and FRET measurements
On each day of screening, cells were harvested, washed three times with PBS, and centrifuged at 300g for 5min. Cells were filtered using a 70μm cell strainer and diluted to 1–2 × 106 cells/mL. Cell concentration and viability were assessed using the Cell countess (Invitrogen) and trypan blue assay. During assays, cells were constantly and gently stirred using a magnetic stir bar at room temperature, keeping the cells in suspension and evenly distributed to avoid clumping. Cells were dispensed at 5μL or 50μL per well into assay plates (dispensed into 40 assay plates, each containing 1536 wells) pre-plated with either compound or DMSO using a Multdrop Combi liquid dispenser (Thermo Fisher Scientific, Pittsburg, PA) and sealed until needed. Because the kinetics of membrane permeability, diffusion, and/or binding of the compound to live cells may be compound-dependent, we tested two incubation times, 20 min and 120 min, for the FLT CRC. FRET EC50 values determined from both incubations were similar, but the 120 min incubation yielded a more reproducible and sigmoidal curve. Plates containing eight-point concentration curves of three tool compounds (known SERCA effectors) were also included on the plates as positive controls for the HTS FRET assay. The FRET HTS screen was performed over two days with a custom HTS fluorescence lifetime plate reader (FLT-PR) and spectral plate reader (SUPR) provided by Photonic Pharma LLC (Minneapolis, MN)18.
The same methods were applied for subsequent FRET retesting of the reproducible hit compounds identified in the pilot screen, except that the compound tested at 10μM and 30 μM [compound]. 158 hit compounds were picked from the library master plates and reloaded onto new assay plates for retesting with 2CS and a null biosensor. Then 18 hit compounds were selected and purchased from ChemBridge Corporation to determine CRC from FRET, ATPase, and Ca-transport assays using at least ten different concentrations by repeatedly scanning the 1536-well plates.
## FRET HTS instrumentation and data analysis
A detailed description of the high-throughput fluorescence lifetime plate reader (FLT-PR) and spectral unmixing plate reader (SUPR), manufactured by Fluorescence Innovations Inc and provided by Photonic Pharma, LLC was described previously18,21. Briefly, for lifetime measurement with the FLT-PR, the observed donor-fluorescence waveform, F (t) was fit by a convolution of the measured instrument response function (IRF) and a single-exponential decay to obtain the lifetime (τ) of the donor fluorophore57 in the absence (τD) and presence (τDA) of the acceptor as described in Eq. [ 1]: 1 F(t)=Ae(−tτ) FRET efficiency (E) was determined as the fractional decrease of donor FLT in the absence and in the presence of acceptor as in Eq. [ 2]: 2 $E = 1$−τDAτD E was determined in the presence and absence of compound and normalized relative to E of the DMSO control. For spectral detection with the SUPR, the observed fluorescence emission spectrum F(λ) was fit by least-squares minimization of a linear combination of component spectra for donor (D), acceptor (A), cellular autofluorescence (C) and water Raman (W) as described previously17.
## HTS data analysis
FLT-PR data was used as the primary metric for flagging potential hit compounds. After fitting waveforms with a single exponential decay to quantify donor lifetime, the change in fluorescence lifetime (Δ FLT) was computed by performing a moving average subtraction in the order the plate was scanned with a window size of half a plate row (24 columns). The reasons for this are twofold: 1) plate gradients are often observed due to heating of the digitizer during acquisition and 2) performing ΔFLT computations with DMSO controls alone can sometimes result in artifacts as a half of the DMSO wells are on the edge of plates, which occasionally exhibit artifacts due to processes needed for the preparation of the drug library being tested. As most compounds are likely to be non-hits, and therefore DMSO like, computation of a moving average is an effective alternative to solving both gradient issues and edge-effect distortion of the primary metric for hit selection, Δτ. As hit compounds from FLT-PR were to be further triaged with a secondary technique (using the spectral plate reader), a generous cutoff was set at a robust z-score of 3 on a plate-by-plate basis. The robust z-score was used, where the median (M) and median absolute deviation (MAD) are used in place of the mean and standard deviation (Eq. 3), to best capture the most hits, as the standard z-score is more subject to strong outliers (compounds that fall outside of the defined upper and lower limits)18.
To remove “false positive” fluorescent compounds, the similarity index (SI17 was computed by comparing a region (500–540nm) of the donor only spectrum (l(a)) for each well to that of the plate-wide average DMSO spectrum (l(b)) in the same wavelength band as described in Eq. 421. Compounds that exceeded an SI robust z-score of 5 (corresponding to an SI of 2×10−4) were deemed likely fluorescent compounds and removed from consideration.
Spectral (SUPR) data was processed similarly to FLT-PR data, with the ΔR/G ratio being computed by applying the same moving average filter on the initial measurement of the ratio of the acceptor amplitude over the donor amplitude as found by fitting basis sets of the component spectra through least squares minimization. The hit threshold was also set using a robust z-score of 3. While the FLT-PR data and SUPR data showed a robust correlation, the FLT-PR data exhibited some strong outliers, presumably due to compounds directly modifying the donor lifetime. To eliminate these likely interfering compounds, correlation was enforced by eliminating compounds that exceed a robust z-score of 3 from the median value of the ratio of ΔFLT over the ΔR/G ratio metric. Additional interfering compounds were removed using two-channel lifetime detection18.
## Cardiac SR preparation
Cardiac SR vesicles were isolated from fresh porcine left ventricular tissue using differential centrifugation of the homogenized tissue as previously described58. The SR vesicles were flash-frozen and stored at −80°C until needed.
## Enzymatic SERCA activity assays of FRET hit compounds
Functional assays were performed using porcine cardiac SR vesicles16. An enzyme-coupled, NADH-linked ATPase assay was used to measure SERCA ATPase activity in 384-well microplates. Each well contained 50 mM MOPS (pH 7.0), 100 mM KCl, 1 mM EGTA, 0.2 mM NADH, 1 mM phosphoenol pyruvate, 10 IU/mL of pyruvate kinase, 10 IU/mL of lactate dehydrogenase, 7 μM of the calcium ionophore A23187 (Sigma), and CaCl2 was added to set free [Ca2+] to three different concentrations59. The Ca2+-ATPase were measured at VMAX (saturating, pCa 5.4), VM!D (subsaturating, midpoint, pCa 6.2), and basal (non-activating, pCa 8.0) [Ca2+]. 10 μg/mL of SR vesicle, calcium, compound (0.048 to 50 μM), and assay mix were incubated for 20 min at room temperature before measurement of functional assays with each of the 18 hit compounds, because a shorter incubation time than the FRET live-cell assays achieved optimal responses. The assay was started upon the addition of MgATP at a final concentration of 5 mM (total volume to 80 μL), and absorbance was read in a SpectraMax Plus microplate spectrophotometer from Molecular Devices (Sunnyvale, CA) at 340nm.
## Ca2+-transport assays of FRET hit compounds
Ca2+-transport assays were performed with similar porcine SR samples as in the Ca2+-ATPase assays described above. The compound effect on the Ca2+-transport activity of SERCA2a was determined using an oxalate-supported assay in which the change in fluorescence in a Ca-sensitive dye, Fluor-4, was determined as previously described18. A buffered solution containing 50 mM MOPS (pH 7.0), 100 mM KCl, 30 mg/mL sucrose, 1 mM EGTA, 10 mM potassium oxalate, 2 μM Fluo-4, 30 μg/mL porcine cardiac SR vesicles, CaCl2 calculated to reach the free [Ca2+] (pCa 8.0, 6.2, and 5.4), and compound (0.048 to 50μM) was dispensed into 384-well black walled, transparent bottomed plates (Greiner Bio-One) containing the tested small molecule and incubated at 22°C for 20 minutes while covered and protected from light. To start the reaction, MgATP was added to a final concentration of 5 mM, and the decrease in 485-nm excited fluorescence of Fluo-4 was monitored at 520 nm for 15 min using a FLIPR Tetra (Molecular Devices, San Jose, CA).
## Data analysis of FRET CRC assays of hit compounds
FRET efficiency (E) (Eq. 2) was determined as the fractional decrease of donor (1CS) lifetimes (τD) in the presence of acceptor (2CS) fluorophore (τDA) due to FRET as described in Eq. 1 and normalized to DMSO controls.
## Data analysis of Ca2+-ATPase and Ca2+-transport CRC assays
SERCA2a ATPase (or Ca-transport) activity at pCa 8.0 was subtracted from pCa 5.4 and pCa 6.2 values. The % effect ATPase (or Ca-transport transport) activity was normalized to the DMSO only (or 2CS in the absence of compound), and then were plotted against [Ca], and the curves were fitted using the Hill function, where V is the initial ATPase rate (or fluorescence rate), VMAX is the ATPase (or Ca2+-transport) at saturating [Ca2+], and EC50 or pKCa, or VM!D is the apparent Ca2+ dissociation constant as described previously ATPase (or Ca2+-transport) at (Midpoint Ca2+)44. These parameters and the [Ca2+] at $10\%$ (C10) above or below baseline pCa (8.0) are reported in Table 1.
## Cheminformatic analysis of hit compounds
An online interactive program was used to perform cheminformatics analysis60 to determine whether the hit compounds had structural similarity by identifying common chemical scaffolds (core structural feature) using binning, multidimensional scaling (MDS), and compound similarity methods where the Tanimoto coefficient31 and maximum common substructure31 values were used to determine clustering (Supplementary Table S1). The physicochemical properties (for e.g. Lipinski Rule of 5) and bioactivity properties of the compounds were also used in the clustering analysis34. A cluster contained two or more compounds with similarity score > 0.4, while a unique compound with a similarity score < 0.4 was referred to as a singleton.
## Statistical analysis
Analysis of two-group comparisons was done by a two-tailed unpaired Student’s t-test (*$p \leq 0.05$) using the data analysis program Microsoft Excel (Santa Rosa, CA). Data are presented as mean ± SEM calculated from a minimum of three separate experiments ($$n = 3$$).
## Data availability
All the data discussed are presented within the article and Supplementary Information and are available from the corresponding authors (OR and DDT) on reasonable request.
## References
1. Bers D. M., Eisner D. A., Valdivia H. H.. **Sarcoplasmic reticulum Ca2+ and heart failure: roles of diastolic leak and Ca2+ transport**. *Circ Res* (2003) **93** 487-490. DOI: 10.1161/01.RES.0000091871.54907.6B
2. Bers D. M.. **Cardiac sarcoplasmic reticulum calcium leak: basis and roles in cardiac dysfunction**. *Annu Rev Physiol* (2014) **76** 107-127. DOI: 10.1146/annurev-physiol-020911-153308
3. Viskupicova J., Rezbarikova P. **Natural Polyphenols as SERCA Activators: Role in the Endoplasmic Reticulum Stress-Related Diseases**. *Molecules* (2022) **27**. DOI: 10.3390/molecules27165095
4. Bers D. M., Despa S.. **Cardiac myocytes Ca2+ and Na+ regulation in normal and failing hearts**. *J Pharmacol Sci* (2006) **100** 315-322. PMID: 16552170
5. Rathod N.. **Nothing Regular about the Regulins: Distinct Functional Properties of SERCA Transmembrane Peptide Regulatory Subunits**. *Int J Mol Sci* (2021) **22**. DOI: 10.3390/ijms22168891
6. Li A.. **The transmembrane peptide DWORF activates SERCA2a via dual mechanisms**. *J Biol Chem* (2021) **296** 100412. DOI: 10.1016/j.jbc.2021.100412
7. Fisher M. E.. **Dwarf open reading frame (DWORF) is a direct activator of the sarcoplasmic reticulum calcium pump SERCA**. *Elife* (2021) **10**. DOI: 10.7554/eLife.65545
8. Rustad M. D., Roopnarine O., Cornea R. L., Thomas D. D.. **Interaction of DWORF with SERCA and PLB as determined by EPR spectroscopy**. *Biochem Biophys Res Commun* (2023) **645** 97-102. DOI: 10.1016/j.bbrc.2023.01.041
9. Nelson B. R.. **A peptide encoded by a transcript annotated as long noncoding RNA enhances SERCA activity in muscle**. *Science* (2016) **351** 271-275. DOI: 10.1126/science.aad4076
10. Makarewich C. A.. **The DWORF micropeptide enhances contractility and prevents heart failure in a mouse model of dilated cardiomyopathy**. *Elife* (2018) **7**. DOI: 10.7554/eLife.38319
11. Hadri L., Hajjar R. J.. **Calcium cycling proteins and their association with heart failure**. *Clin Pharmacol Ther* (2011) **90** 620-624. DOI: 10.1038/clpt.2011.161
12. Sato D., Uchinoumi H., Bers D. M.. **Increasing SERCA function promotes initiation of calcium sparks and breakup of calcium waves**. *J Physiol* (2021) **599** 3267-3278. DOI: 10.1113/JP281579
13. Guarnieri A. R., Benson T. W., Tranter M.. **Calcium cycling as a mediator of thermogenic metabolism in adipose tissue**. *Mol Pharmacol* (2022). DOI: 10.1124/molpharm.121.000465
14. Nowack J., Giroud S., Arnold W., Ruf T.. **Muscle Non-shivering Thermogenesis and Its Role in the Evolution of Endothermy**. *FrontPhysiol* (2017) **8** 889. DOI: 10.3389/fphys.2017.00889
15. Michelangeli F., East J. M.. **A diversity of SERCA Ca2+ pump inhibitors**. *Biochem Soc Trans* (2011) **39** 789-797. DOI: 10.1042/BST0390789
16. Gruber S. J.. **Discovery of enzyme modulators via high-throughput time-resolved FRET in living cells**. *J BiomolScreen* (2014) **19** 215-222. DOI: 10.1177/108705711351074019/2/215
17. Schaaf T. M.. **High-Throughput Spectral and Lifetime-Based FRET Screening in Living Cells to Identify Small-Molecule Effectors of SERCA**. *SLAS Discov* (2017b) **22** 262-273. DOI: 10.1177/1087057116680151
18. Schaaf T. M.. **Live-Cell Cardiac-Specific High-Throughput Screening Platform for Drug-Like Molecules that Enhance Ca(2+) Transport**. *Cells* (2020) **9**. DOI: 10.3390/cells9051170
19. Cornea R. L.. **High-throughput FRET assay yields allosteric SERCA activators**. *J Biomol Screen* (2013) **18** 97-107. DOI: 10.1177/1087057112456878
20. Schaaf T. M.. **Red-Shifted FRET Biosensors for High-Throughput Fluorescence Lifetime Screening**. *Biosensors (Basel)* (2018) **8**. DOI: 10.3390/bios8040099
21. Schaaf T. M., Peterson K. C., Grant B. D., Thomas D. D., Gillispie G. D.. **Spectral Unmixing Plate Reader: High-Throughput, High-Precision FRET Assays in Living Cells**. *SLAS Discov* (2017a) **22** 250-261. DOI: 10.1177/1087057116679637
22. Lo C. H.. **Noncompetitive inhibitors of TNFR1 probe conformational activation states**. *Sci Signal* (2019) **12**. DOI: 10.1126/scisignal.aav5637
23. Lo C. H., Schaaf T. M., Thomas D. D., Sachs J. N.. **Fluorescence-Based TNFR1 Biosensor for Monitoring Receptor Structural and Conformational Dynamics and Discovery of Small Molecule Modulators**. *Methods Mol Biol* (2021) **2248** 121-137. DOI: 10.1007/978-1-0716-1130-2_9
24. Lo C. H.. **Discovery of Small Molecule Inhibitors of Huntingtin Exon 1 Aggregation by FRET-Based High-Throughput Screening in Living Cells**. *ACS Chem Neurosci* (2020) **11** 2286-2295. DOI: 10.1021/acschemneuro.0c00226
25. Rebbeck R. T.. **Novel drug discovery platform for spinocerebellar ataxia, using fluorescence technology targeting beta-III-spectrin**. *J Biol Chem* (2021) **296** 100215. DOI: 10.1074/jbc.RA120.015417
26. Baell J. B., Holloway G. A.. **New substructure filters for removal of pan assay interference compounds (PAINS) from screening libraries and for their exclusion in bioassays**. *J Med Chem* (2010) **53** 2719-2740. DOI: 10.1021/jm901137j
27. Autry J. M.. **Sarcoplasmic Reticulum from Horse Gluteal Muscle Is Poised for Enhanced Calcium Transport**. *Vet Sci* (2021) **8**. DOI: 10.3390/vetsci8120289
28. Bidwell P. A.. **A Large-Scale High-Throughput Screen for Modulators of SERCA Activity**. *Biomolecules* (2022) **12**. DOI: 10.3390/biom12121789
29. Racker E., Eytan E.. **A coupling factor from sarcoplasmic reticulum required for the translocation of Ca2+ ions in a reconstituted Ca2+ATPase pump**. *J Biol Chem* (1975) **250** 7533-7534. PMID: 126239
30. Yu X., Inesi G.. **Variable stoichiometric efficiency of Ca2+ and Sr2+ transport by the sarcoplasmic reticulum ATPase**. *J Biol Chem* (1995) **270** 4361-4367. DOI: 10.1074/jbc.270.9.4361
31. Willett P. **The Calculation of Molecular Structural Similarity: Principles and Practice**. *Mol Inform* (2014) **33** 403-413. DOI: 10.1002/minf.201400024
32. Vitaku E., Smith D. T., Njardarson J. T.. **Analysis of the structural diversity, substitution patterns, and frequency of nitrogen heterocycles among U.S. FDA approved pharmaceuticals**. *J Med Chem* (2014) **57** 10257-10274. DOI: 10.1021/jm501100b
33. Shinde S. R.. **A systematic appraisal on catalytic synthesis of 1,3-oxazole derivatives: A mechanistic review on metal dependent synthesis**. *Synthetic Communications* (2022) **52** 1-36. DOI: 10.1080/00397911.2021.1989596
34. Lipinski C. A., Lombardo F., Dominy B. W., Feeney P. **J. Experimental and computational approaches to estimate solubility and permeability in drug discovery and development settings**. *Adv Drug DelivRev* (2001) **46** 3-26. DOI: 10.1016/s0169-409x(00)00129-0
35. Kranias E. G., Hajjar R. J.. **Modulation of cardiac contractility by the phospholamban/SERCA2a regulatome**. *Circ Res s* (2012) **110** 1646-1660. DOI: 10.1161/CIRCRESAHA.111.259754
36. Balamurugan R., Dekker F. J., Waldmann H.. **Design of compound libraries based on natural product scaffolds and protein structure similarity clustering (PSSC)**. *Mol Biosyst* (2005) **1** 36-45. DOI: 10.1039/b503623b
37. Koch M. A.. **Charting biologically relevant chemical space: a structural classification of natural products (SCONP)**. *Proc Natl Acad Sci U S A* (2005) **102** 17272-17277. DOI: 10.1073/pnas.0503647102
38. Ma B., Shatsky M., Wolfson H. J., Nussinov R.. **Multiple diverse ligands binding at a single protein site: a matter of pre-existing populations**. *Protein Sci* (2002) **11** 184-197. DOI: 10.1110/ps.21302
39. Lebeche D., Davidoff A. J., Hajjar R. J.. **Interplay between impaired calcium regulation and insulin signaling abnormalities in diabetic cardiomyopathy**. *Nat Clin Pract Cardiovasc Med* (2008) **5** 715-724. DOI: 10.1038/ncpcardio1347
40. Krajnak K., Dahl R.. **A new target for Alzheimer’s disease: A small molecule SERCA activator is neuroprotective in vitro and improves memory and cognition in APP/PS1 mice**. *Bioorg Med Chem Let* (2018) **28** 1591-1594. DOI: 10.1016/j.bmcl.2018.03.052
41. Nogami K.. **Pharmacological activation of SERCA ameliorates dystrophic phenotypes in dystrophin-deficient mdx mice**. *Hum Mol Genet* (2021) **30** 1006-1019. DOI: 10.1093/hmg/ddab100
42. Shareef M. A., Anwer L. A., Poizat C.. **Cardiac SERCA2A/B: therapeutic targets for heart failure**. *Eur J Pharmacol* (2014) **724** 1-8. DOI: 10.1016/j.ejphar.2013.12.018
43. Sordi G., Goti A., Young H. S., Palchetti I., Tadini-Buoninsegni F.. **Stimulation of Ca(2+) -ATPase Transport Activity by a Small-Molecule Drug**. *ChemMedChem* (2021) **16** 3293-3299. DOI: 10.1002/cmdc.202100350
44. Stroik D. R.. **Targeting protein-protein interactions for therapeutic discovery via FRET-based high-throughput screening in living cells**. *Sci Rep* (2018) **8** 12560. DOI: 10.1038/s41598-018-29685-z
45. Rocchetti M.. **Modulation of sarcoplasmic reticulum function by PST2744 [istaroxime; (E,Z)-3-((2-aminoethoxy)imino) androstane-6,17-dione hydrochloride)] in a pressure-overload heart failure model**. *J Pharmacol Exp Ther* (2008) **326** 957-965. DOI: 10.1124/jpet.108.138701
46. Gheorghiade M., Ambrosy A. P, Ferrandi M., Ferrari P. **Combining SERCA2a activation and Na-K ATPase inhibition: a promising new approach to managing acute heart failure syndromes with low cardiac output**. *Discov Med* (2011) **12** 141-151. PMID: 21878191
47. Luraghi A.. **Highly Selective SERCA2a Activators: Preclinical Development of a Congeneric Group of First-in-Class Drug Leads against Heart Failure**. *J Med Chem* (2022) **65** 7324-7333. DOI: 10.1021/acs.jmedchem.2c00347
48. Arici M.. **Istaroxime Metabolite PST3093 Selectively Stimulates SERCA2a and Reverses Disease-Induced Changes in Cardiac Function**. *J Pharmacol Exp Ther* (2023) **384** 231-244. DOI: 10.1124/jpet.122.001335
49. Avvisato R., Jankauskas S. S., Santulli G.. **Istaroxime and Beyond: New Therapeutic Strategies to Specifically Activate SERCA and Treat Heart Failure**. *J Pharmacol Exp Ther* (2023) **384** 227-230. DOI: 10.1124/jpet.122.001446
50. Bal N. C., Periasamy M.. **Uncoupling of sarcoendoplasmic reticulum calcium ATPase pump activity by sarcolipin as the basis for muscle non-shivering thermogenesis**. *Philos Trans R Soc Lond B Biol Sci* (2020) **375** 20190135. DOI: 10.1098/rstb.2019.0135
51. Meizoso-Huesca A., Pearce L., Barclay C. J., Launikonis B. S.. **Ca(2+) leak through ryanodine receptor 1 regulates thermogenesis in resting skeletal muscle**. *Proc Natl Acad Sci U S A* (2022) **119**. DOI: 10.1073/pnas.2119203119
52. Maurya S. K., Periasamy M.. **Sarcolipin is a novel regulator of muscle metabolism and obesity**. *Pharmacol Res* (2015) **102** 270-275. DOI: 10.1016/j.phrs.2015.10.020
53. Pagliaro L., Marchesini M., Roti G.. **Targeting oncogenic Notch signaling with SERCA inhibitors**. *J Hematol Oncol* (2021) **14** 8. DOI: 10.1186/s13045-020-01015-9
54. Nikolaienko R.. **New N-aryl-N-alkyl-thiophene-2-carboxamide compound enhances intracellular Ca(2+) dynamics by increasing SERCA2a Ca(2+) pumping**. *Biophys J* (2022). DOI: 10.1016/j.bpj.2022.12.002
55. Hou Z.. **2-Color calcium pump reveals closure of the cytoplasmic headpiece with calcium binding**. *PLoS One* (2012) **7** e40369. DOI: 10.1371/journal.pone.0040369P0NE-D-12-07470
56. Pallikkuth S.. **Phosphorylated phospholamban stabilizes a compact conformation of the cardiac calcium-ATPase**. *Biophys J* (2013) **105** 1812-1821. DOI: 10.1016/j.bpj.2013.08.045S0006-3495(13)01017-5
57. Talbot C. B.. **Correction Approach for Delta Function Convolution Model Fitting of Fluorescence Decay Data in the Case of a Monoexponential Reference Fluorophore**. *J Fluoresc* (2015) **25** 1169-1182. DOI: 10.1007/s10895-015-1583-4
58. Fruen B. R., Bardy J. M., Byrem T. M., Strasburg G. M., Louis C. F.. **Differential Ca(2+) sensitivity of skeletal and cardiac muscle ryanodine receptors in the presence of calmodulin**. *Am J Physiol Cell Physiol* (2000) **279** C724-733. DOI: 10.1152/ajpcell.2000.279.3.C724
59. Mueller B., Karim C. B., Negrashov I. V., Kutchai H., Thomas D. D.. **Direct detection of phospholamban and sarcoplasmic reticulum Ca-ATPase interaction in membranes using fluorescence resonance energy transfer**. *Biochemistry* (2004) **43** 8754-8765. PMID: 15236584
60. Backman T. W., Cao Y., Girke T.. **ChemMine tools: an online service for analyzing and clustering small molecules**. *Nucleic Acids Res* (2011) **39** W486-491. DOI: 10.1093/nar/gkr320
|
---
title: Polygenic risk of Social-isolation and its influence on social behavior, psychosis,
depression and autism spectrum disorder
authors:
- Adam Socrates
- Niamh Mullins
- Ruben Gur
- Raquel Gur
- Eli Stahl
- Paul O’Reilly
- Abraham Reichenberg
- Hannah Jones
- Stan Zammit
- Eva Velthorst
journal: Research Square
year: 2023
pmcid: PMC10002835
doi: 10.21203/rs.3.rs-2583059/v1
license: CC BY 4.0
---
# Polygenic risk of Social-isolation and its influence on social behavior, psychosis, depression and autism spectrum disorder
## Abstract
Social-isolation has been linked to a range of psychiatric issues, but the behavioral component that drives it is not well understood. Here, a GWAS is carried out to identify genetic variants which contribute to Social-isolation behaviors in up to 449,609 participants from the UK Biobank. 17 loci were identified at genome-wide significance, contributing to a $4\%$ SNP heritability estimate. Using the Social-isolation GWAS, polygenic risk scores (PRS) were derived in ALSPAC, an independent, developmental cohort, and used to test for association with friendship quality. At age 18, friendship scores were associated with the Social-isolation PRS, demonstrating that the genetic factors are able to predict related social traits. LD score regression using the GWAS demonstrated genetic correlation with autism spectrum disorder, schizophrenia, and major depressive disorder. However, no evidence of causality was found using a conservative Mendelian randomization approach other than that of autism spectrum disorder on Social-isolation. Our results show that Social-isolation has a small heritable component which may drive those behaviors which is associated genetically with other social traits such as friendship satisfaction as well as psychiatric disorders.
## Introduction
Social contact is essential for surviving and thriving in human societies1. As such, having limited contact with other people, or Social-isolation, can have detrimental effects on both physical and mental health. There is evidence that lack of social contact is associated with schizophrenia2,3, autism spectrum disorder3, and depression4, as well as with medical conditions such as cardiovascular disease1,5 and diabetes6. Longitudinal studies indicate that Social-isolation can predate mental issues and have a strong causal effect on poor mental health outcomes4,7,8. These issues have been acutely brought to light in the context of the Covid-19 pandemic, in which forced social isolation has had a substantial negative effect on mental health9. Social-isolation has been found to be strongly associated with the development of psychosis, and it has been hypothesized that this contribution may be due to negative, delusional or paranoid thoughts not being tested in reality and therefore corrected in social interactions10,11 Despite the impact of Social-isolation on mental and physical health, it remains one of the least studied factors in psychiatric disorders, limiting understanding of aetiology and causality with regards to psychiatric disorders 12,13,14,15. Associations between genetics and traits related to social contact such as feelings of loneliness (feelings of distress or discomfort from being alone) and sociability (the ability to connect and socialize with others) have been noted16. However, the existence and influence of an exclusive genetic predisposition towards Social-isolation behaviors, i.e., action that leads to isolation, is yet to be established. Consequently, there is a fundamental gap in our knowledge about the extent to which Social-isolation may represent a causal and independent risk for poor mental and physical health instead of being merely a direct consequence of other (clinical) symptomatology, for example due to stress or feelings of paranoia.
Twin studies have demonstrated that there is a similar genetic influence on both social isolation ($40\%$) and loneliness ($38\%$), but that they are only moderately genetically correlated, reflecting partially distinct constructs17. However, to our knowledge no prior study has carried out a genome-wide association study (GWAS) to elucidate the polygenic component of the purely behavioral aspects of Social-isolation, as distinct feelings relating to social behavior such as loneliness. This is pertinent, as these behaviors could provide modifiable early intervention targets if found to be on the causal pathway between inherited genetic variation and psychiatric disorders18.
In order to better understand the genetic factors that influence Social-isolation, the present study [1] conducted a novel GWAS for Social-isolation behavior in the UK Biobank cohort; [2] Polygenic Risk Scores (PRS) were derived from this GWAS for individuals in the Avon Longitudinal Study of Parents and Children (ALSPAC, UK) and used to examine associations with social traits for GWAS validation; [3] the genetic correlation between Social-isolation and psychiatric disorders was examined using GWAS results from the Psychiatric Genomics Consortium (PGC), and [4]). Finally, Mendelian Randomization (MR) was applied to estimate causal effects between Social-isolation and psychiatric disorders.
## GWAS
To investigate genetic propensity towards social isolation behavior (Social-isolation), a GWAS was performed in the UK Biobank, based on a composite of 4 self-reported behavioral traits pertaining to this behavior. LD Score regression revealed that the individual traits were genetically correlated (see Supplementary Table 4). These were meta-analyzed with MTAG before being conditioned on schizophrenia (SCZ), major depressive disorder (MDD), and autism spectrum disorder (ASD), using mtCOJO. The initial GWAS identified 19 loci, post-conditioning 17 loci remained at genome-wide significance ($P \leq 5$x10−08; see Fig. 1).
The majority of the SNPs found to be associated with Social-isolation were not previously associated with psychiatric or neurodevelopmental disorders. However, there are several exceptions. For example, the top lead SNP (rs67777906; $$P \leq 1.80$$x10−15) is situated in the ARFGEF2 gene, implicated in distinguishing between bipolar disorder (BD) and SCZ32, as well as post-traumatic stress disorder (PTSD)33,34. The second top SNP in chromosome 8, and the fourth top hit overall (rs2721942; 1.47x10−10), has also been associated with Post Traumatic Stress Disorder (PTSD)35. In chromosome 19, the lead SNP (rs28567442; $$P \leq 6.31$$x10−10) is embedded in ZNF536, implicated in the development of the forebrain, and associated with SCZ36. Other genome-wide significant SNPs are in genes associated with SCZ (rs6125539; 4.72x10−09; CSE1L)32 and impulsivity (rs1248860; 9.51x10−09; CADM2)37. In chromosome 13, rs17057528 ($$P \leq 8.82$$x10−09) is in DIAPH3, identified as an autism risk gene38, and is also implicated in hearing loss and impairment of speech perception39.
## ALSPAC
To validate the Social-isolation GWAS and PRS in an independent cohort, as well as explore its generalizability to a developmental cohort, PRS were generated in ALSPAC using the 13 significance thresholds for SNP inclusion. The PRS were used to examine associations with friendship scores, comprising the 5 items relating to peer contact in $$n = 4$$,934 (at age 12) and $$n = 2$$,909 (at age 18) participants of the ALSPAC cohort.
The Social-isolation PRS were not associated friendship scores at age 12. At age 18, friendship score was significantly associated with the Social-isolation PRS at the PT = 0.05 and PT = 0.1 threshold, with the latter being the most strongly associated (r2 = 0.006, $$P \leq 0.001$$; see Supplementary Tables 6 and 7 for full results). The fewer SNPs were included, the less the predictive the model in terms of p-value, with the genome-wide significant only SNPs not associated with the friendships scores. This demonstrates the signal included in SNPs that did not reach genome-wide significance in contributing towards social isolation behavior.
## LD score regression
LD score regression was performed to investigate genetic correlations between Social-isolation in the UK Biobank and schizophrenia (SCZ), major depressive disorder (MDD) and autism spectrum disorder (ASD) from the Psychiatric Genomics Consortium (PGC). All 3 psychiatric disorders were correlated with Social-isolation, with ASD having the strongest genetic correlation (rg = 0.23, SE = 0.048, $$P \leq 2.25$$x10−06), followed by SCZ (rg = 0.102, SE = 0.028, $$P \leq 0.0002$$) and MDD (rg = 0.093, SE = 0.035, $$P \leq 0.009$$). The results indicate that Social-isolation genetics are associated with the genetics of these psychiatric disorders and may form part of the genetic basis for them. This could occur if the genetics of Social-isolation have downstream effects on behavior that could increase risk of symptoms and eventual diagnosis, or if the diagnosis itself leads to increased Social-isolation. Using LD score regression, the SNP-heritability of Social-isolation after conditioning on psychiatric disorders was estimated to be h2 = 0.04 (SE = 0.0022, $$P \leq 8.95$$x10−77), suggesting a small but significant SNP-based heritable component.
Genetic correlations and heritability estimates were conducted using LD score regression31, to investigate associations between Social-isolation and SCZ, MDD, and ASD, using GWAS summary statistics from the Social-isolation GWAS conducted in the UK Biobank and each psychiatric disorder from the Psychiatric Genomics Consortium (PGC).
## Mendelian randomization
Using the MR-Egger method, there was no evidence of causal relationships between Social-isolation and psychiatric disorders with the exception of ASD having a causal effect on Social-isolation when SNPs were selected as instruments at the 5x10−05 threshold (Beta = 0.019, SE = 0.0052, $$P \leq 0.00041$$). However, MR-*Egger is* a conservative method and may be underpowered to detect causal associations in complex behavioral traits. See Supplementary Table 8 for full results.
To test for causality between Social-isolation and psychiatric outcomes, bi-directional Mendelian Randomization was conducted using the package TwoSampleMR (https://github.com/MRCIEU/TwoSampleMR). Instrumental variables for the exposures (both Social-isolation and the psychiatric disorders SCZ, MDD, and ASD) were extracted at genome-wide significance and at $p \leq 5$x10−05 after strict LD clumping at 10,000kb windows and LD r2 < 0.001 to ensure instruments were independent. Exposure and outcomes were harmonized and MR-Egger was used in the primary analyses to account for horizontal pleiotropy. The inverse variance weighted (IVW) was also used as a less conservative, more powerful approach. To account for multiple testing, a Bonferroni corrected p-value threshold of $P \leq 0.004$ was used to ascertain significance.
## Discussion
This is the first study of the genetic factors that contribute to the behavior of social isolation. A GWAS identified 17 genetic loci which predispose towards social isolation behavior. Some of these were in genes previously associated with psychiatric and neurological disorders, as well as neurotransmitter and brain function. However, most were not previously found to be associated with other mental health, neurodevelopmental, or personality traits. Polygenic risk scores (PRS) derived from the GWAS were associated with the friendship scores at age 18 and there was strong evidence supporting shared genetic etiology between Social-isolation and major psychiatric disorders, based on genetic correlations.
The PRS generated in ALSPAC were associated with friendship scores at age 18 but not at age 12. These results suggest Social-isolation GWAS is valid indicator of social-related traits, with higher PRS associated with worse friendship satisfaction and outcomes. The PRS being associated scores at age 18 as opposed to 12 might indicate that genetically influenced personal social behavior does not necessarily manifest until later in adolescence. This finding could be due to confounding by gene-environment correlation40. At younger ages, children may have less control over their own social environments and interactions than at age 18, as their parents would likely select their environments for them, in which case behavior would be less strongly influenced by their own genetic predispositions. A similar effect is observed in intelligence genetics, in which heritability increases over time41. It is considered that genetic predisposition leads to active and passive correlations with school selection or teacher attention, for example, creating a “snowball” effect in which those genetic influences are amplified over time. It is possible that similar effects are at play with behavioral genetics, in which Social-isolation genetic predisposition lead to development, or lack thereof, of social skills and sociability, modulating social isolation over time.
Social-isolation was found to be genetically correlated with SCZ, as well as with ASD and MDD. This pattern of results suggest that Social-isolation is a feature that cuts across multiple psychiatric disorders and mental health generally. It is well known that social isolation is linked to poorer mental health42, but here it is shown that there is a genetic association which indicates that Social-isolation may form part of the aetiological basis of these disorders. Further studies with psychiatric cases will be required to test this hypothesis, but considering that social engagement is an easily modifiable intervention target43, identifying those with a genetic predisposition towards Social-isolation may be a useful strategy in mitigating mental health issues.
The current study was able to demonstrate a heritable genetic component to Social-isolation by utilizing a large sample size and detailed phenotype information, allowing a comprehensive and valid Social-isolation trait to be developed. This was confirmed by the PRS generated from the GWAS being validated in an independent sample, and several genome-wide significant SNPs found associated with Social-isolation. However, LD Score Regression only estimated $4\%$ heritability for Social-isolation and PRS were only able to explain $0.6\%$ of the variance in friendship scores in ALSPAC replication sample. The SNP-heritability is likely to be a lower bound estimate, as this only takes into account the common SNPs genotyped and not rare variants or de novo mutations44. Further, despite having up to 450,000 individuals available for the GWAS, the most powerful GWAS such as educational attainment are becoming increasingly predictive with approximately 3 million participants45. Thus, increasing sample size will allow the detection of more SNPs that contribute to Social-isolation behavior and increase both heritability estimates and the predictive power of PRS. In ALSPAC the target sample also had relatively few participants at age 18 ($$n = 2$$,909) compared to age 12 ($$n = 4$$,934), which likely contributed to lower bound variance explained.
In order to further investigate how the genetic component of Social-isolation manifests in behavior and the development of psychiatric disorders, further studies will be required which investigate whether or not Social-isolation PRS are able to predict case control status for disorders such as SCZ, MDD and ASD. If so, it will be necessary to consider which specific behaviors are influenced by genetics, and how these manifest in the development and diagnosis of psychiatric disorders. By targeting behavior, our present study has laid the foundation for identifying a possible target for intervention that can be addressed in real world scenarios. However, the relatively small effect sizes of individual SNPs and the resulting low predictive power of PRS mean further investigation is necessary.
## Discovery sample
The UK Biobank (UKB) is a detailed prospective study with over 502,650 participants aged 40–69 years when recruited in 2006–2010, and includes both genetic and phenotypic data on complex traits19. The recruitment process was coordinated around 22 centers in the UK (between 2007 and 2010). Individuals within travelling distance of these centers were identified using NHS patient registers (response rate = $5.47\%$). Invitations were sent using a stratified approach to ensure demographic parameters were in concordance with the general population. All participants provided written informed consent and the current study was ethically approved by the UK Biobank Ethics and Governance Council (REC reference 11/NW/0382; UK Biobank application reference 18177).
## Genetic data
Blood samples from 488,366 UK Biobank participants were genotyped using the UK BiLEVE array or the UK Biobank axiom array. Further details on the genotyping and quality control (QC) can be found on the UK Biobank website (http://www.ukbiobank.ac.uk/scientists-3/genetic-data/). In the current study, SNPs were removed if they had missingness < 0.02 and a minor allele frequency (MAF) < 0.01. Exclusions based on heterozygosity and missingness were implemented according to UK Biobank recommendations (http://biobank.ctsu.ox.ac.uk/showcase/label.cgi?id=100314). Samples were removed if they were discordant for sex. SNPs deviating from Hardy-*Weinberg equilibrium* (HWE) were removed at a threshold of $P \leq 1$x10−8. Genotype data was imputed according to standard UK Biobank procedure, on 487,442 samples20, excluding variants with an MAF < 0.01 and an imputation quality score < 0.3. After basic QC procedures and exclusions, 488,337 samples with phenotype data remained for genetic analysis. Excluding those of non-European ancestry using 4-mean clustering on the first two principal components, 449,609 samples remained for genetic analysis.
## Social isolation:
To derive a comprehensive measure of Social Isolation (Social-isolation), we ran a data-driven principal component analyses (using Promax rotation) on self-reported answers to questions that [1] directly probed for the quantity or quality of social engagement, and [2] were available for at least $90\%$ of study participants. Based on these criteria, we included data on the following 3 items, that all loaded on a single factor: “Frequency of family/friend visits”, “Being able to confide in others”, and “Number of social activities a week”. The items “Frequency of family/friend visits” and “Being able to confide with others” were both rated on a seven-point Likert scale (i.e. ‘Almost daily’, ‘2–4 times a week’, ‘about once a week’, ‘about once a month’, ‘once every few months’, ‘never or almost never’, and ‘no friends/ family outside of household’). The items “Frequency of family/friend visits” and “Being able to confide with others” were considered continuously and recoded so that higher values corresponded to greater social isolation. Answer options for the item “Number of a/social activities a week” included attending a sports club, pub, social club, religious group, adult educational classes, or other group activities and were summed to represent the ‘total number of social activities a week’, also considered continuously.
To complement the answers to the self-report, sociodemographic information about the number of people in the household was added as additional proxy of social contact. This “Number in household” item was dichotomized as a binary trait representing living alone, with 0 others in household coded as ‘1’ for Social-isolation and any greater number in household as ‘0’. See supplementary material for full phenotype and coding details. For all items, individuals with missing data, or who preferred not to answer were excluded. Participants who were wheelchair users or morbidly obese (BMI > 40) were also excluded from the analysis, as these factors may arguably hamper the level of social activity, but are unrelated to genetic or psychiatric vulnerability.
## ALSPAC cohort
The Avon Longitudinal Study of Parents and Children (ALSPAC) is a prospective birth cohort which recruited pregnant women with expected delivery dates between April 1991 and December 1992 from Bristol UK. 14,541 pregnant women were initially enrolled with 14,062 children born and 13,988 alive at 1 year of age. Detailed information on health and development of children and their parents were collected from regular clinic visits and completion of questionnaires. Please note that the study website contains details of all the data that is available through a fully searchable data dictionary and variable search tool” and reference the following webpage: http://www.bristol.ac.uk/alspac/researchers/our-data/. A detailed description of the cohort has been previously published21,22. Ethical approval for the study was obtained from the ALSPAC Ethics and Law Committee and the Local Research Ethics Committees.
## Genotype data
9,115 participants in ALSPAC have genotype data available, after individuals with non-European ancestry were removed. ALSPAC children were genotyped using the Illumina HumanHap550 quad chip genotyping platforms. SNPs with a MAF of < 0.01, a call rate of < 0.95 or evidence for violations of Hardy-*Weinberg equilibrium* ($P \leq 5$x10−07) were removed. Data was imputed using standard ALSPAC procedure using the HapMap 2 reference panel, keeping SNPs with MAF > 0.02 and an INFO score > 0.9. This resulted in 4,731,235 SNPs in the analysis. Full quality control procedures can be found at: https://alspac.github.io/omics_documentation/alspac_omics_data_catalogue.html
## Phenotype data
To test the validity of the Social-isolation construct, 2 friendship scores were derived from 5 questions from clinical questionnaires based on questions from the Cambridge Hormones and Moods Project Friendship Questionnaire23, completed by the parents of offspring at ages 12 and 18 respectively e.g. “*Teenager is* happy with number of friends”. Each question consisted of 4 to 6 categorical responses, corresponding to a 4 to 6 point scale e.g. “1 = Very happy, 2 = Quite happy, 3 = Quite unhappy, 4 = Unhappy, 5 = No friends”. Responses were summed to create a continuous scale, with higher scores corresponding to lower friendship quality and greater Social-isolation. 4, 934 of the cohort had the phenotype information at age 12, and 2,909 at age 18. See supplementary table 2 for full details on questions.
## GWAS summary statistics
To test for genetic correlations between Social-isolation and associated psychiatric disorders using LD score regression, the Social-isolation GWAS based on UK *Biobank data* was used along 3 base genome-wide association summary statistics for schizophrenia (SCZ), depression (MDD), and autism spectrum disorder (ASD). These were the Psychiatric Genomics Consortium Wave 3 (PGC3) SCZ GWAS24, the 2019 PGC MDD Working Group GWAS25, and the 2017 PGC ASD Working Group GWAS26.
## GWAS analysis
Association testing of autosomal SNPs was carried out on each of the 4 Social-isolation traits using BOLT Bayesian linear mixed models (BOLT-LMM)27 to account for relatedness and cryptic population stratification, while increasing power and controlling for false positives. Age, sex, batch, and center were included as covariates, as well as education, income, and Townsend deprivation index (TDI) to account for socio-economic status (SES). The top 15 principal components (PCs) were also included to control for main population stratification. MTAG28 was used to meta-analyze the individual “Frequency of family/friend visits”, “Being able to confide in others”, “Number of social activities a week”, and “Number in household” outcomes to form a single, composite Social-isolation GWAS. This score is achieved by leveraging power across correlated GWAS estimates in overlapping samples. Finally, multitrait-based conditional and joint analysis (mtCOJO)29 was used to adjust the Social-isolation GWAS summary statistics for the effects of psychiatric disorders, specifically schizophrenia (SCZ), major depressive disorder (MDD), and autism spectrum disorder (ASD), using European ancestry GWAS summary statistics for each. These are the psychiatric disorders which are commonly considered to lead to increased risk of social withdrawal and isolation2,3,4,7,8 and were conditioned on to remove potential downstream effects of psychiatric disorders. SNPs were selected as instruments at 5x10−05, clumped 1MB apart or with LD r2 < 0.2 based on the 1000 Genomes Project Phase 3 reference panel for independence. mtCOJO uses these SNPs Generalized Summary-data-based Mendelian Randomization (GSMR) to estimate the effect of the exposures (psychiatric disorders) on the outcome (Social-isolation), producing conditioned effect sizes and p-values. Statistically significant independent signals were identified using 1MB clumping and a genome-wide significance threshold of $P \leq 5$x10−08.
## Polygenic risk score analysis
Polygenic risk scores (PRS) were generated in ALSPAC using PRSice-230, using the Social-isolation GWAS to sum and weight risk alleles for individuals in each cohort. Social-isolation GWAS results were pruned for linkage disequilibrium (LD) using the p-value informed clumping method in PLINK (--clump-p1 1 -- clump-p2 1 --clump-r2 0.1 --clump-kb 250). This method preferentially retains SNPs with the strongest evidence of association and removes SNPs in LD (r2 > 0.1) that show weaker evidence of association within 250Kb windows, based on LD structure from the HRC reference panel. Subsets of SNPs were selected from the results at 13 increasingly liberal P value thresholds (ranging from $p \leq 5$x10−08, to $p \leq 0.5$). Risk alleles were included and tested to predict outcomes at 13 different significance thresholds, allowing the utilization of the most predictive PRS and threshold. These PRS were tested for associations with the friendship scores in ALSPAC, using linear regression models and including age, sex and 10 PCs as covariates. To account for the multiple testing of 13 PRS thresholds and 2 friendship scores, a Bonferroni correct significance threshold of $P \leq 0.002$ was used.
## Data availability
UK *Biobank data* are available through a procedure described at http://www.ukbiobank.ac.uk/using-the-resource/ ALSPAC data access is through a system of managed open access. The steps below highlight how to apply for access to the data included in this paper and all other ALSPAC data.
If you have any questions about accessing data, please [email protected].
Schizophrenia, *Autism spectrum* disorder, and major depressive disorder GWAS summary statistics are publicly available from the PGC (https://www.med.unc.edu/pgc/download-results/)
## Code availability
Software code for PRSice-2 is available at https://www.prsice.info/. All other code used is available upon request.
## References
1. **The cooperative human**. *Nat Hum Behav* (2018) **2** 427-428. DOI: 10.1038/s41562-018-0389-1
2. Green M. F.. **Social Disconnection in Schizophrenia and the General Community**. *Schizophr Bull* (2018) **44** 242-249. DOI: 10.1093/schbul/sbx082
3. Velthorst E.. **Genetic risk for schizophrenia and autism, social impairment and developmental pathways to psychosis**. *Transl Psychiatry* (2018) **8** 204. DOI: 10.1038/s41398-018-0229-0
4. Santini Z. I., Jose P. E., Cornwell E. Y., Koyanagi A., Nielsen L., Hinrichsen C., Koushede V.. **Social disconnectedness, perceived isolation, and symptoms of depression and anxiety among older Americans (NSHAP): a longitudinal mediation analysis**. *The Lancet Public Health* (2020) **5** e62-e70. PMID: 31910981
5. Carpenter W. T.. **Social withdrawal as psychopathology of mental disorders**. *Neuroscience and Biobehavioral Reviews* (2019) **97** 85-86. DOI: 10.1016/j.neubiorev.2017.12.011
6. Brinkhues S., Dukers-Muijrers N. H., Hoebe C. J., Van Der Kallen C. J., Dagnelie P. C., Koster A., Schram M. T.. **Socially isolated individuals are more prone to have newly diagnosed and prevalent type 2 diabetes mellitus-the Maastricht study**. *BMC Public Health* (2017) **17** 1-12. PMID: 28049454
7. Noguchi T., Saito M., Aida J., Cable N., Tsuji T., Koyama S., Kondo K.. **Association between social isolation and depression onset among older adults: a cross-national longitudinal study in England and Japan**. *BMJ open* (2021) **11** e045834
8. Velthorst E., Fett A. K. J., Reichenberg A., Perlman G., van Os J., Bromet E. J., Kotov R.. **The 20-year longitudinal trajectories of social functioning in individuals with psychotic disorders**. *American Journal of Psychiatry* (2017) **174** 1075-1085. PMID: 27978770
9. Pancani L., Marinucci M., Aureli N., Riva P.. **Forced social isolation and mental health: a study on 1,006 Italians under COVID-19 lockdown**. *Frontiers in Psychology* (2021) **12** 1540
10. Velthorst E., Nieman D. H., Becker H. E., van de Fliert R., Dingemans P. M., Klaassen R., Linszen D. H.. **Baseline differences in clinical symptomatology between ultra high risk subjects with and without a transition to psychosis**. *Schizophrenia research* (2009) **109** 60-65. PMID: 19272756
11. Velthorst E., Meijer C.. **The association between social anhedonia, withdrawal and psychotic experiences in general and high-risk populations**. *Schizophrenia Research* (2012) **138** 290-294. PMID: 22484023
12. Kamenov K., Cabello M., Caballero F. F., Cieza A., Sabariego C., Raggi A., Ayuso-Mateos J. L.. **Factors related to social support in neurological and mental disorders**. *PLoS One* (2016) **11** e0149356. PMID: 26900847
13. Kennedy N., Foy K., Sherazi R., McDonough M., McKeon P.. **Long-term social functioning after depression treated by psychiatrists: a review**. *Bipolar disorders* (2007) **9** 25-37. PMID: 17391347
14. Simons C. J., Bartels-Velthuis A. A., Pijnenborg G. H.. **Cognitive performance and long-term social functioning in psychotic disorder: a three-year follow-up study**. *PloS one* (2016) **11** e0151299. PMID: 27082629
15. Bellido-Zanin G., Pérez-San-Gregorio M. Á., Martín-Rodríguez A., Vázquez-Morejón A. J.. **Social functioning as a predictor of the use of mental health resources in patients with severe mental disorder**. *Psychiatry research* (2015) **230** 189-193. PMID: 26343834
16. Day F. R., Ong K. K., Perry J. R.. **Elucidating the genetic basis of social interaction and isolation**. *Nature communications* (2018) **9** 1-6
17. Matthews T., Danese A., Wertz J., Odgers C. L., Ambler A., Moffitt T. E., Arseneault L.. **Social isolation, loneliness and depression in young adulthood: a behavioural genetic analysis**. *Social psychiatry and psychiatric epidemiology* (2016) **51** 339-348. PMID: 26843197
18. Lewis C. M., Vassos E.. **Polygenic risk scores: from research tools to clinical instruments**. *Genome medicine* (2020) **12** 1-11
19. Sudlow C., Gallacher J., Allen N., Beral V., Burton P., Danesh J., Collins R.. **UK biobank: an open access resource for identifying the causes of a wide range of complex diseases of middle and old age**. *PLoS medicine* (2015) **12** e1001779. PMID: 25826379
20. Bycroft C., Freeman C., Petkova D., Band G., Elliott L. T., Sharp K., Marchini J.. **The UK Biobank resource with deep phenotyping and genomic data**. *Nature* (2018) **562** 203-209. PMID: 30305743
21. Boyd A., Golding J., Macleod J., Lawlor D. A., Fraser A., Henderson J., Davey Smith G.. **Cohort profile: the ‘children of the 90s’—the index offspring of the Avon Longitudinal Study of Parents and Children**. *International journal of epidemiology* (2013) **42** 111-127. PMID: 22507743
22. Fraser A., Macdonald-Wallis C., Tilling K., Boyd A., Golding J., Davey Smith G., Lawlor D. A.. **Cohort profile: the Avon Longitudinal Study of Parents and Children: ALSPAC mothers cohort**. *International journal of epidemiology* (2013) **42** 97-110. PMID: 22507742
23. Goodyer I. M., Wright C., Altham P. M. E.. **Recent friendships in anxious and depressed school age children**. *Psychological medicine* (1989) **19** 165-174. PMID: 2727204
24. Ripke S., Walters J. T., O’Donovan M. C.. **Mapping genomic loci prioritises genes and implicates synaptic biology in schizophrenia**. *MedRxiv* (2020)
25. Howard D. M., Adams M. J., Clarke T. K., Hafferty J. D., Gibson J., Shirali M., McIntosh A. M.. **Genome-wide meta-analysis of depression identifies 102 independent variants and highlights the importance of the prefrontal brain regions**. *Nature neuroscience* (2019) **22** 343-352. PMID: 30718901
26. Anney R. J., Ripke S., Anttila V., Grove J., Holmans P. A., Huang H., Chakravarti A.. **Meta-analysis of GWAS of over 16,000 individuals with autism spectrum disorder highlights a novel locus at 10q24. 32 and a significant overlap with schizophrenia**. *Molecular Autism* (2017) **8** 21. PMID: 28540026
27. Loh P. R., Tucker G., Bulik-Sullivan B. K., Vilhjalmsson B. J., Finucane H. K., Salem R. M., Price A. L.. **Efficient Bayesian mixed-model analysis increases association power in large cohorts**. *Nature genetics* (2015) **47** 284-290. PMID: 25642633
28. Turley P., Walters R. K., Maghzian O., Okbay A., Lee J. J., Fontana M. A., Benjamin D. J.. **Multi-trait analysis of genome-wide association summary statistics using MTAG**. *Nature genetics* (2018) **50** 229-237. PMID: 29292387
29. Zhu Z., Zheng Z., Zhang F., Wu Y., Trzaskowski M., Maier R., Yang J.. **Causal associations between risk factors and common diseases inferred from GWAS summary data**. *Nature communications* (2018) **9** 1-12
30. Choi S. W., O'Reilly P. F.. **PRSice-2: Polygenic Risk Score software for biobank-scale data**. *Gigascience* (2019) **8** giz082. PMID: 31307061
31. Bulik-Sullivan B., Finucane H. K., Anttila V., Gusev A., Day F. R., Loh P. R., Neale B. M.. **An atlas of genetic correlations across human diseases and traits**. *Nature genetics* (2015) **47** 1236-1241. PMID: 26414676
32. Ruderfer D. M., Ripke S., McQuillin A., Boocock J., Stahl E. A., Pavlides J.M. W., Freedman R.. **Genomic dissection of bipolar disorder and schizophrenia,including 28 subphenotypes**. *Cell* (2018) **173** 1705-1715. PMID: 29906448
33. Huckins L. M., Chatzinakos C., Breen M. S., Hartmann J., Klengel T., da Silva Almeida A. C.. **Analysis of genetically regulated gene expression identifies a prefrontal PTSD gene, SNRNP35,specific to military cohorts**. *Cell reports* (2020) **31** 107716. PMID: 32492425
34. Maihofer A. X., Choi K. W., Coleman J. R., Daskalakis N. P., Denckla C. A., Ketema E., Ursano R. J.. **Enhancing discovery of genetic variants for posttraumatic stress disorder through integration of quantitative phenotypes and trauma exposure information**. *Biological psychiatry* (2022) **91** 626-636. PMID: 34865855
35. Xie P., Kranzler H. R., Yang C., Zhao H., Farrer L. A., Gelernter J.. **Genome-wide association study identifies new susceptibility loci for posttraumatic stress disorder**. *Biological psychiatry* (2013) **74** 656-663. PMID: 23726511
36. Thyme S. B., Pieper L. M., Li E. H., Pandey S., Wang Y., Morris N. S., Schier A. F.. **Phenotypic landscape of schizophrenia-associated genes defines candidates and their shared functions**. *Cell* (2019) **177** 478-491. PMID: 30929901
37. Sanchez-Roige S., Fontanillas P., Elson S. L., Gray J. C., de Wit H., MacKillop J., Palmer A. A.. **Genome-wide association studies of impulsive personality traits (BIS-11 and UPPS-P) and drug experimentation in up to 22,861 adult research participants identify loci in the CACNA1I and CADM2 genes**. *Journal of Neuroscience* (2019) **39** 2562-2572. PMID: 30718321
38. Vorstman J. A., van Daalen E., Jalali G. R., Schmidt E. R., Pasterkamp R. J., de Jonge M., Ophoff R. A.. **A double hit implicates DIAPH3 as an autism risk gene**. *Molecular psychiatry* (2011) **16** 442-451. PMID: 20308993
39. Santarelli R.. **Information from cochlear potentials and genetic mutations helps localize the lesion site in auditory neuropathy**. *Genome Medicine* (2010) **2** 1-10. PMID: 20193046
40. Plomin R.. **Commentary: Missing heritability, polygenic scores, and gene–environment correlation**. *Journal of Child Psychology and Psychiatry* (2013) **54** 1147-1149. PMID: 24007418
41. Plomin R., Deary I. J.. **Genetics and intelligence differences: five special findings**. *Molecular psychiatry* (2015) **20** 98-108. PMID: 25224258
42. George M., Maheshwari S., Chandran S., Manohar J. S., Rao T. S.. **Understanding the schizophrenia prodrome**. *Indian Journal of Psychiatry* (2017) **59** 505. PMID: 29497198
43. Cornblatt B. A., Auther A. M., Niendam T., Smith C. W., Zinberg J., Bearden C. E., Cannon T. D.. **Preliminary findings for two new measures of social and role functioning in the prodromal phase of schizophrenia**. *Schizophrenia bulletin* (2007) **33** 688-702. PMID: 17440198
44. Gibson G.. **Rare and common variants: twenty arguments**. *Nature Reviews Genetics* (2012) **13** 135-145
45. Okbay A., Wu Y., Wang N., Jayashankar H., Bennett M., Nehzati S. M., Young A. I.. **Polygenic prediction of educational attainment within and between families from genome-wide association analyses in 3 million individuals**. *Nature genetics* (2022) **54** 437-449. PMID: 35361970
|
---
title: 'Protocol to evaluate the effectiveness and cost-effectiveness of an environmental
nutrition and physical activity intervention in nurseries (Nutrition and Physical
Activity Self Assessment for Child Care - NAP SACC UK): a multicentre cluster randomised
controlled trial'
authors:
- Ruth Kipping
- Miranda Pallan
- Kim Hannam
- Kate Willis
- Alex Dobell
- Chris Metcalfe
- Russell Jago
- Laura Johnson
- Rebecca Langford
- Corby K Martin
- William Hollingworth
- Madeleine Cochrane
- James White
- Pete Blair
- Zoi Toumpakari
- Jodi Taylor
- Dianne Ward
- Laurence Moore
- Tom Reid
- Megan Pardoe
- Liping Wen
- Marie Murphy
- Anne Martin
- Stephanie Chambers
- Sharon Anne Simpson
journal: Research Square
year: 2023
pmcid: PMC10002848
doi: 10.21203/rs.3.rs-2370293/v1
license: CC BY 4.0
---
# Protocol to evaluate the effectiveness and cost-effectiveness of an environmental nutrition and physical activity intervention in nurseries (Nutrition and Physical Activity Self Assessment for Child Care - NAP SACC UK): a multicentre cluster randomised controlled trial
## Abstract
### Background
One in seven UK children have obesity when starting school, with higher prevalence associated with deprivation. Most pre-school children do not meet UK recommendations for physical activity and nutrition. Formal childcare settings provide opportunities to deliver interventions to improve nutritional quality and physical activity to the majority of 3–4-year-olds. The nutrition and physical activity self-assessment for childcare (NAP SACC) intervention has demonstrated effectiveness in the USA with high acceptability in the UK. The study aims to evaluate the effectiveness and cost-effectiveness of the NAP SACC UK intervention to increase physical activity, reduce sedentary time and improve nutritional intake.
### Methods
Multi-centre cluster RCT with process and economic evaluation. Participants are children aged 2 years or over, attending UK early years settings (nurseries) for ≥ 12 hours/week or ≥ 15 hours/week during term time and their parents, and staff at participating nurseries. The 12-month intervention involves nursery managers working with a Partner (public health practitioner) to self-assess policies and practices relating to physical activity and nutrition; nursery staff attending one physical activity and one nutrition training workshop and setting goals to be achieved within six months. The Partner provides support and reviews progress. Nursery staff receive a further workshop and new goals are set, with Partner support for a further six months. The comparator is usual practice. Up to 56 nurseries will be stratified by area and randomly allocated to intervention or comparator arm with minimisation of differences in level of deprivation. Primary outcomes: accelerometer-assessed mean total activity time on nursery days and average total energy (kcal) intake per eating occasion of lunch and morning/afternoon snacks consumed within nurseries. Secondary outcomes: accelerometer-assessed mean daily minutes of moderate-to-vigorous physical activity and sedentary time per nursery day, total physical activity on nursery days compared to non-nursery days, average serving size of lunch and morning/afternoon snacks in nursery per day, average percentage of core and non-core food in lunch and morning/afternoon snacks, zBMI, proportion of children who are overweight/obese and child quality-of-life. A process evaluation will examine fidelity, acceptability, sustainability and context. An economic evaluation will compare costs and consequences from the perspective of the local government, nursery and parents.
### Trial registration:
ISRCTN33134697
## Background
Childhood obesity is one of the major public health challenges of the 21st century and has been exacerbated by the COVID-19 pandemic. In England and Scotland around $14\%$ of children [1, 2] are living with obesity when they start primary school (age 4–5 years), with the highest rates of obesity in the most deprived areas. This is an increase of nearly $5\%$ from pre-pandemic levels [1]. Consequently, there is a need for effective obesity prevention approaches to target the pre-school age group, that can be delivered at scale.
Diet and physical activity (PA) are key behavioural targets for obesity prevention interventions. PA in childhood is associated with lower levels of cardio-metabolic risk factors and improved psychological well-being [3], and PA intervention studies in young children have consistently reported improved motor and cognitive development, and psychosocial and cardiometabolic health [4]. PA behaviours track from childhood into adulthood, thus positive patterns should be established in the early years of life [5]. The UK Chief Medical Officer recommends that children aged 1–5 years should be physically active for at least three hours per day, with one hour of this being moderate-to-vigorous PA (MVPA) [6]. In 2015, fewer than $9\%$ of 2–4 year olds in England were meeting the PA guidelines and $83\%$ of children had ‘low’ activity levels [7]. This signals a need for improvement in children’s PA levels.
The amount of PA in which a young child engages is influenced by the activity undertaken at pre-school and particularly the time spent outside [8]. A UK study reported that children aged three to four years, spend more time in MVPA and are less sedentary in pre-school and nursery settings compared with time spent at home [9]. However, the time spent in MVPA is still low, with one UK study reporting only $9\%$ of children’s time at preschool being spent in MVPA [10]. The lack of MVPA in pre-school settings may be influenced by space and equipment, policies (including scheduled times for free/outdoor play), and staff training in PA promotion [11].
Dietary patterns established during childhood influence those in later life [12, 13]. Obesogenic dietary patterns, characterized by low intake of fruits, vegetables, high-fibre breakfast cereals (core foods) [14] and high intake of chocolate, confectionery, low-fibre bread, biscuits and cakes (non-core foods), have been observed in young children and are associated with a higher risk of adiposity in later childhood [15, 16]. Food provided in early years settings potentially contributes to these obesogenic dietary patterns, with non-core foods contributing to around $40\%$ of energy intake in younger children (aged 1–3 years) when in attendance at these settings [17]. In terms of nutrient intake, only $15\%$ of younger children meet the UK recommendation of free sugars contributing to no more than $5\%$ of total energy intake, and this falls to $2\%$ in the 4-10-year age-group. For fibre intake, only $12\%$ of the younger children and $14\%$ of the older children meet the recommended intake [18]. In addition to diet quality, meal size is a critical driver of weight gain. In children aged 2–5 years, every extra 10 kcal consumed per meal are associated with a $7\%$ faster rate of weight gain [19]. A survey 10 years ago of early years settings in the UK found those in the most deprived areas reported serving more healthy food (whole grains, legumes, pulses, and lentils), however many were not meeting national guidelines [20]. There are no data on portion size of servings to children in early years settings in the UK.
Early-years settings provide scalable opportunities to deliver diet and PA interventions at the population level [21] and are considered important environments for early intervention to establish positive health behaviours and prevent childhood obesity [22]. In England and Scotland, $94\%$ and $97\%$ of 3–4 year-old children respectively, access the Government-funded early-years education [23, 24] and a large proportion of children under 5-years also attend private- or voluntary-funded childcare organisations (nurseries) beyond this funded provision [25], presenting an opportunity to positively influence eating and PA. However, currently there is limited guidance for PA and nutrition in early-years settings. UK voluntary guidelines exist for food in these settings [21, 26], but in contrast to school settings, where national school food standards legislation [27] applies, there are no statutory standards for food provision. Similarly, schools have a statutory requirement to provide physical education [28], but there are no PA-related requirements in early years settings.
## Physical activity and nutrition interventions in early years settings
Several syntheses of obesity prevention, PA, and nutrition intervention studies in children under 5 years of age provide some evidence of the effectiveness of these interventions, including those delivered in early years settings [29-33]. These evidence syntheses have identified a clear need for more robust research in this area, including evaluation of interventions with explicit theoretical underpinning and the inclusion of economic evaluation. A systematic review by Larson et al. [ 30] explored policies, practices and interventions for promoting healthy eating and PA in childcare settings and concluded that there was little focus on promoting these behaviours. They emphasised the opportunity for interventions and policies to support healthy eating and PA in these settings.
The 2019 Cochrane review of interventions for preventing obesity in children [34], which included 22 studies of interventions in childcare settings, highlighted the need for a better understanding of intervention implementation and collection of data to allow exploration of the impact of intervention on health inequalities. The latter is particularly important as obesity prevalence in UK children is more than double in deprived areas compared with the most affluent areas [1]. This also signals the need for early-years settings interventions to be designed to particularly benefit low-income families and other groups with poorer nutrition, lower physical activity and higher obesity prevalence.
In the UK, there have been few randomised-controlled trials (RCT) of early-years setting interventions targeting diet, physical activity and obesity [35-37], with mixed findings. One cluster-RCT of an early-years educational intervention showed a small reduction in BMI standard deviation score (zBMI) in the intervention, compared with the control group [35]. Another cluster-RCT evaluating a PA intervention in early-years settings reported no difference in physical activity or BMI between intervention and control groups; the authors suggested the intervention was probably of inadequate dose [37].
## The Nutrition and Physical Activity Self Assessment for Child Care (NAP SACC) programme
The NAP SACC programme [38] was developed in the USA and is an early-years setting intervention which aims to improve policies, practices, and the nutrition and PA environment, through a process of self-assessment and targeted assistance. NAP SACC is informed by social cognitive theory (SCT), which identifies the interrelationship between the environment, people, and behaviour [39], within a socio-ecological framework, which identifies multiple, interdependent influences at policy, community, organisational, interpersonal, and intrapersonal levels [40, 41]. The goals of the NAP SACC programme are to improve the nutritional quality of food served, the amount and quality of PA, staff-child interactions, and nutrition and PA policies [38].
Several RCTs of the NAP SACC programme in the USA have demonstrated its feasibility, acceptability and effectiveness, reporting: improvements in environmental audit nutrition scores [42]; increases in staff knowledge of childhood obesity, healthy eating, personal health, and working with families; decreases in children’s zBMI (−0.14; $95\%$ CI – 0.26,−0.02) [43]; and increased accelerometer-measured PA by $17\%$ [44], following delivery of the NAP SACC programme. No studies included an economic evaluation or assessed dietary intake as an outcome.
## Feasibility of the NAP SACC UK intervention
In partnership with stakeholders, the NAP SACC programme was adapted for use in the UK (NAP SACC UK, details of which are published elsewhere [45]) and a feasibility cluster-RCT conducted with 168 children aged 2–4 years in 12 nurseries in North Somerset and Gloucestershire, England in 2015 to 2016 [45-47]. Overall, the NAP SACC UK intervention was delivered as planned, except for the home component (designed to support parents with their child’s PA and nutrition which was found not to be feasible), and the trial methods and design were found to be acceptable and feasible.
Post intervention (8–10 months after baseline), total activity in nursery settings was higher in the intervention nurseries compared with the control nurseries by 18.7 minutes/day ($95\%$ CI 3.8, 41.3) using the (underpowered) complete-case multi-level linear regression model adjusted for baseline outcome, age, gender and average hours of attendance. Evidence was less clear for improvements in anthropometry and dietary practices in the nurseries; however, the nursery managers reported improvements in several areas of feeding practice.
Given these promising outcomes of the NAP SACC UK feasibility study, together with the limited UK-based evidence for early-years settings interventions targeting diet and physical activity (particularly those targeting the early-years settings environment), there is a need to conduct a more definitive evaluation of the NAP SACC UK intervention. In response to the feasibility study findings, the intervention has been further refined by adapting the timings of intervention processes, extending the intervention period to one year (to include two Review and Reflect and goal setting cycles and a top-up day workshop), including lunchboxes in the Review and Reflect and workshop content and allowing for flexibility around the types of Local Authority staff who can deliver the intervention. Additionally, to improve dietary assessment, detailed data on food served and consumed using food photography will be captured to improve the assessment of diet. This will provide evidence for effectiveness and cost-effectiveness of an intervention that targets the eating and PA environments of early-years settings, which could potentially be rolled out at scale in the UK.
## Study aims and objectives
The aim of the trial is to evaluate the effectiveness and cost-effectiveness of the NAP SACC UK intervention to increase physical activity and diet quality, while reducing sedentary time and portion size to nationally recommended levels, using a cluster RCT design with embedded process and economic evaluations. The trial will take place in early years settings (referred to throughout as “nurseries”). The co-primary objectives are to determine whether the NAP SACC UK intervention at 12 months: a) increases mean accelerometer-measured total physical activity on nursery days compared with usual practice b) reduces the energy (kcal) per eating occasion averaged across snack and lunch eating occasions that occur within nurseries compared with usual practice, within Nationally recommended levels.
The secondary objectives are to determine whether the NAP SACC UK intervention, compared with usual practice at 12 months:
## Study design
The NAP SACC UK study is a multicentre, parallel-group, two-arm, cluster RCT with a repeat cross-sectional design. Clusters (nurseries) will be randomised to receive either the one-year NAP SACC UK intervention or continue with usual practice. The effectiveness and cost-effectiveness of NAP SACC UK will be assessed immediately after the one-year intervention. Figure 1 provides a study overview.
Separate cross-sectional samples of children attending the participating nurseries will be taken at two time points. The first will be prior to randomisation (T0) and the second, immediately after the intervention (T1). NAP SACC UK is designed to be an environmental intervention influencing the whole nursery which will impact all children and not just those present at T0. Due to the movement of children during the intervention period, the children at T0 may not be representative of the cluster, thus the repeat cross-sectional design may minimise bias and has been used in previous studies [48].
## Setting and participants
The trial will include up to 56 nurseries from four areas across England and Scotland: Ayrshire and Arran (Scotland), Sandwell, Somerset and Swindon (England). These areas were selected to ensure a broad range of deprivation status and ethnicity (non-white population varying from $2\%$ in Somerset to $30\%$ in Sandwell [UK 2011 census]) to enable exploration of generalisability. The research will be managed from three University ‘Hubs’ local to the study areas (University of Bristol, University of Glasgow and University of Birmingham).
Nurseries within the four study areas are eligible to participate if they are: day nurseries, private nursery schools, maintained nurseries (including nurseries within Children’s Centres), nursery classes attached to primary schools and pre-schools where children consume lunch (provided by the nursery or family). Nurseries will be excluded from participating if they are: childminders, creches, playgroups, primary school reception classes, solely outdoor nursery settings, solely Special Educational Needs and Disabilities (SEND) nursery settings, au pairs, or settings taking part in a research study or other initiative that would interfere with the NAP SACC UK study.
Participants within each nursery setting include nursery staff (managers and childcare staff), parent/carers and children. Child (and associated parent/carer participants) inclusion criteria are: aged 2 years or over at the time of assessment, not yet attending Reception (England) or Primary One (Scotland), attending the participating nurseries for a minimum of 12 hours/week across the year or 15 hours/week during term time, and consuming at least lunch within the setting.
## Recruitment of nurseries
The study aims to recruit nurseries across the Index of Multiple Deprivation (IMD) or Scottish Index of Multiple Deprivation (SIMD). Within each of the four sites (S)IMD scores will be assigned to all nurseries and invitations sent across the range. An email will be sent to potentially eligible settings which will include a short informative film about the study, study summary and participant information sheet. The email will be followed up by a telephone call from the research staff to the nursery manager to undertake a screening check for eligibility and offer a meeting to discuss further study details. If eligible nursery managers decide to participate in the study, they will be provided with a consent form and letter of agreement to sign. Up to 56 nurseries will be recruited with the aim of a balance across the four sites. All nursery staff from recruited settings who work directly with children aged two years and over will be invited to participate.
## Recruitment of parent/carers and children
In selected and consented nurseries, all parent/carers of children aged 2-years or over will be informed about the study and invited to participate if their child attends for the minimum eligible hours. Parents will have the opportunity to review study documents as hard copies or online and view a short informative online film. Opt-in consent will be obtained for parents as participants, as well as on behalf of their child. The research team will be available by email or telephone to answer questions and, if appropriate, in person at a convenient time for the nursery.
## Random allocation
Each nursery will be randomly allocated to the NAP SACC UK intervention or usual practice comparator group once all T0 data have been collected from the children, parents and staff at the nursery. Allocation will be conducted by a statistician, blind to the identity of nurseries and otherwise uninvolved in the ongoing study.
Within each hub separately, the allocation of nurseries will be conducted to minimise differences on an average IMD score (created for each nursery using the postcodes of the children recruited) at each site. Each random allocation will attempt to balance the IMD score between the two groups per site. This minimisation procedure will be written in Stata and code included in the Statistical Analysis Plan.
## Blinding
Two statisticians and two health economists will support this trial. The senior statistician and health economist will be blinded throughout the trial and will not have access to any identifying data. A study statistician and a health economist will perform all disaggregated analyses according to a pre-specified statistical analysis plan and health economic analysis plan, respectively. In addition, the study statistician will attend Trial Steering Committee meetings as required and prepare all interim reports, e.g., on recruitment and data completeness. The remaining members of the study team will remain blinded to aggregate data only.
## Sample size
Our aim is to recruit up to 56 nurseries (784 children), allowing for two nurseries withdrawing from the study and an average of 14 children per nursery, allowing for up to $35\%$ failing to provide valid accelerometer data on nursery days. Assuming nine children will provide valid primary outcome data at each nursery, 27 nurseries in each of the intervention and control arms will provide $90\%$ power at the $5\%$ significance level to detect a 17-minute difference (0.4 standard deviations) in total daily physical activity on nursery days. In the absence of a good estimate of the variation in mean total activity per day between nurseries, we allowed for variation up to a magnitude corresponding to an intra-cluster correlation of 0.087. The coefficient of variation of cluster size is 0.3 to account for slightly variable cluster size, i.e., different numbers of children in nurseries. As our measure of nutrition is on a continuous scale, a trial of 56 nurseries will be able to detect a 0.4 standard deviation difference in kcal, under the same assumptions. From our feasibility data, this is about 45kcal which equates to approximately half a banana or half a cup of milk.
## Comparator group provision
All consenting nurseries, nursery staff, children and parents/carers will participate in the baseline (T0) and follow-up (T1) data collection. However, the control nurseries and participants within them, will not receive the NAP SACC UK intervention and will continue practice as usual. Data will not be collected from control nurseries for the economic analysis.
## Data collection
Table 1 outlines the data collection time points at the nursery, child and parent level. Details of data collection methods for PA, diet and anthropometric data are also listed here.
## Physical activity
Accelerometers will be worn for five week days.
## Diet data
Morning/afternoon snack and lunch diet data will be collected in nursery from consented children using the Remote Food Photography Method (RFPM) to give direct estimation of portion sizes of foods and drinks through visual comparison of photographs to standard portion size photographs. [ 49, 50] A researcher will take one photograph of a child’s eating occasion before consumption and one photo after. Photos are taken at a 45-degree angle and approximately at arm’s-length distance from the plate, including all the food on the plate. If a child receives additional portions of food, a photograph will be taken of the plate before the serving is added and again after to capture a photograph of the new portion. A final photo is then taken when the child finishes eating to capture any leftovers or confirm the entire serving was consumed. The photographs will be annotated to assist with food identification or portion size estimation where relevant and submitted for analysis to Pennington Biomedical Research Centre (PBRC) [51]. The photos will be processed to estimate nutrient composition of foods consumed. In the absence of the meal being weighed or specific recipe being provided from the nursery, standard recipes or volumetric portion sizes will be used based on the National Diet and Nutrition Survey (NDNS) nutrient databank [52].
## Anthropometric measures
All anthropometric measurements will be completed with children in a private area with a member of nursery staff. Weight will be measured without shoes in light clothing to the nearest 0.1kg using a calibrated medical grade digital scale. Height will be measured to the nearest 0.1cm, without shoes, using a portable stadiometer. Fieldworkers will be trained to ensure correct position for height assessment.
## Process evaluation
Two elements have been identified as critical to successful implementation: 1) the valued relationship formed between the nursery manager and NAP SACC UK Partner and 2) the motivation and “buy in” created among nursery staff at the workshops [47]. Local Authorities (or the NHS Board in Scotland) have identified relevant health or health improvement staff to take on the roles of Partners to replicate what is likely to happen in any future implementation. However, each group of staff will be trained to the same specifications. The process evaluation will explore this variation to understand its impact on how the intervention was implemented and received. The process evaluation will explore the following components: The process evaluation will use a combination of methods to collect detailed information to contextualise the results of the trial and inform any potential roll-out plans should the intervention prove effective:
## Economic evaluation
Data will be collected to capture the costs and consequences from the perspective of the local government, nurseries and parents. The primary economic outcome will be child health-related quality of life, measured at baseline and 12 month follow-up using the parent-reported PedsQL for 2–4 year olds. The developer of the PedsQL measurement tool advised that this measure would be suitable for parents in our study who are completing the measure for a 5 year old child. Versions of the PedsQL for older children include questions about school which would not be relevant to our participants. We will capture costs related to each perspective using various methods, including: Our feasibility study indicated insufficient value in collecting information from parents on their children’s use of healthcare during the intervention to justify the burden. Healthcare use in this generally healthy population was infrequent and believed to be very unlikely to be causally related to the intervention.
## Participant appreciation:
To thank all participating nursery schools, in both the intervention and control arms, we will provide £300 on completion of the study, along with a summary of results at nursery level. Following data collection, participating children will receive a small token of thanks in the form of a sticker and a children’s book; parents will receive a £10 voucher on return of their child’s accelerometer.
## Intervention
Table 2 outlines the detail of the intervention using the TIDieR reporting guidance as a framework. Local Authorities have chosen the most appropriate locally employed staff to deliver the intervention (public health practitioners with expertise in public health, nutrition or physical activity who are referred to as NAP SACC UK Partners), to enable us to test the effectiveness of the intervention as it might be delivered outside a trial.
## Quantitative analysis:
Valid accelerometer data will be at least two days of data worn for at least 6 hours per day on nursery days. Periods of 60-minutes with zero values will be interpreted as time that the monitor is not worn. A day will be considered valid if ≥ 6 hours of data are recorded on a day when the child attended nursery. Days on which children are absent from nursery or spend < $50\%$ of the nursery day in the childcare setting will be excluded from the analyses, informed by the methodology used by Pate et al. [ 53]. Children with ≥ 2 nursery days of accelerometer data will be included in the analyses. Mean minutes of sedentary time (using two thresholds of 0–25 and 0–199 counts per 15 seconds using the criteria proposed by Evenson and Puyau [54, 55]) will be used and mean minutes of light, moderate to vigorous intensity physical activity will be processed (thresholds of 200–799; and > = 800 counts per 15 seconds).[56] Mean accelerometer counts per minute, which provides an indication of the overall volume of physical activity in which the children engage will also be calculated as this approach facilitates comparison with studies that may have applied a different cutpoint. The accelerometer data will be checked for outliers. Informed by previous studies with children we will exclude implausibly high values, such as might occur when a participant uses a trampoline, using a cap of 11,714 counts per minute (cpm)[57].
Total eating occasion size (kcal per occasion) will be computed from the sum of energy in each portion food or drink consumed for each snack (morning or afternoon) or lunch consumed in nursery. The average total size of eating occasions consumed within nursery for each child will then be derived (primary outcome). Specific foods will also be classified as core or non-core and the total intake (kcal) of core and non-core foods will be separately summed in each eating occasion consumed at nursery and expressed as a percentage of total energy consumed in an eating occasion for each child. [ 14] To represent the balance of healthy to less-healthy food intake consumed, the average percentage of core and non-core food in lunch and morning/afternoon snacks consumed by each child will be calculated.
The primary and secondary analyses will be pre-specified in a statistical analysis plan which will be written whilst blinded to the accumulating outcome data and will be made publicly available before the conclusion of the follow-up period. The evidence for an overall intervention effect on the primary outcomes (physical activity and nutrition) will be estimated using a multilevel linear regression model, which will include the following nursery level covariates: intervention group, IMD as used to stratify the allocation, local authority, and a random effect to accommodate variation between nurseries (clustering). The intervention effects will be presented as differences in average total activity and total energy consumed per eating occasion with their $95\%$ confidence intervals. The exact specification of the primary analysis will be informed by an inspection of the baseline measurements of the primary outcomes.
The primary analysis approach will be adapted to estimate the intervention effect on each of the secondary outcomes, utilizing univariate multilevel linear regression (continuous outcome measures) or univariate multilevel logistic regression (binary outcome measures).
We will examine whether the intervention effect on the two primary outcome measures varies by sub-groups of participants. These sub-groups will be pre-specified in the statistical analysis plan and may include parental employment status, geographical area, child’s gender and time spent in nursery.
Sensitivity analyses will repeat the primary analysis with variations to the method that include the following (i) additional covariates where one or more measures was found to be unbalanced at baseline; (ii) missing data imputed under different assumptions about the mechanisms leading to those data being missing; and (iii) excluding outliers if inspection reveals that outliers are genuine.
## Qualitative analysis:
Information collected from document analysis, questionnaires and structured elements of the training sessions/workshops observations will be entered into the REDCap data management system or an Excel file. Interview transcripts, qualitative observations and fieldnotes will be uploaded into NVivo 12 to aid data management and analysis and analysed to identify key themes. An initial coding framework will be developed by two staff including both deductive codes derived from research questions and inductive codes identified from initial readings of early transcripts. This framework will be independently applied to two to four further transcripts depending on the consistency of coding; any discrepancies in coding will be discussed and appropriate revisions made. The final framework will be applied to all subsequent transcripts with any additions or revisions recorded. We will triangulate between different process evaluation data sources (observations, questionnaires, documentary data and interviews) to identify confirmatory or contradictory results. For example, we will compare data from the observations of training workshops with the staff evaluation forms and comments from nursery manager and/or Partner interviews to understand how the workshops were received and their importance within the intervention as a whole.
## Health economic analysis:
The primary economic analysis will consist of a within-trial cost consequences analysis (CCA) from the perspective of the local government, nursery and parents. Results from the within-trial CCA will allow the costs and consequences to be presented clearly in a disaggregated format rather than summarised into a single index. If there is an important difference in physical activity and/or diet at T1, a secondary analysis considering the potential longer-term costs and outcomes of the intervention will be considered. This would include a review of the economic evidence on the medium- and long-term costs and consequences of changes in physical activity and diet in young children.
## Discussion
The trial aims to evaluate the effectiveness and cost-effectiveness of the NAP SACC UK intervention to increase physical activity, reduce sedentary time and improve the quality and quantity of nutritional intake. The trial builds on the success of the feasibility study, the evaluations of the intervention and subsequent adoption across the US. The trial initially started in July 2019 and was paused from March 2020 to January 2022 because of the disruption to research in childcare settings arising from COVID-19. Following discussion with nurseries, collaborators and the funder, we restarted the study in February 2022. Some of the initial outcome measures and processes were refined and this protocol represents the study upon restarting in 2022.
If the NAP SACC UK intervention is found to be effective, this will have important policy and practice implications for the commissioning of programmes to prevent obesity, improve physical activity and nutrition in early years settings in the UK. We also have the scope to apply for additional funding to explore longer term impacts on z-BMI, height and weight beyond the early years’ settings. The trial is deliberately pragmatic in the use of public health staff or staff from commissioned services who work on physical activity and nutrition beyond health visitors as used in the feasibility trial.
## Funding
The NAP SACC UK study is funded by the UK National Institute for Health and Care Research (NIHR) Public Health Research Programme ($\frac{12}{75}$/51). This study was designed and delivered in collaboration with the Bristol Trials Centre (BTC), a UKCRC Registered Clinical Trials Unit which is in receipt of NIHR CTU Support Funding. Pennington Biomedical’s Remote Food Photography Method© was utilised during the study, and Pennington *Biomedical is* supported by NORC Center Grant P30 DK072476 entitled “Nutrition and Metabolic Health Through the Lifespan” sponsored by the National Institute of Diabetes and Digestive and Kidney Disease and by grant U54 GM104940 from the National Institute of General Medical Sciences, which funds the Louisiana Clinical and Translational Science Center. None of the funders nor the study sponsor had involvement in the Trial Steering Committee, the collection, analysis, or interpretation of data or writing of the paper. Intervention costs have been funded by the Office for Health Improvement and Disparities (OHID), Sandwell Metropolitan Borough Council, Swindon Borough Council and Ayrshire and Arran Council. The views and opinions expressed in this publication are those of the authors and do not necessarily reflect those of the funding bodies listed here.
## References
1. 1.Population Health Team ND. National Child Measurement Programme, England 2020/21 School Year. NHS Digital, Government Statistical Service; 2021.. *National Child Measurement Programme, England 2020/21 School Year* (2021.0)
2. 2.Government S. Scottish Health Survey 2019 - volume 1: Main report 2020 [Available from: http://www.gov.scot/publications/scottish-health-survey-2019-volume-1-main-report/.. *Scottish Health Survey 2019 - volume 1: Main report 2020*
3. Strong WB, Malina RM, Blimkie CJ, Daniels SR, Dishman RK, Gutin B. **Evidence based physical activity for school-age youth**. *J Pediatr* (2005.0) **146** 732-7. PMID: 15973308
4. Carson V, Lee EY, Hewitt L, Jennings C, Hunter S, Kuzik N. **Systematic review of the relationships between physical activity and health indicators in the early years (0-4 years) (vol 17, 854, 2017)**. *Bmc Public Health* (2017.0) 17
5. Hayes G, Dowd KP, MacDonncha C, Donnelly AE. **Tracking of Physical Activity and Sedentary Behavior From Adolescence to Young Adulthood: A Systematic Literature Review**. *J Adolesc Health* (2019.0) **65** 446-54. PMID: 31248803
6. 6.Officers UCM. UK Chief Medical Officers’ Physical Activity Guidelines. In: Care DoHaS, editor. 2019.. *UK Chief Medical Officers’ Physical Activity Guidelines* (2019.0)
7. Digital N. *Health Survey for England England2016*
8. Pate RR, McIver K, Dowda M, Brown WH, Addy C. **Directly observed physical activity levels in preschool children**. *J Sch Health* (2008.0) **78** 438-44. PMID: 18651931
9. Hesketh KR, Griffin SJ, van Sluijs EMF. **UK Preschool-aged children's physical activity levels in childcare and at home: a cross-sectional exploration**. *Int J Behav Nutr Phy* (2015.0) **12**
10. Dobell AP, Eyre ELJ, Tallis J, Chinapaw MJM, Altenburg TM, Duncan MJ. **Examining accelerometer validity for estimating physical activity in pre-schoolers during free-living activity**. *Scand J Med Sci Sports* (2019.0) **29** 1618-28. PMID: 31206785
11. Copeland KA, Khoury JC, Kalkwarf HJ. **Child Care Center Characteristics Associated With Preschoolers' Physical Activity**. *Am J Prev Med* (2016.0) **50** 470-9. PMID: 26585052
12. Luque V, Escribano J, Closa-Monasterolo R, Zaragoza-Jordana M, Ferre N, Grote V. **Unhealthy Dietary Patterns Established in Infancy Track to Mid-Childhood: The EU Childhood Obesity Project**. *J Nutr* (2018.0) **148** 752-9. PMID: 29982656
13. Ventura AK, Worobey J. **Early influences on the development of food preferences**. *Curr Biol* (2013.0) **23** R401-8. PMID: 23660363
14. Toumpakari Z, Haase AM, Johnson L. **Adolescents' non-core food intake: a description of what, where and with whom adolescents consume non-core foods**. *Public Health Nutr* (2016.0) **19** 1645-53. PMID: 26878965
15. Johnson L, Mander AR Jones LR, Emmett PM, Jebb SA. **Energy-dense, low-fiber, high-fat dietary pattern is associated with increased fatness in childhood**. *Am J Clin Nutr* (2008.0) **87** 846-54. PMID: 18400706
16. Sirkka O, Fleischmann M, Abrahamse-Berkeveld M, Halberstadt J, Olthof MR, Seidell JC. **Dietary Patterns in Early Childhood and the Risk of Childhood Overweight: The GECKO Drenthe Birth Cohort**. *Nutrients* (2021.0) **13**
17. Ziauddeen N, Page P, Penney TL, Nicholson S, Kirk SFL, Almiron-Roig E. **Eating at food outlets and leisure places and "on the go" is associated with less-healthy food choices than eating at home and in school in children: cross-sectional data from the UK National Diet and Nutrition Survey Rolling Program (2008-2014)**. *Am J Clin Nutr* (2018.0) **107** 992-1003. PMID: 29741556
18. England PH. *National Diet and Nutrition Survey Rolling programme Years 9 to 11 (2016/2017 to 2018/2019)* (2020.0)
19. Syrad H, Llewellyn CH, Johnson L, Boniface D, Jebb SA, van Jaarsveld CHM. **Meal size is a critical driver of weight gain in early childhood**. *Sci Rep-Uk* (2016.0) **6**
20. Neelon SE, Burgoine T, Hesketh KR, Monsivais P. **Nutrition practices of nurseries in England. Comparison with national guidelines**. *Appetite* (2015.0) **85** 22-9. PMID: 25450898
21. 21.Children Af. Eat Better Start Better: Voluntary Food and Drink Guidelines for Early Years Settings in England - A Practical Guide. London; 2017.. *Eat Better Start Better: Voluntary Food and Drink Guidelines for Early Years Settings in England - A Practical Guide* (2017.0)
22. Maher EJ, Li GH, Carter L, Johnson DB. **Preschool child care participation and obesity at the start of kindergarten**. *Pediatrics* (2008.0) **122** 322-30. PMID: 18676550
23. Albakri M, Basi T, Forsyth E., Patel R., Skipp A., Tanner E. *Take-up of free early education entitlements* (2018.0) 97
24. 24.Government S. Chapter 5: Early Learning and Childcare. 2021 [Available from: https://www.gov.scot/publications/summary-statistics-schools-scotland/pages/6/.. *Government S. Chapter 5: Early Learning and Childcare* (2021.0)
25. 25.Education Df. Survey of Childcare and Early Years Providers 2021.. 2021.. *Survey of Childcare and Early Years Providers 2021* (2021.0)
26. Scotland NH. *Setting the table* (2018.0) 143
27. 27.Education TSoSf. The Requirements for School Food Regulations 2014 (legislation.gov.uk). London: Uk Government; 2014.. *The Requirements for School Food Regulations 2014 (* (2014.0)
28. 28.Education Df. National curriculum in England: physical education programmes of study. In: Education Do, editor. London: Department of Education; 2013.. *Education Do* (2013.0)
29. Hesketh KD, Campbell KJ. **Interventions to prevent obesity in 0-5 year olds: an updated systematic review of the literature**. *Obesity (Silver Spring)* (2010.0) **18** S27-35. PMID: 20107458
30. Larson N, Ward DS, Neelon SB, Story M. **What role can child-care settings play in obesity prevention? A review of the evidence and call for research efforts**. *J Am Diet Assoc* (2011.0) **111** 1343-62. PMID: 21872698
31. Narzisi K, Simons J. **Interventions that prevent or reduce obesity in children from birth to five years of age: A systematic review**. *J Child Health Care* (2021.0) **25** 320-34. PMID: 32295414
32. Ward DS, Vaughn A, McWilliams C, Hales D. **Interventions for increasing physical activity at child care**. *Med Sci Sports Exerc* (2010.0) **42** 526-34. PMID: 20068495
33. Waters E, de Silva-Sanigorski A, Hall BJ, Brown T, Campbell KJ, Gao Y. **Interventions for preventing obesity in children**. *Cochrane Database Syst Rev* (2011.0) CD001871. PMID: 22161367
34. Brown T, Moore TH, Hooper L, Gao Y, Zayegh A, Ijaz S. **Interventions for preventing obesity in children**. *Cochrane Database Syst Rev* (2019.0) **7** CD001871. PMID: 31332776
35. Hodgkinson A, Abbott J, Hurley MA, Lowe N, Qualter P. **An educational intervention to prevent overweight in pre-school years: a cluster randomised trial with a focus on disadvantaged families**. *Bmc Public Health* (2019.0) **19** 1430. PMID: 31675942
36. Malden S, Reilly JJ, Gibson AM, Bardid F, Summerbell C, De Craemer M. **A feasibility cluster randomised controlled trial of a preschool obesity prevention intervention: ToyBox-Scotland**. *Pilot Feasibility Stud* (2019.0) **5** 128. PMID: 31728203
37. Reilly JJ, Kelly L, Montgomery C, Williamson A, Fisher A, McColl JH. **Physical activity to prevent obesity in young children: cluster randomised controlled trial**. *BMJ* (2006.0) **333** 1041. PMID: 17028105
38. Benjamin SE, Ammerman A, Sommers J, Dodds J, Neelon B, Ward DS. **Nutrition and physical activity self-assessment for child care (NAP SACC): results from a pilot intervention**. *J Nutr Educ Behav* (2007.0) **39** 142-9. PMID: 17493564
39. Bandura A. *Social foundations of thought and action: A social cognitive theory* (1986.0)
40. Bronfenbrenner U. *The ecology of human development: Experiments by nature and design* (1979.0)
41. McLeroy KR, Bibeau D, Steckler A, Glanz K. **An ecological perspective on health promotion programs**. *Health Educ Q* (1988.0) **15** 351-77. PMID: 3068205
42. Ward D, Hales D, Haverly K, Marks J, Benjamin S, Ball S. **An instrument to assess the obesogenic environment of child care centers**. *Am J Health Behav* (2008.0) **32** 380-6. PMID: 18092898
43. Alkon A, Crowley AA, Neelon SE, Hill S, Pan Y, Nguyen V. **Nutrition and physical activity randomized control trial in child care centers improves knowledge, policies, and children's body mass index**. *Bmc Public Health* (2014.0) **14** 215. PMID: 24580983
44. Bonis M, Loftin M, Ward D, Tseng TS, Clesi A, Sothern M. **Improving physical activity in daycare interventions**. *Child Obes* (2014.0) **10** 334-41. PMID: 25055027
45. Kipping R, Langford R, Brockman R, Wells S, Metcalfe C, Papadaki A. **Child-care self-assessment to improve physical activity, oral health and nutrition for 2- to 4-year-olds: a feasibility cluster RCT**. *Public Health Research* (2019.0) **7**
46. Kipping R, Jago R, Metcalfe C, White J, Papadaki A, Campbell R. **NAP SACC UK: protocol for a feasibility cluster randomised controlled trial in nurseries and at home to increase physical activity and healthy eating in children aged 2-4 years**. *BMJ Open* (2016.0) **6** e010622
47. Langford R, Jago R, White J, Moore L, Papadaki A, Hollingworth W. **A physical activity, nutrition and oral health intervention in nursery settings: process evaluation of the NAP SACC UK feasibility cluster RCT**. *Bmc Public Health* (2019.0) **19** 865. PMID: 31269926
48. Bonell C, Allen E, Warren E, McGowan J, Bevilacqua L, Jamal F. **Modifying the secondary school environment to reduce bullying and aggression: the INCLUSIVE cluster RCT**. *Public Health Research* (2019.0) **7**
49. Nicklas T, Islam NG, Saab R, Schulin R, Liu Y, Butte NF. **Validity of a Digital Diet Estimation Method for Use with Preschool Children**. *J Acad Nutr Diet* (2018.0) **118** 252-60. PMID: 28637631
50. Nicklas T, Saab R, Islam NG, Wong W, Butte N, Schulin R. **Validity of the Remote Food Photography Method Against Doubly Labeled Water Among Minority Preschoolers**. *Obesity (Silver Spring)* (2017.0) **25** 1633-8. PMID: 28758370
51. Fitt E, Cole D, Ziauddeen N, Pell D, Stickley E, Harvey A. **DINO (Diet In Nutrients Out) - an integrated dietary assessment system**. *Public Health Nutr* (2015.0) **18** 234-41. PMID: 24674815
52. 52.University of Cambridge NSR. National Diet and Nutrition Survey Years 1-11, 2008-2019 [data collection]. 2021.. *National Diet and Nutrition Survey Years 1-11, 2008-2019 [data collection]* (2021.0)
53. Pate RR, Brown WH, Pfeiffer KA, Howie EK, Saunders RP, Addy CL. **An Intervention to Increase Physical Activity in Children: A Randomized Controlled Trial With 4-Year-Olds in Preschools**. *Am J Prev Med* (2016.0) **51** 12-22. PMID: 26803357
54. Evenson KR, Catellier DJ, Gill K, Ondrak KS, McMurray RG. **Calibration of two objective measures of physical activity for children**. *J Sports Sci* (2008.0) **26** 1557-65. PMID: 18949660
55. Puyau MR, Adolph AL, Vohra FA, Butte NF. **Validation and calibration of physical activity monitors in children**. *Obes Res* (2002.0) **10** 150-7. PMID: 11886937
56. van Cauwenberghe E, Labarque V, Trost SG, de Bourdeaudhuij I, Cardon G. **Calibration and comparison of accelerometer cut points in preschool children**. *Int J Pediatr Obes* (2011.0) **6** e582-9. PMID: 21121867
57. Rich C, Geraci M, Griffiths L, Sera F, Dezateux C, Cortina-Borja M. **Quality control methods in accelerometer data processing: identifying extreme counts**. *PLoS One* (2014.0) **9** e85134. PMID: 24454804
|
---
title: Dual Role of Mitogen-Activated Protein Kinase 8 Interacting Protein-1 in Inflammasome
and Pancreatic β-Cell Function
authors:
- Rania Saeed
- Abdul Khader Mohammed
- Sarra E. Saleh
- Mohammad M. Aboulwafa
- Khaled M. Aboshanab
- Jalal Taneera
journal: International Journal of Molecular Sciences
year: 2023
pmcid: PMC10002854
doi: 10.3390/ijms24054990
license: CC BY 4.0
---
# Dual Role of Mitogen-Activated Protein Kinase 8 Interacting Protein-1 in Inflammasome and Pancreatic β-Cell Function
## Abstract
Inflammasomes have been implicated in the pathogenesis of type 2 diabetes (T2D). However, their expression and functional importance in pancreatic β-cells remain largely unknown. Mitogen-activated protein kinase 8 interacting protein-1 (MAPK8IP1) is a scaffold protein that regulates JNK signaling and is involved in various cellular processes. The precise role of MAPK8IP1 in inflammasome activation in β-cells has not been defined. To address this gap in knowledge, we performed a set of bioinformatics, molecular, and functional experiments in human islets and INS-1 ($\frac{832}{13}$) cells. Using RNA-seq expression data, we mapped the expression pattern of proinflammatory and inflammasome-related genes (IRGs) in human pancreatic islets. Expression of MAPK8IP1 in human islets was found to correlate positively with key IRGs, including the NOD-like receptor (NLR) family pyrin domain containing 3 (NLRP3), Gasdermin D (GSDMD) and Apoptosis-associated speck-like protein containing a CARD (ASC), but correlate inversely with Nuclear factor kappa β1 (NF-κβ1), Caspase-1 (CASP-1), Interleukin-18 (IL-18), Interleukin-1β (IL-1β) and Interleukin 6 (IL-6). Ablation of Mapk8ip1 by siRNA in INS-1 cells down-regulated the basal expression levels of Nlrp3, NLR family CARD domain containing 4 (Nlrc4), NLR family CARD domain containing 1 (Nlrp1), Casp1, Gsdmd, Il-1β, Il-18, Il-6, Asc, and Nf-κβ1 at the mRNA and/or protein level and decreased palmitic acid (PA)-induced inflammasome activation. Furthermore, Mapk8ip1-silened cells substantially reduced reactive oxygen species (ROS) generation and apoptosis in palmitic acid-stressed INS-1 cells. Nonetheless, silencing of Mapk8ip1 failed to preserve β-cell function against inflammasome response. Taken together, these findings suggest that MAPK8IP1 is involved in regulating β-cells by multiple pathways.
## 1. Introduction
Type 2 diabetes (T2D) is an autoinflammatory metabolic disease caused by low-grade chronic inflammation due to overabundant nutrients and excessive metabolic stress [1,2,3]. The innate immune system appears to be primarily involved in evoking this metabolic inflammation (i.e., metaflammation) [4]. Inflammasomes are cytosolic multi-protein signaling complexes assembled upon recognition of various physiological and pathological stimuli. Inflammasome assembly triggers downstream signaling pathways that activate caspase-1, subsequently releasing pro-inflammatory cytokines (IL-1β and IL-18) and causing pyroptosis [5].
Pyroptosis is a programmed inflammatory cell death characterized by DNA fragmentation and the formation of pores in the plasma membrane resulting in the release of the cytosolic contents and the reinforcement of the immune response [6]. Although optimal inflammasome activation is favorable to the well-being of the host, aberrant inflammasome signaling can lead to an exaggerated innate immune response and the development of autoimmune and inflammatory disorders [7].
Several mechanistic studies have supported the involvement of inflammasome activation in the pathogenesis of T2D and its complications [8,9,10,11]. For example, a high glucose level was shown to induce NLRP3 [12,13]. Furthermore, NLRP3 and the secreted IL-1β were reported to be associated with insulin resistance [14,15], β-cell dysfunction, and cell death [16,17,18].
Interestingly, the protein expression of NLRP3, ASC, Caspase-1, IL-1β, and IL-18 were up-regulated in newly diagnosed T2D patients [19]. Another line of evidence has shown that inflammasome-mediated pyroptosis plays a key role in the occurrence and evolution of diabetes and its complications [20]. Furthermore, the inhibition or genetic deletion of inflammasome components has been found to improve glucose tolerance and insulin secretion and to reduce islet-cell apoptosis [8,10,18,21]. Thus, targeting inflammasome could be an early preventive strategy for diabetes and its complications [22,23]. However, the expression and function of inflammasome in pancreatic islets are still not well-characterized [24].
MAPK8IP1 (also known as Islet-Brain1 (IB1) protein or c-Jun N-terminal kinase (JNK) interacting protein-1 (JIP1)) is a scaffold protein that is highly expressed in the brain [25] and pancreatic β-cells [26]. There is accumulating evidence that MAPK8IP1 plays an essential role in β-cell survival and function. Although MAPK8IP1 has been identified as a potential candidate gene for T2D [27], other studies have demonstrated that the loss of MAPK8IP1 function did not contribute to the development of diabetes [28,29]. Recently, we demonstrated that MAPK8IP1 expression is reduced in human diabetic islets and that the silencing of Mapk8ip1 in INS-1 cells impaired insulin secretion and reduced glucose uptake levels [30]. Furthermore, MAPK8IP1 has been reported to mediate the JNK signaling pathway [31], and the latter was implicated in inflammasome activation [32,33].
To our knowledge, no studies have investigated the role of MAPK8IP1 in pancreatic β-cell inflammasome regulation. Therefore, in this study, we aimed to explore the functional role of MAPK8IP1 in β-cell inflammasomes activation/regulation and its impact on β-cell survival and function.
## 2.1. Expression Profiles of Pro-Inflammatory and Inflammasome-Related Genes (IRGs) in Human Pancreatic Islets
Herein, we investigated the expression of several IRGs using published RNA-seq and qPCR expression analyses of human pancreatic islets. As is shown in Figure 1A, an RNA-seq expression analysis of 26 IRG genes showed that IL-6, NF-κβ1, MAPK8IP1, and pannexin1 (PANX1) were among the most highly expressed genes in the human islets, whereas NLR family pyrin domain containing 12 (NLRP12), NLR family pyrin domain containing 9 (NLRP9), absent in melanoma 2 (AIM2), NLRP3, and NLRC4 were among the lowest expressed genes. For further confirmation, we evaluated the expression of 11 key genes involved in inflammasome activation in healthy human islets obtained from non-diabetic donors by qPCR. Based on ∆Ct values, an mRNA expression analysis revealed that NF-κβ1, ASC, and IL-18 were among the more highly expressed genes in human islets, while AIM2 and NLRC4 were among those with the lowest expression (Figure 1B). Next, we correlated the expression of MAPK8IP1 with IRGs in the pancreatic islets. As illustrated in Figure 1C–E, the expression of MAPK8IP1 correlated positively ($p \leq 0.05$) with that of NLRP3, GSDMD, and ASC (also known as PYCARD), whereas it correlated inversely with that of IL-18, IL-1β, IL-6, CARD17+CASP1, and NF-κβ1. ( Figure 1F–J). On the other hand, AIM2, NLRP9, and NLRC4 showed no expression correlation with MAPK8IP1 (Figure 1K–M). Together, these results reveal that the expression of MAPK8IP1 is correlated with key IRGs in human islets.
## 2.2. Mapk8ip1 Silencing Influences IRGs in INS-1 (832/13) Cells
Having established the expression profiles of these IRGs in human islets, we analyzed the expression profiles of 11 IRGs in rat INS-1 cells. As is shown in Figure 2A, Nf-κβ1 and Nlrp1 were highly expressed, whereas Casp-1 and Il-6 were expressed at lower levels compared with other genes. To further explore the impact of Mapk8ip1 on the expression of IRGs, we silenced Mapk8ip1 in the INS-1 cells using a pool of siRNA.
The results showed a significant reduction ($p \leq 0.05$) in Mapk8ip1 mRNA levels (~$82\%$) 48 h post-transfection compared with negative control cells (Figure 2B). Subsequently, we observed a significant decrease ($p \leq 0.05$) in the mRNA levels of the IRGs, including Il-1β (~$32\%$), Nlrp3 (~$22\%$), Casp1 (~$22\%$), Nlrc4 (~$31\%$), Gsdmd (~$44\%$), Nlrp1 (~$20\%$), Il-18 (~$35\%$), Il-6 (~$30\%$), Asc (~$25\%$), and the transcriptional activator Nf-κβ1 (~$16\%$) compared with the control cells (Figure 2C). No significant alteration in the expression of Aim2 was documented (Figure 2C). At the protein levels, a significant down-regulation ($p \leq 0.05$) was observed in NLRP3 (~$30\%$; Figure 2D), GSDMD (full-length ~$40\%$ and cleaved N-terminal GSDMD ~$34\%$; Figure 2E), and IL-1β (pro IL-1β ~$26\%$ and mature IL-1β ~$28\%$; Figure 2F) in the Mapk8ip1-silenced cells versus controls. On the other hand, the protein expression of un-cleaved CASP-1 was not affected, whereas the expression level of cleaved caspase was reduced (~$25\%$) ($p \leq 0.05$) (Figure 2G). The replicas of the full-length Western blot expressions after Mapk8ip1 silencing are displayed in Supplementary Figures S1–S4.
Overall, these findings suggest that the silencing of Mapk8ip1 leads to decreased expression levels of key IRGs at the mRNA and/or protein levels.
## 2.3. Inflammasome Activation Reduces Cell Viability and Alters the Expression of Pancreatic Β-Cell Function Genes
To investigate the potential of Mapk8ip1 silencing to interfere with inflammasome activation, we initially assessed the impact of inflammasome activation on un-transfected INS-1 cells. Typically, inflammasome activation requires two signals. The first is for upregulating inflammasome components such as pro-IL-1β and NLRP3, which are brought by endotoxin lipopolysaccharide (LPS). The second signal is necessary to promote inflammasome assembly and is typically initiated by substances such as palmitic acid (PA). Therefore, we cultured INS-1 cells in the presence of LPS (1 μM) for 4 h and then stimulated them with various concentrations of palmitate conjugated to fatty acid-free bovine serum albumin (PA–BSA) for 24 h (100 μM, 200 μM, and 500 μM) [8]. Treatment with LPS/PA–BSA resulted in a significant reduction in cell viability ($p \leq 0.05$) at concentrations of 200 μM (~$15\%$) and 500 μM (~$37\%$), while no significant difference was observed at 100 μM PA–BSA compared with the vehicle control (Figure 3A). Based on these results, we selected a 1 μM LPS (4 h) followed by 200 μM PA–BSA stress regimen for further experiments, due to its impact on cell viability.
An expression analysis determined by qPCR revealed a substantial increase ($p \leq 0.05$) in the mRNA levels of most of the genes involved in inflammasome assembly and activation at 200 μM PA–BSA compared with the vehicle control (Figure 3B). The up-regulated genes included Il-1β, Nlrp3, Casp1, Nlrc4, Gsdmd, Nf-κβ1, Nlrp1, Aim2, Il-18, Il-6, Mapk8ip1, and Jnk (Figure 3B). Notably, the expression of most β-cell function genes showed a significant down-regulation ($p \leq 0.05$) compared with the control (Figure 3C). The down-regulated genes included Ins1, Ins2, Glut2, InsR, Cacna1a, and Mafa.
These data indicate that exposing the cells to 1 μM LPS for 4 h, followed by 200 μM PA–BSA led to the up-regulation of most IRGs and the down-regulation of most β-cell function genes.
## 2.4. Expression Silencing of Mapk8ip1 Impairs β-Cell Inflammasome Activation in Stressed INS-1 Cells
Inflammasome assembly results in the cleavage of pro-caspase-1 and the formation of active caspase-1. Activated caspase-1 plays a crucial role in converting pro-IL-1β to mature IL-1β and cleaves GSDMD to form pores in the plasma membrane, which triggers pyroptosis [34]. In order to investigate and understand the impact of Mapk8ip1 silencing on inflammasome activation, we analyzed the mRNA and protein expression of essential genes involved in inflammasome activation in LPS/PA–BSA stressed cells. An expression analysis using qPCR showed a significant reduction in the mRNA levels of Il-1β (~$15\%$), Nlrp3 (~$16\%$), Gsdmd (~$30\%$), Nf-κβ1 (~$16\%$), Nlrp1 (~$10\%$), Il-18 (~$12\%$), Il-6 (~$17\%$), and Asc (~$30\%$) in the Mapk8ip1-silenced LPS/PA–BSA stressed INS-1 cells compared with the negative controls ($p \leq 0.05$) (Figure 4A). However, the expression of Casp-1, Nlrc4, Aim2, and Jnk were not significantly affected (Figure 4A).
At the protein level, we observed a significant reduction in the expression and intracellular processing of pro-IL-1β to mature IL-1β (pro-IL-1β ~$20\%$ and mature IL-1β ~$30\%$, $p \leq 0.01$) in the Mapk8ip1-silenced stressed INS-1 cells when compared with negative controls (Figure 4B). Additionally, the expression and cleavage of GSDMD and NLRP3 were also reduced (full-length GSDMD ~$28\%$, $p \leq 0.05$; cleaved N-terminal GSDMD ~$27\%$, $p \leq 0.001$; and NLRP3 ~$31\%$, $p \leq 0.05$) (Figure 4C,D). Although the protein expression of activated CASP-1 and phosphorylated JNK showed a trend towards reduction in the Mapk8ip1-silenced stressed INS-1 cells, the data were not statistically significant (Figure 4E,G). The protein expression of JNK remained unaffected (Figure 4F). The replicas of the full-length Western blot expressions after Mapk8ip1 silencing and LPS/PA–BSA treatment are displayed in Supplementary Figures S5–S7. Together, these findings strongly indicate that Mapk8ip1 silencing impairs stimulation-induced inflammasome activation.
## 2.5. Expression Silencing of Mapk8ip1 Influences β-Cell Physiology
Herein, we investigated the effects of Mapk8ip1 silencing on apoptosis, ROS production, glucose uptake, and GSIS in INS-1 cells stressed with LPS/PA–BSA. The Mapk8ip1-silenced stressed cells exhibited a significant decrease ($p \leq 0.05$) in apoptosis rate (early and late apoptosis = $20\%$ of the total number of cells) compared with the negative control silenced cells (~$28\%$) (Figure 5A), indicating that the down-regulation of MAPK8IP1 counteracts, at least in part, the pro-apoptotic effect of PA. Additionally, we observed a significant reduction ($p \leq 0.05$) in intracellular ROS production and glucose uptake level in the LPS/PA–BSA-stimulated Mapk8ip1-silenced cells compared with the negative controls (Figure 5B,C).
Moreover, the Mapk8ip1-silenced stressed cells showed no change in their basal insulin secretion (2.8 mM glucose) but exhibited an impaired ability to augment the release of insulin at higher glucose concentrations (16.7 mM glucose) compared with the negative control cells (~$18\%$, $p \leq 0.05$) (Figure 5D). Furthermore, insulin secretion stimulated by potassium chloride (KCl) was significantly reduced (~$22\%$, $p \leq 0.05$) in the Mapk8ip1-silenced stressed cells compared with the negative controls. In contrast, no significant decrease in insulin secretion was observed upon alpha-ketoisocaproic acid (α-KIC) stimulation (Figure 5D).
An analysis of the mRNA expression of β-cell function genes revealed significant reductions in Ins1 (~$25\%$), Ins2 (~$25\%$), Glut2 (~$22\%$), and Cacna1a (~$22\%$) ($p \leq 0.05$) in the Mapk8ip1-silenced stressed cells compared with the negative control, whilst Gck, Pdx-1, Insr, Vamp2, Snap25, Syt5, Cacnb, Mafa, and NeuroD remained unaffected (Figure 6A). These findings indicate that silencing Mapk8ip1 reduced reactive oxygen species (ROS) generation, apoptosis, GSIS, and glucose uptake in stressed INS-1 cells and altered the expression of several pancreatic β-cell function genes.
## 2.6. Mapk8ip1 Silencing Reduces GSDMD Expression in INS-1 Cells
GSDMD is a component of the inflammasome responsible for forming membrane pores and the induction of pyroptosis. When stimulated, caspase-1 cleaves GSDMD, which releases the N-terminal p30 domain. This domain binds to phospholipids on the plasma membrane, resulting in the formation of large oligomeric pores that facilitate the release of cellular contents and mature IL-1β [34]. Therefore, we sought to examine the expression of GSDMD in control and Mapk8ip1-silenced INS-1 cells via confocal microscopy. As is shown in Figure 7A, the confocal microscopic analysis confirmed the expression of GSDMD in unstimulated INS-1 cells (Figure 7A, upper panel).
Upon stimulation with LPS/PA–BSA, GSDMD translocates towards the plasma membranes, resulting in the appearance of pyroptotic bodies in the membranes where GSDMD accumulates (Figure 7B, upper panel, indicated by white arrows). In contrast, Mapk8ip1 silencing led to a decrease in the expression of GSDMD in both the untreated (Figure 7A, lower panel) and the treated cells (Figure 7B, lower panel). Furthermore, the accumulation of activated GSDMD in the plasma membrane was also reduced in the Mapk8ip1-silenced cells. Thus, these findings suggest that Mapk8ip1 silencing reduces the expression of GSDMD in both stressed and unstressed INS-1 cells.
## 3. Discussion
In this study, we specifically evaluated the regulatory role of MAPK8IP1 in inflammasome activation in pancreatic β-cells. Our data described the expression profiles of several IRGs in human islets and INS-1 cells and identified significant correlations with MAPK8IP1. The study demonstrated that reduced expression of Mapk8ip1 in INS-1 cells decreased the expression of IRGs, such as Nlrp3, Nlrp1, and Nlrc4, and impaired stimulation-induced inflammasome activation. Furthermore, the silencing of Mapk8ip1 reduced ROS generation and attenuated stress-induced apoptosis. Despite the observed down-regulation of the inflammatory pathway in stressed INS-1 cells, the silencing of Mapk8ip1 failed to restore β-cell function, as evidenced by the decreased insulin secretion, glucose uptake, and altered expression of several pancreatic β-cell function genes. These findings suggest that MAPK8IP1 plays an important role in inflammasome regulation.
It is well-documented that upon activation, NLR genes form a complex with the adaptor protein, ASC, which facilitates the activation of pro-caspase-1, forming active caspase-1 p20 tetramer. Activated caspase-1 is responsible for the maturation of the active forms of proinflammatory cytokines IL-18 and IL-1β [34] in addition to the cleaving of GSDMD to trigger pyroptosis [34]. Our findings revealed that the reduction in the expression of NLR genes in the Mapk8ip1-silenced cells was associated with reduced expression levels of Asc, Casp-1, and the three caspase-substrates Il-1β, Il-18, and Gsdmd. Moreover, we also noticed reduced expression levels of Nf-κβ1 and Il-6. NF-κβ1 is the transcriptional activator of NLRP3 and pro-IL-1β [11,35], while IL-6 is a downstream effector of IL-1β [36]. In line with previous findings [10], we hypothesize that the reduced Il-6 and Nf-κβ1 levels in the Mapk8ip1-silenced cells might reflect impaired IL-1β bioactivity or inflammasome activity.
Several IRGs showed decreased expression at the mRNA and/or protein levels when the Mapk8ip1-silenced cells were stressed with LPS/PA–BSA. Among these genes are the cleaved GSDMD N-terminal fragment and mature IL-1β, which are essential effectors of inflammasome activation and mediators of the inflammatory cascade [34]. Furthermore, the accumulation of activated GSDMD in the plasma membrane of the LPS/PA–BSA-treated cells was also reduced in the Mapk8ip1-silenced cells, as was shown by the confocal microscopy. Hence, it seems that Mapk8ip1 silencing influences the expression of genes involved in pyroptosis.
It has been stated that MAPK8IP1 protein functions as a regulator of the JNK signal transduction pathway [28,31]. Phosphorylation of JNK is a critical step for NLRP3 assembly [32] and ASC transcriptional regulation [33]. Therefore, it is conceivable that the impact of MAPK8IP1 on inflammasome activation might result from a MAPK8IP1-induced modulation of JNK. In support of this, our data revealed that Mapk8ip1-silenced cells exhibited a trend towards reduced stress-induced JNK activation following LPS/PA–BSA stimulation. Consistent with these findings, several studies have demonstrated the requirement of the MAPK8IP1 scaffold protein for stress-induced JNK activation [28,37]. On the other hand, the observed impairment in inflammasome activation following Mapk8ip1 silencing might be attributed to the down-regulation of different inflammasome subtypes, such as NLRP1 or NLRC4, which are not substantially affected by JNK [38,39]. Future studies are thus required to fully define the mechanism of inflammasome regulation via the JNK–MAPK8IP1 signaling axis, possibly by testing different JNK isoform knockdowns and various stressors [40].
Circulating free fatty acids have been linked to the pathogenesis of T2D and metabolic inflammation in various tissues in the body [41,42]. Previous studies have suggested that free fatty acids may activate toll-like receptors (TLR), leading to inflammasome activation and the production of proinflammatory cytokines [43,44]. It has been reported that IL-1β elevates the risk for T2D by inducing insulin resistance [14] and increasing β-cell apoptosis [16]. Our findings confirm that fatty acids (e.g., PA) exert their pro-inflammatory effects by activating inflammasomes and causing IL-1β release in β-cells, exacerbating cell death. Mapk8ip1 silencing down-regulated inflammasome activation and decreased PA-induced cell death, indicating that the inflammasome signaling axis is involved in PA-induced β-cell death. Thus, an inflammasome antagonist could be a promising therapy for T2D [23].
Similarly, ROS have been identified as one of the early triggers of inflammasome activation [45,46], and they play a pivotal role in promoting β-cell dysfunction [47,48]. PA is a potent inducer of ROS [8,49,50] and contributes to inflammasome activation and β-cell loss [8,44,51]. Our results confirm that PA-induced inflammasome activation is associated with increased ROS generation. However, the silencing of Mapk8ip1 was found to attenuate ROS production in PA-stressed INS-1 cells. Therefore, the reduced ROS generation following Mapk8ip1 silencing could contribute to the down-regulation of inflammasome activation, as indicated by decreased NLRP3, IL-1β and GSDMD expression.
Despite the evidence supporting the critical role of MAPK8IP1 in regulating inflammasome activation, the translation of such findings to β-cell function has yielded disappointing outcomes with respect to insulin secretion, glucose uptake, and the expression of key β-cell functional genes. A previous study reported that MAPK8IP1 is required for GLUT2 expression and is a candidate for T2D [27]. In contrast, Whitemarsh et al. demonstrated that the loss of MAPK8IP1 function does not directly cause diabetes [28]. Our results support the notion that MAPK8IP1 is involved in regulating insulin secretion.
While it is undeniable that there is a reduction in insulin secretion in both unstressed [30] and stressed Mapk8ip1-silenced INS-1 cells, we noticed an improvement in the siMapk8ip1-induced decrease in GSIS under stress. Mapk8ip1-silenced stressed INS-1 cells showed ~$18\%$, ~$22\%$, and $12\%$ reductions in GSIS with 16.7 mM glucose, KCL, and α-KIC stimulation, respectively, while unstressed siMapk8ip1 cells showed ~$30\%$, ~$33\%$, and ~$40\%$ reductions in GSIS with 16.7 mM glucose, KCL, and α-KIC stimulation, respectively [30].
It is worth noting that this study has certain limitations. Although caspase-1 is the primary canonical intracellular enzyme responsible for the maturation of proIL-1β and GSDMD cleavage, there is also a noncanonical pathway involving caspase-4/-5/-11 that may be responsible for the processing of IL-1β and which occurs independently of inflammasome assembly [33,52]. Thus, it is necessary to assess the contribution of other caspases to the inflammatory pathway.
## 4.1. RNA-Seq Expression Data from Human Pancreatic Islets
A publicly available RNA-seq transcriptomic dataset (GSE50398) was retrieved from NCBI’s Gene Expression Omnibus (GEO; “https://www.ncbi.nlm.nih.gov/bioproject/?term=GSE50398 (accessed on 1 February 2020)” [53]. The expression data were obtained from 89 cadaver donors (European ancestry). Of these, 45 were non-diabetic/normoglycemic donors (HbA1c < $6\%$) and 33 were diabetic/hyperglycemic ($6\%$ ≤ HbA1c < $6.5\%$).
## 4.2. Culturing of INS-1 Cell Line and Palmitic Acid Treatment
Rat insulinoma INS-1 ($\frac{832}{13}$) cells (Research Resource Identifier RRID:CVCL_7226) were kindly provided by Dr. C. B. Newgaard of Duke University, USA [54]. As was previously described, rat insulinoma INS-1 ($\frac{832}{13}$) cells were cultured in RPMI-1640 medium [55]. For the palmitic acid treatment, PA (Sigma-Aldrich, Darmstadt, Germany) was dissolved in 0.1 mmol/l NaOH at 70 °C for 30 min and then conjugated to appropriate amounts of fatty acid-free bovine serum albumin (BSA) at 60 °C for 30 min at a molar ratio of 5:1 [56]. Next, the palmitate–BSA (PA–BSA) conjugate was diluted in serum-free RPMI-1640 medium supplemented with $1\%$ BSA to final concentrations of 100 μmol/L, 200 μmol/L, and 500 μmol/L palmitic acid and added to the cells. For the control, we used the same concentration of vehicle (100 mM NaOH–BSA) in RPMI medium.
The INS-1 cells were cultured in 24-well plates (2.0 × 105 cells/well) for inflammasome activation until they reached $80\%$ confluence. The cells were then stimulated with LPS (1 μg/mL) (LPS, Sigma-Aldrich L4391, from *Escherichia coli* 0111:B4) for four hours, then incubated with 200 μM PA–BSA [24,57]. Following incubation, the cells were used for functional validation assays.
## 4.3. siRNA Transfection
The INS-1 ($\frac{832}{13}$) cells were seeded in a 24-well plate (200,000 cells/well) in a complete RPMI 1640 medium and transfected with two sets of siRNA sequences for MAPK8IP1 (s137914 and s137915) (Thermo Fisher Scientific, Waltham, MA, USA) or scramble negative control siRNA, as previously described [55]. To test the effect of inflammasome activation, 24 h post-transfection, the cells were pretreated with LPS for 4 h and incubated with 200 μM PA–BSA for 24 h. A separate group of transfected cells for the control was similarly treated with vehicle (NaOH in serum-free RPMI containing $1\%$ BSA). Following incubation, the mRNA from the treated transfected cells was isolated for further analysis.
## 4.4. RNA Extraction and qRT-PCR
A High-Capacity cDNA Reverse Transcription Kit (Thermo Fisher Scientific, Waltham, MA, USA) was used to synthesize complementary DNA (cDNA) from the extracted RNA. An expression analysis of the key genes involved in β-cell function from the transfected and non-transfected treated cells was assessed with qPCR using TaqMan gene expression assays through the use of gene-specific primer probes for Mapk8ip1 (Rn00587215_m1), Glut2 (Rn00563565_m1), Ins1 (Rn02121433_g1), Ins2 (Rn01774648_g1), Pdx1 (Rn00755591_m1), Insr (Rn00690703_m1), Gck (Rn00561265_m1), and Rat Hprt1 (Rn01527840_m1). A SYBR green gene expression analysis for several β-cell function genes was conducted using the corresponding primers (Table 1). An expression analysis of the principle genes implicated in β-cell inflammasome assembly/activation was assessed using qRT-PCR via SYBR green gene expression assays using the primers listed in Table 1. Rat Hprt1 was used as an endogenous control for normalizing the expression of the target mRNA. *Relative* gene expression was assessed using the 2−ΔΔCt method. All qPCR reactions were run in 96-well plates in triplicate using the QuantStudio 3 qPCR system (Applied Biosystems, Waltham, MA, USA).
## 4.5. Insulin Secretion Assay
GSIS measurements were performed 48 h post-transfection, as previously described [58]. First, INS-1 ($\frac{832}{13}$) β-cells were incubated in pre-warmed secretion assay buffer (SAB) with 2.8 mM glucose for 2 h. The cells were then stimulated with SAB containing either 2.8 mM glucose, 16.7 mM glucose, 2.8 mM glucose plus 10 mM α-KIC, or 35 mM KCl for 1 h. Next, the amount of secreted insulin was determined using the rat insulin ELISA kit (Mercodia, Uppsala, Sweden) and normalized to the total amount of protein.
## 4.6. Western Blot Analysis
To detect activated inflammasome proteins, INS-1 cells were stimulated with 1 µL LPS/200 μM PA–BSA for 4 h [24,59]. Total protein extraction was performed using ice-cold NP-40 ($1.0\%$ NP-40, 150 mM NaCl, 50 mM Tris-Cl, pH 8.0) lysis buffer containing a protease inhibitor cocktail (Thermo Fisher Scientific, Waltham, MA, USA). A Western blot analysis was performed as previously described [58] with the following antibodies: MAPK8IP1 (anti-rabbit; 1:1000, #Ab24449, Abcam, Cambridge, UK), NLRP3 (anti-rabbit; 1:1000, #A12694, Abclonal, Woburn, MA, USA), CASPASE-1 (anti-rabbit; 1:1000, #A0964, Abclonal, Woburn, MA, USA), IL-1β (anti-rabbit; 1:1000, #A162888, Abclonal, USA), GSDMD (anti-rabbit; 1:1000, #A10164, Abclonal, USA), JNK (anti-rabbit; 1:1000, #A48567, Abclonal, USA), pJNK (anti-rabbit; 1:1000, #AP0631, Abclonal, USA), β-actin (anti-mouse, 1:1000, #A5441, Sigma-Aldrich, Darmstadt, Germany), and secondary anti-mouse (#7076S) or anti-rabbit (#7074S, from Cell Signaling Technology, Danvers, MA, USA). Chemiluminescence was detected using the Clarity ECL substrate kit (Bio-Rad, Hercules, CA, USA). β-actin was used as an endogenous control.
## 4.7. Apoptosis Assay
The transfected and non-transfected cells were cultured in RPMI medium in the presence of vehicle (control) or 1 µL LPS followed by 200 μM PA–BSA, as mentioned earlier. Following 24 h incubation, the cells were re-suspended in 500 μL of Annexin-V (1X) Binding Buffer (BD Biosciences, San Jose, CA, USA) and then stained with 2 μL of Annexin V-FITC and 2 μL of Propidium Iodide (PI) (15 min) in the dark. The cells were analyzed using a BD FACS Aria III flow cytometer (Becton Dickinson, Biosciences, Franklin Lakes, NJ, USA).
## 4.8. Cell Viability Assay
An MTT colorimetric assay (Sigma-Aldrich, Saint Louis, MO, USA) was used to assess cell viability. In brief, transfected and non-transfected INS-1 ($\frac{832}{13}$) cells, seeded in 96-well plates (20 × 104/well), were cultured in RPMI medium in the presence of vehicle (control) or 1 µL LPS followed by PA–BSA for 24 h, as mentioned earlier. An aliquot (10 µL) of MTT solution was added to each well and incubated at 37 °C for 2 h. The formed MTT formazan crystals were dissolved in 100 μL DMSO and the absorbance was measured using a microplate reader at an optical density of 570 nm. The cell viability percentage was calculated.
## 4.9. Glucose Uptake
The glucose uptake in the cultured cells was assessed using 2-NBDG (Invitrogen #N13195, Carlsbad, CA, USA). Briefly, the transfected and non-transfected INS-1 cells were cultured in the presence of vehicle (control) or 1 µL LPS followed by 200 μM PA–BSA for 24 h, as mentioned earlier. Forty-eight hours post-transfection, 100 µM of 2-NBDG was added to each well and incubated at 37 °C for one hour, as previously described [58]. The cells were then trypsinized and analyzed using flow cytometry (BD FACS AriaTM III flow cytometer, San Jose, CA, USA).
## 4.10. ROS generation
According to the manufacturer’s instructions, the intracellular generation of ROS was detected using a ROS-Glo H2O2 assay kit (Cat #G8820, Promega, Madison, WI, USA). Briefly, the transfected and non-transfected INS-1 cells were treated with 1 μg/mL LPS followed by 200 μM PA–BSA or vehicle (control) for 24 h. Forty-eight hours post-transfection, the cells were incubated with H2O2 substrate for 3 h at 37 °C. ROS-Glo detection reagent was added, and the cells were incubated for 20 min at room temperature. The relative luminescence was then detected using a plate reader [58].
## 4.11. Immunofluorescence Assay
The transfected INS-1 cells were plated on glass coverslips and treated with 1 μg/mL LPS/200 μM PA–BSA or vehicle (control). The cells were then fixed using $4\%$ paraformaldehyde for 15 min at room temperature and permeabilized with $0.2\%$ Triton X-100 in phosphate-buffered saline (PBS) for 5 min. Glass coverslips were blocked using $1\%$ Triton X-100 + $2\%$ BSA in phosphate-buffered saline (PBS) for 1 h followed by overnight incubation with a primary antibody against GSDMD (anti-rabbit; 1:1000, #A10164, Abclonal, USA). The cells were washed 3 times with $0.1\%$ Triton X-100 in phosphate-buffered saline (PBS) for 5 min and then labeled with respective secondary antibodies tagged with Alexa 488 for 1.5 h. The coverslips were mounted on slides using mounting media with DAPI (Invitrogen, Carlsbad, CA, USA) to stain the nucleus. The slides were then observed under a confocal microscope (A1R Confocal Laser Microscope System, Nikon Inc., Tokyo, Japan).
## 4.12. Statistical Analysis
A Student t-test or a nonparametric Mann-Whitney test was used for the differential expression analysis between the diabetic and non-diabetic donors. The correlation between the variables was calculated using a nonparametric Spearman’s test. All statistical analyses were performed using GraphPad Prism (version 8.0.0, “www.graphpad.com (accessed on 1 February 2020)”).
## 5. Conclusions
In summary, our data suggest that MAPK8IP1 could be an important mediator of β-cell inflammasome. However, despite our promising mechanistic studies identifying MAPK8IP1 as an inflammasome regulator, the therapeutic potential of using MAPK8IP1 to ameliorate T2D seems to be impeded by its role in β-cell function and insulin secretion. This indicates that MAPK8IP1 is involved in multiple pathways that regulate pancreatic β-cell function.
## References
1. Gonzalez L.L., Garrie K., Turner M.D.. **Type 2 diabetes—An autoinflammatory disease driven by metabolic stress**. *Biochim. Biophys. Acta Mol. Basis Dis.* (2018) **1864** 3805-3823. DOI: 10.1016/j.bbadis.2018.08.034
2. Hameed I., Masoodi S.R., Mir S.A., Nabi M., Ghazanfar K., Ganai B.A.. **Type 2 diabetes mellitus: From a metabolic disorder to an inflammatory condition**. *World J. Diabetes* (2015) **6** 598-612. DOI: 10.4239/wjd.v6.i4.598
3. Hotamisligil G.S.. **Inflammation, metaflammation and immunometabolic disorders**. *Nature* (2017) **542** 177-185. DOI: 10.1038/nature21363
4. Chawla A., Nguyen K.D., Goh Y.P.S.. **Macrophage-mediated inflammation in metabolic disease**. *Nat. Rev. Immunol.* (2011) **11** 738-749. DOI: 10.1038/nri3071
5. Pirzada R.H., Javaid N., Choi S.. **The Roles of the NLRP3 Inflammasome in Neurodegenerative and Metabolic Diseases and in Relevant Advanced Therapeutic Interventions**. *Genes* (2020) **11**. DOI: 10.3390/genes11020131
6. Jorgensen I., Miao E.A.. **Pyroptotic cell death defends against intracellular pathogens**. *Immunol. Rev.* (2015) **265** 130-142. DOI: 10.1111/imr.12287
7. Lamkanfi M., Dixit V.M.. **Mechanisms and functions of inflammasomes**. *Cell* (2014) **157** 1013-1022. DOI: 10.1016/j.cell.2014.04.007
8. Wen H., Gris D., Lei Y., Jha S., Zhang L., Huang M.T., Brickey W.J., Ting J.P.. **Fatty acid-induced NLRP3-ASC inflammasome activation interferes with insulin signaling**. *Nat. Immunol.* (2011) **12** 408-415. DOI: 10.1038/ni.2022
9. Sepehri Z., Kiani Z., Afshari M., Kohan F., Dalvand A., Ghavami S.. **Inflammasomes and type 2 diabetes: An updated systematic review**. *Immunol. Lett.* (2017) **192** 97-103. DOI: 10.1016/j.imlet.2017.10.010
10. Sokolova M., Sahraoui A., Høyem M., Øgaard J., Lien E., Aukrust P., Yndestad A., Ranheim T., Scholz H.. **NLRP3 inflammasome mediates oxidative stress-induced pancreatic islet dysfunction**. *Am. J. Physiol. Endocrinol. Metab.* (2018) **315** E912-E923. DOI: 10.1152/ajpendo.00461.2017
11. Ding S., Xu S., Ma Y., Liu G., Jang H., Fang J.. **Modulatory Mechanisms of the NLRP3 Inflammasomes in Diabetes**. *Biomolecules* (2019) **9**. DOI: 10.3390/biom9120850
12. Wang J., Shen X., Liu J., Chen W., Wu F., Wu W., Meng Z., Zhu M., Miao C.. **High glucose mediates NLRP3 inflammasome activation via upregulation of ELF3 expression**. *Cell Death Dis.* (2020) **11** 383. DOI: 10.1038/s41419-020-2598-6
13. Luo B., Li B., Wang W., Liu X., Xia Y., Zhang C., Zhang M., Zhang Y., An F.. **NLRP3 gene silencing ameliorates diabetic cardiomyopathy in a type 2 diabetes rat model**. *PLoS ONE* (2014) **9**. DOI: 10.1371/journal.pone.0104771
14. Vandanmagsar B., Youm Y.H., Ravussin A., Galgani J.E., Stadler K., Mynatt R.L., Ravussin E., Stephens J.M., Dixit V.D.. **The NLRP3 inflammasome instigates obesity-induced inflammation and insulin resistance**. *Nat. Med.* (2011) **17** 179-188. DOI: 10.1038/nm.2279
15. Stienstra R., van Diepen J.A., Tack C.J., Zaki M.H., van de Veerdonk F.L., Perera D., Neale G.A., Hooiveld G.J., Hijmans A., Vroegrijk I.. **Inflammasome is a central player in the induction of obesity and insulin resistance**. *Proc. Natl. Acad. Sci. USA* (2011) **108** 15324-15329. DOI: 10.1073/pnas.1100255108
16. Maedler K., Sergeev P., Ris F., Oberholzer J., Joller-Jemelka H.I., Spinas G.A., Kaiser N., Halban P.A., Donath M.Y.. **Glucose-induced beta cell production of IL-1beta contributes to glucotoxicity in human pancreatic islets**. *J. Clin. Investig.* (2002) **110** 851-860. DOI: 10.1172/JCI200215318
17. Alfadul H., Sabico S., Al-Daghri N.M.. **The role of interleukin-1β in type 2 diabetes mellitus: A systematic review and meta-analysis**. *Front. Endocrinol.* (2022) **13** 1694. DOI: 10.3389/fendo.2022.901616
18. Youm Y.H., Adijiang A., Vandanmagsar B., Burk D., Ravussin A., Dixit V.D.. **Elimination of the NLRP3-ASC inflammasome protects against chronic obesity-induced pancreatic damage**. *Endocrinology* (2011) **152** 4039-4045. DOI: 10.1210/en.2011-1326
19. Lee H.M., Kim J.J., Kim H.J., Shong M., Ku B.J., Jo E.K.. **Upregulated NLRP3 inflammasome activation in patients with type 2 diabetes**. *Diabetes* (2013) **62** 194-204. DOI: 10.2337/db12-0420
20. Mamun A.A., Wu Y., Nasrin F., Akter A., Taniya M.A., Munir F., Jia C., Xiao J.. **Role of Pyroptosis in Diabetes and Its Therapeutic Implications**. *J. Inflamm. Res.* (2021) **14** 2187-2206. DOI: 10.2147/JIR.S291453
21. Larsen C.M., Faulenbach M., Vaag A., Vølund A., Ehses J.A., Seifert B., Mandrup-Poulsen T., Donath M.Y.. **Interleukin-1-receptor antagonist in type 2 diabetes mellitus**. *N. Engl. J. Med.* (2007) **356** 1517-1526. DOI: 10.1056/NEJMoa065213
22. Hong P., Gu R.N., Li F.X., Xiong X.X., Liang W.B., You Z.J., Zhang H.F.. **NLRP3 inflammasome as a potential treatment in ischemic stroke concomitant with diabetes**. *J. Neuroinflamm.* (2019) **16** 121. DOI: 10.1186/s12974-019-1498-0
23. Bai Y., Mu Q., Bao X., Zuo J., Fang X., Hua J., Zhang D., Jiang G., Li P., Gao S.. **Targeting NLRP3 Inflammasome in the Treatment Of Diabetes and Diabetic Complications: Role of Natural Compounds from Herbal Medicine**. *Aging Dis.* (2021) **12** 1587-1604. DOI: 10.14336/AD.2021.0318
24. Lebreton F., Berishvili E., Parnaud G., Rouget C., Bosco D., Berney T., Lavallard V.. **NLRP3 inflammasome is expressed and regulated in human islets**. *Cell Death Dis.* (2018) **9** 726. DOI: 10.1038/s41419-018-0764-x
25. Pellet J.B., Haefliger J.A., Staple J.K., Widmann C., Welker E., Hirling H., Bonny C., Nicod P., Catsicas S., Waeber G.. **Spatial, temporal and subcellular localization of islet-brain 1 (IB1), a homologue of JIP-1, in mouse brain**. *Eur. J. Neurosci.* (2000) **12** 621-632. DOI: 10.1046/j.1460-9568.2000.00945.x
26. Bonny C., Nicod P., Waeber G.. **IB1, a JIP-1-related nuclear protein present in insulin-secreting cells**. *J. Biol. Chem.* (1998) **273** 1843-1846. DOI: 10.1074/jbc.273.4.1843
27. Waeber G., Delplanque J., Bonny C., Mooser V., Steinmann M., Widmann C., Maillard A., Miklossy J., Dina C., Hani E.H.. **The gene MAPK8IP1, encoding islet-brain-1, is a candidate for type 2 diabetes**. *Nat. Genet.* (2000) **24** 291-295. DOI: 10.1038/73523
28. Whitmarsh A.J., Kuan C.Y., Kennedy N.J., Kelkar N., Haydar T.F., Mordes J.P., Appel M., Rossini A.A., Jones S.N., Flavell R.A.. **Requirement of the JIP1 scaffold protein for stress-induced JNK activation**. *Genes Dev.* (2001) **15** 2421-2432. DOI: 10.1101/gad.922801
29. Standen C.L., Kennedy N.J., Flavell R.A., Davis R.J.. **Signal transduction cross talk mediated by Jun N-terminal kinase-interacting protein and insulin receptor substrate scaffold protein complexes**. *Mol. Cell Biol.* (2009) **29** 4831-4840. DOI: 10.1128/MCB.00155-09
30. Saeed R., Mohammed A.K., Saleh S.E., Aboshanab K.M., Aboulwafa M.M., Taneera J.. **Expression silencing of Mitogen-Activated Protein Kinase 8 Interacting Protein-1 conferred its role in pancreatic β-cell physiology and Insulin Secretion**. *Metabolites* (2023) **13**. PMID: 36837926
31. Whitmarsh A.J., Cavanagh J., Tournier C., Yasuda J., Davis R.J.. **A mammalian scaffold complex that selectively mediates MAP kinase activation**. *Science* (1998) **281** 1671-1674. DOI: 10.1126/science.281.5383.1671
32. Song N., Liu Z.S., Xue W., Bai Z.F., Wang Q.Y., Dai J., Liu X., Huang Y.J., Cai H., Zhan X.Y.. **NLRP3 Phosphorylation Is an Essential Priming Event for Inflammasome Activation**. *Mol. Cell* (2017) **68** 185-197e6. DOI: 10.1016/j.molcel.2017.08.017
33. Ghiasi S.M., Dahllöf M.S., Osmai Y., Osmai M., Jakobsen K.K., Aivazidis A., Tyrberg B., Perruzza L., Prause M.C.B., Christensen D.P.. **Regulation of the β-cell inflammasome and contribution to stress-induced cellular dysfunction and apoptosis**. *Mol. Cell Endocrinol.* (2018) **478** 106-114. DOI: 10.1016/j.mce.2018.08.001
34. Kelley N., Jeltema D., Duan Y., He Y.. **The NLRP3 Inflammasome: An Overview of Mechanisms of Activation and Regulation**. *Int. J. Mol. Sci.* (2019) **20**. DOI: 10.3390/ijms20133328
35. Bauernfeind F.G., Horvath G., Stutz A., Alnemri E.S., MacDonald K., Speert D., Fernandes-Alnemri T., Wu J., Monks B.G., Fitzgerald K.A.. **Cutting Edge: NF-κB Activating Pattern Recognition and Cytokine Receptors License NLRP3 Inflammasome Activation by Regulating NLRP3 Expression1**. *J. Immunol.* (2009) **183** 787-791. DOI: 10.4049/jimmunol.0901363
36. Dinarello C.A.. **Interleukin-1 in the pathogenesis and treatment of inflammatory diseases**. *Blood* (2011) **117** 3720-3732. DOI: 10.1182/blood-2010-07-273417
37. Kant S., Standen C.L., Morel C., Jung D.Y., Kim J.K., Swat W., Flavell R.A., Davis R.J.. **A Protein Scaffold Coordinates SRC-Mediated JNK Activation in Response to Metabolic Stress**. *Cell Rep.* (2017) **20** 2775-2783. DOI: 10.1016/j.celrep.2017.08.025
38. Jenster L.-M., Lange K.-E., Normann S., vom Hemdt A., Wuerth J.D., Schiffelers L.D.J., Tesfamariam Y.M., Gohr F.N., Klein L., Kaltheuner I.H.. **P38 kinases mediate NLRP1 inflammasome activation after ribotoxic stress response and virus infection**. *J. Exp. Med.* (2022) **220** e20220837. DOI: 10.1084/jem.20220837
39. Place D.E., Samir P., Karki R., Briard B., Vogel P., Kanneganti T.D.. **ASK Family Kinases Are Required for Optimal NLRP3 Inflammasome Priming**. *Am. J. Pathol.* (2018) **188** 1021-1030. DOI: 10.1016/j.ajpath.2017.12.006
40. Prause M., Christensen D.P., Billestrup N., Mandrup-Poulsen T.. **JNK1 protects against glucolipotoxicity-mediated beta-cell apoptosis**. *PLoS ONE* (2014) **9**. DOI: 10.1371/journal.pone.0087067
41. Ralston J.C., Lyons C.L., Kennedy E.B., Kirwan A.M., Roche H.M.. **Fatty Acids and NLRP3 Inflammasome-Mediated Inflammation in Metabolic Tissues**. *Annu. Rev. Nutr.* (2017) **37** 77-102. DOI: 10.1146/annurev-nutr-071816-064836
42. Nemecz M., Constantin A., Dumitrescu M., Alexandru N., Filippi A., Tanko G., Georgescu A.. **The Distinct Effects of Palmitic and Oleic Acid on Pancreatic Beta Cell Function: The Elucidation of Associated Mechanisms and Effector Molecules**. *Front. Pharmacol.* (2019) **9** 1554. DOI: 10.3389/fphar.2018.01554
43. Shi H., Kokoeva M.V., Inouye K., Tzameli I., Yin H., Flier J.S.. **TLR4 links innate immunity and fatty acid-induced insulin resistance**. *J. Clin. Investig.* (2006) **116** 3015-3025. DOI: 10.1172/JCI28898
44. Legrand-Poels S., Esser N., L’homme L., Scheen A., Paquot N., Piette J.. **Free fatty acids as modulators of the NLRP3 inflammasome in obesity/type 2 diabetes**. *Biochem. Pharmacol.* (2014) **92** 131-141. DOI: 10.1016/j.bcp.2014.08.013
45. Zhou R., Yazdi A.S., Menu P., Tschopp J.. **A role for mitochondria in NLRP3 inflammasome activation**. *Nature* (2011) **469** 221-225. DOI: 10.1038/nature09663
46. Zhou R., Tardivel A., Thorens B., Choi I., Tschopp J.. **Thioredoxin-interacting protein links oxidative stress to inflammasome activation**. *Nat. Immunol.* (2010) **11** 136-140. DOI: 10.1038/ni.1831
47. Volpe C.M.O., Villar-Delfino P.H., Dos Anjos P.M.F., Nogueira-Machado J.A.. **Cellular death, reactive oxygen species (ROS) and diabetic complications**. *Cell Death Dis.* (2018) **9** 119. DOI: 10.1038/s41419-017-0135-z
48. Eguchi N., Vaziri N.D., Dafoe D.C., Ichii H.. **The Role of Oxidative Stress in Pancreatic β Cell Dysfunction in Diabetes**. *Int. J. Mol. Sci.* (2021) **22**. DOI: 10.3390/ijms22041509
49. Sergi D., Naumovski N., Heilbronn L.K., Abeywardena M., O’Callaghan N., Lionetti L., Luscombe-Marsh N.. **Mitochondrial (Dys)function and Insulin Resistance: From Pathophysiological Molecular Mechanisms to the Impact of Diet**. *Front. Physiol.* (2019) **10** 532. DOI: 10.3389/fphys.2019.00532
50. Li N., Frigerio F., Maechler P.. **The sensitivity of pancreatic beta-cells to mitochondrial injuries triggered by lipotoxicity and oxidative stress**. *Biochem. Soc. Trans.* (2008) **36** 930-934. DOI: 10.1042/BST0360930
51. Gerber P.A., Rutter G.A.. **The Role of Oxidative Stress and Hypoxia in Pancreatic Beta-Cell Dysfunction in Diabetes Mellitus**. *Antioxid. Redox Signal.* (2017) **26** 501-518. DOI: 10.1089/ars.2016.6755
52. Pellegrini C., Antonioli L., Lopez-Castejon G., Blandizzi C., Fornai M.. **Canonical and Non-Canonical Activation of NLRP3 Inflammasome at the Crossroad between Immune Tolerance and Intestinal Inflammation**. *Front. Immunol.* (2017) **8** 36. DOI: 10.3389/fimmu.2017.00036
53. Fadista J., Vikman P., Laakso E.O., Mollet I.G., Esguerra J.L., Taneera J., Storm P., Osmark P., Ladenvall C., Prasad R.B.. **Global genomic and transcriptomic analysis of human pancreatic islets reveals novel genes influencing glucose metabolism**. *Proc. Natl. Acad. Sci. USA* (2014) **111** 13924-13929. DOI: 10.1073/pnas.1402665111
54. Hohmeier H.E., Mulder H., Chen G., Henkel-Rieger R., Prentki M., Newgard C.B.. **Isolation of INS-1-derived cell lines with robust ATP-sensitive K+ channel-dependent and -independent glucose-stimulated insulin secretion**. *Diabetes* (2000) **49** 424-430. DOI: 10.2337/diabetes.49.3.424
55. Taneera J., Prasad R.B., Dhaiban S., Mohammed A.K., Haataja L., Arvan P., Hamad M., Groop L., Wollheim C.B.. **Silencing of the FTO gene inhibits insulin secretion: An in vitro study using GRINCH cells**. *Mol. Cell. Endocrinol.* (2018) **472** 10-17. DOI: 10.1016/j.mce.2018.06.003
56. Joshi-Barve S., Barve S.S., Amancherla K., Gobejishvili L., Hill D., Cave M., Hote P., McClain C.J.. **Palmitic acid induces production of proinflammatory cytokine interleukin-8 from hepatocytes**. *Hepatology* (2007) **46** 823-830. DOI: 10.1002/hep.21752
57. Dong Z., Zhuang Q., Ning M., Wu S., Lu L., Wan X.. **Palmitic acid stimulates NLRP3 inflammasome activation through TLR4-NF-κB signal pathway in hepatic stellate cells**. *Ann. Transl. Med.* (2020) **8** 168. DOI: 10.21037/atm.2020.02.21
58. El-Huneidi W., Anjum S., Mohammed A.K., Unnikannan H., Saeed R., Bajbouj K., Abu-Gharbieh E., Taneera J.. **Copine 3 “CPNE3” is a novel regulator for insulin secretion and glucose uptake in pancreatic β-cells**. *Sci. Rep.* (2021) **11** 20692. DOI: 10.1038/s41598-021-00255-0
59. Gurung P., Li B., Subbarao Malireddi R.K., Lamkanfi M., Geiger T.L., Kanneganti T.-D.. **Chronic TLR Stimulation Controls NLRP3 Inflammasome Activation through IL-10 Mediated Regulation of NLRP3 Expression and Caspase-8 Activation**. *Sci. Rep.* (2015) **5** 14488. DOI: 10.1038/srep14488
|
---
title: miRNome and Proteome Profiling of Human Keratinocytes and Adipose Derived Stem
Cells Proposed miRNA-Mediated Regulations of Epidermal Growth Factor and Interleukin
1-Alpha
authors:
- Hady Shahin
- Sallam Abdallah
- Jyotirmoy Das
- Weihai He
- Ibrahim El-Serafi
- Ingrid Steinvall
- Folke Sjöberg
- Moustafa Elmasry
- Ahmed T. El-Serafi
journal: International Journal of Molecular Sciences
year: 2023
pmcid: PMC10002856
doi: 10.3390/ijms24054956
license: CC BY 4.0
---
# miRNome and Proteome Profiling of Human Keratinocytes and Adipose Derived Stem Cells Proposed miRNA-Mediated Regulations of Epidermal Growth Factor and Interleukin 1-Alpha
## Abstract
Wound healing is regulated by complex crosstalk between keratinocytes and other cell types, including stem cells. In this study, a 7-day direct co-culture model of human keratinocytes and adipose-derived stem cells (ADSCs) was proposed to study the interaction between the two cell types, in order to identify regulators of ADSCs differentiation toward the epidermal lineage. As major mediators of cell communication, miRNome and proteome profiles in cell lysates of cultured human keratinocytes and ADSCs were explored through experimental and computational analyses. GeneChip® miRNA microarray, identified 378 differentially expressed miRNAs; of these, 114 miRNAs were upregulated and 264 miRNAs were downregulated in keratinocytes. According to miRNA target prediction databases and the Expression Atlas database, 109 skin-related genes were obtained. Pathway enrichment analysis revealed 14 pathways including vesicle-mediated transport, signaling by interleukin, and others. Proteome profiling showed a significant upregulation of the epidermal growth factor (EGF) and Interleukin 1-alpha (IL-1α) compared to ADSCs. Integrated analysis through cross-matching the differentially expressed miRNA and proteins suggested two potential pathways for regulations of epidermal differentiation; the first is EGF-based through the downregulation of miR-485-5p and miR-6765-5p and/or the upregulation of miR-4459. The second is mediated by IL-1α overexpression through four isomers of miR-30-5p and miR-181a-5p.
## 1. Introduction
Skin is a self-renewing organ that covers the entire surface area of the body. It forms an anatomical guard that works as a barrier from the outer environment. Skin broadly consists of 3 layers: epidermis, dermis, and subcutaneous adipose tissue [1,2]. Because of the complex organization, wound healing involves orchestrated synergy between different skin cells with interplay between a plethora of signaling chemokines, growth factors, and cytokines. The healing process starts with the formation of a fibrin clot, followed by the recruitment of inflammatory cells. Granulation tissue then starts to form alongside angiogenesis. Re-epithelialization takes place with the recruitment of proliferating fibroblasts and migrating keratinocytes causing the dermis to contract [3,4]. Epithelialization is an integral component of wound healing used as a decisive factor of its success. Impaired epithelialization is a characterizing feature of chronic wounds [5]. Keratinocytes, the predominant cellular constituent of the epidermis, play a central role in restoring the epidermis after injury as epithelization is largely mediated by local keratinocytes at the wound edges and by epithelial stem cells in the hair follicles or sweat glands [3,4].
Adipose-derived stem cells (ADSCs), known for their plasticity and low immunogenicity, can be isolated by minimally invasive procedures from subcutaneous adipose tissue with a remarkable yield of ADSCs [6]. These cells have shown potential to differentiate into cell types from different lineages, including osteogenic, chondrogenic, neural, cardiomyogenic, hepatic, endocrinal, as well as epithelial lineages [7,8,9,10]. Additionally, the plasticity and immunomodulatory capacity of ADSCs could increase the chances for the success of cellular transplantation [11]. Establishing sustainable cultures of epidermal cells while eliminating the progressive loss of epidermal stem cell population can be a challenging process due to rapid clonal conversion [12]. As a result, in vitro differentiation protocols were considered to convert ADSCs into epidermal-like cells [13]. Several reports tried to optimize epidermal differentiation protocol with organic and non-organic culture media additives [14,15]. Unfortunately, the effectiveness, reproducibility, and compatibility with the regulations for clinical use are always challenging.
Cross-talk within the epidermal niche between keratinocytes, fibroblasts, stem cells, endothelium, immune cells, and other cell types is vital for successful cutaneous wound repair [5]. Keratinocytes release signaling molecules that act in an autocrine and paracrine manner to stimulate the epithelialization through modulating the proliferative and migratory pattern of the surrounding cells. Multiple signaling molecules, including cytokines, chemokines, growth factors, integrins, and coding and non-coding RNAs function as active mediators of cell–cell communication and are crucial for its effectiveness [5,16]. Furthermore, dermal adipocytes and ADSCs play important roles during skin repair via endocrine secretions and signaling [17,18,19,20,21]. Extracellular vesicles (EVs) can promote early stages of healing through pleiotropic effects, including enhancing fibroblasts migratory and proliferative capacity, as well as inducing collagen deposition. EVs derived from ADSCs have been reported to accelerate murine cutaneous wound healing when applied through local and intravenous injections [20]. MicroRNAs (miRNAs) are small non-coding RNAs that are a major constituent of the EVs cargo; miRNAs can regulate gene expression in neighboring cells when released extracellularly, either freely or within Evs. Additionally, miRNAs play a role in all major cellular processes, including metabolism, cell proliferation, differentiation, and apoptosis [2,22]. A single miRNA can regulate several mRNAs at the post-transcriptional level, but several miRNAs can conversely bind to a single gene and cooperatively fine-tune its expression, which attests to the complexity of miRNA-mediated gene regulation [22,23]. Epigenetically, miRNAs can activate/repress gene expression by controlling the rate of transcription or translation [24,25]. Additionally, miRNAs play a central role in epidermal development where various miRNAs have been detected throughout skin cell lineages during embryonic skin morphogenesis [26]. Studies involving conditional depletion of either of the major regulators of miRNAs biogenesis, Dicer and Drosha, in murine epidermis showed the loss of barrier function, aberrations in hair follicle growth, and impaired epidermal differentiation [26,27]. Various studies identified a multitude of miRNAs and their role in regulating key epidermal developmental processes by creating feedback loops that modulate the proliferation, differentiation, and the migration of epidermal cells in both normal and disease conditions [27,28,29,30]. Additionally, miRNAs are involved in the overlapping phases of cutaneous wound healing, including the inflammatory phase, and as regulators of angiogenesis during re-epithelialization [2,31]. In the context of stem cells, miRNAs are involved in the regulatory pathways modulating the differentiation process of ADSCs into various cell lineages including osteogenic, chondrogenic, neuronal and adipogenic differentiation [32,33,34].
In this study, the effect of a direct co-culture model on stem cell differentiation is reported. The model consists of human keratinocytes and ADSCs co-cultivated at equal numbers. Additionally, this study aims to characterize the differential repertoire of miRNAs and proteins in the lysate of cultured human keratinocytes and ADSCs. A combined approach of computational and experimental analyses was employed (Figure 1) in order to establish a miRNome-proteome axis with a focus on potential regulations that can alter the fate of ADSCs to differentiate into keratinocyte-like cells.
## 2.1. Direct Co-Culture Can Enhance the Commitment of ADSCs towards the Epidermal Lineage
HaCaT, the epidermal cell line, was considered a positive control ($100\%$ group) and ASC52, the immortalized adipose-derived stem cell line, was considered a negative control ($0\%$ group). The study group consisted of a mixture of an equal number of both cell lines ($50\%$ group) as a direct co-culture system, as shown in Figure 2a–c.
After 7 days in culture, gene expression analysis indicated that direct co-cultures of HaCaT with ASC52 expressed higher levels of the transcription factor p63 and the early epithelial marker KRT18 (0.9 folds) while both markers could not be detected in the $0\%$ group (p-value = 0.0004 and 0.01 respectively). The expression of both markers in the co-culture ($50\%$ group) was similar to f HaCaT cells ($100\%$ group) with a p-value = 0.51 and 0.67, respectively (Figure 2d,e). Similar up-regulation pattern was shown for the basal-specific epidermal marker KRT5 (p-value = 0.0004; Figure 2f). Interestingly the co-culture group showed a trend of upregulation of the basal-specific epidermal marker KRT14 with 0.4 folds over the $100\%$ group (p-value = 0.051; Figure 2g) while it kept the same upregulation pattern in comparison to the $0\%$ group (p-value = 0.0005; Figure 2g). On the contrary, cells at day 7 in a co-culture did not express the late epidermal differentiation markers KRT1 and 10 compared with that of the ASC52 cells in the $0\%$ group (p-value = 0.1883 and 0.1841, respectively; Figure 2h,i).
To confirm our initial findings, the expression of the selected epidermal-specific differentiation markers was evaluated with immunocytochemistry (ICC) after 7 days in culture. Monocultures of HaCaT ($100\%$ group) and ASC52 ($0\%$ group) were used as positive and negative controls, respectively. ASC52 cells in the $0\%$ group exhibited no expression for KRT18 or 5 (Figure 2j,n) and faint expression of KRT14 (Figure 2r). Interestingly, in the $50\%$ group, some of the ASC52 cells, with their distinct “spindle-like” morphology, were clearly expressing the three studied markers, especially the cells in close proximity to HaCaT colonies (Figure 2k,o,s). The relative expression of the immunohistochemical marker (Figure 2m,q,u) showed that cells in the $50\%$ group collectively expressed higher levels of KRT18, 5, and 14 with 0.7, 0.8, and 0.6 folds respectively compared to the $100\%$ group (p-value = 1.73 × 10−9, 0.004 and 1.76 × 10−9 respectively). The $0\%$ group comprising solely of an adipose-derived stem cell line showed the lowest expression levels of KRT18, 5, and 14 with fold changes of 0.1, 0.01, and 0.4, respectively (p-value = 2.95 × 10−16; 1.46221 × 10−12; and 1.21 × 10−3; respectively) compared to the $50\%$ group.
## 2.2. 378 miRNAs Differentially Expressed between Primary Keratinocytes and ADSCs
miRNAs represent an important vehicle for communication between cells. To evaluate the differentially expressed miRNA, GeneChip® miRNA arrays were used to profile the expression in primary human keratinocytes and ADSCs. A total of 378 miRNAs were identified as differentially expressed miRNAs (DEmiRNAs) between primary keratinocytes and ADSCs (Figure 3a,b; Supplementary Figure S1 and Table S5). Of these, 114 miRNAs ($30.16\%$) were upregulated in keratinocytes while 264 miRNAs ($69.84\%$) were downregulated. Microarray data was validated with qPCR analysis for the selected miRNA targets using individual miRNA assays. The following miRNAs, miR-30b-5p (3.1 fold, p-value = 0.002), miR-30c-5p (2.5 fold, p-value = 0.002), and miR-203a (5018 fold, p-value = 0.03), followed the same pattern of significant differential expression and showed upregulation in keratinocytes. The expression of miR-34a-3p (0.1 fold, p-value = 0.008) showed a significant downregulation in keratinocytes in a similar pattern to that in the microarray analysis. Furthermore, miR-29b-3p (0.6 fold, p-value = 0.12), miR-195-5p (1.1 fold, p-value = 0.48) and miR-374a-5p (0.9 fold, p-value = 0.24) were non-significant in both assays. Our validation of selected individual miRNA revealed overall agreement with the microarray data.
## 2.3. Differentially Expressed miRNAs–mRNAs Interactome and Enrichment Analysis of the Upregulated miRNA Targets in Keratinocytes Reveals 14 Pathways
The target genes of the identified significant differentially expressed miRNAs (DEmiRNAs) were investigated separately for upregulated and downregulated miRNAs. A total of 659 unique target mRNAs related to 33 upregulated miRNAs in keratinocytes were identified by combining the three target regions, UTRs (3′ and 5′) and CDS (Supplementary Table S6). A similar analysis was performed on the downregulated miRNAs in keratinocytes and a total number of 555 unique target mRNAs for 58 downregulated miRNAs were observed (Supplementary Table S7). To investigate the relation between the upregulated miRNAs and their target genes expression in skin, we explored the Expression Atlas database “https://www.ebi.ac.uk/gxa/home (accessed on 2 December 2022)” and collected the tissue related gene expression data from the FANTOM5 [35] dataset with a cut-off score. The cut-off score (CS) calculated as follows:[1]CS=xi−x¯, where xi: is the expression value of the gene and x¯: is the mean value of total 13,476 genes expressed in Skin, x¯=∑$i = 1$nxiN.
In this case, a list of 109 unique target genes related to 26 upregulated miRNAs were identified, including AGO2, CDKN1A, MAPK1, MCL1, SEPTIN2, SMAD5, TP53, and TSC1 (Figure 4b; Supplementary Table S8). Additionally, 14 pathways were found to be significantly enriched with this list of 109 genes, including signaling by interleukins, RUNX3 regulated CDKN1A transcription, membrane trafficking, vesicle-mediated transport, and others (Figure 4b; Supplementary Table S9).
## 2.4. Proteome Profiling and Integrated Analysis with Differentially Expressed miRNAs Highlights miRNA-Mediated Regulations
The protein content in the lysate of primary keratinocytes and ADSCs was experimentally explored, using proteome profiler arrays (Supplementary Table S10). Interestingly, only two proteins were significantly upregulated in keratinocytes, which were Epidermal growth factor (EGF) and Interleukin 1-alpha (IL-1α), as shown in Figure 5a.
Integrated analysis between DEmiRNAs and differentially expressed proteins (DEproteins) in the lysate of the studied cell types was conducted. The fisher’s exact test suggested a strong association between a number of DEmiRNAs and the upregulated proteins (p-value = 0.04; Supplementary Table S11). This integrated analysis presented two predictions for miRNA-mediated gene regulations and their protein products. EGF is likely to be directly controlled by the downregulated miRNAs miR-485-5p and miR-6765-5p or the upregulated miR-4459 (Figure 5b). On the other hand, IL-1α can be controlled by 5 of the upregulated miRNAs, 4 isomers of miR-30 (b, c, d, and e)-5p, and miR-181a-5p (Figure 5c).
## 3. Discussion
The interaction between keratinocytes and stem cells is not only crucial to understand the wound healing process, but also to identify the key regulators of stem cell differentiation into the epidermal lineage. Culturing keratinocytes alongside ADSCs in monolayer allow for cell–cell contact and communication. In our study, a co-culture of keratinocyte and stem cell lines showed genotypic and phenotypic changes allowed by the cross-communication between cells from the two sources. The expected result was the activation of intracellular signaling cascades that enhanced ADSCs differentiation. The effect of direct co-culture on ADSCs differentiation toward other target cell lineages has been previously reported. For example, ADSCs’ potential for osteogenic differentiation was enhanced when co-cultured at a 50:50 ratio with dental pulp stem cells while the effect was diminished with the use of EVs release inhibitor. In support of our findings, this study confirmed the positive effect of co-culture on enhancing cell differentiation, through cell signaling mediators exchanged between the co-cultured cells [36]. Similarly, direct co-culture has also promoted adipogenesis in ADSCs when seeded at a ratio of 70:30 with umbilical vein endothelial cells and cultured in adipogenic differentiation media [37]. Furthermore, the capacity of ADSCs to differentiate into keratinocyte-like cells was described in another co-culture system where keratinocytes were cultured in a transwell above a monolayer of ADSCs. In this model, secreted molecules and growth factors travel through the pores by gravity and induced the differentiation of ADSCs monolayer. In addition, the authors investigated the effect of keratinocyte conditioned medium on ADSCs. Starting at day 7, ADSCs demonstrated gene and protein expression of epidermal markers KRT5, 14, involucrin, filaggrin, and stratifin, comparable to those of keratinocytes. Furthermore, it was not until day 10, when 20–$30\%$ of ADSCs changed their morphology from the typical spindle-like appearance of stem cells into a polygonal morphology resembling that of keratinocytes [38].
In our direct co-culture model, upregulation of p63 expression was detected after 7 days to a level resembling that of HaCaT. This marker can be considered as a key transcription factor for the epidermal lineage, being the first gene product distinguishing epidermal progenitor cells, as well as a prevalently expressed marker in proliferating keratinocytes [39,40,41]. The upregulation of p63 suggests that the ASC52 in the co-culture started the commitment of differentiation into epidermal-like cells. In agreement, a previous study showed nuclear p63 expression at day 7 when cultivating MSCs derived from umbilical cord in keratinocyte-specific media with EGF and calcium [42]. KRT18 is a major component of intermediate filaments that acts as an early epithelial differentiation marker, expressed exclusively in simple epithelium prior to stratification [40,43]. *Our* gene and protein expression data showed the upregulation of KRT18 in the co-cultured cells in a similar trend to p63 gene expression, which supported the evidence for early epidermal differentiation. KRT18 has been shown in differentiating MSCs induced by a cocktail of growth factors, including KGF, EGF, HGF, and IGF-2 for 14 days and altered their fibroblastic morphology to epithelial-like [44]. As none of these differentiation inducers were added to the culture, the intracellular communication was expected to trigger the same effect. Furthermore, KRT5 and 14 were upregulated in our direct co-culture on both the gene expression and protein levels. Dos Santos et al. [ 2019] reported that the expression of KRT14 in umbilical cord MSCs cultured in keratinocyte media reached its peak at the first day of cultivation followed by abrupt descend at day 4 and maintained a steady state until day 14 [42]. This could be attributed to the role of KRT14 in sustaining proliferation in mitotically active basal keratinocytes followed by downregulation when cells become committed to differentiation [45]. Undifferentiated ADSCs in monoculture showed expression of the epithelial basal marker KRT14 at the protein level, in agreement with previous studies [46,47]. KRT1 and KRT10 are epidermal stratification markers expressed by differentiated keratinocytes in the suprabasal cutaneous layers, including the stratum corneum [5,41]. Longer differentiation protocol in literature showed upregulation of KRT10 starting at day 11, which could explain the absence of upregulation in our 7-day culture [42]. A temporal expression analysis of various epidermal differentiation markers in our system could be considered as an interesting future analysis.
Cell-to-cell communication can occur through several approaches, including receptor-mediated events, direct cell–cell contact, and cell–cell synapses. Often released within EVs, miRNAs are one important mediator in communication, and they disseminate through the extracellular fluid to act as signaling molecules by altering gene expression and protein production in the recipient cell. miRNA-mediated cell-cell communication can be achieved through direct exchange of exosomes between adjacent cells, as well as by shuttling exosomes through the systemic circulation [48,49,50]. Additionally, miRNAs have been shown to regulate various aspects of wound healing including cell proliferation, migration, collagen biosynthesis, and vascularization. Moreover, the field of miRNA-based therapeutics is emerging with vast potential to improve wound healing through targeting of antagomir treatments [51].
In keratinocytes, differential miRNA expression showed that miR-203a was among the most highly expressed. The miRNA miR-203a is one of the most abundant miRNA species in the skin and plays a major role in keratinocytes proliferation and differentiation, alongside the miR-30 family [52,53,54]. On the other hand, miR-34a was downregulated. This result was expected as the cells involved in this study were normal healthy keratinocytes, as the overexpression of miR-34a is known to inhibit keratinocyte proliferation and promote apoptosis [55].
Other major mediators of cell–cell communication are cytokines, which are responsible for a wide range of functions across non-immune cells, including a trophic role in the cell repair and regeneration [56,57]. In the context of stem cell differentiation, chemokines mediate vital cellular processes by establishing the cell communication between proliferating and migrating cells [56]. In this study, both cell types are known for their natural ability to produce cytokines. The secretion can aim at modulating the surrounding tissues in physiological conditions or in response to stimuli, such as cellular stresses imposed by infection, inflammation, tissue damage, or specific culture conditions [58,59]. The expression patterns of a group of cytokines and growth factors as a relevant part of the proteome was explored, with a focus on ADSCs differentiation into epidermal-like cells in response to keratinocytes signaling. Out of the 105 studied proteins, EGF and IL-1α were upregulated in keratinocytes. Keratinocytes are known to both produce and respond to EGF, as it is considered as a major regulator for epidermal homeostasis. The production of EGF by keratinocytes was in agreement with our findings [60,61]. Endogenous growth factors play a major role in orchestrating the proliferative phase in epithelization and are essential for effective wound healing [5,62]. In the epidermis, EGF regulates the barrier function, terminal cell differentiation, cell adhesion, protease secretion, and wound healing [63]. Nevertheless, EGF has long been considered as a crucial additive in epidermal cell culture systems as it stimulates keratinocyte migration and proliferation, as well as ADSCs differentiation into the epidermal lineage [5,42]. Adding the cell culture supernatant of HaCaT cells that are induced to overexpress EGF to ADSCs was associated with enhanced proliferation, migration, and invasion of ADSCs. These findings were abolished when HaCaT were transfected with the EGF inhibitor small interfering RNA (siEGF) [64]. Clinically, the topical application of EGF accelerated healing of split-thickness cutaneous wounds through the stimulation of keratinocytes migration across the wound bed [65].
Epidermal keratinocytes, similar to all epithelial cells with a barrier function, are rich in IL-1α in the physiological state. Upon skin injury, trauma, or infection, IL-1α is released promptly, inducing the local inflammation necessary to initiate wound healing [3,66]. Interestingly, IL-1α is amongst the most frequently reported keratinocyte secretion in culture supernatant, which supports our finding. In skin wounds, IL-1 can mobilize locally located stem cells, as well as enhance the keratinocyte migration [58].
The cluster related to miR-30 is known to be functionally involved in cell fate determination and lineage differentiation of mesenchymal stem cells into adipogenic, chondrogenic, and osteogenic lineages [67]. However, to the best of our knowledge, there are no reports of the direct association of this family to epidermal homeostasis or ADSCs differentiation into epidermal-like cells. The miRNA mir-30a has been shown, in systems other than skin, to block the release of inflammatory cytokines, including IL-1α. However, our results showed the upregulation of miR-30b, c, d, and e and not miR-30a. Different members of miR-30 clusters share a common seed sequence near the 5′ end, but they differ in compensatory sequences near the 3′ end, targeting different genes, and pathways [68,69,70]. The positive regulation of miRNA on proteins can be explained by: [1] miRNA-mediated post-transcriptional upregulation, [2] translation upregulation, or [3] competing with repressive proteins, preventing them from binding to their target sites, leading to increased mRNA stability, thus promoting the expression of the target protein [71]. The upregulation of miR-181a was found to deaccelerate keratinocyte proliferation and promote keratinocyte differentiation when induced by high calcium or UVA irradiation [72,73]. On the other hand, miR-4459 plays a role in decreasing the stemness of human embryonic stem cells through inhibiting its target proteins Cell Division Cycle Protein 20 Homolog B (CDC20B) and Autophagy-Related Protein 13 (ATG13) [74]. To the best of our knowledge, there was no evidence pointing in the direction of miR-4459 mediated EGF regulation in the literature.
On the other hand, miR-485 downregulation in human skin has been previously reported, specifically in terminally differentiated keratinocytes [53]; however, the miR-485 mediated EGF interaction in the context of epidermal development or stem cell differentiation has not been reported before, to the best of our knowledge. KRT17 is known to be a direct target of miR-485, while the signaling cascade involving miR-485/KRT17 may result in suppressing EGFR in oral squamous cell carcinoma cell lines [75]. The accumulated evidence, including the miR485-EGF inhibitory regulation shown here, postulate that miR-485 may constitute an appealing target to be investigated in MSCs differentiation into epidermal-like cells. The inhibitory regulation between downregulated miR-6765 and upregulated EGF is another predicted miRNA-target interaction, which should be explored. To the best of our knowledge, this interaction has not been reported in the context of epidermal development, skin repair, or stem cell differentiation.
In summary, this study reported a direct co-culture model that can be used to study the cell-to-cell interaction in monolayer, including stem cell differentiation to the epidermal lineage. The integrated analysis of miRNA–protein characterization predicted novel pathways for the regulation of EGF and IL-1α in keratinocytes. The investigation of these pathways may help in providing new concepts for stem cell differentiation into epidermal cells as well as for wound repair. Based on our findings and the known role of IL-1α in regulation of cell differentiation, this cytokine should be studied as a media additive for stem cells differentiation into keratinocytes. The limitations of this study included the use of cell lines in the co-culture experiments rather than primary cells. Obtaining enough primary cell numbers to conduct these experiments would be challenging, which is the reason beyond modelling with cell lines instead. The co-culture duration was limited to only 7 days in order to prove the efficiency of the system and to detect early differentiation. Extension of the culture could help in showing more positive cells for the studied markers as well as the expression of late keratinocyte markers. Another limitation for analyzing the study finding was the vast possibilities of intermediate effectors, potential feedback loops, and the post-translational changes affecting the predicted miRNA-mediated gene regulations. Future studies should focus on the experimental validation of the newly proposed miRNA-mediated gene and protein regulations to provide a better understanding of their associated signaling cascades. Loss/gain function studies for the effect of suggest-miRNA to EGF and IL-1α can provide biological evidence for our computational model. Additionally, miRNA-based approach can be explored for in vitro differentiation of stem cells into epidermal cell lineage, as well as in non-healing in vivo wound models. The application of EGF, IL-1α, or a combination of them may be investigated for therapeutic potential.
## 4.1. Epidermal Differentiation Induction in Direct Co-Culture Model
HaCaT cell line (accession: CVCL_U602, Cellosaurus database) of spontaneously transformed, non-tumorigenic keratinocytes isolated from histologically normal skin (Elabscience Biotechnology Inc., Houston, TX, USA) and ASC52 (accession: CVCL_0038, Cellosaurus database) and hTERT immortalized adipose-derived Mesenchymal stem cells (ATCC, Manassas, VA, USA) were used in this study. Both cell lines were cultured in DMEM (Gibco, Billings, MT, USA) with $10\%$ fetal bovine serum (FBS; Life Technologies, São Paulo, Brazil) and 10,000 units penicillin and 10 mg streptomycin/mL (Sigma-Aldrich, St. Louis, MO, USA). Upon reaching $90\%$ confluence, the cells were dissociated with Trypsin-EDTA (Sigma Aldrich, St. Louis, MO, USA), stained with $0.4\%$ trypan blue (1:1) and counted using a TC20 automated cell counter (Bio-Rad Inc., Singapore). Cells from both cell lines were mixed with a concentration of 1:1 in a direct co-culture system. Control groups were either HaCaT alone as a positive control or ASC52 alone as a negative control. The groups were designated as 0, 50, & $100\%$ representing the percentage of HaCaT in the mixture. Cells were seeded at a density of 4 × 103 cell/cm2 and kept at 37 °C and $5\%$ CO2 for 7 days, with media change every second day. On day 7, cells were either fixed for immunocytochemistry or harvested for RNA extraction and gene expression analysis.
## 4.2. Primary Cell Isolation
Full thickness skin biopsies were obtained from healthy donors during abdominoplasty and/or breast reduction procedures under the ethical approval no. $\frac{2015}{177}$-31 by the Swedish Ethical Review Authority. The fat portion was carefully separated. Skin cut into 2–3 mm2 then incubated in 1:1 volume of 10 mg/mL Dispase II solution (Gibco, Tokyo, Japan) overnight at 4 °C. The epidermis was then gently peeled from the dermis and incubated with Trypsin-EDTA (Sigma-Aldrich, St. Louis, MO, USA) on a tube rotator at 37 °C for 30 min. Trypsin was deactivated with media with $10\%$ FBS. The cell suspension was allowed to pass through a 70 µm cell strainer (Corning, New York, NY, USA), and then, the keratinocyte suspension was centrifuged at 700 RCF for 4 min. The cell pellets were washed twice with phosphate buffered saline (PBS) (Life Technologies, Grand Island, NY, USA) before cells were resuspended in keratinocyte serum free media (Life Technologies, Grand Island, NY, USA) supplemented with bovine pituitary extract and epidermal growth factor (Life Technologies, Grand Island, NY, USA), and the media was changed every other day. The fat tissue was cut into 0.5–1 cm2 slices. Collagenase I (1mg/mL) (Life Technologies, Grand Island, NY, USA) was then added at a 3:1 ratio of the tissue volume and incubated on a tube rotator at 37 °C for 90 min. The tissue solution was then centrifuged at 700 RCF for 4 min, the oil phase was removed, and pre-warmed DMEM with $10\%$ FBS was added to stop the enzymatic digestion. Digested tissue was then passed through a 70 µm cell strainer and washed twice with serum-free DMEM. ASDCs were re-suspended in DMEM with $10\%$ FBS and $1\%$ penicillin-streptomycin and seeded in monolayer culture.
## 4.3. Human miRNome Profiling with Microarrays
Total RNA, including small RNAs, were extracted from primary keratinocytes and ADSCs using miRNeasy kit (Qiagen, Hilden, Germany). Briefly, cell pellets were lysed using QIAzol lysis reagent with mechanical agitation. Cell lysates were dissolved in chloroform and the aqueous phase supernatant was loaded into RNeasy mini column, washed and eluted in RNase free water. RNA yield was measured using NanoDrop 1000 spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA).
Low molecular weight RNA molecules were labelled with the FlashTag Biotin RNA Labeling Kit (Affymetrix, Santa Clara, CA, USA). Briefly, 500 ng RNA from each sample was ligated to a poly (A) tail followed by binding to a biotinylated signal molecule. miRNA microarray hybridization was then performed with Affymetrix GeneChip miRNA Array 3.0 (Affymetrix, Santa Clara, CA, USA), according to manufacturer’s instructions. Briefly, biotin-labeled samples were incubated with hybridization master mix at 99 °C for 5 min, followed by 45 °C for another 5 min. Hybridization was performed in a rotating hybridization oven (60 rpm) at 48 °C for 18 h. The arrays were washed and stained on GeneChip automated fluidics station and scanned with an Affymetrix GCS 3000 7G-plus scanner (Affymetrix, Santa Clara, CA, USA). The microarray data were analysed using the Transcriptome Analysis Console (TAC)® (Thermo Fisher Scientific, Waltham, MA, USA). Following quality check (Supplementary Table S4), differential expression (DE) analysis was performed between the two study groups for homo sapiens specific miRNA with ID contain “hsa-miR” and log2 fold change ≥2 with p-value < 0.05 (Figure 1a,b). DE result was annotated using the Affymetrix GeneChip® miRNA 4.0 array annotation, version HG38.
## 4.4. Gene and miRNA Expression
Quantitative real-time PCR was performed to detect the expression of early and late epidermal markers from 5 independent cell line replicates of direct co-culture ($$n = 5$$). Total cellular RNA was extracted using RNeasy mini kit (Qiagen, Hilden, Germany). RNA was reverse transcribed into cDNA using Maxima First Strand cDNA Synthesis Kit (Thermo Fisher Scientific, Waltham, MA, USA) following the elimination of double-stranded DNA as recommended by the manufacturer. Gene expression was determined by the PowerUp© SYBR green master mix (Applied biosystem, Waltham, MA, USA). Sequences for the oligonucleotide primers of target genes are listed in Supplementary Table S1.
For miRNA quantification in keratinocytes and ADSCs, total RNA, including small RNAs, as described above were reverse transcribed with the miScript II or miRCURY LNA Reverse-Transcription kits (Qiagen, Hilden, Germany) with RNA input of 250 ng or 10 ng respectively, according to the manufacturer’s instructions. The miScript SYBR® Green or the miRCURY® LNA SYBR® Green were used according to the availability of the marker of interest in either workflow (Supplementary Table S2) following the manufacturer’s protocols. Gene and miRNA expression levels were quantified in a 7500 Fast Real-Time PCR System (Applied Biosystem, Thermo Fisher) and the assays were performed in a minimum of three technical replicates. Gene and miRNA expression were normalized against the endogenous controls Glyceraldehyde 3-phosphate Dehydrogenase (GAPDH) or U6 snRNA, respectively and the fold change was calculated with the 2−ΔΔCT method [76].
## 4.5. DEmiRNAs-mRNA Interaction Network and Pathway Enrichment Analysis
To identify the target genes of the significant DEmiRNAs, the list of up and downregulated miRNAs were uploaded separately to the miRwalk database v3 “http://mirwalk.umm.uni-heidelberg.de (accessed on 1 December 2022)” [77]. The targeted genes were determined according to the following criteria, [1] the targeted mRNAs should be present on the three databases—TargetScan [78], miRDB [79] and miRTarBase [80], [2] the miRwalk score should be ≥0.95. In order to obtain the “Skin” related expression from the upregulated DEmiRNA-targeted mRNAs in keratinocytes, an expression dataset from the FANTOM5 project in the Expression Atlas [35] was collected. A mean cut-off score was applied on the dataset to draw genes which were highly expressed in skin tissues. The resulting DEmiRNAs-mRNA interactome diagram was generated using the Cytoscape software (v3.8.2) [81]. The Reactome database v83 “https://reactome.org (accessed on 2 December 2022)” [82] was considered to identify the enriched pathways from the up- and downregulated DEmiRNA-targeted mRNAs.
## 4.6. Immunocytochemistry Staining
Immunocytochemistry (ICC) was used to detect the expression of the key epidermal differentiation markers (KRT5, KRT14, and KRT18). Following cell fixation in ice-cold methanol $99\%$ for 15 min at −20 °C, endogenous peroxidase activity was quenched with H2O2 (10 min). Cells were then permeabilized by incubation for 15 min at room temperature with $0.05\%$ Triton X-100 (Thermo Scientifc, IL, USA). Non-specific binding was blocked using $3\%$ bovine serum albumin in PBS, for 1 h. Primary antibodies, listed in Supplementary Table S3, were incubated overnight with the fixed cells at 4 °C followed by incubation with the biotinylated mouse and rabbit specific secondary antibody for an hour at room temperature. The immune complex was visualized using the streptavidin-biotin immunoenzymatic antigen detection system where the streptavidin-enzyme conjugate binds to the biotin present on the secondary antibody. Positive cells were stained with chromogen 3-Amino-9-Ethylcarbazole (AEC) using detection immunohistochemistry kit (Abcam, Cambridge, UK) following the manufacturer’s instruction. Then the cells were rinsed with Tris-buffered saline (TBS) and counterstained with 8GX alcian blue solution (Sigma-Aldrich, St. Louis, MO, USA) for 2 min before a final wash and mounting with anhydrous mounting medium. For negative controls, PBS was applied instead of the primary antibody. ICC were run in triplicate for each individual antibody stain ($$n = 3$$). Stained cells photographed under inverted microscope (CKX53, Olympus Corp., Tokyo, Japan) using the Imageview software version X64 (Olympus Corp., Japan). To quantify stain intensity, colour deconvolution was performed using ImageJ 1.53c with the necessary plugins (Wayne Rasband National institutes of health, Bethesda, MD, USA) then setting a unified threshold for integrated pixel density.
## 4.7. Proteome Profiling
Keratinocytes and ADSCs from 3 donors were harvested and solubilized in lysis buffer 17, according to the manufacture instructions (R&D Systems Inc., Minneapolis, MN, USA). Total protein concentration was determined using the DC protein assay (Bio-Rad Inc., Hercules, CA, USA), following the manufacturer’s instructions. Proteome Profiler™ Human XL Cytokine Array (R&D Systems, Inc., Minneapolis, MN, USA) was used to simultaneously assess soluble human proteins and their differential expression between the two cell types. Following the manufacturer’s instructions, the array membranes were blocked for 1 h on a rocking platform, and 150 µg of the cell lysates were incubated with array membranes overnight at 4 °C. The arrays were then incubated with a cocktail of biotinylated detection antibodies for 1 h followed by chemiluminescent detection with Streptavidin-HRP. Membranes were imaged using the ChemiDoc™ MP imaging system (Bio-Rad Inc., Hercules, CA, USA). The pixel densities at each capture spot were quantified and normalized to the reference spots of each blot. Images were analyzed using ImageJ 1.53c (Wayne Rasband National institutes of health, MD, USA) where the mean intensity corresponded to the relative expression of each blotted protein in the cell lysate.
## 4.8. miRNA-Mediated Gene and Protein Regulations (Integrated Analysis)
Integrated analysis was performed between DEmiRNAs and DEproteins in keratinocytes and ADSCs with the assumption that the resulting protein products were directly affected by miRNA-target interaction. To maintain a reliable resource for miRNA target genes, only experimentally validated targets from the miRwalk database were curated and cross-checked with the miRTarBase as well as the miRNA-gene interactions annotated in the Affymetrix GeneChip® miRNA 4.0 array annotation, version HG38. All possible alternative nomenclature was considered for each gene and protein and were used to search for possible miRNA-gene interactions on the aforementioned databases.
## 4.9. Statistical Analysis
For PCR-based assays and ICC, the data were analysed using the Data Analysis ToolPak (Microsoft® Excel, Microsoft® Office 365, Redmond, DC, USA), and the graphs were created using GraphPad Prism Version 9 (GraphPad Software Inc., San Diego, CA, USA). Statistical significance was evaluated using Student’s t-test for unequal variance. Bar charts showed the mathematical mean and the standard error of mean. To estimate the DEmiRNAs from the microarray analysis, p-value < 0.05 was considered as the level of significance. To identify the enriched pathways in the analysis, Benjamin-Hochberg (BH) corrected p-value < 0.05 was applied. Fisher’s exact test was used to test a null-hypothesis, assuming there was no regulation between DEmiRNAs and the genes associated with DEproteins.
## References
1. Hassan W.U., Greiser U., Wang W.. **Role of adipose-derived stem cells in wound healing**. *Wound Repair Regen.* (2014) **22** 313-325. DOI: 10.1111/wrr.12173
2. Soliman A.M., Das S., Ghafar N.A., Teoh S.L.. **Role of MicroRNA in Proliferation Phase of Wound Healing. Front Genet**. *Front. Genet.* (2018) **9** 38. DOI: 10.3389/fgene.2018.00038
3. Reinke J., Sorg H.. **Wound Repair and Regeneration**. *Eur. Surg. Res.* (2012) **49** 35-43. DOI: 10.1159/000339613
4. Roh C., Lyle S.. **Cutaneous Stem Cells and Wound Healing**. *Pediatr. Res.* (2006) **59** 100R-103R. DOI: 10.1203/01.pdr.0000203572.51876.ba
5. Pastar I., Stojadinovic O., Yin N.C., Ramirez H., Nusbaum A.G., Sawaya A., Patel S.B., Khalid L., Isseroff R.R., Tomic-Canic M.. **Epithelialization in Wound Healing: A Comprehensive Review**. *Adv. Wound Care* (2014) **3** 445-464. DOI: 10.1089/wound.2013.0473
6. Kapur S., Katz A.J.. **Characterization of Adipose-Derived Stem Cells: An Update**. *Curr. Stem Cell Res. Ther.* (2010) **5** 95-102. PMID: 19941461
7. Shi J.-G., Fu W.-J., Wang X.-X., Xu Y.-D., Li G., Hong B.-F., Hu K., Cui F.-Z., Wang Y., Zhang X.. **Transdifferentiation of human adipose-derived stem cells into urothelial cells: Potential for urinary tract tissue engineering**. *Cell Tissue Res.* (2012) **347** 737-746. DOI: 10.1007/s00441-011-1317-0
8. Alghfeli L., Parambath D., Manzoor S., Roach H.I., Oreffo R.O., El-Serafi A.T.. **Synthesis of scaffold-free, three dimensional, osteogenic constructs following culture of skeletal osteoprogenitor cells on glass surfaces**. *Bone Rep.* (2021) **15** 101143. DOI: 10.1016/j.bonr.2021.101143
9. Elsharkawi I., Parambath D., Saber-Ayad M., Khan A.A., El-Serafi A.T.. **Exploring the effect of epigenetic modifiers on developing insulin-secreting cells**. *Hum. Cell* (2019) **33** 1-9. DOI: 10.1007/s13577-019-00292-y
10. Karlsson M., Steinvall I., Olofsson P., Thorfinn J., Sjöberg F., Åstrand L., Fayiz S., Khalaf A., Divyasree P., El-Serafi A.T.. **Sprayed cultured autologous keratinocytes in the treatment of severe burns: A retrospective matched cohort study**. *Ann. Burn. Fire Disasters* (2020) **33** 134-142
11. Sierra-Sánchez A., Ordonez-Luque A., Ibanez O.E., Ruiz-Garcia A., Santiago S.A.. **Epithelial In vitro Differentiation of Mesenchymal Stem Cells**. *Curr. Stem Cell Res. Ther.* (2018) **13** 409-422. DOI: 10.2174/1574888X13666180501120416
12. Enzo E., Seconetti A.S., Forcato M., Tenedini E., Polito M.P., Sala I., Carulli S., Contin R., Peano C., Tagliafico E.. **Single-keratinocyte transcriptomic analyses identify different clonal types and proliferative potential mediated by FOXM1 in human epidermal stem cells**. *Nat. Commun.* (2021) **12** 2505. DOI: 10.1038/s41467-021-22779-9
13. El-Serafi A.T., El-Serafi I., Elmasry M., Steinvall I., Sjöberg F.. **Skin regeneration in three dimensions, current status, challenges and opportunities**. *Differentiation* (2017) **96** 26-29. DOI: 10.1016/j.diff.2017.06.002
14. Roxburgh J., Metcalfe A.D., Martin Y.H.. **The effect of medium selection on adipose-derived stem cell expansion and differentiation: Implications for application in regenerative medicine**. *Cytotechnology* (2015) **68** 957-967. DOI: 10.1007/s10616-015-9848-y
15. Petry L., Kippenberger S., Meissner M., Kleemann J., Kaufmann R., Rieger U.M., Wellenbrock S., Reichenbach G., Zöller N., Valesky E.. **Directing adipose-derived stem cells into keratinocyte-like cells: Impact of medium composition and culture condition**. *J. Eur. Acad. Dermatol. Venereol.* (2018) **32** 2010-2019. DOI: 10.1111/jdv.15010
16. Lau K., Paus R., Tiede S., Day P., Bayat A.. **Exploring the role of stem cells in cutaneous wound healing**. *Exp. Dermatol.* (2009) **18** 921-933. DOI: 10.1111/j.1600-0625.2009.00942.x
17. Segalla L., Chirumbolo S., Sbarbati A.. **Dermal white adipose tissue: Much more than a metabolic, lipid-storage organ?**. *Tissue Cell* (2021) **71** 101583. DOI: 10.1016/j.tice.2021.101583
18. Merrick D., Seale P.. **Skinny Fat Cells Stimulate Wound Healing**. *Cell Stem Cell* (2020) **26** 801-803. DOI: 10.1016/j.stem.2020.04.021
19. Shook B.A., Wasko R.R., Mano O., Rutenberg-Schoenberg M., Rudolph M.C., Zirak B., Rivera-Gonzalez G.C., López-Giráldez F., Zarini S., Rezza A.. **Dermal Adipocyte Lipolysis and Myofibroblast Conversion Are Required for Efficient Skin Repair**. *Cell Stem Cell* (2020) **26** 880-895.e6. DOI: 10.1016/j.stem.2020.03.013
20. Hu L., Wang J., Zhou X., Xiong Z., Zhao J., Yu R., Huang F., Zhang H., Chen L.. **Exosomes derived from human adipose mensenchymal stem cells accelerates cutaneous wound healing via optimizing the characteristics of fibroblasts**. *Sci. Rep.* (2016) **6** 32993. DOI: 10.1038/srep32993
21. Van Dongen J.A., Harmsen M.C., Van der Lei B., Stevens H.P.. **Augmentation of Dermal Wound Healing by Adipose Tissue-Derived Stromal Cells (ASC)**. *Bioengineering* (2018) **5**. DOI: 10.3390/bioengineering5040091
22. Gebert L.F.R., Macrae I.J.. **Regulation of microRNA function in animals**. *Nat. Rev. Mol. Cell Biol.* (2018) **20** 21-37. DOI: 10.1038/s41580-018-0045-7
23. O’Brien J., Hayder H., Zayed Y., Peng C.. **Overview of MicroRNA Biogenesis, Mechanisms of Actions, and Circulation**. *Front. Endocrinol.* (2018) **9** 402. DOI: 10.3389/fendo.2018.00402
24. Makarova J.A., Shkurnikov M.U., Wicklein D., Lange T., Samatov T.R., Turchinovich A.A., Tonevitsky A.G.. **Intracellular and extracellular microRNA: An update on localization and biological role**. *Prog. Histochem. Cytochem.* (2016) **51** 33-49. DOI: 10.1016/j.proghi.2016.06.001
25. El-Serafi A.T., Hayat M.. *Epigenetic Modifiers and Stem Cell Differentiation, in Stem Cells and Cancer Stem Cells* (2012) 147-154
26. Yi R., Pasolli H.A., Landthaler M., Hafner M., Ojo T., Sheridan R., Sander C., O’Carroll D., Stoffel M., Tuschl T.. **DGCR8-dependent microRNA biogenesis is essential for skin development**. *Proc. Natl. Acad. Sci. USA* (2009) **106** 498-502. DOI: 10.1073/pnas.0810766105
27. Andl T., Murchison E.P., Liu F., Zhang Y., Yunta-Gonzalez M., Tobias J.W., Andl C.D., Seykora J.T., Hannon G.J., Millar S.E.. **The miRNA-processing enzyme dicer is essential for the morphogenesis and maintenance of hair follicles**. *Curr. Biology* (2006) **16** 1041-1049. DOI: 10.1016/j.cub.2006.04.005
28. Lerman G., Sharon M., Leibowitz-Amit R., Sidi Y., Avni D.. **The crosstalk between IL-22 signaling and miR-197 in human keratinocytes**. *PLoS ONE* (2014) **9**. DOI: 10.1371/journal.pone.0107467
29. Peng H., Kaplan N., Hamanaka R.B., Katsnelson J., Blatt H., Yang W., Hao L., Bryar P.J., Johnson R.S., Getsios S.. **microRNA-31/factor-inhibiting hypoxia-inducible factor 1 nexus regulates keratinocyte differentiation**. *Proc. Natl. Acad. Sci. USA* (2012) **109** 14030-14034. DOI: 10.1073/pnas.1111292109
30. Yi R., Poy M.N., Stoffel M., Fuchs E.. **A skin microRNA promotes differentiation by repressing ‘stemness’**. *Nature* (2008) **452** 225-229. DOI: 10.1038/nature06642
31. Banerjee J., Chan Y.C., Sen C.K.. **MicroRNAs in skin and wound healing**. *Physiol. Genom.* (2011) **43** 543-556. DOI: 10.1152/physiolgenomics.00157.2010
32. Chen J., Deng S., Zhang S., Chen Z., Wu S., Cai X., Yang X., Guo B., Peng Q.. **The role of miRNAs in the differentiation of adipose-derived stem cells**. *Curr. Stem Cell Res. Ther.* (2014) **9** 268-279. DOI: 10.2174/1574888X09666140213203309
33. Yang C., Luo M., Chen Y., You M., Chen Q.. **MicroRNAs as Important Regulators Mediate the Multiple Differentiation of Mesenchymal Stromal Cells**. *Front. Cell Dev. Biol.* (2021) **9** 619842. DOI: 10.3389/fcell.2021.619842
34. Alghfeli L., Parambath D., Eldeen L.A.T., El-Serafi I., El-Serafi A.T.. **Non-additive effect of the DNA methylation inhibitor, 5-Aza-dC, and glass as a culture surface on osteogenic differentiation**. *Heliyon* (2022) **8** e12433. DOI: 10.1016/j.heliyon.2022.e12433
35. Lizio M., Harshbarger J., Shimoji H., Severin J., Kasukawa T., Sahin S., Abugessaisa I., Fukuda S., Hori F.. **Gateways to the FANTOM5 promoter level mammalian expression atlas**. *Genome Biol.* (2015) **16** 22. DOI: 10.1186/s13059-014-0560-6
36. Jin Q., Li P., Yuan K., Zhao F., Zhu X., Zhang P., Huang Z.. **Extracellular vesicles derived from human dental pulp stem cells promote osteogenesis of adipose-derived stem cells via the MAPK pathway**. *J. Tissue Eng.* (2020) **11** 2041731420975569. DOI: 10.1177/2041731420975569
37. Zhu Z., Guo L., Yeltai N., Xu H., Zhang Y.. **Chemokine (C-C motif) ligand 2-enhanced adipogenesis and angiogenesis of human adipose-derived stem cell and human umbilical vein endothelial cell co-culture system in adipose tissue engineering**. *J. Tissue Eng. Regen. Med.* (2022) **16** 163-176. DOI: 10.1002/term.3264
38. Chavez-Munoz C., Nguyen K.T., Xu W., Hong S.-J., Mustoe T.A., Galiano R.D.. **Transdifferentiation of adipose-derived stem cells into keratinocyte-like cells: Engineering a stratified epidermis**. *PLoS ONE* (2013) **8**. DOI: 10.1371/journal.pone.0080587
39. Pellegrini G., Dellambra E., Golisano O., Martinelli E., Fantozzi I., Bondanza S., Ponzin D., McKeon F., De Luca M.. **p63 identifies keratinocyte stem cells**. *Proc. Natl. Acad. Sci. USA* (2001) **98** 3156-3161. DOI: 10.1073/pnas.061032098
40. Koster M.I., Roop D.R.. **Roop, The role of p63 in development and differentiation of the epidermis**. *J. Dermatol. Sci.* (2004) **34** 3-9. DOI: 10.1016/j.jdermsci.2003.10.003
41. Truong A.B., Kretz M., Ridky T.W., Kimmel R., Khavari P.A.. **p63 regulates proliferation and differentiation of developmentally mature keratinocytes**. *Genes Dev.* (2006) **20** 3185-3197. DOI: 10.1101/gad.1463206
42. dos Santos J.F., Borçari N.R., Araújo M.D.S., Nunes V.A.. **Mesenchymal stem cells differentiate into keratinocytes and express epidermal kallikreins: Towards an in vitro model of human epidermis**. *J. Cell. Biochem.* (2019) **120** 13141-13155. DOI: 10.1002/jcb.28589
43. Cheng Y., Qin K., Huang N., Zhou Z., Xiong H., Zhao J., Zhang Y., Yu S.. **Cytokeratin 18 regulates the transcription and alternative splicing of apoptoticrelated genes and pathways in HeLa cells**. *Oncol. Rep.* (2019) **42** 301-312. PMID: 31115582
44. Păunescu V., Deak E., Herman D., Siska I.R., T˘anasie G., Bunu C., Anghel S., Tatu C.A., Oprea T., Henschler R.. **In vitro differentiation of human mesenchymal stem cells to epithelial lineage**. *J. Cell. Mol. Med.* (2007) **11** 502-508. DOI: 10.1111/j.1582-4934.2007.00041.x
45. Alam H., Sehgal L., Kundu S.T., Dalal S.N., Vaidya M.M.. **Novel function of keratins 5 and 14 in proliferation and differentiation of stratified epithelial cells**. *Mol. Biol. Cell* (2011) **22** 4068-4078. DOI: 10.1091/mbc.e10-08-0703
46. Maeda Y., Hasegawa T., Wada A., Fukai T., Iida H., Sakamoto A., Ikeda S.. **Adipose-derived stem cells express higher levels of type VII collagen under specific culture conditions**. *Arch. Dermatol. Res.* (2017) **309** 843-849. DOI: 10.1007/s00403-017-1781-9
47. Xu X., Li L., Wang C., Liu Y., Chen C., Yan J., Ding H., Tang S.-Y.. **The expansion of autologous adipose-derived stem cells in vitro for the functional reconstruction of nasal mucosal tissue**. *Cell Biosci.* (2015) **5** 54. DOI: 10.1186/s13578-015-0045-7
48. Bayraktar R., Van Roosbroeck K., Calin G.A.. **Cell-to-cell communication: MicroRNAs as hormones**. *Mol. Oncol.* (2017) **11** 1673-1686. DOI: 10.1002/1878-0261.12144
49. Conti I., Varano G., Simioni C., Laface I., Milani D., Rimondi E., Neri L.M.. **miRNAs as Influencers of Cell-Cell Communication in Tumor Microenvironment**. *Cells* (2020) **9**. DOI: 10.3390/cells9010220
50. Valadi H., Ekström K., Bossios A., Sjöstrand M., Lee J.J., Lötvall J.O.. **Exosome-mediated transfer of mRNAs and microRNAs is a novel mechanism of genetic exchange between cells**. *Nat. Cell Biol.* (2007) **9** 654-659. DOI: 10.1038/ncb1596
51. Veith A.P., Henderson K., Spencer A., Sligar A.D., Baker A.B.. **Therapeutic strategies for enhancing angiogenesis in wound healing**. *Adv. Drug Deliv. Rev.* (2018) **146** 97-125. DOI: 10.1016/j.addr.2018.09.010
52. Ma X., Li L., Jia T., Chen M., Liu G., Li C., Li N., Yang D.. **miR-203a controls keratinocyte proliferation and differentiation via targeting the stemness-associated factor DeltaNp63 and establishing a regulatory circuit with SNAI2**. *Biochem. Biophys. Res. Commun.* (2017) **491** 241-249. DOI: 10.1016/j.bbrc.2017.07.131
53. Hildebrand J., Rütze M., Walz N., Gallinat S., Wenck H., Deppert W., Grundhoff A., Knott A.. **A comprehensive analysis of microRNA expression during human keratinocyte differentiation in vitro and in vivo**. *J. Investig. Dermatol.* (2011) **131** 20-29. DOI: 10.1038/jid.2010.268
54. Lawrence P., Ceccoli J.. **Advances in the Application and Impact of MicroRNAs as Therapies for Skin Disease**. *BioDrugs* (2017) **31** 423-438. DOI: 10.1007/s40259-017-0243-4
55. Wu J., Li X., Li D., Ren X., Li Y., Herter E.K., Qian M., Toma M.-A., Wintler A.-M., Sérézal I.G.. **MicroRNA-34 Family Enhances Wound Inflammation by Targeting LGR4**. *J. Investig. Dermatol.* (2020) **140** 465-476.e11. DOI: 10.1016/j.jid.2019.07.694
56. Deverman B.E., Patterson P.H.. **Cytokines and CNS Development**. *Neuron* (2009) **64** 61-78. DOI: 10.1016/j.neuron.2009.09.002
57. Aurora A.B., Olson E.N.. **Immune Modulation of Stem Cells and Regeneration**. *Cell Stem Cell* (2014) **15** 14-25. DOI: 10.1016/j.stem.2014.06.009
58. El-Serafi A.T., El-Serafi I., Steinvall I., Sjöberg F., Elmasry M.. **A Systematic Review of Keratinocyte Secretions: A Regenerative Perspective**. *Int. J. Mol. Sci.* (2022) **23**. DOI: 10.3390/ijms23147934
59. Santos J., Dalla P.V., Milthorpe B.K.. **A Molecular Analysis of Cytokine Content across Extracellular Vesicles, Secretions, and Intracellular Space from Different Site-Specific Adipose-Derived Stem Cells**. *Int. J. Mol. Sci.* (2021) **23**. DOI: 10.3390/ijms23010397
60. Blumenberg M.. **Profiling and metaanalysis of epidermal keratinocytes responses to epidermal growth factor**. *BMC Genom.* (2013) **14**. DOI: 10.1186/1471-2164-14-85
61. Choi S.Y., Lee Y.J., Kim J.M., Kang H.J., Cho S.H., Chang S.E.. **Epidermal Growth Factor Relieves Inflammatory Signals in Staphylococcus aureus-Treated Human Epidermal Keratinocytes and Atopic Dermatitis-Like Skin Lesions in Nc/Nga Mice**. *BioMed Res. Int.* (2018) **2018** 9439182. DOI: 10.1155/2018/9439182
62. Barrientos S., Stojadinovic O., Golinko M.S., Brem H., Tomic-Canic M.. **Growth factors and cytokines in wound healing**. *Wound Repair Regen.* (2008) **16** 585-601. DOI: 10.1111/j.1524-475X.2008.00410.x
63. Tomic-Canic M., Komine M., Freedberg I.M., Blumenberg M.. **Epidermal signal transduction and transcription factor activation in activated keratinocytes**. *J. Dermatol. Sci.* (1998) **17** 167-181. DOI: 10.1016/S0923-1811(98)00016-4
64. Mao Y., Ma J., Xia Y., Xie X.. **The Overexpression of Epidermal Growth Factor (EGF) in HaCaT Cells Promotes the Proliferation, Migration, Invasion and Transdifferentiation to Epidermal Stem Cell Immunophenotyping of Adipose-Derived Stem Cells (ADSCs)**. *Int. J. Stem Cells* (2020) **13** 93-103. DOI: 10.15283/ijsc18146
65. Chen J.D., Kim J.P., Zhang K., Sarret Y., Wynn K.C., Kramer R.H., Woodley D.T.. **Epidermal growth factor (EGF) promotes human keratinocyte locomotion on collagen by increasing the alpha 2 integrin subunit**. *Exp. Cell Res.* (1993) **209** 216-223. DOI: 10.1006/excr.1993.1304
66. Cavalli G., Colafrancesco S., Emmi G., Imazio M., Lopalco G., Maggio M.C., Sota J., Dinarello C.A.. **Interleukin 1alpha: A comprehensive review on the role of IL-1alpha in the pathogenesis and treatment of autoimmune and inflammatory diseases**. *Autoimmun. Rev.* (2021) **20** 102763. DOI: 10.1016/j.autrev.2021.102763
67. Xie H., Lim B., Lodish H.F.. **MicroRNAs induced during adipogenesis that accelerate fat cell development are downregulated in obesity**. *Diabetes* (2009) **58** 1050-1057. DOI: 10.2337/db08-1299
68. Mao L., Liu S., Hu L., Jia L., Wang H., Guo M., Chen C., Liu Y., Xu L.. **miR-30 Family: A Promising Regulator in Development and Disease**. *BioMed Res. Int.* (2018) **2018** 9623412. DOI: 10.1155/2018/9623412
69. Jiang X., Xu C., Lei F., Liao M., Wang W., Xu N., Zhang Y., Xie W.. **MiR-30a targets IL-1alpha and regulates islet functions as an inflammation buffer and response factor**. *Sci. Rep.* (2017) **7** 5270. DOI: 10.1038/s41598-017-05560-1
70. Fu X., Shen Y., Wang W., Li X.. **MiR-30a-5p ameliorates spinal cord injury-induced inflammatory responses and oxidative stress by targeting Neurod 1 through MAPK/ERK signalling**. *Clin. Exp. Pharmacol. Physiol.* (2017) **45** 68-74. DOI: 10.1111/1440-1681.12856
71. Valinezhad Orang A., Safaralizadeh R., Kazemzadeh-Bavili M.. **Mechanisms of miRNA-Mediated Gene Regulation from Common Downregulation to mRNA-Specific Upregulation**. *Int. J. Genom.* (2014) **2014** 970607. DOI: 10.1155/2014/970607
72. Lee A.-Y.. **The Role of MicroRNAs in Epidermal Barrier**. *Int. J. Mol. Sci.* (2020) **21**. DOI: 10.3390/ijms21165781
73. Neu J., Dziunycz P.J., Dzung A., Lefort K., Falke M., Denzler R., Freiberger S.N., Iotzova-Weiss G., Kuzmanov A., Levesque M.P.. **miR-181a decelerates proliferation in cutaneous squamous cell carcinoma by targeting the proto-oncogene KRAS**. *PLoS ONE* (2017) **12**. DOI: 10.1371/journal.pone.0185028
74. Lu W., Han L., Su L., Zhao J., Zhang Y., Zhang S., Zhao B., Miao J.. **A 3′UTR-associated RNA, FLJ11812 maintains stemness of human embryonic stem cells by targeting miR-4459**. *Stem Cells Dev.* (2015) **24** 1133-1140. DOI: 10.1089/scd.2014.0353
75. Jang T.H., Huang W.C., Tung S.L., Lin S.C., Chen P.M., Cho C.Y., Yang Y.Y., Yen T.C., Lo G.H., Chuang S.E.. **MicroRNA-485-5p targets keratin 17 to regulate oral cancer stemness and chemoresistance via the integrin/FAK/Src/ERK/beta-catenin pathway**. *J. Biomed. Sci.* (2022) **29** 42. DOI: 10.1186/s12929-022-00824-z
76. Livak K.J., Schmittgen T.D.. **Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method**. *Methods* (2001) **25** 402-408. DOI: 10.1006/meth.2001.1262
77. Sticht C., De La Torre C., Parveen A., Gretz N.. **miRWalk: An online resource for prediction of microRNA binding sites**. *PLoS ONE* (2018) **13**. DOI: 10.1371/journal.pone.0206239
78. Agarwal V., Bell G.W., Nam J.-W., Bartel D.P.. **Predicting effective microRNA target sites in mammalian mRNAs**. *eLife* (2015) **4** e05005. DOI: 10.7554/eLife.05005
79. Wong N., Wang X.. **miRDB: An online resource for microRNA target prediction and functional annotations**. *Nucleic Acids Res.* (2014) **43** D146-D152. DOI: 10.1093/nar/gku1104
80. Hsu S.D., Lin F.M., Wu W.Y., Liang C., Huang W.C., Chan W.L., Tsai W.T., Chen G.Z., Lee C.J., Chiu C.M.. **miRTarBase: A database curates experimentally validated microRNA-target interactions**. *Nucleic Acids Res.* (2011) **39** D163-D169. DOI: 10.1093/nar/gkq1107
81. Shannon P., Markiel A., Ozier O., Baliga N.S., Wang J.T., Ramage D., Amin N., Schwikowski B., Ideker T.. **Cytoscape: A software environment for integrated models of biomolecular interaction networks**. *Genome Res.* (2003) **13** 2498-2504. DOI: 10.1101/gr.1239303
82. Yu G., He Q.-Y.. **ReactomePA: An R/Bioconductor package for reactome pathway analysis and visualization**. *Mol. Biosyst.* (2015) **12** 477-479. DOI: 10.1039/C5MB00663E
83. Zhang Y., Roos M., Himburg H., Termini C.M., Quarmyne M., Li M., Zhao L., Kan J., Fang T., Yan X.. **PTPsigma inhibitors promote hematopoietic stem cell regeneration**. *Nat. Commun.* (2019) **10** 3667. DOI: 10.1038/s41467-019-11490-5
84. Omori-Miyake M., Yamashita M., Tsunemi Y., Kawashima M., Yagi J.. **In vitro assessment of IL-4- or IL-13-mediated changes in the structural components of keratinocytes in mice and humans**. *J. Investig. Dermatol.* (2014) **134** 1342-1350. DOI: 10.1038/jid.2013.503
85. Herfs M., Hubert P., Suarez-Carmona M., Reschner A., Saussez S., Berx G., Savagner P., Boniver J., Delvenne P.. **Regulation of p63 isoforms by snail and slug transcription factors in human squamous cell carcinoma**. *Am. J. Pathol.* (2010) **176** 1941-1949. DOI: 10.2353/ajpath.2010.090804
86. Lin K.Y., Zhang X.J., Feng D.D., Zhang H., Zeng C.W., Han B.W., Zhou A.D., Qu L.H., Xu L., Chen Y.Q.. **miR-125b, a target of CDX2, regulates cell differentiation through repression of the core binding factor in hematopoietic malignancies**. *J. Biol. Chem.* (2011) **286** 38253-38263. DOI: 10.1074/jbc.M111.269670
|
---
title: Measurement of Postoperative Quality of Pain in Abdominoplasty Patients—An
Outcome Oriented Prospective Study
authors:
- Sascha Wellenbrock
- Matthias Michael Aitzetmüller
- Marie-Luise Klietz
- Philipp Wiebringhaus
- Gabriel Djedovic
- Tobias Hirsch
- Ulrich M. Rieger
journal: Journal of Clinical Medicine
year: 2023
pmcid: PMC10002866
doi: 10.3390/jcm12051745
license: CC BY 4.0
---
# Measurement of Postoperative Quality of Pain in Abdominoplasty Patients—An Outcome Oriented Prospective Study
## Abstract
[1] Background: Postoperative pain is a frequently underestimated complication significantly influencing surgical outcome and patient satisfaction. While abdominoplasty is one of the most commonly performed plastic surgery procedures, studies investigating postoperative pain are limited in current literature. [ 2] Methods: *In this* prospective study, 55 subjects who underwent horizontal abdominoplasty were included. Pain assessment was performed by using the standardized questionnaire of the Benchmark Quality Assurance in Postoperative Pain Management (QUIPS). Surgical, process and outcome parameters were then used for subgroup analysis. [ 3] Results: We found a significantly decreased minimal pain level in patients with high resection weight compared to the low resection weight group ($$p \leq 0.01$$ *). Additionally, Spearman correlation shows significant negative correlation between resection weight and the parameter “Minimal pain since surgery” (rs = −0.332; $$p \leq 0.013$$). Furthermore, average mood is impaired in the low weight resection group, indicating a statistical tendency ($$p \leq 0.06$$ and a Χ2 = 3.56). We found statistically significant higher maximum reported pain scores (rs = 0.271; $$p \leq 0.045$$) in elderly patients. Patients with shorter surgery showed a statistically significant (Χ2 = 4.61, $$p \leq 0.03$$) increased claim for painkillers. Moreover, “mood impairment after surgery” shows a dramatic trend to be enhanced in the group with shorter OP duration (Χ2 = 3.56, $$p \leq 0.06$$). [ 4] Conclusions: While QUIPS has proven to be a useful tool for the evaluation of postoperative pain therapy after abdominoplasty, only continuous re-evaluation of pain therapy is a prerequisite for constant improvement of postoperative pain management and may be the first approach to develop a procedure-specific pain guideline for abdominoplasty. Despite a high satisfaction score, we detected a subpopulation with inadequate pain management in elderly patients, patients with low resection weight and a short duration of surgery.
## 1. Introduction
Being essential for postoperative complications, morbidity, mortality as well as rehabilitation capacity, guideline-based pain therapy has become an integral part for almost all surgical disciplines [1].
While not only perioperative morbidity was found to be reduced by adequate pain medication, several studies describe a significant decrease in complications with a verifiable reduction of hospitalization days [1,2].
Therefore, postoperative pain management is essential not only for individual patients but chronification of underestimated postoperative pain represents an economic burden, with enormous potential for optimization.
Pain management can be divided into non-medicinal and medicinal factors.
While non-medicinal factors include psychological and physical procedures, such as the application of cold to reduce the swelling of an extremity after postoperative decongestion of an extremity after surgery, medical factors mainly focus on systemic pharmacotherapy.
Among these, based on international guidelines, treatment of severe to moderate pain should be based on a combination of opioids (tramadol, piritramide) and non-opioid analgesics (paracetamol, metamizole, NSAIDs, COX-2 inhibitors) [3].
Although this seems to be standardized, many studies have shown insufficient pain management [4,5]. For further standardized assessment and improvement, the “Quality Improvement in Postoperative Pain Therapy” (abbreviated as follows: “QUIPS”) as an interdisciplinary project was initiated. Being the world’s largest acute pain registry and including data on process and quality outcomes, this system allows collection, evaluation and improvement of acute pain therapy in participating institutions [6].
Although it is undoubted, that adequate pain management is essential for individual outcome, there exist almost no studies evaluating postoperative pain in plastic surgery. Especially in semi-elective surgeries, such as body contouring surgeries, characterized by large wound areas, postoperative well-being can lead to faster mobilization and reduction of hospitalization time. Nevertheless, hardly any literature is available on this topic.
Single case reports describe the existence of neuropathic pain syndromes of the N. iliohypogastricus and cutaneous femoris lateralis after abdominoplasty and their avoidability [7]. Feng et al. describe the reduction of pain by combination of local nerve blocks during abdominoplasty surgery [8].
Regarding pain medication, a recommendation of reduced opioid consumption after abdominal wall surgery can be found, but there exist no concrete guidelines and this recommendation has not been evaluated [9].
To sum up, although postoperative pain management has been excessively described to be of utmost importance for outcome and well-being, no standardized study within body contouring patients has been carried out up to now.
Therefore, we used QUIPS in abdominoplasty patients for analyzing pain characteristics as well as to define risk factors for enhanced postoperative pain.
## 2. Materials and Methods
This study was carried out following the guidelines of the declaration of Helsinki as well as by the dean of the university.
## 2.1. Inclusion and Exclusion Criteria
All patients undergoing abdominoplasty according to Pitanguy at the Department for Plastic and Aesthetic Surgery, Reconstructive and Hand Surgery at the Markus Hospital in Frankfurt am Main from January 2010 to December 2015 were included in this study.
For the study, all patients were excluded who underwent combination or revision procedures such as autologous breast reconstruction or repair of rectus diastasis.
## 2.2. Data Collection
A standardized pain questionnaire (QUIPS) was carried out on postoperative day one by a single study nurse focusing on outcome (Appendix A) and processing parameters (Appendix B) using a Numeric Rating Scale from 0 to 10, or a dichotomous Yes/No categorization. This questionnaire was filled out manually under surveillance. Additionally, preoperative anesthesiologic assessments were screened for general data such as age, sex, weight and for specific risk factors, comorbidities and ASA score. Surgical protocols were screened for resection weight as well as for surgery time.
## 2.3. Surgical Procedure
All surgical procedures were carried out by one senior doctor with one or two residents. The surgical procedure was standardized to prevent any technical related bias: Preoperatively, the patient is marked in a standing position to define the resection lines. After proper positioning, the surgical area is sterilely covered. The skin incision is made with the scalpel and the subcutaneous preparation by using the monopolar diathermy. The belly button is incised and sutured cranially with silk as holding suture. Epifascial dissection of the fat-skin soft tissues below and above the umbilicus up to the xiphoid while sparing the lateral sub- and intercostal perforator vessels. Hereby no focus is given on nerve sparing. Insertion of wound drains and drainage at the mons pubis. Collapsing the patient at the hip and re-defining the resection area. Resection of the skin fat flap and adaptation of the skin and tissue with Vicryl 2-0 and 3-0.Placement of the new umbilical position and suturing of the umbilicus with Vicryl 4-0 subcutaneous and Prolene 4-0.Wound closure with continuous intradermal Biosyn 3-0 suture and sterile wound dressing, abdominal belt.
## 2.4. Pain Mangement
All patients received pain medication via a standardized protocol following the official German guidelines for pain management (Appendix C) [3].
## 2.5. Statistical Analysis
Data analysis was performed using SPSS version 22.0 (IBM Corporation, New York, NY, USA). Initially, the database created by QUIPS was completed with the operation- and patient-related data. After conversion, analysis of descriptive data was performed. All data are given as mean and standard deviation (=SD). Nominal-distributed data were analyzed using Pearson’s chi-square test, Fisher’s exact test, and Spearman-rho correlation.
To analyze variables and individual subgroups of this population, a univariate and multivariate correlation analysis as well as the Mann Whitney-U test and the Kruskal Wallis test were used. The median was used for division of groups. A p level of <0.05 was considered as statistically significant.
## 3.1. Demography
In total, 268 patients underwent abdominoplasty within the given timeframe. Further, 110 were excluded due to surgical technique. Of the remaining 158 patients, 55 showed a complete QUIPS and gave written consent for participation. Among those, 41 ($75\%$) were female and 14 ($25\%$) were male, aged between 21 and 67 years. Mean age and mean height was 42.93 ± 9.9 and 169.22 cm ± 8.13 cm, retrospectively. Average weight of patients was 87.05 kg ± 19.73 kg.
## 3.2. Surgical Procedures
Average resection weight was 2913 g ± 2226 g and average duration of surgery was 129.49 ± 37.48 min, with 215 min representing the longest and 56 min the minimal surgical time.
## 3.3. Preoperative Measurements
Patients were preoperatively classified using the ASA sore. Thereby, five ($9\%$) subjects were categorized as ASA-I, 42 ($76\%$) as ASA-II, and eight ($15\%$) as ASA-III. No ASA-IV or V subjects underwent surgery. In total, seven ($13\%$) patients stated regular intake of painkillers before surgery, due to chronic diseases.
## 3.4. QUIPS Outcome Parameters Overall
Table 1 depicts the overall outcome results in our population.
Mean pain on exertion was reported as 4.42 ± 1.54, with the maximum pain being 5.35 ± 2.04 and the minimum pain 1.95 ± 1.43. Thirty-eight of the 55 patients ($69\%$) exceeded the pain level of 4, which normally is considered as the tolerance pain threshold for demanding pain killers. Patient satisfaction was reported to be 11.95 ± 3.03 in average.
In terms of mobility, 34 ($62\%$) patients reported being significantly limited due to pain. When breathing or coughing, 27 ($49\%$) mentioned pain while they in- or exhaled. Thirteen of the respondents ($24\%$) felt their sleep was disturbed and 12 ($23\%$) reported their mood being affected by postoperative pain. Twenty-six ($47\%$) of all respondents reported postoperative pain fatigue. Both nausea and vomiting were mentioned by only 12 ($23\%$) and nine ($13\%$), respectively. Chronic pain was previously described by seven ($13\%$) of the total collective.
Nevertheless, only 13 ($24\%$) demanded extra painkillers. In addition, 26 ($47\%$) of all respondents felt postoperative pain fatigue.
## 3.5. QUIPS Process Parameters Overall
Preoperatively, midazolam 7.5 mg per os was offered to all patients as a sedative but was taken only by six patients ($11\%$).
For intraoperative pain relief, an opioid (Sufentanil) was used in all except five ($9\%$) cases. Non-opioids were used in seven patients ($13\%$).
In the postoperative care unit, opioids were used in 53 ($96\%$) patients and in 35 ($64\%$) patients, non-opioids were injected intravenously.
Postoperatively, midazolam was injected in 35 ($64\%$) cases. Diclofenac was used as a second-line agent in five cases ($9\%$). In addition, 15 ($27\%$) patients did not require any analgesics. In the majority of subjects ($$n = 31$$, $56\%$), no opioid was needed. However, in 24 patients ($44\%$), Piritramid 7.5 mg i.v. was used postoperatively.
## 3.6. Influence of Surgical Parameters on Postoperative Pain Outcome Parameters
For a profound analysis of the quality of postoperative pain therapy, subpopulations were created based on surgical parameters and patient characteristics. Therefore, means were used as dividing factors: Resection weight: Mean weight is 2180 g, Age: Mean age is 43 years, Duration of surgery: Mean operation time 125 min.
## 3.6.1. Resection Weight
When considering the factor “resection weight”, we found significant decreased minimal pain in patients with high resection weight compared to the low resection weight group ($$p \leq 0.01$$) as shown in Table 2. Additionally, Spearman correlation analysis shows a significant negative correlation between resection weight and the parameter “Minimal pain since the Surgery” (Spearman Rho coefficient rs = −0.332; significance value $$p \leq 0.013$$ *).
Additionally average mood was impaired in the low weight resection group. A p-value of 0.06 and a Χ2-value of 3.56 indicate a trend without being statistically significant.
## 3.6.2. Age
With a range of 21 to 67 years, the average age is 42.93 ± 9.9 years with a median of 43 years.
We found statistically significant higher maximum reported pain scores (Spearman Rho coefficient rs = 0.271; significance value $$p \leq 0.045$$ *) in older patients, indicating enhanced pain within this group as shown in Table 3.
## 3.6.3. Duration of Surgery
The average operation time in this study was 129.49 ± 37.48 min, with a range of 159 min and a median of 125 min.
Table 4 shows that Patients with a shorter surgery had a statistically significant (Χ2 = 4.61, $$p \leq 0.03$$ *) increased claim for painkillers. Additionally, “mood impairment after surgery” showed a dramatic trend to be enhanced in the group with shorter OP duration (Χ2 = 3.56, $$p \leq 0.06$$).
## 4. Discussion
Aimed at a redefinition of the body contour, abdominoplasty is performed by wide undermining of tissue of the abdominal wall with its high density of thoracolumbal sensitive nerves [10]. Due to this fact, it remains uncertain how much nerve injury is caused by wide preparation. Pogatzski-Zahn emphasize that intraoperative nerve irritation can cause chronic pain syndromes [11] and nerve injury during this procedure represents an underestimated problem [12].
Despite SOPs and guidelines, as well as an improved patient education, pain is constantly considered as the fifth vital sign [13] and its postoperative management is far from being sufficient [5,6].
With a few exceptions in plastic surgery, there exists no literature of postoperative pain management in standard procedures. One of the reasonable tools for pain relief, published with a small cohort, is additional regional nerve block of the area of interest (e.g., breast [14] or abdomen [15]) or the use of lidocaine-infusing pain pumps [16]). These steps lead to reduced hospital stay, overall pain reduction and a reduced pain medication compared to the control group [17].
Nevertheless, literature of an analysis of pain quality in abdominoplasty with its outcome parameters as a benchmarking tool for evaluation is missing.
Therefore, for the first time, we implemented QUIPS for the analysis of plastic surgery pain management.
In our analysis, the mean maximal pain intensity overall was 5.35 out of 10 in a NRS. Pain levels in 38 subjects ($69\%$) were above a value of 4 and according to the S3-guidelines of perioperative pain treatment [18], therefore needed to be addressed for prevention of long-term functional impairment.
By comparing our cohort result of maximal pain intensity with other different common procedures, such as an appendectomy 5.20 or a functional endoscopic sinusitis surgery 3.96, this level shows a relatively high maximum pain level. Nevertheless, in comparison to traumatological procedures (such as the cruciate ligament-plasty up to 6.0 out of 10), it shows lower maximum pain intensity [19].
We used the median to split participants into two subpopulations regarding their resection-weight, age and duration of surgery, being consistent with our clinically experience.
The cohort with medical indication can be divided into patients with a relevant, functional impairing dermatochalasis of the abdomen, a longer operation time, a higher ASA status and generally the younger patient after bariatric surgery or self-induced massive weight loss.
In contrast to the previous mentioned, the aesthetic patient population comes along with a moderate low resection weight, less ptosis of skin, normally younger women after pregnancy asking for a mommy makeover or the “best ager”, who are older but healthier with a profound focus on their outward appearance and self-motivated initiative for a tummy tuck.
A direct comparison between those two groups is desired and needs further research.
Nevertheless, we evaluated the outcome parameters in relation to the patient specific parameters of our subgroups.
A higher resection weight correlates with a higher bodyweight and this is regarded as a predictor for a higher complication rate in abdominoplasty [20,21].
Interestingly, none of these studies analyzed the influence of the factor “resection weight” itself (in our study ranging from 610 g up to 9600 g).
In contrast to our prediction, our observations show that high resection weight goes along with significantly less pain markers and the mood impairment of patients with low resection weights.
There are no data in literature which confirm those findings for this procedure.
In reduction mammaplasty, Strong et al. found that patients with higher resection weights have significantly less pain than those with low resection rates [22].
A reasonable explanation of our results is the decreased sensibility of the abdominal wall in patients with a large apron of fat and high resection weight. We assume that patients with hanging bulge of skin and fat tissue have a higher basic pain level and they tolerate postoperative surgical pain better than in lower resection weights [23]. We can further postulate that a chronic local hypoxia, with a consecutive increasing lactate level and a lowering of the pH-value in the apron of fat, elevates the excitation threshold of the peripheral nervous system in this local bulge, causing delayed or even failing to trigger an action potential. This hypothesis has been initially drawn by Kim et al., postulating an ischemic-related pain mechanism when showing a significantly elevated lactate level in postoperative wounds [24].
However, one has to be aware that resection weight can be high in patients with massive weight loss with normal BMI.
The correlation analysis shows that older patients have significantly higher maximal pain levels, which is contrary to current literature, emphasizing young age to be a risk factor for postoperative pain [11,25,26,27]. A higher tolerance of pain is awarded to older patients with reduced analgetic consumption, reduced nociceptive activity and lower needs for morphine medication [28]. Elderly patients vary in distribution of medication, metabolism and excretion of pain medication compared to young people [29].
Morphology-wise, older skin shows dermal atrophy with less perfusion, less elastic fibers and less Meissner and Vater Pacini bodies, resulting in a lower tactile and pressure sensibility [30,31]. These factors cause decreased tolerance of shear forces and tension taking place in an abdominoplasty. Young age as a risk factor for developing pain could not be confirmed in our study.
Enlarged duration of surgery is broadly accepted as a risk factor for pain as well [24,32]. Interestingly, patients with lower operation time have significantly higher desire for pain medication and a high tendency of mood disturbance. While those probands with lower operation time mostly have a lower resection weight, reflecting aesthetic abdominoplasty cases, the expectations for this procedure might be higher due to their payment and probably the conditions of their pain-ranking are much stricter than those in the insurance paid comparative group. Additionally, due to shorter surgery-time, subjects mobilize themselves earlier and this might lead to wound tension, with worse pain outcome portrayed by an increased claim for painkillers and mood disturbance [33]. The removal of the suction drain itself causes pain, discomfort and anxiety [34]. The survey took place on the first postoperative day. Patients with shorter duration of surgery are probably more likely to get rid of suction drains earlier in temporary connection to the QUIPS interview, causing high pain intensity.
The statement: “Longer operation time is a predictor for increased postoperative pain” can be rejected in our investigation.
Limitation is the relatively small cohort and a single center study, as well as a relatively small time frame, which needs to be verified in ongoing investigations. Additionally, due to the structure of QUIPS, no conclusion can be drawn about pre-existing chronic disorders influencing postoperative pain medication.
## 5. Conclusions
Abdominoplasty is a standardized procedure that is, due to its large wound area, well suited for the evaluation of pain levels. QUIPS has proven to be a successful tool for the first elevation of postoperative pain quality in abdominoplasty procedures.
Despite a high overall satisfaction score, we detected a subpopulation with inadequate pain management in elderly patients, patients with low resection weight and a short duration of surgery. To what extent newly found risk factors will be deemed suitable to adapt tailored pain management requires further study to pave the way for a procedures-specific pain guideline.
## References
1. Savoia G., Alampi D., Amantea B., Ambrosio F., Arcioni R., Berti M., Bettelli G., Bertini L., Bosco M., Casati A.. **Postoperative pain treatment SIAARTI Recommendations 2010. Short version**. *Minerva Anestesiol.* (2010.0) **76** 657-667. PMID: 20661210
2. Osterbrink J., Evers G.C.. **The influence of nursing measures on incision pain and the use of opioids in the postoperative period**. *Pflege* (2000.0) **13** 306-314. DOI: 10.1024/1012-5302.13.5.306
3. **S3-Leitlinie: Behandlung; Deutsche Interdisziplinäre Vereinigung Für Schmerztherapie**
4. 4.AWMF-Reg.-Nr. 041/001Akuter Perioperativer und Posttraumatischer SchmerzenAWMFBerlin, Germany2007Available online: http://www.awmf.org/leitlinien/detail/ll/001-025.html(accessed on 19 March 2018). *Akuter Perioperativer und Posttraumatischer Schmerzen* (2007.0)
5. Lehmkuhl D., Meissner W., Neugebauer E.A.M.M.. **Evaluation of the “initiative pain-free clinic” for quality improvement in postoperative pain management. A prospective controlled study**. *Schmerz* (2011.0) **25** 508-515. DOI: 10.1007/s00482-011-1054-z
6. Apfelbaum J.L., Chen C., Mehta S.S., Gan A.T.J.. **Postoperative Pain Experience: Results from a National Survey Suggest Postoperative Pain Continues to be Undermanaged**. *Anesth. Analg.* (2003.0) **97** 534-540. DOI: 10.1213/01.ANE.0000068822.10113.9E
7. Meissner W., Komann M., Erlenwein J., Stamer U., Scherag A.. **The Quality of Postoperative Pain Therapy in German Hospitals**. *Dtsch. Arztebl. Int.* (2017.0) **114** 161-167. DOI: 10.3238/arztebl.2017.0161
8. Chowdhry S., Davis J., Boyd T., Choo J., Brooks R.M., Kelishadi S.S., Tutela J.P., Yonick D., Wilhelmi B.J.. **Safe Tummy Tuck: Anatomy and Strategy to Avoid Injury to the Lateral Femoral Cutaneous Nerve During Abdominoplasty**. *Eplasty* (2015.0) **15** e22. PMID: 26171094
9. Feng L.-J.. **Painless abdominoplasty: The efficacy of combined intercostal and pararectus blocks in reducing postoperative pain and recovery time**. *Plast. Reconstr. Surg.* (2010.0) **126** 1723-1732. DOI: 10.1097/PRS.0b013e3181ef8fe5
10. Khansa I., Koogler A., Richards J., Bryant R., Janis J.E.. **Pain Management in Abdominal Wall Reconstruction**. *Plast. Reconstr. Surg. Glob. Open* (2017.0) **5** e1400. DOI: 10.1097/GOX.0000000000001400
11. Rozen W.M., Tran T.M.N., Ashton M.W., Barrington M.J., Ivanusic J.J., Taylor G.I.. **Refining the course of the thoracolumbar nerves: A new understanding of the innervation of the anterior abdominal wall**. *Clin. Anat.* (2008.0) **21** 325-333. DOI: 10.1002/ca.20621
12. Pogatzki-Zahn E.M.. **Prevention and therapy of prolonged, chronic pain after surgery**. *Anästhesiol. Intensiv. Notf. Schmerzther* (2010.0) **45** 496-504. DOI: 10.1055/s-0030-1262479
13. Ducic I., Zakaria H.M., Felder J.M., Arnspiger S.. **Abdominoplasty-related nerve injuries: Systematic review and treatment options**. *Aesthetic Surg. J.* (2014.0) **34** 284-297. DOI: 10.1177/1090820X13516341
14. Phillips D.M.. **Jcaho pain management standards are unveiled**. *JAMA* (2000.0) **284** 428-429. DOI: 10.1001/jama.284.4.423b
15. Woodworth G.E., Ivie R.M.J., Nelson S.M., Walker C.M., Maniker R.B.. **Perioperative Breast Analgesia: A Qualitative Review of Anatomy and Regional Techniques**. *Reg. Anesth. Pain Med.* (2017.0) **42** 609-631. DOI: 10.1097/AAP.0000000000000641
16. Grape S., Kirkham K.R., Albrecht E.. **Transversus abdominis plane block versus local anaesthetic wound infiltration for analgesia after caesarean section: A systematic review and meta-analysis with trial sequential analysis**. *Eur. J. Anaesthesiol.* (2022.0) **39** 244-251. DOI: 10.1097/EJA.0000000000001552
17. Chavez-Abraham V., Barr J.S., Zwiebel P.C.. **The efficacy of a lidocaine-infused pain pump for postoperative analgesia following elective augmentation mammaplasty or abdominoplasty**. *Aesthetic Plast. Surg.* (2011.0) **35** 463-469. DOI: 10.1007/s00266-010-9633-4
18. Sforza M., Andjelkov K., Zaccheddu R., Nagi H., Colic M.. **Transversus abdominis plane block anesthesia in abdominoplasties**. *Plast. Reconstr. Surg.* (2011.0) **128** 529-535. DOI: 10.1097/PRS.0b013e31821e6f51
19. Laubenthal H., Becker M., Neugebauer E.. **Guideline: Treatment of acute perioperative and posttraumatic pain. Updating from the S2- to the S3-level: A preliminary report**. *Anästhesiol. Intensiv. Notf. Schmerzther* (2006.0) **41** 470-472. DOI: 10.1055/s-2006-949507
20. Vastine V.L., Morgan R.F., Williams G.S., Gampper T.J., Drake D.B., Knox L.K., Lin K.Y.. **Wound complications of abdominoplasty in obese patients**. *Ann. Plast. Surg.* (1999.0) **42** 34-39. DOI: 10.1097/00000637-199901000-00006
21. Kim J., Stevenson T.R.. **Abdominoplasty, liposuction of the flanks, and obesity: Analyzing risk factors for seroma formation**. *Plast. Reconstr. Surg.* (2006.0) **117** 773-779. DOI: 10.1097/01.prs.0000200056.57357.3f
22. Strong B., Hall-Findlay E.J.. **How Does Volume of Resection Relate to Symptom Relief for Reduction Mammaplasty Patients?**. *Ann. Plast. Surg.* (2015.0) **75** 376-382. DOI: 10.1097/SAP.0000000000000190
23. Jensen M.P., Gould E.M., Victor T.W., Gammaitoni A.R., White R.E., Galer B.S.. **The relationship of changes in pain quality to pain interference and sleep quality**. *J. Pain* (2010.0) **11** 782-788. DOI: 10.1016/j.jpain.2009.11.009
24. Kim T.J., Freml L., Park S.S., Brennan T.J.. **Lactate concentrations in incisions indicate ischemic-like conditions may contribute to postoperative pain**. *J. Pain* (2007.0) **8** 59-66. DOI: 10.1016/j.jpain.2006.06.003
25. IP H.Y.V., Abrishami A., Peng P.W.H., Wong J., Chung F.. **Predictors of postoperative pain and analgesic consumption: A qualitative systematic review**. *Anesthesiology* (2009.0) **111** 657-677. DOI: 10.1097/ALN.0b013e3181aae87a
26. Gagliese L., Katz J.. **Age differences in postoperative pain are scale dependent: A comparison of measures of pain intensity and quality in younger and older surgical patients**. *Pain* (2003.0) **103** 11-20. DOI: 10.1016/S0304-3959(02)00327-5
27. González-Roldán A.M.M., Terrasa J.L., Sitges C., van der Meulen M., Anton F., Montoya P.. **Age-related changes in pain perception are associated with altered functional connectivity during resting state**. *Front. Aging Neurosci.* (2020.0) **12** 116. DOI: 10.3389/fnagi.2020.00116
28. Lautenbacher S., Peters J.H., Heesen M., Scheel J., Kunz M.. **Age changes in pain perception: A systematic-review and meta-analysis of age effects on pain and tolerance thresholds**. *Neurosci. Biobehav. Rev.* (2017.0) **75** 104-113. DOI: 10.1016/j.neubiorev.2017.01.039
29. Greenblatt D.J., Sellers E.M., Shader R.I.. **Drug therapy: Drug disposition in old age**. *N. Engl. J. Med.* (1982.0) **306** 1081-1088. DOI: 10.1056/NEJM198205063061804
30. Yaar M., Eller M.S., Gilchrest B.A.. **Fifty years of skin aging**. *J. Investig. Dermatol. Symp. Proc.* (2002.0) **7** 51-58. DOI: 10.1046/j.1523-1747.2002.19636.x
31. Green H.A., Drake L.. **Aging, sun damage, and sunscreens**. *Clin. Plast. Surg.* (1993.0) **20** 1-8. DOI: 10.1016/S0094-1298(20)30767-7
32. Gagliese L., Gauthier L.R., Macpherson A.K., Jovellanos M., Chan V.W.S.. **Correlates of postoperative pain and intravenous patient-controlled analgesia use in younger and older surgical patients**. *Pain Med.* (2008.0) **9** 299-314. DOI: 10.1111/j.1526-4637.2008.00426.x
33. Rivas E., Cohen B., Pu X., Xiang L., Saasouh W., Mao G., Minko P., Mosteller L., Volio A., Maheshwari K.. **Pain and opioid consumption and mobilization after surgery: Post hoc analysis of two randomized trials**. *Anesthesiology* (2022.0) **136** 115-126. DOI: 10.1097/ALN.0000000000004037
34. Chao J.W., Raveendran J.A., Maly C., Rogers G., Boyajian M., Oh A.K.. **Closed-Suction Drains after Subcutaneous Mastectomy for Gynecomastia: Do They Reduce Complications?**. *Aesthetic Plast. Surg.* (2017.0) **41** 1291-1294. DOI: 10.1007/s00266-017-0959-z
|
---
title: Evaluation of antioxidant properties of nanoencapsulated sage (Salvia officinalis
L.) extract in biopolymer coating based on whey protein isolate and Qodumeh Shahri
(Lepidium perfoliatum) seed gum to increase the oxidative stability of sunflower
oil
authors:
- Behnaz Safarpour
- Reza E. Kenari
- Jamshid Farmani
journal: Food Science & Nutrition
year: 2022
pmcid: PMC10002883
doi: 10.1002/fsn3.3177
license: CC BY 4.0
---
# Evaluation of antioxidant properties of nanoencapsulated sage (Salvia officinalis L.) extract in biopolymer coating based on whey protein isolate and Qodumeh Shahri (Lepidium perfoliatum) seed gum to increase the oxidative stability of sunflower oil
## Abstract
Sage leaf extract (SLE) is considered an excellent source of bioactive compounds mainly because of its high content of phenolics, widely known as natural antioxidants. This study aimed to compare the performance of free/encapsulated SLE by different coatings in protecting sunflower oil against oxidative deterioration. The coating materials were whey protein isolate and qodumeh seed gum at different ratios (1:0, 1:1, and 0:1). Each nanocapsule was analyzed for particle size, zeta potential, encapsulation efficiency, phenolics release, and SEM images. The total phenolic compounds of SLE were 31.12 mg GA/g. The antioxidant activity of SLE was increased in both DPPH and FRAP assays by increasing extract concentration from 50 to 250 ppm. All nanoparticles exhibited nanometric size, negative zeta potential, encapsulation efficiency higher than $60\%$, and gradual release during storage. The oxidative stability of sunflower oil with or without the incorporation of 250 ppm of free/encapsulated SLE was evaluated during 24 days of storage at 60°C. Peroxide value (PV), thiobarbituric acid value (TBA), oxidative stability index (OSI), color index (CI), and conjugated dienes (CD) were determined. COPM nanoparticles showed the lowest PV, TBA, CI, and CD but both SGUM and WHEY were more effective in delaying oil oxidation than TBHQ and free extract. Higher OSI was observed in oil‐containing nanoparticles with composite coating. Results obtained reinforce the use of whey protein isolate and qodumeh seed gum as a coating for encapsulating SLE to increase the shelf life of sunflower oil as a natural antioxidant.
Nano‐encapsulation is one of the very useful methods for preserving antioxidant compounds, which leads to the controlled release of antioxidant compounds in oil. In our research, Nano‐encapsulation was very effective in preserving and controlling the release of sage extract.
## INTRODUCTION
Sunflower (Helianthus annus) is one of the most important oil crops grown worldwide due to high‐yield oil, and lack of antinutritional factors (Aly et al., 2021; Jafari et al., 2022). Sunflower oil (SFO) is a kind of nutritious vegetable oil that contains more than $85\%$ polyunsaturated fatty acids (PUFA), especially linoleic acid which is used for medical treatment (Meng et al., 2021; Sayyari & Farahmandfar, 2017). The ratio of omega‐3 and omega‐6 fatty acids is prominent for providing cardiovascular and heart health benefits (Aly et al., 2021).
However, due to its fatty acid composition with high PUFA, it is one of the most susceptible to suffering rancidity and oxidation progress. Fat oxidation results in unpleasant flavors, discoloration, changes in texture, nutritional value, shelf life, and appearance of SFO, so synthetic antioxidants such as tert‐butyl hydroquinone (TBHQ), propyl gallate (PG), butylated hydroxytoluene (BHT), and butylated hydroxy anisole (BHA) were used (Razavi & Kenari, 2021). Although synthetic antioxidants are attractive due to their low cost, wide availability, great stability, and effectiveness, their use is limited as they may generate health risks, gastrointestinal tract problem, and cancer risk (Xu et al., 2021). Today, there is growing interest to explore natural antioxidants like plant extracts which provide higher antioxidant activity, and improved sensory properties (Kenari & Razavi, 2022; Wang et al., 2020).
Antioxidant properties of plants are effective in delaying oxidation and rancidity in fats and oils and they have similar activity as chemically synthetic antioxidants (Aly et al., 2021; Wang et al., 2020). Natural extracts from different herbs, such as *Heracleum persicum* (Kenari et al., 2020), *Fumaria parviflora* L. (Razavi & Kenari, 2021), sesame (Esmaeilzadeh Kenari & Razavi, 2022), and *Rosmarinus officinalis* L. (Jafari et al., 2022), are stable for oxidation which is related to the presence of natural phenolic compounds.
Sage (*Salvia officinalis* L.), an evergreen shrub, belongs to the mint family (Labiatae). It is known for its aroma, flavor, and taste. Sage contains a wide array of bioactive compounds like phenolics, terpenoids, and organic acids that have shown antioxidant, antimicrobial, anticancer, and anti‐inflammatory activities (El‐Sayed & Youssef, 2019; Naziruddin et al., 2022). The extraction of bioactive compounds from plant materials with conventional methods such as maceration, shaker, and hydro‐distillation is laborious due to long extraction time, low efficiency, and hazardous solvents (Wrona et al., 2017). Ultrasound‐assisted extraction (UAE) process is a potentially useful technique for the purification and isolation of bioactive compounds. The high‐intensity and high‐frequency sound waves and also their interaction with plant materials distinguish UAE from the conventional methods (Sadat et al., 2021).
The efficiency of plant extracts pertains to biological activities and physicochemical properties. Low stability and water solubility, and the unpleasant taste of plant extract limit their application in food formulation. Encapsulation is a technology for maintaining the biological activities, control release, and bioavailability of bioactive compounds from plant materials which allow their application in different food formulations and preserving their functional properties (Reddy et al., 2022). It also enclosed bioactive compounds from light, oxygen, pH, water, and other adverse conditions (Jamshidi et al., 2020). A range of food‐grade biopolymers is used to create nanoparticles such as polysaccharides, proteins, and a combination of them (Razavi et al., 2021). Seed gums are new and plentiful polysaccharides. The *Lepidium perfoliatum* seed, which is known as Qodumeh Shahri in Iran, produces a high amount of mucilage. It can immobilize and bind a lot of water, and increase the viscosity of foods (Jamshidi et al., 2020). Whey protein isolate is obtained during the production of cheese or casein and it is a by‐product of the dairy industry which is widely used in the food industry because of its functional properties, emulsification, gelatinization, film formation, and solubility in water (Tavares & Noreña, 2019).
Considering that sunflower oil is sensitive to oxidation like other vegetable oils, it is necessary to increase its shelf life by adding natural antioxidants as safe preservatives. The use of extract encapsulation controls the release of antioxidant compounds from the extract during the storage. To the best of our knowledge, studies carried out so far have predominantly focused on using free extracts to increase the shelf life of vegetable oils. Also, no research has been published about the antioxidant activity of the encapsulated sage extract in whey protein isolate and Qodumeh Shahri (Lepidium perfoliatum) seed gum in sunflower oil. Therefore, the present study aimed to evaluate [1] the antioxidant activity of the sage extract, [2] the effect of coating material on the properties of nanocapsules, and [3] the effect of free and nanoencapsulated extract on the extension of oxidative stability of sunflower oil during the accelerated thermal condition.
## Material
The common sage was collected from the local field area near Sari (Mazandaran, Iran) in the summer of 2021. Sunflower oil without antioxidant was purchased from North Agro‐industrial Oil Company. All solvents and chemicals were purchased from Sigma‐Aldrich Company (Sigma). Qodumeh shahri seed gum was purchased from *Reyhan gum* parsian.
## Preparation of sage leaf extract
The leaves of sage were dried immediately after harvesting in a shady place for 1 week and the moisture content was below $10\%$. The dried sage leaves were ground into powder using a mechanical grinder (Habi, Pars‐Khazar). The powder was sieved using a 200‐μm sieve to remove any large pieces. To prepare sage leaf extract, 50 g of sage leaves was mixed with 250 ml of ethanol: water (70:30) solvent. The extraction was done using a ultrasonic bath (6.5l200 H, Dakshin, India) at 35°C for 30 min at a frequency of 35 kHz. The mixture was filtered using Whatman paper No. 1. Then, the solvent was evaporated using a rotary evaporator (RE 120) at 35°C and the final extract was kept at −18°C (Razavi & Kenari, 2021).
## Total phenolic content of sage leaf extract
The total phenolic content (TPC) of sage leaf extract was calculated according to the method reported by Doymaz and Karasu [2018]. Initially, 2.5 ml of Folin–Ciocalteu phenol reagent (0.2 N) was added to 0.5 ml of extract and mixed with 2 ml of Na2CO3 ($7.5\%$). This mixture was kept for 20 min at room temperature in a dark place. After incubation, the absorbance was recorded at 760 nm using a ultraviolet–vis spectrophotometer (Cintra 6, GBS Scientific). The total phenolic content was expressed as a gallic acid calibration curve (Doymaz & Karasu, 2018).
## Determination of antioxidant activity
The antioxidant activity of the extract was determined using 2,2‐diphenyl‐1‐picrylhydrazyl radical scavenging method (DPPH) and ferric reduction antioxidant power (FRAP). Briefly, 0.1 ml of extract and 4.9 ml of DPPH solution (0.1 mM in ethanol) were mixed toughly and held at 25°C for 30 min. Then, the absorbance was recorded at 517 nm and 3 ml of freshly prepared FRAP solution including FeCl3.6H2O (0.02 M in water), TPTZ (0.01 M dissolved in 0.04 M HCL), and acetate buffer (0.3 M, pH = 3.6) at the ratio 1:1:10 was mixed with 10 μl of extract. An increase in absorbance was recorded after 30 min at 593 nm. Antioxidant activity was expressed as mmol/g Trolox (Doymaz & Karasu, 2018).
## Sage leaf extract encapsulation
Whey protein isolate and qodumeh shahri seed gum solution at different ratios (1:0, 1:1, and 0:1) were used as coating materials. Initially, 0.05 g of coating powders was dispersed in deionized water at 30°C and after cooling, mixed overnight to enhance hydration. Then, 10 ml of sage extract was combined with 40 ml of tween 80 and 50 ml of sunflower oil during homogenizing with a magnetic stirrer at 100 rpm for 15 min. After that, the formed emulsion was homogenized again using Ultra‐Turrax homogenizer (IKA Labortechnik) at 15,000 rpm for 10 min followed by adding coating solution to nanoemulsion at a 5:1 ratio (Jafari et al., 2022).
## Properties of encapsulated sage extract
Nanoemulsions were dried using a freeze dryer (SP Scientific) at −50°C and 0.017 mPa for 48 h. The particle size, polydispersity index, and zeta potential of nanoemulsions were measured using a master‐sizer light scattering (Malvern Instrument Ltd.). To evaluate the encapsulation efficiency (EE) of sage extract, 200 mg of different nanoemulsions was mixed with hexane: water: methanol (50:42:8 v/v/v) to destroy the coat of nanocapsules. The surface phenolic content (SPC) and the total phenolic content (TPC) were measured. The EE was calculated using Equation 1: [1] EE%=TPC–SPCTPC×100 The surface morphology of nanoemulsions was examined by SEM (Malvern Instrument Ltd.). Different nanoemulsions were fixed onto double‐sided adhesive carbon tabs mounted on SEM stubs, coated with gold (Kenari et al., 2020).
## Release rate of phenolic compounds
The release rate of phenolic compounds was measured according to the method described by Esmaeilzadeh Kenari et al. [ 2020]. Initially, 20 g of different nanoparticles was poured into separate bottles and kept in an incubator at 60°C for 24 days. Then, 5 ml of phosphate buffer was mixed with 5 g of nanoparticles and centrifuged for 90 min at 1500 g and room temperature. The TPC of the lower phase was determined. The release rate was calculated using Equation 2 (Kenari et al., 2020): Release rate%=100–100×EncapsulatedTPCin the outer phaseEncapsulatedTPCin the inner phase The gradual release rate of phenolic compounds was observed in all samples (Table 2) and differences were significant. There is a positive correlation between size diameter and release rate of phenolic compounds from nanoparticles. This result is in line with the reports of other researchers on the gradual release of phenolic compounds from extracts of Iranian golpar (Kenari et al., 2020), rosemary leaf (Jafari et al., 2022), olive leaf (Mohammadi et al., 2016), and *Ferula persica* into soybean oil (Estakhr et al., 2020).
**TABLE 2**
| Sample | 0 | 4 | 8 | 12 | 16 | 20 | 24 |
| --- | --- | --- | --- | --- | --- | --- | --- |
| SGUM | 5.17 ± 1.1a | 10.21 ± 1.2a | 16.48 ± 2.0a | 25.76 ± 3.5a | 39.91 ± 4.2a | 51.42 ± 5.1a | 66.70 ± 5.3a |
| WHEY | 5.02 ± 0.9b | 8.22 ± 1.0b | 11.35 ± 2.4b | 20.76 ± 1.2b | 28.70 ± 3.2b | 34.91 ± 4.8b | 48.52 ± 2.1b |
| COMP | 4.81 ± 0.8c | 7.45 ± 1.1c | 10.36 ± 1.5c | 17.08 ± 2.5c | 22.19 ± 2.7c | 30.25 ± 2.7c | 43.22 ± 3.5c |
## Oil storage and tests
Free (FREE) and nanoencapsulated sage extract in different seed gum (SGUM), whey protein isolate (WHEY), and complex coatings (COMP) were added to sunflower oil at 250 ppm. Synthetic TBHQ antioxidant (TBHQ) was employed at 100 ppm of concentration to compare the efficiency of sage extract. A control (CONT) sample without antioxidant and other samples were placed in separate bottles and kept in an incubator at 60°C for 24 days. Oil samples were removed periodically every 0, 4, 8, 12, 16, 20, and 24 days for analysis. The release rate of phenolic compounds (Jafari et al., 2022), peroxide value (PV), thiobarbituric acid value (TBA), conjugated dienes (CD) (AOCS, 2009), oxidative stability index (OSI) (Farahmandfar et al., 2018), and color index (CI) were determined (Kenari et al., 2020) every 4 days.
## Statistical analysis
All experiments were performed in triplicate. Experimental data were analyzed using SPSS software (Statistical Program for Social Sciences) version 22. Significant differences ($p \leq .05$) were calculated using Duncan's multiple tests.
## Total phenolic content of sage extract
The total phenolic content (TPC) of sage extract was 31.12 mg GA/g. Nutrizio et al. [ 2020] explored high‐voltage electrical discharge and conventional method for extracting bioactive compounds from sage. They reported 19.67 and 42.13 mg GAE/g for conventional and electrical discharge extraction, respectively (Nutrizio et al., 2020). The TPC of aqueous extract of sage obtained by hot water extraction was 89.65 mg CA/g DW (Kontogianni et al., 2022). A value of 73.7 mg CA/g DW was reported by Kontogianni et al., 2013 for sage extract (Kontogianni et al., 2013). The difference in TPC may be attributed to the extraction time and temperature, type of solvent, extraction method, and variety of sage plants. Hamrouni‐Sellami et al. [ 2013] measured the effect of different drying temperatures on TPC of sage extract. They reported TPC from 0.4 to 2.5 mg GAE/g DW (Hamrouni‐Sellami et al., 2013).
## Antioxidant activity of sage extract
The antioxidant activity of sage extract was determined by the DPPH radical scavenging and FRAP assay. Figure 1a,b presents the antioxidant properties of different concentrations of sage extract. The antioxidant activity of extract was increased by increasing extract concentration. A statistically significant difference was observed between samples in the DPPH method. In the FRAP assay, the concentration of 50 and 100 ppm of extract has no statistically significant difference. Notably, the sage extract at 250 ppm had higher antioxidant activity than TBHQ in both DPPH and FRAP methods. Hamrouni‐Sellami et al. [ 2013] reported higher antioxidant activity of sage extract than BHA, HT, and ascorbic acid in DPPH, FRAP, and β‐carotene assay (Hamrouni‐Sellami et al., 2013) which is in line with the results of our study. The finding of the present study demonstrated that 250 ppm of sage extract could exhibit antioxidant activity equal to synthetic THQ antioxidant. Bigi et al. [ 2021] incorporated the sage extract into biopolymeric chitosan/hydroxypropyl methylcellulose coating and reported antioxidant activity due to the presence of bioactive phenolic compounds such as phenolic and flavonoids (Bigi et al., 2021). The antioxidant activity of sage extract related to presence of carnosol, rosmarinic acid, rosmanol, quinic acid, and carnosic acid (Generalić et al., 2012; Kontogianni et al., 2013; Oudjedi et al., 2019). Kontogianni et al., 2013 reported antioxidant activity for sage extract in both DPPH and FRAP methods which was IC50 = 27.41 μg DW/ml, and 536.81 mg Trolox/DW (Kontogianni et al., 2013). The antioxidant activity of sage extract obtained by electrical discharge, conventional method, and microwave also was reported by other researchers (Generalić et al., 2012; Hamrouni‐Sellami et al., 2013; Nutrizio et al., 2020). Similarly, literature reported a significant increase in both DPPH and FRAP antioxidant activity by an increase in the TPC of extract (Esmaeilzadeh Kenari & Razavi, 2022; Kenari et al., 2020; Razavi & Kenari, 2021).
**FIGURE 1:** *Antioxidant activity of sage extract. (a) DPPH radical scavenging activity, (b) ferric reduction antioxidant power*
## Properties of nanocapsules
The results of the particle size of different nanocapsules are shown in Table 1. All nanocapsules showed a size below 270 nm and a statistically significant difference was observed. The pressure of ultra‐turrax beside sonication energy caused nanosize of particles (Razavi et al., 2020). PDI is among the most important characteristic of nanocarrier systems. PDI of all samples was below 0.300 which indicates the normal distribution of particle size. The zeta potential is helpful to determine the net charge of nanocapsules. Zeta potential of all nanocapsules was negative. It is because of negative nature of whey protein isolate and anionic compounds in seed gum. Tavares and Noreña [2019] reported a negative charge for encapsulated extract in whey protein isolate and chitosan which is due to the negative charge of whey protein isolate (Tavares & Noreña, 2019). The lower zeta potential was observed in nanocapsule prepared using complex coating which attributed to intensifying the negative charge. EE of extract ranged from $61.54\%$ to $74.77\%$. The higher and lower EE was observed in nanocapsule prepared by seed gum followed by whey protein isolate, respectively. The EE higher than $50\%$ was also reported by other researchers (Hosseinialhashemi et al., 2020; Razavi et al., 2020; Rezaei Savadkouhi et al., 2020).
**TABLE 1**
| Sample | Particle size (nm) | PDI | Zeta potential (mV) | EE (%) |
| --- | --- | --- | --- | --- |
| SGUM | 270.0 ± 6.7a | 0.288 ± 0.02c | −35.2 ± 2.1b | 74.77 ± 4.2a |
| WHEY | 255.3 ± 5.4b | 0.294 ± 0.04b | −24.17 ± 1.8a | 61.54 ± 4.0c |
| COMP | 217.4 ± 6.2c | 0.300 ± 0.01a | −41.36 ± 3.6c | 68.16 ± 3.5b |
## Morphology of nanocapsules
The morphological structure of nanocapsules depends on the interactions between the coating components, which affect the final physiochemical properties. The surface morphology of nanocapsules is presented in Figure 2. The surface of all nanocapsules was smooth and did not show cracks, pores, and bubbles. Figure 2c indicates the formation of high compatibility between gum and protein to form wall coating. These surface morphology images confirmed that the sage extract was well encapsulated into the polymer matrix (Esmaeilzadeh Kenari & Razavi, 2022). A similar result was observed in a nanocapsule of Iranian golpar (Kenari et al., 2020), *Fumaria parviflora* (Razavi & Kenari, 2021), rosemary leave (Jafari et al., 2022), and sesame seed extract (Esmaeilzadeh Kenari & Razavi, 2022).
**FIGURE 2:** *SEM images of nanoencapsulated sage extract in different coatings. (a) Whey protein isolate, (b) seed gum, and (c) complex of protein and gum*
## Oil oxidation
Oils with high degree of unsaturation are prone to autooxidation. The simplest test for evaluating the oil autooxidation is PV and TBA. Figure 3a shows the values of PV for each sample in relation to the days of storage at 60°C. In all samples, a continuous increase in PV was observed over time. In the control sample after primary oxidation and maximum PV, a decrease in PV was observed which indicates the stage where the rate of decomposition of peroxide is higher than the rate of peroxide formation. The PV of all samples at the initial time was 1.86 meq/kg. Therefore, the rate of oil oxidation during storage depends on the type of antioxidants being added. The control sample exhibited the highest level of peroxides during storage (76.48 meq/kg) and at 20 days of storage, a decrease in PV was observed. During the storage, sunflower oil containing nanoencapsulated sage extract showed lower PV than oil containing the free sage extract. In other words, the nanoencapsulated extract was more effective to delay the oxidation process than the free extract during the first stage. A similar result was observed by Royshanpour et al. [ 2020] who reported lower PV in soybean oil enriched with nanoencapsulated M. piperita than in free extract (Royshanpour et al., 2020). The control sample exhibited higher PV followed by FREE, TBHQ, SGUM, WHEY, and COMP. In a study conducted by Dauber et al. [ 2022], the antioxidant activity of olive leaf extract in canola oil was measured. A higher PV was observed in the sample without antioxidant and the oil‐containing extract exhibited lower PV due to the presence of phenolic compounds in the extract (Dauber et al., 2022). Hosseinialhashemi et al. [ 2020] stated the higher efficiency of encapsulated Pistacia khinjuk extract than TBHQ on extension of sunflower oil stability (Hosseinialhashemi et al., 2020).
**FIGURE 3:** *Oxidation of sunflower oil during storage. (a) Peroxide value, and (b) thiobarbituric acid value*
TBA value gives a measure of oil oxidation development, in terms of secondary products of oil oxidation. The results of the TBA value of different samples in Figure 3b show an increasing rate in TBA of all samples. The control sample exhibited a higher TBA value followed by FREE, TBHQ, SGUM, WHEY, and COMP. Aleena et al. [ 2020] measured the oxidative stability of sunflower oil in high‐temperature cooking. Their results showed an increase in lipid oxidation during heating (Aleena et al., 2020) which is in accordance with the results of the present study. Binsi et al. [ 2017] increased the oxidative stability of fish oil using sage extract and oil encapsulation. Their results revealed that sage could extend the shelf life of fish oil according to the PV and TBA (Binsi et al., 2017). An increasing trend in the TBA value of plant oil was also reported for soybean oil containing free and nanoencapsulated olive leaf extract (Taghvaei et al., 2014), and potato skin extract in soybean oil (Tavakoli et al., 2021).
The oxidative stability index (OSI) is defined as the point of maximum variation of oil oxidation rate. The results of oxidative stability of sunflower oil samples which were performed at 110°Care illustrated in Figure 4. The OSI of all samples decreased over time. Continued decrease in OSI with an increase in storage time was reported for sunflower oil with pussy willow extract (Sayyari & Farahmandfar, 2017). Control sample showed a lower OSI. Also, oil samples containing encapsulated sage extract exhibited higher OSI, which is related to the antioxidant activity of sage extract. In a study conducted by Upadhyay and Mishra [2015], the sage extract was found to have protective effect on oxidative stability of sunflower oil (Upadhyay & Mishra, 2015). Taghvaei et al. [ 2014] found that the thermal stability of soybean oil containing olive leaf extract in both free and encapsulated forms is higher than blank oil. However, oil containing encapsulated extract showed higher OSI (Taghvaei et al., 2014) which is in accordance with the results of the present study.
**FIGURE 4:** *Oxidative stability of sunflower oil during storage*
Color is considered a vital indicator contributing to the quality of edible oils and consumer preference. The results of color index of different oil samples are presented in Figure 5. The color index of all samples increased over time. At the end of storage time, control sample showed a higher color index. During storage time under thermal conditions, the yellow color of sunflower oil turns dark which is related to the oil oxidation process. Therefore, the oil samples containing TBHA and encapsulated extract showed a lower color index. The decomposition of secondary metabolites to smaller compounds and the formation of polymeric triglycerides was much more in control sample. The colorful compounds present in the sage extract cause higher color index of oil than the control sample on days 0 and 4 of storage. The encapsulation process led to the placement of extract's compounds inside the coating and decreased color index. These results are in accordance with the results reported by Kenari et al. [ 2020], who reported higher color index in soybean oil without antioxidant followed by oil containing TBHQ, nanoencapsulated Iranian golpar extract, and free extract (Kenari et al., 2020). Similar results were reported by Salami et al. [ 2020] for lower color index of canola oil containing TBHQ than oil with pumpkin peel extract (Salami et al., 2020).
**FIGURE 5:** *Color index of sunflower oil during storage*
Another indicator for evaluating oil oxidation is conjugated dienes (CD). These compounds are formed by the rearrangement of double bounds of hydroperoxides during oil oxidation. The results of CD of different oil samples are shown in Figure 6. A continuous increase in CD value was observed in line with the lengthening of the storage period for all samples. Similar to PV, the oil samples containing nanoencapsulated sage extract showed lower CD value. The CD value represents the primary degradation products of oil and confirms the PV content of oil samples. Talón et al. [ 2019] reported lower CD value in sunflower oil containing encapsulated eugenol (Talón et al., 2019). An increasing trend in CD value of plant oil during thermal processing and storage time, and also low CD value of oil containing plant extracts previously reported by other publications (Kenari et al., 2020; Maghsoudlou et al., 2017; Salami et al., 2020; Talón et al., 2019).
**FIGURE 6:** *Conjugated dienes of sunflower oil during storage*
## CONCLUSION
In this study, the antioxidant effect of free and nanoencapsulated sage extract was compared to TBHQ synthetic antioxidant. All coating materials could extend the antioxidant properties of sage extract by controlling gradual release of phenolic compounds, and protecting phenolic compounds from environmental stresses. According to the results of oxidation parameters of oil, the use of complex coating of whey protein isolate and qodumeh shahri seed gum was suggested to encapsulate sage seed gum as a natural antioxidant to extend the shelf life and oxidative stability of sunflower oil.
## CONFLICT OF INTEREST
The authors declare no conflict of interest.
## ETHICS STATEMENT
The study does not involve any human or animal testing.
## DATA AVAILABILITY STATEMENT
Research data are not shared.
## References
1. Aleena K., Divya M., Beena A., Rachana C., Divya K.. **Oxidative stability of sunflower oil on high temperature cooking**. *Frontiers in Nutrition* (2020) **9** 552-554
2. Aly A. A., Zaky E. A., Elhabeby B. S., Alessa H., Hameed A. M., Aljohani M., Alghamdi A. A.. **Effect of thyme addition on some chemical and biological properties of sunflower oil**. *Arabian Journal of Chemistry* (2021) **14**
3. AOCS
. (2009). Official methods and recommended practices of the AMOS. AMOS Press Champaign.. *Official methods and recommended practices of the AMOS* (2009)
4. Bigi F., Haghighi H., Siesler H. W., Licciardello F., Pulvirenti A.. **Characterization of chitosan‐hydroxypropyl methylcellulose blend films enriched with nettle or sage leaf extract for active food packaging applications**. *Food Hydrocolloids* (2021) **120**
5. Binsi P., Nayak N., Sarkar P., Jeyakumari A., Ashraf P. M., Ninan G., Ravishankar C.. **Structural and oxidative stabilization of spray dried fish oil microencapsulates with gum arabic and sage polyphenols: Characterization and release kinetics**. *Food Chemistry* (2017) **219** 158-168. PMID: 27765212
6. Dauber C., Carreras T., González L., Gámbaro A., Valdés A., Ibañez E., Vieitez I.. **Characterization and incorporation of extracts from olive leaves obtained through maceration and supercritical extraction in Canola oil: Oxidative stability evaluation**. *LWT* (2022) **160**
7. Doymaz İ., Karasu S.. **Effect of air temperature on drying kinetics, colour changes and total phenolic content of sage leaves (**. *Quality Assurance and Safety of Crops & Foods* (2018) **10** 269-276
8. El‐Sayed S. M., Youssef A. M.. **Potential application of herbs and spices and their effects in functional dairy products**. *Heliyon* (2019) **5**. PMID: 31338458
9. Esmaeilzadeh Kenari R., Razavi R.. **Phenolic profile and antioxidant activity of free/bound phenolic compounds of sesame and properties of encapsulated nanoparticles in different wall materials**. *Food Science & Nutrition* (2022) **10** 525-535. PMID: 35154689
10. Estakhr P., Tavakoli J., Beigmohammadi F., Alaei S., Mousavi Khaneghah A.. **Incorporation of the nanoencapsulated polyphenolic extract of Ferula persica into soybean oil: Assessment of oil oxidative stability**. *Food Science & Nutrition* (2020) **8** 2817-2826. PMID: 32566199
11. Farahmandfar R., Asnaashari M., Pourshayegan M., Maghsoudi S., Moniri H.. **Evaluation of antioxidant properties of lemon verbena (**. *Food Science & Nutrition* (2018) **6** 983-990. PMID: 29983961
12. Generalić I., Skroza D., Šurjak J., Možina S. S., Ljubenkov I., Katalinić A., Katalinić V.. **Seasonal variations of phenolic compounds and biological properties in sage (**. *Chemistry & Biodiversity* (2012) **9** 441-457. PMID: 22344920
13. Hamrouni‐Sellami I., Rahali F. Z., Rebey I. B., Bourgou S., Limam F., Marzouk B.. **Total phenolics, flavonoids, and antioxidant activity of sage (**. *Food and Bioprocess Technology* (2013) **6** 806-817
14. Hosseinialhashemi M., Tavakoli J., Rafati A., Ahmadi F.. **The aplication of**. *Food Science & Nutrition* (2020) **9** 920-928. PMID: 33598175
15. Jafari S. Z., Jafarian S., Hojjati M., Najafian L.. **Evaluation of antioxidant activity of nano‐and microencapsulated rosemary (**. *Food Science Nutrition* (2022) **10** 2111-2119. PMID: 35702297
16. Jamshidi M., Kenari R. E., Motamedzadegan A., Biparva P.. **Encapsulation of unsaponifiable matter of rice bran oil Bychitosan and**. *Journal of the American Oil Chemists' Society* (2020) **97** 1131-1140
17. Kenari R. E., Amiri Z. R., Motamedzadegan A., Milani J. M., Farmani J., Farahmandfar R.. **Optimization of Iranian golpar (**. *Journal of Food Measurement and Characterization* (2020) **14** 2828-2839
18. Kenari R. E., Razavi R.. **Encapsulation of bougainvillea (**. *Food Science & Nutrition* (2022) **10** 3436-3443. PMID: 36249979
19. Kontogianni V. G., Kasapidou E., Mitlianga P., Mataragas M., Pappa E., Kondyli E., Bosnea L.. **Production, characteristics and application of whey protein films activated with rosemary and sage extract in preserving soft cheese**. *LWT* (2022) **155**
20. Kontogianni V. G., Tomic G., Nikolic I., Nerantzaki A. A., Sayyad N., Stosic‐Grujicic S., Tzakos A. G.. **Phytochemical profile of**. *Food Chemistry* (2013) **136** 120-129. PMID: 23017402
21. Maghsoudlou E., Esmaeilzadeh Kenari R., Raftani Amiri Z.. **Evaluation of antioxidant activity of fig (**. *Journal of Food Processing and Preservation* (2017) **41**
22. Meng Y., Yang H., Wang D., Ma Y., Wang X., Blasi F.. **Improvement for oxidative stability and sensory properties of sunflower oil flavored by Huai**. *Processes* (2021) **9** 1199
23. Mohammadi A., Jafari S. M., Esfanjani A. F., Akhavan S.. **Application of nano‐encapsulated olive leaf extract in controlling the oxidative stability of soybean oil**. *Food Chemistry* (2016) **190** 513-519. PMID: 26213004
24. Naziruddin M., Kian L., Jawaid M., Aziman N., Yusof N., Abdul‐Mutalib N., Tverezovskaya O.. **Development of encapsulated sage extract powder: Inter‐comparison with commercially available powder for physical properties and metabolites composition**. *The Journal of Supercritical Fluids* (2022) **184**
25. Nutrizio M., Kljusurić J. G., Sabolović M. B., Kovačević D. B., Šupljika F., Putnik P., Maltar‐Strmečki N.. **Valorization of sage extracts (**. *Innovative Food Science & Emerging Technologies* (2020) **60**
26. Oudjedi K., Manso S., Nerin C., Hassissen N., Zaidi F.. **New active antioxidant multilayer food packaging films containing Algerian sage and bay leaves extracts and their application for oxidative stability of fried potatoes**. *Food Control* (2019) **98** 216-226
27. Razavi R., Kenari R. E.. **Antioxidant evaluation of**. *Journal of Food Measurement and Characterization* (2021) **15** 2448-2457
28. Razavi R., Kenari R. E., Farmani J., Jahanshahi M.. **Fabrication of zein/alginate delivery system for nanofood model based on pumpkin**. *International Journal of Biological Macromolecules* (2020) **165** 3123-3134. PMID: 33127546
29. Razavi R., Kenari R. E., Farmani J., Jahanshahi M.. **Preparation of double‐layer nanoemulsions with controlled release of glucose as prevention of hypoglycemia in diabetic patients**. *Biomedicine & Pharmacotherapy* (2021) **138**. PMID: 33725590
30. Reddy C. K., Agarwal R. K., Shah M. A., Suriya M.. *Plant extracts: Applications in the food industry* (2022) 75-88
31. Rezaei Savadkouhi N., Ariaii P., Charmchian Langerodi M.. **The effect of encapsulated plant extract of hyssop (**. *Food Science & Nutrition* (2020) **8** 1264-1271. PMID: 32148832
32. Royshanpour F., Tavakoli J., Beigmohammadi F., Alaee S.. **Improving antioxidant effect of phenolic extract of**. *Journal of Food Measurement and Characterization* (2020) **15** 23-32
33. Sadat A. N., Ali M., Sultana A., Hasan M. M., Sharma D., Atiqur M.. **Comparative study of a proposed green extraction method named aqueous ultrasound assisted extraction from fresh leaves of**. *International Journal of Innovative Science & Research Technology* (2021) **6** 946-951
34. Salami A., Asefi N., Kenari R. E., Gharekhani M.. **Addition of pumpkin peel extract obtained by supercritical fluid and subcritical water as an effective strategy to retard canola oil oxidation**. *Journal of Food Measurement and Characterization* (2020) **14** 2433-2442
35. Sayyari Z., Farahmandfar R.. **Stabilization of sunflower oil with pussy willow (**. *Food Science & Nutrition* (2017) **5** 266-272. PMID: 28265361
36. Taghvaei M., Jafari S. M., Mahoonak A. S., Nikoo A. M., Rahmanian N., Hajitabar J., Meshginfar N.. **The effect of natural antioxidants extracted from plant and animal resources on the oxidative stability of soybean oil**. *LWT‐Food Science and Technology* (2014) **56** 124-130
37. Talón E., Vargas M., Chiralt A., González‐Martínez C.. **Antioxidant starch‐based films with encapsulated eugenol. Application to sunflower oil preservation**. *LWT* (2019) **113**
38. Tavakoli J., Abbasi H., Zarei Jelyani A., Mousavi Khaneghah A.. **The use of**. *Journal of Food Quality* (2021) **2021**
39. Tavares L., Noreña C. P. Z.. **Encapsulation of garlic extract using complex coacervation with whey protein isolate and chitosan as wall materials followed by spray drying**. *Food Hydrocolloids* (2019) **89** 360-369
40. Upadhyay R., Mishra H. N.. **Multivariate analysis for kinetic modeling of oxidative stability and shelf life estimation of sunflower oil blended with sage (**. *Food and Bioprocess Technology* (2015) **8** 801-810
41. Wang D., Meng Y., Wang C., Wang X., Blasi F.. **Antioxidant activity and sensory improvement of**. *Processes* (2020) **8** 403
42. Wrona O., Rafińska K., Możeński C., Buszewski B.. **Supercritical fluid extraction of bioactive compounds from plant materials**. *Journal of AOAC International* (2017) **100** 1624-1635. PMID: 28707617
43. Xu X., Liu A., Hu S., Ares I., Martínez‐Larrañaga M.‐R., Wang X., Martínez M.‐A.. **Synthetic phenolic antioxidants: Metabolism, hazards and mechanism of action**. *Food Chemistry* (2021) **353**. PMID: 33714793
|
---
title: 'Seasonal Changes in Midlife Women’S Percentage Body Fat: A 1-Year Cohort Study'
authors:
- A.M. Nelson
- S.L. Casperson
- L. Jahns
- D.G. Palmer
- J.N. Roemmich
journal: JAR life
year: 2022
pmcid: PMC10002894
doi: 10.14283/jarlife.2022.4
license: CC BY 4.0
---
# Seasonal Changes in Midlife Women’S Percentage Body Fat: A 1-Year Cohort Study
## Abstract
### Objective
The purpose of this longitudinal, observational study was to examine whether age and seasonal changes in sedentary activity (sedAct), moderate-to-vigorous physical activity (MVPA), and energy intake (EI) predict changes in body composition among midlife women. We hypothesized that reductions in MVPA and increases in sedAct and EI in winter, along with greater baseline age would predict increases in percentage body fat (%BF) across seasons.
### Design
This study used a longitudinal, within-subjects design. Setting: This study took place in Grand Forks, North Dakota.
### Participants
Participants included 52 midlife women (aged 40-60 years) who were observed over the course of one year.
### Measurements
Percentage body fat measures were obtained via whole body Dual Energy X-ray absorptiometry. Participants were scanned once per season. We measured EI using the ASA24®. We used a GTX3 accelerometer to measure physical activity. Each season, participants wore the monitors for 7 days, 12 hours per day. All measures began in summer.
### Results
Results of hierarchical multiple regression (MR) analyses showed that age increases (β = 0.310, $$p \leq 0.021$$) and summer-to-fall increases in EI (β = 0.427, $$p \leq 0.002$$) predicted seasonal increases in %BF (R2 =.36, F[5, 42]= 4.66, $$p \leq 0.02$$). Changes in MVPA and sedAct were not significant predictors. Repeated measures ANCOVA revealed that summer ($M = 37.7263$, $95\%$ CI [35.8377, 39.6149]) to winter ($M = 38.1463$, $95\%$ CI [36.1983, 40.0942]) increases in %BF are not reversed by spring ($M = 37.8761$, $95\%$ CI [35.9365, 39.8157]).
### Conclusions
To minimize increases in %BF and maintain health, midlife women, particularly older women, should be encouraged to pay extra attention to their diet in the fall months.
## Introduction
Approximately $76\%$ of US women have overweight (body mass index (BMI) ≥ 25) or obesity (BMI ≥ 30; 1). The greatest prevalence of obesity is among women aged 40 years and older, with $43.3\%$ classified as obese [2]. Obesity is associated with greater risk for heart disease, diabetes, and some cancers [3-6]. Women with obesity have a greater risk of type 2 diabetes than men [7], and the risk of diabetes increases with age [8]. As such, understanding factors that contribute to greater rates of obesity in midlife women is crucial.
Age is positively correlated with weight gain in women [9, 10]. This is true even when exercise remains constant [10]. Weight gain is especially prevalent in women who are midlife and of menopausal age [11]; however, changes in body composition (e.g., increases in body fat [12] and visceral fat [13] can occur even in the absences of weight gain [12]. While this may be partly due to biological changes with aging [13], it is important to examine additional predictors that could be exacerbating midlife weight gain.
Weight gain may be more likely during certain seasons due to alterations in usual eating and physical activity. Indeed, diet, physical activity, and body weight change with season [14-16]. Over the course of a year, American adults consumed more energy (kcal) per day in fall relative to spring [14]. Physical activity (PA) differs across seasons as well; PA is lowest [14, 15, 17-19] and sedentary activity (sedAct) is greatest [17, 20]during winter relative to the other seasons. Weather is likely to drive physical activity changes, with conditions such as snowfall [21] and extreme weather [19] being frequently cited as barriers to engaging in physical activity. Another contributing factor to physical activity changes may be daylight; women who experience more than 14 hours of daylight engage in more moderate-to-vigorous physical activity (MVPA) than women who live in an area that is receiving less than 10 hours of daylight [22]. As a result, body weight is greatest in winter [14]. A study of Mexican American women reported similar findings; women gained the most weight in fall [15]. Fall-winter weight gain presents a risk for gradual increases in body weight during adulthood as weight gained is not lost during spring and summer [23].
The purpose of the present secondary analysis of a longitudinal, observational study was to examine whether age and changes in sedAct, MVPA, and energy intake (EI) across seasons predict changes in body composition among midlife women. We hypothesized that reductions in MVPA and increases in sedAct, and EI in winter, along with greater baseline age would predict increases in %BF from summer to spring. As secondary aims, we investigated the changes in each of the predictor variables across seasons. As these data were derived in a location where the spring months can be as intemperate as winter months, we hypothesized that there would be greater levels of EI and sedAct in winter and spring relative to summer and fall. We also hypothesized that MVPA would be lower in winter and spring relative to summer and fall.
## Participants
The study was completed by a total of 52 ambulatory women ranging in age from 40-60 years as previously reported [16, 24-26]. Women were non-overweight, overweight, or obese as classified by BMI ranging from 18-35 kg/m2. Most women were menopausal at the beginning of the study ($$n = 27$$), as measured by follicle-stimulating hormone (FSH) levels of 25.8 mIU/ mL or greater, with 5 additional participants reaching menopausal status by winter. FSH measurement methods have been previously described [25, 26].
Participants were required to have stable weight, defined as fluctuation not exceeding ±4.5 kg for at least 6 months prior to the beginning of the study. Women were excluded from the study if they were smokers, pregnant or lactating, or had health conditions that would limit their physical activity. Furthermore, women who took medications that could potentially influence weight/ appetite were excluded. Participants were asked to refrain from engaging in intentional changes in diet or physical activity while the study was in progress.
Participants were recruited from the Grand Forks, North *Dakota area* through advertising throughout the community. This study was reviewed and approved by the University of North Dakota Institutional Review Board and registered with ClinicalTrials.gov(#NCT01674296). Informed consent was documented prior to the beginning of the study.
## Body Composition
Percentage body fat (%BF) was estimated with whole body Dual Energy X-ray Absorptiometry (DXA, GE Lunar, Madison, WI, enCORE Software Version 13.60.033). The instrument was calibrated before each session using the manufacturer’s calibration phantom. For the 248 calibrations over the 21 months of the study, mean calibration %BF was $60.53\%$ with a standard deviation of 0.01 and a coefficient of variation of $0.016\%$ body fat. All calibration results were within the tolerance limits recommended by the manufacturer. Participants were scanned wearing light clothing or scrubs. Analysis was conducted using iDXA proprietary software.
## Energy Intake
EI, defined as mean calories reported consumed across each season, was derived using the National Cancer Institute’s Automated Self-Administered 24-hour Dietary Recall [27]. The ASA24® is an online measure in which participants self-report the food that they have consumed over the last 24 hours, including all meals, snacks, and drinks. From these data, outcomes such as total intake of energy, carbohydrates, fat, and protein are calculated. Participants completed the ASA24® 36 times throughout the study, each completion spaced approximately 10 days apart.
## Physical Activity
We measured physical activity using a GTX3 accelerometer (ActiGraph Corp., Pensacola, FL USA). Each season, participants wore the monitors at the hip for 7 consecutive days, 12 hours per day. Data were cleaned to remove non-wearing data (i.e., periods during which consecutive zeros were recorded for 20 min). Epochs of 15s were used for data collection. From this, we calculated total minutes of sedAct and MVPA for each season using the Crouter, Kuffel [28] algorithm and Freedson cut-points [29].
## Procedures
The present study had two cohorts; the first began in July of 2012, the second began in July of 2013. Participants visited the Research Center weekly. For the purposes of this study, we defined seasons as summer (June, July, August), fall (September, October, November), winter (December, January, February), and spring (March, April, May). All visits were conducted in the middle month of each season (i.e., July, October, January, and April).
While the visits to the Research Center were otherwise identical, the participants’ very first visit, Day 0 of the summer visit, had two unique components that occurred only at the first visit: [1] participants signed an informed consent document and [2] participants were trained during this visit on how to wear accelerometers.
On Day 0 of each season, participants came to the Research Center after a 12 h fast to complete the DXA body scan, along with other tests including completing a series of online questionnaires. The additional Day 0 tests are described in previous publications [24, 25]. Before leaving the Research Center, participants were given their physical activity monitors and instructed to wear them 12 hours per day for the following 7 days. On Day 8, participants returned the physical activity monitors to the Research Center.
## Statistical Analysis
To determine whether there were changes in %BF across seasons, a multivariate repeated measures analysis of covariance (ANCOVA) was conducted with season as the repeated measure and age as the continuous covariate. Due to a violation of the sphericity assumption of compound symmetry, we used Greenhouse-Geisser corrected p values in the ANCOVA tables. For pairwise multiple comparisons of least squares means of age by season, we used Tukey-Kramer adjusted p values. We calculated difference scores for %BF, EI, sedAct, and MVPA between summer and fall, winter, and spring. We then mean-centered these scores for regression analyses. A significance level of ∝ = 0.05 was chosen a priori to determine significant p values. We used SAS 9.4 TS1M7 for these analyses.
We used 3 hierarchical multiple regression/correlation (MRC) models to predict summer to spring Δ%BF. In each model, age was placed in Step 1 to serve as the variable to be controlled because it was a salient predictor of Δ%BF, and all other predictors were placed in Step 2. Model 1 measured summer-fall difference scores for sedAct, MVPA, and EI. Model 2 measured summer-winter difference scores for sedAct, MVPA, and EI. Model 3 measured summer-spring difference scores for sedAct, MVPA, and EI. As secondary analyses, we used repeated measures ANOVA to investigate whether there were seasonal changes in sedAct and MVPA; one person was excluded from this analysis due to incomplete data. We used bivariate correlational analyses to further assess the relationships between the predictors and Δ%BF across seasons, as well as FSH and %BF/Δ%BF. We used SPSS Version 27.0 for these analyses.
## Primary Outcomes
There was a significant season by age interaction for %BF, F(1.93, 94.53) = 4.09, $$p \leq .021.$$ No pairwise comparisons were statistically significant. Most ($60\%$) participants experienced increases in %BF from summer ($M = 37.7263$, $95\%$ CI [35.8377, 39.6149]) to winter ($M = 38.1463$, $95\%$ CI [36.1983, 40.0942]). Summer to winter increases in %BF (ΔM = 0.4200, adjusted $95\%$ CI [-0.1669, 1.0069]) were not reversed by spring (ΔM = -0.2702, adjusted $95\%$ CI [-0.7754, 0.2350]). These %BF increases were not reversed by spring in $30\%$ of the sample that were younger than the mean age of 49 years (i.e., 3 of the 10 people younger than 49 years who gained %BF,), whereas $37\%$ the women who were older than 49 years (i.e., 7 of the 19 people over the age of 49 who gained %BF from summer to winter) did not reverse increases in %BF by the following spring. Age in the between-subjects table yielded a test statistic of F[1, 49] =3.77, $$p \leq .058.$$ See Table 1 for means and standard deviations and Table 2 for frequencies of season-to-season changes for primary predictors.
Hierarchical MRC model 1 predicted Δ%BF, R2 =.25, F[4, 47] = 3.89, $$p \leq .008.$$ Age was a significant predictor in Step 1, R2 =.12, F[1, 50] = 6.90, $$p \leq .011.$$ Adding seasonal changes in sedAct, MVPA, and EI from summer to fall [ΔR2 =.13, Fchange[3, 47] = 2.66, $$p \leq .059$$] did not increase the model’s ability to predict Δ%BF. In the context of the full model, increases in age and increases in EI had significant unique contributions in predicting increased %BF in spring. See Table 3 for coefficients.
Model 2 also predicted Δ%BF, R2 =.26, F[4, 46] = 3.93, $$p \leq .008.$$ Age was a significant predictor in Step 1, R2 =.14, F[1, 49] = 7.99, $$p \leq .007.$$ However, adding seasonal changes in sedAct, MVPA, and EI from summer to winter did not improve the model, ΔR2 =.12, Fchange[3, 46] = 2.36, $$p \leq .084.$$ In the context of the full model, age was the only significant predictor of Δ%BF. See Table 3 for coefficients.
**Table 3**
| Unnamed: 0 | Step | Variables | B | β | t | p | pr2 | sr2 | 95% CI for B |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| Summer to Fall Changes | Summer to Fall Changes | Summer to Fall Changes | Summer to Fall Changes | Summer to Fall Changes | Summer to Fall Changes | Summer to Fall Changes | Summer to Fall Changes | Summer to Fall Changes | Summer to Fall Changes |
| 1 | | Age | .128 | .348 | 2.63 | .011* | .348 | .348 | .030, .226 |
| 2 | | Age | .101 | .274 | 2.11 | .040* | .294 | .267 | .005, .197 |
| | | sedAct | .002 | .067 | 0.46 | .650 | .066 | .058 | -.008, .012. |
| | | MVPA | -.014 | -.111 | 0.76 | .454 | -.110 | -.096 | -.051, .023 |
| | | EI | .002 | .335 | 2.62 | .012* | .357 | .331 | .001, .004 |
| Summer to Winter Changes | Summer to Winter Changes | Summer to Winter Changes | Summer to Winter Changes | Summer to Winter Changes | Summer to Winter Changes | Summer to Winter Changes | Summer to Winter Changes | Summer to Winter Changes | Summer to Winter Changes |
| 1 | | Age | .134 | .374 | 2.83 | .007** | .374 | .374 | .039, .229 |
| 2 | | Age | .119 | .332 | 2.43 | .019* | .337 | .309 | .020, .218 |
| | | sedAct | .008 | .269 | 1.86 | .069 | .265 | .237 | -.001, .017 |
| | | MVPA | .009 | .058 | 0.38 | .705 | .056 | .048 | -.037, .054 |
| | | EI | .002 | .231 | 1.78 | .082 | .254 | .257 | .000, .004 |
| Summer to Spring Changes | Summer to Spring Changes | Summer to Spring Changes | Summer to Spring Changes | Summer to Spring Changes | Summer to Spring Changes | Summer to Spring Changes | Summer to Spring Changes | Summer to Spring Changes | Summer to Spring Changes |
| 1 | | Age | .128 | .348 | 2.62 | .011* | .348 | .348 | .030, .226 |
| 2 | | Age | .121 | .328 | 2.45 | .018* | .337 | .325 | .022, .219 |
| | | sedAct | .000 | .006 | 0.37 | .970 | .005 | .005 | -.009, .010 |
| | | MVPA | -.014 | -.107 | 0.73 | .469 | -.106 | -.097 | -.052, .024 |
| | | EI | .001 | .194 | 1.37 | .176 | .196 | .182 | -.001, .003 |
Model 3 did not yield a better fit than the reduced model (age as only predictor) for Δ%BF, R2 =.18, F[4, 47] = 2.51, $$p \leq .054.$$ Though age was a significant predictor in Step 1, R2 =.12, F[1, 50] = 6.90, $$p \leq .011$$, adding seasonal changes in sedAct, MVPA, and EI from summer to spring did not significantly improve the model, ΔR2 =.10, Fchange[3, 47] = 1.05, $$p \leq .382.$$ See Table 3 for coefficients.
## Secondary Outcomes
Though they were not found to predict Δ%BF, we investigated whether there were seasonal changes in sedAct and MVPA. We found that sedAct changed across seasons [F(2.55, 127.62) = 5.48, $$p \leq .003$$], increasing during winter relative to summer, F[1, 50] = 9.56, $$p \leq .003.$$ Likewise, we found changes in MVPA across seasons [F[3, 150] = 4.74, $$p \leq .004$$]; MVPA in winter was lower than summer MVPA, F[1, 50] = 15.52, $p \leq .001.$
Results of bivariate correlational analyses revealed that summer-fall increases in EI were associated with sustained increases in %BF into both Winter [[50] =.37, $$p \leq .009$$] and spring, r[50] =.37, $$p \leq .007.$$ SedAct in summer, fall, and winter were positively correlated with %BF in spring (ps <.01). There was no relationship between %BF and FSH levels seasonally, nor in the change in %BF and change in FSH levels across seasons.
## Discussion
This one-year, longitudinal study examined predictors of changes in %BF of midlife women from summer to the following spring. While we hypothesized that age and changes in EI, sedAct, and MVPA would all predict %BF changes, the most important predictors of increases in %BF from summer to spring were greater age and increases in EI from summer to fall. The magnitude of yearly increases in %BF were greater in the older women, consistent with past research [9, 10]. Likewise, we found that greater EI in fall relative to summer was associated with greater increases in %BF, again consistent with past research [14, 15, 17-19]. These findings suggest that while %BF increases with age, it is exacerbated by greater EI, especially during the fall. Notably, for the group as a whole there were no increases in EI from summer to fall, yet those women who reported the greatest increases in EI from summer to fall had the greatest increases in %BF during this period. [ 30, 31]Humans may experience similar circannual weight gain patterns to hibernating mammals; storing fat in fall seasons in preparation for food shortage in winter [30, 31]. As such, excess EI in fall may be more likely to be stored as excess body fat. To our knowledge, there has not yet been work done to study this hypothesis. Though the mechanism behind the relationship between EI and %BF remains unclear, these data suggest that limiting seasonal increases in EI, especially in older women, during the fall is important because not all of the weight gained may be lost the following spring leading to the gradual weight gain with age [23].
Though not significant predictors of body composition changes, secondary analyses showed that sedAct and MVPA differed across seasons, with sedAct being greatest and MVPA being lowest in winter relative to summer. This is consistent with past research that shows people have the lowest PA and greatest sedentary behavior during winter months [14, 15, 17-20]. Given the health concerns of sedentary activity (e.g., increased risk of cardiovascular disease, cancer, and diabetes; 32) and health benefits of physical activity independent of weight management (e.g., greater cardiorespiratory fitness is associated with decreased risk of mortality regardless of BMI; 33) health benefits would likely occur by limiting increases in sedAct and increasing MVPA during winter months. Contrary to our hypothesis seasonal changes in MVPA did not predict changes in %BF. Depending on the season, the women in our study engaged in 37 to 44 min/week of MVPA, far below the recommended amount of 150 min/week [34]. The small magnitude of absolute MVPA limited the ability for energy expenditure to have an impact on %BF of the women in our study and the finite range of seasonal change in MVPA reduced its ability to predict change in %BF.
Strengths of the current study include the use of accelerometers to measure PA, DEXA to assess the primary outcome, %BF, and collecting three dietary recalls per month to estimate usual EI during each season. Retention was high, with only 2 women dropping out, and compliance was high for completion of PA and EI study tasks.
This study has limitations as well. The study lasted 9 months, and while it represents all four seasons over the course of one year, it is unknown whether participants would have modified EI or PA or reduced %BF if followed through a second summer. The sample is small and specialized regarding age and gender, which limits generalizability but does provide evidence for a group at risk of gaining excess adipose tissue [12, 13]. The study is also limited in that the sample was from northern North Dakota, USA, an area which has great changes in both weather and sunlight throughout the year, and therefore not applicable to areas with long, hot summers and cool winters, or places where it is temperate year-round.
## Conclusions
Overall, the results of this study suggest that as women age, attention should be given to achieving or maintaining appropriate energy intake and exercise during the fall and winter months to reduce increases in %BF. Limiting increases in sedentary behavior and energy intake during the fall and winter may help women reduce seasonal increases in %BF.
## Ethical Standards
This study complies with current laws of the country in which it was performed. This study received IRB approval.
## Competing Interests
The authors (AMN, SLC, LJ, DGP, & JNR) have no competing interests to report. This work was supported by the U.S. Department of Agriculture, Agricultural Research Services #5450-51530-057-00D. The U.S. Department of Agriculture prohibits discrimination in all its programs and activities on the basis of race, color, national origin, age, disability, and where applicable, sex, marital status, familial status, parental status, religion, sexual orientation, genetic information, political beliefs, reprisal, or because all or part of an individual’s income is derived from any public assistance program. ( Not all prohibited bases apply to all programs.) Persons with disabilities who require alternative means for communication of program information (Braille, large print, audiotape, etc.) should contact USDA’s TARGET Center at [202] 720-2600 (voice and TDD). To file a complaint of discrimination, write to USDA, Director, Office of Civil Rights, 1400 Independence Avenue, S.W., Washington, D.C. 20250-9410, or call [800] 795-3272 (voice) or [202] 720-6382 (TDD). USDA is an equal opportunity provider and employer.
## References
1. Fryar CD, Carroll MD, Afful J. *Statistics NCfH* (2020) 1208-10
2. Hales CM, Carroll MD, Fryar CD, Ogden CL. *Prevalence of Obesity and Severe Obesity Among Adults: United States* (2020) 2017-2018
3. Wang YC, McPherson K, Marsh T, Gortmaker SL, Brown M. **Health and economic burden of the projected obesity trends in the USA and the UK**. *The Lancet* (2011) **378** 815-25
4. Zhang H, Rodriguez-Monguio R. **Racial disparities in the risk of developing obesity-related diseases: A cross-sectional study**. *Ethnicity & Disease* (2012) **22** 308-16. PMID: 22870574
5. Dehal A, Garrett T, Tedders SH, Arroyo C, Afriyie-Gyawu E, Zhang J. **Body mass index and death rate of colorectal cancer among a national cohort of U.S. adults**. *Nutr Cancer.* (2011) **63** 1218-25. PMID: 21981584
6. Vranian M, Blaha M, Silverman M, Michos E, Minder C, Blumenthal R. **The interaction of fitness, fatness, and cardiometabolic risk**. *Journal of the American College of Cardiology* (2012) **59**
7. Kautzky-Willer A, Harreiter J, Pacini G. **Sex and gender differences in risk, pathophysiology and complications of type 2 diabetes mellitus**. *Endocrine Reviews* (2016) **37** 278-316. PMID: 27159875
8. Censin JC, Peters SAE, Bovijn J, Ferreira T, Pulit SL, Magi R. **Causal relationships between obesity and the leading causes of death in women and men**. *PLoS Genetics* (2019) **15** e1008405. PMID: 31647808
9. Stenholm S, Vahtera J, Kawachi I, Pentti J, Halonen JI, Westerlund H. **Patterns of weight gain in middle-aged and older US adults, 1992-2010**. *Epidemiology* (2015) **26** 165-8. PMID: 25643097
10. Williams PT, Wood PD. **The effects of changing exercise levels on weight and age-related weight gain**. *International Journal of Obesity* (2006) **30** 543-51. PMID: 16314878
11. Kapoor E, Collazo-Clavell ML, Faubion SS. **Weight gain in women at midlife: A concise review of the pathophysiology and strategies for management**. *Mayo Clinic Proceedings* (2017) **92**
12. St-Onge MP. (2005) **8** 523-8
13. Hunter GR, Gower BA, Kane BL. **Age related shift in visceral fat**. *International Journal of Body Composition Research* (2010) **8** 103-8. PMID: 24834015
14. Ma Y, Olendzki BC, Li W, Hafner AR, Chiriboga D, Hebert JR. **Seasonal variation in food intake, physical activity, and body weight in a predominantly overweight population**. *European Journal of Clinical Nutrition* (2006) **60** 519-28. PMID: 16340952
15. Lloyd L, Miller B. **The impact of seasonality on changes in body weight and physical activity in Mexican-American women**. *Women Health* (2013) **53** 262-81. PMID: 23705758
16. Jahns L, Johnson LK, Scheet AJ, Raatz SK, Subar AF, Tande D. **Measures of diet quality across calendar and winter holiday seasons among midlife women: A 1-Year longitudinal study using the automated self-administered 24-hour recall**. *Journal of the American Academy of Nutrition and Dietetics* (2016) **116** 1961-9
17. Buchowski MS, Choi L, Majchrzak KM, Acra S, Matthews CE, Chen KY. **Seasonal changes in amount and patterns of physical activity in women**. *Journal of Physical Activity and Health* (2009) **6** 252-61. PMID: 19420404
18. Pivarnik JM, Reeves MJ, Rafferty AP. **Seasonal variation in adult leisure-time physical activity**. *Medicine & Science in Sports & Exercise* (2003) **35** 1004-8. PMID: 12783049
19. Tucker P, Gilliland J. **The effect of season and weather on physical activity: a systematic review**. *Public Health* (2007) **121** 909-22. PMID: 17920646
20. O’Connell SE, Griffiths PL, Clemes SA. **Seasonal variation in physical activity, sedentary behaviour and sleep in a sample of UK adults**. *Annals of Human Biology* (2014) **41** 1-8. PMID: 23992280
21. Ogawa S, Seko T, Iro T, Mori M. **Differences in physical activity between seasons with and without snowfall among elderly individuals residing in areas that receive snowfall**. *Journal of Physical Therapy Science* (2019) **31** 12-6. PMID: 30774197
22. Schepps MA, Shiroma EJ, Kamada M, Harris TB, Lee IM. **Day length is associated with physical activity and sedentary behavior among older women**. *Scientific Reports* (2018) **8** 6602. PMID: 29700376
23. Yanovski JA, Yanovski SZ, Sovik KN, Nguyen TT, O’Neil PM, Sebring NG. **A prospective study of holiday weight gain**. *New England Journal of Medicine* (2000) **342** 861-7. PMID: 10727591
24. Jahns L, Conrad Z, Johnson LK, Scheett AJ, Stote KS, Raatz SK. **Diet quality is lower and energy intake is higher on weekends compared with weekdays in midlife women: A 1-year cohort study**. *Journal of the Academy of Nutrition and Dietetics* (2017) **117** 1080-6. PMID: 28254199
25. Jahns L, Johnson LK, Conrad Z, Bukowski M, Raatz SK, Jilcott Pitts S. **Concurrent validity of skin carotenoid status as a concentration biomarker of vegetable and fruit intake compared to multiple 24-h recalls and plasma carotenoid concentrations across one year: a cohort study**. *Nutrition Journal* (2019) **18** 78. PMID: 31752882
26. Laugero KD, Paul DR, Jahns L. **Energy intake and season interact to influence physiological stress load among midlife women.**. *The International Journal on the Biology of Stress* (2021)
27. Baranowski T, Baranowski JC, Watson KB, Martin S, Beltran A, Islam N. **Children’s accuracy of portion size estimation using digital food images: effects of interface design and size of image on computer screen**. *Public Health Nutrition* (2011) **14** 418-25. PMID: 21073772
28. Crouter SE, Kuffel E, Haas JD, Frongillo EA, Bassett DR. **Refined two-regression model for the ActiGraph accelerometer**. *Med Sci Sports Exerc* (2010) **42** 1029-37. PMID: 20400882
29. Freedson PS, Melanson E, Sirard J. **Calibration of the Computer Science and Applications, Inc. accelerometer**. *Medicine & Sciene in Sports & Exerise* (1998) **30** 777-81
30. Martin SL. **Mammalian hibernation: a naturally reversible model for insulin resistance in man?**. *Diabetes and Vascular Disease Research.* (2008) **5** 76-81. PMID: 18537093
31. Gangwisch JE. **Seasonal variation in metabolism: evidence for the role of circannual rhythms in metabolism?**. *Hypertens Res.* (2013) **36** 392-3. PMID: 23344125
32. Park JH, Moon JH, Kim HJ, Kong MH, Oh YH. **Sedentary lifestyle: Overview of updated evidence of potential health risks**. *Korean Journal of Family Medicine* (2020) **41** 365-73. PMID: 33242381
33. Barry VW, Baruth M, Beets MW, Drustine JL, Liu J, Blair SN. **Fitness vs. fatness on all-cause mortality: A meta-analysis**. *Progress in Cardiovascular Diseases* (2014) **56** 382-90. PMID: 24438729
34. Bull FC, Al-Ansari SS, Biddle S, Borodulin K, Buman MP, Cardon G. **World Health Organization 2020 guidelines on physical activity and sedentary behaviour**. *Br J Sports Med.* (2020) **54** 1451-62. PMID: 33239350
|
---
title: Plant‐based diets could ameliorate the risk factors of cardiovascular diseases
in adults with chronic diseases
authors:
- Mostafa Lotfi
- Mehran Nouri
- Abduladheem Turki Jalil
- Abbas Rezaianzadeh
- Siavash Babajafari
- Masoumeh Ghoddusi Johari
- Shiva Faghih
journal: Food Science & Nutrition
year: 2022
pmcid: PMC10002912
doi: 10.1002/fsn3.3164
license: CC BY 4.0
---
# Plant‐based diets could ameliorate the risk factors of cardiovascular diseases in adults with chronic diseases
## Abstract
Adherence to plant‐based diets is recommended to prevent and control chronic diseases. However, not all plant‐based foods are healthy for this purpose. This study investigated the relationship between plant‐based diets and risk factors for cardiovascular diseases (CVDs) in adults with chronic diseases. This cross‐sectional study was performed on 3678 males and females (age range: 40–70 years) with chronic diseases who participated in the Kharameh cohort study. A validated semiquantitative food‐frequency questionnaire was used to calculate the plant‐based diet index (PDI), healthy plant‐based diet index (hPDI), and unhealthy plant‐based diet index (uPDI). Lipid profile, fasting blood sugar (FBS), blood pressure, and anthropometric indices were measured. Multivariable‐adjusted logistic regression analysis was performed to determine the association between plant‐based diets and CVDs risk factors. Higher adherence to the PDI was inversely associated with the level of FBS (odds ratio [OR] = 0.42; $95\%$ confidence interval [CI]: 0.33–0.53; $p \leq .001$). A significant decrease was observed for total cholesterol in those with higher adherence to hPDI (OR = 0.80; $95\%$ CI: 0.65–0.98; $$p \leq .035$$). Additionally, the score of uPDI was positively related to FBS (OR = 1.23; $95\%$ CI: 1.00–1.53; $$p \leq .01$$), total cholesterol (OR = 1.23; $95\%$ CI: 1.01–1.49; $$p \leq .061$$), and low‐density lipoprotein (OR = 1.39; $95\%$ CI: 1.13–1.71; $$p \leq .009$$). It was concluded that adherence to PDI and hPDI was related to a lower level of FBS and total cholesterol, respectively. Moreover, the findings suggested that regular intake of the uPDI was correlated with some risk factors for CVDs in adults with chronic diseases.
It was concluded that adherence to the PDI and hPDI was related to decreasing FBS and TC, respectively. Moreover, the findings suggested that regular intake of the uPDI was correlated with some risk factors of CVDs in adults with chronic diseases.
## INTRODUCTION
Cardiovascular diseases (CVDs) are among the leading causes of mortality and the main causes of health system costs (Abdulmuhsin, 2016). In addition, the prevalence of CVDs has increased in recent decades. Numerous investigations have shown that the control of CVD risk factors, such as blood pressure, blood lipids, obesity, and blood sugar, could lead to the improvement of patients’ complications (Khunti et al., 2018; Randhawa et al., 2020; Shmakova et al., 2022).
Lifestyle risk factors, such as smoking, inadequate physical activity, alcohol consumption, and unhealthy diets, are modifiable and known factors for the control of some chronic noncommunicable diseases (Bauman, 2004; Farazian et al., 2019; Wu et al., 2015). Diet plays a key role in the incidence and control of chronic diseases. Various studies on the relationship of plant‐ and animal‐based diets with chronic diseases have reported different results. Some studies have shown that restricting animal foods and consuming more plant‐based foods could reduce body mass index (BMI), blood pressure, blood sugar, or inflammation (Hematdar et al., 2018; Picasso et al., 2019; Yokoyama et al., 2014). However, another study showed that the moderate consumption of animal foods, such as red meat, has not adversely affected cardio‐metabolic factors (Hassanzadeh‐Rostami et al., 2021), and some others have not reported any association (Chiang et al., 2013; Vinagre et al., 2013; Yang et al., 2011).
In comparison to omnivorous diets, vegetarian food regimens are full of fibers, phytosterols, magnesium, Fe3+, folic acid, vitamins C and E, omega‐6 polyunsaturated fatty acids (PUFAs), phytochemicals, and antioxidants but low in total fat and saturated fatty acids (SFAs), sodium, Fe2+, zinc, vitamins A, B12, and D, and little or no cholesterol (Borazjani et al., 2022; Kamalipour & Akhondzadeh, 2011). Therefore, plant‐based diets could have numerous healthy effects (Jalilpiran et al., 2017; Wang et al., 2015). Recently, it has been known that not every plant food retains the same beneficial features. Some plant foods, such as refined grains and sweetened beverages, have adverse effects on health (Hemler & Hu, 2019). *In* general, due to the contradictory results and limitations of previous studies, the present study examined three types of plant‐based diet indices, including the general plant‐based diet index (PDI), healthy plant‐based diet index (hPDI), and unhealthy plant‐based diet index (uPDI), and their relationships with blood lipids, blood pressure, glycemic control, and anthropometric indices in patients with chronic disease.
## Study design and study population
The present cross‐sectional study was carried out on the information of participants of the Kharameh cohort study. The Kharameh cohort study is a branch of Prospective Epidemiological Research Studies in Iran (PERSIAN) (Poustchi et al., 2018) in which 10,663 individuals aged 40–70 years were enrolled from 2014 to 2017 (Rezaianzadeh et al., 2021). Demographic information, physical activity, smoking status, and medical history of the participants were assessed in the PERSIAN cohort study. Additionally, physical examinations (i.e., weight, height, waist circumference [WC], hip circumference [HC], systolic blood pressure [SBP], and diastolic blood pressure [DBP]), biochemical assessments (i.e., fasting blood sugar [FBS], total cholesterol, triglyceride [TG], high‐density lipoprotein [HDL], low‐density lipoprotein [LDL], and alkaline phosphatase), and dietary evaluation were performed (Nikbakht et al., 2020).
Inclusion criteria for the Kharameh cohort study were the age of 40–70 years, living in Kharameh for the last 9 months, and having Iranian nationality. In addition, having a history of one or more types of CVDs (heart failure, angina, and myocardial infarction), hypertension, or diabetes was needed to be included in the present study.
Participants with mental disorders or untreated illnesses in the acute phase and also who did not complete the assessments were excluded from the study. Furthermore, we excluded the data of the participants who under‐ or over‐reported their energy intakes (<800 kcal or > 4200 kcal/day) (Figure 1). This study was approved by the Ethics Committee of Shiraz University of Medical Sciences, Fars, Iran (code: IR.SUMS.REC.1399.1115).
**FIGURE 1:** *Flow diagram of the study.*
## Dietary intake and plant‐based diet assessment
Dietary intakes were collected using a 130‐item semiquantitative food‐frequency questionnaire (FFQ). Consistent with household measures, the records from completed FFQs were transformed into grams. Then, energy and nutrient intakes were calculated using the adapted version of Nutritionist IV for Iranian software (version 7.0; N‐Squared Computing).
In this study, Satije et al. approach was used to calculate the PDI, hPDI, and uPDI (Satija et al., 2016). A total of 130 food items were categorized into 18 food groups; then, the food groups were divided into three principal categories, including healthy plant foods (i.e., whole grains, nuts, vegetables, vegetable oils, fruits, legumes, and tea/coffee), unhealthy plant foods (i.e., refined grains, sugar‐sweetened drinks, fruit juices, potatoes, and sweets/desserts), and animal foods (i.e., egg, dairy, fish/kinds of seafood, animal fats, meat, and miscellaneous animal‐based foods). In the PDI and hPDI, the maximum and minimum intakes of plant foods and healthy plant foods received scores of 10 and 1, respectively. In the uPDI, scores of 1 and 10 were considered for maximum and minimum intakes of unhealthy plant foods, respectively. Rankings were summed up to get a score ranging from 18 to 180 for each of the PDI, hPDI, and uPDI. Overall, a higher score for each index indicated higher adherence to that dietary pattern.
## Anthropometric and biochemical assessments
The trained workforce measured participants’ height, weight, WC, HC, and blood pressure. Height, weight, HC, and WC were measured to the nearest 0.1 cm or 0.1 kg in light clothing with no shoes. BMI was calculated by dividing the weight by height squared. Blood pressure was measured after a 10‐min rest in the sitting position using a standard calibrated sphygmomanometer (Riester Model, Germany). For laboratory assessments, after 10–14 h overnight fast, a 20‐ml blood sample was taken from each participant. Serum separation was done in a minimum of 30 min or a maximum of 2 h. Then, 0.5 ml of serum was used to do the biochemical tests, and the remaining was transferred to cryotubes for storage at −80°C until further analyses. Serum FBS, TG, and cholesterol were measured using the Mindray BS‐380 device by Pars Azmoon kits. HDL, TG, and total cholesterol levels were determined using an enzymatic technique, and LDL level was calculated by *Friedewald formula* (Friedewald et al., 1972).
## Covariates
We considered age, gender, physical activity, educational level, and smoking status as covariates. Demographic characteristics, including age, gender, educational level, and smoking status of the participants, were obtained by a questionnaire. The duration of the participants’ education was asked to determine their educational level. The participants’ smoking status was determined by answering yes or no. Participants’ daily physical activity was assessed by a self‐reported validated questionnaire. Participants reported the time spent on all activities such as sleeping, running, and walking on a typical day (24 h), during the previous year. Each activity was given a value in metabolic equivalent tasks (METs), and then total metabolic equivalent tasks per day were computed (Kazemi Karyani et al., 2019).
## Statistical analyses
BMI ≥30 kg/m2, SBP ≥135 mmHg, DBP ≥85 mmHg, FBS ≥126 mg/dl, TG ≥150 mg/dl, total cholesterol ≥200 mg/dl, LDL‐C ≥ 130 mg/dl, HDL‐C ˂ 40 mg/dl for men and 50 mg/dl for women, TG to HDL ratio ≥5, and total cholesterol to HDL ratio ≥3 were considered as abnormal levels (Huang, 2009; Kohansal et al., 2022; Salcedo‐Cifuentes et al., 2020). All data were analyzed using SPSS for Windows software (version 20.0), and a p‐value less than.05 was considered statistically significant. The Kolmogorov–Smirnov test was used to examine the normal distribution of the variables. For comparing the baseline characteristics between the male and female patients, an independent‐samples t‐test and chi‐squared test were used for continuous variables and categorical variables, respectively. One‐way analysis of variance was used to compare the intake of nutrients and food groups across the quartiles of plant‐based diet indices (i.e., PDI, hPDI, and uPDI). Two different multivariable logistic regression models were used to assess the relationship between plant‐based diet indices and odds of CVD risk factors. Gender, age, physical activity, total energy intake, and smoking status were included in the regression models as confounders.
## RESULTS
The data of 3687 participants with one or more types of CVDs, hypertension, or diabetes were included in the analysis (Figure 1). Table 1 shows the baseline characteristics of the participants. Male participants had higher age ($$p \leq .01$$), educational level ($p \leq .001$), weight ($p \leq .001$), height ($p \leq .001$), and physical activity ($$p \leq .004$$). However, female participants had higher WC, HC, and waist‐to‐hip ratios ($p \leq .001$ for all).
**TABLE 1**
| Variables | Male | Female | p value |
| --- | --- | --- | --- |
| Age (years) | 56.20 ± 7.74 | 55.50 ± 8.01 | .01 |
| Education (years) | 5.56 ± 5.01 | 2.48 ± 3.24 | <.001 |
| Weight (kg) | 74.55 ± 12.46 | 68.02 ± 11.64 | <.001 |
| Height (cm) | 170.36 ± 6.77 | 156.46 ± 5.55 | <.001 |
| Waist circumference (cm) | 97.50 ± 11.43 | 99.67 ± 11.34 | <.001 |
| Hip circumference (cm) | 100.89 ± 7.29 | 102.19 ± 8.97 | <.001 |
| WHR | 0.96 ± 0.06 | 0.97 ± 0.07 | <.001 |
| Physical activity (Met. h/day) | 37.74 ± 7.64 | 36.87 ± 3.31 | .004 |
| Smoking | Smoking | Smoking | Smoking |
| Yes | 571 (84.7%) | 103 (15.3%) | <.001 |
| No | 603 (20%) | 2410 (80%) | <.001 |
According to Table 2, the participants in the highest PDI quartiles had higher intakes of carbohydrates, fiber, omega‐6/omega‐3 fatty acid (FA), beta carotene, vitamin E, C, B9, potassium, and magnesium ($p \leq .001$ for all, except for vitamins E and B9). In addition, the participants in the highest PDI quartiles had lower intakes of energy, protein, fat, cholesterol, SFAs, monounsaturated fatty acids (MUFAs), calcium, and selenium ($p \leq .001$ for all).
**TABLE 2**
| Food groups | PDI | PDI.1 | PDI.2 | PDI.3 | hPDI | hPDI.1 | hPDI.2 | hPDI.3 | hPDI.4 | uPDI | uPDI.1 | uPDI.2 | uPDI.3 | uPDI.4 | uPDI.5 |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| Food groups | Q1 | Q2 | Q3 | Q4 | p value | Q1 | Q2 | Q3 | Q4 | p value | Q1 | Q2 | Q3 | Q4 | p value |
| Energy (Kcal/d) | 2432.29 ± 20.11 | 2116.28 ± 21.38 | 2081.52 ± 21.73 | 2165.77 ± 21.65 | <.001 | 2147.11 ± 19.95 | 2059.54 ± 21.05 | 2223.67 ± 20.89 | 2480.36 ± 10.88 | <.001 | 2233.15 ± 20.68 | 2097.04 ± 22.02 | 2109.58 ± 20.68 | 2393.72 ± 22.31 | <.001 |
| CHO (gr/d) | 354.06 ± 0.94 | 364.48 ± 0.91 | 370.77 ± 0.80 | 379.58 ± 0.94 | <.001 | 363.26 ± 0.81 | 365.90 ± 0.86 | 367.36 ± 0.99 | 371.78 ± 1.24 | <.001 | 359.68 ± 0.98 | 363.75 ± 0.80 | 367.75 ± 0.90 | 376.27 ± 1.02 | <.001 |
| Protein (gr/d) | 76.21 ± 0.32 | 72.15 ± 0.30 | 69.16 ± 0.26 | 65.14 ± 0.31 | <.001 | 72.51 ± 0.27 | 70.63 ± 0.29 | 70.57 ± 0.34 | 69.26 ± 0.43 | <.001 | 72.96 ± 0.32 | 71.96 ± 0.30 | 69.78 ± 0.32 | 68.68 ± 0.35 | <.001 |
| Fat (gr/d) | 58.02 ± 0.38 | 55.66 ± 0.38 | 54.45 ± 0.33 | 52.75 ± 0.38 | <.001 | 55.42 ± 0.32 | 55.39 ± 0.35 | 55.45 ± 0.39 | 55.07 ± 0.48 | 0.90 | 58.30 ± 0.39 | 56.43 ± 0.31 | 55.33 ± 0.36 | 50.90 ± 0.39 | <.001 |
| Fiber (gr/d) | 24.32 ± 0.19 | 25.01 ± 0.21 | 25.33 ± 0.18 | 25.76 ± 0.20 | <.001 | 23.31 ± 0.18 | 23.75 ± 0.16 | 25.84 ± 0.17 | 28.55 ± 0.25 | <.001 | 26.82 ± 0.20 | 25.64 ± 0.17 | 24.77 ± 0.19 | 22.78 ± 0.20 | <.001 |
| Cholesterol (mg/d) | 251.88 ± 3.01 | 209.54 ± 2.68 | 187.23 ± 2.20 | 159.00 ± 2.27 | <.001 | 232.78 ± 2.68 | 211.78 ± 2.61 | 192.35 ± 2.51 | 164.84 ± 3.15 | <.001 | 222.70 ± 2.99 | 217.79 ± 2.51 | 201.92 ± 2.65 | 171.75 ± 2.83 | <.001 |
| Trans fat (mg/d) | 0.23 ± 0.007 | 0.21 ± 0.006 | 0.21 ± 0.005 | 0.21 ± 0.007 | .22 | 0.25 ± 0.006 | 0.22 ± 0.006 | 0.20 ± 0.006 | 0.18 ± 0.007 | <.001 | 0.23 ± 0.006 | 0.23 ± 0.007 | 0.22 ± 0.007 | 0.18 ± 0.006 | <.001 |
| SFA (gr/d) | 20.65 ± 0.19 | 19.54 ± 0.19 | 18.81 ± 0.17 | 18.03 ± 0.19 | <.001 | 19.44 ± 0.16 | 19.50 ± 0.18 | 19.43 ± 0.19 | 18.78 ± 0.24 | .05 | 20.43 ± 0.19 | 19.57 ± 0.17 | 19.43 ± 0.19 | 17.69 ± 0.20 | <.001 |
| MUFA (gr/d) | 16.40 ± 0.16 | 15.64 ± 0.16 | 15.37 ± 0.14 | 15.00 ± 0.16 | <.001 | 15.66 ± 0.13 | 15.67 ± 0.14 | 15.63 ± 0.16 | 15.59 ± 0.20 | .98 | 16.96 ± 0.16 | 16.21 ± 0.13 | 15.63 ± 0.14 | 13.57 ± 0.16 | <.001 |
| PUFA (gr/d) | 9.14 ± 0.11 | 9.10 ± 0.11 | 9.05 ± 0.10 | 8.99 ± 0.11 | .77 | 8.94 ± 0.09 | 8.83 ± 0.10 | 9.02 ± 0.11 | 9.65 ± 0.14 | <.001 | 10.06 ± 0.11 | 9.53 ± 0.09 | 9.00 ± 0.10 | 7.54 ± 0.11 | <.001 |
| Omega 6/3 | 15.52 ± 0.51 | 18.66 ± 0.59 | 24.49 ± 0.96 | 33.28 ± 1.57 | <.001 | 18.22 ± 0.81 | 21.31 ± 0.95 | 23.19 ± 0.74 | 30.58 ± 1.50 | <.001 | 21.64 ± 1.13 | 20.50 ± 0.65 | 24.07 ± 1.08 | 24.51 ± 0.94 | .01 |
| Beta carotene (IU/d) | 4527.90 ± 80.33 | 4758.50 ± 85.89 | 4810.83 ± 81.01 | 5103.98 ± 109.77 | <.001 | 4293.87 ± 66.06 | 4433.73 ± 74.17 | 4932.61 ± 79.71 | 5823.78 ± 141.35 | <.001 | 5380.57 ± 87.84 | 4967.06 ± 77.13 | 4783.80 ± 103.53 | 3919.23 ± 78.18 | <.001 |
| Vit E (mg/d) | 7.11 ± 0.07 | 7.18 ± 0.07 | 7.28 ± 0.07 | 7.43 ± 0.08 | .01 | 6.93 ± 0.05 | 7.02 ± 0.06 | 7.31 ± 0.07 | 7.93 ± 0.10 | <.001 | 7.77 ± 0.07 | 7.58 ± 0.06 | 7.24 ± 0.07 | 6.29 ± 0.07 | <.001 |
| Vit c (mg/d) | 109.95 ± 1.57 | 119.07 ± 1.93 | 120.03 ± 1.70 | 126.74 ± 1.87 | <.001 | 108.83 ± 1.40 | 112.97 ± 1.61 | 121.61 ± 1.70 | 136.69 ± 2.45 | <.001 | 131.48 ± 1.76 | 126.05 ± 1.66 | 117.68 ± 1.67 | 96.63 ± 1.72 | <.001 |
| Vit B9 (mcg/d) | 561.47 ± 3.25 | 561.12 ± 3.09 | 569.26 ± 2.96 | 571.67 ± 3.49 | .04 | 551.30 ± 2.87 | 554.02 ± 2.75 | 571.66 ± 3.01 | 596.14 ± 4.29 | <.001 | 562.39 ± 3.01 | 558.40 ± 2.91 | 563.79 ± 3.35 | 579.05 ± 3.54 | <.001 |
| K (mcg/d) | 3260.77 ± 25.15 | 3332.48 ± 28.20 | 3340.08 ± 24.48 | 3440.74 ± 28.13 | <.001 | 3170.34 ± 21.95 | 3212.42 ± 22.69 | 3405.30 ± 25.61 | 3681.40 ± 35.27 | <.001 | 3597.82 ± 26.82 | 3447.61 ± 21.24 | 3319.26 ± 25.60 | 2950.46 ± 26.31 | <.001 |
| Ca (mg/d) | 1090.86 ± 6.83 | 1056.69 ± 6.97 | 1022.08 ± 6.23 | 958.77 ± 6.47 | <.001 | 1034.62 ± 5.91 | 1026.73 ± 6.35 | 1044.79 ± 6.92 | 1033.70 ± 8.84 | .31 | 1032.07 ± 6.51 | 1022.03 ± 6.32 | 1025.48 ± 6.75 | 1061.00 ± 7.72 | <.001 |
| Mg (mg/d) | 318.36 ± 1.56 | 322.50 ± 1.70 | 323.79 ± 1.51 | 328.31 ± 1.85 | <.001 | 312.26 ± 1.40 | 313.70 ± 1.30 | 326.43 ± 1.52 | 347.40 ± 2.31 | <.001 | 338.38 ± 1.66 | 327.96 ± 1.29 | 321.46 ± 1.75 | 301.79 ± 1.59 | <.001 |
| Se (mcg/d) | 107.05 ± 0.69 | 100.02 ± 0.65 | 97.27 ± 0.61 | 90.75 ± 0.67 | <.001 | 102.38 ± 0.58 | 99.72 ± 0.61 | 97.48 ± 0.71 | 95.69 ± 0.91 | <.001 | 97.95 ± 0.63 | 98.26 ± 0.62 | 97.82 ± 0.66 | 102.92 ± 0.83 | <.001 |
The highest consumption of energy, carbohydrate, fiber, omega‐6/omega‐3 FA, PUFAs, beta carotene, vitamin E, vitamin C, vitamin B9, potassium, and magnesium and the lowest consumption of protein, cholesterol, trans fatty acids, and selenium were observed in the highest hPDI quartile ($p \leq .001$ for all). Moreover, the highest intake of energy, carbohydrate, omega‐6/omega‐3 FA, vitamin B9, calcium, and selenium and the lowest intake of protein, fat, fiber, cholesterol, trans fatty acids, SFAs, MUFAs, PUFAs, beta carotene, vitamin E, vitamin C, potassium, and magnesium were observed in the highest uPDI quartile ($p \leq .001$ for all, except for omega‐6/omega‐3 FA) (Table 2).
Adherence to the PDI was significantly associated with higher consumption of fruits, vegetables, legumes, vegetable oils, tea/coffee, fruit juices, potatoes, sugar‐sweetened beverages, and sweet desserts and lower consumption of animal fat, dairy, egg, fish/seafood, meat, and animal‐based foods ($p \leq .001$ for all, except for fruits and fruit juices). Furthermore, adherence to the hPDI was significantly associated with higher consumption of whole grains, fruits, vegetables, nuts, legumes, vegetable oils, and tea/coffee and lower consumption of refined grains, potatoes, sugar‐sweetened beverages, sweet desserts, animal fat, dairy, egg, fish/seafood, meat, and animal‐based foods ($p \leq .001$ except for dairy and fish/seafood). In addition, the participants in the highest quartile of the uPDI had higher levels of refined grains ($p \leq .001$), potatoes ($$p \leq .03$$), sugar‐sweetened beverages ($$p \leq .009$$), and sweet desserts ($p \leq .001$) and lower intakes of whole grains, fruits, vegetables, nuts, legumes, vegetable oils, animal fat, dairy, egg, fish/seafood, meat, and animal‐based foods ($p \leq .001$ for all, except for vegetable oils) (Table 3).
**TABLE 3**
| Food groups | PDI | PDI.1 | PDI.2 | PDI.3 | hPDI | hPDI.1 | hPDI.2 | hPDI.3 | hPDI.4 | uPDI | uPDI.1 | uPDI.2 | uPDI.3 | uPDI.4 | uPDI.5 |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| Food groups | Q1 | Q2 | Q3 | Q4 | p value | Q1 | Q2 | Q3 | Q4 | p value | Q1 | Q2 | Q3 | Q4 | p value |
| Whole grains (g/d) | 32.03 ± 2.20 | 32.71 ± 2.01 | 34.63 ± 1.86 | 33.55 ± 1.66 | .79 | 27.73 ± 1.62 | 25.33 ± 1.20 | 30.59 ± 1.79 | 55.35 ± 3.30 | <.001 | 45.12 ± 2.14 | 35.10 ± 1.81 | 31.14 ± 1.99 | 19.68 ± 1.79 | <.001 |
| Fruits (g/d) | 304.18 ± 5.81 | 321.31 ± 6.62 | 326.98 ± 5.84 | 332.66 ± 6.23 | .005 | 284.73 ± 5 | 296.47 ± 5.24 | 339.16 ± 6.14 | 383.83 ± 8.25 | <.001 | 362.68 ± 6.23 | 344.22 ± 5.69 | 317.36 ± 5.92 | 251.15 ± 5.83 | <.001 |
| Vegetables (g/d) | 439.38 ± 6.72 | 469.11 ± 8.47 | 469.75 ± 6.40 | 510.58 ± 7.70 | <.001 | 419.53 ± 6.47 | 440.66 ± 6.40 | 493.72 ± 6.77 | 558.25 ± 9.33 | <.001 | 515.44 ± 6.99 | 488.97 ± 6.60 | 468.87 ± 7.27 | 400.80 ± 7.78 | <.001 |
| Nuts (g/d) | 3.70 ± 0.18 | 3.83 ± 0.15 | 3.99 ± 0.15 | 4.14 ± 0.18 | .28 | 3.13 ± 0.11 | 3.65 ± 0.16 | 4.13 ± 0.18 | 5.17 ± 0.25 | <.001 | 4.77 ± 0.17 | 4.37 ± 0.17 | 3.86 ± 0.16 | 2.48 ± 0.16 | <.001 |
| Legumes (g/d) | 26.11 ± 0.71 | 27.72 ± 0.70 | 29.70 ± 0.63 | 30.90 ± 0.72 | <.001 | 26.61 ± 0.58 | 26.34 ± 0.57 | 29.61 ± 0.68 | 32.84 ± 1.01 | <.001 | 32.51 ± 0.74 | 29.24 ± 0.63 | 27.37 ± 0.66 | 24.29 ± 0.69 | <.001 |
| Vegetable oils (g/d) | 14.38 ± 0.30 | 15.32 ± 0.31 | 16.43 ± 0.30 | 17.95 ± 0.33 | <.001 | 14.42 ± 0.25 | 16.13 ± 0.30 | 16.46 ± 0.32 | 17.44 ± 0.41 | <.001 | 16.13 ± 0.30 | 15.94 ± 0.31 | 16.46 ± 0.33 | 15.21 ± 0.31 | .04 |
| Tea and coffee (g/d) | 583.25 ± 16.27 | 673.72 ± 18.80 | 754.14 ± 16.95 | 929.77 ± 22.01 | <.001 | 664.60 ± 13.97 | 728.93 ± 17.90 | 738.85 ± 18.88 | 810.25 ± 27.03 | <.001 | 740.66 ± 20.40 | 707.97 ± 17.44 | 738.25 ± 17.93 | 721.21 ± 18.78 | .57 |
| Fruit juices (g/d) | 1.35 ± 0.25 | 2.54 ± 0.35 | 2.32 ± 0.28 | 2.17 ± 0.28 | .01 | 2.42 ± 0.26 | 2.28 ± 0.28 | 1.55 ± 0.22 | 1.85 ± 0.41 | .12 | 2.24 ± 0.29 | 2.55 ± 0.31 | 1.87 ± 0.26 | 1.55 ± 0.28 | .08 |
| Refined grains (g/d) | 485.74 ± 4.68 | 484.65 ± 4.90 | 481.18 ± 4.38 | 473.98 ± 4.66 | .29 | 493.57 ± 3.91 | 495.18 ± 4.27 | 481.72 ± 4.59 | 445.13 ± 6.24 | <.001 | 433.62 ± 4.57 | 466.47 ± 3.60 | 485.38 ± 4.26 | 548.25 ± 2.19 | <.001 |
| Potatoes (g/d) | 20.29 ± 0.72 | 23.27 ± 0.77 | 26.59 ± 0.77 | 33.27 ± 0.84 | <.001 | 29.76 ± 0.76 | 27.77 ± 0.74 | 23.73 ± 0.74 | 18.50 ± 0.85 | <.001 | 24.01 ± 0.74 | 25.51 ± 0.72 | 25.75 ± 0.81 | 27.26 ± 0.87 | .03 |
| Sugar‐sweetened beverages (g/d) | 52.45 ± 2.56 | 60.51 ± 2.97 | 60.79 ± 2.33 | 80.09 ± 3.31 | <.001 | 83.38 ± 2.72 | 71.10 ± 2.88 | 49.49 ± 2.22 | 36.47 ± 3 | <.001 | 55.14 ± 2.22 | 64.88 ± 3.03 | 66.78 ± 2.78 | 65.32 ± 3.16 | .009 |
| Sweet desserts (g/d) | 29.19 ± 1.12 | 37.09 ± 1.26 | 44.14 ± 1.22 | 54.61 ± 1.33 | <.001 | 43.01 ± 1.02 | 44.88 ± 1.18 | 38.94 ± 1.24 | 33.85 ± 1.74 | <.001 | 36.58 ± 1.18 | 38.58 ± 1.02 | 42.90 ± 1.19 | 45.32 ± 1.59 | <.001 |
| Animal fat (g/d) | 2.40 ± 0.13 | 1.64 ± 0.10 | 1.27 ± 0.07 | 0.77 ± 0.05 | <.001 | 1.89 ± 0.09 | 1.87 ± 0.10 | 1.53 ± 0.10 | 0.68 ± 0.08 | <.001 | 1.96 ± 0.11 | 1.68 ± 0.09 | 1.58 ± 0.09 | 0.97 ± 0.09 | <.001 |
| Dairy (g/d) | 264.01 ± 4.76 | 241.75 ± 4.90 | 206.85 ± 4.12 | 169.55 ± 3.72 | <.001 | 231.86 ± 3.93 | 220.75 ± 4.32 | 224.99 ± 4.76 | 206.57 ± 5.72 | .002 | 250.48 ± 4.70 | 233.58 ± 4.33 | 219.74 ± 4.74 | 181.79 ± 4.21 | <.001 |
| Egg (g/d) | 26.79 ± 0.62 | 20.17 ± 0.59 | 17.16 ± 0.47 | 12.63 ± 0.43 | <.001 | 24.83 ± 0.53 | 20.85 ± 0.56 | 16.84 ± 0.49 | 13.03 ± 0.62 | <.001 | 21.51 ± 0.60 | 21.77 ± 0.56 | 18.95 ± 0.53 | 15.66 ± 0.54 | <.001 |
| Fish or sea foods (g/d) | 4.16 ± 0.20 | 3.32 ± 0.19 | 2.67 ± 0.15 | 1.83 ± 0.13 | <.001 | 3.55 ± 0.15 | 3.04 ± 0.16 | 2.86 ± 0.17 | 2.52 ± 0.22 | .001 | 4.22 ± 0.19 | 3.28 ± 0.17 | 2.70 ± 0.17 | 1.81 ± 0.15 | <.001 |
| Meats (g/d) | 64 ± 1.07 | 54.20 ± 1.03 | 47.60 ± 0.84 | 41.21 ± 0.92 | <.001 | 57.24 ± 0.90 | 53.76 ± 0.95 | 50.82 ± 1.06 | 44.54 ± 1.18 | <.001 | 57.72 ± 1.01 | 53.13 ± 0.94 | 50.29 ± 0.99 | 44.23 ± 1.06 | <.001 |
| Animal‐based foods (g/d) | 4.16 ± 0.19 | 3.29 ± 0.17 | 2.77 ± 0.11 | 2.31 ± 0.14 | <.001 | 3.90 ± 0.16 | 3.65 ± 0.18 | 2.82 ± 0.14 | 1.89 ± 0.14 | <.001 | 3.79 ± 0.19 | 3.51 ± 0.16 | 3.27 ± 0.16 | 2.04 ± 0.12 | <.001 |
As shown in Table 4, adherence to the PDI was related to a lower level of FBS ($p \leq .001$) in the crude and adjusted models. However, there was no significant association between adherence to the PDI and any CVD risk factors. Additionally, after adjusting for gender, age, energy intake, smoking, and physical activity, higher scores of the hPDI were associated with a lower total cholesterol level (odds ratio [OR] = 0.80; $95\%$ confidence interval [CI]: 0.65–0.98; $$p \leq .035$$) and cholesterol to HDL ratio (OR = 0.74; $95\%$ CI: 0.57–0.95; $$p \leq .05$$). In the adjusted model, the scores of the uPDI were positively related to FBS (OR = 1.25; $95\%$ CI: 1.01–1.55; $$p \leq .01$$), total cholesterol (OR = 1.24; $95\%$ CI: 1.0–1.50; $$p \leq .05$$), and LDL (OR = 1.40; $95\%$ CI: 1.13–1.72; $$p \leq .009$$).
**TABLE 4**
| Variables | PDI | PDI.1 | PDI.2 | PDI.3 | PDI.4 | hPDI | hPDI.1 | hPDI.2 | hPDI.3 | Unnamed: 10 | uPDI | uPDI.1 | uPDI.2 | uPDI.3 | uPDI.4 |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| Variables | Q1 | Q2 | Q3 | Q4 | p trend | Q1 | Q2 | Q3 | Q4 | p trend | Q1 | Q2 | Q3 | Q4 | p trend |
| BMI (k/m2) | | | | | | | | | | | | | | | |
| Crude | 1.00 | 1.01 (0.83, 1.23) | 1.07 (0.88, 1.29) | 0.93 (0.76, 1.13) | .69 | 1.00 | 0.95 (0.79, 1.14) | 1.12 (0.93, 1.35) | 1.18 (0.96, 1.44) | .05 | 1.00 | 1.09 (0.90, 1.33) | 0.92 (0.75, 1.11) | 0.88 (0.73, 1.07) | .10 |
| Adjusted | 1.00 | 0.95 (0.77, 1.16) | 0.96 (0.79, 1.18) | 0.84 (0.69, 1.03) | .14 | 1.00 | 0.93 (0.77, 1.13) | 1.14 (0.94, 1.38) | 1.18 (0.95, 1.45) | .04 | 1.00 | 1.01 (0.83, 1.24) | 0.87 (0.71, 1.06) | 0.90 (0.74, 1.10) | .16 |
| SBP (mmHg) | | | | | | | | | | | | | | | |
| Crude | 1.00 | 0.96 (0.77, 1.21) | 0.95 (0.77, 1.18) | 1.08 (0.87, 1.34) | .57 | 1.00 | 1.20 (0.98, 1.48) | 0.90 (0.72, 1.12) | 0.89 (0.71, 1.13) | .13 | 1.00 | 1.13 (0.91, 1.40) | 0.94 (0.75, 1.17) | 1.01 (0.81, 1.26) | .74 |
| Adjusted | 1.00 | 0.94 (0.75, 1.19) | 0.96 (0.77, 1.20) | 1.08 (0.87, 1.35) | .21 | 1.00 | 1.16 (0.94, 1.43) | 0.85 (0.69, 1.06) | 0.87 (0.69, 1.10) | .07 | 1.00 | 1.16 (0.93, 1.44) | 0.93 (0.74, 1.16) | 1.00 (0.80, 1.25) | .62 |
| DBP (mmHg) | | | | | | | | | | | | | | | |
| Crude | 1.00 | 0.90 (0.68, 1.20) | 0.92 (0.70, 1.20) | 1.15 (0.88, 1.50) | .36 | 1.00 | 1.28 (0.99, 1.65) | 0.97 (0.74, 1.27) | 0.87 (0.65, 1.17) | .23 | 1.00 | 1.15 (0.88, 1.50) | 0.90 (0.68, 1.19) | 1.03 (0.78, 1.35) | .75 |
| Adjusted | 1.00 | 0.93 (0.70, 1.24) | 0.96 (0.73, 1.26) | 1.20 (0.92, 1.57) | .20 | 1.00 | 1.29 (1.00, 1.67) | 0.97 (0.74, 1.27) | 0.87 (0.65, 1.17) | .27 | 1.00 | 1.18 (0.90, 1.55) | 0.92 (0.69, 1.21) | 1.03 (0.78, 1.35) | .76 |
| FBS(mg/dl) | | | | | | | | | | | | | | | |
| Crude | 1.00 | 0.73 (0.60, 0.91) | 0.56 (0.45, 0.69) | 0.43 (0.34, 0.55) | <.001 | 1.00 | 0.71 (0.57, 0.87) | 1.15 (0.94, 1.42) | 1.06 (0.85, 1.33) | .09 | 1.00 | 0.90 (0.72, 1.13) | 1.11 (0.89, 1.38) | 1.23 (1.00, 1.53) | .01 |
| Adjusted | 1.00 | 0.71 (0.57, 0.87) | 0.53 (0.43, 0.66) | 0.42 (0.33, 0.53) | <.001 | 1.00 | 0.69 (0.55, 0.86) | 1.13 (0.92, 1.38) | 1.05 (0.84, 1.31) | .12 | 1.00 | 0.88 (0.70, 1.10) | 1.09 (0.87, 1.35) | 1.25 (1.01, 1.55) | .01 |
| Triglycerides (mg/dl) | | | | | | | | | | | | | | | |
| Crude | 1.00 | 1.01 (0.83, 1.22) | 0.95 (0.79, 1.14) | 1.01 (0.83, 1.22) | .91 | 1.00 | 0.90 (0.75, 1.09) | 1.08 (0.90, 1.30) | 1.15 (0.94, 1.40) | .07 | 1.00 | 0.99 (0.82, 1.20) | 1.06 (0.88, 1.29) | 0.90 (0.74, 1.09) | .49 |
| Adjusted | 1.00 | 1.01 (0.83, 1.23) | 0.94 (0.78, 1.13) | 1.01 (0.83, 1.22) | .87 | 1.00 | 0.91 (0.75, 1.09) | 1.09 (0.90, 1.31) | 1.16 (0.95, 1.41) | .06 | 1.00 | 0.97 (0.80, 1.17) | 1.06 (0.88, 1.29) | 0.90 (0.74, 1.10) | .55 |
| Total cholesterol (mg/dl) | | | | | | | | | | | | | | | |
| Crude | 1.00 | 1.16 (0.95, 1.40) | 0.86 (0.71, 1.04) | 1.06 (0.88, 1.29) | .74 | 1.00 | 0.87 (0.72, 1.05) | 0.88 (0.73, 1.06) | 0.80 (0.65, 0.98) | .03 | 1.00 | 1.17 (0.96, 1.41) | 1.11 (0.91, 1.34) | 1.23 (1.01, 1.49) | .06 |
| Adjusted | 1.00 | 1.07 (0.88, 1.31) | 0.78 (0.65, 0.95) | 0.96 (0.79, 1.17) | .18 | 1.00 | 0.84 (0.70, 1.01) | 0.86 (0.72, 1.04) | 0.79 (0.64, 0.96) | .02 | 1.00 | 1.13 (0.93, 1.38) | 1.06 (0.88, 1.29) | 1.24 (1.02, 1.50) | .05 |
| LDLc (mg/dl) | | | | | | | | | | | | | | | |
| Crude | 1.00 | 1.14 (0.92, 1.40) | 0.91 (0.74, 1.20) | 1.05 (0.85, 1.29) | .84 | 1.00 | 0.96 (0.79, 1.17) | 0.89 (0.73, 1.09) | 0.89 (0.72, 1.11) | .22 | 1.00 | 1.31 (1.06, 161) | 1.15 (0.93, 1.42) | 1.39 (1.13, 1.71) | .009 |
| Adjusted | 1.00 | 1.07 (0.87, 1.32) | 0.85 (0.69, 1.05) | 0.97 (0.79, 1.20) | .37 | 1.00 | 0.93 (0.76, 1.14) | 0.88 (0.72, 1.07) | 0.88 (0.71, 1.10) | .19 | 1.00 | 1.28 (1.04, 1.59) | 1.11 (0.90, 1.38) | 1.40 (1.13, 1.72) | .009 |
| HDLc (mg/dl) | | | | | | | | | | | | | | | |
| Crude | 1.00 | 0.84 (0.68, 1.04) | 0.87 (0.72, 1.06) | 0.90 (0.73, 1.10) | .32 | 1.00 | 0.95 (0.78, 1.16) | 1.06 (0.87, 1.29) | 1.09 (0.88, 1.34) | .29 | 1.00 | 0.86 (0.70, 1.06) | 0.89 (0.73, 1.10) | 0.89 (0.73, 1.09) | .32 |
| Adjusted | 1.00 | 1.01 (0.81, 1.25) | 1.07 (0.87, 1.32) | 1.13 (0.91, 1.40) | .22 | 1.00 | 1.05 (0.85, 1.29) | 1.14 (0.93, 1.40) | 1.15 (0.92, 1.43) | .13 | 1.00 | 0.90 (0.73, 1.12) | 0.99 (0.80, 1.22) | 0.87 (0.70, 1.08) | .34 |
| TG/HDL ratio | | | | | | | | | | | | | | | |
| Crude | 1.00 | 0.91 (0.70, 1.17) | 0.71 (0.71, 1.16) | 0.96 (0.74, 1.24) | .72 | 1.00 | 0.78 (0.60, 1.01) | 1.01 (0.79, 1.29) | 1.14 (0.88, 1.47) | .18 | 1.00 | 0.99 (0.77, 1.28) | 1.07 (0.83, 1.38) | 0.94 (0.73, 1.22) | .85 |
| Adjusted | 1.00 | 1.01 (0.78, 1.31) | 1.01 (0.78, 1.30) | 1.08 (0.83, 1.40) | .57 | 1.00 | 0.82 (0.64, 1.07) | 1.06 (0.83, 1.35) | 1.18 (0.91, 1.54) | .13 | 1.00 | 1.00 (0.77, 1.29) | 1.13 (0.88, 1.46) | 0.93 (0.72, 1.21) | .90 |
| Cholesterol/HDL ratio | | | | | | | | | | | | | | | |
| Crude | 1.00 | 0.98 (0.77, 1.25) | 0.89 (0.71, 1.12) | 0.80 (0.63, 1.01) | .05 | 1.00 | 0.75 (0.60, 0.94) | 0.89 (0.71, 1.13) | 0.78 (0.61, 1.00) | .14 | 1.00 | 1.14 (0.91, 1.44) | 1.06 (0.84) | 1.29 (1.02, 1.63) | .06 |
| Adjusted | 1.00 | 1.11 (0.87, 1.43) | 1.01 (0.80, 1.28) | 0.90 (0.71, 1.14) | .33 | 1.00 | 0.79 (0.63, 0.99) | 0.90 (0.71, 1.14) | 0.74 (0.57, 0.95) | .05 | 1.00 | 1.18 (0.94, 1.49) | 1.12 (0.89, 1.41) | 1.26 (0.99, 1.59) | .08 |
## DISCUSSION
In the present study, participants in the highest quartile of PDI had lower FBS levels. Furthermore, adherence to hPDI was associated with lower total cholesterol levels and cholesterol‐to‐HDL ratio. However, adherence to uPDI was associated with elevated levels of FBS, total cholesterol, and LDL.
The benefits of plant‐based diets for cardiovascular health have been shown in numerous studies. A meta‐analysis by Huang et al. reported that vegetarians have lower mortality from coronary heart disease (CHD) (Huang et al., 2012). Another meta‐analysis of five cohort studies demonstrated that vegetarians had a $24\%$ lower risk of CHD mortality compared to nonvegetarians (Key et al., 1999). The above‐mentioned studies have defined vegetarian diets as consuming no or very limited amounts of meat and its products, which mimics the PDI presented in the current study.
Recently, it has been noticed that not every plant food has the same beneficial features. Accordingly, healthy (i.e., mostly whole grains, fruits, vegetables, nuts, legumes, vegetable oils, and tea/coffee) and unhealthy (i.e., sugar‐sweetened beverages, fruit juices, refined grains, potatoes, and desserts) plant food indices have been designed. In the present study, we found that PDI and hPDI were related to lower cholesterol and FBS levels, thereby tipping the balance in favor of heart health. Nevertheless, the uPDI was related to elevated cholesterol, LDL, and FBS levels. High level of cholesterol in the circulation, especially those carried via LDL, is a major cause of heart disease. It causes the accumulation of fatty deposits within the arteries, which reduces the blood flow to the heart and other critical organs, and increases the risk of stroke or heart attack (Clarke et al., 1997). LDL is also prone to oxidation which could further complicate the situation (Holvoet, 2004). On the other hand, a constant high blood sugar level, which is usually ensued from poor food choices, could also produce the same outcomes. Studies showed that high blood sugar often goes hand in hand with elevated blood pressure and cholesterol levels (Cheung & Li, 2012).
Generally, our findings were in agreement with previous studies. Bhupathiraju et al. in a cross‐sectional study showed that adherence to the PDI and hPDI was inversely associated with LDL levels. Also, adherence to the PDI, but not hPDI, was inversely associated with FBS levels (Bhupathiraju et al., 2022). Another cross‐sectional study showed that plant‐based diets were associated with more optimal blood lipid concentrations (Martin et al., 2022). Furthermore, Shin et al. showed that the highest adherence to uPDI had $22\%$ greater odds of dyslipidemia and $48\%$ higher odds of hypertriglyceridemia in Korean adults (Shin & Kim, 2022). A meta‐analysis of observational studies and clinical trials revealed that adherence to vegetarian diets was associated with lower serum LDL, HDL, and total cholesterol levels (Yokoyama et al., 2017).
In addition, a cohort study on South Korean adults revealed that high adherence to the hPDI was inversely associated with the risk of dyslipidemia characterized by high levels of TG, LDL, and cholesterol. Nonetheless, adherence to the uPDI resulted in a remarkable significant increased risk of lipid disorders (Lee et al., 2021). Another cohort study in South Korea showed that the uPDI was associated with a higher risk of metabolic syndrome; however, participants with the highest adherence to the PDI had a lower FBS level (Kim et al., 2020). Furthermore, a case–control study by Zamani et al. following an overall plant‐based diet was associated with a lower risk of gestational diabetes (Zamani et al., 2019).
Several mechanisms have been introduced to justify the above‐mentioned findings, most of which center around the beneficial components of plant‐based diets. Similar to the case in the present study, greater adherence to the PDI and hPDI often marks a dietary plan high in fiber, antioxidants, unsaturated fats, and some micronutrients. Dietary fiber could decrease glucose absorption and impose a beneficial effect on glucose metabolism. It also enhances cholesterol removal by binding cholesterol and bile acids (Brown et al., 1999). Other nutrients, such as vitamin C and magnesium, could also increase insulin sensitivity which resulted in better glycemic control. Moreover, high amounts of polyphenols in healthy plant foods improve the lipid profile by inhibiting the oxidation of LDL (Quiñones et al., 2012). Additionally, decreased intake of animal‐based foods, associated with higher adherence to all three plant‐based indices in this study, results in a consequent low intake of saturated fat and heme iron, which could also justify the benefits of the PDI and hPDI on FBS and lipid profile (de Oliveira Otto et al., 2012).
On the other hand, adherence to the uPDI usually results in lower consumption of fibers and antioxidants, including vitamins C and E and beta carotene, which were particularly lower in the participants with the highest adherence to the uPDI. Low consumption of antioxidants could lead to endothelial dysfunction and oxidative stress, which ultimately pave the way for the development of CVDs (Liu et al., 2005). Moreover, the regular consumption of added sugars, another common characteristic of the uPDI, can lead to poor glycemic control and lipid metabolism (Fried & Rao, 2003).
However, the present study failed to find a significant correlation between none of the plant‐based diet indices with blood pressure, TG, and HDL, among which the lack of association with blood pressure could be the most controversial factor in previous studies. Clinical trial studies abound as to the efficacy of plant‐based diets in lowering blood pressure (Crimarco et al., 2020; Jenkins et al., 2008; Rouse et al., 1986).
By recruiting a large number of participants and using validated questionnaires, the present study could further broaden the understanding of how plant‐based indices interact with single individual CVD risk factors. However, the current study suffers from some limitations. First, the recall bias in reporting dietary intake can affect the results. The cross‐sectional nature of this study was another limitation as it prevented from inferring causality. Moreover, there might have been some other potential confounders affecting the results which could not be measured or adjusted, such as lack of oxidized LDL measurement and the information on medicine intake.
## CONCLUSION
The present study demonstrated that adherence to the PDI and hPDI was associated with more optimal blood glucose, cholesterol levels, and cholesterol to HDL ratio. Nevertheless, the uPDI was related to increased FBS, cholesterol, and LDL levels. The findings of this study further support the knowledge regarding the benefits of dietary patterns mainly composed of healthy plant‐based foods, such as whole grains and vegetables, while discouraging the regular intake of unhealthy plant foods, including refined grains and sugar‐sweetened beverages.
## CONFLICT OF INTEREST
All authors declare that they have no conflict of interest.
## DATA AVAILABILITY STATEMENT
Data available on request from the authors.
## References
1. Abdulmuhsin M. K.. **Global regional and national life expectancy all cause mortality and cause specific mortality for 249 causes of death**. *The Lancet* (2016) **388** 1459-1544
2. Bauman A. E.. **Updating the evidence that physical activity is good for health: an epidemiological review 2000–2003**. *Journal of Science and Medicine in Sport* (2004) **7** 6-19. PMID: 15214597
3. Bhupathiraju S. N., Sawicki C. M., Goon S., Gujral U. P., Hu F. B., Kandula N. R., Kanaya A. M.. **A healthy plant–based diet is favorably associated with cardiometabolic risk factors among participants of South Asian ancestry**. *The American Journal of Clinical Nutrition* (2022) **116** 1078-1090. PMID: 35731596
4. Borazjani M., Nouri M., Venkatakrishnane K., Najafi M., Faghih S.. **Association of plant‐based diets with lipid profile and anthropometric indices: a cross‐sectional study**. *Nutrition & Food Science* (2022) **52** 830-842
5. Brown L., Rosner B., Willett W. W., Sacks F. M.. **Cholesterol‐lowering effects of dietary fiber: a meta‐analysis**. *The American Journal of Clinical Nutrition* (1999) **69** 30-42. PMID: 9925120
6. Cheung B. M., Li C.. **Diabetes and hypertension: is there a common metabolic pathway?**. *Current Atherosclerosis Reports* (2012) **14** 160-166. PMID: 22281657
7. Chiang J.‐K., Lin Y.‐L., Chen C.‐L., Ouyang C.‐M., Wu Y.‐T., Chi Y.‐C., Huang K. C., Yang W. S.. **Reduced risk for metabolic syndrome and insulin resistance associated with ovo‐lacto‐vegetarian behavior in female Buddhists: a case‐control study**. *PLoS One* (2013) **8**. PMID: 23951247
8. Clarke R., Frost C., Collins R., Appleby P., Peto R.. **Dietary lipids and blood cholesterol: quantitative meta‐analysis of metabolic ward studies**. *BMJ* (1997) **314** 112-117. PMID: 9006469
9. Crimarco A., Springfield S., Petlura C., Streaty T., Cunanan K., Lee J., Fielding‐Singh P., Carter M. M., Topf M. A., Wastyk H. C., Sonnenburg E. D., Sonnenburg J. L., Gardner C. D.. **A randomized crossover trial on the effect of plant‐based compared with animal‐based meat on trimethylamine‐N‐oxide and cardiovascular disease risk factors in generally healthy adults: Study With Appetizing Plantfood‐Meat Eating Alternative Trial (SWAP‐MEAT)**. *The American Journal of Clinical Nutrition* (2020) **112** 1188-1199. PMID: 32780794
10. de Oliveira Otto M. C., Alonso A., Lee D.‐H., Delclos G. L., Bertoni A. G., Jiang R., Lima J. A., Symanski E., Jacobs D. R., Nettleton J. A.. **Dietary intakes of zinc and heme iron from red meat, but not from other sources, are associated with greater risk of metabolic syndrome and cardiovascular disease**. *The Journal of Nutrition* (2012) **142** 526-533. PMID: 22259193
11. Farazian F., Emami Moghadam Z., Heshmati Nabavi F., Behnam V. H.. **Effect of self‐care education designed based on banduraâ s self‐efficacy model on patients with hypertension: A randomized clinical trial**. *Evidence Based Care* (2019) **9** 44-52
12. Fried S. K., Rao S. P.. **Sugars, hypertriglyceridemia, and cardiovascular disease**. *The American Journal of Clinical Nutrition* (2003) **78** 873 S-80 S
13. Friedewald W. T., Levy R. I., Fredrickson D. S.. **Estimation of the concentration of low‐density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge**. *Clinical Chemistry* (1972) **18** 499-502. PMID: 4337382
14. Hassanzadeh‐Rostami Z., Hemmatdar Z., Pishdad G. R., Faghih S.. **Moderate consumption of red meat, compared to soy or non‐soy legume, has no adverse effect on cardio‐metabolic factors in patients with type 2 diabetes**. *Experimental and Clinical Endocrinology and Diabetes* (2021) **129** 429-437. PMID: 31207663
15. Hematdar Z., Ghasemifard N., Phishdad G., Faghih S.. **Substitution of red meat with soybean but not non‐soy legumes improves inflammation in patients with type 2 diabetes; a randomized clinical trial**. *Journal of Diabetes and Metabolic Disorders* (2018) **17** 111-116. PMID: 30918843
16. Hemler E. C., Hu F. B.. **Plant‐based diets for cardiovascular disease prevention: all plant foods are not created equal**. *Current Atherosclerosis Reports* (2019) **21** 1-8. PMID: 30631963
17. Holvoet P.. **Oxidized LDL and coronary heart disease**. *Acta Cardiologica* (2004) **59** 479-484. PMID: 15529550
18. Huang P. L.. **A comprehensive definition for metabolic syndrome**. *Disease Models & Mechanisms* (2009) **2** 231-237. PMID: 19407331
19. Huang T., Yang B., Zheng J., Li G., Wahlqvist M. L., Li D.. **Cardiovascular disease mortality and cancer incidence in vegetarians: A meta‐analysis and systematic review**. *Annals of Nutrition & Metabolism* (2012) **60** 233-240. PMID: 22677895
20. Jalilpiran Y., Hezaveh E., Bahmanpour S., Faghih S.. **Healthy plant foods intake could protect against prostate cancer risk: a case‐control study**. *Asian Pacific Journal of Cancer Prevention: APJCP* (2017) **18** 1905-1912. PMID: 28749619
21. Jenkins D. J., Kendall C. W., Faulkner D. A., Kemp T., Marchie A., Nguyen T. H.. **Long‐term effects of a plant‐based dietary portfolio of cholesterol‐lowering foods on blood pressure**. *European Journal of Clinical Nutrition* (2008) **62** 781-788. PMID: 17457340
22. Kamalipour M., Akhondzadeh S.. **Cardiovascular effects of saffron: An evidence‐based review**. *The Journal of Tehran Heart Center* (2011) **6** 59-61. PMID: 23074606
23. Kazemi Karyani A., Karmi Matin B., Soltani S., Rezaei S., Soofi M., Salimi Y., Moradinazar M., Hajizadeh M., Pasdar Y., Hamzeh B., Barzegar L., Haghdoost A. A., Malekzadeh R., Poustchi H., Mohammadi Z., Faramarzi E., Safarpour A. R., Pourfarzi F., Moosazadeh M., Najafi F.. **Socioeconomic gradient in physical activity: Findings from the PERSIAN cohort study**. *BMC Public Health* (2019) **19** 1-11. PMID: 30606151
24. Key T. J., Fraser G. E., Thorogood M., Appleby P. N., Beral V., Reeves G., Burr M. L., Chang‐Claude J., Frentzel‐Beyme R., Kuzma J. W., Mann J., McPherson K.. **Mortality in vegetarians and nonvegetarians: Detailed findings from a collaborative analysis of 5 prospective studies**. *The American Journal of Clinical Nutrition* (1999) **70** 516-524
25. Khunti K., Kosiborod M., Ray K. K.. **Legacy benefits of blood glucose, blood pressure and lipid control in individuals with diabetes and cardiovascular disease: time to overcome multifactorial therapeutic inertia?**. *Diabetes, Obesity and Metabolism* (2018) **20** 1337-1341
26. Kim H., Lee K., Rebholz C. M., Kim J.. **Plant‐based diets and incident metabolic syndrome: Results from a South Korean prospective cohort study**. *PLoS Medicine* (2020) **17**. PMID: 33206633
27. Kohansal A., Zangene A., Turki Jalil A., Hooshang H., Leilami K., Gerami S., Najafi M., Nouri M., Faghih S.. **Association between plant and animal proteins intake with lipid profile and anthropometric indices: A cross‐sectional study**. *Nutrition and Health* (2022) 2601060221104311. PMID: 35656771
28. Lee K., Kim H., Rebholz C. M., Kim J.. **Association between different types of plant‐based diets and risk of dyslipidemia: A prospective cohort study**. *Nutrients* (2021) **13** 220. PMID: 33466664
29. Liu S., Song Y., Ford E. S., Manson J. E., Buring J. E., Ridker P. M.. **Dietary calcium, vitamin D, and the prevalence of metabolic syndrome in middle‐aged and older US women**. *Diabetes Care* (2005) **28** 2926-2932. PMID: 16306556
30. Martin Z. T., Olvera G., Villegas C. A., Campbell J. C., Akins J. D., Brown K. K., Brothers R. M.. **The impact of a plant‐based diet on indices of cardiovascular health in African Americans: A cross‐sectional study**. *Applied Physiology, Nutrition, and Metabolism* (2022) **47** 903-914
31. Nikbakht H.‐A., Rezaianzadeh A., Seif M., Ghaem H.. **Prevalence of metabolic syndrome and its components among a population‐based study in south of Iran, PERSIAN Kharameh cohort study**. *Clinical epidemiology and global health.* (2020) **8** 678-683
32. Picasso M. C., Lo‐Tayraco J. A., Ramos‐Villanueva J. M., Pasupuleti V., Hernandez A. V.. **Effect of vegetarian diets on the presentation of metabolic syndrome or its components: A systematic review and meta‐analysis**. *Clinical Nutrition* (2019) **38** 1117-1132. PMID: 29907356
33. Poustchi H., Eghtesad S., Kamangar F., Etemadi A., Keshtkar A.‐A., Hekmatdoost A., Mohammadi Z., Mahmoudi Z., Shayanrad A., Roozafzai F., Sheikh M., Jalaeikhoo A., Somi M. H., Mansour‐Ghanaei F., Najafi F., Bahramali E., Mehrparvar A., Ansari‐Moghaddam A., Enayati A. A., Malekzadeh R.. **Prospective epidemiological research studies in Iran (the PERSIAN Cohort Study): rationale, objectives, and design**. *American Journal of Epidemiology* (2018) **187** 647-655. PMID: 29145581
34. Quiñones M., Miguel M., Aleixandre A.. **The polyphenols, naturally occurring compounds with beneficial effects on cardiovascular disease**. *Nutrición Hospitalaria* (2012) **27** 76-89. PMID: 22566306
35. Randhawa H., Ghaedi Y., Khan S., Al‐Sharbatti S.. **The prevalence of overweight and obesity among health care providers in the emirate of Ajman, UAE**. *Journal of Complementary Medicine Research* (2020) **11** 40-50
36. Rezaianzadeh A., Jafari F., Sadeghi S. E., Rahimikazerooni S.. **The prevalence and predictors of pre‐hypertension and hypertension in Kherameh cohort study: A population based study on 10,663 persons in south of Iran**. *Journal of Human Hypertension* (2021) **35** 257-264. PMID: 32203076
37. Rouse I. L., Beilin L. J., Mahoney D. P., Margetts B. M., Armstrong B. K., Record S. J., Vandongen R., Barden A.. **Nutrient intake, blood pressure, serum and urinary prostaglandins and serum thromboxane B2 in a controlled trial with a lacto‐ovo‐vegetarian diet**. *Journal of Hypertension* (1986) **4** 241-250. PMID: 3011891
38. Salcedo‐Cifuentes M., Belalcazar S., Acosta E. Y., Medina‐Murillo J. J.. **Conventional biomarkers for cardiovascular risks and their correlation with the Castelli Risk Index‐Indices and TG/HDL‐C**. *Archivos de Medicina (Manizales)* (2020) **20** 11-22
39. Satija A., Bhupathiraju S. N., Rimm E. B., Spiegelman D., Chiuve S. E., Borgi L., Willett W. C., Manson J. A. E., Sun Q., Hu F. B.. **Plant‐based dietary patterns and incidence of type 2 diabetes in US men and women: results from three prospective cohort studies**. *PLoS Medicine* (2016) **13**. PMID: 27299701
40. Shin N., Kim J.. **Association between different types of plant‐based diet and dyslipidaemia in Korean adults**. *The British Journal of Nutrition* (2022) **128** 542-548. PMID: 34503592
41. Shmakova N. N., Puzin S. N., Zarariy N. S., Abol A. V.. **The characteristics of impaired functions and life limitations of disabled people due to coronary heart disease**. *Journal of Population Therapeutics and Clinical Pharmacology* (2022) **29** 62-70
42. Vinagre J., Vinagre C., Pozzi F., Slywitch E., Maranhão R.. **Metabolism of triglyceride‐rich lipoproteins and transfer of lipids to high‐density lipoproteins (HDL) in vegan and omnivore subjects**. *Nutrition, Metabolism, and Cardiovascular Diseases* (2013) **23** 61-67
43. Wang F., Zheng J., Yang B., Jiang J., Fu Y., Li D.. **Effects of vegetarian diets on blood lipids: a systematic review and meta‐analysis of randomized controlled trials**. *Journal of the American Heart Association* (2015) **4**. PMID: 26508743
44. Wu F., Guo Y., Chatterji S., Zheng Y., Naidoo N., Jiang Y., Biritwum R., Yawson A., Minicuci N., Salinas‐Rodriguez A., Manrique‐Espinoza B., Maximova T., Peltzer K., Phaswanamafuya N., Snodgrass J. J., Thiele E., Ng N., Kowal P.. **Common risk factors for chronic non‐communicable diseases among older adults in China, Ghana, Mexico, India, Russia and South Africa: the study on global AGEing and adult health (SAGE) wave 1**. *BMC Public Health* (2015) **15** 1-13. PMID: 25563658
45. Yang S.‐Y., Zhang H.‐J., Sun S.‐Y., Wang L.‐Y., Yan B., Liu C.‐Q., Zhang W., Li X. J.. **Relationship of carotid intima‐media thickness and duration of vegetarian diet in Chinese male vegetarians**. *Nutrition and Metabolism* (2011) **8** 1-7. PMID: 21241467
46. Yokoyama Y., Levin S. M., Barnard N. D.. **Association between plant‐based diets and plasma lipids: a systematic review and meta‐analysis**. *Nutrition Reviews* (2017) **75** 683-698. PMID: 28938794
47. Yokoyama Y., Nishimura K., Barnard N. D., Takegami M., Watanabe M., Sekikawa A., Okamura T., Miyamoto Y.. **Vegetarian diets and blood pressure: a meta‐analysis**. *JAMA Internal Medicine* (2014) **174** 577-587. PMID: 24566947
48. Zamani B., Milajerdi A., Tehrani H., Bellissimo N., Brett N. R., Azadbakht L.. **Association of a plant‐based dietary pattern in relation to gestational diabetes mellitus**. *Nutrition and Dietetics* (2019) **76** 589-596. PMID: 30680868
|
---
title: Costunolide alleviates hyperglycaemia‐induced diabetic cardiomyopathy via inhibiting
inflammatory responses and oxidative stress
authors:
- Bo Jin
- Yi Chen
- Jiong Wang
- Yue Chen
- Mengpei Zhang
- Jianxiong Huang
- Yi Wang
journal: Journal of Cellular and Molecular Medicine
year: 2023
pmcid: PMC10002915
doi: 10.1111/jcmm.17686
license: CC BY 4.0
---
# Costunolide alleviates hyperglycaemia‐induced diabetic cardiomyopathy via inhibiting inflammatory responses and oxidative stress
## Abstract
Hyperglycaemia‐induced myocardial injury promotes the induction of heart failure in diabetic patients. Impaired antioxidant capability and sustained chronic inflammation play a vital role in the progression of diabetic cardiomyopathy (DCM). Costunolide (Cos), a natural compound with anti‐inflammatory and antioxidant properties, has exhibited therapeutic effects in various inflammatory diseases. However, the role of Cos in diabetes‐induced myocardial injury remains poorly understood. In this study, we investigated the effect of Cos on DCM and explored the potential mechanisms. C57BL/6 mice were administered intraperitoneal streptozotocin for DCM induction. Cos‐mediated anti‐inflammatory and antioxidation activities were examined in heart tissues of diabetic mice and high glucose (HG)‐stimulated cardiomyocytes. Cos markedly inhibited HG‐induced fibrotic responses in diabetic mice and H9c2 cells, respectively. The cardioprotective effects of Cos could be correlated to the reduced expression of inflammatory cytokines and decreased oxidative stress. Further investigations demonstrated Cos reversed diabetes‐induced nuclear factor‐κB (NF‐κB) activation and alleviated impaired antioxidant defence system, principally via activation of nuclear factor‐erythroid 2 p45‐related factor‐2 (Nrf‐2). Cos alleviated cardiac damage and improved cardiac function in diabetic mice by inhibiting NF‐κB‐mediated inflammatory responses and activating the Nrf‐2‐mediated antioxidant effects. Therefore, Cos could be a potential candidate for the treatment of DCM.
## INTRODUCTION
Diabetes mellitus (DM) is a common metabolic disorder worldwide and has gradually become an epidemic in recent years. Elevated blood glucose levels typically cause a series of tissue injury‐related events during diabetes, resulting in diabetic complications such as diabetic cardiomyopathy (DCM), diabetic nephropathy, diabetic foot disease and retinopathy. Among these diabetic complications, DCM is considered a risk factor of heart failure, known to be associated with disability and mortality in patients with diabetes. 1 DCM persistently presents as diastolic dysfunction during the early stage, progressing to systolic dysfunction in the advanced phase. Hyperglycaemia‐induced cardiac fibrosis and hypertrophy are the most frequently proposed pathological mechanisms underlying structural and functional alterations in DCM. 2 Subcellular and low‐grade inflammation caused by metabolic disturbances are key pathogenic features of diabetes. 3 Studies have suggested that hyperglycaemia‐induced cardiac adverse remodelling is highly associated with hyperglycaemia‐induced inflammatory responses and oxidative stress. 4 On one hand, hyperglycaemia activates nuclear factor‐κB (NF‐κB) and mitogen‐activated protein kinases (MAPKs), which are crucial regulators of pro‐inflammatory signalling pathways, resulting in the release of chemokines and inflammatory cytokines, such as monocyte chemoattractant protein (MCP)‐1, tumour necrosis factor (TNF)‐α, interleukin (IL)‐6 and IL‐1β, in the hearts of diabetic mice. Subsequently, the released cytokines recruit macrophages, which reside in the cardiac tissues. Infiltrated macrophages express additional inflammatory cytokines under hyperglycaemic conditions, which further aggravate myocardial inflammation and cardiac injury, contributing to the deterioration of DCM. On the contrary, numerous studies have shown that reactive oxygen species (ROS) levels are elevated in the diabetic hearts, which ultimately cause cardiac oxidative injury. 5 Nuclear factor‐erythroid 2 p45‐related factor‐2 (Nrf‐2), a key transcription factor regulating antioxidant stress, plays an important role in the antioxidant response induction. Nrf‐2 is reportedly downregulated in the diabetic heart. Excessive ROS production and weakened antioxidant capacity cause mitochondrial damage and lipid peroxidation, contributing to cardiomyocyte damage in DCM. 6 Thus, hyperglycaemia‐induced inflammation and oxidative stress are primary factors inducing myocardial cell death, thereby resulting in a range of downstream events associated with cardiac injury and remodelling. Thus, blocking inflammatory and oxidative responses may afford a potential strategy to hinder the development of DCM.
Costunolide (Cos), a natural sesquiterpene lactone, has been shown to possess diverse biological activities. 7 Cos alleviates dextran sulfate sodium‐induced acute ulcerative colitis by inhibiting the NF‐κB signalling pathway. 8 In addition, Cos attenuates heat‐killed Staphylococcus aureus‐induced acute lung injury via inhibition of macrophage activation and displays anti‐fibrotic activity in bleomycin‐induced pulmonary fibrosis by regulating NF‐κB and Nrf‐2/NADPH oxidase 4 (NOX4) signalling pathways. 9, 10 These findings suggest that Cos could be a promising candidate for the treatment of DCM.
Herein, using a streptozotocin (STZ)‐induced diabetic mouse model and HG‐challenged H9c2 cardiomyocytes, we examined the potential cardioprotective effect of Cos against DCM and elucidated the underlying mechanisms.
## Reagents
Costunolide was obtained from Chengdu Alfa Biotechnology Co.,Ltd (Chengdu, China) and dissolved in DMSO for experiments in vitro and in $0.5\%$ CMC‐Na for studies in vivo. The chemical structure of *Costunolide is* shown in (Figure 1A). Antibodies against inhibitor of IκB‐α (1:1000, cat. no. 4814), NF‐κB P65 subunit (1:1000, cat. no. 8242), p‐P65 (1:1000, cat. no. 3033), p‐P38 (1:1000, cat. no. 9211), P38 (1:1000, cat. no. 9212) and GAPDH (1:1000, cat. no. 97166) were purchased from Cell Signaling Technology (Danvers, MA, USA). Anti‐TGF‐β (1:1000, cat. no. ab92486) and anti‐β‐MyHC (1:1000, cat. no. ab50967) antibodies were acquired from Abcam (Cambridge, MA, USA). Antibodies against MCP‐1 (1:200, cat. no. sc‐32771) and F$\frac{4}{80}$ (1:200, cat. no. sc‐377009) were obtained from Santa Cruz (CA, USA), and HO‐1 (1:1000, cat. no. 66743‐1‐Ig), COL‐1 (1:1000, cat. no. 66761‐1‐Ig) and Nrf‐2 (1:1000, cat. no. 16396‐1‐AP) were also purchased from Proteintech (Wuhan, China). The assay kits for CK‐MB (cat. no. E006‐1‐1), LDH (cat. no. A020‐2‐2) were obtained from Nanjing Jiancheng Bioengineering Institute (Nanjing, China). Biochemical detection kits of superoxide dismutase (SOD, cat. no. S0109) and malondialdehyde (MDA, cat. no. S01315) were acquire form Beyotime (Shanghai, China). ML385 was purchased from TargetMol Chemicals Inc. (Boston, MA, USA). Bovine serum albumin (BSA, cat. no. A1933) was purchased from Sigma (St. Louis, MO, USA). High glucose solution (1 M) was prepared. Dissolving 198.17 μg D‐ (+)‐glucose monohydrate (cat. no. 5250278, aladdin, USA) in 1 mL ddH2O, and then filtered with a 0.22 μm sieve.
**FIGURE 1:** *Cos improves cardiac function and alleviates hyperglycaemia‐induced cardiac pathological injuries. (A) Chemical structure of Cos. (B,C) Body weight (B) and serum glucose levels (C) of control and diabetic mice. (D) Heart weight/body weight ratio was established after sacrificing mice. (E,F) Serum creatine kinase (CK)‐MB (CK‐MB) and lactate dehydrogenase (LDH) levels were determined using biochemical detection kits. (G) Echocardiography showing the ejection fraction (%) of mice. (H) Representative images of M‐mode echocardiogram. Data are expressed as mean ± SEM, n = 6. ##p < 0.01, ###p < 0.001 between CON and STZ groups; *p < 0.05, **p < 0.01, ***p < 0.001 between STZ and STZ + Cos groups.*
## Cell culture
The embryonic rat heart‐derived cell line, H9c2, was purchased from the Shanghai Institute of Biochemistry and Cell Biology (CSTR:19,375.09.3101RATGNR5, Shanghai, China). H9c2 cells were grown in Dulbecco's Modified Eagle Medium (cat. no. 11995040, Gibco; Eggenstein, Germany), containing 1 g/L glucose and supplemented with $12\%$ foetal bovine serum (FBS; cat. no. 14160063, Gibco; Eggenstein, Germany), 100 U/mL penicillin and 100 mg/mL streptomycin (Invitrogen; CA, USA). The cells were incubated at 37°C in a $5\%$ CO2 humidified incubator.
## Animals
Male C57BL/6 mice (18–20 g) were obtained from the Animal Center of Wenzhou Medical University. All animal care and experimental procedures were approved by the Wenzhou Medical University Animal Policy and Welfare Committee (Approval Document No. wydw2019‐0145). Mice were maintained at 22°C and $60\%$ relative humidity and fed a standard rodent diet and purified water.
Mice were subsequently divided into three groups ($$n = 6$$ per group): non‐diabetic control group (CON), STZ‐induced diabetic group (STZ) and Cos‐treated STZ‐induced diabetic group (STZ + Cos). After a one‐week adaptation, diabetes was induced in the STZ and STZ + Cos groups by administering a single intraperitoneal injection of STZ (cat. no. B2001, Boaigang Biological Technology Co., Ltd. Beijing, China) at the dose of 170 mg/kg dissolved in $0.5\%$ citrate buffer (pH 4.5). The control mice received an intraperitoneal injection of equivalent citrate buffer. Tail blood glucose levels were measured once every 2 weeks after a 12 h fasting period. Mice with fasting‐blood glucose ˃12.2 mmol/L for 8 weeks were considered hyperglycaemic, indicating the successful establishment of the diabetic model. The mice from STZ + Cos group were administered 20 mg/kg Cos orally, once every 2 days. Eight weeks after Cos treatment, mice were sacrificed, and heart tissues and serum were harvested.
## Echocardiographic measurements
One week prior to sacrifice, echocardiography was performed in M‐mode using Vevo 770™ High‐Resolution Imaging System (Visualsonics, Toronto, Canada). Left ventricular ejection fraction and left ventricular fractional shortening were expressed as percentages (Vevo 2100, Visual Sonics imaging system, Toronto, ON, Canada). LV end‐diastolic volume (LVVd), ejection fraction (EF) and LV Mass (AW) were automatically calculated by the ultrasound machine.
## MTT assay
MTT powder (cat. no. M8180, Solarbio; Beijing, China) was dissolved in phosphate‐buffered saline. H9c2 cells (1 × 104 per well) were seeded in 96‐well plates. After cell adherence, Cos was added to the wells at various doses (2.5–20 μM), followed by incubation for 48 h. Next, MTT was added to each well (1 mg/mL) and incubated for 4 h. The formazan crystals were then dissolved in dimethyl sulfoxide (DMSO; 150 μL/well), and absorbance was examined at 490 nm using SpectraMax M5 microplate reader (Molecular Devices, CA, USA).
## Histological and immunohistochemical analyses
Harvested heart tissues were fixed in $4\%$ buffered formalin, subsequently embedded and sliced into 5 μm thick sections for histologic analyses. Haematoxylin–eosin (cat. no. G1120), Sirius red (cat. no. S8060) and Masson's trichrome (cat. no. G1340) staining were performed according to the respective kit instructions (Solarbio).
For immunohistochemistry, antigen retrieval was performed after deparaffinization and hydration of sections. Next, sections were blocked with $5\%$ bovine serum albumin for 1 h, followed by overnight incubation with F$\frac{4}{80}$ antibody (1:50) at 4°C. After overnight incubation, sections were incubated with horseradish peroxidase (HRP)‐conjugated secondary antibodies (cat. no. A0216, Beyotime). Finally, tissues were counterstained with haematoxylin and covered with neutral resin. Images were captured using a Nikon microscope (Nikon, Tokyo, Japan), and the mean percentage of the staining‐positive area was measured and calculated using the ImageJ software (National Institutes of Health, Bethesda, MD).
## Immunofluorescence
H9c2 cells were pretreated with 2.5 or 5 μM Cos for 1 h, followed by exposure to 33 mM glucose for 3 h. Cells were fixed with $4\%$ paraformaldehyde for 15 min and permeabilized with $0.5\%$ Triton X‐100 for 10 min. Subsequently, cells were incubated with an anti‐p65 antibody (1:1000) at 4°C overnight. A FITC‐conjugated secondary antibody was used for detection. Finally, nuclei were counterstained with an anti‐fluorescent quencher containing DAPI. Images were observed and captured using an inverted fluorescence microscope (TE2000U, Nikon, Tokyo, Japan).
## Rhodamine‐phalloidin staining
H9c2 cells were pretreated with 2.5 or 5 μM Cos for 1 h, followed by exposure to 33 mM glucose for 36 h. Cells were fixed with $4\%$ paraformaldehyde and permeabilized with $0.5\%$ Triton X‐100 for 10 min. Subsequently, cells were stained with rhodamine‐phalloidin (cat. no. CA1610, Solarbio) for 30 min. Finally, nuclei were counterstained with an anti‐fluorescent quencher containing DAPI. Images were observed and captured using a fluorescence microscope (Nikon).
## Determination of ROS generation
Dihydroethidium staining (DHE, Beyotime) was utilized to detect ROS generation as previously described. 11 Heart tissue sections were incubated with 2 μM DHE at 37°C for 1 h in a dark humidified chamber. For the in vitro experiment, H9c2 cells were incubated with 2 μM DHE at 37°C for 30 min in a dark humidified chamber, Fluorescent images were obtained using fluorescence microscope (Nikon).
## Measurement of enzyme activities
Serum CK‐MB, LDH and myocyte enzyme activities of SOD and MDA were measured using corresponding detection kits according to the manufacturers' instructions.
## Reverse transcription‐quantitative PCR (RT‐qPCR)
Cultured cells or heart tissue samples were lysed with TRIzol (cat. no. 15596026, Thermo Fisher; CA, USA). Total RNA was extracted and separated using chloroform and isopropanol and purified with ethanol. Reverse transcription was performed using the PrimeScript™ RT reagent Kit (cat. no. DRR037A, Takara Bio Inc., Kusatsu, Japan). Quantitative PCR was performed using TB Green® Premix Ex Taq™ II (cat. no. RR820B, Takara Bio Inc.). Primers (Table S1) were obtained from Sangon Biotech (Shanghai, China). mRNA levels were detected and normalized using β‐actin as the loading control.
## Western blot analysis
Cells or heart tissue samples were lysed using RIPA buffer. Total proteins were separated using $10\%$ sodium dodecyl sulfate‐polyacrylamide gel electrophoresis (SDS‐PAGE) and then electro‐transferred to polyvinylidene fluoride (PVDF) membranes (cat. no. 1620177, Bio‐Rad Laboratory; Hercules, CA). Membranes were blocked in $5\%$ non‐fat milk for 2 h at room temperature and cut into bands, followed by incubation with respective antibodies. After overnight incubation, protein bands were incubated with HRP‐conjugated secondary antibodies and scanned using an image analyser (Quantity One System; Bio‐Rad, Richmond, CA, USA).
## Statistical analysis
All data were presented as mean ± SEM. Prism 8.0 software (GraphPad, San Diego, CA, USA) was used for the statistical analysis. Student's t‐test was used to compare two groups of data. One‐way anova followed by Dunnett's post hoc test was used to compare more than two groups of data. A p value < 0.05 was considered as significant.
## Cos improves cardiac function and alleviates hyperglycaemia‐induced cardiac pathological injuries
To explore the therapeutic effect of Cos on DCM‐mediated cardiac injury, we established a type 1 diabetic mouse model by administering STZ intraperitoneally. Compared with the diabetic group, Cos treatment at the dose of 20 mg/kg did not alter body weight and serum glucose levels in STZ + Cos group mice (Figure 1B,C). After persistent hyperglycaemia for 16 weeks, diabetic mice showed an increased ratio of heart weight to body weight when compared to those of non‐diabetic control mice (Figure 1D). However, Cos treatment significantly reduced this ratio in diabetic mice, indicating that Cos could reverse diabetes‐induced cardiac hypertrophy. Serum levels of lactate dehydrogenase (LDH) and creatine kinase (CK)‐MB are common indicators of heart damage. 12 We found that hyperglycaemia increased serum LDH and CK‐MB levels, which were reversed by Cos treatment (Figure 1 E,F). In addition, Cos reduced HG‐induced upregulation of brain natriuretic peptide (BNP) in cardiac tissue samples (Figure S1). M‐mode echocardiography was performed to assess cardiac function. The ejection fraction (EF) was significantly reduced in diabetic mice, indicating an impaired systolic function during diabetes, while Cos treatment rescued these pathological changes (Figure 1G,H, Table S2). Overall, these findings suggest that Cos improves cardiac function in diabetic mice and alleviates hyperglycaemia‐induced cardiac pathological injuries.
## Cos attenuates diabetes‐induced cardiac fibrosis and hypertrophy
Next, we evaluated structural alternations in the heart tissues of diabetic mice. H&E staining revealed that diabetic hearts exhibited structural abnormalities, including broken fibres, disordered myocardial structures and the presence of foci with necrotic myocytes. Treatment with Cos reversed diabetes‐induced structural alternations in the heart (Figure 2A). Sirius red‐ and Masson's trichrome‐stained sections were used to examine collagen deposition and fibrosis in cardiac tissues. Compared with the heart tissues of the control group, the hearts of diabetic mice showed collagen deposition. However, Cos treatment significantly decreased the upregulation of collagen fibres (Figure 2A–C). Consistent with the staining results, Western blot analysis also demonstrated the Cos‐mediated protective effects against cardiac fibrosis and hypertrophy. The heart tissue samples of Cos‐treated diabetic mice exhibited decreased protein levels of pro‐hypertrophic (β‐Myhc) and profibrotic (type 1 collagen; transforming growth factor β) markers when compared with those in the STZ‐induced diabetic mice (Figure 2D,E). Additionally, Cos treatment downregulated transcription levels of Col1a1, Col4a1, Tgfb1 and Myh7 (Figure 2F–I). These findings indicate that Cos administration attenuates cardiac fibrosis and hypertrophy in DCM mice.
**FIGURE 2:** *Cos attenuates diabetes‐induced cardiac fibrosis and hypertrophy. (A) H&E, Sirius red and Masson's trichrome staining were performed. (B,C) Quantification of positive collagen area in Sirius red and Masson's trichrome staining results. (D) Western blot analysis was used to detect protein levels of Col‐1, TGF‐β and β‐Myhc in heart tissues. GAPDH was used as a loading control. Relative density was quantified using ImageJ (E). (F–I) Relative mRNA levels of Col1a1, Col4a1, Tgfb1 and Myh7. Data were normalized to β‐actin and expressed as mean ± SEM, n = 6. #p < 0.05, ##p < 0.01, ###p < 0.001 between CON and STZ groups; *p < 0.05, **p < 0.01, ***p < 0.001 between STZ and STZ + Cos groups.*
## Cos mitigates hyperglycaemia‐mediated myocardial inflammation and oxidative stress
Macrophage infiltration and myocardial inflammation are well‐established events in the pathogenesis of DCM. 13, 14 We then performed F$\frac{4}{80}$ immunohistochemical staining to examine macrophage infiltration in the diabetic heart tissues. As shown in Figure 3A,B, Cos treatment significantly reduced macrophage infiltration in cardiac tissues of diabetic mice. To clarify the influence of Cos on hyperglycaemia‐induced myocardial inflammation, we examined the activation of the NF‐κB signalling pathway. As shown in Figure 3C,D, Cos treatment markedly suppressed hyperglycaemia‐induced IκB‐α degradation and inhibited the phosphorylation of TBK1 and P65, indicating the anti‐inflammatory effect of Cos in diabetic hearts. MCP‐1 is a key chemokine that plays a pivotal role in diabetes‐induced tissue inflammation. 15 Cos treatment markedly decreased the MCP‐1 protein level in the diabetic hearts when compared with those of the STZ group (Figure 3 E,F). Moreover, Cos treatment reduced transcription levels of inflammatory cytokines (Tnf, Il6 and Il1b) and inflammatory‐related genes (Inos and Ccl2) (Figure 3G). Taken together, Cos mitigates diabetes‐induced cardiac inflammation by inhibiting NF‐κB activation and macrophage infiltration.
**FIGURE 3:** *Cos mitigates hyperglycaemia‐induced myocardial inflammation and oxidative stress. (A) Immunohistochemistry for F4/80 was performed to indicate macrophage infiltration into the hearts of diabetic mice. (B) Quantification of F4/80 positive staining intensity. (C) Representative immunoblot for p‐P65, P65, p‐TBK1, TBK1 and IκB‐α in cardiac tissues of mice. Densitometric quantification was performed using ImageJ (D). (E) Western blotting was performed to determine the protein level of MCP‐1. Relative density is shown in (F). (G) Relative mRNA levels of Tnf, Il6, Il1b, Inos and Ccl2 in cardiac tissues. Data were normalized to β‐actin. (H) Protein levels of Nrf‐2 and HO‐1 were determined by Western blotting. (I) Relative levels of Nrf‐2 and HO‐1 mRNA were normalized to β‐actin. Data are expressed as mean ± SEM, n = 6. #p < 0.05, ##p < 0.01, ###p < 0.001 between CON and STZ groups; *p < 0.05, **p < 0.01, ***p < 0.001 between STZ and STZ + Cos groups.*
Finally, we examined the Cos‐mediated antioxidant activity and determined whether this antioxidant activity partly contributes to the cardioprotective effect in diabetic heart. Nrf‐2 and haem oxygenase (HO‐1) were examined as indicators of oxidative stress. Protein levels of Nrf‐2 and HO‐1 were reduced in the hearts of STZ‐induced diabetic mice, indicating the impaired antioxidation capability in diabetic hearts. However, the protein and mRNA levels of both Nrf‐2 and HO‐1 were rescued by Cos treatment (Figure 3H,I). In addition, increased ROS level has been well established to play a key role in the pathogenesis of DCM. 16, 17 We further determined ROS level in the cardiac tissues. We found that Cos treatment markedly reduced the production of ROS in the diabetic hearts (Figure S2). Taken together, the cardioprotective effects of Cos might be mediated by its anti‐inflammatory and antioxidant effects.
## Pretreatment with Cos inhibits HG‐induced fibrotic and hypertrophic responses in H9c2 cells
H9c2 myocardial cells were selected to evaluate the effect of Cos in vitro. First, optimal Cos dosages were determined using the MTT assay (Figure S3). The dosages of 2.5 and 5 μM of Cos were chosen as appropriate dose for the further in vitro experiments. As shown in Figure 4A–C, HG stimulation for 24 h significantly upregulated mRNA levels of fibrosis‐ (Col‐1 and TGF‐β) and hypertrophy‐related (β‐Myhc) factors, whereas pretreatment with Cos significantly reversed these changes in the HG‐treated H9c2 cells. Likewise, Cos pretreatment reduced protein levels of Col‐1, TGF‐β and β‐Myhc (Figure 4D,E). Furthermore, rhodamine‐phalloidin staining revealed that Cos reduced HG‐induced cellular hypertrophy in H9c2 cells (Figure 4F). These findings indicate that Cos exhibits cardiac protective effect by attenuating HG‐induced fibrotic and hypertrophy responses in myocardial cells.
**FIGURE 4:** *Cos pretreatment inhibits HG‐induced fibrotic and hypertrophic responses in H9c2 cells. (A–C) mRNA levels of Col1a1, Tgfb1 and Myh7 in HG‐challenged H9c2 cells. Cells were pretreated with Cos for 1 h, followed by exposure of HG for 24 h. (D) H9c2 cells were exposed to HG for 36 h after pretreatment with Cos for 1 h. The protein levels of Col‐1, TGF‐β and β‐Myhc were detected in the cell lysis. The relative density of protein levels was quantified (E). (F) Rhodamine‐phalloidin staining was performed to assess the changes of cell size. H9c2 cells were pretreated with Cos for 1 h, following HG‐treatment for 36 h. Data are expressed as mean ± SEM, n = 3. #p < 0.05, ##p < 0.01, ###p < 0.001 between CON and STZ groups; *p < 0.05, **p < 0.01, ***p < 0.001 between STZ and STZ + Cos groups.*
## Cos reduced HG‐induced inflammation and oxidative stress in H9c2 cells
We next investigated the effect of Cos on HG‐induced inflammatory and antioxidative responses in H9c2 cells. As shown in Figure 5A–C, Cos pretreatment reduced HG‐induced upregulation of pro‐inflammatory cytokines such as Tnf, Il6 and Il1b. In order to examine NF‐κB activation in HG‐treated H9c2 cells, we performed p65 immunofluorescence staining and assessed NF‐κB activation using Western blotting. As shown in Figure 5D and Figure S4, pretreatment of Cos decreased nuclear p65 levels in HG‐challenged H9c2 cells. Furthermore, we observed that Cos inhibited the phosphorylation of p65 and p38 and reduced IκB‐α degradation (Figure 5E,F). These results suggest that Cos suppresses NF‐κB‐ and p38‐MAPK‐mediated inflammatory cytokine production.
**FIGURE 5:** *Cos pretreatment reduces HG‐induced H9c2 cell inflammation. (A–C) H9c2 cells pretreated with Cos for 1 h were lysed after a 12‐h exposure to HG. The mRNA levels of Tnf, Il6 and Il1b were measured via RT‐qPCR. (D) Immunofluorescence staining of p65 in H9c2 cells. Cells were pretreated with Cos for 1 h and then exposed to HG for 3 h [scale bar = 20 μm]. (E) Protein levels of p‐P65, P65, IκB‐α, p‐P38 and P38 were determined. Cells were pretreated with Cos for 1 h, followed by exposure of HG for 1 h. Densitometric quantification was performed using ImageJ (F). Data are expressed as mean ± SEM, n = 3. #p < 0.05, ##p < 0.01, ###p < 0.001 between CON and STZ groups; *p < 0.05, **p < 0.01, ***p < 0.001 between STZ and STZ + Cos groups.*
Finally, we confirmed the effect of Cos on HG‐induced oxidative stress in H9c2 cells. As shown in Figure 6A and Figure S5, HG stimulation reduced expression of Nrf‐2 and HO‐1, indicating that HG condition inhibits the Nrf‐2/HO‐1 antioxidation signalling pathway. However, pretreatment with Cos effectively activated the Nrf‐2/HO‐1 signalling pathway in HG‐challenged H9c2 cells. Cos also enhanced the enzyme activity of SOD and decreased the levels of MDA, a natural bi‐product of lipid peroxidation in HG‐treated H9c2 cells (Figure 6B,C). In addition, DHE staining revealed that increased ROS levels caused by HG were significantly reduced by pretreatment of Cos, further confirming the antioxidation of Cos (Figure 6D). Overall, our findings indicate that Cos suppresses HG‐induced inflammation and oxidative stress in H9c2 cells, thereby reducing fibrotic and hypertrophic responses in H9c2 cells.
**FIGURE 6:** *Cos pretreatment reduces HG‐induced oxidative stress in H9c2 cells. (A) Western blot analysis was used to determine the protein levels of Nrf‐2 and HO‐1. GAPDH was used as the loading control. (B–C) Enzymatic activity of SOD and levels of MDA in lysates prepared from H9c2 cells. Cells were exposed to HG for 4 h after pretreatment of Cos for 1 h. (D) Representative images of ROS staining in H9c2 cells. Cells with red fluorescence indicated intracellular ROS. (E–J) H9c2 cells were pretreated with Cos (5 μM) with or without pretreatment of ML385 (4 μM), followed by exposure of HG for 24 h. mRNA levels of Col1a1, Tgfb, Myh7, Tnf, Il6 and Il1b were determined. Data were normalized to β‐actin. Data are expressed as mean ± SEM, n = 3 per group. *p < 0.05, **p < 0.01, and ***p < 0.001, ns, not significant.*
In order to figure out the underlying links between the anti‐inflammatory and antioxidant effects of Cos, ML385, a Nrf‐2 inhibitor was used in our study. As show in Figure 6E–G, Nrf‐2 inhibition aggravated HG‐induced fibrotic responses and impaired the protective effect of Cos. Meanwhile, the production of inflammatory cytokines was increased by Nrf‐2 inhibition, and the anti‐inflammatory effect of Cos was also weakened by ML385 (Figure 6H–J). However, compared to HG group, the cardioprotective effect and anti‐inflammatory effect of Cos were still exist even Nrf‐2 was inhibited. These data indicate that Cos, at least partly, exhibited its cardiac protection effects via activating Nrf‐2 pathway.
## DISCUSSION
In recent years, natural products have gained considerable momentum owing to their high efficiency, low toxicity and high accessibility. These properties facilitate the development of natural products into therapeutic drugs for various diseases. Several natural compounds are well‐known to show therapeutic effects against DCM. For example, crocin, mangiferin and kaempferol ameliorate diabetes‐induced cardiac injuries and improve cardiac function in different DCM models. 18, 19, 20 These effects are mainly attributed to their anti‐inflammatory and antioxidant properties. Cos is a natural sesquiterpene lactone exhibiting potent anti‐inflammatory activity, demonstrating potential therapeutic utility in diverse diseases such as acute liver injury, osteoarthritis and ulcerative colitis. 8, 21, 22 In addition, Cos has been shown to protect against doxorubicin‐induced toxicity in rats by modulating oxidative stress, inflammation and apoptosis. 23 Thus, the anti‐inflammatory and antioxidative effects of Cos make it a potential candidate for the treatment of DCM. In this study, we show that Cos improved cardiac function and mitigated adverse cardiac remodelling in the STZ‐induced diabetic mouse model. The potential mechanism of Cos in alleviating hyperglycaemia‐induced diabetic cardiomyopathy is shown in Figure 7. Treatment with Cos markedly reversed hyperglycaemia‐associated cardiac fibrosis and hypertrophy. This Cos‐mediated cardioprotective effect was associated with decreased inflammation and oxidative stress via the suppression of NF‐κB‐ and p38‐MAPK‐mediated inflammatory responses, along with the activation of Nrf‐2/HO‐1‐mediated antioxidative responses. Our data indicate that Cos, which has been shown to effectively disrupt both inflammatory response and oxidative stress caused by diabetes, could be a potential therapeutic candidate for the treatment of DCM.
**FIGURE 7:** *Schematic illustration indicates the potential mechanism of Cos in alleviating hyperglycaemia‐induced diabetic cardiomyopathy.*
The most prevalent therapeutic strategy for diabetes is treatment of hypoglycaemic drugs with diet and exercise management. 24 However, hyperglycaemia‐induced tissue injury is inevitable in the progress of diabetes, and there are no clinical drugs for the treatment of diabetic complications. 25, 26, 27 Inflammation and oxidative stress are the crucial mechanisms that contribute to hyperglycaemia‐induced cardiac injury. 28 Blockade of inflammation and oxidative stress can provide potent therapeutic effect in DCM. 17, 29 There are two main advantages of Cos for DCM treatment. Firstly, Eliza et al show that Cos significantly decreased glycosylated haemoglobin (HbA1c) and markedly increased plasma insulin level in diabetic Wistar rats. 30, 31 In their model, the mice were treated orally with Cos every day for 30 or 60 days on the first day the diabetes model was established, indicating that costunolide exhibits hypoglycaemic effect in early stage of diabetes. Secondly, in our study, Cos was orally administered once every 2 days for 8 weeks after T1DM model was established for 8 weeks. Our data show that Cos treatment did not alter body weight and serum glucose levels when compared to the diabetic group, indicating that Cos might not alleviate glycaemic dysfunction if given at the late stage of diabetes. However, Cos treatment inhibited hyperglycaemia‐induced inflammation and oxidative stress, leading to reduced cardiac damage in the end stage of diabetes. All these data prove that Cos can improve DCM during the whole pathology process of diabetes, making Cos as an excellent candidate for the treatment of DCM.
Excessive production of pro‐inflammatory factors and elevated ROS levels have been observed in DCM. 32 *There is* an intricate crosstalk between diabetes‐induced inflammatory responses and oxidative stress. 33, 34, 35 Recently, Nrf‐2 knockdown was shown to eliminate piceatannol‐ and fortunellin‐mediated anti‐inflammatory effects in DCM. 36, 37 Furthermore, SnPP IX, an HO‐1 inhibitor, reportedly inhibits the anti‐inflammatory function of beta‐naphthoflavone in lipopolysaccharide (LPS)‐induced inflammation in BV‐2 cells. 38 These findings suggest that the Nrf2/HO‐1 singling pathway may regulate inflammatory responses. Interestingly, oxidative stress may also be influenced by inflammatory responses. TLR4 knockdown was found to weaken NADPH oxidase activity and reduce ROS production in diabetic mice. 39 Yuan et al. have demonstrated that TAK242, a TLR4 inhibitor, decreases ROS accumulation and inhibits tubular cell apoptosis in diabetic mice. 40 BAY‐11‐7082, an NF‐κB inhibitor, was shown to upregulate the protein level of Nrf‐2 and reduce ROS production in human osteoarthritic chondrocytes. 41 These findings suggest that antioxidation may be regulated by TLR4/NF‐κB signalling, thereby providing a new perspective that NF‐κB‐mediated inflammation is closely associated with oxidative stress during the progression of DCM.
The anti‐inflammatory activity of Cos has been extensively reported and attributed to the inhibition of the NF‐κB and MAPK signalling pathways. 22, 42 Cos was also shown to exhibit cardioprotective effect in HFD‐induced obesity mice model 43 and doxorubicin‐induced toxicity in rats via inhibiting NF‐κB inflammatory signalling pathway. 23 However, the antioxidant activity of Cos has not been precisely elucidated. An ethanolic extract of Saussureae Radix was shown to attenuate neuroinflammation via induction of the Nrf‐2/HO‐1 signalling pathway. 44 Using high‐performance liquid chromatography analysis, it was confirmed that *Cos is* one of two major sesquiterpenoids composing the ethanolic extract of Saussureae Radix. Therefore, we speculated that Cos exhibits antioxidant activity. Furthermore, *Cos is* reported to trigger the antioxidative defence system by increasing Nrf‐2 and HO‐1 expression in the LPS‐ and D‐galactosamine‐induced acute liver injury model. 45 These findings indicate that the antioxidant activity of Cos may be mediated via Nrf‐2 and HO‐1 activation. In the present study, we demonstrated that Cos exhibited cardioprotective effect through inhibiting NF‐κB activation and restoring the suppressed antioxidant Nrf‐2 system. Although we cannot figure out whether the anti‐inflammatory or the antioxidant effect of Cos contributes more important role in treating DCM, we prove that the Nrf2 inhibitor, ML385 significantly reversed the anti‐inflammatory effects of Cos, indicating that Cos, at least partly, exhibited its cardiac protection effects via activating Nrf‐2 pathway.
There are some limitations in the present study. First, we only examined the effect of Cos in a single dosage in vivo. The dosage of Cos at a range of 10–30 mg/kg is frequently used in the treatment of inflammatory diseases. Among them, the dose of 20 mg/kg shows stable anti‐inflammatory activity and strong therapeutic effects in most of the tested models. 22, 31, 46 In consideration of the safety and effectiveness, 20 mg/kg was used as the administration dose in our in vivo study. In future investigations, multiple dosages will be examined to establish the pharmacological function of Cos in DCM treatment. In addition, we did not clarify the effect of Cos on cardiomyocyte apoptosis in DCM. It is well‐accepted that myocardial apoptosis is crucial for inducing cardiac fibrosis and hypertrophy. We speculate that the cardioprotective effect of Cos may be partly attributed to its anti‐apoptotic effect. Moreover, studies have shown that Cos suppresses apoptosis of tumour cells in various tumour models. 20, 47, 48 Mao et al. have reported the anti‐apoptotic capability of Cos in LPS‐induced acute lung injury, 45 indicating the anti‐apoptotic potential of Cos in inflammatory diseases. N‐acetyl‐L‐cysteine, an antioxidant and glutathione precursor, can prevent myocardial apoptosis in a type 1 diabetes mouse model. 49 Therefore, Cos may also suppress myocardial apoptosis in an antioxidative stress manner. However, a comprehensive study is needed to elucidate the possible mechanisms elucidating how Cos regulates both inflammatory and oxidative responses in DCM. Finally, our study demonstrated that Cos inhibited IκB‐α degradation and nuclear translocation of P65. Nevertheless, Cos may inhibit the upstream NF‐κB protein. Cos treatment can reduce TAK1 phosphorylation and inhibit the interaction between TAK1 and TAB1. 42 In addition, IKK‐β may be another target of Cos, given that Cos inhibits LPS‐induced BV2 microglial inflammation by inhibiting the IKK‐β/NF‐κB signalling pathway. 50 Parthenolide, a sesquiterpene lactone with a highly similar parent nucleus and double‐bond structure to Cos, can decrease advanced oxidation protein product (AOPP)‐induced MCP‐1 expression by inhibiting the phosphorylation of IKK‐β and NF‐κB p65. 51 Using molecular docking technology, Mondawood et al. have shown that parthenolide potently binds with IKK‐β. These results indicate that IKK‐β may be the direct target of Cos. 52 Accordingly, Cos may be a pan‐NF‐κB inhibitor and confer protection against DCM via its anti‐inflammatory activities.
In summary, our study suggests that Cos attenuates cardiac fibrosis and hypertrophy by inhibiting inflammation and oxidative stress in the STZ‐induced mouse model. Furthermore, Cos can reverse HG‐induced H9c2 fibrosis and hypertrophy in vitro. Mechanistically, Cos‐induced anti‐inflammatory and antioxidative effects were mediated by simultaneously inhibiting the NF‐κB signalling pathway and inducing the Nrf2/HO‐1 pathway. Since Cos shows significant hypoglycaemic effect in the early stage of diabetes and provides cardioprotective effect via anti‐inflammatory and antioxidant activities in end stage of diabetes, Cos may be a potential therapeutic candidate for treating DCM.
## AUTHOR CONTRIBUTIONS
Bo Jin: Data curation (equal); investigation (lead); methodology (lead). Yi Chen: Investigation (equal); methodology (equal); writing – original draft (equal). Jiong Wang: Investigation (supporting); validation (supporting); visualization (supporting). Yue Chen: *Formal analysis* (equal); investigation (supporting). Mengpei Zhang: Data curation (supporting); investigation (supporting); writing – original draft (supporting). Jianxiong Huang: Conceptualization (equal); data curation (equal); project administration (equal); writing – review and editing (equal).
## CONFLICT OF INTEREST STATEMENT
The authors declare no conflict of interest.
## DATA AVAILABILITY STATEMENT
All the data are available from the authors on request.
## References
1. Chavali V, Tyagi SC, Mishra PK. **Predictors and prevention of diabetic cardiomyopathy**. *Diabetes Metab Syndr Obes* (2013) **6** 151-160. DOI: 10.2147/DMSO.S30968
2. Kang YJ. **Molecular and cellular mechanisms of cardiotoxicity**. *Environ Health Perspect* (2001) **109** 27-34. DOI: 10.1289/ehp.01109s127
3. Nishida K, Otsu K. **Inflammation and metabolic cardiomyopathy**. *Cardiovasc Res* (2017) **113** 389-398. DOI: 10.1093/cvr/cvx012
4. Bugger H, Abel ED. **Molecular mechanisms of diabetic cardiomyopathy**. *Diabetologia* (2014) **57** 660-671. DOI: 10.1007/s00125-014-3171-6
5. Cai L, Kang YJ. **Cell death and diabetic cardiomyopathy**. *Cardiovasc Toxicol* (2003) **3** 219-228. DOI: 10.1385/ct:3:3:219
6. Evangelista I, Nuti R, Picchioni T, Dotta F, Palazzuoli A. **Molecular dysfunction and phenotypic derangement in diabetic cardiomyopathy**. *Int J Mol Sci* (2019) **20** 3264. DOI: 10.3390/ijms20133264
7. Kim DY, Choi BY. **Costunolide‐a bioactive sesquiterpene lactone with diverse therapeutic potential**. *Int J Mol Sci* (2019) **20** 2926. DOI: 10.3390/ijms20122926
8. Xie F, Zhang H, Zheng C, Shen XF. **Costunolide improved dextran sulfate sodium‐induced acute ulcerative colitis in mice through NF‐kappaB, STAT1/3, and Akt signaling pathways**. *Int Immunopharmacol* (2020) **84**. DOI: 10.1016/j.intimp.2020.106567
9. Liu B, Rong Y, Sun D. **Costunolide inhibits pulmonary fibrosis via regulating NF‐kB and TGF‐beta1/Smad2/Nrf2‐NOX4 signaling pathways**. *Biochem Biophys Res Commun* (2019) **510** 329-333. DOI: 10.1016/j.bbrc.2019.01.104
10. Chen YT, Du Y, Zhao B. **Costunolide alleviates HKSA‐induced acute lung injury via inhibition of macrophage activation**. *Acta Pharmacol Sin* (2019) **40** 1040-1048. DOI: 10.1038/s41401-018-0192-6
11. Kong SS, Liu JJ, Yu XJ, Lu Y, Zang WJ. **Protection against ischemia‐induced oxidative stress conferred by vagal stimulation in the rat heart: involvement of the AMPK‐PKC pathway**. *Int J Mol Sci* (2012) **13** 14311-14325. DOI: 10.3390/ijms131114311
12. Danese E, Montagnana M. **An historical approach to the diagnostic biomarkers of acute coronary syndrome**. *Ann Transl Med* (2016) **4**. DOI: 10.21037/atm.2016.05.19
13. Lafuse WP, Wozniak DJ, Rajaram MVS. **Role of cardiac macrophages on cardiac inflammation, fibrosis and tissue repair**. *Cell* (2020) **10** 51. DOI: 10.3390/cells10010051
14. Elia E, Ministrini S, Carbone F, Montecucco F. **Diabetic cardiomyopathy and inflammation: development of hostile microenvironment resulting in cardiac damage**. *Minerva Cardiol Angiol* (2022) **70** 357-369. DOI: 10.23736/S2724-5683.20.05454-7
15. Navarro‐Gonzalez JF, Mora‐Fernandez C, Muros de Fuentes M, Garcia‐Perez J. **Inflammatory molecules and pathways in the pathogenesis of diabetic nephropathy**. *Nat Rev Nephrol* (2011) **7** 327-340. DOI: 10.1038/nrneph.2011.51
16. Kaludercic N, Di Lisa F. **Mitochondrial ROS formation in the pathogenesis of diabetic cardiomyopathy**. *Front Cardiovasc Med* (2020) **7**. DOI: 10.3389/fcvm.2020.00012
17. Tang Z, Wang P, Dong C, Zhang J, Wang X, Pei H. **Oxidative stress signaling mediated pathogenesis of diabetic cardiomyopathy**. *Oxid Med Cell Longev* (2022) **2022** 5913374-5913378. DOI: 10.1155/2022/5913374
18. Chen X, Qian J, Wang L. **Kaempferol attenuates hyperglycemia‐induced cardiac injuries by inhibiting inflammatory responses and oxidative stress**. *Endocrine* (2018) **60** 83-94. DOI: 10.1007/s12020-018-1525-4
19. Hou J, Zheng D, Zhong G, Hu Y. **Mangiferin mitigates diabetic cardiomyopathy in streptozotocin‐diabetic rats**. *Can J Physiol Pharmacol* (2013) **91** 759-763. DOI: 10.1139/cjpp-2013-0090
20. Yaribeygi H, Maleki M, Mohammadi MT, Sathyapalan T, Jamialahmadi T, Sahebkar A. **Crocin improves diabetes‐induced oxidative stress via downregulating the Nox‐4 in myocardium of diabetic rats**. *Adv Exp Med Biol* (2021) **1328** 275-285. DOI: 10.1007/978-3-030-73234-9_18
21. He Y, Moqbel SAA, Xu L. **Costunolide inhibits matrix metalloproteinases expression and osteoarthritis via the NFkappaB and Wnt/betacatenin signaling pathways**. *Mol Med Rep* (2019) **20** 312-322. DOI: 10.3892/mmr.2019.10239
22. Wang Y, Zhang X, Zhao L. **Costunolide protects lipopolysaccharide/d‐galactosamine‐induced acute liver injury in mice by inhibiting NF‐kappaB signaling pathway**. *J Surg Res* (2017) **220** 40-45. DOI: 10.1016/j.jss.2017.06.083
23. Christidi E, Brunham LR. **Regulated cell death pathways in doxorubicin‐induced cardiotoxicity**. *Cell Death Dis* (2021) **12**. DOI: 10.1038/s41419-021-03614-x
24. **14. Management of Diabetes in pregnancy: standards of medical Care in Diabetes‐2020**. *Diabetes Care* (2020) **43** S183-S192. DOI: 10.2337/dc20-S014
25. El Hayek MS, Ernande L, Benitah JP, Gomez AM, Pereira L. **The role of hyperglycaemia in the development of diabetic cardiomyopathy**. *Arch Cardiovasc Dis* (2021) **114** 748-760. DOI: 10.1016/j.acvd.2021.08.004
26. Papachristoforou E, Lambadiari V, Maratou E, Makrilakis K. **Association of Glycemic Indices (hyperglycemia, glucose variability, and hypoglycemia) with oxidative stress and diabetic complications**. *J Diabetes Res* (2020) **2020**. DOI: 10.1155/2020/7489795
27. Forbes JM, Cooper ME. **Mechanisms of diabetic complications**. *Physiol Rev* (2013) **93** 137-188. DOI: 10.1152/physrev.00045.2011
28. Tan Y, Zhang Z, Zheng C, Wintergerst KA, Keller BB, Cai L. **Mechanisms of diabetic cardiomyopathy and potential therapeutic strategies: preclinical and clinical evidence**. *Nat Rev Cardiol* (2020) **17** 585-607. DOI: 10.1038/s41569-020-0339-2
29. Kaur N, Guan Y, Raja R, Ruiz‐Velasco A, Liu W. **Mechanisms and therapeutic prospects of diabetic cardiomyopathy through the inflammatory response**. *Front Physiol* (2021) **12**. DOI: 10.3389/fphys.2021.694864
30. Eliza J, Daisy P, Ignacimuthu S, Duraipandiyan V. **Normo‐glycemic and hypolipidemic effect of costunolide isolated from Costus speciosus (Koen ex. Retz.)Sm. in streptozotocin‐induced diabetic rats**. *Chem Biol Interact* (2009) **179** 329-334. DOI: 10.1016/j.cbi.2008.10.017
31. Eliza J, Daisy P, Ignacimuthu S. **Antioxidant activity of costunolide and eremanthin isolated from Costus speciosus (Koen ex. Retz) Sm**. *Chem Biol Interact* (2010) **188** 467-472. DOI: 10.1016/j.cbi.2010.08.002
32. Guo Y, Zhuang X, Huang Z. **Klotho protects the heart from hyperglycemia‐induced injury by inactivating ROS and NF‐kappaB‐mediated inflammation both in vitro and in vivo**. *Biochim Biophys Acta Mol Basis Dis* (2018) **1864** 238-251. DOI: 10.1016/j.bbadis.2017.09.029
33. Domingueti CP, Dusse LM, Carvalho M, de Sousa LP, Gomes KB, Fernandes AP. **Diabetes mellitus: the linkage between oxidative stress, inflammation, hypercoagulability and vascular complications**. *J Diabetes Complications* (2016) **30** 738-745. DOI: 10.1016/j.jdiacomp.2015.12.018
34. Yuan T, Yang T, Chen H. **New insights into oxidative stress and inflammation during diabetes mellitus‐accelerated atherosclerosis**. *Redox Biol* (2019) **20** 247-260. DOI: 10.1016/j.redox.2018.09.025
35. Newsholme P, Cruzat VF, Keane KN, Carlessi R, de Bittencourt PI. **Molecular mechanisms of ROS production and oxidative stress in diabetes**. *Biochem J* (2016) **473** 4527-4550. PMID: 27941030
36. Li H, Shi Y, Wang X. **Piceatannol alleviates inflammation and oxidative stress via modulation of the Nrf2/HO‐1 and NF‐kappaB pathways in diabetic cardiomyopathy**. *Chem Biol Interact* (2019) **310**. DOI: 10.1016/j.cbi.2019.108754
37. Zhao C, Zhang Y, Liu H, Li P, Zhang H, Cheng G. **Fortunellin protects against high fructose‐induced diabetic heart injury in mice by suppressing inflammation and oxidative stress via AMPK/Nrf‐2 pathway regulation**. *Biochem Biophys Res Commun* (2017) **490** 552-559. DOI: 10.1016/j.bbrc.2017.06.076
38. Gao X, He D, Liu D. **Beta‐naphthoflavone inhibits LPS‐induced inflammation in BV‐2 cells via AKT/Nrf‐2/HO‐1‐NF‐kappaB signaling axis**. *Immunobiology* (2020) **225**. DOI: 10.1016/j.imbio.2020.151965
39. Zhang Y, Peng T, Zhu H. **Prevention of hyperglycemia‐induced myocardial apoptosis by gene silencing of toll‐like receptor‐4**. *J Transl Med* (2010) **8**. DOI: 10.1186/1479-5876-8-133
40. Yuan S, Liu X, Zhu X. **The role of TLR4 on PGC‐1alpha‐mediated oxidative stress in tubular cell in diabetic kidney disease**. *Oxid Med Cell Longev* (2018) **2018**. DOI: 10.1155/2018/6296802
41. Cheleschi S, Tenti S, Mondanelli N. **MicroRNA‐34a and MicroRNA‐181a mediate Visfatin‐induced apoptosis and oxidative stress via NF‐kappaB pathway in human osteoarthritic chondrocytes**. *Cell* (2019) **8** 874. DOI: 10.3390/cells8080874
42. Chen Z, Zhang D, Li M, Wang B. **Costunolide ameliorates lipoteichoic acid‐induced acute lung injury via attenuating MAPK signaling pathway**. *Int Immunopharmacol* (2018) **61** 283-289. DOI: 10.1016/j.intimp.2018.06.017
43. Ye B, Chen X, Chen Y. **Inhibition of TAK1/TAB2 complex formation by costunolide attenuates obesity cardiomyopathy via the NF‐kappaB signaling pathway**. *Phytomedicine* (2023) **108**. DOI: 10.1016/j.phymed.2022.154523
44. Oh YC, Li W, Choi JG. **Saussureae radix attenuates neuroinflammation in LPS‐stimulated mouse BV2 microglia via HO‐1/Nrf‐2 induction and inflammatory pathway inhibition**. *Mediators Inflamm* (2021) **2021**. DOI: 10.1155/2021/6687089
45. Mao J, Yi M, Wang R, Huang Y, Chen M. **Protective effects of Costunolide against D‐galactosamine and lipopolysaccharide‐induced acute liver injury in mice**. *Front Pharmacol* (2018) **9**. DOI: 10.3389/fphar.2018.01469
46. Zheng H, Chen Y, Zhang J. **Evaluation of protective effects of costunolide and dehydrocostuslactone on ethanol‐induced gastric ulcer in mice based on multi‐pathway regulation**. *Chem Biol Interact* (2016) **250** 68-77. DOI: 10.1016/j.cbi.2016.03.003
47. Xu C, Huang X, Lei X. **Costunolide‐induced apoptosis via promoting the reactive oxygen species and inhibiting AKT/GSK3beta pathway and activating autophagy in gastric cancer**. *Front Cell Dev Biol* (2021) **9**. DOI: 10.3389/fcell.2021.722734
48. Huang H, Yi JK, Lim SG. **Costunolide induces apoptosis via the reactive oxygen species and protein kinase B pathway in Oral cancer cells**. *Int J Mol Sci* (2021) **22** 7509. DOI: 10.3390/ijms22147509
49. Fiordaliso F, Bianchi R, Staszewsky L. **Antioxidant treatment attenuates hyperglycemia‐induced cardiomyocyte death in rats**. *J Mol Cell Cardiol* (2004) **37** 959-968. DOI: 10.1016/j.yjmcc.2004.07.008
50. Liu YC, Feng N, Li WW. **Costunolide plays an anti‐neuroinflammation role in lipopolysaccharide‐induced BV2 microglial activation by targeting cyclin‐dependent kinase 2**. *Molecules* (2020) **25** 2840. DOI: 10.3390/molecules25122840
51. Zhao Y, Chen SJ, Wang JC. **Sesquiterpene lactones inhibit advanced oxidation protein product‐induced MCP‐1 expression in podocytes via an IKK/NF‐kappaB‐dependent mechanism**. *Oxid Med Cell Longev* (2015) **2015**. DOI: 10.1155/2015/934058
52. Dawood M, Ooko E, Efferth T. **Collateral sensitivity of Parthenolide via NF‐kappaB and HIF‐alpha inhibition and epigenetic changes in drug‐resistant cancer cell lines**. *Front Pharmacol* (2019) **10**. DOI: 10.3389/fphar.2019.00542
|
---
title: 'Diet quality and dietary acid load in relation to cardiovascular disease mortality:
Results from Fasa PERSIAN cohort study'
authors:
- Sahar Fereidouni
- Najmeh Hejazi
- Reza Homayounfar
- Mojtaba Farjam
journal: Food Science & Nutrition
year: 2022
pmcid: PMC10002926
doi: 10.1002/fsn3.3197
license: CC BY 4.0
---
# Diet quality and dietary acid load in relation to cardiovascular disease mortality: Results from Fasa PERSIAN cohort study
## Abstract
Dietary intake is a determining factor in the morbidity and mortality of chronic disorders. However, not many documents have investigated this relationship. The aim of this study was to evaluate the associations of the Mediterranean dietary score (MDS), Alternative Healthy Eating Index (AHEI), Dietary Inflammatory Index (DII), DASH score, and dietary acid load with cardiovascular disease (CVD) mortality. A total of 2158 CVD patients (mean age of 54.73 ± 8.62 years) from the Fasa cohort study, Iran, participated in the current study. Diet quality indices including DII, AHEI, MDS, DASH, and dietary acid load (NEAP score) were computed using a validated 125‐item Food Frequency Questionnaire (FFQ). Cox regression analyses were used to determine HRs and $95\%$ CIs. During a follow‐up of 3 years, we documented 59 CVD deaths. After adjusting for relevant confounders (age, gender, family history of CVD, smoking, physical activity, alcohol intake, and HTN) in the final model, we found that higher DII scores and dietary acid load were significantly related to increased mortality due to CVD (HR = 1.11; $95\%$ CI = 1.01–1.24; and HR = 1.02; $95\%$ CI = 1.01–1.03). However, the DASH score was insignificantly associated with decreased CVD mortality by $20.4\%$ (HR = 0.79; $95\%$ CI = 0.57–1.09). There was no significant relationship among AHEI score, MDS, and CVD mortality. This study showed that increasing dietary acidity and the use of inflammatory food compounds could contribute to CVD mortality. Also, adherence to the DASH diet may be associated with reduced CVD mortality.
The aim of this study was to evaluate the associations of the Mediterranean dietary score (MDS), Alternative Healthy Eating Index (AHEI), Dietary Inflammatory Index (DII), DASH score, and dietary acid load with cardiovascular disease (CVD) mortality. This study showed that increasing dietary acidity and the use of inflammatory food compounds could contribute to CVD mortality. Also, adherence to the DASH diet may be associated with reduced CVD mortality.
## INTRODUCTION
Cardiovascular disease (CVD) is referred to as a group of related disorders that include ischemic heart disease, hypertension, atherosclerosis, heart failure, and peripheral arterial disease. The prevalence of CVD and its mortality have been 422.7 million cases and 17.9 million, during a year in the world, respectively (Emamian et al., 2020; Sarrafzadegan & Mohammmadifard, 2019). Evidence has shown that age, genetics, ethnicity, gender, hypertension, smoking history, abdominal obesity, physical inactivity, and poor diet are the risk factors for CVD; nevertheless, by identifying these risk factors in individuals, we can adopt an appropriate strategy for treatment and prevention (Sarrafzadegan & Mohammmadifard, 2019). Among the risk factors, dietary modification was an important approach for the management and prevention of CVD and its mortality. A diet with excessive sodium, saturated fatty acids, cholesterol, refined grains, red and processed meat, and alcohol could increase the risk of CVD and its mortality, while consumption of fruits, seafood, vegetables, whole grains legumes, nuts, and unsaturated fatty acids is shown to be linked to reduced mortality of CVD (Aigner et al., 2018). Evidence suggested that Mediterranean dietary patterns and the dietary approach to stop hypertension (DASH) diet could reverse the CVD progression (Levitan et al., 2013). Previous documents for the control and management of CVD focused mainly on single micronutrients or food items. However, these days, instead of evaluating single nutrients or groups of food, analysis of dietary patterns and dietary indices considers interactions and dependencies among nutrients and provides appropriate strategies for chronic disease management and prevention (Aigner et al., 2018).
The quality of the diet can be assessed by the Alternative Healthy Eating Index (AHEI), Mediterranean diet score (MED), and DASH scores indices. According to a meta‐analysis, the higher the score of these indices was significantly inversely associated with CVD progression and all‐reason mortality (Aigner et al., 2018; Akbaraly et al., 2011).
Recently, studies introduce acid–base dietary imbalance as a risk factor for chronic diseases, so a higher dietary acid load increases the incidence of CVD, hypertension, and related mortality (Akter et al., 2017). Consuming low amounts of vegetables and fruits simultaneously with eating excessive processed red meat can reduce the quality of the diet and increase its acid load (Fatahi & Azadbakht, 2019). The dietary acid load could be elevated using dietary acid load (DAL), potential renal acid load (PRAL), and net endogenous acid production (NEAP). Higher scores of PRAL, DAL, and NEAP scores show more acid‐loading potential (Fatahi & Azadbakht, 2019).
Evidence has demonstrated that inflammation is involved in the stages of atherosclerosis leading to plaque rupture and thrombosis which increase the probability of CVD mortality. In this regard, a diet rich in antiinflammatory ingredients plays an important role in reducing the severity and consequence of inflammatory chronic diseases the same as CVDs and this mortality. For evaluating the antiinflammatory and inflammatory strength of a diet, a dietary inflammatory index (DII) is recommended. A higher score of DII, indicating a higher inflammatory potential of diet, has been related to the risk of CVD and its mortality (Hodge et al., 2018; Shivappa et al., 2014).
The recent study aimed to investigate the prospective relationships of the DII, DASH, MDS, AHEI score, and dietary acid load with the mortality of CVD in the context of the epidemiologic Persian cohort study.
## Study population
The present study included 1622 women and 536 men with cardiovascular diseases who participated in the Fasa PERSIAN cohort as a branch of the Prospective Epidemiological Research Study in Iran. It was conducted between November 2014 and June 2019. In this cohort study, 10,135 subjects aged 35 to 70 years who were not physically or mentally disabled, and lived in Sheshdeh, a district of Fasa, participated for more than 9 months each year (Farjam et al., 2016). A total of 2222 subjects were excluded from the study due to incomplete data on the intake of diet and mortality status as well as reports of abnormal energy intake (less than 800 kcal and more than 4200 kcal). Finally, a total of 2158 participants with CVD took part in the current research. Information on demographic, behavioral, anthropometric data, medical history, and intake of foods was assessed through questionnaires biennially in the cohort study; likewise, mortality status and the cause of mortality were determined during the annual follow‐up (Farjam et al., 2016). In this study, all procedures involving patients were approved by the Ethics Committee of Shiraz University of Medical Sciences, Shiraz, Iran (code: IR.SUMS.REC.1399.1116). Written informed consent was provided by all patients.
## Dietary intake assessment and indices
The usual food intake of patients was determined using a validated 125‐item Food Frequency Questionnaire (FFQ) which was modified by the Iranian food culture (Farjam et al., 2016; Willett et al., 1985). An expert nutritionist in a face‐to‐face interview registered the amount of food consumption in the last year at the baseline of the cohort study. Nutritionist IV software (version 7.0) was used to determine the nutrient contents and energy of foods (Farjam et al., 2016).
## Alternative healthy eating index‐2010 (AHEI‐2010)
Kennedy et al. designed the AHEI index to assess the quality of the diet (Schwingshackl, Bogensberger & Hoffmann, 2018). This index includes 11 components and the individuals are scored based on the consumption of these items. Legumes, vegetables, whole grains, nuts and fruits, DHA and EPA, and polyunsaturated fatty acids are positive components, while transfatty acids, sweetened beverages, sodium, and red and processed meats are the negative components; alcohol is the moderate component. For calculating, first, we categorized the individuals based on consumption deciles and then inverted the negative component decile score, and finally, summed the total score. The overall score was between 9 and 81. A higher AHEI score indicates a healthier diet (McCullough et al., 2002).
## DASH score
There are different methods for calculating the DASH diet score. In this study, Mellen's DASH Index was applied. Its components included fiber, calcium, magnesium, cholesterol, sodium, potassium, protein, total fat, and saturated fat. The score was considered 1 point for the goal consumption and 0.5 for the intermediate intakes. The overall score ranged from 0 to 9. A higher score indicates greater adherence to the DASH diet (Miller et al, 2013).
## Mediterranean dietary score (MDS)
In this study, we used MEDI‐LITE scoring for this pattern, which was described by Francesco Sophie et al. This score focuses on nine components (vegetable, dairy products, fruit, fish, meat, grains, meat products, legumes, olive oil, and alcohol). Higher intake of cereals, legumes, fruit, fish, and vegetables is scored 2 points, intermediate consumption 1 points, and lower intake of them 0 points. The scoring of dairy products, processed meat, and meat was the opposite of the previous groups. Medium consumption of alcohol was scored 2 points, the lowest consumption 1 point, and the highest intake 0 points. The final score was between 0 and 18 points. Higher scores indicate more adherence to the MDS (Sofi, Macchi, Abbate, Gensini & Casini, 2014).
## Dietary inflammatory index (DII)
We computed the DII score based on 30 groups of food including energy, fat, transfat, cholesterol, carbohydrate, protein, saturated fat, vitamin B12, iron, MUFA, PUFA, vitamin D, vitamin B6, fiber, vitamin B9, vitamin C, niacin, thiamin, riboflavin, vitamin A, magnesium, vitamin E, β‐carotene, onion, garlic, tea, caffeine, selenium, and zinc. Consumption of groups such as eugenol, turmeric, saffron, ginger, pepper, rosemary, polyphenols, and anthocyanin was not available for the calculation of this index. First, the energy intake of the participants was adjusted based on 1000 kcal. Then, to calculate DII, we subtracted the dietary parameters from global average and divided it by the “global standard deviation” to get a Z score. The Z‐score values were then converted into percentiles. The percentile values were then multiplied by 2 minus 1. Finally, the scores obtained from each of the 30 parameters were multiplied by the overall inflammatory score; then, we summed up all food items to calculate the total DII score (Shivappa et al., 2014).
## Dietary acid load
We constructed the acid load of the diet based on the consumption of several nutrients using three different methods: PRAL (mEq/d) = (0.49 × protein [g/d]) + (0.037 × Phosphorous [mg/d]) – (0.021 × potassium [mg/d]) – (0.013 × calcium [mg/d]) – (0.026 × magnesium [mg/d]), NEAP (mEq/day) = (protein [g/d] × 54.5/potassium [mEq/d])‐10.2, and DAL (mEq/day) = (body surface area [m2] × 41 [mEq/day]/1.73 m2) + PRAL. Du Bois formula: height0.725 × weight0.425 × 0.00718 was used for computing the surface of the body. Acid load score obtained from these three methods was used for statistical analysis (Han et al., 2016).
## Assessment of other variables
Cardiovascular patients were followed up clinically during a 3‐year period. They were identified based on previous medical history, electrocardiography, laboratory sampling, validated screening questionnaires, and physical exams. Physical activity measurement was done by the International Physical Activity Questionnaire (IPAQ). Using a digital scale (Tanita BC‐418, Tanita Corp, Japan), we recorded the weight and height. Body mass index (BMI) was then computed by weight (kg)/height (m)2. Waist circumference was measured with a precision of 0.1 cm. The medical history and lifestyle factors that are associated with CVD risk were recorded (Farjam et al., 2016).
## Outcome assessment
In this study, our outcome was cardiovascular mortality; based on the 10th edition of the International Classification of Diseases (ICD), deaths due to the following diseases are considered CVD mortality: hypertension (ICD I10‐I15), peripheral vascular disease (ICD I70‐I89), ischemic heart disease (ICD I20‐I25), stroke (ICD I60‐I69), and pulmonary heart disease (ICD I26‐I28). The patients' CVD mortality was followed for 3 years after the first year of enrollment in the study.
## Statistical analysis
SPSS software version 21 was used for data analysis. Quantitative variables are reported as mean ± standard error and qualitative data are presented as frequency (percentage). The normality of the distribution of variables was assessed using the Kolmogorov–Smirnov test. Chi‐Square tests and Mann–Whitney U tests were used to compare qualitative and quantitative data among the participants based on gender, respectively. A Cox regression analysis was used to assess the relationship among the four dietary quality indices (DII, DASH score, MDS, and AHEI), dietary acid load, and CVD mortality. The outcome was defined as mortality, the temporal factor was time to the event, and exposure was one of the dietary indices. We did not adjust for any covariates in the first model (basic model). The second model was adjusted for covariates including age, gender, smoking, level of physical activity, alcohol intake, family history of CVD, and history of HTN. The third model was adjusted by total energy, waist‐to‐hip ratio (WHR), weight, BMI, DBP, SBP, TG, cholesterol, LDL, and HDL. Finally, the model was adjusted with all the components of models 2 and 3. The hazard ratio (AR) and $95\%$ confidence interval (CI) were presented to show the strength of the relationship between dietary quality indices including DII, AHEI, MDS, and DASH score, and dietary acid load and CVD mortality. p‐values of <.05 were considered significant.
## RESULTS
A total of 2158 individuals ($24.8\%$ males and $75.2\%$ females), with a mean age of 54.73 ± 8.62 years, participated in the current study. During a 3‐year follow‐up of the present study participants, 59 deaths (30 males and 29 females) due to CVD got recorded.
Table 1 displays the basic characteristics of the study patients. As shown in Table 1, the mean weight, age, and serum cholesterol were significantly higher in men with CVDs compared to women (p = <.05), but their BMI, WHR, and the serum level of TG, HDL, and LDL were significantly lower (p = <.05). Based on Table 1, the frequency distribution of the participants in terms of the history of hypertension was significantly higher in women with CVDs compared to men. ( p = <.05). However, the frequency distribution of the records of alcohol consumption and active smoking was significantly lower (p = <.05). Table 1 also shows that 5.6 and 1.8 percent of men and women with CVDs died due to cardiovascular events, which is significantly higher in men (p = <.05) (Table 1).
**TABLE 1**
| Unnamed: 0 | Total | Men (n = 536) | Women (n = 1622) | p‐value |
| --- | --- | --- | --- | --- |
| Age (year) a | 54.73 ± 8.62 | 57.04 ± 8.81 | 54.41 ± 9.02 | <.0001 |
| Weight (Kg) a | 67.76 ± 1.32 | 71.13 ± 13.47 | 66.67 ± 12.83 | <.0001 |
| BMI (kg/m2) a | 27.09 ± 4.94 | 25.30 ± 4.28 | 27.7 ± 4.96 | <.0001 |
| WHR a | 0.96 ± 0.06 | 0.94 ± 0.06 | 0.96 ± 0.06 | <.0001 |
| Physical activity (MET) a | 38.45 ± 8.38 | 40.68 ± 12.47 | 37.7 ± 6.3 | .65 |
| DBP (mmHg) a | 79.97 ± 13 | 79.65 ± 13.83 | 80.07 ± 12.73 | .63 |
| SBP (mmHg) a | 123.04 ± 21.64 | 122.67 ± 22.28 | 123.1473 ± 21.44 | .82 |
| Daily energy intake (kcal) a | 2670.32 ± 750.17 | 2985.86 ± 693.26 | 2564.09 ± 739.32 | <.0001 |
| History of HTN (%) b | 79.1 | 64.6 | 84.0 | <.0001 |
| alcohol consumption (%) b | 0.9 | 3.5 | 0.0 | <.0001 |
| Active smoking (%) b | 18.6 | 50.4 | 8.1 | <.0001 |
| FBS | 102.9 ± 41.44 | 101.91 ± 37.50 | 103.32 ± 42.73 | .96 |
| Cholesterol (mg/dl) a | 189.18 ± 43.27 | 176.13 ± 45.58 | 193.47 ± 41.59 | <.0001 |
| TG (mg/dl) a | 142.13 ± 86.10 | 139.57 ± 97.57 | 143.11 ± 82.08 | .01 |
| LDL (mg/dl) a | 108.86 ± 36.05 | 101.50 ± 37.52 | 111.22 ± 35.19 | <.0001 |
| HDL (mg/dl) a | 81.85 ± 15.49 | 46.71 ± 14.29 | 53.57 ± 15.53 | <.0001 |
| CVD mortality (%) b | 3.1 | 5.6 | 1.8 | <.0001 |
Dietary indices and intake of the participants of the study are reported in Table 2. As Table 2 indicates, the level of energy intake and macro‐ and micronutrients are significantly higher in men compared to women (p = <.05). The Table 2 did not show any significant difference in dietary indices among male and female participants of the study (Table 2).
**TABLE 2**
| Unnamed: 0 | Total | Men (n = 536) | Women (n = 1622) | p‐value a |
| --- | --- | --- | --- | --- |
| Daily energy intake (kcal) a | 3667.97 ± 16.16 | 2985.86 ± 693.26 | 2564.09 ± 739.32 | <.0001 |
| Carbohydrate (gr) | 465.01 ± 2.97 | 521.35 ± 5.61 | 446.40 ± 3.37 | <.0001 |
| Protein (gr) | 81.54 ± 0.55 | 91.81 ± 1.08 | 78.15 ± 0.62 | <.0001 |
| Fat (gr) | 58.04 ± 0.51 | 63.78 ± 1.10 | 56.14 ± 0.57 | <.0001 |
| Monounsaturated fatty acid (gr) | 16.98 ± 0.17 | 19.15 ± 0.41 | 16.26 ± 0.18 | <.0001 |
| Polyunsaturated fatty acid (gr) | 8.21 ± 0.08 | 9.12 ± 0.17 | 7.91 ± 0.09 | <.0001 |
| Saturated fatty acid (gr) | 22.09 ± 0.26 | 23.53 ± 0.55 | 21.62 ± 0.30 | .001 |
| Alpha‐linolenic (gr) | 0.33 ± 0.005 | 0.35 ± 0.011 | 0.32 ± 0.006 | .008 |
| Linoleic (gr) | 0.13 ± 0.002 | 0.35 ± 0.01 | 0.32 ± 0.006 | .008 |
| Cholesterol (mg) | 205.52 ± 2.50 | 250.47 ± 5.60 | 190.67 ± 2.67 | <.0001 |
| DHA (gr) | 0.03 ± 0.0005 | 0.04 ± 0.001 | 0.02 ± 0.0005 | <.0001 |
| EPA (gr) | 0.012 ± 0.0003 | 0.015 ± 0.0007 | 0.011 ± 0.0004 | <.0001 |
| Sodium (mg) | 1281.49 ± 4.13 | 1316.18 ± 2.12 | 1356 ± 2.42 | <.0001 |
| Potassium (mg) | 1157.35 ± 4.41 | 1212.12 ± 2.45 | 1102 ± 2.87 | <.0001 |
| Fiber (gr) | 27.06 ± 0.23 | 28.56 ± 0.46 | 26.56 ± 0.26 | <.0001 |
| Calcium (mg) | 513.88 ± 5.74 | 549.15 ± 10.89 | 502.22 ± 6.71 | <.0001 |
| Vitamin D (IU) | 36.24 ± 0.61 | 48.89 ± 1.43 | 32.05 ± 0.63 | <.0001 |
| Vitamin C (mg) | 141.47 ± 2.08 | 146.08 ± 3.78 | 139.95 ± 2.47 | .039 |
| Vitamin E (mg) | 36.24 ± 0.61 | 32.05 ± 0.63 | 12.56 ± 0.23 | <.0001 |
| Vitamin K (Ug) | 239.89 ± 5.91 | 235.13 ± 9.53 | 241.46 ± 7.21 | .510 |
| Folate (mg) | 374.92 ± 4.73 | 403.90 ± 9.48 | 365.35 ± 5.43 | <.0001 |
| Iron (mg) | 17.61 ± 0.12 | 19.26 ± 0.24 | 17.06 ± 0.14 | <.0001 |
| Zinc (mg) | 6.40 ± 0.05 | 7.24 ± 0.11 | 6.12 ± 0.059 | <.0001 |
| Magnesium (mg) | 259.82 ± 2.26 | 288.42 ± 4.48 | 250.37 ± 2.57 | <.0001 |
| Selenium (mg) | 41.52 ± 0.47 | 49.89 ± 1.06 | 38.75 ± 0.49 | <.0001 |
| AHEI score a | 62.05 ± 10.41 | 62.81 ± 10.96 | 61.79 ± 10.20 | .06 |
| MDS score a | 9.12 ± 2.42 | 9.09 ± 2.45 | 9.14 ± 2.42 | .47 |
| DASH score a | 3.45 ± 0.80 | 3.45 ± 0.82 | 3.44 ± 0.79 | .70 |
| DII score a | 0.05 ± 2.62 | −0.06 ± 2.44 | 0.08 ± 2.67 | .14 |
| NEAP a | 49.42 ± 22.10 | 50.57 ± 24.88 | 49.00 ± 21.00 | .58 |
The association between different variables and CVDs mortality in a Cox univariate model is shown in Table 3. It was revealed that being older (HR = 1.065; $95\%$ CI = 1.02–1.1; $$p \leq .001$$), having a history of HTN (HR = 4.62; $95\%$ CI = 2.69–7.92; p = <.0001), being an active smoker (HR = 1.78; $95\%$ CI = 1.02–3.11; $$p \leq .041$$), and being a man (HR = 3.46; $95\%$ CI = 2.12–5.64; p = <.0001) were significantly associated with the increased risk of CVDs mortality. Also, there was an inverse significant relationship between the serum HDL level and CVDs mortality (HR = 0.97; $95\%$ CI = 0.95–0.99; $$p \leq .02$$) (Table 3).
**TABLE 3**
| Variable | β | p‐value | Hazard ratio (95% CI) |
| --- | --- | --- | --- |
| Age (year) | 0.063 | .001 | 1.065 (1.02–1.1) |
| Gender | 1.24 | <.0001 | 3.46 (2.12–5.64) |
| Weight (Kg) | 0.030 | .129 | 1.030 (0.99–1.07) |
| BMI (kg/m2) | −0.084 | .16 | 0.91 (0.81–1.034) |
| WHR | 0.364 | .88 | 1.44 (0.01–1.97) |
| Physical activity (MET) | −0.027 | .17 | 0.12 (0.94–1.007) |
| DBP (mmHg) | −0.026 | .109 | 0.974(0.943–1.00) |
| SBP (mmHg) | 0.008 | .409 | 1.008(0.98–1.02) |
| Daily energy intake (kcal) | 0.00 | .24 | 1.00 (1.00–1.001) |
| History of HTN (%) | 1.53 | <.0001 | 4.62 (2.69–7.92) |
| alcohol consumption (%) | 0.28 | .78 | 1.32 (0.18–3.71) |
| Active smoking (%) | 0.58 | .041 | 1.78 (1.02–3.11) |
| FBS | 0.007 | <.0001 | 1.00 (1.00–1.01 |
| Cholesterol (mg/dl) | 0.005 | .241 | 1.00 (0.99–1.01) |
| TG (mg/dl) | 0.009 | .95 | 1.00 (0.75–1.35) |
| LDL (mg/dl) | 0.04 | .95 | 1.04 (0.23–4.59) |
| HDL (mg/dl) | −0.02 | .02 | 0.97 (0.95–0.99) |
Table 4 demonstrates the associations among baseline dietary acid load, different dietary indices scores (DII, AHEI, MDS, and DASH scores), and CVDs mortality. As shown in Table 4, there was no significant relationship between dietary indices and CVDs mortality in the main univariate Cox regression models. The confidence interval included 1.00 for NEAP score. However, the fully adjusted model (model 4) reports that there was a significant positive relationship between the baseline DII score and dietary acid load (NEAP score) level with CVDs mortality (HR = 1.11; $95\%$ CI = 1.01–1.24; $$p \leq .04$$ and HR = 1.02; $95\%$ CI = 1.01–1.03; p = <.0001, respectively). We did not observe any correlation among AHEI, MDS scores, and CVD mortality (Table 4).
**TABLE 4**
| Unnamed: 0 | β | p‐value | Hazard ratio (95% CI) |
| --- | --- | --- | --- |
| Model 1 | Model 1 | Model 1 | Model 1 |
| AHEI score | 0.00 | .80 | 1.03 (0.97–1.02) |
| MED score | 0.04 | .42 | 1.04 (0.93–1.16) |
| DASH score | 0.24 | .12 | 0.78 (0.57–1.07) |
| DII score | 0.07 | .15 | 1.07 (0.97–1.19) |
| NEAP | 0.01 | <.0001 | 1.01 (1.00–1.02) |
| Model 2 | Model 2 | Model 2 | Model 2 |
| AHEI score | 0.006 | .667 | 1.00 (0.98–1.03) |
| MED score | 0.054 | .313 | 1.05 (0.95–1.17) |
| DASH score | −0.233 | .15 | 0.79 (0.57–1.08) |
| DII score | 0.08 | .122 | 1.083 (0.97–1.19) |
| NEAP | 0.01 | .001 | 1.01 (1.00–1.01) |
| Model 3 | Model 3 | Model 3 | Model 3 |
| AHEI score | 0.008 | .527 | 1.00 (0.98–1.03) |
| MED score | 0.046 | .396 | 1.04 (0.94–1.16) |
| DASH score | −0.274 | .09 | 0.76 (0.55–1.04) |
| DII score | 0.115 | .035 | 1.12 (1.00–1.24) |
| NEAP | 0.013 | <.0001 | 1.01 (1.00–1.02) |
| Model 4 | Model 4 | Model 4 | Model 4 |
| AHEI score | 0.006 | .65 | 1.00 (0.98–1.03) |
| MED score | 0.07 | .19 | 1.07 (0.96–1.20) |
| DASH score | −0.22 | .16 | 0.79 (0.57–1.09) |
| DII score | 0.12 | .047 | 1.11 (1.01–1.24) |
| NEAP | 0.02 | <.0001 | 1.02 (1.01–1.03) |
## DISCUSSION
This retrospective cohort study was conducted to investigate the association between dietary indices and dietary acid load with CVD mortality in cardiovascular patients of Fasa PERSIAN cohort. The results of the current study showed that the CVD mortality rate increased significantly with age, being a man, and simultaneous onset of hypertension; there was also a significant positive relationship between tobacco use and mortality from cardiovascular diseases. However, it was found that with increasing serum HDL levels, the mortality rate of CVD decreased significantly. In addition, there was a direct and significant association between DII score and dietary acid load with CVD mortality; thus, with an increase in DII score and dietary acid load, the rate of CVD mortality increased by $11\%$ and $2\%$, respectively, while there was no significant correlation between AHEI and MDS scores with CVD mortality. The current study also revealed that adherence to the DASH diet in cardiovascular patients can decrease the risk of CVD mortality by $20.4\%$, but this decrease was not statistically significant although it is clinically significant.
In line with our study, Jibin et al. ( Tan et al., 2018) linked tobacco use and hypertension levels to mortality due to heart disease, stroke, and IHD. Nicotine and carbon monoxide available in tobacco promote the development of atherosclerosis by affecting myocardial oxygen capacity and increasing endothelial damage. Previous studies have also shown that tobacco use is closely linked to high hypertension and stroke. In addition, one of the main risk factors for heart disease is hypertension, and the severity of this disease can be controlled by reducing it (Tan et al., 2018; Wei et al., 1996).
This study also showed an inverse relationship between serum HDL levels and CVD mortality rates in patients. In this regard, Chantal et al. ( Kopecky et al., 2015) reported high levels of serum HDL as a protective factor against mortality and cardiovascular diseases in persons with diabetes. HDL plays a protective role against cardiovascular diseases and their mortality by clearing cholesterol from the macrophages and increasing endothelial function and antioxidant activities (Kopecky et al., 2015).
Mikkola et al. [ 2013] in their study, like the present study, showed that the risk of CVD mortality was higher in men compared to women. In this study, we observed higher rate of smoking and alcohol consumption along with lower levels of HDL in men compared to women, which are all risk factors for CVD and the resulting mortality. In addition, higher levels of estrogen in women compared to men is a protective factor against cardiovascular disorders (Mikkola et al., 2013).
Similar to our study, (Hodge et al., 2018; Shivappa et al., 2014) observed a direct association between CVD mortality and DII score. To explain this association, we can point to the relationship between high DII scores and increased risk of obesity, metabolic syndrome, and insulin resistance.(Garcia‐Arellano et al., 2015; Hébert et al., 2014; Ramallal et al., 2017; Shivappa et al., 2014). These chronic diseases are associated with inflammatory conditions in the body. Inflammatory biomarkers including hs‐CRP, IL‐6, IL‐4, IL‐1B, IL‐10, and TNF‐α have also been shown to be connected with obesity, diabetes, and CVD. In this regard, previous studies have linked increased DII scores to high levels of cytokines such as IL‐1, TNF‐α, and CRP. On the other hand, the guidelines of AHA published in 2019 have introduced obesity and diabetes as two risk factors related to CVD mortality (Arnett et al., 2019; Choi et al., 2013; Shivappa et al., 2014).
According to the AHA‐2019 guideline, energy, saturated fatty acids, cholesterol, transfatty acids, red meat, and refined grains in the diet not only increased the DII score but are also linked to increased mortality due to CVDs (Arnett et al., 2019). In contrast, the antiinflammatory compounds such as vitamin C, zinc, vitamin E, and beta‐carotene are associated with a decrease in DII score, and the antioxidant role of these compounds is a factor in the primary and secondary prevention of cardiovascular diseases (Hodge et al., 2018). In this regard, omega‐3 fatty acids and polyphenols (found in vegetables and fruits) are associated with a decrease in DII score and inflammation rate in the body. They are involved in regulating the body's inflammatory processes, improving lipid profile, oxidative stress, and endothelial function, and in this way, decreasing the risk of chronic disorders such as CVD disease. As a result, if dietary intake leads to inflammatory conditions, it can increase the risk of platelet aggregation and plaque formation at the endothelial cell, predisposing a patient to vascular damage and occlusion and increasing the risk of mortality (Ramallal et al., 2017).
The results of our study demonstrated that there was a significant positive association between dietary acid load and risk of CVD mortality, so that by increasing one score in dietary acid load, the rate of CVD mortality increased by $2\%$. In line with the present study, Shamima et al. ( Akter et al., 2017) detected a strong relationship among PRAL, NEAP scores, and CVD mortality rates. Minseon et al. ( Park et al., 2015) also linked an increased dietary acid load and a higher risk of CVD and all‐cause mortality. Previous research have shown that an increase in consumption of meat and its products, eggs, cheese, refined grains, and fish, and a decrease in consumption of vegetables and fruits lead to an increase in the dietary acid load and risk of chronic disorders, including cardiovascular diseases, hypertension, type 2 diabetes, and their mortality due to these diseases (Akter et al., 2017; Park et al., 2015.) In contrast the consumption of potassium bicarbonate, magnesium, fiber, vitamin C, calcium, and phytochemicals, which exist abundantly in fruits and vegetables (as part of a healthy diet); are probably associated with decreased acid load of diet and a lower risk of CVD disorders. ( Akter et al., 2017; Fatahi & Azadbakht, 2019; Han et al., 2016) Western dietary pattern contains abundant cheese and meat (acidogenic foods) and is deficient in vegetables and fruits (alkalizing foods) which increase the acid load of the dietary intake. ( Kahleova et al., 2021) In addition, previous studies have also shown that a vegan dietary pattern is related to reducing dietary acid load. ( Kahleova et al., 2021; Müller et al., 2021) Partial examination shows that a rise in acid load in diet, which is associated with an increase in blood acid levels, leads to the excretion of sodium in the body and an increase in cortisol secretion and insulin resistance. In this regard, the serum level of potassium reduces due to increased potassium excretion, which leads to an increase in hypertension by affecting vasodilation. ( Adrogué & Madias, 2007) On the other hand, increased cortisol levels are related to metabolic syndrome and increased cardiovascular diseases and mortality risk. ( Carnauba et al., 2017; Hur et al., 2015) Insulin resistance is also caused by a reduced tendency for insulin to bind to its receptor in an acidic environment, which is directly related to the risk of CVD and all‐cause mortality. ( Ormazabal et al., 2018; Zhang et al., 2017) Phillips et al. [ 2019] and Fung et al. [ 2008] showed that when adherence to the DASH diet increases, the risk of cardiovascular diseases and mortality decreases. This decrease is due to the consumption of more vegetables, fruits, nuts and seeds, whole grains, and low‐fat dairy products and restrictions on the consumption of red meat, sugars, sweets, beverages, and total and saturated fat. The mechanism of impact of this diet can be attributed to its compounds such as potassium, sodium, magnesium, calcium, fiber, and antioxidants. In this diet, high fiber intake leads to reduction in the levels of LDL, cholesterol, triglycerides, hypertension, CRP, and the risk of obesity and overweight, and improves insulin sensitivity and endothelial function (Levitan et al., 2013). DASH diet with emphasis on magnesium and potassium sources such as dark‐green leafy vegetables, seeds, nuts, and whole grains can play a role in controlling hypertension and stroke. Magnesium and potassium reduce inflammatory cytokines and increase nitric oxide levels, thereby affecting the severity of heart diseases and mortality due to them (Dokken, 2008; Rifai et al., 2015). According to our results, it was observed that adherence to the DASH diet in cardiovascular patients could reduce the risk of CVD mortality by $20.4\%$; however, this decrease was not statistically significant. In line with our study, (Aigner et al., 2018) did not find a strong relationship between DASH score and mortality from stroke.
In the present study, we did not find a significant correlation between AHEI score and CVD‐related deaths. Contrary to our study, Emily et al. ( Hu et al., 2020) reported that higher AHEI scores led to lower risk of cardiovascular diseases and mortality by $16\%$ and $34\%$, respectively. The AHEI score can assess the quality of the diet and predict the risk of death due to chronic diseases. The index measures the consumption of whole grains, fruits, vegetables, fish, dairy, processed red meat, and alcohol. Also, (Akbaraly et al., 2011) in their study found that higher AHEI score had a significant inverse relationship with CVD mortality, while similar to the current study, no significant association was observed between this score and cancer‐related mortality. The most important reason for the inconsistency of the results of our study with other studies can be the short follow‐up time.
In their study over 4 years, Emily et al. ( Levitan et al., 2013) recorded 1385 mortality cases from HF. In their study, similar to ours, there was no statistically significant relationship between MDS score and mortality. While (Dokken, 2008; Hodge et al., 2018) found that the more the adherence to the Mediterranean diet, the lower the risk of mortality at a young age. The Mediterranean diet improves systolic hypertension and lowers CRP, fibrinogen, oxidative stress, and serum cholesterol, thereby reducing the incidence of chronic disorders and their mortality.
One of the main weaknesses of this study was the short follow‐up time. Also, in this study, recall diet was not examined for nutritional assessment and only FFQ was used. Another weakness of this study was the recording of nutritional data in the baseline state and no other dietary record was recorded during the follow‐up period; as a result, we were unable to assess the patients' dietary changes. Using self‐reporting for food evaluation can also increase overestimation and underestimation. In this study, we failed to assess important factors such as disease severity, type of treatment during illness, and evaluation of health care that are associated with mortality. One of the strengths of this study is the study of dietary indices instead of evaluating specific micronutrients in the controls and inhibition of cardiovascular disease and its mortality. Also, in this study, four important indices were examined simultaneously, and this is of great importance. To summarize and comment definitively in this regard, it is suggested that studies with more follow‐up years and the use of cohort studies should be carried out in other regions simultaneously.
## CONCLUSION
The current study indicated that increasing dietary acid load and the use of inflammatory food compounds increase the risk of CVD mortality. Adherence to the DASH diet can also be associated with decreased risk of mortality due to cardiovascular diseases in patients. Therefore, paying attention to the dietary acid load and acid balance along with the use of diets with antiinflammatory properties and focusing on the DASH diet in CVD patients may be effective in preventing CVD mortality. However, further studies are needed in this regard.
## CONFLICT OF INTEREST
None declare.
## DATA AVAILABILITY STATEMENT
The data supporting the finding of this study are available from the corresponding author upon reasonable request.
## References
1. Adrogué H. J., Madias N. E.. **Sodium and potassium in the pathogenesis of hypertension**. *New England Journal of Medicine* (2007) **356** 1966-1978. PMID: 17494929
2. Aigner A., Becher H., Jacobs S., Wilkens L. R., Boushey C. J., Le Marchand L., Haiman C. A., Maskarinec G.. **Low diet quality and the risk of stroke mortality: The multiethnic cohort study**. *European Journal of Clinical Nutrition* (2018) **72** 1035-1045. PMID: 29426930
3. Akbaraly T. N., Ferrie J. E., Berr C., Brunner E. J., Head J., Marmot M. G., Singh‐Manoux A., Ritchie K., Shipley M. J., Kivimaki M.. **Alternative healthy eating index and mortality over 18 y of follow‐up: Results from the Whitehall II cohort**. *The American Journal of Clinical Nutrition* (2011) **94** 247-253. PMID: 21613557
4. Akter S., Nanri A., Mizoue T., Noda M., Sawada N., Sasazuki S., Tsugane S.. **Dietary acid load and mortality among Japanese men and women: The Japan public health center–based prospective study**. *The American Journal of Clinical Nutrition* (2017) **106** 146-154. PMID: 28539378
5. Arnett D. K., Blumenthal R. S., Albert M. A., Buroker A. B., Goldberger Z. D., Hahn E. J., Himmelfarb C. D., Khera A., Lloyd‐Jones D., McEvoy J. W., Michos E. D.. **2019 ACC/AHA guideline on the primary prevention of cardiovascular disease: A report of the American College of Cardiology/American Heart Association task force on clinical practice guidelines**. *Circulation* (2019) **74** e177-e232
6. Carnauba R. A., Baptistella A. B., Paschoal V., Hübscher G. H.. **Diet‐induced low‐grade metabolic acidosis and clinical outcomes: A review**. *Nutrients* (2017) **9** 538. PMID: 28587067
7. Choi J., Joseph L., Pilote L.. **Obesity and C‐reactive protein in various populations: a systematic review and meta‐analysis**. *Obesity Reviews* (2013) **14** 232-244. PMID: 23171381
8. Dokken B. B.. **The pathophysiology of cardiovascular disease and diabetes: Beyond blood pressure and lipids**. *Diabetes Spectrum: A Publication of the American Diabetes Association* (2008) **21** 160-165
9. Emamian M. H., Hashemi H., Fotouhi A.. **Predicted 10‐year risk of cardiovascular disease in the Islamic Republic of Iran and the body mass index paradox**. *Eastern Mediterranean Health Journal* (2020) **26** 1465-1472. PMID: 33355385
10. Farjam M., Bahrami H., Bahramali E., Jamshidi J., Askari A., Zakeri H., Homayounfar R., Poustchi H., Malekzadeh R.. **A cohort study protocol to analyze the predisposing factors to common chronic non‐communicable diseases in rural areas: Fasa Cohort Study**. *BMC Public Health* (2016) **16** 1-8. PMID: 26728978
11. Fatahi S., Azadbakht L.. **Association between dietary acid load with alternative Mediterranean diet and dietary approaches to stop hypertension among Tehranian women**. *Journal of Advanced Biomedical Sciences* (2019) **8** 1036-1045
12. Fung T. T., Chiuve S. E., McCullough M. L., Rexrode K. M., Logroscino G., Hu F. B.. **Adherence to a DASH‐style diet and risk of coronary heart disease and stroke in women**. *Archives of Internal Medicine* (2008) **168** 713-720. PMID: 18413553
13. Garcia‐Arellano A., Ramallal R., Ruiz‐Canela M., Salas‐Salvadó J., Corella D., Shivappa N., Schröder H., Hébert J., Ros E., Gómez‐Garcia E., Estruch R., Lapetra J., Arós F., Fiol M., Serra‐Majem L., Pintó X., Babio N., González J., Fitó M., Martínez‐González M.. **Dietary inflammatory index and incidence of cardiovascular disease in the PREDIMED study**. *Nutrients* (2015) **7** 4124-4138. PMID: 26035241
14. Han E., Kim G., Hong N., Lee Y.‐H., Kim D. W., Shin H. J., Lee B. W., Kang E. S., Lee I. K., Cha B. S.. **Association between dietary acid load and the risk of cardiovascular disease: Nationwide surveys (KNHANES 2008–2011)**. *Cardiovascular Diabetology* (2016) **15** 1-14. PMID: 26739706
15. Hébert J. R., Shivappa N., Tabung F. K., Steck S. E., Wirth M. D., Hurley T. G.. **On the use of the dietary inflammatory index in relation to low‐grade inflammation and markers of glucose metabolism in the Cohort study on Diabetes and Atherosclerosis Maastricht (CODAM) and the Hoorn study**. *The American Journal of Clinical Nutrition* (2014) **99** 1520
16. Hodge A. M., Bassett J. K., Dugué P.‐A., Shivappa N., Hébert J. R., Milne R., English D. R., Giles G. G.. **Dietary inflammatory index or Mediterranean diet score as risk factors for total and cardiovascular mortality**. *Nutrition, Metabolism, and Cardiovascular Diseases* (2018) **28** 461-469
17. Hu E. A., Steffen L. M., Coresh J., Appel L. J., Rebholz C. M.. **Adherence to the healthy eating index–2015 and other dietary patterns may reduce risk of cardiovascular disease, cardiovascular mortality, and all‐cause mortality**. *The Journal of Nutrition* (2020) **150** 312-321. PMID: 31529069
18. Hur K. Y., Kim J. H., Kim B. J., Kim M.‐S., Lee E. J., Kim S. W.. **Clinical guidelines for the diagnosis and treatment of Cushing's disease in Korea**. *Endocrinology and Metabolism* (2015) **30** 7. PMID: 25827452
19. Kahleova H., McCann J., Alwarith J., Rembert E., Tura A., Holubkov R., Barnard N. D.. **A plant‐based diet in overweight adults in a 16‐week randomized clinical trial: The role of dietary acid load**. *Clinical Nutrition ESPEN* (2021) **44** 150-158. PMID: 34330460
20. Kopecky C., Genser B., Drechsler C., Krane V., Kaltenecker C. C., Hengstschläger M., März W., Wanner C., Säemann M. D., Weichhart T.. **Quantification of HDL proteins, cardiac events, and mortality in patients with type 2 diabetes on hemodialysis**. *Clinical Journal of the American Society of Nephrology* (2015) **10** 224-231. PMID: 25424990
21. Levitan E. B., Lewis C. E., Tinker L. F., Eaton C. B., Ahmed A., Manson J. E., Snetselaar L. G., Martin L. W., Trevisan M., Howard B. V., Shikany J. M.. **Mediterranean and DASH diet scores and mortality in women with heart failure: The Women's Health Initiative**. *Circulation: Heart Failure* (2013) **6** 1116-1123. PMID: 24107587
22. McCullough M. L., Feskanich D., Stampfer M. J., Giovannucci E. L., Rimm E. B., Hu F. B., Spiegelman D., Hunter D. J., Colditz G. A., Willett W. C.. **Diet quality and major chronic disease risk in men and women: Moving toward improved dietary guidance**. *The American Journal of Clinical Nutrition* (2002) **76** 1261-1271. PMID: 12450892
23. Mikkola T. S., Gissler M., Merikukka M., Tuomikoski P., Ylikorkala O.. **Sex differences in age‐related cardiovascular mortality**. *PLoS One* (2013) **8**. PMID: 23700418
24. Miller P. E., Cross A. J., Subar A. F., Krebs‐Smith S. M., Park Y., Powell‐Wiley T., Hollenbeck A., Reedy J.. **Comparison of 4 established DASH diet indexes: Examining associations of index scores and colorectal cancer**. *The American Journal of Clinical Nutrition* (2013) **98** 794-803. PMID: 23864539
25. Müller A., Zimmermann‐Klemd A. M., Lederer A.‐K., Hannibal L., Kowarschik S., Huber R., Storz M. A.. **A vegan diet is associated with a significant reduction in dietary acid load: Post hoc analysis of a randomized controlled trial in healthy individuals**. *International Journal of Environmental Research and Public Health* (2021) **18** 9998. PMID: 34639299
26. Ormazabal V., Nair S., Elfeky O., Aguayo C., Salomon C., Zuñiga F. A.. **Association between insulin resistance and the development of cardiovascular disease**. *Cardiovascular Diabetology* (2018) **17** 1-14. PMID: 29301528
27. Park M., Jung S. J., Yoon S., Yun J. M., Yoon H. J.. **Association between the markers of metabolic acid load and higher all‐cause and cardiovascular mortality in a general population with preserved renal function**. *Hypertension Research* (2015) **38** 433-438. PMID: 25762414
28. Phillips C. M., Harrington J. M., Perry I. J.. **Relationship between dietary quality, determined by DASH score, and cardiometabolic health biomarkers: A cross‐sectional analysis in adults**. *Clinical Nutrition* (2019) **38** 1620-1628. PMID: 30219609
29. Ramallal R., Toledo E., Martínez J. A., Shivappa N., Hébert J. R., Martínez‐González M. A., Ruiz‐Canela M.. **Inflammatory potential of diet, weight gain, and incidence of overweight/obesity: The SUN cohort**. *Obesity* (2017) **25** 997-1005. PMID: 28544794
30. Rifai L., Pisano C., Hayden J., Sulo S., Silver M. A.. *Baylor University medical center proceedings* (2015)
31. Sarrafzadegan N., Mohammmadifard N.. **Cardiovascular disease in Iran in the last 40 years: prevalence, mortality, morbidity, challenges and strategies for cardiovascular prevention**. *Archives of Iranian Medicine* (2019) **22** 204-210. PMID: 31126179
32. Schwingshackl L., Bogensberger B., Hoffmann G. J.. **Diet quality as assessed by the healthy eating index, alternate healthy eating index, dietary approaches to stop hypertension score, and health outcomes: An updated systematic review and meta‐analysis of cohort studies**. *Journal of the Academy of Nutrition and Dietetics* (2018) **118** 74-100. PMID: 29111090
33. Shivappa N., Steck S. E., Hurley T. G., Hussey J. R., Hébert J. R.. **Designing and developing a literature‐derived, population‐based dietary inflammatory index**. *Public Health Nutrition* (2014) **17** 1689-1696. PMID: 23941862
34. Sofi F., Macchi C., Abbate R., Gensini G. F., Casini A.. **Mediterranean diet and health status: An updated meta‐analysis and a proposal for a literature‐based adherence score**. *Public Health Nutrition* (2014) **17** 2769-2782. PMID: 24476641
35. Tan J., Zhang X., Wang W., Yin P., Guo X., Zhou M.. **Smoking, blood pressure, and cardiovascular disease mortality in a large cohort of Chinese men with 15 years follow‐up**. *International Journal of Environmental Research and Public Health* (2018) **15** 1026. PMID: 29783678
36. Wei M., Mitchell B. D., Haffner S. M., Stem M. P.. **Effects of cigarette smoking, diabetes, high cholesterol, and hypertension on all‐cause mortality and cardiovascular disease mortality in Mexican Americans: The San Antonio Heart Study**. *American Journal of Epidemiology* (1996) **144** 1058-1065. PMID: 8942437
37. Willett W. C., Sampson L., Stampfer M. J., Rosner B., Bain C., Witschi J., Hennekens C. H., Speizer F. E.. **Reproducibility and validity of a semiquantitative food frequency questionnaire**. *American Journal of Epidemiology* (1985) **122** 51-65. PMID: 4014201
38. Zhang X., Li J., Zheng S., Luo Q., Zhou C., Wang C.. **Fasting insulin, insulin resistance, and risk of cardiovascular or all‐cause mortality in non‐diabetic adults: A meta‐analysis**. *Bioscience Reports* (2017) **37**. PMID: 28811358
|
---
title: GC–MS analysis and pharmacological evaluations of Phoenix sylvestris (Roxb.)
seeds provide new insights into the management of oxidative stress and hyperglycemia
authors:
- Md. Shafiullah Shajib
- Shanta Islam
- Safaet Alam
- Ridwan Bin Rashid
- Mirola Afroze
- Mala Khan
- Bidyut Kanti Datta
- Lutfun Nahar
- Satyajit Dey Sarker
- Mohammad A. Rashid
journal: Food Science & Nutrition
year: 2022
pmcid: PMC10002931
doi: 10.1002/fsn3.3196
license: CC BY 4.0
---
# GC–MS analysis and pharmacological evaluations of Phoenix sylvestris (Roxb.) seeds provide new insights into the management of oxidative stress and hyperglycemia
## Abstract
Phoenix sylvestris Roxb. ( Arecaceae) seeds are used in the treatment of diabetes in the traditional system of medicine. The present study evaluated antihyperglycemic and antioxidant activities as well as the total phenolic and flavonoid content of the methanol extract of P. sylvestris seeds (MEPS). The constituents of the extract were identified by GC–MS analysis. MEPS demonstrated strong antioxidant activity against 2,2‐diphenyl‐1‐picrylhydrazyl (DPPH) (IC50 = 162.70 ± 14.99 μg) and nitric oxide (NO) (IC50 = 101.56 ± 9.46 μg/ml) free radicals. It also possesses a substantial amount of phenolics and flavonoids. It significantly ($p \leq .05$) reduced blood glucose levels in glucose‐loaded and alloxan‐induced diabetic mice at the doses of 150 and 300 mg/kg b.w., respectively. A total of 46 compounds were detected and identified by gas chromatography–mass spectroscopy (GC–MS) analysis, among which 8‐methylisoquinoline N‐oxide ($32.82\%$) was predominant. The phytochemical study by GC–MS revealed that the MEPS possesses compounds which could be related to its antidiabetic and antioxidant activities. To recapitulate, P. sylvestris seeds can be a very good option for antidiabetic and antioxidant activity though further studies are still recommended to figure out the responsible phytochemicals and establish their exact mechanism of action.
Phoenix sylvestris seeds are used in the treatment of diabetes in traditional system of medicine. The present study evaluated antihyperglycemic and antioxidant activities as well as total phenolic and flavonoid content of methanol extract of P. sylvestris seeds. ( MEPS). The constituents of the extract were also identified by GC–MS analysis.
## INTRODUCTION
Diabetes mellitus is the metabolic syndrome of the human body manifested by chronic hyperglycemia along with impaired metabolism of carbohydrates, protein, and fats due to diminished insulin secretion and/or action (Alam et al., 2022; Nayak & Roberts, 2006). Chronic hyperglycemia exacerbates the antioxidant action by increasing oxidative stress and reactive oxygen species (ROS) in islets of the pancreas (Savu et al., 2012). Furthermore, it has been reported that diabetes is responsible for the excess generation of free radicals due to the reduction of antioxidant levels in the body (Ali & Agha, 2009). Multiple antihyperglycemic agents along with insulin are currently available in the market, but they are not devoid of significant undesirable side effects (Pari & Saravanan, 2004). Recently, the use of plants and plant materials has attracted the attention of researchers for the development of new antihyperglycemic due to their promising efficacy and limited toxicity (Rates, 2001). In addition, antioxidants derived from plants have been shown to play important roles in improving diabetes‐associated disorders (Rahimi et al., 2005).
Phoenix slylvestris (L.) Roxb., a plant of the palm family Arecaceae, is commonly known as “Khejur” in Bangladesh. The plant seeds have been reported to be bacteriostatic against Gram‐positive and Gram‐negative organisms (Kothari, 2011). They are used in the treatment of dysentery, ague, and diabetes in the traditional medicine system (Beg & Singh, 2015; Ghani, 1998). Although traditional use advocates the use of P. sylvestris as a candidate for treating diabetes, no scientific report exists to corroborate this claim. Therefore, the present study aimed to determine the antioxidant action, total phenolic and flavonoid contents, and antihyperglycemic activity of seeds of P. sylvestris for the first time. The constituents of seed extract have also been identified by gas chromatography‐mass spectroscopic (GC–MS) analysis so that future researchers can find a nifty clue to identify responsible phytochemicals from the plant seeds to discover and develop novel therapeutics against diabetes and oxidative stress.
## Plant materials and extraction
The fully matured fruits of P. sylvestris were collected from Akabpur, Mainamati, Comilla, Bangladesh in July 2013. The fruits were identified by the authorities of Bangladesh National Herbarium, Mirpur, Dhaka, Bangladesh, and a voucher specimen has been deposited (accession no: DACB: 38499) for future reference. The seeds of P. sylvestris were separated from the fruits, dried, and ground to a coarse powder using a mechanical grinder. About 500 g of powdered seeds was mixed with 1200 ml of methanol (MeOH). The mixture was occasionally stirred and kept at 25 ± 2°C for 72 h. The extract was then filtered through the Whatman filter paper, number 41. The solvent was removed by using a rotary evaporator under reduced pressure at 40°C temperature and 50 rpm. Finally, 12.4 g ($2.48\%$ yield) concentrated extract was obtained, which was used for phytochemical and biological studies.
## Chemicals and drugs
Chemicals and reagents used in this study were ‐ MeOH, 1,1‐diphenyl‐2‐picrylhydrazyl (DPPH), Griess reagent, quercetin, gallic acid, ascorbic acid, pentobarbital sodium (Sigma Co.), sodium carbonate (Na2CO3), Na‐K tartrate, aluminum chloride (AlCl3), Folin–Ciocalteu's reagent (Merck Co.), alloxan monohydrate (Loba Chemie Pvt. Ltd.). Metformin hydrochloride was obtained as a gift sample from Square Pharmaceuticals Ltd.
## Ethical statements
The protocols for the current study were endorsed by the Ethics Committee of Stamford University Bangladesh (SUB/IAEC/13.05). The animals were treated according to the guidelines provided by The Swiss Academy of Medical Sciences and Swiss Academy of Sciences. After the experiments, animals were euthanized using pentobarbital sodium following the AVMA Guiding Principles for the Euthanasia of animals: 2013 edition. Necessary steps were taken to minimize animal suffering.
## Preliminary screening
MEPS was qualitatively screened for the detection of carbohydrates, reducing sugars, steroids, alkaloids, proteins, saponins, tannins, and flavonoids following the standard procedures (Ghani, 1998).
## GC–MS (gas chromatography–mass spectroscopy) analysis
GC–MS analysis of the MeOH extract of P. sylvestris seeds was performed using Agilent 7890A (Agilent Technologies) capillary gas chromatograph interfaced to a 5975C inert XL EI/CI triple‐axis mass detector. The gas chromatograph was equipped with an HP‐5MSI fused capillary column of $5\%$ phenyl, $95\%$ dimethyl‐poly‐siloxane (film: 0.25 μm, length: 90 m, and diameter: 0.250 mm). The parameters of GC were programmed as follows: inlet temperature: 250°C; oven temperature; 90°C at 0 min raised to 200°C for 2 min (3°C/min) then 280°C for 2 min (15°C/min); carrier gas (Helium) flow rate: 1.1 ml/min; auxiliary temperature: 280°C. Total retention time for the chromatographic analysis was 46 min. The MS parameters were set as follows: quad temperature: 150°C; source temperature: 230°C; mode: scan mode; mass range: 50–550 m/z. The “NIST‐MS Library” was used for mass spectra analysis and identification of compounds. The relative percentage of separated compounds was determined from the peak areas of the total ionic chromatogram.
## Determination of total phenolic content (TPC)
The total phenolics present in the MEPS were quantified using Folin–Ciocalteu's reagent (Singleton et al., 1999). An aliquot (0.5 ml) of Folin–Ciocalteu's reagent was taken and mixed with 1 ml of (200 μg/ mL) MEPS. After 5 min, 4 ml of $7.5\%$ (w/v) Na2CO3 prepared in distilled water was added to the mixture. The solution was mixed well and incubated at 20°C for 1 h. The absorbance was measured at 765 nm using DR 5000™ (Hach) spectrophotometer. A calibration curve ($y = 0.0086$x + 0.2546, R 2 = 0.9998) of gallic acid was prepared using solutions of varying concentrations ranging from 25 to 400 mg/L. Then, the amount of total phenolics present in the extract was measured in gallic acid equivalents (GAE) using the formula: A = (C × V)/m, where, A is the total amount of phenolics equivalent to gallic acid present in the extract, C is the concentration of gallic acid (mg/ml) measured from the calibration curve, V is the extract volume (ml) and m denotes extract weight (g). The process was conducted in triplicate, and the mean value of TPC was determined.
## Determination of total flavonoid content (TFC)
A solution (1 ml) of extract (200 μg/ml) was taken in a test tube, and 2 ml of MeOH was added to it. Then the solution was mixed well with 0.1 ml of $10\%$ of aluminum chloride (w/v, prepared in distilled water) followed by 1 M of Na‐K tartrate, 2.8 ml of distilled water, and incubated at 25°C. After 30 min, the absorbance of the mixture was measured at 415 nm (Selim et al., 2014). The calibration curve of quercetin ($y = 0.0178$x + 0.6152, R 2 = 0.9975) was prepared by measuring the absorbance of its different concentrations (25–400 mg/L). Then, the total flavonoid content of the extract was calculated using the standard calibration curve and expressed as mg of flavonoid present per gm of extract equivalent to quercetin. The experiment was conducted three times, and the mean value of flavonoid content was calculated.
## DPPH free radical scavenging capacity assay
The effect of MEPS on free radicals was determined by analyzing its scavenging effect on stable 1,1‐diphenyl‐2‐picrylhydrazyl (DPPH) free radicals. The plant extract or standard drug (ascorbic acid) was prepared at a concentration ranging from 400 to 1.5625 μg/ml in MeOH. A 0.1 mM solution of DPPH in MeOH was prepared, and 2 ml of this solution was added to 2 ml of the test solution. The mixture was mixed properly and incubated for 30 min at room temperature in a dark place. The absorbances for the standard and experimental solutions were measured against blank (without test sample or drug) DPPH solution using a spectrophotometer at 517 nm (Wang et al., 2013). The scavenging of DPPH free radicals was expressed as a percentage of inhibition was determined from the following equation: %inhibition=absorbance of blank−absorbance of test sampleabsorbance of blank×100 then, IC50 value was calculated from % inhibition vs log concentration curve.
## Nitric oxide (NO) scavenging capacity assay
Exactly 4 ml of MEPS or standard (ascorbic acid) solution at the concentration 400–1.5625 μg/ml in methanol was taken in different test tubes. Then, 1.0 ml of sodium nitroprusside (5 mM) was added to the samples and incubated for 2 h at 30°C. After incubation, 2 ml solution was taken, and 1.2 ml Griess reagent ($1\%$ sulfanilamide, $0.1\%$ napthylene diamine dihydrochloride in $2\%$ H3PO4) was added to it. The absorbances for the standard and test solutions were measured against blank using a spectrophotometer at 550 nm (Alisi & Onyeze, 2008). The percentage of inhibition was calculated as described earlier in DPPH free radical scavenging assay, and the IC50 value was calculated.
## Study animals
Swiss albino mice of either sex, weighing 25–30 g, 6–8 weeks, were used for the antihyperglycemic study. They were procured from the International Centre for Diarrhoeal Disease Research, Bangladesh (ICDDR, B) and housed in appropriate cages with wood flakes bedding. The mice were allowed to acclimatize for 2 weeks in standard laboratory conditions and were maintained at 25°C ± 2°C temperature, $55\%$–$60\%$ relative humidity, and 12 h light/dark cycle. They had access to water and feed ad libitum. The feed was formulated by authorities of ICCDR,B. The animals were randomly divided into five groups (normal control, diabetic control, and three experimental groups), each group consisting of five mice ($$n = 5$$). The normal and diabetic control groups received oral treatment of vehicle (physiological saline). The positive control and experimental groups were orally treated (p.o) with metformin and MEPS, respectively. The experimental mice starved from feed for 12 h but had free access to water before experiments. The tests were performed between 9.00 a.m. and 5.00 p.m., and the investigators had no information about the experimental groups.
## Acute toxicity test
The acute toxic effect of MEPS on animals was assessed before studying the antihyperglycemic activity. Experimental animals were divided into four experimental and one control group ($$n = 5$$). The experimental animals were orally treated with MEPS at the doses of 500, 1000, 2000, and 3000 mg/kg b.w. Control group animals received physiological saline only. Animals were housed and adequately provided with ICCDR,B formulated food and water ad libitum. They were carefully observed for 72 h after administration of MEPS, and any adverse reactions (skin rashes, swelling, itching), behavioral changes, and mortality were documented (Walker et al., 2008).
## Oral glucose tolerance test (OGTT)
The mice of control, diabetic control, standard drug treatment (positive control), and MEPS treatments (experimental groups) fasted overnight. The blood samples of each animal were collected from the tail vein, and glucose level was measured using Accu‐Chek® (Roche) one‐touch glucometer as baseline (0 min). Then animals of diabetic control, positive control, and experimental groups received vehicle (10 ml/kg b.w.), metformin (60 mg/kg b.w.), and MEPS (50, 150, 300 mg/kg b.w.), respectively. Primarily, antihyperglycemic activities were evaluated with the lower doses (50 mg/kg b.w.) of MEPS and the dose was randomly selected based on observing the effect of P. sylvestris fruits in the previous study (Shajib et al., 2015). The higher dose limit (300 mg/kg b.w.) was selected based on the significant glucose‐lowering effect of MEPS. After 30 min, each group of mice received $10\%$ glucose solution at the dose of 2 gm/kg b.w. Then, blood glucose level was measured at 30, 60, 90, and 120 min following glucose treatment (Chaturvedi et al., 2004).
## Assay for alloxan‐induced diabetes
The experimental mice were randomly divided into control, diabetic control, standard drug treatment (positive control), and MEPS treatments (experimental groups). Positive control and experimental group animals were induced with diabetes by intraperitoneal (i.p.) injection of alloxan‐monohydrate at the dose of 60 mg/kg b.w. The blood glucose level was measured before alloxan treatment. The glucose levels were monitored every day after alloxan treatment. Alloxan induces type 1 or insulin‐dependent diabetes (Macdonald Ighodaro et al., 2017). In fasting conditions, blood glucose level of more than 7 mmol/L is indicative of diabetes (Adeyi et al., 2015; Mathew & Tadi, 2021; Njogu et al., 2016). After 3 days of alloxan administration, fasted mice with blood sugar levels ≥8 mmol/L were considered diabetic (Ezeja et al., 2015). The sustained hyperglycemia of the alloxan‐induced diabetic mice was observed for the next 5 days and selected for the study. Alloxan may increase blood glucose levels by more than 11 mmol/L in consecutive days after administration (Macdonald Ighodaro et al., 2017; Njogu et al., 2016). However, the time required to reach the blood glucose level can vary on the alloxan administration route, dose, and experimental animal species (Hansen et al., 2007; Kim et al., 2006; Lips et al., 1988; Njogu et al., 2016). The hyperglycemic mice received vehicle (10 ml/kg b.w.), metformin (60 mg/kg b.w.), or MEPS (50, 150, and 300 mg/kg b.w.). Blood samples were collected from the tail vein of each group of mice, and glucose level was measured at 0 h (as baseline), 4, 8, and 24 h following treatments (Semwal et al., 2010).
## Statistical analysis
All the experimental data were presented as mean ± SEM (standard error of the mean). IC50 values were determined by utilizing GraphPad Prism 6.01 (GraphPad Software, Inc.). The comparison of different groups against the control group was performed by one‐way analysis of variance (ANOVA) followed by Dunnett's test as the post hoc test using SPSS 22 (IBM) software. $p \leq .05$ was set as the level of statistical significance.
## Phytochemical analysis
Preliminary screening for different phytochemical groups reveals that the plant seed contained alkaloids, steroids, carbohydrates, proteins, flavonoids, and tannins. The most abundant compound revealed by the GC–MS analysis of the extract was 8‐methylisoquinoline N‐oxide ($32.82\%$). Other major constituents were as follows: methyl oleate ($12.19\%$), methyl linoleate ($7.44\%$), dodecanoic acid, methyl ester ($5.59\%$), palmitic acid, methyl ester ($4.62\%$), 9‐octadecenoic acid (Z)‐,2,3‐dihydroxypropyl ester ($3.11\%$), 5,8‐dimethyl‐1,4‐dihydro‐1,4‐methanonaphthalene ($2.93\%$), tetradecanoic acid, methyl ester ($2.88\%$), alpha‐bisabolol ($2.41\%$),linalool ($1.69\%$), (+)‐(4 S, 8R)‐8‐epi‐beta‐bisabolol ($1.59\%$), 1‐fluoro‐4‐acetylbenzene ($1.56\%$), methyl stearate ($1.41\%$), 11‐eicosenoic acid, methyl ester ($1.39\%$), and alpha‐bisabolol oxide B ($1.15\%$). The identified compounds, peak area (%), and retention time (min) of MEPS by GC–MS analysis are presented in Table 1. The total ionic chromatograph of the methanol extract of P. sylvestris seed is shown in Figure 1.
## Antioxidant activity
Quantitative analysis of the crude extract demonstrated that there are 91.32 ± 5.20 mg total phenolics equivalent to gallic acid and 21.99 ± 4.70 mg total flavonoids equivalent to quercetin present in per gram extract. The anti‐radical activity of MEPS against DPPH and NO was found to have IC50 values of 162.70 ± 14.99 and 101.56 ± 9.46 μg/ml, respectively. Standard drug ascorbic acid demonstrated IC50 values of 8.71 ± 0.02 and 7.39 ± 0.43 μg/ml, respectively. The highest percent inhibition of DPPH radical exhibited by MEPS was 61.67 ± 1.74 at the maximum experimental concentration (400 μg/ml). Ascorbic acid inhibited DPPH radical by 96.41 ± $0.00\%$ (Figure 2). MEPS and ascorbic acid displayed a maximum of 68.20 ± 1.00 and 96.78 ± $0.38\%$ nitric oxide (NO) scavenging activity at higher concentrations, respectively (Figure 3). The results show that MEPS is capable of arresting the free radicals generated by DPPH and NO, which are harmful to human health (Hasan et al., 2009). It has been reported that plant phenolics and flavonoids may exert significant antioxidant activities (Rice‐Evans et al., 1997; Saija et al., 1995). The presence of a considerable amount of phenolics and flavonoids in MEPS can be attributed to its strong antioxidant activity.
**FIGURE 2:** *DPPH (2,2‐diphenyl‐1‐picrylhydrazyl) free radicals scavenging activity of ascorbic acid (standard) and MEPS* **FIGURE 3:** *Nitric oxide (NO) free radicals scavenging activity of ascorbic acid (standard) and MEPS*
## Acute toxicity
Oral administration of MEPS up to 3000 mg/kg did not cause any adverse reactions, behavioral changes, or mortality during the observational period. This suggests that MEPS possesses a low toxicity profile (LD50 > 3000 mg/kg b.w.). The doses of the MEPS for antihyperglycemic studies were selected from trial experiments. The observations from the acute toxicity study indicate that the experimental doses of MEPS selected for the study were safe.
## Oral glucose tolerance
Oral glucose tolerance test (OGTT) measures the ability to utilize sugars by the body and is commonly performed to evaluate pre‐diabetes, post‐diabetes, and gestational diabetes (Hartling et al., 2012; Ziegler et al., 2009). The additional glucose load causes the excess plasma glucose level, characterized as hyperglycemia and early clinical manifestation of diabetes. Fasted mice showed glucose levels below 5.5 mmol/L, which was in the normal range (Andrikopoulos et al., 2008). After 30 min of oral glucose treatment, the plasma glucose level was significantly increased in mice and then gradually declined throughout the observation period. Oral treatment of MEPS and the standard drug metformin caused a marked reduction of the elevated blood glucose level in OGTT (Figure 4). The result was significant over the observation period (30–120 min) for both metformin (60 mg/kg b.w.) and MEPS at the doses of 150 and 300 mg/kg b.w. The rate of plasma glucose level reduction of MEPS was dose dependent. The result indicates that MEPS may exert protective action against the hyperglycemic condition of diabetes mellitus.
**FIGURE 4:** *Effect of MEPS and metformin in oral glucose tolerance test. Data are expressed as mean ± SEM (n = 5). MEPS = methanol extract of P. sylvestris seeds. **p < .001 and *p < .05, compared to diabetic control group (Dunnett's test)*
## Alloxan‐induced diabetes
Oral ingestion of MEPS (150, 300 mg/kg b.w.) and standard drug metformin (60 mg/kg b.w.) exhibited significant ($p \leq .001$) antihyperglycemic effect in alloxan‐induced diabetic mice throughout the experimental period as shown in Table 2. Intraperitoneal treatment of alloxan (60 mg/kg b.w.) caused marked increases in glucose levels in the mice compared to the vehicle treatment group. The blood sugar level was steady at different measurement times from 0 to 24 h by nearly15 mmol/l for the alloxan‐induced diabetic control mice. The standard drug metformin significantly reduced the blood glucose level after 4 h of treatment (10 mg/kg b.w., p.o). The glucose level of alloxan‐induced diabetic mice also started to decline significantly following 4 h of oral treatment of MEPS at lower doses (50 mg/kg) compared to the diabetic control mice. However, the antihyperglycemic effect of MEPS was noticeably different from the metformin‐treated diabetic mice. The glucose‐lowering effect of MEPS was highest at the maximum dose (300 mg/kg) after 24 h of oral treatment.
**TABLE 2**
| Group | Treatment | Blood glucose level (mmol/L) | Blood glucose level (mmol/L).1 | Blood glucose level (mmol/L).2 | Blood glucose level (mmol/L).3 |
| --- | --- | --- | --- | --- | --- |
| Group | Treatment | 0 h | 4 h | 8 h | 24 h |
| Control | Vehicle (10 ml/kg) | 5.25 ± 0.45 | 5.44 ± 0.39 | 5.19 ± 0.41 | 5.31 ± 0.45 |
| Diabetic control | Vehicle (10 ml/kg) | 15.48 ± 0.39 | 15.04 ± 0.46 | 14.97 ± 0.59 | 15.46 ± 0.19 |
| Positive control | Metformin (10 mg/kg) | 15.15 ± 0.59 | 6.88 ± 0.30* | 4.97 ± 0.07* | 4.46 ± 0.16* |
| Experimental 1 | MEPS (50 mg/kg) | 15.14 ± 0.42 | 13.83 ± 0.30* | 13.73 ± 0.35 | 13.07 ± 0.43* |
| Experimental 2 | MEPS (150 mg/kg) | 15.33 ± 0.34 | 12.47 ± 0.37* | 12.09 ± 0.35* | 11.08 ± 0.33* |
| Experimental 3 | MEPS (300 mg/kg) | 15.85 ± 0.23 | 11.54 ± 0.36* | 10.76 ± 0.21* | 9.76 ± 0.28* |
## DISCUSSION
Plants are gifts of nature housed thousands of important biochemical playing major roles in the regulation and maintenance of body's homeostasis (Alam et al., 2020, 2021; Islam et al., 2022). The present study investigates the antihyperglycemic activities of crude extract of P. sylvestris seed (MEPS) and the rationale for its use in diabetes as claimed in traditional medicine. The plant P. sylvestris is grown in the wild and cultivated in different regions of southeast Asia, including Bangladesh (Lamia & Mukti, 2021). The plant is also economically valued for its multiple households, industrial purposes, and nutritional and medicinal significance in Bangladesh (Chowdhury et al., 2008; Lamia & Mukti, 2021). Previous studies reported that the plant seeds are enriched with antioxidants (Kothari et al., 2012) and protective oil (Qidwai et al., 2018). Recently published literature demonstrated that the alcohol extract seed of Phoenix dactilyfera, a native date palm of the Arecaceae family, possess promising free radical scavenging and reduced blood glucose in diabetic rats (Abiola et al., 2018). The current study reveals the phytochemicals possibly responsible for oxidative radical scavenging capacity and antihyperglycemic activities of P. sylvestris seed extensively grown in Bangladesh.
The plant extracts and compounds with profound antioxidant capacity could be promising candidates for the management of recovery of oxidative stress‐induced diseases such as diabetes (Ashrafi et al., 2022; Sultana et al., 2022; Vinayagam et al., 2016). The phenolics and flavonoids are the significant phytochemicals evidenced to remarkably restore oxidative damage by scavenging free radicals produced in diabetic patients (Emon et al., 2020, 2021; Sarian et al., 2017; Vinayagam et al., 2016). Pre‐clinical studies showed that plant phenolics could elevate plasma insulin levels and increase glucose uptake by accelerating hepatic glycolysis, glucogenesis, and gluconeogenesis (Chakrabarty et al., 2022; Rudra et al., 2020; Vinayagam et al., 2016). The antioxidant defense mechanism of flavonoids involves the mitigation of reactive oxidative species‐induced endothelial cell damage and endoplasmic reticulum stress responsible for impaired insulin and hyperglycemia (Sarian et al., 2017). The presence of a substantial amount of total phenolic and flavonoid contents and the prominent antioxidant capacity of the crude extract of P. sylvestris seed has been evidenced in the recently published literatures (Kothari et al., 2012; Qidwai et al., 2018). However, it was noticeable that the phytochemical contents varied with the extraction methods (Kothari et al., 2012; Qidwai et al., 2018). The variability could also be responsible for the geographical, ecological, and botanical conditions and harvesting times. The result of the present study indicates P. sylvestris seed (MEPS) grown in Bangladesh contains substantial amounts of phenolics and flavonoids. The results showed that the scavenging of DPPH and NO free radicals by MEPS was also noticeable. Furthermore, several antioxidant compounds, including nerolidol (Neto et al., 2013), citronellol (Jagdale et al., 2015), and phytol (Santos et al., 2013) were identified from the GC–MS analysis of MEPS. The substantial retention of phenolics, flavonoid compounds, and promising free radicals detaining capacity of MEPS further encouraged to proceed the investigation of its effect against oxidative stress‐related hyperglycemia.
Hyperglycemia and fluctuation of blood glucose levels are critical pathological indicators of the development and progression of diabetes (Mathew & Tadi, 2021). Oral glucose tolerance test primarily indicates the impairment of glucose tolerance indicates insulin resistance and associated problems of carbohydrate metabolism (Andrikopoulos et al., 2008). The test is also commonly performed to evaluate the glucose tolerance improvement capability of drug candidates or plant extracts before assessment into the additional diabetic model (Abiola et al., 2018; Dauki et al., 2022; Sornalakshmi et al., 2016). In glucose‐ingested non‐diabetic mice, MEPS treatment showed a significant reduction in plasma glucose level. The result indicates that MEPS could be effective for the improvement of metabolic uptake of glucose and re‐establish the normal blood glucose level. To justify the enhancement of glucose tolerance in diabetic‐associated condition, MEPS was further challenged in alloxan‐induced diabetic mice. Alloxan selectively causes damage to a large number of pancreatic beta cells, inhibiting the sensitivity of pancreatic glucokinase enzyme, which results in reduced insulin release and glucose uptake by the tissues. Therefore, the glucose level of blood is significantly raised, and the consequence is characterized as hyperglycemia (Saravanan & Pari, 2005). Besides, alloxan administration induces excessive generation of free radicals such as reactive oxygen species (ROS) by the activation of hydroperoxides, and lipid peroxidation system, which leads to pancreatic tissue injury as well as promotes the pathogenic consequences of diabetes (Halliwell & Gutteridge, 2015; Sabu & Kuttan, 2004). Both mechanisms of alloxan action lead to a pathological state of type 1‐like diabetes or insulin‐dependent diabetes in cells (Macdonald Ighodaro et al., 2017). The significant decrease in blood level by the MEPS (Table 2) indicates that it remarkably alleviated the hyperglycemic effect produced by alloxan. Its antioxidant potential may play a pivotal role in the effects. The presence of antidiabetic agent linalool (More et al., 2014) as well as antioxidant compounds nerolidol (Neto et al., 2013), citronellol (Jagdale et al., 2015) and phytol (Santos et al., 2013) in MEPS (Table 1) further supports the outcome of the study.
## CONCLUSION
The present study revealed that the methanol extract of P. sylvestris (MEPS) possesses strong antioxidant and antihyperglycemic activities. Quantitative analysis of MEPS indicated that it contains a considerable amount of phenolics and flavonoids. Besides, MEPS showed potent scavenging activity against the free radicals generated by DPPH and NO. MEPS significantly reduced the hyperglycemic effect induced by glucose and alloxan. This effect could be associated with its antioxidant action as well as the presence of the bioactive compounds, which were confirmed by GC–MS analysis. Therefore, further studies on the isolation as well as analysis of the biological activities of the isolated compounds, are required. The results of the present study indicate that P. sylvestris seed could be a potential natural source for developing antidiabetic compounds.
## FUNDING INFORMATION
The investigation was partially done in the Molecular Pharmacology and Herbal Drug Research Laboratory, which was established through financial support from the Higher Education Quality Enhancement Project (HEQEP), AIF, Round‐III, Window‐2, CP‐3258, University Grants Commission (UGC) of Bangladesh.
## CONFLICT OF INTEREST
No potential conflict of interest was reported by the authors.
## DATA AVAILABILITY STATEMENT
All analyzed data during this research are included in the published manuscript. *The* generated datasets during this research are not publicly available, although they can be provided from the corresponding author upon reasonable request.
## References
1. Abiola T., Dibie D. C., Akinwale O. J., Shomuyiwa O. A.. **Assessment of the antidiabetic potential of the ethanolic extract of date palm (**. *Diabetes & Metabolism Journal* (2018) **9** 1-9
2. Adeyi A. O., Nneji L. M., Idowu B. A.. **Ameliorative potentials of medicinal plants on the pathophysiological complications of diabetes mellitus: A review**. *Journal of Medicinal Plants Research* (2015) **9** 262-288
3. Alam S., Emon N. U., Shahriar S., Richi F. T., Haque M. R., Islam M. N., Sakib S. A., Ganguly A.. **Pharmacological and computer‐aided studies provide new insights into Millettia peguensis Ali (Fabaceae)**. *Saudi Pharmaceutical Journal* (2020) **28** 1777-1790. PMID: 33424267
4. Alam S., Sarker M. M. R., Afrin S., Richi F. T., Zhao C., Zhou J. R., Mohamed I. N.. **Traditional herbal medicines, bioactive metabolites, and plant products against COVID‐19: Update on clinical trials and mechanism of actions**. *Frontiers in Pharmacology* (2021) **12**. PMID: 34122096
5. Alam S., Sarker M. M. R., Sultana T. N., Chowdhury M. N. R., Rashid M. A., Chaity N. I., Zhao C., Xiao J., Hafez E. E., Khan S. A., Mohamed I. N.. **Antidiabetic phytochemicals from medicinal plants: Prospective candidates for new drug discovery and development**. *Frontiers in Endocrinology* (2022) **13** 1-35
6. Ali M. M., Agha F. G.. **Amelioration of streptozotocin‐induced diabetes mellitus, oxidative stress and dyslipidemia in rats by tomato extract lycopene**. *Scandinavian Journal of Clinical and Laboratory Investigation* (2009) **69** 371-379. PMID: 19148834
7. Alisi C., Onyeze G.. **Nitric oxide scavenging ability of ethyl acetate fraction of methanolic leaf extracts of**. *African Journal of Biochemistry Research* (2008) **2** 145-150
8. Andrikopoulos S., Blair A. R., Deluca N., Fam B. C., Proietto J.. **Evaluating the glucose tolerance test in mice**. *American Journal of Physiology, Endocrinology and Metabolism* (2008) **295** E1323-E1332. PMID: 18812462
9. Ashrafi S., Alam S., Emon N. U., Ahsan M.. **Isolation, characterization and pharmacological investigations of a new phenolic compound along with four others firstly reported phytochemicals from**. *Molecules* (2022) **27** 5972. PMID: 36144708
10. Beg M. J., Singh S. K.. **Indigenous medicinal plants used in ayurved for the treatment of diabetes in Chitrakoot area of Satna district, mp**. *Indian Journal of Life Sciences* (2015) **4** 107
11. Chakrabarty N., Chung H. J., Alam R., Emon N. U., Alam S., Kabir M. F., Islam M. M., Hong S. T., Sarkar T., MMR S., Rahman M. M.. **Chemico‐pharmacological screening of the methanol extract of**. *Molecules* (2022) **27** 3474. PMID: 35684412
12. Chaturvedi P., George S., Milinganyo M., Tripathi Y.. **Effect of**. *Phytotherapy Research: PTR* (2004) **18** 954-956. PMID: 15597317
13. Chowdhury M. S. H., Halim M., Muhammed N., Haque F., Koike M.. **Traditional utilization of wild date palm (**. *Journal of Forestry Research* (2008) **19** 245-251
14. Dauki A. M., Hsueh C.‐H., Cherala G., Othman A. A.. **Oral glucose tolerance test: An informative endpoint or an added burden in metformin drug‐drug interaction studies?**. *Clinical Pharmacology and Therapeutics* (2022) **112** 453-455. PMID: 35687738
15. Emon N. U., Alam S., Rudra S., Al Haidar I. K., Farhad M., Rana M. E. H., Ganguly A.. **Antipyretic activity of**. *Saudi Journal of Biological Sciences* (2021) **28** 5302-5309. PMID: 34466108
16. Emon N. U., Alam S., Rudra S., Chowdhury S., Rajbangshi J. C., Ganguly A.. **Evaluation of pharmacological potentials of the aerial part of**. *Advances in Traditional Medicine* (2020) 1-14
17. Ezeja M. I., Anaga A. O., Asuzu I. U.. **Antidiabetic, antilipidemic, and antioxidant activities of**. *Pharmaceutical Biology* (2015) **53** 605-614. PMID: 25330778
18. Ghani A.. *Medicinal plants of Bangladesh: Chemical constituents and uses* (1998)
19. Halliwell B., Gutteridge J. M.. *Free radicals in biology and medicine* (2015)
20. Hansen P. S., Clarke R. J., Buhagiar K. A., Hamilton E., Garcia A., White C., Rasmussen H. H.. **Alloxan‐induced diabetes reduces sarcolemmal Na+‐K+ pump function in rabbit ventricular myocytes**. *American Journal of Physiology‐Cell Physiology* (2007) **292** C1070-C1077. PMID: 17020934
21. Hartling L., Dryden D. M., Guthrie A., Muise M., Vandermeer B., Aktary W. M., Pasichnyk D., Seida J. C., Donovan L.. **Screening and diagnosing gestational diabetes mellitus**. *Evidence Report/Technology Assessment* (2012) **210** 1
22. Hasan S. R., Mariam J., Majumder M. M., Raushanara A., Hossain M. M., Mazumder M. E. H., Alam M. A., Rumana J., Rana M. S., Arif M.. **Analgesic and antioxidant activity of the hydromethanolic extract of**. *American Journal of Pharmacology and Toxicology* (2009) **4** 1-7
23. Islam M. M., Alam R., Chung H. J., Emon N. U., Fazlul Kabir M., Rudra S., Alam S., Ullah A., Hong S. T., Aktar Sayeed M.. **Chemical, pharmacological and computerized molecular analysis of Stem's extracts of**. *Nutrients* (2022) **14** 265. PMID: 35057446
24. Jagdale A. D., Kamble S. P., Nalawade M. L., Arvindekar A. U.. **Citronellol: A potential antioxidant and aldose reductase inhibitor from**. *International Journal of Pharmacy and Pharmaceutical Sciences* (2015) **7** 203-209
25. Kim J.‐M., Chung J.‐Y., Lee S.‐Y., Choi E.‐W., Kim M.‐K., Hwang C.‐Y., Youn H.‐Y.. **Hypoglycemic effects of vanadium on alloxan monohydrate induced diabetic dogs**. *Journal of Veterinary Science* (2006) **7** 391-395. PMID: 17106233
26. Kothari V.. **In vitro antibacterial activity in seed extracts of**. *Current Trends in Biotechnology and Pharmacy* (2011) **5** 993-997
27. Kothari V., Gupta A., Naraniwal M.. **Comparative study of various methods for extraction of antioxidant and antibacterial compounds from plant seeds**. *Journal of Natural Remedies* (2012) **12** 162-173
28. Lamia F. S., Mukti R. F.. **Bangladeshi wild date palm fruits (**. *International Journal of Applied Sciences and Biotechnology* (2021) **9** 32-37
29. Lips J., Jongsma H., Eskes T.. **Alloxan‐induced diabetes mellitus in pregnant sheep and chronic fetal catheterization**. *Laboratory Animals* (1988) **22** 16-22. PMID: 3280870
30. Macdonald Ighodaro O., Mohammed Adeosun A., Adeboye Akinloye O.. **Alloxan‐induced diabetes, a common model for evaluating the glycemic‐control potential of therapeutic compounds and plants extracts in experimental studies**. *Medicina* (2017) **53** 365-374. PMID: 29548636
31. Mathew T. K., Tadi P.. *StatPearls [Internet]* (2021)
32. More T. A., Kulkarni B. R., Nalawade M. L., Arvindekar A. U.. **Antidiabetic activity of linalool and limonene in streptozotocin‐induced diabetic rat: A combinatorial therapy approach**. *International Journal of Pharmacy and Pharmaceutical Sciences* (2014) **6** 159-163
33. Nayak B. S., Roberts L.. **Relationship between inflammatory markers, metabolic and anthropometric variables in the Caribbean type 2 diabetic patients with and without microvascular complications**. *Journal of Inflammation* (2006) **3** 17. PMID: 17187674
34. Neto J. D. N., de Almeida A. A. C., da Silva Oliveira J., dos Santos P. S., de Sousa D. P., de Freitas R. M.. **Antioxidant effects of nerolidol in mice hippocampus after open field test**. *Neurochemical Research* (2013) **38** 1861-1870. PMID: 23765368
35. Njogu S., Arika W., Nyamai D., Ngugi M., Machocho A., Ngeranwa J., Njagi E.. **Hypoglycemic effect of aqueous and ethyl acetate leaf extract of**. *Diabetes & Metabolism Journal* (2016) **7** 2
36. Pari L., Saravanan R.. **Antidiabetic effect of diasulin, a herbal drug, on blood glucose, plasma insulin and hepatic enzymes of glucose metabolism in hyperglycaemic rats**. *Diabetes, Obesity and Metabolism* (2004) **6** 286-292
37. Qidwai A., Kumar R., Dikshit A.. **Green synthesis of silver nanoparticles by seed of**. *Green Chemistry Letters and Reviews* (2018) **11** 176-188
38. Rahimi R., Nikfar S., Larijani B., Abdollahi M.. **A review on the role of antioxidants in the management of diabetes and its complications**. *Biomedicine & Pharmacotherapy* (2005) **59** 365-373. PMID: 16081237
39. Rates S. M. K.. **Plants as source of drugs**. *Toxicon* (2001) **39** 603-613. PMID: 11072038
40. Rice‐Evans C., Miller N., Paganga G.. **Antioxidant properties of phenolic compounds**. *Trends in Plant Science* (1997) **2** 152-159
41. Rudra S., Sawon S. U., Emon N. U., Alam S., Tareq S. M., Islam M. N., Uddin M., Md Sazid A., Hasbe A.N., Shakil M., Sakib S.A. **Biological investigations of the methanol extract of Tetrastigma leucostaphylum (Dennst.) Alston ex Mabb.(Vitaceae): In vivo and in vitro approach**. (2020)
42. Sabu M., Kuttan R.. **Antidiabetic activity of Aegle marmelos and its relationship with its antioxidant properties**. *Indian Journal of Physiology and Pharmacology* (2004) **48** 81-88. PMID: 15270373
43. Saija A., Scalese M., Lanza M., Marzullo D., Bonina F., Castelli F.. **Flavonoids as antioxidant agents: Importance of their interaction with biomembranes**. *Free Radical Biology and Medicine* (1995) **19** 481-486. PMID: 7590397
44. Santos C. C. D. M. P., Salvadori M. S., Mota V. G., Costa L. M., De Almeida A. A. C., De Oliveira G. A. L., Costa J. P., De Sousa D. P., De Freitas R. M., de Almeida R. N.. **Antinociceptive and antioxidant activities of phytol in vivo and in vitro models**. *Neuroscience Journal* (2013) **2013** 1-9
45. Saravanan R., Pari L.. **Antihyperlipidemic and antiperoxidative effect of Diasulin, a polyherbal formulation in alloxan induced hyperglycemic rats**. *BMC Complementary and Alternative Medicine* (2005) **5** 14. PMID: 15969768
46. Sarian M. N., Ahmed Q. U., Mat So'ad S. Z., Alhassan A. M., Murugesu S., Perumal V., Syed Mohamad S. N. A., Khatib A., Latip J.. **Antioxidant and antidiabetic effects of flavonoids: A structure‐activity relationship based study**. *BioMed Research International* (2017) **2017** 1-14
47. Savu O., Ionescu‐Tirgoviste C., Atanasiu V., Gaman L., Papacocea R., Stoian I.. **Increase in total antioxidant capacity of plasma despite high levels of oxidative stress in uncomplicated type 2 diabetes mellitus**. *Journal of International Medical Research* (2012) **40** 709-716. PMID: 22613434
48. Selim S. A., Adam M. E., Hassan S. M., Albalawi A. R.. **Chemical composition, antimicrobial and antibiofilm activity of the essential oil and methanol extract of the Mediterranean cypress (**. *BMC Complementary and Alternative Medicine* (2014) **14** 179. PMID: 24890383
49. Semwal D. K., Rawat U., Badoni R., Semwal R., Singh R.. **Antihyperglycemic effect of**. *Journal of Medicine* (2010) **11** 17-19
50. Shajib M. S., Akter S., Ahmed T., Imam M. Z.. **Antinociceptive and neuropharmacological activities of methanol extract of**. *Frontiers in Pharmacology* (2015) **6** 212. PMID: 26483687
51. Singleton V. L., Orthofer R., Lamuela‐Raventós R. M.. *Methods in enzymology* (1999) **299** 152-178
52. Sornalakshmi V., Tresina Soris P., Paulpriya K., Packia Lincy M., Mohan V.. **Oral glucose tolerance test (OGTT) in normal control and glucose induced hyperglycemic rats with**. *Group* (2016) **1** 59-62
53. Sultana N., Chung H. J., Emon N. U., Alam S., Taki M. T. I., Rudra S., Tahamina A., Alam R., Ahmed F., Al Mamun A.. **Biological functions of**. *Frontiers in Nutrition* (2022) **9** 1-13
54. Vinayagam R., Jayachandran M., Xu B.. **Antidiabetic effects of simple phenolic acids: A comprehensive review**. *Phytotherapy Research* (2016) **30** 184-199. PMID: 26634804
55. Walker C. I., Trevisan G., Rossato M. F., Franciscato C., Pereira M. E., Ferreira J., Manfron M. P.. **Antinociceptive activity of Mirabilis Jalapa in mice**. *Journal of Ethnopharmacology* (2008) **120** 169-175. PMID: 18761072
56. Wang S.‐S., Wang D.‐M., Pu W.‐J., Li D.‐W.. **Phytochemical profiles, antioxidant and antimicrobial activities of three potentilla species**. *BMC Complementary and Alternative Medicine* (2013) **13** 321. PMID: 24252124
57. Ziegler D., Rathmann W., Dickhaus T., Meisinger C., Mielck A.. **Neuropathic pain in diabetes, pre‐diabetes and normal glucose tolerance: The MONICA/KORA Augsburg surveys S2 and S3**. *Pain Medicine* (2009) **10** 393-400. PMID: 19207236
|
---
title: 'The “fruit and whole‐grain” pattern is associated with a low prevalence of
hypertriglyceridemia among middle and older‐aged Korean adults: Using Korea National
Health and Nutrition Examination Survey 2013–2018 data'
authors:
- SoHyun Park
- Sangwon Chung
- Seong‐Ah Kim
- Sangah Shin
journal: Food Science & Nutrition
year: 2022
pmcid: PMC10002937
doi: 10.1002/fsn3.3128
license: CC BY 4.0
---
# The “fruit and whole‐grain” pattern is associated with a low prevalence of hypertriglyceridemia among middle and older‐aged Korean adults: Using Korea National Health and Nutrition Examination Survey 2013–2018 data
## Abstract
Hypertriglyceridemia is a well‐known risk factor of various chronic diseases including diabetes mellitus, metabolic syndrome, obesity, and cardiovascular diseases. This study aimed to determine dietary patterns and explore the relationship between dietary patterns and hypertriglyceridemia in the Korean adult population. We utilized a cross‐sectional and nationally representative survey, the Korea National Health and Nutrition Examination Survey 2013–2018 database. From 47,217 subjects who participated in the survey between 2013 and 2018, only subjects over 40 years old were included. Subjects lacking 24‐h recall data and data on hypertriglyceridemia and body mass index, and who had implausible energy intake were excluded. A total of 19,806 participants' data were analyzed. Dietary data were based on 24‐h recall data, and dietary patterns were derived using factor analysis. Triglyceride levels greater than 200 mg/dl were considered hypertriglyceridemia, according to the Korean Society of Lipid and Atherosclerosis. Three dietary patterns— “oil and fats & seasoning”, “soybean paste and vegetable”, and “fruit and whole‐grain”— explained $7.9\%$, $6.3\%$, and $5.8\%$ of variation in food intake, respectively. Comparing the lowest and highest dietary pattern score groups after adjusting for potential confounders revealed an inverse relationship between “fruit and whole‐grain” dietary pattern and hypertriglyceridemia in men (odds ratio [OR]: 0.61, $95\%$ confidence interval [CI]: 0.45–0.82, p for trend <.0001); which was only marginal in women (OR: 0.78, $95\%$ CI: 0.58–1.07, p for trend:.628). A diet containing high proportions of fruit and whole‐grain may have preventive effects on hypertriglyceridemia in middle and older aged Korean adults.
This study aimed to derive dietary patterns and explore the relationship between dietary patterns and hypertriglyceridemia in the Korean middle and older aged adult population using data from a representative national survey in Korea. As a result, an inverse relationship between “fruit and whole‐grain” dietary pattern and hypertriglyceridemia was observed in men. A diet containing high proportions of fruit and whole‐grain may prevent hypertriglyceridemia in Korean adults.
## INTRODUCTION
Hypertriglyceridemia, an elevation in triglyceride (TG) levels, is an important biomarker of cardiovascular disease (Brunzell, 2007). Consistently high TG levels may lead to changes in the composition and metabolism of low‐ and high‐density lipoprotein (LDL and HDL) cholesterol, thereby causing an imbalance in serum cholesterol levels (Miller et al., 2011). Moreover, hypertriglyceridemia in conjunction with impaired fasting glucose is a well‐known risk of the development of type 2 diabetes (D'Agostino et al., 2004). Since hypertriglyceridemia may affect other metabolic factors, it is currently considered a major risk factor for various chronic diseases.
For a long time, dietary control has been advocated for managing hypertriglyceridemia. Studies have examined the association between dietary factors and hypertriglyceridemia. According to a previous study, the TG level was expected to increase by $6\%$ for every $5\%$ decrease in the percent of total fat intake in those who consumed a low‐fat and high‐carbohydrate diet (Institute of Medicine, 2005). Furthermore, studies have reported a link between the consumption of several food items and the development of hypertriglyceridemia (Guasch‐Ferre et al., 2019; Takahashi et al., 2010; Yuan et al., 2015). The consumption of more than three servings of fruits and vegetables and whole grains led to a reduction in TG levels in Brazilian adults (Takahashi et al., 2010). Additionally, a high intake of fruit, but not vegetables, was related to a decrease in the prevalence of elevated TG levels in Korean adults (Yuan et al., 2015). A meta‐analysis of randomized controlled trials showed that the consumption of red meat resulted in a greater decrease in the level of TG than that of low‐quality refined grains and simple sugars (Guasch‐Ferre et al., 2019).
The dietary pattern analysis can be used as an adjunct method to explore the combined effect of nutrients and foods in a particular population and examine the relationship between overall diet and risk of metabolic diseases (Hu, 2002). In this regard, the association of dietary patterns with hypertriglyceridemia has been evaluated. A healthy dietary pattern, characterized by high consumption of whole grains, legumes, fruits, and vegetables, was inversely associated with low levels of TG in the Western population (Panagiotakos et al., 2007). Moreover, a similar dietary pattern was negatively associated with fasting TG in a multiethnic Asian population (Whitton et al., 2018).
Importantly, the relationship between specific dietary patterns and triglycerides has not been clearly elucidated in the Korean population. In a previous study involving Korean adults, a healthy dietary pattern, including grains, vegetables, and fish, was associated with a lower risk of hypertriglyceridemia than a traditional dietary pattern (Kim & Jo, 2011), while in another similar study, it showed a nonsignificant relationship (Song & Joung, 2012). Therefore, we aimed to derive dietary patterns and explore the relationship between dietary patterns and hypertriglyceridemia in the Korean adult population using data from a representative national survey in Korea.
## Study population
In this study, we utilized the database of the Korea National Health and Nutrition Examination Survey (KNHANES)—a cross‐sectional and nationally representative survey conducted by the Division of Chronic Disease Surveillance, The Korea Disease Control and Prevention Agency. The KNHANES was approved by the Institutional Review Board of the Korea Disease Control and Prevention Agency (2013‐07CON‐03‐4C, 2013‐12EXP‐03‐5C, and 2018‐01‐03‐P‐A). Among the 47,217 subjects who participated in the survey between 2013 and 2018, subjects who were less than 40 years old ($$n = 20$$,420) and lacked 24‐h recall data ($$n = 3068$$), data on hypertriglyceridemia ($$n = 2657$$) and body mass index (BMI) ($$n = 49$$) were excluded. In addition, 1217 subjects who had implausibly high (male, >5000 kcal/day; female, >3500 kcal/day) or low (male, <500 kcal/day; female, <800 kcal/day) energy intake were excluded. Finally, a total of 19,806 participants' data were included in the analysis (Figure 1).
**FIGURE 1:** *Flow diagram of analytical samples in this study: Korea National Health and Nutrition Examination Survey 2013–2018*
## Dietary intake and dietary pattern
Dietary intake and the dietary patterns were based on the 24‐h recall data. Data on intake of major macronutrients, carbohydrate, protein, and fat were obtained and percentage energy from these macronutrient intakes was assessed as follows: macronutrient intake (g/day) * kcal/g (carbohydrate and protein: 4 kcal/g, fat: 9 kcal/g)/total energy intake (kcal/day). The food items were categorized into 24 food groups based on the similarity of food type and nutrient composition. The specific food components of each food group are described in Table S1. Using factor analysis, different dietary patterns were identified from daily food intake within each food group. Principal component analysis was conducted to identify the distinct dietary patterns. The number of factors was determined using eigenvalues >1.3 and scree plots. The factor score for each dietary pattern was calculated through the sum of food intake weighted by their factor loadings. Three factors with high positive loadings were used to derive and label dietary patterns: “oil and fats & seasoning,” “soybean paste and vegetable,” and “fruit and whole‐grain.” The subjects were divided into three groups based on their dietary pattern scores.
## Hypertriglyceridemia
Hypertriglyceridemia was defined if triglyceride level of participants is greater than 200 mg/dl, based on the guidelines of the Korean Society of Lipid and Atherosclerosis (Committee for the Korean Guidelines for the Management of Dyslipidemia, 2016), or participants are taking medication for lowering blood lipids.
## Covariates
Sociodemographic and health‐related behavior data, including household income level, physical activity, alcohol consumption, and smoking were obtained via a self‐administered questionnaire. Household income level was classified as low, medium‐low, medium‐high, and high. Physical activity status was defined according to the frequency of walking and categorized as none, <3 days/week, and 4–7 days/week. Current alcohol consumption was determined as the presence or absence of alcohol intake in the recent 1 year (“yes” and “no”). Based on the smoking status, the subjects were categorized as “non‐smokers” and “smokers.” Height and weight were measured using standard protocol by trained staff. BMI was calculated as the weight divided by the square of the height (kg/m2).
## Statistical analysis
Statistical analyses were performed using SAS software (version 9.4; SAS Institute, Inc.). Statistical significance was set at $p \leq .05.$ *Categorical data* were presented as numbers and percentages and continuous data as means with standard error. The differences across quantiles of dietary pattern scores were determined using the generalized linear model for continuous data and chi‐square test for categorical data. Multivariable‐adjusted logistic regression analysis was used to calculate the odds ratio (OR) and $95\%$ confidence interval (CI) for hypertriglyceridemia.
Model 1 was adjusted for age (continuous) and other dietary pattern scores (continuous). Model 2 was additionally adjusted for total energy intake (continuous), household income (low, middle‐low, middle‐high, high), current smoking status (yes, no), alcohol consumption status (yes, no), and physical activities (none, <3 days/week, and 4–7 days/week). Model 3 was additionally adjusted for energy‐adjusted carbohydrate (continuous) using the residual method (Willett & Stampfer, 1986).
## RESULTS
The factor loadings for three dietary patterns are shown in Table 1. The “oil and fats & seasoning” pattern included high consumption of oil and fats, seasoning, egg, flour, and low consumption of white rice. The “soybean paste and vegetable” pattern included high consumption of soybean paste, vegetables, fish and seafood, kimchi, and low consumption of pizza, hamburger, bread, sandwich, cereal, and dairy products. The “fruit and whole‐grain” pattern included high consumption of fruit, whole grains, dairy products, potato, and low consumption of meat, alcoholic beverages, and white rice.
**TABLE 1**
| Unnamed: 0 | Dietary patterns | Dietary patterns.1 | Dietary patterns.2 |
| --- | --- | --- | --- |
| | Factor 1 | Factor 2 | Factor 3 |
| | Oil and fats & seasoning pattern | Soybean paste and vegetable pattern | Fruit and whole‐grain pattern |
| White rice | −0.158 | 0.453 | −0.457 |
| Whole grains | | 0.217 | 0.468 |
| Flours | 0.366 | | |
| Noodles | 0.261 | | |
| Bread, sandwiches, and cereal | | −0.239 | |
| Pizza and hamburgers | 0.318 | −0.208 | |
| Potatoes | | | 0.372 |
| Fruits | | | 0.487 |
| Vegetables | | 0.524 | 0.265 |
| Kimchi | | 0.361 | |
| Mushrooms | 0.165 | | |
| Legumes and soybean products | | 0.292 | |
| Soybean paste | | 0.575 | |
| Salted foods | 0.249 | | |
| Eggs | 0.396 | | |
| Meats | 0.330 | | −0.248 |
| Fish and seafood | | 0.398 | 0.174 |
| Dairy products | | −0.162 | 0.416 |
| Beverages | 0.359 | | |
| Alcoholic beverages | 0.265 | | −0.339 |
| Sugar and syrup | 0.317 | | |
| Nuts | | | 0.295 |
| Oils and fats | 0.673 | | |
| Seasonings | 0.488 | | |
| The variance of intake explained (%) | 7.9 | 6.3 | 5.8 |
*The* general characteristics of this study population according to quantiles of three pattern scores are described in Table 2. The “oil and fats & seasoning” dietary pattern was positively associated with men who were younger, more active, and had higher alcohol intake, and women who were younger and smoked and consumed alcohol frequently. The “soybean‐paste and vegetable” dietary pattern group included men who were old and tended to drink and smoke often and women who were old and smoked and consumed alcohol infrequently. Likewise, the “fruit and whole‐grain” dietary pattern group included men in a high‐income household who also smoked and consumed alcohol infrequently and women who were young, active, and smoked and consumed alcohol infrequently.
**TABLE 2**
| Unnamed: 0 | Quantile (Q) of the dietary pattern | Quantile (Q) of the dietary pattern.1 | Quantile (Q) of the dietary pattern.2 | Quantile (Q) of the dietary pattern.3 | p‐Value a |
| --- | --- | --- | --- | --- | --- |
| | Q1 | Q2 | Q3 | Q4 | p‐Value a |
| Men (N = 7933) | 1983 | 1983 | 1984 | 1983 | |
| Oil and fats & seasoning pattern | | | | | |
| Triglyceride range (mg/dl) | 83–178 | 88–190 | 89–193 | 94–202 | |
| Age (year) | 61.2 ± 0.5 | 57.2 ± 0.4 | 54.1 ± 0.3 | 51.4 ± 0.3 | <.0001 |
| BMI (kg/m2) | 23.9 ± 0.1 | 24.4 ± 0.1 | 24.2 ± 0.1 | 24.5 ± 0.1 | .1149 |
| Household income (high) | 290 (18.1) | 474 (27.6) | 602 (34.0) | 776 (42.5) | <.0001 |
| Physical activity (>3 days/week) | 335 (28.8) | 407 (29.6) | 399 (29.1) | 424 (31.6) | <.0001 |
| Current alcohol drinkers | 1331 (78.4) | 1454 (82.7) | 1580 (87.4) | 1635 (89.0) | <.0001 |
| Current smokers | 1560 (81.5) | 1559 (81.3) | 1585 (83.8) | 1580 (81.7) | <.0001 |
| Soybean‐paste & vegetable pattern | | | | | |
| Age (year) | 54.2 ± 0.4 | 55.8 ± 0.4 | 55.8 ± 0.4 | 56.6 ± 0.4 | .0379 |
| BMI (kg/m2) | 24.3 ± 0.1 | 24.4 ± 0.1 | 24.2 ± 0.1 | 24.2 ± 0.1 | .3260 |
| Household income (high) | 568 (33.5) | 517 (29.9) | 511 (30.2) | 546 (32.7) | <.0001 |
| Physical activity (>3 days/week) | 411 (31.0) | 384 (28.9) | 399 (30.8) | 371 (28.7) | <.0001 |
| Current alcohol drinkers | 1478 (83.4) | 1485 (84.8) | 1502 (86.8) | 1535 (85.1) | <.0001 |
| Current smokers | 1553 (81.5) | 1581 (82.2) | 1567 (82.3) | 1583 (82.4) | <.0001 |
| Fruit and whole‐grain pattern | | | | | |
| Age (year) | 54.6 ± 0.4 | 55.9 ± 0.4 | 55.4 ± 0.4 | 56.3 ± 0.4 | <.0001 |
| BMI (kg/m2) | 24.1 ± 0.1 | 24.2 ± 0.1 | 24.4 ± 0.1 | 24.3 ± 0.1 | <.0001 |
| Household income (high) | 476 (28.1) | 445 (26.6) | 542 (32.9) | 679 (39.2) | <.0001 |
| Physical activity (>3 days/week) | 344 (27.2) | 359 (26.6) | 407 (31.5) | 455 (34.7) | <.0001 |
| Current alcohol drinkers | 1657 (87.9) | 1596 (83.0) | 1552 (80.4) | 1479 (76.8) | <.0001 |
| Current smokers | 1641 (91.4) | 1511 (85.9) | 1411 (80.4) | 1437 (81.6) | <.0001 |
| Women (N = 11,873) | 2968 | 2968 | 2969 | 2968 | |
| Oil and fats & seasoning pattern | | | | | |
| Age (year) | 62.3 ± 0.4 | 56.9 ± 0.4 | 54.1 ± 0.3 | 51.7 ± 0.3 | <.0001 |
| BMI (kg/m2) | 24.3 ± 0.1 | 23.9 ± 0.1 | 23.8 ± 0.1 | 23.5 ± 0.1 | <.0001 |
| Household income (high) | 352 (13.3) | 702 (26.4) | 909 (33.3) | 1110 (38.8) | <.0001 |
| Physical activity (>3 days/week) | 549 (27.5) | 675 (33.4) | 757 (36.9) | 759 (37.8) | <.0001 |
| Current alcohol drinkers | 1239 (66.1) | 1630 (74.3) | 1828 (79.0) | 1999 (80.3) | <.0001 |
| Current smokers | 188 (6.9) | 225 (8.1) | 231 (8.6) | 242 (8.4) | <.0001 |
| Soybean‐paste & vegetable pattern | | | | | |
| Age (year) | 54.1 ± 0.4 | 56.3 ± 0.4 | 56.8 ± 0.4 | 57.0 ± 0.4 | .0012 |
| BMI (kg/m2) | 23.8 ± 0.1 | 23.8 ± 0.1 | 23.8 ± 0.1 | 23.8 ± 0.1 | .8551 |
| Household income (high) | 950 (34.0) | 747 (27.7) | 690 (26.0) | 686 (26.0) | .0031 |
| Physical activity (>3 days/week) | 725 (35.0) | 669 (34.4) | 668 (33.1) | 678 (33.9) | <.0001 |
| Current alcohol drinkers | 1815 (77.5) | 1659 (75.3) | 1638 (75.2) | 1584 (74.3) | <.0001 |
| Current smokers | 278 (10.2) | 212 (7.4) | 212 (7.6) | 184 (6.8) | <.0001 |
| Fruit and whole‐grain pattern | | | | | |
| Age (year) | 56.8 ± 0.4 | 56.1 ± 0.4 | 55.7 ± 0.4 | 55.4 ± 0.3 | <.0001 |
| BMI (kg/m2) | 24.1 ± 0.1 | 23.8 ± 0.1 | 23.8 ± 0.1 | 23.6 ± 0.1 | <.0001 |
| Household income (high) | 561 (20.7) | 710 (27.7) | 810 (29.8) | 992 (35.8) | <.0001 |
| Physical activity (>3 days/week) | 559 (28.1) | 637 (31.1) | 740 (35.9) | 804 (41.6) | <.0001 |
| Current alcohol drinkers | 1753 (80.3) | 1656 (75.6) | 1622 (73.7) | 1665 (73.2) | <.0001 |
| Current smokers | 317 (11.7) | 234 (8.6) | 179 (6.4) | 156 (5.8) | <.0001 |
The energy and the macronutrient intake according to three dietary pattern scores are presented in Table 3. The participants following the “oil and fats & seasoning” dietary pattern tended to have high energy, carbohydrate, protein, and fat intake and high percentage of energy from protein and fat intake in both men and women. The participants following the “soybean paste and vegetable” dietary pattern tended to have high energy, carbohydrate, protein, and fat intake, and high percentage of energy from carbohydrate intake in men and women. The participants following the “fruit and whole‐grain” dietary pattern tended to have a high proportion of three main macronutrients, which are carbohydrate, protein, and fat. Men who had a high intake of “fruit and whole‐grain” pattern tended to have low energy while women tended to have high energy.
**TABLE 3**
| Unnamed: 0 | Quantile (Q) of the dietary pattern | Quantile (Q) of the dietary pattern.1 | Quantile (Q) of the dietary pattern.2 | Quantile (Q) of the dietary pattern.3 | p‐Value a |
| --- | --- | --- | --- | --- | --- |
| | Q1 | Q2 | Q3 | Q4 | p‐Value a |
| Men (N = 7933) | 1983 | 1983 | 1984 | 1983 | |
| Oil and fats & seasoning pattern | Oil and fats & seasoning pattern | Oil and fats & seasoning pattern | Oil and fats & seasoning pattern | Oil and fats & seasoning pattern | Oil and fats & seasoning pattern |
| Total energy intake | 1748.1 ± 24.1 | 1944.5 ± 21.9 | 2195.5 ± 20.8 | 2538.4 ± 21.8 | <.0001 |
| Carbohydrate (g/day) | 319.9 ± 4.7 | 319.3 ± 3.9 | 338.3 ± 3.9 | 354.0 ± 4.2 | .0011 |
| % energy from carbohydrate | 73.2 ± 0.4 | 66.1 ± 0.4 | 62.2 ± 0.5 | 56.3 ± 0.5 | <.0001 |
| Protein (g/day) | 52.4 ± 0.9 | 63.7 ± 0.9 | 74.1 ± 0.9 | 93.6 ± 1.3 | <.0001 |
| % energy from protein | 12.0 ± 0.1 | 13.3 ± 0.2 | 13.6 ± 0.2 | 14.8 ± 0.1 | <.0001 |
| Fat (g/day) | 22.0 ± 0.6 | 33.4 ± 0.8 | 42.3 ± 0.8 | 57.0 ± 1.1 | <.0001 |
| % energy from fat | 11.1 ± 0.2 | 15.2 ± 0.3 | 17.3 ± 0.3 | 20.2 ± 0.3 | <.0001 |
| Soybean paste and vegetable pattern | Soybean paste and vegetable pattern | Soybean paste and vegetable pattern | Soybean paste and vegetable pattern | Soybean paste and vegetable pattern | Soybean paste and vegetable pattern |
| Total energy intake | 1952.8 ± 26.6 | 2012.3 ± 24.9 | 2152.1 ± 21.2 | 2471.7 ± 22.8 | <.0001 |
| Carbohydrate (g/day) | 293.1 ± 4.3 | 316.5 ± 4.1 | 340.4 ± 3.8 | 393.0 ± 4.4 | .0012 |
| % energy from carbohydrate | 61.5 ± 0.6 | 64.5 ± 0.6 | 64.5 ± 0.5 | 64.6 ± 0.5 | <.0001 |
| Protein (g/day) | 63.9 ± 1.2 | 65.5 ± 1.1 | 73.0 ± 1.1 | 89.7 ± 1.2 | <.0001 |
| % energy from protein | 13.0 ± 0.2 | 13.0 ± 0.1 | 13.6 ± 0.2 | 14.5 ± 0.1 | .1928 |
| Fat (g/day) | 41.0 ± 1.0 | 36.3 ± 1.0 | 39.7 ± 0.9 | 43.7 ± 1.1 | .0039 |
| % energy from fat | 18.0 ± 0.3 | 15.7 ± 0.3 | 16.1 ± 0.3 | 15.5 ± 0.3 | .1126 |
| Fruit and whole‐grain pattern | Fruit and whole‐grain pattern | Fruit and whole‐grain pattern | Fruit and whole‐grain pattern | Fruit and whole‐grain pattern | Fruit and whole‐grain pattern |
| Total energy intake | 2322.3 ± 24.6 | 1991.1 ± 25.8 | 2032.1 ± 24.2 | 2214.6 ± 23.3 | <.0001 |
| Carbohydrate (g/day) | 330.9 ± 4.2 | 311.2 ± 3.8 | 328.7 ± 4.2 | 368.2 ± 4.2 | <.0001 |
| % energy from carbohydrate | 58.3 ± 0.6 | 64.2 ± 0.5 | 65.5 ± 0.4 | 67.1 ± 0.4 | <.0001 |
| Protein (g/day) | 74.4 ± 1.2 | 67.1 ± 1.2 | 71.3 ± 1.2 | 78.0 ± 1.3 | <.0001 |
| % energy from protein | 12.7 ± 0.1 | 13.4 ± 0.1 | 14.0 ± 0.2 | 14.1 ± 0.2 | <.0001 |
| Fat (g/day) | 39.4 ± 0.9 | 36.4 ± 1.1 | 40.2 ± 0.9 | 44.9 ± 1.0 | <.0001 |
| % energy from fat | 14.7 ± 0.3 | 15.6 ± 0.3 | 17.3 ± 0.3 | 17.8 ± 0.3 | <.0001 |
| Women (N = 11,873) | 2968 | 2968 | 2969 | 2968 | |
| Oil and fats & seasoning pattern | Oil and fats & seasoning pattern | Oil and fats & seasoning pattern | Oil and fats & seasoning pattern | Oil and fats & seasoning pattern | Oil and fats & seasoning pattern |
| Total energy intake | 1388.9 ± 17.2 | 1499.8 ± 15.6 | 1644.6 ± 15.7 | 2093.2 ± 17.1 | <.0001 |
| Carbohydrate (g/day) | 268.3 ± 3.6 | 266.2 ± 3.3 | 272.7 ± 3.2 | 313.0 ± 3.2 | <.0001 |
| % energy from carbohydrate | 77.0 ± 0.3 | 70.6 ± 0.4 | 66.1 ± 0.3 | 60.0 ± 0.4 | <.0001 |
| Protein (g/day) | 39.5 ± 0.5 | 48.1 ± 0.6 | 57.1 ± 0.7 | 79.0 ± 1.0 | <.0001 |
| % energy from protein | 11.5 ± 0.1 | 13.0 ± 0.1 | 14.0 ± 0.1 | 15.1 ± 0.1 | <.0001 |
| Fat (g/day) | 16.0 ± 0.4 | 25.3 ± 0.5 | 33.1 ± 0.6 | 52.7 ± 0.9 | <.0001 |
| % energy from fat | 10.5 ± 0.2 | 15.3 ± 0.3 | 18.3 ± 0.3 | 22.5 ± 0.3 | <.0001 |
| Soybean paste and vegetable pattern | | | | | |
| Total energy intake | 1482.8 ± 20.9 | 1529.8 ± 16.8 | 1684.0 ± 18.0 | 2012.4 ± 19.7 | <.0001 |
| Carbohydrate (g/day) | 239.2 ± 3.6 | 256.2 ± 2.8 | 288.9 ± 3.0 | 343.6 ± 3.5 | <.0001 |
| % energy from carbohydrate | 65.4 ± 0.4 | 68.3 ± 0.4 | 69.5 ± 0.4 | 69.0 ± 0.4 | .0497 |
| Protein (g/day) | 49.7 ± 1.0 | 50.3 ± 0.8 | 56.7 ± 0.8 | 71.4 ± 1.0 | .0053 |
| % energy from protein | 13.3 ± 0.2 | 13.1 ± 0.1 | 13.4 ± 0.1 | 14.2 ± 0.1 | .2677 |
| Fat (g/day) | 33.0 ± 0.8 | 30.2 ± 0.8 | 30.4 ± 0.7 | 37.0 ± 0.8 | .0441 |
| % energy from fat | 19.2 ± 0.3 | 16.8 ± 0.3 | 15.6 ± 0.3 | 16.0 ± 0.3 | .5500 |
| Fruit and whole‐grain pattern | | | | | |
| Total energy intake | 1616.4 ± 17.7 | 1526.0 ± 19.2 | 1644.4 ± 20.4 | 1897.7 ± 17.5 | <.0001 |
| Carbohydrate (g/day) | 268.8 ± 3.1 | 255.2 ± 3.1 | 275.5 ± 3.6 | 323.5 ± 3.6 | <.0001 |
| % energy from carbohydrate | 67.8 ± 0.5 | 67.7 ± 0.4 | 67.9 ± 0.4 | 68.7 ± 0.4 | <.0001 |
| Protein (g/day) | 51.7 ± 0.8 | 51.6 ± 0.8 | 57.0 ± 0.9 | 66.8 ± 1.0 | <.0001 |
| % energy from protein | 12.6 ± 0.1 | 13.5 ± 0.1 | 13.8 ± 0.1 | 14.1 ± 0.2 | <.0001 |
| Fat (g/day) | 28.2 ± 0.7 | 29.5 ± 0.8 | 33.7 ± 0.8 | 39.0 ± 0.8 | <.0001 |
| % energy from fat | 14.9 ± 0.3 | 16.8 ± 0.3 | 17.9 ± 0.3 | 18.1 ± 0.3 | <.0001 |
Table 4 shows the ORs and $95\%$ CIs of hypertriglyceridemia across the quantiles of three dietary pattern scores. There were no significant associations between “oil and fats & seasoning” and “soybean paste and vegetable” patterns and the prevalence of hypertriglyceridemia in both men and women. The ORs of the highest quantile (Q4) for the “fruit and whole‐grain” pattern showed a statistically significant association with the prevalence of hypertriglyceridemia in men (OR: 0.61, $95\%$ CI: 0.45–0.82, p for trend <.0001; model 3). Likewise, for women participants in Q4, the prevalence of hypertriglyceridemia was significantly decreased in model 1 (OR: 0.68, $95\%$ CI: 0.52–0.90, p for trend =.0051) but in model 2 (OR: 0.79, $95\%$ CI: 0.58–1.07 p for trend =.0856) and model 3 (OR: 0.78, $95\%$ CI: 0.58–1.07, p for trend =.062) it was on the borderline.
**TABLE 4**
| Unnamed: 0 | Quantile (Q) of the dietary pattern | Quantile (Q) of the dietary pattern.1 | Quantile (Q) of the dietary pattern.2 | Quantile (Q) of the dietary pattern.3 | p for trend |
| --- | --- | --- | --- | --- | --- |
| | Q1 | Q2 | Q3 | Q4 | p for trend |
| Men (N = 7933) | Men (N = 7933) | Men (N = 7933) | Men (N = 7933) | Men (N = 7933) | Men (N = 7933) |
| Oil and fats & seasoning pattern | Oil and fats & seasoning pattern | Oil and fats & seasoning pattern | Oil and fats & seasoning pattern | Oil and fats & seasoning pattern | Oil and fats & seasoning pattern |
| N | 1983 | 1983 | 1984 | 1983 | |
| Triglyceride range (mg/dl) | 83–178 | 88–190 | 89–193 | 94–202 | |
| Model 1 a | 1.00 (ref.) | 0.99 (0.74–1.32) | 1.03 (0.77–1.36) | 1.12 (0.84–1.48) | .2780 |
| Model 2 b | 1.00 | 0.95 (0.71–1.28) | 1.04 (0.76–1.41) | 1.09 (0.77–1.53) | .4864 |
| Model 3 c | 1.00 | 0.91 (0.67–1.23) | 0.96 (0.70–1.32) | 0.97 (0.68–1.39) | .9635 |
| Soybean paste and vegetable pattern | Soybean paste and vegetable pattern | Soybean paste and vegetable pattern | Soybean paste and vegetable pattern | Soybean paste and vegetable pattern | Soybean paste and vegetable pattern |
| N | 1983 | 1983 | 1983 | 1984 | |
| Triglyceride range (mg/dl) | 90–198 | 89–191 | 88–185 | 88–190 | |
| Model 1 | 1.00 | 0.95 (0.72–1.25) | 1.01 (0.77–1.32) | 0.95 (0.72–1.26) | .9247 |
| Model 2 | 1.00 | 0.93 (0.70–1.24) | 1.02 (0.761–1.36) | 0.93 (0.67–1.28) | .8974 |
| Model 3 | 1.00 | 0.95 (0.72–1.26) | 1.05 (0.78–1.40) | 0.97 (0.70–1.34) | .5518 |
| Fruit and whole‐grain pattern | Fruit and whole‐grain pattern | Fruit and whole‐grain pattern | Fruit and whole‐grain pattern | Fruit and whole‐grain pattern | Fruit and whole‐grain pattern |
| N | 1983 | 1983 | 1984 | 1983 | |
| Triglyceride range (mg/dl) | 99–218 | 90–189 | 85–182 | 85–177 | |
| Model 1 | 1.00 | 0.72 (0.54–0.95) | 0.65 (0.49–0.85) | 0.55 (0.41–0.73) | <.0001 |
| Model 2 | 1.00 | 0.71 (0.53–0.95) | 0.64 (0.47–0.85) | 0.56 (0.42–0.75) | .0002 |
| Model 3 | 1.00 | 0.73 (0.55–0.99) | 0.67 (0.50–0.90) | 0.61 (0.45–0.82) | <.0001 |
| Women (N = 11,873) | Women (N = 11,873) | Women (N = 11,873) | Women (N = 11,873) | Women (N = 11,873) | Women (N = 11,873) |
| Oil and fats & seasoning pattern | Oil and fats & seasoning pattern | Oil and fats & seasoning pattern | Oil and fats & seasoning pattern | Oil and fats & seasoning pattern | Oil and fats & seasoning pattern |
| N | 2968 | 2968 | 2969 | 2968 | |
| Triglyceride range (mg/dl) | 84–165 | 78–153 | 72–147 | 70–145 | |
| Model 1 | 1.00 | 1.10 (0.82–1.48) | 1.00 (0.76–1.34) | 1.17 (0.88–1.57) | .4361 |
| Model 2 | 1.00 | 1.16 (0.85–1.59) | 1.08 (0.79–1.46) | 1.30 (0.91–1.84) | .2195 |
| Model 3 | 1.00 | 1.18 (0.86–1.62) | 1.11 (0.80–1.53) | 1.36 (0.92–2.01) | .1619 |
| Soybean paste and vegetable pattern | Soybean paste and vegetable pattern | Soybean paste and vegetable pattern | Soybean paste and vegetable pattern | Soybean paste and vegetable pattern | Soybean paste and vegetable pattern |
| N | 2968 | 2968 | 2969 | 2968 | |
| Triglyceride range (mg/dl) | 71–147 | 74–150 | 78–154 | 78–159 | |
| Model 1 | 1.00 | 0.99 (0.75–1.32) | 0.90 (0.67–1.20) | 1.10 (0.84–1.45) | .6100 |
| Model 2 | 1.00 | 1.06 (0.78–1.42) | 0.92 (0.68–1.25) | 1.16 (0.84–1.61) | .4598 |
| Model 3 | 1.00 | 1.05 (0.78–1.42) | 0.92 (0.68–1.24) | 1.16 (0.84–1.61) | .6994 |
| Fruit and whole‐grain pattern | Fruit and whole‐grain pattern | Fruit and whole‐grain pattern | Fruit and whole‐grain pattern | Fruit and whole‐grain pattern | Fruit and whole‐grain pattern |
| N | 2968 | 2968 | 2969 | 2968 | |
| Triglyceride range (mg/dl) | 76–158 | 76–160 | 74–147 | 73–147 | |
| Model 1 | 1.00 | 0.85 (0.65–1.12) | 0.71 (0.54–0.92) | 0.68 (0.52–0.90) | .0051 |
| Model 2 | 1.00 | 0.91 (0.69–1.21) | 0.79 (0.60–1.04) | 0.79 (0.58–1.07) | .0856 |
| Model 3 | 1.00 | 0.91 (0.69–1.21) | 0.79 (0.59–1.04) | 0.78 (0.58–1.07) | .0628 |
## DISCUSSION
In this study, using data from 2013–2018 KNHANES, a negative relationship was observed between the “fruit and whole‐grain” dietary pattern and hypertriglyceridemia in middle and older aged Korean adults. Among men following the “fruit and whole‐grain” pattern, the highest dietary pattern score group was significantly associated with a $39\%$ lower prevalence of hypertriglyceridemia than the lowest dietary pattern score group. Among women, there was a marginal negative association between hypertriglyceridemia and “fruit and whole‐grain” dietary pattern.
The assessment of dietary patterns rather than a single nutrient or specific food intake is known to be a better approach to evaluate the association of overall diet with chronic diseases in a population. Dietary guidelines focusing on cardiovascular health, such as dietary approaches to stop hypertension (DASH) and the Mediterranean diet, recommend overall consumption of healthy foods, including fruits, vegetables, whole grains, fish, nuts, legumes, and olive oil (Salehi‐Abargouei et al., 2013; Zazpe et al., 2008). Similarly, the definition of “healthy diet” by the American Heart Association also focuses on whole foods and dietary patterns (Lloyd‐Jones et al., 2010). The relationship between dietary pattern and cardiometabolic risk factors has been assessed previously in various studies. Specifically, healthy dietary patterns, characterized by a high content of vegetables, fish, poultry, fruit, and whole grains, were associated with reduced risk for metabolic syndrome in Asian countries (Fabiani et al., 2019).
As described above, fruit consumption is generally included in healthy dietary patterns to control cardiometabolic parameters (Panagiotakos et al., 2007). Indeed, a high intake of fruits was negatively associated with hypertriglyceridemia in a recent meta‐analysis (Kodama et al., 2018). In addition, increased frequency of fruit consumption (≥3–4 servings/day) was negatively associated with hypertriglyceridemia in the Asian and South American populations (Lim & Kim, 2020; Takahashi et al., 2010). The consumption of total fruit and specific fruit subgroups, such as citrus, noncitrus, and carotene‐rich fruits, also showed an inverse association with hypertriglyceridemia in an observational study (Yuan et al., 2015). In an intervention study, grapefruit supplementation led to a decrease in serum TG in hyperlipidemic patients (Gorinstein et al., 2006).
However, the association between fruit intake and hypertriglyceridemia should be interpreted with caution since a high amount of fructose in fruits may lead to an increase in blood TG levels (Sharma et al., 2016). Dietary fructose led to an increase in postprandial TG levels in the short term, controlled feeding studies (Schaefer et al., 2009), and intake of only vegetables, not fruit, was related to a reduced risk of type 2 diabetes mellitus, which has been linked to hypertriglyceridemia (Colditz et al., 1992; Villegas et al., 2008; Williams et al., 1999), indicating that fruits and vegetables rich in phytochemicals or fibers have an antihypertriglyceridemia effect that is attenuated by fructose (Kodama et al., 2018). Therefore, we evaluated energy‐adjusted carbohydrate intake to exclude the potential confounding effects of simple sugar intake. The results suggesting a favorable effect of fruit intake on hypertriglyceridemia in this study can be useful evidence for a dietary recommendation.
Whole‐grain consumption also plays a potential role in the prevention of hypertriglyceridemia. In a 12‐week nutritional intervention study, a whole grain cereal‐based diet led to a greater decrease in postprandial TG levels than a refined cereal diet (Giacco et al., 2014). In another randomized crossover study, a lower value for incremental area under the curve for TG level was observed in subjects on a whole‐grain diet with dairy, chicken, and nuts than in subjects on a refined grain diet with red meat (Kim et al., 2016). Meanwhile, a whole‐grain diet also had favorable effects on other lipid profiles. According to Giacco et al., intake of whole‐meal wheat foods for 3 weeks significantly reduced postprandial plasma LDL cholesterol in their randomized clinical study (Giacco et al., 2010). In a meta‐analysis of randomized controlled studies, whole‐grain diets led to a greater decrease in total and LDL cholesterol than nonwhole‐grain diets (Hollaender et al., 2015).
In particular, the consumption of a whole‐grain diet can be an important dietary factor among Koreans who consume rice as a staple food. Asian populations, including Korean, Japanese, and Chinese, consume white rice, which is approximately 30–$40\%$ of their total carbohydrate intake (Eshak et al., 2014; Rebello et al., 2014). Koreans derive about $60\%$ of their energy from carbohydrate intake, while adults in the US and UK derive about 45–$47\%$ energy from carbohydrate intake (U.S. Department of Agriculture & Agricultural Research Service, 2020; Ministry of Health and Welfare, 2020; Public Health England, 2020; Song & Song, 2019). Indeed, white rice consumption was positively associated with the prevalence of dyslipidemia in Korean adults (Song & Song, 2019). In addition, among subjects following the “Korean healthy” dietary pattern, characterized by whole grains, legumes, vegetables, and fruits, the highest score group showed significantly lower TG levels than the lowest score group, while among the subjects following the “rice and vegetables” dietary pattern, consisting of white rice, vegetables, and eggs, showed no significant relationship with triglyceride levels in Korean population with type 2 diabetes (Lim et al., 2011). In summary, the whole grain dietary pattern can play a major role in preventing hypertriglyceridemia among Koreans who consume a large amount of rice‐based carbohydrate‐rich food items.
However, the mechanism underlying the negative correlation between triglyceride level and whole‐grain diet remains unclear. Nevertheless, dietary fibers in whole grain appear to reduce blood lipid levels. In particular, the viscous fibers, such as β‐glucans, a major nutrient of whole grains, including barley and oat, have shown hypocholesterolemic effects (Tosh & Bordenave, 2020). In addition, whole grain consumption may result in either lipoprotein lipase activation or synthesis of these lipoproteins and induce a decrease in postprandial TG levels (Frayn, 2002; Giacco et al., 2014). Further studies exploring the effect of whole grain and its components on TG levels are required.
To the best of our knowledge, this is the first study to reveal the inverse association between hypertriglyceridemia and the “fruit and whole‐grain” dietary pattern in a nationally representative sample of the middle‐aged and older aged Korean adult population. However, several limitations exist in this study. First, the 1‐day, 24‐h dietary recall method, which was used for dietary assessment in KNHANES, may not accurately reflect the usual dietary intake and its long‐term effect in subjects. Second, several decisions used in the factor analysis, such as the number of extracted dietary pattern groups, categorization of food items into food groups, and the rotation method, can be subjective (Martinez et al., 1998). Finally, this study was not able to examine the causal relationship between dietary pattern and hypertriglyceridemia due to the cross‐sectional study design of KNHANES. In addition, despite adjusting for confounding variables, there may have been potential confounding bias due to the observational study design of this study. These reasons may limit the possibility of generalizing the results. Therefore, additional studies with different study designs, such as clinical or prospective cohort study are required to confirm the causal relationship of the “fruit and whole‐grain” dietary pattern with hypertriglyceridemia.
## CONCLUSIONS
This study among middle and older aged Korean adults found that the “fruit and whole‐grain” dietary pattern was associated with a significantly low prevalence of hypertriglyceridemia in men, whereas the effect was only marginal in women. Our findings indicate that a high proportion of fruit and whole grain in a diet may prevent hypertriglyceridemia in middle and older aged Korean adults.
## FUNDING INFORMATION
This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MEST) (No. 2020R1C1C1014286).
## CONFLICT OF INTEREST
The authors declare that they do not have any conflict of interest.
## ETHICAL APPROVAL
This study utilized a cross‐sectional and nationally representative survey, the Korea National Health and Nutrition Examination Survey (KNHANES) 2013–2018 database. The KNHANES was approved by the Institutional Review Board of The Korea Disease Control and Prevention Agency (2013‐07CON‐03‐4C, 2013‐12EXP‐03‐5C, and 2018‐01‐03‐P‐A). Written informed consent was obtained from each participant when the surveys were conducted. This study did not require additional review and approval because the KNHANES database is publicly available.
## DATA AVAILABILITY STATEMENT
Data are available in a public, open access repository. Website: https://knhanes.kdca.go.kr/knhanes/eng/index.do.
## References
1. Brunzell J. D.. **Clinical practice. Hypertriglyceridemia**. *New England Journal of Medicine* (2007) **357** 1009-1017. DOI: 10.1056/NEJMcp070061
2. Colditz G. A., Manson J. E., Stampfer M. J., Rosner B., Willett W. C., Speizer F. E.. **Diet and risk of clinical diabetes in women**. *American Journal of Clinical Nutrition* (1992) **55** 1018-1023. DOI: 10.1093/ajcn/55.5.1018
3. **2015 Korean guidelines for the management of dyslipidemia: Executive summary (English translation)**. *Korean Circulation Journal* (2016) **46** 275-306. DOI: 10.4070/kcj.2016.46.3.275
4. D’Agostino R. B., Hamman R. F., Karter A. J., Mykkanen L., Wagenknecht L. E., Haffner S. M.. **Cardiovascular disease risk factors predict the development of type 2 diabetes: The insulin resistance atherosclerosis study**. *Diabetes Care* (2004) **27** 2234-2240. DOI: 10.2337/diacare.27.9.2234
5. Eshak E. S., Iso H., Yamagishi K., Kokubo Y., Saito I., Yatsuya H., Sawada N., Inoue M., Tsugane S.. **Rice consumption is not associated with risk of cardiovascular disease morbidity or mortality in Japanese men and women: A large population‐based, prospective cohort study**. *American Journal of Clinical Nutrition* (2014) **100** 199-207. DOI: 10.3945/ajcn.113.079038
6. Fabiani R., Naldini G., Chiavarini M.. **Dietary patterns and metabolic syndrome in adult subjects: A systematic review and meta‐analysis**. *Nutrients* (2019) **11** 2056. DOI: 10.3390/nu11092056
7. Frayn K. N.. **Insulin resistance, impaired postprandial lipid metabolism and abdominal obesity. A deadly triad**. *Medical Principles and Practice* (2002) **11** 31-40. DOI: 10.1159/000066414
8. Giacco R., Clemente G., Cipriano D., Luongo D., Viscovo D., Patti L., Di Marino L., Giacco A., Naviglio D., Bianchi M. A., Ciati R., Brighenti F., Rivellese A. A., Riccardi G.. **Effects of the regular consumption of wholemeal wheat foods on cardiovascular risk factors in healthy people**. *Nutrition, Metabolism, and Cardiovascular Diseases* (2010) **20** 186-194. DOI: 10.1016/j.numecd.2009.03.025
9. Giacco R., Costabile G., Della Pepa G., Anniballi G., Griffo E., Mangione A., Cipriano P., Viscovo D., Clemente G., Landberg R., Pacini G., Rivellese A. A., Riccardi G.. **A whole‐grain cereal‐based diet lowers postprandial plasma insulin and triglyceride levels in individuals with metabolic syndrome**. *Nutrition, Metabolism, and Cardiovascular Diseases* (2014) **24** 837-844. DOI: 10.1016/j.numecd.2014.01.007
10. Gorinstein S., Caspi A., Libman I., Lerner H. T., Huang D., Leontowicz H., Leontowicz M., Tashma Z., Katrich E., Feng S., Trakhtenberg S.. **Red grapefruit positively influences serum triglyceride level in patients suffering from coronary atherosclerosis: Studies in vitro and in humans**. *Journal of Agricultural and Food Chemistry* (2006) **54** 1887-1892. DOI: 10.1021/jf058171g
11. Guasch‐Ferre M., Satija A., Blondin S. A., Janiszewski M., Emlen E., O'Connor L. E., Campbell W. W., Hu F. B., Willett W. C., Stampfer M. J.. **Meta‐analysis of randomized controlled trials of red meat consumption in comparison with various comparison diets on cardiovascular risk factors**. *Circulation* (2019) **139** 1828-1845. PMID: 30958719
12. Hollaender P. L., Ross A. B., Kristensen M.. **Whole‐grain and blood lipid changes in apparently healthy adults: A systematic review and meta‐analysis of randomized controlled studies**. *The American Journal of Clinical Nutrition* (2015) **102** 556-572. DOI: 10.3945/ajcn.115.109165
13. Hu F. B.. **Dietary pattern analysis: A new direction in nutritional epidemiology**. *Current Opinion in Lipidology* (2002) **13** 3-9. DOI: 10.1097/00041433-200202000-00002
14. Institute of Medicine
. (2005). Dietary reference intakes for energy, carbohydrate, fiber, fat, fatty acids, cholesterol, protein, and amino acids. The National Academies Press.. *Dietary reference intakes for energy, carbohydrate, fiber, fat, fatty acids, cholesterol, protein, and amino acids* (2005)
15. Kim J., Jo I.. **Grains, vegetables, and fish dietary pattern is inversely associated with the risk of metabolic syndrome in south korean adults**. *Journal of the American Dietetic Association* (2011) **111** 1141-1149. DOI: 10.1016/j.jada.2011.05.001
16. Kim Y., Keogh J. B., Clifton P. M.. **Differential effects of red meat/refined grain diet and dairy/chicken/nuts/whole grain diet on glucose, insulin and triglyceride in a randomized crossover study**. *Nutrients* (2016) **8** 687. DOI: 10.3390/nu8110687
17. Kodama S., Horikawa C., Fujihara K., Ishii D., Hatta M., Takeda Y., Kitazawa M., Matsubayashi Y., Shimano H., Kato K., Tanaka S., Sone H.. **Relationship between intake of fruit separately from vegetables and triglycerides ‐ A meta‐analysis**. *Clinical Nutrition ESPEN* (2018) **27** 53-58. DOI: 10.1016/j.clnesp.2018.07.001
18. Lim J. H., Lee Y. S., Chang H. C., Moon M. K., Song Y.. **Association between dietary patterns and blood lipid profiles in Korean adults with type 2 diabetes**. *Journal of Korean Medical Science* (2011) **26** 1201-1208. DOI: 10.3346/jkms.2011.26.9.1201
19. Lim M., Kim J.. **Association between fruit and vegetable consumption and risk of metabolic syndrome determined using the Korean genome and epidemiology study (KoGES)**. *European Journal of Nutrition* (2020) **59** 1667-1678. DOI: 10.1007/s00394-019-02021-5
20. Lloyd‐Jones D. M., Hong Y., Labarthe D., Mozaffarian D., Appel L. J., Van Horn L., Greenlund K., Daniels S., Nichol G., Tomaselli G. F., Arnett D. K., Fonarow G. C., Ho P. M., Lauer M. S., Masoudi F. A., Robertson R. M., Roger V., Schwamm L. H., Sorlie P.. **Defining and setting national goals for cardiovascular health promotion and disease reduction: The American Heart Association's strategic impact goal through 2020 and beyond**. *Circulation* (2010) **121** 586-613. PMID: 20089546
21. Martinez M. E., Marshall J. R., Sechrest L.. **Invited commentary: Factor analysis and the search for objectivity**. *American Journal of Epidemiology* (1998) **148** 17-19. DOI: 10.1093/oxfordjournals.aje.a009552
22. Miller M., Stone N. J., Ballantyne C., Bittner V., Criqui M. H., Ginsberg H. N., Goldberg A. C., Howard W. J., Jacobson M. S., Kris‐Etherton P. M., Lennie T. A., Levi M., Mazzone T., Pennathur S.. **Triglycerides and cardiovascular disease: A scientific statement from the American Heart Association**. *Circulation* (2011) **123** 2292-2333. DOI: 10.1161/CIR.0b013e3182160726
23. Ministry of Health and Welfare, Korea Disease Control and Prevention Agency
. (2020). Korea health statistics 2019: Korea National Health and nutrition examination survey (KNHANES VII‐1). Ministry of Health and Welfare, Korea Disease Control and Prevention Agency.. *Korea health statistics 2019: Korea National Health and nutrition examination survey (KNHANES VII‐1)* (2020)
24. Panagiotakos D. B., Pitsavos C., Skoumas Y., Stefanadis C.. **The association between food patterns and the metabolic syndrome using principal components analysis: The ATTICA study**. *Journal of the American Dietetic Association* (2007) **107** 979-987. DOI: 10.1016/j.jada.2007.03.006
25. Public Health England
. (2020). National diet and nutrition survey rolling programme years 9 to 11 (2016/2017 to 2018/2019). Public Health England.. *National diet and nutrition survey rolling programme years 9 to 11 (2016/2017 to 2018/2019)* (2020)
26. Rebello S. A., Koh H., Chen C., Naidoo N., Odegaard A. O., Koh W. P., Butler L. M., Yuan J. M., van Dam R. M.. **Amount, type, and sources of carbohydrates in relation to ischemic heart disease mortality in a Chinese population: A prospective cohort study**. *The American Journal of Clinical Nutrition* (2014) **100** 53-64. DOI: 10.3945/ajcn.113.076273
27. Salehi‐Abargouei A., Maghsoudi Z., Shirani F., Azadbakht L.. **Effects of Dietary Approaches to Stop Hypertension (DASH)‐style diet on fatal or nonfatal cardiovascular diseases–incidence: A systematic review and meta‐analysis on observational prospective studies**. *Nutrition* (2013) **29** 611-618. DOI: 10.1016/j.nut.2012.12.018
28. Schaefer E. J., Gleason J. A., Dansinger M. L.. **Dietary fructose and glucose differentially affect lipid and glucose homeostasis**. *The Journal of Nutrition* (2009) **139** 1257 S-1262 S. DOI: 10.3945/jn.108.098186
29. Sharma S. P., Chung H. J., Kim H. J., Hong S. T.. **Paradoxical effects of fruit on obesity**. *Nutrients* (2016) **8** 633. DOI: 10.3390/nu8100633
30. Song S., Song Y.. **Three types of a high‐carbohydrate diet are differently associated with cardiometabolic risk factors in Korean adults**. *European Journal of Nutrition* (2019) **58** 3279-3289. DOI: 10.1007/s00394-018-1871-2
31. Song Y., Joung H.. **A traditional Korean dietary pattern and metabolic syndrome abnormalities**. *Nutrition, Metabolism, and Cardiovascular Diseases* (2012) **22** 456-462. DOI: 10.1016/j.numecd.2010.09.002
32. Takahashi M. M., de Oliveira E. P., Moreto F., Portero‐McLellan K. C., Burini R. C.. **Association of dyslipidemia with intakes of fruit and vegetables and the body fat content of adults clinically selected for a lifestyle modification program**. *Archivos Latinoamericanos de Nutrición* (2010) **60** 148-154. PMID: 21425719
33. Tosh S. M., Bordenave N.. **Emerging science on benefits of whole grain oat and barley and their soluble dietary fibers for heart health, glycemic response, and gut microbiota**. *Nutrition Reviews* (2020) **78** 13-20. DOI: 10.1093/nutrit/nuz085
34. U.S. Department of Agriculture & Agricultural Research Service
. (2020). Energy intakes: Percentages of energy from protein, carbohydrate, fat, and alcohol, by gender and age, what we eat in America, NHANES 2017–2018. U.S. Department of Agriculture & Agricultural Research Service.. *Energy intakes: Percentages of energy from protein, carbohydrate, fat, and alcohol, by gender and age, what we eat in America, NHANES 2017–2018* (2020)
35. Villegas R., Shu X. O., Gao Y. T., Yang G., Elasy T., Li H., Zheng W.. **Vegetable but not fruit consumption reduces the risk of type 2 diabetes in Chinese women**. *The Journal of Nutrition* (2008) **138** 574-580. DOI: 10.1093/jn/138.3.574
36. Whitton C., Rebello S. A., Lee J., Tai E. S., van Dam R. M.. **A healthy Asian a posteriori dietary pattern correlates with a priori dietary patterns and is associated with cardiovascular disease risk factors in a multiethnic Asian population**. *The Journal of Nutrition* (2018) **148** 616-623. DOI: 10.1093/jn/nxy016
37. Willett W., Stampfer M. J.. **Total energy intake: Implications for epidemiologic analyses**. *American Journal of Epidemiology* (1986) **124** 17-27. DOI: 10.1093/oxfordjournals.aje.a114366
38. Williams D. E., Wareham N. J., Cox B. D., Byrne C. D., Hales C. N., Day N. E.. **Frequent salad vegetable consumption is associated with a reduction in the risk of diabetes mellitus**. *Journal of Clinical Epidemiology* (1999) **52** 329-335. DOI: 10.1016/s0895-4356(99)00006-2
39. Yuan C., Lee H. J., Shin H. J., Stampfer M. J., Cho E.. **Fruit and vegetable consumption and hypertriglyceridemia: Korean National Health and nutrition examination surveys (KNHANES) 2007‐2009**. *European Journal of Clinical Nutrition* (2015) **69** 1193-1199. DOI: 10.1038/ejcn.2015.77
40. Zazpe I., Sanchez‐Tainta A., Estruch R., Lamuela‐Raventos R. M., Schroder H., Salas‐Salvado J., Corella D., Fiol M., Gomez‐Gracia E., Aros F., Ros E., Ruíz‐Gutierrez V., Iglesias P., Conde‐Herrera M., Martinez‐Gonzalez M. A.. **A large randomized individual and group intervention conducted by registered dietitians increased adherence to Mediterranean‐type diets: The PREDIMED study**. *Journal of the American Dietetic Association* (2008) **108** 1134-1144. DOI: 10.1016/j.jada.2008.04.011
|
---
title: 'Phytochemical analysis and nephroprotective potential of Ajwa date in doxorubicin‐induced
nephrotoxicity rats: Biochemical and molecular docking approaches'
authors:
- Othman A. S. Baothman
- Hisham N. Altayb
- Mustafa A. Zeyadi
- Salman B. Hosawi
- Mohamed Kamel Abo‐Golayel
journal: Food Science & Nutrition
year: 2023
pmcid: PMC10002938
doi: 10.1002/fsn3.3199
license: CC BY 4.0
---
# Phytochemical analysis and nephroprotective potential of Ajwa date in doxorubicin‐induced nephrotoxicity rats: Biochemical and molecular docking approaches
## Abstract
The purpose of this study is to evaluate the likely defensive impact of Ajwa date aqueous extract (AJDAE) in alleviating the nephrotoxicity generated by doxorubicin (DOX) injection in rats. Sixty male Wister albino rats were randomly and equally separated into six groups ($$n = 10$$), and they were treated as follows: untreated control group, extract groups administered with 0.75 and 1.5 mg kg bw of AJDAE, toxicant control group administered with DOX, and prophylactic groups were treated with 0.75 and 1.5 mg/kg of AJDAE and 15 mg/kg DOX. Biochemical parameters, antioxidant enzymes, renal functions, DNA integrity, and histopathology were studied to evaluate the nephroprotective activity of AJDAE. Furthermore, bioactive compounds were utilized for in silico molecular docking. AJDAE treatment resulted in significant improvements in the amended renal biomarkers (urea, creatinine, calcium, phosphorous, and uric acid), antioxidative markers, and MDA. Noticeable histopathological improvements supported this result. Results of in silico studies revealed that d‐Mannitol, 6TMS derivative, palmitic acid, and TMS derivative had a higher docking score with human soluble epoxide hydrolase (−10.9 kcal/mol) and NF‐κB‐DNA (−7 kcal/mol). The present findings indicated that AJDAE could decrease ROS generation and lipid peroxidation (LPO) and repair the DOX injection‐related DNA damage.
The nephroprotective effect of the Ajwa date extract could be attributed to its potential antioxidant effect that contains phytochemicals with potential therapeutic value for the nephrotoxicity.
## INTRODUCTION
Increasing in prevalence globally, chronic kidney disease is a significant public health concern that can potentially result in kidney failure. Kidney disease entails a progressive reduction in kidney function attributable to nephron decreases (Hrenák et al., 2013). Although it has strong antitumor effectiveness, doxorubicin (DOX) is limited as a chemotherapeutic drug due to its cardiac, pulmonary, testicular, and renal toxicity (Fadillioǧlu et al., 2003; Sabbah et al., 2018). In chronic kidney disease animal models, DOX‐induced nephrotoxicity is renowned and widespread; its detrimental impact is managed through proximal tubules' discriminatory cellular injury via mechanisms that stay as the core (Grant et al., 2019). Despite doxorubicin's acute cellular toxicity mechanism ambiguity, ROS is a primary element of DOX toxicity; thus, the factors that manage this oxidative harm are comprehensively researched (Reddy et al., 2007). DOX injection produces hydrogen peroxide, superoxide anions, and hydroxyl radicals. DOX is transformed into a semiquinone free radical by NADPH‐cytochrome P‐450, which then initiates superoxide anion and hydroxyl radicals' generation, producing LPO (Oz & Ilhan, 2006). Although many studies have examined the DOX‐initiating molecular mechanism, there are gaps surrounding its causal factors. Possibly, the basis of this is DOX producing DOX semiquinone. While the resultant semiquinone radical does not have a long life span, it creates a torrent of reactions producing ROS upon reacting with O2 ROS (El‐Shitany et al., 2008; Liu et al., 2007; Mohan et al., 2010; Rašković et al., 2011). Evidently, using doxorubicin results in improved production of superoxide molecules, hydrogen peroxide, and hydroxyl radicals, which can rapidly interact with membrane lipids, resulting in LPO (Oz & Ilhan, 2006). LPO is a core component of the toxic indicators of DOX administration and is assessed in line with MDA intensities. In rats, DOX‐induced nephrotoxicity exhibits severe renal LPO (Akyol et al., 2014; Yagmurca et al., 2004).
Date palm fruits (DPFs) are eaten in numerous countries and are particularly important in Middle Eastern and North African nations. DPF extracts contain many pharmacological properties, including antiallergic (Karasawa & Otani, 2012), antibacterial (Zhang et al., 2013), antifungal (Boulenouar et al., 2011), antioxidants (El Arem et al., 2014), anticancer (Karasawa & Otani, 2012), anti‐inflammatory (Borochov‐Neori et al., 2015), antimicrobial (Mahdhi et al., 2013), cardioprotective (Mubarak et al., 2018a; Sabbah et al., 2018), immune‐boosting (Karasawa et al., 2011), nephroprotective (Wang et al., 2019), and neurologically protective (Pujari et al., 2011). Additionally, research has shown that DPF extract can essentially hunt for oxidants due to its antioxidants and antimutagenic compounds (Allaith, 2008; Shireen et al., 2008).
Recently, natural antioxidant consumption has generated vast interest as a means of averting oxidative harm in numerous oxidative stress‐related diseases. In nephrotic disease, consumption of medicinal plant nutrients in conventional herbal medicine provides therapeutic benefits. This study was conducted to explore the possibility of prophylactic effect of AJDAE in improving doxorubicin‐induced nephrotoxicity in Wister albino rats. Furthermore, molecular docking was used to identify and compare the interaction between the active compounds and the observed biological activity through molecular docking.
## Ajwa date (Phoenix dactylifera L.) aqueous extract preparation
Ajwa date was bought from the date market in Jeddah, KSA; established and recognized by a professor of plant taxonomy; and banked in the Herbarium of Biological Sciences Department, KAU (specimen voucher number: P. dactylifera L. #PD17569). The flesh of Ajwa date (10 g) was drenched in 100 ml of double distilled water for 18 h and mixed up in a blender at a room temperature (22–25°C) (Abdelaziz & Ali, 2014). Ajwa date aqueous extracts were sieved and centrifuged (Hettich ZENTRIFUGEN D‐7200 Tuttlingen Type 1200 220V Benchtop Centrifuge) at 4°C and 5000 g for 15 min. The supernatants were withdrawn and stored at 4°C till usage, and the precipitant was removed. Fresh AJDAE was prepared daily throughout the experiment days (Vayalil, 2002) just prior oral administration by using intragastric gavage throughout the progress of the experimentation.
## Chemicals
Doxorubicin vial was obtained from EBEWE Pharma Ges.m.b. H “Ebewe” (50 mg/25 ml Nfg. KG, A‐4866Unterach). Urea (URE120240), creatinine (CRE106240), calcium (CAL103120), phosphorous (PH123100), and uric acid (UA121240) were obtained from EGY CHEM LAB Technology. Superoxide dismutase, glutathione peroxidase, glutathione‐S‐transferase, glutathione reductase, catalase, and malonaldehyde competitive enzyme immunoassay technique were purchased from Bioassay Technology Laboratory, Shanghai Korain Biotech Co. DNA tissue extraction kit was provided by Qiagen, DNeasy, RNeasy, QIAGEN Group. Top Vision Agarose‐R0491, PBS‐1314‐87‐0, EDTA‐E1161, Tris Acetate‐EDTA buffer (TEA‐T8280), hematoxylin, and eosin were supplied from Merck, Sigma‐Aldrich.
## Chromatographic analysis of Ajwa date using GC‐MS
Gas chromatography–mass spectrometry (GC‐MS) chromatographic analysis of Ajwa date was performed using Agilent Technologies 7890B GC Systems combined with 5977A Mass Selective Detector. Capillary column (HP‐5MS Capillary; 30.0 m × 0.25 mm ID × 0.25 μm film) and helium as a carrier gas with a rate of flow of 1.7 ml/min with 1 μl injection were also used. Analysis of the sample was carried out withholding the column initially for 4 min at 40°C postinjection, and then the temperature was elevated to 300°C (20°C/min heating ramp) along with a 3.0‐min hold. The injection was done in split‐less mode at 300°C. MS scan range was (m/z) 50–550 atomic mass units under electron impact ionization (70 eV).
## Silylation agent: BSA N, O‐Bis (trimethylsilyl) acetamide
The reaction was carried out by adding 100 μl of BSA plus amount of the sample after extraction and heating in water bath at 70°C for 2 h and then injected into GC‐MS under the above conditions. The constituents were determined by mass fragmentations with the NIST mass spectral search program for the NIST/EPA/NIH mass spectral library (June 2014).
## Experimental design
This study was approved by the Research Ethics Committee (REC), Faculty of Medicine, King Abdul Aziz University (KAU). Two‐month‐old male Wistar rats (150–200 g), which were bred in the animal house of King Fahd Medical Research Center (KFMR), King Abdulaziz University, KSA, were accommodated in an experimental animals care facility, including room temperature (25 ± 1°C), 12‐h light/dark cycles, and suitable humidity, and they were allowed free path to a standard pellet diet and tap water. Prior starting the experiment, rats were kept for 7 days to familiarize the surrounding environment in stainless‐steel mesh‐covered plain polypropylene cages. The experiment was approved, and the rats had animal carefulness according to the guidelines of the Committee for the Purpose of Control and Supervision of Experiments on Animals (CPCSEA), Government of KSA. Sixty male Wistar rats were randomly and equally distributed into six groups ($$n = 10$$). Group 1 received only a standard pellet diet. Group 2 received an oral prophylactic dose of 2 ml of AJDAE (0.75 mg/kg bw) daily using intragastric gavage for 30 days according to Vayalil [2002] and Mubarak et al. ( 2018b) with modifications. Group 3 received an oral prophylactic dose of 4 ml of AJDAE (1.5 mg/kg bw) daily using intragastric gavage for 30 days. Group 4 was intraperitoneally injected with a single dose of DOX (15 mg/kg, i.p.) at the end of the 28th day of the study to induce a nephrotoxic injury (Ellison, 2002). Group 5 was intraperitoneally injected with a single dose of DOX (15 mg/kg, i.p.) at the end of the 28th day of the study and received 2 ml of AJDAE (0.75 mg/kg bw) daily for 30 days using intragastric gavage according to Vayalil [2002] and Mubarak et al. ( 2018b). Group 6 was intraperitoneally injected with a single dose of DOX (15 mg/kg, i.p.) at the end of the 28th day of the study and received 4 ml of AJDAE (1.5 mg/kg bw) daily for 30 days using intragastric gavage according to Vayalil [2002] and Mubarak et al. ( 2018b).
## Sample collection
Rats were left fasting 12 h before sampling and decapitation. Twenty‐four hours after intraperitoneal injection of DOX, rats were anesthetized using pentobarbitone sodium (60 mg/kg), and then, blood specimen from each rat was withdrawn via the optic vein, saved in a centrifuge tube, and remained at room temperature for 20 min. Sera were obtained by centrifuging tubes at 4500 g for 10 min using cooling centrifuge. Serum samples were utilized for determination of serum urea, creatinine, calcium, phosphorous, and uric acid by direct colorimetric method. Then, each rat's abdomen was dissected; the left kidney was excised and split up into three specimens. One kidney specimen was submerged instantly into $10\%$ buffered solution of neutral formaldehyde and handled for histopathological inspection, the second kidney specimen was utilized for DNA extraction, and the third kidney specimen was employed for preparation of kidney tissue homogenate.
## Preparation of kidney tissue homogenates
The homogenization of left kidneys' tissue was performed instantly following kidney tissue excision in a Teflon‐glass homogenizer. The kidney tissue specimens were maintained at 2–8°C in a bucket containing ice. A 200 mg weight kidney tissue specimen was excised from the left kidney of each studied rat and submerged in 2 ml of PBS/1 mM EDTA. The tissue specimens were homogenized completely and kept for one round of freezing at −80°C in a deep freezer. Using a cooling centrifuge, the homogenate samples were centrifuged at 18,000 g (+4°C) for 30 min. The supernatants of the homogenized kidney tissue samples were assembled instantly, distributed in Eppendorf tubes, and preserved at −80°C ready for use (Sabbah et al., 2018).
## Oxidative and antioxidative markers
The activities of superoxide dismutase (SOD), glutathione peroxidase (GPx), and kidney tissue homogenates were assayed according to the methodology of Madkour & Abdel‐Daim [2013] and Mubarak et al. ( 2018b). Glutathione‐S‐transferase (GST), glutathione reductase (GR), and catalase (CAT) kidney tissue homogenates were estimated according to the methods of Khan and Sultana [2009]. Malonaldehyde (MDA) was estimated in the kidney tissue homogenates according to the method of Ohkawa et al. [ 1979] that was modified by Mubarak et al. ( 2018b).
## Detection of genomic DNA of rats' kidney abnormality
For studying the kidneys' tissue genomic DNA integrity of all the studied rats and according to the purification protocol of total DNA, the DNA extraction of each rat's kidney tissue specimen was performed using the QIAGEN tissue extraction kit (Sabbah et al., 2018). Preparation of agarose gel of molecular biology grade ($2\%$ agarose gel in 1× TAE buffer) was arranged in accordance with Kumar Gothwal et al. [ 2007]. Gel electrophoresis was conducted at 100 V constant potential difference for 1 h. DNA fragments were pictured by UVI tech. photo‐documentation system.
## Tissue preparation procedure
Histopathological evaluation was performed on all candidate rats in this study (control and tested groups) for renal tissue sections. All studied renal tissue samples were excised in compliance to the timings of the study design according to the planned scarification schedule, followed by fixation in $10\%$ neutral buffered formalin solution, routinely processed, embedded in paraffin, sectioned at 4 μm thickness, and finally stained with hematoxylin and eosin (H&E).
## Hematoxylin and eosin staining procedure
Heating for 1 h in a 60°C oven preceded the staining step for tissue fixation on the slide. After xylene deparaffinization and rehydration in grades alcohol (absolute ethanol, $90\%$ ethanol, and $70\%$ ethanol), the kidney sections were stained with hematoxylin then further washed in running tap water until the sections were blue, followed by eosin staining. Slides were then dipped in $90\%$ ethanol once, transferred to absolute alcohol. Finally, the sections were cleared in 2 changes of xylene, mounted using Canada balsam, and covered with clean glass slide covers.
## Evaluation procedure
Each representative H&E‐stained slide was thoroughly reviewed by the pathologist (LSS) at low power examination for screening as well as higher power magnification for further characterization. Any observed morphological alterations were recorded, and they were compared with the control group for reference evaluation. Histopathological findings were evaluated in a modified semiquantitative four‐tier scoring system focusing on glomerular injury, tubular cyst/cast formation, and interstitial inflammation, and other abnormal observations were also considered when encountered (Zheng et al., 2005). Microscopic variations were assessed as tabulated in Table 1. Finally, a comparative analysis of data was done.
**TABLE 1**
| Glomerular injury, tubular cyst cast formation, atrophy or dysplasia, and interstitial inflammation | Glomerular injury, tubular cyst cast formation, atrophy or dysplasia, and interstitial inflammation.1 |
| --- | --- |
| 0 | No disease |
| 1 | 1%–25% of tissue affected |
| 2 | 26%–50% of tissue affected |
| 3 | 51%–75% of tissue affected |
| 4 | 76%–100% of tissue affected |
## Ligand preparation
The structures of GC‐MS identified bioactive compounds and the antioxidant silymarin which is used as a standard for molecular docking were obtained from NCBI PubChem database and NIST Chemistry WebBook. Energy minimization for the compounds was done by MOE software to remove clashes.
## Selection and preparation of proteins
Two proteins (NF‐κB‐DNA and human soluble epoxide hydrolase) were chosen, which have a linkage with the renal health (Liu, 2019; Tamada et al., 2003). The 3D structures of NF‐κB‐DNA (PDB ID:1NFK) and human soluble epoxide hydrolase (PDB ID:3ANS) were obtained from (RCSB) Protein Data Bank (PDB). The synthetic inhibitors and water molecules were removed from the proteins structures. Energy minimization, protonation, and addition of hydrogen atoms were performed for the 3D structures of proteins by MOE software. The active site residues were determined by MOE site finder according to previous publications (Kpemissi et al., 2019).
## Molecular docking
The molecular docking experiment was done to generate the best binding affinity between proteins and ligands. The proteins were set rigid, and ligands were flexible. Conformations of different ligands with target protein were generated, and the best docking pose with the least binding energy was selected for the prediction of the interacting residues and bond types using Discovery Studio (Biovia, 2017).
## Statistical analysis
All obtained data analyses were stated as mean ± standard error by one‐way analysis of variance using SPSS 21. Illustrating the differences between the means, the t‐test of significance was verified and the difference was deemed statistically significant when p ≤.05.
## GC‐MS analysis of Ajwa date extract
Of 39 phytoconstituents of Ajwa dates, 18 key compound peaks were acquired via pairing the ingredients' mass spectra with the NIST library, as shown in Figure 1. These compounds were glycerol, l‐threitol, 4TMS derivative, l‐(+)‐threose, tris(trimethylsilyl) ether, trimethylsilyloxime, d‐(+)‐Arabitol, 2‐pentenedioic acid, 2‐[(trimethylsilyl)oxy]‐,bis(trimethylsilyl) ester, d‐(−)‐tagatofuranose, pentakis(trimethylsilyl) ether (isomer 1), d‐Pinitol, pentakis (trimethylsilyl) ether, d‐Sorbitol, 6TMS derivative, d‐(+)‐Galactose, pentakis (trimethylsilyl) ether, pentafluorobenzyloxime (isomer 1), d‐Mannitol, 6TMS derivative, d‐glucopyranose, 5TMS derivative, l‐(+)‐Tartaric acid, 4TMS derivative, palmitic acid, TMS derivative, 3‐Heptadecen‐5‐yne, (Z), stearic acid, 9‐Octadecenoic acid, (E)‐TMS derivative, fumaric acid, di (2‐propylphenyl) ester, and α‐linoleic acid. Table 2 provides the identified compounds' chemical/formulae, M/Z ratio, molecular weight, peak area, and retention time.
**FIGURE 1:** *GC‐MS analysis of Ajwa date extract.* TABLE_PLACEHOLDER:TABLE 2
## Group renal function profiles
Group 4 demonstrated a greater elevation ($p \leq .01$) in calcium, creatinine, phosphorus, serum urea, and uric acid than group 1 (Table 3). Additionally, comparison between the renal profiles' serum levels in groups 2 and 3 and the control group 1 was insignificant ($p \leq .01$). Further comparison with the rats administered with DOX showed significant constraint ($p \leq .01$) at both AJDAE levels for calcium (−$31.78\%$, −$31.71\%$), creatinine (−$50.91\%$, −$23.19\%$), phosphorus (−$40.58\%$, −$51.59\%$), serum urea (−$24.36\%$, −$39.9\%$), and uric acid (−$24.35\%$, −$36.43\%$). Also, comparison between the serum level fluctuations in groups 3 and 2 was insignificant ($p \leq .01$) (Table 3).
**TABLE 3**
| Groups | Urea, mg/dl | Creatinine, mg/dl | Ca, mg/dl | Ph, mg/dl | UA, mg/dl |
| --- | --- | --- | --- | --- | --- |
| Control | 32.52 ± 3.74 | 0.24 ± 0.04 | 5.42 ± 0.28 | 5.95 ± 0.37 | 2.34 ± 0.40 |
| AJDAE (0.75 g/kg bw) | 30.35 ± 2.60 | 0.27 ± 0.04 | 5.52 ± 0.12 | 5.79 ± 0.41 | 2.32 ± 0.19 |
| AJDAE (1.5 g/kg bw) | 29.52 ± 3.02 | 0.28 ± 0.02 | 5.55 ± 0.12 | 5.63 ± 0.21 | 2.28 ± 0.30 |
| DOX | 50.43 ± 4.23*** | 0.59 ± 0.05*** | 11.63 ± 1.08*** | 8.76 ± 0.97*** | 3.87 ± 0.34*** |
| DOX + AJDAE (0.75 g/kg bw) | 38.14 ± 3.22### | 0.35 ± 0.03### | 5.71 ± 0.15### | 6.73 ± 0.63### | 2.64 ± 0.27### |
| DOX + AJDAE (1.5 g/kg bw) | 34.43 ± 4.05### | 0.35 ± 0.03### | 5.63 ± 0.18### | 6.63 ± 0.25### | 2.46 ± 0.25### |
## Antioxidant markers and LPO
Group 4 demonstrated significant depletion ($p \leq .01$) in the kidney homogenate levels of antioxidant markers (SOD, GR, GST, GPx, and CAT) and significant elevation ($p \leq .01$) of MDA levels in the kidney tissue homogenate compared with group 1 (Figure 2).
**FIGURE 2:** *Effects of AJDAE on tissue SOD, GR, GST, GPx, CAT, and MDA levels in DOX‐treated rats. The values are mean ± SEM (n = 10). Statistical analysis was calculated via t‐test analysis. For estimation of p values, DOX‐treated group was compared with the control group, and AJDAE‐protected groups were compared with the DOX‐treated group. All data were characterized as mean ± SEM. Statistical data were tested using t‐test, and differences were expressed at p < .05, p < .01, and p < .0001 as indicated by (*), (**), and (***) compared with normal control and (#), (##), and (###) compared with DOX‐treated group.*
Fluctuations in the kidney tissue homogenate levels of antioxidant markers and MDA of groups 2 and 3 were insignificant ($p \leq .05$) when compared with group 1 (Figure 2).
When compared with the kidney tissue homogenate levels of group 4, those of groups 2 and 3 showed significantly increased ($p \leq .01$) depletion for SOD ($31\%$, $63\%$), GR ($41\%$, $62\%$), GST ($26\%$, $35\%$), GPx ($67\%$, $41\%$), and CAT ($71\%$, $30\%$) (Figure 2). Further comparisons of the same groups revealed significantly decreased ($p \leq .01$) elevation of the kidney tissue homogenate levels of MDA (−$47\%$, −$60\%$).
Additionally, comparisons of the changes in the kidney tissue homogenate levels of SOD, GR, GST, GPX, CAT, and MDA of groups 3 and 2 were insignificant ($p \leq .01$) (Figure 2).
## Electrophoretic pattern of the groups' DNA
DNA extracted from the kidney tissues revealed a variety of banding forms (Figure 3). Group 1's genomic DNA presented a unique sharp band with no disintegration and tail formation. Groups 4's genomic DNA showed an entirely diverse style of banding; a classical band DNA fragmentation was identified that was not in Group 1. Groups 5 and 6 showed substantial kidney‐DNA recovery. Groups 2 and 3 did not exhibit any kidney tissue DNA disintegration.
**FIGURE 3:** *DNA fragmentation of rats treated with different concentrations of AJDAE following DOX‐induced nephrotoxicity. Lane M is a DNA marker with 10,000 bp. Lane 1 is normal group. Lane 2 is AJDAE group (0.75 g/kg bw). Lane 3 is AJDAE group (1.5 g/kg bw). Lanes 4 and 5 are fragmented DNA streaks (DOX‐treated group). Lanes 6 and 7 are DNA of rats' kidneys (0.75 g/kg bw of AJDAE + DOX group). Lanes 8 and 9 are DNA of rats' kidneys (1.5 g/kg bw of AJDAE protected + DOX group).*
## Histopathology results
Some segments of group 1 were unexceptional and were established as reference tissue for the comparative evaluations. Segments of groups 2 and 3 displayed slight fluctuations, primarily in terms of endocapillary glomerular proliferation (Figure 4a). Group 4 tissue sections showed varying degrees of glomerular damage, such as mesangial hyperplasia and segmental sclerotic alterations (Figure 4b), and atrophic glomeruli alongside compensatory hyperplasia and neutrophilic glomerular proliferation (Figure 4c). Tubular changes encompassed hyaline change and cystic tubular atrophy (Figure 4d), but one segment demonstrated focal tubular epithelial dysplasia (Figure 4e). This entire group demonstrated remarkable effects in over $50\%$ of the renal tissue sections investigated; only this group recorded a score of 4; the other five groups had no record. Measurably, the alterations in groups 5 and 6 were weaker than those in group 4. However, a score of 3 was recorded in a single specimen for each member of groups 5 and 6 (Figure 4f). Moreover, a minor swing toward normalcy was observed in group 6 compared with group 5, with a $100\%$ increase in renal sections demonstrating <$25\%$ change. Conversely, the extreme pathological changes in renal tissue observed in group 4 (atrophy and dysplasia) were not present in groups 5 and 6. These adjustments were mesangial proliferation, some retained tubular hyaline fluctuations, and a lack of interstitial inflammation. Also the more grave injurious manifestations (cystic strophic dilation and tubular epithelial dysplasia) were exceptionally detected in group 4 (Table 4).
**FIGURE 4:** *(a) Representative renal section in group 2 showing focal glomerular proliferation. Same changes were observed in group 3 (H&E × 400). (b) Renal section in group 4 showing interstitial inflammation (arrowhead) and focal tubular hyaline cast formation (arrow) (H&E × 400). (c) Renal tissue section in group 4 showing focal glomerular atrophy and sclerosis (H&E × 100). (d) Renal section in group 4 showing tubular cysts formation (H&E × 200). (e) Renal section in group 4 showing focal dysplastic tubular epithelium and pronounced nearby hyaline changes along with interstitial inflammation (H&E × 400). (f) Renal section in group 5 showing glomerular mesangial proliferation and tubular hyaline change (H&E × 400).* TABLE_PLACEHOLDER:TABLE 4
## In silico studies
GC‐MS analysis of proteins and bioactive compounds' interaction showed that these ligands have diverse binding affinity to the chosen proteins. The strongest binding affinity was noted between the D‐Mannitol, 6TMS derivative, and the human soluble epoxide hydrolase (PDB ID:3ANS) protein (Figure 5). Additionally, the top docking energy of −7 kcal/mol was observed in palmitic acid and TMS derivative, with NF‐κB‐DNA (PDB ID:1NFK) (Figure 6 and Table 5).
**FIGURE 5:** *3D and 2D interaction of D‐Mannitol, 6TMS derivative, and human soluble epoxide hydrolase (PDB ID:3ANS).* **FIGURE 6:** *3D and 2D interaction of palmitic acid, TMS derivative, and NF‐κB‐DNA (PDB ID:1NFK).* TABLE_PLACEHOLDER:TABLE 5
## DISCUSSION
This study examines the AJDAE's nephroprotective impact on DOX‐induced nephrotic harm associated with oxidative stress, production of free radicals, and histopathological studies, confirming that an oxidative stress injury causes such effects.
This study verified DOX‐induced nephropathy, as evidenced by a significant increase in the levels of serum of urea, calcium, creatinine, phosphorus, and uric acid and supported by toxic histopathological fluctuations in noncontrol groups, providing quantitative and qualitative proof. For instance, group 4 administered with DOX without AJDAE protection demonstrated the highest nephrotoxicity score [4]. The severe injuries (tubular epithelial dysplasia and cystic atrophic dilation) and interstitial inflammatory reactions occurred solely in this group. These findings correspond with existing research (Ayla et al., 2011; Refaie et al., 2016).
The GC‐MS results underscore that AJDAE is comprised of several phytoconstituents with antioxidant activities, including fumaric, linoleic, palmitic, and stearic acids (Table 2). These results correspond with previous research maintaining that Ajwa date contains many quantities of the aforementioned acids, typified by their antioxidant activity (Al‐Farsi & Lee, 2008a, 2008b; Ashrafian et al., 2012; Hayes et al., 2010; Nehdi et al., 2010). The AJDAE GC‐MS results confirm octadecanoic acid, an effectual anti‐inflammatory tool. It is especially beneficial for subduing the pro‐inflammatory signaling that leads to decreases in cytokines expression and proinflammatory mediators and inflammatory‐related ailments (Kang et al., 2018). The most common saturated free fatty acid is palmitic acid, a significant anti‐cancerous tool and potentially beneficial for human breast cancer treatment (Zafaryab et al., 2019). This is consistent with docking results that found palmitic acid demonstrates a strong inhibitory effect on NF‐κB protein. Additionally, stearic acid can better protect cortical neurons against oxidative harm by improving the endogenous antioxidant enzymatic system (Wang et al., 2007).
According to GC‐MS results, α‐linolenic acid is an active AJDAE ingredient. It can decrease the pervasiveness of chronic renal disorders, and it contributes to re‐establishing a healthy renal function (Gopinath et al., 2011). It aids with the prevention and management of other ailments, including autoimmune disease, ischemic heart disease, and strokes (Calder & Yaqoob, 2009; Connor, 2000). Furthermore, it has been confirmed that α‐linolenic acid decreases renal oxidative stress, and a synergistic effect was observed in its isomers (Saha & Ghosh, 2013). Kidney impairment diagnoses are aided by key biomarkers, urea, and creatinine (Khan & Sultana, 2004; Mohan et al., 2010). The findings agree with previous research that reported elevations in renal biomarkers' serum values (creatinine, urea, and uric acid) are related to tubular blockade and weakened renal architecture (Afsar et al., 2020; El‐Sheikh et al., 2012). Intraperitoneal DOX‐injected rats demonstrated some renal function issues, verifying that this drug can avert tubular cells' protein synthesis or improve renal tissues' LPO and free radical production (Naqshbandi et al., 2012).
Orally administered AJDAE could lessen the nephrotoxic impact generated by intraperitoneal DOX injection through substantial decreases in the serum values of calcium, creatinine, phosphorus, urea, and uric acid, compared with the DOX‐administered groups' renal function improving toward a normal level. This is attributable to DPFs antioxidant capability and anti‐inflammatory impacts, which decreases DOX‐related oxidative stress, inflammation, and tissue harm. DPF aqueous extracts contain high concentrations of polyphenolic elements that help prevent kidney intoxication and significantly improve the elevated levels of creatinine and urea caused by a range of chemotherapeutic medications (Abdelaziz et al., 2015; Yasin et al., 2015).
The serological and biochemical results, confirmed at the histopathological scope, explained the AJDAE protection. Although groups 4, 5, and 6 continued to show DOX‐induced oxidative cell injury, the AJDAE showed a clear antidotal effect on the nephrotoxicity in groups 5 and 6 only, as confirmed by the degeneration in the injury scope (score 4 was not documented; most recorded at score ≤ 2). This highlights AJDAE's protective role in combatting DOX nephrotoxicity, which is possibly due to its antioxidant, anti‐inflammatory, and regenerative impact, as confirmed in the past studies (Al‐Asmari et al., 2020; Younas et al., 2020).
In this study, DOX intoxication in rats resulted in raised levels of serum uric acid, conflicting with the results of Salah et al. [ 2012], which showed a reduction in plasma uric acid following dimethoate poisoning in rats. Furthermore, the increased uric acid level signified vascular disease composed of thickened preglomerular arteries and spread of smooth muscle cell (Kang et al., 2002). Blood uric acid is a strong antioxidant that is highly effective in foraging singlet oxygen and free radicals (Ames et al., 1981). Renal oxidative injury could result in damaged renal function.
This study noted that intraperitoneal DOX injections induced nephrotoxicity in rats because of oxidative stress and production of ROS, resulting in substantial decreases in the values of antioxidant enzymes SOD, GR, GST, GPx, and CAT. Additionally, the MDA serum level was substantially elevated compared with the normal control, along with tubular atrophy and raised glomerular capillary permeability. These latter results correspond with other research proposing that kidney intoxication is due to LPO, with the ensuing biological macromolecules injury through iron‐dependent oxidative harm and degenerative changes in the kidney depend on doses amassed and length of treatment, as DOX metabolites are partly expelled via the kidney (Wang et al., 2019). Another mechanism for the potential DOX‐related kidney harm is the transforming of DOX to semiquinone free radical through NADPH‐cytochrome P‐450, which generates superoxide anion and hydroxyl radical that triggers LPO (Rashid et al., 2013).
A substantial reduction in the CAT level in the nephrotic tissue of the DOX‐intoxicated group was found compared with the control group, which agrees with previous research that documented decreased CAT antioxidant activity with DOX treatment in rats (Ayla et al., 2011).
Orally administered AJDAE generated substantial improvement in the antioxidants' enzyme activities, alongside a significant decrease in the kidneys' MDA levels. This agrees with other research that identified substantial improvement of antioxidant enzyme activities for SOD, CAT, GST, and GSH level (Abdelaziz et al., 2015; El‐Far et al., 2016; Sahyon & Al‐Harbi, 2020). The improvement was attributable to DPF extracts' effectiveness as an antioxidant (El‐Far et al., 2016).
AJDAE's positive effects could be due to its active bio‐constituents capacity to remove free radicals and prevent LPO. DPF can chelate superoxide and hydroxyl radicals and very effectively suppresses macromolecule harm, such as LPO and protein oxidation in vitro (Vayalil, 2012). Another in vitro study showed that the flavonoid glycoside content of date extract is effectual as an LPO inhibitor (Zhang et al., 2013). Additionally, various anthocyanins, flavonoids, and phenolic compounds were established as protective agents for the kidney because of LPO harm (Allaith, 2008; Pandey & Rizvi, 2009; Sandhar et al., 2011).
This study identified a classical band disintegration of genomic DNA in the DOX‐intoxicated rats, but not in the control group. Groups 5 and 6 showed substantial recovery in kidney DNA. Groups 2 and 3 did not show any kidney tissue DNA disintegrations. These molecular findings correspond with other research that found that intraperitoneal DOX‐injected rats resulted in a classical fragmentation of DNA band not identified in the control group (Afsar et al., 2020).
DPF date has proved effective in decreasing micronuclei reductions, which is widely considered an indicator of improved DNA restoration in cells, or due to cell death or apoptosis of extreme DNA harm (Vukicevic et al., 2004). A previous study showed that DPF extract affects cellular recovery speed by regenerating the injured DNA areas (Diab & Aboul‐Ela, 2012).
The current histopathological findings verified that AJDAE treatment restored the tissue damaged by DOX injection, which agrees with recent research (Wang et al., 2019) in which the histopathological investigation of the kidney showed damaged tissue following DOX injection. Furthermore, microscopic scrutinization of tissue sections identified substantial enhancement in rats administered with AJDAE by increasing doses of 1 and 1.5 g.
Molecular docking established that d‐Mannitol and 6TMS derivative demonstrated high activity (docking energy recorded at −10.9 kcal/mol) on soluble epoxide hydrolase (sEH) enzyme. This enzyme is common in human tissues and is responsible for hydrolysis in many (Enayetallah et al., 2004), including renal tissues (Parrish et al., 2009). Additionally, as the GC‐MS process established, most bioactive compounds present in Ajwa extract demonstrated stronger binding affinity with human soluble epoxide hydrolase than the regular Silymarin (−7.8 kcal/mol).
## CONCLUSIONS
The findings show that AJDAE can combat DOX toxicity; improve the renal histology and serum benefits of calcium, urea, creatinine, phosphorous, and uric acid; and stop further renal injury. AJDAE's protective effect could be attributable to its antioxidant impact through free radical chelating actions, declining inflammation, and membrane‐stabilizing activity. Potentially, Ajwa dates are therapeutically valuable for tackling doxorubicin nephrotoxicity.
## CONFLICT OF INTEREST
The authors declare no conflict of interest.
## ETHICAL STATEMENT
Ethical approval for this research was obtained from the Research Ethics Committee of Faculty of Medicine, King Abdulaziz University (reference no. 442‐16). The rats received animal care according to the guidelines of the Committee for the purpose of Control and Supervision of Experiments on Animals, KSA.
## DATA AVAILABILITY STATEMENT
All data analyzed during this study are included in this published article.
## References
1. Abdelaziz D. H., Ali S. A.. **The protective effect of**. *Journal of Ethnopharmacology* (2014) **155** 736-743. PMID: 24945397
2. Abdelaziz D. H., Ali S. A., Mostafa M. M.. **Phoenix dactylifera seeds ameliorate early diabetic complications in streptozotocin‐induced diabetic rats**. *Pharmaceutical Biology* (2015) **53** 792-799. DOI: 10.3109/13880209.2014.942790
3. Afsar T., Razak S., Almajwal A., Aldisi D.. **Doxorubicin‐induced alterations in kidney functioning, oxidative stress, DNA damage, and renal tissue morphology; improvement by**. *Saudi Journal of Biological Sciences* (2020) **27** 2251-2260. DOI: 10.1016/j.sjbs.2020.07.011
4. Akyol S., Ugurcu V., Altuntas A., Hasgul R., Cakmak O., Akyol O.. **Caffeic acid phenethyl ester as a protective agent against nephrotoxicity and/or oxidative kidney damage: A detailed systematic review**. *The Scientific World Journal* (2014) **2014** 1-16. DOI: 10.1155/2014/561971
5. Al‐Asmari A. K., Al‐Said M. S., Abbasmanthiri R., Al‐Buraidi A., Ibrahim K. E., Rafatullah S.. **Impact of date palm pollen (**. *Clinical Phytoscience* (2020) **6** 1-12
6. Al‐Farsi M. A., Lee C. Y.. **Optimization of phenolics and dietary fibre extraction from date seeds**. *Food Chemistry* (2008a) **108** 977-985. DOI: 10.1016/j.foodchem.2007.12.009
7. Al‐Farsi M. A., Lee C. Y.. **Nutritional and functional properties of dates: A review**. *Critical Reviews in Food Science and Nutrition* (2008b) **48** 877-887. DOI: 10.1080/10408390701724264
8. Allaith A. A. A.. **Antioxidant activity of Bahraini date palm (**. *International Journal of Food Science and Technology* (2008) **43** 1033-1040. DOI: 10.1111/j.1365-2621.2007.01558.x
9. Ames B. N., Cathcart R., Schwiers E., Hochstein P.. **Uric acid provides an antioxidant defense in humans against oxidant‐ and radical‐caused aging and cancer: A hypothesis**. *Proceedings of the National Academy of Sciences of the United States of America* (1981) **78** 6858-6862. DOI: 10.1073/pnas.78.11.6858
10. Ashrafian H., Czibik G., Bellahcene M., Aksentijević D., Smith A. C., Mitchell S. J., Dodd M. S., Kirwan J., Byrne J. J., Ludwig C., Isackson H., Yavari A., Støttrup N. B., Contractor H., Cahill T. J., Sahgal N., Ball D. R., Birkler R. I., Hargreaves I., Watkins H.. **Fumarate is cardioprotective via activation of the Nrf2 antioxidant pathway**. *Cell Metabolism* (2012) **15** 361-371. DOI: 10.1016/j.cmet.2012.01.017
11. Ayla S., Seckin I., Tanriverdi G., Cengiz M., Eser M., Soner B., Oktem G.. **Doxorubicin induced nephrotoxicity: Protective effect of nicotinamide**. *International journal of cell biology, 2011* (2011) **2011** 390238. DOI: 10.1155/2011/390238
12. Biovia D. S.. **Discovery studio visualizer**. (2017)
13. Borochov‐Neori H., Judeinstein S., Greenberg A., Volkova N., Rosenblat M., Aviram M.. **Antioxidant and antiatherogenic properties of phenolic acid and flavonol fractions of fruits of ‘Amari' and ‘Hallawi' date (**. *Journal of Agricultural and Food Chemistry* (2015) **63** 3189-3195. PMID: 25765921
14. Boulenouar N., Marouf A., Cheriti A.. **Antifungal activity and phytochemical screening of extracts from**. *Natural Product Research* (2011) **25** 1999-2002. DOI: 10.1080/14786419.2010.536765
15. Calder P. C., Yaqoob P.. **Omega‐3 polyunsaturated fatty acids and human health outcomes**. *BioFactors* (2009) **35** 266-272. DOI: 10.1002/biof.42
16. Connor W. E.. **Importance of n− 3 fatty acids in health and disease**. *The American Journal of Clinical Nutrition* (2000) **71** 171 S-175 S. DOI: 10.1093/ajcn/71.1.171 S
17. Diab K., Aboul‐Ela E.. **In vivo comparative studies on antigenotoxicity of date palm (**. *Toxicology International* (2012) **19** 279-286. DOI: 10.4103/0971-6580.103669
18. el Arem A., Saafi E. B., Ghrairi F., Thouri A., Zekri M., Ayed A., Zakhama A., Achour L.. **Aqueous date fruit extract protects against lipid peroxidation and improves antioxidant status in the liver of rats subchronically exposed to trichloroacetic acid**. *Journal of Physiology and Biochemistry* (2014) **70** 451-464. DOI: 10.1007/s13105-014-0323-6
19. El‐Far A., Shaheen H., Abdel‐Daim M., Al‐Jaouni S., Mousa S.. **Date palm (**. *Curr Trends Nutraceuticals* (2016) **1** 2
20. Ellison N.. **Goodman & Gilman’s The pharmacological basis of therapeutics, 10th Edition**. *Anesthesia & Analgesia* (2002) **94** 1377. DOI: 10.1097/00000539-200205000-00085
21. El‐Sheikh A. A., Morsy M. A., Mahmoud M. M., Rifaai R. A., Abdelrahman A. M.. **Effect of coenzyme‐Q10 on doxorubicin‐induced nephrotoxicity in rats**. *Advances in Pharmacological Sciences* (2012) **2012**. DOI: 10.1155/2012/981461
22. El‐Shitany N. A., El‐Haggar S., El‐Desoky K.. **Silymarin prevents adriamycin‐induced cardiotoxicity and nephrotoxicity in rats**. *Food and Chemical Toxicology* (2008) **46** 2422-2428. DOI: 10.1016/j.fct.2008.03.033
23. Enayetallah A. E., French R. A., Thibodeau M. S., Grant D. F.. **Distribution of soluble epoxide hydrolase and of cytochrome P450 2C8, 2C9, and 2J2 in human tissues**. *Journal of Histochemistry & Cytochemistry* (2004) **52** 447-454. DOI: 10.1177/002215540405200403
24. Fadillioǧlu E., Erdoǧan H., Söǧüt S., Kuku I.. **Protective effects of erdosteine against doxorubicin‐induced cardiomyopathy in rats**. *Journal of Applied Toxicology: An International Journal* (2003) **23** 71-74
25. Gopinath B., Harris D. C., Flood V. M., Burlutsky G., Mitchell P.. **Consumption of long‐chain n‐3 PUFA, α‐linolenic acid and fish is associated with the prevalence of chronic kidney disease**. *British Journal of Nutrition* (2011) **105** 1361-1368. PMID: 21255476
26. Grant M. K., Seelig D. M., Sharkey L. C., Choi W. S., Abdelgawad I. Y., Zordoky B. N.. **Sexual dimorphism of acute doxorubicin‐induced nephrotoxicity in C57Bl/6 mice**. *PLoS One* (2019) **14**. PMID: 30785938
27. Hayes J. D., McMahon M., Chowdhry S., Dinkova‐Kostova A. T.. **Cancer chemoprevention mechanisms mediated through the Keap1–Nrf2 pathway**. *Antioxidants & Redox Signaling* (2010) **13** 1713-1748. DOI: 10.1089/ars.2010.3221
28. Hrenák J., Arendášová K., Rajkovičová R., Aziriová S., Repová K., Krajčírovičová K., Celec P., Kamodyová N., Bárta A., Adamcová M., Paulis L., Simko F.. **Protective effect of captopril, olmesartan, melatonin and compound 21 on doxorubicin‐induced nephrotoxicity in rats**. *Physiological Research* (2013) **62** S181-S189. PMID: 24329698
29. Kang D. H., Nakagawa T., Feng L., Watanabe S., Han L., Mazzali M., Truong L., Harris R., Johnson R. J.. **A role for uric acid in the progression of renal disease**. *Journal of the American Society of Nephrology* (2002) **13** 2888-2897. DOI: 10.1097/01.asn.0000034910.58454.fd
30. Kang M.‐C., Ham Y.‐M., Heo S.‐J., Yoon S.‐A., Cho S.‐H., Kwon S.‐H., Jeong M. S., Jeon Y.‐J., Sanjeewa K. K. A., Yoon W.‐J., Yoon W.‐J.. **Anti‐inflammation effects of 8‐oxo‐9‐octadecenoic acid isolated from**. *EXCLI Journal* (2018) **17** 775. PMID: 30190667
31. Karasawa K., Otani H.. **Anti‐allergic properties of a matured fruit extract of the date palm tree (**. *Journal of Nutritional Science and Vitaminology* (2012) **58** 272-277. DOI: 10.3177/jnsv.58.272
32. Karasawa K., Uzuhashi Y., Hirota M., Otani H.. **A matured fruit extract of date palm tree (**. *Journal of Agricultural and Food Chemistry* (2011) **59** 11287-11293. DOI: 10.1021/jf2029225
33. Khan N., Sultana S.. **Induction of renal oxidative stress and cell proliferation response by ferric nitrilotriacetate (Fe‐NTA): Diminution by soy isoflavones**. *Chemico‐Biological Interactions* (2004) **149** 23-35. DOI: 10.1016/j.cbi.2004.06.003
34. Khan T. H., Sultana S.. **Antioxidant and hepatoprotective potential of**. *Journal of Enzyme Inhibition and Medicinal Chemistry* (2009) **24** 320-327. DOI: 10.1080/14756360802167754
35. Kpemissi M., Eklu‐Gadegbeku K., Veerapur V. P., Negru M., Taulescu M., Chandramohan V., Hiriyan J., Banakar S. M., Nv T., Suhas D. S., Puneeth T. A., Vijayakumar S., Metowogo K., Aklikokou K.. **Nephroprotective activity of**. *Biomedicine & Pharmacotherapy* (2019) **116**. DOI: 10.1016/j.biopha.2019.108961
36. Kumar Gothwal R., Kumar Nigam V., Mohan M. K., Sasmal D., Ghosh P.. **Extraction of bulk DNA from Thar Desert soils for optimization of PCR‐DGGE based microbial community analysis**. *Electronic Journal of Biotechnology* (2007) **10** 400-408
37. Liu J.‐Y.. **Inhibition of soluble epoxide hydrolase for renal health**. *Frontiers in Pharmacology* (2019) **9**. DOI: 10.3389/fphar.2018.01551
38. Liu L. L., Li Q. X., Xia L., Li J., Shao L.. **Differential effects of dihydropyridine calcium antagonists on doxorubicin‐induced nephrotoxicity in rats**. *Toxicology* (2007) **231** 81-90. DOI: 10.1016/j.tox.2006.11.067
39. Madkour F. F., Abdel‐Daim M.. **Hepatoprotective and antioxidant activity of**. *Indian Journal of Pharmaceutical Sciences* (2013) **75** 642-648. PMID: 24591738
40. Mahdhi A., Bahi A., Mzah D., Bakhrouf A.. **Use of mixture design to construct a consortium of date palm (**. *Journal of the Science of Food and Agriculture* (2013) **93** 3850-3855. DOI: 10.1002/jsfa.6323
41. Mohan M., Kamble S., Gadhi P., Kasture S.. **Protective effect of**. *Food and Chemical Toxicology* (2010) **48** 436-440. DOI: 10.1016/j.fct.2009.10.042
42. Mubarak S., Hamid S. A., Farrag A. R., Samir N., Hussein J. S.. **Atherogenic coefficient and atherogenic index in doxorubicin–induced cardiotoxicity: Impact of date palm extract**. *Comparative Clinical Pathology* (2018a) **27** 1515-1522
43. Mubarak S., Hamid S. A., Farrag A. R., Samir N., Hussein J. S.. **Cardioprotective effect of date palm against doxorubicin‐induced cardiotoxicity**. *Asian Journal of Pharmaceutical and Clinical Research* (2018b) **11** 141-146
44. Naqshbandi A., Khan M. W., Rizwan S., ur Rehman S., Khan F.. **Studies on the protective effect of dietary fish oil on cisplatin induced nephrotoxicity in rats**. *Food and Chemical Toxicology* (2012) **50** 265-273. DOI: 10.1016/j.fct.2011.10.039
45. Nehdi I., Omri S., Khalil M., Al‐Resayes S.. **Characteristics and chemical composition of date palm (**. *Industrial Crops and Products* (2010) **32** 360-365. DOI: 10.1016/j.indcrop.2010.05.016
46. Ohkawa H., Ohishi N., Yagi K.. **Assay for lipid peroxides in animal tissues by thiobarbituric acid reaction**. *Analytical Biochemistry* (1979) **95** 351-358. DOI: 10.1016/0003-2697(79)90738-3
47. Oz E., Ilhan M. N.. **Effects of melatonin in reducing the toxic effects of doxorubicin**. *Molecular and Cellular Biochemistry* (2006) **286** 11-15. DOI: 10.1007/s11010-005-9003-8
48. Pandey K. B., Rizvi S. I.. **Plant polyphenols as dietary antioxidants in human health and disease**. *Oxidative Medicine and Cellular Longevity* (2009) **2** 270-278. DOI: 10.4161/oxim.2.5.9498
49. Parrish A. R., Chen G., Burghardt R. C., Watanabe T., Morisseau C., Hammock B. D.. **Attenuation of cisplatin nephrotoxicity by inhibition of soluble epoxide hydrolase**. *Cell Biology and Toxicology* (2009) **25** 217-225. DOI: 10.1007/s10565-008-9071-0
50. Pujari R. R., Vyawahare N. S., Kagathara V. G.. **Evaluation of antioxidant and neuroprotective effect of date palm (**. *Indian Journal of Experimental Biology* (2011) **49** 627-633. PMID: 21870431
51. Rashid S., Ali N., Nafees S., Ahmad S. T., Arjumand W., Hasan S. K., Sultana S.. **Alleviation of doxorubicin‐induced nephrotoxicity and hepatotoxicity by chrysin in Wistar rats**. *Toxicology Mechanisms and Methods* (2013) **23** 337-345. DOI: 10.3109/15376516.2012.759306
52. Rašković A., Stilinović N., Kolarović J., Vasović V., Vukmirović S., Mikov M.. **The protective effects of silymarin against doxorubicin‐induced cardiotoxicity and hepatotoxicity in rats**. *Molecules* (2011) **16** 8601-8613. PMID: 21993249
53. Reddy P. N., Reddy P. S., Rao M. R.. **Studies on the effect of doxorubicin on MDA, NO 2, NO 3, Se‐GSH peroxidase and SOD levels in albino rat tissues**. *African Journal of Biotechnology* (2007) **6** 2303-2309
54. Refaie M. M., Amin E. F., El‐Tahawy N. F., Abdelrahman A. M.. **Possible protective effect of diacerein on doxorubicin‐induced nephrotoxicity in rats**. *Journal of Toxicology* (2016) **2016**. DOI: 10.1155/2016/9507563
55. Sabbah M. F., Alshubali F., Baothman O. A., Zamzami M. A., Shash L., Hassan I. A.. **Cardioprotective effect of Ajwa date aqueous extract on doxorubicin‐induced toxicity in rats**. *Biomedical and Pharmacology Journal* (2018) **11** 1521-1536
56. Saha S. S., Ghosh M.. **Protective effect of conjugated linolenic acid isomers present in vegetable oils against arsenite‐induced renal toxicity in rat model**. *Nutrition* (2013) **29** 903-910. DOI: 10.1016/j.nut.2012.12.013
57. Sahyon H. A., Al‐Harbi S. A.. **Chemoprotective role of an extract of the heart of the**. *Food and Chemical Toxicology* (2020) **135**. DOI: 10.1016/j.fct.2019.111045
58. Salah E. B. S.‐B., El Arem A., Louedi M., Saoudi M., Elfeki A., Zakhama A., Najjar M. F., Hammami M., Achour L.. **Antioxidant‐rich date palm fruit extract inhibits oxidative stress and nephrotoxicity induced by dimethoate in rat**. *Journal of Physiology and Biochemistry* (2012) **68** 47-58. PMID: 21983806
59. Sandhar H. K., Kumar B., Prasher S., Tiwari P., Salhan M., Sharma P.. **A review of phytochemistry and pharmacology of flavonoids**. *Internationale Pharmaceutica Sciencia* (2011) **1** 25-41
60. Shireen K., Pace R., Mahboob M., Khan A.. **Effects of dietary vitamin E, C and soybean oil supplementation on antioxidant enzyme activities in liver and muscles of rats**. *Food and Chemical Toxicology* (2008) **46** 3290-3294. PMID: 18706466
61. Tamada S., Nakatani T., Asai T., Tashiro K., Komiya T., Sumi T., Okamura M., Kim S., Iwao H., Kishimoto T., Yamanaka S., Miura K.. **Inhibition of nuclear factor‐κB activation by pyrrolidine dithiocarbamate prevents chronic FK506 nephropathy**. *Kidney International* (2003) **63** 306-314. PMID: 12472797
62. Vayalil P. K.. **Antioxidant and antimutagenic properties of aqueous extract of date fruit (**. *Journal of Agricultural and Food Chemistry* (2002) **50** 610-617. PMID: 11804538
63. Vayalil P. K.. **Date fruits (**. *Critical Reviews in Food Science and Nutrition* (2012) **52** 249-271. DOI: 10.1080/10408398.2010.499824
64. Vukicevic V., Kampfinger K., Stopper H.. **Influence of altered apoptosis in human lymphoblastoid cell lines on micronucleus frequency**. *Toxicology Letters* (2004) **147** 187-195. DOI: 10.1016/j.toxlet.2003.11.005
65. Wang Y., Chao X., Shi H., Mehboob H., Hassan W.. *Cardiology Research and Practice* (2019) **2019**. PMID: 31929900
66. Wang Z. J., Liang C. L., Li G. M., Yu C. Y., Yin M.. **Stearic acid protects primary cultured cortical neurons against oxidative stress**. *Acta Pharmacologica Sinica* (2007) **28** 315-326. DOI: 10.1111/j.1745-7254.2007.00512.x
67. Yagmurca M., Erdogan H., Iraz M., Songur A., Ucar M., Fadillioglu E.. **Caffeic acid phenethyl ester as a protective agent against doxorubicin nephrotoxicity in rats**. *Clinica Chimica Acta* (2004) **348** 27-34. DOI: 10.1016/j.cccn.2004.03.035
68. Yasin B. R., El‐Fawal H. A., Mousa S. A.. **Date (**. *International Journal of Molecular Sciences* (2015) **16** 30075-30090. PMID: 26694370
69. Younas A., Naqvi S. A., Khan M. R., Shabbir M. A., Jatoi M. A., Anwar F., Inam‐Ur‐Raheem M., Saari N., Aadil R. M.. **Functional food and nutra‐pharmaceutical perspectives of date (**. *Journal of Food Biochemistry* (2020) **44**. DOI: 10.1111/jfbc.13332
70. Zafaryab M., Fakhri K. U., Khan M. A., Hajela K., Rizvi M. M. A.. **In vitro assessment of cytotoxic and apoptotic potential of palmitic acid for breast cancer treatment**. *International Journal of Life Sciences Research* (2019) **7** 166-174
71. Zhang C.‐R., Aldosari S. A., Vidyasagar P. S., Nair K. M., Nair M. G.. **Antioxidant and anti‐inflammatory assays confirm bioactive compounds in Ajwa date fruit**. *Journal of Agricultural and Food Chemistry* (2013) **61** 5834-5840. DOI: 10.1021/jf401371v
72. Zheng Z., Schmidt‐Ott K. M., Chua S., Foster K. A., Frankel R. Z., Pavlidis P., Barasch J., D'Agati V. D., Gharavi A. G.. **A Mendelian locus on chromosome 16 determines susceptibility to doxorubicin nephropathy in the mouse**. *Proceedings of the National Academy of Sciences of the United States of America* (2005) **102** 2502-2507. DOI: 10.1073/pnas.0409786102
|
---
title: Fibrosis of Peritoneal Membrane, Molecular Indicators of Aging and Frailty
Unveil Vulnerable Patients in Long-Term Peritoneal Dialysis
authors:
- Patrícia Branco
- Rita Calça
- Ana Rita Martins
- Catarina Mateus
- Maria João Jervis
- Daniel Pinto Gomes
- Sofia Azeredo-Lopes
- Antonio Ferreira De Melo Junior
- Cátia Sousa
- Ester Civantos
- Sebastian Mas-Fontao
- Augusta Gaspar
- Sância Ramos
- Judit Morello
- Fernando Nolasco
- Anabela Rodrigues
- Sofia Azeredo Pereira
journal: International Journal of Molecular Sciences
year: 2023
pmcid: PMC10002940
doi: 10.3390/ijms24055020
license: CC BY 4.0
---
# Fibrosis of Peritoneal Membrane, Molecular Indicators of Aging and Frailty Unveil Vulnerable Patients in Long-Term Peritoneal Dialysis
## Abstract
Peritoneal membrane status, clinical data and aging-related molecules were investigated as predictors of long-term peritoneal dialysis (PD) outcomes. A 5-year prospective study was conducted with the following endpoints: (a) PD failure and time until PD failure, (b) major cardiovascular event (MACE) and time until MACE. A total of 58 incident patients with peritoneal biopsy at study baseline were included. Peritoneal membrane histomorphology and aging-related indicators were assessed before the start of PD and investigated as predictors of study endpoints. Fibrosis of the peritoneal membrane was associated with MACE occurrence and earlier MACE, but not with the patient or membrane survival. Serum α-Klotho bellow 742 pg/mL was related to the submesothelial thickness of the peritoneal membrane. This cutoff stratified the patients according to the risk of MACE and time until MACE. Uremic levels of galectin-3 were associated with PD failure and time until PD failure. This work unveils peritoneal membrane fibrosis as a window to the vulnerability of the cardiovascular system, whose mechanisms and links to biological aging need to be better investigated. Galectin-3 and α-Klotho are putative tools to tailor patient management in this home-based renal replacement therapy.
## 1. Introduction
Peritoneal dialysis (PD) is a home-based modality of renal replacement therapy and a good option for patients with chronic kidney disease (CKD). Independently of the chronological age, this population commonly presents an accelerated aging process affecting skeletal, immune, renal, and cardiovascular systems [1]. Therefore, the risk of mortality in CKD patients is increased by 10- to 20-fold in comparison to individuals with normal renal function [2]. Moreover, cardiovascular toxicity caused by uremia represents a major factor for the increased mortality in dialysis programs [3].
The dialytic capacity of the peritoneal membrane is pivotal in PD, but the integrity of the membrane in the uremic patient might have been overlooked. The existence of a high person-to-person variability in the status of the membrane before the start of PD was recently reported and related to the anti-aging molecule α-Klotho [4].
Deficiency of α-*Klotho is* well known to be involved in damage of the cardiovascular system, atherosclerosis, skin atrophy and osteoporosis, traits commonly associated with human aging [5,6]. Those traits also overlap with the manifestations of CKD [7], suggesting that α-Klotho might be an important player in PD outcomes. α-*Klotho is* changed in uremia [8], which is recognized as a disbalance between protective and harmful molecules [9]. In uremia, the concentrations of proteins associated with mechanisms underlying early aging can be affected, such as those related to inflammation and fibrosis in multiple organs/tissues [10].
In fact, α-Klotho was shown to be a uremic molecule implicated in the vulnerability of the peritoneal membrane, expressed as submesothelial fibrosis [4]. As more vulnerable peritoneal membranes were associated with low circulating α-Klotho, we herein hypothesized that α-Klotho might represent a multifaceted marker of both the survival of the membrane and the survival of the patient. Therefore, we conducted a prospective longitudinal observational study in a cohort of incident PD patients to investigate the impact of uremic toxins related to aging, peritoneal membrane status and the patient’s frailty in long-term PD outcomes.
## 2.1. Baseline Characterization of Study Population
This observational prospective cohort study included 58 patients, followed for 60 months. A total of $31\%$ were female. At baseline, patients were 56 (30–79) years old with a median renal residual function assessed by rGFR of 7 (4–10) mL/min/1.73 m2. The underlying renal diseases were diabetic renal disease ($20\%$), chronic glomerulonephritis ($20\%$), hypertensive nephrosclerosis ($23\%$), autosomal dominant polycystic kidney disease ($11\%$) and chronic pyelonephritis ($10\%$). Twenty-two patients had fibrosis of the peritoneal membrane at the baseline of the study.
Concerning dialysis parameters, 2 and 32 patients were fast and average-fast transporters, respectively, and $94\%$ of patients had good efficacy of dialysis.
Regarding therapeutics, patients with atherosclerosis artery diseases ($40\%$) were treated with the highest tolerated dose of statins and antiplatelet therapy. In addition, all patients were on inhibitors of renin–angiotensin axis (IECA or ARA) and diuretic therapy. Eighteen patients ($31\%$) were on spironolactone, which was mainly added in those with fibrosis of the peritoneal membrane before the start of PD. A total of 12 patients were on beta-blockers and $30\%$ were on other antihypertensive drugs. The number of patients in treatment for mineral bone disease was low.
The normalized protein catabolic rate (nPCR) was 0.99 (0.79–1.09) g/Kg/day and $47\%$ had proper nutrition. A total of 10 patients were vulnerable and 5 were frail according to the Edmonton scale.
The baseline variables of the study were analyzed according to the biopsy score of the membrane (Table 1), which considers submesothelial compact zone thickness (STM), vasculopathy and inflammation [4]: Score 0 represents no fibrosis, no vasculopathy, nor inflammation; Score 1: no fibrosis, but vasculopathy and/or inflammation; Score 2: fibrosis with/without vasculopathy and/or inflammatory changes.
Overall, at baseline, patients with membrane fibrosis received more spironolactone, antiplatelet and statins therapy. In the S2 group (fibrosis), more than half of the patients had peripheral arterial disease (PAD). While the biopsy score was not related to the age of the patients, the cutoff for the level of circulating α-Klotho (anti-aging molecule) that discriminated the existence of peritoneal membrane fibrosis before the start of PD was defined by performing a ROC curve (AUC = 0.860, $$p \leq 4$$ × 10−6). This cutoff was established at 742 pg/mL, with $83\%$ sensitivity (to detect fibrosis) and $71\%$ specificity (to detect no fibrosis).
## 2.2. Impact of the Status of the Peritoneal Membrane and Age-Related Indicators in PD-Related Outcomes
Regarding the long-term outcomes of the study, the minimum time on PD was 13 months and the median time was 42 (30–58) months. Technical failure during the follow-up period occurred in $41\%$ of patients, with a median time until failure of 40 (26–56) months.
A total of 27 patients ($47\%$) had a MACE during the study, with a minimum time for MACE of 8 months and a median time of 17 (12–31) months.
Next, we investigated the relation of study outcomes with the status of the membrane (biopsy score, STM, α-Klotho levels with a 742 pg/mL cutoff as a surrogate of fibrosis) and the age-related baseline indicators (age, serum biomarkers, frailty).
## 2.3. Status of Peritoneal Membrane, Age-Related Indicators and Technical Failure of Peritoneal Dialysis
Contrary to our initial hypothesis, the status of the membrane was not associated with technical failure (Table 2). Overall, the patients with PD failure, compared to those without, were older, had higher frailty scores, were more likely to be on calcium channel blockers (Table 2) and presented higher circulating galectin-3 at the study baseline. The use of icodextrin solutions, glucose applied, or diabetes were not associated with failure (Table 2) or time to PD failure (Table 3).
In addition, the galectin-3 was also related to the time until PD failure (Table 3). A cut-off of galectin-3 to discriminate PD failure was established at 8.88 ng/mL (sensitivity = $92\%$ and specificity of $46\%$), which was also associated with the survival of the peritoneal membrane (Figure 1A). This cut-off was independently associated with PD failure in an adjusted model to age, PAD, and calcium channel blockers (CCB) (Figure 1B), wherein age, frailty score and icodextrin did not account for the prediction of time to PD failure.
## 2.4. Peritoneal Membrane, Age-Related Indicators and Major Cardiovascular Event
While not related to PD failure, the presence of membrane fibrosis at the study baseline was associated with occurrence of MACE (Table 2) and time to MACE (Table 3, Figure 2A). Both endpoints were also related to age, frailty score, arterial atherosclerotic disease, use of statins, nPCR, beta-blockers and Kt/v (Table 2 and Table 3).
The existence of fibrosis in the peritoneal membrane at the study baseline was independently associated with time to MACE in an adjusted model to age, nutritional status and PAD (Figure 2B), wherein the frailty score or heart failure did not account for the prediction of the time to PD failure. This multivariate association was maintained when the membrane status was inferred by the non-invasive surrogate α-Klotho, using the identified cut-off for α-Klotho of 742 pg/mL instead of biopsy score (Figure 2C). The association of time until the occurrence of MACE with atherosclerotic disease might be inferred by the use of antiplatelet therapy (Figure 2D), maintaining α-Klotho cutoff as an independent factor in the model, together with age and antiplatelet use at the study baseline. The estimated survival probability for time to MACE discriminated by α-Klotho levels in an adjusted model to age, frailty, nPCR, rGFR and use of antiplatelet drugs is represented in Figure 2E.
Overall, our results suggest a link between the vulnerability of a patient’s cardiovascular system and the status of the peritoneal membrane. In addition to age, lower α-Klotho and PAD were also predictors of cardiovascular risk over time in different multivariate models.
## 2.5. Peritoneal Membrane, Age-Related Indicators and All-Causes Mortality
A total of six deaths occurred during the study, five related to cardiovascular disease and 1 to malignancy, which were not related to the biopsy score of the peritoneal membrane. Cardiovascular mortality and survivor groups had similar age, scores of biopsies and frailty as well as similar levels of serum biomarkers.
## 3. Discussion
Our data provides new information about the links between the peritoneal membrane, uremia and PD outcomes. We found that blood levels of galectin-3 represent a putative tool to identify patients at higher risk of PD failure. In addition, and contrary to our initial hypothesis, the baseline membrane fibrosis was not a predictor of technical failure, time to failure or all-causes mortality in PD. Instead, the status of the peritoneal membrane was related to MACE and time until occurrence of MACE, which can be inferred by circulating α-Klotho.
The rationale for the choice of the pre-PD molecules was driven by the hypothesis that prematurely aged phenotypes of the peritoneal membrane could be associated with poorer long-term PD outcomes. These phenotypes are difficult to predict only from demographic characteristics, but could be favored by a uremic toxic environment, patients’ frailty and aging. Therefore, a group of aging-related indicators was investigated as predictors of PD outcomes. PD outcomes were associated with uremic molecules, but not with the frailty test applied. This test was chosen due to its simplicity and ease for daily clinical practice and validation in Portuguese [11,12].
The person-to-person variability in membrane status and functions, even before the start of PD [4], is likely to be driven by genetic and non-genetic factors [13,14,15,16]. The latter includes exposure to glucose, peritonitis, loss of residual renal function, inflammation and uremia [4,17,18]. In this context, better knowledge about aging-related uremic molecules might fulfill clinicians’ aims for accessible risk stratification tools for tailored prescriptions. Foreseeing a proof of concept that uremia-related mechanisms impact both membrane and patient survival, we selected a panel of proteins reported to be associated with aging, inflammation and fibrosis in other organs/tissues [19,20,21,22,23,24,25,26,27,28,29,30,31,32,33].
We found that the status of the membrane (evaluated by histomorphology, STM and by a surrogate α-Klotho cutoff) was not associated with changes in the functions of the peritoneal transport. Moreover, the pre-PD membrane status was not predictive of long-term survival of both peritoneal membrane and patients.
As α-*Klotho is* associated with fibrosis of the peritoneal membrane [4], the absence of association between α-Klotho and PD failure was an unexpected finding. While it did not consider fibrosis, a previous study about the arteriolar structure concluded that membrane arteriolar frailty in CKD stage 5 patients follows with cardiovascular system damage [34]. Therefore, as α-*Klotho is* associated with arteriosclerosis and aging, our results might suggest that the peritoneal biopsy score reflects a vascular vulnerability more than the integrity of the membrane. This novel and overlooked dimension might account for the shared mechanisms of persistent uremic phenotype, premature aging, and fibrosis of different tissues. In fact, the membrane might not represent a risk factor but a marker of a particular cardiovascular vulnerability profile.
Substantial cardiovascular risk persists in CKD patients, despite the treatment of established cardiovascular risk factors such as arterial hypertension and dyslipidemia. Knowledge about the uremia profiles that might be predictors of these risks will pave the way for personalized interventions. Moreover, this knowledge aligns with the need for novel drugs to control the unbalanced status of protective and deleterious molecules that constitutes uremia.
Our data might support that even older, frail, at higher cardiovascular risk and/or with a worsened status of peritoneal membrane patients might take advantage of this home-based modality of renal replacement therapy because we did not find any association between frailty or peritoneal membrane status and mortality or survival in the technique. Attention must be paid to the combination of atherosclerotic arterial disease, namely PAD and low α-Klotho levels. α-*Klotho is* an anti-aging molecule that exerts beneficial effects on the endothelium [35]. Moreover, α-Klotho-deficient mice show increased vascular calcification [36,37], further supporting a beneficial cardiovascular role of α-Klotho and putative relevance of recombinant α-Klotho to control the burden of comorbidities in PD patients.
Differently from α-Klotho, there was a clear association with galectin-3 and PD failure. Galectin-3, which is secreted by macrophages, has been associated with an inflammatory and fibrotic phenotype [19,25,26,27,28,29]. Moreover, Béllon et al. [ 2011] showed that alternative activated macrophages or M2 phenotypes were present in the peritoneal effluent drained from patients, and were able to stimulate fibroblast proliferation and the loss of peritoneal function [30].
α-Klotho and galectin-3 share common characteristics, e.g., both are uremic toxins and have been related to fibrosis and inflammation. However, unlike α-Klotho, galectin-3 was associated with PD failure. Therefore, the differences found in PD outcomes between poor α-*Klotho versus* rich galectin-3 uremic profiles suggest different underlying mechanisms. Moreover, while low baseline α-Klotho was highly associated with cardiovascular disease, such an association was not found for galectin-3 (Table 2). Instead, our data indicated galectin-3 as a predictor of earlier PD failure. Further studies are necessary to validate this data, but a putative explanation for the galectin-3 result is that this molecule is a high-affinity binding protein for advanced glycation products [38] whose relation to poor membrane efficiency and survival is well accepted [39,40]. Of note, inhibitors of galectin-3 are currently investigated in clinical trials [41,42,43], although in areas other than PD.
Our study has several strengths. All biomarker measurements were performed in the same laboratory to ensure measurement consistency across the pooled cohort, and we analyzed an anatomical territory with fibrosis and achieved a long follow-up period.
However, the study has some limitations. Firstly, the strict inclusion criteria from a single PD center implied that a rather small sample was studied; serum biomarkers were only measured at baseline, which might have hampered finding associations with time-dependent outcomes (PD and MACE). Secondly, other parameters of adequacy such as nutrition and volemia were neglected in our research, which may have influence data analysis and affect our prediction of long-term outcomes of patients. Moreover, our data only focused on clinical examinations and basic personal information, not including environmental conditions such as psychosocial and economic dimensions, which can affect their clinical outcomes.
Further research might focus on the putative role of galectin-3 and α-Klotho as tools to tailor patient management in this home-based renal replacement therapy.
## 4.1. Study Design and Participants
This was a single center, prospective study with 60 months of follow-up that included incident patients at the PD Unit of Santa Cruz Hospital, Centro Hospitalar de Lisboa Ocidental, Portugal. The study was approved by the Ethics Committee of the NOVA Medical school, Faculdade de Ciências Médicas, NOVA University of Lisbon (Approval number $\frac{50}{2019}$). The study was conducted according to the Declaration of Helsinki and Good Clinical Practices and complied with the European Union GDPR Legislation.
At the enrollment, patients were referred from the Nephrology consultation inside or outside the hospital for an information consultation. Patients were enrolled in a consecutive manner. The main purposes of the consultation were to assess the eligibility criteria for the renal substitution technique and to provide information to allow an informed choice. This consultation included a multidisciplinary team composed of a doctor, nurse, nutritionist, and social worker. Inclusion criteria were being over 18 years of age and having a stable clinical condition, defined by the absence of serious abdominal infections (diverticulitis, pancreatitis and cholecystitis) or active neoplasia, and on PD with biopsy of the peritoneal membrane. The exclusion criteria were having had previous aggressions to the peritoneal membrane (such as surgeries or peritonitis).
Non-autonomous patients were included for assisted PD whenever there was a caretaker. All patients signed informed consent.
## 4.2. PD Prescription
PD was started within 30 (21–44) days after the implantation of the catheter. All patients started on continuous ambulatory peritoneal dialysis (CAPD) and after the first year, $30\%$ of patients switched to automated peritoneal dialysis (APD) and remained stable over the observation period. All patients were treated with dialysis solutions with a reduced content of glucose degradation products and a normal pH (Baxter®, Deerfield, MA, USA and Fresenius®, Bad Homburg, Germany). At baseline, no hypertonic solutions were used and a total of $38\%$ of the patients received polyglucose. The major reasons to include polyglucose in the prescription was hydration status ($41\%$), diabetes and the presence of basal peritoneal membrane fibrosis ($30\%$). Prescriptions of amino-acid-containing solutions for PD were exclusive for diabetic patients. The daily quantity of administered glucose was maintained for CAPD, but increased in APD patients over time to achieve adequate ultrafiltration and fluid balance.
## 4.3. Baseline Variables
The following variables were assessed at study baseline Peritoneal and renal Kt/V urea and creatinine clearances, glomerular filtration rate (GFR), body surface area (BSA), and protein catabolic rate were calculated using Patient onLine (POL) software version 6.3 (Fresenius®, Bad Homburg, Germany). These variables were investigated as factors with impact on PD outcomes. All patients were followed up until death, PD drop-out, or 30 June 2019.
## 4.4. Study Outcomes
The primary outcomes were PD-related outcomes:-PD technique failure refers to ultrafiltration failure, peritonitis, or dialysis inefficacy. Patients were considered with no technical failure when achieving 60 months of follow-up.-Time for technique failure is the time on PD of each patient in the study until technical failure. Participants dropping PD out for reasons other than technical failure (switching to hemodialysis by option, kidney transplantation, transference to other PD centers or loss to follow-up) were censored.
The secondary outcomes were Cardiovascular Outcomes:(a) All-cause mortality(b) Major Cardiovascular event-Major Cardiovascular event (MACE) after 3 months on PD. MACEs were defined according to validated clinical criteria and included coronary heart disease (CHD), congestive heart failure (HF), acute myocardial infarction (AMI), acute cerebral infarction (ACI) and cardiac death caused by AMI, arrhythmias or HF. CHD was defined as ≥$50\%$ diameter stenosis of coronary arteries by either coronary angiography or CT angiography [48]. HF was diagnosed according to ESC guidelines for the diagnosis and treatment of chronic heart failure [49]. AMI was diagnosed according to ESC guidelines for the management of acute coronary syndromes [50]. ACI was defined as an acute neurological event lasting more than 24 h associated with clinical evidence of ischemic focus of the brain [51]. Cardiac death was defined as death caused by AMI, arrhythmias or CHF.-Time for MACE was defined for each patient as the time in the study until a MACE. Censored data were defined for those dropping out of the study without MACE or those achieving the end of the study without MACE.
## 4.5. Statistical Analyses
Categorical variables are presented as absolute (n) and relative frequencies (%); continuous non-normally distributed data are expressed as median (interquartile range). The Kruskal–Wallis test was used to assess differences between three or more independent groups. The Mann–Whitney U-test was used to assess differences between two independent groups. Potential associations between categorical data were analyzed using the Chi-Squared test. ROC curves were also used to identify cut-offs for potential blood biomarkers. Multiple Cox proportional hazards regression models were performed to assess potential predictors of survival, technique survival and the time to the occurrence of a cardiovascular event. The proportional hazards assumption was assessed through the Schoenfeld residual plots. All models were fit using the ‘survival’ R package [52,53].
The optimal cutpoints were obtained through the maximally selected rank statistics method (see [54] for more details) using the ‘maxstat’ R package [55] and considering the time until PD failure with the censor variable, indicating whether the patient suffered a technique failure or not.
Survival curves were generated using the Kaplan–Meier technique and tested using the log-rank test.
A confidence level alpha of 0.05 was considered throughout the study. Statistical analyses were performed with the R software, R version 4.2.0 and SPSS.
## References
1. Dai L., Qureshi A.R., Witasp A., Lindholm B., Stenvinkel P.. **Early Vascular Ageing and Cellular Senescence in Chronic Kidney Disease**. *Comput. Struct. Biotechnol. J.* (2019.0) **17** 721-729. DOI: 10.1016/j.csbj.2019.06.015
2. Lanzani C., Citterio L., Vezzoli G.. **Klotho: A Link between Cardiovascular and Non-Cardiovascular Mortality**. *Clin. Kidney J.* (2020.0) **13** 926-932. DOI: 10.1093/ckj/sfaa100
3. Lozier M.R., Sanchez A.M., Lee J.J., Tamariz L.J., Valle G.A.. **Comparison of Cardiovascular Outcomes by Dialysis Modality: A Systematic Review and Meta-Analysis**. *Perit. Dial. Int. J. Int. Soc. Perit. Dial.* (2019.0) **39** 306-314. DOI: 10.3747/pdi.2018.00227
4. Branco P., Martins A.R., Calça R., Mateus C., Jervis M.J., Rodrigues A., Lopes S.A., Civantos E., Mas-Fontao S., Gaspar A.. **Alpha-klotho and peritoneal membrane status: A hypothesis generating study**. *Eur. J. Clin. Investig.* (2022.0) **53** e13903. DOI: 10.1111/eci.13903
5. Bi X., Yang K., Zhang B., Zhao J.. **The Protective Role of Klotho in CKD-Associated Cardiovascular Disease**. *Kidney Dis.* (2020.0) **6** 395-406. DOI: 10.1159/000509369
6. Kuro-o M.. **The Klotho Proteins in Health and Disease**. *Nat. Rev. Nephrol.* (2019.0) **15** 27-44. DOI: 10.1038/s41581-018-0078-3
7. Kadoya H., Satoh M., Nishi Y., Kondo M., Wada Y., Sogawa Y., Kidokoro K., Nagasu H., Sasaki T., Kashihara N.. **Klotho Is a Novel Therapeutic Target in Peritoneal Fibrosis via Wnt Signaling Inhibition**. *Nephrol. Dial. Transplant.* (2020.0) **35** 773-781. DOI: 10.1093/ndt/gfz298
8. Bao J.-F., Hu P.-P., She Q.-Y., Li A.. **A Land of Controversy: Fibroblast Growth Factor-23 and Uremic Cardiac Hypertrophy**. *J. Am. Soc. Nephrol.* (2020.0) **31** 1423-1434. DOI: 10.1681/ASN.2020010081
9. Zoccali C., Vanholder R., Massy Z.A., Ortiz A., Sarafidis P., Dekker F.W., Fliser D., Fouque D., Heine G.H., Jager K.J.. **The Systemic Nature of CKD**. *Nat. Rev. Nephrol.* (2017.0) **13** 344-358. DOI: 10.1038/nrneph.2017.52
10. Roh E., Hwang S.Y., Yoo H.J., Baik S.H., Cho B., Park Y.S., Kim H.J., Lee S.-G., Kim B.J., Jang H.C.. **Association of Plasma FGF21 Levels with Muscle Mass and Muscle Strength in a National Multicentre Cohort Study: Korean Frailty and Aging Cohort Study**. *Age Ageing* (2021.0) **50** 1971-1978. DOI: 10.1093/ageing/afab178
11. Dent E., Kowal P., Hoogendijk E.O.. **Frailty Measurement in Research and Clinical Practice: A Review**. *Eur. J. Intern. Med.* (2016.0) **31** 3-10. DOI: 10.1016/j.ejim.2016.03.007
12. Perna S., Francis M.D., Bologna C., Moncaglieri F., Riva A., Morazzoni P., Allegrini P., Isu A., Vigo B., Guerriero F.. **Performance of Edmonton Frail Scale on Frailty Assessment: Its Association with Multi-Dimensional Geriatric Conditions Assessed with Specific Screening Tools**. *BMC Geriatr.* (2017.0) **17**. DOI: 10.1186/s12877-016-0382-3
13. Gillerot G., Goffin E., Michel C., Evenepoel P., Van Biesen W., Tintillier M., Stenvinkel P., Heimbarger O., Lindholm B., Nordfors L.. **Genetic and Clinical Factors Influence the Baseline Permeability of the Peritoneal Membrane**. *Kidney Int.* (2005.0) **67** 2477-2487. DOI: 10.1111/j.1523-1755.2005.00357.x
14. Morelle J., Marechal C., Yu Z., Debaix H., Corre T., Lambie M., Verduijn M., Dekker F., Bovy P., Evenepoel P.. *N. Engl. J. Med.* (2021.0) **385** 1570-1580. DOI: 10.1056/NEJMoa2034279
15. Rumpsfeld M., McDonald S.P., Purdie D.M., Collins J., Johnson D.W.. **Predictors of Baseline Peritoneal Transport Status in Australian and New Zealand Peritoneal Dialysis Patients**. *Am. J. Kidney Dis.* (2004.0) **43** 492-501. DOI: 10.1053/j.ajkd.2003.11.010
16. Teitelbaum I.. **Peritoneal Dialysis**. *N. Engl. J. Med.* (2021.0) **385** 1786-1795. DOI: 10.1056/NEJMra2100152
17. Zhou Q., Bajo M.-A., del Peso G., Yu X., Selgas R.. **Preventing Peritoneal Membrane Fibrosis in Peritoneal Dialysis Patients**. *Kidney Int.* (2016.0) **90** 515-524. DOI: 10.1016/j.kint.2016.03.040
18. Mehrotra R., Devuyst O., Davies S.J., Johnson D.W.. **The Current State of Peritoneal Dialysis**. *J. Am. Soc. Nephrol.* (2016.0) **27** 3238-3252. DOI: 10.1681/ASN.2016010112
19. Campbell S.. **The Role of TWEAK/Fn14 IN the Pathogenesis of Inflammation and Systemic Autoimmunity**. *Front. Biosci.* (2004.0) **9** 2273. DOI: 10.2741/1395
20. Winkles J.A.. **The TWEAK–Fn14 Cytokine–Receptor Axis: Discovery, Biology and Therapeutic Targeting**. *Nat. Rev. Drug Discov.* (2008.0) **7** 411-425. DOI: 10.1038/nrd2488
21. Ebert T., Pawelzik S.-C., Witasp A., Arefin S., Hobson S., Kublickiene K., Shiels P.G., Bäck M., Stenvinkel P.. **Inflammation and Premature Ageing in Chronic Kidney Disease**. *Toxins* (2020.0) **12**. DOI: 10.3390/toxins12040227
22. Lu H., Jia C., Wu D., Jin H., Lin Z., Pan J., Li X., Wang W.. **Fibroblast Growth Factor 21 (FGF21) Alleviates Senescence, Apoptosis, and Extracellular Matrix Degradation in Osteoarthritis via the SIRT1-MTOR Signaling Pathway**. *Cell Death Dis.* (2021.0) **12** 865. DOI: 10.1038/s41419-021-04157-x
23. Moe O.W.. **Fibroblast Growth Factor 23: Friend or Foe in Uremia?**. *J. Clin. Investig.* (2012.0) **122** 2354-2356. DOI: 10.1172/JCI64184
24. Castillo-Rodríguez E., Pizarro-Sánchez S., Sanz A., Ramos A., Sanchez-Niño M., Martin-Cleary C., Fernandez-Fernandez B., Ortiz A.. **Inflammatory Cytokines as Uremic Toxins: “Ni Son Todos Los Que Estan, Ni Estan Todos Los Que Son”**. *Toxins* (2017.0) **9**. DOI: 10.3390/toxins9040114
25. Takenaka T., Kobori H., Miyazaki T., Suzuki H., Nishiyama A., Ishii N., Yamashita M., Hayashi M.. **Klotho Protein Supplementation Reduces Blood Pressure and Renal Hypertrophy in Db/Db Mice, a Model of Type 2 Diabetes**. *Acta Physiol.* (2019.0) **225** e13190. DOI: 10.1111/apha.13190
26. Pedersen L., Pedersen S.M., Brasen C.L., Rasmussen L.M.. **Soluble Serum Klotho Levels in Healthy Subjects. Comparison of Two Different Immunoassays**. *Clin. Biochem.* (2013.0) **46** 1079-1083. DOI: 10.1016/j.clinbiochem.2013.05.046
27. Moreno J.A., Izquierdo M.C., Sanchez-Niño M.D., Suárez-Alvarez B., Lopez-Larrea C., Jakubowski A., Blanco J., Ramirez R., Selgas R., Ruiz-Ortega M.. **The Inflammatory Cytokines TWEAK and TNFα Reduce Renal Klotho Expression through NFκB**. *J. Am. Soc. Nephrol.* (2011.0) **22** 1315-1325. DOI: 10.1681/ASN.2010101073
28. Grabner A., Faul C.. **The Role of Fibroblast Growth Factor 23 and Klotho in Uremic Cardiomyopathy**. *Curr. Opin. Nephrol. Hypertens.* (2016.0) **25** 314-324. DOI: 10.1097/MNH.0000000000000231
29. de Boer R.A., van Veldhuisen D.J., Gansevoort R.T., Muller Kobold A.C., van Gilst W.H., Hillege H.L., Bakker S.J.L., van der Harst P.. **The Fibrosis Marker Galectin-3 and Outcome in the General Population**. *J. Intern. Med.* (2012.0) **272** 55-64. DOI: 10.1111/j.1365-2796.2011.02476.x
30. Salgado J.V., Goes M.A., Salgado Filho N.. **FGF21 and Chronic Kidney Disease**. *Metabolism* (2021.0) **118** 154738. DOI: 10.1016/j.metabol.2021.154738
31. Sanz A.B., Sanchez-Niño M.D., Ortiz A.. **TWEAK, a Multifunctional Cytokine in Kidney Injury**. *Kidney Int.* (2011.0) **80** 708-718. DOI: 10.1038/ki.2011.180
32. Sanz A.B., Aroeira L.S., Bellon T., del Peso G., Jimenez-Heffernan J., Santamaria B., Sanchez-Niño M.D., Blanco-Colio L.M., Lopez-Cabrera M., Ruiz-Ortega M.. **TWEAK Promotes Peritoneal Inflammation**. *PLoS ONE* (2014.0) **9**. DOI: 10.1371/journal.pone.0090399
33. Blanco-Colio L.M., Martín-Ventura J.L., Muñóz-García B., Orbe J., Páramo J.A., Michel J.-B., Ortiz A., Meilhac O., Egido J.. **Identification of Soluble Tumor Necrosis Factor-Like Weak Inducer of Apoptosis (STWEAK) as a Possible Biomarker of Subclinical Atherosclerosis**. *Arterioscler. Thromb. Vasc. Biol.* (2007.0) **27** 916-922. DOI: 10.1161/01.ATV.0000258972.10109.ff
34. Donderski R., Stróżecki P., Sulikowska B., Grajewska M., Trafny R., Bodnar M., Marszałek A., Stefańska A., Siódmiak J., Odrowąż-Sypniewska G.. **Assessment of Peritoneal Membrane Arteriolar Structure in Conjunction with Traditional Cardiovascular System Evaluation in Chronic Kidney Disease (CKD) Stage 5 Patients**. *Kidney Blood Press. Res.* (2018.0) **43** 1042-1052. DOI: 10.1159/000490827
35. Vila Cuenca M., Hordijk P.L., Vervloet M.G.. **Most Exposed: The Endothelium in Chronic Kidney Disease**. *Nephrol. Dial. Transplant.* (2020.0) **35** 1478-1487. DOI: 10.1093/ndt/gfz055
36. Hu M.C., Shi M., Cho H.J., Adams-Huet B., Paek J., Hill K., Shelton J., Amaral A.P., Faul C., Taniguchi M.. **Klotho and Phosphate Are Modulators of Pathologic Uremic Cardiac Remodeling**. *J. Am. Soc. Nephrol.* (2015.0) **26** 1290-1302. DOI: 10.1681/ASN.2014050465
37. Kuro-o M.. **Klotho and Calciprotein Particles as Therapeutic Targets against Accelerated Ageing**. *Clin. Sci.* (2021.0) **135** 1915-1927. DOI: 10.1042/CS20201453
38. Vlassara H., Li Y.M., Imani F., Wojciechowicz D., Yang Z., Liu F.T., Cerami A.. **Identification of Galectin-3 as a High-Affinity Binding Protein for Advanced Glycation End Products (AGE): A New Member of the AGE-Receptor Complex**. *Mol. Med.* (1995.0) **1** 634-646. DOI: 10.1007/BF03401604
39. Masola V., Bonomini M., Borrelli S., Di Liberato L., Vecchi L., Onisto M., Gambaro G., Palumbo R., Arduini A.. **Fibrosis of Peritoneal Membrane as Target of New Therapies in Peritoneal Dialysis**. *Int. J. Mol. Sci.* (2022.0) **23**. DOI: 10.3390/ijms23094831
40. Roumeliotis S., Dounousi E., Salmas M., Eleftheriadis T., Liakopoulos V.. **Unfavorable Effects of Peritoneal Dialysis Solutions on the Peritoneal Membrane: The Role of Oxidative Stress**. *Biomolecules* (2020.0) **10**. DOI: 10.3390/biom10050768
41. Lau E.S., Liu E., Paniagua S.M., Sarma A.A., Zampierollo G., Lopez B., Diez J., Wang T.J., Ho J.E.. **Galectin-3 Inhibition with Modified Citrus Pectin in Hypertension**. *JACC Basic Transl. Sci.* (2021.0) **6** 12-21. DOI: 10.1016/j.jacbts.2020.10.006
42. **A Study to Investigate the Safety and Efficacy of GB1211 (a Galectin-3 Inhibitor) in Combination with Atezolizumab in Patients with Non-Small Cell Lung Cancer (NSCLC)**
43. **Blocking Extracellular Galectin-3 in Patients with Osteoarthritis**
44. **NKF-K/DOQI Clinical Practice Guidelines for Peritoneal Dialysis Adequacy: Update 2000**. *Am. J. Kidney Dis.* (2001.0) **37** S65-S136. DOI: 10.1016/S0272-6386(01)70006-6
45. Blake P.G., Bargman J.M., Brimble K.S., Davison S.N., Hirsch D., McCormick B.B., Suri R.S., Taylor P., Zalunardo N., Tonelli M.. **Clinical Practice Guidelines and Recommendations on Peritoneal Dialysis Adequacy 2011**. *Perit. Dial. Int. J. Int. Soc. Perit. Dial.* (2011.0) **31** 218-239. DOI: 10.3747/pdi.2011.00026
46. Twardowski Z.. **Peritoneal equilibration test**. *Perit. Dial. Int.* (1987.0) **7** 138-148
47. Rolfson D.B., Majumdar S.R., Tsuyuki R.T., Tahir A., Rockwood K.. **Validity and Reliability of the Edmonton Frail Scale**. *Age Ageing* (2006.0) **35** 526-529. DOI: 10.1093/ageing/afl041
48. Fihn S.D., Blankenship J.C., Alexander K.P., Bittl J.A., Byrne J.G., Fletcher B.J., Fonarow G.C., Lange R.A., Levine G.N., Maddox T.M.. **2014 ACC/AHA/AATS/PCNA/SCAI/STS Focused Update of the Guideline for the Diagnosis and Management of Patients with Stable Ischemic Heart Disease**. *Circulation* (2014.0) **130** 1749-1767. DOI: 10.1161/CIR.0000000000000095
49. McDonagh T., Damy T., Doehner W., Lam C.S.P., Sindone A., van der Meer P., Cohen-Solal A., Kindermann I., Manito N., Pfister O.. **Screening, Diagnosis and Treatment of Iron Deficiency in Chronic Heart Failure: Putting the 2016 European Society of Cardiology Heart Failure Guidelines into Clinical Practice**. *Eur. J. Heart Fail.* (2018.0) **20** 1664-1672. DOI: 10.1002/ejhf.1305
50. Collet J.-P., Thiele H., Barbato E., Barthélémy O., Bauersachs J., Bhatt D.L., Dendale P., Dorobantu M., Edvardsen T., Folliguet T.. **2020 ESC Guidelines for the Management of Acute Coronary Syndromes in Patients Presenting without Persistent ST-Segment Elevation**. *Eur. Heart J.* (2021.0) **42** 1289-1367. DOI: 10.1093/eurheartj/ehaa575
51. Smith D., English J., Jonhston C.. **Cerebrovascular Disease**. *Harrisons Manual of Medicine* (2012.0) **Volume 2** 3270-3271
52. Therneau T.. *A Package for Survival Analysis in R* (2023.0)
53. Therneau T.M., Grambsch P.M.. *Modeling Survival Data: Extending the Cox Model* (2000.0)
54. Hothorn T., Lausen B.. **Bagging Tree Classifiers for Laser Scanning Images: A Data- and Simulation-Based Strategy**. *Artif. Intell. Med.* (2003.0) **27** 65-79. DOI: 10.1016/S0933-3657(02)00085-4
55. Hothorn T., Lausen B.. *Maxstat:Maximally Selected Rank Statistics* (2002.0) **Volume 2**
|
---
title: Safety evaluation of Balanced Health Care Dan—A medicinal formulation containing
traditional edible ingredients in lung tumor‐loaded mice
authors:
- Feng Dong
- Changhui Zhao
- Xiaoyun He
- Yueyang Dong
- Haiyan Liu
- Peng Yao
- Wentao Xu
journal: Food Science & Nutrition
year: 2022
pmcid: PMC10002941
doi: 10.1002/fsn3.3195
license: CC BY 4.0
---
# Safety evaluation of Balanced Health Care Dan—A medicinal formulation containing traditional edible ingredients in lung tumor‐loaded mice
## Abstract
Chinese formulation‐based medicinal food has been widely used in clinical trials, but its safety is not well studied. In this research, the edible safety assessment of Balanced Health Care Dan—a formulation containing traditional edible ingredients that were initially formulated to reduce side effects for lung cancer patients—was studied in mice based on biochemical and gut microbial analyses. The experimental mice were subcutaneously loaded with lung tumor A549 cells and then administrated with Balanced Health Care Dan (200 mg/kg or 400 mg/kg b.w. in gavage feeding) for 4 weeks. The body weight, blood parameters, and pathogenic phenotype in tissues were examined. No toxicological symptom was found in experimental mice compared with the normal control. Comprehensive analyses were also conducted to evaluate intestinal microbiota that are associated with many diseases. Balanced Health Care Dan modified the gut microbiota structure in a positive way. In conclusion, the Chinese formulation‐based medicinal food has shown no toxicological effect in mice within 4 weeks of feeding experiment and has the potential to be used in clinical trials.
Chinese formulation‐based medicinal food has been widely used in clinical trials, but its safety is not well studied. The current research evaluated a medicinal food for cancer patients using biochemical and gut microbial analyses to support the use of such a formula.
## INTRODUCTION
Chinese formulation‐based medicinal food has made great progress in clinical uses. There are increasing varieties of traditional Chinese formulations used in cancer treatment, but the safety of these formulations has not been systematically evaluated.
Tumors are seriously threatening human life and health, leading to increased mortality and morbidity worldwide (Ferlay et al., 2021; Siegel et al., 2019). Among all diseases, cancer has the second highest death rate, only next to cardiovascular and cerebrovascular diseases (Siegel et al., 2018; Zhang et al., 2021). Lung cancer is a malignant tumor characterized by a rapid proliferation rate, less survivability, and high mortality. Lung cancer has the highest mortality rate among all cancers (Su et al., 2021). Surgery, radiotherapy, and chemotherapy are the most common clinical treatment strategies (Couzin‐Frankel, 2013; Ma et al., 2019). These treatments have greatly improved the prognosis of lung cancer patients, but also have brought about many side effects (Wang et al., 2020, 2021). These side effects greatly reduce the quality of life of the patients during treatment.
Traditional medicinal food (TMF) offers a promising option to reduce the side effects during cancer treatment (Luo et al., 2019; Zhang et al., 2022). “ Balanced Health Care Dan” is a formula that is designed to improve patient's quality of life, and decrease chemotherapy‐induced adverse effects. Although most of the TMFs are botanical and have been traditionally considered to be nontoxic, the ingredients of TMF are generally complex with certain substances that might cause additive toxicities (Lin et al., 2018; Zhang & Yuan, 2012). Therefore, edible safety evaluation and clinical studies of TMFs are equally important. Animal‐based toxicity evaluation is necessary before the clinical application of these TMFs added to the diet of cancer patients (Shen et al., 2014; Wang, 2015). The current research focuses on the edible safety evaluation of “Balanced Health Care Dan,” which can be considered a model for traditional formula‐based medicinal food.
## Animals and cells
Totally, 72 BALB/C Nude mice (36 males and 36 females) of 5‐week‐old with an average body weight of 40–60 g were purchased from Vital River Laboratory Animal Technology Co., Ltd. The animal room was maintained at 23 ± 2°C, with relative humidity of 50 ± $5\%$. A 12 h light/dark cycle was provided by automated fluorescent illumination. All mice were provided with their diet and water ad libitum. The animal studies were approved by the Animal Care and Use Committee at China Agricultural University and all experiments were performed in accordance with relevant guidelines and regulations (Approval Number: AW02110202‐4).
A549 lung cancer cells (ATCC) were cultured in a carbon dioxide cell incubator at 37°C. The medium was Dulbecco's modified *Eagle medium* basal medium, complemented with $10\%$ fetal bovine serum and 100 U/ml penicillin and streptomycin. The cells were digested with $0.25\%$ trypsin and passaged on alternate days.
## Traditional formula ingredients
The Balanced Health Care Dan was prepared using the following ingredients: Dangshen, Astragalus membranaceus, Atractylodes macrocephala, white lentil, Ligusticum chuanxiong, bezoar, musk, Rhodiola, Platycodon grandiflorum, mulberry bark, licorice, Poria cocos, wood incense, Sichuan pepper, aloes, polygonatum, purple Ganoderma lucidum, Hedyotis diffusa, Dendrobii Officmalis Caulis, pollen, and honey. Each ingredient was washed, dried, disinfected, and ground into very fine powder. The powder was then mixed with condensed honey followed by pellet making for serving. Currently, the formulation is in the application for patent protection (Application No. CN20180374829.6).
## Mouse lung cancer tumor modeling
Each mouse was subcutaneously inoculated with A549 cells (5 × 106 cells/mouse) on the right back. The animals were used for the experiment when the average tumor volume of the group reached over 100 mm3. The successful model mice were randomly divided into groups as follows: Control model mice were gavaged with sterile water, while low‐dose group and high‐dose groups of mice were gavaged with 200 or 400 mg/kg of Balanced Health Care Dan solution for 4 weeks, respectively. The doses used were physiologically relevant to that administered to humans. The treatment was carried out in six consecutive days per week. The daily body weight of the mice was recorded.
## Blood biochemistry
At the end of the experiment, mice were fasted for 12 h before blood collection. Blood samples were collected from the inner canthus under anesthesia. The serum samples were analyzed for alkaline phosphatase (ALP), aspartate aminotransferase (AST), alanine aminotransferase (ALT), albumin (ALB), creatinine (CREA), urea, triglycerides (TG), and cholesterol (CHO) as previously reported (He et al., 2020).
## Pathology examination
The fresh tissues of the tumor, liver, kidney, spleen, and lung of mice were fixed with $4\%$ paraformaldehyde for 24–48 h. The tissues were dehydrated, waxed, embedded, and sliced. Subsequent HE staining was performed and then examined microscopically by a professional staff.
## Fecal metagenomic analysis
The intestinal contents of mice were collected in sterilized tubes and frozen at −80°C. DNA was then extracted from the intestinal contents according to the instructions in the kit (FDA6512, Beijing Ford Press Technology Co., LTD). The 16S rDNA sequencing and data analysis were performed as reported (He et al., 2020; Xu et al., 2020). Briefly, The V3‐V4 region of the 16S rDNA was amplified by PCR with specific primers linked to the barcode. Thermal cycling consisted of initial denaturation at 98°C for 1 min, followed by 30 cycles of denaturation at 98°C for 10 s, annealing at 50°C for 30 s, and elongation at 72°C for 30 s. Finally, 72°C for 5 min. Sequencing libraries were generated using TruSeq® DNA PCR‐Free Sample Preparation Kit (Illumina) following the manufacturer's recommendations, and index codes were added. The library quality was assessed on the [email protected] Fluorometer (Thermo Scientific) and Agilent Bioanalyzer 2100 system. At last, the library was sequenced on an Illumina NovaSeq platform. After data filtering, UPARSE software (uparsev7.0.1001) was used to cluster valid data into Operational Taxonomic Units. Mothur method and SILVA138 (http://www.arb‐silva.de/)'s SSUrRNA database were used for species annotation analysis. Qiime software (Version 1.9.1) was used to analyze diversity. R software (Version 2.15.3) was used for PCA analysis. LEfSe software was used for LEfSe analysis, and the filter value of LDA Score was 4 by default (Segata et al., 2011).
## Statistical analysis
The experimental data were presented as mean value ± standard deviation, and the data were analyzed by GraphPad Prism 8. A one‐way analysis of variance (ANOVA) was applied with Dunnet‐1 post hoc analysis. Differences between values were considered statistically significant at *$p \leq .05$, and extremely significant at **$p \leq .01.$
## Tumor incidence
A total of 72 mice (36 female and 36 male mice) were used in this experiment, among which 54 mice (27 males and 27 females) were successfully loaded with tumors that met the requirement and were included in the follow‐up experiment. The 54 mice were randomly divided into six groups (3 male groups and 3 female groups; 9 mice/group) following a computerized randomization scheme based on body weight. Four weeks treatment with Balanced Health Care Dan did not significantly affect the body weight of these mice (Figure 1a).
**FIGURE 1:** *Body weight of male and female mice during treatment. (a) Male mice; (b) female mice. CK, control group; Low, low‐dose group; High, high‐dose group*
## Analysis of clinical appearance
During the 4‐week experiment, no treatment‐related adverse effects in the clinical appearance of the animals were observed. The body weight of male and female mice in the experimental groups was comparable with that of the control group on day 7, day 14, day 21, and at the end of the experiment. Balanced Health Care Dan slightly increased body weight nonsignificantly, demonstrating that the formula did not exhibit any acute toxicity effects on the animals' growth and development (Hamaguchi et al., 2019).
## Analysis of hematology
Hematology index is an important indicator in safety evaluation (Rosa et al., 2018). Serum biochemical profile derangement can reflect nutrient metabolism abnormality (Ca Llens & Bartges, 2015; Gwinn et al., 2020; Paiano et al., 2019), as damaged tissues or organs modify the serum parameters. We detected blood biochemical indexes including ALB, ALP, ALT, AST, CREA, Urea, CHO, and TG at day 28 (Figures 2 and 3). These indicators can reflect liver, kidney function, and lipid metabolism. Liver and kidney are important target organs of many toxins, which can alter relevant indicators after intragastric administration (Calle‐Toro et al., 2020; Ursell et al., 2012). As the important indexes of liver and kidney function, the values of ALP, ALT, AST, and Urea were not significantly different between groups. In female groups, the mean values of ALB and CREA were slightly lower in low‐dose and high‐dose groups, respectively, compared with control. However, these differences were within the normal range, which possibly resulted from individual differences between mice. Such differences were not seen in male mice.
**FIGURE 2:** *Blood biochemistry of male and female mice during treatment. (a) Male mice; (b) female mice* **FIGURE 3:** *Lipid profile of male and female mice during treatment. (a) Male mice; (b) female mice. CHO, cholesterol; TG, total triglycerides*
The lipid profile reflects basic metabolism of the body. CHO and TG are the common indicators for lipid metabolism. In the female group, the mean value of TG in the low‐dose group reduced slightly compared with that of the control group indicating the treatment might have a hypolipidemic effect. Since this beneficial effect was not shown in the high‐dose group, we think the difference was more likely attributed to the background variability and sporadic deviation.
## Analysis of pathology in organs
A complete gross necropsy and microscopic anatomic pathological analysis of organs were conducted on all animals after the 4‐week feeding study. The liver, kidney, lung, and spleen showed no pathological lesions (data not shown) in the low‐dose or high‐dose groups. From the pathological point of view, Balanced Health Care Dan does not have any toxic or adverse effects.
## Analysis of microbiota
Intestinal flora is known as the “second fingerprint” of the body (Duffy et al., 2015; Ursell et al., 2012), which has an enormous impact on the nutrition and health status of the host. Diet directly affects the balance of intestinal microbiota. For edible safety evaluation, analyzing the changes in intestinal microbiota can directly reflect the effects of tested materials on body health (Barko et al., 2017). Therefore, the evaluation of intestinal microbiota is an important part of edible safety evaluation. Thus, in this study, the intestinal contents were used to evaluate the effects of Balanced Health Care Dan.
Based on metagenomic sequencing, in both male and female mice, the relative abundance of intestinal microbiota changed after TMF treatment (Figure 4a,b). Principal coordinates analysis showed significant changes in the intestinal microbiota of male mice in the TMF‐treated group compared with the control group, while the effect of TMF on the intestinal microbiota of female mice was relatively small (Figure 4c,d). The Shannon index was used to measure species diversity (Nielsen, 2021). From Figure 4e,f, TMF treatment significantly altered the structure and composition of intestinal microbiota in both male and female mice.
**FIGURE 4:** *The overall level of intestinal flora change. (a, b) The overall level of intestinal flora changed; (c) OUT analysis of each dose group in male mice; (d) OUT analysis of each dose group in female mice; (e) principal coordinate analysis of male mouse samples; (f) principal component analysis of female mouse samples*
From the analysis of the genus level, the most predominant 10 genera were identified (Figure 5a). According to the species annotation and abundance information of all samples at the genus level, correlation heatmap was applied to represent the top 35 genera (Figure 5b). Compared with the control group, Bacteroides, Ralstonia, Bilophila, Muribaculum, Prevotellaceae, Alistipes, and Anaerotruncus all showed significant changes in the TMF treatment groups (Figure 5c). The results showed that there were different effects on the mice by gender. For example, in male mice groups as shown in Figure 6a,b, Deterribacteres, Deferribacteraceae, Deferribacterales, and Deferribacteres increased significantly compared with the control group. Bacteroides acidifaciens and Proteobacteria decreased significantly. In addition, the proportion of Firmicutes to Bacteroidetes did not change significantly. While in the female mice groups as shown in Figure 6c,d, Muribaculaceae, Blautia, Bacteroides caccae, Clostridia, Firmicutes, Lachnospiraceae Bacterium, Lachnospiraceae, Oscillibacter, Lachnospirales, Oscillospirales, and Osillospiraceae increased significantly. Mucispirillum, Deferribacteraceae, Deferribacterales, Deferribacteres, Bacteroides, Alistipes Bacteroidaceae, Bacteroides acidifaciens, Bacteroidales, Bacteroidota, Bacteroidia, and Rikenellaceae significantly decreased. In addition, the proportion of Firmicutes to Bacteroidetes increased significantly. In order to further confirm the gender difference in the effects of TMF, functional predictive analysis was performed. As shown in Figure 7a, there was no difference in the top 10 functions between male and female mice although the function of the top 35 varied by gender (Figure 7b).
**FIGURE 5:** *Changes in the level of intestinal flora. (a) Top 10 dominant bacteria analysis; (b) top 35 dominant bacteria thermogram analysis; (c) abundance distribution box map between groups* **FIGURE 6:** *LEfSe analysis in mice (a) and (b) Male mice; (c) and (d) female mice.* **FIGURE 7:** *Function prediction analysis. (a) Relative abundance analysis of top 10 function annotation; (b) cluster analysis of relative abundance of top 35 function; (c) functional notes of Venn diagram; (d) function annotation of petal diagram*
There was more Bacteroides in the gut of colorectal cancer patient, which indirectly proved that the decrease in Bacteroides had a positive effect on the body's resistance to cancer (Garrett, 2019). Proteobacteria have been found to dominate the intestinal microbiota in acute and chronic inflammation caused by infectious pathogens or protozoan parasites, and the same phenomenon has been found in colorectal cancer associated with enteritis in animal and human experiments (Da et al., 2020). Balanced Health Care Dan significantly reduced Proteobacteria, indicating that Balanced Health Care Dan could help improve body health. The abundance of Muribaculaceae was negatively correlated with proinflammatory factors (Chung et al., 2020), and the increasing abundance of Muribaculaceae in the study indicated that it could protect intestinal health. The abundance of Lachnospiraceae was increased in the TMF treatment groups. As a potentially beneficial bacterium, Lachnospiraceae participates in the metabolism of a variety of carbohydrates, among which acetic acid, the fermentation product, is the main source of energy for the host (Vacca et al., 2020). In contrast to the control groups, the proportion of Firmicutes to Bacteroidetes in male mice groups did not change but significantly increased in female mice group, suggesting that after high‐dose treatment, female mice were more likely to absorb heat to maintain body weight (John & Mullin, 2016; Zhao et al., 2021). Therefore, to a certain extent, Balanced Health Care Dan improved the intestinal microbiota of mice and then might increase body immunity. This effect was more obvious in male mice. Therefore, from the point of view of intestinal health, Balanced Health Care *Dan is* not potentially harmful to the body. It needs to be noted that there are huge differences in gut microbiota between human and animal models, which are caused by species‐specific differences, evidenced by host–microbial interactions, environment, diet, and genetic responses (Nguyen et al., 2015). Further clinical trial is still necessary for accurate risk assessment.
## CONCLUSION
In a 4‐week animal feeding test, we evaluated edible safety of a traditional formula‐based medicinal food called Balanced Health Care Dan by examining behavior performance, body weight, relevant blood parameters, pathological phenotype, and intestinal microbiota in lung tumor‐loaded mice. We found no treatment‐related adverse effects in this short‐term toxicity evaluation experiment. This research provides a new strategy for the edible safety evaluation of TMFs which have the potential to be used for cancer patients.
## CONFLICT OF INTEREST
The authors declare no conflict of interest in this study.
## DATA AVAILABILITY STATEMENT
The datasets used and/or analyzed during this study are available from the corresponding author upon reasonable request.
## References
1. Barko P. C., Mcmichael M. A., Swanson K. S., Williams D. A.. **The gastrointestinal microbiome: A review**. *Journal of Veterinary Internal Medicine* (2017) **32** 9-25. DOI: 10.1111/jvim.14875
2. Ca Llens A. J., Bartges J. W.. **Urinalysis**. *Veterinary Clinics of North America: Small Animal Practice* (2015) **45** 621-637. DOI: 10.1016/j.cvsm.2015.02.001
3. Calle‐Toro J. S., Back S. J., Viteri B., Andronikou S., Kaplan S. L.. **Liver, spleen, and kidney size in children as measured by ultrasound: A systematic review**. *Journal of Ultrasound in Medicine* (2020) **39** 223-230. DOI: 10.1002/jum.15114
4. Chung Y. W., Gwak H. J., Moon S., Rho M., Ryu J. H.. **Functional dynamics of bacterial species in the mouse gut microbiome revealed by metagenomic and metatranscriptomic analyses**. *PLoS One* (2020) **15**. DOI: 10.1371/journal.pone.0227886
5. Couzin‐Frankel J.. **Cancer immunotherapy**. *Science* (2013) **342** 1432-1433. DOI: 10.1126/science.342.6165.1432
6. Da J. K., Yang J., Seo H., Lee W. H., Cho J. Y.. **Colorectal cancer diagnostic model utilizing metagenomic and metabolomic data of stool microbial extracellular vesicles**. *Scientific Reports* (2020) **10**. DOI: 10.1038/s41598-020-59529-8
7. Duffy L. C., Raiten D. J., Hubbard V. S., Starke‐Reed P.. **Progress and challenges in developing metabolic footprints from diet in human gut microbial cometabolism**. *Journal of Nutrition* (2015) **145** 1123S-1130S. DOI: 10.3945/jn.114.194936
8. Ferlay J., Colombet M., Soerjomataram I., Parkin D. M., Bray F.. **Cancer statistics for the year 2020: An overview**. *International Journal of Cancer* (2021) **149** 778-789. DOI: 10.1002/ijc.33588
9. Garrett W. S.. **The gut microbiota and colon cancer**. *Science* (2019) **364** 1133-1135. DOI: 10.1126/science.aaw2367
10. Gwinn W. M., Auerbach S. S., Parham F., Stout M., Waidyanatha S., Mutlu E., Collins B., Paules R. S., Merrick B. A., Ferguson S., Ramaiahgari S., Bucher J. R., Sparrow B., Toy H., Gorospe J., Machesky N., Shah R. R., Balik‐Meisner M. R., Mav D., DeVito M.. **Evaluation of 5‐day in vivo rat liver and kidney with high‐throughput transcriptomics for estimating benchmark doses of apical outcomes**. *Toxicological Sciences* (2020) **176** 343-354. DOI: 10.1093/toxsci/kfaa081
11. Hamaguchi T., Tsutsui‐Kimura I., Mimura M., Saito T., Saido T. C., Tanaka K. F.. **App NL‐G‐F/NL‐G‐F mice overall do not show impaired motivation, but cored amyloid plaques in the striatum are inversely correlated with motivation**. *Neurochemistry International* (2019) **129**. DOI: 10.1016/j.neuint.2019.104470
12. He X., Zheng S., Sheng Y., Miao T., Zhao C.. **Chlorogenic acid ameliorates obesity by preventing energy balance shift in high fat diet induced obese mice**. *Journal of the Science of Food and Agriculture* (2020) **101** 631-637. DOI: 10.1002/jsfa.10675
13. John G. K., Mullin G. E.. **The gut microbiome and obesity**. *Current Oncology Reports* (2016) **18**. DOI: 10.1016/j.jff.2020.104061
14. Lin A. X., Chan G., Hu Y., Ouyang D., Ung C., Shi L., Hu H.. **Internationalization of traditional Chinese medicine: Current international market, internationalization challenges and prospective suggestions**. *Chinese Medicine* (2018) **13**. DOI: 10.1186/s13020-018-0167-z
15. Luo H., Vong C. T., Chen H., Gao Y., Lyu P., Qiu L., Zhao M., Liu Q., Cheng Z., Zou J., Yao P., Gao C., Wei J., Ung C. O. L., Wang S., Zhong Z., Wang Y.. **Naturally occurring anti‐cancer compounds: Shining from Chinese herbal medicine**. *Chinese Medicine* (2019) **14**. DOI: 10.1186/s13020-019-0270-9
16. Ma Z., Fan Y., Wu Y., Kebebe D., Zhang B.. **Traditional Chinese medicine‐combination therapies utilizing nanotechnology‐based targeted delivery systems: A new strategy for antitumor treatment**. *International Journal of Nanomedicine* (2019) **14** 2029-2053. DOI: 10.2147/IJN.S197889
17. Nguyen T. L. A., Vieira‐Silva S., Liston A., Raes J.. **How informative is the mouse for human gut microbiota research?**. *Disease Models & Mechanisms* (2015) **8** 1-16. DOI: 10.1242/dmm.017400
18. Nielsen F.. **On a variational definition for the Jensen‐Shannon symmetrization of distances based on the information radius**. *Entropy* (2021) **23**. DOI: 10.3390/e23040464
19. Paiano R. B., Birgel D. B., Ollhoff R. D., Junior E.. **Biochemical profile and productive performance in dairy cows with lameness during postpartum period**. *Acta Scientiae Veterinariae* (2019) **47**. DOI: 10.22456/1679-9216.93775
20. Rosa C. V. D. D., Campos J. M., Sá Nakanishi A. B., Comar J. F., Martins I. P., Mathias P. C. F., Pedrosa M. M. D., Godoi V. A. F., Natali M. R. M.. **Food restriction promotes damage reduction in rat models of type 2 diabetes mellitus**. *PLoS One* (2018) **13**. DOI: 10.1371/journal.pone.0199479
21. Segata N., Izard J., Waldron L., Gevers D., Miropolsky L., Garrett W. S., Huttenhower C.. **Metagenomic biomarker discovery and explanation**. *Genome Biology* (2011) **12**. DOI: 10.1186/gb-2011-12-6-r60
22. Shen J., Ke W., Hu Q., Su J., Zhang W.. **Establishment quality control method of traditional Chinese medicine based on efficacy and safety (in Chinese)**. *Modernization of Traditional Chinese Medicine and Materia Medica‐World Science and Technology* (2014) **16** 502-505
23. Siegel R. L., Miller K. D., Jemal A.. **Cancer statistics, 2018**. *CA: A Cancer Journal for Clinicians* (2018) **68** 7-30. DOI: 10.3322/caac.21442
24. Siegel R. L., Miller K. D., Jemal A.. **Cancer statistics, 2019**. *CA: A Cancer Journal for Clinicians* (2019) **69** 7-34. DOI: 10.3322/caac.21551
25. Su X. L., Wang J. W., Che H., Wang C. F., Jiang H., Lei X., Zhao W., Kuang H. X., Wang Q. H.. **Clinical application and mechanism of traditional Chinese medicine in treatment of lung cancer**. *Chinese Medical Journal* (2021) **133** 2987-2997. DOI: 10.1097/CM9.0000000000001141
26. Ursell L. K., Metcalf J. L., Parfrey L. W., Knight R.. **Defining the human microbiome**. *Nutrition Reviews* (2012) **70** S38-S44. DOI: 10.1111/j.1753-4887.2012.00493.x
27. Vacca M., Celano G., Calabrese F. M., Portincasa P., Angelis M.. **The controversial role of human gut Lachnospiraceae**. *Microorganisms* (2020) **8**. DOI: 10.3390/microorganisms8040573
28. Wang K., Chen Q., Shao Y., Yin S., Yu H.. **Anticancer activities of TCM and their active components against tumor metastasis**. *Biomedicine & Pharmacotherapy* (2021) **133**. DOI: 10.1016/j.biopha.2020.111044
29. Wang Q., Jiao L., Wang S., Chen P., Bi L., Zhou D., Yao J., Li J., Wang L., Chen Z., Jia Y., Zhang Z., Shen W., Zhu W., Xu J., Gao Y., Xu L., Gong Y.. **Adjuvant chemotherapy with Chinese herbal medicine formulas versus placebo in patients with lung adenocarcinoma after radical surgery: A multicenter, randomized, double‐blind, placebo‐controlled trial**. *Biological Procedures Online* (2020) **22**. DOI: 10.1186/s12575-020-00117-5
30. Wang X.. **Methodology for systematic analysis of in vivo efficacy material base of traditional Chinese medicine – Chinmedomics (in Chinese)**. *China Journal of Chinese Materia Medica* (2015) **40** 13-17. PMID: 25993780
31. Xu J., Ge J., He X., Sheng Y., Huang K.. **Caffeic acid reduces body weight by regulating gut microbiota in diet‐induced‐obese mice**. *Journal of Functional Foods* (2020) **74**. DOI: 10.1016/j.jff.2020.104061
32. Zhang L., Zhang F., Li G.. **Traditional Chinese medicine and lung cancer—From theory to practice**. *Biomedicine & Pharmacotherapy* (2021) **137**. DOI: 10.1016/j.biopha.2021.111381
33. Zhang M., Yuan Y. F.. **Research on effects of the combination of traditional Chinese and Western medicine to pharmacokinetics and pharmacodynamics (in Chinese)**. *Medical Recapitulate* (2012) **18** 4212-4215
34. Zhang Y., Chen N., Xin N., Li Q., Zhang T., Ye H., Zhao C.. **Complexation of chlorogenic acid enhances the antiproliferative effect of lactoferrin to colon cancer cells**. *Food Bioscience* (2022) **46**. DOI: 10.1016/j.fbio.2022.101601
35. Zhao F., Jin H., Shen X., Li Q., Liu X., Zhang L., Sun Z., Yu J.. **Effect of the administration of probiotics on the fecal microbiota of adult individuals**. *Food Science & Nutrition* (2021) **9** 6471-6479. DOI: 10.1002/fsn3.2547
|
---
title: 4‐PBA inhibits hypoxia‐induced lipolysis in rat adipose tissue and lipid accumulation
in the liver through regulating ER stress
authors:
- Yanlei Xiong
- Yueming Wang
- Yanlian Xiong
- Lianghong Teng
journal: Food Science & Nutrition
year: 2023
pmcid: PMC10002945
doi: 10.1002/fsn3.3156
license: CC BY 4.0
---
# 4‐PBA inhibits hypoxia‐induced lipolysis in rat adipose tissue and lipid accumulation in the liver through regulating ER stress
## Abstract
High‐altitude hypoxia may disturb the metabolic modulation and function of both adipose tissue and liver. The endoplasmic reticulum (ER) is a crucial organelle in lipid metabolism and ER stress is closely correlated with lipid metabolism dysfunction. The aim of this study is to elucidate whether the inhibition of ER stress could alleviate hypoxia‐induced white adipose tissue (WAT) lipolysis and liver lipid accumulation‐mediated hepatic injury. A rat model of high‐altitude hypoxia (5500 m) was established using hypobaric chamber. The response of ER stress and lipolysis‐related pathways were analyzed in WAT under hypoxia exposure with or without 4‐phenylbutyric acid (PBA) treatment. Liver lipid accumulation, liver injury, and apoptosis were evaluated. Hypoxia evoked significant ER stress in WAT, evidenced by increased GRP78, CHOP, and phosphorylation of IRE1α, PERK. Moreover, Lipolysis in perirenal WAT significantly increased under hypoxia, accompanied with increased phosphorylation of hormone‐sensitive lipase (HSL) and perilipin. Treatment with 4‐PBA, inhibitor of ER stress, effectively attenuated hypoxia‐induced lipolysis via cAMP‐PKA‐HSL/perilipin pathway. In addition, 4‐PBA treatment significantly inhibited the increase in fatty acid transporters (CD36, FABP1, FABP4) and ameliorated liver FFA accumulation. 4‐PBA treatment significantly attenuated liver injury and apoptosis, which is likely resulting from decreased liver lipid accumulation. Our results highlight the importance of ER stress in hypoxia‐induced WAT lipolysis and liver lipid accumulation.
Enhanced ER stress mediated WAT lipolysis was observed in a rat model of high‐altitude hypoxia, which contribute to hepatic dysfunction and apoptosis through excess release of FFA. Our findings highlight the vital role of 4‐PBA in WAT lipolysis and liver dysfunction via regulating ER stress, which may provide novel insights into systemic metabolic disturbances in high‐altitude area.
## INTRODUCTION
Ascent to high altitude is associated with multi physiological and metabolic responses to counter with the stress of hypobaric hypoxia. White adipose tissue (WAT) is the largest reservoir of energy reserves, which stores energy in the form of triglyceride in lipid droplets. WAT plays an essential role in maintaining the whole‐body lipid metabolism homeostasis and accumulated evidence has demonstrated the functional association between adipose tissue and liver (Natarajan et al., 2017; Sun et al., 2012). In our previous work, hypobaric hypoxia was proved to accelerate lipolysis and suppress lipogenesis of WAT (Xiong et al., 2014). Under normal conditions, the lipid metabolism is a dynamic equilibrium process between different organs. However, under hypoxia environment, the activation of lipolysis promotes excessive free fatty acids (FFA) release, which is taken up by the liver, contributing to ectopic lipid accumulation and pathogenesis of liver (Lefere et al., 2016). Adipose tissue dysfunction could lead to increased delivery of FFA and glycerol to the liver which drives hepatic gluconeogenesis and facilitates the accumulation of lipids and insulin signaling inhibiting lipid intermediates (Bosy‐Westphal et al., 2019). Herein, hypoxia caused lipid metabolism disorder of WAT may further influence liver function, leading to the maladaptation to high‐altitude environment and increasing the incidence of acute mountain sickness (AMS).
The endoplasmic reticulum (ER) is an organelle that functions to synthesize, fold, and transport proteins. It is also the site of triglyceride synthesis and nascent lipid droplet formation (Nettebrock & Bohnert, 2019). The sensing, metabolizing, and signaling mechanisms for lipid metabolism exist within or on the ER membrane domain (Balla et al., 2020). Dysregulation of ER homeostasis led to accumulation of misfolded proteins in the ER lumen and evoke ER stress (Henne, 2019). To reduce ER stress, the unfolded protein response (UPR) signal pathways are activated. Recently, accumulated evidence suggested that ER homeostasis and UPR activation play an important homeostatic role in lipid metabolism (Basseri & Austin, 2012; Mohan et al., 2019). As reported by Deng et al., ER stress could induce lipolysis by activating cAMP/PKA and ERK$\frac{1}{2}$ pathways (Deng et al., 2012). Previous study also found that burned patients displayed significant ER stress within adipose tissue and ER stress could augment lipolysis in cultured human adipocytes (Bogdanovic et al., 2015).
The disulfide bond formation during protein synthesis is independent of oxygen, however, the post‐translational protein folding and isomerization process is oxygen‐dependent (Koritzinsky et al., 2013). Herein, hypoxia exposure could induce extensive protein modification in the ER and result in the accumulation of misfolded/unfolded proteins, which activate UPR and evoke ER stress (Chipurupalli et al., 2019; Maekawa & Inagi, 2017). We decided to test the hypothesis that ER stress may modulate hypoxia‐induced WAT metabolic derangement and liver dysfunction based on the following evidence: [1] ER is one of the major sites of lipid metabolism. [ 2] lipid metabolism and function are sensitive to oxygen concentration. [ 3] Hypoxia could induce ER stress due to the accumulation of misfolded proteins (Xu et al., 2015; Yang et al., 2014). [ 4] ER stress is closely correlated with lipid metabolism dysfunction (Mohan et al., 2019). [ 5] lipid metabolism in WAT plays a critical role in the progression of liver dysfunction (Dong et al., 2020).
To address this issue, we investigated the effects of ER stress in hypoxia‐induced lipolysis using chemical chaperone 4‐PBA, antagonist of ER stress. The main objective of this study was to clarify the role of ER stress which regulates WAT lipolysis and liver lipid accumulation under continuous high‐altitude hypoxia exposure. An understanding of the interplay between tissues and these proposed mechanisms may provide novel therapeutic strategies for the treatment of the whole‐body metabolism dysfunction at high altitude.
## Animals care
Adult male Sprague–Dawley rats (280–330 g) were purchased from Weitong Lihua Laboratory Animal Limited Company. The rats were housed at room temperature (22°C–25°C) and in a 12–12 h light–dark cycle with free access to food and water and adapted to the condition above for 1 week before experiment. All experiments were conducted in strict accordance with the laboratory animal care guidelines published by the US National Institutes of Health (NIH publication no. 85–23, revised 1996). All protocols concerning animal use were approved by the Institutional Animal Care and Use Committee of Institute of Basic Medical Sciences, Peking Union Medical College and Capital Medical University.
## Hypoxic challenge
Hypoxia group rats were placed in a hypobaric chamber (Guizhou Fenglei Air Ordnance Co., Ltd.) and subjected to hypoxia mimicking an altitude of 5500 m for 10 days. The chamber was opened daily for 30 min to clean and replenish food and water and room temperature was kept at 20°C–22°C. We monitored the body weights of rats every day. 4‐PBA (P21005) was commercially purchased (Sigma‐Aldrich). Rats were randomly divided into four groups: [1] Control group, [2] Hypoxia group, [3] Control + 4‐PBA (30 mg/kg /day), and [4] Hypoxia + 4‐PBA (30 mg/kg/day). The dose of 4‐PBA was set based on previous reports (Luo et al., 2015; You et al., 2019; Zeng et al., 2017). All the rats were sacrificed by decapitation and serum was obtained by centrifugation and stored at −80°C. The perirenal fat pads were collected and weighed immediately, frozen in liquid nitrogen, and stored at −80°C.
## Histology staining
WAT and liver tissue were fixed in $4\%$ paraformaldehyde overnight, followed by embedment in paraffin and longitudinal slicing, with 4‐μm‐thick sections obtained for hematoxylin‐eosin (HE) staining. The stained slides were examined by microscopy for histomorphological analyses. A commercial terminal deoxynucleotidyl transferase‐mediated dUTP nick‐end labeling (TUNEL) kit (Roche) was employed to assess the degree of hepatic cell apoptosis. Histological alterations were assessed in randomly selected histological fields at ×400 magnification and apoptosis index (AI) was calculated.
## Western blotting and densitometry analyses
Homogenized rat WAT was lysed in 200 μl RIPA lysis buffer (Beyotime, P0013B) with $1\%$ phenylmethyl sulfurylfluoride and $4\%$ complete protease inhibitor cocktail mix (Roche). Extracts were centrifuged at 14,000 g for 15 min at 4°C. Eighty micrograms of total protein was used for sodium dodecyl sulfate‐polyacrylamide gel electrophoresis, followed by transferring blotting to nitrocellulose membrane (Millipore Corp., Billerica). Membranes were then blocked with $5\%$ non‐fat‐dried milk in PBS for 1 h with gentle shaking. Membranes were incubated first with primary antibodies (dilution: 1:1000) overnight at 4°C, in $1\%$ BSA in PBS overnight at 4°C with shaking. The following primary antibodies were purchased from Cell Signaling Technology: anti‐p‐HSL (#4139), anti‐HSL (#18381), anti‐pPKA, anti‐perilipin, anti‐Phospho‐PKA Substrate (RRXS*/T*) (#9624), anti‐GRP78 (#3183 S), anti‐CHOP (#2895P), anti‐protein kinase‐like eIF2α kinase (PERK) (#3192 S), and their phosphorylated species. anti‐ATGL antibody (ab109251), anti‐CGI58 antibody (ab111984), and anti‐β‐actin antibody (ab6276) were purchased from Abcam. Then, membranes were washed and incubated with secondary antibodies for 2 h at room temperature. Finally, the samples were visualized by enhanced chemiluminescence using Tanon‐410 automatic gel imaging system (Shanghai Tianneng Corporation). After scanning, band density was analyzed using Image J 1.33 software (National Institutes of Health).
## Reverse‐transcription PCR and quantitative real‐time PCR
Total RNA was prepared from frozen liver tissues with TRIZOL (Invitrogen) reagent and the cDNA was synthesized using TransScript TM First‐Strand cDNA Synthesis Super‐Mix (TransGen Biotech, AT301). The program was run on a S1000 Thermal Cycler. Quantitative real‐time PCR was performed using the SYBR®Pre‐mix Ex TaqTMkit (Takara, RR420A) and analyzed in a step‐one plus RT‐PCR system (life science, Applied Biosystems). The primer sequences are listed in Table 1.
**TABLE 1**
| Primer ID | Primer sequence 5′–3′ | Accession No. | Product size (bp) |
| --- | --- | --- | --- |
| CD36 | Fed: TCCTCGGATGGCTAGCTGATT | NC_051339.1 | 150 |
| CD36 | Rev: TGCTTTCTATGTGGCCTGGTT | NC_051339.1 | 150 |
| FABP1 | Fed: CTTCTCCGGCAAGTACCAAGT | NM_012556.2 | 162 |
| FABP1 | Rev: CATGCACGATTTCTGACACCC | NM_012556.2 | 162 |
| FABP4 | Fed: GTAGAAGGGGACTTGGTCGTC | NM_053365.2 | 234 |
| FABP4 | Rev: GCCTTTCATGACACATTCCAC | NM_053365.2 | 234 |
| β‐Actin | Fed: CGTTGACATCCGTAAAGACC | NM_031144.3 | 260 |
| β‐Actin | Rev: GCTAGGAGCCAGGGCAGTA | NM_031144.3 | 260 |
## Serum measurements
Serum levels of non‐esterified fatty acid (NEFA) and glycerol were measured using NEFA kit (A042, Jiancheng Biotechnology) and Glycerol Assay kit (F005‐1, Jiancheng Biotechnology), respectively. These assays were performed according to manufacturer's instructions. Serum levels of triglyceride (TG), total cholesterol (TC), high‐density lipoprotein cholesterol (HDL‐C), and low‐density lipoprotein cholesterol (LDL‐C) were measured by an automatic biochemical analyzer (Chemray 240, Rayto Life and Analytical Sciences).
Serum alanine (ALT), aspartate aminotransferase (AST), and alkaline phosphatase (ALP) microplate test kits were obtained from Nanjing Jiancheng Bioengineering Institute. These assays were performed as previously described (Wang et al., 2020). Briefly, ALT, AST, and ALP activities were evaluated at 37°C for 15 min by assessing for a decrease in absorbance at a wavelength of 510 nm, with Chemi Lab ALT, AST, and ALP assay kits, respectively.
## Statistical analysis
The data are presented as mean ± standard error (SE). For Western blot, protein levels were normalized to β‐actin. Statistical significance is determined by one‐way Analysis of variance (ANOVA) or nonparametric for more than three groups. p‐Value <.05 was considered statistically significant (SPSS 18.0 software).
## Hypoxia exposure induces endoplasmic reticulum stress in WAT
To investigate the role of ER stress in WAT under hypoxia treatment, we first examined the expression of ER stress markers, namely GRP78 and CHOP (Figure 1a). Under ER stress conditions, increased GRP78 is dissociated from unfolded proteins and activates ER stress receptors triggering the UPR. As shown in Figure 1b,c, hypoxia exposure significantly increased levels of GRP78 and CHOP. Continuous hypoxia treatment also activated ER stress‐related pathways in rat adipose tissue, evidenced by enhanced p‐PERK/PERK ratio (Figure 1d) and p‐IRE1α/IRE1α ratio (Figure 1e). 4‐PBA treatment significantly attenuated hypoxia‐induced ER stress, evidenced by decreased GRP78, CHOP, p‐PERK/PERK ratio, and p‐IRE1α/IRE1α ratio in 4‐PBA + hypoxia group as compared with hypoxia group.
**FIGURE 1:** *Hypoxia exposure induces endoplasmic reticulum stress in the WAT. The expression levels of ER stress‐related genes in WAT are shown. (a) GRP78, CHOP, p‐PERK, PERK, p‐IRE1α, and IRE1α protein expression levels; (b) Relative GRP78 protein expression levels; (c) relative CHOP protein expression levels; (d) p‐PERK/PERK ratio; (e) p‐IRE1α/IRE1α ratio. Data are shown as the mean ± SE of at least two independent western blots, *p < .05, **p < .01, and ***p < .001 (control group vs. hypoxia group, n = 6/group). #
p < .05, ##
p < .01 (hypoxia group vs. hypoxia + 4‐PBA group, n = 6/group)*
## PBA treatment attenuates enhanced lipolysis in WAT induced by hypoxia
Compared with control group, exposure to hypoxia equivalent to an altitude of 5500 m for 10 days significantly reduced the body weight of rat and wet weight of perirenal fat (Figure 2a,b). Both serum levels of glycerol and FFA significantly increased in hypoxia group rats, indicating enhanced lipolysis under hypoxia exposure (Figure 2c,d). In support of these findings, histological analysis of WAT showed that continuous hypoxia significantly reduced the volume of adipocytes compared with that in control group rats (Figure 2e,f). Hypoxia exposure led to increased serum levels of triglycerides (TG), low‐density lipoprotein cholesterol (LDL‐C), while the levels of total cholesterol (TC) level and high‐density lipoprotein cholesterol (HDL‐C) did not change significantly (Figure 2g–j).
**FIGURE 2:** *4‐PBA treatment attenuates enhanced lipolysis in white adipose tissue under hypoxia. 4‐PBA treatment (30 μg/kg body weight by intra‐peritoneal injection) significantly attenuates hypoxia‐induced body weight (a) and WAT loss (b); (c) serum levels of glycerol; (d) serum levels of FFA; (e) representative images of HE‐stained sections of WAT (magnification, 400×); (f) changes in the adipocytes volume in WAT. (g) Serum levels of TG (h); (i) serum levels of TG; (j) serum levels of HDL‐C; (J) serum levels of LDL‐C. Data are shown as mean ± SE, *p < .05, **p < .01, (control group vs. hypoxia group, n = 6/group). #p < .05, ##p < .01 (hypoxia group vs. hypoxia + 4‐PBA group, n = 6/group)*
To investigate the effect of inhibition of ER stress on WAT lipolysis under hypoxia, we first evaluated the body weight and wet weight of perirenal fat in hypoxia rats with or without 4‐PBA treatment. 4‐PBA significantly attenuated the reduction of body weight and wet weight of perirenal fat after 10 days exposure to hypoxia (Figure 2a,b). In addition, inhibition of ER stress via 4‐PBA was associated with a significant reduction of lipolysis, evidenced by a significant reduction in serum glycerol and FFA levels (Figure 2c,d). Moreover, 4‐PBA treatment significantly attenuated hypoxia caused reduction of adipocyte volume (Figure 2f). 4‐PBA treatment effectively attenuated hypoxia‐induced increased levels of TG (Figure 2j).
## ER stress inhibition ameliorate hypoxia‐induced WAT lipolysis via cAMP/PKA pathway
Endoplasmic reticulum stress has been suggested to trigger lipolysis in adipocytes. The lipolysis process is closely correlated with the production of cAMP and activation of cAMP‐dependent protein kinase A (PKA). In our study, hypoxia challenge significantly increased pPKA production (Figure 3a,b), which phosphorylates HSL and perilipin (Miyoshi et al., 2006; Sztalryd et al., 2003). The p‐HSL/HSL ratio (Figure 3c) and p‐Perilipin/Perilipin (Figure 3d) significantly increased in the hypoxic group, which were attenuated by 4‐PBA treatment. Although the abundance of ATGL remained unchanged in the WAT of the hypoxia rats, the level of CGI‐58 significantly increased in the hypoxia rats compared with the control rats (Figure 3e,f). Taken together, these data indicated that the inhibition of ER stress was shown to alleviate hypoxia‐induced lipolysis mainly by blocking the activation of cAMP‐PKA‐pHSL/Perilipin pathway.
**FIGURE 3:** *ER stress inhibition ameliorates WAT lipolysis in hypoxia rats via cAMP/PKA pathway. 4‐PBA treatment significantly downregulated expression levels of WAT lipolysis‐related genes induced by hypoxia. (a) pPKA, p‐HSL, HSL, p‐Peri, Perilipin, CGI‐58, and ATGL protein expression levels; (b) Relative pPKA protein expression levels; (c) p‐HSL/HSL ratio; (d) p‐Peri/Peri ratio; (e) Relative CGI‐58 protein expression levels; (f) Relative ATGL protein expression levels. Data are shown as the mean ± SE, *p < .05, **p < .01, and ***p < .001(control group vs. hypoxia group, n = 6/group). #p < .05, ##p < .01 (hypoxia group vs. hypoxia + 4‐PBA group, n = 6/group)*
## PBA treatment ameliorated hypoxia‐induced liver lipid transport and accumulation
Under continuous hypoxia exposure, increased delivery of free fatty acids (FFA) caused by enhanced lipolysis in WAT may contribute to the lipid accumulation in the liver. As shown in Figure 4a, levels of FFA content significantly increased in hypoxia group rat liver, which was attenuated by 4‐PBA treatment. Lipid uptake in the liver was regulated by many transporters, including cluster of differentiation (CD36), fatty acid binding protein 1(FABP1), and FABP4. mRNA levels of CD36, FABP1, and FABP4 that regulate the entry of fatty acids into hepatocyte, are generally upregulated to cope with increased circulation FFAs (Figure 4b–d).
**FIGURE 4:** *4‐PBA treatment ameliorates hypoxia‐induced liver lipid accumulation in the liver. (a) 4‐PBA treatment significantly attenuates hypoxia‐induced FFA accumulation in the liver. Relative mRNA levels of (b) CD36 (c) FABP1 and (d) FABP4. Data are shown as the mean ± SE, *p < .05, **p < .01, and ***p < .001(control group vs. hypoxia group, n = 6/group). #p < .05, ##p < .01 (hypoxia group vs. hypoxia + 4‐PBA group, n = 6/group)*
## PBA treatment ameliorated hypoxia‐induced live hepatic injury and apoptosis
Hypoxia‐induced liver lipid accumulation may further trigger the pathogenesis of liver injury, serum levels of liver enzyme were tested to confirm our speculation. As shown in Figure 5a–c, the hypoxia group rat exhibited a marked increase in the levels of AST, ALT, and ALP ($p \leq .05$), indicating potential liver injury. However, the hypoxia + 4‐PBA group significantly decreased the levels of AST and ALT ($p \leq .05$) when compared with hypoxia group, indicating that 4‐PBA inhibits hypoxia‐induced hepatocellular injury.
**FIGURE 5:** *4‐PBA ameliorates hypoxia‐induced hepatic injury and apoptosis. (a) Serum levels of AST in rats exposed to hypoxic (n = 6) or normoxic (n = 6) conditions. (b) Serum levels of ALT; (c) serum levels of ALP; (d)apoptosis index of four group rats; (e) representative images of TUNEL‐stained sections of liver (magnification, 400×). Data are shown as mean ± SE, *p < .05, **p < .01, (control group vs. hypoxia group). #p < .05, ##p < .01 (hypoxia group vs. hypoxia + 4‐PBA group)*
The apoptosis status of rat liver exposed to hypoxia was evaluated with a TUNEL assay. As shown in Figure 5d,e, the percentage of apoptotic cells was significantly increased in hypoxia group as compared with control group, which was effectively attenuated by 4‐PBA treatment.
## DISCUSSION
Lipid metabolism in white adipose tissue played an essential role in maintaining energy homeostasis at high‐altitude area. In this study, WAT ER stress‐mediated lipolysis is enhanced in a rat model of high‐altitude hypoxia. Moreover, we found that increased FFA release results in liver lipid accumulation and liver dysfunction, which was attenuated by the inhibition of ER stress using 4‐PBA.
As ER membrane are located with a variety of lipid metabolism‐related enzymes and ER is the major site of lipid metabolism, ER is involved in the control of metabolic homeostasis via regulating lipid metabolism. Under normal conditions, ER in the adipocyte functions to meet the demands of protein synthesis and secretion, triglyceride synthesis, nascent lipid droplet formation, and nutrient sensing. However, ER function is overwhelmed and the UPR is activated under stressful conditions (Menikdiwela et al., 2019; Sikkeland et al., 2019). Therefore, perturbations in ER homeostasis exerts a vital pathogenic mechanism in multi metabolic disorders of adipose tissue (Khan & Wang, 2014; Suzuki et al., 2017). Adverse stimuli like hypoxia may pose challenges to adipocyte and induce ER stress. In the present study, continuous hypoxia exposure evoked ER stress in adipose tissue, evidenced by increased GRP78, CHOP, p‐PERK, and p‐IRE1α expression in rat WAT. Our finding is in accordance with previous studies showing that hypoxia exposure induce ER stress in 3 T3‐F442A and 3 T3‐L1 adipocytes (Mihai & Schroder, 2015). UPR pathways were activated to ameliorate the overload of unfolded proteins under ER stress, which in turn influence lipid metabolism (Song et al., 2016). The activation of ER stress in adipose tissue may further induce lipolysis and elevated circulating FFAs (Song et al., 2017).
To confirm the potential role of the ER stress and UPR in the modulation of the lipolysis, we treated rat with 4‐PBA, an ER stress inhibitor. 4‐PBA treatment led to significant reduction in lipolysis, which blocked the phosphorylation of HSL and perilipin. As the results from upstream regulation, 4‐PBA treatment then effectively reduced glycerol and FFA release from adipose tissue, suggesting that ER stress‐mediated lipolysis mainly by regulating cAMP‐PKA/HSL under hypoxia. Similar to our study, enhanced lipolysis and ER stress occurred in the visceral WAT and inhibition of ER stress alleviated lipolysis in a rat model of chronic kidney disease(Zhu et al., 2014). In addition, curcumin was reported to suppress the ER stress‐mediated lipolysis via cAMP/PKA/HSL pathway (Wang et al., 2016). Deng et al., also reported that ER stress involved lipolysis through up‐regulation of GRP78 and activation of phosphorylation status of PERK and eIF2α in rat adipocytes (Deng et al., 2012).
Since the liver is the largest metabolic organ and regulates various physiological and metabolic processes, it also performs a key role in high‐altitude adaptation (Xu et al., 2019). Adipose dysfunction is closely associated with metabolism‐related liver diseases, an understanding of the interplay between tissues and these proposed mechanisms is still necessary (Da Silva Rosa et al., 2020). Accumulating data are pointing out the pathophysiological role of ectopic fat accumulation in different organs, including the liver (Bosy‐Westphal et al., 2019). In this study, the increased uptake of circulating lipids induced by WAT lipolysis significantly stimulated hepatic expression of lipid uptake and transport proteins CD36 and FABP4, which resulted in excess fatty acid uptake and lipid over accumulation in the liver. As a result, hypoxia‐treated rats displayed increased liver enzymes and hepatic apoptosis. As shown in Figure 6, 4‐PBA effectively attenuated hypoxia‐induced lipolysis via cAMP‐PKA‐HSL/perilipin pathway. The protective effect of 4‐PBA on liver injury and apoptosis, is likely resulting from decreased liver lipid accumulation via inhibiting FFA transport. Lines of evidence proved that excess FFA may modify the biology and function of hepatocyte and play an essential role in the pathogenesis liver dysfunction (Pereira et al., 2021). A high serum level of saturated FFAs is associated with hepatocyte lipo‐apoptosis (Takahara et al., 2017). In line with our fundings, Hubel, E., et al. found that repetitive Amiodarone treatment led to ER stress and aggravated lipolysis in adipose tissue while inducing a lipotoxic hepatic lipid environment and hepatic injury (Hubel et al., 2021).
**FIGURE 6:** *Protective mechanisms of 4‐PBA inhibit hypoxia‐induced lipolysis in WAT and lipid accumulation in the liver through regulating ER stress. Treatment with 4‐PBA, inhibitor of ER stress, effectively attenuated hypoxia‐induced lipolysis via cAMP‐PKA‐HSL/perilipin pathway. In addition, 4‐PBA treatment significantly attenuated liver injury and apoptosis, which is likely resulting from decreased liver lipid accumulation via inhibiting FFA transport.*
In conclusion, enhanced ER stress‐mediated WAT lipolysis was observed in a rat model of high‐altitude hypoxia, which contributes to hepatic dysfunction and apoptosis through excess release of FFA. Our findings highlight the vital role of 4‐PBA in WAT lipolysis and liver dysfunction via regulating ER stress, which may provide novel insights into systemic metabolic disturbances in high‐altitude area.
## CONFLICT OF INTEREST
The authors declare no conflict of interests.
## DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available on request from the corresponding author.
## References
1. Balla T., Sengupta N., Kim Y. J.. **Lipid synthesis and transport are coupled to regulate membrane lipid dynamics in the endoplasmic reticulum**. *Biochimica et Biophysica Acta – Molecular and Cell Biology of Lipids* (2020) **1865**. DOI: 10.1016/j.bbalip.2019.05.005
2. Basseri S., Austin R. C.. **Endoplasmic reticulum stress and lipid metabolism: Mechanisms and therapeutic potential**. *Biochemistry Research International* (2012) **2012**. DOI: 10.1155/2012/841362
3. Bogdanovic E., Kraus N., Patsouris D., Diao L., Wang V., Abdullahi A., Jeschke M. G.. **Endoplasmic reticulum stress in adipose tissue augments lipolysis**. *Journal of Cellular and Molecular Medicine* (2015) **19** 82-91. DOI: 10.1111/jcmm.12384
4. Bosy‐Westphal A., Braun W., Albrecht V., Muller M. J.. **Determinants of ectopic liver fat in metabolic disease**. *European Journal of Clinical Nutrition* (2019) **73** 209-214. DOI: 10.1038/s41430-018-0323-7
5. Chipurupalli S., Kannan E., Tergaonkar V., D'Andrea R., Robinson N.. **Hypoxia induced ER stress response as an adaptive mechanism in cancer**. *International Journal of Molecular Sciences* (2019) **20** 1-17. DOI: 10.3390/ijms20030749
6. da Silva Rosa S. C., Nayak N., Caymo A. M., Gordon J. W.. **Mechanisms of muscle insulin resistance and the cross‐talk with liver and adipose tissue**. *Physiological Reports* (2020) **8**. DOI: 10.14814/phy2.14607
7. Deng J., Liu S., Zou L., Xu C., Geng B., Xu G.. **Lipolysis response to endoplasmic reticulum stress in adipose cells**. *The Journal of Biological Chemistry* (2012) **287** 6240-6249. DOI: 10.1074/jbc.M111.299115
8. Dong M., Li L., Li G., Song J., Liu B., Liu X., Wang M.. **Mangiferin protects against alcoholic liver injury via suppression of inflammation‐induced adipose hyperlipolysis**. *Food & Function* (2020) **11** 8837-8851. DOI: 10.1039/d0fo01436b
9. Henne M.. **And three's a party: Lysosomes, lipid droplets, and the ER in lipid trafficking and cell homeostasis**. *Current Opinion in Cell Biology* (2019) **59** 40-49. DOI: 10.1016/j.ceb.2019.02.011
10. Hubel E., Fishman S., Holopainen M., Käkelä R., Shaffer O., Houri I., Shibolet O., Zvibel I.. **Repetitive amiodarone administration causes liver damage via adipose tissue ER stress‐dependent lipolysis, leading to hepatotoxic free fatty acid accumulation**. *American Journal of Physiology. Gastrointestinal and Liver Physiology* (2021) **321** G298-G307. DOI: 10.1152/ajpgi.00458.2020
11. Khan S., Wang C. H.. **ER stress in adipocytes and insulin resistance: Mechanisms and significance (review)**. *Molecular Medicine Reports* (2014) **10** 2234-2240. DOI: 10.3892/mmr.2014.2532
12. Koritzinsky M., Levitin F., Van den Beucken T., Rumantir R. A., Harding N. J., Chu K. C., Boutros P. C., Braakman I., Wouters B. G.. **Two phases of disulfide bond formation have differing requirements for oxygen**. *The Journal of Cell Biology* (2013) **203** 615-627. DOI: 10.1083/jcb.201307185
13. Lefere S., Van Steenkiste C., Verhelst X., Van Vlierberghe H., Devisscher L., Geerts A.. **Hypoxia‐regulated mechanisms in the pathogenesis of obesity and non‐alcoholic fatty liver disease**. *Cellular and Molecular Life Sciences* (2016) **73** 3419-3431. DOI: 10.1007/s00018-016-2222-1
14. Luo T., Chen B., Wang X.. **4‐PBA prevents pressure overload‐induced myocardial hypertrophy and interstitial fibrosis by attenuating endoplasmic reticulum stress**. *Chemico‐Biological Interactions* (2015) **242** 99-106. DOI: 10.1016/j.cbi.2015.09.025
15. Maekawa H., Inagi R.. **Stress signal network between hypoxia and ER stress in chronic kidney disease**. *Frontiers in Physiology* (2017) **8** 74. DOI: 10.3389/fphys.2017.00074
16. Menikdiwela K. R., Ramalingam L., Allen L., Scoggin S., Kalupahana N. S., Moustaid‐Moussa N.. **Angiotensin II increases endoplasmic reticulum stress in adipose tissue and adipocytes**. *Scientific Reports* (2019) **9** 8481. DOI: 10.1038/s41598-019-44834-8
17. Mihai A. D., Schroder M.. **Glucose starvation and hypoxia, but not the saturated fatty acid palmitic acid or cholesterol, activate the unfolded protein response in 3T3‐F442A and 3T3‐L1 adipocytes**. *Adipocytes* (2015) **4** 188-202. DOI: 10.4161/21623945.2014.989728
18. Miyoshi H., Souza S. C., Zhang H. H., Strissel K. J., Christoffolete M. A., Kovsan J., Rudich A., Kraemer F. B., Bianco A. C., Obin M. S., Greenberg A. S.. **Perilipin promotes hormone‐sensitive lipase‐mediated adipocyte lipolysis via phosphorylation‐dependent and ‐independent mechanisms**. *The Journal of Biological Chemistry* (2006) **281** 15837-15844. DOI: 10.1074/jbc.M601097200
19. Mohan S., PRM R., Brown L., Ayyappan P., RK G.. **Endoplasmic reticulum stress: A master regulator of metabolic syndrome**. *European Journal of Pharmacology* (2019) **860**. DOI: 10.1016/j.ejphar.2019.172553
20. Natarajan S. K., Rasineni K., Ganesan M., Feng D., McVicker B. L., McNiven M. A., Osna N. A., Mott J. L., Casey C. A., Kharbanda K. K.. **Structure, function and metabolism of hepatic and adipose tissue lipid droplets: Implications in alcoholic liver disease**. *Current Molecular Pharmacology* (2017) **10** 237-248. DOI: 10.2174/1874467208666150817111727
21. Nettebrock N. T., Bohnert M.. **Born this way – biogenesis of lipid droplets from specialized ER subdomains**. *Biochimica et Biophysica Acta ‐ Molecular and Cell Biology of Lipids* (2019) **1865** 158448. DOI: 10.1016/j.bbalip.2019.04.008
22. Pereira S., Moore J., Li J. X., Yu W. Q., Ghanim H., Vlavcheski F., Joseph Y. D., Dandona P., Volchuk A., Cummins C. L., Tsiani E., Giacca A.. **4‐Phenylbutyric acid improves free fatty acid‐induced hepatic insulin resistance in vivo**. *Endocrine Connections* (2021) **10** 861-872. DOI: 10.1530/EC-21-0248
23. Sikkeland J., Lindstad T., Nenseth H. Z., Dezitter X., Qu S., Muhumed R. M., Ertunc M. E., Gregor M. F., Saatcioglu F.. **Inflammation and ER stress differentially regulate STAMP2 expression and localization in adipocytes**. *Metabolism* (2019) **93** 75-85. DOI: 10.1016/j.metabol.2019.01.014
24. Song Y. F., Hogstrand C., Wei C. C., Wu K., Pan Y. X., Luo Z.. **Endoplasmic reticulum (ER) stress and cAMP/PKA pathway mediated Zn‐induced hepatic lipolysis**. *Environmental Pollution* (2017) **228** 256-264. DOI: 10.1016/j.envpol.2017.05.046
25. Song Y. F., Huang C., Shi X., Pan Y. X., Liu X., Luo Z.. **Endoplasmic reticulum stress and dysregulation of calcium homeostasis mediate Cu‐induced alteration in hepatic lipid metabolism of javelin goby**. *Aquatic Toxicology* (2016) **175** 20-29. DOI: 10.1016/j.aquatox.2016.03.003
26. Sun X., Tang Y., Tan X., Li Q., Zhong W., Sun X., Jia W., CJ M. C., Zhou Z.. **Activation of peroxisome proliferator‐activated receptor‐gamma by rosiglitazone improves lipid homeostasis at the adipose tissue‐liver axis in ethanol‐fed mice**. *American Journal of Physiology. Gastrointestinal and Liver Physiology* (2012) **302** G548-G557. DOI: 10.1152/ajpgi.00342.2011
27. Suzuki T., Gao J., Ishigaki Y., Kondo K., Sawada S., Izumi T., Uno K., Kaneko K., Tsukita S., Takahashi K., Asao A., Ishii N., Imai J., Yamada T., Oyadomari S., Katagiri H.. **ER stress protein CHOP mediates insulin resistance by modulating adipose tissue macrophage polarity**. *Cell Reports* (2017) **18** 2045-2057. DOI: 10.1016/j.celrep.2017.01.076
28. Sztalryd C., Xu G., Dorward H., Tansey J. T., Contreras J. A., Kimmel A. R., Londos C.. **Perilipin a is essential for the translocation of hormone‐sensitive lipase during lipolytic activation**. *The Journal of Cell Biology* (2003) **161** 1093-1103. DOI: 10.1083/jcb.200210169
29. Takahara I., Akazawa Y., Tabuchi M., Matsuda K., Miyaaki H., Kido Y., Kanda Y., Taura N., Ohnita K., Takeshima F., Sakai Y., Eguchi S., Nakashima M., Nakao K.. **Toyocamycin attenuates free fatty acid‐induced hepatic steatosis and apoptosis in cultured hepatocytes and ameliorates nonalcoholic fatty liver disease in mice**. *PLoS One* (2017) **12**. DOI: 10.1371/journal.pone.0170591
30. Wang L., Zhang B., Huang F., Liu B., Xie Y.. **Curcumin inhibits lipolysis via suppression of ER stress in adipose tissue and prevents hepatic insulin resistance**. *Journal of Lipid Research* (2016) **57** 1243-1255. DOI: 10.1194/jlr.M067397
31. Wang Y., Xiong Y., Zhang A., Zhao N., Zhang J., Zhao D., Yu Z., Xu N., Yin Y., Luan X., Xiong Y.. **Oligosaccharide attenuates aging‐related liver dysfunction by activating Nrf2 antioxidant signaling**. *Food Science & Nutrition* (2020) **8** 3872-3881. DOI: 10.1002/fsn3.1681
32. Xiong Y., Qu Z., Chen N., Gong H., Song M., Chen X., du J., Xu C.. **The local corticotropin‐releasing hormone receptor 2 signalling pathway partly mediates hypoxia‐induced increases in lipolysis via the cAMP‐protein kinase a signalling pathway in white adipose tissue**. *Molecular and Cellular Endocrinology* (2014) **392** 106-114. DOI: 10.1016/j.mce.2014.05.012
33. Xu L. H., Xie H., Shi Z. H., du L. D., Wing Y. K., Li A. M., Ke Y., Yung W. H.. **Critical role of endoplasmic reticulum stress in chronic intermittent hypoxia‐induced deficits in synaptic plasticity and long‐term memory**. *Antioxidants and Redox Signaling* (2015) **23** 695-710. DOI: 10.1089/ars.2014.6122
34. Xu Z., Jia Z., Shi J., Zhang Z., Gao X., Jia Q., Liu B., Liu J., Liu C., Zhao X., He K.. **Transcriptional profiling in the livers of rats after hypobaric hypoxia exposure**. *PeerJ* (2019) **7**. DOI: 10.7717/peerj.6499
35. Yang D., Gao L., Wang T., Qiao Z., Liang Y., Zhang P.. **Hypoxia triggers endothelial endoplasmic reticulum stress and apoptosis via induction of VLDL receptor**. *FEBS Letters* (2014) **588** 4448-4456. DOI: 10.1016/j.febslet.2014.09.046
36. You Y. D., Deng W. H., Guo W. Y., Zhao L., Mei F. C., Hong Y. P., Zhou Y., Yu J., Xu S., Wang W. X.. **4‐Phenylbutyric acid attenuates endoplasmic reticulum stress‐mediated intestinal epithelial cell apoptosis in rats with severe acute pancreatitis**. *Digestive Diseases and Sciences* (2019) **64** 1535-1547. DOI: 10.1007/s10620-018-5437-1
37. Zeng M., Sang W., Chen S., Chen R., Zhang H., Xue F., Li Z., Liu Y., Gong Y., Zhang H., Kong X.. **4‐PBA inhibits LPS‐induced inflammation through regulating ER stress and autophagy in acute lung injury models**. *Toxicology Letters* (2017) **271** 26-37. DOI: 10.1016/j.toxlet.2017.02.023
38. Zhu Y., Chen Y. L., Li C., Ding X. Y., Xu G. Y., Hu L. L., Hou F. F., Zhou Q. G.. **The effect of inhibition of endoplasmic reticulum stress on lipolysis in white adipose tissue in a rat model of chronic kidney disease**. *Acta Pharmacologica Sinica* (2014) **35** 356-362. DOI: 10.1038/aps.2013.177
|
---
title: Organic vegetable juice supplement alleviates hyperlipidemia in diet‐induced
obese mice and modulates microbial community in continuous colon simulation system
authors:
- Hyeon Ji Kim
- Sung Joon Mo
- Jisoo Kim
- Bora Nam
- Soo‐Dong Park
- Jae‐Jung Sim
- Jaehun Sim
- Jung‐Lyoul Lee
journal: Food Science & Nutrition
year: 2023
pmcid: PMC10002948
doi: 10.1002/fsn3.3193
license: CC BY 4.0
---
# Organic vegetable juice supplement alleviates hyperlipidemia in diet‐induced obese mice and modulates microbial community in continuous colon simulation system
## Abstract
In this study, we investigated the effects of organic vegetable juice (OVJ) supplementation on modulating the microbial community, and how its consumption ameliorated blood‐lipid profiles in diet‐induced obese mice. Here, we studied the alleviating effect of hyperlipidemia via animal experiments using diet‐induced obese mice and analyzed the effect of OVJ on the microbial community in continuous colon simulation system. OVJ consumption did not have a significant effect on weight loss but helped reduce the weight of the epididymis fat tissue and adipocytes. Additionally, blood‐lipid profiles, such as triglyceride, high‐density lipoprotein, and glucose, were improved in the OVJ‐fed group. Expression levels of genes related to lipid synthesis, including SREBP‐1, PPARγ, C/EBPα, and FAS, were significantly decreased. In addition, OVJ treatment significantly reduced inflammatory cytokines and oxidative stress. OVJ supplement influenced intestinal bacterial composition from phylum to genus level, including decreased Proteobacteria in the ascending colon in the phylum. At the family level, Akkermansia, which are associated with obesity, were significantly augmented in the transverse colon and descending colon compared to the control juice group. In addition, treatment with OVJ affected predicted lipid‐metabolism‐function genes related to lipid synthesis. These results suggest that OVJ supplementation may modulate gut microbial community and reduce the potential symptom of hyperlipidemia in diet‐obese mice.
The antihyperlipidemic effect of organic vegetable juice (OVJ) was studied through animal experiments using diet‐induced obese mice, and the effect of OVJ treatment on the microbiome was analyzed in a continuous colon simulation system. Supplementation of OVJ may alleviate hyperlipidemia by improving blood lipid profile and antioxidant effects in obese mice, and alter the microbiome to affect predicted lipid metabolism function genes involved in lipid synthesis. These results suggest that OVJ supplementation may modulate the gut microbiome and reduce the potential role of hyperlipidemia.
## INTRODUCTION
Vegetables are a major source of phytochemicals with potential beneficial health properties (Nuutila et al., 2003; Singh et al., 2009). Among the major phytochemicals of vegetables are polyphenols and carotenoids. Polyphenols have been studied for their antioxidant properties of preventing the damage caused by reactive oxygen species (ROS), such as hydroxyl radicals, hydrogen peroxide (H2O2), and superoxide (Williams et al., 2013). Carotenoid compounds have been studied for their anti‐obesity and anti‐inflammatory effects (González‐Castejón & Rodriguez‐Casado, 2011). Intake of vegetables prevented damage by oxidative stress and improved hyperlipidemia according to previous studies. The relationship between lipid levels in the blood and the level of antioxidants from vegetables has been recently suggested (Kim et al., 2008; Yang et al., 2008).
Obesity occurs because of many factors, such as energy imbalance, neurosecretion factors, and environmental factors. Recently, the number of obese patients has sharply increased due to westernized diet, stress, and lack of exercise, and obesity has been a major health problem worldwide (Jeung et al., 2019). Obesity causes various diseases, such as metabolic syndrome, cholelithiasis, cardiovascular disease, hypertension, diabetes, and hyperlipidemia (Bray, 2000). Hyperlipidemia refers to a condition that causes inflammation due to the presence of more fatty substances than necessary in the blood (Williams et al., 2013). Although hyperlipidemia does not have any specific symptoms, it is a risk factor for high blood pressure, arteriosclerosis, and stroke, and significantly increases the mortality rate in cardiovascular diseases (Klop et al., 2013; Nelson, 2013).
The human gut microbiota is composed of more than 1 million bacteria and a complex community of over 100 trillion diverse bacteria (Doré & Blottière, 2015; Graf et al., 2015). The gastrointestinal tract (GIT) includes a diverse and complex microorganism community, which are major contributors to human health (Mangal et al., 2017). The human digestive system has no digestive enzymes for plant‐derived complex carbohydrates, but the human gut microbiota can decompose and utilize them (Graf et al., 2015). In addition, gut microbiota plays a role in producing organic acid and short‐chain fatty acids (SCFAs) including propionate, butyrate, and acetate in the human GIT. These SCFAs are known to affect and regulate the microbial composition of the human intestinal microbiota (Den Besten et al., 2013). For these reasons, studies of human intestinal microbiome using genomic and metagenomic analyses on various topics have been actively conducted in recent years.
The in vivo analysis of human or animal gut is an ideal method to investigate intestinal microbiota, but these methods have technical and ethical difficulties, including expensive experimental cost, long time consumption, and difficulty in standardization due to limitations in controlling individual diets (Cha et al., 2018). In addition, in vivo studies are limited to fecal samples, which do not provide information about dynamic microbial changes in the fermentation site of the gut (Sousa et al., 2008). Furthermore, the gut microbiome of in vivo is often impaired due to inter‐individual differences associated with numerous factors such as age, gender, diet, geography, genetic background, and antibiotic use (Williams et al., 2015). To resolve these problems, an in vitro gut fermentation model has been developed and characterized by applying simple batch culture conditions and using a more complex apparatus for human fecal samples to control pH, temperature, and anaerobic conditions. A simple culture system, such as the TNO in vitro model, replicates the proximal colon in a single‐segment fermenter, whereas the three‐stage continuous system replicates the whole large intestine (Cinquin et al., 2006; Feria‐Gervasio et al., 2014).
Batch colon simulation models are generally closed systems, and these models are used with sealed vessels containing suspensions of fecal samples and culture system medium under anaerobic conditions. Batch culture systems have the advantage of being easy to set up and useful for fermentation, but these systems create a short time frame for fermentation research and are unable to control microorganisms. In contrast, continuous culture systems are open systems, where the fresh medium is injected and waste is released periodically. These systems simulate the major parts of the large intestine, including the ascending colon (AC), transverse colon (TC), and descending colon (DC). Continuous colon simulation systems are well‐controlled environmental parameters, which enable the detection of changes in metabolites and microbial composition in each part of the large intestine (Adamberg et al., 2014; Costabile et al., 2015; Maccaferri et al., 2012). Therefore, continuous colon simulation systems are more similar to the human GIT compared to the batch model.
This study investigated whether organic vegetable juice (OVJ) can modulate large intestinal microbiota and affect the predicted lipid metabolism function genes using continuous colon simulation systems. Based on the results of continuous simulation systems, we studied the effects of OVJ on the blood‐lipid levels, liver gene expression, and adipocytes in the epididymal fat of diet‐induced obese mice. In addition, we investigated the effects of OVJ on the antioxidant levels of diet‐induced obese mice.
## Animal experiments
Figure 1a shows the design of the in vivo experiments performed in this study. Six‐week‐old male C57BL/6 was purchased from Laonbio. All mice were fed at constant humidity (55 ± $10\%$) and temperature (22 ± 1°C) with a 12 h light/dark cycle. After 7 days of acclimatization, mice were fed a normal diet group ($$n = 10$$; AIN‐93G), high‐fat diet (HFD) group ($$n = 10$$; Rodent diet with 60 Kcal% fat), and high‐fat with organic vegetable juice (HFD‐OVJ) ($$n = 10$$, group) for 9 weeks. The composition of the ND was formulated based on the AIN‐93G purified rodent diet. In this study, the recommended daily intake of OVJ obtained from hy Co., Ltd. was applied for mice to examine the health effects of OVJ. Table S1 shows the configuration of the OVJ. The OVJ was freeze‐dried, evenly mixed with a HFD in powder form, and then supplied to mice in the HFD‐OVJ group. Body weight and food intake were measured weekly. The food efficiency ratio (FER) was calculated by applying the equation: Blood samples were taken from the inferior vena cava and immediately placed at room temperature (20–23°C) and then centrifuged at 3000 g at 4°C for 10 min. The serum was separated from the blood sample. The harvested serum was stored at −80°C until analysis. The liver and epididymis fat tissues were collected, rinsed with sterilized PBS, and weighed. The partial liver tissue was stored in a deep freezer at −80°C immediately after collection for gene expression analysis using real‐time PCR. The animal experimental plan was approved by the Ethics Committee at the R&D Center, hy Co., Ltd. (AEC‐2020‐00003‐Y).
**FIGURE 1:** *Schematic of experimental design. (a) Six‐week‐old male C57BL/6 mice were randomly assigned to either Normal (n = 10), HFD (n = 10), and HFD‐OVJ (n = 10) groups. Each group were fed the following diet for 9 weeks: Normal group, rodent diet; HFD group, HFD (rodent diet with 60 kcal% fat); HFD‐OVJ group, HFD with 2.5% OVJ. (b) Schematic representation of the continuous colon simulation system. AC, ascending colon; DC, descending colon; HFD, high‐fat diet; HFD‐OVJ, organic vegetable juice with high‐fat diet; TC, transverse colon*
## Blood biochemistry analysis
Serum samples were collected at T&P Bio for blood analysis. The serum total cholesterol (T‐CHOL), triglyceride (TG), high‐density lipoprotein cholesterol (HDL), low‐density lipoprotein cholesterol (LDL), glucose (GLU), aspartate transaminase (AST), and alanine transaminase (ALT) levels were determined using Beckman Coulter AU480 analyzer (Beckman Coulter Inc.). Serum samples remaining after the blood biochemistry tests were used to measure the following antioxidant biomarkers: (a) 8‐hydroxy‐2‐deoxyguanosine (8‐OHdG) ELISA Kit (Abcam) was used. ( b) malondialdehyde (MDA) concentration in serum samples was measured using a lipid peroxidation Assay Kit (Abcam). ( c) H2O2 level was measured using a Catalase Assay Kit (Colorimetric/Fluorometer) (Abcam).
## Histological tissue analysis
The liver and epididymis fat tissues were washed with sterilized PBS and fixed in $10\%$ formalin. Tissue samples were obtained from T&P Bio for histological analysis. Fixed tissues were implanted in paraffin for hematoxylin and eosin staining. The liver and epididymis fat tissues were observed under a fluorescence microscope (Axiovert 200M, Carl Zeiss) at a magnification of ×20 and ×10, respectively.
## RNA extraction and gene expression analysis
Total tissue RNA was extracted from liver tissues using the easy spin Total RNA Extraction Kit (iNtRON) via bead‐beating. The liver tissues were mixed with 1 ml lysis buffer and transferred into lysing matrix tubes, containing specialized beads (MP Biomedicals), and pulverized through Fastprep24. After bead‐beating, the remaining procedure of easy spin Total RNA Extraction Kit was followed. Consequently, total RNA was eluted with 50 μl elution buffer. Total RNA samples were quantified using Nanodrop and stored at 20°C until gene expression analysis. The extracted total RNA was reverse‐transcribed into cDNA using an Omniscript RT Kit (Qiagen). Reverse transcription PCR conditions were set at 37°C for 60 min. The cDNA was amplified using a QuantStudio 6 Flex‐Real Time Instrument with a gene expression master mix. Real‐time PCR was performed using mouse‐specific TaqMan Gene Expression Assays and normalized by the expression of GAPDH (Applied Biosystems). Table S2 shows the catalog numbers of the genes and names.
## The continuous colon simulation system
Figure 1b shows the workflow of the continuous colon simulation system performed in this study. Fecal samples were collected from three healthy adults who had not taken antibiotics for 3 months. Fecal samples were diluted with phosphate‐buffered saline (PBS) at a 1:10 (v/v) ratio in an anaerobic chamber. The fecal slurry sample was separated into each vessel at a final concentration of $2\%$. The continuous colon simulation system medium contained 1 g/L peptone, 4 g/L mucin, 0.5 g/L l‐cystein‐HCl, 1 g/L xylan, 0.5 g/L inulin, 0.4 g/L bile salt, 0.0025 g/L resazurin, 3 g/L yeast extract, 0.4 g/L d‐glucose, 2 g/L pectin, 1 g/L arabinogalactan, and 3 g/L starch. To simulate the environment of the large intestine, conditions were created for the three major parts of the large intestine; AC, TC, and DC. The volume and pH were as follows: 300 ml, pH 5.5 for AC; 400 ml, pH 6.2 for TC; 325 ml, pH 6.8 for DC (Cha et al., 2018). To maintain the anaerobic conditions in the simulation system, nitrogen gas (N2) was constantly injected at a flow rate of 10 ml/min. The temperature of each vessel was maintained at 37°C and the pH was automatically adjusted with 1 N HCl and 1 N NaOH. In addition, the pump on/off time interval was controlled to 12.5 ml/h to maintain a continuous flow. The colon system was pre‐run to achieve chemical and microbial stabilization for 2 weeks. After the stable step, the sample was treated in the AC and the washout step proceeded for 2 weeks.
## Juice treatment and fecal slurry sampling from the culture system
Organic vegetable juice was obtained from the hy Co., Ltd. Control juice was diluted with water and mixed carbohydrates (glucose 4.44 g/200 ml, fructose 4.52 g/200 ml, and sucrose 7.84 g/200 ml). The amount of these monosaccharides and disaccharides was the same as the OVJ. Two hundred microliters OVJ and control juice were injected into the AC of the continuous colon simulation system per day for 2 weeks. The same amount of sample of fecal slurry from the colon simulation system was carried out at a constant time point at the same amount in each vessel. Sampling was performed a total of 8 times, once a day from 2 days before the end of the stable period, 4 days before the end of the juice treatment period, and 2 days before the end of the washout period. These fecal slurries were used to analyze microbial community.
## Bacterial 16S rRNA amplification for next‐generation sequencing
The bioinformatics analysis of fecal DNA samples from the colon simulation system was carried out at Chunlab. Total fecal DNA samples were extracted using the QIAamp DNA Stool Mini Kit (Qiagen). PCR amplification of 16S rRNA sequences was conducted to prepare DNA sequencing templates. The V3–V4 region of the 16S rRNA sequence was amplified using the 341‐forward (341F)/805‐reverse (805R) primer set (341F: 5′‐TCGTCGGCAGCGTCAGATGTGTATAAGAGACAGCCTACGGGNGGCWGCAG‐3′; 805R: 5′‐GTCTCGTGGGCTCGGAGATGTGTATAAGAGACAGGACTACHVGGGTATCTAATCC‐3′). PCR conditions for the 341F/805R primer set were as follows: initial denaturation at 95°C for 3 min; 25 cycles of denaturation at 95°C for 30 s; annealing at 55°C for 30 s; extension at 72°C for 30 s; final elongation at 72°C for 5 min. After amplification of the 16S rRNA gene, the 16S rRNA amplicon library was used as a template DNA for next‐generation sequencing (NGS). Illumina MiSeq sequencing platform (Illumina) was used for NGS.
## Bioinformatics analysis
The microbiome taxonomic profiling was analyzed using the EzBioCloud database provided by Chunlab. The raw data were analyzed using the QIIME 1.9.1 package program. The sequencing data were filtered for low‐quality reads and mismatched indexes using “Trim sequence” during quality control. High‐quality sequences of 300 bp were extracted using open reference clustering, one of several operational taxonomic unit clustering (OTU clustering) methods. After OTU clustering, Chimera checking was performed using ChimerSlayer. β‐diversity analysis was performed using unweighted UniFrac distances and visualized based on unweighted PCoAs. Linear discriminant analysis effect size (LEfSe) analysis (Segata et al., 2011) and PICRUSt platform were used for metagenome prediction (Langille et al., 2013). The data were analyzed using R version 3.6.2 (https://www.r‐project.org). All datasets have been deposited in NCBI Gene Expression Omnibus with the accession code PRJNA720297 and GSE171609. All data were expressed as mean ± standard error (SE).
## Statistical analysis
All data in this study were presented as the mean ± SE. For blood analysis, gene expression data analysis, and metabolic analysis, differences between groups (Normal vs. HFD, HFD vs. HFD‐OVJ) were evaluated using unpaired Student's t‐tests. $p \leq .05$ was considered statistically significant.
## Effects of organic vegetable juice on body weight, food efficiency ratio, epididymal fat and liver mass, and the size of adipocytes in diet‐induced obese mice
Body weight of the HFD group was significantly increased after only 1 week of the HFD. Final body weight was $30\%$ higher in the HFD group than in the normal group ($p \leq .05$). Body weight of the HFD‐OVJ group was slightly lower than that of the HFD group, but the difference was not significant (Figure 2a). After 9 weeks, the weight gain in the HFD group was significantly increased compared to the normal group ($p \leq .001$). Weight gain of mice in the HFD‐OVJ group was significantly lower than that of the HFD group ($p \leq .05$) (Figure 2b). The food efficiency ratio of the HFD group was significantly higher than that of the normal group ($p \leq .001$), but was significantly decreased when the HFD was supplemented with OVJ ($p \leq .05$) (Figure 2c).
**FIGURE 2:** *Effects of OVJ treatment on diet‐induced obese mice. (a) Change in body weight. (b) Body weight gain. (c) Food efficiency ratio. (d) Total liver mass. (e) Epididymal fat mas. Results are presented as the mean ± SE. Significant differences are indicated as #
p < .05, ##
p < .01, and ###
p < .001 when compared with the normal group. *p < .05, **p < .01, ***p < .001 when compared with the HFD group. HFD, high‐fat diet; HFD‐OVJ, organic vegetable juice with high‐fat diet*
The HFD group showed increased liver mass compared to normal group ($p \leq .001$). The liver mass of the HFD‐OVJ group was slightly lower than that of the HFD group, but there was no significant difference (Figure 2d). The epididymal fat mass of the HFD group was also increased by $78.5\%$ compared to that in the normal group ($p \leq .001$). The weight of epididymal fat in the HFD‐OVJ group was significantly lower than that in the HFD group ($p \leq .01$) (Figure 2e).
The adipocyte size in the liver and epididymal fat was measured using histological tissue analysis. The epididymal fat adipocyte was markedly enlarged in HFD group compared with the normal group. However, adipose tissue was reduced in the HFD‐OVJ group (Figure 3a). The degree of hepatic steatosis in the HFD group developed compared with the Normal group, but mice fed OVJ showed reduced steatosis compared to the HFD group (Figure 3b). The adipocyte area of epididymal fat showed significant reductions in the HFD‐OVJ group compared to the HFD group ($p \leq .01$) (Figure 3c). Intake of OVJ decreased epididymal fat mass and adipocyte formation in diet‐induced obese mice.
**FIGURE 3:** *Effects of OVJ treatment on liver and epididymal fat of diet‐induced obese mice. (a) Morphology of epididymal fat, and (b) liver was analyzed using a microscope. (c) Adipocytes of epididymal fat. Effects of fruit‐vegetable drink on serum biochemistry and lipid and cholesterol levels of diet‐induced obese mice. (d) Serum biochemistry. (e) Lipid and cholesterol levels in serum. Results are presented as the mean ± SE. #
p < .05, ##
p < .01, and ###
p < .001 when compared with the normal group, *p < .05 **p < .01 when compared with the HFD group. ALT, alanine transferase; AST, aspartate transferase; GLU, glucose; HDL, high‐density lipoprotein cholesterol; HFD, high‐fat diet; HFD‐OVJ, organic vegetable juice with high‐fat diet; LDL, low‐density lipoprotein cholesterol; T‐CHOL, Total cholesterol; TG, triglyceride*
## Effects of OVJ on serum biochemistry and lipid and cholesterol levels in diet‐induced obese mice
The levels of liver toxicity biomarkers AST and ALT were increased in the HFD group compared with the normal group, but only statistically significant for ALT ($p \leq .01$). AST and ALT levels were significantly reduced in the HFD‐OVJ group ($p \leq .05$) (Figure 3d). GLU levels were increased in the HFD group compared to the normal group ($p \leq .01$) and significantly decreased in the HFD‐OVJ group compared to the HFD group ($p \leq .05$). T‐CHOL and LDL were both increased in high‐fat diet‐induced obese mice ($p \leq .05$ and $p \leq .01$, respectively). The levels of T‐CHOL and LDL in the HFD‐OVJ group were similar to those in the HFD mice. A high‐fat diet increased TG levels in the serum to 211.83 ± 13.88 mg/dl compared with 145.43 ± 15.54 mg/dl in the normal group ($p \leq .01$). The level of TG in the HFD‐OVJ group (204.50 ± 4.19 mg/dl) was slightly lower than that in the HFD group, but the difference was not significant. The levels of HDL in the HFD and HFD‐OVJ groups were both higher than those in the normal group; especially, the HDL level of HFD‐OVJ fed mice was significantly increased compared to that of the HFD group ($p \leq .01$) (Figure 3e).
## Effects of OVJ on lipid synthesis and inflammatory gene expression in the liver tissue of diet‐induced obese mice
We examined gene expression related to lipid synthesis in liver tissues. The HFD group showed increased expression of genes involved in the regulation of sterol regulatory element‐binding protein 1 (SREBP‐1), peroxisome proliferator‐activated receptor (PPARγ), CCAAT/enhancer‐binding protein α (C/EBPα) and fatty acid synthesis (FAS) (Figure 4a). The expression of SREBP‐1 and FAS was significantly increased in HFD‐fed mice ($p \leq .001$ and $p \leq .05$, respectively). Lipid synthesis‐related gene expression, including FAS, SREBP1, PPARγ, and C/EBPα was significantly lower in the HFD‐OVJ group than in the HFD group (FAS and SREBP‐1, $p \leq .001$; PPARγ and C/EBPα, $p \leq .05$). We measured the mRNA levels of genes related to inflammation (Figure 4b). IL‐6 expression was significantly higher in the HFD group compared to that in the normal group ($p \leq .05$). In addition, HFD‐OVJ intake decreased the expression of SREBP‐1, PPARγ, and C/EBPα, and IL‐6 compared to HFD group.
**FIGURE 4:** *HFD and OVJ effect on the gene expression in the liver. (a) FAS, SREBP‐1, PPARγ, and C/EBPα, and (b) IL‐6. Results are presented as the mean ± SE. #
p < .05 and ###
p < .001 when compared with the normal group, *p < .05 ***p < .001 when compared with the HFD group. HFD, high‐fat diet; HFD‐OVJ, organic vegetable juice with high‐fat diet*
## Antioxidant effects of OVJ in diet‐induced obese mice
8‐OHdG causes oxidative DNA damage via ROS. Therefore, 8‐OHdG is an established biomarker of oxidative stress. A high‐fat diet significantly increased 8‐OHdG in the serum compared with the normal group ($p \leq .05$). The level of 8‐OHdG in the HFD‐OVJ group was significantly lower than that in the HFD group ($p \leq .05$). In addition, the intake of OVJ in diet‐induced obese mice reduced 8‐OHdG activity to the level of the normal group (Figure 5a). MDA was used as a marker of lipid peroxidation by oxidative degradation of lipids. MDA concentration in the HFD group was significantly higher than that in the normal group ($p \leq .01$), while the MDA level in the HFD‐OVJ group was significantly lower than that in the HFD group ($p \leq .01$) (Figure 5b). H2O2 level was significantly elevated in the HFD group compared to the normal group ($p \leq .05$), but slightly reduced in the HFD‐OVJ group; there was no significant difference (Figure 5c).
**FIGURE 5:** *Antioxidant effects of OVJ. (a) 8‐OHdG, (b) MDA, and (c) H2O2 in diet‐induced obese mice. Results are presented as the mean ± SE. #
p < .05 and ##
p < .01 when compared with the normal group, *p < .05, **p < .01 when compared with the HFD group. HFD, high‐fat diet; HFD‐OVJ, organic vegetable juice with high‐fat diet*
## Overview and diversity analysis of the microbial community in the culture system
The microbial composition of the continuous colon simulation was analyzed using the Illumina MiSeq platform, targeting the variable region from V3 to V4 of the bacterial 16S rRNA gene with the 341F/805R primer set. After low‐quality sequence filtering, chimera checking, and OTU clustering, we confirmed an average of 190 OTUs with $97\%$ sequence identity. We obtained an average of 35,512 high‐quality sequences per sample and classified reads were obtained. Alpha diversity, such as the Shannon diversity index, was used to measure the biodiversity and richness of all groups. Table 1 shows the alpha diversity of AC, TC, and DC for each control juice group and OVJ group. When control juice and organic juice treatment were processed, the average alpha‐diversity was identified as 2.66 and 2.79, respectively. These results indicated that the microbial community after OVJ treatment was slightly more diverse compared to that after control juice treatment. The Shannon index was significantly increased in the TC and DC after organic juice treatment ($p \leq .01$ and $p \leq .05$, respectively). After treatment with control juice, the Shannon diversity was also significantly increased in the TC, but there was no significant difference in the DC compared to the AC. The beta‐diversity showed by principal coordinates analysis, including classified OTUs, confirmed a difference in the microbial composition as a result of colon simulation parts after organic juice and control juice treatment (Figure 6a). When control juice and organic juice were added, the composition of the intestinal flora was changed according to different parts of the colon and sample type. The AC showed a similar beta‐diversity in the control juice and organic juice groups. Beta‐diversity of the transverse and descending parts showed alterations in the microbial community caused by control juice and organic juice treatment.
## Effects of OVJ on the microbial community in the culture system
The microbiota of the OVJ and control juice‐treated groups was analyzed at the phylum level (Figure S1A, Table S3). We analyzed four major phyla; Actinobacteria, Bacteroidetes, Firmicutes, and Proteobacteria. Bacteroidetes and Firmicutes occupied most of the microbial composition in each group (Figure 6b). Bacteroidetes were significantly increased in the TC and DC of OVJ group compared with the control juice group ($p \leq .001$ and $p \leq .05$, respectively). Firmicutes abundance was significantly lower in the TC and DC of the OVJ group than in those of the control juice treatment but significantly increased in the AC compared with the control juice treatment ($p \leq .05$). Proteobacteria abundance was reduced significantly in the AC of the OVJ‐treated group compared to the control juice group ($p \leq .01$).
At the family level, the top 30 species obtained from the colon simulation system were analyzed (Figure S1B). The abundance of Bacteroidaceae was significantly higher in TC and DC of OVJ group than in those of the control juice group ($p \leq .001$ and $p \leq .01$, respectively). Bacteroidaceae belongs to Bacteroidetes, and the abundance of these species was decreased in the intestines of obese people (Figure 6c). The abundance of this species was higher in the TC and DC of OVJ group than in those of the control juice group. In particular, there was a significant difference in DC ($p \leq .05$) (Figure 6d).
We compared changes in the relative abundance of the top 40 at the genus level (Figure S1C). The abundance of Holdemania was significantly higher in TC and DC of control juice group than in those of the OVJ‐treated group ($p \leq .05$) (Figure 6e). Butyricimonas occupied a very small percentage of the OVJ‐treated group, but this species did not exist in the control juice group. There was a significant difference in DC ($p \leq .05$) (Figure 6f). Lachnospira abundance was significantly increased in the AC of the OVJ group compared to the control juice group ($p \leq .05$) and was higher in the TC and DC of the control juice‐treated group (Figure 6g).
The LDA score was used to confirm the significant difference in the microbial composition in the colon section. There was a significant difference in microbial taxa level with a log LDA score above 3.0 for each colon section. In the AC of the OVJ treatment group, Lactobacillus abundance was significantly higher than in that of the other groups. Bacteroidales, Frisingicoccus, Lachnospira, and Alistipes abundance was significantly higher in the TC of the OVJ group than in that of the control group. Clostridium_g24, Clostridium_g35, Ruminococcus_g4, and Agathobaculum abundance were significantly increased in the DC of the OVJ group. Escherichia abundance was significantly higher in the AC of the control juice treatment group than in that of the OVJ group. Fusicatenibacter, Faecalicatena, and Anaerostipes abundance were significantly increased in the TC of the control juice group. In DC of control group, Eubacterium, Hungatella, Elsenbergiella, and Blautia abundance were significantly higher than in the other groups (Figure 7a).
**FIGURE 7:** *Effects of OVJ on the bacterial abundance and metabolic pathway of continuous colon simulation system. (a) Cladogram based on the relative abundance of microbial composition using LEfSe. (b) Heatmap showing the relative abundance of microorganism at the family level for each group. (c) Heatmap representing PICRUSt analysis with KEGG pathway of each group. AC, ascending colon; Con, control juice; DC, descending colon; OVJ, organic vegetable juice; TC, transverse colon*
At the family level, 28 bacterial genera showed differences among different groups. The abundance of Akkermansiaceae was increased in the TC and DC of the OVJ group compared to the control juice group (Figure 7b). We predicted the metagenome function of each taxonomy group using the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways. We investigated the predicted lipid metabolism function genes of the microbiome in OVJ and control juice groups with PICRUSt (Figure 7c). OVJ significantly changed the microbial functions, such as primary and secondary bile acid biosynthesis (LDA score: 4.13, 4.22, respectively). However, in the control juice group fatty acid biosynthesis (LDA score: 4.02), fatty acid metabolism (LDA score: 5.18), and lipid metabolism (LDA score: 4.83) were significantly shifted (Figure S2).
## DISCUSSION
Obesity causes hyperlipidemia, which increases mortality by causing cardiovascular disease when the degree is severe. Therefore, it is necessary to lower the risk of cardiovascular disease by regulating blood lipids (Klop et al., 2013; Nelson, 2013). The human gut microbiota has been a major research topic in human health. A growing number of studies have suggested that the gastrointestinal microbiome is not only important for gut health but also for diseases such as obesity, diabetes, and atopy (Larsen et al., 2010; Penders et al., 2007; Zhao, 2013). Vegetables are a functional source that reduces the risk of diseases, such as cancer, cardiovascular disease, as well as aging (Liu, 2003). Major phytochemicals in the OVJ used in this study were β‐carotene and lycopene (β‐carotene; 113.1 ± 7.6, lycopene; 30.8 ± 0.9). β‐carotene and lycopene are carotenoids that have anti‐obesity properties (González‐Castejón & Rodriguez‐Casado, 2011). Here, we aimed to study the alleviating effect of hyperlipidemia effect of OVJ through in vivo experiment. Furthermore, we analyzed the gut microbiome using a continuous colon simulation system and investigated the correlation between the community and the metabolic pathways related to lipid synthesis predicted gene functions through OVJ consumption.
A HFD with OVJ slightly reduced body and liver weight gain and significantly decreased epididymal fat gain. Adipocytes of epididymal fat were also decreased, indicating that OVJ supplement can reduce adipocyte size in the epididymal fat of obese mice (Duwaerts et al., 2017). Serum lipid and cholesterol levels of diet‐induced obese mice tended to increase, which are correlated with hyperlipidemia (Neyrinck et al., 2013; Zhang et al., 2007). OVJ decreased GLU and TG levels and increased blood HDL. When GLU concentration increases, insulin ratio increases, and the excess glucose is stored in the subcutaneous fat, causing hyperlipidemia (Koopmans et al., 2001). According to previous studies, fruit and vegetable juice may contribute to improving blood lipid profiles and further prevent cardiovascular diseases, including hyperlipidemia (Chang & Liu, 2009; Zheng et al., 2017). In other words, the results of this study suggested that the juice used in the study might also improve blood lipids. Lipid synthesis‐related gene expression was significantly decreased in the liver tissue of HFD‐OVJ mice, including SREBP‐1, PPARγ, C/EBPα, and FAS, which suggested lower lipid levels and lower fatty acid synthesis in OVJ‐treated mice (Hu et al., 2010; Park et al., 2013). FAS is a key factor in determining the maximum capacity to synthesize fatty acids via the de novo pathway (Clarke, 1993). PPARγ is a transcription factor that is mainly expressed in the adipose tissue and regulates the accumulation of fat in adipocytes by being involved in the production of insulin‐sensitive adipokines, such as adiponectin. It is also involved in the synthesis of TG (Medina‐Gomez et al., 2007; Park et al., 2019). SREBP‐1 is a transcription factor that plays an important role in the synthesis of TG in adipose and liver tissues. The expression of SREBP‐1 is dominant in liver tissues, and it regulates the expression of enzymes related to TG synthesis in hepatocytes (Shimano et al., 1999; Yuan et al., 2009). Hyperlipidemia is accompanied by an increase in free fatty acids in the blood. Increased blood‐free fatty acids are directly toxic to hepatocytes (Feingold & Grunfeld, 1992). The increase in free fatty acids in the liver increases the activity of enzymes that generate free radicals, lipid peroxidation, and the production of inflammatory cytokines such as IL‐6. Oxidative stress is associated with obesity‐related complications, such as hyperlipidemia. Production of 8‐OHdG is caused by oxidative DNA damage, which increases in overweight and obese people. The level of 8‐OHdG in obese women is significantly higher than that in lean women (Devries et al., 2008). MDA is the main product of lipid peroxidation, which is a free radical‐generating process by oxidants (Garcia‐Sanchez et al., 2020). The concentration of MDA was significantly reduced in patients with healthy BMI compared to that in obese individuals (BMI above 40 kg/m2) (Olusi, 2002). In this study, treatment with OVJ significantly decreased the lipid synthesis‐related genes and oxidative stress. It could be the effect of β‐carotene and lycopene. The anti‐obesity effect of β‐carotene is related to the provitamin A effect. This effect is associated with reduced expression of PPARγ in the adipose tissue through the involvement of retinol X receptor signaling (Mounien et al., 2019). Lycopene blocks lipid accumulation in the adipose tissue by decreasing the expression of lipogenesis‐related genes, which is also related to the reduction of inflammatory cytokines (Fenni et al., 2017; Wang et al., 2019). Given this, it is possible to predict vegetables influenced improvement of blood lipids and antioxidant activity in animal experiments.
We examined the positive alterations of the microbiome and metagenomic functions in each section of the continuous colon simulation system. OVJ treatment significantly improved the richness of microorganisms in TC and DC and caused distinct alterations in the composition of the gut microbiome. The relative abundance of the family Erysipelotrichaceae and the genus Holdemania was significantly reduced in the DC of OVJ‐treated group compared to the control group. Previous studies have reported that Erysipelotrichaceae exhibits high abundance in obese individuals (Zhang et al., 2009), and that there is a correlation between Erysipelotrichaceae levels and host cholesterol metabolites (Martínez et al., 2013). In addition, it has been reported that supplementation of flavonol quercetin inhibits the growth of Erysipelotrichaceae (Etxeberria et al., 2015). The genus Holdemania has been reported to correlate with clinical markers of impaired lipid and glucose metabolism (Lippert et al., 2017). Our study shows that supplementation of OVJ can inhibit the growth of taxa associated with obesity and lipid metabolism, such as Erysipelotrichaceae and Holdemania. The three taxa whose relative abundance were increased significantly in the OVJ‐treated group compared to the control group were the family Bacteroidaceae in TC and DC, the genus Butyricimonas in DC and the genus Lachnospira in AC. Bacteroidaceae family is known to be significantly decreased in obesity, and the genus Bacteroides spp. has been reported to indicate a negative correlation between energy intake and obesity (Chávez‐Carbajal et al., 2019). Butyricimonas is known as butyrate‐producing bacteria with anti‐inflammatory effects (Den Besten et al., 2013). Lachnospira also belongs to butyrate‐producing bacteria and is known for the fermentation of polysaccharides of SCFAs (Ferrario et al., 2014). Butyrate has very beneficial effects on energy metabolism, intestinal homeostasis, and regulation of immune response, and has the potential to alleviate obesity and related comorbidities by regulating liver and intestinal lipid metabolism (Coppola et al., 2021). In addition, OVJ consumption increases the relative abundance of Lactobacillus and Akkermansia, known as beneficial bacteria. According to Wiese et al. lycopene‐rich food and flavonol consumption increase the relative abundance of Lactobacillus and change liver metabolism and vascular functions (Wiese et al., 2019). Previous research supports the link between the gut and vascular systems, which also links risk factor‐mediated cardiovascular diseases (Li et al., 2017). According to Chang et al., Akkermansia, belonging to the Akkermansiaceae family, is related to reduced weight gain and maintenance of metabolic homeostasis (Baldwin et al., 2016) and is enriched in healthy people rather than people with metabolic syndrome (Lim et al., 2017). We also observed that the OVJ‐treated group and control juice‐treated group showed a shift in metagenome function, related to the lipid metabolism. KEGG pathway analysis showed a lower number of genes related to lipid metabolism in OVJ‐induced vessels. This suggested that the administration of OVJ improved the microbiome environment and potentially reduced lipid metabolism‐related gene expression. In our results, treatment with OVJ increased the abundance of butyrate‐producing bacteria and therefore OVJ treatment could increase endogenous butyrate production and could be a useful strategy for the prevention of obesity and related metabolic diseases.
In this study, we showed that the microbial community and lipid metabolism were altered in the culture system upon treatment with OVJ, and that blood lipid profiles and antioxidant ability were alleviated in diet‐induced obese mice. These results suggest that OVJ may represent a natural way of alleviating hyperlipidemia. In future studies, it may be necessary to determine whether OVJ influences lipid metabolism‐related genes directly or whether its effects are predominantly mediated by changes in the microbiome. In addition, it may be necessary to investigate the effects of the ingredients on the lipid metabolic gene pathways of microbiome to improve blood lipid profiles and hyperlipidemia in future studies.
## FUNDING INFORMATION
This research received no external funding.
## CONFLICT OF INTEREST
The authors declare no conflict of interest.
## DATA AVAILABILITY STATEMENT
Data are contained within the article or supplementary material.
## References
1. Adamberg S., Sumeri I., Uusna R., Ambalam P.. **Survival and synergistic growth of mixed cultures of**. *Microbial Ecology in Health and Disease* (2014) **25** 23062
2. Baldwin J., Collins B., Wolf P. G., Martinez K.. **Table grape consumption reduces adiposity and markers of hepatic lipogenesis and alters gut microbiota in butter fat‐fed mice**. *The Journal of Nutritional Biochemistry* (2016) **27** 123-121. PMID: 26423887
3. Bray G. A.. **A concise review on the therapeutics of obesity**. *Nutrition* (2000) **16** 953-960. PMID: 11054601
4. Cha K. H., Lee E. H., Yoon H. S., Lee J. H.. **Effects of fermented milk treatment on microbial population and metabolomic outcomes in a three‐stage semi‐continuous culture system**. *Food Chemistry* (2018) **263** 216-224. PMID: 29784310
5. Chang W. H., Liu J. F.. **Effects of kiwifruit consumption on serum lipid profiles and antioxidative status in hyperlipidemic subjects**. *International Journal of Food Sciences and Nutrition* (2009) **60** 709-716. PMID: 19919518
6. Chávez‐Carbajal A., Nirmalkar K., Pérez‐Lizaur A., Hernández‐Quiroz F., Ramírez‐del‐Alto S., García‐Mena J., Hernández‐Guerrero C.. **Gut microbiota and predicted metabolic pathways in a sample of Mexican women affected by obesity and obesity plus metabolic syndrome**. *International Journal of Molecular Sciences* (2019) **20** 438. PMID: 30669548
7. Cinquin C., Le Blay G., Fliss I., Lacroix C.. **New three‐stage in vitro model for infant colonic fermentation with immobilized fecal microbiota**. *FEMS Microbiology Ecology* (2006) **57** 324-336. PMID: 16867149
8. Clarke S. D.. **Regulation of fatty acid synthase gene expression: An approach for reducing fat accumulation**. *Journal of Animal Science* (1993) **71** 1957-1965. PMID: 8349524
9. Coppola S., Avagliano C., Calignano A., Berni Canani R.. **The protective role of butyrate against obesity and obesity‐related diseases**. *Molecules* (2021) **26** 682. PMID: 33525625
10. Costabile A., Walton G. E., Tzortzis G., Vulevic J.. **Effects of orange juice formulation on prebiotic functionality using an in vitro colonic model system**. *PLoS One* (2015) **10**. PMID: 25807417
11. Den Besten G., van Eunen K., Groen A. K., Venema K.. **The role of short‐chain fatty acids in the interplay between diet, gut microbiota, and host energy metabolism**. *Journal of Lipid Research* (2013) **54** 2325-2340. PMID: 23821742
12. Devries M. C., Hamadeh M. J., Glover A. W., Raha S.. **Endurance training without weight loss lowers systemic, but not muscle, oxidative stress with no effect on inflammation in lean and obese women**. *Free Radical Biology & Medicine* (2008) **45** 503-511. PMID: 18502211
13. Doré J., Blottière H.. **The influence of diet on the gut microbiota and its consequences for health**. *Current Opinion in Biotechnology* (2015) **32** 195-199. PMID: 25615931
14. Duwaerts C. C., Amin A. M., Siao K., Her C.. **Specific macronutrients exert unique influences on the adipose‐liver axis to promote hepatic steatosis in mice**. *Cellular and Molecular Gastroenterology and Hepatology* (2017) **4** 223-236. PMID: 28649594
15. Etxeberria U., Arias N., Boqué N., Macarulla M., Portillo M., Martínez J., Milagro F.. **Reshaping faecal gut microbiota composition by the intake of trans‐resveratrol and quercetin in high‐fat sucrose diet‐fed rats**. *The Journal of Nutritional Biochemistry* (2015) **26** 651-660. PMID: 25762527
16. Feingold K. R., Grunfeld C.. **Role of cytokines in inducing hyperlipidemia**. *Diabetes* (1992) **41** 97-101. PMID: 1526345
17. Fenni S., Hammou H., Astier J., Bonnet L.. **Lycopene and tomato powder supplementation similarly inhibit high‐fat diet induced obesity, inflammatory response, and associated metabolic disorders**. *Molecular Nutrition & Food Research* (2017) **61** 1601083
18. Feria‐Gervasio D., Tottey W., Gaci N., Alric M.. **Three‐stage continuous culture system with a self‐generated anaerobia to study the regionalized metabolism of the human gut microbiota**. *Journal of Microbiological Methods* (2014) **96** 111-118. PMID: 24333608
19. Ferrario C., Taverniti V., Milani C., Fiore W., Laureati M., De Noni I., Stuknyte M., Chouaia B., Riso P., Guglielmetti S.. **Modulation of fecal**. *The Journal of Nutrition* (2014) **144** 1787-1796. PMID: 25332478
20. Garcia‐Sanchez A., Gamez‐Nava J. I., Diaz‐de la Cruz E. N., Cardona‐Munoz E. G.. **The effect of visceral abdominal fat volume on oxidative stress and proinflammatory cytokines in subjects with normal weight, overweight and obesity**. *Diabetes, Metabolic Syndrome and Obesity* (2020) **13** 1077-1087
21. González‐Castejón M., Rodriguez‐Casado A.. **Dietary phytochemicals and their potential effects on obesity: A review**. *Pharmacological Research* (2011) **64** 438-455. PMID: 21798349
22. Graf D., Di Cagno R., Fåk F., Flint H. J.. **Contribution of diet to the composition of the human gut microbiota**. *Microbial Ecology in Health and Disease* (2015) **26** 26164. PMID: 25656825
23. Hu Y., Kutscher E., Davies G. E.. **Berberine inhibits SREBP‐1‐related clozapine and risperidone induced adipogenesis in 3T3‐L1 cells**. *Phytotherapy Research* (2010) **24** 1831-1838. PMID: 20564506
24. Jeung W. H., Nam W., Kim H. J.. **Oral Administration of**. *Preventive Nutrition and Food Science* (2019) **24** 136-143. PMID: 31328117
25. Kim S. Y., Yoon S., Kwon S. M., Park K. S.. **Kale juice improves coronary artery disease risk factors in hypercholesterolemic men**. *Biomedical and Environmental Sciences* (2008) **21** 91-97. PMID: 18548846
26. Klop B., Elte J. W., Cabezas M. C.. **Dyslipidemia in obesity: Mechanisms and potential targets**. *Nutrients* (2013) **5** 1218-1240. PMID: 23584084
27. Koopmans S. J., Jong M. C., Que I., Dahlmans V. E. H.. **Hyperlipidaemia is associated with increased insulin‐mediated glucose metabolism, reduced fatty acid metabolism and normal blood pressure in transgenic mice overexpressing human apolipoprotein C1**. *Diabetologia* (2001) **44** 437-443. PMID: 11357474
28. Langille M. G., Zaneveld J., Caporaso J. G., McDonald D.. **Predictive functional profiling of microbial communities using 16S rRNA marker gene sequences**. *Nature Biotechnology* (2013) **31** 814-821
29. Larsen N., Vogensen F. K., Van Den Berg F. W., Nielsen D. S.. **Gut microbiota in human adults with type 2 diabetes differs from non‐diabetic adults**. *PLoS One* (2010) **5**. PMID: 20140211
30. Li R., Yang J., Saffari A., Jacobs J.. **Ambient ultrafine particle ingestion alters gut microbiota in association with increased atherogenic lipid metabolites**. *Scientific Reports* (2017) **7** 42906. PMID: 28211537
31. Lim M. Y., You H. J., Yoon H. S., Kwon B.. **The effect of heritability and host genetics on the gut microbiota and metabolic syndrome**. *Gut* (2017) **66** 1031-1038. PMID: 27053630
32. Lippert K., Kedenko L., Antonielli L., Kedenko I., Gemeier C., Leitner M., Kautzky‐Willer A., Paulweber B., Hackl E.. **Gut microbiota dysbiosis associated with glucose metabolism disorders and the metabolic syndrome in older adults**. *Beneficial Microbes* (2017) **8** 545-556. PMID: 28701081
33. Liu R. H.. **Health benefits of fruit and vegetables are from additive and synergistic combinations of phytochemicals**. *The American Journal of Clinical Nutrition* (2003) **78** 517S-520S. PMID: 12936943
34. Maccaferri S., Klinder A., Brigidi P., Cavina P.. **Potential probiotic**. *Applied and Environmental Microbiology* (2012) **78** 956-964. PMID: 22156412
35. Mangal P., Khare P., Jagtap S., Bishnoi M.. **Screening of six ayurvedic medicinal plants for anti‐obesity potential: An investigation on bioactive constituents from**. *Journal of Ethnopharmacology* (2017) **197** 138-146. PMID: 27469197
36. Martínez I., Perdicaro D. J., Brown A. W., Hammons S., Carden T. J., Carr T. P., Eskridge K. M., Walter J.. **Diet‐induced alterations of host cholesterol metabolism are likely to affect the gut microbiota composition in hamsters**. *Applied and Environmental Microbiology* (2013) **79** 516-524. PMID: 23124234
37. Medina‐Gomez G., Gray S., Vidal‐Puig A.. **Adipogenesis and lipotoxicity: Role of peroxisome proliferator‐activated receptor γ (PPARγ) and PPARγcoactivator‐1 (PGC1)**. *Public Health Nutrition* (2007) **10** 1132-1137. PMID: 17903321
38. Mounien L., Tourniaire F., Landrier J. F.. **Anti‐obesity effect of carotenoids: Direct impact on adipose tissue and adipose tissue‐driven indirect effects**. *Nutrients* (2019) **11** 1562. PMID: 31373317
39. Nelson R. H.. **Hyperlipidemia as a risk factor for cardiovascular disease**. *Primary Care* (2013) **40** 195-211. PMID: 23402469
40. Neyrinck A. M., Van Hee V. F., Bindels L. B., De Backer F.. **Polyphenol‐rich extract of pomegranate peel alleviates tissue inflammation and hypercholesterolaemia in high‐fat diet‐induced obese mice: Potential implication of the gut microbiota**. *The British Journal of Nutrition* (2013) **109** 802-809. PMID: 22676910
41. Nuutila A. M., Puupponen‐Pimiä R., Aarni M., Oksman‐Caldentey K. M.. **Comparison of antioxidant activities of onion and garlic extracts by inhibition of lipid peroxidation and radical scavenging activity**. *Food Chemistry* (2003) **81** 485-493
42. Olusi S. O.. **Obesity is an independent risk factor for plasma lipid peroxidation and depletion of erythrocyte cytoprotectic enzymes in humans**. *International Journal of Obesity and Related Metabolic Disorders* (2002) **26** 1159-1164. PMID: 12187391
43. Park D. Y., Ahn Y. T., Park S. H., Huh C. S.. **Supplementation of**. *PLoS One* (2013) **8**. PMID: 23555678
44. Park M., Yoo J. H., Lee Y. S., Lee H. J.. *Nutrients* (2019) **11** 494. PMID: 30813654
45. Penders J., Stobberingh E. E., Brandt P. V. D., Thijs C.. **The role of the intestinal microbiota in the development of atopic disorders**. *Allergy* (2007) **62** 1223-1236. PMID: 17711557
46. Segata N., Izard J., Waldron L., Gevers D.. **Metagenomic biomarker discovery and explanation**. *Genome Biology* (2011) **12** 1-18
47. Shimano H., Yahagi N., Amemiya‐Kudo M., Hasty A. H.. **Sterol regulatory element‐binding protein‐1 as a key transcription factor for nutritional induction of lipogenic enzyme genes**. *The Journal of Biological Chemistry* (1999) **274** 35832-35839. PMID: 10585467
48. Singh D., Singh R., Singh P., Gupta R. S.. **Effects of embelin on lipid peroxidation and free radical scavenging activity against liver damage in rats**. *Basic & Clinical Pharmacology & Toxicology* (2009) **105** 243-248. PMID: 19496777
49. Sousa T., Paterson R., Moore V., Carlsson A., Abrahamsson B., Basit A. W.. **The gastrointestinal microbiota as a site for the biotransformation of drugs**. *International Journal of Pharmaceutics* (2008) **363** 1-25. PMID: 18682282
50. Wang J., Suo Y., Zhang J., Zou Q.. **Lycopene supplementation attenuates western diet‐induced body weight gain through increasing the expressions of thermogenic/mitochondrial functional genes and improving insulin resistance in the adipose tissue of obese mice**. *The Journal of Nutritional Biochemistry* (2019) **69** 63-72. PMID: 31060024
51. Wiese M., Bashmakov Y., Chalyk N., Nielsen D. S.. **Prebiotic effect of lycopene and dark chocolate on gut microbiome with systemic changes in liver metabolism, skeletal muscles and skin in moderately obese persons**. *BioMed Research International* (2019) **2019** 4625279. PMID: 31317029
52. Williams C. F., Walton G. E., Jiang L., Plummer S., Garaiova I., Gibson G. R.. **Comparative analysis of intestinal tract models**. *Annual Review of Food Science and Technology* (2015) **6** 329-350
53. Williams D. J., Edwards D., Hamernig I., Jian L.. **Vegetables containing phytochemicals with potential anti‐obesity properties: A review**. *Food Research International* (2013) **52** 323-333
54. Yang R. L., Shi Y. H., Hao G., Li W.. **Increasing oxidative stress with progressive hyperlipidemia in human: Relation between malondialdehyde and atherogenic index**. *Journal of Clinical Biochemistry and Nutrition* (2008) **43** 154-158. PMID: 19015749
55. Yuan H., Shyy J. Y., Martins‐Green M.. **Second‐hand smoke stimulates lipid accumulation in the liver by modulating AMPK and SREBP‐1**. *Journal of Hepatology* (2009) **51** 535-547. PMID: 19556020
56. Zhang H., DiBaise J. K., Zuccolo A., Kudrna D., Braidotti M., Yu Y., Parameswaran P., Crowell M. D., Wing R., Rittmann B. E.. **Human gut microbiota in obesity and after gastric bypass**. *Proceedings of the National Academy of Sciences of the United States of America* (2009) **106** 2365-2370. PMID: 19164560
57. Zhang Y., Guo K., LeBlanc R. E., Loh D.. **Increasing dietary leucine intake reduces diet‐induced obesity and improves glucose and cholesterol metabolism in mice via multimechanisms**. *Diabetes* (2007) **56** 1647-1654. PMID: 17360978
58. Zhao L.. **The gut microbiota and obesity: From correlation to causality**. *Nature Reviews. Microbiology* (2013) **11** 639-647. PMID: 23912213
59. Zheng J., Zhou Y., Li S., Zhang P.. **Effects and mechanisms of fruit and vegetable juices on cardiovascular diseases**. *International Journal of Molecular Sciences* (2017) **18** 555. PMID: 28273863
|
---
title: Clinical Study of Metabolic Parameters, Leptin and the SGLT2 Inhibitor Empagliflozin
among Patients with Obesity and Type 2 Diabetes
authors:
- Zsolt Szekeres
- Barbara Sandor
- Zita Bognar
- Fadi H. J. Ramadan
- Anita Palfi
- Beata Bodis
- Kalman Toth
- Eszter Szabados
journal: International Journal of Molecular Sciences
year: 2023
pmcid: PMC10002958
doi: 10.3390/ijms24054405
license: CC BY 4.0
---
# Clinical Study of Metabolic Parameters, Leptin and the SGLT2 Inhibitor Empagliflozin among Patients with Obesity and Type 2 Diabetes
## Abstract
Obesity is a major public health problem worldwide, and it is associated with many diseases and abnormalities, most importantly, type 2 diabetes. The visceral adipose tissue produces an immense variety of adipokines. Leptin is the first identified adipokine which plays a crucial role in the regulation of food intake and metabolism. Sodium glucose co-transport 2 inhibitors are potent antihyperglycemic drugs with various beneficial systemic effects. We aimed to investigate the metabolic state and leptin level among patients with obesity and type 2 diabetes mellitus, and the effect of empagliflozin upon these parameters. We recruited 102 patients into our clinical study, then we performed anthropometric, laboratory, and immunoassay tests. Body mass index, body fat, visceral fat, urea nitrogen, creatinine, and leptin levels were significantly lower in the empagliflozin treated group when compared to obese and diabetic patients receiving conventional antidiabetic treatments. Interestingly, leptin was increased not only among obese patients but in type 2 diabetic patients as well. Body mass index, body fat, and visceral fat percentages were lower, and renal function was preserved in patients receiving empagliflozin treatment. In addition to the known beneficial effects of empagliflozin regarding the cardio-metabolic and renal systems, it may also influence leptin resistance.
## 1. Introduction
According to the latest data, nearly 2 billion adults ($39\%$ of the world’s adult population) were estimated to be obese or overweight. If current trends continue, it is expected that 1 billion adults, nearly $20\%$ of the world’s population, will clinically be declared obese by 2025 [1]. Obesity is associated with many diseases and abnormalities, such as type 2 diabetes [2], dyslipidemia [3], cardiovascular diseases [4], hypertension [5], certain types of cancer [6,7], pneumological [6], nephrological [8], skeletal muscle [9], rheumatologic [10], dermatologic [11], and neuropsychologic [11] complications, and is it associated with premature mortality. Obesity, especially the dysfunctional visceral adipose tissue (VAT), is the main driver of many metabolic abnormalities including insulin resistance, hyperinsulinemia, glucose intolerance, atherogenic dyslipidemia (high triglyceride and apolipoprotein B levels, increased proportion of small, dense LDL [low-density lipoprotein] particles, low HDL [high-density lipoprotein] cholesterol levels, and small HDL particles), and is associated with a low-grade inflammation [6].
Leptin was the first identified adipokine in the 1990s known to suppress food intake through the suppression of appetite and mediate energy homeostasis including glucose and lipid metabolism [12]. The serum level of leptin is elevated paradoxically in obesity [13], and this high level of leptin may induce leptin resistance and result in altered glucose metabolism and insulin resistance [14]. Hyperleptinemia has also been associated with increased inflammation, oxidative stress, endothelial dysfunction, atherogenesis, and thrombosis [15]. Based on these effects, leptin is attributed to a significant role in the development of cardiovascular diseases. Additionally, patients with type 2 diabetes mellitus scored a higher percentage of hypertension, obesity, metabolic syndrome, and endothelial dysfunction if they had elevated leptin levels [16].
The link between obesity and type 2 diabetes mellitus [T2DM] has long been recognized and explains the high prevalence of type 2 diabetes mellitus. Type 2 diabetes mellitus is associated with many vascular complications. Microvascular complications include diabetic kidney disease, retinopathy, and neuropathy, whereas the macrovascular complications include coronary artery, cerebrovascular, and peripheral vascular diseases. The main goals of treatment in patients with T2DM are to achieve adequate glycemic control, reduce body weight and prevent vascular damage, and target organ damage [17]. Novel antidiabetic therapies such as sodium glucose co-transporter 2 (SGLT2) inhibitors provide a new approach to preventing or ameliorating the complications that insulin resistance and hyperglycemia create [18]. SGLT2 inhibitors are potent antihyperglycemic drugs, which inhibit glucose reabsorption in the proximal tubules of the kidney inducing glycosuria and improving blood glucose levels, and may reduce body weight through calorie loss. Numerous studies have shown they are associated with reduced cardiovascular morbidity and mortality, including vascular diseases and heart failure [19]. Furthermore, SGLT2 inhibitors have also demonstrated positive reno-metabolic effects [20]. In a cardiovascular outcome trial, the SGLT2 inhibitor empagliflozin proved superior to conventional antidiabetic therapy in reducing the rate of MACE, mortality, and hospitalization due to heart failure [21]. SGLT2 inhibitor therapy has been associated with a decrease in serum triglycerides, an increase in HDL cholesterol, and also a small increase in LDL cholesterol level was observed [20]. The presence of metabolic disturbances in obese patients results in oxidative stress [22]. Since obesity and insulin resistance is a major component of metabolic syndrome, it is strongly associated with oxidative stress [23]. The oxidative modification of lipoproteins can result in more atherogenic compounds, which may have a key role in the development of cardiovascular dysfunction in patients with diabetes mellitus [24,25].
The aim of our study was to investigate certain laboratory parameters such as lipids, inflammatory markers, blood glucose level, glycated hemoglobin [HbA1c] level, kidney function, leptin level, as well as body mass index [BMI], body fat and visceral fat percentage among patients afflicted with obesity and diabetes. We also investigated a subgroup of patients receiving empagliflozin treatment.
## 2.1. Body Mass Index, Body Fat, and Visceral Fat Were Significantly Lower in the Empagliflozin Treated Group
BMI was significantly lower in the control group (C) when compared to the obese (O) ($p \leq 0.001$), to the obese and diabetic (OD) ($p \leq 0.001$), and to the empagliflozin treated (ODE) group ($p \leq 0.001$). It was also significantly lower in the diabetic (D) group when compared to the obese (O) ($p \leq 0.001$), to the obese and diabetic (OD) ($p \leq 0.001$), and to the empagliflozin-treated group ($p \leq 0.001$). BMI was significantly lower in the empagliflozin-treated group (ODE) when compared to the obese and diabetic (OD) group ($p \leq 0.001$). There was no significant difference between the other groups.
Body fat was significantly lower in the control group (C), when compared to the obese (O) ($p \leq 0.001$), and to the obese and diabetic (OD) ($p \leq 0.001$) groups. It was also significantly lower in the diabetic (D) group when compared to the obese (O) ($$p \leq 0.001$$) and to the obese and diabetic (OD) ($$p \leq 0.001$$) groups. Body fat was significantly lower in the empagliflozin-treated group (ODE) when compared to the obese and diabetic (OD) group ($$p \leq 0.002$$). There were no significant differences between the other groups.
Visceral fat was significantly lower in the control group (C) when compared to the obese (O) ($p \leq 0.001$), to the obese and diabetic (OD) ($p \leq 0.001$), and to the empagliflozin-treated group (ODE) ($p \leq 0.001$). It was also significantly lower in the diabetic (D) group when compared to the obese (O) ($p \leq 0.001$), to the obese and diabetic (OD) ($p \leq 0.001$), and to the empagliflozin-treated group ($p \leq 0.001$). Visceral fat was significantly lower in the empagliflozin-treated group (ODE) when compared to the obese and diabetic (OD) group ($p \leq 0.014$). There were no significant differences between the other groups (Table 1).
## 2.2. Hemoglobin Levels Were Significantly Higher among the Empagliflozin Treated Patients
Hemoglobin levels were significantly higher in the empagliflozin-treated group (ODE) when compared to the diabetic (D), and obese and diabetic (OD) groups ($$p \leq 0.004$$ and $p \leq 0.001$, respectively). There was no significant difference between the diabetic (D) and the obese and diabetic (OD) group ($$p \leq 0.850$$). The obese group (O) had a significantly higher hemoglobin when compared to the obese and diabetic group (OD) and a significantly lower level when compared to the empagliflozin-treated obese group (ODE) ($$p \leq 0.033$$ and $$p \leq 0.007$$ respectively) (Table 2).
## 2.3. Renal Parameters Were Significantly Higher in Diabetic Patients, Yet Were Reduced in the Empagliflozin Treated Group
Urea nitrogen level increases significantly with the appearance of diabetes in obesity (O vs. OD) ($$p \leq 0.002$$). In the empagliflozin-treated group (ODE), the urea nitrogen level was significantly lower when compared to the obese and diabetic (OD) group ($$p \leq 0.008$$) (Table 2).
Creatinine significantly increases with the appearance of diabetes in the obese groups (O vs. OD) ($$p \leq 0.011$$). In the empagliflozin-treated group (ODE), the creatinine level was significantly lower when compared to the obese and diabetic (OD) group ($$p \leq 0.012$$) (Table 2).
## 2.4. Blood Glucose and HbA1c Levels Were Significantly Higher in Diabetic Patients, Yet There Was No Significant Difference between the Different Diabetic Groups
Blood glucose and HbA1c levels were significantly lower in the control group (C) when compared with the diabetic (D), the obese and diabetic (OD), and the empagliflozin-treated obese and diabetic groups ($$p \leq 0.029$$, $$p \leq 0.005$$, and $p \leq 0.001$, respectively). Blood glucose and HbA1c levels were significantly lower in the obese group when compared with the diabetic (D), the obese and diabetic (OD), and the empagliflozin-treated obese and diabetic groups ($$p \leq 0.015$$, $$p \leq 0.008$$, and $p \leq 0.001$, respectively). There were no significant differences between the other groups regarding blood glucose and HbA1c levels.
## 2.5. Leptin Levels Were Significantly Higher in Obese Patients, Yet Were Reduced in the Empagliflozin-Treated Group
Leptin levels were significantly higher with the appearance of obesity (O) ($$p \leq 0.003$$) even if obesity was present with diabetes (OD) ($p \leq 0.001$) when compared to the control (C) group. It was also significantly higher in diabetic patients (D) when compared with the control group (C) ($$p \leq 0.029$$). Obese and diabetic patients (OD) had a significantly higher level of leptin when compared to diabetic yet not obese (D) patients ($$p \leq 0.001$$). In the empagliflozin-treated group (ODE), the leptin level was significantly lower when compared to the obese and diabetic (OD) group ($$p \leq 0.048$$) (Table 2).
## 2.6. There Were No Significant Differences between the Other Measured Parameters
There was no significant difference in body muscle percentage, white blood cell count, red blood cell count, platelet count, fibrinogen levels, uric acid, triglyceride, sodium and potassium levels, and thyroid-stimulating hormone levels among the groups.
There was no significant difference in the cholesterol levels among the groups. It bears mentioning, cholesterol levels were strongly affected by the antihyperlipidemic agents.
The continuous variables did not differ from the normal distribution. Data are shown as means ± standard deviation.
## 3. Discussion
In our clinical study, we examined metabolic and inflammatory parameters, kidney function, and leptin levels among patients afflicted with hypertension, obesity, type 2 diabetes, and cardiovascular diseases. The aim of our study was to detect the severity of the metabolic state among these patients and to examine a subgroup of patients treated with empagliflozin. In our study, we found empagliflozin-treated obese, diabetic patients had significantly lower BMI, body fat, and visceral fat values as well as lower serum creatinine and leptin levels when compared to patients with obesity and type 2 diabetes treated with usual antidiabetics (such as biguanides and sulfonylureas). Leptin levels were already higher among patients with type 2 diabetes even with normal BMI, and were significantly higher in obese non-diabetic patients and were the highest in obese patients with type 2 diabetes. Furthermore, we discovered that increased visceral fat and leptin levels predicted diabetes similarly to HbA1c.
Excess visceral adiposity is a major risk factor for metabolic and cardiovascular disorders. It plays a crucial role in the development of a diabetogenic and atherogenic metabolic profile inducing insulin resistance and increased cardiometabolic risk [26]. In our study, BMI, body fat, and visceral fat percentage were the highest among patients with obesity and type 2 diabetes (Group OD). In the empagliflozin-treated obese, diabetic patients (Group ODE), BMI, body fat, and visceral fat were significantly lower when compared with obese and diabetic patients (OD) treated with usual antidiabetics (Table 1). In an animal study, empagliflozin suppressed weight gain by shifting energy metabolism towards fat utilization, elevated adenosine monophosphate-activated [AMP] protein kinase, and acetyl coenzyme A [acetyl-CoA] carboxylase phosphorylation in skeletal muscle. Furthermore, empagliflozin increased energy expenditure, heat production and browning, and attenuated obesity-induced inflammation and insulin resistance by polarizing M2 macrophages in white adipose tissue [WAT] and liver [27]. Thus, empagliflozin suppressed weight gain by enhancing fat utilization and browning and attenuated obesity-induced inflammation and insulin resistance.
White adipose tissue is an endocrine organ capable of producing and releasing numerous bioactive substances known as adipokines or adipocytokines. Dysregulated production of adipocytokines is involved in the development of obesity-related diseases. Leptin is one of the most examined adipokines. An increased leptin level is associated with insulin resistance and T2DM development [28]. In T2DM, a link has also been reported between high leptin concentrations and increased cardiovascular [CV risk], including the presence of microvascular complications and cardiac autonomic dysfunction [29]. Furthermore, obesity, hypertension, metabolic syndrome, and endothelial dysfunction are more frequent in T2DM patients with increased leptin levels [30]. In chronic heart disease (CHD) patients, elevated leptin levels were significantly associated with an increased risk of cardiac death, acute coronary syndrome, non-fatal MI, stroke, and hospitalization for congestive heart failure [31,32]. Similarly, higher leptin levels were significantly related to the number of stenotic coronary arteries and arterial stiffness in CHD patients [33]. The presence, severity, extent, and lesion complexity of coronary atherosclerosis have been associated with higher leptin levels in CHD patients [34]. Leptin may also affect cardiac remodeling, metabolism, and contractile function [35]. Other effects of leptin include activation of inflammatory responses, oxidative stress, thrombosis, and atherosclerosis, thereby resulting in endothelial dysfunction and atherosclerotic plaque [16].
In our study, the leptin level was already higher among patients with type 2 diabetes even with normal BMI (Group D), was significantly higher in obese non-diabetic patients (Group O), and was the highest in obese patients with type 2 diabetes (Group OD) when compared to the control group.
A link between increased plasma leptin concentrations and chronic kidney disease (CKD) has been reported, which is possibly due to reduced renal clearance [36]. Leptin concentrations gradually increased with the severity of CKD [37]. In CKD patients, plasma leptin levels have been inversely associated with glomerular filtration rate and directly associated with urinary albumin levels as well as age and obesity markers (BMI and waist circumference) [38]. Overall, hyperleptinemia has been linked to the presence, severity, and progression of CKD. In our study, creatinine levels were significantly higher with the appearance of diabetes and were the highest among obese patients with type 2 diabetes. Among the empagliflozin-treated obese and type 2 diabetic patients, the creatinine level was significantly lower eliciting improved renal function (Table 2).
We possess a vast amount of knowledge regarding the cardiovascular and renal effects of SGLT2 inhibitors [20,39,40,41,42]. In addition to their direct effect on glucose homeostasis, they have many other underlying mechanisms from which not all are fully understood. For instance, SGLT2 inhibitors may also act upon visceral adipose tissue. Dapagliflozin therapy was associated with a decreased circulating leptin level and an increased circulating adiponectin level among patients with type 2 diabetes, which, may contribute to the beneficial effects of SGLT2 inhibitors on metabolic homeostasis, such as improved insulin resistance and reduced cardiovascular risk [43,44,45]. Furthermore, dapagliflozin displayed significantly lower arterial stiffness in diabetic mice treated with dapagliflozin when compared to untreated diabetic mice [46]. The effects of empagliflozin on adipocytokines were examined in an animal study conducted on obese rats. Empagliflozin dose-dependently reduced body weight, body fat, adiponectin, and leptin following the 28-day treatment [39]. In our study, the leptin level was significantly lower in the empagliflozin-treated obese and type 2 diabetic patients (ODE) when compared to the obese, diabetic patients (OD) treated with other antidiabetics (Table 2). To the best of our knowledge, this is the first time the beneficial effect of empagliflozin on the leptin level has been demonstrated in a clinical setting.
HbA1c is a well-known screening and diagnostic tool in detecting diabetes. A score higher than 5.7 % value implies prediabetes, and consequently, higher than 6.5 % confirms diabetes. Our receiver operating characteristic [ROC] analysis has proven the recommended 5.7 % cut-off value effectively predicted altered glucose homeostasis with very high sensitivity and acceptable specificity. In the same analysis, leptin was found to be similar in the prediction of diabetes. This is congruent with previous observations stating elevated leptin levels are associated with insulin resistance and T2DM development [28].
The second ROC analysis with the composite endpoint diabetes and obesity showed, in addition to HgA1c, leptin, and visceral fat may have a role in the diagnosis of diabetes among obese adults. These findings emphasize patients with increased visceral fat, which is easily measured using a smart weight scale, are prime candidates to be screened for insulin resistance or diabetes with HbA1c and fasting glucose value.
Hemoglobin values were the highest in the empagliflozin-treated group, which, may imply a slight hemoconcentration, and may be related to the osmotic diuretic effect of empagliflozin treatment. It is worthwhile to draw the attention of patients to the need for adequate fluid intake during SGLT2 inhibitor treatment. Unexpectedly, HbA1c levels were the highest in the empagliflozin-treated group. Presumably, this is due to the fact that, in Hungary, SGLT2 inhibitor treatment can only be prescribed to patients with an HbA1c level above $7\%$. This also means this group is a more severe patient group in terms of diabetes, thus, the results obtained prove even more crucial.
There was no significant difference in C-reactive protein (CRP) levels among the examined groups; however, some differences were detected. The CRP level was the lowest in the non-obese, non-diabetic group (C). Although many factors can influence the CRP level, it may be important that it was higher among obese and diabetic patients, which may indicate a low level of inflammation and corresponds to previous observations [19]. Among patients receiving empagliflozin treatment (ODE), the CRP level was lower when compared to the obese and diabetic group (OD), which may reflect lower inflammation status, likely due to the empagliflozin treatment. It has been previously reported, that empagliflozin reduced renal inflammation and oxidative stress in spontaneously hypertensive rats [47] In the EMPA-CARD trial patients with type2 diabetes and coronary artery disease treated with empagliflozin had lower levels of interleukin 6, interleukin 1β and CRP levels compared to a placebo. There were elevations in superoxidase dismutase (SOD) activity, glutathione (GSHr), and total antioxidant capacity (TAC) with empagliflozin [48].
Notably, there was no significant difference in LDL cholesterol levels. This may be due to the fact in which LDL cholesterol levels were greatly influenced by antihyperlipidemic drugs. Previous literature data indicated a moderate increase in LDL level can be detected with SGLT2 inhibitor treatment. In our study, we did not observe higher LDL values in the empagliflozin-treated group when compared to the other groups. Additionally, in our study, CV disease incidence was provided primarily to describe the patient population. Although it was lower in the empagliflozin-treated group, it was not intended to examine this correlation.
The main strength of our study is that, to the best of our knowledge, this is the first examination that has demonstrated that empagliflozin treatment has a beneficial effect on serum leptin levels under clinical conditions. However, our study was conducted on a relatively small number of patients, so further studies on a larger patient population are needed to confirm our results.
## 4.1. Ethics
The study protocol was approved by the Regional Ethics Committee of Pecs (No. 7622—PTE 2019) and was conducted in accordance with the ethical principles stated in the Declaration of Helsinki. Written informed consent was obtained from all patients.
## 4.2. Patients
102 patients (35 female, 67 male) were enrolled in our study. Patients were recruited from different internal medicine and outpatient departments by various physicians. They voluntarily agreed to participate in our study in which they signed an informed consent letter. Subgroup analysis was performed based on different metabolic states. Patients who did not have type 2 diabetes and were not obese were assigned to group C (20 patients), declared as the control group. Obese patients without diabetes were assigned to group O (obese), (20 patients). Non-obese patients with type 2 diabetes were selected into group D (diabetic), (19 patients). Obese and diabetic patients were assigned into group OD (obese and diabetic), (19 patients). Obese, diabetic patients receiving empagliflozin therapy for at least 3 months were assigned to group ODE (20 patients). Patients were considered obese if their BMI was 30.0 kg/m2 or higher. Antihypertensive, antidiabetic, and antihyperlipidemic therapies were recorded from the patient’s history as well as their comorbidities, such as diabetes mellitus, hypertension, and cardiovascular diseases. Exclusion criteria include the following: previous SGLT2 inhibitor therapy for groups C, O, D, OD; active cancer disease; and refusing to sign the consent form. Four patients were excluded from the study for different reasons (low compliance, severe epileptic seizure, withdrawal of their consent, and urgent psychiatric ward admission).
Patients’ general characteristics were as follows. The mean age for different groups was: 65.95 for group C, 66.40 for group O, 74.58 for group D, 70.90 for group OD, and 65.20 for group ODE. The distribution of sex (male to female percentage) in the groups was as follows: 75–$25\%$ for group C, 50–$50\%$ for group O, 52.60–$47.40\%$ for group D, 68.40–$31.60\%$ for group OD, and 75–$25\%$ for group ODE. Mean BMI values for different groups were as follows: 26.01 kg/m2 for group C, 34.75 kg/m2 for group O, 26.50 kg/m2 for group D, 35.78 kg/m2 for group OD, and 31.61 kg/m2 for group ODE. All patients had high blood pressure in their medical history. All patients in the diabetic groups (D, OD, ODE) had identified type 2 diabetes mellitus in their medical history, whereas none were reported in the remaining groups (C, O). The percentage of patients with identified cardiovascular disease was $70.40\%$ in group C, $69.40\%$ in group O, $78.60\%$ in group D, $64.30\%$ in group OD, and $73.48\%$ in group ODE. All patients received antihypertensive therapy. All diabetic patients (D, OD, ODE) received antidiabetic therapy, whereas none were administered in the non-diabetic groups (C, O). Empagliflozin was administered only in the ODE group. No other SGLT2 inhibitors were used in our study. The percentage of patients with antihyperlipidemic therapy was as follows: $70\%$ for group C, $75\%$ for group O, $89.47\%$ for group D, $84.20\%$ for group OD, and $80\%$ for group ODE.
## 4.3. Study Design
102 patients were recruited into this clinical study. We assessed their body composition, followed by pre-prandial venous blood collected using a peripheral venous catheter in the cubital vein. The preparation and laboratory procedures were in full accordance with the recommendations of the laboratory kits. Laboratory tests were performed at the Department of Laboratory Medicine, University of Pecs, Pecs, Hungary. The leptin levels were determined using the immunoassay method (Human Leptin ELISA, Biovendor, Czech Republic) at the Department of Biochemistry and Medical Chemistry, University of Pecs, Pecs, Hungary.
## 4.4. Anthropometric Measurements
The patients’ body composition was assessed using an Omron HBF-511 body composition scale (Omron HealthCare Co., Ltd., Kyoto, Japan). We measured weight, BMI, body fat percentage, and visceral fat percentage. Height was measured using a measuring tape.
## 4.5. Laboratory Tests
Pre-prandial laboratory tests were performed on every patient. These include complete blood count (red and white blood cell count, platelet count, hemoglobin level, hematocrit), fibrinogen, basic metabolic panel (pre-prandial glucose, sodium, potassium, calcium, blood urea nitrogen, and creatinine levels), lipid panel (total cholesterol, HDL cholesterol, LDL cholesterol, and triglyceride levels), liver panel (aspartate transaminase (AST), alanine transaminase (ALT), gamma-glutamyl transferase (GGT) levels), hemoglobin A1C level, and the thyroid stimulating hormone level.
## 4.6. Immunoassay Tests
Plasma leptin 1 levels were measured in duplicate using enzyme-linked immunosorbent assay (ELISA) kits (Cat. No. RD191001100). The blood samples were centrifuged at 2500× g for 10 min. The recovered plasma was stored at −70 °C in aliquots until assayed. The tests were performed in full accordance with the recommendations of the manufacturer, with a detection limit of 0.08 and 0.2 ng/mL, respectively. ( BioVendor GmbH., Brno, Czech Republic).
## 4.7. Statistical Analysis
IBM SPSS statistics, version 28.0.0. ( SPSS, Chicago, IL, USA, 2022); software for statistical; was used to conduct descriptive analyses and to describe the sample. Data are shown as means ± standard deviation.
Differences in the continuous variables were evaluated using a one-way repeated ANOVA statistical test (Tamhane post-hoc test) following the administering of the Kolmogorov–Smirnov test to check the normality of the data distribution. The continuous variables did not differ from the normal distribution.
In the case of categorical variables, data are shown as percentages and incidence (absolute number compared to total number). Differences were evaluated by using chi-square test analyses.
Multivariate linear regression and stepwise analyses of the data were performed regarding the leptin values for HbA1c, LDL, triglyceride, creatinine, hemoglobin, and visceral fat.
Multiple regression analysis with various models including leptin, HbA1c, and visceral fat considering the principle of multicollinearity was performed to reveal which factors predict the occurrence of diabetes and obesity.
The diagnostic power of variables was assessed using the area under the curve (AUC) of the receiver operating characteristic (ROC) curve. The predicted probabilities were calculated from the variables produced by binary logistic regression analysis, in which p ≤ 0.05 was considered statistically significant.
## 5. Conclusions
BMI, body fat, and visceral fat values as well as serum creatinine and leptin levels were improved with empagliflozin treatment. High leptin levels and leptin resistance in obesity are associated with insulin resistance, type 2 diabetes, increased risk of CV diseases, low-grade inflammation, and thrombosis. The markedly decreased circulating leptin levels observed in the empagliflozin-treated group may contribute to the known beneficial cardiovascular effects of empagliflozin treatment.
## References
1. Loos R.J.F., Yeo G.S.H.. **The genetics of obesity: From discovery to biology**. *Nat. Rev. Genet.* (2022) **23** 120-133. DOI: 10.1038/s41576-021-00414-z
2. Klein S., Gastaldelli A., Yki-Järvinen H., Scherer P.E.. **Why does obesity cause diabetes?**. *Cell Metab.* (2022) **34** 11-20. DOI: 10.1016/j.cmet.2021.12.012
3. Vekic J., Zeljkovic A., Stefanovic A., Jelic-Ivanovic Z., Spasojevic-Kalimanovska V.. **Obesity and dyslipidemia**. *Metabolism* (2019) **92** 71-81. DOI: 10.1016/j.metabol.2018.11.005
4. Powell-Wiley T.M., Poirier P., Burke L.E., Després J.-P., Gordon-Larsen P., Lavie C.J., Lear S.A., Ndumele C.E., Neeland I.J., Sanders P.. **Obesity and Cardiovascular Disease: A Scientific Statement From the American Heart Association**. *Circulation* (2021) **143** e984-e1010. DOI: 10.1161/CIR.0000000000000973
5. Mendoza M.F., Kachur S.M., Lavie C.J.. **Hypertension in obesity**. *Curr. Opin. Cardiol.* (2020) **35** 389-396. DOI: 10.1097/HCO.0000000000000749
6. Piché M.-E., Tchernof A., Després J.-P.. **Obesity Phenotypes, Diabetes, and Cardiovascular Diseases**. *Circ. Res.* (2020) **126** 1477-1500. DOI: 10.1161/CIRCRESAHA.120.316101
7. Ackerman S.E., Blackburn O.A., Marchildon F., Cohen P.. **Insights into the Link between Obesity and Cancer**. *Curr. Obes. Rep.* (2017) **6** 195-203. DOI: 10.1007/s13679-017-0263-x
8. Kovesdy C.P., Furth S.L., Zoccali C.. **Obesity and kidney disease: Hidden consequences of the epidemic**. *J. Nephrol.* (2017) **30** 1-10. DOI: 10.1007/s40620-017-0377-y
9. Urano T., Inoue S.. **Recent genetic discoveries in osteoporosis, sarcopenia and obesity**. *Endocr. J.* (2015) **62** 475-484. DOI: 10.1507/endocrj.EJ15-0154
10. George M.D., Baker J.F.. **The Obesity Epidemic and Consequences for Rheumatoid Arthritis Care**. *Curr. Rheumatol. Rep.* (2016) **18** 6. DOI: 10.1007/s11926-015-0550-z
11. Lause M., Kamboj A., Faith E.F.. **Dermatologic manifestations of endocrine disorders**. *Transl. Pediatr.* (2017) **6** 300-312. DOI: 10.21037/tp.2017.09.08
12. Farooqi I.S., O’Rahilly S.. **Human disorders of leptin action**. *J. Endocrinol.* (2014) **223** T63-T70. DOI: 10.1530/JOE-14-0480
13. D’Souza A.M., Neumann U.H., Glavas M.M., Kieffer T.J.. **The glucoregulatory actions of leptin**. *Mol. Metab.* (2017) **6** 1052-1065. DOI: 10.1016/j.molmet.2017.04.011
14. **Intensive Blood-Glucose Control with Sulfonylureas or Insulin Compared with Conventional Treatment and Risk of Complications in Patients with Type 2 Diabetes**. *Endocrinologist* (1999) **9** 149. DOI: 10.1097/00019616-199903000-00016
15. Ouchi K.W.N., Parker J.L., Lugus J.J., Walsh K.. **Adipokines in inflammation and metabolic disease**. *Nat. Rev. Immunol.* (2011) **11** 85-97. DOI: 10.1038/nri2921
16. Katsiki N., Mikhailidis D.P., Banach M.. **Leptin, cardiovascular diseases and type 2 diabetes mellitus review-article**. *Acta Pharmacol. Sin.* (2018) **39** 1176-1188. DOI: 10.1038/aps.2018.40
17. Artasensi A., Pedretti A., Vistoli G., Fumagalli L.. **Type 2 Diabetes Mellitus: A Review of Multi-Target Drugs**. *Molecules* (2020) **25**. DOI: 10.3390/molecules25081987
18. Vesa C.M., Popa L., Popa A.R., Rus M., Zaha A.A., Bungau S., Tit D.M., Corb Aron R.A., Zaha D.C.. **Current Data Regarding the Relationship between Type 2 Diabetes Mellitus and Cardiovascular Risk Factors**. *Diagnostics* (2020) **10**. DOI: 10.3390/diagnostics10050314
19. Musso G., Gambino R., Cassader M., Pagano G.. **A novel approach to control hyperglycemia in type 2 diabetes: Sodium glucose co-transport (SGLT) inhibitors. Systematic review and meta-analysis of randomized trials**. *Ann. Med.* (2012) **44** 375-393. DOI: 10.3109/07853890.2011.560181
20. Szekeres Z., Toth K., Szabados E.. **The Effects of SGLT2 Inhibitors on Lipid Metabolism**. *Metabolites* (2021) **11**. DOI: 10.3390/metabo11020087
21. Fitchett D., Inzucchi S.E., Cannon C.P., McGuire D.K., Scirica B.M., Johansen O.E., Sambevski S., Kaspers S., Pfarr E., George J.T.. **Empagliflozin Reduced Mortality and Hospitalization for Heart Failure Across the Spectrum of Cardiovascular Risk in the EMPA-REG OUTCOME Trial**. *Circulation* (2019) **139** 1384-1395. DOI: 10.1161/CIRCULATIONAHA.118.037778
22. Jakubiak G.K., Osadnik K., Lejawa M., Kasperczyk S., Osadnik T., Pawlas N.. **Oxidative Stress in Association with Metabolic Health and Obesity in Young Adults**. *Oxid. Med. Cell. Longev.* (2021) **2021** 9987352. DOI: 10.1155/2021/9987352
23. Jakubiak G.K., Osadnik K., Lejawa M., Osadnik T., Golawski M., Lewandowski P., Pawlas N.. **“Obesity and Insulin Re-sistance” Is the Component of the Metabolic Syndrome Most Strongly Associated with Oxidative Stress**. *Antioxidants* (2022) **11**. DOI: 10.3390/antiox11010079
24. Tsatsoulis A., Paschou S.A.. **Metabolically Healthy Obesity: Criteria, Epidemiology, Controversies, and Consequences**. *Curr. Obes. Rep.* (2020) **9** 109-120. DOI: 10.1007/s13679-020-00375-0
25. Genovesi S., Antolini L., Orlando A., Gilardini L., Bertoli S., Giussani M., Invitti C., Nava E., Battaglino M.G., Leone A.. **Cardiovascular Risk Factors Associated with the Metabolically Healthy Obese (MHO) Phenotype Compared to the Metabolically Unhealthy Obese (MUO) Phenotype in Children**. *Front. Endocrinol.* (2020) **11** 27. DOI: 10.3389/fendo.2020.00027
26. Mahabadi A.A., Massaro J.M., Rosito G.A., Levy D., Murabito J.M., Wolf P.A., O’Donnell C.J., Fox C.S., Hoffmann U.. **Association of pericardial fat, intrathoracic fat, and visceral abdominal fat with cardiovascular disease burden: The Framingham Heart Study**. *Eur. Heart J.* (2009) **30** 850-856. DOI: 10.1093/eurheartj/ehn573
27. Xu L., Nagata N., Nagashimada M., Zhuge F., Ni Y., Chen G., Mayoux E., Kaneko S., Ota T.. **SGLT2 Inhibition by Empagliflozin Promotes Fat Utilization and Browning and Attenuates Inflammation and Insulin Resistance by Polarizing M2 Macrophages in Diet-induced Obese Mice**. *Ebiomedicine* (2017) **20** 137-149. DOI: 10.1016/j.ebiom.2017.05.028
28. Andrade-Oliveira V., Câmara N.O.S., Moraes-Vieira P.M.. **Adipokines as Drug Targets in Diabetes and Underlying Disturbances**. *J. Diabetes Res.* (2015) **2015** 681612. DOI: 10.1155/2015/681612
29. Vavruch C., Länne T., Fredrikson M., Lindström T., Östgren C.J., Nystrom F.H.. **Serum leptin levels are independently related to the incidence of ischemic heart disease in a prospective study of patients with type 2 diabetes**. *Cardiovasc. Diabetol.* (2015) **14** 62. DOI: 10.1186/s12933-015-0208-1
30. Morioka T., Emoto M., Yamazaki Y., Kawano N., Imamura S., Numaguchi R., Urata H., Motoyama K., Mori K., Fukumoto S.. **Leptin is associated with vascular endothelial function in overweight patients with type 2 diabetes**. *Cardiovasc. Diabetol.* (2014) **13** 10. DOI: 10.1186/1475-2840-13-10
31. Puurunen V.-P., Kiviniemi A., Lepojärvi S., Piira O.-P., Hedberg P., Junttila J., Ukkola O., Huikuri H.. **Leptin predicts short-term major adverse cardiac events in patients with coronary artery disease**. *Ann. Med.* (2017) **49** 448-454. DOI: 10.1080/07853890.2017.1301678
32. Bickel C., Schnabel R.B., Zeller T., Lackner K.J., Rupprecht H.J., Blankenberg S., Sinning C., Westermann D.. **Predictors of leptin concentration and association with cardiovascular risk in patients with coronary artery disease: Results from the AtheroGene study**. *Biomarkers* (2016) **22** 210-218. DOI: 10.3109/1354750X.2015.1130745
33. Tsai J.-P., Wang J.-H., Chen M.-L., Yang C.-F., Chen Y.-C., Hsu B.-G.. **Association of serum leptin levels with central arterial stiffness in coronary artery disease patients**. *BMC Cardiovasc. Disord.* (2016) **16**. DOI: 10.1186/s12872-016-0268-5
34. Hasan-Ali H., El-Mottaleb N.A.A., Hamed H.B., Abd-Elsayed A.. **Serum adiponectin and leptin as predictors of the presence and degree of coronary atherosclerosis**. *Coron. Artery Dis.* (2011) **22** 264-269. DOI: 10.1097/MCA.0b013e3283452431
35. Feijóo-Bandín S., Portoles M., Roselló-Lletí E., Rivera M., Juanatey J.R.G., Lago F.. **20years of leptin: Role of leptin in cardiomyocyte physiology and physiopathology**. *Life Sci.* (2015) **140** 10-18. DOI: 10.1016/j.lfs.2015.02.016
36. Alix P.M., Guebre-Egziabher F., Soulage C.O.. **Leptin as an uremic toxin: Deleterious role of leptin in chronic kidney disease**. *Biochimie* (2014) **105** 12-21. DOI: 10.1016/j.biochi.2014.06.024
37. Ambarkar M.. **Adipokines and their Relation to Endothelial Dysfunction in Patients with Chronic Kidney Disease**. *J. Clin. Diagn. Res.* (2016) **10** BC04-BC08. DOI: 10.7860/JCDR/2016/15867.7060
38. De Oliveira R.B., Liabeuf S., Okazaki H., Lenglet A.. **The clinical impact of plasma leptin levels in a cohort of chronic kidney disease patients**. *Clin. Kidney J.* (2012) **6** 63-70. DOI: 10.1093/ckj/sfs176
39. Packer M., Zannad F., Butler J., Filippatos G., Ferreira J.P., Pocock S.J., Brueckmann M., Zeller C., Hauske S., Anker S.D.. **Influence of endpoint definitions on the effect of empagliflozin on major renal outcomes in the EMPEROR-Preserved trial**. *Eur. J. Heart Fail.* (2021) **23** 1798-1799. DOI: 10.1002/ejhf.2334
40. Packer M., Januzzi J.L., Ferreira J.P., Anker S.D., Butler J., Filippatos G., Pocock S.J., Brueckmann M., Jamal W., Cotton D.. **Concentration-dependent clinical and prognostic importance of high-sensitivity cardiac troponin T in heart failure and a reduced ejection fraction and the influence of empagliflozin: The EMPEROR -Reduced trial**. *Eur. J. Heart Fail.* (2021) **23** 1529-1538. DOI: 10.1002/ejhf.2256
41. Zinman B., Wanner C., Lachin J.M., Fitchett D., Bluhmki E., Hantel S., Mattheus M., Devins T., Johansen O.E., Woerle H.J.. **Empagliflozin, Cardiovascular Outcomes, and Mortality in Type 2 Diabetes**. *N. Engl. J. Med.* (2015) **373** 2117-2128. DOI: 10.1056/NEJMoa1504720
42. Zinman B., Inzucchi S., Lachin J.M., Wanner C., Ferrari R., Fitchett D., Bluhmki E., Hantel S., Kempthorne-Rawson J., Newman J.. **Rationale, design, and baseline characteristics of a randomized, placebo-controlled cardiovascular outcome trial of empagliflozin (EMPA-REG OUTCOME™)**. *Cardiovasc. Diabetol.* (2014) **13** 102. DOI: 10.1186/1475-2840-13-102
43. Wu P., Wen W., Li J., Xu J., Zhao M., Chen H., Sun J.. **Systematic Review and Meta-Analysis of Randomized Controlled Trials on the Effect of SGLT2 Inhibitor on Blood Leptin and Adiponectin Level in Patients with Type 2 Diabetes**. *Horm. Metab. Res.* (2019) **51** 487-494. DOI: 10.1055/a-0958-2441
44. Vickers S.P., Cheetham S., Headland K., Dickinson K., Grempler R., Mayoux E., Mark M., Klein T.. **Combination of the sodium-glucose cotransporter-2 inhibitor empagliflozin with orlistat or sibutramine further improves the body-weight reduction and glucose homeostasis of obese rats fed a cafeteria diet**. *Diabetes, Metab. Syndr. Obesity Targets Ther.* (2014) **7** 265-275. DOI: 10.2147/DMSO.S58786
45. Bailey C.J., Iqbal N., T’Joen C., List J.F.. **Dapagliflozin monotherapy in drug-naïve patients with diabetes: A randomized-controlled trial of low-dose range**. *Diabetes, Obes. Metab.* (2012) **14** 951-959. DOI: 10.1111/j.1463-1326.2012.01659.x
46. Lee D.M., Battson M.L., Jarrell D.K., Hou S., Ecton K.E., Weir T.L., Gentile C.L.. **SGLT2 inhibition via dapagliflozin improves generalized vascular dysfunction and alters the gut microbiota in type 2 diabetic mice**. *Cardiovasc. Diabetol.* (2018) **17** 62. DOI: 10.1186/s12933-018-0708-x
47. Malínska H., Hütti M., Miklánková D., Hojná S., Papaoušek F., Šilhavý J., Mlejnek P., Zicha J., Hrdlička J., Pravenec M.. **Beneficial Effects of Empagliflozin Are Mediated by Reduced Renal Inflammation and Oxidative Stress in Spontaneously Hypertensive Rats Expressing Human C-Reactive Protein**. *Biomedicines* (2022) **10**. DOI: 10.3390/biomedicines10092066
48. Gohan S., Reshadmanesh T., Khodabandehloo H., Karbalaee-Hasani A., Ahangar H., Arsang-Jang S., Ismail-Beigi F., Dadashi M., Ghanban S., Taheri H.. **The effect of EMPAgliflozin on markers of inflammation in patients with concomitant type 2 diabetes mellitus and Coronary ARtery Disease: The EMPA-CARD randomized controlled trial**. *Diabetol. Metab. Syndr.* (2022) **13** 170. DOI: 10.1186/s13098-022-00951-5
|
---
title: Combination of gamma irradiation and storage condition for improving mechanical
and physical postharvest characteristics of fresh garlic cloves
authors:
- Seyedeh Hoda Yoosefian
- Ebrahim Ahmadi
- Ayat Mohammad‐Razdari
journal: Food Science & Nutrition
year: 2022
pmcid: PMC10002959
doi: 10.1002/fsn3.3186
license: CC BY 4.0
---
# Combination of gamma irradiation and storage condition for improving mechanical and physical postharvest characteristics of fresh garlic cloves
## Abstract
The aim of this study was the discrimination and optimization of irradiation effect under physical and mechanical experiments on garlic. The samples were irradiated with 0, 75, and 150 Gy doses and stored at 4 and 18°C for 5 months. Physical, mechanical, and color properties were measured in the period of storage. Based on the results, all irradiated garlic samples had less quality variation than control samples. Response surface methodology (RSM) optimized dose, storage time, and temperature of the stored garlic which was 75 Gy, 2 months, and 17°C, respectively. In addition, after finding the optimal dose, time, and temperature, the most effective factor as weight loss was obtained and the data were classified by the principal component analysis (PCA) approach. The results showed that the PCA method had a high ability to classify and separate the data obtained from measuring the physicochemical properties of garlic and cover $99\%$ variance of data. Moreover, partial least square (PLS) was applied for predicting weight loss data with R2 0.9999. As well, a mechanical test was investigated for finding the best situation and duration of storage condition. Finally, irradiation prevented the destruction of garlic and saved garlic in the best quality as compared with control or nonirradiated samples. After all this, it can be decided to keep garlic in warehouses and transfer this product with minimum damage.
Irradiated garlic storage for 5 months under controlled condition. Optimization of effective parameters was prepared by RSM. Discrimination of irradiated garlic was done by chemometric approach. Mechanical test was completed for storage situation and duration.
## INTRODUCTION
Preservation of garlic as an export basic crop for a long time is very important. Garlic is rich in folic acid, vitamin C, calcium, iron, magnesium, potassium, zinc, and vitamins B2, B1, and B3 (Balmori et al., 2019; Eugster et al., 2018). It is necessary to maintain the quality of this type of food that is rich in nutrients for a long time (Noda et al., 2018). In order to preserve foods, there are different ways, such as thermal, freezing, drying, fermentation, coating, packaging film, irradiation, atmosphere control, etc. ( Li et al., 2016).
Irradiation is an advanced technical approach to preserve the physical and chemical properties of the food product (Bearth & Siegrist, 2019; Güler et al., 2017). Gamma irradiation is a nonthermal method and has emerged for the preservation of fruits and vegetables to prevent the use of chemical preservatives (Olanya et al., 2015). Many studies have described the useful effect of low‐dose irradiation (50–150 Gy) for sprouting inhibition (Zhang et al., 2016). Some studies showed that storage of irradiated garlic with 100 Gy dose in a refrigerator environment reduces weight loss and spoilage (Martins et al., 2016).
Designing experiments using response surface methodology (RSM) was performed for the first time in 1950. It was initially used for chemical industries, but recently, RSM has been widely used to improve quality, design product, and analyze uncertainty. RSM is a set of statistical and applied mathematics techniques to build experimental models. The aim of this type of design is to optimize the response (output variable) which is influenced by several independent variables (input variables) (Mohammad‐Razdari et al., 2021).
To better understand the data of chemistry, the application of statistical sciences, computers, mathematics, and graphics is necessary. Chemometrics methods are used to gain chemical information obtained in the laboratory, in such a way that useful information is extracted by analyzing the obtained chemical data. Based on this information, the desired experiments can be designed with high efficiency. For data analysis, researchers use PCA (principal component analysis), LDA (linear Discriminant Analysis), ANN (Artificial Neural Network), K‐NN (Nearest Neighbor), and multivariate regression (Tazi et al., 2019).
Due to the cost of keeping garlic in warehouses, limited storage environment, high garlic production in Iran, and its high consumption, it is very important to try to improve the quality of garlic. Also, irradiation is considered as an effective way to preserve garlic and improve its safety (Hassan et al., 2019). Research on the application of irradiation treatment on the garlic product has been carried out to control pests and stop microbial activity, but the effect of treatment on the quality properties, especially the mechanical properties of this product had not been done. Herein, based on our knowledge, for the first time was investigated discrimination of the results based on the chemometric approach and optimization by response surface method in gamma‐irradiated garlic.
## Samples preparation
Fresh garlic bulbs were harvested (30 kg) from a local place in Hamedan, Iran in 2015, and transported (15 kg of all garlic bulbs) to the Atomic Energy Organization of Iran. Fresh samples were stored to uniform size (mean 45 mm), color free from visual defects, or damages and then packed into paper bags. Samples were irradiated by γ‐ray with Gamma Cell 220 (60Co) with a dosage rate of 3.05 Gy/s and the irradiation dose was 0 Gy (control), 75, and 150 Gy (Fernandes et al., 2016). The nonirradiated and irradiated garlic samples were transferred to the laboratory and further divided into paper bags (three packages for each treatment as replication and 10 garlic bulbs were placed inside each paper package) and stored under two different storage conditions: (a) 18°C temperature and $65\%$–$70\%$ relative humidity and (b) refrigerator (second place) was with 4°C temperature and $70\%$–$80\%$ relative humidity.
## Physical and chemical parameters
The garlic clove color was measured with the portable colorimeter (HP‐200, China). In addition, the humidity and weight loss of garlic cloves were investigated through storage (Borchert et al., 2014). In order to estimate the allicin, the amount of total pyruvate and nonenzymatic pyruvate were calculated. After removing the contaminants, 50 g of the sample was crushed well with 100 ml of distilled water in an electric mixer (German Yuro‐Sonic). After 10 min, the resulting solution was filtered through Erlenmeyer filter paper. From the filtered extract mixture, 50 μl was taken and poured into the tube with 2 ml of distilled water and 2 ml of dinitrophenylhydrazine solution. The resulting solution was placed in a hot water bath at 37°C for 10 min, then 2 ml of 1.5 M sodium hydroxide was added to each tube. After adding NaOH, the samples must be read quickly. The absorption of the samples at a wavelength of 515 nm was read with a spectrophotometer (Cary 100 model, manufactured by Varian Company, USA). Sodium pyruvate was also used as a standard in the concentration range of 0–60 mM instead of garlic extract (Gruhlke et al., 2019; Reiter et al., 2020).
## Optimization by RSM
The effect of three independent variables (dose, storage time, and temperature) was investigated by Box Behnken design on the dependent variables (humidity, weight loss, color, and allicin). Assumed that there are three mathematical functions fk for yk which are in equation [1]: [1] yk=fkε1,ε2,ε3 where, ε1, ε2, and ε3 are natural variables. ε1, ε2, and ε3 are dose, temperature, and storage time, respectively. In the response surface method, natural variables change to coded variables (x1, x2, …) equation [2]: [2] yk=fkx1x2x3… This research used a quadratic model for modeling equation [3] (Sukumar & Athmaselvi, 2019): [3] yk=β0+∑$i = 12$βixi+∑$i = 12$βiixi2+∑$i = 12$∑j=i+12βijxixj In equation [3], yk is the predicted response which was considered as dependent variables ($k = 1$, 2, …, 8). Xi is the input coded variable or the same independent variable ($i = 1$, 2, 3). The values of the independent variables were coded between −2 and + 2. All coefficients are parameters of regression coefficients. Using the quadratic model, five mathematical models were evaluated for each dependent variable. Each factor in the design was measured at three different levels (2−, 0, +2), two axial points, and six repetitions at the central point (Table 1). Experimental design as well as process optimization was performed using Design Expert11 software. Analysis of variance on quadratic model coefficients was also performed using this software. Significant sentences in the model were obtained using analysis of variance for each response.
**TABLE 1**
| Independent variables | Coded variables | Natural variables | Levels | Levels.1 | Levels.2 | Levels.3 | Levels.4 |
| --- | --- | --- | --- | --- | --- | --- | --- |
| Storage time (month) | X1 | ɛ1 | −2 | −1 | 0 | 1 | 2.0 |
| Dose (Gy) | X2 | ɛ2 | | −1 | 0 | 1 | |
| Temperature (°C) | X3 | ɛ3 | | −1 | 0 | 1 | |
## Principal component analysis (PCA)
The principal component analysis is a statistical method that uses orthogonal transfer for the conversion of a set of observed correlated variables into a set of uncorrelated linear variables which are principal components. This conversion is performed in a way that the first component has the highest variance, and then, the other components also have high variance with limitations of course, and all components are perpendicular to the previous components. PCA is a high precession sensitive method to find main variables. PCA is one of the common methods in data analysis and dimension reduction in multivariate systems (Tazi et al., 2019). Additionally, loading plots provide information on the relative importance of this set of sensors in the analysis of principal components (Barbosa et al., 2020).
## Partial least square (PLS)
One of the most powerful techniques in the field of chemometric is factor analysis. Factor analysis is a multivariate method that provides important and useful information by reducing the size of the data and the minimum number of perpendicular vectors as PLS. PLS is a linear combination method for main variables in the matrices and instead of the I × J matrix and its variables can be defined as a linear combination of the J factor and finally, new variables are defined for the matrix (Rambo et al., 2020).
## Relaxation test
In order to measure the viscoelastic properties, a relaxation test was performed on the samples in a pressure plate accordingly. A texturometer (Zwick/Roell, bt1_fr0.5th.d14 model, Germany, xforce hp load cell with 500 N capacity) with a 25‐mm‐diameter probe pressure plate was used to perform the test. These experiments were done at room temperature under the following conditions: Initial loading force 0.1 N, start speed test at 70 mm.min−1, load speed at 2 mm.min−1, and time of the test was 400 s.
The basic requirement of designing machines for the processing of fruit and food is having information about the physical characteristics of the fruit. In the lines of transportation and packing, the fruit is subjected to different loads that may damage its tissue. The stress and strain state under static and dynamic loads that are related to the nature and mechanical behavior of the material is considered as the first step in quantitative analysis of the characteristics of agricultural products (Mahiuddin et al., 2018). One of the best models for investigating load damages of fruit is the general Maxwell model stress relaxation test design. To overcome this drawback in agricultural products, springiness with coefficient Ee was added to the general Maxwell model (Mohsenin, 1986). Equation [4] shows the expression of Maxwell's mathematical model: [4] σt=σ1.e−tTrel1+σ2.e−tTrel2+σe where the consideration σi = ε0Ei, σe = ε0Ee and can be written as equation [5]: [5] σt=E1.e−tTrel1+E2.e−tTrel2+Eeε0 The obtained model coefficients were determined and evaluated from relaxation stress curves. Residues were determined using the sequential model. All stress relaxation time models were calculated by MATLAB R.13 software.
Table 5 shows the mean values of stress and relaxation time in the viscoelastic element for the two‐component Maxwell model that is calculated with equation [1]. MATLAB R.13 software was used to fit data on the models. In this way, with the insertion of variables, the best model and relaxation time was concluded with CF Tool command and definition of stress function. According to the result, relaxation time increased with increasing storage time and this process was greater for the irradiated samples than for control samples. Also, the relaxation time for samples placed at 18°C was $27\%$ longer than samples placed at 4°C. The results showed that irradiation caused humidity loss, and samples lose their viscous state and become more elastic (Calado et al., 2018).
**TABLE 5**
| Dose treatment (Gy) | Temperature (°C) | Stress (MPa) | Stress (MPa).1 | Stress (MPa).2 | Relaxation time (s) | Relaxation time (s).1 |
| --- | --- | --- | --- | --- | --- | --- |
| Dose treatment (Gy) | Temperature (°C) | σ1 | σ2 | σe | Trel1 | Trel2 |
| 150 | 4 | 5.0112 | 5.2523 | 4.9878 | 1 × 10−6 | 11 × 10−4 |
| 150 | 18 | 4.8796 | 5.1121 | 4.8318 | 3 × 10−7 | 11 × 10−5 |
| 75 | 4 | 4.8983 | 5.117 | 4.90 | 4 × 10−7 | 17 × 10−5 |
| 75 | 18 | 4.9293 | 5.1460 | 5.11 | 6 × 10−8 | 2 × 10−6 |
| 0 | 4 | 4.8874 | 5.1178 | 4.8873 | 6 × 10−8 | 11 × 10−6 |
| 0 | 18 | 4.94 | 5.092 | 4.9426 | 8 × 10−8 | 8 × 10−6 |
Relaxation time is different based on the characteristics of the viscoelastic or viscous substances, but this time is longer in elastic material. The results of chemical studies (Calado et al., 2018) showed that irradiation is the cause of humidity loss in the product and evaporation of water within tissue. With increasing storage time, the relaxation time average in the control and irradiated samples with 75 and 150 Gy doses are $5\%$, $9\%$, and $12\%$ at 4°C and $8.5\%$, $11\%$, and $17\%$ at 18°C, respectively. The relaxation time depends on the humidity content of the product and is reduced because of the humidity (Danalache et al., 2015). However, critical stress decreases with an increase in storage time, but this change does not occur in irradiated samples. Thus, irradiated garlic samples with 150 Gy dose have more critical stress too.
## Shear test
In order to determine enough shear force for the garlic product, an experimental shear tool with a commercial single sickle knife section and a counter shear was used as the twofold shear test. Finally, the test was done using a Zwick/Roell texturometer. The diameter cutter probe, preload, distance of probe to the bottom of the page, and speed at start position are 0.5 mm, 0.3 N, 65, mm, and 300 mm. min−1, respectively (Figure 1).
**FIGURE 1:** *3D couture of humidity and weight loss changing with dose, time, and temperature*
The shear stress was calculated in MPa using equation [6] (Mohsenin, 1986): [6] τ=FmaxA where *Fmax is* the maximum shear force (N) and A is the cross‐sectional area of the stalk at shear planes (mm2). The shear energy was calculated by the area under these curves to the maximum force of the curve (Chen et al., 2004). The cross‐sectional area for calculating the shear test was calculated by Solid Works R.2015 software. For this purpose, garlic cloves were designed in the software to obtain the main dimensions (length, width, and thickness) of the samples. By drawing a three‐dimensional image, the rectangular area was measured in these dimensions (Figure 1).
As shown in Figure 7a, shear stress in garlic is reduced with increasing storage time, but irradiation keeps the shear stress almost constant. According to the results, shear stress was reduced by about $27\%$, $23\%$, and $18\%$ at 4°C for control samples and $32\%$, $29\%$, and $24\%$ at 18°C for irradiated samples (75 and 150 Gy). Lower temperature (4°C) does not allow the shear stress to reduce, and as a result, irradiated and control samples stored at 4°C have higher values of shear stress.
**FIGURE 7:** *Effect of storage time, irradiation dose, and temperature on (a) shear force, (b) shear stress (c) Young's modulus, and (d) energy of rupture*
Shear strain was significantly reduced, because of the increase in storage time and storage temperature (Figure 7b). But in contrast, irradiation prevents shear strain reduction, because of the tissue damage and softening of the stored sample (Calado et al., 2018). As shear strain in control, irradiated with 75 and 150 Gy doses decreased $29\%$, $23\%$, and $20\%$ at 4°C and $35\%$, $28\%$, and $26\%$ at 18°C, respectively.
According to the results (Table 6), energy of rupture reduced with increasing storage time and storage temperature. But, when increased irradiation dose, the energy of rupture has less reduction, because of the preservation of tissues from penetration. Thus, the values for control and irradiated samples were (75 and 150 Gy) $14\%$, $13\%$, and $13\%$ for storage at 4°C, and $17\%$, $16\%$, and $15\%$ for storage at 18°C, respectively. Figure 7c shows the preserved tissues in irradiated samples and they have higher values.
**TABLE 6**
| Variables | Df | τmax (MPa) | ɛmax | Emax (MPa) | σrel (MPa) | Trel |
| --- | --- | --- | --- | --- | --- | --- |
| Irradiation | 2.0 | 0.01523 a | 0.2213 a | 0.001025 a | 0.00 b | 0.011 a |
| Temperature | 1.0 | 0.02136 a | 0.1145 a | 0.005621 a | 0.00 b | 0.498 a |
| Storage | 5.0 | 0.345 c | 0.223 c | 0.528 c | 0.0131 a | 0.711 c |
| Irradiation × temperature | 2.0 | 0.02232 b | 0.1256 b | 0.01002 a | 0.0311 a | 0.0304 a |
| Irradiation × storage | 10.0 | 0.02136 a | 0.1145 a | 0.005621 a | 0.601 c | 0.609 c |
| Temperature × storage | 5.0 | 0.01964 b | 0.1056 a | 0.003216 b | 0.192 c | 0.457 c |
| Irradiation × temperature × storage | 10.0 | 0.01687 a | 0.0451 b | 0.006321 a | 0.122 c | 0.647 c |
| Test error | 41.0 | 0.001247 | 0.0225 | 0.2654 | 0.06254 | 1.012 |
| CV (%) | | 9.52 | 3.254 | 0.987 | 21.45 | 2.365 |
As mentioned, increasing storage time can decrease the shear stress, because the garlic samples lose their intracellular water with increasing storage time. Of course, due to biological activity, breathing decreased humidity content (Pérez et al., 2007). Compressibility is most of the original strains because the cells are full of water in fresh garlic (Zhou et al., 2010). In a study that had been done on garlic, there was a decrease in the shear stress and energy of rupture when tissue was destroyed (Llamas et al., 2013).
## Statistical analysis
These experiments were designed based on a factorial design with three factors of gamma ray (control, 75, and 150 Gy doses), storage time (1–5 months), and the storage temperature (4 and 18°C). The analysis of variance was carried out on data using SPSS software (IBM SPSS Statistics 22, IBM, NY). Also, differences between means were determined with Tukey post hoc comparison tests. p‐values of 0.05 or less were considered significant.
## Optimization
According to the results of (Figure 1), the effects of radiation dose and storage time variables were significant on humidity variables. With a 150 Gy irradiation dose and 5 months of storage time, the lowest amount of humidity was observed in garlic samples. With increasing storage time, the humidity in the control and irradiated samples of 75 and 150 Gy decreased to $14.51\%$, $17.37\%$, and $18.97\%$, respectively. Also, with increasing the intensity of irradiation, the percentage of humidity in irradiated samples with intensities of 75 and 150 Gy decreased to $4.15\%$ and $6.94\%$, respectively, compared to the control sample at the end of storage. The amount of initial humidity indicates water loss of the product and water loss mainly depends on the intensity of product respiration and ethylene production (Sukumar & Athmaselvi, 2019).
Also, (Figure 1) shows weight loss changes after 5 months of storage and the high effect of 150 Gy dose on garlic samples. According to the results, storage time is a very important and influential indicator on weight loss. Over the storage time and increasing the intensity of irradiation, the percentage of weight loss increased significantly during 5 months. Weight loss will be followed by dehydration, metabolic activity, respiration, and transpiration (Ghasemi & Chayjan, 2019). Respiration intensity in fruits and vegetables increases with ripening, which reduces the storage of nutrients in products and weight loss (Durante et al., 2020). Also, the application of radiation, first with the application of 75 Gy intensity and then 150 Gy intensity increased the percentage of weight loss (Figure 2). One of the most important factors in weight loss during irradiation is water loss due to energy absorption caused by waves (Sukumar & Athmaselvi, 2019). In a study, the effect of storage time for 6 months in cold and traditional warehouses on garlic cultivars in Hamadan was investigated. According to the results, with increasing storage time, the percentage of weight loss of samples increased and in cold warehouses, these changes were less than traditional warehouses (Tazi et al., 2019). Another study was investigated the effect of radiation intensity of 50 and 150 Gy and storage time for 45 days on garlic samples. According to the results, with increasing storage time, the percentage of weight loss increased and irradiated samples had a higher percentage of weight loss than the control (Sukumar & Athmaselvi, 2019).
**FIGURE 2:** *3D couture of color parameters and allicin changing with dose, time, and temperature*
Analysis of variance (Table 2) on humidity and weight loss indices is presented the effect of independent variables of radiation dose, storage temperature, and time. Statistically, it is a significant model with a p‐value of less than 0.05. Based on the results of p‐value in modeling, the humidity content of the sample is equal to 0.0144 and shows a significant effect of independent variables on humidity. Also, the main effects of three independent parameters of time and temperature on humidity are significant. Also, the main and secondary effects of independent variables of A, B, and C on weight loss are significant.
**TABLE 2**
| Parameter | Source | Sum of squares | Df | Mean square | F value | p‐value |
| --- | --- | --- | --- | --- | --- | --- |
| Parameter | Source | Sum of squares | Df | Mean square | F value | Prob > F |
| Humidity | Model | 0.42 | 3 | 0.14 | 3.37 | 0.0414 |
| Humidity | A‐Time | 0.13 | 1 | 0.13 | 3.20 | 0.0467 |
| Humidity | B‐Dose | 0.13 | 1 | 0.13 | 3.16 | 0.0488 |
| Humidity | C‐Temp | 0.16 | 1 | 0.16 | 3.76 | 0.0747 |
| Humidity | Residual | 0.54 | 13 | 0.042 | | |
| Humidity | Lack of Fit | 0.054 | 9 | 0.060 | | |
| Humidity | Pure Error | 0.000 | 4 | 0.000 | | |
| Humidity | Cor Total | 0.97 | 16 | | | |
| Weight loss | Model | 0.098 | 9 | 0.011 | 238.32 | <0.0001 |
| Weight loss | A‐Time | 0.072 | 1 | 0.072 | 1570.36 | <0.0001 |
| Weight loss | B‐Dose | 0.011 | 1 | 0.011 | 232.97 | <0.0001 |
| Weight loss | C‐Temp | 1.826 E‐003 | 1 | 1.826 E‐003 | 39.94 | 0.0004 |
| Weight loss | AB | 6.074 E‐004 | 1 | 6.074 E‐004 | 13.28 | 0.0082 |
| Weight loss | AC | 6.702 E‐004 | 1 | 6.702 E‐004 | 14.66 | 0.0065 |
| Weight loss | BC | 7.957 E‐003 | 1 | 7.957 E‐003 | 174.03 | <0.0001 |
| Weight loss | A2 | 5.218 E‐004 | 1 | 5.218 E‐004 | 11.41 | 0.0118 |
| Weight loss | B2 | 2.804 E‐003 | 1 | 2.804 E‐003 | 61.32 | 0.0001 |
| Weight loss | C2 | 1.466 E‐003 | 1 | 1.466 E‐003 | 32.06 | 0.0008 |
| Weight loss | Residual | 3.200 E‐004 | 7 | 4.572 E‐005 | | |
| Weight loss | Lack of Fit | 3.200 E‐004 | 3 | 1.067 E‐004 | | |
| Weight loss | Pure Error | 0.000 | 4 | 0.000 | | |
| Weight loss | Cor Total | 0.098 | 16 | | | |
Changes in the brightness of garlic samples (Figure 2) at the beginning of storage or in the early months of storage have higher values and decreased the brightness of the samples after 5 months. According to the results of the analysis of variance (Table 3), the storage time on the L* index was not significant but had a great effect on the main effects of dose and temperature on the L* sample. Among the interactions, the interaction of the AB parameter on L* was not significant but the effects of AC and BC on L* were significant. The results of (Figure 2) show the three‐dimensional contour L*. According to the results, the control sample and 150 Gy had higher brightness.
**TABLE 3**
| Parameter | Source | Sum of squares | Df | Mean square | F value | p‐value |
| --- | --- | --- | --- | --- | --- | --- |
| Parameter | Source | Sum of squares | Df | Mean square | F value | Prob > F |
| L* | Model | 31.40 | 9 | 3.49 | 13.86 | 0.0011 |
| L* | A‐Time | 0.074 | 1 | 0.074 | 0.29 | 0.6042 |
| L* | B‐Dose | 3.25 | 1 | 3.25 | 12.92 | 0.0088 |
| L* | C‐Temp | 11.26 | 1 | 11.26 | 44.74 | 0.0003 |
| L* | AB | 4.225 E‐003 | 1 | 4.225 E‐003 | 0.017 | 0.9005 |
| L* | AC | 1.14 | 1 | 1.14 | 4.55 | 0.0704 |
| L* | BC | 1.33 | 1 | 1.33 | 5.30 | 0.0548 |
| L* | A2 | 0.79 | 1 | 0.79 | 3.13 | 0.1202 |
| L* | B2 | 11.88 | 1 | 11.88 | 47.22 | 0.0002 |
| L* | C2 | 0.81 | 1 | 0.81 | 3.20 | 0.1166 |
| L* | Residual | 1.76 | 7 | 0.25 | | |
| L* | Lack of Fit | 1.76 | 3 | 0.59 | | |
| L* | Pure Error | 0.000 | 4 | 0.000 | | |
| L* | Cor Total | 33.16 | 16 | | | |
| a* | Model | 1.30 | 9 | 0.14 | 20.60 | 0.0003 |
| a* | A‐Time | 0.43 | 1 | 0.43 | 61.68 | 0.0001 |
| a* | B‐Dose | 0.021 | 1 | 0.021 | 3.00 | 0.1270 |
| a* | C‐Temp | 5.513 E‐003 | 1 | 5.513 E‐003 | 0.79 | 0.4047 |
| a* | AB | 0.000 | 1 | 0.000 | 0.000 | 1.0000 |
| a* | AC | 0.036 | 1 | 0.036 | 5.15 | 0.0575 |
| a* | BC | 0.099 | 1 | 0.099 | 14.15 | 0.0071 |
| a* | A2 | 0.18 | 1 | 0.18 | 26.17 | 0.0014 |
| a* | B2 | 0.21 | 1 | 0.21 | 30.07 | 0.0009 |
| a* | C2 | 0.33 | 1 | 0.33 | 46.67 | 0.0002 |
| a* | Residual | 0.049 | 7 | 7.011 E‐003 | | |
| a* | Lack of Fit | 0.049 | 3 | 0.016 | | |
| a* | Pure Error | 0.000 | 4 | 0.000 | | |
| a* | Cor Total | 1.35 | 16 | | | |
| b* | Model | 39.47 | 6 | 6.58 | 24.91 | <0.0001 |
| b* | A‐Time | 34.07 | 1 | 34.07 | 129.01 | <0.0001 |
| b* | B‐Dose | 8.000 E‐004 | 1 | 8.000 E‐004 | 3.029 E‐003 | 0.9572 |
| b* | C‐Temp | 0.60 | 1 | 0.60 | 2.27 | 0.1628 |
| b* | AB | 1.225 E‐003 | 1 | 1.225 E‐003 | 4.638 E‐003 | 0.9470 |
| b* | AC | 4.62 | 1 | 4.62 | 17.50 | 0.0019 |
| b* | BC | 0.17 | 1 | 0.17 | 0.65 | 0.4382 |
| b* | Residual | 2.64 | 10 | 0.26 | | |
| b* | Lack of Fit | 2.64 | 6 | 0.44 | | |
| b* | Pure Error | 0.000 | 4 | 0.000 | | |
| b* | Cor Total | 42.11 | 16 | | | |
| Allicin | Model | 155.59 | 9 | 17.29 | 37.02 | <0.0001 |
| Allicin | A‐Time | 126.69 | 1 | 126.69 | 271.34 | <0.0001 |
| Allicin | B‐Dose | 0.067 | 1 | 0.067 | 0.14 | 0.7156 |
| Allicin | C‐Temp | 0.55 | 1 | 0.55 | 1.17 | 0.3154 |
| Allicin | AB | 4.40 | 1 | 4.40 | 9.43 | 0.0180 |
| Allicin | AC | 2.06 | 1 | 2.06 | 4.42 | 0.0736 |
| Allicin | BC | 0.46 | 1 | 0.46 | 0.99 | 0.3539 |
| Allicin | A2 | 12.64 | 1 | 12.64 | 27.07 | 0.0012 |
| Allicin | B2 | 7.12 | 1 | 7.12 | 15.24 | 0.0059 |
| Allicin | C2 | 2.69 | 1 | 2.69 | 5.76 | 0.0475 |
| Allicin | Residual | 3.27 | 7 | 0.47 | | |
| Allicin | Lack of Fit | 3.27 | 3 | 1.09 | | |
| Allicin | Pure Error | 0.000 | 4 | 0.000 | | |
| Allicin | Cor Total | 158.86 | 16 | | | |
According to the analysis of variance (Table 3), the main effect of storage time on the parameter a* is significant and the main effects of B and C on a* are not significant. Also, the interactions of AC and BC on a* are significant. The effects of the square on a* are significant. According to (Figure 2), in the control sample and 150 Gy, a* is more than in the 75 Gy sample.
With increasing storage time, the value of a* increased, which is due to the reduction of humidity in the sample tissue and the shrinkage process (Kainthola et al., 2020). With increasing the amount of radiation intensity, the value of a* increased compared to nonirradiated samples. Because, as mentioned, with increasing the intensity of radiation and absorption of this energy by intracellular water molecules and its evaporation, the intracellular water decreases and the amount of redness increases over time (Mohammad‐Razdari et al., 2021).
According to the results of analysis of variance (Table 3), the main effect of time on b* is significant and the main effects of dose and temperature on b* were not significant. Also, among the interactions, only AC is significant. Due to the three‐dimensional contour (Figure 2), the dose parameter is ineffective and only has increased b* in all samples after 5 months.
On the other hand, with increasing storage time, b* or the amount of jaundice increased and with the application of irradiation intensity, b* of the irradiated sample decreased by 75 Gy compared to the control sample but increased the amount of b*with increasing irradiation intensity up to 150 Gy. Increased b* with increasing radiation intensity is due to the fact that most plant products or organs typically lose large amounts of volatiles during puberty and aging. The yellowing of the white garlic sample also has a linear relationship with the time elapsed after harvest, which is also proportional to the intensity used (Olanya et al., 2015). In a study, by increasing the amount of radiation intensity to a certain value and putting the samples at 4°C, the amount of b* of the irradiated sample increased less than the control sample (Durante et al., 2020). This means that the L* index decreases with increasing storage time and increasing the amount of irradiation intensity (Tables 1 and 2). This is due to the decrease in humidity and the reduction of water between cells during the storage period (Noda et al., 2018). On the other hand, with increasing radiation intensity, the intracellular water, as mentioned above, decreases and evaporation takes place (Sukumar & Athmaselvi, 2019). Similar results were obtained in another study that decreased the value of L* with increasing the amount of radiation intensity and storage time (Durante et al., 2020).
Table 4 shows the results of RSM analysis of variance and the proposed model. According to the results, there are significant changes in storage time on allicin. But the main changes in irradiation dose and storage temperature on the allicin are not significant. Also, among the interactions, the interaction effect of storage time at irradiation dose is significant on allicin. But the quadratic effects of all independent variables (time–dose and temperature) on allicin are significant. According to the three‐dimensional contour (Figure 2), it can be said that at the beginning of storage, the allicin was high, but after 5 months, the allicin has decreased sharply, and this reduction has occurred in all samples and treatments.
**TABLE 4**
| Response | Intercept | A | B | C | AB | AC | BC | A2 | B2 | C2 |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| Humidity | 7.6009 | −0.1294 | 0.1285 | 0.1401 | | | | | | |
| P | | 0.0467 | 0.0488 | 0.0747 | | | | | | |
| Weight Loss | 0.2747 | 0.0947 | −0.0364 | −0.0151 | −0.012 | 0.0129 | −0.0446 | −0.0111 | 0.0258 | −0.0186 |
| P | | <0.0001 | <0.0001 | 0.0004 | 0.008 | 0.0065 | <0.000 | 0.0118 | 0.0001 | 0.0008 |
| L | 79.15 | −0.09625 | −0.6375 | −1.18625 | −0.0325 | −0.535 | 0.5775 | 0.4325 | 1.68 | 0.4375 |
| P | | 0.6042 | 0.0088 | 0.0003 | 0.9005 | 0.0704 | 0.0548 | 0.1202 | 0.0002 | 0.1166 |
| a* | −0.15 | 0.2325 | 0.05125 | 0.02625 | −0.0000 | −0.095 | 0.1575 | −0.20875 | 0.2237 | 0.27875 |
| P | | 0.0001 | 0.1270 | 0.4047 | 1.0000 | 0.0575 | 0.0071 | 0.0014 | 0.0009 | 0.0002 |
| b* | 11.7194 | 2.06375 | 0.01 | 0.27375 | 0.0175 | 1.075 | −0.2075 | | | |
| P | | <0.0001 | 0.9572 | 0.1628 | 0.9470 | 0.0019 | 0.4382 | | | |
| Allicin | 10.1767 | −3.97954 | 0.0916 | −0.2612 | 1.0491 | −0.7184 | −0.3391 | −1.73254 | 1.3 | 0.799125 |
| P | | <0.0001 | 0.7156 | 0.3154 | 0.0180 | 0.0736 | 0.3539 | 0.0012 | 0.0059 | 0.0475 |
| Legend | | p < .01 | .01 <= p < .05 | .05 <= p < .10 | p >= .10 | | | | | |
As the shelf life increased, the stiffness of the samples decreased (Figure 2). Measurement of pyruvate during 5 months of storage shows the enzymatic degradation and decomposition of allicin, which results in a reduction in the aroma and flavoring compositions of garlic during storage. On the other hand, for irradiated samples, the intensity of allicin decrease was less. This means that by irradiating the samples, the allicin is maintained and the gamma ray prevents the decomposition and degradation of enzymatic compounds, and with increasing intensity, further degradation of enzymes is prevented (Chen et al., 2020). In one study, a decrease in pyruvate was observed during storage, which is due to the breakdown of tasteless compounds by peptidase enzymes and in fact reduces the activity of the enzyme alliinase and the aroma and flavoring compounds of garlic (Chen et al., 2020; Mandal et al., 2019).
## Introducing optimal points and desirability index
Based on the results of the desirability index, the best and most optimal answer is obtained for the variables of time, temperature, and dose so that it can be decided based on the results of the variables of humidity, weight loss, color, and allicin to maintain garlic samples. Desirability index is the way to find the best points that are 0.76. According to (Figure 3), the desirability index had the highest value at the beginning of the storage period. According to the contour of (Figure 3), the humidity was high at the beginning of storage time and decreased after 5 months. The best humidity point for garlic samples is 63.9281. Also, there is weight loss in irradiated samples and control during storage. Proper storage of garlic products should have an optimal value of 0.0071 weight loss. Moreover, according to the contour results, the best values of L*, a*, b*, and allicin are 81.082, 0.2047, 8.6841, 8.7033, and 13.6084, respectively, and the best storage temperature is 17°C.
**FIGURE 3:** *2D couture of desirability, color parameters, and allicin changing with dose, time, and temperature and best optimal point*
According to the results and expressions of quadratic equations, there is a linear relationship between dependent and independent variables. The positive sign of the coefficients indicates the direct effect and the negative sign indicates the inverse effect of the independent and dependent variables. According to (Table 4), for the humidity‐dependent variable, the intercept is 7.6009 and the negative coefficients A indicate that the storage time has an inverse effect on humidity and humidity decreases with increasing storage time. According to the equations, the increase of A, AC, and B2 decreases B, C, AB, BC, and A2 variables, and C2 increases weight loss. Also for the variable L*, the increase of A, B, C, AB, and AC decreases L* and the BC, A2, B2, and C2 variables are directly related to L*. According to (Table 4), A, B, C, BC, B2, and C2 parameters are directly related to a* and have reverse relation to the other parameters. Also, all parameters except BC are directly related to b* and as these parameters increase, b* also increases.
Finally, in the study of the degree of allicin, the independent parameters A, C, AC, BC, and A2 have an inverse effect on the allicin and if these parameters increase independently, the reverse downward trend occurs for the allicin and the parameters B, AB, B2, and C2 are also related. It has a steep downward trend and increases allicin with their increase.
## Classification of weight loss by PCA
Due to the large effect of independent variables on weight loss, PCA results of weight loss changes have been investigated. The results of PCA on weight loss show a good and significant difference between the control samples, 75 and 150 Gy at two temperatures of 4 and 18°C and in different months. ( Figure 4a) of PC1 and PC2 covers the ability of $71\%$ and $21\%$ of the data, respectively, and observed a total of $92\%$ of the variance between the data, and is able to make a good distinction between the samples. It is noteworthy that the control groups, 75 and 150 Gy, have been identified together and can be grouped together. In the second month of storage, PC1 and PC2 were able to distinguish $66\%$ and $31\%$ of the variance between the data and $97\%$ of the total variance between the data, respectively, and were able to differentiate the data. This indicates the high accuracy during testing and the ability of PC detection in data classification. Also in months 3, 4, and 5, PC values have $99\%$, $95\%$, and $85\%$ variance between data, respectively.
**FIGURE 4:** *PCA classification for control and irradiated samples in 5‐month storage based on storage time (a) month 1, (b) month 2, (c) month 3, (d) month 4, and (e) month 5*
The best classification was done in month 3. In other words, it can be said that from month 3 onwards and with the aging process of garlic samples, the data in almost all groups are closer to each other and the weight loss of samples is obvious. Also, (Figure 5) shows the PC results for separating the samples at 5 months of storage for the control samples, 75 and 150 Gy. PC was able to cover a total of $93\%$ of the variance between the data and in the samples of irradiated garlic with a dose of 75 and 150 Gy, the PC was able to cover $94\%$ and $85\%$ of the variance between the data and it can be said that the overall optimization results with RSM and high PCA accuracy, the best radiation dose is 75 Gy to see the best weight loss results.
**FIGURE 5:** *PCA classification for control and irradiated samples in 5‐month storage based on treatments*
## Prediction by PLS
In this section, (Figure 6) shows the PLS results of weight loss data in the control and irradiated samples at 5 months of storage at 4 and 18°C. Predicted weights versus actual weights are shown and the results show that PLS has the ability to predict data in 5 months of high‐performance R2 storage.
**FIGURE 6:** *PLS results for weight loss of all treatments*
## CONCLUSIONS
Irradiation preserves mechanical and physical properties of the food. In this study, irradiated samples stored at 18 and 4°C had lower changes in color parameters (a*, b*, and L*). Also, based on the results, irradiation has a significant effect on garlic tissues and preserves the tissues of the sample, but a high dose has the reverse effect on shear stress, shear strain, and energy of rupture. Therefore, the storage time had a significant effect on mechanical properties. Likewise, results showed that chemometrics methods like PCA and PLS can classify and predict storage condition data exactly. Then, in a relaxation test, two components of the Maxwell model were obtained for garlic cloves. Moreover, with increasing storage time, the relaxation time average increased in irradiated samples than in the control samples. Finally, the results showed that the irradiation preserved some mechanical and physical properties of garlic during 5 months of storage.
## FUNDING INFORMATION
This study was supported by the office of vice chancellor for research at Bu‐Ali Sina University (Thesis No. 4375).
## CONFLICT OF INTEREST
The authors have no conflict of interest to declare.
## PRACTICAL APPLICATIONS
Irradiation is an effective way for increasing the shelf life of food products. Irradiated garlic stored for 5 months under controlled condition and optimization of effective parameters was done by RSM. Also, discrimination of irradiated garlic was done by chemometric approaches. Moreover, a mechanical test was completed for the storage simulation by the General Maxwell model and shear test.
## DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available on request from the corresponding author.
## References
1. Balmori D. M., Domínguez C. Y. A., Carreras C. R., Rebatos S. M., Farías L. B. P., Izquierdo F. G., Berbara R. L. L., García A. C.. **Foliar application of humic liquid extract from vermicompost improves garlic (**. *International Journal of Recycling of Organic Waste in Agriculture* (2019) **8** 103-112
2. Barbosa S., Saurina J., Puignou L., Núñez O.. **Classification and authentication of paprika by UHPLC‐HRMS fingerprinting and multivariate calibration methods (PCA and PLS‐DA)**. *Food* (2020) **9** 486
3. Bearth A., Siegrist M.. **“As long as it is not irradiated”–influencing factors of US consumers' acceptance of food irradiation**. *Food Quality and Preference* (2019) **71** 141-148
4. Borchert N. B., Cruz‐Romero M. C., Mahajan P. V., Ren M., Papkovsky D. B., Kerry J. P.. **Application of gas sensing technologies for non‐destructive monitoring of headspace gases (O2 and CO2) during chilled storage of packaged mushrooms (**. *Food Packaging and Shelf Life* (2014) **2** 17-29
5. Calado T., Fernández‐Cruz M. L., Verde S. C., Venâncio A., Abrunhosa L.. **Gamma irradiation effects on ochratoxin a: Degradation, cytotoxicity and application in food**. *Food Chemistry* (2018) **240** 463-471. PMID: 28946299
6. Chen K., Nakasone Y., Xie K., Sakao K., Hou D. X.. **Modulation of allicin‐free garlic on gut microbiome**. *Molecules* (2020) **25** 682. PMID: 32033507
7. Chen Y., Gratton J. L., Liu J.. **Power requirements of hemp cutting and conditioning**. *Biosystems Engineering* (2004) **87** 417-424
8. Danalache F., Beirão‐da‐Costa S., Mata P., Alves V. D., Moldao‐Martins M.. **Texture, microstructure and consumer preference of mango bars jellified with gellan gum**. *LWT‐ Food Science and Technology* (2015) **62** 584-591
9. Durante M., Ferramosca A., Treppiccione L., Di Giacomo M., Zara V., Montefusco A., Lenucci M. S.. **Application of response surface methodology (RSM) for the optimization of supercritical CO2 extraction of oil from patè olive cake: Yield, content of bioactive molecules and biological effects in vivo**. *Food Chemistry* (2020) **332**. PMID: 32603919
10. Eugster A., Murmann P., Känzig A., Breitenmoser A.. **A specific but nevertheless simple real‐time PCR method for the detection of irradiated food shown detailed at the example of garlic (**. *European Food Research and Technology* (2018) **244** 1-7
11. Fernandes Â., Barreira J. C., Antonio A. L., Oliveira M. B. P., Martins A., Ferreira I. C.. **Extended use of gamma irradiation in wild mushrooms conservation: Validation of 2 kGy dose to preserve their chemical characteristics**. *LWT‐ Food Science and Technology* (2016) **67** 99-105
12. Ghasemi A., Chayjan R. A.. **Optimization of pelleting and infrared‐convection drying processes of food and agricultural waste using response surface methodology (RSM)**. *Waste and Biomass Valorization* (2019) **10** 1711-1729
13. Gruhlke M. C., Antelmann H., Bernhardt J., Kloubert V., Rink L., Slusarenko A. J.. **The human allicin‐proteome: S‐thioallylation of proteins by the garlic defence substance allicin and its biological effects**. *Free Radical Biology and Medicine* (2019) **131** 144-153. PMID: 30500420
14. Güler S. K., Bostan S. Z., Çon A. H.. **Effects of gamma irradiation on chemical and sensory characteristics of natural hazelnut kernels**. *Postharvest Biology and Technology* (2017) **123** 12-21
15. Hassan A. B., Ahmed S. M., Elkhatim K. A. S., Abdelhalim T. S., Fawale S. O., Adiamo O. Q., Ahmed I. A. M.. **Effect of gamma irradiation and microwave heating treatments on microbial load and antioxidant potentials in cinnamon, fennel and hot pepper**. *Journal of Food Measurement and Characterization* (2019) **13** 1-9
16. Kainthola J., Kalamdhad A. S., Goud V. V.. **Optimization of process parameters for accelerated methane yield from anaerobic co‐digestion of rice straw and food waste**. *Renewable Energy* (2020) **149** 1352-1359
17. Li L., Lichter A., Chalupowicz D., Gamrasni D., Goldberg T., Nerya O., Porat R.. **Effects of the ethylene‐action inhibitor 1‐methylcyclopropene on postharvest quality of non‐climacteric fruit crops**. *Postharvest Biology and Technology* (2016) **111** 322-329
18. Llamas D. P., Patón L. G., Díaz M. G., Serna J. G., Sáez S. B.. **The effects of storage duration, temperature and cultivar on the severity of garlic clove rot caused by**. *Postharvest Biology and Technology* (2013) **78** 34-39
19. Mahiuddin M., Khan M. I. H., Kumar C., Rahman M. M., Karim M. A.. **Shrinkage of food materials during drying: Current status and challenges**. *Comprehensive Reviews in Food Science and Food Safety* (2018) **17** 1113-1126. PMID: 33350150
20. Mandal S. K., Das A., Dey S., Sahoo U., Bose S., Bose A., Ramadan M. A.. **Bioactivities of Allicin and related organosulfur compounds from garlic: Overview of the literature since 2010**. *Egyptian Journal of Chemistry* (2019) **62** 1-11
21. Martins N., Petropoulos S., Ferreira I. C.. **Chemical composition and bioactive compounds of garlic (**. *Food Chemistry* (2016) **211** 41-50. PMID: 27283605
22. Mohammad‐Razdari A., Ebrahimi R., Zarein M., Mahfeli M., Yoosefian S. H.. **Investigation of**. *Journal of Food Process Engineering* (2021) **44** e13697. DOI: 10.1111/jfpe.13697
23. Mohsenin N. N.. (1986)
24. Noda Y., Asada C., Sasaki C., Nakamura Y.. **Effects of hydrothermal methods such as steam explosion and microwave irradiation on extraction of water soluble antioxidant materials from garlic husk**. *Waste and Biomass Valorization* (2018) **9** 1-6
25. Olanya O. M., Niemira B. A., Phillips J. G.. **Effects of gamma irradiation on the survival of Pseudomonas fluorescens inoculated on romaine lettuce and baby spinach**. *LWT‐ Food Science and Technology* (2015) **62** 55-61
26. Pérez M. B., Aveldaño M. I., Croci C. A.. **Growth inhibition by gamma rays affects lipids and fatty acids in garlic sprouts during storage**. *Postharvest Biology and Technology* (2007) **44** 122-130
27. Rambo M. K. D., Ferreira M. M. C., Melo P. M. D., Santana Junior C. C., Bertuol D. A., Rambo M. C. D.. **Prediction of quality parameters of food residues using NIR spectroscopy and PLS models based on proximate analysis**. *Food Science and Technology* (2020) **40** 444-450
28. Reiter J., Hübbers A. M., Albrecht F., Leichert L. I. O., Slusarenko A. J.. **Allicin, a natural antimicrobial defence substance from garlic, inhibits DNA gyrase activity in bacteria**. *International Journal of Medical Microbiology* (2020) **310**. PMID: 31585716
29. Sukumar A., Athmaselvi K. A.. **Optimization of process parameters for the development of finger millet based multigrain extruded snack food fortified with banana powder using RSM**. *Journal of Food Science and Technology* (2019) **56** 705-712. PMID: 30906028
30. Tazi I., Isnaini N. L., Mutmainnah M., Ainur A.. **Principal component analysis (PCA) method for classification of beef and pork aroma based on electronic nose**. *Indonesian Journal of Halal Research* (2019) **1** 5-8
31. Zhang J., Ding Z., Liang J., Cheng J., Yang S.. **Suitable duration of microwave pre‐treatment for maintaining postharvest qualities on pecan nuts during storage**. *Transactions of the Chinese Society of Agricultural Engineering* (2016) **32** 284-292
32. Zhou G. H., Xu X. L., Liu Y.. **Preservation technologies for fresh meat–A review**. *Meat Science* (2010) **86** 119-128. PMID: 20605688
|
---
title: 'Dual PTP1B/TC-PTP Inhibitors: Biological Evaluation of 3-(Hydroxymethyl)cinnoline-4(1H)-Ones'
authors:
- Kira V. Derkach
- Maxim A. Gureev
- Anastasia A. Babushkina
- Vladimir N. Mikhaylov
- Irina O. Zakharova
- Andrey A. Bakhtyukov
- Viktor N. Sorokoumov
- Alexander S. Novikov
- Mikhail Krasavin
- Alexander O. Shpakov
- Irina A. Balova
journal: International Journal of Molecular Sciences
year: 2023
pmcid: PMC10002984
doi: 10.3390/ijms24054498
license: CC BY 4.0
---
# Dual PTP1B/TC-PTP Inhibitors: Biological Evaluation of 3-(Hydroxymethyl)cinnoline-4(1H)-Ones
## Abstract
Dual inhibitors of protein phosphotyrosine phosphatase 1B (PTP1B)/T-cell protein phosphotyrosine phosphatase (TC-PTP) based on the 3-(hydroxymethyl)-4-oxo-1,4-dihydrocinnoline scaffold have been identified. Their dual affinity to both enzymes has been thoroughly corroborated by in silico modeling experiments. The compounds have been profiled in vivo for their effects on body weight and food intake in obese rats. Likewise, the effects of the compounds on glucose tolerance, insulin resistance, as well as insulin and leptin levels, have been evaluated. In addition, the effects on PTP1B, TC-PTP, and Src homology region 2 domain-containing phosphatase-1 (SHP1), as well as the insulin and leptin receptors gene expressions, have been assessed. In obese male Wistar rats, a five-day administration of all studied compounds led to a decrease in body weight and food intake, improved glucose tolerance, attenuated hyperinsulinemia, hyperleptinemia and insulin resistance, and also compensatory increased expression of the PTP1B and TC-PTP genes in the liver. The highest activity was demonstrated by 6-Chloro-3-(hydroxymethyl)cinnolin-4(1H)-one (compound 3) and 6-Bromo-3-(hydroxymethyl)cinnolin-4(1H)-one (compound 4) with mixed PTP1B/TC-PTP inhibitory activity. Taken together, these data shed light on the pharmacological implications of PTP1B/TC-PTP dual inhibition, and on the promise of using mixed PTP1B/TC-PTP inhibitors to correct metabolic disorders.
## 1. Introduction
Protein phosphotyrosine phosphatase 1B (PTP1B) is a negative regulator of metabolic pathways activated by insulin (which is produced by pancreatic beta cells) and by adipokine leptin (which is produced by adipose tissue) [1,2]. In response to insulin stimulation, PTP1B dephosphorylates the active phosphorylated forms of the insulin receptor and insulin receptor substrate proteins-1 and -2 (IRS-1, IRS-2) [3,4]. Under leptin stimulation, the enzyme dephosphorylates the leptin-activated non-receptor Janus kinase 2 (JAK2) and IRS$\frac{1}{2}$-proteins, the key components of leptin signaling [2,5]. Suppression of PTP1B phosphatase activity abolishes its inhibitory effect on insulin and leptin signaling. In obesity, type 2 diabetes mellitus (T2DM), and metabolic syndrome, the insulin and leptin signaling pathways are attenuated as a result of long-term exposure to hyperinsulinemia and hyperleptinemia; thus, the use of PTP1B inhibitors may become one of the approaches to restore them and increase tissue sensitivity to insulin and leptin [1,6]. PTP1B can be inhibited by small molecules targeting either the catalytic or the allosteric site of the enzyme. However, inhibitors aimed at the allosteric site are expected to exert more specific inhibition because the allosteric site of PTP1B, unlike its catalytic site, is more distinct from the allosteric sites of other PTP1B-related phosphatases, such as the T-cell protein phosphotyrosine phosphatase (TC-PTP) and endothelial Src homology region 2 domain-containing phosphatase-1 (SHP1) [1,6,7]. TC-PTP is also capable of attenuating the insulin and leptin signaling pathways by dephosphorylating their receptor and post-receptor components [8,9]. TC-PTP inhibitors, similarly to PTP1B inhibitors, can also improve glucose homeostasis and prevent obesity in metabolic syndrome and T2DM [10].
Previously, selective PTP1B phosphatase inhibitors have been developed [11,12,13,14], which had little effect on TC-PTP activity. The therapeutic potential of such inhibitors may not be fully justified. Inhibition of PTP1B alone can lead to undesirable toxic effects, which have been observed in previous in vivo experiments [15] and, similarly, can cause a compensatory increase in the activity of other phosphatases, primarily TC-PTP, which can functionally replace PTP1B [16,17,18,19]. Consequently, currently, there are no approved PTP1B inhibitors [20], despite the high demand for the effective treatment of the diseases associated with insulin and leptin resistance (T2DM, metabolic syndrome, and obesity). Accordingly, a paradigm shift might be needed to create such inhibitors, which would act on both PTP1B and TC-PTP, though not necessarily with high potency.
The aim of the present work is to identify small molecule inhibitors that would potentially be able to suppress the functional activity of both tyrosine phosphatases, PTP1B and TC-PTP, and thereby affect the metabolic and hormonal parameters in rats with diet-induced obesity.
As a lead structure in our quest for dual PTP1B/TC-PTP inhibitors, we relied on the earlier reported compound PI4 (ethyl 3-(hydroxymethyl)-4-oxo-1,4-dihydrocinnoline-6-carboxylate), a 4-oxo-1,4-dihydrocinnoline derivative, that demonstrated an inhibitory activity towards PTP1B, and exerted a stimulating effect on components of the insulin and leptin signaling pathways in rat hypothalamic neurons, including the serine/threonine protein kinase Akt and transcription factor STAT3 (signal transducer and activator of transcription 3) [21]. In addition, PI4 reduced the body and fat weights in diet-induced obese rats, suppressed their food intake, improved metabolic parameters, and increased their sensitivity to insulin and leptin [22]. In this study, we hypothesized that by exploring an expanded set of 3-(hydroxymethyl)-4-oxo-1,4-dihydrocinnoline analogs 1–4 (Figure 1), we might not only obtain yet another set of PTP1B inhibitors but could potentially identify dual inhibitors of PTP1B/TC-PTP, which are sought after in the context of metabolic disease treatments (vide supra).
## 2.1. Compounds 1–4
Compounds 1–4 were obtained, as described previously [23].
## 2.2. Inhibitory Effects of Compounds 1–4 towards PTP1B and TC-PTP
Compounds 1–4 displayed dose-dependent inhibition of both phosphatases, PTP1B and TC-PTP, as expressed in the IC50 values (Table 1) calculated from the respective dose-response curves (Figure 2). In each case, the range from 0.1 to 80 μM of the tested compound was studied. As can be seen from Table 1, compounds 1 and 2 showed a ~5–7-fold higher selectivity for PTP1B compared to that for TC-PTP, and in the case of compound 2, the differences were significant ($$p \leq 0.001$$). In the case of compounds 3 and 4, the IC50 values for PTP1B and TC-PTP were comparable, indicating their close selectivity for both enzymes. In the case of compound 3, the IC50 value for TC-PTP was significantly lower than the IC50 values for TC-PTP inhibition by compounds 1 and 2 ($$p \leq 0.01$$ and $p \leq 0.0001$, respectively), although they did not differ significantly from that of compound 4 ($$p \leq 0.12$$), which indicates a more pronounced TC-PTP inhibitory effect of compound 3 as compared to compounds 1 and 2. In the case of PTP1B, no significant differences in the IC50 values were found between all the studied compounds. The obtained IC50 values for compounds 1–4 were close to the IC50 values for a number of natural tyrosine phosphatase inhibitors and their semisynthetic derivatives [24,25,26,27], which may indicate a similar dose dependence to their inhibitory effect on the activity of PTP1B and TC-PTP. At the same time, compounds 1–4 had higher IC50 values in comparison with those for some recently developed synthetic phosphatase inhibitors, yet they were inferior to them in their binding affinity to the allosteric sites of these enzymes [18,19].
## 2.3. The Effects of Compounds 1–4 on the Body Weight, Food Intake, Glucose Tolerance and Insulin and Leptin Levels in the Blood of Male Rats with HFD-Induced Obesity
Based on the results of the in vitro experiments and our earlier data on the pharmacodynamics of compound PI4 [21,22], we studied the anorexigenic effect of compounds 1–4 and their influence on glucose homeostasis, as well as insulin and leptin resistance, in rats with diet-induced obesity.
The five-day treatment of obese male rats with all the studied compounds (1 and 2 at a dose of 7 mg/kg/day, and 3 and 4 at doses of 8 and 10 mg/kg/day, respectively) led to a decrease in the body weight and food consumption. This indicates a pronounced anorexigenic effect of compounds 1–4. The highest effect was demonstrated by compounds 3 and 4, which reduced the body weight in obese rats by 22.2 ± 4.5 and 22.6 ± 4.0 g, respectively, ($p \leq 0.05$ as compared to untreated obese animals) and reduced the consumption of dry standard food by $46\%$ and $39\%$, respectively (Figure 3).
Compounds 1–4 had no significant effect on glucose levels, although, of the most active compounds (3 and 4), compound 4 improved the glucose sensitivity, previously reduced in obesity, as shown by the glucose tolerance test. This was indicated by the values of the glucose levels 120 min after glucose load, which were reduced by $20\%$ and $17\%$ in the groups treated with compounds 3 and 4, respectively, in comparison with the untreated obese animals, as well as the values of AUC0-120 for the curves “glucose concentration (mM)”—time (minutes)”. The AUC0-120 values in the groups treated with compounds 3 and 4 were reduced by $24\%$ and $20\%$, respectively, as compared with the untreated animals ($p \leq 0.05$ as compared to the Ob group) (Figure 4).
The levels of Insulin and leptin were elevated in obese rats, indicating the development of insulin and leptin resistance. Measurement of the levels of insulin and leptin in the blood of animals, before and after the treatments with compounds 1–4, showed that the compounds caused a decrease in the levels of these hormones (except for the insulin level in the Ob + 2 group), which indicates an increase in the sensitivity to insulin and leptin. The inhibiting effect on insulin and leptin levels exerted by compounds 3 and 4 was more pronounced compared to compounds 1 and 2. For the purpose of normalization, insulin sensitivity was indicated by the index of insulin resistance (IR), calculated as the product of the blood concentrations of glucose and insulin. In the group treated with compound 3, the insulin resistance index was significantly reduced compared to the Ob group, yet did not differ from that in the control animals (Table 2).
## 2.4. The Effects of Compounds 1–4 on Gene Expression in the Livers of Rats with HFD-Induced Obesity
Based on the in vitro IC50 values for PTP1B and TC-PTP inhibition by compounds 1–4, and on their ability to increase insulin sensitivity and reduce hyperleptinemia, established in the in vivo experiments, we studied the effect of these compounds on the expression of the genes encoding PTP1B and TC-PTP, as well as for insulin and leptin receptors, in the livers of obese rats. Along with this, we studied the expression of cytoplasmic tyrosine phosphatase SHP1, the catalytic site of which differs from those of PTP1B and TC-PTP [28].
The treatments with all the studied compounds led to an overall increase in the expression of the PTP1B and TCPTP genes in the liver. However, a significant difference to the Ob group was shown only for the groups treated with compounds 3 and 4, and, in the case of the TCPTP gene, for the group treated with compound 1. Compounds 3 and 4 did not affect the expression of the *Shp1* gene, which encodes the protein tyrosine phosphatase SHP1, while compounds 1 and 2 increased the expression of the *Shp1* gene, and the differences from the control were significant. These data suggest that the unsubstituted and fluoro-substituted analogs, 1 and 2, are likely to also inhibit phosphatase SHP1.
Compounds 3 and 4 caused a slight decrease in the expression of the genes encoding the insulin and leptin receptors, but the difference with the Ob group was not significant. This tendency, albeit rather weak, could be due to an increase in the sensitivity of hepatocytes to insulin and leptin (Figure 5).
## 2.5. Docking Simulation of Compounds 1–4 as PTP1B/TC-PTP Inhibitors
The evaluation of the specific biological activity of compounds 1–4 showed their capability to interact with PTP1B and TC-PTP. Based on the gene expression data of phosphatases in the liver, there are reasons to believe that compounds 1 and 2 are also potentially able to interact with the SHP1 phosphatase, which has a catalytic site different from that of other closely related phosphatases, PTP1B and TC-PTP. It is necessary to establish the causes of this phenomenon to reduce the spectrum of side interactions.
As a first step, we studied the active sites of the phosphatases PTP1B, TC-PTP, and SHP1. The amino acids that form the catalytic sites of the enzymes are shown in Figure 6 and are highlighted in blue.
As can be seen from Figure 6, some identical amino acid residues are not included in the catalytic site formation.
The main reason is the difference in the protein loop conformations and the presence of secondary structure elements. The TC-PTP enzyme is less structured in the domain between Arg114 and Cys123 (unstructured loop). This domain in PTP1B (Arg114–Cys123) and SHP-1 (Arg352–Cys361) is more rigid due to the β-sheet secondary structure element presence (Figure 7A). An unstructured loop in the case of TC-PTP can be useful for studies into ligand-induced conformational changes. The second domain, very different in the studied phosphatases, is located between the following residues: Thr180–Pro187 (PTP1b), Thr179–Pro186 (TC-PTP), and Ser416–Pro423 (SHP-1) (Figure 7B).
Primarily, in the PTP1b and TC-PTP structures, this domain is identical by sequence (TWPDFGVP), but in the SHP-1 phosphatase, it is different (SWPDHGVP). The main ligand-interacting sequences «WPD» and «GVP» are conserved, although the phenylalanine residue is changed to histidine and the threonine residue is replaced by serine. The substitution of phenylalanine for histidine provides a decrease in the efficiency of the hydrophobic interactions and an increase in the efficiency of polar interactions (also pH-dependent), in this part of the active site of the enzymes.
If we compare the geometry of the TC-PTP and PTP1B loops, we will identify significant differences in their positions. However, their sequences remain the same. The key differences, here, are hidden in the different rotamer states of the Phe–Asp pair. Another difference lies in the amino acid properties forming the active site cavity. Here, we can observe the difference in the interacting amino acids profile: SHP1 is less hydrophobic and more polar with the addition of a histidine residue. The homology degree in these segments of PTP1B and TC-PTP is also much higher (Figure 8).
Molecular docking results and the binding free energies of compounds 1–4 are shown in Table 3. Studied compounds bind stably within the active site of all the observed phosphatases. With regard to PTP1B, compounds 1–4 can be considered site-specific ligands, because the ligand efficiency (LE) value (Table 3) is superior to that of the structure of the reference compound. In contrast to the used reference compound, the active inhibitor co-crystallized with PTP1B (pdb model 1Q1M, ligand structure, Figure 9). At the same time, a binding mode for the studied structures 1–4 also reproduces interactions specific to the reference compound (see Figure 9, PTP1B inhibitor showed in PDB model 1Q1M).
When switching to the alternative targets of TC-PTP and SHP1, we observe a decrease in the active site specificity. At the same time, in the case of SHP1, the compounds also change the binding area (which remains within the active site). Compounds 1 and 2 retain high levels of site-specificity towards SHP1 (Table 3—highlighted by orange). Conversely, compounds 3 and 4 show a potential capability to selectively interact with PTP1B and TC-PTP (Table 3—highlighted by green). More clearly, the specificity of the observed targets is corroborated by the free energy value (ΔG). Such significant differences in comparison to the docking results, in theory, may indicate a significant solvent role in the binding process (the MM-GBSA method considers this, implicitly). The binding selectivity to PTP1B and TC-PTP for compounds 3 and 4 agrees with our experimental data.
Ligand interaction diagrams analysis, presented in Figure 9, shows that compounds 1–4 interact mainly within the amino acids: Asp181, Arg221, Phe182, Cys215, and Ala217, in the same manner as the reference compound. Thus, compounds 1–4 mimic the key lipophilic and polar contacts, from which PTP1B site-specificity is achieved.
The diagrams of the ligand–protein (enzyme) interactions with TC-PTP were similarly analyzed.
Studying the protein–ligand interactions of compounds 1–4 with TC-PTP showed that the interaction profile for these compounds is almost identical. A distinctive feature is the weakening of the network of lipophilic ligand–enzyme contacts in the cavity of the active site. Moreover, unlike PTP1B, a pi–cation interaction with Arg222 is realized (Figure 10). In the TC-PTP structure, it is more accessible for any subsequent interaction.
Ligand interaction diagrams of compounds 1–4 with SHP1 (Figure 11) showed that these compounds are more than capable of interacting with the enzyme, to form a stable protein–ligand complex. However, the binding region is strongly shifted towards Tyr276, relative to the binding site. Both PTP1B and TC-PTP contain the residues Tyr46 and Tyr48, which have a similar arrangement in each phosphatase molecule. In regard to the reference structure of the PTP1B ligand (Figure 9), then, Tyr46 interacts with the lipophilic fluorophenyl part of the molecule, sometimes forming a pi–stacking interaction. Compounds 3 and 4 are distinguished by a lower intensity of lipophilic contacts with SHP1 and by the absence of the pi–stacking interaction with Tyr276 (Figure 11).
## 3. Discussion
Our in vitro data indicate that all 4-oxo-1,4-dihydrocinnoline derivatives 1–4, which are structural analogs of the previously studied compound PI4 [21,22], are inhibitors of the PTP1B and TC-PTP tyrosine phosphatases. At the same time, they differ in selectivity for these tyrosine phosphatases since compounds 1 and 2 were found to be more selective towards PTP1B, while compounds 3 and 4 had no significant differences in selectivity for either phosphatase. In addition, compound 3 was far more effective as an inhibitor of TC-PTP than compounds 1 and 2 (Table 1). These observations are supported by the molecular docking results.
As noted above, unlike compounds 1 and 2, compounds 3 and 4, according to the results of the in vitro experiments, are similar in their ability to inhibit the PTP1B and TC-PTP phosphatases. Notably, the inhibitory effect for PTP1B, as judged by the IC50 values, did not fully correlate with the performance of the compounds in the in vivo experiments. Thus, the IC50 value for compound 1 was the lowest among all the derivatives studied and was 2.5 times inferior to that for compound 4. At the same time, the anorexigenic effect of compound 1 was less pronounced compared to compounds 3 and 4. Importantly, the effects of compounds 3 and 4 on the food intake and metabolic parameters in obese rats are similar, while, according to the in vitro experiments, the effectiveness of compound 3 as an inhibitor of TC-PTP is more pronounced. It can be assumed that the similar affinity of compounds 3 and 4, with respect to both phosphatases PTP1B and TC-PTP, is important for the metabolic effects of these inhibitors. This distinguishes compounds 3 and 4 from compounds 1 and 2, which are more selective for PTP1B.
The PTP1B and TC-PTP phosphatases, being negative regulators of insulin and leptin signaling, are involved in the development of insulin and leptin resistance and mediate an increase in appetite, and the accumulation of excess adipose tissue in metabolic disorders [1,6,8,9,10]. At the same time, there are a number of common downstream targets for PTP1B and TCPTP, which makes them at least partly interchangeable. For instance, both phosphatases dephosphorylate the hormone-activated phosphorylated forms of the leptin and insulin receptors, as well as the non-receptor-associated JAK2-tyrosine kinase associated with the leptin receptor [6,10]. However, there are also significant differences in the intracellular targets of the PTP1B and TCPTP phosphatases. The PTP1B dephosphorylates the IRS$\frac{1}{2}$ proteins that couple the insulin receptor to downstream SH2 domain-containing proteins [3,6], while the TCPTP dephosphorylates the STAT3 transcription factor, which is activated via the leptin receptor and controls the expression of STAT3-dependent genes [10]. The fact that compound 3, which is the most active with respect to TC-PTP, significantly reduces leptin levels is likely due to the fact that phosphatase TC-PTP is even more involved in the regulation of leptin signaling than in the regulation of insulin signaling [10].
Our results on the high efficiency of the dual inhibitors of the PTP1B and TCPTP phosphatases are supported by numerous studies, whereby a pronounced anorexigenic effect of low-selective inhibitors of PTP1B and TC-PTP was identified, although this phenomenon remains poorly understood. Celastrol, a naturally occurring pentacyclic triterpene, when administered to mice, reduced the activity of both phosphatases in the hypothalamic arcuate nucleus, through an allosteric mechanism, significantly reducing their food intake and body weights. The anorexigenic effects of Celastrol are mainly due to the activation of leptin signaling in hypothalamic neurons [29]. Simultaneous knockout of the PTP1B and TC-PTP genes in the hypothalamus of obese mice, as well as the combined administration of the PTP1B inhibitor and the glucocorticoid hormone antagonist RU486, which attenuates TC-PTP expression, suppressed food intake, normalized body weight and adipose tissue, improved glucose tolerance, alongside insulin and leptin sensitivity [9]. At the same time, inhibiting the two phosphatases separately was significantly less effective. Thus, simultaneous, and similarly effective inhibition of both PTP1B and TC-PTP, which we have shown for 4-oxo-1,4-dihydrocinnolines 1–4, primarily for compounds 3 and 4, does not allow for the compensatory switching of the mechanisms, from one phosphatase to another, in the inhibition of leptin and insulin signaling. However, one cannot exclude the involvement in these compensatory mechanisms of additional phosphatases, which are less specific in their targeting of the insulin and leptin receptors and their downstream signaling proteins.
Pharmacological or genetic suppression of PTP1B and TC-PTP activity can trigger a number of compensatory mechanisms, which weaken the effects of the inhibitors of these enzymes. Among them, are changes in gene expression of both phosphatases and the components of the insulin and leptin signaling cascade. In the liver, we studied the expression of the genes that encode PTP1B, TC-PTP, and SHP1, as well as genes encoding insulin and leptin receptors. It was shown that in the liver of rats treated with compounds 3 and 4, the expressions of the PTP1B and TC-PTP genes were significantly increased, while the expression of the *Shp1* gene did not change. In the case of compounds 1 and 2, which are more specific to PTP1B, there was a trend towards an increase in the PTP1B and TC-PTP gene expressions, yet there was a significant increase in the *Shp1* gene expression. An increase in the expression of the SHP1 phosphatase, in obese rats with impaired glucose tolerance, is consistent with its negative role in the regulation of feeding behaviors and glucose homeostasis [30]. Thus, it was found that in mice with diet-induced obesity, the expression of SHP1 is increased [31], and inhibition or knockout of this enzyme prevents the development of metabolic disorders [30,32]. Thus, there is reason to believe that this phosphatase may be involved in the mechanisms through which insulin and leptin signaling is weakened. Our data on the expression of the SHP1 gene, however, seem somewhat unexpected and require further study. Based on these data, it can be concluded that inhibitors with comparable selectivity towards the PTP1B and TC-PTP phosphatases (compounds 3 and 4) have little effect on the expression of SHP1, while inhibitors predominantly selective for PTP1B (compounds 1 and 2), increase it, and this effect does not depend on the IC50 values.
## 4.1. Compound Synthesis
Compounds 1–4 were synthesized, as described previously [23].
## 4.2.1. The Phosphatases Activity Assay and the Determination of IC50 for Tested Compounds
The measurement of the activity of the phosphatases PTP1B and TC-PTP and their inhibition was carried out using 6,8-difluoro-4methylumbelliferyl phosphate (DiFMUP), as previously described [33]. The stock solution of recombinant human PTP1B protein (#ab51277, Abcam, Cambridge, UK), at a concentration of 1 µg/µL, was prepared in 25 mM Tris–HCl (pH 7.5), $20\%$ glycerol, 2 mM β-mercaptoethanol, 1 mm EDTA, and 1 mM DTT, and stored at −20 °C. Active human recombinant TC-PTP protein was purchased from Sigma-Aldrich (#SRP0218, Saint Louis, MO, USA) as the aqueous buffer solution at a concentration of 2.9 mg/mL, and aliquots were stored at −80 °C. The fluorogenic substrate DiFMUP (#D6567, Molecular Probes, Thermo Fisher Scientific, Waltham, MA, USA) was dissolved at a concentration of 10 mM in N,N-dimethylformamide and stored in aliquots at −20 °C. The 6,8-difluoro-7-hydroxy-4-methylcoumarin (DiFMU, #D6566, Molecular Probes, Thermo Fisher Scientific, Waltham, MA, USA) was used as a reference fluorescent standard. For the IC50 determination, the assay was carried out in black flat bottom 96-well plates using a reaction volume of 100 µL. The phosphatases (PTP1B or TC-PTP) were preincubated with the tested compounds (0.1–80 µM) in 50 mM HEPES (pH 6.9), 100 mM NaCl, 1 mM EDTA, 2 mM DTT, 0.1 mg/mL BSA for 5 min at 37 °C. The following concentrations of the compounds were studied: 0.1, 0.2, 0.5, 1, 2, 5, 10, 20, 40, and 80 μM. The reactions were initiated by the addition of the fluorogenic substrate DiFMUP and diluted in the assay buffer. Progress curves for hydrolysis reaction were obtained with 25 µM of DiFMUP for 80 ng/mL PTP1B, and 35 µM of DiFMUP for 80 ng/mL TC-PTP. Fluorescence excitation of hydrolyzed DiFMUP and the fluorescent standard DiFMU was monitored at 355 nm and emission was detected at 460 nm for 6–10 min at 30 s intervals in a Fluoroskan Ascent FL microplate reader (Thermo Electron Corporation, Vantaa, Finland). Initial velocities of the reactions were used to calculate the IC50 values using GraphPad Prism.
## 4.2.2. Animals and the Induction of Obesity
The male Wistar rats (at the start of the experiment the age of the animals was 2-months, and at the end of the experiment their age was 6-months) were obtained from the Rappolovo animal facility (Russia). The animals were housed in plastic cages, five animals in each, with a normal light–dark cycle (12 h/12 h), a temperature of 22 ± 3 °C, and free access to food and water. All experiments were approved by the Institutional Animal Care and Use Committee at the Sechenov Institute of Evolutionary Physiology and Biochemistry (St. Petersburg, Russia) (protocol No $\frac{02}{02}$-2020, 19 February 2020) and according to The Guide for the Care and Use of Laboratory Animals and the European Communities Council Directive of 1986 ($\frac{86}{609}$/EEC). All efforts were made to minimize animal suffering and reduce the number of experimental animals.
The obesity model was induced at 16-weeks (starting at two months of age) using a high-fat diet. Control animals received standard laboratory chow pellets. The high-fat diet included supplements of 5–7 g of a fat mixture containing $52.4\%$ pork lard, $41.7\%$ curd, $5\%$ liver, $0.5\%$ L-methionine, $0.2\%$ baker’s yeast, and $0.2\%$ sodium chloride [34]. After 16-weeks on a high-fat diet, the animals with increased body weight, glucose intolerance, and hyperinsulinemia, and hyperleptinemia were selected for further experiments. The glucose intolerance was estimated according to the results of the glucose tolerance test (GTT). In the GTT, 120 min after glucose loading, the blood glucose levels in obese rats were above 7 mM, and the AUC0-120 values for the curve “glucose concentration (mM)–time (minutes)” were above $30\%$, as compared to the average AUC0-120 values in the control group.
Further, six groups were formed: control (Con, $$n = 10$$), obesity (Ob, $$n = 10$$), obese rats with five-day treatment with compounds 1 (Ob + 1, $$n = 5$$), 2 (Ob + 2, $$n = 5$$), 3 (Ob + 3, $$n = 10$$), and 4 (Ob + 4, $$n = 10$$). All compounds were administered in DMSO (300 µL) at a daily dose of 7 mg/kg (1 and 2), 8 mg/kg [3], and 10 mg/kg [4] (i.p.), based on the results of the preliminary experiments. The used doses of compounds 1–4, in terms of the number of moles of each compound per kg of animal body weight, were equivalent. Control and obese rats received DMSO in the same volumes (300 µL), instead of the tested compounds. During the five days of the experiment, all animals were transferred to standard food and had free access to food and drinking water. Five animals each from the control group and the groups with obesity and treated with the most active compounds 3 and 4 were selected for the study of glucose tolerance using the GTT. The test was carried out on the morning of the next day after the last (fifth) injection of the compounds or DMSO, after a 10 h fast. The rest of the animals ($$n = 5$$ in each group) were anesthetized and decapitated on the last day of the experiment. The blood and liver samples were taken to measure blood glucose and hormone levels and gene expression in the liver.
## 4.2.3. The Determination of Blood Glucose and Hormones Levels and GTT
The glucose levels in the blood obtained from the tail vein were measured using a glucometer (Life Scan Johnson & Johnson, Denmark) and the test strips “One Touch Ultra” (USA). The levels of insulin and leptin in rat serum were estimated with the “Rat Insulin ELISA” (Mercodia AB, Uppsala, Sweden) and “ELISA for Leptin, Rat” (Cloud-Clone Corporation, Houston, TX, USA) kits. The GTT was carried out using a single injection of glucose (2 g/kg, i.p.) after 10 h of fasting, as described earlier [34]. The blood glucose levels were measured before (0 min) and 15, 30, 60, and 120 min after the glucose load.
## 4.2.4. The Determination of Gene Expression in the Liver of Rats
The total RNA was isolated from the liver samples of rats using the “ExtractRNA Reagent” (Evrogen, Moscow, Russia), and the samples containing 1 μg of RNA were transcribed to cDNA using the random oligodeoxynucleotide primers and the “MMLV RT kit” (Evrogen, Russia). The amplification procedure was carried out in the mixture containing 10 ng of reverse transcribed product, 0.4 μM of the forward and reverse primers, and the “qPCRmix-HS SYBR + LowROX kit” (Evrogen, Russia). The amplified signals were detected using the “Applied Biosystems® 7500 Real-Time PCR System” (Life Technologies, Carlsbad, CA, USA, Thermo Fisher Scientific Inc., USA). The primers that were used to assess the expression of genes encoding the phosphatase PTP1B, TC-PTP, and SHP1 and the insulin and leptin receptors are presented in Table 4. The obtained data were calculated with the delta–delta Ct method and expressed as a fold expression, relative to the corresponding control [35]. The expression of the gene encoding 18S RNA was used as an endogenous control.
## 4.2.5. Statistical Analysis of Biological Experiments
The data on food intake, body weight, and biochemical parameters in rats, as well as the PCR data, were analyzed using the IBM SPSS Statistics 22 software (“IBM”, Armonk, NY, USA), and the results are presented as mean ± standard error of the mean (M ± SEM). All differences are considered significant at $p \leq 0.05.$ The calculation of the IC50 values of the studied inhibitors for the initial velocities of enzymatic reactions (PTP1B, TC-PTP) was carried out using the nonlinear regression analysis with GraphPad Prism 8 (“GraphPad Software, Inc.”, Boston, MA, USA). The statistical analysis was carried out using the Wilcoxon test for pairwise comparison and Dunn’s test for multiple comparisons.
## 4.3.1. Used Enzyme Models
Observed protein structures were taken from the RCSB Protein Data Bank database [36]. PDB IDs: 1Q1M (PTP1b), 1L8K (TC-PTP), 4GRZ (SHP1).
## 4.3.2. Protein and Ligand Structure Preparation
All proteins (enzymes) were preprocessed before the calculations using the protein prep wizard tool from Schrodinger suite 2021-4 [37]. During preprocessing, the following errors were fixed: missing amino acid sidechains, incorrect protonation states, missing hydrogens, incorrect bond orders, bond angles, bond length, and torsion angles. Solvent molecules were removed from all protein structure models.
The three-dimensional structure of the compounds was generated using the LigPrep module with the computation of possible ionization states, tautomers, and stereoisomers at the physiological pH (7.4) All manipulations with the protein structures and small molecules were performed in the OPLS4 forcefield.
## 4.3.3. Molecular Docking and MM-GBSA
All protein models (PTP1B, TC-PTP, and SHP1) were superimposed with the use of the protein to PTP1B structure, using the protein alignment tool (PTP1B was used as the reference). The active site was defined by positioning the reference ligand, which was present in the PTP1B model (PDB id: 1Q1M). The grid box was placed in accordance with the centroid of the workspace ligand structure. The grid box side size was 12 Å (in accordance with ligand size and the addition of the non-bonded interactions distance). The scaling factor was 1.0; the partial charge cutoff was 0.25, without the excluded volumes.
The reference structure and the other studied compounds were docked in the PTP1B, TC-PTP, and SHP1 active sites. The Glide program [38] included in Schrodinger Suite was used for the docking. For each ligand, 15 docking solutions in standard precision (SP) mode were generated, without the use of constraints. Best-fitting binding pose was selected by comparing ligand–protein interactions with those for the reference ligand (see Figure 9). For each docked structure a ligand interaction diagram was generated, describing the protein–ligand contacts and types.
Gibbs free energy was calculated with the use of the MM–GBSA method [39], including the implicit solvent model. This method considers the influence of the solvent and analyzes the free energy components, such as the energy increments of the strained contacts, solvation energy, and the ligand-induced conformational changes in protein amino acids surrounding the active site that interact with the ligand.
For calculations, the best protein–ligand complexes obtained by docking were used. The VSGB solvation model was used along with the OPLS4 forcefield. Protein flexibility was allowed at a distance of 6 Å from the ligand. The calculations were performed with the use of the Prime module [40] from Schrodinger Suite 2021-4.
## 5. Conclusions
We have studied the inhibitory effects of a small series of 4-oxo-1,4-dihydrocinnolines 1–4 in vitro, which exhibited dual inhibitory profiles towards two phosphatases, PTP1B and TC-PTP, which are both involved in insulin and leptin signaling. While two compounds (1 and 2) demonstrated different selectivity for the studied tyrosine phosphatases, PTP1B and TC-PTP, the selectivity of compounds 3 and 4 for these phosphatases was comparable. The affinity of the compounds to both phosphatases was corroborated by in silico modeling experiments. In obese rats, compounds 3 and 4 restored metabolic and hormonal parameters to a greater extent than compounds 1 and 2, which were more selective for PTP1B than TC-PTP. Thus, the development of phosphatase inhibitors with similar selectivity for PTP1B and TC-PTP may become a promising direction for the development of drugs for the correction of hyperphagia and obesity and the treatment of metabolic disorders, such as type 2 diabetes mellitus and metabolic syndromes.
## References
1. Krishnan N., Koveal D., Miller D.H., Xue B., Akshinthala S.D., Kragelj J., Jensen M.R., Gauss C.M., Page R., Blackledge M.. **Targeting the disordered C terminus of PTP1B with an allosteric inhibitor**. *Nat. Chem. Biol.* (2014) **10** 558-566. DOI: 10.1038/nchembio.1528
2. Villamar-Cruz O., Loza-Mejía M.A., Arias-Romero L.E., Camacho-Arroyo I.. **Recent advances in PTP1B signaling in metabolism and cancer**. *Biosci. Rep.* (2021) **41** BSR20211994. DOI: 10.1042/BSR20211994
3. Yip S.C., Saha S., Chernoff J.. **PTP1B: A double agent in metabolism and oncogenesis**. *Trends Biochem. Sci.* (2010) **35** 442-449. DOI: 10.1016/j.tibs.2010.03.004
4. Sorokoumov V.N., Shpakov A.O.. **Protein phosphotyrosine phosphatase 1B: Structure, function, role in the development of metabolic disorders and their correction by the enzyme inhibitors**. *J. Evol. Biochem. Physiol.* (2017) **3** 259-270. DOI: 10.1134/S0022093017040020
5. Köhn M.. **Turn and Face the Strange: A New View on Phosphatases**. *ACS Cent. Sci.* (2020) **6** 467-477. DOI: 10.1021/acscentsci.9b00909
6. Liu R., Mathieu C., Berthelet J., Zhang W., Dupret J.M., Rodrigues L.F.. **Human Protein Tyrosine Phosphatase 1B (PTP1B): From Structure to Clinical Inhibitor Perspectives**. *Int. J. Mol. Sci.* (2022) **23**. DOI: 10.3390/ijms23137027
7. Zhang Z.Y., Dodd G.T., Tiganis T.. **Protein Tyrosine Phosphatases in Hypothalamic Insulin and Leptin Signaling**. *Trends Pharmacol. Sci.* (2015) **36** 661-674. DOI: 10.1016/j.tips.2015.07.003
8. Loh K., Fukushima A., Zhang X., Galic S., Briggs D., Enriori P.J., Simonds S., Wiede F., Tiganis T.. **Elevated hypothalamic TCPTP in obesity contributes to cellular leptin resistance**. *Cell Metab.* (2011) **14** 684-699. DOI: 10.1016/j.cmet.2011.09.011
9. Dodd G.T., Xirouchaki C.E., Eramo M., Mitchell C.A., Andrews Z.B., Henry B.A., Cowley M.A., Tiganis T.. **Intranasal Targeting of Hypothalamic PTP1B and TCPTP Reinstates Leptin and Insulin Sensitivity and Promotes Weight Loss in Obesity**. *Cell Rep.* (2019) **28** 2905-2922. DOI: 10.1016/j.celrep.2019.08.019
10. Wang Y.N., Liu S., Jia T., Feng Y., Xu X., Zhang D.. **T Cell Protein Tyrosine Phosphatase in Glucose Metabolism**. *Front. Cell Dev. Biol.* (2021) **9** 682947. DOI: 10.3389/fcell.2021.682947
11. Liu P., Du Y., Song L., Shen J., Li Q.. **Discovery of novel, high potent, ABC type PTP1B inhibitors with TCPTP selectivity and cellular activity**. *Eur. J. Med. Chem.* (2016) **118** 27-33. DOI: 10.1016/j.ejmech.2016.04.014
12. Chen X., Gan Q., Feng C., Liu X., Zhang Q.. **Investigation of selective binding of inhibitors to PTP1B and TCPTP by accelerated molecular dynamics simulations**. *J. Biomol. Struct. Dyn.* (2019) **37** 3697-3706. DOI: 10.1080/07391102.2018.1526117
13. Jin J., Ye X., Boateng D., Dai K., Ye F., Du P., Yu H.. **Identification and characterization of potent and selective inhibitors targeting protein tyrosine phosphatase 1B (PTP1B)**. *Bioorg. Med. Chem. Lett.* (2019) **29** 2358-2363. DOI: 10.1016/j.bmcl.2019.06.011
14. Rocha S., Lucas M., Silva V.L.M., Gomes P.M.O., Silva A.M.S., Araújo A.N., Aniceto N., Guedes R.C., Corvo M.L., Fernandes E.. **Pyrazoles as novel protein tyrosine phosphatase 1B (PTP1B) inhibitors: An in vitro and in silico study**. *Int. J. Biol. Macromol.* (2021) **181** 1171-1182. DOI: 10.1016/j.ijbiomac.2021.04.061
15. Brown A.P., Saravanan C., Devine P., Magnifico M., Gao J., Beaulieu V., Ma F., Yasoshima K., Barnes-Seeman D., Yamada K.. **Correlation Between Nasal Epithelial Injury and In Vitro Cytotoxicity Using a Series of Small Molecule Protein Tyrosine Phosphatase 1B Inhibitors Investigated for Reversal of Leptin Resistance in Obesity**. *Int. J. Toxicol.* (2017) **36** 303-313. DOI: 10.1177/1091581817711877
16. Sharma B., Xie L., Yang F., Wang W., Zhou Q., Xiang M., Zhou S., Lv W., Jia Y., Pokhrel L.. **Recent advance on PTP1B inhibitors and their biomedical applications**. *Eur. J. Med. Chem.* (2020) **199** 17376. DOI: 10.1016/j.ejmech.2020.112376
17. Li X., Wang L., Shi D.. **The design strategy of selective PTP1B inhibitors over TCPTP**. *Bioorg. Med. Chem.* (2016) **24** 3343-3352. DOI: 10.1016/j.bmc.2016.06.035
18. Li X., Niu M., Wang A., Lu L., Englert U., Feng S., Zhang L., Yuan C.. **Synthesis, structure and in vitro biological properties of a new copper(II) complex with 4-{[3-(pyridin-2-yl)-1H-pyrazol-1-yl]methyl}benzoic acid**. *Acta Crystallogr. C Struct. Chem.* (2021) **77** 641-648. DOI: 10.1107/S2053229621009748
19. Ma Y., Ding T.T., Liu Y.Y., Zheng Z.H., Sun S.X., Zhang L.S., Zhang H., Lu X.H., Cheng X.C., Wang R.L.. **Design, synthesis, biological evaluation and molecular dynamics simulation studies of imidazolidine-2,4-dione derivatives as novel PTP1B inhibitors**. *Biochem. Biophys. Res. Commun.* (2021) **579** 40-46. DOI: 10.1016/j.bbrc.2021.09.047
20. Campos-Almazán M.I., Hernández-Campos A., Castillo R., Sierra-Campos E., Valdez-Solana M., Avitia-Domínguez C., Téllez-Valencia A.. **Computational Methods in Cooperation with Experimental Approaches to Design Protein Tyrosine Phosphatase 1B Inhibitors in Type 2 Diabetes Drug Design: A Review of the Achievements of This Century**. *Pharmaceuticals* (2022) **15**. DOI: 10.3390/ph15070866
21. Zakharova I.O., Sorokoumov V.N., Bayunova L.V., Derkach K.V., Shpakov A.O.. **4-Oxo-1,4-dihydrocinnoline derivative with phosphatase 1B inhibitor activity enhances leptin signal transduction in hypothalamic neurons**. *J. Evol. Biochem. Physiol.* (2018) **54** 273-280. DOI: 10.1134/S0022093018040038
22. Derkach K.V., Sorokoumov. V. N., Bakhtyukov A.A., Bondareva V.M., Shpakov A.O.. **Insulin and leptin levels in blood and brain structures of rats with diet-induced obesity and the effect of various drugs on them**. *J. Evol. Biochem. Physiol.* (2022) **58** 1892-1904. DOI: 10.1134/S0022093022060187
23. Babushkina A.A., Mikhaylov V.N., Novikov A.S., Sorokoumov V.N., Gureev M.A., Kryukova M.A., Shpakov A.O., Balova I.A.. **Synthesis, X-ray and DFT Studies of 6-halo-3-(hydroxymethyl)cinnolin-4(1**. *Chem. Heterocycl. Comp.* (2022) **58** 432-437. DOI: 10.1007/s10593-022-03109-3
24. De-La-Cruz-Martínez L., Duran-Becerra C., González-Andrade M., Páez-Franco J.C., Germán-Acacio J.M., Espinosa-Chávez J., Torres-Valencia J.M., Pérez-Villanueva J., Palacios-Espinosa J.F., Soria-Arteche O.. **Indole-and pyrazole-glycyrrhetinic acid derivatives as ptp1b inhibitors: Synthesis, in vitro and in silico studies**. *Molecules* (2021) **26**. DOI: 10.3390/molecules26144375
25. Martínez-Aldino I.Y., Villaseca-Murillo M., Morales-Jiménez J., Rivera-Chávez J.. **Absolute configuration and protein tyrosine phosphatase 1B inhibitory activity of xanthoepocin, a dimeric naphtopyrone from Penicillium sp. IQ-429**. *Bioorg. Chem.* (2021) **115** 105166. DOI: 10.1016/j.bioorg.2021.105166
26. Wu J.M., Zhou Q.Q., Xie X.Y., Xu J.B.. **Khayalactone- and phragmalin-type limonoids with PTP1B inhibitory activity from Trichilia sinensis Bentv**. *Fitoterapia* (2021) **154** 105025. DOI: 10.1016/j.fitote.2021.105025
27. Yamazaki H., Tsuge H., Takahashi O., Uchida R.. **Germacrane sesquiterpenes from leaves of Eupatorium chinense inhibit protein tyrosine phosphatase**. *Bioorg. Med. Chem. Lett.* (2021) **53** 128422. DOI: 10.1016/j.bmcl.2021.128422
28. Sharma Y., Bashir S., Bhardwaj P., Ahmad A., Khan F.. **Protein tyrosine phosphatase SHP-1: Resurgence as new drug target for human autoimmune disorders**. *Immunol. Res.* (2016) **64** 804-819. DOI: 10.1007/s12026-016-8805-y
29. Kyriakou E., Schmidt S., Dodd G.T., Pfuhlmann K., Simonds S.E., Lenhart D., Geerlof A., Schriever S.C., De Angelis M., Schramm K.W.. **Celastrol Promotes Weight Loss in Diet-Induced Obesity by Inhibiting the Protein Tyrosine Phosphatases PTP1B and TCPTP in the Hypothalamus**. *J. Med. Chem.* (2018) **61** 11144-11157. DOI: 10.1021/acs.jmedchem.8b01224
30. Sharma Y., Ahmad A., Yavvari P.S., Kumar M.S., Bajaj A., Khan F.. **Targeted SHP-1 Silencing Modulates the Macrophage Phenotype, Leading to Metabolic Improvement in Dietary Obese Mice**. *Mol. Ther. Nucleic Acids.* (2019) **16** 626-636. DOI: 10.1016/j.omtn.2019.04.020
31. Xu E., Charbonneau A., Rolland Y., Bellmann K., Pao L., Siminovitch K.A., Neel B.G., Beauchemin N., Marette A.. **Hepatocyte-specific Ptpn6 deletion protects from obesity-linked hepatic insulin resistance**. *Diabetes* (2012) **61** 1949-1958. DOI: 10.2337/db11-1502
32. Krüger J., Wellnhofer E., Meyborg H., Stawowy P., Östman A., Kintscher U., Kappert K.. **Inhibition of Src homology 2 domain-containing phosphatase 1 increases insulin sensitivity in high-fat diet-induced insulin-resistant mice**. *FEBS Open Bio.* (2016) **6** 179-189. DOI: 10.1002/2211-5463.12000
33. Welte S., Baringhaus K.H., Schmider W., Müller G., Petry S., Tennagels N.. **6,8-Difluoro-4-methylumbiliferyl phosphate: A fluorogenic substrate for protein tyrosine phosphatases**. *Anal. Biochem.* (2005) **338** 32-38. DOI: 10.1016/j.ab.2004.11.047
34. Derkach K.V., Bondareva V.M., Chistyakova O.V., Berstein L.M., Shpakov A.O.. **The effect of long-term intranasal serotonin treatment on metabolic parameters and hormonal signaling in rats with high-fat diet/low-dose streptozotocin-induced type 2 diabetes**. *Int. J. Endocrinol.* (2015) **2015** 245459. DOI: 10.1155/2015/245459
35. Schmittgen T.D., Livak K.J.. **Analyzing real-time PCR data by the comparative CT method**. *Nat. Protoc.* (2008) **3** 1101-1108. DOI: 10.1038/nprot.2008.73
36. Berman H.M.. **The Protein Data Bank**. *Nucleic Acids Res.* (2000) **28** 235-242. DOI: 10.1093/nar/28.1.235
37. Madhavi Sastry G., Adzhigirey M., Day T., Annabhimoju R., Sherman W.. **Protein and ligand preparation: Parameters, protocols, and influence on virtual screening enrichments**. *J. Comput. Aided. Mol. Des.* (2013) **27** 221-234. DOI: 10.1007/s10822-013-9644-8
38. Friesner R.A., Banks J.L., Murphy R.B., Halgren T.A., Klicic J.J., Mainz D.T., Repasky M.P., Knoll E.H., Shelley M., Perry J.K.. **Glide: A New Approach for Rapid, Accurate Docking and Scoring. 1. Method and Assessment of Docking Accuracy**. *J. Med. Chem.* (2004) **47** 1739-1749. DOI: 10.1021/jm0306430
39. Wang E., Sun H., Wang J., Wang Z., Liu H., Zhang J.Z.H., Hou T.. **End-Point Binding Free Energy Calculation with MM/PBSA and MM/GBSA: Strategies and Applications in Drug Design**. *Chem. Rev.* (2019) **119** 9478-9508. DOI: 10.1021/acs.chemrev.9b00055
40. Sherman W., Day T., Jacobson M.P., Friesner R.A., Farid R.. **Novel procedure for modeling ligand/receptor induced fit effects**. *J. Med. Chem.* (2006) **49** 534-553. DOI: 10.1021/jm050540c
|
---
title: Analyzing cross‐talk of EPO and EGF genes along with evaluating therapeutic
potential of Cinnamomum verum in cigarette‐smoke‐induced lung pathophysiology in
rat model
authors:
- Haseeb Anwar
- Soha Navaid
- Humaira Muzaffar
- Ghulam Hussain
- Muhammad Naeem Faisal
- Muhammad Umar Ijaz
- Sanel Riđanović
journal: Food Science & Nutrition
year: 2023
pmcid: PMC10002988
doi: 10.1002/fsn3.3188
license: CC BY 4.0
---
# Analyzing cross‐talk of EPO and EGF genes along with evaluating therapeutic potential of Cinnamomum verum in cigarette‐smoke‐induced lung pathophysiology in rat model
## Abstract
The integrity of the distal alveolar epithelium is crucial for lung regeneration following an injury. The present study aimed to evaluate the effect of *Cinnamomum verum* extract; cross‐talk of epidermal growth factor (EGF) and erythropoietin (EPO) genes in a smoke‐induced lung injury rat model. For experimentation ($$n = 27$$), albino rats were divided equally into three groups, i.e., negative control (NC), positive control (PC), and treatment group (TG). Cigarette smoke was exposed to PC and TG (4 CG/day). C. verum was given orally (350 mg/kg body weight) for 21 days. Decapitation ($$n = 3$$) was done on 14th, 18th, and 21st days, respectively. Analyses (hematology, biochemical, high performance liquid chromatography [HPLC], histology, and gene expression) were carried out and results were statistically analyzed by two‐way analysis of variance. HPLC analysis of ethanolic extract of C. verum was done to identify the presence of phenolic constituents which showed high concentrations of quercetin and P‐coumaric acid. Serum oxidative parameters such as total oxidant status, malondialdehyde, and hematological parameters such as red blood cells, hemoglobin, hematocrit, and white blood cells were significantly ($p \leq .05$) elevated in the PC group; however, these parameters were significantly ($p \leq .05$) improved in TG. While total antioxidant capacity and serum parameters such as total protein, albumin, and globulin were significantly ($p \leq .05$) reduced in the PC group but significantly improved ($p \leq .05$) in TG. Histological analysis revealed that smoke exposure resulted in a measurable increase in alveolar septal thickening while ethanolic extract of C. verum greatly ameliorated the histopathological changes in the lung alveoli. *The* gene expression analysis of EGF and EPO genes showed a significant upregulation ($p \leq .05$) of both genes in PC group while in TG, the level of both genes downregulated, in which lung damage was ameliorated due to cytoprotective effects of ethanolic extract of C. verum.
Histological analysis revealed that smoke exposure resulted in a measurable increase in alveolar septal thickening while ethanolic extract of *Cinnamomum verum* greatly ameliorated the histopathological changes in the lung alveoli.
## INTRODUCTION
The adult lung usually remains in a dormant state during normal homeostatic phenomenon but as soon as lung tissue comes across to any sort of injury, the regenerative response is incited (Kotton & Morrisey, 2014). In lung injury, the most susceptible area of damage is lung epithelium which becomes a primary target in injury and also becomes the first one to come across with repairing process. Lung injury either could be acute or chronic, in each case; it results in damage to the lungs which then ultimately lead to the alteration of normal epithelial homeostasis (Reiss et al., 2012). The repair mechanism of adult lungs depends upon an injury response (Crosby & Waters, 2010). Tissues, such as pancreas, lungs, and liver, exhibit a low steady‐state cellular turnover although they have a responsibility to robustly replace lost or damaged cells after an injury (Hogan et al., 2014).
The role of epidermal growth factor (EGF) is reported in prenatal as well as postnatal lung development but less is known about its role in adult lung growth (Foster et al., 2002). Based upon this, the present study proposed that the EGF gene may have a potential role in regeneration of the damaged alveolar epithelium, as it is considered as a mitogenic factor having the ability to proliferate epithelial cells. *Another* gene selected for our proposed study is the erythropoietin gene (EPO gene); the reason why we have selected it as our gene of interest is because of its antioxidant, anti‐inflammatory, effects to evaluate that either it has any relevant role in attenuating lung injury (Haine et al., 2021).
Presently, many herbal drugs are available for treating various illnesses and the list of herbal therapies is steadily growing (Huntley & Ernst, 2000) as medicinal plant extracts have fewer side effects than synthetic drugs. From the wide list of beneficial plants, *Cinnamomum verum* is also considered as a medicinally beneficial plant having a variety of therapeutic potentials as having an antioxidant potential, as it is rich in certain effective phyto‐constituents (Faix et al., 2009). The present study includes various parameters including gene expression analysis, oxidative stress parameters in both serum and bronchoalveolar lavage (BAL), electrocardiography (ECG) and histopathological analyses were done in order to evaluate the antioxidant potential of ethanolic extract of C. verum in ameliorating the cigarette‐smoke‐induced lung damage. Furthermore, our proposed study aimed to elucidate the regenerative capacities of C. verum and also to analyze the level of expression of the epidermal growth factor gene (EGF gene) and also the erythropoietin gene (EPO gene) in the alveolar epithelium, after inducing lung damage by cigarette smoke.
## Hypothesis
The EGF and EPO genes may have a potential role in regeneration of the damaged alveolar epithelium after inducing lung damage by cigarette smoke. Moreover, C. verum may have therapeutic potential against lung injury.
## Experimental animal model and experimental conditions
A total of 27 albino rats weighing 150–200 g and 6–8 weeks of age were arranged from the Animal rearing nursery of the Department of Physiology, GCUF and brought to the experimental Animal station, Department of Physiology. Consent was taken from the institutional animal ethical committee GCUF (Ref. No. GCUF/ERC/07). The experimental rats were acclimatized to the animal house with standardized environmental conditions of 25°C ± 5 temperature with appropriate humidity of $50\%$ ± 5. The illumination was also maintained at a 12‐h cycle of daylight and dark. Feed and water will be given ad libitum to rats as per their requirement during the experimental trial.
The experimental trial was composed of three groups with nine rats in each group; negative control (NC) group, positive control (PC) group, and treatment group (TG) as shown in Table 1. All were subjected to acclimatization for 1 week. For the induction of smoke‐induced lung injury, the protocol was followed as described by Yu et al. [ 2018] with slight modification. Rats were combined in a chamber (60 × 60 × 60 cm) having two holes at the top of the chamber (outlet), lower part of the chamber (input) connected to a small air pump with a flexible hose. Lung injury was induced by exposing the rats to cigarette smoke (4 CG/day having nicotine content: 1.8 mg/cigarette) for 21 days. During the whole experimental trial of 21 days, hyperactivity, increased sweating, and whistle sound in breathing were observed. The TG ($$n = 9$$) contains rats that were orally administered with ethanolic extract of C. verum (350 mg/kg body weight) along with exposure to cigarette smoke. Another group of healthy rats was considered as the NC group ($$n = 9$$). All three groups were subdivided into three subgroups ($$n = 3$$) for the periodical sampling on the 14th, 18th, and 21st day, respectively.
**TABLE 1**
| S.no | Groups (n = 9) | Subgroups and No. of rats per subgroup (n = 3) | Decapitation day/sampling day |
| --- | --- | --- | --- |
| 1 | Negative control (NC) (n = 9) (Normal healthy rats) Fresh air No smoke exposure | N.S.B‐1 | Day 14 |
| 1 | Negative control (NC) (n = 9) (Normal healthy rats) Fresh air No smoke exposure | N.S.B‐2 | Day 18 |
| 1 | Negative control (NC) (n = 9) (Normal healthy rats) Fresh air No smoke exposure | N.S.B‐3 | Day 21 |
| 2 | Positive control (PC) (n = 9) CS exposure: 4 CG/day/group (Nicotine content: 1.8 mg/cigarette) | P.S.B‐1 | Day 14 |
| 2 | Positive control (PC) (n = 9) CS exposure: 4 CG/day/group (Nicotine content: 1.8 mg/cigarette) | P.S.B‐2 | Day 18 |
| 2 | Positive control (PC) (n = 9) CS exposure: 4 CG/day/group (Nicotine content: 1.8 mg/cigarette) | P.S.B‐3 | Day 21 |
| 3 | Treatment group (TG) CS exposure: 4 CG/day/group Treatment: Ethanolic extract of Cinnamon 350 mg/kg B.W (Oral administration) | T.S.B‐1 | Day 14 |
| 3 | Treatment group (TG) CS exposure: 4 CG/day/group Treatment: Ethanolic extract of Cinnamon 350 mg/kg B.W (Oral administration) | T.S.B‐2 | Day 18 |
| 3 | Treatment group (TG) CS exposure: 4 CG/day/group Treatment: Ethanolic extract of Cinnamon 350 mg/kg B.W (Oral administration) | T.S.B‐3 | Day 21 |
## Preparation of ethanolic extract of C. verum
Cinnamon bark was procured from the herbal market and then assessed by an Expert Botanist and a voucher specimen numbered 272‐bot‐21 was allotted. It was kept in the herbarium of the Department of Botany, Government College University Faisalabad, Pakistan. For the preparation of cinnamon extract, cinnamon bark was milled into fine coarse powder which was macerated in $85\%$ ethanol and kept at room temperature for 3 days (72 h), then the soaked material was filtered and the filtrate was then evaporated using a Rotatory Evaporator (SCILOGEX; Model RE100‐Pro) at 40°C and 30 RPM. The left‐over filtrate was transferred into Petri dishes and kept in an incubator at about 40°C to yield the completely dried powder.
## Body and organ weight
The weekly body weight of each rat was recorded during the experimental tenure while the weight of the lungs of each decapitated animal model was recorded on decapitation days, i.e., Days 14, 18, and 21.
## Collection of blood for hematology
After cervical decapitation, blood samples were taken in an anticoagulant (ehylenediaminetetraacetic acid) tube on the specified days of decapitation to analyze the complete blood profile.
## Collection of serum for oxidative stress measurements
For serum collection, blood samples were taken in clotting vials and were subjected to centrifugation at 1500 g for about 10 min. Serum was collected in Eppendorf's tubes and further analysis of oxidative stress evaluation such as total oxidant status (TOS) and total antioxidant capacity (TAC) by following the calorimetric protocols explained by Anwar et al. [ 2012].
## Total oxidant status (μmol H2O2
equiv./L)
Serum total oxidants were measured spectrophotometrically (BIOLAB‐310) on the basis of the oxidation of ferrous ion to ferric ion in the presence of various oxidant species and was calibrated with hydrogen peroxide. The TOS of samples was determined equivalent to H2O2 standards (Mustafa et al., 2022).
## Total antioxidant capacity (mmol Trolox equiv./L)
Total antioxidants in serum samples were measured spectrophotometrically (BIOLAB‐310) using a novel automated colorimetric method, using o‐dianisidine dihydrochloride as the substrate and the TAC of samples was determined equivalent to Trolox standards (Mustafa et al., 2022).
## Serum analysis for total proteins and albumin
By using commercially available kits, total proteins (Bioclin® Total Protein Monoreagent Diagnostic Kit; K031) and albumin (Bioclin Albumin Monoreagent Diagnostic Kit; K040) were measured in serum samples spectrophotometrically using BIOLAB‐310 manufactured by Biorays. Total Protein assay is done via biuret reaction, which takes place in an alkaline medium in which a chelate is formed between the Cu2+ ion and the peptide bonds present in proteins, and as a result, a violet‐colored complex is formed which is then measured spectrophotometrically at 546 nm (Cov–erdale et al., 2019). Cu+2+Serum Protein→25–37°CpH>12Copper–Protein Complex While the serum concentration of albumin is detected via colorimetric assay using Albumin Assay Kit in which a selective interaction is formed between bromocresol green and albumin as a result a chromophore is generated which is then measured spectrophotometrically at 620 nm. Final results were obtained using the following formula: Total Protein Concentrationg/dl=ΔAbsorbance of specimenΔAbsorbance of standard×6 Albuming/dl=ΔAbsorbance of specimenΔAbsorbance of standard×4
## Collection of bronchoalveolar lavage fluid
Bronchoalveolar lavage fluid (BALF) was performed by lavaging the lungs three to four times with normal saline which was then furtherly centrifuged for 10 min at 1500 g at 4°C (Dianat et al., 2018) and after that, the collected BALF was used to evaluate parameters such as TAC, TOS, and malondialdehyde (MDA).
## Histopathology
Lung tissues were collected in $10\%$ paraformaldehyde for 24 h to do fixation, and then they were dehydrated by different concentrations of ethanol. The tissues were embedded in paraffin wax thereafter and sectioned into 2‐ to 4‐μm‐thin slices with the help of a microtome. The de‐paraffinized sections will then be subjected to eosin and hematoxylin stains, which were further assessed under a light microscope (Huang et al., 2015).
## RNA extraction
A small piece of lung tissue was taken in Trizol (Thermo Fisher Scientific) and kept at −40°C until further processing. By using Nanodrop, isolated RNA was quantified. The cDNA synthesis was done by using Revert‐Aid cDNA synthesis kit (Thermo Fisher Scientific) by using equal concentrations of RNA from each sample according to the manufacturer's instructions. Finally, gene expression analyses of EGF and EPO genes were done using qRT‐PCR (Biorad; BM10‐QPCR96 RT 2).
## Gene expression analysis (qRT‐PCR)
*The* gene expression analysis using primers selected for the genes (EPO and EGF genes) as shown in Table 2 was performed by using the technique of qRT‐PCR.
**TABLE 2**
| S. no | Primer | Sequence | Source |
| --- | --- | --- | --- |
| 1 | EPO‐F | CAGGCGCGGAGATGGG | Macrogen, South Korea |
| 1 | EPO‐R | GGCCCAGAGGAATCAGTAGC | Macrogen, South Korea |
| 2 | EGF‐F | CAGCAACGTGAGCAGTAACG | Macrogen, South Korea |
| 2 | EGF‐R | CAAACCAAGGTTGGGGACCA | Macrogen, South Korea |
## Phytochemical analysis of C. verum by high performance liquid chromatography
To evaluate the phytochemical constituents present in the prepared ethanolic extract of C. verum, high performance liquid chromatography (HPLC) was done according to the conditions as expressed in Table 3 to evaluate the identification of effective constituents responsible for attenuating the adverse effects of acute lung injury induced by cigarette smoke.
**TABLE 3**
| Ethanolic extract | Injection volume (μl) | Flow rate (ml/min) | Detection wavelength (nm) | Run time (min) |
| --- | --- | --- | --- | --- |
| Cinnamomum verum | 10 | 0.8 | 275 | 10 |
## Interpretation of ECG
ECG of rats from each group was assessed after the completion of trial with the help of Powerlab 15T; Model ML 4818 to elucidate the variations in heart rate by the effect of cigarette‐smoke‐induced lung injury and also the effect of treatment was elucidated on the heart rate (Dianat et al., 2014).
## Statistical analysis
Data were subjected to a two‐way analysis of variance (two‐way ANOVA) by using the statistical software and the level of significance between different groups was determined by applying Duncan's multiple range test in CoStat. The result was considered statistically significant at p ≤.05. The data were represented as a mean ± standard error and the statistical evaluation was performed by using Graph Pad prism 8 and CoStat computer software.
## RESULTS AND DISCUSSION
Lung injury was induced in rats via cigarette smoke and then herbal treatment of ethanolic extract of C. verum was introduced to evaluate the therapeutic efficacy of C. verum to ameliorate smoke‐induced lung pathophysiology. The values of all the parameters were analyzed in three groups, i.e., NC group (normal healthy rats), PC group (rats subjected to cigarette smoke), and TG (rats subjected to cigarette smoke along with treatment) at three different days of treatment (Days 14, 18, and 21).
HPLC analysis of ethanolic extract of C. verum was done to identify the presence of certain phenolic and flavonoid constituents in the prepared extract. As a result, high concentrations of polyphenolic flavonoid, i.e., quercetin along with phenolic acid, i.e., P‐coumaric acid were found on chromatogram obtained after HPLC. Chromatogram of ethanolic extract of C. verum is shown in Figure 1 and the relative quantity of phenolic constituents is represented in Table 4. The flavonoids and phenolic constituents of cinnamon may potentially be used as a dietary source of bioactive phytochemicals for improving health. The ethanolic extract of C. verum exhibited strong anti‐inflammatory, antioxidant, and cytoprotective properties. P‐coumaric acid is highly abundant phenolic acid constituent of C. verum and possess biological functions such as anti‐inflammatory, antioxidant, antidiabetic, and anticancerous activities (Sultana et al., 2009). Quercetin is basically a flavonoid and many health benefits are associated with its consumption such as reducing inflammation, neutralizing free radicals, exhibiting immune‐modulating activities, and preventing upper respiratory tract ailments (Li et al., 2016).
**FIGURE 1:** *Representative chromatogram of ethanolic extract of Cinnamomum verum.* TABLE_PLACEHOLDER:TABLE 4 The body weight, organ weight, and relative organ weight of rats among different groups (Table 5) were analyzed with the help of two‐way ANOVA. Cigarette smoke proves to be very injurious to the health status of the body which causes weight loss (Chiolero et al., 2008). While in the TG, despite exposure to cigarette smoke, the treatment, i.e., ethanoic extract of C. verum contributed to ameliorating the damaged or lost cells by the potential of rejuvenating the lung injury which implies that the C. verum also contributes to have regenerative properties.
**TABLE 5**
| Body weight (g) | Body weight (g).1 | Body weight (g).2 | Body weight (g).3 | Body weight (g).4 |
| --- | --- | --- | --- | --- |
| Groups | Day 14 | Day 18 | Day 21 | Overall mean |
| Negative control (NC) | 230.67 ± 4.91a | 250.67 ± 5.46a | 273.33 ± 7.75a | 251.56 ± 12.32A |
| Positive control (PC) | 222 ± 2.31a | 186.67 ± 13.25b | 234.33 ± 6.98b | 214.33 ± 14.28B |
| Treatment group (TG) | 162.67 ± 18.19b | 150.33 ± 5.70c | 153.67 ± 4.70c | 155.56 ± 3.68C |
| Overall mean | 205.11 ± 21.37AB | 195.89 ± 29.33B | 220.44 ± 35.24A | |
The level of total oxidant status (Figure 2a) was significantly higher in the PC group than that of the TG. The level of total oxidant status at different days, i.e., days 14, 18, and 21 was also significant (p ≤ 0.05). The level of total oxidant status was significantly higher on Days 18 and 21 as compared to Day 14, while the level of TAC (Figure 2b) was significantly lower in the PC group than that of the other two groups. Cigarette smoke results in overproduction of free radicals which ultimately causes a reduction in the level of antioxidants. So, in order to combat the diminished functioning of lung proteins due to oxidation, various enzymes and low‐molecular‐weight scavengers are present in the alveolar lining fluid of lung tissue which then knocks out the toxic species in order to ameliorate the lung damage (Lang et al., 2002) but in case of lung injury, the level of antioxidants diminished leading to a high level of oxidative stress (Abdul‐razaq & Ahmed, 2013).
**FIGURE 2:** *(a, b) Total oxidant status (μmol/L; mean ± SE) and total antioxidant capacity (mmol/L; mean ± SE) in negative control (NC), positive control (PC), and Cinnamomum verum treatment group (TG) at different days in cigarette‐smoke‐induced lung injury rat model (different superscripts a–c on bars show a significant difference between groups at p ≤ .05). SE, standard error*
Serum proteins are the health markers that indicate the health condition of the body. Both excessive and reduced levels of proteins cause serious concerns about various health ailments. Cigarette smoke includes large quantities of radicals that generate oxidative stress and generate toxic chemicals which tend to accumulate in the liver progressively and affect serum biochemical parameters related to liver efficiency (Osman et al., 2017).
Serum albumin is considered an acute phase protein having antioxidant properties, contributes as a marker of inflammation. Thus, it implies that whenever a body is under stress condition due to any sort of injury or inflammatory response, the amount of albumin gets reduced indicating the presence of an inflammatory reaction going on in the body (Ranasinghe et al., 2013). In the present study, level of serum total protein, albumin, and globulin was significantly reduced in the PC group but significantly improved in TG (Figure 3a–c).
**FIGURE 3:** *(a–c) Total protein (g/dl; mean ± SE), albumin (g/dl; mean ± SE), and globulin (g/dl; mean ± SE) in negative control (NC), positive control (PC), and Cinnamomum verum treatment group (TG) at different days in cigarette‐smoke‐induced lung injury rat model (different superscripts a–c on bars show significant difference between groups at p ≤ .05). SE, standard error*
Bronchoalveolar lavage is an important diagnostic tool for both research purposes and in clinical medicine related to lung ailments because the BAL fluid contains various cytological and biochemical indicators related to the cellular response to various infections or toxicants. When the toxicity is limited to the lung, useful information can be gathered through BALF analysis, because only the blood analysis cannot always indicate the state of the lung (Song et al., 2013). In the present study, the level of total oxidant status and MDA was significantly higher in PC group than that of the other two groups (Figure 4a–c). However, the level of TAC was significantly lower in PC group which then seemed to be improved in the TG.
**FIGURE 4:** *(a–c) Mean ± SE of total oxidant status (μmol/L H2O2 equiv./L; mean ± SE), malondialdehyde (μmol/L; mean ± SE), and total antioxidant capacity (mmol/L Trolox equiv./L; mean ± SE) in bronchoalveolar lavage fluid among negative control (NC), positive control (PC), and Cinnamomum verum treatment group (TG) at different days in cigarette‐smoke‐induced lung injury rat model (different superscripts a–c on bars show significant difference between groups at p ≤ .05). SE, standard error*
The overall mean of serum red blood cells, serum hematocrit, and serum hemoglobin level among different groups (Table 6) were analyzed with the help of a two‐way ANOVA. The data showed that there was a significant difference (p ≤.05) among all the groups. The level of red blood cells, hematocrit, and hemoglobin was significantly higher in PC group than that in the other two groups. The level of red blood cells and hematocrit at different days, i.e., days 14, 18, and 21 was also significant (p ≤.05). In case of cigarette‐smoke‐induced lung injury, the amount of carboxyhemoglobin increased which then leads to the state of hypoxia which causes excessive secretion of erythropoietin to initiate the production of RBCs (Khan et al., 2014) which further leads to an increase in hemoglobin level (Leifert, 2008). The carbon monoxide present in cigarette smoke elevates the permeability of capillaries which then results in a decrease in plasma volume. This state mimics the state of polycythemia which is characterized by an increase in the share of RBCs in the blood volume. Thus, the hematocrit level increased due to cigarette‐smoke‐induced lung injury (Khan et al., 2014).
**TABLE 6**
| Red blood cells: (106/μl) | Red blood cells: (106/μl).1 | Red blood cells: (106/μl).2 | Red blood cells: (106/μl).3 | Red blood cells: (106/μl).4 |
| --- | --- | --- | --- | --- |
| Groups | Day 14 | Day 18 | Day 21 | Overall mean |
| Negative control (NC) | 7.00 ± 0.32a | 7.02 ± 0.31b | 6.97 ± 0.37b | 6.99 ± 0.02B |
| Positive control (PC) | 7.67 ± 0.14a | 8.29 ± 0.19a | 8.78 ± 0.13a | 8.25 ± 0.32A |
| Treatment group (TG) | 6.88 ± 0.23a | 7.03 ± 0.13b | 7.22 ± 0.19b | 7.04 ± 0.09B |
| Overall mean | 7.18 ± 0.25B | 7.45 ± 0.42AB | 7.66 ± 0.57A | |
The serum white blood cells, lymphocytes, and platelets among different groups were analyzed by two‐way ANOVA as shown in Table 7. The level of white blood cells and lymphocytes was significantly higher in PC group than that of other two groups. The inflammatory stimulation of the respiratory tract induces an increase in the activation of inflammatory cytokines which then affects the number of leukocytes. Leukocytosis is the marker of smoking‐induced tissue damage. While the level of platelets was significantly lower in PC group than that of the other two groups, the level of platelets at different days, i.e., days 14, 18, and 21 was also significant (p ≤.05). There was a gradual increase in platelets values from Day 14 to 21. Thrombocytopenia is an eminent marker of adverse outcomes in patients suffering from pneumonia, as decreased platelet count is associated with severe sepsis and severe intravascular coagulation (Ghoneim et al., 2020).
**TABLE 7**
| White blood cells (103/μl) | White blood cells (103/μl).1 | White blood cells (103/μl).2 | White blood cells (103/μl).3 | White blood cells (103/μl).4 |
| --- | --- | --- | --- | --- |
| Groups | Day 14 | Day 18 | Day 21 | Overall mean |
| Negative control (NC) | 8.22 ± 0.17b | 8.48 ± 0.07c | 8.67 ± 0.08c | 8.46 ± 0.13C |
| Positive control (PC) | 12.87 ± 0.21a | 14.72 ± 0.46a | 15.99 ± 0.55a | 14.53 ± 0.91A |
| Treatment group (TG) | 11.49 ± 0.67a | 11.44 ± 0.51b | 11.93 ± 0.21b | 11.62 ± 0.16B |
| Overall mean | 10.86 ± 1.38B | 11.55 ± 1.80A | 12.20 ± 2.12A | |
Microscopic images of lung tissues from the NC group, PC group, and TG were taken via Light microscope at Magnification 100× (Figures 5, 6, 7). Histological analysis of the present study revealed that smoke exposure resulted in a measurable increase in alveolar septal thickening. Cigarette smoke caused a greater extent of septal thickening and cellular inflammation. Ethanolic extract of C. verum greatly ameliorated the histopathological changes in the lung alveoli caused by cigarette‐smoke‐induced lung injury because it is proved to be enriched with phenolic and flavonoid contents such as P‐coumaric acid and quercetin identified via HPLC analyses. Cigarette smoke exposure led to the infiltration of inflammatory cells into lung interstitium and alveolar spaces and alveolar wall thickening along with the appearance of neutrophil infiltration (Wang et al., 2015). Furthermore, septal thickening and cellular inflammation to a greater extent were also observed in smoke‐exposed rats (Gotts et al., 2017).
**FIGURE 5:** *Representative hematoxylin and eosin stained photomicrographs (100×) of lung tissue at day 14; in negative control (NC), positive control (PC), and Cinnamomum verum treatment group (TG) in cigarette‐smoke‐induced lung injury rat model.* **FIGURE 6:** *Representative hematoxylin and eosin stained photomicrographs (100×) of lung tissue at day 18; in negative control (NC), positive control (PC), and Cinnamomum verum treatment group (TG) in cigarette‐smoke‐induced lung injury rat model.* **FIGURE 7:** *Representative hematoxylin and eosin stained photomicrographs (100×) of lung tissue at day 21; in negative control (NC), positive control (PC), and Cinnamomum verum treatment group (TG) in cigarette‐smoke‐induced lung injury rat model.*
In the present study as shown in Figure 8, the observations obtained by electrocardiograms (ECG) obtained from PC group depicted a shortened R‐R interval along with hidden P‐wave and shortened QRS complex which ultimately depicted a condition of supraventricular arrhythmia which is characterized by irregular heart rate that begins above the ventricles which causes the heart to beat too fast and in irregular manner (Mutiso et al., 2014). While in the TG, this condition is quite improved with normal R‐R interval and decreased heart rate as compared to that of PC group.
**FIGURE 8:** *The electrocardiogram (ECG) charts in negative control (NC), positive control (PC), and Cinnamomum verum treatment group (TG) in cigarette‐smoke‐induced lung injury rat model.*
The mRNA expression level of EGF and EPO genes among different groups (Figure 9a,b), respectively, was analyzed with the help of two‐way ANOVA. The expression level of EGF and EPO genes was significantly higher in PC group, while the expression level was downregulated in the TG. However, on different days, there was no significant difference observed in the expression level of both genes.
**FIGURE 9:** *(a, b) mRNA expression level of EGF and EPO genes (mean ± SE) in negative control (NC), for review only positive control (PC), and Cinnamomum verum treatment group (TG) at different days in cigarette‐smoke‐induced lung injury rat model. a,b different letters on the bars show a significant difference from each other (p ≤ .05). SE, standard error*
One study suggested that the expression capacity of epidermal growth factor receptor, i.e., EGFR is not concentrated in the case of the alveolar membrane of pathogen‐free rodents like rats, mice and also there is less expression of EGFR in the upper and lower of mammalian lung airways during the normal homeostatic phenomenon. On the contrary, EGFR expression in lung airway epithelium is increased in response to any inflammatory stimuli like any sort of inflammatory mediator (TNF‐α), cigarette smoke, or any mechanical injury (Takeyama et al., 2001). Another study elaborated that in the case of asthma, the epidermal growth factor receptor expression is elevated in the alveolar epithelium and its activation leads to repairment in the alveolar epithelium (Burgel & Nadel, 2004). Furthermore, the hematopoietic growth factor erythropoietin (EPO) exhibits all‐tissue‐protective pleiotropic properties. Besides the hematopoietic effect of EPO, it has also been shown to possess certain pleiotropic properties, such as cytoprotective, anti‐inflammatory, and antiapoptotic activities in the cardiovascular system, as well as in the liver and kidneys (Jelkmann, 2004).
## CONCLUSION
The present study concluded that the cross‐talk of EGF and EPO genes plays a crucial role in lung pathophysiology progression. The expression level of both EGF and EPO genes elevated with the alveolar epithelial damage in smoke‐induced lung injury (PC) group in order to activate the compensatory mechanisms to attenuate the damage caused by noxious stimuli (cigarette smoke) while the level of both EGF and EPO genes downregulated in the TG (ethanolic extract of C. verum) as the HPLC analysis of C. verum showed the presence of phyto‐antioxidants like flavonoids (quercetin) and phenolic constituents (P‐coumaric acid) exhibiting cytoprotective, anti‐inflammatory, and radical scavenging properties.
## ACKNOWLEDGMENTS
Authors are thanknful to all the faculty members of Department of Physiology, Government College University Faisalabad, Pakistan for providing technical support for this work.
## FUNDING INFORMATION
Author(s) received no financial support for conducting this research work.
## CONFLICT OF INTEREST
The authors declare that they have no conflict of interest.
## DATA AVAILABILITY STATEMENT
The dataset supporting the conclusions of this article is included within this article.
## References
1. Abdul‐razaq S. N., Ahmed B. M.. **Effect of cigarette smoking on liver function test and some other related parameters**. *Zanco Journal of Medical Sciences* (2013) **17** 556-562
2. Anwar H., Rahman Z. U., Javed I., Muhammad F.. **Effect of protein, probiotic, and symbiotic supplementation on serum biological health markers of molted layers**. *Poultry Science* (2012) **91** 2606-2613. DOI: 10.3382/ps.2012-02172
3. Burgel P. R., Nadel J. A.. **Roles of epidermal growth factor receptor activation in epithelial cell repair and mucin production in airway epithelium**. *Thorax* (2004) **59** 992-996. DOI: 10.1136/thx.2003.018879
4. Chiolero A., Faeh D., Paccaud F., Cornuz J.. **Consequences of smoking for body weight, body fat distribution, and insulin resistance**. *American Journal of Clinical Nutrition* (2008) **87** 801-809. DOI: 10.1093/ajcn/87.4.801
5. Coverdale J. P., Barnett J. P., Adamu A. H., Griffiths E. J., Stewart A. J., Blindauer C. A.. **A metalloproteomic analysis of interactions between plasma proteins and zinc: Elevated fatty acid levels affect zinc distribution**. *Metallomics* (2019) **11** 1805-1819. PMID: 31612889
6. Crosby L. M., Waters C. M.. **Epithelial repair mechanisms in the lung**. *American Journal of Physiology ‐ Lung Cellular and Molecular Physiology* (2010) **298** L715-L731. DOI: 10.1152/ajplung.00361.2009
7. Dianat M., Radan M., Badavi M., Mard S. A., Bayati V., Ahmadizadeh M.. **Crocin attenuates cigarette smoke‐induced lung injury and cardiac dysfunction by anti‐oxidative effects: The role of Nrf2 antioxidant system in preventing oxidative stress**. *Respiratory Research* (2018) **19** 58. DOI: 10.1186/s12931-018-0766-3
8. Dianat M., Radan M., Badavi M., Sarkaki A.. **The evaluation of inotropic properties and antidysrhythmic effect of vanillic acid and exercise on CaCl**. *Research Journal of Pharmaceutical, Biological and Chemical Sciences* (2014) **5** 1545-1555
9. Faix Š., Faixová Z., Plachá I., Koppel J.. **Effect of**. *Acta Veterinaria Brno* (2009) **78** 411-417. DOI: 10.2754/avb200978030411
10. Foster D. J., Yan X., Bellotto D. J., Moe O. W., Hagler H. K., Estrera A. S., Hsia C. C. W.. **Expression of epidermal growth factor and surfactant proteins during postnatal and compensatory lung growth**. *American Journal of Physiology ‐ Lung Cellular and Molecular Physiology* (2002) **283** L981-L990. DOI: 10.1152/ajplung.00053.2002
11. Ghoneim A. H. A., Mohammad M. A., Elghamrawy M. A., Embarak S.. **Platelet count as a predictor of outcome of hospitalized patients with community‐acquired pneumonia at Zagazig University Hospitals, Egypt**. *The Egyptian Journal of Bronchology* (2020) **14** 1-7. DOI: 10.1186/s43168-020-00007-0
12. Gotts J. E., Abbott J., Fang X., Yanagisawa H., Takasaka N., Nishimura S. L., Calfee C. S., Matthay M. A.. **Cigarette smoke exposure worsens endotoxin‐induced lung injury and pulmonary edema in mice**. *Nicotine and Tobacco Research* (2017) **19** 1033-1039. DOI: 10.1093/ntr/ntx062
13. Haine L., Yegen C. H., Marchant D., Richalet J. P., Boncoeur E., Voituron N.. **Cytoprotective effects of erythropoietin: What about the lung?**. *Biomedicine and Pharmacotherapy* (2021) **139**. DOI: 10.1016/j.biopha.2021.111547
14. Hogan B. L. M., Barkauskas C. E., Chapman H. A., Epstein J. A., Jain R., Hsia C. C. W., Niklason L., Calle E., Le A., Randell S. H., Rock J., Snitow M., Krummel M., Stripp B. R., Vu T., White E. S., Whitsett J. A., Morrisey E. E.. **Repair and regeneration of the respiratory system: Complexity, plasticity, and mechanisms of lung stem cell function**. *Cell Stem Cell* (2014) **15** 123-138. DOI: 10.1016/j.stem.2014.07.012
15. Huang R., Zhong T., Wu H.. **Quercetin protects against lipopolysaccharide‐induced acute lung injury in rats through suppression of inflammation and oxidative stress**. *Archives of Medical Science.* (2015) **11** 427-432. DOI: 10.5114/aoms.2015.50975
16. Huntley A., Ernst E.. **Herbal medicines for asthma: A systematic review**. *Thorax* (2000) **55** 925-929. DOI: 10.1136/thorax.55.11.925
17. Jelkmann W.. **Molecular biology of erythropoietin**. *Internal Medicine* (2004) **43** 649-659. DOI: 10.2169/internalmedicine.43.649
18. Khan M. I., Bukhari M. H., Akhtar M. S., Brar S.. **Effect of smoking on red blood cells count, hemoglobin concentration and red cell indices**. *Pakistan Journal of Medical and Health Sciences* (2014) **8** 361-364
19. Kotton D. N., Morrisey E. E.. **Lung regeneration: Mechanisms, applications and emerging stem cell populations**. *Nature Medicine* (2014) **20** 822-832. DOI: 10.1038/nm.3642
20. Lang J. D., McArdle P. J., O'Reilly P. J., Matalon S.. **Oxidant‐antioxidant balance in acute lung injury**. *Chest* (2002) **122** 314S-320S. DOI: 10.1378/chest.122.6_suppl.314 S
21. Leifert J. A.. **Anaemia and cigarette smoking**. *International Journal of Laboratory Hematology* (2008) **30** 177-184. DOI: 10.1111/j.1751-553X.2008.01067.x
22. Li Y., Yao J., Han C., Yang J., Chaudhry M. T., Wang S., Liu H., Yin Y.. **Quercetin, inflammation and immunity**. *Nutrients* (2016) **8** 1-14. DOI: 10.3390/nu8030167
23. Mustafa I., Anwar H., Irfan S., Muzaffar H., Ijaz M. U.. **Attenuation of carbohydrate metabolism and lipid profile by methanolic extract of**. *BMC Complementary Medicine and Therapies* (2022) **22** 1-12. DOI: 10.1186/s12906-022-03507-2
24. Mutiso S. K., Rono D. K., Bukachi F.. **Relationship between anthropometric measures and early electrocardiographic changes in obese rats**. *BMC Research Notes* (2014) **7** 1-7. DOI: 10.1186/1756-0500-7-931
25. Osman G., Galal M., Mohammed A.. **Polycyclic aromatic hydrocarbons (PAHS) accumulation and histopathological biomarkers in gills and mantle of**. *Punjab University Journal of Zoology* (2017) **32** 39-50
26. Ranasinghe P., Pigera S., Premakumara G. S., Galappaththy P., Constantine G. R., Katulanda P.. **Medicinal properties of “true” cinnamon (**. *BMC Complementary and Alternative Medicine* (2013) **13** 1. DOI: 10.1186/1472-6882-13-275
27. Reiss L. K., Uhlig U., Uhlig S.. **Models and mechanisms of acute lung injury caused by direct insults**. *European Journal of Cell Biology* (2012) **91** 590-601. DOI: 10.1016/j.ejcb.2011.11.004
28. Song J. A., Yang H. S., Lee J., Kwon S., Jung K. J., Heo J. D., Cho K. H., Song C. W., Lee K.. **Standardization of bronchoalveolar lavage method based on suction frequency number and lavage fraction number using rats**. *Toxicological Research* (2013) **26** 203-208. DOI: 10.5487/TR.2010.26.3.203
29. Sultana B., Anwar F., Ashraf M.. **Effect of extraction solvent/technique on the antioxidant activity of selected medicinal plant extracts**. *Molecules* (2009) **14** 2167-2180. DOI: 10.3390/molecules14062167
30. Takeyama K., Jung B., Shim J. J., Burgel P. R., Dao‐Pick T., Ueki I. F., Protin U., Kroschel P., Nadel J. A.. **Activation of epidermal growth factor receptors is responsible for mucin synthesis induced by cigarette smoke**. *American Journal of Physiology ‐ Lung Cellular and Molecular Physiology.* (2001) **280** L165-L172. DOI: 10.1152/ajplung.2001.280.1.l165
31. Wang Q., Wu X., Tong X., Zhang Z., Xu B., Zhou W.. **Xuebijing ameliorates sepsis‐induced lung injury by downregulating HMGB1 and RAGE expressions in mice**. *Evidence‐based Complementary and Alternative Medicine* (2015) **2015**. DOI: 10.1155/2015/860259
32. Yu N., Sun Y. T., Su X. M., He M., Dai B., Kang J.. **Treatment with eucalyptol mitigates cigarette smoke‐induced lung injury through suppressing ICAM‐1 gene expression**. *Bioscience Reports* (2018) **38** 1-10. DOI: 10.1042/BSR20171636
|
---
title: A PRECEDE‐PROCEED model‐based educational intervention to promote healthy eating
habits in middle school girls
authors:
- Asma Arshad
- Fouzia Shaheen
- Waseem Safdar
- Muhammad R. Tariq
- Muhammad T. Navid
- Asma S. Qazi
- Mohammad A. Awan
- Muhammad W. Sajid
- Humphrey K. Garti
journal: Food Science & Nutrition
year: 2022
pmcid: PMC10002990
doi: 10.1002/fsn3.3167
license: CC BY 4.0
---
# A PRECEDE‐PROCEED model‐based educational intervention to promote healthy eating habits in middle school girls
## Abstract
The present study was designed to develop Nutrition Education Program (NEP) based on PRECEDE‐PROCEED model (PPM) to address healthy eating behavior among middle school girls aged between 4 and 12 years. For this, middle school girls from grade 1 to 8 ($$n = 900$$) were consulted for their eating behaviors, followed by the analysis of their health problems. From 15 different schools of three large cities (Faisalabad, Lahore, and Rawalpindi) of Pakistan, students were divided into two groups: control group ($$n = 30$$) and intervention group ($$n = 30$$) from each school. The data were collected through interview‐based questionnaires according to the phases of PRECEDE Model and evaluated based on PROCEED model. Implementation of NEP was carried out through lectures. Lessons were prepared to enhance student's awareness about nutritious food and healthy lifestyle through educational pamphlets and influenced their attitude towards selection of food choices from My‐Plate. Results showed that NEP was quite successful for long‐term results. A significant increase in total caloric intake was observed after 8 weeks of NEP intervention (1694 ± 217 Kcal) as compared to before intervention (1329 ± 318 Kcal). Similarly, carbohydrate, protein, and fat content was also increased in daily diet. Conclusively, NEP based on PPM has great impact on healthy lifestyle of middle school girls. Significant difference was observed in score of health variables before and after NEP intervention.
The present study was designed to develop Nutrition Education Program (NEP) based on PRECEDE‐PROCEED model (PPM) to address healthy eating behavior among middle school girls aged between 4 and 12 years. NEP based on PPM has great impact on healthy lifestyle of middle school girls. Significant difference was observed in score of health variables before and after NEP intervention.
## INTRODUCTION
The early childhood (aged 4–12 years) represents the largest generation of the world population, $90\%$ of which reside in low‐ or middle‐income countries. Many studies show that dietary behaviors during early childhood contribute to the establishment of lifelong eating patterns (Alfaro et al., 2020; Aziz et al., 2018; Barasheh et al., 2017; Liu et al., 2021; Melián‐Fleitas et al., 2021; Sirasa et al., 2019; Xi et al., 2021). During the early transition period from infant to early childhood, intake of soft, semi‐solid, or solid nutrient‐dense foods is essential due to high nutritional needs. Children aged 2–5 years required adequate protein, micronutrients, and essential fatty acids marked as a significant stage for creating dietary patterns can reach out to adulthood (Motevalli et al., 2021). When children start schooling, significant weight changes are observed in them which is of public health concern. The deficiencies that are present in early age also become the health issue later in life. Healthy eating habits among these early and middle childhood stages are essential for healthy growth, cognitive development, as well as various other aspects of good physical health and mental wellbeing (Liu et al., 2021; Xi et al., 2021). A study focused on dietary habits of high school children showed that they were exposed to high‐density fast foods and their meals had vegetable and fruits in very less frequency (Alfaro et al., 2020). Their access to fast foods was much easier during school hours. The study also showed that most of them did not bring lunch to school.
In developing countries, dietary intake of school children is limited to consumption of fruits and vegetables. In Pakistan, the life style has drastically changed especially in the urban areas where the shift is seen due to lack of time among working population (Almas et al., 2020). With the sudden expansion of technology and increased production as well as ready availability of highly processed foods, have shifted preadolescent's dietary patterns from healthier to less healthy foods (Ochola et al., 2014). More intake of the processed food has been found to result in development of chronic diseases (Aziz et al., 2018). Research in the diet of children indicates that nutritional deficiency in primary school students is among the causes of low enrolment in school, high absenteeism on daily basis, early dropout, as well as poor classroom performance (Almas et al., 2020). The National Nutrition Survey (NNS) conducted by the Government of Pakistan along with UNICEF shows that $41.5\%$ of children under the age of 5 years are underweight, $11.6\%$ of them have wasting, of those underweight, $31\%$ have stunted growth and about half of them are anemic (Achakzai, 2016; FAO‐UN, 2019).
Lack of knowledge about the dietary intake is of great concern in many areas of Pakistan. This lack of information leads to low nutritional intake which in turn is affecting the nutritional status especially of the individuals in the growing age. While limited resources are seen addressing this issue, intervention in the form of nutritional education by using a theoretical model increases the affectivity of the programs that make the population aware about their dietary intake (Pereira & Oliveira, 2021). PRECEDE‐PROCEED model has been used in various studies to help those health program planners that are responsible for policy making. It also helps evaluators and analyzers in different situations to design health programs efficiently. PRECEDE involves Predisposing, Reinforcing, and Enabling Constructs in Educational Diagnosis and Evaluation. PROCEED means Policy, Regulatory, and Organizational Constructs in Educational and Environmental Development (Handyside et al., 2021; Jeihooni et al., 2019; Khorsandi et al., 2012). It guides planners in comprehensive process starting with desired outcomes and going backwards to assess health and quality of life, identify strategies and design, and implement as well as evaluate health promotion programs (Marques et al., 2020; Solhi et al., 2016).
Thus, proper and targeted intervention at early stage of child's life is required. It is important to design appropriate strategies to control malnutrition and improve dietary intake to meet the requirements. The PRECEDE‐PROCEED model is a cost‐benefit evaluation framework that could be used by the health and nutrition policy makers, school administrators, and nutrition educators to analyze situations and design health and nutrition education programs effectively. The present study focused on nutrition education based on PRECEDE‐PROCEED model considering the dietary requirements of middle school students aged 4–12 years from different areas of Pakistan. The study aims to develop the nutrition education program based on healthy food choices and to assess its impact on dietary intake and knowledge of students. This study further provides a comprehensive structure for analyzing nutrition education needs of the school going children.
## Study design, schools, and participants
In this study, PRECEDE‐PROCEED Model (PPM) was used as a planning tool to assess the intake and nutritional knowledge before and after educational intervention. In order to educate the children, the nutrition education program consisted of lesson plans and worksheets each formulated according to the stages of PRECEDE‐PROCEED Model. For this, 15 middle schools were selected in three big cities (Faisalabad, Lahore, and Rawalpindi) of Punjab province, Pakistan, based on available facilities to implement educational lessons and no other research project was conducted in these schools related to nutritional intervention. School principals were contacted and informed comprehensively about the objectives and procedures of this study. Quasi‐experimental study was performed on 900 school girls, 60 from each school. The cluster analysis method was used for selection of children aged between 4 and 12 years if they were not participating any other research projects, were not having any clinical conditions/symptoms except malnutrition or growth impairment, and had not using vitamin and mineral supplements. Informed written consent was taken from each child's parents/guardians.
## PRECEDE model stages
The data for the study were collected through questionnaires according to the phases of PRECEDE Model starts with diagnostic activities (social, behavioral, and educational diagnosis), predisposing factors, enabling factors, and reinforcing factors. At the stage of educational diagnosis, the predisposing, enabling, and reinforcing factors were reviewed. Predisposing factors were knowledge and diet intake of children. Enabling factors include socioeconomic status, father's occupation, family income, parent's education level, and access to the fast food. The reinforcing factors include encouragement from teachers, family, and peers. Soliciting input from key informants was cross verified from their parents/guardians and school management. The baseline data were collected for age, height, weight, and Body Mass Index (BMI). Predisposing factors were assessed by Nutrition Knowledge Questionnaire (NKQ; Table S1) consisted closed‐ended questions in which subjects were asked to either agree to or disagree with the statements about the knowledge. Enabling factors consisted of seven closed‐ended questions (Table S2). These questions included socioeconomic status, father's occupation, family income, parent's education level, and access to the fast food. For food Frequency Questionnaire (FFQ), all those behaviors were considered that were affecting the dietary intake like breakfast habits, meal skipping, and snacking according to Pakistan dietary guidelines, 2019 (Table S4). The FFQ questions were asked about the usual intake from each food group over a specified time period. Questionnaire was distributed before the start of lectures among students and was read aloud for better understanding. Students were given pretest questionnaire in the class along with the brief introduction about the purpose of researcher's study. At the end of the intervention, an immediate post‐test was conducted.
## PROCEED model stages
The phases of PROCEED Model include impact evaluation and outcome evaluation. Evaluation was conducted after the educational intervention. A post‐test questionnaire was given to the students, which was similar to the questionnaire used in pretesting. After a time period of 1 and 3 months, a follow‐up (outcome) testing was conducted. Lectures given to the students were developed by a team of researchers and middle school teachers according to Pakistan dietary guidelines [2019] to make the intervention effective to improve the student's knowledge. Lesson plans were consisting of introduction to nutrition and balanced diet, nutrients in food (carbohydrates, protein, and fats), vitamins, minerals, My Plate, food groups (cereals, meat, dairy, fruits, and vegetables), recommended allowance from each food group, food servings, adverse effects of fast food, healthy foods choices for snacks, meal planning, physical activity, and safe food handling. The nutrition education program was a 6‐week plan with two lessons per week included PowerPoint presentations, educational posters, short films, group discussions, and question answer sessions. The time period for each lesson was 45 minutes. Worksheets were given to the students after each lecture that were prepared according to the lesson content as a reinforcing activity.
## Data analysis
Data were managed and analyzed using Statistical Package for the Social Sciences version 25 (SPSS Inc.). Chi‐Square test was used to analyze demographic and anthropometric characteristics concerning different dietary patterns. Analysis of variance (anova) was used to evaluate statistical differences in control and intervention groups at significance $p \leq .05$, confidence level is $95\%$.
## RESULTS AND DISCUSSIONS
The present study was designed to develop Nutrition Education Program (NEP) based on PRECEDE‐PROCEED model (PPM) to address healthy eating behavior among middle school girls aged between 4 and 12 years in light of Pakistan Dietary Guidelines 2019 (FAO‐UN, 2019). For this, students from 15 different schools of three large cities Faisalabad, Lahore, and Rawalpindi, Pakistan, were consulted for their eating behaviors. Numerous NEP had been set, developed, and implemented worldwide for multiple populations. However, nutritional assessment paired with instruction/lectures for promoting healthy eating habits was rarely implemented on school students. It is well understood that dynamic interplay of environmental, personal, and social behavioral factors has great impact on individual's health. More importantly, a major role played by environmental factors directly or indirectly may hinder self‐efficacy of individuals seeking healthy behaviors. We believe that an eight‐phase PPM would be helpful in creating a conceptual framework of healthy lifestyle among middle school students.
## PRECEDE model
Prior to intervention of Nutritional Education Program through PROCEED model, four phases of PRECEDE model were used for planning of soliciting relevant information as depicted in cover figure. It provided a framework to determine nutritional/health problem in middle school girls aged 4–12 years and in formulation of educational content which addresses their needs. These systematic sequential steps of PRECEDE‐PROCEED model increase the sustainability and effectiveness of education programs intervention (Pourhaji et al., 2020).
## Social assessment (Phase‐1)
Before selection of participants, consultation was carried out with students of different grades, teaching staff, and school administrations. Demographic information for different grades was taken to understand the key issues related to health and quality of life. All of the respondents were school girls from grade one to grade eight. Data were collected through questionnaires and/or oral interviews where required. Demographic data of the participants are presented in Table 1. Cluster sampling was done for selection of participants. A total of 30 students for intervention and 30 students for control group were considered in this study from each school. Control group was taken for comparison of NEP at baseline. Participants were divided in to three subgroups based on their grade of study. Demographic data were collected for family size, birth order, parent's education, parent's occupation, and average family income.
**TABLE 1**
| Characteristics | Intervention group n = 450 | Control group n = 450 | p value |
| --- | --- | --- | --- |
| Class of students (%) | Class of students (%) | Class of students (%) | Class of students (%) |
| Group‐1 (1–3 grade) | 26.7 | 33.3 | .291 |
| Group‐2 (1–3 grade) | 33.3 | 33.3 | .291 |
| Group‐3 (1–3 grade) | 40.0 | 33.3 | .291 |
| Family size (%) | Family size (%) | Family size (%) | Family size (%) |
| Three | 24.4 | 22.2 | .246 |
| Four | 40.0 | 31.1 | .246 |
| Five | 26.7 | 33.3 | .246 |
| Six or more | 8.9 | 13.3 | .246 |
| Birth order (%) | Birth order (%) | Birth order (%) | Birth order (%) |
| First | 35.6 | 26.7 | .381 |
| Second | 37.8 | 33.3 | .381 |
| Third | 22.2 | 26.7 | .381 |
| Fourth or higher | 4.4 | 13.3 | .381 |
| Father's occupation (%) | Father's occupation (%) | Father's occupation (%) | Father's occupation (%) |
| Govt. employee | 24.4 | 31.1 | .133 |
| Private employee | 44.4 | 40.0 | .133 |
| Business man | 31.1 | 26.7 | .133 |
| Unemployed | 0.0 | 2.2 | .133 |
| Mother's occupation (%) | Mother's occupation (%) | Mother's occupation (%) | Mother's occupation (%) |
| Employed | 15.6 | 26.7 | .116 |
| Housewife | 84.4 | 73.3 | .116 |
| Father's education (%) | Father's education (%) | Father's education (%) | Father's education (%) |
| ≤10th grade | 8.9 | 15.6 | .208 |
| 10th to 12th grade | 13.3 | 11.1 | .208 |
| <12th grade to Bachelors | 55.6 | 53.3 | .208 |
| More than Bachelors | 22.2 | 17.8 | .208 |
| Mother's Education (%) | Mother's Education (%) | Mother's Education (%) | Mother's Education (%) |
| ≤10th grade | 26.7 | 11.1 | .151 |
| 10th to 12th grade | 6.7 | 17.8 | .151 |
| <12th grade to Bachelors | 55.6 | 64.4 | .151 |
| More than Bachelors | 11.1 | 6.7 | .151 |
| Average income (%) | Average income (%) | Average income (%) | Average income (%) |
| <20,000 | 0.0 | 0.0 | .012 |
| 31,000–40,000 | 4.4 | 0.0 | .012 |
| 41,000–50,000 | 15.6 | 24.4 | .012 |
| 51,000–70,000 | 62.2 | 51.1 | .012 |
| More than 70,000 | 17.8 | 24.4 | .012 |
There was no significance difference in demographic characteristics between control and intervention group. Almost an equal number of respondents were included in both groups with respect to their grade of study. Majority of the students belong to a small family size (3–4 family members, $22.2\%$–$40\%$) with moderate income status. Fathers of majority of the respondents work as private employee ($40\%$–$44.4\%$) followed by Bossiness man ($26.7\%$–$31.1\%$) in control and intervention group, respectively. Majority of the participants have educated parents. Similar pattern of participants was observed in previous researches at their own capacity (Handyside et al., 2021; Jeihooni et al., 2019; Mosavi et al., 2020; Nejhaddadgar et al., 2019; Sezgin & Esin, 2018). During application of PPM on different communities to assess their quality of life, similar strata was designed with equal capabilities of respondents (Handyside et al., 2021).
## Epidemiological assessment (Phase‐2)
In addition to demographic properties, baseline characteristics of anthropometric measures with respect to studied population were also evaluated. Results of anthropometric data are represented in Table 2. Analysis showed that there was no significant difference in baseline data of both control and intervention group. Mean age of the respondents was 9.16 ± 4.12 and 9.62 ± 5.46 years, respectively, in control and intervention group. Similarly, height weight and BMI of the respondent of both groups were statistically similar ($p \leq .05$). It was also observed that, on the average, all participants have poor health status (based on body mass index). Nevertheless, regarding their different grade level, health evaluation was carried out within their respective age percentile. These results suggested that quality of life of participants in both groups was analogously poor and hence, PRECEDE phase three and four needed to be applied for further planning of intervention (i.e., NEP). Similar studies were conducted previously on adult community with different heath issues needed education intervention (Franceschi et al., 2021; Labyak et al., 2021). According to dietary guideline of Pakistan, 2019, $43.9\%$ of the young female were suffering malnutrition, especially deficiency of some minerals like iron and calcium. Anthropometric results of the present study also showed malnutrition status of middle school girls. Nutrition knowledge questionnaire was developed based on students' class and their daily life style, which is used in phase 3 of PRECEDE model.
**TABLE 2**
| Characteristics | Intervention group n = 450 | Control group n = 450 | p value |
| --- | --- | --- | --- |
| Age (years) | 9.62 ± 5.46 | 9.16 ± 4.12 | 0.291 |
| Height (cm) | 135.91 ± 9.64 | 134.63 ± 12.4 | 0.116 |
| Weight (pounds) | 57.91 ± 1.247 | 56.14 ± 1.49 | 0.172 |
| BMI | 16.94 ± 2.99 | 16.98 ± 2.71 | 0.319 |
## Educational and ecological assessment (Phase‐3)
Educational Assessment was conducted according to three factors of PRECEDE model: predisposing factors, enabling factors, and reinforcing factors. For the students, the predisposing factor was knowledge and food intake. Enabling factors included socioeconomic status, father's occupation, family income, parent's education level, and access to the fast food. The reinforcing factors included encouragement from teachers, family, and peers. Educational assessment was done through nutrition knowledge questionnaire (Table S1). Questions were designed with consultation of field experts and school teachers. Generally, question responses were recorded as agree, disagree, and unsure. Questionnaire was filled by the examiners for low‐grade classes (1–5 grade); however, participants were encouraged to fill themselves. Questionnaire was prepared according to the knowledge about their daily dietary routine gained during consultation sessions with students and school staff. School cafeteria was also inspected with due permission of school administration for availability of food and food products. There was no special restriction for selling junk food such as fries, sugar drinks, samosa, burger, and shawarma, which was regularly consumed by the students during school timings. Although drinking water facility was provided by the school administration at easy access, conversely, availability of sugary soft‐drinks offers unhealthy competitive food against water.
Before intervention, data for knowledge and lifestyle were gathered and compared with control. No significant difference was observed between control and intervention group participants before the intervention of NEP. However, after intervention NEP, significant improvement was noticed. Student t‐test was performed with SPSS software to evaluate the difference in control and intervention group before NEP (Table 3). These results suggested that before intervention, students have very low knowledge about nutrition and healthy lifestyle. Similar kind of preintervention assessment was carried out on diabetic patients to maintain healthy life style (Kołota & Głąbska, 2021). The selected participants had same nutritional knowledge, attitude, and enabling factors. In another study on development of culinary nutrition knowledge among school students, analysis on intervention and control group was carried out together for comparison (Afrin et al., 2021; Maximova et al., 2021). Similar results were obtained in different studies conducted on different community problems (Azar et al., 2017; Handyside et al., 2021; Jeihooni et al., 2019). Six different reinforcing factors were evaluated on the grounds of five‐point scale of strongly agree, agree, neutral, disagree, and strongly disagree. Results of reinforcing factors are represented in Figure 1.
Results of reinforcing factors showed that most of the students were encouraged by their parents (approximately $80\%$) to eat fruits and vegetables and discourage to eat fast food which is not healthy at all. Similar results were obtained for teacher enforcement about fruits and vegetables. There is a mixed response of the students about friends eating behavior. For the question “My friend like to eat fruit and vegetables,” disagreement of the student was stronger as compared to agreement. Furthermore, most of the participants have friends who like to eat fast/cafeteria food. This unhealthy company/environment had great influence on student's health. ( Jeihooni et al., 2019) also reported direct influence of friend circle and social pressure on eating behavior of school going students. However, after holding 3 months educational sessions for teachers and parents and also presenting educational content to social network groups enhanced score of reinforcing factors. Peer groups in schools directly effect on choosing and continuing lifestyle patterns (Pourhaji et al., 2020). Overall results of reinforcing factors were in accordance with the anthropometric results. Malnutrition status of the school girls was due to unhealthy food availability and friend circle having bad eating habits. Group discussions from 251 students of elementary school, America, revealed that girls were less physically active than boys and highlighted the effect of social support on eating behavior of students (Jeihooni et al., 2019). In another study, increase in scores of reinforcing factors was observed when nutritional education was presented to parents, school officials, and teachers (Nejhaddadgar et al., 2019).
## Administrative and policy assessment (Phase‐4)
Administrative management policies were also consulted for batter health of school students. Availability of healthy food, fruits and vegetables, nuts, physical activities of students facilitated by the administration, and hygiene conditions of food area were evaluated. Data about regular eatables were collected through food frequency questionnaire (Table S3). In this last phase of PRECEDE model, availability of facilities, support, and resources were consulted for the implementation on actual intervention during PROCEED model of the study. Coordination was established with management of the schools and designated facilitators for distribution of responsibilities, budgeting, administrative barriers, personnel availability, and all other necessary supports from school, participants, and their parents. During this phase, venue of interventions/lecture sessions and timetable were also designed with consent of teachers and informants. Learning material was developed under supervision of education experts according to the objectives of this study.
## PROCEED model
The phases of PROCEED Model included implementation of the planning during PRECEDE model, evaluation of the process of implementation, impact of the evaluation, and final outcome evaluation. Evaluations were carried out on similar grounds of knowledge, attitude, enabling factor, reinforcing factors, nutrition knowledge questions, and food frequency questionnaire assessment.
## Implementation and process evaluation (Phase 5–6)
Implementation of the nutrition education program (NEP) was carried out through lectures. Lectures were delivered by poster presentations and multimedia presentations. Content and schedule of the lectures are given in Table 4. During lectures, students exhibit enthusiasm about interesting facts for nutrition and healthy diet. After each intervention, evaluation of the process was carried out through worksheets and results are presented in Figure 2. Twelve lessons were prepared, two lessons per week, and each lesson is of 45–60 min according to the content of lesson. Knowledge assessment after interventions was compared with before intervention as well as control group. After each session, participants were encouraged to have colloquy sessions for clarity of concepts. Lessons were prepared to enhance students' awareness about nutritious food and healthy life style through educational pamphlets and influenced their attitude towards selection of food choices from My‐Plate. Students were guided about three major components of diet and their right proportion with respect to optimal calories.
A special session was also arranged at the end of NEP for parents and teachers to reinforce their kids/students for regular eating of fruits and vegetables and drink plenty of water. They were also guided to ask students for daily physical exercise and avoid junk/fast food from canteen. Impact of the evaluation and final outcome evaluation was carried out after 4 and 8 weeks of interventions. Numerous studies have been published on PPM of health care education; most of them evaluated outcome of the exercise immediately after a month or two (Azar et al., 2017; Azar et al., 2018; Handyside et al., 2021; Jeihooni et al., 2019; Nejhaddadgar et al., 2019; Sezgin & Esin, 2018; Solhi et al., 2016).
## Impact evaluation and final outcome evaluation (phase 7–8)
After NEP, evaluation was carried out immediately after the lessons, 4 weeks after the interventions and 8 weeks of the interventions. Results of predisposing factors including knowledge and attitude showed significant improvement after intervention ($$p \leq .011$$). Surprising, huge improvement mean score of the knowledge and attitude towards healthy diet along with intervention sessions. Highest mean score for both was observed in after 8 weeks of intervention. While comparing with control group (8.92), significant improvement in knowledge was observed in intervention group (35.41). Similar results were noticed in attitude variable. Comparing with control group (18.71) significant improvement ($$p \leq .0019$$) in attitude was observed in intervention group (47.30). The mean score of the enabling factors also shows significant improvement ($$p \leq .006$$) towards healthy diet along with intervention sessions. Highest mean score was observed after 8 weeks of intervention (51.99) as compared to before intervention.
While comparing with control group (23.36), significant improvement in enabling factors was observed in intervention group (51.99). Mean score of the reinforcing factor towards healthy diet along with intervention sessions also showed significant difference ($$p \leq .0012$$). Highest mean score was observed after 8 weeks of intervention. While comparing with control group (26.13), significant improvement in knowledge was observed in intervention group (55.96). Similar results were noticed in attitude variable. Similarly, highest mean score was observed in performance after 8 weeks of intervention. While comparing with control group (16.49), significant improvement in knowledge was observed in intervention group (53.99). Similar results were noticed in attitude variable. These results showed that NEP based on PPM has high impact on middle school girls. Although in advanced countries PPM was used by many researchers, a very few interventions were reported in developing countries (Pourhaji et al., 2020). Results from previous studies showed desirable general health status and quality of life score by similar strategies of PPM. However, (Pourhaji et al., 2020) presented relatively unfavorable quality of life scores while using PPM‐based educational intervention. In several recent studies, significant increase was observed in knowledge, attitude, and of self‐efficacy scores of intervention group by PPM (Handyside et al., 2021; Jeihooni et al., 2019; Lee & Lee, 2020; Nejhaddadgar et al., 2019).
Comparative analysis of predisposing factors is presented in Figure 2b,c. Significant difference was observed before and after intervention. However, there was no significant difference immediately, after 4 weeks and after 8 weeks of intervention. These results showed that NEP was quite successful for long‐term results. Nutrient intake was also calculated from food frequency questionnaire. Results are depicted in Tables 5 and 6. Total caloric intake before intervention was 1329 ± 318 Kcal; after 8 weeks of NEP intervention (1694 ± 217 Kcal), significant difference was observed in total calories. Similarly, carbohydrate, protein, and fat content was increased in daily diet.
Conclusively, NEP based on PPM has great impact on healthy lifestyle of middle school girls. Significant difference was observed in score of health variables before and after NEP intervention. Previous studies also show successful application of this model for health education programs on different communities. Results from present study showed that school going students especially girls have nutritional deficiencies. Furthermore, they have unrestricted access to junk and fast food with unbalanced dietary components, high fat content, and even unhygienic servings. During this study, school administration was suggested for abiding by these processed foods and to facilitate students with fresh fruits and vegetables. Similarly, easy access of sugar beverages limits water consumption in school students. Therefore, students should be encouraged by the parents, teachers, and school administrations adopting healthy lifestyle. PPM‐based nutrition education can be effective in reducing malnutrition in school going students. PPM improvisation, according to the participants, was often required to reduce potential barrier in successful outcomes. Findings of this study can be used to improve healthy eating behavior across populations among all age groups. Especially, PPM would be very effective as a primary intervention in school going children to enhance quality of life. To maintain long‐term healthy behavior, social media approach along with face‐to‐face intervention would be a more successful way in future. Further research is required to consolidate best and easiest channels, and face‐to‐face interventions and alleviated time constraints may develop a better way to increase program effectiveness.
## INSTITUTIONAL REVIEW BOARD STATEMENT
The current study was reviewed and approved by Institutional Review Board (IRB), Faisalabad Medical university (FMU), Faisalabad, Pakistan, and Department of Food sciences, University of the Punjab, Lahore, Pakistan. Informed consent was obtained from all participants (parents/guardians of children) involved in the study for collection and analysis of dietary data.
## CONFLICT OF INTEREST
All authors declare no conflict of interest in this manuscript.
## DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request.
## References
1. Achakzai A. B. K.. *Pakistan adolescent nutrition strategy and operational plan* (2016)
2. Afrin S., Mullens A. B., Chakrabarty S., Bhowmik L., Biddle S. J. H.. **Dietary habits, physical activity, and sedentary behaviour of children of employed mothers: A systematic review**. *Preventive Medicine Reports* (2021) **24**. DOI: 10.1016/j.pmedr.2021.101607
3. Alfaro B., Rios Y., Arranz S., Varela P.. **Understanding children's healthiness and hedonic perception of school meals via structured sorting**. *Appetite* (2020) **144**. DOI: 10.1016/j.appet.2019.104466
4. Almas A., Iqbal R., Sabir S., Ghani A., Kazmi K.. **School health education program in Pakistan (SHEPP) – a threefold health education feasibility trial in schoolchildren from a lower‐middle‐income country**. *Pilot and Feasibility Studies* (2020) **6** 1-10. DOI: 10.1186/s40814-020-00625-x
5. Azar F. E., Solhi M., Darabi F., Rohban A., Abolfathi M., Nejhaddadgar N.. **Effect of educational intervention based on PRECEDE‐PROCEED model combined with self‐management theory on self‐care behaviors in type 2 diabetic patients**. *Diabetes and Metabolic Syndrome: Clinical Research and Reviews* (2018) **12** 1075-1078. DOI: 10.1016/j.dsx.2018.06.028
6. Azar F. E., Solhi M., Nejhaddadgar N., Amani F.. **The effect of intervention using the PRECEDE‐PROCEED model based on quality of life in diabetic patients**. *Electronic Physician* (2017) **9** 5541-5550. PMID: 29238496
7. Aziz A., Pervaiz M., Khalid A., Khan A. Z., Rafique G.. **Dietary practices of school children in Sindh, Pakistan**. *Nutrition and Health* (2018) **24** 231-240. DOI: 10.1177/0260106018791859
8. Barasheh N., Shakerinejad G., Nouhjah S., Haghighizadeh M. H.. **The effect of educational program based on the precede‐proceed model on improving self‐care behaviors in a semi‐urban population with type 2 diabetes referred to health centers of Bavi, Iran**. *Diabetes and Metabolic Syndrome: Clinical Research and Reviews* (2017) **11** S759-S765. DOI: 10.1016/j.dsx.2017.05.012
9. **Pakistan dietary guidelines for better nutrition**. *Ministry of Planning, Development and Reform, Government of Pakistan* (2019) **21** 24-25
10. Franceschi R., Fornari E., Ghezzi M., Buzzi E., Toschi M., Longhi S., Maimone R., Forti S., Carneri S., Pirous F. M., Agostini B., Iori T., Gibin M., Porchia S., Soffiati M., Maffeis C.. **Educational intervention of healthy life promotion for children with a migrant background or at socioeconomic disadvantage in the north of Italy: Efficacy of telematic tools in improving nutritional and physical activity knowledge**. *Nutrients* (2021) **13** 1-13. DOI: 10.3390/nu13103634
11. Handyside L., Warren R., Devine S., Drovandi A.. **Utilisation of the PRECEDE‐PROCEED model in community pharmacy for health needs assessment: A narrative review**. *Research in Social and Administrative Pharmacy* (2021) **17** 292-299. DOI: 10.1016/j.sapharm.2020.03.021
12. Jeihooni A. K., Heidari M. S., Harsini P. A., Azizinia S.. **Application of PRECEDE model in education of nutrition and physical activities in obesity and overweight female high school students**. *Obesity Medicine* (2019) **14**. DOI: 10.1016/j.obmed.2019.100092
13. Khorsandi M., Ghofranipour F., Hidarnia A., Faghihzadeh S., Ghobadzadeh M.. **The effect of PRECEDE PROCEED model combined with the health belief model and the theory of self‐efficacy to increase normal delivery among nulliparous women**. *Procedia‐Social and Behavioral Sciences* (2012) **46** 187-194. DOI: 10.1016/j.sbspro.2012.05.091
14. Kołota A., Głąbska D.. **Analysis of food habits during pandemic in a polish population‐based sample of primary school adolescents: Diet and activity of youth during covid‐19 (day‐19) study**. *Nutrients* (2021) **13** 3711. DOI: 10.3390/nu13113711
15. Labyak C. A., Kaplan L. G., Johnson T. M., Moholland M.. **Practical school nutrition program may reduce food neophobia**. *Nutrients* (2021) **13** 1-14. DOI: 10.3390/nu13103541
16. Lee G. Y., Lee D. Y.. **The sexual risk behaviors of middle school students according to school nurse placement levels in Korea**. *Asian Nursing Research* (2020) **14** 212-220. DOI: 10.1016/j.anr.2020.08.001
17. Liu K. S. N., Chen J. Y., Ng M. Y. C., Yeung M. H. Y., Bedford L. E., Lam C. L. K.. **How does the family influence adolescent eating habits in terms of knowledge, attitudes and practices? A global systematic review of qualitative studies**. *Nutrients* (2021) **13** 3717. DOI: 10.3390/nu13113717
18. Marques A., Loureiro N., Avelar‐Rosa B., Naia A., de Matos M. G.. **Adolescents' healthy lifestyle**. *Jornal de Pediatria* (2020) **96** 217-224. DOI: 10.1016/j.jped.2018.09.002
19. Maximova K., Khan M. K., Dabravolskaj J., Maunula L., Ohinmaa A., Veugelers P. J.. **Perceived changes in lifestyle behaviours and in mental health and wellbeing of elementary school children during the first COVID‐19 lockdown in Canada**. *Public Health* (2021) **202** 35-42. DOI: 10.1016/j.puhe.2021.10.007
20. Melián‐Fleitas L., Franco‐Pérez Á., Caballero P., Sanz‐Lorente M., Wanden‐Berghe C., Sanz‐Valero J.. **Influence of nutrition, food and diet‐related interventions in the workplace: A meta‐analysis with meta‐regression**. *Nutrients* (2021) **13** 3945. DOI: 10.3390/nu13113945
21. Mosavi F., Aliakbari F., Rabiei L.. **Effect of education based on “PRECEDE” model on self‐care behavior in hemodialysis patients**. *Journal of Education and Health Promotion* (2020) **9** 69. DOI: 10.4103/jehp.jehp_534_19
22. Motevalli M., Drenowatz C., Tanous D. R., Khan N. A., Wirnitzer K.. **Management of childhood obesity—Time to shift from generalized to personalized intervention strategies**. *Nutrients* (2021) **13** 1200. DOI: 10.3390/nu13041200
23. Nejhaddadgar N., Darabi F., Rohban A., Solhi M., Kheire M.. **Effectiveness of self‐management program for people with type 2 diabetes mellitus based on PRECEDE PROCEED model**. *Diabetes and Metabolic Syndrome: Clinical Research and Reviews* (2019) **13** 440-443. DOI: 10.1016/j.dsx.2018.08.016
24. Ochola S., Masibo P. K.. **Dietary intake of schoolchildren and adolescents in developing countries**. *Annals of Nutrition & Metabolism* (2014) **64** 24-40. DOI: 10.1159/000365125
25. Pereira A. R., Oliveira A.. **Dietary interventions to prevent childhood obesity: A literature review**. *Nutrients* (2021) **13** 1-17. DOI: 10.3390/nu13103447
26. Pourhaji F., Delshad M. H., Tavafian S. S., Niknami S., Pourhaji F.. **Effects of educational program based on precede‐proceed model in promoting low back pain behaviors (EPPLBP) in health care workers Shahid Beheshti university of medical sciences: Randomized trial**. *Heliyon* (2020) **6**. DOI: 10.1016/j.heliyon.2020.e05236
27. Sezgin D., Esin M. N.. **Effects of a PRECEDE‐PROCEED model based ergonomic risk management programme to reduce musculoskeletal symptoms of ICU nurses**. *Intensive and Critical Care Nursing* (2018) **47** 89-97. DOI: 10.1016/j.iccn.2018.02.007
28. Sirasa F., Mitchell L. J., Rigby R., Harris N.. **Family and community factors shaping the eating behaviour of preschool‐aged children in low and middle‐income countries: A systematic review of interventions**. *Preventive Medicine* (2019) **129**. DOI: 10.1016/j.ypmed.2019.105827
29. Solhi M., Hamedan M. S., Salehi M.. **A PRECEDE‐PROCEED based educational intervention in quality of life of women‐headed households in Iran**. *Medical Journal of the Islamic Republic of Iran* (2016) **30** 417. PMID: 28210582
30. Xi Y., Lin Q., Yang Q., Li F., Liu H., Luo J., Ouyang Y., Sun M., Yong C., Xiang C., Zheng W.. **Association between free sugars intake and excessive daytime sleepiness among Chinese adolescents**. *Nutrients* (2021) **13** 3959. DOI: 10.3390/nu13113959
|
---
title: Establishment and Validation of Predictive Model of Tophus in Gout Patients
authors:
- Tianyi Lei
- Jianwei Guo
- Peng Wang
- Zeng Zhang
- Shaowei Niu
- Quanbo Zhang
- Yufeng Qing
journal: Journal of Clinical Medicine
year: 2023
pmcid: PMC10002994
doi: 10.3390/jcm12051755
license: CC BY 4.0
---
# Establishment and Validation of Predictive Model of Tophus in Gout Patients
## Abstract
[1] Background: A tophus is a clinical manifestation of advanced gout, and in some patients could lead to joint deformities, fractures, and even serious complications in unusual sites. Therefore, to explore the factors related to the occurrence of tophi and establish a prediction model is clinically significant. [ 2] Objective: to study the occurrence of tophi in patients with gout and to construct a predictive model to evaluate its predictive efficacy. [ 3] Methods: The clinical data of 702 gout patients were analyzed by using cross-sectional data of North Sichuan Medical College. The least absolute shrinkage and selection operator (LASSO) and multivariate logistic regression were used to analyze predictors. Multiple machine learning (ML) classification models are integrated to analyze and identify the optimal model, and Shapley Additive exPlanations (SHAP) interpretation was developed for personalized risk assessment. [ 4] Results: Compliance of urate-lowering therapy (ULT), Body Mass Index (BMI), course of disease, annual attack frequency, polyjoint involvement, history of drinking, family history of gout, estimated glomerular filtration rate (eGFR), and erythrocyte sedimentation rate (ESR) were the predictors of the occurrence of tophi. Logistic classification model was the optimal model, test set area under curve (AUC) ($95\%$ confidence interval, CI): 0.888 (0.839–0.937), accuracy: 0.763, sensitivity: 0.852, and specificity: 0.803. [ 5] Conclusions: We constructed a logistic regression model and explained it with the SHAP method, providing evidence for preventing tophus and guidance for individual treatment of different patients.
## 1. Introduction
Gout is an inflammatory disease caused by the deposition of monosodium urate (MSU) crystals in joint and non-joint structures [1]. Patients with gout experience a variety of symptoms, including severe pain, acute and persistent inflammatory arthritis, and symptoms associated with chronic disease [2]. As gout progresses, clinical symptoms of advanced disease characterized by a tophus may appear, primarily the recurrence of chronic granulomas resulting from a continuous deposition of MSU [3,4]. Formation of a tophus can lead to joint deformities and associated joint injury, fracture and skin rupture, or infection [5,6]. In addition, tophi can occur in unusual sites (such as the heart valve, carpal canal, larynx, and spine) and cause complications. Growing studies indicate that multiple factors may influence the development of tophi, such as course of disease, estimated glomerular filtration rate (eGFR), compliance of urate-lowering therapy (ULT), etc. [ 7,8]. The main treatment for tophi is pharmaceutical intervention including purine and non-purine xanthine oxidase inhibitors, uric acid excretion agents, uric acid enzymes, and whole-human anti-IL-1β monoclonal antibodies, as well as other interventions such as surgical removal [9,10,11,12]. However, if the current treatment regimen is not effective, the presence of a tophus can lead to significant complications. Therefore, an early detection of risk factors and establishment of a prediction model has great significance to improve early prevention of tophi. Machine learning (ML) is an emerging field of medicine that represents a powerful set of algorithms capable of representing, adapting, learning, predicting, and analyzing data; moreover, ML is considered as the future of biomedical research, personalized medicine, and computer-aided diagnosis [13,14]. Therefore, this study used a variety of ML classification models to build a prediction model [15,16]. Through collecting and sorting the clinical data of gout patients, influencing factors of tophus were analyzed to provide clinical evidence for early treatment of tophus formation. The application of the ML model has shown accurate individual prediction and promising clinical application prospects [17]. However, its application in real clinical practice and interpretable evidence for risk prediction models are limited [18,19]; therefore, we also used the Shapley Additive exPlanations (SHAP) interpretation tool to provide an intuitive explanation of risk leading to patient predictions [20]. The tool can generate individual probabilities of clinical events by integrating determinants, and it also meets the need for a combination of biological and clinical models, contributing to the development of personalized medicine. The aim of this study is to establish a more suitable clinical study of tophi in gout patients, and establish a corresponding predictive model. It is helpful to improve the diagnosis system of the tophus and provide more reference for clinicians.
## 2.1.1. Subjects
A total of 792 gout patients from 1 January 2018 to 30 June 2022 attended Rheumatic Immunity Department, Affiliated Hospital of North Sichuan Medical College.
## 2.1.2. Inclusion Criteria
Inclusion criteria are as follows: [1] compliance with the diagnostic criteria for gout established by ACR/EULAR in 2015; [2] informed consent and voluntary participation in the study; and [3] having complete clinical data.
## 2.1.3. Exclusion Standards
Exclusion criteria are as follows: [1] those with serious diseases such as chronic cardiac insufficiency, liver diseases, malignant tumors, and mental diseases; [2] intake of certain drugs (such as diuretics, aspirin, cytotoxic drugs, antituberculosis drugs, etc.); and [3] patients who were unable to cooperate, unwilling to participate, or whose clinical data were incomplete.
## 2.2.1. Grouping Methods and Diagnosis of Tophi
All the patients were divided into two groups. Diagnosis of tophi: Light yellow or white uplift or neoplasm of various sizes, hard subcutaneous lump, or/and color, Doppler ultrasonography showing evidence of dual-track sign, tophus, bone erosion, or/and X-ray computed tomography showing high-density lump. These were comprehensively evaluated in combination with patients’ clinical histories.
## 2.2.2. Study Indicators
There were 43 variables: [1] General data, including tophus, sex, compliance of ULT, Body Mass Index (BMI), course of disease, annual attack frequency, polyjoint involvement, history of drinking, history of smoking, history of hypertension, history of high altitude residence, history of high purine diet, history of sugary diet, history of diabetes, history of hyperlipidemia, history of kidney stones, history of kidney crystallization, family history of gout. [ 2] Laboratory examination, including eGFR, Erythrocyte sedimentation rate (ESR), White cell rate (WBC), Granulocte (GR), Lymphocyte (LY), Monocyte (MO), Red blood cell (RBC), Hemoglobin (HGB), Hematocrit (HCT), Mean Corpuscular Volume (MCV), Mean corpuscular hemoglobin concentration (MCHC), Platelet (PLT), Mean platelet volume (MPV), Plateletcrit (PCT), Platelet distribution width (PDW), Uric acid (UA), Urea, Creatinine (Crea), Alanine aminotransferase (ALT), Aspartate aminotransferase (AST), serum albumin (ALB), globulin (GLOB), Cystatin C (CysC), Urine Ph, urine specific gravity.
## 2.2.3. Construction and Evaluation of Predictive Models
After selecting characteristic factors from all independent variables, we divided gout patients into training set and testing set. Multiple ML classification models were applied for comprehensive analysis, comparison on the importance of each index in training set and testing set of different models. Furthermore, we utilized the optimal model to evaluate and verify the results. The SHAP presentation model as a whole and single sample interpretation were also developed. Detailed steps were as follows: [1] Screening characteristic factors: First, R software (glmnet4.1.2) was used to conduct the least absolute shrinkage and selection operator (LASSO) regression analysis and adjusting the variable screening and complexity. Then, LASSO regression analysis results were used to conduct multifactor logistic regression analysis with SPSS, and finally, we obtained the characteristic factors of $p \leq 0.05.$ [ 2] Data division: Pyskthon (0.22.1) random number method was used to randomly divide the gout patients into training set and test set according to the ratio of 7:3, of which 491 were in the training set and 211 were in the testing set. [ 3] Classified multi-model comprehensive analysis: eXtreme Gradient Boosting (XGBoost), Logistic regression, Light Gradient Boosting Machine (LightGBM), RandomForest, Adaptive Boostint (AdBoost), Multilayer Perceptron (MLP), support vector machine (SVM), K-Nearest-Neighbors (KNN), Gaussian Naïve Bayes (GNB) were built by using python (sklearn 0.22.1, xgboost 1.2.1, lightgbm 3.2.1). We then trained and tested the above parameter model (Repeat 10 samples), analyzed the importance of the training set and testing set indicators in different models, and selected the optimal model. Python (sklearn 0.22.1) was used to construct the area under the receiver operating characteristics (ROC) curve and is often used to describe tools for diagnostic testing or the identification accuracy of predictive models [21]. R software (rmda 1.6) was used to plot the decision curve analysis (DCA) that is essentially the decision analysis. Thus, it is possible to decide whether to use one model, or which one of several models was the optimal, with significant advantages in assessing the clinical applicability of the model [22]. Python (sklearn 0.22.1) calibration curves were used to measure the model’s prediction power, and comprehensive assessment of the predictive model was employed to validate its usefulness in decision support or more general simulation modeling [23]. Python (sklearn 0.22.1) was used to plot precision recall (PR) curves, which were widely used to evaluate the performance of models. PR and area under PR (AP) curve can provide a valuable complement to existing model evaluation methods [24]. [ 4] Training, verification, and testing of the optimal model: the training set was cross-verified with 10 folds and evaluated with the testing set. Python (sklearn 0.22.1) Draw learning curves were used to evaluate the model fit and stability of training and validation sets [25]. [ 5] Python (shap 0.39.0) was used to draw the SHAP interpretation of importance and contribution to the model and interpret the model results by calculating the contribution of each feature to the predicted results. In addition, the SHAP was built for a single sample and tries to calculate the prediction performance [26].
## 3. Statistical Analysis
Variables were all included in comparison between training and testing sets. Continuous variables were expressed as median and Inter-Quartile Range (IQR) and compared using the Mann–Whitney U-test. Categorical variables were expressed in number and percentage and compared using chi-square tests. Bilateral p values less than 0.05 were considered statistically significant. SPSS (version 25.0), R (version 3.6.1), and Python (version 3.4.3) were used for statistical analysis.
## 4.1. Comparison of Baseline Data
In this study, we excluded a total of 90 gout patients with other serious diseases, See Supplementary Data S1. Regarding the analysis of 702 gout patients, all variables were investigated at the initial diagnosis, and the compliance of ULT was defined as poor compliance if the medication possession ratio (MPR) [27,28,29,30] was lower than $60\%$ and as high compliance if the MPR ≥ $60\%$. The annual attack frequency could be divided into the severity of at least 12 times per year and less than 6 times per year, with less than 6 as low degree, 6–12 as medium degree, and more than 12 as high degree. History of drinking was defined by no history of drinking, drinking less than 70 g per week as moderate history of drinking, drinking ≥ 70 g per week and drinking years ≥ 10 years as excessive drinking. Polyjoint involvement was defined by the presence of a tophus above three joints. The specific baseline data of the final training set and the test set are shown in Table 1. There was no significant difference between the two groups ($p \leq 0.05$).
## 4.2. Screening of Characteristic Factors for Risk of Tophi in Gout Patients
LASSO regression analysis was conducted on the remaining independent variables with presence of a tophus as the dependent variable (Figure 1). LASSO can compress variable coefficients to prevent overfitting and solve severe collinearity problems [31]. The results showed that (lambda with minimum mean square error = 0.024) 42 independent variables were reduced to 11, including sex, compliance of ULT, BMI, course of disease, annual attack frequency, history of drinking, family history of gout, polyjoint involvement, eGFR, ESR, and UA. To further control the influence of confounding factors, the above 11 independent variables were analyzed using multivariate logistic regression [32]. Finally, only compliance of ULT, BMI, course of disease, annual attack frequency (>12 times), history of drinking (drinking ≥ 70 g per week/drinking years ≥ 10 years), family history of gout, polyjoint involvement, eGFR, and ESR were determined as characteristic factors ($p \leq 0.05$), as Table 2.
## 4.3. Comprehensive Analysis of Classified Multi-Model
XGBoost, Logistic, LightGBM, RandomForest, AdaBoost, MLP, SVM, KNN, and GNB were trained and repeated 10 times. The model was evaluated using area under curve (AUC) values [21], and the results indicated that XGBoost, LightGBM, and RandomForest were the highest in the training set and Logistic was the highest in the testing set (Figure 2a,b); see more details in Supplemental Table S1. The AUC indicator focuses on the predictive accuracy of the model and does not tell whether the model is clinically usable or which one of the two is more preferable [21,33]. Therefore, the DCA, calibration curves, and PR curve were analyzed. The DCA evaluates Logistic and RandomForest for a better clinical suitability (Figure 2c). Calibration curves showed a higher accuracy of GNB and Logistic model predictions (Figure 2d). In training and test sets, the Logistic model showed the optimal performance, with the highest AP value in the test set (Figure 2e,f). Comprehensive analysis demonstrated that Logistic could be considered the optimal model.
## 4.4. The Best Model Building and Evaluation
Logistic regression analysis and 10-fold cross validation were performed on the training set. The results showed that the average AUC of the training set was 0.876 (0.838–0.914), the average AUC of the verification set was 0.854 (0.733–0.972), and the AUC of the testing set was 0.888 (0.839–0.937) (Figure 3a–c). The AUC of the training set, the verification set, and the testing set was finally stable at about 0.85, and the model prediction effect was accurate. As the performance of the verification set under the AUC index was lower than the test set or the ratio was lower than $10\%$, the model fitting could be considered successful, and the learning curve indicated that the training set and the verification set had a strong fitting and high stability [25] (Figure 3d). These results indicated that the logistic regression model could be used for the classification modeling task of the data set.
## 4.5. The SHAP to Model Interpretation
To visually explain the selected variables, we used SHAP to illustrate how these variables predicted the formation of a tophus in the model [26]. Figure 4a shows the nine most important features in our model. In each feature important line, the attributions of all patients to the results are plotted with different colored dots, where red dots represent high risk values and blue dots represent low risk values. Decreased BMI and compliance of ULT (MRP < $60\%$), longer course of disease, high annual attack frequency (>12 times), history of excessive drinking, family history of gout, polyjoint involvement, decreased eGFR, and increased ESR would elevate the formation of tophi in gout patients. Figure 4b shows the ranking of nine risk factors evaluated by the average absolute SHAP value, with the x-axis SHAP value indicating the importance of the forecast model. In addition, we provided two typical examples to illustrate the interpretability of the model, one was a gout patient without a tophus with a low SHAP predictive score (0.133) (Figure 4c), while another gout patient with a tophus had a higher SHAP score (0.722) (Figure 4d).
## 5. Discussion
In this study, we excluded a total of 90 patients; of these patients, only one had heart disease, two had liver damage, and one had lung cancer. The prevalence of tophus was about $4.4\%$. The risk of hyperuricemia in heart disease is high and may be due to decreased renal perfusion and UA excretion [34,35,36]. Elevated levels of xanthine oxidase (XO) were also reported in patients with heart failure [37]. In addition, some patients with decompensated heart failure (DHF) develop sodium retention that stimulates renal urate anion exchangers that affect UA [38]. Diuretic doses are usually higher than baseline doses, resulting in reduced UA excretion and possible hyperuricemia [39]. The liver is the main site of UA biosynthesis. XO participates in the formation of UA and may releases XO after impaired liver function [40]. Most patients with advanced liver disease have hypoproteinemia. It should be noted that the presence of carboxylic acid groups in albumin is necessary for the positive effect of albumin on MSU nucleation [41]. Chemotherapy in cancer patients can lead to increased cell destruction, significantly raising UA levels, which in turn can lead to gout [42]. In the case of mental illness, this part of the population is excluded because it is unable to provide regular outpatient care. At the same time, gout treatment drugs are harmful to the liver. In our investigation, many of the liver diseases associated with gout patients were not treated properly, which may interfere with our study of tophus formation in gout patients. In addition, there are drugs that can cause hyperuricemia, which can lead to gout symptoms [43]. The gout manifestations in these patients may be transient, so we do not consider them a risk factor for tophi.
Our results show that nine clinical characteristic variables were screened by LASSO and multivariate logistic regression analysis from 42 clinical variables (compliance of ULT, course of disease, polyjoint involvement, history of drinking (drinking ≥ 70 g per week/years of drinking ≥ 10 years), eGFR, annual attack frequency (>12 times), BMI, ESR, and family history of gout to assess the risk of tophi in patients. About $25\%$ of gout patients in our study developed a tophus. Several studies have reported predictive risk factors as the clinical presentation for tophus patients [7,44,45,46,47]. For example, a Chinese retrospective study has shown that disease duration and joint involvement in the upper extremities affected joints and kidney stones, and that hypertension is a risk factor for the development of subcutaneous tophi, while BMI may be a protective factor for tophus [44]. Beilei Lu et al. reported a lower eGFR and a longer disease duration as independent risk factors for tophus formation in gout patients. Double Profile Sonography (DCS) was higher in patients with tophus than those without [7]. Another study has shown that age and DCS are potential risk factors for tophi [45]. A simple study of metabolic markers associated with tophi has shown that UA, eGFR, γ-GT, and ALT are related to tophi, and that the γ-GT/ALT ratio can be used as a predictor or monitor of tophi [47]. A recent study reported that high serum free fatty acid level is independently correlated with risk of tophi, which may promote tophus deposition by lowering urine pH [46].These findings often rely on data labeled by human experts. Despite the differences, it is indicated that course of disease, eGFR, and DCS may play a more significant role. Unfortunately, this study did not include clinical manifestations of joint ultrasound in gout patients. In our study, the course of disease and the role of eGFR in the formation of tophus were consistent.
In addition, we found clinical factors that might influence the formation of tophi, such as compliance of ULT, polyjoint involvement, history of drinking, annual attack frequency, BMI, ESR, and family history of gout. In this study, compliance of ULT was considered the most significant predictor. A multicenter prospective study reported that ultrasound monitored a reduction in urate deposition after ULT in gout [8]. Another prospective study also found a gradual reduction in the size of tophi after lesinurad plus fipronil treatment [48]. The 2012 American Academy of Rheumatology Gout Management Guidelines recommends ULT as an initial treatment for gout with tophi [49]. Reasonable ULT can reduce serum UA level and pathological MSU deposition [50]. Alcohol is an important risk factor for gout. Ethanol consumes ATP, increases lactic acid production, increases UA production, and reduces UA excretion from the kidneys [51]. BMI was positively correlated with body temperature, probably because of the thicker subcutaneous fat tissue and better thermal insulation [52,53]. However, lower temperatures result in lower urate solubility [54]. In a large data analysis, both men and women had U-shaped UA–BMI relationships, which was positively correlated with a BMI of 20 kg/m2 and negatively correlated with a BMI of 20 kg/m2 [55]. Interestingly, in elderly patients, BMI was positively associated with quadriceps muscle mass [56]. Albumin was positively correlated with muscle mass in males, and negatively correlated with muscle mass in females [57]. Albumin is a large molecule that may increase the solubility of UA [58,59]. Elevated levels of hyaluronic acid in the blood of obese people can lead to a slight increase in urine solubility [60,61]. Another cross-sectional study from China also suggests that BMI may be a protective factor. This evidence suggests that those with lower BMI may be more likely to form tophi [44]. Polyjoint involvement, annual attack frequency, and more ESR may indicate the frequency and severity of acute spasms, reflecting the potential deposition of MSU. At the same time, genetic factors and immune status may also affect the formation of tophi.
Although there are many risk factors for tophi, no predictive model has been established. In this study, we used several ML models, and found that the logistic regression model performed better than other ML models after analyzing the AUC, DCA, calibration curves, and PR curves. However, it has always been a challenge to interpret the ML prediction model more comprehensively and to visually present the predictive results to clinicians. Therefore, we applied the SHAP method to the logistic regression model to achieve the optimal predictive effect and interpretability. We identified some important variables associated with the development of tophi in gout patients.
However, our study has several limitations. Firstly, there was no gold standard inclusion or exclusion criteria for tophi. Secondly, the sample size was relatively small in the study; the data were collected in a single institution, it was not a multi-center study. Therefore, the results were of limited generality. Furthermore, although a high consistency was achieved in the repeatability analysis within the training and testing set, some inevitable errors may occur due to segmentation uncertainty. Finally, the design of the study did not include some variables such as 24 h quantitative UA and joint ultrasound in the analysis. Longitudinal or prospective case controlled studies are also needed to further explain the relationship between risk factors and tophus formation.
## 6. Conclusions
In conclusion, this study constructed a predictive model based on the ML model, and the logistic regression model showed a better performance in this study. In addition, we provided a personalized risk assessment for the development of tophi in gout patients explained by SHAP. This effective computer-aided approach can help first-line clinicians and patients identify and intervene in the occurrence of tophi.
## References
1. Dalbeth N., Gosling A.L., Gaffo A., Abhishek A.. **Gout**. *Lancet* (2021) **397** 1843-1855. DOI: 10.1016/S0140-6736(21)00569-9
2. Singh J.A.. **The impact of gout on patient’s lives: A study of African-American and Caucasian men and women with gout**. *Arthritis Res. Ther.* (2014) **16** R132. DOI: 10.1186/ar4589
3. Dalbeth N., Stamp L.. **Hyperuricaemia and gout: Time for a new staging system?**. *Ann. Rheum. Dis.* (2014) **73** 1598-1600. DOI: 10.1136/annrheumdis-2014-205304
4. Grassi W., De Angelis R.. **Clinical features of gout**. *Reumatismo* (2012) **63** 238-245. DOI: 10.4081/reumatismo.2011.238
5. Aati O., Taylor W.J., Horne A., Dalbeth N.. **Toward development of a Tophus Impact Questionnaire: A qualitative study exploring the experience of people with tophaceous gout**. *J. Clin. Rheumatol.* (2014) **20** 251-255. DOI: 10.1097/RHU.0000000000000127
6. Khanna P.P., Nuki G., Bardin T., Tausche A.K., Forsythe A., Goren A., Vietri J., Khanna D.. **Tophi and frequent gout flares are associated with impairments to quality of life, productivity, and increased healthcare resource use: Results from a cross-sectional survey**. *Health Qual. Life Outcomes* (2012) **10** 117. DOI: 10.1186/1477-7525-10-117
7. Lu B., Lu Q., Huang B., Li C., Zheng F., Wang P.. **Risk factors of ultrasound-detected tophi in patients with gout**. *Clin. Rheumatol.* (2020) **39** 1953-1960. DOI: 10.1007/s10067-020-04947-2
8. Ebstein E., Forien M., Norkuviene E., Richette P., Mouterde G., Daien C., Ea H.K., Brière C., Lioté F., Petraitis M.. **Ultrasound evaluation in follow-up of urate-lowering therapy in gout: The USEFUL study**. *Rheumatology* (2019) **58** 410-417. DOI: 10.1093/rheumatology/key303
9. Sriranganathan M.K., Vinik O., Pardo Pardo J., Bombardier C., Edwards C.J.. **Interventions for tophi in gout**. *Cochrane Database Syst. Rev.* (2021) **8** Cd010069. DOI: 10.1002/14651858.CD010069.pub3
10. Schlesinger N.. **Difficult-to-treat gouty arthritis: A disease warranting better management**. *Drugs* (2011) **71** 1413-1439. DOI: 10.2165/11592290-000000000-00000
11. Mandell B.F., Yeo A.E., Lipsky P.E.. **Tophus resolution in patients with chronic refractory gout who have persistent urate-lowering responses to pegloticase**. *Arthritis Res. Ther.* (2018) **20** 286. DOI: 10.1186/s13075-018-1782-x
12. Doscher M., Beutel B.G., Lovy A., Alba B., Melone C.. **En Masse Excision and Curettage for Periarticular Gouty Tophi of the Hands**. *Eplasty* (2022) **22** e25. PMID: 36000011
13. Greener J.G., Kandathil S.M., Moffat L., Jones D.T.. **A guide to machine learning for biologists**. *Nat. Rev. Mol. Cell Biol.* (2022) **23** 40-55. DOI: 10.1038/s41580-021-00407-0
14. Handelman G.S., Kok H.K., Chandra R.V., Razavi A.H., Lee M.J., Asadi H.. **eDoctor: Machine learning and the future of medicine**. *J. Intern. Med.* (2018) **284** 603-619. DOI: 10.1111/joim.12822
15. Moons K.G., Royston P., Vergouwe Y., Grobbee D.E., Altman D.G.. **Prognosis and prognostic research: What, why, and how?**. *BMJ* (2009) **338** b375. DOI: 10.1136/bmj.b375
16. Adams S.T., Leveson S.H.. **Clinical prediction rules**. *BMJ* (2012) **344** d8312. DOI: 10.1136/bmj.d8312
17. Churpek M.M., Yuen T.C., Winslow C., Meltzer D.O., Kattan M.W., Edelson D.P.. **Multicenter Comparison of Machine Learning Methods and Conventional Regression for Predicting Clinical Deterioration on the Wards**. *Crit. Care Med.* (2016) **44** 368-374. DOI: 10.1097/CCM.0000000000001571
18. Shortliffe E.H., Sepúlveda M.J.. **Clinical Decision Support in the Era of Artificial Intelligence**. *JAMA* (2018) **320** 2199-2200. DOI: 10.1001/jama.2018.17163
19. Vayena E., Blasimme A., Cohen I.G.. **Machine learning in medicine: Addressing ethical challenges**. *PLoS Med.* (2018) **15**. DOI: 10.1371/journal.pmed.1002689
20. Lundberg S.M., Erion G., Chen H., DeGrave A., Prutkin J.M., Nair B., Katz R., Himmelfarb J., Bansal N., Lee S.I.. **From Local Explanations to Global Understanding with Explainable AI for Trees**. *Nat. Mach. Intell.* (2020) **2** 56-67. DOI: 10.1038/s42256-019-0138-9
21. Obuchowski N.A., Bullen J.A.. **Receiver operating characteristic (ROC) curves: Review of methods with applications in diagnostic medicine**. *Phys. Med. Biol.* (2018) **63** 07tr01. DOI: 10.1088/1361-6560/aab4b1
22. Vickers A.J., Elkin E.B.. **Decision curve analysis: A novel method for evaluating prediction models**. *Med. Decis. Mak.* (2006) **26** 565-574. DOI: 10.1177/0272989X06295361
23. Fenlon C., O’Grady L., Doherty M.L., Dunnion J.. **A discussion of calibration techniques for evaluating binary and categorical predictive models**. *Prev. Vet. Med.* (2018) **149** 107-114. DOI: 10.1016/j.prevetmed.2017.11.018
24. Li W., Guo Q.. **Plotting receiver operating characteristic and precision-recall curves from presence and background data**. *Ecol. Evol.* (2021) **11** 10192-10206. DOI: 10.1002/ece3.7826
25. Belkin M., Hsu D., Ma S., Mandal S.. **Reconciling modern machine-learning practice and the classical bias-variance trade-off**. *Proc. Natl. Acad. Sci. USA* (2019) **116** 15849-15854. DOI: 10.1073/pnas.1903070116
26. Lundberg S.M., Lee S.I.. **A unified approach to interpreting model predictions**. *Advances in Neural Information Processing Systems 30, Proceedings of the 31st International Conference on Neural Information Processing Systems, Long Beach, CA, USA, 4–9 December 2017* (2017) 4765-4774
27. Yin R., Li L., Zhang G., Cui Y., Zhang L., Zhang Q., Fu T., Cao H., Li L., Gu Z.. **Rate of adherence to urate-lowering therapy among patients with gout: A systematic review and meta-analysis**. *BMJ Open* (2018) **8** e017542. DOI: 10.1136/bmjopen-2017-017542
28. Scheepers L., van Onna M., Stehouwer C.D.A., Singh J.A., Arts I.C.W., Boonen A.. **Medication adherence among patients with gout: A systematic review and meta-analysis**. *Semin. Arthritis Rheum.* (2018) **47** 689-702. DOI: 10.1016/j.semarthrit.2017.09.007
29. McGowan B., Bennett K., Silke C., Whelan B.. **Adherence and persistence to urate-lowering therapies in the Irish setting**. *Clin. Rheumatol.* (2016) **35** 715-721. DOI: 10.1007/s10067-014-2823-8
30. Perez-Ruiz F., Perez-Herrero N., Richette P., Stack A.G.. **High Rate of Adherence to Urate-Lowering Treatment in Patients with Gout: Who’s to Blame?**. *Rheumatol. Ther.* (2020) **7** 1011-1019. DOI: 10.1007/s40744-020-00249-w
31. Sauerbrei W., Royston P., Binder H.. **Selection of important variables and determination of functional form for continuous predictors in multivariable model building**. *Stat. Med.* (2007) **26** 5512-5528. DOI: 10.1002/sim.3148
32. Pourhoseingholi M.A., Baghestani A.R., Vahedi M.. **How to control confounding effects by statistical analysis**. *Gastroenterol. Hepatol. Bed Bench* (2012) **5** 79-83. PMID: 24834204
33. Muschelli J.. **ROC and AUC with a Binary Predictor: A Potentially Misleading Metric**. *J. Classif.* (2020) **37** 696-708. DOI: 10.1007/s00357-019-09345-1
34. Leyva F., Anker S., Swan J.W., Godsland I.F., Wingrove C.S., Chua T.P., Stevenson J.C., Coats A.J.. **Serum uric acid as an index of impaired oxidative metabolism in chronic heart failure**. *Eur. Heart J.* (1997) **18** 858-865. DOI: 10.1093/oxfordjournals.eurheartj.a015352
35. Sakai H., Tsutamoto T., Tsutsui T., Tanaka T., Ishikawa C., Horie M.. **Serum level of uric acid, partly secreted from the failing heart, is a prognostic marker in patients with congestive heart failure**. *Circ. J.* (2006) **70** 1006-1011. DOI: 10.1253/circj.70.1006
36. Von Lueder T.G., Girerd N., Atar D., Agewall S., Lamiral Z., Kanbay M., Pitt B., Dickstein K., Zannad F., Rossignol P.. **Serum uric acid is associated with mortality and heart failure hospitalizations in patients with complicated myocardial infarction: Findings from the High-Risk Myocardial Infarction Database Initiative**. *Eur. J. Heart Fail.* (2015) **17** 1144-1151. DOI: 10.1002/ejhf.419
37. Otaki Y., Watanabe T., Kinoshita D., Yokoyama M., Takahashi T., Toshima T., Sugai T., Murase T., Nakamura T., Nishiyama S.. **Association of plasma xanthine oxidoreductase activity with severity and clinical outcome in patients with chronic heart failure**. *Int. J. Cardiol.* (2017) **228** 151-157. DOI: 10.1016/j.ijcard.2016.11.077
38. Enomoto A., Kimura H., Chairoungdua A., Shigeta Y., Jutabha P., Cha S.H., Hosoyamada M., Takeda M., Sekine T., Igarashi T.. **Molecular identification of a renal urate anion exchanger that regulates blood urate levels**. *Nature* (2002) **417** 447-452. DOI: 10.1038/nature742
39. Ritter F., Franzeck F., Geisshardt J., Walker U.A., Osthoff M.. **Gout Arthritis During Admission for Decompensated Heart Failure-A Descriptive Analysis of Risk Factors, Treatment and Prognosis**. *Front. Med.* (2022) **9** 789414. DOI: 10.3389/fmed.2022.789414
40. Furuhashi M., Mori K., Tanaka M., Maeda T., Matsumoto M., Murase T., Nakamura T., Koyama M., Moniwa N., Ohnishi H.. **Unexpected high plasma xanthine oxidoreductase activity in female subjects with low levels of uric acid**. *Endocr. J.* (2018) **65** 1083-1092. DOI: 10.1507/endocrj.EJ18-0127
41. Perl-Treves D., Addadi L.. **A structural approach to pathological crystallizations. Gout: The possible role of albumin in sodium urate crystallization**. *Proc. R. Soc. Lond. B Biol. Sci.* (1988) **235** 145-159. DOI: 10.1098/rspb.1988.0069
42. Gelber A.C.. **Among the Company That Gout Keeps, Is Cancer on the List?**. *J. Rheumatol.* (2021) **48** 1358-1360. DOI: 10.3899/jrheum.210138
43. Leung N., Yip K., Pillinger M.H., Toprover M.. **Lowering and Raising Serum Urate Levels: Off-Label Effects of Commonly Used Medications**. *Mayo Clin. Proc.* (2022) **97** 1345-1362. DOI: 10.1016/j.mayocp.2022.02.027
44. Ma L., Sun R., Jia Z., Zou Y., Xin Y., Cheng X., Liu T., Cui L., Liu Z., Wu X.. **Clinical characteristics associated with subcutaneous tophi formation in Chinese gout patients: A retrospective study**. *Clin. Rheumatol.* (2018) **37** 1359-1365. DOI: 10.1007/s10067-017-3969-y
45. Sun C., Qi X., Tian Y., Gao L., Jin H., Guo H.. **Risk factors for the formation of double-contour sign and tophi in gout**. *J. Orthop. Surg. Res.* (2019) **14** 239. DOI: 10.1186/s13018-019-1280-0
46. Si K., Wei C., Xu L., Lv W., Dong B., Wang Z., Huang Y., Chen Y., Wang Y.. **Association between serum free fatty acid levels and tophus in patients with gout: A cross-sectional study**. *Clin. Exp. Rheumatol.* (2022). DOI: 10.55563/clinexprheumatol/a3i566
47. Liu W., Song H., Man S., Li H., Gao S.. **Simple metabolic markers associated with tophaceous gout**. *Clin. Rheumatol.* (2021) **40** 5047-5053. DOI: 10.1007/s10067-021-05861-x
48. Dalbeth N., Jones G., Terkeltaub R., Khanna D., Fung M., Baumgartner S., Perez-Ruiz F.. **Efficacy and safety during extended treatment of lesinurad in combination with febuxostat in patients with tophaceous gout: CRYSTAL extension study**. *Arthritis Res. Ther.* (2019) **21** 8. DOI: 10.1186/s13075-018-1788-4
49. Khanna D., Fitzgerald J.D., Khanna P.P., Bae S., Singh M.K., Neogi T., Pillinger M.H., Merill J., Lee S., Prakash S.. **2012 American College of Rheumatology guidelines for management of gout. Part 1: Systematic nonpharmacologic and pharmacologic therapeutic approaches to hyperuricemia**. *Arthritis Care Res.* (2012) **64** 1431-1446. DOI: 10.1002/acr.21772
50. Ellmann H., Bayat S., Araujo E., Manger B., Kleyer A., Cavallaro A., Lell M., Schenker H., Simon D., Tascilar K.. **Effects of Conventional Uric Acid-Lowering Therapy on Monosodium Urate Crystal Deposits**. *Arthritis Rheumatol.* (2020) **72** 150-156. DOI: 10.1002/art.41063
51. Nielsen S.M., Zobbe K., Kristensen L.E., Christensen R.. **Nutritional recommendations for gout: An update from clinical epidemiology**. *Autoimmun. Rev.* (2018) **17** 1090-1096. DOI: 10.1016/j.autrev.2018.05.008
52. Waalen J., Buxbaum J.N.. **Is older colder or colder older? The association of age with body temperature in 18,630 individuals**. *J. Gerontol. A Biol. Sci. Med. Sci.* (2011) **66** 487-492. DOI: 10.1093/gerona/glr001
53. Hoffmann M.E., Rodriguez S.M., Zeiss D.M., Wachsberg K.N., Kushner R.F., Landsberg L., Linsenmeier R.A.. **24-h core temperature in obese and lean men and women**. *Obesity* (2012) **20** 1585-1590. DOI: 10.1038/oby.2011.380
54. Loeb J.N.. **The influence of temperature on the solubility of monosodium urate**. *Arthritis Rheum.* (1972) **15** 189-192. DOI: 10.1002/art.1780150209
55. Zhou H., Liu Z., Chao Z., Chao Y., Ma L., Cheng X., Wang Y., Li C., Chen Y.. **Nonlinear relationship between serum uric acid and body mass index: A cross-sectional study of a general population in coastal China**. *J. Transl. Med.* (2019) **17** 389. DOI: 10.1186/s12967-019-02142-9
56. Akazawa N., Kishi M., Hino T., Tsuji R., Tamura K., Hioka A., Moriyama H.. **Higher Body Mass Index in Hospitalized Older Patients Is Related to Higher Muscle Quality**. *J. Nutr. Health Aging* (2022) **26** 495-500. DOI: 10.1007/s12603-022-1785-9
57. Chen Z., Song C., Yao Z., Sun J., Liu W.. **Associations between albumin, globulin, albumin to globulin ratio and muscle mass in adults: Results from the national health and nutrition examination survey 2011–2014**. *BMC Geriatr.* (2022) **22**. DOI: 10.1186/s12877-022-03094-4
58. Kippen I., Klinenberg J.R., Weinberger A., Wilcox W.R.. **Factors affecting urate solubility in vitro**. *Ann. Rheum. Dis.* (1974) **33** 313-317. DOI: 10.1136/ard.33.4.313
59. Katz W.A., Schubert M.. **The interaction of monosodium urate with connective tissue components**. *J. Clin. Investig.* (1970) **49** 1783-1789. DOI: 10.1172/JCI106396
60. Romo M., López-Vicario C., Pérez-Romero N., Casulleras M., Martínez-Puchol A.I., Sánchez B., Flores-Costa R., Alcaraz-Quiles J., Duran-Güell M., Ibarzábal A.. **Small fragments of hyaluronan are increased in individuals with obesity and contribute to low-grade inflammation through TLR-mediated activation of innate immune cells**. *Int. J. Obes.* (2022) **46** 1960-1969. DOI: 10.1038/s41366-022-01187-z
61. Burt H.M., Dutt Y.C.. **Growth of monosodium urate monohydrate crystals: Effect of cartilage and synovial fluid components on in vitro growth rates**. *Ann. Rheum. Dis.* (1986) **45** 858-864. DOI: 10.1136/ard.45.10.858
|
---
title: Highly Specific Detection of Oxytocin in Saliva
authors:
- Muhit Rana
- Nimet Yildirim
- Nancy E. Ward
- Stephanie P. Vega
- Michael J. Heffernan
- Avni A. Argun
journal: International Journal of Molecular Sciences
year: 2023
pmcid: PMC10003004
doi: 10.3390/ijms24054832
license: CC BY 4.0
---
# Highly Specific Detection of Oxytocin in Saliva
## Abstract
Oxytocin is a peptide neurophysin hormone made up of nine amino acids and is used in induction of one in four births worldwide (more than 13 percent in the United States). Herein, we have developed an antibody alternative aptamer-based electrochemical assay for real-time and point-of-care detection of oxytocin in non-invasive saliva samples. This assay approach is rapid, highly sensitive, specific, and cost-effective. Our aptamer-based electrochemical assay can detect as little as 1 pg/mL of oxytocin in less than 2 min in commercially available pooled saliva samples. Additionally, we did not observe any false positive or false negative signals. This electrochemical assay has the potential to be utilized as a point-of-care monitor for rapid and real-time oxytocin detection in various biological samples such as saliva, blood, and hair extracts.
## 1. Introduction
Oxytocin (OT) is a neuropeptide hormone best known for its role in the facilitation of childbirth through the induction of myometrial smooth muscle contractions [1,2]. It also plays an essential role in the later stages of life, affecting various health conditions and complex social behaviors. These include affiliation, sexual behavior, social recognition, social bonding, parturition, lactation, appetite regulation, aggression, depression, obesity, and social deficit of autism spectrum disorder (ASD) [3,4,5,6,7]. Recent scientific evidence indicates that dysfunction of the oxytocin system could be the underlying cause for the pathogenesis of insulin resistance and dyslipidemia and contribute to weight gain in some genetic obesity conditions such as the Prader–Willi syndrome (PWS). The circulating peripheral oxytocin levels were reported to be higher in children with PWS as compared to their healthy siblings; in contrast, the oxytocin levels were lower in individuals with anorexia [5,8,9,10,11,12,13,14,15]. Oxytocin is also involved in regulation of metabolic energy and linked to late-onset obesity in an oxytocin receptor-deficient mice model [12]. Therefore, monitoring oxytocin levels could play a therapeutic role in management of obesity and diabetes. Besides, it has been suggested that oxytocin, known to promote mother-infant bonds, may be implicated in the social deficit of autism [6]. The researchers recently reported that oxytocin levels were significantly lower in individuals with ASD as compared to control subjects [16,17,18]. Some features of ASD have also been linked to disturbance of the oxytocin system in the body [19,20,21]. Exogenous administration of oxytocin has improved various outcomes associated with social responsiveness, including eye contact, emotion recognition, social cognition, and neural circuitry associated with social awareness [17,18,22].
OT is of particular interest in the study of childbearing women, as it has a role in the onset and course of labor and breastfeeding. One in four births worldwide (more than 13 percent in the United States) is induced with oxytocin [23]. Exogenous administration of oxytocin is critical in a clinical setting for induction and augmentation of labor as well as management of postpartum uterine atony/hemorrhage [24,25,26,27,28]. When oxytocin levels are high, strong contractions occur that reduce the chance of bleeding or postpartum hemorrhage. A double-blinded clinical trial on 200 participants showed oxytocin’s role in reducing blood loss during cesarean delivery and the investigators reported that oxytocin infusion is an appropriate regimen [29,30]. Another clinical study by the Cohen Group showed that obese patients required more oxytocin than lean women during the first stage of successful labor induction, indicating that the current clinical practice can benefit from dosage optimization [31].
Oxytocin levels also have great significance during the perinatal period. For example, endogenous oxytocin is a potential biomarker for the prediction of the type of labor and risk assessment of premature labor. Perinatal screening after the 32nd week of pregnancy can help predict premature labor in high-risk pregnancies [32]. Increased levels of circulating peripheral oxytocin levels are linked to postpartum breastmilk production as well as a decrease in the frequency of migraine headaches over the course of pregnancy [33,34,35]. Additionally, a group of physicians and researchers from Northwestern and Indiana University conducted multi-centered clinical research on 66 pregnant women and reported that elevated oxytocin levels during pregnancy may signal postpartum depression (PPD) [36].
Oxytocin also plays a critical role as a neurotransmitter. It mediates the brain’s dynamic function and various complex social behaviors, including affiliation, sexual behavior, social recognition, and aggression. Oxytocin is secreted in the hypothalamus along with a similarly structured nonapeptide called vasopressin. Due to the similar structure of these two neuropeptides, they often compete to bind with existing antibodies, resulting in poor specificity for current immunoassays. Currently, two existing immunoassays (radioimmunoassay-RIA and enzyme immunoassay-EIA) are insufficient for sensitive and specific detection of oxytocin. This problem can be resolved with mass spectrometry combined with liquid chromatography (LC/MS); however, micro dialysis of the sample and lengthy retention times make this method unsuitable for practical oxytocin monitoring.
Despite being one of the most widely utilized drugs in obstetrics, there is currently no instrument capable of point-of-care (POC) detection of oxytocin. While its pharmacokinetics has been extensively studied for both intravascular (IV) and intranasal (IN) administration, its dose-effect response has been poorly understood. The research studies, as well as the clinical significance of perinatal oxytocin, suggest that accurate and real-time measurement of peripheral oxytocin levels may help develop pharmacokinetic models to facilitate a better understanding of the effects of oxytocin and optimize oxytocin use. [ 37,38] Therefore, it would be extremely valuable for researchers and medical professionals to have a simple and practical assay that would accurately determine the peripheral levels of oxytocin in pregnant women and guide clinical plans for oxytocin administration. Clinical researchers could also benefit from the development of such a tool to aid in quantifying peripheral oxytocin levels toward a better understanding of the long-term effects of exogenous oxytocin on mother and child.
The biological levels of oxytocin in bodily fluids such as saliva are very low. The physiological level of peripheral oxytocin is only on the order of 1–300 pg/mL, making it difficult to detect it with high specificity using the current immunoassay-based methods. These antibody-based methods also suffer from significant cross-reactivity by arginine vasopressin, another similar neuropeptide hormone [39,40,41]. Other laboratory-based methods, such as LC-MS/MS, exist but they require complex instrumentation and sample processing steps, increasing the cost and turnaround times.
As illustrated in Figure 1, we have developed an aptamer-based electrochemical assay that enables the measurement of oxytocin in minimally invasive biological samples (e.g., commercially available pooled saliva samples) with high sensitivity and specificity while lowering the detection limits to pg/mL levels. Upon running a rapid (<2 min) electrochemical algorithm, the oxytocin content is quantified. This study demonstrates the electrochemical oxytocin detection in both lab samples and exogenously enriched saliva samples with a limit of detection (LOD) of 1 pg/mL. Table 1 shows the comparative performance of our electrochemical sensor assay with other available technologies, including the commercially available ELISA kit.
## 2.1. Aptamer Development
Aptamers are synthetic, single-stranded DNA or RNA oligonucleotides with very high affinity, selectivity, and specificity to low molecular weight molecules, macromolecules such as proteins, and even whole cells [46]. Aptamers have been generated with binding constants (Kd) to their targets that are in the nanomolar range, comparable to antibody-antigen values. Commonly used bioreceptors (enzymes and antibodies) are mostly unavailable for small peptide targets, especially for short-chain peptides, making aptamers excellent candidates for bimolecular recognition due to the small size of nucleic acids and their versatile in vitro development and synthesis for any targeted peptide. In comparison to antibodies and enzymes, aptamers are also less prone to degradation and denaturation.
Aptamer development has traditionally been via an iterative process called systematic evolution of ligands by exponential enrichment (SELEX); however, this approach has limited aptamer development studies for new targets to academic laboratories or specialized companies. The emergence of oxytocin aptamer as “Raptamers” from Raptamer Discovery Group has been a game changer in this field to allow efficient aptamer development for a wide range of targets. We have employed this non-SELEX strategy in our study to develop the first aptamer specific to the oxytocin molecule. In contrast to SELEX, the Raptamer strategy employs a bead-based library as the basis for the rapid selection of affinity agents for targeted biomarkers with standard laboratory practices. The Raptamer selection process has the advantage of using a single round of PCR amplification; this is in contrast to the multiple rounds of PCR in SELEX which can lead to PCR bias in the aptamer selection. In addition, Raptamer library beads incorporate proprietary modified nucleotides in the random region; these modified bases provide a more functionally diverse composition for enhancement of interactions with target molecules. In our study to develop an oxytocin aptamer, the combinatorial library (typically ~10 × 10−7 members) was initially mixed with magnetic particles functionalized (tagged) with oxytocin molecule. Isolation from magnetic separation provided the first stage selection of library beads, and the putative Raptamers were released and subjected to a secondary pull-down to remove ‘false-positive’ candidates. The true Raptamers were identified using next-generation sequencing (NGS) methods and the comparison of the amount of each sequence after pull-down to its initial presence in the solution pool. The most abundant sequences were then synthesized with the appropriate oligonucleotide modifications and end modifications such as biotinylation.
## 2.2. Selection of Aptamers for Oxytocin
After the initial bead assay and the NGS stage, eight putative Raptamer sequences for oxytocin were obtained. All of the eight putative Raptamers were biotinylated, immobilized on streptavidin-coated carbon screen-printed electrodes (SPE), and characterized for oxytocin binding using both electrochemical impedance and direct electrochemical oxidation as demonstrated in Figure 1. Oxytocin was initially introduced to the Raptamer-modified SPEs (a-SPE) in controlled buffer solutions and incubated for durations varying between 1 and 10 min. After a brief rinsing step, aptamer-bound oxytocin resulted in impedance changes on the surface of the electrode (Figure A1) and the magnitude of this change was used to rank the affinity of each Raptamer as shown in Table 2 and Figure A2. This initial Raptamer validation step allowed us to rapidly down select four Raptamers for further characterization using an electrochemical oxidation method (utilizing the electrochemically active tyrosine group in oxytocin) and an optical particle aggregation method (using wavelengths shifts of functionalized gold nanoparticles).
All the measurements were recorded as Nyquist plots in a 0.1 M PBS buffer solution containing 5 mM [Fe(CN)6]$\frac{3}{4}$ redox pair (1:1 M ratio). The electrochemical impedance spectroscopy (EIS) spectra were conducted over a frequency range from 10 kHz to 0.1 Hz using an AC voltage with amplitude of 10 mV, superimposed on a DC potential of 0.15 V vs. Ag/AgCl. Affinity levels of each oxytocin aptamer (Raptamer) to oxytocin is denoted as “-” for no affinity, “+” for weak affinity and “++” for strong affinity as shown in Table 2.
## 2.3. Validation of Aptamers via Spectroscopic Characterization
We performed an independent validation of one of the candidates Raptamers, and then we have developed a robust, non-electrochemical procedure in house to validate the performance of this Raptamer. We utilized a well-established gold nanoparticle colorimetric assay, which proved to be an independent confirmation of aptamer binding. Our detection strategy [47] along with the spectroscopic information to characterize and validate each *Raptamer is* demonstrated in Figure 2. Briefly, citrate-reduced gold nanoparticles (AuNP) possess negative charges and their strong inter-particle electrostatic repulsive forces make them retain a characteristic red color in the solution. Upon mixing, the aptamer adsorbs on negatively charged AuNP and protects the nanoparticle against positively charged salt (Na+)-induced aggregation with its negative phosphate backbone. Conversely, when target biomarker (oxytocin) is introduced, the adsorbed aptamer desorbs from AuNP surface and strongly binds to the target, leaving AuNP unprotected in the solution. In presence of ~150 mM NaCl, the remaining AuNP negative charge is easily neutralized with Na+, leading to a loss in electrostatic repulsion. As a result, the inter-particles distance reduces and a salt-induced aggregation takes place and leads to a plasmon effect reflecting in color transition (red to purple to clear) in less than a minute [48,49,50]. In fact, the gold nanoparticles surface plasmon resonance peak (OD520) is reduced and the peak is shifted to a longer wavelength region (OD700) with increasing amount of Na+ ions. This simple mechanism allowed us to discriminate aptamer functionality in complex biological matrices like saliva based on the quantitative information obtained using UV-Vis spectroscopy for cross-validation.
With the colorimetric assay, we performed sensitivity and specificity analysis of Raptamer gOT-1B. Figure 2 and Figure A3 with inset demonstrates a dose-dependent linear correlation between the absorbance reading (OD520) and various oxytocin levels. As expected, a more drastic color change was observed when higher dosing of OT was introduced in buffer. According to the absorbance reading at 520 nm wavelength, a calibration curve was obtained with a coefficient of determination (R2 value) of over 0.98 after linear regression. These findings prove that Raptamer gOT-1B (one of the putative aptamers) is capable of distinguishing different levels of oxytocin target.
In this gold nanoparticle aptamer assay, when oxytocin was exogenously exposed to Raptamer-bound nanoparticles, a signal reduction at OD520 confirmed the binding event as shown in Figure 3. When a cocktail of oxytocin and vasopressin was tested exogenously, the absorbance reading showed nearly no difference compared to the result when only oxytocin was present. It is crucial to distinguish oxytocin from vasopressin since the two molecules differ only by two peptide residues while both contain the signal-generating tyrosine in their structures [1]. In this case, the absence of any false positive or false negative detection verified the specificity of the chosen aptamer. The specificity demonstrated by our novel technology offers a distinct advantage over commercially available immunoassays such as EIA or RIA [41]. A reduction in absorption was also observed when additional oxytocin was added endogenously, confirming the proper function of aptamer in real saliva environment. At this point, we have successfully validated the functionality of the Raptamer candidate gOT-1B identified previously for oxytocin detection both exogenously in buffer and endogenously in saliva.
## 2.4. Electrochemical Detection of Oxytocin
With the selected and validated Raptamer, we continued to develop an electrochemical assay for oxytocin detection. As highlighted in Figure 4, oxytocin aptamer gOT-1bB produced sufficient and distinguishable signals when binding with exogenously expressed oxytocin. The sensitivity analysis showed a dose-dependent response curve as demonstrated in Figure 4 (right panel). We obtained a calibration curve to correlate the electrochemical signal with oxytocin concentrations in reaction buffer environment.
As Figure 5 shows, when oxytocin was exogenously administrated in saliva, the aptamer on the carbon SPE surface bound specifically to oxytocin, and the tyrosine residue of the peptide produced a peak current as signal readout as shown in Figure 4. Such peak current (readout signal) was only observed when exogenously oxytocin was introduced into test samples both in buffer and control saliva environment [51,52]. On the other hand, adding vasopressin to saliva did not produce any signals, confirming that oxytocin was the only target.
## 3. Discussion
In this study, we successfully established an electrochemical sensor and a technology platform for future development of a rapid and accurate instrument capable of measuring oxytocin level in peripheral body fluids at point-of-care. This novel technology utilizes Raptamer-modified disposable carbon electrodes to achieve preeminent sensitivity of 1 pg/mL, which is on the order of laboratory-based technologies such as LC-MS and more sensitive than the commercially available Enzo Oxytocin ELISA Kit [41].
Cross-validated by electrochemical impedance spectroscopy and a nanoparticle colorimetric assay, we confirmed that this electrochemical detection method is also highly specific to oxytocin (1 µg/mL) with 100× specificity over vasopressin (>100 µg/mL) as shown in Figure 5. While the current commercially available immunoassays such as RIA and EIA often fail to distinguish vasopressin and oxytocin, our approach is able to capture the structural difference of these two similar molecules.
## 4.1. Materials and Equipment
Invitrogen’s Ultrapure DNase-free, RNase-free DEPC treated water (catalog # 4387937) was used in all studies. 10X PBS, NaCl, MgCl2, and all other reagents were purchased from Sigma-Aldrich, St. Louis, MO 63103, USA. Oxytocin peptide (ab120186) is purchased from Abcam Inc. Cambridge, MA 02139, USA. To avoid any DNase contamination, DNA Away (DNA Surface Decontaminant) was purchased from Thermo Scientific and used before performing any experiment. Streptavidin screen-printed carbon electrodes (Catalog# Dropsens DRP-STR110) were purchased from Metrohm USA Inc., Riverview, FL, USA. The USB-powered potentiostat (Model number: EmStat3+, Potential range ±3 V or ±4 V, and current ranges 1 nA to 10 mA or 100 mA) was obtained from PalmSens BV, Houten, Netherlands. A Raptamer (formerly X-Aptamer) Selection Kit was purchased from Raptamer Discovery Group, Houston, TX (RDG; formerly AM Biotechnologies; Houston, TX). This kit employs a proprietary bead-based library, containing modified DNA nucleotides within the random region, as the basis for the rapid selection of affinity agents for several targets in parallel using standard laboratory tools. All commercial de-identified pooled saliva samples were obtained from BioIVT (Westbury, NY, USA) and tested at a BSL-2 lab facility.
## 4.2. Raptamer Selection
For the primary Raptamer selection, a non-SELEX bead-based selection approach was utilized. In this selection, a bead-based DNA oligonucleotide library was mixed with oxytocin-coated magnetic particles and incubated for 90 min at room temperature. The library beads containing oligonucleotides that bound to oxytocin were isolated via magnetic separation. The isolated library beads were resuspended in 1N NaOH and incubated at 65 °C for 30 min to cleave the oligonucleotides from the beads. The cleaved oligonucleotides were then subjected to a secondary pull-down selection to remove ‘false-positive’ binders and to enrich the pool for Raptamers with high affinity to oxytocin. Following PCR amplification of the enriched and control pools and next-generation sequencing (PrimBio Research Institute, Garnet Valley, PA, USA), the candidate oxytocin Raptamers were identified by a proprietary analysis method (Raptamer Discovery Group, LLC, Houston, TX, USA). This method identifies the sequences enriched in the primary target pool compared to the control(s), which consist of any negative target controls and the magnetic particle (not containing target) control. These enriched oxytocin candidate Raptamers were then synthesized with the appropriate modified nucleotides included in the sequences.
## 4.3. Gold Nanoparticles (AuNPs) Synthesis for Nanoplasmonic Assay
Gold nanoparticles (AuNPs) were synthesized using the standard citrate reduction method. This nano-plasmonic test was designed according to the published articles [49,53,54,55]. Briefly, 2 mL of 50 mM HAuCl4 was added into 98 mL of boiling DI water in an Erlenmeyer flask. Then 10 mL of 38.8 mM sodium citrate was added, and the mixture was stirred until the color turned wine-red. The synthesized homogenous gold nanoparticles were characterized using UV-Vis spectroscopy and stored at 4 °C. All aptamers were reconstituted in 1× PBS, 2 mM MgCl2, pH 7.4, and targets were resuspended in 1 × PBS. All oxytocin aptamers were pre-heated at 95 °C for 5 min to remove any dimerization before utilizing in any experiment.
## 4.4. Nanoplasmonic Aptamer Characterization and Identification
For aptamer validation, 1 µL of 10 µM aptamer is added to 98 μL of 11 nM AuNPs (~13 nm size) to a final volume of 99 µL and incubated at room temperature (RT) for 5 min. After 5 min, 1 µL of 10 µM target (OT) is added to the 99 μL of pre-incubated Aptamer/AuNP solution for a final volume of 100 μL, resulting in final aptamer and target concentrations of 100 nM. After an additional 15–20 min of incubation at RT, 3 µL of 1 M of NaCl is added to 100 µL of the nanoparticle solution to a final concentration of ~30 mM Na+. After the addition of NaCl, the color transition was observed within 1 min or less and recorded with a photograph. In the presence of salt addition, when the aptamer binds to the specific target, it desorbs from the gold nanoparticle surface, leaving gold nanoparticles unprotected and easily neutralized by Na+ and showing a color changed from red to purple. Similarly, if the aptamer does not bind to its target, the gold nanoparticle’s color will be unchanged upon salt (Na+) addition. The resulting nanoparticle assembly’s change in the optical density (OD) at $\frac{520}{700}$ nm (Abs $\frac{520}{700}$) was used to plot the aggregation rate and degree. The UV-*Vis spectrum* of each sample was measured in 96 well plates using BioTek microplate reader (Gen5 microplate data collection and analysis software, BioTek Science Company, Winooski, VT). Control experiments were performed in the absence of target (only Aptamer + AuNP + adjusted reaction buffer, 1 × PBS, 2 mM mgCl2, pH 7.4). We utilized a similar procedure for detecting and validating all eight-oxytocin candidate Raptamers in buffer and saliva samples. In the previous study, the change in OD ratio at $\frac{520}{700}$ nm of the resulting nanoprobe complex assembly was used to determine the limit of detection (3 σ/slope) where σ is the standard deviation of controls while slope is obtained by linearly fitting the calibration curve [53,54]. The limit of detection (LOD) is calculated, followed by the 3-sigma rule. The equation is, LOD =3.3 × standard deviation of the regression line (σ) /Slope(S). A 3σ-rule is widely used to determine the signal-to-noise ratio for estimating the detection limit [54,55,56,57,58,59].
## 4.5. Sensitivity Measurements
For sensitivity measurement, studies were also performed using various amounts (0, 10, 40, 70, 100, 140 ng/mL) of target OT in 100 μL of solution and, color changes were recorded with in 1 min after incubation with ~30 mM NaCl. Control experiments were performed in the absence of target OT. We performed similar procedure to detect oxytocin in Saliva where we spiked various amounts of target in presence of fixed aptamer concentration. The OD value at $\frac{520}{700}$ nm and the pictures of the nanoparticle suspensions were recorded. All experiments were performed in triplicate ($$n = 3$$) using 96 well plates. For specificity measurements, individual OT Raptamers with their target and/or non-target analyte with a ratio of Raptamer: target = 1.4:1 (for buffer/Saliva) and evaluated to verify its false positive and false negative binding performance. The change in OD value at $\frac{520}{700}$ nm was measured and plotted.
## 4.6. Electrochemical Measurements (Sensitivity, Specificity and Cross-Reactivity)
For electrochemical sensitivity measurement, various concentrations (1 pg/mL to 100 pg/mL) of target oxytocin were tested, and a dose-dependent calibration curve with improved sensitivity (R2 = 0.9921) is observed from saliva samples. The assay’s specificity and cross-reactivity is evaluated in presence of non-target (Vasopressin) with saliva samples. A complex cocktail mixture of non-target peptide was prepared to show that the assay can detect only the targeted oxytocin in vitro PBS buffer as well as in exogenously enriched commercial saliva samples.
## 4.7. Electrochemical Detection of Oxytocin in Spiked Saliva Sample
The unprocessed (no anticoagulant/filtration, storage condition at −20 °C or colder) human saliva samples (de-identified, pooled samples sourced from BioIVT) were utilized to demonstrate the assay preclinical utility. Prior testing, 70 µL of 400 nM oxytocin aptamer in 1× PBS 2 mM MgCl2 is loaded on s-SPEs (prior washed with 1× PBS) for at least 10 min, then 70 µL (35 µL saliva + 35 µL of 1× PBS, 2 mM MgCl2) loaded on oxytocin aptamer immobilized streptavidin (s)-SPEs. Detection time is less than 5 s. The electrochemical Square Wave Voltammetry (SWV) parameters are: Current: 1 mA; T equilibration: 0 s; E begin: 0.0 V; E end: 1.5 V; E step: 0.005 V; Amplitude: 0.05 V; Frequency: 15.0 Hz.
## References
1. Asai K., Ivandini T.A., Einaga Y.. **Continuous and selective measurement of oxytocin and vasopressin using boron-doped diamond electrodes**. *Sci. Rep.* (2016.0) **6** 32429. DOI: 10.1038/srep32429
2. Macdonald K., Macdonald T.M.. **The peptide that binds: A systematic review of oxytocin and its prosocial effects in humans**. *Harv. Rev. Psychiatry* (2010.0) **18** 1-21. DOI: 10.3109/10673220903523615
3. Meyer-Lindenberg A., Domes G., Kirsch P., Heinrichs M.. **Oxytocin and vasopressin in the human brain: Social neuropeptides for translational medicine**. *Nature Revs. Neurosci.* (2011.0) **12** 524-538. DOI: 10.1038/nrn3044
4. Ding C., Leow M.K., Magkos F.. **Oxytocin in metabolic homeostasis: Implications for obesity and diabetes management**. *Obes. Rev.* (2019.0) **20** 22-40. DOI: 10.1111/obr.12757
5. Pataky Z., Guessous I., Caillon A., Golay A., Rohner-Jeanrenaud F., Altirriba J.. **Variable oxytocin levels in humans with different degrees of obesity and impact of gastric bypass surgery**. *Int. J. Obes. (Lond)* (2019.0) **43** 1120-1124. DOI: 10.1038/s41366-018-0150-x
6. Andari E., Duhamel J.-R., Zalla T., Herbrecht E., Leboyer M., Sirigu A.. **Promoting social behavior with oxytocin in high-functioning autism spectrum disorders**. *Proc. Nat.Acad. Sci.* (2010.0) **107** 4389-4394. DOI: 10.1073/pnas.0910249107
7. De Cagna F., Fusar-Poli L., Damiani S., Rocchetti M., Giovanna G., Mori A., Politi P., Brondino N.. **The Role of Intranasal Oxytocin in Anxiety and Depressive Disorders: A Systematic Review of Randomized Controlled Trials**. *Clin Psychopharmacol. Neurosci.* (2019.0) **17** 1-11. DOI: 10.9758/cpn.2019.17.1.1
8. Qian W., Zhu T., Tang B., Yu S., Hu H., Sun W., Pan R., Wang J., Wang D., Yang L.. **Decreased circulating levels of oxytocin in obesity and newly diagnosed type 2 diabetic patients**. *J. Clin. Endocrinol. Metabol.* (2014.0) **99** 4683-4689. DOI: 10.1210/jc.2014-2206
9. Coiro V., Passeri M., Davoli C., d’Amato L., Gelmini G., Fagnoni F., Schianchi L., Bentivoglio M., Volpi R., Chiodera P.. **Oxytocin response to insulin-induced hypoglycemia in obese subjects before and after weight loss**. *J. Endocrinol. Investig.* (1988.0) **11** 125-128. DOI: 10.1007/BF03350119
10. Schorr M., Marengi D.A., Pulumo R.L., Yu E., Eddy K.T., Klibanski A., Miller K.K., Lawson E.A.. **Oxytocin and Its Relationship to Body Composition, Bone Mineral Density, and Hip Geometry Across the Weight Spectrum**. *J. Clin. Endocrinol. Metabol.* (2017.0) **102** 2814-2824. DOI: 10.1210/jc.2016-3963
11. Monteleone A.M., Scognamiglio P., Volpe U., Di Maso V., Monteleone P.. **Investigation of Oxytocin Secretion in Anorexia Nervosa and Bulimia Nervosa: Relationships to Temperament Personality Dimensions**. *Eur. Eat. Disord. Rev.* (2016.0) **24** 52-56. DOI: 10.1002/erv.2391
12. Camerino C.. **The New Frontier in Oxytocin Physiology: The Oxytonic Contraction**. *Int. J. Mol. Sci.* (2020.0) **21**. DOI: 10.3390/ijms21145144
13. Johnson L., Manzardo A.M., Miller J.L., Driscoll D.J., Butler M.G.. **Elevated plasma oxytocin levels in children with Prader-Willi syndrome compared with healthy unrelated siblings**. *Am. J. Med. Genet. A* (2016.0) **170** 594-601. DOI: 10.1002/ajmg.a.37488
14. Swaab D., Purba J., Hofman M.. **Alterations in the hypothalamic paraventricular nucleus and its oxytocin neurons (putative satiety cells) in Prader-Willi syndrome: A study of five cases**. *J. Clin. Endocrinol. Metabol.* (1995.0) **80** 573-579
15. Zayan U., Caccialupi Da Prato L., Muscatelli F., Matarazzo V.. **Modulation of the thermosensory system by oxytocin**. *Front. Mol. Neurosci.* (2022.0) **15** 1075305. DOI: 10.3389/fnmol.2022.1075305
16. Guastella A.J., Boulton K.A., Whitehouse A.J.O., Song Y.J., Thapa R., Gregory S.G., Pokorski I., Granich J., DeMayo M.M., Ambarchi Z.. **The effect of oxytocin nasal spray on social interaction in young children with autism: A randomized clinical trial**. *Mol. Psy.* (2023.0) **28** 834-842. DOI: 10.1038/s41380-022-01845-8
17. De Mayo M.M., Young L.J., Hickie I.B., Song Y.J.C., Guastella A.J.. **Circuits for social learning: A unified model and application to Autism Spectrum Disorder**. *Neurosci. Biobehav. Rev.* (2019.0) **107** 388-398. DOI: 10.1016/j.neubiorev.2019.09.034
18. Sikich L., Kolevzon A., King B.H., McDougle C.J., Sanders K.B., Kim S.-J., Spanos M., Chandrasekhar T., Trelles M.D.P., Rockhill C.M.. **Intranasal Oxytocin in Children and Adolescents with Autism Spectrum Disorder**. *New Engl. J. Med.* (2021.0) **385** 1462-1473. DOI: 10.1056/NEJMoa2103583
19. Parker K.J., Garner J.P., Libove R.A., Hyde S.A., Hornbeak K.B., Carson D.S., Liao C.-P., Phillips J.M., Hallmayer J.F., Hardan A.Y.. **Plasma oxytocin concentrations and OXTR polymorphisms predict social impairments in children with and without autism spectrum disorder**. *Proc. Nat. Acad. Sci. USA* (2014.0) **111** 12258-12263. DOI: 10.1073/pnas.1402236111
20. Feldman R., Gordon I., Zagoory-Sharon O.. **Maternal and paternal plasma, salivary, and urinary oxytocin and parent–infant synchrony: Considering stress and affiliation components of human bonding**. *Develop. Sci.* (2011.0) **14** 752-761. DOI: 10.1111/j.1467-7687.2010.01021.x
21. Clark C.L., St John N., Pasca A.M., Hyde S.A., Hornbeak K., Abramova M., Feldman H., Parker K.J., Penn A.A.. **Neonatal CSF oxytocin levels are associated with parent report of infant soothability and sociability**. *Psychoneuroendocrinology* (2013.0) **38** 1208-1212. DOI: 10.1016/j.psyneuen.2012.10.017
22. Guastella A.J., Einfeld S.L., Gray K.M., Rinehart N.J., Tonge B.J., Lambert T.J., Hickie I.B.. **Intranasal oxytocin improves emotion recognition for youth with autism spectrum disorders**. *Biol. Psychiatry* (2010.0) **67** 692-694. DOI: 10.1016/j.biopsych.2009.09.020
23. Harman J.H., Kim A.. **Current trends in cervical ripening and labor induction**. *Am. Fam. Physician* (1999.0) **60** 477-484. PMID: 10465223
24. Tran G., Kanczuk M., Balki M.. **The association between the time from oxytocin cessation during labour to Cesarean delivery and postpartum blood loss: A retrospective cohort study**. *Can. J. Anaesth.* (2017.0) **64** 820-827. DOI: 10.1007/s12630-017-0874-4
25. Buckley D.S.J.. **Pain in Labour: Your Hormones Are Your Helpers**
26. Cauldwell M., Steer P.J., Swan L., Uebing A., Gatzoulis M.A., Johnson M.R.. **The management of the third stage of labour in women with heart disease**. *Heart* (2017.0) **103** 945-951. DOI: 10.1136/heartjnl-2016-310607
27. Othman E.R., Fayez M.F., El Aal D.E., El-Dine Mohamed H.S., Abbas A.M., Ali M.K.. **Sublingual misoprostol versus intravenous oxytocin in reducing bleeding during and after cesarean delivery: A randomized clinical trial**. *Taiwan J. Obstet. Gynecol.* (2016.0) **55** 791-795. DOI: 10.1016/j.tjog.2016.02.019
28. Pakniat H., Khezri M.B.. **The Effect of Combined Oxytocin-Misoprostol Versus Oxytocin and Misoprostol Alone in Reducing Blood Loss at Cesarean Delivery: A Prospective Randomized Double-Blind Study**. *J. Obstet. Gynaecol. India* (2015.0) **65** 376-381. DOI: 10.1007/s13224-014-0607-3
29. Ghulmiyyah L.M., Usta I.M., Ghazeeri G., Taher N., Abu-Ghannam G., Tamim H., Nassar A.H.. **Intravenous Oxytocin Use to Decrease Blood Loss during Scheduled Cesarean Delivery: A Randomized Double-Blinded Controlled Trial (OXYTRIAL)**. *Am. J. Perinatol.* (2017.0) **34** 379-387. DOI: 10.1055/s-0036-1592130
30. Ghulmiyeh D.L.. **Oxytocin to Decrease Blood Loss during Cesarean Section. In NCT00891150, N.C. g**. (2013.0)
31. Hill M., Reed K.L., Cohen W.R.. **Oxytocin utilization for labor induction in obese and lean women**. *J. Perinat. Med.* (2015.0) **43** 703-706. DOI: 10.1515/jpm-2014-0134
32. Prevost M., Zelkowitz P., Tulandi T., Hayton B., Feeley N., Carter C.S., Joseph L., Pournajafi-Nazarloo H., Yong Ping E., Abenhaim H.. **Oxytocin in pregnancy and the postpartum: Relations to labor and its management**. *Front. Public Health* (2014.0) **2** 1. DOI: 10.3389/fpubh.2014.00001
33. Hoshiyama E., Tatsumoto M., Iwanami H., Saisu A., Watanabe H., Inaba N., Hirata K.. **Postpartum migraines: A long-term prospective study**. *Intern. Med.* (2012.0) **51** 3119-3123. DOI: 10.2169/internalmedicine.51.8542
34. Tzabazis A., Kori S., Mechanic J., Miller J., Pascual C., Manering N., Carson D., Klukinov M., Spierings E., Jacobs D.. **Oxytocin and Migraine Headache**. *Headache* (2017.0) **57** 64-75. DOI: 10.1111/head.13082
35. Forest Tennant M.. **DrPH Oxytocin, an Opioid Alternative, Ready for Regular Clinical Use to Manage Chronic Pain**
36. Massey S.H., Schuette S.A., Pournajafi-Nazarloo H., Wisner K.L., Carter C.S.. **Interaction of oxytocin level and past depression may predict postpartum depressive symptom severity**. *Arch. Womens Ment. Health* (2016.0) **19** 799-808. DOI: 10.1007/s00737-016-0616-6
37. Devikarani D., Harsoor S.. **Are we using right dose of oxytocin?**. *Indian J. Anaesth.* (2010.0) **54** 371-433. PMID: 21189871
38. McCullough M.E., Churchland P.S., Mendez A.J.. **Problems with measuring peripheral oxytocin: Can the data on oxytocin and human behavior be trusted?**. *Neurosci. Biobehav. Rev.* (2013.0) **37** 1485-1492. DOI: 10.1016/j.neubiorev.2013.04.018
39. Carson D.S., Howerton C.L., Garner J.P., Hyde S.A., Clark C.L., Hardan A.Y., Penn A.A., Parker K.J.. **Plasma vasopressin concentrations positively predict cerebrospinal fluid vasopressin concentrations in human neonates**. *Peptides* (2014.0) **61** 12-16. DOI: 10.1016/j.peptides.2014.08.003
40. Carson D.S., Garner J.P., Hyde S.A., Libove R.A., Berquist S.W., Hornbeak K.B., Jackson L.P., Sumiyoshi R.D., Howerton C.L., Hannah S.L.. **Arginine Vasopressin Is a Blood-Based Biomarker of Social Functioning in Children with Autism**. *PLoS ONE* (2015.0) **10**. DOI: 10.1371/journal.pone.0132224
41. **Enzo Lifesciences Oxytocin ELISA Kit**
42. Jong T.R., Menon R., Bludau A., Grund T., Biermeier V., Klampfl S.M., Jurek B., Bosch O.J., Hellhammer J., Neumann I.D.. **Salivary oxytocin concentrations in response to running, sexual self-stimulation, breastfeeding and the TSST: The Regensburg Oxytocin Challenge (ROC) study**. *Psychoneuroendocrinology* (2015.0) **62** 381-388. DOI: 10.1016/j.psyneuen.2015.08.027
43. Altemus M., Roca C., Galliven E., Romanos C., Deuster P.. **Increased vasopressin and adrenocorticotropin responses to stress in the midluteal phase of the menstrual cycle**. *J. Clin. Endocrinol. Metab.* (2001.0) **86** 2525-2530. DOI: 10.1210/jcem.86.6.7596
44. Bello D., White-Traut R., Schwertz D., Pournajafi-Nazarloo H., Carter C.S.. **An exploratory study of neurohormonal responses of healthy men to massage**. *J. Alternative Complement. Med.* (2008.0) **14** 387-394. DOI: 10.1089/acm.2007.0660
45. Zhang G., Zhang Y., Fast D.M., Lin Z., Steenwyk R.. **Ultra sensitive quantitation of endogenous oxytocin in rat and human plasma using a two-dimensional liquid chromatography-tandem mass spectrometry assay**. *Anal. Biochem.* (2011.0) **416** 45-52. DOI: 10.1016/j.ab.2011.04.041
46. Strehlitz B., Reinemann C., Linkorn S., Stoltenburg R.. **Aptamers for pharmaceuticals and their application in environmental analytics**. *Bioanal. Rev.* (2012.0) **4** 1-30. DOI: 10.1007/s12566-011-0026-1
47. Li H., Rothberg L.. **Colorimetric detection of DNA sequences based on electrostatic interactions with unmodified gold nanoparticles**. *Proc. Nat. Acad. Sci. USA* (2004.0) **101** 14036-14039. DOI: 10.1073/pnas.0406115101
48. Pfeiffer F., Mayer G.. **Selection and Biosensor Application of Aptamers for Small Molecules**. *Front. Chem.* (2016.0) **4** 25. DOI: 10.3389/fchem.2016.00025
49. Mondal B., Ramlal S., Lavu P.S., Kingston J.. **Highly Sensitive Colorimetric Biosensor for Staphylococcal Enterotoxin B by a Label-Free Aptamer and Gold Nanoparticles**. *Front. Microbiol.* (2018.0) **9** 179. DOI: 10.3389/fmicb.2018.00179
50. Rana M., Balcioglu M., Robertson N.M., Hizir M.S., Yumak S., Yigit M.V.. **Low picomolar, instrument-free visual detection of mercury and silver ions using low-cost programmable nanoprobes**. *Chem. Sci.* (2017.0) **8** 1200-1208. DOI: 10.1039/C6SC03444F
51. Xu J., Wang Y., Xian Y., Jin L., Tanaka K.. **Preparation of multiwall carbon nanotubes film modified electrode and its application to simultaneous determination of oxidizable amino acids in ion chromatography**. *Talanta* (2003.0) **60** 1123-1130. DOI: 10.1016/S0039-9140(03)00214-5
52. Varmira K., Mohammadi G., Mahmoudi M., Khodarahmi R., Rashidi K., Hedayati M., Goicoechea H.C., Jalalvand A.R.. **Fabrication of a novel enzymatic electrochemical biosensor for determination of tyrosine in some food samples**. *Talanta* (2018.0) **183** 1-10. DOI: 10.1016/j.talanta.2018.02.053
53. Rana M., Augspurger E.E., Hizir M.S., Alp E., Yigit M.V.. **Molecular logic gate operations using one-dimensional DNA nanotechnology**. *J. Mat. Chem. C* (2018.0) **6** 452-455. DOI: 10.1039/C7TC04957A
54. Rana M., Balcioglu M., Kovach M., Hizir M.S., Robertson N.M., Khan I., Yigit M.V.. **Reprogrammable multiplexed detection of circulating oncomiRs using hybridization chain reaction**. *Chem. Commun.* (2016.0) **52** 3524-3527. DOI: 10.1039/C5CC09910B
55. Kumar N., Rana M., Geiwitz M., Khan N.I., Catalano M., Ortiz-Marquez J.C., Kitadai H., Weber A., Dweik B., Ling X.. **Rapid, Multianalyte Detection of Opioid Metabolites in Wastewater**. *ACS Nano* (2022.0) **16** 3704-3714. DOI: 10.1021/acsnano.1c07094
56. Novick S.J., Zhao W., Yang H.. **Setting Alert and Action Limits in the Presence of Significant Amount of Censoring in Data**. *PDA J. Pharm. Sci. Technol.* (2017.0) **71** 20-32. DOI: 10.5731/pdajpst.2016.006684
57. Kim J.H., Cho C.H., Ryu M.Y., Kim J.G., Lee S.J., Park T.J., Park J.P.. **Development of peptide biosensor for the detection of dengue fever biomarker, nonstructural 1**. *PLoS ONE* (2019.0) **14**. DOI: 10.1371/journal.pone.0222144
58. Desai D., Kumar A., Bose D., Datta M.. **Ultrasensitive sensor for detection of early stage chronic kidney disease in human**. *Biosens. Bioelectron.* (2018.0) **105** 90-94. DOI: 10.1016/j.bios.2018.01.031
59. Cipra T., Fuchs A., Formanek J., Kubat J., Mikiskova H., Zajicek P., Dvorak J.. **Detection and interpolation of outliers in biosignals**. *Act Nerv. Super.* (1990.0) **32** 283-291
|
---
title: Phenolic compounds, antioxidant activity and sensory evaluation of sea buckthorn
(Hippophae rhamnoides L.) leaf tea
authors:
- Qian He
- Kailin Yang
- Xinyan Wu
- Chunhong Zhang
- Chunnian He
- Peigen Xiao
journal: Food Science & Nutrition
year: 2022
pmcid: PMC10003008
doi: 10.1002/fsn3.3155
license: CC BY 4.0
---
# Phenolic compounds, antioxidant activity and sensory evaluation of sea buckthorn (Hippophae rhamnoides L.) leaf tea
## Abstract
Sea buckthorn leaf tea, an emerging potential functional beverage product, has not yet had appropriate product standards and corresponding quality evaluation methods, and its poor taste directly affects the acceptance of the population, thus limiting its market consumption potential. In this study, two major packaging forms of sea buckthorn leaf tea available in the Chinese market were selected. The contents of total phenolics, total flavonoids, and 10 phenolic compounds, as well as the in vitro antioxidant capacity and sensory characteristics of sea buckthorn leaf tea were analyzed. Results showed that the quality of sea buckthorn leaf tea in the Chinese market varied widely. The total phenolic content, total flavonoid content, antioxidant activity, and consumer acceptance of bagged sea buckthorn leaf tea were higher than those of bulk sea buckthorn leaf tea. Multifactorial statistical analysis showed that the taste astringency of sea buckthorn leaf tea was closely related to ellagic acid and isorhamnetin‐3‐O‐neohesperidin. Furthermore, isorhamnetin‐3‐O‐neohesperidin had a greater effect on the antioxidant activity of sea buckthorn leaf tea. Therefore, ellagic acid and isorhamnetin‐3‐O‐neohesperidin can be used as potential quality markers for sea buckthorn leaf tea. This work provides a reference for taste improvement and quality control of sea buckthorn leaf tea.
In this study, the quality and taste of sea buckthorn leaf tea in the Chinese market were analyzed by the UPLC technique, in vitro antioxidant activity assay, and sensory evaluation. The results showed that ellagic acid was an important factor affecting the bitterness of sea buckthorn leaf tea. Ellagic acid and isorhamnein‐3‐O‐neohesperidin could be used as potential quality markers for buckwheat leaf tea. This work provides a reference for taste improvement and quality control of sea buckthorn leaf tea for the promotion and application of sea buckthorn leaf tea as a potential functional beverage product.
## INTRODUCTION
Sea buckthorn (Hippophae rhamnoides L.), belonging to the Elaeagnaceae family, is a thorny deciduous shrub (Büyükokuroǧlu & Gülçin, 2009). Sea buckthorn is widely distributed in many countries, such as China, India, and Mongolia (Morikawa et al., 2015). It is a drought‐tolerant and hardy plant that is used in land reclamation and farmland conservation (Li & Schroeder, 1996). Sea buckthorn berries have been used in traditional medicine for a long history. In the modern food industry, sea buckthorn berries can be processed into fruit powder, beverage, and oil, which have been widely studied and applied (Pundir et al., 2021).
The leaves of sea buckthorn, as another part of sea buckthorn with abundant resources, are not fully utilized. The leaves have attracted increasing attention because of its potential health value. Studies on in vitro and in vivo pharmacological activities have shown that sea buckthorn leaves have antioxidant, antibacterial, antiviral, antitumor, and immunomodulatory effects (Geetha et al., 2003; Jain et al., 2008; Li, Liu, et al., 2021). Population application history investigations and animal studies have shown that sea buckthorn leaves have a good safety profile and no adverse effects in rats (Wang et al., 2017). Sea buckthorn leaves contain a variety of phytochemicals, such as polyphenols, polysaccharides, flavonoids, carotenoids, and saponins (Trivedi & Sati, 2022). In this regard, sea buckthorn leaves are a potential ingredient because of its nutritional and medicated components that are beneficial to human health. As a result, the Chinese government approved the use of sea buckthorn leaves as a common food ingredient in 2013. The development and application of sea buckthorn leaf as a raw material in food, medicine, animal feed, and even cosmetics have rapidly developed (Beveridge et al., 1999; Li, Liu, et al., 2021).
In recent years, scholars have focused on utilizing sea buckthorn leaves and converting them into a potential functional beverage product due to their health benefits. After drying at high temperatures, sea buckthorn leaf tea retains considerable nutritional value and is comparable with commonly consumed vegetables (Tanwar et al., 2018). The common forms of sea buckthorn leaf tea on the market are bulk sea buckthorn leaf tea (unpackaged) and bagged sea buckthorn leaf tea. Sea buckthorn leaf tea contains a variety of functional ingredients with many health benefits, such as lipid lowering, weight loss, and antioxidation (Cho et al., 2014; Lee et al., 2011).
Sea buckthorn leaf tea is an emerging commodity and is available in various forms such as loose sea buckthorn leaf tea and bagged sea buckthorn leaf tea when sold in the market. Since sea buckthorn is widely distributed in China, differences in ecological environment, genetic germplasm, and processing method may significantly affect the chemical composition and activity of sea buckthorn leaf tea (Tanwar et al., 2018). However, no suitable quality evaluation methods and standards are available yet for this product. As a potential functional beverage, the taste of sea buckthorn leaf tea has a direct impact on its market consumption and acceptance. Few reports are available on factors affecting the taste of sea buckthorn leaf tea. Therefore, sea buckthorn leaf tea currently available in the Chinese market should be investigated to establish quality evaluation methods and explore key factors affecting taste. In particular, the correlation between the phenolic composition of sea buckthorn leaves and their taste should be established to expand the utilization of sea buckthorn leaf tea as a potential health resource.
The aim of this study was to investigate the phenolic composition, sensory quality, and antioxidant activity of sea buckthorn leaf tea available in the Chinese market, to investigate the important factors affecting the taste of sea buckthorn leaf tea, and to provide valuable data for the quality evaluation of sea buckthorn leaf tea.
## Chemicals and materials
The following standards with purity higher than $98\%$ were purchased from Yuanye Biotechnology Co., Ltd.: isoquercitrin, isorhamnetin‐3‐O‐neohesperidoside, narcissin, quercitrin, ellagic acid, gallocatechin, apigenin, rutin, and kaempferol. Epicatechin was obtained from Chengdu Pulse Biotechnology Co., Ltd. Citric acid, sucrose, tannic acid, sodium glutamate, and quinine hydrochloride was acquired from Adamas Beta Chemical Reagents Co., Ltd. 2,2′‐Azino‐bis‐(3‐ethylbenzthiazoline‐6‐sulphonate acid; ABTS) and ferric ion‐reducing antioxidant power (FRAP) kits were provided by Shanghai Biyuntian Biotechnology Co., Ltd. Diphenyl‐1‐picrylhydrazyls (DPPH) was supplied by Sigma‐Aldrich. Chromatography‐grade formic acid was obtained from Shanghai Aladdin Reagent Co., Ltd. Methanol was acquired from Tianjin Beilian Fine Chemical Depot. Experimental ultrapure water was produced with a Mili‐Q system (Millipore Corp.).
## Sample materials
Sea buckthorn leaf tea samples (No. S1–S18) were purchased in the Chinese market and identified by Prof. Chunnian He, a researcher at the Institute of Medicinal Plant Development in Beijing, China. The source and lot numbers of the samples are shown in Table 1.
**TABLE 1**
| Number | Manufacturer (China) | Factory time | Type |
| --- | --- | --- | --- |
| S1 | Xining, Qinghai | June 2020 | Sea buckthorn leaf tea bag |
| S2 | Dingxi, Gansu | July 2020 | Sea buckthorn leaf tea bag |
| S3 | Zhangye, Gansu | July 2020 | Sea buckthorn leaf tea bag |
| S4 | Xinzhou, Shanxi | July 2020 | Sea buckthorn leaf tea bag |
| S5 | Lvliang, Shanxi | July 2020 | Sea buckthorn leaf tea bag |
| S6 | Da Hinggan Ling Prefecture, Heilongjiang | July 2020 | Sea buckthorn leaf tea bag |
| S7 | Xi'an, Shaanxi | January 2021 | Sea buckthorn leaf tea bag |
| S8 | Xinjiang | March 2021 | Bulk sea buckthorn leaf tea |
| S9 | Tacheng, Xinjiang | June 2021 | Bulk sea buckthorn leaf tea |
| S10 | Aksu, Xinjiang | January 2021 | Bulk sea buckthorn leaf tea |
| S11 | Urumqi, Xinjiang | March 2021 | Bulk sea buckthorn leaf tea |
| S12 | Kashgar, Xinjiang | June 2020 | Bulk sea buckthorn leaf tea |
| S13 | Jilin City, Jilin Province | July 2020 | Bulk sea buckthorn leaf tea |
| S14 | Chaoyang, Liaoning | June 2021 | Bulk sea buckthorn leaf tea |
| S15 | Chaoyang, Liaoning | June 2020 | Bulk sea buckthorn leaf tea |
| S16 | Zibo, Shandong | June 2020 | Bulk sea buckthorn leaf tea |
| S17 | Tongliao, Inner Mongolia | June 2021 | Bulk sea buckthorn leaf tea |
| S18 | Tongliao, Inner Mongolia | June 2021 | Bulk sea buckthorn leaf tea |
## Preparation of extracts
In an ultrasonic extractor, about 1 g of tea powder was extracted with 25 ml of water at a temperature of 50°C and a power of 300 W for 40 min. The extracts were stored at 4°C prior to further assay.
## Quantitative determination of 10 compound contents by UPLC‐DAD
Quantification of 10 phenolic compounds was performed on a Thermo Ultimate 3000 UPLC system equipped with a DAD‐3000RS (Thermo Fisher Scientific). A Waters ACQUITY UPLC BEH C18 (2.1 mm × 100 mm, 1.7 μm) column was used. The mobile phase consisted of formic acid–water (0.1:100, v/v; A) and acetonitrile (B), and the gradient elution procedure was as follows: 0–10 min, $2\%$–$10.7\%$ B; 10–18 min, $10.7\%$–$10.7\%$ B; 18–25 min, $10.7\%$–$16.3\%$ B; 25–30 min, $16.3\%$–$29.8\%$ B; 30–35 min, $29.8\%$–$45\%$ B; and 35–40 min, $45\%$–$90\%$ B. The flow rate was 0.3 ml/min, and the injection volume was 2 μl. The temperatures of the column and sample tray were 30 and 10°C, respectively. The detection wavelengths were set at 360, 280, and 254 nm. Chromeleon 7 software was used to acquire and analyze data.
A mixture stock standard solution containing 90.53 μg/ml epicatechin, 5.92 μg/ml isorhamnetin‐3‐O‐neohesperidoside, 873.88 μg/ml narcissin, 19.01 μg/ml isoquercitrin, 1500.32 μg/ml ellagic acid, 27.09 μg/ml rutin, 25.60 μg/ml quercitrin, 6.76 μg/ml catechin, 9.47 μg/ml kaempferol, and 6.33 μg/ml apigenin was diluted by 1, 2, 2.5, 3, and 4 times to obtain standard solutions for plotting of standard curves. Prior to UPLC analysis, all solutions were stored at 4°C and filtered through 0.22 μm nylon micropore membranes.
The method was validated for linearity, precision, repeatability, stability, and recovery rates following the International Conference on Harmonization (ICH) guideline (ICH Q2(R1), 2005).
## The content of flavonoids and tannins
In order to more fully reflect the content of polyphenols in sea buckthorn leaf tea, a semiquantitative method was temporarily adopted. First, the chromatographic peaks of sea buckthorn leaf tea were identified based on the UV characteristic spectra of flavonoids and tannins (the UV absorption spectra of flavonoids in methanol mainly appear between 300 and 400 nm and 240 and 280 nm; and the tannin has strong characteristic absorption at about 275 nm). Furthermore, the contents of individual flavonoids and tannins were calculated by comparing their peak areas with those of isoquercitrin and ellagic acid standards, respectively. Finally, the contents of all single flavonoid or tannin were added to obtain the total contents of flavonoids and tannins in sea buckthorn leaf tea.
## Antioxidant activity assays
In vitro antioxidant activity of ultrasonic extracts of sea buckthorn leaf tea was determined by FRAP, DPPH, and ABTS methods, respectively.
## FRAP method
Total antioxidant capacity was evaluated using the FRAP method described by Li, Li, et al. [ 2021]. About 15 μl of sea buckthorn leaf tea solution was placed separately in a 96‐well plate, mixed with 180 μl of FRAP working solution, and shaken thoroughly for 10 s. The solution was incubated for 6 min at 37°C, and absorbance was recorded at 734 nm. The results were expressed as mg TE/g.
## DPPH method
DPPH method was used to determine the antioxidant activity with reference to literature (Gülçin et al., 2011; Köksal & Gülçin, 2008). In brief, 0.2 mM DPPH·solution in ethanol was prepared, and 15 μl of sea buckthorn leaf tea solution extract plus 180 μl of 0.2 mM DPPH solution was incubated at 37°C for 30 min under light protection. Absorbance was recorded at 517 nm. The standard curve of Trolox was developed. The antioxidant capacity of the samples was expressed as milligrams of Trolox equivalents per gram of dried sample (mg TE/g).
## ABTS method
The ABTS radical scavenging activity of sea buckthorn leaf tea was referenced from the method of Li, Li, et al. [ 2021]. ABTS was blue‐green in color and had a characteristic absorbance at 734 nm. About 10 μl of sea buckthorn leaf tea solution was placed separately in a 96‐well plate, mixed with 200 μl of ABTS + working solution, and shaken thoroughly for 10 s. The sample was incubated at room temperature for 5 min, and absorbance was recorded at 734 nm. The results were expressed as mg TE/g.
## Analysis of the antioxidant potency composite (APC) index
The APC index was used to evaluate the overall antioxidant activity of sea buckthorn leaf tea (Peng et al., 2019). APC index was calculated using the following formula: APC index (%) = ([DPPH value]/[the maximum DPPH value] + [ABTS value]/[the maximum ABTS value] + [FRAP value]/[the maximum FRAP value])/3 × 100.
## Sensory evaluation
Each sea buckthorn leaf tea (1:50, w/v) was steeped at 80°C for 5 min, and 5 ml of each tea infusion was poured into a disposable plastic cup and cooled to room temperature. Fifteen trained panelists including 11 women and 4 men aged between 20 and 45 years evaluated the product in the laboratory. Each person evaluated 18 sea buckthorn leaf tea samples. The following attributes were evaluated: color, aroma, sour, bitter, sweet, umami, astringent, and overall acceptance. The scoring criteria were based on the 10‐component table method described in a previous work (Liu et al., 2018), as shown in Table S2. Each panelist was asked to wash their taste buds with drinking water between different samples at intervals of 1–2 min. Sensory profiles of the samples were developed based on their average score. Overall acceptance was evaluated based on the taste and color of sea buckthorn leaf tea. The standard references for bitterness, astringency, freshness, and sweetness were quinine hydrochloride, tannin, sodium glutamate, and sucrose, respectively.
Taste is one of the important quality factors that influence consumer preferences after variety (Yu et al., 2020). Therefore, sensory analysis of sea buckthorn leaf tea was carried out. The experiment is illustrated in Figure 2. Color is an important quality criterion that influences consumer preferences because it is the first attribute perceived by consumers (Lima et al., 2019). The results showed that sea buckthorn leaf tea infusion was yellow‐brown, and the color difference was not significant ($p \leq .05$). The common explanation for the yellowish‐brown color of tea infusion is that it is determined by water‐soluble flavonoids such as kaempferol, isoquercitrin, and rutin. The result showed no significant difference in the aroma evaluation of the sea buckthorn leaf teas.
**FIGURE 2:** *Sea buckthorn leaf tea taste evaluation chart*
Flavor is one of the important factors of acceptance among consumers. Astringency was found in all tea infusion from 1.29 ± 1.39 to 5.64 ± 1.62. The sweet scores pattern showed the highest score at S18 (Tongliao, Inner Mongolia, 3.38), followed by S4 (Xinzhou, Shanxi, 1.24). Acid flavor was found in all tea infusion from 1.09 ± 1.63 to 3.04 ± 2.36. The fresh scores pattern showed the highest score at S18 (Tongliao, Inner Mongolia, 3.26), followed by S4 (Xinzhou, Shanxi, 1.24). Bitterness was found in all tea infusion from 2.00 ± 1.53 to 5.96 ± 2.23. Based on the score of the sensory characteristics, the corresponding radar chart was created. Statistical differences were found in bitterness, astringency, and acid in sea buckthorn leaf tea samples ($p \leq .05$). Ma et al. [ 2019] claimed that the astringency perception threshold of sea buckthorn leaves is low. However, their results were inferred from this fact. This fact is that sea buckthorn leaves are rich in flavonol glycosides, which have little effect on astringency (Scharbert & Hofmann, 2005). Therefore, our experimental results on the predominantly bitter and astringent taste of sea buckthorn leaf tea are more reliable. It is currently reported that the astringent substances of tea are mainly polyphenols, and the primary phenolic acids in tea are ellagic acid, chlorogenic acid, etc. ( Ceci et al., 2018). At the same time, flavonoids are one of the important factors affecting the bitter taste of tea (Zhang et al., 2020). At present, technologies to improve the bitter taste of tea mainly include enzyme treatment, microcapsule technology, food irradiation, and other technologies (Ye et al., 2022). Therefore, these methods can be used to improve the taste of sea buckthorn leaf tea.
Acceptability was found in all 18 tea infusion from 2.27 ± 1.62 to 6.24 ± 1.56. The top three teas with the acceptability were S11 (Urumqi, Xinjiang, 6.36), S17 (Tongliao, Inner Mongolia, 6.03), and S5 (Lvliang, Shanxi, 5.67). S8 (Xinjiang, 2.15) has the lowest acceptance. According to Figure 2, it can be shown that the bitterness and astringency of sea buckthorn leaf tea are closely related to the overall acceptability. Meanwhile, the mean value of acceptability of the bagged sea buckthorn leaf tea (S1–S7) was 5.8 and the mean value of acceptability of the bulk sea buckthorn leaf tea (S8–S10) was 5.4. Hence, bagged tea is more popular among consumers. In summary, the results showed the influence of taste and packaging form on the acceptance of sea buckthorn leaf tea.
## Statistical analysis
Standard deviation was calculated for each experiment after three repetitions. One‐way analysis of variance (ANOVA, $p \leq .05$) with Tukey's HSD and Tamhane's test was used to evaluate the results using SPSS 20.0. Pearson's correlation coefficients were determined using SPSS 20.0.
## Composition and contents of phenolic compounds
An analytical UPLC‐DAD method was developed for the simultaneous determination of 10 phenolic compounds in sea buckthorn leaf tea. The method was validated by determining linearity, precision, repeatability, stability, and recovery rates. Good linear correlations were obtained for the phenolic compounds using this method with R 2 >.999. Moreover, the relative standard deviations of the repeatability, precision, stability, and recovery of the method were all below $5.00\%$, and the recovery was within the range of $97.91\%$–$102.70\%$. The results confirm the validity of the method for the evaluation of sea buckthorn leaf tea (Table S1). The UPLC‐DAD chromatogram of a representative sample S3 mixed with a standard solution is shown in Figure S1.
The results of ultrasonic‐assisted extraction of sea buckthorn leaf tea are as follows: ten phenolic compounds including two catechins (catechin and epicatechin), one phenolic acid (ellagic acid), one flavonoid (apigenin), and six flavonols (isoquercitrin isorhamnete‐3‐O‐neohesperidin, aquaporin, quercetin, rutin, and kaempferol) were detected from the sea buckthorn leaf tea. Ellagic acid was the major compound in all the test sample (Ciesarová et al., 2020), with the highest content in S1 (Xining, Qinghai, 5.77 mg/g), S2 (Xining, Qinghai, 5.87 mg/g), and S13 (Jilin City, Jilin Province, 4.83 mg/g; Table 2).
**TABLE 2**
| Number | Isoquercitrin (mg/g) | Ellagic acid (mg/g) | Rutin (mg/g) | Narcissin (mg/g) | Quercitrin (mg/g) | Kaempferol (mg/g) | Catechin (mg/g) | Epicatechin (mg/g) | Isorhamnetin‐3‐O‐neohesperidoside (mg/g) | Apigenin (mg/g) |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| S1 | 0.82 ± 0.02b | 5.77 ± 0.05a | 0.29 ± 0.01ef | 0.58 ± 0.01b | 0.74 ± 0.01b | 0.056 ± 0.001fg | 3.38 ± 0.13a | 0.40 ± 0.02a | 0.002 ± 0.001a | – |
| S2 | 0.35 ± 0.03fg | 5.87 ± 0.25a | 0.69 ± 0.06b | 0.27 ± 0.02d | 1.05 ± 0.05d | 0.051 ± 0.001g | – | 0.22 ± 0.01ef | 0.007 ± 0.001g | – |
| S3 | 0.35 ± 0.03fg | 3.32 ± 0.21f | 0.09 ± 0.01k | 0.23 ± 0.01de | 0.25 ± 0.02de | 0.036 ± 0.001h | – | 0.33 ± 0.03b | 0.005 ± 0.001h | – |
| S4 | 0.26 ± 0.01j | 1.93 ± 0.06ij | 0.19 ± 0.02ghi | 0.11 ± 0.01fg | 0.39 ± 0.02fg | 0.032 ± 0.001h | – | – | 0.002 ± 0.001j | – |
| S5 | 0.37 ± 0.01f | 3.33 ± 0.06f | 0.22 ± 0.01gh | 0.20 ± 0.01e | 0.40 ± 0.01e | 0.069 ± 0.003d | – | 0.29 ± 0.02cd | 0.008 ± 0.001g | – |
| S6 | 0.31 ± 0.02ghi | 2.86 ± 0.13g | 0.30 ± 0.01e | 0.25 ± 0.01de | 0.50 ± 0.02de | 0.116 ± 0.004a | 1.38 ± 0.06b | 0.25 ± 0.01 e | 0.017 ± 0.001bc | 0.003 ± 0.001a |
| S7 | 0.74 ± 0.02c | 2.37 ± 0.19h | 0.15 ± 0.01ij | 0.61 ± 0.03b | 0.30 ± 0.01b | 0.059 ± 0.002ef | – | – | 0.012 ± 0.001ef | – |
| S8 | 0.29 ± 0.0hij | 1.57 ± 0.02j | 0.63 ± 0.02c | 0.22 ± 0.01 e | 1.19 ± 0.03 e | 0.063 ± 0.003 e | – | 0.09 ± 0.01h | 0.003 ± 0.001ij | – |
| S9 | 0.43 ± 0.01 e | 1.59 ± 0.06j | 0.87 ± 0.02a | 0.63 ± 0.03b | 1.84 ± 0.05b | – | 1.35 ± 0.15b | 0.13 ± 0.01g | 0.013 ± 0.001de | – |
| S10 | 0.64 ± 0.01d | 1.69 ± 0.18j | 0.45 ± 0.02d | 0.46 ± 0.03c | 0.63 ± 0.01c | 0.108 ± 0.001b | 0.62 ± 0.03c | ‐ | 0.014 ± 0.001d | – |
| S11 | 0.27 ± 0.02ij | 4.34 ± 0.13cd | 0.08 ± 0.01k | 0.11 ± 0.01fg | 0.36 ± 0.02fg | 0.035 ± 0.001h | – | 0.23 ± 0.01ef | 0.017 ± 0.001c | – |
| S12 | 0.29 ± 0.02hij | 2.25 ± 0.03hi | 0.60 ± 0.03c | 0.24 ± 0.01de | 1.11 ± 0.01de | 0.033 ± 0.001h | – | 0.25 ± 0.02 e | 0.018 ± 0.001bc | – |
| S13 | 0.31 ± 0.02hi | 4.83 ± 0.22b | 0.17 ± 0.01ij | 0.11 ± 0.01fg | 0.39 ± 0.01fg | 0.057 ± 0.005f | 1.35 ± 0.09b | 0.25 ± 0.02 e | 0.017 ± 0.001bc | – |
| S14 | 0.21 ± 0.01k | 3.93 ± 0.18 e | 0.15 ± 0.01ij | 0.07 ± 0.01g | 0.39 ± 0.01g | 0.032 ± 0.001h | – | 0.21 ± 0.01f | 0.014 ± 0.001d | – |
| S15 | 0.34 ± 0.01fgh | 4.07 ± 0.13de | 0.24 ± 0.02fg | 0.13 ± 0.01f | 0.49 ± 0.01f | 0.072 ± 0.001d | – | 0.25 ± 0.01de | 0.019 ± 0.001ab | – |
| S16 | 0.88 ± 0.04a | 2.42 ± 0.09h | 0.18 ± 0.01hi | 0.94 ± 0.05a | 0.21 ± 0.01a | 0.098 ± 0.005c | 0.68 ± 0.02c | 0.32 ± 0.01bc | 0.011 ± 0.001f | – |
| S17 | 0.30 ± 0.01hij | 2.38 ± 0.04h | 0.03 ± 0.01L | 0.24 ± 0.01de | 0.17 ± 0.01de | 0.056 ± 0.002fg | – | 0.20 ± 0.01f | 0.004 ± 0.001hi | – |
| S18 | 0.29 ± 0.01hij | 4.60 ± 0.16bc | 0.12 ± 0.01jk | 0.12 ± 0.01fg | 0.42 ± 0.01fg | – | – | 0.21 ± 0.02f | 0.017 ± 0.001bc | 0.002 ± 0.001b |
| Sum total | 7.45 | 59.12 | 5.45 | 5.52 | 10.83 | 0.973 | 8.76 | 3.63 | 0.2 | 0.007 |
In this study, on the other hand, quercetin was higher in sea buckthorn leaf tea, which could be due to different processing methods and origins. Samples with higher isorhamnetin derivatives (isorhamnetin and isorhamnete‐3‐O‐neohesperidin) were S15 (Chaoyang, Liaoning) and S11 (Urumqi, Xinjiang). The study reported that the isorhamnetin derivatives content of sea buckthorn leaves was superior to that of quercetin derivatives (Pop et al., 2013). Overall, the highest quantity of nine flavonoids was found in the sea buckthorn leaf tea of S9 (Tacheng, Xinjiang), S8 (Xinjiang), and S1 (Xining, Qinghai).
Among all the test samples, ellagic acid total content was the highest at 59.12 mg/g, consistent with the report that tannin components are mainly found in sea buckthorn leaves (Wang et al., 2021). The study evaluated the sea buckthorn leaf profiles and reported that phenolic acid content is proportional to the flavonoid content (Raudone et al., 2021). However, in the present work, the samples with a more excellent ellagic acid content of sea buckthorn leaf tea all had lower total flavonoid content. This finding could be due to high‐temperature enzyme inactivation, thus better preserving the phenolic acids in the sea buckthorn leaf tea (Ma et al., 2019). The contents of catechins and apigenin were low in the sea buckthorn leaf tea.
The content of sea buckthorn leaf tea polyphenolic compounds is shown in Figure 1. The content of tannins ranged from 12.93 to 2.28 mg/g and the content of flavonoids ranged from 12.02 to 4.26 mg/g. S1 (Xining, Qinghai, 11.98 mg/g), S2 (Dingxi Gansu, 10.09 mg/g), and S9 (Tacheng, Xinjiang, 10.53 mg/g) had higher flavonoid content. S1 (Xining, Qinghai, 12.93 mg/g), S2 (Dingxi, Gansu, 9.05 mg/g), and S6 (Da Hinggan Ling Prefecture, Heilongjiang, 7.12 mg/g) had higher tannin content. In addition, the flavonoid and tannin contents of S17 (Tongliao, Inner Mongolia) were low. S17 was identified as sea buckthorn leaf black tea. *In* general, black tea decreases flavonoids and flavonoid glycosides after a specific processing step during fermentation (Feng et al., 2020). The bagged sea buckthorn leaf tea (S1–S7, 7.35 ± 2.85 mg/g) contained higher levels of tannins than the bulk sea buckthorn leaf tea (S8–S18, 3.01 ± 0.59 mg/g; $p \leq .01$). However, there was no significant difference in total flavonoids content between the two types of tea ($p \leq .05$). This finding illustrates the higher degree of crushing in bagged tea than in bulk tea. It also illustrates that proper crushing of tea leaves can increase the extraction rate of active ingredients (Danna et al., 2016).
**FIGURE 1:** *Polyphenols content of sea buckthorn leaf tea*
## Analysis of antioxidant activity
According to literature reports, evaluation of the antioxidant activities of natural antioxidants is difficult using a single method. Therefore, the processes involved in employing different assay principles are necessary (Higdon & Frei, 2003; Liu et al., 2019; Xu et al., 2017). In the present study, three commonly used different assays such as FRAP (reducing Fe3+ to Fe2+), ABTS (scavenging ATBS radical), and DPPH (removes DPPH free radicals) were combined to evaluate the antioxidant activities of tea extracts (Table 3).
**TABLE 3**
| Number | ABTS (mmol TE/g) | DPPH (mmol TE/g) | FRAP (mmol TE/g) | APC comprehensive index (%) |
| --- | --- | --- | --- | --- |
| S1 | 5.72 ± 0.34bc | 50.21 ± 2.71a | 17.94 ± 0.24a | 88.48 |
| S2 | 5.37 ± 0.61bc | 50.12 ± 1.28a | 16.72 ± 0.33bc | 84.8 |
| S3 | 9.84 ± 0.08 a | 45.01 ± 1.47bc | 16.81 ± 0.98bc | 96.73 |
| S4 | 6.75 ± 0.68b | 42.04 ± 0.83bcdef | 14.08 ± 0.34g | 78.85 |
| S5 | 9.58 ± 0.25a | 42.63 ± 1.14bcde | 14.51 ± 0.22efg | 89.68 |
| S6 | 0.81 ± 0.55h | 42.78 ± 0.38bcde | 13.95 ± 0.31g | 58.96 |
| S7 | 1.42 ± 0.33gh | 44.48 ± 1.77bcd | 16.18 ± 0.30cd | 66.6 |
| S8 | 4.91 ± 0.37cd | 45.64 ± 1.31bc | 17.77 ± 0.50a | 82.36 |
| S9 | 2.91 ± 0.78efg | 40.26 ± 1.12def | 13.95 ± 0.15g | 64.4 |
| S10 | 5.98 ± 0.62bc | 46.22 ± 4.25bc | 17.34 ± 0.56ab | 85.51 |
| S11 | 2.06 ± 0.86fgh | 44.69 ± 0.52bcd | 14.84 ± 0.12efg | 66.23 |
| S12 | 2.38 ± 0.79efgh | 43.20 ± 0.69bcde | 15.16 ± 0.39ef | 66.97 |
| S13 | 2.30 ± 0.72efgh | 41.36 ± 0.99cdef | 15.27 ± 0.35def | 65.69 |
| S14 | 1.44 ± 0.33gh | 41.97 ± 0.19bcdef | 15.13 ± 0.23ef | 62.9 |
| S15 | 3.29 ± 0.25ef | 38.19 ± 0.18fg | 14.42 ± 0.19fg | 65.25 |
| S16 | 1.24 ± 0.43h | 20.69 ± 0.61h | 12.00 ± 0.04h | 41.86 |
| S17 | 6.01 ± 0.66bc | 39.04 ± 1.68efg | 15.54 ± 0.13de | 77.26 |
| S18 | 3.82 ± 0.73de | 35.74 ± 1.17g | 14.10 ± 0.22g | 64.78 |
The FRAP values were 12.00 ± 0.04 to 17.94 ± 0.24 mmol TE/g for different tea beverages. The ABTS values varied widely among the different tea beverages, with the maximum value being 12 times higher than the minimum value. The DPPH values varied widely among the different tea beverages with a minimum value of 20.69 ± 0.61 and the maximum value of 50.21 ± 2.71 mmol TE/g. The results showed that sea buckthorn leaf tea had higher ability to scavenge DPPH free radicals than the two other modalities. The difference in correlation coefficients between the three different antioxidant methods may be due to the other measurement principles of different evaluation methods.
Antioxidant potency composite index was calculated to comprehensively compare the antioxidant activity from sea buckthorn leaf tea (Table 3). The ranked antioxidant activity from the largest to the smallest was as follows: S3 (Zhangye, Gansu) > S5 (Lvliang, Shanxi) > S1 (Xining, Qinghai) > S10 (Aksu, Xinjiang) > S2 (Dingxi, Gansu) > S8 (Xinjiang) > S4 (Xinzhou, Shanxi) > S17 (Tongliao, Inner Mongolia) > S12 (Kashgar, Xinjiang) > S7 (Xi'an, Shaanxi) > S11 (Urumqi, Xinjiang) > S13 (Jilin City, Jilin Province) > S15 (Chaoyang, Liaoning) > S18 (Tongliao, Inner Mongolia) > S9 (Tacheng, Xinjiang) > S14 (Chaoyang, Liaoning) > S6 (Da Hinggan Ling Prefecture, Heilongjiang) > S16 (Zibo, Shandong). The difference between the maximum and minimum values of the APC index is $33.83\%$. This finding was closely related to the content of flavonoids and phenolic acids in sea buckthorn leaf tea. Phenolic compounds, such as flavonoids, phenolic acids, and tannins, are known to be the major providers of antioxidant capacity in plants (Yang et al., 2019).
The average APC index of the bagged sea buckthorn leaf tea (S1–S7, 80.59 ± 13.48) was found to be stronger than that of the bulk sea buckthorn leaf tea (S8–S18, 67.56 ± 11.60; $p \leq .05$). This result highlighted that the antioxidant capacity of the bagged sea buckthorn leaf tea was stronger than that of the loose sea buckthorn leaf tea ($p \leq .05$). We speculate that this might be due to the higher degree of grinding tea leaves for bagged tea than that for loose sea buckthorn leaf tea; the degree of grind is positively correlated with antioxidant capacity (Q. Xu et al., 2021).
## Correlation analysis
To explore the relationship between antioxidant activity, sensory evaluation, and individual phenols and flavonoids, we calculated the Pearson's correlation coefficient (Tables 4 and 5). This is because polyphenolic compounds are an important factor affecting the antioxidant properties and taste of sea buckthorn leaf tea (Barbe et al., 2019; Delius et al., 2017).
As shown in Table 4, the results indicated that the stronger antioxidant capacity of sea buckthorn leaf tea samples may be due to the strong correlation between antioxidant values with tannins. The Pearson's correlation coefficients between total flavonoid content and antioxidant values were all lower than the Pearson's correlation coefficients between tannin content and antioxidant values. For individual compounds, isorhamnetin‐3‐O‐neohesperidoside has a certain correlation with antioxidant activity. In contrast to earlier findings, we find that ellagic acid was less correlated with antioxidant. We believe that the antioxidant activity of sea buckthorn leaf tea is influenced by species factors. For example, different tea steeping temperatures may all be responsible for the antioxidant activity of the compounds being affected (Pérez‐Burillo et al., 2018).
As shown in Table 5, polyphenolic compounds were positively correlated with bitterness, astringency, and arithmetic and negatively correlated with sweet and fresh. Importantly, ellagic acid and isorhamnetin‐3‐O‐neohesperidoside have correlation with astringent. Most of the available reported studies suggest that the astringency of tea liquids is caused by whole saliva or the interaction of proline‐rich proteins (PRPs) with polyphenols (Georgiades et al., 2014). Meanwhile, catechin and isorhamnetin‐3‐O‐neohesperidin showed moderate correlation with bitterness. This is consistent with the current report that the catechins are important in affecting the bitter taste of the tea infusion (J. Li et al., 2017).
## CONCLUSION
In this study, sea buckthorn leaf teas circulating in the China market were analyzed by UPLC, sensory evaluation, and in vitro antioxidant assay. The results indicate that the quality of sea buckthorn leaf tea varies greatly, and the polyphenol content, antioxidant activity, and sensory evaluation of bagged sea buckthorn leaf tea were better than those of bulk sea buckthorn leaf tea. Importantly, the taste astringency of sea buckthorn leaf tea was closely related to ellagic acid and isorhamnetin‐3‐O‐neohesperidin. Meanwhile, isorhamnetin‐3‐O‐neohesperidin had a greater effect on the antioxidant activity of sea buckthorn leaf tea. Therefore, ellagic acid and isorhamnetin‐3‐O‐neohesperidin are recommended as quality control indicators for sea buckthorn leaf tea. The study provides a reference for the taste improvement of sea buckthorn leaf tea and valuable data for the quality control of sea buckthorn leaf tea. It is beneficial for promoting and applying sea buckthorn leaf tea as potential functional beverage products.
## FUNDING INFORMATION
This research was funded by the Major Project of Strategic Research and Consultation of the Chinese Academy of Engineering (2021‐XZ‐10), the National Key Research and Development program (2021YFE0190100), the CAMS Innovation Fund for Medical Sciences (CIFMS ID: 2021‐I2M‐1‐071), and the Innovation Team and Talents Cultivation Program of National Administration of Traditional Chinese Medicine (ZYYCXTD‐D‐202005).
## CONFLICT OF INTEREST
All authors disclosed no relevant relationships.
## ETHICS STATEMENT
This article does not contain any studies performed with human participants or animals by any of the authors.
## CONSENT TO PARTICIPATE
Corresponding and all the coauthors are willing to participate in this manuscript.
## CONSENT FOR PUBLICATION
All authors are willing for publication of this manuscript.
## DATA AVAILABILITY STATEMENT
Even though adequate data have been given in the form of tables and figures, all authors declare that if more data are required, then the data will be provided on request basis.
## References
1. Barbe A., Rame C., Mellouk N., Estienne A., Bongrani A., Brossaud A., Riva A., Guérif F., Froment P., Dupont J.. **Effects of grape seed extract and proanthocyanidin B2 on In vitro proliferation, viability, steroidogenesis, oxidative stress, and cell signaling in human granulosa cells**. *International Journal of Molecular Sciences* (2019) **20** 4215. DOI: 10.3390/ijms20174215
2. Beveridge T., Li T. S., Oomah B. D., Smith A.. **Sea buckthorn products: Manufacture and composition**. *Journal of Agricultural and Food Chemistry* (1999) **47** 3480-3488. PMID: 10552673
3. Büyükokuroǧlu M. E., Gülçin I.. **In vitro antioxidant and antiradical properties of**. *Pharmacognosy Magazine* (2009) **4** 189-195
4. Ceci C., Lacal P. M., Tentori L., De Martino M. G., Miano R., Graziani G.. **Experimental evidence of the antitumor, antimetastatic and antiangiogenic activity of ellagic acid**. *Nutrients* (2018) **10** 1756. DOI: 10.3390/nu10111756
5. Cho H., Cho E., Jung H., Yi H. C., Lee B., Hwang K. T.. **Antioxidant activities of sea buckthorn leaf tea extracts compared with green tea extracts**. *Food Science and Biotechnology* (2014) **23** 1295-1303. DOI: 10.1007/s10068-014-0178-1
6. Ciesarová Z., Murkovic M., Cejpek K., Kreps F., Tobolková B., Koplík R., Belajová E., Kukurová K., Daško Ľ., Panovská Z., Revenco D., Burčová Z.. **Why is sea buckthorn (**. *Food Research International* (2020) **133**. DOI: 10.1016/j.foodres.2020.109170
7. Danna H., Xiaolei Z., Jianzhi S., Gong X., Wang S.. **Studies on cellulase‐ultrasonic assisted extraction technology for flavonoids from**. *Chemistry Central Journal* (2016) **10** 56. DOI: 10.1186/s13065-016-0202-z
8. Delius J., Frank O., Hofmann T.. **Label‐free quantitative 1H NMR spectroscopy to study low‐affinity ligand‐protein interactions in solution: A contribution to the mechanism of polyphenol‐mediated astringency**. *PLoS One* (2017) **12**. DOI: 10.1371/journal.pone.0184487
9. Feng L., Liu P., Zheng P., Zhang L., Zhou J., Gong Z., Yu Y., Gao S., Zheng L., Wang X., Wan X.. **Chemical profile changes during pile fermentation of Qingzhuan tea affect inhibition of alpha‐amylase and lipase**. *Scientific Reports* (2020) **10** 3489. DOI: 10.1038/s41598-020-60265-2
10. Geetha S., Sai Ram M., Mongia S. S., Singh V., Ilavazhagan G., Sawhney R. C.. **Evaluation of antioxidant activity of leaf extract of seabuckthorn (**. *Journal of Ethnopharmacology* (2003) **87** 247-251. DOI: 10.1016/s0378-8741(03)00154-5
11. Georgiades P., Pudney P. D., Rogers S., Thornton D. J., Waigh T. A.. **Tea derived galloylated polyphenols cross‐link purified gastrointestinal mucins**. *PLoS One* (2014) **9**. DOI: 10.1371/journal.pone.0105302
12. Gülçin L., Topal F., Beyza S., Öztürk S., Gören A.. **Polyphenol contents and antioxidant properties of Medlar (**. *Records of Natural Products* (2011) **5** 158-175
13. Higdon J. V., Frei B.. **Tea catechins and polyphenols: Health effects, metabolism, and antioxidant functions**. *Critical Reviews in Food Science and Nutrition* (2003) **43** 89-143. DOI: 10.1080/10408690390826464
14. **Validation of analytical procedures: Text and methodology**. (2005)
15. Jain M., Ganju L., Katiyal A., Padwad Y., Mishra K. P., Chanda S., Karan D., Yogendra K. M. S., Sawhney R. C.. **Effect of**. *Phytomedicine* (2008) **15** 793-799. DOI: 10.1016/j.phymed.2008.04.017
16. Köksal E., Gülçin I.. **Purification and characterization of peroxidase from cauliflower (**. *Protein and Peptide Letters* (2008) **5** 320-326
17. Lee H. I., Kim M. S., Lee K. M., Park S. K., Seo K. I., Kim H. J., Choi M. S., Lee M. K.. **Anti‐visceral obesity and antioxidant effects of powdered sea buckthorn (**. *Food and Chemical Toxicology* (2011) **49** 2370-2376. DOI: 10.1016/j.fct.2011.06.049
18. Li J., Xiao Q., Huang Y., Ni H., Wu C., Xiao A.. **Tannase application in secondary enzymatic processing of inferior Tieguanyin oolong tea**. *Electronic Journal of Biotechnology* (2017) **28** 87-94. DOI: 10.1016/j.ejbt.2017.05.010
19. Li T., Schroeder W. R.. *Sea buckthorn (Hippophae rhamnoides L.): A multipurpose plant* (1996)
20. Li Y., Li P., Yang K. L., He Q., Wang Y., Sun Y., He C., Xiao P. G.. **Impact of drying methods on phenolic components and antioxidant activity of sea buckthorn (**. *Molecules* (2021) **26** 7189. DOI: 10.3390/molecules26237189
21. Li Y., Liu Q., Wang Y., Zu Y. H., Wang Z. H., He C. N., Xiao P. G.. **Application and modern research progress of sea buckthorn leaves**. *China Journal of Chinese Materia Medica* (2021) **46** 1326-1332. DOI: 10.19540/j.cnki.cjcmm.20201211.601
22. Lima E. M. F., Madalao M. C. M., Dos Santos W. C., Bernardes P. C., Saraiva S. H., Silva P. I.. **Spray‐dried microcapsules of anthocyanin‐rich extracts from**. *Journal of Food Science and Technology* (2019) **56** 4147-4157. DOI: 10.1007/s13197-019-03885-5
23. Liu Q., Tang G. Y., Zhao C. N., Gan R. Y., Li H. B.. **Antioxidant activities, phenolic profiles, and organic acid contents of fruit vinegars**. *Antioxidants (Basel)* (2019) **8** 78. DOI: 10.3390/antiox8040078
24. Liu Y., Luo L., Liao C., Chen L., Wang J., Zeng L.. **Effects of brewing conditions on the phytochemical composition, sensory qualities and antioxidant activity of green tea infusion: A study using response surface methodology**. *Food Chemistry* (2018) **269** 24-34. DOI: 10.1016/j.foodchem.2018.06.130
25. Ma X., Moilanen J., Laaksonen O., Yang W., Tenhu E., Yang B.. **Phenolic compounds and antioxidant activities of tea‐type infusions processed from sea buckthorn (**. *Food Chemistry* (2019) **272** 1-11. DOI: 10.1016/j.foodchem.2018.08.006
26. Morikawa T., Akaki J., Ninomiya K., Kinouchi E., Tanabe G., Pongpiriyadacha Y., Yoshikawa M., Muraoka O.. **Salacinol and related analogs: New leads for type 2 diabetes therapeutic candidates from the Thai traditional natural medicine**. *Nutrients* (2015) **7** 1480-1493. DOI: 10.3390/nu7031480
27. Peng L. P., Men S. Q., Liu Z. A., Tong N. N., Imran M., Shu Q. Y.. **Fatty acid composition, phytochemistry, antioxidant activity on seed coat and kernel of**. *Food* (2019) **9** 30. DOI: 10.3390/foods9010030
28. Pérez‐Burillo S., Giménez R., Rufián‐Henares J. A., Pastoriza S.. **Effect of brewing time and temperature on antioxidant capacity and phenols of white tea: Relationship with sensory properties**. *Food Chemistry* (2018) **248** 111-118. DOI: 10.1016/j.foodchem.2017.12.056
29. Pop R. M., Socaciu C., Pintea A., Buzoianu A. D., Sanders M. G., Gruppen H., Vincken J. P.. **UHPLC/PDA‐ESI/MS analysis of the main berry and leaf flavonol glycosides from different Carpathian**. *Phytochemical Analysis* (2013) **24** 484-492. DOI: 10.1002/pca.2460
30. Pundir S., Garg P., Dviwedi A., Ali A., Kapoor V. K., Kapoor D., Kulshrestha S., Lal U. R., Negi P.. **Ethnomedicinal uses, phytochemistry and dermatological effects of**. *Journal of Ethnopharmacology* (2021) **266**. DOI: 10.1016/j.jep.2020.113434
31. Raudone L., Puzeryte V., Vilkickyte G., Niekyte A., Lanauskas J., Viskelis J., Viskelis P.. **Sea buckthorn leaf powders: The impact of cultivar and drying mode on antioxidant, phytochemical, and chromatic profile of valuable resource**. *Molecules* (2021) **26** 4765. DOI: 10.3390/molecules26164765
32. Scharbert S., Hofmann T.. **Molecular definition of black tea taste by means of quantitative studies, taste reconstitution, and omission experiments**. *Journal of Agricultural and Food Chemistry* (2005) **53** 5377-5384. DOI: 10.1021/jf050294d
33. Tanwar H., Singh D., Singh S. B., Ganju L.. **Anti‐inflammatory activity of the functional groups present in**. *Inflammopharmacology* (2018) **26** 291-301. DOI: 10.1007/s10787-017-0345-0
34. Trivedi V. L., Sati J.. **Seabuckthorn (**. *National Academy Science Letters* (2022) **45** 195-198. DOI: 10.1007/s40009-021-01095-2
35. Wang B., Zhang Q. S., Yang X. F., Zhang Y. Y., Guo H. L., Zhang X. M.. **Toxicologic study on corosolic acid concentrations of Seabuckthorn leaves**. *Journal of Shanxi Agricultural Sciences* (2017) **45** 203-205. DOI: 10.3969/j.issn.1002-2481.2017.02.14
36. Wang N. N., Zheng W. H., Zhang K. X., Wen X. F., Lu S. G., Yang Z. G.. **Research progress on chemical constituents and pharmacological activities of seabuckthorn and prediction of its Q‐markers**. *China Journal of Chinese Materia Medica* (2021) **46** 5522-5532. DOI: 10.19540/j.cnki.cjcmm.20210520.201
37. Xu D. P., Li Y., Meng X., Zhou T., Zhou Y., Zheng J., Zhang J. J., Li H. B.. **Natural antioxidants in foods and medicinal plants: Extraction, assessment and resources**. *International Journal of Molecular Sciences* (2017) **18** 96. DOI: 10.3390/ijms18010096
38. Xu Q., Huang R., Yang P., Wang L., Xing Y., Liu H., Wu L., Che Z., Zhang P.. **Effect of different superfine grinding technologies on the physicochemical and antioxidant properties of tartary buckwheat bran powder**. *RSC Advances* (2021) **11** 30898-30910. DOI: 10.1039/d1ra05093a
39. Yang H., Kim Y. J., Shin Y.. **Influence of ripening stage and cultivar on physicochemical properties and antioxidant compositions of Aronia grown in South Korea**. *Food* (2019) **8** 598. DOI: 10.3390/foods8120598
40. Ye J. H., Ye Y., Yin J. F., Jin J., Liang Y. R., Liu R. Y., Tang P., Xu Y. Q.. **Bitterness and astringency of tea leaves and products: Formation mechanism and reducing strategies**. *Trends in Food Science & Technology* (2022) **123** 130-143. DOI: 10.1016/j.tifs.2022.02.031
41. Yu P., Huang H., Zhao X., Zhong N., Zheng H., Gong Y.. **Distinct variation in taste quality of congou black tea during a single spring season**. *Food Science & Nutrition* (2020) **8** 1848-1856. DOI: 10.1002/fsn3.1467
42. Zhang L., Cao Q. Q., Granato D., Xu Y. Q., Ho C. T.. **Association between chemistry and taste of tea: A review**. *Trends in Food Science & Technology* (2020) **101** 139-149. DOI: 10.1016/j.tifs.2020.05.015
|
---
title: Stimulation of ACE inhibitory and improving α‐amylase and α‐glucosidase and
antioxidant activities of semi‐prepared and dry soup by incorporating with date
kernel powder
authors:
- Maryam Mousavi
- Vajiheh Fadaei
- Behrouz Akbari‐adergani
journal: Food Science & Nutrition
year: 2022
pmcid: PMC10003009
doi: 10.1002/fsn3.3170
license: CC BY 4.0
---
# Stimulation of ACE inhibitory and improving α‐amylase and α‐glucosidase and antioxidant activities of semi‐prepared and dry soup by incorporating with date kernel powder
## Abstract
Date kernel as a functional food component has a special importance due to its rich nutritional profile, low price, and ease of access. For this, in this research, the sub‐product was used for formulation of semi‐prepared dry soup (SPDS); the effect of adding 0 (S1 = control), 2 (S2), and 4 (S3) %w/w date kernel powder (DKP) on physicochemical, nutritional, and organoleptic properties and beneficial effects of SPDS samples were evaluated. The results revealed that S2 and S3 samples were different from the control sample in some physicochemical properties so that viscosity increased 1.27 and 1.52 times and a* raised 5.6 and 8.5 times, respectively, while L* decreased 0.94 and 0.88 times and b* reduced 0.92 and 0.8 times, respectively. The nutritional properties of S2 and S3 samples compared with the control sample improved. Also, differences were observed in the beneficial effects of S2 and S3 compared with the control sample as total polyphenol content (TPC) increased 1.06 and 1.11 times, respectively ($p \leq .05$); antioxidant activities (AA) of S2 and S3 samples were 8.04 and 6.01 mg/ml and angiotensin‐converting enzyme (ACE) inhibitory activities were measured to be 8.2 and 7.86 mg/ml, respectively; also, α‐amylase and α‐glucosidase inhibitory activities of S2 and S3 samples were observed $4.48\%$ and $5.70\%$, and $4.59\%$ and $6.36\%$, respectively. From the organoleptic aspect, S3 had the highest acceptability. Generally, it is concluded that with the addition of DKP (maximally $4\%$w/w) to SPDS formulation, a functional soup could be produced considering the rich nutritional profile of DKP.
Date kernel powder (DKP) could be considered as the potential functional ingredient in healthy food products. DKP improved α‐amylase and α‐glucosidase and antioxidant activities of semi‐prepared and dry soup (SPDS). SPDS with improved nutritional properties was provided using DKP incorporation.
## INTRODUCTION
Semi‐prepared dry soup (SPDS) is produced in a dried form after processing and packaged in a suitable way; it is consumable after the addition of water or other suitable liquids as well as desirable food components and boiling. SPDSs have received a lot of attention in recent years because of their favorable shelf life, easy transportation, and short time required for their preparation (Niththiya et al., 2014). Now, if functional ingredients are employed in preparation of this type of soups, their value will be doubled. Since eating soup is the easiest way for receiving nutrients and with increasing workload and not having enough time for most people in the community to prepare this type of product in the food industry, ready‐to‐eat soups are produced in powder form, it is expected that by using a functional component as a value‐added food ingredient, soups with high nutritional and healthful values are prepared and people can receive a major portion of nutrients and healthful components through the consumption of soup.
In recent years, with the introduction of functional foods and increasing consumer awareness of the relationship between health and food consumption, the demand for foods with significant health‐giving effects has increased. Currently, food factories are trying to generate a product with numerous health‐giving effects. The health‐giving effects of functional foods are attributed to their healthful bioactive compounds, and date kernel, due to its nutritional and health‐giving properties mentioned below, is among the compounds that can be used as a functional food ingredient.
Depending on the date variety, date kernel comprises $10\%$ (on average) of the total weight of the date and it is produced in large quantities as waste in date processing factories. Date kernel is odorless and has a light to dark brown color and a slightly mild bitter taste (Hamada et al., 2002). Phenolic acids, flavonoids (epicatechin, catechins, procyanidins, quercetin, naringinin, rutin, luteolin, chrysoerio, tricin, and apigenin), anthocyanins, carotenoids (β‐carotene, lutein), tocopherols (α and γ‐tocopherol, tocotrienol), phytosterols (β‐sitosterols), fatty acids (oleic, linoleic, palmitic, and lauric), and carboxylic acids have been reported among bioactive compounds of this sub‐product (Alharbi et al., 2021). Date kernel contains $7.1\%$–$10.3\%$ moisture, $46\%$–$51\%$ acid detergent fiber, $65\%$–$69\%$ neutral detergent fiber, $5\%$–$6.3\%$ protein, $9.9\%$–$13.5\%$ fat, and $1\%$–$1.8\%$ ash (Hamada et al., 2002). TPC of date kernel is 48.64 mg gallic acid/100 g date. The dietary fiber of date kernel is $58\%$ from which $53\%$ is soluble dietary fiber. Semi‐processed date kernel has equal protein to wheat flour (Wahini, 2016). Date kernel protein plays an important biological role in our body due to its essential amino acids (Bouaziz, 2020). Different insoluble fibers including cellulose and hemi‐cellulose are major carbohydrates of date kernel (Ghnimi et al., 2017). From ancient time, in the Iran's regions under date cultivation, date kernel powder (DKP) has been traditionally used to treat some diseases such as hypertension, headache, etc. Date kernel has unique properties such as antioxidant, antimicrobial, antiviral, hypoglycemic, and anti‐cancer characteristics (Al‐Qarawi et al., 2005).
Considering the available literature, most of the studies are devoted to the effect of DKP on bread quality and bakery products and except in one case (Ambigaipalan & Shahidi, 2015), functional properties of prepared food products have not been discussed (Almana & Mahmoud, 1994; Bouaziz et al., 2010, 2020; Ghasemi et al., 2020; Najjar et al., 2022; Platat et al., 2014, 2015; Vinita, 2016). Of course, some researchers evaluated physicochemical and organoleptic properties of non‐bakery products rich in DKP or date kernel fiber (Amany et al., 2011; Asghari‐pour et al., 2018; Bouaziz, 2020; El Sheikh et al., 2014; Fikry et al., 2019; Wedad & El‐Khol, 2018). It should be noted that none of these studies have mentioned the adverse toxicological effects of date kernel powder; however, El Sheikh et al. [ 2014] have confirmed in vitro cytotoxic activity of roasted date kernels. Also, considering the available literature, no research on the evaluation of the effect of this valuable and cheap functional sub‐product on the quality of dry soup has still been undertaken (Abdel‐Haleem & Omran, 2014; Niththiya et al., 2014; Olubi et al., 2021; Upadhyay et al., 2017). On the other hand, based on the studies, the highest content of antioxidant compounds in dates is included in their kernels, which encompass phenolic compounds (Wahini, 2016). Also, date kernel has therapeutic effects and high level of dietary fibers which can be a good source for investigating α‐amylase and α‐glucosidase inhibitory activities. Nevertheless, a large amount of this functional sub‐product is wasted in factories producing date by‐products. For this reason, in this research, DKP is used as a health‐giving food component in catering functional SPDS. On this basis, the goal of this research was to formulate SPDS based on cereals and DKP and evaluation of its properties from different aspects.
## Materials
Olein palm oil from Margarin Public Company, wheat and barley flakes from Abshar Sabz, Barbari wheat flour from Setareh Mortazavi Company, refined iodine‐free salt from Pars Namak Kaveh Company, orange lentil from Golestan Company, date (*Phoenix dactylifera* L.) kernel powder (variety Mazafati) from Pilar Company, onion powder from Rashak Company, whey protein powder from Shirpooyan Company, citric acid from Ensign Company, garlic powder from Rashak Company, Meat Flavor from Symrise Company, monosodium glutamate from Foodchem and spices from Behsab Company were prepared. All chemicals required for carrying out tests were supplied by Merck Company except sucrose, angiotensin‐converting enzyme (ACE), diphenyl 2‐picrylhydrazyl, hypuryl‐His‐Leu, α‐amylase and α‐glucosidase purchased from Sigma Company.
## Preparation of semi‐prepared dry soup samples
All required materials were approved by the quality control unit were weighed individually (according to Table 1) for the preparation of SPDS samples. Subsequently, all weighed materials were mixed in a special mixer and transferred to a particular silo for packaging in 75 g laminated bags (three layers of low‐density polyethylene foil/aluminum/low‐density polyethylene). Random sampling was carried out from the produced soups, and the relevant tests were performed 7 days after production.
**TABLE 1**
| Integration | Treatment | Treatment.1 | Treatment.2 |
| --- | --- | --- | --- |
| Integration | S1 (%) | S2 (%) | S3 (%) |
| DKP | 0 | 2 | 4 |
| Wheat and barley flakes | 64.2 | 62.2 | 60.2 |
| Barbari wheat flour | 12 | 12 | 12 |
| Olein palm oil | 8 | 8 | 8 |
| Salt | 6 | 6 | 6 |
| Orange lentil | 2.7 | 2.7 | 2.7 |
| Onion powder | 2 | 2 | 2 |
| Whey protein powder | 1.6 | 1.6 | 1.6 |
| Monosodium glutamate | 1 | 1 | 1 |
| Citric acid | 0.6 | 0.6 | 0.6 |
| Spices (cumin, cinnamon, turmeric) | 1.1 | 1.1 | 1.1 |
| Garlic powder | 0.7 | 0.7 | 0.7 |
| Meat flavor | 0.1 | 0.1 | 0.1 |
| Total | 100 | 100 | 100 |
It should be mentioned that since the color of DKP had a great effect on the color of soups, first, the formulation was designed to prepare a control soup sample so that its color would not be substantially affected by the addition of low percentages of DKP. After the approval of organoleptic properties of control soup sample by 10 trained panelists, the next stage of research was started for the determination of experimental treatments. Indeed, 15 treatments having $0\%$–$7\%$ DKP ($0\%$, $0.5\%$, $1\%$, $1.5\%$, $2\%$, $2.5\%$, $3\%$, $3.5\%$, $4\%$, $4.5\%$, $5\%$, $5.5\%$, $6\%$, $6.5\%$, and $7\%$) were prepared as soup samples enriched with higher than $4\%$ DKPs had an astringent taste and a dark color; thus, they were unacceptable by panelists and removed, and samples were selected according to Table 1.
## Experiments on date kernel powder and soup samples
For preparation of soups, 75 g SPDS was mixed with 1 L cold water and boiled for 20 min at mild temperature. Then, the prepared and cooled soup was poured in a mixer, and the mixed sample (with ambient temperature) was tested.
## Physicochemical properties
Titratable acidity was calculated by titrating the samples with 0.1 N NaOH (Niththiya et al., 2014) and pH was measured by a pH meter (Schott GERATE; Fikry et al., 2019).
Moisture content was evaluated using dry air oven (Fikry et al., 2019).
Color parameters were measured with a Chroma‐meter (TES 135A, Taiwan) and reported in terms of L*, a*, and b* values (Fikry et al., 2019).
The viscosity of soup samples was measured by a rotational Brookfield viscometer (RVDV‐ II+ pro). After initial tests, DV64 was selected as the suitable spindle and viscosity was reported at room temperature in terms of Pa/s after 30 s (Abdel‐Haleem & Omran, 2014).
For calculation of rehydration, 2 g of functional semi‐prepared dry soup was mixed with 20 ml distilled water at a fixed speed (100 rpm) and rehydrated in a water bath at a given temperature (45°C) for 10 min; then, samples were filtered, weighed and the relevant factor was calculated (Abdel‐Haleem & Omran, 2014).
## Nutritional properties
Fat content was determined using Soxhlet extraction (El Sheikh et al., 2014), protein content by Kjeldahl method (Bouhlali et al., 2015), carbohydrate content by titration (McCleary et al., 2019), minerals (iron, zinc, sodium, potassium, calcium, magnesium, copper, and selenium) by Bouhlali et al. [ 2015] method, ash content by direct ignition (El Sheikh et al., 2014), and soluble, insoluble as well as total fibers by Abdel‐Haleem and Omran [2014] method were measured.
Total calories were estimated by Equation [1] (Bouhlali et al., 2015): [1] Total energycalories=Total fatsper100g×9+Total carbohydratesper100g×4+Total proteinsper100g×4.
## Beneficial effects
The total polyphenol content (TPC) was calculated through the Folin–Ciocalteu method and reported in terms of mg gallic acid per 100 g of sample (Fikry et al., 2019). According to the Folin–Ciocalteu procedure, the sample and Folin–Ciocalteu reagent were thoroughly mixed in a volumetric flask. After 3 min, 5 ml of $10\%$ Na2CO3 solution was added, and the mixture was left for 1 h. The absorbance of the mixture was determined at 760 nm by using a spectrophotometer (UV2100). The total concentration of phenolic compounds was determined by comparison with the absorbance of chlorogenic acid as standard.
Antioxidant activity (AA) was measured through the evaluation of free radical scavenging capacity of 2,2‐diphenyl−1‐picrylhydrazyl (DPPH) and reported in terms of IC50 (Bouhlali et al., 2015). IC50 is inhibitory concentration (mg/ml) required for the reduction of DPPH radicals to $50\%$ of its original content and able to inhibit $50\%$ of its activity. The lower IC50, the higher antioxidant capacity. The sample was mixed with 0.4 mmol/LDPPH radical in ethanol. The mixture was vigorously shaken and left for 10 min. The absorbance of the mixture was determined at 525 nm with a spectrophotometer (UV2100). The radical scavenging activity was calculated using Equation [2] as follows: [2] Inhibition%=1−abssample/abscontrol×100.
To measure ACE inhibition, first, 75 g of sachet content was mixed in individual containers with 1 L cold water and boiled gradually by thermal energy for 20 min while stirring; after cooling down to room temperature, it was filtered through Whatman filter paper and the filtered solution was assayed through Balthazar et al. [ 2019] method. In brief, to determine the amount of ACE inhibitory activity, 80 μl of the filtered solution was mixed with 200 μl of 5 mM hyporyl L‐histidyl L‐leucine (HHL) solution and incubated for 3 min at 37°C. Samples and HHL were prepared in 100 mM borate buffer (pH = 3.8) containing 300 mM sodium chloride. Next, 20 μl of 0.1 U/ml solution of ACE was added. Then, they were incubated for 30 min; and after that, the enzymatic reaction was stopped by adding 250 μl of 0.05 M HCl solution. The freed hippuric acid was extracted with about 2 ml of ethyl acetate and evaporated at a temperature of about 90°C for about 10 min. The residue was dissolved in 1 ml of distilled water and its absorbance was read at 228 nm by an ELISA reader. ACE inhibitory activity reported as IC50 was calculated according to Equation [3]. IC50 is inhibitory concentration for inhibition of $50\%$ of ACE activity in terms of mg/ml. Lower IC50 values show higher inhibition percentages. [ 3] ACEI$.\%$=B−AB−C×100. A: hippuric acid absorption in the presence of ACE inhibitor. B: absorption in the absence of ACE inhibitor. C: absorption in the absence of ACE.
α‐amylase and α‐glucosidase inhibitory activities were measured through Laoufi et al. [ 2017] method. The reaction mixture for determining the α‐amylase inhibitory activity contains the following: 200 μl of the filtered solution, 200 μl of 0.02 M sodium phosphate buffer (pH 6.9; 6.7 mM NaCl) containing 1.3 U/ml of porcine pancreatic α‐ amylase solution (PPA). The reaction medium was pre‐incubated at 37°C for 5 min, and then 200 μl of $0.4\%$ starch solution in the above buffer was added and incubated at 37°C for 10 min. Six hundred microlitersl of DNSA solution was added to the reaction and placed in a boiling water bath for 7 min, then cooled down in cold water. The reaction mixture was then diluted after adding 1 ml of distilled water and the absorbance was measured at 540 nm. To eliminate the absorbance produced by filtered solution, appropriate solution controls with the filtered solution and except the enzyme were also included. Commercial inhibitor acarbose was used as a positive control at a concentration range of 0.040–2.670 mg/ml. As a blank buffer solution was used instead of substrate. The tube with enzyme solution but without the filtered solution/acarbose served as the control with total enzyme activity. For determining the α‐glucosidase inhibitory activity, the filtered solution (500 μl) and 1000 μl of 0.1 M potassium phosphate buffer (pH 6.90) containing α‐glucosidase solution (1.0 U/ml) were incubated in water bath (25°C) for 10 min. p‐Nitrophenyl‐α‐d‐glucopyranoside solution (5 mM, 500 μl) in 0.1 M potassium phosphate buffer (pH 6.90) was then added to each tube and the mixtures were re‐incubated at 25°C for 5 min. Absorbance readings were recorded at 405 nm and the control (buffer in place of the filtered solution) was recorded as well.
The enzyme inhibition rate expressed as a percentage of inhibition was calculated using Equation [4]. [ 4] Inhibition of enzyme activity%=C−S/C×100. C: absorbance of the control ($100\%$ enzyme activity). S: absorbance of the tested sample.
## Organoleptic properties
Organoleptic properties of DKP were measured through evaluation of its color, flavor, taste, and odor.
Organoleptic properties such as flavor, odor, oral and non‐oral texture (mouthful, pouring, stirring, and spoon able), color, and overall acceptability of soup samples were measured by a 5‐point hedonic system. Evaluation levels included (from 1 to 5): 1 = inconsumable or very weak, 2 = unacceptable or weak, 3 = acceptable or average, 4 = satisfactory or good, 5 = very satisfactory or very good (Abdel‐Haleem & Omran, 2014).
Sensory evaluation was carried out by 100 untrained panelists (30 men and 70 women in the range of 25–60 years).
Table 6 shows the changes in the scores of organoleptic properties of soup samples. The scores of organoleptic properties (except color) of soup samples increased through increasing the percentage of DKP ($p \leq .05$). While the highest scores of organoleptic properties (except color) allocated to S3 sample, the highest score of color property belonged to S2 and the lowest score of color property to S3. Date kernel could be used as a food component to improve the sensory properties of food products, and the decrease in water activity affects product stability (Bouaziz et al., 2010). In this research, brightness was reduced as a result of the increase in the percentage of DKP and samples became darker. Texture improvement could be due to higher consistency as a result of the increase in the amount of DKP in soup samples. Food components containing dietary fibers might play a role as texture‐modifying properties in cereal‐based food products, in addition to their functional characteristics. The satisfaction level of panelists of taste might be owing to the interaction of special taste of DKP with other formulation ingredients in soup samples. The soup sample containing $4\%$ of DKP had the highest overall acceptability because of its favorable taste and odor as well as suitable texture. Increased score (of sensory evaluation) and darkness intensity of other products enriched with appropriate amounts of DKP have been reported as well; It should be noted that the color darkness caused by DKP may be desirable or undesirable depending on the product type (Almana & Mahmoud, 1994; Amany et al., 2011; Ambigaipalan & Shahidi, 2015; Ghasemi et al., 2020; Halaby et al., 2014; Shokrollahi & Taghizadeh, 2016; Vinita, 2016).
**TABLE 6**
| Characteristics treatment | Flavor | Odor | Color | Texture | Overall acceptability |
| --- | --- | --- | --- | --- | --- |
| S1 (DKP = 0) | 3.7b ± 0.5 | 3.7b ± 0.5 | 4.0b ± 0.0 | 4.0b ± 0.0 | 3.7b ± 0.5 |
| S2 (DKP = 2) | 3.9b ± 0.8 | 3.9b ± 0.8 | 4.8a ± 0.4 | 4.4a ± 0.5 | 4.0b ± 0.7 |
| S3 (DKP = 4) | 4.7a ± 0.5 | 4.7a ± 0.5 | 4.1b ± 0.6 | 4.6a ± 0.5 | 4.8a ± 0.4 |
## Statistical analysis
A completely randomized design was used to analyze and evaluate the data obtained from the different experiments. Analysis of variance and comparison of means (three replicates for each experiment) were carried out through SPSS software version 16.0 (Duncan's multiple range test and confidence level of $95\%$).
## Titratable acidity and pH
Table 3 shows pH and acidity changes of soup samples. With the increase in DKP percentage in functional soup samples, pH decreased and acidity increased ($p \leq .05$), and S1 and S3 had the highest pH value and acidity, respectively. Reduced pH and elevated pH could be attributed to the low pH and high acidity of DKP (Table 2). The reason for high acidity of DKP is the presence of acidic compounds such as caffeic acid, gallic acid, p‐coumaric acid, ferulic acid, and vanillic acid in it (Al Harthi et al., 2015; Al‐Farsi & Lee, 2011).
**TABLE 2**
| Characteristics | Treatment |
| --- | --- |
| pH | 5.63 |
| Titratable acidity (g/100 g) | 0.003 |
| Moisture (g/100 g) | 2.93 |
| L a | 16.91 |
| a a | 14.28 |
| b a | 7.18 |
| Fat (g/100 g) | 15.07 |
| Protein (g/100 g) | 6.98 |
| Carbohydrate (g/100 g) | 77.03 |
| Ash (g/100 g) | 0.94 |
| IDF (g/100 g) | 23.15 |
| SDF (g/100 g) | 34.18 |
| TDF (g/100 g) | 57.20 |
| K (ppm) | 1857 |
| Cu (ppm) | 6455.8 |
| Ca (ppm) | 1187 |
| Mg (ppm) | 1244 |
| Na (ppm) | 146.7 |
| Fe (ppm) | 2492 |
| Zn (ppm) | 265.83 |
| Se (ppm) | 2682 |
| Calori (kcal/kg) | 4237 |
| AA (mg/ml) | 0.074 |
| TPC (mg GAE/g) | 1.67 |
| ACE inhibitory activity (mg/ml) | 9.1 |
| α‐glucosidase Inhibition (%) | 19.9 |
| α‐amylase Inhibition (%) | 34.57 |
| Sensory properties | Good |
## Moisture, viscosity, and rehydration
Table 3 shows the changes in moisture content, viscosity, and rehydration of soup samples. Increasing the percentage of DKP in soup samples decreased moisture content and rehydration ($p \leq .05$) and promoted viscosity ($p \leq .05$); viscosity of S2 and S3 was 1.27 and 1.52 times higher than control sample, and, in fact, S3 had the highest viscosity.
**TABLE 3**
| Treatment characteristics | S1 (DKP = 0) | S2 (DKP = 2) | S3 (DKP = 4) |
| --- | --- | --- | --- |
| pH | 4.6a ± 0.0 | 4.6a ± 0.01 | 4.45b ± 0.05 |
| Acidity (g/100 g) | 0.0076b ± 0.00037 | 0.0074b ± 0.00000 | 0.0082a ± 0.00037 |
| Moisture (g/100 g) | 5.14a ± 0.053 | 4.98a ± 0.624 | 4.80a ± 0.264 |
| L * | 61.9a ± 0.1 | 57.9b ± 0.4 | 54.4c ± 0.8 |
| a * | −1.47c ± 0.115 | 3.15b ± 0.359 | 6.03a ± 0.302 |
| b * | 17.86a ± 0.066 | 16.37b ± 0.865 | 14.30c ± 0.36 |
| Viscosity (Cp) | 969c ± 51 | 1221b ± 87 | 1472a ± 127 |
| Rehydration (g/100 g) | 874a ± 108 | 599ab ± 154 | 542b ± 172 |
Reduced moisture content is ascribed to the availability of insoluble cellulose fibers, hemi‐cellulose, and lignin in DKP, which, owing to the lack of ability to absorb and hold water, caused the decrease in moisture content of DKP (Al‐Farsi & Lee, 2008). Also, hydration (swelling and water retention capacity) characteristics of date kernel dietary fiber have been reported to be lower than fibrex (Shokrollahi & Taghizadeh, 2016). The dwindled moisture content of soup samples enriched with DKP is due to the replacement of DKP with wheat ($7\%$ moisture) and barley ($5\%$ moisture) as these ingredients possess higher moisture content than DKP ($3\%$ moisture).
The viscosity of final product is dependent on the type of raw/initial material. Abdel‐Haleem and Omran [2014] introduced viscosity as the indicator of concentration in dried vegetable soup fortified with legumes so the addition of thickening components to the soup formulation affects the viscosity of the product; also, they reported that when dry foods are recombined, they should exhibit acceptable textural, visual, and sensory properties with a minimum of hydration time. Niththiya et al. [ 2014] reported increased viscosity after formulating instant soup powder with Borassus flabellifer tuber flour and local vegetables. Compounds with the capacity of moisture absorption result in the increase in viscosity and stability of product (Elleuch et al., 2011; Figuerola et al., 2005). The ability to absorb water in DKP might be attributed to soluble dietary fiber and protein. Water‐soluble polysaccharides of date kernel also own the ability to absorb water and increase viscosity. Because of hydroxyl groups and ability to interact with water molecules through hydrogen bonds and, as a result, increased ability of water holding, the fiber of DKP causes increased viscosity and moisture of enriched product (Bouaziz et al., 2010, 2020; Najjar et al., 2022).
The results of researches related to the effect of moisture are different based on the formulation of products enriched with DKP. Ambigaipalan and Shahidi [2015] reported that the increase in DKP leads to the increase in moisture content in muffin samples. Ghasemi et al. [ 2020] and Platat et al. [ 2015] expressed no significant change in moisture content of sponge cake and pita bread after elevating the percentage of DKP. However, Asghari‐pour et al. [ 2018] and Vinita [2016] stated the reduced moisture content of puff and biscuit samples, respectively, after the addition of DKP. Also, Shokrollahi and Taghizadeh [2016] reported an increase in firmness of bread fortified with higher than $2.5\%$ of DKP.
## Color
Table 3 displays the changes in color indices of soup samples. L*, a*, and b* are representatives of brightness, redness‐greenness, and blueness‐yellowness, respectively. Intensification of the percentage of DKP in soup samples decreased L* and b* values but increased a* ($p \leq .05$). Brightness index of S2 and S3 was 0.94 and 0.88 lower compared with the control sample, while their a* was 5.6 and 8.5 times higher and b* was 0.92 and 0.80 lower than those of the control sample.
Natural colorants of date kernel such as flavonoids, anthocyanins, and carotenoids affect its color (Alharbi et al., 2021). These colorants are polyphenols of DKP (Bouaziz et al., 2020). The intensity of the soup color increased with increasing the level of replacement of barley and wheat flakes with DKP, which indicates a significant direct relationship between the amount of DKP and darkness of soup color. Also, the darkness of the product can be related to the presence of short‐chain carbohydrates and proteins in DKP, which intensify the Millard reaction in the product, and, as a result, decreases the brightness of product (Bouaziz et al., 2010, 2020). The color of DKP influences a* value too. The reduction of b* value reveals yellowness of soup samples. It should be noted that the color of DKP is very different from the color of wheat and barley flakes, and, as a result, impacts the darkness of the product.
Platat et al. [ 2014] stated that the color of date kernel could be representative of its antioxidant and nutritional properties so that the more intense the greenness‐redness is, the higher the phenolic compounds and total antioxidant capacity will be and the more intense the blueness‐yellowness is, the higher the phenolic compounds, flavonoids, catechins, total antioxidant capacity and the lower the dietary fiber will be. Bouaziz [2020] reported similar results about color values after the addition of insoluble fibers of date kernel as insoluble fiber concentrate to turkey burgers. Also, Shokrollahi and Taghizadeh [2016] and Bouaziz et al. [ 2010, 2020] confirmed the decrease in brightness of bread enriched with fiber and DKP. Najjar et al. [ 2022] attributed the darkness of cookies to the darkness of added date kernel.
## Fat, protein, carbohydrate, and ash
Table 4 displays the changes in fat, protein, carbohydrate, and ash content of soup samples. Increasing the percentage of DKP resulted in the increase in fat content of functional soup samples ($p \leq .05$), partial decrease in protein content and no significant change in carbohydrate and ash contents.
**TABLE 4**
| Treatment characteristics | S1 (DKP = 0) | S2 (DKP = 2) | S3 (DKP = 4) |
| --- | --- | --- | --- |
| Fat (g/100 g) | 13.49b ± 0.04 | 13.57a ± 0.04 | 13.61a ± 0.04 |
| Protein (g/100 g) | 11.03a ± 0.11 | 10.96a ± 0.04 | 10.91a ± 0.04 |
| Carbohydrate (g/100 g) | 70.4a ± 0.11 | 70.4a ± 0.03 | 70.4a ± 0.10 |
| Ash (g/100 g) | 5.052a ± 0.03 | 5.053a ± 0.02 | 5.052a ± 0.02 |
| IDF (g/100 g) | 11.6c ± 0.15 | 12.1b ± 0.06 | 12.7a ± 0.11 |
| SDF (g/100 g) | 8.6c ± 0.17 | 8.9b ± 0.08 | 9.3a ± 0.02 |
| TDF (g/100 g) | 20.2c ± 0.28 | 20.9b ± 0.08 | 22.0a ± 0.12 |
| K (ppm) | 11381a ± 51 | 11083a ± 51 | 11083a ± 51 |
| Na (ppm) | 26958a ± 15 | 26596ab ± 15 | 26234b ± 27 |
| Ca (ppm) | 800b ± 0.00 | 813ab ± 11 | 827a ± 11 |
| Mg (ppm) | 976b ± 12.2 | 996ab ± 7.0 | 1013a ± 21.1 |
| Fe (ppm) | 1100c ± 25 | 1158b ± 14 | 1217a ± 14 |
| Zn (ppm) | 20.8c ± 2.9 | 30.8b ± 1.4 | 48.3a ± 3.8 |
| Cu (ppm) | 63.5c ± 0.00 | 201.1b ± 18.3 | 328.1a ± 18.3 |
| Se (ppm) | 417.6c ± 11.3 | 482.7b ± 22.6 | 534.8a ± 11.3 |
| Calori (KCal/Kg) | 4250 ± 1 | 4240 ± 1 | 4230 ± 1 |
It is worth noting that date kernel oil has high antioxidant activity as well as considerable phenolic compounds (Besbes et al., 2005). Tafti et al. [ 2017] reported that amount of extracted oil date kernel is quite low. The slight decrease in the protein content of soup samples containing DKP compared with the control soup is due to the low protein content of DKP. Considering the replacement of wheat and barley flakes with DKP in the formulation of soup samples, reduction of the protein amount is pretty reasonable. Most of date kernel proteins are insoluble (Hamada et al., 2002). The lack of change in ash and carbohydrate contents of enriched soup samples is due to the similarity of ash and carbohydrate content of DKP to wheat flour ash (Hamada et al., 2002). Generally, the ash content of date kernel is low.
Almana and Mahmoud [1994] reported slight decrease in protein content of Mafrood flat bread containing DKP compared with the control sample. Ambigaipalan and Shahidi [2015] did not observe any change in fat content of muffin samples after the addition of DKP but reported a significant ash increase compared with the control sample. Ghasemi et al. [ 2020] expressed an increase in protein content and no change in acid‐insoluble ash of sponge cake samples after intensifying the percentage of DKP. The results of the research of Asghari‐pour et al. [ 2018] also showed the low ash content of puff samples enriched with DKP. Vinita [2016] did not notice any significant change in protein, fat, and ash contents of biscuit samples enriched with DKP compared with the control sample. Platat et al., 2015 did not observe any significant change in the protein content of pita bread enriched with DKP but its fat content increased.
## Soluble, insoluble, and total fibers
Table 4 shows the changes in different types of fibers of soup samples. With the increase in the percentage of DKP, different types of fibers raised in soup samples ($p \leq .05$). The comparison of different types of fibers of enriched soup samples (S2 and S3) with the control soup shows that their SDFs are 1.04 and 1.1 times, IDFs are 1.03 and 1.05 times, and TDFs are 1.03 and 1.09 times higher, respectively; S3 had the highest amount of different fiber types. The higher fiber content of DKP compared with wheat and barley is the reason for the observed results. The high fiber content of DKP provides the possibility of its application as a rich source of fibers in different food formulations (Al‐Farsi & Lee, 2008, 2011; Baliga et al., 2011; Bouaziz et al., 2010; Elleuch et al., 2011; Habib & Ibrahim, 2009; Hamada et al., 2002).
In line with our results, the higher fiber content of foods enriched with date kernel powder has been reported (Almana & Mahmoud, 1994; Ambigaipalan & Shahidi, 2015; Asghari‐pour et al., 2018; Ghasemi et al., 2020; Platat et al., 2015; Vinita, 2016).
## Minerals
Table 4 shows the changes in mineral contents of soup samples. With an increase in the percentage of DKP, calcium, iron, zinc, copper, and selenium content of functional soup samples went up ($p \leq .05$). The comparison of these minerals in enriched soup samples (S2 and S3) with the control soup sample shows their calcium content was 1.01 and 1.03 times, the iron content was 1.05 and 1.11 times, the zinc content was 3.2 and 5.2 times and selenium content was 1.16 and 1.28 times higher than those of the control sample, respectively; S3 allocated the highest minerals contents to itself. The sodium content of soup samples dwindled as a consequence of the increase in percentage of DKP ($p \leq .05$), while no significant change was observed in potassium and magnesium content ($p \leq .05$).
The reason for the increase in calcium, magnesium, iron, zinc, copper, and selenium content in enriched soup samples is the presence of DKP; as it is observed in the compounds of DKP (Table 2), DKP is rich in these minerals, and the availability of these minerals is in this order: copper > selenium > iron > potassium > magnesium > calcium > zinc > sodium. Since the iron level is high in date kernel, the importance of this sub‐product in the enrichment of food products feels great. The reason for the low sodium content of enriched soup samples is the reduced level of this mineral in DKP. It should be noted that selenium owns antioxidant capacity and is regarded as an anticancer ingredient; the amount of which depends on the cultivation environment of date (Hamada et al., 2002).
Martinez‐Ballesta et al. [ 2010] reported that among minerals of date kernel, potassium, and iron are available in higher amounts. Ghasemi et al. [ 2020] observed an increase in sodium, zinc, calcium, iron, and potassium content of the sponge cake sample with the increase in the percentage of DKP.
## Calorie
Table 4 displays calorie changes in soup samples. The increase in the percentage of DKP descended the calorie amount of soup samples ($p \leq .05$). Fiber‐rich cereals such as wheat and barley cause satiation. In this research, with the replacement of DKP, the amount of barley and wheat flakes was reduced compared with the control sample due to the uniformity of the raw material weight in the formulation of the samples. Despite this, the reduction in calorie was observed; thus, the reason for the slight reduction in calorie of enriched soup samples could be attributed to DKP. Al Juhaimi et al., 2012 reported the energy of kernel powder of some date varieties to be 4340–4795 kcal/kg which is similar to that of DKP used in this study (4237 kcal/kg).
## Antioxidant activity and total polyphenol content
The changes in AA and TPC of soup samples are inserted in Table 5. AA and TPC of functional soup samples increased through increasing the percentage of DKP ($p \leq .05$). By comparing TPC of enriched soup samples (S2 and S3) with the control soup sample, it is observed that their TPC was 1.06 and 1.11 times higher, respectively, and S3 had the highest TPC. AA (IC50) of S2 and S3 was measured to be 8.04 and 6.01 mg/ml, respectively.
**TABLE 5**
| Characteristics treatment | AA (mg/ml) (IC50) | TPC (mg GAE/g) | ACE inhibitory activity (mg/ml) | α‐Glucosidase inhibition (%) | α‐Amylase inhibition (%) |
| --- | --- | --- | --- | --- | --- |
| S1 (DKP = 0) | 13.08a ± 0.00 | 0.456c ± 0.00 | 8.43a ± 0.02 | 1.06b ± 0.00 | 2.49c ± 0.21 |
| S2 (DKP = 2) | 8.04b ± 0.34 | 0.482b ± 0.00 | 8.20b ± 0.02 | 4.59a ± 0.00 | 4.48b ± 0.27 |
| S3 (DKP = 4) | 6.01c ± 0.09 | 0.507a ± 0.00 | 7.86c ± 0.02 | 6.36a ± 1.77 | 5.70a ± 0.13 |
One of the reasons for the application of DKP as a functional food ingredient is its great phenolic compounds (Al‐Farsi & Lee, 2008); the AA of DKP has been reported to be quite high due to its high phenolic content (Ghnimi et al., 2017; Shams Ardekani et al., 2010). In fact, TPC is the most effective factor in AA. Antioxidant compounds with two or more electron donor groups have higher AA (Simić et al., 2007). There is a strong correlation between AA and TPC as well as flavonoids (Bouhlali et al., 2015). Flavonoids of date kernel include luteolin, quercetin, kaempferol, apigenin, and isorhamnetin (Hinkaew et al., 2021). The presence of carotenoids with AA in date kernel has been reported (Alharbi et al., 2021). It should be mentioned that TPC and AA capacity of date kernel are higher than date paste, and is comparable to that of tea and grape seed (Platat et al., 2014). Some compounds such as gallic acid, vanillic acid, ferulic acid, caffeic acid, p‐coumaric acid, syringic acid (Amany et al., 2011; Hinkaew et al., 2021), protocatechuic acid and p‐hydroxybenzoic acid (Amany et al., 2011) have been found in the kernel of date varieties. Radfar et al. [ 2019] detected and measured seven phenolic compounds including chlorogenic acid, caffeic acid, vanillic acid, gallic acid, cinnamic acid, 3,5‐DHB, and 2,5‐DHB in four varieties of date kernel while they reported cinnamic acid as the most frequent phenolic compound in date kernel extract. Besides, polysaccharides (Fan et al., 2010; Jiao et al., 2011; Luo et al., 2010) and plant fibers (Hamada et al., 2002) have AAs as well.
Ansari and Kumar [2012] reported that soups could be functional enriched foods through replacement of its common ingredients with healthful ingredients; for example, the presence of onion (rich of quercetin) in the formulation of soup samples increases their AA. Substantial amount of phenolic compounds, mainly flavan‐3‐ols as well as flavonoids, in the bread having DKP has been reported (Platat et al., 2015). The intensification of AA caused by the increase in TPC of food products enriched with DKP has been reported before (Amany et al., 2011; Ambigaipalan & Shahidi, 2015; Asghari‐pour et al., 2018; Ghasemi et al., 2020; Platat et al., 2015).
## ACE inhibitory activity
Hypertension is an important factor in cardiovascular diseases and due to the important roles of ACE (EC number: 3.4.15.1; Peptidyl‐dipeptidase A) in regulating blood pressure, inhibition of this enzyme is used to treat high blood pressure. ACE inhibitors, similar to synthetic drugs, are widely used to control blood pressure. Blood pressure medications, especially ACE inhibitors, are employed to regulate blood pressure in renin–angiotensin system. Owing to some of the side effects of drug inhibitors in case of long‐term use, natural sources such as plant extracts and bioactive peptides have been considered as alternatives to chemical drugs (Daskaya‐Dikmen et al., 2017).
Table 5 represents the changes in ACE inhibition of soup samples. With the increase in the percentage of DKP, ACE inhibitory activity of soup samples promoted ($p \leq .05$). ACE inhibitory activities (IC50) of S2 and S3 were measured to be 8.2 and 7.86 mg/ml. The addition of DKP to soup samples eventuates minimal ACE inhibition but this activity increases as the percentage of DKP addition intensifies, which could be attributed to the higher polyphenolic compounds and dietary fiber. In fact, phenols act as ACE inhibitors (Iwaniak et al., 2014); Ferulic acid is a weak ACE inhibitor (Al Shukor et al., 2013).
Karakaya, El, Simsek, and Buyukkestelli [2015] confirmed the ACE inhibitory activity in plant product containing seeds and sprouts of lentil, pea, and caseinomacropeptide extracted from whey, and ACE inhibitory activity ($37\%$) of salad dressing containing this plant product was attributed to peptides resulting from hydrolysis of storage proteins by endopeptidases as a result of germination, bioactive peptide glycomacropeptide as well as polyphenol content of dietary plants (Karakaya, El, & Simsek, 2015). Ambigaipalan and Shahidi [2015] reported that the addition of $5\%$ DKP to muffin did not show any ACE inhibition but the level of $2.5\%$ DKP hydrolysate displayed ACE inhibitory activity (app. $13\%$, IC50 = 66 mg/ml); ACE inhibitory activity of the control muffin sample was significantly (app. $36\%$, IC50 = 28 mg/ml) higher than samples containing DKP. Patten et al. [ 2012] found that ACE inhibitory activity was directly related to TPC and that the processed form of dietary plants showed a wider capacity to regulate the renin–angiotensin system in vitro. Godos et al. [ 2017] confirmed ACE inhibitory activity by phenolic acids. Hung et al. [ 2020] reported the highest ACE inhibitory activity in the extracts of different types of bitter watermelon to be 15.8 mg/ml and concluded that with increasing TPC and peptides and triterpenoids in bitter watermelon extract, ACE inhibitory activity intensified. Ambigaipalan et al. [ 2015] confirmed that protein hydrolysates of date kernel can be used to inhibit ACE activity; IC50 values of these hydrolysates were much higher than captopril (a synthetic drug that inhibits ACE).
## α‐Amylase and α‐glucosidase inhibitory activities
Slowdown the intestinal absorption of glucose using the inhibition of α‐amylase (EC 3.2.1.1; 1,4‐α‐D‐glucan glucanohydrolase) and α‐glucosidase (EC 3.2.1.20; α‐D‐glucoside glucohydrolase) that decreases the post‐prandial hyperglycemia might represent a noteworthy method for treating type 2 diabetes (Ali Asgar, 2013). The results of inhibition of soup samples against α‐amylase and α‐glucosidase (Table 5) demonstrated a DKP concentration‐depended reduction in α‐ amylase and α‐glucosidase activities ($p \leq .05$). Samples S2 and S3 showed a significant increase in the inhibition of α‐amylase activity ($4.48\%$ and $5.70\%$, respectively) as compared to control ($2.49\%$). Also, the presence of $2\%$ and $4\%$ DKP improved the inhibition of α‐glucosidase activity in the soups to $4.59\%$ and $6.36\%$, respectively, in comparison to control ($1.06\%$). So, the highest inhibition appeared in S3, while control showed the weakest effect. The soup samples enriched with DKP exhibited a higher α‐glucosidase inhibitory activity than α‐amylase inhibitory activity. This observation suggests that bioactive compounds inhibiting α‐amylase and α‐glucosidase activity are present in DKP which could be attributed to the polyphenolic compounds and flavonoids. Phenolic compounds have been widely studied as α‐amylase and α‐glucosidase inhibitors (Abdelli et al., 2020; Kang et al., 2014; Kwon et al., 2008; Laoufi et al., 2017; Lou et al., 2018; Meng et al., 2016; Oboh et al., 2016); these compounds interact with digestive enzymes, α‐amylase, and α‐glucosidase (Abdelli et al., 2020); the most cited polyphenolic compounds are quercetin, kaempferol, rosmarinic acid, cyanidin, rutin, catechin, luteolin, and ellagic acid (Egbuna et al., 2021). However, Fratianni et al. [ 2021] did not find any correlation between the content of total polyphenols and the α‐glycosidase inhibition in honey. Also, the inhibitory effect of flavonoids on key enzymes linked to type 2 diabetes, α‐amylase, and α‐glucosidase has been demonstrated (Kim et al., 2000; Laoufi et al., 2017; Ng et al., 2015; Tadera et al., 2006; Williams, 2013). *In* general, the weak inhibitory activity of α‐amylase and α‐glucosidase in the soups could be attributed to their low polyphenolic content. α‐amylase and α‐glucosidase inhibitory activities were measured $34.57\%$ and $19.9\%$ for DKP, respectively (Table 1).
## CONCLUSIONS
Dry soup is one of the most popular types of soup due to its longer shelf life and ease of transportation. Currently, the enrichment of foods and beverages with functional natural substances has received much attention from nutritionists, and the use of nutraceuticals in food production brings about human health by reducing the risk of diseases. This study demonstrated that increasing the percentage of DKP in soup samples eventuated the reduction in pH value and promotion of acidity; it resulted in a significant intensification of viscosity while color changes were observed in samples and their fat, fibers, and minerals (iron, zinc, calcium, magnesium, copper, and selenium) raised. With the increase in the percentage of DKP, AA, TPC, and ACE inhibitory activity of soup samples improved. The soup samples exhibited a weak inhibitory activity of digestive enzymes, α‐amylase, and α‐glucosidase.
Finally, the results of this study hint that DKP can be used as a cheap natural source for the production of functional foods such as SPDS based on cereals due to their valuable and functional compounds. For this purpose, the utmost amount of DKP for functional soup production is suggested to be $4\%$ as the color change in products with higher than this threshold amount of DKP is unpleasant; besides, a high level of incorporation causes astringent taste and flavor. It is proposed that dark food products such as meat products, confectionery, chocolates, bakery products, cream‐filled of cakes and biscuits especially biscuits with cocoa and coffee flavors are the best choices for fortification with DKP.
## FUNDING INFORMATION
This research received no specific grant from any funding agency in the public, commercial, or not‐for‐profit sectors.
## CONFLICT OF INTEREST
The authors declare that there is no conflict of interest.
## RESEARCH INVOLVING HUMAN AND ANIMAL PARTICIPANTS
The manuscript does not contain experiments using animals or human studies.
## DATA AVAILABILITY STATEMENT
The authors declare that data supporting the findings of this study are available within the article.
## References
1. Abdel‐Haleem A. M. H., Omran A. A.. **Preparation of dried vegetarian soup supplemented with some legumes**. *Food and Nutrition Sciences* (2014) **5** 2274-2285
2. Abdelli I., Benariba N., Adjdir S., Fekhikher Z., Daoud I., Terki M., Benramdane H., Ghalem S.. **In silico evaluation of phenolic compounds as inhibitors of α‐amylase and α‐glucosidase**. *Journal of Biomolecular Structure and Dynamics* (2020) **39** 816-822. PMID: 31955660
3. Al Harthi S. S., Mavazhe A., Al Mahroqi H., Khan S. A.. **Quantification of phenolic compounds, evaluation of physicochemical properties and antioxidant activity of four date (**. *Journal of Taibah University Medical Sciences* (2015) **10** 346-352
4. Al Juhaimi F., Ghafoor K., Ozcan M. M.. **Physical and chemical properties, antioxidant activity, total phenol and mineral profile of seeds of seven different date fruit (**. *International Journal of Food Sciences and Nutrition* (2012) **63** 84-89. PMID: 21762027
5. Al Shukor N., Van Camp J., Gonzales G. B., Staljanssens D., Struijs K., Zotti M. J., Raes K., Smagghe G.. **Angiotensin‐converting enzyme inhibitory effects by plant phenolic compounds: A study of structure activity relationships**. *Journal of Agricultural and Food Chemistry* (2013) **61** 11832-11839. PMID: 24219111
6. Al‐Farsi M. A., Lee C. Y.. **Optimization of phenolies and dietary fibreextraction from date seeds**. *Food Chemistry* (2008) **108** 977-985. PMID: 26065761
7. Al‐Farsi M. A., Lee C. Y.. *Usage of date (Phoenix dactylifera L.) seeds in human health and animal feed, in nuts and seeds in health and disease prevention* (2011) 447-452
8. Alharbi K. L., Raman J., Shin H. J.. **Date fruit and seed in nutricosmetics**. *Cosmetics* (2021) **8** 59
9. Ali Asgar M. D.. **Anti‐diabetic potential of phenolic compounds: A review**. *International Journal of Food Properties* (2013) **16** 91-103
10. Almana H. A., Mahmoud R. M.. **Palm date seeds as an alternative source of dietary fiber in Saudi bread**. *Ecology of Food and Nutrition* (1994) **32** 261-270
11. Al‐Qarawi A. A., Abdel‐Rahman H., Ali B. H., Mousa H. M., El‐Mougy S. A.. **The ameliorative effect of dates (**. *Journal of Ethnopharmacology* (2005) **98** 313-317. PMID: 15814265
12. Amany M. M. B., Shaker M. A., Abeer A.. **Antioxidant activities of date pits in a model meat system**. *International Research Journal of Biochemistry and Bioinformatics* (2011) **1** 89-94
13. Ambigaipalan P., Abdulrahman S., Al‐Khalifa A., Shahidi F.. **Antioxidant and angiotensin I converting enzyme (ACE) inhibitory activities of date seed protein hydrolysates prepared using Alcalase Flavourzyme and Thermolysin**. *Journal of Functional Foods* (2015) **18** 1125-1137
14. Ambigaipalan P., Shahidi F.. **Date seed flour and hydrolysates affect physicochemical properties of muffin**. *Food Bioscience* (2015) **12** 54-60
15. Ansari M. M., Kumar D. S.. **Fortification of food and beverages with phytonutrients**. *Food and Public Health* (2012) **2** 241-253
16. Asghari‐pour S., Noshad M., Nasehi B., Jooyandeh H., Beiraghi‐Toosi S.. **Optimization of physicochemical and functional properties of corn‐based snacks containing date kernel flour**. *Journal of Food Processing and Preservation* (2018) **42** 11
17. Baliga M. S., Baliga B. R. V., Kandathil S. M., Bhat H. P., Vayalil P. K.. **A review of the chemistry and pharmacology of the date fruits (**. *Food Research Internatinal* (2011) **44** 1812-1822
18. Balthazar C. F., Santillo A., Guimarães J. T., Capozzi V., Russo P., Caroprese M., Marino R., Esmerino E. A., Raices R. S. L., Silva M. C., Silva H. L. A., Freitas M. Q., Granato D., Cruz A. G., Albenzio M.. **Novel milk–juice beverage with fermented sheep milk and strawberry (Fragaria × ananassa): Nutritional and functional characterization**. *Journal of Dairy Science* (2019) **102** 10724-10736. PMID: 31521367
19. Besbes S., Blecker C., Deroanne C., Lognay G., Drira N. E., Attia H.. **Heating effects on some quality characteristics of date seed oil**. *Food Chemistry* (2005) **91** 469-476
20. Bouaziz F., Ben Abdeddayem A., Koubaa M., Ellouz Ghorbel R., Ellouz Chaabouni S.. **Date seeds as a natural source of dietary fibers to improve texture and sensory properties of wheat bread**. *Food* (2020) **9** 737
21. Bouaziz M. A.. **Use of endemic date palm (**. *Journal of Food Quality* (2020) **13** 12
22. Bouaziz M. A., Amara W. B., Attia H., Blecker C., Besbes S.. **Effect of the addition of defatted date seeds on wheat dough performance and bread quality**. *Journal of Texture Studies* (2010) **41** 511-531
23. Bouhlali E. D. T., Alem C., Ennassir J., Benlyas M., Mbark A. N., Zegzouti Y. F.. **Phytochemical compositions and antioxidant capacity of three date (**. *Journal of the Saudi Society of Agricultural Sciences* (2015) **16** 350-357
24. Daskaya‐Dikmen C., Yucetepe A., Karbancioglu‐Guler F., Daskaya H., Ozcelik B.. **Angiotensin‐i‐converting enzyme (ACE)‐inhibitory peptides from plants**. *Nutrients* (2017) **9** 316. PMID: 28333109
25. Egbuna C., Awuchi C. G., Kushwaha G., Rudrapal M., Patrick‐Iwuanyanwu K. C., Singh O., Odoh U. E., Khan J., Jeevanandam J., Kumarasamy S., Chukwube V. O., Narayanan M., Palai S., Găman M. A., Uche C. Z., Ogaji D. S., Ezeofor N. J.. **Bioactive compounds effective against type 2 diabetes mellitus: A systematic review**. *Current Topics in Medicinal Chemistry* (2021) **21** 1067-1095. PMID: 33966619
26. El Sheikh D. M., El‐Kholany E. A., Kamel S. M.. **Nutritional value, cytotoxicity, anti‐carcinogenic and beverage evaluation of roasted date pits**. *World Journal of Dairy & Food Sciences* (2014) **9** 308-316
27. Elleuch M., Bedigian D., Roiseux O., Besbes S., Blecker C., Attia H.. **Dietary fibre and fibre rich by‐products of food processing: Characterisation, technological functionality and commercial applica tions: A review**. *Food Chemistry* (2011) **124** 411-421
28. Fan H., Mazza G., Liao X.. **Purification, composition and antioxidant activity of polysaccharides from wolfberry, cherry, kiwi and cranberry fruits**. *Journal of Food Science and Technology* (2010) **2** 9-17
29. Figuerola F., Hurtado M. L., Estévez A. M., Chiffelle I., Asenjo F.. **Fibre concentrates from apple pomaceand citrus peel as potential fibre sources for food enrichment**. *Food Chemistry* (2005) **91** 395-401
30. Fikry M., Aniza Y. Y., Alhussein M., Awaadh A., Rahman R., Ling Chin N., Mousa E., Sin Chang L.. **Effect of the roasting conditions on the physicochemical, quality and sensory attributes of coffee‐like powder and brew from defatted palm date seeds**. *Food* (2019) **8** 61
31. Fratianni F., Ombra M. N., d'Acierno A., Caputo L., Amato G., De Feo V., Coppola R., Nazzaro F.. **Polyphenols content and in vitro α‐glycosidase activity of different Italian monofloral honeys, and their effect on selected pathogenic and probiotic bacteria**. *Microorganisms* (2021) **9** 1694. PMID: 34442773
32. Ghasemi E., Loghmanifar S., Salar S.. **The effect of adding date kernel powder on the qualitative and sensory properties of spongy cake**. *Journal of Novel Applied Sciences* (2020) **9** 47-55
33. Ghnimi S., Umer S., Karim A., Kamal‐Eldin A.. **Date fruit (**. *NFS Journal* (2017) **6** 1-10
34. Godos J., Sinatra D., Blanco I., Mulè S., La Verde M., Marranzano M.. **Association between dietary phenolic acids and hypertension in a Mediterranean cohort**. *Nutrients* (2017) **9**. PMID: 28953227
35. Habib H. M., Ibrahim W. H.. **Nutritional quality evaluation of eighteen date pit varieties**. *International Journal of Food Sciences & Nutrition* (2009) **60** 99-111
36. Halaby M. S., Farag M. H., Gerges A. H.. **Potential effect of date pits fortified bread on diabetic rats**. *International Journal of Nutrition and Food Sciences* (2014) **3** 49-59
37. Hamada J. S., Hashimb I. B., Sharif B. F. A.. **Preliminary analysis and potential uses of date pits in foods**. *Food Chemistry* (2002) **76** 135-137
38. Hinkaew J., Aursalung A., Sahasakul Y., Tangsuphoom N., Suttisansanee U.. **A comparison of the nutritional and biochemical quality of date palm fruits obtained using different planting techniques**. *Molecules* (2021) **26** 2245. PMID: 33924574
39. Hung C. Y., Yang S. H., Jaw‐Neng Lin J. N.. **Determination of contents of total phenol, flavonoid, triterpenoid and peptide and in vitro angiotensin‐converting enzyme inhibitory activity of bitter melon extract**. *Journal of Taiwan Agricultural Research* (2020) **69** 185-192
40. Iwaniak A., Minkiewicz P., Darewicz M.. **Food‐originating ACE inhibitors, including antihypertensive peptides, as preventive food components in blood pressure reduction**. *Comprehensive Reviews in Food Science and Food Safety* (2014) **13** 114-134. PMID: 33412648
41. Jiao G., Yu G., Zhang J., Stephen Ewart H.. **Chemical structures and bioactivities of sulfated polysac charides from marine algae**. *Drugs* (2011) **9** 196-223
42. Kang B. H., Racicot K., Pilkenton S. I., Aporrolidis E.. **Evaluation of the in vitro antihyperglycemic effect of**. *Plant Foods for Human Nutrition* (2014) **69** 155-160. PMID: 24706251
43. Karakaya S., El S. N., Simsek S.. **Functional salad dressing as an excipient food**. *Turkish Journal of Agriculture – Food Science and Technology* (2015) **3** 849-855
44. Karakaya S., El S. N., Simsek S., Buyukkestelli H. I.. **Vegetable product containing caseinomacropeptide and germinated seed and sprouts**. *Journal of Food Science and Technology* (2015) **53** 880-887. PMID: 26788011
45. Kim J. S., Kwon C. S., Sou K. H.. **Inhibition of alpha glucosidase and amylase luteolin, a flavonoid**. *Bioscience, Biotechnology, and Biochemistry* (2000) **64** 2458-2461. PMID: 11193416
46. Kwon Y. I., Apostolidis E., Shetty K.. **Inhibitory potential of wine and tea against α‐amylase and α‐glucosidase for management of hyperglycemia linked to type 2 diabetes**. *Journal of Food Biochemistry* (2008) **32** 15-31
47. Laoufi H., Benariba N., Adjdir S., Djaziri R.. **In vitro α‐amylase and α‐glucosidase inhibitory activity of**. *Journal of Applied Pharmaceutical Science* (2017) **7** 191-198
48. Lou W., Chen Y., Ma H., Liang G., Liu B.. **Antioxidant and α‐amylase inhibitory activities of tannic acid**. *Journal of Food Science and Technology* (2018) **55** 3640-3646. PMID: 30150823
49. Luo A., He X., Zhou S., Fan Y., Luo A., Chun Z.. **Purification, composition analysis and antioxidant activity of the polysaccharides from**. *Carbohydrate Polymers* (2010) **7** 1014-1019
50. Martinez‐Ballesta M. C., Dominguez‐Perles R., Moreno D. A., Muries B., Alcaraz‐Lopez C., Bastías E., Carvajal M.. **Minerals in plant food: Effect of agricultural practices and role in human health. A review**. *Agronomy for Sustainable Development.* (2010) **30** 295-309
51. McCleary B. V., McLoughlin C., Charmier L. M. J., McGeough P.. **Measurement of available carbohydrates, digestible, and resistant starch in food ingredients and products**. *Cereal Chemistry* (2019) **97** 114-137
52. Meng Y., Su A., Yuan S., Zhao H., Tan S., Hu C., Guo Y.. **Evaluation of total flavonoids, myricetin, and quercetin from Hovenia dulcis Thunb. as inhibitors of α‐ amylase and α‐glucosidase**. *Plant Foods for Human Nutrition* (2016) **71** 444-449. PMID: 27787697
53. Najjar Z., Alkaabi M., Alketbi K., Stathopoulos C., Ranasinghe M.. **Physical chemical and textural characteristics and sensory evaluation of cookies formulated with date seed powder**. *Food* (2022) **11**
54. Ng K., Gu C., Zhang H., Patrim C. Y.. **Evaluation of alpha amylase and alpha glucosidase inhibitory activity of flavonoids**. *International Journal of Nutrition Sciences* (2015) **2** 1-6
55. Niththiya N., Vasantharuba S., Subajini M., Srivijeindran S.. (2014)
56. Oboh G., Ogunsuyi O. B., Ogunbadejo M. D., Adefegha S. A.. **Influence of gallic acid on α‐amylase and α‐glucosidase inhibitory properties of acarbose**. *Journal of Food and Drug Analysis* (2016) **24** 627-634. PMID: 28911570
57. Olubi O., Felix‐Minnaar J. V., Jideani V. A.. **Physicochemical, mineral and sensory characteristics of instant**. *Food* (2021) **10** 1817
58. Patten G. S., Abeywardena M. Y., Head R. J., Bennett L. E.. **Processed dietary plants demonstrate broad capacity for angiotensin converting enzyme and angiotensin II receptor binding inhibition in vitro**. *Journal of Functional Foods* (2012) **4** 851-863
59. Platat C., Habib H. M., Hashim I. B., Kamal H., AlMaqbali F., Souka U., Ibrahim W. H.. **Production of functional pita bread using date seed powder**. *Journal of Food Science and Technology* (2015) **52** 6375-6384. PMID: 26396382
60. Platat C., Habib H. M., Maqbali F. D. A., Jaber N. N., Ibrahim W. H.. **Identification of date seeds varieties patterns to optimize nutritional benefits of date seeds**. *Journal of Nutrition & Food Sciences* (2014) **S8**
61. Radfar R., Farhoodi M., Ghasemi I., Mousavi Khaneghah A., Farzaneh Shahraz F., Hosseini H.. **Assessment of phenolic contents and antioxidant and antibacterial activities of extracts from four varieties of Iranian date palm (**. *Applied Food Biotechnology* (2019) **6** 173-184
62. Shams Ardekani M. R., Khanavi M., Hajimahmoodi M., Jahangiri M., Hadjiakhoondi A.. **Comparison of antioxidant activity and total phenol contents of some date seed varieties from Iran**. *Iranian Journal of Pharmaceutical Research* (2010) **9** 141-146. PMID: 24363720
63. Shokrollahi F., Taghizadeh M.. **Date seed as a new source of dietary fiber: Physicochemical and baking properties**. *International Food Research Journal* (2016) **23** 2419-2425
64. Simić A., Manojlović D., Šegan D., Todorović M.. **Electrochemical behavior and antioxidant and Prooxidant activity of natural Phenolics**. *Molecules* (2007) **12** 2327-2340. PMID: 17978760
65. Tadera K., Minami Y., Takanatsu K., Matsuoka T.. **Inhibition of α‐glucosidase and α‐amylase by flavonoids**. *Journal of Nutritional Science and Vitaminology* (2006) **52** 99-103
66. Tafti G., Dahdivan S., Ardakani Y.. **Physicochemical properties and applications of date seed and its oil**. *International Food Research Journal* (2017) **24** 1399-1406
67. Upadhyay S., Tiwari R., Kumar S., Kohli D.. **Production and evaluation of instant herbal mix soup**. *International Journal of Agricultural* (2017) **7** 37-42
68. Vinita D.. **Sensory characteristics and nutrient composition of biscuits prepared by using seed powder of date fruit (**. *International Journal of Food Science and Nutrition* (2016) **1** 1-3
69. Wahini M.. **Exploration of making date seed's flour and its nutritional contents analysis**. (2016)
70. Wedad M., El‐Khol Y.. **Production of probiotic yoghurt fortified with date seeds (**. *Egyptian Journal of Agricultural Research* (2018) **96** 159-173
71. Williams G.. **Possible effects of datary polyphenols on sugar absorption and digestion**. *Molecular Nutrition & Food Research* (2013) **57** 48-57. PMID: 23180627
|
---
title: Synthesis of Bio-Based Polyester from Microbial Lipidic Residue Intended for
Biomedical Application
authors:
- Ana P. Capêto
- João Azevedo-Silva
- Sérgio Sousa
- Manuela Pintado
- Ana S. Guimarães
- Ana L. S. Oliveira
journal: International Journal of Molecular Sciences
year: 2023
pmcid: PMC10003017
doi: 10.3390/ijms24054419
license: CC BY 4.0
---
# Synthesis of Bio-Based Polyester from Microbial Lipidic Residue Intended for Biomedical Application
## Abstract
In the last decade, selectively tuned bio-based polyesters have been increasingly used for their clinical potential in several biomedical applications, such as tissue engineering, wound healing, and drug delivery. With a biomedical application in mind, a flexible polyester was produced by melt polycondensation using the microbial oil residue collected after the distillation of β-farnesene (FDR) produced industrially by genetically modified yeast, Saccharomyces cerevisiae. After characterization, the polyester exhibited elongation up to $150\%$ and presented Tg of −51.2 °C and Tm of 169.8 °C. In vitro degradation revealed a mass loss of about $87\%$ after storage in PBS solution for 11 weeks under accelerated conditions (40 °C, RH = $75\%$). The water contact angle revealed a hydrophilic character, and biocompatibility with skin cells was demonstrated. 3D and 2D scaffolds were produced by salt-leaching, and a controlled release study at 30 °C was performed with Rhodamine B base (RBB, 3D) and curcumin (CRC, 2D), showing a diffusion-controlled mechanism with about $29.3\%$ of RBB released after 48 h and $50.4\%$ of CRC after 7 h. This polymer offers a sustainable and eco-friendly alternative for the potential use of the controlled release of active principles for wound dressing applications.
## 1. Introduction
The transition to a higher degree of sustainability lies in the industrial development of bio-based and biodegradable plastics to counteract the absence of new extensive sources of conventional non-renewable crude oil, avoiding the related detrimental effect on climate change and environmental pollution [1,2].
In this context, resource circulation, a concept that combines circular economy and sustainable development goals, has been used to replace the conventional petroleum refinery with an economically viable biorefinery targeting the production of several bio-based materials, including plastics [3]. New platform chemicals [4,5] were created, based on carbon-rich precursors (nucleic acids, proteins, carbohydrates, lipids, etc.) derived from the second and third-generation expanded range of feedstocks, such as lignocellulosic residues (e.g., from agriculture, forest, and industrial by-products), and extracts obtained from wastes of industrial, and municipal origin [6,7].
Natural (e.g., alginate, cellulose, starch, etc.) and synthetic biodegradable polymers (e.g., polyesters, polyamides, etc.) are produced for high-value applications in the food, pharmaceutical, and biomedical fields through microbial agents, biopolymer blending, or/and chemical methods [8,9,10].
The most well-known and characterized bio-based polymers and copolymers [11,12,13,14] are synthesized from monomers, directly or indirectly obtained by fermentation of biomass, such as α-hydroxy acids (e.g., citric, lactic, acetic, malic, glycolic, and tartaric acids) [15]; alcohols and glycols (e.g., ethanol, butanediol, propanediol, glycerol, etc.); dicarboxylic acids (e.g., succinic, sebacic, itaconic, azelaic acids, etc.); and diamines [16,17,18].
Polylactide (PLA) and poly lactide-co-glycolide (PLGA) are good examples of biodegradable poly(α-hydroxy acids) aliphatic polyesters with extensive biomedical applications in disposable and support products (e.g., syringes, blood bags, sutures, bone plates, and sealants), hard and soft tissue engineering, surgical implants, reconstructive surgery, wound dressings, controlled drug delivery systems, among others [19,20]. These biopolymers allow precise control of their properties, replicability of results, and easy adaptation to industrial production [21,22]. Beyond the economic and technical restrictions to this kind of application, polymers need to meet other hard limits such as biocompatibility, mechanical/thermal performance, wettability, or stability [23,24].
Skin diseases are a serious burden in healthcare and chronic wound treatment, representing a global care cost varying from $13 to $15 billion annually [25]. Based on this, the use of novel technologies for the treatment of skin diseases and injuries is highly encouraged. The development of site-specific drug delivery represents an advantage because it is painless and noninvasive, while improving drug bioavailability, and it minimizes the systemic toxicity and drug exposure to non-desired sites [26].
Recent advances in green technological processes allowed for the development of new bio-based materials or composites to produce semi-permeable films, foams, hydroactive dressings, and hydrogels [27,28,29], designed to have a more active role in advanced cutaneous wound healing [30,31]. These smart materials provide a physical shield that avoids external bacterial contamination of the wound, allowing gas exchange, while maintaining proper moisture by absorbing wound exudates [32,33]. Additionally, these materials are suitable for drug delivery, controlling the release kinetics, and increasing the performance and effectiveness of the treatment [34,35,36,37,38].
Several natural synthetic polymers have been used for tissue engineering and wound healing applications since they combine properties such as biocompatibility and or biodegradability, and despite development strategies, performance and price are still hot topics, since to achieve the required properties, biopolymers have high production costs [39]. Within this context, the utilization of waste material or renewable resources may overcome this problem [40].
The aim of this work was to develop a flexible and biocompatible aliphatic copolyester from a biowaste, a microbial oil, to produce a porous scaffold for wound dressing, suitable for the controlled release of bioactive compounds.
## 2.1. Synthesis and Characterization of FDR-Based Epoxide
The microbial oil residue collected after distillation of β-farnesene (FDR) industrial production by a fermentation process was characterized elsewhere [41] by the presence of hydrocarbons ($43.29\%$ wt. lipids), terpenes (e.g., farnesene and farnesol), a small percentage of fatty acids ($0.33\%$), complex lipids such as triglycerides ($4.29\%$), diglycerides ($2.62\%$), among other compounds.
FDR has a biphasic nature with a certain amount of waxes esters ($0.22\%$) that increase the viscosity of the derived polymeric materials due to the formation of crystals at room temperature [42]. The material’s viscosity is an important parameter that affects productivity and product quality, along with polymer processability, polymerization kinetics (mass and heat transfer coefficients, residence time, etc.), pumping and stirrer power input, cooling capacity, etc. [ 43].
With that issue in mind, the crude biowaste was in a first stage treated by winterization, a process usually applied in the industrial refinement of vegetable oils to remove waxes through crystallization [44]. The procedure promoted a considerable removal of the suspended waxes esters in the raw FDR (Figure 1A,B), along with a reduction in the viscosity (about $83\%$) and color of the epoxide (Figure 1C,D) under the same operating conditions.
FDRw (FDRwinterized) was then submitted to a typical epoxidation to increase thermal and oxidative stability, therefore improving the lubricant [45] and plasticizing properties [46,47]. To increase the hydroxyl content of the epoxidized oil, castor oil was incorporated (ratio 3:1), a non-edible natural polyol commonly used in the synthesis of bio-based plasticizers [48], thermosetting elastomers and cross-linked polymers [49,50,51].
The physicochemical properties of FDRw, derived epoxide, and of the formulated polyol, are shown in Table 1. All these products were characterized by acid and hydroxyl values as well as viscosity and density. The molecular weight and functionality were also determined for the polyol.
FDRw presented an average value of viscosity of 123 mPa·s, density of 0.962 g·cm−3, and hydroxyl (–OH) value around 45 mg KOH·g−1. The iodine value is parameter correlated to the percentage of unsaturation, i.e., the presence of reactive sites, the C=C bonds [52]. The obtained value of 130 g I$\frac{2}{100}$ g is typical of semi-drying vegetable oils such as soybean, corn, and sunflower oils [53].
The iodine value of the treated residue (FDRw) showed the expected sharp decrease after epoxidation (from 130 to 32 g I$\frac{2}{100}$ g), i.e., a conversion of unsaturated bonds (Ivr) of about $76\%$. This variation is related to the breakdown of the double bonds in unsaturated fatty acids when oxidation reactions occur [54].
The epoxide presented a yield value of 89.2 % wt., based on FDRw input and a relative conversion of the oxirane oxygen content (RCO) of about $61.4\%$.
The initial content in FDRw hydroxyl groups (−OH = 42.5 mg KOH·g−1), derived from the presence of fatty alcohols and terpenes (e.g., farnesol), has increased after epoxidation (63.5 mg KOH·g−1). When castor oil was added (ratio 3:1), a value of 104.2 mg KOH·g−1 was achieved. The content of hydroxyl groups with the ability to form intramolecular H-bonding or the higher content of oligomers may contribute to the increase in the viscosity value observed on the formulated co-polyol [55], hence, the melting at relatively high temperatures under inert atmosphere. This operation allowed the reduction in water content (from 2.8 to $0.4\%$), reduction in viscosity, and the generally improved thermal properties (e.g., pour point and cloud point), seeking improved oxidation and hydrolytic stability [56].
After this treatment, the polyol became clear with a lower viscosity ($15\%$) than the castor oil/epoxide mixture) with an enhanced flowability at room temperature. In fact, the values found for polyol’s viscosity and density (1350 mPa·s, 0.989 g·cm−3) were lower than the ones found for the epoxide (1600 mPa·s and 1.04 g·cm−3).
The polyol presented an average molecular weight of 2978.9 Da and the obtained value of 5.5 hydroxyl groups per molecule is higher than the value found for castor oil of 2.7 hydroxyl groups per molecule [57], i.e., about the same order of magnitude as castor oil-based polyols obtained by thiol-ene reaction with 2-mercaptoethanol [58] and soybean oil polyols obtained using ethanol and ethylene glycol as hydroxylation agents [59].
The initial acid value for FDRw (4.4 mg KOH·g−1), a reflex of the free fatty acids content (0.33 g/100 g lipids) [41,60], increased to 10.7 mg KOH·g−1 after epoxidation, probably due to the presence of residual carboxylic acid used in that step. After the addition of castor oil with a lower acid value (0.9 mg KOH·g−1), the value was reduced to 4.2 mg KOH·g−1.
## 2.2. Polyester Synthesis and Characterization
The polyester was synthesized by melt polycondensation between the FDRw-based polyol and azelaic acid, with citric acid (the crosslinker), in the presence of 1,4-butanediol and sorbitol as chain extenders and mechanical enhancers. The reaction took place under nitrogen purge and is catalyst and solvent-free to minimize any unwanted cytotoxicity, which is a prime parameter for biomedical applications [61].
The yield based on the initial input of polyol was about 78.6 % wt., a good result attesting to the economic viability of any industrial process.
Comparing both FDR and FDRw structural profiles (Figure 2), no relevant changes are apparent. Visible throughout the entire process are the typical bands assigned to alkyl C-H bands at 2920, 2855, and 2973 cm−1. As expected, the bands at 831 and 863 cm−1 assigned to the highly reactive oxirane-ring structures (epoxy groups) already present in FDRw are accentuated after epoxidation, evidence that some of the C=C double bonds remaining in the oily residue were broken and converted [62].
In the screening of the formulated polyol, there is an increment in the intensity of the bands at 3462 cm−1 and 1734 cm−1 assigned to –OH and C=O stretching, respectively, an expected result from the addition of the castor oil with higher hydroxyl content.
After esterification, the characteristic ester C=O stretching band at 1734 cm−1 suffers an exponential growth, and a new band emerged at 1171 cm−1 assigned to C-O stretching [63]. These are the signature bands of ester linkages typical of polyesters. The hydroxyl groups still attached to the carbon backbone of the cured polyester contribute to the hydrophilicity of the polymer [64].
The polyester presented a smooth and flexible surface, with an amber/brown color (Figure 3).
To verify the surface wettability, the contact angle of the polyester was measured using water and squalane (Figure 4). The contact angle value stabilized within the time frame of 60 s for both fluids. For water (Figure 3), the contact angle showed a variation between 70.2° and 54.9°, thus confirming the hydrophilic character (moderate wettability) of this biopolymer [65]. Regarding the squalane oil, the change in the angle between 31.5 to 9.1° suggests good compatibility. That is a relevant fact since squalane and its natural counterpart, squalene, have been reported to be beneficial for skin health, exhibiting antioxidant, detoxifying, and regeneration activities. Squalane also acts as a drug carrier in both in vivo models and in vitro environments [66,67].
The droplet’s behavior indicated a spreading mechanism specifically observed with squalane, while absorption played a predominant role in the interface water/polymer surface, i.e., the surface wetting [68].
The polyester revealed a gel content (Table 2) in DMSO of $78.5\%$ after 24 h, a sign of a relatively low cross-linking density that can be explained by the presence of unreacted oligomers [69].
The water absorption exhibited an average value of about $22.8\%$ after 24 h. This value corroborates the hydrophilic nature of the polymer and the presence of hydroxyl terminations visible in the structural analysis [70].
Before applying any material in tissue engineering it is important to investigate their mechanical properties and thermal behavior. The results of the polyester tensile properties (Table 2) showed a tensile strength of about 0.19 MPa paired with an elongation at break with values between 102.1 and $153.0\%$ (average $127.5\%$) and an average Young elastic modulus of 1.9 × 10−3 MPa. This behavior confirmed the polyester elastomeric properties [71].
Mechanical analysis of insulin-loaded PLGA nanofibers displayed elongation at a break of 164.3 ± $27.2\%$ and tensile strength of approximately 2.87 ± 0.07 MPa, similar to human native skin [72]. A recent review [73] of commercial wound dressing with mechanical properties determined using tensile testing concluded that monolayered electrospun fiber dressings made of polyvinyl alcohol, carboxymethylcellulose, and alginate possessed a Young’s modulus between 0.24 and 0.95 MPa, with the total elongation varying between 68 and $134\%$.
Strategies to increase tensile strength can lie in increasing cross-linking density through curing conditions [74], or, for example, with the addition of glycol derivatives such as isosorbide [75] in the polymer composition. Additionally, the incorporation of natural fibers such as cellulose [76] or chitin [77] is also possible.
Analyzing the polyester thermogram (Figure 5), the material showed glass transition temperature (Tg) taken at midpoint of −51.2 °C and a melting temperature (Tm) of 169.6 °C, confirming a semi-crystalline nature, and non-glassy behavior at room temperature. An amorphous phase with a low glass transition temperature (Tg) promotes the flexibility of the polymer network and physical net points that govern the form stability after elastic deformation [78].
## 2.3. Polyester in vitro Biocompatibility Study
Prior to the bio-application of any new material, it is necessary to check the cytocompatibility. The biocompatibility of this bio-based polyester was evaluated on human keratinocytes that were exposed to conditioned media with polymer discs (Figure 6). After an incubation period of 3 min and 24 h, the cell viability wasn’t visibly affected.
The copolyester didn’t show any signs of toxicity when in contact with keratinocyte cells; hence, it was considered to be safe for skin applications. Keratinocytes are the predominant cell type found in the epidermis and the first to be affected by toxic substances through direct contact. Similar results were found by a polyester derived from palm fatty acids [79] and by a polyurethane film prepared with palm kernel oil-based polyester [80].
## 2.4. In Vitro Degradation Study
For successful tissue engineering applications, the polymeric biomaterial should degrade within a particular time span to allow the release of drugs in a sustained manner [81]. Hereupon, an in vitro degradation study was conducted under accelerated conditions (40 °C, RH = $75\%$). Two weeks after the beginning of the experiment (Figure 7), the polyester sample did not suffer mass loss; actually, there was an increment of about $41.4\%$ with a visible expansion in the sample volume. Only after three weeks did the sample present a slight erosion around the edges, shown by a perceptible mass loss of $5.4\%$. In the following weeks, the mass loss was gradual; however, eight weeks later, the mass loss became exponential. After 11 weeks under accelerated conditions, i.e., one year in real-time, the bio-based polyester disintegrated completely in PBS solution. In the meantime, the pH value showed a slow reduction from the initial value of pH = 7.4 to pH = 6.7 until all semblance of integrity disappeared.
Since the diffusion of water into the polymer network preceded the hydrolytic degradation, the polyester degradation occurred in a two-step bulk erosion, a behavior similar to other aliphatic polyesters such as PLA and PCL [82,83,84]. In this type of erosion, after exceeding a critical value of water intake, the cleavage of the polymer chains, especially the hydrolytically unstable ester chemical bonds, results in water-soluble fragments; hence, the mass loss is fast, with a sudden release of degradation products [85]. Considering the results obtained so far, it is possible to conclude that lower cross-linking densities tend to promote a faster degradation [86].
There are several biopolymers extensively studied for applications in tissue engineering such as polycaprolactone (PCL), poly-l-lactic acid (PLA), and poly(glycolic acid) (PGA), whose biodegradation properties limit their clinical application [87,88]. The degradation of two years observed for PCL restricted its use for in vivo tissue engineering applications. High molecular weight crystalline PLA exhibited a degradation time of over 50 weeks [89,90].
The high degree of structural degradation observed with this particular bio-based polyester after one year, in closed containers and in vitro physiological conditions, suggests that this polyester will potentially undergo a higher hydrolytic degradation rate and biodegradation promoted by a microorganism-rich, moist, and warm environment such as a landfill [91,92].
## 2.5. Polyester Scaffold Properties
The design of a bio-based scaffold can be tuned to enhance mechanical properties and biodegradability through processing strategies directed to control pore size, pore quantity, and pore connectivity with the inherent change in density [93].
SEM micrographs (Figure 8) revealed the morphology of the scaffold porous surface along the structure of the pore walls. The surface of both scaffolds was found to be free from any irregularities such as cracking and delamination.
While 2D scaffolds exhibited a homogenous structure with well-formed pores (size 500 μm) only present on the surface, 3D scaffolds presented a longitudinal pore size gradient in the scaffold’s inner core. The difference in the distribution of the pores can be justified by the utilization of centrifugal forces in the production of the cylindrical scaffold [94].
## 2.6. In Vitro Dye Release Study
Several factors condition the drug delivery from a degradable polymeric matrix, such as the degree of the polymer hydrophilicity, the surface erosion and cleavage of polymer bonds within the matrix, and the diffusion mechanism of the entrapped drug [95].
The capacity of the prepared polyester to adsorb and release a drug and the rate of that delivery was studied using two dyes.
Curcumin (CRC) was selected for its recognized potential as a wound healing agent along with anti-infectious, anti-inflammatory, and antioxidant properties [96,97]. Quite recently, curcumin-loaded delivery systems for wound healing applications were the subject of an updated review [98].
Additionally, to evaluate the capacity of the manufactured polyester to entrap water-soluble molecules, Rhodamine B base (RBB) was selected as a model [51,99]. This dye was used with the same purpose in the case of polylactic acid microchamber arrays and liquid resins in 3D-printed medical devices [100,101].
In the end, two data sets were obtained using the manufactured scaffolds. Figure 9 illustrates the release curves for both dyes and the plots regarding the application of the Korsmeyer–Peppas model. In Table 3 are the compiled results.
Concerning RBB dye, the 3D scaffolds presented a loading capacity after 7 h of 1.60 mg dye·g−1, and within 48 h about $29.3\%$ of the dye was released. Nevertheless, when the contact water was replaced after 10 days, the dye was still being released to the medium. Regarding CRC, the 2D structures revealed a loading capacity of 0.64 mg dye·g−1 after 7 h, and in the same period about $50.4\%$ of the dye was released in the ethanolic contact solution. No further release occurred when replacing the contact solution, although the polymer still presented some coloration.
After application of the Korsmeyer–Peppas model Equation [11] to the release data set, correlation coefficients (R2) were 0.987 (RBB) and 0.996 (CRC), respectively, indicating a good adjustment between the model and the experimental data. From the data fitting, the transport constant for CRC was $k = 29.1$ h−1 and the transport exponent $$n = 0$.31$ and for RBB, $k = 11.9$ h−1 and $$n = 0$.24$ (Table 3). Since its n < 0.5, the diffusional model is a quasi-Fickian model, i.e., a non-swellable matrix-diffusion [102].
These results concur with the curcumin release from poly(lactic acid)/polycaprolactone electrospun fibers which also followed a diffusion-controlled mechanism, with a good fit to the Korsmeyer–Peppas model, and a diffusion coefficient, n ≤ 0.5 [89]. A poly(lactic acid)/polycaprolactone polyester, loaded with a complex of curcumin and β-cyclodextrin with an efficacy release of $20\%$ was developed with potential use for wound dressings, drug delivery, and regenerative systems [90]. The Korsmeyer–Peppas model has been widely used to describe bioactive compound release from polymers, namely grapeseed extract from nanofibrous membrane made with polylactic acid and polyethylene oxide [92], and rutin release from cellulose acetate/poly(ethylene oxide) [93].
The manufactured 3D scaffolds showed a lower release rate and a higher loading capacity to entrap hydrophilic molecules than the 2D structures for the lipophilic molecule. This information suggests that this polyester could potentially be used to deliver drugs with hydrophilic characteristics into the skin, as in the case of hyaluronic acid and ascorbic acid in PCL [103].
## 3.1. Reagents
β-Farnesene distillation oily residue (FDR) was kindly provided by Amyris and sugarcane-based Squalane (NeossanceTM) from Aprinnova LLC, Emeryville, CA, USA. Formic acid $99\%$ was supplied by VWR; sodium chloride, ethanol $96\%$ (EtOH), ethyl acetate, hydrogen peroxide ($35\%$), sodium hydrogen carbonate, and citric acid monohydrated (CA) from Labchem. Pressurized nitrogen of high purity was supplied by Gasin. Azelaic acid (AzA), sorbitol (S), and 1,4-Butanediol (BTN) were all supplied by Sigma. Acofarma supplied castor oil containing 85–$90\%$ of OH-bearing ricinoleic acid (C18:1). This non-edible oil was characterized by an acid value of 0.9 mg KOH·g−1, hydroxyl value (166 mg KOH·g−1), iodine value of 83 gI$\frac{2}{100}$ g, $0.11\%$ of water, and viscosity of 986 cp (20 °C).
## 3.2. Epoxide Synthesis and Polyol Formulation
FDR pre-treatment (FDRw) Crude FDR was treated by winterization, a procedure where waxes are removed through crystallization with a suitable solvent at low temperatures. FDR was thoroughly mixed with ethanol $96\%$ (mass ratio EtOH: FDR = 4:1) for 30 min, followed by overnight decantation at 4.0 ± 2.0 °C in a refrigerator (Thermo ScientificTM, Waltham, MA, USA) and additional clarification by centrifugation (Thermo ScientificTM, Heraeus Megafuge 16 R). The ethanol was recovered to be recycled in a rotary evaporator (Heidolph, Hei-Vap Precision, Schwabach, Germany). The clarified product (FDRw) was reserved for further analysis and used in the next stage.
Epoxidation (Epoxide) was carried out with in situ performic acid according to [104] with slight modifications. FDRw (200 g), ethyl acetate (100 g), and formic acid (12 g) were weighed in a closed borosilicate vessel with a three-way lid and placed on a heating/stirring plate (Hei-Tech, Heidolph Instruments GmbH & Co.KG, Schwabach, Germany) with temperature control. After a few minutes of constant stirring, refrigerated hydrogen peroxide (320 g) was slowly poured to avoid exothermic foaming. The epoxidation took place at 85 °C for 3 h with constant stirring, followed by decantation of the organic phase in a separatory funnel and a neutralization/washing operation with water and sodium bicarbonate solution ($10\%$). The epoxide was then dried overnight at 80 °C in a vacuum oven (Binder, model VD23).
Polyol formulation (Polyol) FDRw-based epoxide (300 g) was mixed with castor oil (100 g) and heated at 160 °C for 7 h under nitrogen inert atmosphere with constant stirring.
## 3.3. Polyester Synthesis
The selected reagents were the formulated polyol (P), azelaic acid (AzA, saturated diacid); citric acid (CA, cross-linker); and 1,4-butanediol (BTN) and sorbitol (S) both acting as chain-extender/mechanical enhancers. The prepolymer was prepared with a reagent mass ratio P: Aza: CA: BTN = 6:4:3:1:1, using an eco-friendly solvent and catalyst-free melt-polycondensation method [51], under nitrogen purge. In the first phase, all reagents except sorbitol were melted together at 160 °C for 1.5 h with constant stirring. After that period, the sorbitol was added, and the reaction was prolonged for 30 min. The reactor content was then quickly poured into a PTFE mold, and the curing was carried out in a vacuum drying oven for 3 days at 140 °C. The result was a thermoset copolyester.
## 3.4. Scaffolds Production
The scaffold production was carried out by salt leaching, an easily available and inexpensive technique used here as a proof-of-concept. A known quantity of sieved (Sieve shaker AS200, Retsch GmbH, Düsseldorf, Germany) sodium chloride crystals, with particle size between 315 and 500 μm, was mixed with the prepolymer right before casting. The content was partially poured into the square PTFE mold, resulting in a flat 2D membrane with a thin, porous layer. To obtain cylindrical structures with a different pore distribution (noted as 3D), another fraction was poured into polypropylene tubes (10 mL) and quickly centrifuged at 10,000 rpm. Both molds were placed in the vacuum drying oven and cured at 140 °C for 3 days. After cooling to room temperature, the tubes were cut into slices of similar dimensions and cooled down in the refrigerator at 4.0 ± 2.0 °C to allow the polymer detachment. The retained salt porogens were then dissolved with lukewarm deionized water, and to control the leaching process, conductivity measurements were periodically taken (pH/mV Mettler-Toledo Seven Excellence Multiparameter), until the registered value was similar to the solvent conductivity.
## 3.5. FDR, Epoxide, and Polyol Characterization
Density was determined by the pycnometer method following ASTM D792-08.
Dynamic viscosity was determined using a vibrational viscometer (SC-10, Scansci, Vila Nova de Gaia, Portugal). This parameter was evaluated in the initial and intermediate products (FDRw, epoxide, and polyol).
Epoxidation and hydroxylation yields were based on the FDRw initial input and were determined by Equation [1]. [ 1]Yield (%)=mfmi×100 where mf is the weight of the final product, and mi the weight of the initial input.
Iodine value variation or, just another way of saying, the conversion of unsaturated bonds (Ivr) was determined by the Equation [2]:[2]Ivr (%)=(Iv0−Ivf)Iv0×100 where *Ivo is* the initial (FDRw) iodine value, and *Ivf is* the iodine value after epoxidation stage.
Oxirane oxygen content (O.O.C.), was determined by titrimetric analysis following the HCl-acetone ultrasonic assisted method [105] and computed by Equation [3] after [106,107]. [ 3]O.O.C. exp. (%) =(V0−V)×M×16(1000×W)×100=(V0−V)×M×1.6W where V0 = Volume of NaOH solution required for the blank; V = Volume of NaOH solution required for the sample; M = NaOH solution molarity (mol/L); W = sample weight (g).
The relative conversion of the oxirane oxygen content (RCO) was determined by the expression [4]:[4]RCO (%)=O.O.C. exp. O.O.C theor.×100 where [5]O.O.C. theoretical (%)=[(IV02AI2)(100+IV02AI2)×AOx]×AOx×100 where, AI2 = iodine atomic mass (126.9); AOx = oxygen atomic mass (16.0); Ivo = iodine value for the initial sample (130 gI$\frac{2}{100}$ g).
Acid and hydroxyl (–OH) values, were determined by titrimetric analysis following ASTM D4662-08 and ASTM D1957-86, respectively.
Structural analysis was performed using Fourier Transform Infrared Spectroscopy (FTIR) to envision the chemical structure of all compounds in the near infrared range, between 4000 and 700 cm−1. The equipment used was a Perkin Elmer FT-IR spectrophotometer, fitted with Pike Miracle ATR accessory containing ZnSe crystal.
Polyol molecular weight distribution was determined by size-exclusion chromatography using a high performance liquid chromatograph (model 1260 Infinity II, Agilent Technologies, Santa Clara, CA, USA) attached to an Evaporative Light Scattering Detector (ELSD, 1290 Infinity II, Agilent Technologies, Santa Clara, CA, USA) with evaporation temperature at 70 °C and nebulization at 65 °C, using nitrogen as nebulizing gas coupled to a TSK gel GMHxL column for insoluble polymers. The isocratic analysis was carried out with tetrahydrofuran as the mobile phase; flow rate of 0.6 mL·min−1; sample concentrations of 20–25 mg·mL−1 dissolved in THF and injection volumes of 20 μL. The molecular weight was estimated by a calibration curve of polystyrene-standards 400–303,000 Da (Agilent (Waldbronn, Germany).
Polyol functionality (f), defined by Equation [6], was calculated based on the average molecular weight (Mw) and equivalent weight (E) was determined using –OH number. [ 6]f=MwE where E=(1000×56.1)[−OH]
## 3.6. Polyester Characterization
Static contact angle of the liquid−solid interface was determined using water, squalane oil, and cured polyester. This property was measured by the sessile drop method at room temperature using the tensiometer (Attention® Theta Lite, Biolin Scientific, Gothenburg, Sweden) with OneAttension software version 4.0.2. Before the experiment, the polyester film was cleaned, dried, and cut into 20 × 20 mm squares. A small drop of fluid (20 μL) was placed in the polymer surface and the angle made with the tangent was recorded. Since for biopolymers, no contact angle equilibrium is expected, but rather a pronounced variation during the first 60 s (Farris et al., 2011), the analysis was focused on that time window. The measurements were carried out at three different positions on the sample surface and at three different moments in time.
Gel content was calculated following [75], as the fraction of the polyester insoluble in dimethyl sulfoxide (DMSO). Cured polyester strips were cut out with dimensions 40 × 10 × 2 mm, weighed (Wi), and immersed in the solvent for 24 h at room temperature. The insoluble fraction was filtered in a pre-weighed filter paper and dried under vacuum at 80 °C overnight. We is the recorded weight of the dried samples after being extracted. The gel content determined using Equation [7] was taken as the average of three samples. [ 7]Gel content (%)=WeWi×100 Water absorption was determined following ASTM-D 570-98, replacing water by PBS solution to mimic physiological conditions. Dried disk samples with a known weight (Wd) were immersed in PBS (pH = 7.4 ± 0.2) solution at 37 °C for 24 h. After that, the samples were withdrawn from the liquid; the excess of solution was removed from the surface with filter paper and weighed (Wt). The water absorption was determined by Equation [8] and is expressed as a percentage (%). [ 8]Water absorption (%)=Wt−WdWd×100 Mechanical performance was evaluated using a Texture Analyser (model TA.XT plus C) with data acquisition and treatment software Express Connect v7.3 (Texture Technologies Corp. and Stable Microsystems Ltd., Hamilton, MA, USA). Several longitudinal strips with dimensions 100 × 10 × 2 mm were cut from the polyester mat and each one was attached to miniature tensile grips. The experiment was carried out at room temperature using a 30 Kg load cell with a strain rate of 2 mm·min−1. The elastic modulus (Young modulus) and % elongation were the determined properties.
Thermal analysis was performed by differential scanning calorimetry (DSC) using a calorimeter (DSC 204, NETZSCH GmbH Co.Holding KG, Bayern, Germany), and nitrogen as the purge gas (40 mL·min−1). Approximately 2 mg of each sample was placed in aluminum pans and the thermal properties were recorded between −70 and 300 °C at 10 °C ·min−1 to observe the melt (Tm) and glass transition (Tg) temperatures. The latter was measured on the second heating ramp to erase the thermal history of the polymer.
In vitro cytotoxicity of the cured polyester was evaluated by indirect contact. The human keratinocyte cell line HaCaT (CLS) was kept in culture in DMEM media (Gibco, Waltham, MA, USA) supplemented with 10 % FBS (Gibco) and $1\%$ antibiotic (Gibco) at 37 °C, with $5\%$ CO2 in a humidified atmosphere. To perform assays with cell lines, polymer discs were sterilized by immersion in ethanol for 1 h and briefly washed with sterile PBS prior to the experiment. Polymer discs were incubated with media in 24 well plates for 3 min and 24 h at 37 °C, with $5\%$ CO2 in a humidified atmosphere. Then, a previously seeded 96-well plate with HaCaT at 1 × 104 cells/well was incubated with the conditioned media in quadruplicate. Wells with only conditioned media (without cells) were used to subtract a possible influence of the samples in the PrestoBlue fluorescence signal. Cells treated with $10\%$ DMSO (dimethyl sulfoxide) were used as a negative control. After 24 h of exposure to the conditioned media, cytotoxicity was evaluated by the metabolic inhibition using a PrestoBlue assay (Thermo Fischer), according to the manufacturer’s instructions. PrestoBlue reagent was added to the media and incubated for 2 h at 37 °C, with $5\%$ CO2 in a humidified atmosphere. The fluorescence signal was read in a Synergy H1 microplate reader (BioTek, Winooski, VT, USA). The results are expressed in percentage of cell viability as compared with a control (cells with plain media). At least two independent experiments were performed.
In vitro degradation in physiological conditions was determined in accordance with ISO 175:2010, in a climatized chamber (Climacell 11 Ecoline, MMM Medcenter Einrichtungen, GmbH, München, Germany) under accelerated conditions, i.e., with a temperature of 40.0 ± 2.0 °C and relative humidity (RH) of 75 ± $5\%$. Pre-weighed (mi) dried disks of very similar dimensions (diameter = 10.7 ± 0.01 mm and thickness of 1.18 ± 0.02 mm) were individually placed in closed glass vessels containing 10 mL of PBS solution (pH of 7.4 ± 0.1) and placed in the controlled environment chamber. Weekly, one of the vessels was retrieved and the disk was dried at 50 °C for 48 h. The sample final weight (mf) was registered, along with the pH of the remaining saline solution. The mass loss (%) related with sampling time was determined using Equation [9]. [ 9]mass loss (%)=mi−mfmi×100
## 3.7. Scaffold Properties
Surface morphology was visualized using a JEOL-5600 LV Scanning Electron Microscope (Tokyo, Japan) from JEOL, Japan, equipped with a SPRITE HR Four Axis Stage controller (Deben Research). Samples were placed directly on double-sided adhesive carbon tape (NEM tape, Nisshin, Japan), placed on metallic stubs covered with adhesive carbon tape (NEM tape, from Nisshin, Japan) and coated with gold/palladium using a Sputter Coater (Polaron, from Bad Schwalbach, Germany). All observations were performed in high-vacuum with an acceleration voltage of 30 kV, at working distance of 12–13 mm and a spot-size of 20.
In vitro dye adsorption/release was monitored with a UV-Vis spectrophotometer (Shimadzu model UV-1900), using water-soluble dye (Rhodamine B base, RBB) on 3D scaffolds and hydrophobic curcumin (CRC) in 2D scaffolds to study the behavior of previously prepared scaffold disks. With that purpose, 100 mL of each dye were prepared dissolving a certain quantity of RBB in deionized water, and CRC in PBS ethanolic solution (EtOH 35 % v/v). Sets of pre-weighed and dried scaffold disks were immersed in 20 mL of dye solution at 20.0 mg /mL and 7.2 mg/mL for RBB and CRC, respectively, and the vessels were placed in an incubator shaker (Innova 40, series S) at 30 °C and 120 rpm for 7 h.
After that period, the scaffolds were removed, the excess liquid absorbed with paper filter and the disks dried overnight at 50 °C. The 3D scaffolds (RBB dye) were then placed in 20 mL of aqueous solution and the 2D disks (CRC loaded) were immersed in 20 mL of the PBS ethanolic solution. The amount of dye released was quantified hourly until equilibria conditions were achieved or within an established period of 48 h.
The concentration of the entrapped dyes was determined using standard calibration curves in the linear range at the wavelength of maximum absorbance, (RBB, 544 nm) and (CRC, 430 nm).
The loading capacity (L.C.) was calculated from the following Equation [10]:[10]Loading capacity (mg dyeg scf)=mdi−mdfmscf×100 mdi and mdf are the initial and final amount of dye in the contact solution (mg), and mscf the scaffold initial weight.
To evaluate the diffusional mechanism regarding the release kinetics of both dyes, we applied the Korsmeyer–Peppas model [108]. This empirical Equation [11] allows us to analyze both Fickian and non-Fickian release of drug from swelling and non-swelling polymeric delivery systems [102]. [ 11]Ln(MtM∞)=Ln(k)+n Ln(t) *In this* equation, Mt/M∞ is the fraction of dye (expressed in % of release) delivered at time t; k is the transport constant (dimension of time−1); and n is the transport exponent (dimensionless). The release constant k provides mostly information on the drug formulation such as structural characteristics, whereas n is important since it is related to the release mechanism (i.e., Fickian diffusion or non-Fickian diffusion) [109].
## 4. Conclusions
In this study, a fermentation by-product derived from the production of β-farnesene was used as feedstock in the synthesis of an aliphatic copolyester by polycondensation.
This polymer exhibited interesting elastomeric properties described by the relevant elasticity and low *Young modulus* along with low Tg, and high Tm.
The hydrophilic nature of this polymer was translated to relevant water uptake, followed by hydrolytic degradation by bulk erosion in saline solution within one year. This behavior suggests the possibility of easier degradation in landfill conditions (acidic, warm, and microorganism-rich), an added benefit concerning environmental impact.
The bio-based polyester showed evidence of non-toxicity to human keratinocyte cells, and dye release studies of the polyester scaffold showed a diffusion-controlled mechanism compatible with drug delivery applications.
These properties show a promising possibility, for instance, in the production of a nanofiber layer designed to carry bioactive molecules to integrate into a dressing for wounded or burnt skin tissue. Further studies will be necessary.
Considering these conclusions, this polymer offers a sustainable and eco-friendly alternative for the potential use of controlled release of active principles for wound dressing applications.
## References
1. Alhazmi H., Almansour F.H., Aldhafeeri Z.. **Plastic Waste Management: A Review of Existing Life Cycle Assessment Studies**. *Sustainability* (2021) **13**. DOI: 10.3390/su13105340
2. Wojnowska-Baryła I., Kulikowska D., Bernat K.. **Effect of Bio-Based Products on Waste Management**. *Sustainability* (2020) **12**. DOI: 10.3390/su12052088
3. Duan Y., Tarafdar A., Kumar V., Ganeshan P., Rajendran K., Shekhar Giri B., Gómez-García R., Li H., Zhang Z., Sindhu R.. **Sustainable Biorefinery Approaches towards Circular Economy for Conversion of Biowaste to Value Added Materials and Future Perspectives**. *Fuel* (2022) **325** 124846. DOI: 10.1016/j.fuel.2022.124846
4. Roux M., Varrone C.. **Assessing the Economic Viability of the Plastic Biorefinery Concept and Its Contribution to a More Circular Plastic Sector**. *Polymers* (2021) **13**. DOI: 10.3390/polym13223883
5. Talan A., Pokhrel S., Tyagi R.D., Drogui P.. **Biorefinery Strategies for Microbial Bioplastics Production: Sustainable Pathway towards Circular Bioeconomy**. *Bioresour. Technol. Rep.* (2022) **17** 100875. DOI: 10.1016/j.biteb.2021.100875
6. Hughes S.R., Gibbons W.R., Moser B.R., Rich J.O.. **Sustainable Multipurpose Biorefineries for Third-Generation Biofuels and Value-Added Co-Products**. *Biofuels—Economy, Environment and Sustainability* (2012) 13
7. Wellenreuther C., Wolf A.. *Innovative Feedstocks in Biodegradable Bio-Based Plastics: A Literature Review* (2020) 1-40
8. Baranwal J., Barse B., Fais A., Delogu G.L., Kumar A.. **Biopolymer: A Sustainable Material for Food and Medical Applications**. *Polymers* (2022) **14**. DOI: 10.3390/polym14050983
9. Spalvins K., Zihare L., Blumberga D.. **Single Cell Protein Production from Waste Biomass: Comparison of Various Industrial by-Products**. *Energy Procedia* (2018) **147** 409-418. DOI: 10.1016/j.egypro.2018.07.111
10. Yadav B., Chavan S., Atmakuri A., Tyagi R.D., Drogui P.. **A Review on Recovery of Proteins from Industrial Wastewaters with Special Emphasis on PHA Production Process: Sustainable Circular Bioeconomy Process Development**. *Bioresour. Technol.* (2020) **317** 124006. DOI: 10.1016/j.biortech.2020.124006
11. Aliko K., Aldakhlalla M.B., Leslie L.J., Worthington T., Topham P.D., Theodosiou E.. **Poly(Butylene Succinate) Fibrous Dressings Containing Natural Antimicrobial Agents**. *J. Ind. Text.* (2021) **51** 6948S-6967S. DOI: 10.1177/1528083720987209
12. Prabakaran R., Marie J.M., Xavier A.J.M.. **Biobased Unsaturated Polyesters Containing Castor Oil-Derived Ricinoleic Acid and Itaconic Acid: Synthesis, In Vitro Antibacterial, and Cytocompatibility Studies**. *ACS Appl. Bio Mater.* (2020) **3** 5708-5721. DOI: 10.1021/acsabm.0c00480
13. Vogt L., Ruther F., Salehi S., Boccaccini A.R.. **Poly(Glycerol Sebacate) in Biomedical Applications—A Review of the Recent Literature**. *Adv. Healthc. Mater.* (2021) **10** e2002026. DOI: 10.1002/adhm.202002026
14. Zamboulis A., Nakiou E.A., Christodoulou E., Bikiaris D.N., Kontonasaki E., Liverani L., Boccaccini A.R.. **Polyglycerol Hyperbranched Polyesters: Synthesis, Properties and Pharmaceutical and Biomedical Applications**. *Int. J. Mol. Sci.* (2019) **20**. DOI: 10.3390/ijms20246210
15. Bruggeman J.P.. *Biodegradable Polyol-Based Polymers A Polymer Platform for Biomedical Applications* (2008)
16. Chae T.U., Ahn J.H., Ko Y.S., Kim J.W., Lee J.A., Lee E.H., Lee S.Y.. **Metabolic Engineering for the Production of Dicarboxylic Acids and Diamines**. *Metab. Eng.* (2020) **58** 2-16. DOI: 10.1016/j.ymben.2019.03.005
17. Rebolledo-Leiva R., Moreira M.T., González-García S.. **Environmental Assessment of the Production of Itaconic Acid from Wheat Straw under a Biorefinery Approach**. *Bioresour. Technol.* (2022) **345** 126481. DOI: 10.1016/j.biortech.2021.126481
18. Todea A., Deganutti C., Spennato M., Asaro F., Zingone G., Milizia T., Gardossi L.. **Azelaic Acid: A Bio-Based Building Block for Biodegradable Polymers**. *Polymers* (2021) **13**. DOI: 10.3390/polym13234091
19. Rocha C.V., Gonçalves V., da Silva M.C., Bañobre-López M., Gallo J.. **PLGA-Based Composites for Various Biomedical Applications**. *Int. J. Mol. Sci.* (2022) **23**. DOI: 10.3390/ijms23042034
20. Sevim K., Pan J.. **A Model for Hydrolytic Degradation and Erosion of Biodegradable Polymers**. *Acta Biomater.* (2018) **66** 192-199. DOI: 10.1016/j.actbio.2017.11.023
21. RameshKumar S., Shaiju P., O’Connor K.E., Ramesh Babu P.. **Bio-Based and Biodegradable Polymers—State-of-the-Art, Challenges and Emerging Trends**. *Curr. Opin. Green Sustain. Chem.* (2020) **21** 75-81. DOI: 10.1016/j.cogsc.2019.12.005
22. Schoubben A., Ricci M., Giovagnoli S.. **Meeting the Unmet: From Traditional to Cutting-Edge Techniques for Poly Lactide and Poly Lactide-Co-Glycolide Microparticle Manufacturing**. *J. Pharm. Investig.* (2019) **49** 381-404. DOI: 10.1007/s40005-019-00446-y
23. Hinchliffe J.D., Parassini Madappura A., Syed Mohamed S.M.D., Roy I.. **Biomedical Applications of Bacteria-Derived Polymers**. *Polymers* (2021) **13**. DOI: 10.3390/polym13071081
24. Chilakamarry C.R., Sakinah A.M.M., Zularisam A.W., Pandey A.. **Glycerol Waste to Value Added Products and Its Potential Applications**. *Syst. Microbiol. Biomanuf.* (2021) **1** 378-396. DOI: 10.1007/s43393-021-00036-w
25. Fonder M.A., Lazarus G.S., Cowan D.A., Aronson-Cook B., Kohli A.R., Mamelak A.J.. **Treating the Chronic Wound: A Practical Approach to the Care of Nonhealing Wounds and Wound Care Dressings**. *J. Am. Acad. Dermatol.* (2008) **58** 185-206. DOI: 10.1016/j.jaad.2007.08.048
26. Singh Malik D., Mital N., Kaur G.. **Topical Drug Delivery Systems: A Patent Review**. *Expert Opin. Ther. Pat.* (2016) **26** 213-228. DOI: 10.1517/13543776.2016.1131267
27. Bu Y., Ma J., Bei J., Wang S.. **Surface Modification of Aliphatic Polyester to Enhance Biocompatibility**. *Front. Bioeng. Biotechnol.* (2019) **7** 98. DOI: 10.3389/fbioe.2019.00098
28. Niculescu A.G., Grumezescu A.M.. **An Up-to-Date Review of Biomaterials Application in Wound Management**. *Polymers* (2022) **14**. DOI: 10.3390/polym14030421
29. Okur M.E., Karantas I.D., Şenyiğit Z., Üstündağ Okur N., Siafaka P.I.. **Recent Trends on Wound Management: New Therapeutic Choices Based on Polymeric Carriers**. *Asian J. Pharm. Sci.* (2020) **15** 661-684. DOI: 10.1016/j.ajps.2019.11.008
30. Darie-Niță R.N., Râpă M., Frąckowiak S.. **Special Features of Polyester-Based Materials for Medical Applications**. *Polymers* (2022) **14**. DOI: 10.3390/polym14050951
31. Suarato G., Bertorelli R., Athanassiou A.. **Borrowing from Nature: Biopolymers and Biocomposites as Smart Wound Care Materials**. *Front. Bioeng. Biotechnol.* (2018) **6** 137. DOI: 10.3389/fbioe.2018.00137
32. Ijaola A.O., Akamo D.O., Damiri F., Akisin C.J., Bamidele E.A., Ajiboye E.G., Berrada M., Onyenokwe V.O., Yang S.-Y., Asmatulu E.. **Polymeric Biomaterials for Wound Healing Applications: A Comprehensive Review**. *J. Biomater. Sci. Polym. Ed.* (2022) **33** 1998-2050. DOI: 10.1080/09205063.2022.2088528
33. Negut I., Dorcioman G., Grumezescu V.. **Scaffolds for Wound Healing Applications**. *Polymers* (2020) **12**. DOI: 10.3390/polym12092010
34. Calori I.R., Braga G., de Jesus P.D.C.C., Bi H., Tedesco A.C.. **Polymer Scaffolds as Drug Delivery Systems**. *Eur. Polym. J.* (2020) **129** 109621. DOI: 10.1016/j.eurpolymj.2020.109621
35. Akhlaq M., Azad A.K., Fuloria S., Meenakshi D.U., Raza S., Safdar M., Nawaz A., Subramaniyan V., Sekar M., Sathasivam K.V.. **Fabrication of Tizanidine Loaded Patches Using Flaxseed Oil and Coriander Oil as a Penetration Enhancer for Transdermal Delivery**. *Polymers* (2021) **13**. DOI: 10.3390/polym13234217
36. Alven S., Peter S., Mbese Z., Aderibigbe B.A.. **Polymer-Based Wound Dressing Materials Loaded with Bioactive Agents: Potential Materials for the Treatment of Diabetic Wounds**. *Polymers* (2022) **14**. DOI: 10.3390/polym14040724
37. Gosai H., Patel P., Trivedi H., Joshi U., Kumar P., Kothari V.. **Role of Biodegradable Polymer-Based Biomaterials in Advanced Wound Care**. *Wound Healing Research* (2021) 599-620
38. Guimarães I., Baptista-Silva S., Pintado M., Oliveira A.L.. **Polyphenols: A Promising Avenue in Therapeutic Solutions for Wound Care**. *Appl. Sci.* (2021) **11**. DOI: 10.3390/app11031230
39. Williams D.F.. **Challenges With the Development of Biomaterials for Sustainable Tissue Engineering**. *Front. Bioeng. Biotechnol.* (2019) **7** 127. DOI: 10.3389/fbioe.2019.00127
40. Zamri M.F.M.A., Bahru R., Amin R., Aslam Khan M.U., Razak S.I.A., Hassan S.A., Kadir M.R.A., Nayan N.H.M.. **Waste to Health: A Review of Waste Derived Materials for Tissue Engineering**. *J. Clean. Prod.* (2021) **290** 125792. DOI: 10.1016/j.jclepro.2021.125792
41. Teixeira F.S., Vidigal S.S.M.P., Pimentel L.L., Costa P.T., Tavares-Valente D., Azevedo-Silva J., Pintado M.E., Fernandes J.C., Rodríguez-Alcalá L.M.. **Phytosterols and Novel Triterpenes Recovered from Industrial Fermentation Coproducts Exert In Vitro Anti-Inflammatory Activity in Macrophages**. *Pharmaceuticals* (2021) **14**. DOI: 10.3390/ph14060583
42. Marcovich N.E., Kurańska M., Prociak A., Malewska E., Bujok S.. **The Effect of Different Palm Oil-Based Bio-Polyols on Foaming Process and Selected Properties of Porous Polyurethanes**. *Polym. Int.* (2017) **66** 1522-1529. DOI: 10.1002/pi.5408
43. Moritz H.U.. **Increase in Viscosity and Its Influence on Polymerization Processes**. *Chem. Eng. Technol.* (1989) **12** 71-87. DOI: 10.1002/ceat.270120112
44. Gharby S.. **Refining Vegetable Oils: Chemical and Physical Refining**. *Sci. World J.* (2022) **2022** 6627013. DOI: 10.1155/2022/6627013
45. Salimon J., Abdullah B.M., Salih N.. **Optimization of the Oxirane Ring Opening Reaction in Biolubricant Base Oil Production**. *Arab. J. Chem.* (2016) **9** S1053-S1058. DOI: 10.1016/j.arabjc.2011.11.002
46. Hosney H., Nadiem B., Ashour I., Mustafa I., El-Shibiny A.. **Epoxidized Vegetable Oil and Bio-Based Materials as PVC Plasticizer**. *J. Appl. Polym. Sci.* (2018) **135** 46270. DOI: 10.1002/app.46270
47. Thampi A.D., John A.R., Arif M.M., Rani S.. **Evaluation of the Tribological Properties and Oxidative Stability of Epoxidized and Ring Opened Products of Pure Rice Bran Oil**. *Proc. Inst. Mech. Eng. Part J J. Eng. Tribol.* (2021) **235** 1093-1100. DOI: 10.1177/1350650120950535
48. Jia P., Zhang M., Hu L., Feng G., Bo C., Zhou Y.. **Synthesis and Application of Environmental Castor Oil Based Polyol Ester Plasticizers for Poly(Vinyl Chloride)**. *ACS Sustain. Chem. Eng.* (2015) **3** 2187-2193. DOI: 10.1021/acssuschemeng.5b00449
49. Kunduru K.R., Basu A., Zada M.H., Domb A.J.. **Castor Oil-Based Biodegradable Polyesters**. *Biomacromolecules* (2015) **16** 2572-2587. DOI: 10.1021/acs.biomac.5b00923
50. Ma Y., Wang R., Li Q., Li M., Liu C., Jia P.. **Castor Oil as a Platform for Preparing Bio-Based Chemicals and Polymer Materials**. *Green Mater.* (2021) **10** 99-109. DOI: 10.1680/jgrma.20.00085
51. Kolanthai E., Sarkar K., Meka S.R.K., Madras G., Chatterjee K.. **Copolyesters from Soybean Oil for Use as Resorbable Biomaterials**. *ACS Sustain. Chem. Eng.* (2015) **3** 880-891. DOI: 10.1021/acssuschemeng.5b00001
52. Seneviratne K., Jayathilaka N.. *Coconut Oil: Chemistry and Nutrition* (2016)
53. Adekunle K.F.. **A Review of Vegetable Oil-Based Polymers: Synthesis and Applications**. *Open J. Polym. Chem.* (2015) **5** 34-40. DOI: 10.4236/ojpchem.2015.53004
54. Neswati N.. **Combination of Temperature and Time in Epoxidation for Producing Epoxidized Palm Oil as Source of Bio Polyol**. *IOP Conf. Ser. Earth Environ. Sci.* (2021) **757** 012069. DOI: 10.1088/1755-1315/757/1/012069
55. Coman A.E., Peyrton J., Hubca G., Sarbu A., Gabor A.R., Nicolae C.A., Iordache T.V., Averous L.. **Synthesis and Characterization of Renewable Polyurethane Foams Using Different Biobased Polyols from Olive Oil**. *Eur. Polym. J.* (2021) **149** 110363. DOI: 10.1016/j.eurpolymj.2021.110363
56. Gomna A., N’Tsoukpoe K.E., Le Pierrès N., Coulibaly Y.. **Thermal Stability of a Vegetable Oil-Based Thermal Fluid at High Temperature**. *Afr. J. Sci. Technol. Innov. Dev.* (2020) **12** 317-326. DOI: 10.1080/20421338.2020.1732080
57. Darie-Niță R.N., Irimia A., Grigoraș V.C., Mustață F., Tudorachi N., Râpă M., Ludwiczak J., Iwanczuk A.. **Evaluation of Natural and Modified Castor Oil Incorporation on the Melt Processing and Physico-Chemical Properties of Polylactic Acid**. *Polymers* (2022) **14**. DOI: 10.3390/polym14173608
58. Ionescu M., Radojčić D., Wan X., Shrestha M.L., Petrović Z.S., Upshaw T.A.. **Highly Functional Polyols from Castor Oil for Rigid Polyurethanes**. *Eur. Polym. J.* (2016) **84** 736-749. DOI: 10.1016/j.eurpolymj.2016.06.006
59. Beltrán A.A., Boyacá L.A.. **Preparation of Oleochemical Polyols Derived from Soybean Oil**. *Lat. Am. Appl. Res.* (2011) **41** 69-74
60. **Manual of Methods of Analysis of Foods—Oils and Fats**. *Laboratory Manual* (2015) 96
61. Veloso-Fernández A., Laza J.M., Ruiz-Rubio L., Martín A., Taguado M., Benito-Vicente A., Martín C., Vilas J.L.. **Towards a New Generation of Non-Cytotoxic Shape Memory Thermoplastic Polyurethanes for Biomedical Applications**. *Mater. Today Commun.* (2022) **33** 104730. DOI: 10.1016/j.mtcomm.2022.104730
62. Jalil M.J., Md Zaini M.S., Mohd Yamin A.F., Azmi I.S., Chang S.H., Morad N., Hadi A.. **Synthesis and Physicochemical Properties of Epoxidized Oleic Acid-Based Palm Oil**. *IOP Conf. Ser. Earth Environ. Sci.* (2019) **291** 012046. DOI: 10.1088/1755-1315/291/1/012046
63. El-Ghazawy R.A.M., Farag R.K., Elsaeed S.M., Abde-Halim E.-D.A., Yossef M.A., Toyor W.E.. **Castor Oil Based Organogels: I. Synthesis, Swelling, and Network Parameters**. *J. Dispers. Sci. Technol.* (2014) **35** 350-357. DOI: 10.1080/01932691.2013.788450
64. Rai R., Tallawi M., Grigore A., Boccaccini A.R.. **Synthesis, Properties and Biomedical Applications of Poly(Glycerol Sebacate) (PGS): A Review**. *Prog. Polym. Sci.* (2012) **37** 1051-1078. DOI: 10.1016/j.progpolymsci.2012.02.001
65. Menzies K.L., Jones L.. **The Impact of Contact Angle on the Biocompatibility of Biomaterials**. *Optom. Vis. Sci.* (2010) **87** 387-399. DOI: 10.1097/OPX.0b013e3181da863e
66. Kim S.-K., Karadeniz F.. **Biological Importance and Applications of Squalene and Squalane**. *Adv. Food Nutr. Res.* (2012) **65** 223-233. DOI: 10.1016/B978-0-12-416003-3.00014-7
67. Yamaguchi Y., Nagasawa T., Kitagawa A., Nakamura N., Matsumoto K., Uchiwa H., Hirata K., Igarashi R.. **New Nanotechnology for the Guided Tissue Regeneration of Skin—Potential of Lyotropic Liquid Crystals**. *Pharmazie* (2006) **61** 112-116. PMID: 16526557
68. Farris S., Introzzi L., Biagioni P., Holz T., Schiraldi A., Piergiovanni L.. **Wetting of Biopolymer Coatings: Contact Angle Kinetics and Image Analysis Investigation**. *Langmuir* (2011) **27** 7563-7574. DOI: 10.1021/la2017006
69. Dai J., Ma S., Wu Y., Han L., Zhang L., Zhu J., Liu X.. **Polyesters Derived from Itaconic Acid for the Properties and Bio-Based Content Enhancement of Soybean Oil-Based Thermosets**. *Green Chem.* (2015) **17** 2383-2392. DOI: 10.1039/C4GC02057J
70. Witono J.R., Noordergraaf I.W., Heeres H.J., Janssen L.P.B.M.. **Water Absorption, Retention and the Swelling Characteristics of Cassava Starch Grafted with Polyacrylic Acid**. *Carbohydr. Polym.* (2014) **103** 325-332. DOI: 10.1016/j.carbpol.2013.12.056
71. Tang S., Li J., Wang R., Zhang J., Lu Y., Hu G., Wang Z., Zhang L.. **Current Trends in Bio-based Elastomer Materials**. *SusMat* (2022) **2** 2-33. DOI: 10.1002/sus2.45
72. Lee C.-H., Hung K.-C., Hsieh M.-J., Chang S.-H., Juang J.-H., Hsieh I.-C., Wen M.-S., Liu S.-J.. **Core-Shell Insulin-Loaded Nanofibrous Scaffolds for Repairing Diabetic Wounds**. *Nanomed. Nanotechnol. Biol. Med.* (2020) **24** 102123. DOI: 10.1016/j.nano.2019.102123
73. Minsart M., Van Vlierberghe S., Dubruel P., Mignon A.. **Commercial Wound Dressings for the Treatment of Exuding Wounds: An in-Depth Physico-Chemical Comparative Study**. *Burn. Trauma* (2022) **10** tkac024. DOI: 10.1093/burnst/tkac024
74. Lang K., Regina J.S., Gross R.A., Linhardt R.J.. **Review on the Impact of Polyols on the Properties of Bio-Based Polyesters**. *Polymers* (2020) **2969**. DOI: 10.3390/polym12122969
75. Pantic O., Spasojevic M., Dzunuzovic E., Nikolic M.S., Savic S., Markovic M., Spasojevic P.. **The Effect of Glycol Derivatives on the Properties of Bio-Based Unsaturated Polyesters**. *Polymers* (2022) **14**. DOI: 10.3390/polym14152970
76. Tian J., Cao Z., Qian S., Xia Y., Zhang J., Kong Y., Sheng K., Zhang Y., Wan Y., Takahashi J.. **Improving Tensile Strength and Impact Toughness of Plasticized Poly(Lactic Acid) Biocomposites by Incorporating Nanofibrillated Cellulose**. *Nanotechnol. Rev.* (2022) **11** 2469-2482. DOI: 10.1515/ntrev-2022-0142
77. Wang M., Xue H., Feng Z., Cheng B., Yang H.. **Increase of Tensile Strength and Toughness of Bio-Based Diglycidyl Ether of Bisphenol A with Chitin Nanowhiskers**. *PLoS ONE* (2017) **12**. DOI: 10.1371/journal.pone.0177673
78. Neffe A.T., Izraylit V., Hommes-Schattmann P.J., Lendlein A.. **Soft, Formstable (Co)Polyester Blend Elastomers**. *Nanomaterials* (2021) **11**. DOI: 10.3390/nano11061472
79. Ling Y.T.Q., Yap Y.J., Heng Y.X., Lee S.Y., Koh R.Y., Ang D.T.C., Chia C.H., Gan S.N.. **Physiochemical and In-Vitro Cytotoxicity Properties of Biocompatible Palm Fatty Acid-Based Polyesters**. *Sains Malays.* (2021) **50** 395-407. DOI: 10.17576/jsm-2021-5002-11
80. Zulkifli N.N.B., Badri K.B.H., Nor M.A.A.M., Amin K.A.M.. **Palm Kernel Oil-Based Polyurethane Film: Biocompatibility and Antibacterial Activity Studies**. *AIP Conf. Proc.* (2017) **1817** 020005. DOI: 10.1063/1.4976757
81. Song R., Murphy M., Li C., Ting K., Soo C., Zheng Z.. **Current Development of Biodegradable Polymeric Materials for Biomedical Applications**. *Drug Des. Devel. Ther.* (2018) **12** 3117-3145. DOI: 10.2147/DDDT.S165440
82. Rowe M.D., Eyiler E., Walters K.B.. **Hydrolytic Degradation of Bio-Based Polyesters: Effect of PH and Time**. *Polym. Test.* (2016) **52** 192-199. DOI: 10.1016/j.polymertesting.2016.04.015
83. Winzenburg G., Schmidt C., Fuchs S., Kissel T.. **Biodegradable Polymers and Their Potential Use in Parenteral Veterinary Drug Delivery Systems**. *Adv. Drug Deliv. Rev.* (2004) **56** 1453-1466. DOI: 10.1016/j.addr.2004.02.008
84. Woodard L.N., Grunlan M.A.. **Hydrolytic Degradation and Erosion of Polyester Biomaterials**. *ACS Macro Lett.* (2018) **7** 976-982. DOI: 10.1021/acsmacrolett.8b00424
85. Tatu R.R., Oria M., Rao M.B., Peiro J.L., Lin C.Y.. **Biodegradation of Poly(l-Lactic Acid) and Poly(ε-Caprolactone) Patches by Human Amniotic Fluid in an in-Vitro Simulated Fetal Environment**. *Sci. Rep.* (2022) **12** 3950. DOI: 10.1038/s41598-022-07681-8
86. Şucu T., Shaver M.P.. **Inherently Degradable Cross-Linked Polyesters and Polycarbonates: Resins to Be Cheerful**. *Polym. Chem.* (2020) **11** 6397-6412. DOI: 10.1039/D0PY01226B
87. Liu S., Qin S., He M., Zhou D., Qin Q., Wang H.. **Current Applications of Poly(Lactic Acid) Composites in Tissue Engineering and Drug Delivery**. *Compos. Part B Eng.* (2020) **199** 108238. DOI: 10.1016/j.compositesb.2020.108238
88. Sawadkar P., Mohanakrishnan J., Rajasekar P., Rahmani B., Kohli N., Bozec L., García-Gareta E.. **A Synergistic Relationship between Polycaprolactone and Natural Polymers Enhances the Physical Properties and Biological Activity of Scaffolds**. *ACS Appl. Mater. Interfaces* (2020) **12** 13587-13597. DOI: 10.1021/acsami.9b19715
89. Lam C.X., Teoh S.H., Hutmacher D.W.. **Comparison of the Degradation of Polycaprolactone and Polycaprolactone–(β-Tricalcium Phosphate) Scaffolds in Alkaline Medium**. *Polym. Int.* (2007) **56** 718-728. DOI: 10.1002/pi.2195
90. Bartnikowski M., Dargaville T.R., Ivanovski S., Hutmacher D.W.. **Degradation Mechanisms of Polycaprolactone in the Context of Chemistry, Geometry and Environment**. *Prog. Polym. Sci.* (2019) **96** 1-20. DOI: 10.1016/j.progpolymsci.2019.05.004
91. Kliem S., Kreutzbruck M., Bonten C.. **Review on the Biological Degradation of Polymers in Various Environments**. *Materials* (2020) **13**. DOI: 10.3390/ma13204586
92. Chamas A., Moon H., Zheng J., Qiu Y., Tabassum T., Jang J.H., Abu-Omar M., Scott S.L., Suh S.. **Degradation Rates of Plastics in the Environment**. *ACS Sustain. Chem. Eng.* (2020) **8** 3494-3511. DOI: 10.1021/acssuschemeng.9b06635
93. Prasadh S., Wong R.C.W.. **Unraveling the Mechanical Strength of Biomaterials Used as a Bone Scaffold in Oral and Maxillofacial Defects**. *Oral Sci. Int.* (2018) **15** 48-55. DOI: 10.1016/S1348-8643(18)30005-3
94. Oh S.H., Park I.K., Kim J.M., Lee J.H.. **In Vitro and in Vivo Characteristics of PCL Scaffolds with Pore Size Gradient Fabricated by a Centrifugation Method**. *Biomaterials* (2007) **28** 1664-1671. DOI: 10.1016/j.biomaterials.2006.11.024
95. Kamaly N., Yameen B., Wu J., Farokhzad O.C.. **Nanoparticles: Mechanisms of Controlling Drug Release**. *Chem. Rev.* (2016) **116** 2602-2663. DOI: 10.1021/acs.chemrev.5b00346
96. Dai J., Gu L., Su Y., Wang Q., Zhao Y., Chen X., Deng H., Li W., Wang G., Li K.. **Inhibition of Curcumin on Influenza A Virus Infection and Influenzal Pneumonia via Oxidative Stress, TLR2/4, P38/JNK MAPK and NF-ΚB Pathways**. *Int. Immunopharmacol.* (2018) **54** 177-187. DOI: 10.1016/j.intimp.2017.11.009
97. Hussain Z., Thu H.E., Amjad M.W., Hussain F., Ahmed T.A., Khan S.. **Exploring Recent Developments to Improve Antioxidant, Anti-Inflammatory and Antimicrobial Efficacy of Curcumin: A Review of New Trends and Future Perspectives**. *Mater. Sci. Eng. C* (2017) **77** 1316-1326. DOI: 10.1016/j.msec.2017.03.226
98. Sideek S.A., El-Nassan H.B., Fares A.R., ElMeshad A.N., Elkasabgy N.A.. **Different Curcumin-Loaded Delivery Systems for Wound Healing Applications: A Comprehensive Review**. *Pharmaceutics* (2023) **15**. DOI: 10.3390/pharmaceutics15010038
99. Balla E., Daniilidis V., Karlioti G., Kalamas T., Stefanidou M., Bikiaris N.D., Vlachopoulos A., Koumentakou I., Bikiaris D.N.. **Poly(Lactic Acid): A Versatile Biobased Polymer for the Future with Multifunctional Properties—From Monomer Synthesis, Polymerization Techniques and Molecular Weight Increase to PLA Applications**. *Polymers* (2021) **13**. DOI: 10.3390/polym13111822
100. Bloomquist C.J., Mecham M.B., Paradzinsky M.D., Janusziewicz R., Warner S.B., Luft J.C., Mecham S.J., Wang A.Z., DeSimone J.M.. **Controlling Release from 3D Printed Medical Devices Using CLIP and Drug-Loaded Liquid Resins**. *J. Control. Release* (2018) **278** 9-23. DOI: 10.1016/j.jconrel.2018.03.026
101. Liu Y., Gai M., Sukvanitvichai D., Frueh J., Sukhorukov G.B.. **PH Dependent Degradation Properties of Lactide Based 3D Microchamber Arrays for Sustained Cargo Release**. *Colloids Surf. B Biointerfaces* (2020) **188** 110826. DOI: 10.1016/j.colsurfb.2020.110826
102. Paarakh M.P., Jose P.A.N.I., Setty C.M., Peter G.. **V Release Kinetics—Concepts and Applications**. *Int. J. Pharm. Res. Technol.* (2019) **8** 12-20. DOI: 10.31838/ijprt/08.01.02
103. Janmohammadi M., Nourbakhsh M.S., Bonakdar S.. **Electrospun Skin Tissue Engineering Scaffolds Based on Polycaprolactone/Hyaluronic Acid/L-Ascorbic Acid**. *Fibers Polym.* (2021) **22** 19-29. DOI: 10.1007/s12221-021-0036-8
104. Uprety B.K., Reddy J.V., Dalli S.S., Rakshit S.K.. **Utilization of Microbial Oil Obtained from Crude Glycerol for the Production of Polyol and Its Subsequent Conversion to Polyurethane Foams**. *Bioresour. Technol.* (2017) **235** 309-315. DOI: 10.1016/j.biortech.2017.03.126
105. He Z., Wang Y., Zhao T., Ye Z.. **Huang, Ultrasonication-assisted rapid determination of epoxide values in polymer mixtures containing epoxy resin**. *Anal. Methods* (2014) **6** 4257-4261. DOI: 10.1039/C4AY00439F
106. Jalil M.J., Jamaludin S.K., Daud A.R.M.. **Degradation Oxirane Ring Kinetics of Epoxidized Palm Kernel Oil-Based Crude Oleic Acid**. *Chem. Chem. Technol.* (2018) **12** 296-299. DOI: 10.23939/chcht12.03.296
107. Silviana D.D., Anggoro A.C.. **Waste Cooking Oil Utilisation as Bio-Plasticiser through Epoxidation Using Inorganic Acids as Homogeneous Catalysts**. *Chem. Eng. Trans.* (2017) **56** 1861-1866. DOI: 10.3303/CET1756311
108. Korsmeyer R.W., Gurny R., Doelker E., Buri P., Peppas N.A.. **Mechanisms of Solute Release from Porous Hydrophilic Polymers**. *Int. J. Pharm.* (1983) **15** 25-35. DOI: 10.1016/0378-5173(83)90064-9
109. Wu I.Y., Bala S., Škalko-Basnet N., di Cagno M.P.. **Interpreting Non-Linear Drug Diffusion Data: Utilizing Korsmeyer-Peppas Model to Study Drug Release from Liposomes**. *Eur. J. Pharm. Sci.* (2019) **138** 105026. DOI: 10.1016/j.ejps.2019.105026
|
---
title: Relation of amino acid composition, hydrophobicity, and molecular weight with
antidiabetic, antihypertensive, and antioxidant properties of mixtures of corn gluten
and soy protein hydrolysates
authors:
- Homaira Mirzaee
- Hassan Ahmadi Gavlighi
- Mehdi Nikoo
- Chibuike C. Udenigwe
- Faramarz Khodaiyan
journal: Food Science & Nutrition
year: 2022
pmcid: PMC10003021
doi: 10.1002/fsn3.3160
license: CC BY 4.0
---
# Relation of amino acid composition, hydrophobicity, and molecular weight with antidiabetic, antihypertensive, and antioxidant properties of mixtures of corn gluten and soy protein hydrolysates
## Abstract
New mixed Alcalase‐hydrolysates were developed using corn gluten meal (CP) and soy protein (SP) hydrolysates, namely CPH, SPH, SPH30:CPH70, SPH70:CPH30, and SPH50:CPH50. Amino acid profile, surface hydrophobicity (H 0), molecular weight (MW) distribution, antioxidant activity, angiotensin‐converting enzyme (ACE), α‐amylase, and α‐glucosidase inhibitory activities, and functional characteristics of hydrolysates were determined. Hydrolysis changed the amount of hydrophilic and hydrophobic amino acid composition and significantly increased the H 0 values of hydrolysates, especially for CPH. The DPPH radical scavenging activity (RSA) was higher for CPH, SPH30:CPH70, and SPH50:CPH50 than SPH and SPH70:CPH30. Moreover, SPH, SPH70:CPH30, and SPH50:CPH50 showed lower MW than CPH, and this correlated with the higher hydrophilicity, and ABTS and hydroxyl RSA values obtained for SPH and the mixed hydrolysates with predominantly SPH. SPH70:CPH30 exhibited higher ACE, α‐glucosidase, and α‐amylase inhibitory activities among all samples due to its specific peptides with high capacity to interact with amino acid residues located at the enzyme active site and also low binding energy. At $15\%$ degree of hydrolysis, both SPH and CPH showed enhanced solubility at pH 4.0, 7.0 and 9.0, emulsifying activity, and foaming capacity. Taken together, SPH70:CPH30 displayed strong antioxidant, antihypertensive, and antidiabetic attributes, emulsifying activity and stability indexes, and foaming capacity and foaming stability, making it a promising multifunctional ingredient for the development of functional food products.
In this study, new formulation of mixed Alcalase‐hydrolysates was developed using corn gluten meal (CP) and soy protein (SP) hydrolysates, namely CPH, SPH, SPH30:CPH70, SPH70:CPH30, and SPH50:CPH50. Amino acid profile, surface hydrophobicity (H0), molecular weight (MW) distribution, antioxidant activity, angiotensin‐converting enzyme (ACE) and antidiabetic activities, and functional characteristics of hydrolysates were determined. For CPH, protein or peptide bands with MW of 30–40, 50–55, 70–80, and 150–170 kDa were observed. For SPH, most of the bands at 70–200 kDa assigned to SP disappeared, and new bands were formed at 10–15, 15–20, 30, 40–50, and 50–60 kDa. SPH70:CPH30 revealed higher ACE, a‐glucosidase, and a‐amylase inhibitory activities among all samples. At $15\%$ degree of hydrolysis, both SPH and CPH showed enhanced solubility at pH 3.0, 7.0, and 9.0, emulsifying activity, and foaming capacity. Taken together, SPH70:CPH30 with high antioxidant, antihypertensive, and antidiabetic attributes. These results have important implications for the designing of mixed protein hydrolysate as by‐product used for the development of novel antidiabetic, antihypertensive, and antioxidant additives for food industrial application.
## INTRODUCTION
Plant proteins have been widely considered as sustainable ingredients for the development of bioactive peptides and hydrolysates with antidiabetic, antihypertensive, and antioxidant activities (Das et al., 2022; Jin, Liu, et al., 2016). Enzymatic hydrolysis is a mild process that does not damage the amino acid composition compared with acidic hydrolysis, which could break down amino acids to toxic substances, such as 3‐chloropropane‐1,2‐diol with carcinogenic effects (Adler‐Nissen, 1979, 1984; Lee & Khor, 2015; Nikoo et al., 2022; Wong et al., 2020). Thus, enzymatic hydrolysis is widely used in food industry as a controllable process for limited hydrolysis of proteins (Rezvankhah et al., 2021a, 2021b; Rezvankhah, Yarmand, et al., 2022).
The activity of α‐glucosidase and α‐amylase is associated with diabetes mellitus (Chandrasekaran & Gonzalez de Mejia, 2022; de Matos et al., 2022; Jiang et al., 2018; Qiao et al., 2020). These enzymes, secreted by the pancreas, break down dietary disaccharides and polysaccharides (Fadimu, Gill, et al., 2022; Karimi et al., 2020). The resulting glucose is absorbed at a higher rate into the blood, resulting in increased blood glucose level (Chandrasekaran & Gonzalez de Mejia, 2022; Tacias‐Pascacio et al., 2020). Moreover, hypertension is associated with angiotensin I‐converting enzyme (ACE), an enzyme of the renin‐angiotensin system pathway and important target of antihypertensive agents (Gharibzahedi & Smith, 2021; Guo et al., 2020; Ozón et al., 2022; Wang et al., 2019). Plant protein hydrolysates have shown strong inhibitory activities against α‐glucosidase, α‐amylase, and ACE toward the prevention and management of diabetes and hypertension (Karimi et al., 2020, 2021; Liu et al., 2020; Qiao et al., 2020). These inhibitory activities have been attributed to the presence of specific peptides, predominantly those composed of highly hydrophobic amino acids, released by commercial proteases (Das et al., 2022). The interaction between these peptides and amino acid residues at the active site of enzymes leads to inhibition of enzymatic activity (Quintero‐Soto et al., 2021). Strong ACE‐inhibitory activity has been reported for lentil protein hydrolysates obtained from sequential hydrolysis with Alcalase and Flavourzyme (Rezvankhah et al., 2021a, 2021b). In addition, cross‐linked lentil protein hydrolysates showed improved ACE‐inhibitory activity (Rezvankhah, Emam‐Djomeh, et al., 2022; Rezvankhah, Yarmand, et al., 2022).
Corn gluten meal (CGM) is a major by‐product of the corn wet milling process (Liu et al., 2020). It contains $62\%$–$71\%$ protein with zein as the prominent protein fraction, accounting for $68\%$ of the total protein, and glutelin as the residual part (~$28\%$ of zein weight) (Hu et al., 2022; Ren et al., 2018). Zein limits the application of CGM in various foods due to its poor water solubility (Yang et al., 2007). Furthermore, CGM contains several hydrophobic amino acid residues, which are buried inside the protein structure (Shen et al., 2020). Corn gluten meal is deficient in lysine and tryptophan, limiting its use in human nutrition (Zhu et al., 2019). Some studies have reported that enzymatic modification of CGM improved its solubility and bioavailability (Jiang et al., 2020; Jin, Liu, et al., 2016; Jin, Ma, et al., 2016; Liu et al., 2015). Corn protein hydrolysates (CPH) consist of small peptides with different molecular weight (MW) profiles, including di‐ and tripeptides, which can be effectively absorbed into blood circulation (Jin, Liu, et al., 2016; Jin, Ma, et al., 2016). Antioxidant and ACE‐inhibitory activities of CPH have been reported (Li et al., 2019; Ren et al., 2018; Wang et al., 2020; Yang et al., 2007). When subjected to gastrointestinal digestion, CPH showed $12.9\%$ increased antioxidant activity while retaining $77.5\%$ of peptides compared with the undigested hydrolysates (Ren et al., 2018). Moreover, Yang et al. [ 2007] reported that peptide Ala‐Tyr in Alcalase‐hydrolyzed CGM at 50 mg/kg body weight decreased the systolic blood pressure of rats by 9.5 mmHg at 2 h after oral administration. Furthermore, corn germ protein hydrolysates (CGPH) and associated peptidic fraction (F1) with MW <2 kDa showed higher radical scavenging and α‐glucosidase inhibitory activities than F2 fraction with MW of 2–10 kDa (Karimi et al., 2020). These inhibitory activities can be attributed to the different amino acid sequences, which determine the interactions with the active site residues of the enzyme (Quintero‐Soto et al., 2021).
Soy proteins (SPs) are one of the most utilized plant proteins in foods due to their nutritional quality, availability, and affordability compared with other sources of plant proteins (Xu et al., 2021). Native SPs are composed of a mixture of globulins and albumins (Tian et al., 2020). Ninety percent of the proteins are storage proteins with a globular structure consisting mainly of 7S (β‐conglycinin) and 11S (glycinin) globulins (Chen et al., 2011a, 2011b). Soy proteins have been considered for their emulsifying activity and gelling potential compared with other plant proteins (Bessada et al., 2019). However, native SP has compact globular structures, leading to low molecular flexibility, relatively low emulsifying properties, and antioxidant activities compared with its modified states (Zhang et al., 2021). Moreover, dietary SPs have shown antidiabetic activity in humans, indicating potential involvement of α‐amylase and α‐glucosidase inhibition (Das et al., 2022; Wang et al., 2019; Zhang et al., 2021; Zhao et al., 2021). Enzymatic hydrolysis of SP has been shown to increase solubility, antioxidant and α‐glucosidase inhibitory, and antihypertensive activities of SPH (Jiang et al., 2018; Tian et al., 2020; Wang et al., 2019). α‐Glucosidase inhibitory activity of SPH was reported to be higher than that of flaxseed, rapeseed, sunflower, and sesame protein hydrolysates (Han et al., 2021). Furthermore, novel peptides IY, YVVF, LVF, WMY, LVLL, and FF were identified from ACE‐inhibiting Alcalase‐derived SPH (Xu et al., 2021). The high hydrophobicity scores of the peptides might be the main contributor to the activity of SPH. The C‐terminal hydrophobic residues showed important interactions that may have contributed to ACE inhibition (Xu et al., 2021).
A combination of plant protein hydrolysates can compensate for the deficiency of individual hydrolysates (Akharume et al., 2021). For instance, CGM has a low amount of lysine but is rich in methionine and cysteine and hydrophobic amino acids (Hu et al., 2022). Conversely, SP has a high content of lysine but limited sulfur‐containing and hydrophobic amino acids (Tian et al., 2020). The combination of hydrolysates not only improves the deficiencies and balances the amino acid composition but also can augment biological (i.e., antioxidant, antidiabetic, and antihypertensive activities), and functional properties such as emulsifying activity. Since CGM and CPH demonstrate weaker ACE inhibitory activity than SP and SPH, it is postulated that combinations of SPH and CPH can produce improved hypertension property (Xu et al., 2021).
Hence, the objective of this study was to investigate the influence of combinations of CGM and SP hydrolysates on the bioactivities, including in vitro antidiabetic, antihypertensive and antioxidant activities, and functional properties. There is a dearth of studies on the combination of complementary bioactive hydrolysates; thus, the mixed protein hydrolysates can be explored as novel plant‐based ingredients with augmented antioxidant, antihypertensive, antidiabetic, and techno‐functional properties.
## Materials
Soy protein isolate (SPI, ~$90\%$ protein) and CGM (~$62\%$ protein) were supplied by Shansong Industrial Chinese Co. Ltd. and a grain processing refinery, Golshahd Co. Ltd., respectively. Alcalase 2.4 L from Bacillus licheniformis, with the activity of 2.4 Anson Units (AU)/g, and a density of 1.18 g/ml was purchased from Novozymes. 2,2 Diphenyl‐1‐picrylhydrazyl (DPPH), 2,2′‐azino‐bis (3‐ethylbenzthiazoline‐6‐sulphonic acid) diammonium salt (ABTS), 4‐nitrophenyl α‐d‐glucopyranoside (PNPG), porcine pancreatic α‐amylase, rat intestinal α‐glucosidase, ACE (5 UN), hippuryl‐his‐leu (HHL), and ammonium salt of 1‐anilino‐8‐naphtalene‐sulphonic acid (ANS) were purchased from Sigma‐Aldrich. Soluble starch ACS reagent was purchased from Merck.
## Enzymatic hydrolysis of proteins
Protein solutions of SPI (~$90\%$ protein) and CGM (~$62\%$ protein) were prepared at $5\%$ w/w. Soy protein isolate solution was heated at 90°C for 15 min to unfold the protein while the CGM solution was heated at 100°C for 30 min due to its low water solubility, thus the intense thermal treatment for denaturation (Tian et al., 2020; Yang et al., 2007). Enzymatic hydrolysis using Alcalase was conducted at temperature 60°C, pH 8.0, and enzyme/substrate ratio of $2.5\%$ w/w. This condition was optimized in preliminary studies. Hydrolysis time was considered based on DH reaching $15\%$; thus, 90 and 210 min were obtained for hydrolysis of SPI and CGM, respectively. Corn gluten meal have zein and glutelin as the main proteins (heat resistant) and thus required longer hydrolysis time to reach DH of $15\%$ (Yang et al., 2007). After the reaction, CPH and SPH solutions were heated at 95°C for 10 min to terminate enzymatic activity. Then, the protein/peptide solutions were centrifuged at 15,000 g for 10 min and the obtained supernatants (rich in soluble peptides) were collected and adjusted to pH 7.0 using 1 M HCl, and spray‐dried (DORSA tech) through a drying air of 180°C (the exhausting temperature of 75–80°C) and an air flow of 0.3–0.4 MPa (Akbarbaglu et al., 2019; Rezvankhah, Emam‐Djomeh, et al., 2022; Rezvankhah, Yarmand, et al., 2022; Sarabandi et al., 2019). The powdered hydrolysates were stored at −18°C until the next experiments.
## Determination of the degree of hydrolysis (DH)
Degree of hydrolysis was determined using the pH‐stat protocol reported by Adler‐Nissen [1986] and calculated using the equation: [1] DH%=B×Nb×1α×1Mp×1htot where B is the volume (ml) of NaOH needed to maintain the pH constant; N b is the normality of the consumed base; M p is the mass of protein in CP and SP; h tot is the total number of peptide bonds in the protein substrates (considered 9.2 for CP and 7.75 for SP) (Adler‐Nissen, 1979, 1986; Jin et al., 2015; Tian et al., 2020; Xu et al., 2021); and α is the amount of α‐NH2 released during the proteolysis reaction.
## Preparation of mixture hydrolysates
CPH and SPH were mixed homogeneously using a mixer (Moulinex, LM238125) at three proportions, including 30:70, 50:50, and 70:30 (% w/w), which were referred to as SPH30:CPH70, SPH70:CPH30, and SPH50:CPH50, respectively. Unhydrolyzed proteins (CP and SP) were used as control in all analyses.
## Amino acid composition
The amino acid profiles were determined using reversed‐phase high‐performance liquid chromatography (RP‐HPLC, Agilent 1100 HPLC; Agilent Ltd.), as described by Liu et al. [ 2012]. First, the samples were hydrolyzed in the glass tubes using 6 M HCl at 120°C for 12 h. Thereafter, the digests were filtered through 0.22 μm pore size filter. The separation was performed using a Zorbax analytical column (C18, 4 × 250 mm, 5 μm particle size; Agilent) at the temperature of 40°C with a UV detector spectra monitored at 338 nm. The elution of column with the flow rate of 1 ml/min was conducted with mobile phases comprising 7.40 mmol/L of sodium acetate/triethylamine/tetrahydrofuran (400:0.10:2, v/v/v), set at pH 7.1 using acetic acid and 7.40 mmol/L of sodium acetate/methanol/acetonitrile (1.5:2.5:2.5, v/v/v), set at pH 7.1. A standard solution comprising of 17 amino acids was used as an external standard.
Amino acid composition of the hydrolysates is presented in Table 1. The RP‐HPLC amino acid profile of CP showed higher hydrophobic amino acid contents, while the SP exhibited higher hydrophilic amino acid contents (Table 1). Similar findings have been reported in earlier investigations (Reyes Jara et al., 2018; Zhou et al., 2012). Enzymatic hydrolysis by Alcalase significantly changed the amino acid profiles of both proteins (Fadimu et al., 2021; Rezvankhah, Emam‐Djomeh, et al., 2022; Rezvankhah, Yarmand, et al., 2022). CPH had higher content of hydrophilic amino acids, while SPH showed higher content of hydrophobic amino acids than their respective native proteins (CP and SP) (Table 1).
**TABLE 1**
| Amino acid composition (g/100 g protein) | CP | CPH | SP | SPH | SPH30:CPH70 | SPH70:CPH30 | SPH50:CPH50 |
| --- | --- | --- | --- | --- | --- | --- | --- |
| hydrophobicity | hydrophobicity | hydrophobicity | hydrophobicity | hydrophobicity | hydrophobicity | hydrophobicity | hydrophobicity |
| Aspartic acid | 5.49 | 6.07 | 10.93 | 12.36 | 7.92 | 10.52 | 10.08 |
| Glutamic acid | 21.99 | 24.53 | 16.36 | 27.07 | 25.28 | 26.16 | 25.27 |
| Serine | 4.35 | 4.82 | 4.31 | 4.82 | 4.95 | 5.0 | 4.75 |
| Glycine | 2.42 | 2.73 | 19.63 | 3.79 | 3.11 | 3.33 | 3.06 |
| Histidine | 2.09 | 2.24 | 2.01 | 2.45 | 2.10 | 2.54 | 2.43 |
| Arginine | 8.96 | 8.51 | 5.54 | 6.50 | 7.45 | 7.10 | 7.47 |
| Threonine | 2.93 | 3.06 | 3.59 | 3.52 | 3.21 | 3.47 | 3.34 |
| Cysteine | 2.71 | 2.19 | 0.64 | 0.74 | 1.56 | 0.99 | 1.53 |
| Tyrosine | 4.38 | 4.32 | 2.05 | 2.15 | 3.22 | 3.04 | 3.21 |
| Lysine | 1.32 | 1.80 | 7.00 | 8.19 | 3.72 | 6.09 | 5.03 |
| | 56.64 | 60.27 | 72.06 | 71.38 | 62.52 | 68.24 | 66.17 |
| Hydrophobic | Hydrophobic | Hydrophobic | Hydrophobic | Hydrophobic | Hydrophobic | Hydrophobic | Hydrophobic |
| Alanine | 8.31 | 9.10 | 5.28 | 5.21 | 7.86 | 6.68 | 7.04 |
| Proline | 1.49 | 1.65 | 2.57 | 3.11 | 2.09 | 2.60 | 2.76 |
| Valine | 4.42 | 4.24 | 4.77 | 4.43 | 4.11 | 4.27 | 4.31 |
| Methionine | 3.88 | 3.51 | 1.20 | 1.08 | 2.58 | 1.83 | 2.59 |
| Isoleucine | 3.38 | 3.03 | 2.99 | 3.49 | 3.19 | 3.35 | 3.16 |
| Leucine | 16.09 | 13.83 | 8.73 | 8.23 | 11.92 | 9.77 | 10.28 |
| Phenylalanine | 5.78 | 4.38 | 2.40 | 2.87 | 5.72 | 3.23 | 3.69 |
| Tryptophan | ‐ | ‐ | ‐ | ‐ | ‐ | ‐ | ‐ |
| | 43.35 | 39.74 | 27.94 | 28.42 | 37.47 | 31.73 | 33.83 |
The variation in amino acid profiles could be related to the unfolding of the protein structures and exposure of the hydrophobic segments during enzymatic hydrolysis (Jin et al., 2015; Liu et al., 2012). This variation could also be due to the separation of some unhydrolyzed polypeptides during the hydrolysis and removal by centrifugation (Fadimu et al., 2021; Rezvankhah, Emam‐Djomeh, et al., 2022; Rezvankhah, Yarmand, et al., 2022). Furthermore, the amino acid composition varied among all mixed hydrolysates including SPH70:CPH30, SPH50:CPH50, and SPH30:CPH70. Indeed, the contents of hydrophilic and hydrophobic amino acids of the mixed hydrolysates were determined by the dominant hydrolysate portion. CPH was the major constituent in SPH30:CPH70, and the addition of SPH increased the quantity of hydrophilic amino acids when compared to CPH alone (Table 1). Previous studies have reported high hydrophobic and sulfur‐containing amino acid contents for CGM (CP) and CPH, and high hydrophilic amino acid and lysine contents for SP and SPH (Jin, Liu, et al., 2016; Jin, Ma, et al., 2016; Reyes Jara et al., 2018).
## Surface hydrophobicity (
H
0
)
The protein and hydrolysate samples were investigated for the H 0 using 1‐anilinonaphthalene‐8‐sulfonic (ANS) according to the method of He et al. [ 2021]. The samples were diluted to 0.01–0.02 mg/ml in 10 mM phosphate buffer (pH 7). The fluorescence intensity was measured at a wavelength of 390 nm (excitation) and 470 nm (emission) using a fluorescence spectrophotometer (LS‐55, Perkin Elmer). The slope of the plot of fluorescence vs. concentration of samples was expressed as the surface hydrophobicity (H 0).
## SDS‐polyacrylamide gel electrophoresis (SDS‐PAGE)
SDS‐polyacrylamide gel electrophoresis was used to estimate the MW profile of CP, SP, and their hydrolysates following the method of Rezvankhah et al. ( 2021b). Briefly, a sample solution (5 mg/ml) of proteins and respective hydrolysates was mixed with an equal amount of Laemmli sample buffer (960 μl of 66 mM Tris–HCl, pH 6.9, $27.3\%$ w/v glycerol, $2.2\%$ SDS, $0.01\%$ bromophenol blue). Then, the prepared samples were combined with 2‐mercaptoethanol and heated for denaturation at 95°C for 5 min before the electrophoresis. The concentration of $12\%$ Mini‐Protean™ precast gels (Bio‐Rad) was used to run the electrophoresis. Thereafter, 10 μl of cooled samples was loaded on the gels, and then subjected to a constant voltage of 150 V. Additionally, a marker with MW standards (Bio‐Rad Broad Range Marker) was run alongside the samples. When the process finished, the gels were stained with $0.1\%$ Coomassie Brilliant Blue R‐250 in a mixture of $10\%$ acetic acid and $40\%$ methanol for 2 h. The protein/peptide bands were visualized by discoloring the gels using a mixture of $40\%$ methanol and $10\%$ acetic acid solutions.
## MW distribution
Gel permeation chromatography (GPC) was applied to determine MW distribution of the CP, SP, and hydrolysate samples following the method of Rezvankhah et al. ( 2021b). The Waters Breeze HPLC system (Waters Corporation) equipped with a Waters UV detector and Superdex Peptide HR column (30 cm × 10 mm and 13–15 μm particle size) was used to evaluate the protein/peptide sizes. The analytes were dissolved in ultrapure water and stirred for 30 min at 25°C. The solution was centrifuged at 12,000 g for 10 min, and the supernatant was filtered through a 0.22 μm membrane filter. The injection volume of sample was 50 μl, and the column was eluted with 20 mM phosphate buffer containing 0.15 M NaCl (pH 7) at a flow rate of 0.5 ml/min. The spectra were monitored at a wavelength of 210 nm. Standard compounds of known MW including reduced glutathione (300 Da), glutathione disulfide (600 Da), cyanocobalamin (1355 Da), aprotinin (6500 Da), and cytochrome C (12,500 Da) were used to prepare a standard curve, which was used to determine the MW.
## DPPH radical scavenging activity
Antioxidant activity was determined according to the method of Zheng et al. [ 2015]. 2,2 Diphenyl‐1‐picrylhydrazyl solution at 0.2 mM in $95\%$ ethanol and the protein/peptide solution at 7 mg/ml were prepared. Then, 2 ml of DPPH solution was combined with 2 ml of samples and stored for 30 min in the dark. The absorbance of the mixtures was read at 517 nm using a spectrophotometer. To compare the antioxidant activity of the analytes, ascorbic acid (0.01 mg/ml) was used as a positive control. The RSA% was calculated using the equation: [2] RSA%=AC−ASAC−AB×100 where the absorbance values were for control (A C), sample (A S), and blank (A B).
## ABTS radical cation scavenging activity
The ABTS·+ scavenging activity of hydrolysates was evaluated according to the protocol reported by Amini Sarteshnizi et al. [ 2021]. The ABTS solution (940 μl) was combined with 60 μl of samples (7 mg/ml) and vigorously shaken, and then incubated at 25°C for 10 min in the dark. The absorbance was read at 734 nm using a spectrophotometer, and ascorbic acid (0.01 mg/ml) was used as a positive control.
## Hydroxyl radical scavenging activity (RSA)
The hydroxyl RSA was evaluated using the protocol reported by Zheng et al. [ 2015]. Two milliliter of samples (7 mg/ml), 2 ml of FeSO4 (6 mM), and 2 ml of H2O2 (6 mM) were thoroughly mixed and kept for 10 min at room temperature. Then, 2 ml of salicylic acid (6 mM) was added and incubated for 30 min. The absorbance was then read at 510 nm (A S). Distilled water was used as the blank (instead of salicylic acid solution) and the control (instead of sample solution). Ascorbic acid (0.01 mg/ml) was used as a positive control. The RSA was calculated using equation 2.
## ACE inhibition assay
The potential antihypertensive activity of CP, SP, and hydrolysates was assessed by the in vitro inhibition of angiotensin I‐converting enzyme (ACE) based on the method of Boye et al. [ 2010]. ACE inhibition (%) was computed by the equation: [3] Inhibitory activity%=AC−ASAC−AB×100 where the absorbance values were for control (A C), sample (A S), and blank (A B). Also, the IC50 value, the concentration of sample that inhibited $50\%$ of ACE activity, was determined using sample concentrations of 0.1–2 mg/ml.
## α‐Glucosidase inhibition assay
The inhibitory activity of CP, SP, and their hydrolysates against rat intestinal α‐glucosidase was assessed following the method of Karimi et al. [ 2020]. Briefly, the enzyme was extracted from acetone powder from rat intestine, and the obtained solution was diluted to 90 mU/ml. Then, 150 μl of different concentrations of the samples (10–500 μg/ml) was combined with 250 μl of α‐glucosidase and incubated at 37°C for 10 min. To carry on the reaction, 100 μl of PNPG solution (5 mM) was added and the mixture was incubated at 37°C for 30 min while scanning the absorbance at 405 nm every 2 min. Instead of analyte solution, phosphate buffer was utilized as a control. To compare the inhibitory activity, acarbose (0.5 mg/ml), a synthetic antidiabetic compound, was used as a positive control. The enzyme inhibition was calculated using the equation: [4] Inhibtion ofα−glucosidase%=AC−ASAC×100 where the absorbance values were for control (A C) and sample (A S). Sample concentrations of 10–500 μg/ml were used to determine the IC50 values.
## α‐Amylase inhibition assay
The α‐amylase inhibitory activity of CP, SP, and hydrolysates was determined using the method reported by Rahimi et al. [ 2022]. Briefly, 100 μl of different concentrations of the samples (10–500 μg/ml) was combined with 120 μl of α‐amylase solution (0.6 U/ml) and incubated at 37°C for 5 min. Then, 120 μl of $0.5\%$ (w/v) starch solution was added. The enzyme activity was terminated by heating the reaction mixture at 100°C for 10 min followed by cooling to ambient temperature. The undigested starch was separated by centrifugation at 15,000 g for 2 min. Then, 20 μl of the supernatant was mixed with 1 ml of PAHBAH and the solution was heated to 70°C for 10 min. The sample solutions were cooled, and absorbance values were read at 410 nm. The inhibitory activity was determined using the following equation: [5] Inhibition ofα−amylase%=1−AS−ABAC×100 where the absorbance values were for sample (A S), blank (A B, phosphate buffer, enzyme, sample), and control (A C, starch, buffer, enzyme). Furthermore, IC50 values were determined as previously described. Acarbose, at its IC50 value (0.125 mg/ml), was used as a positive control.
## Solubility
The solubility of CP, SP, and hydrolysates was assessed by the method of Fathollahy et al. [ 2021] with slight modifications. Briefly, 10 mg/ml of samples at three pH values (4.0, 7.0, and 9.0) was centrifuged at 8000 g for 20 min. The supernatants were taken for protein determination based on the Bradford protocol (Bradford, 1976). To prepare the standard curve and calculate the protein content, bovine serum albumin was used as a reference protein. The solubility was determined by the following equation: [6] Solubility%=Protein content in the supernatantTotal protein content in the sample×100
## Emulsifying properties
Two emulsifying properties including emulsifying activity index (EAI, m2/g) and emulsifying stability index (ESI, min) were assessed by the method of Rezvankhah et al. ( 2021b). Samples (10 mg/ml) were prepared and combined with 1 ml of sunflower oil and homogenized at 19,000 rpm for 1 min using a laboratory‐scale homogenizer (IKA, T25). The emulsion (100 μl) was taken from the container bottom immediately after production to determine EAI. Also, ESI was determined by taking 100 μl of the emulsion from the container bottom after 10 min. The aliquots were combined with 5 ml of SDS ($0.1\%$), and the absorbance of the diluted solutions was read at 500 nm using a UV–Vis spectrophotometer. EAI and ESI were computed using the equations: [7] EAIm2g=22.303A0DFIθC [8] ESImin=A0∆A∆t where A 0 is the absorbance of diluted emulsion at 500 nm immediately after homogenization, DF is the dilution factor [50], I is the path length of the cuvette (m), θ is the oil volume fraction (0.25), C is the protein concentration in the aqueous phase (g/m3), ∆A=A0−A10, and ∆$t = 10$min.
## Foaming properties
Two foaming properties including the foaming capacity (FC) and foaming stability (FS) of the CP, SP, and hydrolysates were assessed following the procedure reported by Rezvankhah, Emam‐Djomeh, et al. [ 2022] and Rezvankhah, Yarmand, et al. [ 2022]. The 10 mg/ml sample solutions in a 50 ml measuring cylinder was whipped at 19,000 rpm for 2 min using a laboratory‐scale homogenizer (IKA, T25). The total volume (ml) of the initial foam was determined. Also, the foam volume was recorded after storage time of 30 min at room temperature. Foaming capacity and FS were calculated using the following equations: [9] FC%=B−AA×100 [10] FS%=C−AA×100 where the volume before whipping (ml), the volume immediately after whipping (ml), and the volume after standing for 30 min (ml) are denoted by A, B, and C, respectively.
## Statistical analysis
The experimental data were reported as means ± standard deviation. One‐way analysis of variance (ANOVA) was used to analyze the obtained data. The Duncan test was applied to evaluate the comparison of mean difference using the SPSS software (version 26, IBM software).
## Surface hydrophobicity
Surface hydrophobicity (H 0) has an important effect on the macromolecular structural stability, surface, and biological properties of proteins (Rezvankhah, Emam‐Djomeh, et al., 2022; Rezvankhah, Yarmand, et al., 2022; Wang et al., 2020). As shown in Figure 1, enzymatic hydrolysis of CP and SP significantly increased the H 0 values, as previously reported by others (Zheng et al., 2015). The noncovalent, particularly the hydrophobic interactions, and disulfide bonds (SS) are abundantly present in CP (Liu et al., 2015; Wang et al., 2020). Although not considered the prevalent driving force for aggregation, hydrophobic interactions influence the aggregation tendency (Zheng et al., 2015). For CPH, the fluorescence intensity with ANS remarkably increased, indicating a higher H 0 value than CP. The hydrophobic patches are buried inside the zein and glutelin structures (Liu et al., 2015). When the proteins are hydrolyzed, the hydrophobic segments are exposed to the surface. Albeit, it did not lead to aggregation. The insoluble aggregates may have been separated by centrifugation, while the soluble aggregates were maintained (Zheng et al., 2015). According to a previous study, CPH had an emulsion‐like appearance, which indicates that hydrolysis of CP not only increased its surface hydrophobicity but also decreased the MW and disulfide bonds of the protein, thereby transforming the insoluble aggregates into soluble aggregates (Zheng et al., 2015). Hydrolysis also increased the H 0 of SP, but this increase in SPH was remarkably lower than CPH; this may be related to the amino acid composition of SP and CP or their corresponding hydrolysates (Rezvankhah, Emam‐Djomeh, et al., 2022; Rezvankhah, Yarmand, et al., 2022; Wang et al., 2016; Zheng et al., 2015) (Table 1). SPH had lower hydrophobic amino acid composition (28.42 g/100 g) than CPH (39.74 g/100 g), and this is likely due to the dominant hydrophobic amino acid portion in CP. As shown in Figure 1, the combination of SPH with CPH, depending on the dominant part, resulted in different H 0 values. Therefore, the order of H 0 values for combined hydrolysates of SP and CP was SPH30:CPH70 > SPH50:CPH50 > SPH70:CPH30. Therefore, the higher the content of CPH in the mixture, the higher the H 0 value achieved (Figure 1).
**FIGURE 1:** *Surface hydrophobicity values of unhydrolyzed proteins, hydrolysates, and hydrolysate mixtures. The data marked with different letters are significantly different (p < .05). CP and SP indicate unhydrolyzed protein of corn and soy, respectively. CPH, SPH, and different mixing ratios indicate hydrolysates of corn and soy, mixtures, respectively.*
## Molecular weight profile
Approximate MW of CP, SP, CPH, SPH, SPH30:CPH70, SPH70:CPH30, and SPH50:CPH50 was determined using SDS‐PAGE (Figure 2a). The intense bands detected for CP ranged from 80 to 200 kDa. A wide range of MW (250 Da to 250 kDa) have been reported for corn protein samples (He et al., 2021; Ortiz‐Martinez et al., 2017). For SP, the hydrolysates had MW of 40–50 and 70–200 kDa, the former indicating the presence of smaller polypeptides in SP than in CP. The electrophoretic pattern of SP showed β‐conglycinin subunits α′ (~72 kDa), α (~68 kDa), and β (~53 kDa), two subunits of glycinin, the acidic subunit (“A”) at 29–33 kDa and the basic subunit (“B”) at around 18–22 kDa (Meinlschmidt et al., 2016; Zhang et al., 2021). Hydrolysis of proteins significantly altered the bands that correspond to the smaller peptides produced and/or larger peptide cleavage by Alcalase (Jin, Liu, et al., 2016; Jin, Ma, et al., 2016; Liu et al., 2020; Xu et al., 2021). For CPH, bands with MW of 30–40, 50–55, 70–80, and 150–170 kDa were observed. CP did not show bands with MW <80 kDa, while the SDS‐PAGE pattern of CPH revealed polypeptides with MW ~30 kDa that was assigned to α‐zein, glutelin, and dimers (He et al., 2021; Liu et al., 2012). Indeed, zein, due to its low solubility, did not have a band at lower MW but proteolysis resulted in the emergence of peptides with higher solubility (Ortiz‐Martinez et al., 2017). For SPH, most of the SP bands at 70–200 kDa disappeared, and new bands were formed at 10–15, 15–20, 30, 40–50, and 50–60 kDa. The β‐conglycinin subunits α′ (~72 kDa) and α (~68 kDa), which are known as major soy allergens, completely disappeared after SP hydrolysis, as previously reported by Meinlschmidt et al. [ 2016]. As shown in Figure 2a, the dominant portion (SPH or CPH) of the combination of the two hydrolysates determined the MW profiles of the mixture. For instance, the higher SPH ratio in SPH70:CPH30 resulted in the emergence of lanes mostly similar to the lanes detected for SPH, while the higher CPH caused the formation of bands specific to CPH (Figure 2a). The combination of SPH and CPH is hypothesized to present stronger biological activities and functional properties than their respective hydrolysates.
**FIGURE 2:** *SDS‐PAGE patterns (a) and GPC MW distribution (b) of unhydrolyzed proteins, hydrolysates and hydrolysate mixtures. CP and SP indicate unhydrolyzed protein of corn and soy, respectively. CPH, SPH, and different mixing ratios indicate hydrolysates of corn and soy, mixtures, respectively.*
It has been reported that lower MW peptides are not visible on SDS‐PAGE gels, associating with heating effects on the protein conformation (Fadimu, Gill, et al., 2022; Rezvankhah, Emam‐Djomeh, et al., 2022; Rezvankhah, Yarmand, et al., 2022). Heating is performed for various aims (denaturation/unfolding of the protein structures and terminating of enzyme) during the hydrolysis stage or preparation of analytes for SDS‐PAGE analysis. Most of the SDS‐PAGE gels have been designed to determine molecules/peptides with MW above 10 kDa; thus, visualizing molecules with lower MW is possible using techniques such as gel permeation chromatography (Fadimu, Gill, et al., 2022; Rezvankhah, Emam‐Djomeh, et al., 2022; Rezvankhah, Yarmand, et al., 2022).
## Molecular weight distribution
Changes in MW that could not be detected by SDS‐PAGE, especially MW below 10 kDa, were determined by GPC, given the high potential of GPC for accurate determination of MW distribution (Fadimu, Farahnaky, et al., 2022; Fadimu, Gill, et al., 2022). Molecular weight distribution of the proteins and their hydrolysates is shown in Figure 2b. Peptide size is one of the most important factors that influence the functional properties, bioavailability, and bioactivities of peptides (Rezvankhah et al., 2021a; Rezvankhah, Emam‐Djomeh, et al., 2022; Rezvankhah, Yarmand, et al., 2022). The chromatogram of CP showed short and sharp peaks assigned to MW 6683 and 3655 Da, respectively. The chromatogram of SP indicated short and sharp peaks with MW of 2720 and 1052 Da, respectively. These results as similar to previous findings with SDS‐PAGE patterns showing that SP had smaller polypeptides than CP (Zhang et al., 2021). As shown in Figure 2b, enzymatic hydrolysis resulted in the generation of small peptides with CPH showing a sharp peak at MW of 1917 Da and a short peak at MW of <300 Da. According to Figure 2b, SPH showed a peak for peptides with MW of 439 Da as previously reported (Wang et al., 2019). The combination of SPH and CPH, depending on the dominant portion, also altered the MW of the mixed hydrolysates. On this basis, the chromatograms of the SPH30:CPH70, SPH50:CPH50, and SPH70:CPH 30 showed sharp peaks at 1377, 1023, 486 Da, respectively (Figure 2b), indicating that the higher the SPH amount incorporated, the lower the MWs of hydrolysate mixture obtained.
## Antioxidant activity
As shown in Figure 3a, among the proteins and hydrolysates, CPH ($46.25\%$), SPH30:CPH70 ($46.70\%$), and SPH50:CPH50 ($47.30\%$) showed the highest DPPH RSA with no significant difference ($p \leq .05$), followed by SPH70:CPH30 ($33.50\%$), CP ($19.30\%$), SPH ($12.40\%$), and SP ($9.30\%$). Ascorbic acid, however, exhibited higher DPPH RSA at a much lower concentration than the hydrolysates. CP and its hydrolysates had high content of hydrophobic amino acids, which may be related to their higher reactivity with DPPH radicals (Jin et al., 2015; Karimi et al., 2020, 2021). Conversely, SP and SPH, which have high content of hydrophilic amino acids, showed lower antioxidant activity (Figure 3a). Enzymatic hydrolysis significantly increased the antioxidant activity of the proteins, and this is associated with the liberation of medium‐ and small‐sized peptides with exposed hydrophobic and reactive groups with antioxidant power (Zhou et al., 2017). CPH with higher hydrophobicity (Figure 1 and Table 1) had the strongest DPPH radical scavenging activity. Among all samples, SPH30:CPH70 and SPH50:CPH50 showed the strongest antioxidant activity.
**FIGURE 3:** *Antioxidant activity of unhydrolyzed proteins, hydrolysates, and hydrolysate mixtures. The small letters including (a), (b), and (c) illustrate DPPH, ABTS, and hydroxyl radical scavenging activities, respectively. Ascorbic acid (0.01 mg/ml) was used as a positive control. The data marked with different letters are significantly different (p < .05). CP and SP indicate unhydrolyzed protein of corn and soy, respectively. CPH, SPH, and different mixing ratios indicate hydrolysates of corn and soy, mixtures, respectively.*
A previous study on chickpea protein hydrolysates obtained with Alcalase showed that a content of over $50\%$ hydrophobic amino acids resulted in high DPPH RSA (Quintero‐Soto et al., 2021).
ABTS and hydroxyl RSA gave different results. According to Figure 3b, all samples had ABTS RSA higher than $70\%$. The highest antioxidant power was obtained for SPH ($95.01\%$), which was similar to the result of ascorbic acid ($97.40\%$), followed by SPH70:CPH30 ($89.70\%$), SP ($88.03\%$), SPH50:CPH50 ($86.04\%$), SPH30:CPH70 ($83.22\%$), CP ($76.07\%$), and CPH ($74.75\%$). SP and SPH had higher reactivity with ABTS, while CP and CPH had lower reactivity, thus showing lower ABTS RSA. The production of peptides with MW <1 kDa remarkably influences the antioxidant power of hydrolysates (Tian et al., 2020). Also, SPH had higher hydrophilic amino acids that are known to have high interaction with the hydrophilic radical (ABTS) (Hu, Chen, et al., 2020; Hu, Dunmire, et al., 2020). ABTS radical has hydrophilic affinity, while DPPH radical has hydrophobic affinity; thus, SPH indicated higher ABTS RSA, while CPH exhibited higher DPPH RSA, similar to previous findings (Hu, Chen, et al., 2020; Rezvankhah, Yarmand, et al., 2022).
Based on the hydroxyl RSA results in Figure 3c, the lower the MW of the hydrolysates, the higher the antioxidant power obtained. SPH due to its higher hydrophilic amino acids and containing lower MW peptides (Figure 2a,b) exhibited higher hydroxyl RSA, while CP due to its higher hydrophobic amino acid profiles and larger peptides exhibited lower hydroxyl RSA (Rezvankhah, Emam‐Djomeh, et al., 2022; Rezvankhah, Yarmand, et al., 2022). MW of peptides can substantially affect their antioxidant activity. The small‐ and medium‐sized peptides have shown stronger antioxidant power due to their ability to interact with the radicals (Bu et al., 2020; Rezvankhah et al., 2021a; Singh et al., 2014; Tian et al., 2020; Zhao et al., 2021).
The order of hydroxyl RSA values was SPH ($82.30\%$) > SPH70:CPH30 ($77.40\%$) > SPH50:CPH50 ($75.20\%$) > SPH30:CPH70 ($73.40\%$) > CPH ($70.60\%$) > ascorbic acid ($54.20\%$) (as positive control) > SP ($44.60\%$) > CP ($40.50\%$), respectively. Taken together, the results suggest that SPH and CPH combined hydrolysate SPH70:CPH30 possessed the best antioxidant activities and, thus, have the potential to protect food or biological systems against oxidative damages. Zhang et al. [ 2021] reported potent hydroxyl RSA for SPH treated with ultrasound. Similar results were reported for mung bean protein hydrolysates (Liu et al., 2022). Also, CPH indicated potent hydroxyl RSA compared with nonhydrolyzed protein (Zheng et al., 2015).
## In vitro antihypertensive property
Results of the ACE inhibitory activity of the proteins and their hydrolysates are presented in Figure 4a. SPH exhibited the highest ACE inhibitory activity ($95.45\%$) similar to SPH70:CPH30 ($94.76\%$) with no significant difference, followed by SPH30:CPH70 ($89.65\%$), SPH50:CPH50 ($89.28\%$), and SP ($88.64\%$), and the lowest values were obtained for CP ($76.68\%$) and CPH ($74.56\%$). Also, IC50 values of 0.5, 0.25, 0.38, 0.15, 0.23, 0.18, 0.21 mg/ml were obtained for CP, SP, CPH, SPH, SPH30:CPH70, SPH70:CPH30, and SPH50:CPH50, respectively. It was observed that hydrolysis of SP increased the ACE inhibitory activity as previously reported (Wang et al., 2019; Xu et al., 2021). Although proteins (SP and CP) and their hydrolysates (SPH and CPH) exhibited high ACE‐inhibitory activity (higher than $70\%$), the combined SPH and CPH also showed strong ACE‐inhibitory activity. The hydrophobic amino acids positioned at the C‐terminal residues have been shown to contribute to ACE‐inhibitory activity of peptides (Ambigaipalan et al., 2015). CP demonstrated weaker ACE inhibitory activity because the active peptide is locked in within the protein primary structure. Therefore, the activity obtained for the intact protein (CP and SP) could be due to unknown molecules co‐isolated with the proteins. Liu et al. [ 2020] identified 12 peptides from active fractions of Alcalase‐hydrolysate (CPH) obtained from CGM with good ABTS radical scavenging and ACE inhibitory activities (Liu et al., 2020). Similar findings were reported for SPH (Xu et al., 2021). Hence, the combination of CPH and SPH led to alterations in amino acid profiles of the new mixed hydrolysates, which influenced the bioactivities.
**FIGURE 4:** *ACE (a), α‐glucosidase (b), and α‐amylase (c) inhibitory activities of unhydrolyzed proteins, hydrolysates and hydrolysate mixtures. The data marked with different letters are significantly different (p < .05). CP and SP indicate unhydrolyzed protein of corn and soy, respectively. CPH, SPH and different mixing ratios indicate hydrolysates of corn and soy, mixtures, respectively.*
Angiotensin‐converting enzyme such as other enzymes has binding sites that could interact with peptide inhibitors (Quintero‐Soto et al., 2021). Higher interactions and affinity of peptides and ACE, indicated by lower binding energy, often result in stronger inhibitory activity (Quintero‐Soto et al., 2021). The hydrophobic amino acids located on the enzyme active site allow the interaction of uncharged amino acids.
## In vitro antidiabetic properties
The in vitro α‐glucosidase and α‐amylase inhibitory activities of the samples are presented in Figure 4b,c. For α‐glucosidase inhibitory activity (Figure 4b), SPH showed the highest inhibition ($32.26\%$), followed by SPH70:CPH30 ($30.58\%$), SPH30:CPH70 ($27.89\%$), SPH50:CPH50 ($28.28\%$), and SP ($27.48\%$) with no significant difference, and CPH ($26.13\%$) and CP ($25.20\%$) had the lowest values. The sample activities were lower than the effect of acarbose ($50.76\%$). The IC50 values of 24.59, 13.51, 20.09, 5.65, 9.97, 7.15, 8.64, and 0.51 mg/ml were obtained for CP, SP, CPH, SPH, SPH30:CPH70, SPH70:CPH30, and SPH50:CPH50, and acarbose, respectively.
It was observed that the combination of SPH and CPH (SPH70:CPH30) led to mixed hydrolysates with higher α‐glucosidase inhibitory activity than CPH. Acarbose, a synthetic compound, at the same concentration exhibited much higher α‐glucosidase inhibitory activity than the crude hydrolysates. However, it has several side effects (Das et al., 2022), making the natural hydrolysates promising safer alternatives. The strongest α‐glucosidase activity of SPH could be associated with the higher content of smaller peptides produced at the same DH compared with CPH (Figure 2a,b). Also, the composition of amino acid residues influences the activity (Quintero‐Soto et al., 2021). The presence of the basic amino acids (lysine and arginine) at the end of the peptide chains and amino acids with hydroxyl groups (serine, threonine, and tyrosine) contributes to α‐glucosidase inhibition through the interaction with the active site of the enzyme (Karimi et al., 2020, 2021). The prevalent interactions are electrostatic and hydrogen bonds which lead to suppression of the enzyme activity (Rezvankhah, Emam‐Djomeh, et al., 2022; Rezvankhah, Yarmand, et al., 2022). Incorporation of SPH increased the α‐glucosidase inhibitory activity of CPH, likely due to augmentation of the amino acid composition.
As illustrated in Figure 4c, hydrolysis significantly increased the α‐amylase inhibitory activity of the samples. The highest α‐amylase inhibition was obtained for SPH ($58.21\%$), followed by SPH70:CPH30 ($55.88\%$), SPH50:CPH50 ($52.26\%$), SPH30:CPH70 ($49.57\%$), SP ($42.74\%$), CPH ($39.62\%$), and CP ($35.85\%$). The IC50 values were 3.51, 1.22, 1.69, 0.23, 0.6, 0.33, 0.39 mg/ml for CP, SP, CPH, SPH, SPH30:CPH70, SPH70:CPH30, and SPH50:CPH50, respectively. According to Figure 4c, SPH and mixed hydrolysate with high contribution of SPH (SPH70:CPH30) showed higher α‐amylase inhibition than CPH and mixed hydrolysate with high contribution of CPH (SPH30:CPH70).
Bioactive peptides can interact with the active site of enzymes to reduce and/or inhibit substrate binding. Moreover, bioactive peptides can bind the allosteric site of the enzyme. For instance, peptides can interact with calcium and chloride ion binding sites of enzymes to produce unstable conformations, thereby restricting enzyme‐substrate binding (Ngoh & Gan, 2016). Indeed, calcium ions participate in structure formation, functions, and regulation of the stability of α‐amylase (Admassu et al., 2018). Moreover, amino acid residues including glycine, leucine, serine, aspartic and glutamic acids, proline, phenylalanine, tryptophan, and tyrosine have been shown to bind the active site of the enzyme, thus increasing the potential to achieve inhibition (Karimi et al., 2020). SPH and SPH70:CPH30 exhibited higher potential in inhibiting α‐amylase than the other hydrolysates.
## Functional properties
Solubility, emulsifying, and foaming properties of the samples are provided in Table 2. The solubility of proteins and their hydrolysates was determined at pH 5, 7, and 9. Hydrolysis of CP and SP significantly improved the protein solubility. At pH 4, CP and SP were not soluble in water ($0\%$) due to the insolubility of CP, zero net charges of the proteins (CP and SP), or lack of electrostatic repulsions (Rezvankhah et al., 2020; Zheng et al., 2015). Conversely, CPH, SPH, and their combinations exhibited higher solubility ranging from $92\%$ to $96\%$. This result is related to the small‐sized peptides released during the hydrolysis process (Chen et al., 2011a, 2011b). At pH 7, CP and SP had $23.18\%$ and $34\%$ solubility in water, while CPH, SPH, and their combined hydrolysates exhibited higher solubility of $94\%$–$100\%$. At pH 9, CP and SP had $24.50\%$ and $96\%$ solubility, which is likely related to the expected higher positive charge on the protein molecules. However, CP, due to its inherent low solubility in water (for zein and glutelin), still had low solubility at pH 9. Hydrolysis led to the solubility of $96\%$ for CPH and $92\%$–$100\%$ for all hydrolysates.
**TABLE 2**
| Functional properties | Functional properties.1 | Functional properties.2 | Functional properties.3 | Functional properties.4 |
| --- | --- | --- | --- | --- |
| Sample | EAI (m2/g) | ESI (min) | FC (%) | FS (%) after 30 min |
| CP | 12.43 ± 0.65g | 07.60 ± 0.56g | 00.00 ± 0.00f | 00.00 ± 0.00f |
| SP | 41.82 ± 0.52c | 31.20 ± 0.41c | 81.25 ± 6.25b | 62.50 ± 3.50a |
| CPH | 28.48 ± 0.16f | 18.30 ± 0.42f | 25.00 ± 2.10e | 12.50 ± 1.30d |
| SPH | 50.94 ± 0.91b | 40.65 ± 0.91b | 102.5 ± 5.50a | 03.75 ± 1.25e |
| SPH30:CPH70 | 37.24 ± 0.48d | 30.30 ± 0.14d | 32.50 ± 1.25d | 17.50 ± 1.20c |
| SPH70:CPH30 | 53.54 ± 0.62a | 46.80 ± 0.70a | 40.62 ± 3.12c | 24.37 ± 0.62b |
| SPH50:CPH50 | 35.12 ± 0.87e | 28.75 ± 0.63e | 40.00 ± 1.25c | 26.25 ± 1.25b |
| Solubility (%) | Solubility (%) | Solubility (%) | Solubility (%) | Solubility (%) |
| Sample | pH = 4 | pH = 7 | pH = 9 | pH = 9 |
| CP | nsc | 23.18 ± 0.01e | 24.50 ± 1.10e | 24.50 ± 1.10e |
| SP | nsc | 34.00 ± 1.20d | 96.00 ± 1.21c | 96.00 ± 1.21c |
| CPH | 96.00 ± 2.30a | 100.0 ± 0.04a | 96.00 ± 1.11c | 96.00 ± 1.11c |
| SPH | 92.30 ± 0.40b | 94.00 ± 1.02c | 92.00 ± 1.13d | 92.00 ± 1.13d |
| SPH30:CPH70 | 94.00 ± 0.30a | 96.00 ± 1.20b | 96.00 ± 1.30c | 96.00 ± 1.30c |
| SPH70:CPH30 | 94.00 ± 0.30a | 96.00 ± 1.05b | 98.00 ± 1.21b | 98.00 ± 1.21b |
| SPH50:CPH50 | 94.00 ± 0.30a | 96.00 ± 1.10b | 100.0 ± 1.42a | 100.0 ± 1.42a |
The emulsifying properties (EAI) of CP and SP were 12.43 and 41.82 m2/g, respectively, at pH 7 while their hydrolysates CPH and SPH exhibited significantly higher values of 28.48 and 50.94 m2/g, respectively. SP and SPH specifically had higher EAI values than CP and CPH. The mixed hydrolysates including SPH30:CPH70, SPH70:CPH30, and SPH50:CPH50 had EAI values of 37.24, 53.54, 35.12 m2/g, respectively. Notably, the highest EAI obtained for SPH70:CPH30 supported the hypothesis that a combination of $70\%$ SPH with $30\%$ CPH produces a synergistic effect. The EAI is associated with the protein/peptide ability to reduce the interfacial tension, thus generating smaller droplets leading to high emulsion stability (Rezvankhah et al., 2020, 2021b; Rezvankhah, Emam‐Djomeh, et al., 2022; Rezvankhah, Yarmand, et al., 2022). A similar trend was observed for ESI (min). Emulsion stability of CP and SP (7.60 and 31.20 min, respectively) significantly increased after enzymatic hydrolysis, with CPH and SPH showing ESI values of 18.30 and 40.65 min, respectively. This result could be related to the exposed hydrophobic regions, which can keep the oil at the oil–water interface. DH up to $15\%$ improves both EAI and ESI of proteins due to an increase in water solubility and hydrophobic interactions (Wang et al., 2020). The mixed hydrolysates including SPH30:CPH70, SPH70:CPH30, and SPH50:CPH50 gave ESI values of 30.30, 46.80, and 28.75 min, respectively. Hence, the highest ESI value was obtained for SPH70:CPH30, indicating the strong surface‐active properties of SPH.
For foaming properties, CP showed FC ($0\%$) while SP had a FC of $81.25\%$. Hydrolysis significantly increased the FC, which reached $25\%$ and $102.5\%$ for CPH and SPH, respectively. The improvement of FC could be related to an increase in solubility of the hydrolysates. SPH30:CPH70, SPH70:CPH30, and SPH50:CPH50 had FC values of $32.50\%$, $40.62\%$, and $40\%$, respectively. Limited hydrolysis not only enhances protein solubility but also increases hydrophobic interactions, thus increasing FC (Jin et al., 2015). FS showed a slightly different result (Table 2). Hydrolysis significantly increased FS for only CPH ($12.5\%$). FS for SPH was significantly reduced, which might be related to the generation of small‐sized peptides with higher solubility, which influences the hydrophilic–hydrophobic balance. Interestingly, the mixed hydrolysates, including SPH30:CPH70, SPH70:SPH30, and SPH50:CPH50, showed higher FS than CPH and SPH. These results are likely due to the balanced peptide mixture in the combinations with optimum hydrophilic and hydrophobic interactions.
## CONCLUSION
Hydrolysis of CP and SP changed the hydrophilic and hydrophobic amino acid contents of the resulting hydrolysates. CPH had higher hydrophobic amino acid contents, while SPH had higher hydrophilic amino acid contents. Consequently, the hydrophilic–hydrophobic amino acid ratio of SPH70:CPH30, SPH30:CPH70, and SPH50:CPH50 depended on the dominant hydrolysates in the mixture. The combination of CPH with SPH led to increase in DPPH RSA of SPH, and the combination of SPH with CPH led to the increase in ABTS and hydroxyl RSA of CPH. A similar trend was observed for surface hydrophobicity where combined hydrolysates with CPH showed higher surface hydrophobicity than SPH alone. SPH and SPH70:CPH30 had lower MW, and higher ACE, α‐glucosidase and α‐amylase inhibitory activities than CPH and SPH30:CPH70. SPH and CPH showed improved solubility, emulsifying activity, and foaming capacity. Furthermore, SPH70:CPH30 exhibited better functional properties than the other hydrolysates and mixtures. Considering all the biological and functional properties, SPH70:CPH30 can be further explored as a promising multifunctional ingredient with antioxidant, antihypertensive, antidiabetic, and functional properties for utilization in developing novel functional food products.
## FUNDING INFORMATION
This study was supported by the office of vice chancellor for research at Tarbiat Modares University (Grant Number: 9830422005).
## CONFLICT OF INTEREST
None of the authors declare a conflict of interest.
## ETHICAL APPROVAL
This study does not involve any human or animal testing.
## DATA AVAILABILITY STATEMENT
The data used in this paper are available in case of reviewer or editor request.
## References
1. Adler‐Nissen J.. **Determination of the degree of hydrolysis of food protein hydrolysates by trinitrobenzenesulfonic acid**. *Journal of Agricultural and Food Chemistry* (1979) **27** 1256-1262. DOI: 10.1021/jf60226a042
2. Adler‐Nissen J.. **Control of the proteolytic reaction and of the level of bitterness in protein hydrolysis processes**. *Journal of Chemical Technology and Biotechnology* (1984) **34** 215-222. DOI: 10.1002/jctb.280340311
3. Adler‐Nissen J.. *Enzymic hydrolysis of food proteins* (1986)
4. Admassu H., Gasmalla M. A. A., Yang R., Zhao W.. **Identification of bioactive peptides with α‐amylase inhibitory potential from enzymatic protein hydrolysates of red seaweed (**. *Journal of Agricultural and Food Chemistry* (2018) **66** 4872-4882. DOI: 10.1021/acs.jafc.8b00960
5. Akbarbaglu Z., Mahdi Jafari S., Sarabandi K., Mohammadi M., Khakbaz Heshmati M., Pezeshki A.. **Influence of spray drying encapsulation on the retention of antioxidant properties and microstructure of flaxseed protein hydrolysates**. *Colloids and Surfaces B: Biointerfaces* (2019) **178** 421-429. DOI: 10.1016/j.colsurfb.2019.03.038
6. Akharume F. U., Aluko R. E., Adedeji A. A.. **Modification of plant proteins for improved functionality: A review**. *Comprehensive Reviews in Food Science and Food Safety* (2021) **20** 198-224. DOI: 10.1111/1541-4337.12688
7. Ambigaipalan P., Al‐Khalifa A. S., Shahidi F.. **Antioxidant and angiotensin I converting enzyme (ACE) inhibitory activities of date seed protein hydrolysates prepared using Alcalase, Flavourzyme and Thermolysin**. *Journal of Functional Foods* (2015) **18** 1125-1137. DOI: 10.1016/j.jff.2015.01.021
8. Amini Sarteshnizi R., Sahari M. A., Ahmadi Gavlighi H., Regenstein J. M., Nikoo M., Udenigwe C. C.. **Influence of fish protein hydrolysate‐pistachio green hull extract interactions on antioxidant activity and inhibition of α‐glucosidase, α‐amylase, and DPP‐IV enzymes**. *LWT ‐ Food Science and Technology* (2021) **142**. DOI: 10.1016/j.lwt.2021.111019
9. Bessada S. M. F., Barreira J. C. M., Oliveira M. B. P. P.. **Pulses and food security: Dietary protein, digestibility, bioactive and functional properties**. *Trends in Food Science and Technology* (2019) **93** 53-68. DOI: 10.1016/j.tifs.2019.08.022
10. Boye J. I., Roufik S., Pesta N., Barbana C.. **Angiotensin I‐converting enzyme inhibitory properties and SDS‐PAGE of red lentil protein hydrolysates**. *LWT ‐ Food Science and Technology* (2010) **43** 987-991. DOI: 10.1016/j.lwt.2010.01.014
11. Bradford M. M.. **A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein‐dye binding**. *Analytical Biochemistry* (1976) **72** 248-254. DOI: 10.1016/0003-2697(76)90527-3
12. Bu G., Huang T., Li T.. **The separation and identification of the residual antigenic fragments in soy protein hydrolysates**. *Journal of Food Biochemistry* (2020) **44**. DOI: 10.1111/jfbc.13144
13. Chandrasekaran S., Gonzalez de Mejia E.. **Optimization, identification, and comparison of peptides from germinated chickpea (**. *Food Chemistry* (2022) **374**. DOI: 10.1016/j.foodchem.2021.131717
14. Chen L., Chen J., Ren J., Zhao M.. **Effects of ultrasound pretreatment on the enzymatic hydrolysis of soy protein isolates and on the emulsifying properties of hydrolysates**. *Journal of Agricultural and Food Chemistry* (2011a) **59** 2600-2609. DOI: 10.1021/jf103771x
15. Chen L., Chen J., Ren J., Zhao M.. **Modifications of soy protein isolates using combined extrusion pre‐treatment and controlled enzymatic hydrolysis for improved emulsifying properties**. *Food Hydrocolloids* (2011b) **25** 887-897. DOI: 10.1016/j.foodhyd.2010.08.013
16. Das D., Kabir M. E., Sarkar S., Wann S. B., Kalita J., Manna P.. **Antidiabetic potential of soy protein/peptide: A therapeutic insight**. *International Journal of Biological Macromolecules* (2022) **194** 276-288. DOI: 10.1016/j.ijbiomac.2021.11.131
17. de Matos F. M., de Lacerda J. T. J. G., Zanetti G., de Castro R. J. S.. **Production of black cricket protein hydrolysates with α‐amylase, α‐glucosidase and angiotensin I‐converting enzyme inhibitory activities using a mixture of proteases**. *Biocatalysis and Agricultural Biotechnology* (2022) **39**. DOI: 10.1016/j.bcab.2022.102276
18. Fadimu G. J., Farahnaky A., Gill H., Truong T.. **Influence of ultrasonic pretreatment on structural properties and biological activities of lupin protein hydrolysate**. *International Journal of Food Science & Technology* (2022) **57** 1729-1738. DOI: 10.1111/ijfs.15549
19. Fadimu G. J., Gill H., Farahnaky A., Truong T.. **Investigating the impact of ultrasound pretreatment on the physicochemical, structural, and antioxidant properties of lupin protein hydrolysates**. *Food and Bioprocess Technology* (2021) **14** 2004-2019. DOI: 10.1007/s11947-021-02700-4
20. Fadimu G. J., Gill H., Farahnaky A., Truong T.. **Improving the enzymolysis efficiency of lupin protein by ultrasound pretreatment: Effect on antihypertensive, antidiabetic and antioxidant activities of the hydrolysates**. *Food Chemistry* (2022) **383**. DOI: 10.1016/j.foodchem.2022.132457
21. Fathollahy I., Farmani J., Kasaai M. R., Hamishehkar H.. **Characteristics and functional properties of Persian lime (**. *LWT ‐ Food Science and Technology* (2021) **140**. DOI: 10.1016/j.lwt.2020.110765
22. Gharibzahedi S. M. T., Smith B.. **Effects of high hydrostatic pressure on the quality and functionality of protein isolates, concentrates, and hydrolysates derived from pulse legumes: A review**. *Trends in Food Science and Technology* (2021) **107** 466-479. DOI: 10.1016/j.tifs.2020.11.016
23. Guo Y., Wang K., Wu B., Wu P., Duan Y., Ma H.. **Production of ACE inhibitory peptides from corn germ meal by an enzymatic membrane reactor with a novel gradient diafiltration feeding working‐mode and**. *Journal of Functional Foods* (2020) **64**. DOI: 10.1016/j.jff.2019.103584
24. Han R., Hernández Álvarez A. J., Maycock J., Murray B. S., Boesch C.. **Comparison of alcalase‐ and pepsin‐treated oilseed protein hydrolysates – Experimental validation of predicted antioxidant, antihypertensive and antidiabetic properties**. *Current Research in Food Science* (2021) **4** 141-149. DOI: 10.1016/j.crfs.2021.03.001
25. He W., Tian L., Fang F., Chen D., Federici E., Pan S., Jones O. G.. **Limited hydrolysis and conjugation of zein with chitosan oligosaccharide by enzymatic reaction to improve functional properties**. *Food Chemistry* (2021) **348**. DOI: 10.1016/j.foodchem.2021.129035
26. Hu R., Chen G., Li Y.. **Production and characterization of antioxidative hydrolysates and peptides from corn gluten meal using papain, ficin, and bromelain**. *Molecules* (2020) **25**. DOI: 10.3390/molecules25184091
27. Hu R., Dunmire K. M., Truelock C. N., Paulk C. B., Aldrich G., Li Y.. **Antioxidant performances of corn gluten meal and DDGS protein hydrolysates in food, pet food, and feed systems**. *Journal of Agriculture and Food Research* (2020) **2**. DOI: 10.1016/j.jafr.2020.100030
28. Hu R., Xu J., Qi G., Wang W., Sun X. S., Li Y.. **Antioxidative hydrolysates from corn gluten meal may effectively reduce lipid oxidation and inhibit HepG2 cancer cell growth**. *Journal of Agriculture and Food Research* (2022) **7**. DOI: 10.1016/j.jafr.2021.100252
29. Jiang M., Yan H., He R., Ma Y.. **Purification and a molecular docking study of α‐glucosidase‐inhibitory peptides from a soybean protein hydrolysate with ultrasonic pretreatment**. *European Food Research and Technology* (2018) **244** 1995-2005. DOI: 10.1007/s00217-018-3111-7
30. Jiang X., Cui Z., Wang L., Xu H., Zhang Y.. **Production of bioactive peptides from corn gluten meal by solid‐state fermentation with**. *LWT ‐ Food Science and Technology* (2020) **131**. DOI: 10.1016/j.lwt.2020.109767
31. Jin D. X., Liu X. L., Zheng X. Q., Wang X. J., He J. F.. **Preparation of antioxidative corn protein hydrolysates, purification and evaluation of three novel corn antioxidant peptides**. *Food Chemistry* (2016) **204** 427-436. DOI: 10.1016/j.foodchem.2016.02.119
32. Jin J., Ma H., Wang B., Yagoub A. E. G. A., Wang K., He R., Zhou C.. **Effects and mechanism of dual‐frequency power ultrasound on the molecular weight distribution of corn gluten meal hydrolysates**. *Ultrasonics Sonochemistry* (2016) **30** 44-51. DOI: 10.1016/j.ultsonch.2015.11.021
33. Jin J., Ma H., Zhou C., Luo M., Liu W., Qu W., He R., Luo L., Yagoub A. E. G. A.. **Effect of degree of hydrolysis on the bioavailability of corn gluten meal hydrolysates**. *Journal of the Science of Food and Agriculture* (2015) **95** 2501-2509. DOI: 10.1002/jsfa.6982
34. Karimi A., Ahmadi Gavlighi H., Amini Sarteshnizi R., Udenigwe C. C.. **Effect of maize germ protein hydrolysate addition on digestion,**. *Journal of Cereal Science* (2021) **97**. DOI: 10.1016/j.jcs.2020.103148
35. Karimi A., Azizi M. H., Ahmadi Gavlighi H.. **Fractionation of hydrolysate from corn germ protein by ultrafiltration:**. *Food Science & Nutrition* (2020) **8** 2395-2405. DOI: 10.1002/fsn3.1529
36. Lee B. Q., Khor S. M.. **3‐Chloropropane‐1,2‐diol (3‐MCPD) in soy sauce: A review on the formation, reduction, and detection of this potential carcinogen**. *Comprehensive Reviews in Food Science and Food Safety* (2015) **14** 48-66. DOI: 10.1111/1541-4337.12120
37. Li G., Liu W., Wang Y., Jia F., Wang Y., Ma Y., Gu R., Lu J.. **Functions and applications of bioactive peptides from corn gluten meal**. *Advances in Food and Nutrition Research* (2019) **87** 1-41. DOI: 10.1016/bs.afnr.2018.07.001
38. Liu F. F., Li Y. Q., Sun G. J., Wang C. Y., Liang Y., Zhao X. Z., He J.‐X., Mo H. Z.. **Influence of ultrasound treatment on the physicochemical and antioxidant properties of mung bean protein hydrolysate**. *Ultrasonics Sonochemistry* (2022) **84**. DOI: 10.1016/j.ultsonch.2022.105964
39. Liu P., Huang M., Song S., Hayat K., Zhang X., Xia S., Jia C.. **Sensory characteristics and antioxidant activities of Maillard reaction products from soy protein hydrolysates with different molecular weight distribution**. *Food and Bioprocess Technology* (2012) **5** 1775-1789. DOI: 10.1007/s11947-010-0440-3
40. Liu W. Y., Fang L., Feng X. W., Li G. M., Gu R. Z.. *European Food Research and Technology* (2020) **246** 2017-2027. DOI: 10.1007/s00217-020-03552-6
41. Liu X., Zheng X., Song Z., Liu X., Kopparapu N. K., Wang X., Zheng Y.. **Preparation of enzymatic pretreated corn gluten meal hydrolysate and**. *Journal of Functional Foods* (2015) **18** 1147-1157. DOI: 10.1016/j.jff.2014.10.013
42. Meinlschmidt P., Schweiggert‐Weisz U., Eisner P.. **Soy protein hydrolysates fermentation: Effect of debittering and degradation of major soy allergens**. *LWT ‐ Food Science and Technology* (2016) **71** 202-212. DOI: 10.1016/j.lwt.2016.03.026
43. Ngoh Y. Y., Gan C. Y.. **Enzyme‐assisted extraction and identification of antioxidative and α‐amylase inhibitory peptides from Pinto beans (**. *Food Chemistry* (2016) **190** 331-337. DOI: 10.1016/j.foodchem.2015.05.120
44. Nikoo M., Benjakul S., Ahmadi Gavlighi H.. **Protein hydrolysates derived from aquaculture and marine byproducts through autolytic hydrolysis**. *Comprehensive Reviews in Food Science and Food Safety* (2022) **21** 4872-4899. DOI: 10.1111/1541-4337.13060
45. Ortiz‐Martinez M., Otero‐Pappatheodorou J. T., Serna‐Saldívar S. O., García‐Lara S.. **Antioxidant activity and characterization of protein fractions and hydrolysates from normal and quality protein maize kernels**. *Journal of Cereal Science* (2017) **76** 85-91. DOI: 10.1016/j.jcs.2017.05.021
46. Ozón B., Cotabarren J., Valicenti T., Graciela Parisi M., David Obregón W.. **Chia expeller: A promising source of antioxidant, antihypertensive and antithrombotic peptides produced by enzymatic hydrolysis with Alcalase and Flavourzyme**. *Food Chemistry* (2022) **380**. DOI: 10.1016/j.foodchem.2022.132185
47. Qiao H., Bi X., Zhang Y., Liu M., Zu S., Jia N., Jiang S., Lu Q., Zu Y., Bao Y.. **Enzymic polypeptide antioxidant activity and inhibitory activity on α‐glucosidase and α‐amylase from**. *Industrial Crops and Products* (2020) **146**. DOI: 10.1016/j.indcrop.2020.112158
48. Quintero‐Soto M. F., Chávez‐Ontiveros J., Garzón‐Tiznado J. A., Salazar‐Salas N. Y., Pineda‐Hidalgo K. V., Delgado‐Vargas F., López‐Valenzuela J. A.. **Characterization of peptides with antioxidant activity and antidiabetic potential obtained from chickpea (**. *Journal of Food Science* (2021) **86** 2962-2977. DOI: 10.1111/1750-3841.15778
49. Rahimi R., Gavlighi H. A., Sarteshnizi R. A., Barzegar M., Udenigwe C. C.. *LWT ‐ Food Science and Technology* (2022) **154**. DOI: 10.1016/j.lwt.2021.112765
50. Ren X., Liang Q., Zhang X., Hou T., Li S., Ma H.. **Stability and antioxidant activities of corn protein hydrolysates under simulated gastrointestinal digestion**. *Cereal Chemistry* (2018) **95** 760-769. DOI: 10.1002/cche.10092
51. Reyes Jara A. M., Liggieri C. S., Bruno M. A.. **Preparation of soy protein hydrolysates with antioxidant activity by using peptidases from latex of**. *Food Chemistry* (2018) **264** 326-333. DOI: 10.1016/j.foodchem.2018.05.013
52. Rezvankhah A., Emam‐Djomeh Z., Askari G.. **Encapsulation and delivery of bioactive compounds using spray and freeze‐drying techniques: A review**. *Drying Technology* (2020) **38** 235-258. DOI: 10.1080/07373937.2019.1653906
53. Rezvankhah A., Emam‐Djomeh Z., Safari M., Salami M., Askari G.. **Investigating the effects of maltodextrin, gum arabic, and whey protein concentrate on the microencapsulation efficiency and oxidation stability of hemp seed oil**. *Journal of Food Processing and Preservation* (2022) **46**. DOI: 10.1111/jfpp.16554
54. Rezvankhah A., Yarmand M. S., Ghanbarzadeh B.. **The effects of combined enzymatic and physical modifications of lentil protein applying Alcalase, Flavourzyme, microbial transglutaminase, and ultrasound: Antioxidant, antihypertension, and antidiabetic activities**. *Journal of Food Measurement and Characterization* (2022) **16** 3743-3759. DOI: 10.1007/s11694-022-01478-z
55. Rezvankhah A., Yarmand M. S., Ghanbarzadeh B., Mirzaee H.. **Characterization of bioactive peptides produced from green lentil (**. *Journal of Food Processing and Preservation* (2021a) **45**. DOI: 10.1111/jfpp.15932
56. Rezvankhah A., Yarmand M. S., Ghanbarzadeh B., Mirzaee H.. **Generation of bioactive peptides from lentil protein: Degree of hydrolysis, antioxidant activity, phenol content, ACE‐inhibitory activity, molecular weight, sensory, and functional properties**. *Journal of Food Measurement and Characterization* (2021b) **15** 5021-5035. DOI: 10.1007/s11694-021-01077-4
57. Sarabandi K., Gharehbeglou P., Jafari S. M.. **Spray‐drying encapsulation of protein hydrolysates and bioactive peptides: Opportunities and challenges**. *Drying Technology* (2019) **38** 577-595. DOI: 10.1080/07373937.2019.1689399
58. Shen Y., Hu R., Li Y.. **Antioxidant and emulsifying activities of corn gluten meal hydrolysates in oil‐in‐water emulsions**. *Journal of the American Oil Chemists' Society* (2020) **97** 175-185. DOI: 10.1002/aocs.12286
59. Singh B. P., Vij S., Hati S.. **Functional significance of bioactive peptides derived from soybean**. *Peptides* (2014) **54** 171-179. DOI: 10.1016/j.peptides.2014.01.022
60. Tacias‐Pascacio V. G., Morellon‐Sterling R., Siar E. H., Tavano O., Berenguer‐Murcia Á., Fernandez‐Lafuente R.. **Use of Alcalase in the production of bioactive peptides: A review**. *International Journal of Biological Macromolecules* (2020) **165** 2143-2196. DOI: 10.1016/j.ijbiomac.2020.10.060
61. Tian R., Feng J., Huang G., Tian B., Zhang Y., Jiang L., Sui X.. **Ultrasound driven conformational and physicochemical changes of soy protein hydrolysates**. *Ultrasonics Sonochemistry* (2020) **68**. DOI: 10.1016/j.ultsonch.2020.105202
62. Wang K., Luo S., Cai J., Sun Q., Zhao Y., Zhong X., Jiang S., Zheng Z.. **Effects of partial hydrolysis and subsequent cross‐linking on wheat gluten physicochemical properties and structure**. *Food Chemistry* (2016) **197** 168-174. DOI: 10.1016/j.foodchem.2015.10.123
63. Wang R., Zhao H., Pan X., Orfila C., Lu W., Ma Y.. **Preparation of bioactive peptides with antidiabetic, antihypertensive, and antioxidant activities and identification of α‐glucosidase inhibitory peptides from soy protein**. *Food Science & Nutrition* (2019) **7** 1848-1856. DOI: 10.1002/fsn3.1038
64. Wang Y. H., Wang J. M., Wan Z. L., Yang X. Q., Chen X. W.. **Corn protein hydrolysate as a new structural modifier for soybean protein isolate based O/W emulsions**. *LWT ‐ Food Science and Technology* (2020) **118**. DOI: 10.1016/j.lwt.2019.108763
65. Wong S. F., Lee B. Q., Low K. H., Jenatabadi H. S., Wan Mohamed Radzi C. W. J. B., Khor S. M.. **Estimation of the dietary intake and risk assessment of food carcinogens (3‐MCPD and 1,3‐DCP) in soy sauces by Monte Carlo simulation**. *Food Chemistry* (2020) **311**. DOI: 10.1016/j.foodchem.2019.126033
66. Xu Z., Wu C., Sun‐Waterhouse D., Zhao T., Waterhouse G. I. N., Zhao M., Su G.. **Identification of post‐digestion angiotensin‐I converting enzyme (ACE) inhibitory peptides from soybean protein isolate: Their production conditions and**. *Food Chemistry* (2021) **345**. DOI: 10.1016/j.foodchem.2020.128855
67. Yang Y., Guanjun T. A. O., Ping L. I. U., Liu J. I. A.. **Peptide with angiotensin I‐converting enzyme inhibitory activity from hydrolyzed corn gluten meal**. *Journal of Agricultural and Food Chemistry* (2007) **55** 7891-7895. DOI: 10.1021/jf0705670
68. Zhang W., Huang L., Chen W., Wang J., Wang S.. **Influence of ultrasound‐assisted ionic liquid pretreatments on the functional properties of soy protein hydrolysates**. *Ultrasonics Sonochemistry* (2021) **73**. DOI: 10.1016/j.ultsonch.2021.105546
69. Zhao Y., Zhao X., Sun‐Waterhouse D., Ivan Neil Waterhouse G., Zhao M., Zhang J., Wang F., Su G.. **Two‐stage selective enzymatic hydrolysis generates protein hydrolysates rich in Asn‐Pro and Ala‐His for enhancing taste attributes of soy sauce**. *Food Chemistry* (2021) **345**. DOI: 10.1016/j.foodchem.2020.128803
70. Zheng X. Q., Wang J. T., Liu X. L., Sun Y., Zheng Y. J., Wang X. J., Liu Y.. **Effect of hydrolysis time on the physicochemical and functional properties of corn glutelin by Protamex hydrolysis**. *Food Chemistry* (2015) **172** 407-415. DOI: 10.1016/j.foodchem.2014.09.080
71. Zhou C., Hu J., Yu X., Yagoub A. E. G. A., Zhang Y., Ma H., Gao X., Otu P. N. Y.. **Heat and/or ultrasound pretreatments motivated enzymolysis of corn gluten meal: Hydrolysis kinetics and protein structure**. *LWT ‐ Food Science and Technology* (2017) **77** 488-496. DOI: 10.1016/j.lwt.2016.06.048
72. Zhou K., Sun S., Canning C.. **Production and functional characterisation of antioxidative hydrolysates from corn protein via enzymatic hydrolysis and ultrafiltration**. *Food Chemistry* (2012) **135** 1192-1197. DOI: 10.1016/j.foodchem.2012.05.063
73. Zhu B., He H., Hou T.. **A comprehensive review of corn protein‐derived bioactive peptides: Production, characterization, bioactivities, and transport pathways**. *Comprehensive Reviews in Food Science and Food Safety* (2019) **18** 329-345. DOI: 10.1111/1541-4337.12411
|
---
title: Physicochemical, microbial, and functional attributes of processed Cheddar
cheese fortified with olive oil–whey protein isolate emulsion
authors:
- Shafeeqa Irfan
- Mian Anjum Murtaza
- Ghulam Mueen ud Din
- Iram Hafiz
- Mian Shamas Murtaza
- Sobia Rafique
- Kashif Ameer
- Muhammad Abrar
- Isam A. Mohamed Ahmed
journal: Food Science & Nutrition
year: 2022
pmcid: PMC10003027
doi: 10.1002/fsn3.3159
license: CC BY 4.0
---
# Physicochemical, microbial, and functional attributes of processed Cheddar cheese fortified with olive oil–whey protein isolate emulsion
## Abstract
Olive (*Olea europaea* L.) has triacylglycerols, phenolics, and other antioxidants in its composition playing significant roles in maintaining health and reducing the onset of diseases. This study aimed to analyze the quality, antioxidant, textural profile, and sensory properties of processed Cheddar cheese fortified with $0\%$, $5\%$, $10\%$, $15\%$, and $20\%$ (v/w) olive oil–whey protein isolate emulsion during 60 days of storage period. The results showed that processed cheese had significantly higher ($p \leq .05$) antioxidant activity, and total phenolic and flavonoids contents, whereas nonsignificant increase ($p \leq .05$) in moisture and acidity while decreasing tendencies in pH, fat, protein, and ash contents. Sensory analysis showed that processed Cheddar cheese with $5\%$ emulsion had higher taste, aroma, texture/appearance, overall acceptability scores, and hardness. Conclusively, results indicated that olive oil–whey protein isolate emulsion could be beneficial for manufacturing and commercializing processed cheeses, analogs, or spreads with improved nutritional value and sensory characteristics.
Present study aimed at preparing processed Cheddar cheese fortified with $0\%$, $5\%$, $10\%$, $15\%$, and $20\%$ (v/w) olive oil‐whey protein isolate emulsion. Quality, antioxidant, textural profile and sensory properties were analyzed during 60 days of storage. The results showed that processed cheese had significantly higher antioxidant activity, phenolics and flavonoids contents.
## INTRODUCTION
Processed cheese is a homogenous blend of cheese manufactured from several ingredients, such as the same or different natural cheese varieties, vegetable oils, butter oil, milk solids, emulsifying salts, and other dairy or nondairy ingredients with extended shelf life (Gulzar et al., 2020). Processed cheese has low functionality because of its low nutritional profile, which can be enhanced by incorporating valuable ingredients with high content of bioactive and functional components (Shaukat et al., 2022). Several studies have been conducted to evaluate the influence of different functional materials, such as vegetable powders (El‐Loly et al., 2022), fruits (Abbas et al., 2021), grain flour, and cheese fat substitution through the addition of vegetable oils (Hamdy et al., 2021) on processed cheese.
Olive is grown for olive oil and table olives production in Mediterranean regions (Lanza & Ninfali, 2020). Primarily olive oil is composed of more than $98\%$ triacylglycerols, oleic acid esters, and about $0.5\%$–$1.0\%$ nonglyceridic constituents (Makoś et al., 2017). Moreover, olive oil provides phenolic compounds that are well‐known for their health‐beneficial biological properties. In particular, olive oil's phenolic compounds possess anti‐inflammatory, antioxidant, and antimicrobial activities (Melguizo‐Rodríguez et al., 2021).
Currently, the food industry has a great demand for emulsions (Ranjha et al., 2021). The emulsion is thermodynamically unfavorable (breakdown over time) due to physiochemical mechanisms and can be stabilized using emulsifiers as stabilizers specifically (Weiss et al., 2020). Proteins exhibit amphiphilic character, therefore widely used as emulsifying/stabilizing ingredient in food emulsions. Proteins have the ability to create electrostatic and steric repulsive forces by forming an interfacial layer between oil droplets. During long‐term storage, these forces stabilize the droplets as opposed to coalescence and flocculation (Ding et al., 2021; Marhamati et al., 2021). Whey protein isolates (WPI), an emulsifier, enhance the formation and stability of oil‐in‐water emulsions (Hwang et al., 2017). These also have the capacity to prevent prooxidants to access the droplets, and consequently inhibit lipid oxidation (Nooshkam & Varidi, 2021).
So far, the characteristics and importance of processed cheese fortified with olive oil–whey protein isolate emulsion have not been studied. Therefore, the proposed study was designed to prepare and optimize the emulsion from olive oil with whey protein, improve the functional properties, and assess the physicochemical composition, antioxidant potential, microbiological characteristics, texture, and sensory acceptability of processed cheese fortified with olive oil–whey protein isolate emulsion.
## Materials
Natural Cheddar cheese (3 months ripened) was procured from Noon Pakistan Ltd. Bhalwal, Sargodha, Pakistan. Extra virgin olive oil was purchased from the Pansari Premium Herbal Store, Lahore, Pakistan. Whey protein isolate (WPI) was purchased online from Myprotein™ (United Kingdom). All other chemicals were obtained from the Dairy Processing Hall, Institute of Food Science and Nutrition, University of Sargodha, Sargodha, Pakistan.
## Preparation of olive oil emulsion (OOE)
Extra virgin olive oil and whey protein isolate (WPI) (oil‐in‐water) emulsion was prepared by following the procedure described by Kuhn and Cunha [2012] with some modifications. To prepare WPI stock solution, WPI was dissolved into deionized water at room temperature for 90 min with magnetic stirring and was subjected to overnight storage to achieve complete dissolution at room temperature (25 ± 2°C). The pH of WPI stock solution was adjusted to neutral by using 2.0 M buffer solution (NaOH) and stored at 10°C for a night.
To prepare emulsions, WPI stock solution and olive oil were used in $0.5\%$:$5\%$ (v/v) ratio, respectively. For the preparation of extra virgin olive oil and whey protein isolate emulsion, required amount of olive oil was poured drop‐wise into the required amount of encapsulating WPI solution. The mixture was homogenized at 14000 rpm for 4 min. Sodium azide ($0.02\%$ w/v) was mixed into the emulsion. The pH of the emulsions was adjusted to neutral by using 2.0 M buffer solution (NaOH).
## Preparation of processed Cheddar cheese
Processed Cheddar cheese was prepared using ripened natural Cheddar cheese, olive oil emulsion (OOE) at $0\%$ (control), $5\%$, $10\%$, $15\%$, and $20\%$ concentrations, and $2\%$ emulsifying salt by following the process described by Kapoor and Metzger [2008], with some modifications. Shredding of the Cheddar cheese was carried out to prepare processed Cheddar cheese. All the ingredients mentioned above were cumulatively poured in a steam‐jacked cooker followed by mixing aided with thermal treatment at a temperature range = 65 to 75°C for a time interval of 15 min. From the cooker, hot‐processed Cheddar cheese was taken out of the molds made with stainless steel having a depth of 10.16 cm. After that, processed cheese samples were cooled to bring their temperature to normal room temperature followed by slicing and packaging. Small rectangular shape blocks were made of processed cheese by slicing and were subjected to packaging under vacuum in polythene bags. Samples were then transferred to the storage facility and processed Cheddar cheese was stored at 2°C for 60 days.
## Physicochemical analyses
AOAC [2016] methods were employed for analyzing moisture (AOAC Method No. 948.12), protein (AOAC Method No. 2011.04), ash (AOAC Method No. 942.05), and acidity (AOAC Method No. 942.15) (%lactic acid) contents of processed Cheddar cheese. The fat content of processed Cheddar cheese was evaluated by following the Gerber method with some modifications as given by Marshall [1993]. The Gerber method is a volumetric method that employs chemical reagents, such as sulfuric acid to carry out the breakdown of emulsion and fat separation. A special flask was utilized for the measurement of fat content known as butyrometer. Briefly, 10 ml of the pipette was used for transferring sulfuric acid (10 ml) to the butyrometer. The samples were placed in the butyrometer and 1 ml of the amyl alcohol was added using the 1 ml pipette. Then, the butyrometer was placed in the water bath for a time interval of 5 min. After centrifugation, the butyrometers stoppers were oriented downward in the water bath for 3 to 10 min. The difference in the readings was denoted as fat mass in terms of percentage in cheese samples. The pH of processed Cheddar cheese was evaluated for pH by following the method described by Ong et al. [ 2007] using an Electronic Digital pH Meter. All analyses were carried out thrice ($$n = 3$$) at the interval of 0, 30, and 60 days during storage.
## Preparation of water‐soluble extracts (WSEs)
WSEs of processed Cheddar cheese were prepared by following the method developed by Gupta et al. [ 2013] with some modifications. Briefly, WSEs were prepared by first mixing 15 g of grated cheese in water (50 ml) followed by mixture placement in the water bath and thermal treatment at a temperature constraint of 40°C for a time interval of 5 min. Then, the mixture was homogenized using Omni‐Mixer homogenizer (Omni International, Waterburg, CT) for a period of 2 min. Furthermore, HCl (2 M) was employed for pH adjustment at 4.6 and then the mixture was added with distilled water until reaching the sample weight (grated cheese sample mixtures) equivalent to 100 g. Then, the placement of the samples in water bath was again carried out at 40°C temperature for 1 h in order to allow the melting fat of cheese samples followed by centrifugation of samples at 4500 rpm (3000 × g) at 40°C temperature for 1 h. After centrifugation, Whatman filter paper No. 1 was used for filtration. WSEs of processed Cheddar cheese were collected in a round‐bottom flask and then subjected to freeze‐drying. Powdered freeze‐dried samples were weighed, transferred to plastic tubes, and subjected to storage at −20°C.
## Determination of total antioxidant activity
For assessing the ascorbic acid equivalent (AAE) antioxidant capacity, the WSEs were evaluated by following the method described by Prieto et al. [ 1999] with some modifications at the interval of 0, 30, and 60 days during storage. Briefly, 1 ml of WSEs of processed Cheddar cheese was mixed with 4 ml of phosphomolybdate reagent [28 mM/L NaOH +0.6 M/L H2SO4 + 4 mM/L (NH4)2MoO4]. The mixture was vortexed for 30 s. Then, it was incubated in the water bath at 95°C for 90 min. The incubated mixture was cooled to room temperature and again vortexed for 30 s. The absorbance of samples was measured at 695 nm through a spectrophotometer (SpectraMax Plus384, Molecular Devices, Sunnyvale, CA). The ascorbic acid‐equivalent antioxidant capacity of processed Cheddar cheese was recalculated using the ascorbic acid standard curve as mg/100 ml AAE. All the replications and experiments were carried out thrice ($$n = 3$$).
## Determination of total phenolic content
For assessing the total phenolic content (TPC) in terms of gallic acid equivalent (GAE), the WSEs of processed Cheddar cheese were evaluated by following the method of Reis et al. [ 2012] with some modifications at the interval of 0, 30 and 60 days during storage. Briefly, 1 ml of WSEs of processed Cheddar cheese was mixed with 1 ml of $10\%$ (v/v) Folin–Ciocalteu reagent, vortexed for 30 s and left for 10 min at room temperature. Then, 2 ml of $20\%$ (w/v) sodium carbonate (Na2CO3) was added and again vortexed for 30 s. The mixture was incubated in dark at 30°C for 60 min. After incubation, the absorbance of samples was measured at 760 nm through a spectrophotometer. Gallic acid was dissolved in ethanol and used as a standard. The Gallic acid equivalent (GAE) phenolic content of processed Cheddar cheese was recalculated using the Gallic acid standard curve as mg GAE/100 ml. All the replications and experiments were carried out thrice.
## Determination of total flavonoid content
For assessing the total flavonoids content (TFC) in terms of Quercetin equivalent (QE), the WSEs of processed Cheddar cheese were evaluated by following the method described by Zhishen et al. [ 1999] with some modifications at the interval of 0, 30, and 60 days during storage. Briefly, 1.5 ml of WSEs of processed Cheddar cheese was mixed with 75 μl of $5\%$ (w/v) sodium nitrite (NaNO3) and vortexed for 1 min. Then, 150 μl of $10\%$ (wt/v) aluminum chloride (AlCl3) solution was added, again vortexed, and left for 5 min. Later, 0.5 ml of 1 M/L sodium hydroxide (NaOH) was added and vortexed again. Afterward, samples were incubated in dark for 45 min. After incubation, the absorbance of samples was measured at 510 nm through a spectrophotometer. Quercetin was dissolved in ethanol and used as a standard. TFC of processed Cheddar cheese was recalculated using Quercetin standard curve as mg QE/100 ml. All the replications and experiments were carried out thrice ($$n = 3$$).
## Microbiological analysis
The total plate count of processed Cheddar cheese was evaluated by following the method of APHA [1984]. Bismuth sulfite agar (Hi Media Ltd.) was used for the analysis. Briefly, 10 g of cheese sample was taken in a presterilized pestle and mortar and properly mixed with 90 ml of $0.1\%$ sterile peptone water. Furthermore, $0.1\%$ peptone water was used to prepare 10‐fold serial dilutions. To observe the aseptic conditions, the preparation of samples and serial dilutions were carried out inside the laminar flow cabinet.
## Textural profile analysis
Processed Cheddar cheese was evaluated for the texture profile analysis to assess the effect of olive oil–whey protein isolate emulsion and emulsifying salt on the texture of processed cheese by using TA‐XT plus Texture Analyzer and P‐75 compression plate probe by following the method described by O'Mahony et al. [ 2005]. Cheese samples were packed in air‐tight plastic bags and equilibrated for 18 h at 8°C. Samples were cut into cubes of 25 mm height, length, and width through a stainless‐steel wire cutter; before analysis, again equilibrated for 30 min at 8°C. Samples were taken out from the incubator and straightaway compressed in two consecutive cycles at 1 mm/s rate to $30\%$ of the original cheese height.
## Sensory evaluation
Sensory quality properties in terms of sensory characteristics of processed Cheddar cheese, such as color, texture, taste, flavor, appearance, and overall acceptability, were evaluated during storage using a 9‐point Hedonic scale. A sensory panel comprising 15 semi‐trained panelists carried out the sensory analysis of processed Cheddar cheese, and panelists included faculty members and post‐graduate students of the Institute of Food Science and Nutrition, University of Sargodha, Sargodha, Pakistan. Moreover, the necessary ethical approval for sensory evaluation from the Institutional Review Board (IRB) was sought under IRB No. SU/IFSN/IRB/004.
## Statistical analysis
Data were statistically analyzed using Statistix 8.1 software (analytical software, Tallahassee, Florida, USA). The two‐way Analysis of variance (ANOVA) technique was used to compare the means. A probability of $p \leq .05$ was used to establish statistical significance. Data were expressed as means ± SD.
## Physicochemical parameters
The moisture, fat, protein, and ash contents of processed Cheddar cheese samples were increased with the corresponding increase in the concentration of OOE. The change in the composition of processed cheese samples was highly significant ($p \leq .05$) with the addition of OOE. However, nonsignificant ($p \leq .05$) change was observed in the moisture, fat, protein, and ash contents of processed Cheddar cheese during 60 days of storage because each sample was air‐tightly sealed separately (Table 1). Gab‐Allah [2018] reported that during storage, moisture content of cheese was slightly decreased ($p \leq .05$) from different treatments that resulted in increased fat and moisture content of the cheese fortified with olive oil and sunflower oil wax. Khaliq et al. [ 2021] and Shekhar et al. [ 2015] reported a similar trend in the contents of moisture, fat, and protein of cheese fortified with olive oil during storage. Khaliq et al. [ 2021] employed extra virgin olive oil (EVOO) for the fortification of cream cottage cheese and concluded that a rise in total volatile fatty acids may be ascribed to protein breakage which leads to the enhanced flavor of cheese product. This enhanced flavor effect was related to moisture content replacement and high discharge of whey protein from the cottage cheese matrix.
**TABLE 1**
| Treatments | Storage period (days) | Storage period (days).1 | Storage period (days).2 |
| --- | --- | --- | --- |
| Treatments | 0 | 30 | 60 |
| Moisture (%) | Moisture (%) | Moisture (%) | Moisture (%) |
| T 0 (control) | 38.25 ± 0.03L | 38.52 ± 0.04m | 38.64 ± 0.09 n |
| T 1 (5% OOE) | 38.68 ± 0.04j | 38.81 ± 0.05k | 38.91 ± 0.07L |
| T 2 (10% OOE) | 38.96 ± 0.02g | 39.04 ± 0.01h | 39.14 ± 0.03i |
| T 3 (15% OOE) | 39.96 ± 0.03d | 40.25 ± 0.07 e | 40.32 ± 0.05f |
| T 4 (20% OOE) | 40.47 ± 0.05a | 40.63 ± 0.02b | 40.75 ± 0.04c |
| Fat (%) | Fat (%) | Fat (%) | Fat (%) |
| T 0 (control) | 30.57 ± 0.02m | 30.51 ± 0.05 n | 30.45 ± 0.03° |
| T 1 (5% OOE) | 30.71 ± 0.03j | 30.65 ± 0.07k | 30.60 ± 0.02L |
| T 2 (10% OOE) | 30.98 ± 0.08g | 30.93 ± 0.02h | 30.87 ± 0.01i |
| T 3 (15% OOE) | 31.23 ± 0.04d | 31.18 ± 0.02 e | 31.12 ± 0.02f |
| T 4 (20% OOE) | 32.56 ± 0.02a | 32.50 ± 0.03b | 32.45 ± 0.02c |
| Protein (%) | Protein (%) | Protein (%) | Protein (%) |
| T 0 (control) | 26.62 ± 0.06L | 26.59 ± 0.05m | 26.56 ± 0.03 n |
| T 1 (5% OOE) | 26.84 ± 0.02j | 26.81 ± 0.06k | 26.79 ± 0.02k |
| T 2 (10% OOE) | 27.08 ± 0.03g | 27.05 ± 0.01h | 27.02 ± 0.08i |
| T 3 (15% OOE) | 27.30 ± 0.01d | 27.28 ± 0.02 e | 27.25 ± 0.07f |
| T 4 (20% OOE) | 27.53 ± 0.05a | 27.50 ± 0.03b | 27.48 ± 0.04c |
| Ash (%) | Ash (%) | Ash (%) | Ash (%) |
| T 0 (control) | 3.95 ± 0.03m | 3.92 ± 0.03 n | 3.89 ± 0.03° |
| T 1 (5% OOE) | 4.60 ± 0.05j | 4.56 ± 0.03k | 4.53 ± 0.03L |
| T 2 (10% OOE) | 4.74 ± 0.02g | 4.70 ± 0.04h | 4.66 ± 0.07i |
| T 3 (15% OOE) | 4.90 ± 0.04d | 4.87 ± 0.02 e | 4.84 ± 0.05f |
| T 4 (20% OOE) | 4.99 ± 0.03a | 4.95 ± 0.04b | 4.93 ± 0.03c |
| pH | pH | pH | pH |
| T 0 (control) | 5.84 ± 0.03c | 5.86 ± 0.06b | 5.87 ± 0.02a |
| T 1 (5% OOE) | 5.81 ± 0.02de | 5.82 ± 0.05d | 5.84 ± 0.07c |
| T 2 (10% OOE) | 5.79 ± 0.03fg | 5.80 ± 0.06ef | 5.81 ± 0.03def |
| T 3 (15% OOE) | 5.77 ± 0.02ij | 5.78 ± 0.04hi | 5.79 ± 0.05gh |
| T 4 (20% OOE) | 5.72 ± 0.06L | 5.75 ± 0.05k | 5.76 ± 0.05jk |
| Titratable acidity (% lactic acid) | Titratable acidity (% lactic acid) | Titratable acidity (% lactic acid) | Titratable acidity (% lactic acid) |
| T 0 (control) | 0.9062 ± 0.0007L | 0.9045 ± 0.0005m | 0.9034 ± 0.0003 n |
| T 1 (5% OOE) | 0.9215 ± 0.0005i | 0.9116 ± 0.0002j | 0.9089 ± 0.0002k |
| T 2 (10% OOE) | 0.9302 ± 0.0005f | 0.9285 ± 0.0005g | 0.9242 ± 0.0003h |
| T 3 (15% OOE) | 0.9401 ± 0.0005d | 0.9323 ± 0.0009 e | 0.9311 ± 0.0010f |
| T 4 (20% OOE) | 0.9521 ± 0.0012a | 0.9499 ± 0.0010b | 0.9427 ± 0.0015c |
The pH of the processed Cheddar cheese treatments observed a significant reduction ($p \leq .05$) with the addition of OOE, which could be due to the carboxylic group of the emulsion. However, the pH insignificantly ($p \leq .05$) changed during the storage period of 60 days (Table 1). Khaliq et al. [ 2021] found that the decrease in pH of cream cottage cheese fortified with extra virgin is associated with high acidity. Abbas et al. [ 2015] reported a decrease in the pH of cheese yogurt fortified with EVOO during cold storage. Shan et al. [ 2011] reported that the hindrance in pH increase of processed during storage is associated with the higher quantity of phenolic compounds available in the herbal extracts. The acidity of the processed Cheddar cheese increased with increasing the OOE but acidity had a slight decrease ($p \leq .05$) during storage (Table 1). Shan et al. [ 2011] reported pH at the initial stage in the range 5.42–5.58 during 9 days of storage, and the authors reported a significantly rising tendency in the pH of control samples.
## Antioxidant activity
The results of WSEs of processed cheese are given in Table 2 which ranged from 88.98 mg GAE/100 ml to 107.92 mg GAE/100 ml. However, olive oil provides no less than 30 phenolic compounds (Musumeci et al., 2013). WSEs of processed Cheddar cheese showed a reduction in phenolic content ($p \leq .05$) during storage. However, phenolic content was higher in WSEs of processed Cheddar cheese as compared to control directly after the production (Table 2). During storage, WSEs of control processed Cheddar cheese did not show a significant change. Peptides that are naturally produced in cheese (Murtaza et al., 2022), due to the activity of starter and rennet, may also act as phenolic compounds. On other hand, some of these peptides react with phenolic compounds present in the cheese to neutralize as well as inhibit their activity (Fox et al., 2004). These peptides may have interacted with the phenolic content of OOE and could describe the reason for the phenolic content decrease in the WSEs of processed Cheddar cheese. The results were according to the findings of Solhi et al. ( 2020b) in the processed cheese containing asparagus powder. The authors concluded that decreeing tendency in the antioxidant activity of fortified processed cheese samples during storage. This may be attributed to the interactions of phenolic compounds of asparagus and whey proteins with active groups of –SH. Such interactions may lead to a reduction in antioxidant activity. Several enzymes, such as tyrosinase, are usually produced owing to the action of starters, which may cause the conversion of polyphenols into chemical compounds named quinine, which may interact with enzymes and proteins. These secondary reactions may cause alteration of the qualitative and functional properties of proteins even in the sensory characteristics of food products (Solhi et al., 2020a). Fadavi and Beglaryan [2015] reported similar results that UF‐Feta cheese enriched with peppermint extract did not show the expected results of water‐soluble phenolic content due to the rennet concentration.
**TABLE 2**
| Treatments | Storage period (days) | Storage period (days).1 | Storage period (days).2 |
| --- | --- | --- | --- |
| Treatments | 0 | 30 | 60 |
| Total phenolic content (mg GAE/100 ml) | Total phenolic content (mg GAE/100 ml) | Total phenolic content (mg GAE/100 ml) | Total phenolic content (mg GAE/100 ml) |
| T 0 (control) | 88.98 ± 0.03i | 88.56 ± 0.01g | 88.34 ± 0.01j |
| T 1 (5% OOE) | 89.29 ± 0.01h | 78.89 ± 0.01m | 64.36 ± 0.06° |
| T 2 (10% OOE) | 93.94 ± 0.07f | 82.92 ± 0.10L | 76.39 ± 0.10 n |
| T 3 (15% OOE) | 101.66 ± 0.04d | 96.78 ± 0.01 e | 85.94 ± 0.03k |
| T 4 (20% OOE) | 124.42 ± 0.01a | 116.72 ± 0.01b | 107.92 ± 0.01c |
| Total antioxidant content (mg AAE/100 ml) | Total antioxidant content (mg AAE/100 ml) | Total antioxidant content (mg AAE/100 ml) | Total antioxidant content (mg AAE/100 ml) |
| T 0 (control) | 108.74 ± 0.03k | 108.69 ± 0.01k | 108.64 ± 0.02k |
| T 1 (5% OOE) | 112.51 ± 0.01j | 115.62 ± 0.01i | 107.11 ± 0.10L |
| T 2 (10% OOE) | 117.12 ± 0.01h | 122.24 ± 0.21g | 116.78 ± 0.05h |
| T 3 (15% OOE) | 220.56 ± 0.04 e | 227.30 ± 0.03d | 209.43 ± 0.11f |
| T 4 (20% OOE) | 303.71 ± 0.01b | 308.94 ± 0.02a | 302.11 ± 1.74c |
| Total flavonoid content (mg QE/100 ml) | Total flavonoid content (mg QE/100 ml) | Total flavonoid content (mg QE/100 ml) | Total flavonoid content (mg QE/100 ml) |
| T 0 (control) | 43.46 ± 0.02m | 45.88 ± 0.01k | 44.89 ± 0.01L |
| T 1 (5% OOE) | 51.35 ± 0.09i | 44.89 ± 0.09L | 33.81 ± 0.06 n |
| T 2 (10% OOE) | 62.14 ± 0.05f | 54.79 ± 0.08h | 46.78 ± 0.09j |
| T 3 (15% OOE) | 73.51 ± 0.01c | 67.98 ± 0.01 e | 60.54 ± 0.01g |
| T 4 (20% OOE) | 83.71 ± 0.01a | 76.21 ± 0.02b | 69.83 ± 0.12c |
Emulsions stabilized with whey protein isolate (WPI) play the role of the antioxidant system, as α‐lactalbumin and β‐lactoglobulin are its major constituents having thiol function, disulfide bonds, and cysteyl residues, which are able to inhibit lipid oxidation by scavenging free radicals (Cayot & Lorient, 1997). Many in vitro and in vivo studies have reported that cheese naturally possesses various biologically active compounds, such as conjugated linoleic acid (CLA), γ‐aminobutyric acid (GABA), vitamins, organic acids, fatty acids, exopolysaccharides, and peptides, exhibiting antiproliferative, antimicrobial, and antioxidant activities and inhibiting angiotensin‐converting enzyme (ACE) (Geurts et al., 2012).
The total antioxidant capacity (TAC) of produced processed Cheddar cheese' WSEs samples are given in Table 2, was in the range of 108.74 mg/100 ml AAE to 168.82 mg AAE/100 ml. Processed Cheddar cheese' WSEs having OOE had significantly higher ($p \leq .05$) TAC levels compared to control WSEs of processed Cheddar cheese that gradually decreased during 60 days of storage. The decrease in TAC of WSEs of processed Cheddar cheese could be associated with the absorption of phenolic compounds of OOE with proteins' active groups; which might possess the ability to reduce the antioxidant effect of phenolic compounds. Shahidi and Naczk [2003] reported that the polyphenols are turned into quinines due to the activity of enzymes produced by the starters presented in cheese. Quinines are very active and might react with proteins resulting in the change in proteins' nutritional and physicochemical characteristics as well as in the sensory characteristics of food materials. The results are according to the findings of Apostolidis et al. [ 2007], who reported that herbal extract‐supplemented cheeses had significantly ($p \leq .05$) higher antioxidant activity in comparison to nonenriched samples. Solhi et al. ( 2020a) found similar results in their study on processed cheese containing tomato powder. During storage, the antioxidant activity of processed cheese samples exhibited diminished tendency, which could be possibly ascribed to the interaction of protein active groups with phenolic molecules owing to phenolic compounds absorption. In another study on cheese supplemented with extract of dehydrated cranberry fruit, Khalifa and Wahdan [2015] found a decrease in lipolysis, proteolysis, and acid value along with an increase in oxidation stability of the cheese.
The TFC of produced Cheddar processed cheese WSEs samples, given in Table 2, was in the range of 43.46 mg QE/100 ml to 51.74 mg QE/100 ml. WSEs of processed Cheddar cheese showed a reduction in flavonoid content ($p \leq .05$) during storage. The maximum total flavonoid content (83.71 mg QE/100 ml) was present in T4 having $20\%$ OOE at day 0 than control and other treatments. However, T0 showed the presence of TFC (43.46 mg QE/100 ml) but content was much less. In dairy products, the antioxidant activity of flavonoids is little known, however, flavonoids' antioxidant activity has been reported (Nadeem et al., 2013). Fruit or plant extracts/ oils may be the source of a higher quantity of flavonoids in any dairy products. Qureshi et al. [ 2019] reported similar results in a soft cheese (paneer) supplemented with the extracts of date (*Phoenix dactylifera* L.) cultivars and its whey. Authors have also concluded that rich amounts of flavonoids in plant extracts may lead to high concentrations of flavonoids in dairy products or extracts.
## Total plate count
Processed Cheddar cheese samples showed a significant ($p \leq .05$) increase in total plate count during 30 days of storage in control as well as in processed Cheddar cheese containing OOE. Singh et al. [ 2015] reported a similar decrease in the results of total plate count for the chevon cutlets treated with clove oil. Processed cheese samples having OOE showed a significant ($p \leq .05$) increase in total plate count throughout 30 days of storage period, but results were significantly ($p \leq .05$) less compared to the control during storage study. Comparatively, a slow increase in the total plate count in processed Cheddar cheese containing OOE may be credited to the antimicrobial properties of extra virgin olive oil in OOE, as shown in Table 3. Librán et al. [ 2013] also observed a similar antimicrobial effect of different aromatic plants' aqueous extracts in cheese. Authors have also reported that cheese microbial quality is dependent on unhygienic conditions during manufacturing, postmanufacturing conditions (handling, storage, and packaging), milk thermal treatment, and milk quality. Authors also reported that thermal treatment of milk at 82°C for 5 min time period could lead to the destruction of yeasts and molds as well as coliforms, but postmanufacturing conditions may also lead to microbial contamination (Librán et al., 2013; Qureshi et al., 2019). Mahajan et al. [ 2015] conducted a study on cheese fortified with pomegranate rind extract and reported similar antimicrobial effect results in the cheese. Comparatively, a slowly rising tendency in the total plate count of fortified cheese samples might be ascribed to the antimicrobial properties of pomegranate rind extract.
**TABLE 3**
| Treatments | Storage period (days) | Storage period (days).1 | Storage period (days).2 | Storage period (days).3 | Storage period (days).4 |
| --- | --- | --- | --- | --- | --- |
| Treatments | 0 | 7 | 14 | 21 | 30 |
| Total plate count (log cfu/g) | Total plate count (log cfu/g) | Total plate count (log cfu/g) | Total plate count (log cfu/g) | Total plate count (log cfu/g) | Total plate count (log cfu/g) |
| T 0 (control) | 3.07 ± 0.08a | 3.68 ± 0.03a | 4.32 ± 0.03b | 4.74 ± 0.04c | 6.34 ± 0.05d |
| T 1 (5% OOE) | 2.86 ± 0.02a | 3.36 ± 0.04b | 4.03 ± 0.06c | 4.48 ± 0.06c | 6.07 ± 0.06d |
| T 2 (10% OOE) | 2.53 ± 0.05a | 3.14 ± 0.08b | 3.73 ± 0.05c | 4.26 ± 0.05c | 5.62 ± 0.04d |
| T 3 (15% OOE) | 2.07 ± 0.03a | 2.89 ± 0.03b | 3.58 ± 0.05c | 3.96 ± 0.07c | 5.17 ± 0.03d |
| T 4 (20% OOE) | 1.69 ± 0.05a | 2.62 ± 0.09b | 3.27 ± 0.04b | 3.63 ± 0.03c | 4.85 ± 0.02d |
## Textural properties of processed cheese
Quigley et al. [ 2011] reported that texture, a critical characteristic, is the resultant of chemical and physical properties that define the attributes and identity of any food product certainly for the cheese. Processing and compositional parameters greatly affect the textures of cheeses. Hardness has a critical role in the development of cheese texture as Meullenet and Gross [1999] have defined hardness as the required force to bite a sample, positioned in the middle of molar teeth, completely through it. Table 4 shows the hardness of sliced processed Cheddar cheese evaluated with respect to different concentrations of OOE at day 0. The maximum hardness was reported in T1 (3800 ± 31.17 g) having $5\%$ of OOE. T0 (control), T1 ($5\%$ OOE), T2 ($10\%$ OOE), T3 ($15\%$ OOE), and T4 ($20\%$ OOE) exhibited springiness values of $0.61\%$, $0.59\%$, $0.36\%$, $0.39\%$, and $0.28\%$, respectively. T0 (control), T1 ($5\%$ OOE), T2 ($10\%$ OOE), T3 ($15\%$ OOE), and T4 ($20\%$ OOE) exhibited gumminess values of 1.07, 1.34, 1.65, 1.84, and 1.33 N, respectively (Table 4). These results were similar to findings reported by Sołowiej et al. [ 2014], who prepared processed cheese using acid and rennet casein at concentrations ranging $11\%$–$13\%$. Furthermore, the authors investigated the effects of acid and rennet casein substitution on hardness, cohesiveness, and adhesiveness. Processed cheese prepared with the addition of $10\%$ rennet casein and $3\%$ whey protein isolate exhibited a high degree of hardness of 8869.6 g. This increase in hardness might be attributable to the calcium ions' complementary effect on the hydration process as well as para‐casein aggregation, which exerted significant influence on the water‐binding ability of the casein matrix owing to calcium and disulfide bridges formation in conjunction with casein cross‐linkages (Kapoor & Metzger, 2008). Khan et al. [ 2019] reported similar texture evaluation results. The authors evaluated the effect of *Citrus reticulata* Blanco crude flower extract on processed cheese. The processed cheese added with $2\%$ and $4\%$ crude flower extracts exhibited soft and semi‐hard textures, respectively.
**TABLE 4**
| Parameter | 0 day | 0 day.1 | 0 day.2 | 0 day.3 | 0 day.4 |
| --- | --- | --- | --- | --- | --- |
| Parameter | T0 (control) | T1 (5% OOE) | T2 (10% OOE) | T3 (15% OOE) | T4 (20% OOE) |
| Hardness (g) | 3857 ± 30.04d | 3800 ± 31.17c | 3775 ± 73.90b | 3745 ± 48.50a | 3710 ± 49.06a |
| Springiness (%) | 0.61 ± 0.02a | 0.59 ± 0.16d | 0.36 ± 0.06b | 0.39 ± 0.08c | 0.28 ± 0.08a |
| Gumminess (N) | 1.07 ± 0.07a | 1.34 ± 0.03c | 1.65 ± 0.23b | 1.84 ± 0.18a | 1.33 ± 0.26d |
## Sensorial attributes of processed cheese
Sensory evaluation of different attributes revealed that processed Cheddar cheese treatments having OOE had higher sensory scores of for taste, aroma, texture/appearance, and overall acceptability in comparison to the control (Table 5). The best score for taste (7.65 ± 0.41), aroma (7.13 ± 0.26), texture/appearance (7.50 ± 0.41), and overall acceptability (7.80 ± 0.26) were observed for $5\%$ OOE processed Cheddar cheese during 60 days of storage. In processed Cheddar cheese, the taste is the major determinant for quality and customer acceptance. Control was without OOE, resulting in a lightly salty taste. While, cheese made with $5\%$ and $10\%$ OOE resulted in “slight olive,” “buttery,” and “salty” taste. However, cheeses made with $15\%$ and $20\%$ OOE resulted in “slightly bitter and sour” taste. Similar scores were observed by Caspia et al. [ 2006]. The aromas of the product are very sensitive to processing and storage and influence the flavors of food. Control resulted in typical Cheddar cheese aroma while processed Cheddar cheese having $15\%$ and $20\%$ OOE resulted in high‐intense olive oil aroma. On the other hand, processed Cheddar cheese having $5\%$ and $10\%$ OOE resulted in a less intense olive oil aroma. Khan et al. [ 2019] reported similar sensory evaluation results. The authors evaluated the effect of *Citrus reticulata* Blanco crude flower extract on the sensory properties of processed cheese. The processed cheese added with $2\%$ and $4\%$ crude flower extract exhibited bitter and umami tastes, respectively. The probable reason for the bitter and umami taste was attributed to the excessive accumulation of small hydrophobic peptides and masking compounds (salt) in cheese curd. After 12 months of the storage period, the Cheddar cheese texture changed because of accelerated ripening. Market value, quality, and consumer acceptance are critically based on the texture of the food product. Results showed that processed Cheddar cheeses having $15\%$ and $20\%$ of OOE had semi‐soft and crumbly texture/ appearance. On the other hand, results showed that $5\%$ and $10\%$ OOE improved the texture/appearance of processed Cheddar cheese as compared to the control. The results were similar to the study conducted by Prinsloo [2007]. The quality, taste, aroma, texture, appearance, and likeness or dislikeness of judges decide the products’ overall acceptance. Judges preferred T1 and T2 having better taste, aroma, and texture/appearance keeping the typical characteristics of Cheddar cheese. These two treatments are even preferred over processed control (T0) due to better sensory scoring.
**TABLE 5**
| Treatments | Storage period (days) | Storage period (days).1 | Storage period (days).2 |
| --- | --- | --- | --- |
| Treatments | 0 | 30 | 60 |
| Taste | Taste | Taste | Taste |
| T 0 (control) | 7.69 ± 0.57a | 6.81 ± 0.56cd | 5.98 ± 0.40 e |
| T 1 (5% OOE) | 7.65 ± 0.41a | 7.19 ± 0.35bc | 6.21 ± 0.28 e |
| T 2 (10% OOE) | 7.73 ± 0.42a | 6.80 ± 0.33d | 6.26 ± 0.32 e |
| T 3 (15% OOE) | 7.25 ± 0.35b | 6.23 ± 0.22 e | 5.68 ± 0.21f |
| T 4 (20% OOE) | 7.05 ± 0.43b | 5.97 ± 0.23ef | 5.21 ± 0.23g |
| Aroma | Aroma | Aroma | Aroma |
| T 0 (control) | 7.13 ± 0.26b‐f | 6.94 ± 0.26d‐h | 6.74 ± 0.26fgh |
| T 1 (5% OOE) | 7.81 ± 0.62a | 7.67 ± 0.65ab | 7.49 ± 0.62a‐d |
| T 2 (10% OOE) | 7.65 ± 0.41ab | 7.52 ± 0.43abc | 7.37 ± 0.44a‐e |
| T 3 (15% OOE) | 7.15 ± 0.91b‐f | 6.98 ± 0.95c‐g | 6.83 ± 0.91 e‐h |
| T 4 (20% OOE) | 6.75 ± 0.68fgh | 6.56 ± 0.70gh | 6.42 ± 0.68h |
| Texture/appearance | Texture/appearance | Texture/appearance | Texture/appearance |
| T 0 (control) | 7.50 ± 0.41abc | 7.40 ± 0.41bc | 7.30 ± 0.41c |
| T 1 (5% OOE) | 7.79 ± 0.24a | 7.69 ± 0.24ab | 7.58 ± 0.25abc |
| T 2 (10% OOE) | 7.70 ± 0.26ab | 7.60 ± 0.26abc | 7.50 ± 0.26abc |
| T 3 (15% OOE) | 6.65 ± 0.48d | 6.55 ± 0.47de | 6.45 ± 0.47def |
| T 4 (20% OOE) | 6.29 ± 0.48efg | 6.19 ± 0.48fg | 6.08 ± 0.50g |
| Overall acceptability | Overall acceptability | Overall acceptability | Overall acceptability |
| T 0 (control) | 7.80 ± 0.26a‐d | 7.70 ± 0.27b‐e | 7.60 ± 0.29c‐f |
| T 1 (5% OOE) | 8.15 ± 0.34a | 8.04 ± 0.34ab | 7.93 ± 0.34abc |
| T 2 (10% OOE) | 8.00 ± 0.21ab | 7.91 ± 0.20abc | 7.81 ± 0.17a‐d |
| T 3 (15% OOE) | 7.69 ± 0.53b‐e | 7.55 ± 0.58def | 7.41 ± 0.55efg |
| T 4 (20% OOE) | 7.29 ± 0.48efg | 7.14 ± 0.53gh | 7.04 ± 0.53h |
## CONCLUSIONS
It was concluded that processed Cheddar cheese fortified with different concentrations of olive oil–whey protein isolate emulsion is nutritionally excellent as compared to the control. Antioxidant potential was significantly ($p \leq .05$) higher in processed Cheddar cheeses fortified with olive oil emulsion in comparison with the control. In addition, processed Cheddar cheese with OOE showed better antimicrobial activity compared to the control. Processed Cheddar cheese with $5\%$ emulsion showed excellent sensory perception. Olive oil–whey protein isolate emulsion could be used to manufacture and commercialize processed cheese, analogs, or spreads with improved nutritional value and sensory characteristics. It is recommended that quantification of bioactive compounds (exopolysaccharides, organic acids, peptides, fatty acids, γ‐aminobutyric acid, and vitamins) extracted from processed Cheddar cheese should be done for a better understanding of biological activities.
## CONFLICT OF INTEREST
The authors declare no conflict of interest.
## DATA AVAILABILITY STATEMENT
The data supporting the conclusions of this article are included in the manuscript.
## References
1. Abbas H. M., Hassan F. A. M., Kassem J. M., Rasmy N. M., El‐Kaluobi M. H., El‐Okaby M. F.. **Preparation of cheese yoghurt using extracted high virgin olive oil**. *International Journal of Dairy Science* (2015) **10** 288-296
2. Abbas K. A., Abdelmontaleb H. S., Hamdy S. M., Aït‐Kaddour A.. **Physicochemical, functional, fatty acids profile, health lipid indices, microstructure and sensory characteristics of walnut‐processed cheeses**. *Food* (2021) **10** 2274
3. AOAC
. (2016). Official methods of analysis of AOAC international (20th ed.). Association of Official Analytical Chemists.. *Official methods of analysis of AOAC international* (2016)
4. APHA
. (1984). Compendium of method of microbiological examination of foods (2nd ed.). American Public Health Association.. *Compendium of method of microbiological examination of foods* (1984)
5. Apostolidis E., Kwon Y. I., Shetty K.. **Inhibitory potential of herb, fruit, and fungal‐enriched cheese against key enzymes linked to type 2 diabetes and hypertension**. *Innovative Food Science and Emerging Technologies* (2007) **8** 46-54
6. Caspia E. L., Coggins P. C., Schilling M. W., Yoon Y., White C. H.. **The relationship between consumer acceptability and descriptive sensory attributes in cheddar cheese**. *Journal of Sensory Studies* (2006) **21** 112-127
7. Cayot P., Lorient D.. **Structure‐function relationships of whey proteins**. *Food Science and Technology‐New York‐Marcel Dekker* (1997) **11** 225-256
8. Ding Y., Chen L., Shi Y., Akhtar M., Chen J., Ettelaie R.. **Emulsifying and emulsion stabilizing properties of soy protein hydrolysates, covalently bonded to polysaccharides: The impact of enzyme choice and the degree of hydrolysis**. *Food Hydrocolloids* (2021) **113**
9. El‐Loly M. M., Mohamed A. G., Farahat E. S.. **Innovative vegetables‐processed cheese: II. High nutritional and functional attributes**. *Bioactive Compounds in Health and Disease* (2022) **5** 13-32
10. Fadavi A., Beglaryan R.. **Optimization of UF‐feta cheese preparation, enriched by peppermint extract**. *Journal of Food Science and Technology* (2015) **52** 952-959. PMID: 25694705
11. Fox P. F., McSweeney P. L., Cogan T. M., Guinee T. P.. *Cheese: Chemistry, physics and microbiology, Volume 1: General aspects* (2004)
12. Gab‐Allah R. H.. **Manufacture of pickled and un‐pickled high fat soft cheese using olive and sunflower oleogels**. *Sciences* (2018) **8** 223-230
13. Geurts L., Everard A., Le Ruyet P., Delzenne N. M., Cani P. D.. **Ripened dairy products differentially affect hepatic lipid content and adipose tissue oxidative stress markers in obese and type 2 diabetic mice**. *Journal of Agricultural and Food Chemistry* (2012) **60** 2063-2068. PMID: 22300436
14. Gulzar N., Sameen A., Muhammad Aadil R., Sahar A., Rafiq S., Huma N., Nadeem M., Arshad R., Muqadas Saleem I.. **Descriptive sensory analysis of pizza cheese made from mozzarella and semi‐ripened cheddar cheese under microwave and conventional cooking**. *Food* (2020) **9** 214
15. Gupta A., Mann B., Kumar R., Sangwan R. B.. **ACE‐inhibitory activity of cheddar cheeses made with adjunct cultures at different stages of ripening**. *Advances in Dairy Research* (2013) **1** 102
16. Hamdy S. M., Hassan M. G., Ahmed R. B., Abdelmontaleb H. S.. **Impact of oat flour on some chemical, physicochemical and microstructure of processed cheese**. *Journal of Food Processing and Preservation* (2021) **45**
17. Hwang J. Y., Ha H. K., Lee M. R., Kim J. W., Kim H. J., Lee W. J.. **Physicochemical property and oxidative stability of whey protein concentrate multiple nanoemulsion containing fish oil**. *Journal of Food Science* (2017) **82** 437-444. PMID: 28103390
18. Kapoor R., Metzger L. E.. **Process cheese: Scientific and technological aspects—A review**. *Comprehensive Reviews in Food Science and Food Safety* (2008) **7** 194-214
19. Khalifa S. A., Wahdan K. M.. **Improving the quality characteristics of white soft cheese using cranberry (**. *International Food Research Journal* (2015) **22** 2203-2211
20. Khaliq A., Zahid H., Qurat ul ain S., Ansar S., Zeeshan A., Yaqub M., Rabani S., Sajjad Q.. **Development and quality characterization of cream cottage cheese spread fortified with extra virgin olive oil (**. *Life Science Journal* (2021) **18** 42-48
21. Khan U. M., Ahmad I., Inayat S., Amin H. M. A., Selamoglu Z.. **Physicochemical properties of Cheddar cheese made from Citrus reticulata Blanco crude flowers extract**. *Turkish Journal of Agriculture‐Food Science and Technology* (2019) **7** 856-860
22. Kuhn K. R., Cunha R. L.. **Flaxseed oil–whey protein isolate emulsions: Effect of high pressure homogenization**. *Journal of Food Engineering* (2012) **111** 449-457
23. Lanza B., Ninfali P.. **Antioxidants in extra virgin olive oil and table olives: Connections between agriculture and processing for health choices**. *Antioxidants* (2020) **9** 41. PMID: 31906540
24. Librán C. M., Moro A., Zalacain A., Molina A., Carmona M., Berruga M. I.. **Potential application of aromatic plant extracts to prevent cheese blowing**. *World Journal of Microbiology and Biotechnology* (2013) **29** 1179-1188. PMID: 23417280
25. Mahajan D., Bhat Z. F., Kumar S.. **Pomegranate (Punica granatum) rind extract as a novel preservative in cheese**. *Food Bioscience* (2015) **12** 47-53
26. Makoś P., Fernandes A., Boczkaj G.. **Method for the determination of carboxylic acids in industrial effluents using dispersive liquid‐liquid microextraction with injection port derivatization gas chromatography–mass spectrometry**. *Journal of Chromatography A* (2017) **1517** 26-34. PMID: 28855089
27. Marhamati M., Ranjbar G., Rezaie M.. **Effects of emulsifiers on the physicochemical stability of oil‐in‐water Nanoemulsions: A critical review**. *Journal of Molecular Liquids* (2021) **340**
28. Marshall V. M.. **Starter cultures for milk fermentation and their characteristics**. *International Journal of Dairy Technology* (1993) **46** 49-56
29. Melguizo‐Rodríguez L., Illescas‐Montes R., Costela‐Ruiz V. J., Ramos‐Torrecillas J., de Luna‐Bertos E., García‐Martínez O., Ruiz C.. **Antimicrobial properties of olive oil phenolic compounds and their regenerative capacity towards fibroblast cells**. *Journal of Tissue Viability* (2021) **30** 372-378. PMID: 33810929
30. Meullenet J. F., Gross J.. **Instrumental single and double compression tests to predict sensory texture characteristics of foods**. *Journal of Texture Studies* (1999) **30** 167-180
31. Murtaza M. A., Irfan S., Hafiz I., Ranjha M. M. A., Rahaman A., Murtaza M. S., Ibrahim S. A., Siddiqui S. A.. **Conventional and novel technologies in the production of dairy bioactive peptides**. *Frontiers in Nutrition* (2022) **9** 1-11
32. Musumeci G., Trovato F. M., Pichler K., Weinberg A. M., Loreto C., Castrogiovanni P.. **Extra‐virgin olive oil diet and mild physical activity prevent cartilage degeneration in an osteoarthritis model: An in vivo and in vitro study on lubricin expression**. *The Journal of Nutritional Biochemistry* (2013) **24** 2064-2075. PMID: 24369033
33. Nadeem M., Abdullah M., Hussain I., Inayat S., Javid A., Zahoor Y.. **Antioxidant potential of Moringa oleifera leaf extract for the stabilisation of butter at refrigeration temperature**. *Czech Journal of Food Sciences* (2013) **31** 332-339
34. Nooshkam M., Varidi M.. **Physicochemical stability and gastrointestinal fate of β‐carotene‐loaded oil‐in‐water emulsions stabilized by whey protein isolate‐low acyl gellan gum conjugates**. *Food Chemistry* (2021) **347**. PMID: 33493834
35. O'Mahony J. A., Lucey J. A., McSweeney P. L. H.. **Chymosin‐mediated proteolysis, calcium solubilization, and texture development during the ripening of Cheddar cheese**. *Journal of Dairy Science* (2005) **88** 3101-3114. PMID: 16107399
36. Ong L., Henriksson A., Shah N. P.. **Chemical analysis and sensory evaluation of Cheddar cheese produced with lactobacillus acidophilus, Lb. casei, Lb. paracasei or Bifidobacterium sp**. *International Dairy Journal* (2007) **17** 937-945
37. Prieto P., Pineda M., Aguilar M.. **Spectrophotometric quantitation of antioxidant capacity through the formation of a phosphomolybdenum complex: Specific application to the determination of vitamin E**. *Analytical Biochemistry* (1999) **269** 337-341. PMID: 10222007
38. Prinsloo A.. **The relationship between consumer acceptability and descriptive sensory attributes of Cheddar cheese, with special reference to free choice profiling**. (2007)
39. Quigley L., O'Sullivan O., Beresford T. P., Ross R. P., Fitzgerald G. F., Cotter P. D.. **Molecular approaches to analysing the microbial composition of raw milk and raw milk cheese**. *International Journal of Food Microbiology* (2011) **150** 81-94. PMID: 21868118
40. Qureshi T. M., Amjad A., Nadeem M., Murtaza M. A., Munir M.. **Antioxidant potential of a soft cheese (paneer) supplemented with the extracts of date (Phoenix dactylifera L.) cultivars and its whey**. *Asian‐Australasian Journal of Animal Sciences* (2019) **32** 1591-1602. PMID: 31011003
41. Ranjha M. M. A., Irfan S., Lorenzo J. M., Shafique B., Kanwal R., Pateiro M., Arshad R. N., Wang L., Nayik G. A., Roobab U., Aadil R. M.. **Sonication, a potential technique for extraction of phytoconstituents: A systematic review**. *Processes* (2021) **9** 1406
42. Reis F. S., Stojković D., Soković M., Glamočlija J., Ćirić A., Barros L., Ferreira I. C.. **Chemical characterization of**. *Food Research International* (2012) **48** 620-626
43. Shahidi F., Naczk M.. *Phenolics in food and nutraceuticals* (2003)
44. Shan B., Cai Y. Z., Brooks J. D., Corke H.. **Potential application of spice and herb extracts as natural preservatives in cheese**. *Journal of Medicinal Food* (2011) **14** 284-290. PMID: 21142945
45. Shaukat A., Nadeem M., Ranjha M. M. A. N., Teferra T. F., Rukh L., Sultan W.. **Effect of ripening and in vitro digestion on free amino acids and angiotensin I converting enzyme inhibitory (ACE‐I) potential of cow and Buffalo milk cheddar cheeses**. *International Journal of Food Properties* (2022) **25** 948-959
46. Shekhar S., Kanawjia S. K., Khetra Y.. **Effect of milk fat replacement by olive oil on quality attributes of Quarg type cheese during storage**. *Indian Journal of Dairy Science* (2015) **68** 435-441
47. Singh P. K., Kumar S., Bhat Z. F., Kumar P.. **Effect of sorghum bicolour and clove oil on the quality characteristics and storage quality of aerobically packaged chevon cutlets**. *Nutrition and Food Science* (2015) **45** 145-163
48. Solhi P., Azadmard‐Damirchi S., Hesari J., Hamishehkar H.. **Production of the processed cheese containing tomato powder and evaluation of its rheological, chemical and sensory characteristics**. *Journal of Food Science and Technology* (2020a) **57** 2198-2205. PMID: 32431346
49. Solhi P., Azadmard‐Damirchi S., Hesari J., Hamishehkar H.. **Effect of fortification with asparagus powder on the qualitative properties of processed cheese**. *International Journal of Dairy Technology* (2020b) **73** 226-233
50. Sołowiej B., Cheung I. W., Li‐Chan E. C.. **Texture, rheology and meltability of processed cheese analogues prepared using rennet or acid casein with or without added whey proteins**. *International Dairy Journal* (2014) **37** 87-94
51. Weiss J., Ahmad T., Zhang C., Zhang H.. **A review of recent progress on high internal‐phase Pickering emulsions in food science**. *Trends in Food Science and Technology* (2020) **106** 91-103
52. Zhishen J., Mengcheng T., Jianming W.. **The determination of flavonoid contents in mulberry and their scavenging effects on superoxide radicals**. *Food Chemistry* (1999) **64** 555-559
|
---
title: 'Skin Temperature Circadian Rhythms and Dysautonomia in Myalgic Encephalomyelitis/Chronic
Fatigue Syndrome: The Role of Endothelin-1 in the Vascular Tone Dysregulation'
authors:
- Trinitat Cambras
- Maria Fernanda Zerón-Rugerio
- Antoni Díez-Noguera
- Maria Cleofé Zaragozá
- Joan Carles Domingo
- Ramon Sanmartin-Sentañes
- Jose Alegre-Martin
- Jesus Castro-Marrero
journal: International Journal of Molecular Sciences
year: 2023
pmcid: PMC10003028
doi: 10.3390/ijms24054835
license: CC BY 4.0
---
# Skin Temperature Circadian Rhythms and Dysautonomia in Myalgic Encephalomyelitis/Chronic Fatigue Syndrome: The Role of Endothelin-1 in the Vascular Tone Dysregulation
## Abstract
There is accumulating evidence of autonomic dysfunction in myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS); however, little is known about its association with circadian rhythms and endothelial dysfunction. This study aimed to explore the autonomic responses through an orthostatic test and analysis of the peripheral skin temperature variations and vascular endothelium state in ME/CFS patients. Sixty-seven adult female ME/CFS patients and 48 healthy controls were enrolled. Demographic and clinical characteristics were assessed using validated self-reported outcome measures. Postural changes in blood pressure, heart rate, and wrist temperature were recorded during the orthostatic test. Actigraphy during one week was used to determine the 24-h profile of peripheral temperature and activity. Circulating endothelial biomarkers were measured as indicators of endothelial functioning. Results showed that ME/CFS patients presented higher blood pressure and heart rate values than healthy controls in the supine and standing position ($p \leq 0.05$ for both), and also a higher amplitude of the activity rhythm ($p \leq 0.01$). Circulating levels of endothelin-1 (ET-1) and vascular cell adhesion molecule-1 (VCAM-1) were significantly higher in ME/CFS ($p \leq 0.05$). In ME/CFS, ET-1 levels were associated with the stability of the temperature rhythm ($p \leq 0.01$), and also with the self-reported questionnaires ($p \leq 0.001$). This suggests that ME/CFS patients exhibited modifications in circadian rhythm and hemodynamic measures, which are associated with endothelial biomarkers (ET-1 and VCAM-1). Future investigation in this area is needed to assess dysautonomia and vascular tone abnormalities, which may provide potential therapeutic targets for ME/CFS.
## 1. Introduction
Myalgic encephalomyelitis, also known as chronic fatigue syndrome (ME/CFS), is a debilitating multifaceted disorder that affects more than 50 million people worldwide. ME/CFS is a chronic condition that predominantly affects women [1]. The etiology of ME/CFS remains unknown, but it appears to be multifactorial, with immunogenetic and environmental factors influencing disease onset [2]. ME/CFS is characterized by debilitating post-exertional fatigue, neurocognitive impairments, autonomic dysfunction, and nonrestorative sleep, causing a marked reduction in daily activities, especially in women [3,4]. Currently, there are no reliable diagnostic markers for ME/CFS, nor are any FDA-approved disease-modifying drugs available [5].
Previous reports on autonomic dysfunction in ME/CFS suggest an imbalance between the sympathetic and parasympathetic nervous systems, with decreased parasympathetic tone and increased sympathetic output [6]. Autonomic symptoms are highly prevalent in ME/CFS and include dizziness, orthostatic hypotension, palpitations, lipotimia, hot flashes, constipation, and night sweating [7]. Moreover, hemodynamic determinants related to blood pressure (systolic pressure, diastolic volume, cardiac output, heart rate variability, arterial stiffness, cerebral blood flow, etc.) have also been described to be disturbed in ME/CFS [8,9,10]. Dysregulation of the autonomic nervous system also leads to an impaired vasomotor response in ME/CFS, such as postural orthostatic tachycardia syndrome (POTS). However, there are some discrepancies in the prevalence of POTS, ranging from $23.7\%$ in ME/CFS compared with only $4\%$ in healthy controls [11] to $5.7\%$ in ME/CFS, without differences with respect to non-ME/CFS individuals [12]. A recent review paper noted that circadian rhythm disruptions (sleep activity, insomnia, cognition problems, energy disturbances, impaired thermoregulation and dysautonomia) and cytokine profiling (mainly TGF-β) may be implicated in ME/CFS and long COVID. However, to date, no association has been established between peripheral skin temperature circadian rhythms, circulating endothelial biomarkers, and dysautonomia in ME/CFS [13]. The circadian system, whose master clock is the hypothalamic suprachiasmatic nuclei, regulates the manifestations of the circadian rhythms, including sleep schedules. The master clock regulates peripheral clocks and generates rhythms throughout the organism by means of daily variations in hormones, temperature rhythm, and regulation of the balance between the sympathetic nervous system, which peaks during the day, and the parasympathetic, which predominates at night [14,15,16].
Circadian alterations in peripheral temperature have been associated with dysautonomia in ME/CFS, reflecting alterations in the vasoconstriction/vasodilation process [13]. The peripheral temperature increases at night due to vessel dilation, in order to favor sleep, while alterations in this process have been related to difficulty in sleep onset [17]. However, vasoconstriction is due not only to the sympathetic effect on the vessels, but also to the regulation of local vessels by endothelial function, which has been reported to be altered in ME/CFS [18,19,20].
Endothelin-1 (ET-1), a strong endothelial vasoconstrictor, has been related to the regulation of the circadian rhythms [21], and high circulating ET-1 levels have recently been reported in ME/CFS [19]; thus we hypothesized that there would be an alteration of the vasoconstrictor/vasodilation process, which might be reflected in temperature variations, and would indicate dysautonomia in ME/CFS.
In this study, we aimed to explore the autonomic responses of ME/CFS patients, through the study of hemodynamic variables and thermoregulation, and their association with the endothelium state. Thus, we tested the utility of a passive standing test (10-min NASA lean test, NLT) [22] to assess orthostatic intolerance (OI) and also we investigated peripheral temperature changes in the study participants by (a) measuring the wrist temperature rhythms through actigraphy (i.e., the temperature variation between day and night) and (b) recording the skin temperature changes when participants changed from a supine to standing position during the NLT. Moreover, we also explored the functioning endothelial status through the measurement of circulating endothelial biomarkers in the study participants.
## 2.1. Demographic and Clinical Characteristics of Study Population
Table 1 summarizes the demographic and clinical characteristics and the results of routine blood testing of the participants. The groups differed in terms of age ($$p \leq 0.016$$) and BMI ($p \leq 0.001$). In addition, ME/CFS patients showed higher levels of cholesterol, triglycerides (TG), and low-density lipoproteins (LDL) (all $p \leq 0.01$), and also of the hormone 17β-estradiol ($$p \leq 0.044$$), compared to healthy controls, although the values of this last variable were inside the normal range. We did not control for the menstrual cycle stage among participants; however, healthy volunteers were younger than ME/CFS patients ($37.5\%$ vs. $50.5\%$ were aged over 50). No statistically significant differences were reported for menopause between ME/CFS and healthy controls ($34\%$ vs. $31\%$). Thus, since there is a strong association of LDL, cholesterol, and TG with age ($p \leq 0.05$ for all), and also between TG and BMI ($p \leq 0.005$), the comparison of the other variables displayed in Table 2, and the further statistical tests, were always carried out adjusting for age and BMI.
Additionally, the groups differed in the self-reported outcome measures, with ME/CFS patients recording higher scores for fatigue severity, anxiety/depression symptoms, sleep quality, autonomic symptoms, and lower health-related quality of life ($p \leq 0.001$ for all).
## 2.2. Passive Standing Test
All participants finished the orthostatic test (10-min NLT). Blood pressure and HR values were significantly higher in ME/CFS than in healthy controls; however, no differences according to time of the day were observed. As for changes in hemodynamic variables, there were no differences according to groups (Table 2).
Seven ME/CFS patients (four in the morning and three in the afternoon) and eight healthy controls (six in the morning and two in the afternoon) had POTS. Moreover, four ME/CFS patients and one healthy control had OH; thus, $16\%$ of ME/CFS patients ($\frac{11}{67}$) and $13\%$ of healthy controls ($\frac{6}{48}$) had abnormal cardiovascular responses to position changes after NLT, with no differences between the two groups in terms of distribution. However, although no differences were found between groups in the postural autonomic responses, ME/CFS patients had higher values of BP and HR in both the supine and standing position, and in individuals with and without POTS ($p \leq 0.05$ in all ANOVA comparisons) (Figure 1a–c).
## 2.3. Postural Wrist Temperature Changes
Postural wrist temperature-related changes (WT) could only be measured in 42 ME/CFS patients (15 in the morning and 27 in the afternoon) and in 33 healthy controls (17 in the morning and 16 in the afternoon). Mean WT varied according to the time of the day, and was always higher in the afternoon (WT morning: 28.5 ± 0.37 °C and WT afternoon: 31.5 ± 0.27 °C), but no differences were found between the groups (Table 2). Wrist temperature rose significantly during the NLT ($p \leq 0.01$ in both ME/CFS and healthy controls), the increase being higher in the morning (mean value 2.35 ± 0.23 °C vs. 0.76 ± 0.22 °C in the afternoon), with no differences between ME/CFS patients and healthy controls.
Changes in WT during the NLT (ΔWT_NLT) were positively associated with nocturnal temperature (T_M5, $r = 0.430$; $p \leq 0.007$), and with variables related to the stability of the WT circadian rhythm, described in the Section 4 (T_R, $r = 0.417$; $$p \leq 0.012$$, and T_PV, $r = 0.409$; $$p \leq 0.014$$) in ME/CFS, but these associations were not present in controls. In addition, in patients, ΔWT_NLT was associated with hemodynamic variables such as SBP ($r = 0.350$; $$p \leq 0.031$$) and DBP ($r = 0.371$; $$p \leq 0.028$$) and negatively with the decrease in DBP during the NLT. In healthy controls, these associations were not observed. ΔWT_NLT was negatively associated with ET-1 levels both in controls (r = −0.676; $$p \leq 0.006$$) and in ME/CFS (r = −0.635; $$p \leq 0.034$$). However, it should be remembered that these two variables could only be measured together in a few individuals (15 ME/CFS and 17 healthy controls).
## 2.4. Wrist Temperature Rhythms and Motor Activity Measured by Actigraphy
Circadian rhythm shows similar profiles in the two groups (Figure 2a,b). However, once corrected for age and BMI, an ANOVA indicated that variables obtained from activity data differed between groups, but not those obtained from temperature data (Table 2). Specifically, maximum activity, mean daily activity, and the amplitude of the rhythm were lower in ME/CFS. Body mass index emerged as the most important factor for WT rhythm variables.
We also tested the association between the PSQI questionnaire (sleep alterations) and T_M5, since the nocturnal temperature increases with sleep. As expected, in controls, the correlation was negative, but in ME/CFS, it did not reach significance. Moreover, since both activity and WT circadian rhythms tend to be associated, we tested whether this association occurred in a similar way in both groups, finding that, in controls, the nocturnal temperature (T_M5) was associated with high daily motor activity ($r = 0.433$; $$p \leq 0.003$$), high rhythm amplitude ($r = 0.452$; $$p \leq 0.002$$), and stability of the activity rhythm ($r = 0.318$; $$p \leq 0.033$$); these associations were not found for ME/CFS.
## 2.5. Endothelial Function Biomarkers
As shown in Table 2, ME/CFS patients showed significantly higher levels of plasma ET-1 and VCAM-1 proteins than healthy controls ($p \leq 0.05$ adjusted for both age and BMI). However, no differences were found for the ICAM-1 protein. To relate endothelial dysfunction to the rest of the variables, regression models with ET-1, VCAM-1, or ICAM-1 as dependent variables were carried out considering the following variables as predictors: (a) those obtained from temperature rhythms, (b) those obtained from motor activity rhythms, and (c) those obtained from the NLT.
Table 3 shows the model of the selection process with the final coefficients and their statistical significance. Specifically, our results showed the following.
Interestingly, VCAM-1 and ICAM-1 were not associated with the temperature variables. Also of interest, ET-1 levels were positively associated with the self-reported outcome measures in ME/CFS, but not in healthy controls; the higher the ET-1 levels, the worse the symptomatology (Table 4).
## 2.6. Discriminant Analysis
Finally, we tested the association of the subjective perception of the autonomic symptoms with the objective variables. To do so, we carried out a stepwise regression analysis with the global COMPASS-31 score as a dependent variable, and the representative and non-redundant variables for each group as predictors: namely, the hemodynamic variables (DBP, HR, ΔSBP_3L, ΔDBP_3L, and ΔHR_3L), temperature circadian rhythm variables (amplitude, T_PV, and T_M5), and endothelial biomarkers (ET-1 and VCAM-1). Again, a stepwise linear regression model was applied and collinearity was analyzed (Table 5).
The results indicate that the predictors for the best model with COMPASS-31 as the dependent variable were DBP, ET-1, VCAM-1, T_PV, and T_M5. A MANOVA was followed up with discriminant analysis of these variables, which revealed significant differences between the two groups (Wilks’ lambda = 0.372; Chi-square = 53.95; $p \leq 0.0001$). In all, $91.5\%$ of participants of the original grouped cases ($96\%$ of ME/CFS patients and $85\%$ of healthy controls) were correctly classified.
## 3. Discussion
This work is the first to study indicators of autonomic dysfunction through variables related to the vasoconstriction/vasodilation process, also considering the subjective symptomatology of ME/CFS patients. The main outcome of this study is the association of circulating ET-1 levels with the self-reported frequency/severity of autonomic symptoms and WT-related variables, which suggests a potential role of endothelial dysfunction in ME/CFS [19]. Moreover, a discriminant analysis using blood pressure, endothelium biomarkers, and circadian variables allows a good classification of the participants into patient and control groups. This indicates that only a multi-functional approach could shed light on the complexity of autonomic symptoms in ME/CFS.
This study also addresses secondary questions such as the usefulness of NLT, the measurement of peripheral WT, and the analysis of alterations of circadian rhythms in ME/CFS. The inclusion of NLT in the diagnostic evaluation is to add objective values to the autonomic symptoms in ME/CFS patients, in addition to completing the COMPASS-31 and OGS measures. However, in agreement with Roerink et al. [ 12], we did not find differences in OI responses between groups; nor could we correlate the hemodynamic response with the symptom-related questionnaires. In any case, one should bear in mind that the diagnosis of POTS is not based on increases in HR alone, since other OI criteria should also be included [11]. Our results confirmed that ME/CFS patients had higher values of BP and HR than matched healthy controls, independently of the orthostatic symptoms, corroborating the idea that the ME/CFS condition may induce a circulatory decompensation [22]. Increased HR in the supine position has already been reported in ME/CFS and may reflect increased sympatho-adrenomedullary activity [22,23]. Although most ME/CFS patients were not hypertensive, the increase in BP could be related to arterial stiffness, which has also been described in ME/CFS and in turn has been associated with more sympathetic dominance [24,25].
To our knowledge, this is the first study to include the measurement of WT during the NLT. Postural skin temperature changes may be due to blood redistribution and may reflect the individual variability of vasomotor activity [26]. Essentially, the skin (primarily distal site) may serve as an ideal reservoir for blood redistribution due to changes in skin vascular tone [27]. With these considerations in mind, we assumed that the skin temperature would rise in the upright position, partially due to gravity, which favors the presence of blood in the lower part of the arm, but also due to the vasodilator response.
Although we did not find differences in skin temperature between groups during NLT, this variable was of interest since it may be associated with the amplitude of the WT circadian rhythm and with ET-1 concentrations and may suggest a different thermoregulatory response in ME/CFS. However, since postural changes in WT and ET-1 levels could only be measured together in a relatively low number of individuals, further studies are needed to analyze this issue in greater depth.
A previous study by our group found significant differences in the activity rhythm between ME/CFS and healthy controls, with ME/CFS showing less activity during the day, a lower amplitude, and more fragmentation [3]. In agreement with this earlier study, on this occasion, we did not find differences in the WT rhythm. Nevertheless, only WT circadian rhythm variables were predictors of soluble ET-1 levels, a finding that corroborates the association between these variables. Thus, the fact that the soluble ET-1 protein might be correlated with the variables of the WT circadian rhythm but not with variables of the motor activity circadian rhythm in ME/CFS indicates that ET-1 and skin temperature are both markers of vasoconstriction/vasodilation processes, which may be impaired in ME/CFS and may be related to the intrinsic endothelial structure of the blood vessel [28]. Thus, our study points to endothelial dysfunction as a prominent feature that may help to explain the autonomic symptoms in ME/CFS; in our view, measuring the skin temperature might provide useful information for the clinical evaluation in ME/CFS.
On the other hand, endothelial damage alters the balance between vasoconstrictor/vasodilation events and also expresses higher levels of adhesion molecules. In our cohort, VCAM-1 levels were also high in ME/CFS, suggesting endothelial inflammation; however, the levels did not correlate with the severity of the symptomatology. Interestingly, levels of the vasoconstrictor ET-1 marker were elevated in ME/CFS, as other authors have recently reported [19]; since this is the variable that best correlated with the results of the self-reported questionnaires, it could be an indicator of illness severity. However, one must take into account that ME/CFS patients also had high levels of cholesterol, triglycerides, and LDL which may have caused endothelial dysfunction and cardiovascular disease, and may play a crucial role in the pathogenesis of ME/CFS [29]. In addition, none of the ME/CFS patients were taking contraceptive pills; however, mean estrogen levels were higher in ME/CFS patients (menopause: $34\%$), although the mean age was higher in this group. Further clinical and experimental studies are required to assess the sex hormone imbalance and autonomic dysfunction in people with ME/CFS.
Interestingly, sleep perception in healthy controls (as reflected by PSQI scores) was associated with nocturnal WT, an objective variable that increased during sleep due to vasodilation. However, this association was not found in ME/CFS patients, who often report unrefreshing sleep. This may be due to the autonomous nervous system dysfunction in our ME/CFS cohort, or to a decrease in circulating levels of melatonin, a key regulatory hormone of body temperature rhythm and sleep, which also affects autonomic vagal activity. In fact, sleep disturbances have been associated with elevated BP and HR and with lower HR variability, indicating reduced parasympathetic activity at night [30]. Since sleep and the cardiovascular function are regulated by the networks of central nervous system nuclei, their simultaneous study is also important in determining further relationships between sleep–wake control circuitry and the pathways regulating the autonomic function [31].
## Limitations of the Study
The present study has some limitations that should be mentioned. First, the single-center, cross-sectional nature of the design prevents us from identifying causation. In addition, the relatively small sample size, the use of self-reported data, and the inclusion only of women with mild/moderate disease severity may mean that the results are not representative of the entire population. Further limitations include the lack of data on comorbid health conditions, menstrual cycle stage, metabolic syndrome factors, physical activity, lack of endothelial function assessment in the clinical setting, and other lifestyles that may independently explain the endothelial dysfunction in the participants. Finally, the analyses derived from ET-1 associations are exploratory and may not have been appropriately powered.
## 4.1. Study Participants
A single-center, prospective, cross-sectional case–control study of 67 consecutive females with ME/CFS and 48 non-fatigued healthy controls recruited at a single outpatient tertiary-referral center (ME/CFS clinical unit, Vall d’Hebron University Hospital, Barcelona, Spain) from October 2019 to March 2022 was conducted. Following the suggestion made in our previous study, data were not recorded during the summer [3]. After receiving verbal and written information on the study protocol, each study participant gave signed informed consent to participate prior to enrollment, which was approved by the Vall d’Hebron Hospital Institutional Review Board (reference number PR/AG $\frac{201}{2016}$). The study was carried out in accordance with the principles set in the declaration of Helsinki and with all the international literature on harmonization and good clinical practice guidelines.
Patients with ME/CFS were potentially eligible if they were female, aged ≥ 18 years, with a confirmed diagnosis by a specialist of ME/CFS according to the 2011 international consensus criteria [7]. Healthy control subjects were eligible if they were adult females and neither met the case criteria for ME/CFS nor reported orthostatic intolerance symptoms at the time of study inclusion. Healthy controls were recruited through word-of-mouth from the local community. None of the participants were taking contraceptive pills.
Participants were subjected to stringent exclusion criteria, as previously described by our group [3]. The major exclusion criteria were a previous or current diagnosis of an autoimmune disorder, multiple sclerosis, psychosis, major depression disorder, heart disease, hematological disorders, infectious diseases, sleep apnea or thyroid-related disorders; pregnancy or breast-feeding; smoking habit; strong hormone-related drugs; and fatigue-associated symptoms that did not conform to the ME/CFS case criteria used for this study. Demographic and clinical characteristics of the study population are displayed in Table 1.
## 4.2. Experimental Procedures
Thirty-two ME/CFS patients and 29 healthy controls came to the local hospital on Wednesday morning between 8 a.m. and 11 a.m. and another cohort (35 ME/CFS patients and 19 healthy controls) on Tuesday afternoon between 3 p.m. and 6 p.m. for a clinical assessment. Demographic and self-reported outcome measures were recorded, as briefly described below and detailed in our previous study [3]. In participants who attended in the morning, a blood sample was taken for routine biochemical analysis. All participants underwent the same orthostatic test protocol (10-min NLT) to evaluate orthostatic intolerance (see details below), and wrist temperature (WT) was also recorded using a temperature sensor (Thermochron iButton® DS1921H, San Jose, CA, USA) placed on the right wrist of each participant during the NLT procedures.
Participants were asked to wear an ambulatory actigraphy device (ActTrust®, Condor instruments, Sao Paulo, Brazil) on the wrist of the non-dominant arm over the radial artery continuously for seven days, except when showering or at the swimming pool. The same actigraphy was programmed to collect data on activity (arbitrary units), skin temperature (°C), and light intensity (lux) at one-minute intervals. The data were recorded and stored in the device’s memory for further analysis. Subjects returned to the hospital after one week to hand in the actigraphy device, and completed health-related questionnaires.
## 4.3. Measures
Participants were also asked to provide complete validated self-report questionnaires on their current health status one week after the first clinical assessment. Changes in fatigue perception (FIS-40) [32], sleep quality (PSQI) [33], anxiety/depression (HADS) [34], autonomic symptoms (COMPASS-31) [35], and health-related quality of life (SF-36) [36] were assessed as described in our previous study [3], except for the assessment of the frequency and severity and interference of orthostatic symptoms, which was performed using the orthostatic grading scale (OGS).
## 4.4. Orthostatic Grading Scale
The orthostatic grading scale (OGS) is a 5-item validated self-reported questionnaire designed to assess symptoms of orthostatic intolerance due to orthostatic hypotension. The five questions address the frequency/severity and interference of orthostatic symptoms in daily life activities. Respondents rate each item on a scale of 0 to 4. Adding the scores for the individual items produces an overall OGS score ranging from 0 (never or rarely orthostatic symptoms) to 20 (maximum orthostatic symptoms). Higher scores indicate greater severity of autonomic dysfunction [37].
## 4.5. Assessment of Cardiovascular Autonomic Function
Autonomic response was evaluated using a passive standing test (10-min NASA lean test, NLT), a simple and well-established non-invasive procedure used to assess impaired cardiovascular compensatory responses to standing in ME/CFS. The NLT classifies OI phenotypes as orthostatic hypotension (OH) and postural orthostatic tachycardia syndrome (POTS), by measuring hemodynamic parameters, blood pressure (BP), and heart rate (HR) for both clinical and research purposes. The test was conducted in a consistent manner by the same examiner in the morning or in the afternoon, in a quiet room with an average relative temperature of 22.1 ± 1.2 °C and humidity of 55 ± $5.8\%$.
The participants were first asked to lie down for five minutes and then to stand and lean against a wall, with heels 6–8 inches away from the wall. An automated BP cuff with a monitor (Beurer BM-26, Beurer GmbH & Co., Ulm, Germany) was placed on the left arm, recording the systolic blood pressure (SBP) and diastolic blood pressure (DBP) and HR at 1-min intervals, and, simultaneously, a temperature sensor (Thermochron iButton® DS1921H, San Jose, CA, USA) was placed on the right wrist to record the peripheral temperature changes throughout the orthostatic test. SBP, DBP, and HR were recorded every minute for the two last minutes in the supine position and during the full 10-min after attaining the upright position. Throughout the recording, participants were asked to remain still, and any talking or movement was discouraged, except for reporting any symptoms of concern.
The NLT was stopped early at the request of the subject, or in the event of severe pre-syncope [22]. After 10 min upright, each participant was asked about the frequency/severity and impact of orthostatic symptoms on a 5-item OGS score [37].
## 4.6. Concomitant Medication
In order to maintain patients’ functional status, usual medication was not withdrawn during the study. However, information on current medication use was collected from the medical records. Thirty-nine patients ($58\%$) were taking non-steroidal anti-inflammatory drugs (ibuprofen, celecoxib for generalized pain), twenty-six ($39\%$) serotonin-norepinephrine reuptake inhibitors (duloxetine for neuropathic pain), twelve ($18\%$) anticonvulsants (gabapentin, pregabalin for chronic neuropathic pain), thirteen ($19\%$) analgesics (paracetamol), and thirty-four ($51\%$) anxiolytics (alprazolam, quetiapine for generalized anxiety disorder). Only seven patients ($10\%$) were taking antihypertensive drugs, and none on the day of the NLT. None of the healthy controls were taking any medication.
## 4.7. Hemodynamic Definitions Recorded during the Orthostatic Test
Criteria for OI were based on the 2021 expert consensus statement and guidelines on the definition of OH and POTS as follows: (a) orthostatic hypotension was defined as a decrease in SBP of ≥20 mmHg, or a decrease in DBP of ≥10 mmHg in the first three minutes standing compared with resting supine values; and (b) POTS was defined as either an increase in HR ≥ 30 bpm and/or a current HR ≥ 120 bpm based on the average of the last three minutes standing [38,39]. Classification of OI (OH and POTS) during the 10-min NLT was quantified as the difference between supine and standing hemodynamic changes (ΔBP and ΔHR), as previously described [22]. For the purposes of this study, SBP/DBP and HR were used as raw values recorded during the NLT. For correlation analysis and grouping comparison, we calculated the mean values during the last two minutes in the supine position (sp), and the mean values during first three minutes standing (3F), and mean values of minutes 8–10 standing (last three minutes, 3L). The changes in these variables compared with the values in the supine position were also calculated (Δ3L or Δ3L).
## 4.8. Actigraphy Analysis
Skin temperature and activity data obtained with the actigraphy device were analyzed with “El-temps, version 314”, an integrated package for chronobiological analysis (Prof. A. Díez-Noguera, University of Barcelona, Spain; https://www.el-temps.com) (accessed on 15 September 2022). The rhythmic variables were determined by adjusting the data to a 24-h co-sinusoidal curve. Thus, data of each variable (skin temperature and activity) were calculated: the mean 24-h value (MESOR), the acrophase (time of the day when the maximum value of the variable occurs), and the amplitude of an adjusted 24-h sinusoidal rhythm. Non-parametric circadian analysis was also performed as previously described [40]: M10 (or M5 in the case of WT) and L5 (or L10 in the case of WT) denoted the mean temperature in the 10 (or 5) consecutive hours with highest values (night temperature or activity during the day) and the 10 (or 5) hours with lowest values (day temperature or night activity values), respectively. Note that activity increases during the day and the skin temperature during the night. In addition, the intra-daily variability (IV), the stability of the rhythm measured by the grouping of the acrophases (Rayleigh’s test, R), and the percentage of variance explained by the 24-h rhythm (PV) were also calculated. To facilitate comprehension, when these variables were obtained from temperature data, the name of the variables starts with T, and when obtained from activity, they start with A.
## 4.9. Blood Sampling and Processing
A total of twenty milliliters of fasting blood samples from each participant were collected into K2EDTA anticoagulant tubes (BD vacutainer, Becton, Dickinson and Company, ON, Canada) from an antecubital vein with a 19-gauge needle without venous stasis. One tube was transported to the local core laboratory for the assessment of routine blood tests, including a comprehensive metabolic panel. All other blood samples were immediately centrifuged at 2500 rpm for 15 min at 4 °C (Thermo Scientific, Waltham, MA, USA) and then supernatants were collected and stored in aliquots at −80 °C until assayed. No sample was thawed more than twice. Repeated samples from each participant were measured in the same analytical batch.
## 4.10. Measurement of Endothelial Biomarkers
Circulating levels of soluble ET-1, vascular cell adhesion molecule-1 (VCAM-1), and intracellular adhesion molecule-1 (ICAM-1) were measured as indicators of endothelial functioning status. Plasma concentrations of ET-1 (cat n° DET-100), VCAM-1 (cat n° DVC00), and ICAM-1 (cat n° DCIM00) proteins were assayed in each participant using commercially available ELISA kits according to the manufacturer’s instruction manual (Quantikine R&D Systems, Minneapolis, MN, USA), using a Synergy™ H1M, hybrid multi-mode microplate reader (BioTek Instruments, Inc., Winooski, VT, USA) at O.D. 450 nm. Results were analyzed by comparison with standard calibration curves in each well, and are presented as averages of two duplicated samples.
## 4.11. Statistical Analysis
Data were tested for normality and homogeneity of variance with the Shapiro–Wilk and Levene tests, respectively. Data are presented as mean ± standard error of means (SEM) for continuous variables. Statistically significant differences for parametric variables were tested by means of one-way ANOVA. Since the NLT was conducted during the morning in some participants and during the afternoon in others, in the analysis of NLT variables, the time of the day was also included as an independent factor. In cases of non-parametric variables such as those obtained in the questionnaires, the Mann–Whitney U test was assayed to test differences between groups. Meanwhile, differences in categorical variables were analyzed using Fisher’s exact test. Pearson’s correlation analyses were used in paired data to evaluate the associations between the different variables. For each variable studied, paired data with missing values were excluded from the analysis. In addition, stepwise linear regression models with backwards elimination were performed to select significant predictors of endothelial biomarkers (ET-1, VCAM-1, or ICAM-1). Subsequently, we tested the interaction between the autonomic symptoms assessed by COMPASS-31 and the significant predictors of endothelial biomarkers using stepwise linear regression models. Finally, we conducted a discriminant analysis using the variables obtained in the last analysis to evaluate which were the variables that could reliably classify the subjects into the two groups. Univariate F-tests were then calculated to determine the importance of each independent variable in forming the discriminant functions. Examining the Wilk’s lambda values for each of the predictors revealed how important the independent variable was to the discriminant function, with smaller values representing greater importance. Data were analyzed using IBM SPSS Statistics for Windows, version 27.0 (IBM Corp., Armonk, NY, USA). All analyses were adjusted for age and BMI. p-values ≤ 0.05 was considered statistically significant.
## 5. Conclusions
This is the first study reporting indicators of the vasoconstriction and vasodilation process, such as peripheral skin temperature variations and vascular endothelium function biomarkers, considering all participants’ self-reported symptomatology of ME/CFS. Taken together, future studies in this area should aim to provide a better characterization of the underlying mechanisms of dysautonomia and endothelial dysfunction, which may contribute to the development of new biological targets for potential therapeutic interventions in ME/CFS.
## References
1. Nacul L., Authier F.J., Scheibenbogen C., Lorusso L., Helland I.B., Martin J.A., Sirbu C.A., Mengshoel A.M., Polo O., Behrends U.. **European Network on Myalgic Encephalomyelitis/Chronic Fatigue Syndrome (EUROMENE): Expert Consensus on the Diagnosis, Service Provision, and Care of People with ME/CFS in Europe**. *Medicina* (2021) **57**. DOI: 10.3390/medicina57050510
2. Castro-Marrero J., Saez-Francàs N., Santillo D., Alegre J.. **Treatment and management of chronic fatigue syndrome/myalgic encephalomyelitis: All roads lead to Rome**. *Br. J. Pharmacol.* (2017) **174** 345-369. DOI: 10.1111/bph.13702
3. Cambras T., Castro-Marrero J., Zaragoza M.C., Diez-Noguera A., Alegre J.. **Circadian rhythm abnormalities and autonomic dysfunction in patients with Chronic Fatigue Syndrome/Myalgic Encephalomyelitis**. *PLoS ONE* (2018) **13**. DOI: 10.1371/journal.pone.0198106
4. Castro-Marrero J., Zaragoza M.C., Gonzalez-Garcia S., Aliste L., Saez-Francàs N., Romero O., Ferré A., Fernandez de Sevilla T., Alegre J.. **Poor self-reported sleep quality and health-related quality of life in patients with chronic fatigue syndrome/myalgic encephalomyelitis**. *J. Sleep Res.* (2018) **27** e12703. DOI: 10.1111/jsr.12703
5. Cortes Rivera M., Mastronardi C., Silva-Aldana C.T., Arcos-Burgos M., Lidbury B.A.. **Myalgic Encephalomyelitis/Chronic Fatigue Syndrome: A Comprehensive Review**. *Diagnostics* (2019) **9**. DOI: 10.3390/diagnostics9030091
6. Tanaka M., Tajima S., Mizuno K., Ishii A., Konishi Y., Mike T., Watanabe Y.. **Frontier studies on fatigue, autonomic nerve dysfunction, and sleep-rhythm disorder**. *J. Physiol. Sci.* (2015) **65** 483-498. DOI: 10.1007/s12576-015-0399-y
7. Carruthers B.M., van de Sande M.I., De Meirleir K.L., Klimas N.G., Broderick G., Mitchell T., Staines D., Powles A.C., Speight N., Vallings R.. **Myalgic encephalomyelitis: International Consensus Criteria**. *J. Intern. Med.* (2011) **270** 327-338. DOI: 10.1111/j.1365-2796.2011.02428.x
8. Nelson M.J., Bahl J.S., Buckley J.D., Thomson R.L., Davison K.. **Evidence of altered cardiac autonomic regulation in myalgic encephalomyelitis/chronic fatigue syndrome: A systematic review and meta-analysis**. *Medicine* (2019) **98** e17600. DOI: 10.1097/MD.0000000000017600
9. Natelson B.H., Brunjes D.L., Mancini D.. **Chronic Fatigue Syndrome and Cardiovascular Disease: JACC State-of-the-Art Review**. *J. Am. Coll. Cardiol.* (2021) **78** 1056-1067. DOI: 10.1016/j.jacc.2021.06.045
10. Natelson B.H., Lin J.S., Blate M., Khan S., Chen Y., Unger E.R.. **Physiological assessment of orthostatic intolerance in chronic fatigue syndrome**. *J. Transl. Med.* (2022) **20** 95. DOI: 10.1186/s12967-022-03289-8
11. Clayton E.W.. **Beyond myalgic encephalomyelitis/chronic fatigue syndrome: An IOM report on redefining an illness**. *JAMA* (2015) **313** 1101-1102. DOI: 10.1001/jama.2015.1346
12. Roerink M.E., Lenders J.W., Schmits I.C., Pistorius A.M., Smit J.W., Knoop H., van der Meer J.W.. **Postural orthostatic tachycardia is not a useful diagnostic marker for chronic fatigue syndrome**. *J. Intern. Med.* (2017) **281** 179-188. DOI: 10.1111/joim.12564
13. McCarthy M.J.. **Circadian rhythm disruption in Myalgic Encephalomyelitis/Chronic Fatigue Syndrome: Implications for the post-acute sequelae of COVID-19**. *Brain Behav. Immun. Health* (2022) **20** 100412. DOI: 10.1016/j.bbih.2022.100412
14. Buijs R.M., Escobar C., Swaab D.F.. **The circadian system and the balance of the autonomic nervous system**. *Handb. Clin. Neurol.* (2013) **117** 173-191. DOI: 10.1016/B978-0-444-53491-0.00015-8
15. Riganello F., Prada V., Soddu A., di Perri C., Sannita W.G.. **Circadian Rhythms and Measures of CNS/Autonomic Interaction**. *Int. J. Environ. Res. Public Health* (2019) **16**. DOI: 10.3390/ijerph16132336
16. Buijs R.M., Soto Tinoco E.C., Hurtado Alvarado G., Escobar C.. **The circadian system: From clocks to physiology**. *Handb. Clin. Neurol.* (2021) **179** 233-247. DOI: 10.1016/B978-0-12-819975-6.00013-3
17. Krauchi K., Wirz-Justice A.. **Circadian clues to sleep onset mechanisms**. *Neuropsychopharmacology* (2001) **25** S92-S96. DOI: 10.1016/S0893-133X(01)00315-3
18. Kennedy G., Spence V., Khan F., Belch J.J.. **Plasma endothelin-1 levels in chronic fatigue syndrome**. *Rheumatology* (2004) **43** 252-253. DOI: 10.1093/rheumatology/keg462
19. Haffke M., Freitag H., Rudolf G., Seifert M., Doehner W., Scherbakov N., Hanitsch L., Wittke K., Bauer S., Konietschke F.. **Endothelial dysfunction and altered endothelial biomarkers in patients with post-COVID-19 syndrome and chronic fatigue syndrome (ME/CFS)**. *J. Transl. Med.* (2022) **20** 138. DOI: 10.1186/s12967-022-03346-2
20. Paschos G.K., FitzGerald G.A.. **Circadian clocks and vascular function**. *Circ. Res.* (2010) **106** 833-841. DOI: 10.1161/CIRCRESAHA.109.211706
21. Douma L.G., Barral D., Gumz M.L.. **Interplay of the Circadian Clock and Endothelin System**. *Physiology* (2021) **36** 35-43. DOI: 10.1152/physiol.00021.2020
22. Lee J., Vernon S.D., Jeys P., Ali W., Campos A., Unutmaz D., Yellman B., Bateman L.. **Hemodynamics during the 10-minute NASA Lean Test: Evidence of circulatory decompensation in a subset of ME/CFS patients**. *J. Transl. Med.* (2020) **18** 314. DOI: 10.1186/s12967-020-02481-y
23. Timmers H.J., Wieling W., Soetekouw P.M., Bleijenberg G., Van Der Meer J.W., Lenders J.W.. **Hemodynamic and neurohumoral responses to head-up tilt in patients with chronic fatigue syndrome**. *Clin. Auton. Res.* (2002) **12** 273-280. DOI: 10.1007/s10286-002-0014-1
24. Allen J., Murray A., Di Maria C., Newton J.L.. **Chronic fatigue syndrome and impaired peripheral pulse characteristics on orthostasis--a new potential diagnostic biomarker**. *Physiol. Meas.* (2012) **33** 231-241. DOI: 10.1088/0967-3334/33/2/231
25. Slomko J., Estevez-Lopez F., Kujawski S., Zawadka-Kunikowska M., Tafil-Klawe M., Klawe J.J., Morten K.J., Szrajda J., Murovska M., Newton J.L.. **Autonomic Phenotypes in Chronic Fatigue Syndrome (CFS) Are Associated with Illness Severity: A Cluster Analysis**. *J. Clin. Med.* (2020) **9**. DOI: 10.3390/jcm9082531
26. Shilco P., Roitblat Y., Buchris N., Hanai J., Cohensedgh S., Frig-Levinson E., Burger J., Shterenshis M.. **Normative surface skin temperature changes due to blood redistribution: A prospective study**. *J. Therm. Biol.* (2019) **80** 82-88. DOI: 10.1016/j.jtherbio.2019.01.009
27. Krauchi K., Gompper B., Hauenstein D., Flammer J., Pfluger M., Studerus E., Schotzau A., Orgul S.. **Diurnal blood pressure variations are associated with changes in distal-proximal skin temperature gradient**. *Chronobiol. Int.* (2012) **29** 1273-1283. DOI: 10.3109/07420528.2012.719961
28. Suganya N., Bhakkiyalakshmi E., Sarada D.V., Ramkumar K.M.. **Reversibility of endothelial dysfunction in diabetes: Role of polyphenols**. *Br. J. Nutr.* (2016) **116** 223-246. DOI: 10.1017/S0007114516001884
29. Maloney E.M., Boneva R.S., Lin J.M., Reeves W.C.. **Chronic fatigue syndrome is associated with metabolic syndrome: Results from a case-control study in Georgia**. *Metabolism* (2010) **59** 1351-1357. DOI: 10.1016/j.metabol.2009.12.019
30. Jarrin D.C., Ivers H., Lamy M., Chen I.Y., Harvey A.G., Morin C.M.. **Cardiovascular autonomic dysfunction in insomnia patients with objective short sleep duration**. *J. Sleep Res.* (2018) **27** e12663. DOI: 10.1111/jsr.12663
31. Fink A.M., Bronas U.G., Calik M.W.. **Autonomic regulation during sleep and wakefulness: A review with implications for defining the pathophysiology of neurological disorders**. *Clin. Auton. Res.* (2018) **28** 509-518. DOI: 10.1007/s10286-018-0560-9
32. Fisk J.D., Ritvo P.G., Ross L., Haase D.A., Marrie T.J., Schlech W.F.. **Measuring the functional impact of fatigue: Initial validation of the fatigue impact scale**. *Clin. Infect. Dis.* (1994) **18** S79-S83. DOI: 10.1093/clinids/18.Supplement_1.S79
33. Buysse D.J., Reynolds C.F., Monk T.H., Berman S.R., Kupfer D.J.. **The Pittsburgh Sleep Quality Index: A new instrument for psychiatric practice and research**. *Psychiatry Res.* (1989) **28** 193-213. DOI: 10.1016/0165-1781(89)90047-4
34. Herrero M.J., Blanch J., Peri J.M., De Pablo J., Pintor L., Bulbena A.. **A validation study of the hospital anxiety and depression scale (HADS) in a Spanish population**. *Gen. Hosp. Psychiatry* (2003) **25** 277-283. DOI: 10.1016/S0163-8343(03)00043-4
35. Sletten D.M., Suarez G.A., Low P.A., Mandrekar J., Singer W.. **COMPASS 31: A refined and abbreviated Composite Autonomic Symptom Score**. *Mayo. Clin. Proc.* (2012) **87** 1196-1201. DOI: 10.1016/j.mayocp.2012.10.013
36. Alonso J., Prieto L., Antó J.M.. **The Spanish version of the SF-36 Health Survey (the SF-36 health questionnaire): An instrument for measuring clinical results**. *Med. Clin.* (1995) **104** 771-776
37. Schrezenmaier C., Gehrking J.A., Hines S.M., Low P.A., Benrud-Larson L.M., Sandroni P.. **Evaluation of orthostatic hypotension: Relationship of a new self-report instrument to laboratory-based measures**. *Mayo. Clin. Proc.* (2005) **80** 330-334. DOI: 10.4065/80.3.330
38. Vernino S., Bourne K.M., Stiles L.E., Grubb B.P., Fedorowski A., Stewart J.M., Arnold A.C., Pace L.A., Axelsson J., Boris J.R.. **Postural orthostatic tachycardia syndrome (POTS): State of the science and clinical care from a 2019 National Institutes of Health Expert Consensus Meeting—Part 1**. *Auton. Neurosci.* (2021) **235** 102828. DOI: 10.1016/j.autneu.2021.102828
39. Raj S.R., Bourne K.M., Stiles L.E., Miglis M.G., Cortez M.M., Miller A.J., Freeman R., Biaggioni I., Rowe P.C., Sheldon R.S.. **Postural orthostatic tachycardia syndrome (POTS): Priorities for POTS care and research from a 2019 National Institutes of Health Expert Consensus Meeting—Part 2**. *Auton. Neurosci.* (2021) **235** 102836. DOI: 10.1016/j.autneu.2021.102836
40. Van Someren E.J., Mirmiran M., Swaab D.F.. **Non-pharmacological treatment of sleep and wake disturbances in aging and Alzheimer’s disease: Chronobiological perspectives**. *Behav. Brain Res.* (1993) **57** 235-253. DOI: 10.1016/0166-4328(93)90140-L
|
---
title: Resveratrol Modulates Chemosensitisation to 5-FU via β1-Integrin/HIF-1α Axis
in CRC Tumor Microenvironment
authors:
- Aranka Brockmueller
- Sosmitha Girisa
- Ajaikumar B. Kunnumakkara
- Mehdi Shakibaei
journal: International Journal of Molecular Sciences
year: 2023
pmcid: PMC10003050
doi: 10.3390/ijms24054988
license: CC BY 4.0
---
# Resveratrol Modulates Chemosensitisation to 5-FU via β1-Integrin/HIF-1α Axis in CRC Tumor Microenvironment
## Abstract
Frequent development of resistance to chemotherapeutic agents such as 5-flourouracil (5-FU) complicates the treatment of advanced colorectal cancer (CRC). Resveratrol is able to utilize β1-integrin receptors, strongly expressed in CRC cells, to transmit and exert anti-carcinogenic signals, but whether it can also utilize these receptors to overcome 5-FU chemoresistance in CRC cells has not yet been investigated. Effects of β1-integrin knockdown on anti-cancer capabilities of resveratrol and 5-FU were investigated in HCT-116 and 5-FU-resistant HCT-116R CRC tumor microenvironment (TME) with 3D-alginate as well as monolayer cultures. Resveratrol increased CRC cell sensitivity to 5-FU by reducing TME-promoted vitality, proliferation, colony formation, invasion tendency and mesenchymal phenotype including pro-migration pseudopodia. Furthermore, resveratrol impaired CRC cells in favor of more effective utilization of 5-FU by down-regulating TME-induced inflammation (NF-kB), vascularisation (VEGF, HIF-1α) and cancer stem cell production (CD44, CD133, ALDH1), while up-regulating apoptosis (caspase-3) that was previously inhibited by TME. These anti-cancer mechanisms of resveratrol were largely abolished by antisense oligonucleotides against β1-integrin (β1-ASO) in both CRC cell lines, indicating the particular importance of β1-integrin receptors for the 5-FU-chemosensitising effect of resveratrol. Lastly, co-immunoprecipitation tests showed that resveratrol targets and modulates the TME-associated β1-integrin/HIF-1α signaling axis in CRC cells. Our results suggest for the first time the utility of the β1-integrin/HIF-1α signaling axis related to chemosensitization and overcoming chemoresistance to 5-FU in CRC cells by resveratrol, underlining its potential supportive applications in CRC treatment.
## 1. Introduction
Colorectal cancer (CRC) is defined as a malignant neoplasm of the colon or rectal epithelium, the treatment of which represents a major medical challenge worldwide [1]. Currently, in the United States of America, approximately 150,000 new cases of CRC are diagnosed annually [1]. In Germany, the number is more than 60,000, and despite numerous early detection measures and techniques, the relative 5-year survival rate in *Germany is* $63\%$ for men and $65\%$ for women [2]. After diagnosis, most patients receive chemotherapy containing first-line chemotherapeutic agents such as oxaliplatin and 5-flourouracil (5-FU) within the framework of the FOLFOX4 or FOLFOX6 therapy scheme [3,4,5,6].
Oxaliplatin is a platinum derivative chemotherapeutic agent, while 5-FU is one of the main components of the FOLFOX therapy schemes, a synthetic pyrimidine analogue that can be administered intravenously as a prodrug and exerts its effect via fluorinated nucleotides that are incorporated into the patient’s deoxyribonucleic acid (DNA) instead of the pyrimidine nucleoside thymidine [7,8]. It thus inhibits DNA replication and induces cell death in cancer cells [9]. However, in addition to the pronounced aggressiveness of CRC, recognizable by severe metastatic properties of the cells [10], it develops high resistance via activating multiple survival signaling pathways to the mono-target chemotherapeutic agents leading to complicated treatment process with lesser therapeutic outcome or success [6,7,9,11,12]. Moreover, due to the heterogeneity of this disease, it represents a major treatment hurdle, and thus, efforts are being made to explore the exact mechanism or pathways involved in the development of resistance, including dihydropyrimidine dehydrogenase, thymidylate synthase or thymidine phosphorylase signaling pathways [13,14,15,16,17,18,19]. At the same time, biomarkers are being sought that could indicate an unfavorable, chemoresistance-promoting course of FOLFOX therapy at an early stage, with complement compound 3 gene being suggested as an example [12]. Overcoming chemoresistance to mono-target therapies in CRC continues to be intensively explored, and an increasing amount of research is looking at combination treatment of standard anti-cancer drugs and herbal polyphenols [20]. This exciting research strategy is based on a broad, health-promoting spectrum of natural agents such as the blood-glucose-regulative action of chlorella [21] or vitamin C [22] and the anti-oxidative experience of miswak [23] or turmeric-components curcumin and calebin A [24].
Especially for the reduction of CRC-chemoresistance, potential is offered by the secondary plant compound, by the secondary plant compound resveratrol, is a well-known natural polyphenol which can be extracted from various grapes, berries and nuts [25]. Resveratrol has already been shown significant anti-inflammatory [26] and anti-tumor effects [27], especially by modulating the signaling pathways of the important pro-inflammatory nuclear factor ‘kappa-light-chain-enhancer’ of activated B-cells (NF-kB) and NF-kB-related gene cascades [28]. Consistent with the increased number of β1-integrins in CRC cells [29], which function as both cell-survival-protective adhesion and active signaling molecules, resveratrol specifically exploits these β1-integrin receptors in tumor cells by altering their expression pattern and using them as a gateway to transfer its anti-cancer effects and signals into the tumor cells [30]. Interestingly, it was previously described that resveratrol was even able to re-sensitise HCT-116R as well as SW480R CRC cells, which were already resistant to 5-FU by inducing the uptake of this chemotherapeutic agent, thereby overcoming 5-FU-resistance and inducing apoptosis in cancer cells, that would not have been targeted if treated with 5-FU alone [20]. Indeed, the development of chemoresistance is significantly influenced by cancer stem cells (CSC), which have a high renewal and differentiation potential [31].
Moreover, several biomarkers have already been used to detect CSC in HCT116 cells, and it is also known that CSC parameters can be down-regulated by resveratrol treatment in CRC cells [27]. Furthermore, the progression of CRC is also accompanied by an increase in angiogenesis factors, namely hypoxia-induced factors (HIF), as a reaction to hypoxic conditions in the tumor cells, which are responsible for the formation of a new vascular epithelium through the activation of vascular endothelial growth factor (VEGF) cascade [32]. In metastatic CRC, the intratumoral expression of HIF-1α is high as a particularly active subunit, and enables the invasive properties of CRC cells [33,34]. While there is initial evidence of resveratrol-binding HIF-1α restriction in HT-29 [35] and LoVo [36] colon carcinoma cells, thus exploring of the effects of resveratrol on HIF-1α signaling in HCT-116 and 5-FU-resistant HCT-116R CRC cells offers particular scientific interest.
In previous investigations, we have shown the role of β1-integrin receptors in the anti-viability, anti-proliferative and anti-invasive effects of resveratrol [10,30]. However, since it is not yet known whether β1-integrins are also involved in 5-FU-chemosensitisation by resveratrol, we have aimed to address this topic. In addition, we also wanted to shed light on the resveratrol/HIF-1α interaction in CRC cells and to find out whether this connection is influenced by the presence or absence of β1-integrins. Therefore, we chose a pre-clinical, animal-free 3D-alginate tumor microenvironment (TME) culture in vitro and compared the tumor-inhibitory property of resveratrol in two CRC cell lines, HCT-116 and HCT-116R, the later represents a 5-FU-resistant variant of HCT-116 CRC cells.
## 2.1. Resveratrol Increases 5-FU Sensitivity and Acts Anti-Carcinogenic via ß1-Integrin Receptors in CRC Cells
The ability of resveratrol and its related analogs to increase the susceptibility to the chemotherapeutic agent 5-FU as well as to enhance its effect in CRC cells has already been demonstrated by several research groups worldwide [20,37]. Based on our previous findings that resveratrol employs β1-integrin receptors to transfer its anti-tumor signals into CRC cells [10,30], we hypothesized a connection of resveratrol’s chemosensitisation via these β1-integrin receptors and cells were treated as follows: First, basal control (CRC cells without additives), TME control (CRC cells in multicellular milieu, without additives), CRC-TME treated with 2 nM 5-FU or 5 µM resveratrol, CRC-TME treated with 2 nM 5-FU and 5 µM resveratrol, CRC-TME treated with 0.5 µM β1-SO (control compound) or 0.5 µM β1-ASO (knockdown compound). Further, CRC-TME treated with 0.5 µM β1-SO or 0.5 µM β1-ASO was supplemented with 2 nM 5-FU or 5 µM resveratrol or with a combination of both substances. Therefore, four different evaluation methods were chosen and their results are described in more detail below.
## 2.1.1. Reduction of CRC Cell Vitality
Compared to the basal control (Ba. Co., without fibroblasts and T-lymphocytes), HCT-116 cells were significantly more viable in the pro-inflammatory TME containing fibroblasts and T-lymphocytes. This remained when β1-SO or β1-ASO were added to the TME, confirming that these substances alone had no relevant effect on HCT-116 cells in the TME. When HCT-116 cells were treated with resveratrol (5 µM), 5-FU (2 nM) or a combination thereof (2 nM 5-FU with 5 µM resveratrol), their survival-capacity decreased markedly and it also decreased when 5-FU was added to HCT-116 cells in the β1-SO-TME or β1-ASO-TME. Resveratrol, however, was able to exert its inhibitory effect in the β1-SO-TME, but not in the β1-ASO-TME (Figure 1A), suggesting that a knockdown of β1-integrin receptors does not affect 5-FU’s effect, but does predominantly restrict the anti-carcinogenic ability of resveratrol in HCT-116 CRC cells.
Then, HCT-116R cells, resistant to chemotherapeutic agent 5-FU, were investigated with the same treatments as HCT-116 cells. Besides the observation that more vital HCT-116R cells than non-resistant HCT-116 cells were measured in general, the main difference was the treatment failure of 5-FU in a TME containing HCT-116R cells and also in β1-SO-TME or β1-ASO-TME with HCT-116R cells. Surprisingly, an addition of resveratrol to TME or β1-SO-TME reduced the viability of HCT-116R cells (Figure 1B), while a knockdown with β1-ASO eliminated the anti-cancer potential and chemosensitising effect of resveratrol in these particularly combative CRC cells.
## 2.1.2. Reduction of CRC Cell Colony Formation
In TME, HCT-116 cells formed considerably more colonies (black arrows) than in the basal control (Ba. Co.), which was observed to be uninfluenced by an addition of β1-SO or β1-ASO. However, the treatment of these cells with resveratrol, 5-FU, combination of 5-FU and resveratrol, 5-FU and β1-SO or 5-FU and β1-ASO visibly limited the colony formation. Furthermore, resveratrol inhibited proliferative formation of CRC cell colonies in β1-SO-TME but not in β1-ASO-TME (Figure 2A(a); “alginate”). However, when HCT-116R cells were treated with the same agents or additives, the formation of more colonies (black arrows) was noticed compared to HCT-116 cells, while the ineffectiveness of 5-FU in inhibiting colonies was also observed in TME, β1-SO-TME as well as β1-ASO-TME treatments. Interestingly, resveratrol strongly reduced the colony-forming ability of HCT-116R cells in TME as well as in β1-SO, but not in combination with β1-ASO (Figure 2B(a); “alginate”). Overall, this proliferation investigation confirmed the preceding vitality evaluation.
## 2.1.3. Reduction of CRC Cell Invasion
By comparison with CRC cells in the basal control (Ba. Co.), the number of migrated HCT-116 colonies was higher in TME and also in the β1-SO-TME and β1-ASO-TME. When TME was treated with 5-FU, resveratrol, 5-FU and resveratrol, the invasion capacity of HCT-116 cells was distinctly reduced, similar to an addition of 5-FU to β1-SO-TME or β1-ASO-TME. Moreover, resveratrol-treatment weakened CRC colony settling in TME with β1-SO but not in TME with β1-ASO (Figure 2A(b); “T-blue”). Further extended to 5-FU-resistant CRC cells, smaller but much more migrated colonies settled from HCT-116R cells compared to HCT-116 cells. The same treatments showed the inefficacy of 5-FU to contain HCT-116R invasional property in TME and TME with β1-SO or β1-ASO. Noteworthy, addition of resveratrol to TME or β1-SO-TME averted migration as well as formation of HCT-116R cell colonies, however this effect of resveratrol was not observed in β1-ASO-TME treatment (Figure 2B(b); “T-blue”). These behavioural observations of CRC cell invasion confirmed the previously described MTT and colony formation assays that the β1-integrin receptor might play an important role in the resveratrol-sensitizing effect of 5-FU on CRC cell migration.
## 2.1.4. Reduction of CRC Cell’s Mesenchymal Phenotype
Both cell types, HCT-116 and HCT-116R, presented in the basal control with elongated cell bodies, planar surface with small pseudopodia on the surface and moderate cell-cell contact, representing a more epithelial morphology (Figure 3A(a,e)). However, in TME, where CRC cells have been influenced pro-inflammatory, HCT-116 as well as HCT-116R cells showed a distinctly mesenchymal shape, in that both the cell bodies and their extensions appeared rounded, developed many thick pseudopodia on the surface with active nucleus and inclined to emigrate (Figure 3A(b,f)). A treatment with 5-FU reduced this TME-induced change in HCT-116 cells without completely restoring the appearance of the basal control (Figure 3A(c)) and occasionally led to the generation of apoptosis, (Figure 3A(c)). In HCT-116R cells, an addition of 5-FU to the TME showed only a slight effect, so that the CRC cells remained morphologically similar to the TME control (Figure 3A(g)). Contrary to this, TME-fueled CRC cells clearly responded to a treatment with resveratrol, as single additive or combined with 5-FU. In both CRC cell lines, HCT-116 and HCT-116R, the cell bodies remained rather roundish to oval, but the round cell extensions needed for migration regressed (Figure 3A(d,h)) and partially re-extended and they had an epithelial shape without pseudopodia in the slightly less aggressive HCT-116 cells in resemblance to the basal control (Figure 3A(d)). Furthermore, the observation of numerous mitochondrial changes and apoptotic bodies, which were even more visible in HCT-116R cells than in HCT-116 cells (Figure 3A(d,h)), was remarkable when CRC cells were treated with resveratrol in combination with 5-FU. To explore the role of β1-integrin in resveratrol’s chemosensitising signaling, HCT-116 and HCT-116R cells in the TME were further subjected to treatment with the knockdown substance β1-ASO or the control substance β1-SO and along with resveratrol alone or combination of resveratrol and 5-FU (Figure 3B). While in β1-SO-TME resveratrol or resveratrol with 5-FU had an almost unrestricted effect of the CRC cells resulting in a smooth epithelial surface (Figure 3B(a,c,e,g)) and initiation of mitochondrial changes as well as apoptosis (Figure 3B(a,c,e,g)), where this effect was severely limited in β1-ASO treatment (Figure 3B(b,d,f,h)). Numerous TME-induced cell extensions remained, especially at combined treatment of β1-ASO and resveratrol (Figure 3B(b,f)), which was very similar to the TME control.
In the combination treatment of β1-ASO with resveratrol and 5-FU, comparatively, strong but somewhat less round cell extensions were observed (Figure 3B(d,h)). As a whole, resveratrol promotes the tendency to an epithelial-like phenotype in CRC cells, making them more susceptible to treatment with the chemotherapeutic 5-FU, and increases apoptosis initiation in HCT-116 and HCT-116R cells, which coincides with outlined results of vitality, proliferation and invasion assays. Summarising this assay, it remains to be noted that (a) HCT-116R cells were indeed predominantly resistant to treatment with 5-FU, (b) resveratrol was effective in both CRC cell types, alone or synergistically in combination with 5-FU and (c) the anti-viable, anti-proliferative, anti-invasive as well as anti-mesenchymal effect of resveratrol was largely cancelled out by the addition of β1-ASO. All in all, the results suggested an anti-carcinogenic and 5-FU-chemosensitising property of resveratrol in CRC cells at least in part via β1-integrin receptors.
## 2.2. HIF-1α Is Involved in Resveratrol-Promoted Chemosensitising CRC Cells to 5-FU
First of all, a slightly different morphology of the two cell lines should be noted, because while HCT-116R cells proliferated more and presented as many small roundish cells with a migratory mesenchymal character, the HCT-116 were in comparison somewhat larger and more epithelially spread out. Both, HCT-116 and HCT-116R were clearly HIF-1α-marked in the basal control (Ba. Co.) containing cell culture medium only. A resveratrol addition down-regulated the HIF-1α expression unambiguously (Figure 4A). This observation was reproduced also in TME composed of floating T-lymphocytes in the cell culture medium, fibroblast monolayers on the well-plate-bottom and CRC cells on glass coverslips. While the HIF-1α expression was strong in the TME control (TME), resveratrol treated HCT-116 and HCT-116R cells showed barely HIF-1α immunolabeling. Noteworthy, a treatment of the CRC cells with the chemotherapeutic agent 5-FU was ineffective in preventing HIF-1α expression, but the combined administration of 5-FU and resveratrol led to a down-regulation of HIF-1α, thus appearing HIF-1α to be a target of resveratrol but not of 5-FU (Figure 4A).
Furthermore, investigations of the mechanistic action of resveratrol in these cell lines confirmed the anti-HIF-1α efficacy of resveratrol even in the presence of the control substance β1-SO alone as well as in the presence of β1-SO and 5-FU in HCT-116 and HCT-116R cells (Figure 4B). Collectively, these results proposed an increased likelihood of HCT-116 and HCT-116R cells responding to 5-FU through resveratrol’s ability to make the CRC cells vulnerable through the β1-integrin/HIF-1α axis.
## 2.3. ß1-Integrin Participated in Resveratrol-Mediated Down-Regulation of NF-kB Activation and Related Gene End Products
Compatible with previous findings of elevated integrin values in CRC cells [38], a high β1-integrin level has been found in TME in the presence or absence of control β1-SO. But if β1-ASO was added to TME, the level of β1-integrin was down-regulated (Figure 5). All told, β1-integrin knockdown was successfully performed as transient transfection by oligonucleotides listed in Material and Methods.
For the extended examination of protein expression level, HCT-116 (Figure 6A) or HCT-116R (Figure 6B) were separated from alginate drops and immunoblotted by SDS-PAGE (Figure 6).
In both CRC cell lines, a sample verification was carried out by the uniform display of β-actin as loading control and the pan-NF-kB was equally represented as a vitality sign in all rehearsals (Figure 6A,B). Following a known, comprehensible signaling chain, the expression of phosphorylated NF-kB (p-NF-kB) as main inflammation parameter, vascularisation factor VEGF as well as CD44, CD133 and ALDH1 as cancer stem cell marker were comparable within each CRC cell line. In HCT-116 cells (Figure 6A), the expression of these markers was higher in TME-cultivated CRC cells than in the basal control and remained high when β1-SO or β1-ASO were added to the TME. With an addition of β1-SO or β1-ASO to 5-FU-treated HCT-116 cells in TME, the expression of parameters mentioned were comparative with the basal control level. Furthermore, a treatment of TME-HCT-116 cells with 5-FU, resveratrol or both agents in combination, significantly down-regulated inflammation (p-NF-kB), vascularisation (VEGF) and cancer stem cell (CD44, CD133, ALDH1) expression. However, remarkably, these anti-CRC effects of resveratrol were reversed by β1-integrin knockdown using β1-ASO, regardless of the presence or absence of 5-FU (Figure 6A). As another representative of vascularisation, HIF-1α level was investigated whereby a decisive difference became apparent. In contrast to resveratrol, 5-FU could not down-regulate HIF-1α and thus could not prevent the initiation of vascularisation. But interestingly, a dual treatment of resveratrol and 5-FU inhibited HIF-1α expression. Resveratrol’s significantly suppressed HIF-1α expression which was also observed remarkably in β1-SO-TME, but not in β1-ASO-TME, regardless of 5-FU’s presence or absence. In a further Western blot analysis on cleaved-caspase-3, this apoptosis marker was up-regulated in all HCT-116 cells in which resveratrol was able to unfold its effect freely, alone or combined with 5-FU. Accordingly, an increased caspase-3 level was noticed in HCT-116 and HCT-116R cells, treated with resveratrol alone or a resveratrol-5-FU combination in TME or β1-SO-TME, but not in a β1-integrin knockdown via β1-ASO in TME (Figure 6A). The key difference in the dynamic observation of 5-FU-resistant HCT-116R cells showed that 5-FU alone or in combination with β1-SO or β1-ASO had no significant anti-inflammatory, anti-vascularising as well as anti-stemness effect (Figure 6B). This was confirmed by the observation where the expression level of all these parameters were comparatively similar to the expression in the TME control of HCT-116R cells (Figure 6B). In contrast, resveratrol alone as well as in combined treatment with 5-FU induced a strong anti-tumor effect against these parameters, both in TME and β1-SO-TME, but not in β1-ASO-TME. In summary, resveratrol reduces inflammation, vascularisation, particularly by inhibiting HIF-1α, and suppresses cancer stem cell formation and increases apoptosis in both HCT-116 and HCT-116R cells, acting chemosensitising and synergistic agent in combination with chemotherapeutic drug, 5-FU at least proportionally via β1-integrin receptors.
## 2.4. Resveratrol Inhibits β1-Integrin/HIF-1a Axis in CRC Cells
The previous results suggested a functional molecular connection between the pathways of β1-integrin and master transcriptional regulator HIF-1α and to investigate this specifically, immunoprecipitation assay was chosen. For this purpose, HCT-116 (Figure 7A) and HCT-116R (Figure 7B) cells were cultured in alginate drops. After 10 days, CRC-samples were obtained, immunoprecipitated with anti-β1-integrin antibody and immunoblotted against HIF-1α to demonstrate the concatenation of both signaling pathways. Consistent with the known β1-integrin expression in the basal control [39], an expression of HIF-1α was analysed by densitometry. In comparison, the expression of HIF-1α in the TME control was markedly increased, consistent with the already shown high β1-integrin expression in Figure 5 and in agreement with the high HIF-1α expression in Figure 4A in both CRC cell lines. Resveratrol impressively suppressed the β1-integrin coupled HIF-1α expression in HCT-116 as well as HCT-116R cells. The uniform β-actin detection served as loading control. Overall, these results suggested for the first time an attenuation of TME-promoted β1-integrin/HIF-1α axis by resveratrol treatment, indicating the intracellular mode of action of resveratrol in inducing anti-tumor effect in CRC cells.
## 3. Discussion
Colorectal cancer management has benefited significantly over the past decade from the development of both target-based therapies and conventional chemotherapeutic substance, such as 5-FU, which have decisively increased the quality of patients’ lives and their life expectancies [40]. However, the performance of these drugs is seriously compromised by the emergence of resistance and recurrence mechanisms which are observed in more than $50\%$ of patients, in routine clinical practice [11]. Therefore, it is the need of the hour to design novel therapeutic compounds for the effective management of such resistant malignancies.
In the past, we demonstrated the importance of β1-integrin receptors in the anti-viability and anti-invasive action of resveratrol, as a natural chemopreventive compound on various CRC cell lines [30]. Pursuing this, the question was whether resveratrol would have a chemosensitising effect to 5-FU via β1-integrin receptors and related signaling pathways on CRC cell lines in a 3D alginate tumor microenvironment. The following core statements could be derived from the results which suggests that resveratrol enhances, at least in part through the use of β1-integrin receptors, (I) intensifies the effectiveness of 5-FU in CRC cells, (II) paves the way for 5-FU efficacy in therapy-resistant CRC cells, (III) triggers the epithelial phenotype, (IV) targets the vascularisation marker HIF-1α and (V) specifically inhibits the association of β1-integrin with HIF-1α in 5-FU-resistant and non-resistant CRC cells.
The results of our study showed that in all the methods used, a significant containment of CRC cells in vitro was observed by treatment with resveratrol in the presence of β1-integrin. Moreover, resveratrol treatment in CRC wells also significantly enhanced the effect of 5-FU when both agents are used in combination. This finding confirms earlier scientific suspicions and hypotheses of a chemosensitising potential of resveratrol in CRC to 5-FU treatment [20,37,41] and it can be assumed that a combination application could also lead to an optimisation of patient’s clinical therapy, for example through accelerated recovery or later during the process of development of resistance. Tumor cells, including CRC cells, are exposed to more severe environmental conditions such as cytokine storms related to inflammatory processes, increased oxidative and endoplasmic reticulum stress as well as hypoxia and changes in the local pH-value [42,43,44]. Such a complex situation was reproduced in our multicellular 3D model, simulating an advanced carcinogenic body situation without animal testing and, as in vivo, by habituation to this environment, the CRC cells become insensitive to environmental influences, including drugs.
A mesenchymal phenotype resulting from epithelial-mesenchymal transition (EMT) as consequence of a pole reversal is a hallmark of aggressive and therapy-resistant tumor cells. During the present study, we observed resveratrol’s ability to change the migratory, mesenchymal and pseudopodia-rich phenotype of CRC cells to a more localised epithelial phenotype thus lowering migratory as well as invasive attendance. In contrast to 5-FU, resveratrol also operated in chemoresistant cells via β1-integrin receptors, thus repressing EMT and paving the way for 5-FU to exert its effect in HCT-116R cells. This could be a potential key component to develop complementary treatment options to control invasive stages of cancer, what is particularly important as more than $50\%$ of patients endure metastases during the course of CRC disease [45] and these often lead to death from organ failure. These findings are in accordance with other studies that have shown that morphological changes and upregulation of intercellular junctions on the surface of cancer cells by resveratrol as a multitargeting component are strongly associated with its anti-malignant and anti-proliferation behaviour in tumor cells [20,46].
Rapidly growing cancerous mass and metastatic processes require a pronounced vascularisation and the angiogenic factor HIF-1α plays a crucial role in this process [47]. As a sensitive parameter, HIF-1α indicates milieu changes in CRC cells, generated in the event of oxygen deficiency, and induces new vessel formation [34] to increase an oxygen supply. In addition, up-regulated αvβ5-integrin or β1-integrin expression as well as NF-kB phosphorylation are associated with a strong HIF-1α increase in CRC cells [48,49]. Plant-derived polyphenols such as curcumin and its analogs could down-regulate HIF-1α expression by interrupting NF-kB phosphorylation in HCT-116 cells [50]. Moreover, HIF-1α has already been suggested as a potential target for resveratrol in various tumor types such as prostate and pancreatic cancers [51,52] or colon carcinoma [35,36], but to the best of our knowledge, the present work is the first study to demonstrate the potential of resveratrol to target HIF-1α in inducing cancer cell chemosensitisation in CRC. Thus, resveratrol could intervene the cancer cascades when the time for local intervention has already passed and the treatment possibilities in patients are limited, thereby increasing the possibilities of general healing chances for patients with advanced CRC.
Each decoding of cancer mechanisms leads to more precise therapy options, reaching the pathologically changed cells more directly and sparing the healthy cells. Thus, it is becoming increasingly interesting to use natural effects of plant ingredients as a co-treatment for various proliferative diseases. For example, inhibition of liver cancer cells by means of oxidative stress and DNA damage induced by safranel, an active component of the spice saffron, has been demonstrated [53]. Furthermore, the flavone quercetin supported significantly the anti-cancer competence of sorafenib (a standard drug against liver cancer) in vitro as well as in vivo by down-regulation of inflammation and up-regulation of apoptosis [54]. Moreover, photosensitive compounds of the plant Cichorium Pumilum reduced oxidative stress as well as the activation of estrogen receptors in female Sprague-Dawley rats‘ breast tumors [55]. A crucial factor in chronic progressive diseases, including cancer, is the halting of inflammatory processes. This property has been demonstrated in numerous natural products such as hawthorn from *Crataegus oxyacantha* [56], dandelion from Taraxacum officinale [57] and also resveratrol from *Vitis vinifera* [10], that is the focus of this work, in whose detailed effect there is great interest. In establishing the mechanism of action of resveratrol, its effect on tumor suppressor gene p53 was already ruled out in previous study [58]. But interestingly, an in vivo investigation emphasised resveratrol’s significant anti-carcinogenic effect through suppression of oncogenic Kras expression. In this context, the authors describe a $60\%$ inhibition of CRC incidence in mice with activated Kras mutation by supplementing 150 or 300 ppm resveratrol (which in humans would correspond to a dose of 105 and 210 mg) for nine weeks [59]. And now, through this study, we have identified the effect of resveratrol through the β1-integrin signaling pathway as the fulcrum of numerous anti-CRC mechanisms such as proliferation and invasion, and appropriately, showed the importance of β1-integrin receptors in angiogenesis via targeting HIF-1α, and also in chemosensitisation to 5-FU. Especially in view of the fact that β1-integrin is abundant in CRC cells [29], this finding may be helpful in the future for the therapeutical or concomitant use of resveratrol and conventional chemotherapeutics in CRC management.
## 4.1. Trial Substances
The monoclonal antibodies against phospho-p65-NF-kB (#MAB7226), p65-NF-kB (#MAB5078) as well as polyclonal anti-cleaved-caspase-3 (#AF835) were acquired from R&D Systems (Heidelberg, Germany), while monoclonal antibody against β1-integrin (#14-0299-82) was from Thermo Fisher Scientific (Langenselbold, Germany). Monoclonal antibody to β-actin (#A4700), resveratrol, 5-FU, alginate, DAPI, MTT reagent and Fluoromount were purchased at Sigma-Aldrich (Taufkirchen, Germany). Monoclonal anti-HIF-1α (sc-13515), anti-VEGF (sc-7269) and normal mouse IgG were from Santa Cruz (Dallas, TX, USA). Anti-CD44 (ab243894) and anti-CD133 (ab278053), both monoclonal, were from Abcam PLC (Cambridge, UK) and anti-ALDH1 (A248522) was bought as monoclonal antibody from Acris Antibodies GmbH (Herold, Germany). Alkaline phosphatase-linked Western blot antibodies were from EMD Millipore (Schwalbach, Germany) and rhodamine-coupled secondary immunofluorescence antibodies were obtained from Dianova (Hamburg, Germany), while Epon was purchased from Plano (Marburg, Germany). Resveratrol (100 mM in ethanol) and 5-FU (1000 µM in ethanol) were prepared as stock solutions. Both were finally diluted in cell culture medium without exceeding an ethanol concentration of $0.1\%$ during the CRC cell treatment.
## 4.2. Cell Types and Conditioning
HCT-116 (human CRC cells) were obtained from European Collection of Cell Cultures (Salisbury, UK). Chemoresistant tumor cells were produced from these cells by repeated and long-term treatment with 5-FU, and are hereafter referred to as HCT-116R [60]. MRC-5 (human fibroblasts) were from the same institute and additionally, Jurkat cells (human T-lymphocytes) were acquired from Leibniz Institute (Braunschweig, Germany). The preparatory cell culture has already been described in detail [30], as well as the composition of the Dulbecco’s Modified Eagle medium/F-12 cell culture medium from Sigma-Aldrich (Taufkirchen, Germany), which was used with $10\%$ fetal bovine serum (FBS) as growth medium or with $3\%$ FBS as experimental medium [10].
## 4.3. Knockdown of β1-Integrin
Transient transfection was performed, as described earlier [10], with phosphorothioate-specific oligonucleotides from Eurofins MWG Operon (Ebersberg, Germany), incubated in Lipofectin transfection reagent from Invitrogen (Karlsruhe, Germany). The sequences used were β1-integrin-ASO (5′TAGTTGGGGTTGCACTCACACA3′) as antisense oligonucleotide (ASO) and β1-integrin-SO (5′TGTGTGAGTGCAACCCCAACTA3′) as sense oligonucleotide (SO).
## 4.4. Alginate Drop Preparation
Alginate drops were formed, following a method, shown in numerous previous work [30,61]. HCT-116 or HCT-116R were passaged, resuspended in sterile alginate ($2\%$ in 0.15M NaCl) and dripped into CaCl2 (100 mM) for polymerization. Afterwards, the resulting CRC-alginate drops were washed with 0.15M NaCl threefold and twofold with cell culture medium ($10\%$ FBS). Then, they were incubated in serum-starved cell culture medium containing $3\%$ FBS for 30 min and to start the trial, CRC-alginate drops were transferred with bent tweezers to prepared 12-well-plates.
## 4.5. Carcinogenic Tumor Microenvironment
To illuminate β1-integrin’s role in overcoming chemoresistance through resveratrol, a pro-inflammatory, vivo-near TME was constructed in vitro and treated differently. Therefore, CRC-alginate drops were prepared as described before and placed in experimental 12-well-plates with serum starved cell culture medium ($3\%$ FBS), which were changed every second day. The treatments for 10–14 days were as follows: Firstly, a basal control (alginate drops without TME) and a TME control for self-check. Then, concentration-dependent treatments with resveratrol (1, 5 µM) or 5-FU (1, 2 nM) or transfection with β1-ASO/SO (0.5 µM) and finally combinations thereof. Here, TME represents a multicellular, pro-inflammatory composition with a fibroblast-monolayer (MRC-5) on the bottom of the well, floating T-lymphocytes (Jurkat) suspended in the cell culture medium and added CRC-alginate drops. This 3D tumor study model has already been used by our group [10,30,61] as well as, in a similar way, by other research teams [62].
## 4.6. Vitality Assay
Vitality of CRC cells and thus indirect proliferation was detected by detaching the CRC cells from alginate and performing a MTT assay as previously described [30]. Briefly, the CRC-alginate drops were removed from 12-well-plates with bent tweezers and washed in Hank’s salt solution, ensuring that only CRC were examined. Then, CRC cells were extracted from alginate drops by dissolving with sodium citrate. Afterwards, they were washed in Hank’s solution and resuspended in MTT culture medium consisting $3\%$ FBS, but without vitamin C and phenol red. After adding MTT solution and stopping the reaction after 3 h, the Optical Density of samples was measured by Bio-Rad ELISA reader (Munich, Germany) at 550 nm (OD 550).
## 4.7. Proliferation Assay
In order to document the proliferation of HCT-116 and HCT-116R, developed CRC cell colonospheres in alginate drops were photographed with a Zeiss Axiovert 40 CFL (Oberkochen, Germany) phase contrast microscope after 10–14 days of treatment. The images were stored digitally as already done earlier [10,30].
## 4.8. Invasion Assay
CRC cell colonospheres, that had formed in the alginate drops, emigrated from alginate and settled as colonies to the bottom of the 12-well-plates. These colonies were stained with toluidine blue after fixation with Karnovsky solution and stained colonies were manually counted (three independent trials of each treatment), whereas HCT-116 or HCT-116R colonies were certainly distinguishable from fibroblast monolayer, as already published [10].
## 4.9. Immunofluorescence Microscopy
HCT-116 or HCT-116R, grown as monolayer on glass coverslips, were treated 4 h in a modified TME, the detailed procedures were described in our last papers [10,30,61,63], consisting of MRC-5 monolayer on the bottom, floating Jurkat cells in cell culture medium ($3\%$ FBS) and small mesh bridges for placing the glass coverslips in 6-well-plates. Afterwards, they were fixed in methanol and prepared for immunofluorescence microscopy as also already been described in detail [30,63,64]. The slides were immunolabeled with a primary antibody against HIF-1α (dilution 1:80), processed with a described secondary antibody (dilution 1:100), DAPI-stained for the assurance of CRC cell vitality and fixed in Fluoromount. Immunofluorescence images were taken with a Leica (Wetzlar, Germany) DM 2000 microscope and related LAS V4.12 software.
## 4.10. Electron Microscopic Evaluation
To investigate ultrastructural changes of CRC cell lines, HCT-116 and HCT-116R were cultivated according to the same experimental set-up as described in the Section 4.9. Afterward, the glass coverslips covered with CRC cells were fixed in Karnovsky solution for 1 h, transferred into a tube with a cell scraper and fixed in osmium tetroxide (OsO4) for 2 h. Then, the dehydration with an ascending series of alcohols and embedding of the cells with Epon were followed as described earlier [31,65]. With a Reichert-Jung Ultracut E (Darmstadt, Germany), samples were prepared, then contrasted with $2\%$ uranyl acetate/lead citrate solution and evaluated with a transmission electron microscope (TEM) 10 from Zeiss (Jena, Germany).
## 4.11. Western Blot Analysis
For immunoblotting, alginate drops with HCT-116 or HCT-116R cells were removed from 12-well-plates with bent tweezers after 10–14 days of treatment. To ensure that Western blot samples contained only CRC cells, CRC-alginate drops were washed as described in the Section 4.6. After dissolving in sodium citrate (55 mM) and freeing from alginate residues, CRC cells were resuspended in lysis mix, centrifuged and the supernatant was frozen at −80 °C. The specimens were processed as described earlier [31,61,63]. We used mentioned primary antibodies and secondary antibodies in dilution 1:10.000 [30]. Samples were blotted with a transblot apparatus from Bio-Rad (Munich, Germany) and densitometric values were evaluated with related “Quantity One” analysis software. β-actin was used as a loading control.
## 4.12. Immunoprecipitation Assay
To investigate the functional relationship between β1-integrin and HIF-1α signaling pathways, CRC cell samples were obtained as described in Section 4.11. Afterwards, they were incubated with 25 µL of normal mouse or rabbit IgG serum and *Staphylococcus aureus* to preclear, treated with primary antibody against β1-integrin at 4 °C for 2 h and incubated with *Staphylococcus aureus* at 4 °C for 1 h again according to a proven method [64]. Samples were separated by SDS-PAGE with the explained Western blot technique and apparatus, and anti-HIF-1α antibody was used for this experiment.
## 4.13. Statistical Evaluation
Three independent repetitions were performed from all assays and analyzed by unpaired student’s t-test. The results matched by ANOVA (one-way) followed by a post hoc test to compare group parameters. At all outcomes, a p-value < 0.05 was considered as statistically significant.
## 5. Conclusions
The current results demonstrate an inhibition of growth, viability and pathological morphological changes and thus both a chemosensitisation of non-5-FU-resistant HCT-116 cells and an overcoming of chemoresistance in 5-FU-resistant HCT-116R cells by treatment with resveratrol along with β1-integrin receptor. For this purpose, resveratrol not only acts as pro-apoptotic (caspase-3) but also induces its effect against inflammation (NF-kB), vascularisation (VEGF), and cancer stem cells (CD44, CD133, ALDH1), and also targets the β1-integrin/HIF-1α axis that is highly pronounced in CRC cells. In summary, resveratrol represents a multifunctional polyphenol that could complement the therapy options of advanced, metastatic or 5-FU-resistant CRC in the future.
## References
1. Siegel R.L., Miller K.D., Fuchs H.E., Jemal A.. **Cancer Statistics, 2021**. *CA A Cancer J. Clin.* (2021.0) **71** 7-33. DOI: 10.3322/caac.21654
2. **Darmkrebs**
3. **Leitlinien Kolorektales Karzinom**
4. Satake H., Kagawa Y., Shinozaki E., Tanizawa Y., Jin L., Cai Z., Makiyama A.. **Real-World Data Analysis of Second-Line Antiangiogenic Targeted Treatments Following Anti-Epidermal Growth Factor Receptor Monoclonal Antibodies and First-Line FOLFOX for Patients with Metastatic Colorectal Cancer**. *Adv. Ther.* (2022.0) **39** 2596-2613. DOI: 10.1007/s12325-022-02122-4
5. Akdeniz N., Kaplan M.A., Uncu D., İnanç M., Kaya S., Dane F., Küçüköner M., Demirci A., Bilici M., Durnalı A.G.. **The comparison of FOLFOX regimens with different doses of 5-FU for the adjuvant treatment of colorectal cancer: A multicenter study**. *Int. J. Color. Dis.* (2021.0) **36** 1311-1319. DOI: 10.1007/s00384-021-03888-9
6. Yang I.P., Miao Z.F., Huang C.W., Tsai H.L., Yeh Y.S., Su W.C., Chang T.K., Chang S.F., Wang J.Y.. **High blood sugar levels but not diabetes mellitus significantly enhance oxaliplatin chemoresistance in patients with stage III colorectal cancer receiving adjuvant FOLFOX6 chemotherapy**. *Ther. Adv. Med. Oncol.* (2019.0) **11** 1758835919866964. DOI: 10.1177/1758835919866964
7. Escalante P.I., Quiñones L.A., Contreras H.R.. **Epithelial-Mesenchymal Transition and MicroRNAs in Colorectal Cancer Chemoresistance to FOLFOX**. *Pharmaceutics* (2021.0) **13**. DOI: 10.3390/pharmaceutics13010075
8. Geevimaan K., Guo J.Y., Shen C.N., Jiang J.K., Fann C.S.J., Hwang M.J., Shui J.W., Lin H.T., Wang M.J., Shih H.C.. **Patient-Derived Organoid Serves as a Platform for Personalized Chemotherapy in Advanced Colorectal Cancer Patients**. *Front. Oncol.* (2022.0) **12** 883437. DOI: 10.3389/fonc.2022.883437
9. Hammond W.A., Swaika A., Mody K.. **Pharmacologic resistance in colorectal cancer: A review**. *Ther. Adv. Med. Oncol.* (2016.0) **8** 57-84. DOI: 10.1177/1758834015614530
10. Brockmueller A., Mueller A.L., Shayan P., Shakibaei M.. **β1-Integrin plays a major role in resveratrol-mediated anti-invasion effects in the CRC microenvironment**. *Front. Pharmacol.* (2022.0) **13** 978625. DOI: 10.3389/fphar.2022.978625
11. Defeudis A., Cefaloni L., Giannetto G., Cappello G., Rizzetto F., Panic J., Barra D., Nicoletti G., Mazzetti S., Vanzulli A.. **Comparison of radiomics approaches to predict resistance to 1st line chemotherapy in liver metastatic colorectal cancer**. *Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society* 3305-3308. DOI: 10.1109/embc46164.2021.9630316
12. He X.S., Zou S.Y., Yao J.L., Yu W., Deng Z.Y., Wang J.R., Gan W.J., Wan S., Yang X.Q., Wu H.. **Transcriptomic Analysis Identifies Complement Component 3 as a Potential Predictive Biomarker for Chemotherapy Resistance in Colorectal Cancer**. *Front. Mol. Biosci.* (2021.0) **8** 763652. DOI: 10.3389/fmolb.2021.763652
13. Heslin M.J., Yan J., Weiss H., Shao L., Owens J., Lucas V.S., Diasio R.B.. **Dihydropyrimidine dehydrogenase (DPD) rapidly regenerates after inactivation by eniluracil (GW776C85) in primary and metastatic colorectal cancer**. *Cancer Chemother. Pharmacol.* (2003.0) **52** 399-404. DOI: 10.1007/s00280-003-0681-1
14. Salonga D., Danenberg K.D., Johnson M., Metzger R., Groshen S., Tsao-Wei D.D., Lenz H.J., Leichman C.G., Leichman L., Diasio R.B.. **Colorectal tumors responding to 5-fluorouracil have low gene expression levels of dihydropyrimidine dehydrogenase, thymidylate synthase, and thymidine phosphorylase**. *Clin. Cancer Res. Off. J. Am. Assoc. Cancer Res.* (2000.0) **6** 1322-1327
15. Amatori F., Di Paolo A., Del Tacca M., Fontanini G., Vannozzi F., Boldrini L., Bocci G., Lastella M., Danesi R.. **Thymidylate synthase, dihydropyrimidine dehydrogenase and thymidine phosphorylase expression in colorectal cancer and normal mucosa in patients**. *Pharm. Genom.* (2006.0) **16** 809-816. DOI: 10.1097/01.fpc.0000230410.07899.bc
16. Bai W., Wu Y., Zhang P., Xi Y.. **Correlations between expression levels of thymidylate synthase, thymidine phosphorylase and dihydropyrimidine dehydrogenase, and efficacy of 5-fluorouracil-based chemotherapy for advanced colorectal cancer**. *Int. J. Clin. Exp. Pathol.* (2015.0) **8** 12333-12345. PMID: 26722420
17. Gustavsson B., Kaiser C., Carlsson G., Wettergren Y., Odin E., Lindskog E.B., Niyikiza C., Ma D.. **Molecular determinants of efficacy for 5-FU-based treatments in advanced colorectal cancer: mRNA expression for 18 chemotherapy-related genes**. *Int. J. Cancer* (2009.0) **124** 1220-1226. DOI: 10.1002/ijc.23852
18. Shigeta K., Ishii Y., Hasegawa H., Okabayashi K., Kitagawa Y.. **Evaluation of 5-fluorouracil metabolic enzymes as predictors of response to adjuvant chemotherapy outcomes in patients with stage II/III colorectal cancer: A decision-curve analysis**. *World J. Surg.* (2014.0) **38** 3248-3256. DOI: 10.1007/s00268-014-2738-1
19. Soong R., Shah N., Salto-Tellez M., Tai B.C., Soo R.A., Han H.C., Ng S.S., Tan W.L., Zeps N., Joseph D.. **Prognostic significance of thymidylate synthase, dihydropyrimidine dehydrogenase and thymidine phosphorylase protein expression in colorectal cancer patients treated with or without 5-fluorouracil-based chemotherapy**. *Ann. Oncol. Off. J. Eur. Soc. Med. Oncology* (2008.0) **19** 915-919. DOI: 10.1093/annonc/mdm599
20. Buhrmann C., Shayan P., Kraehe P., Popper B., Goel A., Shakibaei M.. **Resveratrol induces chemosensitization to 5-fluorouracil through up-regulation of intercellular junctions, Epithelial-to-mesenchymal transition and apoptosis in colorectal cancer**. *Biochem. Pharmacol.* (2015.0) **98** 51-68. DOI: 10.1016/j.bcp.2015.08.105
21. Amin A., Lotfy M., Mahmoud-Ghoneim D., Adeghate E., Al-Akhras M.A., Al-Saadi M., Al-Rahmoun S., Hameed R.. **Pancreas-protective effects of chlorella in STZ-induced diabetic animal model: Insights into the mechanism**. *J. Diabetes Mellit.* (2011.0) **1** 36. DOI: 10.4236/jdm.2011.13006
22. Dakhale G.N., Chaudhari H.V., Shrivastava M.. **Supplementation of vitamin C reduces blood glucose and improves glycosylated hemoglobin in type 2 diabetes mellitus: A randomized, double-blind study**. *Adv. Pharmacol. Sci.* (2011.0) **2011** 195271. DOI: 10.1155/2011/195271
23. Al-Dabbagh B., Elhaty I.A., Murali C., Al Madhoon A., Amin A.. **Salvadora persica (Miswak): Antioxidant and promising antiangiogenic insights**. *Am. J. Plant Sci.* (2018.0) **9** 1228. DOI: 10.4236/ajps.2018.96091
24. Brockmueller A., Samuel S.M., Mazurakova A., Büsselberg D., Kubatka P., Shakibaei M.. **Curcumin, calebin A and chemosensitization: How are they linked to colorectal cancer?**. *Life Sci.* (2023.0) **318** 121504. DOI: 10.1016/j.lfs.2023.121504
25. Ehala S., Vaher M., Kaljurand M.. **Characterization of phenolic profiles of Northern European berries by capillary electrophoresis and determination of their antioxidant activity**. *J. Agric. Food Chem.* (2005.0) **53** 6484-6490. DOI: 10.1021/jf050397w
26. Meng T., Xiao D., Muhammed A., Deng J., Chen L., He J.. **Anti-Inflammatory Action and Mechanisms of Resveratrol**. *Molecules* (2021.0) **26**. DOI: 10.3390/molecules26010229
27. Buhrmann C., Shayan P., Brockmueller A., Shakibaei M.. **Resveratrol Suppresses Cross-Talk between Colorectal Cancer Cells and Stromal Cells in Multicellular Tumor Microenvironment: A Bridge between In Vitro and In Vivo Tumor Microenvironment Study**. *Molecules* (2020.0) **25**. DOI: 10.3390/molecules25184292
28. Dariya B., Behera S.K., Srivani G., Farran B., Alam A., Nagaraju G.P.. **Computational analysis of nuclear factor-κB and resveratrol in colorectal cancer**. *J. Biomol. Struct. Dyn.* (2021.0) **39** 2914-2922. DOI: 10.1080/07391102.2020.1757511
29. Long X., Wong C.C., Tong L., Chu E.S.H., Ho Szeto C., Go M.Y.Y., Coker O.O., Chan A.W.H., Chan F.K.L., Sung J.J.Y.. **Peptostreptococcus anaerobius promotes colorectal carcinogenesis and modulates tumour immunity**. *Nat. Microbiol.* (2019.0) **4** 2319-2330. DOI: 10.1038/s41564-019-0541-3
30. Brockmueller A., Shayan P., Shakibaei M.. **Evidence That β1-Integrin Is Required for the Anti-Viability and Anti-Proliferative Effect of Resveratrol in CRC Cells**. *Int. J. Mol. Sci.* (2022.0) **23**. DOI: 10.3390/ijms23094714
31. Buhrmann C., Yazdi M., Popper B., Shayan P., Goel A., Aggarwal B.B., Shakibaei M.. **Resveratrol Chemosensitizes TNF-β-Induced Survival of 5-FU-Treated Colorectal Cancer Cells**. *Nutrients* (2018.0) **10**. DOI: 10.3390/nu10070888
32. Wei T.T., Lin Y.T., Tang S.P., Luo C.K., Tsai C.T., Shun C.T., Chen C.C.. **Metabolic targeting of HIF-1α potentiates the therapeutic efficacy of oxaliplatin in colorectal cancer**. *Oncogene* (2020.0) **39** 414-427. DOI: 10.1038/s41388-019-0999-8
33. Azuma M., Shi M., Danenberg K.D., Gardner H., Barrett C., Jacques C.J., Sherod A., Iqbal S., El-Khoueiry A., Yang D.. **Serum lactate dehydrogenase levels and glycolysis significantly correlate with tumor VEGFA and VEGFR expression in metastatic CRC patients**. *Pharmacogenomics* (2007.0) **8** 1705-1713. DOI: 10.2217/14622416.8.12.1705
34. Fan L.F., Dong W.G., Jiang C.Q., Qian Q., Yu Q.F.. **Role of Hypoxia-inducible factor-1 alpha and Survivin in colorectal carcinoma progression**. *Int. J. Color. Dis.* (2008.0) **23** 1057-1064. DOI: 10.1007/s00384-008-0511-3
35. Jung K.H., Lee J.H., Thien Quach C.H., Paik J.Y., Oh H., Park J.W., Lee E.J., Moon S.H., Lee K.H.. **Resveratrol suppresses cancer cell glucose uptake by targeting reactive oxygen species-mediated hypoxia-inducible factor-1α activation**. *J. Nucl. Med. Off. Publ. Soc. Nucl. Med.* (2013.0) **54** 2161-2167. DOI: 10.2967/jnumed.112.115436
36. Wu H., Liang X., Fang Y., Qin X., Zhang Y., Liu J.. **Resveratrol inhibits hypoxia-induced metastasis potential enhancement by restricting hypoxia-induced factor-1 alpha expression in colon carcinoma cells**. *Biomed. Pharmacother. Biomed. Pharmacother.* (2008.0) **62** 613-621. DOI: 10.1016/j.biopha.2008.06.036
37. Colin D., Gimazane A., Lizard G., Izard J.C., Solary E., Latruffe N., Delmas D.. **Effects of resveratrol analogs on cell cycle progression, cell cycle associated proteins and 5fluoro-uracil sensitivity in human derived colon cancer cells**. *Int. J. Cancer* (2009.0) **124** 2780-2788. DOI: 10.1002/ijc.24264
38. Chen J.S., Hsu Y.M., Chen C.C., Chen L.L., Lee C.C., Huang T.S.. **Secreted heat shock protein 90alpha induces colorectal cancer cell invasion through CD91/LRP-1 and NF-kappaB-mediated integrin alphaV expression**. *J. Biol. Chem.* (2010.0) **285** 25458-25466. DOI: 10.1074/jbc.M110.139345
39. Buhrmann C., Shayan P., Goel A., Shakibaei M.. **Resveratrol Regulates Colorectal Cancer Cell Invasion by Modulation of Focal Adhesion Molecules**. *Nutrients* (2017.0) **9**. DOI: 10.3390/nu9101073
40. Ochiai T., Ohta K., Iida M., Kumagai Y., Mitsunori Y., Aihara A., Noguchi N., Tanaka S., Arii S., Yamazaki S.. **High resectability of colorectal liver metastases with aggressive chemotherapy in the era of molecular target-based agents**. *Hepato-Gastroenterol.* (2013.0) **60** 955-960. DOI: 10.5754/hge121213
41. Huang L., Zhang S., Zhou J., Li X.. **Effect of resveratrol on drug resistance in colon cancer chemotherapy**. *RSC Adv.* (2019.0) **9** 2572-2580. DOI: 10.1039/C8RA08364A
42. Bhat A.A., Nisar S., Singh M., Ashraf B., Masoodi T., Prasad C.P., Sharma A., Maacha S., Karedath T., Hashem S.. **Cytokine- and chemokine-induced inflammatory colorectal tumor microenvironment: Emerging avenue for targeted therapy**. *Cancer Commun.* (2022.0) **42** 689-715. DOI: 10.1002/cac2.12295
43. Brockmueller A., Sameri S., Liskova A., Zhai K., Varghese E., Samuel S.M., Büsselberg D., Kubatka P., Shakibaei M.. **Resveratrol’s Anti-Cancer Effects through the Modulation of Tumor Glucose Metabolism**. *Cancers* (2021.0) **13**. DOI: 10.3390/cancers13020188
44. Kang Y., Lv R., Feng Z., Zhu J.. **Tumor-associated macrophages improve hypoxia-induced endoplasmic reticulum stress response in colorectal cancer cells by regulating TGF-β1/SOX4**. *Cell. Signal.* (2022.0) **99** 110430. DOI: 10.1016/j.cellsig.2022.110430
45. Vatandoust S., Price T.J., Karapetis C.S.. **Colorectal cancer: Metastases to a single organ**. *World J. Gastroenterol.* (2015.0) **21** 11767-11776. DOI: 10.3748/wjg.v21.i41.11767
46. Ren J., Hamada J., Takeichi N., Fujikawa S., Kobayashi H.. **Ultrastructural differences in junctional intercellular communication between highly and weakly metastatic clones derived from rat mammary carcinoma**. *Cancer Res.* (1990.0) **50** 358-362. PMID: 2295075
47. Casillas A.L., Toth R.K., Sainz A.G., Singh N., Desai A.A., Kraft A.S., Warfel N.A.. **Hypoxia-Inducible PIM Kinase Expression Promotes Resistance to Antiangiogenic Agents**. *Clin. Cancer Res. Off. J. Am. Assoc. Cancer Res.* (2018.0) **24** 169-180. DOI: 10.1158/1078-0432.CCR-17-1318
48. Crociani O., Zanieri F., Pillozzi S., Lastraioli E., Stefanini M., Fiore A., Fortunato A., D’Amico M., Masselli M., De Lorenzo E.. **hERG1 channels modulate integrin signaling to trigger angiogenesis and tumor progression in colorectal cancer**. *Sci. Rep.* (2013.0) **3** 3308. DOI: 10.1038/srep03308
49. Liang Z., Liu H., Zhang Y., Xiong L., Zeng Z., He X., Wang F., Wu X., Lan P.. **Cyr61 from adipose-derived stem cells promotes colorectal cancer metastasis and vasculogenic mimicry formation via integrin α(V) β(5)**. *Mol. Oncol.* (2021.0) **15** 3447-3467. DOI: 10.1002/1878-0261.12998
50. Rajitha B., Nagaraju G.P., Shaib W.L., Alese O.B., Snyder J.P., Shoji M., Pattnaik S., Alam A., El-Rayes B.F.. **Novel synthetic curcumin analogs as potent antiangiogenic agents in colorectal cancer**. *Mol. Carcinog.* (2017.0) **56** 288-299. DOI: 10.1002/mc.22492
51. Srivani G., Behera S.K., Dariya B., Aliya S., Alam A., Nagaraju G.P.. **Resveratrol binds and inhibits transcription factor HIF-1α in pancreatic cancer**. *Exp. Cell Res.* (2020.0) **394** 112126. DOI: 10.1016/j.yexcr.2020.112126
52. Wang D., Gao Z., Zhang X.. **Resveratrol Induces Apoptosis in Murine Prostate Cancer Cells via Hypoxia-Inducible Factor 1-alpha (HIF-1α)/Reactive Oxygen Species (ROS)/P53 Signaling**. *Med. Sci. Monit. Int. Med. J. Exp. Clin. Res.* (2018.0) **24** 8970-8976. DOI: 10.12659/MSM.913290
53. Nelson D.R., Hrout A.A., Alzahmi A.S., Chaiboonchoe A., Amin A., Salehi-Ashtiani K.. **Molecular Mechanisms behind Safranal’s Toxicity to HepG2 Cells from Dual Omics**. *Antioxidants* (2022.0) **11**. DOI: 10.3390/antiox11061125
54. Abdu S., Juaid N., Amin A., Moulay M., Miled N.. **Effects of Sorafenib and Quercetin Alone or in Combination in Treating Hepatocellular Carcinoma: In Vitro and In Vivo Approaches**. *Molecules* (2022.0) **27**. DOI: 10.3390/molecules27228082
55. Al-Akhras M.A., Aljarrah K., Al-Khateeb H., Jaradat A., Al-Omari A., Al-Nasser A., Masadeh M.M., Amin A., Hamza A., Mohammed K.. **Introducing Cichorium Pumilum as a potential therapeutical agent against drug-induced benign breast tumor in rats**. *Electromagn. Biol. Med.* (2012.0) **31** 299-309. DOI: 10.3109/15368378.2012.662193
56. Hamza A.A., Lashin F.M., Gamel M., Hassanin S.O., Abdalla Y., Amin A.. **Hawthorn Herbal Preparation from Crataegus oxyacantha Attenuates In Vivo Carbon Tetrachloride -Induced Hepatic Fibrosis via Modulating Oxidative Stress and Inflammation**. *Antioxidants* (2020.0) **9**. DOI: 10.3390/antiox9121173
57. Mohammad N.S., Nazli R., Zafar H., Fatima S.. **Effects of lipid based Multiple Micronutrients Supplement on the birth outcome of underweight pre-eclamptic women: A randomized clinical trial**. *Pak. J. Med. Sci.* (2022.0) **38** 219-226. DOI: 10.12669/pjms.38.1.4396
58. Mahyar-Roemer M., Katsen A., Mestres P., Roemer K.. **Resveratrol induces colon tumor cell apoptosis independently of p53 and precede by epithelial differentiation, mitochondrial proliferation and membrane potential collapse**. *Int. J. Cancer* (2001.0) **94** 615-622. DOI: 10.1002/ijc.1516
59. Saud S.M., Li W., Morris N.L., Matter M.S., Colburn N.H., Kim Y.S., Young M.R.. **Resveratrol prevents tumorigenesis in mouse model of Kras activated sporadic colorectal cancer by suppressing oncogenic Kras expression**. *Carcinogenesis* (2014.0) **35** 2778-2786. DOI: 10.1093/carcin/bgu209
60. Yu Y., Kanwar S.S., Patel B.B., Nautiyal J., Sarkar F.H., Majumdar A.P.. **Elimination of Colon Cancer Stem-Like Cells by the Combination of Curcumin and FOLFOX**. *Transl. Oncol.* (2009.0) **2** 321-328. DOI: 10.1593/tlo.09193
61. Buhrmann C., Kunnumakkara A.B., Kumar A., Samec M., Kubatka P., Aggarwal B.B., Shakibaei M.. **Multitargeting Effects of Calebin A on Malignancy of CRC Cells in Multicellular Tumor Microenvironment**. *Front. Oncol.* (2021.0) **11** 650603. DOI: 10.3389/fonc.2021.650603
62. Gao H., Tian Q., Zhu L., Feng J., Zhou Y., Yang J.. **3D Extracellular Matrix Regulates the Activity of T Cells and Cancer Associated Fibroblasts in Breast Cancer**. *Front. Oncol.* (2021.0) **11** 764204. DOI: 10.3389/fonc.2021.764204
63. Mueller A.L., Brockmueller A., Kunnumakkara A.B., Shakibaei M.. **Calebin A, a Compound of Turmeric, Down-Regulates Inflammation in Tenocytes by NF-κB/Scleraxis Signaling**. *Int. J. Mol. Sci.* (2022.0) **23**. DOI: 10.3390/ijms23031695
64. Buhrmann C., Busch F., Shayan P., Shakibaei M.. **Sirtuin-1 (SIRT1) is required for promoting chondrogenic differentiation of mesenchymal stem cells**. *J. Biol. Chem.* (2014.0) **289** 22048-22062. DOI: 10.1074/jbc.M114.568790
65. Shakibaei M., De Souza P.. **Differentiation of mesenchymal limb bud cells to chondrocytes in alginate beads**. *Cell Biol. Int.* (1997.0) **21** 75-86. DOI: 10.1006/cbir.1996.0119
|
---
title: 'Impact of Acute High Glucose on Mitochondrial Function in a Model of Endothelial
Cells: Role of PDGF-C'
authors:
- Adriana Grismaldo Rodríguez
- Jairo Zamudio Rodríguez
- Alfonso Barreto
- Sandra Sanabria-Barrera
- José Iglesias
- Ludis Morales
journal: International Journal of Molecular Sciences
year: 2023
pmcid: PMC10003065
doi: 10.3390/ijms24054394
license: CC BY 4.0
---
# Impact of Acute High Glucose on Mitochondrial Function in a Model of Endothelial Cells: Role of PDGF-C
## Abstract
An increase in plasma high glucose promotes endothelial dysfunction mainly through increasing mitochondrial ROS production. High glucose ROS—induced has been implicated in the fragmentation of the mitochondrial network, mainly by an unbalance expression of mitochondrial fusion and fission proteins. Mitochondrial dynamics alterations affect cellular bioenergetics. Here, we assessed the effect of PDGF-C on mitochondrial dynamics and glycolytic and mitochondrial metabolism in a model of endothelial dysfunction induced by high glucose. High glucose induced a fragmented mitochondrial phenotype associated with the reduced expression of OPA1 protein, high DRP1pSer616 levels and reduced basal respiration, maximal respiration, spare respiratory capacity, non-mitochondrial oxygen consumption and ATP production, regarding normal glucose. In these conditions, PDGF-C significantly increased the expression of OPA1 fusion protein, diminished DRP1pSer616 levels and restored the mitochondrial network. On mitochondrial function, PDGF-C increased the non-mitochondrial oxygen consumption diminished by high glucose conditions. These results suggest that PDGF-C modulates the damage induced by HG on the mitochondrial network and morphology of human aortic endothelial cells; additionally, it compensates for the alteration in the energetic phenotype induced by HG.
## 1. Introduction
Metabolic diseases, including diabetes, are considered the main risk factor for the development of cardiovascular diseases (CVD) [1,2]. The origin of CVD has been related to early loss of vascular endothelial function, which decreases the production or bioavailability of vasodilator molecules such as nitric oxide (NO) and predisposes to a blood vessel contraction phenotype [3]. Transient and sustained glucose levels greater than 5.5 mmol/L considered average fasting glucose levels [4] induce endothelial dysfunction [5,6,7]. Consequently, identification of the initial steps that lead to endothelial dysfunction under high glucose conditions is crucial for early intervention in diabetes.
Reactive oxygen species (ROS) production, especially superoxide radical (O−) by the mitochondria, is one of the intracellular mechanisms that reduce the biodisponibility of NO. Due to the high reactivity of O− and NO, the production of peroxynitrite (ONOO-) is promoted, altering the structure of nucleic acids, proteins, and lipids and leading to cell death [8]. Although the mitochondrial content in endothelial cells (ECs) is low compared to other cell types with higher energy demands [9], they are considered signalling organelles that act as local microenvironmental sensors [9,10,11]. Its optimal function is driven by a balance between fission and fusion processes [3,12,13,14,15]. However, in diabetic patients and hyperglycemic conditions, a decreased expression of OPA1 and MFN$\frac{1}{2}$ proteins (related to the fusion mechanism) and an increased expression of DRP1 and FIS1 proteins (related to the fission mechanisms) are observed. This imbalance, coupled with an impaired autophagy mechanism, leads to a disrupted mitochondrial network and the accumulation of tiny, damaged, and inefficient organelles that contribute to the increased ROS production and loss of ECs function or death [12,13,14,16].
In this context, looking for new therapies that mitigate the mitochondrial damage induced by high glucose is critical for the reduction of CVD risk in diabetic patients.
Recently, we reported the role of PDGF-C on the modulation of mitochondrial oxidative stress induced by high d-glucose in human aortic endothelial cells (HAECs). PDGF-C is a growth factor that exerts its effects by binding to PDGFRα and PDGFR αβ tyrosine kinase receptors. PDGF-C reduced the increase in mitochondrial superoxide production, and it was associated with the up-regulation of SOD2 expression and activity and the modulation of *Keap1* gene expression [5]. Now, here we report the effect of this growth factor on the modulation of the fragmented mitochondrial morphology and the mitochondrial functional changes induced by high glucose in endothelial cells.
## 2.1. PDGF-C Restores the Integrity of Mitochondrial Network of HAECs Going on High d-Glucose Treatment
To better understand the effect of PDGF-C on mitochondrial damage induced by 35 mmol/L of d-glucose for 7 h (herein referred to as HG) in HAECs, mitochondrial network integrity was evaluated by confocal microscopy. As shown in Figure 1A, mitochondria of cells cultured in 5 mmol/L of d-glucose (herein referred to as NG) exhibited a continuous and elongated network with peripheral localization (upper left and right). In contrast, HG induced a shorter and fragmented mitochondrial morphology (lower left), which changes to dense and hyperfused aggregates with nuclear localization when cells were treated with 50 ng/mL of PDGF-C for 1 h (lower right). The reduction of $64\%$ of the count of branches (*** $p \leq 0.001$) (Figure 1A), $63\%$ of the count of junctions (*** $p \leq 0.001$) (Figure 1B), and $71\%$ of the mitochondrial area (**** $p \leq 0.0001$) (Figure 1C)) in HG conditions for 7 h and compared to NG support the morphology observations. Treatment with PDGF-C 50 ng/mL for 1 h significantly increased the number of branches (# $p \leq 0.05$), junctions (# $p \leq 0.05$) and a tendency to increase the total mitochondria area of cells treated with HG. These results suggest that PDGF-C modulates the damage induced by HG on the mitochondrial network and morphology of HAECs.
## 2.2. Mitochondrial Dynamic-Related Proteins Expression
To reinforce these results, the expression of fission and fusion proteins was measured by western blot in the same conditions mentioned above. Regarding NG conditions, and as described in Figure 2A,B (Supplementary Figure S1A,B), HG did not significantly change MFN1 and MFN2 expression. Similarly, PDGF-C did not affect the expression of these fusion proteins in any of the evaluated glucose conditions and times. Opposite, HAECs going on HG for 6 and 7 h diminished the OPA1 protein expression (Figure 2C, Supplementary Figure S1C), regarding NG (* $p \leq 0.05$); PDGF-C treatment restored to the basal level, the expression of this fusion-related protein (# $$p \leq 0.0486$$).
On the other hand, results about mitochondrial fission-related proteins showed that HG did not change the expression of either FIS1 or DRP1 alone or in combination with PDGF-C (Figure 3A,B, respectively. Supplementary Figure S2A,B). Additionally, the phosphorylation of DRP1 at Ser616, which is known for promoting the fission of the mitochondrial network [17,18], also was evaluated by western blot. As shown in Figure 3C (Supplementary Figure S2C), HG for 6 (** $p \leq 0.01$) and 7 h (*** $p \leq 0.001$) increased the phosphorylation of Ser616 residue in DRP1 and PDGF-C treatment diminished this to basal level (### $p \leq 0.001$).
These results suggest that PDGF-C modulates the mitochondrial network and morphology by regulating the fission through phosphorylation and dephosphorylation of DRP1 and intensifying the fusion process by upregulating OPA1 expression in HAECs going on HG conditions.
## 2.3. Bioenergetic Analysis
To know the implications of acute elevated high glucose concentrations on the mitochondrial function of HAECs and the role of PDGF-C in these conditions, cells were treated as mentioned above. The oxygen consumption rates (OCRs) were measured with Agilent Seahorse XFe24 Analyzer Mitostress Test (Seahorse Bioscience, Agilent, Santa Clara, CA, USA), according to the manufacturer’s protocol [16,19]. The live-cell bioenergetics was conducted to determine the basal mitochondrial functions, including oxygen consumption rates (OCR), extracellular acidification rates (ECAR), ATP production, proton leak, maximal respiration, spare respiratory capacity, mitochondrial stress, and nonmitochondrial respiration. Basal OCR and OCR in response to injection of oligomycin (ATP synthase inhibitor), FCCP (mitochondrial uncoupler), and rotenone/antimycin (Complex I and III inhibitors, respectively; Figure 4, central upper panel) were assayed. Evaluation of the six mitochondrial parameters showed that HG significantly reduced basal respiration (** $p \leq 0.01$; Figure 4A), maximal respiration (** $p \leq 0.01$; Figure 4C), spare respiratory capacity (**$p \leq 0.01$; Figure 4D), non-mitochondrial oxygen consumption (* $p \leq 0.05$; Figure 4E) and ATP production (* $p \leq 0.05$; Figure 4F), regarding NG conditions. PDGF-C significantly increased the non-mitochondrial oxygen consumption diminished by HG conditions (# $p \leq 0.05$; Figure 4E).
In the same experiments, the parameters baseline phenotype, stressed phenotype (after oligomycin injection), and metabolic potential were evaluated to assess the cell energy metabolism phenotype (Figure 5, central upper panel) of HAECs going on NG and HG conditions, and the effect of PDGF-C on changes induced by HG. As shown in Figure 5, HG significantly reduced the baseline OCR (* $p \leq 0.05$; Figure 5A), the baseline OCR/ECAR ratio (** $p \leq 0.01$; Figure 5C), the stressed OCR (** $p \leq 0.01$; Figure 5D) and the stressed OCR/ECAR ratio (**** $p \leq 0.0001$). PDGF-C increased the stressed OCR (* $p \leq 0.05$; Figure 5D) and stressed ECAR (* $p \leq 0.05$; Figure 5E) and slightly reduced the metabolic potential (% baseline OCR; non-significant; Figure 5G). Interestingly, PDGF-C significantly reduced the metabolic potential (% baseline OCR; # $p \leq 0.05$; Figure 5G) and increased the metabolic potential (% baseline ECAR; # $p \leq 0.05$; Figure 5H) in basal d-glucose conditions (5 mmol/L), suggesting that PDGF-C potentiates the glycolytic metabolism even in normal glucose conditions.
## 3. Discussion
Although mitochondrial content in endothelial cells is low because of their low energy demand [9] and their mainly glycolytic ATP production [7,20,21,22,23], they have a long and extensive mitochondrial network that undergoes balanced cycles of fission and fusion and exerts essential functions related with environmental sensing and signaling [9,10,11], and maintaining the balance among calcium concentrations, ROS production and nitric oxide [23].
Mitochondrial network fragmentation has been previously reported in endothelial cells and in in vivo models of high glucose environment and diabetes [3,12,13,14,15]; this condition has been associated with the development of vascular dysfunction [3,15]. It is clearly stated the influence of increased ROS production in the induction of mitochondrial fission [24,25]; our results support these affirmations. In a previous study published by our group, we found augmented mitochondrial ROS in HAECs treated with HG for 6 to 9 h. It was related to the diminished expression of the antioxidant enzyme SOD2 and the activity of the Nrf2/Keap1 pathway [5]. Now here, in the same endothelial model, we report the induction of mitochondrial network fragmentation by HG conditions, reflected as short and discontinuous mitochondria localized at the cellular periphery, reduction in the count of branches, junctions, and total area, diminished expression of fusion protein OPA1, and augmented levels of DRP1pSer616, regarding NG conditions (Figure 1 and Figure 2). Concerning the PDGF-C effect, there are no reports about its involvement in the mitochondrial dynamics process associated with any pathology, so this is the first report showing PDGF-C as a mitochondrial morphology modulator in endothelial cells subjected to metabolic stress conditions. The mechanisms could be associated with the induction of SOD2 expression and consequent reduction in mitochondrial ROS production [5], which could regulate the mitochondrial fission mechanisms and maintain mitochondrial integrity and functionality [26].
Although no changes in DRP1 expression were observed, it is known that its pro-fission role depends on its translocation from the cytoplasm to the mitochondrial outer membrane [24]. This process is controlled by the phosphorylation of Ser616 and Ser637 residues [24,27]. In our model, HG conditions induced the phosphorylation of DRP1 in Ser616, which promotes the transit of DRP1 to mitochondria and leads to its fragmentation [17]; interestingly, this effect was reverted by PDGF-C treatment, possibly through the parallel activation of phosphatases whose target is DRP1; however, the main mechanisms remain unclear. In this context, the increased expression of OPA1 and the modulation of phosphorylation of DRP1 in Ser616 residue induced by PDGF-C probably promotes the fusion of dysfunctional mitochondria, leading to a distribution of damaged components, including mitochondrial DNA, uncoupling proteins, and antioxidant enzymes [13,14]. Additionally, the phosphorylation of DRP1 on Ser637 residue is known for reversing the effects of Ser616 phosphorylation [24]. Although the phosphorylation state of this residue was not evaluated in our study, it is known that PDGF-C drives different signalling pathways, including PI3K/Akt, MAPK, and PLCγ [28], leading to the activation of kinases such as AMPK, MAPK, and cyclin-dependent kinase 1/cyclin B1, involved in the phosphorylation of this residue [24].
Changes in mitochondrial morphology induced by environmental conditions, such as increased extracellular glucose levels, can alter the typical mitochondrial bioenergetics profile [29]. As shown in our results, HG-induced alterations in mitochondrial function are evidenced by diminishing basal respiration, maximal respiration, reserve capacity, non-mitochondrial OCR, and ATP-linked OCR (Figure 4). The reduction in maximal respiration, reserve capacity, and ATP-linked OCR has been related to diminished mitochondrial mass, mitochondrial dysfunction, low ATP demand and severe electron transport chain damage, respectively [16,19]; which is according with the diminished total mitochondrial area, mitochondrial network fragmentation (fewer branches and junctions regarding NG conditions) observed by confocal microscopy (Figure 1) and the reduction of mitochondrial fusion evidenced by the diminished expression of OPA1 (Figure 2). Although the non-mitochondrial OCR has been related to the increased production of extramitochondrial ROS (cytosolic) [16,19], in our model, we did not find evidence that suggests the high production of cytosolic ROS in EC exposed to HG [5]. Our results are supported by different studies that indicate that DRP1-induced mitochondrial fission is associated with a diminished OXPHOS capacity and the increased activity of glycolytic metabolism [24,30]. On these affected parameters, PDGF-C recovered the mitochondrial morphology, possibly through increasing the expression of the mitochondrial fusion protein OPA1; however, PDGF-C only exerted a restauration role on the non-mitochondrial OCR parameter (Figure 4E), which could be related to the induction of the initial response of endothelial cells to metabolic stress.
Even though our results indicate a diminished OXPHOS activity in HG cells (Figure 4 and Figure 5), proteomic analysis by [31] shows that energy production in diabetic primary rat cardiac microvascular endothelial cells (RCMVECs) is shifted from glycolysis to OXPHOS after high glucose stress (25 mM by 2 weeks). However, we demonstrated that acute HG stress (7 h) in non-diabetic human aortic endothelial cells decreases OXPHOS metabolism (Figure 4) by the assessment of oxygen consumption and, similarly, Hapsula et al., 2019 [31], reported diminished oxidative phosphorylation and increased glycolysis-related protein expression in non-diabetic RCMVECs after HG exposure, regarding cells in NG conditions. These results suggest differential metabolic responses to HG exposure dependent on cell origin and phenotype (i.e., microvasculature vs microvasculature, diabetic vs non-diabetic).
Typically, when extracellular glucose increases, the endothelial cells enhance the glucose uptake mainly through GLUT 1 transporters and metabolism through glycolysis and glycolytic side branches [32], while the OXPHOS capacity diminishes, as reported by [23] in the EA.hy926 cell line and confirmed by our results (Figure 4). Similarly, in a high glucose HUVECs model, Zeng et al., 2019 [33] evidenced an unbalanced process of mitochondrial dynamics promoting fission through the decreased expression of MFN1 and increased expression of FIS1, which was associated with the decreased expression of complex I (NADH: ubiquinone oxidoreductase core subunit 1) and complex II (Succinate dehydrogenase) of the electron transport chain, leading to a deficient aerobic metabolism.
Our results suggest that PDGF-C modulates the damage induced by HG on the mitochondrial network and morphology; additionally, it compensates for the alteration in the energetic phenotype induced by HG. Nevertheless, our work proposes an initial approach to show the changes that acute HG induces in a macrovascular endothelial cell model and the role that PDGF-C can exert on these changes. It constitutes a guide for future experiments, including the assessment of endothelial function parameters (i.e., angiogenic capacity, nitric oxide production) and the evaluation of the behavior of each mitochondrial complex in the established conditions.
## 4.1. Cells and Reagents
Human Aortic Endothelial Cells (CC-2535), EGM-2 BulletKit (CC-3162) and EBM-2 [00190860] were obtained from Lonza (Walkersville, MD, USA). hrPDGF-C (SRP3139) and Valinomycin (V0627) were obtained from Sigma-Aldrich (St. Louis, MO, USA). Mitotracker Green FM (M7514), Hoechst 34580 (H21486) N,N-dimethyl-4-[5-(4-methyl-1-piperazinyl)[2,5′-bi-1H-benzimidazol]-2′yl] and SuperSignalTM west pico PLUS chemiluminescent substrate [34577] were obtained from Thermo Fisher Scientific/Invitrogen (Chelmsford, MA, USA). Protease and phosphatase inhibitor cocktail [5872], antibodies against OPA1 (D7C1A), MFN1 (D6E2S), MFN2 (D1E9), DRP1 (D6C7), DRP1pS616, β actin (8H10D10), Anti-rabbit IgG HRP-linked and anti-mouse IgG HRP-linked were obtained from Cell Signaling Technology (Danvers, MA, USA). Antibody against (ab96764) were obtained from Abcam (San Francisco, CA, USA).
## 4.2. Cell Culture and Treatments
All experiments were established according to the conditions selected before and reported in [5]. Briefly, Human Aortic Endothelial Cells (HAECs) from passage 4 to passage 7 were grown in standard culture conditions in EGM-2 BulletKit containing 5.5 mmol/L glucose (normal human fasting blood sugar average) [4]. Confluent cells were seeded in multiwell plates, and after 24 h, cells were deprived in an EBM-2 medium containing 5.5 mmol/L glucose and $0.2\%$ fetal bovine serum. After 12 h, cells were treated with 29.5 mmol/L d-glucose to reach a final concentration of 35 mmol/L (HG) for 6–9 h; these times were selected according to a previous study where we identified increased production of mitochondrial ROS after 6–9 h of HG [5]. Treatments with 50 ng/mL of hrPDGF-C were made for 1 h after 6 h of 35 mmol/L d-glucose stress induction, considering the short half-life of PDGF in HUVECs, which has been reported to be between 50 min and 3 h [34]. All comparisons were made from cells treated with glucose 5.5 mmol/L.
## 4.3. Mitochondrial Network Analysis
HAECs were seeded at 3 × 104 cells/well in 35 mm glass-bottom culture dishes (MatTek) coated with $0.2\%$ gelatin and cultured until $60\%$ of confluence. Once deprived for 12 h, cells were treated with d-glucose 35 mmol/L for 6 h and 1 additional hour with 50 ng/mL of PDGF-C to evaluate mitochondrial network integrity. Briefly, live cells were washed once with PBS and stained with 100 nmol/L Mitotracker Green FM and 5 μg/mL Hoechst to define mitochondria and nucleus, respectively [35]. 2D and 3D cell imaging were acquired with an Olympus FV1000 confocal microscope, using a 60× oil immersion objective and an excitation/emission range of $\frac{400}{545}$ for MitoTracker Green FM and $\frac{361}{497}$ for Hoechst. Images were pre-processed according to the protocol suggested by Chaudhry et al., 2020 [36], data about the count and length of branches, count of junctions, and total mitochondria area were obtained according to the protocol suggested by Valente et al., 2017 [37], using the Fiji plugin for Image J.
## 4.4. Mitochondrial Dynamics-Related Proteins Expression
Expression of mitochondrial fusion OPA1, MFN1, MFN2 and fission DRP1, and FIS1 proteins was measured by western blot. HAECs were seeded in 6-well plates, treated as above, and lysed on ice in RIPA buffer supplemented with protease and phosphatase inhibitors cocktail. Total protein was quantified by bicinchoninic acid. Obtained protein was electrophoresed and transferred to PVDF membranes. Membranes were incubated overnight at 4 °C with the antibodies and dilutions mentioned in Table 1. The next day, membranes were washed and incubated with anti-rabbit IgG HRP-linked or anti-mouse IgG HRP-linked (Table 1) antibodies at room temperature for 1 h. Protein bands were detected with SuperSignalTM west pico PLUS chemiluminescent substrate and captured by the iBright 1500 imaging system from ThermoFisher Scientific (Chelmsford, MA, USA). Analysis of obtained bands was evaluated by densitometry with Image J software [38].
## 4.5. Bioenergetics Analysis
Oxygen consumption rate (OCR) and Extracellular acidification rate (ECAR) were measured in a Seahorse XFe24 analyzer (Seahorse Biosciences, MA, USA) through mito stress and energy phenotype tests, respectively. Cells were plated in a Seahorse microplate at a density of 7 × 104 cells/well and treated as mentioned before. After completing the above-mentioned treatments, cells were equilibrated in DMEM without sodium bicarbonate, containing 5 mmol/L or 35 mmol/L (according to cell treatments) of d-glucose, 2 mmol/L of glutamine and 1 mmol/L of sodium pyruvate. Basal OCR and OCR in response to sequential injection of Oligomycin 1.5 μmol/L (ATP synthase inhibitor, mitochondrial complex V), FCCP 1 μmol/L (mitochondrial uncoupler), and Rotenone/Antimycin 0.5 μmol/L (mitochondrial complexes I and III inhibitors, respectively) were registered. Parameters such as proton leak, maximal respiration, spare respiratory capacity, non-mitochondrial oxygen consumption, and ATP-linked OCR were analyzed through the mito stress test to reflect mitochondrial function. Basal ECAR and Stressed ECAR in response to oligomycin injection were registered, and parameters such as baseline and stressed OCR, baseline and stressed ECAR, and the ratio between baseline OCR/ECAR and the metabolic potential were analyzed through the energy phenotype test. OCR values (pmolO2/min) and ECAR values (mpH/min) were normalized to total protein concentration per well (μg/μL) measured by a Bradford assay.
## 4.6. Statistic
All experiments were done at least in triplicate, and data are expressed as mean ± SEM. An unpaired t-test was used for comparisons between the 2 groups. A p-value < 0.05 was considered statistically significant. Graph Pad Software, San Diego, CA, was used for all analyses.
## References
1. Giglio R.V., Stoian A.P., Haluzik M., Pafili K., Patti A.M., Rizvi A.A., Ciaccio M., Papanas N., Rizzo M.. **Novel molecular markers of cardiovascular disease risk in type 2 diabetes mellitus**. *Biochim. Biophys. Acta Mol. Basis Dis.* (2021) **1867** 166148. DOI: 10.1016/j.bbadis.2021.166148
2. Galicia-Garcia U., Benito-Vicente A., Jebari S., Larrea-Sebal A., Siddiqi H., Uribe K., Ostolaza H., Martín C.. **Pathophysiology of Type 2 Diabetes Mellitus**. *Int. J. Mol. Sci.* (2020) **21**. DOI: 10.3390/ijms21176275
3. Joshi M.S., Williams D., Horlock D., Samarasinghe T., Andrews K.L., Jefferis A.-M., Berger P.J., Chin-Dusting J.P., Kaye D.M.. **Role of mitochondrial dysfunction in hyperglycaemia-induced coronary microvascular dysfunction: Protective role of resveratrol**. *Diabetes Vasc. Dis. Res.* (2015) **12** 208-216. DOI: 10.1177/1479164114565629
4. **2. Classification and Diagnosis of Diabetes:**. *Diabetes Care* (2021) **44** S15-S33. DOI: 10.2337/dc21-S002
5. Grismaldo Rodríguez A., Zamudio Rodríguez J.A., Mendieta C.V., Quijano Gómez S., Sanabria Barrera S., Morales Álvarez L.. **Effect of Platelet-Derived Growth Factor C on Mitochondrial Oxidative Stress Induced by High d-Glucose in Human Aortic Endothelial Cells**. *Pharmaceuticals* (2022) **15**. DOI: 10.3390/ph15050639
6. Eelen G., de Zeeuw P., Treps L., Harjes U., Wong B.W., Carmeliet P.. **Endothelial Cell Metabolism**. *Physiol. Rev.* (2018) **98** 3-58. DOI: 10.1152/physrev.00001.2017
7. de Zeeuw P., Wong B.W., Carmeliet P.. **Metabolic Adaptations in Diabetic Endothelial Cells**. *Circ. J.* (2015) **79** 934-941. DOI: 10.1253/circj.CJ-15-0230
8. Tenopoulou M., Doulias P.-T.. **Endothelial nitric oxide synthase-derived nitric oxide in the regulation of metabolism**. *F1000Research* (2020) **9** 1190. DOI: 10.12688/f1000research.19998.1
9. Tang X., Luo Y., Chen H.-Z., Liu D.-P.. **Mitochondria, endothelial cell function, and vascular diseases**. *Front. Physiol.* (2014) **5** 175. DOI: 10.3389/fphys.2014.00175
10. Kim D., Sankaramoorthy A., Roy S.. **Downregulation of Drp1 and Fis1 Inhibits Mitochondrial Fission and Prevents High Glucose-Induced Apoptosis in Retinal Endothelial Cells**. *Cells* (2020) **9**. DOI: 10.3390/cells9071662
11. Chen H., Chomyn A., Chan D.C.. **Disruption of Fusion Results in Mitochondrial Heterogeneity and Dysfunction**. *J. Biol. Chem.* (2005) **280** 26185-26192. DOI: 10.1074/jbc.M503062200
12. Rovira-Llopis S., Bañuls C., Diaz-Morales N., Hernandez-Mijares A., Rocha M., Victor V.M.. **Mitochondrial dynamics in type 2 diabetes: Pathophysiological implications**. *Redox Biol.* (2017) **11** 637-645. DOI: 10.1016/j.redox.2017.01.013
13. Shenouda S.M., Widlansky M.E., Chen K., Xu G., Holbrook M., Tabit C.E., Hamburg N.M., Frame A.A., Caiano T.L., Kluge M.A.. **Altered Mitochondrial Dynamics Contributes to Endothelial Dysfunction in Diabetes Mellitus**. *Circulation* (2011) **124** 444-453. DOI: 10.1161/CIRCULATIONAHA.110.014506
14. Makino A., Scott B.T., Dillmann W.H.. **Mitochondrial fragmentation and superoxide anion production in coronary endothelial cells from a mouse model of type 1 diabetes**. *Diabetologia* (2010) **53** 1783-1794. DOI: 10.1007/s00125-010-1770-4
15. Trudeau K., Molina A.J., Guo W., Roy S.. **High Glucose Disrupts Mitochondrial Morphology in Retinal Endothelial Cells: Implications for Diabetic Retinopathy**. *Am. J. Pathol.* (2010) **177** 447-455. DOI: 10.2353/ajpath.2010.091029
16. Hill B.G., Benavides G.A., Lancaster J.R., Ballinger S., Dell’Italia L., Zhang J., Darley-Usmar V.M.. **Integration of cellular bioenergetics with mitochondrial quality control and autophagy**. *Biol. Chem.* (2012) **393** 1485-1512. DOI: 10.1515/hsz-2012-0198
17. Kashatus J.A., Nascimento A., Myers L.J., Sher A., Byrne F.L., Hoehn K.L., Counter C.M., Kashatus D.F.. **Erk2 Phosphorylation of Drp1 Promotes Mitochondrial Fission and MAPK-Driven Tumor Growth**. *Mol. Cell* (2015) **57** 537-551. DOI: 10.1016/j.molcel.2015.01.002
18. Chang C.R., Blackstone C.. **Cyclic AMP-dependent protein kinase phosphorylation of Drp1 regulates its GTPase activity and mitochondrial morphology**. *J. Biol. Chem.* (2007) **282** 21583-21587. DOI: 10.1074/jbc.C700083200
19. Divakaruni A.S., Paradyse A., Ferrick D.A., Murphy A.N., Jastroch M.. **Analysis and Interpretation of Microplate-Based Oxygen Consumption and pH Data**. *Methods Enzymol.* (2014) **547** 309-354. DOI: 10.1016/b978-0-12-801415-8.00016-3
20. Susan W.S.L., Shi Y.. **The glycolytic process in endothelial cells and its implications**. *Acta Pharmacol. Sin.* (2021) **43** 251-259. DOI: 10.1038/s41401-021-00647-y
21. Li X., Sun X., Carmeliet P.. **Hallmarks of Endothelial Cell Metabolism in Health and Disease**. *Cell Metab.* (2019) **30** 414-433. DOI: 10.1016/j.cmet.2019.08.011
22. Yetkin-Arik B., Vogels I.M.C., Neyazi N., van Duinen V., Houtkooper R.H., van Noorden C.J.F., Klaassen I., Schlingemann R.O.. **Endothelial tip cells in vitro are less glycolytic and have a more flexible response to metabolic stress than non-tip cells**. *Sci. Rep.* (2019) **9** 10414. DOI: 10.1038/s41598-019-46503-2
23. Koziel A., Woyda-Ploszczyca A., Kicinska A., Jarmuszkiewicz W.. **The influence of high glucose on the aerobic metabolism of endothelial EA.hy926 cells**. *Pflüg. Arch.-Eur. J. Physiol.* (2012) **464** 657-669. DOI: 10.1007/s00424-012-1156-1
24. Xie L., Shi F., Li Y., Li W., Yu X., Zhao L., Zhou M., Hu J., Luo X., Tang M.. **Drp1-dependent remodeling of mitochondrial morphology triggered by EBV-LMP1 increases cisplatin resistance**. *Signal Transduct. Target. Ther.* (2020) **5** 1-12. DOI: 10.1038/s41392-020-0151-9
25. Han Y., Kim B., Cho U., Park I.S., Kim S.I., Dhanasekaran D.N., Tsang B.K., Song Y.S.. **Mitochondrial fission causes cisplatin resistance under hypoxic conditions via ROS in ovarian cancer cells**. *Oncogene* (2019) **38** 7089-7105. DOI: 10.1038/s41388-019-0949-5
26. Zhao Q., Lu D., Wang J., Liu B., Cheng H., Mattson M.P., Cheng A.. **Calcium dysregulation mediates mitochondrial and neurite outgrowth abnormalities in SOD2 deficient embryonic cerebral cortical neurons**. *Cell Death Differ.* (2019) **26** 1600-1614. DOI: 10.1038/s41418-018-0230-4
27. Zhang S., Gao Y., Wang J.. **Advanced glycation end products influence mitochondrial fusion-fission dynamics through RAGE in human aortic endothelial cells**. *Int. J. Clin. Exp. Pathol.* (2017) **10** 8010-8022. PMID: 31966653
28. Shen S., Wang F., Fernandez A., Hu W.. **Role of platelet-derived growth factor in type II diabetes mellitus and its complications**. *Diabetes Vasc. Dis. Res.* (2020) **17** 1479164120942119. DOI: 10.1177/1479164120942119
29. Hong S.-G., Shin J., Choi S.Y., Powers J.C., Meister B.M., Sayoc J., Son J.S., Tierney R., Recchia F.A., Brown M.D.. **Flow pattern–dependent mitochondrial dynamics regulates the metabolic profile and inflammatory state of endothelial cells**. *J. Clin. Investig.* (2022) **7** e159286. DOI: 10.1172/jci.insight.159286
30. Cai J., Wang J., Huang Y., Wu H., Xia T., Xiao J., Chen X., Li H., Qiu Y., Wang Y.. **ERK/Drp1-dependent mitochondrial fission is involved in the MSC-induced drug resistance of T-cell acute lymphoblastic leukemia cells**. *Cell Death Dis.* (2016) **7** e2459. DOI: 10.1038/cddis.2016.370
31. Haspula D., Vallejos A., Moore T.M., Tomar N., Dash R.K., Hoffmann B.R.. **Influence of a Hyperglycemic Microenvironment on a Diabetic Versus Healthy Rat Vascular Endothelium Reveals Distinguishable Mechanistic and Phenotypic Responses**. *Front. Physiol.* (2019) **10** 558. DOI: 10.3389/fphys.2019.00558
32. Clyne A.M.. **Endothelial response to glucose: Dysfunction, metabolism, and transport**. *Biochem. Soc. Trans.* (2021) **49** 313-325. DOI: 10.1042/BST20200611
33. Zeng Y., Pan Q., Wang X., Li N., Lin Y., Man F., Xiao F., Guo L.. **Impaired Mitochondrial Fusion and Oxidative Phosphorylation Triggered by High Glucose Is Mediated by Tom22 in Endothelial Cells**. *Oxidative Med. Cell. Longev.* (2019) **2019** 4508762. DOI: 10.1155/2019/4508762
34. Gay C.G., Winkles J.A.. **The half-lives of platelet-derived growth factor A- and B-chain mRNAS are similar in endothelial cells and unaffected by heparin-binding growth factor-1 or cycloheximide**. *J. Cell. Physiol.* (1991) **147** 121-127. DOI: 10.1002/jcp.1041470116
35. Stab B.R., Martinez L., Grismaldo A., Lerma A., Gutiérrez M.L., Barrera L.A., Sutachan J.J., Albarracín S.L.. **Mitochondrial Functional Changes Characterization in Young and Senescent Human Adipose Derived MSCs**. *Front. Aging Neurosci.* (2016) **8** 299. DOI: 10.3389/fnagi.2016.00299
36. Chaudhry A., Shi R., Luciani D.S.. **A pipeline for multidimensional confocal analysis of mitochondrial morphology, function, and dynamics in pancreatic β-cells**. *Am. J. Physiol. Metab.* (2020) **318** E87-E101. DOI: 10.1152/ajpendo.00457.2019
37. Valente A.J., Maddalena L.A., Robb E.L., Moradi F., Stuart J.A.. **A simple ImageJ macro tool for analyzing mitochondrial network morphology in mammalian cell culture**. *Acta Histochem.* (2017) **119** 315-326. DOI: 10.1016/j.acthis.2017.03.001
38. Schneider C.A., Rasband W.S., Eliceiri K.W.. **NIH Image to ImageJ: 25 Years of image analysis**. *Nat. Methods* (2012) **9** 671-675. DOI: 10.1038/nmeth.2089
|
---
title: 'The Effects of a Fasting Mimicking Diet on Skin Hydration, Skin Texture, and
Skin Assessment: A Randomized Controlled Trial'
authors:
- Jessica Maloh
- Min Wei
- William C. Hsu
- Sara Caputo
- Najiba Afzal
- Raja K. Sivamani
journal: Journal of Clinical Medicine
year: 2023
pmcid: PMC10003066
doi: 10.3390/jcm12051710
license: CC BY 4.0
---
# The Effects of a Fasting Mimicking Diet on Skin Hydration, Skin Texture, and Skin Assessment: A Randomized Controlled Trial
## Abstract
Diet and nutrition have been shown to impact dermatological conditions. This has increased attention toward integrative and lifestyle medicine in the management of skin health. Emerging research around fasting diets, specifically the fasting-mimicking diet (FMD), has provided clinical evidence for chronic inflammatory, cardiometabolic, and autoimmune diseases. In this randomized controlled trial, we evaluated the effects of a five-day FMD protocol, administrated once a month for three months, on facial skin parameters, including skin hydration and skin roughness, in a group of 45 healthy women between the ages of 35 to 60 years old over the course of 71 days. The results of the study revealed that the three consecutive monthly cycles of FMD resulted in a significant percentage increase in skin hydration at day 11 ($$p \leq 0.00013$$) and at day 71 ($$p \leq 0.02$$) relative to baseline. The results also demonstrated maintenance of skin texture in the FMD group compared to an increase in skin roughness in the control group ($$p \leq 0.032$$). In addition to skin biophysical properties, self-reported data also demonstrated significant improvement in components of mental states such as happiness ($$p \leq 0.003$$) and confidence (0.039). Overall, these findings provide evidence for the potential use of FMD in improving skin health and related components of psychological well-being.
## 1. Introduction
The utilization of integrative medicine and lifestyle interventions in the management of various health concerns has gained significant interest in recent years. A lifetime prevalence of 35–$69\%$ for the use of complementary medicine was found in patients who seek dermatological treatments [1]. Among the various modalities of integrative medicine, diet and nutrition play a particularly important role. By addressing the underlying dietary and nutritional factors that may contribute to skin concerns, dietary interventions in dermatology may offer a unique inside-out approach to care [2].
The relationship between diet, weight, and dermatological disease has been demonstrated in the literature. For example, high-glycemic load diets are associated with hyperinsulinemia which, in turn, can contribute to acne through increases in inflammation and androgen-mediated increases in sebum production [3]. A randomized controlled trial in those with acne found that intervention with a low-glycemic load diet improved insulin sensitivity and reduced total inflammatory lesion counts compared to control [4]. Furthermore, psoriasis has been associated with increased adiposity, with excess adipose tissue that contributes to a pro-inflammatory state [5]. A study was conducted to investigate the role of weight reduction on psoriasis with the use of a low-calorie, protein-based diet [5]. After four weeks, a significant improvement was found in measures of quality of life, pain, itch, and the extent and severity of psoriasis [5].
Fasting has gained popularity in recent years. Fasting is conventionally defined as the abstinence of food and/or beverages for periods of time, usually for several hours of a day or lasting for days [6]. A recent prospective observational study evaluated the effects of fasting during the Islamic observance of Ramadan in individuals with psoriasis [7]. Ramadan is a month-long observance associated with engaging in intermittent daily fasting, where all food and beverages are avoided from dawn to sunset. This type of fasting was found to be associated with a significant reduction in the extent and severity of psoriasis symptoms [7]. Additionally, there were notable improvements in metabolic markers, including a significant decrease in fasting plasma glucose and triglycerides and a significant increase in HDL cholesterol [7]. These findings indicate that intermittent fasting during Ramadan may have potential therapeutic benefits for individuals with psoriasis.
An alternative method of fasting that has gained attention in recent years is the Fasting Mimicking Diet (FMD). This approach involves the consumption of a specifically formulated, calorie-restricted nutrition regimen with a customized macronutrient composition, ratio, and quantity over a 5-day period. It aims to evade nutrient sensing by cells and to elicit a similar physiological response to fasting. The FMD has been shown to be safe and efficacious in both animal and human studies for various parameters and conditions [8]. A study that assessed the effects of FMD on mice revealed that after lifelong feeding with cycles of FMD, mice had lower insulin and glucose signaling, lower visceral fat and reduced pro-inflammatory cytokines [9]. There is also evidence to suggest that FMD can significantly decrease circulating levels of growth factor, contributing to its potential anti-cancer and anti-aging properties [9].
The benefits of the FMD have also been demonstrated in human clinical trials, including reductions in body weight and visceral adiposity, insulin growth factor-1 (IGF-1), serum glucose and biomarkers associated with aging [10,11]. Moreover, research has found that FMD can support gut health by promoting markers of regeneration and improving intestinal healing under inflammation-driven stress [12]. The FMD has also been shown to consistently reduce intestinal inflammation, increase stem cell number, and promote the growth of protective gut microbiota, such as Lactobacillaceae and Bifidobacteriacea while modifying the presence of other microbes in mice [12].
A number of studies have examined the effects of fasting-mimicking diets in various conditions, including inflammatory bowel disease, diabetes, cardiovascular disease, cancer, multiple sclerosis, Alzheimer’s disease, and depression in animal models and human studies [11,12,13,14,15,16,17,18,19,20,21,22,23]. However, there are currently no human clinical trials assessing the effect of an FMD in dermatology. This is the first randomized controlled trial to assess the efficacy of a five-day FMD protocol done once a month for three cycles on facial skin parameters such as skin hydration and skin roughness in a group of women over the course of 71 days.
## 2.1. Subject Enrollment
A total of 45 female participants, ranging in age from 35 to 60 years old, were recruited for the study. This protocol was approved by the Advarra Institutional Review Board (24 November 2021). All participants provided written consent prior to participation. Inclusion criteria for participants were as follows: good general health and social well-being, Fitzpatrick skin type I-VI, and mild to moderate scores for parameters on the global face (fine dry lines, roughness, uneven skin tone, and lack of radiance). Exclusion criteria included a known allergy to any component of the meal kit, breastfeeding, pregnancy or planned pregnancy during the study, history of gastric bypass, or history of skin cancer. Having a health condition and/or pre-existing or dormant dermatologic disease on the face (ex: psoriasis, rosacea acne, eczema, seborrheic dermatitis, and severe excoriations) that the investigator deemed inappropriate for participation or deemed to interfere with the outcome of the study also served as an exclusion.
## 2.2. Randomization and Blinding
The participants were randomized into two groups, with 24 individuals in the intervention group and 21 subjects in the control group. In order to ensure the integrity of the study, the study products were dispensed by a separate individual, not involved in the investigation or evaluation process. Additionally, the research participants and the research staff responsible for dispensing the study products were instructed to refrain from discussing study products with the investigator or other evaluator(s). This allowed the evaluators to be blinded during evaluations.
## 2.3. Participant Instructions and Group Interventions
All participants were instructed to maintain their regular exercise and physical activity habits, avoid prolonged sun exposure and use of tanning beds or sunless tanning products. Participants were also asked to continue their regular use of cosmetics, makeup, and sunscreen and to refrain from the use of new facial products for the duration of the study.
The intervention group was provided with ProLonTM, a fasting-mimicking diet product, to be consumed for five consecutive days, with the first usage beginning on day 1, followed by an additional usage on day 30, and on day 60. ProLon was intended to replace usual meals during the five-day period. Following the five-day period, participants were instructed to keep meals light on day 6 and resume normal eating habits on day 7. The composition of the test product and daily meal plans are detailed in Table 1. The control group did not receive any specific diet and was instructed to maintain their habitual diet for the duration of the study. Participants were evaluated during five visits throughout the study, with the first evaluation on day 0, the second evaluation on day 11, the third evaluation on day 30, the fourth evaluation on day 60, and the fifth and final evaluation on day 71.
Diaries were provided to each participant in the intervention group at the beginning of the study to record their consumption of the test product. Participants were also asked to document any additional food or beverage items consumed during the five-day period beyond the provided meal kit and to note any changes in eating patterns between the fasting cycles.
## 2.4. Skin Assessments
Prior to any skin assessments, it was ensured that all subjects did not have any makeup or topical products on their face and were allowed to acclimate to ambient conditions within the clinic for at least 15 min. The designated rooms were maintained at a temperature range of 68–75 °F and a relative humidity range of 35–$65\%$.
Individuals in both groups completed a 22-item 5-point Likert scale self-assessment questionnaire to evaluate various skin parameters, including texture, hydration and skin tone. In order to measure skin hydration, Corneometer (CK Electronic, Köln, Germany) measurements were taken on the center of each subject’s right cheek. Antera 3D (Miravex, Dublin, Ireland) imaging was performed on the skin surface with multi-directional illumination and computer-aided reconstruction of the skin surface. Texture (Ra, or mean roughness) was analyzed using the Antera 3D CS software.
## 2.5. Statistics
Sample size estimation was based on prior similar dietary intervention studies [11,24]. Demographic and clinical characteristics were described according to the study group using means and standard deviations for continuous variables and count with frequencies for categorical variables. A descriptive statistical summary was provided for all efficacy grading parameters, bio instrumentation (Corneometer) measurements, Antera image analysis, and self-assessment questionnaires with baseline response data. The descriptive statistical summary included the sample mean, median, SD, MIN, and MAX of scores/values at all applicable time points. Questionnaires were tabulated, and the frequency and percentage of all response options were reported for each question and time point. For questionnaires without baseline response data, a binomial (sign) test was performed to test if the proportion of the combined designated favorable responses is equal to the combined designated unfavorable responses for each applicable question. Corneometer and Antera 3D imaging measurement results were tested using paired t-test and 2 sample t-test for the change from the baseline and between groups, respectively. Self-assessment measures were tested by the Wilcoxon signed rank test and Wilcoxon rank sum test to assess the change from baseline and between groups, respectively. All statistical tests were 2-sided at significance level alpha = 0.05 unless specified otherwise. Statistical analyses were performed using SAS software version 9.4 (SAS Statistical Institute).
## 3. Results
Out of the 45 participants that were screened for eligibility, all 45 met the criteria and were subsequently enrolled in the study. The 21 subjects randomized into the control group all completed the study, whereas 22 out of the 24 subjects randomized into the treatment group completed the study. A CONSORT diagram is presented in Figure 1 to illustrate the flow of participants throughout the study.
## 3.1. Demographics
The demographics of the study population are outlined in Table 2.
## 3.2. Skin Hydration
When assessing the percent increase in skin hydration, there was a significantly greater increase in the treatment group ($25.1\%$) relative to the control ($8.52\%$) at day 11, which corresponds to five days after completing the first ProLon cycle and after the resumption of subjects’ habitual diet ($$p \leq 0.024$$). Additionally, there was a statistically significant increase in percent change for skin hydration at both day 11 and day 71 in the treatment group relative to baseline ($$p \leq 0.00013$$ and $$p \leq 0.02$$, respectively). However, in the control group, a statistically significant increase was only observed at day 71 compared to the baseline ($$p \leq 0.0098$$) (Figure 2).
## 3.3. Skin Roughness
Skin texture, as determined by mean roughness (Ra), did not exhibit a significant change in the treatment group at day 11 or at day 71 relative to baseline. However, mean roughness was found to increase in the control group significantly at day 71 relative to baseline ($$p \leq 0.032$$) (Figure 3).
## 3.4. Subjective Outcomes
After completion of 3 ProLon cycles, subjects in the intervention group reported a significant improvement in various aspects of skin health, including [1] skin texture ($p \leq 0.001$), [2] smoothness ($p \leq 0.001$), [3] hydration ($$p \leq 0.021$$), and [9] skin tone evenness ($$p \leq 0.021$$), along with other skin improvements presented in Figure 4. In addition, there were significant improvements in some aspects of mood and self-perception (Figure 4). Specifically, subjects in the treatment group reported significant improvements in feelings of [17] happiness ($$p \leq 0.003$$), [19] confidence (0.039), [15] empowerment to take control of their health ($p \leq 0.001$), and [14] optimism about the future ($$p \leq 0.002$$).
Overall, the ProLon intervention was well tolerated. The adverse effects reported during the study were not deemed by the investigators to be serious events and were not deemed to be associated with the intervention. They included one case of each of the following: left-hand paresthesia, left leg pain, arthralgia, pyelonephritis, and knee ligament rupture, all of which resolved.
## 4. Discussion
This study aimed to evaluate the effect of fasting-mimicking diet with monthly ProLon on the skin. The results indicate that FMD has a beneficial effect on objective measures of skin parameters such as hydration and texture. Additionally, there was an improvement in multiple self-reported outcomes related to skin appearance and general well-being.
Although FMD has not been previously studied for its effects on skin health, there is support for the use of caloric restriction in improving skin anatomy and function. Caloric restriction with fasting has demonstrated improvement in multiple skin properties, including skin barrier function in both mice and humans [25]. Studies have also suggested that caloric restriction may improve the appearance of wrinkles and decrease the presence of oxidative stress [26,27].
The potential mechanisms by which FMD may exert its effects on the skin are multifaceted. Fasting has been shown to initiate comprehensive cellular and systemic reprogramming in organisms in response to starvation conditions. Biogerontological research in the past 30 years has linked prolonged nutrient deprivation with the downregulation of pro-growth signaling and activation of cellular protection mechanisms, which may have implications for the amelioration of disease-associated factors and the delay of aging [8,28]. The FMD was specifically designed to mimic the effects of water-only fasting and had been shown to induce anti-oxidative stress in cells [29,30], dampen the mTOR-S6K signaling pathway, activate autophagy [31,32,33], promote stem cell-based regenerations in multiple tissues [12,14,34], augment the gut microbiome [12], and reduce risk factors associated with age-related diseases [11]. Further research on the use of period fasting interventions such as FMD may reveal it to be a cost-effective and feasible component of an integrative approach to skin health.
Another potential mechanism by which FMD may impact skin health is through the gut-skin axis. Research suggests that subsequent cycles of FMD may reduce intestinal inflammation and stimulate protective members of the gut microbiome, such as Lactobacillaceae and Bifidobacteriacea [12]. These specific members of the gut microbiota have been found to be relevant to skin health. For example, children with eczema have been found to have less gut colonization by Bifidobacterium and Lactobacillus strains relative to healthy control [35,36]. Furthermore, in an animal study, oral supplementation with Bifidobacterium breve B-3, a member of the Bifidobacteriacea family, has been found to protect against UV-induced changes in transepidermal water loss and changes in skin hydration [37]. However, additional research will be needed to better understand the relationship between fasting-mediated shifts in the gut microbiome and changes in skin outcomes.
In this study, the self-reported data was found to be consistent with the objective skin biophysical measurements. For example, the participants in the treatment group reported improvement in skin hydration, which was in alignment with the objective measure of the studies. This suggests that the extent of skin improvement was to a noticeable degree, which may imply clinical relevance. Furthermore, the results from self-assessments suggest that the FMD impacted aspects of mental well-being and self-esteem, demonstrated by improvements in the feeling of happiness, confidence, and attractiveness. Previous studies investigating FMD have also found improvements in mental states. In one study, patients with depression that received both FMD and psychotherapy had an improvement in self-esteem and psychological quality of life compared to a group receiving psychotherapy alone [23]. It is interesting to note that our FMD study reported improvement in aspects of self-esteem in generally healthy subjects.
This trial serves as a pilot study to demonstrate the effects of FMD on skin health, and the parameters assessed warrant further investigation with a larger sample size. Furthermore, because this study was done with a healthy population, future research should focus on individuals with a skin condition to better understand the use of the FMD in the setting of dermatological disease. Additionally, future research can expand the study population to include male participants. With evidence suggesting improvements in aspects of mental well-being, further research on the FMD and skin health should incorporate validated questionnaires for mood to better understand the skin-mind axis in the context of fasting.
## 5. Conclusions
The results of this study are the first to investigate the role of fasting-mimicking interventions and, specifically, FMD for skin health and appearance in a randomized controlled trial. Skin hydration and skin roughness were found to be supported with the use of the FMD, along with self-reported improvements in components of mental wellness, including happiness, confidence, and optimism. Overall, FMD is found to be a promising intervention in the realm of integrative and lifestyle medicine for skin health. Future studies should evaluate this intervention in individuals with skin diseases to better understand the mechanisms and effects of FMD in clinical dermatology settings.
## References
1. Ernst E.. **The usage of complementary therapies by dermatological patients: A systematic review**. *Br. J. Dermatol.* (2000) **142** 857-861. DOI: 10.1046/j.1365-2133.2000.03463.x
2. Katta R., Desai S.P.. **Diet and dermatology: The role of dietary intervention in skin disease**. *J. Clin. Aesthetic Dermatol.* (2014) **7** 46-51
3. Meixiong J., Ricco C., Vasavda C., Ho B.K.. **Diet and acne: A systematic review**. *JAAD Int.* (2022) **7** 95-112. DOI: 10.1016/j.jdin.2022.02.012
4. Smith R.N., Mann N.J., Braue A., Mäkeläinen H., Varigos G.A.. **A low-glycemic-load diet improves symptoms in acne vulgaris patients: A randomized controlled trial**. *Am. J. Clin. Nutr.* (2007) **86** 107-115. DOI: 10.1093/ajcn/86.1.107
5. Castaldo G., Pagano I., Grimaldi M., Marino C., Molettieri P., Santoro A., Stillitano I., Romano R., Montoro P., D’Ursi A.M.. **Effect of Very-Low-Calorie Ketogenic Diet on Psoriasis Patients: A Nuclear Magnetic Resonance-Based Metabolomic Study**. *J. Proteome Res.* (2021) **20** 1509-1521. DOI: 10.1021/acs.jproteome.0c00646
6. Santos H.O., Macedo R.C.O.. **Impact of intermittent fasting on the lipid profile: Assessment associated with diet and weight loss**. *Clin. Nutr. ESPEN* (2018) **24** 14-21. DOI: 10.1016/j.clnesp.2018.01.002
7. Almutairi N., Shaaban D.. **Clinical implications of intermittent Ramadan fasting on stable plaque psoriasis: A prospective observational study**. *Postep. Dermatol. Alergol.* (2022) **39** 368-374. DOI: 10.5114/ada.2021.107098
8. Longo V.D., Panda S.. **Fasting, Circadian Rhythms, and Time-Restricted Feeding in Healthy Lifespan**. *Cell Metab.* (2016) **23** 1048-1059. DOI: 10.1016/j.cmet.2016.06.001
9. Brandhorst S.. **Fasting and fasting-mimicking diets for chemotherapy augmentation**. *Geroscience* (2021) **43** 1201-1216. DOI: 10.1007/s11357-020-00317-7
10. Brandhorst S., Longo V.D.. **Protein Quantity and Source, Fasting-Mimicking Diets, and Longevity**. *Adv. Nutr.* (2019) **10** S340-S350. DOI: 10.1093/advances/nmz079
11. Wei M., Brandhorst S., Shelehchi M., Mirzaei H., Cheng C.W., Budniak J., Groshen S., Mack W.J., Guen E., Di Biase S.. **Fasting-mimicking diet and markers/risk factors for aging, diabetes, cancer, and cardiovascular disease**. *Sci. Transl. Med.* (2017) **9** eaai8700. DOI: 10.1126/scitranslmed.aai8700
12. Rangan P., Choi I., Wei M., Navarrete G., Guen E., Brandhorst S., Enyati N., Pasia G., Maesincee D., Ocon V.. **Fasting-Mimicking Diet Modulates Microbiota and Promotes Intestinal Regeneration to Reduce Inflammatory Bowel Disease Pathology**. *Cell Rep.* (2019) **26** 2704-2719. DOI: 10.1016/j.celrep.2019.02.019
13. Sulaj A., Kopf S., Rauchhaupt E.v., Kliemank E., Brune M., Kender Z., Bartl H., Cortizo F.G., Klepac K., Han Z.. **A six-month periodic fasting reduces microalbuminuria and improves metabolic control in patients with type 2 diabetes and diabetic nephropathy: A randomized controlled study**. *medRxiv* (2021). DOI: 10.2337/db22-389-P
14. Cheng C.W., Villani V., Buono R., Wei M., Kumar S., Yilmaz O.H., Cohen P., Sneddon J.B., Perin L., Longo V.D.. **Fasting-Mimicking Diet Promotes Ngn3-Driven beta-Cell Regeneration to Reverse Diabetes**. *Cell* (2017) **168** 775-788. DOI: 10.1016/j.cell.2017.01.040
15. Choi I.Y., Piccio L., Childress P., Bollman B., Ghosh A., Brandhorst S., Suarez J., Michalsen A., Cross A.H., Morgan T.E.. **A Diet Mimicking Fasting Promotes Regeneration and Reduces Autoimmunity and Multiple Sclerosis Symptoms**. *Cell Rep.* (2016) **15** 2136-2146. DOI: 10.1016/j.celrep.2016.05.009
16. de Groot S., Lugtenberg R.T., Cohen D., Welters M.J.P., Ehsan I., Vreeswijk M.P.G., Smit V., de Graaf H., Heijns J.B., Portielje J.E.A.. **Fasting mimicking diet as an adjunct to neoadjuvant chemotherapy for breast cancer in the multicentre randomized phase 2 DIRECT trial**. *Nat. Commun.* (2020) **11** 3083. DOI: 10.1038/s41467-020-16138-3
17. Lugtenberg R.T., de Groot S., Kaptein A.A., Fischer M.J., Kranenbarg E.M., Carpentier M.D., Cohen D., de Graaf H., Heijns J.B., Portielje J.E.A.. **Quality of life and illness perceptions in patients with breast cancer using a fasting mimicking diet as an adjunct to neoadjuvant chemotherapy in the phase 2 DIRECT (BOOG 2013-14) trial**. *Breast Cancer Res. Treat.* (2021) **185** 741-758. DOI: 10.1007/s10549-020-05991-x
18. Mishra A., Mirzaei H., Guidi N., Vinciguerra M., Mouton A., Linardic M., Rappa F., Barone R., Navarrete G., Wei M.. **Fasting-mimicking diet prevents high-fat diet effect on cardiometabolic risk and lifespan**. *Nat. Metab.* (2021) **3** 1342-1356. DOI: 10.1038/s42255-021-00469-6
19. Salvadori G., Zanardi F., Iannelli F., Lobefaro R., Vernieri C., Longo V.D.. **Fasting-mimicking diet blocks triple-negative breast cancer and cancer stem cell escape**. *Cell Metab.* (2021) **33** 2247-2259.e2246. DOI: 10.1016/j.cmet.2021.10.008
20. van den Burg E.L., Schoonakker M.P., van Peet P.G., van den Akker-van Marle M.E., Willems van Dijk K., Longo V.D., Lamb H.J., Numans M.E., Pijl H.. **Fasting in diabetes treatment (FIT) trial: Study protocol for a randomised, controlled, assessor-blinded intervention trial on the effects of intermittent use of a fasting-mimicking diet in patients with type 2 diabetes**. *BMC Endocr. Disord.* (2020) **20**. DOI: 10.1186/s12902-020-00576-7
21. Fay-Watt V., O’Connor S., Roshan D., Romeo A.C., Longo V.D., Sullivan F.J.. **The impact of a fasting mimicking diet on the metabolic health of a prospective cohort of patients with prostate cancer: A pilot implementation study**. *Prostate Cancer Prostatic Dis.* (2022). DOI: 10.1038/s41391-022-00528-3
22. Rangan P., Lobo F., Parrella E., Rochette N., Morselli M., Stephen T.L., Cremonini A.L., Tagliafico L., Persia A., Caffa I.. **Fasting-mimicking diet cycles reduce neuroinflammation to attenuate cognitive decline in Alzheimer’s models**. *Cell Rep.* (2022) **40** 111417. DOI: 10.1016/j.celrep.2022.111417
23. Maniaci G., La Cascia C., Giammanco A., Ferraro L., Chianetta R., Di Peri R., Sardella Z., Citarrella R., Mannella Y., Larcan S.. **Efficacy of a fasting-mimicking diet in functional therapy for depression: A randomised controlled pilot trial**. *J. Clin. Psychol.* (2020) **76** 1807-1817. DOI: 10.1002/jclp.22971
24. Huang A.W., Wei M., Caputo S., Wilson M.L., Antoun J., Hsu W.C.. **An Intermittent Fasting Mimicking Nutrition Bar Extends Physiologic Ketosis in Time Restricted Eating: A Randomized, Controlled, Parallel-Arm Study**. *Nutrients* (2021) **13**. DOI: 10.3390/nu13051523
25. Bragazzi N.L., Sellami M., Salem I., Conic R., Kimak M., Pigatto P.D.M., Damiani G.. **Fasting and Its Impact on Skin Anatomy, Physiology, and Physiopathology: A Comprehensive Review of the Literature**. *Nutrients* (2019) **11**. DOI: 10.3390/nu11020249
26. Choi Y.J.. **Shedding Light on the Effects of Calorie Restriction and its Mimetics on Skin Biology**. *Nutrients* (2020) **12**. DOI: 10.3390/nu12051529
27. Bhattacharyya T.K., Hsia Y., Weeks D.M., Dixon T.K., Lepe J., Thomas J.R.. **Association of Diet With Skin Histological Features in UV-B-Exposed Mice**. *JAMA Facial Plast. Surg.* (2017) **19** 399-405. DOI: 10.1001/jamafacial.2017.0060
28. Longo V.D., Mattson M.P.. **Fasting: Molecular mechanisms and clinical applications**. *Cell Metab.* (2014) **19** 181-192. DOI: 10.1016/j.cmet.2013.12.008
29. Wei M., Fabrizio P., Hu J., Ge H., Cheng C., Li L., Longo V.D.. **Life span extension by calorie restriction depends on Rim15 and transcription factors downstream of Ras/PKA, Tor, and Sch9**. *PLoS Genet.* (2008) **4**. DOI: 10.1371/journal.pgen.0040013
30. Raffaghello L., Lee C., Safdie F.M., Wei M., Madia F., Bianchi G., Longo V.D.. **Starvation-dependent differential stress resistance protects normal but not cancer cells against high-dose chemotherapy**. *Proc. Natl. Acad. Sci. USA* (2008) **105** 8215-8220. DOI: 10.1073/pnas.0708100105
31. Eisenberg T., Knauer H., Schauer A., Buttner S., Ruckenstuhl C., Carmona-Gutierrez D., Ring J., Schroeder S., Magnes C., Antonacci L.. **Induction of autophagy by spermidine promotes longevity**. *Nat. Cell Biol.* (2009) **11** 1305-1314. DOI: 10.1038/ncb1975
32. Madeo F., Tavernarakis N., Kroemer G.. **Can autophagy promote longevity?**. *Nat. Cell Biol.* (2010) **12** 842-846. DOI: 10.1038/ncb0910-842
33. Brandhorst S., Choi I.Y., Wei M., Cheng C.W., Sedrakyan S., Navarrete G., Dubeau L., Yap L.P., Park R., Vinciguerra M.. **A Periodic Diet that Mimics Fasting Promotes Multi-System Regeneration, Enhanced Cognitive Performance, and Healthspan**. *Cell Metab.* (2015) **22** 86-99. DOI: 10.1016/j.cmet.2015.05.012
34. Cheng C.W., Adams G.B., Perin L., Wei M., Zhou X., Lam B.S., Da Sacco S., Mirisola M., Quinn D.I., Dorff T.B.. **Prolonged fasting reduces IGF-1/PKA to promote hematopoietic-stem-cell-based regeneration and reverse immunosuppression**. *Cell Stem Cell* (2014) **14** 810-823. DOI: 10.1016/j.stem.2014.04.014
35. Melli L.C.F.L., Carmo-Rodrigues M.S.D., Araújo-Filho H.B., Mello C.S., Tahan S., Pignatari A.C.C., Solé D., Morais M.B.. **Gut microbiota of children with atopic dermatitis: Controlled study in the metropolitan region of São Paulo, Brazil**. *Allergol. Immunopathol.* (2020) **48** 107-115. DOI: 10.1016/j.aller.2019.08.004
36. Watanabe S., Narisawa Y., Arase S., Okamatsu H., Ikenaga T., Tajiri Y., Kumemura M.. **Differences in fecal microflora between patients with atopic dermatitis and healthy control subjects**. *J. Allergy Clin. Immunol.* (2003) **111** 587-591. DOI: 10.1067/mai.2003.105
37. Satoh T., Murata M., Iwabuchi N., Odamaki T., Wakabayashi H., Yamauchi K., Abe F., Xiao J.Z.. **Effect of Bifidobacterium breve B-3 on skin photoaging induced by chronic UV irradiation in mice**. *Benef. Microbes* (2015) **6** 497-504. DOI: 10.3920/BM2014.0134
|
---
title: 'Laparoendoscopic Single-Site Inguinal Herniorrhaphy: Experience of a Single
Institute'
authors:
- Wei-Quen Tee
- Yen-Ting Wu
- Hung-Jen Wang
- Yao-Chi Chuang
- Wei-Chia Lee
- Chia-Hung Tsai
- Long-Yuan Lee
- Chien-Hsu Chen
journal: Journal of Clinical Medicine
year: 2023
pmcid: PMC10003067
doi: 10.3390/jcm12051786
license: CC BY 4.0
---
# Laparoendoscopic Single-Site Inguinal Herniorrhaphy: Experience of a Single Institute
## Abstract
Background: Minimally invasive techniques for inguinal herniorrhaphy have focused on developing the laparoendoscopic single-site (LESS) procedure to improve cosmesis. Outcomes of total extraperitoneal (TEP) herniorrhaphy vary considerably because of being performed by different surgeons. We aimed to evaluate the perioperative characteristics and outcomes of patients undergoing the LESS-TEP approach for inguinal herniorrhaphy and to determine its overall safety and effectiveness. Methods: Data of 233 patients who underwent 288 laparoendoscopic single-site total extraperitoneal approach (LESS-TEP) herniorrhaphies at Kaohsiung Chang Gung Memorial Hospital between January 2014 and July 2021 were reviewed retrospectively. We reviewed the experiences and results of LESS-TEP herniorrhaphy performed by a single surgeon (CHC) using homemade glove access and standard laparoscopic instruments with a 50 cm long 30° telescope. Results: Among 233 patients, 178 patients had unilateral hernias and 55 patients had bilateral hernias. About $32\%$ ($$n = 57$$) of patients in the unilateral group and $29\%$ ($$n = 16$$) of patients in the bilateral group were obese (body mass index ≥ 25). The mean operative time was 66 min for the unilateral group and 100 min for the bilateral group. Postoperative complications occurred in 27 ($11\%$) cases, which were minor morbidities except for one mesh infection. Three ($1.2\%$) cases were converted to open surgery. Comparison of the variables between obese and non-obese patients found no significant differences in operative times or postoperative complications. Conclusion: LESS-TEP herniorrhaphy is a safe and feasible operation with excellent cosmetic results and a low rate of complication, even in obese patients. Further large-scale prospective controlled studies and long-term analyses are needed to confirm these results.
## 1. Introduction
Inguinal herniorrhaphy is one of the most frequently performed surgeries worldwide. Compared to conventional open repair surgery, laparoscopic inguinal herniorrhaphy has demonstrated promising outcomes, including alleviating postoperative pain, allowing an earlier return to normal activities with a shorter length of hospital stay, a better cosmetic result, and improved quality of life in the postoperative period [1,2]. However, it has a longer learning curve and higher costs [2].
A laparoscopic procedure generally has two approaches: transabdominal preperitoneal (TAPP) or total extraperitoneal (TEP). The TEP approach has a steep learning curve due to the cramped and unfamiliar visual field of the preperitoneal space. However, at the same time, the approach reduces unnecessary exposure of the bowel and the risk of serious visceral injury compared to TAPP [3].
Minimally invasive techniques have focused on developing the laparoendoscopic single-site (LESS) procedure to improve cosmetic outcomes. This single-incision technique is a less invasive alternative compared to conventional laparoscopic surgery, requiring only one incision over the umbilical fold. Each incision carries some risk of morbidities, such as bleeding and iatrogenic injury to the internal abdominal organ and vessels during the operation. Therefore, a greater number of incisions is associated with an increased risk of port-related morbidities and poorer cosmetic results [4,5]. Cosmesis is an important issue for many patients. A published survey of 750 patients highlighted that patients desire better cosmetic outcomes [6].
Previous studies regarding the outcomes of LESS-TEP vary considerably because different surgeons performed the procedures. In this study, the experiences and results of LESS-TEP performed in a single hospital by a single surgeon (CHC) have been reviewed.
The purpose of thisstudy was to evaluate the perioperative characteristics and outcomes of patients undergoing the extraperitoneal approach (LESS-TEP) for herniorrhaphy and to determine the overall safety and effectiveness of the procedure.
## 2. Patients and Methods
Data of patients who underwent a laparoendoscopic single-site total extraperitoneal approach (LESS-TEP) herniorrhaphy in Kaohsiung Chang Gung Memorial Hospital (KCGMH) between January 2014 and July 2021 were reviewed retrospectively. All procedures were performed by a single surgeon, Dr. C. H. Chen. The patients’ demographic data, body mass index (BMI), underlying history, history of intra-abdominal operations, American Society of Anesthesiologists (ASA) grade of physical status, hernia size, intraoperative data, and postoperative data were collected retrospectively. The study was approved by the Institutional Review Board of the KCGMH (No.: 202200718B0). Because of the retrospective nature of the study, signed informed consent of the patients was waived.
Hernia size was evaluated in the outpatient department. When evaluating the sizes of the inguinal hernias, patients were requested to stand up and perform the Valsalva maneuver for at least 10 min. The superficial inguinal ring (SIR) was defined as a boundary. The inguinal sac beyond the SIR or into the scrotum was the infra-SIR type, while the hernial sac above the SIR was the supra-SIR type (Figure 1). The pantaloon hernia type referred to a coincidence of indirect and direct hernia over the ipsilateral groin.
The numerical pain rating scale was used to evaluate the pain scale on postoperative day 1. Pain killers with Non-Steroidal Anti-Inflammatory Drugs (NSAIDs) or acetaminophen were administered to all patients. Additional analgesics were defined as intravenous injections of parecoxib or intramuscular injections of morphine. All patients were required to return to the outpatient department within 7 days of discharge. The wound condition was evaluated during the outpatient department visit. If there was no morbidity, treatment was terminated. However, patients with complications had to revisit the outpatient department until the morbidity was resolved. Patients were also requested to revisit the outpatient department if hernia recurrence was suspected.
## 2.1. Surgical Technique
The patient was placed in a supine position under general anesthesia. A 1.5–2 cm incision was made at the infra-umbilical edge as a quadrant circle line. The incision was carried down to the anterior sheath of the abdominal rectus. A slight space was created between the transversalis fascia and peritoneum. A balloon dilator was used to create a preperitoneal space. Then, a wound retractor (LAGIS® WR-60ES) was placed and homemade powder-free glove access (Figure 2) using an 11 mm trocar and two 5 mm trocars were attached to the retractor. The preperitoneal space was insufflated to 10 mmHg. A 50 cm long Hopkins Forward-Oblique Telescope 30° was inserted and manipulated using laparoscopic dissectors.
The preperitoneal space was cleared out through the internal ring to identify landmarks such as the inferior epigastric vessel, pubic bone, and internal inguinal ring. After completing the preperitoneal dissection, sac isolation was performed. In the case of a large hernial sac or adhesion, the hernial sac was divided just beyond the internal inguinal ring, and then transected after the peritoneum was closed using the Weck Hem-o-lock polymer ligation system (Hem-o-lok, Teleflex, Wayne, PA, USA). If peritoneal tears were noted, they were closed using the Hem-o-lok (Figure 3) or sutures. After parietalization of the spermatic cord, mesh positioning was performed. A 15 × 10 cm parietex hydrophilic anatomical polyester mesh (Medtronic, Minneapolis, MN, USA) or 3 × 6 inch monofilament polypropylene mesh (Davol, Bard, Warwick, RI, USA) was placed without wrinkles. The choice of mesh was dependent on the patient’s economic status (the anatomical mesh costs around USD 520 while polypropylene mesh would be reimbursed by the Taiwanese National Health Insurance). We used an Absorbatack fixation device (Covidien, Medtronics, Minneapolis, MN, USA) to fix the monofilament polypropylene mesh. However, the parietex hydrophilic anatomical mesh was designed to encircle the gonadal vessels and vas deferens, which combined 2D and 3D weave to reduce mesh mobilization. Therefore, no fixation device was used to fix the parietex hydrophilic anatomical mesh. The space was deflated while monitoring the mesh to ensure it was in place. Then the homemade glove access and retractor were removed and the fascia was sutured layer by layer. The single skin incision was closed with subcuticular sutures (Figure 4).
## 2.2. Statistical Analysis
Statistical analysis was performed using IBM SPSS statistics Base 22.0 software (IBM Corp. Released 2013. IBM SPSS Statistics for Windows, Version 22.0 Armonk, NT, USA: IBM Corp). For the analysis of clinical characteristics and variables between obese and non-obese groups, we used an independent t-test for continuous variables and Chi-square or Fisher’s exact test for categorical variables. $p \leq 0.05$ was considered statistically significant.
## 3.1. Patients’ Demographic and Clinical Data
A total of 288 LESS-TEP herniorrhaphies were performed on 233 patients (unilateral: 178 patients, bilateral: 55 patients). The patients’ demographic data and hernia characteristics are summarized in Table 1. Most patients were male with a mean age of 57 ± 14 years in the unilateral group and 64 ± 10 years in the bilateral group. About $32\%$ ($$n = 57$$) of the patients in the unilateral group were obese (BMI ≥ 25), while $29\%$ ($$n = 16$$) of the patients in the bilateral group were obese [7]. Most patients in both groups (unilateral: $75.3\%$; bilateral: $78.2\%$) had an ASA physical status of grade II. About $16\%$ of patients had a history of abdominal surgery in the unilateral group and about $18\%$ in the bilateral group. The history of abdominal surgeries included appendectomy, cholecystectomy, colectomy, and others. Twenty-three percent ($$n = 41$$) of patients in the unilateral group had an infra-SIR type hernia while $27\%$ ($$n = 30$$) in the bilateral group had one.
## 3.2. Perioperative Data
The perioperative data were shown in Table 2. Mean operative time was 66 ± 26 min in the unilateral group and 100 ± 39 min in the bilateral group. The most common hernia type was indirect ($75.8\%$) in the unilateral group while in the bilateral group it was the direct type ($54.5\%$). Prior herniorrhaphy cases occurred in less than $5\%$ of both groups. Incarcerated or femoral hernias were not found. Peritoneal tears were $19.6\%$ in the unilateral group and $14.5\%$ in the bilateral group. Any size of peritoneal perforation found during the surgery was marked as a peritoneal tear. The blood loss in this procedure was minimal and no patient needed blood transfusion. A total of three cases were converted to open surgery due to severe adhesion.
## 3.3. Postoperative Outcomes and Complications
Postoperative outcomes and complications are listed in Table 3. The mean hospital stay after surgery was 1 day in both groups. Two patients underwent outpatient surgeries in the unilateral group. The mean numerical pain rating scales on postoperative day 1 in unilateral and bilateral group were 1.6 and 1.9, respectively. A total of $16.85\%$ in the unilateral group and $12.7\%$ in the bilateral group of patients needed an additional analgesic agent. Prolonged spermatic cord pain was defined as the need for oral painkillers one week after surgery. Prolonged spermatic cord pain was $6.18\%$ in the unilateral group and $10.9\%$ in the bilateral group. Among the seven cases that experienced inguinal seroma, two required fine-needle percutaneous aspiration while others resolved spontaneously. The median follow-up period was 1.14 weeks (range, 1~326). One patient experienced a delayed mesh abscess 3 years postoperatively. He was treated successfully with mesh removal and debridement. In this study, only one patient experienced recurrence at the surgical side one year after surgery and he received open herniorrhaphy for recurrence. However, five patients experienced recurrence over another side (non-surgical side) and all of them received open repair in consideration of adhesion due to prior laparoscopic surgery. Currently, we routinely check the contralateral side when performing unilateral LESS-TEP. If a contralateral hernia is noted during surgery, we perform bilateral repair to avoid reoperation in the future.
In the unilateral group, the perioperative and postoperative data between obese patients (BMI ≥ 25) and non-obese patients (BMI < 23) are listed in Table 4. There were 57 patients in the obesity group and 60 patients in the non-obesity group. No significant differences were found between the two groups in the parameters of hernia size, operative time, pain rating scale, and complications.
## 4. Discussion
Controversy still exists about the best surgical repair for inguinal hernias. Therefore, we undertook this study because we considered it worthwhile from the perspective of LESS-TEP safety to clarify the outcomes of LESS-TEP. *We* generally favored LESS-TEP because we were able to manipulate the hernial sac without going through the intra-peritoneal cavity, which lowered the risk of complications such as visceral injury, intestinal obstruction, and port-site hernia, as previously described [3,8,9]. In the recent literature, long-term outcomes of chronic groin pain, hernia recurrence, and quality of life were comparable between TEP and TAPP [10,11]. Another systematic review and meta-analysis showed comparable surgical efficacy and morbidity between LESS-TEP and the conventional multiple-port TEP, except for cosmesis [12]. Therefore, the present study focused on LESS-TEP and reviewed the outcomes of surgical cases that were performed at our institution.
In this study, the total complication rate was $11\%$, which was compatible with those of previous studies [13,14,15]. We defined prolonged spermatic cord pain as spermatic cord induration and continued pain requiring painkillers for relief over one week after surgery. In most patients, painkillers (such as NSAIDs or acetaminophen) were prescribed for 3 days. This pain may be the result of transient inflammation due to intraoperative hernia sac dissection. Roland et al. reported that reduction of the hernia surface area may cause chronic pain [16]. However, all patients with prolonged spermatic cord pain experienced subsided pain 2–3-weeks after postoperative follow-up.
The incidence of seroma formation after herniorrhaphy varies from $0.5\%$ to $12.2\%$ [3]. The risk factors for seroma formation are coagulopathy, congestive liver disease, and cardiac insufficiency [17]. Patients can develop fluid accumulation at the space where the hernia sac used to be and the fluid reabsorbs spontaneously with time in most cases. Therefore, most seromas resolve spontaneously in 6–8 weeks [3]. In the present study, seven seroma cases were noted and two of them needed percutaneous fine-needle aspiration. The factors for reducing seroma formation included a complete reduction of the hernial sac without being transected. A recent prospective randomized controlled study showed that a left-in-situ transected hernial sac might increase exudation, resulting in seroma formation [18].
In this study, a total of three cases were converted to open surgeries with a conversion rate of $1.28\%$, comparable to the rates reported for previous studies (0.48–$1.8\%$) [13]. Two patients had adhesive preperitoneal space due to the previous herniorrhaphy; the other had bulky omentum incarceration. It is known that the abdominal cavity is more likely to develop adhesion when the patient has had previous abdominal operations. We created a preperitoneal cavity to manipulate the hernial sac in this limited space. Therefore, the difficulty of the surgery and the operative time would increase if the patient had had a previous lower abdominal operation [19]. The data between obese and non-obese patients in the unilateral group were also comparable. Although we supposed that the operation time would be relatively higher in the obese group, the results showed no significant differences between the two groups. In a previous study, high BMI correlated with prolonged operative times only during the surgeon’s learning period. Upon reaching the experienced level, the surgeon appeared to handle the challenge easily [20].
Variable access ports are placed at single-incision wounds, such as multi-instrument access or single-access ports. The access port may restrict the operating space of the TEP procedure [21]. Previous studies showed that homemade glove ports were safe and feasible [22,23]. We favored the use of a homemade powder-free glove port, which was easier to make and cost-effective. Due to the texture and elasticity, the homemade glove port provided greater angulation for manipulating the hernial sac. It could also be insufflated to make the tract clearer and larger. There was no significant gas leakage or glove rupture noted during the surgery under 10 mmHg pressure. In addition, the crowded laparoscopic instruments might lead to crashing, and therefore a 50 cm long Hopkins Forward-Oblique Telescope 30° was introduced that increased the working space for the surgeon and camera assistant, thereby reducing the crashing issue.
There are many different groin hernia classifications in the literature. However, most of them are complex and difficult to remember. Frequently used classifications such as Gilbert’s [24] and Nyhus’ [25] classifications were based on findings during an open (anterior) approach. The *European hernia* society (EHS) provided a general and systemic use of hernia classification [26]. However, the EHS classification system did not include the size or descent of the hernia sac. The hernia sac with scrotal extension is a major challenge during surgery, and it is important to determine this feature before surgery. Thus, we used the superficial inguinal ring as a boundary to evaluate the descending level of the hernia sac.
## Limitations
This study has several limitations, including first, that the sample size for this retrospective study was relatively low and limited to one center. Second, all the procedures were performed by a single experienced surgeon, limiting the generalizability of the results to other surgeons. However, the same surgeon performed all the operations, which reduces possible technique bias. Third, due to the short follow-up period, we may underestimate the recurrence rate and some delayed complications. Lastly, we did not analyze the factors that cause complications and affect operative time.
## 5. Conclusions
In conclusion, LESS-TEP herniorrhaphy is a safe and feasible operation with acceptable outcomes. Consequently, it may be a good option for those who care about wound cosmesis. Further large-scale prospective controlled studies and long-term analysis are still needed to further examine related chronic pain, recurrence rates, and effects on quality of life.
## References
1. Memon M.A., Cooper N.J., Memon B., Memon M.I., Abrams K.R.. **Meta-analysis of randomized clinical trials comparing open and laparoscopic inguinal hernia repair**. *Br. J. Surg.* (2003) **90** 1479-1492. DOI: 10.1002/bjs.4301
2. Ielpo B., Nunez-Alfonsel J., Duran H., Diaz E., Fabra I., Caruso R., Malavé L., Ferri V., Barzola E., Quijano Y.. **Cost-effectiveness of Randomized Study of Laparoscopic Versus Open Bilateral Inguinal Hernia Repair**. *Ann. Surg.* (2018) **268** 725-730. DOI: 10.1097/SLA.0000000000002894
3. HerniaSurge G.. **International guidelines for groin hernia management**. *Hernia* (2018) **22** 1-165
4. Ahmed I., Paraskeva P.. **A clinical review of single-incision laparoscopic surgery**. *Surgeon* (2011) **9** 341-351. DOI: 10.1016/j.surge.2011.06.003
5. Kommu S.S.R.A.. **Devices for laparoendoscopic single-site surgery in urology**. *Expert Rev. Med. Devices* (2009) **6** 95-103. DOI: 10.1586/17434440.6.1.95
6. Rao A., Kynaston J., MacDonald E.R., Ahmed I.. **Patient preferences for surgical techniques: Should we invest in new approaches?**. *Surg. Endosc.* (2010) **24** 3016-3025. DOI: 10.1007/s00464-010-1078-y
7. 7.
World Health Organization
Regional Office for the Western Pacific
The Asia-Pacific Perspective: Redefining Obesity and Its TreatmentHealth Communications AustraliaSydney, Australia2000. *The Asia-Pacific Perspective: Redefining Obesity and Its Treatment* (2000)
8. Bringman S., Blomqvist P.. **Intestinal obstruction after inguinal and femoral hernia repair: A study of 33,275 operations during 1992–2000 in Sweden**. *Hernia* (2005) **9** 178-183. DOI: 10.1007/s10029-004-0305-7
9. McCormack K., Wake B.L., Fraser C., Vale L., Perez J., Grant A.. **Transabdominal pre-peritoneal (TAPP) versus totally extraperitoneal (TEP) laparoscopic techniques for inguinal hernia repair: A systematic review**. *Hernia* (2005) **9** 109-114. DOI: 10.1007/s10029-004-0309-3
10. Bansal V.K., Misra M.C., Babu D., Victor J., Kumar S., Sagar R., Rajeshwari S., Krishna A., Rewari V.. **A prospective, randomized comparison of long-term outcomes: Chronic groin pain and quality of life following totally extraperitoneal (TEP) and transabdominal preperitoneal (TAPP) laparoscopic inguinal hernia repair**. *Surg. Endosc.* (2013) **27** 2373-2382. DOI: 10.1007/s00464-013-2797-7
11. Aiolfi A., Cavalli M., Del Ferraro S., Manfredini L., Lombardo F., Bonitta G., Bruni P.G., Panizzo V., Campanelli G., Bona D.. **Total extraperitoneal (TEP) versus laparoscopic transabdominal preperitoneal (TAPP) hernioplasty: Systematic review and trial sequential analysis of randomized controlled trials**. *Hernia* (2021) **25** 1147-1157. DOI: 10.1007/s10029-021-02407-7
12. Lo C.W., Yang S.S., Tsai Y.C., Hsieh C.H., Chang S.J.. **Comparison of laparoendoscopic single-site versus conventional multiple-port laparoscopic herniorrhaphy: A systemic review and meta-analysis**. *Hernia* (2016) **20** 21-32. DOI: 10.1007/s10029-015-1443-9
13. Lee Y.J., Kim J.H., Kim C.H., Lee G.R., Lee Y.S., Kim H.J.. **Single incision laparoscopic totally extraperitoneal hernioplasty: Lessons learned from 1231 procedures**. *Ann. Surg. Treat. Res.* (2021) **100** 47-53. DOI: 10.4174/astr.2021.100.1.47
14. Suzuki Y., Wakasugi M., Mikamori M., Tamaoka K., Nakahara Y., Tei M., Furukawa K., Ohtsuka M., Masuzawa T., Akamatsu H.. **Long-term outcomes of single-incision versus multiport laparoscopic totally extra-peritoneal inguinal hernia repair: A single-institution experience of 186 consecutive cases**. *Surg. Today* (2022) **52** 114-119. DOI: 10.1007/s00595-021-02323-4
15. Li C.-C., Tseng S.-I., Lee H.-Y., Chueh K.S., Tsai C.C., Chou Y.H., Huang C.N., Wu W.J.. **Retrospective comparison of open- versus single-incision laparoscopic extraperitoneal repair of inguinal hernia procedures: A single-institution experience**. *Urol. Sci.* (2020) **31** 77. DOI: 10.4103/UROS.UROS_88_19
16. Kocijan R., Sandberg S., Chan Y.W., Hollinsky C.. **Anatomical changes after inguinal hernia treatment: A reason for chronic pain and recurrent hernia?**. *Surg. Endosc.* (2010) **24** 395-399. DOI: 10.1007/s00464-009-0595-z
17. Bhangu A., Singh P., Pinkney T., Blazeby J.M.. **A detailed analysis of outcome reporting from randomised controlled trials and meta-analyses of inguinal hernia repair**. *Hernia* (2015) **19** 65-75. DOI: 10.1007/s10029-014-1299-4
18. Ruze R., Yan Z., Wu Q., Zhan H., Zhang G.. **Correlation between laparoscopic transection of an indirect inguinal hernial sac and postoperative seroma formation: A prospective randomized controlled study**. *Surg. Endosc.* (2019) **33** 1147-1154. DOI: 10.1007/s00464-018-6374-y
19. Dulucq J.L., Wintringer P., Mahajna A.. **Totally extraperitoneal (TEP) hernia repair after radical prostatectomy or previous lower abdominal surgery: Is it safe? A prospective study**. *Surg. Endosc.* (2006) **20** 473-476. DOI: 10.1007/s00464-006-3027-3
20. Kato J.M., Iuamoto L.R., Suguita F.Y., Essu F.F., Meyer A., Andraus W.. **Impact of Obesity and Surgical Skills in Laparoscopic Totally Extraperitoneal Hernioplasty**. *Arq. Bras. Cir. Dig.* (2017) **30** 169-172. DOI: 10.1590/0102-6720201700030002
21. Kim J.H., Park S.M., Kim J.J., Lee Y.S.. **Initial experience of single port laparoscopic totally extraperitoneal hernia repair: Nearly-scarless inguinal hernia repair**. *J. Korean Surg. Soc.* (2011) **81** 339-343. DOI: 10.4174/jkss.2011.81.5.339
22. Tai H.C., Lin C.D., Wu C.C., Tsai Y.C., Yang S.S.. **Homemade transumbilical port: An alternative access for laparoendoscopic single-site surgery (LESS)**. *Surg. Endosc.* (2010) **24** 705-708. DOI: 10.1007/s00464-009-0620-2
23. Tai H.C., Ho C.H., Tsai Y.C.. **Laparoendoscopic single-site surgery: Adult hernia mesh repair with homemade single port**. *Surg. Laparosc. Endosc. Percutaneous Tech.* (2011) **21** 42-45. DOI: 10.1097/SLE.0b013e31820ad65a
24. Gilbert A.I.. **An anatomic and functional classification for the diagnosis and treatment of inguinal hernia**. *Am. J. Surg.* (1989) **157** 331-333. DOI: 10.1016/0002-9610(89)90564-3
25. Nyhus L.M., Klein M.S., Rogers F.B.. **Inguinal hernia**. *Curr. Probl. Surg.* (1991) **28** 401-450. DOI: 10.1016/0011-3840(91)90028-N
26. Miserez M., Alexandre J.H., Campanelli G., Corcione F., Cuccurullo D., Pascual M.H., Hoeferlin A., Kingsnorth A.N., Mandala V., Palot J.P.. **The European hernia society groin hernia classification: Simple and easy to remember**. *Hernia* (2007) **11** 113-116. DOI: 10.1007/s10029-007-0198-3
|
---
title: Functional Characterization of Transgenic Mice Overexpressing Human 15-Lipoxygenase-1
(ALOX15) under the Control of the aP2 Promoter
authors:
- Dagmar Heydeck
- Christoph Ufer
- Kumar R. Kakularam
- Michael Rothe
- Thomas Liehr
- Philippe Poulain
- Hartmut Kuhn
journal: International Journal of Molecular Sciences
year: 2023
pmcid: PMC10003068
doi: 10.3390/ijms24054815
license: CC BY 4.0
---
# Functional Characterization of Transgenic Mice Overexpressing Human 15-Lipoxygenase-1 (ALOX15) under the Control of the aP2 Promoter
## Abstract
Arachidonic acid lipoxygenases (ALOX) have been implicated in the pathogenesis of inflammatory, hyperproliferative, neurodegenerative, and metabolic diseases, but the physiological function of ALOX15 still remains a matter of discussion. To contribute to this discussion, we created transgenic mice (aP2-ALOX15 mice) expressing human ALOX15 under the control of the aP2 (adipocyte fatty acid binding protein 2) promoter, which directs expression of the transgene to mesenchymal cells. Fluorescence in situ hybridization and whole-genome sequencing indicated transgene insertion into the E1-2 region of chromosome 2. The transgene was highly expressed in adipocytes, bone marrow cells, and peritoneal macrophages, and ex vivo activity assays proved the catalytic activity of the transgenic enzyme. LC-MS/MS-based plasma oxylipidome analyses of the aP2-ALOX15 mice suggested in vivo activity of the transgenic enzyme. The aP2-ALOX15 mice were viable, could reproduce normally, and did not show major phenotypic alterations when compared with wildtype control animals. However, they exhibited gender-specific differences with wildtype controls when their body-weight kinetics were evaluated during adolescence and early adulthood. The aP2-ALOX15 mice characterized here can now be used for gain-of-function studies evaluating the biological role of ALOX15 in adipose tissue and hematopoietic cells.
## 1. Introduction
Lipoxygenases are fatty acid dioxygenases that oxygenate arachidonic acid and related polyenoic fatty acids to the corresponding hydroperoxy derivatives [1,2,3,4,5]. They have been implicated in the differentiation of mesenchymal [6,7,8,9] and ectodermic cells [10,11,12] but may also play a role in the pathogenesis of inflammatory [13,14], hyperproliferative [15,16,17,18], neurodegenerative [19,20,21], and metabolic [22,23,24,25] diseases. The human genome involves six functional ALOX genes (ALOX15 [26], ALOX15B [27], ALOX12 [28,29], ALOX12B [30], ALOX5 [31,32], ALOXE3 [12,33]), and each of the ALOX-isoforms exhibit distinct biological functions. In the mouse genome, a single-copy ortholog exists for each human ALOX gene, but in addition, an *Aloxe12* gene exists that encodes for a functional epidermal Aloxe12 [34]. This enzyme shares a high degree of amino acid identity with Alox15, but in humans, the corresponding ortholog is a corrupted pseudogene [34]. Except for ALOXE12, knockout mice are available for each ALOX-isoform ([11,35,36,37,38,39,40]), but despite the availability of these tools, the biological relevance of the different ALOX-isoforms is still a matter of discussion.
Together with ALOX5, ALOX15 is the most comprehensively characterized ALOX-isoform [26,41,42,43]. It was discovered in 1975 as a protein in immature red blood cells of rabbits that was capable of oxygenating mitochondrial membranes [44]. Because of this property, the enzyme has been implicated in the maturational breakdown of mitochondria during late erythropoiesis [45,46]. To explore the biological roles of Alox15 in vivo, ALOX15−/− mice have been generated [35], and these animals (loss-of-function strategy) have been tested in a large number of mouse models of human diseases [47,48,49,50,51,52,53,54]. In addition, a number of transgenic mouse lines have been created (gain-of-function strategy), in which mouse or human ALOX15 was overexpressed under the control of different regulatory elements, that exhibit interesting phenotypes. Moderate overexpression of the endogenous mouse Alox15 induced spontaneous formation of aortic fatty streaks, which was related to upregulated expression of endothelial cell-adhesion molecules [55]. Endothelium-specific overexpression of human ALOX15 [56] accelerated aortic lipid deposition in LDL-receptor deficient mice [57], but it also inhibited tumor growth and metastasis in two different mouse models of human cancer [58]. More recently, these authors showed that this protective effect may be related to the promotion of apoptosis and necrosis in primary and metastatic tumor cells [59]. In rabbits, overexpression of human ALOX15 under the control of the lysozyme promoter induced macrophage-specific expression of the transgene [60] and protected the animals from aortic lipid deposition when fed a lipid-rich Western-type diet [61]. Transgenic mice overexpressing human ALOX15 under the control of the scavenger receptor A promoter [62] were also protected from aortic lipid deposition, but here, the protective effect was related to the augmented biosynthesis of anti-inflammatory and pro-resolving lipid mediators such as lipoxin A4, resolvin D4, and protectin D1 [63]. Genetically modified mice overexpressing a transgenic version of the endogenous Alox15 under the control of the alpha-cardiac myosin heavy chain promoter developed heart failure and diabetic cardiomyopathy [64,65]. When human ALOX15 was overexpressed in the intestinal epithelium under the control of the villin promoter [66], the resulting transgenic mice were protected from the development of azoxymethane-induced colonic tumors, and expression of the ALOX15 transgene was always impaired in tumor cells when compared with non-tumor controls [67]. Additional mechanistic studies have suggested that expression of the transgene inhibited the expression of tumor necrosis factor alpha and its target, the inducible nitric-oxide synthase. Moreover, activation of nuclear factor kappa B was prevented [67]. More recently, a transgenic mouse line was created in which the expression of the human ALOX15 was controlled by the Cre-lox promoter [68]. The employed strategy ensured ubiquitous overexpression of the transgene, but when these mice and corresponding wildtype controls were used in a diabetic peripheral neuropathy model, the authors did not observe significant differences when compared with wildtype controls [68].
In most of these transgenic animals, tissue-specific expression of the transgene was explored. However, incorporation of the transgene into the host genome was controlled in neither of them, and thus, it is unclear how many copies of the transgene had been incorporated into the host genome and at which positions. Moreover, in many of these studies, activity assays were performed, and thus, it is unclear whether the transgenic enzyme was catalytically active. This is a serious limitation for some of these studies, since expression of a functional ALOX15 is strongly regulated on the translational level [69,70,71], and thus, detection of the transgenic mRNA is not sufficient to conclude the catalytic activity of the enzyme. In fact, in human umbilical vein endothelial cells, high levels of ALOX15 mRNA were detected, but the catalytically active enzyme was missing [72]. To contribute to the discussion on the biological role of Alox15, we here characterize transgenic mice expressing the human ALOX15 under the control of the aP2 (adipocyte fatty acid binding protein-2) promoter (aP2-Alox15 mice). In these mice, transgene expression is directed to mesenchymal cells, particularly to adipocytes, macrophages, and other cells of the myeloic linage. In the present paper we report the breeding of homozygous aP2-ALOX15 mice and found that the transgene was incorporated as single copy gene into the E1-2 region of chromosome 2. Ex vivo activity assays indicated expression of the functional transgene in adipocytes, spleen, bone marrow cells, and peritoneal macrophages. The in vivo activity of transgenic human ALOX15 was indicated by analysis of the plasma oxylipidomes. Because of the high expression levels of the transgene in adipose tissue and in the hematopoietic system, these mice can be used in the future to study the role of ALOX15 in adipocyte differentiation and hematopoiesis.
## 2.1. Creation of Transgenic aP2-ALOX15 Mice
Different ALOX-isoforms (ALOX5 [73], ALOX12 [74], ALOX12B [75], ALOXE3 [76] and ALOX15 [77]) have been implicated in adipocyte differentiation, in the energy metabolism of fat cells, and in adipose tissue remodeling. Moreover, supplementation studies with ALOX products suggested that several oxylipins activate adipogenesis of 3T3 cells in vitro, and these data support a possible role of ALOX15 in adipogenesis [9]. However, 15-HETE formation in mice is limited, since none of the seven functional Alox isoforms produce 15-HETE as a major arachidonic acid oxygenation product. Human ALOX15 effectively oxygenates arachidonic acid to 15-HETE [78,79], and thus, 15-HETE formation can be used as metabolic footprint for expression of the transgene.
To test the impact of endogenous 15-HETE formation in mouse adipose tissue in vivo, a transgenic mouse line was created that overexpresses human ALOX15 under the control of the aP2 (adipocyte fatty acid binding protein 2) promoter. For this purpose, the cDNA of human ALOX15 was ligated behind the aP2 promoter (Figure 1), and this construct was microinjected into fertilized eggs. Cells were reimplanted into pseudo-pregnant mice, and individuals carrying the transgene in the germ line were crossed with wildtype C57BL/6 mice. Heterozygous allele carriers were intercrossed, and homozygous aP2-ALOX15 mice were selected. These animals were used to establish a colony of homozygous aP2-ALOX15 mice, and individuals of this colony were used for our characterization studies.
## 2.2. Genomic Insertion of the Transgene
Since our transgenic strategy involved coincidental incorporation of the transgene into the host genome, we next explored how many copies of the transgene were incorporated into the genome, and we also determined the site(s) of transgene insertion.
For this purpose, we first performed fluorescence in situ hybridization (FISH) using the human ALOX15 cDNA as a probe. Figure 2A shows a representative FISH staining of a heterozygous founder. It can be seen that the transgene was inserted as a single-copy gene into the subband E1-2 on chromosome 2. No specific fluorescence signal was detected on any other chromosome. To describe the site of transgene incorporation more precisely and to exclude transgene insertion having disrupted a functional gene in this region, we carried out whole-genome sequencing. The obtained sequence data confirmed single incorporation of the transgene and also suggested that no functional gene was disrupted during transgene insertion.
## 2.3. Tissue-Specific Expression of Transgenic Human ALOX15
To explore the tissue-specific expression of both endogenous mouse Alox15 and the human ALOX15 transgene, we carried out qRT-PCR using isofom-specific primers. Human and mouse ALOX15 cDNA share a high degree of nucleotide sequence identity ($85\%$), and thus, designing ortholog specific pPCR primers was somewhat difficult. However, we selected cDNA regions with relatively low degrees of nucleotide sequence conservation, and by using these ortholog specific primer pairs, we found (Figure 3A) that the endogenous arachidonic acid 12-lipoxygenating mouse Alox15 was expressed at relatively low levels in most cells and tissues.
The highest expression levels were observed in the lung, but even in this organ, only about 20 copies of Alox15 mRNA were present per 1000 copies of Gapdh mRNA. Lower expression levels were observed in bone marrow cells and in peri-epididymal adipose tissue. In spleen, heart, liver, kidney, and brain, we did not see specific amplification signals. When similar analyses were carried out with the RNA extracts prepared from corresponding tissues of the transgenic aP2-ALOX15 mice, lower steady-state ALOX15 mRNA concentrations were detected in the lungs, but elevated levels were detected in the perirenal adipose tissue (Figure 3A). However, despite these differences, the expression levels of the endogenous mouse Alox15 mRNA were also low (1–5 Alox mRNA copies per 1000 copies of Gapdh) in all tested tissues of the transgenic aP2-ALOX15 mice.
When we repeated these analyses with the human ALOX15-specific amplification primers (Figure 3A), we did not see specific amplification signals in the different tissues of wildtype mice. These results were expected, since wildtype mice do not express human ALOX15, and our primers do not pick up mouse Alox15 mRNA. In contrast, we detected the abundant expression of human ALOX15 in the three types of adipose tissue (peri-epididymal, subcutaneous, perirenal). We also detected high-level expression of the transgene in bone marrow cells, spleen, lungs, and testis. The most interesting outcome of our qRT-PCR data was that the steady state concentrations of the transgenic mRNA, which varied between 800 and 3,500 ALOX15 mRNA copies per 103 copies of Gapdh, were much higher than the mRNA copy numbers of the endogenous Alox15 in wildtype tissues. Taken together, these data indicate overexpression of the transgenic enzyme in adipose tissue, but also in hematopoietic cells (bone marrow), spleen, lung, and testis.
## 2.4. Transgenic Human ALOX15 Is Catalytically Active in Peritoneal Lavage and Bone Marrow Cells
Expression of ALOX15 orthologs is strongly regulated at transcriptional [80,81] and translational levels [69,70]. In fact, in young rabbit reticulocytes, large amounts of ALOX15 mRNA are present, but no functional protein can be detected [45]. Thus, in principle, there is the possibility that ALOX15 mRNA is present but no functional enzyme is expressed [72]. To explore whether a functional transgenic enzyme is expressed in different tissues of the transgenic aP2-ALOX15 mice, we carried out ex vivo activity assays using intact cell suspensions or tissue homogenate supernatants as enzyme source. In peritoneal lavage cells, mouse ALOX15 is highly expressed, and thus, these cells are well-suited to ex vivo activity assays.
When we incubated these cells from wildtype mice with arachidonic acid (Figure 4A), large amounts of 12-HETE were detected. In addition, smaller amounts of 15-HETE were also identified as a minor side product. Interestingly, under these experimental conditions, we did not find any 5-HETE formation, although significant amounts of ALOX5 mRNA were detected in these cells by qRT-PCR.
**Figure 4:** *Ex vivo activity assays of peritoneal lavage cells prepared from mice of different genotype. 107 peritoneal macrophages prepared from mice of different genotype were incubated with arachidonic acid as described in Section 4, and the oxygenation products were analyzed by RP-HPLC. Representative chromatograms are shown; the statistical evaluation of the data is given in Table 1.* TABLE_PLACEHOLDER:Table 1 When these experiments were repeated with peritoneal macrophages prepared from ALOX15−/− mice, 12- and 15-HETE were no longer detectable. Instead, 5-HETE was identified as a major arachidonic acid oxygenation product (Figure 4B). Since ALOX5 is expressed in peritoneal macrophages, the formation of 5-HETE is plausible when the dominant ALOX15 pathway is genetically silenced. When we incubated peritoneal lavage cells from our transgenic aP2-ALOX15 mice with arachidonic acid, 12-HETE remained the major oxygenation product (Figure 4C). However, the relative share of 15-HETE was significantly increased when compared with wildtype cells (Table 1). The most plausible explanation for these data is that both endogenous ALOX15 (forming 12-HETE) and the transgenic ALOX15 (forming 15-HETE) were catalytically active. To provide more compelling evidence for this conclusion, we crossed ALOX15−/− mice with our transgenic aP2-ALOX15 animals, prepared peritoneal lavage cells from the animals, and carried out ex vivo activity assays. In these cells, endogenous Alox15 (forming 12-HETE) was absent, and thus, 15-HETE originating from the transgenic ALOX15 pathway was expected to be dominant. In fact, when we carried out such ex vivo activity assays, 15-HETE was the major arachidonic acid oxygenation product, and small amounts of 12-HETE were also detected (Figure 4D). Since human ALOX15 exhibits dual-reaction specificity [79], the formation of small amounts of 12-HETE by the transgenic enzyme is plausible. Similar ex vivo activity assays were carried out with three individuals of each genotype; the statistical evaluation of the experimental raw data is given in Table 1. In summary, these data confirm our hypothesis that the transgenic human ALOX15 is expressed as a catalytically active enzyme in peritoneal macrophages in addition to the endogenous mouse Alox15. Assuming a similar specific activity of mouse and human ALOX15, more endogenous Alox15 should be present in these cells when compared with the transgenic human ALOX15.
Bone marrow cells are another rich source of endogenous Alox15. However, in these cell types, the arachidonic acid metabolism is somewhat more complex, since the arachidonic acid 12-lipoxygenating ALOX15, the arachidonic acid 12-lipoxygenating ALOX12, and the arachidonic acid 5-lipoxygenating ALOX5 are simultaneously expressed. When wildtype bone marrow cells are incubated with arachidonic acid, 12-HETE was identified as a major arachidonic acid oxygenation product (Table 2). 15-HETE and 5-HETE only contributed minor shares. When bone marrow cells of ALOX15−/− mice were used for the ex vivo activity assays, we did not see any 15-HETE formation (Table 2). These data suggest that the minor share of 15-HETE formation by wildtype bone marrow cells (4.9 ± $1.6\%$) may be related to the catalytic activity of the endogenous mouse Alox15. Although mouse Alox15 is dominantly arachidonic acid 12-lipoxygenating, 15-HETE is a minor side product [82].
The strong but incomplete reduction of 12-HETE formation by ALOX15−/− bone marrow cells suggests that the dominant 12-HETE formation by wildtype bone marrow cells (93.8 ± $3.6\%$) may be related to the mixed catalytic activity of the endogenous ALOX15 and ALOX12 isoforms. Interestingly, ALOX15−/− bone marrow cells produce large amounts of 5-HETE, and these data (Table 2) are consistent with the results of the ex vivo activity assays of peritoneal lavage cells (Figure 4B, Table 1). Expression of transgenic human ALOX15 completely altered the pattern of arachidonic acid oxygenation. In this case, 15-HETE was the major (55.5 ± $1.6\%$) arachidonic acid oxygenation product; these activity data suggest catalytic activity from the transgenic human ALOX15. When we crossed aP2-ALOX15 transgenic mice with ALOX15−/− animals, the relative share of 12-HETE was further reduced (from 44.4 ± $1.6\%$ in aP2-ALOX15 mice to 10.9 ± $0.5\%$ in aP2-ALOX15 + ALOX15−/−); these data suggest that more than $70\%$ of the 12-HETE formed by aP2-ALOX15 bone marrow cells originated from the endogenous ALOX15 pathway. In these cells, ALOX12 may only contribute $30\%$ to 12-HETE formation. As expected, formation of 15-HETE was strongly elevated by bone marrow cells of aP2-ALOX15 + ALOX15−/− mice. Taken together, our ex vivo activity experiments confirm that in peritoneal lavage cells, as well as in bone marrow cells, transgenic human ALOX15 is expressed and the transgenic enzyme is catalytically active.
## 2.5. Transgenic Human ALOX15 Is also Catalytically Active in Solid Tissue
As indicated in Figure 3, the transgenic mRNA is expressed in different solid tissues such as adipose tissue, spleen, lungs, and testis. To explore whether the transgenic enzyme is also expressed in these tissues and whether the protein is catalytically active, we carried out similar ex vivo activity assays. For this purpose, we prepared tissue homogenates, and used the 20,000 g supernatants as enzyme source. After a 15 min incubation period of the homogenate supernatants with arachidonic acid, we quantified by RP-HPLC the formation of 12-HETE and 15-HETE as the major readout parameter.
When homogenate supernatants of wildtype control mice were used for the ex vivo activity assays, only small amounts of 12-HETE and 15-HETE were formed. Only in lungs and spleen, we observed significant formation of 12-HETE, which exceeded the formation of 15-HETE. These data suggest that the endogenous arachidonic acid 12-lipoxygenating mouse Alox15 is expressed in these tissues, but that the enzyme may not be present in adipose tissue, testis, and heart. In contrast, activity assays with homogenate supernatants prepared from these tissues of aP2-ALOX15 mice indicated dominant 15-HETE formation by the homogenate supernatants of adipose tissue, spleen, and lungs, whereas only minor 15-HETE formation was observed in testis and heart. Taken together, our ex vivo activity data indicated that the transgenic human ALOX15 is expressed at high levels in adipose tissue but also in spleen and lungs.
## 2.6. In Vivo Activity of Transgenic Human ALOX15
Our ex vivo activity assays indicated the expression of the catalytically active transgenic human ALOX15 in different tissues, but the data did not prove the in vivo activity of the enzyme. If the transgenic enzyme is catalytically active in vivo and if this in vivo activity is mirrored on blood plasma levels of 15-HETE and other omega-6 oxygenation products of polyenoic fatty acids, the plasma concentrations of 15-HETE, 15-HEPE, 17-HDHA, and 15-HETrE should be higher in aP2-ALOX15 mice when compared with wildtype controls. To test this hypothesis, we analyzed the plasma oxylipidomes of aP2-ALOX15 mice and corresponding wildtype controls and quantified the plasma concentrations of oxygenated polyenoic fatty acids, including the major ALOX15 products [83].
As negative controls, we also quantified the plasma concentrations of other oxylipins that are not formed from arachidonic acid, eicosapentaenoic acid, docosahexaenoic acid, nor 8,11,14-eicosatrienoic acid by human ALOX15 (8-HETE, 8-HEPE, 10-HDHA, 8-HETrE). First, we quantified the sum of all oxygenated polyenoic fatty acids present in the blood plasma of the two genotypes (Figure 5A). Here, we did not find a significant difference between the two genotypes. These data indicate that the degree of oxidative challenge is similar in both genotypes. In other words, overexpression of human ALOX15 did not lead to an increased oxidative stress in the aP2-ALOX15 mice. When we analyzed the major arachidonic acid oxygenation products, we found that the plasma concentrations of 15-HETE in transgenic aP2-ALOX15 mice were almost five-fold higher than those in wildtype control animals (C57BL/6). In contrast, there were no significant differences between the two genotypes when the plasma levels of 8-HETE (not an ALOX15 product) were compared. These data can be interpreted as an indication of the in vivo activity of the transgenic human ALOX15.
Next, we analyzed the oxygenation products of three other polyenoic fatty acids. Human ALOX15 converts 5,8,11,14-eicosapentaenoic acid predominantly to 15-HEPE [83]. We found that the 15-HEPE plasma concentrations were more than five-fold higher in aP2-ALOX15 mice when compared with wildtype controls. Here again, we did not find any difference between the two genotypes for 8-HEPE, which is not formed by human ALOX15 (Figure 6C). Similar results were obtained for the major oxygenation products formed from 4,7,10,13,16,19-docosahexaenoic acid (Figure 6D) and 8,11,14-eicosatrienoic acid (Figure 6E). Here, the differences between aP2-ALOX15 mice and C57Bl/6 control animals were even more pronounced.
In summary, our plasma oxilipidomes suggest the in vivo activity of the transgenic human ALOX15. Interestingly, the in vivo catalytic activity of the transgenic ALOX15 did not induce an elevated oxidative challenge in the transgenic animals. Obviously, the reductive capacity of the ALOX15-expressing cells is high enough to ensure the instantaneous reduction of the hydroperoxy fatty acids formed by the transgenic enzyme.
## 2.7. Reproduction Statistics
ALOX15 has been implicated in spermatogenesis [84], and ALOX15−/− mice are sub-fertile [85,86]. Although we did not find dramatic overexpression of the catalytically active transgene in testis (Figure 7), we compared the reproduction statistics of the aP2-ALOX15 mice with those of wildtype controls. Here, we found that the frequency of pregnancy (litters per female and month) and the reproduction efficiency (litters per female and months) were significantly elevated in aP2-ALOX15 transgenic mice (Figure 7). For the other readout parameters, no significant differences were observed when the two genotypes were compared (Figure 7). In summary, one can conclude that the aP2-ALOX15 mice, which express human ALOX15 in addition to the endogenous mouse ALOX15, were slightly more fertile than wildtype controls, although the differences were rather subtle.
## 2.8. Body-Weight Kinetics
To explore whether systemic overexpression of human ALOX15 impacts the development of mice during adolescence and adulthood, we next profiled the body-weight kinetics of aP2-ALOX15 mice and wildtype control animals starting at the age of 8 weeks. For female individuals, we found that at 8 weeks, aP2-ALOX15 mice were significantly leaner than wildtype controls (Figure 8A). In fact, between 8–30 weeks, the curve of the body-weight kinetics of aP2-ALOX15 mice was consistently below the curve of the wildtype controls, and this difference was statistically highly significant (Wilcoxon test, $p \leq 0.0085$). In contrast, between 31–38 weeks, no significant difference was observed between the two genotypes. These data suggest that female aP2-ALOX15 mice gained less body weight than wildtype controls during the early developmental period but that the transgenic individuals caught up with the wildtype controls at later developmental stages (Figure 8A).
When the body weights of male individuals were profiled, an inverse situation was observed. Here, we did not find significant differences between the two genotypes in early developmental stages (Figure 8B). In fact, between 8 and 20 weeks, the curves of the body-weight kinetics were superimposable, and using the Wilcoxon test, we did not observe a significant difference between the two genotypes. However, at later developmental stages (20–38 weeks), aP2-ALOX15 mice gained significantly more body weight that wildtype controls (Figure 8B). Although the extent of this difference was rather subtle, it was statistically highly significant ($$p \leq 0.0005$$). When the Wilcoxon test was applied for the entire experimental timescale, highly significant differences between aP2-ALOX15 mice and wildtype controls were observed for either sex ($p \leq 0.0001$), but the net effects were opposite in males and females. In females, aP2-ALOX15 mice gained less body weight, whereas male aP2-ALOX15 individuals gained more. The mechanistic basis for the observed gender specific effects have not been explored. We speculate that the transgenic expression of the ALOX15 in adipose tissue might impact the production of sexual hormones and/or of leptin in the adipose tissue. To test this hypothesis, additional experiments must be carried out that exceed the frame of the present study.
## 3.1. Degree of Novelty and Limitations
Mammalian ALOX15 orthologs have been implicated in the differentiation of adipocytes [87], in the oxidative metabolism of fat cells [88], and in the remodeling of the adipose tissue [77]. ALOX15 mRNA expression was dramatically upregulated in white epididymal adipocytes when wildtype mice were fed a high-fat diet [87]. In 3T3 preadipocytes, ALOX15 is virtually absent, but its expression is strongly upregulated when cells were differentiated into adipocytes [87]. When treated with the ALOX15 products, these cells adopt a proinflammatory phenotype and lose their insulin resistance [9,87]. ALOX15−/− mice are resistant to the induction of type-1 diabetes [89] and also to the inflammatory effects of obesity induced by a Western-type diet [90]. ALOX15-deficient nonobese diabetic mice developed diabetes at a markedly reduced rate, demonstrated improved glucose tolerance, reduced severity of insulitis, and improved beta-cell mass when compared with age-matched nondiabetic ALOX15-sufficient controls. These results suggest an important role for ALOX15 in the pathogenesis of autoimmune diabetes [91]. In most of these studies, loss-of-function strategies were employed to evaluate the role of ALOX15 in the pathogenesis of obesity and diabetes, but the application of gain-of-function strategies was rare. To address this problem, we here created transgenic mice that overexpress human ALOX15 under the control of the aP2 promoter. Our qRT-PCR studies (Figure 3) and our ex vivo activity data (Figure 5) indicate the expression of the transgene in different types of adipose tissue but also in other mesenchymal cells such as bone marrow, spleen, and peritoneal macrophages. Although our data indicate that transgene expression may not be specific for adipocytes, we did not detect transgene expression in other major organs of our aP2-ALOX15 mice, such as in liver, skin, bones, kidney, or skeleton muscles.
If one compares the aP2-ALOX15 mice created in this study with previously described ALOX transgenic mouse lines, the advantages and disadvantages of the aP2-ALOX15 mice can be summarized as follows: (i) Expression of the transgene is limited to a small number of special cell types, and thus, these mice are particularly suited for further investigations into the role of the ALOX15 pathway in adipocytes (Figure 5) and in hematopoietic cells (Table 2). In other studies, expression of the ALOX transgenes was controlled by different promoters directing transgene expression to other cell types [56,59,62,63,66]. Thus, for studies on the potential role of the ALOX15 pathway in adipocytes and hematopoietic cells, the previously created ALOX15 transgenic mouse lines are less suitable. On the other hand, the aP2-ALOX15 mice may not be useful to study the metabolic role of this enzyme in endothelial cells, epithelial cells, and/or macrophages. For such experiments, transgenic ALOX15 mice should be used, in which transgene expression is controlled by the preproendothelin [56], the lysozyme [63], scavenger receptor A [62], or the villin [66] promoter.
(ii) In all previously created ALOX15 transgenic mouse lines, incorporation of the transgene into the genome was not controlled. Thus, multiple copies of the transgene might have been inserted, and incorporation of the transgene might have disrupted other genes. For the aP2-ALOX15 mice, we characterized the site of transgene insertion and found that the ALOX15 transgene was incorporated as a single-copy gene into the E1-2 region of chromosome 2 (Figure 2). Moreover, complete genome sequencing suggested that transgene incorporation did not structurally disturb other genes.
(iii) In most previously created ALOX15 transgenic mouse lines, the catalytic activity of the transgenic enzyme was not tested, and thus, it was unclear whether the transgenic enzyme was catalytically active. For the aP2-ALOX15 mice, we carried out ex vivo ALOX15 activity assays with different cells and tissues and showed catalytic activity of the transgene (Figure 4 and Figure 5, Table 2). Moreover, we found that the product pattern formed from exogenously added arachidonic acid was very similar to that formed by recombinant human ALOX15 [79].
(iv) Although our ex vivo activity assays indicated the principle catalytic activity of the transgenic enzyme, such assays do not prove the in vivo activity. To show the in vivo activity of the transgenic ALOX15, we analyzed the plasma oxylipidomes (Figure 6) and found that in the blood plasma of the aP2-ALOX15 mice, the classical ALOX15 products formed from different polyenoic fatty acid were elevated. In contrast, there was no difference between aP2-ALOX15 mice and wildtype controls when unrelated oxylipins (e.g., 8-HETE, 8-HEPE, 8-HeTrE) were compared. These data suggest the in vivo activity of the transgenic enzyme. Corresponding experiments have not been carried out with any of the other ALOX transgenic mouse lines.
(v) ALOX15 has been implicated in spermatogenesis, and ALOX15−/− mice are sub-fertile [84]. Thus, there is the possibility that transgenic overexpression of ALOX15 might impact the reproduction behavior of aP2-ALOX15 mice. To test the fertility of these animals, we evaluated the reproduction statistics and found no dramatic difference with wildtype mice (Figure 7). Similar experiments have not been carried out for any of the other transgenic ALOX15 mice.
(vi) aP2-ALOX15 mice showed gender-specific differences to wildtype controls when their body-weight kinetics were evaluated (Figure 8). This observation is not trivial and must be considered in the interpretation of future experimental data obtained with these mice in animal disease models. Here again, body-weight kinetics have not been reported for any of the other transgenic mouse lines.
In summary, one can conclude that the aP2-ALOX15 mice created in this study constitute the most comprehensively characterized transgenic ALOX15 mouse line currently available.
The aP2-ALOX15 mice may also be used for rescue experiments reversing the effects induced by systemic or tissue-specific ALOX15 knockout. We recently reported that systemic inactivation of the ALOX15 gene induced subtle defects in the erythropoietic systems in ALOX15−/− mice, as indicated by significantly reduced Hb, HK, and Ery counts [92]. When the ALOX15−/− mice were crossed with our aP2-ALOX15 mice, we found that this defective phenotype was rescued, since the above-mentioned erythropoietic parameters were normalized [92]. From these data, we concluded that overexpression of human ALOX15 in hematopoietic cells may compensate for ALOX15 deficiency.
Originally, we created these mice in order to study the role of ALOX15 in adipocytes. In previous cellular studies, different ALOX-isoforms and their metabolites have been implicated in adipogenesis and in the pathogenesis of the metabolic syndrome [9,93,94,95], which is associated with hyperplasia of the adipose tissue. The aP2-ALOX15 mice appear to be a valuable research tool to test these hypotheses in vivo. The present paper describes the production and basic functional characterization of these mice, which can later be used in different animal disease models associated with adipocyte hyperplasia. Since our lab is not specialized in such diseases, and since we do not have the suitable model systems, the aP2-ALOX15 mice may be employed by interested scientists in the frame of scientific collaboration.
## 3.2. Human vs. Mouse ALOX15 as Transgene
When we started this project, we had a long discussion regarding whether we should use the endogenous mouse Alox15 or the corresponding human ortholog (ALOX15) as the transgene. This discussion was prompted by the catalytic differences of the two ALOX15 orthologs. Human ALOX15 converts arachidonic acid mainly to 15S-HETE ($90\%$), and only about $10\%$ is formed as 12S-HETE [79,96]. Under identical experimental conditions, mouse Alox15 exhibits an inverse product pattern. Here, 12S-HETE is dominant, whereas 15S-HETE is a minor side product [82]. The molecular basis for this difference in the reaction specificity has been explored [97,98,99,100], and the Triad Concept [26,101] has been developed as a mechanistic tool for predicting the reaction specificity of mammalian ALOX15 orthologs on the basis of their primary structures. Moreover, the evolutionary hypothesis of mammalian ALOX15 specificity [102,103] suggests that ALOX15 orthologs of those mammalian species ranked in evolution above gibbons, including humans, chimpanzees, and orangutans, express arachidonic acid 15-lipoxygenating ALOX15 orthologs. In mammals ranking in evolution below gibbons, arachidonic acid 12-lipoxygenating ALOX15 orthologs are present. Thus, the vast majority (<$95\%$) of mammals express an arachidonic acid 12-lipoxygenating ALOX15 ortholog, despite their annotation as ALOX15. However, several mammals (about $5\%$) including rabbits [104], mountain hares [103], kangaroo rats [105], anteaters, and bamboo rats [103] violate this concept. Thus, because of the dominance of arachidonic acid 12-lipoxygenating ALOX15 orthologs in mammals, the endogenous mouse ALOX15 should be employed as the transgene. However, the advantage of using the human ALOX15 as transgene is that the catalytic activity of this transgene can easily be followed. In mice, there is no arachidonic acid 15-lipoxygenating ALOX-isoform and thus, 15-HETE formation can be considered as a metabolic footprint of the transgene. In contrast, several mouse ALOX-isoforms, including the endogenous ALOX15, convert arachidonic acid to 12-HETE, and thus, profiling 12-HETE formation does not allow metabolic profiling of the transgene. Thus, we decided to use human ALOX15 as the transgene.
Despite their different reaction specificity with arachidonic acid and 4,7,10,13,16,19-docosahexaenoic acid [83], mouse Alox15 and its human ortholog are very similar. They share a high degree (>$85\%$) of amino acid sequence identity, and both enzymes are capable of oxygenating linoleic acid. For both enzymes, 13-H(p)ODE was identified as the dominant linoleic acid oxygenation product. Similarly, from 5,8,11,14,17-eicosapentaenoic acid, 15-H(p)EPE is formed as the major oxygenation product by the two ALOX15 orthologs [83]. Moreover, both ALOX15 orthologs are capable of oxygenating biomembranes, although human ALOX15 is somewhat more efficient. If ALOX15 orthologs fulfill their biological functions via the formation of specific oxygenation products from arachidonic acid or docosahexaenoic acid, there must be a functional difference between mouse and human ALOX15. In contrast, when product formation from linoleic acid and/or 5,8,11,14,17-eicosapentaenoic acid is more important, both enzymes should induce similar biological effects. If the biological functions of the ALOX15 orthologs are related to their ability to oxygenate complex substrates, there may not be major differences between mouse and human ALOX15.
For clarity, we would like to discuss the following example. Rabbit ALOX15 has been implicated in late erythropoiesis [46]. When synthesized, the enzyme oxygenates mitochondrial membrane lipids, which initiates the maturational proteolytic breakdown of the mitochondria in mature reticulocytes. If this concept is transferred to other mammals, there should not be a major impact whether the ALOX15 ortholog is an arachidonic acid 12-lipoxygenating or an arachidonic acid 15-lipoxygenating enzyme. As long as the enzyme is capable of oxygenating the membrane lipids it will fulfill its biological function. Thus, in this case, the ability of the enzyme to oxygenate complex substrates is more important for the biological function than the reaction specificity with arachidonic acid. Moreover, linoleic acid is the major polyenoic fatty acid of mitochondrial membranes. Thus, the ability of mouse and human ALOX15 to oxygenate this substrate may be more important for the biological role of the enzyme than their reaction specificity with free arachidonic acid.
## 3.3. ALOX15 Expression in Hematopoietic Cells Suppresses the Pro-Inflammatory ALOX5 Pathway
In mouse bone marrow cells, several ALOX-isoforms (ALOX15, ALOX12, ALOX5) are constitutively expressed, and thus, these cells may be good models for the exploration of functional ALOX interaction. When these cells were incubated ex vivo with arachidonic acid (Table 2), we found that 12-HETE was dominant. In addition, small amounts of 15-HETE were also detected, whereas 5-HETE formation was minimal. The most plausible explanation for this product pattern was that 12-HETE formation may be due to the catalytic activity of both endogenous ALOX12 and ALOX15. Mouse ALOX12 exclusively produces 12-HETE, whereas the endogenous ALOX15 forms 15-HETE as a minor side product. When we knocked out ALOX15 expression, the relative share of 12-HETE formation was strongly reduced, and 15-HETE formation completely disappeared. Thus, in mouse bone marrow cells, endogenous ALOX15 is responsible for the formation of 15-HETE and parts of 12-HETE. Most interestingly, however, was the observation that functional inactivation of the ALOX15 pathway strongly upregulated the catalytic activity of endogenous ALOX5 (Table 2). In fact, 5-HETE was the dominant arachidonic acid oxygenation product when bone marrow cells of ALOX15−/− mice were incubated ex vivo with arachidonic acid. In peritoneal macrophages (Table 1), this effect was even more pronounced. Here, ALOX15-derived 12-HETE was dominant when ALOX15-sufficient cells were employed. In contrast, exclusive 5-HETE formation was observed with ALOX15-deficient macrophages.
These data suggest that at least in bone marrow cells and in peritoneal macrophages, a catalytically active ALOX15 suppresses the ALOX5 pathway. Expression of the human ALOX15 transgene induced a similar repressive effect as the endogenous ALOX15 (Table 1 and Table 2). The most straightforward explanation for this observation is that endogenous and transgenic ALOX15 orthologs compete with endogenous ALOX5 for the exogenous substrate. However, there are two lines of experimental evidence arguing against this explanation: (i) the affinity of human ALOX15 [79] and human ALOX5 [106] for arachidonic acid is comparable, and thus, the suppressive effect cannot be related to competition of the two enzymes for the joint substrate; (ii) when we analyzed the free arachidonic acid, which was left over after the ex vivo incubation period, we found that about half of the substrate was not converted. These data suggest suppression of the ALOX5 pathway, even though plenty of exogenous arachidonic acid was present as ALOX5 substrate. Thus, simple substrate competition may not be the major reason for the suppression of the ALOX5 pathway by ALOX15 expression. The molecular basis for the suppressive effect of ALOX15 expression on ALOX5 has not been explored in detail, but it may be possible that primary and/or secondary products of the ALOX15 pathway directly inhibit ALOX5. This effect may be of biological relevance, since it may explain, at least in part, the anti-inflammatory role of ALOX15 in different mouse inflammation models [63,107,108], in addition to the ALOX15-dependent formation of special pro-resolving mediators [109,110]
## 3.4. Expression of ALOX15b in aP2-ALOX15 Mice
As indicated in Table 2, transgenic expression of human ALOX15 in bone marrow cells suppressed the catalytic activity of ALOX5 in our ex vivo activity assays. Unfortunately, we did not quantify the expression levels of endogenous ALOX5 or other ALOX-isoforms such as ALOX15b. In humans, ALOX15B converts AA to the same oxygenation product (15-HETE) as ALOX15, but its mouse ortholog exhibits a different product specificity [111] with free AA (8-HETE formation). Since we did not see major amounts of 8-HETE formation in our ex vivo activity assays using peritoneal lavage and bone marrow cells (Table 2), it may be concluded that ALOX15b expression in these cells may not be very pronounced. Moreover, analyses of the plasma oxylipidomes did not reveal significant differences between aP2-ALOX15 mice and wildtype controls; these data suggest that the endogenous ALOX15b pathway may have minimally altered by our genetic manipulation.
For complex substrates, the situation is somewhat different. When nanodiscs involving AA-containing phospholipids were used as substrate for recombinant mouse and human ALOX15B orthologs, 15S-HETE-containing phospholipids were detected as major reaction products [112]. In other words, with phospholipids as substrate, mouse and human ALOX15B orthologs exhibit similar reaction specificities. Because of these observations, we cannot completely exclude, on the basis of our experimental data, the modification of ALOX15b expression in the aP2-ALOX15 mice.
The biological role of ALOX15B has not been well-defined, neither in mice nor in humans. In a recent review [113], the different hypotheses on the putative physiological and pathophysiological functions of mammalian ALOX15B orthologs were summarized, but because of the lack of systemic ALOX15b−/− mice, most of these hypotheses have not been confirmed under in vivo conditions.
## 4.1. Chemicals
The chemicals used for the different experiments were obtained from the following sources: phosphate-buffered saline without calcium and magnesium (PBS) from PAN Biotech (Aidenbach, Germany); EDTA from Merck KG (Darmstadt, Germany); arachidonic acid (AA) and authentic HPLC standards of HETE-isomers (15R/S-HETE, 12S/R-HETE, 8R/S-HETE, 5S-HETE) from Cayman Chem (distributed by Biomol GmbH, Hamburg, Germany); acetic acid from Carl Roth GmbH (Karlsruhe, Germany); sodium borohydride from Life Technologies, Inc (Eggenstein, Germany); restriction enzymes from ThermoFisher (Schwerte, Germany). Oligonucleotide synthesis was performed at BioTez Berlin Buch GmbH (Berlin, Germany). Nucleic acid sequencing was carried out at Eurofins MWG Operon (Ebersberg, Germany). HPLC-grade methanol, acetonitrile, n-hexane, 2-propanol, ethanol, and water were from Fisher Scientific (Schwerte, Germany).
## 4.2. Animals
A colony of homozygous ALOX15−/− mice [35] that was provided years ago by Dr. C. Funk is kept in our animal house. These mice have been back-crossed into a C57BL/6J background several times [92] and were crossed with homozygous aP2-ALOX15 mice for ex vivo activity assays using peritoneal lavage cells (Figure 4). aP2-ALOX15 transgenic mice expressing human ALOX15 under the control of the aP2 promoter were created as described in Section 2.2.
## 4.3. RNA Extraction and qRT-PCR
A total of 10–30 mg (wet weight) of different tissues were stored in RNAlater solution (Sigma-Aldrich/Merck, Taufkirchen, Germany), after which they were cut into small pieces using a scalpel and then homogenized in 400 µL of LBP buffer (Nucleospin RNA plus kit, Macherey-Nagel, Düren, Germany) using a FastPrep24 homogenizer. Cell debris was spun down, and from the homogenate supernatant, total RNA was extracted following the instructions of the vendor of the Nucleospin RNA plus kit (Macherey-Nagel, Düren, Germany). Subsequently, 500 ng of RNA was reversely transcribed using the Tetro Reverse Transcriptase kit (Meridian Bioscience, Memphis, TN, USA, distributed by BioCat GmbH, Heidelberg, Germany) and Oligo dT18 reagents as recommended by the vendor. qRT-PCR was performed as described before [114]. Briefly, for each target gene, specific intron-spanning amplification primer combinations were synthesized (BioTez GmbH, Berlin, Germany), and external amplification standards were prepared. The following primer combinations were used: mouse ALOX15, 5′-GTACGCGGGCTCCAACAACGA-3′ and 3′-TCTCCGGGGCCCTTCACAGAA-5′; human ALOX15, 5′-ACTGAAATCGGGCTGCAAGGGG-3′ and 3′-TGGCCCACAGCCACCATAACGG-5′. Expression of target genes was quantified using standard curves (known copy numbers of the external amplification standards) and was normalized to GAPDH expression. qRT-PCR was performed on a Rotor Gene 3000 device (Corbett Research, Mortlake, Australia). Amplification products were generated, and the progress of the amplification process was followed using the SensiMixTM SYBR PCR Kit (Meridian Bioscience, Memphis, TN, USA, distributed by BioCat GmbH, Heidelberg, Germany).
## 4.4. Fluorescence In Situ Hybridization (FISH)
Prometaphase chromosomes were prepared from three 8-week-old male aP2-15LOX1 mice. Spleen tissue was disrupted in 3 mL of RPMI 1640 medium using a dounce homogenizer. Six cell-culture flasks (75 cm2) containing 20 mL of RPMI 1640 medium, supplemented with $10\%$ FCS, 7.5 µg/mL concanavalin A, and 5 µg/mL LPS (both from Sigma-Aldrich/Merck, Taufkirchen, Germany) were prepared. Then, 500 µL of homogenate was added to each flask and cultured for 48 h at 37 °C under $5\%$ CO2-containing atmosphere. The cultured cells were harvested, and the cell suspension was filled into four 50 mL blue-cap tubes. Cells were pelleted by centrifugation for 10 min at 1000 rpm. Each cell pellet was resuspended in 10 mL of RPMI 1640 containing $10\%$ FCS, and the two suspensions were combined. Then, 120 µL of Colcemid was added (Karyomax stock 10 µg/mL, ThermoFisher Scientific, Schwerte, Germany), and the cell suspensions were transferred to 15 cm Petri dishes. The dishes were incubated for 10 min at 37 °C, and then the cell suspensions were transferred into 50 mL blue-cap tubes. After centrifugation for 10 min at 1000 rpm, the supernatant was discarded, 10 mL of 75 mM KCl (37 °C) was added, and the samples were incubated for 15 min at 37 °C. Afterwards, 10 droplets of ice-cold fixative (20 mL of acetic acid + 60 mL of methanol) was added to each tube, and cells were pelleted by centrifugation (10 min at 1200 rpm, 4 °C). The supernatant was discarded and all cell pellets were combined and washed three times with 20 mL of fresh fixative at 4 °C. Finally, the cells, which were effectively reduced by the previous preparation to nuclei, were resuspended in 1 mL of fixative and kept at −20 °C until further use.
For FISH, ~0.1–0.2 mL of fixative (with cells/nuclei) were applied to clean and humid slides and air-dried. During this step, spread metaphases were formed by sequential evaporation of methanol and acetic acid. Before vanishing from the slide surface, acetic acid attracts atmospheric water, and the nucleic material spreads, leading to enlarged interphase nuclei and well-spread metaphase chromosomes [115]. Slides were processed using a standard FISH procedure as previously reported [116]. For mapping of the ALOX15, cDNA was used as probe. ALOX15 cDNA was labelled by degenerate oligonucleotide primed polymerase chain reaction (DOP-PCR), incorporating biotin-dUTPs during the reaction. The ALOX15 cDNA probe was applied in a one-color FISH experiment, and the probe was either detected by avidin-tagged Spectrum Orange or Spectrum Green. Then, 20 metaphases were acquired on a Zeiss Axioplan microscope (Carl Zeiss, Jena, Germany) equipped with corresponding filters and ISIS software (MetaSystems, Altlussheim, Germany). The positions of the acquired metaphases were registered; thus, the same 20 metaphases could be evaluated again after a second FISH was conducted on the same slide using a commercial multicolor FISH probe (M-FISH) set staining all 21 different murine chromosomes in specific color combinations (“SkyPaintTM DNA Kit M-10 for Mouse Chromosomes”, Applied Spectral Imaging, Edingen-Neckarhausen, Germany). Accordingly, the chromosome in which the human ALOX15 cDNA was inserted could be identified.
## 4.5. Whole-Genome Sequencing
Genomic DNA was prepared from 58 mg of liver tissue using the Invisorb® Spin Tissue Mini Kit (Invitek Molecular GmbH, Berlin, Germany). An additional RNAse treatment was performed, and the RNA-free DNA preparation was quality-checked with agarose gel electrophoresis. The DNA was sequenced using the shot-gun technology (ATLAS Biolabs GmbH, Berlin, Germany). The whole genome sequence data can be obtained by interested scientists upon request from Dr. K.R. Kakularam
## 4.6. Preparation of Bone Marrow Cells and Peritoneal Macrophages
For preparation of peritoneal macrophages, 10 mL of PBS was injected into the peritoneal cavity of sacrificed mice. The belly was gently massaged for 2 min and the fluid was removed by puncturing the peritoneal cavity. Usually, about 8–9 mL of cell suspension was recovered. Cells were spun down for 15 min at 800 g and were washed twice with PBS. Finally, the cells were reconstituted in 0.5 mL of PBS and were used for ex vivo ALOX activity assays. To prepare bone marrow cells, mice were sacrificed under anesthesia by cervical dislocation, and the two femur bones were prepared. The ends of the bones were cut off, and the bone marrow cavity was rinsed with 10 mL PBS. The cell suspensions were combined, and cells were pelleted (15 min, 800 g), washed twice with PBS, and reconstituted in 1 mL of PBS. Aliquots of this cell suspension were used for ex vivo ALOX activity assays quantifying the formation of oxygenated AA derivatives.
## 4.7. Ex Vivo Activity Assays
To explore whether the transgenic human ALOX15 is expressed in different cells and tissues as a catalytically active enzyme, we carried out ex vivo activity assays using tissue homogenate or cell suspensions (peritoneal macrophages, bone marrow cells) as enzyme source. For the activity assays of solid tissues, 200 mg (wet weight) of tissue were homogenized in 2 mL of PBS using the Fast Prep-24 homogenizer (MP Biomedicals, Eschwege, Germany). The tissue homogenates were centrifuged for 10 min at 15,000× g (4 °C), and the homogenate supernatants were used as enzyme source. Aliquots (20–200 µL depending on the protein content) of the homogenate supernatants were incubated at room temperature in 1 mL of PBS containing 100 µM of arachidonic acid for 15 min. The reaction was terminated by the addition of 1 mg of solid sodium borohydride. After the addition of 35 µL of acetic acid, the lipids were extracted twice with 1 mL of ethyl acetate. Then, 1 mL of 2-propanol was added, and the solvents were evaporated in a rotatory evaporator. The remaining lipids were reconstituted in 0.5 mL of RP-HPLC column solvent (acetonitrile:water:acetic acid, 70:30:0.1, by vol.), the sample was sonicated, debris was spun down, and aliquots were injected for RP-HPLC quantification of the ALOX15 products. A similar method was employed for the ex vivo activity assays of the cell suspensions. For these assays, 1–10 × 106 cells were incubated at room temperature in 0.5 mL of PBS containing 100 µM of arachidonic acid. After 10 min, the hydroperoxy fatty acids that formed were reduced by the addition of 1 mg of solid sodium borohydride, the sample was acidified, and 0.5 mL of ice-cold acetonitrile was added. The protein precipitate was spun down, and aliquots of the protein-free supernatant were injected to RP-HPLC for quantification of the hydroxy fatty acids.
## 4.8. RP-HPLC Analysis of the ALOX Products
To quantify the amounts of ALOX products formed during the incubation period of the ex vivo activity assays, a Shimadzu instrument (LC20 AD) equipped with a diode array detector (SPD M20A) was used, and the hydroxy fatty acids were separated on a Nucleodur C18 Gravity column (Macherey-Nagel, Düren, Germany; 250 × 4 mm, 5 μm particle size), which was coupled with a guard column (8 × 4 mm, 5 μm particle size). The analytes were eluted isocratically using a solvent system consisting of acetonitrile:water:acetic acid (70:30:0.1, by vol) with a flow rate of 1 mL/min at 25 °C. The absorbance at 235 nm (absorbance maximum of the conjugated dienes) was recorded, and the UV spectra of the dominant peaks that were recorded during the chromatographic runs were evaluated.
## 4.9. LC-MS/MS Analysis of the Blood Plasma Oxylipidomes
Mouse Alox15 is an arachidonic acid 12-lipoxygenating enzyme [82], whereas the human ortholog oxygenates the same substrate predominantly to 15-H(p)ETE [78]. If aP2-ALOX15 transgenic mice have significantly elevated 15-HETE plasma levels, these data may be interpreted as an indication of the in vivo activity of the transgenic enzyme. To explore whether the pattern of the plasma oxylipins was impacted by in vivo expression of the transgenic human ALOX15, we quantified the amounts of more than 40 different free-oxygenated PUFAs in the blood plasma [117]. For this purpose, EDTA blood was drawn from sacrificed mice, and after a 15 min incubation period, the blood plasma was prepared by centrifugation. Then, 10 µL of blood plasma was mixed with 450 µL of water and 10 µL of a mixture of internal standards (LTB4-d4, 20-HETE-d6, 15-HETE-d8, 13-HODE-d4, 14,15-DHET-d11, 9,10-DiHOME-d4, 12,13-EpOME-d4, 8,9-EET-d11, PGE2-d4; 10 ng/mL each). Next, 5 µL of a butylhydroxytoluene (BHT) solution were added to prevent PUFA autooxidation during sample workup and storage. Plasma proteins were precipitated by the addition of 100 µL of a 1:4 mixture (by vol.) of glycerol/water and 500 µL of acetonitrile. The pH was adjusted to 6.0 by the addition of 2 mL of phosphate buffer (0.15 M), the precipitated proteins were removed by centrifugation, and the clear supernatant was used for solid-phase lipid extraction on a 200 mg Agilent Bond Elut Certify II cartridge (Agilent Technologies, Santa Clara, USA). Before sample application, the cartridge was conditioned with 3 mL of methanol and 3 mL of phosphate buffer (0.15 M, pH 6.0). After the sample was applied, the column was washed with 3 mL of a 1:1 mixture (by vol.) of methanol: water, and the oxygenated fatty acids were eluted with a 74:25:1 mixture (by vol.) of ethyl acetate: n-hexane:acetic acid. The solvents were evaporated in a stream of nitrogen, and the remaining lipids were reconstituted in 100 µL of a 6:4 mixture (by vol.) of methanol: water and used for LC-MS/MS analysis. The chemical identity of the different analytes was concluded from co-chromatography with authentic standards, and for each of the quantified metabolites, a calibration curve was established.
LC-MS/MS was carried out on an Agilent 1290/II LC-MS system consisting of a binary pump system, an autosampler, and a column oven (Agilent Technologies, Waldbronn, Germany). As a stationary phase, we employed an Agilent Zorbax Eclipse C18 UPLC column (150 × 2.1 mm, 1.8 µm particle size). The column temperature was set at 30 °C. As a mobile phase, we used a solvent gradient that was mixed from two stock solutions. Stock A: water containing $0.05\%$ acetic acid. Stock B: 1:1 mixture (by vol.) of methanol: acetonitrile. The HPLC system was connected to a triple-quadrupole MS system (Agilent 6495 System, Agilent Technologies, Santa Clara, CA, USA). Negative electrospray ionization was carried out. The mass spectrometer was run in dynamic MRM mode, and each metabolite was detected simultaneously by two independent mass transitions that are characteristic for the different analytes. Experimental raw data were evaluated with the Agilent Mass-Hunter software package, version B10.0. For all metabolites analyzed in this study, individual calibration curves were established, and the lower detection limits were also determined. More detailed information on this analytical procedure is given in [117].
## 4.10. Reproduction Statistics
Nineteen breeding pairs (1 male, 2 females/cage) were investigated in a time frame of 4–6 months, and the following reproduction parameters were determined: litters/month, pups/litter, male/female ratio, dead pups before weaning.
## 4.11. Body-Weight Kinetics
The body weights of 5 males and 5 females (aP2-ALOX15 and C57BL/6 as controls) were determined once a week between 8 and 34 weeks of age.
## 4.12. Statistical Evaluation of the Experimental Raw Data
Statistical calculations and figure design were performed using GraphPad prism version 8.00 for Windows (GraphPad Software, La Jolla, CA, USA, (license obtained on 8 January 2021).
## 5. Conclusions
Transgenic mice expressing human ALOX15 under the control of the aP2 (activating protein 2) promoter in addition to the endogenous ALOX15 are viable and reproduce normally, but exhibited gender-specific differences with wildtype controls when their body-weight kinetics were evaluated. These mice can now be used in whole-animal disease models associated with adipose tissue hyperplasia, such as adipositas, diabetes, and metabolic syndrome.
## References
1. Andreou A., Feussner I.. **Lipoxygenases—Structure and reaction mechanism**. *Phytochemistry* (2009) **70** 1504-1510. DOI: 10.1016/j.phytochem.2009.05.008
2. Haeggström J.Z., Funk C.D.. **Lipoxygenase and Leukotriene Pathways: Biochemistry, Biology, and Roles in Disease**. *Chem. Rev.* (2011) **111** 5866-5898. DOI: 10.1021/cr200246d
3. An J.-U., Kim S.-E., Oh D.-K.. **Molecular insights into lipoxygenases for biocatalytic synthesis of diverse lipid mediators**. *Prog. Lipid Res.* (2021) **83** 101110. DOI: 10.1016/j.plipres.2021.101110
4. Chrisnasari R., Hennebelle M., Vincken J.-P., van Berkel W.J., Ewing T.A.. **Bacterial lipoxygenases: Biochemical characteristics, molecular structure and potential applications**. *Biotechnol. Adv.* (2022) **61** 108046. DOI: 10.1016/j.biotechadv.2022.108046
5. Ivanov I., Heydeck D., Hofheinz K., Roffeis J., O’Donnell V.B., Kuhn H., Walther M.. **Molecular enzymology of lipoxygenases**. *Arch. Biochem. Biophys.* (2010) **503** 161-174. DOI: 10.1016/j.abb.2010.08.016
6. Rapoport S., Schewe T.. **The maturational breakdown of mitochondria in reticulocytes**. *Biochim. et Biophys. Acta (BBA) Rev. Biomembr.* (1986) **864** 471-495. DOI: 10.1016/0304-4157(86)90006-7
7. Van Leyen K., Duvoisin R.M., Engelhardt H., Wiedmann M.. **A function for lipoxygenase in programmed organelle degradation**. *Nature* (1998) **395** 392-395. DOI: 10.1038/26500
8. Zahradka P., Neumann S., Aukema H.M., Taylor C.G.. **Adipocyte lipid storage and adipokine production are modulated by lipoxygenase-derived oxylipins generated from 18-carbon fatty acids**. *Int. J. Biochem. Cell Biol.* (2017) **88** 23-30. DOI: 10.1016/j.biocel.2017.04.009
9. Song Y.-S., Lee D.H., Yu J.-H., Oh D.-K., Hong J.T., Yoon D.-Y.. **Promotion of adipogenesis by 15-(S)-hydroxyeicosatetraenoic acid**. *Prostaglandins Other Lipid Mediat.* (2016) **123** 1-8. DOI: 10.1016/j.prostaglandins.2016.02.001
10. Brash A.R., Yu Z., Boeglin W.E., Schneider C.. **The hepoxilin connection in the epidermis**. *FEBS J.* (2007) **274** 3494-3502. DOI: 10.1111/j.1742-4658.2007.05909.x
11. Epp N., Fürstenberger G., Müller K., De Juanes S., Leitges M., Hausser I., Thieme F., Liebisch G., Schmitz G., Krieg P.. **12R-lipoxygenase deficiency disrupts epidermal barrier function**. *J. Cell Biol.* (2007) **177** 173-182. DOI: 10.1083/jcb.200612116
12. Krieg P., Fürstenberger G.. **The role of lipoxygenases in epidermis**. *Biochim. et Biophys. Acta (BBA)—Mol. Cell Biol. Lipids* (2014) **1841** 390-400. DOI: 10.1016/j.bbalip.2013.08.005
13. Giménez-Bastida J., González-Sarrías A., Laparra-Llopis J., Schneider C., Espín J.. **Targeting Mammalian 5-Lipoxygenase by Dietary Phenolics as an Anti-Inflammatory Mechanism: A Systematic Review**. *Int. J. Mol. Sci.* (2021) **22**. DOI: 10.3390/ijms22157937
14. Xu X., Li J., Zhang Y., Zhang L.. **Arachidonic Acid 15-Lipoxygenase: Effects of Its Expression, Metabolites, and Genetic and Epigenetic Variations on Airway Inflammation**. *Allergy, Asthma Immunol. Res.* (2021) **13** 684-696. DOI: 10.4168/aair.2021.13.5.684
15. Gholamalizadeh M., Majidi N., Tajaddod S., Abdollahi S., Poorhosseini S.M., Ahmadzadeh M., Joubani M.N., Dahka S.M., Shafaei H., Hajiesmaeil M.. **Interactions of Colorectal Cancer, Dietary Fats, and Polymorphisms of Arachidonate Lipoxygenase and Cyclooxygenase Genes: A Literature Review**. *Front. Oncol.* (2022) **12** 865208. DOI: 10.3389/fonc.2022.865208
16. Vaezi M.A., Safizadeh B., Eghtedari A.R., Ghorbanhosseini S.S., Rastegar M., Salimi V., Tavakoli-Yaraki M.. **15-Lipoxygenase and its metabolites in the pathogenesis of breast cancer: A double-edged sword**. *Lipids Health Dis.* (2021) **20** 1-19. DOI: 10.1186/s12944-021-01599-2
17. Roffeis J., Hornung D., Kuhn H., Walther M.. **15-Lipoxygenase-2 is differentially expressed in normal and neoplastic ovary**. *Eur. J. Cancer Prev.* (2007) **16** 568-575. DOI: 10.1097/CEJ.0b013e32801023c4
18. Vishnupriya P., Aparna A., Viswanadha V.P.. **Lipoxygenase (LOX) Pathway: A Promising Target to Combat Cancer**. *Curr. Pharm. Des.* (2021) **27** 3349-3369. DOI: 10.2174/1381612826666210101153216
19. Yan M., Zhang S., Li C., Liu Y., Zhao J., Wang Y., Yang Y., Zhang L.. **5-Lipoxygenase as an emerging target against age-related brain disorders**. *Ageing Res. Rev.* (2021) **69** 101359. DOI: 10.1016/j.arr.2021.101359
20. Kumar A., Behl T., Jamwal S., Kaur I., Sood A., Kumar P.. **Exploring the molecular approach of COX and LOX in Alzheimer’s and Parkinson’s disorder**. *Mol. Biol. Rep.* (2020) **47** 9895-9912. DOI: 10.1007/s11033-020-06033-x
21. Biringer R.G.. **The Role of Eicosanoids in Alzheimer’s Disease**. *Int. J. Environ. Res. Public Health* (2019) **16**. DOI: 10.3390/ijerph16142560
22. Dobrian A., Morris M., Taylor-Fishwick D., Holman T., Imai Y., Mirmira R., Nadler J.. **Role of the 12-lipoxygenase pathway in diabetes pathogenesis and complications**. *Pharmacol. Ther.* (2018) **195** 100-110. DOI: 10.1016/j.pharmthera.2018.10.010
23. Das U.N.. **Arachidonic acid in health and disease with focus on hypertension and diabetes mellitus: A review**. *J. Adv. Res.* (2018) **11** 43-55. DOI: 10.1016/j.jare.2018.01.002
24. Samala N., Tersey S.A., Chalasani N., Anderson R.M., Mirmira R.G.. **Molecular mechanisms of nonalcoholic fatty liver disease: Potential role for 12-lipoxygenase**. *J. Diabetes its Complicat.* (2017) **31** 1630-1637. DOI: 10.1016/j.jdiacomp.2017.07.014
25. Tersey S.A., Bolanis E., Holman T.R., Maloney D.J., Nadler J.L., Mirmira R.G.. **Minireview: 12-Lipoxygenase and Islet β-Cell Dysfunction in Diabetes**. *Mol. Endocrinol.* (2015) **29** 791-800. DOI: 10.1210/me.2015-1041
26. Ivanov I., Kuhn H., Heydeck D.. **Structural and functional biology of arachidonic acid 15-lipoxygenase-1 (ALOX15)**. *Gene* (2015) **573** 1-32. DOI: 10.1016/j.gene.2015.07.073
27. Brash A.R., Boeglin W.E., Chang M.S.. **Discovery of a second 15S-lipoxygenase in humans**. *Proc. Natl. Acad. Sci. USA* (1997) **94** 6148-6152. DOI: 10.1073/pnas.94.12.6148
28. Nugteren D.. **Arachidonate lipoxygenase in blood platelets**. *Biochim. Et Biophys. Acta (BBA)—Lipids Lipid Metab.* (1975) **380** 299-307. DOI: 10.1016/0005-2760(75)90016-8
29. Hamberg M., Samuelsson B.. **Prostaglandin endoperoxides. Novel transformations of arachidonic acid in human platelets**. *Proc. Natl. Acad. Sci. USA* (1974) **71** 3400-3404. DOI: 10.1073/pnas.71.9.3400
30. Sun D., McDonnell M., Chen X.-S., Lakkis M.M., Li H., Isaacs S.N., Elsea S.H., Patel P.I., Funk C.D.. **Human 12(R)-Lipoxygenase and the Mouse Ortholog**. *J. Biol. Chem.* (1998) **273** 33540-33547. DOI: 10.1074/jbc.273.50.33540
31. Funk C.D., Hoshiko S., Matsumoto T., Rdmark O., Samuelsson B.. **Characterization of the human 5-lipoxygenase gene**. *Proc. Natl. Acad. Sci. USA* (1989) **86** 2587-2591. DOI: 10.1073/pnas.86.8.2587
32. Rådmark O., Werz O., Steinhilber D., Samuelsson B.. **5-Lipoxygenase, a key enzyme for leukotriene biosynthesis in health and disease**. *Biochim. et Biophys. Acta (BBA)—Mol. Cell Biol. Lipids* (2015) **1851** 331-339. DOI: 10.1016/j.bbalip.2014.08.012
33. Yu Z., Schneider C., Boeglin W.E., Marnett L.J., Brash A.R.. **The lipoxygenase gene**. *Proc. Natl. Acad. Sci. USA* (2003) **100** 9162-9167. DOI: 10.1073/pnas.1633612100
34. Funk C.D., Chen X.-S., Johnson E.N., Zhao L.. **Lipoxygenase genes and their targeted disruption**. *Prostaglandins Other Lipid Mediat.* (2002) **68–69** 303-312. DOI: 10.1016/S0090-6980(02)00036-9
35. Sun D., Funk C.D.. **Disruption of 12/15-lipoxygenase expression in peritoneal macrophages. Enhanced utilization of the 5-lipoxygenase pathway and diminished oxidation of low density lipoprotein**. *J. Biol. Chem.* (1996) **271** 24055-24062. DOI: 10.1074/jbc.271.39.24055
36. Johnson E.N., Brass L.F., Funk C.D.. **Increased platelet sensitivity to ADP in mice lacking platelet-type 12-lipoxygenase**. *Proc. Natl. Acad. Sci.* (1998) **95** 3100-3105. DOI: 10.1073/pnas.95.6.3100
37. Chen X.-S., Shelter J.R., Johnson E.N., Funk C.D.. **Role of leukotrienes revealed by targeted disruption of the 5-lipoxygenase gene**. *Nature* (1994) **372** 179-182. DOI: 10.1038/372179a0
38. Krieg P., Dick A., Latzko S., Rosenberger S., Meyer J., Crumrine D., Hielscher T., Elias P.M., Rauh M., Schneider H.. **Conditional Alox12b Knockout: Degradation of the Corneocyte Lipid Envelope in a Mouse Model of Autosomal Recessive Congenital Ichthyoses**. *J. Investig. Dermatol.* (2019) **140** 249-253.e6. DOI: 10.1016/j.jid.2019.06.134
39. Krieg P., Rosenberger S., de Juanes S., Latzko S., Hou J., Dick A., Kloz U., van der Hoeven F., Hausser I., Esposito I.. **Aloxe3 Knockout Mice Reveal a Function of Epidermal Lipoxygenase-3 as Hepoxilin Synthase and Its Pivotal Role in Barrier Formation**. *J. Investig. Dermatol.* (2013) **133** 172-180. DOI: 10.1038/jid.2012.250
40. Alfardan R., Guo C., Toth L.A., Nie D.. **Impaired Recovery from Influenza A/X-31(H3N2) Infection in Mice with 8-Lipoxygenase Deficiency**. *Med. Sci.* (2019) **7**. DOI: 10.3390/medsci7040060
41. Sigal E., Grunberger D., Highland E., Gross C., A Dixon R., Craik C.S.. **Expression of cloned human reticulocyte 15-lipoxygenase and immunological evidence that 15-lipoxygenases of different cell types are related**. *J. Biol. Chem.* (1990) **265** 5113-5120. DOI: 10.1016/S0021-9258(19)34092-X
42. Singh N.K., Rao G.N.. **Emerging role of 12/15-Lipoxygenase (ALOX15) in human pathologies**. *Prog. Lipid Res.* (2018) **73** 28-45. DOI: 10.1016/j.plipres.2018.11.001
43. Çolakoğlu M., Tunçer S., Banerjee S.. **Emerging cellular functions of the lipid metabolizing enzyme 15-Lipoxygenase-1**. *Cell Prolif.* (2018) **51** e12472. DOI: 10.1111/cpr.12472
44. Schewe T., Halangk W., Hiebsch C., Rapoport S.. **A lipoxygenase in rabbit reticulocytes which attacks phospholipids and intact mitochondria**. *FEBS Lett.* (1975) **60** 149-152. DOI: 10.1016/0014-5793(75)80439-X
45. Rapoport S.M., Schewe T., Wiesner R., Halangk W., Ludwig P., Janicke-Höhne M., Tannert C., Hiebsch C., Klatt D.. **The lipoxygenase of reticulocytes. Purification, characterization and biological dynamics of the lipoxygenase, its identity with the respiratory inhibitors of the re-ticulocyte**. *Eur. J. Biochem.* (1979) **96** 545-561. DOI: 10.1111/j.1432-1033.1979.tb13068.x
46. Schewe T., Rapoport S.M., Kuhn H.. **Enzymology and physiology of reticulocyte lipoxygenase: Comparison with other lipox-ygenases**. *Adv. Enzymol. Relat. Areas Mol. Biol.* (1986) **58** 191-272. PMID: 3087141
47. Cyrus T., Witztum J.L., Rader D.J., Tangirala R., Fazio S., Linton M.F., Funk C.D.. **Disruption of the 12/15-lipoxygenase gene diminishes atherosclerosis in apo E–deficient mice**. *J. Clin. Investig.* (1999) **103** 1597-1604. DOI: 10.1172/JCI5897
48. George J., Afek A., Shaish A., Levkovitz H., Bloom N., Cyrus T., Zhao L., Funk C.D., Sigal E., Harats D.. **12/15-Lipoxygenase gene disruption attenuates athero-genesis in LDL receptor-deficient mice**. *Circulation* (2001) **104** 1646-1650. DOI: 10.1161/hc3901.095772
49. Mackel J.J., Garth J.M., Blackburn J.P., Jones M., Steele C.. **12/15-Lipoxygenase Deficiency Impairs Neutrophil Granulopoiesis and Lung Proinflammatory Responses to**. *J. Immunol.* (2020) **204** 1849-1858. DOI: 10.4049/jimmunol.1900808
50. Gaberel T., Gakuba C., Zheng Y., Lépine M., Lo E.H., van Leyen K.. **Impact of 12/15-Lipoxygenase on Brain Injury After Subarachnoid Hemorrhage**. *Stroke* (2019) **50** 520-523. DOI: 10.1161/STROKEAHA.118.022325
51. Habouri L., El Mansouri F., Ouhaddi Y., Lussier B., Pelletier J.-P., Benderdour M., Fahmi H.. **Deletion of 12/15-lipoxygenase accelerates the development of aging-associated and instability-induced osteoarthritis**. *Osteoarthr. Cartil.* (2017) **25** 1719-1728. DOI: 10.1016/j.joca.2017.07.001
52. Halade G.V., Kain V., Ingle K.A., Prabhu S.D., Lindsey M.L., Gray G.A., Wood S.K., Curran-Everett D., Bolli R., Canty J.M.. **Interaction of 12/15-lipoxygenase with fatty acids alters the leukocyte kinetics leading to improved postmyocardial infarction healing**. *Am. J. Physiol. Circ. Physiol.* (2017) **313** H89-H102. DOI: 10.1152/ajpheart.00040.2017
53. Lauder S.N., Tyrrell V.J., Allen-Redpath K., Aldrovandi M., Gray D., Collins P., A Jones S., Taylor P.R., O’Donnell V.. **Myeloid 12/15-LOX regulates B cell numbers and innate immune antibody levels in vivo**. *Wellcome Open Res.* (2017) **2** 1. DOI: 10.12688/wellcomeopenres.10308.1
54. Liu Y., Zheng Y., Karatas H., Wang X., Foerch C., Lo E.H., Van Leyen K.. **12/15-Lipoxygenase Inhibition or Knockout Reduces Warfarin-Associated Hemorrhagic Transformation After Experimental Stroke**. *Stroke* (2017) **48** 445-451. DOI: 10.1161/STROKEAHA.116.014790
55. Reilly K.B., Srinivasan S., Hatley M.E., Patricia M.K., Lannigan J., Bolick D.T., Vandenhoff G., Pei H., Natarajan R., Nadler J.L.. **12/15-Lipoxygenase Activity Mediates Inflammatory Monocyte/Endothelial Interactions and Atherosclerosis in Vivo**. *J. Biol. Chem.* (2004) **279** 9440-9450. DOI: 10.1074/jbc.M303857200
56. Harats D., Kurihara H., Belloni P., Oakley H., Ziober A., Ackley D., Cain G., Kurihara Y., Lawn R., Sigal E.. **Targeting gene expression to the vascular wall in transgenic mice using the murine preproendothelin-1 promoter**. *J. Clin. Investig.* (1995) **95** 1335-1344. DOI: 10.1172/JCI117784
57. Harats D., Shaish A., George J., Mulkins M., Kurihara H., Levkovitz H., Sigal E.. **Overexpression of 15-Lipoxygenase in Vascular Endothelium Accelerates Early Atherosclerosis in LDL Receptor–Deficient Mice**. *Arter. Thromb. Vasc. Biol.* (2000) **20** 2100-2105. DOI: 10.1161/01.ATV.20.9.2100
58. Harats D., Ben-Shushan D., Cohen H., Gonen A., Barshack I., Goldberg I., Greenberger S., Hodish I., Harari A., Varda-Bloom N.. **Inhibition of carcinogenesis in transgenic mouse models over-expressing 15-lipoxygenase in the vascular wall under the control of murine preproendothelin-1 promoter**. *Cancer Lett.* (2005) **229** 127-134. DOI: 10.1016/j.canlet.2005.02.017
59. Sultan M., Ben-Shushan D., Peled M., Kamari Y., Isman S., Barshack I., Kuban R.-J., Kühn H., Harats D., Shaish A.. **Specific overexpression of 15-lipoxygenase in endothelial cells promotes cancer cell death in an in vivo Lewis lung carcinoma mouse model**. *Adv. Med Sci.* (2020) **65** 111-119. DOI: 10.1016/j.advms.2019.11.006
60. Shen J., Kühn H., Petho-Schramm A., Chan L.. **Transgenic rabbits with the integrated human 15-lipoxygenase gene driven by a lysozyme promoter: Macrophage-specific expression and variable positional specificity of the transgenic enzyme**. *FASEB J.* (1995) **9** 1623-1631. DOI: 10.1096/fasebj.9.15.8529842
61. Shen J., Herderick E., Cornhill J.F., Zsigmond E., Kim H.S., Kühn H., Guevara N.V., Chan L.. **Macrophage-mediated 15-lipoxygenase expression protects against atherosclerosis development**. *J. Clin. Investig.* (1996) **98** 2201-2208. DOI: 10.1172/JCI119029
62. Horvai A., Palinski W., Wu H., Moulton K.S., Kalla K., Glass C.K.. **Scavenger receptor A gene regulatory elements target gene expression to macrophages and to foam cells of atherosclerotic lesions**. *Proc. Natl. Acad. Sci.* (1995) **92** 5391-5395. DOI: 10.1073/pnas.92.12.5391
63. Merched A.J., Ko K., Gotlinger K.H., Serhan C.N., Chan L.. **Atherosclerosis: Evidence for impairment of resolution of vascular inflammation governed by specific lipid mediators**. *FASEB J.* (2008) **22** 3595-3606. DOI: 10.1096/fj.08-112201
64. Kayama Y., Minamino T., Toko H., Sakamoto M., Shimizu I., Takahashi H., Okada S., Tateno K., Moriya J., Yokoyama M.. **Cardiac 12/15 lipoxygenase–induced inflammation is involved in heart failure**. *J. Exp. Med.* (2009) **206** 1565-1574. DOI: 10.1084/jem.20082596
65. Suzuki H., Kayama Y., Sakamoto M., Iuchi H., Shimizu I., Yoshino T., Katoh D., Nagoshi T., Tojo K., Minamino T.. **Arachidonate 12/15-Lipoxygenase–Induced Inflammation and Oxidative Stress Are Involved in the Development of Diabetic Cardiomyopathy**. *Diabetes* (2014) **64** 618-630. DOI: 10.2337/db13-1896
66. Madison B.B., Dunbar L., Qiao X.T., Braunstein K., Braunstein E., Gumucio D.L.. **cis Elements of the Villin Gene Control Expression in Restricted Domains of the Vertical (Crypt) and Horizontal (Duodenum, Cecum) Axes of the Intestine**. *J. Biol. Chem.* (2002) **277** 33275-33283. DOI: 10.1074/jbc.M204935200
67. Tian R., Zuo X., Jaoude J., Mao F., Colby J., Shureiqi I.. **ALOX15 as a suppressor of inflammation and cancer: Lost in the link**. *Prostaglandins Other Lipid Mediat.* (2017) **132** 77-83. DOI: 10.1016/j.prostaglandins.2017.01.002
68. Coppey L., Obrosov A., Shevalye H., Davidson E., Paradee W., Yorek M.A.. **Characterization of Mice Ubiquitously Overexpressing Human 15-Lipoxygenase-1: Effect of Diabetes on Peripheral Neuropathy and Treatment with Menhaden Oil**. *J. Diabetes Res.* (2021) **2021** 1-11. DOI: 10.1155/2021/5564477
69. Ostareck D.H., Ostareck-Lederer A., Wilm M., Thiele B.J., Mann M., Hentze M.W.. **mRNA Silencing in Erythroid Differentiation: hnRNP K and hnRNP E1 Regulate 15-Lipoxygenase Translation from the 3′ End**. *Cell* (1997) **89** 597-606. DOI: 10.1016/S0092-8674(00)80241-X
70. Ostareck-Lederer A., Ostareck D.H., Standart N., Thiele B.J.. **Translation of 15-lipoxygenase mRNA is inhibited by a protein that binds to a repeated sequence in the 3’ untranslated region**. *EMBO J.* (1994) **13** 1476-1481. DOI: 10.1002/j.1460-2075.1994.tb06402.x
71. Reimanna I., Hutha A., Thielea H., Thielea B.J.. **Suppression of 15-lipoxygenase synthesis by hnRNP E1 is dependent on repetitive nature of LOX mRNA 3′-UTR control element DICE**. *J. Mol. Biol.* (2002) **315** 965-974. DOI: 10.1006/jmbi.2001.5315
72. Lee Y.W., Kühn H., Kaiser S., Hennig B., Daugherty A., Toborek M.. **Interleukin 4 induces transcription of the 15-lipoxygenase I gene in human endothelial cells**. *J. Lipid Res.* (2001) **42** 783-791. DOI: 10.1016/S0022-2275(20)31641-2
73. Neels J.G.. **A role for 5-lipoxygenase products in obesity-associated inflammation and insulin resistance**. *Adipocyte* (2013) **2** 262-265. DOI: 10.4161/adip.24835
74. Leiria L.O., Wang C.-H., Lynes M.D., Yang K., Shamsi F., Sato M., Sugimoto S., Chen E.Y., Bussberg V., Narain N.R.. **12-Lipoxygenase Regulates Cold Adaptation and Glucose Metabolism by Producing the Omega-3 Lipid 12-HEPE from Brown Fat**. *Cell Metab.* (2019) **30** 768-783.e7. DOI: 10.1016/j.cmet.2019.07.001
75. Madsen L., Petersen R.K., Sørensen M.B., Jorgensen C., Hallenborg P., Pridal L., Fleckner J., Amri E.-Z., Krieg P., Furstenberger G.. **Adipocyte differentiation of 3T3-L1 preadipocytes is dependent on lipoxygenase activity during the initial stages of the differentiation process**. *Biochem. J.* (2003) **375** 539-549. DOI: 10.1042/bj20030503
76. Hallenborg P., Jørgensen C., Petersen R.K., Feddersen S., Araujo P., Markt P., Langer T., Furstenberger G., Krieg P., Koppen A.. **Epidermis-Type Lipoxygenase 3 Regulates Adipocyte Differentiation and Peroxisome Proliferator-Activated Receptor γ Activity**. *Mol. Cell. Biol.* (2010) **30** 4077-4091. DOI: 10.1128/MCB.01806-08
77. Kwon H.-J., Kim S.-N., Kim Y.-A., Lee Y.-H.. **The contribution of arachidonate 15-lipoxygenase in tissue macrophages to adipose tissue remodeling**. *Cell Death Dis.* (2016) **7** e2285. DOI: 10.1038/cddis.2016.190
78. Sigal E., Grunberger D., Craik C.S., Caughey G.H., A Nadel J.. **Arachidonate 15-lipoxygenase (omega-6 lipoxygenase) from human leukocytes. Purification and structural homology to other mammalian lipoxygenases**. *J. Biol. Chem.* (1988) **263** 5328-5332. DOI: 10.1016/S0021-9258(18)60719-7
79. Kühn H., Barnett J., Grunberger D., Baecker P., Chow J., Nguyen B., Bursztyn-Pettegrew H., Chan H., Sigal E.. **Overexpression, purification and characterization of human recombinant 15-lipoxygenase**. *Biochim. Et Biophys. Acta (BBA)—Lipids Lipid Metab.* (1993) **1169** 80-89. DOI: 10.1016/0005-2760(93)90085-N
80. Conrad D.J., Kuhn H., Mulkins M., Highland E., Sigal E.. **Specific inflammatory cytokines regulate the expression of human monocyte 15-lipoxygenase**. *Proc. Natl. Acad. Sci. USA* (1992) **89** 217-221. DOI: 10.1073/pnas.89.1.217
81. Shankaranarayanan P., Chaitidis P., Kühn H., Nigam S.. **Acetylation by Histone Acetyltransferase CREB-binding Protein/p300 of STAT6 Is Required for Transcriptional Activation of the 15-Lipoxygenase-1 Gene**. *J. Biol. Chem.* (2001) **276** 42753-42760. DOI: 10.1074/jbc.M102626200
82. Freire-Moar J., Alavi-Nassab A., Ng M., Mulkins M., Sigal E.. **Cloning and characterization of a murine macrophage lipoxy-genase**. *Biochim. Biophys Acta* (1995) **1254** 112-116. DOI: 10.1016/0005-2760(94)00199-9
83. Kutzner L., Goloshchapova K., Heydeck D., Stehling S., Kuhn H., Schebb N.H.. **Mammalian ALOX15 orthologs exhibit pronounced dual positional specificity with docosahexaenoic acid**. *Biochim. et Biophys. Acta (BBA)—Mol. Cell Biol. Lipids* (2017) **1862** 666-675. DOI: 10.1016/j.bbalip.2017.04.001
84. A Fischer K., Van Leyen K., Lovercamp K.W., Manandhar G., Sutovsky M., Feng D., Safranski T.. **15-Lipoxygenase is a component of the mammalian sperm cytoplasmic droplet**. *Reproduction* (2005) **130** 213-222. DOI: 10.1530/rep.1.00646
85. Soria-Tiedemann M., Michel G., Urban I., Aldrovandi M., O’Donnell V.B., Stehling S., Kuhn H., Borchert A.. **Unbalanced Expression of Glutathione Peroxidase 4 and Arachidonate 15-Lipoxygenase Affects Acrosome Reaction and In Vitro Fertilization**. *Int. J. Mol. Sci.* (2022) **23**. DOI: 10.3390/ijms23179907
86. Moore K., Lovercamp K., Feng D., Antelman J., Sutovsky M., Manandhar G., van Leyen K., Safranski T., Sutovsky P.. **Altered epididymal sperm maturation and cytoplasmic droplet migration in subfertile male Alox15 mice**. *Cell Tissue Res.* (2010) **340** 569-581. DOI: 10.1007/s00441-010-0972-x
87. Chakrabarti S.K., Cole B.K., Wen Y., Keller S.R., Nadler J.L.. **12/15-Lipoxygenase Products Induce Inflammation and Impair Insulin Signaling in 3T3-L1 Adipocytes**. *Obesity* (2009) **17** 1657-1663. DOI: 10.1038/oby.2009.192
88. Cole B.K., Morris M.A., Grzesik W.J., Leone K.A., Nadler J.L.. **Adipose Tissue-Specific Deletion of 12/15-Lipoxygenase Protects Mice from the Consequences of a High-Fat Diet**. *Mediat. Inflamm.* (2012) **2012** 1-13. DOI: 10.1155/2012/851798
89. Bleich D., Chen S., Zipser B., Sun D., Funk C.D., Nadler J.L.. **Resistance to type 1 diabetes induction in 12-lipoxygenase knockout mice**. *J. Clin. Investig.* (1999) **103** 1431-1436. DOI: 10.1172/JCI5241
90. Nunemaker C.S., Chen M., Pei H., Kimble S.D., Keller S.R., Carter J.D., Yang Z., Smith K.M., Wu R., Bevard M.H.. **12-Lipoxygenase-knockout mice are resistant to inflammatory effects of obesity induced by western diet**. *Am. J. Physiol. Metab.* (2008) **295** E1065-E1075. DOI: 10.1152/ajpendo.90371.2008
91. McDuffie M., Maybee N.A., Keller S.R., Stevens B.K., Garmey J.C., Morris M.A., Kropf E., Rival C., Ma K., Carter J.D.. **Nonobese Diabetic (NOD) Mice Congenic for a Targeted Deletion of 12/15-Lipoxygenase Are Protected From Autoimmune Diabetes**. *Diabetes* (2008) **57** 199-208. DOI: 10.2337/db07-0830
92. Rademacher M., Kuhn H., Borchert A.. **Systemic deficiency of mouse arachidonate 15-lipoxygenase induces defective erythropoiesis and transgenic expression of the human enzyme rescues this phenotype**. *FASEB J.* (2020) **34** 14318-14335. DOI: 10.1096/fj.202000408RR
93. Lefterova M.I., Haakonsson A.K., Lazar M.A., Mandrup S.. **PPARγ and the global map of adipogenesis and beyond**. *Trends Endocrinol. Metab.* (2014) **25** 293-302. DOI: 10.1016/j.tem.2014.04.001
94. Chakrabarti S.K., Wen Y., Dobrian A.D., Cole B.K., Ma Q., Pei H., Williams M.D., Bevard M.H., Vandenhoff G.E., Keller S.R.. **Evidence for activation of inflammatory lipoxygenase pathways in visceral adipose tissue of obese Zucker rats**. *Am. J. Physiol. Metab.* (2011) **300** E175-E187. DOI: 10.1152/ajpendo.00203.2010
95. Iyer A., Fairlie D.P., Prins J.B., Hammock B.D., Brown L.. **Inflammatory lipid mediators in adipocyte function and obesity**. *Nat. Rev. Endocrinol.* (2010) **6** 71-82. DOI: 10.1038/nrendo.2009.264
96. Sloane D.L., A Dixon R., Craik C.S., Sigal E.. **Expression of cloned human 15-lipoxygenase in eukaryotic and prokaryotic systems**. *Adv. Prostaglandin Thromboxane Leukot. Res.* (1991) **21** 25-28
97. Sloane D.L., Leung R., Barnett J., Craik C.S., Sigal E.. **Conversion of human 15-lipoxygenase to an efficient 12-lipoxygenase: The side-chain geometry of amino acids 417 and 418 determine positional specificity**. *Protein Eng.* (1995) **8** 275-282. DOI: 10.1093/protein/8.3.275
98. Sloane D.L., Leung R., Craik C.S., Sigal E.. **A primary determinant for lipoxygenase positional specificity**. *Nature* (1991) **354** 149-152. DOI: 10.1038/354149a0
99. Borngräber S., Kuban R.-J., Anton M., Kühn H.. **Phenylalanine 353 is a Primary Determinant for the Positional Specificity of Mammalian 15-Lipoxygenases**. *J. Mol. Biol.* (1996) **264** 1145-1153. DOI: 10.1006/jmbi.1996.0702
100. Borngraber S., Browner M., Gillmor S., Gerth C., Anton M., Fletterick R., Kuhn H.. **Shape and specificity in mammalian 15-lipoxygenase active site. The functional interplay of sequence determinants for the reaction specificity**. *J. Biol. Chem.* (1999) **274** 37345-37350. DOI: 10.1074/jbc.274.52.37345
101. Vogel R., Jansen C., Roffeis J., Reddanna P., Forsell P., Claesson H.-E., Kuhn H., Walther M.. **Applicability of the Triad Concept for the Positional Specificity of Mammalian Lipoxygenases**. *J. Biol. Chem.* (2010) **285** 5369-5376. DOI: 10.1074/jbc.M109.057802
102. Adel S., Karst F., González-Lafont A., Pekárová M., Saura P., Masgrau L., Lluch J.M., Stehling S., Horn T., Kuhn H.. **Evolutionary alteration of ALOX15 specificity optimizes the biosynthesis of antiinflammatory and proresolving lipoxins**. *Proc. Natl. Acad. Sci. USA* (2016) **113** E4266-E4275. DOI: 10.1073/pnas.1604029113
103. Heydeck D., Reisch F., Schäfer M., Kakularam K.R., Roigas S.A., Stehling S., Püschel G.P., Kuhn H.. **The Reaction Specificity of Mammalian ALOX15 Orthologs is Changed During Late Primate Evolution and These Alterations Might Offer Evolutionary Advantages for Hominidae**. *Front. Cell Dev. Biol.* (2022) **10** 871585. DOI: 10.3389/fcell.2022.871585
104. Bryant R.W., Bailey J.M., Schewe T., Rapoport S.M.. **Positional specificity of a reticulocyte lipoxygenase. Conversion of arachi-donic acid to 15-S-hydroperoxy-eicosatetraenoic acid**. *J. Biol. Chem.* (1982) **257** 6050-6055. DOI: 10.1016/S0021-9258(20)65103-1
105. Kozlov N., Humeniuk L., Ufer C., Ivanov I., Golovanov A., Stehling S., Heydeck D., Kuhn H.. **Functional characterization of novel ALOX15 orthologs representing key steps in mammalian evolution supports the Evolutionary Hypothesis of reaction specificity**. *Biochim. Et Biophys. Acta (BBA)—Mol. Cell Biol. Lipids* (2018) **1864** 372-385. DOI: 10.1016/j.bbalip.2018.12.016
106. Soberman R.J., Harper T.W., Betteridge D., Lewis R.A., Austen K.F.. **Characterization and separation of the arachidonic acid 5-lipoxygenase and linoleic acid omega-6 lipoxygenase (arachidonic acid 15-lipoxygenase) of human polymorphonuclear leukocytes**. *J Biol Chem.* (1985) **260** 4508-4515. DOI: 10.1016/S0021-9258(18)89293-6
107. Krönke G., Katzenbeisser J., Uderhardt S., Zaiss M.M., Scholtysek C., Schabbauer G., Zarbock A., Koenders M.I., Axmann R., Zwerina J.. **12/15-Lipoxygenase Counteracts Inflammation and Tissue Damage in Arthritis**. *J. Immunol.* (2009) **183** 3383-3389. DOI: 10.4049/jimmunol.0900327
108. Ackermann J.A., Hofheinz K., Zaiss M.M., Krönke G.. **The double-edged role of 12/15-lipoxygenase during inflammation and immunity**. *Biochim. Et Biophys. Acta (BBA)—Mol. Cell Biol. Lipids* (2017) **1862** 371-381. DOI: 10.1016/j.bbalip.2016.07.014
109. Serhan C.N., Levy B.D.. **Resolvins in inflammation: Emergence of the pro-resolving superfamily of mediators**. *J. Clin. Investig.* (2018) **128** 2657-2669. DOI: 10.1172/JCI97943
110. Serhan C.N., Chiang N.. **Resolution phase lipid mediators of inflammation: Agonists of resolution**. *Curr. Opin. Pharmacol.* (2013) **13** 632-640. DOI: 10.1016/j.coph.2013.05.012
111. Furstenberger G., Hagedorn H., Jacobi T., Besemfelder E., Stephan M., Lehmann W.D., Marks F.. **Characterization of an 8-lipoxygenase activity induced by the phorbol ester tumor promoter 12-O-tetradecanoylphorbol-13-acetate in mouse skin in vivo**. *J. Biol. Chem.* (1991) **266** 15738-15745. DOI: 10.1016/S0021-9258(18)98471-1
112. Bender G., Schexnaydre E.E., Murphy R.C., Uhlson C., Newcomer M.E.. **Membrane-dependent Activities of Human 15-LOX-2 and Its Murine Counterpart**. *J. Biol. Chem.* (2016) **291** 19413-19424. DOI: 10.1074/jbc.M116.741454
113. Benatzy Y., Palmer M.A., Brüne B.. **Arachidonate 15-lipoxygenase type B: Regulation, function, and its role in pathophysiology**. *Front. Pharmacol.* (2022) **13** 1042420. DOI: 10.3389/fphar.2022.1042420
114. Marbach-Breitrück E., Kalledat A., Heydeck D., Stehling S., Fluhr J.W., Zuberbier T., Kuhn H.. **Atopic Patients Show Increased Interleukin 4 Plasma Levels but the Degree of Elevation Is Not Sufficient to Upregulate Interleukin-4-Sensitive Genes**. *Ski. Pharmacol. Physiol.* (2019) **32** 192-200. DOI: 10.1159/000499431
115. Claussen U., Michel S., Mühlig P., Westermann M., Grummt U.-W., Kromeyer-Hauschild K., Liehr T.. **Demystifying chromosome preparation and the implications for the concept of chromosome condensation during mitosis**. *Cytogenet. Genome Res.* (2002) **98** 136-146. DOI: 10.1159/000069817
116. Karst C., Trifonov V., Romanenko S.A., Claussen U., Mrasek K., Michel S., Avner P., Liehr T.. **Molecular cytogenetic characterization of the mouse cell line WMP2 by spectral karyotyping and multicolor banding applying murine probes**. *Int. J. Mol. Med.* (2006) **17** 209-213. DOI: 10.3892/ijmm.17.2.209
117. Schäfer M., Kakularam K.R., Reisch F., Rothe M., Stehling S., Heydeck D., Püschel G.P., Kuhn H.. **Male Knock-in Mice Expressing an Arachidonic Acid Lipoxygenase 15B (Alox15B) with Humanized Reaction Specificity Are Prematurely Growth Arrested When Aging**. *Biomedicines* (2022) **10**. DOI: 10.3390/biomedicines10061379
|
---
title: Defective COX1 expression in aging mice liver
authors:
- Steffen Witte
- Angela Boshnakovska
- Metin Özdemir
- Arpita Chowdhury
- Peter Rehling
- Abhishek Aich
journal: Biology Open
year: 2023
pmcid: PMC10003073
doi: 10.1242/bio.059844
license: CC BY 4.0
---
# Defective COX1 expression in aging mice liver
## ABSTRACT
Mitochondrial defects are associated with aging processes and age-related diseases, including cardiovascular diseases, neurodegenerative diseases and cancer. In addition, some recent studies suggest mild mitochondrial dysfunctions appear to be associated with longer lifespans. In this context, liver tissue is considered to be largely resilient to aging and mitochondrial dysfunction. Yet, in recent years studies report dysregulation of mitochondrial function and nutrient sensing pathways in ageing livers. Therefore, we analyzed the effects of the aging process on mitochondrial gene expression in liver using wildtype C57BL/6N mice. In our analyses, we observed alteration in mitochondrial energy metabolism with age. To assess if defects in mitochondrial gene expression are linked to this decline, we applied a Nanopore sequencing based approach for mitochondrial transcriptomics. Our analyses show that a decrease of the Cox1 transcript correlates with reduced respiratory complex IV activity in older mice livers.
## Abstract
Summary: Mitochondrial function is severely affected in aging livers. Hence, the current study uses a Nanopore sequencing based approach for mitochondrial transcriptomics. The analyses show that a decrease of the Cox1 transcript correlates with reduced respiratory complex IV activity in older mice livers.
## INTRODUCTION
Mitochondria play a key role in cellular energy metabolism by providing the bulk of ATP to drive cellular activities. Moreover, they carry out additional important metabolic tasks such as the tricarboxylic acid cycle (TCA), β-oxidation of fatty acids, and ketogenesis. Due to their prokaryotic origin, mitochondria contain their own genome and a perfectly adapted transcriptomic machinery that differs substantially from those used to express the nuclear genome. The circular mitochondrial DNA (mtDNA) encodes 2 rRNAs, 22 tRNAs, and 11 mRNAs encoding 13 polypeptides that are core subunits of the oxidative phosphorylation (OXPHOS) complexes in the inner membrane (Anderson et al., 1981). Mitochondrial transcription, translation, and assembly of the OXPHOS complexes are highly complex processes. The mitochondrial gene expression process is coordinated by numerous nuclear-encoded, imported regulatory factors (D'Souza and Minczuk, 2018; Basu et al., 2020; Barshad et al., 2018). Since each mitochondrial-encoded polypeptide is part of an OXPHOS complex, mutations in mitochondrial DNA cause severe disease phenotypes (Taylor and Turnbull, 2005). In addition, accumulating mitochondrial DNA mutations have been linked to cellular aging processes (Münscher et al., 1993; Zhang et al., 1993; Nekhaeva et al., 2002; Taylor et al., 2003; McDonald et al., 2008). However, these dysfunction influence aging is highly controversial and not yet fully elucidated (Kauppila et al., 2018; Wolf, 2021; Vermulst et al., 2007).
Cellular senescence is considered as the process of general decline in cellular physiology leading to morbidity and mortality. Exploring further the connection between cellular aging and mitochondria, one finds that multifaceted pathways are linked with a mitochondrial contribution to aging (Sun et al., 2016). Conversely, age-related changes in the cell contribute to a severe decline in mitochondrial function as well (Chistiakov et al., 2014). The liver shows remarkable resilience to aging, but it is becoming increasingly clear that the liver mitochondria undergo similar cellular changes associated with aging as other tissues (Łysek-Gładysińska et al., 2021; Barazzoni et al., 2000). Aging causes changes to both the nuclear and mitochondrial genomes and the epigenome in liver (Hunt et al., 2019). Yet, also in hepatocytes, mitochondria fulfill crucial metabolic roles. In aged hepatocytes, mitochondria increase in size, show decreased membrane potential, and increased ROS production (Sugrue and Tatton, 2001; Sastre et al., 1996). Increased cell stress due to ROS production as a result of dysfunction of the OXPHOS system could also contribute to cellular aging. Because mitochondrial DNA, unlike nuclear DNA, lacks important repair and quality control mechanisms, mtDNA has been shown to be more susceptible to ROS (Miquel et al., 1980; Huang et al., 2020). However, it is not clear whether mtDNA mutations and mitochondrial malfunction are a cause, side effect, or consequence of aging. Therefore, more detailed insights into the role of mitochondrial DNA maintenance and transcription, and its adaptation to aging processes remain to be obtained.
Analyses of the nuclear transcriptome provided new insights into a variety of molecular pathways and cell type-specific aging markers (The Tabula Muris Consortium, 2020; Zhang et al., 2021). Gene expression studies in aging mouse liver revealed pathways of fibrosis and immune response to be upregulated while those of metabolism and cell cycle appear to be downregulated (Pibiri et al., 2015; White et al., 2015). Furthermore, multi-time-point transcriptomics in rats showed that the most prominent pathway downregulated with aging was oxidative phosphorylation and respiratory electron transport (Shavlakadze et al., 2019). Considering the importance of mitochondrial gene expression for the biogenesis of the OXPHOS system, it is important to examine the effect of ageing on this process. New technologies, such as the recent development of Nanopore sequencing, allow for fast genome- and transcriptome sequencing in a wide range of research areas (Wang et al., 2021). Especially, for future analyses of mitochondrial dysfunction, its adaptation to changing metabolic conditions, and involvement in human disease pathogenesis, a reliable and fast approach for monitoring of the mitochondrial transcriptome would represent a key technical asset. Therefore, we established a method for transcriptome analysis of mitochondrial mRNAs. Together with functional analyses of mitochondria, this approach provides new insights into mitochondrial changes during aging in healthy hepatocytes.
Considering a key role of mitochondria in aging and pathology, we investigated mitochondrial function in 12- and 65-week-old mice. For this, we established a Nanopore sequencing based method for mitochondrial transcriptome analyses. Our analyses revealed a robust decrease of mt-COX1 transcript abundance resulting in reduced protein levels. Loss of the core subunit COX1 is in agreement with reduced complex IV activity in aged mice samples. Our findings provide new insights into how altered gene expression contributes to the functional decline of hepatic mitochondria.
## Changes in mitochondrial membrane potential during ageing
Considering that hepatic cells are claimed to display to a certain extend resilience towards aging processes, we analyzed the correlation between different key parameters of mitochondrial physiology and age in the livers of young and aged mice 12 week (12 W) and 65 week (65 W), respectively (Fig. 1A). Livers excised from the 65 W mice showed a pale appearance. During the mitochondrial isolation the older livers also displayed a higher degree of fat layer abundance. The mitochondrial respiratory chain establishes a proton gradient across the inner membrane that drives ATP production by the F1Fo ATP synthase. To assess the mitochondrial membrane potential (Δσ), the isolated mitochondria from both experimental cohorts were stained with the membrane potential sensitive dye tetramethylrhodamin-methylester-perchlorat (TMRM). Applying the samples to flow cytometric analyses, we observed that the membrane potential was significantly reduced in the 65 W mice compared with that measured for the 12 W mice mitochondria (Fig. 1B). Similarly, when the isolated mitochondria were stained for superoxide production with fluorescent indicator MitoSOX, we found that superoxide levels were significantly increased in mitochondria from the 65 W mice compared to that of the 12 W mice (Fig. 1C). Based on these observations, we concluded that a decrease in the mitochondrial health parameters were apparent with increasing age.
**Fig. 1.:** *Aged mitochondria show a decrease in membrane potential and increased oxidative stress. (A) Overview of the experimental cohort of 12-week-old female mice compared with a cohort of 65-year-old mice and the procedures conducted. (B) Isolated mitochondria stained with TMRM to measure the mitochondrial membrane potential (column scatter plot showing mean and distribution of the biological replicates, * denotes P<0.05 from two-tailed unpaired t-test, n=3). (C) Isolated mitochondria stained with MitoSOX to measure the mitochondrial membrane potential (column scatter plot showing mean and distribution of the biological replicates, P<0.05 from two-tailed unpaired t-test, n=3).*
## Mitochondrial energy metabolism changes with age
A reduction in the mitochondrial membrane potential with increased superoxide production are indicative of dysfunction of the respiratory chain and concomitantly the oxidative phosphorylation process. Therefore, we assessed mitochondrial oxygen consumption by real time respirometry to determine if a decline in oxidative phosphorylation was apparent with age. In the presence of non-limiting amounts of ADP and substrate (state 3) actively respiring mitochondria reach a maximal physiological respiration. We found that state 3 respiration was lower in 65 W liver mitochondria as compared to those from 12 W old mice (Fig. 2A). The addition of oligomycin and CCCP to measure the uncoupled maximal respiration showed a similar difference. When we quantified averages obtained from three measurements in each case, we observed that the rate of oxygen consumption was significantly reduced under all conditions (Fig. 2B). Subsequently, we measured the NADH and NAD+ levels in liver tissue lysates. Interestingly, both metabolites were increased in the 65 W mitochondria (Fig. 2C). However, the NAD+/NADH ratio was similar in both age groups. Surprisingly, quantification of the ATP levels in the experimental cohort liver lysates showed a significantly higher levels of ATP in the 65 W mice samples (Fig. 2D). To assess if the increased amounts of ATP at steady state were the result of increased glycolysis, we also measured the lactate levels in the same samples and found them to be increased in the 65 W mice samples (Fig. 2E). Accordingly, our results indicated that in the liver samples from older mice the activity of the respiratory chain is decreased and cells display a highly glycolytic metabolism.
**Fig. 2.:** *Mitochondrial metabolism changes with age. (A) Freshly isolated mitochondria from both cohorts were subjected to oxygen consumption rate measurement via a Seahorse Flux Analyzer to determine the respiratory capacity of the mitochondria (mean±s.e.m., n=4). (B) Basal respiration was calculated upon substrate injection; uncoupled respiration was calculated after oligomycin injection and maximal respiration was calculated after CCCP injection. Relative oxygen consumption per condition was calculated from the average of three measurements (mean±s.e.m., * denotes P<0.05 from two-tailed unpaired students t-test, n=4). (C) Total NAD and NADH were determined from equal amounts of tissue lysate form both cohorts using a NAD/NADH colorimetric assay kit (Sigma-Aldrich). NAD+ was calculated as the difference between NAD total and NADH. NAD/NADH ratio was also calculated. NAD+, NADH concentration (absorbance) and NAD/NADH ratio were normalized to the mean of the young mice samples (column scatter plot showing mean and distribution of the biological replicates, * denotes P<0.05 from two-tailed unpaired t-test, ns= non-significant, n=3). (D and E) Equal amounts of tissue lysate per cohort were used to determine ATP and Lactate levels respectively via fluorescence assay kits (Abcam) (column scatter plot showing mean and distribution of the biological replicates, * denotes P<0.05 from two tailed unpaired Students t-test, n=3).*
## Mitochondrial respiratory complex IV activity reduces with age
The reduced respiration observed in the mitochondria of 65 W mice led us to examine the amounts of mtDNA in the isolated mitochondria of the different mice. For this, we used equal amounts of purified mitochondria and treated the obtained DNA with RNAse to deplete mitochondrial RNA. Subsequently, we used real time PCR for a quantitative assessment of mtDNA. For complete coverage of mtDNA we used primers designed for several mitochondrial genes. These analyses showed that the mtDNA copy numbers were similar in both 12 W and 65 W mice (Fig. 3A). Next, we analyzed the protein levels of selected subunits of the mitochondrial OXPHOS complexes. Of the mitochondrial proteins addressed, only COX1 showed a reduction in the aged experimental cohort (Fig. 3B). Quantification of the blots confirmed a statistically significant reduction only in the levels of COX1 in the 65 W cohort (Fig. 3C). These findings suggested a selective effect on the cytochrome c oxidase (complex IV) of the respiratory chain. Accordingly, we assessed the activity of complex IV in the isolated mitochondrial. As expected from the steady state protein analyses, we found that the activity of complex IV was significantly reduced in the 65 week mitochondria samples, with a pronounced degree of variability compared to the 12 week samples (Fig. 3D). These findings on the activity of complex IV are in agreement with the reduced steady state levels of COX1.
**Fig. 3.:** *OXPHOS complex IV decreases with age. (A) Equal amounts of isolated mitochondria from both the cohorts subjected to mtDNA isolation and subsequent qPCR to test the different mt-genes in order to estimate the relative mtDNA amounts (mean±s.e.m., n=4). (B) Isolated mitochondria from both the cohorts subjected to Tris Tricine-SDS-Page and immunoblotting to check for OXPHOS (n=3). (C) Area under the curves were calculate with ImageJ and were normalized to Vdac (mean±s.e.m., * denotes P<0.05 from two-tailed unpaired Students t-test, n=3) (D) Complex IV activity was measured using activity assay microplate kit (Abcam). Complex IV activity was normalized to activity of a young mice samples and the mean±s.e.m. is shown (* denotes P<0.05 from two tailed unpaired t-test, n=3).*
## Mitochondrial transcriptome profiling by Nanopore sequencing
To address as to why COX1 levels were reduced in liver mitochondria of 65 week old mice, we decided to assess mitochondrial RNA levels. Since the isolated crude mitochondrial fraction used for our analyses also contained microsomal membranes, we further enriched mitochondria by sucrose density centrifugations. Gradient-purified mitochondria were processed for RNA isolation and subsequent DNAse digestion was performed to avoid mtDNA contamination. The purified RNA was subjected to library preparation and Nanopore sequencing (Fig. 4A). The use of PCR barcoding allowed us to pool all samples and to analyze these together. Although we used a poly dT primer annealing approach for library generation, we were able to obtain sequencing reads for mitochondrial ribosomal RNAs, Rnr1 and Rnr2. A detailed analysis of the reads showed that both RNAs contain a short internal stretch of poly A (Fig. 4B). Thus, the presence of an internal A-rich sequence enables poly dT primer annealing and subsequent recovery of Rnr1 and Rnr2.
**Fig. 4.:** *OXPHOS complex IV decreases with age. (A) Experimental overview of the purification of mitochondrial samples isolated from the livers of the experimental cohorts, RNA isolation, library preparation and Nanopore sequencing. (B) Representative image of mt-Rnr2 sequencing. Bam alignment file visualization with Geneious Prime shows the reference sequence highlighted in yellow and the consensus is marked in blue. Blue box shows a magnified view of the alignment in the region where the poly dT primer anneals during the library preparation.*
Subsequently, the Nanopore sequencing results were analyzed using the Epi2me Labs differential gene expression pipeline (Oxford Nanopore Technologies). Normalization of the data to Rnr2 was carried out and results displayed in a heatmap visualization (Table S2 for Mapped Trancript Counts). Interestingly, these analyses revealed that the Cox1 transcript was strongly decreased in all the 65 W samples (Fig. 5A). Further statistical analysis showed that Cox1 was the only transcript that was significantly different between mitochondria of the 65 W samples and those from the 12 W ones (Fig. 5B). Considering that we had incorporated barcodes using PCR in the experimental setup in order to multiplex the analysis, we decided to further exclude a PCR bias and inaccurate transcript number estimation. Therefore, we carried out qPCR analyses to confirm the reduction of the Cox1 transcript by an alternative approach. Using this second strategy, we were able to confirm the reduction the Cox1 transcript in liver mitochondria of 65 W samples (Fig. 5C). In conclusion, we observed a reduction of complex IV activity due to decrease in the levels of Cox1 mRNA and protein. This finding agrees with the observed decline in OXPHOS activity and changes of mitochondrial health parameters that we observe in the liver mitochondria from mice samples with age.
**Fig. 5.:** *OXPHOS complex IV decreases with age. (A) Heatmap representation of the normalized transcript counts obtained from the differential gene expression workflow from Epi2Me Labs. Counts normalized to mt-Rnr2. Heatmap generated using GraphPad Prism 9 (n=4). (B) Volcano plot representation of (A) generated using GraphPad Prism 9 (n=4). Red denotes highly significant change. Green indicates a fold change over 0.5 and blue indicates no change. (C) Analysis of gene expression of mitochondrial transcripts in total mRNA isolated from 12 W and 65 W liver samples by qPCR. (means±s.e.m.s, ** denotes P<0.01 from two-tailed unpaired Students t-test, n=4).*
## DISCUSSION
Cellular functions decline with age. In addition to DNA damage and increased ROS production, aging has been found to be associated with metabolic dysfunction (Houtkooper et al., 2011). During the aging process, energy demands, lipid metabolism and reaction conditions change. The adaptation of the OXPHOS system to these changing conditions appears to be a highly dynamic process that can be influenced by a variety of (external) factors such as diet or exercise (Perelló-Amorós et al., 2021; Chen et al., 2018; McCoin et al., 2019; Han et al., 2012). In this context, the liver represents a critical organ for energy and lipid metabolism and hepatic mitochondria are a central hub for oxidative phosphorylation, fatty acid metabolism, and ketogenesis. Therefore, to study the effects of adaptation of metabolic pathways to changing conditions hepatocytes represent a suitable cellular system. Moreover, the liver is the only visceral organ that can regenerate and is exposed to a high degree to changing metabolic conditions (Michalopoulos and Bhushan, 2021). Therefore, mitochondria need to adapt to metabolic challenges. They proliferate in a process of mitochondrial biogenesis through fusion and fission from pre-existing mitochondria (Ploumi et al., 2017). Thus, the transcription and translation of nuclear genes coding for mitochondrial proteins appears to be critical for many aspects of cellular physiology (Kotrys and Szczesny, 2020).
Here, we compared mitochondrial functions in the livers of young and old mice. Our analyses find that both ATP and lactate levels are significantly increased in liver in the aged group. This, in conjunction with the reduced oxygen consumption rates, indicates that the livers become increasingly glycolytic with age. Our finding is in line with previous studies which reported increased lactate levels and reduced glycolytic intermediates indicative of elevated anaerobic glycolysis (Houtkooper et al., 2011). Surprisingly, the NAD+ levels were found to be increased in this study. In contrast, other studies report a hepatic NAD+ deficiency in aged mice and humans (Zhou et al., 2016). However, in our analyses we found that the NAD+/NADH ratio remains unaffected in 65 W mice compared to the 12 W age cohort. In addition, we found a decrease in oxidative phosphorylation capacity of the mitochondria that are linked to a loss in COX1, the mitochondrial-encoded core subunit of the cytochrome c oxidase.
Third-generation sequencing has revolutionized many aspects of biology from genome assemblies, metagenomics, epigenetics and transcriptomics (Kraft and Kurth, 2019). Additionally, the use of the MinION sequencer (Oxford Nanopore Technologies) allows rapid, sensitive and real time long read sequencing of nucleic acids (Player et al., 2020; Tyler et al., 2018). Coupled with barcoding, it possible to further reduce the machine run time by pooling all the samples and replicates in a single run. Here we applied this experimental approach to address if a loss COX1 was linked to the availability of the corresponding transcript. For this, we purified mitochondria to eliminate cytosolic RNAs from the samples to be analyzed. This enrichment allowed for extensive sequencing of the mitochondrial transcripts alone. However, on the downside this approach reduced the number of housekeeping genes that could be used for normalizing the data during analysis. Therefore, we started with equal amounts of purified mitochondria for RNA extraction and subsequently used equal amounts of RNA for cDNA preparations and library generation. We chose to base our analyses on polyadenylated RNAs for the initiation for cDNA and library preparation. The presence of an internal poly A stretch, long enough for the oligo dT primer to bind, enabled the analysis of the mitochondrial Rnr1 and Rnr2. Regarding the Cox1 transcript, the analyses revealed that in aged mitochondria the amount of the mRNA was specifically reduced in liver mitochondria of the 65 W mice while other transcripts were not affected. This reduction agrees with the observed decline in the respiratory activity.
## MATERIALS AND METHODS
Key resources are specified in table S1.
## Animals
Maintenance of all mice and their study were performed according to the guidelines from the German Animal Welfare Act and approved by the Landesamt für Verbraucherschutz und Lebensmittelsicherheit, Niedersachsen, Germany (AZ: 33.9-42502-04-$\frac{14}{1720}$). The animals were kept either in high barrier (SPF-specified pathogen free) areas in IVC (individually ventilated caging) on standard rodent chow to WT C57BL/6N mice, with restricted access for animal care staff only. The WT C57BL/6N mice were historically acquired from Charles River Laboratories, Research Models and Services, Germany GmbH, Sulzfeld. They were subsequently maintained at the animal facility at Max Planck Institute for Multidisciplinary Sciences, Göttingen till they reached appropriate age for experimentation.
## Mitochondrial isolation
The animals were sacrificed at the respective ages to isolate the liver. These were then homogenized using glass potters in 15 ml of Isolation Buffer (IB), containing 10 mM Tris-MOPS pH 7.4, 1 mM EGTA/Tris and 200 mM Sucrose. Nuclei, debris and unlyzed cells were removed by centrifugation at 700×g, 10 min, 4°C. Mitochondria were further pelleted at 7000×g, 10 min, 4°C. They were then washed and resuspended in Isolation Buffer and their protein concentration was determined using Bradford assay.
## Mitochondrial sucrose-gradient purification
To remove extramitochondrial nucleic acids from the isolated mitochondria, samples were first treated with 50 U Benzonase for 30 min, at 4°. Centrifugation (2×, 14,000×g, 10 min, 4°C) and resuspension in IB (composition as described in mitochondrial isolation, containing 2.5 mM EDTA) stopped the Benzonase activity. A sucrose gradient was prepared by placing 1 volume of isolation buffer containing 1.7 M sucrose into a centrifuge tube and overlaying this with 2 volumes of isolation buffer containing 1 M sucrose. Mitochondria were finally resuspended in 1 M sucrose isolation buffer and carefully loaded onto the gradient. After centrifugation for 25 min, 25,000 rpm (SW41Ti, Beckman Coulter) at 4°C, mitochondria were isolated from the interphase.
## Mitochondrial membrane potential measurement
Membrane potential measurement was conducted on isolated liver mitochondria (as described above) using a Flow Cytometer (FACSCanto, BD Biosciences). The mitochondria were resuspended in freshly prepared Analysis buffer (pH 7.0; 250 mM Sucrose, 20 mM Tris-MOPS, 100 µM Pi(K), 0.5 mM MgCl2, 5 mM Succinate, 2 µM Rotenone). 100 µM Tetramethylrhodamin-methylester-perchlorat, TMRM (Life Tech; T668) was added to all samples except the unstained control and the samples were incubated for 10 min at room temperature, protected from light. The measurement was done at Ex488/Em590 nm.
## Mitochondrial respiration analysis (Seahorse)
Oxygen Consumption Rate (OCR) of freshly isolated liver mitochondria (isolation as described above) was obtained via respirometry on a Seahorse XFe96 analyzer (Agilent; S7894-10000). The mitochondria were resuspended in Mitochondrial Assay (MAS) Buffer (pH 7.4; 70 mM Sucrose, 210 mM Mannitol, 5 mM HEPES, 1 mM EGTA, 10 mM KH2PO4, 5 mM MgCl2, $0.5\%$ BSA(w/v) to a concentration of 0.1 mg/ml. The mitochondria were aliquoted in the Seahorse XF96 Cell Culture Microplate (Agilent; 101085-004) and the plate was centrifuged at 2000×g for 5 min at room temperature. In port A of the Seahorse XFe96 Sensor Cartridge (Agilent; 101085-004) either a Pyruvate (0.1 M Pyruvate, 40 mM Succinate, 40 mM ADP in MAS buffer) or Succinate (100 µM Succinate, 40 µM ADP in MAS buffer) mix was provided as substrate for the mitochondria. 30 µM of Oligomycin, 40 µM of CCCP, 10 µM of Antimycin and Rotenone dissolved in MAS buffer were added in ports B, C, and D of the Sensor Cartridge respectively. A modified Mito Stress Test protocol was used for the measurement. The modifications were the removal of the Equilibration step for the Cell Plate to minimize the stress time on the isolated mitochondria and the addition of another Port injection to accommodate for the substrate.
## NAD/NADH measurement
Quantification of NAD+/NADH ratio in mouse liver tissue samples was done using a NAD+/NADH Quantification Kit (Sigma-Aldrich; MAK037). The protocol provided by the supplier was followed. NADtotal as well as NADH were detected measuring absorbance at 450 nm on a microplate reader (Synergy H1; BioTek; 8041000) following mechanical lysis and deproteination of the lysates. In order to measure NADH separately from the NADtotal, NAD+ was thermally decomposed by a 30 min incubation at 60°C.
## ATP measurement
Measurement of total tissue ATP content was done using the ATP Assay Kit (Colorimetric/Fluorometric) (Abcam; ab83355) and as specified by the protocol from the supplier. In order to determine total ATP mouse liver tissue was lysed mechanically and deproteinized. Samples were incubated for 30 min at room temperature with the reaction mix in triplicates. The phosphorylation of glycerol resulting in a product detectable at 570 nm was measured using a microplate reader (Synergy H1; BioTek; 8041000).
## Lactate measurement
L-Lactate Assay Kit (Colorimetric/Fluorometric) (Abcam; ab65330) was used in order to detect lactate in mouse liver tissue lysate. The experiment was performed as instructed in the protocol provided by the supplier. Following mechanical lysis, the samples were deproteinized and assayed for 30 min. Lactate was detected calorimetrically at 570 nm. The measurement was done in triplicate and visualized using a microplate reader (Synergy H1; BioTek; 8041000).
## Mitochondrial DNA isolation
The QIAamp® DNA Blood Mini Kit was used to isolate mitochondrial DNA from mitochondria. The isolated mitochondria (see above) were diluted 1:50 in the provided sample buffer and the manufacturer's protocol was executed.
## RNA isolation and purification
The isolated and purified mitochondria were lysed in 1 ml TRIzol using the vendor's protocol. The samples were then treated with DNase I (ThermoFisher Scientific) following the manufacturer's instructions. For purification of the sample, the RNA Clean and Concentrator Kit (R1013, Zymo Research) was used. The RNA quality and concentration were measured using a Nanodrop 2000.
## cDNA synthesis and qPCR
cDNA was synthesized using the RevertAid First Strand cDNA Synthesis Kit (K1621; ThermoFisher Scientific), specifically using the Random Hexamer primers from the isolated RNA (described in the previous section) in a thermocycler (Labcycler Gradient, SensoQuest GmbH). The cDNA was used in quadruplicates for a Real Time PCR reaction with Sensi Mix SYBR Low-ROX Kit (QT625-05; Bioline) on a Quant Studio 6 Flex Real-Time PCR system (4485691; ThermoFisher Scientific). All the primer sequences of the primers used are available upon request.
## Tricine-SDS PAGE and western blot
Tricine-SDS PAGE was performed using standard methods. The samples were lysed in T-PER Tissue Protein Extraction Buffer (ThermoFisher Scientific; 78510). The gels used were gradient 10-$18\%$ gels. Western blotting was preformed using standard semi-dry transfer.
## Cytochrome c oxidase activity assay
The Complex IV Rodent Enzyme Activity Microplate Assay Kit (Abcam; ab109911) was used to assess the activity of the Cytochrome c Oxidase in isolated mouse liver mitochondria. The procedure was performed as instructed by the provider. The principle of the assay is the use of immunoprecipitated mitochondrial complex IV, whose activity gets determined by the oxidation of Cytochrome c. A microplate reader (Synergy H1; BioTek; 8041000) was used to measure the absorbance of cytochrome c, which turns colorless due to its oxidation and is detectable at 550 nm.
## Library preparation
Total mitochondrial RNA was purified and concentrated on an RNA Clean Concentrator™-5 column (Zymo Research, Irvine, CA, USA). cDNA libraries were prepared from a mix of 50 ng RNA according to the Oxford Nanopore Technologies (Oxford Nanopore Technologies Ltd, Oxford, UK) protocol ‘DNA-PCR Sequencing’ with a 14 cycles PCR (8 min for elongation time). ONT adapters were ligated to 650 ng of cDNA.
## Nanopore sequencing and analysis
Nanopore libraries were sequenced using a MinION Mk1b with R9.4.1 flowcells. The data were generated and basedcalled using MinKNOW (Version 21.11.9). The fastq files were then uploaded to and analyzed using the Epi2Me Labs (Version 1.1) Differential Gene Expression workflow. Transcript counts obtained were processed for heatmap and volcano plot generation using Prism 9 software (GraphPad Software, San Diego, CA, USA).
## Statistical analysis
Significant differences between means of two (or more) sets of data were analyzed using either unpaired t-tests or ANOVA tests in Prism 9 software (GraphPad Software, San Diego, CA, USA, unless otherwise noted.
## Ethics approval and consent to participate
Maintenance of all mice and their study were performed according to the guidelines from the German Animal Welfare Act and approved by the Lower Saxony government office ethic committee, Landesamt für Verbraucherschutz und Lebensmittelsicherheit, Niedersachsen, Germany (AZ: 33.9-42502-04-$\frac{14}{1720}$). All methods are reported in accordance with ARRIVE guidelines for the reporting of animal experiments.
## References
1. Anderson S., Bankier A. T., Barrell B. G., De Bruijn M. H. L., Coulson A. R., Drouin J., Eperon I. C., Nierlich D. P., Roe B. A., Sanger F.. **Sequence and organization of the human mitochondrial genome**. (1981) **290** 457-465. DOI: 10.1038/290457a0
2. Barazzoni R., Short K. R., Nair K. S.. **Effects of aging on mitochondrial DNA copy number and cytochrome c oxidase gene expression in rat skeletal muscle, liver, and heart**. (2000) **275** 3343-3347. DOI: 10.1074/jbc.275.5.3343
3. Barshad G., Marom S., Cohen T., Mishmar D.. **Mitochondrial DNA transcription and its regulation: an evolutionary perspective**. (2018) **34** 682-692. DOI: 10.1016/j.tig.2018.05.009
4. Basu U., Bostwick A. M., Das K., Dittenhafer-Reed K. E., Patel S. S.. **Structure, mechanism, and regulation of mitochondrial DNA transcription initiation**. (2020) **295** 18406-18425. DOI: 10.1074/jbc.rev120.011202
5. Chen P. B., Yang J. S., Park Y.. **Adaptations of skeletal muscle mitochondria to obesity, exercise, and polyunsaturated fatty acids**. (2018) **53** 271-278. DOI: 10.1002/lipd.12037
6. Chistiakov D. A., Sobenin I. A., Revin V. V., Orekhov A. N., Bobryshev Y. V. **Mitochondrial aging and age-related dysfunction of mitochondria**. (2014) **2014** 238463. DOI: 10.1155/2014/238463
7. D'Souza A. R., Minczuk M.. **Mitochondrial transcription and translation: overview**. (2018) **62** 309-320. DOI: 10.1042/ebc20170102
8. Han D., Ybanez M. D., Johnson H. S., Mcdonald J. N., Mesropyan L., Sancheti H., Martin G., Martin A., Lim A. M., Dara L.. **Dynamic adaptation of liver mitochondria to chronic alcohol feeding in mice**. (2012) **287** 42165-42179. DOI: 10.1074/jbc.m112.377374
9. Houtkooper R. H., Argmann C., Houten S. M., Cantó C., Jeninga E. H., Andreux P. A., Thomas C., Doenlen R., Schoonjans K., Auwerx J.. **The metabolic footprint of aging in mice**. (2011) **1** 134. DOI: 10.1038/srep00134
10. Huang Z., Chen Y., Zhang Y.. **Mitochondrial reactive oxygen species cause major oxidative mitochondrial DNA damages and repair pathways**. (2020) **45** 84. DOI: 10.1007/s12038-020-00055-0
11. Hunt N. J., Kang S. W. S., Lockwood G. P., Couteur D. G. L., Cogger V. C.. **Hallmarks of aging in the liver**. (2019) **17** 1151-1161. DOI: 10.1016/j.csbj.2019.07.021
12. Kauppila T. E. S., Bratic A., Jensen M. B., Baggio F., Partridge L., Jasper H., Grönke S., Larsson N.-G.. **Mutations of mitochondrial DNA are not major contributors to aging of fruit flies**. (2018) **115** E9620-E9629. DOI: 10.1073/pnas.1721683115
13. Kotrys A. V., Szczesny R. J.. **Mitochondrial gene expression and beyond—novel aspects of cellular physiology**. (2020) **9** 17. DOI: 10.3390/cells9010017
14. Kraft F., Kurth I.. **Long-read sequencing in human genetics**. (2019) **31** 198-204. DOI: 10.1007/s11825-019-0249-z
15. Łysek-Gładysińska M., Wieczorek A., Jóźwik A., Walaszczyk A., Jelonek K., Szczukiewicz-Markowska G., Horbaå„Czuk O. K., Pietrowska M., Widå,Ak P., Gabryå D.. **Aging-related changes in the ultrastructure of hepatocytes and cardiomyocytes of elderly mice are enhanced in ApoE-deficient animals**. (2021) **10** 502. DOI: 10.3390/cells10030502
16. Mccoin C. S., Schulze A. V., Allen J., Fuller K. N. Z., Xia Q., Koestler D. C., Houchen C. J., Maurer A., Dorn G. W., Shankar K.. **Sex modulates hepatic mitochondrial adaptations to high-fat diet and physical activity**. (2019) **317** E298-E311. DOI: 10.1152/ajpendo.00098.2019
17. Mcdonald S. A. C., Greaves L. C., Gutierrez–Gonzalez L., Rodriguez–Justo M., Deheragoda M., Leedham S. J., Taylor R. W., Lee C. Y., Preston S. L., Lovell M.. **Mechanisms of field cancerization in the human stomach: the expansion and spread of mutated gastric stem cells**. (2008) **134** 500-510. DOI: 10.1053/j.gastro.2007.11.035
18. Michalopoulos G. K., Bhushan B.. **Liver regeneration: biological and pathological mechanisms and implications**. (2021) **18** 40-55. DOI: 10.1038/s41575-020-0342-4
19. Miquel J., Economos A. C., Fleming J., Johnson J. E.. **Mitochondrial role in cell aging**. (1980) **15** 575-591. DOI: 10.1016/0531-5565(80)90010-8
20. Münscher C., Rieger T., Müller-Höcker J., Kadenbach B.. **The point mutation of mitochondrial DNA characteristic for MERRF disease is found also in healthy people of different ages**. (1993) **317** 27-30. DOI: 10.1016/0014-5793(93)81484-h
21. Nekhaeva E., Bodyak N. D., Kraytsberg Y., Mcgrath S. B., Orsouw N. J. V., Pluzhnikov A., Wei J. Y., Vijg J., Khrapko K.. **Clonally expanded mtDNA point mutations are abundant in individual cells of human tissues**. (2002) **99** 5521-5526. DOI: 10.1073/pnas.072670199
22. Perelló-Amorós M., Fernández-Borràs J., Sánchez-Moya A., Vélez E. J., García-Pérez I., Gutiérrez J., blasco J.. **Mitochondrial adaptation to diet and swimming activity in Gilthead Seabream: improved nutritional efficiency**. (2021) **12** 678985. DOI: 10.3389/fphys.2021.678985
23. Pibiri M., Sulas P., Leoni V. P., Perra A., Kowalik M. A., Cordella A., Saggese P., Nassa G., Ravo M.. **Global gene expression profile of normal and regenerating liver in young and old mice**. (2015) **37** 9796. DOI: 10.1007/s11357-015-9796-7
24. Player R., Verratti K., Staab A., Bradburne C., Grady S., Goodwin B., Sozhamannan S.. **Comparison of the performance of an amplicon sequencing assay based on Oxford Nanopore technology to real-time PCR assays for detecting bacterial biodefense pathogens**. (2020) **21** 166. DOI: 10.1186/s12864-020-6557-5
25. Ploumi C., Daskalaki I., Tavernarakis N.. **Mitochondrial biogenesis and clearance: a balancing act**. (2017) **284** 183-195. DOI: 10.1111/febs.13820
26. Sastre J., Pallardó F. V., Plá R., Pellín A., Juan G., O'connor J. E., Estrela J. M., Miquel J., Viña J.. **Aging of the liver: Age-associated mitochondrial damage in intact hepatocytes: Aging of the liver: Age-associated mitochondrial damage in intact hepatocytes**. (1996) **24** 1199-1205. DOI: 10.1002/hep.510240536
27. Shavlakadze T., Morris M., Fang J., Wang S. X., Zhu J., Zhou W., Tse H. W., Mondragon-Gonzalez R., Roma G., Glass D. J.. **Age-related gene expression signature in rats demonstrate early, late, and linear transcriptional changes from multiple tissues**. (2019) **28** 3263-3273.e3. DOI: 10.1016/j.celrep.2019.08.043
28. Sugrue M. M., Tatton W. G.. **Mitochondrial membrane potential in aging cells**. (2001) **10** 176-188. DOI: 10.1159/000046886
29. Sun N., Youle R. J., Finkel T.. **The mitochondrial basis of aging**. (2016) **61** 654-666. DOI: 10.1016/j.molcel.2016.01.028
30. Taylor R. W., Barron M. J., Borthwick G. M., Gospel A., Chinnery P. F., Samuels D. C., Taylor G. A., Plusa S. M., Needham S. J., Greaves L. C.. **Mitochondrial DNA mutations in human colonic crypt stem cells**. (2003) **112** 1351-1360. DOI: 10.1172/jci19435
31. Taylor R. W., Turnbull D. M.. **Mitochondrial DNA mutations in human disease**. (2005) **6** 389-402. DOI: 10.1038/nrg1606
32. **A single-cell transcriptomic atlas characterizes ageing tissues in the mouse**. (2020) **583** 590-595. DOI: 10.1038/s41586-020-2496-1
33. Tyler A. D., Mataseje L., Urfano C. J., Schmidt L., Antonation K. S., Mulvey M. R., Corbett C. R.. **Evaluation of Oxford Nanopore's MinION sequencing device for microbial whole genome sequencing applications**. (2018) **8** 10931. DOI: 10.1038/s41598-018-29334-5
34. Vermulst M., Bielas J. H., Kujoth G. C., Ladiges W. C., Rabinovitch P. S., Prolla T. A., Loeb L. A.. **Mitochondrial point mutations do not limit the natural lifespan of mice**. (2007) **39** 540-543. DOI: 10.1038/ng1988
35. Wang Y., Zhao Y., Bollas A., Wang Y., Au K. F.. **Nanopore sequencing technology, bioinformatics and applications**. (2021) **39** 1348-1365. DOI: 10.1038/s41587-021-01108-x
36. White R. R., Milholland B., Macrae S. L., Lin M., Zheng D., Vijg J.. **Comprehensive transcriptional landscape of aging mouse liver**. (2015) **16** 899. DOI: 10.1186/s12864-015-2061-8
37. Wolf A. M.. **MtDNA mutations and aging—not a closed case after all?**. (2021) **6** 56. DOI: 10.1038/s41392-021-00479-6
38. Zhang C. F., Linnane A. W., Nagley P.. **Occurrence of a particular base substitution (3243 A to G) in mitochondrial DNA of tissues of ageing humans**. (1993) **195** 1104-1110. DOI: 10.1006/bbrc.1993.2158
39. Zhang M. J., Pisco A. O., Darmanis S., Zou J.. **Mouse aging cell atlas analysis reveals global and cell type-specific aging signatures**. (2021) **10** e62293. DOI: 10.7554/elife.62293
40. Zhou C.-C., Yang X., Hua X., Liu J., Fan M.-B., Li G.-Q., Song J., Xu T.-Y., Li Z.-Y., Guan Y.-F.. **Hepatic NAD(+) deficiency as a therapeutic target for non-alcoholic fatty liver disease in ageing**. (2016) **173** 2352-2368. DOI: 10.1111/bph.13513
|
---
title: The PPAR-γ Agonist Pioglitazone Modulates Proliferation and Migration in HUVEC,
HAOSMC and Human Arteriovenous Fistula-Derived Cells
authors:
- Carmen Ciavarella
- Ilenia Motta
- Francesco Vasuri
- Teresa Palumbo
- Anthony Paul Lisi
- Alice Costa
- Annalisa Astolfi
- Sabrina Valente
- Piera Versura
- Eugenio F. Fornasiero
- Raffaella Mauro
- Mauro Gargiulo
- Gianandrea Pasquinelli
journal: International Journal of Molecular Sciences
year: 2023
pmcid: PMC10003103
doi: 10.3390/ijms24054424
license: CC BY 4.0
---
# The PPAR-γ Agonist Pioglitazone Modulates Proliferation and Migration in HUVEC, HAOSMC and Human Arteriovenous Fistula-Derived Cells
## Abstract
The failure of arteriovenous fistulas (AVFs) following intimal hyperplasia (IH) increases morbidity and mortality rates in patients undergoing hemodialysis for chronic kidney disease. The peroxisome-proliferator associated receptor (PPAR-γ) may be a therapeutic target in IH regulation. In the present study, we investigated PPAR-γ expression and tested the effect of pioglitazone, a PPAR-γ agonist, in different cell types involved in IH. As cell models, we used Human Endothelial Umbilical Vein Cells (HUVEC), Human Aortic Smooth Muscle Cells (HAOSMC), and AVF cells (AVFCs) isolated from (i) normal veins collected at the first AVF establishment (T0), and (ii) failed AVF with IH (T1). PPAR-γ was downregulated in AVF T1 tissues and cells, in comparison to T0 group. HUVEC, HAOSMC, and AVFC (T0 and T1) proliferation and migration were analyzed after pioglitazone administration, alone or in combination with the PPAR-γ inhibitor, GW9662. Pioglitazone negatively regulated HUVEC and HAOSMC proliferation and migration. The effect was antagonized by GW9662. These data were confirmed in AVFCs T1, where pioglitazone induced PPAR-γ expression and downregulated the invasive genes SLUG, MMP-9, and VIMENTIN. In summary, PPAR-γ modulation may represent a promising strategy to reduce the AVF failure risk by modulating cell proliferation and migration.
## 1. Introduction
The arteriovenous fistula (AVF) is the choice procedure for vascular access in patients subjected to hemodialysis for chronic kidney disease (CKD), whose incidence is increasing worldwide and represents a risk factor for cardiovascular disease (CVD) [1]. Despite the higher patency rate and the lower risk of complications in comparison to the other vascular access options, AVF carries a high percentage of failure because of inadequate maturation or degeneration, resulting in increased morbidity and further surgical operations [2,3,4,5]. AVF degeneration is commonly caused by neointimal hyperplasia (IH), a complex process characterized by progressive intimal thickening and consequent stenosis with lumen occlusion [3,6]. Shear stress and inflammation have been proposed as typical triggering events in the IH development [7,8]. IH pathogenesis involves different steps, starting from endothelial cell (ECs) damage and dysfunction [9], inflammation and smooth muscle cell (SMCs) phenotype transdifferentiation, proliferation and migration, culminating in vascular remodeling and matrix deposition (including calcification) [6]. CKD accelerates these processes, as observed in mice where it promoted endothelial dysfunction and neointimal formation in association with a lower expression of the junction protein Vascular Endothelial (VE)-Cadherin [10]. It has recently been proposed that endothelial-to-mesenchymal transition (End-MT) plays a pathogenic role in cardiovascular diseases, such as atherosclerosis [11] and plaque calcification [12]. End-MT is a phenotype conversion involving ECs that progressively lose the typical endothelial markers (i.e., CD31 and VE-Cadherin), shapes, and features, transiting into mesenchymal cells characterized by specific markers, morphology, and function [13]. A study performed on human atherosclerotic plaques and porcine aortas demonstrated that End-MT, triggered by shear stress, contributes to IH development [14].
The peroxisome-proliferator-associated-receptor-γ (PPAR-γ) belongs to a family of nuclear receptors, involved in the regulation of a broad range of biological mechanisms. Primarily, PPAR-γ is well known as a master regulator of the adipogenic differentiation process, by stimulating pre-adipocyte differentiation in mature adipocytes [15,16]. PPAR-γ also regulates glucose metabolism [17] and a class of PPAR-γ-agonists, the thiadolidondines, which is commonly used for type 2-diabetes treatment [18]. Additionally, PPAR-γ plays a pivotal function in regulating other biological processes, including cell proliferation and migration. In particular, it has been shown that PPAR-γ exerts a protective role in atherosclerosis, acting on vascular SMCs, ECs, and inflammatory cells. For this reason, the use of PPAR-γ agonists, such as pioglitazone or rosiglitazone, have been proposed for the therapy of cardiovascular diseases. However, the role of PPAR-γ in the progression of IH contextually to AVF failure has not been fully elucidated.
Here, we investigated the expression of PPAR-γ and the effects exerted by pioglitazone in cell populations mostly involved in IH pathogenesis, i.e., ECs and SMCs. Further, we established and characterized a model of primary cells isolated from veins subjected to the AVF procedure in order to investigate the nature of cells within the IH lesion and to reproduce the pathological context in vitro.
## 2.1. Patient Characteristics
The present study included 12 end-stage kidney disease patients subjected to AVF surgery, distinguished in two groups (T0: normal veins collected at the first AVF establishment, including 3 males and 3 females, mean age 71 ± 13.7; T1: failed AVF with IH, including 3 males and 3 females, mean age 65 ± 12). The groups were homogenous for sex and age. Obesity, dyslipidemia, and diabetes mellitus were present in $14\%$, $42.8\%$, and $43\%$ of all cases, respectively. Hypertension occurred in all cases, and tobacco smoke risk factor was recorded for the $43\%$ of the patient group.
## 2.2. Altered Distribution and Expression of PPAR-γ in Failed AVF Veins and Cells
In order to explore whether PPAR-γ could be a good candidate target for IH modulation and AVF failure prevention, we analyzed protein expression and localization in AVF tissues. Based on the histological findings, we observed an altered expression and distribution of PPAR-γ in failed AVF (T1). PPAR-γ was broadly expressed in native veins T0, mostly in SMCs within the tunica media (Figure 1a,b), whereas a differential expression pattern was highlighted in failed AVF. As evidenced in Figure 1c, PPAR-γ was not completely absent in the AVF veins and positive areas were detectable within the tissue surrounding the IH lesion. Conversely, all the areas interested by the IH lesion were negative to protein expression (Figure 1d). The percentage of PPAR-γ positive areas was assessed by applying the ImageJ deconvolution color tool on random fields in 10× images, confirming a lower percentage of receptor expression in AVF T1 veins (Figure 1e, percentage of PPAR-γ positive areas: 6.48 ± 0.97 in AVF T0 and 1.18 ± 0.29 in AVF T1; p value 0.0002, unpaired t-test). This PPAR-γ distribution pattern suggests the possible involvement of this transcription factor during the IH disease development and progression through the unbalanced production and function of PPAR-γ in proliferating cells that characterize the IH lesion. This result also found confirmation in mRNA analysis performed in the AVF cell model, where AVFCs T1 exhibited lower transcript levels of PPAR-γ (a $50\%$ reduction) when compared to AVFCs T0 (Figure 1f).
## 2.3. The PPAR-γ Agonist Pioglitazone Affects Proliferation and Migration Property in Endothelial and Smooth Muscle Cells
In order to test whether endothelial and smooth muscle cells, the key players during IH initiation and progression, were sensitive to PPAR-γ modulation through pharmacological approaches, we analyzed the effects of pioglitazone in HUVEC and HAOSMC.
According to the proliferation assay performed through crystal violet staining, we observed differences in dose/time response of HUVEC and HAOSMC to pioglitazone. In detail, HUVEC growth underwent a $30\%$ decrease when exposed to pioglitazone at 10 μM for 24 h. This effect was prevented when pioglitazone was co-administrated with the PPAR-γ inhibitor GW9662 at 5 μM and 10 μM (Figure 2a). HUVEC treated with pioglitazone displayed increased expression of PPAR-γ protein and a down-regulation of nuclear factor kappa-light-chain-enhancer of activated B cells (NF-kB) and tumor necrosis factor α (TNF-α), both target genes of PPAR-γ and early regulators of the inflammatory process (Figure S1a,b). Pioglitazone also partially inhibited HUVEC migration potential, as shown by scratch wound assay performed through an Incucyte S3 assay (Figure 2b; see methods for experimental details). Data were analyzed at 6 h and 12 h post-wounding, showing that HUVEC were able to almost completely close the wound as evidenced by wound width reduction and wound confluence increase (Figure 2c,d). Pioglitazone significantly slowed down the wound closure process, whereas the GW9662 addition at 5 μM concentration contributed to reactivation of the migration features of HUVEC, coherently with wound width and wound confluence values (Figure 2c,d).
Similarly to HUVEC, pioglitazone affected HAOSMC proliferation by reducing cell growth, especially after 72 h administration ($20\%$ growth decrease; Figure 3a). PPAR-γ inhibition with GW9662 mitigated the effect of pioglitazone at each time point, more significantly at 10 μM. The analysis of cell migration evidenced a reducing effect mediated by pioglitazone after 15 h and 24 h from initial scratch (Figure 3b). GW9662 antagonized the effect of pioglitazone and the wound healing capacity of HAOSMC was restored (mainly after 15 h with GW9662 at 5 μM and after 24 h with GW9662 at 10 μM) (Figure 3b). These results were supported by measures of wound width and wound confluence reported in Figure 3c,d.
## 2.4. Characterization of a Primary Cell Model Isolated from Failed AVFs
AVFCs T0 and AVFCs T1 displayed comparable morphology and immunophenotype, including the fibroblast-like morphology, the expression of CD44, α-smooth muscle actin (α-SMA), and the absence of endothelial antigen CD34, coherently with a mesenchymal stem cell (MSC)-like profile (Figure 4a,b). We did not observe significant variations of mesenchymal marker expression between T0 and T1. Conversely, a lower expression of α-SMA was observed in AVFCs T1 compared to AVFCs T0 (Figure 4b), suggesting the phenotype transition is occurring contextually to the IH development. In addition, the analysis of cell proliferation also highlighted divergence between AVFCs T0 and T1 cells. Indeed, AVFCs T1 exhibited higher growth rate than AVFCs T0, as supported by crystal violet stain (Figure 4c) and count of Ki-67 positive cells (Figure 4d).
To further characterize the AVFC model with respect to the pathogenic processes typically associated with the IH development, we proceeded with the analysis of genes involved in vascular remodeling, cell migration, and invasion. AVFCs T1 displayed increased expression of matrix metalloproteinase 9 (MMP-9), snail transcription factor 1 (SLUG), and VIMENTIN than AVFCs T0 (Figure 4e–g).
Altogether, these data were coherent with the unbalanced proliferative and migratory features of cells involved in IH pathogenesis.
## 2.5. Pioglitazone Mitigated IH Pathogenic Mechanisms in AVF T0 and T1 Cell Model
Preliminary results showed a lower expression of PPAR-γ in AVFCs T1 than T0, suggesting a possible regulatory role during IH pathogenesis. In order to explore whether PPAR-γ stimulation could mitigate IH mechanisms, both AVFCs T0 and AVFCs T1 were exposed to pioglitazone at increasing concentrations (0–5–10–20–50 μM). Pioglitazone affected AVFCs proliferation in a dose- and time-dependent manner, especially in AVFCs T0, coherently with the unbalanced proliferative phenotype in AVFCs T1 cells (Figure 5a). However, we did not observe detrimental effects on cell viability in the presence of low drug concentrations (5–10 μM) for 24–48 h. In particular, we found a significant growth decrease in cells treated with 10 μM pioglitazone for 48 h ($49.1\%$ ± 0.225 in AVFCs T1, $p \leq 0.01$; $31\%$ ± 0.036 in AVFCs T0, $p \leq 0.0001$; two-way ordinary Anova test). More drastic effects on cell viability were detected at higher pioglitazone doses and at prolonged exposures (20–50 μM), especially in AVFCs T1 (Figure 5b). The analysis of the cell metabolic activity performed by MTT assay supported the proliferation results. In AVFCs T0, the resulting amount of cell viability and metabolic activity was reduced by $25\%$ and 18 % after 24 h and 48 h, respectively (Figure 5c). In AVFCs T1, the metabolic activity was not drastically affected at 10 μM for 24 h and 48 h (Figure 5d). Based on these data, the pioglitazone administration at 10 μM for 48 h resulted in more effective experimental conditions to modulate cell proliferation without excessive cytotoxic effects.
In addition, pioglitazone affected the cell migration process in AVFCs T0 and AVFCs T1, as detected by the wound healing assay performed by a manual scratch test (Figure S2a).
Next, we explored whether pioglitazone at the selected concentration was able to stimulate PPAR-γ expression in AVFCs T1. It was found that a significant up-regulation of PPAR-γ mRNA and protein in AVFCs T1 treated with pioglitazone at 10 μM for 48 h (Figure 6a,b) and in AVFCs T0 (Figure S2b). These data showed that AVFCs T1 resulted in being more sensitive to PPAR-γ modulation by pioglitazone, confirming PPAR-γ as a potential target for regulating the increased proliferative process occurring in IH. Further, coherently with the ameliorating effect on cell migration, pioglitazone also mediated the down-regulation of the main genes involved in cell migration/invasiveness SLUG, MMP-9, and VIMENTIN in AVFCs T1 (Figure 6c).
## 2.6. PPAR-γ Inhibition Reversed the Regulatory Effects of Pioglitazone in AVFCs T1 Proliferation and Migration
In order to confirm whether pioglitazone regulates IH pathogenesis via PPAR-γ stimulation, AVFCs T1 were exposed to pioglitazone/GW9662 combination. As shown in Figure 7, GW9662 reversed the effect of pioglitazone by reactivating the proliferative (Figure 7a) and migratory (Figure 7b) processes in AVFCs T1. An increasing trend of AVFCs T1 migration was observed with both GW9662 concentrations, particularly evident at 5 μM as highlighted by wound width (Figure 7c) and wound confluence (Figure 7d) measurements. These data supported the regulatory role of PPAR-γ on cell proliferation and migration mechanisms, which are crucial during IH pathogenesis.
## 3. Discussion
IH development is the leading cause of inadequate maturation and consequent failure of AVF, resulting in increased cardiovascular complications and repeated interventions. Vascular damage, shear stress, and surgical trauma contribute to IH initiation, whose cellular and molecular mechanisms involve several cell players and have not been fully elucidated yet. IH is a neointimal lesion typically characterized by unbalanced proliferation and migration processes of vascular cells, including ECs and SMCs. The regulation of cell proliferation and migration represents a possible strategy to prevent IH and AVF failure.
With the intent of identifying a possible regulatory strategy of the IH pathogenesis, the present study investigated the involvement of the nuclear hormone receptor PPAR-γ in AVF through its modulation by the pharmacological agonist pioglitazone. As reported in the literature, PPAR-γ performs anti-proliferative functions in addition to the well-known metabolic regulatory roles [19], indeed PPAR-γ over-expression was shown to inhibit vascular SMC proliferation and IH in mice through the inhibition of the Toll-like Receptor 4 (TLR4)-mediated inflammation [20]. Further, the PPAR-γ agonist rosiglitazone was able to modulate rat aortic VSMCs and to reduce IH after angioplasty in a rat carotid artery model [21]. The PPAR-γ agonist pioglitazone is an insulin-sensitizing drug with anti-fibrotic and anti-inflammatory effects, able to exert beneficial effects in many cardiovascular pathological events by contributing to the reduction of atherosclerotic plaque, neointimal formation, plaque inflammation, and to the repair of endothelial function, as reviewed in Nesti et al. [ 22]. Pioglitazone also preserves beta-cell function and improves the metabolic syndrome through several functions, i.e., by enhancing insulin sensitivity, lowering blood pressure, reducing triglycerides and increasing high-density lipoprotein (HDL) (reviewed in DeFronzo et al. [ 23]). Like rosiglitazone, pioglitazone was effective at reducing IH, by promoting a significant regression of IH in balloon-injured rat carotid artery [24] and by mitigating the IH formation in mouse injured femoral artery [25].
The protein and mRNA analysis performed in veins collected contextually to first surgical procedure for AVF (T0) and repeated intervention due to AVF failure (T1), revealed an altered expression of PPAR-γ in failed AVF. In particular, the histological findings highlighted a differential pattern of localization of PPAR-γ protein in AVF T1, as it was normally expressed in the tissue surrounding the lesion and negative within the neointimal formation. Based on this preliminary and clinical sketch of PPAR-γ distribution in AVF tissues, we aimed at exploring whether its pharmacological modulation in vitro was able to counteract IH pathogenic processes. Firstly, we observed that HUVEC and HAOSMC, representative of endothelial and smooth muscle cell models, were highly responsive to pioglitazone administration undergoing growth decrease and wound healing property impairments. Interestingly, the co-administration of inhibitor/agonist of PPAR-γ (GW9662/pioglitazone) reversed this trend by reactivating both cell proliferation and migration. According to the literature [26,27], these data supported the role of PPAR-γ as potential modulator of vascular remodeling and pathological alterations. We therefore extended the analysis to a primary cell model isolated from AVF patients included in a group homogenous for age, sex, and risk factors. This model consisted of cells characterized by a mesenchymal-like phenotype, given by adherence to plastic, typical elongated fibroblast shape, expression of CD44, and lack of CD34, the endothelial cell marker. Further, we detected differential expression of α-SMA protein, which resulted lower in AVFCs T1 and possibly suggestive of altered differentiation lineage during IH progression. Cells were also characterized according to functional properties, displaying a more prominent growth rate and ki-67 expression and marker of proliferation in AVFCs T1 in comparison to AVFCs T0. This result was consistent with a previous study of our group, which elucidated the increased positivity to ki-67 in AVF tissues belonging to the T1 group [28]. Additionally, AVFCs T1 exhibited high mRNA levels of SLUG, MMP-9 and VIMENTIN, genes typically involved in migration, invasion, vascular remodeling, and End-MT. End-MT is a crucial phenotype conversion occurring in ECs in both physiological and pathological states, and might be a source of key cell contributors to vascular pathology, and, considering the IH localization within the intimal layer, to AVF failure. In this regard, it was demonstrated that blocking Notch signaling, responsible for End-MT activation in ECs, prevented AVF failure [29].
PPAR-γ was lower in AVFCs T1 than AVFCs T0, reflecting the histological expression pattern. Pioglitazone was effective at modulating cell proliferation in both AVFCs T0 and T1, particularly in AVFCs T1 where the decrease of growth rate was associated with the up-regulation of PPAR-γ expression. Pioglitazone also reduced AVFCs T1 migration and relative genes levels (SLUG, MMP-9, VIMENTIN). MMP-9 also represents a crucial factor in AVF failure, as shown by a model of AVF mice where stenosis and inflammation were reduced following MMP-9 knockout [30].
The exposure of AVFCs T1 to pioglitazone/GW9662 counteracted both the proliferation and migration rates observed in presence of pioglitazone alone. These results support the pivotal role of PPAR-γ in the resolution of IH pathogenic mechanisms in vitro, paving the way to further investigations aimed at identifying novel PPAR-γ target genes to be proposed in the prevention of IH and AVF failure.
Future studies will be necessary to unveil alternative approaches involving natural compounds [19] to stimulate PPAR-γ expression and activity, in order to overcome side effects due to drug therapy such as fluid retention and the risk for heart failure associated with pioglitazone administration [22]. To this end, a recent study showed the efficacy of Jujuboside B, a saponin extracted from *Zizyphus jujuba* var. spinosa, at modulating VSMCs proliferation and migration, via AMPK/PPAR-γ signaling [31]. Similarly, fisetin, a plant flavonoid polyphenol, was found to inhibit VSMC proliferation and migration, and to ameliorate IH following injury, by inducing the antioxidant enzyme paraoxonase (PON2) through PPAR-γ activation [32].
Ultimately, a noteworthy feature of the present study is the utilization of cell models isolated from patients subjected to AVF intervention, with the goal to reproduce the disease context in vitro and to characterize the cell populations that participate to IH pathogenesis. However, the high level of biological variability existing among different patients affects data reproducibility and points out the importance of properly evaluating the treatment schedule taking into account of patient specific characteristic for more personalized and effective approaches.
## 4.1. Study Design and Sample Collection
For the present study, vascular tissues were collected from end-stage kidney disease patients subjected to AVF surgical procedure, with the approval of the local Ethical Committee (protocol number $\frac{142}{2019}$/Sper/AOUBo). All samples were anonymous and treated according to the ethical guidelines of the 1975 Declaration of Helsinki and following revisions. After surgery, fresh vein tissues were both fixed with formalin for histological analysis and processed for cell isolation, as described below in this section. Tissues were distinguished as follows: [1] T0, normal veins taken when the first AVF was established, and [2] T1, failed AVF with IH.
## 4.2. Immunohistochemistry
PPAR-γ expression was detected in AVF T0 and AVF T1 tissues by immunohistochemistry using a non-biotin-amplified method (Novolink, Leica Biosystems, Wetzlar, Germany). To this aim, 2 μm thick sections of formalin-fixed and paraffin-embedded tissues (FFPE) were deparaffinized and rehydrated through a series of graded ethanol and rinsed in distilled water. Inactivation of endogenous peroxidase activity was performed with a $3\%$ H2O2 in absolute methanol solution for 10 min (min) at room temperature (rt), antigen retrieval was performed using citrate buffer (pH 6) in microwave, and after cooling, slides were washed with Tris Buffered Saline (TBS). Sections were incubated with PPAR-γ primary antibody (1:100, clone C26H12, Cell Signaling Technology, Danvers, MA, USA) in a moist chamber at 4 °C over/night (o/n), followed by NovoLink Polymer for 30 min at rt and finally exposed to the substrate/chromogen 3,3′-diaminobenzidine (DAB) Novocastra DAB Chromogen and NovoLink DAB buffer. Nuclei were counterstained with Mayer’s hematoxylin. Samples were dehydrated, cover slipped, and observed under a light microscope using the Image Pro Plus program. The quantification of PPAR-γ was performed on digitalized images randomly acquired at 10× magnification by using the Image J (version 1.53) deconvolution color tool. Results were expressed as mean of PPAR-γ-positive areas.
## 4.3. Cell Cultures and Treatments
Cell models used in this study were the following: human endothelial umbilical vein cells (HUVEC, Lonza, Basel, Switzerland), human aortic smooth muscle cells (HAOSMC, Promocell, Heidelberg, Germany), and primary cells isolated from veins. HUVEC were cultured in Dulbecco’s Modified Eagle’s Medium (DMEM) enriched with $10\%$ fetal bovine serum (FBS) and $1\%$ antibiotics (Euroclone, Milano, Italy). HAOSMC were cultured with HAOSMC growth medium (Promocell, Heidelberg, Germany). Primary cells were isolated from vascular tissue (T0 and T1) collected during AVF surgical procedure. The cell isolation procedure was performed through organ culture. To this scope, at time of tissue collection, AVF tissues were rinsed with Phosphate Buffer Saline (PBS, Merck Group, Darmstadt, Germany) and then placed in 12-well plates with 2 mL of DMEM enriched with $20\%$ FBS and $1\%$ antibiotics. Growth medium was freshly replaced twice a week and culture growth was observed daily under the inverted microscope. Tissue cultures were kept in incubator at 37 °C, $5\%$ CO2 for two weeks, when cells started to grown from explant. After tissues were discarded, cells were cultured for further 10 days, when $70\%$ confluence was reached. AVFCs were used at passages 3–10, for analysis of immunophenotype, proliferation, migration, and gene and protein expression.
Pioglitazone hydrochloride (E6910; Merck Group, Darmstadt, Germany) and GW9662 (M6191; Merck Group, Darmstadt, Germany) were used as pharmacological agonist and irreversible inhibitor of PPAR-γ, respectively. For PPAR-γ inhibition, a preliminary exposure to GW9662 was performed for 30 min, followed by combination with pioglitazone. Experimental scheme (concentration/exposure time) was optimized according to viability assay for each cell model. Treatments were administrated with complete growth medium, and controls were grown with dymethyl sulfoxide (DMSO; Merck Group, Darmstadt, Germany), which was used as vehicle solvent for compound preparations.
## 4.4. Immunofluorescence
The analysis of AVF immunophenotype was performed through immunofluorescence. Cells were fixed with $2\%$ paraformaldehyde (membrane antigens) for 4 min at rt. Then, cells were permeabilized with Triton X-100 (Merck Group, Darmstadt, Germany) at $1\%$ in PBS for additional 10 min at rt for cytoplasmic and nuclear antigen detection. Incubation with bovine serum albumin (BSA, Merck Group, Darmstadt, Germany) $1\%$ in PBS was assessed for 30 min at rt for the blocking of non-specific binding sites. Cells were incubated with primary antibodies (CD34 1:80, Dako, Santa Clara, CA, USA; CD44 1:100, BD Biosciences Pharmingen, Franknlin Lakes, NJ, USA; α-SMA 1:100, Sigma Aldrich, St Louis, MO, USA; Ki-67 1:100, Novocastra, Leica Biosystems, Wetzlar, Germany) for 1 h at 37 °C. Samples were then washed with PBS and incubated with anti-mouse Alexa Fluor 488 and anti-rabbit Alexa Fluor 546 (ThermoFisher Scientific, Carlsbad, CA, USA) secondary antibodies in $1\%$ BSA/PBS for 1 h at 37 °C in the dark. After washing with PBS, nuclei were counterstained with DAPI (4′, 6-diamidino-2-phenylindole; Thermo Fisher Scientific, Carlsbad, CA, USA). Images were acquired by a Leica DMI4000 B inverted fluorescence microscope (Leica Microsystems, Milan, Italy). Quantification of Ki-67 positive cells was performed on digitalized images randomly acquired at 40× magnification, and a minimum of 5 fields was examined for each sample. Results were expressed as percentages of nuclei positive to the target protein and expressed as percentage of positive cells/total cells.
## 4.5. Cell Growth and Proliferation Assay
The analysis of cell proliferation was performed through crystal violet stain and MTT assay.
Crystal violet stain was assessed in HUVEC, HAOSMC, AVFCs T0, and AVFCs T1 for testing the effect of pioglitazone, alone and in combination with GW9662, on cell proliferation. Cells were seeded in a 96-well plate in triplicate in complete growth medium at different density seeding for each cell model (HUVEC 5 × 103/well; HAOSMC 10 × 104/well; AVFCs T0, T1: 10 × 104/well). After 24 h, cells were fixed with formalin for 10 min at rt, washed with PBS, and stained with crystal violet for 20 min. Crystal violet excess was removed by four washes with distilled water, air dried, and solubilized with $10\%$ acetic acid. The quantification of cell viability was performed by absorbance analysis at 592 nm optical density (OD) by Spark multimode microplate reader (Tecan, Zurich, Switzerland).
For metabolic activity analysis following pioglitazone administration, MTT assay (Vybrant MTT Cell Proliferation Assay Kit, Thermo Fisher Scientific, Carlsbad, CA, USA) was performed in AVFCs T0 and AVFCs T1, following the manufacturer’s instructions. Cells were seeded in a 96-well plate in triplicate at a density of 104 cells/well in 100 μL complete growth medium. After 24 h, cells were treated with pioglitazone (0–5–10–20 μM) for 24–48–72 h. Then, cell medium was replaced with fresh growth medium and 10 μL of 12 mM MTT component A (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) were added to each well and left in incubator for 4 h. Then, 100 μL of MTT component B sodium dodecyl sulfate (SDS)-hydrochloride acid (HCl) were added for further 18 h at 37 °C. Absorbance was analyzed at 570 nm OD by a Spark multimode microplate reader (Tecan, Zurich, Switzerland).
## 4.6. In Vitro Migration Assay
The analysis of cell migration in HUVEC, HAOSMC, AVFCs T0, and AVFCs T1 under pioglitazone/GW9662 treatments was performed through a scratch wound assay. For manual scratch assay, 1 × 105 AVFCs T0 and T1 were seeded in a 24-well plate. After 24 h, the cell monolayer was wounded with a sterile p200 pipette tip, washed with PBS, and treated with pioglitazone at 10 μM in DMEM $10\%$ FBS. Cell migration was monitored under the light microscope until 48 h, when cells were fixed with formalin at rt, washed with PBS, stained with $0.1\%$ Crystal Violet in $25\%$ methanol for 25 min, and air-dried. Images were taken with a digital camera (Nikon). For automated scratch assay, cells (HUVEC, HAOSMC, AVFCs T1) were seeded in an Incucyte ImageLock 96-well plate (Essen Bioscience, Ann Harbor, MI, USA) and treated with pioglitazone alone or in combination with GW9662 according to the experimental design. The scratch was performed by using the WoundMaker (Essen BioScience, Ann Harbor, MI, USA) device. After a PBS wash, cell treatments were replenished and the plates were placed in an IncuCyte S3 instrument (Essen BioScience, Ann Harbor, MI, USA) equipped with a dedicated incubator. Then, each wound image per well was automatically recorded with a 10× objective lens every three h for 48 h using the IncuCyte S3/SX1 optical module phase contrast. Images were processed by using the IncuCyte 2022B software to analyze, over time, the wound width that is the distance between the migrating edges of the wound and measured in micrometers (μm), and the wound confluence, which represents the percentage (%) of the wound region occupied by cells.
## 4.7. Gene Expression Analysis
Total RNA was extracted from AVF T0 and AVF T1 through PureZOLTM RNA isolation reagent (BioRad Laboratories, Hercules, CA, USA), according to the manufacturer’s instructions. Reverse transcription was performed from 1 µg of total RNA in 20 µL reaction volume using iScriptTM cDNA synthesis kit (BioRad Laboratories, Hercules, CA, USA). Real-time PCR was carried out in a CFX Connect real-time PCR Detection System (BioRad Laboratories, Hercules, CA, USA) using the SYBR green mix (Sso AdvancedTM Universal Sybr Green Supermix; BioRad Laborato-ries, Hercules, CA, USA) and primers sequences were designed using the NCBI BLAST tool and purchased from Merck Group, Darmstadt, Germany (Table 1). Each assay was performed in triplicate and each target gene expression was normalized to glyceraldehyde 3-phospate dehydrogenase (GAPDH). Gene expression levels were determined by the comparative 2−ΔΔCt method and expressed as fold changes relative to controls [33].
## 4.8. Western Blot
Total cellular proteins were extracted from AVFCs T1 treated with pioglitazone (10 μM) for 48 h using a lysis buffer (0.1 M KH2PO4, pH 7.5, $1\%$ NP-40, 0.1 mM β-glycerolphosphate, supplemented with protease inhibitor cocktail; Sigma-Aldrich, St Louis, MO, USA) and quantified through the Bio-Rad Protein Assay (BioRad Laboratories, CA, USA) at the spectrophotometer. Thirty micrograms of proteins were separated on $10\%$ polyacrylamide gel by SDS-PAGE (TGX Stain-Free™ FastCast™ Acrylamide Solutions; BioRad Laboratories, Hercules, CA, USA). Proteins were transferred to a nitrocellulose membrane (GE Healthcare Life Sciences, Chicago, IL, USA), blocked with $5\%$ non-fat dry milk in TBS-Tween for 1 h at rt, and incubated with the primary antibodies PPAR-γ (1:1000; clone C26H12, Cell Signaling Technology, Danvers, MA, USA), β-actin (1:4000; clone AC-74, Sigma-Aldrich, St. Louis, MO, USA) at 4 °C o/n. Incubation with secondary antibody human anti-rabbit/mouse horseradish peroxidase-conjugated (GE Healthcare, Chicago, IL, USA) was performed for 1 h at rt. The protein signal was detected using Westar ηC chemiluminescent substrate (Cyanagen, Bologna, Italy). Membrane imaging and densitometric analysis were performed at the ChemiDoc XRS+ Imaging System (BioRad Laboratories, Hercules, CA, USA) using ImageLab v5.1.1 (BioRad Laboratories, Hercules, CA, USA).
## 4.9. Statistical Analysis
For each experiment, at least three biological and technical replicates were performed. Data were expressed as mean ± standard deviation (SD) or standard error of mean (SEM). Data analysis and graphs were developed with GraphPad Prism 6 statistical software. Statistical analysis was performed through unpaired and paired Student’s t-test for comparison between two groups, whereas ordinary one-way and two-way analyses of variance (ANOVAs) followed by Tukey’s test were applied for multiple comparisons. Results were considered statistically significant at the $95\%$ confidence level ($p \leq 0.05$).
## 5. Conclusions
In the present study, PPAR-γ induction by pioglitazone was performed in three different cell models, represented by HUVEC, HAOSMC, and patient AVFCs. The obtained data were coherent at supporting the PPAR-γ induction strategy as effective at inhibiting cell proliferation and migration, key mechanisms in IH development. These results can be preliminary to future studies addressed at investigating the PPAR-γ downstream signaling pathway and at searching for novel agonists to prevent IH and AVF failure.
## References
1. Hu H., Patel S., Hanisch J.J., Santana J.M., Hashimoto T., Bai H., Kudze T., Foster T.R., Guo J., Yatsula B.. **Future Research Directions to Improve Fistula Maturation and Reduce Access Failure**. *Semin. Vasc. Surg.* (2016.0) **29** 153-171. DOI: 10.1053/j.semvascsurg.2016.08.005
2. Roy-Chaudhury P., Spergel L.M., Besarab A., Asif A., Ravani P.. **Biology of Arteriovenous Fistula Failure**. *J. Nephrol.* (2007.0) **20** 150-163. PMID: 17514619
3. Roy-Chaudhury P., Arend L., Zhang J., Krishnamoorthy M., Wang Y., Banerjee R., Samaha A., Munda R.. **Neointimal Hyperplasia in Early Arteriovenous Fistula Failure**. *Am. J. Kidney Dis.* (2007.0) **50** 782-790. DOI: 10.1053/j.ajkd.2007.07.019
4. Yap Y.-S., Chi W.-C., Lin C.-H., Liu Y.-C., Wu Y.-W.. **Association of Early Failure of Arteriovenous Fistula with Mortality in Hemodialysis Patients**. *Sci. Rep.* (2021.0) **11** 5699. DOI: 10.1038/s41598-021-85267-6
5. Nantakool S., Srisuwan T., Reanpang T., Rerkasem K., Prasannarong M.. **A Randomized Controlled Trial of the Effect of Postoperative Hand Exercise Training on Arteriovenous Fistula Maturation in Patients with Chronic Kidney Disease**. *J. Vasc. Surg.* (2022.0) **75** 230-237. DOI: 10.1016/j.jvs.2021.07.124
6. Ma S., Duan S., Liu Y., Wang H.. **Intimal Hyperplasia of Arteriovenous Fistula**. *Ann. Vasc. Surg.* (2022.0) **85** 444-453. DOI: 10.1016/j.avsg.2022.04.030
7. Ando J., Yamamoto K.. **Effects of Shear Stress and Stretch on Endothelial Function**. *Antioxid. Redox Signal.* (2011.0) **15** 1389-1403. DOI: 10.1089/ars.2010.3361
8. Carracedo M., Artiach G., Arnardottir H., Bäck M.. **The Resolution of Inflammation through Omega-3 Fatty Acids in Atherosclerosis, Intimal Hyperplasia, and Vascular Calcification**. *Semin. Immunopathol.* (2019.0) **41** 757-766. DOI: 10.1007/s00281-019-00767-y
9. Tian D., Zeng X., Wang W., Wang Z., Zhang Y., Wang Y.. **Protective Effect of Rapamycin on Endothelial-to-Mesenchymal Transition in HUVECs through the Notch Signaling Pathway**. *Vasc. Pharmacol.* (2019.0) **113** 20-26. DOI: 10.1016/j.vph.2018.10.004
10. Liang A., Wang Y., Han G., Truong L., Cheng J.. **Chronic Kidney Disease Accelerates Endothelial Barrier Dysfunction in a Mouse Model of an Arteriovenous Fistula**. *Am. J. Physiol. Renal. Physiol.* (2013.0) **304** F1413-F1420. DOI: 10.1152/ajprenal.00585.2012
11. Evrard S.M., Lecce L., Michelis K.C., Nomura-Kitabayashi A., Pandey G., Purushothaman K.-R., d’Escamard V., Li J.R., Hadri L., Fujitani K.. **Endothelial to Mesenchymal Transition Is Common in Atherosclerotic Lesions and Is Associated with Plaque Instability**. *Nat. Commun.* (2016.0) **7** 11853. DOI: 10.1038/ncomms11853
12. Ciavarella C., Motta I., Vasuri F., Fittipaldi S., Valente S., Pollutri D., Ricci F., Gargiulo M., Pasquinelli G.. **Involvement of MiR-30a-5p and MiR-30d in Endothelial to Mesenchymal Transition and Early Osteogenic Commitment under Inflammatory Stress in HUVEC**. *Biomolecules* (2021.0) **11**. DOI: 10.3390/biom11020226
13. Hao Y.-M., Yuan H.-Q., Ren Z., Qu S.-L., Liu L.-S., Wei D.-H., Yin K., Fu M., Jiang Z.-S.. **Endothelial to Mesenchymal Transition in Atherosclerotic Vascular Remodeling**. *Clin. Chim. Acta* (2019.0) **490** 34-38. DOI: 10.1016/j.cca.2018.12.018
14. Moonen J.-R.A.J., Lee E.S., Schmidt M., Maleszewska M., Koerts J.A., Brouwer L.A., van Kooten T.G., van Luyn M.J.A., Zeebregts C.J., Krenning G.. **Endothelial-to-Mesenchymal Transition Contributes to Fibro-Proliferative Vascular Disease and Is Modulated by Fluid Shear Stress**. *Cardiovasc. Res.* (2015.0) **108** 377-386. DOI: 10.1093/cvr/cvv175
15. Rosen E.D., Spiegelman B.M.. **PPARgamma: A Nuclear Regulator of Metabolism, Differentiation, and Cell Growth**. *J. Biol. Chem.* (2001.0) **276** 37731-37734. DOI: 10.1074/jbc.R100034200
16. Rosen E.D., Hsu C.-H., Wang X., Sakai S., Freeman M.W., Gonzalez F.J., Spiegelman B.M.. **C/EBPα Induces Adipogenesis through PPARγ: A Unified Pathway**. *Genes Dev.* (2002.0) **16** 22-26. DOI: 10.1101/gad.948702
17. Wu Z., Xie Y., Morrison R.F., Bucher N.L., Farmer S.R.. **PPARgamma Induces the Insulin-Dependent Glucose Transporter GLUT4 in the Absence of C/EBPalpha during the Conversion of 3T3 Fibroblasts into Adipocytes**. *J. Clin. Investig.* (1998.0) **101** 22-32. DOI: 10.1172/JCI1244
18. Tan M.H.. **Current Treatment of Insulin Resistance in Type 2 Diabetes Mellitus**. *Int. J. Clin. Pract. Suppl.* (2000.0) **113** 54-62
19. Ciavarella C., Motta I., Valente S., Pasquinelli G.. **Pharmacological (or Synthetic) and Nutritional Agonists of PPAR-γ as Candidates for Cytokine Storm Modulation in COVID-19 Disease**. *Molecules* (2020.0) **25**. DOI: 10.3390/molecules25092076
20. Zhang L.-L., Gao C.-Y., Fang C.-Q., Wang Y.-J., Gao D., Yao G.-E., Xiang J., Wang J.-Z., Li J.-C.. **PPARγ Attenuates Intimal Hyperplasia by Inhibiting TLR4-Mediated Inflammation in Vascular Smooth Muscle Cells**. *Cardiovasc. Res.* (2011.0) **92** 484-493. DOI: 10.1093/cvr/cvr238
21. Yang H.-M., Kim B.-K., Kim J.-Y., Kwon Y.-W., Jin S., Lee J.-E., Cho H.-J., Lee H.-Y., Kang H.-J., Oh B.-H.. **PPARγ Modulates Vascular Smooth Muscle Cell Phenotype via a Protein Kinase G-Dependent Pathway and Reduces Neointimal Hyperplasia after Vascular Injury**. *Exp. Mol. Med.* (2013.0) **45** e65. DOI: 10.1038/emm.2013.112
22. Nesti L., Tricò D., Mengozzi A., Natali A.. **Rethinking Pioglitazone as a Cardioprotective Agent: A New Perspective on an Overlooked Drug**. *Cardiovasc. Diabetol.* (2021.0) **20** 109. DOI: 10.1186/s12933-021-01294-7
23. DeFronzo R.A., Inzucchi S., Abdul-Ghani M., Nissen S.E.. **Pioglitazone: The Forgotten, Cost-Effective Cardioprotective Drug for Type 2 Diabetes**. *Diabetes Vasc. Dis. Res.* (2019.0) **16** 133-143. DOI: 10.1177/1479164118825376
24. **Pioglitazone Enhances Cytokine-Induced Apoptosis in Vascular Smooth Muscle Cells and Reduces Intimal Hyperplasia**
25. Osman I., Fairaq A., Segar L.. **Pioglitazone Attenuates Injury-Induced Neointima Formation in Mouse Femoral Artery Partially through the Activation of AMP-Activated Protein Kinase**. *Pharmacology* (2017.0) **100** 64-73. DOI: 10.1159/000471769
26. Zhou Y., Zhang M.-J., Li B.-H., Chen L., Pi Y., Yin Y.-W., Long C.-Y., Wang X., Sun M.-J., Chen X.. **PPARγ Inhibits VSMC Proliferation and Migration via Attenuating Oxidative Stress through Upregulating UCP2**. *PLoS ONE* (2016.0) **11**. DOI: 10.1371/journal.pone.0154720
27. Spigoni V., Picconi A., Cito M., Ridolfi V., Bonomini S., Casali C., Zavaroni I., Gnudi L., Metra M., Cas A.D.. **Pioglitazone Improves In Vitro Viability and Function of Endothelial Progenitor Cells from Individuals with Impaired Glucose Tolerance**. *PLoS ONE* (2012.0) **7**. DOI: 10.1371/journal.pone.0048283
28. Mauro R., Rocchi C., Vasuri F., Pini A., Croci Chiocchini A.L., Ciavarella C., La Manna G., Pasquinelli G., Faggioli G., Gargiulo M.. **Tissue Ki67 Proliferative Index Expression and Pathological Changes in Hemodialysis Arteriovenous Fistulae: Preliminary Single-Center Results**. *J. Vasc. Access* (2021.0) **24** 11297298211015496. DOI: 10.1177/11297298211015495
29. Wang Y., Liang A., Luo J., Liang M., Han G., Mitch W.E., Cheng J.. **Blocking Notch in Endothelial Cells Prevents Arteriovenous Fistula Failure Despite CKD**. *J. Am. Soc. Nephrol.* (2014.0) **25** 773-783. DOI: 10.1681/ASN.2013050490
30. Shih Y.-C., Chen P.-Y., Ko T.-M., Huang P.-H., Ma H., Tarng D.-C.. **MMP-9 Deletion Attenuates Arteriovenous Fistula Neointima through Reduced Perioperative Vascular Inflammation**. *Int. J. Mol. Sci.* (2021.0) **22**. DOI: 10.3390/ijms22115448
31. Ji Z., Li J., Wang J.. **Jujuboside B Inhibits Neointimal Hyperplasia and Prevents Vascular Smooth Muscle Cell Dedifferentiation, Proliferation, and Migration via Activation of AMPK/PPAR-γ Signaling**. *Front. Pharmacol.* (2021.0) **12** 1360. DOI: 10.3389/fphar.2021.672150
32. Pei F., Pei H., Su C., Du L., Wang J., Xie F., Yin Q., Gao Z.. **Fisetin Alleviates Neointimal Hyperplasia via PPARγ/PON2 Antioxidative Pathway in SHR Rat Artery Injury Model**. *Oxidative Med. Cell. Longev.* (2021.0) **2021** 6625517. DOI: 10.1155/2021/6625517
33. Livak K.J., Schmittgen T.D.. **Analysis of Relative Gene Expression Data Using Real-Time Quantitative PCR and the 2(-Delta Delta C(T)) Method**. *Methods* (2001.0) **25** 402-408. DOI: 10.1006/meth.2001.1262
|
---
title: Oral Administration of Lipopolysaccharide Enhances Insulin Signaling-Related
Factors in the KK/Ay Mouse Model of Type 2 Diabetes Mellitus
authors:
- Kazushi Yamamoto
- Masashi Yamashita
- Masataka Oda
- Vindy Tjendana Tjhin
- Hiroyuki Inagawa
- Gen-Ichiro Soma
journal: International Journal of Molecular Sciences
year: 2023
pmcid: PMC10003108
doi: 10.3390/ijms24054619
license: CC BY 4.0
---
# Oral Administration of Lipopolysaccharide Enhances Insulin Signaling-Related Factors in the KK/Ay Mouse Model of Type 2 Diabetes Mellitus
## Abstract
Lipopolysaccharide (LPS), an endotoxin, induces systemic inflammation by injection and is thought to be a causative agent of chronic inflammatory diseases, including type 2 diabetes mellitus (T2DM). However, our previous studies found that oral LPS administration does not exacerbate T2DM conditions in KK/Ay mice, which is the opposite of the response from LPS injection. Therefore, this study aims to confirm that oral LPS administration does not aggravate T2DM and to investigate the possible mechanisms. In this study, KK/Ay mice with T2DM were orally administered LPS (1 mg/kg BW/day) for 8 weeks, and blood glucose parameters before and after oral administration were compared. Abnormal glucose tolerance, insulin resistance progression, and progression of T2DM symptoms were suppressed by oral LPS administration. Furthermore, the expressions of factors involved in insulin signaling, such as insulin receptor, insulin receptor substrate 1, thymoma viral proto-oncogene, and glucose transporter type 4, were upregulated in the adipose tissues of KK/Ay mice, where this effect was observed. For the first time, oral LPS administration induces the expression of adiponectin in adipose tissues, which is involved in the increased expression of these molecules. Briefly, oral LPS administration may prevent T2DM by inducing an increase in the expressions of insulin signaling-related factors based on adiponectin production in adipose tissues.
## 1. Introduction
Lipopolysaccharide (LPS) is a glycolipid found in the outer membrane of Gram-negative bacteria. When LPS is injected into the body, it binds to Toll-like receptors (TLR4) in vivo, even at very low doses. This binding leads to the production of pro-inflammatory cytokines and, in turn, causes a strong systemic inflammatory response. This condition is called a cytokine storm state, which can induce shock symptoms such as fever and diarrhea and may lead to death [1,2,3,4]. Since the discovery of LPS, it has been called an endotoxin [5] and has been used as an inflammation-inducing substance for stimulating immune cells in vitro and inflammation models in animals (in vivo) by injection. Moreover, bacterial translocation has been observed in persistent inflammatory lesions in the intestinal tract and periodontal tissues, and persistent invasion of bacteria and LPS in living organisms induces systemic inflammation [6]. Based on these findings, LPS is considered a cause of chronic inflammatory diseases, including lifestyle-related diseases [7]. Specifically, many studies have reported that LPS is a causative agent of type 2 diabetes mellitus, one of the major lifestyle diseases.
Type 2 diabetes mellitus is a chronic inflammatory disease of adipose tissues and is characterized by impaired glucose tolerance and insulin resistance. Several studies have suggested that LPS induces the onset of type 2 diabetes mellitus. For example, LPS injection was reported to decrease the protein and mRNA expression of Glucose transporter type 4 (Glut 4), the primary transporter for glucose uptake in adipose tissues, and induces symptoms of type 2 diabetes mellitus such as increased fasting blood glucose, impaired glucose tolerance, and insulin resistance [1,2,3,4]. However, in these reports, LPS was injected intraperitoneally or intravenously to induce pathological models of systemic inflammation deliberately. LPS exists within healthy individuals, yet inflammation was not induced [8].
A hundred trillion bacteria reside in the mucosa of animals, and approximately half of those are Gram-negative bacteria that contain LPS on their cell wall. Thus, LPS is permanently present in the mucosa and should not be considered toxic to living beings. Indeed, our group found that oral LPS administration does not induce inflammation, unlike its injections [9,10]. Furthermore, oral LPS administration was found to suppress inflammation induced by a high-fat diet, prevent dementia, and suppress atherosclerosis in senescence-accelerated mice (SAM-P8) and ApoE-deficient atherosclerotic mice fed a high-fat diet [11,12]. These findings suggest that the physiological role of LPS differs from that of inflammation exacerbation by injection. Therefore, we propose that orally administered LPS should not be viewed as an endotoxin that induces inflammation but as a potentially beneficial substance that merits further studies.
Hence, this study aimed to establish this proposition by investigating the effect of oral LPS administration on type 2 diabetes mellitus using the KK-Ay mouse model, which is the standard animal model of type 2 diabetes mellitus [13,14]. As a result, we found that oral LPS administration improved fasting blood glucose levels and glucose tolerance index (HOMA-IR) without exacerbating inflammatory markers in KK/Ay mice with diabetic symptoms. In addition, for the first time, we found that adiponectin, an adipokine important for regulating glucose metabolic function, is induced in adipose tissues.
## 2.1. Verification of the Development of Type 2 Diabetes Mellitus in KK/Ay Mice
To investigate the effect of oral LPS administration on type 2 diabetes mellitus, KK/Ay mice, a standard model for this analysis, were used because they exhibit impaired glucose tolerance and insulin resistance. Initially, an oral glucose tolerance test (OGTT) and measurement of blood glucose parameters were performed to confirm the condition of KK/Ay mice and compared it with the condition of non-type 2 diabetes mellitus model C57BL/6 mice.
The OGTT results show that blood glucose levels were significantly increased in KK/Ay mice compared with those in C57BL/6 mice at 0, 30, 60, 120, and 180 min after oral glucose administration (Figure 1a). The area under the curve (AUC) of the OGTT was also significantly increased in KK/Ay mice compared with that in C57BL/6 mice (Figure 1b). In addition, the blood glucose parameters, such as fasting blood glucose level and hemoglobin A1C (HbA1c) test results, were significantly increased in KK/Ay mice compared with those in C57BL/6 mice (Figure 1c,d). Furthermore, blood insulin and HOMA-IR, an indicator of insulin resistance, were significantly increased in KK/Ay mice compared with those in C57BL/6 mice (Figure 1e,f).The abnormal glucose tolerance and insulin resistance results from these tests confirmed that KK/Ay mice had type 2 diabetes mellitus at the start of the study.
## 2.2. Oral LPS Administration Suppresses Type 2 Diabetes Mellitus in KK/Ay Mice
KK/Ay mice with type 2 diabetes mellitus were divided into two groups: the LPS (−) group received distilled water, and the LPS (+) group received distilled water containing LPS. After 8 weeks, the effects of oral LPS administration on insulin resistance and glucose intolerance in KK/Ay mice were examined by performing an OGTT and measuring blood glucose parameters. In the OGTT, blood glucose levels at 60, 120, and 180 min after oral glucose administration were significantly lower in the LPS (+) group than in the LPS (−) group (Figure 2a), and the AUC of the OGTT was also significantly lower in the LPS (+) group than in the LPS (−) group (Figure 2b). Regarding blood glucose parameters, the fasting blood glucose showed a decreasing trend in the LPS (+) group compared with that in the LPS (−) group, and HbA1c decreased significantly in the LPS (+) group when compared with that in the LPS (−) group (Figure 2c,d). In addition, blood insulin levels appear to decrease in the LPS (+) group compared with those in the LPS (−) group, and HOMA-IR, a marker of insulin resistance, decreased significantly in the LPS (+) group compared with that in the LPS (−) group (Figure 2e,f). These results suggest that oral LPS administration has an ameliorating or inhibitory effect on insulin resistance and glucose intolerance in type 2 diabetic KK/Ay mice.
Next, the OGTT and blood glucose parameters at the start (week 0) and end (week 8) of the study were compared to determine whether oral LPS administration had an ameliorating or inhibitory effect on type 2 diabetes mellitus. The AUC of the OGTT increased significantly in the LPS (−) group at the end of the study compared with that at the start of the study; however, no significant difference was found before and after the start of the study in the LPS (+) group (Figure 3a). Fasting blood glucose levels were not statistically different between the LPS (−) and LPS (+) groups; however, the fasting blood glucose level increased approximately 1.2-fold in the LPS (−) group, whereas it was almost unchanged in the LPS (+) group (Figure 3b). HbA1c increased significantly in both the LPS (−) and LPS (+) groups at the end of the study compared with that at the beginning of the study. However, the increase in HbA1c was approximately 1.6-fold in the LPS (−) group, whereas it was approximately 1.3-fold in the LPS (+) group, with the rate of increase also being lower in the LPS (+) group (Figure 3c). Blood insulin levels in both the LPS (−) and LPS (+) groups also increased significantly at the end of the study compared with that at the beginning of the study. The LPS (−) group had a 2.5-fold increase in blood insulin level, whereas the LPS (+) group had a 1.5-fold increase, with the rate of increase being lower in the LPS (+) group (Figure 3d). HOMA-IR increased significantly in the LPS (−) group at the end of the study compared with that at the beginning of the study, and no significant difference was observed in the LPS (+) group before and after the study (Figure 3e). These results indicate that oral LPS administration suppressed the progression of insulin resistance and glucose intolerance in type 2 diabetic KK/Ay mice, suggesting that oral LPS administration has a suppressive effect on type 2 diabetes mellitus.
## 2.3. Effects of Oral LPS Administration on Adipose Tissues
The analysis of body weight and adipose tissue changes in mice with and without LPS administration showed no significant changes in body weight, mesenteric adipose tissue weight, perirenal adipose tissue weight, peritesticular adipose tissue weight, or total adipose tissue weight (Figure 4a–e). On the contrary, the size of cells in adipose tissues was significantly reduced by oral LPS administration (Figure 4f,g).
## 2.4. Effects of Oral LPS Administration on the Expression Levels of Insulin Signaling-Related Factors in Adipose Tissues of KK/Ay Mice
The adipose tissue is one of the tissues that plays a crucial role in type 2 diabetes mellitus, specifically its insulin signaling. The elevated insulin signaling in adipose tissues is involved in suppressing the progression of type 2 diabetes mellitus. Therefore, we hypothesized that the expression levels of insulin signaling-related factors are upregulated in adipose tissues of KK/Ay mice, in which blood insulin resistance suppression and blood glucose intolerance by oral LPS administration were observed. The expression levels of insulin signaling-related factors in adipose tissues indicated that mRNA and protein expression levels of Glut4, which is involved in insulin-mediated glucose uptake, were much elevated in the LPS (+) group than in the LPS (−) group (Figure 5a,b). In addition, the expression level of insulin receptor (Ir), which is involved in insulin signaling to upregulate Glut4 expression, showed an increasing trend in the LPS (+) group compared with that in the LPS (−) group (Figure 5c). Moreover, the expression levels of thymoma viral proto-oncogene (Akt) and insulin receptor substrate 1 (Irs1) were significantly elevated in the LPS (+) group compared with those in the LPS (−) group (Figure 5d,e). These results suggest that the expression levels of insulin signaling-related factors in adipose tissues are upregulated in KK/Ay mice, in which glucose intolerance and insulin resistance were suppressed by oral LPS administration.
## 2.5. Adiponectin Expression in Adipose Tissues of KK/Ay Mice by Oral LPS Administration
The induction of the expression levels of insulin signaling-related factors in adipose tissues is thought to involve adiponectin, a cytokine synthesized uniquely by adipocytes. Therefore, we hypothesized that when the expression levels of insulin signaling-related factors in adipose tissues were increased by oral LPS administration, adiponectin expression in adipose tissues would be also induced.
The adiponectin gene expression and protein levels in adipose tissues were measured. The results show that adiponectin mRNA expression was significantly elevated in the LPS (+) group compared with that in the LPS (−) group (Figure 6a). In addition, the protein level of adiponectin was also significantly elevated in the LPS (+) group compared with that in the LPS (−) group (Figure 6b). Based on these results, we hypothesized that oral LPS administration induces adiponectin expression and upregulates the expression levels of insulin signaling-related factors in adipose tissues. Further investigations on adiponectin showed that the mRNA expression levels of adiponectin receptors (Adipor1 and Adipor2), which are located on the reaction pathway of adiponectin, were significantly increased in the LPS (+) group compared with that in the LPS (−) group (Figure 6c,d). Adiponectin directly induces increased Glut4 expression; however, whether it directly induces the expression levels of Ir, Irs1, and Akt2 is unclear. Therefore, to investigate whether adiponectin directly induces an increase in the expression levels of insulin signaling-related factors, we stimulated 3T3-L1 adipocytes in vitro with adiponectin (Figure 7a). The results showed that the mRNA expression of Glut4 tended to increase, and the expression levels of Ir, Irs1, and Akt2 mRNAs in 3T3-L1 adipocytes significantly increased upon adiponectin stimulation (Figure 7b–e). Thus, the upregulation of Glut4, Ir, Irs1, and Akt2 observed in the adipose tissues of KK/Ay mice orally treated with LPS may be due to adiponectin induction by LPS.
Based on these findings, we speculated that oral LPS administration induces the expression of adiponectin in adipose tissues and that adiponectin directly induces the upregulation of insulin signaling-related factors in adipose tissues by oral LPS administration.
## 3. Discussion
Enteral LPS is responsible for the onset of obesity-related type 2 diabetes mellitus [15,16,17,18]. This perception is based on the fact that overeating and high-fat diets increase the amount of enteral LPS transferred into the blood [18,19]. In this study, glucose intolerance and insulin resistance are induced in a mouse model receiving continuous infusion of LPS using a subcutaneously implanted infusion pump [16]. However, previous reports have also demonstrated that intestinal LPS is incorporated into chylomicrons and transferred into blood and that chylomicron LPS is barely involved in disease induction [20,21,22]. Based on the reports, evidence is insufficient to support the hypothesis that LPS is the cause of type 2 diabetes mellitus. In this study, oral LPS administration suppressed the onset of type 2 diabetes mellitus in KK/Ay mice by increasing the expression levels of insulin signaling-related factors in adipose tissues.
In our previous study, oral LPS administration to ApoE KO mice fed a high-fat diet and P8 (SAMP8), a mouse model of accelerated aging, reduced insulin resistance and AUC during a glucose tolerance test. This effect was observed at higher oral LPS doses (1 mg > 0.3 mg/kg/day) [11,12]. Based on these results, the LPS dose to be given orally in this study was set at 1 mg/kg/day.
KK/Ay mice were used as experimental animals in this study. These mice were developed by crossing KK mice with C57BL/6-Ay mice carrying the *Agouti* gene (Ay) [23,24]. They exhibit hyperglycemia, high HbA1c levels, insulin resistance, and impaired glucose tolerance, so they are widely used as a standard mouse model of type 2 diabetes mellitus to investigate substances that are beneficial for type 2 diabetes mellitus and analyze the mechanism of action of type 2 diabetes mellitus [13,14]. Therefore, we determined that the KK/Ay mouse model was appropriate for this study to clarify the effects of oral LPS administration on type 2 diabetes mellitus.
The initial state of KK/Ay mice was compared with that of naïve C57BL/6 mice, which were used as the non-type 2 diabetes mellitus model. At the beginning of the study, measurements of OGTT and blood parameters showed that KK/Ay mice had more advanced glucose tolerance and insulin resistance than C57BL/6 mice (Figure 1). Most of the studies using KK/Ay mice have concluded that KK/Ay mice had type 2 diabetes mellitus based on these results [25,26,27]. Therefore, we can say that KK/Ay mice had type 2 diabetes mellitus at the beginning of this study. The results of OGTT and measurements of blood parameters performed after 8 weeks of oral LPS administration were consistent with previous results from Apo E KO mice and SAMP8 [11,12]. Oral LPS administration led to a significant increase in the AUC of OGTT, fasting blood glucose, HbA1c level, and insulin resistance (Figure 2). These results suggest that oral LPS administration has an ameliorating or inhibitory effect on KK/Ay mice with type 2 diabetes mellitus. In this study, the effect of oral LPS administration on type 2 diabetes mellitus was further verified by comparing OGTT and blood parameters before and after oral LPS administration (Figure 3). The results were novel in that oral LPS administration has an inhibitory effect on type 2 diabetes mellitus. On the contrary, in a previous study, we reported that oral LPS administration to pre-diabetic (type 2) humans reduced fasting blood glucose and HbA1c levels compared with those before oral LPS administration, which are slightly different from the results of the present study [28]. However, in this previous study, LPS was orally administered in combination with Salacia tea, which has hypoglycemic effects. The oral LPS administration method was different from the method employed in the present study, in which only LPS was administered orally. This report also indicates that oral LPS administration enhanced the hypoglycemic effect of Salacia tea. Thus, LPS, when administered orally alone, has an inhibitory effect on type 2 diabetes mellitus; however, when taken in combination with other active ingredients or foods, it appeared to have an ameliorating effect on type 2 diabetes mellitus.
Many studies have shown that the adipose tissue is one of the key tissues involved in the suppression of type 2 diabetes mellitus [29,30,31,32,33,34,35,36]. The mRNA expression levels of insulin signaling-related factors such as Ir, Irs1, Akt, and Glut4 declined in adipose tissues in the case of type 2 diabetes mellitus [31,34,36]. Furthermore, mice with adipose tissue-specific knockout of Glut4 mRNA have abnormal glucose tolerance and insulin resistance, even though the expression levels of insulin signaling-related factors in other tissues were comparable to those of wild-type healthy mice [37]. A study reported that glucose intolerance was also observed in mice with adipose tissue-specific knockout of Ir [38]. Conversely, glucose intolerance and insulin resistance were suppressed in mice with adipose tissue-specific enhanced expression of Glut4 [39,40]. These studies have suggested that increased expression levels of insulin signaling-related factors in adipose tissues are crucial in the suppression of type 2 diabetes mellitus. This study revealed that gene and protein expression levels of insulin signaling-related factors were elevated in KK/Ay mice, whose glucose intolerance and insulin resistance were suppressed by oral LPS administration (Figure 5). Therefore, these results suggest that oral LPS administration increases the expression levels of insulin signaling-related factors in adipose tissues of KK/Ay mice and suppresses insulin resistance and glucose tolerance.
The expression levels of insulin signaling-related factors in adipose tissues are thought to be regulated by cytokines produced by adipose tissues [32,41]. Among these factors, adiponectin is a cytokine produced specifically by adipocytes and induces the expression levels of insulin signaling-related factors in adipose tissues [42,43]. For example, in vitro, adiponectin directly induces an increase in Glut4 mRNA and protein expression in adipocytes [43]. In vivo, mice with higher expression levels of adiponectin have increased Glut4 expression levels in adipose tissues and suppressed glucose intolerance and insulin resistance [42]. In this study, the expression of adiponectin in adipose tissues of KK/Ay mice, which showed suppression of type 2 diabetes mellitus by oral LPS administration, was increased, and the adiponectin receptor, a molecule in the pathway for adiponectin to induce increased the gene expression levels of insulin signaling-related factors, was also upregulated (Figure 6). Furthermore, in vitro adiponectin stimulation studies on 3T3–L1 adipocytes confirmed that adiponectin directly induces an increase in Glut4 expression, consistent with previous reports (Figure 7). In addition, adiponectin was demonstrated to directly induce increased mRNA expression levels of Ir, Irs1, and Akt2 (Figure 7). These results suggest that adiponectin induced by oral LPS administration upregulates the expression levels of insulin signaling-related factors in adipose tissues (Figure 8).
LPS injection not only induces type 2 diabetes mellitus by decreasing insulin signaling in adipose tissues but also induces increased expression levels of inflammatory cytokines such as interleukin 1 beta (IL-1β), IL-6, monocyte chemotactic protein 1 (Mcp1), and tumor necrosis factor alpha (TNFα) and induces weight changes, hepatotoxicity, and dyslipidemia. However, oral LPS administration did not induce the mRNA expression levels of IL-1b, IL-6, IL-12b, Mcp1, or TNFα in adipose tissues of KK/Ay mice (Supplementary Figure S1). In addition, it increased the expression levels of insulin signaling-related factors in adipose tissues and suppressed type 2 diabetes mellitus.
No effects were observed on body weight, adipose tissue weight, and blood Alanine transaminase (ALT) and aspartate transaminase (AST) levels, markers of hepatotoxicity (Figure 4 and Figure S2). Furthermore, the results of dyslipidemia markers, such as blood triglyceride (TG), total cholesterol (TC), low-density lipoprotein (LDL), and high-density lipoprotein (HDL), were in agreement with our previous reports [11,12], showing a significant decrease in TC and LDL, consequently a suppressive effect on dyslipidemia (Supplementary Figure S2). Dyslipidemia is suppressed by adiponectin [44], and the results of this study revealed that this effect may be related to the increase in adiponectin levels in adipose tissues induced by oral LPS administration. Furthermore, the results corresponded with our initial proposition that oral LPS administration induces an entirely different effect on organisms compared with LPS injection; thus, we proposed that the conventional idea that LPS is involved in the development of type 2 diabetes mellitus should be revised.
The pathway by which orally administered LPS affects adipose tissues is unknown. Lu et al. found that mice with small intestine-specific TLR4 knockout had impaired glucose tolerance [45], suggesting that orally administered LPS acts starting from TLR4 in the small intestine in vivo. Furthermore, repeated low-dose LPS stimulations mimicking oral LPS administration in vitro do not induce adiponectin expression in 3T3-L1 adipocytes [46]. Thus, we hypothesized that orally administered LPS induces adiponectin expression through an indirect effect on adipose tissues. As possible mediators involved in this signaling process, we found that the membrane-bound Csf1 of blood monocytes is one of the second signaling molecules that transmits the signal of orally administered LPS to distal tissues [10]. In the future, we would like to clarify the mechanism of the inhibitory effect of orally administered LPS on type 2 diabetes mellitus, including analysis of the second signal in the control of adipose tissues.
## 4.1. Animal Experiments
Male KK/Ay mice, aged 7 weeks, were purchased from CLEA Japan (Tokyo, Japan), and male C57BL/6 mice, aged 7 weeks, were purchased from SLC Japan (Hamamatsu, Japan) and maintained in a temperature- and humidity-controlled room under a 12 h light/dark cycle with unrestricted access to food and water. A mouse diet (low-fat diet (LFD); 16.1 kJ/g, $4.3\%$ w/w fat and $0.005\%$ w/w cholesterol; D12450B) was purchased from Research Diets, Inc. (New Brunswick, NJ, USA). All mice were acclimated for 1 week while fed on an acclimation diet (CE-2; CLEA Japan, Tokyo, Japan) and drank sterilized distilled water. KK/Ay mice were assigned to the LPS (+) and LPS (−) groups and fed an LFD for 8 weeks. C57BL/6 mice were assigned to one group (control group) and fed an LFD for 8 weeks. Purified LPS derived from P. agglomerans (obtained from Macrophi Inc., Kagawa, Japan) was dissolved in sterilized distilled water and applied at 1 mg/kg body weight (BW)/day. The LPS dose was estimated from previous in vivo studies, in which the dose required to achieve preventive effects was determined (0.1 ± 1 mg/kg BW/day) [9,10,11,12]. The drinking water was replaced weekly, and the concentration of LPS was adjusted according to the average BW and amount of water consumption. We previously confirmed that LPS degradation in drinking water in a week was not significant [11,12]. At the end of the experiment, the KK/Ay mice were anesthetized under isoflurane vapor and euthanized by decapitation. Whole blood was collected, and a portion was stored at −80 °C until assays were performed. The rest was centrifuged (2000× g for 20 min at 4 °C), and the resulting plasma or serum (supernatant) was stored at −80 °C until assays were performed. Mesenteric, perinephric, and epididymal adipose tissues were collected, weighed, and stored at −80 °C until assays were performed. At the end of the experiment, the C57BL/6 mice were anesthetized under isoflurane vapor and euthanized by decapitation. Serum or plasma was collected in the same manner as with KK/Ay mice and stored at −80 °C until assays were performed.
The animal experiments were reviewed and approved by the Animal Care and Use Committee of the Control of Innate Immunity (approval number: CIITRA 02–08, CIITRA 02–09). This experiment was conducted according to the Law for the Humane Treatment and Management of Animal Standards Relating to the Care and Management of Laboratory Animals and Relief of Pain (Ministry of the Environment, Tokyo, Japan), the Fundamental Guidelines for Proper Conduct of Animal Experiments and Related Activities in Academic Research Institutions (Ministry of Education, Culture, Sports, Science and Technology, Tokyo, Japan), and the Guidelines for Proper Conduct of Animal Experiments (Science Council of Japan).
## 4.2. OGTTs
Mice were fasted overnight and subjected to an OGTT by oral glucose administration (gavage with 2 g of D-glucose/kg BW). Blood samples were collected from the tail vein, and blood glucose levels were monitored using an Accu-Chek Aviva blood glucose meter with Accu-Chek Aviva test strips (Roche Diagnostics K.K., Tokyo, Japan) at 0, 30, 60, 120, and 180 min after glucose loading. The AUC was calculated using the trapezoid rule.
## 4.3. Biochemical Analyses of Serum or Plasma, Whole Blood, and Epididymal ADIPOSE Tissues
TG, TC, LDL, HDL, glucose, AST, and ALT levels in the serum or plasma were measured using commercial enzyme kits (Wako Pure Chemical, Osaka, Japan) according to the manufacturer’s protocol. Insulin was determined using ELISA kits (Shibayagi, Shibukawa, Japan). HbA1c was measured by Oriental Yeast CO., Ltd. (Tokyo, Japan). Epididymal adipose tissue adiponectin levels were measured using ELISA kits (Proteintech Group, Inc., Tokyo, Japan).
## 4.4. Quantitative Reverse-Transcription Polymerase Chain Reaction (qRT-PCR)
RNA was extracted using the RNeasy Mini Kit (QIAGEN, Hilden, Germany), and cDNA was synthesized by reverse transcription using ReverTra Ace qPCR RT Master Mix (TOYOBO, Osaka, Japan), according to the manufacturer’s instructions. RT-PCR assay was conducted using 5 μL of cDNA as the template and 10 μL of Power SYBR Green PCR Master Mix (Thermo Fisher Scientific, Tokyo, Japan) on the Stratagene Mx 3005P QPCR System (Agilent Technologies, Santa Clara, CA, USA). The primers are listed in Table 1. Data were analyzed based on the 2−∆∆Ct method and normalized by GADPH expression. The qPCR amplification was performed with an activation step at 95 °C for 10 min, followed by 40 cycles at 95 °C for 15 s (denaturation) and 60 °C for 1 min (annealing), and a dissociation stage at 95 °C for 15 s, 60 °C for 30 s, and 95 °C for 15 s for each gene.
## 4.5. Western Blot Analysis
Western blot analysis was performed using antibodies that specifically recognize proteins, including Glut4 and glyceraldehyde-3-phosphate dehydrogenase (Gapdh). The epididymal adipose tissue was homogenized, proteins were extracted, and 15 μg of extracted protein was loaded for sodium dodecyl sulfate–polyacrylamide gel electrophoresis immunoblot analysis. Protein bands were then transferred to polyvinylidene fluoride membranes (Bio-Rad Laboratories, Hercules, CA, USA). After blocking the nonspecific sites, the membrane was probed with primary antibodies, followed by a horseradish peroxidase-conjugated secondary antibody (Cell Signaling Technology, Inc., Danvers, MA, USA). Detection of antibody reactions was performed with ECL Western blotting Detection Reagents (Advansta, San Jose, CA, USA). Each band was normalized using the corresponding value of Gapdh as an internal control. The antibodies used were Gapdh (primary antibody (mouse monoclonal, Abcam, Cambridge, UK) 1:4000 dilution, secondary antibody (rabbit polyclonal, Abcam) 1:4000 dilution) and Glut4 (primary antibody (rabbit monoclonal, Cell signaling) 1:1000 dilution, secondary antibody (rabbit polyclonal, Abcam) 1:4000 dilution). Reaction times were overnight at 4 °C for primary antibodies and 1 h at room temperature for secondary antibodies.
## 4.6. Cell Culture
Mouse embryo 3T3-L1 cell line was obtained from the American Type Culture Collection (Manassas, VA, USA). 3T3-L1 pre-adipocytes were cultured in Dulbecco’s modified *Eagle medium* (Wako) supplemented with $10\%$ fetal bovine serum (Sigma-Aldrich, St. Louis, MO, USA) at 37 °C in $5\%$ CO2. AdipoInducer Reagent (Takara Bio, Otsu, Japan) was used for the differentiation of 3T3-L1 pre-adipocytes. For the differentiation of 3T3-L1 pre-adipocytes to mature adipocytes, 3T3-L1 pre-adipocytes were induced with differentiation media (DMEM with low glucose content supplemented with $10\%$ fetal bovine serum, 2.5 μM dexamethasone (DEX), 0.5 mM 3-Isobutyl 1-methylxanthine (IBMX), and 10 μg/mL insulin (days 0–2). On day 2, the medium was replenished with maturation media (DMEM with high glucose content supplemented with $10\%$ fetal bovine serum and 10 μg/mL insulin) and maintained at a 37 °C and $5\%$ CO2 environment. This medium was changed every 2 days until day 8. At this time, the cells exhibited characteristics of mature adipocytes. At day 8, the medium was replenished, and 3T3-L1 mature adipocytes were treated with or without recombinant adiponectin (20 μg/mL, Prospec, Ness-Ziona, Israel). Samples were collected at 24 h to extract RNA after adiponectin treatment.
## 4.7. Statistical Analysis
All statistical analyses were performed using Ekuseru Toukei 2012 (SSRI, Tokyo, Japan). Data are reported as the mean ± standard error of the mean (SE). Statistical analysis was performed by Student’s t test. A difference was considered significant at $p \leq 0.05.$
## 5. Conclusions
This study showed for the first time that oral LPS administration has an inhibitory effect on type 2 diabetes mellitus. In addition, the study showed that LPS increases the expression levels of insulin signaling-related factors in adipose tissues, and the upregulator of these factors is adiponectin in adipose tissues.
## References
1. Leuwer M., Welters I., Marx G., Rushton A., Bao H., Hunter L., Trayhurn P.. **Endotoxaemia leads to major increases in inflammatory adipokine gene expression in white adipose tissue of mice**. *Pflug. Arch.* (2009) **457** 731-741. DOI: 10.1007/s00424-008-0564-8
2. Mehta N.N., McGillicuddy F.C., Anderson P.D., Hinkle C.C., Shah R., Pruscino L., Tabita-Martinez J., Sellers K.F., Rickels M.R., Reilly M.P.. **Experimental endotoxemia induces adipose inflammation and insulin resistance in humans**. *Diabetes* (2010) **59** 172-181. DOI: 10.2337/db09-0367
3. Anhê F.F., Barra N.G., Cavallari J.F., Henriksbo B.D., Schertzer J.D.. **Metabolic endotoxemia is dictated by the type of lipopolysaccharide**. *Cell Rep.* (2021) **36** 109691. DOI: 10.1016/j.celrep.2021.109691
4. Luche E., Cousin B., Garidou L., Serino M., Waget A., Barreau C., André M., Valet P., Courtney M., Casteilla L.. **Metabolic endotoxemia directly increases the proliferation of adipocyte precursors at the onset of metabolic diseases through a CD14-dependent mechanism**. *Mol. Metab.* (2013) **2** 281-291. DOI: 10.1016/j.molmet.2013.06.005
5. Ulmer A.J., Rietschel E.T., Zähringer U., Heine H.. **Lipopolysaccharide: Structure, Bioactivity, Receptors, and Signal Transduction**. *Trends Glycosci. Glycotechnol.* (2002) **14** 53-68. DOI: 10.4052/tigg.14.53
6. Kitamoto S., Nagao-Kitamoto H., Hein R., Schmidt T.M., Kamada N.. **The Bacterial Connection between the Oral Cavity and the Gut Diseases**. *J. Dent. Res.* (2020) **99** 1021-1029. DOI: 10.1177/0022034520924633
7. Festi D., Schiumerini R., Eusebi L.H., Marasco G., Taddia M., Colecchia A.. **Gut microbiota and metabolic syndrome**. *World J. Gastroenterol.* (2014) **20** 16079-16094. DOI: 10.3748/wjg.v20.i43.16079
8. Soma G.-I., Inagawa H., Nishizawa T., Yoshioka N., Taniguchi Y., Kohchi C.. **Preventative and therapeutic potential of lipopolysaccharide derived from edible Gram-negative bacteria to various diseases**. *Current. Drug Ther.* (2008) **3** 26-32. DOI: 10.2174/157488508783331243
9. Inagawa H., Kobayashi Y., Kohchi C., Zhang R., Shibasaki Y., Soma G.. **Primed Activation of Macrophages by Oral Administration of Lipopolysaccharide Derived from Pantoea agglomerans**. *In Vivo* (2016) **30** 205-211. PMID: 27107076
10. Mizobuchi H., Yamamoto K., Yamashita M., Nakata Y., Inagawa H., Kohchi C., Soma G.I.. **Prevention of Diabetes-Associated Cognitive Dysfunction Through Oral Administration of Lipopolysaccharide Derived From Pantoea agglomerans**. *Front. Immunol.* (2021) **12** 650176. DOI: 10.3389/fimmu.2021.650176
11. Kobayashi Y., Inagawa H., Kohchi C., Kazumura K., Tsuchiya H., Miwa T., Okazaki K., Soma G.I.. **Oral administration of Pantoea agglomerans-derived lipopolysaccharide prevents metabolic dysfunction and Alzheimer's disease-related memory loss in senescence-accelerated prone 8 (SAMP8) mice fed a high-fat diet**. *PLoS ONE* (2018) **13**. DOI: 10.1371/journal.pone.0198493
12. Kobayashi Y., Inagawa H., Kohchi C., Kazumura K., Tsuchiya H., Miwa T., Okazaki K., Soma G.I.. **Oral administration of Pantoea agglomerans-derived lipopolysaccharide prevents development of atherosclerosis in high-fat diet-fed apoE-deficient mice via ameliorating hyperlipidemia, pro-inflammatory mediators and oxidative responses**. *PLoS ONE* (2018) **13**. DOI: 10.1371/journal.pone.0195008
13. Tian L.Q., Yu Y.T., Jin M.D., Duan H.L., Huang G., Zhang M.L.. **Early 1,25-Dihydroxyvitamin D(3) Supplementation Effectively Lowers the Incidence of Type 2 Diabetes Mellitus via Ameliorating Inflammation In KK-A(y) Mice**. *J. Nutr. Sci. Vitaminol.* (2021) **67** 84-90. DOI: 10.3177/jnsv.67.84
14. Zhang X.L., Wang Y.N., Ma L.Y., Liu Z.S., Ye F., Yang J.H.. **Uncarboxylated osteocalcin ameliorates hepatic glucose and lipid metabolism in KKAy mice via activating insulin signaling pathway**. *Acta Pharmacol. Sin.* (2020) **41** 383-393. DOI: 10.1038/s41401-019-0311-z
15. Marcano R., Rojo M.. **Pathological and Therapeutic Approach to Endotoxin-Secreting Bacteria Involved in Periodontal Disease**. *Toxins* (2021) **13**. DOI: 10.3390/toxins13080533
16. Cani P.D., Amar J., Iglesias M.A., Poggi M., Knauf C., Bastelica D., Neyrinck A.M., Fava F., Tuohy K.M., Chabo C.. **Metabolic endotoxemia initiates obesity and insulin resistance**. *Diabetes* (2007) **56** 1761-1772. DOI: 10.2337/db06-1491
17. Manco M.. **Endotoxin as a missed link among all the metabolic abnormalities in the metabolic syndrome**. *Atherosclerosis* (2009) **206** 36. DOI: 10.1016/j.atherosclerosis.2009.03.047
18. Cani P.D., Bibiloni R., Knauf C., Waget A., Neyrinck A.M., Delzenne N.M., Burcelin R.. **Changes in gut microbiota control metabolic endotoxemia-induced inflammation in high-fat diet-induced obesity and diabetes in mice**. *Diabetes* (2008) **57** 1470-1481. DOI: 10.2337/db07-1403
19. Kawano M., Miyoshi M., Ogawa A., Sakai F., Kadooka Y.. **Lactobacillus gasseri SBT2055 inhibits adipose tissue inflammation and intestinal permeability in mice fed a high-fat diet**. *J. Nutr. Sci.* (2016) **5** e23. DOI: 10.1017/jns.2016.12
20. Harris H.W., Grunfeld C., Feingold K.R., Read T.E., Kane J.P., Jones A.L., Eichbaum E.B., Bland G.F., Rapp J.H.. **Chylomicrons alter the fate of endotoxin, decreasing tumor necrosis factor release and preventing death**. *J. Clin. Invest.* (1993) **91** 1028-1034. DOI: 10.1172/JCI116259
21. Ghoshal S., Witta J., Zhong J., de Villiers W., Eckhardt E.. **Chylomicrons promote intestinal absorption of lipopolysaccharides**. *J. Lipid Res.* (2009) **50** 90-97. DOI: 10.1194/jlr.M800156-JLR200
22. Kumwenda Z.L., Wong C.B., Johnson J.A., Gosnell J.E., Welch W.J., Harris H.W.. **Chylomicron-bound endotoxin selectively inhibits NF-kappaB activation in rat hepatocytes**. *Shock* (2002) **18** 182-188. DOI: 10.1097/00024382-200208000-00016
23. Kondo K.. **Inbred strains resulting from Japanese mice**. *Bull Exp. Anim.* (1957) **6** 107-112
24. Nakamura M., Yamada K.. **Studies on a diabetic (KK) strain of the mouse**. *Diabetologia* (1967) **3** 212-221. DOI: 10.1007/BF01222198
25. Ishida T., Jobu K., Morisawa S., Kawada K., Yoshioka S., Miyamura M.. **Juzentaihoto Suppresses Muscle Atrophy in KKAy Mice**. *Biol. Pharm. Bull.* (2022) **45** 888-894. DOI: 10.1248/bpb.b22-00039
26. Ohtake K., Nakano G., Ehara N., Sonoda K., Ito J., Uchida H., Kobayashi J.. **Dietary nitrite supplementation improves insulin resistance in type 2 diabetic KKA(y) mice**. *Nitric. Oxide* (2015) **44** 31-38. DOI: 10.1016/j.niox.2014.11.009
27. Mizuno Y., Yamamotoya T., Nakatsu Y., Ueda K., Matsunaga Y., Inoue M.K., Sakoda H., Fujishiro M., Ono H., Kikuchi T.. **Xanthine Oxidase Inhibitor Febuxostat Exerts an Anti-Inflammatory Action and Protects against Diabetic Nephropathy Development in KK-Ay Obese Diabetic Mice**. *Int. J. Mol. Sci.* (2019) **20**. DOI: 10.3390/ijms20194680
28. Nakata K., Taniguchi Y., Yoshioka N., Yoshida A., Inagawa H., Nakamoto T., Yoshimura H., Miyake S., Kohchi C., Kuroki M.. **A mixture of Salacia oblonga extract and IP-PA1 reduces fasting plasma glucose (FPG) and low-density lipoprotein (LDL) cholesterol levels**. *Nutr. Res. Pract.* (2011) **5** 435-442. DOI: 10.4162/nrp.2011.5.5.435
29. Smith U., Kahn B.B.. **Adipose tissue regulates insulin sensitivity: Role of adipogenesis, de novo lipogenesis and novel lipids**. *J. Intern. Med.* (2016) **280** 465-475. DOI: 10.1111/joim.12540
30. Ohlsson C., Hammarstedt A., Vandenput L., Saarinen N., Ryberg H., Windahl S.H., Farman H.H., Jansson J.O., Movérare-Skrtic S., Smith U.. **Increased adipose tissue aromatase activity improves insulin sensitivity and reduces adipose tissue inflammation in male mice**. *Am. J. Physiol. Endocrinol. Metab.* (2017) **313** E450-E462. DOI: 10.1152/ajpendo.00093.2017
31. Ding S., Jiang J., Wang Z., Zhang G., Yin J., Wang X., Wang S., Yu Z.. **Resveratrol reduces the inflammatory response in adipose tissue and improves adipose insulin signaling in high-fat diet-fed mice**. *PeerJ* (2018) **6** e5173. DOI: 10.7717/peerj.5173
32. Burhans M.S., Hagman D.K., Kuzma J.N., Schmidt K.A., Kratz M.. **Contribution of Adipose Tissue Inflammation to the Development of Type 2 Diabetes Mellitus**. *Compr. Physiol.* (2018) **9** 1-58. PMID: 30549014
33. Longo M., Zatterale F., Naderi J., Parrillo L., Formisano P., Raciti G.A., Beguinot F., Miele C.. **Adipose Tissue Dysfunction as Determinant of Obesity-Associated Metabolic Complications**. *Int. J. Mol. Sci.* (2019) **20**. DOI: 10.3390/ijms20092358
34. Li J., Gong L., Liu S., Zhang Y., Zhang C., Tian M., Lu H., Bu P., Yang J., Ouyang C.. **Adipose HuR protects against diet-induced obesity and insulin resistance**. *Nat. Commun.* (2019) **10** 2375. DOI: 10.1038/s41467-019-10348-0
35. Cho S., Lee H., Han J., Lee H., Kattia R.O., Nelson Z.V., Choi S., Kim S.-Y., Park H.-Y., Jeong H.G.. **Viburnum stellato-tomentosum Extract Suppresses Obesity and Hyperglycemia through Regulation of Lipid Metabolism in High-Fat Diet-Fed Mice**. *Molecules* (2021) **26**. DOI: 10.3390/molecules26041052
36. Hauffe R., Stein V., Chudoba C., Flore T., Rath M., Ritter K., Schell M., Wardelmann K., Deubel S., Kopp J.F.. **GPx3 dysregulation impacts adipose tissue insulin receptor expression and sensitivity**. *JCI Insight* (2020) **5** e136283. DOI: 10.1172/jci.insight.136283
37. Abel E.D., Peroni O., Kim J.K., Kim Y.B., Boss O., Hadro E., Minnemann T., Shulman G.I., Kahn B.B.. **Adipose-selective targeting of the GLUT4 gene impairs insulin action in muscle and liver**. *Nature* (2001) **409** 729-733. DOI: 10.1038/35055575
38. Softic S., Boucher J., Solheim M.H., Fujisaka S., Haering M.F., Homan E.P., Winnay J., Perez-Atayde A.R., Kahn C.R.. **Lipodystrophy Due to Adipose Tissue-Specific Insulin Receptor Knockout Results in Progressive NAFLD**. *Diabetes* (2016) **65** 2187-2200. DOI: 10.2337/db16-0213
39. Shepherd P.R., Gnudi L., Tozzo E., Yang H., Leach F., Kahn B.B.. **Adipose cell hyperplasia and enhanced glucose disposal in transgenic mice overexpressing GLUT4 selectively in adipose tissue**. *J. Biol. Chem.* (1993) **268** 22243-22246. DOI: 10.1016/S0021-9258(18)41516-5
40. Carvalho E., Kotani K., Peroni O.D., Kahn B.B.. **Adipose-specific overexpression of GLUT4 reverses insulin resistance and diabetes in mice lacking GLUT4 selectively in muscle**. *Am. J. Physiol. Endocrinol. Metab.* (2005) **289** E551-E561. DOI: 10.1152/ajpendo.00116.2005
41. Kawai T., Autieri M.V., Scalia R.. **Adipose tissue inflammation and metabolic dysfunction in obesity**. *Am. J. Physiol. Cell Physiol.* (2021) **320** C375-C391. DOI: 10.1152/ajpcell.00379.2020
42. Aprahamian T.R.. **Elevated adiponectin expression promotes adipose tissue vascularity under conditions of diet-induced obesity**. *Metabolism* (2013) **62** 1730-1738. DOI: 10.1016/j.metabol.2013.07.010
43. Fu Y., Luo N., Klein R.L., Garvey W.T.. **Adiponectin promotes adipocyte differentiation, insulin sensitivity, and lipid accumulation**. *J. Lipid Res.* (2005) **46** 1369-1379. DOI: 10.1194/jlr.M400373-JLR200
44. Wang X., Chen Q., Pu H., Wei Q., Duan M., Zhang C., Jiang T., Shou X., Zhang J., Yang Y.. **Adiponectin improves NF-κB-mediated inflammation and abates atherosclerosis progression in apolipoprotein E-deficient mice**. *Lipids Health Dis.* (2016) **15** 33. DOI: 10.1186/s12944-016-0202-y
45. Liu X., Perakakis N., Gong H., Chamberland J.P., Brinkoetter M.T., Hamnvik O.P.R., Mantzoros C.S.. **Adiponectin administration prevents weight gain and glycemic profile changes in diet-induced obese immune deficient Rag1-/- mice lacking mature lymphocytes**. *Metabolism* (2016) **65** 1720-1730. DOI: 10.1016/j.metabol.2016.09.003
46. Yamamoto K., Yamashita M., Inagawa H., Kohchi C., Soma G.I.. **Anti-inflammatory and Insulin Signaling Phenotype Induced by Repeated Lipopolysaccharide Stimulation in 3T3-L1 Adipocytes**. *Anticancer. Res.* (2022) **42** 3983-3991. DOI: 10.21873/anticanres.15894
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.