Sex differences in heritable and environmental risk of abdominal obesity: family-based study and case-control study of rural adults in China


 Background: Abdominal obesity (AO) has shown a dramatic increase trend in the past decades. But there were few studies on sex differences in the risk of AO, especially the heritable risk. This study aimed to investigate the sex differences of AO risk in China rural areas.Methods: A family-based study coupled with the Henan Rural Cohort was carried out, including 1,533 people and 39,259 people, respectively. Questionnaires were applied in including included demographics, dietary intake and physical activity. Combined family-based study with the Henan Rural Cohort, sex differences in heritable risk of AO was analyzed among rural people. Chi-square test was applied to study the heritable risk of AO. Environmental factors for AO were assessed in couple case-control study. Logistic regression analysis was carried out to assess the association of lifestyles with AO.Results: Women had higher risk to develop AO than men in Henan Rural Cohort (OR: 3.070, 95% CI: 2.943, 3.203, P<0.001). The female first-degree relatives were more susceptible to AO than those in the Henan Rural Cohort (OR: 1.854, 95% CI: 1.318, 2.608, P<0.001). Moreover, female offspring had higher risk to develop AO than male offspring when their mothers suffered from AO (OR: 1.797, 95% CI: 1.111, 2.907, P=0.016), and this risk increased if their grandmothers suffered from AO (OR: 6.750, 95% CI: 1.667, 27.336, P=0.007). In addition, low physical activity increased the risk to develop AO for women when compared with their husbands (OR: 2.253, 95% CI: 1.283, 3.957, P=0.005). Conclusion: This study indicated females are heritably susceptible to AO than men, especially for those who have maternal family history and reinforce the benefits of increasing physical activity for protecting females from suffering from AO. Proactive action based on maternal family history is a cost-effective strategy for the prevention of AO, although larger family-based studies are required to verify the conclusion.

In addition, a large proportion of the population still live in rural areas with limited access to health but high burden of metabolic disorders. Thus, a family-based study was carried out in rural China. Comparing with the AO prevalence in Henan Rural Cohort study, the AO prevalence among the rst-degree relatives of probands could be applied to speculate the heritable risk. Furthermore, it could provide further evidence by comparison of the AO prevalence between female and male offspring of AO cases. Besides, casecontrol couples could be employed to investigate the sex differences of environmental risk factors of AO.
This study will provide more reliable evidence for sex-speci c prevention and help reduce the burden of metabolic disorders in rural areas.

Study subjects
In this family-based study 1,533 participants from 499 extended families including 130 AO probands and their 359 rst-degree relatives were recruited. From them, 433 offspring of AO cases were selected to assess the heritable risk. On the other hand, 178 case-control couples (40 husband-case and 138 wifecase) out of 377 couples (N = 754) was included to investigate the environmental risk factor of AO ( Fig. 1). All the participants living in rural area of Zhengzhou city or Jiaozuo city in Henan Province, China.
In this study, 39,259 subjects from Henan Rural Cohort were recruited from ve rural regions of Henan Province, China. The registration number of Henan Rural Cohort was ChiCTR-OOC-15006699, detail information of which was available at the website (http://www.chictr.org.cn/showproj.aspx?proj=11375), as well as the introduction article we published previously. [13] This study complied with the Declaration of Helsinki. The Life Science Ethics Review Committee of Zhengzhou University had reviewed and approved the protocol. All the subjects participated in this study had signed an informed consent.
De nition Of Abdominal Obesity WC was measured from the midpoint line between the lowest point of the rib and the upper edge of the iliac crest using non-elastic tape. According to the recommendation of Working Group on Obesity in China (WGOC), the appropriate cut-off point for AO for adults in China was de ned as WC ≥ 85 cm for men and ≥ 80 cm for women. [14] Information Collection By Face-to-face Interview A face-to-face interview with standardized questionnaire was conducted for all the participated individuals in this family-based study by well-trained investigators. The information collected included demographics, dietary intake and physical activity.
Combined with the Chinese Residents Dietary Guidelines and local people dietary habits, food frequency questionnaire was applied to collect dietary intake information. According to the Chinese Food Composition Table, fat intake was gured with multiplying the fat of detailed portion size in each food item by the frequency, divided into < 50, ≥50 g/day, and vegetables intake were classi ed as < 50, ≥50 g/day. [15] The short version of the international physical activity questionnaires (IPAQ) was employed to collect the physical activity information. By showing illustration cards of different physical activities with description, the investigators recorded the self-report physical activity within the last 7 days for all the participants. Based on the IPAQ scoring protocol version 2.0 (http://www.ipaq.ki), physical activity was divided into three levels of low, moderate and high.

Statistical analysis
In order to reveal the sex differences of heritable and environmental risk factors for AO, the strategy of statistical analysis was as follows. First, chi-square test was applied to investigate the heritable risk on AO by comparing the prevalence of AO between the rst-degree relatives and the Henan Rural Cohort. Second, the heritable risk was veri ed by comparing the prevalence of AO between male and female offspring with chi-square test. Third, logistic regression analysis was carried out to assess the association of dietary intake and physical activity with AO. SPSS 21.0 (IBM SPSS, New York, US) was applied to conduct all the statistical analysis. A value of P less than 0.05 was considered statistically signi cant.

Results
Structural characteristic of the family-based study and the couple case-control study In the family-based study, 130 probands and their 359 rst-degree relatives were included, i.e. 130 fathers, 130 mothers, 67 siblings, and 32 offspring. In the couple study, 377 couples were included, i.e. 63 couples without AO, 40 couples with AO for husband, 138 couples with AO for wife, and 136 couples with AO for both husband and wife. There was no distribution difference between the four different couple types (P = 0.061). Then the 40 couples with AO for husband and the 138 couples with AO for wife were applied as case-control studies (Model 1 and Model 2 in Table 4, respectively) to investigate the associations of diet and physical activity with AO.
Association Between Sex And Abdominal Obesity 39,259 subjects and 1,533 subjects were recruited to investigate the association between sex and AO in the Henan Rural Cohort and the family-based study ( Table 1). The result for the Henan Rural Cohort suggested that women had higher risk to develop AO than men (OR: 3.070, 95% CI: 2.943, 3.203, P < 0.001). This signi cant association between sex and AO was also observed in the family-based study (OR: 2.617, 95% CI: 2.127, 3.221, P < 0.001). Note: Chi-square test was applied to investigate the association between sex and abdominal obesity. AO: abdominal obesity. *P < 0.05.

Prevalence comparison of abdominal obesity between the rst-degree relatives and the Henan Rural Cohort
The prevalence of AO in the Henan Rural Cohort and the rst-degree relatives were 51.6% and 57.4%, respectively (Table 2). Compared with the subjects in the Henan Rural Cohort, the rst-degree relatives were at a higher risk to develop AO (OR: 1.265, 95% CI: 1.025, 1.561, P = 0.028). In the subgroup analysis, the female rst-degree relatives were more susceptible to AO than those in the Henan Rural Cohort (OR: 1.854, 95% CI: 1.318, 2.608, P < 0.001). Instead, this phenomenon was not observed in the male rstdegree relatives (P = 0.216).

Discussion
The differences of heritable and environmental risk between male and female AO were investigated in this study. The prevalence of AO was compared between the rst-degree relatives of family-based study and people of the Henan Rural Cohort to assess the heritable risk, which was further determined in offspring. Then a couple case-control study was applied to assess the environmental risk. The results indicated that women were at a higher risk to develop AO than men. Moreover, the risk could be transmitted from mother to daughter and even from grandmother to granddaughter. However, no signi cantly heritable risk of AO was observed from father to daughter or from grandfather to granddaughter. In addition, the results of couple case-control study indicated that low physical activity was a risk factor for women to develop AO. Therefore, women are heritably susceptible to AO, especially for those who have maternal family history. Increasing physical activity may protect them from suffering from AO.
This study showed that women were more susceptible to AO, which was consistent with previous studies.
A national study in China showed that the prevalence of AO increased from 8.5% in 1993 to 27.8% in 2009 in men and from 27.8-45.9% in women.
[16] A review of the dynamic trends in the prevalence of AO between 1989 and 2011 suggested that the prevalence of AO was higher in women than in men in all cycles. [17] This signi cant sex differences may be due to the homeostasis of endocrine regulation between men and women. After menopause, the secretion of sex hormones in women decreases, resulting in the accumulation of visceral fat and an increased risk of metabolic disorders. [18,19] The alarming increase in the prevalence of AO may partly explain the increase in the prevalence of metabolic syndrome in recent years. It was reported that AO with normal BMI was associated with cardiovascular diseases, and was related to dyslipidemia in women. [20] Another South Korean study showed that general obesity without AO was not associated with a higher risk of cardiovascular diseases. [21] Another major nding of this study was that women with maternal AO history have a higher risk of developing AO. This heritable risk included genetic and environmental factors as well as geneenvironmental interaction. However, it is di cult to completely separate the effects of gene and environment on AO, and gene-environment interaction plays an important role in this heritable risk. The important molecular mechanism of gene-environment interaction is epigenetics. For example, the heritability of BMI could be up to 70%. [22] However, GWAS suggested that SNP contributed only 6% to BMI. [23] The unknown part of heritable contribution may be associated with epigenetic transmission. Recent work has indicated that DNA methylation, histone modi cation and other epigenetic changes played a vital role in fetal growth and development. [24,25] An animal experiment showed that experiencing high-fat diet in womb may cause epigenetic changes and the metabolic disorders in offspring, suggesting that a mother's experience during pregnancy may cause epigenetic changes and be transmitted to the next generation. [26] On the other hand, the epigenetic pattern associate with AO may be reversible. The results of this study suggested that female offspring with AO mother had higher risk to develop AO than male offspring. This may be related to epigenetic changes caused by physical activity. Recent studies have shown that epigenetics is highly active and affected by environmental factors. [27] Previous study revealed that moderate exercise could up-regulate the methylation level of related proteins and contributed to reduce in ammation level, thus preventing the occurrence of diseases associated with chronic in ammation. [28] Some studies have shown that physical activity also plays a vital role in histone modi cation and noncoding RNA expression. [29,30] Therefore, physical activity would regulate the epigenetic mechanisms of a variety of human diseases. Due to the traditional family division of labor in China, men are responsible for most of the outdoor work, which results in a signi cantly higher level of physical activity in men than in women. Increased physical activity in men may reverse the epigenetic changes associated with AO, which may be one reason to explain the higher risk of AO in women than in men.
Moreover, transmission of risk lifestyle for AO from mother or grandmother to offspring may be another factor to increase AO prevalence in AO families. Caregivers act as instructors and executors of children's behavior patterns in early childhood, affecting children's development and future lifestyles and habits of behavior. [31] It was reported that parents with AO had less physical activity and higher fat intake than normal parents, which also create an environment for their children to suffer from obesity. [32] Similarly, children with good household routines have a lower risk of obesity. [33] In China rural areas, it is common for men to go out for work, leaving their wives or mothers to take care of their children. With the rapid economic and social development, part of women living in rural area go out to work to reduce the nancial burden. As a result, the grandmother is the primary caregiver of children. Grandmothers' low level of education and lack of health and nutrition knowledge would easily lead to overfeeding of children, which will have a negative impact on the formation of offspring's behavior patterns. [34] This process may be one reason to explain the transmission process of AO risk behavior model from parent to offspring.
In addition, it was suggested that higher fat intake was signi cantly associated with AO for male, but a protective factor for female. It may be attributed to the fact that fat intake of males is higher than that of females in both cases and controls (Fig. 2). This is an original study to integrate a family-based study and a cohort study with large sample in the same district. Thus, the results are reliable and convincing. Present study indicated that women with maternal AO history had a higher risk of developing AO, which revealed information about sex differences in AO and had important public health implications. The results also provided clues for further study of the underlying mechanisms for sex differences in AO risk. On the other hand, limitations of this study must be acknowledged. First, the results would be more credible if the sample size of family-based study was larger. Nevertheless, unlike the recruitment of other cross-sectional subjects, family-based study required complete several generations, coupled with the older age of the target population, which greatly increased the di culty of expanding the sample size. Therefore, polycentric and multi co-operational studies are needed to verify the conclusion of this study in the future. Then, this study did not separate genetic from environmental contributions to AO risk, although this is not our general aim. Future work that focuses on the gene-environment interaction will be a good supplement of current research. More so, the generalizability of our results to the whole population may be uncertain, but it has good applicability to low and middle-income areas represented by rural areas.

Conclusions
In this study combined family-based study and case-control study found that females are heritably susceptible to AO, especially for those who have maternal family history. Increasing physical activity may protect them from suffering from AO. Proactive action based on maternal family history is a costeffective strategy for the prevention of AO, although dim light was shed on the sex differences of heritable and environmental risk for AO and a larger family-based study is required to verify the conclusion.

Declarations Ethics approval and consent to participate
This study complied with the Declaration of Helsinki. The Life Science Ethics Review Committee of Zhengzhou University had reviewed and approved the protocol. All the subjects participated in this study had signed an informed consent.

Consent for publication
Written informed consent for publication was obtained from all participants.

Availability of data and materials
The data used during the current study are available from the corresponding author on reasonable request.

Competing interests
There was no con ict of interest to be declared. Authors' contributions SY designed the research, drafted the original manuscript, performed the statistical analysis. YF drafted the original manuscript, performed the statistical analysis. CQ analyzed and interpreted the data, revised the original manuscript. XL designed the research, revised the original manuscript. WH and ZM designed the research, analyzed and interpreted the data, revised the original manuscript. CC put forward conception, reviewed the manuscript. CW and WL designed the research, reviewed the manuscript, managed and coordinated the planning and execution of research activities and provided nancial support for the study. Family-based study were carried out in Henan Rural areas, China. Comparing with the AO prevalence in Henan Rural Cohort study, the AO prevalence among the rst-degree relatives of probands were applied to assess the heritable impact. The comparison of the prevalence of male and female offspring of AO provided further evidence. On the other hand, case-control couples were employed to investigate the sex differences of environmental risk factors of AO. Note: The designations employed and the presentation of the material on this map do not imply the expression of any opinion whatsoever on the part of Research Square concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. This map has been provided by the authors.