Abstract P178: Methylation Sites Associated With Cumulative Socioeconomic Status And Body Mass Index Among African American Adults From The Atherosclerosis Risk In Communities Study

Abstract

Background: Obesity increases the risk of multiple morbidities and is patterned by socioeconomic status (SES). Lower SES is associated with greater burden of obesity, especially among African American adults. Epigenetic research has provided novel insights into human biology and health; however, few studies have examined whether epigenetic loci mediate the association between SES and obesity. Objective: We sought to determine the union of epigenetic sites associated with cumulative SES and body mass index (BMI) among African American adults from the Atherosclerosis Risk in Communities (ARIC) study, and to examine differences by sex. Hypothesis: Several Cytosine-phosphate-guanine (CpG) sites will be associated with both cumulative SES and BMI, which will provide suggestive evidence of epigenetic mechanisms mediating the association between SES and obesity. Also, sex differences will be detected. Methods: Methylation data from venous blood of 2,684 African American adults (1702 women, 982 men) in the ARIC study (mean age 56.61) was collected from either visit 2 (1990-1992) or visit 3 (1993-1995) using the Illumina HumanMethylation 450k array. For each participant, we created a cumulative SES score comprised of education, income, and occupation from visit 1 (1987-1989). Each SES indicator had 3 categories, which were given a score of 0 (low), 0.5 (medium), or 1 (high), then all SES indicator scores were summed (mean 1.47, SD 0.76, range: 0-3). BMI values (mean 30.12, SD 6.26, range 14.68-62.35) from the same visit as the methylation typing, were then inverse rank normalized. Linear regression analyses were used to test the association of cumulative SES scores and the association of BMI residuals with CpG sites, respectively. Analyses were stratified by sex and combined using inverse variance fixed-effect meta-analysis. The final model adjusted for age, age 2 , alcohol use, smoking, genetic principal components and time between visits. Statistical significance was determined as p <2X10 -7 . Results: Among men and women combined, 3,843 CpG sites were associated with cumulative SES and 212 CpG sites were associated with BMI. Thirty (n=30) CpG sites were statistically significant for both cumulative SES and BMI. Of the 30 sites, 15 sites were previously identified in epigenetic studies of obesity/BMI, and 11 sites were previously identified for other cardiometabolic traits. Sites were statistically more significant among women. Conclusion: Our findings suggest shared patterns of epigenetic variation between SES and BMI and differences by sex. As a follow up, we will replicate our study findings in other cohort studies and test the hypothesis that the effect of SES on obesity is mediated through methylation. Indeed, such findings could reveal how SES gets “under the skin” to influence obesity. Such knowledge could help us better identify therapeutic targets for intervention.