Methylation loci associated with body mass index, waist circumference, and waist-to-hip ratio in Chinese adults: an epigenome-wide analysis. (October 2016)

• Record Type: Journal Article
• Authors: Wang, Biqi; Gao, Wenjing; Li, Jun; Yu, Canqing; Cao, Weihua; Lv, Jun; Pang, Zengchang; Cong, Liming; Wang, Hua; Wu, Xianping; Liang, Liming; Wu, Tangchun; Li, Liming
• Is Part Of: Lancet. Volume 388(2016)Supplement 1
• Page Start: S21
• Publication Date: 2016-10
• Subjects: Medicine -- Periodicals; Medicine -- Periodicals; Medicine; Medicine; Electronic journals; Periodicals; 610.5
• DOI: 10.1016/S0140-6736(16)31948-1
• Languages: English
• ISSNs: 0140-6736

Abstract

Abstract: Background: DNA methylation has been implicated in the pathology of obesity, but little is known about such epigenetic variants in Chinese people. We conducted an epigenome-wide analysis of methylation at CpG sites in relation to body mass index (BMI), waist circumference (WC) and waist-to-hip ratio (WHR) among Chinese adult populations. Methods: 469 twins from the Chinese National Twin Registry were the discovery cohort. After quality control, methylation levels in whole blood leukocyte DNA were tested for association with BMI, WC, and WHR using mixed linear regressions, adjusting for demographics, lifestyle, and family structure. Methylation sites at a false discovery rate (FDR) q value of 0·05 or less were selected for replication in four independent panels of general populations: Dongfeng-tongji cohort (n=688), Coke Oven Workers (n=137), Wuhan-Zhuhai cohort (n=241), and Shiyan participants (n=144). Association of obesity traits with the selected CpG sites in each replication panel was examined using linear regressions with adjustment of covariates similar to that in the discovery cohort, and results were combined using fixed effect meta-analysis with significance level penalised by Bonferroni correction. All participants gave written informed consent, and the Peking University Biomedical Ethics Committee approved the study protocol. Findings: Methylation levels at 6, 3, and 1 CpG sites were significantly associated with BMI, WC, and WHR, respectively in discovery cohort (FDR q value range 0·0010–0·0443); these findings were replicated by meta-analysis in the four panels. We found associations between obesity traits and methylation at genes CPT1A (meta-analysis p=0·0013 for BMI, p=0·0047 for WC, and p=0·0178 for WHR), ABCG1 (p<0·0001 for BMI and WC) and SREBF1 (p<0·0001 for BMI). Additionally, CpG sites at gene ARID1B (meta-analysis effect size β −0·0221, 95% CI −0·0363 to −0·0079, p=0·0022 for BMI), gene TOP1 (−0·0237, −0·0404 to −0·0071, p=0·0053 for BMI), and methylation site cg17061862 (−0·0379, −0·0575 to −0·018, p=0·0002 for BMI; −0·0356, −0·0608 to −0·0103, p=0·0057 for WC) showed inverse associations with BMI or WC. Interpretation: Our findings support the association of DNA methylation at CPT1A, ABCG1, and SREBF1 with obesity traits in Chinese populations. We identified additional associated methylation sites at genes ARID1B and TOP1 and at cg17061862. Funding: This study was supported by theSpecific Research Project of Health Public Service, Ministry of Health, China (201002007, 201502006 ) andKey Grant Project of Chinese Ministry of Education (310006 ).