Abstract 2075: Replication of Association of Newly Identified Loci With Metabolic and Lipid-related Traits in Asian Indian Sikhs
Recent genome-wide association scans (GWAS) have implicated several loci with metabolic and lipid related traits. This investigation seeks to examine the role of six GWAS-derived loci (CETP rs3764241; APOA5-A4-C3-A1 rs12286037; APOE-C1-C4 rs4420638; CELSR2-PSRC1-SORT1 rs599839, CDKN2A-CDKN2B rs1333049; and PON1 rs662) with type 2 diabetes (T2D) and quantitative serum lipids using a case-control cohort from our Sikh Diabetes Study (SDS). We genotyped these SNPs on 2,194 subjects including 1,379 T2D cases and 815 controls. Three of the six loci revealed significant association with T2D; CELSR2-PSRC1-SORT1 (rs599839) [OR 1.55; p=0.018], CDKN2A-CDKN2B (rs1333049) [OR 1.18; p=0.013], and PON1 (rs662) [OR 1.24; p=0.021] in an additive model after adjusting for age, gender and body mass index (BMI). We also examined the impact of these variants on quantitative metabolic traits (fasting blood glucose (FBG) and serum lipids) including age, gender, medication and disease status as covariates in the model. The variant near CELSR2-PSRC1-SORT1 (rs599839) was associated with increased FBG levels in controls (β=0.059; p=0.009). The strongest association with HDL cholesterol was observed at rs3764241 in the CETP gene (β=0.072; p=0.002), where HDL levels were increased by 3.51 mg/dl in two ‘A’ allele carriers vs. no ‘A’ carriers. The ‘A’ allele carriers also revealed a significant decrease in triglycerides (β=−0.063; p=0.013) and VLDL cholesterol (β=−0.08; p=0.009). Our data also replicated the association of ‘T’ allele of APOA5-A4-C3-A1 rs12286037 with increased triglyceride levels (β=0.089; p=0.001). However, we could not confirm the association of APOE-C1-C4 (rs4420638) variant with any lipid or T2D-related trait. The replication of positive association of most of these loci in an independently ascertained sample from South Asia strongly confirm the underlying role of these genes in the pathophysiology of hyperlipidemia, T2D and cardiovascular disease. As the overall contribution of these loci for affecting serum lipids in this study was ≤ 3%, perhaps the causal SNPs in these genes with larger effects still need to be identified. Further dense genotyping and deep sequencing may help identifying putative functional loci of therapeutic importance.