Abstract 4369047: Genome-wide analysis of human lipidome identifies new metabolite QTLs linked with diabetes and cardiovascular disease in Asian Indians and Europeans

Background: Despite the achievements of genome-wide association studies (GWAS), the genetic mechanisms that predispose individuals to type 2 diabetes (T2D) and its cardiovascular (CV) complications remain poorly understood. The prevalence of T2D is three to six times higher among South Asian Indians compared to Europeans. High-throughput metabolite profiling techniques have advanced rapidly, evolving from focusing on single-gene/single-metabolite associations to a genome-wide/metabolome-wide approach. However, data is limited on non-white populations. We conducted an integrated genome-wide lipidomic analysis and Mendelian randomization (MR) to identify gene variants associated with lipid metabolites and determine the causal relationships between circulating lipid species and T2D, as well as CV risk factors, using a discovery cohort of Asian Indian Diabetic Heart Study (AIDHS) (3871) and validation cohorts comprising over 1.14 million European and South Asian Individuals from multiple independent studies. Methods: Untargeted lipidomic analysis was performed on blood samples of AIDHS individuals using liquid chromatography and high-resolution mass spectrometry (LC-MS/MS). We analyzed the association of 269 metabolites with T2D and CVD by conducting a metabolome-GWAS using genotyped and imputed data from 20,006,524 variants in discovery and replication cohorts. Results: Our study identified 261 SNP-metabolite associations of genome-wide significance, of which 59 association signals (p ranging from 8.4×10 −8 to 1.8×10 -18 ) were from genes involved in atherosclerosis, fatty acid biosynthesis, inflammation, insulin signaling, and apoptosis pathways. Two-sample MR and sensitivity analysis using independent of 1,141,060 individuals from European and South Asian populations from the UK Biobank, DIAGRAM, DIAMANT, and Australia, identified species of fatty acids, PC, and LPCs contributing to increased or decreased risk for T2D and CVD. Conclusion: In this first metabolome-GWAS in Asian Indians, we report new mQTLs beyond FADS1/2 associated with lipid subclasses not reported previously. MR analysis in independent datasets identified new molecular signatures of circulating lipid and fatty acid metabolism that are causally associated with inflammation and vascular ischemic disease in dysregulated insulin metabolism and T2D. Funding: The AIDHS study was supported by NIH grants R01DK082766 and R01DK118427 (NIDDK) and grants from the Presbyterian Health Foundation of Oklahoma.