Genomewide association study of C-peptide surfaces key regulatory genes in Indians

Insulin is a commonly used measure of pancreatic β-cell function but exhibits a short half-life in the human body. During biosynthesis, insulin release is accompanied by C-peptide at an equimolar concentration which has a much higher plasma half-life and is therefore projected as a precise measure of β-cell activity than insulin. Despite this, genetic studies of metabolic traits haveneglected the regulatory potential of C-peptide for therapeutic intervention of type-2 diabetes. The present study is aimed to search genomewide variants governing C-peptide levels in genetically diverse and high risk population for metabolic diseases-Indians. We performed whole genome genotyping in 877 healthy Indians of Indo-European origin followed by replication of variants with P ≤ 1 × 10^-3 in an independent sample-set of 1829 Indians. Lead-associated signals were also tested in-silico in 773 Hispanics. To secure biological rationale for observed association, we further carried out DNA methylation quantitative trait loci analysis in 233 Indians and publicly available regulatory data was mined. We discovered novel lncRNA gene AC073333.8 with the strongest association with C-peptide levels in Indians that however missed genomewide significance. Also, noncoding genes, RP1-209A6.1 and RPS3AP5; protein gene regulators, ZNF831 and ETS2; and solute carrier protein gene SLC15A5 retained robust association with C-peptide after meta-analysis. Integration of methylation data revealed ETS2 and ZNF831 single-nucleotide polymorphisms as significant meth-QTLs in Indians. All genes showed reasonable expression in the human lung, signifying alternate important organs for C-peptide biology. Our findings mirror polygenic nature of C-peptide where multiple small-effect size variants in the regulatory genome principally govern the trait biology.