Cis-regulation of antisense non-coding RNA at the JAZF1 locus in type 2 diabetes.

Abstract

Several genomic loci of type 2 diabetes (T2D) nominated in genome-wide association studies (GWAS) have been suggested to regulate metabolism in muscle. However, a large portion of the genetic risk and the underlying regulation remain unexplained. The present study aimed to localize the potentially functional regions or genes at juxtaposed with another zinc finger protein 1 (JAZF1) locus and interpret their possible biological mechanisms in the muscle of T2D. Seven GWAS datasets including 21,897 T2D patients and 32,710 healthy controls of 772 SNPs within JAZF1 locus were meta-analysed using unconditional logistic regression. The Sherlock and GTEx protal online algorithms were implemented to show the significant colocalizations. Multiple omics data were integrated to predict the potential biological functions of JAZF1-AS1 in muscle. The cis regulation of JAZF1-AS1 was analysed using in vitro cloning in Human skeletal muscle cells (HSkMC). With a cross-population meta-analysis of seven GWAS, we identified a linkage disequilibrium (LD) block within intron 1 of JAZF1 that was significantly associated with T2D (false discovery rate < 0.05). The colocalization analysis showed a significant association between genetically determined expression of JAZF1 in skeletal muscle and T2D with a strong probability of colocalization (PP4 = 75.09%). This region also encodes the upstream regulatory region of the antisense non-coding RNA JAZF1-AS1. Expression-quantitative trait loci analysis detected a regulatory SNP within this LD block, rs864745, which is associated with the expression of JAZF1-AS1 and JAZF1. With in vitro cloning, we further reported the role of JAZF1-AS1 in cis-regulating JAZF1 by directly forming RNA double strands. Downregulation of JAZF1, caused by JAZF1-AS1 depletion, inhibited the glucose uptake and lipid oxidation in skeletal muscle. The present study proposes a strategy for identifying a novel T2D gene at the reported locus and generating a model in which polymorphisms at JAZF1 influence T2D risk through antisense-mediated gene regulation.