Bridging the gap between type 2 diabetes GWAS and therapeutic targets

点击翻译按钮获取中文摘要

Type 2 diabetes (T2D) is a heterogeneous disorder characterized by resistance of hepatic, skeletal muscle and adipose tissues to insulin and a relative deficiency of insulin secretion by pancreatic β cells. T2D has a substantial genetic component, and over the past decade human genetic studies have identified over 400 association signals across diverse populations. However, in most cases the specific variants and genes responsible for these association signals are not known. T2D signals include loci for which functions of the protein products encoded by nearby genes are poorly characterized, the closest known gene is distant, or more than one gene appears to be a plausible biological candidate. Identifying the causal variants, the regulatory gene networks affected by the change in DNA sequence, and the mechanisms by which such variation leads to disease are critical steps toward understanding the genetic architecture of T2D, validating potential drug targets, and developing novel therapeutic strategies. Here, we propose large-scale multi-disciplinary functional genomics projects in islet, liver, adipose and muscle cells to determine the contributions and mechanisms underlying T2D risk-associated variants and their downstream effector transcripts. Throughout the project, we leverage our prior and ongoing generation of genomic data sets and genome-wide and targeted screens for function of variants and genes. To complement these efforts, we will first collect genome-wide array and sequencing-based association study results, identify conditionally distinct association signals and construct credible sets of variants. We propose to link variants to effector transcripts through analyses of genome-wide transcriptomic and epigenomic data, perturbation assays that alter thousands of variant-containing regulatory elements and effector transcripts, perturbations of tens of specific variants, and integrative computational analyses. Next, we propose systematic evaluation of hundreds of potential effector transcripts through use of genome-wide and targeted screens of insulin secretion, lipid accumulation, mitochondrial function, glucose uptake, and differentiation state, with assay selection depending on cell type. Based on these results, we propose focused studies on tens to hundreds of potential effector transcripts to evaluate electrophysiology, gluconeogenesis, lipid metabolism and signaling pathways, and we propose thorough investigation the context-specific mechanism of action of individual genes. Finally, we propose to analyze, integrate, and visualize all data by placing effector transcripts into cell-type and environmental context-specific networks, selecting network nodes as candidate biomarkers and modulation points for drugs, and building a framework to understand the tissue-specific contribution of variants and transcripts to individual disease heterogeneity. Successful completion of these aims will translate T2D association signals into biological insights and therapeutic targets.