A Genome-Wide Association Study of IVGTT-Based Measures of First-Phase Insulin Secretion Refines the Underlying Physiology of Type 2 Diabetes Variants

Abstract

Understanding the physiological mechanisms by which common variants predispose to type 2 diabetes requires large studies with detailed measures of insulin secretion and sensitivity. Here we performed the largest genome-wide association study of first-phase insulin secretion, as measured by intravenous glucose tolerance tests, using up to 5,567 individuals without diabetes from 10 studies. We aimed to refine the mechanisms of 178 known associations between common variants and glycemic traits and identify new loci. Thirty type 2 diabetes or fasting glucose–raising alleles were associated with a measure of first-phase insulin secretion at P < 0.05 and provided new evidence, or the strongest evidence yet, that insulin secretion, intrinsic to the islet cells, is a key mechanism underlying the associations at the HNF1A, IGF2BP2, KCNQ1, HNF1B, VPS13C/C2CD4A, FAF1, PTPRD, AP3S2, KCNK16, MAEA, LPP, WFS1, and TMPRSS6 loci. The fasting glucose–raising allele near PDX1, a known key insulin transcription factor, was strongly associated with lower first-phase insulin secretion but has no evidence for an effect on type 2 diabetes risk. The diabetes risk allele at TCF7L2 was associated with a stronger effect on peak insulin response than on C-peptide–based insulin secretion rate, suggesting a possible additional role in hepatic insulin clearance or insulin processing. In summary, our study provides further insight into the mechanisms by which common genetic variation influences type 2 diabetes risk and glycemic traits.

Fine-mapping type 2 diabetes loci to single-variant resolution using high-density imputation and islet-specific epigenome maps

“…For most of these loci, orthogonal data (e.g. associations with continuous metabolic traits3,36,37, cis -eQTL data39) are consistent with a role in islet function. In contrast, at six signals, including three which are likely, on physiological grounds, to be acting, at least partly, through effects on islets, we saw reductions (10% to 76%) in the lead variant PPA after islet-annotation-informed fGWAS (Supplementary Table 8).…”

“…This variant overlaps an islet promoter that lies 3’ to ANK1 but 5’ to the transcription factor NKX6.3 (Supplementary Figure 6). The T2D-risk allele at the rs13262861 signal shows a directionally-consistent association with both in vivo measures of reduced insulin secretion3,37,40, and a cis -eQTL for reduced NKX6-3 expression in human islets32. Members of the NKX6 transcription factor family (including NKX6.3) are implicated in islet development and function44.…”

“…For these analyses, we focused on regulatory annotations from human islets because: (a) most established T2D-risk variants are considered, given patterns of association to continuous metabolic traits, to act through primary effects on beta-cell function3,36,37; (b) previous studies have shown that the strongest signal for regulatory enrichment at T2D association signals is seen for islet-specific regulatory elements31,34, a view supported by the annotation overlaps of the high-PPA (i.e. PPA>80%) variants described above; and (c) we had access to particularly extensive, high-resolution epigenomic and chromatin state annotation maps for human islets combining available histone modification and transcription factor ChIP-seq, ATAC-seq and whole genome bisulphite sequencing33.…”

Patterns of differential gene expression in a cellular model of human islet development, and relationship to type 2 diabetes predisposition

“…The horizontal grey dashed line marks the 5% significance threshold. T2D, type 2 diabetes; DE, definitive endoderm; GT, primitive gut tube; PF, posterior foregut; PE, pancreatic endoderm; EP, endocrine precursor; EN, endocrine-like cells; BLC, beta-like cells evidence from physiological studies points most emphatically to risk of type 2 diabetes mediated via reduced insulin secretion (ESM Table 11) [27,28]. In this analysis, enrichment for genes differentially expressed at the beta-like cell stage became more significant (permuted p value =0.007; Fig.…”

tRNA modifications and islet function

“…For example, a study in a Danish cohort showed that homozygous carriers of the risk allele of rs7756992 had ~22% lower insulin secretion than non‐risk carriers . The inhibitory effect on insulin secretion has also been reported in other risk CDKAL1 SNPs, including rs7754840, rs9368222, rs742642 and rs2206734 . Hyperglycaemic clamp studies have been performed to examine whether the risk alleles affect first‐ or second‐phase glucose‐stimulated insulin secretion in humans .…”