Genetic fine mapping and genomic annotation defines causal mechanisms at type 2 diabetes susceptibility loci

Abstract

We performed fine mapping of 39 established type 2 diabetes (T2D) loci in 27,206 cases and 57,574 controls of European ancestry. We identified 49 distinct association signals at these loci, including five mapping in or near KCNQ1. 'Credible sets' of the variants most likely to drive each distinct signal mapped predominantly to noncoding sequence, implying that association with T2D is mediated through gene regulation. Credible set variants were enriched for overlap with FOXA2 chromatin immunoprecipitation binding sites in human islet and liver cells, including at MTNR1B, where fine mapping implicated rs10830963 as driving T2D association. We confirmed that the T2D risk allele for this SNP increases FOXA2-bound enhancer activity in islet- and liver-derived cells. We observed allele-specific differences in NEUROD1 binding in islet-derived cells, consistent with evidence that the T2D risk allele increases islet MTNR1B expression. Our study demonstrates how integration of genetic and genomic information can define molecular mechanisms through which variants underlying association signals exert their effects on disease.

Figure 1. FOX2A bound sites are a genomic marker of T2D risk variants

(A) Variants in ChIP-seq binding sites for 165 proteins were tested for enrichment of posterior probabilities compared to variants in shifted sites. Variants in FOXA2 ChIP-seq sites were significantly enriched (p<0.00030). (B) FOXA2 ChIP-seq sites were partitioned based on overlap with other genomic features. There was stronger enrichment in: (i) FOXA2 sites overlapping a ChIP-seq site for another protein compared to unique sites; (ii) sites identified in primary islets compared to HepG2 or primary liver cells; and (iii) sites overlapping islet enhancers compared to those that did not (**p<0.00030; *p<0.05). (C) Variants at each signal were tested for FOXA2 enrichment. Nineteen signals had greater enrichment than expected compared to shifted sites; at 15 signals this enrichment was nominally-significant (p<0.05). (D) FOXA2-bound variants disrupting recognition motifs have an increased probability of being causal.

Figure 2. The lone variant in the 99% credible set at the MTNR1B locus affects FOXA2-bound enhancer activity

(A) The intronic variant, rs10830963, has 99.8% probability of driving the association signal at the MTRN1B locus. This variant overlaps a FOXA2 binding site, and the risk allele G is predicted to create a de novo recognition motif, which closely matches the NEUROD1 consensus. (B) Electrophoretic mobility shift assay of a 25bp fragment surrounding both alleles in EndoC-βH1 cell extracts. Proteins were bound to both alleles. In the presence of a NEUROD1 antibody, only the risk allele band was super-shifted, and in the presence of an unlabelled NEUROD1 consensus probe, the signal was competed away. NE: nuclear extract. (C, D) The 224bp sequence surrounding each allele was cloned into a luciferase reporter construct containing a minimal promoter and tested for luciferase activity in (C) EndoC-βH1 and (D) HepG2 cells (n=3 for each cell type). Results are presented as mean ± standard error. The risk allele had significantly increased enhancer activity over the protective allele in both forward and reverse orientations in both cell types.