Genome-wide association studies in the Japanese population identify seven novel loci for type 2 diabetes

Abstract

Genome-wide association studies (GWAS) have identified more than 80 susceptibility loci for type 2 diabetes (T2D), but most of its heritability still remains to be elucidated. In this study, we conducted a meta-analysis of GWAS for T2D in the Japanese population. Combined data from discovery and subsequent validation analyses (23,399 T2D cases and 31,722 controls) identify 7 new loci with genome-wide significance (P<5 × 10(-8)), rs1116357 near CCDC85A, rs147538848 in FAM60A, rs1575972 near DMRTA1, rs9309245 near ASB3, rs67156297 near ATP8B2, rs7107784 near MIR4686 and rs67839313 near INAFM2. Of these, the association of 4 loci with T2D is replicated in multi-ethnic populations other than Japanese (up to 65,936 T2Ds and 158,030 controls, P<0.007). These results indicate that expansion of single ethnic GWAS is still useful to identify novel susceptibility loci to complex traits not only for ethnicity-specific loci but also for common loci across different ethnicities.

Figure 1. Outline of the present study.

(a) GWAS meta-analysis and subsequent validation analysis. (b) Follow-up analyses for seven novel susceptibility loci for T2D in populations other than Japanese. SNP, single nucleotide polymorphism; MAF, minor allele frequency; HWE P, Hardy-Weinberg Equilibrium test P; RSQ, r square.

Figure 2. Regional association plots of the two-stage GWAS meta-analysis for seven novel T2D susceptibility loci in the Japanese.

Each plot shows −log₁₀ P-values against the chromosomal positions of SNPs in the specific region. The SNP with the strongest association signal (lead SNP) in each locus is represented as a purple diamond; the other SNPs are coloured according to the extent of LD with the lead SNP. Estimated recombination rates from the hg19/1000 Genomes Project March 2012 East Asian references are shown as light-blue lines. (a) CCDC85A, (b) FAM60A, (c) DMRTA1, (d) ASB3, (e) ATP8B2, (f) MIR4686 and (g) INAFM2. *Gene position for INAFM2 was added manually.

Figure 3. Prioritized biological T2D risk genes.

The gene score for each gene was calculated by summing up the number of criteria satisfied (filled red box indicates criterion satisfied; 40 genes with a score ≥2 out of 286 genes included in the T2D risk loci were defined as ‘biological candidate genes'; see Supplementary Data 4). Filled blue boxes indicate the nearest gene to the T2D risk SNP. Filled green boxes indicate overlap with drug target genes.

Figure 4. Connection of biological T2D risk genes to drug targets.

Representative connections between T2D risk SNPs (blue), T2D biological genes (green), drug target genes (purple) and targeted drugs are shown (see Supplementary Table 21). *Compounds under clinical trial; **see Supplementary Table 23.