East Asian genome-wide association studies (GWAS) for type 2 diabetes identified 8 loci with genome-wide significance, and 2 loci with a borderline association. However, the associations of these loci except MAEA locus with type 2 diabetes have not been evaluated in independent East Asian cohorts. We performed a replication study to investigate the association of these susceptibility loci with type 2 diabetes in an independent Japanese population.

We genotyped 7,379 Japanese participants (5,315 type 2 diabetes and 2,064 controls) for each of the 9 single nucleotide polymorphisms (SNPs), rs7041847 in GLIS3, rs6017317 in FITM2−R3HDML−HNF4A, rs6467136 near GCCI−PAX4, rs831571 near PSMD6, rs9470794 in ZFAND3, rs3786897 in PEPD, rs1535500 in KCNK16, rs16955379 in CMIP, and rs17797882 near WWOX. Because the sample size in this study was not sufficient to replicate single SNP associations, we constructed a genetic risk score (GRS) by summing a number of risk alleles of the 9 SNPs, and examined the association of the GRS with type 2 diabetes using logistic regression analysis.

With the exception of rs1535500 in KCNK16, all SNPs had the same direction of effect (odds ratio [OR]>1.0) as in the original reports. The GRS constructed from the 9 SNPs was significantly associated with type 2 diabetes in the Japanese population (p = 4.0 × 10^-4, OR = 1.05, 95% confidence interval: 1.02–1.09). In quantitative trait analyses, rs16955379 in CMIP was nominally associated with a decreased homeostasis model assessment of β-cell function and with increased fasting plasma glucose, but neither the individual SNPs nor the GRS showed a significant association with the glycemic traits.

These results indicate that 9 loci that were identified in the East Asian GWAS meta-analysis have a significant effect on the susceptibility to type 2 diabetes in the Japanese population.