Testing the association of novel meta-analysis-derived diabetes risk genes with type II diabetes and related metabolic traits in Asian Indian Sikhs

A recent meta-analysis on three genome-wide association (GWA) scans identified six loci (NOTCH2, THADA, ADAMTS9, JAZF1, CDC123/CAMKID and TSPAN8/LGRS) highly associated with type II diabetes (T2D) in Caucasians. This investigation seeks to confirm this association with diabetes and related metabolic traits in Khatri Sikh diabetics of North India. We genotyped highly significant variants from each locus in a case-control cohort consisting of 680 T2D cases and 637 normoglycemic (NG) controls. Only CDC123/CAMKID (rs12779790) replicated earlier evidence of association with T2D under a dominant model (odds ratio (OR): 1.27; 95% confidence interval (CI): 1.02-1.57; P = 0.031) during initial testing. However, we could not confirm this association using multiple testing corrections. In a multiple linear-regression analysis, the same variant in the CDC123/CAMKID revealed a marked decrease in fasting insulin levels among 'G' (risk) allele carriers independently in NG controls (P = 0.030) and in T2D cases (P = 0.009), as well as in the combined sample (P = 0.003) after adjusting for covariates. Evidence of impaired beta-cell function was also observed among 'G' (risk) allele carriers in T2D cases (P 0.008) and in a combined cohort (P = 0.026). Our data could not confirm the role of the remaining variants with risk either for T2D or quantitative phenotypes measuring insulin secretion or insulin resistance. These findings suggest that CDC123/CAMKID could be a major risk factor for the development of T2D in Sikhs by affecting b-cell function. To our knowledge, this is the first study reporting the role of recently emerging loci in this high-risk population from the South Asian subcontinent. Journal of Human Genetics (2009) 54, 162-168; doi: 10.1038/jhg.2009.7; published online 27 February 2009