Lack of significant effects of the type 2 diabetes susceptibility loci JAZF1, CDC123 / CAMK1D, NOTCH2, ADAMTS9, THADA, and TSPAN8 / LGR5 on diabetes and quantitative metabolic traits

Recently, several novel loci reaching genome-wide significance levels for type 2 diabetes (T2D) were identified through a meta-analysis of three genome-wide scans and large-scale follow-up. The aim of our study was to investigate the association of these loci with T2D and related subphenotypes in two cohorts from Germany. We performed an association study of 9 SNPs in or around JAZF1, CDC123/CAMK1D, NOTCH2, BCL11A, ADAMTS9, VEGFA, DCD, THADA, and TSPAN8/LGR5 with T2D and related quantitative traits (fasting insulin and glucose, indices derived from OGTT) in the isolated population of Sorbs (205 cases and 695 controls) and in a mixed German population (Leipzig) (938 subjects with and 918 without T2D). None of the variants was associated with T2D, but the meta-analysis of both cohorts revealed a modest trend of association of rs7578597 in THADA with T2D (p=0.055). Furthermore, Sorbian subjects homozygous for the rs7578597 T-allele had lower mean 30-minute plasma insulin when compared with carriers of the C-allele (p<0.05). The T-allele was also nominally associated with higher fasting plasma glucose in the Leipzig cohort (p<0.05). Although several other SNPs showed some evidence for association with T2D-related traits the effects were not replicated within our study. Associations of the T2D-risk alleles with T2D or related subphenotypes were overall very weak in the 2700 subjects studied. This is compatible with the modest effect size of these second sweep variants, which will require large-scale association studies on quantitative traits to clarify their role in the pathophysiology of T2D.