IRS1 gene variants, dysglycaemic metabolic changes and type-2 diabetes risk

Abstract

Background and aims: A recent genome-wide association study identified rs2943641C > T, 500 kb from the insulin receptor substrate-1 gene (IRS1), as a type-2 diabetes (T2D) susceptibility locus. We aimed to replicate this association by meta-analysis and examine whether common variants within IRS1, present on the HumanCVD BeadChip, were associated with T2D risk.

Methods and results: We genotyped rs2943641 in 2389 prevalent or incident T2D patients and 6494 controls from two prospective and three case studies based in UK and in the European Atherosclerosis Research Study-II (EARSII; n = 714). Thirty-three IRS1 variants had been genotyped in the prospective Whitehall-II study (n = 4752) using the HumanCVD BeadChip. In a fixed-effects meta-analysis of the UK study cohorts rs2943641T allele was associated with 6% lower risk of T2D (p = 0.18), with T-allele carriers having an odds ratio (OR) of 0.89 (95% confidence interval [CI]: 0.80-1.00, p = 0.056) compared to CC subjects. The T-allele was also associated with lower fasting insulin and homeostasis model assessment index of insulin resistance in Whitehall-II and with lower post-load insulin after an oral glucose tolerance test in EARSII (all p < 0.05). None of the IRS1 variants on the chip showed linkage disequilibrium with rs2943641. In silico analysis with follow-up genotyping (total n = 9313) identified that the rare allele of the IRS1 promoter variant rs6725556A > G showed association with reduced T2D risk (OR per G-allele: 0.82, 95%CI: 0.69-0.96, p = 0.015).

Conclusions: We confirm the association of rs2943641T with T2D protection. There is a possible independent effect on risk of a putative IRS1 promoter variant.

Figure 1

Plasma insulin concentration after an OGTT by rs2943641 genotype in European Atherosclerosis Research Study-II. Data are presented as mean ± SEM in carriers of the CC (white circles), CT (white squares) and TT (black triangles) genotypes. ∗p = 0.004, ∗∗p = 0.03, ∗∗∗p = 0.09, for additive model adjusting for age, “case”/“control” status and region. P-values for AUC insulin: p = 0.005 (additive), p = 0.003 (dominant), confirming result reported by Rung et al. .

Figure 2

Odds ratio (OR) and 95% confidence interval (CI) for T2D per rs2943641T allele (additive genetic model) in the various populations of this study. Heterogeneity chi-squared = 0.73 (d.f. = 3), p = 0.87; I-squared (variation in OR attributable to heterogeneity) = 0.0%. ORs were adjusted for age and where applicable for gender and centre. Abbreviations: NPHSII = Northwick Park Heart Study-II; WHII = Whitehall-II; EW = European Whites (UDACS, EDSC and PREDICT combined); IA = Indian Asians (UDACS, EDSC and PREDICT combined).

Figure 3

Odds ratio (OR) and 95% confidence interval (CI) for T2D per IRS1 rs6725556G allele (additive genetic model) in the various populations of this study. Heterogeneity chi-squared = 4.36 (d.f. = 3), p = 0.23; I-squared = 31.1%. ORs were adjusted for age and where applicable for gender and centre. Abbreviations: NPHSII = Northwick Park Heart Study-II; WHII = Whitehall-II study; EW = European Whites; IA = Indian Asians.