Variants associated with type 2 diabetes identified by the transethnic meta-analysis study: assessment in American Indians and evidence for a new signal in LPP

Abstract

Aim/hypothesis: A recent genome-wide trans-ancestry meta-analysis identified seven new loci associated with type 2 diabetes. We assessed the replication of the seven lead single nucleotide polymorphisms (SNPs) and evaluated these loci for additional signals in American Indians.

Methods: Seven SNPs were genotyped in 7,710 individuals from a longitudinally studied American Indian population, and associations with type 2 diabetes, BMI and related phenotypes were assessed. Previous genome-wide association study (GWAS) data from these individuals were used to screen for additional type 2 diabetes signals at these loci. A variant independent of the trans-ancestry meta-analysis was identified within LPP, and its replication was assessed in an additional 3,106 urban American Indians.

Results: SNP rs6813195 near to TMEM154 was nominally associated with type 2 diabetes (p = 0.01, OR 1.12 [95% CI 1.03, 1.22]) and adiposity: the type 2 diabetes risk allele was associated with a lower percentage body fat (β = -1.451%, p = 4.8 × 10(-4)). Another SNP, rs3130501 near to POU5F1-TCF19, was associated with BMI (β = -0.012, p = 0.004), type 2 diabetes adjusted for BMI (p = 0.02, OR 1.11 [95% CI 1.02, 1.22]), 2 h glucose concentrations (β = 0.080 mmol/l, p = 0.02) and insulin resistance estimated by homeostatic model (β = 0.039, p = 0.009). The independent variant identified at the LPP locus in our American Indian GWAS for type 2 diabetes was replicated in the additional samples (all American Indian meta-analysis, p = 8.9 × 10(-6), OR 1.29 [95% CI 1.15, 1.45]).

Conclusions/interpretation: For two of the seven newly identified variants, there was nominal evidence for association with type 2 diabetes and related traits in American Indians. Identification of an independent variant at the LPP locus suggests the existence of more than one type 2 diabetes signal at this locus.

Fig. 1

Association of rs7649407 in LPP with type 2 diabetes (T2D) in American Indians from the Gila River Indian Community, and replication in cross-sectionally studied urban American Indians. (a) Associations of 46 tag SNPs across the region (Chr3:187,740 kb–188,808 kb) encompassing LPP with T2D in American Indians. Results were analysed in full-heritage Pima Indians (diamonds; N = 3,625), all American Indians from the Gila River Indian community including full-heritage Pima Indians (triangles; N = 7,710) and combined American Indians and urban American Indians (squares; N = 10,816); p values were adjusted for age, sex, birth year, family membership and admixture estimates. (b) T2D prevalence in American Indians from the Gila River Indian Community (p = 4.1 × 10⁻⁴, OR 1.29 [1.12, 1.48]) and urban American Indians (p = 0.007, OR 1.31 [1.08, 1.59]) by LPP SNP rs7649407 (A/G) genotype. All American Indian meta-analysis: p = 8.9 × 10^–6, OR 1.29 [95% CI 1.15, 1.45]. ORs and p values were adjusted for age, sex, birth year, family membership and admixture estimates. Black bars, GG genotype; grey bars, AG genotype; white bars, AA genotype. (c) T2D prevalence in different age groups by LPP SNP rs7649407 genotype. Triangles, GG genotype; squares, AG genotype; diamonds, AA genotype