Association of Cardiometabolic Genes with Arsenic Metabolism Biomarkers in American Indian Communities: The Strong Heart Family Study (SHFS)

Abstract

Background: Metabolism of inorganic arsenic (iAs) is subject to inter-individual variability, which is explained partly by genetic determinants.

Objectives: We investigated the association of genetic variants with arsenic species and principal components of arsenic species in the Strong Heart Family Study (SHFS).

Methods: We examined variants previously associated with cardiometabolic traits (~ 200,000 from Illumina Cardio MetaboChip) or arsenic metabolism and toxicity (670) among 2,428 American Indian participants in the SHFS. Urine arsenic species were measured by high performance liquid chromatography-inductively coupled plasma mass spectrometry (HPLC-ICP-MS), and percent arsenic species [iAs, monomethylarsonate (MMA), and dimethylarsinate (DMA), divided by their sum × 100] were logit transformed. We created two orthogonal principal components that summarized iAs, MMA, and DMA and were also phenotypes for genetic analyses. Linear regression was performed for each phenotype, dependent on allele dosage of the variant. Models accounted for familial relatedness and were adjusted for age, sex, total arsenic levels, and population stratification. Single nucleotide polymorphism (SNP) associations were stratified by study site and were meta-analyzed. Bonferroni correction was used to account for multiple testing.

Results: Variants at 10q24 were statistically significant for all percent arsenic species and principal components of arsenic species. The index SNP for iAs%, MMA%, and DMA% (rs12768205) and for the principal components (rs3740394, rs3740393) were located near AS3MT, whose gene product catalyzes methylation of iAs to MMA and DMA. Among the candidate arsenic variant associations, functional SNPs in AS3MT and 10q24 were most significant (p < 9.33 × 10-5).

Conclusions: This hypothesis-driven association study supports the role of common variants in arsenic metabolism, particularly AS3MT and 10q24. Citation: Balakrishnan P, Vaidya D, Franceschini N, Voruganti VS, Gribble MO, Haack K, Laston S, Umans JG, Francesconi KA, Goessler W, North KE, Lee E, Yracheta J, Best LG, MacCluer JW, Kent J Jr., Cole SA, Navas-Acien A. 2017. Association of cardiometabolic genes with arsenic metabolism biomarkers in American Indian communities: the Strong Heart Family Study (SHFS). Environ Health Perspect 125:15-22; http://dx.doi.org/10.1289/EHP251.

Figure 1

Regional association plot at 10q24 of percent arsenic species. Index single nucleotide polymorphism (SNP) rs12768205 nearby associations according to human genome build 18 for percent inorganic arsenic (iAs%) in top panel, percent monomethylarsonate (MMA%) in middle panel, and percent dimethylarsinate (DMA%) in bottom panel. The solid red line is the MetaboChip-wide significance threshold at –log(4.13 × 10^–7) or 6.38. The dashed orange line is the suggestive MetaboChip-wide linkage disequilibrium (LD) threshold at –log(7.77 × 10^–7). Chr, chromosome.

Figure 2

Regional association plot at 10q24 of principal components of arsenic species. Index single nucleotide polymorphisms (SNPs) for principal components (PCs) (top panel PC1 rs3740394, bottom panel PC2 rs3740393) and nearby associations according to human genome build 18. The solid red line is the MetaboChip-wide significance threshold at –log(4.13 × 10^–7) or 6.38. The dashed orange line is the suggestive MetaboChip-wide linkage disequilibrium (LD) threshold at –log(7.77 × 10^–7).

Figure 3

Distribution of percent arsenic species by rs12768205 genotype. Index single nucleotide polymorphism (SNP) rs12768205 for percent arsenic species principal components shows separation of distribution of percent inorganic arsenic (iAs%), percent monomethylarsonate (MMA%) and percent dimethylarsinate (DMA%) by genotype. Of the 2,428 participants, the distribution of genotypes is homozygous major AA (1,245), heterozygous AG (982), and homozygous minor GG (201).