Genetic association between intronic variants in AS3MT and arsenic methylation efficiency is focused on a large linkage disequilibrium cluster in chromosome 10

Abstract

Differences in arsenic metabolism are known to play a role in individual variability in arsenic-induced disease susceptibility. Genetic variants in genes relevant to arsenic metabolism are considered to be partially responsible for the variation in arsenic metabolism. Specifically, variants in arsenic (3+ oxidation state) methyltransferase (AS3MT), the key gene in the metabolism of arsenic, have been associated with increased arsenic methylation efficiency. Of particular interest is the fact that different studies have reported that several of the AS3MT single nucleotide polymorphisms (SNPs) are in strong linkage-disequilibrium (LD), which also extends to a nearby gene, CYP17A1. In an effort to characterize the extent of the region in LD, we genotyped 46 SNPs in a 347,000 base region of chromosome 10 that included AS3MT in arsenic-exposed subjects from Mexico. Pairwise LD analysis showed strong LD for these polymorphisms, represented by a mean r(2) of 0.82, spanning a region that includes five genes. Genetic association analysis with arsenic metabolism confirmed the previously observed association between AS3MT variants, including this large cluster of linked polymorphisms, and arsenic methylation efficiency. The existence of a large genomic region sharing strong LD with polymorphisms associated with arsenic metabolism presents a predicament because the observed phenotype cannot be unequivocally assigned to a single SNP or even a single gene. The results reported here should be carefully considered for future genomic association studies involving AS3MT and arsenic metabolism.

Keywords: AS3MT; Arsenic; Linkage Disequilibrium; SNP; polymorphism.

Figure 1

Linkage-disequilibrium profiles for the genotyped SNPs in HapMap a) Asian population, and b) European population. LD as r² for SNP pairs is shown inside the squares. Box gray-scale color indicates r² value varying from white (r²=0) to black (r²=1). A schematic illustration of the genomic structure of the genomic region studied is above each LD plot, arrowheads on genes indicate the gene orientation.

Figure 2

Linkage-disequilibrium profile for the genotyped SNPs in the Mexican population in this study. LD as r² for SNP pairs is shown inside the squares. Box gray-scale color indicates r² value varying from white (r²=0) to black (r²=1). A schematic illustration of the genomic structure of the genomic region studied is above the LD plot, arrowheads on genes indicate the gene orientation.

Figure 3

uDMA/uMMA ratio stratified by genotype for AS3MT polymorphism rs10883795. uDMA/uMMA ratio is shown on the Y-axis as the geometric mean, with error bars outlining the 95% confidence interval of the geometric mean values. Genotype groups are shown on the X-axis as homozygous non-variant, heterozygous and homozygous variant (TT, TC and CC respectively). P = 0.00074 (codominant inheritance, linear regression).

Figure 4

Genetic association between the chromosome 10 genomic variants and arsenic metabolism. Negative log-converted P-values for statistical tests of association between genetic variation and uDMA/uMMA ratio are shown (Y-axis) for all the study polymorphisms (X-axis), as well as their location within the genomic region (Lines from bar-graphed polymorphisms leading to genomic map at top of figure). Horizontal line shows P = 0.05.