Association and ancestry analysis of sequence variants in ADH and ALDH using alcohol-related phenotypes in a Native American community sample

Abstract

Higher rates of alcohol use and other drug-dependence have been observed in some Native American (NA) populations relative to other ethnic groups in the US. Previous studies have shown that alcohol dehydrogenase (ADH) genes and aldehyde dehydrogenase (ALDH) genes may affect the risk of development of alcohol dependence, and that polymorphisms within these genes may differentially affect risk for the disorder depending on the ethnic group evaluated. We evaluated variations in the ADH and ALDH genes in a large study investigating risk factors for substance use in a NA population. We assessed ancestry admixture and tested for associations between alcohol-related phenotypes in the genomic regions around the ADH1-7 and ALDH2 and ALDH1A1 genes. Seventy-two ADH variants showed significant evidence of association with a severity level of alcohol drinking-related dependence symptoms phenotype. These significant variants spanned across the entire 7 ADH gene cluster regions. Two significant associations, one in ADH and one in ALDH2, were observed with alcohol dependence diagnosis. Seventeen variants showed significant association with the largest number of alcohol drinks ingested during any 24-hour period. Variants in or near ADH7 were significantly negatively associated with alcohol-related phenotypes, suggesting a potential protective effect of this gene. In addition, our results suggested that a higher degree of NA ancestry is associated with higher frequencies of potential risk variants and lower frequencies of potential protective variants for alcohol dependence phenotypes.

Keywords: admixture; alcohol metabolizing enzymes; alcoholism; low coverage sequencing.

Figure 1

Global and local Native American ancestry estimates. (A) Native American (NA) ancestry distribution in the samples. Global ancestry was estimated using variants distributed across the whole genome; local ancestry was estimated using variants in and around each of the candidate regions (ADH, ALDH1A1, ALDH2). (B) Local NA ancestry of the ADH region plotted against global NA ancestry of all samples, each point representing an individual (Pearson's r = 0.55).

Figure 2

Samples stratified by degree of local NA ancestry and the severity level of alcohol dependence drinking symptoms. (A) Alcohol dependence drinking symptoms of samples with low and high NA heritages. (B/D) Alternative allele frequencies of 69/3 significant variants positively/negatively associated with dependence drinking in each sample subset. Each line represents a variant. (C) Linear regression of allele frequencies of significant variants (positively associated with dependence drinking) in samples of low and high NA heritages onto the number of dependence drinking symptoms.

Figure 3

Samples stratified by degree of local NA ancestry and the alcohol dependence diagnosis. (A) Alcohol dependence of samples with low and high NA heritages. (B) Alternative allele frequencies of 2 significant variants positively associated with alcohol dependence in each sample subset.

Figure 4

Samples stratified by degree of local NA ancestry and the maximum drinks (md) ever consumed in a 24-hour period. (A) Maximum drinks in 24-hour of samples with low and high NA heritages. (B) Alternative allele frequencies of 10 significant variants negatively associated with max drink phenotype in each sample subset. (C) Allele counts of a potentially protective variant in each sample subset (md = maximum drinks).