NEW YORK – Researchers have identified through a genome-wide association study half a dozen loci associated with alcohol consumption.
Excessive alcohol consumption is the result of a combination of genetic factors as well as non-genetic factors like gender and poverty. Previous twin and other studies have found that genetic factors account for about half the variance in alcohol use disorder. While risk alleles at ADH and ALDH have been well replicated, other implicated loci like SNPs near KLB, SLC39A8, and CADM2 have not been well replicated.
Through a genome-wide association study, researchers from the University of Liverpool uncovered a number of genetic loci linked to alcohol consumption. As they reported Wednesday in Science Advances, the researchers further examined the functional relevance of these loci in Caenorhabditis elegans models and teased out causal links and shared genetic architecture with other conditions, including gout, hypertension, and lung cancer.
"Our study offers insight into genes, pathways, and relationships for disease risk associated with high alcohol consumption," first author Andrew Thompson said in a statement.
In a GWAS of 21,967 cases and 103,282 controls from the UK Biobank, the researchers searched for genetic loci associated with heavy alcohol consumption, which they defined as more than 35 alcohol units per week for women and 50 alcohol units per week for men. They then carried forward the lead SNPs they identified into a replication analysis using Genetic Epidemiology Research in Adult Health and Aging data. In a meta-analysis of UK Biobank and GERA data, six loci reached genome-wide significance for association with alcohol consumption: loci in ADH1B, KLB, BTF3P13, GCKR, SLC39A8, and DRD2.
Following conditional analyses, the researchers identified an additional signal at the SLC39A8 as well as other SNPs in or near ADH1A, ADH1B, ADH4, ADH5, TSPAN5, and EIF4E. They also noted that the signal at BTF3P13 did not reach locus-wide significance in the conditional analysis, suggesting that it is a false positive.
In C. elegans, loss-of-function mutations in or RNAi knock-down of ADH1B, KLB, GCKR, and SLC39A8 influenced how ethanol affected the worms. This suggested that these genes do have an effect on alcohol response.
Some of these loci have previously been linked to alcohol-related phenotypes. For instance, the lead SNP at KLB has been associated with alcohol consumption, and the one linked at DRD2 has been linked to addiction phenotypes as well as alcohol use disorder.
Three of the six variants the researchers validated in the UK Biobank and GERA datasets were identified through the Genotype-Tissue Expression database as expression quantitative trait loci. One, for instance, appeared to affect the expression of ANNK1 and TTC12 in various tissues, while another influenced RFC1 expression in the cerebellar hemisphere and skeletal muscle but UDGH expression in blood.
The researchers examined these variants within the GeneATLAS database, which harbors data on associations between traits and gene variants. This indicated that these SNPs contribute to a number of diseases, including alcohol-related ones like alcohol dependence and alcoholic liver disease, but also gout, hypertension, and diabetes.
They further submitted more than three dozen loci associated with heavy drinking in the UK Biobank dataset to the Reactome Knowledgebase for pathway analysis. This implicated six pathways, including PI3K signaling and PI3K/AKT pathways, which are often associated with cancer.
Through Mendelian randomization, the researchers examined the causal relationship between their heavy drinker case-control phenotype and 111 traits and outcomes, finding two lung cancer outcomes that reached significance. Alcohol, the researchers noted, is a known carcinogen.
"Further investigation is required, however, to realize the potential of these outcomes and result in meaningful population — or clinical-level impact," Thompson and his colleagues wrote in their paper.