Genetic analysis in European ancestry individuals identifies 517 loci associated with liver enzymes

Abstract

Serum concentration of hepatic enzymes are linked to liver dysfunction, metabolic and cardiovascular diseases. We perform genetic analysis on serum levels of alanine transaminase (ALT), alkaline phosphatase (ALP) and gamma-glutamyl transferase (GGT) using data on 437,438 UK Biobank participants. Replication in 315,572 individuals from European descent from the Million Veteran Program, Rotterdam Study and Lifeline study confirms 517 liver enzyme SNPs. Genetic risk score analysis using the identified SNPs is strongly associated with serum activity of liver enzymes in two independent European descent studies (The Airwave Health Monitoring study and the Northern Finland Birth Cohort 1966). Gene-set enrichment analysis using the identified SNPs highlights involvement in liver development and function, lipid metabolism, insulin resistance, and vascular formation. Mendelian randomization analysis shows association of liver enzyme variants with coronary heart disease and ischemic stroke. Genetic risk score for elevated serum activity of liver enzymes is associated with higher fat percentage of body, trunk, and liver and body mass index. Our study highlights the role of molecular pathways regulated by the liver in metabolic disorders and cardiovascular disease.

Fig. 1. Overview of study design and findings.

The figure illustrates the genotype and phenotype quality control (QC) within the UK Biobank (UKB) data. Statistical analysis and replication resulted in 517 loci associated with liver enzymes. PC principal component, SNP single-nucleotide polymorphism, GWAS-genome-wide association studies, LD linkage disequilibrium.

Fig. 2. Overview of ALT, ALP, and GGT loci identified within the UKB study (discovery sample).

Manhattan (MH) plots illustrated have been created based on summary statistics of GWAs on liver enzymes where the x-axis demonstrates chromosome number and the y-axis represents −log 10 (P value) for the association of SNP with liver enzymes. Q–Q plots are illustrated to show the inflation of test statistics using the summary statistics of the liver enzyme GWAS. Where the x-axis represents the expected log (P value). The red line shows the expected results under the null association. Y-axis illustrates the observed log (P value). a MH plot based on ALP GWAS summary statistics. b MH plot based on ALT GWAS summary statistics. c MH plot based on GGT GWAS summary statistics. d Q–Q plots for ALP, e Q–Q plots for ALT, and f Q–Q plots for GGT. Inflation of test statistics was represented by lambda (λ) values.

Fig. 3. Overview of nearest genes mapped to known and novel ALT, ALP, and GGT replicated SNPs and their overlap.

Yellow box depicts replicated genes mapped to ALT. Red box depicts replicated genes mapped to ALP. Blue box depicts replicated genes mapped to GGT. Boxes in overlapping sections depict genes identified to be associated with more than one liver enzyme.

Fig. 4. Overview of diseases and traits known to be related to liver enzyme SNPs using DisGeNET.

Previous knowledge on the association of all (pink), known (brown), ALT (stone), ALP (light gray), and GGT(aegean) loci are depicted.

Fig. 5. Overview of tissue enrichment for GGT SNPs using DEPICT.

Illustrated are the tissues and organs enriched with genes mapped to GGT SNPs. False discovery rate <0.05 was used to identify enriched tissue/cells.

Fig. 6. Overview of tissue and physiological systems enrichment using DEPICT.

Illustrated are the tissues and organs enriched with genes mapped to ALT (a) and ALP (b) SNPs. False discovery rate <0.05 was used to identify enriched tissue/cells.