Editors' ChoiceGenetics

The Genome and Metabolome Collide

See allHide authors and affiliations

Science Translational Medicine  14 Sep 2011:
Vol. 3, Issue 100, pp. 100ec148
DOI: 10.1126/scitranslmed.3003167

In the past 5 years, genome-wide association studies (GWASs) have linked thousands of genotypes to hundreds of common conditions and traits and uncovered pharmacogenetic predictors of drug response and toxicity. However, the molecular mechanisms behind most DNA variants’ physiological effects remain elusive. As a first step toward mechanistic enlightenment, Suhre et al. dissect the relation between a boatload of genetic variants and individual human metabolomes—the complete collection of small-molecule metabolites in a biological sample.

The authors reported results from a GWAS of 2820 individuals that assessed 600,000 single-nucleotide polymorphisms (SNPs) for correlations to the plasma concentrations of more than 250 metabolites in 60 well-characterized, disease-relevant metabolic pathways. Thirty-seven of the 600,000 tested SNPs reached genome-wide significance (P < 1.0 × 10–6) in two separate patient cohorts and were found to impact the blood levels of the metabolites by as much as 60%. Notably, 15 of these 37 SNPs reside in genes previously tied to disorders such as gout, cancer, Crohn’s, and coronary artery disease, to name a few. Some of the metabolite, disease, and gene correlations of greatest statistical and clinical relevance included N-acetylornithine (P = 5.4 × 10–252) with kidney disease and NAT8 (N-acetyltransferase 8); mannose/glucose ratios (P = 5.5 × 10–53) with diabetes and GCKR (glucokinase regulator); and bradykinin (P = 6.6 × 10–18) with hypertension and KLKB1 (kallikrein B, plasma 1). Further, SNPs in SLCO1B1 (solute carrier organic anion transporter B1) that were previously tied to hepatic transport of statins and statin-related muscle inflammation were found to impact blood concentrations of the fatty acids tetradecanedioate and hexadecanedioate. Such critical metabolomic data can now be used in the research and development of novel statins with reduced muscle toxicity.

Although disease causation could not be confirmed by the current findings, this study underscores how the coupling of intermediate phenotypes with GWASs can illuminate disease-related biological pathways, validate previous genotype-phenotype correlations, and identify novel targets for therapeutic manipulation. Going forward, we can expect more investigations of the interplay between genome and metabolome.

K. Suhre et al., Human metabolic individuality in biomedical and pharmaceutical research. Nature 477, 54–60 (2011). [Abstract]

Navigate This Article