Untargeted Metabolome- and Transcriptome-Wide Association Study Suggests Causal Genes Modulating Metabolite Concentrations in Urine Article Swipe
YOU?
·
· 2021
· Open Access
·
· DOI: https://doi.org/10.1021/acs.jproteome.1c00585
· OA: W3209901575
Gene products can affect the concentrations of small molecules (aka "metabolites"), and conversely, some metabolites can modulate the concentrations of gene transcripts. While many specific instances of this interplay have been revealed, a global approach to systematically uncover human gene-metabolite interactions is still lacking. We performed a metabolome- and transcriptome-wide association study to identify genes influencing the human metabolome using untargeted metabolome features, extracted from <sup>1</sup>H nuclear magnetic resonance spectroscopy (NMR) of urine samples, and gene expression levels, quantified from RNA-Seq of lymphoblastoid cell lines (LCL) from 555 healthy individuals. We identified 20 study-wide significant associations corresponding to 15 genes, of which 5 associations (with 2 genes) were confirmed with follow-up NMR data. Using metabomatching, we identified the metabolites corresponding to metabolome features associated with the genes, namely, N-acetylated compounds with <i>ALMS1</i> and trimethylamine (TMA) with <i>HPS1</i>. Finally, Mendelian randomization analysis supported a potential causal link between the expression of genes in both the <i>ALMS1</i>- and <i>HPS1</i>-loci and their associated metabolite concentrations. In the case of <i>HPS1</i>, we additionally observed that TMA concentration likely exhibits a reverse causal effect on <i>HPS1</i> expression levels, indicating a negative feedback loop. Our study highlights how the integration of metabolomics, gene expression, and genetic data can pinpoint causal genes modulating metabolite concentrations.
