A Genome-Wide Metabolic QTL Analysis in Europeans Implicates Two Loci Shaped by Recent Positive Selection

Geoffrey C. Nicholson, Mattias Rantalainen, Jia V. Li, Anthony D. Maher, Daniel Malmodin, Kourosh R. Ahmadi, Johan H. Faber, Amy Barrett, Josine L. Min, Nigel W. Rayner, Henrik Toft, Maria Krestyaninova, Juris Vīksna, Sudeshna Guha Neogi, Marc‐Emmanuel Dumas, Uğis Sarkans, Peter Donnelly, Thomas Illig, Jerzy Adamski, Karsten Suhre, Maxine Allen, Krina T. Zondervan, Tim D. Spector, Jeremy K. Nicholson, John C. Lindon, Dorrit Baunsgaard, Elaine Holmes, Mark I. McCarthy, Chris HolmesPLoS 2011 · DOI

Abstract

We have performed a metabolite quantitative trait locus (mQTL) study of the $^1$H nuclear magnetic resonance spectroscopy ($^1$H NMR) metabolome in humans, building on recent targeted knowledge of genetic drivers of metabolic regulation. Urine and plasma samples were collected from two cohorts of individuals of European descent, with one cohort comprised of female twins donating samples longitudinally. Sample metabolite concentrations were quantified by $^1$H NMR and tested for association with genome-wide single-nucleotide polymorphisms (SNPs). Four metabolites' concentrations exhibited significant, replicable association with SNP variation ($8.6 \times 10^{-11} < p < 2.8 \times 10^{-23}$). Three of these—trimethylamine, 3-amino-isobutyrate, and an N-acetylated compound—were measured in urine. The other—dimethylamine—was measured in plasma. Trimethylamine and dimethylamine mapped to a single genetic region (hence we report a total of three implicated genomic regions). Two of the three hit regions lie within haplotype blocks (at 2p13.1 and 10q24.2) that carry the genetic signature of strong, recent, positive selection in European populations. Genes NAT8 and PYROXD2, both with relatively uncharacterized functional roles, are good candidates for mediating the corresponding mQTL associations. The study's longitudinal twin design allowed detailed variance-components analysis of the sources of population variation in metabolite levels. The mQTLs explained 40%–64% of biological population variation in the corresponding metabolites' concentrations. By re-analysing our plasma samples using the Biocrates platform, we replicated the mQTL findings of previous studies and discovered a previously uncharacterized yet substantial familial component of variation in metabolite levels.

Introduction

While thousands of genetic variants affecting gene expression (eQTLs) have been mapped, understanding how DNA dictates the "metabolome"—the complex collection of small molecules in biofluids—is still in its early stages. Metabolites are the functional end-products of biological processes, making them excellent biomarkers for disease. However, a metabolite's concentration is the result of a complex web of enzymes, transporters, and environmental factors like diet, making them mechanistically distant from the genome. This study seeks to identify "heavy hitter" genetic variants, or Metabolic Quantitative Trait Loci (mQTLs), that exert large effects on the human metabolome and to investigate the evolutionary forces shaping these associations.

Methodology

The researchers employed an untargeted $^1$H Nuclear Magnetic Resonance (NMR) spectroscopy approach to provide a global snapshot of abundant hydrogen-containing molecules in urine and plasma.

Study Design and Datasets: * MolTWIN: A cohort of 142 female twins who provided samples longitudinally across multiple visits. This design allowed for a sophisticated variance decomposition to separate mQTL effects from other familial factors, environmental influences, and day-to-day fluctuations. * MolOBB: 69 participants from the Oxford Biobank used for replication. * HapMap (CEU and YRI): Reference populations used to identify signatures of natural selection by comparing allele frequencies between European and African populations.

Analytical Pipeline: The team performed a genome-wide scan testing associations between approximately 2.5 million SNPs and 526 metabolite peaks. To ensure robustness, plasma samples were also re-analyzed using the targeted Biocrates mass-spectrometry platform to replicate findings from previous landmark studies.

Key Results

The study identified high-confidence mQTLs for four key metabolites: trimethylamine (urine), 3-amino-isobutyrate (urine), an N-acetylated compound (urine), and dimethylamine (plasma). These associations were extremely robust, with p-values ranging from $8.6 \times 10^{-11}$ to $2.8 \times 10^{-23}$.

Notably, the identified mQTLs explain 40% to 64% of the biological variation in their respective metabolites. This effect size is significantly higher than what is typically observed in gene expression studies, suggesting these variants are dominant regulators of their metabolic pathways. Furthermore, the authors successfully replicated 12 out of 15 mQTLs previously reported in major targeted metabolomics studies, demonstrating that untargeted NMR is a powerful and complementary tool to mass-spectrometry.

Evolutionary Insights and Candidate Genes

A major finding of this work is that two of the three identified genomic regions (2p13.1 and 10q24.2) carry signatures of strong, recent positive selection in Europeans. The "advantageous" alleles in these populations are associated with decreased expression of PYROXD2 and increased concentrations of trimethylamine.

The study highlights two primary candidate genes:

  • NAT8: Linked to kidney function and N-acetylation, likely mediating the N-acetylated compound association.
  • PYROXD2: A poorly characterized oxidoreductase that appears to be a functional driver for the observed metabolic traits at the 10q24.2 locus.

Discussion and Limitations

Despite the strength of these findings, a substantial "familial" component of variation remains even after accounting for the primary mQTLs. This suggests that while certain SNPs act as major regulators, a broader network of genes and shared environmental factors also contributes to metabolic profiles.

Limitations and Future Directions: * Metabolite Identification: One of the strongest signals was for an "N-acetylated compound" (N-ACu) whose exact chemical identity remains partially unknown. * Tissue Specificity: While metabolites were measured in biofluids, regulation likely occurs in the liver or kidneys. Future studies using tissue biopsies are needed to confirm intracellular mechanisms. * Functional Validation: Molecular biology studies are required to determine how specific mutations alter enzyme kinetics or transporter function. * Population Diversity: The selection signatures are specific to European populations; the marked difference in allele frequencies compared to African populations (HapMap-YRI) suggests divergent evolutionary pressures.

Summary updated

Abstract

We have performed a metabolite quantitative trait locus (mQTL) study of the (1)H nuclear magnetic resonance spectroscopy ((1)H NMR) metabolome in humans, building on recent targeted knowledge of genetic drivers of metabolic regulation. Urine and plasma samples were collected from two cohorts of individuals of European descent, with one cohort comprised of female twins donating samples longitudinally. Sample metabolite concentrations were quantified by (1)H NMR and tested for association with genome-wide single-nucleotide polymorphisms (SNPs). Four metabolites' concentrations exhibited significant, replicable association with SNP variation (8.6×10(-11)<p<2.8×10(-23)). Three of these-trimethylamine, 3-amino-isobutyrate, and an N-acetylated compound-were measured in urine. The other-dimethylamine-was measured in plasma. Trimethylamine and dimethylamine mapped to a single genetic region (hence we report a total of three implicated genomic regions). Two of the three hit regions lie within haplotype blocks (at 2p13.1 and 10q24.2) that carry the genetic signature of strong, recent, positive selection in European populations. Genes NAT8 and PYROXD2, both with relatively uncharacterized functional roles, are good candidates for mediating the corresponding mQTL associations. The study's longitudinal twin design allowed detailed variance-components analysis of the sources of population variation in metabolite levels. The mQTLs explained 40%-64% of biological population variation in the corresponding metabolites' concentrations. These effect sizes are stronger than those reported in a recent, targeted mQTL study of metabolites in serum using the targeted-metabolomics Biocrates platform. By re-analysing our plasma samples using the Biocrates platform, we replicated the mQTL findings of the previous study and discovered a previously uncharacterized yet substantial familial component of variation in metabolite levels in addition to the heritability contribution from the corresponding mQTL effects.
BibTeX
@inproceedings{nicholson2011genome,
  author = {Geoffrey C. Nicholson and Mattias Rantalainen and Jia V. Li and Anthony D. Maher and Daniel Malmodin and Kourosh R. Ahmadi and Johan H. Faber and Amy Barrett and Josine L. Min and Nigel W. Rayner and Henrik Toft and Maria Krestyaninova and Juris Vīksna and Sudeshna Guha Neogi and Marc‐Emmanuel Dumas and Uğis Sarkans and Peter Donnelly and Thomas Illig and Jerzy Adamski and Karsten Suhre and Maxine Allen and Krina T. Zondervan and Tim D. Spector and Jeremy K. Nicholson and John C. Lindon and Dorrit Baunsgaard and Elaine Holmes and Mark I. McCarthy and Chris Holmes},
  title = {{A Genome-Wide Metabolic QTL Analysis in Europeans Implicates Two Loci Shaped by Recent Positive Selection}},
  booktitle = {PLoS},
  year = {2011},
  doi = {10.1371/journal.pgen.1002270},
}
Figure