Evidence of a causal and modifiable relationship between kidney function and circulating trimethylamineN-oxide with implications for heart and kidney disorders

Abstract

AbstractObjectivesThe host-microbiota co-metabolite trimethylamineN-oxide (TMAO) is linked to increased thrombotic and cardiovascular risks. Here we, sought to i) characterize which host variables contribute to fasting serum TMAO levels in real-life settings ii) identify potential actionable therapeutic means related to circulating TMAO.DesignWe applied “explainable” machine learning, univariate-, multivariate- and mediation analyses of fasting plasma TMAO concentration and a multitude of bioclinical phenotypes in 1,741 adult Europeans of the MetaCardis study. We expanded and validated our epidemiological findings in mechanistic studies in human renal fibroblasts and a murine model of kidney fibrosis following TMAO exposure.ResultsNext to age, kidney function was the primary variable predicting circulating TMAO in MetaCardis, with microbiota composition and diet playing minor, albeit significant roles. Mediation analysis revealed a causal relationship between TMAO and kidney function decline that strengthened at more severe stages of cardiometabolic disease. We corroborated our findings in preclinical models where TMAO exposure augmented conversion of human renal fibroblasts into myofibroblasts and increased kidney scarringin vivo. Mechanistically, TMAO aggravated kidney fibrosis due to ERK1/2 hyperactivation synergistically with TGF-β1 signaling. Consistent with our findings, patients receiving next-generation glucose-lowering drugs with reno-protective properties, had significantly lower circulating TMAO when compared to propensity-score matched control individuals.ConclusionAfter age, kidney function is the major determinant of fasting circulating TMAO in adults. Our findings of lower TMAO levels in individuals medicated with reno-protective anti-diabetic drugs suggests a clinically actionable intervention for decreasing TMAO-associated excess cardiovascular risk that merits urgent investigation in human trials.Data availability statementRaw shotgun sequencing data that support the findings of this study have been deposited in the European Nucleotide Archive with accession codes PRJEB37249, PRJEB38742, PRJEB41311 and PRJEB46098. Serum NMR and urine NMR metabolome data have been uploaded to Metabolights with accession number MTBLS3429; serum GC-MS and isotopically quantified serum metabolites (UPLC–MS/MS) are available from MassIVE with accession numbers MSV000088042 and MSV000088043, respectively.