Human microbiome research has extended from studying the microbes themselves to the molecules they produce. A typical metabolomics study can only characterize 10% of the molecular data from a human sample. Here, a discovery strategy called reverse metabolomics was debuted, whereby tandem mass spectrometry spectra acquired from newly synthesized compounds are searched for in public metabolomics datasets to uncover phenotypic associations. To demonstrate the concept, four metabolite classes were synthesized and explored in this study, and they discovered that Glu, Ile/Leu, Phe, Thr, Trp, and Tyr binding cholic acids were elevated in Crohn's disease (CD), which were verified in four inflammatory bowel disease cohorts. Bifidobacterium, Clostridium, and Enterococcus genera produce certain bile amidates which may promote intestinal inflammation by interferon-γ production in CD4+ T cells and agonism of the pregnane X receptor (PXR).
The discovery and analysis of metabolites usually requires separating them from the sample first, which introduces many limitations, especially when studying metabolites from human samples. Reverse metabolomics is a new method for studying microbiome-derived molecules. It combines organic synthesis, data science, and mass spectrometry to investigate biological phenotypes within a specific compound class and, thereby, identify potential diagnostic biomarkers or therapeutic targets. In its first application, hundreds of molecules that had never been observed in the human body were discovered, and after searching for around 2,000 unique compounds in public data, newly discovered bile acids were associated with CD, which may mediate CD processes through the regulation of host immune function and PXR signaling. This groundbreaking work is significant in discovering new metabolites with potential diagnostic and therapeutic applications.
Gentry EC, et al. Reverse metabolomics for the discovery of chemical structures from humans. Nature. 2024 Feb;626(7998):419-426. DOI: 10.1038/s41586-023-06906-8. Epub 2023 Dec 5. PMID: 38052229; PMCID: PMC10849969.