Integrated multi-omics reveals a probiotic-enriched microbiome–metabolome axis associated with hPMSC therapeutic effects in Crohn’s disease

Crohn’s disease (CD) is a chronic, refractory inflammatory bowel disease with limited effective treatments. Human placental mesenchymal stromal cell (hPMSC) therapy shows promising therapeutic potential but marked interindividual variability, hindering clinical standardization. Gut microbiota dysbiosis and host metabolic crosstalk are key drivers of CD pathogenesis, but their roles in hPMSC response remain unclear. Herein, we employed the SAMP1/Yit mouse model (a spontaneous CD-like ileitis model recapitulating human CD pathophysiology) and integrated multi-omics profiling (fecal 16S rRNA gene sequencing, fecal and urine metabolomics via chemical isotope labeling liquid chromatography-mass spectrometry) to characterize dynamic remodeling of the gut microbiome and host metabolome during hPMSC treatment. Our results showed that hPMSCs restored gut microbial alpha diversity, enriched probiotic taxa including Ligilactobacillus , and altered metabolites involved in pathways closely linked to intestinal inflammation and barrier function, such as lysine degradation. We further identified 13 microbial taxa, 11 fecal metabolites, and 15 urine metabolites as candidate non-invasive microbial and metabolic markers associated with hPMSC treatment. Strong positive correlations between Ligilactobacillus and amino acid-related metabolites (e.g., lysopine) highlight coordinated alterations within a probiotic-enriched microbiome–metabolome axis following hPMSC treatment. A machine learning model integrating these multi-omics features achieved an area under the receiver operating characteristic curve (AUC) of 0.901 for discriminating hPMSC-treated mice from disease-model controls. Collectively, our study delineates coordinated gut microbiome and metabolic remodeling following hPMSC treatment in CD, highlights a probiotic-enriched microbiome–metabolome axis, and identifies candidate non-invasive microbial and metabolic markers for future mechanistic and translational validation.