Dual immunomodulatory effects of bacteroides and phocaeicola secretomes revealed by integrative metabolomic and functional screening in HT-29 cells

Abstract

Background The immunomodulatory effects of gut Bacteroides are typically attributed to surface polysaccharides signaling through TLR2 and TLR4, and to outer membrane vesicles (OMVs) that traffic effector molecules to host cells. However, these bacteria also secrete a wide array of metabolites, including short-chain fatty acids (SCFAs), medium-chain fatty acids (MCFAs), phenolic compounds, and aldehydes, whose contributions to immune modulation remain underexplored. Moreover, it is unclear whether different Bacteroides and Phocaeicola species share similar immunomodulatory profiles or display species- and strain-specific activities. Methods We cultivated 18 strains of Bacteroides and Phocaeicola, collected their conditioned media, and applied them to HT-29 intestinal epithelial cells. TLR2 and TLR4 gene expression and IL-8 secretion were measured under basal and LPS-stimulated conditions. The volatile metabolome was characterized using HS-GC/MS, and the data were integrated with cellular responses. Results The strains induced strikingly divergent effects, ranging from strong suppression to marked induction of TLR2/TLR4 expression and IL-8 production. Metabolomic profiling revealed that while SCFAs and MCFAs were associated with anti-inflammatory outcomes, the presence of phenolic compounds and aldehydes correlated with pro-inflammatory effects. Integration of metabolomic and cellular data allowed us to classify the strains into three functional categories: anti-inflammatory (e.g., B. uniformis EBA 5-20), pro-inflammatory (e.g., B. thetaiotaomicron 6-237), and strains with context-dependent mixed effects. Conclusion The immunomodulatory activity of gut Bacteroides and Phocaeicola cannot be predicted solely by their taxonomic affiliation; rather, it depends on the specific repertoire of secreted metabolites and their interplay with host pathways. Evaluating both metabolomic profiles and receptor-mediated responses provides a rational basis for selecting candidate probiotics for personalized microbiome modulation.