Steaming reduces Polygonum multiflorum hepatotoxicity by modulating gut microbiota-bile acid-FXR/NF-κB signaling in the gut-liver axis.
BACKGROUND Polygonum multiflorum Thunb. (PM) can induce hepatotoxicity, particularly with prolonged use. Traditional processing methods, notably steaming (SPM) and steaming with black bean decoction (PPM), are recognized for their ability to mitigate this toxicity; however, the underlying mechanisms, especially from the gut-liver axis perspective, remain poorly understood. PURPOSE This study aims to clarify the mechanism by which processing reduces PM-induced hepatotoxicity, with a particular focus on the regulation of the gut-liver axis. METHODS We employed a comprehensive approach integrating histopathological examination, multi-omics analyses (including 16S rRNA sequencing, transcriptomics, and metabolomics), and molecular analysis to investigate the distinct effects of PM, SPM, and PPM in a mouse model. RESULTS PM caused significant liver injury, characterized by elevated serum alanine aminotransferase (ALT) and aspartate aminotransferase (AST) levels and severe pathological damage. This damage was marked by intestinal barrier disruption (decreased tight junction protein-1 (ZO-1) and occludin), systemic inflammation (elevated tumor necrosis factor-α (TNF-α) and interleukin-6 (IL-6) levels), and activation of the hepatic nuclear factor-kappa B (NF-κB) pathway. Multi-omics analysis revealed that PM induced severe GM dysbiosis (e.g., reduction in the Lactobacillus and a significant increase in the unclassified_f_Oscillospiraceae), a marked decrease in short-chain fatty acids (SCFAs) (e.g., a 72% reduction in butyrate), and disruption of the hepatic bile acid profile. Notably, the two processed products, particularly PPM, significantly reversed these alterations. Compared to the PM group, PPM treatment reduced serum ALT by 82% and exerted extensive protective effects. It restored beneficial bacteria, increased SCFAs levels, normalized bile acid metabolism, and suppressed the hepatic fibrogenic/NF-κB inflammatory axis. This detoxifying effect may be associated with reduced systemic exposure to toxic stilbene glycosides and anthraquinones in the processed PM groups compared to the PM group. CONCLUSION This study suggests that processing, particularly with black bean decoction (PPM), can alleviate PM-induced hepatotoxicity by reshaping the GM, restoring its metabolic products (SCFAs), normalizing bile acid signaling, and downregulating the FXR/NF-κB inflammatory axis within the gut-liver axis. These findings provide a mechanistic basis for the clinical preference for processed PM products and highlight the importance of stringent quality control, while also underscoring that residual risk may persist, particularly in susceptible populations.