A new incompatible combination: Reynoutria multiflora combined with Cullen corylifolium enhances idiosyncratic hepatotoxicity under immunological stress

Ethnopharmacological relevance Compound preparations of traditional Chinese medicines (TCMs) have gained considerable interest. However, ensuring their safety is difficult because of ingredient complexity. Reynoutria multiflora (Thunb.) Moldenke (PM), commonly used in tonic preparations for clinical use, has been frequently reported to cause drug-induced liver injury (DILI). Liver injury caused by PM is idiosyncratic and its underlying mechanisms have been elucidated. However, a systematic evaluation of the safety of PM compound formulations is lacking. Our preliminary research predicted a potential compatibility risk between PM and Cullen corylifolium (Linnaeus) Medikus (PF); however, the specific effects and mechanisms of their combined action on the liver have not been fully studied. Aim of the study To determine the effects of the PM and PF combination on the liver and its mechanisms. Materials and methods This study used a lipopolysaccharide-induced idiosyncratic (IDILI) model. Kits were employed to measure the liver function indices and ELISA was used to detect inflammatory factor content. The histopathological changes of the liver were observed using H&E and TUNEL staining. Serum metabolomics and liver transcriptomics were performed simultaneously to detect overall differences in metabolite and gene expression. Network pharmacology and molecular docking were used to screen hepatotoxic components and their potential pathways. Results Liver function indicators and pathological results demonstrated that the combined treatment with PM and PF may exacerbate immune-mediated liver injury compared to treatment with either herb alone. Inflammatory factor levels indicated that the combination had an immuno-amplifying effect, further increasing inflammatory cytokine levels. Integrated metabolomic and transcriptomic analysis revealed that the combination primarily affected the primary bile acid synthesis and glycerophospholipid metabolism. This exacerbated hepatocyte apoptosis and liver injury by down-regulating the expression of key genes ( BAAT and ALB ) for bile acid metabolism. Network pharmacology and molecular docking identified 2,3,5,4'-Tetrahydroxy stilbene-2-Ο-β-D-glucoside and psoralidin as potential toxic components of PM and PF, respectively. These two ingredients specifically targeted the cytosolic DNA-sensing pathway and NOD-like receptor signalling pathway. When combined, PM and PF triggered immune hyperactivation via dual-pathway cross-regulation, leading to immuno-metabolic disorders, increased hepatocyte apoptosis and enhanced inflammatory cascade responses. Conclusions The PM and PF combination represents a newly identified incompatible pair that can aggravate IDILI under specific conditions. Our findings provide a critical reference for the safe use of compound medication, enriching the TCM combination theory.