Ameliorative effect of Rhus verniciflua Stokes on high-fat diet-induced obesity and advanced analysis with machine...

Obesity and metabolic dysfunction-associated steatotic liver disease (MASLD) arise from impaired redox and energy homeostasis, yet current therapies remain limited. We evaluated the anti-obesity efficacy and mechanisms of Rhus verniciflua Stokes (RVS) extract using integrated in vivo and artificial intelligence (AI)-based approaches.

Male ICR mice were fed a high-fat diet (HFD) for 7 weeks with oral RVS (100, 200, or 400 mg/kg) or simvastatin (10 mg/kg). Body weight, serum/hepatic lipids, adipokines, inflammatory cytokines, NAD(P)(H) redox status, and lipid-metabolic gene expression were assessed. A multi-layer AI/ML pipeline, molecular docking, molecular dynamics, deep learning-based drug-target interaction modeling and reinforcement learning, was applied to the 24 NMR-authenticated RVS constituents to prioritize bioactive candidates.

RVS at 200 mg/kg suppressed HFD-induced body weight gain by ~18%, reduced serum LDL by ~50% and hepatic triglycerides by ~35%, and lowered the NAS Score by ~65% (p < 0.01 vs. HFD), with efficacy comparable to simvastatin. RVS restored hepatic NAD⁺/NADH and NADP⁺/NADPH balance, reactivated the AMPK-SIRT1-PGC1α axis, and suppressed lipogenic regulators (Srebf1, Fasn, C/EBPα) in liver and skeletal muscle. The AI pipeline identified a computationally prioritized triad, fisetin, butein, and ursolic acid glucoside, whose predicted multi-target profile converges on AMPK-SIRT1 activation and ACC/FAS-mediated lipogenesis suppression. Transcriptomic cross-comparison revealed directional concordance between RVS-regulated signatures and human obesity/MASLD modules.

These results demonstrate that RVS ameliorates HFD-induced obesity and MASLD through redox-metabolic pathway modulation, with systemic efficacy likely arising from the synergistic contributions of parent compounds, their Phase II conjugates, and gut microbial metabolites.