香根草根提取物通过调节TLR4/NLRP3信号通路，有望缓解肥胖及胰岛素抵抗症状。

Abstract

Background Polygonati Odorati Rhizoma (Yuzhu) is known to regulate glucose and lipid metabolism and inflammation, but its bioactive components and molecular mechanisms, particularly those involving the TLR4/NLRP3 pathway, remain unexplored. This study integrates chemical profiling, network pharmacology, and experimental validation to investigate whether and how Yuzhu extract ameliorates obesity-associated insulin resistance (IR).Methods The chemical constituents of Yuzhu were characterized by UHPLC-Q-TOF-MS/MS. Network pharmacology integrated with GEO dataset analysis was employed to identify potential active compounds, therapeutic targets, and signaling pathways. Molecular docking was performed to evaluate compound-target interactions. A high-fat diet (HFD)-induced obese mouse model and a TNF-α-induced 3T3-L1 adipocyte inflammation model were used for experimental validation. Metabolic parameters, inflammatory cytokines, and TLR4/NLRP3 pathway-related molecules were assessed by biochemical assays, RT-qPCR, Western blot, immunohistochemistry, ELISA, and TUNEL staining.Results A total of 127 chemical constituents were identified from Yuzhu, of which 67 met the screening criteria for potential bioactivity. Network pharmacology analysis identified 23 overlapping targets and highlighted corchorifatty acid F, 9,10-dihydroxy-12Z-octadecenoic acid, N-trans-feruloyloctopamine, and diosgenin glucoside (DG) as core active components. TLR4 and CCL2 were identified as hub targets, with significant enrichment in the NOD-like receptor signaling pathway. Molecular docking demonstrated favorable binding affinities of the core compounds to TLR4 and NLRP3, with DG exhibiting the strongest binding activity. In vivo, Yuzhu significantly reduced body weight gain, hyperglycemia, dyslipidemia, glucose intolerance, and IR in HFD-fed mice, while suppressing adipose tissue inflammation. Mechanistically, Yuzhu inhibited the expression of TLR4, MyD88, NLRP3, IL-1β, TNF-α, p-p65, p-STAT3, and