Abstract
OBJECTIVE: To investigate the molecular mechanisms of Huanglian Wendan Decoction in treating type 2 diabetes mellitus-associated skeletal muscle lesions via integrated bioinformatics and experimental validation. METHODS: T2DM-related skeletal muscle microarray datasets from the GEO database were analyzed to identify differentially expressed genes (n = 3309). Apoptosis-related targets (n = 887) and T2DM-related targets (n = 3106) were retrieved from GeneCards, and intersection analysis yielded 96 shared targets. Protein-protein interaction networks were constructed using STRING and Cytoscape; via CytoHubba, 10 core targets were first screened based on degree centrality. ROC curves were further used to validate the diagnostic efficacy of these 10 core genes, ultimately confirming 5 key targets (EGFR, PTEN, MDM2, TRAF6, and CCL5) with valid diagnostic value. Functional enrichment analysis revealed pathways including cysteine-type endopeptidase activity regulation and pyroptosis, while immune infiltration analysis linked the key targets to immune cell modulation. Molecular docking was used to assess the interactions between HLWDD compounds and the 5 key targets. A T2DM rat model was established using a high-fat diet combined with streptozotocin; biochemical parameters and skeletal muscle morphology were evaluated, and the protein expression of the 5 key targets was analyzed using immunohistochemistry and western blotting. RESULTS: HLWDD significantly reduced HbA1c, blood lipid, glucose, and renal dysfunction markers (P < 0.05), improved skeletal muscle histology, and downregulated core target proteins. Bioinformatics has highlighted the association of core genes with apoptosis and immune responses. ROC analysis demonstrated strong diagnostic potential (AUC > 0.5). CONCLUSION: HLWDD alleviated T2DM skeletal muscle injury by modulating apoptosis-related pathways and immune interactions, as supported by multi-omics and experimental validation. This study provides novel therapeutic targets and mechanistic insights into HLWDD for T2DM management.