【目的】 探究黄芪来源细胞外囊泡（EVs）样颗粒（AR-EVLP）在糖尿病伤口愈合中的作用及潜在机制，为中医治疗糖尿病 皮肤溃疡提供新的治疗药物模式和理论依据。【方法】 使用超速差速离心法提取 AR-EVLP。使用非靶向代谢组学分析 AREVLP的潜在有效成分，并通过与Herb数据库比对，明确AR-EVLP中的主要活性物质。通过Swiss Target Prediction数据库和 TCMSP 数据库获取有效成分对应靶点。通过 GeneCards 数据库获取糖尿病伤口愈合相关靶点。通过药物靶点和疾病靶点交 集，获得关键靶点。通过DAVID数据库进行基因本体论（GO）功能注释和京都基因与基因组百科全书（KEGG）通路富集分析。 根据通路富集结果绘制“药物成分-靶点-通路”图，明确核心作用靶点。通过Autodock和PyMOL软件进行分子对接和可视 化。【结果】 芒柄花黄素是 AR-EVLP的主要有效成分，其与糖尿病和伤口愈合之间存在 66个关键靶点，包括 SRC、CASP3、 JUN等15个关键节点蛋白。GO功能富集分析提示芒柄花黄素可以调控蛋白磷酸化、基因表达等生物过程，KEGG通路富集 分析提示可能通过 VEGF、TNF等多个信号通路发挥作用。PIK3CA、JUN、MAPK14等 7个靶点为芒柄花黄素发挥作用的核 心靶点。分子对接结果显示，芒柄花黄素与MAPK14结合力最强。【结论】 AR-EVLP可能是糖尿病伤口愈合治疗潜在的有效 药物模式。

Objective To explore the role and potential mechanism of Astragali Radix-derived extracelluar vesicles （EVs）-like particles（AR-EVLP） in diabetic wound healing， providing a novel therapeutic drug mode and theoretical basis for traditional Chinese medicine in treating diabetic skin ulcers. Methods AR-EVLP was extracted using ultracentrifugation. Untargeted metabolomics was used to analyze the potential active components of AR-EVLP，and the main active substances in AR-EVLP were identified by comparison with the Herb database. The targets of the active components were obtained through the Swiss Target Prediction database and the TCMSP database. Targets related to diabetic wound healing were obtained from the GeneCards database. Key targets were identified by intersecting drug targets and disease targets. GO functional annotation and KEGG pathway enrichment analysis were performed using the DAVID database. Based on the pathway enrichment results， a “drug components-targets-pathways” diagram was constructed to identify core targets. Molecular docking and visualization were performed using Autodock and PyMOL software. Results Formononetin was identified as the main active component in AR-EVLP，with 66 key targets related to diabetes and wound healing，including 15 key node proteins such as SRC，CASP3 and JUN. GO functional enrichment analysis suggested that formononetin can regulate biological processes such as protein phosphorylation and gene expression. KEGG pathway enrichment analysis indicated potential involvement in multiple signaling pathways，including VEGF and TNF. Seven targets， including PIK3CA， JUN and MAPK14， were identified as core targets for formononetin. Molecular docking showed that formononetin had the strongest binding affinity with MAPK14. Conclusion AR-EVLP may be a potential effective drug mode for the treatment of diabetic wound healing.

R285

国家自然科学基金资助项目（编号：82174119）