目的 运用网络药理学及动物实验验证探究三果汤 （余甘子、毛诃子、诃子组成） 调节胰岛 β 功能治疗糖尿病的活性成分及潜在分子机制。方法 采用中药系统药理分析平台 （TCMSP） 结合文献检索、SwissADME获取三果汤活性成分及预测靶点；通过 Drugbank、Genecards、OMIM、TTD 数据库获取糖尿病主要靶点；利用韦恩图 （Venn） 分析药物和疾病共有基因，采用 STRING 数据库及 CytoScape 3.8.0 软件构建“三果汤-成分-靶点-基因-疾病”共同靶点的蛋白相互作用 （PPI） 网络，并通过 ClueGO、CytoHubba 可视化分析得到潜在作用基因靶点及其富集通路。将 Wistar雄性大鼠随机分为正常组、模型组及三果汤低、中、高 （0.43、0.86、1.72 g•kg-1）剂量组，采用链脲佐菌素 （STZ） 腹腔注射复制糖尿病模型及相应药物干预。进行胰岛功能指数测算、病理切片染色及 PCR 检测。结果 三果汤共有 23 种活性成分，预测靶点 434 个，疾病基因 1 200 个，维恩 （Venn） 分析三果汤与糖尿病共有基因 146 个；三果汤治疗糖尿病主要与调节胰岛素分泌、葡萄糖跨膜运输、蛋白激酶的激活、促进血管生长以及 MAPK 联级反应等有关，涉及蛋白激酶 Bα （AKT1） 、甘油醛-3-磷酸甘油醛脱氢酶（GAPDH） 、血管内皮生长因子 A （VEGFA） 、丝裂原活化蛋白激酶 1 （MAPK1） 、前列腺素内过氧化物合酶（PTGS2） 、信号转导与转录活化因子 3 （STAT3） 、Toll 样受体 4 （TLR4） 7 个核心靶点。动物实验验证结果显示，三果汤能增加 HOMA-β 指数 （P＜0.01） ，促进胰岛体积、β 细胞恢复，减少 α 细胞数量；能明显上调大鼠胰腺组织中 AKT1、GAPDH 基因的表达 （P＜0.001） ，下调 VEGFA、MAPK1、PTGS2、STAT3、TLR4 基因的表达（P＜0.001，P＜0.000 1） ，验证了网络药理学结果的部分预测。结论 三果汤能多靶点、多通路发挥治疗糖尿病的作用。证实其通过调节 AKT1、GAPDH、VEGFA、MAPK1、PTGS2、STAT3、TLR4 等 7 个关键靶点基因，从而调节胰岛 β 细胞功能，达到治疗糖尿病的效果。

Objective To explore the active components and potential molecular mechanism of Triphala （composed of Phyllanthi Fructus， Terminaliae Belliricae Fructus and Chebulae Fructus， THL） in regulating the function of pancreatic islet β by using the network pharmacology method and animal experiment verification. Methods The active constituents and predicted targets of THL were obtained by Traditional Chinese Medicine Database and Analysis Platform（TCMSP）combined with literature search and SwissADME. The main targets of diabetes were obtained from Drugbank，Genecards，OMIM and TTD databases. Venn diagram was used to analyze the common genes of drugs and disease，and STRING database and CytoScape 3.8.0 software was used to construct the protein-protein interaction （PPI）network of the common targets of “THL- compositions- targets- genes- disease.” The potential targets and their enrichment pathways were visualized using ClueGO and CytoHubba. Wistar male rats were randomly divided into normal group，model group and low-，medium- and high- dose （0.43，0.86，1.72 g•kg-1） THL groups. The diabetic model was replicated by introperitoneally injecting streptozotocin （STZ）. The followup experiment was carried out with corresponding drug. Pancreatic islet function index calculation，pathological section staining，and PCR detection were then carried out. Results There were 23 active ingredients in THL，434 predicted targets，1 200 disease genes. A total of 146 common genes in THL and diabetes were obtained by Venn analysis. It was found that THL in the treatment of diabetes was mainly related to regulating insulin secretion，glucose transmembrane transport，activation of protein kinase，promoting vascular growth and MAPK cascade reaction，which were involved seven core targets including protein kinase B alpha（AKT1），glyceraldehyde -3- phosphate glyceraldehyde dehydrogenase （GAPDH），vascular endothelial growth factor A （VEGFA），mitogen activated protein kinase 1 （MAPK1），prostaglandin endoperoxide synthase （PTGS2），signal transduction and transcription activating factor 3 （STAT3），and Toll-like receptor 4（TLR4）. Animal validation results showed that THL could increase HOMA- β index （P＜0.01），promote recovery of pancreatic islet volume and β cell function， decrease the number of α cell， and significantly up-regulate the expression of AKT1 and GAPDH genes in rat pancreatic tissue（P＜0.001），as well as down-regulate the expression of VEGFA，MAPK1，PTGS2，STAT3，and TLR4 genes（P＜0.001，P＜0.000 1）. The partial prediction of network pharmacology results were validated with the experimental results. Conclusion THL can treat diabetes through multiple targets and pathways. It is confirmed that THL could achieve the effect of treating diabetes by regulating seven key target genes including AKT1， GAPDH， VEGFA， MAPK1， PTGS2， STAT3 and TLR4 to regulate the function of pancreatic islet β cells. This study can provide a reliable basis for the clinical use and basic research of THL in anti-diabetes.

R285.5

云南省教育厅项目 （2019Y0318）；云南省科技厅科技计划项目-应用基础研究计划项目 （202101AZ010001-015）