目的 通过网络药理学预测四君子汤治疗2型糖尿病（T2DM）的主要活性成分和作用靶点，探讨其多成分-多靶点-多通路的潜在作用机制；并选取AMPK信号通路中的InsR、PI3K进行动物实验验证。方法 通过中医药系统药理学数据库和分析平台（TCMSP）、文献挖掘和本课题组已有研究收集四君子汤及其4味中药党参、白术、茯苓、甘草的化学成分，经口服生物利用度（OB）、类药性（DL）、半衰期（HL）3个条件筛选得到药物的有效成分，通过SwissTargetPrediction数据库进行有效成分靶点预测；同时从TTD、Drugbank和DisGeNET数据库对T2DM的靶点进行检索及筛选；成分靶点与疾病靶点映射后使用Cytoscape 3.2.1软件构建药物有效成分-靶点蛋白相互作用网络，对该网络进行拓扑学分析筛选节点自由度（degree）≥平均自由度的节点得到核心有效成分-核心靶点网络，同时使用String数据库绘制核心靶点蛋白-蛋白相互作用（PPI）网络；对核心靶点基因利用DAVID数据库进行GO分析和KEGG分析，构建核心有效成分-核心靶点-代谢通路网络图，探讨四君子汤治疗T2DM的潜在作用机制。采用系统对接网站进行相关度前3蛋白对相关度前5成分的分子对接。并进一步采用动物实验进行验证：选取SD大鼠，运用高糖高脂饲料联合小剂量链脲佐菌素（STZ）诱导T2DM模型，5.65 g/kg四君子汤ig给药28 d，检测各组大鼠的空腹血糖（FBG）及口服葡萄糖耐量（OGTT），RT-PCR法检测AMPK信号通路中InsR、PI3K mRNA的表达量。结果 从四君子汤中筛选出113个化学成分，涉及治疗T2DM的47个靶点；根据节点自由度≥平均自由度，筛选出核心成分22个，核心靶点27个；GO分析结果表明其涉及血糖稳态、脂肪组织发育的正调控等7个生物过程，涉及类固醇激素受体活化、药物结合等4个分子功能，包括质膜、核常染色质2个细胞组成；KEGG分析结果表明其可能通过AMPK信号通路、PPAR信号通路、胰岛素抵抗等14个信号通路治疗糖尿病。分子对接60%成分具有强烈的结合活性，40%成分具有较强的结合活性。四君子汤可显著改善T2DM大鼠OGTT（P<0.001），同时降低FBG（P<0.01），显著升高InsR、PI3K mRNA（P<0.001）的表达量，验证了网络药理学的部分预测结果。结论 四君子汤通过多靶点、多通路治疗T2DM，为其分子机制的研究奠定一定的基础。

Objective The main active components and targets of Sijunzi Decoction for type 2 diabetes mellitus (T2DM) treatment were predicted by network pharmacology, and the potential mechanism of multi-component-multi-target-multi-pathway was discussed, and INSR and PI3K in the AMPK signal pathway were verified in animal experiment. Methods Based on TCM system pharmacology database and analysis platform (TCMSP), literature mining and previous studies of our research group, the chemical components of Sijunzi decoction and its four herbs Codonopsis pilosula, Atractylodes macrocephala, Poria cocos, Glycyrrhiza uralensis were screened. Three conditions of oral absorption rate (OB), drug-like (DL), half-life (HL) were used to screen the active components of the drug; The targets of active components were predicted through the Swiss Target Prediction database; TTD, Drupe and DisGNET databases were used to screen the targets of T2DM. After mapping the component target and disease target, the interaction network of the target proteins of the active components was constructed by using the software of Cytoscape 3.2.1. Through topological analysis of the network, the nodes with degree of freedom ≥ average degree of freedom were screened out, and the core target network of core active ingredient was obtained, while using the String database to draw core-target protein-protein interaction (PPI) networks, using DAVID database for GO analysis and KEGG analysis of core target genes to construct core active components-core target-metabolism Path network diagram to explore the potential mechanism of Sijunzi Decoction against T2DM. The system docking site was used for molecular docking of the top 5 components and the top 3 proteins. Further, animal experiments were used to verify. SD rats were fed with high-sugar high-fat diet combined with low-dose streptozotocin (STZ) to induce T2DM model, and administrated with 5.65 g/kg Sijunzi Decoction for 28 d, and the levels of blood glucose (FBG) and oral glucose tolerance (OGTT) of each group of mice were determinated. The InsR, PI3K mRNA expression in AMPK signaling pathway was detected by RT-PCR. Results A total of 113 chemical components were selected from Sijunzi Decoction, involving 47 targets for the treatment of T2DM. Twenty-two core components and 27 core targets were screened according to node degree of freedom ≥ average degree of freedom. The GO analysis results showed that it involved seven biological processes such as blood glucose homeostasis, positive regulation of adipose tissue development, four molecular functions including steroid hormone receptor activation and drug binding, and cell composition including the plasma membrane and nuclear chromatin. The results of KEGG analysis showed that it might treat diabetes through 14 signaling pathways, such as AMPK signaling pathway, PPAR signaling pathway and insulin resistance. A total of 60% of the molecules had intense binding activity and 40% had stronger binding activity. The results of animal experiments showed that Sijunzi Decoction could significantly improve OGTT (P<0.001), and decrease FBG (P<0.01) in T2DM rats, and significantly increase the expression of INSR mRNA (P<0.001) and PI3K mRNA (P<0.001). Some of the predicted results of network pharmacology were verified. Conclusion This study reflects that Sijunzi Decoction can treat type 2 diabetes mellitus through multi-target and multi-channel, which lays a foundation for the future study of molecular mechanism.

R285.5

山西省基础研究项目（2013011048-6）；山西中医药大学重点培育项目（2019PY-117）