目的 基于网络药理学和分子对接技术探讨天麻治疗脑梗死的作用机制，并采用分子动力学模拟技术进行验证。方法 利用SymMap、BATMAN-TCM 1.0数据库，筛选天麻的活性成分及其潜在靶点。利用GeneCards、DisGeNET数据库收集脑梗死相关靶点，并利用STRING数据库与Cytoscape 3.10.1软件绘制蛋白质相互作用网络，将天麻与脑梗死交集靶点导入DAVID数据库进行基因本体（GO）和京都基因与基因组百科全书（KEGG）通路富集分析。使用Cytoscape 3.10.1构建“天麻–活性成分–潜在靶点–信号通路–脑梗死”网络。利用Autodock Tools 1.5.7将活性成分与核心靶点结合进行分子对接，通过分子动力学模拟分析结合模型的亲和性和稳定性。结果 筛选得到天麻治疗脑梗死的13个关键活性成分，包括间羟基苯甲酸、4-羟基苄胺、γ-谷甾醇、蝙蝠葛碱和香茅醛等。药物与疾病交集靶点89个，PPI网络图中度值最高的5个分别为肿瘤坏死因子（TNF）、胰岛素（INS）、白细胞介素-1β（IL-1β）、前列腺素内过氧化物合酶2（PTGS2）、过氧化物酶体增殖物激活受体γ（PPARG）。GO分析条目566个，其中生物过程（BP）多富集于信号释放、调节炎症反应、有机羟基化合物转运等；细胞组成（CC）多富集于神经元细胞体、分泌颗粒腔、细胞质囊泡腔等；分子功能（MF）方多富集于类固醇、蛋白酶结合和核受体活性等。KEGG富集结果88个，包括神经活性相互作用、人巨细胞病毒、脂质与动脉粥样硬化、5-羟色胺能突触和环磷酸腺苷（cAMP）信号通路。分子对接结果显示，天麻的核心活性成分能够很好地与PPI网络图中度值最高的5个靶蛋白结合，形成较稳定的复合物。分子动力学模拟进一步验证了γ-谷甾醇与INS、蝙蝠葛碱与TNF、香茅醛与TNF具有较好的亲和力与结合稳定性。结论 天麻可能通过γ-谷甾醇、蝙蝠葛碱、香茅醛等活性成分作用于TNF和INS等靶点基因，调控信号释放、调节炎症反应和cAMP信号通路等实现抗炎、抗氧化应激、抗细胞凋亡，从而发挥治疗脑梗死的作用。

Objective To explore the mechanism of Gastrodia elata in treatment of cerebral infarction based on network pharmacology and molecular docking techniques, and to verify it by molecular dynamics simulation technology. Methods The active components and potential targets of G. elata were screened using SymMap and BATMAN-TCM 1.0 databases. The targets related to cerebral infarction were collected from GeneCards and DisGeNET databases. The protein-protein interaction network was constructed using STRING database and Cytoscape 3.10.1 software. The intersection targets of G. elata and cerebral infarction were imported into the DAVID database for gene ontology (GO) and Kyoto encyclopedia of genes and genomes (KEGG) pathway enrichment analysis. Cytoscape 3.10.1 was used to construct a network of “Gastrodia elata-active ingredients-potential targets-signaling pathways-cerebral infarction”. The active components and core targets were docked using Autodock Tools 1.5.7, and the affinity and stability of the binding models were analyzed through molecular dynamics simulation. Results Thirteen key active components of G. elata in treatment of cerebral infarction were screened, including hydroxybenzoic acid, 4-hydroxybenzylamine, γ-sitosterol, dauricine, citronellal, etc. There were 89 intersection target genes between drugs and diseases. The five highest moderate values in PPI network diagram were tumor necrosis factor (TNF), insulin (INS), interleukin-1β (IL-1β), prostaglandin endoperoxide synthase 2 (PTGS2), peroxisome proliferator-activated receptor γ (PPARG). There were 566 items in the GO analysis, in which biological processes (BP) were mostly enriched in signal release, regulation of inflammatory response, transport of organic hydroxyl compounds, etc. The cell composition (CC) was mostly enriched in neuronal cell bodies, secretory granule cavities, cytoplasmic vesicle cavities, etc. Molecular function (MF) was mostly enriched in steroids, protease binding and nuclear receptor activity. There were 88 KEGG enrichment results, including neuroactive interactions, human cytomegalovirus, lipid and atherosclerosis, 5-hydroxytryptaminergic synapses, and cyclic adenosine monophosphate (cAMP) signaling pathways. The results of molecular docking showed that the core active components of G. elata could well bind to the five target proteins with the highest moderate value in the PPI network diagram, forming relatively stable complexes. Molecular dynamics simulation further verified that γ-sitosterol had good affinity and binding stability with INS, dauricine and TNF, and citronellal and TNF. Conclusion G. elata may act on target genes such as TNF and INS through active ingredients such as γ-sitosterol, dauricine and citronellal, regulate signal release, regulate inflammatory response and cAMP signaling pathway to achieve anti-inflammatory, anti-oxidative stress and anti-apoptosis, thus playing a role in the treatment of cerebral infarction.

R965;R971

云南省教育厅科研项目（202101AY070001-283）