基于网络药理学和分子对接探讨桔梗皂苷D干预机械通气相关肺损伤的潜在作用机制

Exploring the Potential Mechanisms of Platycodin D in Ameliorating Mechanical Ventilation-Induced Lung Injury Based on Network Pharmacology and Molecular Docking

摘要: 目的：围绕桔梗皂苷D (Platycodin D, PD)与机械通气相关肺损伤(ventilator-induced lung injury, VILI)的潜在关联，利用网络药理学和分子对接方法筛选候选靶点及相关通路。方法：分别从PharmMapper、SEA、SuperPred、TCMIP等数据库整理PD预测靶点，并结合OMIM、GeneCards获得VILI相关基因。两类靶点经规范化、去重后取交集，进一步构建“药物–靶点–疾病”网络。随后在STRING平台建立蛋白质–蛋白质相互作用(PPI)网络，并借助Cytoscape和CytoNCA进行拓扑筛选。交集基因采用clusterProfiler进行Gene Ontology (GO)和Kyoto Encyclopedia of Genes and Genomes (KEGG)富集分析。对筛出的代表性靶点，则采用分子对接评价其与PD的结合情况。结果：共获得PD潜在靶点223个、VILI相关靶点2712个，交集靶点96个。PPI网络包含97个节点和1002条相互作用边；CytoNCA二轮筛选后得到MMP9、TNF、IL6、IL1B、AKT1等5个核心节点。GO富集结果主要指向脂多糖反应、细菌来源分子反应、白细胞迁移、炎症反应调控及PI3K/AKT信号正调控等生物过程；KEGG富集涉及TNF、IL-17、MAPK、趋化因子信号通路以及AGE-RAGE等与炎症和屏障损伤相关的通路。分子对接中，PD与MMP9的结合能最低(−9.2 kcal/mol)。结论：PD干预VILI的可能机制并非单一靶点作用，而可能与MMP9、TNF、IL6、IL1B、AKT1等节点共同参与的炎症放大、细胞迁移、基质重塑和氧化应激有关。本研究为后续实验验证提供了候选靶点和通路线索。

Abstract: Objective: This study aimed to explore the potential association between Platycodin D (PD) and ventilator-induced lung injury (VILI), and to identify candidate targets and related pathways using network pharmacology and molecular docking. Methods: Predicted targets of PD were collected from PharmMapper, SEA, SuperPred, TCMIP, and other databases. VILI-related genes were obtained from OMIM and GeneCards. After standardization and duplicate removal, the overlapping targets were identified and used to construct a drug-target-disease network. A protein-protein interaction (PPI) network was then built using the STRING platform, followed by topological screening with Cytoscape and CytoNCA. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses of the overlapping genes were performed using clusterProfiler. Representative targets were further selected for molecular docking to evaluate their binding affinity with PD. Results: A total of 223 potential PD targets and 2712 VILI-related targets were obtained, yielding 96 overlapping targets. The PPI network contained 97 nodes and 1002 interaction edges. After two rounds of CytoNCA-based screening, five core nodes were identified: MMP9, TNF, IL6, IL1B, and AKT1. GO enrichment mainly involved biological processes such as response to lipopolysaccharide, response to molecules of bacterial origin, leukocyte migration, regulation of inflammatory response, and positive regulation of PI3K/AKT signaling. KEGG enrichment highlighted pathways related to inflammation and barrier injury, including the TNF, IL-17, MAPK, chemokine signaling, and AGE-RAGE pathways. In the molecular docking analysis, PD showed the lowest binding energy with MMP9 (−9.2 kcal/mol). Conclusion: The potential mechanism by which PD may modulate VILI is unlikely to depend on a single target. Instead, it may involve a coordinated network centered on MMP9, TNF, IL6, IL1B, AKT1, and related nodes, contributing to the regulation of inflammatory amplification, cell migration, matrix remodeling, and oxidative stress. These findings provide candidate targets and pathway-level clues for subsequent experimental validation.