基于整合生物信息学探究黄葵胶囊治疗慢性肾脏病的机制

Exploring the Mechanisms of Huangkui(黄葵)Capsules in Treating Chronic Kidney Disease Based on Integrated Bioinformatics Analysis

目的:整合多种生物信息学方法对黄葵胶囊治疗慢性肾脏病(Chronic kidney disease,CKD)的机制进行综合分析。方法:本研究从文献中收集黄葵胶囊的化学成分,从Genecards、OMIM和TTD数据库中收集CKD疾病靶点,利用Cytoscape 3.8.0软件构建“药物-成分-靶点”网络和蛋白互作网络进行核心成分和靶点筛选。利用GEO数据库中的临床样品对黄葵胶囊治疗CKD的核心靶点进行差异分析,找出差异表达的核心靶点(SDECGs),并对这些靶点进行分析以及免疫细胞浸润分析。将CKD样品进行聚类并进行聚类间SDECGs表达分析、免疫细胞分析及富集分析。再把CKD样品进行加权基因共表达分析,分别得正常组与CKD组、CKD样品聚类组之间的模块关键基因,取交集,将交集基因以4种机器学习算法筛选,构建模型,从而得到重要性评分前五的特征基因,构建列线图。使用GEO数据集及分子对接对预测结果进行验证。结果:检索到黄葵胶囊化学成分43种,靶点393个,其中核心成分为杨梅素、槲皮素及棉皮素等7种;CKD与黄葵胶囊交集靶点247个,其中核心靶点25个;通过GEO差异分析可获得18个SDECGs,该18个靶点与免疫细胞表达多呈正相关;将CKD样品聚类后获得2个聚类(C1、C2),两个聚类涉及的SDECGs表达、免疫细胞表达及通路富集等都分别与正常组和CKD组对应。分析所得涉及的机制主要有:免疫炎症、氧化应激和细胞焦亡等过程。通过WGCNA结合支持向量机构构建模型,得到5个特征基因(Ccdc186、Smc5、Slc30a7、Coil、Dmxl1),该特征基因可预测CKD患者使用黄葵胶囊治疗的敏感度以及CKD的患病风险。结论:黄葵胶囊治疗CKD主要通过多成分作用于多靶点(ALB、AKT1、GAPDH等)及多途径(免疫炎症、氧化应激、细胞焦亡和自噬及组织修复等机制)发挥作用。本研究对CKD的临床治疗以及黄葵胶囊的机制探索有一定的指导意义。

Objective:To comprehensively analyze the mechanism of Huangkui(黄葵)Capsules in treating chronic kidney disease(CKD)by combining multiple bioinformatics methods.Methods:This study collected chemical ingredients of Huangkui Capsules from the literature,gathered CKD disease targets from Genecards,OMIM,and TTD databases,and used Cytoscape 3.8.0 software to construct a"drug-ingredienttarget"network and a protein interaction network for screening core components and targets.The significantly differently expressed core genes(SDECGs)of Huangkui Capsules in treating CKD were obtained through differential analysis of clinical samples in the GEO database,and the expression and immune cell infiltration of these targets were analyzed.Then,CKD samples were clustered and expression analysis,immune cell analysis,and enrichment analysis of SDECGs between clusters were performed.The module key genes of the normal group and CKD group,and the clustering groups of CKD samples were obtained through weighted gene correlation network analysis(WGCNA).The intersection of the key genes was selected,and these intersection genes were screened with four machine learning algorithms to construct a model,resulting in a column chart of the top five feature genes based on importance scores.Finally,the prediction results were validated using GEO dataset and molecular docking.Results:Huangkui Capsules were found to contain 43 chemical components and 393 targets,among which the core components were seven ingredients including myricetin,quercetin,and gossypetin.There were 247 intersection targets between CKD and Huangkui Capsules,25 of which were identified as core targets.Differential analysis using GEO unveiled 18 SDECGs,which were positively correlated with immune cell expression.Clustering of CKD samples resulted in two clusters(C1 and C2),and the expression of SDECGs,immune cell expression,and pathway enrichment of these two clusters basically corresponded to the normal group and CKD group.The mechanisms involved in the analysis mainly included immune inflammation,oxidative stress,cell pyroptosis and other processes.By combining WGCNA with support vector machine(SVM),a model was constructed and 5 feature genes(CCDC186,SMC5,SLC30A7,COIL,DMLX1)were obtained,which could predict the sensitivity of CKD patients to Huangkui capsules treatment and the risk of developing CKD.Conclusion:Huangkui capsules mainly treat CKD by acting on multiple targets(ALB,AKT1,GAPDH,etc.)and multiple pathways(immune inflammation,oxidative stress,cell necrosis,autophagy,and tissue repair)through multiple components(myricetin,quercetin,gossypetin,etc.).This study provides guidance for the clinical treatment of CKD and the exploration of the mechanism of Huangkui capsules.