目的基于miRNA-mRNA网络探讨肝细胞癌和抑郁症的共病机制。方法(1)在GEO数据库中获取与肝细胞癌和抑郁症相关的miRNA微阵列数据;利用在线工具GEO2R分别分析两个数据集的差异表达miRNA(DE-miRNA),然后取交集。在TargetScan、miRDB、miRPathDB和miRWalk数据库中预测各个DE-miRNA的靶基因,然后取交集。(2)利用DAVID数据库对DE-miRNA靶基因进行功能富集分析及通路富集分析,并筛选出风险通路。(3)利用STRING 数据库和Cytoscape 3.9.1软件构建DE-miRNA靶基因的蛋白-蛋白相互作用网络,然后筛选中枢基因及关键基因。(4)利用Microsoft Excel软件及Cytoscape 3.9.1软件构建DE-miRNA-mRNA-风险通路可视化网络。(5)利用GEPIA2数据库进行中枢基因在肝细胞癌中的表达分析及生存分析。结果(1)肝细胞癌和抑郁症共同表达的上调DE-miRNA为miR-1290、miR-3156-5p,下调DE-miRNA为miR-575、miR-6737-3p,四者对789个靶基因进行调控。(2)DE-miRNA靶基因参与的生物功能与免疫、炎症、血管生成、神经元信号传导等相关。共筛选出15条风险通路,包括丝裂原活化蛋白激酶信号通路、磷脂酰基醇3-激酶/蛋白激酶B信号通路、轴突引导、催产素信号通路、神经营养素信号通路等。(3)共获得10个中枢基因,分别为BRCA1、NOTCH1、PTEN、EGFR、ATRX、ERCC4、RAD51B、EME1、RMI1、XRCC2。将富集在风险通路上的3个中枢基因作为关键基因。(4)构建的DE-miRNA-mRNA-风险通路网络包含了10对DE-miRNA-中枢基因组合,以及13个DE-miRNA-关键基因-风险通路组合。(5)与正常组相比,BRCA1、PTEN、ATRX、ERCC4、RAD51B、EME1、RMI1、XRCC2在肝细胞癌组中呈高表达,NOTCH1、EGFR在肝细胞癌组中呈低表达(P<0.01)。BRCA1、EME1、RMI1、XRCC2与肝细胞癌患者的生存率有关(P<0.05)。结论miR-1290、miR-3156-5p、miR-575、miR-6737-3p可能是肝细胞癌和抑郁症共同的调控因子,其通过作用于多个靶基因从而调控炎症反应、肿瘤免疫反应、神经传导等途径,进一步参与肝细胞癌和抑郁症的发生、发展。此外,BACR1、EME1、RMI1、XRCC2这4个靶基因与肝细胞癌患者的预后密切相关。
ObjectiveTo explore the comorbidity mechanism of hepatocellular carcinoma and depression based on miRNA-mRNA network.Methods(1) The microarray data of miRNA related to hepatocellular carcinoma and depression were obtained from the GEO database. The differentially expressed miRNAs (DE-miRNA) of the two datasets were analyzed by using the online tool GEO2R, respectively, and then the intersection was acquired. Target genes of various DE-miRNA were predicted in the databases of TargetScan, miRDB, miRPathDB, and miRWalk for obtaining the intersection. (2) The functional enrichment analysis and pathway enrichment analysis were performed on DE-miRNA target genes by employing the DAVID database, and the risk pathways were screened. (3) The protein-protein interaction network of DE-miRNA target genes was established by the application of the STRING database and Cytoscape 3.9.1 software, and then the central genes and key genes were screened. (4) The Microsoft Excel software and Cytoscape 3.9.1 software was used to establish DE-miRNA-mRNA-risk pathways visualized network. (5) The GEPIA2 database was employed to perform expression and survival analyses of central genes in hepatocellular carcinoma.Results(1) The up-regulated DE-miRNA of hepatocellular carcinoma and depression co-expressions were miR-1290, miR-3156-5p, down-regulated DE-miRNA were miR-575, miR-6737-3p, and the four regulated 789 target genes. (2) The biological functions involved in DE-miRNA target genes were related to immunity, inflammation, angiogenesis, and neuronal signal transduction, etc. A total of 15 risk pathways were screened, including mitogen-activated protein kinase signaling pathway, phosphatidylinositol 3-kinase/protein kinase B signaling pathway, axonal guidance, oxytocin signaling pathway, and neurenergen signaling pathway, etc. (3) A total of 10 central genes were obtained, concerning BRCA1, NOTCH1, PTEN, EGFR, ATRX, ERCC4, RAD51B, EME1, RMI1, and XRCC2, respectively. Three central genes enriched in risk pathways were regarded as key genes. (4) The established network of DE-miRNA-mRNA-risk pathways contained 10 pairs of DE-miRNA-central genes combinations, and 13 combinations of DE-miRNA-key genes-risk pathways. (5) Compared with the normal group, BRCA1, PTEN, ATRX, ERCC4, RAD51B, EME1, RMI1, and XRCC2 were high-expressed in the hepatocellular carcinoma group, and NOTCH1 and EGFR were low-expressed in the hepatocellular carcinoma group (P<0.01). BRCA1, EME1, RMI1, and XRCC2 were related to the survival rate of patients with hepatocellular carcinoma (P<0.05). ConclusionMiR-1290, miR-3156-5p, miR-575, and miR-6737-3p may be the co-regulated factors of hepatocellular carcinoma and depression. They may thus regulate approaches to inflammatory responses, tumor immune responses, nerve conduction, etc.,and further participate the occurrence and development of hepatocellular carcinoma and depression through acting on multiple target genes. In addition, the four target genes of BRCA1, EME1, RMI1, and XRCC2 are closely related to prognosis of patients with hepatocellular carcinoma.