多囊卵巢综合征相关miRNA调控网络的构建及生物信息学分析

Establishment of miRNA-mRNA regulatory network in ovarian granulosa cells of patients with polycystic ovary syndrome

目的 利用生物信息学方法对多囊卵巢综合征(PCOS)患者卵巢颗粒细胞差异表达的miRNA靶基因进行分析,通过构建miRNA-mRNA调控网络,为PCOS潜在发病机制研究提供新思路。方法 选择PCOS患者卵巢颗粒细胞miRNA表达数据集GES84376作为分析对象,并利用Limma分析差异表达miRNA。利用在线分析网站对差异表达的miRNA靶基因进行预测。应用DAVID网站进行生物信息学分析,并利用SRTING网站及Cytoscape软件构建miRNA-mRNA调控网络。结果 共筛选出251个miRNAs,其中3个miRNA显著上调(miR-3188、miR-4433及miR-3135b)。富集通路(KEGG)分析显示,miR-3188、miR-4433及miR-3135b的靶基因在mTOR信号通路、MAPK信号通路及PI3K/Akt信号通路中富集程度最高。通过STRING网站进行蛋白互作分析,其互作程度最高的前10位关键基因分别为:MAPK1、MDM2、FRS2、AGT、IGF1R、HDAC1、AGO1、AKT3、MTOR及CREB1。通过构建miRNA-mRNA调控网络图,结果显示miR-3188、miR-4433及miR-3135b均参与多种靶基因调控并互有联系,且以miR-3188为核心。同时调控网络中以MAPK1、MDM2及FRS2为核心多个miRNA的共同作用。结论 以生物信息学为基础筛选出的miRNA及其相关靶基因可与PCOS的发生发展相关,其具体机制有待于后续的验证及深入探究。

Objective A bioinformatics approach was used to analyze the target genes in ovarian granulosa cells of polycystic ovarian syndrome(PCOS) patients, and to provide new ideas for the study of the potential pathogenesis of PCOS by constructing miRNA-mRNA regulatory networks. Methods The miRNA expression dataset GES84376 from ovarian granulosa cells of PCOS patients was selected for analysis, and differentially expressed miRNAs were analyzed using Limma. miRNA target genes were predicted using an online analysis website. The DAVID website was applied for bioinformatics analysis and miRNA-mRNA regulatory network was constructed. Results A total of 251 miRNAs were screened, among which 3 miRNAs were significantly upregulated(miR-3188, miR-4433 and miR-3135b). The enrichment pathway(KEGG) analysis showed that the target genes of miR-3188, miR-4433 and miR-3135b were the most enriched in mTOR, MAPK and PI3K/Akt signaling pathway. Top 10 key genes with the highest degree of interactions were MAPK1, MDM2, FRS2, AGT, IGF1R, HDAC1, AGO1,AKT3, MTOR and CREB1, respectively. By constructing a miRNA-mRNA regulatory network map, the results showed that miR-3188, miR-4433 and miR-3135b were involved in the regulation of multiple target genes and interconnected, and miR-3188 was the core. Meanwhile, MAPK1, MDM2 and FRS2 are at the core of the regulatory network, and they are subject to the joint action of multiple miRNAs. Conclusion The bioinformatics-based screening of miRNAs and their related target genes can be associated with the development of PCOS, and their specific mechanisms need to be validated and further explored.