Integrated analysis of microRNA and gene expression profiles reveals a functional regulatory module associated with liver fibrosis
Integrated analysis of microRNA and gene expression profiles reveals a functional regulatory module associated with liver fibrosis
复制标题
microRNA和基因表达谱的综合分析揭示了与肝纤维化相关的功能调节模块
DOI：
10.1016/j.gene.2017.09.027
复制
发表时间：
2017-12-15
期刊：
影响因子：
3.5
通讯作者：
You, Hong
中科院分区：
文献类型：
--
作者：
Chen, Wei;Zhao, Wenshan;You, Hong
背景：肝纤维化以细胞外基质（ECM）蛋白过度积累为特征，是慢性肝脏炎症的最终共同途径。越来越多的证据表明 microRNA (miRNA) 失调对肝纤维化的不同阶段具有重要影响。然而，我们对此类疾病的 miRNA 基因调控细节的了解仍不清楚。方法：提取公开的肝硬化患者基因表达综合 (GEO) 数据集进行综合分析。使用 GEO2R 网络工具鉴定差异表达的 miRNA (DEM) 和基因 (DEG)。 DEM 的假定靶基因预测是使用五种主要算法的交集进行的：DIANA-microT、TargetScan、miRanda、PICTAR5 和 miRWalk。基于序列水平的计算目标预测以及DEM和DEG之间的逆表达关系构建了功能性miRNA基因调控网络（FMGRN）。选择 DAVID Web 服务器进行 KEGG 通路富集分析。功能性miRNA基因调控模块是根据生物学解释生成的。利用String数据库确定肝纤维化相关模块中基因之间的内部联系。与肝纤维化相关的 miRNA 基因调控模块在重组人 TGF beta 1 刺激和特定 miRNA 抑制剂处理的 LX-2 细胞中进行了实验验证。结果：与正常对照相比，我们在肝硬化活检样本中总共鉴定出了 85 个和 923 个失调的 miRNA 和基因。所有明显的 miRNA 基因对均被识别并组装成 FMGRN，FMGRN 由 51 个 miRNA 和 275 个基因之间的 990 个调控组成，形成两个大子网络，分别定义为下网络和上网络。 KEGG 通路富集分析显示，up 网络显着参与多个 KEGG 通路，其中“粘着斑”、“PI3K-Akt 信号通路”和“ECM-受体相互作用”显着（调整后 p < 0.001）。这些通路中富集的基因与其调控miRNA结合形成了一个功能性miRNA-基因调控模块，其中包含7个miRNA、22个基因和42个miRNA-基因连接。基于String数据库的基因相互作用分析表明，22个基因中有8个高度聚类。最后，我们通过实验证实了与肝脏相关的包含5个miRNA（miR-130b-3p、miR-148a-3p、miR-345-5p、miR-378a-3p和miR-422a）和6个基因（COL6A1、COL6A2、COL6A3、PIK3R3、COL1A1、CCND2）的功能调节模块。纤维化。结论：我们的 miRNA和基因表达谱的整合分析突出了与肝纤维化相关的功能性miRNA-基因调控模块，这在一定程度上可能为更好地理解肝纤维化的潜在发病机制提供重要线索。
Background: Liver fibrosis, characterized with the excessive accumulation of extracellular matrix (ECM) proteins, represents the final common pathway of chronic liver inflammation. Ever-increasing evidence indicates microRNAs (miRNAs) dysregulation has important implications in the different stages of liver fibrosis. However, our knowledge of miRNA-gene regulation details pertaining to such disease remains unclear.Methods: The publicly available Gene Expression Omnibus (GEO) datasets of patients suffered from cirrhosis were extracted for integrated analysis. Differentially expressed miRNAs (DEMs) and genes (DEGs) were identified using GEO2R web tool. Putative target gene prediction of DEMs was carried out using the intersection of five major algorithms: DIANA-microT, TargetScan, miRanda, PICTAR5 and miRWalk. Functional miRNA-gene regulatory network (FMGRN) was constructed based on the computational target predictions at the sequence level and the inverse expression relationships between DEMs and DEGs. DAVID web server was selected to perform KEGG pathway enrichment analysis. Functional miRNA-gene regulatory module was generated based on the biological interpretation. Internal connections among genes in liver fibrosis-related module were determined using String database. MiRNA-gene regulatory modules related to liver fibrosis were experimentally verified in recombinant human TGF beta 1 stimulated and specific miRNA inhibitor treated LX-2 cells.Results: We totally identified 85 and 923 dysregulated miRNAs and genes in liver cirrhosis biopsy samples compared to their normal controls. All evident miRNA gene pairs were identified and assembled into FMGRN which consisted of 990 regulations between 51 miRNAs and 275 genes, forming two big sub-networks that were defined as down-network and up-network, respectively. KEGG pathway enrichment analysis revealed that up network was prominently involved in several KEGG pathways, in which "Focal adhesion", "PI3K-Akt signaling pathway" and "ECM-receptor interaction" were remarked significant (adjusted p < 0.001). Genes enriched in these pathways coupled with their regulatory miRNAs formed a functional miRNA-gene regulatory module that contains 7 miRNAs, 22 genes and 42 miRNA-gene connections. Gene interaction analysis based on String database revealed that 8 out of 22 genes were highly clustered. Finally, we experimentally confirmed a functional regulatory module containing 5 miRNAs (miR-130b-3p, miR-148a-3p, miR-345-5p, miR-378a-3p, and miR-422a) and 6 genes (COL6A1, COL6A2, COL6A3, PIK3R3, COL1A1, CCND2) associated with liver fibrosis.Conclusions: Our integrated analysis of miRNA and gene expression profiles highlighted a functional miRNA-gene regulatory module associated with liver fibrosis, which, to some extent, may provide important clues to better understand the underlying pathogenesis of liver fibrosis.