1 陕西省人民医院心血管内科，西安 710068
2 空军军医大学第一附属医院临床免疫科，西安 710032
目的：对心肌肥厚模型——内皮素1(endothelin 1，ET1)刺激人诱导的多能干细胞源性心肌细胞(human induced pluripotent stem cell derived cardiomyocytes，hiPSC-CMs)的基因芯片数据进行生物信息学分析，探讨长链非编码RNA(long non-coding RNA，lncRNA)的表达谱及竞争性内源性RNA(competing endogenous RNA，ceRNA)调控网络。方法：在GEO(Gene Expression Omnibus)数据库中筛选心肌肥厚相关基因芯片，对下载的原始探针矩阵信息和平台注释文件进行预处理，获取基因表达矩阵信息，利用实用报表提取语言(practical extraction and report language，Perl)赋予基因不同的生物型属性，从而区分出lncRNA与mRNA。利用R语言limma包分别分析lncRNAs和mRNAs的基因表达差异。使用在线下载的数据库预测与每个lncRNA关联的microRNA(miRNA)。在此基础上，利用基于TargetScan，miRDB和miRTarBase的预测软件对miRNA的靶基因(mRNA)进行预测。以Cytoscape3.7.1软件构建lncRNA-miRNA-mRNA ceRNA调控网络。针对ceRNA调控网络中的靶基因，运用在线数据库DAVID6.8进行基因本体论(Gene Ontology，GO)富集分析，并通过在线工具KOBAS3.0进行京都基因和基因组百科全书(Kyoto Encyclopedia of Genes and Genomes，KEGG)通路富集分析。结果：与对照组相比，ET1刺激的hiPSC-CMs中19个lncRNA和717个mRNA表达差异具有统计学意义。其中，lncRNA表达上调的有4个，下调的有15个；mRNA表达上调的有312个，下调的有405个。成功构建了ceRNA调控网络，该网络中，显著上调的lncRNA包括AC017002和MIR210HG，分别与9，10个miRNA关联密切，显著下调的有LINC00342，其与9个miRNA关联密切。该网络中靶基因共32个，其中18个上调，14个下调。对靶基因进行GO富集分析，发现被显著富集到细胞增殖、质膜和蛋白结合等13个不同的GO子集中。KEGG富集分析发现：靶基因被富集到MAPK信号通路、细胞凋亡、Wnt信号通路、JAK-STAT信号通路、TNF信号通路等20个信号通路上。结论：本研究构建了ceRNA网络，为探讨lncRNA参与心肌肥厚致病机制提供了一个新的视角。LncRNA可能通过ceRNA调控网络，影响心肌肥厚发生发展。在ceRNA网络中有3个lncRNA(表达上调的为AC017002、MIR210HG，表达下调的为LINC00342)可能以竞争性“海绵吸附”miRNA的方式调控下游mRNA的表达水平，使AP1，c-MYC，PIM1表达上调，ASK1(即MAP3K5)，MSK1/2(即RPS6KA5)及SFRP1表达下调，并通过在MAPK信号通路、细胞凋亡、Wnt信号通路、JAK-STAT信号通路、TNF信号通路等途经参与心肌肥厚的病理生理过程。这些基因有可能成为潜在的治疗靶点，研发阻断上述lncRNA或下游靶基因的相关药物可能对治疗心肌肥厚有着深远的临床意义。
Bioinformatics analysis of lncRNA expression profiles and competing endogenous RNA network of myocardial hypertrophy model: human induced pluripotent stem cell derived cardiomyocytes stimulated by endothelin 1
CorrespondingAuthor: DIAO Jiayu
This work was supported by the Youth Science and Technology Rising Star Project of Shaanxi Province, China (2017KJXX-70).
Objective: To perform bioinformatics analysis on the gene chip data of myocardial hypertrophy model-human induced pluripotent stem cell derived cardiomyocytes induced by endothelin-1 (hiPSC-CMs), and to explore the expression profile of long non-coding RNAs (lncRNAs) and competing endogenous RNA (ceRNA) regulatory network. Methods: The gene chip related to cardiac hypertrophy in the Gene Expression Omnibus (GEO) database was screened, and the downloaded original probe matrix information and platform annotation file were preprocessed to obtain the gene expression matrix information. The different biotype attributes were assigned to the genes by using the Practical Extraction and Report Language (Perl), to distinguish lncRNAs from mRNAs. Gene differential expression analysis of lncRNAs and mRNAs was performed using the R language limma package. The microRNAs associated with each lncRNA were predicted using a database downloaded online. On this basis, target genes (mRNAs) of microRNAs were predicted using prediction softwares based on TargetScan, miRDB and miRTarBase. The lncRNA-miRNA-mRNA ceRNA regulatory network was constructed using Cytoscape 3.7.1 software. The Gene Ontology (GO) enrichment analysis and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis of target genes from the ceRNA network were performed by the online database DAVID6.8 and the web servers KOBAS3.0. Results: Compared with the control groups, there were 19 lncRNAs and 717 mRNAs expressed differentially in ET1-stimulated hiPSC-CMs with statistical significance, among which 4 lncRNAs up-regulated, 15 lncRNAs down-regulated, 312 mRNAs up-regulated and 405 mRNAs down-regulated. ceRNA regulatory network was constructed successfully, in which lncRNA AC017002 and MIR210HG were up-regulated, LINC00342 was down-regulated. The 3 lncRNAs related closely to 9, 10 and 9 miRNAs, respectively. There were 32 target genes in the network, among which 18 up-regulated and 14 down-regulated. GO enrichment analysis showed that the target genes were significantly enriched to 13 different GO subsets, including cell proliferation, plasma membrane, protein binding and so on. KEGG enrichment analysis found that the target genes were enriched to 20 signaling pathways, including the MAPK signaling pathway, apoptosis, Wnt signaling pathway, JAK-STAT signaling pathway, TNF signaling pathway and so on. Conclusion: Our study constructed the ceRNA network, providing a new perspective for exploring the pathogenic mechanism of lncRNAs involved in cardiac hypertrophy. LncRNAs may affect the pathogenesis of cardiac hypertrophy by regulating in the ceRNA network. In the ceRNA network, up-regulated AC017002 and MIR210HG and down-regulated LINC00342 may regulate the expression level of downstream mRNA by means of competitive “sponge adsorption” of miRNA, so that the expression of AP1, c-MYC and PIM1 were up-regulated, and the expression of ASK1 (MAP3K5), MSK1/2 (RPS6KA5) and SFRP1 were down-regulated. These genes were involved in the pathophysiological process of cardiac hypertrophy through MAPK signaling pathway, apoptosis, Wnt signaling pathway, JAK-STST signaling pathway and TNF signaling pathway. It may become a potential therapeutic target, and the development of drugs that block the above-mentioned lncRNAs and/or downstream target genes may have profound clinical significance in the treatment of cardiac hypertrophy.
lncRNA; ceRNA; cardiac hypertrophy; gene expression profile; differentially expressed genes; geo database