Objective: To identify the time dependent profiles of gene expression and featured co-expression network modules in peripheral blood of rats after middle cerebral artery occlusion (MCAO).
Methods: Microarray GSE119121 from GEO database was analyzed by R language to identify the significantly changed genes in peripheral blood at different time points (0, 1, 2, 3, 6 and 24 h) after MCAO. Gene expression patterns at different time courses were screened by STEM tools. Then, function annotation and pathway enrichment of differentially expressed genes (DEGs) were performed using the Gene Ontology (GO) database and the Kyoto Gene and Genomic Encyclopedia (KEGG) database. Depending on CEMiTool package, gene expression profile matrix was inputted into R to construct the co-expression networks and to analyze modules, and enrichment analysis was conducted to evaluate the correlation between the modules and different time points.
Results: Comparing with gene at 0 h, the numbers of DEGs in peripheral blood at different time points after MCAO were 163 (1 h), 502 (2 h), 527 (3 h), 550 (6 h), and 75 (24 h), respectively. Moreover, a total of 38 gene expression patterns were enriched, and pattern 65 and pattern 34 were specifically up-regulated or down-regulated at 2-6 h. Hp, Nos2, P2ry10, and Klf12 were representative genes of these two models. The co-expression network module analysis showed that the gene status in the early acute phase (1-6 h) was positively correlated with the Module 2. Module 3 and Module 4 was positively correlated with phase phase 1-3 h and 2-6 h, respectively. Noteworthy, Module 6 gradually changed from positive correlation (0-2 h) to negative correlation (3-24 h) with the MCAO time course, and Module 6 was mainly related to viral response and innate immune response. The hub genes of Module 6 included Mx1, Mx2, and Rtp4.
Conclusions: Our study has identified the featured genes and dynamic co-expression network modules in peripheral blood after acute ischemic stroke, which provides a potential basis for judging the onset time of ischemic stroke.
目的: 研究大脑中动脉栓塞（MCAO）大鼠不同造模时程外周血的特异表达基因及特征共表达网络模块。
方法: 利用GEO数据库的基因表达谱芯片GSE119121联合R语言分析MCAO造模后不同时间点（0、1、2、3、6及24 h）外周血的差异表达基因。通过STEM工具筛选不同造模时程基因表达模式。利用基因本体（GO）数据库和京都基因与基因组百科全书（KEGG）数据库对差异表达的基因进行功能注释和通路富集。在R语言环境下利用CEMiTool包对基因表达谱矩阵进行共表达网络构建及模块分析，并将模块与不同造模时间点进行富集分析。
结果: 与造模0 h相比，造模后1、2、3、6及24 h差异表达基因数分别为163、502、527、550、75。共有38种基因表达模式被富集，其中模式65和模式34分别在2~6 h特异上调或下调， Hp、 Nos2、 P2ry10及 Klf12为两种模式的代表性基因。共表达网络模块分析显示，造模急性期早期（1~6 h）基因状态与模块2正相关，造模1~3 h基因状态与模块3正相关，造模2~6 h基因变化与模块4正相关。基因模块6随着造模时间的迁移，与各时间点从正相关（0~2 h）逐渐转为负相关（3~24 h），模块6主要与病毒应答及固有免疫应答相关，其网络核心节点包括 Mx1、 Mx2、 Rtp4等基因。
结论: 本研究初步筛选了缺血性脑卒中急性发病期大鼠外周血的特征基因及动态共表达网络模块，为探究缺血性脑卒中的病理生理变化规律提供了依据。