Abstract: Objective　To investigate the potential pathogenesis and biomarkers by analyzing the gene chip data from temporal lobe epilepsies（TLE）patients with an history early of complex febrile seizures（CFS）and controls by using bioinformatic tools. Methods　The gene chip （GSE28674） was downloaded from GEO database.The chip data were collected and uploaded by the Paediatrics Department of the University Hospital of Sao Paulo in 2011，including the hippoca3 transcriptomic data of 18 children with temporal lobe epilepsy.R soft ware was employed for the identification of the differentially expressed genes（DEGs）of TLE-HS with CFS compared with those without FS.The gene ontology（GO）and the Kyoto encyclopedia of genes and genomes（KEGG）analysis were performed to establish protein-protein interaction network and hub genes were found，Which were used to make ROC curve performance analysis. Results　GSE28674 consisted of 6 TLE-HS patients with CFS and 12 TLE-HS patients without FS.We captured a total of 254 DEGs，of which 236 were upregulated and 18 were downregulated.The results of GO and KEGG indicated that these DEGs were significantly enriched in the pathways of GABAergic synapse，retrograde endocannabinoid signaling，neurotransmitter secretion and calmodulin binding.The top 10 hub genes with the highest connectivity in the PPI network were GABRG2，NRXN1，GABRA1，KCNQ2，GRIN2A，SYT1，GRIN1，SNAP25，SYP，SLC32A1，among which GABRA1，NRXN1 and GRIN2A had high diagnostic value. Conclusion　The 10 hub genes screened in this study can be used as diagnostic biomarkers of early CFS injury promoting TLE-HS.GABRG2 and GABRA1 may participate in early CFS injury promoting TLE-HS through GABAergic synaptic and retrograde endogenous cannabinoid signaling pathway.

Key words: complex febrile seizures, temporal lobe epilepsies, bioinformatic analysis, gene chip

摘要： 目的　使用生物信息学分析对早期具有复杂性热性惊厥病史的颞叶癫痫伴海马硬化患者及对照样本的基因芯片数据进行分析，探索颞叶癫痫与早期复杂性热性惊厥的诊断生物标志物及潜在发病机制。方法　从GEO数据库中下载基因芯片（GSE28674），该芯片数据是圣保罗大学医院儿科于2011年收集上传，包括了18例颞叶癫痫患儿海马CA3区转录组数据。利用R软件鉴定有CFS的TLE-HS样本相对于无FS的TLE-HS样本的表达差异基因（differentially expressed genes，DEGs），进行功能和通路富集分析，构建蛋白质-蛋白质互相作用网络（PPIs），寻找hub基因，以hub基因为标志物进行ROC曲线效能分析。结果　本研究采用的芯片GSE28674共包含18个样本，其中6例具有CFS病史的TLE-HS患者，12例无FS。从中共鉴定出254个差异基因，236个上调基因，18个下调基因。GO和KEGG结果表明DGEs主要富集在GABA能突触通路，逆行内源性大麻素信号通路，神经递质分泌，钙调蛋白结合等通路。PPI网络显示连接度最高的hub基因10个，分别是GABRG2，NRXN1，GABRA1，KCNQ2，GRIN2A，SYT1，GRIN1，SNAP25，SYP，SLC32A1，其中GABRA1、NRXN1和GRIN2A具有较高诊断价值。结论　本研究筛选的10个hub基因可作为早期CFS损伤促进TLE-HS发生的诊断生物标志物，GABRG2、GABRA1可能通过GABA能突触通路、逆行内源性大麻素信号通路参与早期CFS损伤促进TLE-HS的发生。

关键词: 复杂性热性惊厥, 颞叶癫痫, 生物信息分析, 基因芯片