乳腺癌脑转移差异表达基因的生物学功能分析

乳腺癌脑转移(breast cancer brain metastasis,BCBM)的发病机制尚未明确。为了探究BCBM的发病机制,对BCBM差异表达基因的生物学功能进行研究并筛选关键调控基因。从基因表达综合数据库(gene expression omnibus,GEO)下载4个BCBM基因表达谱数据(GSE12237、GSE100534、GSE125989以及GSE43837),采用R语言筛选差异表达基因,采用富集分析包括基因本体分析(gene ontology,GO)和京都基因与基因组百科全书分析(Kyoto encyclopedia of genes and genomes,KEGG)进行生物学功能分析,采用STRING和Cytoscape分析蛋白质相互作用网络,采用Kaplan-Meier进行生存分析。结果表明,同时存在于2个及以上基因表达谱数据中的差异表达基因261个,GO分析主要涉及细胞外基质组织、细胞外结构组织等生物过程,细胞外基质结构组成、胶原结合等分子功能,含有胶原的细胞外基质、胶原蛋白三聚物等细胞组分;KEGG分析主要涉及蛋白质消化和吸收、局部黏附等通路。蛋白质相互作用网络分析得到9个关键调控基因,其中,DCN、COL6A1与BCBM的生存率显著相关,可作为潜在的BCBM关键调控基因,并为BCBM分子机制的研究提供思路。

Biological function analysis of differentially expressed genes in brain metastasis breast cancer

To explore the pathogenesis of breast cancer brain metastasis (BCBM), the biological functions of differentially expressed genes in BCBM were investigated and the hub genes of BCBM were screened in this study. Four gene expression profiles of BCBM datasets (GSE12237, GSE100534, GSE12598, and GSE43837) were retrieved from the gene expression omnibus (GEO) database and screened by R packages to obtain the differentially expressed genes of BCBM. The functional annotations of differentially expressed genes were conducted by gene ontology (GO) analysis. The enrichment analysis of signal pathways of differentially expressed genes was explored by Kyoto encyclopedia of genes and genomes (KEGG) analysis. The protein-protein interaction network was structured by the STRING database to screen the hub genes that might participate in BCBM. Finally, survival analysis was conducted by Kaplan-Meier survival analysis database. As a result, a total of 261 differentially expressed genes present in 2 or more datasets were screened out. Functional analysis showed that those differentially expressed genes were related to biological processes such as extracellular matrix organization and extracellular structural organization, molecular functions such as extracellular matrix structural composition and collagen binding, and cellular components such as collagen-containing extracellular matrix and collagen trimer. Signal pathways were mainly enriched in protein digestion and absorption, focal adhesion pathways. Nine hub genes were screened by using protein interaction network analysis, among which two hub genes (DCN, COL6A1) were significantly correlated with the survival probability of BCBM. The two hub genes could serve as potential hub genes for BCBM and might provide ideas for molecular mechanism research.