| BackgroundBreast cancer is one of the most common malignancies worldwide and is a leading cause of cancer-related death in women.Although the screening,diagnosis,and treatment of breast cancer have improved,the prognosis is still bleak.Alterations in gene-specific methylation that occur early in carcinogenesis can be used for the early detection and prevention of cancer.Although several studies have reported abnormal DNA methylation modifications of certain genes in breast cancer,the comprehensive network map and functional pathways based on datasets from gene expression profile microarrays and gene methylation microarrays have not yet been revealed.The worldwide incidence of breast cancer continues to rise;thus,more sensitive and specific biomarkers related to the development of breast cancer are urgently needed for more accurate,early diagnosis and prognosis stratification of these patients.Research purposeThis study aimed to identify differentially expressed biomarkers driven by aberrant methylation in breast cancer and to explore potential pathological mechanisms using comprehensive bioinformatics analysis.Materials and methodsFirst,the gene expression profile microarray GSE45827 and the DNA methylation microarray GSE32393 were extracted from the publicly Gene Expression Omnibus(GEO)database.The GSE45827 gene expression data contained 130 breast cancer tissue specimens and 11 normal breast tissue specimens.The GSE32393 gene methylation data enrolled 137 samples,consisting of 114 breast cancer tissue samples and 23 adjacent normal tissue samples.The differentially expressed genes(DEGs)of gene expression GSE45827 were screened using the online GE02R tool(P-value<0.05 and |log FC|?2.0).For gene methylation profile GSE32393 data,the corresponding methylation site was converted to its mapped genome according to the platform annotation file,and then the aberrantly methylated genes were identified.The aberrantly methylated DEGs were selected using VENNY by overlapping the corresponding differential gene lists.The screened genes were subjected to GO analysis and KEGG pathway enrichment analysis using the online STRING database.The protein-protein interaction(PPI)network constructed by the STRING database was visualized with Cytoscape software,and the core genes were screened using the CytoHubba plugin.The results we obtained were then further verified in Oncomine and TCGA databases.The cBioPortal database was then employed to analyze the genetic alterations of the six core genes we identified.The prognostic roles of these six hub genes in breast cancer were evaluated using the Kaplan-Meier plotter tool.Finally,the UALCAN database was utilized to explore the relationship between the mRNA expression of core genes and clinical features(tumor stage,molecular subtype,and histological type).ResultsThrough the above analyses,we screened 618 upregulated genes and 1698 downregulated genes in the GSE45827 gene expression profile.In the gene methylation profile GSE32393,we screened 2604 genes that were hypermethylated and 3754 genes that were hypomethylated.We obtained 18 hypomethylated/highly expressed oncogenes by overlapping the lists of upregulated genes,hypomethylated genes,and oncogenes in VENNY.Similarly,we obtained 21 hypermethylated/downregulated tumor suppressor genes(TSGs)by overlapping the lists of downregulated genes,hypermethylated genes,and TSGs.GO analysis suggested that these 39 aberrantly methylated differentially expressed genes were mainly involved in the biological processes of cellular component movement and cellular metabolism.KEGG enrichment analysis revealed that differential genes were primarily enriched in the NF-?B and ATM signaling pathways.We screened six core genes based on the six algorithms in the CytoHubba plugin.The Hub genes included three hypomethylated upregulated oncogenes(BCL2,KIT,and RARA)and three hypermethylated downregulated TSGs(ATM,DICER1,and DNMT1).To validate the stability of the results we found,we first verified the expression levels of six core genes in the Oncomine database,and the results suggested that they were consistent with the previous data.Subsequently,we further verified the expression of core genes and DNA methylation status in the TCGA database.The results indicated that the promoter regions of BCL2,KIT,and RARA were highly hypomethylated,while those of ATM,DICER1,and DNMT1 were remarkably hypermethylated.Additionally,a negative correlation was found between the expression and the methylation status of the core genes.This finding further demonstrates the reliability and stability of our results.Next,to assess the prognostic value of the selected core genes in breast cancer,we conducted a survival analysis with six core genes,and the results revealed that patients with a high expression of BCL2 or KIT had longer overall survival,while lower expression of ATM or DICER1 was associated with a worse prognosis in breast cancer patients.Finally,further analysis showed that the mRNA expression levels of ATM and DICER1 were significantly different among different subgroups of clinical features,including clinical stage,molecular subtype,and histological type.ConclusionThrough comprehensive bioinformatics analysis,our study revealed a series of abnormally methylated differentially expressed genes and their related pathways in breast cancer,which provides new insight for further investigations into the pathogenesis of breast cancer.These aberrantly methylated oncogenes and TSGs,particularly ATM and DICER1,may serve as biomarkers for more accurate diagnosis and treatment strategies for breast cancer patients in the future.Compared with studies of single data types,this study provides new clues for the relationship between DNA methylation and gene expression alterations.We obtained more reliable and accurate screening results by combining multiple datasets.However,further molecular experiments are still needed to confirm the candidate genes and related pathways screened in our study. |
PDF Full Text Request