Network-based meta-analysis and confirmation of genes ATP1A2, FXYD1, and ADCY3 associated with cAMP signaling in breast tumors compared to corresponding normal marginal tissues.

Breast cancer (BC) is a global health concern with a growing prevalence. Since BC is a heterogeneous cancer, transcriptome analyzes were carried out on breast tumor tissues relative to their corresponding normal tissues in order to identify gene expression signatures and perform meta-analysis. Five expression profiling by array data sets from breast tumor tissues and non-tumor neighboring tissues were retrieved following a search in the GEO database (GSE70947, GSE70905, GSE10780, GSE29044, and GSE42568). Meta-analysis of gene expression using the Network Analyst tool identified common differentially expressed genes and biological pathways in all data sets. Then, the DEGs were analyzed through PPI network construction, gene ontology, and pathway analysis. The detected hub genes underwent Kaplan-Meier (KM) plotter and UALCAN validation. Finally, Real-time PCR analysis was used on BC patients' samples to determine mRNA levels of cAMP signaling pathway members ATP1A2, FXYD1, and ADCY3. Breast tumor tissues showed 710 differentially expressed genes (DEGs), with 392 overexpressed and 318 underexpressed, compared to normal marginal tissues. On the EnrichR library, GO, and KEGG pathway analyses were performed on the DEGs list. Progesterone-mediated oocyte maturation and the NF-kappa B signaling system were upregulated DEGs' top deregulated signaling pathways. In contrast, pathways related to cancer and the cAMP signaling pathway were the most enriched terms for down-regulated genes. Next, Real-time PCR quantification of cAMP signaling cascade members ATP1A2, FXYD1, and ADCY3 was performed on 50 BC tumoral and non-tumoral tissues for validation. Results of meta-analyzed array data sets revealed DEGs representing BC gene signatures, and cAMP signaling pathway members as effective factors in BC. The results of our real-time PCR expression level determination for ATP1A2, FXYD1, and ADCY3 in breast tumor tissues relative to the normal margins contradicted our bioinformatics investigations, which found increased levels for these genes. Of these, only ATP1A2's expression levels were statistically significant. This study focused on identifying gene expression signatures that provide an invaluable source of evidence for BC-related underlying mechanisms to provide new therapeutic targets and biomarkers.