MiRNA-mRNA integrative analysis reveals epigenetically regulated and prognostic miR-103a with a role in migration and invasion of carboplatin-resistant ovarian cancer cells that acquired mesenchymal-like phenotype
DNA methylation, histone modifications, and miRNAs affect ovarian cancer (OC) progression and therapy response. Identification of epigenetically downregulated miRNAs in drug-resistant OC cell lines with a possible role in drug resistance and/or drug-induced mesenchymal-like phenotype. MiRNA profiling was performed on parental and carboplatin-resistant OC cells, MES-OV and MES-OV CBP. RT-qPCR validation, epigenetic modulation and other CBP-resistant OC cell lines were used to select miRNAs of interest. The integration of miRNA-predicted target genes and differentially expressed genes (DEGs), pathway and functional analysis were used for forecasting their biological role. Data mining was performed to determine their possible prognostic and predictive values. MiRNA profiling revealed 48 downregulated miRNAs in OC cells whose drug sensitivity and metastatic potential were impacted by epigenetic modulators. Of the fourteen selected, nine were validated as changed, and seven of these restored their expression upon treatment with epigenetic inhibitors. Only three had similar expression patterns in other OC cell lines. MiRNA-mRNA integrative analysis resulted in 56 target DEGs. Pathway analysis revealed that these genes are involved in cell adhesion, migration, and invasion. The functional analysis confirmed the role of miR-103a-3p, miR-17-5p and miR-107 in cell invasion, while data mining showed their prognostic and predictive values. Only miR-103a-3p was epigenetically regulated at the constitutive level. High throughput miRNA and cDNA profiling coupled with pathway analysis and data mining delivered evidence for miRNAs which can be epigenetically regulated in drug-resistant, mesenchymal-like OC cells as possible markers to combat therapy-induced short overall survival and tumor metastatic potential.
