Identification of a novel experimental model to reveal mechanisms leading to epigenetic changes and subsequent activation of cancer testis genes in cancer

Abstract

Epigenetic aberrations are frequently observed in cancer. Tumor-suppressor genes are often repressed with anomalous hypermethylation in cancer, while DNA hypomethylation has been identified in repetitive sequences and promoter regions of cancer testis (CT) genes resulting in genomic instability. Although it has been shown that CT genes are often regulated by dissociation of repressive proteins from promoter-proximal regions and epigenetic mechanisms, including DNA methylation, histone methylation and acetylation, the process leading to epigenetic changes and de-repression of CT genes remains largely unknown. This study aimed to reveal molecular mechanisms which may have role in coordinating CT gene expression. For this purpose, we designed two groups of experiments. The first was based on extending our previous observations related to two genes (ALAS2, CDR1) which showed inverse expression patterns, compared to CT genes in cancer cell lines. The ex vivo analysis of expression patterns of these genes, however, did not support an inverse relation between their expression and that of CT genes. The second approach was based on categorizing cancer cells into CT-high, CT-intermediate and CT-low groups to define differentially expressed non-CT genes that could help explain mechanisms underlying epigenetic changes and subsequent activation of CT genes. Surprisingly, we could not identify any transcripts that differentially expressed between these subgroups. We therefore, hypothesized that non-overlapping and distinct mechanisms could be involved in the upregulation of CT genes in different tumors. As our earlier work suggested a relationship between epithelial to mesenchymal transition (EMT) and CT expression we asked if an EMT based classification could help elucidate these mechanisms. Indeed, differential genes and differentially activated signaling pathways were discovered when cancer cells were first grouped by their EMT status. This helped us identify candidate proteins (BMI1, PCGF2, RB1 and RBL1) and pathways including MAPK/ERK and PTEN/PI3K pathways which can coordinate CT gene expression in cancer. Thirdly, we investigated clinical relevance of high CT gene expression in triple negative breast cancer by attempting to correlate this with drug sensitivity. Drug sensitivity against panobinostat showed correlation with CT gene expression. In summary, this study suggests new approaches to elucidate mechanisms which coordinate epigenetic aberrations in cancer and how these can be utilized for cancer therapy.