CANCER EPIGENETICS · Comprehensive characterization of epigentic reprogramming in cancer

Comprehensive characterization of epigentic reprogramming in cancer

Project context and objectives Abnormal gene expression is a key characteristic of cancer. Mutations or deletions of the deoxyribonucleic acid (DNA) are genetic changes that can cause such abnormalities. On the other hand, abnormal expression patterns can also be generated by changes in the way the long DNA molecules are folded and packaged in the nucleus of the cell. The processes that govern DNA folding and the resulting gene expression are called epigenetic mechanisms. Epigenetic defects and modifications also contribute to an uncontrolled growth of cells and to cancer formation. The best studied epigenetic modification is the methylation of DNA. In cancer states, regions of the DNA covering many genes are known to be 'hyper-methylated'. This methylation is correlated with silencing (deactivation) of genes in an irreversible manner. There is generally little known about the mechanisms regulating methylation in normal cells and aberrant methylation in cancer cells. Another epigenetic mechanism, which has lately been recognised as playing a role in cancer is a protein complex (named the polycomb complex), which binds (is attached) to specific gene regions and silences them by modifying the structure of the chromatin (a structure composed of DNA wrapped around proteins called histones). This reversible silencing occurs mainly in embryonic and developing cells and is important for normal development. It has recently been shown that genes that are methylated in cancer cells are those that are silenced by the polycomb system in normal cells during early development. The aim of our work was to study the interaction between the two epigenetic systems in order to shed further light on the loss of epigenetic regulation in cancer. This may have practical implications, as epigenetic silencing is potentially reversible and thus represents an attractive target for therapy in cancer. Knowledge of the aberrant mechanisms is the key to designing intervention strategies and targeted therapies. Work performed Our studies included simultaneous mapping of changes in DNA methylation and polycomb occupancy in normal, immortal and cancer cells over the entire genome (genome-wide approach). Main results We found that in some immortalised cells (cells that continuously divide, like cancer cells) there is very little DNA methylation, which is similar to normal cells. As these cells continue to grow and divide they gained more and more DNA methylation in a steady and gradual manner, becoming eventually highly methylated. The same cells had a high level of polycomb occupancy at developmental gene locations in early passages but as the cells grew a gradual depletion of polycomb marks was apparent. We termed this process 'epigenetic switching'. Expression analyses of the cells revealed that epigenetically modified genes were silenced throughout the process. However, reversible silencing by polycomb was switched to irreversible silencing by DNA methylation. We propose that such a switch reduces the ability of cells to respond to various signals and cues, which is one of the hallmarks of cancer cells. A paper reporting these results is currently under review for publication in 'Nature Genetics'. Next, we wanted to determine the generality of our findings from cell culture to real tumours and metastatic tissue from cancer patients. For this reason we were involved in establishing a tumour repository in the Sheba Medical Center with a joint collaboration of surgeons, medical oncologists and scientific personnel. We collected tens of fresh tumour and metastatic samples from patients undergoing curative or palliative procedures. We established methods to prepare the starting material from the small amount of fresh tissue available to be able to perform the polycomb occupancy studies. We then performed genome-wide DNA methylation and polycomb occupany profiling on several normal tissue, primary tumours and metastatic samples. We were able to show that, as we saw in cell lines, an accumulation of DNA methylation and diminished polycomb occupancy exist in metastatic cells (the most violent cancer cells compared to normal cells). We thus established that epigenetic switching is not just a phenomenon seen in a laboratory set-up, but rather that it occurs in real tumours. We are currently performing more experiments to enlarge our sample size and summarising our results for publication.

Data: CORDIS, © European Union