Prostate cancer is the second leading cancer related cause of death in American men, with the vast majority of the mortality arising from the androgen independent and/or metastatic forms of the disease. Although many studies have clearly demonstrated that CpG island hypermethylation plays a significant role in prostate cancer, as in other types of cancer, genomic scanning approaches to identify novel targets of CpG island methylation have been lacking. Genomic scanning approaches have demonstrated cancer type specific methylation changes in many other types of cancers suggesting that similar data could be obtained for prostate cancer. Furthermore, studies focusing specifically on the androgen independent and/or metastatic forms of prostate cancer have been very limited. Addressing these questions in human prostate cancer is confounded by a number of difficulties. We therefore propose to take advantage of the TRAMP mouse model of prostate cancer to identify novel targets of CpG island hypermethylation. We will study primary tumors, metastatic tumors, and recurrent androgen independent tumors, thus surveying a set of phenotypes reflective of the human disease that have high clinical relevance. We will use a restriction landmark genomic scanning approach to identify novel targets in the mouse model. We will then translate the discoveries made in the TRAMP mouse model to the human disease. To accomplish this, we will identify human genomic sequences that are orthologous to the novel targets identified in the mouse genome. We will then use a highly sensitive novel technology, MassArray Quantitative Methylation Analysis (MA-QMA), to determine if the methylation found in the mouse model is conserved in human prostate cancer. This technique uses PCR products that would normally be cloned for bisulfite sequencing analysis and instead uses MALDI-TOF analysis to quantitatively determine the amount of methylation at each CpG dinucleotide within the PCR product. This work will serve 2 major purposes: 1st, this will lead to the identification of a novel set of CpG islands that become hypermethylated in human prostate cancer and may identify events specific for advanced phenotypes; 2nd this work will serve to validate the TRAMP mouse model as a useful tool for further research into the epigenetics of prostate cancer. ? ? ? ?
