Investigating colon tumor cell response to 5-Aza-2'-deoxycytidine: analysis of the CDX2 promoter and its methylation status .

Cytosine methylation silences gene activation and is indicative of an inactive chromatin state. This is especially true in tumors when tumor suppressor genes are silenced by methylation. Because methylation silencing is reversible, DNA methylation has been identified as a potential drug target. Methylation-inhibiting drugs such as 5-AZa-2'-deoxycytidine (5-Aza-CdR) have been shown to reverse methylation and induce differentiation in cell culture, and these drugs are now under investigation in numerous clinical trails. Most clinical investigations of 5-Aza-CdR have focused on malignant hematologic disease, and few clinical trials have been performed in solid tumors. Given the number of methylation-silenced tumor suppressor genes in solid tumors a better understanding the mechanisms and targets of hypomethylating agents is needed. The goal of this dissertation is to better define the cellular responses of colon cancer cells to 5-Aza-CdR. Our findings provide new insights about the effects of hypomethylation by 5-Aza-CdR in colorectal adenocarcinoma. I found the p53 DNA damage response pathway is activated in colorectal tumor cells after 5-Aza-CdR treatment, and that cells that express wild-type p53 are more sensitive to 5-Aza-CdR mediated growth arrest and toxicity. CDX2 expression is greatly reduced in the later stages of human colorectal carcinogenesis and in some human colorectal cancer-derived cell lines. I found that the CDX2 gene is re-expressed in colon tumor cells treatment with 5-Aza-CdR, and that there is a region of the CDX2 promoter that is selectively methylated in tumor cells that lack CDX2 expression and in normal cells of non-epithelial origin. These results indicate the methylation plays a role in regulating CDX2 expression in both normal and neoplastic tissues and suggests that cytosine methylation is important in the establishment and maintenance of tissue-specific gene expression. Building on my observation about 5-Aza-CdR and CDX2, I tested the role of CDX2 activation in the restoration of RA responsiveness in colon tumor sells, and found the CDX2 alone is not sufficient. Lastly, I identified an intergenic region upstream of CDX2 that is transcribed in a manner that correlates with CDX2 expression. These preliminary results in this area help to explain why an isolated intergenic region should be the target of methylation in tumor cells.