Examination of the role of DNA methylation changes in prostate cancer using the Transgenic Adenocarcinoma of Mouse Prostate (TRAMP) model.

摘要

The objectives of this dissertation are (1) to characterize DNA methyltransferase (Dnmt) expression and aberrant DNA methylation throughout TRansgenic Adenocarcinoma of Mouse Prostate (TRAMP) tumor progression; (2) to examine the effects of genetically reduced Dnmt1 expression on TRAMP tumorigenesis; (3) to utilize TRAMP to determine whether green tea polyphenols (GTPs) act as Dnmt inhibitors in vivo. Our overall hypothesis is that TRAMP tumors are characterized by alterations in DNA methylation and that inhibition of Dnmts will alter TRAMP tumor development.;We next identified TRAMP tumor phenotype-specific patterns of DNA hypermethylation, including promoter-specific hypermethylation, coinciding with decreased gene expression. In addition, we found several genes in which downstream hypermethylation correlates with robust mRNA overexpression. We examined the kinetic relationship between these two phenomena in TRAMP and our results indicate that increased transcription precedes, and therefore may initiate, the downstream hypermethylation of these genes. Interestingly, removal of downstream gene hypermethylation in TRAMP cell lines led to decreased expression of these genes, suggesting that downstream hypermethylation contributes to the maintenance of increased gene expression. Global analysis of DNA methylation reveals decreased 5-methyl-2'-deoxycytidine (5mdC) levels in both WD and PRIM samples and hypomethylation of repetitive elements was observed in all sample types analyzed, compared to normal prostate. Overall, we find that DNA hypomethylation begins at early stages and locus-specific DNA hypermethylation, which is associated with altered gene expression, occurs primarily during late stages of TRAMP tumor progression.;We have utilized Dnmt1 hypomorphic mice in a wild type (WT) or TRAMP genetic background to examine normal prostate development and prostate cancer progression in the presence of low level Dnmt1 expression (8). Prostate tissue morphology and differentiation marker staining were normal in Dnmt1 hypomorphic mice. We next confirmed decreased Dnmt1 expression in TRAMP Dnmt1 hypomorphic mouse prostates as compared to TRAMP prostates. We observe increased tumor incidence and pathological progression in TRAMP Dnmt1 hypomorphic mice at 12 and 15 weeks of age. In contrast, we find decreased primary tumor incidence and pathological progression, and reduced metastatic incidence in DNMT1 hypomorphic mice at 24 weeks.;To determine how changes in TRAMP tumor progression may be associated with DNA methylation, we examined global and repetitive element methylation, which were significantly decreased in TRAMP Dnmt1 hypomorphic mouse prostates as compared to TRAMP prostates. Interestingly, we examined locus-specific promoter and downstream hypermethylation and found decreased methylation for only the promoter hypermethylated locus in Dnmt1 hypomorphic samples. This suggests that Dnmt1 plays a more prominent role in establishing or maintaining promoter methylation than it does for downstream methylation. Correlation analyses revealed that the levels of DNA hypomethylation and downstream hypermethylation were associated with tumor progression.;Several studies indicate that consumption of GTPs is associated with decreased prostate cancer incidence in humans and several polyphenolic components of green tea, with the main component epigallocatechin-3-gallate (EGCG), have been shown to reverse locus-specific hypermethylation in cancer cell lines (9-12). We sought to test whether the oral consumption of green tea can affect normal or cancer-specific DNA methylation in vivo. We provided 0% or 0.3% GTPs to WT and TRAMP mice in their drinking water from 4 to 15 or 24 weeks of age. This concentration of green tea is equivalent to drinking approximately 9 cups of green tea per day. We did not observe a clear change in global or locus-specific DNA methylation in either WT or TRAMP prostate samples. Furthermore, GTP consumption did not inhibit TRAMP tumor progression, in spite of activating known molecular markers of GTP activity in the prostate. Even when we utilized a range of GTP concentrations, including 0.1%, 0.3%, or 0.6%, there was no consistent DNA hypomethylation observed. In summary, we have shown that oral infusion of green tea does not alter normal or tumor-specific DNA methylation patterns in the mouse prostate. (Abstract shortened by UMI.)