| Methylation of cytosine is a process carried out by DNA methyltransferases. In mammals, this modification is central to many physiologic processes such as X chromosome inactivation, imprinting, and transcriptional regulation. In certain pathophysiologic states such as carcinogenesis, gene promoter CpG islands become aberrantly hypermethylated, resulting in transcriptional silencing and loss of gene expression. Inactivation of tumor suppressor genes can then lead to cancer formation and progression. One fundamental question in carcinogenesis is how does de novo methylation occur? In this dissertation, we examined the proliferative inflammatory atrophy (PTA) lesion of the prostate in which hypermethylation of GSTP1, the most common somatic change in prostate cancer, is first seen. PTA lesions were analyzed by immunohistochemistry and found to contain activated macrophages expressing inducible nitric oxide synthase. Treatment of cancer cells as well as normal prostate epithelial cells with nitric oxide resulted in increased single copy gene and LINE1 methylation as well as global 5-methylcytosine content. Incubation of purified recombinant DNA methyltransferases with nitric oxide resulted in increased enzymatic activity. These observations suggest that nitric oxide, an important signaling molecule produced by inflammatory cells, can play a significant role in initiating DNA methylation changes observed in epithelial cells. Furthermore, we showed that nitric oxide treatment of DNA methyltransferase 1 resulted in its S-nitrosylation, a post-translational modification that is increasingly becoming recognized as an important modulator of protein conformation and function. Although aberrant DNA methylation is a virtually universal alteration seen in cancer, we have very few classes of pharmacologic agents that act on DNA methyltransferases. The most intensively studied are nucleoside analogs that effectively trap the enzyme onto DNA during a catalytic cycle. These compounds are very effective at causing gene demethylation and reactivation but also have toxic side effects. Thus, there has been much interest in developing non-nucleoside analog inhibitors of DNA methyltransferases. A promising candidate is procainamide, which can reactivate GSTP1 in LNCaP prostate cancer cells. We found that procainamide is a specific partial competitive inhibitor of DNMT1 both in cell culture and in vitro. Procainamide acts by lowering the affinity of DNMT1 for its two substrates, S-adenosylmethionine and hemimethylated DNA. Moreover, procainamide profoundly reduces the processivity of DNMT1. We hope these findings will shed light on the mechanisms by which DNA methyltransferase activity can be enhanced or inhibited, thus providing insight on pathogenesis as well as the design of novel drug therapies. |
