Denise S. O'Keefe, Assistant Professor
PhD, University of Adelaide, Australia
Email: okeefeds@upmc.edu
Research Interest:
The most common molecular changes seen in prostate cancer are epigenetic (non-mutational) in nature, such as the hypermethylation of gene promoter sequences. In fact, more than 90% of prostate cancers exhibit alterations in the DNA structure of particular genes, resulting in the "switching-off" of a number of tumor-suppressor type genes. Our lab investigates the potential mechanisms leading to these epigenetic changes to better understand the etiology of prostate cancer and thus develop novel tools for diagnosis, prognosis and treatment of the disease.
Field Defects and Prostate Cancer
A field defect describes an area of histologically normal tissue that contains epigenetic or mutational defects also seen in tumors from adjacent tissue. Thus, these changes likely represent some of the earliest events in carcinogenesis. The silenced genes marking the field defect may make ideal therapeutic targets since it is theoretically possible to use drugs to reverse epigenetic changes and switch the genes back on. Mining microarray data, we have identified several genes with decreased expression in prostate cancer, for example the gene that encodes the pro-apoptotic protein ASC. Real-time PCR demonstrated the down-regulation of these genes in prostate cancer cell lines, but not in normal prostate lines. This decrease in expression resulted from promoter methylation. Using methylation-specific PCR on microdissected patient specimens, we showed that methylation occurred not only in tumors, but also in high-grade prostatic intraepithelial neoplasia and normal tissue adjacent to the tumor. Importantly, we did not detect methylation in prostate tissue from cancer-free donors. These observations suggest that our genes act as markers for a field defect. Currently, we are further characterizing the other genes we have identified, and their roles in prostate cancer initiation and progression.
Methylation Profiling as a Clinical Tool in Prostate Cancer
Our lab is also working toward establishing prostate cancer methylation profiles, patterns of methylated and inactivated genes, to predict the clinical behavior of particular tumors. This capability would allow physicians to tailor treatment specifically to each patient. Methylation profiling, already proven useful in the management of other cancers, has particular relevance to prostate cancer. Based on clinical T stage, biopsy Gleason score and preoperative serum PSA, prostate cancer patients are stratified into three risk groups for recurrence following radical prostatectomy: low, intermediate and high. These criteria, however, lack prognostic accuracy for the intermediate group, underscoring the need for new markers. Another important problem in prostate cancer is the management of patients with high-grade prostatic intraepithelial neoplasia (HGPIN), a putative precursor to prostate cancer. The presence of HGPIN on a needle biopsy often, but not always, indicates the presence of cancer within the prostate, necessitating additional biopsies. Furthermore, the cancer might not be adjacent to or even on the same side of the prostate that contains the HGPIN. Thus, methylation profiles of HGPIN that could accurately predict the presence of cancer would be very beneficial. We can measure, relatively inexpensively, promoter methylation on both biopsy and paraffin-fixed archived samples. Hence, we are fortunate that, through Urology, we have access to one of the largest and most comprehensive prostate tissue banks in the U.S., allowing us to thoroughly test our hypotheses and develop accurate approaches toward methylation profiling.
Targeted Chromatin Remodeling Therapy for Androgen Independent Prostate Cancer
The inability to adequately treat androgen-independent prostate cancer causes the majority of patient deaths, prompting us to explore new therapeutic strategies. The androgen dihydrotestosterone binds to the androgen receptor (AR), inducing a homodimer formation that activates expression of genes with an androgen response element in their promoter. As androgen stimulates prostate cell growth and inhibits glandular epithelial cell death, androgen ablation therapy decreases tumor size. However, the vast majority of patients eventually develop androgen-independent tumors. How androgen-independent tumors form remains incompletely understood, but in some cases the AR gene is amplified, allowing small amounts of androgens to induce AR target genes, while in other cases the gene is mutated, increasing the promiscuity of AR. The aim of this project is to shut down the AR target genes using a novel technique developed by us: targeted chromatin remodeling.
Diet and Disease: The Role of Folate in the Initiation and Progression of Prostate Cancer
Many factors can reportedly alter the epigenetic status of DNA; however, folic acid is of notable importance. Folate, as part of the one carbon cycle, is critical for the synthesis of S-adenosyl methionine, which contributes the methyl group for methylation reactions. In 1998 the U.S. government mandated fortification of the food supply with folate, resulting in a tripling of the mean serum folate level in the U.S. population. Recent studies indicate that dietary folate may affect epigenetic programming, even in adults. It also seems that the epigenetic changes that occur differ, depending on the tissue type and the time at which dietary manipulation occurs. The prostate has a particularly high requirement for folate and seems especially susceptible to alterations in DNA methylation. Moreover, epidemiological evidence suggests that higher plasma folate levels increase the rate of prostate cancer. These findings have led us to study the effects of dietary folate on prostate cancer. Indeed, we have evidence that two proteins intimately involved in intracellular folate regulation contribute to prostate cancer. PSMA, a unique folate hydrolase that may increase intracellular folate levels, can induce invasive prostate cancer in a mouse model. At the same time, we have identified a folate transporter that is down-regulated in prostate tumors, and undergoes epigenetic silencing in prostate cancer cell lines. Interestingly, this down-regulation occurs under conditions of folate deprivation in an in vitro model of prostate cancer. Currently, we are evaluating global and specific epigenetic changes in primary human prostate tissue under different conditions of dietary folate. We are also studying if PSMA mediates epigenetic changes when it induces invasive cancer in a mouse model and the mechanism by which low folate results in the epigenetic silencing of specific genes.
Recent Publication
Kelavkar UP, Harya N, Hutzley J, Bacich DJ, Monzon FA, Chandran U, Dhir R and O'Keefe, DS (2007). DNA methylation paradigm shift: 15-lipoxygenase-1 upregulation in prostatic intraepithelial neoplasia and prostate cancer by atypical promoter hypermethylation. Prostaglandins & other Lipid Mediators, 82:185-197.
Collard, RL, Harya, NS, Monzon, FA, Maier, CE, O'Keefe, DS Methylation of the ASC Gene Promoter is Associated with Aggressive Prostate Cancer (2006). The Prostate, 66:687-695.
Bacich D.J., Wozniak, K., Lu, M., O'Keefe, DS, Callizot, N, Heston, WDW and Slusher, BS Mice lacking Glutamate Carboxypeptidase II are protected from peripheral neuropathy and ischemic brain injury (2005). J Neurochem., 95(2):314-23.
O'Keefe, DS and Heston, WDW Clearing up the confusion over the Glutamate Carboxypeptidase II gene (2004). American Journal of Medical Genetics A, 327:329-330.
Balaji KC, Rao PS, Smith DJ, Louis S, Smith LM, Sherman S, Bacich D, O'Keefe, DS. Microarray analysis of differential gene expression in androgen independent prostate cancer using a metastatic human prostate cancer cell line model (2004). Urol Oncol., 22(4):313-20.
O'Keefe, DS, Bacich, DJ and Heston, WDW Comparative analysis of Prostate-Specific Membrane Antigen (PSMA) versus a prostate-specific membrane antigen-like gene (2004). The Prostate, 58:200-210.
