Abstract 144: A microarray-based study of DNA methylation changes in prostate cancer

Abstract

Abstract Background: Prostate cancer is the most common cancer in men in the United States. Prior work, based on analyses of selected genes, has shown that epigenetic changes are important in disease progression and pathogenesis. We hypothesized that a global comparative analysis of DNA methylation changes in neoplastic prostate cells would provide an opportunity to discover novel genes involved in prostate carcinogenesis. We designed such a study using DNA from paraffin embedded normal and cancer specimens and from commonly used prostate cancer cell lines. Methods: DNA from LNCaP, VCaP, DU145, DUPRO, PC3, LAPC4, and 22RV cells, paraffin-embedded clinical samples including 78 cancer samples and 3 normal prostate samples from radical prostatectomies, positive control in vitro methylated DNA, and negative control DKO cells, in which the DNMT1 and DNMT3b enzymes were genetically disrupted, were modified with sodium bisulfite. Along with these samples, non bisulfite-modified genomic DNA, which served as an additional negative control, was also hybridized to Illumina Golden Gate BeadArrays after labeling and amplification, which allowed for the assessment of 1505 CpG sites in 807 genes. A Beta methylation value from the BeadStudio Software is assigned to each CpG site. Probes with Beta>0.25 were considered methylated (Beta>0.5 represented a strong methylation level). In order to enrich for cancer-specific methylation, we excluded probes with methylation in normal prostate samples, genomic DNA samples, or DKO negative control samples. Probes unmethylated in the positive control in vitro methylated DNA samples were also excluded. Results: 101 genes were methylated in at least one prostate cancer cell line and clinical sample and in the positive control samples but not in the negative control samples or normal prostate samples. Pathways whose members were silenced by DNA methylation included: growth factor signaling, cell-cell adhesion, differentiation, and apoptosis. The methylation status of single genes or combinations of 2 or more methylated genes resulted in 78%-90% sensitivity and 100% specificity for prostate cancer tissues versus normal tissues. Conclusions: Our results demonstrate that high density 96-well BeadArrays are an efficient and cost-effective method to ascertain the DNA methylation status of many genes simultaneously in cell lines and in paraffin-embedded clinical samples. Of equal importance, our study demonstrates the importance of including positive and negative controls both for DNA methylation and for bisulfite conversion in such studies to more specifically identify cancer-specific DNA methylation changes that will most likely result in cancer- specific gene silencing. Our results have implications not only for improving prostate cancer detection but also for understanding the genes, whose silencing by DNA methylation, may be critical for prostate cancer development and progression. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 101st Annual Meeting of the American Association for Cancer Research; 2010 Apr 17-21; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2010;70(8 Suppl):Abstract nr 144.