Our published experiments have indicated that the DNA demethylating agent deoxazaycytidine (DAC) mediates dramatic induction of CT-X genes such as NY-ESO-1, MAGE-A1, and MAGE-A3 in cultured lung cancer cells, but not in normal human bronchial epithelia (NHBE). Furthermore, we have demonstrated that the HDAC inhibitor Depsipeptide (DP) enhances DAC-mediated induction of CT-X genes in lung cancer cells. In recent studies, pyrosequencing and chromatin immunoprecipitation (ChIP) techniques were used to comprehensively examine chromatin alterations associated with repression/activation of NY-ESO-1, MAGE-A1, and MAGE-A3 in a panel of lung cancer lines exhibiting various CT-X gene expression profiles. Repression of CT-X genes in A549, Calu-6 and H841 cells as well as normal human bronchial epithelial cells coincided with CpG hypermethylation, recruitment of a variety of DNA methyltransferases (DNMTs), histone methyltransferases (HMT) such as KMT6, histone demethylases (HDM) including LSD1, JARID1B and JARID1D, as well as deacetylation of histones H3/H4, and increased levels of H3K9me3 and H3K27me3 within the NY-ESO-1 MAGE-A1 and MAGE-A3 promoters. In contrast, activation of these CT-X genes in untreated H1299 cells, or H841 cells following sequential DAC/DP exposure coincided with marked DNA demethylation, reduced levels of DNMTs, HMTs, HDMs, hyperacetylated core histones, and increased levels of H3K4me2, H3K4me3, and H3K9Ac within the respective promoters. Repression of CT-X genes appeared related to persistence of normal chromatin structure within the respective promoters rather than aberrant silencing mechanisms. Additional studies revealed that knock-down of HMTs, or HDMs including EZH2, LSD1, JARID1B, and JARID1D, but not the Class III HDAC SIRT1 markedly enhanced DAC-mediated activation of NY-ESO-1, MAGE-A1 and MAGE-A3 in lung cancer cells. Subsequent experiments revealed that at doses one log lower than the IC-50, 3-deaza-neplanocin A (DZNep) depleted EZH2 and enhanced DAC-mediated activation of CT-X genes in lung cancer cells. Consistent with these observations, DZNep markedly enhanced recognition and lysis of these cells by allogeneic T cells expressing receptors for NY-ESO-1 and MAGE-A3. In addition, DZNep potentiated apoptosis mediated by DNA demethylating agents and HDAC inhibitors in lung cancer cells. These data, which constituted the first experiments to demonstrating that inhibition of histone methyltransferases may be a novel strategy to augment CT-X gene expression in lung cancer cells for subsequent eradication by adoptive immunotherapy regimens have been published recently in Cancer Research. A series of experiments have also been performed to identify novel epigenetic targets for therapy of malignant pleural mesotheliomas (MPM). Briefly, a panel of pleural mesothelioma lines and normal mesothelial cell cultures either initiated in our laboratory or obtained from commercial sources were processed for micro-array analysis. These experiments revealed aberrant expression of a variety of genes encoding DNA methyltransferases, histone acetyltransferases and histone deacetylases in MPM cells relative to normal mesothelia. Interestingly, this analysis revealed that cultured MPM cells exhibit over-expression of enhancer of zeste 2 (EZH2), suppressor of zeste 12 (SUZ12), and embryonic ectoderm development (EED), which encode key components of Polycomb repressive complex-2 (PRC-2) implicated in mediating stem cell pluripotency, as well as aberrant silencing of tumor suppressor genes during malignant transformation. Quantitative RT-PCR (qRT-PCR), gel-based RT-PCR, and western blot experiments confirmed significant up-regulation of EZH2 and EED in MPM lines;EZH2 and EED over-expression coincided with increased levels of H3K27Me3- a repressive chromatin mark mediated by EZH2. Additional qRT-PCR, RT-PCR, and immunohistochemistry (IHC) experiments utilizing 20 primary MPMs and tissue micro-arrays containing 28 MPMs and 17 peritoneal mesotheliomas revealed over-expression of EZH2 in 85% of mesotheliomas. No EZH2 expression was detected in normal pleura or peritoneum. Additional IHC and micro-array experiments revealed that EZH2 expression correlated with advanced stage of disease, and poor survival of patients with pleural mesothelioma. Knockdown of EZH2 or EED decreased global H3K27Me3 levels, and significantly diminished proliferation, migration, clonogenicity, and tumorigenicity of MPM cells. The effects of EED knockdown were more pronounced than EZH2 inhibition in these cells, possibly due to compensation of EZH2 depletion by EZH1, or the destabilizating effects of EED knock-down on PRC-2. The antiviral agent DZNep mediated dose-dependent depletion of EZH2, EED, and H3K27Me3, and significantly inhibited proliferation, migration, clonogenicity, and tumorigenicity of MPM cells. Micro-array experiments revealed that several tumor suppressors which are stem cell polycomb target were commonly activated in cultured MPM cells by knock-down of EZH2 or EED, or DZNep treatment. Additional micro-array experiments revealed that approximately 120 genes were differentially expressed in both of two mesothelioma lines following DZNep exposure. Less than 15% of these differentially expressed genes are stem cell polycomb targets. These results suggest that the mechanisms by which DZNep mediates cytotoxicity in MPM cells are not solely attributable to inhibition of PRC-2. Overall, these experiments are the first to demonstrate that aberrant expression of PRC-2 contributes to the malignant phenotype of MPM, and suggest that PRC-2 may be a novel target for mesothelioma therapy. A manuscript pertaining to these experiments will be published in Clinical Cancer Research within the next several months. Several reports indicate that over-expression of EZH2 correlates with advanced stage of disease and decreased survival of lung cancer patients. Currently, the mechanisms contributing to up-regulation of EZH2 in lung cancers, and the pathways by which EZH2 mediates progression of these malignancies have not been fully defined. Recently, a series of experiments have been performed to examine these issues. Western blot analysis revealed that EZH2 was highly over-expressed in all lung cancer cell lines relative to normal respiratory epithelia (SAEC or NHBE) or immortalized human bronchial epithelial cells (HBEC). Interestingly, EZH2 expression decreased in SAEC as these cells senesced. Subsequent experiments revealed that knock-down of EZH2 significantly diminished proliferation, clonogenicity, and tumorgenicity of lung cancer cells. Interestingly, knock-down of EZH2 was insufficient to induce apoptosis in lung cancer cells cultured in normal media, but dramatically increased apoptosis when cells were cultured in serum free media. Additional experiments demonstrated that EZH2 depletion increased sensitivity of lung cancer cells to HDACs inhibitors, but not other chemotherapeutic agents including HSP90 antagonists or EGFR inhibitors. Micro-array experiments revealed that nearly 400 genes were differentially expressed in EZH2-depleted lung cancer cells relative to vector controls. Approximately 28% of up-regulated genes are Polycomb target genes in stem cells. Subsequent studies revealed that DZNep depleted EZH2, and inhibited proliferation and clonogenicity of lung cancer cells. Micro-array analysis revealed >200 differentially expressed genes in DZNep-treated lung cancer cells, many of which were also seen in MPM cells following DZNep treatment. Collectively these data demonstrate that EZH2 is a novel epigenetic target for lung cancer therapy. These results, together with our mesothelioma data have provided impetus for dedicated funding from NCI leadership to continue development of DZNep for future clinical trials.
Showing the most recent 10 out of 13 publications
