The emerging relationships between epigenetics, cell cycle regulation, and cancer have prompted considerable research directed toward the evaluation of agents that modulate chromatin structure for cancer therapy. During the past year my research has continued to focus on:1. Evaluation of tumor antigen and tumor suppressor gene expression in lung and esophageal cancer and malignant pleural mesothelioma (MPM) cells following 5 aza-2'deoxycytidine (DAC), Depsipeptide FK228 (DP), and sequential DAC/DP exposure.2. Examination of gene expression profiles in cultured cancer cells and primary cancer specimens following exposure to chromatin remodeling agents.3. Investigation of the mechanisms by which inhibitors of DNA methyltransferase (DNMT) and HDAC activity mediate growth arrest and apoptosis in thoracic malignancies.Previously we have demonstrated induction of NY-ESO-1 cancer testis antigen (CTA) and p16 tumor suppressor gene expression in cultured lung cancer cells following exposure to the DNA demethylating agent DAC, the HDAC inhibitor DP, or sequential DAC/DP. Furthermore, we have shown that DP markedly enhances DAC-mediated apoptosis preferentially in cancer cells, and facilitates their recognition by cytolytic T lymphocytes specific for NY-ESO-1. Recently we have conducted several clinical trials designed to ascertain the feasibility of using these chromatin remodeling agents to alter gene expression in primary thoracic cancers. These include a phase I study of gene induction in thoracic malignancies mediated by 72h DAC infusion, a phase II study of gene induction in lung cancer mediated by 4h DP infusion, and a phase I study of gene induction in thoracic malignancies mediated by sequential DAC/DP infusion. Thirty-five patients received DAC infusions, 19 patients received DP infusions, and 21 patients have received sequential DAC/DP therapy. Whereas no objective clinical responses have been observed, approximately 30-40% of patients on these trials with documented progression of disease prior to protocol entry experienced disease stabilization; several individuals with stage IV lung cancers remained on study for more that a year. Target gene expression in pre- and post- treatment tumor tissues has been evaluated in a blinded manner by quantitative RT-PCR, MSP, and IHC techniques, as well as LCM and cDNA arrays. Steady-state plasma concentrations of DAC and DP approximated 25-50nM and 250-500ng/ml, respectively; these concentrations equaled or exceeded threshold levels for gene induction in cultured cancer cells. In general, RT-PCR and MSP analyses have been unreliable due to extensive stromal cell contamination, necrosis, and inflammation in biopsy specimens. IHC analysis revealed induction of NY-ESO-1 and p16 expression in 60% and 40% of DAC-treated patients; evidence of cumulative treatment effect was observed in one individual exhibiting prolonged stabilization of disease. Post-treatment antibodies were detected in 3 of 13 patients, 2 of whom exhibited NY-ESO-1 induction in their tumors. Enhanced histone acetylation and/or induction of p21 expression were observed in tumors from approximately 30-40% of patients receiving DP therapy. Recently LCM and cDNA array techniques have been used to extend the molecular analysis pertaining to patients receiving DAC, DP, or sequential DAC/DP therapy. Highly variable gene expression profiles have been observed in patient biopsies following drug treatment, reflecting the heterogeneity of the lung cancers as well as the underlying complexity of lung cancer epigenetics. Induction of several novel tumor suppressors has been observed in these biopsies. Whereas the overall concordance rate between tumor tissues and cell lines was only 25%, several regulatory pathways which previously had not been demonstrated to be modulated by chromatin remodeling mechanisms have been consistently altered in patient samples and cell lines. These pathways are the focus of ongoing investigative efforts in the lab. Collectively, these data indicate that DAC and DP can mediate target gene induction in thoracic malignancies. Further evaluation of gene expression profiles in cultured cancer cells and tumor tissues may enable rational design of additional trials evaluating the use of chromatin remodeling agents for augmenting immunogenicity and apoptosis of lung cancer cells. Our experience to date indicates that NY-ESO-1 is the CTA that is most robustly induced by DAC, DP, or sequential DAC/DP. Whereas alterations in chromatin structure facilitate expression of NY-ESO-1 in cancer cells, the precise mechanisms that regulate induction of this CTA during malignant transformation have not been fully defined. Recent studies from our lab indicate that NY-ESO-1 contains no bona fide CpG island within the promoter or proximal coding region, suggesting that expression of this CTA is mediated by other genes which are regulated via chromatin remodeling mechanisms. As such, we performed a series of experiments to examine if Brother of the Regulator of Imprinted Sites (BORIS) contributes to epigenetic regulation of NY-ESO-1 expression in lung cancer cells. Preliminary RT-PCR experiments revealed expression of BORIS in approximately 60% of lung cancer lines, but not cultured normal human bronchial epithelial (NHBE) cells, or normal human dermal fibroblasts. Quantitative RT-PCR experiments revealed profound, coincident induction of BORIS and NY-ESO-1 expression in a panel of cultured lung cancer cells (but not NHBE cells) following DAC, DP, or sequential DAC/DP exposure. IHC analysis confirmed BORIS protein expression in CALU-6 lung cancer cells but not NHBE cells following exposure to DAC, DP, or sequential DAC/DP. Subsequent tissue array experiments indicated that BORIS was expressed in 80-90% of NSCLCs, but not adjacent normal cells. Methylation-specific PCR (MSP), bisulfite sequencing, and chromatin immunoprecipitation (ChIP) experiments confirmed that induction of BORIS expression by DAC, DP, or sequential DAC/DP coincided with direct modulation of chromatin structure within the CpG island of the BORIS promoter. Northern blot analysis indicated that direct augmentation or inhibition of BORIS expression by gene transfer techniques resulted in corresponding changes in NY-ESO-1 expression in lung cancer cells. NY-ESO-1 promoter-reporter experiments confirmed that BORIS enhanced transcription of NY-ESO-1 in lung cancer cells. Electromobility shift assays (EMSA) indicated a direct physical interaction between BORIS protein and an NY-ESO-1 promoter sequence containing a putative CTCF recognition site. Experiments are in progress to comprehensively examine BORIS expression in biopsy specimens from patients on the gene induction trials, and to further ascertain the role of BORIS in lung cancer epigenetics.Additional studies have focused on the mechanisms by which DAC and DP mediate growth arrest and apoptosis in cancer cells. Recently, we conducted a series of experiments to examine if antisense-mediated inhibition of DNMT expression could mediate growth arrest in cultured lung and esophageal cancer, and malignant pleural mesothelioma (MPM) cells, as a prelude to evaluation of these agents in thoracic oncology patients. Cancer cells and normal human bronchial epithelial (NHBE) cells were transfected with phosphorothioate-modified antisense oligos (ASOs) specifically targeting DNMT1 or DNMT3b, or mismatch oligos using lipofectamine techniques. Under exposure conditions 200-fold lower than achieved in phase I clinical studies, ASOs mediated specific, dose-dependent depletion of DNMT1 and DNMT3b, resulting in pronounced (>80%) inhibition of proliferation of all cancer cell lines; these effects were not observed following ASO transfection of NHBE cells. Depletion of DNMT1 and /or DNMT3b coincided with dramatic, caspase- dependent, p53-independent apoptosis in 4 of the 6 cancer lines.
Showing the most recent 10 out of 26 publications
