Search

Background The sensitivity of myelocytomatosis oncogene (MYC) amplified medulloblastoma to class I histone deacetylase (HDAC) inhibition has been shown previously; however, understanding the underlying molecular mechanism is crucial for selection of effective HDAC inhibitors for clinical use. The aim of this study was to investigate the direct molecular interaction of MYC and class I HDAC2, and the impact of class I HDAC inhibition on MYC function. Methods Co-immunoprecipitation and mass spectrometry were used to determine the co-localization of MYC and HDAC2. Chromatin immunoprecipitation (ChIP) sequencing and gene expression profiling were used to analyze the co-localization of MYC and HDAC2 on DNA and the impact on transcriptional activity in primary tumors and a MYC amplified cell line treated with the class I HDAC inhibitor entinostat. The effect on MYC was investigated by quantitative real-time PCR, western blot, and immunofluorescence. Results HDAC2 is a cofactor of MYC in MYC amplified medulloblastoma. The MYC-HDAC2 complex is bound to genes defining the MYC-dependent transcriptional profile. Class I HDAC inhibition leads to stabilization and reduced DNA binding of MYC protein, inducing a downregulation of MYC activated genes (MAGs) and upregulation of MYC repressed genes (MRGs). MAGs and MRGs are characterized by opposing biological functions and by distinct enhancer-box distribution. Conclusions Our data elucidate the molecular interaction of MYC and HDAC2 and support a model in which inhibition of class I HDACs directly targets MYC’s transactivating and transrepressing functions.

Background Pilocytic astrocytoma (PA) is the most common pediatric brain tumor and a mitogen-activated protein kinase (MAPK)-driven disease. Oncogenic MAPK-signaling drives the majority of cells into oncogene-induced senescence (OIS). While OIS induces resistance to antiproliferative therapies, it represents a potential vulnerability exploitable by senolytic agents. Methods We established new patient-derived PA cell lines that preserve molecular features of the primary tumors and can be studied in OIS and proliferation depending on expression or repression of the SV40 large T antigen. We determined expression of anti-apoptotic BCL-2 members in these models and primary PA. Dependence of senescent PA cells on anti-apoptotic BCL-2 members was investigated using a comprehensive set of BH3 mimetics. Results Senescent PA cells upregulate BCL-XL upon senescence induction and show dependency on BCL-XL for survival. BH3 mimetics with high affinity for BCL-XL (BCL-XLi) reduce metabolic activity and induce mitochondrial apoptosis in senescent PA cells at nano-molar concentrations. In contrast, BH3 mimetics without BCL-XLi activity, conventional chemotherapy, and MEK inhibitors show no effect. Conclusions Our data demonstrate that BCL-XL is critical for survival of senescent PA tumor cells and provides proof-of-principle for the use of clinically available BCL-XL-dependent senolytics.

INTRODUCTION: Pilocytic astrocytoma (PA) is a mitogen-activated protein kinase (MAPK)-driven disease. Treatment of sub-totally resected PA remains challenging and relapses occur even after MAPK-targeted therapies such as MEK inhibition. Oncogenic activation of the MAPK-pathway drives the majority of cells into oncogene-induced senescence (OIS). OIS might represent a complementary vulnerability exploitable by senolytic agents. Here we investigated the senolytic properties of BH3-mimetics in PA. METHODS: Four patient-derived PA cell lines, DKFZ-BT66, -BT308, -BT317 (KIAA1549:BRAF-fusion) and DKFZ-BT314 (BRAF V600E-mutation) were treated with different BH3-mimetics, chemotherapeutics and MEK-inhibitors in proliferation (expression of SV40 large T (TAg)) and OIS (repression of TAg) states. Inhibition of metabolic activity (CellTiterGlo® 2.0) and reduction of viability (trypan blue) was assessed after 72 hours. Target expression was determined using gene expression data and Western blot. On-target activity was verified by immunoprecipitation. Dependence on Bcl-xL was investigated by shRNA-mediated knockdown and BH3-profiling. Gene expression data of primary PA and 751 cancer cell lines (GDSC dataset) was analyzed. RESULTS: BH3-mimetics inhibiting Bcl-xL (Bcl-xLi; Navitoclax, A-1131852, A-1155463, AZD4320) showed activity in 3/4 models (DKFZ-BT66, -BT314 and -BT317) in the OIS state (IC50s 20-fold below achievable plasma concentration indicating translatability. Genes from the gene set HALLMARK_XENOBIOTICS_METABOLISM could represent a predictive biomarker.

Pediatric low-grade gliomas, a diverse group of WHO grade 1 and 2 glial or glioneural tumors, comprise the most common category of primary brain tumors in children. The majority of these tumors are driven by alterations in the MAPK pathway, making them in principle susceptible to MAPKi therapy. While patients often benefit from MAPKi during treatment, tumor rebound may occur once treatment is stopped, constituting a significant clinical challenge. BT-40, patient-derived cells with molecular features of pleomorphic xanthoastrocytoma (BRAFV600E, CDKN2Adel), were used to model the rebound growth in vitro, based on viable cell counts in response to treatment and withdrawal of the clinically relevant BRAFV600E specific inhibitor dabrafenib. Standard-of-care chemotherapy (vincristine and carboplatin) was used as a reference. MAPK pathway reactivation upon withdrawal was assessed by WB and qPCR analysis. Based on the observed cell-regrowth and MAPK-reactivation pattern, key-timepoints during withdrawal were identified, which are currently being further analyzed through RNAseq and phospho-/proteomics. BT-40 cells started to proliferate again two days after dabrafenib withdrawal, and earliest five days after chemotherapy withdrawal. MAPK pathway activity, based on Mek and Erk phosphorylation, reached baseline levels three hours after dabrafenib withdrawal, this was associated with 2.5-fold increased c-Fos gene expression two hours after withdrawal. The earlier cell regrowth after dabrafenib withdrawal compared to chemotherapy withdrawal matches clinical observations, making the model suitable to study the rebound. The observed MAPK overactivation suggests the growth rebound might not only be caused by a fast reactivation of the pathway but also by other mechanisms, e.g. accumulation of upstream activators due to loss of negative feedback or parallel pathways. To investigate this, key-timepoints during treatment withdrawal will be analyzed using a multi-omics approach. Based on these findings, possible rebound-driving mechanisms will be identified and further validated using BT-40 and additional PXA models in vitro and in vivo.

Background and purpose Fibrosis is a frequent adverse effect of radiotherapy and no effective treatments are currently available to prevent or reverse fibrotic disease. We have previously identified altered epigenetic patterns at a gene enhancer of the diacylglycerol kinase alpha ( DGKA ) locus in normal skin fibroblasts derived from fibrosis patients. An open chromatin pattern related to radiation-inducibility of DGKA is associated with onset of radiation-induced fibrosis. Here, we explore epigenetic modulation of DGKA as a way to mitigate predisposition to fibrosis. Material and methods We studied the effect of the BET-bromodomain inhibitors (JQ1, PFI-1) on DGKA inducibility in primary fibroblasts. Hence, DGKA transcription was additionally induced by the radiomimetic drug bleomycin, and DGKA mRNA expression, histone H3K27 acetylation and downstream markers of profibrotic fibroblast activation after BET-bromodomain inhibition were determined. Results BET-bromodomain inhibition suppressed induction of DGKA in bleomycin-treated fibroblasts, reduced H3K27ac at the DGKA enhancer and repressed collagen marker gene expression. Alterations in fibroblast morphology and reduction of collagen deposition were observed. Conclusion For the DGKA enhancer, we show that BET-bromodomain inhibitors can alter the epigenetic landscape of fibroblasts, thus counteracting profibrotic transcriptional events. Interference with epigenetic patterns of fibrosis predisposition may provide novel preventive therapies that improve radiotherapy.

Background Medulloblastoma (MB) is one of the most common malignant pediatric CNS tumors. Patients with Group 3 MBs harboring MYC amplification exhibit low survival rates. Surviving patients suffer from therapy-induced sequelae, which calls for new targeted therapy strategies. We and others have previously shown the sensitivity of MYC-amplified MB to class I histone deacetylase (HDAC) inhibition. After demonstrating that the MYC target gene PLK1 is significantly downregulated upon class I HDACi treatment, we hypothesized that inhibition of both HDACs and PLK1 could have synergistic effects. Methods Cell metabolic activity changes upon HDAC and PLK1 inhibitor treatment were measured in MYC-amplified and non-amplified MB cell lines, as well as in an additional MYC-inducible cell line. The interaction effect of both inhibitors was determined by computation of the combination index (CI) using the Chou-Talalay method. Results were validated assessing cell viability, cell cycle, and apoptosis induction. Transcription profile changes after combination treatment were evaluated. Results MYC-amplified MB cell lines were more sensitive than non-amplified cell lines to PLK1i treatment, showing IC50 in clinically achievable concentration ranges. Inhibition of class I HDACs and PLK1 synergistically reduced cell metabolic activity in lower concentrations in MYC-amplified compared to non-amplified MB cell lines. We also observed a significant loss of viability and cells in G1 phase, as well as induction of apoptosis after combination treatment in MYC-amplified cells. MYC target gene sets were significantly downregulated in the MYC-amplified cell line HD-MB03 after treatment with combination. We demonstrated reduction of MYC protein levels upon PLK1i treatment. In vivo evaluation of combination treatment using orthotopic Group 3 MYC-amplified MB PDX models is ongoing. Conclusion Our data suggest that MYC-amplification is a predictive marker for PLK1i treatment in MB. The combination of HDACi and PLKi could be a candidate therapy for future clinical trials for MYC-amplified group 3 MB.

Patients with MYC-driven Group 3 medulloblastoma (MB) show particularly poor outcome. It was previously shown that MYC-driven MBs are highly sensitive to class I histone deacetylase inhibition (HDACi). We studied the molecular effects of the class I HDACi entinostat in MYC-driven MB cells to identify potentially synergistic drug combinations, prioritizing drug clinical availability to enable clinical translation. Gene expression profiles of the MYC-amplified group 3 MB cell line HD-MB03 treated with entinostat were analyzed using bioinformatic approaches, identifying 29 altered biomechanisms. Overlay with a translational drug library of n=76 compounds resulted in 44 compounds targeting 9 biomechanisms. Filtering for publications supporting each drug′s role in MYC-driven entities, or functional interaction with HDACs, without publication of this combination in MBs, resulted in 5 compounds (olaparib, idasanutlin, ribociclib, selinexor, vinblastine). Synergism testing identified olaparib as the drug with the strongest synergism. Validation of the combination olaparib and entinostat by p.H2AX and PI staining as well as trypan blue exclusion showed increased double strand breaks (DSBs), increased cell death, loss of viability and cell numbers. Selectivity of MYC-amplified MB cells was shown by comparison to MYC-non amplified cell lines, which showed higher IC50s, and reacted with cell cycle arrest as opposed to cell death to the combination treatment. The role of HDACis in DNA damage repair was confirmed by increased DSBs when entinostat was added to the combination of olaparib with doxorubicin. Our study identified olaparib as a potential combination partner with entinostat for the treatment of MYC-driven Group 3 MB.

Medulloblastoma (MB) is one of the common malignant brain tumors in children. Patients with MYC-amplified Group 3 MBs exhibit particularly poor survival rates even after intensive therapy. Surviving patients often suffer from long-term side effects. This calls for new therapeutic strategies, such as targeted therapy options. The sensitivity of MYC-amplified MBs to class I histone deacetylase (HDAC) inhibition was previously shown. In order to elucidate potential combination partners, we have identified PLK1 as one of the top regulated genes following treatment of MYC-amplified Group 3 MB cells with class I HDACi entinostat. Our aim here is to investigate possible combination treatments with entinostat and several PLK1i (volasertib, GSK461364, onvansertib). The cell metabolic activity was evaluated using MYC-amplified and nonamplified MB, high-grade glioma (HGG) and neuroblastoma (NB) cell lines after single and combination treatments. The interaction effect was determined by combination index (CI) (Chou-Talalay CI calculation method). Results were validated assessing cell viability, cell cycle profile, and caspase activity upon treatment with single agents or combination. On-target activity was examined using immunoblotting for pTCTP and H3K27ac. We have confirmed our findings in an inducible MYC cell line. The gene expression profile was analyzed in HD-MB03 cell line after entinostat, volasertib, or combination treatment. We demonstrate that the MYC target gene PLK1 is significantly downregulated upon HDACi treatment. Based on this, we hypothesized that inhibition of both class I HDACs and PLK1 could have synergistic effects. MYC-amplified cell lines were more sensitive than nonamplified cell lines to treatment with all PLK1i investigated, showing 3 to 10 times lower IC50. PLK1i and entinostat interacted synergistically (CI below 0.9) at lower concentrations in MYC-amplified compared to non-amplified MB cell lines. Similar results were obtained for MYC or N-MYC-amplified HGG and NB cell lines. We also observed the loss of viability and loss of fraction of cells in G1 phase in MYC-amplified MB cells after treatment with entinostat and PLK1i. MYC target genes were significantly downregulated in the MYC-amplified MB cell line HD-MB03 after treatment with PLK1i and entinostat. Moreover, we demonstrated reduction of MYC levels and faster MYC degradation upon volasertib treatment. Our data suggest that MYC-amplification might serve as a predictive biomarker for PLK1i treatment in MB and other entities. The combination of entinostat and PLKi could be a candidate therapy for future clinical trials for MYC-amplified Group 3 MB, and possibly other tumors harboring MYC amplification. Citation Format: Gintvile Valinciute, Jonas Ecker, Thomas Hielscher, Christin Schmidt, Marc Remke, Gianluca Sigismondo, Jeroen Krijgsveld, Stefan M. Pfister, Olaf Witt, Till Milde. Synergistic interaction of HDACi and PLK1i in Group 3 MYC-amplified medulloblastoma [abstract]. In: Proceedings of the AACR Special Conference on the Advances in Pediatric Cancer Research; 2019 Sep 17-20; Montreal, QC, Canada. Philadelphia (PA): AACR; Cancer Res 2020;80(14 Suppl):Abstract nr A53.

High-risk medulloblastoma (MB) is a deadly disease with poor overall survival. Survivors suffer from severe treatment-associated morbidity. Patients with Group 3 MB with an amplification of MYC show particularly poor outcome. Therefore novel therapies tailored to this subgroup are urgently needed. We have previously shown that MYC amplified MB cell lines are highly susceptible to inhibition of class I histone deacetylases (HDACs), including HDAC2. We here explore the functional interaction of HDAC2 and MYC. METHODS: Chromatin immunoprecipitation with pulldown of MYC, HDAC2, RNApolII and H3K27ac, followed by DNA-sequencing was performed in n=3 primary MYC amplified Group 3 MB tumors and correlated with gene expression profiles (GEP). The GEP of MYC amplified MB cell line HD-MB03 was determined after class I HDAC inhibition. MYC stability upon HDACi treatment was assessed in immunofluorescence microscopy and cycloheximide chase experiments. MYC-HDAC2 interaction was analyzed in co-immunoprecipitation and mass spectrometry studies. RESULTS: In primary tumors 90% of MYC DNA binding sites are co-occupied by HDAC2. Compared to MYC- or HDAC2-only bound genes, MYC-HDAC2 co-bound genes show strong recruitment of RNApolII and higher expression levels. In HD-MB03 cells MYC and HDAC2 show overlapping DNA binding sites of 80%, both proteins are co-localized in a multi-protein complex. GEP analysis revealed a significant downregulation of MYC target genes after treatment with class I HDAC inhibitor MS-275, coinciding aberrant MYC protein levels. Our data suggests that MYC functions directly depend on class I HDAC activity. Inhibiting HDAC2 may offer a therapeutic option to target the currently undruggable oncoprotein MYC.

Medulloblastoma (MB) is the most common malignant brain tumor in children. Patients with MYC-amplified Group 3 MBs exhibit particularly poor survival rates even after intensive therapy, and surviving patients often suffer from long-term sequelae. Poor survival rates and sequelae call for new therapeutic strategies, such as targeted therapy options. We have previously shown that MYC-amplified MBs are sensitive to class I histone deacetylase (HDAC) inhibition. Here we investigate possible combinatorial treatment with HDACi and PLK1i. We evaluated the metabolic activity of MYC-amplified and non-amplified MB, high grade glioma (HGG) and neuroblastoma (NB) cell lines after single and combinatorial treatments with HDACi and PLK1i. The interaction effect was determined by combination index (CI) calculation using the Chou-Talalay method. On-target activity was examined by immunoblotting for pTCTP and H3K27ac. The findings were validated in an inducible MYC cell line. The gene expression profile in the MYC-amplified HD-MB03 cell line was analyzed after MS-275, volasertib or combination treatment. We show that the MYC target gene PLK1 is significantly downregulated upon HDACi treatment. We hypothesized that inhibition of both, HDACs and PLK1 could have synergistic effects. MYC-amplified cell lines were more sensitive than non-amplified cell lines to PLK1i showing 3–10 times lower IC50. PLK1i and HDACi interacted synergistically (CI below 0.9) in lower concentrations in MYC-amplified compared with non-amplified cell lines. Similar results were obtained for MYC or N-MYC-amplified HGG and NB cell lines. MYC target genes were significantly downregulated in the MYC-amplified cell line HD-MB03 after treatment with PLK1i and HDACi. Our data suggests that MYC-amplification is a predictive marker for PLK1i treatment in MB and other entities. The combination of HDACi and PLKi could be a candidate therapy for MYC-amplified group 3 MB, and possibly other tumors harboring MYC amplification, in future clinical trials. Studies investigating the mechanism of action are ongoing.

High-risk medulloblastoma (MB) remains a deadly disease with poor overall survival despite aggressive multi-modal treatment. Survivors suffer from important treatment-associated morbidity. Patients with Group 3 MB tumors with amplification or high expression of the transcription factor MYC have a particularly poor outcome. Previously we have shown that MYC amplified Group 3 MB cell lines are highly susceptible to pan- and class I-selective histone deacetylase (HDAC) inhibitor treatment. We here explore the potentially sensitizing effect of MYC to HDAC inhibition by unraveling the molecular interaction of MYC and HDAC2. In three established MB cell lines, MYC amplified as well as single copy, mRNA and protein expression of MYC and HDAC2 after enzymatic inhibition of class I HDACs and after siRNA-mediated knockdown of HDAC2 was investigated. The direct interaction of HDAC2 and MYC protein was analyzed in co-immunoprecipitation and mass spectrometry studies. Chromatin immunoprecipitation with pulldown of MYC or HDAC2 and subsequent DNA-sequencing (ChIP-Seq) was performed in a MYC amplified cell line. Reduced MYC protein levels were observed after both pharmacological inhibition of class I HDACs (including HDAC2) as well as knockdown of HDAC2, without significant alteration of MYC mRNA levels. Reduced MYC protein levels coincided with increased levels of acetylated MYC protein. MYC and HDAC2 were found to be localized in a protein complex together with other potential drug targets. MYC and HDAC2 co-localize on the DNA with an overlap of DNA binding sites of 80%. Our data suggest that inhibition of class I HDACs destabilizes MYC protein, thus reducing the abundance of one of the cell’s main oncogenic drivers. The direct interaction of MYC/HDAC2 in a multi-protein complex localized on the DNA represents a putative epigenetic target for novel combinatorial treatment approaches. Perturbation of gene expression profiles upon HDACi treatment of MYC amplified MB cell lines are currently under investigation.