Patients with glioma whose tumors carry a mutation in isocitrate dehydrogenase 1 (IDH1R132H) are younger at diagnosis and live longer. IDH1 mutations co-occur with other molecular lesions, such as 1p/19q codeletion, inactivating mutations in the tumor suppressor protein 53 (TP53) gene, and loss-of-function mutations in alpha thalassemia/mental retardation syndrome X-linked gene (ATRX). All adult low-grade gliomas (LGGs) harboring ATRX loss also express the IDH1R132H mutation. The current molecular classification of LGGs is based, partly, on the distribution of these mutations. We developed a genetically engineered mouse model harboring IDH1R132H, TP53 and ATRX inactivating mutations, and activated NRAS G12V. Previously, we established that ATRX deficiency, in the context of wild-type IDH1, induces genomic instability, impairs nonhomologous end-joining DNA repair, and increases sensitivity to DNA-damaging therapies. In this study, using our mouse model and primary patient-derived glioma cultures with IDH1 mutations, we investigated the function of IDH1R132H in the context of TP53 and ATRX loss. We discovered that IDH1R132H expression in the genetic context of ATRX and TP53 gene inactivation (i) increases median survival in the absence of treatment, (ii) enhances DNA damage response (DDR) via epigenetic up-regulation of the ataxia-telangiectasia–mutated (ATM) signaling pathway, and (iii) elicits tumor radioresistance. Accordingly, pharmacological inhibition of ATM or checkpoint kinases 1 and 2, essential kinases in the DDR, restored the tumors’ radiosensitivity. Translation of these findings to patients with IDH1132H glioma harboring TP53 and ATRX loss could improve the therapeutic efficacy of radiotherapy and, consequently, patient survival.
Survival of glioma (GBM) patients treated with the current standard of care remains dismal. Immunotherapeutic approaches that harness the cytotoxic and memory potential of the host immune system have shown great benefit in other cancers. GBMs have developed multiple strategies, including the accumulation of myeloid-derived suppressor cells (MDSCs) to induce immunosuppression. It is therefore imperative to develop multipronged approaches when aiming to generate a robust anti-tumor immune response. Herein, we tested whether combining MDSC depletion or checkpoint blockade would augment the efficacy of immune-stimulatory herpes simplex type-I thymidine kinase (TK) plus Fms-like tyrosine kinase ligand (Flt3L)-mediated immune stimulatory gene therapy. Our results show that MDSCs constitute >40% of the tumor-infiltrating immune cells. These cells express IL-4Rα, inducible nitric oxide synthase (iNOS), arginase, programmed death ligand 1 (PDL1), and CD80, molecules that are critically involved in antigen-specific T cell suppression. Depletion of MDSCs strongly enhanced the TK/Flt3L gene therapy-induced tumor-specific CD8 T cell response, which lead to increased median survival and percentage of long-term survivors. Also, combining PDL1 or CTLA-4 immune checkpoint blockade greatly improved the efficacy of TK/Flt3L gene therapy. Our results, therefore, indicate that blocking MDSC-mediated immunosuppression holds great promise for increasing the efficacy of gene therapy-mediated immunotherapies for GBM.
Fyn, a Src family kinase member (SFK), has oncogenic activities. However, the function of Fyn in malignant brain tumors (GBM) remains contradictory. Bioinformatics network analysis of the transcriptome of GBM neurospheres suggested that Fyn is a highly connected node. Thus, Fyn was selected as a potential regulator of glioma behavior. We investigated the role of Fyn using in vitro and in vivo models. Fyn expression correlates positively with GBM cell aggressiveness. The analysis of the effects of Fyn knockdown in NP mouse glioma neurospheres showed that cell proliferation and migration was decreased compared to NP-control cells. Using an intracranial glioma model, we found that Fyn-knockdown glioma neurospheres displayed reduced tumor growth and prolonged survival compared with NP tumors. We also developed a stringent Fyn-deficient genetic model using the Sleeping Beauty transposase system. The tumors were induced by injecting plasmids encoding: N-Ras/shp53 (NP) or N-Ras/shp53/shFyn (NPF) into the neonatal periventricular region. Fyn knockdown (NPF) increased animal survival. The histopathological evaluation of gliomas indicates that loss of Fyn reduced malignant features such as pseudopalisades, necrosis, hyper vascularization and oncostreams. Nevertheless, the survival analysis of TCGA treated patient data indicates that GBM with lower levels of Fyn had reduced survival. Therefore, we evaluated the effects of therapies on Fyn knockdown cells. The analysis surprisingly showed that NPF neurospheres were more resistant to treatment than NP. Fyn loss decreases glioma aggressiveness but induces less response to therapies. These results indicate an important role for Fyn modulating many glioma cellular processes and its relevance as a novel regulator of GBM behavior and therapy response.
Diffuse intrinsic pontine glioma (DIPG) is a very aggressive pediatric brainstem tumor for which there is no treatment. The most common mutation occurs on the amino terminal tail of histone H3 variants H3.1 and H3.3 changing a lysine to methionine at position 27 and this results in global hypomethylation of H3K27. Additionally, 24% of tumors harbor activating mutations in Activin-A receptor Type 1 (ACVR1) a bone morphogenetic (BMP) receptor of the transforming growth factor beta (TGF-β) family. We used the Sleeping Beauty Transposase (SB) system to deliver plasmids encoding oncogenic NRAS and a short hair pin for p53 (shP53) in combination with ACVR1 G328V or H3.1 K27M, or both, into the lateral ventricles of neonatal mice, to generate endogenous tumors with DIPG mutations. Tumor development in mice was monitored by measuring luciferase activity in vivo; animals that reached moribund stage were perfused and processed for histology. Our preliminary results indicate that tumors with ACVR1 G328V have an increased median survival (MS>133 dpi) as compared to our control groups: NRAS/shP53 group (MS=90 dpi) and NRAS/shp53/H3.1 K27M group (MS=92 dpi). Tumors harboring NRAS/shp53/ and ACVR1 G328V genetic alterations also do not express the oligodendrocyte transcription factor 2 (Olig2) by immunohistochemistry and western blot. In conclusion, we used the SB system to develop a mouse model of DIPG. We observed an increase in median survival and some phenotypic differences in tumors harboring the ACVR1 G328V mutation. The loss of Olig2 NRAS/shp53/ACVR1 G328V tumors is interesting because Olig2 has been associated with delayed tumor progression. In the future we aim to elucidate the mechanisms by which ACVR1 and H3.1 K27M mutations contribute to tumor initiation and progression in order to develop targeted therapies for DIPG. Work supported by NIH/NINDs, Leah’s Happy Hearts and Chad Tough Foundations.
Background: High-grade malignant peripheral nerve sheath tumors (MPNST) frequently lose Polycomb Repressive Complex 2 (PRC2), reducing the repressive histone mark, H3K27me3. We hypothesized that H3K27me3 loss sensitizes MPNST to oncolytic HSV due to impaired transcriptional resistance. HSV replication is also restricted by the integrated stress response, normally countered by HSV ICP34.5, representing an additional resistance pathway to ICP34.5 -attenuated HSV. Methods: PRC2 loss-of-function was evaluated as a biomarker of HSV G47Δ ( ΔICP34.5 ) susceptibility in MPNST using isogenic, pharmacologic inhibition, and genetic rescue models. CUT&RUN and RNA-sequencing defined PRC2-dependent viral and host responses. We engineered a G47Δ-derived vector (G47Δ 2 ) with tumor-regulated ICP34.5 expression [G47Δ 2 ,ins( BIRC5>ICP34.5 )] and evaluated efficacy in MPNST xenografts. Results: H3K27me3 deficiency increased G47Δ sensitivity and viral spread in vitro . CUT&RUN demonstrated PRC2-dependent H3K27me3 accumulation on the viral genome, explaining G47Δ activity in H3K27me3-deficient MPNST. RNA sequencing of G47Δ infected MPNST cells revealed heightened integrated stress response activity and blunted induction of translation-associated programs in the PRC2-deficient state (both p <0.001), consistent with compensatory resistance at the level of viral protein translation. Accordingly, tumor-regulated ICP34.5 HSV limited daily tumor-volume growth to 0.6% (95% CI, −2.9–4.2%), improving upon both G47Δ 2 (8.3%; 4.5–12.2%) and controls (22.5%; 17.4–27.8%). A murine gastrointestinal toxicity phenotype reduced all-cause event-free survival, indicating vector and dosing optimizations are needed to widen the therapeutic window. Conclusions: H3K27me3 loss creates a paradoxical antiviral state that tumor-selective ICP34.5 overcomes to rescue HSV anti-tumor activity, supporting biomarker-driven viral engineering to exploit the permissive transcriptional state of H3K27me3-deficient MPNSTs. Key Points: Loss of a key epigenetic mark (H3K27me3) MPNST is a biomarker of susceptibility to oncolytic HSV H3K27me3-deficient MPNST mounts a compensatory stress response that limits HSV propagation. Engineered viral gene therapy to overcome the stress response rescues antitumor activity. Importance of the Study: High-grade MPNST lacks effective systemic therapies, and biomarkers that can guide oncolytic virotherapy are poorly defined. This study identifies loss of PRC2-dependent H3K27me3 as a marker of susceptibility to oncolytic HSV but shows that this same epigenetic state also primes a compensatory integrated stress response that limits productive viral propagation at the level of protein translation. These findings connect a tumor-specific chromatin defect to both enhanced viral permissiveness and a compensatory resistance mechanism. By restoring tumor-selective expression of ICP34.5 , an HSV-encoded antagonist of the integrated stress response frequently removed to promote oncolytic vector attenuation, we demonstrate that this second barrier can be therapeutically targeted to improve antitumor activity. The work therefore provides a mechanistic framework for matching viral engineering to tumor epigenetic state. Robust anti-tumor activity must be balanced with therapeutic safety, providing avenues for further vector optimization. This biomarker-driven strategy may be relevant to other H3K27me3-deficient nervous system tumors.
A distinctive glioma sub-type harbors mutations in Isocitrate Dehydrogenase 1 (IDH1-R132H: mIDH1), ATRX and p53. IDH1-R132H generates 2-hydroxglutarate (2HG), which gives rise to histone and DNA hypermethylation. Mutant IDH1 has been associated with better patients’ prognosis. We hypothesize that this is mediated by the impact of epigenetic changes on DNA-repair and DNA damage response (DDR) pathways. To test this, we developed a genetically engineered mouse glioma model harboring mIDH1, TP53 knockdown and ATRX knockdown using the Sleeping Beauty Transposase system. We also generated mouse tumor neurospheres (NS) and human glioma cells expressing mIDH1, with or without ATRX loss. Our results show that mIDH1 expression enhances median survival (MS > 160 DPI) when compared with wild type (wt) IDH1 (MS = 72 DPI). Mutant IDH1 tumors exhibit inhibition of cell differentiation; and enhanced histone 3 methylation. Analysis of ChIP-seq, RNA-seq and Bru-seq data indicates that genes involved in DNA-repair and DDR, including homologous recombination DNA-repair, were upregulated in the mIDH1 mouse model. Functional assessment of DNA repair activity was performed after ionizing radiation (IR), i.e., H2AX phosphorylation, GFP-reporter DNA repair assay, and comet assay. Our results demonstrate that mIDH1 mouse NS and human glioma cells exhibited increased DNA-repair and DDR activity and enhanced genomic stability. This phenotype is correlated with the response to IR where cells and mice harboring mIDH1 tumors were more resistant compared with wt-IDH1 tumors. The combination of IR with inhibitors of DDR reverts radio-resistance in mIDH1 tumor bearing mice. In conclusion, genetically engineered mice harboring glioma encoding mIDH1, ATRX and TP53 knockdown displayed increased MS in absence of treatment. This could be attributed to increased DNA-repair, DDR and genomic stability, which in turn hampers radiation efficacy. Our findings thus reveal a potential therapeutic approach targeting DDR for enhancing therapeutic efficacy of radiation in mIDH1 glioma patients