G34R Mutation Research Articles

MPC-5 Characteristics of H3 G34-mutant gliomas

Introduction: Diffuse hemispheric gliomas, H3 G34-mutant (DHG H3G34-mutant) are newly recognized infiltrating gliomas of the cerebral hemispheres of pediatric and young adult patients. We experienced 6 DHG H3G34-mutant cases. In this study, we describe the clinical, radiological and pathological characteristics of these cases. Result: Mean age at diagnosis was 16.8 years (range:10–26). Three patients were male. Among six cases, tumors located in cerebral cortex in five cases and multiple sites including basal ganglia and cortex in a case. All tumors showed no or only a faint contrast-enhancement and harbored restriction of diffusion. One patient underwent total resection, four underwent partial resection and one underwent biopsy. Pathological diagnosis were CNS embryonal tumors (n=3/6), glioblastoma, IDH-wildtype (n=2/6) and anaplastic astrocytoma, IDH-wildtype (n=1/5). All cases were negative for Olig2 and positive for GFAP in immunohistochemistry. Mean Ki-67 index was 38% (range: 10–60%). All cases revealed at least one of mitosis, necrosis or microvascular proliferation. Especially, mitosis was the most frequently found (n=5/6). The H3F3A mutations were G34R mutations in all cases. One case revealed a characteristic mutation pattern, therefore now we are performing further examination. Adjuvant chemoradiotherapies were performed for all cases. Mean progression free survival was 10.1 months (range: 1.6–33.1). Discussion: As published literatures reported, all cases exhibited restriction of diffusion and negative for Olig2. For a cerebral hemispheric tumor of pediatric or young adult patient which shows restriction of diffusion and no contrast-enhancement, and of which pathological findings is malignant and olig2 is negative, genetic analysis of H3F3A gene might be essential.

EPCO-07. HYBRID NEURO-GLIAL CELLULAR ARCHITECTURE IN HIGH-GRADE GLIOMA DRIVEN BY H3-G34R MUTATION

Abstract High-grade gliomas (HGG) with histone H3.3 G34R mutation are rare intractable tumours in the cerebral hemispheres that preferentially affect adolescents and young adults, but have unknown mechanisms of neuroanatomical specificity and tumourigenesis. Here, we performed single-nucleus RNA-sequencing of twenty patient samples, encompassing twelve tumours with G34R mutation and eight H3.3 wildtype HGGs, age- and location-matched. Both classes of HGG were heterogeneous, with malignant cells in multiple states, recapitulating neural and glial developmental trajectories. G34R HGG is distinguished by lack of malignant cells in the oligodendroglial lineage, and aberrant expression of neuronal programs superimposed over cellular states, resulting in hybrid glio-neuronal malignant programs. Singe-cell barcoding supports plasticity between cellular states in HGG with multiple possible transitions. CRISPR-correction of G34R in HGG models followed by scRNA-seq supports that the G34R mutation directly drives these aberrant programs. Our study provides a framework for studying the origin and tumourigenesis of paediatric gliomas.

EPCO-16. ONCOHISTONE INTERACTOME PROFILING UNCOVERS MECHANISMS OF CHROMATIN DISRUPTION AND IDENTIFIES POTENTIAL THERAPEUTIC TARGETS IN PEDIATRIC HIGH-GRADE GLIOMA

Abstract Mutations in histone H3 at amino acids 27 (H3K27M) and 34 (H3G34R) occur with high-frequency in pediatric high-grade glioma. H3K27M mutations have been shown to lead to global disruption of H3K27me3 through dominant negative PRC2 inhibition with accompanying gains in H3K36me3, while H3G34R mutations lead to local losses of H3K36me3 through inhibition of SETD2. However, the mechanism of action of these mutants on the broader landscape of chromatin-associated proteins remains unknown. Importantly, proteins with differential associations with oncohistones could be targeted therapeutically. Here we profiled the interactomes of the H3.1K27M, H3.3K27M and H3.3G34R oncohistones using BioID to gain an unbiased measure of their interaction landscapes. Among the differential interactors all 3 mutants lost interaction with H3K9 methyltransferases, while H3G34R also had reduced interaction with DNA methyltransferases accompanied by genome-wide DNA hypomethylation. In contrast, H3K27M mutants had increased association with transcription factors, consistent with the activation of transcription induced by the global loss of H3K27me3. H3K9me3 was reduced in H3K27M-containing nucleosomes, and cis-H3K9 methylation was required for H3K27M to exert its effect on global H3K27me3. Depletion of H3K9 methyltransferases with shRNA or treatment with H3K9 methyltransferase inhibitors was lethal to H3.1K27M, H3.3K27M and H3.3G34R mutant pHGG cell lines, underscoring the importance of H3K9 methylation for oncohistone-mutant gliomas and suggesting it could make an attractive therapeutic target.

Identification of novel therapeutic targets for histone 3 mutated children’s brain tumour, using unique tumour cell surface proteomic signatures

Abstract Aims Improvements in the treatments for childhood and adolescent brain tumours, High-Grade Glioma (pHGG) and Diffuse Intrinsic Pontine Glioblastoma (DIPG), have not advanced much and they continue to carry a very poor prognosis. These brain tumours are now defined by mutations affecting histone 3 proteins, indeed 80% of DIPGs harbour histone H3.1 and H3.3 K27M somatic mutations whilst 30% of pHGGs exhibit H3.3 G34R or G34V mutations. We hypothesized that the histone 3 mutant tumours will have distinct mutation specific surfactome (cell membrane proteins) signature. Method We therefore analysed the cell surface proteomics of pHGG and DIPG, in order to identify novel targets for therapy. We have at first isolated the cell membrane fractions from a range of patient cells carrying different histone 3 mutations (G34R, G34V), relative to wild type histone 3. A comparative quantitative mass-spectrometry analyses of these cell surface membrane fractions is then performed. Results The results obtained to date demonstrated unique differential cell membrane expression patterns which correlated to specific mutation type. For example, increased expression of Ras-related proteins Rab-3, Rab-3D is detected only in histone H3.3-G34R mutated cell line in comparison. Conclusion Identification and analyses of these unique cell membrane proteins’ association with specific in H3.3 mutation in pHGG, will help to identify precise mutation specific therapeutic targets, benefiting the patients to receive therapy based on tumour’s molecular signature.

Structural Basis of RACK7 PHD Domain to Read a Pediatric Glioblastoma‐Associated Histone Mutation H3.3G34R

Main observation and conclusionHistone point mutations, including missense mutations on histone H3 at positions 27 (K27M), 34 (G34R/V, G34W, G34L) and 36 (K36M), were identified as potential cancer driver mutations. H3.3G34R/V mutations account for pediatric glioblastomas (GBM). RACK7 (also known as ZMYND8, PRKCBP1) was recently reported to specifically bind H3.3G34R through its PHD (plant homedomain) domain (PHDRACK7) in vitro and in H3.3G34R pediatric glioblastoma cells, playing key roles in H3.3G34R‐mediated gene transcription. Herein, we provided both biochemical and NMR structural evidences that PHDRACK7 recognized histone H3.3G34R mutant via a mechanism distinct from all other reported PHD domains. Except the reported residue D104, two new sites D108 and L121 of PHDRACK7 were found necessary for the interactions between PHDRACK7 and histone H3.3G34R peptide. Our results provided a potential molecular basis for pediatric GBM driven by the H3.3G34R mutation.

HGG-31. HISTONE H3.3 G34R/V MUTATIONS STIMULATE PEDIATRIC HIGH-GRADE GLIOMA FORMATION THROUGH THE INDUCTION OF CHROMOSOMAL INSTABILITY

The histone H3.3 G34R/V mutations are known drivers of high-grade pediatric glioma (pHGG). However, the mechanism(s) for H3.3 G34R/V induced tumor formation are unclear. Chk1 phosphorylates H3.3 S31 at the pericentromere during early mitosis, suggesting a novel mitotic function. We observed that H3.3 G34 mutant pHGG cells have reduced mitotic H3.3 S31 phosphorylation compare to WT H3.3 cell lines. The H3.3 G34R mutation reduced Chk1 phosphorylation at S31 by >90% in an in vitro kinase assay. Overexpression of either H3.3 G34R or non-phosphorylatable S31A in H3.3 WT, diploid cells led to a significant increase in chromosome mis-segregations. Likewise, H3.3 G34 mutant pHGG cells have significantly elevated rates of mis-segregation as compare to H3.3 WT pHGG cells. During normal cell division, phospho-S31 is lost in late anaphase. However, when chromosome missegregation occurs, phospho-S31 spreads and stimulates p53 accumulation in G1 – thus suppressing aneuploid cell proliferation. Here we show that cells expressing mutant G34 fail to arrest following mis-segregation, despite having WT p53. These studies demonstrate that the H3.3 G34R/V mutations are sufficient to transform normal, diploid cells into proliferative, chromosomally instable cells. To determine if this process contributes to tumorigenesis, we used the RCAS/TVA mouse model to overexpress H3.3 WT, G34R, or S31A in the glial precursor cells of mice pups. Over 100 days, S31A and G34R mice had drastically reduced survival (averaging 77, 81, and 100 days for S31A, G34R, and WT mice). Furthermore, most G34R and S31A mice developed HGG, while H3.3 WT mice remained tumor-free and did not develop high-grade tumors. Our work strongly indicates that a major factor in H3.3 G34R pHGG formation is the induction of chromosomal instability – which occurs directly through the suppression of H3.3 S31 phosphorylation.

Stem cell models help crack regional oncohistone codes driving childhood gliomas.

In this issue of Cell Stem Cell, Funato etal. (2021) and Bressan etal. (2021) use stem cells as models to define functions of the histone H3.3 G34R mutation in childhood gliomas. Both studies find strong regional specificity to oncohistone activity and implicate specific elements of an aberrantly locked-in neural progenitor transcriptional circuitry.

Dissecting the impact of regional identity and the oncogenic role of human-specific NOTCH2NL in an hESC model of H3.3G34R-mutant glioma

H3.3G34R-mutant gliomas are lethal tumors of the cerebral hemispheres with unknown mechanisms of regional specificity and tumorigenicity. We developed a human embryonic stem cell (hESC)-based model of H3.3G34R-mutant glioma that recapitulates the key features of the tumors with cell-type specificity to forebrain interneuronal progenitors but not hindbrain precursors. We show that H3.3G34R, ATRX, and TP53 mutations cooperatively impact alternative RNA splicing events, particularly suppression of intron retention. This leads to increased expression of components of the Notch pathway, notably NOTCH2NL, a human-specific gene family. We also uncover a parallel mechanism of enhanced NOTCH2NL expression via genomic amplification of its locus in some H3.3G34R-mutant tumors. These findings demonstrate a novel mechanism whereby evolutionary pathways that lead to larger brain size in humans are co-opted to drive tumor growth.

Genetic signature and treatment of pediatric high-grade glioma.

Pediatric high-grade glioma (HGG) is a type of malignancy that carries a poor prognosis. The genetic analysis of HGGs has previously identified useful mutations, the targeting of which has improved prognosis. Thus, further research into the more common mutations, such as H3 histone variants (HIST1H3B and H3F3A) and BRAF V600E, may be useful in identifying tumors with different prognoses, as the mutations are considered to drive two distinct oncogenic programs. The present study performed a retrospective analysis of pediatric HGGs. In total, 42 cases of HGG, including 32 cases (76.1%) of anaplastic astrocytoma and 10 cases (23.8%) of glioblastoma multiforme (GBM), were assessed. The median age of the patients was 7 years (range, 0-32 years). Mutations on histone H3, in particular the K27M and G34R mutations in the distinct variants HIST1H3B and H3F3A, in addition to the presence of the BRAF V600E mutation, were analyzed in 24 patients. The H3F3A K27M mutation was identified in 7 patients (29.1%), while the HIST1H3B K27M mutation was only observed in 1 patient with GBM. In addition, 5 patients harbored a BRAF V600E mutation (21%), while the H3F3A G34R mutation was not recorded in any of the patients. The overall survival of the wild-type patients at 20 months was 68% [confidence interval (CI): 38-85%] compared with 28% (CI: 0.4-60%) in patients with the H3F3A K27M mutation. These results suggested that patients with the H3F3A K27M mutation had a worse prognosis compared with wild-type patients (P=0.0045). Moreover, 3/5 patients with the BRAF V600E mutation had HGGs that were derived from a previous low-grade glioma (LGG; P=0.001). In conclusion, these results suggested that histone H3 mutations may help predict the outcome in patients with HGG. In addition, the BRAF V600E mutation was found to be associated with an increased risk of anaplastic progression. The novel data of the present study may help better define the clinical and radiological characteristics of glioma.

Surprising phenotypic diversity of cancer-associated mutations of Gly 34 in the histone H3 tail.

Sequencing of cancer genomes has identified recurrent somatic mutations in histones, termed oncohistones, which are frequently poorly understood. Previously we showed that fission yeast expressing only the H3.3G34R mutant identified in aggressive pediatric glioma had reduced H3K36 trimethylation and acetylation, increased genomic instability and replicative stress, and defective homology-dependent DNA damage repair. Here we show that surprisingly distinct phenotypes result from G34V (also in glioma) and G34W (giant cell tumors of bone) mutations, differentially affecting H3K36 modifications, subtelomeric silencing, genomic stability; sensitivity to irradiation, alkylating agents, and hydroxyurea; and influencing DNA repair. In cancer, only 1 of 30 alleles encoding H3 is mutated. Whilst co-expression of wild-type H3 rescues most G34 mutant phenotypes, G34R causes dominant hydroxyurea sensitivity, homologous recombination defects, and dominant subtelomeric silencing. Together, these studies demonstrate the complexity associated with different substitutions at even a single residue in H3 and highlight the utility of genetically tractable systems for their analysis.

Correlation Between Immunohistochemistry and Sequencing in H3G34-Mutant Gliomas.

Recurrent glycine-to-arginine/valine alterations at codon 34 (G34R/V) within H3F3A gene characterize a subset of hemispheric high-grade gliomas (HGG) affecting children and young adults. These tumors, defined as G34R/V-mutant gliomas, are histologically heterogenous, with microscopic features of either HGG or embryonal tumors (primitve neuroectodermal tumor-like features). To assess the value of immunohistochemistry (IHC) to detect G34R/V-mutated cases, we tested anti-histone G34V (clone 329E5) and anti-histone G34R (clone RM240) antibodies in a series of 28 formalin-fixed and paraffin-embedded samples. A total of 28 cases of hemispheric, IDH-wt HGG mainly affecting children and young adults were evaluated by IHC and by sequencing. The median age of patients at diagnosis was 17 years (0.1 to 26 y). By IHC, 10 of the 28 cases showed nuclear positivity for G34R and 3 of the 28 cases for G34V. Molecular analysis of G34R/V-mutation status was successful in 24 of the 28 cases. Mutation at glycine 34 of the H3F3A gene was identified in 9 of the 24 tumors (37%) by direct sequencing, revealing 7 of 9 positive case by sequencing and 2 of 9 false negative cases by IHC. Two of 15 negative case by sequencing demonstrated a false positivity by IHC. In total, in 4 (16.6%) of 24 cases, IHC and mutational results were discordant: 2 tumors were negative by IHC (false negative) but harbored G34R mutation by sequencing, and 2 cases were positive by IHC (false positive by IHC) but wild type by sequencing. Moreover, most mutated cases showed loss of ATRX expression and/or p53 expression. The positivity by IHC with specific antibody tested is not highly predictive for presence of G34R/V mutation, but confirmation by sequencing is mandatory; G34R/V mutations should be suspected in all hemispheric tumor IDH1/2 wild type, showing loss of OLIG2 and ATRX and/or p53 expression.

High-grade gliomas with H3 G34R mutation: a clinicopathological study

Objective: To analyze the clinicopathological features and probable mechanisms of high-grade gliomas with H3 G34R mutation. Methods: Five cases of high-grade gliomas with H3 G34R mutation were collected at Xuanwu Hospital, Capital Medical University, Beijing, China, from 2016 to 2019. The clinical and pathological data for each case was retrospectively reviewed. Results: The 5 patients (2 males and 3 females) aged from 15 to 45 years (mean 23 years), and had a history of headache or motor weakness. Four of them were younger than 20 years of age. Magnetic resonance imaging showed that the lesions of 3 cases were seen separately in frontal lobe, parietal lobe or temporal lobe, 1 case involved both frontal lobe and parietal lobe, and otherwise multiple lobes were involved in 1 case. Contrast enhancement could be observed in 2 cases. Pathological examination showed that glioblastoma was the most common entity, with or without primitive neuronal component. All 5 cases showed that H3 G34R was diffusely positive in tumor nuclei with ATRX loss. Moreover, p53 was overexpressed in 4 cases. None of them showed Olig2 expression. Two patients showed disease progression after surgery at 18 months and 24 months, respectively. The latter of the two deceased 3 months after tumor progression. Conclusions: The clinicopathological and molecular genetics features of high-grade gliomas with H3 G34R mutation have relatively similar clinicopathological and genetic features, and more commonly seen in young adults (vs. older adults). Thus, these tumors may be discussed further as a distinct tumor entity.

TBIO-16. NOTCH1 PATHWAY AS TARGET FOR DRUG INTERVENTION FOR HISTONE 3 G34R MUTATED pHGG

There have been no significant improvements in the treatments for childhood High-Grade Glioma (pHGG) and Diffuse Intrinsic Pontine Glioblastoma (DIPG), which continue to have a very poor prognosis. These cancers harbor mutations affecting histone 3 (H3) proteins; 80% of DIPGs with histone H3 K27M somatic mutations whilst 30% of pHGGs exhibit H3.3 G34R or G34V mutations. We have generated and validated a histone 3.3 G34R mutant-specific antibody and investigated the downstream effects of H3.3 G34R mutations in pHGG. In order to identify the genes that may be deregulated by G34R mutant histone expression, we have performed chromatin immunoprecipitation (ChIP) assays with our H3.3 G34R and wild type H3 antibodies, using pHGG H3 G34R mutant and wild-type cell lines. Initial analyses of ChIP data have implicated deregulation of cell-signaling pathways including Notch1, Hedgehog, PPAR-1, PLC-beta and Androgen, in H3 G34R mutated pHGG. We are currently determining the effects of altered expression of Notch1 pathway components on tumorigenesis of H3 G34R mutated pHGG, through gene and protein expression and inhibition assays. Specifically we find that the Notch1 pathway component HES1 shows increased expression in G34R mutant cells compared to controls, directing our evaluation of the utility of gamma-secretase inhibitors as potential therapeutics. These analyses may underpin development of novel treatment strategies for H3 mutated pHGG.

HGG-38. A COMPARATIVE PROTEOMIC-ANALYSIS OF THE CELL MEMBRANE FRACTIONS OF HISTONE 3 MUTATED BRAIN TUMOURS TO IDENTIFY NOVEL THERAPEUTICS

Improvements in the treatments for childhood and adolescent brain tumours, High-Grade Glioma (pHGG) and Diffuse Intrinsic Pontine Glioblastoma (DIPG), have not advanced much and they continue to carry a very poor prognosis. These brain tumours are now defined by mutations affecting histone 3 proteins, indeed 80% of DIPGs harbour histone H3.1 and H3.3 K27M somatic mutations whilst 30% of pHGGs exhibit H3.3 G34R or G34V mutations. We hypothesized that the histone 3 mutant tumours will have distinct mutation specific surfactome. We therefore analysed the cell surface proteomics of pHGG and DIPG, in order to identify novel targets for therapy. We have at first isolated the cell membrane fractions from a range of patient cells carrying different histone 3 mutations (G34R, G34V, K27M), relative to wild type histone 3. A comparative quantitative mass-spectrometry analyses of these cell surface membrane fractions is then performed to identify specific targetable factors, which can be then be used for tumour specific precision-therapy. Results of these experiments will be presented.

IMG-12. CHARACTERISATION OF MODELS OF H3F3A_G34R/V MUTANT PAEDIATRIC GLIOBLASTOMA IN VIVO USING MAGNETIC RESONANCE IMAGING

Approximately 15% of paediatric/young adult cerebral hemispheric glioblastomas (pGBM) harbour G34R/V mutations in H3F3A, encoding the histone H3.3 variant. Development of novel therapeutic interventions demands models that accurately recapitulate this subset of disease and sensitive imaging methods with which to study tumours in situ. Three H3F3A_G34R primary-patient-derived cultures, alongside established cell-line KNS42 (H3F3A_G34V), were implanted orthotopically in immunocompromised mice. KNS42 (TP53_R342*) tumours were clearly detectable using T2-weighted (T2w)-MRI, enhanced following contrast agent administration, indicating impaired blood-brain barrier (BBB) integrity, and demonstrated minimal invasion. OPBG_GBM_001 cells (TP53_89-90X,ATRX_II2133-2144X) formed infiltrative tumours that were hyperintense on T2w-MRI and demonstrated contrast-enhancement suggestive of heterogeneous BBB integrity. HSJD_GBM_002 cells (TP53_P278T,ATRX_R666*) spread diffusely throughout the brain with their full extent typically not discernible by T2w-MRI, the BBB also remaining intact. No evidence of CHOP_GBM_001 tumour was detected by MRI 11months post-implantation. Immunocompetent syngeneic models using tumour cells induced by mutations modelling hemispheric pGBM (NRAS/shP53/shATRX±H3.3G34R) are being explored. Fast growing heterogeneous lesions with variable contrast-enhancement were identified; the H3.3G34R mutation conferred longer median survival (2 clones:25/28days, control:14days). These models have the advantage of an intact immune system and short latency for initial efficacy studies. Primary pGBM cells yield tumours that are more representative of the spectrum of clinical disease; variable hyperintensity on T2w-MRI corresponding to cellular density, with diffusely infiltrative disease less clearly definable, a paucity of oedema and a range of contrast-enhancement. Pathological features including giant multinucleated cells, and mitotic figures were also evident.

Dendritic Cell Vaccination is Effective Against H3.3G34R Mutant Glioblastoma in a Novel Syngeneic Genetically Engineered Mouse Model

Dendritic Cell Vaccination is Effective Against H3.3G34R Mutant Glioblastoma in a Novel Syngeneic Genetically Engineered Mouse Model

Alternative lengthening of telomeres in molecular subgroups of paediatric high-grade glioma

PurposeThe maintenance of telomere length prevents cancer cell senescence and occurs via two mutually exclusive mechanisms: (a) reactivation of telomerase expression and (b) activation of alternative lengthening of telomeres (ALT). ALT is frequently related to alterations on ATRX, a chromatin-remodelling protein. Recent data have identified different molecular subgroups of paediatric high-grade glioma (pHGG) with mutations of H3F3A, TERTp and ATRX; however, differences in telomere length among these molecular subgroups were not thoroughly examined.MethodsWe investigated which genetic alterations trigger the ALT mechanism in 52 IDH-wildtype, 1p/19q-wildtype pHGG. Samples were analysed for telomere length using Tel-FISH. ATRX nuclear loss of expression was assessed by IHC, H3F3A and TERTp mutations by DNA sequencing, and TERTp methylation by MS-PCR.ResultsMutant H3.3 was found in 21 cases (40.3%): 19.2% with K27M mutation and 21.1% with G34R mutation. All H3.3G34R-mutated cases showed the ALT phenotype (100%); on the opposite, only 40% of the H3.3K27M-mutated showed ALT activation. ATRX nuclear loss was seen in 16 cases (30.7%), associated sometimes with the G34R mutation, and never with the K27M mutation. ATRX nuclear loss was always related to telomere elongation. TERTp C250T mutations were rare (5.4%) and were not associated with high intensity Tel-FISH signals, as TERTp hyper-methylation detected in 21% of the cases. H3.3/ATRX/TERTp-wildtype pHGG revealed all basal levels of telomere length.ConclusionOur results show a strong association between H3.3 mutations and ALT, and highlight the different telomeric profiles in histone-defined subgroups: H3.3-G34R mutants always trigger ALT to maintain telomere length, irrespective of ATRX status, whereas only some H3.3-K27M tumours activate ALT. These findings suggest that acquiring the gly34 mutation on H3.3 might suffice to trigger the ALT mechanism.

RACK7 recognizes H3.3G34R mutation to suppress expression of MHC class II complex components and their delivery pathway in pediatric glioblastoma.

Histone H3 point mutations have been identified in incurable pediatric brain cancers, but the mechanisms through which these mutations drive tumorigenesis are incompletely understood. Here, we provide evidence that RACK7 (ZMYND8) recognizes the histone H3.3 patient mutation (H3.3G34R) in vitro and in vivo. We show that RACK7 binding to H3.3G34R suppresses transcription of CIITA, which is the master regulator of MHC (major histocompatibility complex) class II molecules and genes involved in vesicular transport of MHC class II molecules to the cell surface, resulting in suppression of MHC class II molecule expression and transport. CRISPR-based knock-in correction of the H3.3G34R mutation in human pediatric glioblastoma (pGBM) cells significantly reduces overall RACK7 chromatin binding and derepresses the same set of genes as does knocking out RACK7 in the H3.3G34R pGBM cells. By demonstrating that H3.3G34R and RACK7 work together, our findings suggest a potential molecular mechanism by which H3.3G34R promotes cancer.

Abstract B09: Epigenetic changes mediated by H3.3 G34R mutation in a CRISPR-edited pediatric glioblastoma cell line

Abstract Background: Pediatric glioblastomas (pGBM) are aggressive and lethal brain tumors with few effective treatment options. Since the discovery of recurrent histone H3 mutations in pGBMs, multiple groups have sought to uncover the mechanism by which these mutations drive tumorigenesis. One striking aspect of histone H3 mutations is their tumor location and age-specificity: the K27M mutation occurs in midline brain structures in infants and young children, whereas the G34R/V mutations occur in the cerebral hemisphere of adolescents. Efforts to elucidate the mechanism underlying K27M tumorigenesis have centered on K27M-mediated inhibition of the PRC2 complex and the resulting drastic loss of the repressive H3K27me3 mark in these tumors. Studies have shown that G34 mutations affect methylation of the adjacent K36 residue, resulting in a local decrease of the active H3K36me3 mark on the mutant histone. It remains uncertain whether the G34R/V mutations promote global changes in the epigenome that drive tumorigenesis. Methods: To investigate the epigenetic consequences of the G34R mutation, we generated an isogenic in vitro model using CRISPR/Cas9 to correct the G34R mutation in a pGBM-derived cell line. We then comprehensively characterized and compared the epigenome and transcriptome of CRISPR-unedited clones (carrying the G34R mutation) and CRISPR-edited (wild-type) clones. Using ChIP-seq, we profiled H3K36me3 and H3K27me3 marks to study G34R-mediated regulation of these antagonistic marks. Results: ChIP-seq profiling revealed that CRISPR editing of the G34R mutation had minimal effects on total levels or genomic distribution and profile of H3K36me3 and H3K27me3 marks. In contrast, we previously showed that CRISPR editing of the K27M mutation in pGBM cell lines rescued total H3K27me3 levels and genomic distribution to that of histone wild-type pGBMs. In agreement with their similar epigenetic profiles, there were very few transcriptomic differences between G34R and wildtype cells. Conclusions: Our results suggest that the epigenome in G34R-mutant pGBM cells may be less dynamic or amenable to epigenetic perturbation from removal of the initial driver oncomutation than K27M-mutant tumors. These findings underscore that different strategies may be required to target and treat G34R and K27M-mutant pGBM tumors. Citation Format: Shriya Deshmukh, Carol C.L. Chen, Amira Ouanouki, Nada Jabado. Epigenetic changes mediated by H3.3 G34R mutation in a CRISPR-edited pediatric glioblastoma cell line [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 B09.

Reciprocal H3.3 gene editing identifies K27M and G34R mechanisms in pediatric glioma including NOTCH signaling

Histone H3.3 mutations are a hallmark of pediatric gliomas, but their core oncogenic mechanisms are not well-defined. To identify major effectors, we used CRISPR-Cas9 to introduce H3.3K27M and G34R mutations into previously H3.3-wildtype brain cells, while in parallel reverting the mutations in glioma cells back to wildtype. ChIP-seq analysis broadly linked K27M to altered H3K27me3 activity including within super-enhancers, which exhibited perturbed transcriptional function. This was largely independent of H3.3 DNA binding. The K27M and G34R mutations induced several of the same pathways suggesting key shared oncogenic mechanisms including activation of neurogenesis and NOTCH pathway genes. H3.3 mutant gliomas are also particularly sensitive to NOTCH pathway gene knockdown and drug inhibition, reducing their viability in culture. Reciprocal editing of cells generally produced reciprocal effects on tumorgenicity in xenograft assays. Overall, our findings define common and distinct K27M and G34R oncogenic mechanisms, including potentially targetable pathways.