BRD2 bromodomain-mediated regulation of cell state plasticity modulates therapy response in glioblastoma

Raghavendra Vadla; Brett Taylor; Yohei Miyake; Benjamin Lin; Daisuke Kawauchi; Shunichiro Miki; Nidhi Nathwani; Brandon M Jones; Yashpreet Kaur; Abhinaba Banerjee; Philip Pham; Jonathan Tsang; Albert Baldwin; David A Nathanson; Donald P Pizzo; C Ryan Miller; Frank B Furnari

doi:10.1093/neuonc/noaf169

. 2025 Jul 19;27(11):2828–2842. doi: 10.1093/neuonc/noaf169

BRD2 bromodomain-mediated regulation of cell state plasticity modulates therapy response in glioblastoma

Raghavendra Vadla ¹, Brett Taylor ^2,³, Yohei Miyake ⁴, Benjamin Lin ^5,⁶, Daisuke Kawauchi ⁷, Shunichiro Miki ⁸, Nidhi Nathwani ⁹, Brandon M Jones ¹⁰, Yashpreet Kaur ¹¹, Abhinaba Banerjee ¹², Philip Pham ¹³, Jonathan Tsang ¹⁴, Albert Baldwin ¹⁵, David A Nathanson ¹⁶, Donald P Pizzo ¹⁷, C Ryan Miller ¹⁸, Frank B Furnari ^19,^✉

¹ Division of Regenerative Medicine, Department of Medicine, University of California San Diego, La Jolla, California, USA

² Medical Scientist Training Program, University of California, San Diego, La Jolla, California, USA

³ Division of Regenerative Medicine, Department of Medicine, University of California San Diego, La Jolla, California, USA

⁴ Division of Regenerative Medicine, Department of Medicine, University of California San Diego, La Jolla, California, USA

⁵ Medical Scientist Training Program, University of Alabama at Birmingham, Birmingham, Alabama, USA

⁶ Department of Pathology, Division of Neuropathology, Heersink School of Medicine, University of Alabama at Birmingham, Birmingham, Alabama, USA

⁷ Division of Regenerative Medicine, Department of Medicine, University of California San Diego, La Jolla, California, USA

⁸ Division of Regenerative Medicine, Department of Medicine, University of California San Diego, La Jolla, California, USA

⁹ Division of Regenerative Medicine, Department of Medicine, University of California San Diego, La Jolla, California, USA

¹⁰ Division of Regenerative Medicine, Department of Medicine, University of California San Diego, La Jolla, California, USA

¹¹ Division of Regenerative Medicine, Department of Medicine, University of California San Diego, La Jolla, California, USA

¹² Department of Bioengineering, University of California, San Diego, La Jolla, California, USA

¹³ Division of Regenerative Medicine, Department of Medicine, University of California San Diego, La Jolla, California, USA

¹⁴ Departments of Molecular and Medical Pharmacology, University of California, Los Angeles, California, USA

¹⁵ UNC Lineberger Comprehensive Cancer Center, University of North Carolina School of Medicine, Chapel Hill, North Carolina, USA

¹⁶ Department of Bioengineering, University of California, San Diego, La Jolla, California, USA

¹⁷ Department of Pathology, University of California San Diego, La Jolla, California, USA

¹⁸ Department of Pathology, Division of Neuropathology, Heersink School of Medicine, University of Alabama at Birmingham, Birmingham, Alabama, USA

¹⁹ Division of Regenerative Medicine, Department of Medicine, University of California San Diego, La Jolla, California, USA

^✉

Corresponding Author: Frank Furnari, Ph.D., Professor, University of California, San Diego, Sanford Consortium for Regenerative Medicine, 2880 Torrey Pines Scenic Drive, La Jolla, CA 92037, USA (ffurnari@health.ucsd.edu).

PMCID: PMC12700489 PMID: 40686478

Abstract

Background

Glioblastoma (GBM) displays remarkable cell state plasticity, a major contributor to therapeutic resistance and tumor progression. While epigenetic mechanisms play a central role in driving this plasticity, the key regulators remain poorly understood, and developing effective therapeutic strategies targeting them has been challenging.

Methods

We investigated the role of BRD2, a key regulator of NF-κB-mediated mesenchymal (MES) transition, using GBM patient-derived xenograft cell lines, CRISPR-mediated knock-in/knockout approaches, RNA-seq, and in vitro and in vivo modeling. BET inhibitors were employed to target MES gene expression and sensitize GBM to radiation therapy.

Results

We found that PTEN loss induces RelA chromatin localization and acetylation-mediated recruitment of BRD2 to the MES gene promoters. BRD2 binding is essential for maintaining MES gene expression and phenotype. Genetic ablation or loss-of-function mutation of BRD2 bromodomains reverses MES transition, enhances radiation sensitivity, and improves survival in orthotopic xenograft models. Additionally, treatment with a brain-penetrant BD2-selective inhibitor suppresses the MES phenotype and increases radiation sensitivity of GBM stem cells in vitro.

Conclusion

Our study identifies BRD2 as a key mediator of MES transition in GBM, with its bromodomains playing a crucial role in driving cell state plasticity. Targeting BRD2 with BD2-selective inhibitors offers a promising therapeutic strategy to overcome radiation resistance and improve outcomes for GBM patients.

Keywords: BET inhibitors, BRD2, glioblastoma, mesenchymal, NF-κB signaling

Key points.

BRD2 regulates NF-κB-mediated mesenchymal transition in GBM.
BET-BD2 inhibitors inhibit MES transition and sensitize GBM to radiation.

Importance of the Study.

Mesenchymal transition in GBM is strongly linked to therapy resistance and disease recurrence. While NF-κB drives MES gene expression, direct targeting remains unfeasible due to the lack of therapeutically tractable targets. Using CRISPR-based NF-κB modulation, we identified BRD2 as a key regulator of the MES transition and uncovered a mechanism in which GBM cell state plasticity depends on BRD2 bromodomains. Our study provides a compelling rationale for combining BD2-specific inhibitors with standard GBM therapies, highlighting a promising strategy to overcome therapy resistance and improve patient survival.

Glioblastoma (GBM) presents a significant challenge in cancer treatment due to its remarkable ability to switch between different cellular states, a phenomenon known as cell state plasticity.¹ Despite aggressive treatments involving surgical resection, radiotherapy, and chemotherapy, recurrence is almost inevitable.^2,3 Bulk RNA sequencing (RNA-seq) analyses have identified 3 transcriptomic subtypes in IDHwt GBM: classical (CL), mesenchymal (MES), and proneural (PN), with the MES subtype particularly linked to poorer outcomes and survival rates.^1,4 The MES subtype is characterized by higher therapy resistance, increased invasion, and the presence of tumor-associated macrophages/microglia (TAMs).^5–7 Studies indicate that IDHwt tumors with a MES recurrence have a significantly shorter interval between surgeries compared to nonmesenchymal recurrences.⁸ Moreover, tumors tend to transition to the MES subtype upon recurrence.⁹ The biological mechanisms driving this MES transition remain poorly understood.

Multiple oncogenic pathways have been associated with MES transition in GBM.¹⁰ Among them, activation of the NF-κB pathway is crucial in promoting this transition.^6,11 NF-κB signaling enhances the expression of key MES genes as well as various cytokines, chemokines, and growth factors, which collectively modulate the tumor microenvironment and contribute to the enhanced MES phenotype.¹² PTEN is a negative regulator of the NF-κB pathway and is frequently mutated or deleted in GBM.¹³ However, the connection between PTEN and the NF-κB pathway as well as how PTEN may function upstream to influence NF-κB signaling during MES transition in GBM, remains poorly understood.

Bromodomain and Extra Terminal (BET) proteins, which include BRD2, BRD3, BRD4, and BRDT, function as readers of acetylated proteins and play a critical role as coactivators in oncogenic transcriptional programs.¹⁴ BRD4 binds to acetylated RelA at lysine 310, thereby promoting the transcription of inflammatory genes.¹⁵ BRD2 and BRD4 are overexpressed in GBM,¹⁶ but their specific roles in promoting NF-κB-mediated gene expression in GBM remain unclear. BET proteins are involved in epithelial to mesenchymal transition (EMT) in other cancers,^17,18 but their roles in driving MES transition in GBM remain unexplored. Given the growing interest in developing bromodomain-specific inhibitors that can cross the blood–brain barrier,¹⁹ understanding the functions of the RelA/BET complexes in GBM is crucial for designing effective therapeutic strategies.

Herein, we uncover the crucial role of BRD2 and its bromodomains as key modulators of the MES transition in GBM downstream of PTEN/NF-κB signaling. We further demonstrate the effectiveness of BRD2-specific inhibitors in blocking this transition, prolonging mouse survival, and enhancing sensitivity of GSCs to ionizing radiation (IR).

Materials and Methods

Cell Culture and Reagents

Parental U87MG were obtained and cultured as described previously.²⁰ GSCs were obtained as previously described.²¹ GSCs were cultured in DMEM + GlutaMAX medium supplemented with B27 (GIBCO/Life Technologies) and 20 ng/mL human recombinant EGF and 20 ng/mL bFGF. All cells were incubated at 37 °C and 5% CO₂. GSK620 and Ipatasertib were purchased from MedChemExpress. ABBV-744 was provided by Andrew Shiau (UCSD). pLV-IκB-SR vector and pGL4.32[luc2P/NF-κB-RE/Hygro] vector were obtained as described previously.²² Mission shRNA constructs targeting BRD2 were purchased from Sigma.

Protein Subcellular Fractionation and Western Blotting

Subcellular Protein Fractionation Kit for Cultured Cells (ThermoFisher Scientific) was used to separate different protein compartments within cells. Cytoplasmic, nuclear, and chromatin-bound proteins were extracted according to the manufacturer’s instructions (#78840). For whole cell lysates, cells were lysed in RIPA buffer. The following antibodies were used, anti-BRD2 (Cell Signaling Technologies (CST), #5848), anti-RelA/p65 (CST, #8242), anti-BRD4 (CST, #13440), anti-GAPDH (CST, #2118), anti-TBP (CST, #44059), anti-PTEN (Millipore, #04-035), anti-H3 (Novus Biologicals, #NBP1-61519), anti-AKT (CST, #9272), anti-phospho-AKT Thr308 (CST, #9275), anti-acetyl-NF-κB p65 (CST, #3045), p300 (Santa Cruz, #sc-48343), CD44 (CST, #37259), and ANXA1 (CST, #32934).

Generation of CRISPR-engineered Cell Lines

pSpCas9(BB)-2A-GFP (px458) plasmid was used to generate endogenous RELA K310R and PTEN KO GSCs as described.²³ See Supplementary Methods for details.

Immunofluorescence

GSC11 cells expressing wild-type or mutant RelA were seeded on coverslips and treated with TNF-α (20 ng/mL) for 20 min. Cells were then washed with PBS, fixed in 4% PFA for 15 min at room temperature (RT), and permeabilized with 0.3% Triton X-100 for 10 min at RT. Coverslips were then blocked by incubation in 2% BSA in PBS at RT for 30 min, followed by 5% donor bovine serum (ThermoFisher Scientific) for 20 min. Blocked coverslips were then probed with RelA antibodies diluted in PBS containing 2% BSA and incubated in a humified chamber overnight at 4 °C. Next, fluorochrome-conjugated secondary antibody (Invitrogen, #A32731, working concentration 1:1 000) diluted in PBS containing 2% BSA was added and the coverslips were incubated for 2 h at RT in the dark. Coverslips were mounted with Fluoro-Gel II with DAPI (Electron Microscopy Sciences). Imaging was conducted using a Keyence microscope at 20× or 40× magnification.

RNA-seq Data Analysis

Paired-end 150 bp RNA-seq was performed by Novogene Corporation Inc, Sacramento, CA. Fastq file quality was ensured with FastQC, and reads were aligned to the hg38 index using STAR. Alignment quality was confirmed by assessing the final STAR logs. Index and bigwig files were generated from the BAM files with samtools and bamCoverage, respectively. Each BAM file was viewed in IGV to assess the quality and features of mapped reads. The Subread package was used to generate count files, and the raw counts were used as input files to define differentially expressed genes (DEG). Genes with less than 50 total counts across all samples were filtered out for DEG analysis. All the above packages were downloaded and maintained with the Conda package manager. In R, the DESeq2 package was used for DEG analysis. Hierarchical clustering was visualized using heatmaps generated by pheatmap, and volcano plots were created with EnhancedVolcano. Initial DEGs were identified using a Benjamini–Hochberg corrected determined P_adj value of <0.05. For gene ontology (GO) analysis, DEGs were further filtered to be changed by at least 2-fold. GO analysis was performed using IDEP²⁴ and Metascape.²⁵ A gene list Venn diagram was generated using the online tool found at https://www.bioinformatics.org/gvenn/. Gene set variation analysis was performed by using Gene Set Variation Analysis (GSVA; Bioconductor) with gene signatures of interest from MSigDB. GSVA enrichment was performed using counts after variance-stabilizing transformation. Lastly, Gene Set Enrichment Analysis (GSEA) software was used for enrichment analysis of GBM subtypes. Input gene list files were generated with DESeq2. Three biological replicates were used for each experimental condition.

Chromatin Immunoprecipitation-qPCR

Chromatin immunoprecipitation was performed according to the manufacturer’s instructions (Active Motif) with the following modifications. Chromatin was sheared in diluted lysis buffer to 200 to 500 bp using a Covaris M220 Focused Ultrasonicator with the following parameters: 3 min, peak incident power 75%, duty factor 10%, and 200 cycles per burst. ChIP-grade antibodies were obtained from commercially available sources: anti-BRD2 (CST, #5848), anti-RelA/p65 (CST, #8242), anti-BRD4 (CST, #13440). Five percent of the chromatin was not exposed to antibodies and was used as an input control. DNA quantity for each ChIP sample was normalized against input DNA.

Quantitative Real-time PCR

RNA was extracted with the RNeasy Plus kit (Qiagen, #74134) according to the manufacturer’s instructions. Reverse transcription of mRNA was performed using 3-5 μg RNA with RNA to cDNA EcoDry Premix (Takara, #639549). For real-time PCR analysis, 1 μL of cDNA (10 ng of starting RNA) was amplified per reaction using the iTaq Universal SYBR Green Supermix (Bio-Rad, #1725124) and the Bio-Rad CFX96 qPCR system.

shRNA-mediated Knockdown

For lentivirus production, 293T cells were transfected with BRD2 shRNA (Sigma), psPAX2, and pMD2.G packaging constructs using TransIT-VirusGEN (Mirus). Supernatants containing high-titer lentivirus were collected 48 h after transfection and filtered through a 0.45 µm cellulose acetate filter before use. Viral preparations were then purified by LentiX concentrator (Takara Bio). GSCs were infected for 24 h at 37 °C, and the supernatant containing the virus was replaced with fresh culture medium. Infected cells were selected by puromycin.

Matrigel Invasion Assay

GSCs (1 × 10⁵ cells) were suspended in serum-free culture medium and seeded into 24-well Transwell inserts (8.0 µm). Medium with indicated factors was added to the remaining receiver wells. After 16-24 h, the invaded GSCs were fixed and stained with crystal violet (0.05%, Sigma) and counted as cells per field of view using a Keyence BZ-X700 microscope.

GSC Irradiation and Treatment

Cells were irradiated using X-RAD 320 (Precision X-ray; 320 kV, 12.5 mA) with a single dose of either 2 or 4 Gy.

Orthotopic Xenografts

Animal research experiments were conducted in accordance with the regulations of the UCSD Animal Care Program, protocol number S00192M. Female BALB/c nude mice (4-5 weeks old) were used for intracranial tumor implantation. A total of 1-2 × 10⁵ cells in a 3-5 μL volume was injected stereotactically into the brain (1.0 mm anterior and 2.0 mm right to the bregma, and 3 mm deep from the inner plate of the skull). Tumor growth was monitored using the FMT 2500 fluorescence tomography system (PerkinElmer). For drug treatment studies, mice received daily (5 days/week) oral gavage of vehicle or GSK620 at 30 mg/kg. The vehicle formulation consisted of 10% DMSO, 40% PEG, 5% Tween-80, and 45% saline. Mice were euthanized per institutional guidelines for animal welfare and experimental conduct.

Clonogenic Assay

Cells were exposed to varying doses of IR (0-6 Gy) and then seeded at appropriate densities in 6-well plates. Following irradiation, cells were incubated for 14 days to allow colony formation. Colonies were then fixed with 4% formaldehyde and stained with 0.5% crystal violet. The surviving fraction was calculated by normalizing the plating efficiency of irradiated cells to that of non-irradiated control cells.

Immunohistochemistry

The Histology Core Facility at UCSD Pathology prepared 5 µm formalin-fixed, paraffin-embedded (FFPE) tissue sections. Sections were immunostained for Human Nucleoli (clone NM95; Abcam; ab190710) and Iba-1 (Wako; 019-19741). See Supplementary Methods for details.

Cell Growth Assay

Five replicates were plated per each condition in black-walled plate, clear-bottom 96-well plates. Cell growth was analyzed using an ATPlite 1step assay kit (PerkinElmer 6016731) following the manufacturer’s instructions.

Statistical Analyses

All experiments were repeated at least 3 times. Statistical analyses were performed using GraphPad Prism 9. Data sets were analyzed using unpaired t-tests or multiple comparisons by one-way ANOVA or two-way ANOVA as appropriate. In figures, asterisks indicate *P < 0.05, **P < 0.01, and *** P < 0.001, ****P < 0.0001.

Results

PTEN Loss Drives MES Transition by Regulating the Chromatin Localization of RelA, BRD2, and BRD4 Proteins

We previously demonstrated that NF-κB-mediated recruitment of BRD4 promotes IL-6 transcription in heterogeneous GBM models.²² NF-κB activation is frequently observed in GBM and often coincides with activation of the PI3K-AKT signaling pathway.¹² Given the central role of PTEN in regulating the PI3K-AKT signaling, we hypothesized that BET proteins may be regulated by this pathway in GBM. We overexpressed PTEN in PTEN-deficient GSC11, GSC23, and U87 cells, and assessed the chromatin binding of BRD2 and BRD4 by separating the nuclear lysate into soluble nuclear extract (SNE) and chromatin-bound (CB) fractions. PTEN expression resulted in a decreased association of BRD2 and BRD4 with chromatin, compared to WCL (whole cell lysate) and SNE fractions (Figure 1A and Supplementary Figure S1A). Conversely, PTEN knockout in TS576 cells increased the chromatin localization of BRD2 and BRD4 (Figure 1B). We also observed variable levels of BRD2, BRD4, and RelA in the WCL of different GSC lines, depending on PTEN status, indicating that while PTEN consistently modulates the chromatin association of these proteins, its effect on their overall stability is variable and cell line dependent. The phosphatase activity of PTEN is essential for its tumor-suppressive role and PI3K-AKT inhibition.²⁶ Expression of the phosphatase-dead PTEN (G129R) mutant led to increased chromatin association of BRD2 and BRD4 compared to wild-type PTEN, indicating phosphatase-dependent regulation of BET protein activity (Figure 1C). Moreover, pharmacological inhibition of AKT with Ipatasertib decreased the chromatin localization of BRD2 and BRD4 in concert with the known feedback phosphorylation of Akt at T308²⁷ (Figure 1D). Interestingly, RelA and BET proteins exhibited similar patterns of chromatin localization across all conditions (Figure 1A to D, Supplementary Figure S1A). These results suggest that PI3K-AKT activation downstream of PTEN loss enhances chromatin deposition of RelA, BRD2, and BRD4 proteins in GBM.

Immunoblot and gene expression data labelled A to G. Panels A to D shows levels of BRD2, BRD4 and RelA proteins in soluble nuclear and chromatin bound extracts of cell lines with or without PTEN expression. Panel E shows GSEA for NFkB and MES genes in TS576 cells with or without PTEN expression. Panel F shows qPCR analysis of MES genes. Panel G shows a schematic of PTEN mediated MES regulation via BRD2, BRD4 and RelA chromatin binding. — PTEN negatively regulates chromatin binding of BRD2, BRD4 and RelA. (A-D) Immunoblot analyses of BRD2, BRD4, and RelA in SNE, CBE, and WCL from: (A) GSC11 and GSC23 cells stably expressing either empty vector or wild-type PTEN. (B) TS576 cells with or without PTEN deletion; (C) U87 cells stably expressing wild-type (PTEN-WT) or phosphatase dead (PTEN-G129R) PTEN; (D) U87 cells treated with DMSO or Ipatasertib (1 µM, 24 h). Representative blots (n = 3). TBP, H3, and GAPDH serve as loading controls for SNE, CBE, and WCL fractions, respectively. (E) GSEA graphs from bulk RNA-seq in TS576 cells with (PTEN_WT) or without (PTEN_KO) PTEN expression. (F) qPCR analysis showing MES gene expression in TS576 cells with or without PTEN expression (n = 3, mean ± SD). (G) Schematic model illustrating PTEN-mediated regulation of MES transition in GBM.

Given that NF-κB activation is a known driver of MES transition in GBM,⁶ we investigated whether PTEN loss-induced chromatin association of RelA, BRD2, and BRD4 contributes to this transcriptional reprogramming. RNA-seq on PTEN-deleted TS576 cells revealed significant enrichment of MES-associated gene signatures, NF-κB signaling pathways, and increased invasion compared to parental cells (Figure 1E and F, and Supplementary Figure S1B and S1C). Collectively, these results indicate that PTEN inactivation enhances the chromatin association of RelA and BET proteins, which, in turn, promotes a MES-like transition in GBM (Figure 1G).

Acetylation of RelA at Lysine 310 Drives MES Transition in GBM

RelA undergoes various posttranslational modifications that influence its stability, subcellular localization, and interactions with other proteins.²⁸ Among these, acetylation at lysine 310 (K310) has been shown to promote the recruitment of BRD4 to super enhancers and promoters of NF-κB target genes, thereby activating transcription.²⁹ We hypothesized that RelA K310 acetylation following PTEN loss facilitates the recruitment of BRD2 and BRD4 to chromatin via their bromodomains (BD1 and BD2), thereby promoting MES transition. Consistent with this, we observed elevated levels of RelA K310 acetylation in PTEN-deleted TS576, while re-expression of PTEN reduced acetylation in GSC11 and GSC23 (Figure 2A and Supplementary Figure S2A). Given that the histone acetyltransferase p300 has previously been implicated in mediating RelA acetylation at K310,³⁰ we assessed its expression. We found that p300 levels were upregulated in PTEN-deleted cells and downregulated upon PTEN overexpression (Figure 2A). These findings suggest that increased p300 expression may contribute to enhanced RelA K310 acetylation in PTEN-deficient GBM cells. To delineate the functional role of this acetylation, we generated an endogenous knock-in mutation in the RELA gene by substituting lysine 310 with a nonacetylatable arginine (K310R), which is hereafter referred to as RelA-MUT (Supplementary Figure S2B). This mutation led to marked reduction in RelA K310 acetylation in both TS576 and GSC11 cell lines (Supplementary Figure S2C). Notably, the RelA mutation did not affect nuclear translocation in response to TNF-α, nor did it alter NF-κB reporter activation or its inhibition by a IκBα super-repressor (IκBαSR) (Supplementary Figure S2D and E). Notably, RelA-MUT cells showed a marked reduction in BRD2 and BRD4 chromatin binding relative to RelA-WT GSCs (Figure 2B), while chromatin localization of RelA increased. These findings suggest that K310 acetylated chromatin-bound RelA is required for efficient recruitment of BET proteins to chromatin. To define the transcriptional effects of RelA K310 acetylation, we performed RNA-seq on RelA-MUT and RelA-WT GSC11. Disruption of K310 acetylation led to significantly more downregulated (1300, 2-fold decrease, P_adj < 0.05) than upregulated (668, 2-fold increase, P_adj < 0.05) genes (Figure 2C). The GO analysis of the downregulated genes revealed a significant enrichment for matrisome-associated terms (Supplementary Figure S2F), which are functionally linked to the MES gene signature in GBM.^31,32 To further investigate the link between K310 acetylation and MES identity, we assessed MES gene expression by q-PCR and immunoblotting. MES genes were consistently downregulated in RelA-MUT cells compared to WT controls (Figure 2D and Supplementary Figure S2G). GSVA demonstrated enrichment of the MES signature in RelA-WT cells, whereas RelA-MUT cells displayed enrichment for a PN signature (Figure 2E). Functionally, RelA-MUT cells exhibited reduced invasiveness (Figure 2F and Supplementary Figure S2H) and increased sensitivity to IR (Figure 2G), consistent with a less aggressive, non-MES phenotype (Figure 2G). To evaluate in vivo tumorigenic potential, we orthotopically implanted RelA-WT and RelA-MUT cells into nude mice. RelA-MUT GSC11 cells failed to form tumors (data not shown), while TS576 PTEN KO/RelA-MUT implanted mice displayed similar survival compared to RelA-WT tumors (Supplementary Figure S2I). This discrepancy may reflect intrinsic genetic differences between the 2 models. Nevertheless, staining of endpoint RelA-MUT tumors revealed reduced expression of the MES marker CD44 (Figure 2H). Furthermore, IBA1 staining showed decreased infiltration of TAMs, a feature previously associated with the MES phenotype in GBM³³ (Figure 2I).

Immunoblot and gene expression data labelled A to K. Panel A shows the levels of K310 acetylation of RelA and p300 in TS576 AND GSC11 cells with or without PTEN expression. Panel B shows the chromatin binding of BRD2, BRD4 and RelA in wildtype and mutant RelA expressing GSC11 cells. Panels C, D and E shows differential gene expression by RNA-seq, MES gene expression by qPCR, and GBM subtype association in wildtype and mutant RelA expressing GSC11 cells respectively. Panel F shows invasion of wildtype and mutant RelA expressing GSC11 and TS576 cells. Panels G shows clonogenic survival of wildtype and mutant RelA expressing TS576 cells after irradiation. Panels H and I shows in vivo staining of CD44 and IBA1+ cells in wildtype and mutant RelA expressing TS576 xenografts. Panel J shows BRD2 ChIP-qPCR at MES gene promoters in GSC11 cells. Panel K depicts a schematic of RelA K310 acetylation mediated MES gene expression. — RelA Lysine 310 acetylation promotes MES transition. (A) Immunoblot analysis of acetylated RelA (K310Ac) and p300 in TS576 (left panel) and GSC11 (right panel) cells with or without PTEN expression. (B) Immunoblot analysis of indicated proteins in SNE and CBE fractions of GSC11 cells expressing either RelA-WT or RelA-MUT. Representative blots (n = 3). (C) Volcano plot showing differential gene expression in RelA-WT versus RelA-MUT expressing GSC11 cells. Data are presented as log₂ fold change versus −log₁₀ FDR. (D) q-PCR analysis of MES gene expression in GSC11 and TS576 PTEN KO cells expressing either RelA-WT or RelA-MUT (n = 3; mean ± SD). (E) Heatmap of GSVA scores for GSC11 cells expressing either RelA-WT or RelA-MUT, grouped by GBM subtype. (F) Matrigel invasion assay comparing RelA-WT and RelA-MUT expressing GSC11 and TS576 PTEN KO cells. Invasive cells were quantified across five random fields and expressed as relative percentages (n = 3; mean ± SD). Scale bar, 100 µm. (G) Clonogenic survival assay of TS576 PTEN KO cells expressing RelA-WT or RelA-MUT following IR (4 Gy). Cell colonies were analyzed 14 days after IR (n = 3; mean ± SD). (H) IF staining images of CD44 in the brains of patient-derived xenograft tumor-bearing mice derived from TS576 PTEN KO cells expressing RelA-WT or RelA-MUT. Scale bar, 100 μm. (I) Dual IHC staining of human glioma cells (NM95; purple) and IBA1 + macrophages/microglia (green) in mice GBM tumors from (H), scale bar, 50 µm. Quantification of IBA⁺ cells are shown (n = 3 per each group). (J) ChIP-qPCR analysis of BRD2 occupancy at MES gene promoters in GSC11 cells expressing either RelA-WT or RelA-MUT. Data represent fold enrichment over input (n = 3 biological replicates, each with 2 technical replicates; mean ± SD). (K) Schematic model the role of RelA acetylation in promoting MES gene expression in GBM.

RelA Lysine310 Acetylation is Essential for BRD2 Recruitment to MES Gene Promoters

To determine whether the downregulation of MES genes in RelA-MUT cells results from impaired recruitment of BET proteins to MES gene promoters, we performed ChIP-qPCR to assess the localization of RelA, BRD2, and BRD4 at a panel of extracellular matrix (ECM)-associated MES gene promoters. Both BRD2 and BRD4 were enriched at these promoters in RelA-WT cells (Figure 2J and Supplementary Figure S2J). In contrast, RelA-MUT cells exhibited a marked reduction in BRD2 occupancy, while BRD4 binding was only modestly affected (Figure 2J and Supplementary Figure S2J). Interestingly, RelA-MUT cells demonstrated increased RelA binding at these promoters despite diminished transcription of the corresponding genes (Supplementary Figure S2K). These observations support a model in which BRD2 deposition at RelA-bound regulatory elements is required for full transcriptional activation, particularly at MES gene loci where BRD4 may already be constitutively present to support basal gene expression.³⁴ Together, these findings establish that RelA K310 acetylation is a key molecular event that facilitates BRD2 recruitment to MES gene promoters, thereby promoting their transcription in GBM (Figure 2K).

BRD2 Promotes MES Transition in GBM

We performed BRD2 knockdown to evaluate its role in MES transition and observed a significant reduction in ECM gene expression by RT-qPCR (Figure 3A and Supplementary Figure S3A). Transcriptomic profiling using RNA-seq revealed 841 genes that were significantly downregulated in BRD2-depleted cells compared to controls (Figure 3B). GO analysis of these downregulated genes revealed enrichment for matrisome-associated terms and those related to the MES phenotype, which was overrepresented among the top pathways (Supplementary Figure S3B). Notably, 213 genes were commonly downregulated in both RelA-MUT and BRD2 knockdown conditions (Supplementary Figure S3C). GO analysis of these shared targets identified matrisome-associated terms among the top pathways (Supplementary Figure S3D), suggesting that BRD2 and RelA co-regulate MES gene expression. GSVA and GSEA demonstrated a loss of the MES subtype signature and concurrent enrichment of the CL and PN subtype signatures in BRD2-depleted cells (Figure 3C and D and Supplementary Figure S3E), supporting a role for BRD2 in promoting the PN-to-MES transition. Previous studies have identified the AP-1 transcriptional factor complex as a key driver of MES gene expression in GBM.⁹ Consistent with this, GSEA revealed reduced expression of AP-1 target genes in BRD2 knockdown cells (Supplementary Figure S3F), suggesting BRD2 acts as a co-regulator of transcription factors involved in MES transition. To further support our findings, we analyzed an independent, published RNA-seq dataset³⁵ from GSCs and confirmed a reduction in MES gene expression in BRD2-depleted cells (Supplementary Figure S3G). BRD2 knockdown also downregulated NF-κB pathway genes (Supplementary Figure S3H) and impaired cellular invasiveness (Figure 3E and Supplementary Figure S3I).

Gene expression and in vivo data labelled A to L. Panel A shows BRD2 knockdown in GSC11 cells by western blot. Panel B shows volcano plot for differential gene expression in control and BRD2 knock down GSC11 cells by RNA-seq. Panels C to D show GBM subtype enrichment and MES gene expression in control and BRD2 knock down GSC11 cells. Panel E shows matrigel invasion of GSC11 cells expressing control or BRD2 shRNA. Panels F and G show BRD2 deletion effects on cell growth by ATPlite assay in GSC11 and GSC23 cells and in vivo mouse survival in GSC11 xenografts. Panel H shows in vivo imaging of mice tumors in Panel G . Panel I and J shows staining of xenograft tumors for CD44 and invasion of cells in vivo in GSC11 tumors with or without BRD2 deletion. Panel K presents dual IHC for glioma cells and IBA1+ macrophages. Panel L shows cell proliferation of BRD2 expressing or depleted GSC11 and GSC23 cells post-irradiation. — BRD2 promotes MES transition in GBM. (A) Immunoblot analysis of BRD2 in GSC11 with shCONT or shBRD2. (B) Volcano plot showing differential gene expression in shBRD2 versus shCONT expressing GSC11 cells. Data are presented as log₂ (fold change) versus −log₁₀(FDR). (C) Heatmap of GSVA scores for shBRD2 or ShCTRL expressing GSC11 cells, grouped by GBM subtype. (D) q-PCR analysis of MES gene expression in shBRD2 or shCONT expressing GSC11 cells (n = 3; mean ± SD). (E) Matrigel invasion assay comparing shCONT or shBRD2 expressing GSC11 cells. Invasive cells were quantified across 5 random fields and expressed as relative percentages (n = 3; mean ± SD). Scale bar, 100 µm. (F) (Top) Immunoblot showing BRD2 levels in GSC11 and GSC23 cells. (Bottom) ATPlite assay showing cell growth differences in BRD2 deleted GSC11 and GSC23 cells (n = 3 per group; mean ± SD). (G) Kaplan–Meier survival curves of mice bearing intracranial tumors derived from GSC11 with or without BRD2 deletion. (H) FMT imaging of orthotopic xenografts derived from iRFP720 expressing GSC11 cell with or without BRD2 expression (n = 3 per group). (I) IF staining images of CD44 in the brains of patient-derived xenograft tumor-bearing mice derived from (G). Scale bar, 100 μm. (J) IHC showing in vivo growth of GSC11 cells with or without BRD2 expression. Glioma cells (NM95; purple). Scale bar, 2 mm. (K) Dual IHC staining of human glioma cells (NM95; purple) and IBA1 + macrophages/microglia (green) in mice GBM tumors from (G), scale bar, 50µm. Quantification of IBA⁺ cells are shown (n = 3 per each). (L) Cell proliferation assay with ATPlite showing cell growth differences in GSC11 and GSC23 cells with or without BRD2 expression post-IR (4 Gy) (n = 3 per group; mean ± SD).

Next, we generated BRD2 knockout GSC11 and GSC23 lines (Figure 3F). BRD2 KO cells displayed reduced proliferation, consistent with the known role of BRD2 in cyclin regulation³⁶ (Figure 3F). BRD2 KO impaired the upregulation of MES signature genes following TNF-α induced NF-κB activation (Supplementary Figure S3J), reinforcing its role in NF-κB dependent MES gene regulation. Orthotopic implantation of BRD2 KO cells into immunocompromised mice resulted in significantly prolonged survival compared to controls (Figure 3G and H). Immunofluorescence (IF) analysis of endpoint tumors revealed decreased expression of the MES marker CD44, diminished tumor invasion, and a lower abundance of TAMs in BRD2 KO tumors (Figure 3I to K), consistent with a suppressed MES phenotype. Notably, BRD2 KO tumors exhibited increased Ki67 staining, indicative of an enrichment in proliferative, PN-like tumor cells⁹ (Supplementary Figure S3K). Additionally, BRD2-deficient cells exhibited enhanced sensitivity to both IR and temozolomide (TMZ) (Figure 3L and Supplementary Figure S3L). Collectively, these findings identify BRD2 as a critical regulator of MES transition in GBM and its loss promotes a proliferative PN-like state, enhancing sensitivity to standard therapies.

BRD2 Bromodomains are Essential for MES Transition

BRD2 contains 2 N-terminal bromodomains, BD1 and BD2, which recognize and bind to acetyl-lysine residues on histones and other regulatory proteins.¹⁴ To elucidate the role of these bromodomains in regulating the MES phenotype, we re-expressed either wild-type BRD2 (BRD2-WT), or bromodomain-mutant variants-BRD2-Y113F (BD1 mutant), BRD2-Y386F (BD2 mutant), and BRD2- Y113F/Y386F (double mutant)³⁶ in BRD2 knockout GSC11 and GSC23 cells (Figure 4A and Supplementary Figure S4A). In GSC11, re-expression of either wild-type or mutant BRD2 had no significant effect on proliferation (Figure 4B). However, RNA-seq followed by GSVA enrichment analysis for Verhaak GBM subtypes revealed that mutation of either BD1, BD2, or both resulted in loss of the MES signature (Figure 4C and Supplementary Figure S4D). Notably, disruption of BD1 or BD2 individually led to enrichment of the CL subtype signature, while the mutation of both bromodomains resulted in enrichment of the PN subtype signature (Figure 4C and Supplementary Figure S4D). These results demonstrate that both BD1 and BD2 are necessary for the maintenance of CL and MES transcriptional programs in GSC11. In contrast, in GSC23 cells, expression of BRD2 bromodomain mutants led to increased proliferation relative to BRD2-WT (Supplementary Figure S4B). RNA-seq analysis indicated that these cells exhibited reduced MES gene expression and increased expression of hallmark G2/M checkpoint genes, consistent with elevated cell cycle activity (Figure 4C and Supplementary Figure S4E). Importantly, MES gene expression in both GSC11 and GSC23 was dependent on intact BRD2 bromodomains. Inhibition of either domain was sufficient to downregulate the MES subtype signature. Disruption of BD1 resulted in more downregulated genes than BD2 (Figure 4D), with substantial overlap between gene sets co-regulated by both bromodomains (Figure 4E). Functionally, cells expressing BD1, BD2, or double mutants showed a significant reduction in colony numbers, with the BD1/BD2 double mutant having the most pronounced effect (Figure 4F). In vivo, orthotopic implantation of GSCs expressing BRD2 BD mutants in mice resulted in prolonged survival compared to those implanted with BRD2-WT cells (Figure 4G). Collectively, these data underscore the essential role of BRD2 bromodomains in maintaining the MES state in GBM. Disruption of either BD1 or BD2 is sufficient to impair MES gene expression, reduce tumorigenicity, and improve survival in vivo, highlighting BRD2 bromodomain inhibition as a potential therapeutic strategy for MES-subtype GBM.

Graphs and survival data labelled A to G. Panel A shows schematic of BRD2 with mutations of its bromodomains and immunoblots from BRD2 deleted GSC11 cells rescued with BRD2 wildtype and BD1 or BD2 mutants. Panel B shows cell growth differences among BRD2 mutant expressing GSC11 cells via ATPlite assay. Panels C display GBM subtype enrichment across BRD2 mutant expressing cells by RNA-seq in GSC11 and GSC23 cells. Panel D shows differentially expressed genes in cells expressing BD1 and BD2 mutants in GSC11 and GSC23 cells. Panel E shows a Venn diagram of shared downregulated genes between BD1 and BD2 mutants in GSC11 and GSC23 cells. Panel F shows clonogenic growth of BRD2 bromodomain mutants expressing GSC23 cells. Panel G shows survival curves of mice bearing tumors from GSC11 and GSC23 cells expressing BRD2-WT or mutants. — GBM cell MES transition is mediated by BRD2 bromodomains. (A) Schematic illustration of the domains of BRD2 (left). IB analysis of BRD2 deleted GSC11 cells rescued with BRD2-WT (wildt-ype), BRD2-BD1mut (Y113F), BRD2-BD2mut (Y386F), and BRD2-BD1/2mut (Y113F/Y386F). (B) ATPlite assay showing cell growth differences in GSC11 cells expressing BRD2-BD mutants (n = 3 per group; mean ± SD). One-way ANOVA with Dunnett’s multiple-comparison test. (C) Heatmap of GSVA scores for GSC11 cells GSC11 and GSC23 cells expressing BRD2–BD mutants, grouped by GBM subtype. (D) Bar plots showing gene expression differences in GSC11 and GSC23 cells expressing BRD2–BD mutants. (E) Venn diagram showing shared genes that are commonly downregulated in GSC11 and GSC23 expressing BRD2-BD1 vs BRD2-BD2 mutants. (F) Clonogenic assay with GSC23 cells expressing BRD2-BD mutants (n = 3; mean ± SD). (G) Kaplan–Meier survival curves of mice bearing intracranial tumors derived from GSC11 (n = 3) and GSC23 (n = 5) expressing BRD2-WT or BRD2-BD mutants.

BET-BD2 Inhibitors Downregulate MES Gene Expression in GBM

BET inhibitors have shown potent antitumor activity in preclinical studies.³⁷ However, their clinical translation has been hindered by significant toxicity.³⁸ Recently developed BD2-selective inhibitors demonstrated reduced toxicity compared to pan-BET or BD1-selective inhibitors.^34,39 Our results indicate that disruption of BRD2–BD2 is sufficient to downregulate MES gene expression and extend survival in GBM-bearing mice. Based on this, we hypothesized that BD2-selective inhibition may effectively suppress MES gene expression while minimizing cytotoxic effects relative to BD1 or pan-BET inhibitors. Our IC50 analysis across multiple GSC lines found that the BD2-selective inhibitor (GSK620) exhibited higher IC50 values than the BD1 inhibitor (GSK778) (Figure 5A and Supplementary Figure S5A), consistent with the lower cytotoxicity profile previously reported for BD2-specific agents.³⁴ Furthermore, treatment of GSCs with 2 BD2-selective inhibitors (GSK620 and ABBV-744) resulted in significant downregulation of many MES-associated genes, as assessed by RT-qPCR (Figures 5B and Supplementary Figure S5B). We next performed RNA-seq on GSC11 and GSC23 cells treated with GSK620 and identified DEGs( ≥ 2-fold change, padj < 0.05). GSK620 treatment downregulated 194 genes in GSC11 and 384 genes in GSC23 (Figure 5C). Notably, CHI3L2, a key MES marker implicated in immune cell infiltration in glioma^¹⁰, was among the genes commonly downregulated in both lines (Figure 5D). GSEA further confirmed a significant loss of MES gene signatures following GSK620 treatment in both GSCs and in a mouse MES GBM model⁴⁰ (Figure 5E and Supplementary Figure S5C). BD2 inhibition also suppressed transcriptional programs associated with AP-1 and NF-κB signaling pathways (Supplementary Figure S5D). Together, these results demonstrate that BD2-selective BET inhibitors effectively downregulate MES gene expression in GBM models, supporting their translational promise as a therapeutic strategy for targeting the MES subtype of GBM.

Graphs and gene expression labelled A to E. Panel A shows IC50 curves for GSC11 and GSC23 treated with BD1 inhibitor (GSK778) or BD2 inhibitor (GSK620) using ATPlite assays. Panel B shows qPCR analysis of MES genes in GSC11 and GSC23 cells treated with GSK620. Panel C displays bar plots of differential gene expression in GSC11 and GSC23 treated with GSK620. Panel D presents a volcano plot of differentially expressed genes following GSK620 treatment in GSC11 and GSC23 cells. Panel E shows GSEA plots from RNA-seq of GSC11 and GSC23 treated with GSK620 or DMSO. — BET-BD2 inhibitors downregulate MES gene expression. (A) IC50 assays with ATPlite showing cell proliferation in GSCs treated with varying doses of BD1 inhibitor (GSK778) or BD2 inhibitor (GSK620) (n = 3). (B) qPCR analysis of MES gene expression in GSC11 and GSC23 cells treated with GSK620 (0.5-5 µM, 24-48 h). (C) Bar plots showing gene expression differences in GSC11 and GSC23 cells treated with GSK620 (0.5-5 µM, 24-48 h). (D) Volcano plot showing differential gene expression in GSCs treated with GSK620 (0.5-5 µM, 24-48 h). Data are presented as log₂ (fold change) versus −log₁₀ (FDR). (E) GSEA graphs from bulk RNAseq for GSCs treated with GSK620 or DMSO (0.5-5 µM, 24-48 h).

GSK620 is a Brain-Penetrant BET-BD2 Inhibitor that Extends Survival in GBM Models and Sensitizes GSCs to Therapy

To assess the effect of GSK620 on the chromatin localization of the BET proteins, we treated GSC11 cells with the BD2-selective inhibitor GSK620 and performed nuclear fractionation as described before (Figure 1). Treatment with GSK620 led to a more pronounced reduction in chromatin-bound BRD2, compared to BRD4 or RelA (Figure 6A), suggesting that BD2 inhibitor preferentially affects BRD2 chromatin binding. Consistently, ChIP-qPCR analysis revealed decreased BRD2 occupancy at MES gene promoters following GSK620 treatment, whereas BRD4 levels at these loci remained unchanged (Figure 6B). These findings suggest that the BD2 domain of BRD2 is essential for its recruitment to MES gene regulatory regions. Functionally, BD2 inhibition reduced GSC invasiveness (Figure 6C and Supplementary Figure S6A), and significantly impaired clonogenic survival (Figure 6D and Supplementary Figure S6B).

Graphs and survival data labelled A to H. Panel A shows immunoblots of BRD2, BRD4 and RelA in soluble nuclear and chromatin bound fractions after DMSO or GSK620 treatment for 24 hours. Panel B presents ChIP-qPCR of BRD2 and BRD4 at MES gene promoters after DMSO or GSK620 treatment after 24 hours. Panels C to D shows matrigel invasion and clonogenic growth after DMSO or GSK620 treatement in GSC11 and GSC23 cell lines. Panel E shows survival curve for mice xenografted with GSC11 cells treated with vehicle or GSK620. Panel F presents IHC of glioma cells and IBA1+ macrophages in tumors. Panel G shows growth curves of GSC267, GSC23 and GSC11 cells treated with DMSO, GSK620 and IR by ATPlite assay. Panel H illustrates a schematic model of BRD2/RelA-driven MES regulation in GBM. — BET-BD2 inhibitors suppress MES phenotype, prolong survival, and sensitizes GSCs to IR. (A) IB analysis of indicated proteins in SNE and CBE fractions of GSC11 cells treated with DMSO or GSK620 (1 µM, 24 h). Representative blots (n = 3). (B) ChIP-qPCR analysis of BRD2 and BRD4 occupancy at MES gene promoters in GSC11 cells treated DMSO or GSK620 (1 µM, 24 h). Data represent fold enrichment over input (n = 3 biological replicates, each with 2 technical replicates; mean ± SD). C) Matrigel invasion assay comparing GSC11 cells treated with DMSO or GSK620 (0.5 µM, 24 h). Invasive cells were quantified across 5 random fields and expressed as relative percentage (n = 3; mean ± SD). Scale bar, 100 µm. (D) Clonogenic assay with for GSCs treated with DMSO or GSK620 (0.5-5 µM, 24 h). Data represents mean ± SD. (E) Kaplan–Meier survival curves for mice bearing intracranial tumors derived from GSC11 after treatments with vehicle or GSK620 (n = 5). (F) Dual IHC staining of human glioma cells (NM95; purple) and IBA1 + macrophages/microglia (green) in mice GBM tumors derived from (E), scale bar: 250 µm. Quantification of IBA⁺ cells are shown (n = 3 per each). (G) ATPlite assay of GSCs treated with DMSO or GSK620 (0.5-5 µM, 24 h) were exposed to 1 dose of IR (2-4 Gy) and assessed for cell growth on indicated timepoints (n = 3 per group; mean ± SD). (H) Schematic model illustrating RelA/BRD2 mediated regulation of MES transition in GBM. Created in BioRender. Vadla, R. (2025) https://BioRender.com/2alntun

To evaluate the pharmacokinetic properties of GSK620 and its brain penetrance, we conducted in vivo analysis in mice treated with oral GSK620.³⁴ The compound exhibited a brain-to-plasma area under the curve (AUC_0-24h) ratio of 27.2% and a plasma half-life (t½) of 1.05 h (Supplementary Figure S6C), comparable to the kinetics of TMZ⁴¹ and indicating sufficient CNS penetration. To assess therapeutic efficacy in vivo, we orthotopically implanted GSC11 cells into mice and initiated GSK620 treatment after tumor establishment²¹ (Supplementary Figure S6D). GSK620-treated mice lived significantly longer than vehicle-treated controls (Figure 6E). IHC staining showed reduced infiltration of TAMs (Figure 6F). Given the well-documented resistance of MES GBM cells to standard therapies,⁵ we tested whether BD2 inhibition sensitizes GSCs to IR. Pretreatment with GSK620 followed by IR significantly decreased GSC proliferation compared to either treatment alone (Figure 6G). Similar radiosensitization was observed with another BD2 inhibitor, ABBV-744 (Supplementary Figure S6E). Notably, GSK620 did not impair the proliferation of neural progenitor cells (NPC) at the doses that sensitized GSCs to IR, indicating a favorable therapeutic index (Supplementary Figure S6F). Together, these findings establish GSK620 as a brain-penetrant, BD2-selective BET inhibitor that downregulates MES gene expression, suppresses tumor aggressiveness, extends survival in GBM models, and enhances the sensitivity of GSCs to radiotherapy.

Discussion

In this study, we elucidated a critical role of BRD2 and its bromodomains in regulating GBM MES transition. Using CRISPR-mediated editing of the endogenous RELA gene in GSCs, we demonstrate that acetylation of RelA at lysine 310 recruits BRD2 to the promoters of MES genes, thereby driving their expression. BRD2 depletion abrogates the MES-like state and sensitizes GBM cells to standard therapies, specifically IR. Domain-specific mutagenesis revealed that both BD1 and BD2 are essential for maintaining the MES state, with loss of either shifting cells toward other transcriptional subtypes. Pharmacologic inhibition with BD2-selective BET inhibitors effectively downregulated MES gene expression, enhanced therapy response, and extended survival in GBM-bearing mice (Figure 6H).

The role of BET proteins in GBM is well established, with BRD4 being the most extensively studied member.¹⁴ BRD4 regulates key oncogenic programs including MYC, NOTCH1, NF-κB, and cell cycle genes, supporting cell proliferation, stemness, and therapy resistance.^42–44 Its elevated expression in high-grade gliomas, success in other cancers, and correlation with poor prognosis have made it a focus of therapeutic efforts in GBM. However, the essentiality of BRD4 in both tumor and normal cells posed a significant barrier to clinical translation, as pan-BET inhibitors led to adverse toxicities in clinical trials.⁴⁵ Although BRD2 has also been shown to be expressed at high levels in GBM,¹⁶ its role and function have not been well characterized.³⁵

In our study, we observed increased chromatin association of both BRD2 and BRD4 upon PTEN loss, suggesting that BRD4 may be involved in MES transition in GBM. Indeed, BRD4 plays a crucial role in EMT in other cancers,¹⁷ suggesting a similar role in GBM. We were unable to generate homozygous BRD4 knockout clones in GSCs due to its essential role in cell viability, but BRD2 deletion was tolerated and led to a delay in cell growth along with suppression of MES gene expression. These findings suggest that BRD2 plays a key role in maintaining tumor cell plasticity and adaptive potential, unlike the essential survival gene BRD4. These observations prompted us to focus on BRD2, a comparatively less-characterized member of the BET family in GBM. Our data demonstrate that BRD2 loss disrupts MES transition and impairs tumor progression. BRD2 depletion also led to downregulated AP-1 gene signatures associated with MES transition,⁹ suggesting that BRD2 might be a common epigenetic modulator for multiple MES-promoting pathways in GBM. Importantly, we show that targeting BRD2 using a BD2-selective inhibitor sensitizes GBM cells to radiation therapy. This effect may be linked to impaired DNA damage response due to BET inhibition.⁴⁶

IDH-WT gliomas consistently exhibit a limited set of epigenetically defined cell states, including OPC/NPC-like (collectively referred as PN), AC-like, and MES-like that reflect shared adaptive programs.¹ These states differ in their response to therapy, with PN-like cells being more sensitive and MES-like cells more resistant. The concept of “state-selective lethality” proposes trapping GBM cells in a therapeutically vulnerable state, such as the PN, to enhance treatment efficacy.⁴⁷ Our findings show that inhibition of RelA acetylation, BRD2 depletion, or treatment with BD2-selective inhibitors downregulates the MES-like state, promotes a shift toward CL/PN-like states, and sensitizes GSCs to IR. While we tested radiosensitization in vitro, future studies will be necessary to determine whether combining radiation with MES state inhibition using BD2 inhibitors results in a synergistic survival benefit. Notably, even in the absence of radiation, disruption of the MES state through BRD2 deletion, bromodomain mutation, or BD2 inhibition significantly prolonged survival in GBM models, suggesting that MES suppression alone can improve outcomes. The use of BD2-selective BET inhibitors is particularly relevant, as many BET-targeting compounds have already advanced into clinical trials,³⁷ highlighting the translational potential of this strategy for targeting MES cell state in GBM. In GSC11 cells, mutation of either BRD2 bromodomains led to a transition away from the MES state toward a CL state, a shift not observed in GSC23 cells. We reasoned that this difference may be attributed to EGFR amplification in GSC11,⁴⁸ as EGFR signaling is known to promote the CL cell-state. These findings suggest that underlying genetics, such as EGFR amplification, may influence the transcriptional programs adopted by GSCs once the MES state is attenuated. Based on this, it would be interesting to explore whether combining BD2 and EGFR inhibitors could be effective in EGFR-amplified GBMs and/or drive a further shift toward a more therapeutically vulnerable PN-like state.

Recent studies have shown that IR induces significant microenvironmental changes in both GBM patients and mice models, with increased deposition of ECM, TAM accumulation, and MES transition.^49,50 Notably, interactions between immune cells and tumor cells have been implicated in promoting MES transition and tumor aggressiveness.³³ In our study, BRD2 mutant GSC11 and GSC23 cells exhibited similar or even higher cell proliferation rates in vitro yet formed significantly slower growing tumors in vivo. This suggests that the tumor microenvironment, particularly the immune cell component or MES-associated ECM remodeling, may play a crucial role in driving tumor aggressiveness. Moreover, it suggests that these components may be attenuated in BRD2 mutant tumors. Given that IR also enhances ECM deposition, our findings suggest that combining BRD2 bromodomain targeting with BD2-selective inhibitors and standard therapies may help suppress ECM remodeling and prevent aggressive tumor recurrence.

Our findings position BRD2 as a novel, underexplored epigenetic vulnerability in GBM. Unlike pan-BET inhibitors, which have dose-limiting toxicities, we demonstrate that BD2-selective inhibition (BD2i) is less toxic and effectively sensitizes GBM cells to IR. Thus, targeting BRD2, either genetically or pharmacologically, represents a rational and potentially safer strategy to modulate GBM cell state plasticity and improve responsiveness to conventional therapies.

Supplementary material

Supplementary material is available online at Neuro-Oncology (https://academic.oup.com/neuro-oncology).

noaf169_Supplementary_Data

noaf169_supplementary_data.xlsx^{(9.9KB, xlsx)}

noaf169_Supplementary_Figures_S1-S6

noaf169_supplementary_figures_s1-s6.docx^{(5.6MB, docx)}

Acknowledgments

We are grateful to Dr. Andrew Shiau for GSK620. We thank Dr. Krishna Bhat for GBM-PDX neurospheres. We thank Nicole Bergstein for IHC sections. Images were created with Biorender.com

Contributor Information

Raghavendra Vadla, Division of Regenerative Medicine, Department of Medicine, University of California San Diego, La Jolla, California, USA.

Brett Taylor, Medical Scientist Training Program, University of California, San Diego, La Jolla, California, USA; Division of Regenerative Medicine, Department of Medicine, University of California San Diego, La Jolla, California, USA.

Yohei Miyake, Division of Regenerative Medicine, Department of Medicine, University of California San Diego, La Jolla, California, USA.

Benjamin Lin, Medical Scientist Training Program, University of Alabama at Birmingham, Birmingham, Alabama, USA; Department of Pathology, Division of Neuropathology, Heersink School of Medicine, University of Alabama at Birmingham, Birmingham, Alabama, USA.

Daisuke Kawauchi, Division of Regenerative Medicine, Department of Medicine, University of California San Diego, La Jolla, California, USA.

Shunichiro Miki, Division of Regenerative Medicine, Department of Medicine, University of California San Diego, La Jolla, California, USA.

Nidhi Nathwani, Division of Regenerative Medicine, Department of Medicine, University of California San Diego, La Jolla, California, USA.

Brandon M Jones, Division of Regenerative Medicine, Department of Medicine, University of California San Diego, La Jolla, California, USA.

Yashpreet Kaur, Division of Regenerative Medicine, Department of Medicine, University of California San Diego, La Jolla, California, USA.

Abhinaba Banerjee, Department of Bioengineering, University of California, San Diego, La Jolla, California, USA.

Philip Pham, Division of Regenerative Medicine, Department of Medicine, University of California San Diego, La Jolla, California, USA.

Jonathan Tsang, Departments of Molecular and Medical Pharmacology, University of California, Los Angeles, California, USA.

Albert Baldwin, UNC Lineberger Comprehensive Cancer Center, University of North Carolina School of Medicine, Chapel Hill, North Carolina, USA.

David A Nathanson, Department of Bioengineering, University of California, San Diego, La Jolla, California, USA.

Donald P Pizzo, Department of Pathology, University of California San Diego, La Jolla, California, USA.

C Ryan Miller, Department of Pathology, Division of Neuropathology, Heersink School of Medicine, University of Alabama at Birmingham, Birmingham, Alabama, USA.

Frank B Furnari, Division of Regenerative Medicine, Department of Medicine, University of California San Diego, La Jolla, California, USA.

Author contributions

R.V. conceptualized, designed and performed the experiments, analyzed and interpreted the data, generated figures and wrote the manuscript. S.M., Y.M., D.K., N.N., Y.K., A.B.and P.P. performed mouse experiments. B.T. and B.L. performed RNA-seq analysis. D.P.P. performed IHC analysis. J.T. performed PK analysis. D.A.N. supervised PK analysis. C.R.M. supervised RNA-seq analysis and edited the manuscript. F.B.F. conceptualized, designed experiments, provided resources, edited the manuscript, and supervised all aspects of the study.

Funding

This work was supported by NIH [R01NS080939 and R56NS080939 to F.B.F; R01CA258248 and R01NS134798 to F.B.F. and C.R.M], AACR-Stache strong fellowship to R.V., the Japan Society for the Promotion of Science Overseas Research Fellowship to S.M. and Y.M., and the Kurland Family Foundation and P50CA211015 to D.A.N.

Conflicts of interest statement. The authors declare no potential conflicts of interest.

Data availability

Data have been deposited at Gene Expression Omnibus. Bulk RNA-seq data are available under accession number GSE304029.

References

1. Neftel C, Laffy J, Filbin MG, et al. An integrative model of cellular states, plasticity, and genetics for glioblastoma. Cell. 2019;178(4):835–849.e21. [DOI] [PMC free article] [PubMed] [Google Scholar]
2. Stupp R, Hegi ME, Mason WP, et al. ; European Organisation for Research and Treatment of Cancer Brain Tumour and Radiation Oncology Groups. Effects of radiotherapy with concomitant and adjuvant temozolomide versus radiotherapy alone on survival in glioblastoma in a randomised phase III study: 5-year analysis of the EORTC-NCIC trial. Lancet Oncol. 2009;10(5):459–466. [DOI] [PubMed] [Google Scholar]
3. Stupp R, Mason WP, van den Bent MJ, et al. ; European Organisation for Research and Treatment of Cancer Brain Tumor and Radiotherapy Groups. Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma. N Engl J Med. 2005;352(10):987–996. [DOI] [PubMed] [Google Scholar]
4. Verhaak RGW, Hoadley KA, Purdom E, et al. ; Cancer Genome Atlas Research Network. Integrated genomic analysis identifies clinically relevant subtypes of glioblastoma characterized by abnormalities in PDGFRA, IDH1, EGFR, and NF1. Cancer Cell. 2010;17(1):98–110. [DOI] [PMC free article] [PubMed] [Google Scholar]
5. Segerman A, Niklasson M, Haglund C, et al. Clonal variation in drug and radiation response among glioma-initiating cells is linked to proneural-mesenchymal transition. Cell Rep. 2016;17(11):2994–3009. [DOI] [PubMed] [Google Scholar]
6. Bhat KPL, Balasubramaniyan V, Vaillant B, et al. Mesenchymal differentiation mediated by NF-κB promotes radiation resistance in glioblastoma. Cancer Cell. 2013;24(3):331–346. [DOI] [PMC free article] [PubMed] [Google Scholar]
7. Wang Q, Hu B, Hu X, et al. Tumor evolution of glioma-intrinsic gene expression subtypes associates with immunological changes in the microenvironment. Cancer Cell. 2017;32(1):42–56.e6. [DOI] [PMC free article] [PubMed] [Google Scholar]
8. Varn FS, Johnson KC, Martinek J, et al. ; GLASS Consortium. Glioma progression is shaped by genetic evolution and microenvironment interactions. Cell. 2022;185(12):2184–2199.e16. [DOI] [PMC free article] [PubMed] [Google Scholar]
9. Wang L, Jung J, Babikir H, et al. A single-cell atlas of glioblastoma evolution under therapy reveals cell-intrinsic and cell-extrinsic therapeutic targets. Nat. Cancer. 2022;3(12):1534–1552. [DOI] [PMC free article] [PubMed] [Google Scholar]
10. Azam Z, To S shun T, Tannous BA. Mesenchymal transformation: the rosetta stone of glioblastoma pathogenesis and therapy resistance. Adv Sci (Weinh). 2020;7(22):2002015. [DOI] [PMC free article] [PubMed] [Google Scholar]
11. Schmitt MJ, Company C, Dramaretska Y, et al. Phenotypic mapping of pathologic cross-talk between glioblastoma and innate immune cells by synthetic genetic tracing. Cancer Discov. 2021;11(3):754–777. [DOI] [PMC free article] [PubMed] [Google Scholar]
12. Soubannier V, Stifani S. NF- κ B signalling in glioblastoma. Biomedicines. 2017;5(2):29. [DOI] [PMC free article] [PubMed] [Google Scholar]
13. Koul D. PTEN signaling pathways in glioblastoma. Cancer Biol Ther. 2008;7(9):1321–1325. [DOI] [PubMed] [Google Scholar]
14. Belkina AC, Denis GV. BET domain co-regulators in obesity, inflammation and cancer. Nat Rev Cancer. 2012;12(7):465–477. [DOI] [PMC free article] [PubMed] [Google Scholar]
15. Huang B, Yang XD, Zhou MM, Ozato K, Chen LF. Brd4 coactivates transcriptional activation of NF-κB via specific binding to acetylated RelA. Mol Cell Biol. 2009;29(5):1375–1387. [DOI] [PMC free article] [PubMed] [Google Scholar]
16. Pastori C, Daniel M, Penas C, et al. BET bromodomain proteins are required for glioblastoma cell proliferation. Epigenetics. 2014;9(4):611–620. [DOI] [PMC free article] [PubMed] [Google Scholar]
17. Tian B, Zhao Y, Sun H, et al. BRD4 mediates NF-κB-dependent epithelial-mesenchymal transition and pulmonary fibrosis via transcriptional elongation. Am J Physiol Lung Cell Mol Physiol. 2016;311(6):L1183–L1201. [DOI] [PMC free article] [PubMed] [Google Scholar]
18. Serresi M, Kertalli S, Li L, et al. Functional antagonism of chromatin modulators regulates epithelial-mesenchymal transition. Sci Adv. 2021;7(9):eabd7974. [DOI] [PMC free article] [PubMed] [Google Scholar]
19. Engelberg IA, Foley CA, James LI, Frye SV. Improved methods for targeting epigenetic reader domains of acetylated and methylated lysine. Curr Opin Chem Biol. 2021;63:132–144. [DOI] [PMC free article] [PubMed] [Google Scholar]
20. Inda MM, Bonavia R, Mukasa A, et al. Tumor heterogeneity is an active process maintained by a mutant EGFR-induced cytokine circuit in glioblastoma. Genes Dev. 2010;24(16):1731–1745. [DOI] [PMC free article] [PubMed] [Google Scholar]
21. Benitez JA, Finlay D, Castanza A, et al. PTEN deficiency leads to proteasome addiction: a novel vulnerability in glioblastoma. Neuro-Oncology. 2021;23(7):1072–1086. [DOI] [PMC free article] [PubMed] [Google Scholar]
22. Zanca C, Villa GR, Benitez JA, et al. Glioblastoma cellular cross-talk converges on NF-κB to attenuate EGFR inhibitor sensitivity. Genes Dev. 2017;31(12):1212–1227. [DOI] [PMC free article] [PubMed] [Google Scholar]
23. Koga T, Chaim IA, Benitez JA, et al. Longitudinal assessment of tumor development using cancer avatars derived from genetically engineered pluripotent stem cells. Nat Commun. 2020;11(1):550. [DOI] [PMC free article] [PubMed] [Google Scholar]
24. Ge SX, Son EW, Yao R. iDEP: an integrated web application for differential expression and pathway analysis of RNA-Seq data. BMC Bioinf. 2018;19(1):534. [DOI] [PMC free article] [PubMed] [Google Scholar]
25. Zhou Y, Zhou B, Pache L, et al. Metascape provides a biologist-oriented resource for the analysis of systems-level datasets. Nat Commun. 2019;10(1):1523–1523. [DOI] [PMC free article] [PubMed] [Google Scholar]
26. Furnari FB, Lin H, Huang HS, Cavenee WK. Growth suppression of glioma cells by PTEN requires a functional phosphatase catalytic domain. Proc Natl Acad Sci USA. 1997;94(23):12479–12484. [DOI] [PMC free article] [PubMed] [Google Scholar]
27. Yang J, Wu Z, Renier N, et al. Pathological axonal death through a Mapk cascade that triggers a local energy deficit. Cell. 2015;160(1–2):161–176. [DOI] [PMC free article] [PubMed] [Google Scholar]
28. Huang B, Yang XD, Lamb A, Chen LF. Posttranslational modifications of NF-κB: another layer of regulation for NF-κB signaling pathway. Cell Signal. 2010;22(9):1282–1290. [DOI] [PMC free article] [PubMed] [Google Scholar]
29. Brown JD, Lin CY, Duan Q, et al. NFκB directs dynamic super enhancer formation in inflammation and atherogenesis. Mol Cell. 2014;56(2):219–231. [DOI] [PMC free article] [PubMed] [Google Scholar]
30. Chen L, Mu Y, Greene WC. Acetylation of RelA at discrete sites regulates distinct nuclear functions of NF-kappaB. EMBO J. 2002;21(23):6539–6548. [DOI] [PMC free article] [PubMed] [Google Scholar]
31. Yin W, Zhu H, Tan J, et al. Identification of collagen genes related to immune infiltration and epithelial-mesenchymal transition in glioma. Cancer Cell Int. 2021;21(1):1–18. [DOI] [PMC free article] [PubMed] [Google Scholar]
32. Cheng WY, Kandel JJ, Yamashiro DJ, Canoll P, Anastassiou D. A multi-cancer mesenchymal transition gene expression signature is associated with prolonged time to recurrence in glioblastoma. PLoS One. 2012;7(4):e34705. [DOI] [PMC free article] [PubMed] [Google Scholar]
33. Hara T, Chanoch-Myers R, Mathewson ND, et al. Interactions between cancer cells and immune cells drive transitions to mesenchymal-like states in glioblastoma. Cancer Cell. 2021;39(6):779–792.e11. [DOI] [PMC free article] [PubMed] [Google Scholar]
34. Gilan O, Rioja I, Knezevic K, et al. Selective targeting of BD1 and BD2 of the BET proteins in cancer and immunoinflammation. Science. 2020;368(6489):387–394. [DOI] [PMC free article] [PubMed] [Google Scholar]
35. Chen L, Qi Q, Jiang X, et al. Phosphocreatine promotes epigenetic reprogramming to facilitate glioblastoma growth through stabilizing BRD2. Cancer Discov. 2024;14(8):1547–1565. [DOI] [PubMed] [Google Scholar]
36. Hnilicová J, Hozeifi S, Stejskalová E, et al. The C-terminal domain of Brd2 is important for chromatin interaction and regulation of transcription and alternative splicing. Mol Biol Cell. 2013;24(22):3557–3568. [DOI] [PMC free article] [PubMed] [Google Scholar]
37. Trojer P. Targeting BET bromodomains in cancer. Ann Rev Cancer Biol. 2022;6(1):313–336. [Google Scholar]
38. Shorstova T, Foulkes WD, Witcher M. Achieving clinical success with BET inhibitors as anti-cancer agents. Br J Cancer. 2021;124(9):1478–1490. [DOI] [PMC free article] [PubMed] [Google Scholar]
39. Faivre EJ, McDaniel KF, Albert DH, et al. Selective inhibition of the BD2 bromodomain of BET proteins in prostate cancer. Nature. 2020;578(7794):306–310. [DOI] [PubMed] [Google Scholar]
40. Friedmann-morvinski D, Narasimamurthy R, Xia Y, et al. Targeting NF-kB in glioblastoma: a therapeutic approach. Sci Adv. 2016;2(1):e1501292. [DOI] [PMC free article] [PubMed] [Google Scholar]
41. Ballesta A, Zhou Q, Zhang X, Lv H, Gallo JM. Multiscale design of cell-type-specific pharmacokinetic/pharmacodynamic models for personalized medicine: application to temozolomide in brain tumors. CPT Pharmacometrics Syst Pharmacol. 2014;3(4):e112. [DOI] [PMC free article] [PubMed] [Google Scholar]
42. Ma L, Li G, Yang T, et al. An inhibitor of BRD4, GNE987, inhibits the growth of glioblastoma cells by targeting C-Myc and S100A16. Cancer Chemother Pharmacol. 2022;90(6):431–444. [DOI] [PMC free article] [PubMed] [Google Scholar]
43. Ma T, Hu C, Lal B, et al. Reprogramming transcription factors Oct4 and Sox2 induce a BRD-dependent immunosuppressive transcriptome in GBM-propagating cells. Cancer Res. 2021;81(9):2457–2469. [DOI] [PMC free article] [PubMed] [Google Scholar]
44. Tao Z, Li X, Wang H, et al. BRD4 regulates self‐renewal ability and tumorigenicity of glioma‐initiating cells by enrichment in the Notch1 promoter region. Clin Transl Med. 2020;10(6):e181. [DOI] [PMC free article] [PubMed] [Google Scholar]
45. Stathis A, Bertoni F. BET proteins as targets for anticancer treatment. Cancer Discov. 2018;8(1):24–36. [DOI] [PubMed] [Google Scholar]
46. Wang ZQ, Zhang ZC, Wu YY, et al. Bromodomain and extraterminal (BET) proteins: biological functions, diseases, and targeted therapy. Signal Transduct Target Ther. 2023;8(1):420. [DOI] [PMC free article] [PubMed] [Google Scholar]
47. Nicholson JG, Fine HA. Diffuse glioma heterogeneity and its therapeutic implications. Cancer Discov. 2021;11(3):575–590. [DOI] [PubMed] [Google Scholar]
48. Wu S, Gao F, Zheng S, et al. EGFR amplification induces increased DNA damage response and renders selective sensitivity to talazoparib (PARP inhibitor) in glioblastoma. Clin Cancer Res. 2020;26(6):1395–1407. [DOI] [PubMed] [Google Scholar]
49. Watson SS, Duc B, Kang Z, et al. Microenvironmental reorganization in brain tumors following radiotherapy and recurrence revealed by hyperplexed immunofluorescence imaging. Nat Commun. 2024;15(1):3226. [DOI] [PMC free article] [PubMed] [Google Scholar]
50. Jeon HM, Kim JY, Cho HJ, et al. Tissue factor is a critical regulator of radiation therapy-induced glioblastoma remodeling. Cancer Cell. 2023;41(8):1480–1497.e9. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

noaf169_Supplementary_Data

noaf169_supplementary_data.xlsx^{(9.9KB, xlsx)}

noaf169_Supplementary_Figures_S1-S6

noaf169_supplementary_figures_s1-s6.docx^{(5.6MB, docx)}

Data Availability Statement

Data have been deposited at Gene Expression Omnibus. Bulk RNA-seq data are available under accession number GSE304029.

[CIT0001] 1. Neftel C, Laffy J, Filbin MG, et al. An integrative model of cellular states, plasticity, and genetics for glioblastoma. Cell. 2019;178(4):835–849.e21. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0002] 2. Stupp R, Hegi ME, Mason WP, et al. ; European Organisation for Research and Treatment of Cancer Brain Tumour and Radiation Oncology Groups. Effects of radiotherapy with concomitant and adjuvant temozolomide versus radiotherapy alone on survival in glioblastoma in a randomised phase III study: 5-year analysis of the EORTC-NCIC trial. Lancet Oncol. 2009;10(5):459–466. [DOI] [PubMed] [Google Scholar]

[CIT0003] 3. Stupp R, Mason WP, van den Bent MJ, et al. ; European Organisation for Research and Treatment of Cancer Brain Tumor and Radiotherapy Groups. Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma. N Engl J Med. 2005;352(10):987–996. [DOI] [PubMed] [Google Scholar]

[CIT0004] 4. Verhaak RGW, Hoadley KA, Purdom E, et al. ; Cancer Genome Atlas Research Network. Integrated genomic analysis identifies clinically relevant subtypes of glioblastoma characterized by abnormalities in PDGFRA, IDH1, EGFR, and NF1. Cancer Cell. 2010;17(1):98–110. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0005] 5. Segerman A, Niklasson M, Haglund C, et al. Clonal variation in drug and radiation response among glioma-initiating cells is linked to proneural-mesenchymal transition. Cell Rep. 2016;17(11):2994–3009. [DOI] [PubMed] [Google Scholar]

[CIT0006] 6. Bhat KPL, Balasubramaniyan V, Vaillant B, et al. Mesenchymal differentiation mediated by NF-κB promotes radiation resistance in glioblastoma. Cancer Cell. 2013;24(3):331–346. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0007] 7. Wang Q, Hu B, Hu X, et al. Tumor evolution of glioma-intrinsic gene expression subtypes associates with immunological changes in the microenvironment. Cancer Cell. 2017;32(1):42–56.e6. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0008] 8. Varn FS, Johnson KC, Martinek J, et al. ; GLASS Consortium. Glioma progression is shaped by genetic evolution and microenvironment interactions. Cell. 2022;185(12):2184–2199.e16. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0009] 9. Wang L, Jung J, Babikir H, et al. A single-cell atlas of glioblastoma evolution under therapy reveals cell-intrinsic and cell-extrinsic therapeutic targets. Nat. Cancer. 2022;3(12):1534–1552. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0010] 10. Azam Z, To S shun T, Tannous BA. Mesenchymal transformation: the rosetta stone of glioblastoma pathogenesis and therapy resistance. Adv Sci (Weinh). 2020;7(22):2002015. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0011] 11. Schmitt MJ, Company C, Dramaretska Y, et al. Phenotypic mapping of pathologic cross-talk between glioblastoma and innate immune cells by synthetic genetic tracing. Cancer Discov. 2021;11(3):754–777. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0012] 12. Soubannier V, Stifani S. NF- κ B signalling in glioblastoma. Biomedicines. 2017;5(2):29. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0013] 13. Koul D. PTEN signaling pathways in glioblastoma. Cancer Biol Ther. 2008;7(9):1321–1325. [DOI] [PubMed] [Google Scholar]

[CIT0014] 14. Belkina AC, Denis GV. BET domain co-regulators in obesity, inflammation and cancer. Nat Rev Cancer. 2012;12(7):465–477. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0015] 15. Huang B, Yang XD, Zhou MM, Ozato K, Chen LF. Brd4 coactivates transcriptional activation of NF-κB via specific binding to acetylated RelA. Mol Cell Biol. 2009;29(5):1375–1387. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0016] 16. Pastori C, Daniel M, Penas C, et al. BET bromodomain proteins are required for glioblastoma cell proliferation. Epigenetics. 2014;9(4):611–620. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0017] 17. Tian B, Zhao Y, Sun H, et al. BRD4 mediates NF-κB-dependent epithelial-mesenchymal transition and pulmonary fibrosis via transcriptional elongation. Am J Physiol Lung Cell Mol Physiol. 2016;311(6):L1183–L1201. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0018] 18. Serresi M, Kertalli S, Li L, et al. Functional antagonism of chromatin modulators regulates epithelial-mesenchymal transition. Sci Adv. 2021;7(9):eabd7974. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0019] 19. Engelberg IA, Foley CA, James LI, Frye SV. Improved methods for targeting epigenetic reader domains of acetylated and methylated lysine. Curr Opin Chem Biol. 2021;63:132–144. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0020] 20. Inda MM, Bonavia R, Mukasa A, et al. Tumor heterogeneity is an active process maintained by a mutant EGFR-induced cytokine circuit in glioblastoma. Genes Dev. 2010;24(16):1731–1745. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0021] 21. Benitez JA, Finlay D, Castanza A, et al. PTEN deficiency leads to proteasome addiction: a novel vulnerability in glioblastoma. Neuro-Oncology. 2021;23(7):1072–1086. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0022] 22. Zanca C, Villa GR, Benitez JA, et al. Glioblastoma cellular cross-talk converges on NF-κB to attenuate EGFR inhibitor sensitivity. Genes Dev. 2017;31(12):1212–1227. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0023] 23. Koga T, Chaim IA, Benitez JA, et al. Longitudinal assessment of tumor development using cancer avatars derived from genetically engineered pluripotent stem cells. Nat Commun. 2020;11(1):550. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0024] 24. Ge SX, Son EW, Yao R. iDEP: an integrated web application for differential expression and pathway analysis of RNA-Seq data. BMC Bioinf. 2018;19(1):534. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0025] 25. Zhou Y, Zhou B, Pache L, et al. Metascape provides a biologist-oriented resource for the analysis of systems-level datasets. Nat Commun. 2019;10(1):1523–1523. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0026] 26. Furnari FB, Lin H, Huang HS, Cavenee WK. Growth suppression of glioma cells by PTEN requires a functional phosphatase catalytic domain. Proc Natl Acad Sci USA. 1997;94(23):12479–12484. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0027] 27. Yang J, Wu Z, Renier N, et al. Pathological axonal death through a Mapk cascade that triggers a local energy deficit. Cell. 2015;160(1–2):161–176. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0028] 28. Huang B, Yang XD, Lamb A, Chen LF. Posttranslational modifications of NF-κB: another layer of regulation for NF-κB signaling pathway. Cell Signal. 2010;22(9):1282–1290. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0029] 29. Brown JD, Lin CY, Duan Q, et al. NFκB directs dynamic super enhancer formation in inflammation and atherogenesis. Mol Cell. 2014;56(2):219–231. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0030] 30. Chen L, Mu Y, Greene WC. Acetylation of RelA at discrete sites regulates distinct nuclear functions of NF-kappaB. EMBO J. 2002;21(23):6539–6548. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0031] 31. Yin W, Zhu H, Tan J, et al. Identification of collagen genes related to immune infiltration and epithelial-mesenchymal transition in glioma. Cancer Cell Int. 2021;21(1):1–18. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0032] 32. Cheng WY, Kandel JJ, Yamashiro DJ, Canoll P, Anastassiou D. A multi-cancer mesenchymal transition gene expression signature is associated with prolonged time to recurrence in glioblastoma. PLoS One. 2012;7(4):e34705. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0033] 33. Hara T, Chanoch-Myers R, Mathewson ND, et al. Interactions between cancer cells and immune cells drive transitions to mesenchymal-like states in glioblastoma. Cancer Cell. 2021;39(6):779–792.e11. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0034] 34. Gilan O, Rioja I, Knezevic K, et al. Selective targeting of BD1 and BD2 of the BET proteins in cancer and immunoinflammation. Science. 2020;368(6489):387–394. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0035] 35. Chen L, Qi Q, Jiang X, et al. Phosphocreatine promotes epigenetic reprogramming to facilitate glioblastoma growth through stabilizing BRD2. Cancer Discov. 2024;14(8):1547–1565. [DOI] [PubMed] [Google Scholar]

[CIT0036] 36. Hnilicová J, Hozeifi S, Stejskalová E, et al. The C-terminal domain of Brd2 is important for chromatin interaction and regulation of transcription and alternative splicing. Mol Biol Cell. 2013;24(22):3557–3568. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0037] 37. Trojer P. Targeting BET bromodomains in cancer. Ann Rev Cancer Biol. 2022;6(1):313–336. [Google Scholar]

[CIT0038] 38. Shorstova T, Foulkes WD, Witcher M. Achieving clinical success with BET inhibitors as anti-cancer agents. Br J Cancer. 2021;124(9):1478–1490. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0039] 39. Faivre EJ, McDaniel KF, Albert DH, et al. Selective inhibition of the BD2 bromodomain of BET proteins in prostate cancer. Nature. 2020;578(7794):306–310. [DOI] [PubMed] [Google Scholar]

[CIT0040] 40. Friedmann-morvinski D, Narasimamurthy R, Xia Y, et al. Targeting NF-kB in glioblastoma: a therapeutic approach. Sci Adv. 2016;2(1):e1501292. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0041] 41. Ballesta A, Zhou Q, Zhang X, Lv H, Gallo JM. Multiscale design of cell-type-specific pharmacokinetic/pharmacodynamic models for personalized medicine: application to temozolomide in brain tumors. CPT Pharmacometrics Syst Pharmacol. 2014;3(4):e112. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0042] 42. Ma L, Li G, Yang T, et al. An inhibitor of BRD4, GNE987, inhibits the growth of glioblastoma cells by targeting C-Myc and S100A16. Cancer Chemother Pharmacol. 2022;90(6):431–444. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0043] 43. Ma T, Hu C, Lal B, et al. Reprogramming transcription factors Oct4 and Sox2 induce a BRD-dependent immunosuppressive transcriptome in GBM-propagating cells. Cancer Res. 2021;81(9):2457–2469. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0044] 44. Tao Z, Li X, Wang H, et al. BRD4 regulates self‐renewal ability and tumorigenicity of glioma‐initiating cells by enrichment in the Notch1 promoter region. Clin Transl Med. 2020;10(6):e181. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0045] 45. Stathis A, Bertoni F. BET proteins as targets for anticancer treatment. Cancer Discov. 2018;8(1):24–36. [DOI] [PubMed] [Google Scholar]

[CIT0046] 46. Wang ZQ, Zhang ZC, Wu YY, et al. Bromodomain and extraterminal (BET) proteins: biological functions, diseases, and targeted therapy. Signal Transduct Target Ther. 2023;8(1):420. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0047] 47. Nicholson JG, Fine HA. Diffuse glioma heterogeneity and its therapeutic implications. Cancer Discov. 2021;11(3):575–590. [DOI] [PubMed] [Google Scholar]

[CIT0048] 48. Wu S, Gao F, Zheng S, et al. EGFR amplification induces increased DNA damage response and renders selective sensitivity to talazoparib (PARP inhibitor) in glioblastoma. Clin Cancer Res. 2020;26(6):1395–1407. [DOI] [PubMed] [Google Scholar]

[CIT0049] 49. Watson SS, Duc B, Kang Z, et al. Microenvironmental reorganization in brain tumors following radiotherapy and recurrence revealed by hyperplexed immunofluorescence imaging. Nat Commun. 2024;15(1):3226. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0050] 50. Jeon HM, Kim JY, Cho HJ, et al. Tissue factor is a critical regulator of radiation therapy-induced glioblastoma remodeling. Cancer Cell. 2023;41(8):1480–1497.e9. [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

BRD2 bromodomain-mediated regulation of cell state plasticity modulates therapy response in glioblastoma

Raghavendra Vadla

Brett Taylor

Yohei Miyake

Benjamin Lin

Daisuke Kawauchi

Shunichiro Miki

Nidhi Nathwani

Brandon M Jones

Yashpreet Kaur

Abhinaba Banerjee

Philip Pham

Jonathan Tsang

Albert Baldwin

David A Nathanson

Donald P Pizzo

C Ryan Miller

Frank B Furnari

Abstract

Background

Methods

Results

Conclusion

Key points.

Importance of the Study.

Materials and Methods

Cell Culture and Reagents

Protein Subcellular Fractionation and Western Blotting

Generation of CRISPR-engineered Cell Lines

Immunofluorescence

RNA-seq Data Analysis

Chromatin Immunoprecipitation-qPCR

Quantitative Real-time PCR

shRNA-mediated Knockdown

Matrigel Invasion Assay

GSC Irradiation and Treatment

Orthotopic Xenografts

Clonogenic Assay

Immunohistochemistry

Cell Growth Assay

Statistical Analyses

Results

PTEN Loss Drives MES Transition by Regulating the Chromatin Localization of RelA, BRD2, and BRD4 Proteins

Figure 1.

Acetylation of RelA at Lysine 310 Drives MES Transition in GBM

Figure 2.

RelA Lysine310 Acetylation is Essential for BRD2 Recruitment to MES Gene Promoters

BRD2 Promotes MES Transition in GBM

Figure 3.

BRD2 Bromodomains are Essential for MES Transition

Figure 4.

BET-BD2 Inhibitors Downregulate MES Gene Expression in GBM

Figure 5.

GSK620 is a Brain-Penetrant BET-BD2 Inhibitor that Extends Survival in GBM Models and Sensitizes GSCs to Therapy

Figure 6:

Discussion

Supplementary material

Acknowledgments

Contributor Information

Author contributions

Funding

Data availability

References

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases