Research ArticleCell biologyOncologyStem cells Open Access | 10.1172/jci.insight.186344

Chunying Li,1 Qiqi Xie,1 Sugata Ghosh,1,2 Bihui Cao,1 Yuanning Du,1 Giau V. Vo,3 Timothy Y. Huang,3 Charles Spruck,4 Richard L. Carpenter,1,5,6 Y. Alan Wang,7 Q. Richard Lu,8,9 Kenneth P. Nephew,1,6,10 and Jia Shen1,6,11

1Medical Sciences Program, and

2Cell, Molecular, and Cancer Biology Graduate Program, Indiana University School of Medicine, Bloomington, Indiana, USA.

3Degenerative Diseases Program, and

4Cancer Genome and Epigenetics Program, Sanford Burnham Prebys Medical Discovery Institute, La Jolla, California, USA.

5Department of Biochemistry and Molecular Biology, Indiana University School of Medicine, Indianapolis, Indiana, USA.

6Indiana University Melvin and Bren Simon Comprehensive Cancer Center, Indianapolis, Indiana, USA.

7Brown Center for Immunotherapy and Department of Medicine, Indiana University School of Medicine, Indianapolis, Indiana, USA.

8Brain Tumor Center, Division of Experimental Hematology and Cancer Biology, Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio, USA.

9Department of Pediatrics, University of Cincinnati, College of Medicine, Cincinnati, Ohio, USA.

10Department of Anatomy, Cell Biology and Physiology, and

11Department of Medical and Molecular Genetics, Indiana University School of Medicine, Indianapolis, Indiana, USA.

Address correspondence to: Jia Shen, 202 Biology Building, 1001 East 3rd St., Bloomington, Indiana, 47405, USA. Phone: 812.855.4724; Email: shen17@iu.edu.

Find articles by Li, C. in: JCI | PubMed | Google Scholar

1Medical Sciences Program, and

2Cell, Molecular, and Cancer Biology Graduate Program, Indiana University School of Medicine, Bloomington, Indiana, USA.

3Degenerative Diseases Program, and

4Cancer Genome and Epigenetics Program, Sanford Burnham Prebys Medical Discovery Institute, La Jolla, California, USA.

5Department of Biochemistry and Molecular Biology, Indiana University School of Medicine, Indianapolis, Indiana, USA.

6Indiana University Melvin and Bren Simon Comprehensive Cancer Center, Indianapolis, Indiana, USA.

7Brown Center for Immunotherapy and Department of Medicine, Indiana University School of Medicine, Indianapolis, Indiana, USA.

8Brain Tumor Center, Division of Experimental Hematology and Cancer Biology, Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio, USA.

9Department of Pediatrics, University of Cincinnati, College of Medicine, Cincinnati, Ohio, USA.

10Department of Anatomy, Cell Biology and Physiology, and

11Department of Medical and Molecular Genetics, Indiana University School of Medicine, Indianapolis, Indiana, USA.

Address correspondence to: Jia Shen, 202 Biology Building, 1001 East 3rd St., Bloomington, Indiana, 47405, USA. Phone: 812.855.4724; Email: shen17@iu.edu.

Find articles by Xie, Q. in: JCI | PubMed | Google Scholar

1Medical Sciences Program, and

2Cell, Molecular, and Cancer Biology Graduate Program, Indiana University School of Medicine, Bloomington, Indiana, USA.

3Degenerative Diseases Program, and

4Cancer Genome and Epigenetics Program, Sanford Burnham Prebys Medical Discovery Institute, La Jolla, California, USA.

5Department of Biochemistry and Molecular Biology, Indiana University School of Medicine, Indianapolis, Indiana, USA.

6Indiana University Melvin and Bren Simon Comprehensive Cancer Center, Indianapolis, Indiana, USA.

7Brown Center for Immunotherapy and Department of Medicine, Indiana University School of Medicine, Indianapolis, Indiana, USA.

8Brain Tumor Center, Division of Experimental Hematology and Cancer Biology, Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio, USA.

9Department of Pediatrics, University of Cincinnati, College of Medicine, Cincinnati, Ohio, USA.

10Department of Anatomy, Cell Biology and Physiology, and

11Department of Medical and Molecular Genetics, Indiana University School of Medicine, Indianapolis, Indiana, USA.

Address correspondence to: Jia Shen, 202 Biology Building, 1001 East 3rd St., Bloomington, Indiana, 47405, USA. Phone: 812.855.4724; Email: shen17@iu.edu.

Find articles by Ghosh, S. in: JCI | PubMed | Google Scholar

1Medical Sciences Program, and

2Cell, Molecular, and Cancer Biology Graduate Program, Indiana University School of Medicine, Bloomington, Indiana, USA.

3Degenerative Diseases Program, and

4Cancer Genome and Epigenetics Program, Sanford Burnham Prebys Medical Discovery Institute, La Jolla, California, USA.

5Department of Biochemistry and Molecular Biology, Indiana University School of Medicine, Indianapolis, Indiana, USA.

6Indiana University Melvin and Bren Simon Comprehensive Cancer Center, Indianapolis, Indiana, USA.

7Brown Center for Immunotherapy and Department of Medicine, Indiana University School of Medicine, Indianapolis, Indiana, USA.

8Brain Tumor Center, Division of Experimental Hematology and Cancer Biology, Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio, USA.

9Department of Pediatrics, University of Cincinnati, College of Medicine, Cincinnati, Ohio, USA.

10Department of Anatomy, Cell Biology and Physiology, and

11Department of Medical and Molecular Genetics, Indiana University School of Medicine, Indianapolis, Indiana, USA.

Address correspondence to: Jia Shen, 202 Biology Building, 1001 East 3rd St., Bloomington, Indiana, 47405, USA. Phone: 812.855.4724; Email: shen17@iu.edu.

Find articles by Cao, B. in: JCI | PubMed | Google Scholar

1Medical Sciences Program, and

2Cell, Molecular, and Cancer Biology Graduate Program, Indiana University School of Medicine, Bloomington, Indiana, USA.

3Degenerative Diseases Program, and

4Cancer Genome and Epigenetics Program, Sanford Burnham Prebys Medical Discovery Institute, La Jolla, California, USA.

5Department of Biochemistry and Molecular Biology, Indiana University School of Medicine, Indianapolis, Indiana, USA.

6Indiana University Melvin and Bren Simon Comprehensive Cancer Center, Indianapolis, Indiana, USA.

7Brown Center for Immunotherapy and Department of Medicine, Indiana University School of Medicine, Indianapolis, Indiana, USA.

8Brain Tumor Center, Division of Experimental Hematology and Cancer Biology, Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio, USA.

9Department of Pediatrics, University of Cincinnati, College of Medicine, Cincinnati, Ohio, USA.

10Department of Anatomy, Cell Biology and Physiology, and

11Department of Medical and Molecular Genetics, Indiana University School of Medicine, Indianapolis, Indiana, USA.

Address correspondence to: Jia Shen, 202 Biology Building, 1001 East 3rd St., Bloomington, Indiana, 47405, USA. Phone: 812.855.4724; Email: shen17@iu.edu.

Find articles by Du, Y. in: JCI | PubMed | Google Scholar

1Medical Sciences Program, and

2Cell, Molecular, and Cancer Biology Graduate Program, Indiana University School of Medicine, Bloomington, Indiana, USA.

3Degenerative Diseases Program, and

4Cancer Genome and Epigenetics Program, Sanford Burnham Prebys Medical Discovery Institute, La Jolla, California, USA.

5Department of Biochemistry and Molecular Biology, Indiana University School of Medicine, Indianapolis, Indiana, USA.

6Indiana University Melvin and Bren Simon Comprehensive Cancer Center, Indianapolis, Indiana, USA.

7Brown Center for Immunotherapy and Department of Medicine, Indiana University School of Medicine, Indianapolis, Indiana, USA.

8Brain Tumor Center, Division of Experimental Hematology and Cancer Biology, Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio, USA.

9Department of Pediatrics, University of Cincinnati, College of Medicine, Cincinnati, Ohio, USA.

10Department of Anatomy, Cell Biology and Physiology, and

11Department of Medical and Molecular Genetics, Indiana University School of Medicine, Indianapolis, Indiana, USA.

Address correspondence to: Jia Shen, 202 Biology Building, 1001 East 3rd St., Bloomington, Indiana, 47405, USA. Phone: 812.855.4724; Email: shen17@iu.edu.

Find articles by Vo, G. in: JCI | PubMed | Google Scholar

1Medical Sciences Program, and

2Cell, Molecular, and Cancer Biology Graduate Program, Indiana University School of Medicine, Bloomington, Indiana, USA.

3Degenerative Diseases Program, and

4Cancer Genome and Epigenetics Program, Sanford Burnham Prebys Medical Discovery Institute, La Jolla, California, USA.

5Department of Biochemistry and Molecular Biology, Indiana University School of Medicine, Indianapolis, Indiana, USA.

6Indiana University Melvin and Bren Simon Comprehensive Cancer Center, Indianapolis, Indiana, USA.

7Brown Center for Immunotherapy and Department of Medicine, Indiana University School of Medicine, Indianapolis, Indiana, USA.

8Brain Tumor Center, Division of Experimental Hematology and Cancer Biology, Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio, USA.

9Department of Pediatrics, University of Cincinnati, College of Medicine, Cincinnati, Ohio, USA.

10Department of Anatomy, Cell Biology and Physiology, and

11Department of Medical and Molecular Genetics, Indiana University School of Medicine, Indianapolis, Indiana, USA.

Address correspondence to: Jia Shen, 202 Biology Building, 1001 East 3rd St., Bloomington, Indiana, 47405, USA. Phone: 812.855.4724; Email: shen17@iu.edu.

Find articles by Huang, T. in: JCI | PubMed | Google Scholar

1Medical Sciences Program, and

2Cell, Molecular, and Cancer Biology Graduate Program, Indiana University School of Medicine, Bloomington, Indiana, USA.

3Degenerative Diseases Program, and

4Cancer Genome and Epigenetics Program, Sanford Burnham Prebys Medical Discovery Institute, La Jolla, California, USA.

5Department of Biochemistry and Molecular Biology, Indiana University School of Medicine, Indianapolis, Indiana, USA.

6Indiana University Melvin and Bren Simon Comprehensive Cancer Center, Indianapolis, Indiana, USA.

7Brown Center for Immunotherapy and Department of Medicine, Indiana University School of Medicine, Indianapolis, Indiana, USA.

8Brain Tumor Center, Division of Experimental Hematology and Cancer Biology, Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio, USA.

9Department of Pediatrics, University of Cincinnati, College of Medicine, Cincinnati, Ohio, USA.

10Department of Anatomy, Cell Biology and Physiology, and

11Department of Medical and Molecular Genetics, Indiana University School of Medicine, Indianapolis, Indiana, USA.

Address correspondence to: Jia Shen, 202 Biology Building, 1001 East 3rd St., Bloomington, Indiana, 47405, USA. Phone: 812.855.4724; Email: shen17@iu.edu.

Find articles by Spruck, C. in: JCI | PubMed | Google Scholar

1Medical Sciences Program, and

2Cell, Molecular, and Cancer Biology Graduate Program, Indiana University School of Medicine, Bloomington, Indiana, USA.

3Degenerative Diseases Program, and

4Cancer Genome and Epigenetics Program, Sanford Burnham Prebys Medical Discovery Institute, La Jolla, California, USA.

5Department of Biochemistry and Molecular Biology, Indiana University School of Medicine, Indianapolis, Indiana, USA.

6Indiana University Melvin and Bren Simon Comprehensive Cancer Center, Indianapolis, Indiana, USA.

7Brown Center for Immunotherapy and Department of Medicine, Indiana University School of Medicine, Indianapolis, Indiana, USA.

8Brain Tumor Center, Division of Experimental Hematology and Cancer Biology, Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio, USA.

9Department of Pediatrics, University of Cincinnati, College of Medicine, Cincinnati, Ohio, USA.

10Department of Anatomy, Cell Biology and Physiology, and

11Department of Medical and Molecular Genetics, Indiana University School of Medicine, Indianapolis, Indiana, USA.

Address correspondence to: Jia Shen, 202 Biology Building, 1001 East 3rd St., Bloomington, Indiana, 47405, USA. Phone: 812.855.4724; Email: shen17@iu.edu.

Find articles by Carpenter, R. in: JCI | PubMed | Google Scholar

1Medical Sciences Program, and

2Cell, Molecular, and Cancer Biology Graduate Program, Indiana University School of Medicine, Bloomington, Indiana, USA.

3Degenerative Diseases Program, and

4Cancer Genome and Epigenetics Program, Sanford Burnham Prebys Medical Discovery Institute, La Jolla, California, USA.

5Department of Biochemistry and Molecular Biology, Indiana University School of Medicine, Indianapolis, Indiana, USA.

6Indiana University Melvin and Bren Simon Comprehensive Cancer Center, Indianapolis, Indiana, USA.

7Brown Center for Immunotherapy and Department of Medicine, Indiana University School of Medicine, Indianapolis, Indiana, USA.

8Brain Tumor Center, Division of Experimental Hematology and Cancer Biology, Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio, USA.

9Department of Pediatrics, University of Cincinnati, College of Medicine, Cincinnati, Ohio, USA.

10Department of Anatomy, Cell Biology and Physiology, and

11Department of Medical and Molecular Genetics, Indiana University School of Medicine, Indianapolis, Indiana, USA.

Address correspondence to: Jia Shen, 202 Biology Building, 1001 East 3rd St., Bloomington, Indiana, 47405, USA. Phone: 812.855.4724; Email: shen17@iu.edu.

Find articles by Wang, Y. in: JCI | PubMed | Google Scholar

1Medical Sciences Program, and

2Cell, Molecular, and Cancer Biology Graduate Program, Indiana University School of Medicine, Bloomington, Indiana, USA.

3Degenerative Diseases Program, and

4Cancer Genome and Epigenetics Program, Sanford Burnham Prebys Medical Discovery Institute, La Jolla, California, USA.

5Department of Biochemistry and Molecular Biology, Indiana University School of Medicine, Indianapolis, Indiana, USA.

6Indiana University Melvin and Bren Simon Comprehensive Cancer Center, Indianapolis, Indiana, USA.

7Brown Center for Immunotherapy and Department of Medicine, Indiana University School of Medicine, Indianapolis, Indiana, USA.

8Brain Tumor Center, Division of Experimental Hematology and Cancer Biology, Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio, USA.

9Department of Pediatrics, University of Cincinnati, College of Medicine, Cincinnati, Ohio, USA.

10Department of Anatomy, Cell Biology and Physiology, and

11Department of Medical and Molecular Genetics, Indiana University School of Medicine, Indianapolis, Indiana, USA.

Address correspondence to: Jia Shen, 202 Biology Building, 1001 East 3rd St., Bloomington, Indiana, 47405, USA. Phone: 812.855.4724; Email: shen17@iu.edu.

Find articles by Lu, Q. in: JCI | PubMed | Google Scholar

1Medical Sciences Program, and

2Cell, Molecular, and Cancer Biology Graduate Program, Indiana University School of Medicine, Bloomington, Indiana, USA.

3Degenerative Diseases Program, and

4Cancer Genome and Epigenetics Program, Sanford Burnham Prebys Medical Discovery Institute, La Jolla, California, USA.

5Department of Biochemistry and Molecular Biology, Indiana University School of Medicine, Indianapolis, Indiana, USA.

6Indiana University Melvin and Bren Simon Comprehensive Cancer Center, Indianapolis, Indiana, USA.

7Brown Center for Immunotherapy and Department of Medicine, Indiana University School of Medicine, Indianapolis, Indiana, USA.

8Brain Tumor Center, Division of Experimental Hematology and Cancer Biology, Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio, USA.

9Department of Pediatrics, University of Cincinnati, College of Medicine, Cincinnati, Ohio, USA.

10Department of Anatomy, Cell Biology and Physiology, and

11Department of Medical and Molecular Genetics, Indiana University School of Medicine, Indianapolis, Indiana, USA.

Address correspondence to: Jia Shen, 202 Biology Building, 1001 East 3rd St., Bloomington, Indiana, 47405, USA. Phone: 812.855.4724; Email: shen17@iu.edu.

Find articles by Nephew, K. in: JCI | PubMed | Google Scholar |

1Medical Sciences Program, and

2Cell, Molecular, and Cancer Biology Graduate Program, Indiana University School of Medicine, Bloomington, Indiana, USA.

3Degenerative Diseases Program, and

4Cancer Genome and Epigenetics Program, Sanford Burnham Prebys Medical Discovery Institute, La Jolla, California, USA.

5Department of Biochemistry and Molecular Biology, Indiana University School of Medicine, Indianapolis, Indiana, USA.

6Indiana University Melvin and Bren Simon Comprehensive Cancer Center, Indianapolis, Indiana, USA.

7Brown Center for Immunotherapy and Department of Medicine, Indiana University School of Medicine, Indianapolis, Indiana, USA.

8Brain Tumor Center, Division of Experimental Hematology and Cancer Biology, Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio, USA.

9Department of Pediatrics, University of Cincinnati, College of Medicine, Cincinnati, Ohio, USA.

10Department of Anatomy, Cell Biology and Physiology, and

11Department of Medical and Molecular Genetics, Indiana University School of Medicine, Indianapolis, Indiana, USA.

Address correspondence to: Jia Shen, 202 Biology Building, 1001 East 3rd St., Bloomington, Indiana, 47405, USA. Phone: 812.855.4724; Email: shen17@iu.edu.

Find articles by Shen, J. in: JCI | PubMed | Google Scholar

Published March 10, 2025 - More info

Published in Volume 10, Issue 5 on March 10, 2025
JCI Insight. 2025;10(5):e186344. https://doi.org/10.1172/jci.insight.186344.
© 2025 Li et al. This work is licensed under the Creative Commons Attribution 4.0 International License. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/. Published March 10, 2025 - Version history
Received: August 26, 2024; Accepted: January 24, 2025 View PDF Abstract

Glioblastoma (GBM) is the most lethal brain cancer, with GBM stem cells (GSCs) driving therapeutic resistance and recurrence. Targeting GSCs offers a promising strategy for preventing tumor relapse and improving outcomes. We identify SUV39H1, a histone-3, lysine-9 methyltransferase, as critical for GSC maintenance and GBM progression. SUV39H1 is upregulated in GBM compared with normal brain tissues, with single-cell RNA-seq showing its expression predominantly in GSCs due to super-enhancer–mediated activation. Knockdown of SUV39H1 in GSCs impaired their proliferation and stemness. Whole-cell RNA-seq analysis revealed that SUV39H1 regulates G2/M cell cycle progression, stem cell maintenance, and cell death pathways in GSCs. By integrating the RNA-seq data with ATAC-seq data, we further demonstrated that knockdown of SUV39H1 altered chromatin accessibility in key genes associated with these pathways. Chaetocin, an SUV39H1 inhibitor, mimics the effects of SUV39H1 knockdown, reducing GSC stemness and sensitizing cells to temozolomide, a standard GBM chemotherapy. In a patient-derived xenograft model, targeting SUV39H1 inhibits GSC-driven tumor growth. Clinically, high SUV39H1 expression correlates with poor glioma prognosis, supporting its relevance as a therapeutic target. This study identifies SUV39H1 as a crucial regulator of GSC maintenance and a promising therapeutic target to improve GBM treatment and patient outcomes.

Introduction

Glioblastoma (GBM) is the most common primary malignant brain tumor in adults, known for its aggressiveness and lethality (1). Despite standard treatment that includes a combination of surgery, radiotherapy, and/or temozolomide (TMZ), the prognosis for patients with GBM remains poor, with a median survival of less than 16 months (2, 3). There is an urgent need for more effective treatments to improve patient survival.

The resistance of GBM to standard treatment is partly attributed to intratumoral heterogeneity driven by GBM stem cells (GSCs), which possess self-renewal and differentiation properties, and exhibit strong tumorigenic potential (4–6). GSCs can differentiate into various cell types within GBM, such as endothelial cells, pericytes, and non-stem GBM cells (NSGCs). As GSCs can initiate and propagate tumors, resist standard therapies including TMZ, repopulate tumors after treatment, and contribute to disease relapse, eliminating GSCs could overcome chemoresistance and represent an effective therapeutic strategy in GBM.

Epigenetic modifications, including DNA methylation and histone modifications, profoundly impact gene expression and cellular behavior in normal and cancer cells, including stem cells (7, 8). Targeting epigenetic regulators in GSCs has the potential to reverse aberrant gene expression patterns, disrupt stemness, and sensitize GSCs to existing therapies, offering new avenues for effective GBM treatment. Previous studies have identified several epigenetic regulators involved in maintaining GSC stemness and therapeutic resistance, including PRMT6 (9), PRC2 (10, 11), KDM2B (12), DNMT1 (13), HDACs (14, 15), and HELLS (16). SUV39H1 is a key epigenetic regulator and histone methyltransferase responsible for trimethylating histone H3 at lysine 9 (H3K9me3) (17, 18), which promotes heterochromatin formation and transcriptional repression. However, the specific roles of SUV39H1 in GSCs and GBM remain to be determined.

To address this gap, we employed an integrative approach, combining in vitro and in vivo experiments with bioinformatics analyses, to investigate the functional importance of SUV39H1 in GSCs and GBM progression.

Results

SUV39H1 is upregulated in GBM. We first investigated the expression of SUV39H1 in GBM using RNA-seq data from GBM datasets (The Cancer Genome Atlas [TCGA], Murat, Kamoun, and Rembrandt). Compared with nontumor brain tissues, SUV39H1 expression was significantly increased in GBM tissues (P < 0.05; Figure 1, A–D). Overexpression of SUV39H1 in GBM tissues was confirmed by immunohistochemistry (IHC) (P < 0.001; Figure 1E). These findings demonstrate that SUV39H1 is upregulated in GBM compared with normal brain tissues.

Figure 1

SUV39H1 is upregulated in GBM. (A–D) Violin plots showing SUV39H1 expression in nontumor and GBM tissues across 4 datasets: TCGA_GTEx_GBM (A), Murat_GBM (B), Kamoun_GBM (C), and Rembrandt_GBM (D). (E) H&E staining (top row) and IHC staining of SUV39H1 (bottom row) in normal (n = 4) and GBM (n = 9) tissues. Quantification of the number of atypical cells and the percentage of SUV39H1-positive cells is shown on the right. Scale bars: 100 μm (left and middle panels) and 30 μm (high-magnification insets, right panel). Data represent mean ± SD. ***P < 0.001, ****P < 0.0001 by unpaired, 2-tailed t test.

SUV39H1 is preferentially expressed in GSCs. To examine the expression pattern of SUV39H1 in GBM, single-cell RNA-seq analysis was performed. Uniform manifold approximation and projection (UMAP) clustering identified distinct cell types in GBM (Figure 2A and Supplemental Figure 1; supplemental material available online with this article; https://doi.org/10.1172/jci.insight.186344DS1). We used stemness markers, including OLIG2, NES, and SOX2, to further analyze the malignant cells and distinguish GSCs from NSGCs (Figure 2B). The data revealed that SUV39H1 was preferentially overexpressed in GSCs (Figure 2, C and D). Immunofluorescent staining confirmed the colocalization of SUV39H1 with OLIG2 and SOX2 in GBM patient tissues (Figure 2E and Supplemental Figure 2).

Figure 2

Expression patterns of SUV39H1 in GBM. (A) The UMAP clustering of single-cell RNA-seq data from GBM tumors sourced from the CELLxGENE database reveals various cell types. (B) UMAP plot distinguishing GSCs (orange) from non-stem GBM cells (NSGCs, blue). (C) Volcano plot displaying DEGs between GSCs and NSGCs. Genes with significant expression changes (|log2 fold change| > 2 and adjusted P value < 0.05) are shown as red dots, with the top genes (OLIG2, NES, SOX2, and SUV39H1) annotated in the box. (D) UMAP plots showing the expression patterns of the indicated genes in GSCs and NSGCs. The color intensity represents the normalized expression level. (E) Representative images (left panel) and quantification (right panel) of immunofluorescent staining showing colocalization of SUV39H1 (green) and OLIG2 (red) in GBM tissues (n = 3). Scale bars: 50 μm.

Analysis of RNA-seq data for 3 pairs of GSCs and their matched differentiated NSGCs (NCBI Gene Expression Omnibus [GEO] GSE54791) revealed higher SUV39H1 expression in GSCs (P < 0.01; Figure 3A). Using GSC3565 and GSC1914 models for differentiated and undifferentiated cell states (Figure 3B), we found that SUV39H1 levels decreased during serum-induced differentiation, accompanied by downregulation of OLIG2 and upregulation of GFAP, a differentiated-NSGC marker (quantitative PCR [qPCR], Figure 3C; Western blot, Figure 3D).

Figure 3

SUV39H1 expression in GSCs, NSGCs, and normal brain cells. (A) Analysis of RNA-seq data (GSE54791) for SUV39H1 expression in multiple GSC and NSGC pairs (MCG4, MCG6, MCG8). CPM, counts per million. (B–D) Representative images (B), qPCR data (C), and Western blotting data and quantification (D) for GSC3565 and GSC1914 differentiation by serum induction. OLIG2 is a GSC marker, and GFAP is a differentiated NSGC marker. (E) H3K27ac ChIP-seq data showing the SUV39H1 locus in indicated GSCs and NSGCs. (F) Violin plot showing SUV39H1 expression levels in GSCs compared to normal neural stem cells (NSCs). P value was calculated using an unpaired, 2-tailed t test. (G) H3K27ac ChIP-seq signal tracks showing the SUV39H1 locus in indicated GSCs and NSCs. (H) Representative images and qPCR data for GSC3565 cells treated with JQ1 for 48 hours. Scale bars: 100 μm. Data represent mean ± SD. *P < 0.05; **P < 0.01; ****P < 0.0001 by unpaired, 2-tailed t test.

ChIP-seq data of H3K27ac, a histone modification linked to active super-enhancers that boost gene transcription, revealed enriched peaks at the SUV39H1 gene in GSCs compared with differentiated NSGCs, indicating enhanced regulatory activity of SUV39H1 expression in GSCs (Figure 3E). Additionally, RNA-seq analysis of 44 GSC models and 9 normal brain cell lines (GSE119834) demonstrated preferential SUV39H1 expression in GSCs (P = 0.004; Figure 3F). Consistently, ChIP-seq signal tracks for H3K27ac showed increased super-enhancer marks at the SUV39H1 gene in GSCs versus normal brain cells (Figure 3G). To confirm the role of GSC-specific super-enhancer regulation of SUV39H1 expression, we treated GSCs with JQ1, a bromodomain-containing 4 (BRD4) inhibitor known to disrupt super-enhancer function (19). This treatment led to a dose-dependent reduction in SUV39H1 mRNA levels (Figure 3H), demonstrating that SUV39H1 expression in GSCs is indeed super-enhancer driven. We further examined available ChIP-seq data for OLIG2 and SOX2 in GSCs (20) and identified binding peaks for both transcription factors at the SUV39H1 locus, overlapping partially with H3K27ac peaks (Supplemental Figure 3A), suggesting their potential role in regulating SUV39H1 expression. Knockdown (KD) studies showed that silencing SOX2 led to a reduction in SUV39H1 mRNA expression (Supplemental Figure 3, B and C), and OLIG2 KD similarly decreased SUV39H1 mRNA levels (Supplemental Figure 3, D and E). Furthermore, SUV39H1 expression was positively correlated with SOX2 and OLIG2 expression in GBM samples from the TCGA (Supplemental Figure 3F) and Chinese Glioma Genome Atlas (CGGA) databases (Supplemental Figure 3G). Together, these findings indicate that SOX2 and OLIG2 may promote SUV39H1 expression through their interaction with super-enhancers. Further studies are needed to clarify the detailed mechanism.

SUV39H1 is required for GSC maintenance. SUV39H1 upregulation in GSCs suggests a potential dependency on this enzyme. To investigate the functional roles of SUV39H1 in GSCs, we first knocked down SUV39H1 in GSC3565 and GSC1914 cells using 2 different shRNAs and confirmed KD efficiency by Western blotting (Figure 4A). SUV39H1 KD led to decreased GSC proliferation (Figure 4B), which was further validated by a 5-ethynyl-2′-deoxyuridine (EdU) incorporation assay showing slower DNA replication in GSCs with SUV39H1 KD (Figure 4, C and D). In NSGCs differentiated from GSCs (Supplemental Figure 4, A–C) and in U118 (Supplemental Figure 4, D–F), a human GBM cell line, SUV39H1 KD also reduced proliferation and survival, suggesting that both GSCs and NSGCs may depend on SUV39H1 for these functions. Our prior research demonstrated that targeting SUV39H1 had minimal impact on nonneoplastic cells, such as astrocytes and human mammary epithelial cells, in contrast with its significant effects in GSCs (21). Consistent with this, recent research also revealed that SUV39H1 KD did not affect the proliferation of astrocytes or human neural stem cells (22). Furthermore, SUV39H1 KD impaired the self-renewal ability of GSCs. Tumorsphere formation assay showed a significant reduction in the number of tumorspheres formed by SUV39H1-KD cells (P < 0.05; Figure 4, E and F). Extreme limiting dilution assay (ELDA) quantitatively demonstrated a decreased stem cell frequency in SUV39H1-KD cells (P < 0.05; Figure 4G). Notably, in GSC3565 and GSC1914 cells, shSUV39H1-2 achieved higher KD efficiency than shSUV39H1-1, correlating with greater reductions in proliferation (Figure 4B) and stemness (Figure 4, F and G). This dose-dependent relationship underscores the critical role of SUV39H1 in these processes. In addition to using GSC3565 (undefined GBM subtype) and GSC1914 (classical subtype), we assessed SUV39H1 KD in GSCs from other GBM subtypes, including GSC23 (proneural subtype) and GSC839 (mesenchymal subtype). Consistent with previous findings, SUV39H1 KD reduced tumorsphere growth and viability across these subtypes (Supplemental Figure 4, G–L).

Figure 4

SUV39H1 regulates GSC proliferation and stemness. (A) Western blot data (left panel) and quantification (right panel) showing the levels of SUV39H1 and OLIG2 in control (Ctrl) and SUV39H1-KD GSCs. (B) Cell viability analysis for GSC3565 (day 5) and GSC1914 (day 8) following gene KD. Two-way ANOVA and Dunnett’s multiple-comparison test. (C and D) Representative images (C) and quantification (D) of EdU incorporation assay in GSCs. Unpaired, 2-tailed t test. Scale bars: 50 μm. (E and F) Representative images (E) and quantification (F) of tumorsphere formation in GSCs after 72 hours of gene KD. Unpaired, 2-tailed t test. Scale bars: 100 μm. (G) Limiting dilution assay demonstrating the self-renewal capacity of GSCs with various cell numbers after 13 days of gene KD. Data points represent the log fraction of wells without spheres plotted against the number of cells plated per well. Pairwise test. Data represent mean ± SD. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001.

To further examine the effect of inhibiting SUV39H1, GSCs were treated with chaetocin, an inhibitor of SUV39H1 (22, 23). qPCR analysis demonstrated decreased expression of SUV39H1 and OLIG2 in GSCs treated with chaetocin compared with untreated controls (Figure 5A). Tumorsphere formation assay revealed a significant reduction in the number of tumorspheres formed in chaetocin-treated GSCs (P < 0.05; Figure 5, B and C). Notably, chaetocin treatment sensitized both GSC3565 and GSC1914 cells to the GBM chemotherapy drug TMZ, with synergy scores of 10.148 and 16.086, respectively (Figure 5, D and E). These findings demonstrate that SUV39H1 is essential for GSC maintenance.

Figure 5

Chaetocin treatment disrupts GSCs and synergizes with TMZ. (A) qPCR analysis of SUV39H1 and OLIG2 expression in GSCs treated with 0 nM or 50 nM chaetocin. (B and C) Representative images (B) and quantification (C) of tumorsphere formation in GSCs after 48 hours of chaetocin treatment. Unpaired, 2-tailed t test. Scale bars: 100 μm. (D and E) 3D synergy score plots showing the synergistic effect of chaetocin and TMZ on killing GSC3565 (D) and GSC1914 (E). The highest single agent (HSA) synergy score assesses the combination’s efficacy relative to the most effective single agent involved in the combination. An HSA score of greater than 10 is considered indicative of synergy. Data represent mean ± SD. *P < 0.05, **P < 0.01.

SUV39H1 regulates cell cycle, stemness, and cell death pathways in GSCs. To explore the molecular mechanisms by which SUV39H1 maintains GSCs, we performed RNA-seq analysis on GSC3565 and GSC1914 cells with SUV39H1 KD versus control KD (Figure 6A). Gene set enrichment analysis (GSEA) identified several enriched pathways associated with the differentially expressed genes (DEGs) (Figure 6B). SUV39H1-KD GSCs exhibited downregulation of G2/M cell cycle pathways (Figure 6C). Consistently, qPCR detection of G2/M cell cycle–related genes, including CDK16, CDC27, and CUL3, showed decreased expression upon SUV39H1 KD (Figure 6D). Flow cytometry analysis revealed an enrichment of the G2/M phase cell population in SUV39H1-KD GSCs (Figure 6E). A similar increase in the G2/M phase population was also observed in NSGCs differentiated from GSCs and in U118 cells with SUV39H1 KD (Supplemental Figure 5A). GSEA also identified decreased stem cell–related pathways in SUV39H1-KD cells (Figure 6F), which was validated by qPCR showing downregulation of stemness genes, including OLIG2, NES, and MYC (Figure 6G), and by immunofluorescence for OLIG2 (Figure 6H). Additionally, cell death pathways were upregulated, with GSEA indicating enrichment of autophagy, ferroptosis, and pyroptosis pathways (Supplemental Figure 5B), which were validated by qPCR for selected genes (Supplemental Figure 5C). These results demonstrate that SUV39H1 regulates cell cycle, stemness, and cell death pathways in GSCs.

Figure 6

Signaling pathways regulated by SUV39H1 in GSCs. (A) Volcano plot illustrating DEGs in SUV39H1-KD versus control KD GSCs, with key genes highlighted. (B) Enrichment map visualizing the significant pathways affected by SUV39H1 KD. (C) GSEA plot showing enrichment of the G2/M cell cycle–related pathways in GSCs with SUV39H1 KD. P.adjust values indicate the significance of enrichment, which was assessed using the Kolmogorov-Smirnov test. Multiple hypothesis testing correction was applied using the Benjamini-Hochberg method to control the false discovery rate (FDR). (D) qPCR detection of G2/M cell cycle–related genes in GSC3565 and GSC1914 cells. (E) Flow cytometry data showing cell cycle change in GSCs with SUV39H1 KD. (F) GSEA plot showing enrichment of the stem cell–related pathways in SUV39H1-KD GSCs. (G) qPCR detection of stem cell–related genes in GSCs. (H) Immunofluorescent staining for OLIG2 (green) and DAPI (blue) in GSC3565 and GSC1914 cells. Scale bars: 50 μm.

Targeting SUV39H1 alters chromatin accessibility. To investigate how targeting SUV39H1 affects gene expression, we performed an assay for transposase-accessible chromatin followed by sequencing (ATAC-seq), revealing distinct chromatin accessibility profiles between untreated and chaetocin-treated GSCs (Figure 7, A and B). These differentially accessible chromatin regions were distributed across various genomic features (Figure 7C), and heatmap visualization highlighted a significant population of genes associated with these regions, which we named ATAC-seq differential genes (ATAC-seq-DGs) (Figure 7D). Integrating ATAC-seq-DGs with RNA-seq DEGs (Figure 6A) identified overlapping genes (n = 2823, 8.4%) (Figure 7E), with many linked to G2/M cell cycle and stem cell pathways (Figure 7F). For example, CDC27 and CDC6 G2/M cell cycle (Figure 7G) and OLIG2 and NES stemness genes (Figure 7H) showed decreased chromatin accessibility and gene expression upon SUV39H1 targeting, as visualized in the ATAC-seq and RNA-seq signal tracks. These data suggest that targeting SUV39H1 reduces chromatin accessibility at these genes, resulting in downregulation of their expression, contributing to G2/M cell cycle arrest and disruption of stemness.

Figure 7

Targeting SUV39H1 alters chromatin accessibility in GSCs. (A) Principal component analysis (PCA) plot displaying the clustering of untreated (blue) and chaetocin-treated (pink) GSCs. (B) Aggregate plots (top) and heatmaps (bottom) of ATAC-seq signals at transcription start sites (TSS) in untreated and chaetocin-treated GSCs. (C) Pie chart illustrating the distribution of differentially accessible chromatin regions across various genomic features in response to chaetocin treatment. (D) Heatmap demonstrating the genes associated with altered chromatin accessibility upon SUV39H1 targeting. (E) Venn diagram showing the overlap of genes (n = 2823, 8.4%) between RNA-seq DEGs (blue) and genes associated with ATAC-seq differentially accessible regions (pink) after chaetocin treatment. (F) Bubble plot of GO enrichment analysis for the overlapping genes identified in E. (G and H) RNA-seq and ATAC-seq signals for indicated genes related to G2/M cell cycle (G) and stem cell maintenance (H).

Targeting SUV39H1 decreases GSC-driven GBM growth in mice. The impact of targeting SUV39H1 on the in vivo tumor formation ability of GSCs was assessed using a xenograft mouse model. Control or SUV39H1-KD GSC3565 cells expressing luciferase were intracranially injected into the brains of immunodeficient mice (Figure 8, A and B). SUV39H1-KD GSC–derived tumors displayed significantly reduced growth in mice on day 29 (Figure 8, C and D). IHC analysis revealed decreased