Abstract
Glioblastoma (GBM), the most aggressive tumor in the adult central nervous system, remains a major therapeutic challenge due to its high recurrence and resistance to conventional therapies. Recent evidence underscores the pivotal role of glioma stem cells (GSCs) in driving these malignant features. In this study, using intracranial xenograft models established in 4-week-old male BALB/c nude mice and patient-derived primary GSCs, we uncover a critical function of the chromatin assembly factor subunit Chaf1b in sustaining the stemness of GSCs and modulating the tumor immune microenvironment. We show that Chaf1b is markedly overexpressed in high-grade gliomas and GSC populations. Genetic silencing of Chaf1b led to a significant reduction in GSC self-renewal capacity and tumorigenicity, both in vitro and in intracranial xenograft models. Mechanistically, Chaf1b was found to upregulate IL-33 secretion, thereby promoting microglial M2 polarization and activating the PI3K/AKT signaling pathway—effects that were reversible upon IL-33 neutralization. These findings position Chaf1b as a key mediator of GBM aggressiveness and suggest it as a promising target for disrupting the stem-immune axis in GBM.
Significance Statement
This study reveals the critical role of the chromatin assembly factor Chaf1b in glioma stem cells (GSCs). We found that Chaf1b is significantly upregulated in subventricular zone–positive glioblastoma (GBM) patients and in their derived GSCs. Functional experiments demonstrated that Chaf1b knockdown markedly inhibits GSC proliferation, self-renewal, and tumorigenicity in vivo. Mechanistically, Chaf1b induces IL-33 secretion from GSCs, promotes microglial polarization toward the immunosuppressive M2 phenotype, and activates the PI3K/AKT signaling pathway. These processes collectively reshape the immune microenvironment, enhance GSC stemness, and drive GBM progression. This study systematically elucidates the pivotal role of Chaf1b in stemness maintenance and immune modulation of GSCs, highlighting its potential as a therapeutic target for GBM.
Introduction
Gliomas are the most common malignant tumors of the central nervous system, accounting for ∼80% of primary malignant brain tumors in adults (Bhuvanalakshmi et al., 2018; Miller et al., 2021; Siegel et al., 2023). Among them, glioblastoma (GBM) is the most aggressive subtype, with a median overall survival of ∼15 months (Wang and Jiang, 2013). Therefore, investigating the underlying mechanisms of gliomagenesis and identifying novel therapeutic targets are crucial for improving patient prognosis and enhancing long-term survival outcomes.
The tumor immune microenvironment is a central driver of GBM malignancy and a major contributor to its therapeutic resistance and recurrence (Wang et al., 2022). Moreover, microglia—the principal immune cells of the central nervous system—play a pivotal role in remodeling the immune microenvironment of GBM. As the brain's resident macrophages, microglia can sense a wide range of microenvironmental cues and undergo polarization into either proinflammatory (M1) or anti-inflammatory (M2) phenotypes, thereby exerting immunomodulatory effects that are either antitumorigenic or protumorigenic, respectively (Lin et al., 2024). In the subventricular zone (SVZ)—a critical region for GBM initiation and progression—microglia confer a survival advantage to the tumor through multiple mechanisms, including supporting glioma stem cells (GSCs) viability, modulating the immune microenvironment, maintaining tumor heterogeneity, and influencing therapeutic responses. These functions establish microglia as indispensable immune regulators within the GBM ecosystem (Munro et al., 2024).
As a reservoir of neural stem cells within the central nervous system, the SVZ provides critical support for the origin and maintenance of stemness in GSCs (Lee et al., 2018; Epstein et al., 2024). Previous studies have confirmed that the anatomical location of GBM significantly influences patient prognosis. Tumors in contact with the lateral ventricles exhibit increased expression of stemness-related genes, indicative of a functionally distinct immuno-oncological microenvironment (Steed et al., 2020; Bartkowiak et al., 2023). Such tumors are associated with higher rates of distant recurrence and reduced median overall survival, independent of patient age and extent of resection—features likely linked to the presence of the SVZ (Lim et al., 2007; Nestler et al., 2015; Mistry et al., 2017). GSCs represent a subpopulation of glioma cells characterized by self-renewal, multipotency, and high invasiveness. They play a pivotal role in gliomagenesis, tumor progression, recurrence, and therapeutic resistance (Lathia et al., 2015; Gimple et al., 2019). By sustaining stem-like features, GSCs contribute to intratumoral heterogeneity and serve as a major reservoir for relapse (Chen et al., 2012). In addition, they exhibit remarkable resistance to radiotherapy and chemotherapy through enhanced DNA repair capacity, apoptosis evasion, and activation of drug efflux mechanisms (Bao et al., 2006; Gao et al., 2022; Xie et al., 2022; Ranjan et al., 2023; Wu et al., 2023; Fan et al., 2024). More importantly, GSCs actively reshape the tumor immune microenvironment, inducing immune evasion and promoting tumor aggressiveness, thereby rendering gliomas more refractory and difficult to control (Gangoso et al., 2021).
Given the central role of GSCs in tumor growth and recurrence, targeting GSCs has emerged as a critical therapeutic strategy to enhance treatment efficacy and reduce relapse rates. Comparative proteomic analysis of GBMs with versus without SVZ involvement identified four prognosis-associated candidates (Chaf1b, DNAJC30, C2CD4C, and BPNT1), among which Chaf1b emerged as the lead candidate and was prioritized for subsequent mechanistic investigation. In this context, we focused on a key molecule, chromatin assembly factor 1 subunit B (Chaf1b), a core component of the chromatin assembly factor-1 (CAF-1) complex. Chaf1b participates in nucleosome assembly during DNA replication and in chromatin restoration following DNA damage repair. Its expression is specifically upregulated during the S phase, thereby contributing to the coordinated regulation of cell proliferation, genomic stability, and epigenetic inheritance (Volk and Crispino, 2015; Chen et al., 2023). Recent investigations have revealed that Chaf1b is highly expressed across multiple tumor types and is frequently associated with poor prognosis. Its upregulation has been closely linked to radioresistance, therapeutic tolerance, and enhanced tumor invasiveness (Peng et al., 2018; Di et al., 2020; Dean et al., 2023; Saleiro et al., 2023). Although the role of Chaf1b in GSCs remains uncharacterized, our study demonstrates that Chaf1b promotes GSC stemness by sustaining self-renewal and preserving an undifferentiated state. Moreover, Chaf1b drives microglial polarization toward an immunosuppressive M2 phenotype, thereby reprogramming the immune microenvironment, facilitating immune evasion, and ultimately fueling tumor progression.
Materials and Methods
RNA extraction and qRT–PCR
Total RNA was isolated from GBM tissues with (SVZ+) or without (SVZ−) SVZ involvement using the M5 Universal RNA Mini Kit (Mei5bio) in accordance with the manufacturer's instructions. Quantitative reverse transcription PCR (qRT–PCR) was performed with the riboSCRIPT mRNA/lncRNA qRT–PCR Starter Kit (RiboBio) on a CFX96 Touch Real-Time PCR Detection System (Bio-Rad Laboratories).
Glioma samples and ethical approval
To perform Western blotting and primary GSC extraction, we collected six human glioma specimens and two brain tissue samples from patients with traumatic brain injury at the Second Hospital of Lanzhou University. All tissue samples were obtained at the Department of Neurosurgery, Clinical Medical Center of the Second Hospital of Lanzhou University, and the surgeries were performed by an experienced neurosurgical team. Prior to experimental procedures, the study protocol was approved by the Ethics Committee of the Second Hospital of Lanzhou University (Approval No. 2024A-229). Written informed consent was obtained from all participants, covering the purpose of the study, potential risks, and the intended use of the samples.
Of the collected specimens, six GBM tissues and two normal brain tissues were used for immunofluorescence (IF) staining, while four GBM tissues were processed for primary cell isolation and subsequent experiments. Among these, two were classified as SVZ+ GBM and two as SVZ− GBM; the remaining two were WHO Grade 2 and Grade 3 gliomas, respectively. Normal brain tissues were obtained from patients diagnosed with deep-seated GBM who required surgical drainage. This study was approved by the Clinical Ethics Committee of the Second Hospital of Lanzhou University.
Cell culture
The HMC3 human microglial cell line was obtained from the American Type Culture Collection. Cells were maintained in Minimum Essential Medium (Solarbio, 41,500) supplemented with penicillin–streptomycin (Sangon Biotech, E607011) and 10% fetal bovine serum (Sangon Biotech, E510008). GSCs were cultured in serum-free medium based on DMEM/F12. For every 100 ml of medium, 2 ml B27 supplement, 20 µl of EGF, and 20 µl of bFGF (stock concentration, 100 µg/ml; final concentration, 20 ng/ml), and 1 ml of antibiotics (e.g., 1% penicillin–streptomycin) were added. Under sterile conditions, all components were added sequentially and mixed thoroughly. The medium was sterilized using a 0.22 µm filter membrane, protected from light, and stored at 4°C. All prepared medium was used within 1 week.
Transfection
The lentiviruses used in this study were synthesized by Hanbio Biotechnology with a viral titer of 1 × 108 TU/ml. The shRNA sequences were as follows:
shRNA-NC: 5′-TTCTCCGAACGTGTCACGTAA-3′
shRNA-1: 5′-GGAGGAGATGATGCTGTCATCCTAT-3′
shRNA-2: 5′-GGCCACTTAGAAGATGTGTATGATA-3′
shRNA-3: 5′-CAGTGACATTTCATGGTCCAGCGAT-3′
After lentiviral transduction, stable cell lines were established using puromycin selection at a concentration of 2 µg/ml.
Extraction of primary human GSCs
The enzymatic dissociation kit used in this study was the DHBTE-10 Primary (Single-Cell) Preparation Kit from RWD Life Science. The procedure was as follows: Freshly resected human GBM tissues were obtained intraoperatively by the Department of Neurosurgery and immediately transferred to the laboratory in precooled calcium- and magnesium-free D-Hank's balanced salt solution. After rinsing with PBS to remove residual blood, the tissue was minced into ∼1 mm3 fragments using sterile dissection scissors under aseptic conditions. An enzyme mixture was prepared according to the recommended ratio (e.g., 50 µl of Enzyme A, 1,875 µl of Buffer A, 25 µl of Enzyme B, and 50 µl of Enzyme C), with Enzyme A predissolved in a 37°C water bath. The enzyme cocktail was preactivated at 37°C and 100 rpm for 30 min, followed by the addition of the minced tumor tissue into the mixture, which was then processed using a single-cell suspension preparation device. Upon completion of the dissociation program, the sample was gently triturated 8–10 times using a 1 ml pipette. The resulting suspension was filtered through a 70 µm cell strainer and collected into a 50 ml centrifuge tube. The dissociation chamber was rinsed with PBS and filtered again to maximize cell recovery. The cell suspension was centrifuged at 300 × g for 10 min. After discarding the supernatant, red blood cell lysis buffer was added and incubated for 2–3 min before a second centrifugation. The final cell pellet was resuspended in serum-free complete GSC medium (supplemented with B27, 20 ng/ml EGF, and 20 ng/ml bFGF) and cultured under nonadherent conditions to enrich for GSCs.
Mice and tumor model
All experimental mice were 4-week-old male BALB/c nude mice purchased from the Beijing Vital River Laboratory Animal Technology and were housed in a specific pathogen-free facility. BALB/c nude mice carry a homozygous Foxn1nu mutation, which results in an absent thymus and a profound deficiency in T-lymphocytes while retaining functional B-cells and innate immune cells. This immunodeficient background makes them widely used for xenograft studies, as they are unable to mount effective T-cell–mediated immune responses against transplanted human tumor cells. A total of 40 4-week-old male nude mice were used for tumor model establishment. For the intracranial tumor model, mice were anesthetized with isoflurane (RWD Life Science). After anesthesia, the animals were positioned in a digital stereotaxic apparatus (RWD Life Science). Once the skull at the injection site was fully exposed (coordinates relative to the bregma: +0.7 mm anterior, −2.0 mm lateral), primary GSCs (2 × 105 cells in 5 µl volume) were slowly injected to a depth of 3 mm and maintained in situ for 10 min. On Day 12 after injection, mice were killed, and brain tissues were harvested.
Western blot
Radioimmunoprecipitation assay (Beyotime, P0013K) buffer containing protease inhibitors and/or phosphatase inhibitors was used to extract proteins from cells and glioma tissues. A membrane and cytosol protein extraction kit (Beyotime, P0033) was used to extract membrane proteins from cultured cells. After sufficient centrifugation and boiling for denaturation, protein samples were isolated and then electrotransferred onto PVDF membranes. After blocking with 5% bovine serum albumin (Beyotime, ST023-50 g), membranes were successively soaked in diluted primary and secondary antibody solutions. Finally, membranes were visualized with ImageQuant LAS 500 (GE Healthcare). Primary antibodies included rabbit anti-Chaf1b (1:10,000, Abcam, ab109442), mouse anti-Nestin (1:200, Santa Cruz Biotechnology, sc-23927), mouse anti-SOX2 (1:200, Abcam, ab97959), rabbit anti-CD206 (1:500, Proteintech, 18704-1-AP), rabbit anti-CD86 (1:1,000, Proteintech, 13395-1-AP), rabbit anti-bcl-2 (1:1,000, Proteintech, 12789-1-AP), rabbit anti-caspase-3 (1:1,000, Abcam, ab184787), mouse anti-GAPDH (1:40,000, Proteintech, 60004-1-Ig), rabbit anti-Bax (1:8,000, Proteintech, 50599-2-Ig), rabbit anti-Bcl-2 (1:8,000, Proteintech, 26593-1-AP), rabbit anti-cleaved-caspase-3 (1:500, Abmart, TA7022), rabbit anti-pi3k (1:1,000, Cell Signaling Technology, 4255S), rabbit anti-p-pi3k (1:1,000, Cell Signaling Technology, 13857S), rabbit anti-akt (1:1,000, Cell Signaling Technology, 9272S), and rabbit anti-p-akt (1:1,000, Cell Signaling Technology, 9271S). HRP goat anti-rabbit IgG (H + L; 1:4,000, ABclonal, AS014) and HRP goat anti-mouse IgG (H + L; 1:4,000, ABclonal, AS003) were used as secondary antibodies.
Limiting dilution assay (LDA)
The in vitro limiting dilution assay was used to assess the frequency of stem-like cells in primary GSCs. Prior to the experiment, primary GSCs were digested into single-cell suspensions, followed by cell counting and viability assessment. Cells were then seeded at varying densities (100, 50, 20, 10, 5, and 1 cell per well) into 96-well ultralow attachment plates, with 10–12 replicate wells per condition. Each well was supplemented with 100 µl of serum-free GSC medium containing B27, 20 ng/ml EGF, and 20 ng/ml bFGF. Plates were incubated at 37°C in a 5% CO₂ humidified incubator for 7–10 d without medium replacement. At the endpoint, wells were examined under a microscope, and the presence of tumor spheres (diameter >50 µm) was recorded. Wells meeting this criterion were defined as positive. The number of positive wells at each seeding density was subsequently entered into the extreme limiting dilution assay (ELDA) statistical tool (http://bioinf.wehi.edu.au/software/elda/) to calculate the frequency of sphere-forming functional GSCs and to evaluate intergroup differences.
Tumor sphere formation assay
The tumor sphere formation assay was conducted to evaluate the self-renewal and clonogenic capacity of primary GSCs. Prior to the experiment, primary GSCs were enzymatically dissociated into single-cell suspensions and counted using trypan blue exclusion.
Cells were then seeded at a fixed density of 1,000 cells per well into six-well ultralow attachment plates, with three replicate wells per group. Each well was supplemented with 2 ml of serum-free neural stem cell medium containing B27 supplement, 20 ng/ml EGF, and 20 ng/ml bFGF. Cells were incubated at 37°C in a humidified atmosphere of 5% CO₂ for 3 d without medium replacement. At the endpoint, tumor spheres were examined and quantified under an inverted microscope. Spheres with diameters >50 µm were considered positive. The number of spheres per well was recorded, and the average number of spheres per well was calculated to assess the self-renewal capacity of GSCs.
Coverslip cell seeding for IF
Coverslip-based cell culture was performed to facilitate subsequent IF staining for morphological and marker expression analysis. Sterile glass coverslips were placed into 24-well culture plates, and ∼2 × 104 cells were seeded per well in 1 ml of the corresponding culture medium. Cells were incubated at 37°C in a humidified atmosphere with 5% CO₂. Depending on experimental requirements, incubation was continued for 24–72 h until the cells adhered and moderately spread on the coverslips. At the end of the culture period, cells were gently rinsed twice with PBS and fixed in 4% paraformaldehyde for 15 min at room temperature. Following fixation, cells were again washed with PBS. The processed coverslips were then subjected to IF staining and subsequently mounted for observation under a confocal laser scanning or fluorescence microscope.
IF and IHC
Tumor samples were fixed with paraformaldehyde for at least 1 week and then dehydrated by a Fully Enclosed Tissue Processor (Leica, ASP300) before embedding in paraffin. Paraffin sections (4 µm) were dewaxed with different concentrations of xylene and ethanol. After antigen repair and blocking with 5% normal goat serum in PBS for 1 h at room temperature, the tissues were treated with an antibody. After incubation with specific IF labeling reagents, the slides were observed promptly. The cell slides were additionally permeabilized with 0.1% Triton X-100 (Sigma-Aldrich) in PBS for 10 min before blocking. Fluorescence images were captured by a Leica DM4B microscope, Zeiss LSM880 confocal laser scanning microscope, or TissueFAXS Plus quantitative imaging system. For hematoxylin and eosin staining, slides were first placed in Harris's hematoxylin solution (Biosharp, BL700A), washed with PBS for seconds and finally stained with 1% alcoholic eosin (Biosharp, BL700A).
Primary antibodies included mouse anti-IBA1 (1:400, ServiceBio, GB12105), primary antibodies included rabbit anti-Chaf1b (1:50, Abcam, ab109442), mouse anti-Nestin (1:50, Santa Cruz Biotechnology, sc-23927), mouse anti-SOX2 (1:50, Abcam, ab97959), rabbit anti-CD133 (1:30, Abcam, ab19898), and mouse anti-IBA1(1:400, ServiceBio, GB12105). Goat anti-rabbit IgG H&L (Alexa Fluor 488; 1:200, Abcam, ab150077), goat anti-mouse IgG H&L (Alexa Fluor 488; 1:200, Abcam, ab150113), goat anti-rabbit IgG H&L (Alexa Fluor 594; 1:200, Abcam, ab150080), and goat anti-mouse IgG H&L (Alexa Fluor 594; 1:200, Abcam, ab150116) were used as secondary antibodies.
Statistical analysis
All statistical analyses were performed using GraphPad Prism version 8.0.1. Data from at least three independent experiments are presented as mean ± standard deviation (SD). Comparisons between two groups were conducted using Student's t test. A p value of <0.05 was considered statistically significant.
Result
High Chaf1b expression is closely associated with stemness-related gene expression, malignant progression, and poor prognosis in glioma
To identify SVZ-associated candidate molecules, we first performed proteomic profiling of GBM tissues with SVZ involvement (SVZ+) versus those without SVZ involvement (SVZ−). Differentially expressed proteins were observed between the two groups, and four prognosis-related candidates—Chaf1b, DNAJC30, C2CD4C, and BPNT1—were shortlisted for further validation (Fig. 1A). We then quantified their transcript levels by qPCR in an independent cohort (SVZ+ GBM, n = 10; SVZ− GBM, n = 10). Chaf1b mRNA was significantly higher in SVZ+ tumors than in SVZ− tumors (unpaired two-tailed t test, ***p < 0.001), whereas DNAJC30, C2CD4C, and BPNT1 showed no significant differences (p > 0.05; Fig. 1B). These findings nominate Chaf1b as the key candidate for subsequent analyses. To elucidate the expression pattern and clinical relevance of Chaf1b in glioma, we first analyzed the TCGA database by comparing the differences in multiple clinical and molecular features between patients with low (n = 349) and high (n = 350) Chaf1b expression (Table 1). The results revealed that high Chaf1b expression was significantly associated with WHO grade, IDH mutation status, initial therapeutic response, age distribution, 1p/19q codeletion status, histological subtype, survival outcome, expression of stemness-related genes (SOX2, NES, and PROM1), and the immune-related gene AIF1, suggesting a potential role of Chaf1b in promoting malignant progression of glioma. Further analysis using the CCGA database revealed that high Chaf1b expression was significantly associated with reduced overall survival in patients, and its expression increased progressively with higher WHO grades of glioma, suggesting that Chaf1b may serve as a potential biomarker for glioma malignancy and poor prognosis (Fig. 1C,D). In addition, Chaf1b expression was markedly higher in IDH-wild–type gliomas compared with IDH-mutant subtypes, with the most pronounced differences observed in WHO Grade 2 and 3 gliomas. Notably, Chaf1b levels also exhibited a stepwise increase with escalating WHO grades. Given that IDH mutations are typically associated with more favorable clinical outcomes, these findings further support a strong correlation between elevated Chaf1b expression and increased tumor aggressiveness and adverse prognosis (Fig. 1E,F). To validate these findings, we further performed IF staining to examine the expression of Chaf1b and the stemness marker Nestin in a normal brain tissue and glioma samples of varying grades (including Grade 2, Grade 3, and Grade 4, stratified into SVZ− and SVZ+ groups). The results showed that both Chaf1b and Nestin were markedly upregulated in high-grade gliomas, with the highest expression observed in Grade 4 SVZ+ specimens. Notably, coexpression and colocalization of Chaf1b and Nestin were clearly detected (Fig. 1G). In parallel, we assessed the protein expression of Chaf1b and Nestin in glioma tissues of different grades using Western blot analysis. The results demonstrated that Chaf1b, Nestin, CD133, and Sox2 were significantly upregulated in high-grade gliomas, with the highest levels observed in the Grade 4 SVZ+ group, consistent with the IF findings (Fig. 1H). Taken together, these results establish a clear association between high Chaf1b expression and increased glioma malignancy, enhanced stemness, and poor prognosis, suggesting its potential role as a prognostic biomarker and functional tumor-promoting factor.
Expression pattern and clinical relevance of Chaf1b in glioma. A, Proteomic analysis of SVZ+ GBM and SVZ− GBM tissues identified candidate molecules associated with prognosis. B, Validation of candidate molecules at the mRNA level in SVZ+ and SVZ− GBM tissues. C, D, Analysis of Chaf1b expression levels and overall survival (OS) in patients with primary versus recurrent GBM based on the CGGA database. E, F, Comparative analysis of Chaf1b expression across glioma samples of different WHO grades and its association with IDH mutation status in various GBM subtypes. G, Immunofluorescence staining showing the spatial distribution of Chaf1b (red) and the stemness marker Nestin (green) in a normal brain tissue and glioma specimens of varying grades, including both SVZ− and SVZ+ subtypes. DAPI (blue) was used for nuclear counterstaining. Scale bar, 50 µm. H, Western blot analysis of Chaf1b, CD133, Sox2, and Nestin protein expression across glioma tissues of different grades. (*p < 0.05; **p < 0.01; **p < 0.001; ns, not significant).
Association between Chaf1b expression and clinicopathological and molecular features in glioma patients
SVZ+ GSCs exhibit significantly enhanced stemness compared with SVZ− GSCs
To further investigate the biological characteristics of GSCs, we isolated primary GSCs and used them in subsequent experiments (Fig. 2A). The primary GSCs were derived from two patients with SVZ+ GBM and two patients with SVZ− GBM, respectively. Sphere formation assays revealed that GSCs derived from SVZ+ tumors exhibited significantly enhanced sphere-forming capacity and demonstrated more pronounced stem-like features compared with their SVZ− counterparts (Fig. 2B). Next, we performed stemness characterization of the cultured primary GSCs by IF analysis for the stemness markers Nestin, Sox2, and CD133. The results showed markedly higher expression of all three markers in SVZ+-derived GSCs than in SVZ−-derived GSCs, indicating that SVZ+ GSCs possess stronger stem-like properties (Fig. 2C). To exclude the possibility that differences in fluorescence intensity arose from variations in cell density or other nonspecific factors, we further performed Western blot analysis to quantify the expression of Chaf1b and the stemness-associated markers Nestin, CD133, and Sox2 in GSCs derived from SVZ+ and SVZ− regions. The results showed that Chaf1b was expressed at significantly higher levels in SVZ+ GSCs (GSC128, GSC236) compared with SVZ− GSCs (GSC76, GSC82), with Nestin, CD133, and Sox2 displaying similar expression patterns. These findings indicate that SVZ+-derived GSCs possess stronger stemness properties and suggest that elevated Chaf1b expression may contribute to the maintenance of their stem-like phenotype (Fig. 2D). Collectively, these results confirm that SVZ+ GSCs exhibit superior tumorigenic stemness potential.
Isolation and identification of primary GSCs. A, Isolation of primary GSCs. B, Primary GSCs were isolated from two SVZ+ GBM and two SVZ− GBM patients, followed by tumor sphere formation assays to evaluate stemness properties. C, Immunofluorescence staining was performed to assess stem-like characteristics of the isolated GSCs using established stemness markers, including Nestin, Sox2, and CD133. D, Western blot analysis of Chaf1b, Nestin, CD133, and Sox2 protein expression in GSCs lines derived from SVZ− (GSC128, GSC236) and SVZ+ (GSC76, GSC82) tumors. GAPDH was used as a loading control. Quantification of relative protein levels is shown in the right panel. Data are presented as mean ± SD. (*p < 0.05; **p < 0.01; ***p < 0.001; ns, not significant).
Chaf1b knockdown impairs stemness and promotes apoptosis in primary GSCs
To investigate the role of Chaf1b in primary GSCs, we developed a stable Chaf1b knockdown system (sh-Chaf1b) using lentiviral vectors. The transduction efficiency was assessed by ZsGreen fluorescence labeling (Fig. 3A). Concurrently, Western blot analysis confirmed a marked reduction of Chaf1b protein levels in primary GSCs following knockdown (Fig. 3B,E). Two primary GSC lines (GSC82 and GSC236) were selected for limiting dilution assays (ELDA) after Chaf1b silencing. The results showed that, compared with the Chaf1b-NC group, sphere-forming frequencies were significantly reduced in the sh1, sh2, and sh3 groups, indicating that Chaf1b knockdown markedly impaired GSC stemness (Fig. 3C). To further validate these findings, we performed tumorsphere formation assays in GSC236 and GSC82 cells following Chaf1b knockdown. The results demonstrated that silencing Chaf1b significantly reduced the sphere-forming capacity of both GSC lines compared with the control group (Fig. 3D,F). We then examined the expression of stemness markers Nestin and Sox2 by Western blot in Chaf1b-silenced GSC236 and GSC82 cells. Notably, Chaf1b knockdown led to a marked decrease in the expression of both Nestin and Sox2, suggesting that Chaf1b plays a critical role in maintaining GSC stemness (Fig. 3G). We next investigated whether Chafl1b influences GSC survival through assessment of apoptosis-related proteins using Western blot. In both GSC236 and GSC82 cells, silencing Chaf1b led to increased expression of proapoptotic proteins cleaved Caspase-3 and Bax, while the level of antiapoptotic protein Bcl-2 was reduced (Fig. 3H). These results suggest that Chaf1b depletion induces apoptosis-associated molecular changes, implicating a potential antiapoptotic role of Chaf1b in supporting GSC survival. Taken together, our findings demonstrate that Chaf1b promotes the maintenance and survival of primary GSCs by sustaining stemness marker expression and suppressing apoptotic signaling, whereas Chaf1b knockdown markedly impairs stemness and induces apoptosis.
Chaf1b knockdown impairs stemness and promotes apoptosis in primary GSCs. A, Primary GSCs were transduced with lentiviruses, and infection efficiency was evaluated by ZsGreen fluorescence labeling. B, E, Western blot analysis was performed to assess the knockdown efficiency of three distinct shRNAs targeting Chaf1b. Quantification of relative protein levels is shown in the right panel. Data are presented as mean ± SD. (*p < 0.05; **p < 0.01; ***p < 0.001; ns, not significant). C, The self-renewal capacity of two primary GSC lines (GSC82 and GSC236) was determined using ELDA. D, F, Tumorsphere formation assays were conducted to evaluate the stemness of the primary GBM stem cells GSC236 and GSC82. Quantification of relative protein levels is shown in the right panel. Data are presented as mean ± SD. (*p < 0.05; **p < 0.01; ***p < 0.001; ns, not significant). G, Western blot analysis was used to detect the expression of stemness markers Nestin and Sox2 in GSC236 and GSC82 cells following Chaf1b knockdown (sh1, sh2, sh3). Quantification of relative protein levels is shown in the right panel. Data are presented as mean ± SD. (*p < 0.05; **p < 0.01; ***p < 0.001; ns, not significant). H, Western blotting was also performed to assess the expression of apoptotic proteins cleaved Caspase 3 and Bax, as well as the antiapoptotic protein Bcl-2, in the Chaf1b knockdown (Chaf1b-sh) groups of GSC236 and GSC82. (*p < 0.05; **p < 0.01; **p < 0.001; ns, not significant).
High Chaf1b expression promotes microglial enrichment and the remodeling of an immunosuppressive microenvironment
In the immune microenvironment of gliomas, microglia play a pivotal role, and existing literature has demonstrated a close association between GSCs and the host immune milieu (Pang et al., 2023). To further investigate the relationship between GSCs and microglia, we performed a Spearman correlation analysis using the TCGA database. The results revealed that Chaf1b expression was significantly correlated with the macrophage enrichment score (Fig. 4A). Further analyses revealed that the microglia-specific marker P2RY12 showed a weak but significant correlation with the same score (Fig. 4B), whereas TMEM119 exhibited a moderate and highly significant correlation (Fig. 4C). These findings indicate that Chaf1b expression is closely associated with microglial enrichment in the glioma microenvironment, with TMEM119 providing the strongest evidence of this relationship. Subsequently, we further compared macrophage enrichment between tumors with high and low expression of the indicated genes (Fig. 4D–F). The results showed that tumors with high Chaf1b expression exhibited significantly greater macrophage enrichment scores than those with low expression (Fig. 4D). Similarly, higher expression of the microglia-specific markers P2RY12 and TMEM119 was also associated with increased macrophage enrichment (Fig. 4E,F). These findings suggest that elevated Chaf1b expression is linked to enhanced myeloid infiltration, and the consistent patterns for P2RY12 and TMEM119 indicate that this signal is predominantly driven by microglia. To further explore the role of GSCs in the immunological microenvironment, we analyzed the compositional changes in microglial subtypes (M0, M1, M2) in glioma patients with differing Chaf1b expression levels. The results revealed a markedly increased proportion of M2-type immunosuppressive microglia in the Chaf1b-high group (Fig. 4G). To further explore the relationship between Chaf1b and stemness- or immune-related genes, we performed correlation analysis among Chaf1b, IL33, NES, SOX2, PROM1, and the microglia-specific markers TMEM119 and P2RY12. Correlation analysis showed that Chaf1b was positively associated with stemness markers (NES, SOX2, PROM1) and IL33, while both TMEM119 and P2RY12 also correlated with these genes, indicating a link between CHAF1B expression, stemness maintenance, and microglial enrichment.(Fig. 4I). In parallel, we investigated the expression patterns of Chaf1b and its associated stemness and immune regulatory markers (IL33, NES, SOX2, PROM1, PROM1, P2RY12, TMEM119) and found that all target genes were significantly upregulated in tumor tissues, further supporting a positive association between Chaf1b expression and both stemness and immune modulation (Fig. 4H). To validate these bioinformatic findings, we performed IF staining to examine the expression of Chaf1b and Iba-1 in a normal brain tissue and glioma samples of different WHO grades. We observed a progressive increase in Chaf1b and Iba-1 expression with tumor grade, with the highest expression detected in Grade 4 SVZ+ samples, consistent with the bioinformatic analysis (Fig. 4J). We further confirmed these observations through in vivo experiments. Immunofluorescence staining in a murine glioma model revealed that Iba-1 expression was predominantly localized to the peritumoral regions, indicating substantial microglial accumulation at the tumor margins. Moreover, a marked phenotypic shift of microglia was observed within the tumor core, supporting their dynamic regulatory roles in the glioma microenvironment (Fig. 4K).
Chaf1b expression correlates with microglial enrichment in gliomas and associates with stemness- and immune-related signature. A, Spearman correlation analysis of Chaf1b expression with microglial enrichment in gliomas based on the TCGA database. B, Spearman correlation analysis of P2RY12 expression with macrophage enrichment in gliomas based on the TCGA database. C, Spearman correlation analysis of TMEM119 expression with macrophage enrichment in gliomas based on the TCGA database. D–F, Analysis of macrophage enrichment in gliomas with low and high Chaf1b, P2RY12, and TMEM119 expression based on the TCGA database. G, Analysis of the CGGA database demonstrated altered proportions of microglial subtypes (M0, M1, and M2) in patients with high versus low Chaf1b expression. H, Expression levels of Chaf1b, IL33, NES, SOX2, TMEM119, P2RY12, and PROM1 were compared between tumor tissues and normal brain using TCGA transcriptomic data. I, The TCGA database was further used to assess the correlation between Chaf1b and genes associated with stemness and immune regulation, including IL33, NES, SOX2, and PROM1. J, Immunofluorescence staining was used to visualize Chaf1b (red) and the microglial marker Iba1 (green) in normal brain tissue and gliomas of different WHO grades (including SVZ− and SVZ+ GBM); nuclei were counterstained with DAPI (blue), and merged images revealed colocalization signals. The lower bar graph presents quantification of integrated fluorescence density (relative fluorescence units, RFU). Scale bar, 50 µm. K, Immunofluorescence staining was performed to assess Iba1 (green) expression in microglia within the glioma regions of the murine brain; white dashed lines indicate tumor boundaries, and a magnified view of the boxed region is shown on the right. Nuclei were stained with DAPI (blue).
Chaf1b induces microglial M2 polarization through IL-33 secretion and activates the PI3K/AKT signaling pathway
To further investigate the underlying mechanisms driving microglial phenotypic switching, we collected supernatants from primary GSCs with or without Chaf1b knockdown and subjected them to analysis using the Olink inflammation panel. The results revealed differential expression of multiple inflammatory cytokines and chemokines in the Chaf1b knockdown group compared with the control group, with IL-33 showing the most pronounced change (Fig. 5A). To validate this observation, we examined IL-33 expression in Chaf1b-silenced (Chaf1b-sh) and control (Chaf1b-NC) GSCs using enzyme-linked immunosorbent assay (ELISA) and quantitative real-time PCR (qRT-PCR). Both assays confirmed a significant reduction in IL-33 levels following Chaf1b knockdown (Fig. 5B,C). To more accurately simulate the in vivo environment, we employed a Transwell system to coculture microglia with primary GSCs. We then evaluated the stemness of GSCs following coculture using the ELDA. The results demonstrated that GSCs cocultured with M2-type microglia exhibited significantly enhanced sphere-forming capacity, whereas those cocultured with M1-type microglia showed markedly reduced stemness potential (Fig. 5D). To validate the role of Chaf1b in microglial recruitment, we conducted transwell migration assays. The results showed that knockdown of Chaf1b significantly reduced the migratory capacity of HMC3 cells compared with controls, while neutralization of IL-33 partially impaired the ability of GSCs to attract HMC3 cells. These findings suggest that Chaf1b promotes microglial recruitment, at least in part, through IL-33 signaling (Fig. 5E,F). To further investigate the phenotypic changes of microglia induced by GSCs, we examined the expression of the M2-type microglial marker CD206, the M1-type marker CD86, and components of the PI3K/AKT signaling pathway (PI3K, p-PI3K, AKT, and p-AKT) in HMC3 cells cocultured with GSC236 or GSC82. Western blot analysis revealed that Chaf1b knockdown significantly reduced CD206 and p-PI3K/p-AKT levels while increasing CD86 expression, indicating a shift from an immunosuppressive M2-like phenotype toward a more proinflammatory state (Fig. 5G). Moreover, neutralization of IL-33 largely abrogated the effects of Chaf1b-expressing GSCs on microglia, as reflected by reduced CD206 and p-PI3K/p-AKT levels and restored CD86 expression (Fig. 5H). Together, these findings suggest that Chaf1b promotes microglial M2 polarization by upregulating IL-33 and activating the PI3K/AKT pathway, thereby contributing to immune evasion, whereas Chaf1b silencing or IL-33 blockade reverses this effect and enhances immune activation. Collectively, these results reveal a mechanistic pathway whereby Chaf1b promotes microglial recruitment and M2 polarization through IL-33–mediated activation of the PI3K/AKT signaling cascade. This axis plays a pivotal role in sustaining GSC stemness and remodeling an immunosuppressive tumor microenvironment.
Chaf1b promotes microglial recruitment and M2 polarization via IL-33 induction and activation of the PI3K/AKT pathway. A, Olink inflammation panel analysis was performed using conditioned media from primary GSCs with or without Chaf1b knockdown. The results showed differential expression of several inflammatory cytokines and chemokines, with IL-33 being most prominently altered. B, C, ELISA and qRT-PCR were employed to validate IL-33 expression levels in Chaf1b knockdown (Chaf1b-sh) and control (Chaf1b-NC) GSCs. D, ELDA was used to assess the stemness of GSC236 and GSC82 following coculture with M1- or M2-polarized microglia. E, F, Transwell migration assays showing the effect of Chaf1b knockdown and IL-33 neutralization on the migration of HMC3 microglial cells cocultured with GSC236 or GSC82. Data are presented as mean ± SD. (*p < 0.05; **p < 0.01; ***p < 0.001; ns, not significant). G, Western blot analysis of CD206, CD86, PI3K, p-PI3K, AKT, and p-AKT expression in HMC3 microglial cells cocultured with GSC236 or GSC82 after Chaf1b knockdown, without IL-33 neutralization. GAPDH was used as a loading control. Quantification of relative protein levels is shown on the right. Data are presented as mean ± SD. (*p < 0.05; **p < 0.01; ***p < 0.001; ns, not significant). H, Western blot analysis of CD206, CD86, PI3K, p-PI3K, AKT, and p-AKT expression in HMC3 microglial cells cocultured with GSC236 or GSC82 after IL-33 neutralization treatment. GAPDH was used as a loading control. Quantification of relative protein levels is shown on the right. Data are presented as mean ± SD. (*p < 0.05; **p < 0.01; ***p < 0.001; ns, not significant).
Chaf1b knockdown markedly suppresses the in vivo tumorigenic capacity of GSCs and prolongs survival in nude mice
To further assess the tumorigenic potential of glioma cells following Chaf1b knockdown, we performed in vivo experiments using an immunodeficient mouse model.
GSC82 and GSC236 cells transduced with either control (Chaf1b-NC) or Chaf1b-targeting lentiviral vectors (sh1, sh2, sh3) were intracranially injected into immunodeficient mice. Tumor development was subsequently evaluated by magnetic resonance imaging (MRI) and hematoxylin and eosin (H&E) staining. The results showed that, compared with the control group, intracranial tumor volumes in Chaf1b-silenced mice (sh1, sh2, sh3) were significantly reduced, indicating that Chaf1b knockdown markedly impairs the in vivo tumor-forming capacity of GSCs (Fig. 6A,B). We then analyzed the survival of tumor-bearing mice. Kaplan–Meier survival curves were generated for mice injected with GSC82 (left) or GSC236 (right) cells transduced with either control (Chaf1b-NC) or Chaf1b-targeting (sh1, sh2, sh3) lentivirus. The analysis revealed that Chaf1b knockdown significantly prolonged overall survival in both models (Fig. 6C). Consistently, mice in the Chaf1b-shRNA groups (sh1/sh2/sh3) exhibited markedly smaller tumor volumes than those in the control group (Fig. 5D). Together, these findings indicate that Chaf1b plays a critical regulatory role in supporting the in vivo proliferative and tumorigenic capacity of GSCs.
Chaf1b knockdown markedly suppresses GSC tumorigenicity in vivo and prolongs mouse survival. A, Representative MRI scans of orthotopic glioma models in nude mice injected with GSC82 or GSC236 cells transduced with control (Chaf1b-NC) or Chaf1b-targeting shRNAs (sh1, sh2, sh3). B, Representative H&E–stained histological sections from orthotopic tumors generated by Chaf1b-NC or Chaf1b-sh (sh1, sh2, sh3) GSCs (GSC82 and GSC236). C, Kaplan–Meier survival curves of mice bearing intracranial tumors derived from GSC82 or GSC236 cells transduced with either Chaf1b-NC or Chaf1b-sh (sh1, sh2, sh3), showing significantly prolonged survival in the Chaf1b knockdown groups. D, Quantification of intracranial tumor volumes in mice injected with GSC82 or GSC236 cells expressing Chaf1b-NC or Chaf1b-sh constructs. (*p < 0.05; **p < 0.01; **p < 0.001; ns, not significant).
Discussion
With the continuous advancement of tumor biology research, cancer stem cells have increasingly been recognized as central driving units that mediate tumor initiation, sustain intratumoral heterogeneity, and contribute to therapeutic resistance and metastatic recurrence (Bayik and Lathia, 2021). In GBM, GSCs have been recognized as key drivers of tumor growth and therapeutic resistance. Notably, the concept of GSCs has evolved from a discrete cell population to a highly plastic cellular state (Sloan et al., 2024). Building on this foundation, recent studies have increasingly focused on identifying key regulatory factors and specific biomarkers to selectively suppress the stemness of GSCs, thereby curbing the malignant progression of GBM. For instance, one study demonstrated that Chi3l1 binds to CD44 and activates the Akt/β-catenin/MAZ signaling cascade, promoting the mesenchymal transition of GSCs and enhancing their stem-like properties, ultimately contributing to poor prognosis in GBM (Guetta-Terrier et al., 2023). Similarly, another study demonstrated that glycolysis-driven lactylation of PTBP1 in GSCs enhances its stability and RNA-binding activity, thereby sustaining stemness and promoting tumor progression (Zhou et al., 2025). Therefore, targeting the stemness of GSCs holds promise as a potentially effective therapeutic strategy for the treatment of GBM.
In previous studies, Chaf1b has primarily been investigated in the contexts of acute myeloid leukemia, lung squamous cell carcinoma, hepatocellular carcinoma, viral infections, and tissue fibrosis. However, no study to date has systematically explored its role in GBM, particularly its potential function in maintaining the stemness of tumor stem cells (Zhang et al., 2020; Geis et al., 2022; Dean et al., 2023; Gao et al., 2023; Imai et al., 2023; Zheng et al., 2025). In this study, we systematically elucidated for the first time the critical regulatory role of the chromatin assembly factor Chaf1b in maintaining GSCs stemness, remodeling the tumor immune microenvironment, and promoting in vivo tumorigenesis. Analysis of the CCGA database revealed that Chaf1b is significantly upregulated in high-grade gliomas and positively correlates with several stemness markers, including SOX2, NES, and PROM1, suggesting a close association with stemness maintenance. Further validation by IF and Western blot confirmed that Chaf1b is coexpressed with Nestin, particularly in Grade 4 SVZ+ GBM tissues, indicating that Chaf1b plays a functionally important role in regions characterized by enhanced stemness.
We next examined patient-derived GSCs isolated from anatomically distinct regions—SVZ+ and SVZ−—and found that SVZ+-derived GSCs exhibited markedly enhanced sphere-forming capacity and higher expression of stemness markers. These findings suggest that the SVZ microenvironment may promote stemness-associated traits through Chaf1b, further reinforcing the functional relevance of Chaf1b in sustaining GBM stemness. Moreover, this discovery provides important molecular insight into the spatial heterogeneity of GSCs.
At the level of functional validation, we established GSC models with stable Chaf1b knockdown and subsequently evaluated their stemness capacity. The results demonstrated that silencing Chaf1b markedly impaired GSC self-renewal and tumorsphere formation, accompanied by a notable reduction in the expression of stemness-associated factors such as Nestin and Sox2, indicating that Chaf1b exerts a direct role in sustaining stem-like properties. Furthermore, Chaf1b knockdown also triggered programmed cell death in GSCs by modulating apoptotic pathways, providing additional evidence that Chaf1b is a key regulator of GSC stemness maintenance.
Previous studies have firmly established that GSCs interact closely with the tumor immune microenvironment, particularly through dynamic cross talk with microglia (Pang et al., 2023). Modulating the immune microenvironment of GBM has been demonstrated to significantly enhance antitumor cytotoxic responses (Chen et al., 2022). Accordingly, we further investigated the role of Chaf1b in the interplay between GSCs and the immune microenvironment. Database analyses revealed a strong positive correlation between high Chaf1b expression and microglial enrichment, with a notable increase in immunosuppressive M2-like microglia in the high-expression group. These findings were substantiated by in vivo experiments, in which microglia were found to accumulate predominantly around the tumor margins in a murine glioma model, while those within the tumor core exhibited pronounced phenotypic switching. To identify potential mediators of this effect, we performed Olink inflammation panel profiling, which revealed that IL-33 is a key secreted factor induced by Chaf1b. Previous studies have shown that IL-33 contributes to the formation of an immunosuppressive tumor microenvironment and promotes GBM progression by regulating immune cell recruitment and activation; conversely, loss of IL-33 has been reported to suppress tumor growth (De Boeck et al., 2020). Multiple studies have also demonstrated that IL-33 regulates microglial activation, survival, and phagocytic capacity through diverse mechanisms. It plays a pivotal role in processes such as host defense against infection, synaptic pruning, and memory consolidation. As such, IL-33 is recognized as a central mediator linking central immune regulation with neural functional remodeling (Nguyen et al., 2020; He et al., 2022; Han et al., 2023). Coculture experiments further demonstrated that IL-33 derived from GSCs promotes the polarization of microglia toward the M2 phenotype and amplifies this effect through the PI3K/AKT signaling axis, thereby establishing a locally immunosuppressive microenvironment. This finding is consistent with previous reports that GSCs can engage microglia through multiple mechanisms to foster an immunosuppressive tumor milieu (Dumas et al., 2020; Li et al., 2022; Wang et al., 2024; Zhou et al., 2024). This process can be reversed either by Chaf1b knockdown or by administration of a neutralizing antibody against IL-33, suggesting that it represents a tractable and potentially targetable pathway for therapeutic intervention.
In vivo experiments confirmed the pivotal role of Chaf1b in GSCs biology. Knockdown of Chaf1b in GSCs significantly impaired tumor formation in immunodeficient mice, resulting in reduced tumor burden and extended survival. These results underscore the essential function of Chaf1b in sustaining GSCs stemness and driving tumor progression.
However, the molecular mechanisms by which Chaf1b maintains the stem-like phenotype of GSCs remain incompletely defined, warranting further investigation to elucidate its downstream regulatory networks.
Conclusion
This study reveals that Chaf1b promotes glioma progression through dual mechanisms, by sustaining the stemness and tumorigenic potential of GSCs and by inducing the secretion of IL-33, which drives microglial M2 polarization and activates the PI3K/AKT signaling pathway, thereby fostering an immunosuppressive microenvironment. We further demonstrate that elevated Chaf1b expression is strongly associated with higher tumor grade, increased expression of stemness markers, and poorer clinical outcomes, highlighting its potential as a therapeutic target.
Data Availability
All data generated or analyzed during this study are included in this published article and its supplementary information files.
Footnotes
This work was supported by the National Natural Science Foundation of China (Grant Numbers: 82460517, 82060455, 82560489, 82503355), Natural Science Foundation of Gansu Province (Grant Number, 25YFFA054, 25JRRA605), Cuiying Scientific and Technological Innovation Program of Lanzhou University Second Hospital (CY2023-QN-A06) and TechTianjin Health Research Project (GSWZD2024-16).
↵*Y.-L.H. and L.N. contributed equally to this work.
The authors declare no competing financial interests.
- Correspondence should be addressed to Ya-Wen Pan at hylpyw{at}163.com or Guo-Qiang Yuan at ksnabarro{at}163.com.
