Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2025 Apr 17;135(12):e179395.
doi: 10.1172/JCI179395. eCollection 2025 Jun 16.

Disruption of ataxia telangiectasia-mutated kinase enhances radiation therapy efficacy in spatially directed diffuse midline glioma models

Affiliations

Disruption of ataxia telangiectasia-mutated kinase enhances radiation therapy efficacy in spatially directed diffuse midline glioma models

Avani Mangoli et al. J Clin Invest. .

Abstract

Diffuse midline gliomas (DMGs) are lethal brain tumors characterized by p53-inactivating mutations and oncohistone H3.3K27M mutations that rewire the cellular response to genotoxic stress. We used RCAS/tv-a retroviruses and Cre recombinase to inactivate p53 and induce native H3.3K27M mutations in a lineage- and spatially directed manner. We generated primary mouse tumors that recapitulated human DMG. Disrupting ataxia-telangiectasia mutated (ATM) kinase enhanced the efficacy of radiation therapy (RT) in murine and patient-derived DMG models and increased survival. Microscopy-based in situ sequencing was used to spatially resolve transcriptional profiles in more than 750,000 single cells with or without ATM disruption and RT, revealing altered immune-neoplastic and endothelial cell interactions after treatment. An allelic series of primary murine DMG models with different p53 mutations confirmed that transactivation-independent p53 activity was a key mediator of radiosensitivity after ATM disruption. We generated primary DMG mouse models and performed deep profiling that revealed mechanisms of response to ATM disruption and RT that can be utilized as a therapeutic strategy.

Keywords: Brain cancer; Drug therapy; Mouse models; Neuroscience; Oncology.

PubMed Disclaimer

Figures

Figure 1
Figure 1. Atm loss improves radiosensitivity of primary murine DMBs generated using a conditional H3.3K27M allele.
(A) Representation of mouse genotypes used to generate primary mouse DMGs with p53 loss (NestinTVA p53fl/fl [nP]) and mouse DMGs with p53 loss and H3.3K27M (NestinTVA p53fl/fl H3f3aloxP-Stop-loxP-K27M-Tag/+ [nPH]) with or without the conditional H3.3K27M allele. Mice also contained 1 intact and 1 floxed allele of Atm (Atmfl/+, not shown). (B) Dot plot showing the time to tumor formation between nPH and nP mice without any statistical significance. Welch’s t test was used to determine statistical significance. (C) Schematic showing nPHAfl/+ (Atmfl/+) and nPHAfl/fl (Atmfl/fl) within the RCAS/tv-a retrovirus and conditional H3K27M allele. (D) Time to tumor formation showing no statistical difference between nPHAfl/fl and nPHAfl/+ mice. Welch’s t test was used to determine statistical significance. (E) Overall survival of nPHAfl/fl and nPHAfl/+ mice following administration of 3 daily fractions of 10 Gy image-guided focal brain irradiation, with significantly longer median survival of nPHAfl/fl mice. P = 0.03, by Mantel-Cox (log-rank) test. (F) Whole-mount and H&E images of tumor cells from nP, nPH, nPHAfl/+, and nPHAfl/fl (top to bottom) mice exhibiting hypercellularity and infiltration of normal brain tissue. (G) IHC for HA expression indicating the presence of the PDGF-β HA tag. (H) IHC for p53. (I) IHC for histone 3 lysine 27 trimethylation (H3K27me3). (J) IHC displaying the Ki67 proliferation of tumor cells. (K) Anti-FLAG IHC confirmed the presence of the FLAG-HA tag. Scale bars: 100 μm (for all H&E and IHC images in FK); 1,000 μm (for whole-mount images in F)
Figure 2
Figure 2. Spatial clustering in primary mouse DMGs treated with focal brain irradiation or tumoral Atm deletion.
(A) Schematic of DMG-bearing mice were subjected or not to focal brain irradiation and ISS. All mice were of the genotype NestinTVA p53fl/fl H3f3aloxP-Stop-loxP-K27M-Tag/+ with either Atm-intact (Atmfl/+) or Atm-null (Atmfl/fl) tumors. (B) Harmony integration showing clustering of 4 tumor-bearing mice with the H3f3aloxP-Stop-loxP-K27M-Tag/+ genotype with either Atm-intact (Atmfl/+) or Atm-null (Atmfl/fl) tumors. (C) Spatial clustering of cells into 10 cell archetypes based on label transfer in 4 tumor-bearing mouse brains (bottom color panel), H&E images of whole brain (left), and distribution of cells within normal brain, tumor periphery, and tumor core annotated in bar graph (right). Top row indicates Atm intact with and without irradiation. Bottom row indicates Atm null with and without irradiation. Color legend on the bottom corresponds to individual cell type noted on bar graph.
Figure 3
Figure 3. Differentially expressed genes and neighborhood analysis of primary mouse DMGs with tumoral Atm loss and/or focal irradiation.
(A) Spatial identification of tumors by expression of Pdgfra, Olig1, and Olig2 in all conditions (top to bottom): Atm-intact, Atm-intact with irradiation, Atm-null, Atm-null with irradiation. (B) Spatial identification of p53 loss in all tumor conditions: Atm-intact without and with irradiation (top row, left to right). Atm-null without and with irradiation (bottom row, left to right). (C) Key differentially expressed genes in Atm-intact neoplastic tumor cells treated with and without focal brain irradiation. The log2 fold change and P values for all genes are indicated in Supplemental Table 3. (D) Key differentially expressed genes in Atm-null neoplastic tumor cells treated with and without focal brain irradiation. The log2 fold change and P value for all genes are indicated in Supplemental Table 4. (E) Co-occurrence plot of Atm-intact (nPHAfl/+) tumor showing the number compared with the distance of various cell types in relation to neoplastic cells. (F) Co-occurrence plot of Atm-intact (nPHAfl/+) tumor with irradiation showing the number compared with the distance of various cell types in relation to neoplastic cells. (G) Co-occurrence plot of Atm-null (nPHAfl/fl) tumor showing the number compared with the distance of various cell types in relation to neoplastic cells. (H) Co-occurrence plot of Atm-null (nPHAfl/fl) tumor with irradiation showing the number compared with the distance of various cell types in relation to neoplastic cells. Red arrow indicates increased frequency of immune cells compared with neoplastic cells. Color legend for EH is on the right side panel. OPC, oligodendrocyte precursor cell; TLC, T lymphocyte. Neighborhood enrichment and co-occurrence analyses were conducted on the entire slide. All unlabeled cells were removed for analysis.
Figure 4
Figure 4. Pharmacologic inhibition and DNA damage response signaling in primary mouse DMGs with tumoral Atm loss and/or focal irradiation.
(A) Overall survival of mice bearing patient-derived SF8628 DMG xenografts were treated with 20 mg/kg AZD1390 for 2 weeks (ATM inhibitor [ATMi], 5 days per week for 2 weeks) and/or focal brain irradiation (RT, 2 Gy for 3 days per week for a total dose of 12 Gy). (B) Western blot of SF8628 followed by AZD1390 treatment with and without RT (0 hours, 1 hour, 3 hours, 6 hours, 24 hours). (C) Gross dissection and images of the brain with tumor were captured. Representative IHC images of tumor cells from NestinTVA p53fl/fl PDGF-β + H3.3K27M + Cre (PKC) mice treated with vehicle (V) or the ATM inhibitor drug (D) AZD1390, with or without 10 Gy irradiation, for the following (top to bottom): H&E, p-KAP1, total KAP1, and γH2AX. Imaged with Zeiss Axio imager. Original magnification, ×20, insets ×40. (D) PDGF-β + H3.3K27M + p53fl/fl (n = 5 per treatment group) cells stained for p-KAP1 revealed increased p-KAP1 expression in samples treated with 1 dose of 10 Gy and showed significance in vehicle-treated samples (PKC + V +/– RT). P = 0.0079, by Mann-Whitney U test. This expression was significantly reduced in drug-treated tumors subjected to RT compared with vehicle-treated tumors subjected to RT, suggesting the ATM inhibitor AZD1390 sensitized the DIPG tumor–bearing mice to irradiation. *P = 0.0159 and **P = 0.0079, by Mann-Whitney U test. (E) PDGF-β + H3.3K27M + p53fl/fl tumor cells stained for total KAP1 show unchanged levels across all treatment groups (n = 5 per treatment group). (F) PDGF-β + H3.3K27M + p53fl/fl tumor cells stained with yH2AX demonstrated increased expression in samples treated with 1 dose of 10 Gy RT when compared with their respective non-RT-treated samples (n = 5 per treatment group). **P = 0.0079, by Mann-Whitney U test for the vehicle-treated groups and *P = 0.0317, by Mann-Whitney U test for drug-treated groups.
Figure 5
Figure 5. Effect of tumor-specific Atm loss in primary DMGs on a Cdkn1a-null (p21–/–)background.
(A) Overview of the p21–/– genotypes analyzed. (B) Tumor-free survival of nIp21A mice with and without intact Atm using the log-rank test. (C) Post-focal brain irradiation survival of nIp21A mice with and without intact Atm indicating a statistically significant survival benefit in nlp21Afl/+ mice (P < 0.05, by log-rank test). (D) IHC showing p21 expression in nIp21A mouse brains. NestinTVA Ptenfl/fl Atmfl/+ (nPtenA) tumor–bearing brain generated with identical RCAS viruses shown as a control. (E) Plot indicating the percentage of tumor cells that stained positive for p21 compared with the total cell count. (F) IHC images of staining for Ki67 showing proliferation for nlp21Afl/+ and nIp21Afl/fl. (G) TUNEL staining of tumor-bearing brains of nIp21A mice with and without intact Atm in tumors collected 1 hour after focal brain RT. (H) Quantification of TUNEL staining in nlp21Afl/+ mice. *P < 0.05, by unpaired t test. Scale bars: 2 mm (D top row) 50 μm, (D bottom row, F, and G).
Figure 6
Figure 6. Tumor formation in mice expressing a p53 transactivation domain 1 mutant.
(A) Schematic for conditional the p53 transactivation domain 1 mutant and mouse genotypes for expression of a p53 transactivation domain 1 mutant. (B) Tumor-free survival in the np53(25,26) group compared with the np53 control group based on a log-rank test. (C) Time to tumor presentation in the p5325,26/fl group compared with the p53fl/fl control group (Wilcoxon test). (D) IHC for anti-HA staining in the p53 and p53(25,26) groups. Scale bars: 50 μm. (E) IHC for p53 expression in the p53 (scale bar: 100 μm) and p53(25,26) (scale bar: 50 μm) groups.
Figure 7
Figure 7. Effect of Atm loss on survival after fractionated focal brain irradiation in mouse DMGs expressing a p53 transactivation domain 1 mutant.
(A) Schematic showing the p53LSL(25,26) allele and genotypes for NestinTVA p53LSL(25,26)/fl Ink4A/ARFfl/fl mice with either Atmfl/fl or Atmfl/+. (B) IHC images showing p-Atm in Atmfl/+ and Atmfl/fl tumors. Scale bars: 50 μm. (C) IHC images showing p-KAP1 expression in Atmfl/+ and Atmfl/fl tumors. Scale bars: 50 μm. (D) Time to tumor formation in NestinTVA p53LSL(25,26)/fl Ink4A/ARFfl/fl mice with either Atmfl/fl or Atmfl/+ (dot plot). NS, by Wilcoxon test. (E) Overall survival following fractionated brain irradiation in mouse DMGs expressing a p53 transactivation domain 1 mutation with or without Atm loss. The P value is based on a log-rank test.

Update of

References

    1. Wu G, et al. Somatic histone H3 alterations in pediatric diffuse intrinsic pontine gliomas and non-brainstem glioblastomas. Nat Genet. 2012;44(3):251–253. doi: 10.1038/ng.1102. - DOI - PMC - PubMed
    1. Schwartzentruber J, et al. Driver mutations in histone H3.3 and chromatin remodelling genes in paediatric glioblastoma. Nature. 2012;482(7384):226–231. doi: 10.1038/nature10833. - DOI - PubMed
    1. Zhang L, et al. Exome sequencing identifies somatic gain-of-function PPM1D mutations in brainstem gliomas. Nat Genet. 2014;46(7):726–730. doi: 10.1038/ng.2995. - DOI - PMC - PubMed
    1. Werbrouck C, et al. TP53 pathway alterations drive radioresistance in diffuse intrinsic pontine gliomas (DIPG) Clin Cancer Res. 2019;25(22):6788–6800. doi: 10.1158/1078-0432.CCR-19-0126. - DOI - PubMed
    1. Deland K, et al. Tumor genotype dictates radiosensitization after Atm deletion in primary brainstem glioma models. J Clin Invest. 2021;131(1):e142158. doi: 10.1172/JCI142158. - DOI - PMC - PubMed