STING activation induces cytotoxic and immune responses in meningiomas via inflammatory cell death pathways - PubMed (original) (raw)

. 2026 Feb 12;17(1):2685.

doi: 10.1038/s41467-026-69296-1.

Shashwat Tripathi # 1 2, Hinda Najem # 1 2, Harrshavasan Congivaram 1 2, Mateo Gomez 1 2, Thomas K Sears 1 2, Lisa Hurley 1 2, Leah K Billingham 1 2, Caylee Silvers 1 2, Wenxia Wang 1 2, Deanna Tiek 1 2, Nishanth Sadagopan 1 2, Rahul Chaliparambil 1 2, Qianyi Pu 1 2, Ching Man Wai 3, Abrar Choudhury 4, Alicia Steffens 1 2, Kathleen McCortney 1 2, Gustavo Ignacio Vazquez Cervantes 1 2, Daniel Oyon 1, Luis Fernandez 1, Ashley Selner 1, Jawad Fares 1, Matthew Hagan 1, S Joy Trybula 1, Jacky Yeung 5, Matthieu Peyre 6, Michel Kalamarides 6, Peng Zhang 1 2, Alexander H Stegh 7 8, David M Ashley 9 10, Craig M Horbinski 1 2 11, Jared T Ahrendsen 2 11, Pouya Jamshidi 1 2 11, Daniel J Brat 1 2 11, Rimas V Lukas 1 2, Roger Stupp 1 2, Maciej S Lesniak 1 2, Adam M Sonabend 1 2, Gavin P Dunn 12, James P Chandler 1 2, Matthew C Tate 1 2, Stephen T Magill 1 2, Jason Miska 1 2, Michael A Curran 13, David R Raleigh 4 14 15, Amy B Heimberger 16 17

Affiliations

STING activation induces cytotoxic and immune responses in meningiomas via inflammatory cell death pathways

Mark W Youngblood et al. Nat Commun. 2026.

Abstract

Meningiomas are common tumors of the central nervous system that are typically treated with surgery or radiation, but lack established systemic therapies. Activation of the stimulator of interferon genes pathway with an agonist such as 8803 can trigger anti-tumor immune responses. Using integrated molecular approaches, here we show that this pathway is targetable in both neoplastic and immune populations within the meningioma microenvironment. Meningioma tumor cells exhibit promoter hypomethylation and increased chromatin accessibility of the STING genomic locus, associated with robust expression of this gene. Treatment of diverse patient meningiomas ex vivo with 8803 induces direct tumor cytotoxicity through inflammatory cell death pathways, including induction of gasdermin D membrane pore formation. Release of necrotic tumor debris triggered by 8803 activates macrophages and upregulates matrix metalloproteinase production, facilitating degradation of extra-cellular collagen. Injection of preclinical meningiomas with 8803 induces survival benefits, including in an immunocompetent orthotopic setting, through remodeling of the tumor microenvironment, immune infiltration, and downregulation of tumor-mediated immune suppression, thereby nominating 8803 for treatment consideration in meningiomas.

© 2026. The Author(s).

PubMed Disclaimer

Conflict of interest statement

Competing interests: MAC declares the following competing interests: grants and personal fees from ImmunoGenesis that are relevant to the presented study, and unrelated work from Alligator Bioscience, ImmunOS, Oncoresponse, Kineta, Xencor, Astrazeneca. MAC holds the patent Cyclic dinucleotides as agonists of stimulator of interferon gene dependent signaling licensed to ImmunoGenesis, Inc. The remaining authors declare no potential competing interests related to the content of this manuscript.

Figures

Fig. 1

Fig. 1. The immune landscape of the meningioma microenvironment is dominated by innate immune suppressive populations.

A Fifteen distinct immune populations (CD45 positive) were identified using single-cell RNA-sequencing data from meningioma (n = 13 patients) and adjacent dura tissues (n = 9 patients) based on the listed markers. Differentially expressed genes (DEGs) were found using standard parameters in Seurat FindAllMarkers (which use default non-parametric Wilcoxon rank sum test on all clusters), and examples are shown in each cluster box,. 30,206 immune cells were included. B Myeloid and lymphoid populations exhibited differential distribution between adjacent dura and meningiomas, such as the enrichment of phagocytic tumor-associated macrophages (phago.TAMs, shown in olive color) in meningioma tissue. C T and NK cell populations were present at higher levels in dura, whereas macrophages were dominant within meningiomas. Differential abundance testing was conducted on batch corrected sample level clusters using MiloR standard workflow, including: buildgraph, makeNhoods, countcells, calcNhoodDistance, and testNhoods. Visualization was performed using plotDAbeeswarm. Neighborhoods are colored if _p_-value < 0.05; otherwise, shown in grey. Immune cell populations with fewer than 5 significant neighborhoods were removed from the strip plot. A two-sided negative binomial generalized linear model with spatial FDR correction was used. Data was also analyzed using t-tests (with multiple testing correction) and results are further plotted in Fig. S2 (p < 0.01; q < 0.01). D Gene expression dot plot of selected functional markers among the general immune populations shown in (A). 30,206 total immune cells were used for analysis. Bubble size corresponds to the percent of cells expressing each marker; colors indicate average expression. Phagocytic TAMs have high expression of M2-like markers such as CD163, CD204, and CD206.

Fig. 2

Fig. 2. Collagen-mediated immune suppression is present within meningiomas.

A Multiprotein immunofluorescence demonstrates that Phago.TAM cells in meningiomas express immunosuppressive proteins such as TIM-3 (left, yellow label) and LAIR-1 (right, white label). Representative results are shown; n = 4 patient samples were tested from varying WHO grades with similar findings. Scale bar is 100 µM and 20 µM for the inset. B Using the same n = 4 patient samples, collagens were strongly detected in meningioma stromal regions and exhibited concentric laminar patterns (left). Both tumor and stromal areas harbor cells expressing immune suppressive LAIR-1 - a known receptor for collagens (right). The right image shows an overlapping region from panel A. Scale bar is 200 (left) and 100 (right) µM. C Volcano plot showing select chemokines enriched in the dura (left side) and meningioma tissues (right side) based on scRNA-seq of immune clusters shown in Fig. 1A. The annotated chemokines correlate with the elevated myeloid (blue text) and lymphoid (red text) populations in the meningioma and dura, respectively. Differentially expressed genes were calculated through pseudobulk analysis using DESeq2 analysis (two-sided Wald test with Benjamini-Hochberg (BH) procedure for multiple correction) on immune cells grouped to compare between dura and meningioma. D UMAP plot of scRNA-seq data from meningiomas and dura reveals three distinctive NK populations associated with resting (green), activated (blue), and exhausted (red) phenotypes. Cell annotation was performed based on DEGs using Seurat FindAllMarkers. E Dot plot of selected NK marker genes. Bubble size corresponds to the percent of cells expressing each marker; colors indicate average expression. Activated NK cells exhibited higher levels of granzyme and CD247, which were relatively absent in resting populations. Exhausted NK cells had high expression of both activation and inhibition markers and components of the TGF-β pathway.

Fig. 3

Fig. 3. Increased Chromatin accessibility and promoter hypomethylation drives STING expression across meningioma subtypes.

A UMAP cluster analysis of all cells within the meningioma microenvironment distinguishes distinct populations of meningioma, fibroblast, and extracellular matrix (ECM)-producing meningioma cells (n = 22 patients). Cell annotation was based on differentially expressed genes (DEGs) using standard parameters in Seurat FindAllMarkers on all clusters (B) Dot-plot demonstrating immune target expression analysis across meningioma cell populations, including the expression of STING (TMEM173) in meningioma, myeloid, and endothelial cells. Bubble size corresponds to the percent of cells expressing each marker; colors indicate average expression. C STING (orange) is widely expressed in the meningioma microenvironment, specifically in the neoplastic (SSTR2+), endothelial (CD31+), and myeloid (CD163+) populations. The bottom images also show the nuclear expression of pIRF3 (green) - a marker of downstream STING activation in meningioma, endothelial, and myeloid cells. n = 4 patients with similar results. D Pseudo-bulk ATAC-seq signal across myeloid and tumor cell populations from meningiomas reveals active chromatin regions near the STING locus (light blue highlight). Similar activation is seen in myeloid populations from glioma (GSE230389); however, the chromatin near STING is closed in the glioma tumor populations. In the middle, a heatmap depicts the correlation of methylation position with expression in meningioma (top heatmap rows) and glioma (bottom heatmap rows). Significant correlations are indicated by “X” in correlation rows (FDR < 0.05). The strongest methylation probe correlating with STING expression is highlighted (red, bold, italicized): cg16983159. E Correlation of cg16983159 methylation level with STING expression among meningioma (n = 584 patients) and glioblastoma (n = 170 patients) samples, highlighted in green and grey, respectively. Two-sided Pearson correlation was used.

Fig. 4

Fig. 4. Treatment of ex vivo meningioma with the STING agonist 8803 induces programmed necrotic cell death.

A Schema created with BioRender demonstrating ex vivo experiments of resected patient meningiomas, including (i) treatment of meningioma microenvironment with 8803 (B), (ii) removal of CD45+ immune cells before treatment with 8803 (C), and (iii) treatment with inhibitors of cell death pathways before treatment with 8803 (D). B Longitudinal viability assessments of ex vivo meningiomas spanning WHO grades and DNA methylation subtypes treated with 10 μM of the STING agonist 8803 (n = 9 patients). Each color and symbol represents a distinct patient (see further details in Fig. S6). The standard deviation is shown as error bars for each time point of each sample, calculated across a median of 10 replicates. P < 0.01 from paired t-test among treated and control cells (_P_ = 1.33 × 10−3, 3.29 × 10−3, and 1.91 × 10−5 for 24 hr, 48 hr, and 72 hr, respectively). C** Cell viability of ex vivo meningioma specimens (n = 3 patients) in which the CD45+ immune cell population is removed before treatment with 10 μM of 8803. The cell death percentage across replicates for each sample was significantly higher than that of untreated controls in the corresponding sample. Each bar represents the cell killing percentage relative to the untreated control for each patient. Data are presented as mean values +/− SEM. Paired t-test performed between treated and control conditions for NU04113 (P = 5.60 × 10−3), ITA-68 (P = 4.11 × 10−3), and ITA-66 (P = 4.52 × 10−2). D Ex vivo meningiomas (n = 3 patients) treated with cell death pathway inhibitors before administration of 8803. E Single sample Gene Set Enrichment Analysis of bulk RNA-sequencing of patient meningioma samples after treatment with 8803 (n = 4 paired patient samples (control vs 8803 treatment)). Each symbol/color represents a different ex vivo meningioma. F Genes associated with caspases and pyroptosis that were differentially expressed (| logFC | > 1, _p_-value adjusted <0.05) in patient samples (n = 4 paired patient samples, control vs 8803 treatment). The range of changes in expression counts per million (CPM) is shown below each gene (log transformed). In (C–E**), data were analyzed using a two-sided T-test. In E and F, the box bounds the interquartile range divided by the median, with the whiskers extending to a maximum of 1.5 times the interquartile range beyond the box.

Fig. 5

Fig. 5. The STING agonist 8803 triggers meningioma pyroptosis through GSDMD induction.

A Schema created with BioRender showing the cleavage of GSDMD via caspase enzymes results in oligomerization and pore formation in the cell membrane, leading to leakage of cellular contents. B Western blot analysis of ex vivo meningiomas treated with 8803 (n = 5 patients) and then probed for uncleaved or oligodimerized GSDMD with GAPDH as a loading control. The administration of 8803 results in increased oligomerization of GSDMD, consistent with the induction of pyroptosis. C Volumetric densitometry data of the Western blot data indicate an increased ratio of oligomerized GSDMD to total GSDMD, normalized to GAPDH (n = 5 patients). The box bounds the interquartile range divided by the median, with the whiskers extending to a maximum of 1.5 times the interquartile range beyond the box. D Scanning electron microscopy (SEM) of untreated (left) and 8803-treated (right) meningioma cells. Cells treated with 10 μM 8803 for 6 hours display a plasma membrane with irregular pores compromising cellular integrity. For SEM, at least two imaging sessions for each of the time points were performed (control, 6 h, and 48 h) with two replicates per time point. Representative images were selected from what was observed more broadly in the cell sections. E Meningioma cell death induced with 8803 is partially blocked with the ROS inhibitors N-acetyl cysteine (NAC) and glutathione monoethyl ester (GSHee). The percentage of cell death at 48 hours is shown for four samples in triplicate (each color/shape represents a different meningioma). A significant decrease in cell killing is observed when 8803 treatment is blocked with NAC (P = 1.82 × 10−4) or GSHee (P = 0.011). F Representative image of transmission electron microscopy (TEM) of meningioma mitochondria treated for 6 hours with STING-agonist 8803. Internal mitochondrial structures like cristae are collapsed, while a compromised double membrane is evident. For TEM, at least two imaging sessions for each of the time points were performed (control, 6 h, and 48 h) with two replicates per time point. Scale bar: 500 nm.

Fig. 6

Fig. 6. 8803 exerts a therapeutic effect in a preclinical meningioma model.

A Efficacy of 8803 on the IOMM-Lee cell line was evaluated in immunocompromised nude mice. Schema created with BioRender. On day 10, one cohort of mice received unilateral intratumoral injection of 8803 (red), while the contralateral side received a PBS injection (purple). In a second cohort used as controls to evaluate any systemic effects of 8803 in the first cohort, both flank tumors were treated with PBS sham (green and blue). B Western blot analysis of meningioma (CH157, IOMM-LEE, MGS1) and glioblastoma cells (GBM12 and 0827) probed for expression of STING. C Densitometry results from Western blot confirming STING expression in meningioma cell lines (n = 3 cell lines: CH157, IOMM-Lee, MGS1) and glioblastoma cell lines (n = 2 cell lines: 0827, GBM12, 0827) relative to GAPDH (p = 0.037). Data are presented as mean values +/− SEM and analyzed using a two-sided T-test. D Representative photo of a mouse on post-implant day 22 showing minimal tumor burden on the treated right side (red) but an evident tumor on the contralateral PBS-treated left side (purple) (n = 5 mice). E Graph of tumor volumes of bilaterally implanted IOMM-Lee cells. Mice (n = 5 mice/group) were treated with 5 µg of 8803 on day 10 (red). Tumors in all other groups (green, blue, purple) continued to grow, but the 8803-treated tumors remained stable (P < 0.0001 for time and P = 0.000022 for treatment). Data are presented as mean values +/− SEM and analyzed using two-way ANOVA with Tukey’s multiple comparisons. F Graph demonstrating the tumor volumes of the flank model of unilateral implanted IOMM-Lee cells in which only one side is initially treated at a later time point, starting on day 17. Mice (n = 7–10 mice/group) were treated with 5 µg of 8803. The tumor in the untreated mouse continued to grow, but the 8803-treated tumor remained stable (P < 0.0001 for time and treatment). Experiment terminated on Day 28 due to the control tumors ulcerating and reaching a humane endpoint. Data are presented as mean values +/− SEM and analyzed using two-way ANOVA with Tukey’s multiple comparisons.

Fig. 7

Fig. 7. 8803 exerts a therapeutic effect across orthotopic preclinical meningioma models.

A Schematic of orthotopic treatment experiments in immunocompromised mice created with BioRender. B Survival of nude mice orthotopically implanted with IOMM-Lee using Kaplan-Meier analysis. Controls: n = 10 mice (median survival [MS]: 24.5 days), 8803: n = 9 mice (MS: 34 days). Statistics (log-rank test): control versus 8803, P = 0.0125. C Survival of nude mice orthotopically implanted with CH157-MN using Kaplan-Meier analysis. The remaining mice were terminated at day 150. Controls: n = 10 mice (MS: 16 days), 8803: n = 10 mice (MS: 45 days). Statistics (log-rank test): control versus 8803, P = 0.00005.

Fig. 8

Fig. 8. Orthotopic STING activation transforms the meningioma microenvironment.

A Survival of immunocompetent FVB/N mice orthotopically implanted with MGS1 using Kaplan-Meier analysis. Controls: n = 10 mice (median survival [MS]: 39 days), 8803: n = 9 mice (MS: undefined days). Statistics (log-rank test): control versus 8803, P = 0.0009. B Representative images of extracranial (ExtrCr) and intracranial (IntrCr) tumor burden in control and 8803 treated MGS1-bearing mice. Though both conditions exhibited ExtrCr growth of tumors away from the region of injection, only PBS-treated mice had IntrCr tumor burden on post-mortem analysis. C Representative H&E image of control and 8803-treated MGS1-bearing mice demonstrating the presence and absence of intracranial meningiomas. Mice treated with 8803 demonstrated immune infiltration, while PBS-treated mice exhibited brain invasion, as shown in the bottom insets. n = 3 mice for PBS control and n = 3 mice for 8803 treatment. D UMAP plot of scRNA-seq data of 42,150 immune cells (CD45+). n = 6 mice per condition (untreated and 8803). Three mice were pooled for each sequencing run. Cell annotation based on differentially expressed genes (DEGs) using standard parameters in Seurat FindAllMarkers on all clusters (Table S2). Meningioma tumor cells were additionally identified using FindTransferAnchors using human scRNA (analyzed in Fig. 1) as a reference database. ISG: Interferon-Stimulated Genes; TAM: tumor-associated macrophages (E) Stacked bar graph and (F) Strip plot demonstrating that 8803 increases cytotoxic immune populations. Differential abundance testing was conducted on batch corrected sample level clusters using MiloR standard workflow, including: buildgraph, makeNhoods, countcells, calcNhoodDistance, and testNhoods. Visualization was performed using plotDAbeeswarm. Neighborhoods are colored if _p_-value < 0.05; otherwise shown in grey. Immune cell populations with fewer than 5 significant neighborhoods were removed from the strip plot. A two-sided negative binomial generalized linear model with spatial FDR correction was used.

Fig. 9

Fig. 9. Meningioma lysis triggers macrophage activation and extracellular matric breakdown through TLR and STING pathways.

A Volcano plot of DEGs from immune cells showing that cytotoxic genes such as Prf1 and Ifng, and matrix metalloproteinases are highly upregulated, whereas the immune suppressive molecules like TIM3 and LAIR are decreased following 8803 treatment. Differentially expressed genes were calculated using Seurat FindMarkers using the two-sided Wilcoxon Rank Test between 8803 and control from immune cells. Data was also analyzed using DESeq2 and presented in Fig. S8D. B Volcano plot of DEGs from meningioma cells showing a decrease in several notable collagens and an increase in STING (Tmem173), GSDMD, and cell death caspases. Differentially expressed genes were calculated using Seurat FindMarkers using the two-sided Wilcoxon Rank Test between 8803 vs control from meningioma tumor cells. Data was also analyzed using DESeq2 and presented in Fig. S8E. C Dot plot of GSEA results showing increased pyroptosis and ROS generation (‘Reactive oxygen species biosynthetic process’) and decreased ECM assembly in meningioma cells following 8803 treatment. NES: Normalized Enrichment Score. Analyzed using weighted Kolmogorov-Smirnov test with FDR correction (D) Luminex assay reveals elevation in MMP-3 (a protease that degrades extracellular matrix) in response to treatment of wild-type (FvB) macrophages with MGS1 lysate (n = 3 biological replicates). Data are presented as mean values +/− SEM and analyzed using a two-sided T-test. See additional assays and conditions in Fig. S11-S12. E Treatment of MGS1 tumors with 8803 (n = 3 mice) resulted in a decrease in COL6A (right) compared to control (left) (n = 3 mice). COL6A is a key component of human meningioma stromal mass. Scale bars represent 100 μm. Adjacent H&E slides are shown for each tumor slice. These panels are also shown in Fig. S12 at different magnifications, along with additional examples. F Graphical summary created with BioRender demonstrating that the administration of the STING agonist 8803 triggers mitochondrial ROS-mediated programmed necrotic death pathways (pyroptosis, necroptosis, ferroptosis) in meningioma cells, resulting in DAMP release via GSDMD pore formation that activates latent macrophages to produce collagen-degrading enzymes and attract T cells through chemokine elaboration.

References

    1. Price, M. et al. CBTRUS Statistical Report: primary brain and other central nervous system tumors diagnosed in the United States in 2017–2021. Neuro Oncol.26, vi1–vi85 (2024). -DOI -PMC -PubMed
    1. Choudhury, A. et al. Meningioma DNA methylation groups identify biological drivers and therapeutic vulnerabilities. Nat. Genet.54, 649–659 (2022). -DOI -PMC -PubMed
    1. Wang, A. Z. et al. Single-cell profiling of human dura and meningioma reveals cellular meningeal landscape and insights into meningioma immune response. Genome Med.14, 49 (2022). -DOI -PMC -PubMed
    1. Nassiri, F. et al. A clinically applicable integrative molecular classification of meningiomas. Nature597, 119–125 (2021). -DOI -PMC -PubMed
    1. Choudhury, A. et al. Perivascular NOTCH3+ stem cells drive meningioma tumorigenesis and resistance to radiotherapy. Cancer Discov.14, 1823–1837 (2024). -DOI -PMC -PubMed

MeSH terms

Substances

Grants and funding

LinkOut - more resources