Mismatch repair deficiency drives malignant progression and alters the tumor immune microenvironment in glioblastoma models
Montserrat Puigdelloses Vallcorba, Nishant Soni, Seung-Won Choi, Kavita Rawat, Tanvi Joshi, Sam Friedman, Alice Buonfiglioli, Angelo Angione, Zhihong Chen, Gonzalo Piñero, Gabrielle Price, Mehek Dedhia, Raina Roche, Emir Radkevich, Anne M. Bowcock, Deepti Bhatt, Winfried Edelmann, Robert M. Samstein, Timothy E. Richardson, Nadejda M. Tsankova, Alexander M. Tsankov, Ranjit S. Bindra, Raul Rabadan, Juan C. Vasquez, Dolores Hambardzumyan
Montserrat Puigdelloses Vallcorba, Nishant Soni, Seung-Won Choi, Kavita Rawat, Tanvi Joshi, Sam Friedman, Alice Buonfiglioli, Angelo Angione, Zhihong Chen, Gonzalo Piñero, Gabrielle Price, Mehek Dedhia, Raina Roche, Emir Radkevich, Anne M. Bowcock, Deepti Bhatt, Winfried Edelmann, Robert M. Samstein, Timothy E. Richardson, Nadejda M. Tsankova, Alexander M. Tsankov, Ranjit S. Bindra, Raul Rabadan, Juan C. Vasquez, Dolores Hambardzumyan
View: Text | PDF
Research Article Cell biology Immunology Neuroscience Oncology

Mismatch repair deficiency drives malignant progression and alters the tumor immune microenvironment in glioblastoma models

Abstract

Mutations in DNA mismatch repair (MMR) pathway genes (MSH2, MSH6, MLH1, and PMS2) are linked to acquired resistance to temozolomide (TMZ) and high tumor mutation burden (TMB) in high-grade gliomas (HGGs), including glioblastomas (GBMs). However, the specific roles of individual MMR genes in the initiation, progression, TMB, microsatellite instability (MSI), and resistance to TMZ in gliomas remain unclear. Here, we developed de novo mouse models of germline and somatic MMR-deficient (MMRd) HGGs. Surprisingly, loss of Msh2 or Msh6 did not lead to high TMB, MSI, nor did it confer a response to anti–programmed cell death 1 (anti–PD-1) in GBM. Similarly, human GBM showed discordance between MMR gene mutations and the TMB and MSI. Germline MMRd promoted the progression from low-grade to HGG and reduced survival compared with MMR-proficient (MMRp) tumor–bearing mice. This effect was not tumor cell intrinsic but was associated with MMRd in the tumor immune microenvironment, driving immunosuppressive myeloid programs, reduced lymphoid infiltration, and CD8+ T cell exhaustion. Both MMR-reduced (MMRr) and MMRd GBM were resistant to TMZ, unlike MMRp tumors. Our study shows that N3-(2-fluoroethyl) imidazotetrazine (KL-50), an imidazotetrazine-based DNA targeting agent that induces MMR-independent cross-link–mediated cytotoxicity, was effective against germline and somatic MMRr and MMRd GBMs, offering a potential therapy for TMZ-resistant HGG with MMR alterations.

Authors

Montserrat Puigdelloses Vallcorba, Nishant Soni, Seung-Won Choi, Kavita Rawat, Tanvi Joshi, Sam Friedman, Alice Buonfiglioli, Angelo Angione, Zhihong Chen, Gonzalo Piñero, Gabrielle Price, Mehek Dedhia, Raina Roche, Emir Radkevich, Anne M. Bowcock, Deepti Bhatt, Winfried Edelmann, Robert M. Samstein, Timothy E. Richardson, Nadejda M. Tsankova, Alexander M. Tsankov, Ranjit S. Bindra, Raul Rabadan, Juan C. Vasquez, Dolores Hambardzumyan

×

Figure 3

Germline loss of Msh2 drives GBM progression in adult mice without increasing TMB or MSI.

Options: View larger image (or click on image) Download as PowerPoint
Germline loss of Msh2 drives GBM progression in adult mice without incre...
(A) Schematic illustration for the generation of adult mGBM models and (B) Kaplan-Meier survival curves for Nf1 mGBM and (C) PDGFB mGBM. (D) Schematic illustration of the experiment for WES. Data show SNVs in coding regions per Mb of sequenced DNA and MSI scores for WT, Msh2, and Msh6-deficient tumors. (E) Representative images and (F) quantification of IHC for IBA1+ TAMs in tumors from C. Scale bars: 100 μm and 50 μm (inset). (G) Overview of sample processing for scRNA-seq. (H) Uniform manifold approximation and projection (UMAP) plot for all nonimmune cell and immune cell scRNA-seq data in this study, colored by cell-type annotation. (I) Stacked bar plots depicting the proportion of annotated cells within various genotypes. (J) UMAP plots of all cells colored by cell density. Red indicates high cell density, and blue indicates low density. (K) UMAP plots of cells colored by cell-cycle status and stacked bar plots depicting the proportion of annotated cells within various genotypes. (L) Quantification graph of multiplex IF for Ki67+OLIG2+ double-positive cells in the Ki67+ cell population from tumors. (M) 2D quadrant scatter plot representing the GBM cellular state across tumor cells split by Msh2 status. Orange color indicates AC-like cells, purple color indicates MES-like cells, yellow indicates NPC-like cells, and green indicates OPC-like cells (32). (N) Distribution of the proportion of MES- plus AC-like malignant cells (left) and proportion of NPC- plus OPC-like malignant cells (right) across samples split by Msh2 genotypes. Black = WT, blue = Msh2 heterozygous, and red = Msh2-KO. Box plots display the median (central line), interquartile range (box), and whiskers extending to the smallest and largest values within 1.5 times the interquartile range. Individual data points are shown as dots to provide a detailed view of the sample distribution. *P < 0.05, **P < 0.01, and ****P < 0.0001, by MC and GBW tests (B and C) and 1-way ANOVA followed by Tukey’s post hoc analysis (F). Panels A, D and G include illustrations created with BioRender.