Jounce Therapeutics

Aug 31, 2023

Abstract Glioblastoma (GBM) tumors are enriched in immune-suppressive myeloid cells and are refractory to immune checkpoint therapy (ICT). Targeting epigenetic pathways to reprogram the functional phenotype of immune-suppressive myeloid cells to overcome resistance to ICT remains unexplored. Single-cell and spatial transcriptomic analyses of human GBM tumors demonstrated high expression of an epigenetic enzyme—histone 3 lysine 27 demethylase (KDM6B)—in intratumoral immune-suppressive myeloid cell subsets. Importantly, myeloid cell-specific Kdm6b deletion enhanced proinflammatory pathways and improved survival in GBM tumor-bearing mice. Mechanistic studies showed that the absence of Kdm6b enhances antigen presentation, interferon response and phagocytosis in myeloid cells by inhibition of mediators of immune suppression including Mafb, Socs3 and Sirpa. Further, pharmacological inhibition of KDM6B mirrored the functional phenotype of Kdm6b-deleted myeloid cells and enhanced anti-PD1 efficacy. This study thus identified KDM6B as an epigenetic regulator of the functional phenotype of myeloid cell subsets and a potential therapeutic target for enhanced response to ICT. Access options Get Nature+, our best-value online-access subscription $29.99 / 30 days $119.00 per year Prices vary by article type from$1.95 Additional access options: Fig. 2: KDM6Bhigh intratumoral myeloid cells are immune suppressive. Fig. 3: Myeloid-specific Kdm6b deletion improves survival in mice. Fig. 4: Kdm6b deletion alters intratumoral monocytes and macrophages. Fig. 5: Kdm6b regulates the chromatin landscape of myeloid cells. Fig. 6: Kdm6b regulates genes associated with inflammatory response. Fig. 7: Pharmacological inhibition of KDM6B improves anti-PD1 response. Data availability ChIP–seq, scATAC-seq, scRNA-seq and spatial transcriptomics (Visium) data that support the findings of this study have been deposited in the European Genome Phenome Archive under accession code EGAS00001007002 . Mass cytometry data have been deposited in the ImmPort repository (accession no. SDY2295 ). All requests for data should be made to the corresponding authors (S.G. and P. Sharma), following verification of any intellectual property or confidentiality obligations. Human GBM scRNA-seq data are available in Sequence Read Archive with accession no. PRJNA588461 , the GEO dataset with accession no. GSE131928 and at https://www.brainimmuneatlas.org/ . Source data are provided with this paper. All other data supporting the findings of this study are available from the corresponding author on reasonable request. Code availability The scRNA-seq, scATAC-seq and ChIP–seq analyses presented in the manuscript were performed with open-source algorithms as described in Methods . Further details will be made available by the authors on request. References Locati, M., Curtale, G. & Mantovani, A. Diversity, mechanisms, and significance of macrophage plasticity. Annu. Rev. Pathol. 15, 123–147 (2020). Acknowledgements This research is supported by the MD Anderson Physician Scientist Award (S.G.), Khalifa Physician Scientist Award (S.G.), Andrew Sabin Family Foundation Fellows Award (S.G.) and Clinic and Laboratory Integration Program Award (S.G.). We thank L. Xiong, B. Guan, D. N. Tang, S. Kemp and A. Jung for technical assistance. We thank the CATALYST-working group at MD Anderson Cancer Center for human GBM tumor samples. CATALYST is supported by the MD Anderson GBM Moon Sho. P. Sharma is a member of the Parker Institute for Cancer Immunotherapy. P. Sharma and S.G. are members of the James P. Allison Institute. S.G. and P. Sharma are members of the BreakThrough Cancer–GBM Extended Leadership Team. Author information Authors and Affiliations Department of Immunology, The University of Texas MD Anderson Cancer Center, Houston, TX, USA Sangeeta Goswami, Deblina Raychaudhuri, Pratishtha Singh, Seanu Meena Natarajan, Mercedes Hennessey, Aminah J. Tannir & Padmanee Sharma Department of Genitourinary Medical Oncology, The University of Texas MD Anderson Cancer Center, Houston, TX, USA Sangeeta Goswami, Jan Zhang, Swetha Anandhan & Padmanee Sharma James P. Allison Institute, The University of Texas MD Anderson Cancer Center, Houston, TX, USA Sangeeta Goswami & Padmanee Sharma Yulong Chen, Marc D. Macaluso, Zhong He, Sonali Jindal, Sreyashi Basu & Padmanee Sharma Department of Neurosurgery, The University of Texas MD Anderson Cancer Center, Houston, TX, USA Candice Poon, Brittany Parker Kerrigan & Frederick F. Lang Authors Deblina Raychaudhuri Pratishtha Singh Seanu Meena Natarajan Yulong Chen Candice Poon Mercedes Hennessey Aminah J. Tannir Jan Zhang Swetha Anandhan Brittany Parker Kerrigan Marc D. Macaluso Zhong He Sonali Jindal Frederick F. Lang Sreyashi Basu Padmanee Sharma Contributions S.G. developed the project, designed the experiments, analyzed data, wrote the manuscript and acquired funding. D.R., S.M.N. and P. Singh performed experiments, analyzed data and wrote the manuscript. P. Singh and Y.C. performed bioinformatics analyses. J.Z., M.H., A.J.T. and S.A. helped with murine experiments. B.P.K., C.P. and F.F.L. provided human GBM tumor samples. M.D.M. and S.J. performed H&E, IHC and immunofluorescence staining of human GBM samples. S.B. and Z.H. helped with human scRNA-seq and Visium analysis. P. Sharma provided scientific input, edited the manuscript and acquired funding. Corresponding authors Competing interests P. Sharma reports consulting, advisory roles and/or stocks/ownership for Achelois, Apricity Health, BioAlta, Codiak BioSciences, Constellation, Dragonfly Therapeutics, Forty-Seven, Inc., Hummingbird, ImaginAb, Jounce Therapeutics, Lava Therapeutics, Lytix Biopharma, Marker Therapeutics, BioNTx, Oncolytics, Glympse, Infinity Pharma and Polaris; and owns a patent licensed to Jounce Therapeutics. The remaining authors declare no competing interests. Peer review Peer review information Nature Cancer thanks Jason Miska, Michael Platten and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Additional information Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Extended data a, Feature plots of scRNA seq data demonstrating the expression of indicated genes encoding epigenetic enzymes, in intratumoral myeloid cell clusters from Fig. 1c . b, UMAP plot of scRNA seq data showing CD45+ immune cell subsets in the GBM TME derived from patients with GBM (n = 4 patients). c, Violin plots demonstrating the expression level of indicated genes in the different immune cell clusters. d, UMAP plot of scRNA seq data depicting CD45+ immune cell subsets in the GBM TME derived from patients with GBM (n = 20 patients). e, Violin plots representing the expression level of indicated genes in the different immune cell clusters. f, Violin plot demonstrating the expression level of KDM6B in the different myeloid cell clusters in the TME of patients with GBM (n = 5 patients) from Fig. 1c . The dotted lines represent the cutoff KDM6B expression values used to define KDM6Bhigh and KDM6Blow cells. g,h, Volcano plots representing differentially expressed genes between KDM6Bhigh and KDM6Blow myeloid cells in the TME of patients with GBM (g) n = 4 patients (h) n = 13 patients. Volcano Plots show the fold change (log2FC) plotted against the Absolute Confidence log10 adjusted p value (-log10(p)). Two-tailed Poisson’s test was performed using the Seurat R package. a, Hematoxylin and Eosin stained GBM tumor sections (n = 3 patients). Each section represent one patient. Scale included in the images. b, Representative figures showing immunohistochemical staining for KDM6B in GBM tissue samples (n = 3 patients). c,d, Multiplex immunofluorescence staining shows distribution of CD3 (white), KDM6B (blue), CD68 (green) GFAP (yellow) and CD163 (red) in GBM tumor samples. Arrow marks show the colocalization of KDM6B protein expression with CD68 and CD163. Data is representative of two independent experiments. e,i, UMAP plots of matched scRNA seq data showing CD45+ immune cell subsets in the GBM TME used to embed single cells to their spatial coordinates in tissue sections by applying CellTrek. f,j, Gene expression data of all the different immune cell clusters from matched patients plotted on spatial coordinates. g,h,k,l, Visium spatial gene expression score plots for the CD68 + KDM6B + CD163 + KLF2 + CXCL8 expressing myeloid cell clusters (highlighted in teal) and CD3E + CD8A + GZMB + GZMK + ICOS T cell clusters (highlighted in red) in human GBM tumors to assess spatial distribution of KDM6B expressing myeloid cells and activated CD8 T cells. a–c, Dotplots showing the average gene activity score of genes of interest as well as percentage of cells in the cluster expressing the gene in the indicated clusters (a shown in Fig. 2e , and b and c shown in Extended Data Fig. 2e, i ). a, Schematic representation demonstrating generation of the LysMcreKDM6Bfl/fl genetic murine model. b, Representative image of an agarose gel showing bands depicting PCR amplified DNA from Kdm6b deleted homozygous mice (single 400 bp band), Kdm6b deleted heterozygous mice (both 368 and 400 bp bands), and control homozygous mice (single 368 bp band). Data is representative of two independent experiments. c, t-SNE plots and box and whisker plots depicting the identity and abundance of different immune cell populations present in the indicated anatomical locations in control and LysMcreKDM6Bfl/fl mice as determined from CyTOF analysis. n = 3 female mice/group. Two-tailed Student’s t test was performed (ns= non-significant). Whiskers represent minimum to maximum values. a, Heatmap showing the expression of genes of interest in the different Cd45+ immune cell clusters (shown in Fig. 3c ). b, Bar graphs depicting the frequencies of the different intratumoral immune cell subsets in control and LysMcreKDM6Bfl/fl mice. c, Bar graph representing the ratio of intratumoral CTLs and Tregs in control and LysMcreKDM6Bfl/fl mice as determined from scRNA seq. d, Violin plots depicting the expression level of Lyz2 in different immune cell clusters in control and LysMcreKDM6Bfl/fl mice. Data representative of two independent scRNA seq experiments. e–g, Box and whisker plots representing the flow cytometry analysis of abundance of different immune cell populations present in PBMC (e), lymph nodes (f) and spleen (g) derived from control and LysMcreKDM6Bfl/fl GBM tumor bearing female mice (n = 3 biologically independent samples/group for (E), n = 4 biologically independent samples/group for f and g. Two-tailed Student’s t-test was performed. Whiskers represent minimum to maximum values. a, Heatmap showing the expression of protein markers of interest in the indicated immune cell clusters from Fig. 3g determined by mass cytometry. b, Heatmap showing the expression of genes of interest in the different myeloid cell clusters (shown in Fig. 4 A-right panel). c, Bar plots representing the frequencies of intratumoral myeloid clusters from control and LysMcreKDM6Bfl/fl mice. Data representative of two independent experiments. a, UMAP demonstrating the CTL and Treg clusters in the GBM (GL261) TME of control and LysMcreKDM6Bfl/fl mice determined by scATAC seq (as shown in Fig. 5a, b ). Bar graphs depicting the frequencies and ratio of intratumoral CTLs and Tregs in control and LysMcreKDM6Bfl/fl mice as determined from scATAC seq. b, Coverage plots depicting the chromatin accessibility of the indicated genes in the CTL and Treg clusters. c, Heatmap showing the expression of genes of interest in the indicated myeloid cell clusters (shown in Fig. 5d ). d,e, Coverage plots depicting accessibility of indicated chromatin regions (peaks) in genes of interest. a, Representative gating strategy on FlowJo for analysis of flow cytometry data showing GL261 phagocytosis by BMDMs (shown in Fig. 6j ). b, Box and Whisker plots showing the concentration of the indicated cytokines in 24 hour culture supernatants of M2, M1 and M0 BMDMs derived from control and LysMcreKDM6Bfl/fl male mice(n = 6 biologically independent samples/condition). Data is representative of 2 independent experiments. Two-tailed Student’s t-test was performed. Whiskers represent minimum to maximum values. c, Relative average expression of indicated genes in M2 and M1 BMDMs derived from control and LysMcreKDM6Bfl/fl male mice (n = 3 biologically independent samples for control M1 BMDMs and n = 4 for LysMcreKDM6Bfl/fl M1 BMDMs). Two-tailed Student’s t-test was performed. Whiskers represent minimum to maximum values. d,e, Concentration of IL-6 and IL-10 in the supernatants of BMDMs derived from control and LysMcreKDM6Bfl/fl male mice following efferocytosis and subsequent LPS stimulation (n = 4 biologically independent samples/condition). Two-tailed Student’s t-test was performed. Data is representative of two independent experiments. Whiskers represent minimum to maximum values. f,g, Box and Whisker plots demonstrating the relative gene expression of INOS and IL10 in BMDMs derived from control and LysMcreKDM6Bfl/fl male mice following efferocytosis and subsequent LPS stimulation (n = 3 biologically independent samples for control BMDMs and n = 4 LysMcreKDM6Bfl/fl BMDMs for (F), n = 4 biologically independent samples for control BMDMs and n = 3 LysMcreKDM6Bfl/fl BMDMs for G). Two-tailed Student’s t-test was performed. Whiskers represent minimum to maximum values. a, Heatmap representing the expression of genes of interest in the indicated Cd45+ immune cell clusters as determined by scRNA seq (shown in Fig. 7c ). a, Representative axial MRI images of CT-2A tumor from vehicle treated mice (left panel) and GSK-J4 treated mice (right panel), taken on day 14 post tumor inoculation. b, Box and whisker plot showing the difference in CT-2A tumor volumes (determined from MRI) between vehicle and GSK-J4 treated mice (n = 10 female mice/group). Two-tailed Student’s t-test was performed. Whiskers represent minimum to maximum values. c, Box and whisker plot depicting the difference in CT-2A tumor weight (harvested on day 22 post tumor inoculation) between vehicle and GSK-J4 treated mice (n = 10 female mice/group). Two-tailed Student’s t-test was performed. Whiskers represent minimum to maximum values. d, Heatmap demonstrating the expression of protein markers of interest in the indicated CD45+ immune cell clusters in the GBM (CT-2A) TME of vehicle & GSK-J4 treated mice as determined by mass cytometry. e, Heatmap showing the expression of protein markers of interest in the indicated CD45+ immune cell clusters as determined by mass cytometry (shown in Fig. 7j ).