Inhibition of O-GlcNAc transferase activates type I interferon-dependent antitumor immunity by bridging cGAS-STING pathway
- Department of Microbial Infection and Immunity, Infectious Disease Institute, The Ohio State University, United States
- Pelotonia Institute for Immuno-Oncology, The Ohio State University Comprehensive Cancer Center, The Ohio State University, United States
- Department of Periodontics and Oral Medicine, University of Michigan School of Dentistry, University of Michigan Rogel Cancer Center, University of Michigan, United States
- Department of Cancer Biology and Genetics, The Ohio State University, United States
- Department of Radiation Oncology, The Ohio State University, United States
- Department of Chemistry and Biochemistry, University of Delaware, United States
- Eppley Institute for Research in Cancer and Allied Diseases, University of Nebraska Medical Center, United States
Figures
Figure 1 with 1 supplement
OGT is significantly upregulated in human and mouse tumor samples.
(A–C) Boxplot showing mRNA expression level of Ogt, Normal and tumor samples (A), Individual stages (B), Nodal metastasis status (C). The plot was generated using the UALCAN online server. (D–E) Boxplot showing protein expression level of OGT, Normal and tumor samples (A), Individual stages (B). The plot was generated using the UALCAN online server (https://ualcan.path.uab.edu/analysis.html). (F) IHC analysis of OGT expression in normal colon tissues, primary colon tumor samples (from Human Protein Atlas, https://www.proteinatlas.org/), scale bar: 400 μm. (G) Western blot analysis of OGT and O-GlcNAc expression in normal, adjacent and tumor tissues in Apcmin spontaneous tumor mice. (H) HE and IHC staining of OGT in adjacent and tumor tissues in Apcmin spontaneous tumor mice, scale bar: left panel 275 μm, right panel 75 μm, n=3 respectively. (I) Schematic of AOM/DSS model of colitis-associated colorectal cancer (CAC). (J) Western blot analysis of OGT and O-GlcNAc expression in normal, adjacent and tumor tissues in CAC model. (K) HE and IHC staining of OGT in adjacent and tumor tissues in CAC model, scale bar: left panel 275 μm, right panel 75 μm, n=4 respectively. human samples (A–F), mouse samples (G–K). Statistical significance was determined by Pearson test, unpaired Student’s t-test, *p<0.05, **p<0.01, ***p<0.001, ns, no significant difference. Data represent the mean of ± SD.
Figure 1—figure supplement 1
The expression pattern of OGT in TCGA and CTPAC databases.
(A) Boxplot showing mRNA expression level of Ogt in multiple types of cancers. The plot was generated using the GEPIA2 online server. *p<0.05. (B–D) Boxplot showing mRNA expression level of Ogt in LUAD, Normal and tumor samples (B), Individual stages (C), Nodal metastasis status (D). The plot was generated using the UALCAN online server. (E–F) Boxplot showing protein expression level of OGT in LUAD, Normal and tumor samples (E), Individual stages (F). The plot was generated using the UALCAN online server (https://ualcan.path.uab.edu/analysis.html). Statistical significance was determined by Pearson test, unpaired Student’s t-test, *p<0.05, **p<0.01, ***p<0.001, ns, no significant difference. Data represent the mean of ± SD.
Figure 2 with 1 supplement
Epithelial OGT deletion inhibits mouse colorectal tumorigenesis.
(A) Western blot analysis of OGT expression in intestinal tissues and counting tumor numbers in APCmin and Ogt IEC cKO mice. (B) Histology analysis of intestinal carcinogenesis by HE staining, scale bar: up panel 275 μm, bottom panel 20 μm, n=3, respectively. (C) Real-time PCR analysis of cytokines mRNA expression in intestine. (D) ELISA analysis of cytokines expression in intestine. Statistical significance was determined by unpaired Student’s t-test, *p<0.05, **p<0.01, ***p<0.001, ns, no significant difference. Data represent the mean of ± SD.
Figure 2—figure supplement 1
The genotype of APCmin,Villin-Cre and Ogtfl/fl mice.
Genomic DNA was extracted from the tails of APCmin, Villin-Cre, and Ogtfl/fl mice and used for PCR with various primers. The resulting products were separated by agarose gel electrophoresis to determine the genotype.
Figure 3 with 1 supplement
OGT deficiency induces cGAS/STING-dependent the type I IFN pathway.
(A–D) Real-time PCR analysis of cytokines mRNA expression in different Ogt−/− cell lines including MC38 (A), LLC (B), HT29 (C), B16-OVA (D) cells. (E) Western blot analysis of the activation of the interferon signaling pathway in different Ogt−/− cell lines including MC38, LLC and B16-OVA cells. (F) Western blot analysis of the activation of the interferon signaling pathway in Ogt−/− rescued cell lines including MC38, LLC, HT29 and B16-OVA cells. (G–H) Real-time PCR and western blot analysis of cytokines mRNA expression and the activation of the interferon signaling pathway in different Ogt−/−Mavs−/− double knockout clones in MC38 cells. (I–K) Real-time PCR and ELISA analysis of cytokines mRNA expression in Ogt−/−cGAS−/− double knockout clones in MC38 (I–J), HT29 (K) cells. (L) Western blot analysis of the activation of the interferon signaling pathway in Ogt−/−cGAS−/− or Ogt−/−Sting−/− double knockout clones in MC38, HT29 and B16-OVA cells. (M–N) BMDCs pre-treated with B16-OVA-Ogt−/− (L), B16-OVA-Ogt−/−cGAS−/− or B16-OVA-Ogt−/−Sting−/− cells (M) supernatant, and co-cultured with OT-1 T cell, then T cell proliferation was evaluated by flow cytometry, OVA257-264 as a positive control. Representative fluorescence-activated cell sorting histograms and statistical data are shown. Data are representative of two or three independent experiments. Statistical significance was determined by unpaired Student’s t-test, one-way ANOVA, two-way ANOVA, *p<0.05, **p<0.01, ***p<0.001, ns, no significant difference. Data represent the mean of ± SD.
Figure 3—figure supplement 1
In vitro Cross-Priming of T Cells by Ifnar-/- BMDCs.
Ifnar1-/- BMDCs was pre-treated with B16-OVA-Ogt+/+ or B16-OVA-Ogt−/− supernatant, and co-cultured with OT-1 T cell, then T cell proliferation was evaluated by flow cytometry. Representative fluorescence-activated cell sorting histograms and statistical data are shown. Data are representative of three independent experiments. Statistical significance was determined by one-way ANOVA, *p<0.05, **p<0.01, ***p<0.001, ns, no significant difference. Data represent the mean of ± SD.
Figure 4 with 1 supplement
OGT deficiency causes DNA damage and accumulates cytosolic DNA.
(A) The extranuclear dsDNA in different Ogt−/− MC38 cells clones were determined by PicoGreen staining assay was quantified by image J. (B) The extranuclear dsDNA in different Ogt−/− MC38 cells clones were determined by anti-dsDNA fluorescence staining assay and was quantified by image J. (C) Western blot analysis of γH2AX and H2AX expression in different Ogt−/− cell lines including MC38, LLC and B16-OVA cells. (D) Analysis of γH2AX and H2AX expression in different Ogt−/− clones by anti-γH2AX staining assay and was quantified by image (J). (E) The DNA damage was determined by comet assay, and extranuclear dsDNA was analyzed by using CometScore in Ogt−/− MC38 cells. (F) Western blot analysis of γH2AX and H2AX expression in Ogt−/− rescued cells including MC38, LLC, HT29 and B16-OVA cells. (G) The DNA damage in rescued MC38 cells were determined by comet assay, and extranuclear dsDNA was analyzed by using CometScore. Data are representative of three or four independent experiments. Statistical significance was determined by unpaired Student’s t-test, one-way ANOVA, *p<0.05, **p<0.01, ***p<0.001, ns, no significant difference. Data represent the mean of ± SD.
Figure 4—figure supplement 1
OGT deficiency causes DNA damage and accumulates cytosolic DNA.
(A) The extranuclear dsDNA in different Ogt−/− LLC clones were determined by PicoGreen staining assay and was quantified by image J. (B) The extranuclear dsDNA in different Ogt−/− LLC clones were determined by anti-dsDNA fluorescence staining assay and was quantified by image J. Data are representative of three independent experiments. Statistical significance was determined by unpaired Student’s t-test, two-way ANOVA, *p<0.05, **p<0.01, ***p<0.001, ns, no significant difference. Data represent the mean of ± SD.
Figure 5 with 1 supplement
The C terminal of HCF-1 rescue DNA damage and the type I IFN pathway in Ogt−/− cells.
(A) Volcano plot of OGT binding proteins identified by LC–MS/MS from stably expressed exogenous GFP-OGT in OGT knockout HT29 cells. (B) OGT and HCF1 binding was confirmed by immunoprecipitation assay in OGT rescued HT29 cells. (C) OGT and HCF1 binding was confirmed by immunoprecipitation assay in 293T cells. (D) HCF1 cleavage was confirmed by western blot in Ogt rescued MC38 cells. (E) Co-IP analysis of the interaction between OGT and different HCF-1 mutant. (F) Real-time PCR analysis of cytokines mRNA expression effected by HCF-1 isoforms in MC38 OGT knockout cells. (G) Western blot analysis of γH2AX and H2AX expression in exogenous HCF-1C600 expressed MC38 Ogt knockout cells. (H) The extranuclear dsDNA were determined by anti-dsDNA fluorescence staining assay and was quantified by image J in exogenous HCF-1C600 expressed MC38 OGT knockout cells. Data are representative of three or four independent experiments. Statistical significance was determined by one-way ANOVA, two-way ANOVA, *p<0.05, **p<0.01, ***p<0.001, ns, no significant difference. Data represent the mean of ± SD.
Figure 5—figure supplement 1
In vitro pull-down assay analysis of the interaction of HCF-1C600 and OGT.
(A) His pull-down assays were used to analyze the interaction between HCF-1C600 and OGT. (B) His-tagged OGT and HCF1 were expressed and purified, followed by SDS-PAGE separation and staining with Coomassie blue.
Figure 6 with 4 supplements
Ogt deficiency inhibits tumor progression through enhancing infiltration by CD8+ T cells.
(A–B) Tumor volume, weight of Ogt+/+ or Ogt−/− MC38 tumors in C57BL/6 J mice and mice survival, n=5 respectively. (C–D) Tumor volume, weight of Ogt+/+ or Ogt−/− LLC tumors in C57BL/6 J mice and mice survival, n=5 respectively. (E–H) Flow cytometry analysis of percentage of CD4+ and CD8+ T cells population (E) and IFN-γ+ (F), TNF-α+ (G), IFN-γ+TNF-α+ double positive (H) intratumoral CD8+ T cells population in MC38 tumors, subcutaneous tumor isolated at day 18 post-tumor inoculation, n=5 respectively. (I) Tumor volume and weight of Ogt+/+ or Ogt−/− MC38 in Rag2-/- mice, n=5 respectively. (J–K) Tumor volume, weight of Ogt+/+ or Ogt−/− MC38 tumors injected with either control IgG or anti-CD8α at days 0, 7, and 14 post tumor inoculation in C57BL/6 J mice and mice survival, n=5 respectively. (L–M) Tumor volume, weight of Ogt-/- rescued MC38 tumors in C57BL/6 J mice, tumor growth volume and weight (L), mice survival (M). N–O Tumor volume, weight of Ogt−/−cGAS−/− or Ogt−/−Sting−/− double knockout MC38 tumors in C57BL/6 J mice, tumor growth volume and weight (N), mice survival (O).(P) Flow cytometry analysis showing percentage of CD4+ and CD8+ T cells population (N), CD8+ IFN-γ+ (O), CD8+ TNF-α+ T cell population (P) in Ogt−/−cGAS−/− or Ogt−/−Sting−/− double knockout MC38 tumors in C57BL/6 J mice, subcutaneous tumor isolated at day 18 post-tumor inoculation. (Q–R) Tumor volume, weight of Ogt+/+ or Ogt−/− MC38 tumors injected with either control IgG or anti-PD-L1 at days 7, 10, and 13 post tumor inoculation in C57BL/6 J mice and mice survival, n=5 respectively. (S) Kaplan-Meier survival curves for colorectal cancer patients with low (n=207) or high (n=231) OGT transcripts in TCGA dataset. (T) Progression-free survival curves for colorectal cancer patients with low (n=58) or high (n=58) OGT transcripts in TCGA dataset. (U) Scatterplot presenting the association between the mRNA expression level of OGT and CD8+ T infiltration, Spearman’s r=–0.263, P=9.75E-6, Spearman’s rank correlation test. Data are representative of two or three independent experiments. Statistical significance was determined by Spearman’s rank correlation test, unpaired Student’s t-test, one-way ANOVA, two-way ANOVA, *p<0.05, **p<0.01, ***p<0.001, ns, no significant difference. Data represent the mean of ± SD.
Figure 6—figure supplement 1
The cell proliferation of different tumor model in vitro and B16-OVA tumor growth analysis in vivo.
(A–C), The cell proliferation of different Ogt−/− tumor model in vitro. (A) MC38, (B) LLC, (C) B16-OVA Ogt−/− cells proliferation in vitro. (D–E) Tumor volume, weight of Ogt+/+ or Ogt−/− B16-OVA tumors in C57BL/6 J mice, and mice survival, n=5 respectively. Data are representative of two or three independent experiments. Statistical significance was determined by unpaired Student’s t-test, two-way ANOVA, *p<0.05, **p<0.01, ***p<0.001, ns, no significant difference. Data represent the mean of ± SD.
Figure 6—figure supplement 2
Ogt deficiency inhibits tumor progression through enhancing infiltration of CD8+ T cells.
(A) A total schematic of flow cytometry analysis. (B–C) Flow cytometry analysis showing percentage of IFN-γ and TNF-α-expressing intratumoral CD8+ T cells in MC38 tumors with or without PMA and ionomycin stimulation. (D–H) Flow cytometry analysis showing percentage of CD45+ (D), CD11b+ CD11c+ (E), CD11b+ F4/80+ (F), CD11b+ Ly6C+ (G) and Treg cells (H) in MC38 tumor model, n=5 respectively. (I–P) Flow cytometry analysis showing percentage of CD4+ and CD8+ T cells in LLC (I), B16-OVA cells (M). IFN-γ+, TNF-α+ and IFN-γ+ TNF-α+ double positive expressing intratumoral CD8+ T cells in LLC (J–L), or B16-OVA (N–P) tumors isolated at day 18 post-tumor inoculation, n=5 respectively. Data are representative of two or three independent experiments. Statistical significance was determined by unpaired Student’s t-test, one-way ANOVA, *p<0.05, **p<0.01, ***p<0.001, ns, no significant difference. Data represent the mean of ± SD.
Figure 6—figure supplement 3
Ogt deficiency inhibits tumor progression through enhancing infiltration of CD8+ T cells.
(A) Tumor volume and weight of Ogt+/+ or Ogt−/− MC38 tumors injected with control IgG or anti-CD4 antibody at day 0, 7 and 14 post tumor inoculation in C57BL/6 J mice, n=5 respectively. (B–D) The percentage of CD4+ and CD8+ T cells (B), CD8+ IFN-γ+ (C) and CD8+ TNF-α+ (D) in Ogt-/- rescued MC38 tumors isolated at day 18 post-tumor inoculation. (E–F) Flow cytometry analysis showing percentage of CD8+ IFN-γ (E), CD8+ TNF-α+ (F) in Ogt−/−cGAS−/− or Ogt−/−Sting−/− double knockout MC38 tumors in C57BL/6 J mice, subcutaneous tumor isolated at day 18 post-tumor inoculation. (G–H) Tumor volume, weight of Ogt+/+ or Ogt−/− LLC tumors injected with control IgG or anti-PD-L1 antibody at days 7, 10, and 13 post tumor inoculation in C57BL/6 J mice, and mice survival, n=5 respectively. Data are representative of two or three experiments. Statistical significance was determined by unpaired Student’s t-test, one-way ANOVA, two-way ANOVA, *p<0.05, **p<0.01, ***p<0.001, ns, no significant difference. Data represent the mean of ± SD.
Figure 6—figure supplement 4
OGT expression is negatively related to CD8+ T cells infiltration in human colorectal cancer.
(A) Gene Ontology (GO) enrichment and pathway analysis in OGT high and OGT low patients. (B–K) GSEA analysis in OGT high and OGT low patients. T cells activation (B), response to interferon-gamma (C), interferon-gamma production (D), antigen processing and presentation (E), interleukin-1 production (F), interleukin-12 production (G), dectin-1 mediated noncanonical NF-κB signaling (H), mismatch repair (I), covalent chromatin modification (J) and DNA repair complex (K) in OGT high and OGT low patients. (L–Q) RNAseq analysis of mRNA expression pattern in OGT high and OGT low patients, CD8A (L), IFNG (M), ISG15 (N), MX1 (O), CD274 (P) and CXCL10 (Q) mRNA expression patterns in OGT high and OGT low patients. Statistical significance was determined by Pearson test, unpaired Student’s t-test, *p<0.05, **p<0.01, ***p<0.001, ns, no significant difference. Data represent the mean of ± SD.
Figure 7 with 1 supplement
Combination therapy with OSMI-1 and anti-PD-L1 augmented T cells and antitumor immunity.
(A) The extranuclear dsDNA were determined by anti-dsDNA fluorescence staining treated with 50 μM and 100 μM OSMI in MC38 cells respectively and was quantified by image J. (B–C) Western blot analysis of protein expression in MC38 and LLC cells treated with 50 μM and 100 μM OSMI, respectively. (D) Analysis of γH2AX and H2AX expression by anti-γH2AX staining treated with 50 μM and 100 μM OSMI in MC38 cells and was quantified by image J. (E–F) Tumor volume, weight of MC38 tumors injected with either control OSMI-1 or anti-PD-L1 in C57BL/6 J mice and mice survival. (G–H) Tumor volume, weight of LLC tumors injected with either control OSMI-1 or anti-PD-L1 in C57BL/6 J mice and mie survival. (I–K) Flow cytometry analysis showing percentage of CD4+ and CD8+ T cells population (I), and CD8+ IFN-γ+ cells (J), CD8+ TNF-α+ cells (K) population in MC38 subcutaneous tumor isolated at day 18 post-tumor inoculation. (L–N) Flow cytometry analysis showing percentage of CD4+ and CD8+ T cells population (L), and CD8+ IFN-γ+ cells (M), CD8+ TNF-α+ cells (N) population in LLC subcutaneous tumor isolated at day 18 post-tumor inoculation. Data are representative of three or four independent experiments. Statistical significance was determined by unpaired Student’s t-test, one-way ANOVA, two-way ANOVA, *p<0.05, **p<0.01, ***p<0.001, ns, no significant difference. Data represent the mean of ± SD.
Figure 7—figure supplement 1
OSMI-1 could significantly induce a high percentage of DNA damage.
(A–B) The cell proliferation in different treatments in vitro. (A) MC38 cell proliferation, (B) LLC cell proliferation. (C) The extranuclear dsDNA was measured by anti-dsDNA fluorescence staining treated with 50 μM and 100 μM in LLC cells, respectively. (D) The γH2AX expression was measured anti-γH2AX fluorescence staining treated with 50 μM and 100 μM in LLC cells, respectively. Data are representative of three experiments. Statistical significance was determined by one-way ANOVA, *p<0.05, **p<0.01, ***p<0.001, ns, no significant difference. Data represent the mean of ± SD. Schematic. A schematic of critical role of OGT-mediated antitumor immunity. Inhibition of O-GlcNAc transferase promotes the activation of cGAS-STING pathway and the production of type I interferon, which enhances CD8 T cells dependent antitumor immunity.
Additional files
-
Supplementary file 1
Primer sequences for genotype, RT-PCR and molecular cloning.
(a) Primer sequences for genotype. (b) Related to Experimental Procedures. Primer sequences for RT-PCR. (c) Related to CRISPR/Cas9. Primer sequences for molecular cloning.
- https://cdn.elifesciences.org/articles/94849/elife-94849-supp1-v1.docx
-
Supplementary file 2
Mass spectrometry assay of OGT interactome.
- https://cdn.elifesciences.org/articles/94849/elife-94849-supp2-v1.xlsx
-
MDAR checklist
- https://cdn.elifesciences.org/articles/94849/elife-94849-mdarchecklist1-v1.docx
Download links
A two-part list of links to download the article, or parts of the article, in various formats.
Downloads (link to download the article as PDF)
Open citations (links to open the citations from this article in various online reference manager services)
Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)
Inhibition of O-GlcNAc transferase activates type I interferon-dependent antitumor immunity by bridging cGAS-STING pathway
eLife 13:RP94849.
https://doi.org/10.7554/eLife.94849.3
