Fig 2: Intrinsic Rig-I deficiency boosts antitumor activity of adoptively transferred CD8+ T cells.(A and B) 2 × 106 Naive CD8+ T cells from spleens of Rig-I+/+ or Rig-I–/– mice were i.v. transfused into Rag1–/– mice. Saline was used as a vehicle control. 2 × 105 MC38 cells were s.c. inoculated 1 day after T cell injection and tumor volume was measured on days 10, 13, 15, 19, and 21 (n = 6 per group). Experimental design (A) and tumor growth curve (B) are shown. (C) CD45+CD8+ cell number was counted using flow cytometry. (D and E) Annexin V, IFN-γ, and CD107a were detected by flow cytometry. The percentage of Annexin V+ cells (D) and IFN-γ+ or CD107a+ cells (E) in CD45+CD8+ cells are shown. (F and G) Immunofluorescence staining of tumor tissue was performed to detect the nuclei (DAPI; blue), CD8 (green) and granzyme B (red). 10 random fields of view for each group were analyzed. Representative images (F) and combined plots (G) of the percentages of GZMB+CD8+ T cells were shown. Data are representative of 2 independent experiments and expressed as mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001, by 2-way ANOVA (B) or unpaired Student’s t test (C–E and G).

Fig 3: Rig-I ablation strengthens survival and antitumor capacity of tumor-infiltrating CD8+ T cells.(A and B) qRT-PCR analysis of Rig-I mRNA level (A) and Western blotting analysis of RIG-I protein level (B) in CD45+CD8+ cells sorted from spleens or MC38 tumor-infiltrating lymphocytes of WT mice. (C and D) The percentages of Annexin V+ cells (C) or Ki67+ cells (D) in CD45+ CD8+ cells from MC38 tumor-infiltrating lymphocytes in Rig-I+/+ or Rig-I–/– mice were determined by flow cytometry. Representative cytometric plots and quantification plots are shown. (E) The protein levels of proapoptotic and antiapoptotic proteins and granzyme A were detected in freshly isolated tumor-infiltrating CD8+ T cells by Western blotting. The same batch of samples were run contemporaneously in different gels. (F–I) Indicated markers of MC38 tumor-infiltrating CD8+ T cells were determined by flow cytometry. Representative cytometric plots and quantification plots are shown. Data are representative of at least 2 independent experiments and expressed as mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, by unpaired Student’s t test.

Fig 4: RIG-I restrains CD8+ T cell survival and cytotoxicity by dampening HSP90-mediated protective effect on STAT5.(A) Indicated protein levels of splenic naive CD8+ T cells after 48 hours anti-CD3/CD28 stimulation were analyzed by Western blotting. (B) GSEA of RNA-Seq (same as Figure 6, D–G) comparing in vitro activated CD8+ T cells as indicated are shown. (C) Empirical cumulative distribution function for change in expression of all genes (black) expressed in in vitro activated Rig-I–/– CD8+ T cells (change relative to that in Rig-I+/+ CD8+ T cells) and for subsets of genes upregulated (red) or downregulated (blue) by overexpression of constitutively active STAT5 in CD8+ T cells. (D and E) Splenic naive CD8+ T cells were stimulated by anti-CD3/CD28 and STAT5-IN-1 (10 μM) for 2 days and analyzed by flow cytometry. (F and G) Cell lysates of in vitro stimulated CD8+ T cells were immunoprecipitated with dynabeads-coupled control IgG or RIG-I antibody (F). Cell lysates of naive Rig-I+/+ or Rig-I–/– CD8+ T cells stimulated with anti-CD3/CD28 for 2 days were immunoprecipitated with dynabeads-coupled control IgG or STAT5 antibody (G). The precipitates were then probed with indicated antibodies. (H) Lysates of 293T cells cotransfected with indicated plasmids were immunoprecipitated with HA antibody and analyzed by immunoblot with indicated antibodies. (I) Indicated protein levels of MC38 tumor-infiltrating CD45+CD8+ T cells were analyzed by Western blotting. (J) Mice s.c. inoculated with MC38 cells were treated with STAT5-IN-1. Schematic diagram and tumor growth curve are shown (n = 5 per group). Data are representative of at least 2 independent experiments and expressed as mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, by 1-way ANOVA (D and E) or 2-way ANOVA (J). The same batch of samples were run contemporaneously in different gels for A and I.