Cish actively silences TCR signaling in CD8+ T cells to maintain tumor tolerance

Abstract

Improving the functional avidity of effector T cells is critical in overcoming inhibitory factors within the tumor microenvironment and eliciting tumor regression. We have found that Cish, a member of the suppressor of cytokine signaling (SOCS) family, is induced by TCR stimulation in CD8(+) T cells and inhibits their functional avidity against tumors. Genetic deletion of Cish in CD8(+) T cells enhances their expansion, functional avidity, and cytokine polyfunctionality, resulting in pronounced and durable regression of established tumors. Although Cish is commonly thought to block STAT5 activation, we found that the primary molecular basis of Cish suppression is through inhibition of TCR signaling. Cish physically interacts with the TCR intermediate PLC-γ1, targeting it for proteasomal degradation after TCR stimulation. These findings establish a novel targetable interaction that regulates the functional avidity of tumor-specific CD8(+) T cells and can be manipulated to improve adoptive cancer immunotherapy.

Figure 1.

Cish induction is TCR stimulation dependent. (A) Induction of Cish gene expression using real-time PCR at indicated times after αCD3 stimulation of naive CD8⁺ T cells. Results shown as means ± SEM n = 3; two independent experiments. (B) Western Blot of Cish protein expression after αCD3 stimulation of naive CD8⁺ T cells at indicated times; three independent experiments. (C) Relative Cish mRNA expression in different CD8⁺ T cell subsets using real-time PCR; Naive (Tn; CD62L⁺CD44⁻), in vivo–derived central memory (T_cm; CD62L⁺CD44⁺), and effector memory (T_em; CD62L⁻CD44⁺) after vaccination. n = 3; three independent experiments. (D) Representative FACS blot of tumor-resident pmel-1 thy1.1⁺ CD8⁺ T cells 7 d after ACT. n = 3; two independent experiments. (E and F) Representative FACS blot of Cish expression by intracellular staining of tissue-resident pmel-1 CD8⁺ T cells in antigen-negative nondraining axillary lymph node (Irr LN) or antigen-positive tumor 7 d after ACT. *, P < 0.05 by unpaired Student’s t test. n = 3; two independent experiments. (G) Schematic of Cish knockout targeting construct (B, BamHI; E, EcoRI; H, HindIII; P, PstI; X, XhoI). (H) PCR confirmation of genotype. Gel electrophoresis of DNA products after PCR amplification. n = 300+. (I) Representative FACS blot of CD8⁺ and CD4⁺ thymocytes from Cish^−/− or WT mice, enumerated in J. Values represent mean ± SEM. P > 0.05 by unpaired Student’s t test. n = 3; three independent experiments. (K) Enumeration and flow cytometric evaluation of CD4⁺ and CD8⁺ splenocytes for each genotype. Values represent mean ± SEM. P > 0.05 by unpaired Student’s t test. n = 3; three independent experiments.

Figure 2.

Cish inhibits CD8⁺ T cell expansion and functional avidity. (A) Phenotypic memory analysis of CD8⁺ T cells from Cish^−/− and WT pmel-1 littermate mice using flow cytometry. Values represent mean ± SEM. P > 0.05 by unpaired Student’s t test. n = 3; three independent experiments. (B) Enumeration of T cell expansion 6 d after stimulation with hgp100_25-33 peptide-pulsed splenocytes. Values represent mean ± SEM. **, P < 0.01 by unpaired Student’s t test. n = 3; three independent experiments. (C) Evaluation of apoptosis of primed Cish^−/− or WT CD8⁺ T cells as assessed by flow cytometric evaluation of 7-AAD and Annexin V staining after αCD3 restimulation at times indicated. n = 3; three independent experiments. (D–F) Enzyme-linked immunosorbent assay (ELISA) evaluation of IFN-γ, TNF or IL-2 in supernatants after an overnight co-culture of primed Cish^−/− or WT pmel-1 CD8⁺ T cells with peptide-pulsed splenocytes. Values are shown as means ± SEM. ****, P < 0.0001 by two-way ANOVA. n = 3; three independent experiments. (G) Representative FACS blot of intracellular IFN-γ, TNF, and IL-2 in Cish^−/− or WT pmel-1 CD8⁺ T cells 6 h after αCD3 stimulation. n = 3; three independent experiments. (H) Assessment of cytokine polyfunctionality in Cish^−/− or WT pmel-1 CD8⁺ T cells from (G) using Boolean gating strategy. Data are shown as total percentage of two or three concomitant cytokines present after antigenic stimulation. n = 3; three independent experiments.

Figure 3.

Cish deletion augments CD8⁺ T cell–mediated tumor immunity. (A–E) 7 d after s.c. implantation of 5 × 10⁵ B16 melanoma cells, C57BL/6 tumor-bearing mice received the adoptive transfer (ACT) of 10⁶ Cish^−/−, WT pmel-1 naive CD8⁺ T cells, or no cells (NT) in conjunction with rhgp100 VV (10⁷ pfu) and IL-2 (2 × 10⁶ IU BID for 3 d). Tumor progression, survival, and expansion of pmel-1 T cells were followed. (A) Tumor growth as assessed by measurement of the perpendicular diameters over time. Values are shown as means ± SEM. ****, P < 0.0001 by two-way ANOVA. n = 5 per group; 10 independent experiments. (B) Kaplan-Meier survival curve after ACT in A. **, P < 0.01 by Log-rank test for trend. n = 5–6 per group; 10 independent experiments. (C) Induction of ocular autoimmunity 6 d after ACT as assessed by combinatorial scoring of severity of iridiocyclitis, vitritis, and choroiditis as described in Materials and methods. ****, P < 0.0001. n = 6; two independent experiments. (D) In vivo growth kinetics of congenically marked T cells after ACT into C57BL/6 recipients as assessed by flow cytometry in conditions identical to (A). Values are shown as means ± SEM. ****, P < 0.0001 by two-way ANOVA. n = 3; three independent experiments. (E) Ex vivo Annexin V staining by flow cytometry of congenically marked Cish^−/− or WT pmel-1 T cells from the spleen 6 d after ACT. *, P < 0.05 by unpaired Student’s t test. n = 3; two independent experiments.

Figure 4.

Cishdeletion augments long-term CD8⁺ T cell–intrinsic tumor killing. (A–D) Growth of B16 melanoma in Rag1^−/− tumor-bearing hosts after ACT of 2.5 × 10⁵ indicated pmel-1 CD8⁺ T cells in conjunction with rhgp100 VV (3 × 10⁶ pfu) and IL-2 (2 × 10⁵ IU BID for 3 d) or no cell transfer (NT). (A) Tumor growth as assessed by measurement of the perpendicular diameters over time. Values are shown as means ± SEM. ****, P < 0.0001 by two-way ANOVA. n = 5 per group; three independent experiments. (B) Outgrowth of amelanotic tumors from A in five out of five mice, 50+ d after ACT; two independent experiments. (C) IFN-γ production assessed by ELISA 7 d after ACT, after co-culture with ex vivo enriched congenically marked Cish^−/− or WT pmel-1 T cells with cognate peptide-pulsed splenocytes. T cells were normalized for cell number. Values are shown as means ± SEM. ****, P < 0.0001 by two-way ANOVA. n = 3; two independent experiments. (D) Growth of B16 melanoma in Rag1^−/− tumor-bearing hosts as assessed by measurement of the perpendicular diameters over time after treatment. ACT of 2.5 × 10⁵ Cish^−/− pmel-1 CD8⁺ T cells or no cell transfer (NT) with or without administration of αCD8-depleting antibody 38 d after transfer. Values are shown as means ± SEM. **, P < 0.01 by two-way ANOVA. n = 5; two independent experiments.

Figure 5.

Knockdown of Cish in mice and man confers enhanced CD8⁺ T cell tumor reactivity. (A) IFN-γ evaluation by ELISA after an overnight co-culture of Cish shmiR or control shmiR retroviral transduced CD8⁺ pmel-1 WT T cells. Values are shown as means ± SEM. ****, P < 0.0001 by two-way ANOVA. n = 3; two independent experiments. (B) Growth of B16 melanoma in C57BL/6 tumor-bearing hosts as assessed by measurement of the perpendicular diameters over time after treatment. ACT of 1 × 10⁶ of retroviral transduced Cish shmiR WT, control shmiR WT, or control Cish^−/− pmel-1 T cells in conjunction with rhgp100 VV (3 × 10⁶ pfu) and IL-2 (2 × 10⁵ IU BID for 3 d) or no cell transfer (NT). Values are shown as means ± SEM *, P < 0.05 by two-way ANOVA. n = 5; two independent experiments. (C–E) IFN-γ evaluation by intracellular staining of patient PBL transduced with various tumor-specific TCRs, knocked down for CISH, and co-cultured with tumor for 6 h and assessed by flow cytometry. Retroviral transduced PBL with CISH shmiR or control shmiR and cotransduced with previously described TCR-encoding retroviruses directed against cancer-testis antigen, NY-ESO-1 (C), or MART-1 specific highly avid DMF5 (D) or poorly avid DMF4 TCR (E). 5 d after transduction, PBL were co-cultured with antigen-positive or -negative tumors and assessed for intracellular IFN-γ by flow cytometry. *, P < 0.05 by paired Student’s t test. n = 3–5; three independent experiments. (F) Transduced PBL from D were co-stained for IFN-γ, TNF, and IL-2 and evaluated for polycytokine functionality by flow cytometry and using a Boolean gating strategy. *, P < 0.05 by paired Student’s t test. n = 3; three independent experiments. (G) IFN-γ evaluation by ELISA of patient PBL knocked down for CISH and co-cultured overnight with relevant and irrelevant tumors. Retroviral transduced PBL with CISH shmiR or control shmiR and cotransduced with TCR-encoding retroviruses directed against cancer-testis antigens, MAGE-A1. 5 d after transduction, PBL were co-cultured with antigen-positive or -negative tumors. *, P < 0.05 by paired Student’s t test. n = 3; two independent experiments.

Figure 6.

No overt changes in STAT5 activation in the absence of Cish at acute time points. (A) Western blot of Cish, phosphorylated STAT5, and β-actin in naive Cish^−/− or WT pmel-1 T cells after αCD3 stimulation at times indicated; two independent experiments. (B) Schematic of STAT5 reporter self-inactivating (sin) retrovirus coexpressing the congenic marker Thy1.1 under the phospholgycerate kinase 1 promoter. (C) Relative luciferase activity in STAT5 reporter-transduced Cish^−/− or WT-primed pmel-1 T cells after αCD3 stimulation at times indicated. Values are shown as means ± SEM. P > 0.05 by two-way ANOVA. n = 3; two independent experiments. (D) Flow cytometric evaluation of intracellular phosphorylated STAT5 in primed Cish^−/− or WT pmel-1 T cells after the addition of IL-2 at indicated concentrations and times. P > 0.05 by two-way ANOVA. n = 3; three independent experiments.

Figure 7.

TCR-dependent hyperactivation program in the absence of Cish. (A) Enhanced up-regulation of relative Tbx21, Cmyc, and Bcl2l1 gene expression by real-time PCR with or without αCD3 stimulation for 4 h. Values are shown as means ± SEM; n = 3. *, P < 0.05; **, P < 0.01 by two-way ANOVA. n = 3; two independent experiments. (B) Microarray volcano plot comparing in vitro–primed Cish^−/− or WT CD8⁺ T cells 2 h after αCD3-stimulation. n = 3 independent mice per genotype; two independent experiments. (C and D) Gene-set enrichment analysis of genes in Cish-deficient T cells relative to WT littermates reveals a strong ranking with the Goldrath antigen response (C) and Kaech naive versus day 8 effector T cell down-regulated (D) profile.

Figure 8.

Enhanced PLC-γ1 activation and downstream signaling in the absence of Cish. (A) Cish protein evaluation WT or Cish^−/− CD8⁺ T cells with or without TCR stimulation by immunoblotting; three independent experiments. (B) Total phospho-Tyrosine (pTyr) blotting of cell lysates after αCD3 stimulation of Cish^−/− or WT CD8⁺ T cells at times indicated by immunoblotting; three independent experiments. (C) Western blot analysis of phospho-PLC-γ1, total PLC-γ1, phosphor-LAT, whole LAT, phospho-Zap-70, whole Zap-70, and β-actin blot after αCD3 stimulation of Cish^−/− or WT CD8⁺ T cells at times indicated; three independent experiments. (D) Representative Ca²⁺ flux as assessed by fluorometric evaluation after αCD3 stimulation of Cish^−/− or WT CD8⁺ T cells. Kinetic of the ratio of Fluo3-AM by Fura Red over time shown and assessed by flow cytometry; five independent experiments. (E and F) Relative luciferase activity of NFAT and NF-κB reporter transduced Cish^−/− or WT CD8⁺ T cells after αCD3 stimulation. ***, P < 0.001 by two-way ANOVA. n = 3; two independent experiments.

Figure 9.

Cish specifically inhibits Ca²⁺ flux, T cell polyfunctionality, and PLC-γ1 accumulation in TCR microclusters. (A) Schematic of retroviral Cish expression vector. N-terminal Flag-tagged (3×) Cish, self-cleaving furin-2A (f2A) peptide, and congenic marker Thy1.1 driven by the intrinsic LTR promoter. (B) Representative Ca²⁺ flux as assessed by flow cytometry after αCD3 stimulation of Cish, empty control transduced Cish^−/− CD8⁺ T cells, or empty control transduced WT CD8⁺ T cells. Kinetic of the ratio of Fluo3-AM by FuraRed over time shown and assessed by flow cytometry; four independent experiments. (C) Assessment of functional avidity by intracellular IFN-γ using flow cytometry. Cish or empty control transduced Cish^−/− CD8⁺ T cells or empty control transduced WT CD8⁺ T cells from B were stimulated with indicated αCD3 concentrations for 6 h and evaluated by flow cytometry. Values represent mean fluorescence intensity. **, P < 0.01 by paired Student’s t test; three independent experiments. (D) Assessment of cytokine polyfunctionality in T cells from B using Boolean gating strategy. Data are shown as total percentage of two or three concomitant cytokines present after antigenic stimulation. n = 3; three independent experiments. (E and F) Evaluation of PLC-γ1 and phosphotyrosine in TCR microclusters after αCD3 stimulation. (E) Representative confocal images from transduced CD8⁺ T cells from B were dropped on to stimulatory coverslips, fixed after three minutes, immunostained for PLC-γ1 and phosphotyrosine (pTYR), and enumerated for intensity and area of each (F). Bars, 2 µm. ns, P > 0.05; **, P < 0.01 by unpaired Student’s t test. n = 18–24; three independent experiments.

Figure 10.

TCR stimulation–dependent polyubiquitination of PLC-γ1 by Cish. (A) Immunoprecipitation of YFP-tagged PLC-γ1 and immunoblotting of Cish in transfected 293T cells in the absence of TCR signaling complex; two independent experiments. (B) Immunoprecipitation of FLAG-tagged Cish and immunoblotting of PLC-γ1 in indicated transduced CD8⁺ T cells with and without CD3 stimulation (5 min). Whole lysates were blotted for PLC-γ1 and Cish; two independent experiments. (C) Immunoprecipitation of endogenous Cish and immunoblotting of PLC-γ1 in 3 d Cish^−/− or WT CD8⁺ T cell blasts; two independent experiments. (D) 293T cells were transfected with Tagged plasmids expressing PLC-γ1-YFP, Ubiquitin-HA, and Cish-FLAG where indicated in the presence of the proteasome inhibitor, MG-132. After transfection, PLC-γ1 was immunoprecipitated and blotted for HA and YFP. Whole-cell lysates were blotted for Cish and PLC-γ1-YFP; two independent experiments. (E) Immunoprecipitation of endogenous PLC-γ1 and immunoblotting of ubiquitin in indicated reconstituted CD8⁺ T cells with or without TCR stimulation in the presence of the proteasome inhibitor, MG-132; two independent experiments. (F) Immunoprecipitation of endogenous PLC-γ1 and immunoblotting of ubiquitin in native CD8⁺ T cells blasts (3 d) with or without TCR stimulation in the presence of the proteasome inhibitor, MG-132; two independent experiments.