Deficiencies of the T cell and NK cell CD3ζ signaling adapter protein in patients with cancer and autoimmune diseases are well documented, but mechanistic explanations are fragmentary. The stimulatory NKG2D receptor on T and NK cells mediates tumor immunity but can also promote local and systemic immune suppression in conditions of persistent NKG2D ligand induction that include cancer and certain autoimmune diseases. In this paper, we provide evidence that establishes a causative link between CD3ζ impairment and chronic NKG2D stimulation due to pathological ligand expression. We describe a mechanism whereby NKG2D signaling in human T and NK cells initiates Fas ligand/Fas-mediated caspase-3/-7 activation and resultant CD3ζ degradation. As a consequence, the functional capacities of the TCR, the low-affinity Fc receptor for IgG, and the NKp30 and NKp46 natural cytotoxicity receptors, which all signal through CD3ζ, are impaired. These findings are extended to ex vivo phenotypes of T and NK cells among tumor-infiltrating lymphocytes and in peripheral blood from patients with juvenile-onset lupus. Collectively, these results indicate that pathological NKG2D ligand expression leads to simultaneous impairment of multiple CD3ζ-dependent receptor functions, thus offering an explanation that may be applicable to CD3ζ deficiencies associated with diverse disease conditions.

The stimulatory NKG2D receptor and its cell stress-inducible MHC class I-related chain A and B (MICA and MICB) ligands are mediators of protective tumor and infectious disease immunity, yet also promote tumor immune evasion and systemic immune suppression (1). Human NKG2D in complex with the DAP10 signaling adaptor protein is expressed on most CD8 T cells and NK cells, and some CD4 T cells. Depending on the functional differentiation state of a given lymphocyte, NKG2D either costimulates or directly activates T and NK cells (27). Expression of the MIC ligands of NKG2D is tightly regulated to flag only diseased cells and tissues for NKG2D-mediated lymphocyte control (8). Accordingly, MIC ligands are absent from the surface of most normal cells but are induced by conditions of cellular stress caused by malignant transformation, infection, and inflammation (4, 5, 913).

As with other immunoreceptors, NKG2D, together with DAP10, is subject to ligand-induced internalization and lysosomal degradation to prevent chronic lymphocyte stimulation (1417). The resultant NKG2D downmodulation is transient except for conditions that provide for massive and persistent MIC ligand expression, as in advanced cancers and certain autoimmune diseases (5, 1012). In addition to locally affecting NKG2D, tumors and autoimmune disease target tissues shed soluble MICA, thereby causing systemic NKG2D downmodulation (14, 18). Consequently, the presence of MIC in advanced tumors is a reflection of host tumor susceptibility and correlates with negative disease outcomes (19).

A number of observations suggest that sustained NKG2D ligand expression may have broader immunosuppressive effects. Unlike normal peripheral blood NK cells, those from patients with colorectal cancers producing soluble MICA lack the natural cytotoxicity receptor (NCR) NKp46 (15). With mouse NK cells, sustained NKG2D engagement negatively affects CD16, NK1.1, and NKp46 functions (20). In addition to loss of NKG2D and its alternative DAP10 and DAP12 signaling adaptors, mouse NK cells also loose the CD3ζ signaling adaptor, which is irrelevant for NKG2D (21). In humans, CD3ζ is associated with low-affinity Fc receptor for IgG (FcγRIII, CD16) and is essential for NKp30 and NKp46 functions (22, 23). Because CD3ζ is also indispensable for TCR expression and function, its NKG2D-initiated downmodulation could have profound T cell-impairing effects (2427).

Loss of or diminished CD3ζ expression and function is common in T and NK cells from patients with cancer, autoimmune diseases, and infections, and during pregnancy—conditions that are frequently or typically associated with persistent MIC expression (1, 28, 29). Hence these associations, together with the evidence obtained with mouse NK cells, suggest that the CD3ζ deficiencies could be triggered by recurring and widespread ligand engagements of NKG2D.

In this study, we provide evidence that stimulation of human T and NK cells through NKG2D in the context of persistent ligand exposure precipitates functional impairments of CD3ζ-associated immune receptors. Mechanistically, our results implicate a chain of events including NKG2D downmodulation, paracrine Fas ligand (FasL) production, caspase activation, and finally, caspase-mediated CD3ζ cleavage. Our in vitro data correlate with phenotypes observed with T and NK cells from cancer patients and patients with juvenile-onset systemic lupus erythematosus (SLE), thus establishing an NKG2D-initiated mechanism that may promote far-reaching lymphocyte tolerization.

Peripheral blood samples from healthy adult volunteers, pediatric SLE patients, and age-matched control subjects were procured at the Fred Hutchinson Cancer Research Center and Seattle Children’s Hospital. Tumor samples were provided by the Cooperative Human Tissue Network (CHTN). All activities were approved by local Institutional Review Boards. PBMCs were isolated by density centrifugation on Histopaque-1077 (Sigma-Aldrich, St. Louis, MO). Tumor-infiltrating lymphocytes (TILs) were isolated as described previously (10). Aliquots of the SLE patient samples were previously analyzed for soluble MICA (30).

CD8 T cells and NK cells were purified (>98%) from healthy donor-derived PBMCs using MACS microbead cell separation methodology (Miltenyi Biotec, Auburn, CA) and phenotypes confirmed by flow cytometry. NKG2D-licensed T cell (lines 1–8) and NK cell lines (lines 1–3) were established and maintained as described previously (7, 31). In brief, bead-purified T cells or NK cells were cultured in serum-free AIM V medium (Invitrogen, Carlsbad, CA) in the presence of irradiated allogeneic PBMCs and 100 IU/ml IL-2 (Proleukin; Novartis Pharmaceuticals, East Hanover, NJ). After two stimulations at weekly intervals, the cell lines were confirmed for expression of CD3, CD8, and NKG2D, and CD16, CD56, and NKG2D, respectively. The HLA-A2–restricted and MART-1–specific NKG2D+CD8 T cell clones 3, 7, 39, and 80 were established as described previously (32). The TALL-104 leukemia, A375 melanoma, and P815 mouse mastocytoma lines were from the American Type Culture Collection (ATCC) and cultured according to recommendations. LTk-MICA*001, C1R–HLA-A2, C1R–HLA-A2–MICA, and MICB and UL16-binding proteins 1–5 (ULBP1-5) transfectants have been described previously (33).

Ex vivo purified peripheral blood CD8 T cells or NK cells, NKG2D-licensed T cell or NK cell lines, or TALL-104 cells were cultured for 4 d in the presence of adherent LTk-MICA or mock transfectants at a lymphocyte/transfectant ratio of 2:1 in AIM V medium, and T or NK cells harvested at daily intervals. Using the adherent LTk cells facilitated essentially pure lymphocyte harvesting. To correct for possible transfectant cell loss caused by cytotoxic lymphocyte activities, we plated the LTk cells without prior irradiation. Some experiments used purified recombinant soluble MICA at a concentration of 20 ng/ml instead of transfectants (34). NKG2D ligand-masking mAb were anti-MICA/B mAb 6D4 (35) and anti–ULBP1-5 mAb (3F1, 6F6, 4A10, 6E6, and 6D10, all at 10 μg/ml) (36). To assess phenotypic and functional recovery, we cultured 96-h LTk-MICA–exposed T cells for an additional 24 h in the absence of ligand. The MART-1–specific CD8 T cell clones were exposed to M27-peptide pulsed C1R–HLA-A2 or C1R–HLA-A2–MICA transfectants as described previously (14). Solid-phase Ab-mediated stimulations used goat anti-mouse IgG [Fc fragment-specific, F(ab′)2 (10 μg/ml, Jackson ImmunoResearch Laboratories, West Grove, PA)] and anti-NKG2D (clone 1D11, 5–50 ng/ml) (2) alone or together with anti-CD3 (OKT3, 10 ng/ml). Isotype-matched control Ig was used as control. Caspase inhibitors were benzyloxycarbonyl Asp-Glu-Val-Asp (zDEVD)–fluoromethyl ketone (FMK; 20 μM; R&D Systems, Minneapolis, MN) and quinolyl-valyl-O-methylaspartyl-[-2,6-difluorophenoxy]-methyl ketone (Q-VD-OPh; 50 μΜ; R&D Systems), and benzyloxycarbonyl Val-Ala-Asp (zVAD)–FMK (20 μM; Calbiochem, San Diego, CA). Neutralizing anti-FasL (clone NOK-1; BD Biosciences, San Jose, CA) and antagonist anti-Fas/APO-1 (clone SM1-23; Bender MedSystems, Vienna, Austria) were each used at 10 μg/ml.

For surface stainings, T or NK cells were treated with PBS/10% human serum/0.15% sodium azide for 15 min at room temperature, followed by incubation with pretitrated mAb cocktails consisting of combinations of anti-CD3–Alexa Fluor 700 (clone UCHT1), anti-CD4–allophycocyanin–Alexa 750 (clone S3.2; Caltag Laboratories, Burlingame, CA), anti-CD8α–PerCPCy5.5 (clone SK1), anti-CD16–Pacific blue or –FITC (clone 3G8), anti-CD56–Pacific blue (clone MEM-188; Biolegend, San Diego, CA), anti-CD56–PE (clone MY31), anti-NKp30–PE (clone Z25; Beckman Coulter, Fullerton, CA), anti-NKp46–PE (clone BAB281; Beckman Coulter), and anti-NKG2D–allophycocyanin (clone 1D11), for 25 min at 4°C, and washed in PBS/1% BSA/0.15 sodium azide. Unless otherwise specified, mAbs were from BD Pharmingen (San Diego, CA). To determine CD3ζ expression, surface-stained T or NK cells were permeabilized, incubated with anti-CD3ζ–PE (clone 2H2D9; Beckman Coulter) or anti-CD3ζ–FITC (clone G3; Serotec, Oxford, United Kingdom), and washed using cytofix/cytoperm kit reagents (BD Biosciences). Detection of active caspases used caspase-2–specific fluorochrome FAM-Val-Asp-Val-Ala-Asp (VDVAD)-FMK, caspase-3/-7–specific FAM-DEVD-FMK, and caspase-8–specific FAM-Ileu-Glu-Thr-Asp (IETD)-FMK fluorochrome-conjugated cell-permeable inhibitor of caspase (FLICA) kits (Immunohistochemistry Technologies, Bloomington, MN) before surface and intracellular staining steps. Cell viability was assessed using a LIVE/DEAD Fixable Violet Dead Cell Stain Kit (Invitrogen). Apoptotic cell death was assessed with the Annexin V/7-aminoactinomycin D apoptosis detection kit (BD Biosciences). Phosphorylated ZAP70 was detected using anti–phospho-ZAP70–Alexa 488 (clone 17A; BD Biosciences) and Phosflow fix/perm kit reagents (BD Biosciences). Cells were examined with an LSR II (BD Biosciences) flow cytometer. Electronic compensation was with single stains. Cutoffs for background fluorescence were based on “fluorescence minus one” gating controls. Data were analyzed with FlowJo (Tree Star, Ashland, OR) software. Gating for each sample was based on forward scatter area versus a forward scatter height plot to eliminate aggregates and isolate lymphocytes, followed by exclusion of dead cells.

T cell lysates (106 cells) were prepared with LDS sample buffer, treated with reducing agent (both from Invitrogen), and boiled for 10 min at 70°C. Cleared supernatants were subjected to SDS-PAGE (4–12% bis-Tris NuPAGE gels; Invitrogen), and proteins electroblotted onto polyvinylidene difluoride membranes. Immunoblots were probed with Abs to CD3ζ (C-terminal–specific clone 8D3, BD Biosciences; N-terminal–specific clone 6B10.2; Santa Cruz Biotechnology, Santa Cruz, CA), anti–caspase-3 (clone 3G2), anti–caspase-7 (clone C7), and anti–β-Actin, and developed using HRP-conjugated secondary reagent. Unless otherwise specified, Abs were from Cell Signaling Technology (Beverly, MA). Detection was with ECL kit (GE Healthcare, Piscataway, NJ) or with SuperSignal West Dura Extended Duration Substrate (Thermo Fisher Scientific, Waltham, MA). For phospho-ZAP70 induction, T cells were serum starved for 30 h at 37°C in RPMI 1640 before exposure to anti-CD3 (OKT3; 1 μg/ml, 5 min at 4°C) or control Ig, followed by cross-linking with goat anti-mouse IgG F(ab′)2 (20 μg/ml; 10 min at 37°C; Jackson ImmunoResearch Laboratories). Reactions were terminated on ice and cells resuspended in lysis buffer containing fresh protease and phosphatase inhibitors (50 mM Tris-HCl; pH 7.5, 150 mM NaCl, 1% Triton-X100, 1 mM EDTA, 1 mM Na3VO4, 1 mM NaF, and protease inhibitor mixture). Cleared supernatants were subjected to SDS-PAGE under reducing conditions in 4–12% bis-Tris NuPAGE gels, transferred to polyvinylidene difluoride membranes, and probed with anti–phospho-ZAP70 (pY319) and donkey anti-rabbit IgG-HRP. For loading controls, stripped blots were reprobed with anti-ZAP70 (clone 99F2). Abs were from Cell Signaling Technology.

Total RNA was isolated from T cells with the RNeasy Plus Mini Kit (Qiagen, Valencia, CA) and cDNA synthesized using the Superscript First-Strand Synthesis System for RT-PCR (Invitrogen). Real-time RT-PCR used Platinum quantitative PCR supermix with uracil DNA glycosylase (Invitrogen) in low-profile 100-μl tubes in a StepOnePlus Instrument (Applied Biosystems, Foster City, CA). CD3ζ cDNA and 18S rRNA were amplified using commercial TaqMan System primers (Hs00167901_m1 and 433760T; Applied Biosystems). 18S rRNA was used as endogenous loading control. The comparative threshold cycle method was used to determine relative gene expression levels.

Calcium flux was measured as described previously (37). In brief, T cells were labeled with 2 μM Fura red, 1 μM Fluo-4, and 0.02% Pluronic F-127 (Invitrogen) for 30 min at room temperature in the dark, PBS-washed and incubated with anti-CD3 (OKT3, 1 μg/ml) or control Ig for 30 min on ice, washed again in cold PBS, and transferred to prewarmed HBSS/1 mM calcium. CD3 was cross-linked with goat anti-mouse IgG F(ab′)2 (20 μg/ml, Jackson Immunoresearch Laboratories). Samples were examined with an LSR II flow cytometer. Calcium mobilization was determined by measuring the ratio of Fluo-4 (FL1)/Fura red (FL3) using FlowJo software.

Redirected lysis assays used anti-CD16, anti-NKp30 (clone 210845; R&D Systems), and anti-NKp46 and anti-2B4 (clones BAB281 and C1.7; Beckman Coulter), together with FcγR+ P815 target cells (7). Cytotoxicity assays with MART-1–specific T cell clones and M27 peptide-pulsed C1R–HLA-A2 or A375 target cells were as described previously (14). Assays were carried out in triplicate and results scored as percentage of specific lysis, according to the standard formula. Commercial ELISA (R&D Systems) was used to quantitate FasL in T cell culture supernatants. For statistical analysis, Student t test was used to determine significance levels when comparing two groups. The p values <0.05 were considered significant.

To determine whether sustained NKG2D engagement affects CD3ζ protein expression in human NKG2D+CD8 T cells, ex vivo isolated normal peripheral blood CD8 T cells were exposed to LTk-MICA fibroblast transfectants or mock-transfected controls for 1 to 4 d and examined for intracellular CD3ζ and surface CD3 and NKG2D by flow cytometry. Using the adherent LTk transfectants as ligand-expressing matrix allowed for essentially pure T cell harvesting. Consistent with published results, surface NKG2D decreased over time (Fig. 1A, top panel) (14). However, as opposed to IL-2–activated murine NK cells, we observed no loss of CD3ζ expression in the human CD8 T cells (Fig. 1A, top panel) (21). This discrepancy suggested a need for signaling-competent NKG2D to induce the CD3ζlow/− phenotype, as, distinct from activated NK cells, NKG2D cross-linking alone is insufficient and an additional TCR signal is needed to trigger full activation of resting CD8 T cells (4, 6, 7).

FIGURE 1.

NKG2D-initiated CD3ζ downmodulation in CD8 T cells. A, Flow cytometry dot plots show gradual NKG2D downmodulation and unaltered anti-CD3ζ Ab reactivity in peripheral blood-derived NKG2D+CD8 T cells cocultured ex vivo with LTk-MICA transfectants during the time period indicated (top panels). Data shown are representative of duplicate experiments with each three healthy donor samples. Middle and bottom panels show gradually increasing NKG2D and associated CD3ζ downmodulation in NKG2D-licensed CD8 T cells (line 1) and TALL-104 cells during the coculture with LTk-MICA transfectants. Data shown are representative of separate experiments with eight IL-2–primed NKG2D+CD8 T cell lines and of four repeat experiments with TALL-104 cells. Dot plots were derived from a gating tree based on exclusion of LIVE/DEAD fixable violet staining and inclusion of surface CD3. Numbers in graphs specify proportions of CD3ζlow/− cells among total live CD8 T cells (%). B, Mean frequencies (%) of live CD3ζlow/− CD8 T cells recorded by flow cytometry at the indicated time points in separate experiments with the NKG2D-licensed T cell lines 1–8, and in four repeat experiments with TALL-104 cells, cocultured with either LTk-MICA or negative control cells. C, Mean frequencies (%) of live CD3ζlow/− CD8 T cells recorded by flow cytometry at the indicated time points in separate experiments with the NKG2D-licensed T cell lines 1–4, and in three repeat experiments with TALL-104 cells, exposed to solid-phase anti-NKG2D (mAb 1D11) or control Ig. B and C, Error bars indicate SD. Asterisks above bars refer to p values indicated in bar graphs. D, Immunoblots with C-terminal–specific anti-CD3ζ mAb reveal gradually decreasing amounts of the native 16-kDa CD3ζ protein in the NKG2D-licensed T cell line 1 exposed to LTk-MICA cells, and in anti-NKG2D (mAb 1D11) Ab–cross-linked TALL-104 cells. β-actin is shown as loading control. Data are representative of four T cell lines tested and of three repeat experiments with TALL-104 cells.

FIGURE 1.

NKG2D-initiated CD3ζ downmodulation in CD8 T cells. A, Flow cytometry dot plots show gradual NKG2D downmodulation and unaltered anti-CD3ζ Ab reactivity in peripheral blood-derived NKG2D+CD8 T cells cocultured ex vivo with LTk-MICA transfectants during the time period indicated (top panels). Data shown are representative of duplicate experiments with each three healthy donor samples. Middle and bottom panels show gradually increasing NKG2D and associated CD3ζ downmodulation in NKG2D-licensed CD8 T cells (line 1) and TALL-104 cells during the coculture with LTk-MICA transfectants. Data shown are representative of separate experiments with eight IL-2–primed NKG2D+CD8 T cell lines and of four repeat experiments with TALL-104 cells. Dot plots were derived from a gating tree based on exclusion of LIVE/DEAD fixable violet staining and inclusion of surface CD3. Numbers in graphs specify proportions of CD3ζlow/− cells among total live CD8 T cells (%). B, Mean frequencies (%) of live CD3ζlow/− CD8 T cells recorded by flow cytometry at the indicated time points in separate experiments with the NKG2D-licensed T cell lines 1–8, and in four repeat experiments with TALL-104 cells, cocultured with either LTk-MICA or negative control cells. C, Mean frequencies (%) of live CD3ζlow/− CD8 T cells recorded by flow cytometry at the indicated time points in separate experiments with the NKG2D-licensed T cell lines 1–4, and in three repeat experiments with TALL-104 cells, exposed to solid-phase anti-NKG2D (mAb 1D11) or control Ig. B and C, Error bars indicate SD. Asterisks above bars refer to p values indicated in bar graphs. D, Immunoblots with C-terminal–specific anti-CD3ζ mAb reveal gradually decreasing amounts of the native 16-kDa CD3ζ protein in the NKG2D-licensed T cell line 1 exposed to LTk-MICA cells, and in anti-NKG2D (mAb 1D11) Ab–cross-linked TALL-104 cells. β-actin is shown as loading control. Data are representative of four T cell lines tested and of three repeat experiments with TALL-104 cells.

Close modal

Because TCR engagement itself impacts CD3ζ expression, NKG2D-mediated effects on CD3ζ were tested using eight high-dose IL-2–activated NKG2D+CD8 T cell lines (hereafter referred to as NKG2D-licensed T cells) from healthy donors and the NKG2D+ TALL-104 T cell leukemia line, which can be activated through NKG2D alone (5, 38, 39). As with the ex vivo peripheral blood CD8 T cells, the T cell lines were exposed to the LTk-MICA transfectants and examined by flow cytometry at initiation of coculture and 24-h intervals thereafter. With these NKG2D-licensed T cells and TALL-104 cells, exposure to LTk-MICA but not to negative control cells resulted in both gradually decreasing surface NKG2D and corresponding CD3ζ loss (Fig. 1A, 1B and data not shown). T cells with diminished or absent CD3ζ appeared in small numbers during the first 2 d of coculture and increased to more prominent populations after 3 d (Fig. 1A, 1B). The extent of CD3ζ reduction was highly reproducible among the eight NKG2D-licensed T cell lines and in repeat experiments with TALL-104 (Fig. 1B). Surface CD3 remained mostly unchanged (data not shown). Varying degrees of NKG2D downmodulation and associated CD3ζ loss may reflect heterogeneity in regard to activation status and sensitivity to direct NKG2D stimulation within T cell lines (Fig. 1A).

In addition to MICA, NKG2D interacts with the closely related MICB and ULBP family ligands (40, 41). On testing of transfectants expressing MICB or individual ULBP (members 1 through 5), CD3ζ downmodulation was similarly induced by all of these ligands. The same effect was observed after T cell exposure to recombinant soluble MICA and was abrogated in the presence of anti-NKG2D ligand masking mAb (Supplemental Fig. 1 and data not shown). Ab-mediated NKG2D cross-linking also resulted in CD3ζ loss, which was less pronounced presumably because of lower stability of the anti-NKG2D mAb compared with MICA transfectants under the long-term cell culture conditions (Fig. 1C). The CD3ζlow/− flow cytometry phenotype was biochemically confirmed by immunoblot analyses of T cells harvested at daily intervals during a 96-h time course of NKG2D stimulation with LTk-MICA cells or cross-linked mAb (Fig. 1D).

CD3ζ is indispensable for TCR signal transduction (25, 26, 42). We thus tested whether TCR function was compromised in the CD3ζlow/− T cells. Assessment of calcium release after CD3 mAb cross-linking revealed markedly impaired responses in T cells exposed for 96 h to LTk-MICA transfectants as compared with control cells (Fig. 2A, left panels). Consistent with the reduced calcium response, CD3-induced phosphorylation of ZAP70, the protein tyrosine kinase downstream of CD3ζ, was also diminished as shown by immunoblot (Fig. 2B) (26). Phospho-flow cytometry of CD3-induced ZAP70 phosphorylation in CD3ζlow/− and CD3ζhigh T cells confirmed and extended these results by associating CD3ζ downmodulation with TCR functional impairment at the single-cell level (Fig. 2C).

FIGURE 2.

NKG2D-initiated functional impairment of CD3ζ-dependent T cell and NK cell receptors. A, Calcium release (expressed as Fluo-4/Fura red ratio over time in seconds) in response to CD3 cross-linking is attenuated in the NKG2D-licensed T cells (line 1; top left graph) or TALL-104 cells (bottom left graph) after 96 h of exposure to LTk-MICA but not to negative control LTk-neo cells. Calcium release is fully restored after an additional 24 h of culture in the absence of NKG2D ligand (right graphs). Data shown are representative of three T cell lines tested and of duplicate TALL-104 cell experiments. B, Anti–phospho-ZAP70 immunoblot reveals reduced anti-CD3 mAb-induced ZAP70 phosphorylation in the NKG2D-licensed CD8 T cells (line 4) after 96 h of exposure to LTk-MICA, but not to negative control cells. Total ZAP70 confirms equal loading. Data are representative of three T cell lines tested. C, Phospho-flow histograms display differences in CD3-induced ZAP70 phosphorylation between CD3ζlow/− (left histogram) and CD3ζhigh (right histogram) T cells (line 7) after 72 h of exposure to LTk-MICA cells. The top CD3ζ histogram indicates the applied gating. Data shown are representative of experiments with three T cell lines. D, Reduced CD16- and NKp46-mediated redirected lysis of P815 mastocytoma cells by NKG2D-licensed T cells (lines 3 and 6) after exposure for 96 h to LTk-MICA, but not to LTk-neo control cells. Impaired responses are restored after T cell recovery for 24 h in fresh media in the absence of MICA. Data shown are representative of three repeat experiments. E, Reduced CD16-, NKp46-, and NKp30-mediated redirected lysis of P815 mastocytoma cells by NKG2D-licensed NK cells (line 1) after exposure for 96 h to LTk-MIC, but not to LTk-neo control cells. Impaired responses are restored after NK cell recovery for 24 h in fresh media in the absence of MICA. 2B4-mediated cytotoxicity is unaffected by LTk-MICA exposure. Data shown are representative of two repeat experiments with each four NK cell lines. D and E, Data points represent means of triplicates with SD < 6%. F and G, By flow cytometry and immunoblot analysis, CD3ζ protein expression is restored in the NKG2D-licensed T cells (line 1) and TALL-104 cells after 24 h of recovery in fresh media in the absence of MICA. Data shown are representative of three T cell lines tested and of three repeat experiments with TALL-104 cells.

FIGURE 2.

NKG2D-initiated functional impairment of CD3ζ-dependent T cell and NK cell receptors. A, Calcium release (expressed as Fluo-4/Fura red ratio over time in seconds) in response to CD3 cross-linking is attenuated in the NKG2D-licensed T cells (line 1; top left graph) or TALL-104 cells (bottom left graph) after 96 h of exposure to LTk-MICA but not to negative control LTk-neo cells. Calcium release is fully restored after an additional 24 h of culture in the absence of NKG2D ligand (right graphs). Data shown are representative of three T cell lines tested and of duplicate TALL-104 cell experiments. B, Anti–phospho-ZAP70 immunoblot reveals reduced anti-CD3 mAb-induced ZAP70 phosphorylation in the NKG2D-licensed CD8 T cells (line 4) after 96 h of exposure to LTk-MICA, but not to negative control cells. Total ZAP70 confirms equal loading. Data are representative of three T cell lines tested. C, Phospho-flow histograms display differences in CD3-induced ZAP70 phosphorylation between CD3ζlow/− (left histogram) and CD3ζhigh (right histogram) T cells (line 7) after 72 h of exposure to LTk-MICA cells. The top CD3ζ histogram indicates the applied gating. Data shown are representative of experiments with three T cell lines. D, Reduced CD16- and NKp46-mediated redirected lysis of P815 mastocytoma cells by NKG2D-licensed T cells (lines 3 and 6) after exposure for 96 h to LTk-MICA, but not to LTk-neo control cells. Impaired responses are restored after T cell recovery for 24 h in fresh media in the absence of MICA. Data shown are representative of three repeat experiments. E, Reduced CD16-, NKp46-, and NKp30-mediated redirected lysis of P815 mastocytoma cells by NKG2D-licensed NK cells (line 1) after exposure for 96 h to LTk-MIC, but not to LTk-neo control cells. Impaired responses are restored after NK cell recovery for 24 h in fresh media in the absence of MICA. 2B4-mediated cytotoxicity is unaffected by LTk-MICA exposure. Data shown are representative of two repeat experiments with each four NK cell lines. D and E, Data points represent means of triplicates with SD < 6%. F and G, By flow cytometry and immunoblot analysis, CD3ζ protein expression is restored in the NKG2D-licensed T cells (line 1) and TALL-104 cells after 24 h of recovery in fresh media in the absence of MICA. Data shown are representative of three T cell lines tested and of three repeat experiments with TALL-104 cells.

Close modal

In addition to the TCR, CD3ζ controls signaling of human CD16 and the NKp30 and NKp46 NCRs (22, 23). Although mainly NK cell associated, CD16 and the NCRs are also expressed by some T cells (43, 44). Whereas CD16, NKp30, and NKp46 are either absent or poorly expressed in TALL-104 cells (45), three of the eight NKG2D-licensed T cell lines were strongly positive for NKp46 (lines 1, 3, and 6), and one of those (line 3) also expressed CD16. We thus tested these T cell lines in redirected cytotoxicity assays for CD16 and NKp46 functions after NKG2D-initiated CD3ζ downmodulation. As compared with control T cells, those exposed to LTk-MICA transfectants mounted much lower CD16- and NKp46-dependent cytolytic responses that were reduced by ∼60% (Fig. 2D). These effects were due to CD3ζ because expression of the alternative FcεRγ signaling adaptor was not affected (data not shown).

Qualitatively similar results were obtained with NK cells, which typically express CD16, NKp46, and NKp30 (22, 46). As with the CD8 T cells, IL-2–activated but not resting NK cells were susceptible to CD3ζ downmodulation induced by coculture with LTk-MICA cells (data not shown). Loss of CD3ζ corresponded with reduced CD16-, NKp46-, and NKp30-dependent redirected cytolytic responses. In contrast, the functionality of the 2B4 receptor, which is not CD3ζ associated, was unaffected (Fig. 2E) (47).

The observed CD3ζ impairments in T cells and NK cells were fully reversible as determined after termination of the routine coculture with LTk-MICA cells and additional 24 h of culture in the absence of NKG2D ligand. CD3ζ protein expression returned to normal levels, and the functions of TCR, CD16, NKp46, and NKp30 were restored (Fig. 2A, right panels, 2D–G and data not shown). Altogether, our results demonstrate that persistent NKG2D ligand exposure of T or NK cells expressing signaling-competent NKG2D causes reversible CD3ζ downmodulation and functional impairment of immunoreceptors that are associated with this signaling adaptor.

Transient short-term CD3ζ downmodulation after TCR engagement is crucial for the termination of signal transduction and prevention of T cell hyperactivation (48). Long-term diminished CD3ζ, however, which is common in T and NK cells from patients with tumors and autoimmune diseases, contributes to T and NK cell functional abnormalities associated with these disease conditions (29). Several mechanisms have been implicated in contributing to persistent CD3ζ downregulation, including transcriptional defects and increased protein degradation (4958). Moreover, loss of normally detergent-soluble CD3ζ may reflect increased redistribution to the cytoskeleton (59, 60). We found no evidence for decreases of CD3ζ mRNA in the NKG2D-licensed CD8 T cell lines and TALL-104 cells after 96 h of exposure to LTk-MICA cells (data not shown). However, the fact that the CD3ζ downmodulation was revealed using mAbs specific for its carboxyl terminus was suggestive of an involvement of caspase proteases because the recognition sequences of the mAbs include a known caspase cleavage motif (61, 62).

The CD3ζ cytoplasmic domain has several consensus target sequences for caspases, among which caspase-3 and -7 have been shown to cleave in vitro translated CD3ζ (52, 6264). Circumstantial evidence for a physiological involvement of caspase-3 in generating a CD3ζ-deficient T cell phenotype has been described in patients with gastric and liver cancers (51, 65). Similarly, caspase-3 activity is high in CD3ζlow/− SLE patient T cells, and treatment of these T cells with caspase inhibitors restores CD3ζ expression (55). Persistent NKG2D ligand presence may add an additional component to these associations, as cancers frequently express and shed MICA, and soluble MICA is abundant in juvenile-onset SLE patient plasma (10, 30).

To probe for a causal relation between NKG2D engagement, caspase cleavage, and CD3ζ loss, we simultaneously determined caspase activity and CD3ζ expression in the NKG2D-licensed T cell lines. FLICA reagents covalently bind to the reactive cysteine residue in the large subunit of active caspase heterodimers, thereby allowing detection of caspase activity in combination with other cellular proteins by polychromatic surface and intracellular flow cytometry. Although several caspases may cleave CD3ζ, we focused on caspase-3 and -7 because of their previously suggested role in cancer- and SLE-associated CD3ζ degradation (51, 55, 62, 65). Using the T cell lines and the routine time-course exposure to LTk-MICA cells or NKG2D cross-linking Abs, we noted a gradual appearance of caspase activity, as reflected by caspase-3/-7–specific FLICA reagent FAM-DEVD-FMK binding, in parallel with decreases in CD3ζ (Fig. 3A, 3B). Caspase activity was not associated with induction of cell death or apoptosis, as all caspase-active T cells were viable based on LIVE/DEAD fixable violet and Annexin V/7-aminoactinomycin D stainings, respectively (Fig. 3A). Consistent with these flow cytometry results, immunoblot analysis showed that caspase-3 processing to a 19-kDa protein species increased over time (Fig. 3C). Because FAM-DEVD-FMK detects active caspase-3 and -7, cleavage of the latter was also explored. Similar to caspase-3, NKG2D cross-linking resulted in caspase-7 processing to its active 20- and 26-kDa subunits (Fig 3C). In further support of an involvement of caspases in CD3ζ cleavage, immunoblotting using an N-terminal–specific anti-CD3ζ mAb revealed gradual appearance of higher mobility 7- and 10-kDa protein fragments parallel to decrease of the native 16-kDa CD3ζ protein (Fig. 3D) (62).

FIGURE 3.

Activation of caspase-3 and/or -7 is associated with CD3ζ cleavage. A, Flow cytometry dot plots show gradual appearance of caspase-3/-7–specific FLICA (FAM-DEVD-FMK)-binding CD3ζlow/− cells among NKG2D-licensed CD8 T cells (line 2) exposed to LTk-MICA or solid-phase anti-NKG2D (mAb 1D11) during the time period indicated. Data shown are representative of separate experiments with five T cell lines. Dot plots were derived from a gating tree based on exclusion of LIVE/DEAD fixable violet staining (top left plots) and inclusion of surface CD3. Top right dot plot shows absence of Annexin V/7-aminoactinomycin D (7-AAD) reactivity of gated FAM-DEVD-FMK–binding CD3ζlow/− cells. Numbers in graphs specify proportions of FAM-DEVD-FMK–binding CD3ζlow/− cells among total live CD8 T cells (%). B, Mean frequencies (%) of live FAM-DEVD-FMK–binding CD3ζlow/− CD8 T cells recorded by flow cytometry at the indicated time points in separate experiments with five NKG2D-licensed T cells lines cocultured with LTk-MICA or negative control cells. Error bars indicate SD. Asterisks above bars refer to p value indicated in bar graph. C, Immunoblots with anti–caspase-3 or -7 Abs reveal gradual appearance of the active subunits in anti-NKG2D (mAb 1D11) Ab–cross-linked NKG2D-licensed T cells (line 4). β-actin is shown as loading control. Data are representative of four T cell lines tested. D, Immunoblots with N-terminal–specific anti-CD3ζ mAb reveal gradually decreasing amounts of the native 16-kDa CD3ζ protein and the appearance of higher mobility protein bands in the NKG2D-licensed T cell line 1 exposed to LTk-MICA cells, and in anti-NKG2D (mAb 1D11) Ab–cross-linked TALL-14 cells. β-actin is shown as loading control. Data are representative of four T cell lines tested and of three repeat experiments with TALL-104 cells.

FIGURE 3.

Activation of caspase-3 and/or -7 is associated with CD3ζ cleavage. A, Flow cytometry dot plots show gradual appearance of caspase-3/-7–specific FLICA (FAM-DEVD-FMK)-binding CD3ζlow/− cells among NKG2D-licensed CD8 T cells (line 2) exposed to LTk-MICA or solid-phase anti-NKG2D (mAb 1D11) during the time period indicated. Data shown are representative of separate experiments with five T cell lines. Dot plots were derived from a gating tree based on exclusion of LIVE/DEAD fixable violet staining (top left plots) and inclusion of surface CD3. Top right dot plot shows absence of Annexin V/7-aminoactinomycin D (7-AAD) reactivity of gated FAM-DEVD-FMK–binding CD3ζlow/− cells. Numbers in graphs specify proportions of FAM-DEVD-FMK–binding CD3ζlow/− cells among total live CD8 T cells (%). B, Mean frequencies (%) of live FAM-DEVD-FMK–binding CD3ζlow/− CD8 T cells recorded by flow cytometry at the indicated time points in separate experiments with five NKG2D-licensed T cells lines cocultured with LTk-MICA or negative control cells. Error bars indicate SD. Asterisks above bars refer to p value indicated in bar graph. C, Immunoblots with anti–caspase-3 or -7 Abs reveal gradual appearance of the active subunits in anti-NKG2D (mAb 1D11) Ab–cross-linked NKG2D-licensed T cells (line 4). β-actin is shown as loading control. Data are representative of four T cell lines tested. D, Immunoblots with N-terminal–specific anti-CD3ζ mAb reveal gradually decreasing amounts of the native 16-kDa CD3ζ protein and the appearance of higher mobility protein bands in the NKG2D-licensed T cell line 1 exposed to LTk-MICA cells, and in anti-NKG2D (mAb 1D11) Ab–cross-linked TALL-14 cells. β-actin is shown as loading control. Data are representative of four T cell lines tested and of three repeat experiments with TALL-104 cells.

Close modal

To gain direct evidence for an involvement of caspase-3, -7, or both in CD3ζ downmodulation, T cell cultures including LTk-MICA cells or NKG2D cross-linking Abs were supplemented with caspase-3 and -7 selective peptide inhibitor zDEVD-FMK, or the broad-spectrum caspase inhibitors zVAD-FMK or Q-VD-OPh. We confirmed that none of these reagents affected cell viability, T cell NKG2D, or MICA expression on the LTk cells. By flow cytometry and immunoblot analysis, the NKG2D-initiated induction of the CD3ζlow/− phenotype was completely prevented in the continuous presence of any one of the three inhibitory peptide substrates (Fig. 4A–C) and attenuated on addition of the inhibitors after initial 48-h T cell exposure to cross-linked anti-NKG2D mAb or LTk-MICA cells (Supplemental Fig. 2 and data not shown). Because the caspase-3 and -7 inhibitor was no less effective than the pan-reactive reagents, our results implicate caspase-3 or -7, or both, as the primary proteases responsible for the observed CD3ζ loss. In the absence of monospecific inhibitors, their relative contributions cannot be assessed.

FIGURE 4.

Caspase inhibition prevents NKG2D-initiated CD3ζ degradation. A, By flow cytometry, proportions of caspase-3/-7–specific FLICA (FAM-DEVD-FMK)-binding CD3ζlow/− cells among NKG2D-licensed T cells (line 5) exposed for 96 h to solid-phase anti-NKG2D (mAb 1D11) and polycaspase (Q-VD-OPh, zVAD-FMK) or caspase-3/-7 (zDEVD-FMK) inhibitors are minute compared with those among control T cell populations. Dot plots were derived from a gating tree based on exclusion of LIVE/DEAD fixable violet staining and inclusion of surface CD3. Numbers in graphs specify proportions of FAM-DEVD-FKM–binding CD3ζlow/− cells among total live CD8 T cells (%). Data shown are representative of four T cell lines tested. B, Averaged percentages of FAM-DEVD-FKM–binding CD3ζlow/− cells among total live CD8 T cells recorded by flow cytometry in NKG2D-licensed T cell lines 5, 6, 7, and 8 exposed for 96 h to solid-phase anti-NKG2D (mAb 1D11) with or without the caspase inhibitors or DMSO solvent control. Error bars indicate SD. Asterisks above bars refer to p value indicated in graph. C, Immunoblots with anti-CD3ζ C- or N-terminal–specific Abs show reversal of CD3ζ cleavage in NKG2D-licensed T cells (line 7) exposed for 96 h to solid-phase anti-NKG2D (mAb 1D11) and caspase inhibitors. β-actin is shown as loading control. Data are representative of four T cell lines tested.

FIGURE 4.

Caspase inhibition prevents NKG2D-initiated CD3ζ degradation. A, By flow cytometry, proportions of caspase-3/-7–specific FLICA (FAM-DEVD-FMK)-binding CD3ζlow/− cells among NKG2D-licensed T cells (line 5) exposed for 96 h to solid-phase anti-NKG2D (mAb 1D11) and polycaspase (Q-VD-OPh, zVAD-FMK) or caspase-3/-7 (zDEVD-FMK) inhibitors are minute compared with those among control T cell populations. Dot plots were derived from a gating tree based on exclusion of LIVE/DEAD fixable violet staining and inclusion of surface CD3. Numbers in graphs specify proportions of FAM-DEVD-FKM–binding CD3ζlow/− cells among total live CD8 T cells (%). Data shown are representative of four T cell lines tested. B, Averaged percentages of FAM-DEVD-FKM–binding CD3ζlow/− cells among total live CD8 T cells recorded by flow cytometry in NKG2D-licensed T cell lines 5, 6, 7, and 8 exposed for 96 h to solid-phase anti-NKG2D (mAb 1D11) with or without the caspase inhibitors or DMSO solvent control. Error bars indicate SD. Asterisks above bars refer to p value indicated in graph. C, Immunoblots with anti-CD3ζ C- or N-terminal–specific Abs show reversal of CD3ζ cleavage in NKG2D-licensed T cells (line 7) exposed for 96 h to solid-phase anti-NKG2D (mAb 1D11) and caspase inhibitors. β-actin is shown as loading control. Data are representative of four T cell lines tested.

Close modal

Activation of effector caspase-3 and -7 occurs in the extrinsic or intrinsic apoptotic pathway, the ER stress response-mediated caspase cascade, or through granzyme B (66). Although active caspase-3 is crucial for the execution of apoptosis, its cleavage can also occur in the absence of cell death, after TCR-mediated upregulation of the Fas death receptor ligand (FasL) and engagement of surface Fas (6769). T cell costimulation by NKG2D is associated with protection from FasL/Fas-induced apoptosis and growth arrest possibly by induction of the antiapoptotic regulators Bcl-xL and c-FLIP, in conjunction with NKG2D-mediated downregulation of Fas expression (36, 70; our unpublished data). Similarly, NKG2D ligation on human NK cells induces the anti-apoptotic Bcl-2 homolog Bcl2A1 and the inhibitor of apoptosis protein cIAP2 (71). However, NKG2D engagement also promotes expression and shedding of FasL (36). Activated lymphocyte populations heterogeneous for NKG2D, such as those that arise in microenvironments with persistent NKG2D ligand expression and various degrees of resultant NKG2D downmodulation, may thus be differentially affected by these opposing NKG2D functions (36). With no or minimal NKG2D downmodulation, NKG2D signals may induce FasL production, yet simultaneously confer survival and protect from FasL/Fas-initiated caspase activation. NKG2Dlow/− T or NK cells, however, may be susceptible to FasL/Fas activity because of absence of protective signals. If caspase-3/-7 processing and resultant CD3ζ cleavage were downstream of NKG2D-induced FasL activity, differential impairment of NKG2D on our CD8 T cell lines in the LTk-MICA cocultures should be associated with corresponding differences in caspase-3/-7 activation and CD3ζ cleavage. Consistent with this model, we registered loss of CD3ζ staining and associated caspase-3/-7 activity mainly among T cells with little or no NKG2D (Figs. 1A, 5A).

FIGURE 5.

Inhibition of FasL/Fas signaling prevents NKG2D-initiated caspase activation and CD3ζ degradation. A, By flow cytometry, FAM-DEVD-FMK–binding CD3ζlow/− cells among LTK-MICA–exposed NKG2D-licensed T cells (line 4) are NKG2Dlow/−, whereas CD3ζhigh cells are NKG2Dhigh. Left dot plot indicates gating. Dashed profiles in histograms represent isotype Ig control staining. Data are representative of four T cell lines tested. B, Daily media replacement or treatment with neutralizing anti-FasL or antagonist anti-Fas mAb prevent the appearance of caspase-3/-7–active (FAM-DEVD-FMK–binding) CD3ζlow/− cells in NKG2D-licensed T cells (line 4) exposed for 72 h to LTk-MICA transfectants (top panel). Bottom panel shows caspase-8–specific FAM-IETD-FMK binding to T cells. Dot plots were derived from a gating tree based on exclusion of LIVE/DEAD fixable violet staining and inclusion of surface CD3. Numbers in graphs specify proportions of CD3ζlow/− cells with evidence for caspase activity among total live CD8 T cells (%). Data shown are representative of three T cell lines tested. C, Averaged percent of FAM-DEVD-FKM–binding CD3ζlow/− cells among total live CD8 T cells recorded by flow cytometry in the NKG2D-licensed T cell lines 3–6 exposed for 72 (left bar graph) or 96 h (right bar graph) to LTk-MICA cells with or without daily media replacement or addition of neutralizing anti-FasL or antagonist anti-Fas mAb. Error bars indicate SD. Asterisks above bars refer to p value indicated in graph. D, By flow cytometry, NKG2Dhigh FAM-DEVD-FMK T cells (line 3) display less surface Fas than NKG2Dlow/− FAM-DEVD-FMK–binding T cells. Left dot plot indicates gating. Dashed profiles in histograms represent isotype Ig control staining. Data are representative of three T cell lines tested.

FIGURE 5.

Inhibition of FasL/Fas signaling prevents NKG2D-initiated caspase activation and CD3ζ degradation. A, By flow cytometry, FAM-DEVD-FMK–binding CD3ζlow/− cells among LTK-MICA–exposed NKG2D-licensed T cells (line 4) are NKG2Dlow/−, whereas CD3ζhigh cells are NKG2Dhigh. Left dot plot indicates gating. Dashed profiles in histograms represent isotype Ig control staining. Data are representative of four T cell lines tested. B, Daily media replacement or treatment with neutralizing anti-FasL or antagonist anti-Fas mAb prevent the appearance of caspase-3/-7–active (FAM-DEVD-FMK–binding) CD3ζlow/− cells in NKG2D-licensed T cells (line 4) exposed for 72 h to LTk-MICA transfectants (top panel). Bottom panel shows caspase-8–specific FAM-IETD-FMK binding to T cells. Dot plots were derived from a gating tree based on exclusion of LIVE/DEAD fixable violet staining and inclusion of surface CD3. Numbers in graphs specify proportions of CD3ζlow/− cells with evidence for caspase activity among total live CD8 T cells (%). Data shown are representative of three T cell lines tested. C, Averaged percent of FAM-DEVD-FKM–binding CD3ζlow/− cells among total live CD8 T cells recorded by flow cytometry in the NKG2D-licensed T cell lines 3–6 exposed for 72 (left bar graph) or 96 h (right bar graph) to LTk-MICA cells with or without daily media replacement or addition of neutralizing anti-FasL or antagonist anti-Fas mAb. Error bars indicate SD. Asterisks above bars refer to p value indicated in graph. D, By flow cytometry, NKG2Dhigh FAM-DEVD-FMK T cells (line 3) display less surface Fas than NKG2Dlow/− FAM-DEVD-FMK–binding T cells. Left dot plot indicates gating. Dashed profiles in histograms represent isotype Ig control staining. Data are representative of three T cell lines tested.

Close modal

In support of a role of FasL/Fas in caspase-3/-7 activation, culture supernatants from the CD8 T cell cocultures with LTk-MICA cells contained abundant FasL (data not shown). Moreover, the appearance of caspase-3/-7–active and CD3ζlow/− populations was reduced by daily replacement of culture supernatant with fresh media (Fig. 5B, 5C). Direct evidence for FasL/Fas involvement was obtained by treatment of 72 or 96-h T cell cultures with neutralizing anti-FasL or antagonist anti-Fas Abs. In a pattern near identical to that of direct caspase inhibition (Fig. 4A), caspase-3/-7 processing and CD3ζ downmodulation were almost completely inhibited in the presence of either Ab, but not by control Ig (Fig. 5B, 5C). Consistent with a FasL/Fas-initiated caspase cascade, caspase-8 was equally affected as shown by the FLICA reagent binding approach (Fig. 5B, bottom panels) (66). In support of NKG2D-mediated protection from FasL effects via Fas downregulation (36, 70), the T cells with high levels and thus presumably fully functional NKG2D (and no signs of associated caspase-3/-7 activity) expressed less surface Fas than the NKG2Dlow/− and caspase-3/-7–active T cells (Fig. 5D)

NKG2D effector functions also include release of the cytotoxic serine protease granzyme B (72). However, the near-complete inhibition of caspase activity in the presence of neutralizing anti-FasL or antagonist anti-Fas argued against a prominent role of granzyme B in caspase cleavage downstream of NKG2D (66, 73). Moreover, absence of caspase-2–specific FAM- VDVAD-FMK binding indicated that caspase-2, a granzyme B substrate, remained as inactive proenzyme in the NKG2D-stimulated T cells (data not shown) (74). We thus conclude that activation of caspase-3, -7, and -8 and the associated CD3ζ cleavage is a consequence of NKG2D signaling-induced FasL expression.

So far, our studies used cytokine-primed NKG2D-licensed T cell lines because these enabled us to examine NKG2D functions independent of other signals that may impact CD3ζ. Despite this logistic rationale, the results obtained are likely of physiological relevance because similarly NKG2D-licensed CD8 T cells occur in inflammatory celiac disease mucosa and rheumatoid arthritis synovia with aberrant MIC expression (5, 12, 37; our unpublished results). In normal human CD8 T cells, however, NKG2D functions as a coreceptor (4, 6). In comparison with published data on TCR-dependent CD3ζ degradation (38), the timing of NKG2D-initiated effects on CD3ζ expression was delayed in our IL-2 primed T cell lines, suggesting that the effects of the two receptors can be discriminated in a costimulatory setting. Ex vivo purified CD8 T cells were exposed for 4 d to solid-phase anti-CD3 and anti-NKG2D, individually or together, and CD3ζ expression was monitored by flow cytometry in daily intervals. Cross-linking of CD3 alone induced rapid and transient CD3ζ downmodulation (Fig. 6A) (38, 48). With the NKG2D-costimulated T cell lines, however, the short-lived CD3-induced effect was followed by gradually increasing proportions of T cells with little or no CD3ζ (Fig. 6A). As with the NKG2D-licensed T cells, caspase inhibitors, as well as neutralizing anti-FasL or antagonist anti-Fas Abs, prevented the NKG2D but not the TCR-dependent disappearance of CD3ζ (Fig. 6A and data not shown). Solid-phase anti-NKG2D mAb alone had no effect.

FIGURE 6.

CD3ζ impairment subsequent to NKG2D-mediated T cell costimulation. A, Flow cytometry analysis of CD3ζ expression in freshly isolated peripheral blood NKG2D+CD8 T cells after exposure to solid-phase anti-CD3 alone or together with anti-NKG2D for the indicated time periods. The short-term (at 24 h), transient CD3 signal-dependent CD3ζ downmodulation is followed by the gradually increasing NKG2D costimulation-dependent effect. The bottom four histograms assigned to the 24- and 96-h time periods show that treatment with the caspase-3/-7–specific peptide inhibitor zDEVD-FMK or with neutralizing anti-FasL mAb prevents the NKG2D costimulation-induced, but not the CD3-induced CD3ζ downmodulation. Data shown are representative of duplicate experiments with each three healthy donor samples. B, Cytotoxic responses of MART-1–specific NKG2D+CD8 T cells (clones 3 and 39) against M27 peptide-pulsed C1R–HLA-A2 or melanoma A375 cell targets are attenuated by prior 96-h coculture with M27 peptide-pulsed C1R–HLA-A2–MICA (filled squares), but not with C1R–HLA-A2 cells (open squares). Data points represent means of triplicates with SD < 6%. Data shown are representative of duplicate experiments with four T cell clones.

FIGURE 6.

CD3ζ impairment subsequent to NKG2D-mediated T cell costimulation. A, Flow cytometry analysis of CD3ζ expression in freshly isolated peripheral blood NKG2D+CD8 T cells after exposure to solid-phase anti-CD3 alone or together with anti-NKG2D for the indicated time periods. The short-term (at 24 h), transient CD3 signal-dependent CD3ζ downmodulation is followed by the gradually increasing NKG2D costimulation-dependent effect. The bottom four histograms assigned to the 24- and 96-h time periods show that treatment with the caspase-3/-7–specific peptide inhibitor zDEVD-FMK or with neutralizing anti-FasL mAb prevents the NKG2D costimulation-induced, but not the CD3-induced CD3ζ downmodulation. Data shown are representative of duplicate experiments with each three healthy donor samples. B, Cytotoxic responses of MART-1–specific NKG2D+CD8 T cells (clones 3 and 39) against M27 peptide-pulsed C1R–HLA-A2 or melanoma A375 cell targets are attenuated by prior 96-h coculture with M27 peptide-pulsed C1R–HLA-A2–MICA (filled squares), but not with C1R–HLA-A2 cells (open squares). Data points represent means of triplicates with SD < 6%. Data shown are representative of duplicate experiments with four T cell clones.

Close modal

To more accurately model natural NKG2D costimulatory conditions, we cultured HLA-A2–restricted and MART-1 Ag-specific NKG2D+CD8 T cell clones in the presence of MART-1 M27 peptide-pulsed C1R–HLA-A2–MICA or control C1R–HLA-A2 transfectants, and monitored their CD3ζ expression and function over time. As with the T cells costimulated by anti-CD3 and anti-NKG2D Ab cross-linking, NKG2D-dependent effects on CD3ζ expression became increasingly apparent after an initial transient wave of TCR-induced CD3ζ downmodulation (data not shown). Consistent with impaired TCR signaling as a result of the NKG2D-initiated CD3ζ loss, the ability of the 96-h MICA-exposed T cell clones to mount cytolytic responses against the peptide-pulsed C1R–HLA-A2 or A375 melanoma (HLA-A2+) targets was markedly diminished as compared with controls (Fig. 6B). This functional impairment was reversed by caspase inhibitors and neutralizing anti-FasL or antagonist anti-Fas (data not shown). Thus, under conditions of persistent ligand expression, both direct and costimulatory NKG2D signals lead to CD3ζ impairment via induction of FasL and ensuing activation of the caspase cascade.

We sought ex vivo evidence in support of our in vitro observations using freshly isolated TILs from patients with MIC-positive cancers and PBLs from juvenile-onset SLE patients with serum-soluble MICA (14, 30). Unlike the in vitro model that used CD8 T cells only, these bulk lymphocyte populations presumably consisted of T and NK cells heterogeneous for NKG2D expression, downmodulation, and functional engagement. Depending on their activation states, these lymphocytes probably also differed in their susceptibility to FasL effects. However, regardless of these and possibly other factors, at least some T cells and NK cells in these populations should display the CD3ζlow/− phenotype and evidence for caspase activation in conjunction with ligand-induced diminished or constitutively absent NKG2D expression. As with the cultured T cell lines, we measured surface CD3 and NKG2D, and CD3ζ, as well as binding of the caspase-3/-7–specific FAM-DEVD-FMK FLICA reagent by flow cytometry. CD16 was used as marker for NK cells. LIVE/DEAD fixable violet and/or Annexin V-positive cells were excluded. With all of nine TIL samples (extracted from one melanoma, and from one breast, six ovarian, and one colon cancer specimens), diminished CD3ζ was preferentially associated with T and NK cells expressing little or no NKG2D (Fig. 7A, dot plots). Active caspase-3/-7 were prominent only in the NKG2D subset of the CD3ζlow/− populations, and absent in T and NK cells with normal CD3ζ (Fig. 7A, histograms and data not shown). Consistent with its role in CD16 assembly, CD3ζ loss in NK cells was associated with low CD16 expression (data not shown). Overall similar data were obtained with 17 SLE patient-derived lymphocytes but not with age-matched control samples (Fig. 7A). Notably, the proportions of NKG2Dlow/−CD3ζlow/− T cells and NK cells correlated with plasma concentrations of soluble MICA (Fig. 7B). Thus, overall, our results with ex vivo lymphocytes from patients with cancer and SLE replicated those made in cell culture.

FIGURE 7.

Preferential occurrence of CD3ζ loss and caspase-3/-7 activity in NKG2Dlow/− T cells and NK cells from cancer and SLE patients. A, Flow cytometry dot plots show that most CD3ζlow/− T cells or NK cells among TIL and SLE PBMCs express little or no NKG2D. Dot plots were derived from a gating tree based on exclusion of LIVE/DEAD fixable violet/Annexin V staining and inclusion of surface CD3 (T cells) or CD16 (NK cells). Numbers in plots specify proportions of CD3ζlow/− cells among total live/nonapoptotic T cells and NK cells (%). Stainings of normal donor PBMCs are shown for comparison. Histograms display binding of caspase-3/-7–specific FAM-DEVD-FMK to NKG2DCD3ζlow/− (shaded profiles) versus NKG2D+CD3ζlow/− (open profiles) T cells or NK cells. Data shown are representative of 9 TIL, 17 juvenile-onset SLE, and 15 normal donor samples. B, Plotting of CD3ζlow/−NKG2D T cell proportions (%) against soluble MICA plasma concentrations in 17 SLE patient peripheral blood samples (R2 = 0.8238).

FIGURE 7.

Preferential occurrence of CD3ζ loss and caspase-3/-7 activity in NKG2Dlow/− T cells and NK cells from cancer and SLE patients. A, Flow cytometry dot plots show that most CD3ζlow/− T cells or NK cells among TIL and SLE PBMCs express little or no NKG2D. Dot plots were derived from a gating tree based on exclusion of LIVE/DEAD fixable violet/Annexin V staining and inclusion of surface CD3 (T cells) or CD16 (NK cells). Numbers in plots specify proportions of CD3ζlow/− cells among total live/nonapoptotic T cells and NK cells (%). Stainings of normal donor PBMCs are shown for comparison. Histograms display binding of caspase-3/-7–specific FAM-DEVD-FMK to NKG2DCD3ζlow/− (shaded profiles) versus NKG2D+CD3ζlow/− (open profiles) T cells or NK cells. Data shown are representative of 9 TIL, 17 juvenile-onset SLE, and 15 normal donor samples. B, Plotting of CD3ζlow/−NKG2D T cell proportions (%) against soluble MICA plasma concentrations in 17 SLE patient peripheral blood samples (R2 = 0.8238).

Close modal

This study builds on previous fragmentary observations suggesting a link between NKG2D signaling and CD3ζ loss in mouse NK cells, and on the possibly related T and NK cell deficiencies that have been extensively documented in patients with cancer and autoimmune diseases (21, 29). Our in vitro experimental modeling using human CD8 T cells and NK cells, together with evidence from ex vivo analyses of patient lymphocytes, establish a dynamic of events whereby NKG2D stimulation and downmodulation in the context of persistent ligand expression facilitates FasL/Fas-mediated caspase-3/-7 activation and, in turn, CD3ζ protein cleavage. This impairment inflicts NKG2Dlow/− T cells and NK cells, which lack resistance to FasL-induced effects because of absence of NKG2D-dependent survival cues (36). At least in our in vitro experiments, granzyme B, which is also released on NKG2D triggering (72) and may target CD3ζ either directly or indirectly through initiation of caspase cascades, did not appear to play a significant role (52). Consistent with growing evidence for nonapoptotic caspase functions, the chronically MIC ligand-exposed T and NK cells showed no signs of apoptosis induction, despite activation of caspase-3, -7, and -8 (68, 69, 75). The NKG2D-initiated effect on CD3ζ was independent of whether NKG2D signaling was direct or costimulatory (5). Functionally, CD3ζ loss in T and NK cells was associated with reduced TCR, and CD16, NKp30, and NKp46 NCR signaling capacities, respectively. CD3ζ expression and function were restored on ligand removal.

With the ex vivo cancer and SLE patient-derived lymphocytes, the colocalization of active caspase-3/-7 and CD3ζ loss at the single-cell level establishes a correlative link between caspase-3/-7 activity and CD3ζ cleavage. In SLE patients, expansions of NKG2D-bearing T cell populations directly correlate with plasma FasL concentrations (30). In cancer patients, NKG2D-induced FasL may also be relevant for induction of caspase processing, although other NKG2D-induced mediators such as arachidonic acid and/or PGE might contribute as well (29, 37, 66). However, in either of these disease conditions, persistent CD3ζ impairment is almost certainly multifactorial (29). In fact, in all tumor and SLE samples, T cells with low CD3ζ but no caspase activity and normal NKG2D were also detected. We thus do not claim an exclusive role for NKG2D, especially because no lymphocyte sample tested could be qualified as derived from an environment negative for all ligands of NKG2D. In contrast, the positive correlation of NKG2DCD3ζ and caspase-3/-7–active T cell and NK cell proportions with plasma-soluble MICA concentrations in our SLE samples clearly supports a causal relation between NKG2D functions and persistent CD3ζ downmodulation in vivo. In further support, our preliminary data with PBMCs from three SLE patients show recovery of CD3ζ expression in the T and NK cells after a 24-h culture period in the absence of soluble MICA (Supplemental Fig. 3).

Despite the CD3ζ downmodulation and associated functional impairment, CD3 surface expression was largely unaffected. This is reminiscent of similar discrepancies that have been noted with T cells in tumor and autoimmune disease settings (29, 57). With the exception of FcεRγ-facilitated TCR rewiring in SLE T cells, and the possibility of activation-induced recruitment of CD3ζ to the cytoskeleton, these observations remain unexplained (49, 56, 59).

Persistent NKG2D ligand exposure of T and NK cells has been associated with phenotypic and functional alterations beyond NKG2D impairment alone. Peripheral blood NK cells exposed to soluble MICA in vivo are deficient for NKp46 and CXCR1, and exposure of NK cells to soluble MICA in vitro results in CXCR1 and CCR7 downmodulation (15). Moreover, in vitro NKG2D costimulation of peripheral blood CD8 T cells recapitulates the CCR9low phenotype of intestinal epithelial lymphocytes from celiac disease mucosa (76). Sustained NKG2D ligand exposure of murine NK cells negatively affects their CD16-mediated Ab-dependent cell-mediated cytotoxicity, and impairs the activating NK1.1 and NKp46 receptors (20). NKG2D-initiated and caspase-mediated CD3ζ degradation may explain the human and mouse NKp46 impairment but is not applicable to mouse CD16, which unlike human CD16, signals effectively only through FcεRγ (23, 77). However, caspase effects may nevertheless represent a common denominator for some of these phenotypic and functional changes, as at least CXCR1 contains putative caspase-3 cleavage target motifs (64, 78). Indeed, our preliminary data suggest that CXCR1 expression in T cells is regulated by NKG2D and restored after treatment with the caspase-3/-7 inhibitor.

Negative immune regulation by NKG2D facilitates tumor immune evasion and affects autoimmune and inflammatory responses. In advanced tumors, persistent NKG2D ligand expression negatively imprints on local and systemic T and NK cell responses through ligand-induced NKG2D downmodulation and population expansions of NKG2D+CD4 T cells with negative regulatory functions (1, 14, 19, 36). In juvenile-onset SLE, those NKG2D+CD4 T cells are associated with disease remission, thus suggesting that they participate in regulation of effector responses that provoke autoimmunity (30).

Our study identifies an additional strategy whereby NKG2D broadly affects T and NK cell responses in tissue environments with pathological NKG2D ligand expression by simultaneous impairment of multiple CD3ζ-dependent receptor functions. Collectively, our results offer a mechanistic explanation for CD3ζ deficiencies associated with cancers and autoimmune diseases, and may also apply to certain NKG2D ligand-positive infectious conditions and to the fetal–maternal interface.

We thank the Cooperative Human Tissue Network, Western Division (Nashville, TN) for tissue specimens.

Disclosures The authors have no financial conflicts of interest.

This work was supported by the Lupus Research Institute (to V.G.) and National Institutes of Health Grants DK58727 (to B.J.), AI30581 (to T.S.), and AI52319 (to T.S.).

The online version of this article contains supplemental material.

Abbreviations used in this paper:

7-AAD

7-aminoactinomycin D

FasL

Fas ligand

FLICA

fluorochrome-conjugated cell-permeable inhibitor of caspase

FMK

fluoromethyl ketone

MICA

MHC class I-related chain A

MICB

MHC class I-related chain B

NCR

natural cytotoxicity receptor

Q-VD-OPh

quinolyl-valyl-O-methylaspartyl-[-2,6-difluorophenoxy]-methyl ketone

SLE

systemic lupus erythematosus

TIL

tumor-infiltrating lymphocyte

ULBP

UL16-binding protein

zDEVD

benzyloxycarbonyl Asp-Glu-Val-Asp

zVAD

benzyloxycarbonyl Val-Ala-Asp.

1
Burgess
S. J.
,
Maasho
K.
,
Masilamani
M.
,
Narayanan
S.
,
Borrego
F.
,
Coligan
J. E.
.
2008
.
The NKG2D receptor: immunobiology and clinical implications.
Immunol. Res.
40
:
18
34
.
2
Bauer
S.
,
Groh
V.
,
Wu
J.
,
Steinle
A.
,
Phillips
J. H.
,
Lanier
L. L.
,
Spies
T.
.
1999
.
Activation of NK cells and T cells by NKG2D, a receptor for stress-inducible MICA.
Science
285
:
727
729
.
3
Wu
J.
,
Song
Y.
,
Bakker
A. B.
,
Bauer
S.
,
Spies
T.
,
Lanier
L. L.
,
Phillips
J. H.
.
1999
.
An activating immunoreceptor complex formed by NKG2D and DAP10.
Science
285
:
730
732
.
4
Groh
V.
,
Rhinehart
R.
,
Randolph-Habecker
J.
,
Topp
M. S.
,
Riddell
S. R.
,
Spies
T.
.
2001
.
Costimulation of CD8alphabeta T cells by NKG2D via engagement by MIC induced on virus-infected cells.
Nat. Immunol.
2
:
255
260
.
5
Meresse
B.
,
Chen
Z.
,
Ciszewski
C.
,
Tretiakova
M.
,
Bhagat
G.
,
Krausz
T. N.
,
Raulet
D. H.
,
Lanier
L. L.
,
Groh
V.
,
Spies
T.
, et al
.
2004
.
Coordinated induction by IL15 of a TCR-independent NKG2D signaling pathway converts CTL into lymphokine-activated killer cells in celiac disease.
Immunity
21
:
357
366
.
6
Ehrlich
L. I.
,
Ogasawara
K.
,
Hamerman
J. A.
,
Takaki
R.
,
Zingoni
A.
,
Allison
J. P.
,
Lanier
L. L.
.
2005
.
Engagement of NKG2D by cognate ligand or antibody alone is insufficient to mediate costimulation of human and mouse CD8+ T cells.
J. Immunol.
174
:
1922
1931
.
7
Bryceson
Y. T.
,
March
M. E.
,
Ljunggren
H. G.
,
Long
E. O.
.
2006
.
Synergy among receptors on resting NK cells for the activation of natural cytotoxicity and cytokine secretion.
Blood
107
:
159
166
.
8
Venkataraman
G. M.
,
Suciu
D.
,
Groh
V.
,
Boss
J. M.
,
Spies
T.
.
2007
.
Promoter region architecture and transcriptional regulation of the genes for the MHC class I-related chain A and B ligands of NKG2D.
J. Immunol.
178
:
961
969
.
9
Groh
V.
,
Bahram
S.
,
Bauer
S.
,
Herman
A.
,
Beauchamp
M.
,
Spies
T.
.
1996
.
Cell stress-regulated human major histocompatibility complex class I gene expressed in gastrointestinal epithelium.
Proc. Natl. Acad. Sci. USA
93
:
12445
12450
.
10
Groh
V.
,
Rhinehart
R.
,
Secrist
H.
,
Bauer
S.
,
Grabstein
K. H.
,
Spies
T.
.
1999
.
Broad tumor-associated expression and recognition by tumor-derived gamma delta T cells of MICA and MICB.
Proc. Natl. Acad. Sci. USA
96
:
6879
6884
.
11
Groh
V.
,
Bruhl
A.
,
El-Gabalawy
H.
,
Nelson
J. L.
,
Spies
T.
.
2003
.
Stimulation of T cell autoreactivity by anomalous expression of NKG2D and its MIC ligands in rheumatoid arthritis.
Proc. Natl. Acad. Sci. USA
100
:
9452
9457
.
12
Hüe
S.
,
Mention
J. J.
,
Monteiro
R. C.
,
Zhang
S.
,
Cellier
C.
,
Schmitz
J.
,
Verkarre
V.
,
Fodil
N.
,
Bahram
S.
,
Cerf-Bensussan
N.
,
Caillat-Zucman
S.
.
2004
.
A direct role for NKG2D/MICA interaction in villous atrophy during celiac disease.
Immunity
21
:
367
377
.
13
Stern-Ginossar
N.
,
Mandelboim
O.
.
2009
.
An integrated view of the regulation of NKG2D ligands.
Immunology
128
:
1
6
.
14
Groh
V.
,
Wu
J.
,
Yee
C.
,
Spies
T.
.
2002
.
Tumour-derived soluble MIC ligands impair expression of NKG2D and T-cell activation.
Nature
419
:
734
738
.
15
Doubrovina
E. S.
,
Doubrovin
M. M.
,
Vider
E.
,
Sisson
R. B.
,
O’Reilly
R. J.
,
Dupont
B.
,
Vyas
Y. M.
.
2003
.
Evasion from NK cell immunity by MHC class I chain-related molecules expressing colon adenocarcinoma.
J. Immunol.
171
:
6891
6899
.
16
Mincheva-Nilsson
L.
,
Nagaeva
O.
,
Chen
T.
,
Stendahl
U.
,
Antsiferova
J.
,
Mogren
I.
,
Hernestål
J.
,
Baranov
V.
.
2006
.
Placenta-derived soluble MHC class I chain-related molecules down-regulate NKG2D receptor on peripheral blood mononuclear cells during human pregnancy: a possible novel immune escape mechanism for fetal survival.
J. Immunol.
176
:
3585
3592
.
17
Roda-Navarro
P.
,
Reyburn
H. T.
.
2009
.
The traffic of the NKG2D/Dap10 receptor complex during natural killer (NK) cell activation.
J. Biol. Chem.
284
:
16463
16472
.
18
Salih
H. R.
,
Rammensee
H. G.
,
Steinle
A.
.
2002
.
Cutting edge: down-regulation of MICA on human tumors by proteolytic shedding.
J. Immunol.
169
:
4098
4102
.
19
Salih
H. R.
,
Holdenrieder
S.
,
Steinle
A.
.
2008
.
Soluble NKG2D ligands: prevalence, release, and functional impact.
Front. Biosci.
13
:
3448
3456
.
20
Coudert
J. D.
,
Scarpellino
L.
,
Gros
F.
,
Vivier
E.
,
Held
W.
.
2008
.
Sustained NKG2D engagement induces cross-tolerance of multiple distinct NK cell activation pathways.
Blood
111
:
3571
3578
.
21
Coudert
J. D.
,
Zimmer
J.
,
Tomasello
E.
,
Cebecauer
M.
,
Colonna
M.
,
Vivier
E.
,
Held
W.
.
2005
.
Altered NKG2D function in NK cells induced by chronic exposure to NKG2D ligand-expressing tumor cells.
Blood
106
:
1711
1717
.
22
Moretta
A.
,
Bottino
C.
,
Vitale
M.
,
Pende
D.
,
Cantoni
C.
,
Mingari
M. C.
,
Biassoni
R.
,
Moretta
L.
.
2001
.
Activating receptors and coreceptors involved in human natural killer cell-mediated cytolysis.
Annu. Rev. Immunol.
19
:
197
223
.
23
Lanier
L. L.
2008
.
Up on the tightrope: natural killer cell activation and inhibition.
Nat. Immunol.
9
:
495
502
.
24
Weissman
A. M.
,
Baniyash
M.
,
Hou
D.
,
Samelson
L. E.
,
Burgess
W. H.
,
Klausner
R. D.
.
1988
.
Molecular cloning of the zeta chain of the T cell antigen receptor.
Science
239
:
1018
1021
.
25
Weiss
A.
1993
.
T cell antigen receptor signal transduction: a tale of tails and cytoplasmic protein-tyrosine kinases.
Cell
73
:
209
212
.
26
Samelson
L. E.
2002
.
Signal transduction mediated by the T cell antigen receptor: the role of adapter proteins.
Annu. Rev. Immunol.
20
:
371
394
.
27
Myers
M. D.
,
Dragone
L. L.
,
Weiss
A.
.
2005
.
Src-like adaptor protein down-regulates T cell receptor (TCR)-CD3 expression by targeting TCRzeta for degradation.
J. Cell Biol.
170
:
285
294
.
28
Taylor
D. D.
,
Sullivan
S. A.
,
Eblen
A. C.
,
Gercel-Taylor
C.
.
2002
.
Modulation of T-cell CD3-zeta chain expression during normal pregnancy.
J. Reprod. Immunol.
54
:
15
31
.
29
Baniyash
M.
2004
.
TCR zeta-chain downregulation: curtailing an excessive inflammatory immune response.
Nat. Rev. Immunol.
4
:
675
687
.
30
Dai
Z.
,
Turtle
C. J.
,
Booth
G. C.
,
Riddell
S. R.
,
Gooley
T. A.
,
Stevens
A. M.
,
Spies
T.
,
Groh
V.
.
2009
.
Normally occurring NKG2D+CD4+ T cells are immunosuppressive and inversely correlated with disease activity in juvenile-onset lupus.
J. Exp. Med.
206
:
793
805
.
31
Jabri
B.
,
Selby
J. M.
,
Negulescu
H.
,
Lee
L.
,
Roberts
A. I.
,
Beavis
A.
,
Lopez-Botet
M.
,
Ebert
E. C.
,
Winchester
R. J.
.
2002
.
TCR specificity dictates CD94/NKG2A expression by human CTL.
Immunity
17
:
487
499
.
32
Li
Y.
,
Bleakley
M.
,
Yee
C.
.
2005
.
IL-21 influences the frequency, phenotype, and affinity of the antigen-specific CD8 T cell response.
J. Immunol.
175
:
2261
2269
.
33
Steinle
A.
,
Li
P.
,
Morris
D. L.
,
Groh
V.
,
Lanier
L. L.
,
Strong
R. K.
,
Spies
T.
.
2001
.
Interactions of human NKG2D with its ligands MICA, MICB, and homologs of the mouse RAE-1 protein family.
Immunogenetics
53
:
279
287
.
34
Wu
J.
,
Groh
V.
,
Spies
T.
.
2002
.
T cell antigen receptor engagement and specificity in the recognition of stress-inducible MHC class I-related chains by human epithelial gamma delta T cells.
J. Immunol.
169
:
1236
1240
.
35
Groh
V.
,
Steinle
A.
,
Bauer
S.
,
Spies
T.
.
1998
.
Recognition of stress-induced MHC molecules by intestinal epithelial gammadelta T cells.
Science
279
:
1737
1740
.
36
Groh
V.
,
Smythe
K.
,
Dai
Z.
,
Spies
T.
.
2006
.
Fas-ligand-mediated paracrine T cell regulation by the receptor NKG2D in tumor immunity.
Nat. Immunol.
7
:
755
762
.
37
Tang
F.
,
Chen
Z.
,
Ciszewski
C.
,
Setty
M.
,
Solus
J.
,
Tretiakova
M.
,
Ebert
E.
,
Han
J.
,
Lin
A.
,
Guandalini
S.
, et al
.
2009
.
Cytosolic PLA2 is required for CTL-mediated immunopathology of celiac disease via NKG2D and IL-15.
J. Exp. Med.
206
:
707
719
.
38
Valitutti
S.
,
Müller
S.
,
Salio
M.
,
Lanzavecchia
A.
.
1997
.
Degradation of T cell receptor (TCR)-CD3-zeta complexes after antigenic stimulation.
J. Exp. Med.
185
:
1859
1864
.
39
Verneris
M. R.
,
Karami
M.
,
Baker
J.
,
Jayaswal
A.
,
Negrin
R. S.
.
2004
.
Role of NKG2D signaling in the cytotoxicity of activated and expanded CD8+ T cells.
Blood
103
:
3065
3072
.
40
Eagle
R. A.
,
Trowsdale
J.
.
2007
.
Promiscuity and the single receptor: NKG2D.
Nat. Rev. Immunol.
7
:
737
744
.
41
Eagle
R. A.
,
Traherne
J. A.
,
Hair
J. R.
,
Jafferji
I.
,
Trowsdale
J.
.
2009
.
ULBP6/RAET1L is an additional human NKG2D ligand.
Eur. J. Immunol.
39
:
3207
3216
.
42
Weiss
A.
,
Littman
D. R.
.
1994
.
Signal transduction by lymphocyte antigen receptors.
Cell
76
:
263
274
.
43
Meresse
B.
,
Curran
S. A.
,
Ciszewski
C.
,
Orbelyan
G.
,
Setty
M.
,
Bhagat
G.
,
Lee
L.
,
Tretiakova
M.
,
Semrad
C.
,
Kistner
E.
, et al
.
2006
.
Reprogramming of CTLs into natural killer-like cells in celiac disease.
J. Exp. Med.
203
:
1343
1355
.
44
Tang
Q.
,
Grzywacz
B.
,
Wang
H.
,
Kataria
N.
,
Cao
Q.
,
Wagner
J. E.
,
Blazar
B. R.
,
Miller
J. S.
,
Verneris
M. R.
.
2008
.
Umbilical cord blood T cells express multiple natural cytotoxicity receptors after IL-15 stimulation, but only NKp30 is functional.
J. Immunol.
181
:
4507
4515
.
45
Brando
C.
,
Mukhopadhyay
S.
,
Kovacs
E.
,
Medina
R.
,
Patel
P.
,
Catina
T. L.
,
Campbell
K. S.
,
Santoli
D.
.
2005
.
Receptors and lytic mediators regulating anti-tumor activity by the leukemic killer T cell line TALL-104.
J. Leukoc. Biol.
78
:
359
371
.
46
Cooper
M. A.
,
Fehniger
T. A.
,
Caligiuri
M. A.
.
2001
.
The biology of human natural killer-cell subsets.
Trends Immunol.
22
:
633
640
.
47
Nakajima
H.
,
Colonna
M.
.
2000
.
2B4: an NK cell activating receptor with unique specificity and signal transduction mechanism.
Hum. Immunol.
61
:
39
43
.
48
Jang
I. K.
,
Gu
H.
.
2003
.
Negative regulation of TCR signaling and T-cell activation by selective protein degradation.
Curr. Opin. Immunol.
15
:
315
320
.
49
Correa
M. R.
,
Ochoa
A. C.
,
Ghosh
P.
,
Mizoguchi
H.
,
Harvey
L.
,
Longo
D. L.
.
1997
.
Sequential development of structural and functional alterations in T cells from tumor-bearing mice.
J. Immunol.
158
:
5292
5296
.
50
Brundula
V.
,
Rivas
L. J.
,
Blasini
A. M.
,
París
M.
,
Salazar
S.
,
Stekman
I. L.
,
Rodríguez
M. A.
.
1999
.
Diminished levels of T cell receptor zeta chains in peripheral blood T lymphocytes from patients with systemic lupus erythematosus.
Arthritis Rheum.
42
:
1908
1916
.
51
Takahashi
A.
,
Kono
K.
,
Amemiya
H.
,
Iizuka
H.
,
Fujii
H.
,
Matsumoto
Y.
.
2001
.
Elevated caspase-3 activity in peripheral blood T cells coexists with increased degree of T-cell apoptosis and down-regulation of TCR zeta molecules in patients with gastric cancer.
Clin. Cancer Res.
7
:
74
80
.
52
Wieckowski
E.
,
Wang
G. Q.
,
Gastman
B. R.
,
Goldstein
L. A.
,
Rabinowich
H.
.
2002
.
Granzyme B-mediated degradation of T-cell receptor zeta chain.
Cancer Res.
62
:
4884
4889
.
53
Torelli
G. F.
,
Paolini
R.
,
Tatarelli
C.
,
Soriani
A.
,
Vitale
A.
,
Guarini
A.
,
Santoni
A.
,
Foa
R.
.
2003
.
Defective expression of the T-cell receptor-CD3 zeta chain in T-cell acute lymphoblastic leukaemia.
Br. J. Haematol.
120
:
201
208
.
54
Tsuzaka
K.
,
Fukuhara
I.
,
Setoyama
Y.
,
Yoshimoto
K.
,
Suzuki
K.
,
Abe
T.
,
Takeuchi
T.
.
2003
.
TCR zeta mRNA with an alternatively spliced 3′-untranslated region detected in systemic lupus erythematosus patients leads to the down-regulation of TCR zeta and TCR/CD3 complex.
J. Immunol.
171
:
2496
2503
.
55
Krishnan
S.
,
Kiang
J. G.
,
Fisher
C. U.
,
Nambiar
M. P.
,
Nguyen
H. T.
,
Kyttaris
V. C.
,
Chowdhury
B.
,
Rus
V.
,
Tsokos
G. C.
.
2005
.
Increased caspase-3 expression and activity contribute to reduced CD3zeta expression in systemic lupus erythematosus T cells.
J. Immunol.
175
:
3417
3423
.
56
Crispín
J. C.
,
Kyttaris
V. C.
,
Juang
Y. T.
,
Tsokos
G. C.
.
2008
.
How signaling and gene transcription aberrations dictate the systemic lupus erythematosus T cell phenotype.
Trends Immunol.
29
:
110
115
.
57
Vaknin
I.
,
Blinder
L.
,
Wang
L.
,
Gazit
R.
,
Shapira
E.
,
Genina
O.
,
Pines
M.
,
Pikarsky
E.
,
Baniyash
M.
.
2008
.
A common pathway mediated through Toll-like receptors leads to T- and natural killer-cell immunosuppression.
Blood
111
:
1437
1447
.
58
Díaz-Benítez
C. E.
,
Navarro-Fuentes
K. R.
,
Flores-Sosa
J. A.
,
Juárez-Díaz
J.
,
Uribe-Salas
F. J.
,
Román-Basaure
E.
,
González-Mena
L. E.
,
Alonso de Ruíz
P.
,
López-Estrada
G.
,
Lagunas-Martínez
A.
, et al
.
2009
.
CD3zeta expression and T cell proliferation are inhibited by TGF-beta1 and IL-10 in cervical cancer patients.
J. Clin. Immunol.
29
:
532
544
.
59
Caplan
S.
,
Baniyash
M.
.
2000
.
Searching for significance in TCR-cytoskeleton interactions.
Immunol. Today
21
:
223
228
.
60
Nambiar
M. P.
,
Krishnan
S.
,
Warke
V. G.
,
Tsokos
G. C.
.
2004
.
TCR zeta-chain abnormalities in human systemic lupus erythematosus.
Methods Mol. Med.
102
:
49
72
.
61
Takeuchi
T.
,
Tsuzaka
K.
,
Pang
M.
,
Amano
K.
,
Koide
J.
,
Abe
T.
.
1998
.
TCR zeta chain lacking exon 7 in two patients with systemic lupus erythematosus.
Int. Immunol.
10
:
911
921
.
62
Gastman
B. R.
,
Johnson
D. E.
,
Whiteside
T. L.
,
Rabinowich
H.
.
1999
.
Caspase-mediated degradation of T-cell receptor zeta-chain.
Cancer Res.
59
:
1422
1427
.
63
Menné
C.
,
Lauritsen
J. P.
,
Dietrich
J.
,
Kastrup
J.
,
Wegener
A. K.
,
Andersen
P. S.
,
Odum
N.
,
Geisler
C.
.
2001
.
T-cell receptor downregulation by ceramide-induced caspase activation and cleavage of the zeta chain.
Scand. J. Immunol.
53
:
176
183
.
64
Fischer
U.
,
Jänicke
R. U.
,
Schulze-Osthoff
K.
.
2003
.
Many cuts to ruin: a comprehensive update of caspase substrates.
Cell Death Differ.
10
:
76
100
.
65
Maki
A.
,
Matsuda
M.
,
Asakawa
M.
,
Kono
H.
,
Fujii
H.
,
Matsumoto
Y.
.
2004
.
Decreased expression of CD28 coincides with the down-modulation of CD3zeta and augmentation of caspase-3 activity in T cells from hepatocellular carcinoma-bearing patients and hepatitis C virus-infected patients.
J. Gastroenterol. Hepatol.
19
:
1348
1356
.
66
Chowdhury
I.
,
Tharakan
B.
,
Bhat
G. K.
.
2008
.
Caspases - an update.
Comp. Biochem. Physiol. B Biochem. Mol. Biol.
151
:
10
27
.
67
Wilhelm
S.
,
Wagner
H.
,
Häcker
G.
.
1998
.
Activation of caspase-3-like enzymes in non-apoptotic T cells.
Eur. J. Immunol.
28
:
891
900
.
68
Alam
A.
,
Cohen
L. Y.
,
Aouad
S.
,
Sékaly
R. P.
.
1999
.
Early activation of caspases during T lymphocyte stimulation results in selective substrate cleavage in nonapoptotic cells.
J. Exp. Med.
190
:
1879
1890
.
69
Kennedy
N. J.
,
Kataoka
T.
,
Tschopp
J.
,
Budd
R. C.
.
1999
.
Caspase activation is required for T cell proliferation.
J. Exp. Med.
190
:
1891
1896
.
70
Conejo-Garcia
J. R.
,
Benencia
F.
,
Courreges
M. C.
,
Gimotty
P. A.
,
Khang
E.
,
Buckanovich
R. J.
,
Frauwirth
K. A.
,
Zhang
L.
,
Katsaros
D.
,
Thompson
C. B.
, et al
.
2004
.
Ovarian carcinoma expresses the NKG2D ligand Letal and promotes the survival and expansion of CD28− antitumor T cells.
Cancer Res.
64
:
2175
2182
.
71
Sutherland
C. L.
,
Chalupny
N. J.
,
Schooley
K.
,
VandenBos
T.
,
Kubin
M.
,
Cosman
D.
.
2002
.
UL16-binding proteins, novel MHC class I-related proteins, bind to NKG2D and activate multiple signaling pathways in primary NK cells.
J. Immunol.
168
:
671
679
.
72
Li
C.
,
Ge
B.
,
Nicotra
M.
,
Stern
J. N.
,
Kopcow
H. D.
,
Chen
X.
,
Strominger
J. L.
.
2008
.
JNK MAP kinase activation is required for MTOC and granule polarization in NKG2D-mediated NK cell cytotoxicity.
Proc. Natl. Acad. Sci. USA
105
:
3017
3022
.
73
Zapata
J. M.
,
Takahashi
R.
,
Salvesen
G. S.
,
Reed
J. C.
.
1998
.
Granzyme release and caspase activation in activated human T-lymphocytes.
J. Biol. Chem.
273
:
6916
6920
.
74
Harvey
N. L.
,
Trapani
J. A.
,
Fernandes-Alnemri
T.
,
Litwack
G.
,
Alnemri
E. S.
,
Kumar
S.
.
1996
.
Processing of the Nedd2 precursor by ICE-like proteases and granzyme B.
Genes Cells
1
:
673
685
.
75
Los
M.
,
Stroh
C.
,
Jänicke
R. U.
,
Engels
I. H.
,
Schulze-Osthoff
K.
.
2001
.
Caspases: more than just killers?
Trends Immunol.
22
:
31
34
.
76
Olaussen
R. W.
,
Karlsson
M. R.
,
Lundin
K. E.
,
Jahnsen
J.
,
Brandtzaeg
P.
,
Farstad
I. N.
.
2007
.
Reduced chemokine receptor 9 on intraepithelial lymphocytes in celiac disease suggests persistent epithelial activation.
Gastroenterology
132
:
2371
2382
.
77
Tomasello
E.
,
Desmoulins
P. O.
,
Chemin
K.
,
Guia
S.
,
Cremer
H.
,
Ortaldo
J.
,
Love
P.
,
Kaiserlian
D.
,
Vivier
E.
.
2000
.
Combined natural killer cell and dendritic cell functional deficiency in KARAP/DAP12 loss-of-function mutant mice.
Immunity
13
:
355
364
.
78
Wee
L. J.
,
Tan
T. W.
,
Ranganathan
S.
.
2007
.
CASVM: web server for SVM-based prediction of caspase substrates cleavage sites.
Bioinformatics
23
:
3241
3243
.