Logo of pnasPNASInfo for AuthorsSubscriptionsAboutThis Article
Proc Natl Acad Sci U S A. 2004 Feb 10; 101(6): 1696–1701.
Published online 2004 Jan 26. doi:  10.1073/pnas.0308231100
PMCID: PMC341820

T cell receptor ζ reconstitution fails to restore responses of T cells rendered hyporesponsive by tumor necrosis factor α


Expression and function of the antigen T cell receptor (TCR) play a central role in regulating immune responsiveness. Accordingly, targeting the expression of TCRαβ or its associated CD3 subunits profoundly influences T cell development and adaptive immunity. Down-regulation of the invariant TCRζ chain has been documented in a wide variety of chronic inflammatory and infectious diseases, and is thought to contribute to the paradoxical immune suppression observed in these diseases. Previously, we reported that prolonged exposure of T cell hybridoma clones to tumor necrosis factor α (TNF) induces nondeletional and reversible hyporesponsiveness to TCR engagement, associated with down-regulation of TCRζ chain expression, impaired TCR/CD3 complex assembly, and attenuation of TCR-induced membrane proximal tyrosine phosphorylation. Here, we have tested whether receptor specific T cell responses are rescued in TNF-treated T cell hybridomas by retroviral-mediated expression of ζ-chimeric (C2ζ) receptors or wild-type TCRζ. Expression of C2ζ receptors at the cell surface is relatively refractory to chronic TNF stimulation. However, C2ζ receptor function depends on association with endogenous TCRζ chains, whose expression is down-regulated by TNF, and so C2 receptor specific responses are attenuated in TNF-treated T cells. Unexpectedly, overexpression of wild-type TCRζ maintains cell surface TCR/CD3 complex expression but fails to rescue receptor proximal signaling in TNF-treated T cells, suggesting the existence of hitherto unrecognized mechanisms through which TNF regulates T cell responsiveness. We provide additional evidence that TNF also uncouples distal TCR signaling pathways independently of its effects on TCRζ expression.

The assembly and expression of lymphocyte antigen receptors at the cell surface is central to the acquisition of immune responsiveness (reviewed in ref. 1). The mature T cell receptor (TCR) complex comprises at least seven transmembrane subunits. Disulfide-linked TCRα and -β chains serve as clonotypic antigen recognition modules (2), which form noncovalent associations with invariant receptor chains CD3γ, δ, ε, and ζ through electrostatic interactions within their transmembrane domains (1, 3, 4). TCRζ is expressed as a homodimer, linked by a disulfide bond within the transmembrane domain (5). The cytoplasmic domains of the TCR invariant chains contain signal-transducing, immune-receptor tyrosine-based activation motif (ITAM) sequences; CD3γ, δ, and ε each have a single ITAM, whereas each TCRζ chain contains three (6, 7). Upon receptor ligation, tyrosines within these motifs are phosphorylated by Src family kinases to form docking sites for Syk family tyrosine kinases, such as ζ-associated protein of 70 kDa (ZAP-70), which contain Src homology 2 domains. These complexes transmit signals to downstream pathways for T cell differentiation and effector function (8).

Both TCRζ and cell surface TCR/CD3 complexes are down-regulated within minutes of TCR ligation through internalization of receptor complexes (9). Most CD3 subunits recycle to the cell surface, but TCRζ cycles independently, much of the cellular pool being degraded in lysosomes (10, 11). As a result, reexpression of fully functional TCR/CD3 complexes depends, at least in part, on de novo synthesis of TCRζ. These findings are consistent with studies in T cell hybridomas that demonstrate that the rate of TCRζ biosynthesis limits that of TCR/CD3 complex assembly at the cell surface (12, 13). These studies predict that physiological or pathological attenuation of TCRζ expression or signaling function should down-modulate T cell reactivity to repeated antigen stimulation.

Sustained loss of TCRζ expression in CD4+ and CD8+ T cells has been documented in a variety of chronic inflammatory and infectious diseases (14-17), as well as in tumor-infiltrating lymphocytes (18). Mechanisms reported to contribute to this include TCRζ gene mutations (19, 20), repression of TCRζ gene transcription through defective production of the Ets family transcription factor Elf-1 (21), and regulation of TCRζ mRNA stability (22). Proteolytic degradation by caspases, granzyme B, or through ubiquitin-dependent pathways has also been described (23-25). Furthermore, it has been widely acknowledged that loss of TCRζ expression might account for the T cell hyporesponsiveness, attenuation of TCR signaling, and suppression of cell-mediated immunity characteristic of these diseases in man. On the other hand, gene targeting experiments in mice have suggested convincingly that loss of TCRζ function, as opposed to expression, results in rather more modest effects on T cell responsiveness (26). Attempts to establish the precise functional significance of reduced TCRζ expression in chronic inflammatory or infectious diseases have been hindered, at least in part, by the lack of robust models for investigating chronic TCRζ down-regulation.

We recently reported that prolonged exposure of murine T cell hybridomas to the proinflammatory cytokine tumor necrosis factor α (TNF) for up to 8 days induces nondeletional and reversible hyporesponsiveness to TCR ligation (27), findings previously documented in both primary human and murine T cells (28, 29). Of the TCR signaling subunits, only expression of TCRζ was affected. Although the rate of TCRζ protein degradation was unchanged, TNF reduced TCRζ mRNA and protein expression, leading to impaired assembly of functional TCR/CD3 complexes at the cell surface. These findings prompted us to speculate that sustained inflammatory cytokine expression could account for the loss of TCRζ chain expression during terminal differentiation of effector T cells at sites of inflammation.

We have developed our in vitro model to explore the contribution of TCRζ down-regulation to T cell hyporesponsiveness in the context of chronic inflammatory disease. Specifically, we sought to render T cells refractory to the suppressive effects of TNF by generating clones either expressing chimeric receptors with integral TCRζ signaling subunits or overexpressing wild-type TCRζ. Our results demonstrate that overexpression of neither chimeric ζ receptors nor wild-type TCRζ are sufficient to restore membrane proximal TCR signaling in TNF-treated T cells. Unexpectedly, we find that chronic TNF stimulation suppresses distal TCR signaling pathways independently of its effects on TCRζ expression, providing further evidence for the potent immunomodulatory effects of this cytokine.

Materials and Methods

Cells and Cell Culture. The derivation of the mouse T cell hybridoma clone 11A2, which is specific for human cartilage glycoprotein-39 (HCgp-39), restricted by HLA-DR4, and expresses human CD4, has been described (27, 30). 11A2 cells and all derived clones were cultured in RPMI medium 1640 supplemented with 25 mM Hepes/2 mM l-glutamine/10% heatinactivated FCS/100 units/ml penicillin/100 μg/ml streptomycin/1 mM sodium pyruvate/50 μM 2-mercaptoethanol, at 37°C and 5% CO2. Every 48 h, cells were passaged into fresh complete medium in the presence or absence of recombinant mouse TNF (BD Pharmingen), or human TNF (a gift of Z. Kaymakcalan, BASF Bioresearch, Cambridge, MA) at the concentrations indicated. At such concentrations, mouse and human TNF have similar effects on murine T cells (27), and both have been used in these studies.

Retroviral Constructs and Transduction. The chimeric immunoreceptor C2ζ has been described (31). The TCRζ-IRES-eGFP construct was generated as follows. The DNA fragment encoding the internal ribosomal entry site (IRES) of murine encephalomyocarditis virus was excised from pCITE-1 (Novagen) and a fragment encoding enhanced GFP (eGFP) was excised from the expression vector EGFP-1 (Clontech) before subcloning both fragments into the expression vector pcDNA3. The IRES-eGFP fragment was excised from this vector and subcloned together with a PCR-derived fragment of full-length TCRζ cDNA from pGEM-3Z (a gift from A. Weissman, National Cancer Institute, National Institutes of Health, Bethesda) into the retroviral vector miniMFG, to yield the final TCRζ-IRES-eGFP insert. The vector miniMFG was derived from the original vector MFG, as described (31, 32). All inserts were verified by DNA sequencing. Supernatants from PT67 packaging cells (Clontech) expressing replication defective retrovirus were produced as described (33), before centrifugation-assisted transduction of 11A2 T cell hybridomas. C2ζ expressing clones were generated by limiting dilution and screened for C2ζ expression by flow cytometry, as described below. GFP-expressing T cells were identified by flow cytometry, and lines generated by cell sorting using a FACS-Vantage SE cell sorter (Becton Dickinson) before cloning by limiting dilution. Transduced T cells were cultured as described above.

Flow Cytometric Analysis. Cells were washed and labeled with one of the following antibodies, using isotype-matched mAb or normal serum as controls: rabbit anti-C2 antiserum (34); phycoerythrin-conjugated anti-rabbit Ig Ab, FITC-conjugated antihuman CD4 mAb (Sigma); phycoerythrin-conjugated anti-CD3ε (clone 17A2), FITC-conjugated anti-CD3ε (clone 145-2C11), and anti-TCRζ (clone 6B10.2) mAbs (all from BD Pharmingen); or phycoerythrin-conjugated anti-TCRζ mAb (clone 6B10.2) (Santa Cruz Biotechnology). Cells were fixed and permeabilized before staining for intracellular TCRζ. Labeled cells were acquired on an LSR BD fluorescence-activated cell sorter (FACS) analyzer (Becton Dickinson), and data were analyzed by using cellquest software.

Cell Stimulation. Cells were harvested, washed and resuspended in complete medium at 106 cells per ml in the absence of TNF, before incubation with either agonistic anti-CD3ε mAb 145-2C11 (BD Pharmingen), native bovine collagen II (a gift from R. Williams, London) bound to 96-well flat-bottomed plates, or with 10 ng/ml phorbol 12-myristate 13-acetate (PMA) and 50 ng/ml ionomycin (Sigma). After 24 h, IL-2 was assayed by using rat anti-mouse IL-2 antibody pairs (BD Pharmingen), as described (27). Receptor ligation for induction of tyrosine phosphorylation and downstream signaling events using anti-CD3ε mAb145-2C11 or anti-C2 anti-serum (or relevant control Ab), together with mouse anti-human CD4 mAb (clone OKT-4), has been described in detail elsewhere (27).

Cell Lysis, Immunoprecipitation, and Immunoblotting Analysis. Cells were lysed at 107 cells per ml in ice-cold isotonic 1% Nonidet P-40 lysis buffer in the presence of protease and phosphatase inhibitors. A modified lysis buffer was used for immunoprecipitation of C2ζ which included 0.5% Triton X-100 and 100 mM NaCl. For immunoprecipitations, lysates were precleared with protein A agarose (Sigma) before incubation with one of the following antibodies: anti-C2 anti-serum; anti-ZAP-70 sc-574; anti-phospholipase Cγ1 sc-81 (both from Santa Cruz Biotechnology); anti-linker for activation of T cells (LAT) 06-807 (Upstate Biotechnology), all at 10 μg/ml, and immunoprecipitation by standard techniques. Whole cell lysates or immunoprecipitated proteins were resolved by SDS/PAGE in precast Tris-glycine or Bis-Tris gels (Invitrogen), according to the suppliers' instructions. Proteins were transferred onto nitrocellulose and probed with the following primary antibodies: antiphosphotyrosine mAb clone 4G10 (Upstate Biotechnology); anti-TCRζ mAb clone 8D3 (recognizes monomer and oligomers of ζ) (BD Pharmingen) or clone 6B10.2 (does not recognize monomeric ζ); anti-ZAP-70 hybridoma supernatant (a gift from Glaxo-Smithkline, Stevenage, U.K.); anti-LAT and anti-C2 Abs, as above; anti-phospholipase Cγ1 mAb (Upstate Biotechnology); anti-c-fos sc-52-G (Santa Cruz Biotechnology). After incubation with horseradish peroxidase-conjugated anti-Ig Ab (DAKO), proteins were visualized by enhanced chemiluminescence using Hyperfilm (Amersham Pharmacia). Immunoblots were scanned by using a GS-710 Calibrated Imaging Densitometer with quantity one software (Bio-Rad), and densities of individual bands were measured by using phoretix 1d software (Non-Linear Dynamics, Newcastle-upon-Tyne, U.K.). Relative expression of TCRζ was calculated after normalization of band densities in each sample, using α-tubulin as a reference.


Expression of Single-Chain Fv Antibody C2 (scFvC2)-CD8-TCRζ Chimeric Receptors in T Cell Hybridomas. In light of our previous studies indicating that TNF suppresses TCRζ expression at the mRNA level (27), we set out to reconstitute TNF-treated T cells by retroviral gene transduction with TCRζ by using two different approaches. First, we generated clones expressing a collagen II specific receptor C2ζ. This chimeric receptor is comprised of the TCRζ transmembrane and cytoplasmic domains coupled to an ectodomain consisting of the human CD8α hinge region and a single-chain Fv sequence derived from the monoclonal antibody C2, specific for type II collagen (31). Abundant surface C2ζ expression was observed by flow cytometry for clone 2D3, but not for the parental clone 11A2 (Fig. 6A, which is published as supporting information on the PNAS web site), whereas both clones showed comparable expression of CD4 (Fig. 6B). Whereas chronic stimulation of 11A2 cells with nontoxic concentrations of TNF for 8 days induced down-regulation of TCR/CD3 expression in 11A2 cells by ≈50% (Fig. 1A), TNF induced only a modest reduction in C2ζ expression on 2D3 cells (Fig. 1B); in some experiments, this reduction was not detected (data not shown). We next studied the effects of TNF on IL-2 production after receptor engagement. 11A2 cells responded vigorously to TCR but not to C2 ligation (Fig. 1C), and TNF suppressed TCR-induced IL-2 production by 95% when compared to control cells. C2 receptor stimulation with plate-bound collagen type II induced IL-2 production by 2D3 cells, and this too was attenuated by TNF, albeit to a lesser extent (68% inhibition of IL-2 production; mean of eight independent experiments). Thus, although surface C2ζ receptor expression appeared to be relatively refractory to chronic TNF treatment, receptor-induced IL-2 production was nonetheless suppressed.

Fig. 1.
Effects of TNF on expression and function of TCR and C2ζ receptor. 11A2 and 2D3 cells were cultured with or without 2.5 ng/ml TNF for 8 days. (A and B) TCR expression on 11A2 cells was determined by flow cytometry (FACS) using anti-CD3ε ...

C2ζ Receptors Are Expressed as Higher-Order Complexes with Endogenous TCRζ Chains. To characterize the mechanism whereby TNF attenuated C2-specific responses, we examined the stoichiometry of C2ζ receptor complexes in 2D3 cells. Cell lysates were resolved by SDS/PAGE and immunoblotted with anti-ζ Ab. For 11A2 cells, TCRζ resolved as a 32-kDa homodimer under nonreducing conditions, whereas the predominant TCRζ complex in 2D3 cells migrated at 130-140 kDa (Fig. 2A). TCRζ-immunoreactive complexes of intermediate molecular mass (MW) were also detected in 2D3 cells, but at lower levels, and expression of endogenous TCRζ homodimers could only be detected in overexposed blots (data not shown). Immunoblotting of cell lysates from control and TNF treated cells showed that TNF down-regulated expression of TCRζ homodimers in 11A2 cells, as well as the major C2ζ complex in 2D3 cells, by ≈50% (Fig. 2B). To test whether C2ζ receptors were expressed as higher-order complexes with endogenous TCRζ chains, we immunoprecipitated C2ζ receptors with anti-C2 Ab and examined precipitates for the presence of associated TCRζ molecules. Anti-C2 immunoprecipitates contained both the 52-kDa monomeric C2ζ receptor and an 18/19-kDa doublet, likely to represent TCRζ monomers (Fig. 2C). The precise molecular nature of the 18/19-kDa doublet is not known; it does not appear to represent phospho-ζ species because we could not detect constitutively phosphorylated TCRζ in anti-C2 immunoprecipitates (data not shown). Nonetheless, levels of C2ζ-associated TCRζ were reduced in TNF-treated cells. Consistent with this finding, levels of total cellular ζ chain expression were reduced by ≈40% in TNF-treated cells relative to untreated 2D3 cells, as determined by intracellular labeling and flow cytometry (Fig. 7A, which is published as supporting information on the PNAS web site). This reduction in TCRζ was not as marked as that seen in TNF-treated 11A2 cells (≈87%) by using the same method (Fig. 7B).

Fig. 2.
C2ζ receptors associate with endogenous TCRζ, and their function is sensitive to the effects of TNF. 11A2 and 2D3 cells were cultured with or without 2.5 ng/ml TNF for 8 days. Lysates were resolved by SDS/PAGE under nonreducing conditions, ...

Two observations suggested that the association between endogenous TCRζ and C2ζ was functionally significant. First, surface CD3ε expression on 2D3 cells was ≈10-fold lower than levels expressed on 11A2 cells (Fig. 2D), a finding consistent with the inability of TCR agonists to stimulate significant IL-2 production in 2D3 cells (see Fig. 1C). Second, after C2 receptor engagement, tyrosine phosphorylation of ZAP-70, LAT, and phospholipase Cγ1 was reduced in immunoprecipitates from TNF treated cells when compared to control cells (Fig. 8, which is published as supporting information on the PNAS web site). Although we cannot rule out an association between the C2ζ receptor and other CD3 subunits, these data suggested that productive C2ζ signaling depends on its association with endogenous TCRζ chains. They also demonstrated that retroviral-mediated expression of a ζ chimeric receptor is not sufficient to reverse the immunomodulatory effects of TNF.

Stable Overexpression of TCRζ in T Cells. As an alternative strategy to assess the contribution of TCRζ down-regulation to TNF-induced T cell hyporesponsiveness, 11A2 cells were transduced with a bicistronic retroviral construct encoding wild-type TCRζ and eGFP. GFP+ cells were selected by cell sorting and sub-cloned (Fig. 3A). Two stable GFP+ clones, 3F2 and 1E2, were studied in detail and found to express TCRζ mRNA and protein at levels reflecting GFP expression (Fig. 3B and data not shown). Densitometric analysis of immunoblots showed that TCRζ protein expression was 2- to 3-fold greater in clones 3F2 and 1E2 than in parental 11A2 cells. To determine whether ζ overexpression translated to detectable differences in TCR signaling, we first compared TCR-induced proximal tyrosine phosphorylation in each clone but did not detect increased phosphorylation in either 3F2 or 1E2 cells. It is possible that levels of TCRζ in 11A2 cells are sufficient for maximal tyrosine phosphorylation immediately after TCR engagement. Accordingly, we assessed the induction of the AP-1 family member c-fos after more sustained TCR ligation. TCR-induced induction of c-fos and its phosphorylation, indicated by molecular weight shift, were enhanced and more sustained in ζ-overexpressing T cells, and were most noticeable 2 and 4 h after stimulation (Fig. 9 Upper, which is published as supporting information on the PNAS web site). Consistent with this finding, TCR-induced IL-2 production by 3F2 and 1E2 cells was 2- to 3-fold higher than that of parental 11A2 cells (Fig. 3C), differences that could also be seen only 4 h after stimulation via the TCR (Fig. 9 Lower).

Fig. 3.
Stable, functional overexpression of TCRζ in T cells. Clone 11A2 was transduced with retrovirus encoding TCRζ-IRES-eGFP, and GFP+ lines were generated by cell sorting. (A) GFP expression in transduced clones 3F2 and 1E2 was determined ...

Overexpression of TCRζ Maintains TCR Expression but Fails to Rescue TCR Signaling in TNF-Treated Cells. We compared the effect of chronic TNF on the expression of TCRζ protein in clone 1E2 and parallel cultures of parental 11A2 cells. Although TNF induced modest suppression of TCRζ protein in 1E2 cells as detected by immunoblotting, the amount of TCRζ protein in the TNF-treated 1E2 cells was greater than that of either control or TNF-treated 11A2 cells (Fig. 4A, compare lane 1 with lanes 5 and 6). These data indicated that stable retroviral-mediated reconstitution of TCRζ expression had been achieved in TNF-treated 1E2 cells at the protein level. Furthermore, flow cytometry experiments confirmed that increased expression of TCRζ in 1E2 cells could overcome TNF-induced down-regulation of surface TCR/CD3 expression observed in parental 11A2 cells (compare Fig. 1 A with Fig. 4B).

Fig. 4.
Reconstitution of TNF-treated T cells with TCRζ chain. 11A2 and 1E2 cells were cultured with or without TNF (2.5 ng/ml) for 8 days. (A) TCRζ expression relative to that in control 11A2 cells was evaluated by immunoblot as in Fig. 3B.( ...

To establish whether TCRζ reconstitution rescued receptorproximal signaling in TNF treated cells, we compared TCR-induced tyrosine phosphorylation of ZAP-70 and LAT in control and TNF-treated 1E2 cells with the parental clone. We found that TCR-induced phosphorylation of ZAP-70 and LAT was reduced in immunoprecipitates from both 1E2 and parental 11A2 cells after TNF stimulation (Fig. 5A). Furthermore, TCR-induced c-fos was decreased in TNF-treated 11A2 and 1E2 cells (Fig. 10, which is published as supporting information on the PNAS web site). Calcium flux after TCR stimulation of 1E2 cells was also significantly attenuated by TNF (data not shown). Moreover, the suppression by chronic TNF treatment of TCR-induced IL-2 production was as marked in 1E2 as in 11A2 cells (Fig. 5B). These results confirmed that overexpression of TCRζ in TNF-treated T cells was not sufficient to restore TCR-induced IL-2 production. Finally, to determine whether suppression of TCR-induced IL-2 production was due exclusively to changes at the level of TCR proximal signaling, or to additional effects of TNF on distal pathways, we examined IL-2 production by control and TNF-treated cells stimulated with phorbol ester and calcium ionophore. Although PMA and ionomycin induced comparable IL-2 production in 1E2 and 11A2 cells, this response was suppressed in cells pretreated with TNF, to a degree similar to that seen for suppression of TCR-induced IL-2 responses (Fig. 5C). The results of these experiments provide evidence that chronic stimulation of T cells with TNF uncouples TCR signaling pathways through distinct mechanisms that are independent of its effects on TCRζ expression.

Fig. 5.
TCRζ overexpression fails to reconstitute receptor proximal TCR signaling in TNF-treated T cells. (A) Cells were cultured with or without 2.5 ng/ml TNF for 8 days before stimulation with anti-CD3ε and anti-hCD4 for 2 min. Immunoprecipitates ...


Gene transfer into T cell hybridomas has proved to be a useful approach for exploring the contribution of individual CD3 subunits to TCR assembly, expression, and signaling. For instance, the TCRneg BW cell fusion partner, from which immortalized T cell hybridomas are derived, was found to be CD3δ and CD3ζ deficient. Reconstitution experiments demonstrated that both chains were necessary for TCR/CD3 expression and function (35). To gain insights into T cell activation and effector responses in chronic inflammatory disease, we have been investigating the effects of TNF on TCR signaling. In previous experiments, we used a chimeric receptor made up of a singlechain Fv extracellular domain coupled to an immune-receptor tyrosine-based activation motif-deficient FcεRI γ chain signaling subunit (C2γIC-) (27). This receptor construct associated with endogenous TCRζ chains and required TCRζ for signaling, but not for receptor expression. Although cell surface expression of C2γ receptors was unimpaired in TNF-treated T cells, receptor-specific IL-2 production was suppressed (27). Because expression of CD3ε, γ, and δ were stable in TNF-treated T cells, we concluded that the selective effect of TNF on TCRζ expression was responsible for the attenuation of receptor-dependent signaling in T cells. However, expression of C2γ in T cell hybridomas did not permit an analysis of the contribution of TCRζ down-regulation to the immunomodulatory effects of TNF.

The C2ζ receptor provided a surrogate antigen receptor that should signal via its integral TCRζ domain independently of endogenous TCR/CD3 complexes. The findings of Brocker and colleagues (36), and those of Annenkov and Chernajovsky (31) supported this notion. Indeed, both groups demonstrated that chimeric receptors of similar or identical structure were expressed in T cell hybridomas as homodimers, and more importantly, did not associate with TCRζ or other CD3 subunits. It was therefore surprising that in 11A2 T cells, C2ζ receptors were expressed as higher order heteromeric complexes of ≈130-140 kDa. Based on the apparent MW of these complexes, and the fact that endogenous TCRζ chains coimmunoprecipitated with the C2ζ receptor, we believe that the 130- to 140-kDa complex represents a divalent trimer or tetramer comprised of two C2ζ subunits associated with endogenous TCRζ chains. In functional terms, the association between C2ζ and TCRζ molecules had two consequences. Firstly, abundant expression of C2ζ brought about reduced surface expression of TCR/CD3 on 2D3 cells, presumably because of the sequestering of TCRζ from the TCR/CD3 complex, highlighting further the integral role of TCRζ in TCR expression and function. Secondly, because signaling via C2ζ in 2D3 cells appeared to depend on association with endogenous TCRζ, it remained susceptible to chronic stimulation with TNF, despite the fact that cell surface expression of C2ζ was relatively spared.

As an alternative approach we generated stable T cell clones coexpressing TCRζ and eGFP from a bicistronic retroviral construct. Functional overexpression of TCRζ was demonstrated by stabilization of cell surface TCR/CD3, the kinetics of c-fos induction, and robust increases in TCR-induced IL-2 production. Reconstitution of TCRζ was achieved in clones that maintained higher levels of TCRζ mRNA and protein despite chronic TNF stimulation, when compared to the parental clone (Fig. 4 and data not shown). Nonetheless, reconstitution of TCRζ and recovery of TCR/CD3 expression at the cell surface was not sufficient to rescue receptor-proximal signaling in TNF treated cells. The target or targets of TNF that bring about impaired tyrosine phosphorylation, despite ζ overexpression, have yet to be defined. Of the Src family kinases tested, the expression and kinase activity of Lck is unaltered by TNF (27). One mechanism for attenuated proximal phosphorylation, which might occur despite an abundance of TCRζ, could be through dissociation of TCRζ from TCR signaling complexes. For example, TNF might inhibit the translocation of TCRζ into plasma membrane signaling domains upon TCR engagement, perhaps through chronic activation of sphingomyelinase and remodeling of membrane lipid rafts, as reported recently (37). However, when we compared detergent soluble and insoluble fractions for the expression of TCRζ we found no clear differences between control and TNF-treated T cells, either constitutively or upon TCR ligation (J.M.C. and A.P.C., unpublished data). An alternative possibility is suggested by reports that the inflammatory process attenuates TCR-mediated signaling by targeting the expression and function of transmembrane adaptor proteins, such as LAT, through mechanisms driven by chronic oxidative stress (38, 39). In the absence of detectable changes in LAT expression or subcellular localization, we discovered that TNF reduces the expression of another transmembrane adaptor protein, the T cell receptor-interacting molecule (TRIM). TRIM associates directly with TCRζ, comodulates with TCR complexes upon TCR ligation (40, and A.P.C., P.I. and B. Schraven, unpublished data), and has been implicated in regulating the rate of internalization of TCR/CD3 complexes at the cell surface (41). However, its precise role in transducing signals to down-stream pathways has yet to be defined. According to our data, overexpression of TCRζ in T cell hybridomas may be sufficient to compensate to some extent for reduced TRIM expression in terms of TCR/CD3 complex assembly. On the other hand, rescue of receptor proximal tyrosine phosphorylation in TNF-treated T cells may necessitate reconstitution of cells with both TCRζ and TRIM.

Stimulation of T cells with reagents that bypass receptorproximal events revealed that TNF has additional effects on downstream TCR signaling pathways in T cell hybridomas. This is not to say that down-regulation of TCRζ by TNF is functionally insignificant. Rather, the data suggest that TNF attenuates T cell activation through a number of distinct mechanisms. Thus, TNF uncouples receptor proximal events leading to impaired calcium responses (27). Independent of these effects, TNF also targets distal pathways that are required for efficient IL-2 gene transcription. To address the relationship between loss of TCRζ expression and uncoupling of distal signaling pathways, we examined the kinetics and dose-response of TNF-induced T cell hyporesponsiveness and down-regulation of TCRζ. Consistent with the suppression of PMA and ionomycin responses by TNF (Fig. 5C), substantial attenuation of T cell activation (>85%) was observed under culture conditions which induced minimal changes in TCRζ expression (see Fig. 11, which is published as supporting information on the PNAS web site). It follows from this that the contribution of TCRζ down-regulation to T cell hyporesponsiveness might only be accurately determined if the suppressive effects of TNF on distal signaling pathways can be reversed.

If down-modulation of TCRζ and the TCR/CD3 complex through antigen independent mechanisms is a relatively late event during T cell differentiation at sites of inflammation, loss of TCRζ expression could be used as a biomarker for terminally differentiated T cells that are profoundly hyporesponsive to antigenic stimulation in vivo. The abundant TCRζdim T cells infiltrating inflamed synovial joints would support this notion (14). Whether or not the presence of circulating TCRζdim cells in peripheral blood correlates with the activity or severity of inflammatory disease, or merely reflects a state of systemic immunosuppression, remains to be determined. Nevertheless, the availability of markers that document immunosuppression in vivo has clinical implications in the all-too-frequent scenario where patients with chronic inflammatory disease are treated with agents that suppress immune function still further. Although the precise significance of T cell hyporesponsiveness is not known, it may reflect a protective response that has evolved to limit reactivity of lymphocytes at sites of tissue damage. This hypothesis seems in keeping with the potent immunomodulatory and therapeutic effects of chronic TNF in animal models of inflammatory autoimmune diseases such as lupus and type I diabetes (42, 43), as well as the acceleration and increased severity of demyelinating disease observed in TNF-deficient mice (44, 45). We propose that defining the molecular basis of lymphocyte hyporesponsiveness has the potential for uncovering therapeutic approaches for immunosuppression. Our in vitro model of chronic TNF stimulation should facilitate such endeavors.

Supplementary Material

Supporting Figures:


This work was supported by the Wellcome Trust (J.M.C., A.E.A., M.P., P.I., and A.P.C.) and the Arthritis Research Campaign (United Kingdom).


Abbreviations: TCR, T cell receptor; TNF, tumor necrosis factor α; ZAP-70, ζ-associated protein of 70 kDa; LAT, linker for activation of T cells; eGFP, enhanced GFP; IRES, internal ribosome entry site; FACS, fluorescence-activated cell sorter.


1. Ashwell, J. D. & Klausner, R. D. (1990) Annu. Rev. Immunol. 8, 139-167. [PubMed]
2. Meuer, S. C., Cooper, D. A., Hodgdon, J. C., Hussey, R. E., Fitzgerald, K. A., Schlossman, S. F. & Reinherz, E. L. (1983) Science 222, 1239-1242. [PubMed]
3. Weissman, A. M., Samelson, L. E. & Klausner, R. D. (1986) Nature 324, 480-482. [PubMed]
4. Bonifacino, J. S., Chen, C., Lippincott-Schwartz, J., Ashwell, J. D. & Klausner, R. D. (1988) Proc. Natl. Acad. Sci. USA 85, 6929-6933. [PMC free article] [PubMed]
5. Weissman, A. M., Baniyash, M., Hou, D., Samelson, L. E., Burgess, W. H. & Klausner, R. D. (1988) Science 239, 1018-1021. [PubMed]
6. Reth, M. (1989) Nature 338, 383-384. [PubMed]
7. Irving, B. A., Chan, A. C. & Weiss, A. (1993) J. Exp. Med. 177, 1093-1103. [PMC free article] [PubMed]
8. Weiss, A. & Littman, D. R. (1994) Cell 76, 263-274. [PubMed]
9. Valitutti, S., Muller, S., Cella, M., Padovan, E. & Lanzavecchia, A. (1995) Nature 375, 148-151. [PubMed]
10. Ono, S., Ohno, H. & Saito, T. (1995) Immunity 2, 639-644. [PubMed]
11. Liu, H., Rhodes, M., Wiest, D. L. & Vignali, D. A. (2000) Immunity 13, 665-675. [PubMed]
12. Minami, Y., Weissman, A. M., Samelson, L. E. & Klausner, R. D. (1987) Proc. Natl. Acad. Sci. USA 84, 2688-2692. [PMC free article] [PubMed]
13. Sussman, J. J., Bonifacino, J. S., Lippincott-Schwartz, J., Weissman, A. M., Saito, T., Klausner, R. D. & Ashwell, J. D. (1988) Cell 52, 85-95. [PubMed]
14. Maurice, M. M., Lankester, A. C., Bezemer, A. C., Geertsma, M. F., Tak, P. P., Breedveld, F. C., van Lier, R. A. & Verweij, C. L. (1997) J. Immunol. 159, 2973-2978. [PubMed]
15. Liossis, S. N., Ding, X. Z., Dennis, G. J. & Tsokos, G. C. (1998) J. Clin. Invest. 101, 1448-1457. [PMC free article] [PubMed]
16. Stefanova, I., Saville, M. W., Peters, C., Cleghorn, F. R., Schwartz, D., Venzon, D. J., Weinhold, K. J., Jack, N., Bartholomew, C., Blattner, W. A., et al. (1996) J. Clin. Invest. 98, 1290-1297. [PMC free article] [PubMed]
17. Zea, A. H., Ochoa, M. T., Ghosh, P., Longo, D. L., Alvord, W. G., Valderrama, L., Falabella, R., Harvey, L. K., Saravia, N., Moreno, L. H. & Ochoa, A. C. (1998) Infect. Immun. 66, 499-504. [PMC free article] [PubMed]
18. Mizoguchi, H., O'Shea, J. J., Longo, D. L., Loeffler, C. M., McVicar, D. W. & Ochoa, A. C. (1992) Science 258, 1795-1798. [PubMed]
19. Tsuzaka, K., Takeuchi, T., Onoda, N., Pang, M. & Abe, T. (1998) J. Autoimmun. 11, 381-385. [PubMed]
20. Nambiar, M. P., Enyedy, E. J., Warke, V. G., Krishnan, S., Dennis, G., Wong, H. K., Kammer, G. M. & Tsokos, G. C. (2001) Arthritis Rheum. 44, 1336-1350. [PubMed]
21. Juang, Y. T., Tenbrock, K., Nambiar, M. P., Gourley, M. F. & Tsokos, G. C. (2002) J. Immunol. 169, 6048-6055. [PubMed]
22. Rodriguez, P. C., Zea, A. H., Culotta, K. S., Zabaleta, J., Ochoa, J. B. & Ochoa, A. C. (2002) J. Biol. Chem. 277, 21123-21129. [PubMed]
23. Gastman, B. R., Johnson, D. E., Whiteside, T. L. & Rabinowich, H. (1999) Cancer Res. 59, 1422-1427. [PubMed]
24. Wieckowski, E., Wang, G. Q., Gastman, B. R., Goldstein, L. A & Rabinowich, H. (2002) Cancer Res. 62, 4884-4889. [PubMed]
25. Wang, H. Y., Altman, Y., Fang, D., Elly, C., Dai, Y., Shao, Y. & Liu, Y. C. (2001) J. Biol. Chem. 276, 26004-26011. [PubMed]
26. Ardouin, L., Boyer, C., Gillet, A., Trucy, J., Bernard, A. M., Nunes, J., Delon, J., Trautmann, A., He, H. T., Malissen, B. & Malissen, M. (1999) Immunity 10, 409-420. [PubMed]
27. Isomäki, P., Panesar, M., Annenkov, A., Clark, J. M., Foxwell, B. M., Chernajovsky, Y. & Cope, A. P. (2001) J. Immunol. 166, 5495-5507. [PubMed]
28. Cope, A. P., Londei, M., Chu, N. R., Cohen, S. B., Elliott, M. J., Brennan, F. M., Maini, R. N. & Feldmann, M. (1994) J. Clin. Invest. 94, 749-760. [PMC free article] [PubMed]
29. Cope, A. P., Liblau, R. S., Yang, X. D., Congia, M., Laudanna, C., Schreiber, R. D., Probert, L., Kollias, G. & McDevitt, H. O. (1997) J. Exp. Med. 185, 1573-1584. [PMC free article] [PubMed]
30. Cope, A. P., Patel, S. D., Hall, F., Congia, M., Hubers, H. A., Verheijden, G. F., Boots, A. M., Menon, R., Trucco, M., Rijnders, A. W. & Sønderstrup, G. (1999) Arthritis Rheum. 42, 1497-1507. [PubMed]
31. Annenkov, A. & Chernajovsky, Y. (2000) Gene Ther. 7, 714-722. [PubMed]
32. Dranoff, G., Jaffee, E., Lazenby, A., Golumbek, P., Levitsky, H., Brose, K., Jackson, V., Hamada, H., Pardoll, D. & Mulligan, R. C. (1993) Proc. Natl. Acad. Sci. USA 90, 3539-3543. [PMC free article] [PubMed]
33. Annenkov, A. E., Daly, G. M. & Chernajovsky, Y. (2002) J. Gene Med. 4, 133-140. [PubMed]
34. Annenkov, A. E., Moyes, S. P., Eshhar, Z., Mageed, R. A. & Chernajovsky, Y. (1998) J. Immunol. 161, 6604-6613. [PubMed]
35. Wegener, A. M., Letourneur, F., Hoeveler, A., Brocker, T., Luton, F. & Malissen, B. (1992) Cell 68, 83-95. [PubMed]
36. Brocker, T., Peter, A., Traunecker, A. & Karjalainen, K. (1993) Eur. J. Immunol. 23, 1435-1439. [PubMed]
37. Cremesti, A. E., Goni, F. M. & Kolesnick, R. (2002) FEBS Lett. 531, 47-53. [PubMed]
38. Gringhuis, S. I., Leow, A., Papendrecht-Van Der Voort, E. A., Remans, P. H., Breedveld, F. C. & Verweij, C. L. (2000) J. Immunol. 164, 2170-2179. [PubMed]
39. Gringhuis, S. I., Papendrecht-van der Voort, E. A., Leow, A., Nivine Levarht, E. W., Breedveld, F. C. & Verweij, C. L. (2002) Mol. Cell. Biol. 22, 400-411. [PMC free article] [PubMed]
40. Bruyns, E., Marie-Cardine, A., Kirchgessner, H., Sagolla, K., Shevchenko, A., Mann, M., Autschbach, F., Bensussan, A., Meuer, S. & Schraven, B. (1998) J. Exp. Med. 188, 561-575. [PMC free article] [PubMed]
41. Kirchgessner, H., Dietrich, J., Scherer, J., Isomaki, P., Korinek, V., Hilgert, I., Bruyns, E., Leo, A., Cope, A. P. & Schraven, B. (2001) J. Exp. Med. 193, 1269-1284. [PMC free article] [PubMed]
42. Jacob, C. O. & McDevitt, H. O. (1988) Nature 331, 356-358. [PubMed]
43. Jacob, C. O., Aiso, S., Michie, S. A., McDevitt, H. O. & Acha-Orbea, H. (1990) Proc. Natl. Acad. Sci. USA 87, 968-972. [PMC free article] [PubMed]
44. Liu, J., Marino, M. W., Wong, G., Grail, D., Dunn, A., Bettadapura, J., Slavin, A. J., Old, L. & Bernard, C. C. (1998) Nat. Med. 4, 78-83. [PubMed]
45. Kassiotis, G. & Kollias, G. (2001) J. Exp. Med. 193, 427-434. [PMC free article] [PubMed]

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences
PubReader format: click here to try


Save items

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


  • Gene
    Gene records that cite the current articles. Citations in Gene are added manually by NCBI or imported from outside public resources.
  • GEO Profiles
    GEO Profiles
    Gene Expression Omnibus (GEO) Profiles of molecular abundance data. The current articles are references on the Gene record associated with the GEO profile.
  • HomoloGene
    HomoloGene clusters of homologous genes and sequences that cite the current articles. These are references on the Gene and sequence records in the HomoloGene entry.
  • MedGen
    Related information in MedGen
  • PubMed
    PubMed citations for these articles
  • Substance
    PubChem chemical substance records that cite the current articles. These references are taken from those provided on submitted PubChem chemical substance records.

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...