The CD3 Proline-Rich Sequence, and Its Interaction with Nck, Is Not Required for T Cell Development and Function1
http://www.100md.com
免疫学杂志 2005年第13期
Abstract
The CD3 proline-rich sequence (PRS) binds to the cytosolic adaptor molecule Nck after TCR ligation. It has been proposed that this interaction is essential for immunological synapse formation and T cell activation. To assess the physiological importance of the CD3 PRS, we have generated mice that lack this motif (CD3.PRSM). Pull-down experiments demonstrated the inability of Nck to bind to the CD3 PRS in thymocytes from mutant mice after TCR ligation. Surprisingly, no differences were observed in the number and percentage of T cell subsets in the thymus and spleen, and there was no apparent defect in positive or negative selection. Furthermore, the proliferative response of CD3.PRSM T cells to staphylococcal enterotoxin B and anti-CD3 Ab was normal. TCR surface expression, constitutive internalization, and Ag-induced down-modulation were also normal. These data suggest that the interaction between the CD3 PRS and Nck, or any other Src homology 3 domain-containing molecule, is not essential for T cell development and function.
Introduction
Central to the function of the immune system is the ability of the TCR:CD3 complex to recognize and bind to foreign peptides presented by MHC molecules and initiate an immune response. The TCR is a multimolecular complex composed of at least eight polypeptide subunits, heterodimers of TCR, CD3, and CD3, and a CD3 homodimer, all of which are necessary for efficient TCR:CD3 surface expression (1, 2, 3, 4). The TCR subunits constitute the ligand-binding portion, whereas the CD3 complex transduces the engagement signal into the T cell (5). TCR:CD3 surface expression is essential for T cell development and function (6, 7, 8, 9).
Ligation of the TCR leads to the recruitment of many signaling molecules to the T cell:APC interface, which has been termed the immunological synapse (IS)3 (10). In addition, changes in the actin cytoskeleton occur at this interface to ensure a stable contact site that is required for optimal T cell activation (11, 12, 13). TCR ligation induces the recruitment and hyperactivation of the tyrosine kinase ZAP70. This in turn leads to the phosphorylation and recruitment of many signaling and adaptor molecules, such as linker for activation of T cells and Nck (14, 15).
Nck is a 47-kDa cytosolic adaptor molecule expressed in a wide variety of cell types and tissues and acts as a link between extracellular and intracellular signaling molecules and the cytoskeleton (16, 17, 18). Nck is composed of three N-terminal Src homology 3 (SH3) domains, which bind to proline-rich sequences (PRS), and one C-terminal SH2 domain, which binds to phosphotyrosine residues. Consequently, Nck has been reported to interact with >45 different proteins, suggesting that it is used by a wide variety of intracellular signaling pathways (19). In T cells, this allows Nck to link the TCR:CD3 complex with proteins that mediate signaling and cytoskeletal rearrangement. For instance, Nck can interact with SH2 domain-containing leukocyte protein of 76 kDa through its SH2 domain and associate with p21-activating kinase (PAK), Vav1, and Wiskott-Aldrich syndrome protein (WASP) through its SH3 domains (20, 21, 22, 23, 24, 25). Such interactions can mediate multiple events. Nck, through its SH3 domains, interacts with PAK, and its activation is hypothesized to regulate the JNK and MAPK signaling cascade. WASP has been shown to be essential in T cell activation and cytoskeletal rearrangement, because T cells from mice deficient in WASP are defective in F-actin production, IL-2 secretion, and proliferation (26, 27). Another Nck binding partner, dynamin 2, has recently been described to be a critical mediator of these events (28).
Although significant advances have been made in our understanding of TCR signal transduction, there is still significant controversy over how this process is initiated by MHC:TCR interaction. Tyrosine phosphorylation of the CD3 chains has long been thought to be the initial trigger for T cell activation and signaling. However, recent work by Gil et al. (29) has shown that after TCR ligation, a conformational change in the TCR complex occurs that reveals a nascent PRS in the CD3 cytoplasmic domain, resulting in the recruitment and binding of Nck via its first SH3 domain (SH3.1). They suggested that the recruitment of Nck to the TCR:CD3 complex is essential for optimal T cell activation and IS formation. Interestingly, this interaction appeared to precede ITAM tyrosine phosphorylation. Although these studies appear to establish a new paradigm for the initiation of TCR signaling and Nck recruitment to the TCR engagement site, it remains to be established that this interaction is required by normal T cells for development and function in vivo. To assess this directly, we have generated mice with a mutation in the PRS of the CD3 protein that prevents Nck binding and examined its effect on T cell development and function.
Materials and Methods
Plasmids
Mutant and wild-type (WT) 2A peptide-linked TCR:CD3 chains were generated by recombinant PCR and cloned into the murine stem cell virus-based retroviral vector MSCV.IRES.GFP (MIG), as described previously (30). Details of primers and cloning strategy can be obtained from A. L. Szymczak.
Immunoprecipitation and Western blot analysis
The 3A9 T cell hybridoma (hen egg lysozyme 48–63 specific; H-2Ak restricted) was lysed in 1% digitonin (Wako), and the lysate was precleared with heat-killed, formalin-fixed Staphylococcus aureus (Pansorbin) cells (Calbiochem) for 1 h. The lysate was immunoprecipitated with Sepharose beads coupled with peptides corresponding to the CD3 PRS (WT or mutated) for 1 h (CD3 WT, GSRPRGQNKERPPPVPNPDY; CD3 PRSM, GSRPRGQNKERAAAVANADY; generated by the Hartwell Center for Biotechnology and Bioinformatics, St. Jude Children’s Research Hospital). Nck did not interact with a control peptide representing a sequence of CD3 that did not contain the PRS (YEPIRKGQRDLYSGLNQRAV; data not shown). Eluted proteins were resolved by 10% SDS-PAGE, transferred onto polyvinylidene difluoride membrane, and probed with anti-Nck Ab (BD Transduction Laboratories). The Ab was detected using protein A/G-alkaline phosphatase and developed using Vistra substrate (Amersham Biosciences).
Transient transfection of 293T cells and the TCR internalization assay
Transfections and internalization assays were performed as described previously with some modifications (31). 293T cells (human embryonic kidney cells; provided by D. Baltimore, California Institute of Technology, Pasadena, CA, and E. Vanin, Baylor College of Medicine, Houston, TX) were incubated in 10-cm plates at 2 x 106/plate overnight at 37°C. On day 1, 500 μl of DMEM was mixed with 30 μl of Fugene6 transfection reagent (Roche). Each group of transfectants received 3A9.TCR-2A, CD3, and the CD3-2A construct indicated (WT or PRSM). DNA (2 μg of each plasmid) was mixed with the Fugene/DMEM, incubated at room temperature for 15 min, and added to the cells. On day 3, transfectants were analyzed by flow cytometry for TCR expression using anti-TCR-.PE Ab (H57-597). All Abs used were obtained from BD Pharmingen, unless stated otherwise.
For TCR internalization, 293T cells were transfected as described above. On day 3, cells were stained with anti-V8.PE (F23.1) and anti-CD3. Allophycocyanin and GFP+TCR+CD3+ cells were sorted and cultured in triplicate for each time point in a 96-well, flat-bottom plate at 5 x 104/well overnight. Brefeldin A (Epicentre Technologies) was added at 10 μg/ml for 30 min, and TCR modulation was determined by staining with anti-TCR-.CyChrome. The percentage of modulation was calculated from the median fluorescence values using untreated controls as reference. Whole splenocytes were cultured at 2.5 x 105/well. After brefeldin A treatment, cells were stained with Abs to Thy1.1 and TCR-. The percentage of modulation was calculated from the median of TCR on live (propidium iodide–)Thy1.1+GFP+ cells using untreated controls as reference.
Mice
RAG-1–/– mice were obtained from The Jackson Laboratory. CD3P/P mice were provided by C. Terhorst (Harvard Medical School, Boston, MA) (9). All animal experiments were performed in an American Association for the Accreditation of Laboratory Animal Care-accredited, specific pathogen-free facility following national, state, and institutional guidelines. Animal protocols were approved by the St. Jude Institutional Animal Care and Use Committee.
Retroviral-mediated stem cell gene transfer
Retroviral producer cell lines were generated as described previously with some modifications (32, 33, 34). 293T cells were transiently transfected with the construct of interest and retroviral packaging plasmids using Fugene6 transfection reagent (Roche). Retroviral producer cell lines were made by repeatedly transducing GP+E86 cells (six to eight times) until a viral titer of >105/ml after 24 h was obtained. Retroviral transduction of bone marrow was performed as described previously (33). Bone marrow was harvested from 10- to 15-wk-old donor mice 48 h after treatment with 150 mg/kg 5-fluoruracil (Pharmacia). Bone marrow cells were cultured in media, and stem cells were induced to proliferate with 20 ng/ml murine IL-3, 50 ng/ml human IL-6, and 50 ng/ml murine stem cell factor (Biosource). Bone marrow cells were cocultured with the retroviral producer cell lines. After 48 h, the nonadherent, transduced bone marrow cells were collected and washed. Sublethally irradiated (450 rad) recipient RAG-1–/– mice were given injections via the tail vein with 4–6 x 106 bone marrow cells in PBS with 2% FBS and 20 U/ml heparin. Mice were analyzed 4–14 wk after transplant. Splenocytes and thymocytes were processed and stained with the following Abs: anti-TCR-.PE, anti-CD45/B220-allophycocyanin (RA3-6B2), anti-CD4-allophycocyanin or -CyChrome (RM4-4), anti-CD8.PE or allophycocyanin (53-6.7), anti-CD25.PE (PC61), anti-CD44.CyChrome (IM7), anti-CD69.PerCP-Cy5.5, anti-V7.PE (TR310), anti-V8.PE (F23.1), and biotinylated V5.1/5.2 (MR9-4), V11 (RR3-15), or V12 (MR11-1), followed by streptavidin-PE.
Nck pull-down assay
Nck pull-down assays were performed as described previously (29). Thymocytes (2.0 x 107/ml) were stimulated with 10 μg of anti-CD3 (2C11; BD Pharmingen) for 5 min at 37°C in RPMI 1640 plus 10% FBS. Cells were pelleted and lysed 15 min on ice in 1 ml of ice-cold lysis buffer (20 mM Tris-HCl, 150 mM NaCl, 0.3% Brij 58, 1 mM PMSF, and 1 mM iodoacetamide with 1 μg/ml aprotinin and leupeptin). The lysate was centrifuged for 15 min at 4°C and incubated with GST beads for 1 h at 4°C as a preclear step, followed by a 4-h incubation with GST-Nck SH3.1 beads. The beads were washed three times in lysis buffer, boiled for 5 min at 95°C in sample buffer with SDS, and subjected to SDS-PAGE and Western blot analysis. An aliquot of lysate from each sample was incubated with an equal amount of acetone for 15 min at 4°C, pelleted, and resuspended in sample buffer with SDS as a control, demonstrating equal expression of CD3 among samples. Proteins were resolved by 13% SDS-PAGE, transferred onto polyvinylidene difluoride membrane, and probed with rabbit anti-CD3 antiserum 448 (35).
Activation and proliferation assays
Thymocytes (1 x 106) were stimulated 16 h with plate-bound anti-CD3 (2C11). Cells were stained with the anti-CD4-PerCP-Cy5.5 (RM4-4), anti-CD8-allophycocyanin (53-6.7), and anti-CD69.PerCP-Cy5.5 or propidium iodide, and analyzed by flow cytometry.
For stimulation with staphylococcal enterotoxin B (SEB; Toxin Technology), T cells from the spleen of bone marrow recipients were negatively sorted by magnetic bead cell sorting. Briefly, splenocytes were incubated with biotin-coupled Abs to CD45/B220, CD49b (Pan NK; DX5), Mac-1 (CD11b; M1/70), Gr-1 (Ly-6G; RB6-8C5), and TER-119 (Ly-76). Cells were incubated with magnetic beads coupled with anti-streptavidin and negatively sorted on an autoMACS (Miltenyi Biotec) to 90–95% purity. T cells recovered were plated at 105/well and stimulated with varying concentrations of SEB with 5 x 105 irradiated (3000 rad) C57BL/6 splenocytes for 48 h. Cells were then pulsed with [3H]thymidine (1 μCi/well; PerkinElmer) for 24 h, and proliferation was measured.
For stimulation with anti-CD3, splenocytes were incubated with varying concentrations of plate-bound anti-CD3 (2C11) for 24 h. Cells were pulsed with [3H]thymidine for 24 h, and proliferation was measured.
Down-modulation of TCR complexes
Splenocytes were plated in triplicate in a 96-well, round-bottom plate (106/well) and incubated at 37°C with 20 μg/ml SEB (Toxin Technology). After a 12-h stimulation, cells were stained with Abs against CD4, CD8, V7, V8, and CD69, and TCR down-modulation was analyzed by flow cytometry. Down-modulation was calculated from median V7/V8 TCR fluorescence using unstimulated cells as a reference.
Results
To confirm that the CD3 PRS could bind to Nck, we performed pull-down assays with 3A9 T cell hybridoma lysates using Sepharose beads conjugated with peptides corresponding to the region of the CD3 cytoplasmic tail that contains the WT PRS (CD3.WT, GSRPRGQNKERPPPVPNPDY (PRS in bold)) or an alanine mutant version (CD3.PRSM, GSRPRGQNKERAAAVANADY). Western blot analysis using an anti-Nck Ab clearly verified that the mutated PRS sequence (PRSM) is unable to bind to Nck (Fig. 1A).
To investigate the role of Nck recruitment to the CD3 PRS in vivo, we chose to generate mice by retroviral-mediated stem cell gene transfer. For this, we used CD3P/P mice, which lack CD3 and have a severe reduction in CD3 and CD3 gene expression (9). Multicistronic retroviral vectors were constructed that carried WT or PRS-mutated CD3 linked to WT CD3 and CD3 by self-cleaving 2A peptides (30) (CD3-2A.WT and CD3-2A.PRSM, respectively) (Fig. 1B). These were expressed in a murine stem cell virus-based retroviral vector containing an internal ribosomal entry site element and GFP (MIG). We have shown previously that T cell development and function is fully restored in CD3P/P mice using retroviral-mediated stem cell gene transfer and 2A peptide-linked CD3-2A constructs, giving rise to 100% cleavage and stoichiometric expression of each CD3 chain (30).
To confirm that the PRSM mutation did not affect surface expression of the TCR, 293T cells were transiently transfected with CD3.WT or CD3.PRSM and the remaining TCR:CD3 chains. The data clearly show that TCR expression in the CD3.PRSM transfectant was comparable with WT (Fig. 1C). TCR internalization was not affected, suggesting that the PRS mutation had not perturbed the transport of TCR:CD3 complexes (data not shown).
To determine the effect of the PRS mutation on T cell development and function in vivo, retroviral producers were made for both CD3-2A.WT and CD3-2A.PRSM MIG constructs and were used to transduce bone marrow from CD3P/P mice. T cell development and function were analyzed after transfer of transduced bone marrow into RAG–/– mice. Surprisingly, T cell development in these mice was normal despite the importance attached to CD3:Nck interaction (29). To verify that Nck:CD3 interaction was abrogated in T cells from mice expressing CD3.PRSM, thymocytes were stimulated with anti-CD3 and subjected to pull-down assays using GST beads coupled to the SH3.1 domain of Nck (Fig. 1D) (29, 36). Western blot analysis using an anti-CD3 Ab showed that TCR complexes were able to interact with Nck in vitro after stimulation with anti-CD3 in cells from mice expressing the WT CD3. In cells expressing the PRS mutant CD3, however, TCR complexes were not pulled down with the Nck-SH3.1 domain, indicating the interaction between CD3 and Nck depends on the CD3.PRS.
The phenotype and functionality of the thymocytes and splenocytes were subsequently analyzed. The percentage of CD4+ and CD8+ thymocytes in mice expressing CD3.PRSM appeared comparable with that of CD3.WT mice (Fig. 2A). No differences were observed in the phenotype of the mice at either early (4–8 wk after transfer) or late (12–14 wk after transfer) time points. There did appear to be a very slight increase in the percentage of cells in the early double-negative (DN) stages of thymocyte development in the PRSM mice compared with WT (Fig. 2B; DN1–3; CD25–44+, CD25+44+, and CD25+44–). However, the resultant reduction of DN4 cells did not appear to affect the number of double-positive or mature single-positive thymocytes, because these were comparable between both groups (Fig. 2C). In addition, stimulation of thymocytes with anti-CD3 induced comparable levels of CD69 expression in CD4+CD8+ cells from both groups of mice (Fig. 2D and data not shown). Finally, there appeared to be no obvious defects in negative selection between the two groups of mice, as measured by the deletion of T cells bearing mammary tumor virus-specific TCRs (Fig. 2E) (37).
No differences were seen in the percentage or number of T or B cells in the spleens of the two groups of mice (Fig. 3, A and B). In addition, both groups had equivalent numbers of CD4+ and CD8+ peripheral T cells (Fig. 3, A and B) and comparable levels of TCR expression (C). Functional analysis was also performed on splenocytes from mice expressing either WT or PRSM CD3. No differences were observed in the ability of splenic T cells from these mice to constitutively internalize or down-modulate their TCR after SEB stimulation (Fig. 3D and data not shown). Likewise, CD69 expression after SEB stimulation on T cells from both groups of mice was comparable (Fig. 3E). Finally, T cells from both groups of mice proliferated effectively after stimulation with SEB and anti-CD3 (Fig. 3F). Cytokine analysis was performed on T cells stimulated with both anti-CD3 and SEB. Although no discernible differences were observed, additional studies must be performed to address this in more detail (data not shown). These data suggest that the interaction between Nck and the CD3 PRS is not required for T cell development, the establishment of peripheral T cells, and T cell function.
Discussion
These data suggest that the interaction between the CD3 PRS and Nck, or presumably any other SH3 domain-containing molecule that could bind to this sequence, is not required for T cell development and function. These results were surprising given the potential significance attached to a recent study indicating a role for CD3:Nck interaction in T cell activation and function in vitro (29, 38). In this study, the authors inhibited CD3:Nck interaction by overexpressing the SH3.1 domain of Nck in Jurkat T cells. This inhibited formation of T cell:APC conjugates as well as IS formation and greatly reduced activation of the cells as measured by IL-2 production and proliferation. What may account for the differences seen here?
One possibility is that this discrepancy may be due to differences in the systems used. Perhaps in mice this interaction is not as critical as in human cells. An alternative possibility, which we favor, is that overexpression of Nck SH3.1 not only interferes with Nck:CD3 interaction but also abrogates interaction with multiple proteins that mediate IS formation, such as cytoskeletal regulatory molecules, and T cell activation. Thus, the observations made by these authors demonstrate an important role for Nck rather than Nck:CD3 interaction (29). This has been suggested by others as well (39, 40). Recently, it has been demonstrated, using molecular imaging techniques and fluorescence resonance energy transfer analysis, that recruitment of Nck to the site of TCR engagement and subsequent actin polymerization requires the adaptor proteins linker for activation of T cells and SH2 domain-containing leukocyte protein of 76 kDa and that TCR-induced tyrosine phosphorylation of these proteins must first take place (38).
Nck interacts with CD3 through the first N-terminal SH3.1 domain, whereas interaction with PAK and WASP, for instance, occurs via its second and third SH3 domains (SH3.2 and SH3.3), respectively (29, 41). It could therefore be argued that overexpression of SH3.1 would not interfere with Nck binding to these proteins, which are key regulators of cytoskeletal rearrangement and thus IS formation. However, it has been shown that although actual binding occurs through a single SH3 domain, multiple SH3 domains of Nck are required for productive interaction to occur, suggesting cooperative use of multiple SH3 domains for tight complex formation. This has been observed with Nck binding to Cbl, Bcr/Abl, and p155 as well as WASP and PAK (42, 43). Using various Nck GST fusion proteins, only combined SH3 domains of Nck or full-length Nck facilitated binding. In addition, using an Ab-based system to alter concentrations of Nck at the plasma membrane, all three SH3 domains were required for actin polymerization (44). Our system alleviates these potential problems by specifically abolishing the interaction of CD3 and Nck through mutation of CD3 rather than alteration of Nck.
What is the purpose of Nck:CD3 interaction in T cells? Some have suggested that this interaction forms a bridge between the TCR and misshapen NIK-related kinase (MINK), a kinase that interacts with Nck, that has recently been described to play a role in negative selection in the thymus (37, 45). Cells in which MINK expression was knocked down had a defect in deletion of thymocytes reactive to the endogenous superantigen Mtv as well as T cells expressing self-reactive TCR transgenes. If this hypothesis is correct, mice unable to recruit Nck to CD3 might display a defect in negative selection. However, no apparent defects in selection were observed in the CD3.PRSM mice generated in this study. Given our findings, the possible link between MINK and the TCR complex should be re-evaluated.
Our data indicate that the interaction between Nck and the CD3 PRS is not required for T cell development or function in mice. However, it is important to note that these data do not necessarily contradict the suggestion by Gil et al. (29, 36) that the CD3 complex undergoes a conformational change after TCR engagement. In these studies, binding of the TCR led to the exposure of the CD3 PRS that was detected in vitro using the Nck SH3.1 domain. Thus, a conformational change may still occur after ligation of TCR complexes containing the CD3 PRS mutation, although this would not be detected using this method. Exposure of the CD3 PRS is likely to be only one of several changes that occur in the CD3 complex as a result of TCR ligation. Indeed, we have shown that ligation of the TCR leads to the exposure of the CD3 N terminus, which ordinarily is buried within the complex, and dissociation of CD3 from the TCR complex (46).
We cannot completely rule out the possibility that this interaction does have some function that was not revealed in our assays, although it clearly does not have a major, significant role in T cell development and function. It is possible that it functions redundantly with another interaction involving the TCR:CD3 complex. We considered the possibility, for instance, that Nck might also be recruited to the PRS in pT. However, pull-down experiments showed that Nck binds to this sequence very weakly, if at all (S. Dilioglou, A. L. Szymczak, and D. A. A. Vignali, unpublished data). Thus, although Nck does interact with CD3, our study clearly shows that T cell development and function is not altered in mice expressing the CD3 PRS mutation, inferring a less significant role for this interaction than was originally suggested.
Acknowledgments
We are very grateful to Cox Terhorst and David Baltimore for reagents. We also thank Richard Cross and Jennifer Hoffrage for assistance with flow cytometry, the staff in the Hartwell Center for oligonucleotide and peptide synthesis and DNA sequencing, and members of the Vignali laboratory for constructive criticisms, comments, and bone marrow isolation.
Disclosures
The authors have no financial conflict of interest.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This work was supported by National Institutes of Health Grant AI-52199, Cancer Center Support Center of Research Excellence Grant CA-21765, and by a grant from the American Lebanese Syrian Associated Charities.
2 Address correspondence and reprint requests to Dr. Dario A. A. Vignali, Department of Immunology, St. Jude Children’s Research Hospital, 332 North Lauderdale, Memphis, TN 38105-2794. E-mail address: dario.vignali@stjude.org
3 Abbreviations used in this paper: IS, immunological synapse; SH3, Src homology 3; PRS, proline-rich sequence; PRSM, PRS alanine mutant; PAK, p21-activating kinase; WASP, Wiskott-Aldrich syndrome protein; WT, wild type; SEB, staphylococcal enterotoxin B; DN, double negative; MINK, misshapen NIK-related kinase.
Received for publication August 4, 2004. Accepted for publication April 19, 2005.
References
Call, M. E., J. Pyrdol, M. Wiedmann, K. W. Wucherpfennig. 2002. The organizing principle in the formation of the T cell receptor-CD3 complex. Cell 111: 967-979.
Bonifacino, J. S., C. K. Suzuki, J. Lippincott-Schwartz, A. M. Weissman, R. D. Klausner. 1989. Pre-Golgi degradation of newly synthesized T-cell antigen receptor chains: intrinsic sensitivity and the role of subunit assembly. J. Cell Biol. 109: 73-83.
Geisler, C.. 1992. Failure to synthesize the CD3- chain: consequences for T cell antigen receptor assembly, processing, and expression. J. Immunol. 148: 2437-2445.
Sussman, J. J., J. S. Bonifacino, J. Lippincott-Schwartz, A. M. Weissman, T. Saito, R. D. Klausner, J. D. Ashwell. 1988. Failure to synthesize the T cell CD3- chain: structure and function of a partial T cell receptor complex. Cell 52: 85-95.
Clevers, H., B. Alarcon, T. Wileman, C. Terhorst. 1988. The T cell receptor/CD3 complex: a dynamic protein ensemble. Annu. Rev. Immunol. 6: 629-662.
Love, P. E., E. W. Shores, M. D. Johnson, M. L. Tremblay, E. J. Lee, A. Grinberg, S. P. Huang, A. Singer, H. Westphal. 1993. T cell development in mice that lack the chain of the T cell antigen receptor complex. Science 261: 918-921.
Dave, V. P., Z. Cao, C. Browne, B. Alarcon, G. Fernandez-Miguel, J. Lafaille, A. de la Hera, S. Tonegawa, D. J. Kappes. 1997. CD3 deficiency arrests development of the but not the T cell lineage. EMBO J. 16: 1360-1370.(Andrea L. Szymczak*, Creg)
The CD3 proline-rich sequence (PRS) binds to the cytosolic adaptor molecule Nck after TCR ligation. It has been proposed that this interaction is essential for immunological synapse formation and T cell activation. To assess the physiological importance of the CD3 PRS, we have generated mice that lack this motif (CD3.PRSM). Pull-down experiments demonstrated the inability of Nck to bind to the CD3 PRS in thymocytes from mutant mice after TCR ligation. Surprisingly, no differences were observed in the number and percentage of T cell subsets in the thymus and spleen, and there was no apparent defect in positive or negative selection. Furthermore, the proliferative response of CD3.PRSM T cells to staphylococcal enterotoxin B and anti-CD3 Ab was normal. TCR surface expression, constitutive internalization, and Ag-induced down-modulation were also normal. These data suggest that the interaction between the CD3 PRS and Nck, or any other Src homology 3 domain-containing molecule, is not essential for T cell development and function.
Introduction
Central to the function of the immune system is the ability of the TCR:CD3 complex to recognize and bind to foreign peptides presented by MHC molecules and initiate an immune response. The TCR is a multimolecular complex composed of at least eight polypeptide subunits, heterodimers of TCR, CD3, and CD3, and a CD3 homodimer, all of which are necessary for efficient TCR:CD3 surface expression (1, 2, 3, 4). The TCR subunits constitute the ligand-binding portion, whereas the CD3 complex transduces the engagement signal into the T cell (5). TCR:CD3 surface expression is essential for T cell development and function (6, 7, 8, 9).
Ligation of the TCR leads to the recruitment of many signaling molecules to the T cell:APC interface, which has been termed the immunological synapse (IS)3 (10). In addition, changes in the actin cytoskeleton occur at this interface to ensure a stable contact site that is required for optimal T cell activation (11, 12, 13). TCR ligation induces the recruitment and hyperactivation of the tyrosine kinase ZAP70. This in turn leads to the phosphorylation and recruitment of many signaling and adaptor molecules, such as linker for activation of T cells and Nck (14, 15).
Nck is a 47-kDa cytosolic adaptor molecule expressed in a wide variety of cell types and tissues and acts as a link between extracellular and intracellular signaling molecules and the cytoskeleton (16, 17, 18). Nck is composed of three N-terminal Src homology 3 (SH3) domains, which bind to proline-rich sequences (PRS), and one C-terminal SH2 domain, which binds to phosphotyrosine residues. Consequently, Nck has been reported to interact with >45 different proteins, suggesting that it is used by a wide variety of intracellular signaling pathways (19). In T cells, this allows Nck to link the TCR:CD3 complex with proteins that mediate signaling and cytoskeletal rearrangement. For instance, Nck can interact with SH2 domain-containing leukocyte protein of 76 kDa through its SH2 domain and associate with p21-activating kinase (PAK), Vav1, and Wiskott-Aldrich syndrome protein (WASP) through its SH3 domains (20, 21, 22, 23, 24, 25). Such interactions can mediate multiple events. Nck, through its SH3 domains, interacts with PAK, and its activation is hypothesized to regulate the JNK and MAPK signaling cascade. WASP has been shown to be essential in T cell activation and cytoskeletal rearrangement, because T cells from mice deficient in WASP are defective in F-actin production, IL-2 secretion, and proliferation (26, 27). Another Nck binding partner, dynamin 2, has recently been described to be a critical mediator of these events (28).
Although significant advances have been made in our understanding of TCR signal transduction, there is still significant controversy over how this process is initiated by MHC:TCR interaction. Tyrosine phosphorylation of the CD3 chains has long been thought to be the initial trigger for T cell activation and signaling. However, recent work by Gil et al. (29) has shown that after TCR ligation, a conformational change in the TCR complex occurs that reveals a nascent PRS in the CD3 cytoplasmic domain, resulting in the recruitment and binding of Nck via its first SH3 domain (SH3.1). They suggested that the recruitment of Nck to the TCR:CD3 complex is essential for optimal T cell activation and IS formation. Interestingly, this interaction appeared to precede ITAM tyrosine phosphorylation. Although these studies appear to establish a new paradigm for the initiation of TCR signaling and Nck recruitment to the TCR engagement site, it remains to be established that this interaction is required by normal T cells for development and function in vivo. To assess this directly, we have generated mice with a mutation in the PRS of the CD3 protein that prevents Nck binding and examined its effect on T cell development and function.
Materials and Methods
Plasmids
Mutant and wild-type (WT) 2A peptide-linked TCR:CD3 chains were generated by recombinant PCR and cloned into the murine stem cell virus-based retroviral vector MSCV.IRES.GFP (MIG), as described previously (30). Details of primers and cloning strategy can be obtained from A. L. Szymczak.
Immunoprecipitation and Western blot analysis
The 3A9 T cell hybridoma (hen egg lysozyme 48–63 specific; H-2Ak restricted) was lysed in 1% digitonin (Wako), and the lysate was precleared with heat-killed, formalin-fixed Staphylococcus aureus (Pansorbin) cells (Calbiochem) for 1 h. The lysate was immunoprecipitated with Sepharose beads coupled with peptides corresponding to the CD3 PRS (WT or mutated) for 1 h (CD3 WT, GSRPRGQNKERPPPVPNPDY; CD3 PRSM, GSRPRGQNKERAAAVANADY; generated by the Hartwell Center for Biotechnology and Bioinformatics, St. Jude Children’s Research Hospital). Nck did not interact with a control peptide representing a sequence of CD3 that did not contain the PRS (YEPIRKGQRDLYSGLNQRAV; data not shown). Eluted proteins were resolved by 10% SDS-PAGE, transferred onto polyvinylidene difluoride membrane, and probed with anti-Nck Ab (BD Transduction Laboratories). The Ab was detected using protein A/G-alkaline phosphatase and developed using Vistra substrate (Amersham Biosciences).
Transient transfection of 293T cells and the TCR internalization assay
Transfections and internalization assays were performed as described previously with some modifications (31). 293T cells (human embryonic kidney cells; provided by D. Baltimore, California Institute of Technology, Pasadena, CA, and E. Vanin, Baylor College of Medicine, Houston, TX) were incubated in 10-cm plates at 2 x 106/plate overnight at 37°C. On day 1, 500 μl of DMEM was mixed with 30 μl of Fugene6 transfection reagent (Roche). Each group of transfectants received 3A9.TCR-2A, CD3, and the CD3-2A construct indicated (WT or PRSM). DNA (2 μg of each plasmid) was mixed with the Fugene/DMEM, incubated at room temperature for 15 min, and added to the cells. On day 3, transfectants were analyzed by flow cytometry for TCR expression using anti-TCR-.PE Ab (H57-597). All Abs used were obtained from BD Pharmingen, unless stated otherwise.
For TCR internalization, 293T cells were transfected as described above. On day 3, cells were stained with anti-V8.PE (F23.1) and anti-CD3. Allophycocyanin and GFP+TCR+CD3+ cells were sorted and cultured in triplicate for each time point in a 96-well, flat-bottom plate at 5 x 104/well overnight. Brefeldin A (Epicentre Technologies) was added at 10 μg/ml for 30 min, and TCR modulation was determined by staining with anti-TCR-.CyChrome. The percentage of modulation was calculated from the median fluorescence values using untreated controls as reference. Whole splenocytes were cultured at 2.5 x 105/well. After brefeldin A treatment, cells were stained with Abs to Thy1.1 and TCR-. The percentage of modulation was calculated from the median of TCR on live (propidium iodide–)Thy1.1+GFP+ cells using untreated controls as reference.
Mice
RAG-1–/– mice were obtained from The Jackson Laboratory. CD3P/P mice were provided by C. Terhorst (Harvard Medical School, Boston, MA) (9). All animal experiments were performed in an American Association for the Accreditation of Laboratory Animal Care-accredited, specific pathogen-free facility following national, state, and institutional guidelines. Animal protocols were approved by the St. Jude Institutional Animal Care and Use Committee.
Retroviral-mediated stem cell gene transfer
Retroviral producer cell lines were generated as described previously with some modifications (32, 33, 34). 293T cells were transiently transfected with the construct of interest and retroviral packaging plasmids using Fugene6 transfection reagent (Roche). Retroviral producer cell lines were made by repeatedly transducing GP+E86 cells (six to eight times) until a viral titer of >105/ml after 24 h was obtained. Retroviral transduction of bone marrow was performed as described previously (33). Bone marrow was harvested from 10- to 15-wk-old donor mice 48 h after treatment with 150 mg/kg 5-fluoruracil (Pharmacia). Bone marrow cells were cultured in media, and stem cells were induced to proliferate with 20 ng/ml murine IL-3, 50 ng/ml human IL-6, and 50 ng/ml murine stem cell factor (Biosource). Bone marrow cells were cocultured with the retroviral producer cell lines. After 48 h, the nonadherent, transduced bone marrow cells were collected and washed. Sublethally irradiated (450 rad) recipient RAG-1–/– mice were given injections via the tail vein with 4–6 x 106 bone marrow cells in PBS with 2% FBS and 20 U/ml heparin. Mice were analyzed 4–14 wk after transplant. Splenocytes and thymocytes were processed and stained with the following Abs: anti-TCR-.PE, anti-CD45/B220-allophycocyanin (RA3-6B2), anti-CD4-allophycocyanin or -CyChrome (RM4-4), anti-CD8.PE or allophycocyanin (53-6.7), anti-CD25.PE (PC61), anti-CD44.CyChrome (IM7), anti-CD69.PerCP-Cy5.5, anti-V7.PE (TR310), anti-V8.PE (F23.1), and biotinylated V5.1/5.2 (MR9-4), V11 (RR3-15), or V12 (MR11-1), followed by streptavidin-PE.
Nck pull-down assay
Nck pull-down assays were performed as described previously (29). Thymocytes (2.0 x 107/ml) were stimulated with 10 μg of anti-CD3 (2C11; BD Pharmingen) for 5 min at 37°C in RPMI 1640 plus 10% FBS. Cells were pelleted and lysed 15 min on ice in 1 ml of ice-cold lysis buffer (20 mM Tris-HCl, 150 mM NaCl, 0.3% Brij 58, 1 mM PMSF, and 1 mM iodoacetamide with 1 μg/ml aprotinin and leupeptin). The lysate was centrifuged for 15 min at 4°C and incubated with GST beads for 1 h at 4°C as a preclear step, followed by a 4-h incubation with GST-Nck SH3.1 beads. The beads were washed three times in lysis buffer, boiled for 5 min at 95°C in sample buffer with SDS, and subjected to SDS-PAGE and Western blot analysis. An aliquot of lysate from each sample was incubated with an equal amount of acetone for 15 min at 4°C, pelleted, and resuspended in sample buffer with SDS as a control, demonstrating equal expression of CD3 among samples. Proteins were resolved by 13% SDS-PAGE, transferred onto polyvinylidene difluoride membrane, and probed with rabbit anti-CD3 antiserum 448 (35).
Activation and proliferation assays
Thymocytes (1 x 106) were stimulated 16 h with plate-bound anti-CD3 (2C11). Cells were stained with the anti-CD4-PerCP-Cy5.5 (RM4-4), anti-CD8-allophycocyanin (53-6.7), and anti-CD69.PerCP-Cy5.5 or propidium iodide, and analyzed by flow cytometry.
For stimulation with staphylococcal enterotoxin B (SEB; Toxin Technology), T cells from the spleen of bone marrow recipients were negatively sorted by magnetic bead cell sorting. Briefly, splenocytes were incubated with biotin-coupled Abs to CD45/B220, CD49b (Pan NK; DX5), Mac-1 (CD11b; M1/70), Gr-1 (Ly-6G; RB6-8C5), and TER-119 (Ly-76). Cells were incubated with magnetic beads coupled with anti-streptavidin and negatively sorted on an autoMACS (Miltenyi Biotec) to 90–95% purity. T cells recovered were plated at 105/well and stimulated with varying concentrations of SEB with 5 x 105 irradiated (3000 rad) C57BL/6 splenocytes for 48 h. Cells were then pulsed with [3H]thymidine (1 μCi/well; PerkinElmer) for 24 h, and proliferation was measured.
For stimulation with anti-CD3, splenocytes were incubated with varying concentrations of plate-bound anti-CD3 (2C11) for 24 h. Cells were pulsed with [3H]thymidine for 24 h, and proliferation was measured.
Down-modulation of TCR complexes
Splenocytes were plated in triplicate in a 96-well, round-bottom plate (106/well) and incubated at 37°C with 20 μg/ml SEB (Toxin Technology). After a 12-h stimulation, cells were stained with Abs against CD4, CD8, V7, V8, and CD69, and TCR down-modulation was analyzed by flow cytometry. Down-modulation was calculated from median V7/V8 TCR fluorescence using unstimulated cells as a reference.
Results
To confirm that the CD3 PRS could bind to Nck, we performed pull-down assays with 3A9 T cell hybridoma lysates using Sepharose beads conjugated with peptides corresponding to the region of the CD3 cytoplasmic tail that contains the WT PRS (CD3.WT, GSRPRGQNKERPPPVPNPDY (PRS in bold)) or an alanine mutant version (CD3.PRSM, GSRPRGQNKERAAAVANADY). Western blot analysis using an anti-Nck Ab clearly verified that the mutated PRS sequence (PRSM) is unable to bind to Nck (Fig. 1A).
To investigate the role of Nck recruitment to the CD3 PRS in vivo, we chose to generate mice by retroviral-mediated stem cell gene transfer. For this, we used CD3P/P mice, which lack CD3 and have a severe reduction in CD3 and CD3 gene expression (9). Multicistronic retroviral vectors were constructed that carried WT or PRS-mutated CD3 linked to WT CD3 and CD3 by self-cleaving 2A peptides (30) (CD3-2A.WT and CD3-2A.PRSM, respectively) (Fig. 1B). These were expressed in a murine stem cell virus-based retroviral vector containing an internal ribosomal entry site element and GFP (MIG). We have shown previously that T cell development and function is fully restored in CD3P/P mice using retroviral-mediated stem cell gene transfer and 2A peptide-linked CD3-2A constructs, giving rise to 100% cleavage and stoichiometric expression of each CD3 chain (30).
To confirm that the PRSM mutation did not affect surface expression of the TCR, 293T cells were transiently transfected with CD3.WT or CD3.PRSM and the remaining TCR:CD3 chains. The data clearly show that TCR expression in the CD3.PRSM transfectant was comparable with WT (Fig. 1C). TCR internalization was not affected, suggesting that the PRS mutation had not perturbed the transport of TCR:CD3 complexes (data not shown).
To determine the effect of the PRS mutation on T cell development and function in vivo, retroviral producers were made for both CD3-2A.WT and CD3-2A.PRSM MIG constructs and were used to transduce bone marrow from CD3P/P mice. T cell development and function were analyzed after transfer of transduced bone marrow into RAG–/– mice. Surprisingly, T cell development in these mice was normal despite the importance attached to CD3:Nck interaction (29). To verify that Nck:CD3 interaction was abrogated in T cells from mice expressing CD3.PRSM, thymocytes were stimulated with anti-CD3 and subjected to pull-down assays using GST beads coupled to the SH3.1 domain of Nck (Fig. 1D) (29, 36). Western blot analysis using an anti-CD3 Ab showed that TCR complexes were able to interact with Nck in vitro after stimulation with anti-CD3 in cells from mice expressing the WT CD3. In cells expressing the PRS mutant CD3, however, TCR complexes were not pulled down with the Nck-SH3.1 domain, indicating the interaction between CD3 and Nck depends on the CD3.PRS.
The phenotype and functionality of the thymocytes and splenocytes were subsequently analyzed. The percentage of CD4+ and CD8+ thymocytes in mice expressing CD3.PRSM appeared comparable with that of CD3.WT mice (Fig. 2A). No differences were observed in the phenotype of the mice at either early (4–8 wk after transfer) or late (12–14 wk after transfer) time points. There did appear to be a very slight increase in the percentage of cells in the early double-negative (DN) stages of thymocyte development in the PRSM mice compared with WT (Fig. 2B; DN1–3; CD25–44+, CD25+44+, and CD25+44–). However, the resultant reduction of DN4 cells did not appear to affect the number of double-positive or mature single-positive thymocytes, because these were comparable between both groups (Fig. 2C). In addition, stimulation of thymocytes with anti-CD3 induced comparable levels of CD69 expression in CD4+CD8+ cells from both groups of mice (Fig. 2D and data not shown). Finally, there appeared to be no obvious defects in negative selection between the two groups of mice, as measured by the deletion of T cells bearing mammary tumor virus-specific TCRs (Fig. 2E) (37).
No differences were seen in the percentage or number of T or B cells in the spleens of the two groups of mice (Fig. 3, A and B). In addition, both groups had equivalent numbers of CD4+ and CD8+ peripheral T cells (Fig. 3, A and B) and comparable levels of TCR expression (C). Functional analysis was also performed on splenocytes from mice expressing either WT or PRSM CD3. No differences were observed in the ability of splenic T cells from these mice to constitutively internalize or down-modulate their TCR after SEB stimulation (Fig. 3D and data not shown). Likewise, CD69 expression after SEB stimulation on T cells from both groups of mice was comparable (Fig. 3E). Finally, T cells from both groups of mice proliferated effectively after stimulation with SEB and anti-CD3 (Fig. 3F). Cytokine analysis was performed on T cells stimulated with both anti-CD3 and SEB. Although no discernible differences were observed, additional studies must be performed to address this in more detail (data not shown). These data suggest that the interaction between Nck and the CD3 PRS is not required for T cell development, the establishment of peripheral T cells, and T cell function.
Discussion
These data suggest that the interaction between the CD3 PRS and Nck, or presumably any other SH3 domain-containing molecule that could bind to this sequence, is not required for T cell development and function. These results were surprising given the potential significance attached to a recent study indicating a role for CD3:Nck interaction in T cell activation and function in vitro (29, 38). In this study, the authors inhibited CD3:Nck interaction by overexpressing the SH3.1 domain of Nck in Jurkat T cells. This inhibited formation of T cell:APC conjugates as well as IS formation and greatly reduced activation of the cells as measured by IL-2 production and proliferation. What may account for the differences seen here?
One possibility is that this discrepancy may be due to differences in the systems used. Perhaps in mice this interaction is not as critical as in human cells. An alternative possibility, which we favor, is that overexpression of Nck SH3.1 not only interferes with Nck:CD3 interaction but also abrogates interaction with multiple proteins that mediate IS formation, such as cytoskeletal regulatory molecules, and T cell activation. Thus, the observations made by these authors demonstrate an important role for Nck rather than Nck:CD3 interaction (29). This has been suggested by others as well (39, 40). Recently, it has been demonstrated, using molecular imaging techniques and fluorescence resonance energy transfer analysis, that recruitment of Nck to the site of TCR engagement and subsequent actin polymerization requires the adaptor proteins linker for activation of T cells and SH2 domain-containing leukocyte protein of 76 kDa and that TCR-induced tyrosine phosphorylation of these proteins must first take place (38).
Nck interacts with CD3 through the first N-terminal SH3.1 domain, whereas interaction with PAK and WASP, for instance, occurs via its second and third SH3 domains (SH3.2 and SH3.3), respectively (29, 41). It could therefore be argued that overexpression of SH3.1 would not interfere with Nck binding to these proteins, which are key regulators of cytoskeletal rearrangement and thus IS formation. However, it has been shown that although actual binding occurs through a single SH3 domain, multiple SH3 domains of Nck are required for productive interaction to occur, suggesting cooperative use of multiple SH3 domains for tight complex formation. This has been observed with Nck binding to Cbl, Bcr/Abl, and p155 as well as WASP and PAK (42, 43). Using various Nck GST fusion proteins, only combined SH3 domains of Nck or full-length Nck facilitated binding. In addition, using an Ab-based system to alter concentrations of Nck at the plasma membrane, all three SH3 domains were required for actin polymerization (44). Our system alleviates these potential problems by specifically abolishing the interaction of CD3 and Nck through mutation of CD3 rather than alteration of Nck.
What is the purpose of Nck:CD3 interaction in T cells? Some have suggested that this interaction forms a bridge between the TCR and misshapen NIK-related kinase (MINK), a kinase that interacts with Nck, that has recently been described to play a role in negative selection in the thymus (37, 45). Cells in which MINK expression was knocked down had a defect in deletion of thymocytes reactive to the endogenous superantigen Mtv as well as T cells expressing self-reactive TCR transgenes. If this hypothesis is correct, mice unable to recruit Nck to CD3 might display a defect in negative selection. However, no apparent defects in selection were observed in the CD3.PRSM mice generated in this study. Given our findings, the possible link between MINK and the TCR complex should be re-evaluated.
Our data indicate that the interaction between Nck and the CD3 PRS is not required for T cell development or function in mice. However, it is important to note that these data do not necessarily contradict the suggestion by Gil et al. (29, 36) that the CD3 complex undergoes a conformational change after TCR engagement. In these studies, binding of the TCR led to the exposure of the CD3 PRS that was detected in vitro using the Nck SH3.1 domain. Thus, a conformational change may still occur after ligation of TCR complexes containing the CD3 PRS mutation, although this would not be detected using this method. Exposure of the CD3 PRS is likely to be only one of several changes that occur in the CD3 complex as a result of TCR ligation. Indeed, we have shown that ligation of the TCR leads to the exposure of the CD3 N terminus, which ordinarily is buried within the complex, and dissociation of CD3 from the TCR complex (46).
We cannot completely rule out the possibility that this interaction does have some function that was not revealed in our assays, although it clearly does not have a major, significant role in T cell development and function. It is possible that it functions redundantly with another interaction involving the TCR:CD3 complex. We considered the possibility, for instance, that Nck might also be recruited to the PRS in pT. However, pull-down experiments showed that Nck binds to this sequence very weakly, if at all (S. Dilioglou, A. L. Szymczak, and D. A. A. Vignali, unpublished data). Thus, although Nck does interact with CD3, our study clearly shows that T cell development and function is not altered in mice expressing the CD3 PRS mutation, inferring a less significant role for this interaction than was originally suggested.
Acknowledgments
We are very grateful to Cox Terhorst and David Baltimore for reagents. We also thank Richard Cross and Jennifer Hoffrage for assistance with flow cytometry, the staff in the Hartwell Center for oligonucleotide and peptide synthesis and DNA sequencing, and members of the Vignali laboratory for constructive criticisms, comments, and bone marrow isolation.
Disclosures
The authors have no financial conflict of interest.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This work was supported by National Institutes of Health Grant AI-52199, Cancer Center Support Center of Research Excellence Grant CA-21765, and by a grant from the American Lebanese Syrian Associated Charities.
2 Address correspondence and reprint requests to Dr. Dario A. A. Vignali, Department of Immunology, St. Jude Children’s Research Hospital, 332 North Lauderdale, Memphis, TN 38105-2794. E-mail address: dario.vignali@stjude.org
3 Abbreviations used in this paper: IS, immunological synapse; SH3, Src homology 3; PRS, proline-rich sequence; PRSM, PRS alanine mutant; PAK, p21-activating kinase; WASP, Wiskott-Aldrich syndrome protein; WT, wild type; SEB, staphylococcal enterotoxin B; DN, double negative; MINK, misshapen NIK-related kinase.
Received for publication August 4, 2004. Accepted for publication April 19, 2005.
References
Call, M. E., J. Pyrdol, M. Wiedmann, K. W. Wucherpfennig. 2002. The organizing principle in the formation of the T cell receptor-CD3 complex. Cell 111: 967-979.
Bonifacino, J. S., C. K. Suzuki, J. Lippincott-Schwartz, A. M. Weissman, R. D. Klausner. 1989. Pre-Golgi degradation of newly synthesized T-cell antigen receptor chains: intrinsic sensitivity and the role of subunit assembly. J. Cell Biol. 109: 73-83.
Geisler, C.. 1992. Failure to synthesize the CD3- chain: consequences for T cell antigen receptor assembly, processing, and expression. J. Immunol. 148: 2437-2445.
Sussman, J. J., J. S. Bonifacino, J. Lippincott-Schwartz, A. M. Weissman, T. Saito, R. D. Klausner, J. D. Ashwell. 1988. Failure to synthesize the T cell CD3- chain: structure and function of a partial T cell receptor complex. Cell 52: 85-95.
Clevers, H., B. Alarcon, T. Wileman, C. Terhorst. 1988. The T cell receptor/CD3 complex: a dynamic protein ensemble. Annu. Rev. Immunol. 6: 629-662.
Love, P. E., E. W. Shores, M. D. Johnson, M. L. Tremblay, E. J. Lee, A. Grinberg, S. P. Huang, A. Singer, H. Westphal. 1993. T cell development in mice that lack the chain of the T cell antigen receptor complex. Science 261: 918-921.
Dave, V. P., Z. Cao, C. Browne, B. Alarcon, G. Fernandez-Miguel, J. Lafaille, A. de la Hera, S. Tonegawa, D. J. Kappes. 1997. CD3 deficiency arrests development of the but not the T cell lineage. EMBO J. 16: 1360-1370.(Andrea L. Szymczak*, Creg)