SEARCH

SEARCH BY CITATION

Keywords:

  • Biochemistry;
  • Developmental immunology;
  • Signal transduction ;
  • TCR

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements

T cell receptor (TCR) signal transduction is mediated by the immunoreceptor tyrosine-based activation motifs (ITAM). The ten ITAM in the TCR complex are distributed in two distinct signaling modules termed TCR ζζ and CD3 γϵ/δϵ. To delineate the specific role of the ζ ITAM in T cell development and TCR signal transmission, we compared the properties of T cells from different TCR ζ-transgenic lines wherein tyrosine-to-phenylalanine substitutions had been introduced in the ζ subunit. These lines lack selected phosphorylated forms of TCR ζ including just p23, both p21 and p23, or all phospho-ζ derivatives. We report herein that the efficiency of positive selection in HY TCR-transgenic female mice was directly related to the number of ζ ITAM in the TCR. In contrast, TCR-mediated signal transmission and T cell proliferative responses following agonist peptide stimulation were similar and independent of the ζ ITAM. Only the duration of MAPK activation was affected by multiple ζ ITAM substitutions. These results strongly suggest that the ITAM in the CD3 γϵ/δϵ module can provide normal TCR signal transmission, with ζ ITAM providing a secondary function facilitating MAPK activation and positive selection.

Abbreviations:
ITAM:

immunoreceptor tyrosine-based activations motifs

LAT:

linker for activation of T cells

PTK:

protein tyrosine kinase

SEB:

staphylococcal enterotoxin B

SLP-76:

SH2-containing leukocyte protein of 76 kDa

ZAP-70:

ζ-associated protein of 70 kDa

Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements

The TCR is a multimeric complex comprised of the ligand-binding αβ heterodimer and the associated signaling subunits, TCR ζ and CD3 γ, δ and ϵ 1, 2. These signaling subunits contain ten copies of a semi-conserved cytoplasmic amino acid sequence (YxxLx6–8YxxL) termed the immunoreceptor tyrosine-based activation motif (ITAM) 3, 4. Upon receptor ligation, two tyrosine residues within each ITAM are rapidly phosphorylated by a member of the Src-family protein tyrosine kinases (PTK), transforming them into high-affinity ligands for the Syk PTK (reviewed in 5, 6. The coordinated actions of the Src and Syk PTK initiate a cascade of signals that ultimately lead to T cell proliferation, cytokine secretion and effector functions 6.

The TCR complex has a structural design that includes two independent signaling modules, the TCR ζζ (six of the ten ITAM) or the CD3 γϵ/δϵ complexes (four of the ten TCR ITAM) 79. The phosphorylated ITAM can activate similar signaling processes, suggesting they function additively (reviewed in 5, 9, 10). However, biochemical studies revealed a hierarchy of distinct protein interactions with phospho-ζ vs. phospho-CD3 γ, δ and ϵ, implying a functional dichotomy between the modules 1113. A common premise is that the TCR ζζ module predominates in signal transmission. First, TCR engagement results in the tyrosine phosphorylation of the 16-kDa TCR ζ subunit, resulting in the presence of two distinct phospho-ζ forms of 21 and 23 kDa (p21 and p23) 1417. The CD3 γ, δ and ϵ subunits are difficult to detect as tyrosine-phosphorylated proteins 16, 18, 19. Second, p21 is constitutively expressed when isolated from thymocytes and peripheral T cells, complexing with an inactive pool of ζ-associated protein of 70 kDa (ZAP-70) PTK 20. The constitutive expression of p21 results from TCR interactions with peptide/MHC complexes in situ and is only found with TCR ζ and not the γ, δ and ϵ subunits 2022.

Given these observations, it was somewhat surprising to find that T cells were fully functional in their ability to respond to TCR stimulation, even when all of the signaling capabilities of TCR ζ were eliminated by truncations or ITAM substitutions 7, 2328. There are several explanations for these findings. First, the CD3 γϵ/δϵ module could be the principal signaling module in T cells. This idea was supported by reports that TCR engagement by peptide/MHC ligands induced a conformational change in CD3 ϵ, facilitating the initiation of TCR-mediated signaling 29, 30. Although the conformational change in CD3 ϵ subunit resulted in CD3 ϵ complexing Nck, this interaction was not required for T cell functions 29, 31. Second, maximal T cell receptor signal transmission could be elicited by a small number of ITAM from any combination of invariant chains.

In all these studies, neither explanation could be adequately addressed since the comparative signaling and functional studies were undertaken with peripheral T cells that had undergone normal development in the thymus. To precisely determine the contribution of the TCR ζ ITAM in TCR-mediated signal transmission and development, we compared T cell functions with distinct TCR ζ-transgenic lines wherein particular tyrosine-to-phenylalanine substitutions were engineered in ζ. These lines lacked either p23 while maintaining the constitutively phosphorylated p21, both p21 and p23, and/or all phospho-ζ intermediates 5, 17. These transgenic lines were crossed with the HY TCR-transgenic mice, wherein a direct dependence on ζ ITAM for positive selection has previously been noted 23, 27. We have identified a critical role for the ζ ITAM in the positive selection of HY-restricted cells that is unaffected by the constitutive expression of p21. Interestingly, TCR-mediated signal transmission and T cell effector functions were independent of ζ ITAM numbers. The TCR ζ ITAM did contribute to prolonged MAPK activation. Taken together, our findings reveal a normal functional role for the CD3 γϵ/δϵ module in TCR signal transmission when all the ζ ITAM are eliminated.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements

The CD3 γϵ/δϵ module is sufficient for normal proliferative responses of T cells

To examine how the number of TCR ζ ITAM in the TCR complex regulate T cell responses, we compared the proliferative responses of peripheral T cells isolated from different TCR ζ-transgenic lines where particular tyrosine residues were substituted with phenylalanine. The transgenic lines are called YF1,2, YF5,6 and YF1–6. The YF1,2 line retains the constitutively phosphorylated p21 with no induced p23, the YF5,6 line contains weak phospho-ζ forms of p19 and p20, and the YF1–6 line has no phospho-ζ intermediates 5, 17.

Peripheral lymph node T cells from a wild-type mouse exhibited a dose-dependent increase in proliferation following stimulation with either an mAb against CD3 ϵ, the T cell mitogen Con A, or the superantigen staphylococcal enterotoxin B (SEB) (Fig. 1A–C). T cells from the YF1,2 line, where the constitutively phosphorylated p21 form of TCR ζ is maintained, had a slightly augmented proliferative response to all the stimuli, while T cells expressing only weakly phosphorylated species of ζ (YF5,6) had comparable proliferative responses as wild-type T cells (Fig. 1A–C). Interestingly, T cells with no phospho-ζ forms had equivalent proliferative responses when compared to wild-type T cells (Fig. 1A–C) 5. These data suggest that the signals generated through the TCR complex in response to mAb, mitogen, superantigen or peptide/MHC stimulation can proceed with/without the TCR ζ ITAM 17, 23, 25, 26.

thumbnail image

Figure 1. T cell proliferation can occur independently of the TCR ζ ITAM. Peripheral T cells (2.5×105) from wild-type and TCR ζ-transgenic mice with selected tyrosine-to-phenylalanine substitutions in the first (YF1,2), third (YF5,6) or all three (YF1–6) TCR ζ ITAM were incubated with increasing concentrations of anti-CD3 ϵ (A), Con A (B) or SEB (C) for 72 h. Proliferation was assessed by [3H]thymidine uptake.

Download figure to PowerPoint

TCR ζ ITAM facilitate the development of HY TCR-transgenic T cells

The findings described above as well as previous publications suggested that the TCR signaling either functioned through the CD3 module and/or required a limited number of ITAM in the TCR complex, provided by any ITAM combination. Yet, all the signaling and functional studies, including those described above, utilized T cells that developed normally in the absence of the TCR ζ ITAM 5, 23, 26, 28. To definitely address the contribution of TCR ζ ITAM to T cell functions, we crossed our mice with the HY αβ TCR-transgenic line. The positive selection of male-reactive T cells in the HY TCR-transgenic female mice is directly dependent on ITAM numbers 27.

Since the contributions of particular phospho-ζ intermediates including p21 and p23 had not been examined in this system, we compared the efficiency of T cell development in the various HY and HY/YF sets of female mice (Fig. 2A). Positive selection of male-specific T cells in the HY TCR-transgenic female mice resulted in the development of CD4CD8+ thymocytes expressing the HY TCR, which was detected by the clonotypic mAb T3.70 (Fig. 2A) 32. In age-matched females that expressed p21 in the absence of p23 (HY/YF1,2), the efficiency of positive selection of thymocytes was reduced 2.4-fold compared to wild-type HY mice (5% vs. 12.1%; p<0.001) (Fig. 2A, B; Table 1). The elimination of both p21 and p23 (HY/YF5,6) also resulted in a 2.4-fold reduction in CD4CD8+T3.70+ thymocytes (5.1% vs. 12.1%; p<0.001) (Fig. 2A, B). Finally, the absence of all phospho-ζ intermediates in the HY/YF1–6 female mice almost completely abolished the development of mature HY-specific CD4CD8+ cells since most of the cells were T3.70 (1.3% vs. 12.1%; p<0.001) (Fig. 2A; Table 1).

thumbnail image

Figure 2. Positive selection in HY/TCR ζ-transgenic female mice is dependent on the number of ITAM available in the TCR ζ subunit. (A) Thymocytes were isolated from the indicated 6-wk-old HY and HY/YF female mice. Single-cell suspensions were prepared and aliquots stained with fluorochrome-labeled mAb detecting CD4, CD8 and the clonotypic TCR (T3.70). The cells were analyzed by flow cytometry. (B) The percentage and statistical variation in the CD4CD8+ thymocytes were compared among the HY, HY/YF1,2, HY/YF5,6 and HY/YF1–6 female mice. At least seven mice per group were analyzed. (C) Lymph node cells were isolated from 6-wk-old HY, HY/YF1,2, HY/YF5,6 and HY/YF1–6 female mice and analyzed by flow cytometry as described for thymocytes.

Download figure to PowerPoint

Table 1. The positive selection of HY-restricted T cells in HY TCR-transgenic female mice is directly dependent on ITAM numbers in the TCRa)
 Thymocytes (×107)% CD4+CD8+% CD4+CD8% CD4CD8+
  1. a) The HY and HY/YF series of mice were analyzed once they were between 4 and 8 wk of age. The statistical error is derived from at least six mice/group.

HY9.3 ± 2.857.0 ± 5.45.2 ± 1.112.0 ± 5.5
HY/YF1,29.1 ± 5.372.0 ± 4.96.2 ± 1.64.8 ± 2.1
HY/YF5,613.0 ± 572.0 ± 4.06.9 ± 1.44.7 ± 1.8
HY/YF1–6.229.3 ± 3.472.7 ± 4.98.2 ± 1.71.7 ± 0.8

The reduced efficiency of positive selection in HY/YF lines relative to the HY female mice resulted in fewer HY-specific T cells in the peripheral lymphoid organs. In the HY female mice, the male-reactive CD8+ T cells were approximately 30% of CD4CD8+ cells in the lymph node (Fig. 2C). In the HY/YF1,2 and HY/YF5,6 mice, these percentages were reduced threefold while only 6% of the CD4CD8+ cells in the HY/YF1–6 mice expressed the transgenic TCR (Fig. 2C). The similar percentage and number of T3.70+ T cells in both the HY/YF1,2 and HY/YF5,6 lines (each retains eight of the ten ITAM in the TCR) suggested that the TCR ζ ITAM function additively, rather than p21 and p23 contributing differentially, to positive selection.

The CD3 γϵ/δϵ signaling module is fully functional in the HY TCR-transgenic T cells in the absence of TCR ζ ITAM

The HY/TCR ζ-transgenic mice represented the best experimental system to determine the contributions of the TCR ζζ and CD3 γϵ/δϵ modules on TCR signal transmission since the TCR ζ ITAM were directly involved in positive selection, implying that they were coupled to the strength of signal transmission. To address how TCR signal transmission is affected by the TCR ζ ITAM, we assessed the phosphorylation state of multiple signaling proteins following TCR interactions with a control peptide or agonist peptide (Smcy)-loaded APC. We analyzed first the phosphorylation state of the TCR ζ and the CD3 subunits.

In wild-type HY thymocytes, p21 was constitutively phosphorylated (Fig. 3A, lane 1). Stimulation with Smcy elicited the expression of p23 (Fig. 3 A, lane 2). In the HY/YF1,2 line, p21 was constitutively expressed but no p23 was induced with the agonist peptide (Fig. 3A, lanes 3–4). Thymocytes from the HY/YF5,6 line contained weak phosphorylated intermediates of 19 and 20 kDa, while no phospho-ζ intermediates were detected in the HY/YF1–6 cells (Fig. 3A, lanes 5–8). Notably, the CD3 ϵ and CD3 δ subunits were inducibly phosphorylated in all the HY and HY/YF female mice following agonist peptide treatment when compared to control peptide stimulations (Fig. 3B, lanes 2, 4, 6, 8 vs. 1, 3, 5, 7). These findings suggested that the CD3 ITAM were fully functional in the absence of phosphorylated TCR ζ intermediates.

thumbnail image

Figure 3. The TCR/CD3 ITAM are phosphorylated in the HY and HY/YF thymocytes. (A) The constitutive and inducible phosphorylation of TCR ζ was assessed by stimulating thymocytes from the HY, HY/YF1,2, HY/YF5,6, and HY/YF1–6 mice with either control peptide (AV)- or agonist peptide (Smcy)-loaded cells for 10 min at 37°C (control peptide: lanes 1, 3, 5, 7; Smcy peptide: lanes 2, 4, 6, 8). The cells were then lysed, and TCR ζ was immunoprecipitated. The precipitates were resolved by SDS-PAGE and Western-immunoblotted with anti-phosphotyrosine (mAb 4G10) followed by anti-ζ mAb. (B) The tyrosine-phosphorylated status of the CD3 subunits was assayed by stimulating the indicated thymocytes for 10 min with control or agonist peptides. The CD3 subunits were immunoprecipitated with anti-CD3 ϵ mAb (145–2C11). Precipitates were Western-immunoblotted with anti-phosphotyrosine.

Download figure to PowerPoint

We compared next the phosphorylation state of ZAP-70 and SH2-containing leukocyte protein of 76 kDa (SLP-76), two key molecules involved in TCR signal transduction. In wild-type HY thymocytes incubated with a null peptide, ZAP-70 was only weakly tyrosine-phosphorylated but did complex with p21 (Fig. 4A, lane 1). Stimulation of these cells with agonist peptide increased the tyrosine phosphorylation of ZAP-70 and resulted in the co-precipitation of p23 (Fig. 4A, lane 2).

thumbnail image

Figure 4. The tyrosine phosphorylation of early TCR signaling intermediates does not require the ζ ITAM. (A) The induction of ZAP-70 PTK tyrosine phosphorylation and the presence of co-precipitating phospho-ITAM was assessed by stimulating thymocytes from the HY, HY/YF1,2, HY/YF5,6 and HY/YF1–6 mice with either control peptide (AV)- or agonist peptide (Smcy)-loaded dendritic cells for 10 min at 37°C (control peptide: lanes 1, 3, 5, 7; Smcy peptide: lanes 2, 4, 6, 8). The cells were then lysed, and ZAP-70 was immunoprecipitated with an anti-ZAP-70 mAb. The immunoprecipitates were resolved by SDS-PAGE and Western-immunoblotted with anti-phosphotyrosine (mAb 4G10) followed by anti-ZAP-70 mAb. (B) The tyrosine-phosphorylated status of SLP-76 and the associated adaptor protein LAT was assayed using the stimulation conditions described in (A). Precipitates were Western-immunoblotted with anti-phosphotyrosine followed by anti-SLP-76 antisera. (C) Thymocytes from the various mice were stimulated with the indicated peptides for 0, 3, 10, 30 and 90 min and the duration of ZAP-70 phosphorylation was assayed as in (A). The data are representative of at least three independent experiments.

Download figure to PowerPoint

In cells from the HY/YF1,2 mice, ZAP-70 was again inducibly phosphorylated following agonist peptide stimulation (Fig. 4A, lane 4). Unlike in wild-type thymocytes, no p23 was expressed in the HY/YF1,2 line (Fig. 4A, lane 4). Agonist peptide stimulation of thymocytes from the HY/YF5,6 line also caused an increased phosphorylation of ZAP-70, at levels equal to or higher than the wild-type HY female cells (Fig. 4A, lane 6). The phosphorylated forms of TCR ζ that co-precipitated with ZAP-70 were weak phospho-ζ intermediates of 19 and 20 kDa (Fig. 4A, lanes 5, 6). Surprisingly, agonist peptide stimulation of thymocytes from the HY/YF1–6 line led to an almost equivalent phosphorylation of ZAP-70 when compared with the HY controls (Fig. 4A, lane 8). This occurred in spite of the absence of any phospho-ζ intermediates and suggested that the small degree of CD3 ϵ and CD3 δ phosphorylation was sufficient for normal ZAP-70 activation (Fig. 4A, lanes 7, 8).

We also analyzed the phosphorylation state of SLP-76, a physiological substrate of catalytically active ZAP-70. In all the HY and HY/YF series of mice analyzed, TCR interactions with the agonist peptide resulted in an induced phosphorylation of SLP-76 (Fig. 4B, lanes 2, 4, 6, 8 vs. 1, 3, 5, 7). In addition, tyrosine-phosphorylated linker for activation of T cells (LAT) was found to co-precipitate with SLP-76 in all the HY and HY/YF series of cells following TCR triggering (Fig. 4B, lanes 2, 4, 6, 8 vs. 1, 3, 5, 7) 33. We did note that the magnitude of SLP-76 phosphorylation was reduced in the HY/YF1,2 line relative to all the other HY and HY/YF series of mice (four of five experiments). The experiments were also repeated with thymic epithelial cells loaded with null or agonist peptide with similar results, although the magnitude of phosphorylation was less than with dendritic cells (data not shown).

Since the preceding analyses of ZAP-70 and SLP-76 were undertaken at a single time point, it was conceivable that signaling differences among the HY and HY/YF series of T cells could have occurred at later time points following TCR engagement. Hence, we also examined the kinetics of ZAP-70 tyrosine phosphorylation. In wild-type HY females, ZAP-70 was inducibly phosphorylated within 3 min of stimulation and maintained at 30 min (Fig. 4C, lanes 2–6). The null peptide (AV) had minimal effects on ZAP-70 phosphorylation (Fig. 4C, lanes 6, 7). Very similar kinetics of ZAP-70 phosphorylation was detected in the HY/YF1,2, HY/YF5,6 and HY/YF1–6 lines (Fig. 4C). Similar kinetics of SLP-76 phosphorylation was observed in all the HY and HY/YF series mice (data not shown). Altogether, these data indicate that the induced phosphorylation of ZAP-70, SLP-76 and LAT could occur normally in the absence of the TCR ζ ITAM, and was unaffected by the expression of the constitutively phosphorylated p21.

MAPK activation is facilitated by the TCR ζ ITAM

We extended our analyses of signal transmission by comparing the kinetics of MAPK (p42/p44) phosphorylation in the various HY and HY/YF transgenic lines following agonist peptide stimulations. We selected p42/p44 because their phosphorylation depends on the integration of multiple signaling pathways initiated by ZAP-70 activation that can include Ras-GRP, SLP-76-SOS and Shc. Thus, differences among the HY and HY/YF series of cells might be revealed with a more distal target that involves multiple signaling pathways. Additionally, the prolonged activation of MAPK has been correlated previously with positive selection of thymocytes 34, 35.

In thymocytes from the HY and HY/YF series of female mice, an increase in the phosphorylation of MAPK (p42,p44) was detected within 3 min of agonist peptide encounter (Fig. 5, lanes 2, 8, 14, 20). For the HY, HY/YF1,2 and HY/YF5,6 thymocytes, phospho-p42/p44 was still evident at 10, 30, 90 and even 180 min following TCR engagement (Fig. 5, lanes 3–6, 9–12, 15–18). However, MAPK phosphorylation was reduced at 90 and 180 min in cells from the HY/YF1–6 line (Fig. 5, lanes 22–24; three independent experiments). These data suggest that phospho-ζ can contribute to MAPK activation.

thumbnail image

Figure 5. MAPK phosphorylation is facilitated by the TCR ζ ITAM. Thymocytes (5×105 cells/lane) from HY, HY/YF1,2, HY/YF5,6 and HY/YF1–6 female mice were stimulated with agonist peptide-loaded dendritic cells. The cells were either lysed immediately (t=0) or lysed at 3, 10, 30, 90 or 180 min after stimulation. The lysates were resolved on SDS-PAGE and Western-immunoblotted with anti-phospho-MAPK followed by anti-MAPK polyclonal antisera. The data represent the most dramatic differences from three independent experiments and were consistent in each case.

Download figure to PowerPoint

T cell effector functions can proceed normally in the absence of TCR ζ ITAM

We examined next the functional responses of the T cells from the various HY and HY/YF female mice. Thymocytes from the various mice were stimulated with varying doses of agonist peptide or the positively selecting antagonist peptide Ube1x 36. We compared the late signaling events, including up-regulation of CD69 and down-regulation of TCR. CD69 expression was induced in a dose-dependent manner in the CD4+CD8+ thymocytes from all of the HY and HY/YF lines (Fig. 6A, B). This was true whether the peptide used was the agonist (Smcy) or positively selecting peptide (Ube1x). Similar patterns of TCR down-regulation were also detected in the distinct HY/TCR ζ-transgenic lines following agonist peptide stimulation (data not shown). The levels of CD5 on the T3.70+ T cell populations were equivalent, indicating that modifications in the signaling threshold could not account for our findings (data not shown) 37.

thumbnail image

Figure 6. The induction of CD69 by antagonist peptides is normal whether or not the ζ ITAM are present. Thymocytes were cultured for 19 h with (A) agonist (Sncy) or (B) antagonist (Ube1x) peptide-pulsed antigen splenocytes. The cells were subsequently stained with a number of fluorochrome-labeled mAb and the percentage of CD4+CD8+ thymocytes expressing CD69 was assessed by flow cytometry. The fold induction over control cultures is indicated and the results are representative of four independent experiments.

Download figure to PowerPoint

To characterize the proliferative responses of the HY cells, we isolated T cells from the lymph nodes of the various mice and stimulated them with increasing doses of the Smcy peptide. Small numbers of CD4CD8+T3.70+ cells could develop in the HY/YF1–6 female mice, enabling us to directly examine the proliferative responses of these T cells. Similar dose-dependent proliferative responses were revealed when we compared the HY, HY/YF1,2, HY/YF5,6 and HY/YF1–6 lines (Fig. 7). In fact, a slightly augmented proliferation was noted in the HY/YF1–6 line (Fig. 7).

thumbnail image

Figure 7. Proliferative responses of HY T cells can be mediated by just the CD3 γϵ/δϵ module. Single-cell suspensions of lymph nodes (adjusted for equal numbers of CD4CD8+T3.70+ cells) were prepared from the female HY, HY/YF1,2, HY/YF5,6 and HY/YF1–6 TCR ζ-transgenic mice. The cells were cultured for 72 h with increasing concentrations of agonist peptide (10–4–10–10 M). Proliferation was measured by the addition of [3H]thymidine for the last 16 h of culture.

Download figure to PowerPoint

As a second measure of T cell effector functions, we compared the cytolytic activity of HY, HY/YF1,2 and HY/YF5,6 T cells vs. peptide-loaded EL-4 target cells. For these experiments, the cells were first expanded in vitro since no cytolytic activity could be detected with ex vivo populations of cells. The HY and HY/YF1,2 T cells had identical cytolytic activities at effector-to-target ratios of 1/1, 10/1 and 20/1 (Fig. 8). The HY/YF5,6 T cells had a 20% higher level of cytolytic activity compared to the HY control cells. Since we were unable to obtain sufficient numbers of HY/YF1–6 cells for these assay, we repeated the experiments with the P14 TCR-transgenic system 28. P14 and P14/YF1,2 T cells again exhibited almost equivalent cytolytic functions at all three effector-to-target ratios (Fig. 8). Interestingly, the cytolytic function of the P14/YF1–6 T cells was reduced 50% relative to the P14 controls (Fig. 8).

thumbnail image

Figure 8. Cytolytic functions are relatively normal and independent of the TCR ζ ITAM. T cells were isolated from HY, HY/YF1,2, HY/YF5,6 or P14, P14/YF1,2 and P14/YF1–6 mice and stimulated with the appropriate agonist peptides Smcy (HY system) or p33 (P14 lines). The T cells were subsequently tested for cytolytic activity against EL-4 loaded target cells using a standard chromium release assay. The data are representative of two independent experiments. Standard errors of the mean are included in the figure, but were generally less than 4% of the percent specific lysis.

Download figure to PowerPoint

As a completely independent assay for T cell effector functions, we examined the levels of IFN-γ production in T cells isolated from the spleen following a secondary infection with Listeria monocytogenes. Using intracellular staining, we determined that 11.2, 6.7 and 9.6% of the CD8+ T cells from wild-type, YF5,6 and YF1–6 mice, respectively, produced IFN-γ. Taken together, these data suggest that the major traditional signal transmission pathways leading to T cell effector functions can occur without the TCR ζ ITAM.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements

The TCR complex encompasses ten ITAM that are distributed in two distinct signaling modules (TCR ζζ and CD3 γϵ/δϵ). We used a series of mice that express just the constitutively phosphorylated p21 in the absence of p23, two weak phosphorylated intermediates of p19 and p20, or no phospho-ζ intermediates. In the latter mice, only the ITAM in the CD3 γϵ/δϵ module are left intact. A comparative analysis of these various TCR ζ-transgenic lines revealed a direct role for the TCR ζ ITAM in the positive selection of HY TCR-transgenic T cells. The HY/YF1,2 and HY/YF5,6 female mice had a two- to threefold reduction in the positive selection of HY-restricted CD8+ T cells relative to the wild-type HY mice, while the HY/YF1–6 line had an almost complete absence of mature HY+CD8+ T cells. The ability of the TCR ζ ITAM to support positive selection has been previously reported in TCR-transgenic mice containing low-avidity TCR, such as the HY, DO11.10 and the AND TCR-transgenic lines 23, 27.

We extended these initial findings by determining that the efficiency of positive selection does not directly correlate with the patterns of phospho-ζ intermediates that are detected immediately ex vivo. The YF1,2 line maintained a normal constitutive level of p21 while the YF5,6 line expresses only very weakly induced phospho-ζ intermediates of 19 and 20 kDa. Since ZAP-70 is complexed with p21 in T cells from wild-type, wild-type HY and HY/YF1,2 mice, one might expect that such a pre-existing complex would enhance TCR signal transmission. Yet, the induction of TCR-mediated signaling pathways was equivalent whether or not p21, p23 or any phosphorylated TCR ζ intermediates were present. This was true whether the cells were stimulated with peptide-loaded dendritic cells, thymic epithelial cells, or splenocytes. These findings differ from previous reports showing that p21 can either enhance or inhibit T cell responses, respectively 38, 39. A role for phospho-ζ intermediates was only revealed when MAPK phosphorylation was analyzed, consistent with a role for MAPK activation in the positive selection and the reported interaction of Shc and phospho-ζ, 12, 34, 35. The results suggested that some form of signal transmission is regulated by phospho-ζ.

There are several potential interpretations for our findings. First, the CD3 γϵ/δϵ subunits could form the primary route for TCR-induced signaling pathways involving ZAP-70, SLP-76 and MAPK activation. This is supported by our signaling studies, CD69 induction analyses, equivalent proliferative responses and similar cytolytic properties of peripheral T cells from the HY and HY/YF sets of mice. It is interesting to note that the cytolytic functions of the P14/YF1–6 T cells were reduced 50% relative to wild-type cells. This may indicate a partial contribution of the TCR ζ ITAM-MAPK pathway to cytotoxic activity. Our data also concur with previous studies where signal transmission was independent of phospho-ζ, although none of these earlier studies analyzed mice where a phospho-ζ dependence for positive selection was evident 7, 23, 2528. Supporting our interpretation is a report which identified an important conformational change in CD3 ϵ, leading to the initiation of TCR signals 29. In addition, earlier studies comparing the signaling capacity of the TCR ζ subunit and the CD3 ϵ subunit revealed a unique signaling role for CD3 ϵ in the induction of particular phosphoproteins 40. Finally, the CD3 δ subunit has important, ITAM-independent roles in the activation of MAPK 34, 35.

A second possibility that could account for our findings is that a small number of ITAM, provided by any combination of invariant chain ITAM, might contribute to a threshold level of signal transmission 10. The involvement of additional ITAM would not actually provide for increased signaling beyond the maximum already elicited. Investigations assessing these possibility will necessitate the development of viral transduction systems where combinatorial substitutions of the CD3 γ, δ and ϵ ITAM are introduced 41. Alternatively, knock-ins of the mutated TCR ζ molecules in peripheral T cells might be required.

A third potential explanation for our findings is that the constitutive expression of p21 and its association with ZAP-70 resulted in a small percentage of surface TCR complexes that are no longer involved in signal transmission. When the levels of p21 are sufficiently elevated, TCR signal transmission might be suppressed. This could account for our recent report that the expression of p21, in the absence of p23, attenuated the deletion of thymocytes in HY TCR-transgenic male mice 42. Receptor attenuation has also been described for the IgE receptor signaling system. Weak ligand interactions with the ITAM-containing IgE receptor hoard a critical kinase in the signaling pathway such that subsequent signaling induced by a strong agonist is blocked 43. Finally, the ITAM-containing LMP2A protein from Epstein-Barr virus is expressed at the plasma membrane and constitutively phosphorylated in B cells 44. At high levels, the phospho-ITAM from LMP2A sequesters kinases critical for proper functioning of the ITAM in the B cell receptor.

In summary, we propose that the TCR ζζ and CD3 γϵ/δϵ modules contribute both redundant and non-redundant functions to T cells. Within this model, the CD3 γϵ/δϵ module provides for a normal and major mechanism of signal transmission, while the ITAM from both modules contribute to positive selection. We are currently addressing how the CD3 γϵ/δϵ module dominants in this capacity. This will require a complete analysis of mice bearing combinations of ITAM and non-ITAM sequence mutations in the CD3 γϵ/δϵ subunits. The findings from our study have yielded important insights into the evolutionary design of the TCR complex with two independent and partially redundant signaling modules.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements

Antibodies, cell lines and mice

The fluorochrome-labeled antibodies used for staining and the mAb used for immunoprecipitations and/or Western blotting have been described elsewhere 17, 28. Dendritic cells (DC2.4) were obtained from Dr. K. Rock (Dana-Farber Cancer Institute, Boston, MA). The thymic epithelial cell line 427.1 was graciously provided by Dr. S. Vukmanovic (University of Maryland). TCR ζ-transgenic mice bearing selected tyrosine-to-phenylalanine substitutions in the TCR ζ ITAM were described elsewhere 5, 17. The mice were crossed with the HY and P14 αβ TCR-transgenic lines and genotyped by PCR 17, 28. These TCR ζ-transgenic lines were retained on a C57BL/6 background with the endogenous TCR ζ locus knocked out.

Cell stimulations, immunoprecipitations and Western blotting

The peptides used for stimulating T cells from the different HY/TCR ζ-transgenic lines included the agonist peptide Smcy (KCSRNRQYL) and the natural positively selecting antagonist peptide Ube1x (KSNLNRQFL) 36, 45. The AV peptide (SGPSNTPPEI) was used as a control. Peptides used for P14 experiments were described 17, 28. Biochemical assays for assessing signal transduction in thymocytes were described previously 28. Briefly, 2×106 DC2.4 or 427.1 cells were pulsed for 2 h with control or agonist peptide at 37ºC, washed and used to stimulate thymocytes (7.5×107 cells) for 0, 3, 10, 30, 90 and/or 180 min at 37°C. The cells were lysed in a 1% Triton X-100-containing lysis buffer and the TCR ζ subunit, ZAP-70 or SLP-76 was immunoprecipitated and analyzed as described 28. MAPK activation was assessed with DC2.4 or 427.1 cells fixed with 0.1% glutaraldehyde following peptide loading. Whole-cell lysates were prepared and immunoblotted with antibodies against phospho-p42/p44 or total p42/p44 MAPK.

T cell effector function assays

APC (irradiated ζ–/– female splenocytes) were prepulsed with increasing doses of HY TCR-specific (Smcy and Ube1x) or control peptide (AV) for 1 h at 37ºC. The cells were washed and cultured for 19 h at 37ºC with 5×105 HY/TCR ζ-transgenic thymocytes. Up-regulation of CD69 was assessed in CD4+CD8+ thymocytes by three-color flow cytometry.

T cells were isolated from the lymph nodes of wild-type and TCR ζ-transgenic mice, stained with mAb against CD3 ϵ or T3.70 and analyzed by flow cytometry. T cell numbers were normalized and 2.5×105 T cells/well were plated in triplicate and stimulated with varying concentrations of plate-bound anti-CD3 ϵ (145-2C11), Con A (Sigma) or superantigen SEB for 72 h. Peptide-induced proliferation of HY/TCR ζ-transgenic peripheral T cells was measured after culturing 1×104 T3.70+CD8+ T cells with peptide-pulsed APC for 72 h in the absence of exogenous IL-2. Proliferation was measured by the addition of 1 μCi of [3H]thymidine for the last 16 h of culture as described 28. Cytolytic functions were undertaken as described elsewhere 46. L. monocytogenes infections and assays of T cell functions following secondary infections were undertaken as described 47.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements

This work was supported by grants from the NIH (N.S.C.v.O. and C.W.). We would like to thank Megan Durham, Philip Wrage and Jessica Murphy for technical assistance. We would like to thank Amy White for critical reading of the manuscript.

  • 1

    WILEY-VCH

  • 2

    WILEY-VCH

  • 3

    WILEY-VCH

  • 4

    WILEY-VCH

  • 5

    WILEY-VCH

  • 6

    WILEY-VCH

  • 7

    WILEY-VCH

  • 8

    WILEY-VCH