| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Regulates Fyn Activity and Cbp/PAG Phosphorylation in Thymocyte Lipid Rafts1
* Department of Pediatrics and
Departments of Pathology and Laboratory Medicine, British Columbia Research Institute for Children’s and Women’s Health, University of British Columbia, Vancouver, Canada; and
Clinical Research Institute of Montreal, Montreal, Canada
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
(PTP) in immune cell function and regulation of Src family kinases was investigated using thymocytes from PTP-deficient mice. PTP-null thymocytes develop normally, but unstimulated PTP–/– cells exhibit increased tyrosine phosphorylation of specific proteins, increased Fyn activity, and hyperphosphorylation of Cbp/PAG that promotes its association with C-terminal Src kinase. Elevated Fyn activity in the absence of PTP is due to enhanced phosphorylation of Fyn tyrosines 528 and 417. Some PTP is localized in lipid rafts of thymocytes, and raft-associated Fyn is specifically activated in PTP–/– cells. PTP is not a Cbp/PAG phosphatase, because it is not required for Cbp/PAG dephosphorylation in unstimulated or anti-CD3-stimulated thymocytes. Together, our results indicate that PTP, likely located in lipid rafts, regulates the activity of raft Fyn. In the absence of PTP this population of Fyn is activated and phosphorylates Cbp/PAG to enhance association with C-terminal Src kinase. Although TCR-mediated tyrosine phosphorylation was apparently unaffected by the absence of PTP, the long-term proliferative response of PTP–/– thymocytes was reduced. These findings indicate that PTP is a component of the complex Src family tyrosine kinase regulatory network in thymocytes and is required to suppress Fyn activity in unstimulated cells in a manner that is not compensated for by the major T cell PTP and SFK regulator, CD45. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
(PTP)3 is a widely expressed transmembrane enzyme that functions to regulate Src family tyrosine kinases (SFKs) (reviewed in Ref.1). Overexpression of PTP in fibroblasts, embryonal carcinoma cells, PC12 cells, and A431 cells respectively induces transformation (2), promotes neuronal differentiation (3) and outgrowth (4), or enhances adhesion of cells to the substratum (5) that is concomitant with Src activation. PTP expression can also induce Fyn activation (6). Conversely, mice with targeted disruption of the PTP gene have substantially reduced brain Src and Fyn activity (7, 8). Embryonic fibroblasts from PTP-deficient mice have reduced activity of these SFKs and consequently impaired integrin signaling that is manifested by delayed cell spreading, migration, and haptotaxis (7, 8, 9). PTP-mediated SFK activation is achieved through dephosphorylation of the C-terminal regulatory tyrosine residue of the SFKs (2, 3, 7, 8). The above studies have established a role for PTP as an SFK phosphatase and positive regulator of the SFKs in multiple contexts.
PTP has never been implicated in immune cell signaling, in which SFKs play key initial roles. In T cells, regulated tyrosine phosphorylation is key to TCR signaling and consequent cell proliferation (10). An early recognizable event upon TCR engagement is the induction of protein tyrosine phosphorylation by the SFKs Lck and Fyn. Indeed, Lck- or Fyn-deficient cell lines and mice show defects in T cell signaling and proliferation (11, 12, 13, 14, 15). Activated Lck and Fyn phosphorylate immunoreceptor-based tyrosine activation motifs in the TCR/CD3 subunits, resulting in recruitment and activation of the Syk family tyrosine kinase Zap70 (16, 17). Phosphorylation by Zap70 or other proteins such as linker for activation of T cells (LAT) and SH2 domain-containing leukocyte protein of 76 kDa (18, 19) mediates the formation of multicomponent complexes and activates several T cell-signaling pathways (20, 21).
Lck and Fyn activities are closely regulated before and upon T cell stimulation. The receptor PTP CD45 can positively regulate Lck activity by dephosphorylating an inhibitory phosphotyrosine in the C-terminal tail of Lck (Tyr505) and can also negatively regulate Lck by dephosphorylating the autophosphorylation loop tyrosine (Tyr394) (22, 23, 24, 25, 26, 27, 28, 29, 30, 31). Studies using CD45-deficient cell lines and thymocytes indicate that in an unstimulated T cell, CD45 dephosphorylates both regulatory tyrosines of Lck with an overall effect of limiting kinase activity (28, 32). It is proposed that upon T cell activation, CD45 loses its proximity to Lck, resulting in rephosphorylation of Tyr394 and activation of Lck (33). Activated Lck colocalizes with and activates Fyn (34, 35). T cell SFK phosphorylation and activities are also modulated by the tyrosine kinase C-terminal Src kinase (Csk). In unstimulated T cells, constitutive tyrosine phosphorylation of the transmembrane protein Csk-binding protein (Cbp)/PAG (phosphoprotein associated with glycosphingolipid-enriched microdomains) promotes formation of a Csk-Cbp/PAG complex mediated through Csk-Src homology 2 (SH2) domain binding to phospho-Tyr314 of Cbp/PAG (36, 37). In this way, Csk is localized at the membrane where it can phosphorylate the C-terminal tyrosine residue of SFKs to inhibit their activities (36, 37, 38), perhaps counteracting the activity of CD45 on this site and contributing to overall low SFK activity. Csk also associates through its SH3 domain with PEP (39), a PTP implicated in the negative regulation of TCR signaling through its dephosphorylation of the activation loop tyrosine in SFKs (40). A membrane-localized Csk-PEP complex could augment the negative regulation of SFKs in unstimulated T cells. T cell stimulation rapidly induces dephosphorylation of Cbp/PAG, with release and relocalization of Csk to the cytosol, thus relieving Csk-mediated SFK inhibition (36, 41). CD45 itself is required for efficient TCR-stimulated Cbp/PAG dephosphorylation (38). Thus, the phosphorylation and activities of Lck and Fyn are modulated by an interactive network of kinases and phosphatases that is still not well understood. Functional outcomes are likely to be orchestrated by dynamic, regulated localization of components of this kinase-phosphatase signaling module to the membrane and, in particular, to the lipid rafts that are proposed to serve as platforms for T cell signaling (42, 43).
The role of PTP as an activator of Src and Fyn in nonimmune cells together with the critical roles played by activated SFKs in TCR-mediated signaling led us to investigate a potential T cell function of PTP as a regulator of SFK activity. Altered Fyn phosphorylation and enhanced Fyn activity were detected in unstimulated thymocytes from PTP-deficient mice, specifically involving a population of Fyn that colocalizes with PTP in lipid rafts. Also, PTP indirectly regulates phosphorylation of the lipid raft protein Cbp/PAG and its association with the SFK kinase inhibitor Csk. Although TCR-mediated tyrosine phosphorylation was apparently unaffected by the absence of PTP, the long-term proliferative response of PTP–/– thymocytes was reduced. PTP thus plays a novel and unique role in resting thymocytes as a positive and negative Src family kinase regulator, and is required for optimal TCR-mediated thymocyte proliferation.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
The PTP-deficient mice have been previously described (7) and were maintained as an advanced intercross line (129SvEv x Black Swiss, 50:50 mixed background). Animal care and use followed the guidelines of the Canadian Council on Animal Care and were reviewed and approved by the University of British Columbia. All experiments were conducted with 8- to 9-wk-old mice. The Jurkat cell line JE6.1 and a derivative cell line lacking CD45 expression (J45.01) were provided by Dr. P. Johnson (University of British Columbia, Vancouver, Canada).
Reagents and Abs
Abs to phosphotyrosine (PY-20), CD45, CD28, Fyn, Lck, and Csk were purchased from BD Transduction Laboratories. Other Abs to phosphotyrosine (4G10; Upstate Biotechnology), Fyn (Santa Cruz Biotechnology), and transferrin receptor (Zymed Laboratories) were also used. Fyn phosphorylation status was investigated using phosphosite-specific Abs raised to the C-terminal phosphorylation site (pTyr529) and autophosphorylation site (pTyr418) of Src (BioSource International) that can recognize homologous phosphorylation sites in Fyn. Anti-LAT Ab was purchased from Santa Cruz Biotechnology. FITC-conjugated anti-CD4 and PE-conjugated anti-CD8a Abs, FITC-conjugated anti-CD25 and -CD69 Abs, and appropriate isotype controls, were purchased from eBioscience. Anti-
-actin Ab, anti-hamster IgG, n-octyl -glucopyranoside, PMA, and ionomycin were obtained from Sigma-Aldrich. PP2 (4-amino-5-(4-chlorophenyl)-7-(t-butyl)pyrazolo [3,4-d]pyrimidine) was purchased from Tocris Cookson, and PP3 (4-amino-7-phenylpyrazol[3,4-]pyrimidine) was obtained from Calbiochem. Purified anti-CD3
Ab (clone 145.2C11) was prepared in our laboratory. Rabbit anti-PTP antiserum 2205 (44) and mouse anti-Cbp/PAG antiserum (38) have been previously described.
T cell preparation
Thymus, spleen, or mesenteric lymph nodes were compressed with a syringe plunger and passed through a 100-µm pore size nylon cell strainer (Falcon; BD Biosciences) to produce single-cell suspensions. After depletion of RBC, thymocytes, splenocytes, and lymph node lymphocytes were used for experiments. For proliferation studies, CD3+ and naive CD4+ T cells were purified from thymocyte suspensions using murine T Cell and CD4+ T Cell Enrichment kits (StemCell Technologies).
Flow cytometry
Single-cell suspensions of 1 x 106 thymocytes or splenocytes were incubated (30 min, 4°C) with saturating amounts of FITC-conjugated anti-CD4 and PE-conjugated anti-CD8a Abs and appropriate isotype controls. Cells were washed twice with PBS and resuspended in equal amounts of RPMI 1640 for analysis on a FACSCalibur flow cytometer (BD Biosciences).
T cell stimulation
For proliferation experiments, purified CD3+ and CD4+ T cells (2 x 105) or lymph node T cells (5 x 104) were cultured in 100 µl of RPMI 1640 medium supplemented with 10% FCS, 2 mM L-glutamine, 100 U/ml penicillin, and 100 µg/ml streptomycin. Cells were stimulated in triplicate with plate-bound anti-CD3 (10 µg/ml) and/or anti-CD28 (5 µg/ml) Abs or with 1 ng/ml PMA and 200 ng/ml ionomycin for 48 h at 37°C, then pulsed with [3H]thymidine (PerkinElmer) for the final 18 h of culture and harvested onto printed glass-fiber filtermates (Wallac) using a cell harvester. [3H]Thymidine incorporation was measured using a multidetector liquid scintillation counter. The IL-2 concentration in the supernatants was assayed using an ELISA kit from eBioscience.
For stimulation before lysis, 1 x 107 thymocytes in 0.5 ml of RPMI 1640 medium were incubated for 10 min with 20 µg/ml anti-CD3, then cross-linked with anti-hamster IgG at 37°C for various times as indicated in the figures. The cells were pelleted by brief (20-s) centrifugation and lysed with Triton-based buffer (20 mM Tris (pH 7.4), 120 mM NaCl, 1% Triton X-100, 5 mM EDTA, 50 mM sodium fluoride, 1 mM Na3VO4, and 10 µg/ml aprotinin and leupeptin) or modified RIPA buffer (20 mM Tris (pH 7.4), 120 mM NaCl, 1% Triton X-100, 0.5% deoxycholate, 0.1% SDS, 10% glycerol, 5 mM EDTA, 50 mM sodium fluoride 1 mM Na3VO4, and 10 µg/ml aprotinin and leupeptin).
Kinase assay
Fyn immunoprecipitates were prepared from 200–300 µg of RIPA lysates using anti-Fyn (FYN 3-G; Santa Cruz Biotechnology). These were used in in vitro kinase assays as previously described (7) and were immunoblotted with anti-Fyn Ab (BD Transduction Laboratories).
Isolation of a raft fraction
Raft fractions were prepared as previously described (45). In brief, 1 x 108 thymocytes were lysed with 0.5 ml of MBS (25 mM MES (pH 6.5), 150 mM NaCl, 1% Triton X-100, 5 mM EDTA, 2 mM Na3VO4, 50 mM sodium fluoride, 1 mM PMSF, and 10 µg/ml aprotinin and leupeptin), incubated on ice for 30–45 min, gently mixed with an equal volume of 80% sucrose (w/v) in MBS, and placed in the bottom of a clear tube (344062; Beckman Coulter). The samples were then overlaid with 2 ml of 30% sucrose and 1 ml of 5% sucrose in MBS and centrifuged at 40,000 rpm in a SW60 rotor for 16–18 h at 4°C. After centrifugation, nine fractions of 450 µl each were collected, starting at the top of the gradient. Equal aliquots of each fraction were resolved by SDS-PAGE for immunoblotting. For immunoprecipitation experiments, equal aliquots from fractions 3 and 4 (the lipid raft fractions) or from fractions 8 and 9 (the Triton X-100-soluble fractions) were combined. The lipid raft pool was incubated with 20 mM n-octyl -glucopyranoside for 1 h before immunoprecipitation.
COS cell transfection
COS-1 cells were transfected with 4 µg of plasmid expressing wild-type (WT) or mutant Fyn (46) with or without 2 µg of plasmid expressing WT PTP or inactive mutant PTP-C414S/C704S (6), using Lipofectamine reagent (Invitrogen Life Technologies). Twenty-four hours after transfection, cells were lysed in modified RIPA buffer.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
–/– T cells
PTP is readily detectable in thymocytes and splenocytes of WT mice, albeit at lower levels than in brain (Fig. 1A). CD45 is a major transmembrane PTP of nucleated hemopoietic cells that plays a key role in T cell development and activation (47). CD45 expression was not altered in PTP-null thymocytes or splenocytes (Fig. 1A). T cell development appeared normal in PTP–/– mice; there were equivalent populations of CD4 and CD8a single-positive and CD4/CD8a double-negative and double-positive cells (Fig. 1B) and similar expressions of CD25 and CD69 (data not shown) in the thymi and spleens of these and WT mice. The sizes and cellularity of these tissues did not significantly differ in WT and PTP-deficient mice (data not shown).
|
–/– thymocytes
An early event upon TCR engagement is increased protein tyrosine phosphorylation that is initially mediated through the activities of the SFKs Lck and Fyn. This response was examined in WT and PTP–/– thymocytes. Anti-CD3-induced phosphorylation of proteins of 
120, 55, 42, and 20 kDa was observed in both WT and PTP–/– cells (Fig. 2A). The phosphotyrosine content of the 120-kDa protein 1 and 10 min after stimulation was somewhat less in the PTP–/– cells, and those of the 55- and 20-kDa proteins were elevated and disappeared more rapidly in stimulated PTP–/– cells. Equivalent ERK activation was detected in WT and PTP–/– cells (Fig. 2B). However, the major difference observed was between unstimulated WT and PTP–/– thymocytes, with a dramatically increased phosphorylation of a band(s) of 75–80 kDa and a lesser increase in 60- and 120-kDa bands observed in the PTP–/– cells (Fig. 2A, arrows). Zap70 and LAT were not detectably phosphorylated in the unstimulated PTP–/– cells (data not shown). The p75–80 tyrosine phosphorylation was reduced in response to anti-CD3 stimulation of PTP–/– cells, whereas in WT cells, p75–80 tyrosine phosphorylation rapidly increased upon anti-CD3 stimulation and was maintained at a high level for at least 10 min.
|
–/– thymocytes
PTP-deficient mice exhibit reduced Src and Fyn activity in brain and embryonic fibroblasts, indicating that PTP is a positive regulator of these Src family kinases (7, 8). We thus investigated the activity of Fyn, a Src family kinase that is involved in very early events of TCR signaling, in resting and stimulated PTP–/– thymocytes. Fyn activity was significantly elevated in unstimulated PTP–/– cells compared with WT cells (1.64 ± 0.1 vs 1 U, respectively; p < 0.05), as assessed by Fyn immunoprecipitation and in vitro kinase assays (Fig. 3A, top panel, and B). Anti-CD3 stimulation initially increased the low basal Fyn activity in WT cells, but progressively reduced the high basal Fyn activity in PTP–/– thymocytes (Fig. 3, A and B). Thus, despite very different Fyn activities in resting WT and PTP–/– thymocytes, the Fyn activities in both cell types after anti-CD3 stimulation were similar. This closely parallels the protein tyrosine phosphorylation patterns in these cells and suggests that the enhanced Fyn activity in resting PTP–/– thymocytes may be the cause of the enhanced tyrosine phosphorylation of certain proteins, such as that of those represented by p75–80.
|
–/– thymocytes compared with WT cells (Fig. 3A, middle panel, and B). PTP can dephosphorylate the regulatory C-terminal Tyr528 of Fyn (6), and in accord with this, we observed that at least part of the increased phosphotyrosine content of Fyn in PTP–/– thymocytes was due to 40% increased phosphorylation at Tyr528 (Fig. 3, C–E). Phosphorylation at Tyr417 in the activation loop of Fyn was also increased by 70% in the PTP–/– cells (Fig. 3, C–E). This suggests that PTP can dephosphorylate both phosphorylation sites on Fyn, with an overall effect of reducing Fyn activity in unstimulated thymocytes. In contrast, no significant differences in Fyn Tyr528 or Tyr417 phosphorylation were detected between anti-CD3-stimulated WT and PTP–/– thymocytes (Fig. 3, C and D).
Heterologously expressed WT and mutant forms of FynB and FynT were used to verify the specificity of phosphosite-specific Abs raised to the C-terminal and autophosphorylation sites of Src (Fig. 4A). There was some residual detection of FynB-K299M, an inactive Fyn mutant that cannot autophosphorylate at Y417, by the phospho-Y417-specific Ab (anti-Src pTyr418). Similarly, we observed residual detection of FynB-Y528F and FynT-Y528F, forms of Fyn possessing a Phe substitution at Tyr528 (FynT-Y528F) that eliminates the C-terminal tail phosphorylation site, by the phospho-Y528 Ab (anti-Src pTyr529). However, these signals were minor compared with those detected for Fyn possessing the appropriately phosphorylated tyrosine residue. Thus, these Abs can be used to detect the two phosphoforms of Fyn in WT and PTP–/– thymocytes as in the experiments shown in Fig. 3. In addition, these Abs were used in PTP/Fyn cotransfection experiments to confirm that PTP could indeed dephosphorylate Fyn-Tyr417, because this specificity of PTP has not been previously reported. Probing FynT immunoprecipitates for total phosphotyrosine content showed that PTP, but not a catalytically inactive PTP mutant, dephosphorylated FynT or FynT-Y528F (Fig. 4, B–D). Probing with phosphosite-specific Abs verified that both FynT and FynT-Y528F were dephosphorylated at Tyr417 in the presence of PTP, and as expected, FynT was also dephosphorylated at Tyr528 (Fig. 4, B–D).
|
is present in thymocyte cell lipid rafts
Areas of the plasma membrane enriched in cholesterol and sphingolipids, termed lipid rafts, are believed to serve as signaling platforms in which T cell signaling proteins reside and to which other proteins are recruited upon TCR stimulation to form signaling complexes (42, 43). Because T cell Fyn is a likely PTP substrate, and because a significant portion of Fyn in T cells is present in raft fractions (34, 48), we investigated whether PTP was localized to lipid rafts. Fyn and Lck, together with LAT, an adaptor protein involved in TCR signaling that is targeted to lipid rafts (49), were present in raft fractions isolated from WT mouse thymocytes (Fig. 5). As reported previously (38, 50), a small population of the receptor PTP CD45 was detected in raft fractions, with the majority of the CD45 localizing to nonraft fractions. Similar to CD45, a small amount of PTP was present in raft fractions, with most PTP segregating in the nonraft, Triton X-100-soluble fractions (Fig. 5). The transferrin receptor is reported to be a Triton X-100-soluble membrane protein (51, 52) and was indeed exclusively detected in nonraft fractions of thymocytes (Fig. 5), indicating that the presence of CD45 and PTP in raft fractions was not due to contamination of these fractions with detergent-soluble material.
|
–/– thymocytes
Populations of PTP and Fyn are present in raft and nonraft thymocyte fractions. We therefore assessed the activities of both pools of Fyn from WT and PTP–/– thymocytes. Raft-associated Fyn was twice as active in unstimulated PTP–/– cells as in WT cells (Fig. 6, A and C). Anti-CD3 stimulation of PTP–/– cells reduced Fyn activity in rafts. In WT cells, anti-CD3 treatment increased Fyn activity in the raft fractions (Fig. 6, A and C). In contrast, anti-CD3 induced little to no changes in nonraft Fyn activity in WT and PTP–/– thymocytes (Fig. 6B). These results indicate that TCR stimulation in these cells is mainly linked to raft Fyn activation. They also suggest that PTP differentially regulates the activity of lipid raft-associated and nonraft-localized Fyn populations, with preferential and significant inhibitory effects on Fyn in lipid rafts.
|
–/– thymocytes
Cbp/PAG is a transmembrane protein localized in lipid rafts (36, 37). Tyrosine-phosphorylated Cbp/PAG recruits Csk, a tyrosine kinase that phosphorylates and negatively regulates Src family kinases, to rafts. The roles of PTP and Cbp/PAG-Csk as Src family kinase regulators prompted us to investigate whether Cbp/PAG phosphorylation and association with Csk were altered in thymocytes lacking PTP. Probing raft fractions of PTP–/– samples with anti-phosphotyrosine Ab detected enhanced phosphorylation of a band that comigrated with Cbp/PAG (Fig. 7A). In accord with this, more Csk was present in the lipid raft fractions of PTP–/– cells than of WT cells (Fig. 7A, bottom panels). To confirm these observations, Cbp/PAG immunoprecipitates from WT and PTP-deficient thymocytes were probed for phosphotyrosine, revealing increased Cbp/PAG phosphorylation in the absence of PTP–/– (Fig. 7B). An increased amount of Csk was associated with the hyperphosphorylated Cbp/PAG in PTP–/– cells (Fig. 7B).
|
is not required for TCR-stimulated Cbp/PAG dephosphorylation
TCR stimulation induces rapid tyrosine dephosphorylation of Cbp/PAG (36, 41). To determine whether PTP plays a role in this process, the Cbp/PAG phosphorylation status was determined in PTP–/– cells after anti-CD3 treatment. Although Cbp/PAG in unstimulated PTP–/– thymocytes was highly phosphorylated, it was rapidly dephosphorylated upon stimulation (Fig. 8A). By 1 min of stimulation, only minimal tyrosine phosphorylation of Cbp/PAG was detectable in PTP–/– cells, and this was similar to that in stimulated WT cells. In both cell types, Cbp/PAG remained dephosphorylated for at least 10 min after stimulation. Thus, PTP is not required for efficient TCR-stimulated Cbp/PAG dephosphorylation.
|
–/– thymocytes is not due to the absence of PTP-mediated Cbp/PAG dephosphorylation
Cbp/PAG is phosphorylated by SFKs (36, 37, 48). The enhanced Cbp/PAG phosphorylation observed in resting thymocytes lacking PTP could be due to increased activity of Fyn and/or a reduction in Cbp/PAG phosphatase activity, with PTP being a potential Cbp/PAG phosphatase in unstimulated cells. To distinguish between these possibilities, PTP–/– cells were incubated with the Src family kinase inhibitor PP2 or the inactive analog PP3, and Cbp/PAG phosphorylation was monitored by immunoprecipitation and probing with anti-phosphotyrosine Ab. If Cbp/PAG hyperphosphorylation in PTP–/– thymocytes is a consequence of higher Fyn activity, PP2-mediated inhibition of Fyn would allow any Cbp/PAG phosphatase activity to predominate, and Cbp/PAG dephosphorylation would be predicted to occur. However, if, in addition to its role as a regulator of Fyn, PTP functioned as a Cbp/PAG phosphatase, the predicted Cbp/PAG dephosphorylation could not occur in the cells lacking PTP. As shown in Fig. 8B, PP2, but not PP3, treatment of unstimulated PTP–/– thymocytes resulted in reduced Cbp/PAG phosphorylation, consistent with an endogenous phosphatase activity that is manifested upon blockade of Src family kinase action. This suggests that enhanced Cbp/PAG phosphorylation in PTP–/– thymocytes is primarily due to the loss of the negative regulatory action of PTP on Fyn, rather than to the loss of a direct action of PTP on Cbp/PAG as a phosphatase.
Cbp/PAG phosphorylation is also elevated in T cells lacking CD45 (38). To determine whether CD45 functions as a Cbp/PAG phosphatase in unstimulated T cells, thus potentially acting in this manner in PP2-treated PTP–/– thymocytes, the CD45-negative Jurkat cell line J45.01 was treated with PP2, and Cbp/PAG tyrosine phosphorylation was monitored. As shown in Fig. 8C, the inhibition of SFK resulted in Cbp/PAG dephosphorylation and significantly reduced association with Csk. Treatment of the same cells with the control compound PP3 had no effect on these parameters. Thus, the enhanced Cbp/PAG phosphorylation observed in T cells lacking CD45 or PTP is not due to the lack of direct dephosphorylation of Cbp/PAG by either of these PTPs.
Impaired PTP–/– T cell proliferation
To determine whether PTP is involved in long-term parameters of T cell activation, the proliferative ability of PTP–/– thymocytes was tested. Purified CD3+ and CD4+ cells lacking PTP proliferated less well than WT cells in response to anti-CD3 stimulation (Fig. 9, A and C). Costimulation with either anti-CD28 or PMA increased proliferation, but PTP–/– thymocytes still responded less well than WT cells. Stimulation with PMA and the calcium ionophone ionomycin resulted in equivalent proliferation of WT and PTP–/– mutant cells. The above stimulations involving anti-CD3 also elicited reduced IL-2 production by PTP–/– thymocytes compared with WT cells (Fig. 9, B and D). Costimulation with PMA and ionomycin resulted in equivalent IL-2 production by WT and PTP-deficient cells. These results demonstrate that the proliferative and IL-2 production capabilities of thymocytes lacking PTP are intact. However, PTP is specifically required for efficient TCR-stimulated proliferation and IL-2 synthesis. In contrast, no defects in proliferation or IL-2 production were detected in stimulated lymph node T cells from WT and PTP–/– animals (Fig. 9, E and F). Consistent with this, protein tyrosine phosphorylation profiles were similar in PTP–/– and WT lymph node T cells before and after anti-CD3 stimulation (Fig. 10A). Furthermore, Fyn tyrosine phosphorylation and activity were unaltered in lymph node T cells by ablation of PTP (Fig. 10B). This suggests that PTP is not involved in Fyn regulation in resting peripheral T cells or in TCR-mediated signaling in these cells.
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
in T cells has revealed several differences linked to regulated protein tyrosine phosphorylation in unstimulated PTP–/– mouse thymocytes compared with WT mouse thymocytes. First, PTP–/– thymocytes exhibited enhanced tyrosine phosphorylation of several proteins. Second, Fyn phosphorylation and activity were higher in the PTP–/– cells. Third, the phosphorylation of Cbp/PAG, an Src family kinase substrate, was increased in the absence of PTP. Thus, Fyn and Cbp/PAG constitute two hyperphosphorylated proteins in resting PTP–/– thymocytes. In contrast to these alterations, no remarkable differences in protein tyrosine phosphorylation profiles or in Fyn activities were detected after anti-CD3-stimulation of PTP-deficient thymocytes compared with WT thymocytes. Although not ruling out the possibility that PTP is involved in TCR-stimulated signaling events, these results strongly suggest an important role for PTP in regulating tyrosine phosphorylation in thymocytes before TCR activation.
The elevated protein tyrosine phosphorylation in unstimulated PTP–/– thymocytes indicates that the absence of PTP skews the basal control of phosphorylation, either through a lack of PTP-mediated protein dephosphorylation or by increasing tyrosine kinase activity. Consistent with the latter possibility, the kinase activity of Fyn was elevated in PTP-deficient thymocytes. PTP, therefore, seems to function in resting thymocytes as a negative regulator of Fyn. This is in contrast to its role in brain and mouse embryo fibroblasts as a positive regulator of Src and Fyn (7, 8). Clearly, Src and Fyn activities were reduced in PTP–/– brain and fibroblasts, in conjunction with increased phosphorylation of the negative regulatory C-terminal tyrosine residue. However, Fyn from PTP–/– thymocytes possessed higher Fyn activity and increased phosphotyrosine content due to enhanced phosphorylation at the C-terminal negative regulatory Tyr528 site and at the positive regulatory Tyr417 site in the activation loop. Conversely, reduced FynT Tyr528 and Tyr417 phosphorylation occurred upon coexpression with PTP in COS cells. These findings suggest that PTP can act as either a positive or a negative regulator of thymocyte Fyn. In PTP–/– thymocytes, elevated Tyr417 phosphorylation probably predominates to effect Fyn activation, as is the case with and has been discussed for Lck in CD45–/– mouse thymocytes (53). In studies that directly measured Lck activity and phosphorylation status in CD45–/– thymocytes, the kinase activity of Lck was increased 2- to 4-fold, and Lck was hyperphosphorylated on negative (Tyr505) and positive (Tyr394) regulatory residues (31, 32, 54). Fyn activity and its phosphorylation status at specific sites have not been reported in CD45–/– thymocytes, but, like Lck, the overall tyrosine phosphorylation of Fyn is increased by 2-fold (54). As we observed in PTP–/– thymocytes, the basal tyrosine phosphorylation of certain proteins is also increased in CD45–/– thymocytes (31, 54). Our results indicate that PTP and CD45 regulate SFK activity, at least in unstimulated thymocytes, in similar ways and with comparable magnitudes.
A consequence of the loss of PTP is hyperphosphorylation of Cbp/PAG. This enhances the interaction of Cbp/PAG with Csk, mediated through recognition by the Csk SH2 domain of phospho-Tyr314 of Cbp/PAG (36, 37). Inhibiting SFK activity by PP2 treatment of PTP–/– thymocytes permits a rapid dephosphorylation of Cbp/PAG and abrogates Csk binding to Cbp/PAG, demonstrating that increased Cbp/PAG phosphorylation in the PTP–/– cells is probably a consequence of Fyn activation rather than due to the absence of PTP-mediated Cbp/PAG dephosphorylation. Indeed, Fyn can phosphorylate Cbp/PAG (36) and appears to be responsible for most Cbp/PAG phosphorylation, because this is greatly reduced in Fyn–/– T cells (48). Some Cbp/PAG tyrosine phosphorylation is maintained during SFK inhibition in PTP–/– thymocytes. This indicates either that PTP acts as a phosphatase on these sites or that TCR engagement activates a specific PTP that is not constitutively functional in resting cells. The latter possibility seems most likely, because the residual level of Cbp/PAG phosphorylation and the low amount of still detectable bound Csk approximate those in unstimulated WT cells where PTP is present. TCR stimulation induces rapid and considerable dephosphorylation of Cbp/PAG (36, 41), and this is not altered in PTP–/– cells. Together, our results indicate that PTP is not a Cbp/PAG phosphatase in resting or stimulated thymocytes.
Cbp/PAG phosphorylation is also increased in unstimulated CD45–/– thymocytes and in a T cell line lacking CD45 (38). As in PTP–/– thymocytes, this appears to be due to increased SFK activity rather than to a lack of CD45-mediated Cbp/PAG dephosphorylation, because ablation of SFK activity by PP2 treatment of CD45–/– Jurkat cells results in Cbp/PAG dephosphorylation (Fig. 8C and Ref.36). However, unlike PTP, CD45 does play a role in anti-CD3 stimulated Cbp/PAG dephosphorylation, because this is inhibited in CD45–/– thymocytes (38). This difference might contribute to the more severely impaired TCR-mediated signaling and proliferation of T cells lacking CD45 (54, 55, 56, 57, 58, 59) than of those lacking PTP.
CD45 is an abundant membrane protein of T cells and is a key regulator of Lck and Fyn. Nevertheless, CD45 cannot compensate for the lack of PTP as a regulator of Fyn activity in resting thymocytes, highlighting an unsuspected and unique action of PTP on Fyn in these cells. Conversely, PTP cannot compensate for CD45 function in T cells, as is obvious from the major defects in thymocyte development observed in CD45-deficient mice and the severely impaired proliferation of CD45–/– thymocytes upon TCR engagement (54, 58, 59, 60). Although this could be due to what are probably much lower amounts of PTP than of CD45 in T cells or to a lower specific activity of PTP than CD45 (61), these factors cannot account for the inability of CD45 to compensate for the absence of PTP. Generally, CD45-null mice share some T cell developmental and proliferation defects with mice lacking Lck, whereas PTP–/– mice are more similar to mice lacking Fyn in having normal T cell development and less severely impaired proliferation. Perhaps this reflects relatively greater contributions of CD45 to Lck regulation and of PTP to Fyn regulation. It is also possible that CD45 and PTP regulate different populations of Fyn that are distinguished by the localization of various phosphatase- kinase pools within the T cell. Increased Fyn activity in PTP–/– thymocytes is specific to the pool of Fyn associated with detergent-insoluble lipid rafts. Our preliminary results do not reveal differences in Lck activity in PTP–/– and WT thymocytes (data not shown). Notably, Fyn and Lck have overlapping, yet distinct, subcellular localizations in resting T cells, with most Lck segregated outside lipid rafts, in contrast to the preferential presence of Fyn in raft microdomains (34, 35). We have shown that a subpopulation of PTP is present in raft T cell fractions; however, so is some CD45. Conceivably, additional factors such as the interaction of PTP or CD45 with other raft proteins could determine the abilities of these PTPs to regulate raft SFKs. For example, CD45 activity toward Lck can be enhanced by the isoform-specific association of CD45 with CD4 (62). PTP can form a complex with contactin, a GPI-linked cell surface molecule of neuronal cells (63). GPI-anchored proteins are typical components of lipid rafts (64, 65), indicating the potential for PTP to interact with other raft receptor molecules in T cells.
Our study also reveals a novel role for PTP in TCR-stimulated thymocyte proliferation. Anti-CD3, alone or with costimulators, induced significantly less proliferation and reduced IL-2 production by PTP–/– thymocytes compared with WT cells. In contrast, proliferation and IL-2 production were normal in PTP–/– T cells from lymph nodes, suggesting that PTP is not essential for these TCR-mediated responses in peripheral T cells. Furthermore, Fyn in resting PTP–/– lymph node T cells was not hyperphosphorylated or activated compared with that in WT lymph node T cells. Studies of Fyn–/– T cells have found that Fyn is required for optimal TCR-stimulated thymocyte, but not splenic T cell, proliferation (11, 13). Our results are consistent with PTP acting as a Fyn regulator, with functional effects on TCR responses in thymocytes, but not in mature peripheral T cells.
The impaired proliferation of thymocytes lacking Fyn reflects a requirement for functional Fyn for optimal proliferation-linked signaling events. In contrast, Fyn is hyperactive in PTP–/– thymocytes before treating the cells with proliferation stimuli, yet proliferation is also impaired. This is reminiscent of the defective TCR-mediated responses of thymocytes with hyperactive Lck before TCR stimulation (CD45–/–) (54) and of thymocytes and peripheral T cells lacking Lck (12, 15). Like PTP–/– thymocytes, CD45–/– cells also possess hyperphosphorylated Cbp/PAG (38). In unstimulated CD45–/– and PTP–/– thymocytes, this may be a down-regulatory mechanism to reduce the elevated Lck and Fyn activities. Thus, although Lck and Fyn are physically and functionally required for optimal TCR-induced proliferation, these observations indicate that these SFKs must be maintained at a lower activity or threshold state before TCR stimulation to function properly in subsequent TCR-linked responses. This is exemplified by the physiological situation in which thymocytes and other T cells become anergic after overstimulation or inappropriate stimulation, such as by anti-CD3 in the absence of costimulators. Interestingly, increased activity or expression of Fyn has been associated with anergy induction in T cells (66, 67, 68, 69), although the molecular mechanisms involved in anergization are not clear. It is possible, therefore, that the absence of PTP results in inappropriate and constitutive activation of Fyn and perhaps of other proteins, leading to partial anergy or an anergic-like state of the thymocytes and subsequent hypoproliferative responses to TCR stimulation. Other mechanisms may be alternatively or additionally involved as well. For T cells to successfully proliferate, the early phosphorylation and/or activation of multiple signaling proteins are required. The absence of critical regulators, such as CD45 and Lck (i.e., CD45–/– or Lck–/– T cells), results in a dramatic block of TCR-proximal phosphorylation/activation events. In contrast, the absence of other proteins linked to TCR-mediated activation, such as Fyb/Slap/ADAP (adhesion and degranulation promoting adaptor protein) (70, 71) or EphB6 (72), and as we report in this study, PTP, does not impact these immediate events despite resulting in significant in vitro proliferative defects of the same T cells. Costimulatory and/or cell attachment-regulated signaling may contribute to differences between short- and long-term T cell activation parameters. For example, the absence of ADAP also profoundly affects TCR-stimulated, integrin-mediated adhesion (70, 71). Integrin receptors provide costimulatory signals to naive T cells by stabilizing contacts between the TCR and immobilized anti-CD3 Ab and decreasing the threshold for activation (73, 74, 75). Because PTP positively regulates integrin signaling in fibroblasts (8, 9), it might play a role in Ag-mediated interactions between the TCR and T cell integrins as well. Additional investigation is required into the precise signaling interplay that compromises the proliferation of PTP-null thymocytes.
|
Acknowledgments |
|---|
|
Disclosures |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
|
Footnotes |
|---|
1 This work was supported by the National Cancer Institute of Canada (to A.V.) and the Canadian Institutes of Health Research (to A.V. and C.J.P.). L.M. is the recipient of a Bertram Hoffmeister Postdoctoral Fellowship, and H.T.L. is the recipient of a Graduate Studentship from the British Columbia Research Institute for Children’s and Women’s Health. A.V. is a Senior Investigator with the Canadian Institutes of Health Research and holds the Canada Research Chair in Signaling in the Immune System. C.J.P. is the recipient of an Investigatorship Award from the British Columbia Research Institute for Children’s and Women’s Health. 
2 Address correspondence and reprint requests to Dr. Catherine J. Pallen, British Columbia Research Institute for Children’s and Women’s Health, 950 West 28th Avenue, Vancouver, British Columbia, Canada V5Z 4H4. E-mail address: cpallen{at}interchange.ubc.ca
3 Abbreviations used in this paper: PTP, protein tyrosine phosphatase ; Csk, C-terminal Src kinase; LAT, linker for activation of T cell; SFK, Src family tyrosine kinase; SH2, Src homology 2; WT, wild type.
Received for publication May 20, 2005. Accepted for publication September 27, 2005.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
(PTP): a src family kinase activator and mediator of multiple biological effects. Curr. Top. Med. Chem. 7: 821-835. activates pp60c-src and is involved in neuronal differentiation. EMBO J. 12: 3789-3798. [Medline] regulates src family kinases and alters cell-substratum adhesion. J. Biol. Chem. 273: 31890-31900. and p59fyn. J. Biol. Chem. 273: 8691-8698. leads to constitutive downregulation of the kinases src and fyn. Curr. Biol. 9: 535-538. [Medline] activates Src-family kinases and controls integrin-mediated responses in fibroblasts. Curr. Biol. 9: 505-511. [Medline] regulates integrin-stimulated FAK autophosphorylation and cytoskeletal rearrangement in cell spreading and migration. J. Cell Biol. 160: 137-146. [Medline]
chain. Cell 71: 649-662. [Medline]
, WT; 
, PTP
, WT; 
, no coexpressed PTP
