PTP-PEST, a scaffold protein tyrosine phosphatase, negatively regulates lymphocyte activation by targeting a unique set of substrates
2001; Springer Nature; Volume: 20; Issue: 13 Linguagem: Inglês
10.1093/emboj/20.13.3414
ISSN1460-2075
Autores Tópico(s)Neutrophil, Myeloperoxidase and Oxidative Mechanisms
ResumoArticle2 July 2001free access PTP-PEST, a scaffold protein tyrosine phosphatase, negatively regulates lymphocyte activation by targeting a unique set of substrates Dominique Davidson Dominique Davidson Laboratory of Molecular Oncology, IRCM, 110 Pine Avenue West, Montréal, Québec, H2W 1R7 Canada McGill Cancer Centre, 3655 Drummond Street, Montréal, Québec, H3G 1Y6 Canada Search for more papers by this author André Veillette Corresponding Author André Veillette Laboratory of Molecular Oncology, IRCM, 110 Pine Avenue West, Montréal, Québec, H2W 1R7 Canada McGill Cancer Centre, 3655 Drummond Street, Montréal, Québec, H3G 1Y6 Canada Department of Medicine, 3655 Drummond Street, Montréal, Québec, H3G 1Y6 Canada Department of Biochemistry, McGill University, 3655 Drummond Street, Montréal, Québec, H3G 1Y6 Canada Search for more papers by this author Dominique Davidson Dominique Davidson Laboratory of Molecular Oncology, IRCM, 110 Pine Avenue West, Montréal, Québec, H2W 1R7 Canada McGill Cancer Centre, 3655 Drummond Street, Montréal, Québec, H3G 1Y6 Canada Search for more papers by this author André Veillette Corresponding Author André Veillette Laboratory of Molecular Oncology, IRCM, 110 Pine Avenue West, Montréal, Québec, H2W 1R7 Canada McGill Cancer Centre, 3655 Drummond Street, Montréal, Québec, H3G 1Y6 Canada Department of Medicine, 3655 Drummond Street, Montréal, Québec, H3G 1Y6 Canada Department of Biochemistry, McGill University, 3655 Drummond Street, Montréal, Québec, H3G 1Y6 Canada Search for more papers by this author Author Information Dominique Davidson1,2 and André Veillette 1,2,3,4 1Laboratory of Molecular Oncology, IRCM, 110 Pine Avenue West, Montréal, Québec, H2W 1R7 Canada 2McGill Cancer Centre, 3655 Drummond Street, Montréal, Québec, H3G 1Y6 Canada 3Department of Medicine, 3655 Drummond Street, Montréal, Québec, H3G 1Y6 Canada 4Department of Biochemistry, McGill University, 3655 Drummond Street, Montréal, Québec, H3G 1Y6 Canada *Corresponding author. E-mail: [email protected] The EMBO Journal (2001)20:3414-3426https://doi.org/10.1093/emboj/20.13.3414 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info There is increasing interest in elucidating the mechanisms involved in the negative regulation of lymphocyte activation. Herein, we show that the cytosolic protein tyrosine phosphatase PTP-PEST is expressed abundantly in a wide variety of haemopoietic cell types, including B cells and T cells. In a model B-cell line, PTP-PEST was found to be constitutively associated with several signalling molecules, including Shc, paxillin, Csk and Cas. The interaction between Shc and PTP-PEST was augmented further by antigen receptor stimulation. Overexpression studies, antisense experiments and structure–function analyses provided evidence that PTP-PEST is an efficient negative regulator of lymphocyte activation. This function correlated with the ability of PTP-PEST to induce dephosphorylation of Shc, Pyk2, Fak and Cas, and inactivate the Ras pathway. Taken together, these data suggest that PTP-PEST is a novel and unique component of the inhibitory signalling machinery in lymphocytes. Introduction Antigen receptors on T and B cells play critical roles in lymphocyte development and activation (Weiss and Littman, 1994; Qian and Weiss, 1997; Benschop and Cambier, 1999). Depending on the biological context, their engagement can lead to cell activation, differentiation, anergy, tolerance or death. Accumulating evidence indicates that these various responses invariably are initiated by intracellular protein tyrosine phosphorylation. Whereas antigen receptors are devoid of intrinsic protein tyrosine kinase (PTK) activity, they are associated with subunits containing in their cytoplasmic region a short tyrosine-based signalling motif termed ITAM (for immunoreceptor tyrosine-based activation motif). This motif allows the recruitment of cytoplasmic PTKs. Following antigen receptor stimulation, the ITAMs undergo rapid phosphorylation at two distinct tyrosines (Chan et al., 1994). Genetic and biochemical data show that this phosphorylation is mediated by the Src family of cytoplasmic PTKs (Peri and Veillette, 1994; Weiss and Littman, 1994; Chow and Veillette, 1995). In turn, ITAM tyrosine phosphorylation creates docking sites permitting the binding and activation of another group of PTKs, the Zap-70/Syk family (van Oers and Weiss, 1995). Through their capacity to activate phosphatidylinositol (PI) 3′ kinase, activated Src and Zap-70/Syk kinases also trigger the membrane recruitment and activation of a third class of PTKs, the Tec/Btk family (Yang et al., 2000). In combination, these three types of kinases are responsible for phosphorylation of several substrates during cell activation, including phospholipase C (PLC)-γ, Vav, Cbl, SLP-76, Blnk/SLP-65, LAT and Shc. These targets link antigen receptor stimulation to downstream pathways such as the Ras–MAPK cascade, calcium-regulated events, lipid metabolism and the actin cytoskeleton. Ultimately, they trigger effector functions. Over the past few years, significant interest has been directed towards understanding the mechanisms involved in the negative regulation of lymphocyte activation. There is clear indication that protein tyrosine phosphatases (PTPs) such as SHP-1 and PEP play significant roles in this process (Cyster and Goodnow, 1995; Pani et al., 1995, 1996; Carter et al., 1999; Cloutier and Veillette, 1999; Gjorloff-Wingren et al., 1999; Zhang et al., 1999). SHP-1 and PEP apparently act by dephosphorylating proximal components of the antigen receptor signalling cascade, including Src kinases, ITAMs, Zap-70/Syk-related PTKs and/or central linker proteins such as Blnk/SLP-65 and SLP-76. Consequent to their actions, they induce a global reduction in immunoreceptor-mediated protein tyrosine phosphorylation, thereby preventing lymphocyte activation. PTP-PEST is a 120 kDa cytosolic PTP belonging to the PEP family (Yang et al., 1993; Charest et al., 1995). Unlike its relatives PEP and PTP-HSCF, which are contained in selected haemopoietic cell types, PTP-PEST is expressed ubiquitously. However, higher levels of PTP-PEST are found in haemopoietic tissues (Davidson et al., 1997). Previous studies have revealed that PTP-PEST can physically associate with several signal transduction molecules, including the adaptor molecule Shc, the focal adhesion proteins paxillin, Hic-5, Cas, CasL (or HEF-1) and Sin, and the inhibitory PTK Csk (Habib et al., 1994; Charest et al., 1996; Davidson et al., 1997; Garton et al., 1997; Shen et al., 1998; Coté et al., 1999; Nishiya et al., 1999). These interactions occur via sequences positioned outside the catalytic domain of PTP-PEST (Figure 1A). Figure 1.Structure and expression pattern of PTP-PEST in haemopoietic cells. (A) The primary structure of PTP-PEST is depicted graphically. Two residues in the catalytic domain (Cys231 and Arg237), that are critical for phosphatase activity, are highlighted. The positions of the P1, P2, P3 and NPLH sequences in the C-terminal non-catalytic domain are also shown. (B) Expression pattern of PTP-PEST in mouse haemopoietic cells. The accumulation of PTP-PEST in various purified populations of mouse haemopoietic cells was examined by immunoblotting of equivalent amounts of cell lysates with a rabbit anti-PTP-PEST (PEST) serum. The identity of the higher molecular weight polypeptides found to react with the anti-PTP-PEST sera in this immunoblot is not known. The presence of comparable quantities of cellular proteins in each lane was verified by staining of the immunoblot membrane with Coomassie blue (data not shown). The position of PTP-PEST is indicated on the left. Exposure: 25 s (ECL). Download figure Download PowerPoint A significant insight into the function of PTP-PEST has been provided by analyses of fibroblast cell lines having altered levels of PTP-PEST. Overexpression of PTP-PEST in rat fibroblasts was shown to cause a reduction in cell migration (Garton and Tonks, 1999). Conversely, whereas ablation of PTP-PEST expression in the mouse germline causes early embryonic lethality, embryo fibroblasts derived from these mice were demonstrated to exhibit an increase in cell spreading and a strong defect in motility (Angers-Loustau et al., 1999). Hence, PTP-PEST would appear to be a critical regulator of cytoskeletal organization in non-haemopoietic cells, possibly through its capacity to dephosphorylate focal adhesion proteins. Here we have examined the possible involvement of PTP-PEST in the regulation of haemopoietic cell functions. We find that PTP-PEST is expressed abundantly in various haemopoietic cell types, including B cells and T cells. Through overexpression and antisense experiments, evidence is provided that PTP-PEST is an efficient negative regulator of antigen receptor-induced B- and T-cell activation. Furthermore, biochemical studies and structure–function analyses indicate that this function correlates with the ability of PTP-PEST to dephosphorylate a restricted set of tyrosine phosphorylation substrates in activated lymphocytes, including Shc, Cas, Pyk2 and Fak, and prevent activation of the Ras signalling cascade. Results PTP-PEST is broadly expressed in haemopoietic cells Even though PTP-PEST is expressed in a wide range of tissues, it accumulates in greater amounts in spleen and thymus (Davidson et al., 1997). To examine further the distribution of PTP-PEST in haemopoietic cells, normal mouse haemopoietic cell types were purified as outlined in Materials and methods. Equivalent quantities of cell lysates were then probed by immunoblotting with an antiserum directed against PTP-PEST (Figure 1B). This analysis revealed that the PTP-PEST protein was present in splenic B cells (lane 1), thymocytes (lane 2), splenic T cells (lane 3), natural killer (NK) cells (lane 4) and bone marrow-derived mast cells (BMMCs) (lane 5). The abundance of PTP-PEST in these cells was in the same range as that observed in spleen (lane 6) and thymus (lane 7). Association of PTP-PEST with signalling molecules in lymphocytes Due to the importance of protein tyrosine phosphorylation for lymphocyte activation, we investigated whether PTP-PEST might have a regulatory role in this process. First, the capacity of PTP-PEST isolated from lymphocytes to interact with signalling proteins known to associate with PTP-PEST in non-haemopoietic cells was ascertained (Figure 2). To this end, we used the IgG+ mouse B-cell line A20. This cell line expresses PTP-PEST, but not the related phosphatases PEP and PTP-HCSF (our unpublished results). A20 cells were lysed in non-ionic detergent-containing buffer, and the ability of PTP-PEST to associate with these molecules was studied by immunoblotting of the appropriate immunoprecipitates with anti-PTP-PEST antibodies (Figure 2A). As documented for non-haemopoietic cells (Charest et al., 1996; Davidson et al., 1997; Garton et al., 1997; Shen et al., 1998; Coté et al., 1999), a significant proportion of PTP-PEST polypeptides expressed in A20 cells was found to be associated with Shc (lane 1), Csk (lane 2) and paxillin (lane 3). Small amounts were also observed in anti-Cas immunoprecipitates (lane 4), although their detection required longer autoradiographic exposures (data not shown). No PTP-PEST was detected in immunoprecipitates generated with normal rabbit serum (lane 5). Taking into account the total quantity of PTP-PEST in A20 cells, it was estimated that at least 25, 10, 5 and 2% was complexed with Shc, paxillin, Csk and Cas, respectively (data not shown). Figure 2.Association of PTP-PEST with signalling molecules in mouse A20 B cells. IgG+ A20 B cells were lysed in NP-40-containing buffer, and the ability of PTP-PEST to associate with other polypeptides was determined by immunoblotting of the indicated immunoprecipitates with an anti-PTP-PEST (PEST) serum. The position of PTP-PEST is highlighted by an arrow on the left. (A) Association of PTP-PEST with signalling molecules in unstimulated A20 cells. NRS: normal rabbit serum. Exposure: 60 s (ECL). (B) Effects of BCR stimulation or PMA treatment on the association of PTP-PEST with signalling molecules. A20 cells were either left unstimulated or stimulated for 10 min with F(ab′)2 fragments of SAM IgG antibodies (20 μg/ml) or PMA (100 ng/ml), prior to cell lysis and immunoprecipitation. Exposures: first panel, 3 h; second, third and fourth panels, 10 s (ECL). (C) Time course of BCR stimulation. Cells were stimulated for the indicated times with F(ab′)2 fragments of SAM IgG antibodies. The accumulation of phosphotyrosine-containing proteins was monitored by immunoblotting of total cell lysates with anti-phosphotyrosine (P.tyr) antibodies (first panel), whereas the presence of Shc–PTP-PEST complexes was detected by immunoblotting of Shc immunoprecipitates with an anti-PTP-PEST serum (second panel). Tyrosine phosphorylation of total Shc (third and fourth panels) and PTP-PEST-bound Shc (fifth and sixth panels) was analysed in parallel. Note that whereas both the 52 and 46 kDa Shc isoforms were recovered in Shc immunoprecipitates (fourth panel), only the 52 kDa variant was associated with PTP-PEST (sixth panel). This presumably is because the 46 kDa Shc protein lacks part of the PTB domain, which mediates the binding to PTP-PEST. The migration of pre-stained molecular mass markers is indicated on the right. Exposures: first panel, 14 h; second panel, 4 h; third panel, 15 h; fourth panel, 2.5 h; fifth panel, 15 h; sixth panel, 2 h. Download figure Download PowerPoint The impact of antigen receptor-induced activation on these interactions was characterized next (Figure 2B). A20 cells were stimulated for 10 min with F(ab′)2 fragments of sheep anti-mouse (SAM) IgG antibodies, and the associations were monitored as detailed for Figure 2A. B-cell receptor (BCR) stimulation was noted to provoke a 2- to 3-fold increase in the interaction of PTP-PEST with Shc (first panel, compare lanes 1 and 2). Between 50 and 60% of PTP-PEST became associated with Shc under these conditions. A similar effect was seen when cells were treated with phorbol 12-myristate 13-acetate (PMA) (lane 3), in agreement with an earlier report (Habib et al., 1994). In contrast, the degree of association of PTP-PEST with paxillin (second panel), Csk (third panel) and Cas (fourth panel) was not influenced by treatment with either anti-BCR antibodies (lane 2) or PMA (lane 3). In a time course of BCR stimulation (Figure 2C), it was observed that the induction of the Shc–PTP-PEST interaction was maximal after 10 min of activation (second panel, lane 3), and persisted for at least 40 min (lane 5). This increased stoichiometry was maintained at the later time points despite the progressive disappearance of overall protein tyrosine phosphorylation (first panel) and, in particular, tyrosine phosphorylation of Shc (third panel). It is noteworthy that the pool of Shc molecules that was associated specifically with PTP-PEST (fifth and sixth panels) underwent some degree of tyrosine phosphorylation in response to BCR stimulation. This observation implied that, in the event that PTP-PEST was responsible for Shc dephosphorylation (see below), there could be a window of co-existence of Shc tyrosine phosphorylation and association with PTP-PEST. Inhibition of immunoreceptor-mediated cellular activation by PTP-PEST In light of the high expression levels of PTP-PEST in lymphocytes and its ability to associate with molecules implicated in immunoreceptor signalling, we wanted to evaluate more specifically its effect on lymphocyte activation. Given the lethal impact of ptp-pest gene ablation in the mouse germline (Angers-Loustau et al., 1999), transfection experiments in model cell lines were used for this purpose. A20 B cells were stably transfected with a cDNA encoding wild-type mouse PTP-PEST, as detailed in Materials and methods. Monoclonal cell lines overexpressing PTP-PEST were generated by limiting dilution and identified by immunoblotting of total cell lysates with anti-PTP-PEST antibodies (data not shown). Clones expressing PTP-PEST at levels >5 times those observed in control cells were retained (Figure 3A, top). All the cell lines selected for further studies expressed unaltered levels of BCR, CD45, CD40 and FcγRIIB at the cell surface (data not shown). Figure 3.Effect of PTP-PEST on antigen receptor-induced activation. (A) Stable overexpression of PTP-PEST in A20 B cells. Stable transfectants overexpressing wild-type PTP-PEST (wt PEST) or expressing the neomycin resistance marker alone (Neo) were stimulated with the indicated concentrations of F(ab′)2 fragments of RAM IgG antibodies (abscissa), as detailed in Materials and methods. The accumulation of IL-2 in the supernatant was assessed by measuring tritiated thymidine incorporation in the IL-2-dependent indicator cell line HT-2 (ordinate). All assays were done in triplicate and repeated at least 10 times. The quantities of PTP-PEST contained in the various cell lines used in this report were determined by immunoblotting of total cell lysates with anti-PTP-PEST antibodies (top panel). Lane 1, Neo.1; lane 2, Neo.2; lane 3, Neo.3; lane 4, Neo.5; lane 5, Neo.6; lane 6, wtPEST.70 (18); lane 7, wtPEST.35 (4); lane 8, wtPEST.30 (9); lane 9, wtPEST.38 (11); lane 10, wtPEST.42 (15); lane 11, wtPEST.64 (16); lane 12, wtPEST.70 (18); lane 13, wtPEST.31 (4.5); lane 14, wtPEST.41 (5); lane 15, wtPEST.71 (6) (fold PTP-PEST overexpression is shown in parentheses). The position of PTP-PEST is indicated on the left. Exposure: 6 h. (B) Transient overexpression of PTP-PEST in A20 B cells. A20 cells were transiently transfected with the vector pXM139 alone or bearing a wild-type mouse ptp-pest cDNA, in the presence of an IL-2 promoter–luciferase reporter construct. Cells were then stimulated with F(ab′)2 fragments of SAM IgG antibodies for 6 h, and luciferase activity was measured in a luminometer as detailed in Materials and methods. Results are represented as the percentage of maximal stimulation obtained with PMA plus ionomycin. Expression levels of PTP-PEST are shown at the top. Exposure: 3 h. Download figure Download PowerPoint Previous studies have shown that BCR-induced activation of A20 cells results in production of the lymphokine interleukin-2 (IL-2) (Justement et al., 1989; Muta et al., 1994). Hence, in order to study the impact of PTP-PEST in these cells, transfectants overexpressing PTP-PEST or expressing the neomycin resistance marker alone (Neo) were stimulated with various concentrations of F(ab′)2 fragments of rabbit anti-mouse (RAM) IgG antibodies. After 24 h, supernatants were harvested and assayed for production of IL-2 using the IL-2-dependent indicator cell line HT-2 (Figure 3A). This experiment showed that all A20 clones overexpressing PTP-PEST exhibited a pronounced reduction in BCR-induced IL-2 release, in comparison with control Neo clones. To ensure that the effect of PTP-PEST was not due to compensatory modifications induced in stably transfected cells, transient transfection assays were also performed (Figure 3B). A20 cells were transfected by electroporation with the vector pXM139 alone or bearing a wild-type ptp-pest cDNA, in combination with pIL-2–luciferase, which contains the IL-2 promoter linked to a luciferase reporter. After 40 h, cells were stimulated for 6 h with F(ab′)2 fragments of SAM IgG and assayed for luciferase activity. We found that transient overexpression of PTP-PEST strongly inhibited (∼5-fold) BCR-induced IL-2 promoter activation in A20 cells. In similar experiments conducted using T-cell lines, we observed that PTP-PEST was also able to inhibit T-cell antigen receptor signalling (data not shown). To help ensure that the inhibitory impact of PTP-PEST on immunoreceptor signalling was not an artefactual consequence of overexpression, the effect of ptp-pest antisense expression was also examined (Figure 4A–C). A similar approach has been used successfully by others to examine the function of endogenous molecules in immune cell signalling (Pimentel-Muinos and Seed, 1999; Suzu et al., 2000). For our studies, a cDNA fragment corresponding to the first 321 nucleotides of the coding region of mouse ptp-pest was cloned in the antisense or sense orientation in the expression vector pXM139. This portion of the ptp-pest cDNA encodes a short non-catalytic region at the N-terminus of PTP-PEST (residues 1–54) and the first 53 amino acids of the phosphatase domain. The short protein potentially encoded by the sense fragment lacks most of the catalytic domain and, thus, is not expected to possess any phosphatase activity. Figure 4.Expression of ptp-pest antisense in A20 B cells. (A) Effect of antisense ptp-pest expression on the abundance of PTP-PEST in A20 cells. A20 cells were transiently transfected with 40 μg of empty pXM139 or pXM139 bearing the first 321 nucleotides of the coding sequence of a mouse ptp-pest cDNA, in either the antisense or the sense orientation. After 48 h, equivalent numbers of viable cells were lysed and the abundance of PTP-PEST was examined by immunoblotting of total cell lysates with anti-PTP-PEST antibodies. The migration of PTP-PEST is indicated on the left. Exposure: 7 h. (B) Influence of antisense ptp-pest expression on BCR signalling. A20 cells were transfected as detailed in (A), in the presence of the IL-2 promoter–luciferase reporter construct. After 66 h, they were stimulated with F(ab′)2 fragments of SAM IgG antibodies for 6 h, and luciferase activity was measured as explained in Materials and methods. Data are shown as the percentage of maximal stimulation achieved with PMA plus ionomycin. (C) Impact of antisense syk expression on BCR signalling. As in (B), except that an antisense syk construct was also used. In keeping with the positive regulatory role of Syk in BCR signalling, a small decrease in activation of the reporter plasmid was observed in BCR-stimulated A20 cells expressing the antisense syk plasmid. Download figure Download PowerPoint A20 B cells were transiently transfected with these plasmids, and the impact on the expression levels of endogenous PTP-PEST was monitored by immunoblotting of total cell lysates with anti-PTP-PEST antibodies (Figure 4A). This analysis revealed that introduction of antisense ptp-pest (lane 3) caused a decrease (∼2.5-fold) in the abundance of PTP-PEST in A20 cells, in comparison with pXM139 alone (lane 1). This effect was observed in several different experiments. As expected, though, transfection of the sense construct (lane 2) had no influence on the abundance of PTP-PEST. Staining of the immunoblot membrane with Coomassie blue confirmed that equivalent amounts of cellular proteins were loaded in each lane (data not shown). Next, the effect of antisense ptp-pest on BCR signalling was evaluated (Figure 4B). Cells were transfected as outlined for Figure 4A, also in the presence of the IL-2 promoter reporter plasmid. After 3 days, they were stimulated or not with F(ab′)2 fragments of SAM IgG antibodies, and changes in IL-2 promoter-driven luciferase activity were monitored. We found that antisense ptp-pest, but not the sense construct, provoked an appreciable increase in BCR-induced IL-2 promoter activation (Figure 4B). This effect was seen at all concentrations of anti-BCR antibodies used; it was observed in seven independent experiments (data not shown). It is noteworthy that a small increase in IL-2 promoter activity was observed in unstimulated cells expressing antisense ptp-pest. This finding may suggest that PTP-PEST is also part of the machinery repressing the BCR-induced signalling cascade in unstimulated B cells. Note that an increase in BCR-induced lymphokine production was also observed in stable A20 transfectants expressing antisense ptp-pest (data not shown). Lastly, to ensure that the stimulatory influence of antisense ptp-pest on BCR signalling was not a non-specific consequence of antisense RNA expression, the effect of transient transfection of another antisense construct was examined (Figure 4C). When constructing this vector, we ensured that it was directed against another polypeptide expressed in A20 cells (i.e. Syk), and that it covered a sequence equivalent in length and position to that used for the ptp-pest antisense construct. This experiment showed that, unlike the antisense ptp-pest, the antisense syk plasmid (corresponding to the first 321 nucleotides of the mouse syk cDNA sequence) had no stimulatory effect on BCR-induced activation of the IL-2 promoter. Therefore, in combination, the results shown in Figures 3 and 4 provided a compelling indication that PTP-PEST is a physiological negative regulator of antigen receptor signalling, and that this inhibitory function exists both in B cells and in T cells. Selective inhibition of immunoreceptor-mediated protein tyrosine phosphorylation by PTP-PEST To clarify the mechanism(s) by which PTP-PEST caused an inhibition of lymphocyte activation, its impact on BCR-induced protein tyrosine phosphorylation was examined (Figure 5). Stable A20 transfectants were stimulated with F(ab′)2 fragments of SAM IgG antibodies, and the accumulation of phosphotyrosine-containing proteins over time was monitored by immunoblotting of total cell lysates with anti-phosphotyrosine antibodies (Figure 5A). This evaluation showed that the extent and kinetics of tyrosine phosphorylation of most BCR-regulated substrates were not significantly affected by PTP-PEST overexpression (compare lanes 6–10 with lanes 1–5). Notably, however, the BCR-induced tyrosine phosphorylation of polypeptides of ∼125 (p125), 115 (p115) and 54 (p54) kDa were markedly reduced or abolished by augmented PTP-PEST expression. A partial diminution (∼50%) in the phosphotyrosine content of a 150 kDa polypeptide (p150) was also observed. Figure 5.Impact of PTP-PEST overexpression on BCR-induced protein tyrosine phosphorylation. (A) Overall protein tyrosine phosphorylation. Representative cell lines expressing the neomycin marker alone (Neo; lanes 1–5) or in combination with wild-type PTP-PEST (lanes 6–10) were stimulated for the indicated times with F(ab′)2 fragments of SAM IgG antibodies. Changes in intracellular protein tyrosine phosphorylation were assessed by immunoblotting of total cell lysates with anti-phosphotyrosine (P.tyr) antibodies. The migrations of pre-stained molecular mass markers are shown on the right, while those of p150, p125, p115 and p54 are indicated on the left. Exposure: 15 h. (B) Tyrosine phosphorylation of specific substrates. Cells expressing the neomycin marker alone (Neo; lane 1) or in combination with wild-type PTP-PEST (PEST; lane 2) were activated for 2.5 min with F(ab′)2 fragments of SAM IgG antibodies. Specific substrates were then immunoprecipitated from cell lysates using the appropriate antibodies, and their phosphotyrosine content was determined by anti-phospho tyrosine (P.tyr) immunoblotting. Equivalent amounts of the various substrates were immunoprecipitated in control cells and in cells overexpressing PTP-PEST (data not shown). The positions of the various polypeptides are indicated on the left. Exposures: SHIP, 9 h; PLCγ2, 30 h; Cas, 21 h; Fak, 30 h; Cbl, 9 h; Pyk2, 9 h; Vav, 9 h; Blnk, 30 h; Syk, 12 h; Dok, 12 h; and Shc, 12 h. (C) Effect of PTP-PEST overexpression on association of Shc with SHIP and Grb2. Cells were activated for the specified times. The extent of association of Shc with SHIP and Grb2 was established by immunoblotting of anti-Shc immunoprecipitates with anti-SHIP (first panel) or anti-Grb2 (second panel) antibodies. The abundance of Shc was also verified by anti-Shc immunoblotting of Shc immunoprecipitates (third panel). The migrations of SHIP, Grb2 and Shc are noted on the left. Exposures: first panel, 13 h; second panel, 39 h; third panel, 2.5 h. Download figure Download PowerPoint These results implied that PTP-PEST prevented B-cell activation by inducing selective, rather than global, dephosphorylation of BCR-regulated substrates. To identify the targets of PTP-PEST more accurately, individual proteins were immunoprecipitated from activated cells and their phosphotyrosine content was measured by anti-phosphotyrosine immunoblotting (Figure 5B). This experiment demonstrated that enforced PTP-PEST expression abrogated BCR-induced tyrosine phosphorylation of four substrates: Cas (p130), Fak (p125), Pyk2 (p115) and Shc (p54). Furthermore, it abolished the binding of Shc to a 150 kDa tyrosine-phosphorylated protein representing SHIP, a 5′ inositol phosphatase (see below). Overall tyrosine phosphorylation of SHIP was also diminished by ∼50% in PTP-PEST-overexpressing cells. In contrast, PTP-PEST had little (<25% reduction) or no impact on the phosphotyrosine content of several other substrates, including phospholipase C-γ2, Cbl, Vav, Blnk/SLP-65, Syk and Dok. No tyrosine phosphorylation of paxillin was observed either in parental A20 cells or in its PTP-PEST-overexpressing variants (data not shown). Previous reports have shown that tyrosine phosphorylation of Shc in activated B cells elicits its binding to the Src homology 2 (SH2) domains of SHIP and of the adaptor molecule Grb2 (Harmer and DeFranco, 1997, 1999; Pradhan and Coggeshall, 1997; Ingham et al., 1999). The interaction with SHIP is secured further by an association between the phosphotyrosine-binding (PTB) domain of Shc and sites of tyrosine phosphorylation on SHIP (Lamkin et al., 1997; Pradhan and Coggeshall, 1997). In view of the ability of PTP-PEST to inhibit tyrosine p
Referência(s)