Portal de Periódicos da CAPES

Receptor-stimulated oxidation of SHP-2 promotes T-cell adhesion through SLP-76–ADAP

2005; Springer Nature; Volume: 24; Issue: 13 Linguagem: Inglês

10.1038/sj.emboj.7600706

1460-2075

Jaeyul Kwon, Cheng‐Kui Qu, Jin-Soo Maeng, Rustom Falahati, Chung-Hee Lee, Mark S. Williams,

Galectins and Cancer Biology

Article2 June 2005free access Receptor-stimulated oxidation of SHP-2 promotes T-cell adhesion through SLP-76–ADAP Jaeyul Kwon Jaeyul Kwon Department of Microbiology and Immunology, University of Maryland School of Medicine, Rockville, MD, USA Search for more papers by this author Cheng-Kui Qu Cheng-Kui Qu Department of Pathology, University of Maryland School of Medicine, Rockville, MD, USA Search for more papers by this author Jin-Soo Maeng Jin-Soo Maeng Laboratory of Biophysical Chemistry, National Heart Lung and Blood Institute, National Institutes of Health, Bethesda, MD, USA Search for more papers by this author Rustom Falahati Rustom Falahati Department of Immunology, George Washington University School of Medicine, Washington, DC, USA Search for more papers by this author Chunghee Lee Chunghee Lee Veritas Inc., Rockville, MD, USA Search for more papers by this author Mark S Williams Corresponding Author Mark S Williams Department of Microbiology and Immunology, University of Maryland School of Medicine, Rockville, MD, USA Search for more papers by this author Jaeyul Kwon Jaeyul Kwon Department of Microbiology and Immunology, University of Maryland School of Medicine, Rockville, MD, USA Search for more papers by this author Cheng-Kui Qu Cheng-Kui Qu Department of Pathology, University of Maryland School of Medicine, Rockville, MD, USA Search for more papers by this author Jin-Soo Maeng Jin-Soo Maeng Laboratory of Biophysical Chemistry, National Heart Lung and Blood Institute, National Institutes of Health, Bethesda, MD, USA Search for more papers by this author Rustom Falahati Rustom Falahati Department of Immunology, George Washington University School of Medicine, Washington, DC, USA Search for more papers by this author Chunghee Lee Chunghee Lee Veritas Inc., Rockville, MD, USA Search for more papers by this author Mark S Williams Corresponding Author Mark S Williams Department of Microbiology and Immunology, University of Maryland School of Medicine, Rockville, MD, USA Search for more papers by this author Author Information Jaeyul Kwon1, Cheng-Kui Qu2, Jin-Soo Maeng3, Rustom Falahati4, Chunghee Lee5 and Mark S Williams 1 1Department of Microbiology and Immunology, University of Maryland School of Medicine, Rockville, MD, USA 2Department of Pathology, University of Maryland School of Medicine, Rockville, MD, USA 3Laboratory of Biophysical Chemistry, National Heart Lung and Blood Institute, National Institutes of Health, Bethesda, MD, USA 4Department of Immunology, George Washington University School of Medicine, Washington, DC, USA 5Veritas Inc., Rockville, MD, USA *Corresponding author. Department of Microbiology and Immunology, University of Maryland School of Medicine, 15601 Crabbs Branch Way, Rockville, MD 20855, USA. Tel.: +1 301 738 0468; Fax: +1 301 517 0344; E-mail: [email protected] The EMBO Journal (2005)24:2331-2341https://doi.org/10.1038/sj.emboj.7600706 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info Receptor-stimulated generation of intracellular reactive oxygen species (ROS) modulates signal transduction, although the mechanism(s) is unclear. One potential basis is the reversible oxidation of the active site cysteine of protein tyrosine phosphatases (PTPs). Here, we show that activation of the antigen receptor of T cells (TCR), which induces production of ROS, induces transient inactivation of the SH2 domain-containing PTP, SHP-2, but not the homologous SHP-1. SHP-2 is recruited to the LAT–Gads–SLP-76 complex and directly regulates the phosphorylation of key signaling proteins Vav1 and ADAP. Furthermore, the association of ADAP with the adapter SLP-76 is regulated by SHP-2 in a redox-dependent manner. The data indicate that TCR-mediated ROS generation leads to SHP-2 oxidation, which promotes T-cell adhesion through effects on an SLP-76-dependent signaling pathway to integrin activation. Introduction Regulation of cellular functions in response to exogenous stress, for example, oxidative stress, irradiation, etc., has been characterized in evolutionarily conserved systems from microbial organisms to plants and mammals. It is now clear that cell surface receptors also utilize changes in redox balance to regulate signal transduction. Receptor stimulation with ligands as diverse as PDGF (Meng et al, 2002), insulin (Mahadev et al, 2001) or angiotensin II (Ushio-Fukai et al, 1998) induces the intracellular production of reactive oxygen species (ROS). In these studies, ROS function as requisite second messengers, regulating protein kinase activation, gene expression and/or proliferative responses. The mechanism(s) for this redox-dependent regulation of biologic responses, however, remains unclear. One redox-sensitive target that regulates signaling is the family of protein tyrosine phosphatases (PTPs), which have an oxidation-sensitive, active site cysteine (Rhee et al, 2000). Insulin-induced ROS generation leads to oxidative inactivation of the PTPs, PTP1B and TC45 (Mahadev et al, 2001; Meng et al, 2004). Both phosphatases control phosphorylation of the insulin receptor or associated proteins, and insulin-induced PTP oxidation regulates downstream signaling. Oxidative inactivation of PTPs was also induced by ROS generation associated with stimulation with EGF (PTP1B) (Lee et al, 1998) or PDGF (LM-PTP, SHP-2, PTEN) (Chiarugi et al, 2001; Meng et al, 2002; Kwon et al, 2004) and this is a pivotal step in the biological effects of these ligands. Previous studies have shown that signal transduction through the antigen receptor of T cells (TCR) can also be regulated by receptor-mediated production of ROS (Jackson et al, 2004). In mature T cells, TCR crosslinking induces rapid (within 2–4 min) ROS generation and the data indicate that TCR-stimulated ROS production regulates activation of MAPK, cytokine secretion and gene expression (Devadas et al, 2002; Kwon et al, 2003). In particular, TCR-induced production of hydrogen peroxide selectively inhibits activation of the MEK/Erk pathway (Kwon et al, 2003). However, the mechanisms by which ROS regulate TCR signaling are still poorly understood. Exposure to exogenous oxidants leads to oxidative inactivation of PTPs in T cells (O'Shea et al, 1992), but there is no evidence to date on the effects of TCR-induced ROS production on PTP function. TCR signaling depends upon coordinated interactions of multiple signaling pathways, including PTPs, protein kinases and adapter proteins (e.g., LAT, SLP-76, Grb2, Gads) that 'nucleate' key proteins (Jordan et al, 2003; Mustelin et al, 2003). The membrane-localized adapter LAT (linker for activation of T cells), upon tyrosine phosphorylation, recruits a number of proteins via their SH2 domains (e.g., Grb2, Gads, PLCγ1) (Wange, 2000). Gads (Grb2-like adapter downstream of Shc) is an adapter protein that, via one of its SH3 domains, interacts with the proline-rich domain of SLP-76 and recruits it to LAT and the membrane (Liu et al, 2001). In addition to the proline-rich domain, SLP-76 (SH2-containing leukocyte-specific protein of 76 kDa) has multiple functional interaction domains, including phosphorylated tyrosines and an SH2 domain (Myung et al, 2001) that bind proteins, including Vav1, Nck, Itk and ADAP (Myung et al, 2001). ADAP (adhesion- and degranulation-promoting adaptor protein), also known as Fyn SH2-binding protein (Fyb) or SLAP-130, binds the SH2 domain of SLP-76 upon ADAP phosphorylation (Peterson, 2003). ADAP is necessary for optimal T-cell responses and the inside-out activation of integrins (Griffiths et al, 2001; Peterson et al, 2001). Thus, the adapters Gads, SLP-76 and ADAP are critical regulators of T-cell development and TCR signaling pathways. The goal of the current report was to investigate the molecular basis for redox control of TCR signal transduction by intracellular production of ROS. In the current study, we show that SHP-2 is oxidized by ROS produced upon TCR stimulation and that the active site cysteine is the primary site of oxidation. We further identify a site of action for SHP-2 in dephosphorylation of Vav1 and ADAP associated with SLP-76, although it does not dephosphorylate SLP-76 itself. The data support the hypothesis that SHP-2 exerts a negative role in dephosphorylation of ADAP and Vav1, which limits TCR-induced adhesion and LFA-1 clustering. TCR-induced ROS production selectively inhibits SHP-2-mediated dephosphorylation of Vav1 and ADAP associated with SLP-76, leading to redox-dependent changes in TCR-stimulated adhesion and integrin clustering. Results TCR-induced oxidation of SHP-2 Our previous data have demonstrated that TCR-induced production of hydrogen peroxide selectively inhibited activation of the MEK/Erk kinase pathway (Kwon et al, 2003; Jackson et al, 2004). SHP-2 has been shown to promote ERK activation in T cells (Frearson and Alexander, 1998). Therefore, we analyzed TCR-induced oxidation of SHP-2. A biotinylated thiol reactive probe was used to label reduced/nonoxidized thiols in whole-cell lysates under anaerobic conditions. The labeling conditions lead to selective alkylation of reactive thiols, such as active site cysteines of PTPs (Kim et al, 2000). Loss of biotin incorporation into immunoprecipitated SHP-2 indicates oxidation of reactive thiols (Figure 1A). Decreased biotin labeling in SHP-2 was detected upon TCR crosslinking in Jurkat T cells (Figure 1B) and primary murine T-cell blasts (Figure 1D), which occurs coincident with the kinetics of TCR-stimulated ROS generation (Kwon et al, 2003; Jackson et al, 2004). The loss was transient, however, indicating that TCR-stimulated thiol oxidation was reversible. Figure 1.TCR stimulation induces oxidative modification of SHP-2. (A) Reduced thiols (PTP-S−) are labeled with PEO-iodoacetyl biotin, but oxidized thiols (PTP-S-OH; sulfenic acid) are not. (B, C) After anti-CD3 stimulation for the indicated times, Jurkat T cells were lysed and labeled with PEO-iodoacetyl biotin as described in Materials and methods. SHP-2 (B) or SHP-1 (C) was immunoprecipitated and biotinylation detected with HRP-conjugated streptavidin and ECL. Equal application of protein was confirmed by immunoblot analysis. The level of biotin incorporation (by densitometry) was corrected for total protein levels and the signal at time=0 was normalized to 100%. The graphs represent the average of three separate experiments±s.e.m. **Significantly different from unstimulated controls; P<0.01. (D) Mouse T-cell blasts were stimulated for the indicated times by anti-CD3 crosslinking. SHP-2 oxidation was measured as in (B). (E) Jurkat T cells were transiently transfected with an expression vector encoding (His)6-SHP-2(C/S) and cells were left unstimulated. Incorporation of biotin into endogenous or tagged SHP-2 was measured as in (B). Download figure Download PowerPoint To test the selectivity of TCR-induced ROS generation in PTP oxidation, oxidation of SHP-1 was measured. SHP-1 also has tandem SH2 domains, exhibits similar substrate specificity as SHP-2 and has been proposed to affect TCR signal transduction (reviewed in Neel et al, 2003). Under conditions in which SHP-2 oxidation was observed, little or no oxidation of SHP-1 was detected (Figure 1C). Although the labeling procedure is designed to detect oxidation of the active site cysteine, iodoacetamide could react with other free thiols in SHP-2. Therefore, Jurkat T cells were transfected to express (His)6-tagged SHP-2 in which the active site cysteine was mutated to serine (SHP-2(C/S)). Due to the presence of the (His)6 tag, the ectopically expressed SHP-2 has a slower mobility in gels and can be distinguished from endogenous SHP-2. In unstimulated cells, biotin labeling of endogenous SHP-2 was readily detected, but there was little incorporation of the label into the SHP-2(C/S) protein (Figure 1E). The data suggest that the majority of the iodoacetamide labeling of SHP-2, under these conditions, was at the active site. Thus, the transient loss of iodoacetamide labeling in SHP-2 upon TCR stimulation indicates reversible oxidation of the active site cysteine of SHP-2. SHP-2 phosphatase activity was also inhibited under conditions in which thiol oxidation was detected. Using a colorimetric assay (Dechert et al, 1994), phosphatase activity in SHP-2 immunoprecipitates was decreased upon TCR stimulation in both Jurkat T cells (16.72 to 11.1 pmol/min/106 cells) and mouse T-cell blasts (14.35 to 10.72 pmol/min/106 cells). The labeling procedure above measures a loss of reactivity in protein thiols under nondenaturing conditions. The results could be affected by accessibility of thiols, and interpretation of the data relies upon the ability to detect decreased biotin incorporation. Therefore, SHP-2 oxidation was assayed using a method in which reduced thiols were blocked under denaturing conditions and the reversibly oxidized thiols were re-reduced and labeled with iodoacetamide-biotin (Kwon et al, 2004). This 'positive labeling' of oxidized thiols will show a gain of signal upon thiol oxidation induced by TCR-stimulated ROS generation. Consistent with the results in Figure 1, TCR crosslinking of human T blasts induced a transient oxidation of SHP-2 (Figure 2A). The level of biotin incorporation was similar to that observed upon treatment of cells with 100 μM exogenous hydrogen peroxide. Under the same conditions, neither TCR stimulation nor treatment with 100 μM hydrogen peroxide induced SHP-1 oxidation. Exposure to 1 mM H2O2, however, did lead to detectable oxidation of SHP-1 (Figure 2B). TCR stimulation in Jurkat T cells also led to marked oxidation of SHP-2 (Figure 2C), but not SHP-1 (Figure 2D). Thus, TCR-stimulated production of ROS selectively induces oxidation of SHP-2 in T cells. Figure 2.SHP-2 oxidation measured by 'positive labeling'. Human T blasts (A, B) or Jurkat T cells (C, D) were stimulated by anti-CD3 crosslinking for the indicated times. Oxidized thiols were labeled with PEO-iodoacetyl biotin in a 'positive' labeling method as described in Materials and methods. Biotinylated proteins were isolated on Neutravidin-Sepharose, separated by SDS–PAGE and immunoblotted for SHP-2 (A, C) or SHP-1 (B, D). An aliquot of the lysates was reserved to blot for total SHP-2 or SHP-1. (E, F) Antioxidants inhibit SHP-2 oxidation. (E) Mouse T-cell blasts were preincubated in the absence (−) or presence (+) of 20 mM NAC for 60 min. (F) Jurkat T cells were transfected with an empty vector (−) or one encoding Prx II (+) and prepared as described in Materials and methods. Cells were stimulated by anti-CD3 crosslinking for the indicated times and SHP-2 oxidation was detected as in Figure 1. The level of biotin incorporation (by densitometry) was corrected for total protein levels and the signal at time=0 was normalized to 100%. The graphs represent the average of three separate experiments±s.e.m. **Significantly different from unstimulated controls; P<0.01. Download figure Download PowerPoint The two labeling approaches suggested that SHP-2 was more sensitive to intracellular ROS than SHP-1. This was directly tested by exposing Jurkat cells to graded concentrations of hydrogen peroxide and measuring oxidation of both SHP-1 and SHP-2 by the loss of biotin-iodoacetamide incorporation as in Figure 1 (Supplementary Figure 1). Oxidation of both SHP-1 and SHP-2 is detected, but at least two- to three-fold more hydrogen peroxide is required to detect oxidation of SHP-1. Antioxidants and antioxidant enzymes were used to indicate a causal role for TCR-induced ROS generation in SHP-2 oxidation. Coincubation of mouse T cells with the antioxidant N-acetyl-cysteine (NAC) inhibited the loss of SHP-2 labeling, suggesting that ROS produced upon TCR stimulation were responsible for the loss of labeling (Figure 2E). The presence of NAC also produced a modest increase in biotin labeling of SHP-2 in unstimulated cells, suggesting a basal level of SHP-2 oxidation in cells. To further establish a role for hydrogen peroxide, the antioxidant enzyme peroxiredoxin II (Prx II) was overexpressed in Jurkat T cells. Previous studies have used overexpression of Prx II to eliminate hydrogen peroxide produced upon TCR stimulation (Devadas et al, 2002). As opposed to that observed in vector-transfected cells, TCR stimulation of cells overexpressing Prx II did not show decreased incorporation of biotin into SHP-2 (Figure 2F). The data indicate that TCR-induced production of hydrogen peroxide leads to transient oxidation of SHP-2. Expression of inactive SHP-2 in T cells The active site mutant of SHP-2 (SHP-2(C/S)) was expressed in cells to mimic the conditions in which SHP-2 was oxidatively inactivated. Expression of such mutated PTPs has also been used to identify potential phosphatase substrates (Neel and Tonks, 1997). Overexpression of wild-type SHP-2 (SHP-2 WT) was used to control for the effects of increased SHP-2 protein and the possible effects of increased SHP-2 activity in cells. Anti-HA immunoprecipitation from anti-CD3-stimulated cells expressing HA-tagged SHP-2(C/S) showed increased association of multiple phosphoproteins (Figure 3A). The literature suggests that some of these SHP-2-associated proteins may be LAT (Frearson and Alexander, 1998), Vav1 (Wakino et al, 2004) and/or Gab2 (Yamasaki et al, 2001). The presence of LAT and the adapter protein ADAP in the anti-HA immunoprecipitates was confirmed by Western blot (Figure 3A). The current results and previous reports support an interaction of SHP-2 with LAT and the multiprotein complex formed around LAT. Therefore, to investigate the potential implications of SHP-2 oxidation, the effect of SHP-2(C/S) expression was examined on TCR signaling in and around the complexes dependent upon LAT and SLP-76. Figure 3.Association of proteins with SHP-2(C/S) upon TCR stimulation. (A) Jurkat T cells were transiently transfected with an empty vector (Vector) or an expression vector encoding HA-SHP-2(C/S) or HA-SHP-2 WT. Cells were stimulated by anti-CD3 crosslinking for 5 min and lysed. Anti-HA immunoprecipitates were blotted with anti-phosphotyrosine and the membrane was stripped and reprobed with antibodies to the indicated proteins. (B–F) Effect of SHP-2(C/S) expression on association of signaling molecules with LAT and Grb2. Jurkat T cells were transfected and stimulated as in (A). (C) Levels of SHP-2 in total cell lysates were determined by immunoblot analysis. (B, D) Anti-LAT or (E, F) anti-Grb2 immunoprecipitates were blotted with (B, E) anti-phosphotyrosine and (D, F) the membranes were stripped and reprobed for the indicated proteins. Download figure Download PowerPoint Immunoprecipitation of LAT from Jurkat cells transfected to express SHP-2(C/S) and SHP-2 WT showed minor changes in associated phosphoproteins upon stimulation with anti-CD3 (Figure 3B). In these studies, the level of SHP-2(C/S) or ectopically expressed SHP-2 WT was titered to avoid overwhelming the system (Figure 3C). There was a modest increase in signals from phosphoproteins comigrating with SLP-76, Vav1 and ADAP in cells expressing SHP-2(C/S). There were no apparent changes in phosphorylation of LAT or ZAP-70, suggesting a selective effect. Immunoblots suggested that the minor increases in the phosphotyrosine signal in Vav1 were not mirrored by changes in Vav1 protein associated with LAT, but SLP-76 protein levels in the immunoprecipitates did show similar increases as in the phosphotyrosine blot (Figure 3D). In general, expression of wild-type SHP-2 did not markedly change either phosphorylation or levels of proteins co-immunoprecipitating with LAT. Grb2 and Gads are the major adapter proteins that bind LAT. Grb2 immunoprecipitates showed minor differences in the association of tyrosine phosphorylated proteins in cells expressing either wild-type or mutant SHP-2 (Figure 3E and F). There was an increased intensity in the phosphotyrosine blot for a band that comigrated with Vav1, although there was no change in Vav1 protein in the immunoprecipitates by immunoblot. Thus, expression of the inactive SHP-2(C/S) induced small changes in phosphorylation and association of proteins with LAT, and the effect(s) does not appear to be centered upon proteins known to be associated with Grb2. Effect of SHP-2(C/S) on Gads- and SLP-76-associated proteins Analysis of phosphoproteins that co-immunoprecipitated with Gads from TCR-stimulated cells expressing SHP-2(C/S) extended the results observed in LAT precipitates. There was an increase (approximately two-fold by densitometry) in tyrosine phosphorylation of bands corresponding to ADAP and Vav1 as compared to cells expressing SHP-2 WT or vector (Figure 4A and Supplementary Figure 2A). In both cases, overexpression of SHP-2 WT led to small decreases in phosphorylation of the proteins as compared to vector control cells. Immunoblot for total ADAP protein showed commensurate changes in ADAP protein (Figure 4A). The data are consistent with the model of SHP-2 regulating binding of phosphorylated ADAP to SLP-76. The changes in Vav1 phosphorylation, however, were not reflected by increased Vav1 protein associated with Gads (Figure 4A). These results were consistent with the model of the SH2 domain of Vav1 binding to phosphorylated SLP-76, and SHP-2 acting to dephosphorylate Vav1. Identical results were observed in SLP-76 immunoprecipitates (Figure 4B and Supplementary Figure 2B). The effects of SHP-2 were selective in that total protein and tyrosine phosphorylation of both SLP-76 and LAT and association of other proteins including PLCγ1 were largely unchanged by expression of either active site mutant or wild-type SHP-2. These results suggested that SHP-2 dephosphorylated proteins associated with SLP-76, but did not target SLP-76 itself. The data further suggest that Vav1 and ADAP are SHP-2 substrates and SHP-2 regulates their activation and/or association with SLP-76. Figure 4.Effect of SHP-2(C/S) on phosphorylation and association of proteins with Gads and SLP-76. Jurkat T cells were transiently transfected with an empty vector (Vector) or an expression vector encoding SHP-2(C/S) or SHP-2 WT. Cells were stimulated for 5 min by anti-CD3 crosslinking. Lysates were immunoprecipitated with (A) anti-Gads, (B) anti-SLP-76, (C) anti-Vav1 or (D) anti-ADAP antibody. Immunoprecipitates were separated by SDS–PAGE, immunoblotted for phosphotyrosine and then stripped and reprobed for the indicated proteins. (C) Anti-Vav1 immunoprecipitates were also blotted with a phosphospecific antibody to Vav1 phosphorylated at Y174. Download figure Download PowerPoint The consistent effect on Vav1 phosphorylation led us to measure directly changes in Vav1 phosphorylation in Vav1 immunoprecipitates (Figure 4C). Vav1 has multiple possible phosphorylation sites, and SHP-2(C/S) expression induced a modest increase in total Vav1 phosphorylation. When analyzed with a phosphospecific antibody to tyrosine 174 (Vav1 Y174), however, the presence of SHP-2(C/S) showed an increase (greater than 50% over vector control) in Vav1 phosphorylation. Thus, SHP-2 activity seems to have a selective effect on the phosphorylation of Vav1 at tyrosine 174, which is critical in activation of Vav1 GEF function (Turner and Billadeau, 2002). Direct immunoprecipitation of ADAP supported the observation that SHP-2(C/S) expression led to increased ADAP phosphorylation (Figure 4D). These data further substantiate the model of SHP-2 regulating ADAP–SLP-76 association. Immunoprecipitation of Gads from anti-CD3-stimulated SLP-76-deficient Jurkat cells (J.14) did not show evidence of Vav1 and low levels of ADAP in Gads immunoprecipitates (Supplementary Figure 2C). Co-immunoprecipitation of SHP-2 with Gads, on the other hand, was largely unchanged in J.14 cells. The SH2 domains of SHP-2 and SHP-1 show selective recognition, and are generally not interchangeable (Neel et al, 2003). Nevertheless, expression of high levels of SHP-2(C/S) might alter SHP-1 function. To test the potential role of SHP-1 in TCR-mediated changes in phosphorylation of proteins in the Gads–SLP-76 complex, SHP-1(C/S) was expressed side by side with SHP-2(C/S) in Jurkat cells (Supplementary Figure 3). In cells expressing SHP-1(C/S), anti-SLP-76 immunoprecipitates did not show the increase in phosphorylated Vav1 or phospho-ADAP that was observed in cells expressing SHP-2(C/S). In contrast, there was decreased phosphorylation of bands corresponding to ADAP, LAT and Vav1. Therefore, the data support a model in which SHP-2 associates with a Gads–SLP-76 complex but does not target SLP-76 itself as a substrate. The data suggest that SHP-2 controls phosphorylation of Vav1 bound to SLP-76 and regulates the phosphorylation of ADAP and therefore its association with SLP-76. Role of ROS in protein association with Gads–SLP-76 SHP-2(C/S) was expressed to mimic oxidatively inactivated SHP-2 in cells where ROS generation was occurring. Quenching TCR-induced ROS production with antioxidants, which inhibited SHP-2 oxidation (Figure 2), should show the opposite effect produced by SHP-2(C/S) expression. Thus, the effects of Prx II overexpression in TCR-stimulated Jurkat T cells were analyzed in immunoprecipitates of both Gads and SLP-76 (Figure 5). As was suggested by our previous findings (Kwon et al, 2003), overall anti-CD3-induced tyrosine phosphorylation was not markedly affected by Prx II overexpression as measured in total cell lysates (Supplementary Figure 4A). In Gads (Figure 5A) and SLP-76 (Figure 5B) immunoprecipitates, anti-CD3-induced ADAP phosphorylation and association were decreased nearly two-fold in cells expressing Prx II as compared to vector control. Direct immunoprecipitation of ADAP confirmed these observations, as Prx II overexpression inhibited TCR-stimulated phosphorylation of total ADAP (Figure 5C). As above, these effects were selective since phosphorylation of SLP-76 and association of proteins such as PLCγ1 were not affected by expression of Prx II. In contrast, the results with Vav1 in Prx II-overexpressing cells were not the converse of those with SHP-2(C/S). Vav1 phosphorylation and Vav1 protein levels in Gads immunoprecipitates showed little to no change as compared to vector control (Figure 5A). Thus, redox regulation of Vav1 phosphorylation in Gads–SLP-76 complexes appears to be multifaceted. The data clearly show that tyrosine phosphorylation of ADAP and its association with SLP-76 are influenced by ROS, potentially through oxidation of SHP-2. Figure 5.Effect of Prx II on association of proteins with Gads and SLP-76. Jurkat T cells were transfected with an empty vector or an expression vector encoding Prx II, and cells were stimulated by anti-CD3 crosslinking. Lysates were immunoprecipitated with (A) anti-Gads, (B) anti-SLP-76 or (C) anti-ADAP antibody. Immunoprecipitates were separated by SDS–PAGE, immunoblotted for phosphotyrosine and were stripped and reprobed for the indicated proteins. (A) Total cell lysates were probed for Prx II to measure overexpression and β-tubulin to control for loading. (D) Effects of coexpression of SHP-2(C/S) and Prx II. Jurkat T cells were transfected with an empty vector or an expression vector encoding SHP-2(C/S) alone or in combination with one encoding Prx II (SHP-2(C/S)+Prx II) and the cells were stimulated by anti-CD3 crosslinking. Lysates were immunoprecipitated with anti-SLP-76 antibody and probed with antibodies to the indicated proteins. Total lysates were probed for SHP-2 and Prx II to measure expression levels and β-tubulin was probed to control loading. Download figure Download PowerPoint The effects of Prx II (Figure 5A) or SHP-2(C/S) (Figure 4A) on SLP-76–ADAP association might result from redox regulation of SHP-2 localized with the Gads–SLP-76 complex. Cotransfection to express both SHP-2(C/S) and Prx II showed an intermediate effect as compared to either protein alone. Levels of ADAP protein co-immunoprecipitating with SLP-76 were nearly 2.5-fold less in cells coexpressing Prx II and SHP-2(C/S) as compared to that observed in cells expressing SHP-2(C/S) alone (Figure 5D). The formation of an SLP-76–ADAP complex in cells expressing both proteins was still almost twice as much as in vector control cells as compared to the effects of Prx II alone, which was two-fold less than vector control (Figure 5A). Similarly, levels of SHP-2 in SLP-76 immunoprecipitates from cells expressing both SHP-2(C/S) and Prx II were half that in cells expressing the mutant SHP-2 alone, although the levels were still considerably higher than that observed in vector control. The data suggest that localized changes in redox balance control the availability of active SHP-2 to the Gads–SLP-76 signaling complex. Effects of ROS and SHP-2 on adhesion ADAP association with SLP-76 has been proposed to regulate TCR-induced adhesion of T cells through inside-out signaling to β-integrins (Griffiths et al, 2001; Peterson et al, 2001). Using adhesion to fibronectin-coated plates, SHP-2(C/S) expression significantly increased TCR-stimulated adhesion (Figure 6A). In contrast, overexpression of SHP-2 WT or Prx II, both of which lead to increased functional SHP-2 in TCR-stimulated cells, significantly inhibited adhesion. Figure 6.Effect of SHP-2 or Prx II on TCR-stimulated adhesion to fibronectin and LFA-1 clustering. (A) Jurkat T cells were transiently transfected with a β-gal expression plasmid in the presence of an empty vector (Vector) or an expression vector encoding SHP-2(C/S), SHP-2 WT or Prx II. Adhesion of anti-CD3-stimulated cells to fibronectin-coated wells was performed as describe