Portal de Periódicos da CAPES

Molecular Events Associated with CD4-mediated Down-regulation of LFA-1-dependent Adhesion

2002; Elsevier BV; Volume: 277; Issue: 2 Linguagem: Inglês

10.1074/jbc.m110064200

1083-351X

Fabienne Mazerolles, Christiane Barbat, Maÿlis Trucy, Waldemar Kolanus, Alain Fischer,

Monoclonal and Polyclonal Antibodies Research

We have previously shown that CD4 ligand binding inhibits LFA-1-dependent adhesion between CD4+ T cells and B cells in a p56lck- and phosphatidylinositol 3-kinase (PI3-kinase)-dependent manner. In this work, downstream events associated with adhesion inhibition have been investigated. By using HUT78 T cell lines, CD4 ligands were shown to induce a dissociation of LFA-1 from cytohesin, a cytoplasmic protein known to bind LFA-1 and to enhance the affinity/avidity of LFA-1 for its ligand ICAM-1. A dissociation of PI3-kinase from cytohesin is also observed. In parallel, we have found that CD4 ligand binding induced a redistribution of PI3-kinase and of the tyrosine phosphatase SHP-2 to the membrane and induced a transient formation of protein interactions including PI3-kinase; an adaptor protein, Gab2; SHP-2; and a SH2 domain-containing inositol phosphatase, SHIP. By using antisense oligonucleotides or transfection of transdominant mutants, down-regulation of adhesion was shown to require the Gab2/PI3-kinase association and the expression of SHIP and SHP-2. We therefore propose that CD4 ligands, by inducing these molecular associations, lead to sustained local high levels of D-3 phospholipids and possibly regulate the cytohesin/LFA-1 association. We have previously shown that CD4 ligand binding inhibits LFA-1-dependent adhesion between CD4+ T cells and B cells in a p56lck- and phosphatidylinositol 3-kinase (PI3-kinase)-dependent manner. In this work, downstream events associated with adhesion inhibition have been investigated. By using HUT78 T cell lines, CD4 ligands were shown to induce a dissociation of LFA-1 from cytohesin, a cytoplasmic protein known to bind LFA-1 and to enhance the affinity/avidity of LFA-1 for its ligand ICAM-1. A dissociation of PI3-kinase from cytohesin is also observed. In parallel, we have found that CD4 ligand binding induced a redistribution of PI3-kinase and of the tyrosine phosphatase SHP-2 to the membrane and induced a transient formation of protein interactions including PI3-kinase; an adaptor protein, Gab2; SHP-2; and a SH2 domain-containing inositol phosphatase, SHIP. By using antisense oligonucleotides or transfection of transdominant mutants, down-regulation of adhesion was shown to require the Gab2/PI3-kinase association and the expression of SHIP and SHP-2. We therefore propose that CD4 ligands, by inducing these molecular associations, lead to sustained local high levels of D-3 phospholipids and possibly regulate the cytohesin/LFA-1 association. phosphatidylinositol 3-kinase Src homology 2 Src homology 3 SH2-containing phosphotyrosine phosphatase SH2 domain-containing inositol phosphatase pleckstrin homology phosphatidylinositol phosphatidylinositol 3,4,5-trisphosphate T cell receptor antibody high pressure liquid chromatography human histocompatibility leukocyte antigen Leukocyte adhesion is a fundamental process in leukocyte physiology, which is strictly regulated and involves a number of interactions between different adhesion proteins (1Gahmberg C.G. Tolvanen M. Kotovuori P. Eur. J. Biochem. 1997; 245: 215-232Google Scholar). As shown in blocking experiments with specific monoclonal Abs and transfection experiments (2Springer T.A. Nature. 1990; 346: 425-434Google Scholar), these adhesion processes are required for T cell activation and effector functions. We have previously shown that the LFA-1-dependent adhesion between CD4+ T cells (resting or T cell lines) and B cells was down-regulated by CD4 ligands. This process requires the activities of the tyrosine kinase p56lck associated with CD4 (3Mazerolles F. Barbat C. Meloche S. Graton S. Soula M. Fagard R. Fischer S. Hivroz C. Bernier J. Sekaly R.P. Fischer A. J. Immunol. 1994; 152: 5670-5679Google Scholar) and of phosphatidylinositol 3-kinase (PI3-kinase)1 associated with the CD4·p56lck complex (4Mazerolles F. Barbat C. Hivroz C. Fischer A. J. Immunol. 1996; 157: 4844-4854Google Scholar). However, the molecular events and the relationship between these kinases required for the down-regulation induced by CD4 binding and the modification of the affinity/avidity of LFA-1 are not clearly understood. PI3-kinase is an intracellular enzyme mainly consisting of two subunits: the p110 catalytic subunit (α and β isoforms) and the p85 regulatory subunit (α and β isoforms) (5Escobedo J.A. Navankasattusas S. Kavanaugh W.M. Milfay D. Fried A.V. Williams L.T. Cell. 1991; 65: 75-82Google Scholar, 6Hiles I.D. Otsu M. Volinia S. Fry M.J. Gout I. Dhand R. Panayatou G. Ruiz-Larrea F. Thompson A. Totty N.F. Hsuan S.A. Courtneidge S. Parker P.J. Waterfield M.J. Cell. 1992; 70: 419-429Google Scholar). This enzyme phosphorylates the d-3 position of the inositol ring of phosphatidylinositol (PI), phosphatidylinositol 4-phosphate (PI 4-P), and phosphatidylinositol 4,5-bisphosphate (PI 4,5-P2) (7Whitman M. Downes C.P. Keeler M. Keller T. Cantley L. Nature. 1988; 332: 644-646Google Scholar). These phospholipids are one of the intermediary messengers for the CD4-mediated down-regulation of LFA-1-dependent adhesion since inhibitors of PI3-kinase activity abrogate the regulatory event (8Mazerolles F. Barbat C. Fischer A. Eur. J. Immunol. 1997; 27: 2457-2465Google Scholar). The mechanism by which PI3-kinase upon CD4 ligand binding mediates down-regulation of LFA-1 adhesion is unknown. PI3-kinase can exert an opposite effect: Kolanus et al. (9Kolanus W. Nagel W. Schiller B. Zeitlmann L. Godar S. Stockinger H. Seed B. Cell. 1996; 86: 233-242Google Scholar) have recently shown that thed-3 phospholipids synthesized by PI3-kinase increase the affinity of LFA-1 for its ligand ICAM-1 by a modification of association between the β2 cytoplasmic domain of LFA-1 and a cytoplasmic protein, cytohesin-1. Cytohesin-1 is a 47-kDa intracellular protein that interacts specifically with the cytoplasmic domain of the leukocyte integrin αLβ2. The overexpression of full-length cytohesin-1 resulted in a constitutive adhesion of αLβ2 (9Kolanus W. Nagel W. Schiller B. Zeitlmann L. Godar S. Stockinger H. Seed B. Cell. 1996; 86: 233-242Google Scholar), and PI3-kinase activates the β2 integrin adhesion pathway at least partially through cytohesin-1. Phosphatidylinositol 3,4,5-trisphosphate (PIP3) could directly recruit cytohesin with its pleckstrin homology (PH) domain and could contribute to the localization of this protein to the plasma membrane in close relationship with LFA-1 at cell interaction sites. Therefore PI3-kinase seems to play a complex role in regulating cell adhesion. However, the regulation of LFA-1 adhesion could also involve other signaling molecules. The precise role of tyrosine kinases and tyrosine phosphatases in adhesion complex formation remains unclear. A key mechanism could involve the binding of SH2 domains present in many signaling molecules and some tyrosine-phosphorylated cytoskeleton molecules. We have previously shown that inhibition of tyrosine kinases with herbimycin A neutralized the down-regulation of LFA-1 adhesion induced by CD4 ligands suggesting that tyrosine kinases are part of this regulatory pathway. Furthermore, p56lck and PI3-kinase (p110 catalytic subunit), which are involved in this mechanism, are transiently tyrosine-phosphorylated after CD4 binding with kinetics similar to that of the down-regulation of adhesion. The SH2-containing phosphotyrosine phosphatase (SHP-2) has been proposed to act as a negative regulator of T cell signaling based on its association with CTLA-4 (10Marengere L. Waterhouse P. Duncan G. Mittrucker H. Feng G. Mak T. Science. 1996; 272: 1170-1173Google Scholar). This intracellular phosphotyrosine phosphatase is characterized by the presence of tandem SH2 domains at its N terminus followed by a single catalytic domain and unique region (11Tonks N. Neel B. Cell. 1996; 87: 365-373Google Scholar). p56lck and PI3-kinase can be dephosphorylated by SHP-2 as suggested in B cells after BCR-FcγRIIβ coclustering (12Sarmay G. Koncz G. Pecht I. Gergely J. Immunol. Lett. 1999; 68: 25-34Google Scholar). Another SH2 containing protein could be involved in the adhesion regulatory process induced by CD4 ligands, i.e.the SH2 domain-containing inositol phosphatase (SHIP) that dephosphorylates the PIP3 at position 5, thereby linking its activity to PI3-kinase activity. SHIP is a cytosolic protein composed of a single SH2 domain at its N terminus, a catalytic domain, two phosphotyrosine binding domain consensus sequences, and several putative SH3-interacting motifs at the C terminus for interacting with many different proteins. (13Lioubin M. Algate P. Tsai S. Carlberg K. Acbersold A. Rohrschneider L. Genes Dev. 1996; 101084Google Scholar). A negative role for SHIP signaling has been described in B cells (14Chacko G.W. Tridandapani S. Damen J. Liu L. Krystal G. Coggeshall K. J. Immunol. 1996; 157: 2234-2238Google Scholar, 15Ono M. Bolland S. Tempst P. Ravetch J. Nature. 1996; 383: 263-266Google Scholar). It has been demonstrated that during FcγRIIβ1-mediated inhibition of B cell receptor signaling, SHIP recruits the p85 subunit of PI3-kinase (16Gupta N. J. Biol. Chem. 1999; 274: 7489-7494Google Scholar). A p85 SH2 domain recognition is present in SHIP. In addition, SHIP has been demonstrated to be a target for protein tyrosine kinase activated in response to multiple cytokines as well as TCR engagement (17Liu L. Damen J.E. Ware M. Hugues M. Krystal G. Leukemia. 1997; 11: 181-184Google Scholar). SHIP was also shown to be able to precipitate SHP-2 after cytokine stimulation (18Liu L. Damen J.E. Ware M.D. Krystal G. J. Biol. Chem. 1997; 272: 10998-11001Google Scholar). PI3-kinase and SHP-2 have been also shown to associate to Gab2, a member of a subfamily of scaffolding adaptors that includesDrosophila Dos and mammalian Gab1 (19Nishida K. Yoshida Y. Itoh M. Fukada T. Ohtani T. Shirogane T. Atsumi T. Takahashi-Tezuka M. Ishihara K. Hibi M. Hirano T. Blood. 1999; 93: 1809-1816Google Scholar, 20Gu H. Pratt J.C. Burakoff S.J. Neel B.G. Mol. Cell. 1998; 2: 729Google Scholar, 21Gadina M. Sudarshan C. Visconti R. Zhou Y.J. Gu H. Neel B.G. O'Shea J.J. J. Biol. Chem. 2000; 275: 26959-26966Google Scholar). These proteins have an N-terminal PH domain followed by multiple potential tyrosine phosphorylation sites and several proline-rich sequences. Initial characterization of Gab2 suggested that it functions in a variety of signaling pathways, including those emanating from receptor tyrosine kinases, cytokine receptors, and antigen receptors (20Gu H. Pratt J.C. Burakoff S.J. Neel B.G. Mol. Cell. 1998; 2: 729Google Scholar, 22Wickrema A. Uddin S. Sharma A. Chen F. Alsayed Y. Ahmad S. Sawyer S.T. Krystal G. Yi T. Nishada K. Hibi M. Hirano T. Platanias L.C. J. Biol. Chem. 1999; 274: 24469-24474Google Scholar). It has been observed that Gab2, in response to stimulation by different cytokines, becomes rapidly tyrosyl-phosphorylated and associates to SHP-2 and PI3-kinase. Indeed Gab2 expresses three YXXM motifs that constitute potential sites for interaction with p85-PI3-kinase. Mutations in these motifs abrogate PI3-kinase binding (23Pratt J.C. Igras V.E. Maeda H. Baksh S. Gelfand E.W. Burakoff S.J. Neel B.G. Gu H. J. Immunol. 2000; 165: 4158-4163Google Scholar). Gab2, by association with SHP-2 or PI3-kinase, either plays a positive signaling role (19Nishida K. Yoshida Y. Itoh M. Fukada T. Ohtani T. Shirogane T. Atsumi T. Takahashi-Tezuka M. Ishihara K. Hibi M. Hirano T. Blood. 1999; 93: 1809-1816Google Scholar, 20Gu H. Pratt J.C. Burakoff S.J. Neel B.G. Mol. Cell. 1998; 2: 729Google Scholar) or inhibits TCR signaling. This negative signal requires Gab2/PI3-kinase but not Gab2/SHP-2 interaction (23Pratt J.C. Igras V.E. Maeda H. Baksh S. Gelfand E.W. Burakoff S.J. Neel B.G. Gu H. J. Immunol. 2000; 165: 4158-4163Google Scholar). Potential interactions between PI3-kinase and LFA-1/cytohesin and the formation of protein association, including SHP-2, SHIP, or Gab2, induced by CD4 binding were therefore investigated. The following Abs were used: 13B8.2 and 25.3 (IgG1; anti-CD4 and anti-LFA-1α monoclonal Abs, respectively; Immunotech, Marseille, France), anti-SHP-2 Ab (Transduction Laboratories), anti-SHIP Ab (Santa Cruz Biotechnology, TEBU S.A., Le Perray-en-Yvelines, France), and anti-p85-PI3-kinase polyclonal Ab and anti-Gab2 polyclonal Ab (Upstate Biotechnology, Inc., Lake Placid, NY); anti-cytohesin polyclonal Ab was a kind gift of Dr. W. Kolanus. Tetramethylrhodamine B isothiocyanate-conjugated anti-mouse IgG and fluorescein isothiocyanate-conjugated sheep anti-rabbit IgG were from Jackson Immunoresearch Laboratories, Inc. The DR134–148 peptide (NGQEEKAGVVSTGLI) is analogous to residues 134–148 of the HLA class II β2 domain, and the control peptide V15A (VQKSIENVTGLGEGA) is a random sequence analogous to DR134–148 (24Cammarota G. Scheirle A. Takacs B. Doran D.M. Knorr R. Bannwarth W. Guardiola J. Sinigaglia F. Nature. 1992; 356: 799-801Google Scholar, 25Mazerolles F. Barbat C. Fischer A. Int. Immunol. 1996; 8: 267-274Google Scholar, 26Nag B. Wada H.G. Passmore D. Clartk B.R. Sharma S.D. Mcconnell H. J. Immunol. 1993; 150: 1358-1364Google Scholar). All were synthesized by Neosystem (Strasbourg, France) and Genosphere Biotechnologies (Paris, France) according to the solid-phase synthesis method FMOC and further purified by means of HPLC, resulting in >96% purity as shown by HPLC analysis. Peptide sequences were validated by amino acid analysis of the purified preparations. The specific inhibitor of PI3-kinase, Ly294002 (27Vlahos C. Matter W.F. Hui K.Y. Brown R.F. J. Biol. Chem. 1994; 269: 5241-5248Google Scholar), was a kind gift from Lilly Research Laboratory (Indianapolis, IN). The Gab2 construct (Gab2-3YF, 3Y>F mutation of the three YXXM motifs), unable to bind p85-PI3-kinase used in these studies, was generated by PCR and subcloned into the pEBB vector developed by Gu et al. (20Gu H. Pratt J.C. Burakoff S.J. Neel B.G. Mol. Cell. 1998; 2: 729Google Scholar). The SHP-2 constructs used in these studies (SHP-2wt and SHP-2C/S, the catalytically inactive Cys to Ser mutant form) were built with the pRC/CMV vector developed by Zhao et al. (28Zhao R. Zhao Z.J. J. Biol. Chem. 2000; 275: 5453-5459Google Scholar). The HUT78 T cell line (CD4−, LFA-1+, CD2−) was obtained as described previously (3Mazerolles F. Barbat C. Meloche S. Graton S. Soula M. Fagard R. Fischer S. Hivroz C. Bernier J. Sekaly R.P. Fischer A. J. Immunol. 1994; 152: 5670-5679Google Scholar). HUT78-CD4− was infected with the wild type CD4 cDNAs. The transfected cells expressed CD4 normally as shown in a previous report (3Mazerolles F. Barbat C. Meloche S. Graton S. Soula M. Fagard R. Fischer S. Hivroz C. Bernier J. Sekaly R.P. Fischer A. J. Immunol. 1994; 152: 5670-5679Google Scholar). Cells were cultured in RPMI 1640 medium (Invitrogen) supplemented with 10% fetal calf serum, 2 mm l-glutamine, and 0.5 mg/ml G418 (Invitrogen). The HUT78-CD4+ T cell line (107) were also preincubated for 2 h in OPTIFECT medium (Bio Media, Boussens, France) and then transfected with the indicated plasmids in combination with 10 μg of green fluorescent protein at 260 V/1200 microfarads using a Bio-Rad electroporator. The transfected cells were grown for 20 h and then washed, and the green fluorescent protein-positive T cells (25–30%) were tested in adhesion assay with B cell lines preincubated for 20 min at 37 °C with hydroethidine (40 μg/ml; Polysciences, Warrington, PA). No modification of CD4 expression was observed after the different transfections (data not shown). SenseS-oligonucleotides were synthesized to nucleotides 1–21 of the human SHIP mRNA (5′-TGT CAG CAC GGC CGC AGA AGA-3′) and to nucleotides 1–18 of the human SHP-2 mRNA (29Walter A.O. Peng Z.Y. Cartwright C.A. Oncogene. 1999; 18: 1911-1920Google Scholar) (5′-ATG ACA TCG CGG AGA TGG-3′) (Genset, La Jolla, CA). The antisenseS-oligonucleotides were synthesized to the complementary strand (antisense SHIP, 5′-TCT TCT GCG GCC GTG CTG ACA-3′; antisense SHP-2, 5′-CCA TCT CCG CGA TGT CAT-3′). Cells were cultured for 48 h in the presence of 15 μm antisense and sense oligonucleotides in medium supplemented with 10% fetal calf serum. After this incubation, cells were washed and tested in adhesion assays. No modification of CD4 expression was observed after incubation with the different oligonucleotides (data not shown). The adhesion assay or conjugate formation between T and B cells was performed as described previously (30Mazerolles F. Lumbroso C. Lecomte O. Le Deist F. Fischer A. Eur. J. Immunol. 1988; 18: 1229-1233Google Scholar). The HUT 78-CD4+ T cell line was incubated with hydroethidine (40 μg/ml; Polysciences) for 20 min at 37 °C, and B cells were incubated with sulfate fluorescein diacetate (100 μg/ml; Molecular Probes) for 20 min at 37 °C. After washing, 3 × 105 T cells were preincubated during for the time indicated with 3 × 105 B cells at 37 °C. After this incubation, cells were cooled at 4 °C to block attachment and detachment. Before counting to maintain preformed conjugates, incubation at 4 °C was followed by centrifugation at 250 × g for 5 min, and cells were gently resuspended in 50 μl of RPMI 1640 medium. Conjugates were identified as red/green pairs of cells under a fluorescence microscope. Two hundred to 350 fluorescent cells were counted blindly in each experiment. Results are expressed as the percentage of T-B conjugates among all T cells. HUT78 T cell lines (20 × 106) were washed in RPMI 1640 medium (Invitrogen) and incubated for different times with soluble CD4 ligands without cross-linking. They were then pelleted, washed in RPMI 1640 medium, and lysed for 20 min on ice in lysis buffer A (20 mm Tris, pH 7.5; 140 mmNaCl; 50 mm NaF; 1 mmNa3VO4; 1% Nonidet P-40; antipain, pepstatin, and leupeptin (each at 2 μg/ml); 2 μg/ml aprotinin; and 1 mm phenylmethylsulfonyl fluoride). For immunoprecipitation with anti-LFA-1 Ab or anti-cytohesin Ab, T cells were lysed in 1 ml of ice-cold lysis buffer B (20 mm Tris, pH 8.2; 150 mm NaCl; 100 μmNa3VO4; 1% Brij 96; antipain, pepstatin, and leupeptin (each at 1 μg/ml); 100 μg/ml aprotinin; 1 mmphenylmethylsulfonyl fluoride; 5 mm iodoacetamide; and 2 mm MgCl2. Lysates were clarified by centrifugation at 12,000 × g for 15 min. The protein concentration was determined in postnuclear supernatants using the Bio-Rad kit with bovine serum albumin as standard. The same amount of each postnuclear lysate was incubated for 1 h at 4 °C with 3 μg of rabbit, rat, or mouse immunoglobulins and recovered by incubation with 40 μl of protein G-Sepharose beads for 45 min at 4 °C. Then the supernatant was incubated for 2 h or overnight at 4 °C with the specific antibodies: anti-p85-PI3-kinase (1:2000), anti-LFA-1α or -β (1:60), anti-cytohesin-1 (1:40), anti-SHIP (5 μg/ml), anti-SHP-2 (1 μg/ml), or anti-Gab2 (1 μg/ml). Immunoprecipitates were recovered by incubation with 40 μl of protein G-Sepharose beads for 45 min at 4 °C and washed three times in lysis buffer (A or B) prior to dissociation in reduced Laemmli sample buffer before resolution by 7% SDS-PAGE. Immunoprecipitates were electrophoretically transferred for 2 h at 150 V to a polyvinylidene difluoride membrane (Immobilon P, Millipore, Bedford, MA). Nonspecific binding was blocked with 5% bovine serum albumin in phosphate-buffered saline, and blots were hybridized with the different antibodies. Proteins were visualized using a chemiluminescence detection system (ECL+; Amersham Biosciences, Inc.) with an anti-rabbit, anti-rat, or anti-mouse Ig coupled to horseradish peroxidase as secondary Ab (Amersham Biosciences, Inc.). Experiments have been quantified by densitometric scanning. Immunoprecipitations were performed on extracts from control or cells incubated with CD4 ligands as described above. Proteins were released by boiling the immune complexes in 1% SDS. After centrifugation, the resulting supernatant was diluted to 0.1% using immunoprecipitation buffer and incubated for 1 h at 4 °C. Secondary immunoprecipitations were performed using the specific antibody as described above. 8 × 105 cells were washed in phosphate-buffered saline, cytospun onto slides for 8 min at 800 rpm, and then fixed and permeabilized for 5 min in ethanol. Subsequently cells were incubated with first antibody for 30 min at room temperature, washed 15 min in phosphate-buffered saline, and incubated with tetramethylrhodamine B isothiocyanate-conjugated goat anti-mouse IgG or fluorescein isothiocyanate-conjugated sheep anti-rabbit IgG for 30 min at room temperature in darkness. After the final wash with phosphate-buffered saline, slides were mounted on a 9:1 mixture of glycerol and 100 mm Tris/HCl, pH 9.0 containing n-propyl-gallate at 20 mg/ml as antifading reagent. Then samples were examined on a confocal laser scanning apparatus (LSM 510, Zeiss). It has been previously shown by Kolanus et al.(9Kolanus W. Nagel W. Schiller B. Zeitlmann L. Godar S. Stockinger H. Seed B. Cell. 1996; 86: 233-242Google Scholar) that LFA-1 in its active form is associated to an intracytoplasmic protein, cytohesin-1, expressing a PH domain to which thed-3 phospholipids synthesized by PI3-kinase could bind. Cytohesin-1 is thought to up-regulate LFA-1 avidity to its ligand ICAM-1. As the HLA-DR β-related 134–148 peptide as well as anti-CD4 Ab can inhibit LFA-1-mediated CD4+ T cell adhesion, we have tested whether the CD4-dependent down-regulation of LFA-1-mediated HUT78 T cell line adhesion was dependent upon cytohesin-1/LFA-1 association/dissociation. Following incubation with DR134–148 peptide, a dissociation of cytohesin-1 from LFA-1 was indeed observed (Fig.1A, lane 2). We have also investigated a hypothetic interaction between cytohesin-1 and PI3-kinase that could account for a local increase of d-3 phospholipids necessary to modify the β2integrin-dependent adhesion. As shown in Fig.1B, an association between PI3-kinase and cytohesin is observed in the absence of CD4 ligand binding, while DR134–148 rapidly induced a rapid partial dissociation of PI3-kinase from cytohesin-1. Activity of CD4-associated PI3-kinase (4Mazerolles F. Barbat C. Hivroz C. Fischer A. J. Immunol. 1996; 157: 4844-4854Google Scholar) could regulate the association between cytohesin-1 and LFA-1. We have tested this hypothesis by preincubating cells with a PI3-kinase activity inhibitor (Ly294002). As shown in Fig. 1A, the dissociation of LFA-1 from cytohesin-1 observed following cell incubation with DR134–148 (Fig. 1A, lane 2) was no longer detected when cells were preincubated with Ly294002 (lane 5). In contrast, the DR134–148 peptide induced an increased association between both proteins in the presence of Ly294002. Similarly, as shown in Fig.1C, Ly294002 preincubation inhibited the dissociation of cytohesin-1 from PI3-kinase observed after DR134–148 incubation (lane 4). We have previously observed that the p110 catalytic subunit of PI3-kinase was transiently tyrosine-phosphorylated following CD4 binding, suggesting the involvement of a tyrosine phosphatase. We have therefore investigated whether an interaction between PI3-kinase and the tyrosine phosphatase SHP-2, as described in another setting, is induced (12Sarmay G. Koncz G. Pecht I. Gergely J. Immunol. Lett. 1999; 68: 25-34Google Scholar). In Fig.2A, it is shown that DR134–148 can indeed trigger a transient association between SHP-2 and PI3-kinase that was maximal after a 10-min incubation and decreased thereafter. The same blot revealed with an anti-SHP-2 Ab showed that this increase was not related to an increase in SHP-2 immunoprecipitation since the same amount of SHP-2 was detected in each lane. No increase was detected with the control peptide V15A (data not shown). Sequential immunoprecipitation with an anti-SHP-2 Ab followed by an anti-PI3-kinase Ab also showed an increase in association following incubation with soluble anti-CD4 Ab (Fig.2B). No increase was detected by using an anti-LFA-1 Ab as a control (Fig. 2B). Similar results were observed by using the DR134–148 peptide (data not shown). Analysis by immunofluorescence imaging showed that p85-PI3-kinase and SHP-2 were cytoplasmic in unstimulated cells (Fig.3, A and B) after incubation with an anti-LFA-1 Ab (Fig. 3, C andD) or with the control peptide V15A (Fig. 3, Eand F). After incubation with anti-CD4 Ab (Fig. 3,G and H) or with DR134–148 peptide (Fig. 3,I and J), a translocation of p85-PI3-kinase and SHP-2 from cytosol to membrane was observed. No association was detected between goat anti-mouse Ig-tetramethylrhodamine B isothiocyanate and anti-LFA-1 Ab or anti-CD4 Ab used for stimulation (data not shown). It has been reported that SHIP can associate to PI3-kinase and SHP-2 and regulate the level of PIP3synthesized by PI3-kinase (13Lioubin M. Algate P. Tsai S. Carlberg K. Acbersold A. Rohrschneider L. Genes Dev. 1996; 101084Google Scholar). We have therefore investigated a hypothetic interaction between SHIP and PI3-kinase following incubation with DR134–148 peptide. PI3-kinase was detected after a sequential immunoprecipitation with an anti-PI3-kinase Ab followed by an anti-SHIP Ab (Fig. 4A). This interaction was transient and peaked after a 5-min incubation. A similar increase was observed after incubation with an anti-CD4 Ab (data not shown). SHP-2 was also detected in association with SHIP after sequential immunoprecipitation with anti-SHP-2 Ab followed by anti-SHIP Ab. A transient increase in this association was induced by anti-CD4 Ab (Fig.4B) or by DR134–148 peptide (data not shown). Sequential immunoprecipitation experiments following DR134–148 incubation did not lead to the detection of a ternary complex between PI3-kinase, SHIP, and SHP-2, suggesting that associations were mutually exclusive (data not shown). PI3-kinase and SHP-2 could also associate to adaptor proteins such as Gab2 known to be able to exert inhibitory activities (23Pratt J.C. Igras V.E. Maeda H. Baksh S. Gelfand E.W. Burakoff S.J. Neel B.G. Gu H. J. Immunol. 2000; 165: 4158-4163Google Scholar). We have investigated whether an association to Gab2 was detectable during events associated with the negative signaling induced by CD4 ligands. It was indeed observed that a soluble anti-CD4 Ab induced a transient association between Gab2 and PI3-kinase (Fig.5A). This transient interaction was maximal following a 30-s incubation with anti-CD4 Ab and decreased rapidly thereafter. No association was detected between Gab2 and SHP-2 (Fig. 5B). In contrast, in parallel to Gab2/PI3-kinase association, a transient association was induced between Gab2 and SHIP following incubation with anti-CD4 Ab (Fig.5C). The role of SHP-2, SHIP, and Gab2 in the down-regulation of LFA-1-mediated adhesion was investigated. For this purpose, the kinetics of adhesion of HUT78-CD4+ T cells preincubated with relevant antisense oligonucleotides or of HUT78-CD4+ T cells transfected with negative transdominant cDNAs were determined. By using antisense oligonucleotides to the SHP-2 protein encoding RNA, we found that, when expression of SHP-2 was reduced (as shown in Blot anti-SHP-2 of Fig.6A, top), down-regulation of adhesion was neutralized (Fig. 6A). In contrast, kinetics of adhesion was not altered when cells were treated with the control sense oligonucleotide. The role of SHIP in the down-regulation of LFA-1-mediated T cell adhesion was similarly analyzed by using sense or antisense oligonucleotides directed against SHIP mRNA. While the anti-SHIP immunoblot (Fig. 6B,top) showed a reduced expression of SHIP following antisense incubation, down-regulation of LFA-1-mediated adhesion was no longer detectable (Fig. 6B). In contrast, kinetics of adhesion was not modified when cells were treated with the control sense oligonucleotide (Fig. 6B). It was verified that incubation with the SHP-2 or SHIP antisense oligonucleotides did not modify CD4 surface expression (data not shown). The role of the association between PI3-kinase and Gab2 was investigated by transfecting a mutated form of Gab2 (Gab2-3Y>F) that is unable to bind PI3-kinase. After transfection of HUT78-CD4+ T cells with this mutated form of Gab2, adhesion to HLA class II+ B cells was significantly modified in contrast to T cell lines transfected with wild type Gab2 (Fig. 6C). The role of SHP-2 activity in the down-regulation of adhesion was also analyzed following transfection with a catalytically inactive SHP-2 protein (SHP-2C/S). Kinetics of adhesion between transfected SHP-2C/S-HUT78-CD4+ T cells and HLA class II+ B cells was found unchanged as compared with the control (Fig.6D). We have previously shown that CD4 ligands and especially the HLA-DR β-related 134–148 sequence (DR134–148 peptide) inhibits LFA-1-dependent adhesion (3Mazerolles F. Barbat C. Meloche S. Graton S. Soula M. Fagard R. Fischer S. Hivroz C. Bernier J. Sekaly R.P. Fischer A. J. Immunol. 1994; 152: 5670-5679Google Scholar). This negative signal requires the activities of the tyrosine kinase p56lck and the lipid kinase PI3-kinase. To account for the role of PI3-kinase on LFA-1-dependent adhesion, we have investigated whether CD4-dependent PI3-kinase activation could modify molecular associations potentially required in mediating LFA-1-dependent adhesion. Kolanus et al. (9Kolanus W. Nagel W. Schiller B. Zeitlmann L. Godar S. Stockinger H. Seed B. Cell. 1996; 86: 233-242Google Scholar) have demonstrated that a cytosolic protein, cytohesin-1, by associating to LFA-1 up-regulates the LFA-1-dependent adhesion in modifying the affinity of LFA-1 for its ligand ICAM-1. We herein show that DR134–148 induces a partial dissociation of LFA-1 from cytohesin in HUT78 T cells, an effect that is PI3-kinase-dependent. The kinetics of dissociation induced by DR134–148 correlates well to the kinetics of deadhesion between CD4+ T cells and HLA class II+ B cells we have previously reported (31Mazerolles F. Amblard F. Lumbroso C. Lecomte O. Van De Moortele P. Barbat C. Piatier-Tonneau D. Auffray C. Fischer A. Eur. J. Immunol. 1990; 20: 637-644Google Scholar). In parallel to CD4/PI3-kinase down-regulation of LFA-1/cytohesin-1 association, a translocation of PI3-kinase from the cytosol to the membrane and a partial dissociation of PI3-kinase from cytohesin-1 occur. These results fit the model proposed by Kolanus et al. (9Kolanus W. Nagel W. Schiller B. Zeitlmann L. Godar S. Stockinger H. Seed B. Cell. 1996; 86: 233-242Google Scholar) in which cytohesin association to LFA-1 correlates with up-regulation of LFA-1-mediated adhesion. PI3-kinase thus appears to be involved in both positive and negative events regulating LFA-1/cytohesin-1 association. On one hand, PI3-kinase activation leads to an increase in d-3 phospholipids enabling anchoring of cytohesin by its PH domain to the membrane in close vicinity to LFA-1 (32Nagel W. Zeitlmann L. Schilcher P. Geiger C. Kolanus J. Kolanus W. J. Biol. Chem. 1998; 273: 14853-14861Google Scholar). On the other hand, CD4-dependent activation and recruitment of PI3-kinase have the opposite effect. These results strongly suggest that PI3-kinase in the latter case is recruited to a different compartment to which cytohesin is excluded. This hypothesis is consistent with the finding of CD4-mediated PI3-kinase dissociation from cytohesin but remains to be directly tested. Krauss et al. (33Krauss K. Altevogt P. J. Biol. Chem. 1999; 274: 36921-36927Google Scholar) have shown that LFA-1-mediated binding of T cells to ICAM-1 is rapidly induced by clustering of membrane lipid rafts as a function of PI3-kinase activation. Krauss et al. (33Krauss K. Altevogt P. J. Biol. Chem. 1999; 274: 36921-36927Google Scholar) suggest that rafts preformed adhesion platforms, which would be important for the rapid regulation of lymphocyte adhesion. Krauss et al. (33Krauss K. Altevogt P. J. Biol. Chem. 1999; 274: 36921-36927Google Scholar) have also implied a close association of the cytohesin-1 system with raft cluster formation. We therefore suggest that CD4 ligand binding induces a delocalization of PI3-kinase from rafts containing cytohesin. It is also possible that CD4 ligand binding induces the recruitment of cytoskeleton proteins or phosphatases in rafts and thereby modifies the integrin activity. Cytoskeleton reorganization is also related to the formation of rafts (34Moran M. Polakis P. McCormick F. Pawson T. Ellis C. Mol. Cell. Biol. 1991; 111804Google Scholar). This hypothesis is presently under investigation. Another possibility would be that separate PI3-kinase compartments play distinct roles in cell adhesion regulation (35Constantin G. Majeed M. Giagulli C. Piccio L. Kim J.Y. Butcher E.C. Laudanna C. Immunity. 2000; 13: 759-769Google Scholar). In previous studies, we have also observed that DR134–148 induces a transient increase in tyrosine phosphorylation of p56lck and p110 subunit of PI3-kinase (36Mazerolles F. Fischer A. Int. Immunol. 1998; 10: 1897-1905Google Scholar). This suggests that tyrosine kinases and tyrosine phosphatases could also be involved in the negative signal induced by CD4 ligands. The adhesion down-regulation induced by CD4 ligands was found neutralized by a preincubation with an inhibitor of tyrosine kinase activity (3Mazerolles F. Barbat C. Meloche S. Graton S. Soula M. Fagard R. Fischer S. Hivroz C. Bernier J. Sekaly R.P. Fischer A. J. Immunol. 1994; 152: 5670-5679Google Scholar). Furthermore, several groups (37Gesbert F. Guenzi C. Bertoglio J. J. Biol. Chem. 1998; 273: 18273-18281Google Scholar, 38Craddock B.L. Welham M. J. Biol. Chem. 1997; 272: 29281-29289Google Scholar) have described interaction between PI3-kinase and the SH2-containing phosphotyrosine phosphatase family. We have therefore investigated whether the tyrosine phosphatase SHP-2 could associate to PI3-kinase and whether CD4 ligands were able to modify this interaction. It has been shown that CD4 ligands induce a transient increase in SHP-2 and PI3-kinase association. After CD4 triggering, a translocation of PI3-kinase and SHP-2 from the cytoplasm to the membrane is concomitantly observed. The SHP-2 expression was found necessary to observe a down-regulation of LFA-1-dependent adhesion induced by CD4, whereas SHP-2 activity is dispensable as shown by transfection experiments with T cells transfected with a catalytically inactive SHP-2 (SHP-2C/S). Xuet al. (39Xu F. Zhao R. Peng Y. Guerrah A. Zhao Z.J. J. Biol. Chem. 2001; 276: 29479-29484Google Scholar) have recently suggested that the catalytic domain is responsible for the localization of SHP-2 in different membrane compartments, although the SHP-2 activity was not required. SHP-2 could thus be required as an adaptor protein to associate PI3-kinase with other proteins independently of its catalytic activity in mediating inhibition of LFA-1-mediated adhesion. SHIP down-regulates PI3-kinase activity by hydrolyzing thed-3 phospholipids synthesized by PI3-kinase (40Damen J.E. Liu L. Rosten P. Humphries R.K. Jefferson A.B. Majerus P.W. Krystal G. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 1689-1693Google Scholar) and thus could be involved in the down-regulation of LFA-1-dependent adhesion. In addition, SHIP can bind SHP-2 (15Ono M. Bolland S. Tempst P. Ravetch J. Nature. 1996; 383: 263-266Google Scholar), and the role of SHIP in regulating LFA-1-dependent adhesion has been recently described in a murine myeloid cell line (41Rey-Ladino J.A. Huber M. Liu L. Damen J.E. Krystal G. Takei F. J. Immunol. 1999; 162: 5792-5799Google Scholar). The role of SHIP has thus been investigated, and a rapid transient association was detected between PI3-kinase and SHIP after CD4 binding. This association could favor the conversion of PI 3,4,5-P3 to PI 3,4-P2 by SHIP. This conversion could thus be important in signaling for the membrane localization or turnover of PH domain-containing proteins such as cytohesin. A role for SHIP in the down-regulation of T cell adhesion induced by CD4 was found as down-regulation was neutralized in the absence of SHIP expression. SHIP is also a substrate of SHP-2 and plays a major role as a negative regulator of intracellular signal transduction (15Ono M. Bolland S. Tempst P. Ravetch J. Nature. 1996; 383: 263-266Google Scholar), while its localization appears to be the determining factor in its mechanism of action (12Sarmay G. Koncz G. Pecht I. Gergely J. Immunol. Lett. 1999; 68: 25-34Google Scholar). SHP-2 was described as one of the proteins able to attract SHIP to the membrane close to lipid substrates. In parallel to PI3-kinase/SHIP association, an interaction between SHIP and SHP-2 has been shown, and a colocalization of both proteins, close to the membrane, has been found to be triggered by CD4 ligands. 2F. Mazerolles, unpublished data. One can therefore propose that SHIP binds to both SHP-2 and PI3-kinase in this setting. However, we did not succeed in detecting an association of these three proteins together. This association might be transient, or alternatively the association between SHIP and SHP-2 could be exclusive of SHIP-PI3-kinase. It is also possible that distinct pools of SHIP are involved, one associated to PI3-kinase and another to SHP-2. The adaptor protein Gab2 was also found to interact with PI3-kinase after CD4 ligand binding and prior to PI3-kinase/SHP-2 association. SHP-2 was not found associated to Gab2, suggesting that PI3-kinase was either sequentially associated to Gab2 and SHP-2 or that these associations were distributed in different compartments. The fact that the Gab2·PI3-kinase complex was detected and not Gab2/SHP-2 after CD4 binding suggests that the negative signaling is similar to the one regulating TCR signaling described recently by Gu et al.(23Pratt J.C. Igras V.E. Maeda H. Baksh S. Gelfand E.W. Burakoff S.J. Neel B.G. Gu H. J. Immunol. 2000; 165: 4158-4163Google Scholar). Indeed, Gu et al. (23Pratt J.C. Igras V.E. Maeda H. Baksh S. Gelfand E.W. Burakoff S.J. Neel B.G. Gu H. J. Immunol. 2000; 165: 4158-4163Google Scholar) show that Gab2 inhibits TCR activation in Jurkat cell, and this Gab2-mediated inhibition requires an interaction with PI3-kinase but not with SHP-2. In the absence of Gab2/PI3-kinase association, down-regulation of adhesion was not observed. Gab2 has a PH domain that is required for TCR signal inhibition (20Gu H. Pratt J.C. Burakoff S.J. Neel B.G. Mol. Cell. 1998; 2: 729Google Scholar). We therefore propose that the Gab2 PH domain binding to PIP3 localizes the Gab2·PI3-kinase complex to a specific subcellular site where Gab2 can exert its inhibitory effect. Gab2 could also mediate indirectly an association between PI3-kinase and SHP-2 with another protein. We have indeed observed that CD4 binding induces a transient association of Gab2 with SHIP in parallel to the transient association of Gab2 with PI3-kinase. Gab2 could thus bring together PI3-kinase and SHP-2 by the intermediary SHIP. In summary, we have shown that CD4 ligand binding, which induces a down-regulation of LFA-1-mediated adhesion in a T cell line, in parallel induces a dissociation of LFA-1 and PI3-kinase from cytohesin. In addition, a transient association of PI3-kinase with SHP-2, Gab2, and SHIP was observed. SHP-2, SHIP, and the Gab2/PI3-kinase association have also been described as important events in the negative regulation of lymphocyte signaling. These differential sequential associations could be required to maintain a stability in the level of PI 3,4-P2/PI 3,4,5-P3 involved in the regulation of LFA-1-dependent adhesion between T and B cells as proposed in the model pictured in Fig. 7. These association/dissociation events induced by CD4 ligand binding could also reflect distinct membrane compartmentalization of these proteins in raft domains and be associated to the negative signaling induced by CD4 by regulating cytohesin association to LFA-1. We thank Dr. Gu (Beth Israel-Deaconess Medical Center, Boston, MA) for the kind gift of Gab2 constructs and Dr. Zhao (Vanderbilt University Medical Center, Nashville, TN) for the kind gift of SHP-2 constructs. We also thank Dr. C. Hivroz for helpful discussions and Y. Goureau for excellent technical assistance for the confocal analysis.