β-Chemokine Receptor CCR5 Signals through SHP1, SHP2, and Syk*

The β-chemokine receptor CCR5 has been shown to modulate cell migration, proliferation, and immune functions and to serve as a co-receptor for the human immunodeficiency virus. We and others have shown that CCR5 activates related adhesion focal tyrosine kinase (RAFTK)/Pyk2/CAK-β. In this study, we further characterize the signaling molecules activated by CCR5 upon binding to its cognate ligand, macrophage inflammatory protein-1β (MIP1β). We observed enhanced tyrosine phosphorylation of the phosphatases SHP1 and SHP2 upon MIP1β stimulation of CCR5 L1.2 transfectants and T-cells derived from peripheral blood mononuclear cells. Furthermore, we observed that SHP1 associated with RAFTK. However, using a dominant-negative phosphatase-binding mutant of RAFTK (RAFTKm906), we found that RAFTK does not mediate SHP1 or SHP2 phosphorylation. SHP1 and SHP2 also associated with the adaptor protein Grb2 and the Src-related kinase Syk. Pretreatment of CCR5 L1.2 transfectants or T-cells with the phosphatase inhibitor orthovanadate markedly abolished MIP1β-induced chemotaxis. Syk was also activated upon MIP1β stimulation of CCR5 L1.2 transfectants or T-cells and associated with RAFTK. Overexpression of a dominant-negative Src-binding mutant of RAFTK (RAFTKm402) significantly attenuated Syk activation, whereas overexpression of wild-type RAFTK enhanced Syk activity, indicating that RAFTK acts upstream of CCR5-mediated Syk activation. Taken together, these results suggest that MIP1β stimulation mediated by CCR5 induces the formation of a signaling complex consisting of RAFTK, Syk, SHP1, and Grb2.

CCR5 is a G-protein-coupled seven-transmembrane receptor that belongs to the pro-inflammatory ␤-chemokine receptor family. It is a co-receptor for macrophage-tropic HIV-1 isolates (4,11,12). CCR5 is activated by the ␤-chemokines MIP1␤, MIP1␣, and RANTES. MIP1␤ has higher specificity for CCR5 than do MIP1␣ and RANTES, which also bind to CCR1 and CCR3 (1,3,12). Knockout mice lacking CCR5 revealed partial defects in macrophages and also showed enhanced T-cell-dependent immune response and delayed type hypersensitivity reaction, suggesting that CCR5 may play an important role in down-regulating T-cell-dependent immune responses (13).
Despite the emerging role of CCR5 and its ligands in HIV infection and the immune response, relatively little is known about the signaling pathways mediated by this receptor. CCR5 has been shown to induce calcium signals and chemotaxis upon binding to the macrophage-tropic HIV gp160 recombinant envelope protein (14). We (15) and others (16) have also recently shown that CCR5, upon binding to its cognate ligand (MIP1␤) or to the HIV-1 envelope glycoprotein from a macrophagetropic strain, activates a member of the focal adhesion kinase family called related adhesion focal tyrosine kinase (RAFTK; also known as Pyk2 and CAK-␤). We further demonstrated that RAFTK associates with the cytoskeletal protein paxillin upon CCR5 activation (15).
Syk, a cytoplasmic protein-tyrosine kinase, plays an important role in the signaling pathways mediated by the B-cell antigen, Fc receptor, T-cell antigen receptor, and integrin receptor and thereby modulates cell growth and chemotaxis (33)(34)(35)(36)(37). Syk interacts with various tyrosine-phosphorylated proteins (38,39). Prior analysis of RAFTK indicated that tyrosine 402 in its N-terminal domain binds to the SH2 domain of Src kinases and activates such Src kinases upon lipopolysaccharide or bradykinin treatment (40).
This study demonstrates that MIP1␤ stimulation of CCR5 transfectants induces the tyrosine phosphorylation of SHP1 and SHP2 and activates Syk. This induction also results in the formation of a signaling complex consisting of RAFTK, Syk, SHP1, and Grb2. These results indicate that RAFTK acts upstream of Syk and suggest that it does not regulate the SHP1 and SHP2 phosphorylation mediated by CCR5.

EXPERIMENTAL PROCEDURES
Reagents and Materials-Anti-RAFTK antibodies were generated as described previously (41). This antiserum recognized both human and murine forms of RAFTK and did not cross-react with focal adhesion kinase. Antibodies to Syk, SHP1, and SHP2 were obtained from Santa Cruz Biotechnology (Santa Cruz, CA). Monoclonal anti-phosphotyrosine antibody (4G10) was a generous gift from Dr. Brian Druker (Oregon Health Sciences University, Portland, OR). Electrophoresis reagents and nitrocellulose membranes were obtained from Bio-Rad. The protease inhibitors leupeptin and ␣ 1 -antitrypsin and all other reagents were acquired from Sigma. Indo-1/AM was purchased from Molecular Probes, Inc. (Eugene, OR).
Construction of CCR5 Stable Transfectants-CCR5 was transfected into the pre-B lymphoma cell line L1.2 as described previously (42,43), and transfectants were selected in medium containing mycophenolic acid. FACS analysis was used to monitor the cell-surface expression of CCR5. CCR5 was expressed at a high level in these cells (42). The ␤-chemokines MIP1␣, MIP1␤, and RANTES bind with high affinity to the expressed CCR5 receptors. These cells possess properties characteristic of native CCR5, as they bind to macrophage-tropic HIV-1 gp120 in the presence of soluble human CD4 (42).
RAFTK Transfectants-Transfectants of mutants RAFTK m402 and RAFTK m906 and wild-type RAFTK (RAFTK WT ) were produced by transfection of the CCR5 L1.2 cells with the RAFTK m402 , RAFTK m906 , or RAFTK WT construct, respectively (15). pcDNA vector without an RAFTK construct was used as a control. Mutants RAFTK m906 and RAFTK m402 were generated by replacing Tyr 906 and Tyr 402 with Phe, respectively, by site-directed mutagenesis. Tyr 906 of RAFTK is the putative binding site for phosphatases, and Tyr 402 has previously been shown to bind to Src kinases (40). Plasmids carrying RAFTK m906 , RAFTK m402 , RAFTK WT , or pcDNA control vector were transfected by electroporation into the CCR5 L1.2 cells using Bio-Rad electroporation equipment. The transfectants were selected in medium containing mycophenolic acid and G418. The double mutants expressed equal amounts of CCR5 as determined by FACS analysis. Several clones of double transfectants were used in the signaling studies.
Primary Lymphocyte Culture-T-cell-enriched, monocyte-depleted cultures were generated from peripheral blood mononuclear cells as described (15,44). Briefly, peripheral blood mononuclear cells were separated by Ficoll-Hypaque gradient centrifugation and two rounds of adherence to plastic. Non-adherent cells were stimulated with phytohemagglutinin (5 g/ml) for 3 days. Cells were removed to fresh medium supplemented with recombinant human interleukin-2 (Advanced Biotechnologies, Columbia, MD). Three-week-old activated T-cells, which were found to be ϳ35% positive for CCR5 by FACS analysis, were used for further studies.
Immunoprecipitation and Western Blot Analysis-Immunoprecipitation studies were conducted as described (45). Briefly, identical amounts of protein from each time point were clarified by incubation with protein A-Sepharose CL-4B (Amersham Pharmacia Biotech) for 1 h at 4°C. Protein A-Sepharose was removed by brief centrifugation, and the supernatants were incubated with different primary antibodies as described below for each experiment for 2 h at 4°C. Immunoprecipitation of the antibody-antigen complexes was performed by incubation at 4°C overnight with 50 l of protein A-Sepharose (10% suspension). Nonspecific bound proteins were removed by washing the Sepharose beads three times with modified radioimmune precipitation assay buffer and one time with phosphate-buffered saline. Immune complexes were solubilized in 30 l of 2ϫ Laemmli buffer, and samples were separated on 8 or 12% SDS-polyacrylamide gel and then transferred to nitrocellulose membranes. The membranes were blocked in 5% nonfat milk protein for 1 h and probed with primary antibody for 3 h at room temperature or at 4°C overnight. Immunoreactive bands were visualized using horseradish peroxidase-conjugated secondary antibody and the enhanced chemiluminescence system (ECL, Amersham Pharmacia Biotech). The densitometric scanning of films was done using a Bio-Rad Model G5-700 imaging densitometer.
Syk Kinase Assay-The Syk kinase assay was performed by first immunoprecipitating lysates with anti-Syk antibody. The immune complexes were then washed twice with radioimmune precipitation assay buffer and twice with Syk kinase buffer (20 mM Hepes, 50 mM NaCl, 10 M Na 3 VO 4 , 5 mM MgCl 2 , and 5 mM MnCl 2 ). The complex was incubated in Syk kinase buffer and 5 Ci of [␥-32 P]ATP for 20 min at 30°C. The reaction was terminated by adding 4ϫ SDS sample buffer and boiling the samples for 5 min at 100°C. Proteins were then separated on 8% SDS-polyacrylamide gel and detected by autoradiography. Rabbit IgG was used as a negative control.
Glutathione S-Transferase Fusion Protein Binding Studies-The RAFTK C-terminal domain (amino acids 681-1009)-glutathione S-transferase (GST) fusion protein was produced as described (46). Briefly, the fusion protein was amplified by polymerase chain reaction and cloned into the pGEX-2T expression vector (Amersham Pharmacia Biotech). The GST fusion protein was produced by 1 mM isopropyl-␤-Dthiogalactopyranoside induction and purified by affinity chromatography on a glutathione-Sepharose column (Amersham Pharmacia Biotech) according to the manufacturer's recommendations. Grb2-GST fusion proteins were purchased from Santa Cruz Biotechnology. For the binding experiments, 5 g of GST fusion protein were mixed with 1 mg of cell lysate and incubated for 1 h at 4°C on a rotatory shaker. GST protein (Santa Cruz Biotechnology) was used as a control. The complex was pre-absorbed by adding 50 l of glutathione-Sepharose 4B beads (Amersham Pharmacia Biotech). The samples were incubated overnight at 4°C on a rotatory shaker. The beads were centrifuged and washed three times with modified radioimmune precipitation assay buffer and once with 1ϫ phosphate-buffered saline. The bound proteins were eluted by boiling in Laemmli sample buffer and subjected to 8% SDS-polyacrylamide gel electrophoresis and Western blot analysis.
Chemotaxis Assay-The assay was performed in 24-well plates containing 5-m porosity inserts (Costar Corp., Cambridge, MA) as described (45). Briefly, cells were resuspended at 6.6 ϫ 10 6 /ml in RPMI 1640 medium containing 2.5% bovine serum albumin. 50 ng of MIP1␤ were added to the bottom wells, and 150 l of cells (1 ϫ 10 6 ) untreated or pretreated with different concentrations of sodium orthovanadate were loaded onto the inserts. Cells migrating to the bottom wells were collected, centrifuged, and counted. The results shown are representative of findings from three independent experiments.

SHP1 and SHP2 Are Tyrosine-phosphorylated upon MIP1␤
Stimulation-Cytoplasmic tyrosine phosphatases have been shown to be important positive and negative regulators of signaling pathways. To characterize the role that phosphatases play in CCR5-mediated signal transduction pathways, CCR5 transfectants were stimulated with MIP1␤, and lysates were analyzed for SHP1 and SHP2 tyrosine phosphorylation. As shown in Fig. 1 (A and C, respectively), MIP1␤ stimulation of CCR5 transfectants resulted in an increase in tyrosine phosphorylation of SHP1 and SHP2. Equal amounts of SHP1 and SHP2 proteins were present in each lane (Fig. 1, A and C, lower  panels). We also found that MIP1␤ stimulation induced a slight increase in tyrosine phosphorylation of SHP1 and SHP2 (Fig. 1,  B and D, respectively) in activated T-cells derived from peripheral blood mononuclear cells. These T-cells were found to be ϳ35% positive for CCR5 (data not shown). The reduced SHP1 and SHP2 phosphorylation in T-cells as compared with CCR5 L1.2 cells may be due to differences in the number of CCR5 receptors on these cells and/or to differences between the cell types. In addition, RAFTK has also been shown to have lower tyrosine phosphorylation in MIP1␤-stimulated T-cells as compared with the similarly stimulated CCR5 L1.2 transfectants (15).
SHP1 and SHP2 Associate with Various Signaling Molecules-Tyr 906 of RAFTK may act as a putative binding site for SHP proteins. Therefore, we examined the association of RAFTK with SHP1. As shown in Fig. 2A, SHP1 associated with RAFTK constitutively; this association was modestly increased upon MIP1␤ stimulation. To further confirm this association, we used a GST fusion protein containing the C-terminal domain of RAFTK. As shown in Fig. 2B, SHP1 was observed to associate with this fusion protein. SHP1 also associated with Grb2 upon MIP1␤ stimulation (Fig. 2C). Since SHP2 was also tyrosine-phosphorylated following MIP1␤ stimulation, we studied whether it associated with the adaptor protein Grb2. As shown in Fig. 2D, SHP2 associated with Grb2, and this association increased upon chemokine stimulation.
RAFTK m906 Has No Effect on the SHP1 or SHP2 Phosphorylation Induced by MIP1␤-Since SHP1 was shown to associate with RAFTK upon MIP1␤ stimulation, we were interested in whether RAFTK modulated the MIP1␤-stimulated phosphorylation of SHP1. To address this question, we created doubletransfected L1.2 cells that expressed human CCR5 and RAFTK m906 , in which Tyr 906 was replaced with Phe. Tyr 906 of RAFTK is a putative binding site for phosphatases. As shown in Fig. 3 (A and B, respectively), there was no significant increase in tyrosine phosphorylation of SHP1 and SHP2 in the RAFTK m906 transfectants as compared with the pcDNA transfectants. An equal amount of SHP1 and SHP2 proteins was present in all samples (Fig. 3, A and B, lower panels).
Effect of Phosphatase Inhibitor on MIP1␤-induced Migration of CCR5 Cells-Tyrosine phosphatases have been shown to Syk Is Phosphorylated and Activated upon MIP1␤ Stimulation-The Src-related kinase Syk has been shown to participate in various signaling pathways regulating cell growth and adhesion. To characterize the role of Syk in CCR5-mediated signal transduction pathways, CCR5 transfectants or T-cells were stimulated with MIP1␤, and the lysates were analyzed for Syk kinase activation. As shown in Fig. 5A (upper panel), MIP1␤ stimulation of the CCR5 L1.2 transfectants resulted in enhanced tyrosine phosphorylation of Syk. Equal amounts of Syk protein were present in each lane (Fig. 5A, lower panel). In addition, MIP1␤ stimulation induced Syk-autophosphorylating activity in CCR5 L1.2 transfectants (Fig. 5B) and, to a lesser degree, in T-cells (Fig. 5C). Similar to SHP1 and SHP2 phosphorylation, Syk-autophosphorylating activity was reduced in T-cells as compared with the CCR5 L1.2 transfectants upon MIP1␤ stimulation.
Syk Associates with RAFTK, Grb2, SHP1, and SHP2-To further characterize the role that Syk might play in CCR5mediated signaling, we sought to identify proteins that associate with Syk upon MIP1␤ stimulation. We observed a constitutive association of Syk with RAFTK, an association that was enhanced upon MIP1␤ stimulation (Fig. 6A). RAFTK has previously been shown to be phosphorylated and activated by MIP1␤ (15). Furthermore, Syk was also shown to associate with the SH2 domain of Grb2. This association was enhanced upon MIP1␤ stimulation (Fig. 6B). Since we observed that Syk was activated and associated with RAFTK upon MIP1␤ stimulation, we subsequently studied the association of Syk with SHP1. As shown, Syk also associated with SHP1 (Fig. 7), and this association was enhanced upon MIP1␤ stimulation. Syk Stimulation by MIP1␤ Is Mediated by RAFTK-Since RAFTK associated with Syk, we wanted to see whether this association was important for Syk activation. To address this question, we created double-transfected L1.2 cells that expressed human CCR5 and RAFTK m402 (which lacks the Srcbinding site) or CCR5 and RAFTK WT . Up to a 3-fold increase in Syk activation was observed in the MIP1␤-treated CCR5-RAFTK WT cells as compared with the pcDNA-transfected cells (Fig. 8A). We also observed an ϳ2-fold decrease in Syk activity in the CCR5 L1.2 transfectants overexpressing the dominantnegative mutant RAFTK m402 as compared with transfectants expressing the pcDNA vector (Fig. 8B). These studies suggest that RAFTK may regulate Syk kinase activation in these cells. more, HIV gp120 binding to CCR5 and chemokine stimulation have also been shown to induce RAFTK/Pyk2 phosphorylation (16,47). In the present study, we have further characterized the role of RAFTK and investigated the possible involvement of the phosphatases SHP1 and SHP2 and the Src-related kinase Syk in signaling pathways mediated by CCR5. Cytoplasmic tyrosine phosphatases and Src-related kinases are known to modulate regulatory pathways of cell spreading, migration, and cytoskeletal organization (48 -52).
MIP1␤ stimulation enhanced tyrosine phosphorylation of the phosphatases SHP1 and SHP2. Both of these phosphatases have been shown to participate in various receptor-mediated pathways. SHP1 has been shown to act as a negative regulator of signaling pathways, whereas SHP2 appears to function as a positive mediator (17,18,(21)(22)(23)(24). Recent studies of hematopoietic cells from motheaten mice indicate that SHP1 plays an important role in the regulation of the stromal cell-derived factor-1-induced signaling pathway (32). We have observed that SHP1 is associated with RAFTK. However, RAFTK does not appear to mediate the MIP1␤-stimulated phosphorylation of SHP1 or SHP2 since overexpression of a dominant-negative phosphatase-binding mutant of RAFTK had little effect on MIP1␤-stimulated SHP1 or SHP2 phosphorylation.
We also observed that chemokine stimulation resulted in the enhanced association of SHP1 and SHP2 with Syk and Grb2. In different signaling pathways, SHP1 and SHP2 are known to associate with various signaling molecules, including Vav, Grb2, SOS, and SLP-76 via their SH2 domains (53)(54)(55)(56). In addition to their role as phosphatases, SHP1 and SHP2 may act as adaptor proteins by providing docking sites for the recruitment of downstream signaling molecules. The present study indicates that tyrosine phosphatases may play an important role in the regulation of MIP1␤-induced migration, as orthovanadate treatment markedly attenuated MIP1␤-stimulated chemotaxis. Recently, platelet-derived growth factor-induced migration was shown to be regulated by SHP2 (48).
MIP1␤ stimulation also induced activation of Syk, a Srcrelated kinase. Syk has been shown to participate in signal transduction pathways mediated by B-and T-cell antigen receptors, the Fc receptor, various growth factor receptors, and integrin receptors (33)(34)(35)(36)(37). However, the G-protein-coupled m1 muscarinic receptor does not activate Syk. Tyr 402 (autophosphorylation site) of RAFTK has been shown to bind to Src kinases, which results in their activation (40). In the present study, we observed that Syk associated with RAFTK and that this association was enhanced upon MIP1␤ stimulation. RAFTK association with another Src-related kinase, Fyn, has been shown to play an important role in mediating T-cell receptor signal transduction (46,57). Furthermore, RANTES has been shown to induce ZAP-70 activity and its association with focal adhesion kinase in T-cells (58). In this study, RAFTK appeared to partially mediate Syk activation, as overexpression of a Src-binding mutant of RAFTK resulted in the reduced activation of Syk, whereas overexpression of wild-type RAFTK enhanced Syk activity. Recently, Syk was shown to be upstream of RAFTK/Pyk2 phosphorylation in Fc⑀ receptor-1-induced tyrosine phosphorylation in mast cells, whereas Pyk2 phosphorylation by thrombin and the adenosine G-proteincoupled receptor was independent of RAFTK/Pyk2 in these cells (59).
We also observed that Syk associated with SHP1 and that this association was enhanced by chemokine stimulation. Tyrosine phosphatases can cause activation of Src kinases (60). SHP1 has been shown to regulate Syk activity, as overexpression of SHP1-inactive mutants in B lymphoma cell lines results in enhanced Syk kinase activity (61).
Taken together, our results provide new information regarding various downstream signaling molecules involved in CCR5mediated signaling pathways. We have found that MIP1␤ stimulation of CCR5 activates Syk and increases the tyrosine phosphorylation of the SH2-domain containing phosphatases SHP1 and SHP2. This results in the formation of a multimeric complex consisting of RAFTK, Syk, Grb2, and SHP1. RAFTK was shown to partially mediate the activation of Syk, but had no significant effect on SHP1 or SHP2 tyrosine phosphorylation. These results suggest that RAFTK differentially regulates several downstream signaling targets that are activated upon CCR5 stimulation.