Fc receptor-mediated platelet activation is dependent on phosphatidylinositol 3-kinase activation and involves p120(Cbl).

The platelet receptor for the Fc domain of IgGs (FcgammaRIIa) triggers intracellular signaling through protein tyrosine phosphorylations leading to platelet aggregation. In this study, we focused on the adaptor protein p120(cbl) (Cbl), which became tyrosine-phosphorylated after platelet activation induced by antibodies. Cbl phosphorylation was dependent on Fc receptor engagement. An association of Cbl with the p85 subunit of phosphatidylinositol 3-kinase (PI 3-K) occurred in parallel with Cbl tyrosine phosphorylation. We showed by in vitro experiments that Cbl/p85 association was mediated by the Src homology 3 domain of p85/PI 3-K and the proline-rich region of Cbl. Inhibition of PI 3-K activity by wortmannin led to the blockade of both platelet aggregation and serotonin release mediated by FcgammaRIIa engagement, whereas it only partly inhibited those induced by thrombin. Thus, PI 3-K may play a crucial role in the initiation of platelet responses after FcgammaRIIa engagement. Our results suggest that Cbl is involved in platelet signal transduction by the recruitment of PI 3-K to the FcgammaRIIa pathway, possibly by increasing PI 3-K activity.

The Cbl protein is the product of the c-cbl protooncogene, the cellular homologue of the v-cbl oncogene present in the Cas-NS-1 retrovirus, which induces pre-B cell lymphomas and myeloid leukemias (1,2). Cbl is found in a wide range of hemopoietic cell lineages and some nonhemopoietic tissues such as lung, brain, and testis (3). A deletion in the c-cbl sequence (62% of the C-terminal domain) involving functional domains, such as the leucine zipper motif and proline-rich region, converts this protooncogene into the transforming one (4). Unlike the product of v-cbl localized in both the cytoplasm and the nucleus, the c-cbl product (p120 cbl ) is exclusively cytoplasmic (5). A tumorigenic form of Cbl was detected in the 70Z/3 pre-B cell lymphoma, in which Cbl tyrosine phosphorylation is increased as a result of a deletion of 17 amino acids in the Cbl sequence (6). Cbl is also heavily tyrosine-phosphorylated in v-src-transformed hemopoietic cells (7).
A regulator activity has recently been described for Cbl when overexpressed in mast cells, in which it regulates p72 syk activity (36). Cbl is also proposed to regulate the T cell receptormediated Ras pathway activation via its association with Grb2 in T cells (37). In interleukin 4-treated B cells, Cbl is tyrosinephosphorylated and associated with p85/PI 3-K, and overexpression of Cbl enhances PI 3-K activity, mitogenic activity, and cell survival (38). Taken together, the data suggest a role for Cbl in multiple signaling pathways of different cell types.
In human platelets, Cbl has been identified and shown to be constitutively associated with the Grb2 adaptor protein and tyrosine-phosphorylated after thrombopoietin activation (27). Thus, Cbl seems to be implicated in signal transduction after thrombopoietin binding to c-mpl. However, the role of Cbl or its possible involvement in platelet signal transduction mediated by other receptors has not yet been documented. Platelet activation is mediated by a wide variety of agonists, including thrombin, thromboxane A2, ADP, and adhesion molecules such as von Willebrand factor and collagen. Some antibodies directed against antigens on the platelet membrane (e.g. the tetraspanin CD9, glycoprotein IV, and the integrin ␣ IIb -␤ 3 ) are also able to induce platelet activation. In most cases, the activation induced by IgGs is dependent on the binding of their Fc domain on the specific receptor, Fc␥RIIa (39).
In the present study, we have investigated the involvement of Cbl in platelet signaling after Fc␥RIIa engagement. Two models of platelet activation involving the Fc receptor were used: the cross-linking of Fc␥RIIa and bridging of the CD9 antigen with Fc␥RIIa by an activating monoclonal antibody (mAb) (anti-CD9, Syb). Our results demonstrate that after Fc␥RIIa engagement, Cbl was heavily tyrosine-phosphorylated. In parallel with Cbl phosphorylation, we found that Cbl was associated with p85/PI 3-K. Moreover, the use of wortmannin, an inhibitor of PI 3-K, abolished platelet aggregation and release induced by antibodies, underlining the crucial role of PI 3-K during immunological activation. The results suggest an * The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18  important role for Cbl in Fc␥RIIa-mediated platelet activation, possibly through the regulation of PI 3-K activity.
Platelet Preparation-Human platelets were isolated from fresh platelet concentrates obtained from healthy donors who did not taken aspirin for at least 1 week. The concentrates were centrifuged at room temperature for 15 min at 130 ϫ g to eliminate other cell types and then subjected to a washing process as described previously (41). Briefly, platelets were isolated on a metrizamide gradient, collected, and resuspended in 10 mM HEPES buffer, pH 7.4, containing 140 mM NaCl, 5 mM NaHCO 3 , 0.5 mM MgCl 2 , 3 mM KCl, and 10 mM glucose. Platelet concentration was adjusted to 10 9 /ml for immunoprecipitation studies or to 5.10 8 /ml for studying the total platelet lysates. CaCl 2 (1 mM) was added 10 min before platelet stimulation.
Platelet Activation, Aggregation, and Release-Platelets were stimulated with either 10 g/ml of antibodies or 1 unit/ml human thrombin at 37°C in an aggregometer (Coulter, Havertown, PA) with constant stirring (1100 rpm). For Fc receptor pathway inhibition, platelets were preincubated for 1 min at 37°C with IV.3 (10 g/ml) before addition of Syb antibody. For Fc receptor cross-linking, platelets were preincubated for 1 min with IV.3 (10 g/ml) and then stimulated by addition of RAM-F(ab)Ј 2 (80 g/ml) for various periods. To study the total platelet proteins, the reactions were stopped by addition of 25% (v/v) of a buffer containing 10% SDS and 5 mM EDTA, and the samples were transferred to ice for complete lysis. After 30 min, 25% (v/v) of 4 ϫ concentrated Laemmli's buffer and 5% of 2-mercaptoethanol were added, and the samples were subjected to Western blot analysis.
To study the platelet aggregation and release, the platelet-rich plasma was adjusted to 5 ϫ 10 8 platelets/ml and incubated with 0.6 M 14 C-labeled serotonin (Amersham Pharmacia Biotech) for 30 min at room temperature, and then the platelets were isolated as described above. Imipramine was added to the platelet-rich plasma 10 min before the agonist to prevent reuptake of the serotonin during the experiment. Inhibition of PI 3-K was performed by the preincubation of platelets with wortmannin (50 nM) at 37°C for 15 min before platelet activation. The aggregation was determined by measuring changes in light transmission through a stirred volume of platelets at 37°C in aggregometer. The aggregation was monitored for 5 min, and the reaction was stopped by transfer into 0.2 volume of ice-cold 0.1 M EDTA and immediate centrifugation for 1 min at 12,000 ϫ g. The [ 14 C]serotonin was measured in the supernatant by liquid scintillation counting. Release was expressed as percent [ 14 C]serotonin liberated compared with the total unstimulated platelet content.
Immunoprecipitation-For immunoprecipitation (IP) studies, platelet stimulation was stopped by the addition of one-third volume of cold 3 ϫ concentrated Nonidet P-40 lysis buffer containing 3% (v/v) Nonidet P-40, 150 mM Tris, 450 mM NaCl, 15 g/ml leupeptin, 15 g/ml aprotinin, 3 mM EGTA, 3 mM Na 3 VO 4 , and 3 mM phenylmethylsulfonyl fluoride. The mixture was transferred to ice for 30 min for complete lysis. Insoluble material was removed by centrifugation for 10 min at 16,000 ϫ g at 4°C, and the supernatant was incubated with antibodies against Cbl or phosphotyrosine proteins (2.5 g/ml) for 2 h at 4°C. Immune complexes were incubated with protein A-Sepharose beads (30 l of 50% slurry) for 1 h at 4°C and then isolated by brief centrifugation. After washing three times with 0.5 ml of 1 ϫ concentrated cold Nonidet P-40 lysis buffer (described above), immunoprecipitates were resuspended in Laemmli's sample buffer containing 5% 2-mercaptoethanol and analyzed by Western blot.
For the preclearing experiments, antibody concentrations used for protein immunodepletion were raised to 5 g/ml. The Nonidet P-40soluble fraction was incubated with anti-Cbl, anti-phosphotyrosine (4G10), or control antibodies for 2 h, followed by addition of 40 l of protein A-Sepharose (50% slurry). This step was reproduced after removing the immuncomplexes. Before addition of each antibody for IP, an aliquot (50 l) of the platelet lysate was conserved in Laemmli's buffer at Ϫ20°C. Both immunoprecipitates and aliquot samples were subjected to Western blot analysis.
Western Blot Analysis-Samples were boiled for 5 min, and proteins were subjected to 10% SDS-PAGE. Separated proteins were transferred electrophoretically to a nitrocellulose membrane (Bio-Rad system). The membrane was incubated 1 h in a blocking buffer containing 5% low dry milk, 2% Tween 20, 100 mM NaCl, and 20 mM Tris, pH 7.4. Specific antibodies against proteins of interest were added for 2 h, followed by a 1-h incubation with horseradish peroxidase-conjugated secondary antibody. An enhanced chemiluminescence system was used for signal detection.
In some experiments, the membranes were stripped; the bound antibody was removed by incubation in buffer containing 2% SDS, 62.5 mM Tris, pH 6.8, and 100 mM 2-mercaptoethanol for 40 min at 60°C. After extensive washing, the membrane was reprobed with another antibody as described above.
The GST fusion proteins were subjected to a competition study. In this case, the GST-p85 and GST-SH3 were preincubated 30 min with the proline-rich peptides corresponding to residues 82-96 and 300 -314 of p85, which were previously shown to bind to the SH3 domain of the p85 itself (43), using an unrelated peptide as control (GSQVVRIVG-GRD). Preincubations were performed at 4°C for 30 min with constant stirring, added to platelet lysates, and treated as described above.

RESULTS
Cbl Tyrosine Phosphorylation during Platelet Activation-To examine whether activation of platelets through Fc␥RIIa involved Cbl, we activated platelets by Fc receptor cross-linking, using IV.3 (anti-Fc receptor) in the presence of RAM-F(ab)Ј 2 or by an anti-CD9 antibody (Syb) known to induce platelet activation through an Fc␥RIIa-dependent pathway. We also compared these results with those observed after thrombin stimulation.
Platelet activation mediated through Fc␥RIIa or the thrombin receptor induced an increase in the level of phosphotyrosine proteins (PYs). We focused our attention on a band at ϳ120 kDa, which was tyrosine-phosphorylated after 2 min of platelet activation (Fig. 1, upper panel). To see whether this band could correspond to p120 cbl , the nitrocellulose membrane was reprobed with anti-Cbl antibody. This experiment confirmed the presence of Cbl in platelets, which is localized within the 120-kDa band (Fig. 1, lower panel). This result was further confirmed by preclearing experiments. Lysates of Syb-activated platelets were subjected twice to anti-PY or anti-Cbl immunoprecipitations. Many PYs were absent after immunodepletion of total PY, including the tyrosine-phosphorylated 120-kDa band (Fig. 2a, upper panel, lanes 4 and 6). A fraction of Cbl remained in the total lysate subjected to anti-PY immunodepletion, which probably corresponds to unphosphorylated Cbl (Fig.  2a, lower panel, lanes 4 and 6). In platelet lysates depleted of Cbl, the tyrosine-phosphorylated band at ϳ120 kDa remained unchanged (Fig. 2a, upper panel, lanes 2 and 5), indicating the presence of additional proteins in the 120-kDa band. Anti-Cbl immunoblotting of the same membrane confirmed the total depletion of Cbl in the lysate (Fig. 2a, lower panel, lane 5). Furthermore, Cbl was present in the anti-PY immunoprecipi-tates of Syb-activated platelets (Fig. 2b, lanes 3 and 6). Altogether, these data confirmed the presence of Cbl within the 120-kDa band and indicated that in Syb-activated platelets, the 120-kDa phosphotyrosine band corresponds to several PYs including Cbl.
To determine the levels of Cbl tyrosine phosphorylation induced by the former agonists, Cbl was immunoprecipitated from lysates of resting and activated platelets. As shown in Fig.  3, upper panel, Cbl was not significantly tyrosine-phosphorylated in resting platelets. By contrast, a high level of Cbl tyrosine phosphorylation was observed after Fc␥RIIa cross-linking. Cbl was also strongly tyrosine-phosphorylated in Syb-activated platelets, although to a lesser level than after Fc␥RIIa crosslinking. Cbl was only minimally phosphorylated in thrombinactivated platelets (Fig. 3a). Densitometer scanning of the autoradiographs of five experiments confirmed this difference (Fig. 3b). Reprobing the membrane with anti-Cbl antibody (Fig.  3, lower panel) confirmed that the increase in tyrosine phosphorylation of Cbl was not the consequence of a change in the recovery of the protein. The addition of IV.3 alone did not induce any tyrosine phosphorylation. Preincubation of IV.3 to block the binding of Syb Fc domain to Fc␥RIIa totally inhibited Syb-induced tyrosine phosphorylation of Cbl. The results demonstrate that Cbl was heavily tyrosine-phosphorylated after Fc␥RIIa engagement and suggest a Cbl involvement in Fc␥RIIa-mediated platelet signaling.
To verify that the different levels of Cbl tyrosine phosphosphorylation (depending on the agonist used) were not attributable to different kinetics and to determine when Cbl phosphorylation occurred, we studied the time course of Cbl tyrosine phosphorylation during platelet stimulation. Strong Cbl tyrosine phosphorylation was already reached at 30 s after platelet activation induced by Fc␥RIIa cross-linking, with a plateau obtained between 1 and 2 min and a decrease to lower levels thereafter (Fig. 4, upper panel). Syb-induced Cbl tyrosine phosphorylation kinetics was similar to that observed after Fc␥RIIa  1 and 4), first and second anti-Cbl IPs (lanes 2 and 5), and first and second anti-PY IPs (lanes 3 and 6) were solubilized and resolved on SDS-PAGE, transferred to a nitrocellulose membrane, and blotted with anti-Cbl. Results are representative of two experiments. cross-linking (Fig. 4, middle panel), with or without a lag depending on the donor used. Indeed, the Fc␥RIIa His-Arg-131 polymorphism has been shown to play a crucial role in the ability of Fc␥RIIa to bind the Fc domain of mAb-IgG1 and, consequently, in the cell activation induced by these antibodies (44). We and others have shown that platelets from homozygous His donors respond more slowly than platelets from homozygous Arg donors to anti-CD9 antibodies, whereas the lag phase of platelets from heterozygous donors is intermediate (41,44). Thrombin induced a weak and slow Cbl tyrosine phosphorylation, which peaked between 2 and 5 min after platelet stimulation and decreased thereafter (Fig. 4, lower panel). The data suggest that Cbl was tyrosine-phosphorylated in the early stages of platelet activation and could participate in the first events triggered by Fc receptor engagement. During thrombin activation, Cbl would be involved to a lesser extent and in later stages after platelet activation.
Cbl Association with p85/PI 3-K in Activated Platelets-As an adaptor protein, tyrosine-phosphorylated Cbl was shown to associate with various signaling proteins. Among them, Cbl was reported in various cells to associate with PI 3-K (9,28,29,45) and to enhance PI 3-K activity (27,38). Because PI 3-K plays an important role in platelet function (46), we searched for an association between Cbl and PI 3-K in platelets. Cbl was immunoprecipitated from resting, and stimulated platelets and samples were analyzed with an anti-p85/PI 3-K antibody. In resting platelets, p85/PI 3-K was hardly detectable in the Cbl immunoprecipitates (Fig. 5). In contrast, p85/PI 3-K co-immunoprecipitated with Cbl in platelets activated through the Fc receptor. In thrombin-activated platelets, Cbl/PI 3-K associa-tion was insignificant (Fig. 5a). The amount of p85 protein present in the anti-Cbl IP of platelets activated by Fc␥RIIa cross-linking or by Syb reached a plateau between 30 s and 2 min of platelet stimulation and subsequently decreased back to resting level (Fig. 5b). Notably, the time course of Cbl/p85 association paralleled that of Cbl phosphorylation (Fig. 5, compare with Figs. 3 and 4), suggesting that Cbl/PI 3-K association was dependent on Cbl tyrosine phosphorylation.
To determine which domain of p85 was involved in Cbl/PI 3-K association, we studied Cbl binding to bacterial GST fusion proteins corresponding to full p85 (GST-P85), the SH3 domain of p85 (GST-SH3), or the N-and C-terminal SH2 domains of p85. Incubation of the different GST fusion proteins with platelet lysates showed that none of the two p85 SH2 domains (Cand N-terminal) bound Cbl (Fig. 6a). By contrast, an association of p85 (full) or p85 SH3 domain with Cbl was observed in both resting and activated platelet lysates. These data suggested an interaction between the p85 SH3 domain and the Cbl proline-rich region.
To determine whether Cbl/PI 3-K association was mediated by the p85 SH3 domain and the Cbl proline-rich region, we used competitive proline-rich peptides derived from p85 that were previously shown to bind the SH3 domain of p85 itself (43). The two competitive peptides strongly abolished Cbl association with GST-SH3 (p85) and GST-p85 (full). The inhibition was total when the two peptides were added together (Fig.  6b), indicating that, in vitro, the p85 SH3 domain mediates Cbl/p85 association through its interaction with the Cbl proline-rich region.
Inhibition of PI 3-K Activity Abolished the Platelet Responses after Fc␥RIIa Engagement-To determine whether PI 3-K plays a role in platelet activation mediated by the Fc receptor, we studied the effect of wortmannin (50 nM), an inhibitor of PI 3-K activity, on platelet aggregation and serotonin release. Platelet aggregation induced by Fc␥RIIa cross-linking or by Syb was strongly inhibited by wortmannin (100 and 88%, respectively; Fig. 7). By contrast, platelet aggregation induced by thrombin was poorly inhibited in the presence of wortmannin and became reversible. Serotonin release induced by Fc␥RIIa engagement was greatly inhibited by preincubation of platelets with wortmannin. Indeed, 85% inhibition of serotonin release was observed in platelets activated by cross-linking, and 70% inhibition was observed after activation by Syb. Wortmannin had no significant effect on thrombin-induced serotonin release (Fig. 7). The data suggest a key role for PI 3-K in Fc receptormediated platelet activation. If tyrosine-phosphorylated Cbl enhances PI 3-K activity, Cbl would also play an important role in platelet activation. Washed platelets were activated for 30 s to 10 min by cross-linking Fc␥RIIa using mAb IV.3 (10 g/ml) plus F(ab)Ј2 RAM (80 g/ml), by Syb (10 g/ml), or by thrombin (1 unit/ml). Platelets were then lysed by Nonidet P-40 containing buffer, and Cbl was immunoprecipitated. The immunoprecipitates were subjected to SDS-PAGE, followed by anti-phosphotyrosine immunoblotting. Results are representative of three experiments.

DISCUSSION
The adaptor protein Cbl has been identified in a variety of cells, including platelets, but its involvement in platelet signaling remains uncharacterized. The present work was devoted to studying the involvement of Cbl in signal transduction after platelet activation induced by Fc␥RIIa cross-linking or Syb antibody (anti-CD9), which activates platelets via Fc␥RIIa. We demonstrated strong and rapid tyrosine phosphorylation of Cbl in platelets activated through Fc␥RIIa. In addition, we showed that after platelet activation, p85/PI 3-K association with Cbl correlates with the intensity of Cbl tyrosine phosphorylation. Furthermore, we showed that the PI 3-K inhibitor wortmannin abolished antibody-mediated platelet responses, indicating a crucial role for PI 3-K in antibody-induced platelet activation.
Cbl was not significantly tyrosine-phosphorylated in resting platelets, but it became phosphorylated during platelet activation depending on the agonist used. After activation by Fc␥RIIa cross-linking, Cbl was strongly and rapidly tyrosine-phosphorylated. To a lesser extent, Syb induced a similar Cbl phosphorylation to that obtained after Fc␥RIIa cross-linking. The former difference in the Cbl phosphorylation was probably attributable to distinct modes of platelet activation induced by Syb and Fc␥RIIa cross-linking, as suggested by others (47). That specific binding of Syb antibody to its antigen (CD9), in the presence of IV.3 (anti-Fc␥R), did not induce Cbl phosphorylation indicates that Cbl tyrosine phosphorylation occurred after Fc␥RIIa engagement. These results suggest that, unlike thrombin, which induced a faint and slow Cbl tyrosine phosphorylation, Fc receptor engagement strongly involves Cbl at the first steps of platelet signaling. Another protein involved in the first steps of Fc␥RIIa-mediated signal transduction is the tyrosine kinase Syk (48). The latter could be a potential candidate to phosphorylate Cbl in platelets, because it was previously demonstrated to participate in Cbl phosphorylation in activated T cells (49). We could not, however, detect any Cbl association with Syk after platelet activation. Cbl tyrosine phosphorylation was transient, suggesting an action of tyrosine phosphatase(s) on phosphorylated Cbl. This is supported by the fact that in the presence of the protein tyrosine phosphatase inhibitor phenylarsine oxide, Cbl phosphorylation remained stable for up to 10 min of platelet activation (data not shown).
Because Cbl association with p85/PI 3-K has been suggested to increase PI 3-K activity in a number of cell systems (14,27,38), we examined the association of Cbl with PI 3-K in platelets. We found that Cbl/PI 3-K association was negligible in resting and thrombin-activated platelets. In contrast, Cbl was strongly associated with PI 3-K after Fc␥RIIa-mediated platelet activation. In in vitro experiments, using GST fusion proteins, we did not find any association between the N-SH2 or C-SH2 domains of p85 and Cbl in resting or Syb-activated platelets. In contrast, full p85 and the p85 SH3 domain precipitated Cbl from resting and Syb-activated platelet lysates. It is thus most likely that Cbl and p85 associate via the SH3 domain of p85 and the proline-rich region of Cbl.
We could not exclude, however, that in vivo Cbl/PI 3-K association may require tyrosine phosphorylation of Cbl. Indeed, our experiments favor a relationship between these two events. In Nb2 cells, constitutive Cbl/PI 3-K association is mediated by the Cbl proline-rich region and the p85 SH3 domain, whereas increased Cbl/p85 association was proposed to occur through both the p85 SH2 and SH3 domains after cell activation and Cbl tyrosine phosphorylation (27). In fact, both SH2 and SH3 domains of p85 interact with Cbl in other cells (19,22,33,50). Interestingly, Soltoff and Cantley (19) suggested that engagement of the p85 SH2 domain exposes the SH3 domain, which can then further interact with Cbl and increase the affinity of p85 for Cbl. The authors proposed that Cbl could act as an adaptor protein that recruits PI 3-K in the epidermal growth factor-mediated activation of PC12 cells (19). Moreover, Cbl has a Tyr-X-X-Met motif, which could associate with a p85 SH2 domain if phosphorylated on tyrosine (51). Alternatively, the tyrosine phosphorylation of Cbl could be necessary for its relocalization near PI 3-K. In that respect, Tanaka et al. (25) showed that in epidermal growth factor-activated macrophages and fibroblasts, Cbl tyrosine phosphorylation may be accompanied by its subcellular translocation. A last hypothesis could be that in resting platelets, the Cbl proline-rich region may not be in a conformation that allows its association with the p85 SH3 domain. After platelet activation and Cbl phosphorylation, a conformational change in Cbl would render possible the association between the two proteins. By analogy, in stimulated fibroblasts, a phosphotyrosine-dependent conformational change of Cbl was proposed to transiently expose the Cbl N-terminal region, permitting interaction with platelet-derived growth factor receptor ␣ (52).
To determine whether Cbl tyrosine phosphorylation and association with PI 3-K occurred before PI 3-K activation, we used wortmannin to inhibit PI 3-K activity. We found that wortmannin had no effect on Cbl phosphorylation or on Cbl/PI 3-K association induced by different agonists, which indicates that the two events occurred upstream of the lipid kinase activation (data not shown). However, a noncovalent association between PI 3-K and Fc␥RIIa has been previously shown (53). Therefore, tyrosine-phosphorylated Cbl would play a role in Fc␥RIIa-mediated platelet signaling by linking PI 3-K with the Fc receptor pathway, possibly by enhancing PI 3-K activity.
That the level of Cbl/p85 association was stronger after Fc receptor engagement than after thrombin addition suggests a differential role for PI 3-K in the signaling induced by the two types of platelet activation. In the presence of wortmannin, platelet aggregation mediated through Fc␥RIIa was abolished. In contrast, and as previously demonstrated with thrombin receptor activating peptide (54,55), platelet aggregation induced by thrombin was only partly inhibited and became reversible. These data indicate that PI 3-K participates in the control of platelet aggregation, especially that which occurs after Fc receptor engagement. The crucial role of PI 3-K in Fc␥RIIa-mediated platelet activation was also confirmed by the demonstration that wortmannin strongly inhibited antibody- induced serotonin release from dense granules but only weakly inhibited serotonin release induced by thrombin. These results demonstrated that in platelets, PI 3-K activation was required to initiate platelet responses after Fc␥RIIa engagement. Thus, if Cbl increases PI 3-K activity as previously proposed, Cbl would also play a crucial role in platelet activation mediated through Fc␥RIIa .
In conclusion, Cbl was strongly tyrosine-phosphorylated during Fc␥RIIa-mediated platelet activation, and levels of Cbl tyrosine phosphorylation paralleled levels of Cbl/PI 3-K association. Because PI 3-K activity appeared crucial in platelet responses dependent on Fc␥RIIa engagement, we suggest that Cbl participates in signal transduction mediated through the Fc receptor by enhancing PI 3-K activity. Thus, Cbl could be one of the key adaptor and regulator proteins in this system. FIG. 6. Binding of Cbl to GST fusion proteins of p85/PI 3-K. a, GST or GST fusion proteins containing full p85, p85 SH3 domain, or p85 N-terminal SH2 domain or C-terminal SH2 domain coupled to agarose beads were added to lysates of both resting and Syb-activated platelets (10 g/ml). The samples were resolved on SDS-PAGE and immunoblotted with anti-Cbl antibody. b, GST-p85 (full) and GST-SH3 (p85) were incubated 30 min with 0.1, 0.5, or 1 p85 proline-rich peptides (PRP1, 80 -96; PRP2, 300 -314 amino acids of p85 sequence) or 1 mM unrelated peptide (control P). The mixtures were then added to lysates of resting and activated platelets and incubated for an additional 1 h. The samples were resolved by SDS-PAGE and immunoblotted with anti-Cbl antibody. Results are representative of three experiments.