Platelet activation by von Willebrand factor requires coordinated signaling through thromboxane A2 and Fc gamma IIA receptor.

Interaction of von Willebrand Factor with glycoprotein Ib-IX-V induces platelet activation through a still poorly defined mechanism. Previous studies have suggested a possible role for the low affinity receptor for immunoglobulin, Fc gamma RIIA, in GPIb-IX-V signaling. Here we show that binding of vWF to platelets induces the tyrosine phosphorylation of Fc gamma RIIA by a Src kinase. Treatment of platelets with the anti-Fc gamma RIIA monoclonal antibody IV.3 specifically inhibits vWF-induced but not thrombin-induced pleckstrin phosphorylation and serotonin secretion. Moreover, vWF fails to induce pleckstrin phosphorylation in mouse platelets, lacking Fc gamma RIIA, and serotonin secretion is impaired. Pleckstrin phosphorylation and serotonin secretion in human platelets stimulated with vWF are blocked by the cyclooxygenase inhibitor acetylsalicylic acid. However, release of arachidonic acid and synthesis of TxA(2) induced by vWF are not affected by the anti-Fc gamma RIIA monoclonal antibody IV.3. Similarly, vWF-induced tyrosine phosphorylation of Fc gamma RIIA, as well as of Syk and PLC gamma 2, occurs normally in aspirinized platelets. Inhibition of the tyrosine kinase Syk by piceatannol does not affect vWF-induced tyrosine phosphorylation of Fc gamma RIIA but prevents phosphorylation of PLC gamma 2. Pleckstrin phosphorylation and platelet secretion induced by vWF, but not by thrombin, are also inhibited by piceatannol. Pleckstrin phosphorylation is also sensitive to the phosphatidylinositol 3-kinase inhibitor wortmannin. These results indicate that PLC gamma 2 plays a central role in platelet activation by vWF and that the stimulation of this enzyme requires coordinated signals through endogenous TxA(2) and Fc gamma RIIA.

von Willebrand Factor (vWF) 1 is a large glycoprotein synthesized by endothelial cells and megakaryocytes and plays an important role in platelet adhesion and thrombus formation (1). Under conditions of high shear stress or in the presence of modulators like ristocetin or botrocetin, vWF binds to the platelet membrane GPIb-IX-V and initiates signals leading to platelet activation (1). This process involves the activation of PLC and PLA 2 , the cytoskeleton reorganization and the interaction of several signaling molecules with the actin filaments, the tyrosine phosphorylation of several proteins, the activation of GPIIb-IIIa, and leads to platelet secretion and aggregation (2)(3)(4)(5)(6)(7). Some of these events, such as the activation of PLC, have been demonstrated to be promoted by the secondary action of TxA 2 produced from arachidonic acid through the cyclooxygenase pathway (2). For this reason, endogenous TxA 2 is believed to play a central role in platelet activation by vWF. However, other events, such as the activation of some tyrosine kinases, are independent of endogenous TxA 2 and are directly mediated by the interaction of vWF with GPIb-IX-V (4). Although considerable progress has been made in understanding the molecular basis of vWF interaction with GPIb-IX-V, very little is known about the mechanism by which such interaction initiates signals leading to platelet activation. GPIb-IX-V is a member of the leucine-rich glycoprotein gene family and is composed of four different polypeptides, GPIb␣, GPIb␤, GPIX, and GPV, at a ratio of 2:2:2:1 (8). The complex is not associated to GTP-binding proteins, does not possess tyrosine kinase activity, and is not tyrosine-phosphorylated upon binding of vWF. The cytoplasmic domain of GPIb-IX-V can interact with the cytoskeleton through actin-binding protein, and with the signaling protein 14-3-3, but the role of these molecules in GPIb-IX-V-mediated platelet activation is unclear (9,10). Recently, it has been reported that recruitment of GPIb-IX-V by vWF or by the snake venom lectin alboaggregin-A leads to the tyrosine phosphorylation of the ITAM sequence of the Fc receptor ␥-chain and to its interaction with the tyrosine kinase Syk, suggesting that immune receptors could participate in GPIb-IX-V signaling (11). In addition to the Fc receptor ␥-chain, human platelets express another ITAM-containing protein, the low affinity receptor for IgG, Fc␥RIIA (12). Several lines of evidence suggest that this 40-kDa transmembrane glycoprotein is involved in platelet activation following the recruitment of GPIb-IX-V. A physical association between the two receptors on the platelet membrane has been demonstrated by flow cytometric fluorescence energy transfer, co-immunoprecipitation, and yeast two-hybrid system (13,14). Moreover, evidence for a functional correlation between the two receptors arise from the observation that anti-GPIb antibodies could inhibit Fc␥RIIAmediated platelet aggregation (13). Finally, we have previously demonstrated that the blockage of Fc␥RIIA prevents vWFinduced translocation of the small GTP-binding proteins rap1b and rap2b to the cytoskeleton and the tyrosine phosphorylation of selected substrates including the tyrosine kinase Syk and PLC␥2 (7).
In this work we further investigate the functional interplay between Fc␥RIIA and GPIb-IX-V in vWF-stimulated platelets. We show that Fc␥RIIA is actually tyrosine phosphorylated upon vWF binding to platelets by a mechanism independent of TxA 2 production. Moreover, blockage of Fc␥RIIA by a specific mAb prevents PLC activation and platelets secretion induced by vWF under conditions in which the production of TxA 2 occurs normally. These results indicate that, in addition to endogenous TxA 2 , Fc␥RIIA plays an essential role in platelet activation by vWF.
Platelet Preparation and Stimulation-For immunoprecipitation studies, human platelets were isolated from freshly drawn blood by gel filtration on Sepharose CL-2B, and eluted with Hepes buffer (10 mM HEPES, 137 mM NaCl, 2.9 mM KCl, 12 mM NaHCO 3 , pH 7.4) as previously described (15). Platelet concentration was adjusted to 10 9 platelets/ml. For studies with labeled cells, the platelet pellet, obtained by centrifugation of the platelet-rich plasma, was resuspended in 135 mM NaCl, 2.7 mM KCl, 12 mM NaHCO 3 , 0.16 mM NaH 2 PO 4 , 2 mM MgCl 2 , 0.2 mM EGTA, 5.5 mM glucose, 0.34% bovine serum albumin, pH 6.5, at the concentration of 0.5 ϫ 10 9 platelets/ml. Labeling was with 0.2 mCi/ml 32 P for 90 min, 0.2 Ci/ml [ 14 C]serotonin for 30 min, or 1.5 Ci/ml [ 3 H]arachidonic acid for 120 min at room temperature. Platelets were then washed with the same buffer without glucose and EGTA and finally resuspended in Hepes buffer, pH 7.4, at the final concentration of 10 9 cells/ml. Platelet samples were stimulated at 37°C under constant stirring with 10 g/ml vWF in the presence of 0.5 mg/ml ristocetin or with 1 unit/ml thrombin for 3 min unless otherwise indicated. Direct crosslinking of Fc␥RIIA was achieved by incubation of platelets with the mAb IV.3 (2 g/ml) followed by addition of 30 g/ml of goat anti-mouse IgG. Preincubation of platelets with antibodies and inhibitors before addition of the agonists was as follows: 1 mM acetylsalicylic acid, or 10 M indomethacin for 30 min; 20 g/ml mAb IV.3 or 20 g/ml unrelated IgG for 3 min; 5, 10, or 20 g/ml piceatannol for 15 min; 10 M PP1 for 10 min; 100 nM wortmannin for 15 min. Control samples were incubated with the corresponding volume of buffer or Me 2 SO used as vehicle for piceatannol, PP1 and wortmannin.
Immunoprecipitation and Immunoblotting-Samples of resting and stimulated platelets (0.4 ml) were lysed by addition of an equal volume of immunoprecipitation buffer 2ϫ (100 mM Tris/HCl pH 7.4, 200 mM NaCl, 2 mM EGTA, 20 g/ml leupeptin, 20 g/ml aprotinin, 2 mM phenylmethylsulfonyl fluoride, 2 mM Na 3 VO 4 , 2 mM NaF, 2% Nonidet P40, 0.5% sodium deoxycholate). The cleared lysates were immunoprecipitated with 2 g of anti-Fc␥RIIA mAb IV.3 or with 2 g of the anti-Syk and anti-PLC␥2 antisera, as previously described (7). Upon electrophoresis on 10 or 7.5% acrylamide gels, immunoprecipitated proteins were transferred to nitrocellulose and probed with the antiphosphotyrosine antibody 4G10 as previously described (7). All the shown results are representative of at least three different experiments.
Measurement of Pleckstrin Phosphorylation-Samples of 32 P-labeled platelets (0.1 ml) treated with the indicated inhibitors and stimulated with vWF or with thrombin were lysed by addition of an equal volume of SDS-sample buffer (25 mM Tris, 192 mM glycine, pH 8.3, 4% SDS, 1% dithiothreitol, 20% glycerol, and 0.02% bromophenol blue), and heated at 95°C for 5 min. Aliquots of total platelet proteins (20 l) were separated by SDS-PAGE on 5-15% acrylamide gradient gels followed by staining with Coomassie Brilliant Blue. Gels were then dried, and phosphorylation of pleckstrin was evaluated upon autoradiography for about 16 h at Ϫ80°C.
Measurement of [ 14 C]Serotonin Secretion-[ 14 C]serotonin-labeled platelets (0.1-ml samples) were stimulated with the appropriated agonist in the presence of 5 M imipramine. A 10-l aliquot was withdrawn before addition of the agonist to evaluate the total incorporated radioactivity. Stimulation was stopped by addition of 0.9 ml of 1.2% paraformaldehyde and 100 mM EDTA and cooling on ice. Platelets were recovered by centrifugation at 10,000 ϫ g for 3 min, and the radioactivity of [ 14 C]serotonin released in the supernatant was determined by liquid scintillation counting.

Measurement of TxA 2 Production and [ 3 H]Arachidonic
Acid Release-Stimulation of platelet samples (0.1 ml) was stopped by addition of 0.9 ml of 10 mM EDTA and 10 M indomethacin. Samples were centrifuged at 3.000 ϫ g for 3 min, and supernatants were collected. Aliquots of 50 l were used for determination of TxB 2 , a stable breakdown product of TxA 2 , using a commercial enzyme immunoassay kit according to the manufacturer's instructions. Release of arachidonic acid was evaluated on 0.2-ml samples of [ 3 H]arachidonic acid-labeled platelets. Stimulation was stopped by addition of 0.8 ml of 1.2% paraformaldehyde and 100 mM EDTA. After centrifugation at 10,000 ϫ g for 3 min, the [ 3 H]arachidonic acid released in the supernatant was measured by liquid scintillation counting.
Studies with Mouse Platelets-Blood from anesthetized mice was collected from the inferior vena cava into syringes containing heparin solution (5 units/ml). Blood was diluted with 1 volume of Hepes buffer and centrifuged at 90 ϫ g for 10 min. For maximal platelet recovery, the red blood cell pellet was diluted with 1 ml of CGS (0.038% trisodium citrate, 0.6% glucose, 0.72% NaCl, pH 7.0), and centrifuged as above. Pooled diluted PRP was centrifuged at 150 ϫ g for 10 min, and platelets were washed with CGS containing 25 ng/ml PGE 1 . Platelets were finally resuspended in Hepes buffer at the final concentration of 10 9 cells/ml and rested for 30 min at 37°C before stimulation. Labeling was performed for 1 h at 37°C by incubating the PRP with 0.25 Ci/ml [ 14 C]serotonin or platelets resuspended in CGS with 0.2 mCi/ml 32 P. Samples of [ 14 C]serotonin-labeled platelets (0.05 ml) were stimulated with agonists from a 10-fold concentrated stock solution for 3 min under constant stirring. The reaction was stopped by brief centrifugation, and the [ 14 C]serotonin released in the supernatant was determined by liquid scintillation counting. Stimulation of 32 P-labeled platelets (0.05-ml samples) was performed for increasing times and was stopped by addition of an equal volume of SDS-sample buffer. Phosphorylation of pleckstrin was analyzed by SDS-PAGE on 10 -20% acrylamide gradient gels followed by autoradiography.

Tyrosine Phosphorylation of Fc␥RIIA in vWF-stimulated
Platelets-We have previously shown that Fc␥RIIA mediates the translocation of the small G-proteins rap1b and rap2b to the cytoskeleton and the tyrosine phosphorylation of selected substrates including Syk and PLC␥2 in vWF-stimulated platelets (7). To clarify the role of Fc␥RIIA in platelet stimulation by vWF, we examined whether Fc␥RIIA itself was actually activated upon stimulation of human platelets with vWF. Activation of Fc␥RIIA involves the phosphorylation of tyrosine residues in its cytoplasmic ITAM that enables the receptor to bind and activate Syk leading to the tyrosine phosphorylation of downstream substrates, such as PLC␥2 (12,16,17). Gel-filtered platelets were treated with buffer or stimulated with 10 g/ml vWF in the presence of 0.5 mg/ml ristocetin for 3 min. Samples were lysed and immunoprecipitated with the anti-Fc␥RIIA mAb IV.3 or with an unrelated serum. Control platelet samples were stimulated by direct cross-linking of Fc␥RIIA with 2 g/ml IV.3 and 30 g/ml goat anti-mouse IgG. The immunoprecipitated proteins were analyzed by immunoblotting with an anti-phosphotyrosine antibody. The results are shown in Fig. 1. The IV.3 mAb was able to immunoprecipitate a tyrosine-phosphorylated protein of about 40 kDa from vWFstimulated but not from resting platelets. This protein was not observed in the immunoprecipitates obtained with an unrelated serum from vWF-stimulated platelets. Moreover, a tyrosine-phosphorylated protein with a similar electrophoretic mobility and reactivity to anti-phosphotyrosine was immunoprecipitated by the IV.3 mAb from platelets stimulated by direct cross-linking of Fc␥RIIA. Therefore, we conclude that stimulation of platelets with vWF induces the tyrosine phosphorylation of Fc␥RIIA. Such phosphorylation, however, is clearly weaker than that produced by direct activation of the receptor, which most likely represents the maximal response.
We next tried to identify the kinase responsible for tyrosine phosphorylation of Fc␥RIIA induced by vWF. Gel-filtered platelets were preincubated with 20 g/ml of the Syk-selective inhibitor piceatannol or with 10 M of the Src kinases inhibitor PP1 and then stimulated with vWF and ristocetin. The tyrosine phosphorylation of immunoprecipitated Fc␥RIIA was evaluated by immunoblotting with an anti-phosphotyrosine antibody. Fig. 2 shows that vWF-induced tyrosine phosphorylation of Fc␥RIIA was not affected by piceatannol, but was completely inhibited by the preincubation of platelets with PP1. These results indicate that a Src kinase is responsible for the tyrosine phosphorylation of Fc␥RIIA in vWF-stimulated platelets. Moreover, the lack of effects of piceatannol is in agreement with the evidence that Syk lies downstream of the activated Fc␥RIIA (12,16).
It is known that many events related to platelet activation by vWF are actually mediated by TxA 2 produced from arachidonic acid through the cyclooxygenase pathway (2). To investigate the possible role of TxA 2 on vWF-induced tyrosine phosphorylation of Fc␥RIIA, we preincubated platelets with two different cyclooxygenase inhibitors, acetylsalicylic acid and indomethacin. Fig. 3 shows that neither of the inhibitors affected vWFinduced tyrosine phosphorylation of Fc␥RIIA. Moreover, tyrosine phosphorylation of Fc␥RIIA induced by vWF occurred normally in platelets preincubated with the TxA 2 receptor antagonist SQ29,548 (data not shown). Therefore, activation of Fc␥RIIA is directly promoted by vWF and does not require the action of produced TxA 2 .
We have previously shown that tyrosine phosphorylation of Syk and PLC␥2 induced by vWF is mediated by Fc␥RIIA (7). Because activation of Fc␥RIIA is independent of TxA 2 synthesis, we verified the effects of acetylsalicylic acid and indomethacin on Syk and PLC␥2 tyrosine phosphorylation. Fig. 4 shows that, similarly to Fc␥RIIA, the tyrosine phosphorylation of Syk and PLC␥2 induced by vWF did not require the production of TxA 2 .
PLC␥2 has been reported to represent a substrate for Syk in collagen-stimulated platelets (18). To investigate the possible correlation between Syk and PLC␥2 in vWF-stimulated platelets, we examined the agonist-induced tyrosine phosphoryla-tion of PLC␥2 in platelets preincubated with the Syk-selective inhibitor piceatannol. Fig. 5 shows that tyrosine phosphorylation of PLC␥2 induced by vWF was totally prevented when Syk was inhibited by piceatannol. These results identify a signaling pathway that is activated in vWF-stimulated platelets by a mechanism independent of TxA 2 and which involves Fc␥RIIA, the tyrosine kinase Syk, and PLC␥2.
Effect of the Blockage of Fc␥RIIA and of the Inhibition of Syk on vWF-induced PLC Activation and Platelet Secretion-We next investigated the role of Fc␥RIIA and Syk on two important events promoted by vWF binding to GPIb-IX-V: activation of PLC and serotonin secretion. Activation of PLC leads to the generation of diacylglycerol, which activates protein kinase C. The main platelet substrate for protein kinase C is the 47 kDa protein pleckstrin whose phosphorylation represents a reliable marker for PLC activation. Fig. 6 shows the phosphorylation of pleckstrin induced by vWF in 32 P-labeled platelets. Such phosphorylation was not affected by preincubation of intact platelets with control IgG but was almost totally blocked by the anti-Fc␥RIIA mAb IV. 3. Fig. 6 also shows that pleckstrin phosphorylation induced by thrombin was not affected by either control IgG or the IV.3 mAb.   We have shown that inhibition of the tyrosine kinase Syk by piceatannol prevented vWF-induced tyrosine phosphorylation of PLC␥2 (Fig. 5). As shown in Fig. 7, piceatannol also completely prevented vWF-induced activation of PLC, as revealed by the inability of the agonist to promote phosphorylation of pleckstrin. By contrast, piceatannol had no effects on pleckstrin phosphorylation induced by thrombin (data not shown). These results suggest that PLC␥2 represents the main PLC isoenzyme activated by vWF. It has been reported that tyrosine phosphorylation of PLC␥2 is not sufficient to support its activation, and that additional factors are required. Among these factors are the lipid products of PI 3-kinase (19,20). Fig. 8 shows that inhibition of PI 3-kinase by wortmannin totally prevented vWF-induced PLC activation and the consequent pleckstrin phosphorylation.
We next investigated the involvement of Fc␥RIIA and of the tyrosine kinase Syk on platelet secretion induced by vWF. In time course experiments, we found that release of [ 14 C]serotonin induced by vWF was rapid and maximal after 3 min of stimulation (data not shown). Pretreatment of platelets with the anti-Fc␥RIIA antibody IV.3 but not with unrelated IgG, totally inhibited the ability of vWF to induce release of [ 14 C]se-rotonin (Fig. 9). By contrast, the blockage of Fc␥RIIA had no effect on [ 14 C]serotonin release induced by thrombin (Fig. 9). The release of [ 14 C]serotonin induced by vWF, but not by thrombin was also inhibited in a dose-dependent manner by the Syk inhibitor piceatannol (Fig. 10). Therefore, both Fc␥RIIA and the tyrosine kinase Syk are involved in the signaling pathway leading to PLC activation and platelet secretion induced by vWF.
To clarify the physiological significance of Fc␥RIIA in platelet activation by vWF we analyzed pleckstrin phosphorylation and serotonin secretion in mouse platelets that can bind to vWF, but do not express Fc␥RIIA. Because mouse platelets have been shown to interact with human vWF activated by botrocetin but not by ristocetin (21), stimulation was performed with 10 g/ml vWF and 6 g/ml botrocetin. Fig. 11 shows that stimulation of mouse platelets with 1 unit/ml thrombin caused a rapid and strong phosphorylation of pleckstrin that was already maximal after 30 s. By contrast, treatment of mouse platelets with vWF for up to 10 min did not cause any significant phosphorylation of pleckstrin (Fig. 11). Even when 10-fold higher doses of vWF were used, no significant pleckstrin phos-   P-labeled platelets were preincubated with 20 g/ml piceatannol for 15 min or with an equal volume of Me 2 SO. Samples were stimulated with 10 g/ml vWF and 0.5 mg/ml ristocetin for the indicated times. Upon separation of platelet whole proteins by SDS-PAGE, phosphorylated pleckstrin (indicated by the arrow on the right) was visualized by autoradiography of the dried gels. phorylation was observed (data not shown). We next investigated the release of [ 14 C]serotonin. Treatment of mouse platelets with vWF caused the release of about 50% of the incorporated serotonin (Fig. 12). This response was clearly impaired when compared with that elicited by thrombin, which was able to promote the release of more than 90% of the incorporated serotonin. Interestingly, secretion of [ 14 C]serotonin induced by thrombin and vWF in human platelets was comparable and almost maximal (Fig. 12). Even treatment of platelets with 10-fold higher concentrations of vWF and botrocetin did not elicite a response comparable with that observed upon stimulation with thrombin or that measured in human platelets. These results indicate that vWF-induced serotonin secretion and PLC activation in mouse platelets lacking Fc␥RIIA are impaired.
Correlation among Fc␥RIIA, TxA 2 , PLC Activation, and Platelet Secretion-It has been previously shown that pleckstrin phosphorylation induced by vWF is inhibited by indomethacin, demonstrating that PLC activation is totally dependent on arachidonic acid metabolism through the cyclooxygenase pathway (2). We confirmed these results, and we found that pretreatment of platelets with the cyclooxyge-nase inhibitor acetylsalicylic acid totally prevented pleckstrin phosphorylation induced by vWF (Fig. 13A). Moreover, as shown in Fig. 13B, the release of [ 14 C]serotonin induced by vWF was blocked in acetylsalicylic acid-treated platelets. Finally, vWF-induced serotonin secretion was also inhibited by the TxA 2 receptor antagonist SQ29,548 (data not shown).  11. Analysis of pleckstrin phosphorylation in mouse platelets. 32 P-labeled mouse platelets were stimulated with 10 g/ml vWF and 6 g/ml botrocetin (vWF), or with 1 unit/ml thrombin (THR) for the indicated times. Total platelet proteins were separated on 10 -20% acrylamide gradient gels, and phosphorylation of pleckstrin (indicated by the arrow on the right) was visualized by autoradiography.
Taken together, these results indicate that PLC activation and platelet secretion induced by vWF require not only the Fc␥RIIA, but also the action of TxA 2 . Therefore, we investigated the correlation between the Fc␥RIIA-and TxA 2 -dependent signaling pathways. We have shown in this study that activation of Fc␥RIIA by vWF is independent of the production of TxA 2 (Fig. 3). We examined whether the signaling pathway initiated by activation of Fc␥RIIA could lead to the release of arachidonic acid by PLA 2 and to its conversion to TxA 2 . A significant accumulation of the stable metabolite TxB 2 in vWF-stimulated platelets versus resting platelets was measured by an immunoassay method (Fig. 14A). However, the accumulation of TxB 2 was not reduced when Fc␥RIIA was blocked by the IV.3 mAb. We also measured directly the release of arachidonic acid from [ 3 H]arachidonate-labeled platelets (Fig. 14B). Even in this case no significant differences in the release of [ 3 H]arachidonate were observed in platelets stimulated with vWF in the absence or in the presence of the anti-Fc␥RIIA mAb IV.3. Therefore, we conclude that the generation of TxA 2 induced by vWF is independent of Fc␥RIIA activation.

DISCUSSION
Stimulation of PLC plays a central role in platelet activation by extracellular agonists and is essential to promote the release of internal granules. Platelet stimulation by vWF induces the activation of PLC by a mechanism that has been recognized to be totally dependent on the release of arachidonic acid and its conversion to TxA 2 (2). In this work we have demonstrated that production of TxA 2 is necessary, but not sufficient for vWF to induce PLC activation and granule secretion. We have described the activation of an additional signaling pathway in vWF-stimulated platelets, involving the membrane Fc␥RIIA, the tyrosine kinase Syk and PLC␥2. Such a pathway, in addition to the generation of TxA 2 , is absolutely required for the agonist-induced PLC activation and granule secretion. These data confirm the essential role of Fc␥RIIA in mediating platelet response to vWF. Such a role has been initially hypothesized based on the association between Fc␥RIIA and GPIb-IX-V, the main vWF receptor on the platelet membrane (13,14). We have previously shown that vWF-induced translocation of the small GTP-binding proteins rap1b and rap2b to the cytoskeleton as well as tyrosine phosphorylation of selected substrates, including Syk and PLC␥2 are dependent on Fc␥RIIA (7). In the   FIG. 13. Effect of acetylsalicylic acid on vWF-induced pleckstrin phosphorylation and serotonin secretion. A, 32 P-labeled platelets were incubated with or without acetylsalicylic acid (ASA) for 30 min and then stimulated with 10 g/ml vWF and 0.5 mg/ml ristocetin for the indicated times. Total platelet proteins were separated on 5-15% acrylamide gradient gel, and phosphorylation of pleckstrin (indicated by the arrow on the right) was evaluated upon autoradiography of the dried gel. B, [ 14 C]serotonin-labeled platelets were treated without (Ϫ) with (ϩ) acetylsalicylic acid (ASA) and then stimulated with 10 g/ml vWF and 0.5 mg/ml ristocetin for 3 min. The release of [ 14 C]serotonin in the supernatant is expressed as percentage of the total incorporated radioactivity. Results are the mean Ϯ S.D. of three separate experiments. present work, we have provided evidence for a direct tyrosine phosphorylation (i.e. activation) of Fc␥RIIA upon binding of vWF to its receptor. Phosphorylation of Fc␥RIIA induced by vWF does not require the secondary action of synthesized TxA 2 and is probably mediated by a tyrosine kinase belonging to the Src family. Although the identity of this kinase is not known, possible candidates include Fyn and Lyn, which have been proposed to be involved in Fc receptor ␥-chain phosphorylation induced by collagen (22).
The molecular mechanism of Fc␥RIIA activation following vWF binding to platelets is unclear. Normally Fc␥RIIA is activated upon clusterization induced by immunocomplexes through the phosphorylation of two tyrosine residues in its intracellular ITAM (12). vWF does not directly bind to Fc␥RIIA, and we have previously demonstrated that its ability to initiate platelet activation through Fc␥RIIA is not because of potential contaminating immunoglobulins (7). However, vWF is a multimeric adhesive protein and its ability to stimulate platelets is correlated to the size of the multimers. Because of its multimeric nature, the platelet-activating vWF is most likely able to promote the clusterization of its receptor, GPIb-IX-V, on the platelet surface. Based on the reported physical proximity between GPIb-IX-V and Fc␥RIIA, it is possible for Fc␥RIIA itself to be incorporated into these clusters and, as a consequence of this, become activated. In this context, the inhibitory effect of the anti-Fc␥RIIA mAb IV.3 on platelet activation reported in the present work, may derive from an interference with the clustering process as a consequence of its binding to the extracellular region of the receptor. In preliminary experiments, we have actually found that activation of Fc␥RIIA (i.e. tyrosine phosphorylation) in vWF-stimulated platelets is prevented by binding of the mAb IV.3 (data not shown).
It is very well known that the activated Fc␥RIIA binds to the tandem SH2 domains of Syk through the phosphotyrosine residues in its ITAM (12,16). Bound Syk becomes activated, undergoes autophosphorylation and may promote the phosphorylation of downstream substrates, including PLC␥2 (18,23). We have provided evidence for this signaling pathway to be activated in vWF-stimulated platelets. In a previous study we have shown that vWF-induced tyrosine phosphorylation of Syk and PLC␥2 is regulated by Fc␥RIIA and can be prevented by the blockage of the receptor with the mAb IV.3 (7). Here we demonstrate that, similarly to Fc␥RIIA, the tyrosine phosphorylation of Syk and PLC␥2 is directly promoted by vWF and does not require the production of TxA 2 . Moreover, inhibition of Syk by piceatannol has no effect on the tyrosine phosphorylation of Fc␥RIIA but completely blocks the phosphorylation of PLC␥2 in vWF-stimulated platelets, suggesting that, as expected, Syk may link activated Fc␥RIIA to PLC␥2.
The impact of this signaling pathway on PLC activation and platelet secretion induced by vWF was investigated by blocking either the Fc␥RIIA with the mAb IV.3 or the tyrosine kinase Syk with the selective inhibitor piceatannol, and by analyzing mouse platelets that lack Fc␥RIIA. PLC activation was monitored by measuring the phosphorylation of pleckstrin, the main substrate for diacylglycerol-activated PKC. This approach is an indirect, but reliable method to measure PLC activation and has been exploited by many investigators. The results presented in this work indicate that both blockage of Fc␥RIIA and inhibition of Syk totally prevent vWF-induced PLC activation and platelet secretion. By contrast neither treatment affects the same cellular responses evoked by thrombin. Although piceatannol has been widely used in the past to inhibit the tyrosine kinase Syk, its specificity has been recently questioned (24). In this study, piceatannol was used to prevent PLC␥2 phosphorylation, an event that is very well known to be mediated by Syk (18). Moreover, the lack of effects of piceatannol on Fc␥RIIA phosphorylation induced by vWF, and on PLC activation and platelet secretion induced by thrombin, adds selectivity to the effects of this inhibitor.
We found that in mouse platelets lacking Fc␥RIIA, vWF was unable to stimulate pleckstrin phosphorylation and promoted a reduced release of serotonin. Interestingly, although mouse platelets have been found to bind and to aggregate in response to human vWF (21), very few data are available on cell activation induced by this agonist (25). A recent report documented the expression of P-selectin in mouse platelets treated with ristocetin (26). This is consistent with our findings that vWF can actually induce some release of serotonin in mouse platelets. Despite this, the compared analysis of vWF-and thrombin-induced secretion in mouse and human platelets revealed that this response is clearly impaired in mouse platelets lacking Fc␥RIIA. Such defective activation of mouse platelets by vWF is even more evident from the analysis of pleckstrin phosphorylation, which is almost undetectable. These results strengthen the importance of Fc␥RIIA on the platelet surface for efficient activation by vWF. The signaling pathway responsible for serotonin secretion induced by vWF in mouse platelets could be related to the Fc receptor ␥-chain that is expressed in these cells and has been proposed to play a role in vWF stimulation of human platelets (11). It is also possible that the contribution of the Fc receptor ␥-chain in vWF-induced activation of mouse platelets may be more relevant than in human cells, thus partially compensating for the absence of Fc␥RIIA.
The complete inhibition of pleckstrin phosphorylation and granule secretion by the mAb IV.3 or by piceatannol in human platelets suggests that PLC␥2 may be the main PLC isoenzyme activated by vWF. It is known that activation of PLC␥2 requires, in addition to its tyrosine phosphorylation, a number of factors, including the lipid products of PI 3-kinase (19,20). In this context, it is interesting to note that inhibition of PI 3-kinase by wortmannin completely blocks vWF-mediated pleckstrin phosphorylation.
Previous studies have demonstrated that PLC activation induced by vWF is totally dependent on the secondary action of endogenous TxA 2 (2). We also have confirmed the essential role of TxA 2 in platelet response to vWF, because we have shown in this work that inhibition of cyclooxygenase totally prevents pleckstrin phosphorylation as well as granule secretion. However, our results stimulate a revaluation of the role of TxA 2 in platelet activation by vWF. First of all, we have shown that tyrosine phosphorylation of Fc␥RIIA, Syk, and PLC␥2 induced by vWF occurs independently of TxA 2 synthesis. Moreover, the total inhibition of vWF-induced PLC activation and granule secretion in platelets pretreated with the mAb IV.3, under conditions in which production of TxA 2 is not affected, indicates that the endogenous TxA 2 is not per se sufficient to support platelet activation when the Fc␥RIIA-mediated signaling pathway here described is blocked. Such inability of endogenous TxA 2 to fully support platelet activation may be related to its low concentration. Based on the measurement of the amount of accumulated TxB 2 we calculated that endogenous TxA 2 reached the concentration of about 50 nM. It is possible that such concentrations allow TxA 2 to behave like a weak agonist that needs costimulatory factors to elicit its effects. The lack of effects of the mAb IV.3 on arachidonic acid release and TxB 2 accumulation induced by vWF, together with the evidence that tyrosine phosphorylation of Fc␥RIIA (as well as of Syk and PLC␥2) is not affected by inhibition of cyclooxygenase, indicate that TxA 2 -and Fc␥RIIA-mediated signaling pathways are independent events, both of which are directly promoted by bind-ing of vWF to its platelet receptor. We have shown here that a Src kinase is responsible for Fc␥RIIA phosphorylation, and we have obtained evidence that the Src kinase inhibitor PP1 is also able to prevent vWF-induced release of arachidonic acid from labeled platelets (data not shown). A central role for the Src tyrosine kinases in platelets activation promoted by recruitment of GPIb-IX-V has also been suggested by the evidence that PP1 inhibits alboaggregin-induced platelet secretion and aggregation as well as phosphorylation of PLC␥2 (11). Therefore, it seems most likely that binding of vWF to GPIb-IX-V stimulates by a still unknown mechanism a Src kinase, which simultaneously promotes the activation of Fc␥RIIA, leading to PLC␥2 tyrosine phosphorylation and the activation of PLA 2 , leading to the release of arachidonic acid and to its conversion to TxA 2 . Integration of signals generated by the activated Fc␥RIIA and the synthesized TxA 2 is required to elicit full platelet activation. The molecular mechanism of this integration is not known. One possibility arises from the consideration that activation of PLC␥2 requires the meeting of several events and factors, including the lipid products of PI 3-kinase and the proteins LAT and SLP-76 (19,20,27,28). Although we have demonstrated that phosphorylation of PLC␥2 induced by vWF is dependent on Fc␥RIIA-mediated signaling, other factors necessary for the full activation of the enzyme could be recruited specifically through the action of endogenous TxA 2 . This possibility, however, deserves further investigation.
In conclusion, our results demonstrate for the first time that the release of endogenous TxA 2 is required but not sufficient to support full platelet activation induced by vWF and recognize a novel essential role for Fc␥RIIA in this process.