Lipid Products of Phosphoinositide 3-Kinase and Phosphatidylinositol 4′,5′-Bisphosphate Are Both Required for ADP-dependent Platelet Spreading*

We have shown previously that ADP released upon platelet adhesion mediated by αIIbβ3integrin triggers accumulation of phosphatidylinositol 3′,4′-bisphosphate (PtdIns-3,4-P2) (Gironcel, D., Racaud-Sultan, C., Payrastre, B., Haricot, M., Borchert, G., Kieffer, N., Breton, M., and Chap, H. (1996) FEBS Lett. 389, 253–256). ADP has also been involved in platelet spreading. Therefore, in order to study a possible role of phosphoinositide 3-kinase in platelet morphological changes following adhesion, human platelets were pretreated with specific phosphoinositide 3-kinase inhibitors LY294002 and wortmannin. Under conditions where PtdIns-3,4-P2synthesis was totally inhibited (25 μm LY294002 or 100 nm wortmannin), platelets adhered to the fibrinogen matrix, extended pseudopodia, but did not spread. Moreover, addition of ADP to the medium did not reverse the inhibitory effects of phosphoinositide 3-kinase inhibitors on platelet spreading. Although synthetic dipalmitoyl PtdIns-3,4-P2 and dipalmitoyl phosphatidylinositol 3′,4′,5′-trisphosphate restored only partially platelet spreading, phosphatidylinositol 4′,5′-bisphosphate (PtdIns-4,5-P2) was able to trigger full spreading of wortmannin-treated adherent platelets. Following 32P labeling of intact platelets, the recovery of [32P]PtdIns-4,5-P2 in anti-talin immunoprecipitates from adherent platelets was found to be decreased upon treatment by wortmannin. These results suggest that the lipid products of phosphoinositide 3-kinase are required but not sufficient for ADP-induced spreading of adherent platelets and that PtdIns-4,5-P2 could be a downstream messenger of this signaling pathway.

Platelets play a key role in hemostasis by their capacities to adhere and to aggregate in response to vascular injury. The most abundant platelet integrin, the ␣ IIb ␤ 3 complex, is largely responsible for platelet aggregation after binding of soluble fibrinogen. Moreover, ␣ IIb ␤ 3 integrin is required for a complete and irreversible platelet adhesion to the subendothelial matrix (1). In this case, its preferential ligand is the von Willebrand factor, but under certain conditions fibrinogen or fibrin could also act as adhesion substrates. In its resting state, ␣ IIb ␤ 3 integrin is able to recognize immobilized fibrinogen. However, the interaction of soluble fibrinogen with ␣ IIb ␤ 3 complex requires a previous conformational change of the integrin due to an inside-out signaling pathway. When platelets adhere in vitro to a fibrinogen matrix, they undergo several irreversible morphological changes such as rounding and spreading. These responses are sustained by a cytoskeletal reorganization including extension of filopodia, lamellipodia, and controlled orientation of stress fibers. It has been shown that a concomitant granular secretion of ADP from adherent platelets was necessary for spreading (2), and that it controlled specific signals, i.e. p125 FAK and PtdIns 3-kinase 1 activations (2,3). Data from Haimovich et al. (4) show that tyrosine phosphorylation of p125 FAK tyrosine kinase seems to be correlated with cell spreading upon platelet adhesion to a fibrinogen matrix. On the other hand, PtdIns 3-kinase activity has been involved in cytoskeletal rearrangements occurring during cell motility or platelet aggregation (5)(6)(7)(8)(9). Moreover, studies in whole cells have demonstrated an association of PtdIns 3-kinase with p125 FAK (10,11) and the small G proteins Rac and Cdc42 (12), all of them being involved in the regulation of cytoskeleton organization. Taking advantage of specific PtdIns 3-kinase inhibitors, LY294002 and wortmannin (8,13,14), we herein demonstrate that PtdIns 3-kinase is involved in the ADP-signaling pathway that controls platelet spreading. Nevertheless, our results suggest that PtdIns-4,5-P 2 , a phospholipid tightly associated with actin-binding proteins in focal contacts and a key regulator of actin polymerization (15), could be a downstream messenger of this signaling pathway.
Preparation of Platelets-Human platelets were isolated from fresh platelet concentrates (Centre Régional de Transfusion Sanguine, Toulouse, France) by centrifugation as described previously (16). All washing procedures were performed at 37°C in the presence of apyrase (1 unit/ml) as an ADP scavenger. In some experiments, platelet-rich plasma was incubated with 100 M aspirin for 20 min to block cyclooxygenase activity. Platelets were labeled for 90 min with 0.4 mCi/ml [␥-32 P]phosphate (Amersham Pharmacia Biotech, Bucks, United Kingdom), as described previously (16). They were finally resuspended in modified Tyrode's buffer (pH 7.4) containing 2.5 mM CaCl 2 .
Cell Adhesion Assays and Lipid Extract Analysis-Cell culture flasks (75 cm 2 , Greiner Labortechnik, Poitiers, France) were precoated or not (control) with 100 g/ml of fibrinogen and were then blocked with fatty acid-free bovine serum albumin (3). The cell adhesion assay was performed using 5 ml of human platelets (3 ϫ 10 7 platelets/ml) that were added for 60 min at 37°C to the fibrinogen-coated flasks or to the control flasks. In some experiments, the ADP scavenger pyruvate kinase plus phosphoenolpyruvate (14.3 units/ml and 1 mM, respectively; 10 min) or the protein kinase C (PKC) inhibitor GF109203X (12 M; 60 min) or the PtdIns 3-kinase inhibitors LY294002 (0 -25 M; 10 min) or wortmannin (0 -100 nM; 15 min) were added to the platelet suspension before adhesion. GF109203X, wortmannin, and LY294002 were dissolved in Me 2 SO, which did not exceed 0.06% (v/v). Recovering of adherent cells, evaluation of the extent of cell adhesion, and lipid extract analysis by HPLC were performed as described previously (3).
Optical Microscopy-At the end of the adhesion step or the spreading restoration assay, unattached platelets were removed by washing with PBS and the buffer was replaced by 1% glutaraldehyde in 0.1 M Na 2 HPO 4 . Fixation was continued at room temperature for 15 min. After washing, adherent platelets were examined by interference light microscopy with a Reichert EMF4 microscope. Micrographs were taken at original magnification ϫ1250.
Immunoprecipitation of Talin-Adherent platelets (4.5 ϫ 10 8 platelets) were scraped off at 4°C in a lysis buffer containing 20 mM Tris-HCl, pH 8, 137 mM NaCl, 10% glycerol, 1 mM Na 3 VO 4 , 1 mM PMSF, 10 M pepstatin, 10 g/ml leupeptin, and 1% (v/v) Triton X-100. Resting platelets in suspension (4.5 ϫ 10 8 platelets) were centrifuged and resuspended in 600 l of the lysis buffer. After sonication (20 kHz for 2 ϫ 10 s) and centrifugation (12,000 ϫ g for 10 min at 4°C), the soluble fraction was collected and subsequently precleared for 30 min at 4°C with protein G-Sepharose 4B fast flow (Sigma). Precleared suspensions were then incubated overnight at 4°C with the polyclonal anti-talin antibody prepared as described previously at a 1:50 dilution (17). Capture of immune complex was performed by adding 50 l of protein G-Sepharose 4B fast flow. The immunoprecipitates were then washed once with PBS without calcium and magnesium, supplemented with anti-proteases as described above and 0.1% (v/v) Triton X-100, and twice with the same buffer without Triton.
Lipid Extraction and Western Blotting on Talin Immunoprecipitates-For lipid analysis, immunoprecipitation of talin was performed after plating of 32 P-labeled platelets as described above. Lipids were extracted as described previously (3) and separated by TLC following the procedure established by Pignataro and Ascoli (18). Briefly, lipid extracts were applied on oxalate-EDTA-impregnated silica gel plates, which were developed twice for 120 min with CHCl 3 , CH 3 OH, 9.15 M NH 4 OH (40:40:15). Individual lanes containing commercial standards PtdIns-4-P or PtdIns-4,5-P 2 were stained with iodine vapors. After exposition of the plates for 3-7 days, the radioactive spots were visualized and quantitated by a PhosphorImager 445 SI (Molecular Dynamics, Inc). Quantification was also performed after scraping the appropriate areas of the plate and counting in a liquid scintillation counter.
For protein analysis, anti-talin immunoprecipitates were solubilized, separated on 7.5% SDS-polyacrylamide gels, and blotted onto nitrocellulose as described previously (11). Immunodetection of talin was performed with the mouse monoclonal anti-talin antibody 8d4 from Sigma. Antibody reaction was visualized using the ECL chemiluminescence system (Amersham Pharmacia Biotech). Quantification of the different bands was performed by a densitometric analysis, which determines the pixel volume in each area (Gel Doc 1000, Bio-Rad).

FIG. 1. Dose-dependent inhibition of adhesion-induced [ 32 P]-PtdIns-3,4-P 2 synthesis by wortmannin and LY294002.
Washed platelets were pretreated with wortmannin (15 min) or LY294002 (10 min) at different concentrations and then were plated on the fibrinogen matrix for 60 min. After washing, adherent cells were scraped off and their lipid extract was analyzed by HPLC after deacylation. [ 32 P]PtdIns-3,4-P 2 was quantified as described under "Experimental Procedures." pendent manner with 80% inhibition achieved at 12 M and 50 nM, respectively. The production of [ 32 P]PtdIns-3,4-P 2 was totally abrogated at 25 M LY294002 and 100 nM wortmannin, at which concentrations platelet adhesion was not significantly affected (Table I). At these concentrations, among other phosphoinositides (PtdIns, PtdOH, PtdIns-4-P, and PtdIns-4,5-P 2 ), only the PtdIns-4-P level was found to be somewhat decreased, but not significantly in comparison with control Me 2 SO-treated platelets (Fig. 2).
PtdIns 3-Kinase Activation Is Necessary for ADP-dependent Platelet Spreading-As described previously (21), upon adhesion, adherent platelets undergo the following steps of morphological changes: disk to sphere shape change, extension of pseudopodia, and a much slower process, cell spreading (Fig. 3, A  and B). Pretreatment of platelets with wortmannin or LY294002 inhibited platelet spreading on fibrinogen (Fig. 3, C  and D). However, it should be noted that, although pretreated platelets did not fully spread, they still extended pseudopodia. The inhibitory effect of LY294002 and wortmannin on cell spreading was already detectable at 6 M and 25 nM, respectively. Concentrations of 25 M LY294002 and 100 nM wortmannin completely prevented platelet spreading. After removing wortmannin and LY294002 from the adhesion medium by two washes, we observed that only the inhibitory effect of LY294002 was reversible after 30 min (data not shown). Indeed, LY294002 is a competitive inhibitor at the ATP-binding site of PtdIns 3-kinase (13), whereas wortmannin induces a covalent modification of the catalytic site of the enzyme (14). In agreement with Haimovich et al. (2,4), pretreatment of platelets with the ADP scavenger pyruvate kinase plus phosphoenolpyruvate just before the adhesion assay induced the same effects as treatment with the PtdIns 3-kinase inhibitors, i.e. absence of spreading but persistence of pseudopodal extension (Fig. 3E). After washing, ADP (20 M) was added to adherent platelets to overcome the ADP scavenging system. Addition of ADP restored full spreading of all adherent platelets as shown in Fig. 3F. On the other hand, addition of 20 M ADP to adherent platelets pretreated with wortmannin did not reverse inhibition of platelet spreading (Fig. 3G). These data demonstrate that PtdIns 3-kinase signaling pathway is required for ADP-induced spreading of adherent platelets.
PtdIns-4,5-P 2 but Not Lipid Products of PtdIns 3-Kinase Is Sufficient for a Full Platelet Spreading-Our previous measurements of PtdIns 3-kinase activation in adherent platelets (3) have shown that, although the PtdIns-3,4,5-P 3 level was not significantly modified between 5 and 30 min of adhesion, Pt-dIns-3,4-P 2 accumulated as a function of the adhesion time. Time course of PtdIns-3,4-P 2 production closely paralleled platelet spreading upon adhesion (data not shown). In order to determine whether products of PtdIns 3-kinase could be involved in platelet spreading, we used Di-C 16 -PtdIns-3,4-P 2 and Di-C 16 -PtdIns-3,4,5-P 3 , which were reported to trigger biologic responses when added to whole cells (7,22). As shown in Fig.  4A, addition of Di-C 16 -PtdIns-3,4-P 2 (20 M) on adherent platelets pretreated with wortmannin only partially restored platelet spreading. After 30 min of incubation with Di-C 16 -PtdIns-3,4-P 2 , some adherent platelets have lost their round shape and have undergone pseudopodal and hyalomere extension. Nevertheless, these modifications concerned only a small proportion of adherent platelets (5%), as compared with 35% of spread platelets obtained with non-pretreated control cells (Fig. 4B). Using amounts of Di-C 16 -PtdIns-3,4-P 2 between 10 and 30 M, we obtained a dose-dependent increase in the response rate (detected as early as 15 min of adhesion) and in the proportion of responsive cells (Fig. 4B). In no case did we observe a full spreading of platelets, even after 60 min of incubation. Addition of Di-C 16 -PtdIns-3,4,5-P 3 or PtdIns-4-P or both (data not shown) together with Di-C 16 -PtdIns-3,4-P 2 was not more efficient in restoring full spreading (Fig. 4B). Surprisingly, addition of PtdIns-4,5-P 2 to adherent platelets pretreated with wortmannin triggered full spreading (Fig. 4, A and B). Moreover, the number of fully spread platelets was increased by addition of both Di-C 16 -PtdIns-3,4-P 2 and PtdIns-4,5-P 2 (Fig.  4B). Nevertheless, under these conditions, the amount of fully spread platelets was far below that observed in the control situation. PtdIns-4,5-P 2 effects on cell spreading are unlikely to be dependent on induction of platelet release reaction, as ADP (20 M), LPA (5 M), the thrombospondin-1 cell binding domain peptide (H-RFYVVMWK-OH; 50 M), could not restore spreading of platelets treated with 100 nM wortmannin (Table II). By contrast, the TXA 2 analog U46619 (5 M) triggered full spreading of adherent platelets pretreated with 100 nM wortmannin. However, addition of PtdIns-4,5-P 2 (20 M) to adherent platelets pretreated with aspirin (100 M) and wortmannin (100 nM) was still able to restore full platelet spreading (Table II).
The fact that strong platelet agonists, such as thrombin (1 unit/ml), the TXA 2 analog U46619, and the PKC activator PMA (10 nM), were able to restore spreading of platelets pretreated with 100 nM wortmannin (Table II) might suggest a PtdIns 3-kinase-independent pathway of platelet spreading. In order to determine a possible activation of PKC isoforms by high concentrations of PtdIns-4,5-P 2 (23), we used the specific PKC inhibitor GF109203X (24). Preincubation of platelets with GF109203X (12 M) induced inhibition of platelet spreading as described previously by Haimovich and co-workers (4). When platelets had been pretreated by both GF109203X (12 M) and wortmannin (100 nM) together, exogenous PtdIns-4,5-P 2 was still able to restore full spreading of adherent platelets with a similar efficiency as in the case of a pretreatment with wortmannin alone (Table II). These experiments argue in favor of a specific PtdIns-4,5-P 2 effect on the restoration of cell spreading that is independent on PKC activation. In some experiments, the ADP scavenger pyruvate kinase plus phosphoenolpyruvate was added to the platelet suspension before adhesion (E and F) as described under "Experimental Procedures." Adhesion was performed for 60 min, and after two washes, ADP (20 M) was added to the adherent platelets for 30 min (F and G). After fixation, adherent cells were observed by interference light microscopy.

FIG. 4. Effects of different phosphoinositides on spreading of platelets pretreated with wortmannin.
A, platelets were incubated in presence of 0.03% Me 2 SO (A) or 100 nM wortmannin (W, PtdIns-3,4-P 2 , PtdIns-4,5-P 2 ) as described previously in the legend to Fig. 2. Adhesion was performed for 60 min and after two washes, Di-C 16 -PtdIns-3,4-P 2 or PtdIns-4,5-P 2 (20 M) was added to the adherent platelets for 30 min as described under "Experimental Procedures." After fixation, adherent cells were observed by interference light microscopy. Platelets in a fully spread stage are indicated by large arrowheads, whereas those in a partial spread stage are indicated by thin arrowheads. B, different lipids were prepared and added to adherent platelets pretreated with wortmannin (W) as described under "Experimental Procedures." Adherent platelets pretreated with 0.03% Me 2 SO (A) were taken as a control. The histogram was used to express the percentages of platelets in a partial spread stage (black box) or in a full spread stage (hatched box). One hundred randomly chosen platelets were counted from each experiment (n ϭ 2-6), and each experiment has a paired control.
The partial rescue of platelet spreading with lipids suggests that there may be additional components to the PtdIns 3-kinasedependent pathway. We addressed the possibility that the myosin light chain kinase (MLCK) could be involved in platelet spreading. Indeed, inhibitors of MLCK have been shown to disassemble focal adhesions and to reduce their phosphotyrosine staining (25), and wortmannin is a MLCK inhibitor, but at micromolar concentrations (26). Preincubation of platelets with 10 M ML-7 (specific inhibitor of MLCK from Biomol; Ref. 27) during the 30 min before adhesion did not modify platelet spreading (data not shown). We concluded that MLCK does not seem to be involved in platelet spreading.
In Vivo Association of PtdIns-4,5-P 2 with Talin Is Decreased upon Wortmannin Treatment-Since the level of [ 32 P]PtdIns-4,5-P 2 was not significantly modified in vivo upon treatment of platelets with wortmannin, we looked for possible changes of the lipid content of focal adhesion sites. Talin is a prominent actin-binding protein that links integrins and the cytoskeleton and has been shown to be required for cell spreading (28,29). Although some variations could be observed in the amount of talin immunoprecipitated from platelets (Fig. 5A), mean values obtained from two experiments were essentially the same (Fig.  5B). Under these conditions, only traces of PtdIns-4-P were recovered in anti-talin immunoprecipitates from resting platelets, whereas, after adhesion, the amount of PtdIns-4-P increased, PtdIns-4,5-P 2 and traces of PtdIns-3,4-P 2 becoming detectable (Fig. 5C). For instance, the amount of talin isolated from activated platelets was 1.4-and 0.9-fold (two experiments) the amount obtained from control platelets, whereas associated phospholipids were increased by 7.5-and 11-fold, respectively. This indicated that adhesion promoted specific association of phosphoinositides with talin. Moreover, by comparison with the relative amounts of polyphosphoinositides found in total platelets, there was an enrichment in PtdIns-4-P and PtdIns-3,4-P 2 in the anti-talin immunoprecipitates (compare Figs. 2  and 5D). Treatment of platelets with wortmannin induced a strong decrease (60 -100%) of these three lipids associated with talin (Fig. 5, C and D). Thus, we conclude that PtdIns 3-kinase inhibitors may modify the association of PtdIns-4,5-P 2 with talin and therefore influence its possible function in platelet spreading. DISCUSSION We have shown previously that synthesis of PtdIns-3,4-P 2 in adherent platelets is under the control of the released ADP, since addition of ADP reversed the inhibitory effects of an ADP  A and B) and plated on the fibrinogen matrix (A, W) or kept in suspension (C) for 60 min as described in the experimental procedure. In some cases, platelets were pretreated with 100 nM wortmannin before plating (W). There was no detectable difference in the amount of lipid-incorporated 32 P between the adherent and suspended samples. Immunoprecipitation of talin from 4.5 ϫ 10 8 platelets was performed with a polyclonal anti-talin antibody followed by Western blotting (A), and lipids were extracted and separated by TLC (C) as described under "Experimental Procedures." No lipid was precipitated by protein G-Sepharose alone. Quantification of the amounts of talin immunoprecipitated was performed by a densitometric analysis (B), and the lipid content (D) is expressed taking the amount of PtdIns-4-P in assay A as 100% (74 cpm). B and D are means from two independent experiments.

Treatment
Agonist Spreading restoration PtdIns-4,5-P 2 ϩ Wortmannin ϩ aspirin PtdIns-4,5-P 2 ϩ Wortmannin ϩ GF109203X PtdIns-4,5-P 2 ϩ scavenger on PtdIns-3,4-P 2 synthesis (3). Our present results, in agreement with those of Haimovich et al. (2,4), demonstrate that ADP release occurring upon platelet adhesion is required for full platelet spreading. Thus, ADP released by adherent platelets controls both PtdIns-3,4-P 2 synthesis and cell spreading. In conditions where PtdIns-3,4-P 2 synthesis was totally abolished by inhibitors of PtdIns 3-kinase, we observed an inhibition of platelet spreading while pseudopodal extension was maintained. We demonstrate here that the PtdIns 3-kinase signaling pathway is required for ADP-induced spreading of adherent platelets. In order to determine which lipid is responsible for platelet spreading, we have added synthetic PtdIns-3,4-P 2 and PtdIns-3,4,5-P 3 to adherent wortmannin-treated platelets. PtdIns-3,4-P 2 appears to be the most efficient in restoring partial spreading, i.e. pseudopodal and hyalomere extension. However, among all phosphoinositides tested, PtdIns-4,5-P 2 alone was able to trigger the full spreading of wortmannin-treated platelets. Although all phospholipid solutions were prepared under similar conditions, their packing into micelles or vesicules was not characterized. Moreover, differences in the acyl chains of PtdIns-3,4-P 2 (palmitate) and PtdIns-4,5-P 2 (stearate and arachidonate) may have influenced their activity. Nevertheless, our results suggest a role for both PtdIns-4,5-P 2 and lipid products of PtdIns 3-kinase in the signal transduction pathway leading to platelet spreading.
We have observed a trend toward a decrease of [ 32 P]PtdIns-4-P in adherent platelets pretreated with PtdIns 3-kinase inhibitors. This result suggests that a wortmannin-sensitive Pt-dIns 4-kinase might exist in platelets, as has been shown in other models (30,31). Nevertheless, in a cell system, wortmannin was reported to inhibit PtdIns 4-kinase at M concentration, and LY294002 has not been described as an inhibitor of known PtdIns 4-kinases (13). Thus, PtdIns 3-kinase activity could be upstream of a PtdIns 4-kinase and/or a PtdIns-4-P phosphatase. Finally, even though we have not measured a significant decrease of [ 32 P]PtdIns-4,5-P 2 in whole platelets pretreated with PtdIns 3-kinase inhibitors, a decrease of Pt-dIns-4-P level could impair the synthesis of a particular pool of PtdIns-4,5-P 2 required for platelet spreading.
Recent studies from the Schlessinger and Rhee laboratories (32,33) demonstrate that ␥ isoforms of phospholipase C (PLC) could be activated by PtdIns-3,4,5-P 3 , either by targeting to cell membrane through their PH domain or by direct activation through their SH2 domain. One could thus expect an increase of PtdIns-4,5-P 2 level when adherent platelets have been pretreated with PtdIns 3-kinase inhibitors. Nevertheless, at least two reasons could explain why in our experiments this variation is not observed. First, in our previous paper (3), we have shown that upon platelet adhesion on a fibrinogen matrix a PLC active on PtdIns-4,5-P 2 was rapidly and transiently stimulated. Maximal increase of PtdOH production and PtdIns-4,5-P 2 decrease was observed as early as 5 min of adhesion. Thereafter, these two metabolites returned gradually to their basal level, and that corroborates the absence of PtdOH and PtdIns-4,5-P 2 variations after 60 min of adhesion, as shown in Fig. 2 of our present article. We thus believe that in the late steps of platelet adhesion PLC activity is not involved. However, it should be of importance to check PLC activity during the early steps of adhesion of platelets treated with PtdIns 3-kinase inhibitors. Second, since our present data show a decrease of PtdIns-4-P level upon platelet treatment with Pt-dIns 3-kinase inhibitors, an eventual increase of the PtdIns-4,5-P 2 level might be impaired.
PtdIns-4,5-P 2 regulates several actin-binding proteins as profilin, gelsolin, ␣-actinin, and vinculin (34). One of the major proteins of focal adhesions, talin, has been shown to be involved in cell spreading (28,29). Its interaction with lipids has been documented in vitro and could be of importance for talin nucleated actin polymerization (34). Here, we show that PtdIns-4,5-P 2 as well as PtdIns-4-P and PtdIns-3,4-P 2 become associated with talin upon platelet adhesion. Moreover, treatment of platelets by wortmannin strongly reduces the amounts of polyphosphoinositides recovered in the anti-talin immunoprecipitate. Even though it remains to be determined whether this association is direct or not, our results support the notion of a possible regulation by PtdIns 3-kinase of a pool of PtdIns-4,5-P 2 potentially involved in cell spreading.
Recent results from King et al. (40), showing that spreading of COS 7 cells attached to fibronectin is delayed after treatment by wortmannin and LY294002, support the view that the Pt-dIns 3-kinase signaling pathway is required for cell spreading, as controlled by integrins and/or by tyrosine kinase receptors (41). It has been suggested that ADP released from adherent platelets supports some specific signals such as Vav phosphorylation via an indirect mechanism involving activation of ␣ IIb ␤ 3 (42). Furthermore, an integrin-associated protein agonist peptide triggers activation of ␣ IIb ␤ 3 integrin resulting in platelet spreading on immobilized fibrinogen (43). Thus, upon platelet adhesion to immobilized fibrinogen, it remains to be clarified whether platelet spreading is secondary to the ␣ IIb ␤ 3 integrin engagement.