Distinct roles of ADP receptors in von Willebrand factor–mediated platelet signaling and activation under high flow

Platelets adhere and aggregate at sites of vascular injury, contributing to the arrest of bleeding but also to the pathologic occlusion of diseased vessels.¹  At high shear rates, equivalent to those generated by blood flow in arterioles or stenotic arteries, adhesion requires von Willebrand factor (VWF) endogenously present in the subendothelial matrix or absorbed onto injured tissue components exposed to plasma. The membrane glycoprotein (GP) Ibα, a component of the GP Ib-IX-V complex,²  can initiate platelet adhesion to the immobilized VWF A1 domain (VWF-A1) while opposing elevated shear forces, but cannot support permanent attachment. Nonetheless, this step is instrumental in the formation of additional bonds that lead to stable platelet adhesion and aggregation mediated by the activated integrin α_IIbβ₃ and the collagen receptors, GP VI and α₂β₁, among others.^3-6  Recent studies have demonstrated the existence of different Ca⁺⁺ signals associated with the engagement of GP Ib-IX-V and α_IIbβ₃ during platelet adhesion under flow.^7,8  In particular, the interaction of GP Ibα with VWF-A1 leads to sequential elevations in intracytoplasmic Ca⁺⁺ concentration ([Ca⁺⁺]_i) of which the first (α/β peak) may be associated with mechanical stimulation and rapid Ca⁺⁺ release from intracellular stores. This signal appears to be coupled to a first stage of α_IIbβ₃ activation that mediates the shift from transient to a more prolonged adhesion and is regulated by cyclic adenosine monophosphate (cAMP) and cyclic guanosine monophosphate (cGMP) levels. A second stage of α_IIbβ₃ activation must then be reached for irreversible adhesion and aggregation to occur, and this appears to be associated with a second [Ca⁺⁺]_i elevation (γ peak) of greater amplitude and duration involving a transmembrane ion flux.⁸  The signaling cascade responsible for generating the GP Ibα and VWF-dependent Ca⁺⁺ peaks remains to be elucidated.

In previous studies it was established that secreted adenosine diphosphate (ADP) is necessary for the shear-induced platelet aggregation initiated by the interaction of soluble VWF with GP Ibα.⁹  ADP binds to different G protein–linked P2 receptors, of which 2 (P2Y₁ and P2Y₁₂) are present on platelets.^10,11  Ligation of P2Y₁, linked to G_q, activates phospholipase C (PLC) and mobilizes Ca⁺⁺ from intracellular stores, leading to the activation of protein kinase C and subsequent platelet aggregation. Ligation of P2Y₁₂, linked to G_i, inhibits adenylyl cyclase, lowers cAMP levels, and potentiates ADP-induced platelet aggregation.^12,13  The function of both P2Y₁ and P2Y₁₂ is required for platelet aggregation after adhesion to collagen-bound VWF under flow conditions.¹⁴  With the present studies we have defined the distinct roles that P2Y₁ and P2Y₁₂ play in the early events that follow the initial platelet interaction with immobilized VWF-A1 under high flow conditions. In particular, we have examined the consequences of specific pharmacologic inhibition of P2 receptors and selected signaling pathways on the occurrence of intracytoplasmic calcium signals related to α_IIbβ₃ activation and the transition from initial platelet contacts to stable adhesion and subsequent aggregation into thrombi. Our findings demonstrate a differential role of P2Y₁ and P2Y₁₂, respectively, in platelet adhesion and aggregation onto immobilized VWF under elevated shear stress, and highlight the distinct contribution of signaling pathways dependent on Src family kinases, PLC, and phosphoinositide 3-kinase (PI 3-K) to these processes.

Venous blood from medication-free, healthy volunteers, who gave their informed consent according to the declaration of Helsinki, was collected into a 1:6 final volume of citric acid/citrate/dextrose, pH 4.5, or a specific α-thrombin inhibitor, either hirudin (Iketon, Milan, Italy) at the final concentration of 400 U/mL or d-phenylalanyl-l-prolyl-l-arginine chloromethyl ketone dihydrochloride (PPACK; Calbiochem, La Jolla, CA) at the final concentration of 50 μM. Acid-citrate-dextrose (ACD)–containing blood (50 mL) was centrifuged at 1000g for 50 seconds at 22°C to 25°C, and the supernatant platelet-rich plasma (PRP) was collected. The procedure was repeated twice to obtain approximately 15 mL PRP. The platelets were incubated for 20 minutes at 37°C with the fluorescent calcium probe fluo-3 acetoxymethyl ester-AM (FLUO 3–AM; Molecular Probes, Eugene, OR) at the final concentration of 8 μM. Erythrocytes separated from the same blood were washed 3 times in a divalent cation-free HEPES-Tyrode buffer (10 mM Hepes [N-2-hydroxyethylpiperazine-N′-2-ethanesulfonic acid], 140 mM NaCl, 2.7 mM KCl, 0.4 mM NaH₂PO₄, 10 mM NaHCO₃, and 5 mM dextrose, pH 6.5) and finally resuspended in the same buffer and with the addition of 1.75 mM Probenecid (Sigma, St Louis, MO), used to prevent leakage of FLUO 3–AM from the platelets.¹⁵  An adequate volume of PRP containing 2 × 10⁸ to 8 × 10⁸ FLUO 3–AM–loaded platelets/mL was mixed with an aliquot of the erythrocyte suspension (50% hematocrit) and apyrase (grade III; 142 ATPase U/mg protein; Sigma) at the final concentration of 5 ATPase U/mL. The mixture was centrifuged at 1000g for 15 minutes, the supernatant was discarded, and the cell pellet was suspended in autologous plasma (prepared from the blood collected in hirudin or PPACK by centrifugation at 1650g for 15 minutes at 22°C to 25°C) or Hepes-Tyrode buffer pH 7.4 and 1.75 mM Probenecid in a proportion such that the hematocrit was 42% to 45%. The labeling procedure did not significantly alter the function of platelets as evidenced by the response to agonists and the expression of surface activation markers.⁸ 

Human VWF was prepared from plasma as previously described.¹⁶  The protein was diluted in phosphate-buffered saline (20 mM Na₂HPO₄, 20 mM NaH₂PO₄, 2.7 mM KCl, 0.15 M NaCl, pH 7.4) to a final concentration of 100 μg/mL, and 100 μL of the solution was used to coat glass coverslips for 60 minutes at 22°C to 25°C. These were then washed with the coating buffer and kept in a moist environment until assembled in a modified Hele-Shaw flow chamber.^4,17  The chamber was positioned in an inverted microscope equipped with epifluorescent illumination (Diaphot-TMD; Nikon Instech, Kanagawa, Japan), an intensified CCD videocamera (C-2400-87; Hamamatsu Photonics, Shizuoka, Japan), and appropriate filters. The total area of an optical field corresponded to approximately 0.007 mm². Blood cells were aspirated through the chamber with a syringe pump (Harvard Apparatus, Boston, MA) at a flow rate calculated to obtain a wall shear rate of 3000 s^–1 at the inlet. When indicated, various substances were added to the blood cell suspension before perfusion. These included prostaglandin (PG) E₁ (which increases the cAMP levels in platelets; Sigma); PP1, PP2, and PP3 (PP1 and PP2 inhibit Src family kinases, and PP3 is a noninhibitory analog control; Calbiochem); apyrase (grade III, 142 ATPase U/mg, or purified to 7641 U/mg; Sigma); wortmannin and LY294002 (which inhibit PI 3-K; both from Sigma); U73122 (which inhibits PLC) and the structurally related inactive analog, U73343 (Calbiochem-Nova Biochem, Bad Soden, Germany); MRS2216 (which inhibits P2Y₁; kindly provided by Dr K. A. Jacobson, National Institutes of Health, Bethesda, MD)¹⁸ ; 2 additional P2Y₁ inhibitors (A3P5P and MRS2179, both from Sigma)¹² ; AR-C69931MX (which inhibits P2Y₁₂; from Astra Zeneca Pharmaceutical LP, Wilmington, DE)^19,20 ; and acetyl salicylic acid (ASA; which inhibits cyclooxygenase-1 and, thus, the generation of thromboxane A₂). To prepare a solution of the latter, powdered lysin-acetyl-salicylate (Sanofi-Synthelabo S.p.A., Milano, Italy) was dissolved into a 0.9% NaCl solution at a final concentration of 10 mM (stock solution). Monoclonal immunoglobulins G (IgG), used when indicated, were prepared and characterized as described previously; in particular, the monoclonal antibody LJ-CP8 blocks the binding of α_IIbβ₃ to both VWF and fibrinogen.²¹  Human fibrinogen, used in perfusion experiments, was prepared from human plasma as previously described.^22,23  Experiments were recorded in real time on videotape at the rate of 25 frames/second, which resulted in a time resolution of 0.08 seconds. Selected video sequences were also digitized in real time using a Matrox-Digisuite board (Matrox Graphics, Dorval, QC, Canada).

Image analysis was performed either on recorded experiments or in-line using custom-made software (Amplimedical, Casti Imaging Division, Venice, Italy). The program tracks the area of single platelets and determines the position of the corresponding centroid on all the frames collected at a sampling rate of 25 per second, then calculates instant velocity and variations of light intensity measured as the total of all the pixels in a platelet. Thus, information on the measured parameters was obtained every 0.04 seconds. Instantaneous velocity was calculated according to the general equation v = s/t, modified with appropriate algorithms to minimize the effect of variations in centroid position resulting from changes in platelet shape or orientation.⁸  Arrest times were calculated from the sum of all the frames in which a platelet had zero velocity, thus was not displaced from its position according to a definition previously published.³  Stable adhesion was defined as zero velocity for 30 seconds or more. In this case, the surface imprint of a platelet in the first frame of the observation period was overlapping at least partially with the imprints in all subsequent 749 frames. The variations in intensity of the FLUO 3–AM fluorescence were converted into [Ca⁺⁺]_i using the equation [Ca⁺⁺]_i = K_d F-F_min/F_max-F where K_d is the dissociation constant of FLUO 3–AM for the interaction with Ca⁺⁺, corresponding to 864 nM at 37°C¹⁵ ; F is the measured fluorescence intensity of a single platelet; F_max is the fluorescence intensity of single platelets treated with the Ca⁺⁺ ionophore A23187 (10 μM; Sigma) in the presence of 2 mM CaCl₂; and F_min is the fluorescence intensity of an unstimulated single platelet. In agreement with previous findings,⁸  the analysis of single platelets translocating on immobilized VWF allowed the identification of 2 characteristic [Ca⁺⁺]_i oscillations associated with specific signaling events. The earlier type α/β peaks are caused by Ca⁺⁺ released from intracellular stores and are associated with the typical platelet translocation on VWF mediated by GP Ibα; the later and longer lasting type γ peaks, which always follow at least one α/β oscillation, are caused by a transmembrane Ca⁺⁺ flux and are associated with stationary platelet adhesion to VWF mediated by the activated integrin α_IIbβ₃.⁸  The computerized program also measured the size of formed platelet aggregates, expressed as the corresponding cross-sectional area assuming an average platelet diameter of 2.4 μm (range, 1.6-4 μm). Consequently, the objects observed on the surface were arbitrarily classified as single platelets, 2 μm² to 13 μm²; microaggregates, 14μm² to 50 μm²; small aggregates, 51 μm² to 200 μm²; and large aggregates, more than 200 μm². ²⁴ 

Platelets loaded with FLUO 3–AM and suspended in reconstituted blood at a count of 1 × 10⁷/mL (to analyze single platelet events) or 8 × 10⁷/mL to 16 × 10⁷/mL (to analyze aggregation) were perfused over immobilized VWF at the wall shear rate of 3000 s^–1. The P2Y₁ antagonist, MRS2216, as well as the PI 3-K inhibitor, wortmannin, completely abolished formation of platelet aggregates, whereas the P2Y₁₂ antagonist, AR-C69931MX, exerted a partial but significant inhibitory effect only on the formation of larger aggregate, and ASA had no effect (Figure 1). Two additional P2Y₁ receptor antagonists, A3P5P and MRS2179, both used at the final concentration of 250 μM, inhibited platelet aggregate formation as effectively as MRS2216 (data not shown). Measurement of the instantaneous velocity of individual platelets translocating on VWF revealed a typical stop-and-go motion with alternating rapid deceleration and acceleration, and permitted to calculate the mean translocation velocity and mean arrest time. The P2Y₁ receptor antagonist, MRS2216, greatly increased the platelet translocation velocity onto VWF and decreased the arrest time at sites of transient adhesion, results similar to those observed with the addition of PGE₁ (Figure 2, Table 1). The 2 other P2Y₁ inhibitors, A3P5P and MRS2179, caused the same effect as MRS2216 (data not shown). Increased platelet translocation velocity and shortened arrest times were also observed after treating the platelets with the PLC inhibitor, U73122 (Figure 2, Table 1), or the Src family kinase inhibitors, PP2 (Table 1) or PP1 (used at the same final concentration of 25 μM, not shown). In contrast, the P2Y₁₂ receptor antagonist as well as apyrase and ASA caused no change of these platelet kinetic parameters (Figure 2, Table 1), nor did PP3, a noninhibitory analog of PP2 and PP1 used as a negative control (not shown). To ensure that maximal effects were detected, in selected experiments we verified that the ADP receptor inhibitors gave consistent results when tested at concentrations 2- to 5-fold in excess of those resulting in the greatest inhibition of platelet adhesion or aggregation (not shown). Apyrase was used at a concentration 10-fold in excess of that required to abolish platelet aggregation induced by 2 μM ADP. Complete inhibition of aggregation was observed only when apyrase and ADP were preincubated before addition to the platelets, since the primary response to exogenous ADP could not be prevented without preincubation (not shown). Blocking of PI 3-K function with 2 different inhibitors, wortmannin and LY294002,²⁵  gave different results depending on the concentration used. LY294002 at the highest concentration tested (25 μM) increased the translocation velocity of platelets interacting with immobilized VWF and decreased the arrest time, whereas wortmannin at 100 nM only slightly increased the translocation velocity and had essentially no effect on the arrest time (Figure 2, Table 1). Higher concentrations of wortmannin, up to 1 μM, increased the translocation velocity to 4.98 μm/second ± 0.68 μm/second and decreased the arrest time to 1.56 seconds ± 0.64 seconds, but under these conditions the inhibitory effect may not be limited to PI 3-K.

In agreement with previous findings,⁸  we found that a proportion of the platelets translocating on VWF exhibited transient [Ca⁺⁺]_i elevations, identified as α/β peaks, which are known to be dependent on Ca⁺⁺ release from intracellular stores. When these platelets established irreversible adhesion, additional [Ca⁺⁺]_i elevations occurred, identified as γ peaks, which have previously been shown to involve a transmembrane ion flux.⁸  ASA and the P2Y₁₂ antagonist had no effect on any of the Ca⁺⁺ signals exhibited by the platelets interacting with immobilized VWF, whereas the P2Y₁ antagonist, MRS2216, completely blocked the appearance of type γ peaks (Figure 3).

A quantitative analysis of the percentage of platelets showing intracytoplasmic Ca⁺⁺ elevations while interacting with immobilized VWF demonstrated that inhibition of P2Y₁ with MRS2216 caused a relatively modest (∼ 30%) but statistically significant decrease in α/β peaks (Figure 4A). Similar results were obtained with the other 2 P2Y₁ inhibitors, A3P5P and MRS2179 (data not shown). Essentially the same result was also obtained when platelets were treated with U73122, an inhibitor of PLC (Figure 4A), in agreement with the concept that the latter enzyme functions downstream of P2Y₁ ligation.¹²  The structurally related but inactive U73122 analog, U73343, used at the same final concentration of 20 μM, had no effect on platelet adhesion to VWF and calcium signaling (not shown). Scavenging of released ADP with apyrase had no effect in this regard (Figure 4A), possibly reflecting the relatively slow kinetics of this enzyme in relation to the rapid ADP effect on P2Y₁. Inhibition of P2Y₁₂ or PI 3-K, or inhibition of thromboxane A₂ synthesis with ASA, also did not influence the number of platelets showing α/β peaks while interacting with immobilized VWF, but PGE₁ completely abolished these events (Figure 4A). Likewise, the Src family kinase inhibitor, PP2, prevented the appearance of α/β peaks when used at the concentration of 25 μM (Table 1). In contrast, only the P2Y₁₂ inhibitor and ASA did not affect the frequency of type γ Ca⁺⁺ peaks, whereas inhibition of P2Y₁, PI 3-K, and PLC, as well as scavenging of ADP with apyrase, completely blocked these events, as did treatment with PGE₁ (Figure 4B). In fact, the P2Y₁₂ receptor antagonist caused an increase in the proportion of platelets exhibiting γ peaks (Figure 4B), possibly because under these conditions a greater amount of ADP was available for binding to P2Y₁. Inhibition of Src family kinases with PP2 obliterated the appearance of γ peaks (Table 1), and the inhibitor concentration required to achieve this effect was 4-fold lower than that required to abolish α/β peaks (not shown). To investigate the influence of P2Y₁ and P2Y₁₂ on the frequency of α/β peaks in single platelets that translocated on immobilized VWF, we performed experiments in the presence of the anti-α_IIbβ₃ monoclonal antibody LJ-CP8, which completely abolished the appearance of γ peaks. There was no change in the α/β peak frequency in single platelets treated with the P2Y₁ or the P2Y₁₂ receptor antagonists, and the same was true for platelets treated with the PLC inhibitor, U73122 (Figure 4C).

The maximal intracytoplasmic Ca⁺⁺ concentration corresponding to individual type α oscillations was partially but significantly decreased in platelets treated with the P2Y₁ receptor antagonist, MRS2216 (1037 nM ± 322 nM compared with 1543 nM ± 312 nM of the control; P < .05). The same was true for platelets treated with the PLC inhibitor, U73122 (902 nM ± 110 nM; P < .01), whereas no such effect was observed with the P2Y₁₂ antagonist, the ADP scavenger, apyrase, the PI 3-K inhibitors, wortmannin and LY294002, or ASA (Figure 5A). In contrast, only the P2Y₁₂ antagonist and ASA had no effect on the peak intracytoplasmic Ca⁺⁺ concentration during type γ oscillations, which were obliterated by all the other inhibitors (Figure 5B).

We demonstrate here the distinct roles of 2 ADP receptors in the activation of platelets interacting with immobilized VWF in a flow field. Others have used apyrase as a scavenger or receptor antagonists to examine whether ADP contributes to VWF-induced α_IIbβ₃ activation. In one study,²⁶  blocking ADP-dependent signaling reduced significantly the number of platelets forming stationary adhesion contacts on VWF at the shear rate of 150 s^–1 but not 600 s^–1 or 1880 s^–1. This finding prompted the conclusion that signal amplification by ADP is not required for VWF-induced platelet activation under intermediate and high shear stress. In a separate report⁸  we identified 2 sequential Ca⁺⁺ signals associated with the activation of flowing platelets interacting with immobilized VWF. The earlier α/β peaks, linked to the engagement of GP Ibα by VWF-A1, involved Ca⁺⁺ release from intracellular stores and preceded stationary adhesion. Since platelets treated with apyrase showed normal α/β Ca⁺⁺ peaks and attached firmly to immobilized VWF, we concluded that ADP stimulation is not needed for the activation of α_IIbβ₃ that stabilizes adhesion. In contrast, scavenging ADP abrogated the later γ peaks that appeared in firmly adherent platelets, suggesting that signal amplification by released ADP is required for full α_IIbβ₃ activation compatible with aggregation.⁸ 

The results now obtained with selective antagonists of the 2 platelet ADP receptors modify the concepts reported in the literature,^8,23  demonstrating a previously unrecognized role of P2Y₁ in generating the early signals associated with a first level of α_IIbβ₃ activation required for stable adhesion to VWF. Such a conclusion rests on the observation that platelets with blocked P2Y₁ exhibited the relatively rapid translocation velocity and short transient arrest times sustained by GP Ibα binding to VWF-A1 without α_IIbβ₃ engagement, as seen with PGE₁–treated platelets. The apparent contradiction, observed previously⁸  and confirmed here, that scavenging ADP with apyrase cannot prevent platelets from achieving α_IIbβ₃-mediated stable adhesion may reflect the inability of the enzyme to remove releasedADPcompletely. In support of this hypothesis, adenosine triphosphate (ATP) can be detected with cell surface–bound firefly luciferase even in the presence of apyrase.²⁷  Since the ATP/ADP ratio in human platelets is approximately 2,²⁸  and the nucleotides are thought to undergo secretion with the same kinetics, it is likely that some released ADP may persist locally in quantities sufficient to activate P2Y₁ even in the presence of apyrase. In the latter situation, the reduced agonist concentration may result in a limited activation sufficient to support stable adhesion but not generalized aggregation.

Blocking P2Y₁ function decreased by approximately 30% the proportion of platelets that exhibited type α oscillations while translocating on VWF, and reduced by approximately 500 nM the [Ca⁺⁺]_i of a type α peak. The reduction in activation mirrors the proportion of platelets that show a γ peak after an α peak,⁸  indicating that signaling from P2Y₁ following GP Ibα-dependent ADP release may be crucial for the increase of [Ca⁺⁺]_i to the levels needed for α_IIbβ₃ activation. This hypothesis is supported by the observation that inhibition of PLC_β, which acts downstream of P2Y₁ and generates an effector of Ca⁺⁺ release from intracellular stores (Figure 6),^12,13  inhibits platelet adhesion to VWF and related Ca⁺⁺ oscillations as effectively as the P2Y₁ antagonist. PLC_γ2, which is also inhibited by U73122,²⁹  may contribute to GP Ib–dependent signaling as previously inferred from studies on platelet adhesion to VWF under static conditions.³⁰  Thus, as also suggested elsewhere,³¹  intracytoplasmic Ca⁺⁺ levels may be related to the degree of activation of platelets interacting with VWF under high shear stress conditions: less than 200 nM equals resting; 400 nM equals β peak, subactivation; 1000 nM equals α peak induced by GP Ibα signaling, leading to ADP release; 1500 nM equals α peak+P2Y₁ signal, first level of α_IIbβ₃ activation and stable adhesion; more than 1500 nM equals γ peak, full α_IIbβ₃ activation with platelet aggregation modulated by P2Y₁₂ function (Figure 6). A recent report³²  indicated that concurrent blockade of both platelet ADP receptors cannot prevent α_IIbβ₃ activation on platelets adherent to VWF-A1 under static conditions, suggesting that direct signaling through GP Ibα may be sufficient for the functional modulation of this integrin. Our present studies confirm that a Ca⁺⁺ signal induced by the engagement of GP Ibα, albeit reduced in intensity, can be detected in P2Y₁-blocked platelets, and yet cannot lead to firm adhesion to VWF under high shear stress conditions (Table 1). The contribution of P2Y₁ to GP Ibα–initiated α_IIbβ₃ activation, therefore, may have different functional relevance in relation to hemodynamic parameters of the blood circulation. Our studies, in agreement with a previous report using platelets with blocked ADP receptors,³²  also indicate that a Src family kinase–dependent pathway is initially involved in the transduction of the signal originated by VWF-A1 binding to GP Ibα, as shown by the abrogation of all Ca⁺⁺ peaks caused by the Src inhibitors PP1 and PP2. Similar conclusions have been reached by others using different experimental conditions.^33,34  Because greater concentrations of the inhibitors were required to block α/β as opposed to γ peaks, it appears that one or more Src family kinases may participate in sequential steps of platelet activation linked first to stable adhesion and then to aggregation (Figure 6).

In contrast to the early role of P2Y₁ in promoting stable platelet adhesion to immobilized VWF, which is an absolute requirement for subsequent aggregation, we found that P2Y₁₂ is not involved in generating any of the Ca⁺⁺ transients initially associated with platelet activation. In particular, we could not detect any effect of blocking P2Y₁₂ on the firm adhesion of platelets to VWF, as others have reported.³⁵  Of note, the translocation velocity of P2Y₁₂-blocked platelets in the above mentioned publication³⁵  was lower than that of control platelets reported here and previously.^3,36  We suggest that measuring platelet-VWF interactions for 2 seconds, as opposed to the 30 seconds of the present study, may increase the experimental error particularly when translocation is minimal during the observation period (Figure 6 in Goto et al³⁵ ). On the other hand, we found that P2Y₁₂ function is critical for the full development of platelet aggregates on immobilized VWF, in agreement with its reported role in shear-induced platelet aggregation initiated by soluble VWF binding to GP Ibα and in thrombus formation on collagen-bound VWF.¹⁴  Others have reported that P2Y₁₂ is involved in maintaining elevated Ca⁺⁺ levels in aggregating platelets,³⁷  a function that is likely to be important for thrombus growth, as we confirm here, but appears to be distinct from earlier signaling events that initiate aggregation.

The contribution of type I PI 3-K to GP Ibα signaling remains controversial. Others reported that stationary platelet adhesion to VWF was impaired by wortmannin and LY294002, 2 distinct PI 3-K inhibitors, used at concentrations (100 nM and 25 μM, respectively) proven to block PtdIns(3,4,5)P₃ production.³⁸  Previously⁸  and here we found that wortmannin used at 100 nM has a minimal effect on platelet adhesion to VWF and cannot prevent α/β Ca⁺⁺ peaks, but even at 10 nM blocks γ peaks linked to α_IIbβ₃ activation. In the present study, we found that LY294002³⁹  dose dependently increases the translocation velocity of flowing platelets interacting with immobilized VWF and shortens their arrest time. Similar effects could be obtained with wortmannin, but only at concentrations at which the specificity for PI 3-K inhibition is not assured. In any case, even LY294002 had no effect on type α/β Ca⁺⁺ peaks, while inhibiting completely γ peaks. A previous report demonstrated a marked effect of PI 3-K inhibitors on the Ca⁺⁺ transients measured in platelets interacting with immobilized VWF under flow,³⁸  a finding that can now be ascribed more precisely to inhibition of the more sustained γ peaks. It is apparent from our results that PI 3-K is not involved in generating the Ca⁺⁺ signals linked to the binding of GP Ibα to immobilized VWF-A1, but has a role in the induction of early α_Iibβ₃ activation. The reason why LY294002 blocks the latter event better than wortmannin remains to be explained. The 2 are structurally and mechanistically distinct inhibitors,^39-41  and may affect differentially other components of the signaling pathways linked to PI 3-K, as seen in other biologic systems,⁴²  either coupled to the VWF-GP Ibα interaction⁴³  or to P2Y₁ signaling. In conclusion, our findings identify a number of sequential steps in the signaling cascade initiated by the VWF-GP Ibα interaction that may contribute to the participation of platelets in hemostasis and atherothrombosis. This information may help in the selection and evaluation of potential targets for pharmacologic intervention aimed at preventing arterial thrombosis.

We thank Dr Ken Jacobson and Dr Marco Cattaneo for useful discussion on the biology and pharmacology of ADP receptors, and for providing a P2Y₁ inhibitor.