Integrin a IIb b 3 -Dependent Calcium Signals Regulate Platelet-Fibrinogen Interactions Under Flow: Involvement of PLC g 2

Platelet adhesion to fibrinogen is important for platelet aggregation and thrombus growth. In this study we have examined the mechanisms regulating platelet adhesion on immobilized fibrinogen under static and shear conditions. We demonstrate that integrin a IIb b 3 engagement of immobilized fibrinogen is sufficient to induce an oscillatory calcium response, necessary for lamellipodial formation and platelet spreading. Released ADP increases the proportion of platelets exhibiting a cytosolic calcium response, but is not essential for calcium signaling or lamellipodial extension. Pretreating platelets with the Src kinase inhibitor, PP2, the IP 3 receptor antagonist, APB-2, or the PLC inhibitor, U73122, abolished calcium signaling and platelet spreading, suggesting a major role for Src-kinase-regulated PLC isoforms in these processes. Analysis of PLC g 2 -/- mouse platelets revealed a major role for this isoform in regulating cytosolic calcium flux and platelet spreading on fibrinogen. Under flow conditions, platelets derived from PLC g 2 -/- mice formed less stable adhesive interactions with fibrinogen, particularly in the presence of ADP antagonists. Our studies define an important role for PLC g 2 in integrin a 3 -dependent calcium flux necessary for stable platelet adhesion and spreading on fibrinogen. Furthermore, they establish an important co-operative signaling role for PLC g 2 and ADP in regulating platelet adhesion efficiency on fibrinogen. distinct to describe platelet adhesion on fibrinogen and vWf under Platelet adhesion to vWf is a multi-step process involving GPIb/V/IX and integrin a IIb b 3 . Adhesion is initiated by a reversible tethering step between the A1 domain of vWf and GPIb a , followed by a secondary adhesive interaction between the C1 domain of vWf and integrin a IIb b 3 . This two-step model of platelet adhesion is critically dependent on the co-operative signaling function of both receptors, with GPIb a -derived calcium spikes initiating integrin a IIb b 3 activation and transient bond formation with vWf, whereas subsequent integrin a IIb b 3 -dependent calcium signals sustain firm adhesion. An important difference between immobilized vWf and fibrinogen is that the latter ligand can engage integrin a IIb b 3 in its low affinity state, and according to previous findings, maintain firm adhesion independent of platelet activation. However, our findings suggest a reinterpretation of this model in which the mechanisms regulating platelet adhesion on vWf are also relevant to fibrinogen. Thus, initial bond formation between fibrinogen and integrin a IIb b 3 is potentially transient and reversible, depending on the tensile stress applied to the formed bonds. Conversion of these bonds to firm adhesion contacts is dependent on integrin a IIb b 3 -dependent calcium flux and also on the release of ADP. These calcium changes are likely to support firm adhesion through multiple mechanisms, including effects on integrin a IIb b 3 affinity, receptor clustering and through reorganization of the cytoskeleton. Elucidating the mechanisms co-ordinating integrin a IIb b 3 calcium signals with ADP release will be important to fully understand the processes regulating platelet adhesion efficiency on fibrinogen. 50-minute period was quantified and expressed as a percentage of the total number of adherent cells in a given field. These results are taken from one experiment representative of three. B , Representative calcium traces are from two individual cells, demonstrating integrin a IIb b 3 -dependent calcium flux in fibrinogen-adherent platelets in the presence ( Apyrase/aspirin, 50 minutes) or absence of inhibitors ( Control, 30 minutes). as of These are taken from one of and These represent – from three individual

In this study we have examined the mechanisms regulating platelet adhesion and activation on a fibrinogen matrix. In contrast to previous reports, our studies do not demonstrate an absolute requirement for ADP for lamellipodial extension and platelet spreading on fibrinogen.
Rather, they suggest that integrin α IIb β 3 outside-in signaling linked to Src kinase-mediated PLCγ2 activation is critical for platelet spreading, whereas ADP release serves a secondary role, potentiating platelet activation. Furthermore, we demonstrate that integrin α IIb β 3 -dependent calcium flux, combined with ADP release, plays an important role in sustaining platelet fibrinogen interactions under flow. These findings challenge previous concepts on the mechanisms regulating platelet adhesion and activation on fibrinogen, defining a pivotal role for integrin α IIb β 3 -dependent calcium flux in these processes.
Measurement of Integrin α IIb β 3 Activation -Integrin α IIb β 3 activation during platelet adhesion to fibrinogen under static conditions, was assessed using the integrin α IIb β 3 -activation specific antibody, PAC-1. Platelets were allowed to adhere to fibrinogen in the presence of PAC-1 (2 µg/ml), followed by fixation (3.7 % formaldehyde), and incubation with a FITC-conjugated antimouse IgG F(ab')2 fragment. PAC-1 immunofluorescence was visualized using confocal fluorescence microscopy (100X magnification, Leica, TCS-SP, Leica Germany), and quantified using the Leica TCS NT software, as previously described (26). In some studies, platelets were preincubated with the indicated concentrations of APB-2, U73122, PP2, apyrase alone or in combination with aspirin, prior to the performance of adhesion studies.

Platelet Adhesion to Fibrinogen under Flow
Conditions -Flow assays were performed as previously described (26). In studies examining the ability of platelets to adhere to the fibrinogen matrix, washed platelets (5 x 10 7 /ml) were reconstituted with washed RBCs (50% v/v) (containing 0.04 units/ml apyrase and 1 unit/ml hirudin), as described previously (26). Platelets were perfused through fibrinogen (100 µg/ml)-coated microcapillary tubes at a wall shear rate of 150 s -1 for 2 minutes at 37 o C. Platelet-matrix interactions were visualized using epi-fluorescence microscopy (Leica DMIRB, Leica Germany) and video-recorded for off-line analysis. Platelet tethering was analyzed at 30, 60 and 90 seconds, and each platelet that interacted with the fibrinogen matrix for ≥ 2 frames (40 milli-seconds) was scored as tethered.
In studies examining cell displacement, DiOC 6 -labeled platelets were perfused through fibrinogen-coated microcapillary tubes for 5 minutes at 150 s -1 . Platelets were considered as dislodged when exhibiting spatial displacement on the surface greater than one platelet diameter from their initial attachment point (12). Similar analysis was used for calcium-dye loaded human and murine platelets when perfused across a fibrinogen matrix (see below), except that platelet adhesion was monitored for 100 frames (0.586 frames/sec) using confocal fluorescence microscopy (TCS-SP, Leica Germany). Similar analysis was used to examine adhesion strength at 7 high shear (1800 s -1 ), however in this case, platelets were first perfused at 150 s -1 for 5 minutes, followed by an increase in wall shear rate to 1800 s -1 for a further 60 seconds.

Analysis of Cytosolic Calcium Flux under Static and Flow Conditions -Changes in cytosolic
calcium levels were monitored according to published methods (14,26). Briefly, washed platelets (1 x 10 9 /ml) were loaded with Oregon Green 488 BAPTA-AM (1 µM) and Fura-Red AM (1.25 µM) for 30 minutes at 37 o C. For mouse platelets, the calcium dyes were loaded at a platelet density of 2 x 10 8 /ml in the presence of 1.25 mM Probenicid. Dye-loaded platelets (1 x 10 7 /ml) were then either allowed to adhere to fibrinogen under static conditions, or reconstituted with RBCs (50%) prior to perfusion through fibrinogen-coated microcapillary tubes at 150 s -1 for human platelets, and 600 s -1 for mouse platelets. To examine changes in calcium flux, sequential confocal images of adherent platelets were captured at a scan rate of 0.586 frames/s for 37.5 seconds at the indicated time points. Real time platelet calcium flux was calculated from ratiometric fluorescence measurements and converted to intracellular calcium concentrations as previously described (14,26).

Role of ADP in Promoting Integrin α IIb β 3 Activation and Platelet Spreading on Immobilized
Fibrinogen -Immobilized fibrinogen supports the adhesion and activation of platelets through engagement of integrin α IIb β 3 . Previous studies have suggested an important role for ADP in promoting lamellipodial formation in fibrinogen-adherent platelets, however its role in promoting integrin α IIb β 3 activation remains less clear (18,23). To investigate this, we examined the effect of apyrase on integrin α IIb β 3 activation during platelet adhesion to fibrinogen by performing indirect immunofluorescence studies using the activation-specific α-integrin α IIb β 3 antibody, PAC-1. As demonstrated in figure 1A, robust PAC-1 binding and platelet spreading was observed in control and apyrase-treated platelets following adhesion to fibrinogen. Using confocal imaging, we demonstrated that PAC-1 staining occurred predominantly at the granulomere on the apical surface of spreading platelets ( Fig 1A) and quantification of PAC-1 fluorescence on the surface of spread platelets revealed no difference between control and apyrase-treated platelets ( Fig 1B).
The inability of apyrase to inhibit PAC-1 binding and spreading was unlikely to be due to incomplete inhibition of ADP, as blocking the two major ADP purinergic receptors, P2Y 1 and P2Y 12 , with AR-C69931MX and A3P5PS, respectively, did not inhibit these platelet responses (data not shown). These findings suggest that integrin α IIb β 3 engagement of fibrinogen can modulate the affinity status of integrin α IIb β 3 and induce lamellipodial extensions independent of ADP.
To examine in further detail the relationship between ADP release and platelet spreading, time course adhesion assays were performed. As demonstrated in figure 1C, platelets pretreated with apyrase spread significantly slower than control platelets, with half maximal spreading of control platelets occurring within <10 minutes compared to 20-30 minutes for apyrase-treated platelets, however by 50 minutes there was no difference in spreading between control and apyrase-treated platelets. In both the fixed end-point and time-course adhesion assays, there was no further decrease in the rate or extent of platelet spreading when aspirin was combined with apyrase, suggesting that TxA 2 was not making a significant contribution to this response.

Role of ADP in Promoting Cytosolic Calcium Flux during Platelet Adhesion on Fibrinogen -
Mobilization of intracellular calcium during platelet adhesion to fibrinogen is important for cytoskeletal remodeling and has been demonstrated to be ADP-dependent (32). To investigate the absolute requirement for ADP for fibrinogen-induced cytosolic calcium flux, confocal imaging studies were performed on adherent platelets labeled with the calcium-indicator dyes Oregon green BAPTA and Fura red. Consistent with previous reports (26,33), platelets firmly adherent to fibrinogen elicited a sustained, oscillatory calcium response that coincided with lamellipodial extension and platelet spreading. As demonstrated in figure 2A, pretreating platelets with apyrase and aspirin markedly delayed the onset of the calcium response, however by 50 minutes the majority of platelets exhibited a sustained calcium response and spread. Furthermore, analysis of the pattern of the cytosolic calcium response in individual platelets ( Fig 2B) revealed no difference in the frequency or magnitude of calcium oscillations between control and apyrase/aspirin-treated platelets (data not shown). In all studies, combining aspirin with apyrase had no further inhibitory effect beyond that observed with apyrase alone, excluding a major role for TXA 2 in promoting cytosolic calcium flux (data not shown). Overall, these studies define an important, albeit non-essential role for ADP in promoting cytosolic calcium flux during platelet adhesion on fibrinogen.

Src Kinases are Essential for Integrin α IIb β 3 Activation and Calcium Mobilization Following
Platelet Adhesion to Fibrinogen -The demonstration that ADP antagonists did not eliminate platelet activation induced by immobilized fibrinogen, raised the possibility that fibrinogen engagement of integrin α IIb β 3 was sufficient to induce cytosolic calcium flux through outside-in signaling processes. Src kinases play a central role in integrin signaling and a recent study (22) has demonstrated an important role for Src kinases in integrin α IIb β 3 -dependent cytoskeletal remodeling. To examine the role of Src kinases in promoting integrin α IIb β 3 activation and calcium flux, platelets were pretreated with the Src kinase inhibitor, PP2 (34). As demonstrated in figures 3A and B, treatment of platelets with PP2 completely eliminated calcium responses in all platelets adhering to fibrinogen. This reduction in calcium flux was associated with an inhibition of PAC-1 binding and platelet spreading (Fig. 3C). Dose-response studies demonstrated a strong correlation between inhibition of PAC-1 binding and platelet spreading, over a concentration range previously demonstrated to be selective for Src kinase inhibition ( Fig. 3D and data not shown) (34). These findings suggest an important role for Src kinases in promoting cytosolic calcium flux.

PLC and IP 3 Contribute to Integrin α IIb β 3 Activation and Platelet Spreading on Immobilized
Fibrinogen -The demonstration that Src kinases promote calcium signals on fibrinogen raised the possibility that one or more phospholipase C (PLC) isoforms may be regulated downstream of integrin α IIb β 3 . Src kinases phosphorylate and activate PLCγ1 and PLCγ2 (35), however PLCγ2 is the predominant isoform present in platelets (36). To investigate changes in tyrosine phosphorylation of PLCγ2, platelets were allowed to adhere to fibrinogen in the presence or absence of PP2, and the phosphorylation status of PLCγ2 assessed by performing antiphosphotyrosine immunoblots on PLCγ2 immunoprecipitates. As shown in figure 4A, adhesion of platelets to fibrinogen evoked tyrosine phosphorylation of PLCγ2, whereas PP2-treatment completely abolished this phosphorylation event. To investigate the functional importance of PLC for integrin α IIb β 3 activation and platelet spreading on fibrinogen, platelets were treated with the PLC inhibitor, U73122 or the IP 3 antagonist, APB-2. As demonstrated in figure 4B, both pharmacological inhibitors completely blocked PAC-1 binding and platelet spreading on fibrinogen and eliminated cytosolic calcium flux (data not shown). Taken together, these studies suggest a potentially important role for one or more Src kinase-regulated PLC isoforms in integrin α IIb β 3 outside-in signaling.

Role of PLCγ2 in Promoting Integrin α IIb β 3 -dependent Calcium
Flux -To investigate the potential role for PLCγ2 in platelet spreading on fibrinogen and integrin α IIb β 3 calcium signaling, adhesion studies were performed with platelets derived from PLCγ2 -/mice. These platelets have been demonstrated to have a major defect in their activation response to soluble and fibrillar collagen, although their responsiveness to a variety of soluble agonists, including ADP, appears intact (37). PLCγ2 -/platelets adhered normally to immobilized fibrinogen, however spreading was significantly delayed relative to PLCγ2 +/+ platelets ( Fig 5A). Maximal spreading was observed within 30 minutes for PLCγ2 +/+ platelets compared to 60 minutes for PLCγ2 -/platelets.
As demonstrated in figure 5B, this reduction in the rate of platelet spreading correlated with a delay in the proportion of platelets exhibiting a cytosolic calcium response. Within 30-40 minutes of adhesion, 80-100% of PLCγ2 +/+ platelets elicited an oscillatory calcium response compared with <15% of the PLCγ2 -/platelets, however by 60-70 minutes, essentially all PLCγ2 -/platelets had undergone a sustained calcium response ( Fig. 5B and C) and adopted a spread morphology.
Examination of the role of ADP in promoting mouse platelet cytosolic calcium flux and spreading on a fibrinogen matrix revealed similar findings to that observed with human platelets, in that both the rate of spreading and proportion of platelets exhibiting a cytosolic calcium response was markedly lower in apyrase-treated platelets (compare Fig 5 with 6, and data not shown). Significantly, PLCγ2 -/platelets treated with apyrase failed to extend lamellipodia and spread, however these platelets still underwent morphological changes following adhesion with the majority of platelets becoming spherical and extending filopodia (Fig 6A). Consistent with their morphological defects, these PLCγ2 -/platelets also exhibited minimal calcium flux (Fig 6B   and C). These studies suggest an important role for PLCγ2 in integrin α IIb β 3 outside-in signaling relevant to platelet spreading on fibrinogen.

Investigation of the Role of ADP in Promoting Platelet-fibrinogen Interactions under Flow -
Our studies to date have demonstrated an important co-operative signaling role for integrin α IIb β 3 and ADP in promoting platelet activation on fibrinogen under static adhesion conditions. To investigate the significance of these findings with respect to platelet adhesion in a shear field, flow-based adhesion assays were performed on fibrinogen-coated microcapillary tubes. In preliminary studies, we confirmed that adhesion of platelets to fibrinogen inversely correlated with shear rate, with maximal adhesion at low shear (150 s -1 ) and progressively fewer plateletmatrix interactions at higher shears (600-1800 s -1 ) (data not shown). Analysis of the cytosolic calcium response under low shear conditions (150 s -1 ) demonstrated that adherent platelets exhibited an oscillatory calcium response that was associated with firm platelet adhesion and spreading. Consistent with previous reports (33), the onset of the calcium response was heterogeneous across the platelet population with lag times ranging between 10 and 200 secs following platelet adhesion. As demonstrated in figure 7A and 7B, pretreating platelets with apyrase had no effect on the ability of platelets to tether to the fibrinogen matrix, however, there was a small but significant reduction (~20%) in the proportion of tethered platelets exhibiting a sustained calcium response. Consistent with our static adhesion data, the reduced calcium signaling in apyrase-treated platelets was the result of a delay in the onset the calcium response, however at later time points the percentage of platelets exhibiting a sustained calcium response was not significantly different between control and apyrase-treated platelets (Fig. 7B). The decrease in calcium flux at the earlier time points was associated with a reduction in the number of platelets forming sustained adhesion contacts with the fibrinogen matrix. As demonstrated in figure 7C, up to 12% of platelets tethering to the matrix at 150 s -1 , were displaced within 10 seconds of tethering. Pretreating platelets with apyrase approximately doubled the proportion of platelets detaching from the point of initial contact. Furthermore, increasing the tensile stress on formed bonds by exposing platelets initially adherent at 150 s -1 to rapid increases in shear up to 1800 s -1 , resulted in up to 60% of apyrase-treated platelets detaching from their point of initial contact compared to 20% of control platelets (Fig 7D). conditions, platelets were pretreated with PP2 prior to perfusion through fibrinogen-coated microcapillary tubes. As demonstrated in figure 8A, PP2 had no effect on the ability of platelets to tether to fibrinogen, however, it completely abolished cytosolic calcium flux in all adherent platelets (Fig. 8B), resulting in defective platelet spreading (data not shown). Analysis of the stability of platelet adhesion contacts at 150 s -1 , revealed that inhibition of calcium flux by PP2 resulted in 30% of platelets displacing from the point of initial attachment (Fig 8C), whereas >90% of platelets were displaced when exposed to sudden increases in shear (Fig 8D). Several lines of evidence suggest that the defect in the stability of platelet adhesion following PP2 or apyrase treatment was primarily a result of reduced calcium signaling. First, analysis of control and apyrase-treated platelets following exposure to sudden increases in shear, revealed that all platelets exhibiting a sustained oscillatory calcium response resisted the detaching effects of increased shear, whereas platelets without a detectable calcium response were easily detached ( Fig 9A). Second, pretreating platelets with intracellular calcium chelators resulted in the formation of unstable adhesion contacts in 100% of platelets ( Fig 9B). Finally, pretreating platelets with the PLC inhibitor, U73122 (Fig 9B), or the IP 3 receptor antagonist, APB-2, abolished stable platelet-fibrinogen interactions. In initial studies we demonstrated that adhesion of wild-type murine platelets to fibrinogen was also shear rate dependent, however in contrast to human platelets, maximal adhesion occurred at 600 s -1 instead of 150 s -1 , with progressively less adhesion at higher shears (data not shown). The ability of murine platelets to adhere efficiently at higher shear rates may reflect the smaller dimensions of these cells relative to human platelets, leading to reduced tensile stress on adhesive bonds. Alternatively, these differences may reflect distinct binding kinetics between murine integrin α IIb β 3 and human fibrinogen. As observed with human platelets, pretreating PLCγ2 +/+ murine platelets with apyrase had no effect on the initial adhesion of these cells to fibrinogen (data not shown), however it reduced the capacity of these platelets to maintain sustained adhesion contacts in response to rapid increases in shear (Fig 10A). This reduction in stable adhesion was associated with a decreased proportion of platelets exhibiting sustained calcium oscillations during the early stages of adhesion (Fig 10B). It should be noted that displacement of mouse platelets was distinct from that observed with human platelets in that the latter typically detached from the matrix with sudden increases in shear, whereas mouse platelets translocated slowly over the fibrinogen matrix. This difference may be attributable to the species variation outlined above.

Role of PLCγ2 in Promoting Sustained Platelet Adhesion under Flow
Analysis of PLCγ2 -/platelets demonstrated a small non-significant reduction in the proportion of platelets exhibiting a sustained oscillatory calcium response in the absence of apyrase (Fig 10B). Thus, >85% of adhesion contacts formed by PLCγ2 -/platelets remained stable following exposure to a sudden increase in tensile stress (Fig 10A). In the presence of apyrase, <5% of PLCγ2 -/platelets elicited a sustained calcium flux following adhesion to fibrinogen ( Fig   10B) and exposure of these platelets to sudden increases in shear resulted in a high proportion of these platelets (65%) becoming displaced from the point of initial contact (Fig. 10A). Together, these findings demonstrate an important role for PLCγ2 in integrin α IIb β 3 outside-in signaling relevant to stable platelet adhesion on fibrinogen. Furthermore, similar to human platelets, they suggest an important role for ADP in stabilizing platelet-fibrinogen interactions under flow.

DISCUSSION
The studies presented here provide new insight into the mechanisms regulating platelet adhesion on immobilized fibrinogen. In contrast to previous reports (10-13), our studies demonstrate that firm platelet adhesion on fibrinogen is not an instantaneous irreversible event, but is in fact activation-dependent. More specifically, our studies suggest that maintenance of integrin α IIb β 3fibrinogen bonds is dependent on integrin α IIb β 3 -derived calcium signals. These calcium signals appear to be regulated by one or more Src kinase family members linked to the activation of PLCγ2 -/-. Our studies also define an important role for ADP in potentiating integrin α IIb β 3 calcium signals. The release of dense granule ADP, in concert with the activation of PLCγ2, appears to play a major role in promoting irreversible platelet adhesion and spreading on fibrinogen.
Our studies define an important role for integrin α IIb β 3 outside-in calcium signals in promoting firm platelet adhesion and spreading on fibrinogen. In contrast to previous reports demonstrating that cytosolic calcium flux on immobilized fibrinogen is dependent on released ADP (24,25), we have provided several lines of evidence suggesting that integrin α IIb β 3 outside-in signaling per se is sufficient to induce calcium flux. First, a sustained oscillatory calcium response was observed in fibrinogen-adherent platelets under experimental conditions eliminating the platelet activating effects of ADP and TxA 2 . Second, the magnitude, pattern and duration of the cytosolic calcium response in these platelets was similar to that previously described for integrin α IIb β 3 calcium signals (14,15) but distinct from soluble agonist-induced calcium flux (38,39) (unpublished observations). Third, the cytosolic calcium response was strictly dependent on Src kinases. These non-receptor tyrosine kinases play a critical role in integrin signal transduction but are not essential for soluble agonist-induced calcium flux (21) (unpublished observations). Finally, the demonstration that PLCγ2 -/mouse platelets have a defective calcium response during platelet adhesion to fibrinogen is consistent with its previously defined role in adhesion receptor signal transduction (40).
The finding for an important role for PLCγ2 -/in integrin α IIb β 3 outside-in signaling is consistent with recent findings that PLCγ2 becomes tyrosine phosphorylated following integrin α IIb β 3 ligation (41), a finding confirmed in the present study. It is also consistent with the proposed involvement of ITAM-like signaling processes in integrin α IIb β 3 signaling. For example, there is strong evidence for an important role for the non-receptor tyrosine kinases including Src family kinases and Syk in initiating integrin α IIb β 3 signaling. The Src kinase Fyn, has been demonstrated to phosphorylate ITAM tyrosine motifs as well as the β 3 tail of integrin α IIb β 3 (22,31,42). Similarly Syk, which binds to the cytoplasmic tails of ITAM-bearing receptors (22,43) also binds integrin α IIb β 3 . Other similarities between integrin α IIb β 3 and ITAM signaling include the recruitment of adaptor molecules to ligated receptors, such as LAT in the case of ITAM-receptors (35,44) and Shc with integrin α IIb β 3 (45). These molecules have been demonstrated to facilitate the binding and activation of signaling molecules such as PLCγ2 and PI 3-kinase, promoting PI turnover and cytosolic calcium flux.
It is generally assumed that ADP release is essential for a sustained oscillatory calcium response and platelet spreading on fibrinogen (24,25). While our findings are consistent with an important role for ADP in promoting these platelets responses, they suggest that its role is not absolute, at least under the experimental conditions employed in this study. The reasons for the apparent differences between our findings and others are not clear, but may primarily reflect  (13). Without information on the percentage of tethered cells forming immediate firm adhesion contacts or detaching from the matrix and no information on the effects of increased tensile strength on adhesive interactions, it is difficult to reconcile potential differences between our studies and previous results. Two distinct models have been proposed to describe platelet adhesion on fibrinogen and vWf under flow conditions. Platelet adhesion to vWf is a multi-step process involving GPIb/V/IX and integrin α IIb β 3 . Adhesion is initiated by a reversible tethering step between the A1 domain of vWf and GPIbα, followed by a secondary adhesive interaction between the C1 domain of vWf and integrin α IIb β 3 . This two-step model of platelet adhesion is critically dependent on the cooperative signaling function of both receptors, with GPIbα-derived calcium spikes initiating integrin α IIb β 3 activation and transient bond formation with vWf, whereas subsequent integrin α IIb β 3 -dependent calcium signals sustain firm adhesion. An important difference between immobilized vWf and fibrinogen is that the latter ligand can engage integrin α IIb β 3 in its low affinity state, and according to previous findings, maintain firm adhesion independent of platelet activation. However, our findings suggest a reinterpretation of this model in which the mechanisms regulating platelet adhesion on vWf are also relevant to fibrinogen. Thus, initial bond formation between fibrinogen and integrin α IIb β 3 is potentially transient and reversible, depending on the tensile stress applied to the formed bonds. Conversion of these bonds to firm adhesion contacts is dependent on integrin α IIb β 3 -dependent calcium flux and also on the release of ADP. These calcium changes are likely to support firm adhesion through multiple mechanisms, including effects on integrin α IIb β 3 affinity, receptor clustering and through reorganization of the cytoskeleton. Elucidating the mechanisms co-ordinating integrin α IIb β 3calcium signals with ADP release will be important to fully understand the processes regulating platelet adhesion efficiency on fibrinogen.       Tyrode's buffer were reconstituted with 50% RBCs, and labeled with DiOC 6 (1 mM), as described under "Experimental Procedures". Platelet suspensions were treated with buffer (Control) or apyrase (1.5 U/ml, Apyrase) prior to perfusion through fibrinogen-coated microcapillary tubes, at a wall shear rate of 150 s -1 . The total number of platelets tethering to the fibrinogen substrate over a 90 second period was calculated as outlined under "Experimental Procedures". B, Calcium dye-loaded platelets were perfused through fibrinogen-coated microcapillary tubes as described above in the presence (Apyrase) or absence (Control) of apyrase (1.5 U/ml). The percentage of cells exhibiting a sustained oscillatory calcium response was determined using confocal software as outlined under "Experimental Procedures". C, The percentage of cells displaced was determined after five minutes of perfusion, as described under "Experimental Procedures". This data represents the mean ± SEM from three individual experiments. D, DiOC 6 -labeled platelets treated with buffer (Control) or apyrase (Apyrase) were perfused through fibrinogen-coated microcapillary tubes at 150 s -1 for 5 minutes, followed by an increase in shear rate to 1800 s -1 . The ability of platelets to remain firmly adherent to the fibrinogen matrix was determined as described in "C". Results represent the percentage of fibrinogen-adherent platelets displaced following the shear increase (arrow).

Immobilized Fibrinogen under Low Shear Conditions -Washed platelets (5 x 10 7 /ml) in
Tyrode's buffer were reconstituted with 50% RBCs, and labeled with DiOC 6 (1 mM), or calcium dyes, where indicated. Platelet suspensions were pre-treated with vehicle alone (Control) or 10 µM PP2 ( PP2) for 10 minutes at 37 o C, prior to perfusion through fibrinogen-coated microcapillary tubes at a wall shear rate of 150 s -1 . A, The total number of DiOC 6 -labeled platelets tethered to the fibrinogen matrix over a 90 second period was calculated as outlined under "Experimental Procedures". B, The percentage of calcium dye-loaded platelets exhibiting a sustained oscillatory calcium response was determined using confocal software as outline under "Experimental Procedures". This data represents the mean ± SEM of three experiments. C, The percentage of cells displaced was determined after five minutes of perfusion, as described under "Experimental Procedures". This data represents the mean ± SEM from eight separate experiments. D, DiOC 6 -labeled platelets treated with vehicle alone (Control) or PP2 (PP2, 10 υM) were perfused through fibrinogen-coated microcapillary tubes at 150 s -1 for 5 minutes, followed by an increase in shear rate to 1800 s -1 . Platelet displacement was quantified as described in "C". Results represent the percentage of fibrinogen-adherent platelets displaced following the shear increase (arrow).

Figure 9. Cytosolic Calcium Flux Maintains Firm Platelet Adhesion on Fibrinogen -A,
Calcium dye-loaded platelets were perfused for 5 minutes through fibrinogen-coated microcapillary slides at 150 s -1 , followed by an increase in shear rate up to 1800 s -1 (see Figure   7D). The cytosolic calcium levels in platelets that displaced or remained adherent were examined, and divided into cells exhibiting Low Ca 2+ (platelets with mean calcium levels <100nM) or Elevated Ca 2+ (platelets with mean calcium levels >100nM). The data represents the % of platelets exhibiting Low or Elevated Ca 2+ that became displaced following the shear increase.
Mean ± SEM from three separate experiments. B, Calcium dye-loaded platelets, treated with either vehicle (control), U73122 (1 µM) or DM-BAPTA (70 µM), were initially perfused through fibrinogen-coated microcapillary slides at 150 s -1 , followed by an increase in shear rate up to 1800 s -1 . Calcium flux was monitored by capturing consecutive confocal images 2 seconds before and 28 seconds after the increase in shear. Results represent the mean ± SEM of 3 individual experiments, and depict the percentage of fibrinogen-adherent platelets displaced following the shear increase (arrow).

Figure 10. PLCγ2 Promotes Platelet Calcium Mobilization and Stable Adhesion to Immobilized
Fibrinogen under Shear Conditions -A, Platelets initially perfused through fibrinogen-coated microcapillary tubes at 600 s -1 were subsequently subjected to an increase in shear rate to 10,000 s -1 . Consecutive confocal images were taken 2 seconds before and 28 seconds after the increase in shear rate, and the percentage of fibrinogen-adherent platelets displaced following the shear increase (arrow) quantified, as described under "Experimental Procedures". Results represent the mean ± SEM of three individual experiments ( ** p<0.01). B, Calcium dye-loaded platelets from PLCγ2 +/+ or PLCγ2 -/mice were perfused through fibrinogen-coated microcapillary tubes (100 µg/ml) at 600 s -1 in the absence (-) or presence (+) of apyrase (1.5 U/ml). The percentage of cells exhibiting a sustained oscillatory calcium response was determined. The data represents the mean ± SEM of three individual experiments.