In the 1880s, Bizzozero 1 and Hayem, 2 working independently,

Transcription

1 Platelets and Thrombin Generation Dougald M. Monroe, Maureane Hoffman, Harold R. Roberts Abstract This review examines the evidence that platelets play a major role in localizing and controlling the burst of thrombin generation leading to fibrin clot formation. From the first functional description of platelets, it has been recognized that platelets supply factors that support the activation of prothrombin. Studies have demonstrated that on activation, the amount of one specific lipid, phosphatidylserine, is significantly increased on the outer leaflet of platelet membranes. When it was found that phosphatidylserine containing lipid extracts could be substituted for platelets in clotting assays, this suggested the possibility that changes in platelet lipid composition were necessary and sufficient to account for platelet surface thrombin generation. Because a growing body of data suggest that platelet-binding proteins provide much of the specificity for platelet thrombin generation, we review in this report data suggesting that changes in lipid composition are necessary but not sufficient to account for platelet surface regulation of thrombin generation. Also, we review data suggesting that platelets from different individuals differ in their capacity to generate thrombin, whereas platelets from a single subject support thrombin generation in a reproducible manner. Individual differences in platelet thrombin generation might be accounted for by differences in platelet-binding proteins. (Arterioscler Thromb Vasc Biol. 2002;22: ) Key Words: platelets thrombin lipids phosphatidylserine coagulation In the 1880s, Bizzozero 1 and Hayem, 2 working independently, wrote about a blood particle, previously observed by others as a colorless corpuscle smaller than red or white cells, that they called a hematoblast or platelet. Bizzozero wrote that accumulation of this cell might account for the 1875 observation of Zahn 3 that bleeding from an injury to a blood vessel was initially blocked by a white thrombus. These studies showed that fibrin was associated with these cells and led them to conclude that platelets supplied a factor that was required for coagulation. 1,2 Subsequently, a number of investigators demonstrated that the rate of clotting and prothrombin conversion to thrombin was decreased in platelet-poor plasma and increased as a function of platelet number. 4 6 See Cover The initial studies of clotting assays relied on patient platelets in the platelet-rich plasma as the surface for thrombin generation. This led to considerable variability in the results from patient to patient. 6 To provide an assay reagent that gave more reproducible results, platelet-poor plasma was clotted with phospholipid. 7 Assay reproducibility was also enhanced by the addition of suspensions of diatomaceous earth to activate clotting. 8 This modified assay gave reproducible clotting values and was valuable for the diagnosis of bleeding disorders as well as for establishing the relationships between the coagulation proteins. But as a result, the use of lipids as a substitute for platelets became widespread. In studies involving how lipids influence coagulation assays, it was observed that there was a correlation between the amount of a specific lipid, phosphatidylserine, and the ability of a lipid surface to promote thrombin generation. 9 Other studies looking at platelet activation showed that circulating unactivated platelets had low levels of phosphatidylserine on the outer leaflet of their membranes. 10 However, when platelets were activated, the amount of phosphatidylserine on the outer leaflets increased dramatically from 2% of the phospholipid content to as much as 12%. 11,12 Resting platelets do not significantly promote factor X or prothrombin activation. 13 Even on activated platelets, procoagulant activity is lost on treatment with phospholipases that cleave the head groups from phosphatidylserine. 14 Because there was an obvious correlation between the observation that activated platelets express phosphatidylserine and the observation that phosphatidylserine in purified lipids is required for thrombin generation, it was concluded by many that phosphatidylserine exposure provided the primary mechanism for regulating coagulation reactions and thrombin generation. 10 Cascade Model of Coagulation Coagulation, as shown in Figure 1, has often been represented as 2 somewhat independent pathways that converge to a common pathway, with thrombin generation as the end point of the reactions. This model represents an evolution of the initial models of coagulation based on a cascade or waterfall Received April 25, 2002; revision accepted July 17, From the Center for Thrombosis and Hemostasis (D.M.M., H.R.R.), University of North Carolina, Chapel Hill, and the Department of Pathology (M.H.), Duke University, Durham, NC. Correspondence to Dr Dougald M. Monroe, University of North Carolina at Chapel Hill, Hematology/Oncology, 932 Mary Ellen Jones Bldg, CB 7035, Chapel Hill, NC dmonroe@med.unc.edu 2002 American Heart Association, Inc. Arterioscler Thromb Vasc Biol. is available at DOI: /01.ATV

2 1382 Arterioscler Thromb Vasc Biol. September 2002 Figure 1. Cascade model of coagulation. This scheme is a modification of previously proposed models 15,16 and reflects the components tested for in the prothrombin time assay (extrinsic pathway) and the activated partial thromboplastin time assay (intrinsic pathway). The intrinsic pathway consists of HK, factor XII, prekallikrein (PK), factor XI, factor IX, factor VIII, factor X, factor V, and prothrombin (II), which is converted to thrombin (IIa). The activated form of these factors is indicated by adding the letter a as a suffix (eg, IXa). The extrinsic pathway consists of TF, factor VII, factor X, factor V, and II. Reactions requiring a phospholipid surface are indicated by lipid. hypothesis. 15,16 These models give a good representation of the processes observed in clinical coagulation laboratory tests. The prothrombin time measures the factors of the so-called extrinsic pathway, and activated partial thromboplastin time measures factors in the intrinsic pathway. Although this model reflects some of the interactions of the proteins, it has inadequacies as a model of the in vivo hemostatic processes. For example, deficiencies of factor XII, high molecular weight kininogen (HK), or prekallikrein do not cause clinical bleeding. Furthermore, others have shown that under normal circumstances, hemostasis is initiated by tissue factor (TF). In addition, if the intrinsic and extrinsic pathways are essentially separate, then activation of factor X by the extrinsic pathway should compensate for a lack of factor VIII or IX. However, factors VIII and IX are clearly essential for hemostasis, because their absence results in the bleeding seen in hemophilia, and this bleeding is not compensated by the intact extrinsic system. Interestingly, factor XI deficiency results in a bleeding disorder that has significant variability between patients and is significantly milder than bleeding in patients with hemophilia A or B lacking factors VIII or IX, respectively It has long been established that coagulation reactions do not occur physiologically in solution but are localized to a surface. 6,20 22 The critical advantage to evolving such a mechanism is that the reactions are confined to a specific site of injury rather than being completely disseminated. Implicit in the view of coagulation shown in Figure 1 is the assumption that the level and amount of coagulation factors drive and regulate coagulation and that the surfaces are essentially passive agents. Also, implicit in the view of coagulation shown in Figure 1 is that the role of cells is primarily to provide a lipid-rich surface for coagulation complex assembly and that all surfaces with similar lipid composition are essentially identical. This supposition is contradicted by a number of studies, including those showing that endothelial cells, which are not considered procoagulant, nonetheless express significant levels of phosphatidylserine on their surface. 23 This phosphatidylserine is important for maintaining the anticoagulant activity of endothelium by enhancing thrombomodulin activity. 23 A direct comparison of platelets and phospholipids has shown that platelets have complex coagulant activities that are not completely mimicked by phospholipids. 24,25 Also, a careful analysis of the kinetics of platelet activity was not consistent with the kinetics seen on lipid surfaces. 26 This and other recent evidence suggests that platelet protein components provide virtually all of the specificity and control mechanisms that determine the expression of procoagulant activity and subsequent thrombin generation. Cell-Based Model of Coagulation A useful model to describe coagulation is shown in Figure 2. This model is derived in part from experiments that use cells such as monocytes or fibroblasts as a source of TF (rather than relipidated TF) and activated platelets as a surface for thrombin generation. 27,28 In this model, coagulation occurs in 3 overlapping phases: initiation, priming, and propagation During the process of hemostasis, a break in the vessel wall brings plasma into contact with TF-bearing cells. This TF may be derived entirely from extravascular sources, such as fibroblasts, or may in part be derived from encrypted sources in blood through a CD62 (P-selectin)/CD15 (Pselectin glycoprotein ligand 1) mediated mechanism Factor VII binds to TF and is rapidly activated 35 by coagulation proteases 36 and by noncoagulation proteases, depending on the cellular location of the TF. 37 The factor VIIa/TF complex activates factor X and factor IX (Figure 2). The activated forms of these 2 proteins, even though activated at the same site, play very different and distinct roles in subsequent coagulation reactions. Factor Xa can activate plasma factor V on the TF cell 38 (as can other cellular proteases 39 ). If factor Xa diffuses from the protected environment of the cell surface from which it was activated, it can be rapidly inhibited by the TF pathway inhibitor or antithrombin. However, the factor Xa that remains on the TF cell surface can combine with factor Va to produce small amounts of thrombin. 30,40 This thrombin, although not sufficient to cleave fibrinogen throughout a wound, nonetheless plays a critical role in amplifying the initial thrombin signal, as shown in Figure The initial factor VIIa/TF complex is subsequently inhibited by the action of the TF pathway inhibitor in complex with factor Xa. 41,42 In the priming phase (Figure 2), the small amount of initial thrombin binds to platelets that have adhered to extravascular matrix components at the site of injury mediated in part by the binding of von Willebrand factor to collagen. 43 The process of binding to matrix proteins, especially collagen, partially activates platelets 44 and also localizes them near a site of TF exposure. Thrombin enhances platelet activation via proteaseactivated receptor (PAR) mechanisms. There appears to be a synergy between the collagen activation of platelets and the thrombin activation of platelets, such that dual stimulation by those 2 agonists results in platelet activity higher than that

3 Monroe et al Platelets and Thrombin Generation 1383 Figure 2. Cell-based model of coagulation. In this scheme, coagulation occurs in 3 phases: initiation, priming, and propagation. In the initiation, factor VIIa bound to TF activates factor IX and also factor X. Factor Xa then activates factor V on the TF-bearing cell, complexes with factor Va, and converts a small amount of II to IIa. 30 In the priming phase, the small amount of initial IIa activates platelets, causing release of granule contents including factor V, activates factor V, activates factor XI, and activates factor VIII by cleaving it from von Willebrand factor (vwf). Cofactors bind to the platelet surface before their respective enzymes. 29 The factor VIIa/TF complex is shut down through the action of the TF pathway inhibitor (TFPI) in complex with factor Xa. In the propagation phase, factor IXa generated by factor VIIa/TF binds to the activated platelets and subsequently activates factor X. This factor IXa is supplemented by factor IXa generated on the platelet surface by factor XIa. Factor Xa then moves directly into a protected complex with factor Va, resulting in a burst of thrombin generation. seen with stimulation by either agonist alone. 45 Thrombin activation of platelets results in degranulation that releases partially active factor V from platelet granules. 45,46 Costimulation of the platelet collagen receptor results in a subset of the platelet population expressing high levels of factor V. 45 Thrombin cleaves the partially activated factor V to a fully active form. 47 Thrombin also cleaves factor VIII, releasing it from von Willebrand factor. 47 In addition, thrombin activates factor XI bound to the platelet surface. 48,49 The result of this stage is a primed activated platelet that rapidly binds the cofactors Va and VIIIa as well as factor XIa. 29 In the propagation phase (Figure 2C), the factor IXa/VIIIa complex assembles when factor IXa reaches the platelet surface. The initial factor IXa formed by the factor VIIa/TF complex can diffuse to the platelet surface, because factor IXa is not rapidly inhibited by antithrombin or other plasma protease inhibitors. Additionally, platelet surface factor XIa can then provide additional factor IXa directly on the platelet surface. 48 Factor X is recruited to the activated platelet surface and is activated by the factor IXa/VIIIa complex. This allows factor Xa to move directly into a protected complex with factor Va, 50 where (in the presence of prothrombin) factor Xa is protected from the TF pathway inhibitor 51 and antithrombin, even in the presence of heparin. 52 Platelet surface factor Xa/Va complexes generate a burst of thrombin sufficient to form a stable hemostatic fibrin clot. This model of coagulation provides a rational explanation for the bleeding tendency in hemophilia. The TF pathway is intact in hemophilic patients so that the initiation and amplification steps occur. This leads to normal platelet activation and may account for the tendency of hemophiliacs to stop bleeding as a result of initial platelet plug formation This also accounts for the normal bleeding time characteristic of hemophilia. However, severe delayed bleeding occurs in hemophilia because there is deficient platelet-dependent factor Xa and thrombin generation throughout the platelet plug. 54,55 Factor Xa generated by the factor VIIa/TF complex cannot diffuse to the platelet surface without being inhibited. The only factor Xa that can be incorporated efficiently into prothrombinase complexes is that formed on the platelet surface by the factor IXa/VIIIa complex in proximity to factor Va. 50,56 Thus, in hemophilia, there is a failure of platelet-surface factor X activation, leading to a decrease in platelet-surface thrombin generation and ineffective clot formation. Role of Platelet Lipids Exposure of phosphatidylserine as platelets become activated is necessary for platelet procoagulant activity and is regulated by active transport mechanisms, including a flip-flop mechanism in which phosphatidylserine from the inner leaflet becomes exposed on the outer leaflet of the activated platelet surface. 14,57 59 Bleeding has been observed in individuals who have defects in this flip-flop mechanism, as seen in Scott syndrome. 60,61 This defect is also seen in a canine model with a platelet procoagulant defect. 62 Although initial work has indicated that the most significant role of phosphatidylserine in membranes is to bind the -carboxyglutamic acid (Gla) residues in coagulation proteins in a calcium-dependent fashion, 63 recent studies have indicated that the phosphatidylserine head group may instead bind factor Xa at a non-gla site to act as an allosteric regulator that turns on coagulation by enhancing prothrombin cleavage Exposure of phosphatidylserine on the outer leaflet of platelets is necessary for platelet procoagulant activity, and data support the idea that this exposure is sufficient to

4 1384 Arterioscler Thromb Vasc Biol. September 2002 account for the procoagulant nature of activated platelets (for review, see Heemskerk et al 59 ). However, there is increasing evidence that specific binding sites on platelets regulate formation of the coagulation complexes This conclusion is based in part on our own studies of differences in factor IXa/VIIIa and factor Xa/Va activity on platelets from different normal individuals. 69 These studies examined platelets from different subjects and showed differences between normal individuals in phosphatidylserine expression even when their platelets were activated in the same way. Although there were differences in the generation of factor Xa and thrombin between individuals, there was no correlation between levels of phosphatidylserine expression and generation of factor Xa or thrombin, as shown in Figure 3. These data support the concept that factors other than phosphatidylserine are necessary for platelet-dependent thrombin generation. Platelet-Binding Proteins Important for the Priming Step Although the clotting factors bind to platelets, not all this binding transduces a signal to the platelet. Therefore, some of the high-affinity binding sites for coagulation factors on platelets do not fit the classic definition of receptors and in this review will be referred to as binding proteins (Figure 4). The small amount of thrombin generated in the initiation step is critical in the development of the procoagulant response. 45,47 In the absence of a protected environment, thrombin is inhibited by plasma levels of antithrombin with a half-life of 1 minute. Unactivated platelets appear to contain at least 3 binding proteins for thrombin. One is a site on the glycoprotein (GP) Ib-IX-V complex that binds to thrombin through the heparin-binding site (thrombin anion binding exosite 2). 70 The second site is one of the PARs, PAR1. 71 PAR1 binds to thrombin through the extended substrate-binding site and anion-binding exosite 1. After this 7-transmembrane-domain protein is cleaved by thrombin, the new PAR amino terminus is a tethered ligand that binds to Figure 3. Phosphatidylserine exposure does not correlate with factor Xa or thrombin (IIa) generation on platelets. Data for this figure were taken from Sumner et al, 69 and detailed methods can be found in that study. Platelets from 14 normal individuals were activated with thrombin. Phosphatidylserine expression was measured by flow cytometry as calcium-dependent binding of fluorescein-labeled annexin V. Phosphatidylserine expression was characterized as low (1.4 to 1.7), medium (1.7 to 2.0), or high (2.1 to 2.4) on the basis of fluorescence units. Factor Xa generation was measured with factor IXa, factor VIIIa, and factor X. IIa generation was measured with factor Xa, and prothrombin (II) was measured with factor Va supplied by the platelets. Note that the y-axis scale for IIa generation does not go to zero. Figure 4. Platelet binding proteins important in coagulation. Known platelet proteins are labeled below the protein. Proteins that have not yet been identified but that are suspected are indicated by question marks. Binding proteins thought to be important in the priming phase are shown in panel A. Binding proteins thought to be important in the propagation phase are shown in panel B. Activity associated with proteins in the propagation phase is not found on unactivated platelets, and activation appears to be required either for physical appearance of the protein on the outer leaflet of platelets or for activation of these proteins as in the case of GP IIb-IIIa. f.xi indicates factor XI; f.x(a), either factor X or factor Xa; f.va, factor Va; f.ix(a), either factor IX or factor IXa; and f.viiia, factor VIIIa. another site on the same or neighboring PAR and triggers a signaling cascade. 71 Some signaling also occurs through PAR4, but this signal appears to be mainly associated with platelet aggregation rather than the development of platelet procoagulant activity. 72 The binding of thrombin to the GP Ib-IX-V complex significantly enhances the ability of thrombin to cleave PAR1 and catalyze platelet activation. 73 The binding of thrombin to the GP Ib-IX-V complex may also signal platelet activation through a fibrin-dependent mechanism. 74 Also, one group has tentatively identified a thrombinbinding site that is thought to be independent of the 2 sites mentioned above. 75 Unactivated factor VIII circulates in association with von Willebrand factor and can bind to platelets through the von Willebrand factor binding site on the GP Ib-IX-V complex. 76 Previous studies have shown that platelets potentiate thrombin activation of factor VIII. 77 This potentiation is maximal when von Willebrand factor is present. 78 We speculate that binding to the GP Ib-IX-V complex brings factor VIII/von Willebrand factor into proximity with thrombin and allows for efficient activation of factor VIII and release of the factor VIIIa from von Willebrand factor to the platelet surface. Factor XI shows saturable, specific, reversible binding to platelets with a K d of 10 nmol/l. 79 This binding is mediated through a cluster of residues on the third apple domain of factor XI 80,81 and occurs even in the absence of HK. 82 Baglia and Walsh 82 have shown that HK or prothrombin can enhance the binding of factor XI to platelets. Factor XI bound to the platelet surface is efficiently activated by thrombin, 48,82 which appears to be the physiological activator. 49 One recent

5 Monroe et al Platelets and Thrombin Generation 1385 study suggests that the binding site on platelets is on the GP Ib subunit of the GP Ib-IX-V complex. 83 Platelet-Binding Proteins Important for the Propagation Phase Factor Va can bind tightly to lipids, and lipid-binding may account for most or all of the procoagulant activity of factor Va on platelets. 84,85 Other studies suggest that there may be a factor Va binding protein on platelets. These studies show the following: (1) the kinetics of prothrombin activation are different on platelets and lipids 26 ; (2) factor Va can displace factor V from platelets, whereas factor V cannot completely displace factor Va 86 ; (3) platelets from 1 patient appear to be deficient in factor Va binding 87 ; and (4) a monoclonal antibody can block factor Va binding to platelets but does not block binding to lipid vesicles. 88 Studies of factor Va binding to platelets are complicated by the presence of factor V(a) in platelet granules so that, at present, the data are not compelling for a platelet-binding protein for factor Va. Studies by a number of workers have shown that factor Va acts as a binding protein for factor Xa on platelets. 25,88 90 There are also studies suggesting that effector cell protease receptor-1 (EPR1) may act as a platelet-binding protein to coordinate factor Xa interaction with factor Va on platelets. Bouchard et al 91 have shown that EPR1 is present on platelets and that an anti-epr1 antibody inhibited the formation of the factor Xa/Va complex in a dose-dependent and platelet donor dependent fashion. 91 Furthermore, they showed that formation of the factor Xa/Va complex blocked the binding of this anti-epr1 antibody. However, these data are contrasted by work that either failed to show the presence of EPR1 mrna 92 or that showed that EPR1 antigen was present on platelets but did not contribute to the assembly of the prothrombinase complex. 90 Combined, these studies suggest that further work is required to elucidate the role of EPR1 or other binding proteins in factor Xa activity on platelets. Factor IXa, in the absence of factor VIIIa, binds to activated platelets with a dissociation constant of 2 to 3 nmol/l. 93,94 This binding is in contrast to the binding of factor IXa to lipids, which has a K d that depends on the phosphatidylserine content of the lipids but is 500 nmol/l at 12% phosphatidylserine (the composition of activated platelets). 95 When the Gla domain of factor IXa is removed by proteolytic cleavage, the resulting des-gla factor IXa binds to platelets with a K d similar to that seen for normal factor IXa. 96 By contrast, des-gla factor IXa binding to lipids has a K d of nmol/l. 95 Also, the homologous protein factor VIIa binds to activated platelets with a K d 100 nmol/l, consistent with lipid surface binding. 97 Factor IXa binds to 500 sites on activated platelets. 93 In the presence of factor VIIIa, the dissociation constant decreases to 0.6 nmol/l. 93 Unactivated factor IX can bind to 250 of these sites with the same affinity as factor IXa. 93 Although factor IXa binding to platelets is enhanced by factor VIIIa, the affinity of zymogen factor IX for platelets is not altered by factor VIIIa. 93 This saturable, specific, reversible binding is consistent with a platelet-binding protein for factor IXa. Factor VIIIa binds rapidly (within 90 seconds) to thrombin- or collagen-stimulated platelets and plateletderived microparticles. 29,98 Factor VIII binding to microparticles occurred with on and off rates similar to those for the binding of factor VIII to lipid vesicles containing phosphatidylserine and was competed by lipid vesicles containing phosphatidylserine. 98 Factor Va, up to 20 nmol/l, enhanced factor VIII binding to microparticles but inhibited binding at higher concentrations. 98 By contrast, 2 other groups looking at factor VIIIa have shown that it binds to activated but not unactivated platelets, with 500 sites that have a dissociation constant of 3 nmol/l. 26,99,100 The binding occurs on platelets activated by thrombin or the PAR1 agonist peptide but not on platelets activated by ADP. 100 This binding is blocked by annexin but not by factor V. 99,100 In these studies, the lack of competition by factor V argues for a protein-mediated binding as opposed to a purely phosphatidylserine-dependent mechanism, 99,100 although one group has suggested that a specific moiety of phosphatidylserine has sufficient specificity to distinguish between factor Va and factor VIIIa. 101,102 Fibrinogen binding to platelets is mediated by the integrin GP IIb-IIIa ( IIB 3 ). 103 On unactivated platelets, GP IIb-IIIa has little affinity for soluble fibrinogen. 104 Platelet activation leads to activation of GP IIb-IIIa through inside-out signaling, resulting in a significant increase in the affinity for fibrinogen. 104 Platelet fibrinogen binding may be stabilized by covalent attachment of serotonin to fibrinogen. 105 Fibrinogen binding promotes platelet aggregation, presumably by a fibrinogen molecule acting as a bridge between GP IIb-IIIa molecules on different platelets. 104,106 There also appears to be outside-in signaling by the GP IIb-IIIa complex once fibrinogen is bound. 107 This signaling may alter the procoagulant ability of platelets, because blocking fibrinogen binding with abciximab, a monoclonal antibody directed against GP IIb-IIIa, decreased thrombin generation by 40% to 70% One group has suggested that prothrombin binding to GP IIb-IIIa may provide a pool of prothrombin for activation by the factor Xa/Va complex and that abciximab may decrease thrombin generation by blocking prothrombin binding. 112 Figure 5. Individual variability in platelet procoagulant function. Figure reprinted from Sumner et al, 69 with permission from Elsevier Science. Thrombin generation (IIase activity) and factor Xa generation (Xase activity) were measured as described in the legend for Figure 3. There was no significant correlation between thrombin generation and factor Xa generation. Note that neither the x-axis nor the y-axis goes to zero.

6 1386 Arterioscler Thromb Vasc Biol. September 2002 Figure 6. Individual variability in thrombin generation. Platelets from 2 different individuals were isolated and counted as described previously. 94 Thrombin generation was measured in a highly controlled setting in which platelets were incubated with plasma levels of proteins (prothrombin; factors XI, IX, X, VIII, and V; antithrombin III; and TF pathway inhibitor) and a cellular source of TF as described previously. 29 Individuals 1 and 2 are the same in panels A and B. Panel A shows thrombin generation as a function of time by using platelets from individual 1 (circles) and individual 2 (triangles). Panel B shows platelets from individual 1 and individual 2 in the complete model (filled symbol) or under identical conditions, except that factor XI was not included in the protein mixture (open symbol). These results are representative of the range of differences seen in studies of a larger number of individuals. 48 Variability of Platelet Procoagulant Response in Individuals As early as 1949, Buckwalter et al 6 noted wide variations in the ability of platelets from different individuals to convert prothrombin to thrombin, even when the platelet number was held constant. Subsequently, other workers have noted individual differences in (1) platelet procoagulant activity, 91,108 (2) thrombin generation in platelet-rich plasma even when clotting factors were relatively constant, 113 and (3) platelet activation triggered by thrombin receptor activating peptide. 114 Of particular interest is the variability seen in patients with factor XI deficiency. 19 Some patients with modest levels of factor XI antigen have severe bleeding symptoms, 17 whereas other patients with no detectable antigen lack significant bleeding symptoms. 18 Studies looking at the contribution of individual differences in platelet procoagulant function have at least 3 difficulties. One is that the assays are dependent on platelet number. 4,6,115 Another difficulty is that all the platelets from an individual may not have an identical function, possibly as a result of differences in platelet age. For example, some workers have seen platelets from a single individual display distinct subpopulations with respect to factor X binding 116 or factor V binding, 45 although other investigators have not seen these subpopulations. 29 And a third difficulty in studying individual differences in platelet procoagulant function is that the results are very dependent on the levels of coagulation factors. 117,118 With an awareness of these difficulties, we have investigated the ability of platelets from different individuals to support the conversion of factor X to factor Xa and of prothrombin to thrombin. 69 Using platelets isolated from 17 different individuals, we have observed differences in factor IXa/ factor VIIIa activation of factor X of up to 3-fold (Figure 5). We have also observed differences in factor Xa/factor Va activation of prothrombin of up to 5-fold (Figure 5). Platelets from the same individuals gave similar results in repeated assays even though there was wide variation between individuals. A number of factors were examined in an attempt to determine a mechanism that would account for this variability. Unexpectedly, as shown in Figure 5, the differences in factor X or prothrombin activation did not correlate with each other; ie, high levels of factor Xa generation did not correlate with high levels of thrombin generation. 69 The differences also did not correlate with age, sex, platelet factor Va levels, binding of plasma factor Va, platelet microparticle production, or phosphatidylserine levels (see Figure 3). 69 Overall, in a model system of coagulation consisting of purified platelets with a highly controlled concentrations of procoagulant proteins and inhibitors at plasma concentrations, the differences in thrombin generation between normal individuals were significant (Figure 6A). The experiments were designed such that the proteins were tightly controlled, and the platelet number was normalized so that the source of the platelets would be the only variable. The peak level of thrombin for individual 2 is almost half the peak level in individual 1 (Figure 6A). These data, representative of results on a larger population, demonstrate that thrombin generation in this system is dependent on the source of platelets. By use of this same cell-based model system, the contribution of factor XI to thrombin generation was measured by using platelets from different individuals. Platelets from some normal individuals show a large increase in thrombin generation on adding factor XI (individual 1 of Figure 6), whereas platelets from other normal individuals show very little response to factor XI (individual 2 of Figure 6). These differences reflect real differences between individuals, inasmuch as the results have been shown to be reproducible in repeated experiments over a period of years. 48 In conclusion, previous and recent studies show that the platelet surface plays a central role in the promotion and regulation of thrombin generation. This regulation extends beyond the expression of phosphatidylserine on the outer leaflet of the platelets and requires binding proteins and receptors that contribute to promoting and controlling the extent of thrombin generation. Cell-based models of coagulation provide the starting point for examining the mechanisms by which platelets can produce a burst of thrombin in a regulated fashion. References 1. Bizzozero J. Ueber einen neuen formbestandtheil des blutes und dessen rolle bei der thrombose und der blutgerinnung. Virchows Arch Pathol Anat Physiol Klin Med. 1882;90:

7 Monroe et al Platelets and Thrombin Generation Hayem G. Sur le mécanisme de l arrêt des hémorrhagies. C R Acad Sci. 1882;95: Zahn FW. Untersuchungen über thrombose: bildung der thromben. Virchows Arch Pathol Anat Physiol Klin Med. 1875;62: Eagle H. Studies on blood coagulation, IV: the nature of the clotting deficiency in hemophilia. J Gen Physiol. 1935;18: Quick AJ, Stanley-Brown M, Bancroft FW. A study of the coagulation defect of hemophilia and in jaundice. Am J Med Sci. 1935; 190: Buckwalter JA, Blythe WB, Brinkhous KM. Effect of blood platelets on prothrombin utilization of dog and human plasmas. Am J Physiol. 1949;159: Bell WN, Alton HG. A brain extract as a substitute for platelet suspensions in the thromboplastin generation test. Nature. 1954;174: Proctor RR, Rapaport SI. The partial thromboplastin time with kaolin. Am J Clin Pathol. 1961;36: Lindhout T, Govers-Riemslag JW, van de Waart P, Hemker HC, Rosing J. Factor Va-factor Xa interaction: effects of phospholipid vesicles of varying composition. Biochemistry. 1982;21: Bevers EM, Rosing J, Zwaal RF. Membrane phospholipids are the major determinant of the binding site for factor X activating- and prothrombinase complexes at the surface of human platelets. Agents Actions Suppl. 1986;20: Comfurius P, Williamson P, Smeets EF, Schlegel RA, Bevers EM, Zwaal RF. Reconstitution of phospholipid scramblase activity from human blood platelets. Biochemistry. 1996;35: Solum NO. Procoagulant expression in platelets and defects leading to clinical disorders. Arterioscler Thromb Vasc Biol. 1999;19: Bevers EM, Rosing J, Zwaal RF. Development of procoagulant binding sites on the platelet surface. Adv Exp Med Biol. 1985;192: Bevers EM, Comfurius P, Zwaal RF. The nature of the binding for prothrombinase at the platelet surface as revealed by lipolytic enzymes. Eur J Biochem. 1982;122: Davie EW, Ratnoff OD. Waterfall sequence for intrinsic blood clotting. Science. 1964;145: MacFarlane RG. An enzyme cascade in the blood clotting mechanism, and its function as a biological amplifier. Nature. 1964; 202: Leiba H, Ramot B, Many A. Heredity and coagulation studies in ten families with factor XI (plasma thromboplastin antecedent). Br J Haematol. 1965;11: Edson JR, White JG, Krivit W. The enigma of severe factor XI deficiency without haemorrhagic symptoms. Thromb Diath Haemorrh. 1967;18: Bolton-Maggs PH. Factor XI deficiency and its management. Haemophilia. 2000;6: Mertens K, Bertina RM. The contribution of Ca 2 and phospholipids to the activation of human blood-coagulation factor X by activated factor IX. Biochem J. 1984;223: Krishnaswamy S, Jones KC, Mann KG. Prothrombinase complex assembly: kinetic mechanism of enzyme assembly on phospholipid vesicles. J Biol Chem. 1988;263: Bom VJ, Bertina RM. The contributions of Ca 2, phospholipids and tissue-factor apoprotein to the activation of human blood-coagulation factor X by activated factor VII. Biochem J. 1990;265: Ravanat C, Archipoff G, Beretz A, Freund G, Cazenave JP, Freyssinet JM. Use of annexin-v to demonstrate the role of phosphatidylserine exposure in the maintenance of haemostatic balance by endothelial cells. Biochem J. 1992;282: Walsh PN, Lipscomb MS. Comparison of the coagulant activities of platelets and phospholipids. Br J Haematol. 1976;33: Miletich JP, Jackson CM, Majerus PW. Properties of the factor Xa binding site on human platelets. J Biol Chem. 1978;253: Nesheim ME, Furmaniak-Kazmierczak E, Henin C, Côtè G. On the existence of platelet receptors for factor V(a) and factor VIII(a). Thromb Haemost. 1993;70: Kjalke M, Monroe DM, Hoffman M, Oliver JA, Ezban M, Roberts HR. Active site-inactivated factors VIIa, Xa, and IXa inhibit individual steps in a cell-based model of tissue factor-initiated coagulation. Thromb Haemost. 1998;80: Hoffman M, Monroe DM. A cell based model of hemostasis. Thromb Haemost. 2001;85: Monroe DM, Roberts HR, Hoffman M. Platelet procoagulant complex assembly in a tissue factor-initiated system. Br J Haematol. 1994;88: Monroe DM, Hoffman M, Roberts HR. Transmission of a procoagulant signal from tissue factor-bearing cell to platelets. Blood Coagul Fibrinolysis. 1996;7: Butenas S, Brummel KE, Branda RF, Paradis SG, Mann KG. Mechanism of factor VIIa-dependent coagulation in hemophilia blood. Blood. 2002;99: Rauch U, Bonderman D, Bohrmann B, Badimon JJ, Himber J, Riederer MA, Nemerson Y. Transfer of tissue factor from leukocytes to platelets is mediated by CD15 and tissue factor. Blood. 2000;96: Furie B, Furie BC, Flaumenhaft R. A journey with platelet P-selectin: the molecular basis of granule secretion, signalling and cell adhesion. Thromb Haemost. 2001;86: Bouchard BA, Tracy PB. Platelets, leukocytes, and coagulation. Curr Opin Hematol. 2001;8: Nemerson Y, Repke D. Tissue factor accelerates the activation of coagulation factor VII: the role of a bifunctional coagulation cofactor. Thromb Res. 1985;40: Wildgoose P, Kisiel W. Activation of human factor VII by factors IXa and Xa on human bladder carcinoma cells. Blood. 1989;73: Kazama Y, Hamamoto T, Foster DC, Kisiel W. Hepsin, a putative membrane-associated serine protease, activates human factor VII and initiates a pathway of blood coagulation on the cell surface leading to thrombin formation. J Biol Chem. 1995;270: Monkovic DD, Tracy PB. Activation of human factor V by factor Xa and thrombin. Biochemistry. 1990;29: Allen DH, Tracy PB. Human coagulation factor V is activated to the functional cofactor by elastase and cathepsin G expressed at the monocyte surface. J Biol Chem. 1995;270: Tracy PB, Rohrbach MS, Mann KG. Functional prothrombinase complex assembly on isolated monocytes and lymphocytes. J Biol Chem. 1983;258: Warn-Cramer BJ, Maki SL, Zivelin A, Rapaport SI. Partial purification and characterization of extrinsic pathway inhibitor (the factor Xa-dependent plasma inhibitor of factor VIIa/tissue factor). Thromb Res. 1987;48: Broze GJ, Warren LA, Novotny WF, Higuchi DA, Girard JJ, Miletich JP. The lipoprotein-associated coagulation inhibitor that inhibits the factor VII-tissue factor complex also inhibits factor Xa: insight into its possible mechanism of action. Blood. 1988;71: Andrews RK, Shen Y, Gardiner EE, Berndt MC. Platelet adhesion receptors and (patho)physiological thrombus formation. Histol Histopathol. 2001;16: Baumgartner HR. Platelet interaction with collagen fibrils in flowing blood, I: reaction of human platelets with alpha chymotrypsindigested subendothelium. Thromb Haemost. 1977;37: Alberio L, Safa O, Clemetson KJ, Esmon CT, Dale GL. Surface expression and functional characterization of alpha-granule factor V in human platelets: effects of ionophore A23187, thrombin, collagen, and convulxin. Blood. 2000;95: Viskup RW, Tracy PB, Mann KG. The isolation of human platelet factor V. Blood. 1987;69: Pieters J, Lindhout T, Hemker HC. In situ-generated thrombin is the only enzyme that effectively activates factor VIII and factor V in thromboplastin-activated plasma. Blood. 1989;74: Oliver JA, Monroe DM, Roberts HR, Hoffman M. Thrombin activates factor XI on activated platelets in the absence of factor XII. Arterioscler Thromb Vasc Biol. 1999;19: Baglia FA, Walsh PN. Thrombin-mediated feedback activation of factor XI on the activated platelet surface is preferred over contact activation by factor XIIa or factor XIa. J Biol Chem. 2000;275: Scandura JM, Walsh PN. Factor X bound to the surface of activated human platelets is preferentially activated by platelet-bound factor IXa. Biochemistry. 1996;35: Franssen J, Salemink I, Willems GM, Wun TC, Hemker HC, Lindhout T. Prothrombinase is protected from inactivation by tissue factor pathway inhibitor: competition between prothrombin and inhibitor. Biochem J. 1997;323:33 37.

8 1388 Arterioscler Thromb Vasc Biol. September Rezaie AR. Prothrombin protects factor Xa in the prothrombinase complex from inhibition by the heparin-antithrombin complex. Blood. 2001;97: Wester J, Sixma JJ, Geuze JJ, Heijnen HF. Morphology of the hemostatic plug in human skin wounds: transformation of the plug. Lab Invest. 1979;41: Sixma JJ, van den Berg A. The haemostatic plug in haemophilia A: a morphological study of haemostatic plug formation in bleeding time skin wounds of patients with severe haemophilia A. Br J Haematol. 1984;58: Vander Velden P, Giles AR. A detailed morphological evaluation of the evolution of the haemostatic plug in normal, factor VII and factor VIII deficient dogs. Br J Haematol. 1988;70: Hoffman M, Monroe DM, Oliver JA, Roberts HR. Factors IXa and Xa play distinct roles in tissue factor-dependent initiation of coagulation. Blood. 1995;86: Sune A, Bette-Bobillo P, Bienvenue A, Fellmann P, Devaux PF. Selective outside-inside translocation of aminophospholipids in human platelets. Biochemistry. 1987;26: Bevers EM, Tilly RH, Senden JM, Comfurius P, Zwaal RF. Exposure of endogenous phosphatidylserine at the outer surface of stimulated platelets is reversed by restoration of aminophospholipid translocase activity. Biochemistry. 1989;28: Heemskerk JWM, Bevers EM, Lindhout T. Platelet activation and blood coagulation. Thromb Haemost. 2002;88: Weiss HJ, Vicic WJ, Lages BA, Rogers J. Isolated deficiency of platelet procoagulant activity. Am J Med. 1979;67: Rosing J, Bevers EM, Comfurius P, Hemker HC, van Dieijen G, Weiss HJ, Zwaal RF. Impaired factor X and prothrombin activation associated with decreased phospholipid exposure in platelets from a patient with a bleeding disorder. Blood. 1985;65: Brooks MB, Catalfamo JL, Brown HA, Ivanova P, Lovaglio J. A hereditary bleeding disorder of dogs caused by a lack of platelet procoagulant activity. Blood. 2002;99: Furie BC, Blumenstein M, Furie B. Metal binding sites of a gammacarboxyglutamic acid-rich fragment of bovine prothrombin. J Biol Chem. 1979;254: Koppaka V, Wang J, Banerjee M, Lentz BR. Soluble phospholipids enhance factor Xa-catalyzed prothrombin activation in solution. Biochemistry. 1996;35: Srivastava A, Quinn-Allen MA, Kim SW, Kane WH, Lentz BR. Soluble phosphatidylserine binds to a single identified site in the C2 domain of human factor Va. Biochemistry. 2001;40: Banerjee M, Majumder R, Weinreb G, Wang J, Lentz BR. Role of procoagulant lipids in human prothrombin activation, 2: soluble phosphatidylserine upregulates and directs factor X(a) to appropriate peptide bonds in prothrombin. Biochemistry. 2002;41: Srivastava A, Wang J, Majumder R, Rezaie AR, Stenflo J, Esmon CT, Lentz BR. Localization of phosphatidylserine binding sites to structural domains of factor Xa. J Biol Chem. 2002;277: Wu JR, Zhou C, Majumder R, Powers DD, Weinreb G, Lentz BR. Role of procoagulant lipids in human prothrombin activation, 1: prothrombin activation by factor X(a) in the absence of factor V(a) and in the absence and presence of membranes. Biochemistry. 2002; 41: Sumner WT, Monroe DM, Hoffman M. Variability in platelet procoagulant activity in healthy volunteers. Thromb Res. 1996;81: De Cristofaro R, De Candia E, Landolfi R, Rutella S, Hall SW. Structural and functional mapping of the thrombin domain involved in the binding to the platelet glycoprotein Ib. Biochemistry. 2001; 40: Vu TK, Hung DT, Wheaton VI, Coughlin SR. Molecular cloning of a functional thrombin receptor reveals a novel proteolytic mechanism of receptor activation. Cell. 1991;64: Andersen H, Greenberg DL, Fujikawa K, Xu W, Chung DW, Davie EW. Protease-activated receptor 1 is the primary mediator of thrombin-stimulated platelet procoagulant activity. Proc Natl Acad Sci U S A. 1999;96: De Candia E, Hall SW, Rutella S, Landolfi R, Andrews RK, De Cristofaro R. Binding of thrombin to glycoprotein Ib accelerates the hydrolysis of Par-1 on intact platelets. J Biol Chem. 2001;276: Soslau G, Class R, Morgan DA, Foster C, Lord ST, Marchese P, Ruggeri ZM. Unique pathway of thrombin-induced platelet aggregation mediated by glycoprotein Ib. J Biol Chem. 2001;276: Hayes KL, Tracy PB. The platelet high affinity binding site for thrombin mimics hirudin, modulates thrombin-induced platelet activation, and is distinct from the glycoprotein Ib-IX-V complex. J Biol Chem. 1999;274: Suzuki H, Shima M, Kamisue S, Nakai H, Nogami K, Shibata M, Morichika S, Tanaka I, Giddings JC, Yoshioka A. The role of platelet von Willebrand factor in the binding of factor VIII to activated platelets. Thromb Res. 1998;90: Hultin MB. Modulation of thrombin-mediated activation of factor VIII:C by calcium ions, phospholipid, and platelets. Blood. 1985;66: Beguin S, Kumar R, Keularts I, Seligsohn U, Coller BS, Hemker HC. Fibrin-dependent platelet procoagulant activity requires GPIb receptors and von Willebrand factor. Blood. 1999;93: Greengard JS, Heeb MJ, Ersdal E, Walsh PN, Griffin JH. Binding of coagulation factor XI to washed human platelets. Biochemistry. 1986;25: Baglia FA, Jameson BA, Walsh PN. Identification and characterization of a binding site for platelets in the Apple 3 domain of coagulation factor XI. J Biol Chem. 1995;270: Ho DH, Baglia FA, Walsh PN. Factor XI binding to activated platelets is mediated by residues R(250), K(255), F(260), and Q(263) within the apple 3 domain. Biochemistry. 2000;39: Baglia FA, Walsh PN. Prothrombin is a cofactor for the binding of factor XI to the platelet surface and for platelet-mediated factor XI activation by thrombin. Biochemistry. 1998;37: Baglia FA, Badellino KO, Li CQ, Lopez JA, Walsh PN. Factor XI binding to the platelet glycoprotein Ib-IX-V complex promotes factor XI activation by thrombin. J Biol Chem. 2002;277: Rosing J, Tans G, Govers-Riemslag JW, Zwaal RF, Hemker HC. The role of phospholipids and factor Va in the prothrombinase complex. J Biol Chem. 1980;255: Sims PJ, Wiedmer T, Esmon CT, Weiss HJ, Shattil SJ. Assembly of the platelet prothrombinase complex is linked to vesiculation of the platelet plasma membrane: studies in Scott syndrome: an isolated defect in platelet procoagulant activity. J Biol Chem. 1989;264: Tracy PB, Peterson JM, Nesheim ME, McDuffie FC, Mann KG. Interaction of coagulation factor V and factor Va with platelets. J Biol Chem. 1979;254: Miletich JP, Kane WH, Hofmann SL, Stanford N, Majerus PW. Deficiency of factor Xa-factor Va binding sites on the platelets of a patient with a bleeding disorder. Blood. 1979;54: Tracy PB, Nesheim ME, Mann KG. Platelet factor Xa receptor. Methods Enzymol. 1992;215: Nesheim ME, Taswell JB, Mann KG. The contribution of bovine factor V and factor Va to the activity of prothrombinase. J Biol Chem. 1979;254: Briedé JJ, Heemskerk JW, van t Veer C, Hemker HC, Lindhout T. Contribution of platelet-derived factor Va to thrombin generation on immobilized collagen- and fibrinogen-adherent platelets. Thromb Haemost. 2001;85: Bouchard BA, Catcher CS, Thrash BR, Adida C, Tracy PB. Effector cell protease receptor-1, a platelet activation-dependent membrane protein, regulates prothrombinase-catalyzed thrombin generation. J Biol Chem. 1997;272: Zaman GJ, Conway EM. The elusive factor Xa receptor: failure to detect transcripts that correspond to the published sequence of EPR-1. Blood. 2000;96: Ahmad SS, Rawala-Sheikh R, Walsh PN. Comparative interactions of factor IX and factor IXa with human platelets. J Biol Chem. 1989;264: Hoffman M, Monroe DM, Roberts HR. Coagulation factor IXa binding to activated platelets and platelet-derived microparticles: a flow cytometric study. Thromb Haemost. 1992;68: Jones ME, Griffith MJ, Monroe DM, Roberts HR, Lentz BR. Comparison of lipid binding and kinetic properties of normal, variant, and gamma-carboxyglutamic acid modified human factor IX and factor IXa. Biochemistry. 1985;24: Rawala-Sheikh R, Ahmad SS, Monroe DM, Roberts HR, Walsh PN. Role of gamma-carboxyglutamic acid residues in the binding of factor IXa to platelets and in factor-x activation. Blood. 1992;79: