SEARCH

SEARCH BY CITATION

Keywords:

  • coagulation;
  • COAT platelets;
  • coated platelets;
  • factor VIIa;
  • hemophilia

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosure of Conflict of Interest
  9. References

Summary.  Background:  Activation of platelets with a combination of collagen and thrombin generates a subpopulation of highly procoagulant ‘coated’ platelets characterized by high surface expression of fibrinogen and other procoagulant proteins.

Objectives:  To analyze the interaction of recombinant factor VIIa (rFVIIa) with coated platelets.

Methods and results:  rFVIIa localized to the coated platelets in flow cytometry experiments, while minimal rFVIIa was found on platelets activated with adenosine diphosphate, thrombin or via glycoprotein VI individually, and essentially no rFVIIa was found on non-stimulated platelets. Removal of the γ-carboxyglutamic acid (Gla) domain of rFVIIa, and addition of EDTA, annexin V or excess prothrombin inhibited rFVIIa localization to the coated platelets, indicating that the interaction was mediated by the calcium-dependent conformation of the Gla domain and platelet exposure of negatively charged phospholipids. A reduced level of platelet fibrinogen exposure was observed at hemophilia A-like conditions in a model system of cell-based coagulation, indicating that coated platelet formation in hemophilia may be diminished. Addition of rFVIIa dose-dependently enhanced thrombin generation and partly restored platelet fibrinogen exposure.

Conclusions:  The data suggest that rFVIIa localized preferentially on platelets activated with dual agonists, thereby ensuring enhanced thrombin generation localized at the site of injury where both collagen and tissue factor are exposed, the latter ensuring the formation of thrombin necessary for coated platelet formation.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosure of Conflict of Interest
  9. References

Vascular damage and disruption of the endothelial cell layer causes exposure of extracellular matrix proteins including collagen and tissue factor (TF) to the flowing blood. TF exposure will result in generation of thrombin. Both collagen and thrombin are strong platelet agonists, and dual activation of platelets via the collagen and thrombin receptors results in the formation of so-called ‘coated’ platelets (reviewed by Dale [1]) with high levels of α-granule proteins such as factor (F) V, fibrinogen/fibrin, von Willebrand factor (VWF), fibronectin, α2-antiplasmin and thrombospondin exposed on their surfaces. The formation of coated platelets correlates with enhanced thrombin generation [2,3]. Increased levels of the plasma proteins FVIII, FIX and FX on the coated platelets correlate with increased platelet-associated FXa and thrombin activity [3]. These data indicate that the most intense hemostatic activity occurs where coated platelets are formed, i.e. at the interface between the flowing blood and the denuded endothelial layer.

Because the coated platelets provide the maximal density of the procoagulant proteins required for thrombin generation and are expected to be formed only immediately at the site of injury, we speculated that a hemostatic agent targeting the coated platelets would provide optimal localized and minimal systemic effect. Recombinant activated FVII (rFVIIa; NovoSeven®, NovoNordisk A/S, Bagsværd, Denmark [4]) is registered for the prevention and treatment of acute and surgery-related bleeding episodes in hemophilia patients with inhibitors, and has potential as a treatment option for the control of critical bleeding in other types of patients [5]. Pharmacologically relevant concentrations of rFVIIa ensure thrombin generation in the absence of a functional tenase complex at hemophilia-like conditions once the coagulation has been triggered with TF [6–8]. The enhanced thrombin generation leads to the formation of a tight fibrin network [9] and ensures activation of the thrombin activatable fibrinolysis inhibitor [10], protecting the hemostatic clot against premature fibrinolysis.

In the present study we evaluated the interaction between rFVIIa and coated platelets. We further used a model system of cell-based coagulation [8] to analyze the functional consequences of rFVIIa interaction with platelets at conditions allowing the formation of coated platelets via endogenously formed thrombin and added glycoprotein (GP) VI agonist.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosure of Conflict of Interest
  9. References

Proteins

FVIII in complex with VWF was Haemate (ZLB Behring, Copenhagen, Denmark), prothrombin fragment 1.2 was from Haematological Technologies Inc. (Essex Junction, VT, USA) and rFVIIa lacking the γ-carboxyglutamic acid (Gla) domain (rFVIIa des-Gla) was prepared as described [11]. Phe-Pro-Arg-rFVIIa (rFVIIai), fluorescein-(FITC) FVIIai and FITC-FVIIai des-Gla were prepared as described [12] using chloromethyl ketones from Bachem AG (Bubendorf, Switzerland) and Haematological Technologies Inc. All other proteins were purchased and processed as described [13]. Rabbit antihuman fibrinogen immunoglobulin G (IgG) and rabbit preimmune IgG (Dako A/S, Glostrup, Denmark) were labeled with R-phycoerythrin (RPE) using Phycolink R-Phycoerythrin conjugation kit (Prozyme, San Leandro, CA, USA) and dialyzed against 20 mm HEPES, pH 7.4, with 150 mm NaCl HEPES-buffered saline (HBS). Rabbit anti-rFVIIa IgG, goat anti-FVIII and rabbit or goat anti-TF IgG were prepared by immunization with rFVIIa, FVIII light chain [14] or the extracellular domains of TF [15] using standard procedures. Rabbit antihuman rFVIIa IgG was purified by protein A-agarose chromatography (Kem-En-Tec, Copenhagen, Denmark), labeled using Alexa Fluor 488 protein labeling kit (Molecular Probes, Eugene, OR, USA), and dialyzed against HBS. Affinity of the anti-rFVIIa IgG to various rFVIIa forms was controlled in an ELISA with immobilized anti-rFVIIa IgG. The antibody bound rFVIIa and rFVIIai with similar affinity, but showed 3- to 6-fold decreased affinity for rFVIIa des-Gla. Therefore, controls with FITC-rFVIIai des-Gla were included.

Flow cytometry of coated platelets

Peripheral blood was drawn from healthy volunteers as described [13]. This study was in accordance with the institutional guidelines and informed consent was obtained from all participants. Platelet-rich plasma was prepared by 20 min centrifugation at 130 × g at room temperature, and plasma proteins were removed by gel filtration on a Sepharose CL2B (Amersham Biosciences, Piscataway, NJ, USA) column equilibrated with 15 mm HEPES, 138 mm NaCl, 2.7 mm KCl, 1 mm MgCl2, 5 mm CaCl2, 5.5 mm dextrose and 1 mg mL−1 bovine serum albumin (BSA), pH 7.4. The platelets were counted on a flow cytometer (FACScan or FACS Canto, BD, Franklin Lakes, NJ, USA) using TruCount tubes (BD), and the platelet density adjusted to 40 000 μL−1. Aliquots of 10 μL platelets were diluted tenfold with buffer containing 3 μg mL−1 RPE-labeled antifibrinogen IgG, rFVIIa, rFVIIai, or rFVIIa des-Gla, and the platelet agonists (final concentrations noted in parentheses): adenosine diphosphate (ADP; 10 μm, Sigma-Aldrich, Brøndby, Denmark), thrombin (5 nm, Roche Diagnostics, Mannheim, Germany), convulxin (0.1 μg mL−1, Pentapharm (Basel, Switzerland) [16]), cross-linked collagen-related peptide (CRP; 10 μm, kindly provided by R. W. Farndale, Cambridge University [17]), or collagen (10 μg mL−1 Collagen Horm, Nycomed, Roskilde, Denmark) alone or in the noted combinations, and incubated 10 min at 37 °C with shaking. In some experiments, annexin V (1.5 μm, Sigma) or goat antihuman TF IgG (0.5 mg mL−1) were added 2 min prior to rFVIIa. In experiments with prothrombin or prothrombin fragment 1.2, the platelets were incubated for 2 min with agonists before addition of prothrombin or prothrombin fragment 1.2 at the noted concentration, followed by addition of rFVIIa after 2 or 5 min. For detection of TF antigen, rabbit antihuman TF was added at 10 μg mL−1. The reactions were stopped with 1.5% paraformaldehyde in HBS containing 5 mm CaCl2 (HBS/CaCl2). After 20 min, samples were washed with HBS/CaCl2 containing 1 mg mL−1 BSA, and resuspended in 50 μL of the buffer containing 10 μL PerCP-labeled anti-CD61 (BD) and 10 μg mL−1 labeled anti-rFVIIa IgG. FITC-conjugated F(ab′)2 fragments of swine antirabbit IgG (Dako) were used for detection of anti-TF IgG binding. Samples were analyzed by flow cytometry with forward- and side-scatter light channels and fluorescence channels set on log. A gate was set on PerCP-CD61 positive cells (all platelets), and cells within this gate analyzed for fibrinogen exposure and anti-rFVIIa IgG binding. Single colored activated platelets were used for compensation of overlapping fluorescence. Rainbow calibration particles (eight peaks; Sphero, BD Biosciences, San Jose, CA, USA) were used to convert fluorescence to number of FITC-rFVIIai per platelet. Kd was calculated by fitting data to an equation for one-site binding with variable Hill coefficient [inline image, where Bmax is the maximal binding and n the Hill coefficient] using graphpad prism software (GraphPad, San Diego, CA, USA) . A Hill coefficient of 1.9 ± 0.3 (n = 3) was obtained. As depletion of FITC-rFVIIai was excluded and equilibrium binding was obtained, a positive Hill coefficient indicates positive cooperativity. The mechanism behind this is not clear at the moment.

Cell-based thrombin generation assay

TF-expressing monocytes at a density of 80 000 monocytes/well were prepared as described [13]. rFVIIa at the noted concentrations was mixed with coagulation proteins and platelets (final density of 100 000 μL−1) as described [13]. FVIII was omitted from samples mimicking hemophilia A-like conditions, and the platelets used for these samples were preincubated with 10 BU mL−1 goat anti-FVIII IgG. Immediately after mixing, the platelets and a GPVI agonist (0.1 μg mL−1 convulxin or 10 μg mL−1 CRP) were added to the monocytes. Aliquots of 10 μL were withdrawn at timed intervals and analyzed for fibrinogen exposure by flow cytometry (see above) and thrombin activity as described [13]. The thrombin activity curves were fitted using a modified Gaussian equation as described [18].

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosure of Conflict of Interest
  9. References

It has previously been shown that rFVIIa binds to platelets activated with the thrombin receptor agonist peptide SFLLRN but not to non-stimulated platelets [19]. We hypothesized that rFVIIa binds coated platelets more efficiently than platelets activated only with the thrombin receptor agonist peptide. Platelets were activated with ADP, thrombin, a collagen GPVI receptor agonist (convulxin [16] or CRP [17]), or a combination of thrombin and a GPVI agonist in the presence or absence of rFVIIa (Fig. 1). Flow cytometry analysis showed that non-stimulated platelets exposed low amounts of fibrinogen, i.e. did not form coated platelets. Essentially no rFVIIa was found on these platelets (compare fluorescence in the FITC channel in Fig. 1A,B). Platelets activated with single agonist, either ADP, thrombin, convulxin, or CRP, demonstrated only a minor increase in fibrinogen exposure as compared with the non-stimulated platelets, and only minimal rFVIIa was observed on the majority of platelets activated with a single agonist. Notably, a part of the platelets stimulated with convulxin (13% in the current experiment) or with CRP (4%) appeared to interact with rFVIIa; however, as these platelets do not express high fibrinogen levels they are by definition [1,2] not coated platelets. Further work is required to characterize this platelet subpopulation. In the thrombin-stimulated samples a small fraction (2%) of the platelets demonstrated high fibrinogen expression, i.e. exhibited a phenotype as coated platelets, and did interact with rFVIIa. The presence of a small subpopulation with the coated platelet phenotype after stimulation with thrombin alone has also been observed by others [1,20]. However, it is only after dual stimulation of the platelets by thrombin and a GPVI agonist that a larger subpopulation of the platelets is converted into coated platelets (40–60% in samples with convulxin and around 13% in samples with CRP). A significant amount of rFVIIa was found on these coated platelets. Stimulation of platelets with collagen combined with thrombin also produced coated platelets, and rFVIIa interacted with this platelet subpopulation to the same extent as following stimulation with thrombin and convulxin or CRP (data not shown). Platelets stimulated with thrombin combined with convulxin were used in the following experiments aiming at characterizing the rFVIIa–coated platelet interaction.

image

Figure 1.  Recombinant factor VIIa (rFVIIa) localization to coated platelets. Platelets were left non-stimulated (control) or activated with adenosine diphosphate (ADP; 10 μm), thrombin (IIa, 5 nm), convulxin (cvx, 0.1 μg mL−1), collagen-related peptide (10 μg mL−1), or the combination of thrombin and a glycoprotein (GP) VI agonist as noted in the absence (A) or presence (B) of 100 nm rFVIIa. Fibrinogen exposure and rFVIIa on the platelets were analyzed by flow cytometry. Coated platelets are the platelet subpopulation with high expression of fibrinogen after stimulation of platelets with thrombin combined with a GPVI agonist.

Download figure to PowerPoint

rFVIIa labeled in the active site with a fluorophor (FITC-FVIIai) was used for directly evaluating the dose-response of rFVIIa interaction with the coated platelets (Fig. 2). Active site-occupancy did not interfere with rFVIIa binding to the coated platelets when anti-rFVIIa IgG was used to detect rFVIIa on platelets (Fig. 3). FITC-rFVIIai bound to the coated platelets in a dose-dependent manner in concentrations up to around 2 μm (Fig. 2). When binding of rFVIIa to non-stimulated platelets was subtracted as background, saturation was obtained and a Kd of 1.2 ± 0.2 μm (n = 3) was calculated. This is similar to the Kd for binding of prothrombin to ionomycin-activated platelets (1.4 μm in the presence of 120 nm sodium chloride [21]). The maximal binding capacity was 1.8 ± 0.2 × 105 FITC-rFVIIai molecules per coated platelet. At a pharmacological relevant rFVIIa concentration of 25 nm, the approximate Cmax after dosing 90 μg kg−1 [22], the number of FITC-rFVIIai molecules found on coated platelets was 2321 ± 717 per platelet. The difference between the FITC-rFVIIai fluorescence intensity of the coated platelets and that of the subpopulation of activated platelets with low fibrinogen exposure or the non-stimulated platelets was observed following addition of as little as 1–2 nm FITC-rFVIIai.

image

Figure 2.  Saturation binding of recombinant factor VIIa (rFVIIa) to coated platelets. Active site-labeled rFVIIa [fluorescein isothiocyanate (FITC)-FVIIai] was added to non-stimulated platelets or platelets stimulated with thrombin combined with convulxin, and FITC-rFVIIai on the platelets quantified by flow cytometry. The coated platelets were identified as the population of activated platelets with high fibrinogen exposure (open circles). FITC-rFVIIai on non-stimulated platelets is indicated by closed circles, and specific binding (the difference between FITC-rFVIIai binding to coated platelets and to non-stimulated platelets) is indicated by squares. Data are representative for three individual experiments.

Download figure to PowerPoint

image

Figure 3.  Involvement of negatively charged phospholipids and the Gla-domain of recombinant factor VIIa (rFVIIa) in localization to coated platelets. Platelets were activated with a combination of thrombin and convulxin in the presence of rFVIIa, active-site blocked rFVIIa (rFVIIai), or rFVIIa des-Gla at 100 nm. Excess EDTA was added to a sample with 100 nm rFVIIa prior to fixing the cells. Annexin V (1.5 μm), prothrombin (II, 10 μm), or antitissue factor immunoglobulin G (0.5 mg mL−1) were added to platelets 2 min prior to addition of 100 nm rFVIIa. Data are representative of at least three experiments.

Download figure to PowerPoint

Vitamin K-dependent proteins interact with negatively charged phospholipid surfaces via their Gla domains. rFVIIa des-Gla did not bind to the coated platelet (Fig. 3). Similar data were obtained with FITC-rFVIIai des-Gla (data not shown). Addition of excess EDTA abolished the rFVIIa–coated platelet interaction (Fig. 3), indicating that the calcium-dependent structure of the Gla domain was essential for the interaction. The involvement of negatively charged phospholipids was further analyzed by preincubating the stimulated platelets with annexin V or 100-fold excess (relative to rFVIIa) of prothrombin, an alternative Gla domain-containing protein, before adding rFVIIa. Both bind to negatively charged phospholipids and inhibited localization of rFVIIa to the coated platelets. As prothrombin interaction with platelets is essential for normal hemostasis, the effect of physiologically relevant prothrombin concentrations on the rFVIIa–coated platelets interaction was evaluated by adding various concentrations of prothrombin prior to rFVIIa (Fig. 4). Preincubation with prothrombin at 1 μm, close to the physiological level (1.2 μm), had a modest influence on the amount of rFVIIa found on the coated platelets (70 ± 27% anti-rFVIIa IgG bound compared with samples without prothrombin, P < 0.05, two-tailed paired Student's t-test, n = 6). Similar results were obtained with prothrombin fragment 1.2 (data not shown). It should be noted that in our experiments it was not possible to inhibit rFVIIa binding to the coated platelets completely by preincubating the platelets with 1000-fold excess prothrombin, as 16 ± 5% FITC-rFVIIai was found on the platelets at these conditions. As the plasma concentrations of other Gla domain-containing proteins are lower than that of prothrombin, physiologically relevant concentrations of these proteins are not expected to influence the interaction of rFVIIa with negatively charged phospholipids on the coated platelet.

image

Figure 4.  Inhibition of recombinant factor VIIa (rFVIIa) localization to coated platelets by an alternative Gla domain-containing protein. Platelets were activated with a combination of thrombin and convulxin, prothrombin at increasing concentrations was added 2 min before rFVIIa (100 nm), and the amount of rFVIIa on the platelets analyzed by flow cytometry. The graph shows the median fluorescence of anti-rFVIIa immunoglobulin G binding to the coated platelets (solid bars) and to the remaining activated platelets with low fibrinogen exposure (open bars). Data are mean and SD from six individual experiments.

Download figure to PowerPoint

Although TF antigen has been found on non-stimulated platelets [23], only a very low amount of functional TF has been found associated with platelets in normal blood (10.6 ± 5.3 pg mL−1, which corresponds to approximately one TF molecule per platelet [24]). In our hands, the binding of anti-TF antibodies increased slightly after stimulation of platelets with thrombin combined with convulxin (data not shown). The potential involvement of TF in the localization of rFVIIa to the coated platelets was analyzed by preincubating the platelets with an antibody to TF that is known to block binding of rFVIIa to TF-expressing cells (Fig. 3). The antibody did not prevent localization of rFVIIa to the coated platelets. Similar results were obtained with lower rFVIIa concentrations (2–25 nm rFVIIa, data not shown). This demonstrates that TF was not involved in rFVIIa localization to coated platelets formed in a pure system in vitro.

The functional implications of rFVIIa interaction with coated platelets were evaluated in a model system of cell-based coagulation [8]. In agreement with earlier data [3] the addition of a GPVI agonist (convulxin or CRP) to the model system shortened the time to maximal thrombin activity from 18–20 min to around 10 min and enhanced the maximal level of thrombin activity 2- to 3-fold (data not shown). Thrombin generation was impaired at hemophilia A-like conditions as compared with the sample reflecting normal conditions (Fig. 5). Addition of a pharmacologically relevant concentration of rFVIIa (25 nm) improved the thrombin generation, although not to the level seen in the normal sample. rFVIIa at 100 nm further increased thrombin generation; however, thrombin generation curves similar to normal conditions were obtained only after addition of higher rFVIIa concentrations (1–2 μm). In the absence of GPVI agonists a similar dose-response of rFVIIa was observed as in the presence of a GPVI agonist, the main difference being that the thrombin generation in all samples proceeded more slowly and to a lower maximal level in the absence of a GPVI agonist (data not shown). This indicates not only that the relative effect of rFVIIa is similar in the presence of coated platelets and with thrombin-stimulated platelets, but also that all processes are enhanced at conditions allowing the formation of coated platelets. Our data demonstrating an rFVIIa dose-dependent enhancement of thrombin generation in a model system triggered by TF alone is in accordance with data from Allen et al. [25].

image

Figure 5.  Recombinant factor VIIa (rFVIIa) enhanced thrombin generation at hemophilia A-like conditions. rFVIIa at 0.1 nm (circles, solid lines), 25 nm (open diamonds, dotted line), 100 nm (open triangles, solid line), or 2 μm (open squares, dotted line) was mixed with plasma concentrations of factor (F) VII, FX, FIX, FXI, FV, prothrombin, antithrombin III, tissue factor pathway inhibitor and calcium, and added to tissue factor-expressing monocytes combined with either (A) convulxin (0.1 μg mL−1) or (B) collagen-related peptide (10 μg mL−1) together with non-stimulated platelets. FVIII was included to mimic normal conditions (closed circles). The data are representative of five experiments.

Download figure to PowerPoint

The formation of coated platelets in the model system was reduced at hemophilia A-like conditions (Fig. 6). When platelets from the model system were analyzed for fibrinogen exposure, a gradual increase in fluorescence intensity (i.e. coated platelet formation) was observed over time. Notably, in the model system the majority of platelets (> 80–90%) were found in the platelet subpopulation with high fibrinogen exposure, i.e. more than observed when the agonists were added directly to the platelets (see Fig. 1). At hemophilia A-like conditions the increase in fibrinogen exposure was delayed compared with the normal sample. Addition of rFVIIa to the hemophilia A-like samples enhanced the fibrinogen exposure in a dose-dependent manner, although addition of even high rFVIIa concentration (2 μm) failed to restore the level of fibrinogen exposure detected in the samples mimicking normal coagulation. rFVIIa at a pharmacologically relevant concentration (25 nm) increased platelet fibrinogen exposure at hemophilia A-like conditions from 27 ± 6% of the control to 40 ± 14% (P < 0.05, two-tailed paired Student's t-test, samples with CRP, n = 5).

image

Figure 6.  Recombinant factor VIIa (rFVIIa) enhanced fibrinogen exposure on platelets at hemophilia A-like conditions. Samples from the experiments in Fig. 5 were analyzed for fibrinogen exposure by flow cytometry. (A) Representative time-course of fibrinogen exposure on platelets in samples reflecting normal conditions with FVIII (solid circles), and hemophilia A-like conditions (open circles), with 100 nm (open triangles) or 2 μm rFVIIa (open squares) added. (B) Median fibrinogen exposure at 10 min in samples reflecting normal conditions and hemophilia A-like conditions with the noted amounts of rFVIIa added. Data are normalized to the fibrinogen exposure in the sample reflecting normal conditions. Solid bars are from samples with convulxin and open bars are from samples with collagen-related peptide (CRP). Mean and SD from four experiments with convulxin and five experiments with CRP are shown. (C–E) Samples initiated with tissue factor-expressing monocytes alone (to ensure initial thrombin generation), convulxin alone, or the combination as indicated and stained with either antifibrinogen immunoglobulin G (IgG; line histogram) or isotype control IgG (filled histogram). (F) Anti-rFVIIa IgG binding to platelets in the presence of monocytes and convulxin at hemophilia-like conditions in the presence (line histogram) or absence (filled histogram) of 100 nm rFVIIa.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosure of Conflict of Interest
  9. References

Coated platelets have been described as a highly procoagulant subpopulation of platelets formed after dual stimulation of platelets with the two strong physiologically relevant agonists, thrombin and collagen [1,2]. rFVIIa localized preferentially to these coated platelets as compared with platelets stimulated with a single agonist. Only a minor amount of rFVIIa was found on non-stimulated platelets when pharmacologically relevant concentrations of rFVIIa were applied. The rFVIIa localization was concentration-dependent with a Kd value of 1.2 ± 0.2 μm, similar to that obtained for prothrombin binding to activated platelets (1.4 μm [21]). Negatively charged phospholipids were involved, as annexin V and prothrombin or prothrombin fragment 1.2 inhibited the localization of rFVIIa to the coated platelets. Furthermore, addition of EDTA or deletion of the Gla-domain abolished binding to the coated platelets, indicating that rFVIIa localized to the coated platelets via Gla domain-mediated interaction with negatively charged phospholipids. It has previously been demonstrated that rFVIIa bound to platelets stimulated with the thrombin receptor agonist peptide SFLLRN and that omission of calcium abolished the interaction [19]. Our data demonstrate significant enhanced retention of rFVIIa on platelets stimulated with dual agonists compared with that observed for platelets stimulated only with thrombin.

Coated platelet formation at hemophilia A-like conditions was studied in a reconstituted cell-based model system, where the coagulation reactions were initiated with TF-expressing monocytes combined with a GPVI agonist. The TF-expressing monocytes will provide endogenously formed thrombin required for generation of coated platelets. The level of fibrinogen exposure was reduced at hemophilia A-like conditions, indicating that the formation of thrombin under these conditions was not sufficient to provide adequate stimulation of the platelets. This indicates that suboptimal exposure of procoagulant proteins on coated platelets upon injury may be a part of the pathophysiology of hemophilia. Thrombin at 5 nm (in combination with a GPVI agonist) was sufficient to ensure formation of coated platelets when added directly to the platelets. However, because significantly higher thrombin concentrations were generated in the reconstituted model system, optimal formation of coated platelets may require formation of adequate thrombin concentrations within a limited time-span as described [26]. At hemophilia A-like conditions the rate of thrombin generation is decreased and consequently this critical level of thrombin may not be reached in time. Addition of rFVIIa improved thrombin generation and enhanced platelet fibrinogen exposure at hemophilia A-like conditions in a dose-dependent manner. Previously, a dose-dependent increase in thrombin generation was reported in the model system in the absence of a GPVI agonist [25], while a maximal effect of rFVIIa on thrombin generation was observed at 10 nm in a reconstituted model system using TF embedded in phospholipid vesicles [27]. Our data showing increased thrombin generation by adding rFVIIa at concentrations above 25 nm support the possibility of enhanced efficacy with increased rFVIIa dosing [28,29].

The preferential localization of rFVIIa to the coated platelets, the weak localization to platelets activated with a single agonist, either ADP, thrombin or a GPVI agonist, and essentially no interaction with non-stimulated platelets may help to explain the observed safety of rFVIIa therapy in hemophilia patients [30], i.e. rFVIIa would be most effective only at a location with exposure of both collagen and TF, the latter ensuring generation of a small amount of thrombin for activation of platelets. TF and collagen are simultaneously exposed on the subendothelium upon vascular injury. In areas of the platelet thrombus more distal to the disrupted endothelial cell layer, platelets are activated by thrombin formed via the platelet prothrombinase complex. Because of the absence of collagen, coated platelets are not expected to be found in distal parts of the growing platelet thrombus, and the platelets found here may therefore have lower potential for thrombin generation. Overall, our data suggest that the preferential localization of rFVIIa to coated platelets may ensure maximal hemostatic effect of rFVIIa at the site of injury where enhancement of thrombin formation is needed to ensure hemostasis.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosure of Conflict of Interest
  9. References

L. Odborg and C. L. Christoffersen, Novo Nordisk A/S, are acknowledged for expert technical assistance. L. P. Rojkjaer is thanked for useful discussions.

Disclosure of Conflict of Interest

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosure of Conflict of Interest
  9. References

All authors are employees of NovoNordisk, the manufacturer of rFVIIa (NovoSeven®), which is studied in this work.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosure of Conflict of Interest
  9. References
  • 1
    Dale GL. Coated-platelets: an emerging component of the procoagulant response. J Thromb Haemost 2005; 3: 218592.
  • 2
    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: 1694702.
  • 3
    Kempton CL, Hoffman M, Roberts HR, Monroe DM. Platelet heterogeneity: variation in coagulation complexes on platelet subpopulations. Arterioscler Thromb Vasc Biol 2005; 25: 8616.
  • 4
    Hedner U, Glazer S, Pingel K, Alberts KA, Blomback M, Schulman S, Johnsson H. Successful use of recombinant factor VIIa in patient with severe haemophilia A during synovectomy. Lancet 1988; 2: 1193.
  • 5
    Levi M, Peters M, Buller HR. Efficacy and safety of recombinant factor VIIa for treatment of severe bleeding: a systematic review. Crit Care Med 2005; 33: 88390.
  • 6
    Lisman T, De Groot PG. Mechanism of action of recombinant factor VIIa. J Thromb Haemost 2003; 1: 11389.
  • 7
    Butenas S, Brummel KE, Paradis SG, Mann KG. Influence of factor VIIa and phospholipids on coagulation in ‘‘acquired’’ hemophilia. Arterioscler Thromb Vasc Biol 2003; 23: 1239.
  • 8
    Hoffman M. A cell-based model of coagulation and the role of factor VIIa. Blood Rev 2003; 17: S15.
  • 9
    He S, Blomback M, Ekman GJ, Hedner U. The role of recombinant factor VIIa (FVIIa) in fibrin structure in the absence of FVIII/FIX. J Thromb Haemost 2003; 1: 12159.
  • 10
    Lisman T, Mosnier LO, Lambert T, Mauser-Bunschoten EP, Meijers JCM, Nieuwenhuis HK, De Groot PG. Inhibition of fibrinolysis by recombinant factor VIIa in plasma from patients with severe hemophilia A. Blood 2002; 99: 1759.
  • 11
    Persson E, Olsen OH, Ostergaard A, Nielsen LS. Ca2+ binding to the first epidermal growth factor-like domain of factor VIIa increases amidolytic activity and tissue factor affinity. J Biol Chem 1997; 272: 1991924.
  • 12
    Sorensen BB, Persson E, Freskgard PO, Kjalke M, Ezban M, Williams T, Rao LVM. Incorporation of an active-site inhibitor in factor VIIa alters the affinity for tissue factor. J Biol Chem 1997; 272: 118638.
  • 13
    Kjalke M, Ezban M, Monroe DM, Hoffman M, Roberts HR, Hedner U. High-dose factor VIIa increases initial thrombin generation and mediates faster platelet activation in thrombocytopenia-like conditions in a cell-based model system. Br J Haematol 2001; 114: 11420
  • 14
    Ezban M, Chapman B, Petersen J, Hansen K, Rasmussen PB, Thomsen J, Kjalke M, Kongerslev L. Characterization of recombinant human two chain factor VIII. Thromb Haemost 1993; 69 (Suppl. 1): 1088 Abstract 1946.
  • 15
    Freskgard PO, Olsen OH, Persson E. Structural-changes in factor VIIa induced by Ca2+ and tissue factor studied using circular-dichroism spectroscopy. Protein Sci 1996; 5: 153140.
  • 16
    Jandrot-Perrus M, Lagrue AH, Okuma M, Bon C. Adhesion and activation of human platelets induced by convulxin involve glycoprotein-VI and integrin α2β1. J Biol Chem 1997; 272: 2703541.
  • 17
    Morton LF, Hargreaves PG, Farndale RW, Young RD, Barnes MJ. Integrin α2β1-independent activation of platelets by simple collagen-like peptides: collagen tertiary (triple-helical) and quaternary (polymeric) structures are sufficient alone for α2β1-independent platelet reactivity. Biochem J 1995; 306: 33744.
  • 18
    Allen GA, Wolberg AS, Oliver JA, Hoffman M, Roberts HR, Monroe DM. Impact of procoagulant concentration on rate, peak and total thrombin generation in a model system. J Thromb Haemost 2004; 2: 40213.
  • 19
    Monroe DM, Hoffman M, Oliver JA, Roberts HR. Platelet activity of high-dose factor VIIa is independent of tissue factor. Br J Haematol 1997; 99: 5427.
  • 20
    Dale GL, Friese P. Bax activators potentiate coated-platelet formation. J Thromb Haemost 2006; 4: 26649
  • 21
    Bevers EM, Janssen MP, Comfurius P, Balasubramanian K, Schroit AJ, Zwaal RFA, Willems GM. Quantitative determination of the binding of β2-glycoprotein I and prothrombin to phosphatidylserine-exposing blood platelets. Biochem J 2005; 386: 2719
  • 22
    Lindley CM, Sawyer WT, Macik BG, Lusher J, Harrison JF, Bairdcox K, Birch K, Glazer S, Roberts HR. Pharmacokinetics and pharmacodynamics of recombinant factor VIIa. Clin Pharm Therapeutics 1994; 55: 63848.
  • 23
    Zillmann A, Luther T, Muller I, Kotzsch M, Spannagl M, Kauke T, Oelschlagel U, Zahler S, Engelmann B. Platelet-associated tissue factor contributes to the collagen-triggered activation of blood coagulation. Biochem Biophys Res Comm 2001; 281: 3069.
  • 24
    Bach RR. Tissue factor encryption. Arterioscler Thromb Vasc Biol 2006; 26: 45661.
  • 25
    Allen GA, Hoffman M, Roberts HR, Monroe DM. Manipulation of prothrombin concentration improves response to high-dose factor VIIa in a cell-based model of hemophilia. Br J Haematol 2006; 134: 31419.
  • 26
    Batar P, Dale GL. Simultaneous engagement of thrombin and Fc gamma RIIA receptors results in platelets expressing high levels of procoagulant proteins. J Lab Clin Med 2001; 138: 393402.
  • 27
    Butenas S, Brummel KE, Branda RF, Paradis SG, Mann KG. Mechanism of factor VIIa-dependent coagulation in hemophilia blood. Blood 2002; 99: 92330.
  • 28
    Kenet G, Lubetsky A, Luboshitz J, Martinowitz U. A new approach to treatment of bleeding episodes in young hemophilia patients: a single bolus megadose of recombinant activated factor VII (NovoSeven®). J Thromb Haemost 2003; 1: 4505.
  • 29
    Parameswaran R, Shapiro AD, Gill JC, Kessler CM. Dose effect and efficacy of rFVIIa in the treatment of haemophilia patients with inhibitors: analysis from the Hemophilia and Thrombosis Research Society Registry. Haemophilia 2005; 11: 1006.
  • 30
    Abshire T, Kenet G. Recombinant factor VIIa: review of efficacy, dosing regimens and safety in patients with congenital and acquired factor VIII or IX inhibitors. J Thromb Haemost 2004; 2: 899909.