SEARCH

SEARCH BY CITATION

Keywords:

  • collagen;
  • platelet adhesion;
  • thrombosis;
  • von Willebrand factor

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: Collagen and von Willebrand factor (VWF) are considered essential to initiate platelet deposition at sites of vascular injury, but their respective roles remain to be elucidated. Methods: We used a model of carotid artery thrombosis induced by a ferric chloride injury to compare the time to first occlusion and occlusion rate at 25 min postinjury in mice lacking the collagen receptor, glycoprotein (GP) VI, or the ligand-binding domain of the VWF receptor, GP Ibα. Results: In normal mice used as controls (n = 12), a complete obstruction of blood flow developed within 8.05 ± 0.47 min (mean ± SEM), and the occlusion rate was 100%. The results were variable in 26 GP VI−/− mice. The artery never occluded in eight mice, but the time to first occlusion in the remaining 18 (8.36 ± 0.27 min) was not different from normal (P = 0.556). Nonetheless, the occlusion rate was 42%, because in seven mice the occluded artery reopened and stayed patent at 25 min. In contrast, the artery never occluded in 12 mice lacking GP Ibα. In ex vivo perfusion experiments, GP VI−/− platelets failed to form thrombi onto collagen type I fibrils, but formed thrombi of normal size when exposed to endothelial or fibroblast extracellular matrix. Conclusions: Absence of GP Ibα function has a more profound antithrombotic effect in vivo than absence of the GP VI-dependent pathway of collagen-induced adhesion/activation. Components of the extracellular matrix may elicit a thrombogenic response in the absence of GP VI but not GP Ibα.


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

The normal function of platelets is to survey the inner lining of vessels and initiate a hemostatic response if lesions are detected that cause bleeding [1]. However, in pathological conditions, particularly in arteries with atherosclerotic plaques, platelets may contribute to the onset of an acute thrombotic occlusion [2]. In agreement with this concept, antiplatelet treatment is beneficial in patients with acute coronary syndromes [3–7], as well as in preventing recurrent stroke and thrombotic cardiovascular complications in patients with non-cardioembolic ischemic stroke or transient ischemic attacks [8,9]. Several, apparently redundant mechanisms support the initial adhesion of platelets to altered vascular surfaces and their subsequent activation and aggregation into a hemostatic plug or pathological thrombus. The extent of this thrombogenic response may depend on the nature of the lesion and be regulated by such variables as the exposure of collagens and/or the generation of α-thrombin through the tissue factor-initiated pathway. Collagens and α-thrombin, independently, are among the most potent inducers of platelet adhesion and aggregation ex vivo, but their respective and likely concurrent roles in modulating hemostasis and, particularly, thrombosis remains to be fully elucidated in vivo [2].

Platelets have two main collagen receptors, the integrin α2β1 and glycoprotein (GP) VI [10,11]. Their respective functional roles are not yet fully understood [12–14] but may depend, at least in part, on the collagen structure [15]. In fact, the response of both human [15] and mouse [16] platelets to pepsin-modified (acid soluble) collagen involves both α2β1 and GP VI, while α2β1 can be inhibited on human platelets [15] or deleted from mouse platelets [16] with no or relatively minor effects on the interaction with native fibrillar collagen. On the latter substrate, however, GP VI-deficient mouse platelets exhibit a markedly impaired response, as they can still adhere under rapid flow conditions but fail to become activated and, thus, cannot form thrombi [17]. Patients with deficiency of either receptor have a mild bleeding tendency [18,19]. In mice, ablation of the GP VI gene results in a variable alteration of the bleeding time [17], while the defect in α2−/− mice is less severe [16,20]. Experimental results have suggested that GP VI may play a critical role in initiating platelet adhesion at sites of vascular injury exposed to high shear rates [21]. However, it is not clear how such a potential function relates to that of GP Ibα and its ligand, von Willebrand factor (VWF), which is present in the subendothelial matrix and rapidly binds from plasma to exposed collagen fibrils [22]. Indeed, results from clinical [23,24] as well as experimental studies [25,26] support the concept that the GP Ibα–VWF interaction is required to initiate platelet adhesion in rapidly flowing blood. In the studies reported here, we have compared the relative contributions of GP VI and GP Ibα to the development of platelet thrombi in an injured artery with rapid blood flow, such as the mouse common carotid artery. We have chosen this thrombosis model, based on a severe lesion induced by ferric chloride, because it is rapidly and consistently occlusive in wild-type mice and reported to be mostly dependent on collagen as the thrombogenic stimulus [27]. Moreover, we have evaluated how complex extracellular matrices, as compared with isolated collagen, interact with GP VI-deficient platelets. The evidence we obtained reinforces the concept that GP Ibα is functionally critical for arterial thrombosis, and demonstrates that one or more extracellular matrix components can induce platelet aggregation independently of GP VI, possibly explaining the variable thrombogenic response induced by different vascular lesions.

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

Experimental animals

The generation of GP VI−/− mice [17]; of mice lacking the extracytoplasmic ligand-binding domain of GP Ibα [28] (designated IL4Rα/GP Ibα); and of mice expressing human instead of murine GP Ibα (designated mGP Ibα−/−/htgGP Ibα+) [26] has been reported previously. All animal care and experimental procedures complied with the Guide for the Care and Use of Laboratory Animals, US Department of Health and Human Services, and were approved by the Animal Care and Use Committee of The Scripps Research Institute.

Carotid injury and thrombosis

The carotid artery of anesthetized mice was injured with FeCl3 according to a standardized protocol [29]. In brief, the common carotid artery was dissected free from the surrounding tissue and a miniature ultrasound flow probe (0.5 VB; Transonic Systems, Ithaca, NY, USA) was positioned around the vessel. The flow probe was interfaced with a flow meter (T106; Transonic Systems) and a computer-based data acquisition program (WinDaq Lite; DATAQ Instruments, Akron, OH, USA). After measuring baseline flow in the artery, a 0.5 × 1.0-mm strip of filter paper (Whatman no. 1) soaked in 10% FeCl3 was applied to the surface of the adventitia for 3 min. After removing the strip, the area surrounding the vessel was carefully washed with warm physiologic saline solution, and carotid blood flow monitoring was resumed and continued for 25 min after the injury. The artery was considered occluded when flow was below 0 ± 0.2 mL min−1, a range corresponding to the accuracy of the system (zero offset) as specified by the manufacturer. At the end of the experiment, mice were killed by an overdose of halothane.

Mouse fibroblast and endothelial cell isolation and culture

For fibroblast preparation, skin from the back of mice younger than 12 weeks was harvested and incubated overnight at 4 °C with 0.25% trypsin and antibiotics. Dermis was isolated, cleaned of fat, washed with saline and digested 1–2 h at 37 °C with 0.25% collagenase in the presence of Ca++ and Mg++. Cells were washed and resuspended in Dulbecco's modified Eagle's medium (DMEM) (Cambrex Bio Science, Walkersville, Inc., Walkersville, MD, USA) supplemented with 10% fetal bovine serum and 50 μg mL−1 ascorbic acid. After reaching confluence, the cells were expanded twice before plating on 24 × 50-mm glass coverslips, precoated with 50 μg mL−1 purified fibronectin, at a density of 5 × 104 cells per coverslip. The coverslips were placed in four-well rectangular dishes with a lid (Nunc, Roskilde, Denmark) and used for experiments after 7–10 days in culture. Extracellular fibroblast matrix was prepared by incubating the cells on a coverslip for 3 min at 22–25 °C with a solution containing 20 mm NH4OH and 0.2% Triton X-100, and removal of cells was monitored with an optical microscope. After washing with phosphate-buffered saline (PBS; 0.04 m sodium phosphate buffer, pH 7.4, 0.15 m NaCl), the coverslips coated with extracellualr matrix were used in flow studies.

Mouse microvascular endothelial cells were prepared from lungs harvested from 10 mice, minced with fine scissors and incubated with 0.5% collagenase in a volume sufficient to allow stirring for 20–30 min at 37 °C. At this point, an equal volume of cell culture medium was added to quench the collagenase action, and the suspension of cells released from the tissue was filtered twice. The suspension was poured into a culture dish and incubated for 2 h to allow cell attachment. The medium was changed, and the incubation continued for a few days changing the medium every day. The cultures were split as needed. To isolate endothelial cells, the growing cells were incubated with acetylated low-density lipoprotein (Ac-LDL) labeled with the fluorescent probe 1,1′-dioctadecyl-3,3,3′,3′-tetramethyl-indocarbocyanine perchlorate (DiI-Ac-LDL), which binds specifically to the endothelial cell LDL scavenger receptor. Positive cells were collected by cell sorting. The resulting endothelial cells were expanded, and finally seeded on glass coverslips were they were grown to confluence. The purity of the endothelial cell monolayer was verified by specific staining with an antimouse CD31 (PECAM-1) antibody. Typically, > 95% of the cells on the coverslip were positively identified with this specific endothelial cell marker. Extracellular matrix for flow experiments was prepared by mechanical disruption of the cell layer or by chemical removal of cells as described above for fibroblasts.

Ex vivo perfusion studies

Blood (0.5 mL) was obtained from the retroorbital venous plexus of anesthetized mice through a heparinized glass capillary tube and collected into heparin at a final concentration of 40 U mL−1. The blood from several mice was pooled to perform an experiment. Mepacrine (10 μm) was added to render platelets fluorescent before perfusing the blood over immobilized fibrillar type I collagen (acid insoluble) coated on a glass coverslip [25] or extracellular matrices prepared as described above. The coverslips were assembled at the bottom of a Hele-Shaw parallel plate flow chamber with linearly variable shear rate decreasing from the inlet to the outlet [30], or a rectangular chamber with constant shear rate across the flow field. Events on the surface were visualized by epifluorescence and confocal videomicroscopy, and analyzed as previously described to obtain direct quantitative thrombus volume measurements [25,31].

Statistical analysis

We compared the time to first thrombotic occlusion after arterial injury in the different groups of animals using both median (Mann–Whitney non-parametric test) and mean values (Student's t-test). In the latter case, we used the time corresponding to the total monitoring period postinjury, 25 min, when no occlusion occurred. Occlusion rates at the end of the monitoring period were compared using Fisher's exact test. The results of ex vivo perfusion studies were evaluated using the Student's t-test, as were baseline flow rates in the carotid artery of all tested animals. All statistical tests were two-sided; P < 0.05 was considered significant.

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

Carotid artery occlusion in normal and GP VI−/− mice

We selected three endpoints to assess the thrombotic response in the FeCl3 model of carotid artery injury: (i) the time to first thrombotic occlusion was used to evaluate thrombus initiation and growth; (ii) the occurrence of embolization with resumption of flow after the first occlusion was used to evaluate thrombus stability; and (iii) the proportion of occluded vessels at the end of the 25-min postinjury monitoring period (occlusion rate) was used to assess the overall thrombogenic response. In the last case, non-occluded (patent) arteries included those that recanalized after thrombotic occlusion (Fig. 1) and those that never occluded completely. Blood flow rate in the uninjured common carotid artery was comparable in all anesthetized animals (mean ± SEM: control = 1.13 ± 0.12 mL min−1, n = 12; GP VI−/− = 1.11 ± 0.07 mL min−1, n = 26; mGP Ibα−/−/htgGP Ibα+ = 1.35 ± 0.37 mL min−1, n = 6; IL4Rα/GP Ibα = 1.15 ± 0.07 mL min−1, n = 12; P > 0.05 for all comparisons).

image

Figure 1.  Doppler tracing of blood flow in the carotid artery of wild-type and glycoprotein (GP) VI−/− mice after FeCl3 injury. After recording baseline carotid blood flow (not shown), ultrasound flow monitoring was interrupted (between −5 and 0 min) to permit application to the vessel of a 0.5 × 1.0-mm strip of filter paper soaked in 10% FeCl3 solution. The filter was removed after 3 min and the exposed area rinsed with physiologic saline solution. Arterial flow was then monitored from time 0 for the next 25 min. In the wild-type control shown here, flow begins to decrease approximately 6 min after application of FeCl3 to the artery and a complete obstruction (no flow) is established during the subsequent 3 min with no successive resumption of flow. In the GP VI−/− mouse, in contrast, flow reduction begins between 8 and 9 min after FeCl3 application to the artery and a complete obstruction develops over the next 1–2 min. However, blood flow resumes afterwards with alternating cycles of progressively less efficient partial re-occlusion (flow decrease) followed by reopening of the vessel (flow increase), indicating thrombus instability and embolization. Arterial flow is normal at the end of the observation period in the GP VI−/− mouse.

Download figure to PowerPoint

The median time to first thrombotic occlusion was 8.32 min in 12 control mice, with a range from 4.9 to 10.9 min (mean ± SEM = 8.05 ± 0.47 min). Eight wild-type mice had a persistent arterial thrombosis after the occlusion first occurred, while four exhibited transient periods of flow resumption indicating thrombus embolization. However, at the end of the monitoring period, all wild-type animals had persistent absence of flow in the injured carotid artery, thus the occlusion rate in the control group was 100% (Fig. 2).

image

Figure 2.  Bar graph representation of the time to first occlusion, embolization and persistence of occlusion at the end of the observation period in wild-type and glycoprotein (GP) VI−/− mice. See the legend to Fig. 1 for details on the carotid artery injury procedure and monitoring. The results shown were obtained in 12 wild-type and 26 GP VI−/− mice.

Download figure to PowerPoint

The median time to first thrombotic occlusion in 26 GP VI−/− mice was 8.4 min, thus not significantly different from normal (P > 0.1); the range from 6.2 to 10.8 min, after excluding eight mice that had no occlusion throughout the entire monitoring period, was also similar to normal (Fig. 2). In contrast, the mean time to occlusion considering all 26 animals, and assigning a value of 25 min (corresponding to the total monitoring period postinjury) to those in which no occlusion occurred, was 13.5 ± 1.86 min (SEM), and was significantly longer than normal (P = 0.002). GP VI deficiency was associated with instability and frequent embolization of arterial thrombi. As a result, at the end of the flow-monitoring period, the occlusion rate in GP VI−/− mice was 11/26 (42.3%), and thus significantly lower (P < 0.05) than the 100% rate seen in wild-type mice (Fig. 2). In seven out of the 26 GP VI−/− mice (26.9%), the artery occluded within the normal time frame and did not reopen, but in eight (30.8%) it never occluded over the 25-min monitoring period (Fig. 2).

Platelet thrombus formation in normal or GP VI−/− mouse blood perfused over isolated collagen type I or extracellular matrices

In agreement with previous results [15], perfusion of blood from GP VI−/− mice over collagen type I fibrils at the initial wall shear rate of 1500 s−1 resulted in the adhesion of a monolayer of platelets but essentially no aggregation. The difference in the volume of deposited platelets, as compared with that seen with normal mouse blood that formed large thrombi, was highly significant (Fig. 3). Unexpectedly, however, platelet thrombus volume was normal when the same GP VI−/− blood was perfused over extracellular matrix deposited by mouse skin fibroblasts or lung microvascular endothelial cells (Fig. 3). In control experiments, we verified that there was only minimal and transient platelet adhesion with no aggregation onto the fibronectin used for cell plating.

image

Figure 3.  Platelet thrombus formation on isolated collagen type I fibrils or extracellular matrix under flow conditions. These experiments were performed with pooled mouse blood, containing 40 U mL−1 heparin to prevent clotting and 10 μm mepacrine to render platelet fluorescent, perfused for 2 min over the indicated substrate at the initial wall shear rate of 1500 s−1. At the end of the perfusion, thrombus volume was measured over an area of 65 025 μm2 in at least three locations for each experiment. The mean ± SEM of the indicated number of experiments is shown.

Download figure to PowerPoint

In vivo and ex vivo thrombus formation in the absence of GP Ibα function

To evaluate the consequences of GP Ibα deficiency on the prothrombotic function of platelets in vivo, consideration had to be given to the complex phenotype associated with absence of this receptor subunit. It has previously been shown that disruption of the GP Ibα gene results in a mouse phenotype that closely resembles the human Bernard–Soulier syndrome [24]. This is characterized by absent membrane expression of GP Ib–IX–V complex, abnormal megakaryocytic development, giant circulating platelets and decreased platelet count (macrothrombocytopenia). It has also been shown that the abnormal phenotype caused by GP Ibα deficiency in mouse platelets can be rescued by the transgenic expression of human GP Ibα, which assembles properly with mouse GP Ibβ and GP IX to form a functioning complex. These mice are here designated mGP Ibα−/−/htgGP Ibα+. The macrothrombocytopenia in the absence of GP Ibα is caused by disruption of the linkage between GP Ibα cytoplasmic tail and platelet cytoskeleton [22]. Thus, it has been possible to correct the abnormalities of platelet morphology and count in GP Ibα-deficient mice, without recovering VWF-binding function, by transgenic expression of a fusion molecule consisting of the interleukin-4 (IL-4) receptor α chain, a linker sequence, and the carboxyl terminal portion of the human GP Ib α chain (Fig. 4) [28]. These mice, here designated IL4Rα/GP Ibα, allow evaluation of the functional consequences of GP Ibα deficiency without the confounding contribution of additional abnormalities in platelet morphology and count. Before performing in vivo experiments, we characterized the functional defect of IL4Rα/GP Ibα platelets by perfusing blood over fibrillar type I collagen and analyzing the results in the range of shear rates between 5000 and 500 s−1. In agreement with previous results obtained with functional GP Ibα blockade on human platelets [25], there was essentially no detectable reactivity on the collagen surface when the blood of IL4Rα/GP Ibα mice was perfused at the highest shear rate tested of 5000 s−1 (Fig. 4). In contrast, considerable thrombus formation occurred, in spite of the absence of functional GP Ibα, at the lower shear rates of 2000 and 500 s−1. Large thrombi formed at all shear rates tested when blood from mGP Ibα−/−/htgGP Ibα+ mice was perfused (Fig. 4). Results comparable with those observed on a pure fibrillar collagen surface were obtained when blood from these mice was perfused over fibroblast extracellular matrix (not shown).

image

Figure 4.  Functional characterization by perfusion over collagen type I fibrils of platelets from mice expressing normal human glycoprotein (GP) Ibα (mGP Ibα−/−/htgGP Ibα+) or GP Ibα lacking the functional extracytoplasmic domain (IL4Rα/GP Ibα). The scheme on top shows the GP Ib structure in IL4Rα/GP Ibα mice. The α chain is replaced by a fusion molecule consisting of the interleukin 4 (IL-4) receptor α chain, a linker sequence, and the carboxyl terminal portion of human GP Ibα with the Cys residue that forms the interchain bond with GP Ibβ, followed by the transmembrane and cytoplasmic domains. In mGP Ibα−/−/htgGP Ibα+ mice, the entire mouse GP Ibα chain is replaced with the human counterpart (not shown). The images shown are single frames from a real time recording of heparinized mouse blood, containing mepacrine to render platelets fluorescent, perfused over immobilized fibrillar type I collagen (acid insoluble) in a Hele-Shaw parallel plate chamber with linearly variable shear rate decreasing from the inlet to the outlet. Events on the surface were visualized by epifluorescence videomicroscopy. The three upper images show normal thrombus formation obtained at all shear rates tested by perfusing blood from mGP Ibα−/−/htgGP Ibα+ mice. The three lower images show the absence of platelet reactivity when blood from IL4Rα/GP Ibα mice was perfused, but only at the highest shear rate tested.

Download figure to PowerPoint

We then evaluated the occurrence of thrombosis after FeCl3-induced carotid artery injury in the two groups of mice with modified GP Ibα subunit in the GP Ib–IX–V platelet complex. In the six mice whose endogenous GP Ibα subunit was replaced by the human counterpart (mGP Ibα−/−/htgGP Ibα+), the median time to carotid artery occlusion after injury was 8.7 min (range 7.5–10.3 min) and the mean ± SEM was 8.97 ± 0.42; the occlusion rate was 100% (Fig. 5). These results were not significantly different from those observed in wild-type mice (P > 0.05 for all comparisons), demonstrating that genetic manipulations per se were not responsible for functional platelet abnormalities. In marked contrast, occlusion never occurred in the injured carotid artery of 12 mice lacking the functional extracytoplasmic domain of GP Ibα (IL4Rα/GP Ibα mice), and thus their occlusion rate was 0% (Fig. 5). The difference between mice with functionally deficient GP Ibα and all other groups tested, including GP VI-deficient mice, was highly significant (P < 0.01). Of note, in preliminary studies we also found that no occlusion developed in six GP Ibα-deficient mice with a typical Bernard–Soulier phenotype including macrothrombocytopenia (not shown).

image

Figure 5.  Bar graph representation of the time to first occlusion, embolization and persistence of occlusion at the end of the observation period in mGP Ibα−/−/htgGP Ibα+ and IL4Rα/GP Ibα mice. See the legend to Fig. 1 for details on the carotid artery injury procedure and monitoring. The results shown were obtained in six mGP Ibα−/−/htgGP Ibα+ and 12 IL4Rα/GP Ibα mice.

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

In the studies reported here, we demonstrate that absence of platelet GP VI alters the course of acute thrombosis in the carotid artery but with variable response, ranging from comparable with that seen in normal mice (complete and persistent flow blockage) to markedly abnormal (no reduction in flow). The fact that the majority of GP VI−/− mice studied (18 out of 26; 69.2%) developed arterial occlusion within a normal timeframe, regardless of subsequent evolution, is a clear indication that, unlike previously suggested [21], initial platelet recruitment at a site of injury exposed to rapidly flowing blood is not strictly dependent on GP VI. This is in agreement with the results of ex vivo perfusion experiments that demonstrated an efficient initial tethering of GP VI−/− platelets to type I collagen, even though spreading and activation were absent and no aggregation ensued [17]. Nonetheless, 57.7% of GP VI−/− mice had abnormal thrombus formation with lack of arterial occlusion at the end of the monitoring period postinjury, either because flow resumed after occlusion as a consequence of thrombus instability or the artery never occluded.

The reasons for the observed variability in the thrombotic phenotype of GP VI−/− mice are not clear at present, but a similar behavior has been reported when measuring the tail bleeding time [17], a parameter of hemostatic function, which was within the normal range in 76.9% of the mice tested but extremely prolonged in the remaining 23.1%. Recent studies have shown that treatment with hirudin, a potent α-thrombin inhibitor, changes the phenotype of FcRγ−/− platelets, which also lack GP VI expression on the membrane, by causing a more profound defect of thrombus formation than seen in comparably treated wild-type mice [32]. The activity of α-thrombin, therefore, appears to ameliorate the platelet defect caused by the functional deficiency of GP VI. Other investigators, also using FcRγ−/− mice, have shown that experimental thrombosis induced by vascular lesions rich in tissue factor and with relatively minor collagen exposure to flowing blood is less dependent on GP VI than in the case of collagen-rich lesions [27]. On the basis of these results, one could postulate that variations in the thrombotic phenotype of our GP VI-deficient mice could result from uncontrolled differences in the ferric chloride-induced carotid artery lesions, with variable α-thrombin generation and collagen exposure. While these possibilities cannot be formally excluded at present, they are not supported by our own experience with the reproducibility of this model [29,33–35], which can be surmised from the fairly tight range of occlusion times in wild-type mice. Alternatively, the effect of one or more presently unidentified gene products variably influencing the phenotype associated with GP VI deficiency should be considered. Such a possibility is supported by preliminary evidence we have obtained on the hereditary characteristics of the normal or prolonged tail bleeding time in GP VI−/− mice. Future studies will clarify this issue.

It is generally thought that collagen is the main thrombogenic determinant of platelet adhesion and aggregation onto exposed extracellular matrices [2]. This concept is apparently contradicted by the observation that platelets exhibiting a complete lack of activation on collagen fibrils ex vivo, as shown previously [17] and confirmed here (Fig. 3), can form thrombi, albeit variably, in the setting of an experimental vascular injury in vivo. In this regard, our results provide direct experimental evidence for the presence in extracellular matrices of components that can induce normal adhesion and aggregation of GP VI−/− platelets. There are several possible explanations for such a finding. Collagen receptors other than GP VI – for example α2β1 (see Introduction) or others [10] – could be involved; or one or more matrix constituents other than collagen [2] could contribute to the adhesion and activation initiated by VWF-GP Ibα [36,37], with subsequent amplification by locally released agonists such as adenosine 5′-diphosphate (ADP) [36]. Among the matrix components that could contribute to a thrombogenic response are fibronectin, laminin, fibulin-1 and proteoglycans [2,38–40]. Effects of α-thrombin are not likely to have influenced the ex vivo perfusion studies because these were conducted with blood rendered essentially unclottable by heparin, although α-thrombin certainly plays a role in platelet activation in vivo [41]. Our findings suggest that variations in thrombogenic responses may be linked to changes in structural components and/or supramolecular assembly of the extracellular matrix. In any case, it is apparent that a defect in the collagen-initiated platelet activation pathway may affect the overall stability of formed thrombi, resulting in a significantly reduced rate of arterial occlusion. This observation highlights the needed synergy of multiple mechanisms for a fully normal platelet thrombogenic response [25].

The findings presented here, obtained in a thrombosis model in vivo and on extracellular matrices ex vivo, confirm that GP Ibα is essential in the process of arterial thrombosis regardless of possible variations in the severity of experimental lesions. Such a conclusion is in agreement with previous results showing that GP Ibα binding to VWF is required for the initial platelet tethering to surfaces exposed to rapidly flowing blood [25,42]. The mechanism is thought to involve rapid immobilization of plasma VWF onto exposed collagen [15], with the contribution of collagen-associated matrix VWF in the subendothelium [43]. Ex vivo studies have previously suggested that functional deficiency of the GP Ibα–VWF adhesion pathway can be overcome by platelet interaction with other substrates, such as collagen, when shear rates are below a threshold between 500 and 1500 s−1 in human blood [25]. In the ex vivo studies presented here we found that the equivalent value in mouse blood is between 2000 and 5000 s−1, possibly because the smaller platelet size and lower viscosity as compared with human blood concur to decrease the drag imposed by hydrodynamic forces on surface-immobilized platelets. In the setting of a mouse carotid artery lesion, however, thrombus formation is critically dependent on GP Ibα function even though the normal wall shear rate is in the order of 1500 s−1. In fact, assuming an average flow rate of 1.2 mL min−1, as experimentally determined in the 56 mice evaluated in these studies, and an average lumen diameter of the mouse common carotid artery of 0.05 cm, then the mean blood flow velocity is 10.2 cm s−1 (from v = V/A, where v is the mean flow velocity, V is the volumetric flow rate in mL s−1, and A is the luminal area of the vessel in cm2), and the wall shear rate is 1632 s−1 (from γw = 8v/D, where γw is the wall shear rate, v is the mean flow velocity in cm s−1, and D is the vessel lumen diameter in cm). It should be noted that this value expresses the shear rate on the luminal surface of the unobstructed vessel. However, GP Ibα binding to VWF is required not only to initiate platelet tethering to the vascular wall but also to sustain platelet–platelet interactions at the surface of the growing thrombus [31]. Indeed, histological examination of FeCl3-induced occlusive lesions has demonstrated diffuse staining for VWF within the thrombus [35], compatible with a role of this protein in platelet cohesion. Moreover, intravital microscopy studies in VWF-deficient mice have shown that a FeCl3 injury in mesenteric arterioles, in which wall shear rates are of the same order of magnitude as in the carotid artery, produces an initial platelet accumulation at the vessel wall but without leading to obstruction of flow, owing to the persistence of a narrow channel with rapid flow that cannot be occluded [44]. Thus, thrombus development may be influenced not only by the initial wall shear rate in the vessel, but also by the increasing shear rate that develops on the surface of the growing platelet aggregate as it narrows the vessel lumen.

Animal thrombosis models [45] have inherent limitations in that experimental lesions are unlikely to reflect accurately the properties of atherosclerotic plaques that cause acute thrombotic complications in human arterial diseases. Moreover, the relationship between composition of the thrombogenic surfaces and ensuing thrombotic process is still superficially understood. Thus, uncontrolled variations in the induced lesion could cause inconsistent results. In spite of these limitations, coupling in vivo and ex vivo studies can provide essential information on the complex mechanisms responsible for the process of thrombosis. Our results support the notion that the GP Ibα–VWF interaction is essential for normal platelet adhesion and aggregation at sites of arterial injury. GP VI appears to play a role in thrombus growth and stabilization, and the relevance of its contribution may depend, at least in part, on the extent to which vascular lesions alter components of the extracellular matrix in addition to collagen. Because of this, individual antithrombotic responses to selective GP VI inhibition, possibly considered as a therapeutic target, could be inconsistent.

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

This work was supported by grants HL50545 (JW), HL75736 (DJL), HL31950, HL42846 and HL78784 (ZMR) from the National Institutes of Health.

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

The authors state that they have no conflict of interest.

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
    Savage B, Cattaneo M, Ruggeri ZM. Mechanisms of platelet aggregation. Curr Opin Hematol 2001; 8: 2706.
  • 2
    Ruggeri ZM. Platelets in atherothrombosis. Nat Med 2002; 8: 122734.
  • 3
    Bhatt DL, Topol EJ. Scientific and therapeutic advances in antiplatelet therapy. Nat Rev Drug Discov 2003; 2: 1528.
  • 4
    Mehta SR, Yusuf S. Short- and long-term oral antiplatelet therapy in acute coronary syndromes and percutaneous coronary intervention. J Am Coll Cardiol 2003; 41: 79S88S.
  • 5
    Moliterno DJ, Chan AW. Glycoprotein IIb/IIIa inhibition in early intent-to-stent treatment of acute coronary syndromes: EPISTENT, ADMIRAL, CADILLAC, and TARGET. J Am Coll Cardiol 2003; 41: 49S54S.
  • 6
    Quinn MJ, Byzova TV, Qin J, Topol EJ, Plow EF. Integrin alphaIIbbeta3 and its antagonism. Arterioscler Thromb Vasc Biol 2003; 23: 94552.
  • 7
    Aljaroudi WA, Halabi AR, Harrington RA. Platelet inhibitor therapy for patients with cardiovascular disease: looking toward the future. Curr Hematol Rep 2005; 4: 397404.
  • 8
    Sacco RL, Adams R, Albers G, Alberts MJ, Benavente O, Furie K, Goldstein LB, Gorelick P, Halperin J, Harbaugh R, Johnston SC, Katzan I, Kelly-Hayes M, Kenton EJ, Marks M, Schwamm LH, Tomsick T, American Heart Association/American Stroke Association Council on Stroke, Council on Cardiovascular Radiology and Intervention, American Academy of Neurology. Guidelines for prevention of stroke in patients with ischemic stroke or transient ischemic attack. Circulation 2006; 113: e40949.
  • 9
    Sandercock P, Gubitz G, Foley P, Counsell C. Antiplatelet therapy for acute ischaemic stroke. Ann Emerg Med 2003; 42: 4235.
  • 10
    Clemetson KJ, Clemetson JM. Platelet collagen receptors. Thromb Haemost 2001; 86: 18997.
  • 11
    Nieswandt B, Watson SP. Platelet–collagen interaction: is GPVI the central receptor? Blood 2003; 102: 44961.
  • 12
    Keely PJ, Parise LV. The α2β1 integrin is a necessary co-receptor for collagen-induced activation of syk and subsequent phosphorylation of phospholipase Cy2 in platelets. J Biol Chem 1996; 271: 2666876.
  • 13
    Nieswandt B, Brakebusch C, Bergmeier W, Schulte V, Bouvard D, Mokhtari-Nejad R, Lindhout T, Heemskerk JW, Zirngibl H, Fassler R. Glycoprotein VI but not α2β1 integrin is essential for platelet interaction with collagen. EMBO J 2001; 20: 212030.
  • 14
    Sarratt KL, Chen H, Zutter MM, Santoro SA, Hammer DA, Kahn ML. GPVI and α2β1 play independent critical roles during platelet adhesion and aggregate formation to collagen under flow. Blood 2005; 106: 126877.
  • 15
    Savage B, Ginsberg MH, Ruggeri ZM. Influence of fibrillar collagen structure on the mechanisms of platelet thrombus formation under flow. Blood 1999; 94: 270415.
  • 16
    Holtkotter O, Nieswandt B, Smyth N, Muller W, Hafner M, Schulte V, Krieg T, Eckes B. Integrin alpha 2-deficient mice develop normally, are fertile, but display partially defective platelet interaction with collagen. J Biol Chem 2002; 277: 1078994.
  • 17
    Kato K, Kanaji T, Russell S, Kunicki TJ, Furihata K, Kanaji S, Marchese P, Reininger A, Ruggeri ZM, Ware J. The contribution of glycoprotein VI to stable platelet adhesion and thrombus formation illustrated by targeted gene deletion. Blood 2003; 102: 17017.
  • 18
    Nieuwenhuis HK, Akkerman JWN, Houdijk WPM, Sixma JJ. Human blood platelets showing no response to collagen fail to express surface glycoprotein Ia. Nature 1985; 318: 4702.
  • 19
    Moroi M, Jung SM, Okuma M, Shinmyozu K. A patient with platelets deficient in glycoprotein VI that lack both collagen-induced aggregation and adhesion. J Clin Invest 1989; 84: 14405.
  • 20
    He L, Pappan LK, Grenache DG, Li Z, Tollefsen DM, Santoro SA, Zutter MM. The contributions of the α2β1 integrin to vascular thrombosis in vivo. Blood 2003; 102: 36527.
  • 21
    Massberg S, Gawaz M, Gruner S, Schulte V, Konrad I, Zohlnhofer D, Heinzmann U, Nieswandt B. A crucial role of glycoprotein VI for platelet recruitment to the injured arterial wall in vivo. J Exp Med 2003; 197: 419.
  • 22
    Ruggeri ZM. Von Willebrand factor. Curr Opin Hematol 2003; 10: 1429.
  • 23
    Ruggeri ZM. Pathogenesis and classification of von Willebrand disease. Haemostasis 1994; 24: 26575.
  • 24
    Lopez JA, Andrews RK, Afshar-Kharghan V, Berndt MC. Bernard–Soulier syndrome. Blood 1998; 91: 4397418.
  • 25
    Savage B, Almus-Jacobs F, Ruggeri ZM. Specific synergy of multiple substrate–receptor interactions in platelet thrombus formation under flow. Cell 1998; 94: 65766.
  • 26
    Ware J, Russell S, Ruggeri ZM. Generation and rescue of a murine model of platelet dysfunction: the Bernard–Soulier Syndrome. Proc Natl Acad Sci USA 2000; 97: 28038.
  • 27
    Dubois C, Panicot-Dubois L, Merrill-Skoloff G, Furie B, Furie BC. Glycoprotein VI-dependent and -independent pathways of thrombus formation in vivo. Blood 2006; 107: 39026.
  • 28
    Kanaji T, Russell S, Ware J. Amelioration of the macrothrombocytopenia associated with the murine Bernard–Soulier syndrome. Blood 2002; 100: 21027.
  • 29
    Konstantinides S, Schafer K, Koschnick S, Loskutoff DJ. Leptin-dependent platelet aggregation and arterial thrombosis suggests a mechanism for atherothrombotic disease in obesity. J Clin Invest 2001; 108: 153340.
  • 30
    Usami S, Chen HH, Zhao Y, Chien S, Skalak R. Design and construction of a linear shear stress flow chamber. Ann Biomed Eng 1993; 21: 7783.
  • 31
    Ruggeri ZM, Dent JA, Saldivar E. Contribution of distinct adhesive interactions to platelet aggregation in flowing blood. Blood 1999; 94: 1728.
  • 32
    Mangin P, Yap CL, Nonne C, Sturgeon SA, Goncalves I, Yuan Y, Schoenwaelder SM, Wright CE, Lanza F, Jackson SP. Thrombin overcomes the thrombosis defect associated with platelet GPVI/FcRγ receptor deficiency. Blood 2006; 107: 434653.
  • 33
    Konstantinides S, Schafer K, Thinnes T, Loskutoff DJ. Plasminogen activator inhibitor-1 and its cofactor vitronectin stabilize arterial thrombi after vascular injury in mice. Circulation 2001; 103: 57683.
  • 34
    Schafer K, Konstantinides S, Riedel C, Thinnes T, Muller K, Dellas C, Hasenfuss G, Loskutoff DJ. Different mechanisms of increased luminal stenosis after arterial injury in mice deficient for urokinase- or tissue-type plasminogen activator. Circulation 2002; 106: 184752.
  • 35
    Konstantinides S, Schafer K, Neels JG, Dellas C, Loskutoff DJ. Inhibition of endogenous leptin protects mice from arterial and venous thrombosis. Arterioscler Thromb Vasc Biol 2004; 24: 2196201.
  • 36
    Mazzucato M, Cozzi MR, Pradella P, Ruggeri ZM, De Marco L. Distinct roles of ADP receptors in von Willebrand factor-mediated platelet signaling and activation under high flow. Blood 2004; 104: 32217.
  • 37
    Kasirer-Friede A, Cozzi MR, Mazzucato M, De Marco L, Ruggeri ZM, Shattil SJ. Signaling through GP Ib-IX-V activates αIIbβ3 independently of other receptors. Blood 2004; 103: 340311.
  • 38
    Inoue O, Suzuki-Inoue K, McCarty OJ, Moroi M, Ruggeri ZM, Kunicki TJ, Ozaki Y, Watson SP. Laminin stimulates spreading of platelets through integrin alpha6beta1-dependent activation of GPVI. Blood 2006; 107: 140512.
  • 39
    Godyna S, Diaz-Ricart M, Argraves WS. Fibulin-1 mediates platelet adhesion via a bridge of fibrinogen. Blood 1996; 88: 256977.
  • 40
    Mazzucato M, Cozzi MR, Pradella P, Perissinotto D, Malmstrom A, Morgelin M, Spessotto P, Colombatti A, De Marco L, Perris R. Vascular PG-M/versican variants promote platelet adhesion at low shear rates and cooperate with collagens to induce aggregation. FASEB J 2002; 16: 190312.
  • 41
    Sambrano GR, Weiss EJ, Zheng Y-W, Huang W, Coughlin SR. Role of thrombin signalling in platelets in haemostasis and thrombosis. Nature 2001; 413: 748.
  • 42
    Savage B, Saldivar E, Ruggeri ZM. Initiation of platelet adhesion by arrest onto fibrinogen or translocation on von Willebrand factor. Cell 1996; 84: 28997.
  • 43
    Rand JH, Glanville RW, Wu X-X, Ross JM, Zangari M, Gordon RE, Schwartz E, Potter BJ. The significance of subendothelial von Willebrand factor. Thromb Haemost 1997; 78: 44550.
  • 44
    Ni H, Denis CV, Subbarao S, Degen JL, Sato TN, Hynes RO, Wagner DD. Persistence of platelet thrombus formation in arterioles of mice lacking both von Willebrand factor and fibrinogen. J Clin Invest 2000; 106: 38592.
  • 45
    Dorffler-Melly J, Schwarte LA, Ince C, Levi M. Mouse models of focal arterial and venous thrombosis. Basic Res Cardiol 2000; 95: 5039.