Variation in Human Platelet Glycoprotein VI Content Modulates Glycoprotein VI–Specific Prothrombinase Activity
Glycoprotein VI (GPVI) is a platelet-specific receptor for collagen that figures prominently in signal transduction. An addition to binding to type I and III collagens, GPVI is also bound specifically by collagen-related peptide and convulxin (CVX), a snake venom protein. We developed a quantitative assay of platelet GPVI in which biotin-conjugated CVX binds selectively to GPVI in separated total platelet proteins by a ligand blot procedure. Using this approach, we have documented a 5-fold range in platelet GPVI content among 23 normal healthy subjects. In addition, we have determined that CVX-induced or collagen-related peptide–induced prothrombinase activity is directly proportional to the platelet content of GPVI. A statistically significant correlation was observed at 2 CVX concentrations: 14.7 ng/mL (R2=0.854 and P<0.001, n=11) and 22 ng/mL (R2=0.776 and P<0.001, n=12). In previous studies, we established a similar range of expression of the integrin collagen receptor α2β1 on platelets of normal subjects. Among 15 donors, there is a direct correlation between platelet α2β1 density and GPVI content (R2=0.475 and P=0.004). In view of the well-documented association of GPVI with platelet procoagulant activity, this study suggests that the variation in GPVI content is a potential risk factor that may predispose individuals to hemorrhagic or thromboembolic disorders.
The initial events in vascular thrombosis are platelet adhesion to collagen and collagen-induced platelet activation. Three receptors play pivotal roles in the initiation of platelet adhesion to collagen: the glycoprotein Ib/IX/V complex, the integrin α2β1, and glycoprotein VI (GPVI). Any variation in the expression or function of these collagen receptors may lead to excessive bleeding or thrombus formation in pathological conditions. In this regard, we first reported significant variations in function and antigen levels of integrin α2β1 on platelets from normal subjects.1 We have shown that the level of platelet α2β1 is an inherited trait2,3 that is governed by the existence of 3 α2 alleles: allele 1 (807T, 1648G) is associated with increased levels of α2β1; allele 2 (807C, 1648G) and allele 3 (807C, 1648A) are each associated with lower levels of α2β1. The rate of platelet attachment to type I collagen in whole blood under high shear rate (1500/s) is proportional to the density of α2β1 receptors on the platelet surface. Thus, the density of platelet α2β1 could have an important impact on platelet adhesion to collagen in whole blood and, therefore, on platelet function in vivo, contributing to an increased risk of thrombosis or to bleeding in relevant disease states.
This assumption was recently confirmed by us4 and others,5 who have respectively demonstrated that high α2β1 density associated with 807T is a risk factor for coronary artery disease and for stroke in younger patients. In contrast, in patients with von Willebrand’s disease type 1 and borderline to normal ristocetin cofactor activity, collagen receptor density correlates inversely with closure time in a high–shear stress system (platelet function analyzer-100).6
Recently, Polgar et al7 reported that convulxin (CVX) bound to GPIV and induced platelet activation.7 Jandrot-Perrus et al8 reported that purified CVX induced platelet aggregation at levels between 15 and 35 pmol/L, but they noted a significant heterogeneity in platelet sensitivity to CVX among normal individuals. In the present study, we quantified platelet GPVI content by a ligand blotting assay with the use of CVX and found significant variation among 23 normal subjects. Moreover, we found that platelet GPVI content correlates directly with GPVI-mediated platelet procoagulant activity. On the basis of these observations, we conclude that platelet GPVI levels could represent yet another genetic risk factor for vascular thrombosis or excessive bleeding.
All reagents were purchased from Sigma Chemical Co unless otherwise noted. Synthetic peptides, including collagen-related peptide (CRP [Gly-Pro-hydroxyproline]n), cross-linked CRP, Gly-Pro-Ala (GPA)n, and Gly-Pro-Pro (GPP)n, were generously provided by Dr Michael Barnes (Cambridge University, Cambridge, UK).9 Murine monoclonal antibody 8C12 specific for α2β1 was a gift from Dr M. Ginsberg (The Scripps Research Institute, La Jolla, Calif). AP2 is a murine monoclonal antibody specific for αIIbβ3 developed in our laboratory.
This project was approved by the Human Subjects Committee of The Scripps Research Institute, and informed consent was first obtained from each healthy donor (18 whites, 3 Hispanic Americans, and 2 Asians). None of the volunteers exhibited apparent episodes of bleeding or thrombosis. Whole blood (30 mL) was drawn and mixed with 1/10 vol of 3.2% sodium citrate or 1/6 vol of acid citrate dextrose (ACD)-A at the General Clinical Research Center of The Scripps Research Institute. Platelet rich plasma (PRP) or platelet lysates were prepared from the blood as described in each section.
Purification and Biotin Labeling of CVX
Convulxin was purified from Crotalus durissus terrificus venom (Miami Serpentarium Laboratories) by the method of Polgar et al,7 with minor modifications. Briefly, 250 mg of lyophilized venom was dissolved in 8 mL of 300 mmol/L NaCl and 100 mmol/L ammonium formate at pH 3.5 (buffer A) containing the following protease inhibitors: Pefabloc-SC (1 μg/mL, Boehringer-Mannheim), N-α-Tosyl-l-lysine chloromethyl ketone (50 μg/mL), Tosyl-l-phenylalanine chloromethyl ketone (100 μg/mL), benzamidine (10 mmol/L), leupeptin (4 μmol/L), aprotinin (1 μg/mL), EDTA (5 mmol/L), and 0.05% NaN3. Insoluble components were removed by centrifugation, and the supernatant was loaded on a column (16×1200 mm) packed with Sephacryl S-300HR (Pharmacia) and equilibrated with buffer A. Fractions (2 mL) were collected at a flow rate of 0.5 mL/min, and each was analyzed by SDS-PAGE followed by silver staining of the gels. Fractions that showed a band at 85 kDa under nonreduced conditions and 2 bands of 14 and 16 kDa under reduced conditions were further analyzed for their ability to induce platelet aggregation. Before the platelet aggregation assays, each fraction was neutralized with 1/10 vol of 1 mol/L Tris, pH 8.5. Fractions positive by SDS-PAGE and platelet-aggregating activity were pooled, concentrated in a Vacufage (Eppendorf), and then dialyzed against PBS, pH 7.4, containing 0.05% sodium azide. Purified CVX was made to 2 mmol/L EDTA and kept at 4°C.
Purified CVX in PBS, pH 7.4, at a concentration of 1 mg/mL was mixed with 1/100 vol of Sulfo-NHS-LC biotin (Molecular Probes) dissolved in dimethyl sulfoxide at a concentration of 10 mg/mL to give a final concentration of 0.1 mg/mL. After 2 hours of incubation at ambient temperature, the mixture was extensively dialyzed against PBS, pH 7.4, containing 0. 0 5% sodium azide to remove free biotin.
Semiquantification of Platelet GPVI
PRP was isolated from whole blood anticoagulated with 1/10 vol of 3.2% sodium citrate by centrifugation at 250g for 12 minutes. Prostaglandin E1 (50 ng/mL final concentration) was then added to the PRP. After a 10-minute incubation at ambient temperature, platelets were pelleted by centrifugation at 750g for 12 minutes. The platelets were washed 3 times in Ringer citrate-dextrose, pH 6.5, containing 50 ng/mL prostaglandin E1, and resuspended in 15 mmol/L Tris-HCl, pH 7.4, containing 1 mmol/L EDTA and 145 mmol/L NaCl (THEN buffer). The platelets were solubilized by addition of 1/10 vol of 10% SDS. Protein concentration was determined by the method of Markwell et al,10 and the samples were kept at 4°C.
From each platelet lysate, 40 μg of protein was separated by SDS-PAGE on a 12% slab gel under nonreduced conditions. The separated proteins were transferred to a polyvinylidine difluoride membrane at 250 mA overnight, and the membrane was then blocked by incubation for 90 minutes in a solution of 5% fat-free milk in 10 mmol/L Tris-HCl, pH 7.4, containing 145 mmol/L NaCl (TBS) plus 0.05% (vol/vol) Tween 20 (TBS-T). The membrane was then incubated in 1 μg/mL (final) biotin-CVX diluted in TBS-T for 120 minutes at ambient temperature. After 4 washings with TBS-T, the membrane was further incubated with a 1:5000 dilution of horseradish peroxidase–conjugated streptavidin (Zymed Laboratories) in TBS-T at ambient temperature for 30 minutes. After final 4 washes, GPVI was visualized by enhanced chemiluminescence (ECL, Amersham Pharmacia Biotech). From the developed image, the density and area of bands corresponding to GPVI were optically scanned and quantified by using Scion Image software (Scion Corp).
Platelet Prothrombinase Assay
Platelet prothrombinase assay was performed according to the method of Alberio et al11 with modifications. Platelets were isolated from ACD-A–anticoagulated whole blood, washed 2 times in Ringer citrate-dextrose/ACD-A (5:1), and then resuspended in HEPES buffer (10 mmol/L HEPES, pH 7.5, containing 140 mmol/L NaCl, 2 mmol/L CaCl2, and 1 mmol/L MgCl2) at a density of 5.5×107/mL. Purified CVX, cross-linked CRP, and Ca2+ ionophore A23187 were each diluted in the HEPES buffer containing 1 mg/mL BSA to concentrations of 220 and 147 ng/mL (CVX), 5 μg/mL (CRP), or 10 μmol/L (A23187). The platelet suspension (180 μL) was mixed with 20 μL of either HEPES buffer alone or 1 of the following stimulants: CVX, cross-linked CRP, or Ca2+ ionophore A23187. After a 10-minute incubation at ambient temperature without agitation, 22 μL of factor Xa (Enzyme Research Laboratories) was added to a final concentration of 1 nmol/L, and the mixture was further incubated at ambient temperature for 5 minutes to obtain maximum prothrombinase activities. To this mixture was added 22 μL of human prothrombin (14 μmol/L, Enzyme Research Laboratories), and then 20 μL aliquots were removed at 30-second intervals and mixed with 80 μL stop solution (10 mmol/L HEPES, 140 mmol/L NaCl, and 0.5% [wt/vol] BSA, pH 7.5, containing 10 mmol/L EDTA) to stop the reaction. Fifty microliters of chromogenic thrombin substrate (CBS 34.47, Diagnostica Stago) was added to a concentration of 0.4 mmol/L to measure the rate of hydrolysis of the substrate by generated thrombin on a microplate reader (Molecular Devices). Serially diluted human α-thrombin (Enzyme Research Laboratories) was used as a standard.
Platelet Aggregation Induced by CVX or Cross-Linked CRP
PRP was obtained from sodium citrate–anticoagulated whole blood, as described above, and the platelet concentration was adjusted to 3×108/mL with platelet-poor plasma. Two hundred seventy microliters of the PRP was stimulated with 30 μL of either serially diluted CVX or cross-linked CRP, and the platelet aggregation curve was recorded on an aggregometer (Monitor IV Plus, Helena Laboratories).
Platelet Adhesion Assay
Flat-bottom microtiter wells were coated at 4°C overnight with 1 of the following substrates: 10 μg/mL CVX in PBS, 10 μg/mL CRP in 10 mmol/L acetic acid, or 10 μg/mL of the negative control peptide GPA in 10 mmol/L acetic acid. The wells were blocked with 2% BSA in TBS at ambient temperature for 1 hour. Washed platelets were resuspended in TBS, pH 7.6, containing 0.5% BSA and either 2 mmol/L EDTA or MgCl2 at a platelet concentration of 108/mL and added to triplicate wells at 50 μL per well. After a 60-minute incubation at ambient temperature, the plate was gently rinsed 3 times in TBS. Adherent platelets were detected by the colorimetric method of Bellavite et al.12 Absorbance at 405 nm was measured on a plate reader (Molecular Devices).
Quantification of Platelet Surface α2β1 by Flow Cytometry
A murine monoclonal antibody specific for the α2β1 complex, 8C12, was used to quantify levels of this receptor on platelets by flow cytometry as previously described.1 AP2 (anti-αIIbβ3) was used to establish gating parameters for platelets. Briefly, bound murine antibodies 8C12 or AP2 were detected by using a 1:50 dilution of fluorescein isothiocyanate–F(ab′)2 goat anti-mouse IgG (heavy and light chains, Zymed Laboratories). Levels of bound 8C12 or AP2 were then expressed as mean fluorescence intensity after subtraction of the mean fluorescence intensity observed for nonimmune murine IgG.
Sequencing of Platelet GPVI mRNA
Platelets were obtained from 20 mL whole blood as described above and solubilized in 1 mL of denaturing solution (4 mol/L guanidinium isothiocyanate and 5 mmol/L sodium citrate, pH 7.0, containing 0.1 mol/L 2-mercaptoethanol and 0.5% [wt/vol] N-lauroylsarcosine). The homogenate was mixed with 50 μL of 2 mol/L sodium acetate, pH 4, 500 μL of water-saturated phenol, and 100 μL of 49:1 chloroform/isoamyl alcohol and then incubated for 15 minutes on ice. After centrifugation at 10 000g for 15 minutes at 4°C, the upper aqueous phase was mixed with the same volume of 2-propanol, and total platelet RNA was precipitated and then dissolved in diethyl pyrocarbonate–treated water. GPVI cDNA was amplified from the platelet RNA by reverse transcription–polymerase chain reaction with the use of the Superscript One Step RT-PCR kit (GIBCO-BRL) with primers GP6F (5′-CTC AGG ACA GGG CTG AGG AA-3′, nucleotides 4 to 23) and GP6R (5′-CCA TGA TCC CTC CCT TGG AT-3′, nucleotides 1089 to 1070). The amplified cDNA was subcloned into pGEM-T EZ vector (Promega) and sequenced.
All statistical analyses were performed on a PC by using the software package SIGMASTAT. Comparisons between groups were made by using the Student t test. A value of P<0.05 was considered statistically significant.
Specificity of CVX-Ligand Blotting
When platelets are lysed directly with SDS and analyzed by ligand blot assay, biotin-CVX binds to 2 proteins with apparent molecular masses of ≈60 and 66 kDa (Figure 1A). On the other hand, when identical platelet samples are lysed first in Triton X-100 and then SDS is added to the soluble platelet fraction before ligand blotting, biotin-CVX binds to a single protein with an apparent molecular mass of 60 kDa. These results are consistent with the findings of Clemetson et al13 and are interpreted to mean that the slower mobility band (66 kDa) represents intact GPVI, whereas the faster mobility band (60 kDa) represents a major proteolytic product. We elected to solubilize platelets in SDS to maximize the recovery of GPVI. The total content of GPVI is then recorded as the sum of the density of the 60-kDa and 66-kDa bands.
Relative Quantification of Platelet GPVI Content by Ligand Blotting
Aliquots of SDS-solubilized platelet protein from donor 1 were loaded in the wells of a slab gel to provide 50, 40, 20, 10, and 5 μg of protein per lane. GPVI was detected with biotin-labeled CVX (Figure 1A). After optical scanning, the areas of the 60-kDa and 66-kDa bands were added to determine the total GPVI content in each sample. In the case of donor 1, within the protein range of 5 to 50 μg, density correlates well with the amount of platelet protein (Figure 1B; y=19.594x−4.2383, R2=0.9922). In every subsequent quantitative ligand blotting assay, donor 1 protein was used as a reference standard, and the GPVI level of each additional platelet sample was expressed relative to that of donor 1. Platelets were obtained from the each subject on at least 2 separate occasions, and each platelet sample was assayed at least in triplicate. In Figure 2, the mean platelet GPVI content of each donor is represented, relative to the content of donor 1. The range of platelet GPVI content determined by this approach was ≈5-fold (mean 0.57, SD 0.23, and variance 79.90%).
The 23 donors could be subdivided roughly into 3 groups with respect to GPVI content: high (for donors 1 and 2, mean 0.975 and SD 0.0354), intermediate (for donors 3 to 16, mean 0.651 and SD 0.114), and low (for donors 17 to 23, mean 0.297 and SD 0.0761). Statistically significant differences were observed by using the Student t test between any of these 3 groups (P≤0.001).
Platelet Prothrombinase Assay
We measured platelet prothrombinase activity essentially according to the method of Alberio et al,11 except that factor Xa was incubated with the activated platelets for 5 minutes before the addition of prothrombin. This minor modification permits a more direct measurement of the activity of the prothrombinase complex because surface expression of factor V on platelets is complete within 2 minutes,11 and a 5-minute incubation should thus be sufficient for maximum assembly of the prothrombinase complex. In addition, the 5-minute incubation allows factor Xa to activate factor V so that the maximum activity of the prothrombinase complex is attained.14 Purified CVX did not induce prothrombinase activity in the absence of platelets, indicating that there were no contaminating venom proteins that affected prothrombinase activity in the CVX preparations.
Platelets from 4 different pairs of donors were tested: donors 1 and 23, donors 1 and 17, donors 4 and 16, and donors 2 and 19. Each assay was carried out in duplicate, and 2 of the paired donor analyses were conducted on 2 separate occasions. Figure 3 represents the results of a typical paired comparison between donors 1 and 17. Donor 1 platelets showed significantly higher prothrombinase activities induced by CVX at any time point compared with those from donor 17, at 2 different CVX concentrations (14.7 and 22 ng/mL). The initial slope of the reaction in each assay was plotted against the relative platelet GPVI content (Figure 4). The correlation between CVX-induced prothrombinase activity and GPVI content was analyzed by simple linear regression. A statistically significant correlation was observed at both CVX concentrations: 14.7 ng/mL (R2=0.854 and P<0. 001, n=11) and 22 ng/mL (R2=0.776 and P<0.001, n=12). Thus, CVX induced prothrombinase activity in a dose-dependent manner. In contrast, we did not observe statistically significant difference in ionophore A23187–induced prothrombinase activities among the donors (data not shown).
Additional Platelet Functions Related to GPVI
Platelets of several high and low donors were tested for CVX-induced aggregation, as described.7,8 Platelets from all the donors (n=4) showed a similar rate and extent of aggregation in response to CVX (10 ng/mL), without an effect of GPVI content (data not shown). Similarly, we did not observe a statistically significant difference in platelet adhesion to either CVX or CRP (n=4, data not shown).
Correlation Between Platelet GPVI Content and α2β1
As we have previously reported,1 platelet α2β1 density varies up to 10-fold among randomly selected normal individuals. Recently, a new model was proposed in which collagen binds initially to either α2β1 or GPVI, leading to subsequent binding to the other receptor and conversion of the integrin to a high-affinity state.15 We were interested in looking for any correlation between levels of these 2 platelet collagen receptors. Fifteen donors among the 23 were available for quantification of surface α2β1 and αIIbβ3 by flow cytometry. By simple linear regression analysis, a statistically significant correlation was observed between the levels of platelet α2β1 and GPVI content (R2=0.475, P=0.004; Figure 5A). In contrast, there was only a modest variance in platelet surface αIIbβ3 levels among the donors (mean 131.82, SD 16.86, and variance 25. 58%), and no correlation was found between αIIbβ3 levels and GPVI content (R2=0.030, P=0.538; Figure 5B).
Determination of GPVI cDNA Sequence
No difference was observed in the GPVI cDNA nucleotide sequence between donor 1 and donor 23, the donors with the most disparate GPVI content. Both sequences had in common 1 nucleotide difference, G936C, compared with the published sequence (numbered according to Clemetson et al13). This nucleotide substitution does not change the corresponding amino acid (both CTG and CTC encode leucine). 936C was also detected in each of the chromosome 19 clones, CTC-550B14 and RP11-700B5, of the Human Genome Project, as reported by Ezumi et al.16
Platelet GPVI is a key receptor for collagens that belongs to the killer cell receptor subgroup of the immunoglobulin gene family.13,17 As with the other members of this gene family, the optimum activity of GPVI depends on the presence of a coreceptor, Fc receptor (FcR) γ-chain. Reduced GPVI expression or function would be predicted to result in abnormal platelet adhesiveness and clinical bleeding symptoms. Indeed, a patient with autoimmune thrombocytopenia who developed antibodies specific for GPVI presented with a mild bleeding tendency.18 In addition, in genetically engineered mice that do not express FcR γ-chain, there is a reproducible impairment of collagen-induced platelet activation.19,20
CVX is a protein from the venom of C durissus terrificus that binds uniquely to GPVI. Because it is polymeric, CVX readily induces platelet activation.7 The first hint that GPVI may be variably expressed among normal individuals came from the studies of Jandrot-Perrus et al,8 who noted, but could not explain, a substantial heterogeneity in platelet sensitivity to CVX. In addition, Polgar et al7 observed that CVX was a powerful platelet agonist and, at concentrations as low as 3 to 10 ng/mL, induced maximal aggregation of isolated human platelets. Given our established interest in the influence of receptor polymorphisms on platelet responses to collagens,1 these intriguing observations prompted us to determine whether the platelet GPVI content might vary between individuals by using the only existing and reliable reagent, CVX.
We purified CVX from snake venom to >99% purity, such that the product exhibits no endogenous activity to affect prothrombinase. Using biotin-conjugated CVX, we then developed a quantitative ligand blotting assay to determine total platelet GPVI content among normal healthy individuals. By this assay, it became obvious that there was at least a 5-fold variation in platelet GPVI content among normal individuals. The 23 subjects in the present study were arbitrarily subdivided into 3 groups: high, intermediate, and low, with phenotypic frequencies of 0.09, 0.61, and 0.30, respectively. The mean GPVI content within each subgroup was significantly different from that of the other subgroups (P≤0.001). The nucleotide sequences of GPVI mRNA were identical between those donors whose platelets expressed the highest (subject 1) and lowest (subject 23) GPVI levels. This indicated that the observed difference in ligand blotting reflected quantitative and not qualitative differences in GPVI that might influence its affinity for CVX.
One might speculate that the GPVI variation is merely a reflection of the difference in platelet volume or membrane surface area between normal individuals. However, the mean platelet volume (MPV) of normal healthy individuals falls within a much more narrow range, 9.5±1.7 fL in 1 report21 and 7.5±1.5 fL in another study.22 Even when platelet size is extremely large, as in the Bernard-Soulier syndrome, the MPV that is observed may be only as high as 1.7 times the normal value (12.5 fL22). The range of MPV among normal subjects is not large enough to account for the 5-fold variation in platelet GPVI content that we observed in the present study. Moreover, whether the platelet is considered to represent a roughly spherical object in suspension or a circular disk, the mean difference in membrane surface area will remain less than the mean difference in volume. Aside from these hypothetical considerations, further evidence that MPV does not make a significant contribution to our findings comes from the fact that neither prothrombinase activity induced by the GPVI-independent agonist A23187 nor platelet surface αIIbβ3 density demonstrated a statistically significant variation among the donors in the present study. Thus, differences in platelet size or MPV cannot account for the GPVI variation that we observed.
If GPVI variation influences platelet functions, a high or a low phenotype could be a risk factor for thromboembolism or excessive hemorrhage, respectively, in certain pathological conditions. To address this question, we investigated CVX-induced or cross-linked CRP–induced platelet aggregation, platelet adhesion to CVX or CRP, and CVX-induced prothrombinase activities. Even though Jandrot-Perrus et al8 reported that the sensitivity of platelets to CVX-induced aggregation is quite variable, we did not observe a direct correlation between platelet aggregability and GPVI content under the conditions that we tested. Similarly, platelet adhesion to GPVI-specific ligands under static conditions did not show an apparent correlation with GPVI content. These platelets responses are relatively insensitive to receptor density and apparently require only a threshold stimulus to induce a maximal response. Moreover, the rate and extent of platelet aggregation is affected by many other factors in addition to the density of the primary receptor (in this case, GPVI), including the patency of several signal transduction pathways, the density of αIIbβ3, and the concentration of platelet and plasma fibrinogen and von Willebrand factor.
On the other hand, prothrombinase assay is more quantitative and much more sensitive to the level of the collagen receptor GPVI, inasmuch as it measures one of the earliest events that results from the engagement of receptor with agonist, namely, the exposure of a procoagulant phospholipid bilayer. We did observe a significant difference in CVX-induced platelet prothrombinase activity that correlated well with platelet GPVI content. After exposure to collagens, activation of the coagulation system is initiated on the platelet surface, in part, by the exposure of negatively charged phospholipids on the outer leaflet of the cell membrane and by the secretion of factor V from the platelet α-granules. It is on this negatively charged phospholipid surface that factor Va, factor Xa, and Ca2+ form a prothrombinase complex and generate thrombin. Accelerated activation of the coagulation system may lead to pathological conditions, such as stroke23 and coronary artery disease.24 On the other hand, certain factor VII genotypes associated with lower plasma factor VII levels (and thus lower prothrombinase activity) are thought to protect against myocardial infarction.25
Platelets express this procoagulant activity in response to collagens or CVX in a dose-dependent manner, as reported by Alberio et al11 and confirmed by our results in the present study. At 2 different concentrations (14.7 and 22 ng/mL), CVX induced prothrombinase activity in a dose-dependent manner. On the other hand, among the same donors, we observed no significant difference in prothrombinase activity induced by a GPVI-independent agonist, ionophore A23187.
These concentrations of CVX alone may have induced an attenuated prothrombinase activity compared with that induced by the combination of thrombin and CVX11 or by a solid-phase ligand.26 However, once prothrombinase complexes are formed on the negatively charged platelet surface in vivo, thrombin will be rapidly generated and will quickly accelerate the response to a maximum procoagulant activity. The higher the platelet GPVI content, the more rapid will be the platelet procoagulant response to collagen in vivo and the risk of pathological thromboembolism. Once platelets adhere to collagen in vivo, the GPVI-dependent procoagulant response would be further accelerated.26 Taken together, our findings argue that high GPVI content could represent another risk factor for thromboembolism, as is the case with α2β1.
It is now accepted that 2 platelet collagen receptors, α2β1 and GPVI, cooperate to effect maximal collagen-induced platelet activation. Regarding α2β1, we previously reported that there is as high as a 10-fold variation in platelet α2β1 density among normal individuals.1 It would be reasonable to expect that these 2 receptors might be proximal on the platelet surface and that their expression may be somehow coordinated. Our data, showing that there is a statistically significant correlation between platelet surface α2β1 density and GPVI content, would be consistent with this hypothesis. There is certainly a need to confirm this preliminary correlation through an analysis of a larger donor population.
IgG Fc receptor IIA (FcγRIIA) also exhibits quantitative heterogeneity on platelets27,28 and might be partially responsible for the signal transduction induced by the association of glycoprotein Ib-IX-V complex with von Willebrand factor through an immunoreceptor tyrosine-based activation motif present on its cytoplasmic domain.29 GPVI also relies on an immunoreceptor tyrosine-based activation motif on the cytoplasmic domain of associated FcR γ-chains for maximal signal transduction. Further studies involving the relationship between these platelet receptors are warranted.
This study was supported by grant HL-46979 from the National Heart, Lung, and Blood Institute awarded to T.J.K. K.J.C. was supported by a grant from the Swiss National Science Foundation (No. 31-52396.97). This is manuscript 13735-MEM from The Scripps Research Institute.
Received May 17, 2001; revision accepted July 25, 2001.
Kunicki TJ, Orchekowski R, Annis D, Honda Y. Variability of integrin α2β1activity on human platelets. Blood. 1993; 82: 2693–2703.
Kunicki TJ, Kritzik M, Annis DS, Nugent DJ. Hereditary variation in platelet integrin α2β1density is associated with two silent polymorphisms in the α2gene coding sequence. Blood. 1997; 89: 1939–1943.
Kritzik M, Savage B, Nugent DJ, Santoso S, Ruggeri ZM, Kunicki TJ. Nucleotide polymorphisms in the α2gene define multiple alleles that are associated with differences in platelet α2β1density. Blood. 1998; 92: 2382–2388.
Santoso S, Kunicki TJ, Kroll H, Haberbosch W, Gardemann A. Association of the platelet glycoprotein Ia C807T gene polymorphism with nonfatal myocardial infarction in younger patients. Blood. 1999; 93: 2449–2453.
Carlsson LE, Santoso S, Spitzer C, Kessler C, Greinacher A. The α2gene coding sequence T807/A873 of the platelet collagen receptor integrin α2β1might be a genetic risk factor for the development of stroke in younger patients. Blood. 1999; 93: 3583–3586.
Di Paola J, Federici AB, Mannucci PM, Canciani MT, Kritzik M, Kunicki TJ, Nugent D. Low platelet α2β1 levels in type I von Willebrand disease correlate with impaired platelet function in a high shear stress system. Blood. 1999; 93: 3578–3582.
Polgar J, Clemetson JM, Kehrel BE, Wiedemann M, Magnenat EM, Wells TNC, Clemetson KJ. Platelet activation and signal transduction by convulxin, a C-type lectin from Crotalus durissus terrificus (tropical rattlesnake) venom via the p62/GPVI collagen receptor. J Biol Chem. 1997; 272: 13576–13583.
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: 27035–27041.
Knight CG, Morton LF, Onley DJ, Peachey AR, Ichinohe T, Okuma M, Farndale RW, Barnes MJ. Collagen-platelet interaction: Gly-Pro-Hyp is uniquely specific for platelet Gp VI and mediates platelet activation by collagen. Cardiovasc Res. 1999; 41: 450–457.
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: 1694–1702.
Clemetson JM, Polgar J, Magnenat E, Wells TN, Clemetson KJ. The platelet collagen receptor glycoprotein VI is a member of the immunoglobulin superfamily closely related to FcαR and the natural killer receptors. J Biol Chem. 1999; 274: 29019–29024.
Monkovic DD, Tracy PB. Functional characterization of human platelet-released factor V and its activation by factor Xa and thrombin. J Biol Chem. 1990; 265: 17132–17140.
Jandrot-Perrus M, Busfield S, Lagrue AH, Xiong X, Debili N, Chickering T, Couedic JP, Goodearl A, Dussault B, Fraser C, et al. Cloning, characterization, and functional studies of human and mouse glycoprotein VI: a platelet-specific collagen receptor from the immunoglobulin superfamily. Blood. 2000; 96: 1798–1807.
Sugiyama T, Okuma M, Ushikubi F, Sensaki S, Kanaji K, Uchino H. A novel platelet aggregating factor found in a patient with defective collagen-induced platelet aggregation and autoimmune thrombocytopenia. Blood. 1987; 69: 1 712–720.
Poole A, Gibbins JM, Turner M, van Vugt MJ, van de Winkel JG, Saito T, Tybulewicz VL, Watson SP. The Fc receptor γ-chain and the tyrosine kinase Syk are essential for activation of mouse platelets by collagen. EMBO J. 1997; 16: 2333–2341.
Nieswandt B, Bergmeier W, Schulte V, Rackerbrandt K, Gessner JE, Zirngibi H. Expression and function of the mouse collagen receptor GPVI is strictly dependent on its association with the FcR γ-chain. J Biol Chem. 2000; 275: 23998–24002.
Cole JL, Marzec UM, Gunthel CJ, Karpatkin S, Worford L, Sundell IB, Lennox JL, Nichol JL, Harker LA. Ineffective platelet production in thrombocytopenic human immunodeficiency virus-infected patients. Blood. 1998; 91: 3239–3246.
Okada Y, Copeland BR, Fitridge R, Koziol JA, del Zoppo GJ. Fibrin contributes to microvascular obstructions and parenchymal changes during early focal cerebral ischemia and reperfusion. Stroke. 1994; 25: 1847–1853.
Siljander P, Farndale RW, Feijge MA, Comfurius P, Kos S, Bevers EM, Heemskerk JW. Platelet adhesion enhances the glycoprotein VI–dependent procoagulant response: involvement of p38 MAP kinase and calpain. Arterioscler Thromb Vasc Biol. 2001; 21: 618–627.
Tomiyama Y, Kunicki TJ, Zipf TF, Ford SB, Aster RH. Response of human platelets to activating monoclonal antibodies: importance of FcγRII (CD32) phenotype and level of expression. Blood. 1992; 80: 2261–2268.
Rosenfeld SI, Ryan DH, Looney RJ, Anderson CL, Abraham GN, Leddy JP. Human Fcγ receptors: stable inter-donor variation in quantitative expression on platelets correlates with functional responses. J Immunol. 1987; 138: 2869–2873.
Sullam PM, Hyun WC, Szollosi J, Dong J, Foss WM, Lopez JA. Physical proximity and functional interplay of the glycoprotein Ib-IX-V complex and the Fc receptor FcγRIIA on the platelet plasma membrane. J Biol Chem. 1998; 273: 5331–5336.