Effect of phosphopyridoxylation on thrombin interaction with platelet glycoprotein Ib.

The purpose of this study was to determine the effect of chemical modification of lysyl residues on thrombin interaction with platelet membrane proteins. Modification of lysyl residues by pyridoxal-5'-phosphate affected two different sites on thrombin and resulted in a greatly decreased binding to platelets. Using a crosslinking bifunctional reagent [bis(sulphosuccinimidyl) suberate (BS3)], we show that modified thrombin retained the ability to form high molecular mass (greater than or equal to 400 kDa) complexes with yet unidentified platelet proteins and to bind to platelet protease nexin I, but had lost the ability to bind to platelet glycoprotein Ib (GPIb). As previously reported by others, heparin protected one of the two sites from phosphopyridoxylation. In contrast modified thrombin, heparin-protected modified thrombin retained the ability to bind to GPIb, indicating that the lysyl residue(s) protected by heparin from the modification are essential for GPIb binding. While unprotected modified thrombin failed to bind hirudin, heparin-protected modified thrombin retained its ability to bind the carboxy-terminal hirudin peptide H54-65. Tritium-labelling of the modified lysyl residues and degradation of modified thrombins by CNBr or trypsin confirmed that the lysyl residue(s) protected by heparin and essential for GPIb binding are located in the thrombin binding domain for the carboxyl-terminal tail of hirudin, within the sequence 18-73 of the thrombin B chain.     

Thrombin interaction with platelet glycoprotein Ib: effect of glycocalicin on thrombin specificity.

We describe here the alteration of thrombin specificity induced by its interaction with glycocalicin. Glycocalicin is the external part of platelet glycoprotein Ib alpha (GPIb alpha) and contains binding sites for von Willebrand factor and thrombin. Taking advantage of its solubility, we have used glycocalicin in competition assays on various thrombin activities. Glycocalicin did not inhibit chromogenic substrate hydrolysis nor diisopropylfluorophosphate iPr2 (PF) incorporation, indicating that thrombin binding to GPIb does not alter access to or the conformation of the thrombin catalytic site. Glycocalicin competitively inhibited thrombin binding to fibrin (Ki = 0.1 mumol/L) and blocked fibrinogen clotting activity of thrombin. Glycocalicin also inhibited thrombin binding to thrombomodulin in a competitive manner (Ki = 3 to 5 mumol/L), but failed to prevent thrombin interaction with protein C in the absence of thrombomodulin. Previous results have indicated that GPIb binds to thrombin within the anion binding exosite masked by the carboxy-terminal hirudin peptide 54-65. The present results confirm the implication of the anion binding exosite in GPIb recognition, and further indicate that the thrombin binding site for GPIb overlaps with the thrombin binding sites for fibrin and thrombomodulin, whereas it is distinct from the thrombin binding site for protein C. Some of the structural requirements for thrombin binding to GPIb appear to be very similar to those reported for binding to its platelet receptor. However, thrombin-GPIb interaction does not appear to compete with receptor hydrolysis but rather increases the sensitivity and the rate of platelet responses elicited by the receptor.     

Clinical and functional consequences of platelet membrane glycoprotein polymorphisms

The last several years have seen an abundance of studies of genetic risk factors for vascular disease, and platelet glycoprotein (GP) polymorphisms have been a primary focus of this area of research. This article reviews GP receptor polymorphisms, particularly those on GPIa-IIa (integrin alpha2beta1), GPIb-IX-V, GPIIb-IIIa (integrin alpha(IIb)beta3), and GPVI, and summarizes clinical and functional studies that have attempted to clarify their roles in human disease. Our focus is on recent work relevant to thrombotic and hemostatic processes and advances in pharmacogenetics. We consider issues affecting our ability to derive firm conclusions from these studies, and discuss future directions in this rapidly evolving area.     

Genetic polymorphism of platelet glycoprotein Ib

Platelet glycoprotein (GP) Ib from 131 healthy Japanese was analyzed using SDS-polyacrylamide gel electrophoresis and specific staining with peroxidase-coupled wheat germ agglutinin after it was transferred to nitrocellulose membranes. Four slightly different species of GPIb were observed and designated as A, B, C, and D for glycoproteins with molecular weights of 168,000, 162,000, 159,000, and 153,000 daltons, respectively. The respective gene frequencies were calculated to be .073, .011, .561, and .355 for A-, B-, C-, D-type GPIb. Portions from each type of GPIb molecule (alpha-chain and glycocalicin) showed heterogeneity with the same molecular weight difference, indicating that the variance would be derived from the polypeptide portion that is exposed to the outer medium. The different types of GPIb were the same with respect to their accessibility to lactoperoxidase, reactivity to lectins, and affinity to TLCK-thrombin. Although Bolin et al reported patients with a bleeding tendency whose platelets have double GPIb bands, here we found that platelets with different GPIb phenotypes showed no significant differences in aggregating activity and platelet retention. Analysis of GPIb phenotype should be important for structural and physiologic studies on GPIb and glycocalicin.     

Glycoprotein Ib and glycoprotein IX are fully complexed in the intact platelet membrane

Two new murine monoclonal antibodies, AK 1 and SZ 1, reactive with the human platelet glycoprotein (GP) Ib-IX complex have been produced by the hybridoma technique. Both AK 1 and SZ 1 immunoprecipitated the GP Ib-IX complex from Triton X-100-solubilized, periodate-labeled platelets. With trypsinized, labeled platelets, AK 1, SZ 1, and FMC 25 (epitope on GP IX) immunoprecipitated a membrane-bound proteolytic fragment of the GP Ib-IX complex consisting of GP IX and an congruent to 25,000 mol wt remnant of the alpha-chain of GP lb disulfide-linked to the beta-subunit. Unexpectedly, although AK 1 and SZ 1 immunoprecipitated purified GP Ib-IX complex, neither antibody immunoprecipitated the individual components of this complex, GP Ib or GP IX. When GP Ib and GP IX were recombined, however, AK 1 and SZ 1 again immunoprecipitated the reformed complex, strongly suggesting that both antibodies were recognizing an epitope present only on the intact complex. Cross-blocking studies indicated that AK 1 and SZ 1 recognized a very similar or identical epitope that was proximal to the epitope for FMC 25. Both AK 1 and SZ 1 bound to a similar number of binding sites (congruent to 25,000) on intact platelets as monoclonal antibodies directed against either GP lb or GP IX. The combined data suggests that GP lb and GP IX are fully complexed in the intact platelet membrane.     

Plasmin-induced redistribution of platelet glycoprotein Ib

Platelet membrane glycoprotein Ib (GPIb), a receptor for von Willebrand factor and thrombin, is present on the platelet surface membrane, in intraplatelet stores, and in plasma (as the proteolytic fragment glycocalicin). We examined the hypothesis that after plasmin-mediated cleavage of platelet surface GPIb, platelets can replenish their surface GPIb pool. Incubation of washed platelets with plasmin (1 hour, 22 degrees C) resulted in loss of platelet surface GPIb, but further incubation (3 hours, 37 degrees C) in autologous plasma resulted in restoration of platelet surface GPIb, as determined by ristocetin- induced platelet agglutination and a flow cytometric assay of platelet binding of three GPIb-specific monoclonal antibodies. Despite the restoration of platelet surface GPIb after the 3-hour incubation of plasmin-treated platelets in autologous plasma, the whole platelet GPIb content (measured by enzyme-linked immunosorbent assay [ELISA], sodium dodecyl sulfate-polyacrylamide gel electrophoresis, and flow cytometry) remained reduced, quantitatively corresponding to an increase in plasma glycocalicin concentration (measured by ELISA). The loss and restoration of platelet surface GPIb occurred on all platelets and, as evidenced by lack of inhibition by prostaglandin E1, EDTA, and cytochalasins, was not mediated by cyclic AMP, extracellular Ca2+, or the platelet microfilament system. In summary, this study shows that after plasmin-mediated cleavage of platelet surface GPIb, platelets can replenish their surface GPIb pool by recruitment of GPIb molecules from the intraplatelet pool (or from a sequestered surface site).     

von Willebrand factor bound to glycoprotein Ib is cleared from the platelet surface after platelet activation by thrombin.

We recently reported that after activation of human platelets by thrombin, glycoprotein (GP) Ib-IX complexes are translocated to the surface-connected canalicular system (SCCS) (Blood 76:1503, 1990). As GPIb is a major receptor for von Willebrand factor (vWF) in platelet adhesion, we have now examined the consequences of thrombin activation on the organization of vWF bound to GPIb on the platelet surface. Studies were performed using monoclonal or polyclonal antibodies in either immunogold staining and electron microscopy (Au-EM) or in flow cytometry. When unstirred platelet-rich plasma was incubated with ristocetin, bound vWF was located by Au-EM as discrete masses regularly distributed over the cell surface. Platelets from a patient with Glanzmann's thrombasthenia, lacking GPIIb-IIIa complexes, gave a similar pattern, confirming that this represented binding to GPIb. That ristocetin was not precipitating vWF before their binding to the platelets was shown by the detection of similar masses on the surface of platelets of a patient with type IIB von Willebrand disease. Experiments were continued using washed normal platelets incubated in Tyrode-EDTA, the purpose of the EDTA being to limit the surface expression of endogenous vWF after platelet stimulation. Under these conditions, platelets were treated with ristocetin for 5 minutes at 37 degrees C in the presence of increasing amounts of purified vWF. This was followed by incubation with thrombin (0.5 U/mL) for periods of up to 10 minutes. Flow cytometry showed a time-dependent loss in the surface expression of vWF bound to GPIb and these changes were confirmed by Au-EM. In particular, immunogold staining performed on ultrathin sections showed that the bulk of the vWF was being cleared to internal membrane systems. Surface clearance of vWF during thrombin-induced platelet activation is a potential mechanism for regulating platelet adhesivity.     

Reduction of platelet glycoprotein Ib in uraemia

Patients with uraemia have abnormal platelet function that may be partially corrected by haemodialysis, cryoprecipitate or 1-desamino-8-d-arginine vasopressin (DDAVP). We studied the platelet von Willebrand factor receptor. glycoprotein Ib (GPIb). and plasma von Willebrand factor (vWF) in uraemic patients undergoing chronic haemodialysis. Using the slope of agglutination of formalin-fixed platelets as an index of response to ristocetin (with a constant amount of normal plasma as a source of vWF), we found the response of platelets from uraemic patients, both before (2·7± 1·5. n = 40) and after dialysis (1·2 ± 1·2, n = 40) to be significantly less than that for normal controls (14·1 ± 10·2. n = 20: P < 0·001). In addition, the agglutination response of platelets obtained after dialysis was less than that of platelets obtained before dialysis ( P < 0·001). Immunoblotting demonstrated decreased or absent staining of glycocalicin, a subunit of GPIb, in platelet lysates from 25 patients. All platelet samples with reduced glycocalicin also had decreased responses to ristocetin. Tritium-labelled platelets from seven patients showed decreased labelling of a protein with an electrophoretic mobility equivalent to that of GPIb (140000 daltons). In addition, platelets with the lowest levels of surface GPIb, as demonstrated by flow cytometry. also had decreased ristocetin agglutination and decreased staining on immunoblot. Levels of von Willebrand factor antigen and ristocetin cofactor in plasma from 10 patients were generally within the normal range, although post-dialysis levels tended to be higher than pre-dialysis levels. The pre- and post-dialysis plasma vWF multimeric patterns were normal.     

Cultured human umbilical vein endothelial cells contain a membrane glycoprotein immunologically related to platelet glycoprotein Ib

Using a platelet glycoprotein Ib (GpIb)-specific monoclonal antibody, AP-1, we have studied cultured human umbilical vein endothelial cells (HUVEC) for the presence of GpIb. Radiolabeled AP-1 bound specifically and saturably to HUVEC in suspension and detected a single class of binding sites (100,000/cell). When Triton X-100 extracts of HUVEC were chromatographed on wheat germ agglutinin (WGA)-Sepharose, radioiodinated, precipitated with AP-1, and subjected to reduced sodium dodecyl sulfate-polyacrylamide electrophoresis (SDS-PAGE), major radioactive bands of 228,000, 145,000, and 130,000 were seen. The latter two bands correspond to the 156,000 and 140,000 bands, representing GpIb alpha and glycocalicin, respectively, which are seen when platelets are subjected to the same procedure. The 228,000 band corresponds to a band previously noted in immunoprecipitates of platelet GpIb but not fully explained. When HUVEC were grown in the presence of 35S-methionine, extracted with Triton X-100, chromatographed on WGA-Sepharose, immunoprecipitated with AP-1, and subjected to reduced SDS-PAGE, radioactive bands of 210,000, 156,000, and 90,000 were seen. We conclude that cultured HUVEC synthesize and express on their surface a glycoprotein immunologically related to platelet GpIb.     

Biochemical and functional consequences of dissociation of the platelet membrane glycoprotein IIb-IIIa complex

The platelet membrane glycoproteins, IIb and IIIa, form a Ca2+- dependent heterodimer complex that functions as the fibrinogen receptor in activated platelets to mediate platelet aggregation. Little is known about factors that affect the IIb-IIIa complex within the platelet membrane. It has been observed that platelets incubated with ethylene glycol tetra-acetic acid (EGTA) at 37 degrees C are unable to aggregate or to bind monoclonal antibodies specific for the IIb-IIIa complex. To determine whether this is due to a dissociation of IIb from IIIa, we developed a method for quantitating the complex on nondenaturing, polyacrylamide gradient gels. Platelets were surface-labeled with 125I and then solubilized and electrophoresed in 0.2% Triton and 10 mmol/L CHAPS. Under these conditions and in the presence of 1 mmol/L Ca2+, glycoproteins IIb and IIIa migrated on the gels as a discrete band at Rf = 0.33. Protein that was eluted from this band bound to an immunoaffinity column specific for the IIb-IIIa complex. In contrast, when the IIb-IIIa complex was solubilized and then dissociated with EGTA, the discrete band at Rf = 0.33 was no longer present, and IIb and IIIa were now found in a broad band at Rf = 0.45 to 0.50. To study IIb and IIIa within the surface membrane, the 125I-labeled platelets were first incubated with 0.5 mmol/L EGTA (1 nmol/L free Ca2+) at 22 degrees C and then solubilized in the absence of EGTA. The IIb and IIIa from these platelets migrated at Rf = 0.33, indicating the presence of the intact IIb-IIIa complex. In contrast, when the platelets were incubated at 37 degrees C for one hour with the EGTA, the discrete band at Rf = 0.33 representing the IIb-IIIa complex gradually disappeared. This phenomenon could not be reversed by adding Ca2+ back to the platelets before solubilization and electrophoresis. This loss of the IIb-IIIa complex from intact platelets was accompanied by (a) a progressive and irreversible decrease in adenosine diphosphate (ADP)-induced platelet aggregation and (b) decreased binding of a complex-dependent monoclonal antibody to the platelets. These studies demonstrate that when platelets are exposed to low Ca2+ at 37 degrees C, the IIb-IIIa heterodimer complexes in their surface membranes are irreversibly disrupted. Because intact IIb-IIIa complexes are required for platelet aggregation, the loss of these complexes may account for the failure of these platelets to aggregate in response to ADP.     

Amino acid sequence of the von Willebrand factor-binding domain of platelet membrane glycoprotein Ib.

We report the amino acid sequence of a 299-residue segment from the alpha chain of the human platelet membrane glycoprotein Ib. This includes the complete sequence of the amino-terminal tryptic fragment of 290 residues comprising the von Willebrand factor-binding domain. Two primary sets of overlapping fragments were obtained by cleavage of the S-carboxymethylated protein at methionyl and lysyl bonds following treatment with cyanogen bromide and Achromobacter protease I, respectively. Additional fragments were obtained by treatment of native glycocalicin with trypsin, Staphylococcus aureus V8 protease, and Serratia marcescens protease. Analysis of all these fragments provided data that allowed determination of the continuous sequence corresponding to approximately half of the alpha-chain polypeptide. This region of glycoprotein Ib is largely hydrophobic and contains only two N-linked and one O-linked carbohydrate chains. A hydrophilic region exists between residues 215 and 299, which contains a cluster of 10 negatively charged residues at 269-287. This area is likely to attract positively charged molecules. The hydrophilic, highly glycosylated (at serine and threonine residues) region corresponding to the previously described "macroglycopeptide" and representing the carboxyl-terminal half of the alpha chain is likely to begin at residue 292. The determined sequence of the alpha chain of glycoprotein Ib contains a region (residues 29-193) with seven repeats, which is indicative of gene duplication and is highly homologous to human leucine-rich alpha 2-glycoprotein. This protein sequence agrees completely with that deduced from the cDNA sequence reported by Lopez et al. [Lopez, J.A., Chung, D.W., Fujikawa, K., Hagen, F.S., Papayannopoulou, T. & Roth, G.J. (1987) Proc. Natl. Acad. Sci. USA 84, 5615-5619].     

Differences between platelet phosphoinositide metabolism stimulated by thrombin or SFLLRN are not accounted for by interaction of thrombin with glycoprotein Ib

The formation of inositol phosphates was compared in aspirin-treated, washed human platelets suspended in Tyrode's-albumin solution containing 2 mM calcium and stimulated with SFLLRN (thrombin receptor-activating peptide) or thrombin. SFLLRN (20 μM) and thrombin (1 U/ml) resulted in maximal irreversible aggregation and 80–90% secretion of dense granule contents. SFLLRN (50–100 μM) caused larger increases at 10 sec than 20 μM SFLLRN in the formation of inositol trisphosphate (IP₃, measured as [³H]inositol label). These increases were not significantly less than those caused by thrombin (1 unit/ml). However, whereas the labeling of IP₃ increased from 10–60 sec with thrombin, with SFLLRN it was much less at 60 sec than that at 10 sec. The decrease was not due to degradation of SFLLRN by ectopeptidases, since it was not prevented by amastatin, an inhibitor of ectopeptidases. Degradation of glycoprotein Ib (GPIb) with an O-sialoglycoprotein endopeptidase did not affect the thrombin-stimulated labeling of inositol phosphates, indicating that binding to GPIb is not involved in the sustained thrombin-induced formation of inositol phosphates. The finding that the thrombin-stimulated formation of IP₃ was not dependent on Ca²⁺ in the medium (EGTA added) indicates that the transient SFLLRN-induced formation of IP₃ is not due to failure to cause Ca²⁺ influx. The finding that formation of IP₃ was transient in SFLLRN-stimulated platelets, whereas platelet aggregation and secretion were maximal, indicates that the sustained activation of phospholipase C caused by thrombin may have roles related to later processes in which platelets participate. Am. J. Hematol. 54:288–295, 1997. © 1997 Wiley-Liss, Inc.     

Human platelet glycoprotein V: its role in enhancing expression of the glycoprotein Ib receptor

Platelet adhesion to an injured blood vessel wall is a critical initiating step in hemostasis mediated by a four member receptor complex (glycoprotein Ib/V/IX) interacting with plasma von Willebrand factor (vWF). The function of the GPV subunit within this complex is presently undefined. To study the role of glycoprotein (GP) V within the GPIb receptor complex, we transfected the GPV subunit gene into a hematopoietic cell line that constitutively expresses the other three subunits (human erythroleukemia [HEL] cells). Using flow cytometry, we found transfected GPV was surface expressed in HEL cells; this, in turn, led to increased surface expression of the ligand-binding GPIb alpha and GPIX subunits. Radioligand binding assays showed that GPV- transfected HEL cells bound more vWF than their non- or mock- transfected counterparts. We employed confocal microscopy of GPV- transfected HEL cells to show that GPV colocalizes with GPIb alpha on the cell surface. These findings suggest that the GPV subunit plays a role within the GPIb receptor complex by enhancing Ib alpha surface expression.     

Evaluation of platelet glycoprotein Ib by fluorescence flow cytometry

Platelet glycoprotein Ib (GpIb), a receptor for von Willebrand's factor (vWF), was studied by way of fluorescence flow cytometry. Using a sandwich staining technique, GpIb was identified by a monoclonal antibody (6D1) directed against an epitope close to the vWF binding site. Platelets from normal individuals were symmetrically distributed with respect to GpIb content. Treatment of washed platelets with plasmin resulted in progressive loss of GpIb as measured by fluorescence flow cytometry and by loss of agglutination response when combined with ristocetin in the presence of vWF. In mixing experiments with GpIb-deficient and normal platelets, it was possible to detect a subpopulation of deficient cells comprising 2% of the total population. Streptokinase treatment of platelet-rich plasma caused loss of the agglutination response to ristocetin and the emergence of a population of GpIb-deficient platelets. Fluorescence flow cytometry appears to be an important new technique by which to study platelet surface receptors.     

Pseudo-Bernard-Soulier syndrome: thrombocytopenia caused by autoantibody to platelet glycoprotein Ib

The Bernard-Soulier syndrome is an inherited bleeding disorder that is due to a deficiency in platelet glycoprotein Ib. Bernard-Soulier platelets fail to agglutinate in response to ristocetin despite normal levels of factor VIII:von Willebrand factor. We report a patient who developed severe refractory thrombocytopenia postsurgically while receiving procainamide therapy. Thrombocytopenia was immune mediated since the patient's platelets bore high levels of antiplatelet antibody. Radioimmunoprecipitation studies demonstrated that the autoantibodies had specificity for platelet glycoproteins Ib and V as well as platelet HLA. The patient's plasma as well as purified immunoglobulin G completely inhibited the ristocetin-induced aggregation of normal platelets but did not inhibit adenosine diphosphate-induced aggregation. The laboratory studies revealed that this patient suffered from antibody-mediated thrombocytopenia with unusual characteristics that we have called pseudo-Bernard-Soulier syndrome.     

Inhibition of platelet glycoprotein Ib, glycoprotein IIb/IIIa, or both by monoclonal antibodies prevents arterial thrombosis in baboons

The antithrombotic efficacy of the monoclonal antibodies 6B4-Fab and MA-16N7C2 against platelet glycoprotein (GP) Ib and GP IIb/IIIa, respectively, on acute platelet-mediated thrombosis was evaluated in a baboon model of femoral artery stenosis, which is a modification of the original Folts model: platelet thrombi form on the injured stenosed artery, producing cyclic flow reductions (CFRs). A dose of 0.6 mg/kg 6B4-Fab significantly reduced the CFRs by 59 +/- 15%, whereas 2 mg/kg 6B4-Fab completely abolished the CFRs without prolongation of the bleeding time. MA-16N7C2 inhibited CFRs by 43 +/- 8% at a dose of 0.1 mg/kg and abolished the CFRs at a dose of 0.3 mg/kg but with a significant prolongation of the bleeding time. Finally, the combination of 0.6 mg/kg 6B4-Fab and 0.1 mg/kg MA-16N7C2 fully prevented the CFRs without prolongation of the bleeding time. The present study demonstrates that the inhibition of platelet GP Ib function by 6B4-Fab is a powerful intervention to prevent platelet thrombus formation in injured arteries without prolongation of the bleeding time; the latter is in contrast to the result after the inhibition of GP IIb/IIIa. Moreover, we demonstrate that combining a GP Ib blocker with a GP IIb/IIIa blocker can achieve a strong antithrombotic effect without increasing the bleeding time. This provides new information that will be beneficial in designing clinical therapeutic approaches.     

Reversibility of thrombin-induced decrease in platelet glycoprotein Ib function

Summary. Thrombin induces a redistribution of glycoprotein (GP) Ib/GP IX complex from the platelet surface into the surface connected canalicular system (SCCS). This redistribution results in a reduced interaction of platelet GP Ib with von Willebrand factor (vWF) bound to subendothelium leading to impaired platelet adhesion. In this study we show that the platelet aggregation and degranulation require concentrations of thrombin above 0.05 U/ml, while the decrease in GP Ib function (about 50% of control value), as determined by ristocetin induced platelet agglutination, can be induced by lower concentrations (0.01–0.04 U/ml). Moreover, we show that when adding thrombin inhibitors to the platelets preincubated with <0.04 U/ml thrombin for 5 min, their agglutinability by ristocetin was gradually recovered within 30 min. indicating that in these conditions the decrease in platelet adhesiveness is reversible. Immuno-electromicroscopic study showed that this restoration of platelet GP Ib function was associated with a reversed translocation of GP Ib from the SCCS to the plasma membrane. The data obtained from counting gold particles showed that the ratio of GP Ib immunolabelling on the external membrane versus that on the SCCS was 3.31±0.90 for resting platelets, down-regulated to 0.84±0.13 ( P >0.05 versus resting platelets) for the platelets treated with 0.04 U/ml thrombin and returned to 2.63±2.21 ( P >0.05 versus resting platelets) after incubation for 30 min with hirudin. However, the translocation of GP Ib was poorly reversed by thrombin inhibitors when higher concentrations of thrombin were used which induced platelet aggregation and large extent of degranulation. We conclude that thrombin affects platelets in a dose dependent manner, and that at low concentrations the decrease in platelet GP Ib related function is a reversible phenomenon.