The glycocalicin portion of platelet glycoprotein Ib expresses both high and moderate affinity receptor sites for thrombin. A soluble radioreceptor assay for the interaction of thrombin with platelets

A soluble radioreceptor assay has been developed to characterize thrombin receptor activities of the human platelet membrane. 125I-Thrombin was added to platelet membranes solubilized in 1% Triton X-100, and thrombin bound to platelet receptors was separated from free thrombin by precipitation with wheat germ agglutinin (WGA) in the presence of alpha 1-acid glycoprotein as carrier. Both high affinity binding (Ki, 0.09 nM; R1, 0.30 pmol/mg protein) and moderate affinity binding (K2, 38 nM; R2, 72 pmol/mg protein) were detected in the detergent-solubilized membrane preparations and these binding parameters were in excellent agreement with values previously determined using intact platelets (Harmon, J. T., and Jamieson, G. A. (1985) Biochemistry 24, 58-64). Using the soluble radioreceptor assay, both high and moderate affinity binding was detected in highly purified preparations of glycoprotein Ib (GPIb) and glycocalicin, and the binding isotherms were identical with those of the crude detergent-solubilized membrane preparation. Treatment of detergent-solubilized membranes with increasing concentrations of a monospecific polyclonal antibody to glycocalicin resulted in the stepwise depletion of GPIb and concomitant reductions of thrombin binding activity. These results demonstrate that both high and moderate affinity binding of thrombin to platelets is completely expressed in the glycocalicin portion of GPIb.     

Platelet activation by thrombin in the absence of the high-affinity thrombin receptor

The receptor status of the moderate-affinity platelet binding site for alpha-thrombin has been established by treating platelets with Serratia marcescens protease under conditions causing cleavage of 95-97% glycoprotein Ib (2.5 micrograms for 30 min). High-affinity binding was lost under these conditions, but the platelets continued to show moderate-affinity binding (Kd1 = 16 +/- 5 nM; 930 +/- 300 sites/platelet) and low-affinity binding (Kd2 = 4.6 +/- 3 microM; 170,000 +/- 90,000 sites/platelet), in good agreement with the values previously obtained for moderate- and low-affinity binding in intact platelets [Harmon, J.T., & Jamieson, G.A. (1986) J. Biol. Chem. 261, 15928-15933]. Platelets treated with Serratia protease under these conditions were about 4-fold less sensitive to activation by alpha-thrombin, as measured by serotonin secretion. Crossover studies with analogues showed that binding of alpha-thrombin was compatible by both D-phenyl-alanyl-L-prolyl-L-arginine chloromethyl ketone treated thrombin and N alpha-p-tosyl-L-lysine chloromethyl ketone treated thrombin, and both analogues were capable of inhibiting activation of Serratia-proteolyzed platelets by alpha-thrombin. These studies establish that the moderate-affinity platelet binding site for alpha-thrombin is a receptor, occupancy of which is required for platelet activation in the absence of the high-affinity receptor.     

Interaction of lectins with human platelets. Effects on platelet stimulation by thrombin and ristocetin.

Wheat germ agglutinin induced aggregation and secretion of fresh platelets. Aggregation, but not secretion of serotonin by platelets in plasma, by the lectin was inhibited by 5 mM EDTA. Further, the lectin-induced stimulation of fresh platelets was blocked by prostaglandin E1. Thus, this lectin stimulates platelets by a mechanism which closely mimics thrombin activation and is independent of intercellular crosslinking. Lentil lectin did not stimulate platelets. Each platelet contained about 6 . 10(-5) binding sites for the lectins with an apparent dissociation constant of 3.0 . 10(-7) M. Wheat germ agglutinin, which binds mainly to glycoprotein I (Mr 150 000), increased the subsequent binding of thrombin to fixed platelets while lentil lectin was without effect. It appears that thrombin and wheat germ agglutinin bind to independent but interacting sites. Wheat germ agglutinin, but neither thrombin nor lentil lectin, inhibited the agglutination of platelets by ristocetin. Further, rat platelets were not aggregated by either ristocetin or wheat germ agglutinin. It appears that the interaction sites of ristocetin and wheat germ agglutinin on platelets are overlapping.     

Identification of the disulphide bonds in human platelet glycocalicin

The glycoprotein Ib/IX complex on platelets is responsible for the first stage of haemostasis as an essential component in the primary adhesion of platelets to damaged vessel walls. Glycocalicin is the extracellular part of platelet glycoprotein Ib alpha and contains the von Willebrand factor and thrombin binding sites. Disulphide bonds are implicated in the von Willebrand binding site and studies with peptides point towards a region of glycocalicin with four cysteines as containing the binding sites for both von Willebrand factor and thrombin. The position and linkage of these two disulphide bonds are now determined to be 209-248 and 211-264 and the relevance of this double-loop structure for glycoprotein Ib/IX function is discussed.     

Platelet responses to compound interactions with thrombin.

Catalytic and noncatalytic interactions of thrombin with platelets are investigated with use of thrombin variants with altered specificities and with ligands of thrombin receptors on platelets. Both alpha-thrombin and weakly coagulant meizothrombin-des-fragment-1 (mu-thrombin) hydrolyze proteolytically activated receptor 1 for thrombin (rPAR1(T), recombinant) with catalytic efficiencies of >10(7) M(-)(1) s(-)(1), whereas rPAR1(T) is not a substrate for weakly coagulant beta-thrombin. In contrast, both mu-thrombin and beta-thrombin are weak agonists of platelet dense body (ATP) secretion. Antibodies that block rPAR1(T) cleavage strongly inhibit the secretory reaction to alpha- and mu-thrombins but not to beta-thrombin or to thrombin receptor activating peptide (TRAP). However, catalytically inactive FPR-thrombin, which binds glycoprotein Ib but does not inhibit rPAR1(T) cleavage, inhibits responses to TRAP as well as those to alpha- and mu-thrombins, which indicates that binding of the inactive enzyme to platelets influences the function of PAR1(T). An antibody that inhibits binding of thrombin to platelet glycoprotein Ib inhibits secretory responses to thrombin but not to TRAP, so occupancy of glycoprotein Ib per se accounts for only part of the attenuation. All three thrombins stimulate a rise in cytosolic Ca(II), and the dose response to beta-thrombin is congruent with that for ATP secretion. However, the response of cytosolic Ca(II) is 10-100 times more sensitive to mu-thrombin and alpha-thrombin than ATP secretion is, and is inhibited by neither anti-PAR1(T) Ig nor FPR-thrombin. Thus, alpha-thrombin appears to have an activity not shared by either mu- or beta-thrombins. This activity is owed to more than coupling of independent signals from cleavage of two proteolytically activated receptors, as there is no synergism when mu-thrombin and beta-thrombin costimulate secretion. It is concluded either that alpha-thrombin has a third interaction site on platelets with which neither mu-thrombin nor beta-thrombin interacts or that dual receptors are coordinately cleaved. In either case, the strong secretory response to thrombin appears to be moderated, independently of cytosolic Ca(II), by occupancy of a noncatalytic interaction site such as glycoprotein Ib.     

Essential groups in synthetic agonist peptides for activation of the platelet thrombin receptor.

Thrombin appears to activate platelets by a novel mechanism that involves the cleavage of its receptor, and it has been proposed that the newly generated N-terminal region of the receptor then acts as a tethered ligand [Vu, T. H., Hung, D. T., Wheaton, V. I., & Coughlin, S. R. (1991) Cell 64, 1057-1068]. Peptides with sequences corresponding to those of the tethered ligand are capable of activating the receptor. In the present study, groups within this tethered ligand peptide that are important for activation of the receptor have been identified by synthesizing a series of peptides. A 14-residue peptide based on the tethered ligand stimulated the aggregation of gel-filtered platelets with an EC50 of 7 microM, and a concentration of 10 microM was the minimum concentration necessary to yield a full aggregation response in platelet-rich plasma. Truncation of the peptide from the C-terminus to nine residues did not markedly affect the response to the peptide. Shorter peptides of five, six, and eight amino acids retained their agonist activity, but the minimal concentration necessary to achieve a full aggregation response in platelet-rich plasma was 2-5-fold higher. Side chains within the tethered ligand peptide that are important for receptor activation were identified by synthesizing a series of peptides in which residues were sequentially replaced by alanine. The results indicated that the side chains of phenylalanine, leucine, and arginine in positions 2, 4, and 5, respectively, are essential for full activity. Most notably, substitution of phenylalanine in the second position resulted in complete loss of agonist activity at concentrations up to 800 microM.(ABSTRACT TRUNCATED AT 250 WORDS)     

Interaction between platelet receptor and iloprost isomers

Iloprost, a stable analog of prostacyclin, has been used for studying the interaction between prostacyclin and its effector cells such as platelets and vascular cells. The compound is usually prepared as a mixture of 16(S) and 16(R) stereoisomers. In this work, we compared the biological activity and platelet receptor binding characteristics between the two isomers. The 16(S) isomer was 20-times more potent than the 16(R) in inhibiting collagen-induced platelet aggregation. Equilibrium binding of iloprost isomers to platelet membrane receptors measured by rapid filtration method revealed that the specific binding data of 16(S) isomer was fit for a single binding species with Kd of 13.4 nM and Bmax 665 fmol/mg protein. By contrast, the Kd and Bmax of 16(R) isomer were 288 nM and 425 fmol/mg, respectively. To further assess different binding behavior of these two isomers, association rate was measured. The observed association rate of the S isomer was 0.036 s-1 and 0.001 s-1 for the R isomer at 15 nM iloprost and 2 mg/ml platelet membrane proteins. We postulate that the striking difference in the association rate with resultant difference in binding affinity and biologic activity between the two isomers was due to fitting of the molecule to the receptor channel. The 16(S) form has a more favorable orientation for fitting into the receptor. We conclude that the two iloprost isomers must be considered as two entirely different compounds when iloprost is used as the ligand for quantifying prostacyclin receptor binding.     

Platelet stimulation by thrombin and other proteases.

The mechanism of stimulation of platelets by thrombin and other proteases was studied by following kinetics of secretion of Ca2+ or ATP. The progress-time curves of secretion were analyzed for rate and total amount released. The reaction of thrombin was perturbed by addition of hydroxylamine or a competitive inhibitor and by variation of pH and it was compared with the reactions of other proteases. Trypsin and papain, with specificities for arginyl residues, induced secretion with a time course that was nearly identical with that induced by thrombin when saturating levels of enzyme were used. At low levels of enzyme, trypsin and papain gave extended lags in the progress-time curves. Higher concentrations of trypsin and papain were required for saturation of the measured parameters. Human plasmin (lysly specificity) and bovine chymotrypsin (aromatic amino acid specificity) failed to induce platelet secretion. Active site inhibited thrombin was also ineffective. Both yield and kinetics depended on pH, with the pH profile for each enzyme similar to its profile for hydrolysis of synthetic substrates. Studies at low pH also showed that the early part of the reaction undergoes a change in rate-determining step from enzyme dependent at low enzyme to enzyme indepdenent at high enzyme. Hydroxylamine, a nucleophile that would be expected to accelerate hydrolytic reactions, actually decreased both the rate of initial reactions and yield. A competitive inhibitor of thrombin also decreased both rate and yield; a calculated inhibition constant was in agreement with the value for a synthetic substrate, suggesting that the interaction of thrombin with platelets is analogous to reaction with substrates. A modification of our previous model is proposed in order to accommodate the results described here and to reaoncile the apparent contradictions that enzyme was found not to turn over in the reaction (Detwiler, T. C., and Feinman, R. D. (1973), Biochemistry 12, 282), that catalytic activity is required (Davey, M. G., and Luscher, E. F. (1967), Nature (London) 216, 875; this paper), and that the reaction is characterized by an apparent equilibrium binding (Tollefsen, D. M., Feagler J. R., and Majerus, P. W. (1974), J. Biol. Chem. 249, 2646). The essential feature is a reversible catalytic step with no dissociation of enzyme from product. This is followed by irreversible, thrombin-independent platelet processes leading to secretion, with yield dependent on the equilibrium concentration of the thrombin product. The model thus has aspects of catalysis, stoichiometry, and an agonist-receptor equilibrium.     

Structure-activity relationships of pyrroloquinazolines as thrombin receptor antagonists.

A series of pyrroloquinazolines has been discovered that represent novel small molecule inhibitors of the intramolecular ligand of the thrombin receptor. Analogs were prepared to study the structure-activity relationships of substitution at the N 1, N3, and N7 positions of the heterocycle. Compounds 4e and 4f have been identified with IC50's of 56 and 52 nM, respectively.     

Studies on the mechanism of thrombin. Interaction with fibrin

Fibrin monomer Sepharose was used to investigate the interactions of thrombin with fibrin. Thrombin binding was found to be reversible and saturable and to depend on the thrombin: fibrin ratio. Scatchard analysis indicated a single class of binding sites with K alpha = 4.9 X 10(5) M-1. Ca2+ ions caused rapid desorption and elution of thrombin from fibrin monomer, and the Ca2+ concentration needed for maximal desorption depended on the fibrin:thrombin ratio. Mg2+, Mn2+, and Sr2+ also released thrombin from fibrin monomer but not as efficiently as Ca2+. These results indicate that divalent metal ions induce a physical change in fibrin monomer which results in desorption of thrombin. Thrombin binding to fibrin in a gel was compared to binding to fibrin monomer. These studies showed that as fibrin monomers polymerize to form the gel network, thrombin is released. Under static conditions the released thrombin remains associated with the gel because diffusion is limited by the gel. However, the thrombin can be readily removed when buffer is allowed to flow through the gel. These results lead to the possibility that thrombin binding to fibrin monomer and its subsequent release, either by Ca2+ or by polymerization, may have important consequences for regulating the effective thrombin concentration in vivo.     

Interaction of vitamin K-dependent protein Z with thrombin. Consequences for the amidolytic activity of thrombin and the interaction of thrombin with phospholipid vesicles

Protein Z is a vitamin K-dependent protein of unknown function present in normal bovine plasma at a concentration of approximately 0.1 microM. Quantitative affinity chromatographic studies using diisopropylphosphoryl (DIP)-thrombin-Affi-Gel 10 as the affinity matrix and free DIP-thrombin as the competitor demonstrated that protein Z interacts with DIP-thrombin with a dissociation constant of 0.15 +/- 0.05 microM. Binding was independent of Ca2+. Protein C and factor IX, other vitamin K-dependent clotting proteins with the same domain structure as that of protein Z, did not interact with immobilized DIP-thrombin under these conditions; and factor X interacted with an affinity 20-fold lower than that for protein Z. The Michaelis constant, Km, for hydrolysis of pyro-Glu-Pro-Arg-p-nitroanilide by thrombin was increased 1.8-fold, from 130 to 230 microM, as a result of the binding of protein Z and the Km for H-Val-Leu-Arg-p-nitroanilide 1.4-fold, from 390 to 560 microM. From these kinetic studies, a dissociation constant of 0.11 +/- 0.04 microM was calculated for the binding of protein Z to alpha-thrombin. Protein Z bound to large phospholipid vesicles (25% phosphatidylserine, 75% phosphatidylcholine) with a dissociation constant of 0.39 +/- 0.16 microM at a phospholipid to protein ratio of 82 mol of phospholipid/mol of protein Z at saturation. In the presence of protein Z thrombin associated with phospholipid vesicles, whereas thrombin did not interact with phospholipid vesicles in the absence of protein Z. These studies, therefore, demonstrate a physiologically relevant interaction between protein Z and thrombin. They also suggest a mechanism whereby thrombin is localized to an injury site by virtue of its interaction with protein Z bound to phospholipid surfaces.     

Rapid purification and partial characterization of human platelet glycoprotein IIIb. Interaction with thrombospondin and its role in platelet aggregation

Glycoprotein IIIb (also known as glycoprotein IV) is a major glycoprotein present on the surface of human platelets. Recent studies suggest that glycoprotein IIIb may be a receptor site for thrombospondin. Thrombospondin, a multifunctional adhesive glycoprotein released from stimulated platelets, plays an important role in the stabilization of platelet aggregates. In this study, a new method for the purification of glycoprotein IIIb is described. Glycoprotein IIIb was isolated from Triton X-114 platelet membrane extracts, under nondenaturing conditions, by tandem anion-exchange and size exclusion fast protein liquid chromatography. The purified glycoprotein had the same apparent molecular mass (88 kDa) under nonreducing or reducing conditions. The tryptic peptide map of the purified protein was identical to that of bona fide glycoprotein IIIb as isolated from two-dimensional polyacrylamide gels of platelet membrane proteins. In addition, the purified glycoprotein was recognized by an anti-GPIIIb monoclonal antibody (OKM5). The purified glycoprotein specifically bound to thrombospondin in the presence of calcium. Monospecific anti-GPIIIb antibodies interfered with the expression of endogenous thrombospondin on thrombin-activated platelets and partially inhibited collagen- and thrombin-induced platelet aggregation without a significant effect on platelet secretion. Glycoprotein IIIb, by interacting with thrombospondin on the activated platelet surface, may play an important role in the platelet aggregation process.     

Interaction of thrombin with sucrose octasulfate

The serine protease thrombin plays multiple roles in many important physiological processes, especially coagulation, where it functions as both a pro- and anticoagulant. The polyanionic glycosaminoglycan heparin modulates thrombin's activity through binding at exosite II. Sucrose octasulfate (SOS) is often used as a surrogate for heparin, but it is not known whether it is an effective heparin mimic in its interaction with thrombin. We have characterized the interaction of SOS with thrombin in solution and determined a crystal structure of their complex. SOS binds thrombin with a K(d) of ~1.4 μM, comparable to that of the much larger polymeric heparin measured under the same conditions. Nonionic (hydrogen bonding) interactions make a larger contribution to thrombin binding of SOS than to heparin. SOS binding to exosite II inhibits thrombin's catalytic activity with high potency but with low efficacy. Analytical ultracentrifugation shows that bovine and human thrombins are monomers in solution in the presence of SOS, in contrast to their complexes with heparin, which are dimers. In the X-ray crystal structure, two molecules of SOS are bound nonequivalently to exosite II portions of a thrombin dimer, in contrast to the 1:2 stoichiometry of the heparin-thrombin complex, which has a different monomer association mode in the dimer. SOS and heparin binding to exosite II of thrombin differ on both chemical and structural levels and, perhaps most significantly, in thrombin inhibition. These differences may offer paths to the design of more potent exosite II binding, allosteric small molecules as modulators of thrombin function.     

Complexes of thrombin with secreted platelet proteins

A labeled 77-kDa complex formed when 125I-thrombin was added to platelet suspensions or to the supernatant solution of ionophore-activated platelets. Prostacyclin inhibited complex formation with whole platelets but not with the supernatant solution of ionophore-activated platelets. This is evidence that the complex formed with a factor secreted from activated platelets. Smaller complexes of 70 and 58 kDa formed between labeled thrombin and lysed platelets. The 77-kDa complex was necessary for the formation of a thrombin-thrombospondin complex.     

Interaction of thrombin with PAR1 and PAR4 at the thrombin cleavage site

Investigations determined the critical amino acids for alpha-thrombin's interaction with protease-activated receptors 1 and 4 (PAR1 and PAR4, respectively) at the thrombin cleavage site. Recombinant PAR1 wild-type (wt) exodomain was cleaved by alpha-thrombin with a Km of 28 microM, a kcat of 340 s-1, and a kcat/Km of 1.2 x 10(7). When the P4 or P2 position was mutated to alanine, PAR1-L38A or PAR1-P40A, respectively, the Km was unchanged, 29 or 23 microM, respectively; however, the kcat and kcat/Km were reduced in each case. In contrast, when Asp39 at P3 was mutated to alanine, PAR1-D39A, Km and kcat were both reduced approximately 3-fold, making the kcat/Km the same as that of PAR1-wt exodomain. Recombinant PAR4-wt exodomain was cleaved by alpha-thrombin with a Km of 61 microM, a kcat of 17 s-1, and a kcat/Km of 2.8 x 10(5). When the P5 or P4 position was mutated to alanine, PAR4-L43A or PAR4-P44A, respectively, there was no change in the Km (69 or 56 microM, respectively); however, the kcat was lowered in each case (9.7 or 7.7 s-1, respectively). Mutation of the P2 position (PAR4-P46A) also had no effect on the Km but markedly lowered the kcat and kcat/Km approximately 35-fold. PAR1-wt exodomain and P4 and P3 mutants were noncompetitive inhibitors of alpha-thrombin hydrolyzing Sar-Pro-Arg-pNA. However, PAR1-P40A displayed a mixed type of inhibition. Mutation of P4, P3, or P2 had no effect on the Ki. All PAR4 exodomains were competitive inhibitors of alpha-thrombin. Mutation of P5, P4, or P2 had no effect on the Ki. These investigations show that Leu at P4 in PAR1 or P5 in PAR4 critically influences the kinetics of alpha-thrombin binding and cleavage of PAR1 and PAR4 exodomains. It also implies that factors other than the hirudin-like binding region on PAR1 exodomain predominate in influencing PAR1 cleavage on cells.     

Design and synthesis of para-fluorophenylalanine amide derivatives as thrombin receptor antagonists.

An antagonist specific for the thrombin receptor is expected to be a remedy for thrombosis. Structure-activity studies of thrombin receptor-tethered ligand SFLLRNP have revealed the importance of the Phe-2-phenyl group in receptor recognition and the replacement of the Phe-2 by para-fluorophenylalanine [(p-F)Phe] was found to enhance its activity [Nose, T. et al. (1993) Biochem. Biophys. Res. Commun. 193, 694-699]. In order to obtain a small sized antagonist, a series of (p-F)Phe derivatives was designed and synthesized novel structural elements essential for receptor interactions being introduced at both the N and C-termini. beta-Mercaptopropionyl (betaMp) or its derivative activated by S-3-nitro-2-pyridinesulphenyl (Npys) was introduced at the N-terminus, and phenylmethyl amines were coupled to the C-terminus. All compounds were inactive when assayed for human platelet aggregation, indicating that they are not agonists. beta-Mercaptopropionyl derivatives were also inactive as antagonists. However, Npys-containing analogs were found to inhibit the agonist activity of SFLLRNP. In particular, SNpys-betaMp-(p-F)Phe-NH-R [R = -CH(C6H5)2 and -CH2-CH-(C6H5)2] potently suppressed platelet aggregation. The results suggested that (p-F)Phe can be used as a structural core to construct an effective antagonist conformation.     

Structure and function of the human platelet thrombin receptor. Studies using monoclonal antibodies directed against a defined domain within the receptor N terminus.

Based upon its recently cloned nucleotide sequence, the human platelet thrombin receptor is thought to be formed by a single polypeptide chain with seven transmembrane domains and an extracellular N terminus that can be cleaved by thrombin. As yet, however, little is known from studies of the receptor protein itself. To obtain such information, we have prepared monoclonal antibodies against a peptide corresponding to receptor residues Ser42 through Phe55, the domain immediately distal to the site of cleavage by thrombin. By flow cytometry, all of the antibodies reacted with the thrombin-responsive megakaryoblastic CHRF-288 and HEL cell lines, but not with the T-lymphoid Sup-T1 cell line. Functionally, the antibodies inhibited platelet responses to alpha-thrombin, gamma-thrombin, and trypsin, but had no effect on platelet activation by ADP, epinephrine, or the thromboxane analog U46619. Radioiodinated antibody bound to approximately 1,800 sites/platelet, a value similar to the reported number of moderate affinity thrombin binding sites per platelet. On Western blots, the antibodies recognized a 66-kDa protein in platelet, HEL, and CHRF-288 membranes. The discrepancy between this apparent size and the predicted mass of the receptor suggests that, as with other G protein-coupled receptors, one or more of the potential sites for N-linked glycosylation have been utilized. Therefore, these results suggest that: 1) the cloned thrombin receptor is involved in a broad range of platelet responses to thrombin, as well as gamma-thrombin and trypsin; 2) as predicted, the N terminus of the receptor is accessible on the platelet surface; 3) the moderate affinity thrombin binding site noted in earlier studies may be the receptor; 4) potentially as much as one third of the mass of the receptor is carbohydrate.