Structure of the hirugen and hirulog 1 complexes of alpha-thrombin

The isomorphous structures of the hirugen (N-acetylhirudin 53'-64' with sulfato-Tyr63') and hirulog 1 (D-Phe-Pro-Arg-Pro-(Gly)4 desulfato-Tyr63'-hirugen) complexes of human alpha-thrombin have been determined and refined at 2.2 A resolution to crystallographic R-factors of 0.167 and 0.163, respectively. The binding of hirugen to thrombin is similar to that of the binding of the C-terminal dodecapeptide of hirudin, including that of the terminal 3(10) helical turn. The sulfato Tyr63', which, as a result of sulfation, increases the binding affinity by an order of magnitude, is involved in an extended hydrogen bonding network utilizing all three sulfato oxygen atoms. The hirugen-thrombin complex is the first thrombin structure determined to have an unobstructed active site; this site is practically identical in positioning of catalytic residues and in its hydrogen bonding pattern with that of other serine proteinases. Hirulog 1, which is a poor thrombin substrate, is cleaved at the Arg3'-Pro4' bond in the crystal structure. The Arg3' of hirulog 1 occupies the specificity site, the D-Phe-Pro-Arg tripeptide is positioned like that of D-Phe-Pro-Arg chloromethylketone in the active site and the Pro4'(Gly)4 spacer to hirugen is disordered in the structure, as is the 3(10) turn of hirugen. The latter must be related to the simultaneous absence both of sulfation and of the last residue of hirudin (Gln65'). In addition, the autolysis loop of thrombin (Lys145-Gly150) is disordered in both structures. Changes in circular dichroism upon hirugen binding are therefore most likely the result of the flexibility associated with this loop.

Mechanistic studies on thrombin catalysis

The kinetic mechanism of the cleavage of four p-nitroanilide (pNA) substrates by human alpha-thrombin has been investigated by using a number of steady-state kinetic techniques. Solvent isotope and viscosity effects were used to determine the stickiness of the substrates at the pH optimum of the reaction; a sticky substrate is defined as one that undergoes catalysis faster than it dissociates from the Michaelis complex. Whereas benzoyl-Arg-pNA could be classified as a nonsticky substrate, D-Phe-pipecolyl-Arg-pNA was very sticky. The other two substrates (tosyl-Gly-Pro-Arg-pNA and acetyl-D-Phe-pipecolyl-Arg-pNA) were slightly sticky. The pH profiles of kcat/Km were bell-shaped for all substrates. The pKa values determined from the pH dependence of kcat/Km for benzoyl-Arg-pNA were about 7.5 and 9.1. Similar pKa values were determined from the pH profiles of kcat/Km for tosyl-Gly-Pro-Arg-pNA and acetyl-D-Phe-pipecolyl-Arg-pNA and for the binding of the competitive inhibitor N alpha-dansyl-L-arginine-4-methylpiperidine amide. The groups responsible for the observed pKa values were proposed to be His57 and the alpha-amino group of Ile16. The temperature dependence of the pKa values was consistent with this assignment. The pKa values of 6.7 and 8.6 observed in the pH profile of kcat/Km for D-Phe-pipecolyl-Arg-pNA were displaced to lower values than those observed for the other substrates. The displacement of the acidic pKa value could be attributed to the stickiness of this substrate.(ABSTRACT TRUNCATED AT 250 WORDS)

A thrombin-based peptide corresponding to the sequence of the thrombomodulin-binding site blocks the procoagulant activities of thrombin

Thrombomodulin, a cofactor in the thrombin-catalyzed activation of protein C, blocks the procoagulant activities of thrombin such as fibrinogen clotting, Factor V activation, and platelet activation. The binding site for thrombomodulin within human thrombin has been localized at a region comprising residues Thr147-Ser158 of the B-chain of thrombin. The dodecapeptide sequence, TWTANVGKGQPS, corresponding to these residues inhibits thrombin binding to thrombomodulin with an apparent Ki = 94 microM (Suzuki, K., Nishioka, J., and Hayashi, T. (1990) J. Biol. Chem. 265, 13263-13267). We have found that the inhibitory effect of the dodecapeptide on the thrombin-thrombomodulin interaction is sequence-specific, and that residues Asn151, Lys154, and Gln156 are essential for thrombomodulin binding. The dodecapeptide was also found to directly block thrombin procoagulant activities, fibrinogen clotting (concentration for half-maximum inhibition, 385 microM). Factor V activation (concentration for half-maximum inhibition, 33 microM), and platelet activation (concentration for half-maximum inhibition, 645 microM). This peptide did not block thrombin inhibition by antithrombin III, but blocked thrombin inhibition by hirudin. These findings suggest that the binding site for thrombomodulin in thrombin is shared with the sites for fibrinogen, Factor V, platelets, and hirudin, and that, therefore, the inhibition of thrombin procoagulant activities by thrombomodulin in part results from blocking of the interaction between thrombin and the procoagulant protein substrates by thrombomodulin.

Thrombin-hirudin complex stability: a comparison with the thrombin-antithrombin III complex

In the present in vitro study the stabilities of the thrombin-hirudin and the thrombin-antithrombin III complexes were investigated. After incubation of the complexes with free inhibitors the thrombin-antithrombin III levels were determined by ELISA. The thrombin-hirudin complex proved to be stable in the presence of antithrombin III or heparin. However, in the presence of heparin and plasma equivalent concentrations of antithrombin III, the thrombin-hirudin complex dissociated and hirudin was displaced. In contrast, both thrombin-antithrombin III and thrombin-antithrombin III/heparin complexes are very stable even in the presence of a large excess of hirudin.

Refined structure of the hirudin-thrombin complex

The structure of a recombinant hirudin (variant 2, Lys47) human alpha-thrombin complex has been refined using restrained least-squares methods to a crystallographic R-factor of 0.173. The hirudin structure consists of an N-terminal domain folded into a globular unit and a long 17-peptide C-terminal in an extended chain conformation. The N-terminal domain binds at the active-site of thrombin where Ile1' to Tyr3' penetrates to the catalytic triad. The alpha-amino group of Ile1' of hirudin makes a hydrogen bond with OG of Ser195 of thrombin, the side-chains of Ile1' and Tyr3' occupy the apolar site, Thr2' is at the entrance to, but does not enter, the S1 specificity site and Ile1' to Tyr3' form a parallel beta-strand with Ser214 to Gly219. The latter interaction is antiparallel in all other serine proteinase-protein inhibitor complexes. The extended C-terminal segment of hirudin, which is abundant in acidic residues, makes many electrostatic interactions with the fibrinogen binding exosite while the last five residues are in a 3(10) helical turn residing in a hydrophobic patch on the thrombin surface. The precision of the complementarity displayed by these two molecules produces numerous interactions, which although independently generally weak, together are responsible for the high degree of affinity and specificity. Although hirudin-thrombin and D-Phe-Pro-Arg-chloromethyl ketone-thrombin differ in conformation in the autolysis loop (Lys145 to Gly150), this is most likely due to different crystal packing interactions and changes in circular dichroism between the two are probably due to the inherent flexibility of the loop. An RGD sequence, which is generally known to be involved in cell surface receptor interactions, occurs in thrombin and is associated with a long solvent channel filled with water molecules leading to the surface from the end of the S1 site. However, the RGD triplet does not appear to be able to interact in concert in a surface binding mode.

Thrombin Glu-39 restricts the P'3 specificity to nonacidic residues

Residue 39 of serine proteases neighbors positions P'2 to P'4 of the substrate. When Glu-39 of thrombin is replaced with Lys, the resultant enzyme (E39K) retains similar P1, P2, and P3 specificities but has altered P'3 and/or P'4 specificities. These conclusions are based on analysis of both p-nitroanilide and synthetic peptide hydrolysis. The activity of E39K is nearly normal toward 17 p-nitroanilide substrates. In peptide substrates, an acidic residue at either the P3 or P'3 position reduces the rate of cleavage by thrombin. A single substitution of Asp with Gly in either the P3 or P'3 position of a peptide corresponding to the P7-P'5 residues of protein C increases the rate of cleavage by thrombin 2-3-fold. Replacement of both Asp residues with Gly increases the rate of cleavage 30-fold. With E39K, the inhibitory effect of Asp in P3 remains unchanged, but Asp in the P'3 site is no longer inhibitory. Significant differences in the catalytic activity of E39K are also seen with respect to protein C activation. In the absence of thrombomodulin, E39K activates protein C 2.2 times faster than thrombin. In the presence of thrombomodulin, the rate of protein C activation is similar for E39K and thrombin. The second order rate constant of inhibition by antithrombin III, where P'4 is a Glu, is slightly increased (1.4-fold). The clotting activity is reduced 2.4-fold due to a lower rate of fibrinopeptides A and B release where P'3 is Arg. These data show that the P'3 position is a determinant of thrombin specificity and suggest that thrombomodulin may function in part by alleviating the inhibitory effects that may arise from the proximity of the Asp in P'3 of protein C with Glu-39 of thrombin.

Expression of a synthetic gene for human cap binding protein (human IF-4E) in Escherichia coli and fluorescence studies on interaction with mRNA cap structure analogues

An artificial gene coding for the human cap binding protein (hCBP: human IF-4E) was chemically synthesized and expressed in Escherichia coli under the control of a trp promoter. The DNA duplex of 662 bp was designed and constructed from 44 oligodeoxynucleotide fragments of typically 30 nucleotides in length. Although the hCBP gene was not directly expressed in E. coli HB101, we succeeded in its high-level expression as a fusion protein connected with a portion of human growth hormone through a tetradecapeptide (Asp-Asp-Pro-Pro-Thr-Val-Glu-Leu-Gln-Gly-Leu-Val-Pro-Arg) that contains the recognition sequence for a site-specific protease alpha-thrombin. Upon induction with 3-indoleacrylic acid, the fusion protein accumulated with a yield of about 20% of the total proteins of the host cell. Upon the treatment of the fusion protein with alpha-thrombin, which recognizes the sequence "Val-Pro-Arg," specific proteolysis at the fused junction occurred efficiently. In this system, nonspecific digestion by alpha-thrombin was not marked. About 15 mg of recombinant hCBP was obtained from a 1-liter culture. Association constants between the recombinant hCBP and mRNA cap structure analogues were determined by fluorescence spectroscopy. The values obtained for the m7GpppA, m7GTP, and m7GMP were almost the same as those reported for the IF-4E isolated from human erythrocyte cells.(ABSTRACT TRUNCATED AT 250 WORDS)

Ionic interactions in the formation of the thrombin-hirudin complex

Site-directed mutagenesis has been used to examine the importance of each of the acidic C-terminal residues of hirudin in the formation of its complex with alpha-thrombin. The contribution to binding energy of acidic residues in the 11 C-terminal amino acids varied from 2.3 to 5.9 kJ.mol-1. The differences between the contributions of individual residues were smaller than would be expected from the crystal structures of the thrombin-hirudin complex. In particular, the small effect (2.4 kJ.mol-1) for the replacement of Asp-55 was surprising in view of the two salt bridges made by this residue. The results of studies involving multiple mutations indicated that the additivity of the effects varied with the position of the mutation. Whereas the effect of mutations involving the glutamic acid residues at positions 61 and 62 were additive, non-additivity was observed with the glutamic acid residues at positions 57 and 58.

Effect of platelet activation on the conformation of the plasma membrane glycoprotein IIb-IIIa complex

Platelet activation converts the membrane GP IIb-IIIa complex into a functional receptor for fibrinogen, but the mechanism is poorly understood. We asked whether induction of receptor competency coincides with a conformational change affecting the spatial arrangement of exoplasmic domains of the IIb and IIIa subunits. Epitopes on these subunits were labeled with monoclonal antibodies conjugated to either a donor fluorescein (FITC) or an acceptor tetramethylrhodamine (TR) chromophore. Then, fluorescence resonance energy transfer (RET) between platelet-bound FITC and TR was measured by flow cytometry. In unstimulated platelets, 6-8% RET efficiency was detected between antibody B1B5, bound to GP IIb, and antibody SSA6, bound to GP IIIa, regardless of which antibody served as RET donor. RET was also observed between these antibodies and A2A9, an antibody specific for the GP IIb-IIIa complex. Cell stimulation by thrombin, ADP plus epinephrine or phorbol-ester caused up to a 2-fold increase in RET between chromophore-labeled, platelet-bound B1B5, SSA6, and A2A9 (p less than or equal to 0.05), suggesting a change in the separation or orientation of these epitopes within the GP IIb-IIIa complex. The activation-related conformational change detected by the increase in RET between antibody B1B5 and SSA6 was independent of receptor occupancy since it was unaffected by the addition of fibrinogen or by the inhibition of fibrinogen binding by the antibody, A2A9, or the peptide, RGDS. In contrast to these results with antibodies bound to different epitopes within GP IIb-IIIa, no RET was observed between FITC-A2A9 and TR-A2A9 bound to different GP IIb-IIIa complexes or between a TR-labeled GP Ib antibody and FITC-labeled GP IIb-IIIa antibodies. These studies demonstrate that platelet activation causes a change in the spatial separation or orientation of exoplasmic domains within GP IIb and IIIa, which may serve to convert this integrin into a functional adhesion receptor.

Predominant contribution of surface approximation to the mechanism of heparin acceleration of the antithrombin-thrombin reaction. Elucidation from salt concentration effects

Heparin has been shown to accelerate the inactivation of alpha-thrombin by antithrombin III (AT) by promoting the initial encounter of proteinase and inhibitor in a ternary thrombin-AT-heparin complex. The aim of the present work was to evaluate the relative contributions of an AT conformational change induced by heparin and of a thrombin-heparin interaction to the promotion by heparin of the thrombin-AT interaction in this ternary complex. This was achieved by comparing the ionic and nonionic contributions to the binary and ternary complex interactions involved in ternary complex assembly at pH 7.4, 25 degrees C, and 0.1-0.35 M NaCl. Equilibrium binding and kinetic studies of the binary complex interactions as a function of salt concentration indicated a similar large ionic component for thrombin-heparin and AT-heparin interactions, but a predominantly nonionic contribution to the thrombin-AT interaction. Stopped-flow kinetic studies of ternary complex formation under conditions where heparin was always saturated with AT demonstrated that the ternary complex was assembled primarily from free thrombin and AT-heparin binary complex at all salt concentrations. Moreover, the ternary complex interaction of thrombin with AT bound to heparin exhibited a substantial ionic component similar to that of the thrombin-heparin binary complex interaction. Comparison of the ionic and nonionic components of thrombin binary and ternary complex interactions indicated that: 1) additive contributions of ionic thrombin-heparin and nonionic thrombin-AT binary complex interactions completely accounted for the binding energy of the thrombin ternary complex interaction, and 2) the heparin-induced AT conformational change made a relatively insignificant contribution to this binding energy. The results thus suggest that heparin promotes the encounter of thrombin and AT primarily by approximating the proteinase and inhibitor on the polysaccharide surface. Evidence was further obtained for alternative modes of thrombin binding to the AT-heparin complex, either with or without the active site of the enzyme complexed with AT. This finding is consistent with the ternary complex encounter of thrombin and AT being mediated by thrombin binding to nonspecific heparin sites, followed by diffusion along the heparin surface to a unique site adjacent to the bound inhibitor.

Quantitative characterization of the thrombin-heparin interaction. Discrimination between specific and nonspecific binding models

Equilibrium binding of human alpha-thrombin to heparin was investigated at pH 7.4 as a function of thrombin and heparin concentrations, NaCl concentration, temperature, and heparin chain length with the extrinsic fluorescence probe, p-aminobenzamidine, or by quantitative affinity chromatography, in order to distinguish between sequence-specific and nonspecific electrostatic modes of binding. Analysis of binding data by a nonspecific binding model developed for protein-nucleic acid interactions, or by the discrete binding site model previously used to analyze the thrombin-heparin interaction, indicated that both models described the binding interaction equally well over the range of thrombin binding densities accessible to measurement. However, the strong dependence of the thrombin-heparin binding interaction on NaCl concentration, its minimal dependence on temperature, and the increase in apparent binding affinity with increasing heparin oligosaccharide chain length were best accounted for by a nonspecific electrostatic association of thrombin with 5 to 6 anionic residues contained in a 3-disaccharide binding site of heparin. This interaction was characterized by an intrinsic dissociation constant (KD,obs) of 6-10 microM at physiological ionic strength. Although the nonspecific binding model satisfactorily described the binding of thrombin to heparin chains ranging in size from 3 to approximately 13 disaccharides in terms of a single intrinsic KD,obs, deviations from this model were apparent with longer heparin chains (approximately 22 to approximately 35 disaccharides) from a progressive decrease in the intrinsic KD,obs of up to 4-fold. Sedimentation equilibrium analyses of thrombin-heparin complexes suggested a second weaker binding site on thrombin for heparin, which accounted for these deviations as well as the observed insolubility of thrombin-heparin complexes at high thrombin binding densities.

Isolation and characterization of a thrombin inhibitor from the tick ixodes ricinus

The secretion of the salivary glands of ticks, Ixodes ricinus, contains anticoagulant substances. Some years ago an antithromboplastin, ixodin, was described. The present paper refers the isolation and characterization of a second anticoagulant substance of the ticks named ixin. Ixin proved to be a relatively stable and specific thrombin inhibitor. The multi step procedure for isolation results in a 850-fold purification. The preparation obtained has a specific activity of 250 antithrombin units/mg. It is not homogeneous and still contaminated. The substance was assumed to be a miniprotein.

Thrombin structure and function: why thrombin is the primary target for antithrombotics

Thrombin has both beneficial and harmful effects. In order of importance, at very low concentrations, alpha-thrombin firstly amplifies its own generation through the activation of factors V and VIII, which are the primary targets of antithrombotic agents. It secondly functions at the cellular level where, at low concentrations it activates platelets, and at higher concentrations, induces endothelial cell changes (e.g., shape changes, albumin transport release of plasminogen activators and other substances). It thirdly converts fibrinogen into clottable fibrin and becomes actively incorporated into the forming thrombus. In addition, it activates protein C, which in turn degrades factors V and VIII (and/or their activated forms) and causes the shutdown of thrombin generation. When compared to other serine proteinases of the blood coagulation and fibrinolytic systems, alpha-thrombin is unique in that it loses most of its proenzyme activation fragment and has developed multisite short-ranged bridge-binding interactions, which appear to explain thrombin specificity. To understand thrombin is to understand haemostasis.

Subtle differences in active site structure between bovine and human thrombins: ESR and fluorescence studies

The primary structures of bovine and human alpha-thrombins are highly homologous yet their x-ray structures are not yet complete enough to distinguish differences. In order to probe and compare their dynamic conformations in solution, we examined bovine and human alpha-thrombins with a series of active site directed fluorosulfonylphenyl spin labeled inhibitors and fluorophores which probe a region within 10-15 A of the catalytic serine residue. Overall, the nitroxide moieties were more immobilized in the bovine vs human derivatives reflecting either more apolar binding regions or steric obstructions to the motion of the nitroxide in bovine thrombin. Most of the labels which distinguish indole (apolar ligand) binding in human thrombin were found to display similar interactions in bovine thrombin, although slight differences in the general topography of this region were suggested. The two active site directed fluorophores, dansyl fluoride and p-nitrophenyl anthranilate showed differences in both lambda emmax and lambda exmax of the complexes with bovine and human-alpha-thrombin, respectively, Several of the effects observed i.e., ligand binding (indole or benzamidine) and the subtle hydrophobic interactions between the nitroxide moiety and the protein active site would be difficult to assess from an x-ray structure determination alone.

Beta-fibrinogenase from the venom of Vipera lebetina

An arginine esterase was purified from the venom of Vipera lebetina by gel filtration on Sephadex G-100 and by affinity and DEAE-cellulose chromatography. The enzyme has a mol. wt of 52,500 and pI approximately 3. It is a glycoprotein containing 23% of neutral sugars, and has extremely high thermostability. The esterase activity is inhibited by diisopropylfluorophosphate (DFP) and phenylmethylsulfonyl fluoride (PMSF). The Km and kcat values are for alpha-N-benzoyl-L-arginine ethyl ester (BAEE) 7.7 x 10(-5) M and 43.8 sec-1, for p-tosyl-L-arginine methyl ester (TAME) 3.6 x 10(-4) M and 39.8 sec-1 (pH 8.5, 25 degrees C, and for alpha-N-benzoyl-DL-arginine-4-nitroanilide (BAPNA) 1.8 x 10(-4) M and 0.94 sec-1 (pH 8.3, 25 degrees C), respectively. Lysine esters are not hydrolyzed. The enzyme has weak caseinolytic activity and hydrolyzes glucagon at the sites Lys12-Tyr13, Arg17-Arg18 and Arg18-Ala19. In fibrinogen it cleaves B beta-chain first and later also the A alpha-chain.

Monoclonal antibodies for human thrombomodulin which recognize binding sites for thrombin and protein C

Monoclonal antibodies for human thrombomodulin, a cofactor for thrombin-catalyzed activation of protein C, were prepared and their epitopes characterized. All six antibodies (MFTM-1-MFTM-6) bound to an elastase-digested active fragment of thrombomodulin, which contains six consecutive EGF domains. Binding of thrombomodulin to these antibodies did not depend on Ca2+ concentration. MFTM-4, MFTM-5, and MFTM-6 strongly inhibited protein C activation by thrombin and thrombomodulin. MFTM-4 and MFTM-5 inhibited thrombin binding to fixed thrombomodulin and bound to a recombinant mutant EGF456 protein, which contained the fourth, fifth, and sixth EGF domains of thrombomodulin. However, MFTM-6 did not inhibit thrombin binding to thrombomodulin and did not bind to EGF456 protein. Binding of thrombomodulin to fixed MFTM-4 or MFTM-5 was competitively inhibited by a recombinant mutant EGF45 protein which contained the fifth and sixth EGF-domains. These results suggest that epitopes of MFTM-4 and MFTM-5 are located in the fifth EGF domain of thrombomodulin. Thus, the binding site for thrombin is located in the fifth EGF domain. These results also suggest that an epitope for MFTM-6 is located at a region near the binding site for gamma-carboxyglutamic acid residues of protein C via Ca2+ on thrombomodulin.

Chemical modification of alpha-thrombin and in vitro characterization of its anticoagulant activity

Acetylthrombin, succinylthrombin, maleylthrombin, polyethylene glycol (PEG)-thrombin, and acetylated PEG-thrombin were prepared and characterized. Although clotting activities could be reduced remarkably by the modification, the potencies of all the modified thrombins for protein C activation also decreased. However, in the case of acetylthrombin, succinylthrombin, and acetylated PEG-thrombin, the remaining potencies for protein C activation exceeded largely the remaining clotting activities. Moreover, by the addition of thrombodulin, the potencies for protein C activation were enhanced 16 - 60 fold for succinylthrombin, maleylthrombin, and PEG-thrombin.

[Hirudin inhibition of human thrombin activity]

Thermodynamics and kinetics for hirudin binding to human alpha-, beta- and gamma-thrombin, under experimental conditions which mimic the in vivo conditions (i.e., pH = 7.35, T = 37.0 degrees C, I = 0.1 M), indicate that the inhibitor specificity for the three species of the enzyme is different. Such a finding agrees with the human thrombin:hirudin binding geometry as revealed by X-ray crystal studies. From the therapeutic viewpoint, thermodynamics and kinetics here reported indicate that the inhibitory activity of hirudin in vivo is more effective than that shown by antithrombin III, which is generally considered the most important plasma thrombin inhibitor.

Localization of the structural domain responsible for the chemotactic properties of thrombin on polymorphonuclear leukocytes

Human alpha thrombin at 1.1.10(-5) M is chemotactic for human polymorphonuclear leukocytes. This thrombin property disappears when the alpha thrombin (1.1.10(-5) M) hirudin (1.32.10(-5) M) mixture is realized. The same result is obtained when the thrombin at 1.1. 10(-5) M is inhibited by antithrombin III in a ratio of 1 mol of thrombin for 4.5 mol of antithrombin III. The hirudin and the antithrombin III appear therefore to mask, by their binding the structural domain responsible for the chemotactic properties of thrombin on polymorphonuclear leukocytes.

Specific interaction of vitronectin with the cell-secreted protease inhibitor glia-derived nexin and its thrombin complex

Interaction of vitronectin with glia-derived nexin (GDN), thrombin, and the complex GDN-thrombin was demonstrated in direct binding assays that indicated the formation of binary and ternary complexes. The concentration of vitronectin necessary to obtain 50% saturation of the immobilized GDN-thrombin complex binding sites (EC50) was about 1 nM. Under similar experimental conditions, the EC50 of vitronectin for the immobilized antithrombin-III-thrombin complex was about fivefold higher. A tight complex was also formed between vitronectin and immobilized GDN (EC50 approximately 1.5 nM) but when vitronectin was immobilized, GDN displayed a reduced affinity for vitronectin (EC50 approximately 10 nM). These results suggest differences between the immobilized and free conformations of GDN and/or vitronectin. In contrast, vitronectin displayed negligible affinity for antithrombin III. Biotinylated GDN was used to characterize further the binding of GDN or the GDN-thrombin complex to vitronectin. The interaction of the biotinylated GDN-thrombin complex with immobilized vitronectin (EC50 approximately 2 nM) was completely blocked by nonbiotinylated complexes of thrombin with either GDN or antithrombin III, whereas free GDN, free thrombin and the GDN-trypsin complex were only weak competitors. Active-site-blocked urokinase and the complex GDN-urokinase also strongly competed for binding of the biotinylated GDN-thrombin complex to vitronectin. Binding of biotinylated GDN to immobilized vitronectin was specific, saturable and was competed with decreasing efficiency by the GDN-thrombin complex, free GDN and free antithrombin III. These interactions between the adhesive component vitronectin and the serine protease inhibitor GDN may relate to localized control of thrombin and/or urokinase action at certain extravascular sites. These results are discussed in terms of binding sites for vitronectin on GDN, thrombin, and the GDN-thrombin complex.