HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 277, Issue 9, 6788-6798, March 1, 2002

This Article Abstract Full Text (PDF) All Versions of this Article: 277/9/6788    most recent M110257200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Verhamme, I. M. Articles by Bock, P. E. Search for Related Content PubMed PubMed Citation Articles by Verhamme, I. M. Articles by Bock, P. E. Social Bookmarking                      What's this?

Binding of Exosite Ligands to Human Thrombin

RE-EVALUATION OF ALLOSTERIC LINKAGE BETWEEN THROMBIN EXOSITES I AND II^*

Ingrid M. Verhamme, Steven T. Olson^§, Douglas M. Tollefsen^¶, and Paul E. Bock

From the  Department of Pathology, Vanderbilt University School of Medicine, Nashville, Tennessee 37232, the ^¶ Department of Internal Medicine, Washington University School of Medicine, St. Louis, Missouri 63110, and the ^§ Center for Molecular Biology of Oral Diseases, University of Illinois, Chicago, Illinois 60612

Received for publication, October 24, 2001, and in revised form, November 27, 2001

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The substrate specificity of thrombin is regulated by binding of macromolecular substrates and effectors to exosites I and II. Exosites I and II have been reported to be extremely linked allosterically, such that binding of a ligand to one exosite results in near-total loss of affinity for ligands at the alternative exosite, whereas other studies support the independence of the interactions. An array of fluorescent thrombin derivatives and fluorescein-labeled hirudin^54-65 ([5F]Hir^54-65(SO)) were used as probes in quantitative equilibrium binding studies to resolve whether the affinities of the exosite I-specific ligands, Hir^54-65(SO) and fibrinogen, and of the exosite II-specific ligands, prothrombin fragment 2 and a monoclonal antibody, were affected by alternate exosite occupation. Hir^54-65(SO) and fibrinogen bound to exosite I with dissociation constants of 16-28 nM and 5-7 µM, respectively, which were changed 2-fold by fragment 2 binding. Native thrombin and four thrombin derivatives labeled with different probes bound fragment 2 and the antibody with dissociation constants of 3-12 µM and 1.8 nM, respectively, unaffected by Hir^54-65(SO). The results support a ternary complex binding model in which exosites I and II can be occupied simultaneously. The thrombin catalytic site senses individual and simultaneous binding of exosite I and II ligands differently, resulting in unique active site environments for each thrombin complex. The results indicate significant, ligand-specific allosteric coupling between thrombin exosites I and II and catalytic site perturbations but insignificant inter-exosite thermodynamic linkage.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The specificity of thrombin toward its procoagulant and anticoagulant physiological substrates is allosterically regulated by interactions of macromolecular substrates, inhibitors, and effectors with either of two electropositive sites, exosites I and II, in near-opposition on the enzyme surface (1, 2). Exosite I binds fibrinogen (Fbg)¹ (3), fibrin I and II (3, 4), the 12-residue carboxyl-terminal hirudin^54-65 sequence (5, 6), thrombomodulin (7), the thrombin receptor (8, 9), and an acidic sequence on the serpin, heparin cofactor II (10, 11). Exosite II binds heparin and other glycosaminoglycans (2, 12, 13), prothrombin activation fragment 2 (F2) (14), the chondroitin sulfate moiety of thrombomodulin (15, 16), the leech peptide hemadin (17), and an exosite II-specific human monoclonal antibody (18). Factors V (19-22), Va (21, 22), and VIII (19), platelet glycoprotein Ib (23-25), and the snake venom protein bothrojaracin (26) have been reported to interact with both exosites I and II.

Binding of exosite ligands to thrombin is correlated with significant changes in the kinetics of hydrolysis of peptide ester and peptide p-nitroanilide substrates (3, 7, 9, 18, 27-31) in addition to profound effects on specificity and reactivity toward its natural macromolecular substrates and inhibitors (10, 11, 15, 32-36). These studies indicate that exosite binding of allosteric effectors is coupled to conformational changes affecting the S1-S3 substrate specificity subsites in the thrombin catalytic site (37-39). Binding studies of F2, thrombomodulin, fibrin, and heparin with various active site-labeled thrombin derivatives in which the S1-S4 subsites were occupied (16, 34, 40), and studies of the effect of exosite ligand binding on the hydrolysis of tripeptide p-nitroanilide substrates suggest that structurally different ligands produce ligand-specific changes in the catalytic site. Extreme allosteric linkage between exosites I and II (30, 31, 41) has been reported to prevent simultaneous occupation of exosites I and II (30, 41), whereas other studies provide contrasting evidence for binary and ternary complex formation with similar affinities among thrombin and exosite I and II ligands (18). In favor of inter-exosite linkage, the dissociation constant for fluorescently labeled, Tyr⁶³-sulfated hirudin^53-64 and bovine thrombin was weakened 10-fold by F2 binding, although ternary complex formation was demonstrated (31). Extremely negative inter-exosite interactions were reported for Tyr⁶³-sulfated hirudin^54-65 (Hir^54-65(SO)) and human F2 or a synthetic peptide, F2^63-116 (30, 41). The latter studies were concluded to reflect mutually exclusive binding of F2 and Hir^54-65(SO) by reciprocal, allosteric modulation of ligand affinity between the two exosites (41). By contrast, binding of an exosite II-specific monoclonal antibody (mAb) did not affect detectably the conformation of thrombin exosite I or its affinity for [5F]Hir^54-65(SO) (18).

To understand the mechanism of exosite regulation of thrombin further, the present work was undertaken to resolve whether the affinities of the exosite I-specific ligands, Hir^54-65(SO) and fibrinogen, and of the exosite II-specific ligands, prothrombin fragment 2 and a monoclonal antibody, were affected by alternate exosite occupation. This was an important goal because studies employing hirudin peptides or F2 as probes of exosite involvement in other thrombin interactions could not be interpreted unambiguously, and it was uncertain whether the effects were due to competitive binding of alternate exosite I or II ligands or to extremely negative exosite linkage. Significant differences between human and bovine thrombin have been reported for the affinities of hirudin peptides (6). Also, the broad disagreement among reported affinities of F2 for bovine (42, 43) and human thrombin (30, 40) and its putative linkage to exosite I prompted a detailed quantitative analysis of F2 binding to human thrombin. Binding of Fbg to exosite I and the monoclonal antibody to exosite II were similarly characterized and quantitated for the first time in this context.

The results support the conclusion that binding of the exosite I and II ligands studied here conforms to a model with independent exosite interactions enabling formation of binary and ternary complexes with experimentally indistinguishable affinity. Structurally different exosite ligands produced different effects on the catalytic site, evidenced by different fluorescence changes of active site-labeled thrombins. Non-additive perturbations of the catalytic site accompany the exosite interactions, such that free thrombin and thrombin in binary and ternary complexes have unique properties. Analysis of experimental error in the dissociation constants for an array of ligands and a number of alternate experimental designs provided no evidence to substantiate extremely negative allosteric linkage between the exosites. It is concluded that the affinity of non-interacting, model exosite I and II ligands for thrombin remains unchanged whether the alternative exosite is occupied or not. Changes in binding affinity of factors V and Va, thrombomodulin, fibrin, and high molecular weight heparin affected by model exosite-specific ligands are likely to be due to competitive overlapping binding sites or additional interactions between the ligands themselves, but not to extreme inter-exosite allosteric linkage.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Purification and Characterization of Proteins and Peptides-- Human -thrombin, prepared by activation of prothrombin purified from human plasma (44), was 90% active as determined by active site titration (45). Prothrombin fragment 2 (F2) generated by cleavage of prethrombin 1 by factor Xa (40, 46), human fibrinogen (4, 33), anti- thrombin (47), and the monoclonal antibody (mAb) against thrombin exosite II (51) was purified by the published methods. Protein concentrations were determined at 280 nm with the absorption coefficients and molecular weights of 1.83 (mg/ml)¹ cm¹ in 0.1 M NaOH or 1.74 in buffer, and 36,700, thrombin (48); 1.25 and 12,900, F2 (49); 1.51 and 170,000, Fbg monomer (4); 0.65 and 58,000, antithrombin (50); 1.35 and 150,000, mAb against thrombin exosite II (51).

Fluorescent thrombin derivatives were prepared by stoichiometric incorporation of ATA-FFR-CH₂Cl or ATA-FPR-CH₂Cl into the active site, and labeling of the NH₂OH-generated free thiol with the fluorescence probes 5-(iodoacetamido)fluorescein (5-IAF), 6-(iodoacetamido)-fluorescein (6-IAF), 4'-{[(iodoacetyl)amino]methyl}fluorescein (4'-IAF), and 2-[(4'-iodoacetamido)-anilino]naphthalene-6-sulfonic acid (IAANS) (Molecular Probes) following published methods (40, 44, 52, 53). Fluorescent meizothrombin-des-F1 was prepared by incorporation of ATA-FPR-CH₂Cl during activation of prethrombin 1 by ecarin and subsequent labeling with IAANS (40). The concentration of nonsulfated Hir^54-65 and Hir^54-65(SO) (Sigma or Bachem) in water or reaction buffer was determined from the purity and peptide content specified by the manufacturer. [5F]Hir^54-65(SO) was prepared by labeling Hir^54-65 (SO) with 5-carboxy(fluorescein) as described previously (35).

Fluorescence Studies-- Fluorescence measurements were made with an SLM 8100 spectrofluorometer, using acrylic cuvettes coated with polyethylene glycol 20,000 except in tryptophan fluorescence experiments. Excitation and emission wavelengths are as follows: [5F]Hir^54-65 (SO), 491 and 520 nm, 4-8 nm bandpass; [5F]-, [6F]-, and [4'F]FPR-thrombin, 495 and 520 nm, 4-8 or 8-16 nm bandpass; [ANS]FFR-thrombin and [ANS]FPR-thrombin, 325 and 450 nm, 8-16 nm bandpass; and [5F]- and [6F]FPR-thrombin tryptophan fluorescence, 295 nm excitation and 360 nm emission, 4-8 nm bandpass. Titrations were performed by successive addition of small titrant volumes with 13% dilution and corrected for dilution and background. Individual fluorescence measurements were recorded after 5 min of equilibration and were averaged over 10-20 readings. Multiple titrations using overlapping titrant concentrations were combined to eliminate error propagation typical for multiple (>6) additions. Results were expressed as the fractional changes in the initial fluorescence ((F_obs  F_o)/F_o = F/F_o) as a function of total titrant concentration and were fit by the appropriate binding equation. Direct binding of F2 to [5F]FPR-thrombin, of competitive F2 binding to unlabeled thrombins, and thrombin inactivation by antithrombin was performed in two buffer systems as follows: 50 mM Hepes, 0.11 M NaCl, 5 mM CaCl₂, 1 mg/ml polyethylene glycol, pH 7.4; and the same buffer with 0.125 M NaCl and 1 mM EDTA. FPR-CH₂Cl (1 µM) was added to all titrations except those containing native thrombin.

Binding of F2 to Active Thrombin and Active Site-blocked, Unlabeled and Labeled Thrombin-- Fluorescence titrations of direct binding of F2 to [5F]- and [6F]FPR-thrombin were analyzed by the quadratic equation for binding of a single ligand (40), to obtain the maximum fluorescence intensity change (F_max/F_o) and the dissociation constant (K_D), with one binding site assumed on thrombin (n = 1). Binding of F2 to native thrombin, and the active site-blocked species ATA-FPR-thrombin and ATA-FFR-thrombin (0, 9, or 25 µM unlabeled thrombin), was measured in competitive binding experiments using [5F]FPR-thrombin (0.26 µM) as a probe. The dependence of F/F_o on the F2 concentration in the absence and presence of competing unlabeled thrombin were analyzed by least squares fitting of the cubic competitive binding equation defining the fractions [T*·F2]/(n[T*]_o) and [T·F2]/(n[T]_o) for competitive binding of F2 to labeled (T*) and unlabeled thrombin (T), as described previously (54, 55). The observed fluorescence change is given by the contribution of the T*·F2 complex, weighted by the maximum fluorescence change associated with its formation (Equation 1),

(Eq. 1)

in which [T*]_o is the total concentration of T*, and n is the number of equivalent and independent binding sites for F2. With an assumed 1:1 stoichiometry for the thrombin·F2 complex, the fitted parameters were F_max/F_o, and the dissociation constants were K_T*(F2) and K_T(F2).

Effect of Hir^54-65(SO) on Binding of F2 to Thrombin-- The properties of the thrombin derivatives [5F]FPR-T and [6F]FPR-T in reporting the interactions with F2 and Hir^54-65(SO) with unequal and opposite fluorescence changes were used for monitoring the joint interactions. In separate experiments, 50-100 nM [5F]FPR-thrombin or [6F]FPR-thrombin were titrated with F2, in the absence and presence of 20 µM Hir^54-65(SO). In complementary experiments, the labeled thrombins were also titrated with Hir^54-65(SO) in the absence and presence of 32.5 µM F2. In all of the experiments, the observed fluorescence change was given by Equation 2 for the ternary complex model (Scheme I).

View larger version (14K): [in this window] [in a new window]   Scheme I.

(Eq. 2)

[T*·H] is the T*·Hir^54-65(SO) complex; [T*·F2] is the T*·F2 complex, and [T*·F2·H] is the ternary complex, T*·F2·Hir^54-65(SO). The combined data sets for each labeled thrombin were fit by Equation 2, with the concentrations of the binary and the ternary complexes calculated by simultaneous solution of the expressions for the equilibrium constants defined by the model, and the mass conservation equations. The fitted parameters were the individual F_max/F_o values for the binary and ternary complexes, the dissociation constants for the binary complexes, K_T*(F2) and K_T*(H), and the dissociation constants, K_T*·F2(H) and K_T*·H(F2), for formation of the ternary complex. Because of the small fluorescence changes resulting from the interaction of Hir^54-65(SO) with [5F]FPR-thrombin and [6F]FPR-thrombin, binding was quantitated independently from the changes in tryptophan fluorescence of 100 nM labeled thrombin titrated with Hir^54-65(SO), and K_T*(H) was fixed at the determined value.

To determine the affinity of F2 for unlabeled thrombin species, 100 nM [5F]FPR-thrombin was titrated with F2 in the presence of 10 or 25 µM unlabeled thrombin. This was repeated in titrations with F2 at saturating Hir^54-65(SO) (50 µM). The competition binding data were fit by the cubic equation as described above (54, 55) to obtain F_max/F_o values for the labeled binary and ternary complexes and the dissociation constants for F2 binding to labeled and unlabeled thrombin, respectively, in their free and Hir^54-65(SO)-saturated forms.

Effect of F2 on Binding of Hir^54-65(SO) to Labeled and Native Thrombin-- [ANS]FFR-T and [4'F]FPR-T exhibited large fluorescence changes upon binding of Hir^54-65(SO), whereas F2 binding caused a significantly smaller effect which allowed the simultaneous interactions to be observed. [ANS]FFR-T (0.19 µM) was titrated with Hir^54-65 (SO), in the absence and presence of 32.5 µM F2, and in separate experiments was titrated with F2, in the presence of fixed concentrations of Hir^54-65(SO). [4'F]FPR-T (10 nM) was titrated similarly with Hir^54-65(SO), in the absence and presence of 36 µM F2, and with F2 in the absence and presence of 20 µM Hir^54-65(SO). The combined data sets for each labeled thrombin were fit by Equation 2 and the ternary complex model for binding of F2 and Hir^54-65(SO) to labeled thrombin (Scheme I). In competitive titrations with native thrombin, 10 nM [4'F]FPR-T was titrated with Hir^54-65(SO) in the absence and presence of 143 nM native thrombin. These titrations were repeated in the presence of 36 µM F2.

Effect of F2 on Binding of [5F]Hir^54-65(SO) to Thrombin-- [5F]Hir^54-65(SO) (50 nM) was titrated with native thrombin in the absence and presence of 25 µM F2, and the combined results from titrations with three separate F2 preparations were pooled. The observed fluorescence change for this situation was given by Equation 3,

(Eq. 3)

where H* represents [5F]Hir^54-65(SO); T·H* is the binary thrombin·[5F]Hir^54-65(SO) complex; T·F2·H* is the ternary thrombin·F2·[5F]Hir^54-65(SO) complex; [H*]_o is the total [5F]Hir^54-65(SO) concentration; and F_maxT(H*) and F_maxT·F2(H*) are the maximal relative fluorescence changes for [5F]Hir^54-65(SO) binding to thrombin and the T·F2 complex. The data were fit by Equation 3.

Binding of Fbg to Thrombin and Meizothrombin-des-F1; Effect of F2, Hir^54-65(SO), and Hir^54-65-- [ANS]FPR-T and [ANS]FPR-meizothrombin-des-F1 reported Fbg binding by a large fluorescence enhancement, whereas these probes were insensitive to binding of Hir^54-65, Hir^54-65(SO), and F2. [ANS]FPR-T (0.2 µM) was titrated with Fbg in the absence and presence of fixed concentrations of Hir^54-65(SO) or Hir^54-65. These titrations were repeated at 32.5 µM F2.

Similarly, [ANS]FPR-meizothrombin-des-F1 (0.2 µM) was titrated with Fbg in the absence and presence of fixed concentrations of Hir^54-65(SO). Background corrections were 15-30% at the highest protein concentrations. Under conditions of saturation of [ANS]FPR-T with Hir^54-65 or Hir^54-65(SO), and [ANS]FPR-meizothrombin-des-F1 with Hir^54-65(SO), an exosite I-independent increase in fluorescence was observed as a linear increase in fluorescence with Fbg concentration. In separate experiments, [ANS]FPR-T at near-saturation with 53 µM Fbg was titrated with the exosite I ligands Hir^54-65(SO) or Hir^54-65.

Binding of Fbg to [ANS]FPR-T or [ANS]FPR-meizothrombin-des-F1 and the effect of Hir^54-65(SO) and Hir^54-65 were described by a model (Scheme II) in which two ligands bind competitively to a fluorescent probe, and the interactions were accompanied by unequal fluorescence changes (21), in this case zero for Hir^54-65(SO) and Hir^54-65 binding. This model was used previously for analysis of competitive binding of Hir^54-65(SO) and factor V/Va to [ANS]FPR-T (21). A linear term in Fbg (protomer) concentration was included to account for the exosite I-independent fluorescence increase. The fluorescence change was described by Equation 4,

(Eq. 4)

Simultaneous least squares fitting of Equation 4 to the data with the cubic equations defining the fractional concentrations of the T*·Fbg complex, [T*·Fbg]/(n[T*]_o), and of the T*·hirudin peptide complex, [T*·H]/(n[T*]_o), gave the dissociation constants K_T*(Fbg) and K_T*(H) for the binary complexes, the maximum fluorescence change, and the slope of the exosite I-independent fluorescence increase (F_exo-ind/F_o) (Scheme II). The [ANS]FPR-meizothrombin-des-F1 titrations were analyzed using the same equations, in which T* was labeled meizothrombin-des-F1. The titration data at ~87% saturation of [ANS]FPR-T with F2 were analyzed similarly to obtain the dissociation constant for Fbg and Hir^54-65(SO) binding to the T*·F2 complex.

View larger version (9K): [in this window] [in a new window]   Scheme II.

Effect of Hir^54-65(SO) on Binding of an Anti-exosite II Antibody (mAb) to Thrombin-- [6F]FPR-T exhibited a large fluorescence quench upon binding of the exosite II-specific mAb, whereas it exhibited a modest enhancement upon binding of Hir^54-65(SO). Binding of the mAb to [6F]FPR-T was studied at probe concentrations of 1.5, 25, and 47 nM, and the dissociation constant and the number of independent thrombin binding sites (1/n) on the mAb were determined by simultaneous analysis using the quadratic binding equation (40). [6F]FPR-T was also titrated with antibody in the absence and the presence of 5 µM Hir^54-65(SO). The combined data were fit by Equation 2 and the ternary complex model for binding of mAb and Hir^54-65(SO) to thrombin (Scheme I).

Effect of F2 on the Kinetics of Thrombin Inactivation by Antithrombin-- The effect of F2 on thrombin inactivation by antithrombin was measured from the loss of thrombin chromogenic substrate activity, under pseudo first-order conditions ([AT]_o [T]_o) in the absence and presence of 5, 10, 15, and 25 µM F2. Residual thrombin ([T]_t) was expressed as the fraction of the initial activity ([T]_o). In a control reaction, F2 had no effect on the chromogenic assay rate. The progress curves of [T]_t/[T]_o with time were fit by a single exponential decay to obtain the observed pseudo first-order rate constants (k_obs) and simultaneously by Equation 5 and the quadratic binding equation, defining [T·F2]_o/(n[T]_o), to obtain the apparent second-order rate constants, k and k', respectively for free thrombin and the T·F2 complex reacting with antithrombin, and the K_D for the T·F2 complex,

(Eq. 5)

Least squares fitting was performed with SCIENTIST Software (MicroMath). All reported estimates of error represent ±2 S.D.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Binding of F2 to Native, Active Site-blocked, Unlabeled, and Labeled Thrombin-- The affinity of F2 for an array of fluorescence probe-labeled thrombins was determined previously (40), but the interaction of F2 with native human thrombin has not been characterized quantitatively. To assess the influence of active site labeling on the thrombin-F2 interaction, binding of F2 to human native thrombin, and active site-blocked nonfluorescent ATA-FPR-T and ATA-FFR-T, was characterized in competitive experiments using [5F]FPR-T as a probe. These studies were done in buffer containing 1 mM EDTA to allow comparison with previous results and subsequently in buffer containing 5 mM calcium. The [5F]FPR-T data were fit well by the cubic equation (54, 55) for competitive ligand (F2) binding to labeled and unlabeled thrombin. F2 bound to [5F]FPR-T with a K_D of 6 ± 1 µM, and to native human thrombin with a K_D of 5 ± 1 µM, a 1:1 binding stoichiometry, and a maximum fluorescence change of 15.4 ± 0.3%, as shown in Fig. 1A. The affinities for ATA-FPR-T and ATA-FFR-T were 3 ± 1 and 10 ± 2 µM, respectively. Experiments in buffer with 5 mM CaCl₂ reported indistinguishable F2 binding to labeled and native thrombin. Dissociation constants for ligand binding to active site-labeled thrombins with free and occupied alternate exosites are summarized in Table I; dissociation constants for ligand binding to native and active site-blocked, unlabeled thrombins are listed in Table II; maximal fluorescence changes for active site-labeled thrombins in binary and ternary complexes and for [5F]Hir^54-65(SO) binding to free thrombin and T·F2 complex are listed in Table III. Fig. 1B compares fluorescence changes and dissociation constants for F2 binding to the unlabeled thrombins with the parameters for the panel of fluorescent thrombins (40). The affinities of F2 for the peptide-inhibited, unlabeled thrombins fell within the range defined by F2 binding to the fluorescent derivatives. The dissociation constants for labeled and unlabeled FFR-thrombins were slightly but consistently higher (~2-fold) than those for FPR-thrombins, although within the joint experimental error. These results indicated a small but reproducible allosteric linkage effect between the affinity for F2 and the structure of the tripeptide occupying the active site.

View larger version (21K): [in this window] [in a new window]   Fig. 1.   Competitive titration of F2 binding to [5F]FPR-T and unlabeled thrombin. A, the fractional change in fluorescence of 0.26 µM [5F]FPR-T (F/Fo) is shown as a function of the total concentration of F2 ([F2]o) in the absence () and presence of 9 () and 25 µM () unlabeled thrombin. Experiments were performed in buffer with EDTA. The solid lines represent the least squares fits with the parameters listed in Tables I-III. B, comparison of KD and Fmax/Fo values for labeled and unlabeled thrombins. Values of Fmax/Fo are shown for F2 binding to fluorescently labeled FPR-thrombins (shaded ellipses) and FFR-thrombins (open ellipses) plotted against the KD (40). The fluorescence probes were 4'-{[(iodoacetyl)amino]methyl}fluorescein (4'-IAF), 5-(iodoacetamido)fluorescein (5-IAF), 6-(iodoacetamido)fluorescein (6-IAF), 2-[(4'-iodoacetamido)anilino]naphthalene-6-sulfonic acid (IAANS), tetramethylrhodamine-5-(and-6)-iodoacetamide (TMRIA), rhodamine X iodoacetamide (XRIA), and 7-diethylamino-3-[(4'-iodoacetylamino)phenyl]-4-methylcoumarin (DCIA). The experimental error in the parameters (±2 S.D.) defines the radii of the ellipses. The vertical dotted lines indicate the values for KD of fragment 2 binding to ATA-FPR-T (1), native thrombin (2), and ATA-FFR-T (3). Titrations were performed and analyzed as described under "Experimental Procedures."

View this table: [in this window] [in a new window]   Table I Dissociation constants for exosite I and II ligand binding to active-site-labeled thrombins with free and occupied alternate exosites Dissociation constants are listed from global analysis of titrations of the indicated active-site-labeled thrombins with exosite I (KD(exo I)) and II (KD(exo II)) ligands, in the absence and the presence of saturating alternate exosite ligand, determined as described under "Experimental Procedures." KT(H) for binding of Hir54-65(SC) to [5F]FPR-T was obtained by tryptophan fluorescence and to [6F]FPR-T by tryptophan (19 ± 8 nM) and fluorescein fluorescence (24 ± 5 nM) (see Fig. 2). KT(F2) for binding of F2 to [5F]FPR-T was obtained by competitive titration with unlabeled and labeled thrombin in EDTA buffer (6 ± 1 µM) and was indistinguishable from the value in buffer containing calcium (6 ± 3 µM) (see Fig. 1A). KT(Fbg) for [ANS]FPR-T was determined by competition with Hir54-65(SO) (7 ± 2 µM) (see Fig. 5B) and with Hir54-65 (5 ± 2 µM) (see Fig. 5A).

View this table: [in this window] [in a new window]   Table II Dissociation constants for binding of F2, Hir54-65(SO), and [5F]Hir54-65(SO) to native and unlabeled, active-site-blocked thrombins Dissociation constants (KD(exo I)) for [5F]Hir54-65(SO) binding to native thrombin were from global analysis of fluorescence titrations in the absence and the presence of saturating F2, as described under "Experimental Procedures." Binding constants of Hir54-65(SO) and F2 to native thrombin, and of F2 to unlabeled active-site-blocked thrombins (KD(exo II)) were determined by competitive titrations with fluorescent thrombins in the presence of EDTA or calcium. Binding of F2 to native thrombin was identical (5 ± 1 µM) in the absence and presence of EDTA. KT(F2) values for ATA-FPR-T and ATA-FFR-T were determined in buffer with EDTA.

View this table: [in this window] [in a new window]   Table III Maximal fluorescence changes of binary and ternary complexes of active-site-labeled thrombins with exosite I and II ligands Maximal fluorescence changes (Fmax/Fo) were from global analysis of fluorescence titrations of active-site-labeled thrombins with exosite I (exo I) and II (exo II) ligands, in the absence (binary complexes) and the presence (ternary complex) of saturating alternate exosite ligand, as described under "Experimental Procedures." Fluorescence changes for F2 binding to [5F]FPR-T in buffer with and without EDTA were 15.4 and 9%, respectively. Fluorescence changes for Fbg binding to [ANS]FPR-T were 102 and 91% from competition experiments with Hir54-65(SO) and Hir54-65, respectively. Exosite I-independent fluorescence changes caused by Fbg binding to [ANS]FPR-T in the presence of saturating Hir54-65(SO) were 0.012 ± 0.001 µM1 at 0 µM F2 and 0.007 ± 0.001 µM1 at 32 µM F2, and in the presence of saturating Hir54-65, 0.013 ± 0.002 µM1 at 0 µM F2. The exosite I-independent fluorescence change for [ANS]FPR-meizothrombin-des-F1 was 0.007 ± 0.001 µM1.

Effect of Hir^54-65(SO) on the Binding of F2 to Thrombin-- The effect of Hir^54-65(SO) on F2 binding to thrombin was studied with [5F]FPR-T and [6F]FPR-T. These probes reported Hir^54-65(SO) binding with small fluorescence increases of 5 and 9% and F2 binding with quenches of 9 and 22%, as shown in Fig. 2 for [6F]FPR-T. No reliable estimate of K_T*(H) for the [5F]FPR-T·Hir^54-65(SO) complex could be obtained from the fluorescein fluorescence data alone, due to the large experimental error in the small fluorescence change. However, binding of Hir^54-65(SO) to [5F]FPR-T and [6F]FPR-T was determined accurately by independent tryptophan fluorescence titrations with Hir^54-65(SO) (Fig. 2, inset, and Table I) which gave dissociation constants of 16 ± 12 and 19 ± 8 nM, respectively. These were fixed parameters in the global analysis of the fluorescein fluorescence data. Both thrombin probes were titrated with F2, in the absence and presence of saturating Hir^54-65(SO), and in separate experiments titrated with Hir^54-65(SO), in the absence and presence of ~85% saturating F2. The combined data for each labeled thrombin were fit simultaneously by Equation 2 for the ternary complex model (see Scheme I). In the absence of Hir^54-65(SO), the F2 affinities for [5F]FPR-T and [6F]FPR-T were indistinguishable at 6 ± 1 and 5 ± 1 µM, respectively. In the presence of saturating Hir^54-65(SO), the affinities were 3 ± 1 and 5 ± 1 µM for F2 binding to the [5F]FPR-T·Hir^54-65(SO) and [6F]FPR-T·Hir^54-65 (SO) complexes, respectively. In the absence of F2, K_T*(H) for the [6F]FPR-T·Hir^54-65(SO) complex was 24 ± 5 nM, and at near-saturating F2, K_T*·F2(H) for Hir^54-65(SO) binding to the [6F]FPR-T·F2 complex was an indistinguishable 16 ± 4 nM. Binding of F2 to the labeled thrombins was affected no more than ~2-fold by simultaneous occupation of the alternate exosite with Hir^54-65(SO), and binding of Hir^54-65(SO) to free and F2-bound thrombin was indistinguishable. The fluorescence amplitudes for individual and simultaneous binding of Hir^54-65(SO) and F2 were approximately additive for these two probes (Table III).

View larger version (22K): [in this window] [in a new window]   Fig. 2.   Binding of F2 and Hir54-65(SO) to [6F]FPR-T. The fractional change in fluorescence of 50 nM [6F]FPR-T (F/Fo) is shown as a function of the total concentration of F2 ([F2]o) in the absence () and presence () of 20 µM Hir54-65(SO). The inset shows titration of 50 nM [6F]FPR-T measured by fluorescein fluorescence as a function of the total concentration of Hir54-65(SO) ([Hir54-65(SO)]o) in the absence () and the presence () of 32.5 µM F2, and titrations of 100 nM [6F]FPR-T tryptophan fluorescence () in the absence of F2. Experiments were performed in buffer containing 5 mM CaCl2. The solid lines represent the least squares fits with the parameters listed in Tables I and III. Titrations were performed and analyzed as described under "Experimental Procedures."

Binding of F2 to native thrombin and its complex with Hir^54-65 (SO) was quantitated in competition experiments with [5F]FPR-T as a probe. Binding of F2 to native thrombin was characterized by a K_D of 5 ± 1 µM, and to the thrombin·Hir^54-65(SO) complex by a K_D of 2 ± 1 µM. In these competition experiments, F2 bound to the [5F]FPR-T·Hir^54-65 (SO) complex with a K_D of 3 ± 1 µM. These affinities were indistinguishable from the K_T*·H(F2) of 3 µM, obtained by direct titration of the [5F]FPR-T·Hir^54-65(SO) complex with F2. These results strongly indicated that at saturating Hir^54-65(SO), F2 bound to native thrombin and [5F]FPR-T in a similar fashion.

Effect of F2 on the Binding of Hir^54-65(SO) to Labeled and Native Thrombin-- [ANS]FFR-T exhibited a 66% quench upon binding of Hir^54-65(SO), and reported F2 binding by a 31% quench, dissimilar signals that were used to examine the effect of F2 on binding of Hir^54-65(SO). Fig. 3A shows the combined results for Hir^54-65(SO) and F2 binding to [ANS]FFR-T and the fit by the ternary complex model (Equation 2). Binding of Hir^54-65(SO) to free [ANS]FFR-T and the [ANS]FFR-T·F2 complex was characterized by indistinguishable dissociation constants of 27 ± 4 and 54 ± 28 nM, respectively, whereas F2 bound to free [ANS]FFR-T and the [ANS]FFR-T·Hir^54-65(SO) complex with dissociation constants of 12 ± 4 and 25 ± 15 µM, respectively. Maximum fluorescence changes for separate and simultaneous binding of Hir^54-65(SO) and F2 were non-additive (Table III), indicating that the thrombin active site sensed separate and simultaneous exosite binding differently, but this was not linked to a significant change in affinity.

View larger version (18K): [in this window] [in a new window]   Fig. 3.   Binding of Hir54-65(SO) and fragment 2 to [ANS]FFR-T. A, the fractional change in fluorescence of 190 nM [ANS]FFR-T (F/Fo) is shown as a function of the total concentration of Hir54-65(SO) ([Hir54-65 (SO)]o) in the absence () and presence () of 32.5 µM F2. The inset shows F/Fo as a function of the total concentration of F2 ([F2]o), in the absence () and presence of 0.15 µM (*) and 20 µM () Hir54-65(SO). Experiments were performed in buffer containing EDTA. The solid lines represent the least squares fits with the parameters listed in Tables I and III. B, competitive titration of [4'F]FPR-T and native thrombin with Hir54-65(SO). F/Fo of 10 nM [4'F]FPR-T is shown as a function of the total concentration of [Hir54-65(SO)]o in the absence () and presence () of 0.143 µM native thrombin in buffer with 5 mM CaCl2. The solid lines represent the least squares fit with the parameters listed in Tables I-III. C, competitive titration of [4'F]FPR-T and native thrombin with Hir54-65(SO) and effect of F2. F/Fo of 10 nM [4'F]FPR-T is shown as a function of [Hir54-65(SO)]o in the absence () and presence () of 0.143 µM native thrombin and in the presence of 36.6 µM F2. The inset shows F/Fo as a function of [F2]o in the absence () and presence of 20 µM () Hir54-65(SO). Experiments were performed in buffer with 5 mM CaCl2. The solid lines represent the least squares fits with the parameters listed in Tables I-III. All titrations were performed and analyzed as described under "Experimental Procedures."

Similar experiments with [4'F]FPR-T reported Hir^54-65(SO) binding with a 50% enhancement (Fig. 3, B and C). Titrations of [4'F]FPR-T and competing native thrombin with Hir^54-65 (SO) were done in the absence and presence of ~88% saturating F2. The inset (Fig. 3C) shows titrations of [4'F]FPR-T with F2, in the absence and presence of saturating Hir^54-65 (SO). The global fit demonstrated binding of Hir^54-65(SO) to free [4'F]FPR-T and the [4'F]FPR-T·F2 complex with indistinguishable dissociation constants of 150 ± 16 nM and 114 ± 28 nM, respectively. F2 bound to free [4'F]FPR-T and the [4'F]FPR-T·Hir^54-65(SO) complex with affinities of 5 ± 2 and 6 ± 4 µM. Fluorescence amplitudes for individual and simultaneous binding of Hir^54-65(SO) and F2 were also non-additive (Table III). Competitive binding of Hir^54-65(SO) to native thrombin and [4'F]FPR-T was characterized by dissociation constants of 28 ± 14 and 150 ± 16 nM. Hir^54-65(SO) binding to the respective thrombin·F2 complexes was indistinguishable, with affinities of 23 ± 18 nM for the native thrombin·F2 complex and 117 ± 22 nM for the [4'F]FPR-T·F2 complex. The affinity of Hir^54-65(SO) for [4'F]FPR-T was reduced ~5-fold compared with [5F]-, [6F]-, and [ANS]-labeled thrombins, which were very similar to the affinity for native thrombin. This was the largest linkage effect observed between the catalytic site probe and the affinity of exosite I for Hir^54-65(SO). However, binding of the complementary exosite ligand affected minimally the affinity of Hir^54-65(SO) and F2 for [ANS]FFR-T, [4'F]FPR-T, and native thrombin.

Effect of F2 on Binding of [5F]Hir^54-65(SO) to Native Thrombin-- Titration of [5F]Hir^54-65(SO) with native thrombin, in the absence and the presence of near-saturating F2, gave near-identical fluorescence decreases as shown in Fig. 4. The combined data were fit by the ternary complex model (Scheme I) with K_T(F2) fixed at 5 µM, the mean value for F2 binding to native thrombin. Binding of [5F]Hir^54-65(SO) to active thrombin and the thrombin·F2 complex were characterized by indistinguishable binding constants of 18 ± 3 and 20 ± 7 nM, respectively. The fluorescence change was unaffected by binding of F2 that was devoid of traces of contaminating prethrombin 2 by SDS-gel electrophoresis. Equivalent affinities of [5F]Hir^54-65(SO) and Hir^54-65(SO) for human thrombin have been demonstrated previously (6). Hence, the almost identical results for [5F]Hir^54-65(SO) binding to thrombin and the thrombin·F2 complex found here were a reflection of the similar affinities of Hir^54-65(SO) binding. Moreover, they were in excellent agreement with the data for Hir^54-65(SO) binding to unlabeled thrombin and thrombin·F2 complex, determined with [4'F]FPR-T as a probe as described above.

View larger version (16K): [in this window] [in a new window]   Fig. 4.   Effect of F2 on binding of native thrombin to [5F]Hir54-65 (SO). The fractional change in fluorescence of 50 nM [5F] Hir54-65(SO) (F/Fo) is shown as a function of the total concentration of native thrombin ([Thrombin]o) in the absence () and presence () of 25 µM F2. Experiments were performed in buffer with CaCl2. The solid lines represent the least squares fits with the dissociation constants listed in Table II. Titrations were performed and analyzed as described under "Experimental Procedures."

Binding of Fbg to Thrombin and Meizothrombin-des-F1; Effect of F2, Hir^54-65(SO), and Hir^54-65-- Kinetic studies of the pathway of Fbg cleavage by thrombin have shown that cleavage is inhibited competitively by hirudin peptides (3). This reflects the predominant role of exosite I in mediating productive Fbg binding as a substrate and in determining the K_m of 7.5 µM (4). The ligand-selective reporting properties of [ANS]FPR-T and [ANS]FPR-meizothrombin-des-F1 were used to investigate Fbg binding directly for the first time. Fbg binding to active site-blocked [ANS]FPR-T, in competition with Hir^54-65(SO) and Hir^54-65, respectively, was signaled by maximal fluorescence enhancements of 102 ± 10 and 91 ± 13% (Fig. 5A), whereas Fbg binding to [ANS]FPR-meizothrombin-des-F1 gave 173 ± 12% (Fig. 5C). Binding of F2, Hir^54-65, and Hir^54-65(SO) to [ANS]FPR-T and [ANS]FPR-meizothrombin-des-F1 did not result in detectable fluorescence changes. [ANS]FPR-T was titrated with Fbg in the absence and presence of Hir^54-65(SO) or Hir^54-65 (Fig. 5A) and in complementary titrations in the presence of ~83% saturating F2 (Fig. 5B). [ANS]FPR-meizothrombin-des-F1 was titrated with Fbg in the absence and presence of Hir^54-65(SO) (Fig. 5C). The peptides progressively decreased Fbg binding, revealing the exosite I-dependent interaction. At saturating Hir^54-65(SO) or Hir^54-65, titration with Fbg of [ANS]FPR-T and [ANS]FPR-meizothrombin-des-F1 demonstrated an additional exosite I-independent fluorescence change, as indicated by a linear fluorescence increase with Fbg concentration. In the competitive binding experiments, [ANS]FPR-T in the presence of ~89% saturating Fbg was titrated with Hir^54-65(SO) and with Hir^54-65. Analysis of the data by Equation 4 showed that Fbg bound to [ANS]FPR-T in the exosite I-mediated interaction with a dissociation constant of 7 ± 2 µM, calculated from the data set with competing Hir^54-65(SO), and the indistinguishable value of 5 ± 2 µM, calculated from the data set with competing Hir^54-65. Fbg bound to the [ANS]FPR-T·F2 complex with a binding constant of 13 ± 4 µM, indicating a possible ~2-fold effect of F2 on the affinity for Fbg. In the presence of Fbg, the affinities of Hir^54-65 (SO) for [ANS]FPR-T and the [ANS]FPR-T·F2 complex were indistinguishable, 17 ± 7 and 19 ± 7 nM, respectively. These studies demonstrated that when the active site of thrombin was blocked by the tripeptide label, Fbg bound exosite I with an affinity equivalent to the K_m of Fbg for native thrombin (7.5 µM) (4). [ANS]FPR-meizothrombin-des-F1 bound Fbg with a dissociation constant of 25 ± 3 µM and Hir^54-65(SO) with a dissociation constant of 27 ± 4 nM. These results indicated that meizothrombin-des-F1 was capable of binding Fbg through exosite I, in agreement with crystallographic studies showing that the Fbg recognition site is accessible on meizothrombin-des-F1 (56).

View larger version (16K): [in this window] [in a new window]   Fig. 5.   Titrations of [ANS]FPR-T with Fbg and Hir54-65. A, the fractional change in fluorescence of 200 nM [ANS]FPR-T (