Formation of a Ternary Complex between Thrombin, Fibrin … · 2001-06-11 · Philip J. Hogg and Craig M. Jackson* From the American Red Crass Blood Services, Southeastern Michigan

THE JOURNAL OF BIOLOC~CAL CHEMISTRY 0 1990 by The American Society for Biochemistry and Molecular Biology, Inc.

Vol. 265, No. 1, Issue of January 5. pp. 248-255.1990 Printed in U.S.A.

Formation of a Ternary Complex between Thrombin, Fibrin Monomer, and Heparin Influences the Action of Thrombin on Its Substrates*

(Received for publication, June I, 1989)

Philip J. Hogg and Craig M. Jackson* From the American Red Crass Blood Services, Southeastern Michigan Region, Detroit, Michigan 48232

The consequences of the combined effects of fibrin II monomer (FnIIm) and heparin (H) on the hydrolysis of peptidyl p-nitroanilide substrates by thrombin (IIa), the cleavage of prothrombin by thrombin and the thrombin-catalyzed release of fibrinopeptides from fibrinogen have been studied at pH 7.4 and I 0.15. The effects of fibrin II monomer and heparin on chromogenic substrate hydrolysis can be described by a hyperbolic mixed inhibition model in which substrate can interact with four possible enzyme species (IIa, IIa*H, IIaaFnIIm, and IIa*FnIIm*H) that arise as a result of random formation of a ternary complex among thrombin, fibrin II monomer, and heparin (Hogg, P. J. and Jackson, C. M. (1990) J. Biol. Chem. 265, 241-247). The formation of the ternary IIa*FnIIm.H complex results in an increase in the K,,, values of 7.03 f 1.17- fold (1.37-9.65 pM) and 1.94 2 0.60-fold (38.1-73.9 pM) for H-D-Ile-Pro-Arg-pNA and Cbz-Gly-Pro-Arg- pNA hydrolysis, respectively, and a decrease in the k, values of 0.45 f 0.0%fold (49.5-22.3 s-‘) and 0.52 + 0.05-fold (93.1-48.4 s-l). Fibrin II monomer and heparin in combination also decrease the efficiency (kc/ K,) with which thrombin cleaves prothrombin to pro- duce Fragment 1 and Prethrombin 1 by 2.3-fold from 607 f 30 to 264 f 13 M-’ s-l. In contrast to the effects of fibrin II monomer and heparin on thrombin hydrolysis of chromogenic substrates, its proteolysis of prothrombin and its inactivation by antithrombin III (Hogg, P. J., and Jackson, C. M. (1989) Proc. N&l. Acad. Sci. U. S. A. 86,3619-3623), these components have no discernible influence on the ability of thrombin to cleave fibrinogen. These observations indicate that the substrate specificity of thrombin is altered when it is bound in a complex with fibrin II monomer and heparin and suggest that the catalytic efficiency of thrombin for its physiological substrates will be affected differentially by these interactions. Such ternary complex formation involving thrombin, fibrin II monomer, and heparin may provide a mechanism for selectively regulating thrombin action.

It is generally believed that spatial constraint of the he- mostatic response is primarily the result of the actions of cofactors and modifiers which enhance localized thrombin formation and action and oppose reactions of inhibitors which

* This work was supported by an Established Investigatorship from the American National Red Cross (to C. M. J.) and the Southeastern Michigan Region Blood Services, American Red Cross. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “aduer- tisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

$ To whom requests for reprints should be addressed.

inactivate thrombin (1). This report describes the modulation of some kinetic properties of thrombin by two of these “reg- ulatory factors”: 1) heparin, a complex glycosaminoglycan which has been isolated from a variety of natural sources including the vascular endothelium (2, 3); and 2) fibrin, the product of thrombin action on fibrinogen (4,5).

Heparin and fibrin bind to sites on thrombin distinct from the Michaelis binding interactions with its substrates (6-13) and alter the kinetic properties of this enzyme. The interaction of thrombin with fibrin has been suggested to be a type of antithrombotic mechanism (antithrombin I), whereby the capturing of thrombin by the clot reduces thrombin action on circulating fibrinogen (6). Alteration in the binding of thrombin to fibrin has been associated with thrombosis in some cases of dysfibrinogenemia (14, 15). The ability of heparin to enhance the inactivation of coagulation proteinases has been established (2,16-N) and has proven clinical value (19). From studies on the kinetic mechanism of heparin action it has been shown that the interaction of heparin with thrombin, in addition to its specific interaction with antithrombin III, is a prerequisite for the efficient action of this glycosaminoglycan (18).

Apart from the individual effects of fibrin and heparin on the catalytic properties of thrombin, there are substantial and different alterations in the actions of this proteinase when these two components are combined. Studies of tbrombin binding to fibrin II polymer as a function of heparin and fibrin II polymer concentrations (20) have established the formation of a thrombin-fibrin II polymer-heparin ternary complex. The demonstration of the formation of this ternary complex complements kinetic evidence indicating formation of an analogous ternary complex with fibrin II monomer (21). The incorporation of thrombin into a complex with heparin and fibrin II monomer can explain the >300-fold reduction in the rate of inactivation of thrombin by beparin-antithrombin III in the presence of fibrin II monomer (21). This study characterizes the properties of thrombin in the thrombin- fibrin II monomer-heparin ternary complex from measurements of the effects of fibrin II monomer and beparin on: 1) the hydrolysis of two peptidyl p-nitroanilide substrates; 2) thrombin cleavage of prothrombin; and 3) the release of fibrinopeptides A and B from fibrinogen. The implications of these results for regulation of thrombin action in uiuo are predicted from the relative abilities of fibrin II monomer and heparin or fibrin polymer and heparin to bind thrombin in a ternary complex.

MATERIALS AND METHODS

Proteins-Human a-thrombin was supplied by Dr. John Fenton of the New York State Department of Health, Albany, NY. The active enzyme concentration was determined by active site titration (22).

Human prothrombin, prepared by the method of Braze and Mile-

248

by guest on March 20, 2020

http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

Fibrin and Heparin Influence Thrombin Substrate Specificity 249

tich (23), was a gift from Dr. Paul Bock of the American Red Cross, Detroit, MI. A molecular weight of 71,600 and a E%,, value of 14.7 was used to calculate the concentration of prothrombin (24).

Fibrinogen was prepared from human fresh frozen plasma by a modification of the method of Jakohsen and Kierulf (25) as described by Hogg and Jackson (21). Fibrin II monomer was prepared as described by Hogg and Jackson (21). A molecular weight of 340,000 and E%,, values of 15.1 and 14.0 were used to calculate the concentrations of fibrinogen and fibrin II monomer, respectively (4, 5). Fibrin II was prevented from polymerizing by the presence of the tetrapeptide Gly-Pro-Arg-Pro at concentrations of 6.7-11.3 mM (26).

Heparin-Heparin with high affinity for antithrombin III (fraction D, Sache et al. (27)) was provided by Dr. Edgar Sache of the Institute Choay, Paris. Its specific activity was determined by the producer as 211 USP units/ma. Its concentration was calculated on a weight basis using a molecular-weight of 20,300.

Assay of Chromogenic Substrate Hydrolysis bv Thrombin-Chro- ” ” _ mogenic substrate ieactions were performed in l-ml polystyrene cuvettes (Sarstedt) in 50 mM HEPES, 0.125 M NaCl, 1 mg/ml PEG 6000, pH 7.4, I 0.15 buffer at 25 ‘C. The release of p-nitroaniline that resulted from hydrolysis of the peptide p-nitroanilide substrates was followed by measuring the increase in absorbance at 405 nm with a Cary 219 recording spectrophotometer. The data were captured and analyzed by software developed in this laboratory.’ The kinetic parameters for chromogenic substrate hydrolysis were obtained from nonlinear least-squares analyses of full progress curves (28). The data were best fit by allowing for competitive product inhibition. The K,,,, K,, and k, values for H-rr-Ile-Pro-Arg-pNA2 (S2288, KabiVitrum) and Cbz-Gly-Pro-Arg-pNA (Cbz-Chromozym TH, Boehringer Mannheim GmbH) in the pH 7.4 buffer were determined to be 1.37 + 0.39 FM, 5.08 + 1.51 pM, and 49.5 + 0.3 s-i and 38.1 + 1.1 jiM, 124 f 7 pM, and 93.1 k 0.8 s-i, respectively. The concentrations of substrate were determined spectrophotometrically at 342 nm in distilled water using an absorption coefficient of 8210 M-’ cm-i (29).

Assay of Thrombin Cleavage of Prothrombin-Reactions were performed in 50 mM HEPES, 0.125 M NaCl, 1.5 mM CaC12, 1 mg/ml PEG 6000, pH 7.4, buffer at 25 “C. Thrombin (0.20 pM) was added to prothrombin (1.44 or 6.07 pM) to start the reaction which was then stopped at discrete time intervals by the thrombin inhibitor D-Phe- Pro-Arg-CH&l (PPACK, Calbiochem, 84 PM). Prothrombin and Prothrombin Fragment 1 were separated and quantified by anion- exchange HPLC at room temperature with a Varian 5500 chromatograph using a Pharmacia Mono Q HR 5/5 column (5 X 50 mm) and a O.l-ml (6.07 FM prothrombin) or 0.5-ml (1.44 fiM prothrombin) sample loop. A combination NaCl and CaCb gradient and step elution system was employed. Buffers A, B, and C were 50 mM Tris/HCl, 1 mg/ml PEG 6000, pH 7.5, containing 0.1 M NaCl (Buffer A), 2.0 M NaCl (Buffer B), or 0.1 M NaCl and 0.2 M CaC& (Buffer C). The elution protocol was: O-3 min. 100% A: 3.1-8 min. 92% A/8% B: 8.1- 9 min, iOO% A; l-3 min, 90% A/l% B/9% C; 19 din, 80% A/10% B/ 10% C; 19.1-23 min, 100% B. The flow rate was 1.0 ml/min throughout, and the peptides were detected by measuring the absorbance at 280 nm. Fragment 1 eluted at 15 min (15 ml) as a doublet and prothrombin at 19 min (19 ml). The weights of the Fragment 1 and prothrombin peaks, cut out of photocopies of the strip chart recordings, were related to concentrations of these proteins from standard curves of weights of peaks versus Fragment 1 and prothrombin concentrations. These curves were linear over the prothrombin concentration range employed.

Assay of Fibrinopeptide Release from Fibrinogen-Reactions were performed in 1.5-ml Microfuge tubes (Beckman) in 50 mM HEPES, 0.125 M NaCl, 1 ma/ml PEG 6000, DH 7.4. I 0.15 buffer at 37 “C in total volumes ‘of 1 ml. Thrombin (6.301 nM) was added to fibrinogen (0.406 pM) to start the reaction which was then stopped at discrete time intervals by the addition of the synthetic thrombin inhibitor D-

’ C. M. Jackson and J. K. Labanowski, unpublished work. * The abbreviations used are: pNA, p-nitroanilide; HEPES, N-(2-

hydroxyethyl)piperazine-N’-ethanesulfonic acid; PEG 6000, polyeth- ylene glycol8000 (J. T. Baker Chemical Co., Cleveland, OH); S2288, H-n-Ile-Pro-Arg-p-nitroanilide; Cbz, benzyloxycarbonyl; Cbz-Chro- mozym TH, Cbz-Gly-Pro-Arg-p-nitroanilide; II, human prothrombin; IIa, human u-thrombin; [AaB&], concentration of Aa chains (twice the concentration of fibrinogen); [aB&], concentration of (Y chains: FPA and FPB, fibrinopeptides A-and B; FnIIm, fibrin II monomer; fibrin from which both A and B fibrinonentides have been released: H, M, 20,300 heparin with high affinity ‘for antithrombin III; HPLC; high performance liquid chromatography.

Phe-Pro-Arg-CH,Cl (23 PM), as described by Hofsteenge et al. (31). Fibrinopeptides A and B were separated and quantified by reverse- phase HPLC at room temperature with a Varian 5500 chromatograph using a Varian MicroPak MCH-N-Cap-5 Cl8 silica column (4 x 150 mm) and a 0.5-ml sample loop. The gradient system was a modification of that used by Hofsteenge et al. (31). Buffers A and B were 20 mM sodium phosphate, pH 5.8, containing 2 and 25% (v/v) acetoni- trile (J. T. Baker Chemical Co., Cleveland, OH), respectively. The elution protocol was: 0 min. 100% A; 5 min. 80% A/20% B: 15 min. 75% A/25% B; 20 min, 60% A/40% ‘B; 25 min, 30% A/70% B. The flow rate was 1.0 ml/min throughout, and the peptides were detected by measuring the absorbance at 214 nm. Fibrinopeptide concentrations were calculated by relating the weights of the fibrinopeptide peaks, cut out of photocopies of the strip chart recordings, to standard curves of weights of peaks versus fibrinopeptide A and B concentrations calculated from reactions in which different fibrinogen concentrations (O-O.71 jtM) were hydrolyzed exhaustively by thrombin (25 nM). These curves were linear over the fibrinogen concentration range employed.

RESULTS

Effects of Fibrin II Monomer and Heparin on the Hydrolysis of Chromogenic Substrates by Thrombin-The effects of heparin and fibrin II monomer on the hydrolysis of chromogenic substrates by thrombin can be described by the kinetic model depicted in Scheme 1. Anticipating the results shown subse- quently, this model indicates that fibrin II monomer and heparin alter the efficiency with which thrombin can hydro- lyse peptidyl p-nitroanilide substrates by binding thrombin in a ternary complex with fibrin II monomer and heparin. No effects of either of the binary complexes on the kinetic behav- ior of thrombin are indicated in Scheme 1; consistent with the data given below. Altered efficiency can result from changes in either the K,,, or the k, for substrate hydrolysis, or both.

Scheme IA and the left- and right-hand faces of the box in Scheme 1B describe the formation of the three possible binary and the final ternary complexes in equilibrium mixtures of thrombin, fibrin II monomer, and heparin. From the results of binding studies with fibrin II polymer (20), the binary dissociation constants for the thrombin-heparin, K,I~.H, and thrombin-fibrin II monomer, K iia F”ii,,,, interactions are taken to be 15 and 301 nM, respectively. The adequacy of the extrapolation of these constants from the results with fibrin II polymer to these studies with fibrin II monomer will be addressed in the discussion. A value of 5.7 pM for the dissociation constant for the fibrin II monomer-heparin interaction, KFnIIm.H, is derived from quantitative affinity chromat- ographic studies of this interaction (21). The ternary dissociation constant for the IIa. H intermediate binary complex. FnIIm interaction, Ktlla.~), F,,~I~, is calculated from the results of the effects of heparin and fibrin II monomer on the initial velocity of hydrolysis of two peptidylp-nitroanilide substrates by thrombin (Figs. 1 and 2).

Fig. IA shows the decrease in the initial velocity of S2288 hydrolysis at a constant concentration of heparin, 48.5 nM,

as a function of fibrin II monomer concentration (O-4.33 pM)

for three different concentrations of substrate. Fig. 1C demonstrates the effects of these two reactants, over the same concentrations ranges, on the K,,, for hydrolysis of this substrate. Fig. 1B shows the decrease in the initial velocity of hydrolysis of S2288 at a constant concentration of fibrin II monomer, 1.59 PM, as a function of heparin concentration (O- 97 nM) for three different S2288 concentrations. Similarly, Fig. 1D demonstrates the effects of these two reactants, over the same concentrations ranges, on the K,,, for hydrolysis of S2288. When the data from Figs. 1, A and B, are fitted by nonlinear regression to Equation 1 with cy, & and KcIla.Hj, FnIIp as the unknown parameters (see Scheme 1 and Miniprint


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

250 Fibrin and Heparin Influence Thrombin Substrate Specificity

A 1la.H

K / K~,akh Fdh+f G Ila K

Fn*llm m.=Mrn

Ila.Fnllm

Ila.Fnllm

SCHEME 1. Model for the effects of fibrin II monomer and heparin on the hydrolysis of chromogenic substrates by thrombin. In this model substrate can interact with four possible enzyme species (IIa, IIa. H, IIa .FnIIm, and IIa. FnIIm. H) that arise as a result of random ternary complex formation among thrombin, fibrin II monomer, and heparin (A and the /&- and rig&-hand faces of the box in B). (Y and p represent the extent of the changes in the Km and k, values for substrate hydrolysis, respectively, as a result of thrombin being bound in a complex with fibrin II monomer and heparin. The model assumes that peptide p-nitroanilide substrate binding to thrombin does not affect the interactions of thrombin with fibrin II monomer and heparin. KII~.H, KII~.F~II~, and KF~II,,.H are the dissociation constants governing the binary complex formation between IIa and H, IIa and FnIIm, and FnIIm and H, respectively. KIL+HL ~nnrn, K~II~.F~~I,,,), n, and K,MJ,,,,H), in are the ternary dissociation constants governing the complex formation between 1Ia.H binary complex and FnIIm, IIa.FnIIm binary complex and H, and FnIIm. H binary complex and IIa, respectively.

Supplement3 for explanation of the symbols), using the values for the binary dissociation constants indicated above, estimates of 7.03 f 1.17 for (Y, 0.45 f 0.08 for /3, and 37.8 f 2.8 nM for KoI,.nJ,~niim are obtained. The solid lines in these figures are the calculated best-fit lines using these kinetic parameters and Equation 1. Heparin alone (O-525 nM), or fibrin II monomer alone (O-5.2 FM) have only small (<9%) effects on the K,,, and k, values for this substrate.

To determine whether fibrin II monomer and heparin influence the hydrolysis of chromogenic substrates other than S2288, the effects of fibrin II monomer and heparin on the

’ “Model for the Effects of Heparin and Fibrin II Monomer on the Hydrolysis of Chromogenic Substrates by Thrombin” is presented in miniprint at the end of this paper. Miniprint is easily read with the aid of a standard magnifying glass. Full size photocopies are included in the microfilm edition of the Journal that is available from Waverly Press.

200 . . n

H

0 0

100 8 . a

0- 012246

[Fibrin Ill, uM [Heparinl, nM

C 10

2

0 1-

012245

[Fibrin Ill, uM

aooy- 1

200 m

n

b

0 ml

100 0

.

8

2

0 0 40 80

[Hoparid, nM

FIG. 1. A, effects of varying fibrin II monomer concentration (O- 4.33 PM) on the initial velocity of S2288 hydrolysis ([IIa]r = 6.07 nM) at a constant concentration of heparin, 48.5 nM, at three different concentrations of S2288, l 5.0 PM; 0, 12.0 pM; a, 49.8 PM. The free concentrations of fibrin II monomer are calculated using Equation 2. The solid lines are calculated from the global least-squares fit of the kinetic data to Equation 1 (39). 8, effects of varying heparin concentration (O-97 nM) on the initial velocity of S2288 hydrolysis ([IIa]r = 6.07 nM) at a constant concentration of fibrin II monomer, 1.59 /IM, at three different concentrations of S2288: 0, 4.87 PM; 0, 9.75 PM; n , 48.7 PM. The free concentrations of heparin are calculated using Equation 2. The solid lines are calculated from the global least- squares fit of the kinetic data to Equation 1 (39). The average of the results of Fig. 1, A and B, give estimates of 7.03 2 1.17 for 01, 0.45 + 0.08 for 0, and 37.8 f. 2.8 nM for Kc~~a.~),~n~,m. Values for the other parameters in Equation 1 are given in the text. C, effects of the same reactant concentrations indicated in A on the Km for S2288 hydrolysis. The solid line is calculated from Equation 1 using the values reported in the text. D, effects of the same reactant concentrations indicated in Fig. 1B on the Km for S2288 hydrolysis. The solid line is calculated from Equation 1 using the values reported in the text.

hydrolysis of Cbz-Chromozym TH by thrombin was also studied. Thrombin hydrolyzes Cbz-Chromozym TH with a 15-fold lower efficiency (kc/K,,, of 2.44 X lo6 M-’ s-’ for Cbz- Chromozym TH versus 3.61 X lo7 M-’ s-’ for S2288), which is mostly a result of a 28-fold higher Km for Cbz Chromozym TH (38.1 FM) than for S2288 (1.37 PM). Fig. 2A reports the decrease in the initial velocity of hydrolysis of Cbz-Chromo- zym TH at a constant concentration of heparin, 105 nM, as a function of fibrin II monomer concentration (O-O.82 pM) for three different concentrations of substrate. Similarly, Fig. 2B shows the decrease in the initial velocity of hydrolysis of this substrate at a constant concentration of fibrin II monomer, 0.49 pM, as a function of heparin concentration (O-105 nM) for the same three concentrations of Cbz-Chromozym TH. When the data from Fig. 2, A and B, are fitted by nonlinear


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

Fibrin and Heparin Influence Thrombin Substrate Specificity

0.0 0.5 1.0 0 50 100

[Fibrin II], uM iHepuh1, nM

FIG. 2. A, effects of varying fibrin II monomer concentration (O- 0.82 PM) on the initial velocity of Cbz-Chromozym TH hydrolysis ([IIa]T = 2.47 nM) at a constant concentration of heparin, 105 nM, at three different concentrations of Cbz-Chromozym TH: 0, 9.86 pM; 0,49.5 pM; n , 98.6 PM. The free concentrations of fibrin II monomer are calculated using Equation 2. The solid lines are calculated from the global least-squares fit of the kinetic data to Equation 1 (39). B, effects of varying heparin concentration (O-105 nM) on the initial velocity of Cbz-Chromozym TH hydrolysis ([IIalT = 2.40 nM) at a constant concentration of fibrin II monomer, 0.49 NM, at three different concentrations of Cbz-Chromozym TH: 0, 9.86 pM; 0, 49.5 PM; W, 98.6 j.tM. The free concentrations of heparin are calculated using Equation 2. The solid lines are calculated from the global least-squares fit of the kinetic data to Equation 1 (39). The average of the results of A and B give estimates of 1.94 rt 0.06 for (Y, 0.52 f 0.05 for 0, and 37.8 + 4.7 nM for K~IIa,nj,rniim. Values for the other parameters in Equation 1 are given in the text.

regression to Equation 1 (Miniprint Supplement), using the values for the binary dissociation constants indicated above, estimates of 1.94 f 0.60 for (Y, 0.52 f 0.05 for /3, and 37.8 f 4.7 nM for Kola.~),FnIIm are obtained. The solid lines in the figures are calculated using these kinetic parameters and Equation 1. Heparin alone (O-150 nM), or fibrin II monomer alone (O-l.40 pM) have only small (<14%) effects on the K,,, and k, values for this substrate.

Because of the linked equilibria of the interactions of the left- and right-hand faces of the box in Scheme 1, the following equality applies,

&a.H&Ia.H), FnIlm = K m m**m K ,Ila.FnIIm~. H - - K~ntlrn.~K(~n~~rn.~,.~,a

Using an average value from the S2288 and Cbz-Chromozym TH kinetic data of 37.8 f 3.7 nM for Kala.H),FnII,,, (see above) estimates of 1.88 and 0.10 nM for the ternary KaI,.F,,IIrnj,H and K~F,,II,,,.H), na dissociation constants are calculated from the kinetic data. To ensure that the method of preparation of fibrin II monomer in these studies has not introduced an artifact that might be responsible for the effects reported here, an experiment was conducted looking at the effects of fibrin produced in situ in the presence of heparin (O-100 nM) on the ability of thrombin (2.51 nM) to hydrolyze Cbz-Chro- mozym TH (67 PM). Fig. 3 demonstrates the progressive decrease in the velocity of substrate hydrolysis as a result of the production of fibrin from fibrinogen (1.22 PM). Fibrinogen and the chromogenic substrate compete for thrombin in the assay. The assay conditions were such that approximately 10% of both substrates, Cbz-Chromozym TH and fibrinogen, were hydrolyzed by the end of the reaction. The solid lines were calculated from the least-squares fit of the data to a second-order polynomial. A more extensive analysis of the data was not persued owing to the complexity of the parallel kinetic processes involved. The results are concordant with the conclusions derived from the data reported in Fig. 2 for Cbz-Chromozym TH and argue that fibrin produced in situ is

251

0 30 60 90 120

Time (aed

FIG. 3. Effect of fibrin produced in situ on the hydrolysis of Cbz-Chromozym TH (67 pM) by thrombin (2.51 nM). The reactions were carried out in the absence (0) or presence (0, W, 0, A, A) of 1.22 pM fibrinogen and 0 nM (O), 10.0 nM (B), 40.0 nM (O), 60.0 nM (A), and 99.9 nM (A) heparin. All assays contained 8.0 mM GPRP. The solid lines represent the linear least-squares fit of the data to a second-order polynomial.

0 2 4 6 5 10

Time (hours)

mma hhl TIma Wd

FIG. 4. A, progress curves for the disappearance of prothrombin (0) and the generation of Fragment 1 (0) from 6.07 pM prothrombin by 0.20 pM thrombin. The data for prothrombin loss and Fragment 1 formation has been fitted by least-squares regression to Equations 3 and 4, respectively (solid lines). This analysis gave an average value of 1.22 + 0.06 x lo-’ s-l for k,,. B and C, the effects of 0.47 pM fibrin II monomer (A), 0.20 fiM heparin (m), and fibrin II monomer and heparin (same concentrations) (U), on the initial velocity of Fragment 1 formation from two different concentrations of prothrombin (1.44 pM, B, and 6.07 PM, C) by 0.20 pM thrombin. The control data, i.e. no fibrin II monomer or heparin, are also plotted for comparison (0). The solid lines have been calculated from Equation 4 with values of 1.22 X low4 s-i (O), 0.92 x 10e4 s-i (m), and 0.52 x lo-’ s-’ (0) for k,,.

behaving similarly to the fibrin II produced and added exog- neously.

Effects of Fibrin II Monomer and Heparin on the Cleavage of Prothrombin by Thrombin-Thrombin cleaves prothrombin at Arg155-Ser’56 generating Prothrombin Fragment 1 and Prethrombin 1 (30). Fig. 4A shows the progress curves for the disappearance of prothrombin (6.07 pM) and the formation of Fragment 1 by thrombin (0.20 PM). The data have been fitted by nonlinear least squares regression to Equations 3 and 4


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


0.0 I7 0 10 20 SO

Time (mid

FIG. 5. Progress curves for the release of FPA (circles) and FPB (squares) in the absence (0, n ) and presence (0, Cl) of 99.8 nM heparin. from 0.406 NM fibrinogen by 0.301 nM thrombin. The data for FPA and. FPB release in the absence of heparin has been fitted by nonlinear regression to Equations 5 and 6, respectively (solid lines). This analysis gave values of 0.82 + 0.10 PM for [AaBP],, 3.99 + 0.27 x 10m3 s-l for kFpA and 2.23 + 0.08 x 10m3 SC’ for kFPB.

(Miniprint Supplement), respectively. This analysis gave an average value of 607 f 30 M-’ s-’ for k,,,/K,,,,,. Fig. 4, B and C, shows the effects of 0.47 pM fibrin II monomer (triangles), 0.20 PM heparin (closed squares), and fibrin II monomer and heparin (same concentrations) (open squares), on the initial velocity of Fragment 1 formation from two different concentrations of prothrombin (1.44 PM, Fig. 4B, and 6.07 PM, Fig. 4C) by 0.20 PM thrombin. Fibrin II monomer alone, at this concentration, has no discernible effect on the proteolysis reaction, whereas heparin decreases the efficiency (Iz,,,/K,,,) 1.3-fold, to 467 M-’ s-l. In contrast to these small effects, fibrin II monomer and heparin in combination decrease iz,,,/K,,,,, 2.3-fold, to 264 M-’ s-l. The solid lines were calculated from the best fit of the data to Equation 4.

Effects of Heparin on the Release of Fibrinopeptides from Fibrinogen by Thrombin-Fig. 5 shows the progress curves for the release of FPA and FPB, from 0.406 j.tM fibrinogen by 0.301 nM thrombin under pseudo first-order conditions (see Miniprint Supplement), in the absence and presence of 99.8 nM heparin. The data for FPA and FPB release in the absence of heparin have been fitted by nonlinear regression to Equa- tions 5 and 6, respectively (32). This analysis gave values of 13.3 f 0.9 X 10’ M-’ s-l for keppA/K,,,ppA and 7.4 f 0.3 X lo6 M-’ s-’ for kcFpB/KmFpB, which are in good agreement with previously determined estimates of these parameters (31,32).

DISCUSSION

The affinity of the interactions between thrombin, fibrin polymer, and heparin and the prevalence of these reactants in uivo, particularly at an injury site, suggests that these ternary complex interactions may be involved in regulating the partition of thrombin between fibrin-bound and soluble phases during hemostasis and a thrombosis (20). These and other4 findings indicate that thrombin in a complex with fibrin II monomer and heparin also possesses altered reactivity toward some of its substrates and inhibitors, suggesting that, in addition to regulating its partition, these ternary complex interactions may also be involved in regulating the kinetic actions of this proteinase. A parallel for these proposed effects in blood clotting is seen with the binding of thrombin to thrombomodulin, an endothelial cell intrinsic membrane glycoprotein that binds thrombin and regulates its substrate

4 Studies of the combined effects of heparin and fibrin II monomer on the inhibition of thrombin by hirudin indicate an approximately lOOO-fold increase in the dissociation constant for this interaction, from 15 fM to -10 PM, at saturating concentrations of heparin and fibrin II monomer (manuscript in preparation).

specificity (see Ref. 30 for a recent review). Demonstration of ternary complex formation between

thrombin, heparin, and fibrin II polymer (20) provided the insight for the kinetic model that has enabled the quantitative analysis of kinetic studies investigating the effects of heparin and fibrin II monomer on the catalytic actions of thrombin. By keeping fibrin II monomeric in solution by the tetrapeptide Gly-Pro-Arg-Pro the complication of fibrin polymerization is avoided and it becomes possible to observe modulation of thrombin action by fibrin independent of thrombin entrap- ment within the fibrin gel (21). The quantitative consequences of these two reactants, heparin and fibrin II monomer, on thrombin action are most easily understood from their effects on the hydrolysis of small synthetic chromogenic substrates.

The effects of fibrin II monomer and heparin on the hydrolysis of peptidyl p-nitroanilide substrates by thrombin can be described by the rapid equilibrium, hyperbolic mixed inhibition model shown in Scheme 1. In this model thrombin is bound in a ternary complex with fibrin II monomer and heparin which arises as a result of random formation of intermediate binary complexes between the components. The thrombin in this complex possesses reduced reactivity toward chromogenic substrates, which is reflected as both an increase in the K,,,, 7.03-fold for S2288 and 1.94-fold for Cbz-Chro- mozym TH, and a decrease in kC, 0.45-fold for S2288 and 0.52- fold for Cbz-Chromozym TH, for substrate hydrolysis (Figs. 1 and 2). An interesting aspect of these phenomena is that these ternary complex interactions have differentially affected the ability of thrombin to hydrolyze the two substrates. The efficiency (kc/K,) with which thrombin can hydrolyze S2288 is reduced 16-fold when thrombin is bound in a ternary complex with fibrin II monomer and heparin, in contrast to the 4-fold decrease in kc/K,,, for hydrolysis of Cbz-Chromozym TH. This analysis has also enabled the evaluation of the dissociation constants governing ternary complex formation. The estimate of the dissociation constant for the interaction between the thrombin-heparin binary complex and fibrin II monomer, KCIIe.Hj, Fnllm, from the experiments with S2288,37.8 f 2.8 nM (Fig. l), is in excellent agreement with the value obtained using Cbz-Chromozym TH as the substrate, 37.8 f 4.7 nM (Fig. 2). The close agreement is important in that it further establishes the adequacy of the kinetic model proposed here for the effects of fibrin II monomer and heparin on the hydrolysis of chromogenic substrates by thrombin (Scheme 1). This dissociation constant from kinetic measurements is equivalent within the expected uncertainty to the dissociation constant calculated from binding studies for the same interaction with fibrin II polymer, 47 + 9 nM (20).

The dissociation constant for the binary thrombin-fibrin II monomer interactions used in these studies, 301 nM, has been extrapolated from binding experiments with fibrin polymer (20). Although this value is subject to uncertainty, results of kinetic experiments characterizing the effect of fibrin II monomer on the inactivation of thrombin by antithrombin III (Kd of 750 + 660 nM; see Fig. 1B of Ref. 21) and the adequacy of the fit of the chromogenic substrate data reported here using this value, suggests that the actual dissociation constant for the thrombin-fibrin II monomer interaction is within the same order of magnitude of the dissociation constant for the thrombin-fibrin polymer interaction. Halving or doubling the Kd for the thrombin-fibrin II monomer interaction when fitting the data of Figs. 1 and 2 resulted in statistically poorer analyses, further supporting the adequacy of the estimate for this dissociation constant.

Fibrin II monomer and heparin in combination also alter the efficiency with which thrombin cleaves its macromolecu-


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

Fibrin and Heparin Influence Thrombin Substrate Specificity 253

% f ? L

[Fibrin1 uM tFibrin1 UM

FIG. 6. A, simulation of the fraction of thrombin bound ([IIa]r = 1 nM) in a ternary complex with heparin ([H]r = 10 nM) and either fibrin II monomer (---) or fibrin II polymer (- - - -) as a function of fibrin concentration. The lines are calculated from Equations 2 and 7 using the dissociation constants from the body of this text for fibrin II monomer and from Scheme 2 of Ref. 20 for fibrin II nolvmer. B, simulation of the weight fraction of fibrin monomer (- - I), kbrin dimer (- - -) and fibrin trimer (. . .) as a function of fibrin concentration. The lines are calculated from Eauations 8 and 9 assuming a dissociation constant of 64 nM for the interaction of fibrin I monomer with a fibrin protofibril (5). Inset, abscissa range; 50 nM fibrin.

lar protein substrates. These two reactants decrease the kc/ K,,, for the release of Fragment 1 from prothrombin at least 2.3-fold in the presence of 1.5 mM Ca*+ (Fig. 4). Because the rate of cleavage of prothrombin by thrombin is dramatically reduced in the presence of 1.5 mM Ca*+ ions,* experiments were performed in the presence of this metal ion so as to make prothrombin a more “physiological” substrate for thrombin. Neither hydrolysis of chromogenic substrate by thrombin, cleavage of fibrinogen by thrombin, or the inhibition of thrombin by heparin-antithrombin III (21) was influ- enced by 1.5 mM Ca2+ ions.

In contrast to the effects of fibrin II monomer and heparin on the hydrolysis of chromogenic substrates by thrombin, its cleavage of prothrombin, and its inactivation by antithrombin III (21), these two reactants have no discernible influence on the ability of thrombin to cleave fibrinogen (Fig. 5). Thrombin inhibition by hirudin is also altered in the ternary complex, the dissociation constant for hirudin binding to thrombin is increased more than 1000-fold.5 These effects of fibrin and heparin on thrombin activity have important implications for the regulation of thrombin action in that it suggests that the catalytic efficiency of thrombin for its physiological substrates will be differentially affected by these ternary complex interactions, a finding consistent with the results of the effects of fibrin II monomer and heparin on the hydrolysis of peptidyl p-nitroanilide substrates by thrombin.

Comparison of the binary and ternary dissociation constants characterizing the formation of the complex of thrombin and heparin with fibrin polymer (see Scheme 2 of Ref. 20) and fibrin II monomer (see above) indicate that thrombin is more tightly bound in the complex with fibrin II monomer than it is with fibrin polymer. This is readily seen in Fig. 6A which illustrates the fraction of thrombin bound in a ternary complex with heparin and either fibrin II monomer (solid line) or fibrin polymer (dashed line) in the presence of 10 nM

heparin as a function of fibrin (monomer or polymer) concentration. This simulation, calculated from Equation 7 with values for the dissociation constants for fibrin II monomer from the body of this text or from Scheme 2 of Ref. 20 for fibrin II polymer, demonstrates that more thrombin will be bound to fibrin monomer than fibrin polymer in the presence

5 P. J. Hogg, P. E. Bock, and C. M. Jackson, unpublished work.

of heparin at any given concentration of fibrin. This effect is mostly a consequence of the difference in affinity of heparin for fibrin polymer (Kd of 0.28 pM, Ref. 20) uersus fibrin monomer (Kd of 5.7 pM, Ref. 21). The higher affinity of heparin for fibrin polymer acts to destabilize ternary complex formation at high fibrin polymer concentrations because of the formation of noninteracting IIa. H and FnIIp . H binary complexes (see inset of Fig. IA and Ref. 20). These findings imply that the effects of heparin and fibrin on the kinetic actions of thrombin will diminish as fibrin polymerizes. Therefore, it is informative to estimate what the concentrations of fibrin monomer might be during the initial stages of clot formation. Using the theory developed by Flory (34) for polymerization of a bifunctional monomer, the size distribu- tion of fibrin protofibrils as a function of fibrin concentration can be estimated (see Miniprint Supplement). Fig. 6B repre- sents the weight fraction of fibrin monomer (- - -), fibrin dimer (- - -), and fibrin trimer (. . . . .) as a function of total fibrin concentration. This analysis suggests that at the initial stages of fibrinogen hydrolysis there will be significant concentrations of fibrin monomer. Therefore, the incorporation of thrombin into a ternary complex with fibrin monomer during the initial stages of clot formation may regulate the balance between the action of thrombin on its substrates and its inactivation by inhibitors (21). The results further suggest that the regulation will be temporal, that is, it will vary with time due to the decreased ability of the polymerizing fibrin to bind thrombin in a complex with heparin. It is important to realize the assumptions inherent in this analysis. There is evidence suggesting that at the early stages of fibrinogen hydrolysis the formed fibrin will be predominantly bound in weak complexes with fibrinogen (35,36). Similarly, due to the slower release of fibrinopeptide B from fibrinogen, the initial fibrin formed will be mostly fibrin I (5, 32). These studies have not discriminated between the effects of fibrin and fibrin-fibrinogen complexes and have not directly assessed potential differences between fibrin I and fibrin II. The results of Fig. 3, however, where the fibrin formed in the reaction will be predominantly fibrin I, imply that fibrin I, like fibrin II, can also support complex formation with thrombin and heparin. The question as to whether fibrin-fibrinogen complexes will also bind thrombin and heparin with similar affin- ities as fibrin II monomer will require further work to answer.

An analogy for the effects of fibrin and heparin demonstrated here exists for the components of the fibrinolytic system. Fears (37) has shown that the activation of plasminogen by tissue plasminogen activator is enhanced by heparin, albeit to a small extent, and that this enhancement is negated when plasminogen is bound to fibrin. This parallels the effects of fibrin on the heparin-catalyzed inactivation of thrombin by antithrombin III. We found that the binding of thrombin to fibrin dramatically decreases the ability of heparin to enhance the inactivation of thrombin by antithrombin III (21). Another example of the influence of the thrombin- glycosaminoglycan interaction on thrombin function, in addition to the one demonstrated here, relates to the observations by Bourin et al. (38) who propose that the binding of thrombin to thrombomodulin involves the co-ordinated interaction of thrombin with a sulfated glycosaminoglycan cova- lently attached to thrombomodulin. Removal of this polysac- charide does not demonstrably affect the binding of thrombin but it does abolish the inhibitory effect of thrombomodulin on the hydrolysis of fibrinogen by thrombin and the ability of thrombomodulin to enhance the inactivation of thrombin by antithrombin III.

Acknowledgments-We thank Dr. Edgar Sache for the gift of


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


heparin, Dr. Paul Bock for the gift of prothrombin, and Dr. John Fenton for the gift of human thrombin. The assistance of Jan Berger in preparing the manuscript is greatly appreciated.

1. 2. 3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13. 14.

15.

16. 17.

18.

REFERENCES

Fenton, J. W., II (1981) Ann. N. Y. Acad. Sci. 370,468-495 Rosenberg, R. D. (1977) Fed. Proc. 36, lo-18 Marcum, J. A., and Rosenberg, R. D. (1987) Semin. Thromb.

Hemostasis 13,464-474 Doolittle, R. F. (1987) in Huemostasis and Thrombosis (Bloom,

A. T., and Thomas, D. P., eds) 2nd Ed., pp. 192-215, Churchill Livingstone, London

Lewis, S. D., Shields, P. P., and Shafer, J. A. (1985) J. Biol. Chem. 260, 10192-10199

Seegers, W. H., Nieft, M., and Loomis, E. C. (1945) Science 101, 520-521

Liu, C. Y., Nossel, H. L., and Kaplan, K. L. (1979) J. Biol. Chem. 254,10421-10425

Liu, C. Y., Kaplan, K. L., Markowitz, A. H., and Nossel, H. L. (1980) J. Biol. Chem. 255, 7627-7630

Kaminski, M., and McDonagh, J. (1983) J. Biol. Chem. 258, 10530-10535

Kaminski, M., and McDonagh, J. (1987) Biochem. J. 242,881- 887

Berliner, L. J., Sugawara, Y., and Fenton, J. W., II (1985) Bio- chemistry 24, 7005-7009

Fenton, J. W., II, Olson, T. A., Zabinski, M. P., and Wilner, G. D. (1988) Biochemistry 27, 7106-7112

Vali, Z., and Scheraga, H. A. (1988) Biochemistry 27, 1956-1963 Liu, C. Y., Nossel, H. L., and Kaplan, K. L. (1979) Thromb.

Haemostasis 42, 79 Haverkate, F., Koopman, J., Kluft, C., D’Angelo, A., Cattaneo,

M., and Manucci, P. M. (1986) Thromb. Haemostasis 55,131 Bjork, I., and Lindahl, U. (1982) Mol. Cell. Biochem. 48,161-182 Hoylaerts, M., Owen, W. G., and Collen, D. (1984) J. Biol. Chem.

259,5670-5677 Olson, S. T. (1988) J. Biol. Chem. 263, 1698-1708

19.

20.

21.

22. 23. 24.

25. 26.

27.

28.

29.

30.

31.

32.

33. 34. 35.

36. 37. 38.

39.

40.

Barrowcliffe, T. W., and Thomas, D. P. (1987) in Haemostasis and Thrombosis (Bloom, A. T., and Thomas, D. P., eds) 2nd Ed., pp. 849-869, Churchill Livingstone, London

Hogg, P. J., and Jackson, C. M. (1990) J. Biol. Chem. 265, 241- 247

Hogg, P. J., and Jackson, C. M. (1989) Proc. N&Z. Acad. Sci. U. S. A. 86, 3619-3623

Chase, T., and Shaw, E. (1969) Biochemistry 8,2212-2224 Broze, G., and Miletich, J. (1984) J. Clin. Znuest. 73, 933-938 Mann, K. G., Elion, J., Butkowski, R. J., Downing, M., and

Nesheim, M. E. (1981) Methods Enzymol. 80, 286-311 Jakobsen, E., and Kierulf, P. (1973) Thromb. Res. 3, 145-159 Laudano, A. P., and Doolittle, R. F. (1980) Biochemistry 19,

1013-1019 Sache, E., Maillard, M., Bertrand, H., Maman, M., Kunz, M.,

Choay, J., Fareed, J., and Messmore, H. (1982) Thromb. Res. 25,443-458

Duggleby, R. D., and Morrison, J. F. (1977) Biochim. Biophys. Actu 481, 297-312

Lottenberg, R., and Jackson, C. M. (1983) Biochim. Biophys. Acta 481,297-312

Jackson, C. M. (1987) in Haemostasis and Thrombosis (Bloom, A. T., and Thomas, D. P., eds) 2nd Ed., pp. 165-191, Churchill Livingstone, London

Hofsteenge, J., Taguchi, H., and Stone, S. R. (1986) Biochem. J. 237,243-251

Higgins, D. L., Lewis, S. D., and Shafer, J. A. (1983) J. Biol. Chem. 258,9276-9282

Esmon, C. T. (1989) J. Biol. f-km. 264,4743-4746 Flory, P. J. (1940) J. Am. Chem. Sot. 58,1877-1885 Brass, E. P., Forman, W. B., Edwards, R. V., and Lindan, 0.

(1976) Thromb. Hemostasis 36, 36-48 Wilf, J., and Minton, A. P. (1986) Biochemistry 25,3124-3133 Fears, R. (1988) Biochem. J. 249, 77-81 Bourin, M-C., Ghlin, A-K., Lane, D. A., Stenflo, J., and Lindahl,

U. (1988) J. Biol. Chem. 263,8044-8052 London, W. P., and Steck, T. L. (1969) Biochemistry 8, 1767-

1779 Duggleby, R. G. (1984) Comput. Biol. Med. 14,447-455

If cx~crimcnta are pcrfonncd under mndit~on~ whcrs [H, >> [,,a] and [F”,,m] >> [,,a,, the total c”“Es”lrall”“s of hsparin. [H, T, and fib”” II m”n”mcr, [F”Ih”] ,, Will relate to the free m”ce”tratl”ns Of these reactants accordxng to Quano” 2.


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

Fibrin and Heparin Influence Thrombin Substrate Specificity


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

P J Hogg and C M Jacksoninfluences the action of thrombin on its substrates.

Formation of a ternary complex between thrombin, fibrin monomer, and heparin

1990, 265:248-255.J. Biol. Chem.

http://www.jbc.org/content/265/1/248Access the most updated version of this article at

Alerts:

When a correction for this article is posted•

When this article is cited•

to choose from all of JBC's e-mail alertsClick here

http://www.jbc.org/content/265/1/248.full.html#ref-list-1

This article cites 0 references, 0 of which can be accessed free at


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/content/265/1/248

http://www.jbc.org/cgi/alerts?alertType=citedby&addAlert=cited_by&cited_by_criteria_resid=jbc;265/1/248&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/265/1/248

http://www.jbc.org/cgi/alerts?alertType=correction&addAlert=correction&correction_criteria_value=265/1/248&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/265/1/248

http://www.jbc.org/cgi/alerts/etoc

http://www.jbc.org/content/265/1/248.full.html#ref-list-1

http://www.jbc.org/

Documents

Formation of a Ternary Complex between Thrombin, Fibrin … · 2001-06-11 · Philip J. Hogg and Craig M. Jackson* From the American Red Crass Blood Services, Southeastern Michigan