Activation of human prothrombin by human prothrombinase. Influence ...

VOl. 262, No. 7, Issue of March 5, pp. 3291-3299,1987 Printed in U.S.A.

THEJOURNAL OF BIOLOGICAL CHEMISTRY (01987 by The American Society of Biological Chemists, Inc.

Activation of Human Prothrombinby Human Prothrombinase INFLUENCEOFFACTOR

Va ON T H E REACTIONMECHANISM* (Received for publication, October 29, 1986)

Sriram Krishnaswamy, William R. Church, Michael E. NesheimS, and KennethG . Manne From the Department of Biochemistry, University of Vermont, Burlington, Vermont 05405 and the $Departments of Medicine and Biochemistry, Queens University, Kingston, Ontario, K7L-3N6

The kineticsof the activationof human prothrombin The activationof human prothrombin to the serine protease catalyzed by human prothrombinase was studied usinga-thrombin isaccomplished by the proteolytic cleavage of two the fluorescent a-thrombin inhibitor dansylarginine- peptide bondsin the zymogen (2-6). Prothrombin is converted N-(3-ethyl-1,5-pentanediyl)amide(DAPA). Prothrom- to a-thrombin by the serine protease factor Xa (2-4). Alto a- though factor Xa binaseproteolytically activatesprothrombin possesses the catalyticmachinery to produce thrombin by cleavages at Arg273-Thr274 (bond A) and these cleavages in the zymogen, therate of prothrombin Arg322-Ile323(bond B). The differential fluorescence activation catalyzed by factor Xa alone is negligible when properties of DAPA complexed with the intermediates comparedtotherate observed when the fully assembled and products of human prothrombin activation were prothrombinase complex is the catalyst (7-9). The prothromexploited to study the kinetics of the individual bond binase complex is composed of the protease factor Xa and the cleavages in the zymogen. When the catalyst wascom- protein cofactor factor Va, assembled on a cellular (10-13) or posed of prothrombinase(humanfactor Xa, human factor Va, synthetic phospholipid vesicles, and calcium a phospholipid surface (7, 14) in the presence of calcium ion ion), initial velocity studies of a-thrombin formation and probably constitutes thephysiologically relevant catalyst indicated that the kinetic constants for cleavage the of for prothrombin activation. The available data indicate that the bovine prothrombinase bonds A or B were similar to the constants that were obtained for the overall reaction (bonds A + B). The complex is composed of factor Xa and factor Va complexed progress of the reaction was also monitored by poly- ina 1:l stoichiometry on a membrane surface containing acrylamide gel electrophoresis in the presence of so- negatively charged phospholipids in the presence of calcium ions and catalyzes the activation of prothrombin by an ordium dodecyl sulfate. The results indicated that the activation of human prothrombin catalyzed by pro- dered sequential kineticmechanism (15). Very little direct thrombinase proceeded exclusively via the formation information isavailable regarding the human prothrombinase of meizothrombin (bond B-cleaved) as an intermediate. complex reconstituted in. vitro usingpurified components Kinetic studies of the cofactor dependence of the rates isolated from human plasma. The present study was underof cleavage of the individual bonds indicated that, in taken to evaluate rigorously the kinetic properties of human the absence of the cofactor, cleavage at bond B would prothrombin activation catalyzed by human prothrombinase constitute the rate-limiting step in prothrombin acti- under relatively well-defined conditions. vation. Progress curves for prothrombin activation Factor Xa catalyzes the activation of human prothrombin catalyzed by prothrombinase andmonitored using the by proteolytic cleavages at ArgZ73-Thr274 and at Arp2-Ile323 fluorophore DAPA were typified by the appearanceof (6). The polypeptide chain composed of residues 1 through a transient maximum,indicatingtheformation of meizothrombin as an intermediate. When factor Xa 273 (Fragment 1.2) hasa molecular weight of 34,500 and is released as an activationpeptide. Residues 274-322 comprise alone was the catalyst, progress curves were characterized by an initial burstphase, suggesting the rapid the A chain of a-thrombin which is disulfide-linked to the B production of prethrombin 2 (bond A-cleaved) followed chain of thrombin (residues 323-581). As two cleavages are by its slow conversion to a-thrombin. Gel electropho- required in the prothrombin molecule to produce thrombin, resis followed by autoradiography wasused to confirm the overall activation of prothrombin can be represented by these results. Collectively,the results indicate that thetwo kinetic pathways illustrated in Fig. 1. A single cleavage activation of human prothrombin via the formationof at Arg273-Thr274(Reaction 1) yields Fragment 1.2 and an meizothrombin as an intermediateis a consequence of intermediate species known as prethrombin 2. Prethrombin 2 the association of the cofactor,human factor Va, with is composed of the A and B chains of thrombin with the bond the enzyme,human factor Xa, on the phospholipid at Arg322-Ile323intact. Further proteolysis at this bond (Reacsurface. tion 3) converts prethrombin2 to a-thrombin. In contrast,a single cleavage in the prothrombin molecule at Arg322-Ile323 (Reaction 2) producesaspeciesknown as meizothrombin. Meizothrombin is a catalytically activeintermediate (16) com* This work was supported by National Institutes of Health Grants posed of fragment 1.2-thrombin A chain and the B chain of HL35058 and 34575, Medical Research Council Grants DRG 309, thrombin linked by a disulfide bond. Further proteolysis of 310,and 311 (to M. E. N.), and theBiekell Foundation (to M. E. N.). meizothrombin by factor Xa (cleavage at Arg273-Ile274;ReacA preliminary accountof this work was presented at the 76th annual tion 4) yields the products Fragment 1.2 and the two-chain meeting of the American Society of Biological Chemists, Washington, form of a-thrombin. D. C., June 8-12, 1986, Abstract 913 (1). In additionto the proteolytic cleavages in the prothrombin $ To whom correspondence shouldbe addressed Dept. of Biochemistry, University of Vermont, Health Science Complex, Given Bldg., molecule catalyzed by factor Xa, thrombinformed during the Burlington, VT 05405-0068. course of the reaction is capable of catalyzing two cleavages

3291

3292

Kinetics of Human Prothrombin Activation ?

?

-

increased catalytic efficiency of the enzymeconferred by virtue of its association with the cofactor (20, 21). The experiments described in this paper were undertaken in order to provide a quantitative description of the kinetics of human prothrombin activation catalyzed by prothrombinase and to examine the consequences of complex formation on the kinetic mechanism of the reaction catalyzed by factor Xa.

tIA

-1 F R A M N T 1.2

PRETMOIUBlrl 2

FRAGMENT 1.2

l

i

"

7

MIZOTCROIVBN

EXPERIMENTALPROCEDURES

k-7

THROMBIN

FIG. 1. Schematic illustration of the pathways for the activation of human prothrombin.Cleavage of prothrombin at Ar$i3ThrZi4(Reaction I ) yields Fragment 1 2 and prethrombin 2. Further cleavage of prethrombin 2 a t Arg323-Ser3z4 (Reaction 3 ) yields a disulfide-linked two-chain form of a-thrombin. Cleavage of prothrombin in the oppositeorder (cleavage a t Arg3z3-Ser3z4; Reaction 4 ) yields meizothrombin. Meizothrombin is composed of the Fragment 1.2-A chain and thrombin-B chain that are covalently linked by a disulfide bond. Further cleavage a t Arg2i3-Th174 produces Fragment 1.2 and a-thrombin.The arrows labeled a and b indicatebonds that are susceptible to cleavage by the feedback action of thrombin. a = Arg155b = Arg2=-ThrZai.

indicated by arrows a and b in Fig. 1 (6). Thefeedback action of thrombin results in the cleavage a t Arg'55-Ser'"6 In the Fragment 1 . 2 region to yield Fragment 1 (residues 1-155)and Fragment 2 (residues 156-273). A second feedback reaction catalyzed by thrombin results in a cleavage at Arg286-Thr287 and produces a species of prethrombin 2 or a-thrombin with 13 residues deleted from its amino terminus. The available data indicate that these two cleavages are catalyzed by a thrombin and not factor Xa in the presence or absence of factor Va (17). The analysisof prothrombin activation catalyzed by prothrombinase or by factor Xa alone is therefore complicated by the side-reactions that result from the feedback action of newly formed meizothrombin or thrombin. The use of the fluorescent, reversible thrombin inhibitor dansylarginine-N-(3-ethyl-l,5-pentanediyl)amide (DAPA)' eliminates the thrombin-catalyzed feedback cleavages (18).Due to the fluorescence enhancement thataccompanies the formationof thethrombin-DAPA complex, thisinhibitor providesa method for continuously monitoring the progress of the conversion of prothrombin to thrombin. In addition,DAPA has been demonstrated to exhibit differential fluorescence properties when complexed with the different intermediates of prothrombinactivation (19). Thedifferential fluorescence properties of DAPA complexed withmeizothrombin,prethrombin 2, and a-thrombin allow for the evaluation of the kinetics of cleavage of the individual bonds catalyzed by prothrombinase by the independent analysis of the conversion of meizothrombin or prethrombin 2 to thrombin. The fully assembled prothrombinase complex catalyzes the activation of prothrombin to thrombin a t a rate that is approximately 5 orders of magnitudegreaterthantherate observed when factor Xa alone is used as the catalyst (7, 9). The rate enhancement that accompanies complex formation has been rationalized on the basis of the co-concentration of enzyme and substrate on the phospholipid surface and the '

' The abbreviations used are: DAPA, dansylarginine-N-(3-ethyl1,5-pentanediyl)amide; dansyl, I-dimethylaminonaphthalene 5-sulfonic acid; Gla, y-carboxyglutamic acid NaDodSO1-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis; PCPs, vesicles composed of phosphatidylcholine-phosphatidylserine.

Materials-Tris base,Echis carinatus venom, L-a-phosphatidylcholine (hen egg), and L-a-phosphatidylserine (bovine brain) were obtained from Sigma. Benzamidine-Sepharose was from Pharmacia P-L Biochemicals. Phospholipid vesicles (PCPs) composed of75% (w/w) phosphatidylcholine and 25% (w/w) phosphatidylserine were prepared as previously described (22, 23). DAPA was synthesized as outlined by Nesheim et al. (9).Humanfactor V was purified by immunoaffinity chromatography as previously described (9, 24). Aliquots of factor,V (0.1-0.5 mg/ml) were activated to factor Va by the addition of bovine thrombin by the methods of Nesheim et al. (9). Factor Va stocksolutions were maintainedon ice and were used within 3 hof preparation. Bovine a-thrombin was prepared according to the procedure of Lundblad et al. (25). The human proteins prothrombin, prethrombin 2, Fragment 1.2, and factor X were purified by previously described procedures (4, 26, 27). Human prothrombin wasradiolabeledwith by the IODO-GEN method(28).Factor X was activated to factor Xa according to the procedure of Jesty and Nemerson (29), using the purified Factor X activator from Russell's viper venom. When activation was complete, as judged by clotting assays, the reaction was chilled to 4 "C and applied to a 1.5 X 15-cm column of benzamidine-Sepharose equilibrated in 20 mM Tris, pH 7.4,0.15 M NaCl a t a flow rate of 60 ml. hr". The column was washed with the same buffer to remove unbound proteins, and bound factor Xa was eluted with 20 mM Tris, pH 7.4, 0.15 M NaCl containing 4 mM benzamidine. Fractions containingfactor Xa activitywere pooled and precipitatedby the additionof solid (NH4)zS04to 80% saturation. The precipitated proteinwas collected by centrifugation (53,000 X g, 20 min), resuspended in 50% (v/v) glycerol and stored at -20°C. The prothrombin activator from the venom of E. carinatus was partially purified by the procedures described by Nesheim and Mann (26). The molecularweights and extinction coefficients (E;?),respectively, used to calculate proteinconcentrations were humanFactor V, 330,000, 0.96 (9); human Factor X, 65,300, 1.16 (30); human Factor Xa, 46,000, 1.16 (30); human prothrombin, 72,000, 1.47 (4); human prethrombin 2, 37,000, 1.95 (4); and human Fragment 1.2, 34,000, 1.12 (4). Measurement of theKinetics of ThrombinFormationUsing DAPA-The formation of thrombin using prothrombin, prethrombin 2 plus Fragment 1.2, or meizothrombin as substrates was continuously monitored by the change in the fluorescence intensity of DAPA present in the reaction mixtures. The use of DAPA also prevented any feedback reactions catalyzed by newly formed product. Fluorescence intensity measurements were performedwith an SLM 8000 photon-counting fluorescence spectrophotometer. Reaction temperatures were maintained at 25 "C using a refrigeratedcirculating water bath. The excitation wavelength was 280 nm (band pass, 8 nm), the emission wavelength was 565 nm (band pass, 16 nm), and scattered light was minimized with a long-pass filter (Schott KV 408) in the emission beam. All bufferswerefilteredusing 0.45-pm filters to further reduce scattered light artifacts. When the factor Va concentration was varied, reaction mixtures (2.0 ml) in continuously stirred cuvettes contained 20 mM Tris, pH 7.4,0.15 M NaC1, 2 mM CaClZ,20 W M PCPs, 1.4 p~ prothrombin, 3 FM DAPA, and variable concentrations of Va. The reaction mixtures were initiated by the addition of factor Xa toa final concentration of 1.02 X lo-' M or 2.04 X io-' M. When prethrombin 2 was the substrate, reaction mixtures contained 1.4 WM prethrombin 2 and 1.4 p~ Fragment 1.2. Meizothrombinwas generated in situ using the activator from E. carinatus exactly as described (26) prior to the addition of factors Va and Xa. In these experiments, the initial concentration of prothrombin was 1.4 pM, and the inferred concentration of meizothrombin upon completion of the initial proteolytic event was 1.4 p ~ The . conversion of meizothrombin to thrombin was then followed by the modest decrease in fluorescence intensity that accompanies thisconversion following the initiation of the reaction with factor Xa. Whentheconcentration of substrate wasvaried, concentrated stock solutions containing 6 F M prothrombin and 12 p M DAPA or 4

3293

Kinetics of Human Prothrombin Activation prethrombin 2,4 p~ Fragment 1.2, and8 pM DAPA were prepared in 20 mM Tris,pH 7.4, 0.15 M NaCl, 2 mM CaC12. These stock solutions were diluted to the indicated concentrations into reaction mixtures (final volume, 2.0 ml) of the same buffer, containing 20 p~ PCPs and4 x IO-' M factor Va. The reaction mixtures were initiated by the addition of 1 X lo-' M factor Xa. When meizothrombin was the varied substrate, variable concentrations of the prothrombinDAPA stock solution described above were prepared in 20 mM Tris, pH 7.4, 0.15 M NaCl, 2 mM CaC12, 30 pM PCPs, and meizothrombin was generated in situby the addition of E. carinatus venom activator (20 pg/ml final concentration). When the fluorescence intensity of the reaction mixture reached a plateau value (typically 1.7-fold greater than that expected for a similar concentration of thrombin-DAPA), the conversion of meizothromhin to thrombin was initiated by the addition of 4 X lo-' M factor Va and 1 X lo-' M factor Xa. Fluorescence Emission Spectra for DAPA Complexed with Thrombin, Meizothrombin, and Prethrombin-Fluorescence emission spectra were collected between 450 and 600 nm (emission band pass, 2 nm) using an excitation wavelength of 280 nm (excitation band pass, 4 nm). A long-pass filter (Schott KV-450) was used in the emission beam to minimize further scattered light artifacts. For these measurements, the SLM 8000 fluorescence spectrophotometer was equipped with hardware and software modifications (On-Line Systems). Uncorrected emission spectra were obtained in the ratiometric mode, 600 points were collected per scan with 10 readings averaged per datum using an electronic time constant of 10 ms. A reaction mixture (2 ml) containing prothrombin (1.4 p M ) , P C P s (30 p ~ ) and , DAPA (3.0 p ~ in) 20 mM Tris.HC1, pH 7.4, 0.15 M NaCl, 2.0 mM CaC12 was initially scanned to determine the emission spectrum of DAPA. The conversion of prothrombin to thrombin was then initiated by the addition of factor Va (10 nM) and factor Xa (3 nM). Fluorescence intensity was monitored at 545 nm until a stable value was reached (approximately 12 min) indicating the quantitativeconversion of prothrombin tothrombin (confirmed by NaDodS0,PAGE). The emission monochrometer was then scanned to obtain the emission spectrum of the thrombin-DAPA complex. A second scan obtained after 18 min produced an identical spectrum. In order to obtain the emission spectrum of the meizothrombin-DAPA complex, a reaction mixture identical to thatdescribed above was initiated with partially purified E. carinatus venom activator (10 pg/ml) to convert prothrombin to meizothrombin. When fluorescence intensity (monitored at 545 nm) reached a stable value, the emission monochrometer was scanned to obtain the emission spectrum of the meizothrombin-DAPA complex. The uncorrected emission spectrum of the prethrombin 2-DAPA complex was obtained using a reaction mixture containingprethrombin 2 (1.4 pM),fragment 1.2 (1.4 pM), P C P s (30 p M ) , DAPA (3.0 p M ) in 20 mM Tris.HC1, pH 7.4, 0.15 M NaC1, 2.0 mM CaC12. Analysis of Prothrombin Activation by NaDodS0,-PAGE-The reaction mixture (40 ml) contained 1.4 pM prothrombin, 3 p~ DAPA, 20 p M PCPs, 4 X lo-' M factor Va in 20 mM Tris, pH 7.4, 0.15 M NaC1,2 mM CaC12 and was initiated at25 "C by the addition of factor Xa to a final concentration of 1 X lo-' M. At intervals, aliquots (2.0 ml) were withdrawn and quenched by the addition of an equal volume of glacial acetic acid. The quenched sampleswere dialyzed extensively at 4 "C against 0.2 M acetic acid and lyophilized. Following lyophilization, the residue was dissolved in 0.2 ml of 62.5 mM Tris, pH 6.8, 2% (w/v) NaDodSO, and subjected to polyacrylamide gel electrophoresis with and without reduction with 5% (v/v) 2-mercaptoethanol. NaDodS0,-PAGE was performed in 1.5-mm gels composed of 9.5% acrylamide, 0.5% N,N'-methylenebisacrylamide according to theprocedure described by Laemmli (31). Protein bands were visualized by staining with Coomassie Brilliant Blue R-250, and the gels were analyzed by densitometry using a Shimadzu CS-930 scanner. The relative staining intensities of thrombin and Fragment 1.2 relative to prothrombin were determined from scans of lanes containing only prothrombin or an equimolar mixture of thrombin plus Fragment 1 . 2obtained afterthe completeactivation of prothrombin. Meizothrombin concentrations were inferred both from scans of Fragment 1.2-A chain on reduced gels and from the difference in intensity of the material observed a t the position of prothrombin on reduced and nonreduced gels. The staining intensity of Fragment 1.2-A chain was assumed to be identical to thatof Fragment 1.2. Thetwo methods of analysis gave indistinguishable results. In separate experiments, the activation of prothrombin by the prothrombinase complex or by factor Xa alone was followed using 12sII-labeledprothrombin. For the activation of prothrombin by prothrombinase, the reaction mixture (2.0 ml) contained 1.4 p~ prop~

thrombin (10' cpm 1251-labeledprothrombin), 3 p~ DAPA, 20 p~ PCPs, 4 X M factor Va in 20 mM Tris, pH 7.4,0.15 M NaCl, 2 mM CaClz and was initiated by the addition of factor Xa (1 X lo-' M). When factor Xa alone was the catalyst, the reaction mixture contained 1.4 p~ prothrombin (lo7cpm 1251-labeledprothrombin), 5 p~ DAPA, 30 p~ P C P s in 20 mM Tris-HC1, pH 7.4, 0.15 M NaCl, 2 mM CaC12and was initiated by the addition of 1 X 10" M factor Xa. At the indicated times, 0.1-ml aliquots of the reaction mixture were withdrawn and quenched by the addition of (0.1 ml) 0.125 M Tris, pH 6.8, 4% (w/v) NaDodSO,, 10 mM EDTA. Twenty-five-microliter aliquots were heated at 110 "C in the presence and absence of 2% (v/ v) 2-mercaptoethanol and subjected to NaDodS0,-PAGE as described above. The proteins were visualized by autoradiography. Data Analysis"nitia1 velocity values were calculated from traces of fluorescence intensity versus time from the initial, steady-state portion of the progress curves. There is a short lag phase (approximately 10 s) associated with the progress curves of human prothrombin activation and thiswas not included in the dataused to calculate steady-state rates. Fluorescence intensity at the completion of the reaction(infinite time) was considered to represent quantitative conversion of substrate to product. This intensity value was then used to convert the steady-state rate (fluorescence change per unit time) into units describing product concentration as a function of time. Steady-state kinetic constants were calculated from the databy weighted nonlinear least-squares regression analysis using the Fortran computer program HYPER described by Cleland (32). Dissociation constants and stoichiometries for the interaction between factor Xa and PCPs-bound factor Va were obtained from the dependence of initial rate on the concentration of factor Va according to Equation 1.

where [E,] is the fixed concentration of enzyme (factor Xa), [TJ is the totalconcentration of titrant (factor Va),[Eb] is the concentration of enzyme complexed as prothrombinase, i is the number of moles of enzyme (factor Xa) combining per mole of titrant (factor Va),and & is the dissociation constant that describes the interaction between enzyme (factor Xa) and titrant (factor Va). A detailed description of the derivation of this expression is available in the literature (33). If R, is the catalytic rate observed when [Eb] = 0 (allfactor Xa is uncomplexed with factor Va)and Rmaxis the rateobserved when [E,,] = [E,], then the concentration of bound Xa ([&I) at any concentration of titrant ( [ T J can ) be related to the observed catalytic rate Robs by Equation 2.

When R, is smallin comparison to Rmax,(Itmax - R,) = R,,,, 1 and 2 can be combined to yield Equation 3.

Equations

This relates the observed catalytic rate to theconcentration of factor Xa complexed in prothrombinase. The stoichiometry and thedissociation constant for the factor Xafactor Va interaction were determined by fitting steady-state rates for prothrombinactivationobtained at varying concentrations of factor Va using two fixed concentrations of factor Xa to Equation 3 using a nonlinear least-squares computer program (CET Research Group), written in Basic and utilizing the Marquardt algorithm. RESULTS

Measurement of the Stoichiometry and Dissociation Constant for the Factor Xu-Factor Vu Interaction-The dependence of theinitialrate of prothrombinactivationonthe concentration of the cofactor (factor Va) in the presence of two fixed concentrations of factor Xa is illustratedin Fig. 2. Inboth cases, initialrate increased untilsaturation was

KineticsProthrombin of Human

3294 loo

Activation

ages to the kineticsof the overall conversion of prothrombin ’ 1 to thrombin. Initial rates obtained aasfunction of increasing

concentrations of the substrate prothrombin are replotted in double-reciprocal form in Fig. 3A. The data illustratedin Fig. 3, B and C, were obtained by usingeither meizothrombin (Fig. 3B) or a mixture of prethrombin 2 and Fragment 1.2 (Fig. 3C as substrates. The data obtained in all three cases are fairly similar,andthe kinetic constantsobtained for the overall conversion of prothrombin to thrombin and for the cleavages of the inlvidual bonds are summarized in Table I. The Michaelis constants obtained for the cleavage of the individual bonds (meizothrombin and prethrombin2 as substrates) are slightly lower than those obtainedfor the overall 0 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00 reaction using prothrombin as the substrate. However, the catalytic efficiency of prothrombinase (kJK,) for the overall CVal / CXal FIG. 2. Dependence of initial rateof prothrombin activation reaction is essentially identical to the catalyticefficiency for on the concentration of cofactor. Reaction mixtures containing the cleavage of the individual bonds in the substrate. These cleavage of the individual bonds in the 1.4 g M prothrombin, 3 g M DAPA, 30 g~ PCPs, and varying concen- data indicate that the trations of factor Va in 20 mM Tris.HC1, pH 7.4, 0.15 M NaCl, 2.0 zymogen is kinetically indistinguishable from the overall conmM CaC12 were initiated by the addition of factorXa to a final version of prothrombin to thrombin, hence precluding the concentration of 1.02 X lo-’ M (0)or 2.04 X lo-’ M (0).The lines identification of a single rate-limiting step in the reaction or are drawn according to Equation 3 described under “Data Analysis” and correspond to a Kd of1.04 X lo-’ M f 0.27 X lo-’ M and a the exclusion of one of the two pathways (see Fig. 1) for the stoichiometry of 1.12 f 0.11 mol of factor Va bound per mol of factor activation of prothrombin on a purely kinetic basis. Xa at saturation. Analysis of Prothrombin Activation Catalyzed by Prothrombinase in the Presence of DAPA by NaDodS0,-PAGE-Due obtained with increasing concentrationsof factor Va, as would to the lack of kinetic evidence for a preferred pathway for be predicted if an interaction between factor Xa and phos- prothrombin activationcatalyzed byprothrombinase, samples were removed from an on-going reactionmixture of prothrompholipid-bound factor Va producedamoreactiveenzyme species. The significant shift in the titration curve observed bin activation catalyzed by prothrombinase and subjected to NaDodS0,-PAGE analysis before and after disulfide bond when the fixed concentration of enzyme (factor Xa) is increased by a factor of 2 suggestsstrongly that the dissociation reduction with 2-mercaptoethanol. The results obtained are constant for the interaction between factor Xa and factorVa illustrated inFig. 4, A (before reduction) andB (after disulfide were visualized by staining on the PCPs surface is approachedby the fixed concentration bond reduction), and the proteins of enzyme. The lines illustrated in the figure were drawn with Coomassie Brilliant Blue. The various intermediates according to Equation3 described under “Experimental Pro- were identified by comparison to standard prothrombinfragmentsseparately analyzed by NaDodS0,-PAGE(datanot cedures” using the computer-fittedvalues of Kd = 1.04 x lo-’ shown). The speciesidentified in Fig. 4A are prothrombin M 2 0.27 X lo-’ M and a stoichiometry of 1.12 f 0.11 mol of (substrate) and the terminal products of the reaction, thromfactor Va bound permol of factor Xa at saturation. The initial rate for prothrombin activation observed in the absence of bin and Fragment 1.2.Notably absent are prethrombin1 and factor Va a t both concentrations of factor Xawas essentially the species Fragment 1 and Fragment 2, indicating the effeczero, hence validating the simplifying assumption used to tiveness of DAPA in inhibiting the feedback reactions catais essentially obtain Equation 3. In addition, these data demonstrate that lyzed by thrombin. In addition, prothrombin the presence of the cofactor increases the overall rate of quantitatively consumed to produce a-thrombin during the prothrombin to thrombinconversion by minimally afactor of course of the reaction, indicating the quantitativeconversion 1000. These values are similar to those determined for the of substrate to product. The pattern observed after disulfide factor Xa-factorVa interaction inbovine prothrombinase (7). bond reduction (Fig. 4B) is similar in that the immediately On thebasis of these data, subsequent kinetic studies designed identifiable species correspond to prothrombin, Fragment 1. to examine the propertiesof fully assembled prothrombinase 2, and the A and B chains of thrombin. In addition to these, were performed a t fixed concentrations of 1nM factor Xa and a species is observed only after reduction which migrates between 4 and 5 nM factor Va. At these nominal concentra- between prothrombin and Fragment 1.2 and has been identions of enzyme constituents, approximately 85% of the added tified as the Fragment 1.2-A chain. This species is not detected prior to reduction with 2-mercaptoethanol, which is factor Xa is complexed withfactor Va.As the remaining consistent with the hypothesis that Fragment 1.2-A chain uncomplexed factorXa(approximately15%)isinherently less active andpossesses %ma the activityof factor Xa incor- arises from the presence of meizothrombin as an intermediate porated into prothrombinase, the contribution of free factor of the overall reaction. This hypothesis is also supported by is detectable in a Xa to the observed kinetics of the prothrombinase complex the observation that Fragment 1.2-A chain transient manner during the early portion of the reaction, will be negligible. Kinetic Constants for the Conversion of Prothrombin, Meizo- exactly as would be expectedfor an intermediate. Meizothrombin, and Prethrombin 2 plus Fragment 1.2 to Throm- thrombin present as an intermediate in the early portion of the reaction would co-migrate with prothrombin before rebin-The intermediate meizothrombin containsanintact peptide bond at Arg273-Thr274 whereasthe mixture of inter- duction of the disulfide bonds. Reduction with 2-mercaptoethmediates composed of prethrombin 2 and Fragment 1.2con- anol converts meizothrombin to Fragment 1.2-A chain and tains an intact peptide bond at Ar$22-Ile323.These interme- the B chain of thrombin. No prethrombin 2 or prethrombin diates were therefore used in order to examine individually 2-des-1-13 was detected by this methodof analysis andwould the cleavage of the two peptide bondscatalyzed by prothrom- have been readily resolved from thrombinfollowing reduction binase and to compare the kinetics of these individual cleav- (see Fig. 4B). The dataobserved following NaDodS0,-PAGE

Kinetics of Human Prothrombin Activation A

S O L ' .

0 -

"

c 4.0 E

"

"

'

"

. .

. .

\

3.0

.

4.0.

'.

-

.

3.0

.

or,

5 '0

-

1/[113

. '

B '_

.

. . 1.5

1.0

QJM)"

cE

I I

1.0

a5

K

\

"

1.5

0 0

'

'

0

2.0 .

. - 0: .5 . 1.0. . 2.0' * 3.0* ' 4.0 . *5.0 ' 0 o

'

0

I

>

'

'

2.5.

S

X

3.5

'

3295

2.0

2.5

I/[mIlo~

3.0

3.5

: ~2

4.0

:1.0 0

1.0

2.0

3.0

I / [PRE 21

"

4.0

5.0

+Mi'

FIG. 3. Determination of kinetic constants for the activation of prothrombin catalyzedby prothrombinase. In A , prothrombin concentrations were varied between 0.2 and 6.1 p~ and initial rates were determined using 30 p~ PCPs, 4.1 nM factor Va and 1 nM factor Xa. In B , meizothrombin concentrations were varied between 0.28 and 5.0 p ~ In. C, equimolar concentrations of prethrombin 2 plus Fragment 1.2 (equimolar mixture) were . lines are drawn according to the fitted kinetic constants listed in Table I. varied between 0.2 and 2.9 p ~ The

TABLE I Kinetic constants for theactivation of humon prothrombin Substrate

Bond cleaved"

K," p M f S.E.

k t

S"

f S.E.

A

-

LJKm

II

X l o " M' S"

A + B 1.06 & 0.13 22.4 & 1.15 2.11 Prothrombin 2.29 B 0.66 & 0.09 15.1 0.85 Meizothrombin 2.93 A 0.46 k 0.05 13.5 f 0.5 Prethrombin 2 plus Fragment 1.2 __ Bond cleaved indicates the bond(s) cleaved during the conversion of the substrate to a-thrombin. A = Arg273-Th?7', B = Arg?22-Ser3=. bThe kinetic constantsand standard errors were obtained by nonlinear least-squares regression analysis of the data illustrated in Fig. 3.

*

~~

"

11

12 13 14 15 10 17 10

B

and the lack of detectable levels of prethrombin 2 indicate that the overall activation of prothrombin catalyzed by fully assembled prothrombinase proceeds exclusively via meizothrombin. The gels in Fig. 4, A and B, were subjected to densitometry analysis in ordertoquantifythe levels of the identified Procedures," species. As discussed under"Experimental meizothrombin levels were assigned on the basisof the staining intensity of the material identified as Fragment 1.2-A 1 2 3 4 s o 7 8 e10 chain and from the difference in the staining intensity at the FIG. 4. Analysis of the activationof prothrombin catalyzed position of prothrombin before and after reduction. The re- by prothrombinase by NaDodS0,-PAGE. Aliquots from an onsults obtained by this analysis are illustrated inFig. 5 where going reaction mixture containing 1.4 pM prothrombin, 3 p~ DAPA, the relative concentration of each species is plotted versus 30 pM PCPs, 4 nM factor Va and 1 nM factor Xa were withdrawn, time. The concentrationof prothrombin decreased essentially quenched, and analyzed by NaDodS04-PAGE as described under "Experimental Procedures." Panel A, NaDodS0,-PAGE analysis monotonicdy as a function of time. A short lag was evident prior to disulfide bond reduction. Panel B, samples analyzed after in the densitometry data for the disappearance of prothrom- reduction with 2-mercaptoethanol. Lane I, molecular weightmarkers; bin and agrees well with the lag observed in progress curves lanes 2-18 correspond to aliquots withdrawn at 0,20,40,60,80, 100, and 1200 s after of prothrombin activation using DAPA (see below and Fig. 6, 120,140,160, 180, 200,220,240,300,360,600, trace A ) . The timecourse for the appearance of thrombin was initiation with factor Xa. The proteins were visualized by staining sigmoidal and did not correspond directly to the curve repre- with Coomassie Blue. The indicated bands correspond to: II, prothrombin; Fgl.2-A,Fragment 1.2-A chain; Fgl.2,Fragment 1.2; Ila, senting the depletion of substrate. The relative amount of thrombin; B , thrombin-B chain; and A , thrombin-A chain. meizothrombin rose to a maximum level of approximately 40% in the first 90 s of the reaction and then decreased to undetectable levels within 200 s of the reaction. These profiles human prothrombin catalyzed by factor Xa alone indicated are indicative of a n ordered sequential conversion of human that the conversion of prothrombin to thrombin exclusively prothrombin to thrombinsolely via meizothrombin catalyzed produced prethrombin 2 as an intermediate (2). These data, inaddition totheordered conversion of prothrombinto by human prothrombinase (15,34). Cofactor Dependence of the Cleavages in Prothrombin Cat- thrombin determined when prothrombinase was used as the cofactor, factor Va, influenced the alyzed by Factor Xa-Earlier studies of the activation of catalyst, indicated that the bovine prothrombin (26) indicated that the cleavage of the mechanism of the reaction catalyzed by factor Xa. The data in Table I illustrate the influence of saturating two bonds in prothrombincatalyzed by factor Xa were stimof the factorXa-catalyzed ulated to different extents by the addition of the cofactor, levels of cofactor ontherate factor Va. In addition, studiesof the activationof bovine and cleavage of the individual bonds in prothrombin. As implied

Kinetics ofP rHou~m~ ar on m b i n

3296

80

160

240

320

400

480

560

640

TIME (sed

Activation

450

475

FIG. 5. Quantitation of the intermediates of prothrombin activation by densitometry analysis. The gels illustrated in Fig. 4,A and 8,were scanned in order to determine the concentrations of

500575550525

WAVELENGTH

600