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

J. Biol. Chem., Vol. 276, Issue 31, 28686-28693, August 3, 2001

This Article Abstract Full Text (PDF) All Versions of this Article: 276/31/28686    most recent M101813200v1 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 Boskovic, D. S. Articles by Nesheim, M. E. Search for Related Content PubMed PubMed Citation Articles by Boskovic, D. S. Articles by Nesheim, M. E. Social Bookmarking                      What's this?

Channeling During Prothrombin Activation^*

Danilo S. Boskovic^§, Laszlo S. Bajzar^¶, and Michael E. Nesheim^**

From the Departments of  Biochemistry and  Medicine, Botterell Hall, Queen's University, Kingston, Ontario K7L 3N6, Canada, the ^§ Children's Hospital of Philadelphia, Philadelphia, Pennsylvania 19104, and ^¶ Hamilton Civic Hospitals Research Centre, Hamilton, Ontario L8V 1C3, Canada

Received for publication, February 27, 2001, and in revised form, May 16, 2001

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The plasma zymogen prothrombin (II) is converted to the clotting enzyme thrombin (IIa) by two prothrombinase-catalyzed proteolytic cleavages. Thus, two intermediates, meizothrombin (mIIa) and prethrombin-2 (P2), are possible on the reaction pathway. Measurements of the time courses of II, mIIa, P2, and IIa suggested a channeling phenomenon, whereby a portion of the II is converted directly to IIa without free mIIa and P2 as obligatory intermediates. Evidence for this was that the maximum rate of IIa formation preceded the maximum in the level of either intermediate. In addition, analysis of the data according to a model that included two parallel pathways through mIIa and P2 indicated that about 40% of the II consumed did not yield free mIIa or P2. Further studies were carried out in which II was continuously infused in a reactor at a constant rate. Under these conditions II, mIIa, and P2 reached constant steady-state levels, and IIa was produced at a constant rate, equal to that of II infusion. During the steady state, traces of II, mIIa, and P2 were introduced as radiolabels. Time courses of isotope consumption were first order, thus allowing the rates of consumption of II, mIIa, and P2 to be calculated. Under these conditions the rate of II consumption equaled the rate of IIa formation. Rates of consumption of the free intermediates, however, were only 22 (mIIa) and 15% (P2), respectively, of the rate of thrombin formation. Thus, both the time course experiments and the steady-state experiments indicate that an appreciable fraction of II is channeled directly to IIa without proceeding through the free intermediates mIIa and P2.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The coagulation cascade (1, 2) consists of a series of zymogen to enzyme conversions within the intrinsic, extrinsic, and common pathways that represents a biochemical amplifier, transforming a small initial stimulus into a local, yet explosive activation of prothrombin to thrombin, which in turn activates fibrinogen. Thrombin also activates anticoagulant and antifibrinolytic feedback in the presence of the endothelial surface protein thrombomodulin (3). Thus, the activation of prothrombin is a crucial step of the coagulation system. Its product, thrombin, can lead to procoagulant, anticoagulant, and antifibrinolytic effects subject to local tissue and circulatory conditions (4).

The activation of prothrombin takes place within the prothrombinase enzyme complex that consists of a coagulant-active phospholipid surface, the enzyme factor Xa, the cofactor, factor Va, and Ca²⁺ (5-7). Although factor Xa alone will slowly catalyze prothrombin activation, the activation rate increases 278,000-fold when the full complex is assembled (8-10).

Two bond cleavages are required for prothrombin activation, one at Arg²⁷⁴-Thr²⁷⁵ and the other at Arg³²³-Ile³²⁴ (bovine numbering). Thus, two activation pathways are possible with the characteristic intermediates meizothrombin and prethrombin-2. In the absence of factor Va prethrombin-2 is the predominant intermediate, with or without phospholipid (11). In the presence of both factor Va and phospholipid meizothrombin is the predominant intermediate (12). Studies of the kinetics of activation of prothrombin, meizothrombin, and the pair fragment 1.2 plus prethrombin-2 in the presence of factor Va, but in the absence of phospholipid, are consistent with a predominant meizothrombin pathway (13). Together, these studies suggest that factor Va directs the reaction toward the meizothrombin pathway.

To understand more fully the pathways of prothrombin activation, the present studies were performed in the presence of factor Va but in the absence of the phospholipid surface component. This approach was taken to focus on the effects specifically attributable to factor Xa and factor Va and to simplify the analysis by reducing the system from four to three components.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Materials-- Bovine blood, used for the isolation of prothrombin, factor X, and factor V, was obtained at Ross McFedridge Abattoir of Glenburnie, Ontario, Canada. Acetonitrile, EDTA, and Tricine¹ were obtained from BDH, Toronto, Ontario, Canada; phenylalanylphenylalanyl arginyl chloromethyl ketone was obtained from Behring Diagnostics, La Jolla, CA. PEG-8000, sodium azide, and trichloroacetic acid were obtained from Fisher. Benzamidine, -amino-n-caproic acid, crude lyophilized Echis carinatus venom, HEPES, crude soybean trypsin inhibitor, Tris base, DEAE-cellulose (medium mesh), QAE-cellulose (coarse mesh), cross-linked Sepharose CL-4B, 6-aminohexanoic acid N-hydroxysuccinimide ester (activated CH-Sepharose 4B), and benzamidine-Sepharose were obtained from Sigma. DEAE-Sepharose (fast flow) was obtained from Amersham Pharmacia Biotech. S-2238 was obtained from Kabi Vitrum (Helena Laboratories, Beaumont, TX). Cyanogen bromide and IODO-BEADS were obtained from Pierce. ¹²⁵I was obtained from ICN Pharmaceuticals Canada Ltd., Montreal. Serva Blue G was obtained from SERVA Feinbiochemica, Heidelberg, Germany.

Proteins-- Bovine factor V, prothrombin, and factor X were isolated by procedures published previously (14-16). Soybean trypsin inhibitor is used during the isolation of prothrombin, and measurable levels of this inhibitor of factor Xa remain with the isolated prothrombin. To remove this contaminant, isolated prothrombin was subjected to gel filtration as described previously (13). Factor X isolation was modified by replacing the DEAE-Sephadex (17) with a DEAE-Sepharose column (2.5 × 16.3 cm). Factor Xa was activated with the purified factor X activator from Russell's viper venom, and factor Xa was isolated by chromatography on benzamidine-Sepharose as described previously (18). Ecarin was isolated from lyophilized E. carinatus venom using chromatography on DEAE-cellulose and preparative polyacrylamide gel electrophoresis as described previously (19). The Kunitz inhibitor of factor Xa was isolated from a lyophilized preparation of soybean trypsin inhibitor by an adaptation of the published method (20). 500 mg of crude lyophilized soybean trypsin inhibitor preparation (Sigma, catalog number T-9128) was dissolved in 100 ml of 10 mM KH₂PO₄, pH 7.6, and centrifuged at 30,000 × g for 20 min to remove particulate material. The supernatant was loaded on a pre-equilibrated DEAE-cellulose column (2.5 × 18 cm), while collecting 8-ml fractions, followed by the buffer until the absorbance at 280 nm returned to base line. Elutions were carried out by stepwise increases in NaCl levels (until the absorbance returned to base line) at concentrations of 34, 130, 170, and 250 mM. The collected fractions were then assayed for Kunitz inhibitory activity against bovine factor Xa (17). The peak fraction following elution by 250 mM NaCl was retained.

Radiolabeling of Bovine Prothrombin-- Two IODO-BEADS (Pierce) were rinsed twice with 1 ml of 400 mM Tris, 150 mM NaCl, 25 mM -amino-n-caproic acid, pH 7.4, prior to the final addition of 500 µl of the same buffer. To this was added 10 µl of Na¹²⁵I (1 mCi), and the tube was shaken intermittently for 5 min at 22 °C. This buffer with ¹²⁵I then was transferred into a tube containing 500 µl of 0.2 mg/ml bovine prothrombin (1 ml final). This tube then was incubated on ice for 10 min with intermittent shaking. Iodination was quenched by the addition of 10 µl of 1.0 M sodium metabisulfite followed by intermittent shaking for 2 min. The iodinated prothrombin then was separated from free ¹²⁵I by gel filtration on G-25 Sephadex (7 × 200 mm) while collecting 1-ml fractions. Absorbance at 280 nm was determined for each fraction. Disintegrations from 1 µl of each sample were counted using an LKB Minigamma 1275 gamma counter. The peak containing the ¹²⁵I-prothrombin then was dialyzed against 50% glycerol/H₂O and stored at 20 °C.

Preparation of Ecarin-Sepharose and Factor Xa-Sepharose-- Approximately 1.5 mg of ecarin stock (1 ml) was dialyzed, at 4 °C, against 100 ml of 100 mM NaH₂PO₄, pH 7.0. The volume of the dialysate was brought to 3 ml with the same buffer. 1.5 g of activated CH-Sepharose 4B was washed with 200 ml of 1 mM HCl followed by 200 ml of 100 mM NaH₂PO₄, pH 7.0. This resin then was added to the ecarin solution with stirring, and the intrinsic fluorescence of the supernatant (after a brief centrifugation) was periodically measured. When the coupling was complete, the remaining reactive groups of the resin were blocked by addition of 1.5 ml of 1 M Tris, pH 8.0. To neutralize traces of a contaminating enzyme activity, the ecarin-Sepharose suspension was treated with phenylalanylphenylalanyl arginyl chloromethyl ketone (20 µM final) for 2 h at 22 °C. Following this, it was washed with 100 ml of 20 mM Tris, 150 mM NaCl, 5 mM CaCl₂, pH 7.4. The ecarin-Sepharose then was stored at 4 °C. Stock factor Xa (5 mg) was dialyzed against 500 ml of 0.1 M sodium citrate, pH 6.5, at 4 °C for 2 h. A 5-ml aliquot of packed Sepharose CL-4B was washed with 100 ml of 2 M Na₂CO₃ (ice-cold) in a sintered glass funnel. The resin then was activated with 2 ml of CNBr/acetonitrile (1 g of CNBr in 2 ml of acetonitrile), followed by washing with 200 ml of ice-cold H₂O and 200 ml of 0.1 M sodium citrate, pH 6.5. The resin then was transferred into the factor Xa solution and gently stirred overnight at 4 °C. The resin was blocked by washing it with 200 ml of 1 M Tris, pH 7.8, followed by 200 ml of 20 mM Tris, 150 mM NaCl, 5 mM CaCl₂, 0.002% sodium azide, pH 7.4. The factor Xa-Sepharose was stored at 4 °C.

Chromogenic Substrate-based Time Course Studies-- A 10-ml reaction was set up comprising 1 µM prothrombin, 100 nM factor Va, 4 nM factor Xa, and 10 µM DAPA in 20 mM Tris, 150 mM NaCl, 2 mM CaCl₂, 0.1% PEG, pH 7.4, 22 °C. The DAPA was included in order to inhibit the thrombin and thereby eliminate thrombin feedback cleavages of prothrombin and prothrombin activation products (8). Three control samples (300 µl each) were taken, for time = 0, prior to onset of the reaction by the addition of factor Xa. Following this, further 300-µl samples were removed at specific times and quenched by adding them to 30 µl of 100 µM Kunitz inhibitor, 50 mM sodium citrate, 20 mM Tris, 150 mM NaCl, 0.1% PEG, pH 7.4, and placing them on ice. When all the samples were collected, they were processed for the determination of prothrombin, prethrombin 2, meizothrombin, and thrombin levels by exploiting the overt or latent activity toward the chromogenic substrate S-2238. To distinguish each of these species, all of the above samples were further sub-sampled by diluting 10 µl of each with 190 µl of bovine serum albumin (1 mg/ml), DAPA (526 nM), in 0.20 M Tris, 0.15 M NaCl, 0.02 M CaCl₂, and 0.1% PEG, pH 7.4, to make 4 diluted samples. Each was then treated with one of the following four protocols, A-D.

Protocol A consists of the addition of 5 µl of a solution of 12 µM antithrombin and 10 units/ml heparin, followed by incubation for 10 min; then addition of 5 µl of 500 µg/ml human platelet factor 4 followed by incubation for 3 min; and finally, addition of 5 µl of 0.4 mg/ml ecarin followed by incubation for 5 min. Under these conditions the initial antithrombin/heparin treatment inhibited all thrombin present in the sample. In the absence of heparin, both thrombin and meizothrombin are protected by DAPA from inhibition by antithrombin (21). In the presence of heparin, however, only meizothrombin is protected. The treatment with the human platelet factor 4 neutralized the heparin and thus allowed subsequently produced thrombin to remain active. Treatment with ecarin resulted in the production of thrombin from pre-2 and mIIa from prothrombin. Thus, the final chromogenic substrate activity represents the combined activities of mIIa, mIIa generated from prothrombin, and thrombin generated from pre-2.

Protocol B consists of the addition of 5 µl of a solution of 12 µM antithrombin and 10 units/ml heparin, followed by incubation for 13 min; and then addition of 5 µl of 0.4 mg/ml ecarin followed by incubation for 5 min. The continued presence of antithrombin/heparin ensured that all thrombin was inhibited, while retaining the activities of the original mIIa as well as the mIIa generated from prothrombin by ecarin.

Protocol C consists of simply incubation for 5 min. These samples thus contain the activities due to mIIa and thrombin.

Protocol D consists of the addition of 5 µl of a solution of 12 µM antithrombin and 10 units/ml heparin followed by incubation for 18 min. This treatment allowed only the chromogenic substrate activity of mIIa to remain intact.

After each of these treatments 10 µl of each sample was assayed for chromogenic substrate activity by adding them to respective microtiter wells containing 290 µl of 0.2 mM S-2238, 20 mM Tris, 150 mM NaCl, 2 mM CaCl₂, 0.1% PEG, pH 7.4, and incubation at 22 °C. Time courses of absorbance at 405 nm were then measured in a microtiter plate reader. At this stage the effective concentration of DAPA is 32 nM. Thus, for each sample containing prothrombin (P), prethrombin 2 (P2), thrombin (T), and meizothrombin (M), in various proportions, four rates of hydrolysis of S2238 were obtained (RA, RB, RC, and RD), each corresponding to one of the above protocols A-D. Levels of prothrombin, prethrombin 2, thrombin, and meizothrombin were then calculated according to the following equations. In these equations k is a constant determined from calibration of the assay with thrombin. The factor of 1.13 accounts for an empirically determined 13% difference between the assay responses to meizothrombin and thrombin under these conditions (Equations 1-4).

(Eq. 1)

(Eq. 2)

(Eq. 3)

(Eq. 4)

Steady-state Studies With Continuous Prothrombin Infusion-- A 30-ml reactor consisting of a plastic beaker equipped with a magnetic stirrer was set up partially filled (20 ml) with 0.3 µM prothrombin, 100 nM factor Va, and 4 nM factor Xa (added last to start the reaction) and 10 µM DAPA in 20 mM HEPES, 150 mM NaCl, 5 mM CaCl₂, 0.004% Tween 80, pH 7.4, 22 °C. A 250-µl syringe (Hamilton Co., Reno, NV) was filled with 77.5 µM prothrombin in 20 mM HEPES, 150 mM NaCl, 5 mM CaCl₂, 0.004% Tween 80, pH 7.4, and the prothrombin solution was infused into the reactor at a constant rate of 3.87 µl/min, using a Harvard Apparatus Compact Infusion Pump. Three control samples (50 µl each) were taken, for time = 0, prior to initiation of the reaction by the addition of factor Xa. The onset of infusion coincided with the start of the reaction. Following this, additional 50-µl samples are removed at specific times, quenched by the addition to 5 µl of 100 µM Kunitz inhibitor, 50 mM sodium citrate, 20 mM HEPES, 150 mM NaCl, 0.004% Tween 80, pH 7.4, and placed on ice. At the end of the experiment, the samples were assayed for levels of prothrombin, prethrombin 2, meizothrombin, and thrombin as described above. In order to measure the rate of consumption of prothrombin, meizothrombin, or prethrombin 2 under steady-state conditions, a trace of radiolabeled protein (~10⁷ cpm) was added 15 min after the start of the reaction (by 15 min steady state is attained). Subsequent samples (50 µl) to be used for analysis of isotope were added to the above sampling mixture (5 µl) that also contained 20 µM D-phenylalanyl-L-prolyl-L-arginine chloromethyl ketone. They subsequently were supplemented with 20 µl of a solution containing prothrombin (200 µg/ml), 50 mM Tris, 37.5 mM EDTA, 4% SDS, 2% -mercaptoethanol, 22.5% glycerol, and 0.02% Serva blue. The samples were heated at 90 °C for 4 min and subjected to SDS-PAGE in Tricine buffer (22). The gels were fixed in 50% methanol, 20% ethanol, 6% trichloroacetic acid for 2 h, and then stained with Coomassie Blue and destained. They were subsequently dried using Bio-Gel wrap in Plexiglas frames at 37 °C. Autoradiographs were made with Kodak XAR 5 film, and radioactive bands were excised and counted in a -scintillation counter. ¹²⁵I-Labeled meizothrombin and the pair fragment 1.2:prethrombin 2 were made immediately prior to the start of the experiments by adding radiolabeled prothrombin (4.0 × 10⁷ cpm) in 300 µl of 20 mM HEPES, 0.15 M NaCl, 5.0 mM CaCl₂, 0.004% Tween 80, 20 µM DAPA, pH 7.4, to 300 µl of ecarin-Sepharose or factor Xa-Sepharose, pre-equilibrated in the same buffer. The slurry was mixed and briefly centrifuged, and the supernatant was discarded. The resin with ¹²⁵I-labeled prothrombin was then incubated at 22 °C for 40 min to quantitatively generate meizothrombin or fragment 1.2:prethrombin 2. The resin was then resuspended in 500 µl of the buffer and reprecipitated by microcentrifugation. An aliquot (400 µl) of the supernatant was then added to an ongoing steady-state reaction in the reactor.

Data Analysis-- Time course data were modeled according to Equations 5-7, where C₁ and C₂ are catalytic efficiencies for conversion of P to M and P2, respectively, and C₃ is that for direct conversion of P to T without obligatory free intermediates. C₄ and C₅ are the catalytic efficiencies for the respective conversions of free M and P2 to thrombin.

(Eq. 5)

(Eq. 6)

(Eq. 7)

If [E] of Equations 5-7 were to remain constant over the course of the reaction, Equations 5-7 could be solved as explicit functions of time with free E as a constant. Experience showed, however, that this was not the case because the time course of P could not be fit with randomly distributed residuals to a single exponential decay. Thus, ratios were taken to eliminate [E]. Equation 8 is the ratio of Equations 5 and 6. Separating the variables d[M] and d[P] in Equation 8 yields Equation 9. Integration of Equation 9 yields Equation 10. In these equations, C_T = C₁ + C₂ + C₃.

(Eq. 8)

(Eq. 9)

(Eq. 10)

Division of both sides of Equation 10 by ([P₀]  [P]) results in the linear Equation 11, where Y and X are given by Equations 12 and 13.

(Eq. 11)

(Eq. 12)

(Eq. 13)

The integral included in the term X is evaluated for each time point by determining the area under a plot of M versus lnP between the points ln[P] and ln[P₀]. The data pairs (X and Y) were then subjected to least squares linear regression to Equation 11 (and an identical equation for the Pre-2 data) to find best values of C₁/C_T, C₂/C_T, C₃/C_T = 1  C₁/C_T  C₂/C_T, C₄/C_T, and C₅/C_T. The first three give the relative rates of prothrombin conversions to M, P2, and direct conversion to thrombin, respectively. The last two, respectively, give catalytic efficiencies (k_cat/K_m) of free M and P2 conversion to thrombin, relative to the total catalytic efficiency of prothrombin consumption.

In the steady-state experiments with radiolabeled prothrombin, meizothrombin, or fragment 1.2:prethrombin-2, first order rate constants for consumption of the radiolabeled species were determined by nonlinear regression of the data to a single exponential decay equation versus time of sampling under steady-state conditions. The rate constant was then multiplied by the steady-state concentrations to determine the rates of consumption under steady-state conditions. In these latter experiments flows along each path were therefore determined directly, without the need for a mathematical model of the reaction kinetics whereby rates could be inferred. The only implicit assumptions in the reactor experiments is that the steady-state rate is equal to the rate constant multiplied by the level of the reactant.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Time Course Studies-- The time courses of prothrombin, meizothrombin, prethrombin 2, and thrombin during the conversion of prothrombin to thrombin are shown in Fig. 1. The substrate prothrombin showed a roughly first order decline in concentration. The intermediates showed a characteristic increase over time up to a maximum value, followed by decline thereafter. The meizothrombin concentration peaked at about 3.5 min, and the prethrombin 2 concentration peaked at about 10 min. The time courses of prothrombin, meizothrombin, and prethrombin 2 are qualitatively consistent with free meizothrombin and prethrombin 2 as obligatory intermediates in the conversion of prothrombin to thrombin. The time course of thrombin, however, is not consistent. The thrombin concentration rises immediately upon initiation of the reaction, without the characteristic lag typical of a mechanism with obligatory intermediates. If free meizothrombin and prethrombin 2 were obligatory intermediates, the maximum rate of thrombin formation would be expected to occur sometime in the interval bound by the times in which the levels of the intermediates reached their maxima (i.e. between 3.5 and 10 min). Since this did not occur, the time course of thrombin generation suggests that some direct conversion of prothrombin to thrombin occurred without equilibration of free intermediates with the prothrombinase complex.

View larger version (19K): [in this window] [in a new window]   Fig. 1.   Time courses of prothrombin, meizothrombin, prethrombin-2, and thrombin during prothrombin activation. The concentrations of prothrombin (closed circles), meizothrombin (closed circles, inset), prethrombin-2 (open circles), and thrombin (closed triangles) were measured as described under "Experimental Procedures." The thrombin time course does not show the initial lag expected for a reaction that has free meizothrombin and prethrombin-2 as obligatory intermediates.

In order to quantify the flow of prothrombin to thrombin, the time course data were analyzed as described under "Experimental Procedures." The time courses of five experiments such as the one depicted in Fig. 1 were analyzed. Plots of the parameters indicated by Equations 11-13 under "Experimental Procedures" for meizothrombin (Fig. 2A) and prethrombin 2 (Fig. 2B) were linear, as predicted by Equation 11. Thus, from the vertical intercepts and slopes, the rate constants (relative to the rate constant for prothrombin consumption) for conversion of prothrombin to meizothrombin or prethrombin 2, and relative rate constants for conversion of the intermediates to thrombin could be calculated. The results of these experiments are summarized in Table I. The relative rate constants for conversion of prothrombin to meizothrombin and prethrombin 2 are 0.30 and 0.31, respectively. Thus, according to this analysis, 30% of the prothrombin that is consumed produced meizothrombin and 31% produced prethrombin 2. By contrast, 39% was directly channeled to thrombin without free meizothrombin or prethrombin 2 as obligatory intermediates. Interestingly, whereas the relative rate constants (relative catalytic efficiencies) for cleavage of the Arg³²³-Ile³²⁴ in prothrombin (C₁) and prethrombin 2 (C₅) are about the same (0.30 and 0.40), the rate constant for cleavage of the Arg²⁷⁴-Thr²⁷⁵ bond in meizothrombin (C₄) is 10-fold greater than that for cleavage of the same bond in prothrombin (C₂). In addition, the rate constant for cleavage of Arg²⁷⁴-Thr²⁷⁵ in meizothrombin (C₄) is 7.9-fold greater than the rate constant for cleavage of the Arg³²³-Ile³²⁴ bond in prethrombin 2. 

View larger version (18K): [in this window] [in a new window]   Fig. 2.   Analysis of time course data to determine relative rate constants for the direct conversion of prothrombin to thrombin and the formation and removal of meizothrombin and prethrombin-2. Time courses such as that depicted in Fig. 1 were analyzed and plotted according to Equations 11-13 under "Experimental Procedures" for meizothrombin (A) and prethrombin-2 (B). Rate constants (relative to the rate constant for total prothrombin consumption) are indicated by the values of the vertical intercepts. In this particular experiment, 23% of the prothrombin was converted to meizothrombin, 33% to prethrombin-2, and 46% directly to thrombin. Shown as insets are plots of meizothrombin or prethrombin-2 concentrations versus the natural logarithm of the prothrombin concentrations during the course of prothrombin activation. The integrals of Equation 13 were evaluated by determining the areas under the inset graphs over the appropriate intervals.

Analysis of the Conversion of Prothrombin, Meizothrombin, and Prethrombin 2 to Thrombin in a Reactor Under Steady-state Conditions-- Because the analysis of the flows of prothrombin to thrombin, either directly by channeling or indirectly through intermediates, requires a model of kinetics, the validity of the conclusions reached by the analysis is dependent on the validity of the model. Thus, an alternative approach was sought that does not rely on a model of kinetics. A reactor was set up in which prothrombin was infused at a constant rate, thereby driving the system into the steady state with constant levels of prothrombin, meizothrombin, and prethrombin 2, and with thrombin produced at a rate equal to the rate of infusion of prothrombin. Under these conditions the rate constants for consumption of prothrombin, meizothrombin, and prethrombin 2 could be measured with radiolabeled proteins, and rates of consumption could be calculated from the rate constants multiplied by the steady-state concentration. This provided a direct measure of flow without the need for a model of kinetics. Upon the initiation of prothrombin infusion (Fig. 3), the prothrombin and meizothrombin levels became constant within the first 5-10 min of infusion, the prethrombin 2 approached a steady state within 30-40 min, and thrombin production was linear with respect to time over the entire 1-h course of the infusion. In separate experiments, trace quantities of radiolabeled prothrombin, meizothrombin, or the pair fragment 1.2:prethrombin 2 were added 20 min after the initiation of the infusion (i.e. in the steady state), and time courses of consumption of the radiolabeled species were determined by gel electrophoresis, autoradiography, and -counting of excised bands. Autoradiograms of the experiments with radiolabeled proteins are shown in Fig. 4. All three species were consumed with first order kinetics (Fig. 5). Meizothrombin consumption had the highest rate constant, prothrombin the second highest, and prethrombin 2 the lowest. Under these conditions the half-lives of meizothrombin, prothrombin, and prethrombin 2 were 6.3, 13.3, and 86.4 min, respectively. Three such experiments were performed for prothrombin, meizothrombin, and prethrombin 2 (nine experiments in total). The results are summarized in Table II. The average steady-state concentrations of prothrombin, meizothrombin, and prethrombin 2 were 246, 19.6, and 180 nM, respectively. The average first order rate constants for consumption of these species were 8.5, 21.3, and 1.8 (× 10⁴/s), respectively. Multiplication of the rate constants by the steady-state concentrations thus gave the rates of conversion of the respective species to thrombin. These rates were then compared with the actual rate of thrombin formation determined from the slope of the thrombin time course (eg. Fig. 3). These calculations indicated that the rate of prothrombin consumption, as calculated from the rate constant for isotope consumption, multiplied by the steady-state concentration was nearly equal to the rate of thrombin formation (90 ± 18%, Table II). In contrast, the calculated rates of meizothrombin and prethrombin 2 conversion to thrombin in the steady-state were, respectively, only 22 ± 2 and 15 ± 2% of the rate of thrombin formation.

View larger version (16K): [in this window] [in a new window]   Fig. 3.   Time courses in the prothrombin infusion experiments. Prothrombin was continuously infused into a stirred reaction. Levels of prothrombin (closed circles), meizothrombin (closed circles, inset) and prethrombin-2 (open circles) approached steady-state levels. Thrombin (closed triangles) was formed at a constant rate, equal to the rate of prothrombin infusion.

View larger version (76K): [in this window] [in a new window]   Fig. 4.   Autoradiograms during the steady state. In separate steady-state experiments, such as that depicted in Fig. 3, trace amounts of radiolabeled prothrombin (A), meizothrombin (B), or prethrombin-2 (C) were added at 20 min. Samples were removed at regular intervals over the next 30 min and subjected to SDS-PAGE and autoradiography. The identities of the respective species are indicated to the right. Prothrombin conversion was measured by counting the prothrombin band (II, A); meizothrombin by counting the fragment 1.2-A chain band (F1.2-A, B) and prethrombin-2 by counting the prethrombin-2 band (Pre-2, C).

View larger version (15K): [in this window] [in a new window]   Fig. 5.   Time courses of isotopic species in steady-state experiments. The time courses of prothrombin (open circles), meizothrombin (closed triangles), and prethrombin-2 (closed circles) were determined from autoradiograms such as that depicted in Fig. 4. The data were subjected to nonlinear regression to the equation cpm = cpm0·exp(k·time) + constant to determine best values for the first order rate constants (k). The indicated lines are regression lines.

Thus, the rates of conversion of meizothrombin and prethrombin 2 to thrombin obtained under steady-state conditions also indicate, as did the analyses of time courses of prothrombin, thrombin, and the two intermediates, that some of the prothrombin is converted to thrombin directly, without free meizothrombin and prethrombin 2 as obligatory intermediates. The steady-state experiments also indicate that the rate constant for cleavage of the Arg²⁷⁴-Thr²⁷⁵ bond in meizothrombin is 11.8-fold greater than the rate constant for cleavage of Arg³²³-Ile³²⁴ in prethrombin 2, a value that is similar to the 7.9-fold difference inferred from the time course studies (the ratio of C₄ to C₅, Table I).

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Previous work indicated that both a phospholipid or cellular surface and factor Va profoundly influence the kinetics of prothrombin activation. In the absence of factor Va, and in either the presence or absence of phospholipid, large amounts of prethrombin 2 accumulate during prothrombin activation, an observation from which the conclusion was reached that in the absence of factor Va the reaction proceeds predominantly through the prethrombin 2 intermediate, and phospholipid itself presumably does not alter the reaction pathway (23, 24). Other work indicated that when both factor Va and phospholipid are present, meizothrombin is the only intermediate that accumulates to an appreciable extent from which the conclusion was reached that meizothrombin is the major intermediate of prothrombin activation as catalyzed by the complete prothrombinase complex (18, 25, 26). In addition, the kinetics of activation of meizothrombin and prethrombin 2 conversion to thrombin in the presence of factor Va, but the absence of phospholipid, are consistent with the conclusion that meizothrombin would likely be the favored intermediate in prothrombin activation (13). Thus, previous results are consistent with the concept that factor Va not only profoundly enhances the rate of prothrombin activation but also influences the pathway of prothrombin activation, directing it through the meizothrombin pathway.

The current work was undertaken to determine quantitatively the influence of factor Va on the pathway of prothrombin activation. The studies were performed in the absence of phospholipid in order to focus specifically on the role of factor Va in the process, without the potential for confounding effects of the phospholipid. Initial efforts were directed toward analyzing time courses, such as those depicted in Fig. 1, according to a model based on four first order reactions comprising conversion of prothrombin to the two intermediates and subsequent conversion of the two intermediates to thrombin. The level of the enzyme during the course of the process was considered to be constant. Thus, the time course of prothrombin would be described by a single exponential, and the time courses of the two intermediates would be described by double exponential expressions. From these, the rate constants for the conversion of prothrombin to the two intermediates and the rate constants for converting the two intermediates theoretically could be obtained by fitting the data to the equations describing the time courses of prothrombin, the two intermediates, and thrombin. Although the data are not shown here, the time courses of prothrombin, meizothrombin, and prethrombin 2 fit reasonably well to the appropriate equations, although minor deviations, as reflected by nonrandom residuals, were evident at levels that could not be attributed to minor systematic errors in the data. In addition, the analyses would not yield logically consistent values for the rate constants. For example, the sum of the rate constants for conversion of prothrombin to meizothrombin and prethrombin 2, as inferred from the time courses of the two intermediates, was not equal to the rate constant for prothrombin conversion inferred from the time course of prothrombin itself. Equally confounding was that values of the rate constants inferred from each of the time courses of the two intermediates suggested that the majority of prothrombin activation occurred through the opposite pathway. This clearly illogical inference indicated that the model comprising two independent pathways was not adequate to describe prothrombin activation. In addition, the thrombin time courses clearly did not display a characteristic lag expected for a system functioning exclusively through obligatory intermediates. Thus, the analysis described under "Experimental Procedures" was employed because it allowed for prothrombin to be converted directly to thrombin and did not depend on the assumption that the level of enzyme remain constant throughout the course of the experiment. The analysis indicated that 30% of the prothrombin was converted to meizothrombin; another 31% was converted to prethrombin 2, and 39% was converted directly to thrombin without release and subsequent conversion of the intermediates, thereby inviting the conclusion that prothrombin activation displays the channeling phenomenon. This was further examined by the prothrombin infusion experiments, which allowed consumption of prothrombin, meizothrombin, and prethrombin 2 to be directly measured under steady-state conditions. These experiments also indicated direct conversion of prothrombin to thrombin without free meizothrombin and prethrombin 2 as obligatory intermediates. In these experiments, 22% of thrombin formation came through meizothrombin and 15% through prethrombin 2. The remaining 63% thus appeared to come by direct conversion of prothrombin to thrombin. Thus, the time course and steady-state experiments both indicate the existence of the channeling phenomenon. Although the results of the two approaches agree qualitatively, the steady-state approach indicates a greater extent of channeling (63 versus 39%). This quantitative difference, however, can probably be reconciled somewhat by normalizing the rates of meizothrombin (22%) and prethrombin 2 conversion (15%) to the actual value measured for prothrombin conversion (90%). Relative to the measured rate of prothrombin conversion, meizothrombin and prethrombin 2 conversion would be 24 and 17%, respectively, and therefore, 59% of the prothrombin would have channeled directly to thrombin. In addition, the rate of prethrombin 2 conversion was probably underestimated somewhat because it had not quite reached its steady-state level and was still slowly accumulating when its rate of flow was measured (20-55 min, Fig. 3). Nonetheless, the two types of experiments together strongly indicate that a sizable proportion of the prothrombin in these experiments was converted to thrombin directly via channeling.

The kinetics of cleavage of the two bonds required for prothrombin activation differ, depending on whether the bonds are cleaved in intact prothrombin or in the intermediates, thereby providing a potential mechanistic rationale for channeling. According to the results of the time course studies, cleavage of the bonds Arg²⁷⁴-Thr²⁷⁵ and Arg³²³-Ile³²⁴ in prothrombin, which form free prethrombin 2 and meizothrombin, respectively, occur in prothrombin with the same catalytic efficiency. The catalytic efficiency of cleavage of Arg³²³-Ile³²⁴ in free prethrombin 2, however, is 1.3-fold greater than cleavage of the same bond in prothrombin (which generates free meizothrombin), and the catalytic efficiency of cleavage of Arg²⁷⁴-Thr²⁷⁵ in free meizothrombin is 10-fold greater than that of cleavage of the same bond in prothrombin (which generates free prethrombin 2). This remarkable increase in catalytic efficiency of cleavage of the second bond required to generate thrombin in the meizothrombin pathway is reminiscent of the so-called efficiency-enhanced distributive mechanism (27), wherein very low levels of intermediate are observed because the intermediate is a much better substrate for the enzyme than is the initial substrate. The existence of this phenomenon in prothrombin activation would explain the low levels and early peak level of meizothrombin, compared with those of prethrombin 2; it still would not explain, however, the acquisition of maximal rates of thrombin formation prior to the time of the meizothrombin peak level. A processive mechanism (27, 28), or channeling, however, posits further that the intermediate of a reaction within an enzymatic complex is processed more efficiently than the free intermediate. Furthermore, the accumulation of free intermediate in such a mechanism represents periodic channeling failure; thus, the level of intermediate may rise coincidentally with the product. This phenomenon is consistent with current results whereby thrombin generation occurred without a lag prior to the time at which the levels of meizothrombin and prethrombin 2 were maximal.

A model of prothrombin activation that includes both channeling and free intermediates is shown in Fig. 6. Pathways allowing the free intermediates meizothrombin and prethrombin 2 are indicated by solid lines and those involving channeling are indicated by dashed lines. The free intermediates are in equilibrium with free enzyme, whereas the channeling intermediates stay associated with the enzyme until the reaction is complete. The species (EP)₁ and (EP)₂ represent two different initial encounter complexes between the enzyme and prothrombin, each allowing for cleavage of one of the two bonds in prothrombin required for eventual conversion to thrombin. That prothrombin could interact with the enzyme (the factor Xa-factor Va complex) in two ways is plausible because multiple binding sites for the prothrombin have been identified among the components of the complex (29, 30). The complexes (EP)₁ and (EP)₂ can equilibrate with free E, whereas and 2 are complexes that do not dissociate to free intermediates. Uppercase K indicates dissociation constants, and lowercase k indicates first order rate constants. If equilibrium is assumed for the binding steps in Fig. 6, and a steady state is assumed in the levels of the various enzymatic complexes, the rate equation for this model is shown in Equations 14-17,

(Eq. 14)

where

(Eq. 15)

(Eq. 16)

and the catalytic efficiency is shown in Equation 17,

(Eq. 17)

The kinetics of the levels of prothrombin, meizothrombin, and prethrombin 2 are given by Equations 18-20.

(Eq. 18)

(Eq. 19)

(Eq. 20)

These equations are identical in form to those presented under "Experimental Procedures" where the time course data were analyzed (Equations 5-7 under "Data Analysis"). Thus, the rate constants (C₁, C₂, C₃, C₄, and C₅ of "Experimental Procedures") can be interpreted according to this model as shown in Equations 21-25.

(Eq. 21)

(Eq. 22)

(Eq. 23)

(Eq. 24)

(Eq. 25)

Therefore, values of the relative rate constants C₁/(C₁+ C₂ + C₃) and C₂/(C₁ + C₂ + C₃) (Table I) can be interpreted in terms of this model as indicating that 30% of the prothrombin is converted to thrombin with free meizothrombin as an intermediate along the path indicated by the solid lines at the top of Fig. 6, 31% with free prethrombin 2 as an intermediate along the path indicated by solid lines at the bottom, and 39% by the dashed lines representing direct channeling. The extent of channeling through each of the two intermediates individually cannot be assessed because values for the individual catalytic efficiencies through the two channeling paths (k₁₃/K₁₁(meizothrombin)) and (k₂₃/K₂₁ (prethrombin 2)) are not revealed in the kinetics of this study.

View larger version (13K): [in this window] [in a new window]   Fig. 6.   Model of channeling in prothrombin activation. The model shows prothrombin (P) interacting with prothrombinase (E) to form encounter complexes (EP)1 and (EP)2, with respective Km values of K11 and K21. These can then turn over to free intermediates meizothrombin (M) or prethrombin-2 (P2) with rate constants k11 and k12 or directly channel to thrombin (T) through the species and 2, by first order processes characterized by the corresponding first order rate constants. Free meizothrombin and prethrombin-2 can interact with free E by interactions with Km values K12 and K22 to form EM and EP2, respectively. These can then turn over to yield thrombin and free E with respective first order rate constants of k12 and k22. The results obtained in this work indicate that 40-60% of the prothrombin proceeds to thrombin along the channeling pathways.

The current studies were intentionally carried out in the absence of phospholipid, even though this component very strongly enhances the kinetics of prothrombin activation. Whether phospholipid modifies the extent of channeling, therefore, is not known. Nonetheless, the current work shows that the channeling phenomenon is a feature of the reaction intrinsic to the proteins that participate in the reaction. Further work will be required to determine whether or to what extent the phenomenon is modified by the surface (phospholipid) component of prothrombinase.

Examples of processive enzymes are common in transcription and translation, as well as a number of DNA or RNA modifications (27). Morris et. al. (28) reported that the vitamin K-dependent carboxylase functions in a processive manner to -carboxylate up to 12 glutamate residues on a peptide substrate comprising residues 18 to +41 of the human coagulation factor IX. Recently, evidence was published that factor XIa-catalyzed activation of factor IX also involves a significant level of channeling (31). The activation of factor IX comprises the cleavage of two peptide bonds and the release of the activation peptide (residues 146-180). When these cleavages are catalyzed by factor XIa, rather then VIIa/TF, a rapid product (factor IXa) generation without a lag phase is observed (31). This is analogous to our observations during prothrombin activation presented here. In both cases the most rapid product formation precedes the time point when the maximum concentration of the intermediate is achieved, implying that the generation of a significant portion of the final product could not be dependent on the prior generation of free intermediate.

Channeling is possible, in principle, for any substrate that is multiply modified by the same enzyme or enzyme complex and that has two or more binding sites for the catalyst (28). Since factor XIa is a disulfide-linked dimer (32), it has two possible binding and active sites, thus explaining this phenomenon. In contrast, the prothrombinase complex, consisting of factor Va and factor Xa in a generally accepted 1:1 stoichiometry (33), only has one active site per factor Xa. In addition, however, the substrate or the intermediate(s) also have a binding site on factor Va (29, 30). Perhaps this allows for possible rearrangements within the complex, without the release of the intermediate, prior to the final activation step.

In conclusion, it is clear that a model of the prothrombin activation pathways incorporating obligatory free intermediates is inconsistent with the observations from the time course as well as from the studies carried out in the steady state. Prothrombin activation can occur by two cleavages in tandem, so that the intermediate is not released from the prothrombinase enzyme complex prior to the second cleavage, a phenomenon generally referred to as channeling or processivity.

	FOOTNOTES

^* The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

^** To whom correspondence should be addressed: Queen's University, Dept. of Biochemistry, Botterell Hall, Rm. A212, Kingston, Ontario K7L 3N6, Canada. E-mail: nesheimm@post.queensu.ca.

Published, JBC Papers in Press, May 30, 2001, DOI 10.1074/jbc.M101813200

	ABBREVIATIONS

The abbreviations used are: Tricine, N-tris(hydroxymethyl)methylglycine; DAPA, dansylarginine-N-(3-ethyl-1,5-pentanediyl)amide; II, factor II/prothrombin; mIIa, meizothrombin; PEG, polyethylene glycol 8000; PAGE, polyacrylamide gel electrophoresis; S-2238, H-D-phenylalanyl-L-pipecolyl-L-arginine-p-nitroanilide dihydrochloride; P2, prethrombin-2.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles: