(Received for publication, May 9, 1997)
From the Department of Biochemistry, Health Science Complex, University of Vermont, Burlington, Vermont 05405
The initiation phase of enzyme generation in a reconstituted model of the tissue factor (TF) pathway to thrombin was evaluated. At 1.25 pM added TF, no thrombin generation was observed in the absence of factor V. The substitution of factor Va for factor V increased the rate of thrombin generation. Factor X activation during the initiation phase was not influenced by the absence of factor VIII or thrombin, leading to the conclusion that initially factor Xa is generated exclusively by the factor VIIa-TF complex. When thrombin was eliminated from the system, no contribution of the factor IXa-factor VIIIa complex to factor X activation was observed during the propagation phase. Similarly, factor V activation was also not observed in the absence of thrombin, indicating that thrombin is the only enzyme responsible for factor V and factor VIII activation. Only subnanomolar amounts of factor VII were activated when prothrombin activation was almost complete. In the absence of coagulation inhibitors, factor XI did not influence thrombin generation initiated by 1.25 pM factor VIIa-TF complex. The termination of factor XIa generation by added hirudin in the factor XI experiment indicates that factor XI activation occurs exclusively by thrombin.
The blood coagulation cascade is thought to be triggered when subendothelial tissue factor (TF)1 is exposed as a consequence of vascular damage (1, 2). TF forms an enzymatic complex with preexistent plasma factor VIIa, and the resulting factor VIIa-TF complex activates factors X and IX (3-5). Factor IXa in complex with its cofactor, factor VIIIa, activates factor X at an ~50-fold higher rate than the factor VIIa-TF complex (6). In turn, factor Xa with phospholipids may activate factor VII (7) and further enhance factor IX and factor X activation. The major function of factor Xa is to form the prothrombinase complex with factor Va and a phospholipid membrane surface, leading to prothrombin activation (8). Thrombin cleaves soluble fibrinogen, forming fibrin, which polymerizes to form an insoluble clot (9). The blood coagulation cascade is down-regulated by the synergistic action of the natural inhibitors of blood coagulation: tissue factor pathway inhibitor, antithrombin III, and the activated protein C system (10, 11).
The procoagulant processes described above may be divided into two
phases (see Fig. 1): (a) an initiation phase,
which is characterized by the appearance of small amounts of thrombin
(<1 nM) and other enzymes (<10 pM) and by low
rates of their generation; and (b) a propagation
phase, which may be characterized by rapid, quantitative
prothrombin activation. Coincidentally, when the inhibitors are
present, thrombin generation is terminated, and extant thrombin is
inhibited by the anticoagulation system. The propagation and
termination processes have been investigated and described in
studies in various reconstituted systems (10-15), in blood plasma
(16), and in minimally altered whole blood (17).
The major regulatory steps that occur during the initiation phase are major factor V and factor VIII and limited factor IX and factor X activation (15). The studies presently available, however, do not identify the sources of the initial factors Va, VIIIa, and Xa required to generate the thrombin that is essential in the catalytic feedback activation leading to explosive thrombin generation. Knowledge of these early reactions is critical to the understanding of blood coagulation, and the answers lie in understanding the dynamics of the initiation phase of the process.
The most common methods used for the evaluation of the different phases of the coagulation process have employed peptidyl-p-nitroanilide substrates (10-13, 15, 16). The high concentrations and amidolytic activity of thrombin allow quantitation with high precision. However, direct evaluation of the activation of the other proteins involved in the coagulation cascade is not a simple task even during the propagation phase, due to their low concentrations (15, 18) and/or their relatively low (if any) activities toward synthetic substrates (19-23). The "initiation phase" is defined in these studies as the period in which little thrombin amidolytic activity is observed or a "lag" time. There is no doubt, however, that enzymatic reactions start at the inception of reagent mixing, but the sensitivities of chromogenic assays do not allow the quantitation of the enzymes present.
The most common methods used for the evaluation of low concentrations of serine proteases are multistage, coupled assays that employ additional protein(s) (24-27). These assays are complicated and have inherent difficulties in interpretation due to almost ubiquitous feedback activation reactions (28-33). One strategy to overcome these problems is the development of synthetic substrates and assays that allow direct and selective quantitation of enzymes at picomolar and lower concentrations.
In a previous report (34), we presented substrates containing the fluorescent aminonaphthalenesulfonamide leaving groups, which are easily modifiable in the sulfonamide moiety to gain selectivity for a given serine protease. Using these substrates, structure-efficiency correlations for a number of the serine proteases involved in blood coagulation and fibrinolysis have been established (23, 34, 35). In this study, we describe the use of 6-peptidylamino-1-naphthalenesulfonamide substrates with selective inhibitors in single enzyme microassays of the blood coagulation serine proteases. The assays were employed to reveal the sequence of events occurring during the initiation phase of zymogen activation in a tissue factor-initiated model of blood coagulation.
Materials
The fluorogenic 6-peptidylamino-1-naphthalenesulfonamide substrates (PNS substrates) were synthesized and characterized as described previously (23, 34, 35). For all assays, substrates were initially dissolved in dimethyl sulfoxide to a stock concentration of 10 mM. Phosphatidylserine and phosphatidylcholine were purchased from Sigma. Phospholipid vesicles (PCPS) composed of 75% phosphatidylcholine and 25% phosphatidylserine were prepared as described (36).
Proteins
The thrombin inhibitor hirudin (37) was a gift from Genentech,
and the factor Xa inhibitor TAP (38) was provided as a gift from Sriram
Krishnaswamy (Hematology/Oncology Division, Department of Medicine,
Emory University). Recombinant human TF-(1-242) and recombinant human
factor VIII (free of albumin) were provided as gifts from Shu-Len Liu
and Rodger Lundblad (Hyland Division, Baxter Healthcare Corp.). Human
thrombin and factors Xa, XIa, and XI were provided as gifts from
Hematologic Technologies Inc. Factor XI was treated with 20 µM D-Phe-Pro-Arg chloromethyl ketone and 1 mM diisopropyl fluorophosphate to inhibit traces of factor XIa in the factor XI preparation and subsequently dialyzed. Monoclonal antibodies HFV-9 and
HFV-17 were from the Biochemistry Antibody Core Laboratory (University of Vermont). Recombinant human factor VIIa
was purchased from NOVO Pharmaceuticals. Human factors II, IX, and X
were isolated from fresh frozen plasma as described previously (39) and
were additionally purified and depleted of trace enzymes as described
(10). Human factor V was isolated by the method of Katzmann et
al. (40) and activated to factor Va as described previously (41,
42). TF relipidation on PCPS vesicles and the factor VIIa-TF complex
formation were accomplished as described previously (43, 44).
Quantitation of Individual Enzymes in Mixtures
Experiments were designed to evaluate each enzyme in mixtures similar to those encountered during mixed coagulation factor activation experiments. For validation/recovery studies of factors Xa and VIIa, 10 nM thrombin, 30 pM factor VIIa, 10 pM factor IXa, and 10 pM factor Xa were incubated at room temperature in HBS/CaCl2 (HBS and 2 mM CaCl2, pH 7.4) containing 200 µM PCPS for 5 min (total volume of 800 µl) (mixture A).
Factor Xa AssayA 200-µl aliquot of mixture A was added to 988 µl of HBS/ETDA (HBS and 20 mM EDTA, pH 7.4) containing 6 µM hirudin (standard concentration). 100 µM 6-(methanesulfonyl-D-Leu-Gly-Arg)-amino-1-naphthalenediethylsulfonamide (mLGRnds) was added, and the rate of substrate hydrolysis was evaluated. A second 200-µl aliquot of mixture A was treated the same way, except that 600 nM TAP (standard concentration) was added. The rate observed for the second aliquot was subtracted from the rate observed for the first aliquot, and the difference was assigned to the factor Xa amidolytic activity. Factor Xa concentrations were estimated from a standard curve prepared using serial dilutions of purified factor Xa.
Factor VIIa AssayThe third 200-µl aliquot of mixture A was mixed with 988 µl of HBS/EDTA containing hirudin (6 µM) and TAP (600 nM). 6-(D-Phe-Pro-Arg)-amino-1-naphthalenebutylsulfonamide (FPRnbs) (100 µM final concentration) was added, and the rate of substrate hydrolysis was evaluated. A fourth 200-µl aliquot of mixture A was treated the same way, except that 20 nM TF was added, and the mixture was incubated for 20 min at room temperature to allow complete factor VIIa-tissue factor complexation. 100 µM FPRnbs was added, and the rate of substrate hydrolysis was evaluated. The difference in substrate hydrolysis rates in the presence and absence of TF is specific for factor VIIa (45). Factor VIIa concentrations were estimated from a standard curve prepared using serial dilutions of purified factor VIIa. Similar factor Xa and factor VIIa assays were carried out for an enzyme mixture that contained 1.4 µM thrombin, 0.3 nM factor VIIa, 1 nM factor Xa, and 1 nM factor IXa (mixture B).
Thrombin Assay1 pM thrombin was added to HBS/CaCl2 containing 200 µM PCPS (total volume of 400 µl) and incubated for 5 min at room temperature. A 200-µl aliquot of this mixture was added to 994 µl of HBS/EDTA containing 600 nM TAP; 50 µM 6-(D-Val-Pro-Arg)-amino-1-naphthalenebutylsulfonamide (VPRnbs) was added; and the rate of substrate hydrolysis was evaluated. A second 200-µl aliquot was treated the same way, except that hirudin (6 µM) was added to the mixture. The difference in rates is specific for thrombin amidolytic activity. Thrombin concentrations were evaluated from a standard curve prepared using serial dilutions of purified thrombin.
Factor XIa Assay400 nM thrombin, 30 pM factor VIIa, 30 pM factor IXa, and 30 pM factor Xa were incubated at room temperature in HBS/CaCl2 containing 200 µM PCPS for 5 min (total volume of 200 µl). This mixture was added to 994 µl of HBS/EDTA containing hirudin (6 µM) and TAP (600 nM), followed by the addition of 50 µM 6-(D-Leu-Pro-Arg)-amino-1-naphthalenepropylsulfonamide (LPRnps). The rate of substrate hydrolysis was evaluated. In the second part of this assay, 1 pM factor XIa was added to the enzyme mixture described above, and this mixture was treated the same way as the previous one. The factor XIa concentration was evaluated from the difference in substrate hydrolysis rates and from a standard curve prepared using serial dilutions of purified factor XIa. A similar factor XIa assay was carried out with a mixture containing 1.4 µM thrombin, 1 nM factor VIIa, 1 nM factor IXa, 1 nM factor Xa, and 15 pM factor XIa.
Coagulation Factor Activation Experiments
These experiments were carried out using the procedures
described in previous publications (10, 15). 1) For experiments designed to evaluate the earliest events occurring during TF-induced thrombin generation, prothrombin (1.4 µM), factor IX (90 nM), factor X (170 nM), factor VII (10 nM), and factor VIIa (100 pM) were preincubated
at 37 °C for 3 min and added to 20 nM factor V or factor
Va, 0.7 nM factor VIII, and 1.25 pM TF
relipidated on 200 µM PCPS (final concentrations in the
reaction mixture). When indicated, factor V, factor VIII, or
prothrombin was omitted, or 3 µM hirudin was added. At
selected time points, three aliquots were removed for the following
assays: (a) a 20-µl aliquot for the evaluation of factor V
activation; (b) a 5-µl aliquot for the thrombin
chromogenic assay using 200 µM substrate Spectrozyme TH
(10); and (c) a 1200-µl aliquot for factor VIIa, factor
Xa, and thrombin fluorogenic assays. Prothrombinase concentrations were
calculated using a kcat for prothrombin
activation of 83.6 s1 (46).
2) In an experiment designed to evaluate the influence of factor XI on the activation process, 1.25 pM factor VIIa was incubated in HBS/CaCl2 for 20 min at 37 °C with 0.25 nM TF relipidated on 200 µM PCPS. Following incubation, factors V and VIII at the concentrations indicated in the previous experiment as well as 30 nM factor XI were added, and the activation reaction was started by the addition of prothrombin, factor IX, and factor X at the above concentrations. At selected time points, 5-µl aliquots were removed for the chromogenic thrombin assay (10), and 400-µl aliquots were taken for a factor XIa assay. The factor XIa assay was based upon the observation that factor XI does not alter the thrombin generation rate in this experimental system. In this assay, 400-µl aliquots were quenched in 794 µl of HBS/EDTA containing standard concentrations of hirudin and TAP. The aliquots were transferred into cuvettes; 50 µM substrate LPRnps was added; and the rate of substrate hydrolysis was evaluated. The rate of the corresponding assay from the control experiment performed in the absence of factor XI was also evaluated, and the concentration of factor XIa was estimated from the difference in rates.
Evaluation of Factor V Activation
20-µl samples of the reaction mixture were quenched in 20 µl
of 2% SDS, 0.06 M Tris, 10% glycerol, and 0.1%
bromphenol blue, pH 6.8, and heated for 5 min at 95 °C. 17-µl
subsamples were subjected to SDS-polyacrylamide gel electrophoresis
under nonreducing conditions on 4-12% polyacrylamide gel as described
by Laemmli (47). Following SDS-polyacrylamide gel electrophoresis, the
proteins were transferred to nitrocellulose membranes for immunoblot
analyses using the general techniques of Towbin et al. (48).
Membranes were blocked for nonspecific binding with 5% nonfat dry milk
in Tris-buffered saline containing 0.05% Tween. Generation of factor
Va was followed by incubating blocked membranes for 1.5 h with
either monoclonal antibody HFV-9 (directed against the light chain)
or
HFV-17 (directed against fragment 307-506 of the heavy chain).
The products of factor V recognized by these antibodies were visualized
using peroxidase-conjugated horse anti-mouse IgG and the
chemiluminescence reagent Luminol (DuPont). Films were developed in a
Kodak X-Omat and scanned for densitometric analysis using a ScanJet
4c/T scanner (Hewlett-Packard Co.).
The specific quantitation of relatively low concentrations of various enzymes generated during the activation process is based upon the substrate's selectivity, the high sensitivity of the substrate leaving group for detection, and the selectivity of specific serine protease inhibitors such as hirudin (thrombin) and TAP (factor Xa). The factor VIIa assay is based upon the increase in amidolytic activity of this enzyme caused by TF under conditions that prevent activation of the zymogens present in the system (10 mM EDTA) (45). The results of control experiments performed with enzyme mixtures in the presence of PCPS, TAP, and/or hirudin at the concentrations present in the enzyme assays in coagulation factor activation experiments showed that the amidolytic activities of factor VIIa, factor Xa, thrombin, and factor XIa are recovered almost quantitatively. This observation is valid for low picomolar concentrations of factors VIIa and Xa measured in the presence of 10 nM thrombin (Table I), for 1 pM factor XIa measured in the presence of 400 nM thrombin, and for 0.3 nM factor VIIa and 1 nM factor Xa or 15 pM factor XIa measured in the presence of 1.4 µM thrombin. The amidolytic activities of thrombin and factor XIa were recovered at 100 and 113-120%, respectively. The amidolytic activities of factor Xa and factor VIIa were recovered at 88-112 and 103-109%, respectively. Thus, thrombin and factor XIa can be measured at femtomolar, factor Xa at 0.4 pM, and factor VIIa at 1 pM concentrations individually2 and reliably (±20%) in the presence of relatively high concentrations of other enzymes.
|
In an experiment initiated with 1.25 pM
TF when both factors VII and VIIa were added to the zymogen mixture,
the initiation or lag phase of thrombin generation evaluated using the
chromogenic p-nitroanilide substrate Spectrozyme TH
continued for ~2 min (Fig. 1). Analyses
of this phase employing a fluorogenic PNS substrate, VPRnbs (Fig.
2), showed undetectable thrombin
generation (<0.2 pM) over the first 20 s after the
reactants were mixed. However, the latter substrate allowed
quantitation of the thrombin generated from 20 s to 2 min. By
30 s, the concentration of thrombin reached 0.60 pM.
Subsequently, the thrombin concentration increased at a relatively
constant rate until 60 s (open circles).
No prothrombin activation was observed over an 8-min period when factor V was omitted from the reaction mixture (Fig. 2, open triangles). The substitution of 20 nM factor Va for factor V decreased the interval in which thrombin generation was undetectable from 20 to 10 s (closed circles). The transition to the explosive propagation phase of thrombin generation in the experiment started 18-20 s earlier with factor Va than observed in the experiment with factor V (50-55 and 70-73 s, respectively). In the absence of factor VIII (closed triangles), the thrombin generation rate during the initiation phase was similar to that observed in the presence of factor VIII.
Analyses of the thrombin generation rates indicated that, during the
initial 60 s, the concentration of prothrombinase (deduced from
d[IIa]/dt) in the factor Va experiment increased from
undetectable levels to 7 fM (Fig. 2, inset,
closed circles). In the presence of factor V, whether factor
VIII was present (open circles) or absent (closed
triangles), the rise in prothrombinase concentration started at
~50-60 s. When factor Va was substituted for factor V, the rise in
prothrombinase concentration occurred at a similar rate, but 20 s
earlier. In experiments with and without factor VIII, the first
detectable amounts of factor Va heavy chain were observed at 30 s
(Fig. 3, A and C,
respectively), i.e. approximately at the same time when the
first detectable amounts of thrombin (0.6-0.7 pM) (Fig. 2)
were observed. However, the first detectable amounts of factor Va light
chain (0.36 nM) were observed after a 70-s delay relative
to the appearance of the heavy chain, i.e. at 100 s
(Fig. 3B).
The cleavage of intact factor V in the presence or absence of factor VIII was complete at 100-120 s after the reactants were mixed (Fig. 3, A-C). This activation was obligately dependent on thrombin since in the presence of 3 µM hirudin, no activation of factor V was observed over a 4-min period (Fig. 3D).
In the complete, factor V-containing system, the first detectable
amounts (8.4 pM) of factor Xa were observed 60 s after
mixing the reagents (Fig. 4, open
circles). This initial rate of factor Xa generation was not
altered by the presence of 3 µM hirudin (closed
circles), the absence of prothrombin (open triangles), or the absence of factor VIII (closed triangles).
Significant prothrombinase activity began to form at 50-60 s (Fig. 2, inset), i.e. the time at which 5-10 pM factor Xa was present. A significant increase in the rate of factor Xa generation was observed after a 3-min lag period only in the complete system (Fig. 4, open circles). In experiments when factor VIII was deleted or when factor VIII activation by thrombin was prevented, factor Xa generation remained almost linear over a 4-min period. Factor X activation rates in the absence of prothrombin or factor VIII or in the presence of hirudin were similar. These data lead to the conclusion that thrombin is the only enzyme responsible for factor VIII activation under these experimental conditions.
The data presented in Fig. 5 demonstrate
that, during the initiation phase, the concentration of prothrombinase
(open triangles) is significantly lower than that of factor
Va heavy chain (open circles) or factor Xa (closed
triangles). When factor Va light chain (closed circles)
achieved subnanomolar levels and factor Xa was present at picomolar
concentrations, the concentration of prothrombinase increased to
picomolar levels. During the first 60 s of the propagation phase,
the concentration of prothrombinase appeared similar to that of factor
Xa.
The process of factor VII activation is largely dependent upon the
concentration of active thrombin present in the system. In the absence
of factor VIII, the stable thrombin formation rate of the propagation
phase of prothrombin activation is shifted from 3 to ~4 min in
comparison with the complete system, and thrombin generation during
this phase progresses at a lower rate (15). Detectable increases in
factor VIIa (70 and 80 pM, respectively) were observed at 4 min in the complete system and at 6 min in the absence of factor VIII
(Fig. 6, open circles and
closed triangles, respectively). At that time, however, the
thrombin concentrations were 850 and 270 nM, respectively.
In both cases, the activation of factor VII proceeded at similar rates.
In experiments where thrombin was absent (open triangles) or
inactivated by hirudin (closed circles), the initial rates
of factor VIIa generation were significantly lower than those achieved
when active thrombin was present. These observations allow the
conclusion that the predominant activator of factor VII is
thrombin.
Factor XI Activation and Its Influence on Thrombin Generation
In experiments designed to evaluate the influence of
factor XI, the reaction was initiated by 1.25 pM factor
VIIa-TF complex, and the generation of thrombin and factor XIa was
followed during the first 5 min. The substitution of preformed 1.25 pM factor VIIa-TF complex (Fig.
7) for 1.25 pM TF added to
the protein mixture containing factor VIIa (Fig. 1) decreased the
initiation phase from 2 min to ~30 s. The initiation phase of
thrombin generation was followed by the propagation phase (from 30 to
150 s for the preformed complex and from 2 to 5.5 min for the TF
experiment). The thrombin generation curves (circles) in the
factor VIIa-TF experiment were similar in the presence and absence of
factor XI.
The fluorogenic substrate LPRnps was employed for the factor XI
activation assay to quantitate factor XIa at subpicomolar concentrations. The data presented in Fig. 7 (triangles)
demonstrate that there is no detectable (<0.2 pM) factor
XIa observed within the first minute of the reaction, i.e.
when thrombin is present at relatively low concentrations. The first
detectable traces (0.7 pM) of factor XIa were observed at
1.5 min. In contrast, at that time, thrombin generation was proceeding
in an explosive manner at 20 nM/s. The maximum rate of
factor XI activation was observed at ~5 min after the beginning of
the reaction and was linear for at least 4 h (Fig.
8, closed circles). During
this period of time, the rate of factor XIa generation was 2.9 × 1013 M/s. At the 4-h time point, only ~15%
(4.5 nM) of factor XI was activated to factor XIa. When 3 µM hirudin was added to the reaction at 28 min after the
initiation of activation, further generation of factor XIa was
terminated (Fig. 8, open circles). These observations lead
to the conclusion that thrombin is the only kinetically relevant activator of factor XI in the absence of inhibitors in the system.
The PNS substrates coupled with specific inhibitors and activators permit the evaluation of thrombin and factors Xa, VIIa, and XIa generated during the initiation and propagation phases of the tissue factor-induced procoagulant pathway. The "initiation phase" has been defined relative to the sensitivity of p-nitroanilide substrate-based assays. The present data permit insight into the reaction occurring during this lag phase, which ultimately leads to explosive thrombin generation during the propagation phase of the reaction system. The principal conclusions that can be reached with the present extensions of kinetic data are the following. 1) During the initiation phase, the limiting ingredient for prothrombinase complex formation is factor Va light chain. 2) During the initiation phase of the reaction, factor Xa is solely generated by the factor VIIa-TF complex. 3) Significant factor VIIa generation during the initiation phase of the reaction does not occur either by factor VIIa-TF or by factor Xa-phospholipid. 4) In the absence of factor V(a), the concentration of factor Xa produced is incapable of generating thrombin (<0.2 pM at 8 min). 5) The initial catalyst that generates thrombin is either factor Xa-factor V or factor Xa-factor Va, with the latter derived from factor Va contamination of factor V preparations. 6) Thrombin is the only significant activator of factors V, VIII, VII, and XI. The central theme of the collected data is the significance of thrombin as an essential catalyst in thrombin generation.
Analyses of the initiation phase of the tissue factor pathway to thrombin support the conclusion that, during the initiation phase, the limiting component of the prothrombinase complex in the absence of inhibitors is factor Va light chain. This conclusion is based upon the observations that substitution of factor V by factor Va shortens the initiation phase and that substantial amounts of prothrombinase (close to those of factor Xa) are generated only after factor Va light chain appears at relatively high concentrations. Beyond this point, i.e. during the beginning of the propagation phase, factor Xa becomes the limiting ingredient of the prothrombinase complex. The factor Xa concentration does not approach the concentration of factor Va until the end of the experiment, well after all prothrombin has been consumed.
The source of factor V and/or factor Va activity in the initiation phase of the reaction remains an enigma. The activation of prothrombin on a biologically relevant time scale requires either factor V or factor Va participation with factor Xa in the reaction. Since the factor Xa activation of factor V can be ruled out by the data in this study, one is left with the conclusion either that factor V is active in forming prothrombinase or that sufficient amounts of factor Va are present in the factor V preparation to yield the initial complex enzyme.
Previously published data show that, at best, factor V has 0.25% the activity of factor Va as a contributor to prothrombinase complex formation and activity (49). Mathematical simulation data3 indicate that this activity is sufficient to provide the initial amounts of thrombin in this experimental system. On the other hand, these data also indicate that the presence of <0.01% of contaminant factor Va in a factor V preparation would yield similar results. The linear character of factor Xa generation curves during the initiation phase of activation suggests that factor Xa is initially generated only by the constant amount of the factor VIIa-TF complex formed in the reaction mixture. Later on, increasing amounts of the factor IXa-factor VIIIa complex support the factor Xa generation process, and the rate of factor X activation increases. The fact that, in the presence of hirudin as well as in the absence of factor VIII or prothrombin, the initial rate of factor Xa generation is similar to that in a complete system supports the conclusion that, during the initiation phase, factor X is activated only by the factor VIIa-TF complex without participation of the factor IXa-factor VIIIa complex.
A similar rate of factor Xa generation is observed in the absence of factor VIII and in the absence of active thrombin. This observation leads to the conclusion that thrombin is the only enzyme responsible for factor VIII activation. Furthermore, the observation that there is no detectable factor V activation in the absence of prothrombin or active thrombin clearly indicates that this procofactor is also activated only by thrombin under these experimental conditions. The presence of 10-30 pM factor Xa does not result in factor V activation, whereas 0.6-0.7 pM thrombin causes rapid cleavage of intact factor V.
The relatively late appearance of newly generated factor VIIa suggests that, at the TF concentration used, the activation of factor VII does not play any substantial role in the generation of thrombin. Plasma factor VIIa, in the presence of limited concentrations of TF, is able to trigger thrombin generation in an explosive manner. No additional generation of factor VIIa occurs during the time period relevant to the initiation phase at the TF concentration used.
We have shown in a previous publication (44) that the most efficient activator of factor VII (among serine proteases of blood coagulation) is factor Xa on a phospholipid surface. Thrombin also displays proteolytic activity versus factor VII; however, it is 570-fold less effective than factor Xa. The data presented in this study clearly demonstrate that, in the reconstituted system, factor VII is primarily activated by thrombin. This apparent paradox may be explained by the fact that, at the moment of the initial detectable phase of factor VII activation, thrombin is already present at submicromolar concentration, whereas the concentration of factor Xa never exceeds the low picomolar range. Thus, the relatively low molar efficiency of thrombin in factor VII activation is compensated for by the high concentration of this enzyme. Furthermore, the submicromolar amounts of thrombin generated are involved in the proteolytic cleavage of proteins present at low nanomolar and subnanomolar concentrations (factors V, VIII, and VII), whereas the picomolar amounts of factor Xa are involved in the proteolytic cleavage of at least three substrates, i.e. prothrombin, factor IX, and factor VII, of which only prothrombin is present at a concentration exceeding its Km (50). Additionally, the factor Xa generated during the experiment is involved in the prothrombinase complex, and in this complex, it is partially inhibited from activating factor VII (44).
It has been reported that factor XI activation may be caused by thrombin (51, 52). The relevance of this process for in vivo conditions is still under discussion (53-55). Lawson et al. (15) reported that factor XI slightly influenced the explosive propagation phase of thrombin generation in our procoagulant system. Factor XI activation dynamics, however, were not evaluated. The data presented in this study demonstrate that thrombin generation initiated by 1.25 pM factor VIIa-TF complex in the absence of coagulation inhibitors is not influenced at plasma concentrations of factor XI. This is due to a relatively low rate of factor XI activation by thrombin and thus the failure to support factor IXa generation in the initial phase of the coagulation process. In addition, the rapid prothrombin activation in this experimental system does not permit observation of any influence of the factor XIa generated on this process. The only kinetically relevant enzyme present in the experimental system able to generate factor XIa is thrombin. In addition, the results of this experiment clearly demonstrate that subpicomolar concentrations of factor XIa may be detected directly in an assay employing a PNS substrate. The sensitivity of this method is higher2 than that of the method suggested by von der Borne et al. (56), which employs an amplification assay using multiple proteins.
In summary, the enzyme microassays based upon the use of PNS substrates allow quantitation of the initiation phase of blood coagulation in a partially reconstituted system. The data lead to the identification of enzymes or enzymatic complexes responsible for the generation of the initial amounts of serine proteases of blood coagulation and their cofactors.
We are grateful to Maria DiLorenzo for participation in enzymatic evaluation of substrates and to Kevin Cawthern and Kenneth Jones for providing mathematical simulation data. We thank Sriram Krishnaswamy for providing TAP and Shu-Len Liu for providing tissue factor and factor VIII. We finally thank William Church for providing monoclonal antibodies to factor V(a) and Richard Jenny for providing coagulation enzymes.