Comparison of Sustained Antithrombotic Effects of Inhibitors of Thrombin and Factor Xa in Experimental Thrombosis
Background In the pathogenesis of (recurrent) thrombosis, clot-associated thrombin appears to play an important role. Antithrombin III–independent thrombin inhibitors have been shown to neutralize clot-bound thrombin effectively. We compared the sustained antithrombotic effects and the effects on endogenous fibrinolysis of several of these agents with recombinant tick anticoagulant peptide (rTAP), a selective factor Xa inhibitor, and low-molecular-weight heparin (LMWH) in an experimental venous thrombosis model.
Methods and Results Rabbits received either recombinant hirudin (rHir), Hirulog-1, CVS#995 (a novel direct inhibitor of thrombin), rTAP, LMWH, or saline. The effect on thrombus growth was assessed by measuring the accretion of 125I-labeled fibrinogen onto preformed nonradioactive thrombi, and the effect on endogenous fibrinolysis was assessed by measuring the decline in radioactivity of preformed 125I-labeled thrombi in rabbit jugular veins. All direct thrombin inhibitors induced a sustained antithrombotic effect compared with either LMWH and rTAP. In addition, CVS#995 also further decreased thrombus size after stopping its infusion, which was due to a significant enhancement of endogenous fibrinolysis.
Conclusions Direct thrombin inhibition by rHir, Hirulog-1, or CVS#995 induces a sustained antithrombotic effect compared with rTAP and LMWH, which is most likely due to inhibition of clot-bound thrombin. CVS#995 was shown to also enhance the extent of endogenous fibrinolysis to a greater degree compared with rHir and might therefore be an interesting new antithrombotic agent for the treatment of venous and arterial thrombosis.
Unfractionated heparin is the most commonly used anticoagulant drug in the treatment and prevention of venous and arterial thromboembolic disorders.1 Heparin inhibits thrombin and activated factor X after complex formation with its natural serpin cofactor antithrombin III, which dramatically enhances the affinity of antithrombin III for both enzymes.
Despite the proven clinical efficacy of heparin in the treatment of venous thromboembolism, it has been directly demonstrated by venography that there is a 20% to 30% extension of venous thrombi in patients undergoing treatment with therapeutic doses of unfractionated heparin.2 3 4 5 6 It has been shown that discontinuation of heparin therapy is complicated by recurrent venous thrombosis in approximately 10% of the patients in the following 6 months.4 Furthermore, it has been shown that arterial thrombus formation at sites of vascular damage, such as atherosclerotic lesions in the coronary artery after angioplasty or thrombolysis, is not adequately prevented by the administration of unfractionated heparin.7 Therefore, it has been suggested that clot-bound thrombin, which is not effectively inhibited by the heparin–antithrombin III complex, might be at least partially responsible for continuous fibrin formation and recurrent thrombosis.8
Antithrombin III–independent thrombin inhibitors, such as rHir and its analogues, have been shown to inhibit clot-bound thrombin more effectively than the heparin–antithrombin III complexes by the formation of stable, noncovalent 1:1 stoichiometric complexes.9 Clinical and experimental studies have confirmed that rHir is able to reduce thrombus extension more efficiently than unfractionated heparin in a model of experimental venous thrombosis and showed that it is able to prevent the formation of coronary thrombosis after angioplasty.7 10 11
Recently, several new antithrombotic agents have been developed, such as Hirulog-1 and rTAP,12 13 14 which specifically and independent of plasma cofactors inhibit thrombin and factor Xa, respectively. Hirulog-1 is a synthetic polypeptide consisting of binding regions for the catalytic and exosite binding sites of thrombin.14 Hirulog-1 has been suggested to inhibit clot-bound thrombin more effective than hirudin, presumably due to its reduced molecular size, which facilitates access to this pool of thrombin.15 In clinical and experimental studies, Hirulog-1 has been demonstrated to be a highly effective anticoagulant and antithrombotic agent.16 17 18 19 TAP is a novel serine protease inhibitor originally isolated from the tick Ornithidoros moubata.12 The recombinant form of TAP (rTAP) has been shown to have a strong antithrombotic effect in models of experimental arterial20 21 and venous thrombosis22 mediated though the inhibition of de novo thrombin generation or prothrombinase function. Other than the specific and direct inhibitors of thrombin or factor Xa inhibitors, there has been a great deal of research focused on optimizing the pharmacological profile of standard heparin by fractionating the compound to lower-molecular-weight fragments. These LMWH and synthetic heparinoids have been shown to have a more consistent anticoagulant effect and to induce less bleeding complications than unfractionated heparin in the prevention and treatment of venous thrombosis.23 24 The improved pharmacological profile of the LMWH and heparinoids has been attributed in large part to less protein binding, resulting in a better and more predictable pharmacokinetic profile. However, the mechanism of inhibition remains dependent on antithrombin III. So far, no direct comparison between these new antithrombotic agents has been made in a single experimental model.
In the present study, we compared the antithrombotic effect of LMWH, rHir, Hirulog-1, and rTAP with a newly developed thrombin inhibitor, CVS#995, in a rabbit jugular vein thrombosis model. In vitro, CVS#995 has been demonstrated to inhibit thrombin with a higher affinity and is less sensitive to proteolytic degradation than are similar-sized inhibitors such as Hirulog-1. Because all these new agents have a relatively short half-life in vivo, we chose to investigate the antithrombotic efficacy of each agent during a continuous infusion and after plasma levels have decreased. We focused on the comparison of the sustained antithrombotic capacity of all these new direct thrombin inhibitors with LMWH and rTAP. The effect on thrombus growth was assessed directly after discontinuation of the study medication infusion and 1 or 2 hours afterward. In addition, the effects on endogenous fibrinolysis were compared in the same animal model.
New Zealand White rabbits of approximately 2.5 kg were used for this study. Anesthesia was induced by the administration of 9 mg Ketamin (Aescoket) and 0.5 mL Rompun 2% (Bayer). To maintain adequate anesthesia, intramuscular injections of Ketamin were given repeatedly. The carotid artery and jugular veins were exposed through a medial incision in the neck. The carotid artery was cleared, and a cannula (baby feeding tube, 1.6-mm diameter) was introduced for the administration of the study medication and for blood sampling. The jugular veins were cleared over a distance of 2 cm, and all small side branches were ligated. The veins were clamped both proximally and distally.
Thrombus Growth Model
Nonradioactive thrombi were formed in both jugular veins of the rabbit by injection of 150 μL homologous rabbit blood aspirated in a 1-mL syringe containing 25 μL human thrombin (Human Thrombin T7009, Sigma Chemical Co; 150 IU/mL) and 45 μL CaCl2 (0.25 mol/L) into the isolated venous segment. After 30 minutes of aging of the thrombus, the blood flow was restored by removing the vessel clamps, and 100 μL 125I-radiolabeled human fibrinogen (Amersham, approximately 2 μCi) was injected systemically, followed immediately by the administration of a loading dose and a 2-hour continuous infusion of the study compound. Blood samples were taken every hour to calculate the mean plasma radioactivity per milliliter of blood for each rabbit. One thrombus was removed at the end of the 2-hour infusion (t=120), whereas the contralateral thrombus was removed 1 hour (t=180) or 2 hours (t=240) after discontinuation of the study drug administration. Thrombus growth was assessed by measuring the accretion of 125I-radiolabeled fibrinogen onto the preformed nonradioactive thrombi. Thrombus growth was measured by calculating the blood volume accreted onto the clot by comparing 125I-related blood radioactivity with 125I-related thrombus radioactivity. Thrombus growth was expressed as a percentage of the initial thrombus volume.
Measurement of Endogenous Fibrinolysis
To measure the extent of endogenous fibrinolysis, radiolabeled thrombi were formed in the jugular veins of the rabbit, and the degree of fibrinolysis was determined by assessing the decline in initial radioactivity of the preformed thrombus. Therefore, homologous rabbit blood was mixed with 125I-labeled fibrinogen (final radioactivity, approximately 10 μCi/mL). An aliquot of 150 μL of this mixture was then aspirated into a 1-mL syringe containing 25 μL thrombin (150 IU/mL) and 45 μL CaCl2 (0.25 mol/L) and quickly injected into the isolated venous segment. After 30 minutes of aging, the vessel clamps were removed and the infusion of a bolus injection of study drug was given, followed by a 2-hour continuous infusion. One thrombus was removed directly after the 2-hour infusion and the contralateral thrombus was removed 1 hour later. The extent of endogenous fibrinolysis was assessed by measuring the remaining radioactivity of the thrombus at the end of the study compared with the initial radioactivity and was expressed as a percentage of the initial thrombus volume.
CVS#995 is a synthetic peptide of 19 amino acids with the following structure: (CH3CH2CH2)2-CHCO-D-P-R-ω-[COCO]-(G)5-N-G-D-F-(E)2-I-P-E-Y-C-OH (Mr=2133).25 It contains an activated carbonyl transition state mimetric (α-keto-amide), which serves to bridge the catalytic or active site domain with another sequence that binds to the anion-binding exosite I of α-thrombin. Therefore, CVS#995 is a synthetic inhibitor of thrombin that couples the stability to proteolytic degradation by the incorporation of the α-keto-amide functional group with excellent selectivity resulting from the use of specific binding sequences for the primary substrate or accessory site on α-thrombin. CVS#995 was synthesized using a combination of solid- and solution-phase chemistry and was purified by RP-HPLC. The purity of CVS#995 was determined to be >98% by a combination of analytical criteria that included RP-HPLC, capillary electrophoresis, FAB mass spectrometry, and quantitative amino acid analysis. The three-dimensional crystal structure of CVS#995 bound to human α-thrombin confirms the multisite binding interactions of this inhibitor with the enzyme, showing the interaction of the catalytic-site functionality, the tetrahedral transition state mimetic, and the binding of the carboxyl-terminal sequence to the anion-binding exosite of thrombin. CVS#995, therefore, is a synthetic thrombin inhibitor that couples the stability to proteolytic degradation by the potent transition state intermediate with the selectivity resulting from the use of accessory binding sites on thrombin. CVS#995 is a potent inhibitor of amidolytic substrate hydrolysis by α-thrombin (Ki=2 pmol/L) (G.P. Vlasuk, unpublished observations). It was shown that the thrombin-induced clotting of purified fibrinogen could be blocked by CVS#995 in a dose-dependent fashion. In addition, the ability of CVS#995 to inhibit fluid-phase and clot-associated thrombin activity was demonstrated in a system of citrated plasma and thrombin-induced clot formation, as described previously.9 CVS#995 initially inhibited both fluid-phase and clot-associated thrombin activity in a time-dependent fashion, followed by a stable level of inhibition. Thrombin-induced platelet aggregation could be blocked by CVS#995 at concentrations of 20 nmol/L or higher. As expected, CVS#995 was absolutely specific for α-thrombin with no inhibition of plasmin, rTPA, or any other serine protease involved in blood coagulation or fibrinolysis, seen at concentrations >100 000-fold in excess of that required to fully inhibit α-thrombin.
Because our aim was to study the effect of CVS#995 in a rabbit model, we studied the effects of CVS#995 on thrombin-induced formation of rabbit fibrin and aggregation of rabbit platelets. The results are provided in Table 1⇓ and indicate that CVS#995 is also effective in inhibiting thrombin-induced formation of fibrin and platelet aggregation in rabbits.
rHir (CGP 39393) with a specific activity of 115 000 ATU/mg was kindly provided by Dr R.B. Wallis, CIBA-GEIGY, Horsham, UK, and obtained from CIBA-GEIGY, Basel, Switzerland.
Hirulog-1 is a synthetic peptide14 that was produced by conventional solid-phase chemistry and characterized as described above for CVS#995.
rTAP was also obtained from Corvas International, San Diego, Calif, and prepared as described previously.26
The LMWH used in this study was Fraxiparin, with a specific activity of 25 000 anti–factor Xa IU/mL, purchased from Sanofi, Paris, France.
All study agents were dissolved in saline.
Seven study groups consisting of eight rabbits per group were investigated. The seven study groups were given either (1) CVS#995 administered at a high dose of 1.0 mg/kg bolus injection followed by the continuous infusion of 5 μg/kg per minute, (2) CVS#995 administered at a median dose of 0.5 mg/kg bolus injection followed by a continuous infusion of 5 μg/kg per minute, (3) rHir at a dose of 0.5 mg/kg bolus injection followed by a 5 μg/kg per minute continuous infusion, (4) Hirulog-1 according to the same regimen as rHir, (5) rTAP administered according to the same regimen as rHir, (6) LMWH at a dose of 40 anti-factor Xa U/kg bolus followed by a continuous infusion of 0.33 anti-factor Xa U/kg per minute, or (7) saline control, also administered as a bolus injection and continuous infusion. The animals received the loading dose followed by a continuous administration for 2 hours via the carotid artery cannula. The dosages were selected, based on previous experiments, for their comparable antithrombotic effect directly at the end of drug infusion. One of the thrombi was removed and counted directly after the cessation of the continuous infusion, whereas the contralateral thrombus was taken out 1 hour after the infusion was stopped. In another seven study groups, treated according to the same protocol but consisting of six animals per group, one of the thrombi was again removed immediately after discontinuation of study medication, but the contralateral thrombi were taken out 2 hours after cessation of study drug infusion.
The effect on the endogenous fibrinolysis was assessed in an additional experiment consisting of four groups (four rabbits each) treated with either (1) CVS#995 at a high dose of 1.0 mg/kg bolus and a continuous infusion of 5 μg/kg per minute for 2 hours, (2) CVS#995 at a median dose (0.5 mg/kg bolus plus 5 μg/kg per minute continuous infusion), (3) rHir (0.5 mg/kg bolus injection plus 5 μg/kg per minute continuous infusion), or (4) saline. In addition to this, a small number of experiments(n=4) were performed with higher doses of rHir (ie, 1.0 mg/kg bolus injection plus 5 μg/kg per minute continuous infusion or a dose of 5.0 mg/kg bolus injection plus 25 μg/kg per minute continuous infusion). The effect of these agents on endogenous fibrinolysis was assessed by removing one thrombus directly after stopping the infusion and the contralateral thrombus 1 hour later.
Counting of the radioactivity of the thrombi to determine the effect on thrombus growth and fibrinolysis was performed by a second independent investigator.
Blood Sampling and Plasma Preparation
Blood samples were drawn from the carotid cannula before the administration of the study medication and 60, 120, and 180 minutes thereafter. Blood samples (9 vol) were collected in 3.2% citrate (1 vol) for determination of the aPTT, PA activity, and PAI-1 activity. Platelet-poor plasma was obtained by immediate centrifugation at 1600g for 20 minutes at 4°C and stored at −70°C until assay.
To allow us to assess the pharmacokinetics of CVS#995 and rHir, an additional six rabbits received an intra-arterial bolus injection of either CVS#995 or rHir at a dose of 3 mg/kg, respectively, and a group of six rabbits received a bolus injection of one of these compounds at a dose of 1 mg/kg. Blood samples (9 vol) to assess plasma levels of CVS#995 or rHir were collected in 3.2% citrate (1 vol) that were drawn before and 2, 5, 10, 15, 20, 30, 40, 60, 120, and 240 minutes afterward. Platelet-poor plasma was obtained by centrifugation at 1600g for 20 minutes at 4°C. Plasma levels of CVS#995 and rHir were determined by measuring the inhibition of purified human α-thrombin amidolytic activity in diluted samples of plasma. The plasma levels are expressed in nmol/L.
The aPTT was determined by standard methods on a MLA 900C apparatus using Actin FS as a reagent. PA activity was measured by amidolytic assay.27 Briefly, 25 μL of plasma was mixed to a final volume of 250 μL with 0.1 mol/L Tris·HCl, pH 7.5, 0.1% (vol/vol) Tween-80, 0.3 mmol/L S-2251 (Chromogenix), 0.13 mol/L plasminogen (Chromogenix), and 0.12 mg/mL CNBr-digested fibrinogen fragments (TPA stimulator, Chromogenix). PA activity was assessed in these samples by spectrophotometric measurement. PAI-1 activity was measured with an amidolytic method, as described previously.28 Briefly, plasma samples were incubated with a fixed excess of TPA (40 IU/mL) for 10 minutes at room temperature. The residual TPA activity was determined by incubation with 0.13 μmol/L plasminogen (Chromogenix, 0.12 mg/mL) CNBr fragments of fibrinogen (TPA stimulator, Chromogenix), and 0.1 mmol/L S-2251 (Chromogenix). The PAI-1 activity in the sample was inversely proportional to the plasmin generated in the incubation mixture, determined by the conversion of the chromogenic substrate. Results are expressed in IU, where 1 IU is the amount of PAI-1 that inhibits 1 IU TPA (first international standard of the World Health Organization).
To assess the plasma levels CVS#995 and rHir, a chromogenic assay was used, measuring the inhibition of purified human α-thrombin in diluted samples of plasma. Briefly, the assay was conducted by combining in appropriate wells of a Corning microtiter plate 50 μL of HBSA (10 mmol/L HEPES, pH 7.5, 150 mmol/L sodium chloride, 0.1% bovine serum albumin), 50 μL of the citrated plasma sample diluted 1:1000 in HBSA or standard inhibitor diluted in HBSA (or HBSA alone for Vo measurement), and 50 μL of purified human thrombin (3000 U/mg specific activity; Enzyme Research Laboratories, Inc) diluted in HBSA. After a 30-minute incubation at ambient temperature (23°C), 50 μL of the chromogenic substrate (Pefachrome TPA [CH3SO2-d-hexahydrotyrosine-glycyl-l-Arg-p-nitroaniline] reconstituted in deionized water before use; Pentapharm Ltd) was added to the wells yielding a concentration of 300 μmol/L (5× Km) and a final total volume of 200 μL. The initial velocity of chromogenic substrate hydrolysis was measured by the change in absorbance at 405 nm using a Thermo Max Kinetic Microplate Reader (Molecular Devices) over a 5-minute period in which less than 5% of the added substrate was used. The plasma levels of CVS#995 and rHir were calculated using a standard curve of purified and quantitated inhibitor made up in control, homologous citrated plasma covering a broad concentration range followed by dilution to 1:1000 with HBSA. Under these conditions, the IC50 for inhibition of purified human α-thrombin (0.25 nmol/L final concentration) by the respective inhibitor in diluted plasma was not significantly different from that in HBSA alone. The limit of inhibitor detection in this assay is 75 nmol/L.
ANOVA (for repeated measures) followed by a Newman-Keuls test was applied for statistical analysis. P<.05 was considered to be statistically significant. All values are expressed as mean±SD.
All animal studies were approved by the Institutional Review Board for Animal Experiments and were performed according to the guidelines of the American Physiological Society and the Dutch Law for Animal Experiments.
The effect of the different antithrombotic agents on thrombus growth is shown in Fig 1⇓. The administration of CVS#995 resulted in a dose-dependent reduction of thrombus growth compared with saline control. CVS#995 administered at the high dose (1.0 mg/kg bolus plus 5 μg/kg per minute continuous infusion) resulted in the most pronounced antithrombotic effect, resulting in a reduction in thrombus growth to 25.2±1.8%, whereas CVS#995 administered at the median dose (0.5 mg/kg bolus plus 5 μg/kg per minute continuous infusion) reduced thrombus growth to 32.5±1.8% compared with 47.9±2.9% thrombus growth in the saline control animals (P<.05). The antithrombotic effect of CVS#995 at the median dose was comparable to the antithrombotic effect of LMWH and rTAP (thrombus growth, 36.1±3.2% and 35.3±3.0%, respectively; P=NS). rHir and Hirulog-1 had an equal antithrombotic effect compared with CVS#995 administered at the median dose (thrombus growth, 29.8±5.6% and 31.0±2.5% versus 32.5±1.8%, respectively; P=NS) but were slightly more effective in thrombus growth reduction directly after cessation of their administration than rTAP and LMWH (P<.05; Fig 1⇓). No difference in bleeding from surgical wounds was observed between the various compounds, although this animal model has not been designed to assess the specific effect of antithrombotic agents on bleeding.
To assess the duration of the antithrombotic effect of these compounds, the extent of thrombus growth was determined 1 and 2 hours after termination of the continuous infusion of the test compound (Fig 1⇑). Thrombus growth in the saline control group significantly increased over this period, to 65.5±4.3% and 70.1±3.0%, respectively. The animals treated with LMWH (thrombus growth, 45.7±2.3% and 57.4±3.3%, respectively), rTAP (thrombus growth, 41.6±3.1% and 47.7±2.7%, respectively), rHir (thrombus growth, 35.7±2.7% and 41.6±2.4%, respectively), and Hirulog-1 (thrombus growth, 37.9±2.1% and 37.2±2.3%, respectively) all demonstrated an increased thrombus growth 1 and 2 hours after the end of the infusion, although to a lesser extent compared with saline (Fig 1⇑). In accordance with earlier observations, rHir and Hirulog-1 were shown to have a significantly higher sustained antithrombotic effect after their discontinuation than LMWH and rTAP. In contrast, CVS#995 at both doses not only prevented thrombus growth completely after the end of its administration but also decreased the amount of labeled fibrinogen present at the preformed clot significantly. CVS#995 administered at median dose reduced the amount of accreted fibrinogen from 32.5±1.8% at t=120 to 27.8±2.9% and 27.3±2.3% at t=180 and t=240, respectively (P<.05), whereas the higher dose was shown to have a similar thrombus-reducing effect from 25.2±1.8% at t=120 to 19.3±2.1% and 20.3±1.9% at t=180 and t=240, respectively (P<.05). The effect on reducing thrombus size appeared to level off 1 hour after stopping the administration since the amount of accreted fibrinogen 2 hours after the end of the infusion did not significantly differ from the amount measured 1 hour earlier.
All test compounds induced a very modest and comparable prolongation of the aPTT (Table 2⇓), but no significant difference in prolongation between the different antithrombotic agents was observed.
Since the administration of CVS#995 not only prevented thrombus growth completely but also reduced the amount of fibrinogen already accreted onto the clot, the effect of CVS#995 on the endogenous fibrinolysis was assessed in an additional study. CVS#995, administered at the median dose, resulted in an enhanced endogenous fibrinolysis of 14.4±1.9% at t=120 and 21.8±2.1% at t=180 compared with 6.9±1.2% and 10.6±1.1% for rHir and 3.6±1.4% and 5.9±0.8% for saline, respectively (P<.05; Fig 2⇓). CVS#995, administered at the higher dose, did not improve the extent of endogenous fibrinolytic effect any further (thrombolysis, 14.8±2.4% at t=120 and 19.7±2.8% at t=180; P=NS versus the median dose; Fig 2⇓), indicating that the thrombolytic effect of CVS#995 was already maximal at the median dose. To observe whether the administration of higher doses of rHir might result in a similar increase in endogenous fibrinolysis, an additional series of rabbits (n=4) were studied. However, the administration of rHir at a dose of 1.0 mg/kg bolus plus 5 μg/kg per minute or at a dose of 5.0 mg/kg bolus plus 25 μg/kg per minute did not induce a further increase in thrombolysis (thrombolysis at t=120, 6.4% and 7.0%, respectively; thrombolysis at t=180, 9.5% and 10.9%, respectively).
To assess whether the enhanced extent of endogenous fibrinolysis could be explained by a systemic activation of the fibrinolytic system, the plasma levels of PA- and PAI-1 activity were determined in the animals treated with CVS#995, rHir, and saline. PA activity levels remained below the detection limit in all samples (data not shown), and although a statistically significant difference in PAI-1 activity levels was found between the animals treated with either CVS#995 or rHir compared with the animals receiving saline 1 hour after the infusion was started, no difference in PAI-1 levels was observed between the CVS#995- and rHir-treated animals during the entire experiment (Table 3⇓).
To determine whether the difference in sustained antithrombotic effect between CVS#995 and rHir could be explained by a difference in elimination, the half-life of both compounds at two different doses was assessed in an additional study. The results are presented in Fig 3⇓. No difference in half-life was observed between the two compounds at a dose of 1 and 3 mg/kg, respectively.
Clot-bound thrombin, which has been shown to be inaccessible to the heparin–antithrombin III complex, has been suggested to be partially responsible for recurrent thrombosis after discontinuation of heparin or LMWH treatment and the relative ineffectiveness of heparin in the prevention of rethrombosis after thrombolytic therapy or angioplasty.4 7 8 Experimental and clinical studies with hirudin and hirudin analogues revealed that these antithrombin III–independent direct thrombin inhibitors not only prevented rethrombosis after thrombolytic therapy and angioplasty more effectively than heparin but also had a sustained antithrombotic effect after the end of its administration in a model of experimental venous thrombosis.7 10 11 17 18 19 20 In accordance with these earlier observations, in the present study the antithrombin III–independent direct thrombin inhibitors were shown to have a more intense and sustained antithrombotic effect than LMWH treatment. Recent studies have suggested that Hirulog-1 is able to inhibit clot-bound thrombin more effectively than rHir because of its greater accessibility to this pool of thrombin due to its smaller molecular size15 but also showed that Hirulog-1 is cleared more rapidly from the circulation than rHir because of its higher susceptibility to proteolytic degradation.16 Despite these differences, in the present study both agents had an equal antithrombotic effect during and after their administration. Directly after cessation of its administration, CVS#995, at the median dose, appeared to have a comparable antithrombotic effect to rHir and Hirulog-1 but had a significantly more pronounced and sustained antithrombotic effect after cessation of the infusion. One hour after discontinuation of its administration, CVS#995 not only completely blocked the accretion of radiolabeled fibrinogen but also significantly decreased thrombus size, suggesting a facilitated endogenous fibrinolysis. In contrast, after the discontinuation of either rHir or Hirulog-1, a moderate increase in fibrinogen accretion was demonstrated. Two hours after cessation of the drug administration, no further change in thrombus size was observed in the CVS#995-, rHir–, and Hirulog-1–treated animals. These observed differences between CVS#995 and the other direct thrombin inhibitors are probably due to a combination of greater accessibility to clot-bound thrombin compared with rHir and resistance to proteolytic inactivation compared with Hirulog-1. It should be noted that the pharmacokinetic properties of CVS#995 are similar to those of Hirulog-1 and that the circulating β-phase elimination half-life of CVS#995 were comparable to the half-life of rHir (G.P. Vlasuk GP, unpublished observations). In accordance, in the present study, no difference in elimination half-life was observed between CVS#995 and rHir. Therefore, the difference in pharmacological effect between CVS#995 and the other agents cannot be explained by an extended half-life of this compound, resulting in a more prolonged maintenance of plasma levels. However, a difference in tissue saturation between the two agents cannot be excluded despite the comparable molecular size of these agents. rTAP, which inhibits factor Xa independent of antithrombin III, was more effective than LMWH in the prevention of thrombus growth after its discontinuation but appeared not to be as effective as CVS#995, rHir, or Hirulog-1 in the dosages studied. Apparently, selective inhibition of factor Xa did not block fibrin deposition as efficiently as inhibition of thrombin. It should be realized, however, that the thrombin-induced clots in this model are relatively thrombin rich and could therefore be more sensitive to thrombin inhibition than inhibition of factor Xa. In addition, careful and extensive comparisons between CVS#995 and other antithrombin agents at a wide range of doses should be made before final conclusions can be reached.
The experiments, which were designed to measure the extent of endogenous fibrinolysis, suggested that thrombin inhibition by either CVS#995 or rHir results in a significant enhancement of the endogenous fibrinolysis compared with saline. CVS#995 appeared to induce a significantly stronger enhancement of the extent of endogenous fibrinolysis than rHir, at least at the doses studied. These data appear to explain the observed reduction in thrombus size seen in the fibrin-accretion experiments with CVS#995 compared with rHir. Experiments with higher doses of rHir revealed that this treatment did not induce an additional increase in endogenous thrombolysis. Although these observations suggest that stable thrombin inhibition enhances the endogenous fibrinolysis, the mechanism by which this effect is obtained remains unclear. In vitro experiments have demonstrated that incubation of endothelial cells with thrombin induces a significant elevation of PAI-1 antigen levels in the supernatant and increases the PAI-1 mRNA levels in these cells.29 30 Therefore, we hypothesized that firm thrombin inhibition might stimulate the endogenous fibrinolysis by lowering PAI-1 secretion of endothelial cells and possibly by inhibiting the release of PAI-1 from platelets. However, other than a significant elevated PAI-1 level in the saline-treated group 1 hour after starting the infusion, no differences in PAI-1 activity levels were observed among the different groups in the systemic circulation. The PA activity level remained below the detection limit in all samples. These observations, however, do not rule out potential differences in PA and PAI-1 activity levels at the local level. As alternative explanations for the difference between CVS#995 and rHir on endogenous fibrinolysis, it can be hypothesized that CVS#995 may exert some distinct, yet unidentified, effects on coagulation or fibrinolytic factors or may evoke a direct effect on endothelial cells. Another possible mechanism by which thrombin inhibition could enhance the endogenous clot lysis might be the inhibition of fibrin cross-linking, which is necessary for stable clot formation. In vitro studies have demonstrated that other than the fibrin/fibrin cross-linking, cross-linking of α2-antiplasmin to fibrin monomers is essential for thrombolysis resistance of fibrin clots.31 32 These processes are dependent on the activation of factor XIII to factor XIIIa by thrombin,33 34 and strong inhibition of thrombin might therefore attenuate this process, resulting in clot instability and enhanced endogenous clot lysis. Another explanation for enhanced fibrinolysis is the inhibition of thrombus growth through the stable attenuation of thrombin-mediated fibrin formation. This inhibition of clot extension might shift the hemostatic balance toward fibrinolysis, resulting in an overall reduction in clot size.
In conclusion, the present study in a venous thrombus growth model in rabbits demonstrated that in contrast to rTAP and LMWH, selective thrombin inhibition induces a sustained antithrombotic effect that is most likely explained by neutralization of clot-bound thrombin. Thrombin inhibition by CVS#995 not only was shown to have a strong sustained antithrombotic effect but also appeared to enhance the endogenous fibrinolysis and might therefore be an interesting antithrombotic agent for the treatment of both venous and arterial thrombotic disorders. Although it has been shown that there is good correlation between the results obtained in the rabbit jugular vein thrombosis model and animal models of arterial thrombosis, it should be realized that the results in this study were obtained in a model of venous thrombosis. Therefore, additional comparative studies in different animal models and with different doses of the various compounds need to be performed to further substantiate differences within the growing group of new antithrombotics.
Selected Abbreviations and Acronyms
|aPTT||=||activated partial thromboplastin time|
|PAI-1||=||plasminogen activator inhibitor–1|
|rTAP||=||recombinant tick anticoagulant peptide|
|RP-HPLC||=||reverse-phase high-performance liquid chromatography|
|rTPA||=||recombinant tissue-type plasminogen activator|
Dr Büller is the recipient of a fellowship from the Royal Netherlands Academy of Arts and Sciences. We thank Dr Dan Pearson, Michael Weinhouse, and Pureza Vallar at Corvas International Inc for supplying the CVS#995 and Hirulog-1 and Dr Howard Soule for the stimulating discussions. In addition, we would like to thank Huan Tran for his technical expertise.
- Received September 12, 1994.
- Revision received July 19, 1995.
- Accepted August 14, 1995.
- Copyright © 1996 by American Heart Association
Hull RD, Raskob GE, Hirsh J, Jay RM, Leclerc JR, Geerts WH, Rosenbloom D, Sackett DL, Anderson C, Harrison L, Geert M. Continuous intravenous heparin compared with intermittent subcutaneous heparin in the initial treatment of proximal vein thrombosis. N Engl J Med. 1986;315:1109-1114.
GISSI-2. A factorial randomized trial of alteplase versus streptokinase and heparin versus no heparin among 12,490 patients with acute myocardial infarction. Lancet. 1990;1:65-71.
Agnelli G, Renga C, Weitz JI, Nenci GG, Hirsh J. Sustained antithrombotic activity of hirudin after its plasma clearance: comparison with heparin. Blood. 1992;80:960-965.
Weitz JI, Hudoba M, Massel D, Maraganore J, Hirsh J. Clot-bound thrombin is protected from inhibition by heparin-antithrombin III but is susceptible to inactivation by antithrombin III-independent inhibitors. J Clin Invest. 1990;86:385-391.
van den Bos AA, Deckers JW, Heyndrickx GR, Laarman GJ, Suryapranata H, Zijlstra F, Close P, Rijnierse JJMM, Büller HR, Serruys PW. Safety and efficacy of recombinant hirudin (CGP 39393) versus heparin in patients with stable angina undergoing coronary angioplasty. Circulation. 1993;88:2058-2066.
Heras MH, Chesebro JH, Penny WJ, Kent RB, Badimon L, Fuster V. Effects of thrombin inhibition on the development of acute platelet-thrombus deposition during angioplasty in pigs: heparin versus recombinant hirudin, a specific thrombin inhibitor. Circulation. 1989;79:657-665.
Waxman L, Smith DE, Arcuri KE, Vlasuk GP. Tick anticoagulant peptide (TAP) is a novel inhibitor of blood coagulation factor Xa. Science. 1990;248:593-596.
Neeper MP, Waxman L, Smith DE, Schulman CA, Sardana M, Ellis RW, Schaffer LW, Siegl PKS, Vlasuk GP. Characterisation of recombinant tick anticoagulant peptide: a highly selective inhibitor of blood coagulation factor Xa. J Biol Chem. 1990;265:17746-17752.
Yao SK, McNatt J, Anderson HJ, Eidt J, Cui KX, Golino, Glas-Greenwalt P, Maraganore J, Buja LM, Willerson JT. Thrombin inhibition enhances tissue-type plasminogen activator-induced thrombolysis and delays reocclusion. Am J Physiol. 1992;262:374-379.
Schaffer LW, Davidson JT, Vlasuk GP, Siegl PKS. Antithrombotic efficacy of recombinant tick anticoagulant peptide: a potent inhibitor of coagulation factor Xa in a primate model of arterial thrombosis. Circulation. 1991;84:1741-1748.
Sitko GR, Ramjit DR, Stabilito II, Lehman D, Lynch JJ, Vlasuk GP. Conjunctive enhancement of enzymatic thrombolysis and prevention of thrombotic reocclusion with selective factor Xa inhibitor, tick anticoagulant peptide. Circulation. 1992;85:805-815.
Vlasuk GP, Ramjit D, Fujita T, Dunwiddie CT, Nutt EM, Smith DE, Shebuski RJ. Comparison of the in vivo anticoagulant properties of standard heparin and the highly selective factor Xa inhibitors antistasin and tick anticoagulant peptide (TAP) in a rabbit model of venous thrombosis. Thromb Haemost. 1991;65:257-262.
Nurmohamed MT, Rosendaal FR, Büller HR, Dekker E, Hommes DW, Vanden Broucke JP, Briët E. Low molecular weight heparin versus standard heparin in general and orthopaedic surgery: a meta-analysis. Lancet. 1992;1:152-156.
Prandoni P, Lensing AWA, Büller HR, Carta M, Cogo A, Vigo M, Casara D, Ruol A, ten Cate JW. Comparison of subcutaneous low-molecular weight heparin with intravenous standard heparin in proximal deep-vein thrombosis. Lancet. 1992;1:441-445.
Vlasuk GP, Vallar PL, Weinhouse MI, Bergum PW, Tran HS, Weitz JI, Tulinsky A, Krishnnan R, Rote WE, Oldeschulte GL, Pearson DA. A novel inhibitor of thrombin containing multiple recognition sequences linked by a α-keto amide transition state mimetic. Circulation. 1994;90(suppl I):I-348. Abstract.
Francis CW, Marder VJ. Increased resistance to plasmic degradation of fibrin with highly cross-linked α-polymer chains formed at high factor XIII concentrations. Blood. 1988;71:1361-1365.
Van Giezen JJJ, Minkema J, Bouma BN, Jansen JWCM. Crosslinking of α2-antiplasmin to fibrin is a key factor in regulation blood clot lysis: species difference. Blood Coagul Fibrinol. In press.
Sakata Y, Aoki N. Cross-linking of α2-plasmin inhibitor to fibrin by fibrin-stabilizing factor. J Clin Invest. 1980;65:290-297.
Witting WI, Bourdon P, Brezniak DV, Maraganore JM, Fenton JW. Thrombin-specific inhibition by and slow cleavage of Hirulog-1. Biochem J. 1992;283:737-743.