Abstract
SSR182289A competitively inhibits human thrombin (K i = 0.031 ± 0.002 μM) and shows good selectivity with respect to other human proteases, e.g., trypsin (K i = 54 ± 2 μM), factor Xa (K i = 167 ± 9 μM), and factor VIIa, factor IXa, plasmin, urokinase, tPA, kallikrein, and activated protein C (all K i values >250 μM). In human plasma, SSR182289A demonstrated anticoagulant activity in vitro as measured by standard clotting parameters (EC100thrombin time 96 ± 7 nM) and inhibited tissue factor-induced thrombin generation (IC50 of 0.15 ± 0.02 μM). SSR182289A inhibited thrombin-induced aggregation of human platelets with an IC50 value of 32 ± 9 nM, but had no effect on aggregation induced by other platelet agonists. The anticoagulant effects of SSR182289A were studied by measuring changes in coagulation markers ex vivo after i.v. or oral administration in several species. In dogs, SSR182289A (0.1–1 mg/kg i.v. and 1–5 mg/kg p.o.) produced dose-related increases in clotting times. After oral dosing, maximum anticoagulant effects were observed 2 h after administration with increases in thrombin time, 2496 ± 356%; ecarin clotting time (ECT), 1134 ± 204%; and activated partial thromboplastin time (aPTT), 91 ± 20% for the dose of 3 mg/kg p.o., and thrombin time, 3194 ± 425%; ECT, 2017 ± 341%; and aPTT, 113 ± 9% after 5 mg/kg p.o. Eight hours after administration of 3 or 5 mg/kg SSR182289A, clotting times were still elevated. SSR182289A also showed oral anticoagulant activity in rat, rabbit, and macaque. Hence, SSR182289A is a potent, selective, and orally active thrombin inhibitor.
Thrombin is a multifunctional serine protease that plays an important role in thrombosis and hemostasis. Formed by proteolytic cleavage of prothrombin by the prothrombinase complex, thrombin initiates the final step of the blood coagulation cascade by cleaving soluble fibrinogen to fibrin. Thrombin also retroactivates and amplifies the coagulation cascade by feedback activation of coagulation factors V, VIII, and XI. In addition, thrombin generated at sites of vascular damage exerts cellular effects, for example, platelet aggregation, vasoconstriction, vasorelaxation, and smooth muscle cell proliferation, which further influence vascular function and blood circulation.
Preclinical and clinical studies indicate that direct inhibitors of the enzymatic activity of thrombin have significant therapeutic potential as antithrombotic agents. Recombinant hirudin has been shown to be superior to low-molecular-weight heparin for the prophylaxis of deep vein thrombosis after hip replacement surgery (Eriksson et al., 1997) and is an effective adjuvant to thrombolytic therapy in the treatment of acute myocardial infarction (Antman, 1996). The prototype small molecule thrombin inhibitor argatroban demonstrates antithrombotic activity in a variety of animal models of arterial and venous thrombosis (Bush, 1991; Berry et al., 1994; Duval et al., 1996).
Oral administration remains a major goal for the clinical development of small molecule direct thrombin inhibitors. Argatroban has poor oral bioavailability, however, recently, orally active thrombin inhibitors have been reported, for example, LB30057 (Kim et al., 1997) or L-374087 (Cook et al., 1999). Currently, the most advanced oral thrombin inhibitor is ximelagatran (formerly H376/95) (Gustafsson et al., 2001), which is a double prodrug of melagatran (Gustafsson et al., 1998). In this article, we introduce SSR182289A (Fig.1), a new thrombin inhibitor that, unlike melagatran, is orally active in its own right and does not require to be administered in the form of a prodrug. This article describes the effects of SSR182289A on the activity of thrombin, on coagulation, platelet aggregation, and thrombin generation measured in vitro, and on coagulation parameters measured ex vivo after oral or intravenous administration of SSR182289A to several animal species.
Chemical structure of SSR182289A, a new small-molecule thrombin inhibitor.
Materials and Methods
Determination of Inhibitor Constants for Human Enzymes
The inhibitor constants (K i) were determined for inhibition by SSR182289A and melagatran of a series of human enzymes.
Chromogenic substrate assays were performed using a Labsystems IEMS (Cergy Pontoise, France) microtiter plate reader and Biolise software.K i values were calculated according to the method of Dixon (1953). In each assay, the compound was tested at a minimum of seven concentrations in duplicate to obtain an inhibition curve. Two different substrate concentrations were used (four in the case of SSR182289A/thrombin). Assays were performed according to the following general procedure. In a 96-well microtiter plate, 25 μl of inhibitor solution (10 μl for cholinesterase activity assay) or buffer was added to 50 μl of substrate (100 μl for cholinesterase activity assay). A volume of 25 μl of enzyme solution (10 μl for cholinesterase activity assay) was added just before the plate was placed in the microtiter plate reader for 1 h at 37°C (30°C for cholinesterase activity assay). The hydrolysis of the substrate yields p-nitroaniline, which was continuously monitored spectrophotometrically at 405 nm. Maximal initial reaction rates were calculated and expressed as millioptical density per minute. Curve fitting (Dixon plot of 1/V maxversus inhibitor concentration) was performed by linear regression analysis to calculate the K i value.
Final concentrations of enzymes and substrates used inK i determinations are shown in Table1. Complexes between factor VIIa and its cofactor, soluble tissue factor, were obtained by mixture of factor VIIa with a 5-fold molar excess of soluble tissue factor to ensure complex formation. The mixture was incubated at 37°C for 15 min in the presence of 5 mM CaCl2 and then stored at −80°C until use.
Inhibitory potency of SSR182289A and melagatran against different human enzymes
Evaluation of Anticoagulant Activity in Vitro
The anticoagulant effects of SSR182289A were studied in human, rat, rabbit, dog, and macaque (Macaca fascicularis) plasma in vitro against the following coagulation parameters: thrombin time, ecarin clotting time (ECT), activated partial thromboplastin time (aPTT), and prothrombin time. Measurement of coagulation times in vitro was performed using citrated plasma samples obtained as follows: human (Caucasian) plasma (Instrumentation Laboratory, Paris, France); rat (Sprague-Dawley) plasma (Iffa Credo, L'Arbresle, France); rabbit (New Zealand White) plasma (Elevage Scientifique des Dombes, Chatillon, France); dog (beagle) plasma (Marshall Europe, Lyon, France); and macaque plasma (Primatology Centre, Fort Foch, France).
Coagulation times were determined using a ST4 coagulometer (Stago, Asnières, France). Plasma (human, rat, rabbit, dog, or macaque plasma; 200 μl) containing either vehicle or inhibitor (three concentrations) was incubated in a plastic cuvette for 2 min at 37°C (3 min for aPTT) containing a magnetic stirring bar. Coagulation was triggered by reagents adapted to each parameter. For measurement of thrombin time, coagulation was triggered by addition of bovine thrombin (200 μl at 12 NIHU/ml for human and rabbit plasma and 14 NIHU/ml for rat, dog, and macaque plasma). In a separate study, using SSR182289A compared with melagatran, for human plasma only, the thrombin time was also determined after addition of human thrombin (200 μl at 12 NIHU/ml). ECT measurements were performed after addition of ecarin (200 μl at 3 U/ml for rat and human plasma, 8 U/ml for rabbit plasma, 5 U/ml for dog plasma, and 4 U/ml for macaque plasma). aPTT and prothrombin time were obtained using commercially available kits (ellagic acid + calcium chloride solutions for aPTT, thromboplastin for prothrombin time) and following manufacturer's instructions (Instrumentation Laboratory).
For each parameter, coagulation times were measured in seconds and the percentage of increase measured for each sample. The EC100 value was calculated (concentration producing 100% increase in clotting time).
Inhibition of Thrombin Generation in Plasma
Continuous monitoring of endogenous thrombin generation in human plasma was performed by a method adapted from those described by Hemker et al. (1993) and Nieuwenhuys et al. (2000).
Thrombin generation experiments were carried out in defibrinated plasma obtained by mixing an aliquot of platelet-poor plasma with ancrod (50 U/ml), letting a clot form for 10 min at 37°C, and keeping the clotted plasma at 0°C for 10 min. The fibrin thus formed was discarded before thrombin generation determination. A chromogenic substrate [0.25 mM S2222 (100 μl), which is converted by thrombin sufficiently slowly and yet shows reasonable specificity for thrombin] then 100 μl of recombinant tissue factor and 100 μl of Ca2+ buffer (0.05 M Tris-HCl, 0.1 M NaCl, 100 mM CaCl2, pH 7.35, and 0.05% ovalbumin) were added to a disposable plastic microcuvette. After this, 100 μl of buffer (0.05 M Tris-HCl, 0.1 M NaCl, pH 7.35, and 0.05% ovalbumin) containing either SSR182289A or one of the reference compounds was added to the mixture. The reaction was started at zero time by adding defibrinated plasma. The reagents were prewarmed to 37°C and the cuvette was thermostatically controlled at that temperature during the measurement. The optical density at 405 nm was recorded at the rate of 10 measurements per minute using a spectrophotometer. From the obtained curve, endogenous thrombin potential (ETP) was calculated using the method described by Nieuwenhuys et al. (2000). Percentage inhibition was then calculated according to the formula (ETP control − ETP in presence of compound)/ETP control. Results were expressed as IC50 values ± S.E.M.
Inhibition of Platelet Aggregation
Platelet-Rich Plasma (PRP) Studies.
Platelet aggregation studies were performed in human and rat platelet-rich plasma. Male Sprague-Dawley rats (250 g; Charles River, L'Arbresle, France) were anesthetized with 60 mg/kg i.p. sodium pentobarbitone and 3 to 5 ml of blood was taken from the abdominal aorta after laparotomy. Human blood was collected from the cubital vein of volunteers, following ethical committee approval, at the Centre de Transfusion Sanguine (Toulouse, France). Blood was collected using 3.8% sodium citrate as anticoagulant (1 volume of citrate to 9 volumes of blood). PRP was prepared by centrifugation of blood at 500g for 10 min. The remainder was centrifuged at 1500g for 10 min to prepare platelet-poor plasma.
Platelet aggregation was measured at 37°C, according to the turbidimetric method of Born and Cross (1963), in an aggregometer (Dual Agrego-meter; Chrono-Log, Haverton, PA), after activation of PRP by various agonists such as 2.5 μM adenosine 5′-diphosphate, 200 μg/ml collagen, 0.2 U/ml thrombin for human and 0.5 U/ml for rat, 375 μM arachidonic acid, and 10 μM TRAP14 (SFLLRNPNDKYEPF). Aggregation with thrombin was performed with a PRP diluted four times in 0.9% NaCl. Drugs were added 1 min before activation, in distilled water or dimethyl sulfoxide solution at final concentration of 0.01%. The extent of aggregation was estimated quantitatively by measuring the maximum amplitude above baseline level (except for collagen aggregation where the slope of the aggregation curve was measured).
The antiaggregatory activities of SSR182289A and melagatran were calculated as percentage of inhibition of aggregation compared with the vehicle control. Results are expressed as IC50values ± S.E.M.
Washed Platelet Studies.
Blood samples (20 ml) were obtained from male New Zealand White rabbits (3.5–4.0 kg; ESD, Chatillon, France) by cannulation of a ear artery and drawn into 5-ml tubes containing 0.5 ml of sodium citrate solution as anticoagulant.
Citrated blood was centrifugated at 250g for 13 min. The PRP was drawn off and the red blood cell-containing layer was centrifugated at 1500g for a further 15 min to obtain platelet-poor plasma. The platelet counts were adjusted in the PRP to 200,000 to 300,000 platelets/μl with the aid of a cell counter (Melet Schloesing MS9; Cergy Pontoise, France). Prostacyclin (2 ng/ml final concentration) was added to the PRP, which was then centrifugated at 180g for 10 min. Prostacyclin was again added at a final concentration of 200 ng/ml and the PRP was centrifuged at 800g for 10 min. The plasma supernatant was discarded and the pellet resuspended in Tyrode's solution (135 mM NaCl, 0.05 mM MgCl2, 2.5 mM KCl, 0.41 mM NaH2PO4, 1.35 mM CaCl2, 10 mM NaHCO3, and 5 mM glucose) containing 200 ng/ml prostacyclin. The platelet suspensions were centrifuged at 800g for 10 min. The resuspension and centrifugation steps were repeated under the same conditions before the final pellet was suspended in prostacyclin-free Tyrode's solution and adjusted to 250,000/μl. The washed platelets were left for at least 2 h before use to allow total elimination of the prostacyclin.
Washed platelet suspensions (225 μl) were incubated with bovine serum albumin buffer (75 μl) in an aggregometer cuvette containing a magnetic stirring bar for 1 min at 37°C in a four-channel aggregometer (Regulest, Florange, France). The test compound or its vehicle was then incubated with the platelet suspension for a further minute before the addition of human thrombin (0.6 U/ml). The extent of the aggregation was assessed as the area under the aggregation curve (duration 5 min), calculated using the data acquisition software provided by the manufacturer of the aggregometer.
The antiaggregatory activities of SSR182289A and melagatran were calculated as percentage of inhibition of aggregation compared with vehicle control. Results are expressed as IC50values with 95% confidence limits.
Evaluation of Anticoagulant Activity Ex Vivo
Animals.
These studies were performed in the following animals: male Sprague-Dawley rats (190–250 g; Charles-River), male New Zealand White rabbits (2.5–3.0 kg; ESD), female beagle dogs (9–15 kg; Marshall Europe, Lyon, France), and male and female macaques (2–5 kg; CRP Le Vallon, Mauritius). Animal facilities, animal care, and study programs are in accordance with the principles laid down in the European Convention for the Protection of Vertebrate Animals used for Experimental and other Scientific Purposes and its appendix. Protocols were approved by the Animal Care and Use Committee of Sanofi-Synthélabo Recherche.
Blood sampling methods were as follows. Rats were anesthetized with sodium pentobarbitone (60 mg/kg i.p.) and blood (3 ml) was withdrawn from the abdominal aorta. A single blood sample was taken from each animal, requiring separate groups of animals for each time point. In the other species used (rabbit, dog, and macaque), serial blood sampling was performed enabling pre- and post-treatment measurement of coagulation parameters in the same animals. Rabbits were placed in restraining boxes to enable serial blood sampling from an indwelling catheter placed in the ear artery. In dogs, a silastic catheter (Lambert et Rivière, Fontenay-sous-Bois, France) was implanted in a jugular vein and passed subcutaneously before being exteriorized in the neck as a vascular access port. This surgical procedure was performed under general anesthesia (thiopental 25 mg/kg i.v.) and sterile conditions. A preventative antibiotic treatment was administered for 10 days after surgery and animals were allowed a minimum 2-week recovery period before starting experimental studies. Catheter patency was ensured by regular flushing with 0.9% sodium chloride solutions. During experiments, serial blood sampling was performed using the vascular access ports. In macaques, blood sampling was performed by venopuncture from femoral veins. All blood samples were taken using 3.8% trisodium citrate (1 volume of citrate for 9 volumes of blood) as anticoagulant. Blood samples were rapidly centrifuged at 1000g for 15 min and the plasmas were stored at −20°C until use.
Measurement of Coagulation Parameters.
Coagulation parameters were measured using an automated coagulation laboratory workstation (ACL 3000; Instrumentation Laboratory).
ECTs were determined by preincubation of 75 μl of citrated plasma at 37°C for 4.5 min before the addition of 75 μl of a 2 U/ml ecarin solution to initiate coagulation. aPTTs and thrombin time values were obtained from commercially available kits [ellagic acid + calcium chloride solutions for aPTT and bovine thrombin (2 NIHU/ml final concentration) for thrombin time; Instrumentation Laboratory] following the manufacturer's instructions. In all cases, clotting times were determined in seconds. If no coagulation occurred, a value of 300 s (or 167 s in primates) was ascribed for the purposes of statistical analysis.
Ex vivo coagulation parameters were measured before and at various times after administration of single oral or i.v. doses of SSR182289A in the different species studied. The doses administered were as follows: dog, 1, 3, and 5 mg/kg p.o. and 0.1, 0.3, and 1 mg/kg i.v.; rabbit, 1, 3, and 10 mg/kg p.o. and 1 mg/kg i.v.; rat, 3, 10, and 30 mg/kg p.o. and 1 mg/kg i.v.; and macaque, 1 and 5 mg/kg p.o. and 1 mg/kg i.v. For oral dosing studies SSR182289A was administered in solution in distilled water, with the exception of dog studies, where the compound was administered in a gelatin capsule followed by 10 ml of distilled water. For i.v. studies SSR182289A was administered in physiological saline.
Coagulation times were measured in seconds and all three coagulation parameters were analyzed in the same manner. For each treatment group, the mean ± S.E.M. percentage of increase in the clotting time was determined versus the pretreatment value (dog, macaque, and rabbit) or versus the control group (rat). Statistical treatment involved an analysis of variance followed by a Dunnett's test and was performed using Everstat software. Values of p < 0.05 were considered to be statistically significant.
Drugs and Reagents
The following drugs and reagents were used: S-2238 (H-d-Phe-Pip-Arg-pNA), S-2444 (pyroGlu-Gly-Arg-pNa), S-2765 (2-d-Arg-Gly-Arg-pNA), S-2288 (H-d-Ile-Pro-Arg-pNA), S-2366 (pyroGlu-Pro-Arg-pNA), S-2251 (H-d-Val-Leu-Lys-pNA), S-2302 (HD-Pro-Phe-Arg-pNA), S2222 (Bz-Ile-Glu-Gly-Arg-pNA), tPA, and plasmin were from Biogenic (Montpellier, France). CBS 34-47 (H-d-CHG-But-Arg-pNA), human thrombin, and ecarin were from Diagnostica Stago (Asnières, France). Trypsin, streptokinase, plasminogen, cholinesterase, chymotrypsin, urokinase, kallikrein, butyrylthiocholine, ancrod, human collagen, prostacyclin, and arachidonic acid were from Sigma (St. Quentin Fallavier, France). TRAP14, L-1400 (Suc-Ala-Ala-Pro-Phe-pNA), and L-1335 (MeOSuc-Ala-Ala-Pro-Val-pNA) were from Bachem (Voisins-Le-Bretonneux, France). Sodium pentobarbitone, soluble tissue factor, heparin, hirudin (rh-V2-Lys47 hirudin), and nadroparin were from Sanofi-Synthélabo (Toulouse or Strasbourg, France); adenosine 5′-diphosphate was from Roche Applied Science (Mannheim, Germany); d-cyclohexylG-3-R-pNA (SP970108B) was from Neosystem (Strasbourg, France); bovine and human thrombin were from Euromedex S.A. (Mundolshein, France); recombinant tissue factor (Recombiplastin) was from Ortho Diagnostic System (Issy-Les-Moulineaux, France); factor Xa, factor IXa, and activated protein C were from Enzyme Research Laboratories (Swansea, UK); factor VIIa was from American Diagnostica (Greenwich, CT); and SSR182289A and melagatran were synthesized by the Cardiovascular-Thrombosis Medicinal Chemistry Department, Sanofi-Synthélabo Recherche (Chilly-Mazarin, France). SR182289A was dissolved in distilled water and melagatran in either distilled water or dimethyl sulfoxide, depending on the study. Drug concentrations were calculated using the base form.
Results
Enzyme Inhibition.
The results of a chromogenic substrate assay demonstrate that SSR182289A is a potent, competitive inhibitor of human thrombin. Figure 2 shows a Dixon plot of 1/V max versus inhibitor concentration using four different substrate concentrations, giving aK i value of 31 ± 2 nM. The selectivity of SSR182289A for thrombin compared with a selection of other human enzymes is shown in Table 1. SSR182289A is highly selective for thrombin and demonstrates a selectivity ratio (based on the respective K i values) of >8000 for thrombin versus factor VIIa/tissue factor, factor IXa, activated protein C, tPA, urokinase, plasmin, kallikrein, and cholinesterase, ratios of 5387 versus factor Xa, 1741 versus trypsin, >322 versus elastase, and 18 versus chymotrypsin. Melagatran was more potent than SSR182289A as an inhibitor of thrombin (K i = 14 ± 2 nM) but generally less selective (Table 2). This was most notable for trypsin, which was inhibited at lower concentrations of melagatran than was thrombin. Melagatran also inhibited activated protein C, kallikrein, plasmin, factor Xa, and factor VIIa/tissue factor at low micromolar concentrations (i.e., selectivity ratios 65–300 versus thrombin).
Dixon plot of 1/V maxagainst concentration of SSR182289A showing SSR182289A-induced inhibition of the amidolytic activity of human thrombin. Seven different concentrations of inhibitor were tested for each of the four chromogenic substrate concentrations (S-2238, 10, 30, 100, and 300 μM). Hydrolysis of the substrate yieldsp-nitroaniline, which was monitored continuously spectrophotometrically at 405 nm. Maximal initial reaction rates were calculated and expressed as millioptical density per minute. Points are the means ± S.E.M. of three separate experiments. Linear regression analysis gives a K i value for SSR182289A of 31 ± 2 nM with an intercept above thex-axis, indicating competitive inhibition.
In vitro anticoagulant activity of SSR182289A in plasma of different species
Anticoagulant Activity in Vitro.
Table 2 shows the effects of SSR182289A on in vitro coagulation in the plasma of different species. SSR182289A produced concentration-dependent inhibition of thrombin time, ECT, aPTT, and prothrombin time in all species studied. In general, thrombin time and ECT were the coagulation parameters most sensitive to inhibition by SSR182289A. In a separate study, the effects of melagatran and SSR182289A were compared on thrombin time determined by the addition of human thrombin to human plasma. The EC100 values were 35 ± 1 nM for melagatran and 142 ± 11 nM for SSR182289A.
Inhibition of Thrombin Generation.
SSR182289A produced concentration-related inhibition of the amidolytic activity of thrombin generated in human plasma after triggering of the coagulation cascade by addition of tissue factor. Its IC50 value was 0.15 ± 0.02 μM. The comparative potency of a series of reference antithrombin compounds in this test is shown in Table3.
Effect of SSR182289A and reference anti-thrombin agents on thrombin generation in human plasma
Inhibition of Platelet Aggregation.
The effects of SSR182289A and melagatran on the aggregation of human platelets (PRPs) are shown in Table 4. Both compounds were potent and selective inhibitors of thrombin induced platelet aggregation. SSR182289A (IC50 of 32 ± 9 nM) was 4 times less potent than melagatran in this test. High concentrations of melagatran and SSR182289A had no significant effect on aggregation induced by other aggregatory stimuli (arachidonic acid, collagen, adenosine 5′-diphosphate, and TRAP).
Effect of SSR182289A and melagatran on human platelet aggregation induced by various agonists
In rat platelets (PRPs), SSR182289A (IC50 of 10 ± 2 nM; n = 4) and melagatran (IC50 of 8 ± 4 nM; n = 3) were potent inhibitors of thrombin induced platelet aggregation but had no influence on aggregation produced by adenosine 5′-diphosphate (IC50 > 10 μM).
In rabbit washed platelets, SSR182289A IC50 of 24 nM (confidence interval 18–34; n = 6) and melagatran IC50 of 6 nM (confidence interval 4–10;n = 3) were potent inhibitors of thrombin-induced aggregation.
Anticoagulant Effects ex Vivo.
Basal (absolute) values of the three coagulation markers measured (thrombin time, ECT, and aPTT) in the different species studied are contained in Table5.
Control values of coagulation parameters measured ex vivo in different species
SSR182289A demonstrated dose-related anticoagulant activity in the four species studied after oral administration. Figures3, 5,6, and 7 show the effects of different oral doses of SSR182289A on thrombin time, ECT, and aPTT in the dog, rabbit, and macaque, respectively. For the effects of SSR182289A after i.v. administration (Figs. 4 and 8) we have presented the results as changes in the thrombin time only.
Ex vivo anticoagulant effects of oral administration of SSR182289A in conscious dogs. Serial blood sampling was performed before and at different times after oral dosing with SSR182289A (1, 3, and 5 mg/kg). Clotting times (thrombin time, ECT, and aPTT) were measured from plasma samples using an automated coagulation laboratory workstation. Figures show the percentage of increases in each clotting time versus pretreatment values. Data are means ± S.E.M.,n = 4 dogs. ∗, p < 0.05 versus pretreatment values.
Ex vivo anticoagulant effects of oral administration of SSR182289A (3, 10, and 30 mg/kg) in conscious rats. A single blood sample was taken from each animal, necessitating a separate group of animals for each time point and for each dose of compound. Clotting times [thrombin time (TT), ECT, and aPTT] were measured from plasma samples using an automated coagulation laboratory workstation. Figures show the percentage of increases in each clotting times versus the time-matched vehicle-treated group. Data are means ± S.E.M.,n = 5. ∗, p < 0.05 versus vehicle group.
Ex vivo anticoagulant effects of oral administration of SSR182289A in conscious macaques. Serial blood sampling was performed before and at different times after oral dosing with SSR182289A (1 or 5 mg/kg). Clotting times [thrombin time (TT), ECT, and aPTT] were measured from plasma samples using an automated coagulation laboratory workstation. Figures show the percentage of increases in each clotting time versus pretreatment value. Data are means ± S.E.M., n = 3 (5 mg/kg) orn = 6 (1 mg/kg) primates. ∗,p < 0.05 versus pretreatment values.
Ex vivo anticoagulant effects of oral administration of SSR182289A in conscious rabbits. Serial blood sampling was performed before and at different times after oral dosing with SSR182289A (1, 3, and 10 mg/kg). Clotting times [thrombin time (TT), ECT, and aPTT] were measured from plasma samples using an automated coagulation laboratory workstation. Figures show the percentage of increases in each clotting time versus pretreatment values. Data are means ± S.E.M., n = 4 rabbits. ∗, p< 0.05 versus pretreatment values.
Ex vivo anticoagulant effects of i.v. administration of SSR182289A in conscious dogs. Serial blood sampling was performed before and at different times after i.v. dosing with SSR182289A (0.1, 0.3, or 1 mg/kg). Thrombin times (TT) were measured from plasma samples using an automated coagulation laboratory workstation. Figures show percentage of increases in thrombin time versus pretreatment values. Data are means ± S.E.M., n = 4 dogs. ∗,p < 0.05 versus pretreatment values.
Ex vivo anticoagulant effects of i.v. administration of SSR182289A 1 mg/kg in rats, rabbits, and macaques. Serial blood sampling was performed in rabbits and macaques. In rats a separate group of animals was used for each time point. Thrombin times (TT) were measured from plasma samples using an automated coagulation laboratory workstation. Figures show the percentage of increases in thrombin time versus pretreatment values (macaque and rabbit) or time-matched vehicle group (rat). Data are means ± S.E.M., n = 4 rabbits, n = 3 macaques, and n= 5 rats. ∗, p < 0.05.
In dogs, SSR182289A (0.1–1 mg/kg i.v. and 1–5 mg/kg p.o.;n = 4) produced dose-related increases in clotting times. After oral dosing, maximum anticoagulant effects were observed 2 h after administration. At this time point, the increases in clotting times were thrombin time, 2496 ± 356%; ECT, 1134 ± 204%; and aPTT, 91 ± 20% for the dose of 3 mg/kg p.o., and thrombin time, 3194 ± 425%; ECT, 2017 ± 341%; and aPTT, 113 ± 9% after 5 mg/kg p.o. Eight hours after administration of the two highest doses, clotting times were still markedly elevated. SSR182289A also demonstrated anticoagulant activity in rats (1 mg/kg i.v. and 3–30 mg/kg p.o.; n = 5) and in rabbits (1 mg/kg i.v. and 1–10 mg/kg p.o.; n = 4). In rats, thrombin time values exceeded 300 s in all treated animals after doses of 10 mg/kg p.o. (30 and 60 min after administration) and 30 mg/kg p.o. (all time points after administration). The maximum increases in clotting times obtained 30 min after administration of 10 mg/kg p.o. in the rat were thrombin time, ≥1257%; ECT, 1078 ± 292%; and aPTT, 199 ± 57%. In the rabbit at the same dose, values attained 60 min after administration were thrombin time, 1088 ± 339%; ECT, 564 ± 179%; and aPTT, 79 ± 28%. In macaques, SSR182289A 1 and 5 mg/kg p.o. (n = 3–6) increased thrombin time by 153 ± 76 and 949 ± 41%, ECT by 36 ± 12 and 332 ± 194%, and aPTT by 26 ± 7 and 125 ± 40%, respectively (E max values). Six hours after administration of 5 mg/kg p.o. SSR182289A to macaques residual anticoagulant activity was still evident (e.g., aPTT + 54 ± 13%).
Discussion
SSR182289A is the result of a rational design medicinal chemistry program for which the starting point was argatroban (Bush, 1991), a small-molecule thrombin inhibitor that possesses several good qualities. Argatroban is a potent and competitive inhibitor of the active site of thrombin, possesses excellent selectivity with respect to other key enzymes involved in coagulation or fibrinolysis, shows low plasma binding, and has demonstrated clinical efficacy as an anticoagulant agent (Jeske et al., 1999). A major limitation of argatroban is its poor oral bioavailability, which probably results from the presence of guanidine and carboxylic acid functions in the molecule. The highly basic arginine chain of argatroban is thought to form a critical interaction with the carboxylate of the Asp189 residue situated at the bottom of the S1 pocket in the active site of thrombin (Rewinkel and Adang, 1999). Our objective has been to optimize the structure of argatroban to produce a drug candidate that retains the positive features of argatroban and, in addition, demonstrates good pharmacological activity after oral administration with a long duration of action. SSR182289A fulfills these criteria.
SSR182289A inhibits the amidolytic activity of human thrombin with aK i value of 31 ± 2 nM, thus showing similar potency to argatroban for which aK i value of 39 nM has been reported (Bush, 1991). The K i value calculated for SSR182289A in our study is in good agreement with the concentration range at which SSR182289 inhibits thrombin-induced platelet aggregation in rat and human platelets (IC50 values of 10 and 32 nM, respectively) and rabbit washed platelets (IC50 of 24 nM). A key feature of the biological profile of SSR182289A is its selectivity for thrombin inhibition. High micromolar concentrations of SSR182289A did not influence aggregation induced by a variety of platelet agonists (adenosine 5′-diphosphate, collagen, TRAP, and arachidonic acid) acting through different pharmacological mechanisms. Enzyme inhibition studies demonstrate that SSR182289A possesses >5000 fold selectivity for thrombin with respect to factor Xa, factor VIIa/tissue factor, and factor IXa, the other potential serine protease targets for anticoagulant agents. Therefore, the in vitro and ex vivo anticoagulant effects of SSR182289A observed in the present study can be attributed to thrombin inhibition. Potentially more important in the context of antithrombotic drug development is the high selectivity demonstrated by SSR182289A with regard to the components of the fibrinolytic system (>8000-fold selectivity versus tPA, urokinase, and plasmin) and against the endogenous anticoagulant factor activated protein C (>8000-fold). This should ensure that the antithrombotic potential of SSR182289A is not counterbalanced by simultaneous inhibition of profibrinolytic or anticoagulant processes. Certain thrombin inhibitors, for example, DUP 714 (Callas et al., 1994), compromise fibrinolysis in animal thrombolysis models, observations linked to significant antiprotease activity against fibrinolytic enzymes. Trypsin selectivity was also a goal for SSR182289A. This has clearly been achieved, however; somewhat surprisingly, its selectivity with respect to the related protease chymotrypsin is relatively modest. The absence of inhibition of cholinesterase by SSR182289A was important to demonstrate because poor selectivity against this enzyme may be responsible for adverse cardiopulmonary effects observed with other thrombin inhibitors (Hijikata-Okunomiya and Okamoto, 1992). We have included melagatran as comparator in the biochemical studies because it is the active form of the orally active thrombin inhibitor ximelagatran, which is in advanced clinical studies. In our hands, melagatran was 2 to 4 times more potent than SSR182289A as a thrombin inhibitor when considering either amidolytic or platelet aggregatory activities but was globally less selective with respect to the other enzymes studied. The enzyme selectivity profile of melagatran in our study was similar to that reported in the literature (Gustafsson et al., 1998), particularly with respect to potent nanomolar inhibition of trypsin and effects on several other proteases at low micromolar concentrations (for example, plasmin, kallikrein, and activated protein C). Given the thrombin/plasmin selectivity ratio of 350 (178 in our study) a possible interference of melagatran with fibrinolytic processes has been investigated (Gustafsson et al., 1998). Inhibition of endogenous fibrinolysis was observed only at doses of melagatran that gave plasma concentrations above the proposed therapeutic range.
SSR182289A inhibited coagulation induced by human thrombin in human plasma with an EC100 value of 142 nM compared with a K i value for inhibition of purified enzyme obtained in a plasma-free system of 31 nM. Although these tests are not directly comparable, the relatively low ratio of the two values suggests that SSR182289A possesses plasma stability and modest binding to plasma proteins. Melagatran, which is known to have low plasma protein binding, showed a similar (slightly lower) ratio in our study (Gustafsson et al., 1998). In this context, it is also pertinent to note the high potency observed for SSR182289A to inhibit thrombin-induced platelet aggregation in human and rat platelets in the presence of plasma. In vitro, SSR 182289A inhibited coagulation in the plasmas of human, rat, dog, macaque, and rabbit. In general, the profile of inhibition was similar in all species studied, i.e., thrombin time ≥ ECT > aPTT, prothrombin time. In addition, the EC100 values for inhibition of coagulation in human plasma corresponded well with those determined in the different animal plasmas. The lower sensitivity to inhibition by SSR182289A of aPTT and prothrombin time assays is consistent with these being global coagulation tests where thrombin is continuously produced by the action of the prothrombinase complex, probably resulting in a higher thrombin concentration than in the thrombin time assay.
The thrombin generation test differs significantly from the aforementioned coagulation tests in that clotting is not the endpoint. Clotting times such as prothrombin time and aPTT measure only the initial formation of thrombin necessary for fibrin formation and, as such, do not reflect overall thrombin generation (Hemker and Béguin, 1995). Measurement of ETP in defibrinated plasma enables continuous monitoring of the amidolytic activity of thrombin generated in plasma after activation of the coagulation cascade with tissue factor. Both SSR182289A (IC50 of 150 nM) and melagatran demonstrated concentration-related inhibitory activity in the thrombin generation test performed in human plasma, melagatran being approximately 7 times more potent in this respect. Other reference antithrombin agents such as hirudin, heparin, and nadroparin (Fraxiparine, a low-molecular-weight heparin) were also active in this model (Lormeau and Herault, 1993).
Measurement of changes in ex vivo coagulation parameters provides a simple and reliable method for determining the pharmacological activity of thrombin inhibitors after i.v. or p.o. administration and determining their duration of action. We have chosen to monitor three different coagulation markers, two of which (thrombin time and aPTT) are well documented in the literature, plus ECT, a marker that seems useful for the evaluation of direct thrombin inhibitors. The ECT uses ecarin, a protease isolated from the venom of the snake Echis carinatus, which activates prothrombin to generate meizothrombin that autocatalyses to α-thrombin (Kornalik and Blombäck, 1975;Bucha and Nowak, 1995). ECT values correlate strongly with plasma concentrations of direct thrombin inhibitors in clinical and preclinical studies (Pötzsch et al., 1997; Jeske et al., 1999;Berry et al., 2000) and predict antithrombotic activity in rat models of arterial and venous thrombosis (Berry et al., 1998).
Oral or intravenous administration of SSR182289A produced anticoagulant effects (thrombin time > ECT > aPTT) in all four species studied. The high sensitivity of thrombin time to inhibition by SSR182289A resulted in thrombin time values that were beyond the measurement limit (>300 s) in certain studies. This high sensitivity is recognized as one of the limitations of thrombin time for monitoring the anticoagulant activity of direct thrombin inhibitors in the clinic (Zoldhelyi et al., 1993). In contrast, the intermediate reactivity of ECT suggests that it is well adapted for this purpose. These studies provide a clear demonstration of the oral potency and duration of action of SSR182289A. In the dog and the macaque, significant anticoagulant effects were observed after administration of 1 mg/kg p.o. SSR182289A, the lowest dose tested in these species. Overall, the dog showed the most marked anticoagulant response after oral administration of SSR182289A with maximum pharmacological activity observed 2 h after drug administration and with a substantial anticoagulant effect still present 8 h after dosing. By way of comparison, we have calibrated our dog ex vivo coagulation model using a highly bioavailable, long-acting subcutaneous formulation of argatroban (Berry et al., 2000). In this model, a dose of argatroban (2 mg/kg s.c.) that gave plasma levels within the clinical anticoagulant range (0.04–2.5 μg/ml; Herrman et al., 1996; Bergougnan et al., 1997; Jeske et al., 1999) was associated with maximum increases in thrombin time, ECT, and aPTT of 367, 203, and 25%, respectively, i.e., a much more modest level of anticoagulation than achieved after oral dosing with SSR182289A. The demonstration that SSR182289A is orally active as an anticoagulant agent with a long duration of action in four animal species gives grounds for optimism with respect to the oral bioavailability of this compound in human.
In conclusion, we have shown that SSR182289A is a new, small-molecule thrombin inhibitor that demonstrates good potency and selectivity in vitro and shows marked and long-lasting anticoagulant activity after oral administration to several animal species.
Acknowledgments
We are grateful for technical and administrative assistance from the staff of Cardiovascular-Thrombosis Department (Chilly-Mazarin and Toulouse) and Toxicology Department (Montpellier and Porcheville).
Footnotes
- Received July 3, 2002.
- Accepted July 23, 2002.
DOI: 10.1124/jpet.102.040667
Abbreviations
- SSR182289A
- N-[3-[[[(1S)-4-(5-amino-2-pyridinyl)-1-[[4-difluoromethylene)-1-piperidinyl]carbonyl]butyl]amino] sulfonyl][1,1′-biphenyl]-2-yl], acetamide hydrochloride
- ECT
- ecarin clotting time
- aPTT
- activated partial thromboplastin time
- ETP
- endogenous thrombin potential
- PRP
- platelet-rich plasma
- tPA
- tissue plasminogen activator
- TRAP
- thrombin receptor activator peptide (SFLLRNPNDKYEPF)
- NIHu
- National Institutes of Health units
- LB30057
- 4-[(E)-amino(hydrazono)methyl]-N-cyclopentyl-N-methyl-N-(2-naphthylsulfonyl)-l-phenylalaninamide
- L-374087
- 3-benzylsulfonylamino-6-methyl-1-(2-amino-6-methyl-5-methylenecarboxamidomethylpyridinyl)-2-pyridinone
- The American Society for Pharmacology and Experimental Therapeutics
