Pharmacological characterization of a novel factor Xa inhibitor, FXV673.

FXV673 is a novel, potent, and selective factor Xa (FXa) inhibitor. FXV673 inhibited human, dog, and rabbit FXa with a K(i) of 0.52, 1.41, and 0.27 nM, respectively. FXV673 also displayed excellent specificity toward FXa relative to other serine proteases. It showed selectivity of more than 1000-fold over thrombin, activated protein C (aPC), plasmin, and tissue-plasminogen activator (t-PA). FXV673 prolonged plasma activated partial thromboplastin time (APTT) and prothrombin time (PT) in a dose-dependent fashion. In the APTT assays, the concentrations (microM) required for doubling coagulation time were 0.41 (human), 0.65 (monkey), 1.12 (dog), 0.25 (rabbit), and 0.80 (rat). The concentrations (microM) required in the PT assays were 1.1 (human), 1.32 (monkey), 2.31 (dog), 0.92 (rabbit), and 1.69 (rat). A coupled-enzyme assay was performed to measure thrombin activity following prothrombinase conversion of prothrombin to thrombin. FXV673 showed IC(50)s of 1.38 and 2.55 nM, respectively, when artificial phosphatidylserine/phosphatidylcholine (PS/PC) liposomes or fresh platelets were used as the phospholipid source for prothrombinase complex formation. It was demonstrated that FXV673 could inhibit further thrombin generation in the prothrombinase complex using PS/PC liposomes. FXV673 dose-dependently prolonged the time to vessel occlusion and inhibited thrombus formation in well-characterized canine models of thrombosis. Interspecies extrapolation (approximately 2.5-fold higher sensitivity for FXa inhibition in human than in dog) suggested that 100 ng/ml of FXV673 would be an effective plasma concentration for clinical studies. Currently FXV673 is undergoing clinical studies to be developed as an antithrombotic agent.

In vitro characterization of a novel factor Xa inhibitor, RPR 130737.

RPR 130737 inhibited factor Xa (FXa) with a Ki of 2.4 nM and also displayed excellent specificity toward FXa relative to other serine proteases. It showed selectivity of more than 1000-fold over thrombin, activated protein C, plasmin, tissue-plasminogen activator and trypsin. RPR 130737 prolonged plasma activated partial thromboplastin time and prothrombin time in a dose-dependent fashion. In the activated partial thromboplastin time assay, the concentrations required for doubling coagulation time were 0.32 microM (human), 0.61 microM (monkey), 0.44 microM (dog), 0.15 microM (rabbit), and 0.82 microM (rat). The concentrations required to double prothrombin time were 0.86 microM (human), and 1.26 microM (monkey), 1.15 microM (dog), 0.39 microM (rabbit) and 7.31 microM (rat). Kinetic studies revealed that RPR 130737 was a fast binding, reversible and competitive inhibitor for FXa when Spectrozyme FXa, a chromogenic substrate, was used. A coupled-enzyme assay measuring thrombin activity following prothrombinase conversion of prothrombin to thrombin indicated that RPR 130737 was a potent inhibitor for prothrombinase-bound FXa. In this assay, RPR 130737 showed IC50s of 17 nM and 35.9 nM, respectively when artificial phosphatidylserine/phosphatidylcholine (PS/PC) liposomes or gel-filtered platelets were used as the phospholipid source. An FX-deficient plasma clotting-time correction assay further demonstrated that RPR 130737 was a specific inhibitor of FXa. RPR 130737 showed no effect on platelet aggregation in vitro. These results indicate that RPR 130737 has the potential to be developed as an antithrombotic agent based on its potent and selective inhibitory effect against FXa.

Pharmacodynamic activity and antithrombotic efficacy of RPR120844, a novel inhibitor of coagulation factor Xa.

These studies were designed to examine the pharmacodynamic profile and antithrombotic efficacy of RPR120844, a competitive inhibitor of coagulation factor Xa, with a K(i) of 7 nM against human factor Xa. In vitro, RPR120844 doubled activated partial thromboplastin time (APTT) at concentrations of 1.54, 1.48, and 0.74 microM in plasma obtained from humans, dogs, and rats, respectively. Intravenous bolus administration of RPR 120844 at 0.3, 1, and 3 mg/kg to rats resulted in maximal increases in APTT of 1.8-, 2.6-, and 8.4-fold over baseline, respectively. The effect on prothrombin time (PT) was less pronounced, resulting in a 4.4-fold increase at 3 mg/kg. These effects were rapidly reversible; APTT and PT returned to control values by 30 min after dosing. Intragastric administration to rats at 50, 100, and 200 mg/kg resulted in modest increases in APTT and PT of 1.5- and 1.3-fold over baseline at the highest dose. Plasma levels were estimated by anti-Xa activity by using an amidolytic, chromogenic assay. Plasma levels were 0.65, 1.29, and 2.45 microM at 30 min after dosing at 50, 100, and 200 mg/kg, respectively. Intravenous administration to dogs at 0.1 and 0.3 mg/kg produced maximal increases in APTT of 1.7- and 2.4-fold over baseline, respectively. Intragastric administration to dogs at 50 mg/kg resulted in maximal increases in APTT and PT of 1.7- and 1.1-fold over baseline, with peak plasma levels of 3.9 microM observed at 15 min after dosing. In a rat model of FeCl2-induced carotid artery thrombosis, RPR120844 (3 mg/kg, i.v. bolus + 300 microg/kg/min constant infusion; n = 4) significantly increased time-to-occlusion from 18+/-1 min (vehicle, n = 4) to 60 min (maximal observation time) and reduced thrombus mass from 5.5 +/- 0.2 mg (vehicle) to 1.4 +/- 0.2 mg. These results indicate that RPR120844 is a potent, selective inhibitor of Xa that exhibits oral activity and is efficacious in a standard model of arterial thrombosis.

Design and structure-activity relationships of potent and selective inhibitors of blood coagulation factor Xa.

The discovery of a series of non-peptide factor Xa (FXa) inhibitors incorporating 3-(S)-amino-2-pyrrolidinone as a central template is described. After identifying compound 4, improvements in in vitro potency involved modifications of the liphophilic group and optimizing the angle of presentation of the amidine group to the S1 pocket of FXa. These studies ultimately led to compound RPR120844, a potent inhibitor of FXa (K(i) = 7 nM) which shows selectivity for FXa over trypsin, thrombin, and several fibrinolytic serine proteinases. RPR120844 is an effective anticoagulant in both the rat model of FeCl(2)-induced carotid artery thrombosis and the rabbit model of jugular vein thrombus formation.

Sulfonamidopyrrolidinone factor Xa inhibitors: potency and selectivity enhancements via P-1 and P-4 optimization.

Sulfonamidopyrrolidinones were previously disclosed as a selective class of factor Xa (fXa) inhibitors, culminating in the identification of RPR120844 as a potent member with efficacy in vivo. Recognizing the usefulness of the central pyrrolidinone template for the presentation of ligands to the S-1 and S-4 subsites of fXa, studies to optimize the P-1 and P-4 groups were initiated. Sulfonamidopyrrolidinones containing 4-hydroxy- and 4-aminobenzamidines were discovered to be effective inhibitors of fXa. X-ray crystallographic experiments in trypsin and molecular modeling studies suggest that our inhibitors bind by insertion of the 4-hydroxybenzamidine moiety into the S-1 subsite of the fXa active site. Of the P-4 groups examined, the pyridylthienyl sulfonamides were found to confer excellent potency and selectivity especially in combination with 4-hydroxybenzamidine. Compound 20b (RPR130737) was shown to be a potent fXa inhibitor (K(i) = 2 nM) with selectivity against structurally related serine proteinases (>1000 times). Preliminary biological evaluation demonstrates the effectiveness of this inhibitor in common assays of thrombosis in vitro (e.g. activated partial thromboplastin time) and in vivo (e.g. rat FeCl(2)-induced carotid artery thrombosis model).

RPR120844, a novel, specific inhibitor of coagulation factor Xa inhibits venous thrombosis in the rabbit.

The in vivo antithrombotic activity of RPR 20844, a novel synthetic coagulation factor Xa (fXa) inhibitor (Ki = 7 nM), was assessed by its ability to inhibit thrombus formation in a damaged segment of the rabbit jugular vein. Intravenous dose-response studies were performed and thrombus mass (TM), activated partial thromboplastin time (APTT), prothrombin time (PT), inhibition of ex vivo fXa activity and plasma drug levels (PDL) were determined. TM, measured at the end of a 50 min infusion, was significantly reduced (p<0.05 vs. saline-treated animals) by RPR120844 at 30 and 100 microg/kg/min. At doses of 10, 30 and 100 microg/kg/min, APTT was prolonged by 2.1, 4.2 and 6.1-fold, and PT was prolonged by 1.4, 2.2 and 3.5-fold, respectively. PDL were determined by measuring anti-fXa activity using an amidolytic assay. Peak PDL were 0.8+/-0.3, 1.5+/-0.9 and 2.4+/-0.6 microM, respectively. The drug effect was reversible with APTT, PT and PDL returning toward pretreatment values 30 min after termination of treatment. The results suggest that RPR 120844, or similar compounds, may provide an efficacious, yet easily reversible, means of inhibiting thrombus formation.

Design of a new class of orally active fibrinogen receptor antagonists.

The integrin receptor recognition sequence Arg-Gly-Asp was successfully used as a template from which to develop a series of potent, selective, orally active, peptide-based fibrinogen receptor antagonists with a long duration of action. Simple modifications centered on the Arg and Gly residues quickly led to a modified peptide (1) with significantly enhanced ability to inhibit in vitro platelet aggregation. Substitution of the guanidino group in 1 by piperidine provided 3, which showed not only a further increase in potency but also a modest degree of oral efficacy. Finally, exploration of the nature of the C-terminal amino acid, with respect to its side-chain functionality and the carboxy terminus, yielded a group of molecules that showed excellent in vitro potency for inhibiting platelet aggregation, excellent integrin selectivity, a high level of oral efficacy, and an extended duration of action.

Inhibition of repetitive thrombus formation in the stenosed canine coronary artery by enoxaparin, but not by unfractionated heparin.

Experiments were designed to compare the antithrombotic efficacy of enoxaparin and unfractionated heparin (UH) in a model of platelet-dependent cyclic flow reductions (CFRs) in the stenosed canine circumflex coronary artery. Low-molecular-weight heparins (LMWHs) are safe and effective in the prevention and treatment of venous thromboembolism. The present experiments were designed to evaluate the potential use of LMWHs in arterial thrombotic indications by comparing the antithrombotic effect of an LMWH with that of UH in an animal model of unstable angina. After establishment of consistent CFRs by experimentally induced vascular stenosis and damage, vehicle (saline), enoxaparin, or UH was administered intravenously as a loading dose plus a continuous infusion for 1 hour. The inhibition of CFRs was taken as an indicator of antithrombotic efficacy. Enoxaparin inhibited repetitive platelet thrombus formation in a dose-dependent manner, with significant inhibition of CFRs achieved at 0.5 mg/kg + 5 microg/kg per minute. This dose of enoxaparin resulted in anti-Xa levels of 0.9 to 1.0 IU/mL, anti-IIa levels of 0.2 to 0.3 IU/mL, activated partial thromboplastin time (APTT) of 1.3-fold over baseline, and a 1.4-fold increase (NS) in template bleeding time. Near-complete abolishment of CFRs was achieved with enoxaparin at 1.0 mg/kg + 10 microg/kg per minute. This dose of enoxaparin produced anti-Xa levels of 2 to 2.2 IU/mL, anti-IIa levels of 0.5 to 0.6 IU/mL, an increase in APTT of 1.4- to 1.5-fold over baseline, and a 1.9-fold increase (P<0.05) in template bleeding time. In contrast, UH had no significant effect on CFRs at a dose (100 U/kg + 10 U/kg per minute) that resulted in anti-Xa levels of 1.2 to 1.6 IU/mL, anti-IIa levels of 1.8 to 2.4 IU/mL, an increase in APTT greater than 10-fold over baseline, and a 2.5-fold increase (P<0.05) in template bleeding time. Compared with the vehicle group, circulating platelet count and hematocrit were not changed significantly by any dose of enoxaparin or UH tested. Enoxaparin, unlike UH, prevented repetitive platelet-dependent thrombus formation in the dog, thereby supporting the potential use of enoxaparin as a replacement for heparin in the treatment of arterial thrombotic disorders such as unstable angina.

Identification and initial structure-activity relationships of a novel class of nonpeptide inhibitors of blood coagulation factor Xa.

The discovery and some of the basic structure-activity relationships of a series of novel nonpeptide inhibitors of blood coagulation Factor Xa is described. These inhibitors are functionalized beta-alanines, exemplified by 2a. Docking experiments placing 2a in the active site of Factor Xa implied that the most expeditious route to enhancing in vitro potency was to modify the group occupying the S3 site of the enzyme. Increasing the hydrophobic contacts between the inhibitor and the enzyme in this region led to 8, which has served as the prototype for this series. In addition, an enantioselective synthesis of these substituted beta-alanines was also developed.

Comparison of enoxaparin, hirulog, and heparin as adjunctive antithrombotic therapy during thrombolysis with rtPA in the stenosed canine coronary artery.

A canine model of electrolytic injury-induced coronary artery thrombosis and rtPA-induced thrombolysis was used to evaluate the relative antithrombotic efficacy of enoxaparin (a low molecular weight heparin), conventional therapy (heparin or heparin plus aspirin), and hirulog (a direct thrombin inhibitor), when used as adjunctive therapy during thrombolysis. After 60 min of clot aging, adjunctive therapy was begun at doses which elevated APTT approximately 2-fold over baseline. Fifteen minutes after the start of adjunctive therapy, recombinant tissue plasminogen activator (rtPA) was administered (100 microg/kg i.v. bolus + 20 microg/kg/min for 60 min). Adjunctive therapy continued for 1 h after termination of rtPA and blood flow was monitored for two additional hours. Enoxaparin (1 mg/kg i.v. bolus + 30 microg/kg/min, n = 10 for each treatment group) was the only adjunctive treatment that significantly increased the total minutes of flow (143 +/- 25 min out of a possible 240 min, vs 54 +/- 25 min for vehicle, p <0.05) and decreased thrombus mass (6.0 +/- 1.3 mg vs 11.8 +/- 3.2 mg for vehicle). Although hirulog (2 mg/kg i.v. bolus + 40 microg/kg/min) did not significantly increase the minutes of flow (120 +/- 27 min, p <0.06) or decrease thrombus mass (8.7 +/- 1.7 mg) compared to vehicle, these values were not significantly different than those measured in the enoxaparin group. However, the results with hirulog were achieved at the expense of a significantly greater increase in template bleeding time than that measured during enoxaparin treatment. Minutes of flow for heparin (50 U/kg i.v. bolus + 0.6 U/kg/min) and heparin plus aspirin (5 mg/kg i.v. bolus) were 69 +/- 20 and 60 +/- 23 min, respectively; thrombus masses were 8.2 +/- 1.3 and 7.3 +/- 1.0 mg, respectively. In summary, enoxaparin was more effective than conventional therapy in this model in terms of vessel patency and thrombus mass, and was as effective as hirulog, at least at a dose of hirulog that only modestly impaired hemostasis. Therefore, enoxaparin may prove to be a safe and effective alternative agent for adjunctive therapy during thrombolysis with rtPA.

Anti-thrombotic activity of RG13965, a novel platelet fibrinogen receptor antagonist.

RG13965, a pseudotetrapeptide analogue of Arg-Gly-Asp (RGD), inhibited collagen-induced dog, monkey, human, hamster, mouse, and pig platelet aggregation in vitro with IC50 values of 3.7, 4.6, 6.3, 126, 136 and 1600 microM, respectively. RG13965 (3, 10, and 30 mg/kg, i.v.) decreased the incidence of collagen/epinephrine-induced thrombosis in mice from 90% in untreated animals to 63, 37, and 0%, respectively. In hamsters, RG13965 (10 and 30 mg/kg, i.v.) prolonged the time required for formation of a hemostatic plug in severed mesenteric arteries by 1.6- and 3.6-fold, respectively. In a canine model of repetitive platelet thrombus formation in the coronary artery, RG13965 (0.1, 0.3, and 1 mg/kg, i.v.) reversibly inhibited cyclic flow reductions (CFRs) and inhibited ADP-induced ex vivo platelet aggregation by 29, 57, and 77%, respectively. RG13965 (1 mg/kg) completely inhibited CFRs for at least 40 min. Platelet count was not altered at any dose and template bleeding time was prolonged modestly (1.8-fold) at only the highest dose. RG13965 dose-dependently and reversibly inhibited thrombus formation at doses which did not completely inhibit ex vivo platelet aggregation and only modestly prolonged template bleeding time.

Specific factor Xa inhibition reduces restenosis after balloon angioplasty of atherosclerotic femoral arteries in rabbits.

BACKGROUND: Balloon angioplasty of atherosclerotic arteries results in activation of the coagulation cascade. Several coagulation factors, including factor Xa and thrombin, are mitogenic for vascular smooth muscle cells in vitro and thus may play a role in restenosis after balloon angioplasty. Specific inhibition of factor Xa can be achieved with recombinant antistasin (rATS) or tick anticoagulant peptide (rTAP). We hypothesized that inhibition of Xa would limit restenosis after balloon angioplasty in an atherosclerotic rabbit model. METHODS AND RESULTS: Focal femoral atherosclerosis was induced by air desiccation injury and a high-cholesterol diet in 38 New Zealand White rabbits. Recombinant antistasin (n = 20 arteries) or rTAP (n = 14 arteries) was administered by intravenous bolus at the time of balloon angioplasty and followed by a 2-hour infusion; controls (n = 21 arteries) received bolus heparin alone (150 U/kg). Therapeutic prolongation of the activated partial thromboplastin time occurred, and antithrombotic drug levels were achieved in all animals. Luminal diameter in millimeters by quantitative angiography did not differ between treatment groups before (1.1 +/- 0.2 for controls, 1.1 +/- 0.2 for rATS, and 1.1 +/- 0.3 for rTAP) or after balloon angioplasty (1.5 +/- 0.3 for controls, 1.4 +/- 0.2 for rATS, and 1.4 +/- 0.2 for rTAP). At 28 days, treatment with factor Xa inhibitors tended to result in arteries with larger luminal diameter than controls (1.2 +/- 0.3 for rATS, 1.2 +/- 0.3 for rTAP versus 1.0 +/- 0.3 for control, P = .09 by one-way ANOVA). Restenosis, defined as reduction in angiographic luminal diameter (in mm) from 2 hours after angioplasty to 28 days after angioplasty was less in the rATS group than in controls (-0.2 +/- 0.1 versus -0.5 +/- 0.4, P < .001) and tended to be less in the rTAP group (-0.3 +/- 0.2 versus -0.5 +/- 0.4, P = .07). Quantitative histopathological analysis showed less percent cross-sectional area narrowing by plaque in both rATS- and rTAP-treated arteries compared with controls (42 +/- 21%, 47 +/- 18%, and 63 +/- 14%, respectively; P < .01 by one-way ANOVA). CONCLUSIONS: We conclude that a 2-hour infusion of rATS or rTAP reduced angiographic restenosis and resulted in less luminal cross-sectional narrowing by plaque compared with controls.

Maintenance of canine coronary artery patency following thrombolysis with front loaded plus low dose maintenance conjunctive therapy. A comparison of factor Xa versus thrombin inhibition.

OBJECTIVE: The aim was to examine the abilities of the direct thrombin inhibitor, recombinant hirudin (rHIR), and the coagulation factor Xa inhibitor, recombinant tick anticoagulant peptide (rTAP), given in combination with rt-PA as high dose front loading plus low dose maintenance infusions, to enhance reperfusion and maintain vessel patency in a canine model of left circumflex coronary artery stenosis and electrolytic lesion. METHODS: Occlusive coronary artery thrombosis was induced in anaesthetised dogs by electrical injury (150 microA) of the intimal surface of the vessel. Thirty minutes after occlusive thrombosis, high dose front loading infusions (45 min) of rTAP (200 micrograms.kg-1 x min-1) and rHIR (300 micrograms.kg-1 x min-1) were initiated concomitant with the start of a 90 min infusion of recombinant tissue-type plasminogen activator (rt-PA). Following the termination of front loading infusions, maintenance infusions of rTAP (10 or 20 micrograms.kg-1 x min-1) or rHIR (20 micrograms.kg-1 x min-1) were initiated and continued for the duration of the protocol (180 min after rt-PA termination). RESULTS: Reperfusion was incomplete in the rHIR group (7/9; 78%), whereas all rTAP-treated preparations reperfused (8/8 per group, aggregate 16/16; 100%). Following thrombolysis, the rHIR group had a high incidence of reocclusion, ranging from intermittent to long periods of occlusion, with only 2/7 (29%) of the preparations which initially recanalised remaining patent during the 180 min period following rt-PA termination. In contrast, 5/8 preparations in each of the two rTAP groups [aggregate 10/16; 63%] remained patent during the same period. The greater efficacy of rTAP v rHIR in maintaining vessel patency was also reflected in integrated coronary artery blood flows [91.0(SEM 5.8)% and 84.9(6.1)% of preocclusion flow in rTAP groups v 57.5(12.2)% of preocclusion flow in rHIR group], times to reocclusion [123.3(22.8) and 128.0(6.7) min in rTAP groups v 36.6(23.2) min in rHIR group; p < 0.05], and residual thrombus masses [1.8(0.3) and 2.0(0.3) mg in rTAP groups v 10.4(3.8) mg in rHIR group; p < 0.05]. CONCLUSIONS: With the present front loading plus low dose maintenance infusions designed to limit the duration of "high dose" conjunctive therapy, rTAP was more effective than rHIR at equimolar plasma concentrations in maintaining post-thrombolysis vessel patency, preserving coronary artery blood flow, and reducing residual thrombus mass. These findings further support the therapeutic potential of inhibiting factor Xa in the setting of coronary artery thrombolysis.

Expression, purification and characterization of multigram amounts of a recombinant hybrid HV1-HV2 hirudin variant expressed in Saccharomyces cerevisiae.

Hirudin (HIR), derived from leeches, and tick anticoagulant peptide (TAP) are polypeptide protease inhibitors of thrombin and coagulation factor Xa (fXa), respectively, and they have both shown utility in vitro and in vivo as potent antithrombotic agents. A thorough side-by-side comparison of the in vivo efficacy of factor Xa inhibition compared to thrombin inhibition by TAP and HIR, respectively, required purification and characterization of multigram amounts of hirudin. Therefore, a recombinant Saccharomyces cerevisiae strain was developed using a plasmid containing the gene encoding the MF alpha 1 preproleader, a synthetic hybrid HV1-HV2 HIR gene, and a galactose-inducible promoter which directed the secretion of 44 mg/liter of recombinant HIR (rHIR) after induction. rHIR was purified by a process that consisted of two chromatographic steps and decolorization. Total yield for the purification process was 3.6 g, or 41%. This process gave 59-fold purification of rHIR that was judged to be > 96% pure with regard to polypeptide content by capillary zonal electrophoresis and reversed-phase high-performance liquid chromatography. Single, unique N- and C-termini were obtained by sequencing and were identical to those predicted from the deduced sequence of the cDNA. Determination of the dissociation constant, by thrombin:hirudin inhibition reaction, and anticoagulant activity, by the activated partial thromboplastin time, demonstrated that the hybrid rHIR HV1-HV2 protein discussed in this report was essentially equipotent with rHIR preparations HV1 and HV2 reported by others.

Site-directed analysis of the functional domains in the factor Xa inhibitor tick anticoagulant peptide: identification of two distinct regions that constitute the enzyme recognition sites.

Recombinant tick anticoagulant peptide (rTAP) is a highly selective inhibitor of blood coagulation factor Xa. rTAP has been characterized kinetically as a slow, tight-binding, competitive inhibitor of the enzyme. We used an approach consisting of both recombinant, site-directed mutagenesis and solid-phase chemical synthesis to generate 31 independent mutations in rTAP to identify those regions of the molecule which contribute to the specific, high-affinity binding interaction with factor Xa. Our results demonstrate that the four amino-terminal residues of rTAP constitute the primary recognition determinant necessary for the formation of the high-affinity enzyme-inhibitor complex. The Arg residue in position three is probably not interacting with the S1-specificity pocket of factor Xa in a substrate-like manner since substitution at this position with a D-Arg amino acid produced only a modest decrease in affinity (5-fold). An additional domain in the rTAP molecule located between residues 40 and 54 was identified as a probable secondary binding determinant. Interestingly, this region in rTAP shares significant amino acid sequence homology with a sequence in prothrombin immediately amino-terminal to the factor Xa cleavage site that generates meizothrombin. These observations indicate that specific segments within two different regions of the rTAP molecule contribute to the potent binding interaction between rTAP and factor Xa.

Site-directed mutagenesis of the leech-derived factor Xa inhibitor antistasin. Probing of the reactive site.

Antistasin (ATS) is a leech-derived 119-amino-acid protein which exhibits potent and highly selective inhibition of coagulation Factor Xa. It inhibits Factor Xa according to a common mechanism of serine-proteinase inhibitors in which a conformationally rigid substrate-like reactive site is presented to the enzyme. In this study a recombinant version of ATS was expressed and purified utilizing a yeast expression system in order to probe the reactive site P1 (Arg-34) and P1' (Val-35) residues by site-directed mutagenesis. The results demonstrate the requirement for a positively charged residue in the P1 position of ATS, with an arginine residue preferred over a lysine, yielding K1 values of 61 pM and 1.28 nM respectively. Mutation of the P1 arginine residue to the non-polar amino acid leucine abolished its inhibitory potency toward Factor Xa. The role of the C-terminal domain of ATS, which shares significant amino acid sequence identity with the N-terminal domain, was investigated by creating a second reactive site in the corresponding position of the C-terminal domain. The inhibitory activity of this mutant demonstrated that the C-terminal domain of ATS is not folded into the proper conformation necessary to create a functional inhibitory domain.

Selective factor Xa inhibition by recombinant antistasin prevents vascular graft thrombosis in baboons.

A baboon model of high-shear, platelet-dependent vascular graft thrombosis was used to assess the antithrombotic effect of recombinant antistasin (rATS), a 119-amino acid protein with selective, subnanomolar inhibitory potency against coagulation factor Xa. In this model, a Dacron vascular graft segment of a femoral arteriovenous (AV) shunt provided the thrombogenic stimulus. Antithrombotic efficacy of rATS was assessed by continuous monitoring of 111In-labeled platelet and 125I-labeled fibrin(ogen) deposition onto the graft surface and blood flow through the vascular shunt. Systemic intravenous administration of rATS (2 or 4 micrograms/kg.min-1) dose dependently decreased both platelet and fibrin(ogen) deposition onto the graft. Vascular graft thrombus formation was completely inhibited at a systemic dose of rATS of 4 micrograms/kg.min-1. None of the AV shunts in animals receiving rATS at either dose occluded, and blood flow was maintained at 81 +/- 4% (2 micrograms/kg.min-1 rATS) or 96 +/- 3% (4 micrograms/kg.min-1 rATS) of basal flow. Systemic fibrinopeptide A elevations in response to exposure to the Dacron graft segment were completely suppressed by both doses of rATS. The ex vivo activated partial thromboplastin times were extended to greater than 150 seconds during infusion of both doses of rATS; however, even at fully antithrombotic doses, template bleeding times were not significantly increased. Thus, in this baboon model, rATS is a potent antithrombotic agent that inhibits both platelet and fibrin(ogen) deposition onto a Dacron vascular graft segment. Furthermore, these results demonstrate that selective inhibition of coagulation factor Xa by rATS can completely prevent vascular graft thrombus formation without significantly compromising primary hemostasis as measured by template bleeding time.

Characterization of recombinant antistasin secreted by Saccharomyces cerevisiae.

Secretion from recombinant yeast was used as a potential source of large quantities of the leech protein antistasin (ATS), a potent and highly specific inhibitor of the serine protease coagulation factor Xa. Mature recombinant ATS (r-ATS) is obtained after intracellular cleavage by the yscF protease of the mating factor alpha-1 pre-proleader from the fusion protein at the Lys-Arg sequence junction. Production levels are relatively low (ca. 1 mg/liter). Purification of the secreted product from a complex growth medium involved cell removal by microfiltration and diafiltration, cation-exchange capture and concentration on S-Sepharose Fast Flow, C-4 reverse-phase high-performance liquid chromatography (RP-HPLC), and HPLC cation-exchange chromatography step, and RP-HPLC concentration and desalting. The process was scaled up from the 16- to the 250-liter level with a corresponding increase in amount of r-ATS. From the 250-liter fermentation two major forms, r-ATS-I and r-ATS-II, distributed approximately 60:40, and a minor form, r-ATS-minor (ca. 1% of the purified r-ATS), were characterized. Limited N-terminal sequence analysis by Edman degradation indicated that r-ATS-I has the predicted mature N-terminus starting with Gln, that r-ATS-II is N-terminally blocked with pyroglutamate, and that r-ATS-minor is an incompletely processed form. RP-HPLC, hydrophilic-interaction HPLC, cation-exchange HPLC analysis, and electrophoresis results are consistent with the differences observed by sequencing. Preliminary in vitro characterization by intrinsic Ki determination for factor Xa inhibition indicated that the yeast r-ATS forms are indistinguishable from each other as well as from r-ATS expressed by the insect baculovirus host-vector system. Nevertheless, r-ATS-I and r-ATS-II appear less potent than insect-derived r-ATS in the activated partial thromboplastin time clotting assay. Further characterization indicated that C-terminal cleavage at Pro-116 had occurred in r-ATS-I and r-ATS-II as well as oxidation of methionine residues to methionine sulfoxide. The possible role of the C-terminus in inhibition of the prothrombinase complex is discussed.