International Journal of Biological Macromolecules 78 (2015) 296–303
Contents lists available at ScienceDirect
International Journal of Biological Macromolecules journal homepage: www.elsevier.com/locate/ijbiomac
Ecotin: Exploring a feasible antithrombotic profile Luciana Serrão Wermelinger a,b , Flávia Serra Frattani a,b , Tatiana Correa Carneiro-Lobo b , Charles S. Craik c , Helena Carla Castro d , Russolina Benedeta Zingali b,∗ a
Departamento de Análises Clínicas e Toxicológicas, Faculdade de Farmácia/CCS, Universidade Federal do Rio de Janeiro, Rio do Janeiro, RJ, Brazil Instituto de Bioquímica Médica Leopoldo de Meis/CCS, Universidade Federal do Rio de Janeiro, Rio do Janeiro, RJ, Brazil California Institute of Quantitative Biosciences (QB3), University of California, San Francisco, San Francisco, CA, USA d Laboratório de Antibióticos, Bioquímica e Modelagem Molecular (LABioMol), Departamento de Biologia Celular e Molecular, IB-/CEG, Universidade Federal Fluminense, Niterói, RJ, Brazil b c
a r t i c l e
i n f o
Article history: Received 30 December 2014 Received in revised form 20 March 2015 Accepted 30 March 2015 Available online 13 April 2015 Keywords: Ecotin Thrombin inhibitor Factor Xa inhibitor
a b s t r a c t Ecotin is an Escherichia coli-derived protein that can inhibit serine proteases. It has been suggested that this protein (ecotin-WT) and some of its variants could be used to develop a prototype to treat thrombosis. In this work, the effect of ecotin-WT and a variant of this protein harboring two mutations (Met84Arg and Met85Arg, ecotin-RR) were analyzed to determine their ability to prevent thrombus formation using in vivo models. Ecotins were analyzed in vitro using the coagulation tests. An in vivo venous thrombosis rat model and a pulmonary thromboembolism mouse model were used to investigate the antithrombotic activity. The bleeding time in rats using a tail-transection model was evaluated as a possible side effect caused by the administration of these proteins. Ecotin-RR was more effective in inhibiting thrombin than ecotin-WT. Both ecotins presented similar mechanisms of anticoagulation activity and were able to decrease thrombus formation. In contrast, only ecotin-RR increased survival rates in the in vivo pulmonary thromboembolism model, reinforcing the antithrombotic activity of ecotin-RR. Ecotin-WT and more so ecotin-RR showed potent antithrombotic effects, not associated with bleeding. The presented results indicate that ecotin-WT and ecotin-RR may be new scaffolds that could be used to develop anticoagulation molecules. © 2015 Elsevier B.V. All rights reserved.
1. Introduction The hemostatic system is essential to protect the integrity of the vasculature. Different processes are involved in this role, including the adhesion and activation of platelets and protein coagulation and inhibition factors [1]. The system is tightly regulated, and any minor imbalance may lead to serious pathological conditions, such as thrombosis [2]. Currently, venous thromboembolism (VTE) is a
Abbreviations: Ecotin-WT, ecotin wild type; ecotin-RR, ecotin with two mutations at the primary binding site; TF.FVIIa, tissue factor–factor VIIa; PT, prothrombin time; APTT, activated partial thromboplastin time; TT, thrombin time; HEPES, N-(2-hydroxyethyl)piperazine-N -(2-ethanesulfonic acid); PEG-6000, polyethylene glycol 6000; S-2765, Z-D-Arg-Gly-Arg-p-nitroanilide; S-2238, H-d-Phe-Pip-Arg-pnitroanilide; BAPNA, N␣-benzoyl-dl-arginine 4-nitroanilide hydrochloride; PPP, platelet poor plasma; Tris–HCl, Trizma® hydrochloride; PBS, phosphate-buffered saline; u-PA, urokinase-type plasminogen activator. ∗ Corresponding author. Tel.: +55 21 3938 6782; fax: +55 21 3938 6782. E-mail addresses: [email protected] (L.S. Wermelinger), fl[email protected] (F.S. Frattani), [email protected] (T.C. Carneiro-Lobo), [email protected] (C.S. Craik), [email protected] (H.C. Castro), [email protected] (R.B. Zingali). http://dx.doi.org/10.1016/j.ijbiomac.2015.03.071 0141-8130/© 2015 Elsevier B.V. All rights reserved.
common pathology. Recent research has shown that the overall incidence of deep vein thrombosis (DVT) and pulmonary embolism (PE) involves 55 and 29 per 100,000 individuals each year, respectively [3,4]. In addition, the number of fatal cases observed 1 year after the first VTE event has an incidence rate of approximately 21%, demonstrating the importance of studying molecules to prevent and treat these conditions [5]. The pharmacologic prophylaxis and treatment of venous thrombosis are currently based on three types of anticoagulants: unfractionated heparin, low-molecular-weight heparins and vitamin K antagonists [6]. Their effects at multiple points in the coagulation cascade lead to severe bleeding problems and other side effects, such as thrombocytopenia, and require monitoring. Alternatively, direct thrombin inhibitor dabigatran and factor Xa inhibitors rivaroxaban and apixaban have been shown to be more selective and safer [7–9]. While these new drugs have fewer side effects, they are not ideal anticoagulants because they can cause bleeding and there are not specific antidotes [10,11]. The need for more efficient and safer anticoagulants is evident. Many efforts have been made, including extensive research using molecules from biological sources, such as hirudin (Hirudo medicinalis),
L.S. Wermelinger et al. / International Journal of Biological Macromolecules 78 (2015) 296–303
ixolaris (Ixodes scapularis), bothrojaracin (Bothrops jararaca venom) and ecotin (Escherichia coli) [12–15]. Ecotin is a protein inhibitor of several serine proteases. It exists as a non-covalent homodimer formed by two 16,096 Da monomers in a head to tail association [16,17]. This homodimer is able to form a unique tetrameric enzyme–inhibitor complex with two serine proteases, such as FXa and trypsin [15,17]. Ecotin has two distinct sites that are known to interact with serine proteases: the primary binding site, which involves the reactive loop (80s loop, residues 81–86) and the 50s loop (residues 52–54), and the secondary binding site, which involves the surface loops 60s (residues 67–70) and 100s (residues 108–113) and the protease C-terminal region [18]. In addition, ecotin can inhibit serine proteases that are involved in several steps of the coagulation cascade, including FXa, FXIa, FXIIa and kallikrein [15,19,20]. However, no or little inhibition was observed with thrombin and the TF.FVIIa complex in the presence of ecotin wild type (ecotin-WT) [15]. Although thrombin was not inhibited by ecotin-WT, our group has shown that ecotin binds to human thrombin via its secondary binding site and modulates catalytic activity [21,22]. To improve its ability to inhibit several serine proteases, some mutations were performed in ecotins, including a single mutation of Met 84 to Arg, increasing ecotin’s inhibition against thrombin, FXa, plasmin and FXIa and leading to a loss of its ability to inhibit elastase [15]. First, ecotin Met84Arg and Met85Arg was described as a potent inhibitor of urokinase-type plasminogen activator (uPa) [23]. Then, our group showed that ecotin with the mutations M84R and M85R changes the primary site of ecotin and it can then also interact with the active site of thrombin. An in silico study showed that P1 Arg84 electrostatically interacts with S1 (Asp189 ) and P2 Arg85 forms a hydrogen bond with catalytic residues of thrombin, improving the affinity of this molecules and in vitro studies revealed the inhibitory capacity by preventing the platelet aggregation induced by thrombin [18,21,22,24,25]. While the anticoagulation effects of ecotin have been widely described in the literature, its actual antithrombotic effects were not demonstrated in vivo until now [15,20,22,26,27]. In this study, ecotin-WT and the mutant (Met84Arg and Met85Arg, ecotin-RR) were compared through a series of in vitro and in vivo coagulation, plasma clotting and thrombus formation models. The anticoagulation and antithrombotic properties of ecotin-WT and ecotin-RR were demonstrated and their potential uses in medicine are discussed. 2. Materials and methods 2.1. Animals Male and female adult BALB-C mice (20–25 g) and Wistar rats (200–250 g) were maintained at room temperature under a 12 h light–dark cycle with unlimited access to food and water. The experiments followed the standards of animal care defined by the Center of Medical Sciences Committee at The Federal University of Rio de Janeiro (IBqM 004). 2.2. Proteins and chemicals Ecotin-WT and ecotin-RR were expressed and purified as previously described [28]. Human thrombin was purified according to Ngai and Chang [29]. The following reagents were commercially obtained: human factor Xa (Calbiochem, San Diego, CA, USA), human fibrinogen and Z-D-Arg-Gly-Arg-pnitroanilide (S-2765) and H-d-Phe-Pip-Arg-p-nitroanilide (S-2238) (Chromogenix, Stockholm, Sweden), bovine pancreatic trypsin, N-(2-hydroxyethyl)piperazine-N -(2-ethanesulfonic acid), HEPES, polyethylene glycol (PEG) 6000 and N␣-benzoyl-dl-arginine
297
4-nitroanilide hydrochloride (BAPNA) (Sigma Chemical Co., St. Louis, USA). All of the solutions were freshly prepared. 2.3. Amidolytic activity of serine proteases Hydrolysis of the synthetic substrates by human thrombin, human factor Xa and bovine trypsin were measured in 10 mM HEPES, 100 mM NaCl, 0.1% PEG 6000, pH 7.4, using a Thermomax Microplate Reader (Molecular Devices, Menlo Park, CA) equipped with a microplate mixer and heating system. Human thrombin (2 nM), factor Xa (2 nM) or trypsin (2 nM) were added to the assay medium in the presence or absence of different concentrations of ecotins, and the reactions were started with the addition of S-2238, S-2756 chromogenic substrate (100 M) or BAPNA (200 M) with minor modifications by Mukherjee and Mackessy [30]. The initial rate of the p-nitroaniline release was determined by the increase in the Abs 405 nm in the first seconds. IC50 refers to the ecotin concentration that inhibited 50% of enzyme activity. 2.4. Fibrinogen clotting assay Human fibrinogen clotting by human thrombin was measured using a Thermomax Microplate Reader (Molecular Devices, Menlo Park, CA) [21]. Various ecotin concentrations were pre-incubated with thrombin (2 nM) for 5 min, and the reactions were started by the addition of 4 mg/mL human fibrinogen. IC50 refers to the ecotin concentration that inhibited 50% of clotting activity. 2.5. Anticoagulant activity assays: prothrombin time (PT) and activated partial thromboplastin time (APTT) Wistar rats of 200–250 g were anesthetized with an intramuscular injection of ketamine (100 mg/kg body weight) and xylazine (16 mg/kg body weight). A BD InsyteTM AutoguardTM catheter (BD Medical) coupled to a 3 mL syringe with 3.8% trisodium citrate solution (1:9 citrate/blood, v/v) was inserted into the right carotid artery for blood collection. Platelet poor plasma (PPP) was obtained by centrifugation at 2000g for 20 min at room temperature. The plasma of Wistar rats (50 L) was incubated for 1 min with different concentrations of ecotins for in vitro anticoagulation assays. For APTT tests, cephalin plus kaolin (APTT reagent, BioMeriaux, RJ, Brazil) were incubated for 2 min at 37 ◦ C with pre-warmed plasma in the presence of different concentrations of both ecotins. The reaction was started by the addition of 100 L of pre-warmed 25 mM CaCl2 . For PT tests, 100 L of a pre-warmed solution of thromboplastin and calcium (PT reagent, BioMeriaux, RJ, Brazil) was added to 50 L of pre-warmed plasma in the presence of different concentrations of both ecotins using an Amelung KC4A Coagulometer (Trinity Biotech, Germany) [31]. 2.6. Venous thrombosis—stasis-induced thrombosis after injection of tissue thromboplastin in vivo Thrombus formation through a combination of stasis and hypercoagulability was induced as previously described [32] with minor modifications by Mendes-Silva et al. [33]. Male and female Wistar rats of 200–250 g were anesthetized as described above. The abdomen was carefully opened and dissected, exposing the vena cava. The inferior vena cava just below the left renal vein was loosely tied at two points 1 cm apart from each other. Different concentrations of ecotins were administered intravenously below the distal tie, and the solution was allowed to circulate for 5, 30 or 60 min before thrombosis induction. The proximal tie was tightened, and stasis was induced through the injection of tissue thromboplastin (3 mg/kg body weight) into the vena cava. After 20 min of stasis, the thrombus in the occluded segment was
298
L.S. Wermelinger et al. / International Journal of Biological Macromolecules 78 (2015) 296–303
clamped off using distal and proximal ties. The thrombus was carefully pulled out, washed with phosphate-buffered saline (PBS), blotted on paper, dried overnight, and weighed. 2.7. Thromboembolism model in vivo and histological analysis Male and female BALB-C mice of 20–25 g were subjected to intravenous injections of ecotins (1.0 and 0.5 mg/kg body weight) 30 min prior to the induction of thrombogenic events by the injection of human thrombin (2000 UI/kg animal). The dose of thrombin used was selected from a concentration as the minimal dose (from 1250 up to 2000 U/kg) that resulted in a reproducible 80–90% mortality rate in the control group (data not shown). Three groups of five animals each were used to estimate the survival rate at 15 min after thrombin injection. A complete necropsy was performed on all of the animals. Tissue samples of the lungs were collected, fixed by immersion in 10% neutral buffered formalin for at least 12 h and submitted for histopathologic examination. After, the lungs were processed by conventional methods, embedded in paraffin, and sectioned at a 3-m thickness prior to staining with phosphotungstic acid hematoxylin for microscopic examination [34]. 2.8. Animal model for the effect on bleeding in vivo Male and female Wistar rats of 200–250 g were anesthetized as described above. Then, the right carotid artery was carefully exposed. A BD InsyteTM AutoguardTM catheter (BD Medical) was inserted into the right carotid artery for the administration of ecotins (1 mg/kg of animal). After 5 min of drug administration, 3 mm of the tail tip was cut and carefully immersed in 40 mL distilled water at room temperature. Sixty minutes later, the blood loss was determined by measuring the hemoglobin dissolved in the water [35]. The volume of blood was deduced from a standard curve obtained using blood samples whose absorbance
was measured at 540 nm. The rat group treated with 5 mg/kg (150 UI/mg) heparin from porcine mucus was used as a positive control [36]. 2.9. Statistical analysis Biological replicates were used per treatment, and the data were presented as a mean ± standard deviation. The differences (p < 0.01) were determined after one-way analysis of variance (ANOVA) and Bonferroni’s multiple comparison test using GraphPad Prism Software (version 5; GraphPad Software, San Diego, CA, USA).
3. Results 3.1. Ecotin-WT and ecotin-RR effects on the activity of serine proteases First, ecotin-WT and ecotin-RR were expressed in E. coli (BL21DE3), purified using a hydrophobic Vydac C4 column, and their activities were determined using an amidolytic assay with trypsin; the IC50 for ecotin-WT was 118 and 182 nM for ecotin-RR. These results demonstrated that ecotins were correctly folded (Fig. 1A). Although the ecotin-WT has a well described effect toward coagulation enzymes, here we compare its effect with ecotin-RR (Figs. 1 and 2). The following effects of ecotins were observed in the amidolytic assay with factor Xa and thrombin, which are important enzymes in the blood coagulation cascade and are targets for prospective antithrombotic drugs. We found that ecotin-WT and ecotin-RR had a similar dose-dependent inhibition for factor Xa with an IC50 of 42 and 46 nM, respectively (Fig. 1B). However, only ecotin-RR inhibited thrombin activity on synthetic substrates (S-2238) with an IC50 of 6 nM and on a physiological substrate (fibrinogen) with an IC50 of 5.6 nM (Fig. 1C and D).
Fig. 1. The effect of ecotins WT and RR on serine protease activity. The inhibition of enzymes in the presence of ecotins WT () and RR () of different concentrations in the presence of: (A) bovine trypsin and BAPNA; (B) factor Xa and S-2765; (C) thrombin and S-2238 and (D) thrombin and fibrinogen. Each point represents the mean ± SD of at least three independent experiments.
L.S. Wermelinger et al. / International Journal of Biological Macromolecules 78 (2015) 296–303
299
a high concentration (1 M) of either ecotin-RR or ecotin-WT, the coagulation time was at least eight times longer compared with the control (Fig. 2A). In addition, very similar results were obtained through APTT assays. In this experiment, the plasma samples were exposed to negatively charged surfaces, which activated factor XII and the high molecular weight kininogen. They successively activated the factors XI, IX, VIII, X, V, and prothrombin, leading to clot formation (Fig. 2B). Overall, these results confirm that ecotin-WT and ecotin-RR have anticoagulation activity and both were able to prolong the clotting time. 3.3. Antithrombotic activity in vivo using stasis and hypercoagulability as a model and a thromboembolic model
Fig. 2. In vitro anticoagulant activity of ecotins. (A) Prothrombin time (PT) and (B) activated partial thromboplastin time (APTT) were assayed with citrated rat plasma incubated with ecotins WT () and RR () at different concentrations for 1 min at 37 ◦ C. Each point represents the mean ± SD of at least three independent experiments, ***p < 0.0001.
3.2. Anticoagulant activity Prothrombin time (PT) and activated partial thromboplastin time (APTT) assays were used to compare the effects of ecotin-RR and ecotin-WT on clotting time. These assays needed a more complex composition, using plasma and different inductors that can promote different pathways for clotting. In the PT assay, thrombin formation is triggered by the addition of thromboplastin that binds to factor VIIa, which in turn activates factor Xa. The resulting thrombin leads to clot formation. Both ecotin-RR and ecotin-WT increased the coagulation time in a dose-dependent manner. With
A thrombosis model combining stasis and hypercoagulability as described in Section 2 was used to evaluate the in vivo ecotin effects on thrombus formation. Thrombosis was induced in rats using treatment with ecotin-RR and ecotin-WT immediately before by thromboplastin. Control rats were treated only with drug vehicle. The ecotin-RR and ecotin-WT treated rats showed dose-dependent response; a higher concentration dose led to an approximate 95% thrombus weight reduction, and a low dose of 0.2 mg prevented thrombus formation when compared to the control group (Fig. 3A). To evaluate a time dependent action, the rats were also treated with 0.2 mg/kg ecotin-WT or ecotin-RR, which is the minimum dose that had an effect at different times (5, 30 and 60 min) prior to thrombus induction with thromboplastin. The thrombus isolated from rats treated with either ecotin at 30 min prior to the induction of the thrombus had the same activity compared to the group that received the treatment 5 min prior. In addition, both ecotins lost a significant amount of inhibitory activity after the 60 min injection (Fig. 3B). As previously shown by our group, ecotin-WT does not directly inhibit thrombin in the amidolytic and fibrinogen clotting activities, nevertheless it was able to bind thrombin and to protect it from the heparin-catalyzed inhibition by anti-thrombin. Furthermore, we also showed that ecotin-WT is a dose-dependent inhibitor of thrombin-induced platelet aggregation [22]. Therefore, these set of data could suggest that ecotins would interfere in thrombus formation during thrombin burst. Hence, both ecotins were further tested for their antithrombotic activity using a model of thrombin-induced acute pulmonary embolism as described. In this experiment, the mice were intravenously injected with human thrombin, which causes an acute paralysis and sudden death. The highest dose, 1 mg/kg, of both ecotins was injected 30 min before the thrombosis induction, and only ecotin-RR prevented death in the animals, with survival rates approximately 50% higher
Fig. 3. The effect of ecotins on stasis-induced venous thrombosis in vivo. (A) Ecotin WT () and RR () at different concentrations were administered i.v. 5 min before the induction of thrombosis by thromboplastin (3 mg/kg) and stasis, (B) ecotin WT () and RR () at a dose of 0.2 mg/kg administered i.v. at different times before the induction of thrombosis. The control group received PBS instead of ecotin. Each point represents the mean ± SD of five to eight animals. **p < 0.01, compared to values observed in the absence of ecotin.
300
L.S. Wermelinger et al. / International Journal of Biological Macromolecules 78 (2015) 296–303
Fig. 4. The effects of ecotins on an in vivo model of pulmonary thromboembolism induced by thrombin. (A) Ecotin WT and RR at a dose of 1 mg/kg administered 30 min before the induction of thromboembolism represent a percentage of survival of the population and (B) ecotin RR at a dose of 1 and 0.5 mg/kg administered 30 min before the induction of thromboembolism represent a percentage of survival of the population. The control group received PBS instead of ecotin. The results shown represent the mean ± SD of three groups of five animals each. ***p < 0.001 or **p < 0.01 compared to values observed in the absence of ecotin. Mallory’s phosphotungstic acid hematoxylinstaining of mouse lung before the thrombin-induced acute pulmonary embolism model. Fibrin formed in the blood vessel stained an intense blue, (C and D) pulmonary thromboembolisms were evaluated in control animals, which only received PBS, (E and F) in the group treated with ecotin-WT at a dose of 1 mg/kg and (G and H) in the ecotin-RR treated animals with a dose of 1 mg/kg, (I and J) and in the ecotin-RR treated animals with dose of 0.5 mg/kg. Minor and major magnification on the left and right, respectively. Asterisk represents the completely or partially clear vessels. Arrows show completely obstructed vessels.
compared with the control (Fig. 4A). After, we treated the mice with ecotin-RR at a lower dose of 0.5 mg/kg, which also prevented animals’ death (Fig. 4B). The lung tissues of ecotin-treated mice had a less fibrin-rich thrombus when compared to the control tissues histologically (Fig. 4C–H). The thrombus is clearly observed in the control tissues as an intense blue color after staining and histological analysis (Fig. 4C and D). Taken together, these results indicate that ecotin-WT and ecotin-RR are able to inhibit in vivo thrombosis.
3.4. Bleeding effect The most common side effect of the currently available antithrombotic drugs is their effect on bleeding. Using a tail transection mouse model, ecotin-WT and ecotin-RR were tested to determine their side effects. Control mice were treated with the drug vehicle (PBS buffer) or heparin. Interestingly, mice treated with 1 mg/kg of either ecotin-WT or ecotin-RR, the same concentration that inhibits 100% of thrombus formation, did not result in any
L.S. Wermelinger et al. / International Journal of Biological Macromolecules 78 (2015) 296–303
Fig. 5. Bleeding effect of ecotins. Ecotins (1 mg/kg) were administered i.v. and after 5 min, the rat tail was cut 3 mm from the tip and carefully immersed in 40 mL of distilled water at room temperature. The blood loss (hemoglobin content) was estimated at 540 nm after 60 min. The absorbance detected for a group that received PBS or 750 UI heparin instead of ecotin was used as a negative and a positive control, respectively. The results represent the mean ± SD of 5–11 animals. ***p < 0.001 compared to values observed in the absence of ecotin.
increased bleeding compared to untreated mice; in contrast, mice treated with 750 UI/kg (5 mg/kg) heparin, as a positive control, had a significant increase in bleeding, as expected (Fig. 5). 4. Discussion Although there are currently many drugs available to treat and prevent thrombotic events, the clinically used drugs present some side effects, such as a risk of bleeding, thrombocytopenia, food interaction and an intense laboratory evaluation during the treatment. To evaluate the anticoagulant effect and safety of new potential drugs, we evaluated two serine protease inhibitors (ecotin-WT and ecotin-RR) as new antithrombotic components, using in vitro and in vivo assays. In this study, we verified that both ecotins have important antithrombotic activity without bleeding side effects. We previously used purified enzymes and showed that ecotinRR conserved the ability to inhibit trypsin and FXa, beyond thrombin improved inhibition. It is known that the mutations M84R and M85R change the primary site of ecotin, allowing it to interact with the catalytic site of thrombin [21]. Our data revealed that the two mutations (Met84Arg and Met85Arg) in the variant primary binding site were crucial in directly modulating the potency of ecotin-RR, which demonstrated an efficient inhibition. Ecotin-RR, at a final concentration of 25 nM, was able to inhibit approximately 75% of thrombin (2 nM) activity immediately after its addition. While the ecotin-R (Met84Arg) mutant inhibited 100% of thrombin (3.7 nM) activity but with four times greater protein concentration (100 nM) and 1 h of incubation [15]. In addition, ecotin-RR prevents 100% of fibrinocoagulation at a low dose of 12.5 nM. This concentration is considered to be more physiologically relevant in vivo. Furthermore, ecotin-RR has an IC50 of 46 nM to inhibit FXa similar to ecotin-R [15]. These results are consistent with the literature regarding the mutation of the primary site where inhibition that occurs with ecotin-RR (Met84Arg and Met85Arg) improves its interactions with thrombin and a thrombin-like catalytic site using an in silico study and some in vitro enzymatic experiments [21]. It is important to emphasize that the IC50 provide an apparent inhibition constant under the conditions described and these identical conditions were used for all inhibition studies, making them relative to one another and valid comparisons [37].
301
The results of the in vitro coagulation plasma assays of the ecotins revealed a prolonged PT time and APTT with a similar profile, despite the first assay being triggered by an extrinsic pathway inducer and the second by an intrinsic pathway inducer; both were oriented in the same way. This result is most likely due to the ecotins inhibitory activity toward FXa, common in both pathways. Currently, anticoagulants, such as unfractionated heparin (UFH), the low molecular weight heparins (LMWHs), synthetic pentasaccharide fondaparinux and vitamin K antagonists (VKAs), have some interactions with food and drugs, need monitoring and present a bleeding risk [10]. Attention has been focused on the development of safer and more efficient molecules. Dabigatran, a direct thrombin inhibitor, and rivaroxaban, a direct factor Xa inhibitor, have been licensed in many countries [11,38]. The advantage of direct thrombin inhibitors relies in its capability of blocking its interaction with substrates, thus preventing fibrin formation, the activation of factors V, VIII, XI or XIII, and platelet aggregation. Also direct thrombin inhibitors can be more effectively because they inactivate fibrin-bound thrombin thus attenuating thrombus formation. On the other hand, it was hypothesized that selective inhibition of coagulation factors located upstream of thrombin in the coagulation pathway might be safer in relation to bleeding risk; by not inhibiting thrombin activity directly, the drugs would allow some amount of thrombin to escape neutralization and thereby participate in hemostasis. Therefore factor Xa is also an attractive target for anticoagulants because it is the rate-limiting component in the generation of thrombin. Hence, the inhibition of both targets, such as FXa and FIIa, represent a notable strategy to prevent coagulation without bleeding side effects. Studies have found that dabigatran can prevent the incidence of recurrent venous thromboembolism (VTE) with no difference in the incidence of major bleeding [39]. Similarly, only ecotinRR was able to prevent death by thromboembolism due to the hypercoagulability induced by thrombin administration, in a dosedependent manner (i.v. 0.5–1 mg/kg); this result was confirmed by lung histology of the animals. As described by Gresele et al. and reinforced by Momi et al. the thrombin-thromboembolic model in mice induced the thrombus at least two main mechanisms. First, thrombin exogenous can be directly convert fibrinogen to fibrin that promotes intravascular fibrin accumulation in the lung. In addition, thrombin exogenous can activate a contact phase by the activation of Factor XI-mediated mechanism leading to additional endogenous thrombin formation by feedback. The endogenous thrombin formed would play a major role in making intravascular deposit fibrin more resistant to fibrinolysis by activation of Factor XIII [34,40]. Despite of the inhibitory effects of ecotin-WT toward thrombin, only ecotin-RR prevented death in approximately 80% of the animals in the thromboembolic model induced by thrombin and did not show a bleeding effect in the same dose. These results suggest that ecotin-RR could be a good template for developing a molecule to directly inhibit thrombin and prevent a recurrent thromboembolism without a bleeding side effect, until 1 h after the treatment. Rivaroxaban, the factor-Xa inhibitor that has a favorable balance of efficacy and safety for preventing venous thromboembolism after major orthopedic surgery, is a direct inhibitor of FXa, like the ecotins [41,42]. When we compared the activity of ecotins with rivaroxaban in venous animal models, we observed a similar activity. Rivaroxaban has potent antithrombotic properties (i.v. 0.3 mg/kg dose leads to an almost complete inhibition of thrombus formation) in stasis and hypercoagulability models and does not prolong bleeding in animal models. Bleeding time of the animals was observed after a dose of 3 mg/kg, and the activity was not different from the animals without treatment, suggesting that the molecule is safe [43]. Using a similar model, we verified that ecotins have an antithrombotic effect with a similar potency; both ecotins
302
L.S. Wermelinger et al. / International Journal of Biological Macromolecules 78 (2015) 296–303
were able to inhibit thrombosis in a dose and time dependent manner (i.v. 0.5 mg/kg dose leading to an almost complete inhibition of thrombus formation) until 30 min. This activity was expected because our data show the same profile for both ecotins in the PT assay, which uses the same fuel (thromboplastin). Then, with a higher dose (1 mg/kg), neither ecotins had an effect on bleeding. Remarkably, ecotins WT and RR displayed notable potency (i.v. dose 0.5 mg/kg inhibited more than 95% of thrombus formation) compared to other antithrombotic proteins from animal sources, such as bothrojaracin (inhibits FII and FIIa), and hirudin (inhibits FIIa). Bothrojaracin is a 27 kDa protein and is a thrombin inhibitor isolated from the Bothrops jararaca venom by our group [44]. This protein interacts with thrombin through the thrombin binding exosite I and II, without interacting with the catalytic site of thrombin and is also able to interact with prothrombin [45,46]. Bothrojaracin requires an i.v. dose of 1 mg/kg to prevent at least 95% of thrombus formation in a deep vein thrombosis model (DVTM), and the same dose is able to prevent 100% of the mortality in a thromboembolic model induced by thrombin in contrast to the only 80% protection displayed in this work by ecotin-RR [14]. Hirudin is a peptide approximately 7 kDa in size isolated from the glands of Hirudo medicinalis that has thrombin inhibitor activity, interacts with exosite I and directly interacts with the catalytic site. Recombinant hirudin is administered intravenously, and this molecule is indicated for anticoagulation in patients with heparininduced thrombocytopenia [12,47]. In a deep vein thrombosis model with hypercoagulability and stasis, hirudin is able to prevent approximately 80% of thrombus formation with a dose of 0.3 mg/kg, showing a similar profile to the ecotins. In these models, Herbert and colleagues induced hypercoagulability with six times less thromboplastin (0.5 mg/kg), which most likely requires less inhibitor to prevent thrombus formation, and a dose of 0.3 mg/kg of hirudin was effective [48]. In a thromboembolic model induced by bovine thrombin, hirudin prevented 100% of death in animals with a dose of 0.4 mg/kg, demonstrating more activity in this thrombosis model compared with ecotin-RR from our study because we needed 1 mg/kg to prevent 80% of death [40]. However, bothrojaracin (1 mg/kg) and hirudin (0.1 mg/kg) cause bleeding side effects when tested at a dose that promotes an antithrombotic effect, which was not observed with ecotin (1 mg/kg) treatment [14,40]. These observations lead us to believe that these molecules should be used in a more specific manner to prevent thrombus formation in patients who require surgery or invasive procedures because ecotin is potent and rapidly active and has reduced bleeding side effects. Interestingly, ecotin-WT and ecotin-RR prevent thrombus formation in a model that combines stasis and hypercoagulability without bleeding. Both molecules displayed good activity with an i.v. dose of 0.5 mg/kg, which inhibited more than 95% of thrombus formation. In addition, only ecotin-RR was able to decrease death due hypercoagulability induced by thrombin. The antithrombotic activity with absence of bleeding can be explained by the ability to inhibit FXa because these two ecotins have an antithrombotic profile similar to rivaroxaban. In addition, ecotin-WT was able to inhibit other enzymes in the hemostatic system, such as FXIa, FXIIa and kallikrein [15,20]. It is important to emphasize that ecotin RR also inhibit uPa [23]. Nevertheless, this blockage has no relevance in the thrombosis models used in our study, since it impairs the thrombus formation preventing the activation of fibrinolytic pathway. The literature indicates that inhibition of these molecules can be efficient and safer since they do not promote bleeding, suggesting that ecotins could have a safer profile because they can also inhibit factors from the intrinsic cascade [49]. In conclusion, our results suggests that ecotin-WT prevents thrombus formation through the inhibition of Factor Xa or
upstream, while ecotin-RR, by its potent inhibition of thrombin, is more efficient when the thrombus is associated with thrombin burst. Thus, both ecotins displayed potent in vivo antithrombotic effects that were not associated with bleeding. These results demonstrated the potential of those molecules as templates for the design of new anticoagulation molecules. It is though essential to understand all the potential activity of ecotins, since ecotin is a foreign protein that could induce an immunological response, it would be desirable further studies to develop a prototype that conserves it biological activity and safety. Acknowledgments This work was supported by Conselho Nacional de Desenvolvimento Científico (CNPq), Fundac¸ão de Amparo à Pesquisa do Estado do Rio de Janeiro (FAPERJ), Coordenac¸ão de Aperfeic¸oamento de Pessoal de Nível Superior (CAPES), Universidade Federal Fluminense (UFF) (Brazil) and Department of Defense grant PC111318 from USA. We would like to thank Dione Maria da Silva and Ana Lúcia de Oliveira Carvalho for their technical assistance. The kind donation of heparin was provided by Dr. Paulo Mourão (Universidade Federal do Rio de Janeiro, RJ, Brazil). References [1] A.S. Wolberg, M.M. Aleman, K. Leiderman, K.R. Machlus, Anesth. Analg. 114 (2012) 275–285. [2] B. Furie, Am. Soc. Hematol. Educ. Program (2009) 255–258. [3] S. Liao, T. Woulfe, S. Hyder, E. Merriman, D. Simpson, S. Chunilal, J. Thromb. Haemost. 12 (2014) 214–219. [4] S.Z. Goldhaber, H. Bounameaux, Lancet Haematol. 379 (2012) 1835–1846. [5] I.A. Naess, S.C. Christiansen, P. Romundstad, S.C. Cannegieter, F.R. Rosendaal, J. Hammerstrøm, J. Thromb. Haemost. 5 (2007) 692–699. [6] E.I. Sinauridze, M.A. Panteleev, F.I. Ataullakhanov, Blood Coagul. Fibrinolysis 23 (2012) 482–493. [7] R.C. Becker, J. Alexander, C.K. Dyke, R.A. Harrington, Thromb. Haemost. 92 (2004) 1182–1193. [8] M. Brodmann, Hamostaseologie 33 (2013) 218–224. [9] M. Coppens, J.W. Eikelboom, D. Gustafsson, J.I. Weitz, J. Hirsh, Circ. Res. 111 (2012) 920–929. [10] P.M. Mannucci, M. Levi, N. Engl. J. Med. 356 (2007) 2301–2311. [11] T. Baglin, J. Thromb. Haemost. 11 (2013) 122–128. [12] F. Markwardt, Thromb. Res. 74 (1994) 1–23. [13] I.M. Francischetti, J.G. Valenzuela, J.F. Andersen, T.N. Mather, J.M. Ribeiro, Blood 99 (2002) 3602–3612. [14] R.B. Zingali, M.S. Ferreira, M. Assafim, F.S. Frattani, R.Q. Monteiro, Pathophysiol. Haemost. Thromb. 34 (2005) 160–163. [15] J.L. Seymour, R.N. Lindquist, M.S. Dennis, B. Moffat, D. Yansura, D. Reilly, M.E. Wessinger, R.A. Lazarus, Biochemistry 33 (1994) 3949–3958. [16] M.E. McGrath, W.M. Hines, J.A. Sakanari, R.J. Fletterick, C.S. Craik, J. Biol. Chem. 266 (1991) 6620–6625. [17] M.E. McGrath, T. Erpel, C. Bystroff, R.J. Fletterick, EMBO J. 13 (1994) 1502–1507. [18] S.Q. Yang, C.I. Wang, S.A. Gillmor, R.J. Fletterick, C.S. Craik, J. Mol. Biol. 279 (1998) 945–957. [19] C.H. Chung, H.E. Ives, S. Almeda, A.L. Goldberg, J. Biol. Chem. 258 (1983) 11032–11038. [20] A.A. Stoop, C.S. Craik, Nat. Biotechnol. 21 (2003) 1063–1068. [21] H.C. Castro, D.M. Silva, C.S. Craik, R.B. Zingali, Biochim. Biophys. Acta 1547 (2001) 183–195. [22] H.C. Castro, R.Q. Monteiro, M. Assafim, N.I. Loureiro, C.S. Craik, R.B. Zingali, Int. J. Biochem. Cell Biol. 38 (2006) 1893–1900. [23] C.I. Wang, Q. Yang, C.S. Craik, J. Biol. Chem. 270 (1995) 12250–12256. [24] M.C. Laboissière, M.M. Young, R.G. Pinho, S. Todd, R.J. Fletterick, I. Kuntz, C.S. Craik, J. Biol. Chem. 277 (2002) 26623–26631. [25] P.C. Sathler, C.S. Craik, T. Takeuchi, R.B. Zingali, H.C. Castro, Appl. Biochem. Biotechnol. 160 (2010) 2355–2365. [26] J.S. Ulmer, R.N. Lindquist, M.S. Dennis, R.A. Lazarus, FEBS Lett. 365 (1995) 159–163. [27] A.A. Stoop, R.V. Joshi, C.T. Eggers, C.S. Craik, Biol. Chem. 391 (2010) 425–433. [28] C.I. Wang, Q. Yang, C.S. Craik, Methods Enzymol. 267 (1996) 52–68. [29] P.K. Ngai, J.Y. Chang, Biochem. J. 280 (1991) 805–808. [30] A.K. Mukherjee, S.P. Mackessy, Biochim. Biophys. Acta 1830 (2013) 3476– 3488. [31] X. Wang, Z. Zhang, Z. Yao, M. Zhao, H. Qi, Int. J. Biol. Macromol. 58 (2013) 225–230. [32] J.C. Martinichen-Herrero, E.R. Carbonero, G.L. Sassaki, P.A. Gorin, M. Iacomini, Int. J. Biol. Macromol. 35 (2005) 97–102.
L.S. Wermelinger et al. / International Journal of Biological Macromolecules 78 (2015) 296–303 [33] W. Mendes-Silva, M. Assafim, B. Ruta, R.Q. Monteiro, J.A. Guimarães, R.B. Zingali, Thromb. Res. 112 (2003) 93–98. [34] P. Gresele, S. Momi, M. Berrettini, G.G. Nenci, H.P. Schwarz, N. Semeraro, M. Colucci, J. Clin. Invest. 101 (1998) 667–676. [35] J.M. Herbert, J.P. Hérault, A. Bernat, R.G. van Amsterdam, G.M. Vogel, J.C. Lormeau, M. Petitou, D.G. Meuleman, Circ. Res. 79 (1996) 590–600. [36] J.C. Santos, J.M. Mesquita, C.L. Belmiro, C.B. da Silveira, C. Viskov, P.A. Mourier, M.S. Pavão, Thromb. Res. 121 (2007) 213–223. [37] C.T. Eggers, S.X. Wang, R.J. Fletterick, C.S. Craik, J. Mol. Biol. 308 (2001) 975–991. [38] J.I. Weitz, Am. J. Hematol. 87 (2012) 133–136. [39] S. Schulman, B. Ritchie, J.K. Goy, S. Nahirniak, M. Almutawa, S. Ghanny, Br. J. Haematol. 164 (2014) 308–310. [40] S. Momi, M. Nasimi, M. Colucci, G.G. Nenci, P. Gresele, Haematologica 86 (2001) 297–302. [41] B.I. Eriksson, L.C. Borris, R.J. Friedman, S. Haas, M.V. Huisman, A.K. Kakkar, T.J. Bandel, H. Beckmann, E. Muehlhofer, F. Misselwitz, W. Geerts, N. Engl. J. Med. 358 (2008) 2765–2775.
303
[42] W.D. Fisher, B.I. Eriksson, K.A. Bauer, L. Borris, O.E. Dahl, M. Gent, S. Haas, M. Homering, M.V. Huisman, A.K. Kakkar, P. Kälebo, L.M. Kwong, F. Misselwitz, A.G. Turpie, Thromb. Haemost. 97 (2007) 931–937. [43] E. Perzborn, J. Strassburger, A. Wilmen, J. Pohlmann, S. Roehrig, K.H. Schlemmer, A. Straub, J. Thromb. Haemost. 3 (2005) 514–521. [44] R.B. Zingali, M. Jandrot-Perrus, M.C. Guillin, C. Bon, Biochemistry 32 (1993) 10794–10802. [45] V. Arocas, C. Lemaire, M.C. Bouton, A. Bezeaud, C. Bon, M.C. Guillin, M. JandrotPerrus, Thromb. Haemost. 79 (1998) 1157–1161. [46] V. Arocas, R.B. Zingali, M.C. Guillin, C. Bon, M. Jandrot-Perrus, Biochemistry 35 (1996) 9083–9089. [47] C.C. Liu, E. Brustad, W. Liu, P.G. Schultz, J. Am. Chem. Soc. 129 (2007) 10648–10649. [48] J.M. Herbert, A. Bernat, J.P. Maffrand, Blood 80 (1992) 2281–2286. [49] M.L. van Montfoort, J.C. Meijers, Thromb. Haemost. 110 (2013) 223–232.
Contents lists available at ScienceDirect
International Journal of Biological Macromolecules journal homepage: www.elsevier.com/locate/ijbiomac
Ecotin: Exploring a feasible antithrombotic profile Luciana Serrão Wermelinger a,b , Flávia Serra Frattani a,b , Tatiana Correa Carneiro-Lobo b , Charles S. Craik c , Helena Carla Castro d , Russolina Benedeta Zingali b,∗ a
Departamento de Análises Clínicas e Toxicológicas, Faculdade de Farmácia/CCS, Universidade Federal do Rio de Janeiro, Rio do Janeiro, RJ, Brazil Instituto de Bioquímica Médica Leopoldo de Meis/CCS, Universidade Federal do Rio de Janeiro, Rio do Janeiro, RJ, Brazil California Institute of Quantitative Biosciences (QB3), University of California, San Francisco, San Francisco, CA, USA d Laboratório de Antibióticos, Bioquímica e Modelagem Molecular (LABioMol), Departamento de Biologia Celular e Molecular, IB-/CEG, Universidade Federal Fluminense, Niterói, RJ, Brazil b c
a r t i c l e
i n f o
Article history: Received 30 December 2014 Received in revised form 20 March 2015 Accepted 30 March 2015 Available online 13 April 2015 Keywords: Ecotin Thrombin inhibitor Factor Xa inhibitor
a b s t r a c t Ecotin is an Escherichia coli-derived protein that can inhibit serine proteases. It has been suggested that this protein (ecotin-WT) and some of its variants could be used to develop a prototype to treat thrombosis. In this work, the effect of ecotin-WT and a variant of this protein harboring two mutations (Met84Arg and Met85Arg, ecotin-RR) were analyzed to determine their ability to prevent thrombus formation using in vivo models. Ecotins were analyzed in vitro using the coagulation tests. An in vivo venous thrombosis rat model and a pulmonary thromboembolism mouse model were used to investigate the antithrombotic activity. The bleeding time in rats using a tail-transection model was evaluated as a possible side effect caused by the administration of these proteins. Ecotin-RR was more effective in inhibiting thrombin than ecotin-WT. Both ecotins presented similar mechanisms of anticoagulation activity and were able to decrease thrombus formation. In contrast, only ecotin-RR increased survival rates in the in vivo pulmonary thromboembolism model, reinforcing the antithrombotic activity of ecotin-RR. Ecotin-WT and more so ecotin-RR showed potent antithrombotic effects, not associated with bleeding. The presented results indicate that ecotin-WT and ecotin-RR may be new scaffolds that could be used to develop anticoagulation molecules. © 2015 Elsevier B.V. All rights reserved.
1. Introduction The hemostatic system is essential to protect the integrity of the vasculature. Different processes are involved in this role, including the adhesion and activation of platelets and protein coagulation and inhibition factors [1]. The system is tightly regulated, and any minor imbalance may lead to serious pathological conditions, such as thrombosis [2]. Currently, venous thromboembolism (VTE) is a
Abbreviations: Ecotin-WT, ecotin wild type; ecotin-RR, ecotin with two mutations at the primary binding site; TF.FVIIa, tissue factor–factor VIIa; PT, prothrombin time; APTT, activated partial thromboplastin time; TT, thrombin time; HEPES, N-(2-hydroxyethyl)piperazine-N -(2-ethanesulfonic acid); PEG-6000, polyethylene glycol 6000; S-2765, Z-D-Arg-Gly-Arg-p-nitroanilide; S-2238, H-d-Phe-Pip-Arg-pnitroanilide; BAPNA, N␣-benzoyl-dl-arginine 4-nitroanilide hydrochloride; PPP, platelet poor plasma; Tris–HCl, Trizma® hydrochloride; PBS, phosphate-buffered saline; u-PA, urokinase-type plasminogen activator. ∗ Corresponding author. Tel.: +55 21 3938 6782; fax: +55 21 3938 6782. E-mail addresses: [email protected] (L.S. Wermelinger), fl[email protected] (F.S. Frattani), [email protected] (T.C. Carneiro-Lobo), [email protected] (C.S. Craik), [email protected] (H.C. Castro), [email protected] (R.B. Zingali). http://dx.doi.org/10.1016/j.ijbiomac.2015.03.071 0141-8130/© 2015 Elsevier B.V. All rights reserved.
common pathology. Recent research has shown that the overall incidence of deep vein thrombosis (DVT) and pulmonary embolism (PE) involves 55 and 29 per 100,000 individuals each year, respectively [3,4]. In addition, the number of fatal cases observed 1 year after the first VTE event has an incidence rate of approximately 21%, demonstrating the importance of studying molecules to prevent and treat these conditions [5]. The pharmacologic prophylaxis and treatment of venous thrombosis are currently based on three types of anticoagulants: unfractionated heparin, low-molecular-weight heparins and vitamin K antagonists [6]. Their effects at multiple points in the coagulation cascade lead to severe bleeding problems and other side effects, such as thrombocytopenia, and require monitoring. Alternatively, direct thrombin inhibitor dabigatran and factor Xa inhibitors rivaroxaban and apixaban have been shown to be more selective and safer [7–9]. While these new drugs have fewer side effects, they are not ideal anticoagulants because they can cause bleeding and there are not specific antidotes [10,11]. The need for more efficient and safer anticoagulants is evident. Many efforts have been made, including extensive research using molecules from biological sources, such as hirudin (Hirudo medicinalis),
L.S. Wermelinger et al. / International Journal of Biological Macromolecules 78 (2015) 296–303
ixolaris (Ixodes scapularis), bothrojaracin (Bothrops jararaca venom) and ecotin (Escherichia coli) [12–15]. Ecotin is a protein inhibitor of several serine proteases. It exists as a non-covalent homodimer formed by two 16,096 Da monomers in a head to tail association [16,17]. This homodimer is able to form a unique tetrameric enzyme–inhibitor complex with two serine proteases, such as FXa and trypsin [15,17]. Ecotin has two distinct sites that are known to interact with serine proteases: the primary binding site, which involves the reactive loop (80s loop, residues 81–86) and the 50s loop (residues 52–54), and the secondary binding site, which involves the surface loops 60s (residues 67–70) and 100s (residues 108–113) and the protease C-terminal region [18]. In addition, ecotin can inhibit serine proteases that are involved in several steps of the coagulation cascade, including FXa, FXIa, FXIIa and kallikrein [15,19,20]. However, no or little inhibition was observed with thrombin and the TF.FVIIa complex in the presence of ecotin wild type (ecotin-WT) [15]. Although thrombin was not inhibited by ecotin-WT, our group has shown that ecotin binds to human thrombin via its secondary binding site and modulates catalytic activity [21,22]. To improve its ability to inhibit several serine proteases, some mutations were performed in ecotins, including a single mutation of Met 84 to Arg, increasing ecotin’s inhibition against thrombin, FXa, plasmin and FXIa and leading to a loss of its ability to inhibit elastase [15]. First, ecotin Met84Arg and Met85Arg was described as a potent inhibitor of urokinase-type plasminogen activator (uPa) [23]. Then, our group showed that ecotin with the mutations M84R and M85R changes the primary site of ecotin and it can then also interact with the active site of thrombin. An in silico study showed that P1 Arg84 electrostatically interacts with S1 (Asp189 ) and P2 Arg85 forms a hydrogen bond with catalytic residues of thrombin, improving the affinity of this molecules and in vitro studies revealed the inhibitory capacity by preventing the platelet aggregation induced by thrombin [18,21,22,24,25]. While the anticoagulation effects of ecotin have been widely described in the literature, its actual antithrombotic effects were not demonstrated in vivo until now [15,20,22,26,27]. In this study, ecotin-WT and the mutant (Met84Arg and Met85Arg, ecotin-RR) were compared through a series of in vitro and in vivo coagulation, plasma clotting and thrombus formation models. The anticoagulation and antithrombotic properties of ecotin-WT and ecotin-RR were demonstrated and their potential uses in medicine are discussed. 2. Materials and methods 2.1. Animals Male and female adult BALB-C mice (20–25 g) and Wistar rats (200–250 g) were maintained at room temperature under a 12 h light–dark cycle with unlimited access to food and water. The experiments followed the standards of animal care defined by the Center of Medical Sciences Committee at The Federal University of Rio de Janeiro (IBqM 004). 2.2. Proteins and chemicals Ecotin-WT and ecotin-RR were expressed and purified as previously described [28]. Human thrombin was purified according to Ngai and Chang [29]. The following reagents were commercially obtained: human factor Xa (Calbiochem, San Diego, CA, USA), human fibrinogen and Z-D-Arg-Gly-Arg-pnitroanilide (S-2765) and H-d-Phe-Pip-Arg-p-nitroanilide (S-2238) (Chromogenix, Stockholm, Sweden), bovine pancreatic trypsin, N-(2-hydroxyethyl)piperazine-N -(2-ethanesulfonic acid), HEPES, polyethylene glycol (PEG) 6000 and N␣-benzoyl-dl-arginine
297
4-nitroanilide hydrochloride (BAPNA) (Sigma Chemical Co., St. Louis, USA). All of the solutions were freshly prepared. 2.3. Amidolytic activity of serine proteases Hydrolysis of the synthetic substrates by human thrombin, human factor Xa and bovine trypsin were measured in 10 mM HEPES, 100 mM NaCl, 0.1% PEG 6000, pH 7.4, using a Thermomax Microplate Reader (Molecular Devices, Menlo Park, CA) equipped with a microplate mixer and heating system. Human thrombin (2 nM), factor Xa (2 nM) or trypsin (2 nM) were added to the assay medium in the presence or absence of different concentrations of ecotins, and the reactions were started with the addition of S-2238, S-2756 chromogenic substrate (100 M) or BAPNA (200 M) with minor modifications by Mukherjee and Mackessy [30]. The initial rate of the p-nitroaniline release was determined by the increase in the Abs 405 nm in the first seconds. IC50 refers to the ecotin concentration that inhibited 50% of enzyme activity. 2.4. Fibrinogen clotting assay Human fibrinogen clotting by human thrombin was measured using a Thermomax Microplate Reader (Molecular Devices, Menlo Park, CA) [21]. Various ecotin concentrations were pre-incubated with thrombin (2 nM) for 5 min, and the reactions were started by the addition of 4 mg/mL human fibrinogen. IC50 refers to the ecotin concentration that inhibited 50% of clotting activity. 2.5. Anticoagulant activity assays: prothrombin time (PT) and activated partial thromboplastin time (APTT) Wistar rats of 200–250 g were anesthetized with an intramuscular injection of ketamine (100 mg/kg body weight) and xylazine (16 mg/kg body weight). A BD InsyteTM AutoguardTM catheter (BD Medical) coupled to a 3 mL syringe with 3.8% trisodium citrate solution (1:9 citrate/blood, v/v) was inserted into the right carotid artery for blood collection. Platelet poor plasma (PPP) was obtained by centrifugation at 2000g for 20 min at room temperature. The plasma of Wistar rats (50 L) was incubated for 1 min with different concentrations of ecotins for in vitro anticoagulation assays. For APTT tests, cephalin plus kaolin (APTT reagent, BioMeriaux, RJ, Brazil) were incubated for 2 min at 37 ◦ C with pre-warmed plasma in the presence of different concentrations of both ecotins. The reaction was started by the addition of 100 L of pre-warmed 25 mM CaCl2 . For PT tests, 100 L of a pre-warmed solution of thromboplastin and calcium (PT reagent, BioMeriaux, RJ, Brazil) was added to 50 L of pre-warmed plasma in the presence of different concentrations of both ecotins using an Amelung KC4A Coagulometer (Trinity Biotech, Germany) [31]. 2.6. Venous thrombosis—stasis-induced thrombosis after injection of tissue thromboplastin in vivo Thrombus formation through a combination of stasis and hypercoagulability was induced as previously described [32] with minor modifications by Mendes-Silva et al. [33]. Male and female Wistar rats of 200–250 g were anesthetized as described above. The abdomen was carefully opened and dissected, exposing the vena cava. The inferior vena cava just below the left renal vein was loosely tied at two points 1 cm apart from each other. Different concentrations of ecotins were administered intravenously below the distal tie, and the solution was allowed to circulate for 5, 30 or 60 min before thrombosis induction. The proximal tie was tightened, and stasis was induced through the injection of tissue thromboplastin (3 mg/kg body weight) into the vena cava. After 20 min of stasis, the thrombus in the occluded segment was
298
L.S. Wermelinger et al. / International Journal of Biological Macromolecules 78 (2015) 296–303
clamped off using distal and proximal ties. The thrombus was carefully pulled out, washed with phosphate-buffered saline (PBS), blotted on paper, dried overnight, and weighed. 2.7. Thromboembolism model in vivo and histological analysis Male and female BALB-C mice of 20–25 g were subjected to intravenous injections of ecotins (1.0 and 0.5 mg/kg body weight) 30 min prior to the induction of thrombogenic events by the injection of human thrombin (2000 UI/kg animal). The dose of thrombin used was selected from a concentration as the minimal dose (from 1250 up to 2000 U/kg) that resulted in a reproducible 80–90% mortality rate in the control group (data not shown). Three groups of five animals each were used to estimate the survival rate at 15 min after thrombin injection. A complete necropsy was performed on all of the animals. Tissue samples of the lungs were collected, fixed by immersion in 10% neutral buffered formalin for at least 12 h and submitted for histopathologic examination. After, the lungs were processed by conventional methods, embedded in paraffin, and sectioned at a 3-m thickness prior to staining with phosphotungstic acid hematoxylin for microscopic examination [34]. 2.8. Animal model for the effect on bleeding in vivo Male and female Wistar rats of 200–250 g were anesthetized as described above. Then, the right carotid artery was carefully exposed. A BD InsyteTM AutoguardTM catheter (BD Medical) was inserted into the right carotid artery for the administration of ecotins (1 mg/kg of animal). After 5 min of drug administration, 3 mm of the tail tip was cut and carefully immersed in 40 mL distilled water at room temperature. Sixty minutes later, the blood loss was determined by measuring the hemoglobin dissolved in the water [35]. The volume of blood was deduced from a standard curve obtained using blood samples whose absorbance
was measured at 540 nm. The rat group treated with 5 mg/kg (150 UI/mg) heparin from porcine mucus was used as a positive control [36]. 2.9. Statistical analysis Biological replicates were used per treatment, and the data were presented as a mean ± standard deviation. The differences (p < 0.01) were determined after one-way analysis of variance (ANOVA) and Bonferroni’s multiple comparison test using GraphPad Prism Software (version 5; GraphPad Software, San Diego, CA, USA).
3. Results 3.1. Ecotin-WT and ecotin-RR effects on the activity of serine proteases First, ecotin-WT and ecotin-RR were expressed in E. coli (BL21DE3), purified using a hydrophobic Vydac C4 column, and their activities were determined using an amidolytic assay with trypsin; the IC50 for ecotin-WT was 118 and 182 nM for ecotin-RR. These results demonstrated that ecotins were correctly folded (Fig. 1A). Although the ecotin-WT has a well described effect toward coagulation enzymes, here we compare its effect with ecotin-RR (Figs. 1 and 2). The following effects of ecotins were observed in the amidolytic assay with factor Xa and thrombin, which are important enzymes in the blood coagulation cascade and are targets for prospective antithrombotic drugs. We found that ecotin-WT and ecotin-RR had a similar dose-dependent inhibition for factor Xa with an IC50 of 42 and 46 nM, respectively (Fig. 1B). However, only ecotin-RR inhibited thrombin activity on synthetic substrates (S-2238) with an IC50 of 6 nM and on a physiological substrate (fibrinogen) with an IC50 of 5.6 nM (Fig. 1C and D).
Fig. 1. The effect of ecotins WT and RR on serine protease activity. The inhibition of enzymes in the presence of ecotins WT () and RR () of different concentrations in the presence of: (A) bovine trypsin and BAPNA; (B) factor Xa and S-2765; (C) thrombin and S-2238 and (D) thrombin and fibrinogen. Each point represents the mean ± SD of at least three independent experiments.
L.S. Wermelinger et al. / International Journal of Biological Macromolecules 78 (2015) 296–303
299
a high concentration (1 M) of either ecotin-RR or ecotin-WT, the coagulation time was at least eight times longer compared with the control (Fig. 2A). In addition, very similar results were obtained through APTT assays. In this experiment, the plasma samples were exposed to negatively charged surfaces, which activated factor XII and the high molecular weight kininogen. They successively activated the factors XI, IX, VIII, X, V, and prothrombin, leading to clot formation (Fig. 2B). Overall, these results confirm that ecotin-WT and ecotin-RR have anticoagulation activity and both were able to prolong the clotting time. 3.3. Antithrombotic activity in vivo using stasis and hypercoagulability as a model and a thromboembolic model
Fig. 2. In vitro anticoagulant activity of ecotins. (A) Prothrombin time (PT) and (B) activated partial thromboplastin time (APTT) were assayed with citrated rat plasma incubated with ecotins WT () and RR () at different concentrations for 1 min at 37 ◦ C. Each point represents the mean ± SD of at least three independent experiments, ***p < 0.0001.
3.2. Anticoagulant activity Prothrombin time (PT) and activated partial thromboplastin time (APTT) assays were used to compare the effects of ecotin-RR and ecotin-WT on clotting time. These assays needed a more complex composition, using plasma and different inductors that can promote different pathways for clotting. In the PT assay, thrombin formation is triggered by the addition of thromboplastin that binds to factor VIIa, which in turn activates factor Xa. The resulting thrombin leads to clot formation. Both ecotin-RR and ecotin-WT increased the coagulation time in a dose-dependent manner. With
A thrombosis model combining stasis and hypercoagulability as described in Section 2 was used to evaluate the in vivo ecotin effects on thrombus formation. Thrombosis was induced in rats using treatment with ecotin-RR and ecotin-WT immediately before by thromboplastin. Control rats were treated only with drug vehicle. The ecotin-RR and ecotin-WT treated rats showed dose-dependent response; a higher concentration dose led to an approximate 95% thrombus weight reduction, and a low dose of 0.2 mg prevented thrombus formation when compared to the control group (Fig. 3A). To evaluate a time dependent action, the rats were also treated with 0.2 mg/kg ecotin-WT or ecotin-RR, which is the minimum dose that had an effect at different times (5, 30 and 60 min) prior to thrombus induction with thromboplastin. The thrombus isolated from rats treated with either ecotin at 30 min prior to the induction of the thrombus had the same activity compared to the group that received the treatment 5 min prior. In addition, both ecotins lost a significant amount of inhibitory activity after the 60 min injection (Fig. 3B). As previously shown by our group, ecotin-WT does not directly inhibit thrombin in the amidolytic and fibrinogen clotting activities, nevertheless it was able to bind thrombin and to protect it from the heparin-catalyzed inhibition by anti-thrombin. Furthermore, we also showed that ecotin-WT is a dose-dependent inhibitor of thrombin-induced platelet aggregation [22]. Therefore, these set of data could suggest that ecotins would interfere in thrombus formation during thrombin burst. Hence, both ecotins were further tested for their antithrombotic activity using a model of thrombin-induced acute pulmonary embolism as described. In this experiment, the mice were intravenously injected with human thrombin, which causes an acute paralysis and sudden death. The highest dose, 1 mg/kg, of both ecotins was injected 30 min before the thrombosis induction, and only ecotin-RR prevented death in the animals, with survival rates approximately 50% higher
Fig. 3. The effect of ecotins on stasis-induced venous thrombosis in vivo. (A) Ecotin WT () and RR () at different concentrations were administered i.v. 5 min before the induction of thrombosis by thromboplastin (3 mg/kg) and stasis, (B) ecotin WT () and RR () at a dose of 0.2 mg/kg administered i.v. at different times before the induction of thrombosis. The control group received PBS instead of ecotin. Each point represents the mean ± SD of five to eight animals. **p < 0.01, compared to values observed in the absence of ecotin.
300
L.S. Wermelinger et al. / International Journal of Biological Macromolecules 78 (2015) 296–303
Fig. 4. The effects of ecotins on an in vivo model of pulmonary thromboembolism induced by thrombin. (A) Ecotin WT and RR at a dose of 1 mg/kg administered 30 min before the induction of thromboembolism represent a percentage of survival of the population and (B) ecotin RR at a dose of 1 and 0.5 mg/kg administered 30 min before the induction of thromboembolism represent a percentage of survival of the population. The control group received PBS instead of ecotin. The results shown represent the mean ± SD of three groups of five animals each. ***p < 0.001 or **p < 0.01 compared to values observed in the absence of ecotin. Mallory’s phosphotungstic acid hematoxylinstaining of mouse lung before the thrombin-induced acute pulmonary embolism model. Fibrin formed in the blood vessel stained an intense blue, (C and D) pulmonary thromboembolisms were evaluated in control animals, which only received PBS, (E and F) in the group treated with ecotin-WT at a dose of 1 mg/kg and (G and H) in the ecotin-RR treated animals with a dose of 1 mg/kg, (I and J) and in the ecotin-RR treated animals with dose of 0.5 mg/kg. Minor and major magnification on the left and right, respectively. Asterisk represents the completely or partially clear vessels. Arrows show completely obstructed vessels.
compared with the control (Fig. 4A). After, we treated the mice with ecotin-RR at a lower dose of 0.5 mg/kg, which also prevented animals’ death (Fig. 4B). The lung tissues of ecotin-treated mice had a less fibrin-rich thrombus when compared to the control tissues histologically (Fig. 4C–H). The thrombus is clearly observed in the control tissues as an intense blue color after staining and histological analysis (Fig. 4C and D). Taken together, these results indicate that ecotin-WT and ecotin-RR are able to inhibit in vivo thrombosis.
3.4. Bleeding effect The most common side effect of the currently available antithrombotic drugs is their effect on bleeding. Using a tail transection mouse model, ecotin-WT and ecotin-RR were tested to determine their side effects. Control mice were treated with the drug vehicle (PBS buffer) or heparin. Interestingly, mice treated with 1 mg/kg of either ecotin-WT or ecotin-RR, the same concentration that inhibits 100% of thrombus formation, did not result in any
L.S. Wermelinger et al. / International Journal of Biological Macromolecules 78 (2015) 296–303
Fig. 5. Bleeding effect of ecotins. Ecotins (1 mg/kg) were administered i.v. and after 5 min, the rat tail was cut 3 mm from the tip and carefully immersed in 40 mL of distilled water at room temperature. The blood loss (hemoglobin content) was estimated at 540 nm after 60 min. The absorbance detected for a group that received PBS or 750 UI heparin instead of ecotin was used as a negative and a positive control, respectively. The results represent the mean ± SD of 5–11 animals. ***p < 0.001 compared to values observed in the absence of ecotin.
increased bleeding compared to untreated mice; in contrast, mice treated with 750 UI/kg (5 mg/kg) heparin, as a positive control, had a significant increase in bleeding, as expected (Fig. 5). 4. Discussion Although there are currently many drugs available to treat and prevent thrombotic events, the clinically used drugs present some side effects, such as a risk of bleeding, thrombocytopenia, food interaction and an intense laboratory evaluation during the treatment. To evaluate the anticoagulant effect and safety of new potential drugs, we evaluated two serine protease inhibitors (ecotin-WT and ecotin-RR) as new antithrombotic components, using in vitro and in vivo assays. In this study, we verified that both ecotins have important antithrombotic activity without bleeding side effects. We previously used purified enzymes and showed that ecotinRR conserved the ability to inhibit trypsin and FXa, beyond thrombin improved inhibition. It is known that the mutations M84R and M85R change the primary site of ecotin, allowing it to interact with the catalytic site of thrombin [21]. Our data revealed that the two mutations (Met84Arg and Met85Arg) in the variant primary binding site were crucial in directly modulating the potency of ecotin-RR, which demonstrated an efficient inhibition. Ecotin-RR, at a final concentration of 25 nM, was able to inhibit approximately 75% of thrombin (2 nM) activity immediately after its addition. While the ecotin-R (Met84Arg) mutant inhibited 100% of thrombin (3.7 nM) activity but with four times greater protein concentration (100 nM) and 1 h of incubation [15]. In addition, ecotin-RR prevents 100% of fibrinocoagulation at a low dose of 12.5 nM. This concentration is considered to be more physiologically relevant in vivo. Furthermore, ecotin-RR has an IC50 of 46 nM to inhibit FXa similar to ecotin-R [15]. These results are consistent with the literature regarding the mutation of the primary site where inhibition that occurs with ecotin-RR (Met84Arg and Met85Arg) improves its interactions with thrombin and a thrombin-like catalytic site using an in silico study and some in vitro enzymatic experiments [21]. It is important to emphasize that the IC50 provide an apparent inhibition constant under the conditions described and these identical conditions were used for all inhibition studies, making them relative to one another and valid comparisons [37].
301
The results of the in vitro coagulation plasma assays of the ecotins revealed a prolonged PT time and APTT with a similar profile, despite the first assay being triggered by an extrinsic pathway inducer and the second by an intrinsic pathway inducer; both were oriented in the same way. This result is most likely due to the ecotins inhibitory activity toward FXa, common in both pathways. Currently, anticoagulants, such as unfractionated heparin (UFH), the low molecular weight heparins (LMWHs), synthetic pentasaccharide fondaparinux and vitamin K antagonists (VKAs), have some interactions with food and drugs, need monitoring and present a bleeding risk [10]. Attention has been focused on the development of safer and more efficient molecules. Dabigatran, a direct thrombin inhibitor, and rivaroxaban, a direct factor Xa inhibitor, have been licensed in many countries [11,38]. The advantage of direct thrombin inhibitors relies in its capability of blocking its interaction with substrates, thus preventing fibrin formation, the activation of factors V, VIII, XI or XIII, and platelet aggregation. Also direct thrombin inhibitors can be more effectively because they inactivate fibrin-bound thrombin thus attenuating thrombus formation. On the other hand, it was hypothesized that selective inhibition of coagulation factors located upstream of thrombin in the coagulation pathway might be safer in relation to bleeding risk; by not inhibiting thrombin activity directly, the drugs would allow some amount of thrombin to escape neutralization and thereby participate in hemostasis. Therefore factor Xa is also an attractive target for anticoagulants because it is the rate-limiting component in the generation of thrombin. Hence, the inhibition of both targets, such as FXa and FIIa, represent a notable strategy to prevent coagulation without bleeding side effects. Studies have found that dabigatran can prevent the incidence of recurrent venous thromboembolism (VTE) with no difference in the incidence of major bleeding [39]. Similarly, only ecotinRR was able to prevent death by thromboembolism due to the hypercoagulability induced by thrombin administration, in a dosedependent manner (i.v. 0.5–1 mg/kg); this result was confirmed by lung histology of the animals. As described by Gresele et al. and reinforced by Momi et al. the thrombin-thromboembolic model in mice induced the thrombus at least two main mechanisms. First, thrombin exogenous can be directly convert fibrinogen to fibrin that promotes intravascular fibrin accumulation in the lung. In addition, thrombin exogenous can activate a contact phase by the activation of Factor XI-mediated mechanism leading to additional endogenous thrombin formation by feedback. The endogenous thrombin formed would play a major role in making intravascular deposit fibrin more resistant to fibrinolysis by activation of Factor XIII [34,40]. Despite of the inhibitory effects of ecotin-WT toward thrombin, only ecotin-RR prevented death in approximately 80% of the animals in the thromboembolic model induced by thrombin and did not show a bleeding effect in the same dose. These results suggest that ecotin-RR could be a good template for developing a molecule to directly inhibit thrombin and prevent a recurrent thromboembolism without a bleeding side effect, until 1 h after the treatment. Rivaroxaban, the factor-Xa inhibitor that has a favorable balance of efficacy and safety for preventing venous thromboembolism after major orthopedic surgery, is a direct inhibitor of FXa, like the ecotins [41,42]. When we compared the activity of ecotins with rivaroxaban in venous animal models, we observed a similar activity. Rivaroxaban has potent antithrombotic properties (i.v. 0.3 mg/kg dose leads to an almost complete inhibition of thrombus formation) in stasis and hypercoagulability models and does not prolong bleeding in animal models. Bleeding time of the animals was observed after a dose of 3 mg/kg, and the activity was not different from the animals without treatment, suggesting that the molecule is safe [43]. Using a similar model, we verified that ecotins have an antithrombotic effect with a similar potency; both ecotins
302
L.S. Wermelinger et al. / International Journal of Biological Macromolecules 78 (2015) 296–303
were able to inhibit thrombosis in a dose and time dependent manner (i.v. 0.5 mg/kg dose leading to an almost complete inhibition of thrombus formation) until 30 min. This activity was expected because our data show the same profile for both ecotins in the PT assay, which uses the same fuel (thromboplastin). Then, with a higher dose (1 mg/kg), neither ecotins had an effect on bleeding. Remarkably, ecotins WT and RR displayed notable potency (i.v. dose 0.5 mg/kg inhibited more than 95% of thrombus formation) compared to other antithrombotic proteins from animal sources, such as bothrojaracin (inhibits FII and FIIa), and hirudin (inhibits FIIa). Bothrojaracin is a 27 kDa protein and is a thrombin inhibitor isolated from the Bothrops jararaca venom by our group [44]. This protein interacts with thrombin through the thrombin binding exosite I and II, without interacting with the catalytic site of thrombin and is also able to interact with prothrombin [45,46]. Bothrojaracin requires an i.v. dose of 1 mg/kg to prevent at least 95% of thrombus formation in a deep vein thrombosis model (DVTM), and the same dose is able to prevent 100% of the mortality in a thromboembolic model induced by thrombin in contrast to the only 80% protection displayed in this work by ecotin-RR [14]. Hirudin is a peptide approximately 7 kDa in size isolated from the glands of Hirudo medicinalis that has thrombin inhibitor activity, interacts with exosite I and directly interacts with the catalytic site. Recombinant hirudin is administered intravenously, and this molecule is indicated for anticoagulation in patients with heparininduced thrombocytopenia [12,47]. In a deep vein thrombosis model with hypercoagulability and stasis, hirudin is able to prevent approximately 80% of thrombus formation with a dose of 0.3 mg/kg, showing a similar profile to the ecotins. In these models, Herbert and colleagues induced hypercoagulability with six times less thromboplastin (0.5 mg/kg), which most likely requires less inhibitor to prevent thrombus formation, and a dose of 0.3 mg/kg of hirudin was effective [48]. In a thromboembolic model induced by bovine thrombin, hirudin prevented 100% of death in animals with a dose of 0.4 mg/kg, demonstrating more activity in this thrombosis model compared with ecotin-RR from our study because we needed 1 mg/kg to prevent 80% of death [40]. However, bothrojaracin (1 mg/kg) and hirudin (0.1 mg/kg) cause bleeding side effects when tested at a dose that promotes an antithrombotic effect, which was not observed with ecotin (1 mg/kg) treatment [14,40]. These observations lead us to believe that these molecules should be used in a more specific manner to prevent thrombus formation in patients who require surgery or invasive procedures because ecotin is potent and rapidly active and has reduced bleeding side effects. Interestingly, ecotin-WT and ecotin-RR prevent thrombus formation in a model that combines stasis and hypercoagulability without bleeding. Both molecules displayed good activity with an i.v. dose of 0.5 mg/kg, which inhibited more than 95% of thrombus formation. In addition, only ecotin-RR was able to decrease death due hypercoagulability induced by thrombin. The antithrombotic activity with absence of bleeding can be explained by the ability to inhibit FXa because these two ecotins have an antithrombotic profile similar to rivaroxaban. In addition, ecotin-WT was able to inhibit other enzymes in the hemostatic system, such as FXIa, FXIIa and kallikrein [15,20]. It is important to emphasize that ecotin RR also inhibit uPa [23]. Nevertheless, this blockage has no relevance in the thrombosis models used in our study, since it impairs the thrombus formation preventing the activation of fibrinolytic pathway. The literature indicates that inhibition of these molecules can be efficient and safer since they do not promote bleeding, suggesting that ecotins could have a safer profile because they can also inhibit factors from the intrinsic cascade [49]. In conclusion, our results suggests that ecotin-WT prevents thrombus formation through the inhibition of Factor Xa or
upstream, while ecotin-RR, by its potent inhibition of thrombin, is more efficient when the thrombus is associated with thrombin burst. Thus, both ecotins displayed potent in vivo antithrombotic effects that were not associated with bleeding. These results demonstrated the potential of those molecules as templates for the design of new anticoagulation molecules. It is though essential to understand all the potential activity of ecotins, since ecotin is a foreign protein that could induce an immunological response, it would be desirable further studies to develop a prototype that conserves it biological activity and safety. Acknowledgments This work was supported by Conselho Nacional de Desenvolvimento Científico (CNPq), Fundac¸ão de Amparo à Pesquisa do Estado do Rio de Janeiro (FAPERJ), Coordenac¸ão de Aperfeic¸oamento de Pessoal de Nível Superior (CAPES), Universidade Federal Fluminense (UFF) (Brazil) and Department of Defense grant PC111318 from USA. We would like to thank Dione Maria da Silva and Ana Lúcia de Oliveira Carvalho for their technical assistance. The kind donation of heparin was provided by Dr. Paulo Mourão (Universidade Federal do Rio de Janeiro, RJ, Brazil). References [1] A.S. Wolberg, M.M. Aleman, K. Leiderman, K.R. Machlus, Anesth. Analg. 114 (2012) 275–285. [2] B. Furie, Am. Soc. Hematol. Educ. Program (2009) 255–258. [3] S. Liao, T. Woulfe, S. Hyder, E. Merriman, D. Simpson, S. Chunilal, J. Thromb. Haemost. 12 (2014) 214–219. [4] S.Z. Goldhaber, H. Bounameaux, Lancet Haematol. 379 (2012) 1835–1846. [5] I.A. Naess, S.C. Christiansen, P. Romundstad, S.C. Cannegieter, F.R. Rosendaal, J. Hammerstrøm, J. Thromb. Haemost. 5 (2007) 692–699. [6] E.I. Sinauridze, M.A. Panteleev, F.I. Ataullakhanov, Blood Coagul. Fibrinolysis 23 (2012) 482–493. [7] R.C. Becker, J. Alexander, C.K. Dyke, R.A. Harrington, Thromb. Haemost. 92 (2004) 1182–1193. [8] M. Brodmann, Hamostaseologie 33 (2013) 218–224. [9] M. Coppens, J.W. Eikelboom, D. Gustafsson, J.I. Weitz, J. Hirsh, Circ. Res. 111 (2012) 920–929. [10] P.M. Mannucci, M. Levi, N. Engl. J. Med. 356 (2007) 2301–2311. [11] T. Baglin, J. Thromb. Haemost. 11 (2013) 122–128. [12] F. Markwardt, Thromb. Res. 74 (1994) 1–23. [13] I.M. Francischetti, J.G. Valenzuela, J.F. Andersen, T.N. Mather, J.M. Ribeiro, Blood 99 (2002) 3602–3612. [14] R.B. Zingali, M.S. Ferreira, M. Assafim, F.S. Frattani, R.Q. Monteiro, Pathophysiol. Haemost. Thromb. 34 (2005) 160–163. [15] J.L. Seymour, R.N. Lindquist, M.S. Dennis, B. Moffat, D. Yansura, D. Reilly, M.E. Wessinger, R.A. Lazarus, Biochemistry 33 (1994) 3949–3958. [16] M.E. McGrath, W.M. Hines, J.A. Sakanari, R.J. Fletterick, C.S. Craik, J. Biol. Chem. 266 (1991) 6620–6625. [17] M.E. McGrath, T. Erpel, C. Bystroff, R.J. Fletterick, EMBO J. 13 (1994) 1502–1507. [18] S.Q. Yang, C.I. Wang, S.A. Gillmor, R.J. Fletterick, C.S. Craik, J. Mol. Biol. 279 (1998) 945–957. [19] C.H. Chung, H.E. Ives, S. Almeda, A.L. Goldberg, J. Biol. Chem. 258 (1983) 11032–11038. [20] A.A. Stoop, C.S. Craik, Nat. Biotechnol. 21 (2003) 1063–1068. [21] H.C. Castro, D.M. Silva, C.S. Craik, R.B. Zingali, Biochim. Biophys. Acta 1547 (2001) 183–195. [22] H.C. Castro, R.Q. Monteiro, M. Assafim, N.I. Loureiro, C.S. Craik, R.B. Zingali, Int. J. Biochem. Cell Biol. 38 (2006) 1893–1900. [23] C.I. Wang, Q. Yang, C.S. Craik, J. Biol. Chem. 270 (1995) 12250–12256. [24] M.C. Laboissière, M.M. Young, R.G. Pinho, S. Todd, R.J. Fletterick, I. Kuntz, C.S. Craik, J. Biol. Chem. 277 (2002) 26623–26631. [25] P.C. Sathler, C.S. Craik, T. Takeuchi, R.B. Zingali, H.C. Castro, Appl. Biochem. Biotechnol. 160 (2010) 2355–2365. [26] J.S. Ulmer, R.N. Lindquist, M.S. Dennis, R.A. Lazarus, FEBS Lett. 365 (1995) 159–163. [27] A.A. Stoop, R.V. Joshi, C.T. Eggers, C.S. Craik, Biol. Chem. 391 (2010) 425–433. [28] C.I. Wang, Q. Yang, C.S. Craik, Methods Enzymol. 267 (1996) 52–68. [29] P.K. Ngai, J.Y. Chang, Biochem. J. 280 (1991) 805–808. [30] A.K. Mukherjee, S.P. Mackessy, Biochim. Biophys. Acta 1830 (2013) 3476– 3488. [31] X. Wang, Z. Zhang, Z. Yao, M. Zhao, H. Qi, Int. J. Biol. Macromol. 58 (2013) 225–230. [32] J.C. Martinichen-Herrero, E.R. Carbonero, G.L. Sassaki, P.A. Gorin, M. Iacomini, Int. J. Biol. Macromol. 35 (2005) 97–102.
L.S. Wermelinger et al. / International Journal of Biological Macromolecules 78 (2015) 296–303 [33] W. Mendes-Silva, M. Assafim, B. Ruta, R.Q. Monteiro, J.A. Guimarães, R.B. Zingali, Thromb. Res. 112 (2003) 93–98. [34] P. Gresele, S. Momi, M. Berrettini, G.G. Nenci, H.P. Schwarz, N. Semeraro, M. Colucci, J. Clin. Invest. 101 (1998) 667–676. [35] J.M. Herbert, J.P. Hérault, A. Bernat, R.G. van Amsterdam, G.M. Vogel, J.C. Lormeau, M. Petitou, D.G. Meuleman, Circ. Res. 79 (1996) 590–600. [36] J.C. Santos, J.M. Mesquita, C.L. Belmiro, C.B. da Silveira, C. Viskov, P.A. Mourier, M.S. Pavão, Thromb. Res. 121 (2007) 213–223. [37] C.T. Eggers, S.X. Wang, R.J. Fletterick, C.S. Craik, J. Mol. Biol. 308 (2001) 975–991. [38] J.I. Weitz, Am. J. Hematol. 87 (2012) 133–136. [39] S. Schulman, B. Ritchie, J.K. Goy, S. Nahirniak, M. Almutawa, S. Ghanny, Br. J. Haematol. 164 (2014) 308–310. [40] S. Momi, M. Nasimi, M. Colucci, G.G. Nenci, P. Gresele, Haematologica 86 (2001) 297–302. [41] B.I. Eriksson, L.C. Borris, R.J. Friedman, S. Haas, M.V. Huisman, A.K. Kakkar, T.J. Bandel, H. Beckmann, E. Muehlhofer, F. Misselwitz, W. Geerts, N. Engl. J. Med. 358 (2008) 2765–2775.
303
[42] W.D. Fisher, B.I. Eriksson, K.A. Bauer, L. Borris, O.E. Dahl, M. Gent, S. Haas, M. Homering, M.V. Huisman, A.K. Kakkar, P. Kälebo, L.M. Kwong, F. Misselwitz, A.G. Turpie, Thromb. Haemost. 97 (2007) 931–937. [43] E. Perzborn, J. Strassburger, A. Wilmen, J. Pohlmann, S. Roehrig, K.H. Schlemmer, A. Straub, J. Thromb. Haemost. 3 (2005) 514–521. [44] R.B. Zingali, M. Jandrot-Perrus, M.C. Guillin, C. Bon, Biochemistry 32 (1993) 10794–10802. [45] V. Arocas, C. Lemaire, M.C. Bouton, A. Bezeaud, C. Bon, M.C. Guillin, M. JandrotPerrus, Thromb. Haemost. 79 (1998) 1157–1161. [46] V. Arocas, R.B. Zingali, M.C. Guillin, C. Bon, M. Jandrot-Perrus, Biochemistry 35 (1996) 9083–9089. [47] C.C. Liu, E. Brustad, W. Liu, P.G. Schultz, J. Am. Chem. Soc. 129 (2007) 10648–10649. [48] J.M. Herbert, A. Bernat, J.P. Maffrand, Blood 80 (1992) 2281–2286. [49] M.L. van Montfoort, J.C. Meijers, Thromb. Haemost. 110 (2013) 223–232.
