Antithrombotic Effect of Chikusetsusaponin IVa Isolated from Ilex paraguariensis (Maté)

Abstract

The triterpene chikusetsusaponin IVa was isolated from the fruit of Ilex paraguariensis. Using biochemical and pharmacological methods, we demonstrated that chikusetsusaponin IVa (1) prolongs the recalcification time, prothrombin time, activated partial thromboplastin time, and thrombin time of normal human plasma in a dose-dependent manner, (2) inhibits the amidolytic activity of thrombin and factor Xa upon synthetic substrates S2238 and S2222, (3) inhibits thrombin-induced fibrinogen clotting (50% inhibition concentration, 199.4±9.1 μM), and (4) inhibits thrombin- and collagen-induced platelet aggregation. The results also indicate that chikusetsusaponin IVa preferentially inhibits thrombin in a competitive manner (K_i =219.6 μM). Furthermore, when administered intravenously to rats, chikusetsusaponin IVa inhibited thrombus formation in a stasis model of venous thrombosis, although it did not induce a significant bleeding effect. Chikusetsusaponin IVa also prolonged the ex vivo activated partial thromboplastin time. Altogether, these data suggest that chikusetsusaponin IVa exerts antithrombotic effects, including minor hemorrhagic events. This appears to be important for the development of new therapeutic agents.

Introduction

T hrombosis and cardiovascular diseases are the major causes of morbidity and mortality in the Western world and frequently involve a prothrombotic state such as myocardial infarction, stroke, and atherosclerosis.¹ Anticoagulant and antithrombotic agents, such as heparin, vitamin K antagonists, and acetylsalicylic acid, are currently used to treat and prevent these thromboembolic events.² Considering the high incidence of side effects (e.g., gastrointestinal bleeding and thrombocytopenia), important efforts have been made to find new antithrombotic substances.^2

–5

In the context of hemostasis, glycyrrhizin [3-O-(2-O-β-d-glucopyranuronosyl-α-d-glucopyranuronosyl)-18β-glycyrrhetinic acid (GL)] (Fig. 1), a triterpenoid saponin obtained from the roots of Glycyrrhiza glabra, presented anticoagulant and antithrombotic effects. GL prolonged the in vitro coagulation time and inhibited the thrombin-induced clotting of plasma and fibrinogen.⁶ In addition, the GL antithrombotic effect was also observed in animal models of thrombosis.⁷

FIG. 1.

Chemical structures of chikusetsusaponin IVa and glycyrrhizin. Glc, glucose; GluA, glucuronic acid.

Ilex paraguariensis, a South American tree of the Aquifoliaceae family, is rich in saponins (approximately 10%).⁸ Dried leaves and twigs from this tree are used to prepare a tea known as maté that is commonly consumed in Brazil, Uruguay, Paraguay, and Argentina. The saponins isolated from its leaves are glycosides derived mainly from ursolic and oleanolic acids.⁸ Other triterpenoid glycosides were isolated from the fruits of I. paraguariensis.⁹ Pentacyclic triterpenoids such as ursolic and oleanolic acids and their derivatives possess pharmacological properties, such as anti-human immunodeficiency virus, hepatoprotective, anti-inflammatory, antimalarial, antiglycation, and cytotoxic effects,^10
–12 suggesting their potential use for the design of new bioactive compounds. In this sense, the triterpenoid chikusetsusaponins (identified by numbers I–IV) possess several pharmacological activities such as anti-inflammatory,¹³ antiviral,¹⁴ fibrinolytic,¹⁵ and anticancer^16
–18 properties. In particular, chikusetsusaponin IVa has also been isolated from other species of Ilex, such as Ilex dumosa,¹⁹ Ilex pubescens,¹³ and Ilex rotunda.²⁰

In the present study, we describe the in vitro and in vivo anticoagulant and antiplatelet activity of chikusetsusaponin IVa (Fig. 1).

Materials and Methods

Drugs and reagents

Bovine fibrinogen, human factor Xa, bovine type I collagen, ADP, and GL were purchased from Sigma-Aldrich (St. Louis, MO, USA). Arachidonic acid was purchased from Chrono-Log Co. (Havertown, PA, USA). Human α-thrombin was purified from plasma of healthy volunteer donors according to the precedure of Ngai and Chang.²¹ Synthetic substrates for thrombin (S2238 [H-d-Phe-Pip-Arg-p-nitroanilide]) and factor Xa (S2222 [Benzyl-Ile-Glu-Gly-Arg-p-nitroanilide]) were purchased from Chromogenix (Milan, Italy). Anasedan^® (xylazine) and Dopalen^® (ketamine) were purchased from Cristália Produtos Químicos Farmacêuticos (São Paulo, Brazil). Calcium thromboplastin was purchased from Wiener Lab (Rosario, Argentina).

Apparatus

Mass spectrometry experiments were carried out using a Micromass/Waters (Milford, MA, USA) quadrupole-time of flight mass spectrometer microequipped with a nano-electrospray ionization source. Data were analyzed with the Waters MassLynx software. ¹H and ¹³C nuclear magnetic resonance were recorded on a Varian (Palo Alto, CA, USA) Inova spectrometer (300 MHz) in CD₃OD. The spectrophotometer used was a SpectraMax (Molecular Devices Co., Sunnyvale, CA, USA), equipped with temperature control and shaking systems.

Isolation and structural elucidation of saponin

Fruits of I. paraguariensis A. St. Hil. were harvested in a cultivated area in Rio Grande do Sul, Brazil. A voucher specimen was deposited at the Herbarium of the Federal University of Rio Grande do Sul, Porto Alegre, RS, Brazil, as voucher number ICN/UFRGS (163413). The hydroethanolic extract was prepared using 10 g of dried and powdered fruits in 40% ethanol by maceration (1:10, plant:solvent). After evaporation of the solvent, the extract was subjected to chromatography on a silica gel column (Merck, Darmstadt, Germany) using gradient mixtures of chloroform:ethanol:water. Fractions were collected and pooled according to thin-layer chromatography (silica gel GF₂₅₄; chloroform:ethanol:water, 40:40:5 by volume). Then pooled fractions were subjected to further purification, separately, using Lichroprep^® (Merck; particle size, 40–63 μm) and gradient mixtures of water:ethanol as eluate. Pure saponin was obtained (90 mg), corresponding to 2% of the dried ethanolic extract.

Chikusetsusaponin IVa

The identification of saponin in its native form was mostly deduced from a combination of mass spectrometry and nuclear magnetic resonance spectroscopy. This combination allowed the identification of an already known saponin named chikusetsusaponin IVa, previously isolated by us from leaves of I. dumosa. ¹⁹ It was obtained as an amorphous powder; its main peak in the mass spectrum was m/z 817.4354 Da, corresponding to the chemical species [(C₄₂H₆₆O₁₄) Na]⁺ (expected m/z, 817.4350 Da), together with the peak m/z 833.4175 Da [(C₄₂H₆₆O₁₄) K]⁺ (Fig. 2). ¹H and ¹³C nuclear magnetic resonance (300 MHz, CD₃OD) data were similar to those previously described.¹⁹

FIG. 2.

Mass spectrum of chikusetsusaponin IVa using a Micromass/Waters quadrupole-time of flight (TOF) mass spectrometer (MS) equipped with a nano-electrospray (ES) ionization source.

Effects of chikusetsusaponin IVa on hemostatic system: in vitro studies

Coagulation assays

The anticoagulant activity of chikusetsusaponin IVa (0–2000 μM) was verified by the determination of the following coagulation parameters in plasma: recalcification time (RT), activated partial thromboplastin time (aPTT), prothrombin time (PT), and thrombin time (TT). Anticoagulant experiments were performed using commercial kits following the manufacturer's instructions (Wiener Lab). Human venous blood was collected from healthy volunteer donors in 1:10 (vol/vol) 3.8% trisodium citrate and centrifuged at 1500 g for 10 min to obtain plasma. Assays were conducted using a 96-well microplate SpectraMax as described.²² Results are expressed as the coagulation time (in seconds) and represent the mean±SEM values of three independent experiments performed in quadruplicates.

Amidolytic assay

Saponin (0–2650 μM) was incubated with thrombin (2 μg/mL) or factor Xa (3.4 μg/mL) in 20 mM Tris-HCl (pH 7.5) for 20 min at 37°C. Enzymatic reactions were started by addition of the chromogenic substrate S2238 (0.2 mM) or S2222 (0.2 mM) in a volume of 100 μL for the determination of thrombin and factor Xa activities, respectively. The amount of p-nitroaniline produced was monitored at 405 nm in intervals of 14 s for 30 min using a SpectraMax spectrophotometer. The initial rate of hydrolysis (expressed as milli-optical density units per minute) was used to calculate the percentage of thrombin inhibition. These data were plotted versus the concentration of inhibitor, and a nonlinear regression analysis was performed to calculate the values for concentration causing 50% inhibition (IC₅₀) values. For kinetic analysis, saponin (0–450 μM) was incubated with thrombin prior to the addition of S2238 (50 and 100 μM). Data from the initial rate, expressed as (μM p-nitroaniline generated/min)⁻¹, was plotted versus the inhibitor concentration to construct a Dixon's plot. The curves were analyzed by linear regression to calculate the inhibition constant (K_i ) value and to determine the type of inhibition.²³ Results are presented as mean±SEM values (n=3).

Thrombin-induced fibrinogen clotting

Chikusetsusaponin IVa (0–900 μM) and GL (0–1100 μM) were solubilized in 20 mM Tris-HCl (pH 7.5) and incubated with thrombin (2 μg/mL) for 10 min at 37°C. Reactions were initiated by the addition of fibrinogen (2 mg/mL) in a volume of 100 μL. The kinetics of fibrin formation were measured at 14-s intervals for 30 min at 650 nm using a SpectraMax spectrophotometer. The initial rate of clotting formation (expressed as milli-optical density per minute) was used to calculate the percentage of the thrombin inhibition. These data were plotted versus the concentration of inhibitor, and a nonlinear regression analysis was performed to calculate the IC₅₀ (mean±SEM, n=3).

Platelet aggregation

The platelet function was measured by the in vitro photometric method,²⁴ using a SpectraMax spectrophotometer. Whole human blood was collected from healthy volunteer donors. Washed human platelets were then prepared by gel filtration in a Sepharose 2B column.²⁵ Platelet concentration was adjusted to 350,000 cells/μL in a Neubauer chamber. Platelet aggregation was measured by the decreased rate in washed human platelet optical density in a final volume of 150 μL. The agonists used were as follows: ADP (10 μM), collagen (2.5 μg/mL), arachidonic acid (10 μM), or human thrombin (3 μg/mL). Saponin (0–1000 μM) was preincubated with washed human platelets for 20 min before the addition of agonist. The results are expressed as the percentage area under curves (mean±SEM, n=4).

Effects of chikusetsusaponin IVa saponin on hemostatic system: In vivo studies

Animals

Male Wistar rats (weighing 300–350 g) were housed in a temperature-controlled room (21–25°C, in a 12-h light/dark cycle) with free access to water and food. All procedures involving animals were carried out in accordance with the guiding principles of the International Society of Toxicology and the Brazilian College of Animal Experimentation. The experimental protocol was approved by the Ethical Committee on Research Animal Care of the Federal University of Rio Grande do Sul, Brazil (protocol number 2008177/2009).

In vivo model of deep venous thrombosis

A rat thrombosis model that is a combination of stasis and hypercoagulability induced by a tissue factor-rich component was used according to the literature²⁶ with minor modifications.²⁷ In brief, rats were anesthetized with xylazine (16 mg/kg, i.p.) followed by ketamine (100 mg/kg, i.p.). The abdomen was surgically opened, and the vena cava was exposed and dissected free from surrounding tissues. Subsequently, rats received the following treatments via the left femoral vein (final volume of 0.7 mL): (1) saline, n=4; (2) chikusetsusaponin IVa at 15 and 50 mg/kg, n=4 per dose; or (3) GL at 15 and 50 mg/kg, n=4 per dose. After 3 min, calcium thromboplastin (3 mg/kg) was injected into the vena cava, and stasis was immediately established by the ligation of caudal vena cava (above the insertion point of the right renal vein). The distal ligations of the vena cava, left renal vein, and other major tributaries were conducted 20 min after thromboplastin administration. The isolated segment of caudal vena cava was removed, and the thrombus was separated, rinsed with saline (at 37°C), dried on a filter paper at 60°C (1 h), and weighed. The ratio value of thrombus/rat weight was used for data comparison.

Bleeding effect

Rats were anesthetized as described above, and the following treatments were administered via the left femoral vein (final volume of 0.7 mL): (1) saline, n=4; (2) chikusetsusaponin IVa at 15 and 50 mg/kg, n=4 per dose; or (3) GL at 15 and 50 mg/kg, n=4 per dose. After 5 min, the rat's tail was cut off 3 mm from the tip and carefully immersed in 40 mL of distilled water at room temperature. Blood loss was evaluated 60 min later as a function of hemoglobin concentration in aqueous solution.⁷ Hemoglobin concentration was determined at 540 nm using a standard curve made with hemoglobin. The hemoglobin content in the water of saline-treated animals was used as a control.

Ex vivo determination of aPTT and hemoglobin

Animals were anesthetized as described above, and the following treatments were administered via the left femoral vein (final volume of 0.7 mL): (1) saline, n=4; (2) chikusetsusaponin IVa at 15 and 50 mg/kg, n=4 per dose; or (3) GL at 15 and 50 mg/kg, n=4 per dose. Blood samples were collected from the right femoral vein (0.3 mL in 3.8% trisodium citrate, 1:10 [vol/vol]) at different times posttreatment. Each blood sample was centrifuged, and plasma was used to measure the aPTT as described above. To estimate the hemolytic potential of saponins, hemoglobin content was measured in plasma samples of animals treated with 50 mg/kg chikusetsusaponin IVa or GL. The concentration of free plasma hemoglobin was measured with a kit commercially available following the manufacturer's instructions (Labtest SA, Lagoa Santa, MG, Brazil).

Statistical analysis

Results are expressed as mean±SEM values, and significance was determined by Student's t test. When more than two groups were compared, an analysis of variance was used followed by a Bonferroni's test to compare pairs of means. The data were considered significant with a probability of < .05.

Results

Effects of chikusetsusaponin IVa on hemostatic system: in vitro studies

Chikusetsusaponin IVa showed anticoagulation effects by elongating the time for RT, aPTT, PT, and TT in a dose-dependent manner (Table 1). At 500 μM, a significant inhibition was observed in PT and TT. At 1000 μM, the saponin caused an increase of 1.3-fold in RT, 2.3-fold in aPTT, 1.8-fold in PT, and 1.9-fold in TT compared with respective control values (Table 1). In relation to the hydrolysis of synthetic substrates, chikusetsusaponin IVa inhibited, in a dose-dependent manner, the enzymes tested (S2238 and S2222) with IC₅₀ values of 384.2±44.2 μM for thrombin-catalyzed hydrolysis (Fig. 3A) and 1585.7±96.9 μM for factor Xa-catalyzed hydrolysis (Fig. 3B). Enzyme kinetic studies indicated that chikusetsusaponin IVa inhibited the thrombin catalytic hydrolysis of S2238 in a competitive manner with a K_i value of 219.6±32 μM (Fig. 3C).

FIG. 3.

Effects of chikusetsusaponin IVa on thrombin and factor Xa amidolytic activities. Data are mean±SEM values of three independent determinations. IC₅₀, concentration causing 50% inhibition. (A, B) Effects of increasing concentrations of chikusetsusaponin IVa on the activity of (A) human thrombin upon synthetic substrate S2238 and (B) human factor Xa upon synthetic substrate S2222. (C) Dixon's plot for the inhibitory pattern of chikusetsusaponin IVa on thrombin upon S2238 hydrolysis. Saponin at different concentrations was incubated with thrombin prior to the addition of 50 μM (●) and 100 μM (■) S2238. The K_i value obtained from the graph is indicated.

Table 1.

Effects of Chikusetsusaponin IVa on Blood Coagulation Parameters In Vitro

	Coagulation time (s)
Saponin (μM)	RT	aPTT	PT	TT
0	385.6±9.6	55.7±2.7	17.3±1.6	167.5±7.3
500	374.6±3.1	81.6±2.3	25.5±0.9^*	240.7±4.5^*
1000	485.7±6.6^***	126.5±11.4^**	31.7±2.4^***	330.1±6.3^***
1600	555.5±4.8^***	139.6±3.6^**	40.0±1.5^***	412.2±27.8^***
2000	840.6±25.0^***	158.3±18.6^***	53.3±1.2^***	428.9±1.0^***

After incubation of chikusetsusaponin IVa with human plasma, the following coagulation parameters were determined: recalcification time (RT), activated partial thromboplastin time (aPTT), prothrombin time (PT), and thrombin time (TT). Data are mean±SEM values (in s).

P<.05, ^** P<.01, ^*** P<.001, statistically significant difference compared with coagulation time measured in the absence of saponin (control).

As shown in Figure 4, chikusetsusaponin IVa inhibited the thrombin-induced fibrinogen clotting in a dose-dependent manner with an IC₅₀ value of 199.4±9.1 μM. This inhibitory activity was similar to that obtained for GL (235.7±1.4 μM; Fig. 4). In relation to the platelet aggregation, the saponin had no effect on ADP or on arachidonic acid-induced aggregation (Fig. 5). However, it significantly inhibited collagen- and thrombin-induced platelet aggregation in a dose-dependent manner. The calculated IC₅₀ values for collagen- and thrombin-induced platelet aggregation were 482 and 190 μM, respectively.

FIG. 4.

Effects of chikusetsusaponin IVa and glycyrrhizin on thrombin-induced fibrinogen clotting. Increasing concentrations of chikusetsusaponin IVa or glycyrrhizin were incubated with human thrombin, followed by the addition of fibrinogen. The kinetics of fibrin formation were monitored at 650 nm as described in Materials and Methods. Data are mean±SEM values of three independent determinations.

FIG. 5.

Effects of chikusetsusaponin IVa on platelet aggregation. Increasing concentrations of chikusetsusaponin IVa were incubated with washed human platelets. Platelet aggregation was triggered by the addition of the following agonists: thrombin (THR), collagen (COLL), ADP, or arachidonic acid (AA). Data are mean±SEM values of four independent determinations. **P<.01, ***P<.001, statistically significant difference compared with platelet aggregation in the absence of inhibitor.

Effects of chikusetsusaponin IVa on hemostatic system: in vivo studies

As shown in Figure 6A, chikusetsusaponin IVa reduced the size of the thrombus by 56.2% and 91.2% at doses of 15 and 50 mg/kg, respectively. In comparison, GL was able to reduce the thrombus only 21.2% and 53.1% at doses of 15 and 50 mg/kg, respectively (Fig. 6A). In addition, chikusetsusaponin IVa did not induce a significant bleeding effect after intravascular administration (Fig. 6B). However, GL caused an increase in blood loss of approximately 1.8-fold compared with the control at the dose of 50 mg/kg (Fig. 6B).

FIG. 6.

In vivo antithrombotic and bleeding effects of chikusetsusaponin IVa and glycyrrhizin. (A) The venous thrombosis model. Saponins at the indicated doses were administered intravenously to rats 3 min before the induction of thrombosis by thromboplastin combined with stasis as described in Materials and Methods. Data are mean±SEM values of four animals. *P<.05, ***P<.001, statistically significant difference compared with animals that received saline instead of saponins. (B) The bleeding effect. Rats were injected intravenously with different doses of chikusetsusaponin IVa or glycyrrhizin. Five minutes after injection, the rat's tail was cut off and carefully immersed in distilled water. Blood loss was evaluated 60 min later by measuring the hemoglobin (Hb) concentration in aqueous solution. Data are mean±SEM values of four animals. Statistically significant differences (P<.001) are indicated for comparison with *animals that received only saline solution and ^#animals that received 50 mg/kg glycyrrhizin.

The anticoagulant effect of chikusetsusaponin IVa observed in vitro was confirmed by the measurement of aPTT ex vivo. Compared with the control, chikusetsusaponin IVa at 50 mg/kg was responsible for a significant prolongation of aPTT at different times following intravascular administration (Table 2). A similar result was observed after the injection of GL at 50 mg/kg (Table 2). To verify the hemolytic potential of chikusetsusaponin IVa and GL, we measured the hemoglobin content in the plasma of animals previously treated with saponins. No signs of hemolysis were observed in the plasma of animals treated with 50 mg/kg GL (control, 43.21±8.2 mg/dL; treated, 43.35±4.5 mg/dL) or chikusetsusaponin IVa (control, 42.11±5.6 mg/dL; treated, 40.75±4.7 mg/dL).

Table 2.

Ex Vivo Anticoagulation Effect of Chikusetsusaponin IVa

	aPTT (s)
Treatment, dose (mg/kg)	5 min	15 min	30 min	60 min
Saline	17.2±0.2	16.8±0.2	16.8±0.6	17.4±0.6
Glycyrrhizin
15	17±0.2	17.7±0.4	17.2±0.3	16.8±0.5
50	16.6±0.2	19.6±1.0	20.4±0.7^*	25±0.3^***
Chikusetsusaponin IVa
15	18.6±0.9	17.6±0.6	19.1±0.8	14.8±1.1
50	20.5±0.3^***	24±0.6^***	21.3±1.4^*	20.4±0.2^*

Ex vivo aPTT was measured in plasma of rats 5, 15, 30 and 60 min after intravenous injection of saline (control), glycyrrhizin, or chikusetsusaponin IVa. Data are mean±SEM values (in s).

P<.05, ^*** P<.001, statistically significant difference compared with coagulation time of the control group.

Discussion

Herein, the anticoagulant, antiplatelet, and in vivo antithrombotic activities of chikusetsusaponin IVa were described for the first time. When tested in vitro, chikusetsusaponin IVa caused a dose-dependent prolongation of RT, aPTT, and PT, suggesting that the major targets for this compound were factors Xa, Va, and thrombin. Additionally, chikusetsusaponin IVa was able to inhibit the TT, indicating thrombin as a potential target for this saponin in plasma. A dose-dependent inhibition of thrombin- and factor Xa-catalyzed hydrolysis of specific synthetic substrates (S2238 and S2222, respectively) was observed after incubation with chikusetsusaponin IVa. This observation confirms that the main targets of this saponin are common coagulation factors. Moreover, the IC₅₀ value for factor Xa inhibition was four times higher than that for thrombin inhibition, indicating a preference for thrombin over factor Xa. We demonstrated that chikusetsusaponin IVa is a competitive inhibitor of thrombin, similar to other thrombin inhibitors such as the thrombin receptor peptide or glycoprotein GPIb that are known to act in the micromolar range.^28,29

Chikusetsusaponin IVa also inhibited thrombin-induced fibrinogen clotting activity with an IC₅₀ that was two times lower than that for the thrombin-catalyzed hydrolysis of S2238. These values indicate that chikusetsusaponin IVa most likely interacts with sites in the thrombin molecule other than the catalytic site because a more potent inhibition was observed when the macromolecular substrate fibrinogen was used. Chikusetsusaponin IVa has an in vitro antiplatelet effect when induced by thrombin and collagen as well as an in vivo antithrombotic property. In comparison with GL, chikusetsusaponin IVa had a significantly more potent antithrombotic effect at the doses of 15 and 50 mg/kg. In addition, chikusetsusaponin IVa had no bleeding effect, which may be considered an advantage compared with known antithrombotic agents such as hirudin, argatroban, heparin, or acetylsalicylic acid.³⁰ Furthermore, intravenous injection of chikusetsusaponin IVa or GL had no toxic hemolytic effect in treated animals and caused only a slight increase in ex vivo aPTT. Moreover, this saponin had a significant antithrombotic activity at a dose of 15 mg/kg without affecting the aPTT. Recently, it was reported that potent orally active factor Xa inhibitors such as darexaban and darexaban glucuronide are able to inhibit thrombus formation and prolong the coagulation time in vivo without inducing any bleeding effect.³¹ It was reported¹⁵ that chikusetsusaponins III, IV, and V increased the in vitro activity of urokinase, an important plasminogen activator during fibrinolysis. Possibly, the increase in fibrinolysis induced by chikusetsusaponins contributed to the antithrombotic and anticoagulant effects described in this research.

In conclusion, chikusetsusaponin IVa is a new plant-derived antithrombotic compound that increases the coagulation time and promotes antiplatelet and antithrombotic activity.

Footnotes

Acknowledgments

We are grateful to the Conselho Nacional de Desenvolvimento Científico e Tecnológico, Ministério de Ciência e Tecnologia and to the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior, Ministério da Educação, Brasília, DF, Brazil, for their financial support and fellowships.

Author Disclosure Statement

No competing financial interests exist.

References

Weitz

, Hirsh

, Samama

. New antithrombotic drugs: American College of Chest Physicians evidence-based clinical practice guidelines (8th edition) Chest, 2008; 133,6 Suppl:234S–256SErratum in Chest 2008;134:473.

Laux

, Perzborn

, Heitmeier

et al.

Direct inhibitors of coagulation proteins—the end of the heparin and low-molecular-weight heparin era for anticoagulant therapy?

Thromb Haemost, 2009; 102:892–899.

Toschi

, Lettino

. Inhibitors of propagation of coagulation: factors V and X. Br J Clin Pharmacol, 2011; 72:563–580.

Newman

, Cragg

. Natural products as sources of new drugs over the last 25 years. J Nat Prod, 2007; 70:461–477.

Radnaabazar

, Park

, Kim

, Cha

, Song

Y-S

. Fibrinolytic and antiplatelet aggregation properties of a recombinant Cheonggukjang kinase. J Med Food, 2011; 14:625–629.

Francischetti

, Monteiro

, Guimarães

. Identification of glycyrrhizin as a thrombin inhibitor. Biochem Biophys Res Commun, 1997; 235:259–263.

Mendes-Silva

, Assafim

, Ruta

, Monteiro

, Guimarães

, Zingali

. Antithrombotic effect of glycyrrhizin, a plant-derived thrombin inhibitor. Thromb Res, 2003; 112:93–98.

Heck

, Mejia

. Yerba mate tea (Ilex paraguariensis): a comprehensive review on chemistry, health implications, and technological considerations. J Food Sci, 2007; 72:138–151.

Taketa

, Breitmaier

, Schenkel

. Triterpenes and triterpenoidal glycosides from the fruits of Ilex paraguariensis (maté) J Braz Chem Soc, 2004; 15:205–211.

10.

Liu

. Pharmacology of oleanolic acid and ursolic acid. J Ethnopharmacol, 1995; 49:57–68.

11.

Yogeeswari

, Sriram

. Betulinic acid and its derivatives: a review on their biological properties. Curr Med Chem, 2005; 12:657–666.

12.

Gnoatto

, Susplugas

, Vechia

et al. Pharmacomodulation on the 3-acetylursolic acid skeleton: design, synthesis and biological evaluation of novel N-{3-[4-(3-aminopropyl)piperazinyl]propyl}-3-O-acetylursolamide derivatives as antimalarial agents. Bioorg Med Chem, 2008; 16:771–782.

13.

Wang

, Zhou

, Jiang

, Liu

. Two new triterpene saponins from the anti-inflammatory saponin fraction of Ilex pubescens root. Chem Biodivers, 2008; 5:1369–1376.

14.

Rattanathongkom

, Lee

, Hayashi

, Sripanidkulchai

, Kanchanapoom

, Hayashi

. Evaluation of chikusetsusaponin IVa isolated from Alternanthera philoxeroides for its potency against viral replication. Planta Med, 2009; 75:829–835.

15.

Matsuda

, Samukawa

, Fukuda

, Shiomoto

, Tung

, Kubo

. Studies of Panax japonicus fibrinolysis. Planta Med, 1989; 55:18–21.

16.

Yoo

, Kwon

, Park

. The cytotoxic saponin from heat-processed Achyranthes fauriei roots. Biol Pharm Bull, 2006; 29:1053–1055.

17.

Hosono-Nishiyama

, Matsumoto

, Kiyohara

, Nishizawa

, Atsumi

, Yamada

. Suppression of Fas-mediated apoptosis of keratinocyte cells by chikusetsusaponins isolated from the roots of Panax japonicus. Planta Med, 2006; 72:193–198.

18.

Liang

, Ding

, Tung Nguyen

et al. Oleanane-type triterpenoids from Panax stipuleanatus and their anticancer activities. Bioorg Med Chem Lett, 2010; 2023:7110–7115.

19.

Pires

, Guillaume

, Gosmann

, Schenkel

. Saponin from Ilex dumosa, an Erva-maté (Ilex paraguariensis) adulterating plant. J Agric Food Chem, 1997; 45:1027–1031.

20.

Amimoto

, Yoshikawa

, Arihara

. Triterpenoid saponins of Aquifoliaceous plants. VIII. Ilexosides XXIX–XXXII from the leaves of Ilex rotunda Thunb. Chem Pharm Bull, 1992; 40:3138–3141.

21.

Ngai

, Chang

. A novel one-step purification of human α-thrombin after direct activation of crude prothrombin enriched from plasma. Biochem J, 1991; 280:805–808.

22.

Ribeiro

, Schneider

, Guimarães

. Purification and characterization of prolixin S (nitrophorin 2), the salivary anticoagulant of the blood-sucking bug Rhodnius prolixus. Biochem J, 1995; 308:243–249.

23.

Dixon

. The determination of enzyme inhibitor constants. Biochem J, 1953; 55:170–171.

24.

Born

, Cross

. The aggregation of blood platelets. J Physiol, 1963; 168:178–195.

25.

Berger

, Reck

Jr , Terra

, Pinto

, Termignoni

, Guimarães

. Lonomia obliqua caterpillar envenomation causes platelet hypoaggregation and blood incoagulability in rats. Toxicon, 2010; 55:33–44.

26.

Vogel

, Meuleman

, Bourgondiën

, Hobbelen

. Comparison of two experimental thrombosis models in rats effects of four glycosaminoglycans. Thromb Res, 1989; 54:399–410.

27.

Reck

Jr , Berger

, Marks

et al. Pharmacological action of tick saliva upon haemostasis and the neutralization ability of sera from repeatedly infested hosts. Parasitology, 2009; 136:1339–1349.

28.

De Marco

, Mazzucato

, Masotti

, Ruggeri

. Localization and characterization of an alpha-thrombin-binding site on platelet glycoprotein Ib alpha. J Biol Chem, 1994; 269:6478–6484.

29.

Bouton

, Jandrot-Perrus

, Moog

, Cazenave

, Guillin

, Lanza

. Thrombin interaction with a recombinant N-terminal extracellular domain of the thrombin receptor in an acellular system. Biochem J, 1995; 305:635–641.

30.

Schulman

, Spencer

. Antithrombotic drugs in coronary artery disease: risk benefit ratio and bleeding. J Thromb Haemost, 2010; 8:641–650.

31.

Iwatsuki

, Sato

, Moritani

et al. Biochemical and pharmacological profile of darexaban, an oral direct factor Xa inhibitor. Eur J Pharmacol, 2011; 673:49–55.