Antiplatelet and antithrombotic effects of cordycepin-enriched WIB-801CE from Cordyceps militaris ex vivo, in vivo, and in vitro

Background A species of the fungal genus Cordyceps has been used as a complementary and alternative medicine of traditional Chinese medicine, and its major component cordycepin and cordycepin-enriched WIB-801CE are known to have antiplatelet effects in vitro. However, it is unknown whether they have also endogenous antiplatelet and antithrombotic effects. In this study, to resolve these doubts, we prepared cordycepin-enriched WIB-801CE, an ethanol extract from Cordyceps militaris-hypha, then evaluated its ex vivo, in vivo, and in vitro antiplatelet and antithrombotic effects. Methods Ex vivo effects of WIB-801CE on collagen- and ADP-induced platelet aggregation, serotonin release, thromboxane A2 (TXA2) production and its associated activities of enzymes [cyclooxygenase-1 (COX-1), TXA2 synthase (TXAS)], arachidonic acid (AA) release and its associated phosphorylation of phospholipase Cβ3, phospholipase Cγ2 or cytosolic phospholipase A2, mitogen-activated protein kinase (MAPK) [p38 MAPK, extracellular signal-regulated kinase (ERK)], and blood coagulation time in rats were investigated. In vivo effects of WIB-801CE on collagen plus epinephrine-induced acute pulmonary thromboembolism, and tail bleeding time in mice were also inquired. In vitro effects of WIB-801CE on cytotoxicity, and fibrin clot retraction in human platelets, and nitric oxide (NO) production in RAW264.7 cells or free radical scavenging activity were studied. Results Cordycepin-enriched WIB-801CE inhibited ex vivo platelet aggregation, TXA2 production, AA release, TXAS activity, serotonin release, and p38 MAPK and ERK2 phosphorylation in collagen- and ADP-activated rat platelets without affecting blood coagulation. Furthermore, WIB-801CE manifested in vivo inhibitory effect on collagen plus epinephrine-induced pulmonary thromboembolism mice model. WIB-801CE inhibited in vitro NO production and fibrin clot retraction, but elevated free radical scavenging activity without affecting cytotoxicity against human platelets. Conclusion WIB-801CE inhibited collagen- and ADP-induced platelet activation and its associated thrombus formation ex vivo and in vivo. These were resulted from down-regulation of TXA2 production and its related AA release and TXAS activity, and p38 MAPK and ERK2 activation. These results suggest that WIB-801CE has therapeutic potential to treat platelet activation-mediated thrombotic diseases in vivo.


Analysis of cordycepin in WIB-801CE with HPLC
WIB-801CE was dissolved with 75% methanol, then analyzed by high performance liquid chromatography (HPLC). An Alliance 2695 liquid chromatography system (Waters Co., Milford, MA, USA), equipped with vacuum degasser, quaternary gradient pump, autosampler and photodiode array detector, was connected to Empower software. A hydrosphere C 18 column (250 mm × 4.6 mm id, 5 μm, YMC Co., Ltd., Kyoto, Japan) was used at a column temperature of 30°C. The applied-mobile phase gradient program was 0.01 M KH 2 PO 4 /methanol (95:5, v/v) at 0 min and held for 5 min; 0.01 M KH 2 PO 4 /methanol v/v) at 20 min and held for 6 min; 0.01 M KH 2 PO 4 /methanol (95:5, v/v) at 27 min and held 6 min for chromatographic balance. In this step, 99.8% of methanol was used. The flow rate was at 1.0 mL/min and sample injection volume was 10 μL. The ultra violet detection was operated at 254 nm.

Animals and administration
We investigated the ex vivo and in vivo effects of WIB-801CE using rats (Sprague-Dawley, male, 200 g) and Institute of Cancer Research (ICR) mice (male, 18 g, Daehan Biolink Co., Ltd., Chungbook, Korea). Rats for ex vivo experiment and mice for in vivo observation were divided into as follows, respectively: WIB-801CE-nontreated group (control), WIB-801CE-treated group, aspirin-treated group as positive control of in vivo, and warfarin-treated group as positive control of ex vivo.
Animals were acclimatized for a week at a temperature of 24 ± 1°C and humidity of 55 ± 5%. Before oral administration of substances, all animals were fasted for 12 h, then were fed with standard pellets diet (Purina Inc., Korea) had free access to water. WIB-801CE [200, 400 mg/kgbody weight (BW)] and warfarin (1 mg/kg-BW) for ex vivo experiment were orally administered to the rats one per day for seven days, and WIB-801CE (200, 400 mg/kg-BW) and aspirin (100 mg/kg-BW) for in vivo observation were orally administered to the mice once a day for five days. 200 mg/kg-BW of WIB-801CE is corresponded to the minimum dose that inhibits rat platelet aggregation (data not shown). WIB-801CE, warfarin, and aspirin were Fig. 1 Composition of cordycepin in WIB-801CE and effects of cordycepin and WIB-801CE on cytotoxicity and platelet activation. a Chemical structure of cordycepin (3'-deoxyadenosine). b The chromatogram of WIB-801CE. c The chromatogram of pure cordycepin. d In vitro effects of WIB-801CE and cordycepin on cytotoxicity. e In vitro effects of WIB-801CE and cordycepin on platelet aggregation without agonists. Measurements of cordycepin, cytotoxicity, and platelet aggregation were carried out as described in "Methods" section. As a positive control to LDH cytotoxicity and platelet activation, 0.2% triton X-100 and collagen (10 μg/mL) were used, respectively. The data are expressed as the mean ± standard deviation (n = 4). NS, not significant versus without WIB-801CE and cordycepin, each control dissolved with distilled water. The experiments were proved by the Ethics Committee for Animal Experiments of Whanin Pharmaceutical Corporation (Suwon, Korea/ 15-NE-016 for rats, 15-NE-008 for mice).
Preparation of rat platelet-rich plasma and platelet-poor plasma for ex vivo assay After the final respective administration, all rats were fasted for 24 h, then after 2 h of WIB-801CE-and warfarin-administration were anesthetized with 20% urethane before sacrifice according to the method of Zhang et al. [17]. The blood was collected from the abdominal aorta. The blood was anti-coagulated with acidcitrate-dextrose solution (0.8% citric acid, 2.2% sodium citrate, 2.45% glucose), and was centrifuged at 250 × g for 10 min in order to obtain platelet-rich plasma (PRP). In order to remove residual red blood cells and white cells, the PRP was again centrifuged at 125 × g for 10 min. Platelet-poor plasma (PPP) was prepared by centrifuging the part of PRP at 1,300 × g for 10 min.
PRP was used to investigate ex vivo platelet aggregation, TXA 2 production, serotonin release, COX-1 and TXAS activities, AA release and protein phosphorylation. PPP was used to investigate ex vivo PT and APTT. The number of platelets in PRP was adjusted with PPP to a final concentration of 5 × 10 8 /mL. All of the above procedures were carried out at 25°C to avoid platelet aggregation from any effect of low temperature.

Preparation of human PRP and washed platelets for in vitro assay
To investigate in vitro effects of WIB-801CE and cordycepin on fibrin clot retraction, we used human PRP and washed platelets. PRP from normal healthy human volunteers with informed consent was obtained from the Korean Red Cross Blood Center (KRBC, Changwon, Korea), and its experimental use was approved by the KRBC (Safety Supervisor Team-621-2015.02.26) and the Korea National Institute for Bioethics Policy Public Institutional Review Board (Seoul, Korea/PIRB12-072-01) with informed consent. PRP anticoagulated with acidcitrate-dextrose solution (0.8% citric acid, 2.2% sodium citrate, 2.45% glucose) was centrifuged for 10 min at 125 × g to remove a little red blood cells and white cells, which was used to investigate the effect of WIB-801CE and cordycepin on thrombin-induced fibrin clot retraction. The number of platelets in PRP was adjusted with PPP to a final concentration of 5 × 10 8 /mL.
To observe in vitro effects of WIB-801CE and cordycepin on cytotoxicity and resting platelet aggregation, we prepared human washed platelets. The PRP was centrifuged for 10 min at 1,300 × g to obtain platelet pellets. The platelets were washed twice with washing buffer (138 mM NaCl, 2.7 mM KCl, 12 mM NaHCO 3 , 0.36 mM NaH 2 PO 4 , 5.5 mM glucose, and 1 mM Na 2 EDTA, pH 6.5). The washed platelets were then resuspended in suspension buffer (138 mM NaCl, 2.7 mM KCl, 12 mM NaHCO 3 , 0.36 mM NaH 2 PO 4 , 0.49 mM MgCl 2 , 5.5 mM glucose, 0.25% gelatin, pH 6.9) to a final concentration of 5 × 10 8 /mL. All of the above procedures were carried out at 25°C to avoid platelet aggregation from any effect of low temperature. The Korea National Institute for Bioethics Policy Public Institutional Review Board (Seoul, Korea/PIRB12-072-01) approved these experiments.

In vitro cytotoxicity assay
Platelet cytotoxicity was determined by leakage of LDH from cytosol. Human washed platelets (10 8 /mL) were incubated for 5 min at 37°C in the presence of WIB-801CE or cordycepin, then centrifuged with 12,000 × g at room temperature for 2 min. The supernatant was measured with a synergy HT multi-model microplate reader (BioTek Instruments, Winooski, VT, USA) using LDH assay kit. LDH leakage was expressed as percentage of total LDH activity in platelets completely lysed by 0.2% triton X-100.

Measurement of ex vivo rat platelet aggregation, and in vitro human resting platelet aggregation
To evaluate antiplatelet effect of WIB-801CE under condition that generates maximally platelet aggregation, we used high dose of collagen and ADP as agonists. The concentration of collagen-induced maximal rat (Sprague-Dawley, male) platelet aggregation was 10 μg/mL [18], and 5 μM of ADP was used to aggregate rat platelets [19]. Accordingly, we used 10 μg/mL of collagen, and 5 μM of ADP to cause ex vivo rat platelet aggregation.
To observe in vitro effects of WIB-801CE and cordycepin on resting human platelet aggregation, the aggregation of human washed platelets (10 8 /mL) was performed as described above in the presence of WIB-801CE or cordycepin without agonists. But, collagen (10 μg/mL) was used as positive control. Each aggregation rate was determined as an increase in light transmission. The PPP for PRP aggregation, and platelet suspension buffer (pH 6.9) for washed platelet aggregation were used as the reference (transmission 0) to regulate the base line of aggregometer. WIB-801CE and cordycepin were dissolved in distilled water.

Ex vivo measurement of TXB 2
To investigate the effect on TXA 2 production, the aggregation was terminated by adding ice-cold 5 mM EDTA and 0.2 mM indomethacin to inhibit subsequent conversion of AA to TXA 2 . The amounts of TXB 2 , a stable metabolite of TXA 2 , were determined using a TXB 2 EIA kit according to the procedure described by the manufacturer.

Ex vivo determination of AA release
To investigate the effect on AA release, the aggregation was terminated adding ice-cold 5 mM EDTA and 0.2 mM indomethacin to inhibit subsequent conversion of AA to TXA 2 , and centrifuged with 200 × g at 4°C for 10 min. The supernatants were used for the assay of AA release. AA release was measured with a Synergy HT multi-model microplate reader (BioTek Instruments, Winooski, VT, USA) using AA release ELISA kit.

Preparation of platelet lysates
We prepared platelet lysates to determine ex vivo COX-1 and TXAS activities. Collagen-and ADP-activated rat PRP was centrifuged for 10 min at 1,300 × g to remove PPP and get platelet pellets. The platelets were then suspended twice with suspension buffer (138 mM NaCl, 2.7 mM KCl, 12 mM NaHCO 3 , 0.36 mM NaH 2 PO 4 , 0.49 mM MgCl 2 , 5.5 mM glucose, 0.25% gelatin, pH 7.4). The suspended platelets in the presence of 1% protease inhibitor cocktail were sonicated ten times in sensitivity 100% for 20 s at 4°C with a sonicator (HD 2070, Bandelin Electronic, Bandelin, Germany) to obtain platelet lysates. Next, the platelet lysates were centrifuged at 12,000 × g for 15 min at 4°C to remove cell debris. The supernatant was used to measure COX-1 and TXAS activity.

Ex vivo measurement of COX-1 activity
Platelet lysates containing 10 μg of protein were used. COX-1 activity was measured with a Synergy HT multimodel microplate reader (BioTek Instruments, Winooski, VT, USA) using COX-1 fluorescent activity assay kit according to the procedure described by manufacturer.

Ex vivo measurement of TXAS activity
Platelet lysates containing 20 μg of protein were used. The reaction for assay of TXAS activity was initiated by the addition of TXAS substrate PGH 2 and allowed to proceed for 1 min at 37°C. The reaction was terminated by the addition of 1 M citric acid, then was neutralized with 1 N NaOH. The concentration of TXA 2 was determined as TXB 2 , a stable metabolite of TXA 2 , which was measured with a Synergy HT multi-model microplate reader (BioTek Instruments, Winooski, VT, USA) using TXB 2 EIA kit.

Ex vivo determination of serotonin release
To investigate the effect on serotonin release, the aggregation was centrifuged at 4°C for 10 min at 200 × g. The supernatants were used for the assay of serotonin release. Serotonin release was measured with a Synergy HT multi-model microplate reader (BioTek Instruments, Winooski, VT, USA) using serotonin ELISA kit.

Ex vivo measurement of PT and APTT
To investigate whether WIB-801CE shows anticoagulant characteristics, if any, has bleeding risk as side effect of anticoagulant [21], we measured PT and APTT, markers of blood coagulation. The PPP (100 μL) was preincubated in a two-channel coagulator (Behnk Elektronik GmbH & Co., KG, Norderstedt, Germany) cup (catalog number 95-662, BioMérieux, Marcyl'Etoile, France) with gentle stirring for 1 min at 37°C. PT was determined as the time interval between the addition of PT reagent (100 μL) to the PPP and the formation of a fibrin clot. After preincubation of PPP for APTT measurement, 100 μL of APTT reagent was added to the PPP (100 μL) and incubated for 3 min at 37°C. Following incubation, 100 μL of 25 mM CaCl 2 was immediately added to the PPP containing APTT reagent. APTT was determined as the time required to form a fibrin clot.

In vivo tail bleeding time assay
We investigated whether WIB-801CE has bleeding risk, the side effect of antiplatelet substance [21]. WIB-801CE (200, 400 mg/kg-BW) and aspirin (100 mg/kg-BW), a positive control, were orally administered to mice once a day for five days. In this study, we used mice for measuring tail bleeding time according to the method of Kim and Lee [22]. After 5 min of the respective final administration, mice were anesthetized with zoletil (40 mg/kg, i.p.). The distal 0.5 cm segment of the tail was transected with operating knife, and immediately immersed in a tube containing 37°C of saline. Tail bleeding time was determined as the time required to cause blood coagulation.

In vivo evaluation of anti-acute pulmonary thromboembolism
To confirm the endogenous antithrombotic effect, we used a mice model to generate acute pulmonary thromboembolism [23]. Mice were orally administered with WIB-801CE (200, 400 mg/kg-BW), and aspirin (100 mg/kg-BW). After 1 h of respective administration, the mixture (100 μL) of collagen (300 μg/kg-BW) plus epinephrine (30 μg/kg-BW) were injected via tail vein, and the rate of protection and mortality was observed for 15 min, which were calculated as follow:

In vitro assay of platelet-mediated fibrin clot retraction
We investigated whether WIB-801CE or cordycepin inhibits fibrin clot retraction, an index of thrombi formation [24]. Human PRP 250 μL (10 8 platelets/mL) were transferred into polyethylene tube to avoid clot adherence, then were preincubated with or without WIB-801CE or cordycepin for 10 min at 37°C, and subsequently stimulated with thrombin (0.5 U/mL) for 60 min at 37°C. Pictures of fibrin clot were taken at 0 and 60 min using a digital camera, and its quantification was carried out by measurement of clot area using the NIH Image J Software (V1.46, National Institutes of Health, USA). Percentage of clot retraction was calculated as follows: Retraction (%) by thrombin = [1 -(final clot area/initial clot area)] × 100.

In vitro nitric oxide assay
To observe antiinflammatory effect of WIB-801CE, we used mouse leukemic macrophage RAW264.7 cells. RAW264.7 cells were obtained from the American Type Culture Collection (ATCC, Manassas, VA, USA), and were maintained at 37°C in 5% CO 2 and 95% air in Dulbecco's Modified Eagle's Medium (GE Healthcare, Marlborough, USA) containing 10% fetal bovine serum, and 1% penicillin-streptomycin solution. RAW264.7 cells (5 × 10 4 cells) were preincubated for 30 min with or without WIB-801CE, or inducible nitric oxide synthase (iNOS) inhibitor amino guanidine (AG), and stimulated for 24 h by lipopolysaccharide (10 ng/mL). The supernatant was used for NO assay using Griess reagent. Equal volume of culture supernatant (80 μL) and Griess reagent (80 μL) were mixed. The absorbance of the mixture was measured at 540 nm using spectrophotometer (Spectramax 190, Molecular devices, LLC., Sunnyvale, CA, USA). Nitrite was used as standard of NO.

Ex vivo nitric oxide assay
To investigate the NO production, we used collagenand ADP-stimulated PRP obtained from rats administered with the WIB-801CE (200, 400 mg/kg-BW). The PRP was centrifuged at 4°C for 10 min at 10,000 × g to get plasma. The plasma was incubated 1 h with 400 μL methanol:diethylether (3:1 mixture v/v), and subsequently plasma proteins were precipitated by centrifuging at 4°C for 10 min at 10,000 × g. The supernatant was used for NO assay using Griess reagent. Equal volume of supernatant (80 μL) and Griess reagent (80 μL) were mixed. After 30 min, absorbance of the mixture was measured at 540 nm using a Synergy HT multi-model microplate reader (BioTek Instruments, Winooski, VT, USA). Nitrite was used as standard of NO.

In vitro determination of antioxidant activity
To obtain antioxidant effect of WIB-801CE, we measured scavenging activity of free radical in DPPH according to the method [25,26]. DPPH was dissolved in 99% ethanol to make 200 μM of solution. WIB-801CE and antioxidant AC were dissolved in distilled water. Equal volume of test substances and DPPH were mixed at room temperature. After 30 min, the reduction in DPPH absorbance at 517 nm was measured using spectrophotometer (Optizen 2120UV, Mecasys, Korea). The scavenging activity of DPPH radicals by substances was determined using the following equation [26]: Scavenging activity (%) = [1 -(A sample /A DPPH )] × 100. The absorbance at 517 nm by 99% ethanol, DPPH vehicle, and distilled water, vehicles of WIB-801CE and AC was 0.001 and 0.000.

Ex vivo measurement of cordycepin effect on rat platelet aggregation
This experiment was performed to investigate the effect of cordycepin on ex vivo platelet aggregation. When cordycepin (15 mg/kg per day) was administered orally to the mice for 14 days, antitumor activity was known to observe [27]. Therefore, we selected 5 and 10 mg/kg-BW per day of cordycepin in this experiment as moderate doses for administration. These doses are corresponded to about 36 and 72% of cordycepin in WIB-801CE (200 mg/kg-BW) that inhibited ex vivo rat platelet aggregation. Rats (Sprague-Dawley, male, 200 g) were acclimatized for a week at a temperature of 24 ± 1°C and humidity of 55 ± 5%. Before oral administration of cordycepin, rats were fasted for 12 h, then were fed with standard pellets diet (Purina Inc., Korea) had free access to water. Cordycepin (5 and 10 mg/kg-BW) was orally administered to the rats one per day for seven days. Cordycepin were dissolved with distilled water. The experiments were proved by the Ethics Committee for Animal Experiments of Whanin Pharmaceutical Corporation (Suwon, Korea/15-NE-016 for rats). After the final respective administration, all rats were fasted for 24 h, then were anesthetized with 20% urethane before sacrifice. PRP preparation, platelet aggregation, measurement were performed as described before.

Protein assay
To determine COX-1, TXAS activity, and protein phosphorylation, protein concentration was measured using bicinchoninic acid assay kit (Pierce Biotechnology, USA).

Statistical analyses
The experimental results are indicated as the mean ± standard deviation accompanied by the number of observations. Data were determined by analysis of variance (ANOVA). If this analysis showed significant differences among the group means, then each group was compared by the Newman-Keuls method. Statistical analysis was carried out according to the SPSS 21.0.0.0 (SPSS, Chicago, IL, USA). p < 0.05 was considered to be statistically significant.

Composition of cordycepin in WIB-801CE
Because it is known that Cordyceps militaris, a source of WIB-801CE, has cordycepin (Fig. 1a) [28], we analyzed cordycepin of WIB-801CE with HPLC. As shown in Fig. 1b, peak 1 from WIB-801CE was observed at 19.988 min of the retention time, which was almost in accord with the retention time (19.980 min) of pure cordycepin (Fig. 1c). This means that peak 1 is derived from cordycepin in WIB-801CE. The concentration of peak 1 in WIB-801CE corresponding to cordycepin was 69.30 ± 0.20 mg/g-WIB-801CE (about 6.93 ± 0.02%, Table 1). Whole fruiting body myelia of Cordyceps militaris is known to contain 0.16% of cordycepin, but whole fruiting body, stroma, and larva of Cordyceps sinensis do not contain cordycepin [29]. Therefore, the cordycepin content in WIB-801CE that we used in this study is very higher than those in whole fruiting body myelia of Cordyceps militaris, and in whole fruiting body, stroma, and larva of Cordyceps sinensis.

In vitro effects of WIB-801CE and cordycepin on activation of resting human platelets
Platelet activation is an index of platelet shape change, platelet aggregation and granule secretion, and is the cause of cardiovascular and cerebrovascular disease, and atherosclerosis [33][34][35]. Accordingly, if WIB-801CE activates resting platelets, unstimulated platelets, a question to evaluate the antiplatelet effects of WIB-801CE might be raised. Therefore, the effect of WIB-801CE and cordycepin on platelet activation was determined by measuring platelet aggregation in resting human platelets. As the results, a positive control collagen (10 μg/mL) activated platelets by increasing platelet aggregation up to 83.3 ± 3.1% (Fig. 1e). However, WIB-801CE (200, 400 μg/mL) alone, and cordycepin (56, 112 μM) alone did not increased platelet aggregation (Fig. 1e), as compared with that (1.0 ± 1.0%) by resting platelets. It was evidenced that WIB-801CE and cordycepin alone do not affect the activation of resting human platelets.

Ex vivo effects of WIB-801CE on platelet aggregation and TXA 2 production
It is known that the inhibition of collagen-and ADPinduced platelet aggregation is potential target to develop antithrombotic agent having antiplatelet characteristics [36,37]. Therefore, we used on collagen and ADP as agonists. When PRP (10 8 /mL) from control was activated with collagen and ADP, the aggregation rate was increased up to 82.9 ± 6.6% by collagen (Fig. 2a) and 78.7 ± 4.9% by ADP (Fig. 2b). However, collagen-and ADPinduced rat platelet aggregation was significantly attenuated by WIB-801CE (200, 400 mg/kg-BW) ( Fig. 2a  and b). The inhibitory degrees by 200 mg/kg-BW were 13.0% to that by collagen (Fig. 2a) and 10.9% against that by ADP (Fig. 2b).

Ex vivo and in vivo effects of WIB-801CE on blood coagulation and tail bleeding time
Bleeding is connected to the attenuation of platelet aggregation and blood coagulation, and the inhibition of thrombosis [8,21,47,48]. Accordingly, we investigated the effects   (Table 3). Aspirin (100 mg/kg-BW) also prolonged tail bleeding time to 1,800.0 ± 0.0 s ( Table 3). Aspirin (100 mg/kg-BW) potently prolonged bleeding time to 1,336.5%, on the other hand, WIB-801CE (200, 400 mg/kg-BW) prolonged it up to 111.4% and 187.5% as compared with that (125.3 ± 17.0 s) by control, respectively (Table 3).

In vivo effects of WIB-801CE on acute pulmonary thromboembolism
Because antiplatelet drugs play an important role in protection of thrombus formation, we investigated whether WIB-801CE, inhibiting ex vivo platelet aggregation, has also a protective effect on endogenous thrombus formation. In this study, in vivo venous antithrombotic effect of WIB-801CE was estimated using collagen plus epinephrineinduced acute pulmonary thromboembolism mouse model [23,[49][50][51]. As shown in Table 4, when the mixture of collagen plus epinephrine was treated to mice, the protection rate was 4.2% against acute pulmonary thromboembolism, and the mortality rate was 95.8%. However, in WIB-801CEtreated mice, the protection degree from a pulmonary thromboembolism was increased to 25.0% by WIB-801CE (200 mg/kg-BW), and 35.0% by WIB-801CE (400 mg/kg-BW) in a dose dependent manner (Table 4). In aspirin (100 mg/kg-BW)-treated mice, the protection degree from a pulmonary thromboembolism was increased up to 35.0%, and the mortality was decreased to 65.0%, which were equal to those by WIB-801CE (400 mg/kg-BW)-treatment (Table 4). These mean that WIB-801CE is actually valid to protect venous thromboembolism like aspirin.

In vitro and ex vivo effects of WIB-801CE on NO production
It is well known that monocytes/macrophages and neutrophils produce various inflammatory mediators (i.e. NO, a b c d  The results were expressed as the mean ± standard deviation (n = 8 or 6). ∞, no coagulation; NS not significant versus normal; N.D normal diet, N number of tested rats; BW body weight, PT prothrombin time, APTT activated partial thromboplastin time s, second prostaglandin E 2 ), and subsequently activate platelets to generate atherothrombosis [52,53]. In recent, it is also reported that neutrophil-produced NO activates platelet in chronic renal failure [54]. Accordingly, we investigated whether WIB-801CE inhibits in vitro NO production in RAW264.7 macrophage cells. As shown in Fig. 7a, lipopolysaccharide (LPS), an activator of macrophages, potently produced NO as compared with that of normal. However, WIB-801CE dose (15 to 50 μg/mL)-dependently attenuated LPS-elevated NO production (Fig. 7a). iNOS inhibitor AG potently inhibited NO production (Fig. 7a).
In vitro effects of WIB-801CE on free radical scavenging activity Reactive oxygen species (ROS) activate platelets and thrombogenesis [55][56][57]. In this study, we investigated whether WIB-801CE has an antioxidant effect scavenging ROS, which was evaluated as scavenging activity of free radical in DPPH, a stable free-radical molecules [25]. AC, a positive control, potently reduced the absorbance of DPPH. This means that AC has antioxidant effect by scavenging free radical in DPPH [25]. WIB-801CE also attenuated the absorbance of DPPH in a dose dependent manner (Fig. 7c). This result is indicated that WIB-801CE has antioxidant effect by scavenging free radical in DPPH like AC.

Discussion
WIB-801CE and its component cordycepin did not affect the cytotoxicity (determined as LDH leakage) and platelet activation (determined as platelet aggregation) to resting human platelets in vitro. These mean that there is no problem to evaluate the antiplatelet effects of WIB-801CE ex vivo. It is well established that various agonists (i.e. collagen, ADP, thrombin)-produced TXA 2 generates circulatory disorder such as thrombosis, atherosclerosis, and myocardial infarction by stimulating platelet aggregation, vasoconstriction, and bronchoconstriction [5,58,59]. Therefore, it is essential to inhibit platelet aggregation and TXA 2 production to prevent circulatory disorder in blood vessel. WIB-801CE attenuated ex vivo collagen-and ADP-induced platelet aggregation and TXA 2 production.
These results were connected to the ex vivo inhibition of AA release and TXAS activity by WIB-801CE in collagen-and ADP-activated platelets. WIB-801CE did not also inhibit ex vivo collagen-and ADP-induced PLC β3 (Ser 537 , Ser 1105 ), PLC γ2 (Tyr 1217 ) phosphorylation. These mean that WIB-801CE would produce DG from phosphatidylinositol 4,5-bisphosphate in collagen-and ADPactivated platelets. DG is known to hydrolyze by p 38 MAPK -activated DG-lipase to release AA [45]. If so, it is considered that WIB-801CE may attenuate AA release by inhibiting p 38 MAPK /DG-lipase pathway [45] without affecting inhibition of cPLA 2 and PLC β3 [9,10,42,43]. Because agonist-produced TXA 2 enforces thrombus formation as a positive promoter [5,58,59], a compound or substance that inhibits the activity of COX-1 or TXAS, the production of TXA 2 or the action of  The results were expressed as the mean ± standard deviation (n = 8 or 5). * p < 0.05 compared with control, ** p < 0.001 compared with control. N number of tested mice, BW body weight, s, second Δ (%) = [(WIB-801CE or aspirin) control]/ control × 100 TXA 2 is evaluated as antithrombotic agents. Many studies have been performed to discover therapeutic agents that can counteract the effects of TXA 2 . Various phytochemicals (i.e. epigallocatechin-3-gallate, caffeic acid, chlorogenic acid, caffedymine, sanguinarine) are known to inhibit COX-1 rather than TXAS to suppress the TXA 2 production in vitro or ex vivo [60][61][62][63][64][65]. However, WIB-801CE inhibited the activity of TXAS rather than COX-1 ex vivo, which reflects that WIB-801CE inhibits the TXA 2 production pathway from PGH 2 rather than prostaglandin G 2 production pathway from AA. In this study, we showed that WIB-801CE may involve in down-regulation of both p 38 MAPK phosphorylation to vanish the AA supply from DG and TXAS activity to block the TXA 2 production from PGH 2 ex vivo. Therefore, it is apparent that WIB-801CE can be beneficially used to prevent the TXA 2 -mediated thrombus formation in vivo. It is well known that agonist-released serotonin stimulates irreversibly platelet aggregation, and subsequently causes the thrombosis as well as TXA 2 [66][67][68]. WIB-801CE inhibited collagen-and ADP-induced serotonin release ex vivo, which reflects that WIB-801CE can inhibit the irreversible platelet aggregation in vivo. WIB-801CE potently inhibited ex vivo the phosphorylation of p 38 MAPK and ERK2 (42 kDa), but not Ca 2+ -dependent a b Fig. 6 In vitro effects of WIB-801CE and cordycepin on fibrin clot retraction. a In vitro effects of WIB-801CE and cordycepin on thrombin-induced fibrin clot. b In vitro effects of WIB-801CE and cordycepin on thrombin-retracted clot retraction (%). Quantification of fibrin clot retraction was performed as described in "Methods" section. The data are expressed as the mean ± standard deviation (n = 4). * p < 0.05 versus the thrombin-stimulated platelets. 1 myosin light chain (MLC) phosphorylation (Data not shown) that involves in serotonin release [14,[69][70][71][72] in collagen-and ADP-activated platelets. These results are allowed to consider that WIB-801CE seems to attenuate ex vivo serotonin release by inhibiting the phosphorylation of p 38 MAPK and ERK2 rather than MLC in collagen-and ADP-activated platelets. This is similar to the reports that some phytochemicals (i.e. caffeic acid phenethyl ester, ginsenoside Rp1) inhibit collagen-and ADP-induced ATP release by phosphorylating p 38 MAPK and ERK2 [39,73].
In recent, we found that CE-WIB801C, n-butanol extracts from Cordyceps militalis, and cordycepin purified from CE-WIB801C has in vitro antiplatelet effect by inhibiting fibrinogen binding to glycoprotein IIb/IIIa via stimulation of cAMP-dependent phosphorylation of vasodilatorstimulated phosphoprotein (Ser 157 ), and inhibition of phosphatidylinositol-3 kinase/Akt phosphorylation [74]. Antiplatelet effect of CE-WIB801C was involved in inhibition of collagen-induced serotonin release. Accordingly, we confirmed that the extracts from Cordyceps militaris have antiplatelet effect in vitro and in vivo.
The formation of fibrin clot by intrinsic and extrinsic blood coagulation factors together with platelet aggregation at injured blood vessels is another cause of thrombogenesis. WIB-801CE did not significantly prolong PT and APTT as compared with that by normal ex vivo. This reflects that WIB-801CE has no anticoagulant characteristics. WIB-801CE, however, weakly prolonged average PT as compared with that by normal. The prolongation of PT is associated with the reduction of coagulation factor VII production by inhibition of NADPH-vitamin K reductase [75]. The dose (1 mg/kg-BW) of warfarin, an inhibitor of NADPH-vitamin K reductase, that unlimitedly prologned PT is corresponded to high dose (60 mg/day) in case of giving to human (60 kg), which is more 12 fold than international normalized dose (5 mg/day) of warfarin [76]. In this study, because it is unknown whether the weak extension of average PT by WIB-801CE is clinically safety or risk, it is necessary to investigate PT using international normalized dose (5 mg/day) of warfarin, then the safety of WIB-801CE in the weak extension of PT should be evaluated in the future.
In addition, WIB-801CE without significantly affecting the prolongation of blood coagulation time may not influence on inhibition of fibrin production. If so, because the fibrin is retracted by the interaction with platelet aggregation [77], anybody may apprehend that the thrombus could be generated by WIB-801CE in vivo. But it is considered that its fear can be excluded because WIB-801CE inhibited both thrombogenic TXA 2 production and serotonin release ex vivo, and thrombin-induced fibrin clot retraction in vitro. These results mean that WIB-801CE can strongly inhibit the fibrin clot retraction by down-regulating platelet activation without significantly affecting the blood coagulation. This is also evidenced as the effect that WIB-801CE inhibited collagen plus epinephrine-induced acute pulmonary thromboembolism in vivo, which is a marker of platelet aggregation-generated thrombogenesis. At the present study, however, it is unknown whether cordycepin in WIB-801CE contributed to the inhibition of acute pulmonary thromboembolism. These should be studied in the future.
As well as anticoagulants, antiplatelet drugs (i.e. aspirin, clopidogrel) also cause bleeding, and surprisingly may generate blood loss [21]. It is known that 20-40 mg/day of aspirin is clinically used in human to protect thrombotic disease [78]. The dose (100 mg/kg-BW) of aspirin, a positive control, seriously prolonged tail bleeding time of mice as compared with that by control. This aspirin dose (100 mg/kg-BW) is corresponded to high dose (6,000 mg/ day) in case of giving to human (60 kg), which is more 300-150 fold than clinical dose (20-40 mg/day) of aspirin and impossible to compare with WIB-801CE in tail bleeding time of mice. At the present study, because it is unknown whether the significant extension of tail bleeding time by WIB-801CE is clinically safety or risk, it is necessary to investigate tail bleeding time using the clinical dose (20-40 mg/day) of aspirin as positive control, then the safety of WIB-801CE-prolonged tail bleeding time should be evaluated in the future.
Leukocyte-produced ROS oxidizes low density lipoprotein (LDL) in blood, then oxidized LDL (ox-LDL) is incorporated into macrophage to generate foam cell (See figure on previous page.) Fig. 7 Effects of WIB-801CE on NO production, and antioxidation activity. a In vitro effects of WIB-801CE on LPS-induced NO production. b Ex vivo effects of WIB-801CE on collagen-and ADP-induced NO production. c In vitro effects of WIB-801CE on antioxidation activity. Measurements were carried out as described in "Methods" section. The data are expressed as the mean ± standard deviation (n = 4). ** p < 0.001 versus LPS-stimulated RAW264.7 cells. † † p < 0.001 versus the DPPH. NS, not significant versus the each agonist-stimulated platelets. 1) Δ (%) = [(DPPH + WIB-801CE 200 μg/mL) -DPPH]/DPPH × 100 Table 5 Effects of cordycepin administration in ADP-induced rat platelet aggregation The data are given as the mean ± standard deviation (n = 3). ** p < 0.001 compared with control. Inhibition (%) = [(cordycepin + ADP)control]/ control × 100 (%) which damages vascular wall by inducing inflammation. Because platelet aggregation is caused at injured place of vascular wall, the inflammation by leukocyte-produced ROS and NO is the cause of thrombus formation. This means that the counteraction of agonists-induced platelet aggregation and leukocyte-induced inflammation might be contributed to the inhibition of thrombosis. WIB-801CE had ex vivo inhibitory effects on the release of serotonin that stimulates the uptake of ox-LDL into macrophage [79]. In addition, WIB-801CE inhibited NO production and elevated scavenging activity of free radical in DPPH in vitro. Therefore, it is anticipated that WIB-801CE would not activate inflammatory leukocytes in vivo as evidenced that WIB-801CE dose not affect ex vivo NO production. These results suggest that WIB-801CE might have antithrombotic effects by inhibiting inflammation via antioxidative action. Because we could not identify cordycepin or its metabolites in PRP from WIB-801CE (200, 400 mg/kg-BW) rats, in the present study, we could not explain whether cordycepin in WIB-801CE was absorbed through intestine and subsequently involved in inhibition of platelet aggregation. This study should be performed in the future. Considering antiplatelet effects observed by blood collection of 2 h after final administration of WIB-801CE and 14 days after administration of cordycepin, although cordycepin is metabolized to an inactive 3'-deoxyhypoxanthinosine by adenosine deaminase in rat blood [80][81][82], it could be thought that unknown substances in WIB-801CE or cordycepin-derived unknown substance might involve in inhibition of platelet aggregation in an acute or chronic manner.
WIB-801CE (200, 400 mg/kg-BW)-dose independently exerted its inhibitory effects on platelet aggregation, TXA 2 production, AA release, TXAS activity, serotonin release, p 38 MAPK phosphorylation and ERK2 phosphorylation in ADP-activated platelets. Considering the inhibition of ADP-induced platelet aggregation is a potential target to develop antithrombotic agent having antiplatelet characteristics [36. 37], it is thought that high dose of WIB-801CE (400 mg/kg-BW) might exert undesirable effect on platelets in vivo.

Conclusion
Cordycepin-enriched WIB-801CE from Cordyceps militaris vanished ex vivo thrombogenic molecules (TXA 2 , serotonin), and their associated signaling molecules (AA, TXAS, p 38 MAPK , ERK2) in platelet aggregation. Furthermore, WIB-801CE inhibited in vivo acute pulmonary thromboembolism, an index of thrombotic formation, without having cytotoxicity and risk of serious bleeding, but with antioxidant and antiinflammatory activity. Therefore, we suggest that WIB-801CE may be a beneficial and effective substance to treat or protect thrombosis, atherosclerosis, and myocardial infarction via inhibition of platelet activation in vivo.