* Correspogy, SchooHan-zhongE-mail: hm

PHYTOTHERAPY RESEARCHPhytother. Res. (2013)Published online in Wiley Online Library(wileyonlinelibrary.com) DOI: 10.1002/ptr.5052

Copyright

Anti-Thrombotic Effect of a Novel Formula fromCorni Fructus with Malic Acid, Succinic Acid andCitric Acid

Qi-Chun Zhang,1,2 Yue Zhao1,2 and Hui-Min Bian1,2*1Department of Clinic Pharmacology, School of Pharmacy, Nanjing University of Chinese Medicine, Nanjing, 210029, P.R. China2Jiangsu Key Laboratory for Pharmacology and Safety Evaluation of Chinese Materia Medica, Nanjing University of ChineseMedicine, Nanjing, 210029, P.R. China

Our previous investigation had confirmed the inhibition of platelet aggregation of a novel Corni fructus-derivedformula composed of malic acid, succinic acid and citric acid with a ratio of 3:2:2. The present study was to furtherevaluate the anti-thrombotic effect of the formula in vivo. Mice of acute pulmonary thromboembolism, and ratsof arterial thrombosis were used to determine the anti-thrombotic effect of the formula. Histology analysis ofendothelium was conducted with hematoxylin and eosin stain. TXB2, 6-K-PGF1α, cAMP, cGMP and NO in ratplasma were determined. In vitro assay of αIIbβ3 and phosphorylation of ERK1/2 were performed in ADP-treatedplatelet. The formula significantly reduced the recovery time and mortality rate of mice with acute pulmonary throm-boembolism. Remarkably extended occlusion time, decreased thrombus weight and more integrated endotheliumwere observed in rat with the formula. Enhanced 6-K-PGF1α, cGMP and NO, but not TXB2 and cAMP, weredemonstrated in rat plasmawith treatment of the formula. Finally, the formula was shown to inhibit αIIbβ3 expressionand activation of ERK1/2 in platelet. The formula shows positive anti-thrombotic effect. The direct interference onADP activated signaling in platelet and regulation of endothelium function are two primary pathways involved inthe action on thrombosis. Copyright © 2013 John Wiley & Sons, Ltd.

Keywords: Corni fructus; anti-thrombsis; ADP; platelet; endothelium.

Abbreviations: vWF, von Willebrand factor; ADP, adenosine diphosphate; TXA2, thromboxane A2; ERK1/2, extracellular-signal-related kinase; PGI2, prostacyclin; NO, nitric oxide; cGMP, cyclic guanosine monophosphate; TXB2, thromboxane B2; cAMP, cyclicadenosine monophosphate; 6-keto-prostaglandin F1α, 6-k-PGF1α

INTRODUCTION

Thrombus formation following platelet activation and ad-hesion is a necessary defense mechanism in the process ofhemostasis. Exposure of subendothelial collagen at sites ofvessel injury initially triggers the affinity of characteristicmolecular on the platelet surface such as integrin αIIbβ3and fibrinogen receptor leading to platelet adhesion.Excessive thrombus formation and atheromatous plaquerupture, however, would obstruct blood flow resulting inacute myocardial infarction or stroke. In atherosclerosis,impairment in endothelial function integrity, collagendeposition and shear stress fluctuation of blood flowactivate platelet and promote atherosclerotic plaque for-mation. It is promising that reversing the abnormal activityof platelet and elevated unfavorable inducible factor wouldprevent and therapy coronary artery disease, ischemiccerebrovascular disease and peripheral arterial disease.Following adhesion to collagen, activation of integrin

αIIbβ3 with higher affinity allows it to bind divalentfibrinogen or multivalent von Willebrand factor(vWF), which regulates platelet aggregation. Previousprospective investigations about αIIbβ3 define its role

ondence to: Hui-Min Bian, Department of clinical pharmacol-l of Pharmacy, Nanjing University of Chinese Medicine, 282Road, Nanjing, 210029, P. R. China.bian@sina.com

© 2013 John Wiley & Sons, Ltd.

in platelet aggregation and as a potential target for anti-thrombotic therapy. And now, intravenous administrationsof antibody and nonpeptide antagonists of αIIbβ3 havebeen utilized broadly to therapy and prevent thrombosisunder conditions of percutaneous coronary interventionand acute coronary syndromes (Bhatt and Topol, 2000).Activated platelets further release adenosine diphosphate(ADP) and thromboxaneA2 (TXA2) that in turn transformαIIbβ3 into a high-affinity state and recruit more and moreplatelets to the developing thrombus. Meanwhile, ADP,binding its receptor P2Y1 and P2Y12, induces phosphoryla-tion of extracellular-signal-related kinase (ERK1/2) pro-moting αIIbβ3 expression and TXA2 production (Rogeret al., 2004). On the other hand, endogenous inhibitors suchas prostacyclin (PGI2) and nitric oxide (NO) limit plateletactivation in vivo (Cheng et al., 2002).

Fruit ofCornus officinalis Sieb. et Zucc (Cornaceae) is acommonly used traditional Chinese medicine. The ingre-dients or extracts of Corni fructus have been shown topossess various pharmacological activities such as anti-in-flammatory, anti-oxidative, neuroprotective activities andinfluence on diabetes mellitus. Corni fructus inhibitsinterleukin release from allergic airway inflammationand eosinophil infiltration (Kam et al., 2012), and inhibi-tion of nuclear factor-kappa B activation and negativeregulation of ERK1/2 are the underlying mechanism in-volved in the anti-inflammatory activity (Choi et al., 2011;Sung et al., 2009). Cornuside, a bisiridoid glucosidecompound from Corni fructus, dilates vascular smooth

Received 02 May 2013Revised 14 June 2013Accepted 08 July 2013

2 Q.-C. ZHANG ET AL.

muscle via endothelium-dependent NO/cyclic guanosinemonophosphate (cGMP) signaling pathway (Kang et al.,2007). For diabetes and the complications, Corni fructuswas explored to inhibit α-glucosidase, protect β-cell andimprove kidney functions (He, 2011; Lin et al., 2011;Yokozawa et al., 2010). Furthermore, Corni fructus couldprotect ischemia/reperfusion-induced brain injury throughdecreasing glutamate-induced neuron toxicity, inhibitingapoptosis process, ameliorating mitochondrial energymetabolism and providing a beneficial environment topromote neuronal survival (Wang et al., 2010; Jeonget al., 2012; Jiang et al., 2009; Zhao et al., 2010). Our previ-ous investigation showed that Corni fructus had positiveinfluence on platelet function of aggregation and adhesionin vitro (Zhang et al., 2013). Components of organic acidsincluding malic acid, succinic acid and citric acid werefound to be responsible for that promising activity. Basedon further broaden investigations, an optimum combina-tion of those acids with a rational ratio of 3:2:2 exhibitedmore potent effect on platelet function than those singlecomponents. The present study was attempted to investi-gate the effect of the formula on thrombosis formationin vivo and potential mechanism.

MATERIAL AND METHODS

Animals and agents. Adult male Sprague–Dawley ratsweighing 270–300 g and ICR mice weighting 18–22 g werefrom Shanghai Laboratory Animal Center of ChineseAcademy of Sciences (Shanghai, PRC). Those animalswere maintained at 20–21 °C and 55±5% relative humid-ity under a 12-h light:12-h dark cycle. Standard rodentchow diets were provided ad libitum throughout the study.All procedures and experiments were conducted in accor-dance with the NIH guide for the care and use of labora-tory animals (NIH Publication No. 80-23; revised 1978).Malic acid, succinic acid, citric acid, dimethyl sulfoxide(DMSO), ADP, indomethacin and adrenaline were fromSigma–Aldrich (St. Louis, MO, USA). The formula (malicacid: succinic acid: citric acid, 3: 2: 2) was suspended inCMC-Na (0.5%) for in vivo experiment or dissolved inDMSO for in vitro assay, the final concentration of DMSOin platelet suspension was 0.1%. The activator ADP wasdissolved in saline. The primary antibodies of αIIbβ3,β-actin, ERK1/2 and p-ERK1/2 were from Cell SignalingTechnology Inc. (Danvers, MA). The detection kitof NO was from Jiancheng Bioengineering Co. Ltd,(Nanjing, PRC). These radioimmunoassay kits for throm-boxaneB2 (TXB2), 6-keto-prostaglandin F1α (6-k-PGF1α),cyclic adenosinemonophosphate (cAMP) and cGMPwerefrom Sino-UK (Beijing, PRC).

Acute pulmonary thromboembolism. A previously de-scribed method of Di Minno was modified and adaptedto induced acute pulmonary thromboembolism followingthe oral administration of the formula or vehicle (CMC-Na, 0.5%, w/v) for 6days in mice (DiMinno and Silver,1983). One hour after the last administration, mixture ofADP (0.4μmol · kg�1) and adrenaline (2.73μmol · kg�1)was injected into tail vein at rate of 20μL · s�1. Recoverytime and mortality rate of paralysed mice were recordedwithin 15min. It was counted as 15min as the recoverytime exceeded 15min.

Copyright © 2013 John Wiley & Sons, Ltd.

Induction of arterial thrombosis. Male Sprague–Dawleyrats administrated orally with the formula or vehicle(CMC-Na, 0.5%) for 6days were anesthetized with intra-peritoneal injection of chloral hydrate (300mg · kg�1) andthen fixed on operation table in supine position. Arterialthrombosis was induced by electrical stimulation on com-mon carotid artery according to the previously describedprotocol by Schumacher (Schumacher et al., 1993). A15mm segment of the left common carotid artery was ex-posed and dissected free of surrounding tissue. A piece ofparafilm was placed under the exposed artery to provideelectrical isolation. Both anode and cathode electrodeswere inserted under the vessel. The stimulation (2mA) lastfor 7min. A temperature-sensing probe was placed in con-tact with the carotid artery on the distal side of stimulationelectrodes. Both electrodes and probe were connected tothe thrombosis instrument. The time of thrombotic occlu-sionwas recorded through temperature change of distal ca-rotid artery automatically alarmed by the instrument afterelectrical stimulation. The time from the end of the electri-cal stimulation to the point of alarm was defined as the oc-clusion time. At the end of experiment, the commoncarotid artery was dissected to remove thrombus. Hema-toxylin and eosin stain was performed on the artery. Theformed thrombus were air-dried at 37 °C for 24h and thenweighed.

Determination of plasma TXB2, 6-keto-PGF1α, cAMP,cGMP and NO. Blood samples of rats with arterialthrombosis were taken from heart and drawn into tubescontaining indomethacin and EDTA-Na2. Followingcentrifugation at 2000 g for 10min, plasmawas collected.TXB2, 6-K-PGF1α, cAMP and cGMP were determinedby radioimmunoassay kits, andNOwas analysed by bio-chemistry detection kit according to the manufacturer’sinstructions. TXB2 and 6-K-PGF1α represented theirprecursor TXA2 and PGI2 respectively.

Western blotting assay. Blood sample of rat was col-lected from carotid artery into tubes containing sodiumcitrate (1:9, v/v). Platelet-rich plasma was obtained bycentrifugation at 150 g for 10min. Platelets were ad-justed to 4 × 108 platelets per milliliter with HEPESbuffer (137mM NaCl, 2.7mM KCl, 1mM MgCl2,5.6mM glucose, 0.35% bovine serum albumin and3.8mM HEPES, pH6.5). Prepared platelets suspensionwas incubated with the formula (400 μg ·mL�1) or vehi-cle (DMSO, 0.1%, v/v) at 37 °C for 10min, and thenADP (10 μmol · L�1) was added. After 5min, reactionwas terminated at 4 °C, and platelets were collectedby centrifugation at 2000 g for 10min. Treated plateletswere subsequently lysed in RIPA buffer (500mmol ·L�1 Tris–HCl pH7.4, 1mmol · L�1 EDTA, 150mmol ·L�1 NaCl, 1% NP-40, 0.25% Na-deoxycholate,1mmol · L�1 phenylmethylsulfonyl fluoride and prote-ase inhibitor cocktail) and then centrifuged at 12,000×gfor 10min. The protein concentrations were determinedusing Lowry protein assay reagents. Protein blots fromSDS-PAGE were then transferred to luminescence mem-branes (Millipore Inc. Bedford, MA, USA), and the pro-teins were immunoblotted with the correspondingprimary antibody: αIIbβ3, β-actin, ERK1/2, p- ERK1/2followed by second antibody detection. Bound antibodieswere visualized via an enhanced chemiluminescent detec-tion system. The density of bands was quantified using aGel-pro analyser and was expressed as arbitrary units.

Phytother. Res. (2013)

3ANTI-THROMBOTIC EFFECT OFA CORNI FRUCTUS-DERIVED FORMULA

Statistical analysis. Data are expressed as means ± SEanalysed using one-way ANOVA. A Newman–Keulspost hoc analysis was used for pairwise multiple compar-isons when significance was determined by ANOVA. P0.05 was considered statistically significant.

RESULTS

Anti-thrombotic effect of the formula in mice of acutepulmonary thromboembolism

The mice injected with mixture of platelet aggregationagents (ADP and adrenaline) became paralysed,grasping for breath, and even expiration. The formulaintragastrically administrated 1h before the thromboticchallenge gave significant protection against this attack.Compared with mice of model group, those treated withthe formula recovered from paralysis in amarkedly shorterperiod (Fig. 1A). None mouse of expiration was observedin group treated with the formula at dose of 200mg ·kg�1

(Fig. 1B). These data indicated that the formula couldinhibit thrombus formation and arterial occlusion.

Figure 1. Effect of the formula in mice of acute pulmonary throm-boembolism. Mice were administrated orally with the formula orvehicle (CMC-Na, 0.5%) for 6 days and injected mixture of ADP(0.4μmol · kg�1) and adrenaline (2.73μmol · kg�1) into tail vein1h after the last administration. The recovery time (A) and mortal-ity rate (B) of paralysed mice were recorded within 15min. Thedose of aspirin for mice was 18.33mg · kg�1. Data are expressedas mean±SEM, n=10. * p<0.05, ** p<0.01, vs model.

Copyright © 2013 John Wiley & Sons, Ltd.

Anti-thrombotic effect of the formula in rats ofarterial thrombosis

To further investigate the anti-thrombotic effect of theformula, arterial thrombosis of rats induced by electricalstimulation on common carotid artery was reproduced forthe assay. In contrast to those vehicle-treated rats, the oc-clusion time, which indicated the duration of thrombusformation, was significantly extended in these formula-treated rats. The formula reduced thrombus formationin a dose-dependent manner over a range of dose(140–280mg · kg�1) (Fig. 2A). On the other hand, theformula decreased the arterial thrombus weight ofthese rats pronouncedly (Fig. 2B). Taken together, theformula showed positive anti-thrombotic effect, whichwas due to the inhibition of platelet aggregation and/orthrombolysis action.

Effect of the formula on TXB2, 6-keto-PGF1α, cAMP,cGMP and NO in rats with arterial thrombosis

TXA2 and PGI2 have opposite action on platelet,promoting platelet activation and inhibiting platelet aggre-gation, respectively. The concentrations of 6-keto-PGF1α,metabolism product of PGI2, were significantly enhanced

Figure 2. Effect of the formula in rats of arterial thrombosis.Arterial thrombosis was induced by electrical stimulation on com-mon carotid artery in rats administrated orally with the formula orvehicle (CMC-Na, 0.5%) for 6 days. The occlusion time (A) wasrecorded, and thrombus were weighed (B). The dose of aspirinfor rat was 12.83mg · kg�1. Data are expressed as mean±SEM,n=10. ** p<0.01, vs model.

Phytother. Res. (2013)

4 Q.-C. ZHANG ET AL.

in blood of the formula-treated rats with arterial thrombo-sis (Fig. 3A), which indicated the increase of PGI2secretion following the formula administration.Otherwise,the levels of TXB2 metabolism product of TXA2 were notchanged remarkably (Fig. 3B).It has been demonstrated that PGI2 inhibits platelet

aggregation through regulation cyclic nucleotide gener-ation/degradation resulting in elevating of cAMP andcGMP. The increased cGMP in blood was observed inthe formula-treated rats with arterial thrombosis, andno obvious change of cAMP was demonstrated in ourdata (Fig. 3C, D).It iswell known that injuryof endothelial cell under con-

ditions of pathology or electronic stimulation mentionedabovewould causeplatelet adhesionand thrombus forma-tion, and integrity of vessel wall and normal endothelialfunction suchasNOrelease confront thrombus formation.

Figure 3. Effect of the formula on TXB2, 6-keto-PGF1α, cAMP and cGMPelectrical stimulation on common carotid artery in rats administrated olevels of 6-keto-PGF1α (A), TXB2 (B), cAMP (C), cGMP (D) and NO (Ebiochemistry assay. The endothelium morphology was analysed with0.01, vs sham; * p<0.05, ** p<0.01, vs model. This figure is availab

Copyright © 2013 John Wiley & Sons, Ltd.

In thepresent study,more integral vesselwallwas shown inthe formula-treated rats compared with these vehicle-treated (Fig. 3E). Meanwhile, the levels of NO in bloodof rats were also enhanced significantly following the for-mula administration (Fig. 3F). Therefore, the formulawasdemonstrated toregulate the communicationbetweenplatelet and endothelial cells.

Effect of the formula on αIIbβ3 and ERK1/2 inADP-activated platelet in vitro

In our previous investigation, the formula extremelyinhibited ADP-induced platelet aggregation, and ADPcould activate ERK1/2 and αIIbβ3 expression. Hence,the levels of αIIbβ3 and phosphorylation of ERK1/2were assayed in ADP-activated platelet incubated with

in rats of arterial thrombosis. Arterial thrombosis was induced byrally with the formula or vehicle (CMC-Na, 0.5%) for 6 days. The) in plasma of rats were determined with radioimmunoassay andH–E stain (F). Data are expressed as mean ±SEM, n=10. ## ple in colour online at wileyonlinelibrary.com/journal/ptr.

Phytother. Res. (2013)

Figure 4. Effect of the formula on αIIbβ3 expression and ERK1/2 activation in ADP-activated platelet in vitro. Platelets of rat with concentra-tion of 4 ×108 platelets per milliliter was incubated with the formula (400μg ·mL�1) or vehicle (DMSO, 0.1%, v/v) at 37 °C for 10min, andthen ADP (10μmol · L�1) was added for 5min. αIIbβ3 expression and ERK1/2 activation were analysed with Western blotting assay. Data areexpressed as mean±SEM, n=3. ## p<0.01, vs blank; * p<0.05, ** p<0.01, vs ADP. This figure is available in colour online atwileyonlinelibrary.com/journal/ptr.

5ANTI-THROMBOTIC EFFECT OFA CORNI FRUCTUS-DERIVED FORMULA

the formula. As shown in the data, the formula could re-markably decrease the ADP-induced αIIbβ3 expressionand phosphorylation of ERK1/2 (Fig. 4A, B). Inhibitionof ADP signaling in platelet is the underlying mecha-nism of the formula regulating platelet function.

DISCUSSION

Antiplatelet agents are normal means adopted to pre-vent and therapy cardiovascular and cerebrovasculardisease associated with platelet dysfunction. Thehigher mortality of these diseases and the side effectof current antiplatelet agents promote the develop-ment of novel intriguing agents to inhibit platelet ag-gregation and thrombus formation. Naturalcompounds or products from food or herb legiti-mately become prospective candidates due to thesafety and positive efficacy.The formula shown in the present study was com-

posed of malic acid, succinic acid and citric acid, whichwere derived from Corni fructus, a commonly usedtraditional Chinese medicine. It could inhibit ADP-in-duced platelet aggregation, and the maximum inhibitionis 82.82% in a ratio of 3:2:2. In platelet adhesion assay,up to 75% of adhesion platelets was blocked by theformula at 400 μg/mL. However, no influence of the for-mula on platelet viability was observed (Zhang et al.,2013). This study provides further evidence supportingthe anti-thrombosis effect of the formula in mice ofacute pulmonary thromboembolism and rats of arterialthrombosis, which indicated the direct action on acti-vated platelet and the indirect effect embodied throughregulation endothelial cells function. The endothelium-dependent mechanisms are related to the enhancedrelease of PGI2, NO and cGMP. The other pathwayinvolved in the anti-thrombosis effect of the formula isblock ADP signaling in the platelet.Both NO and PGI2 participate the endothelial

thromboregulatory functions as inhibiting platelet adhe-sion, activation and aggregation. NO primarily released

Copyright © 2013 John Wiley & Sons, Ltd.

from endothelium binds to the heme-containing enzymesoluble guanylyl cyclase and induces a conformationalchange activation of sGC, which leads to an increaseof cGMP in intracellular (Bellamy and Garthwaite,2002). The increase of cGMP results in decrease ofCa2+ flux via activation of cGMP-dependent protein ki-nase G, which subsequently reduces the binding affinityof αIIbβ3 to fibrinogen (Pigazzi et al., 1999). NO can alsopromote phosphorylation of TXA2 receptor throughcGMP-dependent mechanism, which interrupts theactivation of platelet via TXA2 (Wang et al., 1998).S-nitrosylation of cysteine residues of protein in intracel-lular or plasma is a cGMP-independent mechanism forNO-mediated platelet inhibition (Walsh et al., 2002). Thus,NO plays a key role in mediating the interaction betweenendothelium and platelet. Endothelial dysfunction or di-minished NO would contribute to neutrophil adhesion tothe endothelium and result in atherosclerosis and throm-bosis (Cohen, 1995). L-arginine, a substrate for endothe-lial NO synthases, inhibits formation of atheroscleroticlesions in animal models and ADP-dependent plateletaggregation of atherosclerosis in human (Aji et al., 1997).

PGI2, a derivative of the C-20 unsaturated fatty acidarachidonic acid with short half-time, is another intrigu-ing antiplatelet and anti-thrombotic agent. It regulatesplatelet function through the G protein-coupled mem-brane receptors on the platelet surface (Dutta-Roy andSinha, 1987). Furthermore, PGI2 and NO at low dosesshow expected synergistic action on the antiaggregatoryeffect (Radomski et al., 1987). The formula of the pres-ent work was demonstrated to inhibit the formation ofarterial thrombosis. The increase of PGI2, cGMP andNO implied the function of the formula in protectionon endothelium and regulation of the communicationbetween platelet and endothelial cell.

Consistent with the previous investigation that theformula inhibits ADP-induced platelet aggregationin vitro (Zhang et al., 2013), the present work providesevidence about the anti-thrombotic effect of the formulain ADP and adrenaline-induced acute pulmonarythromboembolism model in vivo. Decreased αIIbβ3expression and ERK1/2 phosphorylation were also

Phytother. Res. (2013)

6 Q.-C. ZHANG ET AL.

observed in ADP-activated platelet following treatmentof the formula. ADP binding P2Y12, a Gi-coupled recep-tor, leads to activation of αIIbβ3 and ERK1/2, andstabilization of platelet aggregation (Dorsam andKunapuli, 2004). P2Y12 is also the target of the metaboliteof clopidogrel (Savi et al., 2001). Nevertheless, P2Y12 wasshown to involve in platelet microparticle formation andplatelet–leukocyte conjugates (Storey et al., 2002).αIIbβ3, also known as glycoprotein GPIIb-IIIa, acts as atherapeutic target to inhibit the thrombosis process.Down-regulation of αIIbβ3 protein levels might be an-other undirected pathway to interfere αIIbβ3 activationthat demonstrated in the present investigation. Hence,these findings emphasize that one of the most importantanti-thrombotic roles of the formula is interruption ofADP signaling pathway in platelet.In the present study, higher dose of the formula was

administrated to the animal. The effective dose of theformula is about 100mg · kg�1 for rodent, and the theo-retic dose for human is about 750mg per person. Theequivalent dose of the formula (100mg · kg�1 in mice)showed different potency of pharmacology in modelsof mice and rat, which produced significant effect onrat but no-significant action observed in mice may be

Copyright © 2013 John Wiley & Sons, Ltd.

due to the difference of species. The appropriate clinicdose for human maybe 750mg or lower than the theo-retic one, which should be strictly decided by the clinicpharmacological investigation. And now, further phar-macological evaluations are required for the formulaon other animal model such as rabbit.

In summary, direct regulation of platelet function andamelioration of communication between endothelialcells and platelets are both involved in the underlyingmechanism for the formula, which provides great prom-ise for the use of the formula to inhibit thrombosis and/or atherothrombosis.

Acknowledgements

This work was supported by the Priority Academic Program Develop-ment of Jiangsu Higher Education Institutions (PAPD, ysxk-2010) andthe National Foundation of Science of China (No. 81173190).

Conflict of Interest

The authors have declared that there is no conflict of interest.

REFERENCES

Bhatt DL, Topol EJ. 2000. Current role of platelet glycoprotein IIb/IIIainhibitors in acute coronary syndromes. JAMA 284: 1549–1558.

Roger S, Pawlowski M, Habib A, et al. 2004. Costimulation of theGi-coupled ADP receptor and the Gq-coupled TXA2 receptor isrequired for ERK2 activation in collagen-induced plateletaggregation. FEBS Lett 556: 227–235.

Cheng Y, Austin SC, Rocca B, et al. 2002. Role of prostacyclin inthe cardiovascular response to thromboxane A2. Science296: 539–541.

Kam SC, Do JM, Choi JH, et al. 2012. In vivo and in vitro animalinvestigation of the effect of a mixture of herbal extracts fromTribulus terrestris and Cornus officinalis on penile erection. JSex Med 9: 2544–2551.

Choi YH, Jin GY, Li GZ, et al. 2011. Cornuside suppresseslipopolysaccharide-induced inflammatory mediators byinhibiting nuclear factor-kappa B activation in RAW 264.7macrophages. Biol Pharm Bull 34: 959–966.

Sung YH, Chang HK, Kim SE, et al. 2009. Anti-inflammatory andanalgesic effects of the aqueous extract of corni fructus in mu-rine RAW 264.7 macrophage cells. J Med Food 12: 788–795.

Kang DG, Choi DH, Lee JK, et al. 2007. Endothelial NO/cGMP-de-pendent vascular relaxation of cornuside isolated from the fruitof Cornus officinalis. Planta Med 73: 1436–1440.

He LH. 2011. Comparative study on alpha-glucosidase inhibitoryeffects of total iridoid glycosides in the crude products andthe wine-processed products from Cornus officinalis.Yakugaku Zasshi 131: 1801–1805.

Lin MH, Liu HK, Huang WJ, et al. 2011. Evaluation of the potentialhypoglycemic and Beta-cell protective constituents isolatedfrom Corni fructus to tackle insulin-dependent diabetesmellitus. J Agric Food Chem 59: 7743–7751.

Yokozawa T, Kang KS, Park CH, et al. 2010. Bioactive constituentsof Corni Fructus: The therapeutic use of morroniside, loganin,and 7-O-galloyl-D-sedoheptulose as renoprotective agents intype 2 diabetes. Drug Discov Ther 4: 223–234.

WangW,Xu J, Li L, et al. 2010. Neuroprotective effect ofmorronisideon focal cerebral ischemia in rats. Brain Res Bull 83: 196–201.

Jeong EJ, Kim TB, Yang H, et al. 2012. Neuroprotective iridoid glyco-sides fromCornus officinalis fruits against glutamate-induced tox-icity in HT22 hippocampal cells. Phytomedicine 19: 317–321.

JiangWL, Chen XG, Zhu HB, et al. 2009. Cornuside attenuates ap-optosis and ameliorates mitochondrial energy metabolism inrat cortical neurons. Pharmacology 84: 162–170.

Zhao LH, Ding YX, Zhang L, et al. 2010. Cornel iridoid glycosideimproves memory ability and promotes neuronal survival infimbria-fornix transected rats. Eur J Pharmacol 647: 68–74.

Zhang QC, Zhao Y, Bian HM. 2013. Antiplatelet Activity of a NovelFormula Composed of Malic Acid, Succinic Acid and CitricAcid from Cornus officinalis Fruit. Phytother Res Feb 27. doi:10.1002/ptr.4934. [Epub ahead of print]

DiMinno G, Silver MJ. 1983. Mouse antithrombotic assay: a sim-ple method for the evaluation of antithrombotic agentsin vivo. Potentiation of antithrombotic activity by ethyl alco-hol. J Pharmacol Exp Ther 225: 57–60.

Schumacher WA, Steinbacher TE, Heran CL, et al. 1993. Effects ofantithrombotic drugs in a rat model of aspirin-insensitive arte-rial thrombosis. Thromb Haemost 69: 509–514.

Bellamy TC, Garthwaite J. 2002. The receptor-like properties ofnitric oxide-activated soluble guanylyl cyclase in intact cells.Mol Cell Biochem 230: 165–176.

Pigazzi A, Heydrick S, Folli F, et al. 1999. Nitric oxide inhibitsthrombin receptor-activating peptide-induced phosphoinositide3-kinase activity in human platelets. J Biol Chem 274:14368–14375.

Wang GR, Zhu Y, Halushka PV, et al. 1998. Mechanism of plateletinhibition by nitric oxide: in vivo phosphorylation of thrombox-ane receptor by cyclic GMP-dependent protein kinase. ProcNatl Acad Sci U S A 95: 4888–4893.

Walsh DM, Klyubin I, Fadeeva JV et al. 2002. Naturally secretedoligomers of amyloid beta protein potently inhibit hippocampallong-term potentiation in vivo. Nature 416: 535–539.

Cohen RA. 1995. The role of nitric oxide and other endothelium-derived vasoactive substances in vascular disease. ProgCardiovasc Dis 38: 105–128.

Aji W, Ravalli S, Szabolcs M, et al. 1997. L-arginine preventsxanthoma development and inhibits atherosclerosis in LDLreceptor knockout mice. Circulation 95: 430–437.

Dutta-Roy AK, Sinha AK. 1987. Purification and properties ofprostaglandin E1/prostacyclin receptor of human blood plate-lets. J Biol Chem 262: 12685–12691.

Radomski MW, Palmer RM, Moncada S. 1987. The anti-aggregatingproperties of vascular endothelium: interactions between pros-tacyclin and nitric oxide. Br J Pharmacol 92: 639–646.

Dorsam RT, Kunapuli SP. 2004. Central role of the P2Y12 receptorin platelet activation. J Clin Invest 113: 340–345.

Savi P, Labouret C, Delesque N, et al. 2001. P2y(12), a new plate-let ADP receptor, target of clopidogrel. Biochem Biophys ResCommun 283: 379–383.

Storey RF, Judge HM, Wilcox RG, et al. 2002. Inhibition of ADP-induced P-selectin expression and platelet-leukocyte conjugateformation by clopidogrel and the P2Y12 receptor antagonistAR-C69931MXbut not aspirin.ThrombHaemost 88: 488–494.

Phytother. Res. (2013)