logy 553 (2006) 39–45www.elsevier.com/locate/ejphar

European Journal of Pharmaco

Inhibitory effect of Andrographis paniculata extract and its activediterpenoids on platelet aggregation

Piengpen Thisoda a,b, Nuchanart Rangkadilok b, Nanthanit Pholphana b,Luksamee Worasuttayangkurn b, Somsak Ruchirawat c, Jutamaad Satayavivad a,b,d,⁎

a Toxicology Graduate Programme, Faculty of Science, Mahidol University, Rama VI Road, Bangkok 10400, Thailandb Laboratory of Pharmacology, Chulabhorn Research Institute, Vipavadee-Rangsit Highway, Laksi, Bangkok 10210, Thailand

c Laboratory of Medicinal Chemistry, Chulabhorn Research Institute, Vipavadee-Rangsit Highway, Laksi, Bangkok 10210, Thailandd Department of Pharmacology, Faculty of Science, Mahidol University, Rama VI Road, Bangkok 10400, Thailand

Received 3 March 2006; received in revised form 1 September 2006; accepted 12 September 2006Available online 30 September 2006

Abstract

Andrographis paniculata has been widely used for the prevention and treatment of common cold especially in Asia and Scandinavia. The threeactive diterpenoids from this plant, including aqueous plant extracts, were investigated for the inhibitory effect on platelet aggregation in vitro. Theresults indicated that andrographolide (AP1) and 14-deoxy-11,12-didehydroandrographolide (AP3) significantly inhibited thrombin-induced plateletaggregation in a concentration-(1–100 μM) and time-dependent manner while neoandrographolide (AP4) had little or no activity. AP3 exhibitedhigher antiplatelet activity than AP1 with IC50 values ranging from 10 to 50 μM. The inhibitory mechanism of AP1 and AP3 on platelet aggregationwas also evaluated and the results indicated that the inhibition of extracellular signal-regulated kinase1/2 (ERK1/2) pathway may contribute toantiplatelet activity of these two compounds. In addition, standardized aqueous extracts of A. paniculata containing different amounts of AP3inhibited thrombin-induced aggregation to different degrees. The extracts significantly decreased platelet aggregation in a concentration-(10–100μg/ml)and time-dependent manner. However, the extract with high level of AP3 (Extract B) (IC50 values=50–75 μg/ml) showed less inhibitory activity againstthrombin than the extract with lower level of AP3 (Extract A) (IC50 values=25–50 μg/ml). These results indicate that the standardized A. paniculataextract may contain other antiplatelet compounds rather than AP1 and AP3, which contribute to high antiplatelet activity. Therefore, the consumption ofA. paniculata products may help to prevent or treat some cardiovascular disorders i.e. thrombosis; however, it should be used with caution by patientswith bleeding disorders.© 2006 Elsevier B.V. All rights reserved.

Keywords: Andrographis paniculata; Diterpenoids; Inhibition; Aggregation; Thrombin; ERK1/2

1. Introduction

Andrographis paniculata (Burm. f.) Nees (Acanthaceae) hasbeen used as a traditional medicine in India, China, Thailand,and Scandinavia. The aerial parts of the plant (leaves and stems)are normally used for extraction of the active phytochemicals.Extracts of the plant and their constituents have been reported toexhibit a wide spectrum of biological activities of therapeutic

⁎ Corresponding author. Laboratory of Pharmacology, Chulabhorn ResearchInstitute, 54 Moo 4, Vipavadee-Rangsit Highway, Laksi, Bangkok, 10210,Thailand. Tel.: +66 2 574 0622; fax: +66 2 574 2027.

E-mail address: jutamaad@cri.or.th (J. Satayavivad).

0014-2999/$ - see front matter © 2006 Elsevier B.V. All rights reserved.doi:10.1016/j.ejphar.2006.09.052

importance including antibacterial, antiviral (Chang et al., 1991;Calabrese et al., 2000), anti-inflammatory (Shen et al., 2002),antimalarial (Misra et al., 1992), immuno-stimulant (Puri et al.,1993; See et al., 2002), hepatoprotective (Kapil et al., 1993),antithrombotic (Zhao and Fang, 1991), anticancer (Matsudaet al., 1994), hypoglycemic (Zhang and Tan, 2000), and hy-potensive (Zhang and Tan, 1996) properties. Standardized extractof A. paniculata (SHA-10) containing andrographolide anddeoxyandrographolide has been used to reduce the prevalenceand intensity of the symptoms in uncomplicated common cold(Caceres et al., 1997; Hancke et al., 1995; Melchior et al., 2000).

The active constituents of A. paniculata are diterpenelactones including andrographolide (AP1), 14-deoxy-11,12-

40 P. Thisoda et al. / European Journal of Pharmacology 553 (2006) 39–45

didehydroandrographolide (AP3), neoandrographolide (AP4)(Fig. 1). Andrographolide is considered to be the most activeand important constituent in this plant. However, each com-ponent possesses some different potency in pharmacologicalactivities. For instance, AP1 shows high activity for anti-in-flammation (Shen et al., 2000) and hepatoprotection againstgalactosamine and paracetamol intoxication (Chander et al.,1995), AP3 produces a potent hypotensive effect (Zhang et al.,1998), whilst AP4 has greater activity against malaria (Misra etal., 1992). The differences in pharmacological activities of thesecompounds may be related to their structures. For example,recent study on structure-activity relationships of andrographo-lide analogues and their pharmacological activities indicated thatAP3 synthesized fromAP1 inhibitedα-glucosidase stronger thanAP1 itself (Dai et al., 2006). The flexible chain between thebutyrolactone moiety and the two six-membered rings may becritical to α-glucosidase inhibitory activity. Therefore, theysuggested that probably AP3 might play an important role in theA. paniculata plant extract exerting antidiabetic activity. Inaddition, the different amounts of each active principle in thisplant are also very important for the therapeutic efficacy or sideeffects of this plant. Our previous study reported that amongthese active compounds, the amount of AP3 dramatically in-creased during storage of crude powders and extracts (Phol-phana et al., 2004). An increase of AP3 in the old plant materialsmay be due to the degradation of AP1 during the storage time toyield AP3 as the major degraded product. Lomlim et al. (2003)indicated that the crystalline form of AP1 was highly stable at70 °C (75% relative humidity) over a period of 3 months whileits amorphous form degraded promptly during 2-months underthe same condition. The major degradation product of AP1 wasAP3. He et al. (2003) also reported that AP3 is one of the AP1metabolites isolated from rat urine, faeces, and the content of thesmall intestine. The presence of this possible breakdown productin the stored plant extract may increase or reduce the phar-macological activities of this extract.

Platelet plays a key role in the physiological hemostaticprocess and pathologic thrombosis. AP1 has been demonstratedto possess antiplatelet activity and to inhibit platelet activatingfactor (PAF)-induced human blood platelet aggregation in adose-dependent manner (Amroyan et al., 1999). There is still noreport comparing the antiplatelet activity of these purecompounds (AP1, AP3, and AP4) isolated from this plant. Theantiplatelet activity of these three compounds may be developed

Fig. 1. Structures of andrographolide (AP1), 14-deoxy-11,12-didehydroandro-grapholide (AP3) and neoandrographolide (AP4).

further to use as antiplatelet aggregation drugs to treat somecardiovascular diseases. However, when this plant is used forthe treatment of common cold or taken with other antiplatelet orantithrombotic agents such as aspirin, the signs of plateletdysfunction should be also concerned as it may increase the riskof bruising and bleeding. In addition, if AP3 or AP4 also hassignificant antiplatelet activity, there will be a possibility thatthe adverse effects of long storage of dried powders or extractsmay occur when this plant preparation is used. Therefore, thepresent study investigates the inhibitory effect of three activediterpenoids (AP1, AP3, AP4) isolated from A. paniculata onplatelet aggregation in vitro and some possible mechanismsinvolved in this inhibitory activity. Standardized extracts, bothfresh and old harvest, containing different proportions of thesecompounds have also been investigated.

2. Materials and methods

2.1. Drugs and chemicals

Three standard diterpenoids, andrographolide (AP1), 14-deoxy-11,12-didehydro-andrographolide (AP3), and neoandro-grapholide (AP4), were purified and identified using TLC, UVspectrum, IR and NMR by the Laboratory of Natural Products,Chulabhorn Research Institute, Thailand. Anti-phospho p-44/42MAP Kinase (Thr202/Tyr204) antibody was purchased fromCell Signaling Technology (Beverly, MA, USA). Anti-mouseIg, horseradish peroxidase linked whole antibody was fromAmersham Biosciences (Buckinghamshire,UK). Thrombin andbovine serum albumin (BSA) were purchased from Sigma (St.Louis, MO, USA), and pentobarbital was obtained fromVetbutal (Polfa, Poland). All other reagents were of analyticalgrade.

2.2. Preparation of plant extracts

A. paniculata plant was kindly identified by Dr. WongsatitChuakul and a voucher specimen was deposited at the Phar-maceutical Botany Mahidol Herbarium, Department of Phar-maceutical Botany, Faculty of Pharmacy, Mahidol University,Bangkok, Thailand (PBM 3760). Two dried A. paniculata rawmaterials were selected. One was recently harvested fromNakornpathom province (January, 2005), Extract A; and theother was another raw plant material stored at room temperaturefor 2 years (January, 2003), Extract B. These two plant materialswere selected as they contained different level of only AP3 whileAP1 and AP4 were present at the same levels. Four hundredgrams of aerial parts of A. paniculata were extracted with 4 l ofhot water (70–75 °C) for 1 h. The extract was then filtered andcollected. The residue was re-extracted twice with 4 l of hotwater each time. The combined water extracts were lyoph-ilized by using a freeze dry method. The %yield of extracts wasbetween 27–30%. Five milligrams of plant extracts weredissolved in 5.0 ml of hot water (70–75 °C) and vigorouslyshaken. The extracts were left at room temperature until theycooled down and then filtered through a 0.45 μm nylonmembrane (13 mm, Orange Scientific, Belgium) prior to

Fig. 2. A concentration-dependent inhibition of platelet aggregation byandrographolide (AP1), 14-deoxy-11,12-didehydroandrographolide (AP3) andneoandrographolide (AP4) treatments in washed platelets of rats. Aggregationwas initiated by a threshold concentration of thrombin. Data represent means±S.E.M. (n=6). The first bar of each graph represents the control group of eachtreatment. **Significant difference from the control group at Pb0.01.

41P. Thisoda et al. / European Journal of Pharmacology 553 (2006) 39–45

analysis by using High Performance Liquid Chromatography(HPLC). All extracts were analyzed for the contents of threeactive diterpenoids (AP1, AP3, AP4) by using HPLC analysispreviously reported by Pholphana et al. (2004). Each purecompound was dissolved in dimethylsulfoxide (DMSO) toobtain 100 μM stock solution. Plant extracts, Extract A andExtract B, were dissolved in Tyrode buffer to obtain the finalconcentration of 100 μg/ml.

2.3. Animals

Male Wistar rats, weighing approximately 250–300 g, weresupplied by the National Laboratory Animal Center, Salaya,Nakornpathom, Thailand. The animals had free access to foodpellets (C.P. feed number 082) and tap water. They were kept in aroom with controlled temperature (23±2 °C), humidity (50%),and light cycle (12-h light/12-h dark), and one animal perhanging cage. An acclimatization period of 7 days was allowedbefore experimentation. The experimental protocols were ap-proved by Chulabhorn Research Institute-Animal Care and UseCommittee (CRI-ACUC), Bangkok.

2.4. Preparation of washed platelets

Rats were anesthetized with sodium pentobarbital (50 mg/kg,i.p.). Blood samples were taken from the heart and transferredinto plastic tubes containing acid–citrate–dextrose (ACD)(75 mM trisodium citrate, 38 mM citric acid and 138 mMglucose) as anticoagulant in a volume ratio 6:1. Platelet richplasma was obtained by centrifugation of blood at 400 ×g for10 min at room temperature. Platelet rich plasma was thencentrifuged at 800 ×g for 10 min and the obtained platelets werewashedwith Tyrode buffer (136mMNaCl, 2.7 mMKCl, 10mMHEPES, 12 mM NaHCO3, 0.34 mM Na2HPO4, 1 mM MgCl2,5.5 mM glucose) containing 0.35% albumin and 10% ACD, pH6.5. The washed platelets were then centrifuged again at 800 ×gfor 10 min and finally re-suspended in Tyrode buffer containing0.1% albumin, pH 7.4. Platelet number was counted by a CoulterCounter (JT; Coulter Electronics, Inc., USA) and adjusted to3×108 cells/ml.

2.5. Assay of platelet aggregation in vitro

Aggregation experiments were performed using a lumi-ag-gregometer (Chronolog Corp., Havertown, PA). Washed plate-lets were incubated with various tested compounds or vehicles at37 °C for 15, 30 and 60 min. Then aggregation was induced byaddition of an agonist, thrombin, while stirring at 1000 rpm in asilicone-treated glass cuvette, in a final volume of 250 μl. Thereaction was then allowed to proceed for 4 min, and the extent ofaggregation was expressed as the percentage of aggregation ofthe control values. For antiplatelet experiments, the aggregationwas performed using the threshold concentration of thrombin(0.1 U/ml). Threshold is defined as the lowest concentration ofan agonist that evokes irreversible aggregation, with amplitudebetween 65% and 85% of the potential maximum deflection inlight transmission.

2.6. Analysis of ERK phosphorylation in platelets

Washed rat platelets (3×108/ml) were incubated with100 μM of AP1, AP3, AP4, or vehicle control (DMSO) for 1 hat 37 °C and then stimulated with thrombin (0.1 U/ml) understirring for the 0.5, 1, 2, or 3 min. The platelets were solubilizedin 2X Laemmli sample buffer (62.5 mM Tris–HCl, pH 6.8, 25%glycerol, 2% SDS, 0.01% bromophenol blue) containing 5%2-mercaptoethanol. Samples were then heated at 95 °C for 5min.Proteins were separated in 7.5% SDS-polyacrylamide gels andtransferred to nitrocellulose membranes. The membranes werewashed with TBST (20 mM Tris-buffer saline 0.1 % Tween 20)for 5 min and then incubated in a blocking buffer (5% nonfatdry milk in TBST) for 1 h with agitation at room temperature.The membranes were incubated overnight with anti-phosphop-44/42 MAP Kinase (Thr202/Tyr204) antibody (Cell Signal-ing Technology; Beverly MA, USA) at 4 °C with gentle agi-tation. In the next day, blots were then incubated with anti-mouseIg, horseradish peroxidase linked whole antibody (AmershamBiosciences, Buckinghamshire,UK) for 2 h at room temperature.Immunoreactive bands were visualized using enhanced chemi-luminescence. At least three independent experiments wereperformed for each treatment.

2.7. Statistical analysis

The means and standard errors of means were calculated forall experiment groups. The data were subjected to the Student'st-test and one-way analysis of variance followed by Duncan'smultiple range test to determine whether means were signifi-cantly different from each other or the control group.

3. Results

AP1, AP3 and AP4 at concentrations of 1–100 μM wereadded to a suspension of washed rat platelets for 60 min beforeaddition of threshold concentration of thrombin. AP1 and AP3inhibited the platelet aggregation induced by thrombin in a

Fig. 3. A time-dependent inhibition of platelet aggregation by 14-deoxy-11,12-didehydroandrographolide (AP3) at the concentration of 100 μM in washedplatelets of rats. Aggregation was initiated by a threshold concentration ofthrombin. Data represent means±S.E.M. (n=6). **,***Significant differencefrom the control group at Pb0.01 and 0.001, respectively.

Fig. 5. Effect of the three diterpenoids (AP1, AP3, and AP4) on thephosphorylation of ERK2. Washed rat platelets were preincubated with eachditerpenoid (100 μM) or vehicle control (DMSO) for 60 min at 37 °C and thenstimulated with thrombin (0.1 U/ml) in the presence of 1 mMCaCl2 for 0.5, 1, 2,or 3 min, with stirring. Platelet lysates were analyzed by SDS-PAGE followedby Western-blotting using antibody specifically recognizing phosphorylatedERK1/2. Results are representative of 3 experiments.

42 P. Thisoda et al. / European Journal of Pharmacology 553 (2006) 39–45

concentration-dependent manner while AP4 showed less effector no activity (Fig. 2). AP1 and AP3 significantly decreased(Pb0.01) platelet aggregation at concentrations of 50–100, and10–100 μM, respectively. The IC50 value of AP3 ranged from 10to 50 μMand IC50 of AP1 was between 50–100 μM. In addition,AP3 at a concentration of 100 μMwas added to a rat platelet for15, 30, and 60 min before addition of thrombin to study theeffect of incubation time of the AP3 activity. Compound AP3significantly decreased (Pb0.01 and 0.001) platelet aggregationin a time-dependent manner (Fig. 3). Addition of AP3 60 minbefore thrombin-induced platelet aggregation showed the mosteffective inhibition (N90%) (Fig. 4A).Since extracellular signal-regulated kinase (ERK1/2) has been shown to be activated afterstimulation with thrombin (Papkoff et al., 1994) and thisactivation has been implicated in platelet aggregation, the effectsof AP1, AP3, and AP4 on ERK1/2 phosphorylation induced bythrombin were examined. The present results indicated thatERK2 was not phosphorylated in resting platelets (Fig. 5).However, ERK2 phosphorylation occurred in thrombin activat-ed platelet aggregation starting at 30 s after thrombin activation

Fig. 4. Effect of (A) andrographolide (AP1), 14-deoxy-11,12-didehydroandro-grapholide (AP3), 100 μM, and (B) two aqueous extracts (Extract A and ExtractB), 100 μg/ml on thrombin-induced platelet aggregation. Washed platelets wereincubated with each compound or vehicle control for 60 min. Aggregation wasinitiated by a threshold concentration of thrombin.

and reached maximum at 2 min. Next, we investigated whetherpreincubation with AP1, AP3, or AP4 affected ERK phosphor-ylation in thrombin-activated platelet aggregation. AP3 wasfound to be more potent than AP1 in the inhibition of ERK2phosphorylation while AP4 did not show any inhibitory effect.The inhibitory effect of AP3 on ERK2 phosphorylation wasmore prolonged than that of AP1 as the strong inhibitory effectwas still observed from 1 to 3 min. These results also correlatedwell with our results observed from the inhibition of plateletaggregation. Taken together, the present results indicated that theinhibitory effect of AP1 and AP3 on platelet aggregation wasmediated through the down-regulation of ERK2 phosphoryla-tion in platelets activated by thrombin, and AP3 was the mostpotent antiplatelet agent in in vitro testing.

In addition, fresh and old aqueous extracts of A. paniculatacontaining different amounts of AP3 were used to study theinhibitory effect on thrombin-induced platelet aggregation atconcentrations of 10–100 μg/ml. The contents of three activediterpenoids (AP1, AP3 and AP4) in these two aqueous extractsof A. paniculata are shown in Table 1. Freshly harvested plant(Extract A) contained low level of AP3 (4.08 mg/g dry weight)while plant materials stored up to 2 years (Extract B) containedabout 3-fold higher levels of this compound (11.34 mg/g dryweight). However, the contents of AP1 and AP4 in Extract Awere not much different to those in Extract B. Both A.

Table 1The contents of andrographolide (AP1), 14-deoxy-11,12-didehydroandrographolide(AP3) and neoandrographolide (AP4), in the aqueous extracts of A. paniculata

Extract AP1 content(mg/g dry weight)

AP3 content(mg/g dry weight)

AP4 content(mg/g dry weight)

Extract A(January, 2005)

31.80±0.23 4.08±0.28 10.23±0.59

Extract B(January, 2003)

37.77±0.21 11.34±1.05 7.98±0.92

Values represent means±SEM (n=3).

Fig. 6. (A) Concentration-and (B) time-dependent inhibition of plateletaggregation by the aqueous extracts of A. paniculata (Extract A and Extract B)in washed platelets of rats. Aggregation was initiated by a threshold concentrationof thrombin. Data represent means±S.E.M. (n=6). *,**,***Significant differencefrom the control group at Pb0.05, 0.01 and 0.001, respectively.

43P. Thisoda et al. / European Journal of Pharmacology 553 (2006) 39–45

paniculata aqueous extracts strongly inhibited platelet aggrega-tion (Fig. 4B). The results showed that these extracts alsoinhibited platelet aggregation in a concentration- and time-de-pendent manner (Fig. 6). Extract A (low AP3) significantlydecreased (Pb0.01) platelet aggregation with the IC50 valueranging between 25 to 50 μg/ml while Extract B (high AP3)significantly inhibited (Pb0.01) aggregation with the IC50 valuebetween 50–75 μg/ml (Fig. 6A). At the concentration of100 μg/ml, both A. paniculata extracts showed a highly sig-nificant inhibition (Pb0.01 and 0.001) on platelet aggregationat the incubation time of 60 min as shown in Fig. 6B. Theseresults were similar to those activities of AP3. However, ExtractA was found to be a stronger inhibitor than Extract B. It alsoshould be noted that Extract A has more rapid onset of actionthan Extract B.

4. Discussion

A. paniculata has been widely used in many countries for thetreatment of common cold and other diseases such as diarrhea,inflammation and diabetes. The present study demonstratedthat at least two active diterpenoids, andrographolide (AP1),14-deoxy-11,12-didehydroandrographolide (AP3), isolated from

A. paniculata remarkably decreased platelet aggregation inducedby thrombin in a concentration-and time-dependent manner.Neoandrographolide (AP4) was found to be a very weak in-hibitor. This is the first report indicating antiplatelet activity ofthese three active diterpenoids present in this plant. In addition,the aqueous extracts of A. paniculata also inhibited plateletaggregation induced by thrombin. These results agree with theprevious study by Amroyan et al. (1999) who indicated that AP1and 40% ethanolic extract of A. paniculata inhibited humanplatelet aggregation induced by PAF. Their study reported theIC50 of AP1 with PAF-induced aggregation was about 5 μM;however, the IC50 of AP1 with thrombin in our study wasbetween 50–100 μM. The compound AP3 was found to be amore effective inhibitor of thrombin-induced platelet aggrega-tion than AP1. At the highest concentration (100 μM), AP3strongly inhibited platelet aggregation in vitro bymore than 95%while AP1 inhibited platelet aggregation by only 70%. As theresults showed that AP3 exhibited the most potent antiplateletaggregation followed by AP1, and AP4 had the least activity,therefore, the position of the double bond between thebutyrolactone moiety and the two six-membered rings (C11–C12 in AP3, and C12–C13 in AP1) may be important for theeffective antiplatelet aggregation while the glucose groupconnected to C19 in AP4 may not have any significant role inthis inhibitory activity in in vitro study. The relationship of theantiplatelet activity of AP1, AP3, and AP4 related to their struc-tures should be further investigated.

The present study examined the antiplatelet activity of twoaqueous extracts of A. paniculata, which contained differentamounts of AP3. In contrast to the results of the pure compounds,the extract containing high level of AP3 (Extract B) showed lessantiplatelet activity than that found in the extract with low levelof this compound (Extract A). As our recent study reported thatthe amount of AP3 highly increased during a storage time ofmore than one year (Pholphana et al., 2004), thus, the use of theold plant materials for the treatment of common cold mayproduce some side effects on platelet aggregation. However, thepresent results indicated that the amounts of AP3 inA. paniculataextracts (IC50=25–50 μg/ml contained AP3=0.10–0.20 μg/ml)may not be the major factor contributing to significantantiplatelet activity because the concentration of AP3 in plantextract was relatively small when it was compared to the IC50 ofpure AP3 (10–50 μM or 3.3–16.6 mg/ml). Other compoundssuch as flavones, flavonoids, and other diterpenoids (e.g. 14-deoxyandrographolide, andrographiside) may also contribute tothe high potency of antiplatelet activity, or the synergisticinteraction among those compounds in the extracts may bemuchmore effective than the activity of each pure compound itself.Further study is needed to evaluate the effect of the combinationof these three compounds on platelet aggregation. In addition,from the HPLC fingerprint, there are also other compoundsrather than AP1, AP3, and AP4, present in this plant (Fig. 7). Thefreshly harvested plant, Extract A, contained some othercompounds (at retention time 1.75 and 2.45 min) at higherlevels than the plant stored up to 2 years. These two compoundsmay be more effective inhibitors of platelet aggregation ratherthan the three diterpenoids. The previous study also reported the

Fig. 7. HPLC fingerprints of the active compounds present in the aqueousextracts of A. paniculata, (A) Extract A; old plant material (B) Extract B; freshplant material.

44 P. Thisoda et al. / European Journal of Pharmacology 553 (2006) 39–45

presence of many flavones extracted from A. paniculata (Zhaoand Fang, 1991; Reddy et al., 2005). Flavone extracted from theroot of A. paniculata was found to inhibit platelet aggregationand the production of thromboxane B2 (TXB2) as reported byZhao and Fang (1991). Their results also indicated that flavonemight promote the synthesis of prostaglandin I2 (PGI2), inhibitthe production of TXA2, stimulate the synthesis of cAMP inplatelets, and thereby prevent the formation of thrombi as well asthe development of myocardial infarction. In addition, Zhanget al. (1994) observed 63 patients of cardiac and cerebral vas-cular disease at 3 h and 1 week after takingA. paniculata extractsand reported that platelet aggregation induced by ADP wassignificantly inhibited (Pb0.001) and serotonin (5-HT) releasedfrom the platelet decreased (Pb0.01), but plasma 5-HT levelsremained unchanged. They also suggested that a rise in theplatelet cAMP level might be another mechanism of the anti-platelet effect of this extract.

Among the active components isolated from A. paniculata,the pharmacological properties of andrographolide have beenstudied extensively, and some of them have been linked to theMAPK signaling. The pharmacological properties of androgra-pholide that have been demonstrated to mediate, in part, throughthe inhibition of ERK1/2 signaling pathway including theinhibition of TNF-α production in LPS-stimulated macrophages(Qin et al., 2006); the suppression of complement 5a-inducedmacrophage migration (Tsai et al., 2004); the inhibition ofinterferon-γ production from concanavaline A-activated T-cells

(Burgos et al., 2005b). However, it has been proposed that theinhibitory effect of andrographolide on serum deprivation-induced human umbilical vein endothelial cells apoptosis wasmediated by the activation of phosphatidyl inositol-3-kinase(PI3K), but it did not affect ERK1/2 activity (Chen et al., 2004).This indicated that ERK1/2 activity may not be the mediator ofandrographolide for some certain biological activities.

In platelets, ERKs have been shown to be activated afterthe stimulation by thrombin (Papkoff et al., 1994), collagen(Börsch-Haubold et al., 1996), or phorbal ester (Aharonovitzet al., 1998). In addition, ERKs also play a role during store-mediated Ca2+ entry (Rosado and Sage, 2001). ERK2signaling in platelets is required for the release of stored Ca2+,GPIb-IX-dependent activation of integrin αIIbβ3, and plateletaggregation in response to low doses but not to high doses ofthrombin, collagen arrachidonic acid, and U46619 (Li et al.,2001; McNicol et al., 1998). Activated αIIbβ3 mediates plateletadhesion and aggregation, and plays critical roles in the de-velopment of thrombotic diseases (Shattil et al., 1998; Parise,1999). In our study, we demonstrated that the down-regulation ofERK2 phosphorylation contributes, in part, to the antiplateletaggregation of AP1 and AP3 in thrombin-induced aggregation.

In addition, it was shown that additional entry of Ca2+ isnecessary to reach sufficiently high Ca2+ to allow platelet ac-tivation (Heemskerk et al., 1991). Thrombin and ADP agonistsalso stimulated influx of extracellular Ca2+. Burgos et al. (2000)found that A. paniculata extract selectively blockaded voltage-operated calcium channels, hence inhibiting 45Ca2+ influx. Re-cently, Burgos et al. (2005a) also reported that 14-deoxyandro-grapholide (its structure similar to AP3) isolated from A.paniculata is an effective antagonist of PAF-mediated processesin bovine neutrophils, probably by its calcium channel blockingproperty. Therefore, the reduction of Ca2+ influx into plateletmay be another mechanism of antiplatelet activity of thesediterpenoids and the plant extracts.

In conclusion, many cardiovascular diseases can be attributedto excessive platelet aggregation, which has a critical role inthrombus formation (Lee et al., 1998). It appears that A.paniculata extract and its active diterpenoids, AP1 and AP3, caninhibit platelet aggregation in vitro; therefore, they may be usedto treat or prevent some cardiovascular diseases. The presentstudy may lead to the development or synthesis of the newantiplatelet aggregation drugs from this medicinal plant that isbeneficial for cardiovascular disorder patients. However, thisplant should be used with caution by patients with bleedingdisorders as it may increase the risk of bleeding. Our resultsindicate that in the use of medicinal plant extracts, especiallyA. paniculata, HPLC chemical fingerprint of the active com-pounds is necessary because the standardized extracts using AP1alone as a marker may not be sufficient for the quality control ofthe plant products.

Acknowledgements

The authors would like to thank Mr. Supachai Ritruechai andMs. Jittra Saehun, laboratory of Pharmacology, ChulabhornResearch Institute, for preparing A. paniculata extracts and their

45P. Thisoda et al. / European Journal of Pharmacology 553 (2006) 39–45

technical assistance in HPLC analysis in this study. We alsothank Ms. Wantika Chantara for technical support on WesternBlot analysis and Dr. Piyajit Watcharasit for valuable discussion.

References

Aharonovitz, O., Aboulafia-Etzion, S., Leor, J., Battler, A., Granot, Y., 1998.Stimulation of 42/44 kDa mitogen-activated protein kinases by argininevasopressin in rat cardiomyocytes. Biochim. Biophys. Acta 1401, 105–111.

Amroyan, E., Gabrielian, E., Panossian, A., Wikman, G., Wagner, H., 1999.Inhibitory effect of andrographolide from Andrographis paniculata on PAF-induced platelet aggregation. Phytomedicine 6, 27–31.

Börsch-Haubold, A.G., Kramer, R.M., Watson, S.P., 1996. Inhibition ofmitogen-activated protein kinase does not impair primary activation ofhuman platelets. Biochem. J. 318 (Pt 1), 207–212.

Burgos, R.A., Imilan, M., Sánchez, N.S., Hancke, J.L., 2000. Andrographispaniculata (Nees) selectively blocks voltage-operated calcium channels inrat vas deferens. J. Ethnopharmacol. 71, 115–121.

Burgos, R.A., Hidalgo,M.A.,Monsalve, J., LaBranche, T.P., Eyre, P.,Hancke, J.L.,2005a. 14-Deoxyandrographolide as a platelet activating factor antagonist inbovine neutrophils. Planta Med. 71, 604–608.

Burgos, R.A., Seguel, K., Perez, M., Meneses, A., Ortega, M., Guarda, M.I.,Loaiza, A., Hancke, J.L., 2005b. Andrographolide inhibits IFN-gamma andIL-2 cytokine production and protects against cell apoptosis. Planta Med. 71,429–434.

Caceres, D.D., Hancke, J.L., Burgos, R.A., Wikman, G.K., 1997. Prevention ofcommon colds with Andrographis paniculata dried extract. A pilot doubleblind trial. Phytomedicine 4, 101–104.

Calabrese, C., Berman, S.H., Babish, J.G., Ma, X., Shinto, L., Dorr, M., Wells, K.,Wenner, C.A., Standish, L.J., 2000. A phase I trial of andrographolide in HIVpositive patients and normal volunteers. Phytother. Res. 14, 333–338.

Chander, R., Srivastava, V., Tandon, J.S., Kapoor, N.K., 1995. Antihepatotoxicactivity of diterpenes of Andrographis paniculata (Kal-Megh) againstPlasmodium berghei induced hepatic damage inMastomys natalensis. Int. J.Pharmacogn. 33, 135–138.

Chang, R.S., Ding, L., Chen, G.Q., Pan, Q.C., Zhao, Z.L., Smith, K.M., 1991.Dehydroandrographolide succinic acid monoester as an inhibitor against thehuman immunodeficiency virus. Proc. Soc. Exp. Biol. Med. 197, 59–66.

Chen, J.H., Hsiao, G., Lee, A.R., Wu, C.C., Yen, M.H., 2004. Andrographolidesuppresses endothelial cell apoptosis via activation of phosphatidyl inositol-3-kinase/Akt pathway. Biochem. Pharmacol. 67, 1337–1345.

Dai, G.F., Xu, H.W., Wang, J.F., Liu, F.W., Liu, H.M., 2006. Studies on thenovel alpha-glucosidase inhibitory activity and structure-activity relation-ships for andrographolide analogues. Bioorg. Med. Chem. Lett. 16,2710–2713.

Hancke, J.L., Burgos, R.A., Caceres, D.D., Wikman, G.A., 1995. A double-blindstudywith a newmonodrugKan Jang: decrease of symptoms and improvementin the recovery from common colds. Phytother. Res. 9, 559–562.

He, X., Li, J., Gao, H., Qiu, F., Cui, X., Yao, X., 2003. Six new andrographolidemetabolites in rats. Chem. Pharm. Bull. (Tokyo) 51, 586–589.

Heemskerk, J.W.M., Feijge, M.A.H., Rietman, E., Hornstra, G., 1991. Ratplatelets are deficient in internal Ca2+release and require influx of extra-cellular Ca2+for activation. FEBS Lett. 284, 223–226.

Kapil, A., Koul, I.B., Banerjee, S.K., Gupta, B.D., 1993. Antihepatotoxic effectsof major diterpenoid constituents of Andrographis paniculata. Biochem.Pharmacol. 46, 182–185.

Lee, J.Y., Lee, M.Y., Chung, S.M., Chung, J.H., 1998. Chemically inducedplatelet lysis causes vasoconstriction by release of serotonin. Toxicol. Appl.Pharmacol. 149, 235–242.

Li, Z., Xi, X., Du, X., 2001. A mitogen-activated protein kinase-dependentsignaling pathway in the activation of platelet integrin alpha IIbbeta3.J. Biol. Chem. 276, 42226–42232.

Lomlim, L., Jirayupong, N., Plubrukarn, A., 2003. Heat-accelerated degradationof solid-state andrographolide. Chem. Pharm. Bull. (Tokyo) 51, 24–26.

Matsuda, T., Kuroyanagi, M., Sugiyama, S., Umehara, K., Ueno, A., Nishi, K.,1994. Cell differentiation-inducing diterpenes from Andrographis panicu-lata Nees. Chem. Pharm. Bull. (Tokyo) 42, 1216–1225.

McNicol, A., Philpott, C.L., Shibou, T.S., Israels, S.J., 1998. Effects of themitogen-activated protein (MAP) kinase kinase inhibitor 2-(2′-amino-3′-methoxyphenyl)-oxanaphthalen-4-one (PD98059) on human platelet acti-vation. Biochem. Pharmacol. 55, 1759–1767.

Melchior, J., Spasov, A.A., Ostrovskij, O.V., Bulanov, A.E., Wikman, G., 2000.Double-blind, placebo-controlled pilot and phase III study of activity ofstandardized Andrographis paniculata Herba Nees. extract fixed combina-tion (Kan jang) in the treatment of uncomplicated upper-respiratory tractinfection. Phytomedicine 7, 341–350.

Misra, P., Pal, N.L., Guru, P.Y., Katiyar, J.C., Srivastava, V., Tandon, J.S., 1992.Antimalarial activity of Andrographis paniculata (Kalmegh) against Plas-modium berghei NK 65 in Mastomys natalensis. Int. J. Pharmacogn. 30,263–274.

Papkoff, J., Chen, R.H., Blenis, J., Forsman, J., 1994. p42 mitogen-activatedprotein kinase and p90 ribosomal S6 kinase are selectively phosphorylatedand activated during thrombin-induced platelet activation and aggregation.Mol. Cell. Biol. 14, 463–472.

Parise, L.V., 1999. Integrin alpha(IIb)beta(3) signaling in platelet adhesion andaggregation. Curr. Opin. Cell Biol. 11, 597–601.

Pholphana, N., Rangkadilok, N., Thongnest, S., Ruchirawat, S., Ruchirawat, M.,Satayavivad, J., 2004. Determination and variation of three activediterpenoids in Andrographis paniculata (Burm.f.) Nees. Phytochem.Anal. 15, 365–371.

Puri, A., Saxena, R., Saxena, R.P., Saxena, K.C., Srivastava, V., Tandon, J.S.,1993. Immunostimulant agents from Andrographis paniculata. J. Nat. Prod.56, 995–999.

Qin, L.H., Kong, L., Shi, G.J., Wang, Z.T., Ge, B.X., 2006. Andrographolideinhibits the production of TNF-alpha and interleukin-12 in lipopolysaccha-ride-stimulated macrophages: role of mitogen-activated protein kinases.Biol. Pharm. Bull. 29, 220–224.

Reddy, V.L., Reddy, S.M., Ravikanth, V., Krishnaiah, P., Goud, T.V., Rao, T.P.,Ram, T.S., Gonnade, R.G., Bhadbhade, M., Venkateswarlu, Y., 2005. A newbis-andrographolide ether from Andrographis paniculata Nees. andevaluation of anti-HIV activity. Nat. Prod. Res. 19, 223–230.

Rosado, J.A., Sage, S.O., 2001. Role of the ERK pathway in the activation ofstore-mediated calcium entry in human platelets. J. Biol. Chem. 276,15659–15665.

See, D., Mason, S., Roshan, R., 2002. Increased tumor necrosis factor alpha(TNF-alpha) and natural killer cell (NK) function using an integrativeapproach in late stage cancers. Immunol. Invest. 31, 137–153.

Shattil, S.J., Kashiwagi, H., Pampori, N., 1998. Integrin signaling: the plateletparadigm. Blood 91, 2645–2657.

Shen, Y.C., Chen, C.F., Chiou, W.F., 2000. Suppression of rat neutrophilreactive oxygen species production and adhesion by the diterpenoid lactoneandrographolide. Planta Med. 66, 314–317.

Shen, Y.C., Chen, C.F., Chiou, W.F., 2002. Andrographolide prevents oxygenradical production by human neutrophils: possible mechanism(s) involved inits antiinflammatory effect. Br. J. Pharmacol. 135, 399–406.

Tsai, H.R., Yang, L.M., Tsai, W.J., Chiou, W.F., 2004. Andrographolide actsthrough inhibition of ERK1/2 and Akt phosphorylation to suppresschemotactic migration. Eur. J. Pharmacol. 498, 45–52.

Zhang, C.Y., Tan, B.K., 1996. Hypotensive activity of aqueous extract ofAndrographis paniculata in rats. Clin. Exp. Pharmacol. Physiol. 23,675–678.

Zhang, X.F., Tan, B.K., 2000. Antihyperglycaemic and antioxidant properties ofAndrographis paniculata in normal and diabetic rats. Clin. Exp. Pharmacol.Physiol. 27, 358–363.

Zhang, Y.Z., Tang, J.Z., Zhang, Y.J., 1994. Study of Andrographis paniculataextracts on antiplatelet aggregation and release reaction and its mechanism.Zhongguo Zhong Xi Yi Jie He Za Zhi 14, 28–30, 34, 5.

Zhang, C., Kuroyangi, M., Tan, B.K., 1998. Cardiovascular activity of 14-deoxy-11,12-didehydroandrographolide in the anaesthetised rat and isolatedright atria. Pharmacol. Res. 38, 413–417.

Zhao, H.Y., Fang, W.Y., 1991. Antithrombotic effects of Andrographispaniculata Nees in preventing myocardial infarction. Chin. Med. J. 104,770–775.