Antioxidant activities of aqueous leaf extracts of Toona sinensis on free radical-induced endothelial cell

Other uses, including reproduction and distribution, or selling or

licensing copies, or posting to personal, institutional or third party

websites are prohibited.

In most cases authors are permitted to post their version of the

article (e.g. in Word or Tex form) to their personal website or

institutional repository. Authors requiring further information

regarding Elsevier’s archiving and manuscript policies are

encouraged to visit:

ContentslistsavailableatScienceDirect

Journal

of

Ethnopharmacology

j o u r n al hom ep a g e :w w w . e l s e v i e r . c o m / l o c a t e / j e t h p h a r m

Antioxidant

activities

of

aqueous

leaf

extracts

of

Toona

sinensis

on

free

radical-induced

endothelial

cell

damage

Hsin-Ling

Yang

_,

_Ssu-Ching

_Chen

_,

_Kai-Yuan

_Lin

_,

_Mei-Tsun

_Wang

_,

_Yu-Chang

_Chen

_,

_Hui-Chi

_Huang

_,

Hsin-Ju

Cho

_,

_Lai

_Wang

_,

_K.J.

_Senthil

_Kumar

_,

_You-Cheng

_Hseu

a_{InstituteofNutrition,ChinaMedicalUniversity,Taichung,Taiwan} b_{DepartmentofLifeSciences,NationalCentralUniversity,Chung-Li,Taiwan} c_{DepartmentofMedicalResearch,Chi-MeiMedicalCenter,Tainan,Taiwan}

d_{SchoolofChinesePharmaceuticalSciencesandChineseMedicineResources,CollegeofPharmacy,ChinaMedicalUniversity,Taichung,Taiwan} e_{DepartmentofCosmeceutics,CollegeofPharmacy,ChinaMedicalUniversity,Taichung,Taiwan}

a

r

t

i

c

l

e

i

n

f

o

Articlehistory:

Received5November2010 Receivedinrevisedform11May2011 Accepted12June2011

Available online 28 June 2011 Keywords: Toonasinensis Gallicacid Endothelialcells AAPH Antioxidants

a

b

s

t

r

a

c

t

Ethnopharmacologicalrelavence:InTaiwan,Toonasinensis(Toonasinensis)iswellknownasatraditional Chinesemedicine,whiletheunderlyingpharmacologicalmechanismsofthisdrugarestillamatterof debate.

Materialsandmethods:Thepurposeofthisstudywastoevaluatetheprotectiveeffectsofnon-cytotoxic concentrationsofaqueousleafextractsofToonasinensis(TSextracts;50–100␮g/mL)andgallicacid (5␮g/mL),amajorcomponentoftheseextracts,againstAAPH-inducedoxidativecelldamageinhuman umbilicalveinendothelialcells(ECs).

Results:ExposureofECstoAAPH(15mM)decreasedcellviabilityfrom100%to43%.However,ECswere pre-incubatedwithTSextractspriortoAAPHinductionresultedinincreasedresistancetooxidativestress andcellviabilityinadose-dependentmanner.AnincreaseinECs-derivedPGI2andIL-1␤ inresponseto

AAPHexposurewaspositivelycorrelatedwithcytotoxicityandnegativelywithTSextracts concentra-tions.Inaddition,gallicacidalsosuppressedPGI2andIL-1␤ productioninAAPH-inducedECs.Notably,TS

extracts/gallicacidtreatmentsignificantlyinhibitedROSgeneration,MDAformation,SOD/catalase activ-ity,andBax/Bcl-2dysregulationinAAPH-stimulatedECs.PretreatmentofECswithTSextracts/gallicacid alsosuppressedAAPH-inducedcellsurfaceexpressionandsecretionofVCAM-1,ICAM-1andE-selectin, whichwasassociatedwithabridgedadhesionofU937leukocytestoECs.Moreover,TSextracts/gallic acidtreatmentsignificantlyinhibitedtheAAPH-mediatedupregulationofPAI-1anddownregulationof t-PAinECs,whichmaydecreasefibrinolyticactivity.

Conclusions:Therefore,Toonasinensismaypossessantioxidantpropertiesthatprotectendothelialcells fromoxidativestress.OurresultsalsosupportthetraditionaluseofToonasinensisinthetreatmentof freeradical-relateddiseasesandatherosclerosis.

© 2011 Elsevier Ireland Ltd. All rights reserved.

1. Introduction

ToonasinensisRoem.(Meliaceae;Toonasinensis) is atype of arborthatiswidelydistributedinAsia.Ithaslongbeenusedas a traditional Chinesemedicine for a wide varietyof conditions in Chinesesociety and is very popular in vegetarian cuisinein

Abbreviations: T. sinensis, Toonasinensis;TSextracts, aqueousextractsof Toonasinensis;AAPH,2,2_{-azo-bis(2-amidinopropane)hydrochloride;Cometassay,} single-cellgelelectrophoresisassay;PGI2,prostacyclin;IL-1␤,interleukinIL-1␤; ROS,reactiveoxygenspecies;MDA,malondialdehyde;SOD,superoxidedismutase; VCAM-1,vascularcelladhesionmolecule1;ICAM-1,intercellularadhesionmolecule 1;PAI-1,plasminogenactivatorinhibitor-1;t-PA,tissueplasminogenactivator.

∗ Correspondingauthor.Tel.:+886422053366x5308;fax:+88642207808. E-mailaddress:[email protected](Y.-C.Hseu).

Taiwan.Theedibleleaveshavebeenusedasanorientalmedicine for treating rheumatoid arthritis, cervicitis,urethritis, tympani-tis,gastriculcers,enteritis,dysentery,itchiness,andcancer(Xien, 1996).Notably,peoplesfromSanxiaregioninChina,T.sinesiswere usedfortraditionalmedicinepreparationforcuring cerebrovas-cularandcardiovasculardiseases(LiandChen,2009).Whilethe underlyingpharmacological mechanismsofthis drugare still a matterofdebate,variousbiologicalactivitiesofToonasinensisleaf extractshavebeenreported,includinganti-cancer(Changetal.,

2002,2006;Yangetal.,2006),anti-diabetes(Hsuetal.,2003),and

antioxidant(Hseuetal.,2008b)effects,aswellasinhibiting Ley-digcellsteroidogenesis(Poonetal.,2005)andsuppressingbrain degenerationin senescence-acceleratedmice(Liao etal., 2006). Moreover,thesafetylevelsandnontoxiccharacteristicsofaqueous extractsofToonasinensiswereevaluatedusingabacterialreverse 0378-8741/$–seefrontmatter © 2011 Elsevier Ireland Ltd. All rights reserved.

mutationassay(Amestest),andbothacuteandsubacutetoxicity studieswereconductedinmice(Liaoetal.,2006).

Atherosclerosisisacomplicatedinflammatoryprocessthatcan leadtovascularendothelialdysfunction.Theearlypathogenesis of atherosclerosis appears to be activation of theendothelium in response to a variety of relevant stimuli including oxidized lipids,cytokines,orturbulentflow(Kimetal.,2007).The vascu-larendotheliumplaysacriticalroleinthepreservationofnormal vesselwallstructureandfunction.Vascularendothelialcells(ECs) controlvascularpermeability,vesseltone,coagulation,fibrinolysis andinflammatoryresponses.ChangesinECmembranefunction inducedbyreactiveoxygenspecies(ROS)appeartoplayakeyrole inthepathogenesisofatherosclerosis(LiandShah,2004). Prostacy-clin(PGI2)isproducedbyECsandpossessesanti-aggregatoryand -vasodilatorypropertiesanddecreasingpolymorphonuclear

adhe-sioninECs(Hseuetal.,2008a).Ithasbeenreportedthat,inresponse

tooxidativestress,ECscanproduceandreleaseeicosanoidslike

PGI2(CohenandTong,2010).ProductionofPGI2byECsisdirectly

proportionaltotheseverityofoxidativestressandcelldamage, whichisimportantforthemaintenanceofvascularhomoeostasis

(EganandFitzGerald,2006).Interleukin(IL)-1hasalsobeen

impli-catedinthepathogenesisofatherosclerosis(Kleemannetal.,2008). ECproduceIL-1inresponsetooxidativestress,whichstimulates ECreleaseofchemotacticfactorsandincreasestheexpressionof cellsurfaceadhesionmoleculesimportantforneutrophil, mono-cyteandlymphocyteadhesion.IL-1productionalsoincreasedin parallelwithoxidativestress-inducedcelldamage(Kleemannet

al.,2008).

Localizedaccumulationofmonocytes/macrophagesandT lym-phocytesinthearterialintimaappearstoplayakeyroleinearly atherogenesis,aswellasinplaqueruptureinadvanced atheroscle-roticlesions(Liuzzoetal.,2005).Molecularmechanismsmediating therecruitmentof monocytes/macrophages and T lymphocytes intoatheroscleroticlesionsmaydependonmultipleandcomplex processes;however,adhesionof monocytesandT lymphocytes tothevascularendotheliumisamong theearliest andessential processesduringatherogenesis, aswellasduring inflammatory responses(Liuzzoetal.,2005).Thisprocessappearstobemediated byendothelial–leukocyte adhesion molecules expressedonthe surfaceofthevascularendotheliumcoveringatheroscleroticand inflammatorylesions(Blankenbergetal.,2003).Theseadhesion moleculesinclude vascular celladhesion molecule-1(VCAM-1), intercellularadhesion molecule-1(ICAM-1),and E-selectin, and theirexpressioncanbetranscriptionallyinducedbyROSor inflam-matorycytokines(Zapolska-DownarandNaruszewicz,2009).

Free-radical-mediatedcell injurymaybecriticallyimportant invariouspathologicalphenomena(Piaoetal.,2004).Therefore, antioxidantsthatpreventdamagecausedbyfreeradicalsareworth ofadditionalinvestigation.Thegoalofthisinvestigationwasto exploretheantioxidantpotentialoftheaqueousleafextractsof Toonasinensis(TSextracts)andgallicacid(3,4,5-trihydroxybenoic acid),withnaturalpurifiedphenoliccomponentofTSextracts,in termsofprotectinghumanumbilicalveinendothelialcellsfrom oxidativedamageafterexposuretofreeradicalinitiatorAAPH.The AAPH-inducedDNAdamagewasassayedbycometassay.ELISAwas performedtoquantifyPGI2,IL-1␤,MDAandSOD/catalaseactivity inAAPH-challangedECs.Bax,Bcl-2,ICAM-1,VCAM-1,E-selectin, PAI-1andt-PAwasdeterminedbyWesternblottinganalysisusing appropriateantibodies.ROS and U937adhesion wasmonitored using fluorescence based assay systems. In the present study, 2,2-azobis-2-amidinopropanedihydrochloride (AAPH),a water-soluble free radical generator,was used to stimulate oxidative stressinvitro,andtheperoxylradicalsweregeneratedbythermal decompositionofanazocompoundinoxygen,eventuallycausing theoxidationoflipid,proteinandDNAinbiomolecules(Mayoetal.,

2003).

2. Materialsandmethods

2.1. Chemicals

Fetal bovine serum (FBS), M-199 medium, glutamine, and penicillin–streptomycin–neomycin (PSN) were obtained from GIBCOLaboratories(GrandIsland,NY,USA).AAPH[2,2 -azo-bis(2-amidinopropane)hydrochloride] wasobtainedfromWakepure chemical (Osaka,Japan). PGI2 immunoassay wasobtainedfrom AmershamPharmaciaBiotech(Piscataway,NJ,USA).IL-1␤,VCAM, ICAM,andE-selectinimmunoassaywasobtainedfromR&D Sys-tems(Minneapolis,MN,USA).SODactivitykitwasobtainedfrom RandoxLaboratoriesLtd.(Crumlin,UK).Catalaseactivitykitwas obtained from Calbiochem (La Jolla, CA, USA). t-PA and PAI-1 immunoassaywasobtainedfromAmericandiagostica(Stamford, CT,USA).Anti-_␤-actin,anti-Bax,anti-Bcl-2,anti-ICAM,anti-VCAM, anti-E-selectin,anti-tPA and anti-PAI-1antibody wereobtained from Santa Cruz Biotechnology Inc. (Heidelberg, Germany). All otherchemicalswereofthehighestgradecommerciallyavailable andsuppliedeitherbyMerck(Darmstadt,Germany)orSigma. 2.2. Toonasinensispreparationandextraction

TheaqueousextractsofToonasinensiswerepreparedbyadding 1000mLwaterto1000gfreshToonasinensisleavesandboiling untilitwasreducedto100mL,aspreviouslydescribed(Hsuetal., 2003).Thecrudeextractswerecentrifugedat860×gfor12min andthesupernatantwasusedforthisstudy.Thecrudeextracts (50g) wereconcentratedin avacuumandfreeze driedtoform powder;thestock(2mg/mLindoubledistilledwater)was subse-quentlystoredat−20◦_{Cuntilanalyzedforantioxidantproperties.} ThecrudeTSextractsseparatedfromfreshToonasinensisleaves hada yieldof5%,whichwasbasedontheinitialweightofthe crudeextracts(Yangetal.,2006).

2.3. IsolationofgallicacidfromTSextracts

The TS extracts were dissolved in a mobile phase consist-ing of methanol–water (50:50, v/v) before high performance liquidchromatography(HPLC)analysisandseparation.Eight com-pounds (gallic acid, methyl gallate, ethyl gallate, kaempferol, kaempferol-3-O-␤-d-glucoside,quercetin,quercitrin, quercetin-3-O-_{␤-d-glucoside}andrutin)wereisolatedfromtheTSextracts, aspreviouslydescribed(Yangetal.,2006).Theidentityofthe com-poundswasfullycharacterizedbycomparisonoftheirspectraldata (IR,NMRandmass)withtheanalogousinformationreportedinthe literature(Yangetal.,2006;Hsuetal.,2003).Gallicacid,amajor componentofleafextractsofToonasinensiswascollectedforuse inthisstudyatayieldof6%(Yangetal.,2006;Chiaetal.,2010). Thestock(5mg/mLinDMSO)wassubsequentlystoredat−20◦_C untilanalyzedforantioxidantproperties.

2.4. Cellculture

Humanumbilicalveinendothelialcells(HUVEC)wereprepared fromhumanumbilicalveinsessentiallyasdescribed(Hseuetal.,

2008a).Inbrief,theumbilicalcordwasinfusedwith0.02%

col-lagenasesolutioncontainingtheECs andflushedfromthecord byperfusionwithcordbufferandthencentrifuged.Theresulting cellsuspensionwasdividedequallybetweenseveral10-cmpetri dishes,andgrowntoconfluenceinM-199mediumand supple-mentedwith20%FBSandPNSat37◦Cin5%CO2.Uponconfluence, theprimaryculturecellsweredetachedwithtrypsin–EDTAand subculturedintissueculturewellsat37◦C.Allexperimentswere carriedoutusingcellsatleast4daysafterasinglepassageonly.The cellswereremovedfromeachwellwith0.4%trypsinandcounted

usingahemocytometer.Cellviability(4×105_{cells/12wells)were} checkedaftertreatmentusingtrypanblueexclusionandexamined usingphasecontrastmicroscopy.

Forallexperiments,ECswereincubatedwiththeindicated con-centrationofTSextracts(0,50,75or100␮g/mL)orwith5␮g/mL gallicacidfor1h.Thesupernatantwasremovedfollowing supple-mentation,theECswerewashedwithcordbufferandtheculture mediawasreplacedwithfreshmediacontaining15mMAAPH(in PBS,pH7.4).Thisreactionmixturewasshakengentlywhilebeing incubatedfor 14hat 37◦C; forthe ROSassay,cells were incu-batedfor2h.InordertoinducefreeradicalchainoxidationinECs, aqueousperoxylradicalsweregeneratedbythermal decomposi-tionofAAPH(anazocompound)inoxygen(Mayoetal.,2003). TheadvantagesofthismethodincludedthefactthatAAPH decom-posesthermallytogenerateradicalswithoutbiotransformations orenzymesandtherateofradicalgenerationcouldbeeasily con-trolledbyadjustingtheconcentrationoftheinitiator(Mayoetal.,

2003).

2.5. Cometassay(single-cellgelelectrophoresisassay)

TheassaywasessentiallythesameasthatdescribedbySing

et al. (1988). The ECs (8×105_{cells/6wells) were suspended in}

1%low-melting-pointagaroseinPBS(pH7.4)andpipettedonto superfrostedglassmicroscopeslidesprecoatedwithalayerof1% normal-melting-pointagarose(warmedat37◦Cpriortouse).The agarosewasallowedtosetat4◦Cfor10minbeforeimmersionof theslidesinlysissolution(2.5MNaCl,100mMEDTA,10mMTris [pH10],1%TritonX-100)at4◦Cfor1hinordertoremove cel-lularproteins.Slideswerethenplacedinsinglerowsina30cm widehorizontalelectrophoresistankcontaining0.3MNaOHand 1mMEDTA(pH13.4)at4◦Cfor40mintoallowseparationofthe twoDNAstrands(alkalineunwinding).Electrophoresiswas per-formedintheunwindingsolutionat30V(1V/cm)and300mAfor 30min.Theslideswerethenwashedthreetimesfor5mineach with0.4MTris(pH7.5)at4◦CbeforestainingwithPI(3␮g/mL).The PI-stainednucleoidswereexaminedunderafluroescence micro-scopeusinga510–550nmexcitationfilterat200×magnification. Thedamagewasnothomogeneousandvisualscoringofthe cel-lular DNA oneach slide was based oncharacterization of 100 randomlyselectednucleus/nucleoids.DNAdamageintheECs,as DNA strandbreaks includingdouble and single-strandvariants atalkali-labile sites, wasanalyzed in analkaline condition(pH 13.4).TailMoment:DNAmigrationfromthenucleusofeachcell wasmeasuredwithacomputerprogramusingthecometmoment parametersasfollows:cometmoment=0→n[(amountofDNA atdistanceX)×(distanceX)]/totalDNA.Theabove-described pro-ceduresarereferredtoasthestandardcometassay.Observation andanalysisoftheresultswerealwaysperformedbythesame experiencedperson.Theanalysiswasblinded,withtheobserver havingnoknowledgeofslideidentity.

2.6. Enzyme-linkedimmunosorbentassayforPGI2andIL-1ˇ metabolites

AftertheECs(4×105_{cells/well)in12wellplatewereincubated} withorwithoutTSextracts/gallicacidandAAPH,500␮Laliquotsof supernatantwerecollectedandquantifiedthePGI2concentration inculturemediabymeasuringthelevelofthestablemetaboliteof PGI2,6-keto-PGF1␣,usinganenzyme-linkedimmunosorbentassay (ELISA)kit(AmershamPharmaciaBiotech)accordingtothe manu-facturer’sprotocol.Meanwhile, concentrationsofIL-1␤ secreted intothe mediawere quantified usinga commerciallyavailable ELISAkit(R&DSystems).

2.7. MeasurementofROSgeneration

ProductionofintracellularROSwasdetectedbyfluorescence microscopy and flow cytometry using 2,7 -dihydrofluorescein-diacetate(DCFH-DA).ECs(2×105_{cells/24wells)wereculturedin} M-199 medium,withrenewaloftheculturemediumwhenthe cellsreached80%confluence.Sampleswerethenincubatedwith 10␮MDCFH-DA inculturemedium at37◦C for30min.During loading,theacetategroupsonDCFH-DAwereremovedby intracel-lularesterase,trappingtheprobeinsidetheECs.Afterloading,cells werewashedwithwarmPBS.ProductionofintracellularROScan bemeasuredbychangesinfluorescenceduetointracellular pro-ductionofDCFcausedbyoxidationofDCFH.IntracellularROS,as indicatedbydichlorofluorescein(DCF)fluorescence,wasmeasured withafluorescencemicroscope(Olympus1×71).

2.8. Assayforformationofthiobarbituricacidreactivesubstances (TBARS)

Toestimatethegenerationofmalondialdehyde(MDA) equiva-lentsduringECoxidation,theTBARSassaywasperformed(Nagai et al.,2011).Aftertreatment,ECs(4× 105_{cells/wellin 12 well)} weremixedwith1.5mLof0.67%thiobarbituric(TBA)and1.5mL of20%trichloroaceticacidandheatedat100◦Cfor30min,then thetemperatureofthereactionproductwasmaintainedat25◦C for30min,followedbytheincubationsampleswerecentrifugedat 860×gfor15minat4◦C.Thereactionproductwasassayed fluoro-metricallyusinganF2000opticalspectrofluorometer(HitachiLtd., Tokyo,Japan),withexcitationat515nmandemissionat553nm. Freshlydilutedtetra-ethoxypropane,which producesMDA,was usedasastandard.ResultsareexpressedaspmolofMDA/105_cells. 2.9. MeasurementofSODactivityinECs

This procedure is a variation of the classical NBT method

(Spitz and Oberley, 1989). ECs were seeded on a 12-well

plate with 4×105_{cells/well. ECs suspension (100␮L) was} cen-trifuged at 860×g for 2min, the supernatant was discarded and 35␮L ice water was added for cell disruption. SOD kit (Randox laboratories) wasused to measure SOD activity,with xanthineandxanthineoxidaseemployedtogeneratesuperoxide radicals that react with 2-(4-iodophenyl)-3-(4-nitrophenol)-5-phenyltetrazoliumchloride (I.N.T.)toforma redformazan dye. SODactivityismeasuredbythedegreeofreactioninhibition.The absorbanceofsampleswasmeasuredat505nmagainstareagent blank.Thecalculationwasperformedaccordingtothe manufac-turer’sinstructions;SODactivityisexpressedasunit/mgprotein. 2.10. Catalaseactivity

Toevaluatecatalaseactivity,4×105_{cells/wellwereseededon} a 12-well plate and washed twicewithPBS, homogenized and immunoprecipitatedwithan anti-catalaseantibody (Iwaiet al., 2003).CatalaseactivitywasmeasuredusingaCatalaseAssayKit (Calbiochem)followingthemanufacturer’sprotocol.Briefly, cata-laseproteinsampleswereincubatedinthepresenceofaknown concentrationofH2O2.After1minincubation, thereactionwas quenchedwithsodiumazide.Catalaseactivitywasdeterminedby aspectrophotometricassay(520nm)thatmeasurestherateof dis-mutationofH2O2.

2.11. U937adhesionassay

U937cellswerelabeledwith10␮g/mLBCECF-AMfor30min at 37◦C, washed and resuspended in serum-free media. ECs (2×105_{cells/wells)wereculturedin24wellplateandincubated}

with reagents prior to being co-cultured with 1×105_cells/mL BCECF-AM-labeledU937cellsfor 30minat37◦C.Non-adhering U937cells wereremoved by gentleaspiration,and wells were washedwithPBS.Cellswerelysedusing0.1%TritonX-100in0.1M Tris–HCl,pH7.4,toevaluateU937adhesiontoECs.Fluorescence wasmeasuredusingamicroplatefluorescencereaderwith excita-tionat510nmandemissionat531nm.

2.12. Westernblotanalysis

ECs (4×106_{cells/100-mm dish) were detached and washed} oncein cold PBS and suspended in 100␮L lysis buffer(10mM Tris–HCl [pH 8],0.32M sucrose, 1% Triton X-100, 5mM EDTA, 2mMDTT,and1mMphenylmethylsulfonyflouride).The suspen-sionwasputonicefor20minandthencentrifugedat16,000×gfor 20minat4◦C.TotalproteincontentwasdeterminedusingBio-Rad proteinassayreagent, withbovine serumalbuminas the stan-dard,proteinextractswerereconstitutedinsamplebuffer(0.062M Tris–HCl,2%SDS,10%glycerol,and5%␤-mercaptoethanol),and themixturewasboiledfor 5min.Equalamounts(20_␮g)ofthe denaturedproteinswereloadedintoeachlane,separatedon8% SDSpolyacrylamidegel,followed bytransfer oftheproteinsto PVDFmembranesinovernight.Membraneswereblockedwith0.1% Tween-20inTris-bufferedsalinecontaining5%non-fatdrymilk for20minatroomtemperature,andthemembraneswerereacted withprimaryantibodiesfor2h.Theywerethenincubatedwitha horseradishperoxidase-conjugatedgoatanti-rabbitoranti-mouse antibodyfor 2h beforebeing developed using theSuperSignal ULTRAchemiluminescence substratefrom PierceBiotechnology (Rockford,IL,USA).Bandintensitieswerequantifiedby densito-metrywithabsorbanceofthemixtureat540nmdeterminedusing anELISAplatereader.

2.13. AssessmentofVCAM-1,ICAM-1,E-selectin,PAI-1,andt-PA secretion

ECswereseededona6-wellplateusing8×105_{cells/well,and} thecellswerepretreatedwithTSextractsorgallicacid,followed bytheadditionofAAPH.Cellculturesupernatantwascollectedfor evaluationofICAM-1,VCAM-1andE-selectinusingcommercially availablekits(R&Dsystems).Additionally,thelevel oft-PAand ofPAI-1wasdeterminedinthesupernatantofECsculturesusing commerciallyavailableELISAkits(Americandiagostica)followed bythemanufacturer’sinstructions.

2.14. Statistics

Data is presented as mean±S.D. of at least three indepen-dentexperiments(n=3).OnewayANOVAfollowedbyDunnett’s testwereperformedtodeterminestatisticaldifferencesbetween groups.Statisticalsignificancewasdefinedasp<0.05foralltests.

3. Results

In this study,a humanumbilicalveinendothelialcells (ECs) culturesystem wasused to evaluatethe antioxidant effectsof theaqueousleafextractsofToonasinensis(TSextracts)andgallic acid,thenaturalphenoliccomponentpurifiedfromTSextracts,in responsetooxidativestressinducedbythefree-radicalgenerator, AAPH.AAPHwasusedtostimulateinvitroconditionsofoxidative stressandtheperoxylradicalsweregeneratedbythermal decom-positionofanazocompoundinoxygen(Mayoetal.,2003).Inthe presenceofoxygen,theseradicalscanattackmembrane polyunsat-uratedfattyacidsandinitiatelipidperoxidationchainreactions. Aspreviouslyreported,theconcentrationofTSextractsand

gal-licacidusedinthesestudieswasnotcytotoxictoECs(Hseuetal., 2011).Therefore,forallsubsequentexperiments,weemployedthe non-cytotoxicconcentrationofTSextracts(i.e.,≤100␮g/mL)and ofgallicacid(5␮g/mL),andfocusedontheeffectofTSextractsor gallicacidonAAPH-inducedECsdamage.

3.1. EffectofTSextractsandgallicacidonAAPH-induced cytotoxicityinECs

Inthisstudy,ECsexposedto15mMAAPHinM-199medium for14hdisplayeddecreasedcellviabilityasassessedbytrypan blueexclusion(Fig.1A).ECstreatedwith50–100␮g/mLTSextracts and5␮g/mL gallicacidsignificantlyrestoredcellviabilitywhen comparedwithAAPHalonetreatment(Fig.1A)(p<0.05).Although morphologicalchanges(cellshrinkage)detectedbyphase-contrast microscopywereevidentintheAAPH-exposedcelldamage. How-ever,theextentofthesechangesandcytotoxicityattributableto AAPHexposurewerereduced withincreasingTS extracts/gallic acidpre-treatment(Fig.1A).Theseresultsclearlyindicatethatthe exposureofECstoTSextracts/gallicacidconfersasignificant pro-tectiveeffectagainstAAPH-inducedoxidativedamage.

3.2. EffectsofTSextractsandgallicacidonDNAdamage

To determine whether TS extracts or gallic acid modulates AAPH-inducedDNAdamageinECswasexamined.Theformation ofDNAstrandbreaks(DNAdamage)inthecellularDNAwas evalu-atedusingsingle-cellgelelectrophoresis(cometassay).Asshown in Fig.1B, treatmentwithAAPH significantlyinduced ECsDNA damageafter14htreatment;however,additionof50–100␮g/mL TSextractsor5␮g/mLgallicacidsignificantly(p<0.05)decreased AAPH-inducedDNAdamageinECs.

3.3. EffectsofTSextractsandgallicacidonAAPH-inducedPGI2 andIL-1ˇproductioninECs

It has been reportedthat theproduction of PGI2 and IL-1␤ isdirectlyproportionaltotheextentofAAPH-inducedoxidative stress and cell damage (Mayo et al., 2003).Bear this in mind; furtherwemonitoredAAPH-inducedPGI2 andIL-1␤production in cultured ECs. Results obtainedfrom this study showedthat ECsincubatedwithAAPHalone(15mM)dramaticallyincreased PGI2 andIL-1␤productioncompared withcontrolcells (Fig.2A and B). AAPH-induced PGI2 and IL-1␤ production was signif-icantly (p<0.05) inhibited by TS extracts and gallic acid in a dose-dependentmannerwithanIC50 valueof73 and56␮g/mL of TS extracts, respectively (Fig. 2A and B). Furthermore, cells werepre-incubatedwith100␮g/mLof TSextractsand5␮g/mL gallic acid alone, the PGI2 or IL-1␤ levels were maintained at a backgroundlevel similarto the untreated samples (data not shown). These findings provided evidence that AAPH-mediated oxidativestressandresultinginjuryinECswassignificantly dimin-ishedbyTSextracts/gallicacidtreatment.Ourdatademonstrating that increased PGI2 and IL-1␤ production by ECs in response toAAPHexposurewaspositivelyandnegativelycorrelatedwith cell damage and TS extracts/gallic acid concentration, respec-tively.

3.4. EffectsofTSextractsandgallicacidonAAPH-inducedROS productionandMDAformationinECs

ROScancausedeleteriousoxidativedamagetocellular compo-nentsandalossofECsfunction(LiandShah,2004).Toinvestigate theparticipationof oxidativestressin AAPH-treatedECs, intra-cellularROSgenerationwasmonitored.Fluorescencemicroscopic analysesusing DCFH-DA as a fluorescenceprobe demonstrated

Fig.1.TSextractsandgallicacidinhibitedAAPH-inducedcytotoxicity(A)andDNAdamage(B)inECs.CellsweretreatedwithTSextracts(50–100␮g/mL)and/orgallicacid (5␮g/mL)intheabsenceorpresenceofAAPH(15mM)for14h.(A)Viablecellnumberincultureswasmonitoredbycountingcellsuspensionsusingahemocytometer(200× magnification).(B)DNAdamagewasassessedusinganalkalinecometassayasreportedinSection2.TailMomentofeachcellwasmeasuredwithacomputerprogramusing thecometmomentparameters.Resultsarethemean±SDofthreeassays;*indicatessignificantdifferences(p<0.05)relativetoAAPHgroups.TS:TSextracts;GA:gallicacid.

that incubation of ECs with 15mM AAPH caused a significant increaseinfluorescence2haftertreatment.However,the AAPH-induced increase of ROS generation was significantly (p<0.05) inhibitedbyexposuretoTSextracts(50–100_␮g/mL)and/orgallic acid(5␮g/mL)inaconcentration-dependentmanner(Fig.3A).

Oxidativestressisassociatedwiththeperoxidationofcellular lipids,whichcanbedeterminedbymeasuringMDA.Asshownin

Fig.3B,MDAlevelintheculturemediawassignificantlyincreased withresponsetoAAPH-exposure;whencellswereincubatedwith TS extracts (50–100␮g/mL) and gallic acid (5␮g/mL), the ele-vatedMDAformationwassignificantlydecreasedontobasallevel (p<0.05).Thesefindings providepositiveevidencethatboth TS extractsand gallic acidtreatmentsignificantly inhibited

AAPH-inducedoxidative stressin ECsbythereduction ofintracellular ROSaccumulationandMDAelevation.

3.5. EffectsofTSextractsandgallicacidonSODandcatalase activityinAAPH-inducedECs

Biological systems protect themselves against thedamaging effectsofactivatedROSbyseveralmeans,includingfreeradical scavengers,chainreactionterminatorsandenzymessuchasSOD andcatalase(GutteridgeandHalliwell,2006).Therefore,theeffects ofTSextractsandgallicacidonAAPH-inducedSODdepletionwere evaluatedinECs.AsshowninFig.4A,AAPHexposurereducedthe SODactivityinECsandthisreductionwassignificantly(p<0.05)

(A)

(B)

PGI

(pg/10

cells)

IL-1

(pg/10

cells)

Fig.2. InhibitoryeffectsofTSextractsandgallicacidonPGI2(A)andIL-1␤(B) productioninECsexposedtoAAPH.CellswereharvestedafterincubationwithTS extracts(50–100␮g/mL)andgallicacid(5␮g/mL)intheabsenceorpresenceof AAPH(15mM)for14h.PGI2andIL-1␤concentrationsweremeasuredintheculture mediausingacommercialELISAkit.Resultsarethemean±SDofthreeassays;* indicatessignificantdifferences(p<0.05)relativetoAAPHgroups.

augmentedbythetreatmentofTSextractsorgallicacidina dose-dependentmannerwithanIC50valueof25␮g/mL.Catalaseisa keycomponentoftheantioxidantdefensesystemandinhibition ofthis protective mechanism resultsin enhancedsensitivityto freeradicalinducedcellulardamage(Sampathkumaretal.,2005). AsshowninFig.4B,treatmentwithAAPHsignificantlydecreased catalaseactivityinECswithanIC50valueof35␮g/mL.However, pre-treatmentwithTSextracts(75or50–100␮g/mL)andgallic acid(5␮g/mL)increasedthecatalaseactivityinAAPH-inducedECs (p<0.05).ThepresentstudyrevealsthatthereisanincreaseofSOD andcatalaseactivitysuggestingthatTSextracts/gallicacidhavea protectiveeffectinresponsetoROS.Therefore,administrationofTS extracts/gallicacidincreasestheSODandcatalaseactivityin AAPH-inducedECdamage,thuspreventingtheaccumulationofexcessive freeradicalsandprotectingECsfromAAPH.

3.6. EffectsofTSextractsandgallicacidonAAPH-inducedBax andBcl-2proteinexpression

It hasbeen shown that theBcl-2 family of proteins hasan importantregulatoryroleinapoptosis,bothinactivation(Bax)and inhibition(Bcl-2)ofapoptosis;theBax/Bcl-2proteinratiois

rec-ognizedasakeyfactorintheregulationoftheapoptoticprocess (Yangetal.,2006).Therefore,westudiedtheeffectsofTSextracts (75␮g/mL)andgallicacid(5␮g/mL)onBcl-2andBaxprotein lev-elsinAAPH-treatedECs.CellsincubatedwithAAPHfor14hcaused anincreaseinBaxproteinandadecreaseinBcl-2protein,which wassignificantly(p<0.05)reversedbyTSextractsandgallicacid (Fig.5).AnalysisofourdataindicatesthatTSextractsorgallicacid maypreventthedysregulationofBax/Bcl-2,andtherebyleadto up-regulateanti-apoptoticsignalsinECs.

3.7. EffectsofTSextractsandgallicacidonAAPH-activatedU937 adhesion

ToexploretheeffectsofTSextractsandgallicacidonendothelial cell–leukocyteinteractions,weexaminedadhesionofU937cells, a monocytecellline,toAAPH-activated ECsunderstatic condi-tions. UnstimulatedconfluentECsexhibited minimalbindingto U937;however,U937adhesionwassubstantiallyincreasedwhen ECswereincubatedwithAAPHalone(Fig.6AandB).Concurrent incubationofECsconfluentwithTSextracts(50–100␮g/mL)and gallicacid(5␮g/mL)dose-dependentlyinhibitedU937adhesionto AAPH-activatedECs(p<0.05)(Fig.6AandB).

3.8. EffectsofTSextractsandgallicacidonAAPH-induced adhesionmoleculeexpression/secretioninECs

Toexamine theeffectof TSextract and gallicacidexposure onadhesionmoleculeexpression/secretion,ECsweretreatedwith theindicatedconcentrationofTSextracts(50,75and100␮g/mL) orgallicacid(5␮g/mL)inserum-freemedium.ECswereexposed to 15mM AAPH for 14h, and then evaluated ICAM-1, VCAM andE-selectinexpressionandsecretionbyWesternblottingand ELISAanalysis,respectively.ResultsrevealedthatAAPH-induced increasesintheexpressionandsecretionofICAM-1,VCAM-1and E-selectinwerereducedbyTSextracts(100_␮g/mL)andgallicacid (5␮g/mL)pre-treatmentinastatisticallysignificant(p<0.05) man-ner(Fig.7AandB).

3.9. EffectsofTSextractsandgallicacidonPAI-1andt-PA expression/secretionfromECs

Since thebalance betweenPAI-1 and t-PA determines fibri-nolyticactivity(MatsumotoandHorie,2011).Therefore,wealso investigatedtheeffectsofindicatedconcentrationofTSextracts (50,75and100␮g/mL)orgallicacid(5␮g/mL)ontheproduction ofPAI-1andt-PAfromECs.ECswereexposedto15mMAAPHfor 14h,andthenevaluatedPAI-1andt-PAexpressionandsecretion byWesternblottingandELISA,respectively.Resultsshowedthat incubationofECswithAAPHledtoincreasePAI-1level(Fig.8A). Notably,AAPHalsopromotedt-PAdepletioninECs(Fig.8B).TS extract and gallic acid treatment significantly inhibited AAPH-inducedincreaseinPAI-1andreduction int-PA(Fig.8AandB). Therefore,TSextracts/gallicacidtreatmentsignificantlyinhibited AAPH-induceddisruptionofthebalancebetweenPAI-1andt-PA, whichmaydecreaseECcoagulationandfibrinolysis.

4. Discussion

Thereis evidence that an injury toarterial endothelialcells reflectsone ofthekeyeventsintheinitiationand evolutionof atherosclerosis(Kimetal.,2007).Ithasbeenshownthat antiox-idantsprotectcellsagainstfreeradical-inducedoxidativedamage inseveralinvitromodelsincludingECs(Hseuetal.,2008a).Inour previousstudy,weconcludedthatTSextractsand gallicacid,a majorcomponentofTS extracts,possesses effectiveantioxidant activity,whichincludesscavengingoffreeandsuperoxideanion

Fig.3. TSextractsandgallicacidattenuatedAAPH-inducedROSgeneration(A)andMDAformation(B)inECs.(A)CellsweretreatedwithTSextracts(50–100␮g/mL)and gallicacid(5␮g/mL)intheabsenceorpresenceofAAPH(15mM)for2h.TheintracellularROSlevel,wasindicatedbyDCFfluorescence,wasmeasuredbyfluorescence microscopy(200×magnification).(B)MDAproductionwasmeasuredbytheTBARSassayasdescribedinSection2.Resultsarethemean±SDofthreeassays;*indicates significantdifferences(p<0.05)relativetoAAPHgroups.

radicals,and greaterreducingpowerand metalchelating

activ-ity(Hseuetal.,2008b).Further,supplementationwithTSextracts

andgallicacidappearstoreduceCuSO4,AAPH-andSNP(sodium nitroprusside)-inducedoxidativemodificationofLDLand AAPH-inducederythrocytehemolysis(Hseuetal.,2008b).Inthepresent work,wedemonstratethatbothTSextractsandgallicacidareable toinhibittheAAPH-inducedoxidativedamageinhuman umbili-calveinendothelialcells.ThefreeradicalsgeneratedfromAAPH reactwithoxygenmoleculesrapidlytoyieldperoxylradicals.The lipidperoxylradicals attackotherlipid moleculestoformlipid hydroperoxideandnew lipidradicals.This reactiontakesplace repeatedlyresultinginattacksonvariousbiologicalmoleculesand theproductionofphysiochemicalalterationsandcellulardamage

(Yokozawaetal.,2000).IthasbeenreportedthatAAPH

intoxica-tionresultsdeathofvascularendothelialcellsinvariousorgans andcausedatherosclerosis,ischemia-reoxygenerationinjuryand

inflammatorydiseases(Yokozawaetal.,2000).Therefore, AAPH-intoxicationexperimentmaybeapromisingassaysystemforthe quantitativestudyofcellularresponsetooxidativestressand bio-logicalactivitiesofantioxidants(Yokozawaetal.,2000).Basedon thesefindings,itseemsreasonabletosuggestthatTSextractsand gallicacidprotectsECsfromfreeradical-inducedoxidative dam-age,furtherreducingtheriskoffreeradical-relateddiseasesand atherosclerosis.

ROScantriggersignaltransductionpathwaysinECsthatleadto anincrease intheexpressionof cytokines,suchasPGI2 and IL-1␤,which areknowntoplay importantrolesinatherosclerotic pathogenesis(Mayoetal.,2003).Prostacyclin(PGI2)isproduced byECsand possessesanti-aggregatoryandvasodilatory proper-ties,decreasingpolymorphonuclearadhesiontoECsinvitro(Hseu

etal.,2008a).Underphysiologicalconditions,thesecretionofPGI2

TS ( g/mL)

SOD activity

(units/m

g protein

)

(A)

(B)

Catalase activity (units/mg protein)

0 2 4 6 8 * * * * GA ( g/mL) AAPH 50 75 100 5 *

Fig.4. InhibitoryeffectsofTSextractsandgallicacidonAAPH-inducedSOD(A)and catalase(B)activityinECs.CellswereharvestedafterincubationwithTSextracts (50–100␮g/mL)andgallicacid(5␮g/mL)intheabsenceorpresenceofAAPH (15mM)for14h.SODorcatalaseactivitywasdeterminedusinga spectrophoto-metricassay(520nmor505nm)asreportedinmaterialsandmethods.Resultsare themean±SDofthreeassays;‘*’indicatessignificantdifferences(p<0.05)relative toAAPHgroups.

responsetoROS/oxidativestress orinjury, PGI2 and IL-1␤ pro-ductionwere dramatically increases in ECs, thereby increasing adhesionofcirculatingmonocytesandlymphocytestothe acti-vatedECsatthesite ofinjury (Renieret al.,2003).Thus, IL-1␤ secretionby vascularcells appears tobeintimately involvedin monocyterecruitmentandadhesion,concomitantwiththe produc-tionofothercytokinesandgrowthfactors,whicharecrucialevents intheinitiationandprogressionoftheatheromatousplaque(Singh etal.,2002).Therefore,whenthecellswereenrichedwith antiox-idants, oxidative stress and resultant injury were significantly diminished.Incongruencewiththeselines,ourresultsalsoshowed thatAAPH-inducedPGI2 and IL-1␤production in ECswere sig-nificantlyinhibitedbyTSextracts/gallicacidinadose-dependent manner(Fig.2).TheseinhibitoryeffectsofTSextracts/gallicacid onPGI2andIL-1␤productionmaycontributetotheeffectofToona sinensisat reducingtherisk offree radical-relateddiseases and atherosclerosis.

ROS-inducedendothelialdysfunctionplaysanimportantrole intheinitiationanddevelopmentofvasculardisease(Hseuetal.,

2008a).Mount of scientific literaturesindicatethat antioxidant

administration improvesendothelial function (Siow and Mann,

Fig.5. EffectsofTSextractsandgallicacidonAAPH-inducedBcl-2andBax expres-sioninECs.CellswereharvestedafterincubationwithTSextracts(75␮g/mL)and gallicacid(5␮g/mL)intheabsenceorpresenceofAAPH(15mM)for14h.Protein (20␮g)fromeachsamplewasresolvedona8%SDS-PAGE,andwesternblotwas performed.Atypicalresultfromthreeindependentexperimentsisshown.Relative changesinproteinbandsweremeasuredusingdensitometricanalyseswiththe controlbeing100%asshownjustbelowthegeldata.Resultsarethemean±SDof threeassays;*indicatessignificantdifferences(p<0.05)relativetoAAPHgroups.

2010; Farbstein et al., 2010). ROS cancause deleterious

oxida-tivedamagetocellularcomponentsandalossofECsfunction(Li

andShah,2004).Notably,ROSproduceDNAlesionssuchasbase

modifications,single-strandbreaks,doublestrandbreaks,andthe cross-linkingofbasesinlivingcells(Valkoetal.,2004).DNAdamage inthecellscanberepaired,althoughtherepairmaybeerror-prone withlowfidelityleadingtomutations,genomicinstability,andcell deaththroughapoptoticmechanisms(Shen,2011).Therefore,the presenteddataprovidesconvincingevidencethatTSextractsand gallicacidcanprovideprotectionfromAAPH-inducedintracellular ROSandDNAdamageinECs,whichmayplayanimportantrole inthepreventionofendothelialdysfunctionand atherosclerosis invitro.

Lipidperoxidationinbiologicalsystemshaslongbeenthought ofasatoxicologicalphenomenonthatresultsinpathological conse-quences(Yokozawaetal.,2000).Ithasbeenwelldocumentedthat AAPHenhancedlipidperoxidationincellularsystems(Yokozawa etal.,2000).Consistencewiththeseviews,wemeasuredthelipid peroxidant, MDA, to determine the antioxidant function of TS extractsandgallicacidagainstAAPH-inducedcellulardamagein ECs.Resultsalsoreveledthat,pretreatmentwithTSextractsand gallicaciddecreasedAAPH-inducedMDAformationinECs.Thus TS extracts/gallicacidpreventingtheaccumulationof excessive freeradicals andprotectingtheECsfromAAPH-derivedcellular injury.

SOD and catalase is a key component of the antioxidant defense system.Inhibitionof this protectivemechanism results in enhancedsensitivity to free radical-inducedcellular damage

(Gutteridgeand Halliwell, 2006).The SOD converts superoxide

radicals (O2−)into H2O2 plus O2,thus participatingwithother antioxidant enzymes in the enzymatic defense against oxygen

Fig.6.TSextractsandgallicacidinhibitsAAPH-inducedU937celladhesiontoECs.ECswereharvestedafterincubationwithTSextracts(50–100␮g/mL)andgallicacid (5␮g/mL)intheabsenceorpresenceofAAPH(15mM)for14h.(A)AfterECswerewashed,BCECF-AM-labeledU937cellswereaddedtoeachwellandcultureswere incubatedforanadditionalhourat37◦_{C.(B)ThelevelofU937cellsboundtoECs(shownasapercentageofthecontrol),indicatedbyBCECF-AMfluorescence,wasmeasured} byfluorescencemicroscopy(200×magnification).Resultsarethemean±SDofthreeassays;*indicatessignificantdifferences(p<0.05)relativetoAAPHgroups. toxicity. Further, the reduction in the activity of catalase may,

therefore, result in a number of deleterious effectsdue to the accumulationofO2−andH2O2(Sampathkumaretal.,2005).Our presentdatademonstratesthattheAAPH-treatmentdramatically depleted endogenous SOD/catalase levels in ECs. Pretreatment with TS extracts/gallic acid significantly (p<0.05) augmented AAPH-depleted SOD/catalase levels in ECs. Thus, the potential elevation of SOD/catalase activity by Toona sinensis is accor-dance with the inhibition of AAPH-induced oxidative damage, thus,providingeffectiveprotectionfromoxidativemodificationof ECs.

Growingevidence demonstratesthat theBcl-2 family plays an important regulatory role in apoptosis, either as activators (Bax)orinhibitors(Bcl-2)(Yangetal.,2006).Endothelialapoptosis is a physiologicalprocess that contributes tovessel homeosta-sisbyeliminating damagedcells fromthevesselwalls(Harrod etal.,2006).Ifincreased,endothelialapoptosismayleadto dis-turbedendothelialfunction,whichmaypromoteatherogenesisand

thrombosis(StonemanandBennett,2004).Manyoftheagentsthat induceapoptosisareoxidantsorstimulatorsofcellularoxidative metabolism,whereas antioxidantshave beenshowntoprevent apoptosisinsomesystems(Qinetal.,2001).Recentstudiesreveled thatBcl-2proteinisalsoresponsibleforcellulardefenseagainst oxidative stress.In addition,activationof Bcl-2caninhibitlipid peroxidation. Thereforeit hasbeen supposed that Bcl-2act as free radicalscavenger(Murakoshi etal.,2003).Theseevidences indicatethattheanti-apoptoticBcl-2proteinassociatedwith mito-chondrialmembraneintegritybypreventingcytochromecrelease, caspase activation, Bax redistribution, and apoptosis (Er et al., 2006).Inthisstudy,wealsoobservedthatAAPH-treatment sup-pressedBcl-2andincreasedBaxproteinlevelsinECs.However, AAPH-inducedBax/Bcl-2dysregulationaresignificantlyreversed byTSextracts/gallicacidpretreatmentinECs.Theresultsconcluded thatTSextracts/gallicacidreducesapoptosisinECs,possiblyvia itabilitytoupregulateBcl-2anddownregulateBaxexpression. Therefore,Toonasinensismayhavevaluableantioxidantproperties.

Fig.7. TSextractsandgallicacidsuppressedAAPH-inducedVCAM-1,ICAM-1,andE-selectinexpression(A)andsecretion(B)inECs.Cellswereharvestedafterincubation withtheindicatedconcentrationofTSextractsandgallicacidintheabsenceorpresenceofAAPH(15mM)for14h.(A)Immunoblottingwasperformedagainstanti-VCAM-1, ICAM-1andE-selectin.Relativechangesinproteinbandsweremeasuredusingdensitometricanalysiswiththecontrolbeing100%asshownjustbelowthegeldata.(B) ThesupernatantswerecollectedforanalysesofVCAM-1,ICAM-1andE-selectinconcentrationusingELISA.Resultsarethemean±SDofthreeassays;*indicatessignificant differences(p<0.05)relativetoAAPHgroups.

Leukocyte adhesion to arterial endothelial cells is thought tobe an important step in thedevelopment of atherosclerosis

(Matsumotoetal.,2002).Itiswellknownthatadhesionmolecules

arestrongpredictorsofatheroscleroticlesiondevelopmentand fur-theronsetofcardiovascularevents(Blankenbergetal.,2003).These include both secreted moleculessuch asthe chemokine family ofchemoattractantcytokinesandsurface-expressedcelladhesion moleculesoftheselectinandimmunoglobulinfamilies(Kimetal., 2007).OurpresentdatademonstratesthatTSextractsandgallic aciddose-dependentlydecreasedtheAAPH-inducedadhesionof U937cellstotheendothelialcells.Furthermore,TSextractsand gallicacidsignificantlydiminishedbothexpressionandsecretion levelsofICAM-1,VCAM-1,andE-selectininAAPH-challengedECs. Thus,TSextractsandgallicacidcandownregulateVCAM-1, ICAM-1,andE-selectinassociatedwithreducedadhesionofleukocytes, whichareplayingmajorroleinthepreventionofatherosclerosis andinflammatoryresponses.

VascularECsalsoproducemanymodulatoryproteins,suchas PAI-1andt-PA,whichregulatecoagulationandfibrinolysis(Cheng et al., 2008). PAI-1, a 50kDa glycoprotein secreted by various cellsincludingendothelialcells,hepatocytes,platelets,andsmooth

musclecells(Binderetal.,2002).Increasedlocalexpressionof PAI-1isobservedinrestenosisandinatheroscleroticplaques(Binder et al., 2002). In vitro, PAI-1 plays a critical role in the regula-tion of fibrinolysis and serving as a primary inhibitor of t-PA, whichisresponsiblefortheintravascularplasminogenactivation

(Chengetal.,2008).Endogenoust-PAreleasefromthe

endothe-lium regulatesthedissolution ofintravascularthrombus,andis a critical determinantof cardiovascular outcome (Cheng et al., 2008).ThebalancebetweenPAI-1andt-PAisknowntocontrol thedevelopmentofthrombosis.Inthepresentstudy,weobserved thatAAPHtreatmentmodulatesbalancebetweenPAI-1andt-PA in ECsto becomeprocoagulative and hypofibrinolytic. Interest-ingly,TSextracts/gallicacidtreatmentsignificantlyinhibitedthe AAPH-inducedup-regulationofPAI-1anddown-regulationoft-PA expressioninECs,whichmayreduceECscoagulationand fibrinol-ysis.

Naturalproducts,includingplants,providerichresourcesfor food and drug discovery. In our previous study, a number of compounds, including gallic acid, methyl gallate, ethyl gallate, kaempferol,kaempferol-3-O-␤-d-glucoside,quercetin,quercitrin, quercetin-3-O-␤-d-glucoside and rutin, were isolated from the

Fig.8. InhibitoryeffectsofTSextractsandgallicacidonAAPH-inducedPAI-1(A)andt-PA(B)expressionandsecretioninECs.Cellswereharvestedafterincubationwith theindicatedconcentrationofTSextractsandgallicacidintheabsenceorpresenceofAAPHfor14h.Immunoblottingwasperformedagainstanti-VCAM-1,ICAM-1and E-selectin.Relativechangesinproteinbandsweremeasuredusingdensitometricanalysiswiththecontrolbeing100%asshownjustbelowthegeldata.Thesupernatants werecollected,andthecontentofPAI-1andt-PAwereanalyzedbyELISA.Resultsarethemean±SDofthreeassays;*indicatessignificantdifferences(p<0.05)relativeto AAPHgroups.

leavesofToonasinensis;identityofthecompoundswasdetermined byHPLCandbasedontheanalogousinformationreportedinthe literature(Hsuetal.,2003;Yangetal.,2006).Thetotalphenolic contentofTS extractswasestimated tobe130mg±26mg gal-licacid(pyrocatechol)equivalents/gofplantextracts(Yangetal., 2006).Theyieldofgallicacid,thenaturalphenoliccomponent puri-fiedfromTSextracts,wasabout6%(Yangetal.,2006).Although itremainsunclearwhichofthecomponentsofToonasinensisare activecompounds,inrecentyearsmuchattentionhasbeenfocused forpolyphenolsregardingtheirbiologicalactivates,especiallythe antioxidantactivity ofgallic acid(Owand Stupans,2003).This studyalsoagreementwithpreviousreportsshowingantioxidant propertiesofgallicacid(Hsiehetal.,2004;Hseuetal.,2008b). Gal-licacidiswidelydistributedinvariousplantsandfruits,suchas gallnuts,sumac,oak bark,green tea,applepeels,grapes, straw-berries,pineapples,bananas,lemonsandinredandwhitewine. Eventhoughthetherapeuticutilityofgallicacidinthisregardis unknown,itscommonoccurrenceinfruitsandfoodaswellasits smallmolecularweight(170Da)mightbeanadvantageinterms ofsafetyanddosingdesign.Theseresultsimplythatgallicacidis oneoftheactivecompoundsresponsiblefortheinhibitionoffree radical-inducedECs damagesinvitro. Furtherbioassay-directed fractionationsleadingtotheidentificationandpurificationofthe compoundsresponsiblefortheantioxidanteffectsarewarranted.

Inconclusion,thepresentstudyrevealed thatToonasinensis supplementationreduced AAPH-inducedoxidative modification inECs. Based onthesefindings, itseemsreasonable tosuggest thattheantioxidantpropertiesofToonasinensisprotectvascular endothelialcellsfromoxidativestress.Ourresultsalsocontribute towardsthevalidationofthetraditionaluseofToonasinensisin thetreatmentofcardiovasculardiseases.However,furtherinvivo experimentsarecriticallywarrantedtoconfirmthesepromising results.

Acknowledgements

Thisworkwassupportedbygrants NSC-99-2320-B-039-035-MY3,NSC-98-2320-B-039-037-MY3,CMU97-258,and CMU99-C-02fromtheNationalScienceCouncilandChinaMedicalUniversity, Taiwan.

References

Binder,B.R.,Christ,G.,Gruber,F.,Grubic,N.,Hufnagl,P.,Krebs,M.,Mihaly,J.,Prager, G.W.,2002.Plasminogenactivatorinhibitor.1.Physiologicaland pathophysio-logicalroles.NewsinPhysiologicalSciences17,56–61.

Blankenberg,S.,Barbaux,S.,Tiret,L.,2003.Adhesionmoleculesandatherosclerosis. Atherosclerosis170,191–203.

Chang,H.C.,Hung,W.C.,Huang,M.S.,Hsu,H.K.,2002.Extractfromtheleaves of Toonasinensis Roemor exerts potent antiproliferativeeffecton human lung cancer cells. The American Journal of Chinese Medicine 30, 307– 314.

Chang,H.L.,Hsu,H.K.,Su,J.H.,Wang,P.H.,Chun,Y.F.,Chia,Y.C.,Tsai,L.Y.,Wu,Y.C., Yuan,S.S.,2006.ThefractionatedToonasinensisleafextractinducesapoptosis ofhumanovariancancercellsandinhibitstumorgrowthinamurinexenograft model.GynecologicOncology102,309–314.

Cheng,M.,Li,Y.,Wu,J.,Nie,Y.,Li,L.,Liu,X.,Charoude,H.N.,Chen,H.,2008.IL-8 inducesimbalancesbetweennitricoxideandendothelin-1,andalsobetween plasminogenactivatorinhibitor-1andtissue-typeplasminogenactivatorin cul-turedendothelialcells.Cytokine41,9–15.

Chia,Y.C.,Rajbanshi,R.,Calhoun,C.,Chiu,R.H.,2010.Anti-neoplasticeffectsof gal-licacid,amajorcompoundofToonasinensisleafextractonoralsquamous carcinomacells.Molecules15,8377–8389.

Cohen,R.A.,Tong,X.Y.,2010.Vascularoxidativestress:thecommonlinkin hyper-tensiveanddiabeticvasculardisease.JournalofCardiovascularPharmacology 55,308–316.

Egan, K., FitzGerald, G., 2006. Eicosanoids and the vascular endothelium. The vascular endothelium. Handbook of ExperimentalPharmacology 176, 189–211.

Er,E.,Oliver,L.,Cartron,P.F.,Juin,P.,Manon,S.,Vallette,F.M.,2006.Mitochondria asthetargetofthepro-apoptoticproteinBax.BiochemicaetBiophysicaActa 1757,1301–1311.

Farbstein,D.,Kozak-Blickstein,A.,Levy,A.P.,2010.Antioxidantvitaminsandtheir useinpreventingcardiovasculardiesease.Molecules9,8098–8110.

Gutteridge,J.M.,Halliwell,B.,2006.FreeradicalsandantioxidantsintheYear2000: ahistoricallooktothefuture.AnnalsoftheNewYorkAcademyofSciences899, 136–147.

Harrod,C.G.,Batjer,H.H.,Bendok,B.R.,2006.Deficienciesinestrogen-mediated regulation ofcerebrovascularhomeostasismaycontributeto anincreased riskofcerebralaneurysmpathogenesisandruptureinmenopausaland post-menopausalwomen.MedicalHypotheses66,736–756.

Hseu,Y.C.,Chang,W.H.,Chen,C.S.,Liao,J.W.,Huang,C.J.,Lu,F.J.,Chia,Y.C.,Hsu, H.K.,Wu,J.J.,Yang,H.L.,2008b.AntioxidantactivitiesofToonasinensisleaves extractsusingdifferentantioxidantmodels.FoodandChemicalToxicology46, 105–114.

Hseu,Y.C.,Chen,S.C.,Yech,Y.J.,Wang,L.,Yang,H.L.,2008a.Antioxidantactivityof Antrodiacamphorataonfreeradical-inducedendothelialcelldamage.Journalof Ethnopharmacology118,237–245.

Hseu,Y.C.,Lin,W.H.,Chang,C.S.,Hung,D.Z.,Wang,M.T.,Cho,H.J.,Kuo,Y.H.,Chen, S.C.,Yang,H.L.,2011.Toonasinensis(leafextracts)inhibitvascularendothelial growthfactor(VEGF)-inducedangiogenesisinvascularendothelialcells.Journal ofEthnopharmacology134,111–121.

Hsieh, T.J., Liu, T.Z., Chia, Y.C., Chern, C.L., Lu, F.J., Chuang, M.C., Mau, S.Y., Chen,S.H., Syu,Y.H.,Chen,C.H., 2004.Protective effectof methylgallate fromToonasinensis(Meliaceae)againsthydrogenperoxide-inducedoxidative stress andDNAdamageinMDCKcells.FoodandChemicalToxicology42, 843–850.

Hsu,H.K.,Yang,Y.C.,Hwang,J.H.,Hong,S.J.,2003.EffectsofToonasinensisleaf extractonlipolysisindifferentiated3T3-L1adipocytes.TheKaohsiungJournal ofMedicalSciences19,3855–3890.

Iwai,K.,Kondo,T.,Watanabe,M.,Yabu,T.,Kitano,T.,Taguchi,Y.,Umehara,H., Taka-hashi,A.,Uchiyama,T.,Okazaki,T.,2003.Ceramideincreasesoxidativedamage duetoinhibitionofcaspase-3-dependentproteolysisinHL-60cellapoptosis. TheJournalofBiologicalChemistry278,9813–9822.

Kim,Y.S.,Ahn,Y.,Hong,M.H.,Kim,K.H.,Park,H.W.,Hong,Y.J.,Kim,J.H.,Kim, W.,Jeong,M.H.,Cho,J.G.,Park,J.C.,Kang,J.C.,2007.Rosuvastatinsuppresses theinflammatoryresponsesthroughinhibitionofc-JunN-terminalkinaseand NuclearFactor-␬Binendothelialcells.JournalofCardiovascularPharmacology 49,376–383.

Kleemann,R.,Zadelaar,S.,Kooistra,T.,2008.Cytokinesandatherosclerosis:a com-prehensivereviewofstudiesinmice.CardiovascularResearch79,360–376. Li,J.,Chen,C.,2009.Experimentalstudyonantithrombosisactivityofn-butanol

extract of Toonasinensis seeds. Journalof Sichuan ofTraditional Chinese Medicine127,26–28.

Li,J.M.,Shah,A.J.,2004.Endothelialcellsuperoxidegeneration:regulationand relevanceforcardiovascularpathophysiology.AmericanJournalofPhysiology. Regulatory,IntegrativeandComparativePhysiology287,1014–1030. Liao,J.W.,Hsu,C.K.,Wang,M.F.,Hsu,W.M.,Chan,Y.C.,2006.BeneficialeffectofToona

sinensisRoemoronimprovingcognitiveperformanceandbraindegenerationin senescence-acceleratedmice.TheBritishJournalofNutrition96,400–407. Liuzzo,G.,Giubilato,G.,Pinnelli,M.,2005.Tcellsandcytokinesinatherosclerosis.

Lupus14,732–735.

Matsumoto,K.,Sera,Y.,Nakamura,H.,Ueki,Y.,Miyake,S.,2002.Serum concentra-tionsofsolubleadhesionmoleculaesarerelatedtodegreeofhyperglycemiaand insulinresistanceinpatientswithtype-2diabetsmellitus.DiabetesResearch andClinicalpractice55,131–138.

Matsumoto,T.,Horie,M.,2011.Angiotensin-convertingenzymeinhibitionand fib-rinolyticbalance.HypertensionResearch34,448–449.

Mayo,J.C.,Tan,D.X.,Saing,R.M.,Lopez-Burilli,S.,Reiter,R.J.,2003.Oxidativedamage tocatalaseinducedbyperoxylradicals:functionalprotectionbymelatoninand otherandotherantioxidants.FreeRadicalResearch37,543–553.

Murakoshi,M.,Osamura,R.Y.,Watanabe,K.,2003.Immunocytochemical localiza-tionofglutathione-peroxidase(GSH-PO)andBcl-2intheratventralprostate. ActaHistochemicaetCytochemica36,335–343.

Nagai,M.,Tani,M.,Kishimoto,Y.,Iizuka,M.,Saita,E.,Toyozaki,M.,Kamiya,T., Ikeguchi,M.,Kondo,K.,2011.Sweetpotato(IpomeabatatasL.)leavessuppressed oxidationoflowdensitylipoprotein(LDL)invitroandhumansubjects.Journal ofClinicalBiochemistryandNutrition48,203–208.

Ow,Y.Y.,Stupans,I.,2003.Gallicacidandgallicacidderivatives:effectsondrug metabolizingenzymes.CurrentDrugMetabolism4,241–248.

Piao,X.L.,ParkII,H.,Baek,S.H.,Kim,H.Y.,Park,M.K.,Park,J.H.,2004. Antioxida-tiveactivityoffuranocoumarinsisolatedfromAngelicaedahuricae.Journalof Ethnopharmacology93,243–246.

Poon,S.L.,Leu,S.F.,Hsu,H.K.,Liu,M.Y.,Huang,B.M.,2005.Regulatorymechanism ofToonasinensisonmouseleydigcellsteroidogenesis.LifeSciences76,1473– 1487.

Qin,F.,Rounds,N.K.,Mao,W.,Kawai,K.,Liang,C.S.,2001.Antioxidantvitamins pre-ventcardiomyocyteapoptosisproducedbynorepinephrineinfusioninferrets. CardiovascularResearch51,736–748.

Renier,G.,Maputu,J.C.,Serri,O.,2003.Benefitsofgliclazideintheatherosclerotic process:decreaseinmonocyteadhesiontoendothelialcells.Metabolism52, 13–18.

Sampathkumar,R.,SivaKumar,T., SivaKumar,P., Netaji,R.,Vijayabaskar,M., Perumal,P.,Gupta,M.,Mazumdar,U.K.,2005.Hepatoprotectiveandinvivo antioxidanteffectsofCareyaarboreaagainstcarbontetrachlorideinducedliver damageinrats.InternationalJournalofMolecularMedicineandAdvance Sci-ences1,418–424.

Shen,Z.,2011.Genomicinstabilityandcancer:anintroduction.JournalofMolecular CellBiology3,1–3.

Sing,N.P.,McCoy,M.T.,Tice,R.R.,Schneider,E.L.,1988.Asimpletechniquefor quantitationoflowlevelsofDNAdamageinindividualcells.ExperimentalCell Research175,184–191.

Singh,R.B.,Mengi,S.A.,Xu,Y.J.,Arneja,A.S.,Dhalla,N.S.,2002.Pathogenesisof atherosclerosis:amultifactorialprocess.ExperimentalandClinical Pharmacol-ogy7,40–53.

Siow,R.C.,Mann,G.E.,2010.Dietaryisoflavonesandvascularprotection:activation ofcellularantioxidantdefensesbySERMsorhormesis.MolecularAspectsof Medicine31,468–477.

Spitz,D.R.,Oberley,L.W.,1989.Anassayforsuperoxidedismutaseactivityin mam-maliantissuehomogenates.AnalyticalBiochemistry179,8–18.

Stoneman,V.E.A.,Bennett,M.R.,2004.Roleofapoptosisinatherosclerosisandits therapeuticimplications.ClinicalScience107,343–354.

Valko,M.,Izakovic,M.,Mazur,M.,Rhodes,C.J.,Telser,J.,2004.Roleofoxygenradicals inDNAdamageandcancerincidence.MolecularandCellularBiochemistry266, 37–56.

Xien,W.C.,1996.AssemblyofChineseHerbsinMainland,vol.2.People’sMedical PublishingHouse,Beijing,pp.459–460.

Yang, H.L., Chang,W.H., Chia, Y.C., Huang, C.J.,Lu, F.J., Hsu, H.K., Hseu, Y.C., 2006.Toonasinensisextractsinducesapoptosisviareactiveoxygenspeciesin humanpremyelocyticleukemiacells.FoodandChemicalToxicology44,1978– 1988.

Yokozawa,T.,Cho,E.J.,Hara,Y.,Kitani,K.,2000.Antioxidativeactivityofgreentea treatedwithradicalinitiator2,2’-azobis(2-amidinopropane)dihydrochloride. JournalofAgriculturalandFoodChemistry48,5068–5073.

Zapolska-Downar, D., Naruszewicz, M., 2009. Dropionate reduces the cytokine-induced VCAM-1 and ICAM-1 expression by nuclear factor-␬B (NF-␬B) activation. Journal of Physiology and Pharmcology 60, 123– 131.