Rev. bras. farmacogn. vol.26 número3

(1)

w ww . e l s e v i e r . c o m / l o c a t e / b j p

Original

Article

Sesquiterpene

lactones

from

Hedyosmum

brasiliense

induce

in

vitro

relaxation

of

rat

aorta

and

corpus

cavernosum

Amanda

Leitolis

a,1

,

Solomon

K.S.

Amoah

b,1

,

Maique

W. Biavatti

b

,

J. Eduardo

da

Silva-Santos

a,∗ a_Laboratory_of_{Cardiovascular}_Biology,_Department_of_{Pharmacology,}_Universidade_Federal_de_Santa_Catarina,_{Florianópolis,}_SC,_Brazil

b_Laboratory_of_{Pharmacognosy,}_Department_of_{Pharmaceutical}_Sciences,_Universidade_Federal_de_Santa_Catarina,_{Florianópolis,}_SC,_Brazil

a

r

t

i

c

l

e

i

n

f

o

Articlehistory:

Received16October2015 Accepted19January2016 Availableonline3March2016

Keywords:

Vasodilation Nitricoxide Erectilefunction Aphrodisiac Vascularfunction Smoothmuscle

a

b

s

t

r

a

c

t

HedyosmumbrasilienseMiq.,Chloranthaceae,hasbeenusedinSouthernBrazilasasedative, anti-inflammatory,andaphrodisiac.Inthisstudy,endothelium-intactandendothelium-denudedrataortic ringsandstripsofcorpuscavernosumwereusedtoinvestigatetherelaxanteffectsofanhexane frac-tionofleavesofH.brasilienseanditssesquiterpenelactones13-hydroxy-8,9-dehydroshizukanolide, podoandin,andelemanolide15-acetoxy-isogermafurenolide.Theincubationofhexanefractionofleaves ofH.brasilienseresultedinsignificantrelaxationofendothelium-intactaorticringspreviously con-tractedbyphenylephrine.Inaddition,13-hydroxy-8,9-dehydroshizukanolideandpodoandindisplayed aclearconcentration-dependentabilitytorelaxendothelium-intact(∼85to90%)and endothelium-denuded(∼45to55%) rataorticrings. A lesspronounced vascularrelaxationwas recordedwhen 15-hydroxy-isogermafurenolidewas tested. Interestingly,in tissues previously incubated with the nitric oxide synthase inhibitor l_-NAME ₍₁₀₀_␮_M), _both _{13-hydroxy-8,9-dehydroshizukanolide} _and podoandinhadtheireffectsinendothelium-intactvesselsreducedtothesamedegreeofrelaxation observedinendothelium-denudedaorticrings.Podoandin,13-hydroxy-8,9-dehydroshizukanolide,and 15-acetoxy-isogermafurenolide(100␮M)werealsoabletorelaxprecontractedcorpuscavernosumstrips

by49.5±3.9%,65.9±7.3%and57.9±5.5%,respectively.Ourresultsdemonstratedthat 13-hydroxy-8,9-dehydroshizukanolide,podoandinand15-acetoxy-isogermafurenolide,isolatedfromH.brasiliense, generatebothendothelium-dependentand-independentrelaxationofrataorticrings,aswellasbeing abletoinduceinvitrorelaxationofratcorpuscavernosum.Importantly,theendothelium-dependenteffect isfullydependentonnitricoxideproduction.Consideringthatpenileerectiondependsonboth relax-ationofcavernosalsmoothmuscleandinflowofbloodforthecavernousbodies,thisisthefirststudy reportingexperimentalevidencesupportingtheaphrodisiacpropertiesofH.brasiliense.

Introduction

Hedyosmum brasiliense Miq. is a member of the Chloran-thaceaefamilywidelydistributedinSouthernBrazil.Itispopularly knownas“cidrão”,“cidreira”and“erva-de-bugre”,amongothers. In folk medicine, it is indicated for thetreatment of migraine, depressionanddiseasesofovary,andhasbeenusedassedative, anti-inflammatory, and aphrodisiac (Reitz, 1965). Experimental evidencedemonstratedthatsolutionsobtainedfromH.brasiliense presentanalgesic(Trentinetal.,1999a),hypnotic-,anxiolytic-and

∗ Correspondingauthor.

E-mail:j.e.silva.santos@ufsc.br(J.E.daSilva-Santos).

1_These_two_authors_contributed_equally_to_the_work.

antidepressant-like effects in rodents (Tolardo et al., 2010;

Goncalvesetal.,2012).Nevertheless,atleasttoourknowledge,

theaphrodisiaceffecthasneverbeeninvestigated.

Sesquiterpenelactones areagroupofsecondary metabolites foundinseveralspecies,suchasAcanthaceae,Amaranthaceae, Api-aceae,Aristolochiaceae,Burseraceae,Coriariaceae,Illiciaceae, Mag-noliaceae,Menispermaceae,Lamiaceae,Lauraceae,Polygonaceae, Winteraceae, andChloranthaceae(Picman,1986;Kawabata and

Mizutani, 1988; Cao et al., 2008; Huet al., 2013). These

com-poundsarewellknownfortheirantifeedantandprotectiveroles in plants, mainly against infections by viruses, bacteria, and fungi(Nawrotand Harmatha,2012;Chadwicketal.,2013).The potentialtherapeuticeffectsofsesquiterpenelactoneshavebeen widelyinvestigatedinanimalmodels,revealinganti-inflammatory

(Merfort,2011;Ferrarietal.,2013),antimicrobial(Ordonezetal.,

http://dx.doi.org/10.1016/j.bjp.2016.01.005

(2)

2011;Ciricetal.,2012),antitumor(Zhangetal.,2005),andantiviral effects(Ozceliketal.,2009;Rossinietal.,2012;Mohammedetal., 2014), amongothers.Inaddition,theirbiologicaleffects include vascularrelaxationasdemonstratedbyseveralstudies(Campos

etal.,2003;Hongetal.,2005).Thestructureofsesquiterpene

lac-tonesconsistsofthreeisopreneunitsandalactonegroup,which mayaccordinglybeclassifiedwiththeircarbonskeletons.There areabout100typesofcarbonskeletons,andalthoughthemajority ofthesesquiterpenelactonesareclassifiedintosevenmaingroups knownasgermacranolides,guaianolides,elemanolides, eremofi-lanolides,eudesmanolides,pseudoguaianolidesandxanthanolides

(Schmidt, 2006),phytochemical studieshave shown

lindenano-lidesasthemainchemicalmarkersoftheChloranthaceaefamily

(KawabataandMizutani,1988).Importantly,someofthe

biolog-icaleffectsfoundaftertheadministrationof solutionsobtained fromH.brasiliense in rodentswere,at least inpart,dependent ontheactivityofsesquiterpenelactones,suchasthe lindenano-lideonoseriolideor13-hydroxy-8,9-dehydroshizukanolide(HDS)

(Trentin etal., 1999a), and theguaianolide podoandin(Tolardo

etal.,2010;Goncalvesetal.,2012).

Given the popular usage of H. brasiliense, its high amounts of sesquiterpene lactones and the putative involvement of these compounds to explain some of the popular uses of this plant,wedesigned this studyto investigatetheability ofHDS, podoandin,and therecentlydescribedelemanolide 15-acetoxy-isogermafurenolide(IGM-A;Amoahetal.,2013),toinduceinvitro relaxationofratcorpuscavernosumandaorta.

Materialsandmethods

Plantmaterial

Hedyosmum brasiliense Miq., Chloranthaceae, was collected from the municipal area of Antonio Carlos in the Santa Cata-rinastate(Brazil).ItwasidentifiedbythebotanistProfessorDr. AdemirReisandavoucherspecimen(number2031)wasdeposited at the Lymann Bradford Smith Herbarium (Univali, Itajaí, SC, Brazil).

Preparationofthehexanefractionandisolationofcompounds

Freshleaves(5kg)ofH.brasilienseweresubjectedtoextraction withbi-distilledethanolfor15days.Thesolventwassubsequently removedunderreducedpressureusingarotaryevaporator.The recuperated solvent was used to re-extract the plant material twice, thereby resulting into a total of 210g of crude extract. Thecrudeextract(190g)wassubjectedtoliquid:liquidpartition toyieldhexane (16g),CH2Cl2 (4g),EtOAc(13g),andthe resid-ualaqueous(74g) fraction. Allfractionswerestored at −18◦_C.

About2gofthehexanefraction(HFHB)waskeptfor biological andpharmacologicaltestsandtherestwassubjectedtoflash sil-icagel column chromatography (240–400mesh).It was eluted with a gradient of 0–70% CH2Cl2 in hexane (200ml) followed by0–70%EtOAcin CH2Cl2 (200ml),yieldingeight sub-fractions (A–H).

Thesub-fractionBwasrecrystallizedtoaffordtheguaianolide podoandin (1; 300mg). It was identified comparing its spec-troscopic data with already published data (Blay et al., 2000; Kubo et al., 1992). There was a spontaneous crystallization of the previously described (Trentin et al., 1999b) lindenanolide 13-hydroxy-8,9-dehydroshizukanolide(2,HDS;40mg)from sub-fraction C. The elemanolide 15-acetoxy-isogermafurenolide (3, IGM-A; 35mg) was isolated from sub-fraction F, as previously detailed(Amoahetal.,2013).

O O H

H

O O

HO

O O

2

1

3

Drugsandreagents

Acetylcholine (ACh) chloride, phenylephrine hydrochloride (PE), and N␻-nitro-l-arginine methyl ester hydrochloride (l -NAME), as well as the salts used for the preparation of the physiological saline solution (PSS) were obtained from Sigma–AldrichCo.(StLouis,MI,USA).Allreagentsusedfor prepa-rationofthehexanefractionandtheisolationofsesquiterpenes wereofanalytical grade.ThestocksolutionofHDS,podoandin, andIGM-AweredissolvedinpureDMSO.Aconcentratedsolution ofhexanefractionwaspreparedusingDMSOanddeionizedwater (1:10).AllworkingsolutionsofHDS,podoandin,andIGM-A,which wereappliedintheinvitroexperimentswere,however,prepared inphysiologicalsalinesolution.

Animals

MaleWistarrats(3–4months)wereprovidedbytheCentral Vivarium facilities of the Universidade Federal de Santa Cata-rina.The animalswere maintainedundercontrolled light–dark cycle(12/12h) and temperature(22±2◦_C), _with_free _access_to

waterand chow.Thisstudy wasperformed inaccordance with theNIHguidelines foranimal experimentation.Inaddition,the Institutional Ethics Committee for Animal Use (CEUA) of the UniversidadeFederalde SantaCatarina(UFSC,Brazil) approved all procedures adopted in this study (authorization number 5371190815).

Isolationofcorpuscavernosumandaorta

Immediatelybeforetheexperiments,theanimalswere euth-anizedbyanesthesiaoverdose(ketamine andxylazine,i.p.)and hadtheirthoracicaortaandpeniscarefullyexcisedandkeptin physiologicalsalinesolution(PSS,compositioninmM:130.3NaCl, 4.7KCl,1.6CaCl2·2H2O,1.18KH2PO4,1.17MgSO4,14.9NaHCO3, 5.5d-glucose)warmedat37◦C.Afterremovingadherenttissues, theaortawascutintoringsofapproximately3–4mm.The dor-salpenileveinandspongytissuewerecarefullyremovedfromthe penisandthecorpuscavernosumwasdividedintotwo longitudi-nalstrips.Bothaorticringsandstripsofcorpuscavernosumwere mountedinorganbathscontainingPSS(37◦_C,_continuously_aerated

(3)

InvitroevaluationoftheeffectsofH.brasilienseanditsisolated sesquiterpenelactonesincorpuscavernosumandaorta

Afterthestabilizationperiodof60minbothcorpuscavernosum andaorticringswereexposedtohighpotassiumnutritive solu-tioncontaining80mMor120mMKCl,respectively.Thetension wasrecordedfor15minandthepreparationswerewashedand allowed to stabilize during 30min. The strips of corpus caver-nosum were then exposed to phenylephrine (PE; 10␮M), and after reachingthe tonic phase of contraction,the preparations wereincubatedwithcumulativeconcentrations ofHFHB(3,10, 30, 100 and 300␮g/ml), HDS, podoandin, or IGM-A (10nM to 100␮M,for allcompounds). Time-matchedexperiments where thepreparationswereexposedtothesameamountsofthe vehi-cle used to dilute the HFHB or the isolated compounds were usedascontrol.Theresultingrelaxationwasrecordedand eval-uated.Importantly,eachstripofcorpuscavernosumwasincubated onlywithoneofthecompoundsorfractionsinvestigatedinthis study.

Thesameprotocolwasfollowedinrataorticrings.However, once these experimentswere performed in both endothelium-intact and endothelium-denuded preparation, these vessels underwent a first exposition to PE (1␮M) followed by addi-tionofacetylcholine(ACh;1␮M).TheinabilityofAChtoinduce relaxation in PE-contracted preparations was used to confirm the lack of functional endothelium. On the other hand, aortic ringswhoseACh-inducedrelaxationwasabove80%were consid-eredendothelium-intactpreparations.Aftertheevaluationofthe endothelialfunctionality,a newinterval of60minwasallowed forstabilization,followedbyadditionofPE(1␮M)and cumula-tiveconcentrationsofHFHB(3,10,30,100and300␮g/ml),HDS, podoandin, orIGM-A(10nMto100␮M,for allcompounds). In addition,theeffectsofHDS,podoandinandIGM-Awerealso eval-uatedinendothelium-intactaorticringspreviouslyincubatedwith 100␮Ml-NAME for 20min.For all experiments,therelaxation inducedby HFHB, HDS,podoandin, and IGM-A,in preparations withor withoutfunctionalendothelium,aswellasinthe pres-enceorabsenceofl-NAME,wererecordedandcomparedamong thegroups.

Importantly, for control parameters we performed time-matchedexperimentsinwhichbothcorpuscavernosumstripsand aorticringswereexposedtothesameamountsofthevehicleused todilutetheHFHBortheisolatedcompounds.

Statisticalanalysis

Theresultswereexpressedasmean±standarderrorofmean (S.E.)of5–6preparations(removed fromdifferentanimals)per group.Thegraphsandthestatisticalanalysiswereperformedwith theprogram GraphPad Prism® _version _6.0d_for _Mac _(GraphPad Software,LaJolla,CA,USA).Comparisonsweremadebyoneway analysisofvariance(ANOVA)followedbyBonferroniposthoctest. Avalueofp<0.05wasconsideredstatisticallysignificant.

Resultsanddiscussion

Thefirstsetofresultsobtainedinthisstudyrevealedthatthe hexanefractionpreparedwithleavesofH.brasiliense(HFHB) pre-sentedminoreffectsonthevasculartoneofrataorticringsthat had its endothelial surface layer damaged prior to contraction inducedbytheselective␣-1adrenergicagonistphenylephrine(PE). Indeed,thehighestconcentrationofHFHBtested inour experi-ments(300␮g/ml)decreasedthevasculartoneofPE-contracted vesselsby22.3±4.5%,whichwasnotstatisticallydifferentfrom therelaxation obtainedwhentheseendothelium-denuded (E−) vessels wereexposedtothesameamounts ofdimethyl sulfox-ide,thevehicleusedtodissolvetheHFHB(Fig.1A).Nonetheless, a robustrelaxationwasobtainedwhentheHFHBwastestedin endothelium-intact (E+) aorticrings. In these preparations, the additionof100and300␮g/mloftheHFHBintotheorganbaths reduced thevascular toneby33.4±5.3and70.2±8.6%, respec-tively, which were significantly higher than the 10.2±2.7 and 13.6±4.8%maximumrelaxationinducedbyvehicleonly(Fig.1B). Consideringthepronouncedvascularrelaxationobtainedwith theHFHBandthepreviouslydescribedmaincomponentsisolated fromthisfraction(Amoahetal.,2013),wedidevaluatethe poten-tialroleofthesesquiterpenelactonesHDS,podoandinandIGM-A inHFHB-inducedaortarelaxation.Notably,thefinalconcentration ofDMSOpresentintheworkingsolutionscontainingthese com-poundsdidnotinfluencethetoneofpre-contractedaorticrings, asconfirmedbytime-matchedexperimentsperformedduringthe pharmacologicaltests(Fig.2,diamond-shapedsymbols).BothHDS and podoandindisplayeda clearconcentration-dependent abil-ity torelaxboth endothelium-intactand endothelium-denuded rat aortic rings. However, as demonstrated in Fig. 2A and B (open circles) a more intense vascular relaxation was reached invesselswithfunctionalendothelium.Forinstance,atthefinal

0

50

100

3

A

B

C

Vehicle HFHB (µg/ml)

Aortic rings Endothelium-denuded

Aortic rings

Endothelium-intact Corpus cavernosum

Relaxation, %

HFHB (µg/ml) HFHB (µg/ml)

*

n-hexane fraction (HFHB)

10 30 100 300 3 10 30 100 300 3 10 30 100 300

0

50

100

0

50

100

(4)

A

0

B

C

50

Relaxation, %

100

0

50

100

0

50

100

–8 –7

* #

#

# #

#

# *

*

* *

*

* *

*

* _*

* *

*

–6 HDS log [M]

Vehicle E+ E– E+/L–NAME (100 µM)

Podoandin log [M] IGM-H log [M]

–5 –4 –8 –7 –6 –5 –4 –8 –7 –6 –5 –4

Fig.2. Endothelium-dependentand-independentvascularrelaxationinducedbysesquiterpenelactonesisolatedfromHedyosmumbrasiliense.Cumulativeconcentrations ofHDS(A),podoandin(B),andIGM-A(C)wereobtainedinendothelium-denuded(E+;closedcircles),endothelium-intact(E−;opencircles),orendothelium-intactrataortic ringspreviouslyincubatedwiththenitricoxidesynthaseinhibitorl-NAME(squares),andcontractedbyphenylephrine(1␮M).Thedashedline(diamond-shapedsymbols)

showsthelackofinfluenceofthevehicleusedtodissolvethecompoundsonthevasculartone.Eachpointrepresentsthemean±S.E.of5–6preparationsfromdifferent animalspergroup.Statisticalanalyseswereperformedusingone-wayanalysisofvariance(ANOVA)followedbyt-testsubjectedtoBonferronicorrection.*Indicatesavalue ofp<0.05whencomparedwiththevehiclegroup.#Indicatesavalueofp<0.05whentheE+groupwascomparedwiththeE+/l-NAMEgroup.

concentrationof100␮M,HDSandpodoandinwereabletorelax endothelium-intactvesselsby90.1±5.9%and86.8±8.0%, respec-tively,whileinendothelium-denudedpreparations,themaximum relaxation obtained were 54.6±5.9% and 46.6±5.7%, respec-tively.Interestingly,therewerenodifferencesbetweenHDSand podoandinintheirpotencytoinducevascularrelaxation,either inendothelium-intactorendothelium-denudedaorticrings.For instance,thecalculatedEC50(concentrationthatgeneratedthehalf maximalresponse)foundinendothelium-intactvesselswere8.7 (4.6–17.1)␮M,and 10.6(4.6–24.4)␮M,forHDSandpodoandin, respectively.Regardingtheeffects of IGM-A,a less pronounced vascular relaxation wasrecorded when this sesquiterpene was incubatedinPE-contractedaorticrings(Fig.2C),whichwasable toinducea maximal relaxationof 51.1±11.6%and 26.0±5.1%, in endothelium-intact and endothelium-denuded vessels. Even though it hasbeen observed in previousstudies that different sesquiterpenelactonescausevascular relaxation(Campos etal.,

2003;Hongetal.,2005),thisisthefirsttimereportofthiseffect

induced by a fraction prepared with H. brasiliense, as well as withthesethreesesquiterpenelactones(i.e.HDS,podoandin,and IGM-A).

Endothelium-dependent vascular relaxation has often been associated with the production of nitric oxide, mainly by the endothelialisoform oftheenzyme nitricoxidesynthase(NOS), despitethewellknownexistenceofotherendothelialmediators abletorelaxthesmoothmusclecellsandtheirinvolvementinthe controlofvasculartone,suchascyclooxygenasederivativesand theendothelium-derivedhyperpolarizingfactor,thatremainsto beidentified(DurandandGutterman,2013).Toverifytheroleof nitricoxideinourfindings,l-NAME,anon-selectiveNOSinhibitor, wasincubatedwithendothelium-intactpreparationsbeforethe assessment of thevasorelaxant effects of HDS,podoandin, and IGM-A.BothHDSandpodoandinhadtheirabilitytocause relax-ationpartiallyavoidedbyl-NAME(Fig.2AandB).Indeed,when incubatedwithl-NAME,theeffects inducedbythese sesquiter-penelactonesinendothelium-intactvesselswerereducedtothe samedegreeofrelaxationobservedinendothelium-denuded aor-tic rings. The incubation of l-NAME has been inconclusive on whetherIGM-A-inducedrelaxationdoesordoesnotdependon nitricoxideproduction, perhapsbecausethe windowof differ-ence in the responses for this compound between E+ and E−

preparationswasreduced,aswellasduethewidevariabilityof effectsobtainedwithIGM-Ain aorticrings(Fig. 2C). Neverthe-less,ourresultsclearlydemonstratethatthehighereffectiveness ofbothHDSandpodoandininendothelium-intactarteries,when compared with endothelium-denuded vessels, is dependenton

stimulationofnitricoxideproduction.We didnotperform fur-therapproachestoexplorethedownstreammechanismsinvolved inthevascularrelaxationmediatedbyHDSandpodoandin. How-ever,nitricoxideisclassicallyknownasabiologicalactivatorof thesolubleguanylatecyclase,includinginvascularsmooth mus-clecells(Arnoldetal.,1977).Inaddition,nitricoxideisalsoable todirectly interactwithand opencalcium-activated potassium channels(Bolotinaetal.,1994).Wehavepreviouslydemonstrated theinvolvement of bothsoluble guanylate cyclase andcalcium activatedpotassiumchannelsin thevasodilatoryeffects associ-atedwithothermedicinalplants(Rattmannetal.,2006;DaSilva etal.,2012).Thus,it isreasonabletospeculate thatoncethese sesquiterpenelactoneshadtheirendothelium-dependenteffects fullyavoidedbyNOSinhibition,theactivationoftheclassicalnitric oxide/guanylatecyclasepathwayaccountsforthevascular relax-ationdescribedinthisstudy.Importantly,asignificantpartofthe vascularrelaxationmediatedbythethreesesquiterpenelactones investigatedinherepersistedafterendotheliumdamageorNOS inhibition(Fig.2,closedcirclesandsquares).Ifononehandthe mechanismsinvolvedinsucheffectremainstobeinvestigated,on theotherhanditmayconferonthesecompoundstheabilityto inducesomedegreeofvascularrelaxationeveninthoseconditions characterizedbyendothelialdysfunction.

(5)

0

50

100

–8

Vehicle

*

HDS

Podoandin

Sesquiterpene lactones log [M]

Relaxation, %

IGM-H

–7 –6 –5 –4

Fig.3.Relaxationofstripsofcorpuscavernosumfromratsbysesquiterpenelactones isolatedfromHedyosmumbrasiliense.Phenylephrinecontractedpreparationswere exposedtocumulativeconcentrationsofHDS(opencircles),podoandin(closed cir-cles)orIGM-A(triangles).Thedashedline(diamond-shapedsymbols)showsthe lackofinfluenceofthevehicleusedtodissolvethecompoundsonthetoneof cav-ernosalsmoothmuscle.Eachpointrepresentsthemean±S.E.of5–6preparations fromdifferentanimalspergroup.Statisticalanalyseswereperformedusingone-way analysisofvariance(ANOVA)followedbyt-testsubjectedtoBonferronicorrection. *Indicatesavalueofp<0.05whentheresultsobtainedinallgroupsofcompounds werecomparedwiththevehiclegroup.

only. However, precontracted corpus cavernosum strips relaxed by49.5±3.9%,65.9±7.3%,and57.9±5.5%whenincubatedwith 100␮MHDS,podoandin,andIGM-A,respectively(Fig.3).Among thelocallyproducedmediators,nitricoxidereleasedbyendothelial cellsandnerveendingsplaysamajorroleinpenileerection.Once produced,itdiffusestoadjacentsmoothmusclecellsdilatingthe arteriesandthesmoothmuscleofthecorpuscavernosum.Indeed, thepotentiationofthenitricoxide/guanylatecyclasepathwayby agentssuchassildenafilisthemainpharmacologicalstrategyused inthetreatmentoferectiledysfunction(Andersson,2001;Dean andLue,2005).Ourdatadoesnotallowustostatewhich intra-cellularmechanismsthesesesquiterpenelactonescouldbeusing tocausetherelaxationofthecorpuscavernosum,buttakinginto accounttheresultsobtainedinrataorticrings,theroleofthenitric oxideinthiseffectdeservesfurtherinvestigation.

Conclusions

This work discloses that a hexane fraction prepared with fresh leavesof H. brasiliense, as well as the sesquiterpene lac-tonesHDS,podoandin,andIGM-A,isolatedfromthisfraction,are abletoinducebothendothelium-independentand endothelium-dependentrelaxationofrataorticrings.Ourresultsdemonstrated that,atleastforHDSandpodoandin,theendothelium-dependent relaxationisfullydependentonnitricoxideproduction.Inaddition, althoughfolkmedicinehasalreadypointedtheuseofH.brasiliense asanaphrodisiac(Reitz,1965),thisisthefirsttimethatastudy reportssomeexperimentalevidencethatsupportthiseffect,since wefoundthatHDS,podoandinandIGM-Aareabletoinduceinvitro relaxationofcorpuscavernosum.Consideringthatthepenile erec-tiondependsonbothrelaxationofcavernosalsmoothmuscleand inflowofbloodforthecavernousbodies,additionalstudiesshould bedevelopedtofurtherinvestigatetheeffectsofH.brasiliense,as wellasitsisolatedsesquiterpenelactones,onbothvascular func-tionandpenileerectionusinginvivoapproaches.

Authorship

S.K.S.A. worked in the preparation of extracts and isolation of the sesquiterpene lactones during his Ph.D. studies, under

supervisionofMWB.A.L.performedtheinvitrostudiesand ana-lyzedtheresultsobtained.A.L.andS.K.S.A.draftedthemanuscript. JESSwasresponsiblefortheconceptionofthestudy,including pro-tocolsandinterpretationofdatawhilesupervisingA.L.duringher graduatestudiesinPharmacology,andworkedonthemanuscript.

Conflictsofinterest

Theauthorsdeclarenoconflictsofinterest.

Acknowledgments

A.LeitolisconductedherM.Sc.studiesasrecipientof fellow-shipfromCNPq,Brazil.S.K.S.Amoahreceivedafellowshipfrom CAPES,Brazil.WethankCNPqandFundac¸ãodeAmparoaPesquisa eInovac¸ãodeSantaCatarina(FAPESC,SC,Brazil)forpreviousgrants thatallowedthedevelopmentofthisresearch.

References

Amoah,S.K.,deOliveira,F.L.,daCruz,A.C.,deSouza,N.M.,Campos,F.R.,Barison, A.,Biavatti,M.W.,2013.SesquiterpenelactonesfromtheleavesofHedyosmum brasiliense(Chloranthaceae).Phytochemistry87,126–132.

Andersson, K.E., 2001. Pharmacology of penile erection. Pharmacol. Rev. 53, 417–450.

Andersson,K.E.,Wagner,G.,1995.Physiologyofpenileerection.Physiol.Rev.75, 191–236.

Arnold,W.P.,Mittal,C.K.,Katsuki,S.,Murad,F.,1977.Nitric-oxideoctivates guany-latecyclase andincreases guanosine3′_-5′_-cyclic_{monophosphate}_levels_in

varioustissuepreparations.Proc.Natl.Acad.Sci.U.S.A.74,3203–3207.

Blay,G.,Bargues,V.,Cardona,L.,García,B.,Pedro,J.R.,2000.Stereoselective syn-thesisof7,11-guaien-8,12-olidesfromsantonin.Synthesisofpodoandinand (+)-zedolactoneA.J.Org.Chem.65,6703–6707.

Bolotina,V.M.,Najibi,S.,Palacino,J.J.,Pagano,P.J.,Cohen,R.A.,1994.Nitricoxide directlyactivatescalcium-dependentpotassiumchannelsinvascularsmooth muscle.Nature368,850–853.

Campos,M.,Oropeza,M.,Ponce,H.,Fernández,J.,Jimenez-Estrada,M.,Torres,H., Reyes-Chilpa,R.,2003.Relaxationofuterineandaorticsmoothmuscleby glau-colidesDandEfromVernonialiatroides.Biol.Pharm.Bull.26,112–115.

Cao,C.M.,Peng,Y.,Shi,Q.W.,Xiao,P.G.,2008.Chemicalconstituentsandbioactivities ofplantsofchloranthaceae.Chem.Biodivers.5,219–238.

Chadwick,M.,Trewin,H.,Gawthrop,F.,Wagstaff,C.,2013.Sesquiterpenoids lac-tones:benefitstoplantsandpeople.Int.J.Mol.Sci.14,12780–12805.

Ciric,A.,Karioti,A.,Koukoulitsa,C.,Sokovic,M.,Skaltsa,H.,2012.Sesquiterpene lactonesfromCentaureazuccarinianaandtheirantimicrobialactivity.Chem. Biodivers.9,2843–2853.

DaSilva,R.d.C.,Crestani,S.,deSouza,P.,Boligon,A.A.,Athayde,M.L.,Santos,A.R., Marques,M.C.,Kassuya,C.A.,DaSilva-Santos,J.E.,2012.Endothelium-dependent andindependentvasorelaxationinducedbyann-butanolicfractionofbarkof ScutiabuxifoliaReiss(Rhamanaceae).J.Ethnopharmacol.141,997–1004.

Dean,R.C.,Lue,T.F.,2005.Physiologyofpenileerectionandpathophysiologyof erectiledysfunction.Urol.Clin.N.Am.32,379–380.

Durand,M.J.,Gutterman, D.D.,2013.Diversityinmechanismsof endothelium-dependentvasodilationinhealthanddisease.Microcirculation20,239–247.

Ferrari,F.C.,Ferreira,L.C.,Souza,M.R.,Grabe-Guimaraes,A.,Paula,C.A.,Rezende, S.A.,Saude-Guimaraes,D.A.,2013.Anti-inflammatorysesquiterpenelactones fromLychnophoratrichocarphaSpreng.(BrazilianArnica).Phytother.Res.27, 384–389.

Goncalves,A.E.,Burger,C.,Amoah,S.K.,Tolardo,R.,Biavatti,M.W.,deSouza,M.M., 2012.Theantidepressant-likeeffectofHedyosmumbrasilienseandits sesquiter-penelactone,podoandininmice:evidencefortheinvolvementofadrenergic, dopaminergicandserotonergicsystems.Eur.J.Pharmacol.674,307–314.

Hong,Y.,Yang,M.S.,Pak,Y.,2005.EffectofcumambrinAontherelaxationofrat aorta.KoreanJ.Pharmacog.36,17–20.

Hu,X.R.,Wu,H.F.,Zhang,X.P.,Yang,J.S.,Dai,Z.,Lin,R.C.,Xu,X.D.,2013.Anew sesquiterpenelactonefromSarcandraglabra.Nat.Prod.Res.27,1197–1201.

Kawabata,J.,Mizutani,J.,1988.Studiesonthechemical-constituentsof Chloran-thaceaeplants.DistributionoflindenanolidesintheChloranthaceae.Agric.Biol. Chem.Tokyo52,2965–2966.

Kubo,I.,Ying,B.P.,Castillo,M.,Brinen,L.S.,Clardy,J.,1992.Podoandin,amolluscicidal sesquiterpenelactonefromPodocarpusandina.Phytochemistry31,1545–1548.

Merfort,I.,2011.Perspectivesonsesquiterpenelactonesininflammationandcancer. Curr.DrugTargets12,1560–1573.

Mohammed,M.M.D.,Christensen,L.P.,Colla,P.L.,2014.Isolationandanti-HIV-1 activityofanewsesquiterpenelactonefromCalocephalusbrowniiF.Muell.Nat. Prod.Res.28,221–229.

Nawrot,J.,Harmatha,J.,2012.Phytochemicalfeedingdeterrentsforstoredproduct insectpests.Phytochem.Rev.11,543–566.

(6)

fromGynoxysverrucosaandtheiranti-MRSAactivity.J.Ethnopharmacol.137, 1055–1059.

Ozcelik,B.,Gurbuz,I.,Karaoglu,T.,Yesilada,E.,2009.Antiviralandantimicrobial activitiesofthreesesquiterpenelactonesfromCentaureasolstitialisL.ssp solsti-tialis.Microbiol.Res.164,545–552.

Picman,A.K.,1986.Biologicalactivitiesofsesquiterpenelactones.Biochem.Syst. Ecol.14,255–281.

Rattmann,Y.D.,Cipriani,T.R.,Sassaki,G.L.,Iacomini,M.,Rieck,L.,Marques,M.C., daSilva-Santos,J.E.,2006.Nitricoxide-dependentvasorelaxationinducedby extractivesolutionsandfractionsofMaytenusilicifoliaMartexReissek (Celas-traceae)leaves.J.Ethnopharmacol.104,328–335.

Reitz, R., 1965. Clorantaceas – Flora Ilustrada Catarinense. Herbário Barbosa Rodrigues,Itajaí,pp.4–10.

Rossini,M.,Scotti,M.T.,Correia,M.V.,Fokoue,H.H.,Scotti,L.,Mendonca,F.J.B., daSilva,M.S.,Emerenciano, V.D.,2012. Sesquiterpenelactoneswith anti-hepatitisCvirusactivityusingmoleculardescriptors.Lett.DrugDes.Discov.9, 881–890.

Schmidt,T.J.,2006.Structure–activityrelationshipsofsesquiterpenelactones.Stud. Nat.Prod.Chem.33,309–392.

Tolardo,R.,Zetterman,L.,Bitencourtt,D.R.,Mora,T.C.,deOliveira,F.L.,Biavatti,M.W., Amoah,S.K.,Burger,C.,deSouza,M.M.,2010.Evaluationofbehavioraland pharmacologicaleffectsofHedyosmumbrasilienseandisolatedsesquiterpene lactonesinrodents.J.Ethnopharmacol.128,63–70.

Trentin,A.P.,Santos,A.R.,Guedes,A.,Pizzolatti,M.G.,Yunes,R.A.,Calixto,J.B.,1999a.

AntinociceptioncausedbytheextractofHedyosmumbrasilienseanditsactive principle, the sesquiterpene lactone13-hydroxy-8,9-dehydroshizukanolide. PlantaMed.65,517–521.

Trentin, A.P., Santos, A.R.S., Guedes, A., Pizzolatti, M.G., Yunes, R.A., Cal-ixto, J.B., 1999b. Antinociception caused by the extract of Hedyosmum brasilienseanditsactiveprinciple,thesesquiterpenelactone 13-hydroxy-8,9-dehydroshizukanolide.PlantaMed.65,517–521.