RP-HPLC and LC–MS–MS determination of a bioactive artefact from Ipomoea pes-caprae extract

(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

RP-HPLC and LC–MS–MS determination of a bioactive artefact from Ipomoea pes-caprae extract

Cristiane da Silva Barth

^a

, Hugo Guilherme Tolentino de Souza

^b

, Lilian W. Rocha

^a

,

Camilla Regina de Souza Madeira

^c

, Camila Assis

^d

, Tiago Bonomini

^a

, Angela Malheiros

^a,b

, Louis P. Sandjo

^e

, Rivaldo Niero

^a,b

, Cechinel Filho

^a

, Angelica Garcia Couto

^a,b

,

José Roberto Santin

^a

, Nara Lins Meira Quintão

^a

, Tania Mari Belle Bresolin

^a,b,∗

aUniversidadedoValedoItajaí,ProgramadePós-Graduac¸ãoemCiênciasFarmacêuticas,Itajaí,SC,Brazil

bUniversidadedoValedoItajaí,CursodeFarmácia,Itajaí,SC,Brazil

cUniversidadeFederaldoParaná,DepartamentodeFarmácia,LaboratóriodeControledeQualidade,Curitiba,PR,Brazil

dUniversidadeFederaldeMinasGerais,DepartmentodeQuímica,BeloHorizonte,MG,Brazil

eUniversidadeFederaldeSantaCatarina,DepartamentodeCiênciasFarmacêuticas,Florianópolis,SC,Brazil

a r t i c l e i n f o

Articlehistory:

Received12February2019 Accepted2May2019

Keywords:

Chemometrics Biomonitoring Inﬂammation Nociception Ethylcaffeate

a b s t r a c t

Solventsplayimportantandcriticalroleinnaturalproductchemistryandcouldgenerateartefactsduring theextractionandpuriﬁcationofmetabolitesfromabiologicalmatrix.Thisstudyaimedtocorrelatethe chromatographicproﬁlewithbiologicalactivityofIpomoeapes-caprae(L.)R.Br.,Convolvulaceae,extracts obtainedwithhydroethanolicextraction.Thus,aerialpartsofI.pes-capraewereextractedwithdiffer- entconcentrationofethanol(50,70and90^◦GL)andtheobtainedextractswereanalysedbyHPLC–UV.

HPLCdatawerestudiedemployingchemometricstodiscriminatethesamples.Moreoverthesesamples werefurthercharacterizedbyusingUPLC–QTOF/MSdata.Theextractswerealsobiomonitoredthrough thepaw-oedemaandspontaneousnociceptioninducedbytrypsininmice.Differentchromatographic profileswereobtainedandtheexploratoryanalysisclearlyrevealedhigherlevelofethylcaffeatein extractsoflowerstrengthofethanol(50^◦GL).Thiscompoundwassuggestedtobeanartefactformed bytransesterificationofcaffeoylquinicacidderivativespresentintheplant,onceitwasnotobserved whenothersolventswereemployed.Duringthebiologicalassay,onlytheextractobtainedwithethanol 50^◦GLpresentedsignificantinhibitionofinflammation(45±9%)andnociception(24±3%).Ethylcaffeate seemstobelinkedtotheanti-inflammatoryeffectsinceitreduced86±5%ofpaw-oedemainducedby trypsin.Artefactscouldcontributetothebiologicalactivityofherbalpreparationsandconsequentlylead tomisinterpretationoftheresults.

Introduction

Alargenumberofsolventsarebeingusedtoextractselectively thenaturalproductsfromherbalmatrices.Theinteractionsolute- solventismainlybasedonsolubilityandpolarity(Malteseetal., 2009).However,therearemanyexamplesofartefactsinnatural products,duetothesolventandextractionsconditionsused.

Asexample,lawsoneisanartefactformedduringtheextraction processproducedwhenleafpowderofLawsoniainermisL.(henna) isaddedtowatermadeslightlyacidicwithlemonjuiceorteaduring hennadyepreparation(AshnagarandShiri,2011);eleutherobin,

∗ Correspondingauthor.

E-mail:[email protected](T.M.Bresolin).

fromsoftcoralErythropodiumcaribaeorum,ispresentonlywhen usingmethanolbutnotethanolintheextraction(Brittonetal., 2001).Alcoholscanreactwithcarboxylicgroupstoformesters,and withhemiacetalstoformacetal(Malteseetal.,2009).Therefore, consideringthewiderangeofbiologicalactivityofplantextracts, whichoftheseactivitiescouldbeattributedtoartefactsgenerated duringthestepsofobtainingthesepreparations?Itisimportantto considertheprotocolusedfortheherbalmaterialandcompounds extraction,becauseitcaninterfereintheﬁnalherbalproduct.

Ipomoea pes-caprae (L.) R. Br., Convolvulaceae, in Brazil is knownas“salsa”,“salsa-brava”and“salsa-pé-de-cabra”(Simão- BianchiniandFerreira,2015),andinothercountries,asbayhops, beachmorning glory orgoat’s foot. Theleaveshave beenused asa ﬁrst aidtotreat jellyﬁshstings andalso inritual bathsto alleviateevil spells and Portuguese man-of-war stings-induced

https://doi.org/10.1016/j.bjp.2019.05.005

creativecommons.org/licenses/by-nc-nd/4.0/).

(2)

C.S.Barthetal./RevistaBrasileiradeFarmacognosia29(2019)570–577 571

dermatitis(Manigaunhaetal.,2010).Recentlyourresearchgroup demonstrated that thesoft extract of I.pes-caprae aerial parts (7.5%of aerial parts, ethanol 50^◦GL as solvent), when topically appliedinmicepaw(2%inasemisolid),blockedthemechanical hypersensitivityinducedbythetoxinofPhysaliaphysalis.Theintra- plantarinjectionofthemajorcomponent,3,5-di-O-caffeoylquinic acid, also abolished the mechanical hypersensitization (Barth et al.,2017).Previous studies reportedthe antinociceptive and anti-inﬂammatory topical effects of this plant, relating with apolar herbalderivatives (oil extract bysteam-distillation with petroleumether),whichdonotmimicpopularuseofthespecies (SunthonpalinandWasuwat,1985).

The active components reported for this species were 2- hydroxy-4,4,7-trimethyl-1(4H)-naphthalenone, (-)-mellein, eugenol,4-vinylguaiacol(Pongprayoonetal.,1991),glycosidesof 2-methylpropanoic, (2S)-methylbutyric, n-hexanoic,n-decanoic, andn-dodecanoicacids,aswellaspentasaccharidesofjalapinolic acid, pescaproside A, pescapreins I-IV, stoloniferin III (Pereda- Miranda et al., 2005), and actinidol 1a and 1b (Pongprayoon etal.,1992a).Furthermetabolites,suchastrans-␤-damascenone, E-phytol(Pongprayoonetal.,1992b),glochidone,betulinicacid,

␣-and␤-amyrinacetate,isoquercitrin(ISQ)(Kroghetal.,1999), 3,5-di-O-caffeoyl-quinicacid(isochlorogenicacida,denominated ISA),3,4-di-O-caffeoyl-quinicacid(isochlorogenicacidb,denominated ISB), 4,5-di-O-caffeoyl-quinicacid (isochlorogenicacidc, denominatedISC),amongothercaffeicacidderivativeswerealso identiﬁed(Teramachietal.,2005).

Thus,this studyaimedtoinvestigateifsolventextractionof aerialpartsofI.pes-capraeaffectstopicalantinociceptiveandanti- inﬂammatoryactivitiesusingpreclinicalmodelsofspontaneous nociceptionandpawoedemainducedbytrypsininmice,regarding thechemicalcomposition.

Materialandmethods Drugsandreagents

TrypsinwaspurchasedfromSigma–Aldrich(St.Louis,MO,USA), indomethacinfromDeg(SãoPaulo,Brazil).ForLCanalysis,allsol- ventsusedwereLCgrade(Tedia,Fairfield,Ohio,USA).Thewater was purified using a Milli-Qsystem (Millipore, Massachusetts, USA).AllsolutionswerefilteredthroughRC0.45␮mmembrane (Macherey-Nagel,Dage,Germany).ISQ,ISA,ISBandISC(>95%of puritybyHPLC-UV)werepurchasedfromSigmaAldrich(St.Louis, Missouri,USA).

Ethylcaffeatewasobtainedfromtheesteriﬁcationofcaffeicacid accordingwithCampos-Buzzietal.(2009),withpurity>90%(by HPLC–UV).Thestructuralcharacterizationwasconﬁrmedbasedon thespectraldata(MSand¹Hand¹³CNMR)anddirectcomparison withtheliteraturedata(Jaikangetal.,2011).

Fortheextractandcreampreparationallsolventswereanalyt- icalgrade.Hostacerin^®CG,Phenonip^®,EDTA,propyleneglycoland BHTwerepurchasedfromPharmaSpecial(SantanadeParnaíba,São Paulo,Brazil).Polymols(triglyceridesofcaprylicandcapricacid) waspurchasedfromAllChemistry(SãoPaulo,Brazil).

Plantmaterial

TheaerialpartsofIpomoeapes-caprae(L.)R.Br.,Convolvulaceae, werecollectedintheEsplanadaBeach(Jaguaruna,SantaCatarina, Brazil,28^◦3655Sand49^◦0131W),inJanuary(summer)of2013.

AnexsiccatewasdepositedintheUniversidadeEstadualdeMar- ingáHerbarium(Maringá,PR,Brazil)undernumberHUEM23566.

TheactivityofaccesstotheGeneticPatrimony/AssociatedTradi- tionalKnowledgewasregisteredintheSistemaNacionaldeGestão

do Patrimônio Genético e do Conhecimento Tradicional Associado (SisGen), in compliance with Law 13.123/2015 and its regula- tionsundernumberAFB43EB.Theleavesandstemsweredriedin roomtemperatureforsevendays,powderedinamillofhammers (Marconi^®600)andsieved(10mesh).

Extractspreparation

The hydroethanolic extracts was obtained from aerial parts (leaves:stemat60:40w/w)andpreparedbydynamicmaceration with50,70and90^◦GLethanolataplant:solventratioof7.5%(w/v), followedbyturboextractionfor10min(5cycleswith2mininter- vals).Leaveswerealsousedtoobtainanextractwith50^◦GLethanol inviewofcomparisonwithaerialparts.Thereafter,theextractwas ﬁlteredandtheextractivesolutionwasreducedbyheatingat50^◦C toobtainthesoftextractwithsyrupyappearance(driedresidue of93%,determinedon2g)(FarmacopeiaBrasileira,2017).These extractswerecharacterizedbydryresidue(andquantiﬁcationof secondarymetabolitesbyHPLC-UV.Methanol50%(v/v)andace- tone50%(v/v)softextractswerealsoelaborated,usinganalytical gradesolvents,inthesamewayofabovedescribed.

HPLC-UVanalysis

AShimadzuLC-10ADLCsystem(Shimadzu,Tokyo,Japan)con- sistingofabinarypumpandaShimadzuSPD-M10Aphotodiode array detector were used. Injections (20␮l) were carried out ona Phenomenex^® (Torrance, California, USA) Luna C18,5␮m (250×4.6mm)columnat30^◦Cwithdetectionat338nm,adapting themethodologypreviouslydeveloped(Dutraetal.,2014;Barth etal.,2017).Themobilephase,at1mlmin⁻¹,consistedofacetoni- trile(A)andpH3wateracidiﬁedwithsulphuricacid(B)according to the following gradient (A:B v/v): 10:90 to 15:85 (0–7min);

15:85to20:80(7–18min);20:80(isocratic18–25min);25:75to 60:40(25–40min);60:40 to70:30(40–43min);70:30to10:90 (47–50min),10:90(isocraticuntil60min).

Soft extracts were dissolved in 1:1 v/v of methanol:water pH3withsulphuricacidanda dilutedto2.5mgml⁻¹ solution.

Referencestandardsof3,5-di-O-caffeoylquinicacid(ISA),3,4-di- O-caffeoylquinicacid(ISB),4,5-di-O-caffeoylquinicacid(ISC)and isoquercitrin(ISQ)wereusedtoidentifythesesubstancesinthe extract.In addition,otherunknownpeakswerealsoquantified, expressedinISQ(peaks1-4,8–4)orISA(peaks5–-7),dependingon theUVabsorptionprofile,typicalofflavonoidorphenolic,respectively(Fig.1).

Multivariateanalysis

Principalcomponent(PCA)analysisofchemicaldatawereper- formedwiththeMINITAB^® 15software(StateCollege,PA,USA).

Acorrelationmatrixwasemployed,andthenumberofcompo- nentswasdeterminedfromKaiser’seigenvaluegreaterthan1.0 rule.17.AlldatawereprocessedusingMATLABsoftware,version 7.13(MathWorks,Natick,USA),togetherwiththePLSToolbox,ver- sion6.5(EigenvectorTechnologies,Manson,USA).Thenumberof principalcomponents(PC)wasdeterminedbyobservingtheper- centageofvarianceexplained.

UPLC-ESI-QToF-MS/MS

AnAcquityUPLCsystemH-class(Waters,SãoPaulo)equipped witha samplemanagerand a quaternarysolventmanager was used.TheseparationwascarriedoutinaBEHC18column:50mm, 1.0mm, particle size 1.7␮m (Waters, São Paulo). The column and the sample tray were kept at temperatures of 40^◦C and

(3)

Fig.1. Chromatographicalproﬁle,at338nm,ofIpomoepes-capraeextractsobtainedwithethanolindifferentstrength:(A)50^◦GL,(B)70^◦GL,(C)90^◦GL,(D)ethylcaffeate.

(4)

20^◦C, respectively.Themobilephase,at0.3mlmin⁻¹,consisted ofwater:formicacid,99.9:0.1v/v(A)andacetonitrile(B),accord- ingtothefollowing gradient(A:Bv:v):95:5to80:20(0–5min), 80:20to50:50(5–8min),50:50to45:35(8–10min),45:35to10:90 (10–13min), 10:90 (13–14.5min), 10:90 to95:5 (14.5–15min), 95:5(15–20min).Theinjectionvolumewas2␮l.

MassdatawererecordedwithaXevoG2-SQTof(Waters)mass spectrometerequippedwithanelectrosprayprobe operatingin positive(ESI⁺)and negative(ESI⁻)ionizationmodes.Thenebu- lizergaswas:nitrogen;conegasﬂow100l/h;desolvationgasﬂow 900l/h;samplingcone40V;sourceoffset80V;collisiongas:argon;

Locksprayreferencesample:Leucineenkephalin.Lockmassesm/z 556.2771(ESI⁺)andm/z554.2615(ESI⁻).

Thedesolvationandsourcetemperaturesweresetat200^◦Cand 90^◦C,respectively.Thecapillaryvoltagewasoperatingat3kV.Data wascollectedbetween100and1500Da,withascantimeof1sover ananalysistimeof20min.LC–MS/MSanalyseswereperformed withcollisionenergiesof25eV(formassvalues<400Da)and40eV (formassvalue>400).Themassdatawereacquiredandprocessed withtheMassLynxV4.1software(Waters)

Ipomoeapes-capraeextractsemisolidpreparation

Acreambasedon:12%Hostacerin^®CG,0.7%Phenonip^®,0.1%

EDTA,0.9%Polymol^®,0.003%BHT,5%propyleneglycolandpuri- ﬁedwater.Thisformulationwaspreparedwithorwithoutthesoft extractat 1%(w/w).The semisolidswerepackedin aluminium tubes,withasealandpolypropylenecover.

Pharmacologicalassays

MaleSwissmice(25–30g;8-week-old)wereusedforthisstudy (6–10Swissmicepergroup).Animalswerehousedundercondi- tionsofoptimumlight,temperature,andhumidity(12-hlight-dark cycle,22±1^◦C),withfoodandwateradlibitum.Alltheprocedures usedinthepresentstudyfollowedthe“PrinciplesofLaboratory AnimalCare”ofNationalInstitutesofHealthpublicationn^o85-23 andwereapprovedbytheAnimalEthicsCommitteeofUNIVALI (protocolnumbers024/2012).

Forspontaneousnociceptioninducedbytrypsin,themiceswere topicallypre-treated withthesemisolidcontainingI.pes-caprae extract(1%),ethylcaffeate(0.15mg/g)orplacebo 30minbefore thetest.Then,theyreceiveda20␮l-injectioncontainingtrypsin (300␮g/paw) (Paszcuk et al., 2008)into the plantar surface of therighthind paw.Followed theinjection, theywereindividu- allyobservedforaperiodof10min.Theamountoftimespentby themouselickingorbitingtheinjectedpawwasobtainedwitha chronometerandconsideredindicativeofnociception.

For pawoedemainduced bytrypsin, theanimals werepre- treatedwithI.pes-capraesemisolid(1%) and,30minlater, they receiveda 50␮l-intraplantar-injectionof saline0.9%containing trypsin(30␮g/paw)intotherighthindpaw(Paszcuketal.,2008).

The contralateral hind paw (left paw) received 50␮l of saline and was used as control. Paw oedema was measured using a plethysmometer (Ugo Basile) at different time points afterthe phlogistic-agentinjectionandwasexpressedinmicroliterasthe differencebetweentherightandleftpaws.

Theresultsarepresentedasthemean±SEMof6–12animals ineachgroup.ThepercentagesofinhibitionwerebasedonAUC (areaundercurve)forthelinegraphs,calculatedusingtheentire timecourseofeachexperimentandreportedasthemean±SEM of inhibitionsobtainedfor each individualexperiment. Statisti- calcomparisonofthedatawasperformedbytwo-wayanalysisof variance(ANOVA)followedbyBonferroni’spost-testandone-way ANOVAfollowedbyDunnett’spost-test.p-valuesless than0.05 (p<0.05)wereconsideredsigniﬁcant.

Table1

Quantiﬁcation(average±standarddeviation)ofsecondarymetabolitesofIpomoea pes-capraesoftextractsobtainedwithdifferentstrengthsofethanol.

Markers (mg/g)

Ethanol(^◦GL)

50 70 90

Peak1*ISA 5.17±0.22 3.67±0.16 3.35±0.26 Peak2*ISA 9.51±0.32 25.10±0.81 24.30±1.42 Peak3*ISA 8.21±0.53 5.22±0.06 3.39±0.08 Peak4*ISA 3.44±0.32 1.61±0.05 1.32±0.09 Peak5**ISQ 5.54±0.55 12.08±0.60 12.44±0.73

ISQ 6.55±0.59 11.29±0.16 10.95±0.26

ISB 2.46±0.97 13.03±0.29 13.27±0.52

ISA 3.64±0.67 44.15±2.54 61.61±2.32

Peak6**ISQ 11.13±1.50 13.44±0.41 15.92±1.74 Peak7**ISQ 12.28±1.08 26.94±1.40 43.43±0.93

ISC nd 12.65±1.46 11.07±0.53

Peak8*ISA(mg/g) 11.83±0.24 3.32±0.13 2.04±0.20 nd:undetected;*ISA:assayofpeakwithUVabsorptionprofiletypicalofphenolic compounds,expressedinISA;**ISQ:assayofpeakwithUVabsorptionprofiletypical offlavonoidcompounds,expressedinISQ.

Resultsanddiscussion

Inthisstudy,weobtaineddifferentextractsfromI.pes-caprae aerialpartsbyemployingthreedifferentconcentrationofethanol (50,70and90^◦GL)forextraction.Thechoiceofapolarsolventsuch asethanolaimedtothefutureestablishmentofaphytomedicine aswellastheproximitytothepopularusethatemploysthejuice oftheleaves(Manigaunhaetal.,2010).

The obtained extracts were then submitted to the chro- matographicprofileandsecondarymetabolitesquantificationby HPLC-UV(Fig.1)usingmethodologypreviouslyvalidated(Dutra et al.,2014).The extractobtainedwithethanol 50^◦GL (Fig.1A) shows thepresence ofsevenunknownpeaksrelatedtopheno- liccomponents(peaks1–4withsimilarUVabsorptionprofileof chlorogenicacidderivativesISA,ISBandISC)orflavonoids(peaks 5, 6and 7,withsimilarUV absorptionprofileof isoquercitrin), besidesthepresenceofchlorogenicacidderivative(ISA,ISB)and isoquercitrin(ISQ).Themaximumabsorbanceofthephenoliccom- pounds wasat215, 240, 290and 325nm, whiletheflavonoids exhibitmaximumofabsorptionat210,254and350nm(inserts ofFig.1A).

Theextractobtainedwith70^◦GLethanol(Fig.1B)alsoshows thepresenceoftheabovecompounds,butingreaterconcentration, includingthepresenceofISC,aspreviouslydetectedbyBarthetal.

(2017).Similarproﬁlewasobservedforthe90^◦GLethanolextract (Fig.1C).However,anadditionalpeakwasobservedattheendof thechromatographicrun,at48min,especiallyinthe50^◦GLextract (Fig.1A),anditschromatographicareadecreaseswiththeincrease ofethanolstrength(Fig.1BandC).

Theethylcaffeatepeakshowedthesameretentiontimeand UVabsorptionproﬁlesimilartotheadditionalpeakthatappeared in theextract chromatograms,more pronouncedin theextract obtainedwith50^◦GLethanol(Fig.1A–D).

Due to the large presence of caffeic acid derivatives in the plant, this latter peak was suspected to be an artefact related tocaffeicacid.Thispresumptionwascheckedbycomparingthe extractchromatogramwiththatofethylcaffeateusingthesameLC gradient(Fig.1D).Ethylcaffeateispresumablyaproductoftrans- esteriﬁcationofcaffeicacidestersandethanol,formedduringthe extractionprocess.

Eachsecondarymetaboliteinthedifferentextractswasquan- tiﬁed,includingthesupplementalpeakat48min(calledpeak8, suspectedtobeethylcaffeate)byexternalstandardization,using thepreviousvalidatedmethod(Dutraetal.,2014).Theresultsare presented in Table1.The 50^◦GL ethanol extract showedlower amountofcaffeicacidderivativesandsomepeakswithphenolic-

(5)

Fig.2. (A)Scoreplotofthesamples;(B)LoadingsonPC1pervariable;(C)LoadingsonPC1versusLoadingsonPC2.

UV-absorptionproﬁle.Ontheotherhand,thisextractshowedthe highestareaofpeak8,whichwasquantiﬁedandexpressedinISA (Table1).

PCAwithmeancenteringwasemployedonthedatasetconsti- tutedoftwelvesamplesdescribedbytwelvevariables.According toFig.2,onlytwoPCwerenecessarytoseparatethethreediffer- entclassesofalcoholicstrength(50,70and90^◦GL).Analysingthe Fig.2A(constructedbasedontheﬁrsttwoprincipalcomponents, accountedformorethan99%ofthetotalvariance),PC1succeeded in discriminateextracts obtainedwith50^◦GLethanol of others (negativescorescorrespondtosamplesof50^◦GLethanol).Onthe otherhand,PC2discriminatedtheextractsof70and90^◦GLethanol (positivescores correspondto extract 90 and negative onesto extract70^◦GLethanol).

ExaminingtheloadingsplotofPCA(Fig.2BandC),itisclear thatthevariablesdirectlyrelatedtotheextractof50^◦GLethanol arethepeak8andthepeak3.Thus,theextractof50^◦GLethanol containshighamountofpeak8andpeak3(Fig.2B).Ontheother hand,thevariablesISA,peak7,peak2andISCarealsoimportantto discriminatethesamples,but,byhavingpositivevaluesofloadings, theyareinsmallquantityinextractsofethanol50^◦GL.Inthisway, thislatterextractischaracterizedbyhigherlevelsofpeak8and3 (comparedtotheothers)andlowlevelsofISA,peak7,peak2and ISC.

Todiscriminatethesampleswith70and90^◦GL(Fig.2C),the variables that are most distinctive are peak 7 and peak 2 (ISC isalsoimportant).Extractof90^◦GLethanolhavehigheramount ofpeak 7 andlower contentof peak2 (andISC). On theother hand,extractof70^◦GLethanolhashigheramountofpeak2(and ISC)andlowercontentofpeak7.Thus,thepeak8seemstobe themore important compoundsrelated to thebiological effect observed.

With the aim to illustratethe difference among the chem- ical composition of the three extracts, UPLC-QTOF/MS analysis was performed. The extracts were exclusively sensitiveto the negative mode ionization and the base peak ion (BPI) current chromatogramfromthethreeextractsisshowinFig.3.Thediag- nosticoftheLC-ESI-MSdatashowedthepresenceofquinicacid, 4,5-di-O-caffeoylquinicacid,3,5-di-O-caffeoylquinicacid, 3,4-di- O-caffeoylquinicacid,andisoquercitrin,corroboratingwithdata obtainedfromtheHPLC-UVanalysis(Fig.1).The caffeoylquinic acidderivatives revealed in their tandem mass spectrasimilar fragmentationpatternandtheobservedionswerem/z353.0809 [M-H-caffeoyl]⁻,m/z191.0501[quinicacid-H]⁻and179.0285[caf- feicacid-H]⁻.SimilarMS²patternwerereportedintheliterature forthesamemetabolites(Ncubeetal.,2014).Isoquercetrinwas proposedas aconstituentbased onitsfragments m/z301.0354 andm/z151.0037, whichwerein accordancetothosereported intheliterature(Simirgiotis,2013).However,itisworthtohigh- lightthepresenceofauniquepeakat7.32min,withm/z207.0492 [C₁₁H₁₂O₄-H]⁻thatstandsoutintheextractobtainedwith50^◦GL ethanol(Fig.3A).Thesemetaboliteswerelightlyfoundin70^◦GL ethanol extract (Fig. 3B), and absent in 90^◦GL ethanol extract (Fig.3C).Thepeakat7.32minwasidentifiedasethylcaffeatebased onitsMS² spectrum,whichshowedafragmentatm/z161.0224 [M-H-ethanol]⁻.Theforegoingdataconfirmthequalitativeanaly- sisbyHPLCandidentifiedthepeak8thatdistinguishestheextract of50^◦GLethanolfromthe70and90^◦GL,beinganartefactgenerated duringtheextraction/evaporationprocess(assuggestedinFig.3F).

Thiscanbeexplainedbythefactthatesteriﬁcationisalready a wellknownreactionthatoccursbetweenacarboxylicacidandan alcohol,beingfavoredbytheincreaseintemperature.Itisatyp- icalnucleophilicsubstitutionreactiononacylcarboninwhicha nucleophile(ethanol)attacksthecarbonofcarbonylgroup(caffeic

(6)

Fig.3.Basepeakchromatogram(negativeionizationmode)oftheextractsobtainedwith:(A)90^◦GLethanol,(B)70^◦GLethanoland(C)50^◦GLethanol,(D)methanol,(E) acetone,(F)ethylcaffeateformation.

acid)leadingtotheformationoftheester(Claydenetal.,2007;

Campos-Buzzietal.,2009).

Toverifythehypothesisofartefactformation,theleavesand stemsofI.pes-capraewereextractedwith50^◦GLmethanoland acetone,oncethesesolventscouldnotproduceethylcaffeatefrom caffeicacidderivatives.Infact,thepresenceofethylcaffeatewas notdetectedintheseextracts(Fig.3DandE).

However,itisimportanttounderstandhowthephytochemical alterationintheextractaffectsitsbiologicalactivity.Aspreviously reported(daCunhaetal.,2004),caffeicacidderivatives(includingethylcaffeate)exertedinvitroandinvivoanti-inﬂammatory activities by scavengingNO and alsomodulating iNOS expres- sion. In addition, antinociceptive effectwas reported tocaffeic acidderivativesinaceticacid-inducedabdominalwrithesinmice (Campos-Buzzietal.,2009).Eventually,basedonthehypothesisof thatthepresenceofethylcaffeateisthemaindifferencebetween thethreeobtainedextracts,theywereincorporatedinasemisolid formulation,includingethylcaffeate,toverifythebiologicaleffect.

In a previouswork,wehave described thatthe Ipes-caprae extract obtained with ethanol 50^◦GL as solvent blocked the mechanicalhypersensitivityinducedbyPhysaliaphysalisvenom (Barthetal.,2017).Furthermore,thesameextractwaseffective inreducingthenociceptionandpaw-oedemainducedbytrypsin,a proteasecapableofgeneratinginﬂammatoryresponsesbytheacti- vationofproteasereceptor2(PAR-2).AsshowninFig.4(A–C),the extractsobtainedwithethanol70and90^◦GLdidnotpresentbio- logicalactivityinbothpaw-oedemaandspontaneousnociception inducedbytrypsininmice.

Ontheotherhand,whenethanol50^◦GLwasusedtoobtainthe extract,itwasobservedsigniﬁcantinhibitionofthepaw-oedema and nociception, withinhibition values of45±9% and 24±3%, respectively. As expected, and accordingly withliterature data (Campos-Buzziet al.,2009), ethylcaffeatepresentedprominent anti-inﬂammatoryeffect,withinhibitionof86±5%.However,it seemsnottoberelatedwiththeanti-nociceptionexertedbythe extract(50^◦GLethanol).

(7)

Fig.4.InfluenceofextractionprotocolinthebiologicalactivityoftheI.pes-capraeextract.(AandB)Anti-inflammatoryand(C)anti-nociceptiveeffectsofI.pes-caprae semisolids(1%),obtainedwithetanol50^◦GL,70^◦GLor90^◦GL,orsemisolidcontainingethylcaffeate(0.15mg/g)ontrypsininducedpaw-oedemaandspontaneousnociception inmice.Eachgrouprepresentsthemeanof6–12miceandtheverticallinesindicatetheS.E.M.Significantlydifferentfromcontrolvalues*p< 0.05,**p<0.01and***p<

0.001.(twe-wayANOVAfollowedbytheBonferroni’sposthoctestandone-wayANOVAfollowedbytheDunnett’sposthoctestforcolumngraph).Panels‘B’representsthe AUCoftherespectivelinegraph‘A’.

Conclusions

Thehighamountoftheartefactintheextractofethanol50^◦GL couldbe, at least in part,the responsible for thetopical anti- inflammatory activityofI.pes-caprae ethanolextract. Together, thehereinresultsdemonstratethattheextractionmethodologyis extremelyimportanttodeterminethebiologicalactivityofthefinal herbalproductandthechromatographicfingerprintshavebeen usedasanimportanttooltoqualitycontrolofherbalmedicines.

Funding

This work was financed by FUMDES (Fundo de Apoio à ManutençãoeaoDesenvolvimentodaEducaçãoSuperiordoEstado deSantaCatarina),CAPES,CNPqandFAPESC(FundacãodeAmparo àPesquisaeInovaçãodoEstadodeSantaCatarina).

Authors’contributions

Theauthorscontributionwasasfollow:CSB, HGTSandLWR performedthe pharmacological and technological experiments, CRSMandCAperformedthechemometricanalysis,TB,AMandLPS performedtheLCanalysis,RN,VCFandTMBBadvisedthechemi-

calstudies,AGCadvisedthetechnologicalstudies,JRSandNLMQ advisedthepharmacologicalstudies.Alltheauthorshavereadthe ﬁnalmanuscriptandapprovedthesubmission.

Conﬂictsofinterest

Theauthorsdeclarenoconﬂictsofinterest.

Acknowledgment

The authors wish to thank Prof. Dr. Edemilson Cardoso da Conceic¸ão(UniversidadeFederaldeGoiás)forthesuggestionsthat motivatedthisworkandtoHUEMtokeeptheexsiccate.

References

Ashnagar, A., Shiri,A., 2011. Isolationand characterization of2-hydroxy-1,4- naphthoquinone(lawsone)frompowderedleavesofhennaplantmarketedin AhwazCityofIran.Int.J.Chem.Tech.Res.3,1941–1944.

Barth,C.S.,Tolentino,H.G.S.,Rocha,L.W.,daSilva,G.F.,dosAnjos,M.F.,Pastor,V.D., Bresolin,T.M.B.,Couto,A.G.,Santin,J.R.,Quintão,N.L.M.,2017.Ipomoeapes- caprae(L.)R.Br(Convolvulaceae)relievednociceptionandinﬂammationin mice—atopicalherbalmedicineagainsteffectsduetocnidarianvenom-skin contact.J.Ethnopharmacol.200,156–164.

(8)

Britton,R.,Roberg,M.,Berisch,H.,Andersen,R.J.,2001.Antimitoticditerpenoids fromErythropodiumcaribaeorum:isolationartifactsandputativebiosynthetic intermediates.TetrahedronLett.42,2953–2956.

Campos-Buzzi,F.,Franzoi,C.L.,Antonini,G.,Fracasso,M.,CechinelFilho,V.,Yunes, R.A.,Niero,R.,2009.Antinociceptivepropertiesofcaffeicacidderivativesin mice.Eur.J.Med.Chem.44,4596–4602.

Clayden,J.,Greeves,N.,Warren,S.,Wothers,P.,2007.OrganicChemistry.Oxford UniversityPressInc.,NewYork.

DaCunha,F.M.,Duma,D.,Assreuy,J.,Buzzi,F.C.,Niero,R.,Campos,M.M.,Calixto,J.B., 2004.Caffeicacidderivatives:invitroandinvivoanti-inﬂammatoryproperties.

FreeRadic.Res.38,1241–1253.

Dutra,D.M.,Barth,C.S.,Block,L.C.,Quintão,N.L.M.,Couto,A.G.,CechinelFilho,V., Bresolin,T.M.B.,2014.Simultaneousdeterminationoffourphenoliccompounds inextractsofaerialpartsofIpomoeapes-caprae(L.)R.Br.(Convolvulaceae)by HPLC-UV.Quim.Nova37,1510–1514.

Brasileira,Farmacopeia,2017.2ôSuplementodaFarmacopeiaBrasileira,5âedição.

AgênciaNacionaldeVigilânciaSanitária,Brasília.

Jaikang,C.,Chaiyasut,C.,Narongchai,P.,Niwatananun,K.,Narongchai,S.,Kusirisin, W.I.,2011.Inhibitoryeffectsofcaffeicacidesteranaloguesonfreeradicalsand humanlivermicrosomeCYP1A2activities.Med.Chem.(LosAngeles)7,99–105.

Krogh,R.,Kroth,R.,Berti,C.,Madeira,A.O.,Souza,M.M.,CechinelFilho,V.,Delle- Monache,F.,Yunes,R.A.,1999.Isolationandidentiﬁcationofcompoundswith antinociceptiveactionfromIpomoeapes-caprae(L.)R.Br.Die.Pharmazie.54, 464–466.

Maltese,F.,vanderKooy,F.,Verpoorte,R.,2009.Solventderivedartifactsinnatural productschemistry.Nat.Prod.Comm.4,447–454.

Manigaunha,A.,Ganesh,N.,Kharya,M.D.,2010.Morningglory:anewthirstin- searchofde-novotherapeuticapproach.Int.J.Phytomed.2,18–21.

Ncube,E.N.,Mhlongo,M.I.,Piater,L.A.,Steenkamp,P.A.,Dubery,I.A.,Madala,N.E., 2014.Analysesofchlorogenicacidsandrelatedcinnamicacidderivativesfrom

NicotianatabacumtissueswiththeaidofUPLC-QTOF-MS/MSbasedonthein- sourcecollision-induceddissociationmethod.Chem.Cent.J.8(66),1–10.

Paszcuk,A.F.,Quintão,N.L.M.,Fernandes,E.S.,Juliano,L.,Chapman,K.,Andrade- Gordon, P., Campos, M.M., Vergnolle, N., Calixto, J.B., 2008. Mechanisms underlyingthenociceptiveandinﬂammatoryresponsesinducedbytrypsinin themousepaw.Eur.J.Pharmacol.581,204–215.

Pereda-Miranda,R.,Escalante-Sánchez,E.,Escobedo-Martínez,C.,2005.Charac- terizationoflipophilicpentasaccharidesfrombeachmorningglory(Ipomoea pes-caprae).J.Nat.Prod.68,226–230.

Pongprayoon,U.,Baeckstrom,P.,Jacabsson,U.,Lindstrom,M.,Bohlin,L.,1991.Com- poundsinhibitingprostaglandinsynthesisisolatedfromIpomoeapes-caprae.

PlantaMed.57,515–518.

Pongprayoon,U.,Baeckström,P.,Jacobsson,U.,Lindström,M.,1992a.Inhibitionof ethylphenylpropiolateinducedratearoedemabycompoundsisolatedfrom Ipomoeapes-caprae(L.)R.Br.Phytother.Res.6,104–107.

Pongprayoon,U.,Baeckstrom,P.,Jacobsson,U.,Lindstrom,M.,Bohlin,L.,1992b.

Antispasmodicactivityofß-damascenoneandE-phytolisolatedfromIpomoea pes-caprae.PlantaMed.58,19–21.

Simão-Bianchini, R., Ferreira, P.P.A., Disponivel em:

http://ﬂoradobrasil.jbrj.gov.br/jabot/ﬂoradobrasil/FB7051,2015.

Simirgiotis,M.J.,2013.AntioxidantcapacityandHPLC-DAD-MSproﬁlingofChilean peumo(Cryptocaryaalba)fruitsandcomparisonwithGermanpeumo(Crataegus monogyna)fromSouthernChile.Molecules18,2061–2080.

Sunthonpalin,P.,Wasuwat,S.,1985.Jellyﬁshdermatilistreatedbytheextractof Ipomoeapes-caprae.SirirajHosp.Gaz.37,329–338.

Teramachi,F.,Koyano,T.,Kowithayakor,T.,Hayashi,M.,Komiyama,K.,Ishibashi,M., 2005.CollagenaseinhibitoryquinicacidesterfromIpomoeapes-caprae.J.Nat.

Prod.68,794–796.