Monitoring oxidative stability and phenolic compounds composition of myrtle-enriched extra virgin olive during heating treatment by flame, oven and microwave using reversed …

ContentslistsavailableatScienceDirect

Industrial

Crops

and

Products

j ou rn a l h o m epa g e :w w w . e l s e v i e r . c o m / l o c a t e / i n d c r o p

Monitoring

oxidative

stability

and

phenolic

compounds

composition

of

myrtle-enriched

extra

virgin

olive

during

heating

treatment

by

flame,

oven

and

microwave

using

reversed

phase

dispersive

liquid–liquid

microextraction

(RP-DLLME)-HPLC-DAD-FLD

method

S.

Dairi

_,

_T.

_{Galeano-Díaz}

_,

_M.I.

_{Acedo-Valenzuela}

_,

_M.P.

_{Godoy-Caballero}

_,

F.

Dahmoune

_,

_H.

_Remini

_,

_K.

_Madani

a_{LaboratoiredeBiomathématique,Biophysique,Biochimie,etScientométrie,FacultédesSciencesdelaNatureetdelaVie,UniversitédeBejaia,}

06000Bejaia,Algeria

b_{UniversityofExtremadura,AnalyticalChemistrydepartment,Badajoz,Spain}

a

r

t

i

c

l

e

i

n

f

o

Articlehistory: Received9July2014 Receivedinrevisedform 16November2014 Accepted18November2014 Keywords:

Extravirginoliveoil Phenoliccompounds Heating Myrtuscommunis Lipidoxidation Functionalfood

a

b

s

t

r

a

c

t

Lipidsoxidationisoneofthemainwaysthataffectthenutritionalqualityofextravirginoliveoil(EVOO) bylosingitsphenolicsubstancesandespeciallyduringheatingtreatments.Thisstudywasfocusedon theevaluationoftheeffectofMyrtuscommunisphenoliccompounds-enrichedextravirginoliveoil (McPC-EEVOO)onphenoliccompoundscompositionduringflame,ovenat180◦Candmicrowaveheating, atdifferentexposuretimes,evaluatedbyreversedphasedispersiveliquid–liquidmicroextraction (RP-DLLME)-HPLC-DAD-FLD.TheK232andK270werealsoevaluatedforallheatingtreatments.Theobtained resultsshowedthattheenrichmentofEVOObymyrtleextractssignificantlypreventstheconsumption ofendogenousphenoliccompoundfromEVOOasphenolicalcoholandflavonoidsincomparisontothe control(EVOOwithoutenrichment).Themostprotectiveeffectwasfoundduringflameandmicrowave heating.TheK232valuesweresignificantlyreducedduringflameheatingcomparedtothecontroland followedbyovenheating,althoughK232valuesinmicrowaveheatingweresimilarforalltheoilexamines. K270wasnotaffectedbythisenrichmentofEVOO.

©2014ElsevierB.V.Allrightsreserved.

1. Introduction

Inrecent years, theMediterranean diethasbecome

increas-inglypopular,gainingwidespreadattentionamongthenutrition

andresearchcommunities(HuangandSumpio,2008)and their

replaceableelementin thisdietarystyleisextravirginoliveoil

(EVOO).

ThebeneficialpropertiesofEVOOaremainlyattributedtoits

composition,havingahighpercentageofmono-unsaturatedfatty

acids(oleic acid:C18:1, n9) and significantamountsof minor

componentsasphenoliccompounds(phenolicacids,phenolic

alco-holssuchashydroxytyrosolandtyrosol,secoiridoidsderivatives

andflavonoids,suchluteolinandapigenin)(Alloucheetal.,2007;

Benganaetal.,2013).EVOOphenoliccompoundscontributetothe

∗ Correspondingauthor.Tel.:+213661141500. E-mailaddress:sofianedairi@yahoo.fr(S.Dairi).

antioxidantandbiologicalproperties(TuckandHayball,2002),

oxi-dationresistance,bitterandpungenttasteofEVOO(Bendinietal.,

2007;Benganaetal.,2013).

Virginoliveoil(VOO)maybeconsumedrawasaningredient

intoast, salads,andotherfoodstuffs,although,in addition,itis

highlyconsumedafterdomesticheating,suchasfrying,boiling,and

microwave(Carrasco-Pancorboetal.,2007).Nevertheless,several

studieshavefocusedonpossibledegradationandlossesofphenolic

compoundspresentinEVOOduringdifferentheatingprocedures

by a complex series of chemical reactionsincluding oxidation,

hydrolysis,andpolymerization(Santosetal.,2013).Beˇsteretal.

(2008)haverecentlyfoundthatthetotalphenolcontentdecreased

by55–60%afterheattreatmentat100◦Cfor142h,withanairflow

of10L/h.Regardingsinglecompounds,Carrasco-Pancorboetal.

(2007)reportedthathydroxytyrolanditsderivativesofEVOO

suf-feredsignificantlossesduringheatingat180◦C.Likewise,Brenes

et al.(2002)foundbetween 20and 30% ofhydroxytyrollosses

inEVOOfromSpanishcultivarsafter5and10minofmicrowave

http://dx.doi.org/10.1016/j.indcrop.2014.11.035

heating.In theliterature,a hugedisparityandconflictedresults

werefoundregardingthecomparisonofmicrowaveheatingeffect

andothersprocessesonoliveoilquality(Santosetal.,2013).

Therefore,phenoliccompoundspresentinEVOOdonotremain

enoughto ensurea good stability during heating. For this, the

enrichmentofoliveoilcomplementingtheolivephenolswithother

kindsofphenoliccompoundscanbeagoodstrategytoimprovethe

oxidativecapacityandnutritionalprofileoftheoliveoil,givingrise

tothedevelopmentofnewfood(Rubioetal.,2012).The

incorpo-rationofthymeextractinEVOOincreasedtheoxidativestability

andtheantioxidantactivityofEVOO(Rubioetal.,2012)andithas

beenalsoobservedanenhancingeffectofthymecompoundsonthe

bioaccessibilityofsecoiridoidsfromoliveoil(Rubioetal.,2014).

Regardingsimulatedhouseheatingeffectonenrichedoil,few

studiesarefoundintheliterature.TherecentworkofMalheiro

etal.(2012)showedthattheteaextractsprotectoliveoilfrom

theoxidativeprocess(until3min),butwithhigherheatingperiods

theextractswerepro-oxidants.Thismeansthatthereisacomplex

interactionoccurringinacomplexmixtureofantioxidantswhich

maybetakenintoaccountwhendesigningafunctionalfood.Tothe

bestofourknowledge,thereisanyworkstudyingtheeffectofthe

co-occurringEVOOandplantextractonthecompositionofEVOO

phenoliccompoundsduringdifferentheatingprocesses.

Myrtle,MyrtuscommunisL.,isarichsourceofantioxidant

com-poundsandpossessesstrongantioxidantproperties(Dairietal.,

2014). The main goal of this work is to evaluate for the first

timetheeffectofmyrtleextract,obtainedbytwodifferent

meth-ods,microwave-assistedextractionandconventionalone,onthe

preservationofEVOO phenoliccompoundsconsumption during

threeheatingprocedures.Theheatinginabutane-airflameheating

(2,3,5and10min)orinovenat180◦C(1,2and3h)wasassayed.

Thelastheatingtreatmentwasappliedbyusingforthefirsttime

scientificmicrowave(5,10,15and20min).Theevolutionofthe

phenoliccompoundscontentofEVOOwasmonitoringbyreversed

phasedispersiveliquid–liquidmicroextraction

(RP-DLLME)-HPLC-DAD-FLDmethod.TheK232andK270werealsodeterminedforall

heatingconditions.

2. Materialsandmethods

2.1. Materials

EVOOsampleswerefromAlgerian“Chemlal”cultivarolives,

col-lectedintheregion ofBejaia (Algeria).Theoils wereextracted

by a three phases centrifugation system from mid-ripe olive

fruit.

2.2. Chemicalsandstandards

Forallexperiments,analyticalreagentgradechemicalsand

sol-ventswereused.UltrapurewaterwasobtainedfromaMillipore

Milli-QA10 System (Waters,Germany). Hydroxytyrosol (HYTY),

luteolin(LUT)andapigenin(APIG)wereobtainedfrom

Extrasyn-these(Genay,France);gallicacid(GAL) tyrosol(TY),p-coumaric

acid(p-CUM),ferulicacid(FER)andmyricitrinanalyticalstandards

fromSigma-AldrichChemieGmbH(Steinheim,Germany).All

sol-ventsemployedwereHPLCgrade;ethanolandacetonitrilewere

providedbyPanreac(Spain),aceticacidbyRomilChemicalsLtd

(England).

2.3. Plantmaterialpreparation

M.communisleaveswerecollectedintheregionofBejaia

(Alge-ria),and dried in anoven at 40◦C until constant weight, then

crushedandsievedtohaveasizelessthan125␮m.Thesamples

werestoredinthedarkatroomtemperature.

2.4. Extractionprocedure

2.4.1. Microwaveassistedextraction(MAE)method

Theextractionprocedurewaspreviouslydescribedinour

pre-viousstudy(Dairietal.,2014).Briefly,domesticmicrowaveoven

(NN-S674MF, LG,Japan, 32L, 1000W; variable in 100W

incre-ments,2.45GHz)modifiedinourlaboratorywasusedforextraction

ofmyrtleleavesphenoliccompound(McPCs).Theleave

suspen-sion(1gofleavespowderper20mLof50%EtOH)wasirradiated

bymicrowaves(700Wofpower)for1minaccordingtothe

fol-lowing cycle: 45s power-on, 10s of power-off and again 15s

power-on.Thesamplewasfilteredwithasinteredglassat0.45␮m

usingavacuumpump.Twoothersadditionalextractionswere

car-riedoutforrecoveringthetotalityofPCsandthethreefractions

werecombined.Then,thesolventwasevaporatedtoobtainadry

extract.

2.4.2. Conventionalextraction(CE)method

20mLof50%-ethanolwereaddedto1gofmyrtleleavespowder

andletmacerateduring60minwithmagneticagitation.Afterthat,

theprocesswasthesameasforMAEmethod(Dairietal.,2014).

2.5. Preparationofenrichedextravirginoliveoil

Weighed quantities of myrtle extracts weredissolved in an

appropriatevolumeofethanol(50%).Then,anappropriatevolume

ofthese(500␮L),togive200mgGAEkg−1 ofoil,wereaddedto

extravirginoliveoil(500g)bystepwiseundervigorousstirringfor

30min.Oilsampleswerestoredinthedarkuntiluse(Bouazizetal.,

2008).

2.6. Reversedphasedispersiveliquid–liquidmicroextraction

(RP-DLLME)ofphenoliccompoundsfromEVOO

TheextractionofphenoliccompoundsfromEVOOwascarried

outbyDLLME,andtheirquantificationwasmadebyemployingthe

standardadditioncalibrationmethod.Forthis,2gofEVOOspiked

withincreasinglevelsofthephenoliccompoundswereweighed

in atest tubewithconical bottom and1mLof1,4-dioxane (as

dispersersolvent) plus 150␮Lof ethanol:water 60:40 (v/v) (as

extractionsolvent) wereinjectedrapidlyintotheEVOOsample

using a 2mL syringe. The mixture was then shaken in a

vor-texapparatusfor1minand centrifugedfor15minat4000rpm

usinga Mixtaselcentrifuge. Thedispersedfinedropsof

extrac-tion solvent were separated and settled at the conical bottom

oftheconicaltesttube.Theseparatedphasewastakenusinga

syringeanddirectlyinjectedinthechromatographicsystemfor

itsanalysisintheoptimized conditions(Godoy-Caballeroetal.,

2013).

2.7. Identificationandquantificationofphenoliccompoundsof

EVOOandenrichedEVOObyRP-DLLME-HPLC-FLD-DAD

ThechromatographicstudieswereperformedusinganAgilent

Model1100LCinstrument (AgilentTechnologies,Palo Alto,CA,

USA)equippedwithdegasser,quaternarypump,columnoven,auto

samplerAgilent1290infinitythermostatizedat5◦C,UV–vis

diode-arraydetector(DAD),rapidscanfluorescencedetector(FLD),and

theChemstationsoftwarepackagetocontroltheinstrument,and

fordataacquisitionandanalyses.Theanalyticalcolumnemployed

wasa120ODBS5␮m,150mm×4.6mm(Teknokroma,Barcelona,

Spain).Thecolumntemperaturewassetat25◦C.Themobilephase

componentswerehigh-puritywater/0.5%aceticacid/0.1%

acetoni-trile(phaseA)andacetonitrile/0.5%aceticacid(phaseB)andwere

10%B;20–30min,45%B;30–45min,45% B;45–50min,10%B;

50–65min,10%B.Then, thecolumnwasre-equilibratedduring

15minintheinitialconditions.Theflowratewassetconstantat

0.5mLmin−1andtheinjectionvolumewas10␮L.

Identification of individual polyphenols was carried out by

usingretention times and spectrometricand fluorescencedata.

ThedetectionwasdirectlyperformedbyHPLC-DADatthe

max-imumabsorbancewavelengthforeachpolyphenol.Quantification

ofthesinglepolyphenolswasdeterminedfromstandardaddition

calibrationcurvemethodusingafour-pointregressioncurveand

analyticalfigureswereperformedbymeansoftheACOCprogram

inMatLabcode(Godoy-Caballeroetal.,2012).

2.8. Measurementofspecificabsorbancecoefficient(K232and

K270)

Coefficientsofspecificextinctionat232and270nm(K232and

K270) were determined according to European Union standard

methods(CommissionRegulation(EEC)no.2568/91).Theoil

sam-plesweredilutedinisooctaneplacedintoa1cmquartzcuvette,

andtheirabsorbancevalues weremeasuredatthewavelengths

270and232nm,againstablankofisooctane.Threereplicateswere

preparedandanalyzedpersample.

2.9. Heatingproceduresofun-enrichedandenrichedextravirgin

oliveoil

2.9.1. Conventionalheatingbyflame

Tosimulatetheconditionsofheatinginhousecooking

prac-tices,samples(5g)ofeachEVOOsamplewereweighedinPyrex

beakersof 42mm diameterand placed onthecenterof a

sup-portandwereheatedbyflamefordifferentexpositiontimes,0,

2,3,5and10min.Threeexperimentswerecarriedoutforeach

sample under the same conditions. After each heating period,

theoiltemperaturewasdeterminedwitha thermometer.After

cooling, the samples were stored in sealed tubes at 6◦C until

analysis.

2.9.2. Conventionalheatinginovenat180◦C

Tosimulatefryingconditions,ninealiquots(5g)ofeachsample

wereweighedinPyrextubesandsubjectedtoconventional

heat-ingat180◦CinaHeraeusdryingoven.Thetubeswereremoved

fromtheovenatfixedintervalsof60min,obtainingsampleswith

differentheatingtreatments(60,120,180min)tobeanalyzed.All

heatedsampleswerecooledatroomtemperature(23±1◦C)and

storedat6◦Cuntilchemicalanalysis.

2.9.3. Microwaveheating

AScientificmicrowaveoven(MilestoneETHOS(800W)with

NS14100ATC-FONS14stopperwithquartzthermowell)wasused

toheatingtreatmentofextravirginoliveoil(EVOOor EEVOO).

To simulatethe averagequantity and conventional times used

in home cooking, three aliquots (5g) of each oil were placed

inPyrex beakers(42mmdiameter)attheovenandexposedto

microwaveirradiationof2450Hzfrequencyatmaximumpotency

(1000W)for differentmicrowavingtimes,5,10,15and20min.

Theachievedtemperaturesofheatedoilsweremeasured

imme-diately aftermicrowave exposure, by insertinga thermocouple

(connectedtoanacquisition system)atapproximatelythe

geo-metricalcenterofthesample.Allheatedsampleswereallowedto

coolatroomtemperaturefor60minafterthermaltreatmentand

beforechemicalanalysis.Thethree5galiquotsofeachoilwere

combinedaftermicrowavinginorder toobtaina homogeneous

sample. The sampleswere keptunder refrigeration (4◦C) until

analysis.

Fig.1.DADandFLDchromatogramsofun-enrichedandenrichedEVOObymyrtle extractsubjectedtotheRP-DLLMEanddirectlyinjected.Galloylquinicacid(1),gallic acid(2),hydroxytyrosol(3),tyrosol(4),secoiridoid1(5),secoiridoid2(6),myricitrin (7),luteolin(8),apigenin(9).Similarchromatogramprofilewasobtainedformyrtle CE-enrichedEVOO.

3. Resultsanddiscussion

3.1. Identificationandquantificationof“chemlal”EVOOphenolic

compounds

TheextractionofEVOOphenoliccompoundswascarriedout

usinganewmethoddevelopedbythelaboratorywherethiscurrent

studywasdone. Thus,areversed phasedispersive liquid–liquid

microextraction(RP-DLLME)procedureshavebeenusedwith

1,4-dioxane asdisperser solvent which is misciblewith extraction

solvent (ethanol/water) and withthe EVOOsample. This novel

extractiontechniqueenablestheincrease ofthecontactsurface

areabetweentheextractionsolventandEVOOsample,resulting

inabetterrecoveryefficiencyandshorterextractiontime(

Godoy-Caballeroetal.,2013).Themainphenoliccompoundsidentifiedin

EVOOareshowninFig.1issummarizedinTable1.Thephenolic

compoundsdetectedbelongingtophenolicalcohols(hydroxytyrol,

tyrosol),cinnamicacidderivatives(p-coumaricacid,ferulicacid)

andflavonoids(luteolinandapigenin).Thesecompoundidentified

wereinagreementwithpreviousworksstudyingthesame

culti-var(Benganaetal.,2013;Tamendjarietal.,2014).Twoanother

important peakswere eluted at 36.6and 36.9minrespectively

between luteolin and apigenin and showingthe same spectral

behavior(max280nm).Accordingtotheliterature(Benganaetal.,

2013;Capriottietal.,2014;Lozano-Sánchezetal.,2013)

concern-ingSpanishandAlgerianEVOO,someoleuropeinandligstroside

derivativeswereelutedbetweenluteolinandapigenin.Therefore,

Table1

IdentificationofphenoliccompoundfromAlgerianextravirginoliveoil“Chemlal”cultivarbyRP-DLLME-HPLC-DAD-FLD.

Phenoliccompound Retentiontime(Tr) Equation R2 _{Concentration(mgkg}−1_ofoil)

Quinicacidorderivative 4.01 – – –

Hydroxytyrol 10.09 Y=72.15x+246.1 0.999 4.12±1.57 Tyrosol 17.32 Y=35.98x+222.9 0.997 5.54±1.46 Luteolin 35.92 Y=365.61x+1340.8 0.997 3.67±0.77 Apigenin 38.47 Y=360.72x+589.74 0.996 1.63±0.60 P-coumaricacid 31.95 Y=0.6695x+32.531 0.993 0.05±0.04 Ferulicacid 32.64 Y=0.4119x+13.303 0.997 0.03±0.01 Secoiridoid1 36.65 – – – Secoiridoid2 36.94 – – –

(oleuropeinorligstrosidederivatives)whichoccurintheseEVOO at63–91mgkg−1ofEVOOasfoundintheworkofBenganaetal. (2013)studyingthesameEVOOvarietyatdifferentharvestdates.

The phenolic alcohols fraction was higher than flavonoids one

which is consistent with the finding of Bengana et al. (2013),

whereastheworkof(Tamendjarietal.,2014)studyingthesame

Algerian cultivars shows a different result. These observations

support the effect of various factors such as geographical

ori-gin,ripeningdegreeandextractionprocessonchemicalprofiles

andEVOOphenoliccompoundscomposition(Bendinietal.,2007;

Benganaetal.,2013).

TheanalysisofphenoliccompoundcompositionfromM.

com-munis leaves was carried out by HPLC-DAD-FLD in the same

conditionsutilizedfortheanalysisofEVOOphenoliccompounds

(datanot shown).The enrichmentof EVOOwithmyrtle extract

revealedthepresenceofnewcompoundsintheoilsampleascan

beseenin theFig.1,andthemainimportantcompoundswere

resumedinTable2.At280nm,thechromatogramreportedthe

presenceofanimportantpeakat4.1minwhich wastentatively

attributedtogalloylquinicacidfrommyrtleextract(Romanietal.,

2012)asmentionedaboveinourpreviousstudy(about7mg

gal-licacidequivalentpergramofdrymyrtleextract)(unpublished

data).However,thispeakoverlayswithanothersmallpeakfrom

EVOObothhavingthesameelutiontimeand spectralbehavior.

Intheliterature,thepresenceofapeakelutingbeforegallicacid

hasbeen reported in EVOO “Chemlal” variety and assigned to

quinicacidpresent ata concentrationof 3–4mgkg−1 (Bengana

etal.,2013).Thisletsupposethat thisinterfering peakmaybe

attributedtoquinicacidorderivative.Theconcentrationof

gal-loylquinicacidwasexpressedasgallicacidequivalent(GAE)from

standardadditioncalibrationcurvewithremovingthesignaldueto

theinterferingpeakfromEVOO.Gallicacidwasalsoidentifiedin

enrichedEVOOand inmyrtleextracts(datanotshown)

accord-ingtoitsretention timeand UV–vis datain comparisontothe

standard,andwaspresentata concentrationof1.48±0.76 and

1.68±0.80mgkg−1 forMyrtle MAE or CE-enrichedextravirgin

oliveoil(EEVOO)respectively.Ontheotherhand,whenthe

chro-matogramwasacquiredat350nm,differentnewcompoundswere

observedbutduetothelackofstandards,itwasonlypossibleto

identifyoneflavonoidasmyricitrin3-Orhamnoside(Romanietal.,

2012),oneofthemaincompoundpresentinmyrtleextract,and

waspresentinEEVOOat2.5timeshigherthangallicacidcontent.

Ontheotherhand,theEVOOsamplesenrichedwithMyrtleMAE

andCEextractsgaveclosephenoliccompoundscompositionwhich

impliesthatthemicrowaveextractionmethodmaybean

alterna-tivemethodallowingagoodrecoveryofantioxidantcompounds

fromplantsasreportedpreviouslyinmany studies(Dahmoune

etal.,2013).

3.2. Heatingeffectsonthephenoliccompoundcompositionof

myrtle-enrichedEVOO

Tobetterunderstandtherole ofphenolic compoundsinthe

EVOOoxidativestability,changesintheircompositionwere

mon-itoredduringtheprogressofoxidation.

3.2.1. Flameheating

Flameheating wasusedin ordertosimulatesomedomestic

practices.Theevolutionoftheendogenousphenoliccompoundsas

hydroxytyrol,tyrosol,luteolin,apigeninandunknown1(showing

astrongFLsignal)duringflameheatinginthecontrol(oil

with-outenrichment)andenrichedEVOO(EEVOO)wasconductedby

monitoringDADor FLD signal,and theevolution of exogenous

compoundssuchasgalloylquinicacidandmyricitrinwasalso

mon-itoredinmyrtleMAEorCE-EEVOO.Theresultwasexpressedas

degradation%.

With regard to phenolic alcohols, hydroxytyrol disappeared

progressivelyduringtheheatingtreatmenttoachieveacontent

lossof 69.5% after10min (235◦C),while tyrosolwas degraded

lessquicklyduringthefirstminutesoftreatment.While50%of

hydroxytyrosolwasdegradedafter5min(188◦C),thedecrease

of tyrosol content was about 7.7% in this time, diminishing

later more drastically until a 58.9% after 10min. In any case

the loss of tyrosol is less than that of hydroxytyrol (Fig. 2).

The phenolic compounds were degraded as a consequence of

their antioxidant activity and their degradation rate was

pos-itively correlated to their antioxidant efficacy (Bouaziz et al.,

2008).Infact,ourresultwasinagreementwiththetestsonthe

antioxidativeproperties ofthestudied compounds; the

antiox-idantcapacity decreased in the orderhydroxytyrol>oleuropein

aglycone>ligstrosideaglycone>Tyrosol(Carrasco-Pancorboetal.,

2005).Thus,hydroxytyrosol,havingtwohydroxylgroup,wasthe

firstcompoundtobeoxidizedincomparisontotyrosol(one

OH-groupandithasbeenclaimedpreviouslythatitisthemostactive

Table2

IdentificationofMyrtuscommunisphenoliccompoundinenrichedextravirginoliveoilusingRP-DLLME-HPLC-DAD-FLD.

EVOO+MyrtleMAE EVOO+MyrtleCE

Equation;R2 _{Concentration(mgkg}−1_{ofoil) Equation;R}2 _{Concentration(mgkg}−1_ofoil)

Galloylquinicacid* _{– 3.44±0.08}a _{– 4.17±0.07}b

Gallicacid Y=262.35x+389.72;0.99 1.48±0.76a _{262.35x+440.59;0.99 1.68±0.80}a

Myricitrin Y=183.6x+697.78;0.99 3.73±1.73a _{183.6x+755.75;0.99 4.04±1.81}a

Thevalueswereexpressedasthemeanofthreemeasurements(n=3)±standarddeviation(SD).Meanswithdifferentlettersforeachcompoundaresignificantlydifferent atp≤0.05.

Fig.2.FlameheatingeffectonEVOOorMyrtle-EEVOOphenoliccompoundsdegradation.

antioxidantcompound(Bendinietal.,2007)providingoxidative

stability totheEVOO byprotecting polyunsaturated fattyacids

(PUFAs) (Nissiotis and Tasioula-Margari,2002). Theantioxidant

activityof hydroxytyrosol is due tohydrogen donation and its

ability to scavengefree radicals by formingan intra-molecular

hydrogenbondbetweenthefreehydrogenofitshydroxylgroup

andtheirphenoxylradicals(Bouazizetal.,2008).

Luteolin(which hasa catecholgroup) is theflavonoidmost

affectedduringflameheatingandespeciallyafter5min(188◦C)

oftreatmentwhere thedegradationrateincreasedsignificantly

toachieve a loss of 69.3±1.1% at the highest exposition time.

Regardingapigeninwhichhasnotacatecholgroup,itwasthemost

stablecompoundanditscontentdiminishedina22.8_±2.1%.These

resultsshowedthatluteolinwasmoreefficientantioxidantthan

apigenin,therefore,itmightplayanimportantroleinthe

protec-tionoftheoilagainstthermaloxidationprocess.Theseobservations

confirmedthatthepresence ofa catecholmoiety enhancesthe

abilityofthephenoliccompoundstoactasantioxidantsby

reac-tingwithlipidradicalstoformnon-reactiveradicals,interrupting

thepropagationchain(Bendinietal.,2007)and alsothis

struc-turefeatureallowsabetterstabilizationofthephenoxylradicalby

increasingtheelectrondelocalization.

Thesecoiridoid1(tr36.6)showedthehigheststabilityin

compar-isontotheotherEVOOphenoliccompoundsduringflameheating

andlosingonly9.08%ofitscontentafter10minofheating.Itiswell

knownthatsecoiridoidcompoundsarethemostimportant

antiox-idants occurringin EVOO(90%) and contributeto itsoxidative

stabilityduetotheirantioxidantpropertiesandespecially

oleo-sidicformofhydroxytyrol(NissiotisandTasioula-Margari,2002).

Accordingtotheliterature,theconcentration ofthesecomplex

phenolsdecreasedduringlongstoragetimeLozano-Sánchezetal.

(2013)andalsoduringstrongheatingasreportedpreviouslyby

Beˇsteretal.(2008)showingthatthermaltreatmentofEVOOfor

Fig.3.ThedegradationofMyrtuscommunisphenoliccompoundduringflameheating(A),ovenheatingat180◦_{C(B)andmicrowaveheating(C)inmyrtle-EEVOO.}

ofsecoiridoid biophenols tothesimple biophenols,tyrosol and

hydroxytyrosol.Thissupposesthatourconditionswerefartocause

asignificantlossesofthesecompounds.

TheresultsshowalsothattheenrichmentofEVOObymyrtle

extractpreventsignificantlytheconsumptionofendogenous

phe-noliccompoundfromEVOOduringflameheating,incomparisonto

thecontrol(EVOOwithoutenrichment)asitcanbeseeninFig.2.

MyrtleMAEorCE-EEVOOdecreasedsimilarlythedegradationof

hydroxytyrol,20.04and18.87%respectively,andoftyrosol,14.14

and13.10%ofdecreaserespectively,incomparisontothecontrol.

Inaddition,myrtleMAE-EEVOOdecreasedthediminutionof

lute-olinandapigenin,32.55and15.62%respectively,andtheseeffects

werehigherthantheprotectionaffordedbymyrtleCE-EEVOO:only

aprotectionof8.90 and3.83%forluteolinandapigenin

respec-tivelywas observed.On the otherhand; secoiridoid 1was not

affectedduringflameheatingofEEVOO,incontrasttothecontrol

(EVOOwithoutenrichment).Regardingthephenoliccompounds

ofmyrtleextracts,ascanbeseeninFig.3A,rapiddegradationwas

observedformyricitrinandafter5minofheattreatmentthelevelof

myricitrindecreaseddramaticallyby91%or97%formyrtleMAEor

CE-EEVOOrespectively.Withregardtogalloylquinicacidwhichisa

hydrolysabletannin,itsleveldecreasedmoreslowlythanobserved

formyricitrinandachievelossesof75.0±6.7%or64.0±7.4%for

myrtleMAEorCE-EEVOOrespectivelyafter5minofheating.The

decreaseoftheconcentrationsofthesebothcompoundscouldbe

explainedbytheirantioxidantactivitiesandtheirthermal

degra-dation(Bouazizetal.,2008)andthedifferenceobservedbetween

themcouldbeattributedtotheirdifferenceinchemicalstructure,

reactivityagainstlipid radicalsbutalsotheirlocalizationinthe

oilsincegalloylquinicacidismorepolarthatflavonol(myricitrin).

Thus,myricitrinandgalloylquinicacidmaycontributetostabilize

EVOOduringheating.

3.2.2. Ovenheating

Theconventionalheatingat180◦Csimulatesfrying

tempera-turesduringdifferenttimeintervals(1,2and3h).Ascanbeseenin

Fig.4,thebehaviorofendogenousEVOOphenoliccompounds

dur-ingovenheatingat180◦Cwasdifferenttotheoutcomeobserved

duringflameheatingprocedure.

Thelevelofhydroxytyrosolinthecontrol(EVOOwithoutphenol

addition)remainedconstantduringthefirsthourofthetreatment;

after2and3hofheating at180◦C,thehydroxytyrosolcontent

lostby15.45%and 53%respectivelybutremainedlessthanthe

lossvalueobserved duringflame heating (Fig.3).On theother

hand,theleveloftyrosolincreasedsignificantlyby33.5and49.1%

compared to the unheated EVOO during the first 2 h of

heat-ingat 180◦C respectively, andafter 2h,tyrosol contentstarted

Fig.4.Evolutioncontentofhydroxytyrolandtyrosolduringovenheatingat180◦_C

forEVOOandmyrtle-EEVOO.

beforeheating(+14.24%incomparisontothecontrol).

Carrasco-Pancorboetal.(2007)whostudiedthedeteriorationofEVOOby

heatingat180◦C,reportedalossof91.5%and57.0%for

hydrox-ytyrosolandtyrosolrespectively,whichletussupposeaninitial

compositionolivecultivareffects(Brenesetal.,2002)on

antiox-idantevolution duringheating. In addition,Beˇsteret al.(2008)

studiedthephenoliccompoundchangesofextravirginoliveoils

afterheatingfor142hat100◦Cwithanairflow10L/handshowed

anincreaseoftyrosolcontentwhile hydroxytyrosoland

deriva-tives(oleuropein)andtyrosolderivatives(ligstroside)decreased.

Theseobservationssupposedthatthetransformationof

hyroxyty-rosolandtyrosolderivativesmayoccurduringheatingat180◦C

releasingsimplebiophenols,hydroxytyrolandtyrosolrespectively

(Beˇsteretal.,2008).Therefore,theconcentrationofhydroxytyrol

ateachstageofheatingwouldbethesumofitsformationfrom

thedecompositionofitssecoiridoidderivativesandits

decompo-sitionduetoitsroleasantioxidant(Kricheneetal.,2010).Since

tyrosolwaslesseffectivethanhydroxytyrol,itresultsan

accumu-lationofthis compoundduringtheirderivativestransformation

andafter2hwhenitstartedtardilyactingasantioxidant,itslevel

decreased.WhenEVOOwasenrichedbymyrtleextract,

hydroxy-tyrollevelwasalmostunalteredduringthetwofirststepofheating

(1and2h)similarlytotheobservedinun-enrichedoil,but,only

forMAE-EEVOO,anincreaseofhydroxytyrollevelwasobserved

incomparisontothecontrol(withoutenrichment).Thismaybe

attributedtothepresenceofinterferingcompoundfromoxidation

coelutingwithhydroxytyrolandthusincreasingtheareaofthis

lattercompound.Attheendofthetreatment(3h),hydroxytyrosol

waslessdegradedthaninthecontrol:41.8and38.4%degradation

forMAEandCE-enrichedEVOOrespectivelygivinganadditional

protectionof11.2%and14.58%fortheaboveoilsrespectivelyin

comparisontothecontrol.Thismaybeexplainedbytheactive

contributionof antioxidants from myrtle extract together with

Fig.5.Monitoringofthedegradationofluteolin,apigeninandunknown1during ovenheatingat180◦CinEVOOandmyrtle-EEVOO.

endogenousphenoliccompoundfromEVOOtoinhibitlipid

oxi-dationduringtheseheatingconditions.Withregardtotyrosol,the

enrichmentdoesnotaffectsignificantlytheevolutionofthis

com-poundduringheatingat180◦Cwhichfollowsthesamebehavior

observedabovefortyrosolinEVOOcontrol.Itsfinalconcentration

remainedequalorsuperiortotheinitialforalloiltypesandthisis

relatedtotheweakantioxidantactivityoftyrosolandderivatives

(Carrasco-Pancorboetal.,2005).

Fig.5showsthedegradationofluteolinduringheatingat180◦C.

ThedepletionofluteolininEVOOincreasedslowlyduringthefirst

steps,degradationreaching3.1and 19.6%in1and2 hof

heat-ing,respectively.Thenafurtherrapiddecreasewasobserved,up

to68.3%attheend(3h).However,apigeninshowedaslower

degra-dationandattheendoftheheatingabout49.4%waslost.Therefore,

apigeninshowedhigherstabilitythanluteolinwhichcorrespondto

theresultonflameheatingeffect.Thebehavioroftheseflavonols

wasinagreementwiththepreviousworkofAlloucheetal.(2007)

studyingtheheatingeffectat180◦Conextravirginoliveoil.The

differenceobservedbetweenthesecompoundsmaybeattributed

tothedifferentnumberofphenolichydroxylslinkedtotheir

phe-nol ring. Regarding the enriched EVOO, myrtle extract didnot

significantlystabilizethesecompoundsinthesupplementedoil

asobservedforflameheating.AscanbeobservedinFig.5,the

depletionof luteolin andapigenin wasingeneralthe samefor

alloils,bothun-enriched andenrichedoil,althougha

significa-tivedecreaseofthedegradationofthesecompoundswithrespect

tothecontrolsamplewasobservedafter2hoftreatment,being

anaverageprotectioneffectof9.6%and3.7%forluteolinand

api-geninrespectivelyinmyrtle-EEVOO.Theseobservationssuppose

alsothepossiblecontributionofflavonolstotheoilstabilityas

pre-viouslyreportedintheworkofArtajoetal.(2006)whoshowedthat

refinedoilenrichedbyluteolinandapigeninimprovesignificantly

itsoxidativestability.

Regardingthesecoirioid 1compound,itsleveldecreased

sig-nificantlywithrespecttotheresultofflameheating.Infact,the

Fig.6.MonitoringofEVOOorMyrtle-EEVOOphenoliccompoundduringmicrowave heating.

significantlossesof44.2%whichishigherthanthevaluefoundfor

flameheatingtreatment(only9%ofdegradationwasobserved).

Thisconfirmsourhypothesisadvancedfromflameheatingresults,

thatthiscompoundwasdegradedindrasticconditionswithlonger

time.Tobetterunderstandtheimpactofthedepletionofthis

com-pound,wehaveexaminedthecorrelationbetweentheunknown

1depletion and tyrosol content increase and a negative linear

relationship(R2_{=0.90)wasfoundduringthefirst2hofthermal}

treatment.Thismaysupposethatthesecoiridoid1maybelongto

ligstrosidefamilyanditsthermalhydrolysisaffectstheesterbound

betweenthephenolicportionandtherestofthemoleculeandthus

releasedasimplephenolastyrosol(tyrosolasanaromatic

alco-holicmoiety).Thisisinagreementwithseveralpreviousstudies

(Brenesetal.,2002;Carrasco-Pancorboetal.,2007).Thesecoiridoid

compoundsarethemainresponsibleofbitternessandpungency

sensationsinEVOO(Rubioetal.,2014)andaccordingtothework

ofLozano-Sánchezetal.(2013)whoinvestigatedthephenolic

com-poundsprofileofEVOOduringstorage(10month)thesecomplex

phenolsdecreasednoticeablyattheendofstoragesuggestingthat

thesesecoiridoidcompoundscouldbeusedasbiomarkersofEVOO

freshness.However,thesecoiridoid1contentduringthefirst2h

ofheatingat180◦CwasnotalteredinmyrtleMAEorCE-EEVOO

whichisincontrasttoitsbehaviorincontrolsample.Butafter3hof

treatment,nosignificantdifferencewasobservedforalloilstudied.

ThismeansthatthepresenceofmyrtleextractsinEVOOreducethe

depletionofthecompoundsecoiridoid1andthismaybeexplained

bythefactthatthislattercompoundwasnotonlylostbythermal

hydrolysisbutalsobyapossibleactionasantioxidantsubstance

reducingitsconsumptionwhenmyrtleextractwaspresent.Infact,

theworkofCarrasco-Pancorboetal.(2005)reportedthattyrosol

derivatives(ligstrosideaglycon)hadhigherscavengingeffecton

DPPHradicalsthantyrosol.

Regarding the antioxidants from myrtle extract (Fig. 3b),

galloylquinic acid presented similar degradation behavior that

observedwhenflameheatingwasappliedbut after3hof

heat-ing,galloylquinicacidwaslostby88%whileduringflameheating

(after10minwith235◦C)thiscompoundwascompletelylost.This

maybeattributedtothetemperatureeffectsincetheheatingtime

islongerinoventreatmentassay.Ontheotherhand,myricitrin

decreasedmoreabruptlyduringheatingat180◦Ctoreacha

max-imalloss(99%)after1hoftreatmentwhichisinaccordancewith

flameheatingassayandthusconfirmingthehighreactivityofthis

compoundanditsthermolabilityrelatedtoitshighernumberof

hydroxylgroups (Biesaga,2011).Incomparison tosecoiridoid 1

depletion,thereisatimeheatingeffectinsteadtemperature

heat-ingoneasfoundforgalloylquinicacid,andthus,itcanbededucted

thatthephenoliccompoundtrendsduringthermalstressaremore

Table3

TemperaturesachievedduringflameandmicrowaveheatingofEVOOsamples.

Flameheating Microwaveheating

Time(min) Temperature(◦_{C) Time(min) Temperature(}◦_C)

2 122±3.1 5 93±3.0

3 140±3.5 10 110±0.0

5 188±3.0 15 112±3.0

10 235±5.0 20 118±4.0

Thevalueswereexpressedasthemeanofthreemeasurements(n=3)±standard deviation(SD).

relatedtotheirnatureofmolecularstructure(Brenesetal.,2002; Lozano-Sánchezetal.,2013).

3.2.3. Microwaveheating

To the best of ourknowledge, noinvestigations have been

published on the influence of microwave irradiation on

phe-nolcomposition ofEVOO orEEVOO usingscientific microwave.

In this present study,we have workingat opensystemduring

microwaveheating(thetemperaturewasnotcontrolled)andthe

temperaturewasimmediatelymeasuredafterprocessing.The

tem-peraturesachievedweregiveninTable3.Animportantdisparity

hasbeenobservedregardingthetemperaturesachievedby

domes-ticmicrowave(170◦Cvs313◦C)reportedbymanyauthorsinthe

literaturewhohavestudiedtheimpactofmicrowaveheatingon

chemicalqualityofEVOO(Santosetal.,2013).

Theinfluence ofmicrowave heatingover the phenolic

com-poundsofthesampleswasassessedandtheresultsobtainedare

representedinFigs.6and7.Regardingphenolicalcohols(Fig.6),

itappearedthattheyaremoreresistanttomicrowaveexposure

thanobservedduringflameheating(Cerretanietal.,2009).The

microwaveheatingdecreasedprogressivelythelevelof

hydroxy-tyrosolbycausingalossof40.2%,whiletyrosolshowsmorestability

andwaslost onlyby23.6% atthehighest microwaveexposure

time(20min,118_±4◦C).Brenesetal.(2002)found20 and30%

ofhydroxytyrollossinEVOOfromSpanishcultivarsafter5and

10minofmicrowaveheatingrespectivelywhiletyrosolwas

unal-teredasfoundbythesameauthors(Brenesetal.,2002)which

mightbedue tothelower powersettingemployed(500W)by

theseauthorsthatinducedlesschemicalchangesinthiscompound

(Cerretanietal.,2009).In addition,theselatterauthorsshowed

animportant depletionof hydroxytyrosoland tyrosol by93.93

and62.33%respectivelyduring15minofheatingusingdomestic

microwave(at720W)andtheachievedtemperaturewas313◦C.

Theseobservationsshowedtemperatureeffectondepletionof

phe-nolicalcoholsduringmicrowaveheatingbutalsotheamountof

theoilanalyzedandthetypeofmicrowaveusedmustbetaken

inaccount.Theenrichmenthadasignificantprotectioneffecton

theconcentrationofphenolicalcoholsbyreducingtheirdepletion

duringmicrowaveheatingandtheirlossesstartedafter5minof

treatmentwhichisincontrastwiththecontrolascanbeseenin

Fig.6.Withrespecttohydroxytyrol,anadditionalprotectiveeffect

wasobserved,about11%and14%forMAE-orCE-EEVOO

respec-tivelyafter15minofheatingwithoutasignificantdifferenceat

p<0.05forthesebothoils.Atthehighesttimeofheating

hydroxyty-rosolcontentinCE-EEVOOwasquiteclosetothecontrol(1.9±0.7%

ofdifference).Thedepletionoftyrosolwasreducedby32.2and

30.3%(after15min,112◦C)inMAE-andCE-EEVOOrespectively.

TheworkofMalheiroetal.(2012)showedthatabout50%ofEVOO

totalphenoliccompounds(TPC)waslostduringmicrowave

heat-ing(1000W,10min)andthatteaextractenrichedEVOOhadnot

apositiveeffectontheTPCdepletionandgreenteaextract

pre-sentedaprooxidanteffectwhichisnotobservedinourwork.This

meansthatthetypeofplantextractisadeterminantfactorinthe

designingoffunctionalfoodsinceantagonistorsynergiceffectmay

Fig.7. MonitoringofEVOOorMyrtle-EEVOOphenoliccompoundduringmicrowave heating.

Regarding flavonol fraction (Fig. 7), luteolin and apigenin

contentsremainedunchangedduringmicrowaveheatingforall

exposuretimesassayedwhichisinagreementwiththeworkof

Vallietal.(2010)studyingtheeffectofmicrowaveheating(9minat

750W).Thecomparisonofthisresultwithflameandconventional

heating,supposed that thesethermal treatmentsledtoa more

importantoxidativeandhydrolyticdegradationofflavonolsthan

microwaveheating (Vallietal.,2010)andflavonolscontributed

secondaryontheoilstabilitycomparedtophenolicalcohols.The

flavonolsintheoils enrichedbymyrtle extractspresentedalso

similarpatternstothoseobservedwiththecontrol.

Thesecoiridoid1wasnotchangedduringmicrowaveheating

forallexposuretimeswhichincontrastwiththeresultobtained

whentheoilswasheatedinovenat180◦C.Itwasalreadyreported

byBrenesetal.(2002)thattheoildegradationwasverylowwhen

theoilswereheatedbymicrowave,andaccordingtotheworkof

Cerretanietal.(2009)itwasfoundthatsecoiridoidsderivatives

(ligstrosides)contentremainedconstantduringmicrowave

heat-inguntil9minoftreatment(700W,293◦C).Theseobservations

showedthehighstabilityofthesecompoundsduringthermalstress

whichisveryimportanttothenutritionalandorganolepticquality

ofEVOO,sincethesecompoundcontributedtothetasteand

pun-gencyoftheoil.ThesamepatternwasobservedforenrichedEVOO

bymyrtleextract.

Regardingmyricitrinandgalloylquinicacid(Fig.3C),microwave

heatinggives riseto lessdegradation effectcompared toflame

andovenheating.Adepletionof64.4and72.4%wasobservedfor

myricitrininMAEandCE-EEVOOrespectivelyduringmicrowave

heating.Thismaybeexplainedbytheoxidativestabilityandalso

bytheirhydrolyticdegradationasshownintheworkofBiesaga

(2011)andLiazidetal.(2007)whoreportedthelessstabilityof

flavonoidasmyricetinduringexpositiontomicrowaveirradiation

whichisrelatedtothegreaterdegreeofsubstitution(hydroxylic

groups).Galloylquinicacidwaslostby49.5and66.1%inMAEand

CE-EEVOOrespectivelyduringheatingtreatmentwhichsupposed

thecontributionofthiscompoundintheoilstability.

The enrichment of olive oil with myrtle extract could be

considereda water/oilemulsion inwhich smalldroplets of the

hydro-ethanolic solutionof theextract are dispersed in theoil

(Suárez et al., 2011). On the other hand, the peroxidation in

homogenouslipidsystemispredominantattheoil/airinterface

and thus, when the antioxidants accumulate is this interface,

theantioxidantactivityisgreater.ThefortificationofEVOOwith

myrtle extract increased the total phenolic compounds from

250mgGAEkg−1 (for the control without supplementation) at

about350mgGAEkg−1 for enrichedEVOO (datanot shown).In

addition,inourpreviousstudy,wereportedthatmyrtleextract

andEVOOphenoliccompoundsmayactcooperativelytoinhibit

lipidoxidationinliposomalmodel,inducedbyiron/ascorbicacid

system(unpublished data). Alltheseobservations supposethat

myrtleextractthroughitsdifferentantioxidantsbeinglocatingin

oil/airinterfaceand/orinlipidcorecouldcontributetostabilize

EVOOespeciallyduringflameheatingbyneutralizinglipid

radi-calsformedandthus,decreasedtheconsumptionofendogenous

phenoliccompoundsofEVOO.

3.3. Heatingeffectonspecificextinctioncoefficientsat232nm

and270nm

Theevaluationofspecificextinctioncoefficients(K232andK270)

allowscheckingthedegreeofoliveoiloxidation(Malheiroetal.,

2012).Thesespecificextinctioncoefficientsareindicativeofthe

formationforprimary(K232)andsecondaryproduct(K270)of

oxi-dationrespectivelyduringthermalstress(Malheiroetal.,2013).

ThemaximumvaluesallowedforK232andK270arerespectively

2.50and0.20forextravirginoliveoils(EEC,1991).

Table4showstheevolutionofK232andK270duringflame

heat-ingatdifferentexposuretimes.Theinitialvalueswere2.07and

0.11respectivelyforK232 andK270and thus,this oilwas

classi-fiedinEVOOcategoryaccordingtoEuropeancommunityregulation

(EEC,1991).Theadditionofmyrtleextractstotheoliveoildidnot

broughtsignificantchangesonK232andK270valuesforenrichedoil

beforeheating(t=0min).Duringtheflameheating,K232showed

highervaluesandconstantincreaseduringthermaltreatmentand

especiallyafter5min(188◦C)ofheatingwhereadrasticincrease

ofK232 wasobserved,from2.88to5.96.However,theenriched

oilbymyrtleextractshowedlessoxidativestatusandalloweda

protectiveeffectofabout40%attheendofthetreatment(10min)

comparedtothecontrol.Theseobservationswereinagreement

withtheresultobtainedfortheevolutionofEVOOphenolic

com-poundduringflametreatmentwhichindicatesthatmyrtleextract

mayreducethelossesofEVOOphenoliccompoundsduring

heat-ingand,therefore,myrtleextractmaycontributetodecreasethe

formationofprimaryproducts(hydroperoxides)byprotectingthe

oxidation ofpolyunsaturatedfattyacids(PUFAs)(Bouazizetal.,

2008;Malheiroetal.,2013).

ConcerningK270,theformationofsecondaryproductinthe

con-trolincreasedprogressivelyduringtheflameheatingbutadrastic

increase wasobserved after3min(140◦C)toachieve 1.07

val-uesattheendofthetreatment.Theenrichmentofoilbymyrtle

extractdidnothaveanysignificanteffectonK270values.The

com-parisonoftheK232andK270obtainedwithEEVOOsupposedthat

firstly,thereisotheroxidationcompoundsthatmayinterferewith

thetrieneconjugatedevaluatedat 270nm.Secondly, it maybe

supposedthatthedecompositionofhydroperoxidesformedwas

Table4

Comparisonofflame,ovenandmicrowaveheatingeffectinqualityparameters(K232andK270)ofextravirginoliveoilswithandwithoutmyrtleextracts.

Flameheating Time(min) 0 2 3 5 10 K232 Control 2.07±0.00de _2.46±0.00d _2.48±0.01cd _2.88±0.09c _5.96±0.07a MyrtleMAE 2.04±0.04e _2.13±0.02de _2.27±0.04de _2.28±0.00de _3.13±0.11b MyrtleCE 2.02±0.00e _2.17±0.08de _2.16±0.05de _2.25±0.02de _3.33±0.11b K270 Control 0.10±0.00d _0.14±0.01d _0.15±0.01d _0.58±0.09b _1.05±0.01a MyrtleMAE 0.11±0.00d _0.12±0.00d _0.16±0.01d _0.51±0.04bc _1.09±0.00a MyrtleCE 0.12±0.01d _0.13±0.00d _0.17±0.04d _0.52±0.10c _1.03±0.02a Ovenheating Time(hour) 0 1 2 3 K232 Control 2.07±0.00efg _2.25±0.00d _2.58±0.02b _2.76±0.03c MyrtleMAE 2.04±0.04fg _2.12±0.03e _2.43±0.01c _2.63±0.03b MyrtleCE 2.02±0.00g _2.10±0.00ef _2.42±0.04c _2.65±0.00b K270 Control 0.11±0.00f _0.48±0.01e _0.66±0.05c _0.75±0.00b MyrtleMAE 0.11±0.00f _0.46±0.00e _0.58±0.00d _0.69±0.01c MyrtleCE 0.12±0.01f _0.55±0.00d _0.71±0.03bc _0.83±0.02a Microwaveheating Time(min) 0 5 10 15 20 K232

Control 2.07±0.02gh _2.04±0.04h _2.21±0.04def _2.24±0.02cde _2.33±0.14bcd

MyrtleMAE 1.98±0.04h _2.13±0.01efgh _2.30±0.07cde _2.37±0.02b _2.48±0.02ab

MyrtleCE 2.02±0.00h _2.10±0.00fgh _2.19±0.01defg _2.34±0.00bc _2.53±0.01a

K270

Control 0.10±0.00h _0.14±0.00efg _0.18±0.03cde _0.21±0.01cd _0.27±0.00ab

MyrtleMAE 0.11±0.00gh _0.17±0.01def _0.20±0.00cd _0.22±0.00bc _0.28±0.01a

MyrtleCE 0.12±0.01fgh _0.17±0.00de _0.17±0.00def _0.21±0.01cd _0.29±0.00a

Thevalueswereexpressedasthemeanofthreemeasurements(n=3)±standarddeviation(SD).Meanswithinasameheatingprocedure,ineachparameterstudied,with differentlettersdiffersignificantly(p≤0.05).

2002).Thus,nocorrelationbetweenthisbothparameterscouldbe

established.

Regardingtheheatingprocedureat180◦C,K232valuesincreased

duringthethermalstresstoachieveavalueof2.76after3hwhich

isverylowcomparedtotheresultobtainedwithflameheating.

ROOHaretheproductsoftheactionofalkylperoxyradicalswith

theantioxidant,theirimmediatedecompositionatthehigher

tem-peratureswouldevenleadtochainpropagationratherthanchain

breakingundertheseconditions(VelascoandDobarganes,2002)

andthismayexplainthedifferenceofK232valuesfoundinthese

bothheatingproceduresandatimeeffectmaybethustakenin

considerationsince alongerheatingtime inducedlower

oxida-tivestabilityoftheoil(Carrasco-Pancorboetal.,2007).Enriched

oilshowedaslightprotectiveeffectonhydroperoxideformation

(K232)of 4.79and3.83%for MAE-andCE-EEVOOrespectivelyat

theendoftheheating(after3h)comparedtothecontrolwhich

inthecontrastwiththeresultofflameheating.Asshownabove,

drasticlossesofmyrtlephenoliccompounds(asmyricitrin)occur

duringovenheatingincomparisontoflameheatingandthus

reduc-ingtheantioxidantactivityoftheenrichedoilandtheirstability

becomesclosetothecontrolone.ForthechangesonK270values,it

canbeseeninTable4thattheproductionrateofsecondary

prod-uctsfromoiloxidationwasfasterduringthefirsthourforallthe

oiltypes(controlandenrichedoils)butthereafter,theK270values

increasedslowlywhichmaybeexplainedbutthestrong

volatiliza-tionofsecondaryoxidizedproductasformicacidandaldehydes

inthesedrasticconditions (Carrasco-Pancorboetal.,2007).The

controlandenrichedoilsshowedsimilarevolutionpatternsduring

ovenheatingbutCE-EEVOOshowedslightlyhighervalueattheend

ofthetreatmentthanthecontrolandMAE-EEVOO.

Table4presentsthechangesonspecificcoefficientabsorbance

duringmicrowaveheatingofoils.Themicrowaveheatingduring

allexposuretimesshowedlessdamageeffectonEVOOin

rela-tionwithhydroperoxidesproduction(K232)comparedtotheabove

heatingprocedures,flameandovenheating,andwhichisin

agree-mentwiththedegradationeffectonphenoliccompoundcontents,

i.e.lowerpolyphenollosses,lowerhydroperoxidesformationand

loweroxidationstatus.Ourresultsareinaccordancewithother

workpreviouslyreportedbyBrenesetal.(2002)butincontrast

withtheworkofAlbietal.(1997)whoreporteda moredrastic

effectofmicrowaveheatingthanofovenone.Theseconflicting

resultsmayberelatedtotheconditionsofheatingandthe

instru-mentused.K232valuesincreasedslowlyduringmicrowaveheating

withsimilartrendsfor thecontroland myrtle-enrichedoilsup

to15min,butattheendofthetreatment,asignificant(p<0.05)

highervaluewasobservedforenrichedoilthanforthecontroland

theirvalues wereslightlyhigher thanthelimitvalue(2.5). Our

resultisinagreementwiththefindingofMalheiroetal.(2012)

studyingtheenrichmentofEVOObygreentea.Thismayberelated

tothephoto-oxidationofchlorophyllwhereitmayactas

prooxi-dantsubstanceswhichleadtoincreasetheproductionofoxidation

primaryproduct.Thishypothesismaybesupportedbythework

ofWanasundaraandShahidi(1998)whoreportedthatgreentea

extractexhibitedaprooxidanteffectinedibleoils,namelymarine

oils,butdechlorophyllizedgreenteaextractpresentedanexcellent

antioxidantactivityintheoilstudied.

RegardingK270,theabsorbancevaluesofconjugatedtrienes(CT)

weregraduallyincreasedwiththeincreaseof heatingtime but

after15min,theK270 valuesbecomehigherthanthelimitvalue

toachieve0.27,0.28and0.29forthecontrol,MAE-orCE-EEVOO

respectively and thus disqualifying the oil studied from EVOO

category.ThisparameterisinaccordancewithK232resultwhich

indicatedthatmicrowaveheatingcausedminorchangesonolive

Fig.8.Principalcomponentanalysis(A)andhierarchicalclusteranalysis(B)ofEVOOsamplesbasedonthemainimportantfactors(redcolor)(hydroxytyrol,tyrosol,luteolin, apigeninandunknown1content,K232andK270).Oil1–5:oilcontrolheatedbyflame(0,2,3,5and10min);Oil6–10:Oil+myrtleMAEheatedbyflame(0,2,3,5and10min);

Oil11–15:oil+myrtleCEheatedbyflame(0,2,3,5and10);Oil16–20:oilcontrolheatedbymicrowave(0,5,10,15and20min);Oil21–25:oil+myrtleMAEheatedby microwave(0,5,10,15and20min);Oil26–30:oil+myrtleCEheatedbymicrowave(0,5,10,15and20min):Oil31–34:oilcontrolheatedbyoven(0,1,2and3h);Oil35–38 oil+myrtleMAEheatedbyoven(0,1,2and3h):Oil39–42:oil+myrtleCEheatedbyoven(0,1,2and3h).(Forinterpretationofthereferencestocolorinthisfigurelegend, thereaderisreferredtothewebversionofthisarticle.)

tothestatisticalanalysis,theK270valuesduringmicrowave

heat-ingdidnotshowsignificantdifference(p<0.05)foralloilsstudied

atthedifferentexpositiontimes,i.e.theadditionofmyrtleextract

totheEVOOdidnotaffectsignificantlytheK270valueswhichis

incontrastwiththeworksof Malheiroet al.(2012,2013) who

reportedthatgreenteaandoliveleavesincreasedK270valuesin

EVOOandsoybeanoilundermicrowaveheatingrespectivelyatthe

endoftreatment(15min,1000W)whichmeansthattheseextracts

actedasprooxidantbyfavoringsecondaryoxidationcompounds

formationwhichisnotobservedwithmyrtleextract.

3.4. Principalcomponentanalysis(PCA)andhierarchicalcluster

analysis

Oncetheinformationfromallparametersanalysisoftheoils

hadbeencollected,astatisticalstudywascarriedouttoseethe

distributionthatfollowedthedata.Principalcomponentanalysis

(PCA) was considered as a good tool in establishing

relation-shipsbetweenvariables.Fig.8ashowsthebiplotofthetwofirst

principalcomponentsthatexplain82.06%ofthetotalvariancein

thedatacollectedfromoilsheatedbythedifferentprocedures.As

canbeclearlyseeninthegeneralplotoftheparameters,thereis

apositiverelationshipbetweenluteolin,apigeninandsecoiridoid

1,andnegativecorrelationofthesevariableswithK232andK270.

Itcanbealsoobservedthattyrosolandhydroxytyrosolshowed

aweakcorrelationwiththeothersfactorsthatmeansthatthese

compoundshaddifferentbehaviorsthanotheronesasobserved

duringovenheatingfortyrosol.

Inordertoobservesimilaritiesordissimilaritiesbetweentheoil

samples,hierarchicalclusteranalysiswasusedasacomplementary

tooltoPCA.Thedissimilarityofdifferentclusterswascalculatedby

Ward’smethod.Fig.8bshowsthedendrogramobtainedfromthe

oilsamplesandsixclustersareformed.Ascanbeseen,cluster1

wasformedprincipallybyoilheatedbymicrowave(0–5min)and

flame(0–3min);andthecluster2wasformedbyoilheatedby

microwave(5–20min)andflame(5min).Thisresultmeansthat

theoilheatedbyflame(after5min)presentsimilaritywiththe

heatingis moreseverethan microwaveone.The oilheated for

20min by flame formed two distinct clusters, 3 and 4, which

containedoilcontrol(withoutenrichment)andenrichedoil

respec-tively.Thisresultsuggestthattheseoilshaveadifferencewithin

thecompositionandphytochemicalprofilewhichisinagreement

withthechemicalanalysis,intheotherhand,theenrichedoilwas

lessdegradedthanoilcontrol.Theclusters5and6wereprincipally

formedbyoilsheatedbyovenat180◦C.Acleardistinctionbetween

theoilcontrolandenrichedoilwasnotdeterminedbecausethe

treatmentappliedisquiteseverecomparedtootherprevious

treat-mentandthus,thereisaclosecompositionandoxidationstatusfor

alloilstudied.Accordingtoalltheseobservation,theenrichment

ofoliveoilbymyrtleextractinducesasignificanteffectprincipally

duringflameheating.

4. Conclusions

Ourresultsdemonstratethatheatingtreatmentcouldinduce

significantlossesofEVOOphenoliccompoundswhichleadtothe

deteriorationofitsnutritionalandorganolepticqualities.Themore

safetyheatingprocedurewasmicrowavetreatmentincomparison

withflame orovenones.Achemical structureand degradation

raterelationshipwas observedand therefore hydroxytyrolwas

themain important antioxidant toensure oilstability but also

flavonoidsfractionhavepositive effects. Theadditionof myrtle

extractsinEVOOpreventstronglytheconsumptionofits

endoge-nousphenoliccompoundsashydroxytyrol,tyrosolandflavonoids

(luteolin and apigenin) and the greatest protective effect was

observed when microwaveand flame heating were applied. In

addition,ahigherinhibitionintheformationofprimaryoxidation

compoundsduetoahighercapacitytoprotectPUFAwasobserved

inenrichedoilduringflameandovenheating.Myrtleextractsmay

benefittheoilsbyimprovingtheircompositioninantioxidant

com-poundsandtheiroxidativestability.Thedevelopmentofanew

functionalfoodwithEVOOandmyrtleextractshouldbeexplored.

Acknowledgements

SofianeDAIRIgratefullyacknowledgesElisabethMartinTornero

andMariCarmenHurtado Sánchezfortheirhelpandassistance

duringthiswork,andtheOrganicChemistryLaboratorypersonnel

(ExtremaduraUniversity)fortheirtechnicalsupportandassistance

duringsampleanalysis.

References

Albi,T.,Lanzón,A.,Guinda,A.,Leon,M.,Pérez-Camino,M.C.,1997.Microwaveand conventionalheatingeffectsonthermoxidativedegradationofediblefats.J.Agri. FoodChem.45(10),3795–3798.

Allouche,Y.,Jimenez,A.,Gaforio,J.J.,Uceda,M.,Beltran,G.,2007.Howheatingaffects extravirginoliveoilqualityindexesandchemicalcomposition.J.Agric.Food Chem.55,9646–9654.

Artajo,L.S.,Romero,M.P.,Morelló,J.R.,Motilva,M.J.,2006.Enrichmentofrefined oliveoilwithphenoliccompounds:evaluationoftheirantioxidantactivityand theireffectonthebitterindex.J.Agric.FoodChem.54,6079–6088.

Bendini,A.,Cerretani,L.,Carrasco-Pancorbo,A.,Gómez-Caravaca,A.M., Segura-Carretero,A.,Fernández-Gutiérrez,A.,Lercker,G.,2007.Phenolicmoleculesin virginoliveoils:asurveyoftheirsensoryproperties,healtheffects, antioxi-dantactivityandanalyticalmethods.AnoverviewofthelastdecadeAlessandra. Molecules12,1679–1719.

Bengana,M., Bakhouche,A., Lozano-Sánchez,J., Amir,Y.,Youyou, A., Segura-Carretero,A.,Fernández-Gutiérrez, A.,2013.Influence ofoliveripenesson chemicalpropertiesandphenoliccompositionofChemlalextra-virginoliveoil. FoodRes.Int.54,1868–1875.

Beˇster,E.,Butinar,B.,Buˇcar-Miklavˇciˇc,M.,Golob,T.,2008.Chemicalchangesinextra virginoliveoilsfromSlovenianIstraafterthermaltreatment.FoodChem.108, 446–454.

Biesaga,M.,2011.Influenceofextractionmethodsonstabilityofflavonoids.J. Chro-matogr.A1218,2505–2512.

Bouaziz,M.,Fki,I.,Jemai,H.,Ayadi,M.,Sayadi,S.,2008.Effectofstorageonrefined andhuskoliveoilscomposition:stabilizationbyadditionofnaturalantioxidants fromChemlalioliveleaves.FoodChem.108,253–262.

Brenes,M.,García,A.,Dobarganes,M.C.,Velasco,J.,Romero,C.,2002.Influenceof thermaltreatmentssimulatingcookingprocessesonthepolyphenolcontentin virginoliveoil.J.Agric.FoodChem.50,5962–5967.

Capriotti,A.L.,Cavaliere,C.,Crescenzi,C.,Foglia,P.,Nescatelli,R.,Samperi,R.,Lagana, A.,2014.Comparisonofextractionmethodsfortheidentificationand quantifi-cationofpolyphenolsinvirginoliveoilbyultra-HPLC-QToFmassspectrometry. FoodChem.158,392–400.

Carrasco-Pancorbo,A.,Cerretani,L.,Bendini,A.,Segura-Carretero,A.,DelCarlo,M., Gallina-Toschi,T.,Lercker,G.,Compagnone,D.,Fernandez-Gutierrez,A.,2005.

Evaluationoftheantioxidantcapacityofindividualphenoliccompoundsin virginoliveoil.J.Agric.FoodChem.53,8918–8925.

Carrasco-Pancorbo,A.,Cerretani,L.,Bendini,A.,Segura-Carretero,A.,Lercker,G., Fernandez-Gutierrez,A.,2007.Evaluationoftheinfluenceofthermaloxidation onthephenoliccompositionandontheantioxidantactivityofextra-virginolive oils.J.Agric.FoodChem.55,4771–4780.

Cerretani,L.,Bendini,A.,Rodriguez-Estrada,M.T.,Vittadini,E.,Chiavaro,E.,2009.

Microwaveheatingofdifferentcommercialcategoriesofoliveoil:PartI.Effect onchemicaloxidativestabilityindicesandphenoliccompounds.FoodChem. 115,1381–1388.

Dahmoune,F.,Boulekbache,L.,Moussi,K.,Aoun,O.,Spigno,G.,Madani,K.,2013.

ValorizationofCitruslimonresiduesfortherecoveryofantioxidants:evaluation andoptimizationofmicrowaveandultrasoundapplicationtosolventextraction. Ind.Crop.Prod.50,77–87.

Dairi,S.,Madani,K.,Aoun,M.,KeeHim,J.L.,Bron,P.,Lauret,C.,Cristol,J.-P., Carbon-neau,M.-A.,2014.Antioxidativepropertiesandabilityofphenoliccompounds ofMyrtusCommunisleavestocounteractinvitroLDLandphospholipidaqueous dispersionoxidation.J.FoodSci.,http://dx.doi.org/10.1111/1750-3841.12517. EuropeanUnionCommissionRegulationEEC2568/91,1991.Onthecharacteristics

ofoliveandolivepomaceoilsandtheiranalyticalmethods.Off.J.Eur.Commun., L248.

Godoy-Caballero,M.,Acedo-Valenzuela,M.,Galeano-Díaz,T.,2012.Simple quantifi-cationofphenoliccompoundspresentintheminorfractionofvirginoliveoil byLC-DAD-FLD.Talanta101,479–487.

Godoy-Caballero,M.P.,Acedo-Valenzuela,M.I.,Galeano-Diaz,T.,2013.Newreversed phasedispersiveliquid–liquidmicroextractionmethodforthedeterminationof phenoliccompoundsinvirginoliveoilbyrapidresolutionliquid chromathog-raphywithultraviolet-visibleandmassspectrometrydetection.J.Chromatogr. A1313,291–301.

Huang,C.L.,Sumpio,B.E.,2008.Oliveoil,themediterraneandiet,andcardiovascular health.J.Am.Coll.Surg.207,407–416.

Krichene,D.,Allalout,A.,Mancebo-Campos,V.,Salvador,M.D.,Zarrouk,M., Fre-gapane,G.,2010. Stabilityofvirginolive oilandbehaviour ofitsnatural antioxidantsundermediumtemperatureacceleratedstorageconditions.Food Chem.121,171–177.

Liazid,A.,Palma,M.,Brigui,J.,Barroso,C.G.,2007.Investigationonphenolic com-poundsstabilityduringmicrowave-assistedextraction.J.Chromatogr.A1140 (1),29–34.

Lozano-Sánchez,J.,Bendini,A.,Quirantes-Piné,R.,Cerretani,L.,Segura-Carretero, A.,Fernández-Gutiérrez,A.,2013.Monitoringthebioactivecompoundsstatus ofextra-virginoliveoilandstorageby-productsovertheshelflife.FoodControl 30,606–615.

Malheiro,R.,Casal,S.,Lamas,H.,Bento,A.,Pereira,J.A.,2012.Canteaextractsprotect extravirginoliveoilfromoxidationduringmicrowaveheating?FoodRes.Int. 48,148–154.

Malheiro,R.,Rodrigues,N.,Manzke,G.,Bento,A.,Pereira,J.A.,Casal,S.,2013.Theuse ofoliveleavesandteaextractsaseffectiveantioxidantsagainsttheoxidationof soybeanoilundermicrowaveheating.Ind.Crop.Prod.44,37–43.

Nissiotis,M.,Tasioula-Margari,M.,2002.Changesinantioxidantconcentrationof virginoliveoilduringthermaloxidation.FoodChem.77,371–376.

Romani,A.,Campo,M.,Pinelli,P.,2012.HPLC/DAD/ESI-MSanalysesandanti-radical activityofhydrolyzabletanninsfromdifferentvegetalspecies.FoodChem.130, 214–221.

Rubio,L.,Macia,A.,Castell-Auvi,A.,Pinent,M.,Blay,M.T.,Ardevol,A.,Romero,M.P., Motilva,M.J.,2014.Effectoftheco-occurringoliveoilandthymeextractsonthe phenolicbioaccessibilityandbioavailabilityassessedbyinvitrodigestionand cellmodels.FoodChem.149,277–284.

Rubio,L.,Motilva,M.J.,Macia,A.,Ramo,T.,Romero,M.P.,2012.Developmentofa phenol-enrichedoliveoilwithbothitsownphenoliccompoundsand comple-mentaryphenolsfromthyme.J.Agric.FoodChem.60,3105–3112.

Santos,C.S.P.,Cruz,R.,Cunha,S.C.,Casal,S.,2013.Effectofcookingonoliveoilquality attributes.FoodRes.Int.54,2016–2024.

Suárez, M., Romero, M.-P., Ramo, T., Motilva, M.-J., 2011. Stability of a phenol-enriched olive oil during storage. Eur. J. Lipid Sci. Technol. 113, 894–903.

Tamendjari,A.,Laincer,F.,Laribi,R.,Arrar,L.,Rovellini,P.,Venturini,S.,2014.Olive oilsfromAlgeria:phenoliccompounds,antioxidantandantibacterialactivities. GrasasyAceites65,e001.

Tuck,K.L.,Hayball,P.J.,2002.Majorphenoliccompoundsinoliveoil:metabolism andhealtheffects.J.Nutr.Biochem.13,636–644.

Valli,E.,Bendini,A.,Cerretani,L.,Fu,S.,Segura-Carretero,A.,Cremonini,M.A.,2010.

Effectsofheatingonvirginoliveoilsandtheirblends:focusonmodificationsof phenolicfraction.J.Agric.FoodChem.58,8158–8166.

Velasco,J.,Dobarganes,C.,2002.Oxidativestabilityofvirginoliveoil.Eur.J.Lipid Sci.Technol.104,661–676.

Wanasundara,U.N.,Shahidi,F.,1998.Antioxidantandpro-oxidantactivityofgreen teaextractsinmarineoils.FoodChem.63(3),335–342.