Effect of the extraction cooditioris of virgin olive oil on the lipoxygenase cascade: Chemical and sensory implications

G asas y Acei es Vol. 50. Fase. 2(1999), 114-121 E ec o he ex ac ion coodi io is o i gin oli e oil on he lipoxygenase cascade: Chemical and senso y implica ions By Ma ia T Mo ales (1), F anca Ange osa (2) and Ramó i Apa icio (1) * (1) Ins i u o de la G asa. A da. Pad e Ga cía Teje o 4, 41012 Se illa, Spain. (2) Is i u o Spe imen ale pe la Elaio ecnica, 65013 Ci á S. Angelo, Pesca a, I aly. RESUMEN E ec o de las condiciones de ex acción del acei e de oli a i gen en la u a de la iipoxigenasa: implicaciones químicas y senso iales. Los compues os olá iles p oducidos a a és de la u a de la Iipo- xigenasa son los esponsables de los a ibu os senso iales más impo - an es del acei e de oli a i gen. El p esen e abajo es udia la e olución de es os compues os según di e en es condiciones de em- pe a u a y iempo del p oceso de ex acción. Se ha de e minado la in- luencia de es os pa áme os en la p oducción de compues os olá iles Ce y Os así como el e ec o que ienen es os compues os en los desc ip o es senso iales más des acados del acei e de oli a i gen ( e de, ama go-pican e, dulce e indeseable). Se dan alo es óp imos de iempo y empe a u a pa a la p oducción de acei es de oli a í ge- nes de al a calidad senso ial. Los a ibu os senso iales de es a ieda- des, la española Picual y las i alianas F an oio y Co a ina, ue on e aluados y los compues os olá iles cuan i icados median e dos me- odologías di e en es con el in de e i a que los esul ados pudie an ci cunsc ibi se a una a iedad especí ica o a un único mé odo de cuan i icación. PALABRAS-CLAVE: Acei e de oli a i gen - Compues os olá- iles - Lipoxigenasa ( u a) - Sis emas de ex acción. SUMMARY E ec o he ex ac ion condi ions o i gin oli e oil on he lipoxygenase cascade: Chemical and senso y implica ions. The ola ile compounds p oduced h ough he lipoxygenase cascade a e esponsible o he mos ema kable senso y a ibu es o i gin oli e oil. The pape analyses he e olu ion o hese compounds acco ding o di e en condi ions o empe a u e and ime o he malaxing p ocess. The in luence o hese pa ame e s on he p oduc ion o Ce and Cs ola ile compounds is s a ed oge he wi h he e ec ha he amoun o hese compounds has on he mos ema kable i gin oli e oil senso y desc ip o s (g een, bi e -pungen , swee and undesi able). Op ima alues o empe a u e and ime a e gi en o p oducing high senso y quali y i gin oli e oils. Th ee cul i a s, Spanish Picual and I alian F an oio and Co a ina, we e cha ac e ised by senso y a ibu es and ola ile compounds quan i ied by wo di e en me hodologies in o de o a oid ha esul s could be ci cumsc ibed o a speci ic cul i a o a quan i ica ion me hod. KEY-WORDS: Ex ac ion sys ems - Lipoxygenase (cascade) - Vi gin oli e oil - Vola ile compounds. 1. INTRODUCTION Vi gin oli e oil is he oil ex ac ed om he ui o he oli e ee, Olea eu opea L. This ege able oil is consumed wi hou a u he e ining p ocess so e aining ola iles and o he mino compounds ha gi e ise o a ag an and delica e la ou ha has been app ecia ed by consume s since ancien imes. In p e ious /o ks (Apa icio e al., 1994; Apa icio and Mo ales 1995) di e en senso y no es esponsible o i gin oli e oil la ou we e es ablished h ough he consume app ecia ion, he mos ema kable being g een, swee , ui y, ipe ui , ipe oli es, undesi able and bi e -pungen . Mos o hese senso y pe cep ions a e p oduced by ola ile compounds (Fla h e ai, 1973; Gu ié ez e al., 1975; Solinas é al., 1988; Mo ales e a/., 1995; Apa icio é al., 1996b, Ange osa e al., 1996). F om all o hem, he g een senso y pe cep ion is o g ea impo ance because he ag an la ou o he i gin oli e oil is p oduced by he balance be ween g een and ui y no es (Mo ales e ai, 1996). I has been s a ed ha alipha ic Ce compounds and he co esponding hexyl es e s (Gu h and G osch, 1991; Gu h and G osch, 1993; Mo ales e ai, 1996) a e he main con ibu o s o he «un ipe» componen o he ui la ou . Mo eo e , almos all ola iles esponsible o g een senso y no es a e he majo componen s o he i gin oli e oil headspace (Salinas e al, 1987; Mo ales e al., 1994; Di Gio acchino e al., 1996; Ange osa e al, 1997), These compounds a e biosynîhesised om Gi8-unsa u a ed a y acids in plan s oilowing he lipoxygenase (LOX) pa hway (Vick and Zimme man, 1987; Ha anaka e al, 1987). This pa hway in oi es he ac ua ion o di e en enzymes gi ing ise o di e en amoun s o aldehydes, alcohols and es e s ha ing all o hem senso y p ope ies and so con ibu ing o he o e all la ou . In he case o oli e ui s, i has been demons a ed (AIR, 1994) ha he LOX pa hway p omo es he o ma ion o Ce ola ile compounds agains Cg ola ile compounds and, in consequence, a g ea amoun o hose ola ile compounds esponsible o g een senso y no es (c) Consejo Supe io de In es igaciones Cien í icas Licencia C ea i e Commons 3.0 España (by-nc) h p://g asasyacei es. e is as.csic.es Vol. 50. Fase. 2(1999) 115 (Mo ales e al., 1996) can be ound in esh and high quali y i gin oli e oils. On he o he hand, en a i e p e ious s udies (Gio acchino and Se aiocco, 1995; Ranalli and Se aiocco, 1996; Ranalli and Ange osa, 1996) ha e demons a ed ha he condi ions o he malaxing p ocess du ing oli e oil ex ac ion a ec o he senso y quali y o i gin oli e oil. Thus, he mos impo an goal o his esea ch is o es ablish an index ela ed o he bes ex ac ion condi ions h ough he in o ma ion o ola ile compounds esponsible o he unique senso y a ibu es o i gin oli e oil (Apa icio e al., 1996b). This pape emphasises he impo ance o empe a u e and ime du ing he malaxing p ocess on he p oduc ion o Cs and Ce ola ile compounds. Di e en me hodologies and a ie ies ha e been used o a oid ha esul s could be ci cumsc ibed o a speci ic cul i a o quan i ica ion me hod. 2. MATERIALS AND METHODS Samples Oli e ui s om h ee di e en Eu opean cul i a s -Picual (Spain), Co a ina and F an oio (I aly)- we e used in his s udy. Vi gin oli e oil was p ope ly ob ained om esh, heal hy ui s o good quali y collec ed a ea ly g een s age o ipeness ha co esponds o alue 2 acco ding o he scale sugges ed by F ias e al., (1993). The ex ac ion o he oil samples was ca ied ou a analy ical scale using expe imen al oil mills. Tempe a u e and ime o malaxing we e he pa ame e s s udied. Di e en empe a u e alues we e selec ed, he minimum being 25° G ha co esponds o he «cold» ex ac ion condi ions while he maximum 35° G is in he neighbou o he maximum sugges ed by expe s (Humanes, 1993). Fi e malaxing imes (15, 30, 45, 60, 90 min) we e selec ed in o de o co e he ange o ime habi ually used o ex ac he oils om he di e en ypes o oli es selec ed. Thi y i gin oli e oil samples we e cha ac e ised by hei ola ile con en using dynamic headspace p ocedu es and senso y desc ip o s. Ano he expe imen was ca ied ou wi h in ac and cu oli e ui s in o de o de ec i he LOX pa hway is induced when oli e ui s a e damaged (Gailla d, 1980). Twen y g ams o in ac ui s we e analysed ollowing he me hodology desc ibed below (DHS using Tenax TA aps). Gu oli es we e ob ained placing wen y g ams o in ac ui s in he ex ac ion essel and cu ing i e imes he pulp o each í uü. Cu oli es we e analysed immedia ely hey we e cu using he same me hodology. Dynamic headspace-Gas Ch oma og aphy Vola ile compounds o i gin oli e oil c Picual we e analysed by a dynamic headspace echnique p e iously epo ed (Mo ales e al., 1994; Apa icio and Mo ales, 1994). Samples o 0.5 g we e hea ed a 40° G and swep wi h N2 (200 ml/min) o 15 min and he ola iles adso bed on o a Tenax TA ap (Ch ompack, Middlebu g, The Ne he lands) a oom empe a u e. A Ch ompack he mal deso p ion cold ap injec o (TCT) was employed o ca y ou he he mal deso p ion o he apped ola iles by hea ing a 220° G o 5 min. The ola iles we e hen condensed on o a used silica ap cooled a -110° G wi h liquid ni ogen o 5 min jus be o e injec ion ha was ca ied ou by lash hea ing o he cold ap a 170° G whe e i was held o 5 min. The ola iles we e ans e ed on o a used silica J&W (Folsom, GA) DB-WAX column (60m x 0.25 mm i.d., 0.25 ^m ilm hickness). The o en empe a u e was held a 40° G o 6 min and p og ammed o ise a 2° G/min o a inal empe a u e o 200° G whe e i was held o 10 min. A Hewle -Packa d 5890 se ies II wi h a FID de ec o was employed. Quan i ica ion was ca ied ou using ¡sobu yl ace a e as in e nal s anda d. The iden i ica ion o ola ile compounds was ca ied ou by MS using iden ical condi ions o Gas Ch oma og aphy. A Fisons Mass De ec o MD800 coupled o a GG 8000 se ies was employed. Masslab 1.3 was he so wa e used. Sample componen s we e e i ied by compa ison o mass spec al da a wi h hose o au hen ic e e ence compounds. Vola ile compounds o i gin oli e oil c . Co a ina and F an oio we e analysed by a dynamic headspace- HRGC echnique p e iously de eloped (Ange osa e al., 1990). Fi y g ams o oil was pu in o a 120 ml D echsel gas washing bo le wi h a po ous dis ibu o o which was added a known amoun o nonan-1-ol (in e nal s anda d). Vola iles we e s ipped wi h ni ogen (1.2 dm^ /min, a 37° G) o 2 h, apped on 50 mg o ac i a ed cha coal, and elu ed wi h 1 mL o die hyl e he . All sol en s, o o ganic esidual analysis, we e pu chased om J.T Bake (De en e , Holland); nonan-1-ol (in e nal s anda d) om Ald ich (S einheim, Ge many); ac i a ed cha coal (0.5-0.85 mm; 20-35 mesh ASTM) om E. Me ck (Schucha d , Ge many). Cha coal was cleaned by ea men in a soxhie appa a us wi h die hyl e he and es ed be o e he analyses. Gas ch oma og aphy was ca ied ou wi h a Ca lo E ba Mega Se ies 5160 i ed wi h a No dion silica capilla y Ca bowax 20 M column (50 m leng h; 0.32 mm i.d.; 0.5 mm ilm hickness), and equipped wi h an on~column injec ion sys em, a GO2 c yogenic accesso y o hold he o en a 25° G and a Flame Ioniza ion De ec o (FID). The o en empe a u e p og am was he ollowing: iso he m a 25° C o 7 min, (c) Consejo Supe io de In es igaciones Cien í icas Licencia C ea i e Commons 3.0 España (by-nc) h p://g asasyacei es. e is as.csic.es 116 G asas y Acei es om 25 o 33° C a 0.8° C/min, om 33 o 80° C a 2.4° C/ min and om 80 o 155° C a 3.7° C/min; inal iso he m a 155° C o 20 minu es. The empe a u e o he de ec o was held a 240° C; ca ie gas was H2 a 30 kPa. Injec ion olume was 0.5 |iL. The iden i ica ion o ola ile compounds was ca ied ou by MS unde iden ical condi ions o GC bu now using a Hewle -Packa d Mass De ec o coupled o a GC 5890A model was used. Sample compounds we e iden i ied by compa ison o mass spec al da a wi h hose o au hen ic e e ence compounds and by hei e en ion imes. The au ho s ha e poin ed ou in p e ious s udies ha he e is a good co ela ion be ween he echniques jus desc ibed abo e (Mo ales and Apa icio, 1993). Senso y p ope ies o ola ile compounds Assesso s: The senso y p ope ies o ola ile compounds (odou and as e) we e e alua ed by i e assesso s wo wi h mo e han en yea s expe ience and h ee who we e habi ual consume s o i gin oli e oil. HRGC-ol ac ome y: To assess he a oma no es co esponding o oli e oil ola ile compounds, an High Resolu bn Gas Ch oma og aphy (HRGC)-ol ac ome y echnique was applied o i gin oli e oil samples (Mo ales e al., 1994). The e luen o he GC column was spli ed 1 o 10 o he de ec o and he sni ing po , espec i ely. The desc ip ions o he odou - ac i e egions o he elua e we e no ed on a o m wi h a p e-p in ed ime scale, assesso s did no see he ch oma og am. Assesso s basically ag eed on he odou s o ola iles, al hough di e en seman ic e ms whe e used o desc ibe some o hem. A consensus-building discussion was held wi h assesso s o decide he inal senso y desc ip o s. Th eshold alues and Tas ing o pu e compounds: Odou h eshold alues o ola ile compounds, in a ma ix o deodo ised sun lowe oil, we e de e mined by he au ho s using he me hod desc ibed by Gu h and G osch (1993). The assesso s also ca ied ou he smelling and as ing, in duplica e, a oom empe a u e o ou pu e ola ile compounds ha p e iously had been cha ac e ised as esponsible o senso y pe cep ions o he han g een: hexan-1-ol (Me ck, Da ms ad , Ge many), (E)-2-hexenal, (E)-2-hexen-1-ol and (E)-3-hexen-1-ol (Ald ich, Milwaukee, Wisconsin, USA). In ac , Apa icio e al., (1996b) ound ha (E)-2-hexen-1-ol, (E)-3-hexen-1-ol and hexan-1-ol we e be e cha ac e ised by he global senso y pe cep ion o undesi able. This cha ac e isa ion ag ees wi h Bedoukian (1971) who ound ha he ans o ms o hexenols a e cha ac e ised by a a y senso y pe cep ion while he c/s o ms a e sha p and g een. The ola ile compounds we e p e iously dilu ed in wa e o pa a in oil, depending on hei solubili y, o he same app oxima e concen a ion as ound in i gin oli e oil samples. Panel es The senso y e alua ion o oli e oil is egula ed inside The Eu opean Union (EC,1995). The desc ip o s o he o icial panel we e se up o desc ibing he mos ema kable senso y pe cep ions o i gin oli e oil and no o di e en ia e sub le changes in he «g een» pe cep ion. In o de o dissec and desc ibe di e en la ou sensa ions associa ed wi h g een pe cep ion, eigh ull ained I alian assesso s we e eques ed o e alua e some oils ob ained om un ipe ui s o di e en a ie ies. In all o he aspec s, i was s ic ly ollowed he egula ion o he Eu opean Union (EC, 1995). Assesso s chose eely hei own desc ip o s by means o associa ions o he sensa ions pe cei ed wi h own o me expe iences. The desc ip o s wi h enough ag eemen be ween assesso s we e selec ed o he p o ile shee (Table I). These senso y desc ip o s basically ag eed wi h hose desc ibing he «g een» a oma pe cep ion, by ei he habi ual o po en ial consume s (Apa icio e al., 1994; Apa icio and Mo ales, 1995), al hough some o he desc ip o s, mos ly ela ed o as ing (bi e and pungen ), we e also selec ed o desc ibing he «g een» p o ile o i gin oli e oil. Table I Senso y desc ip o s e alua ed in i gin oli e oil samples (c . Co a ina. Picual and F an oio). S uc u ed scale om 0 o 5 Senso y desc ip o s Cu g een lawn G een lea o wig G een oli es Wild lowe s G een banana G een oma o Almond A ichoke Apple Walnu husk G een hay Bi e Pungen Co a ina 0.7 1.5 2.0 — 0.1 0.6 0.1 0.7 — 1.1 — 2.8 2.5 Picual 1.0 0.8 1.8 0.1 0.2 0.8 0.6 0.3 0.1 0.5 0.1 1.8 1.9 F an oio 1.0 1.0 2.0 0.2 0.2 0.7 0.3 0.6 0.1 1.1 0.1 2.1 2.0 (c) Consejo Supe io de In es igaciones Cien í icas Licencia C ea i e Commons 3.0 España (by-nc) h p://g asasyacei es. e is as.csic.es Vol. 50. Fase. 2(1999) 117 S a is ical Analysis STATISTICA (S a So , 1995) was he s a is ical package employed. Response su aces we e used o plo he i ed model o he selec ed chemical compounds in a su ace along wi h hei obse ed alues in he expe imen s a di e en ime and empe a u e du ing he ex ac ion p ocess. A quad a ic smoo hing algo i hm was applied o he s anda dised alues o each ola ile compound. Canonical Co ela ion (Tabachnick and Fidell, 1983) was applied o assessing he ela ionship be ween senso y a ibu es and ola iles. This algo i hm allows in es iga ing he edundancies be ween bo h se o da a. p esen in he i gin oli e oil headspace we e also p oduced a e cu ing he oli es. This expe imen s a es ha ola iles a e p oduced, h ough he lipoxygenase pa hway, as soon as oli es a e cu o damaged. M JÀL.l JÂn 3. RESULTS AND DISCUSSION The senso y e alua ion o samples by di e en desc ip o s allowed knowing whe he hey we e basically cha ac e ised by g een a ibu es. The esul s o he senso y p o iles showed ha a ie ies we e cha ac e ised by di e en pe cep ions o g een senso y a ibu e. Oil om Picual oli es is usually cha ac e ised by high alues o a ibu es ui y ( oma o) and pungen (Apa icio e al., 1994, 1996a, 1997) while his panel es (Table I) cha ac e ised he Spanish a ie y as g een oli es, g een oma o and cu g een lawn. Co a ina oil is habi ually cha ac e ised by he as ing a ibu es bi e and pungen (Apa icio e al., 1994; McEwan, 1994) bu , conce ning he g een pe cep ion, he desc ip o s we e g een oli es, g een lea o wig and walnu husk (Table I). High in ensi ies o g een, ui y and pungen a ibu es cha ac e ised F an oio oils (FLAIR, 1991) bu ocusing on g een pe cep ions his oil shows high alues o a ibu es g een oli es and cu g een lawn (Table I). F om he la ge da abase o ola iles iden i ied in i gin oli e oil (Mo ales e al., 1994) only a ew a e esponsible o g een pe cep ions (Apa icio e al., 1996b). These ola ile compounds a e p oduced h ough LOX pa hway and hey a e usually known as «g een ola ile compounds». In his pa hway he p oduc ion o he ola iles is associa ed o cell damage (Gailla d, 1980). Figu e 1 clea ly shows he di e ences be ween wo ch oma og ams co esponding o ola iles p oduced in in ac oli es and cu oli e ui s c . Picual. I is no ewo hy he main p esence o hyd oca bons in in ac ui s and he absence o ola ile compounds esponsible o he inal i gin oli e oil a oma, only aces o hexanal and hexan-1 -ol -bo h coming om linoleic acid- could be de ec ed. The ola ile p o ile o cu oli es shows a conside able change due o he p oduc ion o g een ola ile compounds, coming om bo h linoleic and linolenic acids. O he ola ile compounds usually Figu e 1 P oduc ion o some ola ile compounds in in ac (A) and cu (B) oli es. Quan i ica ion by dynamic headspace Table II shows he Ce ola ile compounds analysed in samples. The HRGC-ol ac ome y showed g een senso y no es co e ing he ange be ween swee - ui y-g een o bi e -powe ul-g een. The odou h eshold and odou ac i i y alue (OAV) (G osch, 1994) a e also gi en. A ola ile compound wi h an OAV g ea e han 1.0 con ibu es o he g een senso y pe cep ion, while ola iles ha ing lesse OAVs could con ibu e o i gin oli e oil la ou h ough possible syne gism phenomena. Table II Senso y desc ip ion, odou h esliolds and odou ac i i y alues o he g een ola ile compounds Vola ile Compounds Hexanal (Z)-3-hexenal (E)-2-hexenal Hexyl ace a e (Z)-3-hexenyl ace a e Hexan-1 -ol (E)-3-hexen-1-ol (Z)-3-hexen-1-ol (E)-2-hexen-1 -ol Odo desc ip o s G een, apple, g een ui G een, cu g ass G een, ui y, almonds F ui y, swee F ui y, g een lea es F ui , banana, so G een G ass, banana G een, g assy, swee Odo h esholds" 60 3.0 1200 1040 730 400 1500 5800 8000 OAV" 8.91 184.67 9.76 0.04 0.07 0.13 0.01 0.07 0.01 Legend: ^ Calcula ed in ^g/kg ^ OAV = odou ac i i y alue. (c) Consejo Supe io de In es igaciones Cien í icas Licencia C ea i e Commons 3.0 España (by-nc) h p://g asasyacei es. e is as.csic.es 118 G asas y Acei es The high alue o he canonical co ela ion coe icien (R=0.98) be ween he Ce g een ola ile compounds (Table II) and he senso y a ibu es (Table I) demons a es ha he o me a e esponsible o he la e . Howe e , he ola iles only explained 66% o he senso y a ibu es ( edundancies be ween da a se s) and his could mean ha he whole in o ma ion o g een senso y a ibu es is no only explained by he odou o he ola iles bu also by he as ing o some o hese ola iles as assesso s e alua ed he samples by hei ol ac o y-gus a o y- ac ile no es (EC, 1995). P e ious s udies using SSW (Apa icio e al., 1994, 1996b) demons a ed ha some o hese compounds mainly con ibu e o he g een senso y pe cep ion i.e. (Z)-3-hexenal, hexyl ace a e, (Z)-3-hexenyl ace a e, and (Z)-3-hexen-1-ol. Howe e , o he se o hem could also con ibu e o o he senso y pe cep ions i.e. (E)-2-hexenal, hexan-1-ol, (E)-3-hexen-1-ol, and (E)-2-hexen-1-ol. The as ing o he la e ola iles allowed hei cha ac e isa ion as as ingen and bi e ((E)-2-hexen-1-ol and (E)-3-hexen-1-ol), ough (hexan-1-ol) and sha p, bi e , as ingen ((E)-2-hexenal). The e is good ag eemen among consume s in ela ion o he senso y pe cep ions. They dislike high in ensi ies o as ing pe cep ions bi e and pungen , while hey like almos all a oma desc ip o s quali ied wi h he adjec i e g een (McEwan, 1994; Paglia ini e al., 1994). The e o e, he p esence o ola iles esponsible o pleasan senso y pe cep ions should be p omo ed. Howe e i gin oli e oil mus be ex ac ed solely using physical means, and hen he a io be ween ola iles in he inal p oduc can only be a ied by changing he empe a u e and ime o he malaxing p ocess. Depending on empe a u e and ime o malaxing, each ola ile compound showed di e en beha iou . Figu e 2 shows a scheme o he LOX pa hway wi h i s h ee di e en b anches o he p oduc ion o ola iles. The b anch coming om linoleic acid gi e ise o hexanal, hexan-1-ol and hexyl ace a e, being he o me and he la e esponsible o desi able pe cep ions. This b anch could be seen as he g een-swee aspec o he global g een la ou . The second b anch could be seen as he esponsible o he main g een pe cep ion, being cons i u ed by (Z)-3-hexenal, (Z)-3-hexenol and (Z)-3-hexenyl ace a e. The hi d b anch gi ing ise o (E)-2-hexenal and (E)-2-hexen-1-ol could be conside ed as he bi e -as ingen aspec o he g een senso y pe cep ions. Figu e 2 Response su aces o Ce ola ile compounds p oduced h ough he lipoxygenase cascade. The su ace has been i ed o he da a by a 2 o de polynomial unc ion and ola ile compounds we e p e iously no malised by Z-sco e. The igu es o ola ile compounds ha e been o a ed (clokwise o coun e -clokwise) in o de o expose he esul s be e . Hexanal, esponsible o g een, apple, swee pe cep ions, has an OAV = 8.91 (Table II), he e o e con ibu ing o pleasan senso y pe cep ions. The p oduc ion o hexanal is mo e in luenced by he ime o malaxing han by he empe a u e, hexanal is p oduced du ing he whole p ocess bu he highes concen a ions we e ound a e 45 minu es o malaxing. A di e en beha iou showed ola ile (E)-2-hexen-1-ol ha is cha ac e ised by a g een smelling and an as ingen -bi e as ing, he la e being an undesi able senso y pe cep ion o po en ial consume s (McEwan, 1994). The p oduc ion o (E)-2-hexen-1-ol is mainly a ec ed by empe a u e since he highes concen a ions we e measu ed a he highes empe a u e o malaxing (35° C). Hexyl ace a e is also a ec ed by empe a u e bu , in his (c) Consejo Supe io de In es igaciones Cien í icas Licencia C ea i e Commons 3.0 España (by-nc) h p://g asasyacei es. e is as.csic.es Vol. 50. Fase. 2(1999) 119 case, he highes concen a ion o his es e was measu ed a he lowes ennpe a u e (25° C). The o he compounds esponsible o g een a oma desc ip o s ((Z)-3-hexenal, (Z)-3-hexen-1-ol, and (Z)-3-hexenyl ace a e) also showed he highe concen a ions a low empe a u e, Figu e 2. The in luence o malaxing ime was low o (Z)-3-hexenal, medium o (Z)-3-hexen-1-ol and highe o (Z)-3-hexenyl ace a e. This means ha he p oduc ion o (Z)-3-hexenal is almos immedia e while he p oduc ion o (Z)-3-hexenyl ace a e needs mo e ime. This explana ion ag ees wi h he sequence o p oduc ion o hese ola iles h ough he LOX pa hway. The i s s ep o his biochemical pa hway is he o ma ion o he aldehyde, hen i s ans o ma ion o alcohol -by he ac ion o alcohol dehyd ogenase enzyme (ADH)- and la e he o ma ion o he es e , by means o he alcohol acyl ans e ase enzyme (AAT). I is no ewo hy ha (Z)-3-hexen-1-ol is ep esen ed by a sha p saddle pic u e. I has wo maximum peaks -a low empe a u e and low ime and a high empe a u e and high ime- so sugges ing ha he la e condi ion (high empe a u e and high ime o malaxing) makes easie he o ma ion o his compound in de imen o (Z)-3-hexenyl ace a e while he o me condi ions enables i s ans o ma ion on (Z)-3-hexenyl ace a e, bo h being cha ac e ised as g een ola iles. Conce ning he ola iles esponsible o he undesi able senso y pe cep ions ((E)-2-hexenal, hexan-1-ol, (E)-3-hexen-1-ol), hei concen a ions a e highe when he empe a u e o he malaxing is he highes . Figu e 2. In his case, he ime o malaxing had no in luence on he o ma ion o he alcohols bu i showed a g ea in luence on he o ma ion o he aldehyde. The concen a ion o he aldehyde was lowe a high and low malaxing imes and highe a medium alues o ime o he Picual a ie y hough ce ain di e ences we e ound om he esul s o Co a ina and F an oio a ie ies. The in e media e ola ile compound (E)-2-hexenal is ep esen ed by a ypical saddle pic u e whe e, once again, he e a e wo maximum peaks a low and high empe a u es unde medium ime o malaxing. Acco ding o he senso y cha ac e isa ion o (E)-2-hexenal and (E)-2-hexen-1-ol he low empe a u e gi es he bi e as ing o i gin oli e oil bu when empe a u e eaches high alues han his senso y pe cep ion in i gin oli e oil is accompanied by an undesi able pe cep ion ha can make he oil o be ejec ed. In gene al, he highes concen a ions o aldehydes we e measu ed when he ime o malaxing was sho e , he p oduc ion o alcohols was p omo ed a high empe a u e (35° C) and he concen a ion o es e s was highe a low empe a u e (25° C). This means ha he malaxing p ocess a high empe a u e (T > 35° C) p omo e he o ma ion o g een ola iles esponsible o undesi able senso y pe cep ions while, on he con a y, low empe a u es (T > 25° C) a ou ed he p oduc ion o desi able g een senso y pe cep ions. The empe a u e is hen he main ac o cha ac e ising he malaxing p ocess. Recen ly Ange osa e al., (1998) ha e sugges ed a b anch o he lipoxygenase cascade o Cs ola ile compounds. An alkoxy adical would be p oduced om 13-hyd ope oxide by he ac ion o he lipoxygenase, la e by p-scission would p oduce 1,3-pen ene adical and an unsa u a ed aldehyde. A se ies o pen ene dime s would be p oduced by dime iza ion o 1,3-pen ene while he hyd oxila ion would p oduce 2-pen en-1-ol and 1-pen en-3-ol ha will hen p oduce 1-pen en-3-one and 2-pen enal by he ac ion o he alcohol dehyd ogenase. Cs ola ile compounds ha e a senso y beha iou qui e simila o Ce ola ile compounds, he e a e con ibu o s o g een, (Z)-2-pen en-1-ol and (E)-2-pen enal, swee , 1-pen en-3-one, pungen , 1-pen en-3-ol, and undesi able, (Z)-2-pen enal and 1-pen en-3-ol, senso y pe cep ions (Apa icio e al., 1996b). (Z)-2-pen en-1-ol, esponsible o g een banana, is p omo ed a low alues o empe a u e (25° C) and ime (15 min) while i s lowes alues a e a he highes alues o hese a iables, so showing a simila beha iou o Ce ola ile compounds esponsible o g een pe cep ions. Figu e 3. The nex ola ile in his lipoxygenase cascade, 1-pen en-3-ol, has i s highes alues immedia ely he malaxing begins and i s concen a ion diminishes when malaxing goes on ime; he empe a u e seems o ha e poo in luence on i s p oduc ion. Figu e 3. This beha iou can be in e p e ed as hese compounds a e p oduced in pa allel p ocess and he p oduc ion o hese ola ile compounds begins immedia ely he oli es a e c ushed; hese compounds al eady appea ed in he ch oma og am o cu oli es. Figu e 1. (Z)-2-pen enal is p oduced om he p e ious compounds and Figu e 3 shows ha he ola ile eaches i s op imum a e a ce ain delay; in ac , i does appea s a ace le els in Figu e 1. The p oduc ion o 1-pen en-3-one is inc eased wi h empe a u e, so sugges ing i is p oduced h ough an oxida ion p ocess. Figu e 3 shows ha i s maximum is eached wi h a delay simila o ha o (Z)-2-pen enal. The main conclusion o he p esen s udy is ha low empe a u e (T > 25° C) and medium imes (35-45 min.) a e he bes ex ac ion condi ions o p omo e he o ma ion o g een ola ile compounds esponsible o desi able senso y pe cep ions. High empe a u e (T > 35° C) wi h minimum alues o ime ( < 35 min.) could also be use ul as a al e na i e way o ob ain pleasan g een i gin oli e oils. (c) Consejo Supe io de In es igaciones Cien í icas Licencia C ea i e Commons 3.0 España (by-nc) h p://g asasyacei es. e is as.csic.es 120 G asas y Acei es -^ 1.2: Figu e 3 Response su aces o Cs ola ile compounds a di e en empe a u es and imes o malaxing. The su ace has been i ed o he da a by a 2"^ o de polynomial unc ion. Values o ola ile compounds a e gi en in 10*ppm. All he esul s, coming om ei he ola ile compounds o senso y a ibu es in o ma ion, poin ou ha low empe a u e du ing malaxing (an indus ial example is pe cola ion) is de e minan o ob ain pleasan g een i gin oli e oils in any o he s udied a ie ies. ACKNOWLEDGEMENTS The au ho s would like o acknowledge hei indeb edness o assesso s who ca ied ou he senso y e alua ion o he samples. This wo k has been suppo ed by AIR3-CT94-1967. BIBLIOGRAPHY AIR (1994).—«Eu opean P ojec AIR3-CT94-1967. Oli e oil la ou and a oma: biochemis y and chemis y o senso y ac o s a ec ing consume app ecia ion and hei analysis by a i icial in elligence, The Commission o he Eu opean Communi ies». Ange osa, F., Di Giacin o, L and Solinas, M. (1990).— «In luence o oli e ui mass s o age on he a oma o esul ing oils. E alua ion o « us y» de ec by HPLC and GLC analyses o ola ile compounds».—Ri . Me ceol., 29, 275-294. Ange osa, R, Di Giacin o, L., Wo, R. and Cumi ini, S. (1996).— «Senso y e alua ion o i gin oli e oils by a i icial neu al ne wo k p ocessing o dynamic head-space gas ch oma og aphic da a».—J. Sci. Food Ag ia, 72, 323-328. Ange osa, F., Di Giacin o, L and D'Alessand o, N. (1997).— «Quan i a ion o some la o componen s esponsible o he "g een" a ibu es in i gin oli e oils».—J. High Resol. Ch oma og ., 20, 507-510. Ange osa, F., Came a, L, d'Alessand o, N. and Melle io, G. (1998).—«Cha ac e iza ion o se en new hyd oca bons compounds p esen in he a oma o i gin oli e oils».—J. Ag ie. Food Chem., 46, 648-653. Apa icio, R. and Mo ales, M.T. (1994).—«Op imiza ion o a dynamic headspace echnique o quan i ying i gin oli e oil ola iles. Rela ionships among senso y a ibu es and ola ile peaks».—Food Qual. and P e ., 5, 109-114. Apa icio, R. and Mo ales, M.T. (1995).—«Senso y wheels: a s a is ical echnique o compa ing QDA panels. Applica ion o i gin oli e oil».—J. Sci. Food Ag ie, 67, 247-257. Apa icio, R., Alonso, M.V., Mo ales, M.T. and Cál en e, J.J. (1994).—«Rela ionship be ween COI es and o he senso y p o iles by s a is ical p ocedu es».—G asas Acei es, 45, 26-41. Apa icio, R., Cál en e, J.J. and Mo ales, M.T. (1996a).— «Senso y au hen ica ion o Eu opean ex a- i gin oli e oil a ie ies by ma hema ical p ocedu es».—J. Sci. Food Ag ie, 72, 435-447. Apa icio, R., Mo ales, M.T. and Alonso, V. (1996b).— «Rela ionship be ween ola ile compounds and senso y a ibu es o oli e oils by he senso y wheel».—J. Am. Oil Chem. Soc, 73,1253-1264. Apa icio, R., Mo ales, M.T. and Alonso, V. (1997).— «Au hen ica ion o Eu opean ex a- i gin oli e oils by hei chemical compounds, senso y a ibu es and consume s a i udes».—J. Ag ie. Food Chem., 45, 1076-1083. Bedoukian, P.Z. (1971).—«The se en p ima y hexenols and hei ol ac o y cha ac e is ics».—^J. Ag ie. Food Chem., 19,1111-1114. Di Gio acchino, L., Ange osa, F. and Di Giacin o, L. (1996).—«E ec o mixing lea es wi h oli es on o ganolep ic quali y o oil ob ained by cen i uga ion».— J. Am. Oil Chem Soc, 73, 371-374. Di Gio acchino, L. and Se aiocco, A. (1995).—«In luenza dei sis emi di la o azione delle oli e sulla composizioni dello spazio di es a degli oli».—Ri . I al. Sos anze G asse, 72, 443-450. EC. (1995).—O icial Jou nal o The Commission o The Eu opean Communi ies. Ma ch 28 h, 3-4 Regula ion n.° 1 656/95. FLAIR (1991).—Eu opean P ojec FLAIR-CT91-0046. The s udy o senso y and nu i ional quali y o i gin oli e oil in ela ion o a ie y, ipeness and ex ac ion echnology, The Commission o he Eu opean Communi ies. Fi s Yea Repo . (c) Consejo Supe io de In es igaciones Cien í icas Licencia C ea i e Commons 3.0 España (by-nc) h p://g asasyacei es. e is as.csic.es Vol. 50. Faso. 2 (1999) 121 Fla h, R.A., Fo ey, R.R. and Guadagni, D.G. (1973).— «A oma componen s o oli e oil».—J. Ag ie. Food Ghem, 21, 948-952. F ías, L, Ga cia-O iz, A., He moso, M., Jiménez, A., Lla e o, M.R, Mo ales, J., Ruano, M.T. and Uceda, M. (1993).—Analis as de Labo a o io de Almaza a, Andalusian Minis y o Ag icul u e, Se ille (Spain). Gallia d, T. (1980).—The Biochemis y o Plan s, Vol. 4, Academic P ess, London. G osch, W.(1994).—«De e mina ion o po en odou an s in oods by a oma ex ac dilu ion analysis (AEDA) and calcula ion o odou ac i i y alues (OAVs)».—Fla ou F ag .J.,9,147-158. Gu h, H. and G osch, W. (1991).—«A compa a i e s udy o he po en odo an s o di e en i gin oli e oils».—Fa . Sci. Technol., 93, 335-339. Gu h, H. and G osch, W. (1993).—«Quan i a ion o po en odo an s o i gin oli e oil by s able so ope dilu ion assays».—J. Am. Oil Ghem. Soc, 70, 513-518. Gu ié ez, R., Olías, J.M., Gu ié ez, F., Cab e a, J. and del Ba io, A. (1975).—«The ch oma og aphic and o ganolep ic me hods in he e alua ion o he a oma ic cha ac e is ics o i gin oli e oil».—G asas y Acei es, 26,21-31. Ha anaka, A., Kajiwa a, T. and Sekiya, J. (1987).— «Biosyn he ic pa hway o Ce-aldehydes o ma ion om linolenic acid in g een lea es».—Ghem. Phys. Lipids, 44, 341-361. Humanes, J. (1993).—P oducción de Acei e de Oli a de Calidad: In luencia del cul i o, Jun a de Andalucía, Conseje ía de Ag icul u a y Pesca, Se ille (Spain). McEwan, J. A. (1994).—«Consume a i udes and oli e oil accep ance: he po en ial consume ».—G asas y Acei es, 45, 9-15. Mo ales, M.T. and Apa icio, R. (1993).—«Op imiza ion by ma hema ical p ocedu es o wo dynamic headspace echniques o quan i ying i gin oli e oil ola iles».—Anal. Chim. Ac a, 282, 423-431. Mo ales, M.T., Apa icio, R. and Ríos, J.J. (1994).— «Dynamic headspace gas ch oma og aphic me hod o de e mining ola iles in i gin oli e oil».—J. Ch oma og . A, 668, 455-462. Mo ales, M.T, Cál en e, J.J. and Apa icio, R. (1996).— «In luence o oli e ipeness on he concen a ion o g een a oma compounds in i gin oli e oil».—Fla ou F ag . J.. 11,171-178. Mo ales, M.T, Alonso, M.V, Ríos, J.J. and Apa icio, R. (1995).— «Vi gin oli e oil a oma: ela ionship be ween ola ile compounds and senso y a ibu es by chemome ics».— J. Ag ie. Food Chem., 43, 2925-2931. Paglia ini, E., Be uccioli, M. and Abba, S. (1994).— «Consume a i udes and oli e oil accep ance: he adi ional consume ».—G asas y Acei es, 45,16-19. Ranalli, A. and Ange osa, F. (Î996).—«In eg al cen i uges o oli e oil ex ac ion. The quali a i e cha ac e is ics o p oduc s».—J. Am Oil Chem. Soc, 73, 417-422. Ranalli, A. and Se aiocco, A. (1996).—«E alua ion o cha ac e is ics o oli e oil p oduced by inno a i e and adi ional p ocessing echnologies».—Ri . I al. Sos anze G asse, 73, 303-315. Solinas, M., Ange osa, F. and Ma silio, V. (1988).—« Indagine su alcuni componen i dell'a oma degli oli e gini di oli a in elazione alia a ie à delle oli e».—Ri . I al. Sos anze G asse, 45, 361-368. Solinas, M., Ma silio, V. and Ange osa, F. (1987).—«E oluzione di alcuni componen i dell'a oma degli oli e gini di oli a in elazione al g ado di ma u azione delle oli e».—Ri . I al. Sos anze G asse, 44, 475-480. S a So , (1995).—STATISTICA elease 5, Tulsa, USA. Tabachnick, B.G. and Fidell, LS. (1983).—Using Mul i a ia e S a is ics, Happe and Row Publishe s, New Yo k. Vick, B.A. and Zimme man, D.C.(1987).—«Oxida i e sys ems o modi ica ion o a y acids: he lipoxygenase pa hway».—In The Biochemis y o Plan s: A Comp ehensi e T ea ise, Vol. 9, p. 53. S ump , P.K. and E.E. Conn, (Eds.).—Academic P ess Inc., New Yo k. Recibido: Junio 1998 Acep ado: Sep iemb e 1998 (c) Consejo Supe io de In es igaciones Cien í icas Licencia C ea i e Commons 3.0 España (by-nc) h p://g asasyacei es. e is as.csic.es