Environmental Stress and Aroma Production During Wine Fermentation

Transcription

1 Envirnmental tress an Arma Prutin During Wine Fermentatin.. Fairbaim, A.Y. mit, D. Jabsn, B.A. Prir, F.F. B auer* Institute fr Wine Bitehnlgy, Department f'vitiulture an Enlgy, Private Bag Xl, tellenbsh niversity, Matielan 762, uth Afria ubmitte fr publiatin: July 213 Aepte fr publiatin: Marh 214 Key wrs: Wine arma, hypersmti stress, temperature stress, syntheti grape must, fermentatin The sensry esriptin f wine uses the wiest range f esriptive terminlgy f all f pruts, refleting the mplex nature f a prut whse harater epens n the balane f hunres f iniviual flavur-ative mpuns. There are many tls that an influene flavur prffles r wine styles, ne f whih is the hie f a speifi yeast strain. Yeasts ntribute t wine flavur by pruing vlatile metablites with ifferent flavur prffles. The impat f hanging envirnmental nitins n the prutin f flavur mpuns by yeast strains remains largely unexplre. This is the first stuy investigating the impat f tw mil fermentatin stresses, hypersmti an temperature stress, n arma prutin in syntheti must by mmerial aharmyes erevisiae wine strains. Hypersmti stress was impse by ultivatin f the yeast fr 21 ays in the must ntaining either.3 r.5 M srbitl. The transient temperature stresses were applie fr 16 h at 8 1 r 37 fr either tw r eight ays after mmenement f the fermentatin. Greater glyerl an aeti ai levels were prue by mst yeasts when nly hypersmti stress was applie. Hypersmti an temperature stress nitins prue a limite number f signifiant hanges t the prffle f the esters, higher alhls an vlatile fatty ais. These hanges iffere signifiantly fr eah strain an stress treatment, suggesting that the fermentatin nitins an signifiantly alter the armati prffle f a wine, althugh these stress impats annt be preite in general. The hanges t the armati prffle are speifi t eah iniviual wine yeast strain. ITRODTIO Wine yeast strains meiate the near-mplete nversin f high levels f sugar t ethanl an O 2 espite ntinuus expsure t varius frms f stress (Bissn, 1999). The stresses may inlue temperature hanges, hypersmti stress, ethanl stress, vitamin, mineral, nitrgen an xygen efiienies, an a lw ph, amng thers (Attfiel, 1997; Alexanre & harpentier, 1998; Bissn, 1999; Bauer & Pretrius, 2). There is eviene that suh stresses are als ften assiate with prblemati (stuk an sluggish) fermentatins (Bissn, 1999; Gibsn et al., 27; Malherbe et al., 27). The high initial sugar ntent f grape must results in hypersmti stress, whih is unterate by. erevisiae thrugh the aumulatin f intraellular glyerl in rer t maintain a psitive turgr pressure (Hhmann, 1997; Mager & ierius, 22; Tamás & Hhmann, 23). Temperatures well belw an abve the grwth ptimum f yeast inrease membrane permeability an isrupt transprt systems int an ut f the yeast ell (Bissn, 1999; Beales, 24). Yeast strains iffer in their ability t sense an respn t stress. Mst f these stresses have been stuie extensively with regar t their impat n yeast grwth an fermentatin perfrmane (Ivrra et al., 1999; arras et al., 21; Zuzuarregui & Del Olm, 24). Generally, yeast grwth uner stress is slwe, as its metablism is aapting t meet the new envirnmental nitins. Hwever, the literature prvies few if any insights int the impat f fermentatinrelate stress n metablite prutin that might influene wine flavur. Wine flavur is efine as the umulative effet f smell, taste an muth-feel (Franis & ewtn, 25), an is a iret funtin f the hemial mpsitin f the wine. The flavur- ative metablites that have an impat n wine pereptin are erive frm the grapes an frm mirrganisms uring fermentatin, as well as frm hemial presses uring prutin an maturatin (Rapp & Manery, 1986; Rapp & Versini, 1996). Pereive flavur is the result f mplex interatins between all the vlatile an nn-vlatile mpuns present in wine. The vlatile armas prue e nv by wine yeast uring fermentatin inlue higher alhls, esters an vlatile fatty ais (Rapp & Versini, 1996). Higher alhls are frme by the earbxylatin an reutin f a-ket ais, riginating frm either glylysis r the Ehrlih *rrespning authr: Flrian F Bauer.jb2@sun.a.za [Tel.: ; Fax: Other authrs: amantha. Fairbairn: fairbairn@sun.a.za, Anita Y mit: aysmit@istel..za, Dan Jabsn: jabsn@sun.a.za, Bernar A. Prir: bapsun.a.za Aknwlegements: We thank anie tilwaney fr labratry assistane. The atinal Researh Funatin an Anhr Yeast are thanke fr finanial supprt. Afr. J. EnL Viti., VL 35,. 2,

2 169 Arma Prue by Yeast ner fress pathway (Rapp & Manery, 1986; Rapp & Versini, 1996; wiegers et al., 25; tyger et al., 211). Vlatile fatty ais are prue as a result f aetyl-a earbxylatin an nensatin reatins (Lambrehts & Pretrius, 2). Aetate esters are frme by enzyme-atalyse nensatin reatins between aetyl-a an higher alhls r ethanl (Lambrehts & Pretrius, 2; Bell & Henshke, 25; Bissn & Karpel, 21), while ethyl esters are frme by the nensatin reatin between ethanl an either a fusel ayl-a r a fatty ayl-a (Bissn & Karpel, 21). The metabli pathways leaing t the frmatin f these vlatile arma mpuns are reasnably well mappe an establishe. Hwever, the regulatin f the metabli flux thrugh these netwrks is nt well unerst. tuies have shwn that bvius fatrs, suh as the availability f preursrs, fermentatin nitins (Henshke & Jiranek, 1993; Lambrehts & Pretrius, 2; aerens et al., 28; Bissn & Karpel, 21) an the geneti make-up f iniviual strains (les et al., 1982; Rssuw et al., 28), play an imprtant rle in the mulatin f arma prutin. Hwever, little ata is available regaring the impat f envirnmental stress n the prutin f suh metablites. Here we investigate the impat f tw f the mst mmn stresses experiene uring alhli fermentatin, namely temperature shifts an hypersmti stress, n the arma prutin apaity f five wine yeast strains. T ur knwlege, this is the first stuy investigating the impat f hypersmti an transient temperature stresses n the prutin f fermentatin-erive vlatile arma mpuns in syntheti grape must. The aitin f the nn-metablise sugar alhl, srbitl, t the syntheti grape must t impse a hypersmti stress was nsiere t be the mst apprpriate means t stuy the impat n arma prutin withut inreasing metablise sugar levels (Hhmann, 1997). This preliminary stuy prvies infrmatin n hw the expsure t stress affets wine arma an whether the bserve hanges are nserve amng ifferent mmerial aharmyes yeast strains. MATERIAL AD METHOD Yeast strains an grwth nitins Fermentatins were nute using syntheti grape must (1 g/l gluse an 1 g/l frutse), as previusly esribe (Henshke & Jiranek, 1993), with the exeptin f the nitrgen aitins. The nitrgen sures (amin ai an ammnium; Bely et al., 199) were ae t reah a level f 25 mg/l nitrgen. The meium was estimate t have an smti stress f 1.11 sm. The mmerial aharmyes yeast strains use in this stuy were Lalvin E1118, rss Evlutin (referre t as in this paper) (Lalleman In, Mntreal, anaa), T5, V17 an V113 (Anhr Yeast, ape Twn, uth Afria). trains V17, V1 13 an T5 are. erevisiae hybris, whereas E1 118 is. erevisiae var. bayanus (als knwn as Prise e Musse strain). The rss Evlutin strain is a natural rss between. erevisiae strains selete fr speffi winemaking prperties. trains were stre at -8, an were streake ut nt YPD plates, whih subsequently were inubate at 3. A single lny was use t inulate 5 ml f syntheti grape must, whih was aerbially inubate at 3 vernight. These ultures were use t inulate the must at an initial OD f.1, rrespning t -4 x 16 ells/ml. Fermentatin nitins The fermentatins were perfrme in 1 ml glass bttles, ntaining 7 ml f syntheti grape must. The bttles were seale with rubber stppers an a O 2 utlet was prvie. Fermentatins were nute in tripliate at 2 fr 21 ays, unless state therwise. T inue hypersmti stress, srbitl was ae upn inulatin t a final nentratin f.3 M (51; sm) r.5 M (2; sm). T evaluate the impat f temperature stress, a set f fermentatin bttles was expse t 8 (T8) r 37 (T37) fr 16 h either n ay tw (D2) r ay eight (D8) f the fermentatin, an subsequently returne t 2. Fermentatins nt subjete t any aitinal stresses serve as the ntrl ata set. These stresses were hsen t represent nitins that might be experiene uring must fermentatin. Fermentatin bttles were weighe t mnitr the fermentatin prgress as O 2 weight lss. hemial analysis Resiual gluse anfrutse, ethanl an glyerl yntheti wine samples were sanne using the Winesan FT12 instrument (FO Analytial A/ sftware versin 2.2.1) equippe with a purpse-built Mihelsn interfermeter (FO Analytial A/, Hillerø, Denmark), an Furier transfrm mi-infrare (FT-MIR) spetra were generate. Quantifie hemial ata fr resiual gluse, frutse, ethanl an glyerl levels were preite frm infrare spetra by mmerial alibratins r in-huse ajustments using the Winesan FT12 21 versin sftware. Arma mpuns The liqui-liqui extratin was nute as previusly esribe (Luw et al., 29), exept fr the mffiatin where the wine/ether mixture was entrifuge at 4 rpm fr 3 mm. A spatula f a 254 was ae t the mixture t ai extratin in the syntheti wine matrix an the entrifugatin was repeate. The rgani layer was remve an rie. The vlatile higher alhls, esters an fatty ais were quantffie in upliate using a Hewlett Pakar 689 Plus gas hrmatgraph (Agilent, Little Falls, Wilmingtn, A). The gas hrmatgraph was equippe with a split/ splitless injetr, set t a split rati f 15:1, a split flw rate f 98.7 ml/min an a temperature f 2. The separatin f mpuns was ahieve using a J & W DB-FFAP apillary G lumn (Agilent, Little Falls, Wilmingtn, A) with imensins f 6 m length x.32 mm internal iameter, with a.5 pm ating film thikness an using a hyrgen arrier gas flw rate f 6.6 ml/min. Three mirlitres f the extrate sample were injete. The initial ven temperature was maintaine at 33 fr 8 min. Temperature inreases urre at a rate f2l /min. The ven temperature was maintaine at 13 fr 1.3 mm, at 17 fr 1 min an then at 24 fr 2.5 mm. The flame inizatin etetr was perate at 25, with a hyrgen. Afr. J. Enl. Viti., Vl. 35,. 2,214

3 Arma Prue by Yeast ner tress 17 flw f 3 ml/min, xygen f 35 ml/min an make-up gas flw f nitrgen at 3 ml/min. The ven temperature was maintaine at 24 fr 5 min between runs. After 22 injetins, the lunm was leane bth thermally an hemially by a hexane injetin at an ven temperature f 25, with a hling time f 1 min per injetin (Luw et al., 21). Authenti referene stanars (Merk, ape Twn) were use t alibrate fr eah f the mpuns using the internal stanar mpun 4-methyl-2-pentanl. Data lletin an peak integratin were perfrme manually using the HP hemstatin sftware. tatistial analysis Fennentatin ata fr stresse treatments were mpare t that f the 'unstresse' ntrl by reating pair-wise mparisns fr eah strain. These graphs were visualise using ytsape (versin 2.8.2, rg). These graphs nly inlue infrmatin that iffere signffiantly frm the ntrl, thus any missin was eeme t be statistially similar t its ntrl. In all ases a signffiane level f 5% (p <.5) was use. A blue ne (ellipse) r ege (line) iniates a statistially signifiant reutin mpare t the ntrl, whereas a re ne r ege entes an inrease. The lur intensity f the ne r ege is an iniatin f the magnitue f the fl hange. RELT AD DIIO The hypersmti (.3 an.5 M srbitl) an transient temperature stresses (8 an 37 fr 16 hn ays 2 an 8) evaluate were relatively mil in rer t mimi nitins that might be fun uring fermentatins use in the wine inustry. The syntheti grape must was nt fermente t ryness (<4 g/l resiual sugar) within the 21 ays frm the mmenement f the experiment. An impsitin f a stress apparently ha little r n impat n the sugar nentratin fermente ver 21 ays (Table 1). Between 63% an 93% f the sugar was fermente, but there n nsistent pattern in an effet f stress n sugar fermentatin. Expsure f the five yeast strains t stress generally nly resulte in small variatins in ethanl nentratins when mpare t ntrl nitins (ata nt shwn). The fermentatin rates were nly slightly affete by the appliatin f temperature stress (ata nt shwn). Hypersmti stress with.5 M srbitl slightly reue the fermentatin rates f the T5 an V17 strains, but surprisingly nt that f the ther strains (ata nt shwn). V17 has previusly been reprte t ferment high-sugar grape musts prly (Erasmus & Van Vuuren, 29). The fermentatin rates in the presene f.3 M srbitl were similar t the ntrls f all strains. Effet f hypersmti an temperature stress n glyerl an aeti ai prutin Hypersmti stress le t inrease prutin f glyerl an aeti ai (Table 2). mpare t ntrl nitins, signifiantly greater levels f glyerl were prue by all strains as the egree f hypersmti stress was inrease (Fig. 1). urprisingly, the levels f glyerl prue appeare t be muh greater than the egree f hypersmti stress applie (Table 2), whih might be relate t the use f srbitl t impse the hypersmti stress. Aitinal glyerl is prue intraellularly t unter the hypersmti stress in terms f whih the internal slute ntent f the yeast ells is inrease an water uptake is prmte (Hhmann, 1997). Inrease glyerl prutin is ften ampanie by inrease aeti ai prutin ue t a rex imbalane reate uring glyerl prutin, whih results in a surplus f AD (Jain et al., 212). This rex imbalane is rrete by the xiatin f aetalehye t aeti ai, with subsequent regeneratin f ADH. Our results reveale that, in mst instanes, the aeti ai levels prue by the five strains inrease tgether with the greater hypersmti stress impse (Table 2), whih is nsistent with the reatin f the rex imbalane. Hwever, nly strains (.3 M srbitl), E1118 (.5 M srbitl) an V17 (.5 M srbitl) isplaye a statistially signifiant inrease (Fig. 2). ner ntrl an hypersmti stress nitins, strain T5 was fun t prue the highest levels f glyerl (Table 2), whereas strain prue the lwest levels in mst ases. On the ther han, V17 prue the highest aeti ai levels an ntinue t prue higher aeti ai nentratins when the yeast was plae uner hypersmti stress (Table 2). Temperature stress ha little impat n the levels f glyerl an aeti ai prue (Table 3). A small but signifiant inrease in glyerl was prue by V17 when a temperature stress f 37 fr 16 h was applie n ay 2, whereas glyerl prutin by V113 was fun t be signifiantly lwer when the temperature was reue t 8 fr 16 h n ay 8 (Fig. 1). ne f the small hanges in aeti ai levels in respnse t temperature stress was statistially signffiant. As bserve uner hypersmti stress nitins, T5 prue the highest glyerl levels, TABLE 1 Impat f stress n the amunt f sugar (1 g/l gluse + 1 g/l frutse) metablise (mean ± stanar eviatin f tripliate bilgial repeats) after 21 ays f fermentatin by. erevisiae strains. train Treatment ntrl.3 M.5 M Day 2 Day 2 Day 8 Day 8 srbitl srbitl ± ± ± ± ± ± ± 14.7 E ± ± ± ± ± ± ± 2.7 T ± ± ± ± ± ± ± 13.7 VT ± ± ± ± ± ± ±3.9 V ± ± ± ± ± ± ± 9.4. Afr. J. EnL Viti., VL 35,. 2,214

4 171 Arma Prue by Yeast ner tress 2 I, r.9. I 1 I *1 a 4. t n as... n en., ' fl O 5., 11 ' - 11 u1 - '. -.- l i., i.., i., i '., l '.,., i., ' i i l ' en - l ' ' n n n ', O n.9.9 r r ' Ii E I. ' as as 11 ' n O - 5 O O l i i ' ' l q ' fl., 11 q a en en., l., l., l l i ' l ' l vi 4 ' l ' l ' i en l '., ' l ' ' n ', O 11 v l IT. n ' l 4-4 E E 11 w Z L) z ' L) z ).'1. m E L) rj. Afr. J. Enl. Viti., Vl. 35,. 2,214

5 Arma Prue by Yeast ner tress a In q O 11 en In i' 'r n 'r n ' a n ' If O ' O O O I1 ' ' 11 II q, fl fl 5 fl l:i.-i.-.-.-o.-ooo.-ooo.-'l.- - n n, If 11 O \ en en O '. \.- flfl.-' O,.9.- fl O, O fl 'l -4 5 \ If en ' O O en en If en i.i..- ' I1 ' I1.- fl rl T1 Q I! Me 11 ii: H - l! - - ' n n. ' n n ' ' ' n a p n - ' 2 'I as h (' (' fl fl ' \ 't If \ \ fl - ' ' ' n ' ' O ' n fl O O fl,.- 5 en en as en en.- O h fl 't 't \ ' \ '' '' '.-.-' - ' n ' fl 2 h ' If \ O O 't. \ O ' ' II..1 '' O ' n In O O OO O O O E2 1 ; L)E-.E.E L)E-.E.E L)E-.E.E L)E-.E.E L)E' Z ' Z ' Z ' Z ' Z E.E E E.. (5. Afr. J. EnL Viti., VL 35,. 2,214

6 Arma Prue by Yeast ner fress 173 / VI13) VI13_2 T5O_1 _2 r I) :2 T5O_2 (1118 I V17_1 D2.T ii11 3_D8. LA Min , A A #1 +3 Max 15.8 FIGRE 1 ignifiant hanges (p <.5) in glyerl levels mpare t ntrls at the en f 21-ay ultivatin f. erevisiae strains when hypersmti stress treatments f.3 (i) r.5 (2) M srbitl, r transient temperature hanges frm 2 t 37 n ay 2 (D2.T37) r frm 2 t 8 n ay 8 (D8.T8) were applie. A blue ne (ellipse) iniates a reutin mpare t the ntrl, an a re ne an inrease. The lur intensity entes the magnitue f the fl hange bserve. whereas the highest aeti ai levels were prue by V17 uner all temperature stress nitins (Table 3). Other investigatins f the impat f temperature stress n glyerl prutin reprte an inrease in levels by a mmerial wine yeast (strain vaferm LO) when temperature was rapily inrease frm 18 t 34 within 2 h (Kuke et al., 23; Bervi6 et al., 27). In a subsequent reprt, Bervi et al. (27) reprte that the lnger the uratin f the temperature shk (24 h mpare t 4 h), the greater the amunt f glyerl prue. Hwever, the temperature stress was applie within the first 3 h f fermentatin, whereas in ur stuy the stress was nly applie n ay tw r ay eight f the fermentatin, suggesting that the timing f stress appliatin may play a rle in glyerl prutin. When glyerl prutin by 13 wine yeast strains at 12 an 28 was mpare, n nsistent effet f fermentatin temperature n prutin ul be establishe (Gamer et al., 213). Other researhers have fun the impat f reue temperature fermentatin n aeti ai levels t have variable results, pinting t an effet f strains an fermentatin nitins. Mlina et al. (27) fun minimal ifferenes in the final aeti ai levels when strain E1 118 fermente a syntheti must at 15 an 28, whereas Beltran et al. (28) reprte a marke lwer aeti ai level at 13 mpare t 25 prue by their wine yeast (strain Lalleman QA23) in grape an syntheti musts. Effet f hypersmti stress n prutin f arma mpuns Hypersmti stress ha either a small r n impat n the prutin f the ii arma mpuns (aeti ai exlue) evaluate in this stuy (Table 2). Almst n impat f hypersmti stress was fun in the levels f isamyl aetate, ethyl hexanate, iethyl suinate, 2-phenyl aetate, 3-ethxy-i -prpanl, 4-methyl-i-prpanl, butyri ai an isvaleri ai prue by mst strains (Table 2). Hwever, statistial analysis reveale small but signifiant ereases in the respetive levels f ethyl hexanate an 2-phenyl aetate by V17 an T5 (Fig. 2). ignffiant inreases in levels f 3-ethxy-i -prpanl, 4-methyl-i-prpanl were prue by (Fig. 2). Hwever, these signifiant hanges were nt fun in ther strains. Mst strains shwe small ereases f ethyl aprylate, hexani ai an tani ai when hypersmti stress was applie (Table 2). The ereases f ethyl aprylate (strains Eu 18, V17 an T5 fr.5 M srbitl), tani ai (V17,.5 M srbitl) an hexani ai (V17,.5 M srbitl an E1118,.3 M srbitl) were signffiant (Fig. 2). This suggests that the rex imbalane ause by glyerl prutin nly influenes the prutin f arma mpuns t a limite extent uner the fermentatin nitins evaluate. In the absene f hypersmti stress, strain V17 prue the highest levels f mst arma mpuns (exluing aeti ai). Hwever, when hypersmti stress was applie, strain E1 118 prue the highest levels f mst arma mpuns (Table 2). The rex balane in yeast was prpse t regulate the metabli pathways respnsible fr the prutin f arma mpuns (Lambrehts & Pretrius, 2). Hwever, a reent stuy by Jain et al. (212) fun that the rex balane nly seeme t affet sme f the pathways respnsible fr arma mpun prutin. The eletin f. Afr. J. Enl. Viti., Vl. 35,. 2,214

7 Arma Prue by Yeast ner tress 174 Otani Ai isvaleri ai 4-Methyl-i -pentanl 3-ethxy-1 -prpanl Ethyl Hexanate V7 VI3 Aeti Ai [lil VI7 VI7 Diethyl sui nate VI13 32 V13 Butyri E1118 ai i Ethyl aprylate VI 7 T5O 2-Phenylethyl Aetate T5L VI7 Hexani Ai E1118 LA L A Min Max 15.8 FIGRE 2 The signifiant hanges (p <.5) after 21 ays' ultivatin f. erevisiae strains in the levels f arma mpuns ue t a hypersmti stress impse by.3 (i) an.5 (2) M srbitl treatment mpare t their unstresse ntrls. A blue ege (line) iniates a reutin mpare t the ntrl, an a re ege an inrease. The lur intensity entes the magnitue f the fl hange bserve. glyerl-pruing genes in a labratry. erevisiae strain resulte in a signffiant inrease in prutin f the higher alhl isbutanl an a erease in the ester ethyl aetate. The eletin strain wul be expete t prue an exess f ADH, whih in turn ul prmte the prutin f higher alhls in an attempt t reue ADH t AD. This rex imbalane wul hiner the prutin f esters, whih requires AD. Hwever, the nentratins f the higher alhls, isamyl alhl an 2-phenyl alhl, were nt affete by the eletin (Jain et al., 212). Effet f transient temperature inrease n arma mpun prutin The ata shwe that the influene f expsure t a transient temperature inrease frm 2 t 37 fr 16 h n the vlatile arma prfile was time an strain epenent. A transient inrease in temperature n either ay 2 r ay 8 resulte in a erease in hexani ai an tani ai prutin an an inrease in isvaleri ai prutin by mst strains. me f these arma mpun ereases r inreases were signifiant (Fig. 3). Fr example, the erease in hexani ai prue by strains, V17 an E1 118 n ay 2 an strain V17 n ay 8 was signifiant. In aitin, tani ai prue by E1 118 an V17 n ay 2 an V17 n ay 8 erease signffiantly (Fig. 3). The inrease in isvaleri ai prutin by strain E1 118 n ay 2 an strain V113 n ay 8 was als signffiant. Other signifiant inreases r ereases in 4-methyl-ipentanl, iethyl suinate, ethyl suinate an 3-ethxy-1- prpanl n ay 2 r 8 were bserve withut a nsistent pattern (Fig. 3). imilarly, Bervi6 et al. (27) fun that transient temperature inreases frm 18 t 34 in. erevisiae resulte in ne higher alhl (isamyl alhl) nentratin inreasing an anther (prpanl) ereasing. The temperature inreases n bth ays ha n r little impat n the levels f isamyl aetate, ethyl hexanate, ethyl aprylate, 2-phenyl aetate an butyri ai prue by all strains (Table 3). train E1 118 prue the highest levels f eight f the 11 arma mpuns teste when the transient temperature was inrease n ay 2. Hwever, n ay 8, strains E1118 an V113 bth prue five f the highest levels f arma mpuns. Fermentatin at higher temperature (28 ) is reprte t yiel greater nentratins f higher alhls that give flwery armas t wine (Mlina et al., 27). On the ther han, Gamer et al. (213) fun that the higher alhl nentratins prue by 13 wine yeasts after fermentatin at 28 mpare t 12 varie nsierably between strains, pinting t an innsistent respnse pattern t inrease temperature. Effet f transient erease in temperature n arma mpun prutin imilar t the temperature inrease effet, the effet f expsure t a transient temperature erease frm 2. Afr. J. EnL Viti., VL 35,. 2,214

8 . MR 175 Arma Prue by Yeast ner fress t 8 fr 16 h n the vlatile arma prfile was als time an strain epenent (Table 3). Mst f the iethyl suinate levels prue n ay 8 was lwer than the ntrl levels, althugh nne were signifiant. Mst strains prue higher levels f butyri an isvaleri ais when the temperature was erease n ay 2 an 8 (Table 3), but nly the higher levels prue by V113 were statistially signffiant (Fig. 4). On the ther han, sme strains prue lwer levels f tani ai at the lwer temperature mpare t their respetive ntrls, an the lwer levels prue by E1 118 n ay 2 an 8 were signifiant. The signifiant inreases in 3-ethxy-1- prpanl an ethyl aprylate levels by respetively E1 118 an (Fig. 4) were nt fun nsistently in the ther strains (Table 3). A transient erease in temperature n bth ays ha n r little impat n the levels f isamyl aetate, ethyl hexanate, 2-phenyl aetate, 4-methyl-ipentanl an hexani ai prue by all strains (Table 3). train V113 prue the highest levels f six f the 11 arma mpuns teste when the transient temperature was erease n ay 2, but n ay 8 eight f the highest levels were prue by strain (Table 3). In ntrast t ur stuies, fr whih a transient temperature erease was applie, ther stuies have mpare the impat f yeast fermentatin temperature at 25 t 28 an 12 t 15 n the levels f vlatile arma mpuns (Mlina et al., 27; Beltran et al., 28; Gamer et al., 213). The reutin in the yeast fermentatin temperature ha a majr impat n the levels f vlatile arma mpuns prue, an the levels prue varie between yeast strains. Hwever, the ifferenes in the levels f arma mpuns prue at the ambient an lwer temperatures i nt nur in the respnse t lwer temperature stress bserve in this stuy. Fr example, in ntrast t bservatins that ur strains prue lwer levels f tani ai uring erease transient temperature stress (Table 3). Mlina et al. (27) an Beltran et al. (28) fun that their strains prue higher levels when the fermentatin was nute at 13 t 15 mpare t 25 t 28, respetively. Furthermre, Mlina et al. (27) reprte that fermentatin by strain Otani Ai E1 18 VI7 HexanDi Ai VI7 VI7 E11 18 VI7 E1 118 Butyri ai D2T37 E1118 VI7 T5 isvaleri ai Ethyl Hexanate T5O D8.T3 VI13 VI13 'thyi VI13 suin!e T5 VI13 T5O 2-Phenylethyl Aetate T5O VI13 E Methyl-1-pentanl 'V 3-ethxy-1 -prpanl W A Min i-i +3 Max 15.8 FIGRE 3 ignifiant hanges (p <.5) mpare t the respetive ntrls in the levels f arma mpuns prue by. erevisiae strains after 21 ays' fermentatin when a transient 16 h inrease in temperature frm 2 t 37 n ay 2 (D2.T37) an ay 8 (D8.T37) was impse. A blue ege (line) iniates a reutin mpare t the ntrl, an a re ege an inrease. The lur intensity entes the magnitue f the fl hange bserve.. Afr. J. Enl. Viti., Vl. 35,. 2,214

9 Arma Prue by Yeast ner tress ethxy-1 -prpanl Ethyl Hexanate 4-Methyl-1-pentanl Ethyl aprylate isvaleri ai EY 118 VI13 Diethyl suinate VI13 Butyri VI13 VI13 ai VI13 D2.T8 VI13 D8T8 Otani Ai A Min Max 158 A FIGRE 4 ignffiant hanges (p <.5) mpare t respetive ntrls in the levels f arma mpuns prue by. erevisiae strains after 21 ays' fermentatin when a transient 16 It erease f temperature frm 2 t 8 n ay 2 (D2.T8) an ay 8 (D8.T8) was impse. A blue ege (line) iniates a reutin mpare t the ntrl, an a re ege an inrease. The lur intensity entes the magnitue f the fl hange bserve. E1 118 at 15 resulte in higher levels f esters mpare t fermentatin at 28, whereas erease temperature stress in this stuy ha little impat. Hwever, a mparisn f levels f arma mpuns prue by 13 mmerial wine yeast strains at fermentatin temperatures f 12 an 28 shwe wie ranges (Gamer et al., 213), pinting t a strain-speffi respnse t reue temperature. OLIO This is the first stuy t investigate the influene f envirnmental stress n the vlatile arma prfile in a syntheti grape must. The preliminary results emnstrate that the evaluate stresses ause a limite number f signffiant hanges in the levels f 11 arma mpuns. Mst hanges urre when the transient temperature was inrease (25 hanges f 11 measurements), whereas a temperature erease resulte in nine signffiant hanges, an hypersmti stress resulte in 16 hanges f 11 measurements. The hanges bserve als iffere fr eah f the strains, but hanges ul nt be relate t whether the strains were hybris r. erevisiae strains. Hypersmti stress resulte in a greater prutin f glyerl an aeti ai, but the impat n vlatile arma mpun prutin was innsistent. This stress, tgether with the resultant glyerl prutin, wul lea t a rex imbalane, but with an apparent limite impat n vlatile arma mpun prutin. Aitinally, signifiant hanges may have urre in the levels f mpuns nt etete by the meth use in this stuy. In nlusin, expsure t mil envirnmental stress ause a relatively small number f speifi hanges t iniviual mpuns. These hanges iffere signifiantly fr eah strain an stress treatment, suggesting that the fermentatin nitins an signffiantly alter the armati prfile f a wine, althugh these impats annt be preite in general. The hanges t the armati prfile appear t be speffi t eah iniviual wine yeast strain. Furthermre, as arma mpuns interat with eah ther, they may ause signifiant hanges in the verall arma prfile f the resultant wine. Further analysis n the impat f envirnmental stress uring grape must fermentatin, an the uratin f the stress appliatin, wul prvie infrmatin regaring the extent t whih envirnmental stress affets the arma prfile f wine. LITERATRE ITED Alexanre, H. & haq,entier,., Bihemial aspets f stuk an sluggish fennentatin in grape must. J. m. Mirbil. Bitehnl. 2, Attfiel, P.V., tress tlerane: The key t effetive strains f inustrial baker's yeast. at. Bitehnl. 15, Bauer, F.F. & Pretrius, I., 2. Yeast stress respnse an fennentatin effiieny: Hw t survive the making f wine - A review.. Afr. J. Enl. Viti. 21, Beales,., 24. Aaptatin f mirrganisms t l temperatures, weak ai preservatives, lw ph, an hypersmti stress: A review. mpr. Rev. F i. F afety 3, 1-2. Bell,.J. & Henshke, P.A., 25. Impliatins f nitrgen nutritin fr grapes, fermentatin an wine. Aust. J. Grape Wine Res. 11, Afr. J. EnL Viti., VL 35,. 2,214

10 177 Arma Prue by Yeast ner fress Beltran, G., v, M., Guillamón, J.M., Mas, A. & Rzès,., 28. Effet Lambrehts, M. & Pretrius, I., 2. Yeast an its imprtane t wine f fennentatin temperature an ulture meia n the yeast lipi mpsitin arma - A review.. Afr. J. Enl. Viti. 21, an wine vlatile mpuns. mt. J. F Mirbil. 121, Luw, L., Rux, K., Treux, A., Tmi,., aes, 1', ieuwut, H.H. & Bely, M., ablayrlles, J.M. & Barre, P., 199. Autmati etetin Van Rensburg, P. 29. haraterizatin f selete uth Afrian yung f assimilable nitrgen efiienies uring alhli fennentatin in ultivar wines using FTMIR spetrspy, gas hrmatgraphy, an enlgial nitins. J. Fennent. Bieng. 7, multivariate ata analysis. J. Agri. F hem. 57, Bervi, M., Pive, A., Kmerl, T., Wnra, M. & (elan, 9., 27. Influene f heat shk n glyerl prutin in alhl fermentatin. J. Bisi. Bieng. 13, Bissn, L.F., tuk an sluggish fermentatins. Amer. J. Enl. Viti. 5, Bissn, L.F. & Kaq,el, J.E., 21. Genetis f yeast impating wine quality. F i. Tehnl. 1, arras, P., Querl, A. & Del Ohn, M., 21. Analysis f the stress resistane f mmerial wine yeast strains. Arh. Mirbil. 175, Erasmus, D.J. & Van Vuuren, H.J.J., 29. Geneti basis fr smsensitivity an geneti instability f the wine yeast aharmyes erevisiae VJ7. Amer. J. Enl. Viti. 6, Franis. I. & ewtn, J., 25. Determining wine arma frm mpsitinal ata. Aust. J. Grape Wine Res. 11, Gamer, A., Trnhni, J., Querl, A. & Bellh,., 213. Prutin f arma mpuns by rytlerant aharmyes speies an hybris at lw an merate fermentatin temperatures. J. Appl. Mirbil. 114, Gibsn, B.R., Lawrene,.J., Lelaire, J.P.R., Pwell,.D. & mart, K.A., 27. Yeast respnses t stresses assiate with inustrial brewery hanling. FEM Mirbil. Rev. 31, Henshke, P. & Jiranek, V., Yeasts - metablism f nitrgen mpuns. In: Fleet, G.H. (e). Wine Mirbilgy an Bitehnlgy. Harw Aaemi Publishers, witzerlan. pp Hhmann, 5., haping up: The respnse f yeast t hypersmti stress. In: Hhmann.. & Mager, W.H. (es). Yeast stress respnses. pringer, ew Yrk. pp Ivrra,., Perez Ortmn, J.E. & Del Ohn, M., An inverse rrelatin between stress resistane an stuk fermentatins in wine yeasts. A mleular stuy. Bitehnl. Bieng. 64, Jain, V.K., Divl, B., Prir, B.A. & Bauer, F.F., 212. Effet f alternative AW-regenerating pathways n the frmatin f primary an senary arma mpuns in a aharmyes erevisiae glyerl-efetive mutant. Appl. Mirbil. Bitehnl. 93, Kuke, A., Bervi, M., Wnra, M., (elan, 9. & Kmerl, T., 23. Influene f temperature shk n the glyerl prutin in v. auvignn blan fermentatin. Vitis 42, Luw, L., Treux, A., Van Rensburg, P., Ki, M., aes, T. & ieuwut, , 21. Fermentatin-erive arma mpuns in varietal yung wines frm uth Afria.. Afr. J. Enl. Viti. 31, Mager, W.H. & ierius, M., 22. vel insights int the hypersmti stress respnse f yeast. FEM Yeast Res. 2, Maiherbe,., Bauer, F.F. & Du Tit, M., 27. nerstaning prblem fermentatins -A review.. Afr. J. Enl. Viti. 28, Mlina,A.M., wiegers, J.H., Varela,., Pretrius, I. &Agsin, E., 27. Influene f wine fermentatin temperature n the synthesis f yeast-erive vlatile arma mpuns. Appl. Mirbil. Bitehnl. 77, Rapp, A. & Manery, H., Wine arma. Experientia 42, Rapp, A. & Versini, G Influene f nitrgen mpuns in grapes n arma mpuns f wines. Wein-Wiss. 51, Rssuw, D., s, T. & Bauer, F.F., 28. Linking gene regulatin an the ex-metablme: A mparative transriptmis apprah t ientif' genes that impat n the prutin f vlatile arma mpuns in yeast. BM Genmis 9, aerens, 5MG., Delvaux, F., Verstrepen, K.J., Van Dijk P., Thevelein J.M & Delvaux, FR., 28. Parameters affeting ethyl ester prutin by aharmyes erevisiae uring fermentatin. Appl. Envirn. Mirbil. 74, les, R., Ough,. & Kunkee, R., Ester nentratin ifferenes in wine fermente by varius speies an strains f yeasts. Amer. J. Enl. Viti. 33, tyger, G., Prir, B.A. & Bauer, F.F., 211. Wine flavr an arma: A review. J. In. Mirbil. Bitehnl. 38, wiegers, J., Bartwsky, E.J., Henshke, P. & Pretrius, I., 25. Yeast an baterial mulatin f wine arma an flavur. Aust. J. Grape Wine Res. 11, Tamás, M.J. & Hhmann, 5., 23. The hypersmti stress respnse f aharmyes erevisiae. In: Hhmann,. & Mager, P.W.H. (es). Yeast stress respnses. pringer-verlag, Berlin. pp Zuzuarregui, A. & Del Olm, M., 24. Analyses f stress resistane uner labratry nitins nstitute a suitable riterin fr wine yeast seletin. Antnie van Leeuwenhek 85, Afr. J. Enl. Viti., Vl. 35,. 2,214