Stress, fermentation performance and aroma production by yeast

Stress,fermentationperformanceandaromaproductionbyyeast

by

SamanthaFairbairn

ThesispresentedinpartialfulfilmentoftherequirementsforthedegreeofMasterofScience

at

StellenboschUniversityInstituteofWineBiotechnology,FacultyofAgriSciences

Supervisor:AnitaSmitCo‐supervisor:ProfFlorianBauer

March2012

Declaration

Bysubmittingthisthesiselectronically,Ideclarethattheentiretyoftheworkcontainedthereinismyown,originalwork, that I am the sole author thereof (save to theextent explicitlyotherwisestated),thatreproductionandpublicationthereofbyStellenboschUniversitywillnotinfringeanythirdpartyrightsandthatIhavenotpreviouslyinitsentiretyorinpartsubmitteditforobtaininganyqualification.Date:14/12/2011

Copyright©2012StellenboschUniversityAllrightsreserved

Stellenbosch University http://scholar.sun.ac.za

SummaryYeast strains contend with numerous stresses during winemaking. An inability to perceive and

initiatethephysiologicalchangesneededtoadapttostress,hasbeenlinkedtosloworincomplete

(residualsugar>4g/L)fermentations.Wineyeaststrainsdifferingenotype;thisismanifestedas

differencesintheirstresstolerance,andfermentationperformance.

Thefirstgoalofthisstudywastoevaluatehowtheinitialsugar(200or240g/L)andnitrogen

(50,100,250,or400mg/L)content,andthefermentationtemperature(15°Cor20°C)affectedthe

fermentation performance of 17 commercial wine yeast strains. Fermentation performance was

evaluated based on the fermentation kinetics (lag phase, maximum fermentation rate and total

weightlossbyCO2evolution),residualsugarcontentandyeastdryweight.Theresultsdemonstrate

that the fermentationperformancesofcommercialyeastculturesaresignificantlyanddifferently

affectedbyinitialnitrogenandsugarlevels,aswellasthefermentationtemperature.Additionally,

excess nitrogen had a negative impact on the fermentation kinetics and sugar consumption.

Nitrogen deficiency is a common cause of slow and incomplete fermentations, as it affects yeast

growth and thus fermentation rates. Nitrogen supplements are routinely added at the onset of

fermentation, reducing the risk of problematic fermentations. Therefore characterising the

fermentative ability of a strain over a range of oenologically relevant conditions, could aid

winemakers in selecting a yeast strain capable of fermenting a grapemust (of known sugar and

nitrogenlevels)tocompletionatthedesiredfermentationtemperature.

Investigations on fermentation related stress generally focus on its influence on

fermentation rateandsugar consumption.However, fromawinemakingperspective, the strain’s

ability to produce the desired volatile aroma compounds is equally important. Yet, literature

provideslittleinsightintotheinfluencestresshasonthevolatilearomaprofile;thisissurprisingas

winearomaiscloselylinkedtowinequalityandconsumerliking.

Thefinalgoalofthisstudywastoevaluatechangestothevolatilearomaprofilesproduced

by five commercial yeast strains, in response to hyperosmotic and temperature stress. The

concentrationsofthearomacompoundswerequantifiedusingagaschromatographcoupledtoa

flame ionization detector. The results show that hyperosmotic and temperature stress caused

significant changes in the levels of a number of aroma compounds. Furthermore, the changes

observeddifferedamong theevaluated strains, aswell as for the fermentation stress treatments

studied.

Futureaimsshouldbedirectedtowardsthepotentialapplicationofyeaststrainselectionas

a means to avoid problematic fermentations in grape must; in addition to the further

characterisationoftherelationshipbetweenstressandtheresultantvolatilearomaprofileinwine.


Opsomming

Gisrasse moet verskeie stresfaktore afweer tydens die wynmaak proses. Die onvermoë van ‘n

wyngisomstreswaarteneemendienodigefisiologieseveranderingeteinisieeromaantepasby

die strestoestande wordmet slepende of onvolledige fermentasies (met ‘n residuele suiker van

meeras4g/L)geassosieer.Wyngisrasseverkilingenotipe;watasgrootverskilleindiegraadvan

strestoleransie,endusookfermentasiesuksesgeopenbaarword.

Dieeerstedoelwitvanhierdiestudiewasomteevalueerhoediesuiker(200of240g/L)en

stikstof (50, 100, 250, of 400 mg/L), asook die fermentasie temperatuur (15°C of 20°C) die

fermentasieprestasievan17kommersiëlewyngiskulturebeïnvloed.Diesuksesvanfermentasieis

geëvalueer op grond van fermentasie kinetika (sloerfase,maksimum fermentasiespoed en totale

gewigsverliesasCO2verlies),dieresiduelesuikerinhoudendiegisdroëmassa.

Die resultate demonstreer dat die fermentasie sukses van kommersiële giskulture

beduidend en verskillend beïnvloed word deur die aanvangsstikstof en – suikerkonsentrasies,

asook die fermentasie temperatuur. Daarbenewens,wanneer stikstof in oormaat teenwoordig is

kandit‘nnegatieweimpakopfermentasietempoensuikermetabolismehê.Beperkendevlakkevan

stikstof‘nalgemeneoorsaakvanslependeofonvolledigefermentasies,aangesienstikstofdiegroei

en gevolglik ook die fermentasiespoed van gis beïnvloed. Stikstofaanvullings word dikwels tot

druiwemostoegevoegaandiebeginvangisting,watdierisikovanprobleemfermentasiesverlaag.

Dus kan die karakterisering van die fermentasievermoë van ‘n gisras vir ‘n reeks wynkundig

relevantekondisiesdiewynmakerhelpom‘ngisrasteselekteerwat instaat isom ‘ndruiwemos

(waarvandiesuikerenstikstofvlakkebekendis)droogtegisbydiegewenstetemperatuur.

Meeste studies wat fermentasieverwante stress ondersoek, fokus op die die invloed

daarvanopfermentasietempoensuikerverbruik.Van‘nwynmaakperspektiefisdiegissevermoë

omdiegewensdevlugtigearomakomponenteteproduseeregterewebelangrikasdievermoëom

fermentasietevoltooi.Togverskafdieliteratuurmininsigtotdieinvloedvanstresopdievlugtige

aromaprofiel; wat verbasend is aangesien die aromaprofiel ‘n belangrike faktor is van die

waargenomewynkwaliteitendaaromookverbruikersvoorkeur.

Diefinaledoelwitvanhierdieprojekwasomdieveranderingetotdievlugtigearomaprofiel

geproduseer deur vyf kommersiële gisrasse in reaksie op hiperosmotiese stres en temperatuur

stres te evalueer. Die konsentrasies van die aromakomponente is gekwantifiseer deur gas

chromatografiegekoppelaanvlam‐ioniserendedeteksie.Dieresultatewysdathiperosmotiese‐en

temperatuur stres beduidende veranderinge meebring in die vlakke van ‘n aantal

aromakomponente.Verder isdiewaargenomeveranderingeookverskillendvirdiegeëvalueerde

gisrasse,asookvirdieverskillestresbehandelingswatondersoekis.


Toekomstigestudiesbehoortgerigteweesopdietoepassingvangisseleksieompotensiële

probleemfermentasies in druiwemos te voorkom; asook die verdere karakterisering van die

verhoudingtussenomgewingstresfaktoreendiegevolglikevlugtigearomaprofielinwyn.


Thisthesisisdedicatedto

Myfamily


BiographicalsketchSamantha Fairbairn was born (29 June 1979) in Cape Town and matriculated from HottentotsHollandHigh School in 1997. She enrolled at theUniversity of Stellenbosch, andobtained aBSc.HonoursdegreeinMicrobiologyin2002.Samanthahassinceworkedasateacherandatechnicalassistant before enrolling at Stellenbosch University for a MSc in Wine Biotechnology.


AcknowledgementsIwishtoexpressmysinceregratitudeandheartfeltappreciationtothefollowingindividualsandinstitutions:

Anita Smit andProfFlorianBauer, who asmy supervisors, provided guidance, advice,muchneededencouragement,inadditiontothecriticalevaluationofthismanuscript

AnchorYeastandTHRIPforfundingthisstudy

The InstituteofWineBiotechnologymanagement for affordingme theopportunity tofurthermystudieswhileintheiremploy

ProfMartin Kidd and Dan Jacobson for their help with statistical data analyses andinterpretation

Lauren Jooste andCandiceStilwaney,my fermentation lab co‐workers and friends, fortheirinvaluableassistance,unwaveringunderstanding,encouragement,andsupport

All my friends and colleagues at the Institute of Wine Biotechnology, especiallyLynnEngelbrecht,AlexisEschtruth,TalithaMostertandEldaLerm,forencouragement,coffeebreaks,invaluablediscussions,andmuchneededadvice

Myfamilyandfriendsfortheirlove,empathyandencouragement

MyHeavenlyFather,whorenewsmystrengthdailyandrefreshesmyhopesandfaith inblessedfuture


PrefaceThis thesis is presented as a compilation of 5 chapters. Each chapter is introduced separately.Chapter 3 is written according to the style of the journal South African Journal of Enology andViticulture,andchapter4toAppliedMicrobiologyandBiotechnology.Chapter1 Generalintroductionandprojectaims Chapter2 Literaturereview Winefermentationstress Chapter3 Researchresults Theimpactandinteractionbetweeninitialnitrogen,initialsugar,andtemperature

onthefermentationperformanceofcommercialwineyeast Chapter4 Researchresults Impactofenvironmentalstressonaromaproductionduringwinefermentation Chapter5 Generaldiscussionandconclusions


i

Contents

Chapter1Generalintroductionandprojectaims 1

1.1 Introduction 2

1.2 ProjectAims 3

1.3 Literaturecited 4

Chapter2Winefermentationstress 6

2.1 Introduction 7

2.2 Generalstressresponse 8

2.3 Hyperosmoticstress 10

2.3.1 Physiologicalimpactofhyperosmoticstress 102.3.2 Acquisitionofhyperosmoticstresstolerance 102.3.3 Osmoticstressandredoxbalance:Impactonwinearoma 11

2.4 Ethanoltoxicity 14

2.4.1 Physiologicalimpactofethanoltoxicity 14 2.4.2 Acquisitionofethanoltolerance 14 2.4.3 Potentialroleofethanolinwinearoma 162.5 Lowtemperature 16 2.5.1 Physiologicalimpactoflowtemperature 16 2.5.2 Acquisitionoflowtemperaturetolerance 172.6 Conclusion 18

2.7Literaturecited 18

Chapter3Theimpactandinteractionbetweeninitialnitrogen,initialsugar,andtemperatureonthefermentationperformanceofcommercialwineyeast 23

3.1 Introduction 24

3.2 Materialsandmethods 27

3.2.1 Syntheticgrapemediumandfermentationtreatments 273.2.2 Dryweight 273.2.3 Statisticalanalysis 28

3.3 Resultsanddiscussion 31

3.3.1 Influenceoftreatmentsonfermentationkinetics 313.3.1.1Impactoftreatmentsonfermentationonset 31

3.3.1.2Impactoftreatmentsonfermentationonsetofindividualyeaststrains 32

3.3.1.3Impactoftreatmentsonfermentationrate 36

3.3.1.4Impactoftreatmentsonfermentationrateofindividualyeaststrains 38

3.3.1.5Impactoftreatmentsonthetotalweightloss 41


ii

3.3.1.6Impactoftreatmentsonthetotalweightlossofindividualyeaststrains 41

3.3.2 Influenceoftreatmentsonsugarconsumption 453.3.2.1Impactoftreatmentsonsugarconsumption 45

3.3.2.2Impactoftreatmentsonsugarconsumptionofindividualyeaststrains 46

3.3.3 Influenceoftreatmentsonyeastdryweight 493.3.3.1Impactoftreatmentsonyeastdryweightproduction 49

3.3.3.2Impactoftreatmentsonthedryweightproductionofindividualyeaststrains 49

3.4 Conclusion 53

3.5 Acknowledgements 54


Chapter4Impactofenvironmentalstressonaromaproductionduringwinefermentation 58

4.1 Introduction 60

4.2 Materialsandmethods 61

4.2.1 Yeaststrainsandgrowthconditions 614.2.2 Fermentationconditions 624.2.3 Chemicalanalysis 62 4.2.3.1Residualglucoseandfructose,andglycerol 62

4.2.3.2Aromacompounds 63

4.2.4 Statisticalanalysis 634.3 Results 65 4.3.1 Fermentationperformance 65 4.3.1.1Fermentationrate 65

4.3.1.2Residualglucoseandfructose,andglycerol 65

4.3.2 Aromacompounds 694.4 Discussion 74

4.5 Acknowledgements 76

4.6 References 77

Chapter5Generaldiscussionandconclusions 805.1 Introduction 81



1

Chapter1

Introduction

andprojectaims


2

Chapter1

Generalintroductionandprojectaims

1.1 Introduction

Commercialwineyeaststrainsareprimarilyselectedfortheirabilitytofermentgrapemustto

dryness,wherebyresidualsugarsarereducedtolessthan4g/L(Pretorius,2000).Foragrape

must to be fermented to dryness, the yeast strain is required to adapt and respond to a

multitudeofenvironmentalstresses,eithersimultaneouslyorsuccessively.Theenvironmental

stresses commonly encountered during commercialwine fermentations include a high initial

sugarconcentration,lownitrogenlevels,andpossiblechangesinthefermentationtemperature,

among others. Stress is described as any factor that reduces cell growth. Stress is therefore

often linked to problematic fermentations, which are defined by either a slow rate of sugar

consumption (sluggish fermentations), or as an incomplete fermentationwith ahigh residual

sugar content (stuck fermentations) (Alexandre&Charpentier, 1998;Bisson,1999;Gibsonet

al.,2007;Malherbeetal.,2007).Thecausesofproblematicfermentationshavebeenthesubject

of extensive study due to their economic and logistic consequences (Malherbe et al., 2007;

Pizarroetal.,2007).

Yeast strains differ in their innate abilities to sense and effectively adapt to stressful

environmental conditions.This ability contributes to the cell’s survival, and therefore also its

abilitytofermentgrapemust(Ivorraetal.,1999;Carrascoetal.,2001;Zuzuarregui&delOlmo,

2004). Additionally, yeast strains vary in their nitrogen requirements, and their capacity to

catabolise sugars (Manginotetal., 1998). This variation in yeast strainmetabolic capabilities

highlights the importance of determining the grape must composition (at least sugar and

nitrogencontent),andusingthatinformationalongwiththeintendedfermentationconditions

toselectanappropriateyeaststarterculture.Theselectionoftheyeaststrain“best”adaptedto

fermenting a characterised grape must to dryness may reduce the risk of problematic

fermentations.However, theability of a strain to ferment grapemust to completion isonly a

reflectionofyeast fermentationperformanceanddoesnotprovide insights intothequalityof

thefinalproduct.

Fromtheconsumer’sperspectivewinequalityisessentialinmakingapurchasingdecision

(Swiegers et al., 2005). Wine flavour, consisting of aroma, taste, and mouth‐feel, is a vital

componentofwinequality(Francis&Newton,2005;Swiegersetal.,2005).Wineflavoursare

derived from the grapes, produced during alcoholic fermentation, and depend on the


3

maturation strategy that isused (Rapp&Mandery,1986;Rapp&Versini,1996).Thevolatile

aromasproducedbywineyeastduringfermentationincludehigheralcohols,estersandvolatile

fatty acids, among others (Rapp & Versini, 1996). The availability of aroma precursors, the

fermentation conditions (Henschke & Jiranek, 1993; Rapp & Versini, 1996; Lambrechts &

Pretorius, 2000; Swiegers et al., 2005; Vilanova et al., 2007; Saerens et al., 2008; Bisson &

Karpel,2010)andthegenotypeoftheyeaststrain(Solesetal.,1982;Rossouwetal.,2008)all

contribute to which particular volatile aroma compounds will be produced by yeast during

alcoholic fermentation. However, currently little data is available regarding the impact of

environmentalstressontheproductionofvolatilearomametabolites.

ThisstudyispartofabroaderprogramattheInstituteofWineBiotechnology,Stellenbosch

University,investigatingyeastnutritionalrequirementsinthewinematrix,anditsinfluenceon

fermentationperformanceandaromaproductionunderwinemakingconditions.

1.2 Projectaims

This study focused on the influence of some of the factors commonly associated with

problematic grape must fermentations on fermentation performance of commercial yeast

startercultures,inadditiontotheimpactofenvironmentalstressonthevolatilearomaprofile.

Themainaimsofthisprojectweretherefore:

i. Toinvestigatetheimpactoftheinitialnitrogencontent,osmoticpressure(initialsugar

content), and temperature on the fermentation performance of 17 commercial active

dryyeastcultures.Toourknowledgethisisthefirststudytoevaluatestressbyvarying

stress levels and combinations using a multifactorial experimental design. This will

potentially identify strains that are suited to fermenting a grape must of specific

characteristics,providingwinemakerswithatooltoselecttheyeaststrainbestadapted

tofermentthatspecificgrapemustandensuringacompletefermentation.

ii. To elucidate the effect of hyperosmotic and temperature stresses on fermentation

performance and the production of fermentation derived volatile aroma compounds,

providing informationwhether stress exposure impactswine aroma andwhether the

observed changes are conserved among different commercial Saccharomyces yeast

strains.Toourknowledgethisisthefirststudytoassesstheinfluenceofstressonthe

fermentationaromaunderwinemakingconditionsinasyntheticwinematrix.


4

1.3 Literaturecited

Alexandre,H.&Charpentier,C.,1998.Biochemicalaspectsofstuckandsluggishfermentationingrapemust.J.Ind.Microbiol.Biotechnol.20,20‐27.

Bisson,L.F.,1999.Stuckandsluggishfermentations.Am.J.Enol.Vitic.50,107‐119.

Bisson,L.F.&Karpel, J.E.,2010.Geneticsofyeastimpactingwinequality.FoodSci.Technol.1,139‐162.

Carrasco, P., Querol, A.& del Olmo,M., 2001. Analysis of the stress resistance of commercialwineyeaststrains.Arch.Microbiol.175,450‐457.

Francis,I.&Newton,J.,2005.Determiningwinearomafromcompositionaldata.Aust.J.GrapeWineRes.11,114‐126.

Gibson,B.R.,Lawrence,S.J.,Leclaire,J.P.R.,Powell,C.D.&Smart,K.A.,2007.Yeastresponsestostressesassociatedwithindustrialbreweryhandling.FEMSMicrobiol.Rev.31,535‐569.

Henschke,P.&Jiranek,V.,1993.Yeasts–metabolismofnitrogencompounds.In:Fleet,G.H.(ed).WineMicrobiol.Biotechnol.HarwoodAcademicPublishers,Switzerland.pp.77‐164.

Ivorra, C., Pérez Ortín, J.E. & del Olmo, M., 1999. An inverse correlation between stressresistance and stuck fermentations inwineyeasts.Amolecular study.Biotechnol.Bioeng.64,698‐708.

Lambrechts,M.&Pretorius,I.,2000.Yeastanditsimportancetowinearoma‐areview.S.Afr.J.Enol.Vitic.21,97‐129.

Malherbe,S.,Bauer,F.&DuToit,M.,2007.Understandingproblemfermentations–Areview.S.Afr.J.Enol.Vitic.28,169‐185.

Manginot, C., Roustan, J. & Sablayrolles, J., 1998. Nitrogen demand of different yeast strainsduringalcoholicfermentation.Importanceofthestationaryphase.Enzym.Microb.Technol.23,511‐517.

Pizarro, F., Varela, C., Martabit, C., Bruno, C., Pérez‐Correa, J.R. & Agosin, E., 2007. Couplingkinetic expressions and metabolic networks for predicting wine fermentations. Biotechnol.Bioeng.98,986‐998.

Pretorius,I.,2000.Tailoringwineyeastforthenewmillennium:novelapproachedtotheartofwinemaking.Yeast.16,675‐729.

Rapp,A.&Mandery,H.,1986.Winearoma.Cel.Mol.LifeSci.42,873‐884.

Rapp,A.&Versini,G.,1996.Influenceofnitrogencompoundsingrapesonaromacompoundsofwines.Wein‐Wiss.51,193‐203.

Rossouw, D., Næs, T. & Bauer, F., 2008. Linking gene regulation and the exo‐metabolome: Acomparative transcriptomics approach to identify genes that impact on the production ofvolatilearomacompoundsinyeast.BMCGenomics.9,530‐547.

Saerens,S.,Delvaux,F.,Verstrepen,K.,VanDijck,P.,Thevelein,J.&Delvaux,F.,2008.ParametersaffectingethylesterproductionbySaccharomycescerevisiaeduringfermentation.Appl.Environ.Microbiol.74,454‐461.


5

Soles, R., Ough, C. & Kunkee, R., 1982. Ester concentration differences inwine fermented byvariousspeciesandstrainsofyeasts.Am.J.Enol.Vitic.33,94‐98.

Swiegers,J.,Bartowsky,E.J.,Henschke,P.&Pretorius,I.S.,2005.Yeastandbacterialmodulationofwinearomaandflavour.Aust.J.GrapeWineRes.11,139‐173.

Vilanova,M.,Ugliano,M.,Varela,C.,Siebert,T.,Pretorius,I.S.&Henschke,P.A.,2007.Assimilablenitrogen utilisation and production of volatile and non‐volatile compounds in chemicallydefinedmediumbySaccharomyces cerevisiaewineyeasts.Appl.Environ.Biotechnol. 77,145‐157.

Zuzuarregui,A.&delOlmo,M.,2004.Analysesofstressresistanceunderlaboratoryconditionsconstituteasuitablecriterionforwineyeastselection.AntonievanLeeuwenhoek.85,271‐280.


6

Chapter2

Literaturereview

Winefermentationstress


7

Chapter2

Winefermentationstress

2.1 Introduction

Stressisanyenvironmentalconditioncompromisingacell’scapacitytosurviveandgrow(Ruis

& Schüller, 1995; Hohmann&Mager, 2003). From the onset of alcoholic fermentation, yeast

cells are bombardedwithmany such stresses experienced either simultaneously or in quick

successionofeachother.Furthermore, the inhibitory influenceofan individualstressmaybe

magnified when present in combination with other stresses (Bisson, 1999). Fermentation

associated stresses include hyperosmotic stress (high initial sugar), ethanol toxicity,

temperature extremes, low pH and nutrient limitations (nitrogen, oxygen, vitamins, and

minerals)(Attfield,1997;Bisson,1999;Gibsonetal.,2007;Malherbeetal.,2007).

Commercialwineyeaststrainsareselectedbasedontheirabilitytoconductreliableand

completefermentationsinadditiontotheproductionofdesirablearomacompounds(Pretorius,

2000;Fleet,2008).Aninabilitytoovercomefermentationrelatedstressesisoftenmanifestedas

astuckorasluggishfermentation.Astuckfermentationendsprematurelywithahighresidual

sugarcontent,whereasasluggishfermentationproceedsataveryslowrate(Bisson,1999).

Yeast cells have developed mechanisms to sense environmental cues, and initiate

physiologicalresponsestocounteracttheharmfuleffectsofstress.Theinherentabilityofyeast

strains to conduct fermentation has been inversely correlated with their intrinsic stress

tolerance(Ivorraetal.,1999;Carrascoetal.,2001;Zuzuarregui&delOlmo,2004).Sinceyeast

strainsdifferintheirabilitytodetectandrespondtothedifferentstressesexperiencedduring

fermentation, stress tolerancecouldbeameaningfulmeansof screeningpotentialwineyeast

startercultures(Zuzuarregui&delOlmo,2004).

While the ability to complete fermentation of a high sugar, low pH grape juice is the

mostimportantattributeofwineyeaststrains,yeaststrainsalsocontributesignificantlytowine

flavour,andthereforequality,ofthefinalwine(Francis&Newton,2005;Swiegersetal.,2005).

This impact of yeast is mainly linked to the de novo production of volatile and non‐volatile

aroma and flavour compounds such as esters, aldehydes, higher alcohols and organic acids

(Rapp&Versini,1996).Whilemanystudieshavefocusedontheimpactofstressonthekinetics

andcompletionoffermentation,thereisverylittlepublishedinformationontheimpactofthe

majorfermentationstressesonaromaproductionbyindividualyeaststrains.


8

This chapter will primarily focus on reviewing the general stress response and the

physiologicalchangesthatyeastcellsundergowhenexposedtothemajorstressesexperienced

duringwine fermentation,hyperosmotic stress, ethanol toxicity, andchangesof, orgrowthat

lowtemperatures.Wherepossible,thereviewwillalsoproposehowsuchstressesmayimpact

ontheproductionofaromaticcompounds,basedonourexistingknowledgeoftheregulationof

therelevantgeneticandmetabolicnetworks.

2.2 Generalstressresponse

Stresshasfarreachingconsequencesontheyeastcell.Itmaycompromisemembraneintegrity,

cause a loss of protein and enzyme function, and lead to growth inhibition (Attfield, 1997;

Gasch,2003).

Exposure to stress induces a general and a specific gene expression response, which

mayresult inan improvedtolerancetosubsequentstressexposures.Theexpressionofgenes

responsible for repairing the ravages of stress and the acquisition of stress tolerance are

coordinatedviaanumberofsignaltransductionpathwaysandspecifictranscriptionfactorsthat

activate several gene expression response elements, the stress response element (STRE), the

heatshockelement(HSE)andtheAP‐1responseelement(oxidativestress)(Estruch,2000).

STREmediates the expression of a number of genes falling within the general stress

response(Martinez‐Pastoretal.,1996).TheSTREisfoundinthepromoterregionofnumerous

stress‐responsivegenesandconsistsofeitherasingleormultiplecopiesoftheCCCCT/AGGGG

nucleotidesequence. It servesasabindingsite for the transcription factorsMsn2porMsn4p,

whichareactivatedbyheatshock,osmoticstress,oxidativestressandnutrientstarvation(Ruis

&Schüller,1995).However,theinitiationofgeneexpressionbyMsn2porMsn4pviatheSTRE

doesnotalwaysfallwithinthegeneralstressresponse.Inresponsetohyperosmoticstress,the

high osmolarity glycerol pathway (HOG)modulates the expression of genes containing STRE.

This does not formpart of the general stress response but is an example of a specific stress

response(Figure1).

The heat shock element (HSE) is a second transcription element activated by stress. The

HSEcontainsatleastthreenucleotidesequencerepeatsofnGAAn,andservesasthebindingsite

for the heat shock transcription factor (Hsf1) (Morimoto, 1998; Bauer & Pretorius, 2000;

Estruch, 2000; Rangel, 2010). The Hsf1 induces the expression of heat shock proteins (HSP)

which generally serve as molecular chaperones that aid protein folding. Many of these

molecular chaperones are present in the absence of stress. However, their expression is

increaseduponexposures to stress.TheexpressionofHSP’sare inducedbyother stresses in


9

addition to temperature stress; therefore they also contribute to “general” stress tolerance

(Bauer&Pretorius,2000;Estruch,2000;Rangel,2010).

Msn2/4p

Specific response

Hog1 pathway

Environmental stress

response Acquisition of

STRESS TOLERANCE

Starvation

OxidationpH

Cold stress

Hyperosmotic stress

General response

Heat shock

Hsf1

Figure1Environmentalstressesinitiatebybothageneralandaspecificstressresponse;mayresultinanoverallimprovedstresstolerance.AdaptedfromSideriusandMager(1997)andTeixeiraetal.,(2010)

Genomic expression studies characterising yeast response to numerous stresses,

(oxidative and osmotic stress, heat shock, nitrogen starvation, and stationary phase) have

identified a shared environmental stress response (ESR) (Gasch et al., 2000; Causton et al.,

2001;Gasch,2003).ThisESRischaracterisedbytherepressionofgenesresponsibleforcellular

growth and protein synthesis (Gasch et al., 2000; Causton et al., 2001; Gasch & Werner‐

Washburne, 2002). It is also characterised by an increased expression of genes involved in

carbohydrate metabolism (for energy, glycogen and trehalose generation), fatty acid

metabolism, protein folding and degradation (heat shock proteins), nucleic acid repair, the

maintenance of the internal osmolarity, cytoskeleton reorganisation, signalling, defence to

reactiveoxygenspeciesandmaintenanceofredoxpotential(Gaschetal.,2000;Caustonetal.,

2001). The expression studies conducted by Gasch et al., (2000) and Causton et al., (2001)

evaluatedeachstressindividually,andobservedatransientresponsetostress.However,under

fermentative conditions, where the cell is continually adapting to the ever changing

environment,thislongtermresponseistermedthefermentationstressresponse(Marksetal.,

2008).Inthiscasetheglobalexpressionchangeswereobservedthroughoutthefermentation,

whichisindicativeofthedynamicnatureofmustfermentation(Marksetal.,2008).

Thisgeneral stress response,or ratherESR,hasbeenproposedas themechanismbehind

cross‐protection (Gaschetal., 2000; Caustonetal., 2001;Gasch&Werner‐Washburne, 2002;


10

Gasch, 2003). The exposure tomild levels of stress initiates physiological changes to the cell

conferring tolerance to subsequent exposuresof lethal levels of the same stress andpossibly

also to other stresses. This cross protection also suggests the existence of a general stress

response, which can be initiated by an array of environmental stresses (pH, heat, osmotic,

nitrogen starvation and oxidation) (Ruis & Schüller, 1995; Martinez‐Pastor et al., 1996). A

recent studyreported thatMsn2pandMsn4p initiategeneexpression individually ina stress

specificmanner,andsuchspecificinductionofgeneexpressionmayrequireareassessmentof

thenatureofthe“generic”generalstressresponse(Berry&Gasch,2008).Furthermore,astudy

on the genome expressionprofiles of deletionmutants in response to various typesof stress

revealed that only a small number of the geneproducts thatwere expressed are required to

adapt to thecurrent stress (Giaeveretal.,2002).BerryandGasch (2008)assert thata single

stressexposureinducesgeneexpressionprimarilydirectedtowardsthepre‐emptiveprotection

ofthecellfromfuturestress,whichwouldaccountfortheexistenceofcross‐protection.

2.3 HyperosmoticStress

2.3.1 Physiologicalimpactofhyperosmoticstress

In winemaking, the high initial sugar concentration causes a hyperosmotic stress response

immediatelyuponyeastinoculationintograpemust.Thesuddenlossofcellvolume(orturgor

pressure) damages the plasmamembrane in terms of its structure and permeability (Wood,

1999),aswellastheactincytoskeleton.Theactincytoskeletonisvitalforbudding,anddamage

to this network would also contribute to the cessation of growth observed following

hyperosmoticstress(Hohmann,1997;Tamás&Hohmann,2003).

2.3.2 Acquisitionofhyperosmoticstresstolerance

Inanattempttostemwatereffluxfromthecell,waterisreleasedfromthevacuole,providing

sometimeforthecelltoadapttoitsenvironment(Hohmann,1997;Attfield,1998).Tolimitthe

lossofwater, theglycerolexportchannelFps1closes,andglycerolproductionisactivatedvia

the high osmolarity glycerol pathway (HOG). Glycerol serves as an osmoprotectant by

increasing the internal solute concentration, and in sodoing it limits the efflux ofwater.The

accumulation of glycerol continues until the influx ofwater restores the cell size to a critical

level and yeast growth canbe resumed (Hohmann, 1997;Repetal., 2000;Mager& Siderius,

2002; Tamás & Hohmann, 2003; Hohmann et al., 2007). Consequently, the higher the initial

sugarconcentrationthelongerthelagphasebeforefermentationcommences.


11

Thecapacitytosenseandcounteracttheilleffectsofosmoticstressisdeterminedbystrain

genotype.Ifthecellisunabletorespondrapidlytohyperosmoticstress,itmayresultinstuckor

sluggishfermentations(Lafon‐Lafourcade,1983;Llauradóetal.,2002;Gibsonetal.,2007).

A genome expression study on the adaptation of Saccharomyces cerevisiae to high sugar

stress found that in addition to glycerol uptake and synthesis, the genes for trehalose, and

glycogensynthesiswerealsoupregulated(Repetal.,2000;Erasmusetal.,2003).Trehaloseisa

disaccharideaccumulatedinresponsetoosmoticstress,oxidativestress,heatstress,coldshock,

dehydration, carbon starvation, as well as during the stationary phase (Hounsa et al., 1998;

Caustonetal.,2001;Erasmusetal.,2003;Rangel,2010).Itbindstoproteins,preventingprotein

denaturationandaggregation,aswellasdecreasingmembranepermeability.Trehalose levels

donotnecessarily correlatewith its synthesis, as it isusually rapidlydegraded, releasing the

proteinsandallowingHSP’s to facilitate the foldingofnativeordenaturedproteins (Singer&

Lindquist,1998).Glycogenisastoragecarbohydrateplayingacrucialroleincellsurvivalduring

periods of nutrient limitation. Additionally, glycogen catabolism has been linked with the

formationofsterols(Pretorius,2000),whichareinturnassociatedwithimprovedcellvitality

aswellasanoverallimprovementinethanoltolerance(Pretorius,2000;Gibsonetal.,2007).

2.3.3 Osmoticstressandredoxbalance:Impactonwinearoma

Glycerolisproducedinresponsetoosmoticstressandasaprecursortophospholipid(aplasma

membranecomponent)formation.However,duringalcoholicfermentationitisprimarilyaby‐

product of themaintenance of the redox balance, specifically the relative levels ofNAD+ and

NADH(Figure2)(Hohmann,1997).Thecatabolismofsugartoethanolcausesnonetchangein

thelevelsofNAD+/NADHsincetheNAD+reducedduringglycolysisisre‐oxidisedwhenethanol

isproduced fromacetaldehyde.However, several intermediatesof thispathway, inparticular

pyruvic acid, aremetabolisedby alternativepathways resulting in an imbalance in the redox

potential. Elevated levels of either NAD+ or NADHmust be reversed in order for the cell to

continue to growand ferment. In the caseof aNADHsurplus, this generally achievedvia the

productionofglycerol,wherebyNADHisreducedtoNAD+(Hohmann,1997).

When exposed to osmotic stress, cells respond with the accumulation of glycerol, which

resultsinasurplusofNAD+.Thisredoximbalanceiscorrectedbytheoxidationofacetaldehyde

toaceticacidtoavoidtheadditionalproductionofNAD+viathemetabolismofacetaldehydeto

ethanol. Acetic acid production has been linked to the initial sugar content, and constitutes

approximately90%ofthevolatileacidsinwine.Thisincreaseintheproductionofaceticacid


12

hasanegativeimpactonwineflavour,as itmayimpartavinegarcharactertothewinewhen

presentabovethesensorythreshold(Swiegers&Pretorius,2005).

Glucose

Fructose‐1,6‐Bisphosphate

Dihydroxyacetonephosphate

Glycerol‐3‐phosphate

Glycerol

Glyceraldehyde‐3‐phosphate

Pyruvate

Acetaldehyde

Ethanol

Acetic acid

NADH

NAD+

NAD+

NADH

NADH

NAD+

NAD+

NADH

ATPADP

Produces 1 NAD+ and 1 NADH=No net change

Produces 1 NAD+Produces 1 NADH

Figure 2 A simplified representation of glycolysis, illustrating the driving force of redox balance inmetabolism.AdaptedfromHohmann,(1997)

Themetabolic networks leading to the formation of other volatile aroma compounds are

reasonablyunderstoodanddescribed(Lambrechts&Pretorius,2000).However,theregulation

of these networks is notwell characterised. It has been suggested that redox homeostasis is

involved in the regulation of these aroma producing networks (Lambrechts & Pretorius,

2000).Jainetal.,(2011)evaluatedtheinfluenceofthesubstitutionoftheglycerolbiosynthetic

pathway with alternative NAD+ regenerating pathways, on the production of primary and

secondarymetabolites.Comparedtothewildtypestrain,thegrowthofthemutantstrainswere

significantly affected by the redox imbalance. The alternative NAD+ producing pathways

providedaslight improvement inyeastgrowth;however theycouldnotmatch thegrowthof

thewild type.The imbalance inNAD+/NADH levels, generallydrove theproductionofhigher

alcohols (isobutanol) in an attempt to reduceNADH toNAD+ (Figure 3), this is in agreement

withtheresultsreportedbyStygeretal.,(2011).Theproductionofafuselalcoholversusafusel

acids is therefore also dependent on the redox requirements of the yeast (Bisson & Karpel,

2010).TheproductionofestersrequiresNAD+,henceashortagethereofresultedinlowerlevels


13

ofestersproducedbythemutantstrainsthanthewildtypestrain(Jainetal.,2011).Inawine

context, the accumulation of NAD+ as a consequence of glycerol production may therefore

influencetheproductionofestersandfuselacids(Figure3)inanattempttorestoretheredox

balance.

Amino Acid

α‐keto acid

“Fusel Aldehyde”

Fusel AcidFusel Alcohol

Pyruvate fromglycolysis

Produces 1 NAD+ Produces 1 NADH

2‐oxoglutarateglutamate

CO2

NAD+

NADH

NADH

NAD+

Figure3TheEhrlichpathway:The importanceof redoxbalancing in the catabolismofaminoacids tofuselalcoholsandacids.AdaptedfromHazelwoodetal.,(2008)


14

2.4 EthanolToxicity

2.4.1 Physiologicalimpactofethanoltoxicity

As fermentationprogresses, the sugar concentrationdecreasesand theethanol concentration

increases (Bisson, 1999). Ethanol primarily targets the plasma membrane, by increasing its

permeability.Thiscausesadeclineinthetransportofnitrogenandsugarintothecell(Bisson,

1991; Hallsworth, 1998; Bisson, 1999). Changes in the plasma membrane permeability also

causeanincreasedinfluxofprotonsintothecellwhichdissipatestheprotonmotiveforceused

to transportaminoacids into thecell. Inorder toregulatecytoplasmicpHthecellpumpsout

protonsviatheATPase,andceasesthesimultaneousimportofaminoacidsandprotons(Bisson,

1991) (Figure 4). Similarly, the V‐ATPase pumps protons into the vacuole (Ma& Liu, 2010).

Ethanolalsodenaturesproteins,includingthoseinvolvedintransport(Hallsworth,1998).This

lossinthefunctionalityoftransportsystemsultimatelycontributestoalossofcellviabilityand

areductioninyeastgrowth(Bisson,1999;Gibsonetal.,2007;Stanleyetal.,2010).

2.4.2 Acquisitionofethanoltolerance

Thegenomeexpressionprofile inresponsetoshort termethanolstress(7%v/v)displayeda

reduction in the expression of those genes involved in protein synthesis, cellular growth and

RNAmetabolismaccordingtoastudybyAlexandreetal.,(2001).Conversely,anincreaseinthe

expressionof genes associatedwith energymetabolism,protein transport, ionichomeostasis,

and stress response was observed. The genes within the functional stress response group

includedthoserelatedtotheexpressionofnumerousHSP’sandtrehalosesynthase.Thisalludes

tothenegativeimpactethanolhasonproteinstructure(Alexandreetal.,2001).Trehalosealso

reducesmembranepermeability,improvingthecell’sethanoltolerancebyreducingtheeffluxof

nutrients from the cell (Mansure etal., 1994; Sharma, 1997). A second transcriptomic study,

evaluating the expression patterns after an one hour long exposure to 5% (v/v) ethanol

reported an increase in the expression of genes associated with transport, cell surface

interactions and lipid metabolism in addition to genes involved in energy metabolism, ionic

homeostasis,andstressresponseasstatedabove(Chandleretal.,2004).

Deletion studies have identified numerous genes involved in cellwall andmembrane

synthesisconferringethanoltolerance.Interestingly,thesegenesaregenerallydownregulated

uponexposure to ethanol (Chandleretal., 2004;Ma&Liu,2010).Ethanol tolerancehas also

been correlated with the fatty acid and sterol composition of the membrane


15

(Alexandreetal.,1994),andstrainsadapttoethanolbyincreasingthesterolandunsaturated

fatty acid contentwhich provide structural stability to the plasmamembrane. However, this

process ishinderedby theabsenceofoxygensincesynthesisof sterolsandunsaturated fatty

acidsrequiresthepresenceofmolecularoxygen(Alexandreetal.,1994).

The ethanol induced loss of functional transport systems induces the expression of high

affinity hexose transporters, which are usually only expressed under conditions of glucose

limitation.Thissuggeststhatcellsexperiencingethanolstressenterapseudo‐starvedstate,as

the cell is unable to access nutrients from the surrounding medium (Chandler et al., 2004).

Furthermore,Marksetal.,(2008)proposesthatethanolservesasasignalforthecelltoenter

stationary growth phase once it reaches the 2% (v/v) level. This may be a pre‐emptive

mechanism to ensure long term survival, as cells in the stationary phase are generallymore

stresstolerant.

ATPase

ATPase

ATP

ADP+ Pi

H+ H+

ADP+ Pi

H+

H+

H+

H+

NO Ethanol

H+

Ethanol

Passive Proton influx

Amino acid symport

H+

Amino acid symport

H+

H+

H+

H+

H+

H+

H+H+

H+H+

Amino acid

Amino acid

Amino acid

Amino acid

ATP

Passive Proton influx

H+

H+

H+

H+

H+ H+

Figure4Theimpactofethanolonthetransportofprotonsandaminoacidsintothecell.AdaptedfromBisson(1991)


16

2.4.3 Potentialroleofethanolinwinearoma

The impactofethanolonwinearomaperceptionhasbeenstudied,andat low levels, ethanol

enhances the sensory perception of aroma compounds. However, when in excess it has a

maskingeffect, and inhighamounts candirectly lead to aburning sensation (Swiegersetal.,

2005).

Ontheotherhand,onlyverylimiteddataareavailableregardingtheimpactofethanol

onthedenovoproductionofaromacompoundsbyyeast.Ethanoltoxicitycausesa lossof the

transport of sugar and nitrogen into the cell, both of which are crucial to the formation on

aromacompounds(Bisson&Karpel,2010).Thusitisnotsurprisingthatethanolislistedamong

the factors impactinghigher alcohol formation (Fleet&Heard, 1993), additionally, it is likely

that ethanol toxicity also impacts the formation of esters, acetaldehyde, organic acids and

diacetyl; as their production is also dependent on sugar and nitrogenmetabolism (Bisson &

Karpel,2010).

2.5 Lowtemperature

2.5.1 Physiologicalimpactoflowtemperature

Whitewinefermentationsareoftenconductedbetween10°Cand15°Ctoensuretheretention

of volatile aroma compounds which would be released in larger amounts at higher

temperatures.Whencooledtotemperaturesbelow20°C,thecellexperiencesatemperatureand

durationspecificcoldstressresponse(Aguileraetal.,2007;Gibsonetal.,2007).Thiscoldstress

response is not as conserved as the heat shock response (Piper, 1995; Kregel, 2002; Rangel,

2010), and is also poorly characterised for yeast (Aguilera et al., 2007). Temperature, like

ethanol,primarilyactsupontheplasmamembrane.Exposuretohightemperaturesorethanol

results in an increase in membrane permeability, whereas lower temperatures result in a

decrease inmembranepermeability (Shinitzky,1984;Gibsonetal.,2007).Consequently,cells

fermentingatlowtemperaturesarelesssusceptibletoethanolexposuresthanthosefermenting

athightemperatures.However,thereducedmembranepermeabilityalsohindersthetransport

of essentialnutrients into the cellby trans‐membraneproteins (Hazel, 1995). Several studies

haveshownthatlowtemperatureincreasesfermentationdurationduetoadeclineinmetabolic

activity and, consequently, a lowering of yeast biomass production (Llauradó et al., 2002;

Beltranetal.,2006;Pizarroetal.,2008).


17

2.5.2 Acquisitionoflowtemperaturetolerance

Yeastcellsadapttolowertemperaturesbyattemptingtomaintainaconstantlevelofmembrane

fluidity.Duringtheearlystagesoflowtemperatureadaptation,thecellstabilisesthemembrane

by increasing thedegreeof fatty acidunsaturation, and increasespermeability bydecreasing

fattyacidchainlength(Saharaetal.,2002;Torijaetal.,2003;Al‐Fageeh&Smales,2006;Taiet

al., 2007; Beltran et al., 2008; Redón et al., 2011). Additionally, low temperature causes the

formation of secondary structureswithinRNAmolecules, reducing translation efficiency. The

cell counters this by increasing the expression of genes involved in ribosomal proteins, RNA

processingandtranslation(Saharaetal.,2002;Schadeetal.,2004;Aguileraetal.,2007;Gibson

etal.,2007).

A transcriptomic study comparing the expression patterns during the course of

fermentations conducted at 13C and 25C, reported the increased expression of genes

associated with membrane permeability at 13C relative to 25C during the initial stages of

fermentation.Conversely,duringthelaterstagesoffermentationthelevelofexpressionofthe

genes associatedwithmembrane permeabilitywas greater at 25C than 13C (Beltran etal.,

2008). Despite a lack of cell division, strains fermenting at 13°Cwere better able to survive

compared to those at 25°C, where a decline in viable cells was observed compared to a

maintainedmaximalpopulationsizeat13°Cthroughoutthefermentation.Thistranscriptomic

datasuggeststhattheearlyonsetofthestressresponse,basedontheMSN2expressionlevels,

at lower temperature compared to the induction upon entering the stationary phase for

fermentationsathighertemperatures,betterpreparesthecellstosurvive(Beltranetal.,2008).

Low temperature tolerance is further characterised by trehalose and phospholipid

synthesis,theexpressionofheatshockproteins,inductionofoxidativestressresponseandcell

wallmannoproteinssynthesisduringthelaterstagesofcoldshock(Schadeetal.,2004).


18

2.6 Conclusions

Wineyeaststrainsexperienceadynamicenvironmentduringgrapemustfermentation.Thecell

isrequiredtocounteractthedeleteriouseffectsoftemperaturechange, lowpH,hyperosmotic

stress,ethanoltoxicity,limitedoxygenandnitrogenavailability,andthepresenceofcompeting

or sometimes antagonistic micro‐organisms throughout fermentation (Alexandre &

Charpentier,1998;Bisson,1999;Gibsonetal.,2007).Itsabilitytorespondtothesestresses,via

so‐called “general” and specific means, will determine whether it survives and completes

alcoholicfermentation.

Overall,wineyeaststrainsattain“general”stresstoleranceviatheaccumulationofHSP

andtrehalosetopreventandrepairproteindenaturation.Inadditiontothis“general”response,

hyperosmoticstressalsoinducestheexpressionoftheenzymesofglycerolproductionaspart

of its specific stress response. In the case of ethanol toxicity and low temperature, as both

stresses impact the plasma membrane, it is not surprising that the cell regulates plasma

membranepermeabilitytooffsettheharmfuleffectsofstress.

Future work should be directed towards establishing the impact of major fermentation

stressesontheproductionofvolatilearomacompounds.

2.7 Literaturecited

Aguilera, J.,RandezGil,F.&Prieto, J.A.,2007.Coldresponse inSaccharomycescerevisiae:newfunctionsforoldmechanisms.FEMSMicrobiol.Rev.31,327‐341.

Al‐Fageeh,M.B.&Smales,C.M., 2006.Control and regulationof the cellular responses to coldshock:theresponsesinyeastandmammaliansystems.Biochem.J.397,247‐259.

Alexandre,H.,Ansanay‐Galeote,V.,Dequin,S.&Blondin,B.,2001.Globalgeneexpressionduringshort‐termethanolstressinSaccharomycescerevisiae.FEBSLetters.498,98‐103.


Alexandre, H., Rousseaux, I. & Charpentier, C., 1994. Relationship between ethanol tolerance,lipid composition and plasma membrane fluidity in Saccharomyces cerevisiae and Kloeckeraapiculata.FEMSMicrobiol.Lett.124,17‐22.

Attfield, P., 1998. Physiological and molecular aspects of hyperosmotic stress tolerance inyeasts.RecentRes.Dev.Microbiol.2,427‐441.

Attfield, P.V., 1997. Stress tolerance: the key to effective strains of industrial baker's yeast.NatureBiotechnology.15,1351‐1357.

Bauer,F.&Pretorius,I.,2000.Yeaststressresponseandfermentationefficiency:howtosurvivethemakingofwine‐areview.S.Afr.J.Enol.Vitic.21,27‐51.


19

Beltran, G., Novo, M., Guillamón, J.M., Mas, A. & Rozès, N., 2008. Effect of fermentationtemperatureandculturemediaontheyeastlipidcompositionandwinevolatilecompounds.Int.J.FoodMicrobiol.121,169‐177.

Beltran,G.,Novo,M.,Leberre,V.,Sokol,S.,Labourdette,D.,Guillamon,J.M.,Mas,A.,François,J.&Rozes, N., 2006. Integration of transcriptomic and metabolic analyses for understanding theglobalresponsesoflowtemperaturewinemakingfermentations.FEMSYeastRes.6,1167‐1183.

Berry,D.B.&Gasch,A.P.,2008.Stress‐activatedgenomicexpressionchangesserveapreparativeroleforimpendingstressinyeast.Mol.Biol.Cell.19,4580‐4587.

Bisson,L.F.,1991.Influenceofnitrogenonyeastandfermentationofgrapes.In:Ranz,J.M.(ed).Proceedingsoftheinternationalsymposiumonnitrogeningrapesandwine.Davis,CA,pp.78‐89



Carrasco, P., Querol, A. & del Olmo,M., 2001. Analysis of the stress resistance of commercialwineyeaststrains.Arch.Microbiol.175,450‐457.

Causton, H.C., Ren, B., Koh, S.S., Harbison, C.T., Kanin, E., Jennings, E.G., Lee, T.I., True, H.L.,Lander, E.S. & Young, R.A., 2001. Remodeling of yeast genome expression in response toenvironmentalchanges.Mol.Biol.Cell.12,323‐337.

Chandler,M.,Stanley,G.,Rogers,P.&Chambers,P.,2004.AgenomicapproachtodefiningtheethanolstressresponseintheyeastSaccharomycescerevisiae.AnnMicrobiol.54,427‐454.

Erasmus,D.J.,Merwe,G.K.&Vuuren,H.J.J.,2003.Genomewideexpressionanalyses:MetabolicadaptationofSaccharomycescerevisiaetohighsugarstress.FEMSYeastRes.3,375‐399.

Estruch, F., 2000. Stress controlled transcription factors, stress induced genes and stresstoleranceinbuddingyeast.FEMSMicrobiol.Rev.24,469‐486.

Fleet,G.&Heard,G.,1993.Yeast‐growthduringfermentation.In:Fleet,G.(ed).WineMicrobiol.Biotechnol.HarwoodAcademicPublishers,Switzerland.pp.27‐54.

Fleet,G.H.,2008.Wineyeastsforthefuture.FEMSYeastRes.8,979‐995.

Francis,I.&Newton,J.,2005.Determiningwinearomafromcompositionaldata.Aust.J.GrapeWineRes.11,114‐126.

Gasch, A.P., Spellman, P.T., Kao, C.M., Carmel‐Harel, O., Eisen, M.B., Storz, G., Botstein, D. &Brown,P.O.,2000.Genomicexpressionprogramsintheresponseofyeastcellstoenvironmentalchanges.Mol.Biol.Cell.11,4241‐4257.

Gasch,A.P.&Werner‐Washburne,M.,2002.Thegenomicsofyeastresponsestoenvironmentalstressandstarvation.Funct.Integr.Genomics.2,181‐192.

Giaever, G., Chu, A.M., Ni, L., Connelly, C., Riles, L., Véronneau, S., Dow, S., Lucau‐Danila, A.,Anderson, K.&André, B., 2002. Functional profiling of theSaccharomyces cerevisiae genome.Nat.418,387‐391.


20


Hallsworth,J.E.,1998.Ethanol‐inducedwaterstressinyeast.J.Ferment.Bioeng.85,125‐137.

Hazel,J.R.,1995.Thermaladaptationinbiologicalmembranes:ishomeoviscousadaptationtheexplanation?Annu.Rev.Physiol.57,19‐42.

Hazelwood, L.A., Daran, J.M., vanMaris, A.J.A., Pronk, J.T. &Dickinson, J.R., 2008. The Ehrlichpathway for fusel alcohol production: a century of research on Saccharomyces cerevisiaemetabolism.Appl.Environ.Microbiol.74,2259‐2266.

Hohmann, S., 1997. Shaping up: The response of yeast to osmotic stress. In: Hohmann, S. &Mager,W.H.(ed).YeastStressResponses.Springer.pp.101‐145.

Hohmann,S.,Krantz,M.&Nordlander,B.,2007.Yeastosmoregulation.MethodsEnzymol.428,29‐45.

Hounsa, C.G., Brandt, E.V., Thevelein, J., Hohmann, S. & Prior, B.A., 1998. Role of trehalose insurvivalofSaccharomycescerevisiaeunderosmoticstress.Microbiol.144,671‐680.

Ivorra, C., Pérez Ortín, J.E. & del Olmo, M., 1999. An inverse correlation between stressresistance and stuck fermentations inwineyeasts.Amolecular study.Biotechnol.Bioeng.64,698‐708.

Jain, V.K., Divol, B., Prior, B. & Bauer, F.F., 2011. Effect of alternative NAD+‐ regeneratingpathwayson the formationofprimartyandsecondaryaromacompounds inaSaccharomycescerevisiaeglycerol‐defectivemutant.Appl.Environ.Biotechnol.

Kregel,K.C.,2002.Invitedreview:Heatshockproteins:Modifyingfactorsinphysiologicalstressresponsesandacquiredthermotolerance.J.Appl.Physiol.92,2177‐2186.

Lafon‐Lafourcade, S., 1983.Wine and brandy. In: Rehm, H.J. & Reed, G. (ed). Food and FeedProductionwithMicroorganisms.Biotechnology.VerlagChemie.pp.81‐163.


Llauradó, J., Rozés, N., Bobet, R., Mas, A. & Constantí, M., 2002. Low temperature alcoholicfermentationsinhighsugarconcentrationgrapemusts.J.FoodSci.67,268‐273.

Ma, M. & Liu, Z.L., 2010.Mechanisms of ethanol tolerance in Saccharomyces cerevisiae. Appl.Environ.Biotechnol.87,829‐845.

Mager,W.H.&Siderius,M.,2002.Novelinsightsintotheosmoticstressresponseofyeast.FEMSYeastRes.2,251‐257.


Mansure,J.J.C.,Panek,A.D.,Crowe,L.M.&Crowe,J.H.,1994.Trehaloseinhibitsethanoleffectsonintactyeastcellsandliposomes.Biochim.Biophys.Acta.1191,309‐316.


21

Marks,V.D.,HoSui,S.J.,Erasmus,D.,VanDerMerwe,G.K.,Brumm,J.,Wasserman,W.W.,Bryan,J.& Van Vuuren, H.J.J., 2008. Dynamics of the yeast transcriptome during wine fermentationrevealsanovelfermentationstressresponse.FEMSYeastRes.8,35‐52.

Martinez‐Pastor,M.,Marchler,G., Schüller,C.,Marchler‐Bauer,A.,Ruis,H.&Estruch,F.,1996.The Saccharomyces cerevisiae zinc finger proteins Msn2p and Msn4p are required fortranscriptionalinductionthroughthestressresponseelement(STRE).EMBOJ.15,2227‐2235.

Morimoto,R.I.,1998.Regulationoftheheatshocktranscriptionalresponse:crosstalkbetweena family of heat shock factors,molecular chaperones, and negative regulators. Gene Dev. 12,3788‐3796.

Piper, P.W., 1995. The heat shock and ethanol stress responses of yeast exhibit extensivesimilarityandfunctionaloverlap.FEMSMicrobiol.Lett.134,121‐127.

Pizarro,F.J., Jewett,M.C.,Nielsen, J.&Agosin,E.,2008.Growthtemperatureexertsdifferentialphysiological and transcriptional responses in laboratory andwine strains of Saccharomycescerevisiae.Appl.Environ.Microbiol.74,6358‐6368.

Pretorius,I.,2000.Tailoringwineyeastforthenewmillennium:novelapproachedtotheartofwinemaking.Yeast.16,675‐729.

Rangel, D.E.N., 2010. Stress induced cross‐protection against environmental challenges onprokaryoticandeukaryoticmicrobes.WorldJ.Microbiol.Biotechnol.1‐16.


Redón,M.,Guillamón,J.M.,Mas,A.&Rozès,N.,2011.Effectofgrowthtemperatureonyeastlipidcompositionandalcoholicfermentationatlowtemperature.Eur.F.Res.Technol.1‐11.

Rep, M., Krantz, M., Thevelein, J.M. & Hohmann, S., 2000. The transcriptional response ofSaccharomycescerevisiaetoosmoticshock.J.Biol.Chem.275,8290.

Ruis,H.&Schüller,C.,1995.Stresssignalinginyeast.BioEssays.17,959‐965.

Sahara,T.,Goda,T.&Ohgiya, S., 2002.Comprehensiveexpressionanalysisof time‐dependentgeneticresponsesinyeastcellstolowtemperature.J.Biol.Chem.277,50015.

Schade, B., Jansen, G., Whiteway, M., Entian, K.D. & Thomas, D.Y., 2004. Cold adaptation inbuddingyeast.Mol.Biol.Cell.15,5492‐5502.

Sharma, S.C., 1997. A possible role of trehalose in osmotolerance and ethanol tolerance inSaccharomycescerevisiae.FEMSMicrobiol.Lett.152,11‐15.

Shinitzky,M.,1984.Membranefluidityandcellularfunctions.Physiol.Membr.Fluidity.1,1‐51.

Singer,M.A. & Lindquist, S., 1998. Thermotolerance in Saccharomyces cerevisiae: the Yin andYangoftrehalose.TrendsBiotechnol.16,460‐468.

Stanley,D.,Bandara,A.,Fraser,S.,Chambers,P.&Stanley,G.,2010.TheethanolstressresponseandethanoltoleranceofSaccharomycescerevisiae.J.Appl.Microbiol.109,13‐24.


22

Styger, G., Jacobson, D. & Bauer, F.F., 2011. Identifying genes that impact on aroma profilesproduced by Saccharomyces cerevisiae and the production of higher alcohols. Appl. Environ.Biotechnol.91,713‐730.


Swiegers,J.H.&Pretorius,I.S.,2005.Yeastmodulationofwineflavor.AdvancesAppl.Microbiol.57,131‐175.

Tai, S.L., Daran‐Lapujade, P., Walsh, M.C., Pronk, J.T. & Daran, J.M., 2007. Acclimation ofSaccharomycescerevisiae to low temperature: a chemostat‐based transcriptomeanalysis.Mol.Biol.Cell.18,5100‐5112.

Tamás,M.J.&Hohmann,S.,2003.TheosmoticstressresponseofSaccharomycescerevisiae.In:Hohmann,S.&Mager,P.W.H.(ed).YeastStressResponses.Springer‐VerlagBerlin.pp.121‐200.

Torija,M.J.,Beltran,G.,Novo,M.,Poblet,M.,Guillamón,J.M.,Mas,A.&Rozes,N.,2003.Effectsoffermentation temperature and Saccharomyces species on the cell fatty acid composition andpresenceofvolatilecompoundsinwine.Int.J.FoodMicrobiol.85,127‐136.

Wood, J.M., 1999. Osmosensing by bacteria: signals andmembrane‐based sensors.Microbiol.Molec.Bio.Rev.63,230‐262.

Zuzuarregui,A.&delOlmo,M.,2004.Analysesofstressresistanceunderlaboratoryconditionsconstituteasuitablecriterionforwineyeastselection.AntonievanLeeuwenhoek.85,271‐280.


23

Chapter3

Researchresults

Theimpactandinteractionbetweeninitialnitrogen,initialsugar,andtemperatureonthefermentation

performanceofcommercialwineyeast


24

Chapter3

Theimpactandinteractionbetweeninitialnitrogen,initialsugar,andtemperatureonthefermentationperformanceof

commercialwineyeast

3.1 Introduction

The metabolic process central to winemaking is the conversion of sugar to ethanol and CO2. This

conversionismediatedbyyeast,primarilySaccharomycescerevisiae.Theuseofactivedryyeast(ADY)

iscommonpractice in thewine industry,as itaidsonsetof fermentationaswellas fermentation to

dryness,wheretheresidualsugarcontentisreducedtolessthan4g/L(Pretoriusetal.,1999).During

fermentation,yeastculturesareexposedtonumerousstresses,bothsimultaneouslyandinsuccession

(Bisson,1999).Thesestressesincludenutrientdeficiencies(minerals,vitamins,nitrogen,andoxygen),

lowpH,ethanoltoxicity,temperatureextremes,andahighosmoticpressure(Kunkee,1991;Attfield,

1997;Alexandre&Charpentier,1998;Bauer&Pretorius,2000;Malherbeetal.,2007).Thesearealso

thefactorsassociatedwithproblematicfermentations,whichhavebeenstudiedextensively,duetothe

economiclossesandlogisticalproblemstheycause(Malherbeetal.,2007).

Nitrogen deficiency is reportedly themost common cause of problematic fermentations. As

nitrogenplaysanintegralroleinbiomassproduction,cellmaintenance,andsugarcatabolism,italso

influences the fermentation rate (Bisson, 1991; 1999). Yeast assimilable nitrogen (YAN) consists of

ammonia, free amino acids (excluding proline and hydroxyproline), and low molecular weight

peptides. S. cerevisiae is unable to utilize larger peptides due to its poor extracellular proteolytic

activity, and proline because of the anaerobic state of fermentation (Bell & Henschke, 2005). It is

generallyagreedthat140mg/LYANisthethresholdlevelbelowwhichtheriskofstuckorsluggish

fermentationsincreases(Agenbach,1977;Belyetal.,1990a).However,thislevelwasestablishedfor

clarified must with moderate sugar levels, and thus should only be considered as a guide (Bell &

Henschke, 2005), as a higher sugar concentration requires more nitrogen to ferment to dryness

(Bisson&Butzke,2000).

Low YAN content is the most common cause of problematic fermentations, and is most

commonly overcomewith diammoniumphosphate (DAP) supplementation. It is routinely added to

thegrapemustbeforetheonsetoffermentation,frequentlywithoutfirstdeterminingtheYANcontent

of the grape must. When nitrogen is deficient, DAP addition reduces the risk of problematic

fermentations (Bisson, 1999). However, when nitrogen is in excess it may result in microbial

instability,and,insomecases,adeclineinthefermentationperformanceofayeaststrain(Taillandier

etal.,2007).

Commercial starter cultures differ significantly in their inherent nitrogen requirements


25 (Jiraneketal.,1991;Manginotetal.,1998).Thisstrainspecificresponsetonitrogenemphasizesthe

importance of evaluating all commercial yeast strains to ascertain their respective nitrogen

requirements,andinsodoingpreventexcessiveorinsufficientnitrogensupplementation(Jiraneket

al.,1991;Manginotetal.,1998;Taillandieretal.,2007).

Ahighinitialsugarconcentrationinitiatesahyperosmoticstressresponseimmediatelyupon

inoculation. To limit the loss ofwater, glycerol channels close, and glycerol production is activated.

Glycerolservesasanosmoprotectant,byincreasingtheinternalsoluteconcentration,andinsodoing

limitstheeffluxofwater.Theaccumulationofglycerolcontinuesuntiltheinfluxofwaterrestorescell

size to a critical level and yeast growth is resumed (Hohmann, 1997; Mager & Siderius, 2002).

Consequently,thehighertheinitialsugarconcentrationthelongerthelagphasebeforefermentation

commences,duetothecessationofgrowthwhilethecelladaptstotheprevailingosmoticconditions.

This may in some cases even result in stuck or sluggish fermentations (Lafon‐Lafourcade, 1983;

Llauradó et al., 2002). As the fermentation progresses, the sugar concentration decreases and the

ethanol concentration increases. Ethanol primarily targets the plasma membrane, increasing its

permeability.Thisultimatelyresultsinareductionincellviabilityandgrowth,causedbyadeclinein

thetransportofnitrogenandsugarintothecell(Bisson,1991;Hallsworth,1998;Stanleyetal.,2010).

Thecellsrespondtoethanoltoxicitybysynthesizingtrehalose,unsaturatedfattyacidsandheatshock

proteinstorestoremembranepermeability(Bisson,1999;Gibsonetal.,2007;Stanleyetal.,2010).

Temperatureandethanolbothactupontheplasmamembrane.Exposuretohightemperatures

ortoethanolresultsinanincreaseinmembranepermeability,whereaslowertemperaturesresultina

decrease in membrane permeability (Gibson et al., 2007). Consequently, cells fermenting at high

temperaturesaremoresusceptible toethanolexposure than those fermentingat low temperatures.

White wine fermentations are generally conducted between 10°C and 15°C compared to 25°C or

higher for red wines. Fermentation temperature impacts the retention of aroma compounds;

additionally transcriptomicstudieshave foundtemperaturedependantdifferences intheexpression

ofaromarelatedgenes(Torijaetal.,2003;Beltranetal.,2006,Molinaetal.,2007).

Several studies have shown that low temperature increases fermentation duration due to a

declineinmetabolicactivityand,consequently,aloweringofyeastbiomassproduction(Fleet&Heard,

1993;Llauradóetal.,2005;Beltranetal.,2006;Pizarroetal.,2008).

Yeast strains differ in their ability to sense and effectively respond to all of the

abovementionedstresses(Ivorraetal.,1999;Carrascoetal.,2001).Thisabilitytosenseandrespond

to stress has been linked to fermentation performance (Zuzuarregui & del Olmo, 2004), as stress

resistancewouldcontributetowhetherstrainsareabletosurviveandultimatelyfermentgrapemust.

Strainsalsovaryintheirnitrogenrequirements(Agenbach,1977;Bezenger&Navarro,1988;Belyet

al.,1990b;Belyetal.,1991;Jiraneketal.,1991;Manginotetal.,1998),andtheirabilitytocatabolise

sugars(Manginotetal.,1998).Thisemphasizestheimportanceoftheselectionofanappropriateyeast


26 starterculturecapableoffermentingamust,ofknownparameters(YAN,totalsugarsetc.),adequate

nitrogensupplementation,oracombinationofbothstrategies.

Theaimofthisstudywastoinvestigatetheimpactoftheinitialnitrogencontent(50,100,250

or400mg/L),hyperosmoticpressure(200or240g/Lsugar),andtemperature(15°Cor20°C)onthe

fermentation performance of 17 commercial active dry yeast cultures using a multifactorial

experimental design. Past studies have assessed the response of yeast strains, including somewine

yeaststrains,toindividualstresses.Toourknowledgethisisthefirststudytoevaluatetheimpactof

different stresses when applied at different levels and in different combinations. Considering the

complex and integrated nature of molecular stress response pathways, a combined application of

stressesmay indeedresult inresponses thatarequalitativelyandquantitativelyverydifferent from

thosedescribedfor individualstresses. Inthisstudy, the fermentationperformancesofstrainswere

characterised on the basis of fermentation kinetics data (weight loss due to CO2 evolution), the

residual sugar levels and the dry weight produced as determined at the end of fermentation. The

fermentationkineticsdatawasused togenerate theEC50,hill‐slope, and topvalues for eachof the

fermentations.TheEC50valuerepresentsthetimerequiredforthefermentationtoreachthehalf‐way

mark,illustratinghowrapidlyastrainisabletoadapttoitsenvironmentalconditions.Thehill‐slope

(maximum fermentation rate) equals thegradientof the curve, and reflects theextent towhich the

grapemustand fermentationconditions impactyeastgrowthandgrapemust fermentation.Thetop

valueequalsthetotalweightloss,whichprovidesarelativeindicationoffermentationcompleteness

(dependant on the initial sugar content). This approach will identify strains that are capable of

fermenting a grapemustwith specific nitrogen and sugar levels, providingwinemakers a toolwith

which to select a yeast strainbest adapted to ferment a specific grape juice and ensuring complete

fermentation.


27 3.2 Materialsandmethods

3.2.1 Syntheticgrapemediumandfermentationtreatments

The synthetic grapemust used in this study is described inTables 1 and2. ThepHof themedium

containing the carbon sources, acids, and salts was set to 3.3 with potassium hydroxide, before

autoclaving.Thevitamin,mineral, lipidandaminoacidstockswerefiltersterilizedandaddedtothe

autoclavedmedium.

Thesmallscalefermentationswereperformedintriplicateat15°Cor20°C.Themustdiffered

in the initial sugar content and contained equimolar amountsof glucose and fructose amounting to

either 200 g/L or 240 g/L. Fermentations also differed in the initial nitrogen content, which was

proportionallydecreasedorincreasedtoreachdesiredlevelof50mg/L,100mg/L,250mg/Lor400

mg/Lnitrogen.The16fermentationtreatmentsaresummarisedinTable3.Alltheyeaststrainsused

(Table 4)were rehydrated according tomanufacturers’ instructions and inoculated at 20 g/hL. All

yeast strains were supplied by Anchor yeast, except Lalvin EC1118 (Lallemand) and AWRI796

(Maurivin).

Thefermentationswereweighedregularlytomonitorfermentationprogress,asCO2evolution.

After 21 days, the fermented synthetic wine samples were scanned using the Winescan FT120

instrument (FOSS Analytical A/S software version 2.2.1) equipped with a purpose‐built Michelson

interferometer(FOSSAnalyticalA/S,Hillerød,Denmark)togenerateaFouriertransformmidinfrared

(FT‐MIR) spectra. Quantified chemical data for residual glucose and fructose levelswere predicted

frominfraredspectrabycommercialcalibrationsorin‐houseadjustmentsusingtheWinescanFT120

2001version2.2.1software.

3.2.2 Dryweight

Thebiomasswasdetermined(intriplicate)ascelldryweight.A4mlsamplewasspundowninapre‐

weighedmicrocentrifugetube;theresultingpelletwasdriedinanovenat30°Cforapproximatelytwo

weeks. In order to ensure that allmoisturewas removed themicrocentrifuge tubewas spun in the

speedyvacsetathighheatfor5minutesandweighed.


28 Table1:ThesyntheticgrapemustusedaspreviouslydescribedbyHenschkeandJiranek(1993)forthecarbon,

acids,salts,traceelements,vitamins,andlipidsources.

3.2.3 Statisticalanalysis

The fermentation kinetics (weight loss due to CO2 evolution) datawas used to fit three‐parameter

logisticdoseresponsecurvesacrossallcombinationsoftreatments.ThethreeparameterswereEC50,

hill‐slope,andthetopvalue.EC50valuerepresentsthetimerequiredforthefermentationtolosehalf

ofthetotalweightloss(fermentationmid‐point).Thehill‐slope(maximumfermentationrate)equals

the gradient of the curve and the top value equals the totalweight loss. The effect of the different

perlitre

CarbonSourcesGlucose 100or120g

Fructose 100or120g

Acids

KHTartrate 2.5g

L‐Malicacid 3g

Citricacid 0.2g

Salts

K2HPO4 1.14g

MgSO4.7H20 1.23g

CaCl2.2H2O 0.44g

Minerals

MnCl2.4H2O 200µg

ZnCl2 135µg

FeCl2 30µg

CuCl2 15µg

H3BO3 5µg

Co(NO3)2.6H2O 30µg

NaMoO4.2H2O 25µg

KIO3 10µg

Vitamins

Myo‐Inositol 100mg

Pyridoxine.HCl 2mg

Nicotinicacid 2mg

CaPantothenate 1mg

Thiamin.HCl 0.5mg

PABA.K 0.2mg

Riboflavin 0.2mg

Biotin 0.125mg

FolicAcid 0.2mg

LipidsErgosterol 10mg

Tween80 0.5mL


29 treatments on the abovementionedparameters aswell as residual sugars (sumof residual glucose

andfructose)anddryweightwereassessedusingfactorialanalysisofvariance(ANOVA).Inallcasesa

significancelevelof5%wasused.

Allresultsarepresentedintwoways,firstlookingatthegeneralimpactofthetreatmentsonthe

parameterinquestion,andsecondlycomparingtheperformanceofindividualstrains.Duetothesize

andcomplexityofthedataset,onlythelowerorderinteractionsarereportedon,insomecasesthisis

despite a significant higher order interaction. Additionally,when a variable (temperature, nitrogen,

sugarorstrain)appearstohavebeen“omitted”,thedataforthatvariablehasbeenpooled.Thismay

hideothersignificantpatterns.

Table2:ThenitrogensupplementationsusedaspreviouslydescribedbyBelyetal.,(1990b),wasproportionallyincreasedordecreasedtoobtainafinalconcentrationof50,100,250and400mg/L.

NitrogenSources: 300mg/L

AminoAcids

Tyrosine 18.326mg/L

Tryptophane 179.333mg/L

Isoleucine 32.725mg/L

asparticacid 44.506mg/L

glutamicacid 120.428mg/L

Arginine 374.374mg/L

Leucine 48.433mg/L

Threonine 75.922mg/L

Glycine 18.326mg/L

Glutamine 505.274mg/L

Alanine 145.299mg/L

Valine 44.506mg/L

Methionine 31.416mg/L

phenylalanine 37.961mg/L

Serine 78.54mg/L

Histidine 32.725mg/L

Lysine 17.017mg/L

Cysteine 13.09mg/L

Proline 612.612mg/L

AmmoniumChloride NH4Cl 0.46g/L


30 Table3:Fermentationswereconducted intriplicateat15°Cor20°C.All treatmentscontainedsugars(200or240g/Ltotal),nitrogen(50,100,250or400mg/Lnitrogen),salts,mineralsandfattyacids.

Table4:Strainsusedinthisstudy(compiledfromcommercialspecificationsheets).

1ExoticSPHwillbereferredtobyitsIWBTnamePR7fromthispointforward

2Bradburyetal.(2006)

Treatmentdescription

Temperature C 15 20 15 20 15 20 15 20 15 20 15 20 15 20 15 20

Sugar g/L 200 200 200 200 200 200 200 200 240 240 240 240 240 240 240 240

Nitrogen mg/L 50 50 100 100 250 250 400 400 50 50 100 100 250 250 400 400

Commercialname

Strain Recommendedwinestyles

Recommendedtemperature

IWBTPR71(Exotics

SPH)

Saccharomyces cerevisiae,S.paradoxushybrid

Whitewine 16‐20C

VIN2000 S.cerevisiaehybrid Whitewine 13‐16C

VIN7 S.kudriavzeviiandS.cerevisiaehybrid2

Whitewine 13‐16C

VIN13 S.cerevisiaehybrid White/Rosewine 12‐16C

WE14 S.cerevisiae Natural sweet whitewine

16‐20C

WE372 S.cerevisiae Red wine/ semi‐sweetwhite

18‐28C

AlchemyI Saccharomycesspp.blend Whitewines 13‐16C

AlchemyII Saccharomycesspp.blend Whitewines 13‐16C

228 S.cerevisiae Brandy base wineproduction

15‐20C

AWRI796 S.cerevisiae Red/whitewine 15‐18,

20‐30C

EC1118 S.cerevisiaebayanus Sparkling, fruitwine,andciders

15‐25C

N96 S.cerevisiaebayanus Sparkling and icewines

12‐28C

NT45 S.cerevisiae Redwines 14‐28C

NT50 S.cerevisiaehybrid Redwines 14‐28C


NT116 S.cerevisiaehybrid Whitewines 12‐16C



31 3.3 Resultsanddiscussion

3.3.1 Influenceoftreatmentsonfermentationkinetics

3.3.1.1 Impactoftreatmentsonfermentationonset

The number of days required for the fermentation to reach its halfwaymark is represented as the

EC50 value. This value was used as an estimate of the duration of the lag phase or the onset of

fermentation.Figure1illustratestheinfluenceoftheinitialnitrogenandsugarconcentrations,aswell

astemperatureontheonsetof fermentationwhencombiningtheEC50valuesofall thestrains(see

Table4)withina specific treatment (seeTable3). Similarly, in subsequent figures,whenavariable

(initial sugar,nitrogen, temperatureoryeast strains) isnotdescribed, that is an indication that the

dataforthatvariablehasbeenpooled.

Figure1:Influenceofthenitrogen(50,100,250or400mg/L),sugar(200or240g/L),andtemperature(15°Cor20°C) treatmentson the fermentationonset(EC50).Thedata forall strainswithinaspecific treatmentarepooled. Error bars indicate 95% confidence intervals, and the letters denote a significant difference on a 5%significance(p<0.05)level.

Temperature T15 Temperature T20Sugar: S200

Nitrogen:N50

N100N250

N4003

4

5

6

7

8

9

10

EC

50 (

days

)

Sugar: S240

Nitrogen:N50

N100N250

N400

c

d

b

h

d

e

f

c

hi

g

f

c

a

i

h

b


32

Low temperature (15C), low nitrogen levels (50 and 100 mg/L) and high sugar content

(240 g/L) are environmental stresses commonly linked to reduced yeast growth. This reduction in

yeast growth explains the extended fermentation lag phases observed in Figure 1 for the

fermentations containing a single or combinations of these stresses. At both sugar levels and

temperatures, the increase in the nitrogen concentration caused a decrease in the lag phase

(Belyetal.,1991;Beltranetal.,2005).Whennitrogenwasraisedto400mg/L,arelativeincreasein

the lag time was observed when compared to the 250mg/L lag phases. This suggests that excess

nitrogenhasadetrimentalimpactontheonsetoffermentation.

3.3.1.2 Impactof treatmentson the fermentationonsetof individualyeaststrains

The variation in the initial nitrogen levels resulted in different lag phases for the strains evaluated

(Figure 2). Nitrogen is essential for biomass production, cell maintenance, and sugar catabolism

(Bisson,1991;1999),itisnotsurprisingthenthattheonsetoffermentationwashamperedbythelow

initial nitrogen content (50 mg/L). Within the 50 mg/L nitrogen treatment, WE372, NT1116, and

AlchemyIwereamongthestrainswiththequickestfermentationonset,andN96,VIN7andNT45were

among the strains with the slowest onset. For the other nitrogen levels, 100, 250 and 400 mg/L,

VIN2000,NT116,andWE372wereamongthestrainswithashortlagphase.Conversely;VIN7,NT45,

andWE14wereamongthestrainswiththelongerlagphase.

Asthenitrogencontentincreasedacorrespondingdecreaseinlagphasewasobserved(Belyet

al., 1991;Beltranetal., 2005).However,whennitrogenwas raised to400mg/L, the lagphasewas

either statistically similar to that of the 250mg/L treatments (228, AlchemyII, NT112, NT45, PR7,

VIN2000, and VIN13) or increased significantly, as was the case for the majority of the strains

(EC1118,AWRI796,N96,NT116,WE372,NT202NT116,NT50,WE14,VIN7andAlchemyI).For the

strainsNT116,NT50,VIN7andWE14thelagphaseincreasedtodurationscomparabletothatofthe

100mg/Ltreatments.VIN7wasconsistentlyamongthestrainswiththelongerlagphase.Thisisnot

surprising,asitsinabilitytorapidlyrespondtohyperosmoticstresshasbeenreportedinotherstudies

(Erasmusetal.,2004;Erasmus&vanVuuren,2009).

Fermentationtemperatureinfluencestheonsetonfermentation(Figure3)byaffectingyeast

growth(Fleet&Heard,1993).Fermentationsconductedat20°Chadashorterlagphasecomparedto

those at 15°C. VIN2000, VIN13 andWE372 are among the strainswith the shortest lag phase and

VIN7, NT50 and WE14 were among those with a long lag phase, irrespective of the fermentation

temperature.


33

Fermentations containing the lower sugar level (200 g/L)were initiatedmore readily than

thosecontaining240g/L(Datanotshown).Thisisindicativeofexposuretoincreasedosmoticstress,

where yeast strains require additional time to adapt to the environmental conditions (Mager &

Siderius,2002).Forbothsugarconcentrations,VIN2000,VIN13andWE372wereamongthestrains

withtheshortestlagphase,andVIN7,NT50andWE14wereamongthosewithalonglagphase.


34

Figure2:Thefermentationonset(EC50value)of17commercialculturesfermentingsyntheticgrapemustcontaining50,100,250,or400mg/nitrogen.Thedataforsugarandtemperaturewascombined,providingonevalueperstrainforeachnitrogenlevel.Errorbarsindicate95%confidenceintervalsforthemeans.

Nitrogen N50 Nitrogen N100 Nitrogen N250 Nitrogen N400

EC1118AWRI 796

PR7228

N96NT50

NT112NT116

WE14WE372

VIN7VIN13

VIN2000NT45

NT202AlchemyI

AlchemyII3

4

5

6

7

8

9

10

EC

50 (

days

)


35

Figure3: The fermentationonset (EC50) of 17 commercial cultures fermenting synthetic grapemust at 15°Cor 20°C.Thenitrogenand sugardata has beenpooled,resultinginasinglevalueforeachstrainattheevaluatedtemperatures.Errorbarsindicate95%confidenceintervalsforthemeans.

Temperature T15 Temperature T20

EC1118AWRI 796

PR7228

N96NT50

NT112NT116

WE14WE372

VIN7VIN13

VIN2000NT45

NT202AlchemyI

AlchemyII4.5

5.0

5.5

6.0

6.5

7.0

7.5

8.0

8.5

9.0

EC

50 (

days

)


36

3.3.1.3 Impactoftreatmentsonfermentationrate

The influenceof nitrogen, sugar and temperatureon themaximum fermentation rate isdepicted in

Figure 4. The maximum fermentation rates (hill‐slope) for both sugar levels respond in a similar

pattern to the temperature and nitrogen treatments; however the responses do differ in their

magnitude. At the lowest nitrogen level (50 mg/L) for both sugar levels, there was no significant

difference between fermentations at 15C and 20C, suggesting that nitrogen rather than the other

stresseshadalimitingeffectonthemaximumfermentationrate.Thefermentationrateincreasedas

thenitrogenlevelincreased.However,at400mg/Ladeclineinthefermentationrateoccurred.

Figure4:Influenceofthenitrogen(50,100,250,400mg/L),sugar(200or240g/L),andtemperature(15°Cor20°C) treatments on the maximum fermentation rates (hill‐slope). The data for all strains within a specifictreatmentwerepooled.Errorbars indicate95%confidence intervals forthemeans.Lettersdenotesignificantdifferencesona5%(p<0.05)significancelevel.


Nitrogen:N50

N100N250

N4000.0

0.1

0.2

0.3

0.4

0.5

Hill

Slo

pe (

g/da

y)

Sugar: S240

Nitrogen:N50

N100N250

N400

aa

bc

e

f

i

de

g

a aab

d de

h

c

f


37

Figure 5 shows the correlation between the timing of fermentation onset and maximum

fermentation rate. The data shows that the EC50 value is inversely correlated with the maximum

fermentationrate,i.e.thatarapidonsetoffermentationcorrelateswitharapidmaximalfermentation

rate. Thus when fermentations have a long lag phase, it is generally followed by a sluggish

fermentation rate, possibly related to a fermentation limiting condition (low temperature,

hyperosmoticstress,ethanoltoxicityornutrientdepletion,amongothers).

Figure5:Therelationshipbetweenthehill‐slope(g/day)andtheEC50value(day).

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

2 3 4 5 6 7 8 9 10 11

Hillslope (g/day)

EC50 (day)


38

3.3.1.4 Impactof treatmentson the fermentationratesof individualyeaststrains

Themaximumfermentationratesofindividualstrainsdidnotdiffersignificantlyfromeachotherfor

the 50 and 100 mg/L nitrogen treatments (Figure 6). In most cases the fermentation rate was

significantlygreater for thestrains fermentingmustcontaining250mg/Lnitrogencompared to the

400mg/Lmust.TheonlyexceptionswereNT112andVIN13thatdisplayedsimilarfermentationrates

at250mg/Land400mg/Lnitrogen.Furthermore,nitrogenexcess(400mg/L)reducedthemaximal

fermentation rate of NT116, NT50, PR7, VIN7, WE14, AlchemyI and WE372 to a rate that was

statisticallysimilartothatoftheir100mg/Lnitrogenfermentations.

Inresponsetothetemperature, fermentationsat20°Cfermentedmorerapidlythanthoseat

15°C (Figure7).Theonlyexceptionwas strain228, forwhich fermentationsproceededat a similar

rate at both temperatures.At15°C themaximal fermentation rateof 228 is among the strainswith

rapidfermentationratesandat20°Citisamongthestrainswithslowrates.Thestablefermentation

ratemaybeasaresultofsuperiorlowtemperaturetolerance,relativetotheotherstrains.

Generally,themaximalfermentationrateswerenegativelyaffectedbythehigherinitialsugar

level, except in case of 228, AlchemyI andNT202 (data not shown). Irrespective of the sugar level,

VIN2000,NT45andEC1118wereamongthestrainswithfasterfermentationrates.Whentheinitial

sugar contentwas200g/L;PR7,NT112,Alchemy1andNT202were among the slower fermenters,

whereasNT50,WE372,AWRI796,VIN7wereamongtheslowfermenterswhenthesugarcontentwas

240 g/L. This difference in fermentation rate in response to the increased sugar level may be

indicativeofanincreasedsensitivitytoosmoticstressanditsresultantethanoltoxicity.


39

Figure6:Themaximalfermentationrate(hill‐slope)of17commercialculturesfermentingsyntheticgrapecontaining50,100,250,or400mg/Lnitrogen.Thesugarandtemperaturedatahasbeencombined,providingasinglevalueforeachstrainattheevaluatednitrogenlevel.Errorbarsindicate95%confidenceintervalsforthemeans.


EC1118AWRI 796

PR7228

N96NT50

NT112NT116

WE14WE372

VIN7VIN13

VIN2000NT45

NT202AlchemyI

AlchemyII0.10

0.15

0.20

0.25

0.30

0.35

0.40

0.45

Hill

Slo

pe (

g/da

y)


40

Figure7:Themaximalfermentationrate(hill‐slope)of17commercialculturesfermentingsyntheticgrapefermentingat15°Cor20°C.Thenitrogenandsugardatahasbeenpooled,providingasinglevalueforeachstrainattheevaluatedtemperatures.Errorbarsindicate95%confidenceintervalsforthemeans.


EC1118AWRI 796

PR7228

N96NT50

NT112NT116

WE14WE372

VIN7VIN13

VIN2000NT45

NT202AlchemyI

AlchemyII0.10

0.15

0.20

0.25

0.30

0.35

Hill

Slo

pe (

g/da

y)


41

3.3.1.5 Impactoftreatmentsonthetotalweightloss

The total weight loss (top value) was influenced by the initial nitrogen and sugar levels, and

temperature(Figure8AandB).Thetotalweightlosswasconsistentlygreaterforthefermentationsat

20Ccomparedtothoseconductedat15C(Figure8A).Fermentationsat20Ccontainingthelowest

nitrogen content (50mg/L) had a similar totalweight loss as those conducted at 15 C containing

100 mg/L nitrogen. This is indicative of the profound influence temperature has on fermentation

performance despite the nitrogen deficiency. Fermentations at 20°C reached similar top values for

treatmentscontaining100,250and400mg/Lnitrogen,asdidfermentationsat15°Ccontaining250

and400mg/Lnitrogen.Despite thedifferences in lagphase (Figure1) andmaximum fermentation

rates(Figure4),thefermentationdurationwasgenerallylongenoughforthesluggishfermentations

toreachcompletionwhennitrogenwasabove50mg/L.

The maximum weight loss was similar for both sugar levels when nitrogen was limiting

(Figure8B).Theincreaseinnitrogenresultedinanincreaseinthetotalweightloss.

3.3.1.6 Impact of treatments on the totalweight loss of individual yeast

strains

The fermentation performances of the individual strains were hampered by the nitrogen deficient

conditions(50mg/L),asseenbythelowtotalweightloss(Figure9).Fermentationsconductedwith

the strains WE372, and 228 were most severely hampered by the low nitrogen levels (50 and

100mg/L),intermsoftotalweightloss.ThestrainsVIN7,EC1118andNT50wereamongthosethat

lost themostoverallweight, suggesting that theyused thenitrogenmoreefficiently.Despitea slow

initialresponse(EC50andhill‐slope),VIN7isamongthestrainswiththehighesttotalweightlossat

verylownitrogenlevels(50mg/L),suggestingthatwhengivetimeVIN7isabletoovercomeadverse

fermentation conditions. At the higher nitrogen levels, the similar top values for the weight loss

indicatesasimilardegreeofcompletionoffermentation.

Fermentations at higher temperature resulted in a higher totalweight loss (Figure10). At both

temperatures VIN7 and AlchemyII were among the strains that lost the most weight and 228 and

WE372wereamongthestrainsthatlosttheleastamountofweight.

Theinteractionsbetweentheinitialsugarconcentrationsandthestrainsarenotsignificantforthe

topvalues(datanotshown).


42

Figure8: (A)Influenceofnitrogen(50,100,250,or400mg/L),andtemperature(15°Cor20°C)onthetotalweightloss(topvalues).Thesugarandstraindatahasbeencombined(B)Influenceofnitrogen(50,100,250,or400mg/L),andsugar(200or240g/L)onthetotalweightloss(topvalues).Thetemperatureandstraindatahaswere combined. Error bars indicate 95% confidence intervals for the means. Letters indicate significantdifferencesona5%(p<0.05)significancelevel.


N50 N100 N250 N400

Nitrogen

4.0

4.5

5.0

5.5

6.0

6.5

7.0

7.5

8.0

8.5

9.0

9.5

Top

(g)

d

bb

a aa

cc

Sugar S200 Sugar S240

N50 N100 N250 N400

Nitrogen

5.0

5.5

6.0

6.5

7.0

7.5

8.0

8.5

9.0

9.5

To

p (

g)

aa

bbc

c

dd

e

A

B


43

Figure9:Thetotalweightloss,measuredasthetopvalue,of17commercialculturesfermentingsyntheticgrapecontaining50,100,250,or400mg/Lnitrogen.Thesugarand temperaturedatawerecombined, resulting inasinglevalue foreachstrainat theevaluatednitrogen levels.Errorbars indicate95%confidence intervals for themeans.


EC1118AWRI 796

PR7228

N96NT50

NT112NT116

WE14WE372

VIN7VIN13

VIN2000NT45

NT202AlchemyI

AlchemyII4

5

6

7

8

9

10

Top

(g)


44

Figure10:Thetotalweightloss(topvalue)of17commercialculturesfermentingsyntheticgrapeat15and20°C.Thesugarandnitrogendatawerecombined,resultinginasinglevaluefortheevaluatedtemperatures.Errorbarsindicate95%confidenceintervalsforthemeans.


EC1118AWRI 796

PR7228

N96NT50

NT112NT116

WE14WE372

VIN7VIN13

VIN2000NT45

NT202AlchemyI

AlchemyII

5.5

6.0

6.5

7.0

7.5

8.0

8.5

9.0

Top

(g)


45

3.3.2 Influenceoftreatmentsonsugarconsumption

3.3.2.1 Impactoftreatmentsonsugarconsumption

Fermentation temperature plays a pivotal role in sugar metabolism (Figure 11). The higher

temperature(20°C)consistentlyresultedinalowerlevelofresidualsugarsattheendoffermentation

comparedtothoseat15C.

Nitrogen is essential for the production and maintenance of biomass, therefore nitrogen also

affects the rate of sugar consumption (Bisson, 1991; 1999). As the nitrogen content increased, a

decreaseinresidualsugarsoccurred,thisagreeswithfindingsreportedinotherstudies(Ingledew&

Kunkee,1985;MendesFerreiraetal.,2004).Fermentationproceededtodryness(<4g/Lsugar),for

the 200 g/L sugar treatments, containing higher nitrogen levels (250 and 400 mg/L) at both

fermentation temperatures. When the initial sugar level was raised to 240 g/L, the fermentations

containing250or400mg/Lnitrogenwereonlyabletoreachdrynesswhenfermentedat20C.It is

likelythatthecombinationofhighsugarandlowtemperaturegaverisetoasluggishfermentationthat

possiblyrequiredmorethan21daystofermenttodryness.

Figure11:Influenceofthenitrogen(50,100,250or400mg/L),sugar(200or240g/L),andtemperature(15°Cor20°C)treatmentsonthetotalresidualsugarlevels.Thedataforallthestrainswithineachspecifictreatmentwere combined. Error bars indicate 95% confidence intervals for the means. Letters indicate significantdifferencesona5%(p<0.05)significancelevel.


Nitrogen:N50

N100N250

N4000

20

40

60

80

100

120

Res

idua

l Sug

ar (

g/L)

Sugar: S240

Nitrogen:N50

N100N250

N400

d

abc

f

c

j

g

f

bc bcab a ab

k

i

h

e


46

3.3.2.2 Impactof treatmentson thesugarconsumptionof individualyeaststrains

The yeast strains were unable to ferment to dryness under the nitrogen deficient conditions of

50mg/Lnitrogen(Figure12).However,thelevelofresidualsugarreachedbyVIN7wassignificantly

lower than the other strains when nitrogen was limiting (50 and 100 mg/L) making it the most

effective strain in terms of sugar consumption; this is also reflected in its high total weight loss

(Figure8).

Theincreaseinnitrogenaidedthefermentationkinetics(EC50,hill‐slopeandtopvalue)andinso

doing it alsopromotedsugarmetabolism.ThestrainsAWRI796,NT50,andNT116hadsignificantly

higher levels of residual sugarswhen fermentingmust containing 400mg/L nitrogen compared to

250mg/L.Moststrains(228,AlchemyI,AlchemyII,EC1118,N96,NT112,NT202,NT45,PR7,VIN2000,

VIN13, andWE14) reduced the sugar concentration to similar levels at the 250 and the 400mg/L

nitrogenlevels.

ThestrainsVIN7,EC1118,andNT45displayasimilardegreeofsugarconsumptionat100mg/Las

otherstrainsfermenting400and250mg/Lnitrogen.Atallnitrogenlevels228,AWRI796andWE372

wereamongtheleasteffectivesugarconsumers,andVIN7,NT45andVIN2000wereamongthemost

effective. Ithasbeenproposedthatnitrogenexcess inducesrapidyeastgrowthandaconsequential

spike in ethanol; which cells are unable to respond to rapidly and consequently die

(Chaneyetal.,2006).Thismayexplaintherelativelypoorsugarconsumptionathighernitrogenlevels,

but it does not explain the increase in fermentation lag phase and the decrease in the maximal

fermentationrate,astheethanolconcentrationshouldbetoolowatthatjuncture.

Fermentation at 20°C enhanced sugar consumption when compared to the 15°C fermentations

(Figure13).Atboth fermentation temperaturesVIN7,EC1118,NT45andVIN2000wereamong the

most efficient sugar consumers. The yeast strains 228, andWE372were among the poorest sugar

consumers,where their sugar consumption at 20Cwas comparable to other strains fermenting at

15C.At15°C,VIN7,andN96consumedsugarsmorereadilythantheotherstrainstested(Figure13).

Thisisconfirmedbytheirslightlysuperiortotalweightloss(Figure9).

Thehigherinitialsugarcontentresultedinahigherresidualsugarlevelattheendoffermentation

(data not shown). In response to the initial sugar content, VIN7, EC1118,NT45 andVIN2000were

among themost efficient sugar consumers;whereas the strains 228 andWE372were consistently

amongthepoorestsugarconsumers.

These results highlight the importance of the selection an appropriate yeast starter culture to

conductafermentationofaspecificgrapemust,asthestressesevaluatedhaddramaticconsequences

onthesugarconsumption.


47

Figure12:Influenceofthedifferentnitrogenlevels(50,100,250or400mg/L)onthetotalresidualsugarlevelsof17commercialcultures.Thetemperatureandsugardatawerecombined,resultinginasinglevalueforeachstrainatthenitrogenlevelsevaluated.Errorbarsindicate95%confidenceintervalsforthemeans.Lettersindicatesignificantdifferencesona5%(p<0.05)significancelevel.


EC1118AWRI 796

PR7228

N96NT50

NT112NT116

WE14WE372

VIN7VIN13

VIN2000NT45

NT202AlchemyI

AlchemyII0

10

20

30

40

50

60

70

80

90

100

Res

idua

l Sug

ar (

g/L)


48

Figure13:Influenceofthefermentationtemperature(15°Cor20°C)onthetotalresidualsugarlevelsof17commercialcultures.Thetemperatureandsugardatawerepooled, providing a single value for each strain the evaluated temperatures. Error bars indicate 95% confidence intervals for themeans. Letters indicate significantdifferencesona5%(p<0.05)significancelevel.


EC1118AWRI 796

PR7228

N96NT50

NT112NT116

WE14WE372

VIN7VIN13

VIN2000NT45

NT202AlchemyI

AlchemyII0

10

20

30

40

50

60

Res

idua

l Sug

ar


49

Basedontheresults,VIN7isthestrainwhichmosteffectivelymetabolizedsugarsdespitethe

prevailingfermentationstresses.However,itslonglagphasemaycreateanopportunityforunwanted

non‐Saccharomycesstrainstodominatethefermentationforalongerperiod,andevenmoresoatlow

temperatureswherenon‐Saccharomyces strainswouldbesomewhatprotected fromethanol toxicity

(Gao&Fleet,1988;Heard&Fleet,1988).

3.3.3 Influenceoftreatmentsonyeastdryweight

3.3.3.1 Impactoftreatmentsonyeastdryweightproduction

At both temperatures, low nitrogen content has a greater impact on biomass production than the

initialsugarcontent(Figure13);as illustratedbyasimilardryweightat100and50mg/Lnitrogen

irrespectiveoftheinitialsugarcontent.

As thenitrogen content increases, thehigher initial sugar levelsnegatively impactsbiomass

production. At 15C, the biomass production has a similar pattern for both sugar concentrations,

showing an increase in biomass up to 250 mg/L nitrogen, which was followed by a decrease in

biomasswhennitrogenwasraisedto400mg/L.Asimilarpatternforbiomassproductionisseenat

20Cwhen the initial sugar is 240 g/L; however for the 200 g/L sugar fermentations the biomass

continuestoincreaseat400mg/Lnitrogen.

Biomass and intracellular activity are crucial in determining the fermentation rate; they are

bothaffectedbythenitrogenstatusofthemust(Varelaetal.,2004).The lownitrogenfermentation

treatmentscausea lowamountofbiomasstobeproduced(Figure13),resulting ina long lagphase

(Figure 1), a lowmaximum fermentation rate (Figure 4), and ultimately high residual sugar levels

(Figure10).

3.3.3.2 Impact of treatments on the dry weight production of individual

yeaststrains

At50mg/Lnitrogen,moststrainsproducedasimilaramountofbiomass(Figure15).Forboth50and

100mg/L,WE372andAWRI796producesignificantlymorebiomassthanWE14,VIN7andNT50.At

250mg/L NT116, 228,WE372were among the high biomass producers, and at 400mg/L VIN13,

VIN2000and228wereamongthehighbiomassproducers.Thebiomassincreasedwiththeincreasing

nitrogen levels (Beltran etal., 2005) until the 400mg/L levelwas reached. The dryweight for the

400mg/L treatments were statistically similar (AlchemyII, N96, NT112, NT45, PR7, VIN2000, and

VIN13),orsignificantlygreater,(228,AlchemyIandVIN7,andsignificantlyless,forAWRI796,EC1118,

NT116, NT202, NT50, WE14, WE372) than that of the 250 mg/L treatments. Even though VIN7


50

produced more biomass in response to 400 mg/L compared to 250 mg/L nitrogen, its maximum

biomasswassignificantlylessthanthatofthehighbiomassproducingstrains.

Figure14:Influenceofthenitrogen(50,100,250or400mg/L),sugar(200or240g/L),andtemperature(15°Cor20°C) treatmentson (dryweight (yeastbiomass).Thedata for theyeast strainwithina specific treatmentwere combined. Error bars indicate 95% confidence intervals for the means. Letters indicate significantdifferencesona5%(p<0.05)significancelevel.

In response to fermentation temperature, AWRI796, 228, NT116, WE372, NT202 and

AlchemyI,AlchemyII,VIN13andVIN7producedsignificantlymorebiomassat20Ccomparedto15C,

whereas the other strains had no significant change in their biomass production (Figure 16).

Regardless of the fermentation temperature,WE372 and 228 were the highest biomass producing

strains,whileVIN7,NT45,andWE14producedthelowestbiomass.

The relationship between the initial sugar levels and the yeasts strains evaluated was not

significant(datanotshown).

Sugar S200 Sugar S240Temperature: T15

Nitrogen:N50

N100N250

N4000.0010

0.0015

0.0020

0.0025

0.0030

0.0035

0.0040

0.0045

0.0050

Dry

wei

ght

(g/m

L)

Temperature: T20

Nitrogen:N50

N100N250

N400

ab

d d

ii

g ggh

f

ab

c

e e

h

j


51

Figure15: Influenceof thenitrogen(50,100,250or400mg/L)on thedryweight(biomass)of17commercialyeastcultures.Thesugarand temperaturedatawerepooled.Errorbarsindicate95%confidenceintervalsforthemeans.Lettersindicatesignificantdifferencesona5%(p<0.05)significancelevel.


EC1118AWRI 796

PR7228

N96NT50

NT112NT116

WE14WE372

VIN7VIN13

VIN2000NT45

NT202AlchemyI

AlchemyII0.001

0.002

0.003

0.004

0.005

0.006

Dry

wei

ght (

g/m

L)


52

Figure16:Influenceoffermentationtemperature(15°Cor20°C)onthedryweight(biomass)of17commercialyeastcultures.Thenitrogenandsugardatawaspooled.Errorbarsindicate95%confidenceintervalsforthemeans.Lettersindicatesignificantdifferencesona5%(p<0.05)significancelevel.


EC1118AWRI 796

PR7228

N96NT50

NT112NT116

WE14WE372

VIN7VIN13

VIN2000NT45

NT202AlchemyI

AlchemyII

strain

0.0024

0.0026

0.0028

0.0030

0.0032

0.0034

0.0036

0.0038

0.0040

0.0042

Dry

wei

ght g

/mL


53

3.4 Conclusion

Thisstudyconfirmsthatthefermentationperformancesofcommercialyeastculturesaresignificantly

anddifferently affected by temperature (Fleet&Heard, 1993), initial nitrogen (Jiraneketal., 1991;

Manginotetal.,1998;Taillandieretal.,2007)andsugarlevels(Erasmusetal.,2004),whenfermenting

adefinedsyntheticgrapemust.

From the results obtained in this study it is clear that, under the conditions tested, the

relationshipbetween initialnitrogen leveland fermentationperformance isnotpurely linear,but is

influencedgreatlybyparameterssuchasinitialsugar,fermentationtemperature,andyeaststrain.As

previously reported, when the nitrogen is limiting, the fermentation rate, sugar consumption and

biomassproductionareallnegativelyaffected (Henschke& Jiranek,1993;Blateyron&Sablayrolles,

2001). The 250 mg/L nitrogen treatment showed an improved overall performance in terms of

fermentation onset, maximal fermentation rate, total weight loss, sugar consumption, and biomass

production compared to the fermentations supplemented with 100 or 400 mg/L nitrogen. This

highlights the importance ofmeasuring the YAN content of the grapemust prior to the addition of

DAP,orcomplexnutrients,asitsadditioninexcessmayhinderfermentationefficiencyandcompletion

(Taillandieretal.,2007).

When experiencing hyperosmotic stress, yeast growth stops and only resumes once the

accumulationofglycerolhasresulted insufficientwateruptaketorestorecellsize toacritical level

(Hohmann,1997;Mager&Siderius,2002).Consequently,thehighertheinitialsugarconcentrationthe

longerthelagphase(EC50)beforefermentationstarts,duetothecessationofgrowthwhilethecell

adaptstotheprevailingosmoticconditions.Thisisinagreementwithourresults,asthehigherinitial

sugarcontentgenerallyresultedinanincreaseintheEC50value(fermentationonset),adecreasein

the initial fermentation rate (hill‐slope), an increase in the top value (total weight lost), and an

increaseintheresidualsugarlevels.

Temperatureplaysaninfluentialroleinyeastgrowthandmetabolism(Fleet&Heard,1993).

S. cerevisiae has an optimum growth temperature of 25°C; therefore fermentations at lower

temperaturewouldbeslowerandalsolonger.Itisnotsurprisingthenthatanincreaseintemperature

(20°C)significantlyinfluencedthefermentationratesandfermentationcompleteness,atallnitrogen

levels.Itisinterestingthatatlownitrogenlevels(50mg/Land100mg/L),anincreaseintemperature

(20°C) is able to overcome the nitrogen deficiency to a certain extent and evenmore sowhen the

initial sugar concentration (200 g/L) is low. There seemed to be a correlation between the

fermentationrate,fermentationcompletenessandbiomassformationateachofthesetsofconditions

tested foreachstrain.Somestrains (e.g.228,WE372andAWRI796)appear tobemoresensitive to

nitrogenlimitation,especiallyunderlowertemperaturesorhighsugarconditionsthanothers(VIN7),

pointingtothelatterstrainbeingmorenitrogenefficientandstressresistant.


54

When the fermentation conditions were “more stressful” the selection of the appropriate

commercialstrainmadeasignificantcontributiontofermentationonset,rateandsugarconsumption.

Thustheuseofastrainadepttofermentingacharacterisedgrapemusttodrynessmayreducetherisk

ofproblematicfermentations.

3.5 Acknowledgements

WeareindebtedtoAnchorYeastforfunding.WewouldalsoliketothankLaurenJoosteforlaboratory

assistance,andProf.MartinKiddforthestatisticalanalysis.

3.6 Literaturecited

Agenbach,W.,1977.Astudyofmustnitrogencontent inrelationto incomplete fermentations,yeast

productionandfermentationactivity.In:(ed).ProceedingsoftheSouthAfricanSocietyofEnologyand

Viticulture.Stellenbosch,pp.66‐88

Alexandre,H.&Charpentier,C.,1998.Biochemicalaspectsofstuckandsluggishfermentationingrape

must.J.Ind.Microbiol.Biotechnol.20,20‐27.

Attfield, P.V., 1997. Stress tolerance: the key to effective strains of industrial baker's yeast. Nature

Biotechnology.15,1351‐1357.

Bauer,F.&Pretorius, I.,2000.Yeaststressresponseandfermentationefficiency:howtosurvivethe

makingofwine‐areview.S.Afr.J.Enol.Vitic.21,27‐51.

Bell,S.J.&Henschke,P.A.,2005.Implicationsofnitrogennutritionforgrapes,fermentationandwine.

Aust.J.GrapeWineRes.11,242‐295.

Beltran, G., Esteve‐Zarzoso, B., Rozès, N.,Mas, A.& Guillamón, J.M., 2005. Influence of the timing of

nitrogenadditionsduringsyntheticgrapemustfermentationsonfermentationkineticsandnitrogen

consumption.J.Agric.FoodChem.53,996‐1002.

Beltran,G.,Novo,M.,Leberre,V.,Sokol,S.,Labourdette,D.,Guillamon,J.M.,Mas,A.,François,J.&Rozes,

N.,2006.Integrationoftranscriptomicandmetabolicanalysesforunderstandingtheglobalresponses

oflowtemperaturewinemakingfermentations.FEMSYeastRes.6,1167‐1183.


55

Bely,M.,Sablayrolles,J.&Barre,P.,1990a.Descriptionofalcoholicfermentationkinetics:itsvariability

andsignificance.Am.J.Enol.Vitic.41,319‐324.

Bely,M.,Sablayrolles,J.&Barre,P.,1991.Automaticdetectionandcorrectionofassimilablenitrogen

deficiencyduringalcoholicfermentationunderenologicalconditions.In:Ranz,J.M.(ed).Proceedings

oftheinternationalsymposiumonnitrogeningrapesandwine.Davis,CA,pp.211–214

Bely,M.,Sablayrolles,J.M.&Barre,P.,1990b.Automaticdetectionofassimilablenitrogendeficiencies

duringalcoholicfermentationinoenologicalconditions.J.Ferment.Bioeng.70,246‐252.

Bezenger, M.C. & Navarro, J.M., 1988. Alcoholic fermentation: Model accounting for initial nitrogen

influence.Biotechnol.Bioeng.31,747‐749.

Bisson, L.F., 1991. Influence of nitrogen on yeast and fermentation of grapes. In: Ranz, J.M. (ed).

Proceedingsoftheinternationalsymposiumonnitrogeningrapesandwine.Davis,CA,pp.78‐89


Bisson,L.F.&Butzke,C.E.,2000.Diagnosisandrectificationofstuckandsluggishfermentations.Am.J.

Enol.Vitic.51,168‐177.

Blateyron,L.&Sablayrolles, J.M.,2001.Stuckandslow fermentations inenology:statistical studyof

causes and effectiveness of combined additions of oxygen and diammonium phosphate. J. Biosci.

Bioeng.91,184‐189.

Bradbury, J.E., Richards, K.D., Niederer, H.A., Lee, S.A., Rod Dunbar, P. & Gardner, R.C., 2006. A

homozygousdiploidsubsetofcommercialwineyeaststrains.AntonievanLeeuwenhoek.89,27‐37.

Carrasco,P.,Querol,A.&delOlmo,M.,2001.Analysisofthestressresistanceofcommercialwineyeast

strains.Arch.Microbiol.175,450‐457.

Chaney,D.,Rodriguez,S.,Fugelsang,K.&Thornton,R.,2006.Managinghighdensitycommercialscale

winefermentations.J.Appl.Microbiol.100,689‐698.

Erasmus,D.J.,Cliff,M.&vanVuuren,H.J.J.,2004.Impactofyeaststrainontheproductionofaceticacid,

glycerol,andthesensoryattributesoficewine.Am.J.Enol.Vitic.55,371‐378.

Erasmus,D.J.&vanVuuren,H.J.J.,2009.GeneticBasisforOsmosensitivityandGeneticInstabilityofthe

WineYeastSaccharomycescerevisiaeVIN7.Am.J.Enol.Vitic.60,145‐154.


56

Fleet, G. & Heard, G., 1993. Yeast‐growth during fermentation. In: Fleet, G. (ed). Wine Microbiol.

Biotechnol.HarwoodAcademicPublishers,Switzerland.pp.27‐54.

Gao,C.&Fleet,G.,1988.TheeffectsoftemperatureandpHontheethanoltoleranceofthewineyeasts,

Saccharomycescerevisiae,CandidastellataandKloeckeraapiculata.J.Appl.Microbiol.65,405‐409.

Gibson,B.R.,Lawrence,S.J.,Leclaire,J.P.R.,Powell,C.D.&Smart,K.A.,2007.Yeastresponsestostresses

associatedwithindustrialbreweryhandling.FEMSMicrobiol.Rev.31,535‐569.

Hallsworth,J.E.,1998.Ethanol‐inducedwaterstressinyeast.J.Ferment.Bioeng.85,125‐137.

Heard,G.&Fleet,G.,1988.TheeffectsoftemperatureandpHonthegrowthofyeastspeciesduringthe

fermentationofgrapejuice.J.Appl.Microbiol.65,23‐28.

Henschke,P.&Jiranek,V.,1993.Yeasts–metabolismofnitrogencompounds.In:Fleet,G.H.(ed).Wine

Microbiol.Biotechnol.HarwoodAcademicPublishers,Switzerland.pp.77‐164.

Hohmann,S.,1997.Shapingup:Theresponseofyeasttoosmoticstress.In:Hohmann,S.&Mager,W.H.

(ed).YeastStressResponses.Springer.pp.101‐145.

Ingledew,W.&Kunkee, R.E., 1985. Factors influencing sluggish fermentations of grape juice. Am. J.


Ivorra,C.,PérezOrtín,J.E.&delOlmo,M.,1999.Aninversecorrelationbetweenstressresistanceand

stuckfermentationsinwineyeasts.Amolecularstudy.Biotechnol.Bioeng.64,698‐708.

Jiranek, V., Langridge, P. &Henschke, P., 1991. Yeast nitrogen demand: selection criterion forwine

yeastsforfermentinglownitrogenmusts.In:Rantz,J.M.(ed).Internationalsymposiumonnitrogenin

grapesandwine.1991,Davis,CA.pp.266‐269

Kunkee, R.E., 1991. Relationship between nitrogen content of must and sluggish fermentation. In:

Rantz,J.M.(ed).Internationalsymposiumonnitrogeningrapesandwine.Davis,CA.pp.148‐155

Lafon‐Lafourcade,S.,1983.Wineandbrandy.In:Rehm,H.J.&Reed,G.(ed).FoodandFeedProduction

withMicroorganisms.Biotechnology.VerlagChemie.pp.81‐163.

Llauradó, J., Rozés, N., Bobet, R., Mas, A. & Constantí, M., 2002. Low temperature alcoholic

fermentationsinhighsugarconcentrationgrapemusts.J.FoodSci.67,268‐273.

Llauradó,J.M.,Rozès,N.,Constantí,M.&Mas,A.,2005.StudyofsomeSaccharomycescerevisiaestrains

forwinemakingafterpreadaptationatlowtemperatures.J.Agric.FoodChem.53,1003‐1011.


57

Mager,W.H.&Siderius,M.,2002.Novelinsightsintotheosmoticstressresponseofyeast.FEMSYeast

Res.2,251‐257.

Malherbe,S.,Bauer,F.&DuToit,M.,2007.Understandingproblemfermentations–Areview.S.Afr.J.


Manginot, C., Roustan, J. & Sablayrolles, J., 1998. Nitrogen demand of different yeast strains during

alcoholicfermentation.Importanceofthestationaryphase.Enzym.Microb.Technol.23,511‐517.

Mendes Ferreira, A., Mendes Faia, A. & Leao, C., 2004. Growth and fermentation patterns of

Saccharomyces cerevisiae under different ammonium concentrations and its implications in

winemakingindustry.J.Appl.Microbiol.97,540‐545.

Pizarro, F.J., Jewett, M.C., Nielsen, J. & Agosin, E., 2008. Growth temperature exerts differential

physiological and transcriptional responses in laboratory and wine strains of Saccharomyces

cerevisiae.Appl.Environ.Microbiol.74,6358‐6368.

Pretorius,I.,VanderWesthuizen,T.&Augustyn,O.,1999.Yeastbiodiversityinvineyardsandwineries

anditsimportancetotheSouthAfricanwineindustry.Areview.S.Afr.J.Enol.Vitic.20,61‐75.

Stanley,D.,Bandara,A.,Fraser,S.,Chambers,P.&Stanley,G.,2010.Theethanolstressresponseand

ethanoltoleranceofSaccharomycescerevisiae.J.Appl.Microbiol.109,13‐24.

Taillandier,P.,RamonPortugal,F.,Fuster,A.&Strehaiano,P.,2007.Effectofammoniumconcentration

onalcoholic fermentationkineticsbywineyeasts forhighsugarcontent.FoodMicrobiology.24,95‐

100.

Varela, C., Pizarro, F. & Agosin, E., 2004. Biomass content governs fermentation rate in nitrogen‐

deficientwinemusts.Appl.Environ.Microbiol.70,3392‐3400.

Zuzuarregui, A. & del Olmo, M., 2004. Analyses of stress resistance under laboratory conditions

constituteasuitablecriterionforwineyeastselection.AntonievanLeeuwenhoek.85,271‐280.


58

Chapter4

Researchresults

Impactofenvironmentalstressonaromaproductionduringwinefermentation


59

Chapter4

Impactofenvironmentalstressonaromaproductionduringwinefermentation

Abstract

Winecharacterandstylearemostlydefinedbytheflavourofindividualwinesandareofcrucial

importancetothewineconsumers(Swiegersetal.2005).Whiletheperceptionofwineflavour,

which includesaroma, tasteandmouth‐feel, is somewhat subjective,mostwinesaremade to

reflectortocorrespondtoaspecific flavourprofile.Winemakershaveanumberof important

toolstotrytoachievespecificwinestyles,oneofwhichisthechoiceofaspecificyeaststrains.

Yeast strains contribute to wine flavour by the production of many volatile metabolites

including esters, higher alcohols, carbonyl compounds, volatile fatty acids, and sulphur

containingcompoundsanddifferentyeaststrainsareknowntoproducesignificantlydifferent

flavourprofiles.However,theinteractionofyeastwithspecificgrapejuicesandtheirresponses

to changing environmental conditions remains largely unexplored.During fermentationyeast

strains continually experience stresses that may impact yeast viability, growth, and

fermentation performance. These consequences of stress have been well‐studied under

laboratory conditions, but current literature provides few insights on the impact of

environmentalstressesduringwine fermentationonwinequality,and inparticularonaroma

production. This is the first study to investigate the impact of two common fermentation

associatedstresses,hyperosmoticandtemperaturestress,onfermentationperformanceandon

theproductionofaromacompoundsinsyntheticgrapemust.Theresultsdemonstratethatthe

stressconditionsresultedinanumberofsignificantchangestothearomaprofile.Furthermore,

the changes observed differed for each strain and each stress treatment. This implies that

environmentalstresscausesnumerouschangestothearomaprofilesinamannerwhichisnot

conservedamongthefivewinestrainstested.

Key words: Wine aroma, hyperosmotic stress, temperature stress, synthetic grape must

fermentation


60

4.1 Introduction

Wine yeast strains mediate the conversion of sugar to ethanol and CO2. This conversion

proceedsdespite continuous exposure to various formsof stress (Bisson1999). The stresses

may include temperature changes, hyperosmotic stress, ethanol stress, vitamin, mineral,

nitrogen, and oxygen deficiencies, and a low pH, among others (Alexandre and Charpentier

1998, Attfield 1997, Bauer and Pretorius 2000, Bisson 1999,Malherbe et al. 2007). There is

evidence that such stresses are also often associated with problematic (stuck and sluggish)

fermentations(Bisson1999,Gibsonetal.2007,Malherbeetal.2007).

Thehighinitialsugarcontentofgrapemustresultsinhyperosmoticstress,whichthecell

counteracts through the accumulation of intracellular glycerol (Hohmann 1997, Mager and

Siderius2002,TamásandHohmann2003).Lowtemperaturedecreasesandhightemperature

increasesmembrane permeability, which disrupts transport systems into and out of the cell

(Beales2004,Bisson1999).

Yeast strains differ in their abilities to sense and respond to stress. Most of these

stresses have been extensively studied with regard to their impact on yeast growth and

fermentationperformance(Carrasco et al. 2001, Ivorra et al. 1999, Zuzuarregui anddelOlmo

2004), yet literature provides few insights into the impact of fermentation‐related stress

appliedduringalcoholicfermentationonwineflavour.

Wineflavourisdefinedastheaccumulativeeffectofsmell,tasteandmouth‐feel(Francis

andNewton2005).Itisthecompositeproductofacombinationofmetabolitesderivedfromthe

grapes, frommicroorganismsduring fermentation, aswell as fromchemicalprocessesduring

productionandmaturation(RappandMandery1986,RappandVersini1996).Theperceived

flavour is the result of complex interactions between all the volatile and non‐volatile

compoundspresentinwine.Compoundspresentatlevelsabovetheirdetectionthresholds,may

maskorsuppressthedetectionofothers.However,whenpresentbelowthedetectionthreshold

levels they may act synergistically with other compounds, and in so doing enhance wine

complexity (Francis andNewton2005).The volatile aromasproduceddenovo bywine yeast

duringfermentationincludehigheralcohols,estersandvolatilefattyacids,amongothers(Rapp

andVersini1996).Higheralcoholsareformedbythedecarboxylationandreductionof‐keto

acids, originatingeither fromglycolysisor theEhrlichpathway (Hazelwoodet al. 2008,Rapp

and Mandery 1986, Rapp and Versini 1996, Swiegers et al. 2005). Volatile fatty acids are

produced as a result of acetyl‐CoA decarboxylation and condensation reactions (Lambrechts

and Pretorius 2000). Acetate esters are formed by enzyme catalysed condensation reactions

betweenacetyl‐CoAandhigheralcoholsorethanol(BellandHenschke2005,BissonandKarpel

2010, Lambrechts and Pretorius 2000), while ethyl esters are formed by the condensation


61

reaction between ethanol and either a fusel acyl‐CoA or a fatty acyl‐CoA (Bisson and Karpel

2010).

Themetabolicpathwaysleadingtotheformationofthesevolatilearomacompoundsare

reasonablywellmappedandestablished.However,theregulationofthemetabolicfluxthrough

these networks is notwell understood. Studies have shown that obvious factors such as the

availability of precursors, fermentation conditions (Bisson and Karpel 2010, Henschke and

Jiranek 1993, Lambrechts and Pretorius 2000, Rapp and Versini 1996, Saerens et al. 2008,

Swiegers et al. 2005, Vilanova et al. 2007) and the genetic make‐up of individual strains

(Rossouw et al. 2008, Soles et al. 1982) play an important role in themodulation of aroma

production.However,littledataisavailableregardingtheimpactofenvironmentalfactorsand

inparticularofenvironmentalstressontheproductionofsuchmetabolites.

Hereweinvestigatethe impactoftwoofthemostcommonstressesexperiencedduring

alcoholic fermentation, temperature shifts andhyperosmotic stress, on the aromaproduction

capacityofseveralwineyeaststrains.Toourknowledge,thisisthefirststudyinvestigatingthe

impact of hyperosmotic and temperature stresses on fermentation performance and the

production of fermentation derived volatile aroma compounds in synthetic grapemust. This

preliminarystudyprovidesinformationonhowtheexposuretostressimpactswinearomaand

whethertheobservedchangesareconservedamongdifferentcommercialSaccharomycesyeast

strains.

4.2 MaterialsandMethods

4.2.1 Yeaststrainsandgrowthconditions

The synthetic grapemustwas as described previously by Henschke and Jiranek (1993), and

contained 100 g/L glucose and 100 g/L fructose. The vitamins, minerals, anaerobic factors

(Henschke and Jiranek 1993) as well as the nitrogen sources, amino acids and ammonium

chloride,(Belyetal.1990)werefiltersterilizedandaddedtotheautoclavedcarbon,acidand

salt sources (Henschke and Jiranek 1993). The nitrogen sources (Bely et al. 1990), were

proportionallydecreasedtoreachdesiredlevelof250mg/Lnitrogen.

Theyeaststrainsused inthisstudyare listed inTable1.Strainswerestoredat ‐80°C,

andwerestreakedoutontoYPDplates,whichweresubsequently incubatedat30°C.Asingle

colonywasused to inoculate5mL synthetic grapemust;whichwas aerobically incubated at

30°Covernight.Thesecultureswereused to inoculate the fermentationsatan initialOD600of

0.1,correspondingtoapproximately1x106cells/mL.


62

Table1.ListofSaccharomyceswineyeaststrainsusedinthisstudy

CommercialName Commercial

Source

Strain

LalvinEC1118LallemandInc.

Montréal,Canada

Saccharomycescerevisiaebayanus

IWBT2851

(CrossEvolutionTM)

S.cerevisiaehybrid

NT50AnchorYeast,Cape

Town,SouthAfrica

S.cerevisiae hybrid

VIN7 S.kudriavzevii andS.cerevisiaehybrid2

VIN13 S.cerevisiae hybrid

1CrossEvolutionwillbereferredtobyitsIWBTname285fromthispointforward.

2Bradburyetal.(2006)

4.2.2 Fermentationconditions

Thefermentationswereperformedin100mLglassbottles,containing70mLofsyntheticgrape

must. The bottles were sealed with rubber stoppers and a CO2 outlet. Fermentations were

conductedintriplicateat20°C,unlessstatedotherwise.

Toinducehyperosmoticstress,sorbitolwasaddedtoafinalconcentrationof0.3M(S1)or

0.5M(S2)uponinoculation.Toevaluatetheimpactoftemperaturestressasetoffermentations

wasmoved to8C (T8)or37C (T37) for16hoursonday two (D2)ordayeight (D8)of the

fermentationandsubsequently returned to20C.Fermentationswhichwerenot subjected to

anyadditionalstress,servedasthecontroldataset.

Thefermentationswereweighedtomonitorfermentationprogressfor21days,whichwas

expressedasCO2weightloss.

4.2.3 ChemicalAnalysis

4.2.3.1 Residualglucoseandfructose,andglycerol

SyntheticwinesampleswerescannedusingtheWinescanFT120instrument(FOSSAnalytical

A/S software version 2.2.1) equipped with a purpose‐built Michelson interferometer (FOSS

Analytical A/S, Hillerød, Denmark) to generate Fourier transform mid infrared (FT‐MIR)

spectra.Quantifiedchemicaldataforresidualglucose,fructose,glycerol,andvolatileacidlevels

werepredictedfrominfraredspectrabycommercialcalibrationsorin‐houseadjustmentsusing

theWinescanFT1202001version2.2.1software.


63

4.2.3.2 Aromacompounds

Fortheextractionofthevolatilearomacompoundsfromthesamples,theprotocoldescribedby

Louw et al. (2010) was followed with a few minor modifications. The internal standard 4‐

methyl‐2‐pentanol(100µL)and1mLdiethyletherwereaddedtoa5mLsampleofwine.The

wine/ether mixture was subjected to 5 minutes of sonication. The wine/ether mixture was

centrifuged at4000g for3minutes,Na2SO4wasadded to themixture and the centrifugation

stepwasrepeated.TheetherlayerwasremovedanddriedonadditionalNa2SO4.

The volatile higher alcohols, esters, fatty acids and carbonyl compounds (Table 2) were

quantifiedinduplicateusingaHewlettPackard6890Plusgaschromatograph(LittleFalls,USA)

with a split/splitless injector and a flame ionization detector. The protocol described by

Malherbe (2011) was followed with a fewmodifications. The separation of compounds was

achieved using a DB‐FFAP capillary GC column (Agilent, Little Falls, Wilmington, USA) with

dimensions 20m length x 0.1mm inside diameter x 0.2 m film thickness. The initial oven

temperaturewasmaintainedat33°Cfor8minutesafterwhichthetemperaturewasincreased

by12°C/minuteuntil240wasreached.Thistemperaturewasheldfor5minutes.A1lsample

wasinjectedwhentheoventemperaturereached250°C.Thesplitratio10:1andthesplitflow

ratewas36.7cm/s.Thecolumnflowratewas6.6mL/minusinghydrogenas thecarriergas.

Thedetectortemperaturewas230°C.Aftereachsample,oventemperaturewasmaintainedat

250°C,withacolumnflowrateof30mL/mintocleanthecolumnofallcontaminantswithhigh

boilingpoints.After22injectionsthecolumnwascleanedboththermallyandchemicallybya

hexaneinjectionatanoventemperatureof250(Louwetal.2010).

4.2.4 StatisticalAnalysis

MultivariatestatisticalanalysiswasperformedusingTheUnscramblersoftware(version9.2,

CAMOASA,Norway).Principalcomponentanalysis(PCA)wasusedtoillustrateandevaluate

thedistributionofstresstreatmentsrelativetoeachotherbasedontheirindividualvolatile

aromaprofiles.

Additionally, fermentation data for stressed treatmentswere compared to that of the

‘unstressed’ control by creating pair‐wise comparisons for each strain. These graphs were

visualised using Cytoscape (version 2.8.2, http://www.cytoscape.org). These graphs only

includeinformationthatdifferedsignificantlyfromthecontrol;thusanyomissionwasdeemed

asstatisticallysimilartoitscontrol.Inallcasesasignificancelevelof5%(p<0.05)wasused.


64

A blue node (ellipse) or edge (line) indicates a reduction compared to the control,

whereasarednodeoredgedenotesanincrease.Thecolourintensityofthenodeoredgeisan

indicationofthemagnitudeofthefoldchange.

Table2.The39volatilearomacompoundsidentifiedandquantifiedusingGC‐FID

Esters HigherAlcohols VolatileFattyAcids Carbonylcompounds

Ethylesters

Ethyldecanoate Hexanol Aceticacid Acetoin

Ethylhexanoate Butanol Propionicacid

Ethylbutyrate Methanol Isobutyricacid

Ethyloctanoate 2‐Phenylethanol Butyricacid

Ethyllactate Propanol Isovalericacid

Ethylpropionate Isobutanol Valericacid

Ethyl‐2‐methylpropanoate Isoamylalcohol Hexanoicacid

Ethyl‐2‐methylbutyrate Pentanol Octanoicacid

Ethylisovalerate 4‐Methyl‐1‐pentanol Decanoicacid

Ethyl‐3‐hydroxybutanoate 3‐Methyl‐1‐pentanol

Acetateesters 1‐Ethoxy‐1‐propanol

Ethylacetate 1‐Octen‐3‐ol

Ethylphenylacetate

Hexylacetate

2‐Phenylethylacetate

Diethylsuccinate

2‐Methyl‐propylacetate

Isoamylacetate


65

4.3 Results

4.3.1 Fermentationperformance

4.3.1.1 Fermentationrate

The stresses evaluated were relatively mild so as to mimic those found in industrial

fermentations.Generally,thefermentationrates(expressedasweightlossbyCO2evolution)of

the yeast strains differed from each other, as did their response to both hyperosmotic and

temperaturestresses(Fig.1and2).EC1118(S1andS2)and285(S2)displayedan increased

fermentation speedwhen exposed to the hyperosmotic stress treatmentswhen compared to

their respective controls (Fig. 2A andB). Similarly in response to temperature stress,VIN13

(8Cand37C)andEC1118(37C)displayedenhancedfermentationrates(Fig.3G,HandC).

The severity (Gasch et al. 2000, Gasch and Werner‐Washburne 2002) of the stress

application also had a significant impact on the fermentation rate. This is illustrated by the

similarfermentationratesoftheS1(0.3Msorbitol)treatmentsandthecontrolfermentations,

and differences in the fermentation rates that only emergedwhen the level of hyperosmotic

stresswasincreasedto0.5Msorbitol(Fig.1).

4.3.1.2 Residualglucoseandfructose,andglycerol

Thesyntheticgrapemustwasnot fermentedtodrynesswithinthe21daysoftheexperiment

andstresstreatmentsgenerallydidnotleadtosignificantchangeswithregardtoresidualsugar

levels. The few cases where the treatments differed significantly from the controls are

illustratedinFig.2.Glucoseandfructoseconsumptionwassignificantlydecreasedinthecaseof

VIN7_D2.T37 when compared to the VIN7 control. Exposure to 8C on day 8 significantly

enhancedthesugar(bothglucoseandfructose)consumptionabilityofVIN13comparedtothe

VIN13controltreatment(Fig.3);whichcorrespondswithitsimprovedfermentationrateinFig.

2h.

Inresponsetohyperosmoticstress,allstrainsproducedmoreglycerolincomparisonto

theircontrols(Fig.4),butnoincreaseinaceticacidwasobservedbymeasurementwithGC‐FID.

This was confirmed by analysis with FT‐MIR, which found that the volatile acidity levels

(predominantlyaceticacid)rangedfrom0.5to0.6g/Lforthecontrolfermentations,1.4to1.6

g/LforS1and2to2.2g/LforS2treatments.


66

0

2

4

6

8

0 6 12 17

Wei

ght L

oss

(g)

Time (day)

Control EC1118 S1 EC1118 S2 EC1118

0

2

4

6

8

0 6 12 17

We

ight

Los

s(g)

Time (day)

Control VIN13 S1 VIN13 S2 VIN13

0

2

4

6

8

0 6 12 17

Wei

ght L

oss

(g)

Time (day)

Control 285 S1 285 S2 285

0

2

4

6

8

0 6 12 17

Wei

ght l

oss

(g)

Time (day)

Control NT50 S1 NT50 S2 NT50

0

2

4

6

8

0 6 12 17

Wei

ght

loss

(g

)

Time (day)

Control VIN7 S1 VIN7 S2 VIN7

Figure1Fermentationratesofstrains(a)EC1118,(b)285,(c)NT50,(d)VIN13and(e)VIN7inresponsetohyperosmoticstresstreatments,S1andS2comparedtoacontrol.Resultsaretheaverageoftriplicatefermentations(errorbarsarenotshownfortheclarityofthegraphs)

a

dc

b

e


67

0

2

4

6

8

0 6 12 17

Wei

ght l

oss

(g)

Time (day)

Control 285 D2.T37 285 D8.T37 285

0

2

4

6

8

0 6 12 17

Wei

ght l

oss

(g)

Time (day)

Control 285 D2.T8 285 D8.T8 285

0

2

4

6

8

0 6 12 17

Wei

ght l

oss

(g)

Time (day)

Control EC1118 D2.T37 EC1118 D8.T37 EC1118

0

2

4

6

8

0 6 12 17

Wei

ght l

oss

(g)

Time (day)

Control EC1118 D2.T8 EC1118 D8.T8 EC1118

0

2

4

6

8

0 6 12 17

Wei

ght l

oss

(g)

Time (day)

Control NT50 D2.T37 NT50 D8.T37 NT50

0

2

4

6

8

0 6 12 17

Wei

ght l

oss

(g)

Time (day)

Control NT50 D2.T8 NT50 D8.T8 NT50

0

2

4

6

8

0 6 12 17

Wei

ght l

oss

(g)

Time (day)

Control VIN13 D2.T37 VIN13 D8.T37 VIN13

0

2

4

6

8

0 6 12 17

Wei

ght l

oss

(g)

Time (day)


a

fe

dc

b

g h


68

0

2

4

6

8

0 6 12 17

Wei

ght l

oss

(g)

Time (day)


0

2

4

6

8

0 6 12 17

Wei

ght l

oss

(g)

Time (day)


Figure2Fermentationratesoftemperaturestressedfermentations:(a)285control,D2.T37,D8.T37;(b)285 control, D2.T8, D8.T8; (c) EC1118 control, D2.T37, D8.T37; (d) EC1118, D2.T8, D8.T8; (e) NT50control, D2.T37,D8.T37; (f)NT50 control, D2.T8, D8.T8 (g) VIN13 control, D2.T37,D8.T37; (h) VIN13control; (i) VIN7 control, D2.T37, D8.T37; (j) VIN7, D2.T8, D8.T8. Results are the average of triplicatefermentations;(errorbarsarenotshownfortheclarityofthegraphs)

Figure3Residualglucoseandfructoseforeachstrainwascompareditscontrol.Thegraphdepictsthestrainsandtreatmentswhichdifferedsignificantly(p<0.05)fromtheircontrols,asviewedbyCytoscape.Abluenode(ellipse)indicatesareductioncomparedtothecontrolandarednodeanincreasethecolourintensitydenotesthemagnitudeofthefoldchangeobserved

i j


69

Figure4Glycerollevelsforstresstreatmentssignificantly(p<0.05)differentfromitsunstressedcontrol.Abluenode(ellipse)indicatesareductioncomparedtothecontrolandarednodeanincreasethecolourintensitydenotesthemagnitudeofthefoldchangeobserved

4.3.2 Aromacompounds

Principalcomponentanalysis(PCA)wasperformedonthearomadatasetinitsentirety,using

TheUnscramblersoftware(version9.2,CAMOASA,Norway).PCAanalysiswasusedtoevaluate

whether the changes could be attributed to a specific stress treatment, and whether the

observedchangeswereconservedamongthestrains(Fig.5).Individualcompoundsseparated

anddominatedthemultivariatespace,andstronglyinfluencedthemodelwhencomparingthe

aromaprofilesofeachstrainwithinatreatment,aswellaswhencomparingthearomaprofile

foronestrainacrossall treatments(datanotshown). Ingeneral, these influentialcompounds

could be tentatively associatedwith individual strains: VIN7 associatedwith isoamyl alcohol,

andpropanol;285withisobutanol;EC1118withdecanoicacidandethyllactate;andNT50with

1‐octenol.Themultivariatedataanalysisdid thereforenotallow for the comparisonbetween

individual stress treatments and corresponding controls in this study, as the model was

dominatedbyindividualaromacompounds.


70

Figure5(a)PCAscoresplotshowspoordifferentiationbetweencolourcodedtreatmentsandnumberedstrains (285 = 1; EC1118 = 2; NT50 = 3; VIN7 = 4; and VIN13 = 5) using the GC – FID data. (b)Corresponding loadings for the PCA indicates which variables contribute significantly to thedifferentiationbetweentreatmentsandnumberedstrains

Cytoscape was used to visualise pair‐wise comparisons between the stress and control

treatments.Thedatashowedthattheinfluenceofexposuretolowtemperatureonthevolatile

aromaprofileistimeandstraindependant(Fig.6),generallyshowinganincreaseincompound

levelsondaytwoandadecreaseondayeight.Theonlyconservedchangewasthedecreasein

the levels of octanoic acid producedbyEC1118, irrespective ofwhen the stresswas applied.

Exposure to low temperatureonday2, increased thevolatile fatty acidproductionbyVIN13

(isovaleric,isobutyric,butyricandpropionicacid),andVIN7(isobutyricacid).

Control S1 S2 D2.T8 D2.T37 D8.T8 D8.T37

(a) (b)


71

Figure6 The significant changes (p<0.05) in the levels of the volatile aroma compounds due to low temperature stress, day eight at 8C or day two at 8C, whencomparingthelowtemperaturestressedfermentationstotheircontrols.Ablueedge(line)indicatesareductioncomparedtothecontrolandarededge(line)anincreasethecolourintensitydenotesthemagnitudeofthefoldchangeobserved


72

Figure7 The significant changes (p<0.05) in the levels of volatile aroma compounds due to high temperature stress, day eight at 37C and day two at 37Cwhencomparingthehightemperaturestressedfermentationstotheircontrols.Ablueedge(line)indicatesareductioncomparedtothecontrolandarededgeanincreasethecolourintensitydenotesthemagnitudeofthefoldchangeobserved


73

Figure 8 The significant changes (p<0.05) in the levels of the volatile aroma compounds due to hyperosmotic stress, treatments S1 (0.3 M) and S2 (0.5 M), whencomparingthehyperosmoticstresstreatmentstotheircontrols.Ablueedge(line)indicatesareductioncomparedtothecontrolandarededgeanincreasethecolourintensitydenotesthemagnitudeofthefoldchangeobserved


74

Ofthestressesevaluated,exposureto37°Cresultedinthegreatestnumberofchangestothe

volatilearomaprofile(Fig.7).Whenthestresswasappliedonday8,amixtureofincreasesand

decreaseswasobservedforeachcompoundgroup.Conversely,exposureondaytworesultedin

ageneraldecreaseinthelevelsofthevolatilearomacompounds.Afewconservedchangeswere

noted in response to high temperature. Ethyl caprate levels were reduced for all evaluated

strainsbut285.OctanoicaciddecreasedforEC1118andVIN7,andhexanoicaciddecreasedfor

EC1118, VIN7 and 285. Furthermore, a few compounds displayed a same response (relative

increaseordecrease)to37Cirrespectiveofwhenthestresswasapplied,namely;hexanoicacid

(VIN7); octanoic acid (VIN7); ethyl‐phenylacetate (VIN7); 2‐phenylethyl acetate (NT50 and

285);andethylcaprate(NT50).

Thehigher levelofhyperosmotic stress (S2)broughtabouta largernumberof changes in

thevolatilearomacompositioncomparedtothelowerlevelofhyperosmoticstress(S1)(Fig.8).

The S1 treatment resulted in inconsistent (increase anddecrease) changes in the compounds

groups (three acids, two esters, two higher alcohols). There appears to be no discernable

pattern, as the relative changes usually occurred for a different strain each time. Even the

compoundswhichwere influencedbybothhyperosmotic stress levelswere in each case as a

consequenceoffermentationsbydifferentstrains.

4.4 Discussion

Theimpactofenvironmentalstressonyeastfermentationperformancehasbeenwidelystudied,

but to our knowledge this is the first study to assess the impact of hyperosmotic and

temperature stress on the volatile aroma profile produced by yeast during alcoholic

fermentation.

The residual sugars measured at the end of fermentation served as an indicator of

fermentationperformance(Fig.3),andmostofthestresstreatmentsappliedheredidnotresult

in significant changes when compared to the unstressed controls. The only exceptions were

VIN7which did not adaptwell to high temperature andhyperosmotic stress (S2), andVIN13

following exposure to 8C on day 8. Other studies have reported on the inability of VIN7 to

complete fermentationswith high initial sugar concentrations (Erasmus et al. 2004, Erasmus

andvanVuuren2009).

Glycerol is produced as a compatible soluteproduced in response tohyperosmotic stress,

wherebyitincreasesthecellsinternalsolutecontentpromotingwateruptake.Thereforeitisnot

surprisingthatallstrainsdisplayedarelativeincreaseintheirglycerollevelsproportionaltothe

hyperosmoticstressapplied.


75

Other investigations into the impact of heat shock on glycerol production reported an

increaseinglycerollevelsinresponsetotemperatureshock,whentemperaturewasincreased

from18Cto34C(Berovicetal.2007,Kukecetal.2003).Berovičetal.(2007)reportedthatthe

longer thedurationof the temperatureshock(24hourscompared to4hours) thegreater the

amount of glycerol produced. The only temperature treatment to increase its glycerol

concentrationinourstudywasVIN7_D2.T37.However,Berovičetal.(2007)appliedheatshock

within the first30hoursof fermentation, inourstudythestresswasonlyappliedondaytwo

anddayeightofthefermentation.Itispossiblethattimingofthestressapplicationplayedarole

in glycerol production in the case of the other strains, and that theywould have beenmore

responsivehadthestressbeenappliedearlier.

Thechangesinaromacompoundsafterexposureoffermentationstolowtemperature(8°C)

are time and strain dependant (Fig. 6), as illustrated by an increase in the levels of four

compounds(volatilefattyacids)forVIN13ondaytwo,comparedtooneester(ethylacetate)on

dayeight.Longchainfattyacids(C16andC18)arevitalcomponentsoftheplasmamembrane,

wheretheyaidtransportofvariouscompoundsacrosstheplasmamembrane(Lambrechtsand

Pretorius2000).Inresponsetofermentationatlowtemperaturesyeaststrainsarereportedto

modulate membrane fluidity by increasing the content of unsaturated fatty acids, or the

productionofmediumchain fatty acids (Beales2004,Beltran et al. 2008).Thisproductionof

mediumchainfattyacidsmayaccountfortheincreaseintheproductionofvolatilefattyacidby

VIN13.

Theonlyconservedchangeinresponseto8°Cwasthedecreaseinthelevelsofoctanoicacid

producedbyEC1118,irrespectiveofwhenthestresswasapplied.Anintracellularaccumulation

of medium chain fatty acids may account for the observed decrease seen in octanoic and

decanoic acid (day 8) for EC1118. At low temperatures the cell membrane becomes less

permeable(Bisson1999),furtherhinderingthepassivediffusionofhydrophobicmediumchain

fatty(C8andC10)acidsintothemedium.

Thefermentationsstressedat37°Cdisplayedthegreatestnumberofchangestothevolatile

aromaprofiles (Fig. 7).Whenappliedonday eight, amixtureof increases anddecreaseswas

observed for each compound group. Conversely, exposure on day two resulted in a general

decreaseinthelevelsofthevolatilearomacompounds.

Yeast cells respond to stress in a gradedmanner (Gasch et al. 2000). Similarly, the lower

levelofhyperosmoticstresscondition(S1)inducedfewersignificantchangescomparedtothat

of the higher hyperosmotic stress condition (S2) (Fig. 8). It has been suggested that redox

balancingplaysan important role in theregulationof themetabolicpathwaysresponsible for

the production of aroma compounds (Lambrechts and Pretorius 2000). When experiencing

osmotic stress, cells respond by accumulating of glycerol,which results in a surplus ofNAD+.

This redox imbalance is corrected by the oxidation of acetaldehyde to acetic acid with


76

subsequent regeneration of NADH. This is illustrated by the increase in volatile acidity in

responsetoincreasinglevelsofosmoticstress.

Inrecentstudies,Jainetal.(2011)andStygeretal.(2011)reportedthatanexcessofNADH

promotestheproductionofhigheralcoholsbyalabyeaststrain,inanattempttoreduceNADH

toNAD+.Also,thisredoximbalanceduetoasurplusofNADHhinderstheproductionofesters

which requires NAD+. If the reverse were true, the excess NAD+ produced due to glycerol

formationshouldresultinadecreaseintheformationofhigheralcoholsandanincreaseinester

production. This is generally not the case for this study, as esters and higher alcohols both

increasedanddecreasedsubsequenttoosmoticstress.Thissuggeststhatredoxbalancingonly

influences the production of aroma compounds to a limited extent under the fermentation

conditionsevaluated.

Literature also suggests that the increase in volatile fatty acids would display a

correspondingincreaseinthelevelsofesters(BissonandKarpel2010);howevernosuchtrend

wasobservedinthisstudy.

Exposure to environmental stress caused a relatively small number of specific changes to

individualcompounds;however,asaromacompoundsinteractwitheachother,theymaycause

significantchanges in theoverallaromaprofileof theresultantwine.This is the firststudy to

investigate the influence of environmental stress on the volatile aroma profile in a synthetic

grapemust.The resultsdemonstrate that theevaluatedstresses causedsignificant changes in

thelevelsofanumberofaromacompounds.Thechangesobservedalsodifferedforeachofthe

strains,andthefermentationconditionstested.Furtheranalysisontheimpactofenvironmental

stressduringgrapemustfermentationandthedurationofthestressapplicationwouldprovide

information regarding the extent towhich environmental stress impacts the aroma profile of

wine.

4.5 Acknowledgements

We are indebted to Candice Stilwaney for technical assistance, and Dan Jacobson for the

statisticaldataanalysis.


77

4.6 References

AlexandreH, Charpentier C (1998)Biochemical aspects of stuck and sluggish fermentation ingrapemust.JIndMicrobiolBiotechnol1:20‐27

AttfieldPV(1997)Stresstolerance:thekeytoeffectivestrainsofindustrialbaker'syeast.NatureBiotechnology13:1351‐1357

BauerF,Pretorius I (2000)Yeast stress responseand fermentationefficiency:how to survivethemakingofwine‐areview.SAfrJEnolVitic27‐51

BealesN(2004)Adaptationofmicroorganismstocoldtemperatures,weakacidpreservatives,lowpH,andosmoticstress:areview.ComprRevFoodSciFoodSaf1:1‐20

Bell SJ, Henschke PA (2005) Implications of nitrogen nutrition for grapes, fermentation andwine.AustJGrapeWineRes3:242‐295

BeltranG,NovoM,GuillamónJM,MasA,RozèsN(2008)Effectoffermentationtemperatureandculture media on the yeast lipid composition and wine volatile compounds. Int J FoodMicrobiol2:169‐177

BelyM,SablayrollesJM,BarreP(1990)Automaticdetectionofassimilablenitrogendeficienciesduringalcoholicfermentationinoenologicalconditions.JFermentBioeng4:246‐252

BerovicM,PivecA,KosmerlT,WondraM,CelanS (2007) Influenceofheat shockonglycerolproductioninalcoholfermentation.JBiosciBioeng2:135‐139

BissonLF(1999)Stuckandsluggishfermentations.AmJEnolVitic1:107‐119

BissonLF,KarpelJE(2010)Geneticsofyeastimpactingwinequality.FoodSciTechnol139‐162

Bradbury JE, Richards KD, Niederer HA, Lee SA, Rod Dunbar P, Gardner RC (2006) Ahomozygousdiploidsubsetofcommercialwineyeaststrains.AntonievanLeeuwenhoek1:27‐37

CarrascoP,QuerolA,delOlmoM(2001)Analysisof thestressresistanceofcommercialwineyeaststrains.ArchMicrobiol6:450‐457

ErasmusDJ,CliffM, vanVuurenHJJ (2004) Impactof yeast strainon theproductionof aceticacid,glycerol,andthesensoryattributesoficewine.AmJEnolVitic4:371‐378

ErasmusDJ,vanVuurenHJJ(2009)GeneticBasisforOsmosensitivityandGeneticInstabilityoftheWineYeastSaccharomycescerevisiaeVIN7.AmJEnolVitic2:145‐154

FrancisI,NewtonJ(2005)Determiningwinearomafromcompositionaldata.AustJGrapeWineRes2:114‐126

GaschAP,SpellmanPT,KaoCM,Carmel‐HarelO,EisenMB,StorzG,BotsteinD,BrownPO(2000)Genomic expression programs in the response of yeast cells to environmental changes.MolBiolCell12:4241‐4257

Gasch AP, Werner‐Washburne M (2002) The genomics of yeast responses to environmentalstressandstarvation.FunctIntegrGenomics181‐192

GibsonBR,LawrenceSJ,Leclaire JPR,PowellCD,SmartKA(2007)Yeastresponsestostressesassociatedwithindustrialbreweryhandling.FEMSMicrobiolRev5:535‐569


78

HazelwoodLA,DaranJM,vanMarisAJA,PronkJT,DickinsonJR(2008)TheEhrlichpathwayforfusel alcoholproduction:a centuryof researchonSaccharomycescerevisiaemetabolism.ApplEnvironMicrobiol8:2259‐2266

HenschkeP,JiranekV(1993)Yeasts–metabolismofnitrogencompounds.In:FleetGH(ed)WineMicrobiolBiotechnol,HarwoodAcademicPublishers,Switzerland,pp77‐164

HohmannS(1997)Shapingup:Theresponseofyeasttoosmoticstress.In:HohmannS,MagerWH(ed)YeastStressResponses,Springer,NewYork,pp101‐145

IvorraC,PérezOrtínJE,delOlmoM(1999)Aninversecorrelationbetweenstressresistanceandstuckfermentationsinwineyeasts.Amolecularstudy.BiotechnolBioeng6:698‐708

JainVK,DivolB,PriorB,BauerFF(2011)EffectofalternativeNAD+‐regeneratingpathwaysontheformationofprimartyandsecondaryaromacompoundsinaSaccharomycescerevisiaeglycerol‐defectivemutant.ApplEnvironBiotechnolDOI10.1007/s00253‐011‐3431‐z

KukecA,BerovicM,WondraM,CelanS,KosmerlT(2003)Influenceoftemperatureshockontheglycerolproductionincv.Sauvignonblancfermentation.Vitis4:205‐206

LambrechtsM,PretoriusI(2000)Yeastanditsimportancetowinearoma‐areview.SAfrJEnolVitic97‐129

LouwL,TredouxA,VanRensburgP,KiddM,NaesT,NieuwoudtH(2010)Fermentation‐derivedaromacompoundsinvarietalyoungwinesfromSouthAfrica.SAfrJEnolVitic2:213‐225

MagerWH, SideriusM (2002)Novel insights into the osmotic stress response of yeast. FEMSYeastRes3:251‐257

Malherbe S (2011) Investigation of the impact of commercialmalolactic fermentation startercultures on red wine aroma compounds, sensory properties and consumer preference.Dissertation,StellenboschUniversity

MalherbeS,BauerF,DuToitM(2007)Understandingproblemfermentations–Areview.SAfrJEnolVitic2:169‐185

RappA,ManderyH(1986)Winearoma.CelMolLifeSci8:873‐884

RappA,VersiniG (1996) Influenceofnitrogencompounds ingrapesonaromacompoundsofwines.Wein‐Wiss3‐4:193‐203

Rossouw D, Næs T, Bauer F (2008) Linking gene regulation and the exo‐metabolome: Acomparativetranscriptomicsapproachtoidentifygenesthatimpactontheproductionofvolatilearomacompoundsinyeast.BMCGenomics1:530‐547

Saerens S, Delvaux F, Verstrepen K, Van Dijck P, Thevelein J, Delvaux F (2008) Parametersaffecting ethyl ester production by Saccharomyces cerevisiae during fermentation. ApplEnvironMicrobiol2:454‐461

SolesR,OughC,KunkeeR(1982)Esterconcentrationdifferencesinwinefermentedbyvariousspeciesandstrainsofyeasts.AmJEnolVitic2:94‐98

StygerG,JacobsonD,BauerFF(2011)Identifyinggenesthatimpactonaromaprofilesproducedby Saccharomyces cerevisiae and the production of higher alcohols. Appl EnvironBiotechnol713‐730


79

Swiegers J, Bartowsky EJ, Henschke P, Pretorius IS (2005) Yeast and bacterialmodulation ofwinearomaandflavour.AustJGrapeWineRes2:139‐173

Tamás MJ, Hohmann S (2003) The osmotic stress response of Saccharomyces cerevisiae. In:HohmannS,MagerPWH(ed)YeastStressResponses,Springer‐VerlagBerlin,Heidelberg,pp121‐200

Vilanova M, Ugliano M, Varela C, Siebert T, Pretorius IS, Henschke PA (2007) Assimilablenitrogenutilisationandproductionofvolatileandnon‐volatilecompounds inchemicallydefinedmediumbySaccharomycescerevisiaewineyeasts.ApplEnvironBiotechnol1:145‐157

Zuzuarregui A, del OlmoM (2004) Analyses of stress resistance under laboratory conditionsconstituteasuitablecriterionforwineyeastselection.AntonievanLeeuwenhoek4:271‐280


80

Chapter5

Generaldiscussionandconclusions


81

Chapter5

Generaldiscussionandconclusions

5.1 Introduction

Duringthealcoholicfermentationofgrapemust,yeastsaresubjecttonumerousenvironmental

stresses(hyperosmoticstress,temperaturechanges,nutrientstarvation,pH,etc.).Yeaststrains

unabletorapidlyperceiveandrespondtostressaremorelikelytobeinvolvedinfermentations

thatproceedataslowrate(sluggishfermentation)orresultinhighresidualsugarlevels(stuck

fermentation) (Alexandre&Charpentier,1998;Bisson,1999;Gibsonetal.,2007;Malherbeet

al.,2007).

In chapter 3 the impact of hyperosmotic stress (imparted by the initial sugar

concentration), nitrogen content and low temperature on the fermentation performance of

commercially available yeast strainswas investigated. The results show that the relationship

between initial nitrogen supplementation and fermentation performance is not linear, and is

significantly affected by initial sugar content, fermentation temperature, aswell as the yeast

strainused.Ourresultsshowtheimportanceofinvestigatingnotonlyindividualfermentation

parametersaffectingfermentationperformance,butalsotheinteractiveeffecttheymayhaveon

eachother.Nitrogendeficienciesarethemostcommoncauseofproblematicfermentations.Of

thenitrogen levelstested,250mg/Lresulted inan improvedoverallperformance intermsof

fermentation onset, maximal fermentation rate, total weight loss, sugar consumption, and

biomassproduction.Conversely,fermentationssupplementedwith400mg/Lnitrogenshowed

no comparative improvement and in some cases fermentation performance and sugar

consumption was reduced relative to the 250 mg/L nitrogen treatment. This illustrates the

danger of excessive nitrogen supplementation in terms of fermentation performance; and

furtheradvocatesdeterminingthemustcharacteristicspriortoinoculation.

The effect of stress on the fermentationperformance of yeast strains has been the

focus of much study, but these studies do not address whether stress impacts wine quality;

whichisthebasisofconsumerliking.Winequalityisgenerallyasaconsequenceofperceived

flavour, being the sum of smell, taste and mouth‐feel. Wine yeast strains produce volatile

aromas(higheralcohols,estersandvolatilefattyacids)duringfermentation.Theproductionof

volatilearomacompoundsisaffectedbytheavailabilityofprecursors,fermentationconditions

(Henschke&Jiranek,1993;Rapp&Versini,1996;Lambrechts&Pretorius,2000;Swiegersetal.,


82

2005; Vilanova et al., 2007; Saerens et al., 2008; Bisson & Karpel, 2010) and the individual

strainsyeaststrainsused(Solesetal.,1982;Rossouwetal.,2008).

Generally, wine flavour is a significant component of consumer liking. Consumer

likingissubjectiveandplaysamajorrolewhenmakingthedecisiontopurchasewine.Toour

knowledge, this is the first study to investigate the impact of hyperosmotic and temperature

stresses on fermentation performance and the production of fermentation derived volatile

aromacompounds.Thevolatilearomacompoundswerequantifiedusingagaschromatograph

andaflameionizationdetector(GC‐FID).Theresultsinchapter4showthatexposuretostress

has thepotential to significantly change thewine aromaprofile. Somewhatunexpectedly, the

observedchangesweredifferentforallthestrainsandstressesevaluated.Nevertheless,amore

in‐depthanalysisofvarious fermentationstresses, concurrentlyandsequentially,may lead to

specific guidelines of how individual yeast strains may perform on specific musts and

fermentation conditions. Itmay also providewinemakerswith tools to influence the aroma

profileofawineinaspecificmanner.

Thedatainchapter3couldbeadaptedtoprovidewinemakerswithanindicationof

whatthefermentationcapacityofthespecificstrainswouldbeifasetoffermentationandmust

stressconditions isexpected.However, theperformanceofyeaststrains fermentingsynthetic

grapemustmay varywhen compared to “real” grapemust. Nonetheless, the selection of the

appropriateyeaststrainandthenitrogensupplementationstrategymayreducetheincidenceof

sloworincompletefermentations.Futurestudiesshouldfurtherinvestigatetheuseofsuitable

strain selection as a potential strategy to avoid problematic fermentations in grapemust, in

additiontocharacterisingtherelationshipbetweenenvironmentalstressandthevolatilearoma

profile.


83

5.2 Literaturecited





Henschke,P.&Jiranek,V.,1993.Yeasts–metabolismofnitrogencompounds.In:Fleet,G.H.(ed).WineMicrobiol.Biotechnol.HarwoodAcademicPublishers,Switzerland.pp.77‐164.




Rossouw, D., Næs, T. & Bauer, F., 2008. Linking gene regulation and the exo‐metabolome: Acomparative transcriptomics approach to identify genes that impact on the production ofvolatilearomacompoundsinyeast.BMCGenomics.9,530‐547.

Saerens,S.,Delvaux,F.,Verstrepen,K.,VanDijck,P.,Thevelein,J.&Delvaux,F.,2008.ParametersaffectingethylesterproductionbySaccharomycescerevisiaeduringfermentation.Appl.Environ.Microbiol.74,454‐461.

Soles, R., Ough, C. & Kunkee, R., 1982. Ester concentration differences inwine fermented byvariousspeciesandstrainsofyeasts.Am.J.Enol.Vitic.33,94‐98.


Vilanova,M.,Ugliano,M.,Varela,C.,Siebert,T.,Pretorius,I.S.&Henschke,P.A.,2007.Assimilablenitrogen utilisation and production of volatile and non‐volatile compounds in chemicallydefinedmediumbySaccharomyces cerevisiaewineyeasts.Appl.Environ.Biotechnol. 77,145‐157.


Documents

Stress, fermentation performance and aroma production by yeast