Use of non-Saccharomyces wine yeast as novel sources of mannoproteins in wine

lable at ScienceDirect

Food Microbiology 43 (2014) 5e15

Contents lists avai

Food Microbiology

journal homepage: www.elsevier .com/locate/ fm

Use of non-Saccharomyces wine yeasts as novel sources ofmannoproteins in wine

P. Domizio a,b,*, Y. Liu c, L.F. Bisson a, D. Barile c,d,**

aDepartment of Viticulture & Enology, University of California-Davis, Davis, CA 95616, USAbDipartimento di Gestione dei Sistemi Agrari, Alimentari e Forestali (GESAAF), Università degli Studi di Firenze, 50144 Firenze, ItalycDepartment of Foods Science & Technology, University of California-Davis, Davis, CA 95616, USAd Foods for Health Institute, University of California-Davis, Davis, CA 95616, USA

a r t i c l e i n f o

Article history:Received 12 February 2014Received in revised form22 March 2014Accepted 14 April 2014Available online 30 April 2014

Keywords:WineNon-SaccharomycesYeastMannoproteinPolysaccharideN-glycanMALDI-TOF

* Corresponding author. Permanent address: Dipartemi Agrari, Alimentari e Forestali (GESAAF), UniversDonizetti 6, 50144 Firenze, Italy. Tel.: þ39055 275550** Corresponding author. Department of Food Sciencof California-Davis, One Shields Ave, Davis, CA 95616

E-mail addresses: [email protected] (P. Domizio), db

http://dx.doi.org/10.1016/j.fm.2014.04.0050740-0020/� 2014 Elsevier Ltd. All rights reserved.

a b s t r a c t

Eight non-Saccharomyces wine strains, previously selected for their ability to modulate the final con-centrations of various volatile compounds and to persist with Saccharomyces cerevisiae in mixed inoculafermentations of grape juice, have been analyzed in the present work to test their ability to releasemannoproteins. The eight strains were members of different genera originally isolated from grape:Hansensiaspora osmophila, Lachancea thermotolerans, Metschnikowia pulcherrima, Pichia fermentans,Saccharomycodes ludwigii, Starmerella bacillaris, Torulaspora delbrueckii and Zygosaccharomyces flor-entinus. A synthetic polysaccharide-free grape juice, was used to characterize the mannoproteinsreleased during the alcoholic fermentation. Mannoproteins profiles were characterized by gel electro-phoresis and carbohydrate composition was analyzed both by HPLC and by mass spectrometry. The eightnon-Saccharomyces yeasts demonstrated a higher capacity to release polysaccharides compared toS. cerevisiae. The proteins released by the eight yeast strains showed a wide variety of protein sizes,ranging from 25 kDa to greater than 250 kDa. The mass spectrometric profile of the N-glycans rangedfrom 1600 to 4000 Da and was characteristic for each strain. Detailed investigation of the degree ofpolymerization of released N-glycans revealed variable composition from 8 to 15 units ofmonosaccharides.

� 2014 Elsevier Ltd. All rights reserved.

1. Introduction

Mannoproteins are released from the yeast cell wall during thealcoholic fermentation and wine aging processes (Boivin et al.,1998; Charpentier and Feuillat, 1993; Charpentier et al., 2004a;Llaubères et al., 1987) and represent one of the major poly-saccharides found in wine. The mannoproteins have been recog-nized to have many positive enological properties such asimproving mouth-feel and fullness (Vidal et al., 2004), decreasingastringency (Carvalho et al., 2006), adding complexity and aromaticpersistence (Chalier et al., 2007), increasing sweetness androundness (Guadalupe et al., 2007), and reducing protein and

timento di Gestione dei Sis-ità degli Studi di Firenze, Via2.e and Technology, University, USA. Tel.: þ1 530 752 [email protected] (D. Barile).

tartrate instability (Gonzalez-Ramos et al., 2008). Moreover, man-noproteins can stimulate malolactic fermentation by lactic acidbacteria (Guilloux-Benatier et al., 1995; Rosi et al., 2000), adsorbsome toxic compounds possibly present in the wine such asochratoxin A (Caridi, 2007; Moruno et al., 2005), and improve thefoam quality of sparkling wines (Moreno-Arribas et al., 2000;Nunez et al., 2006).

In order to increase the amount of mannoproteins releasedduring the alcoholic fermentation, some researchers have devel-oped autolytic thermosensitivemutants (Giovani and Rosi, 2007) orgenetically engineered wine yeast strains of Saccharomyces cer-evisiae (Brown et al., 2007; Gonzalez-Ramos and Gonzalez, 2006;Gonzalez-Ramos et al., 2008). However, these organisms are likelyto be viewed as releasing components not normally present inwine.

Recently, some authors (Comitini et al., 2011; Domizio et al.,2010, 2011; Gobbi et al., 2013; Romani et al., 2010) have shownthat other non-Saccharomyceswine yeasts found in grape and winemaking environments have a high capacity to release important

Table 1Origin of the eight non-Saccharomyces strains and the one commercial strain ofS. cerevisiae used in this study.

Strain Species Origin

# EC1118 Saccharomyces cerevisiae Lallemanda

# 4 Pichia fermentans SAIFETb

# 22Starmerella bacillaris(formerly Candida zemplinina)

GESAAFc

# 32 Hanseniaspora osmophila GESAAFc

# 42 Zygosaccharomyces florentinus GESAAFc

# 46 Metschnikowia pulcherrima SAIFETb

# 64 Saccharomycodes ludwigii GESAAFc

# 92 Torulaspora delbrueckii GESAAFc

# 101 Lachancea thermotolerans SAIFETb

a Lallemand Inc. (Montreal, Canada).b Dipartimento SAIFET, Sezione di Microbiologia Alimentare, Industriale ed

Ambientale, Universitá Politecnica delle Marche.c Dipartimento di Gestione dei Sistemi Agrari, Alimentari e Forestali, Università

degli Studi di Firenze.

P. Domizio et al. / Food Microbiology 43 (2014) 5e156

polysaccharides (including mannoproteins) into wine during alco-holic fermentation and these polysaccharides and mannoproteinswould be naturally present in wine.

The possibility to increase the content of mannoproteinsnaturally by the use of non-Saccharomyces yeasts could representan added value to the already heightened interest towards thesewine yeasts for flavor modification (Ciani et al., 2010). Indeed,there are many reports indicating that the use of non-Saccharo-myces yeasts leads to a more complex aroma and an improvedwine quality (Anfang et al., 2009; Domizio et al., 2007; Jolly et al.,2003, 2006; Renouf et al., 2007; Rojas et al., 2001; Swiegers et al.,2005). The use of non-Saccharomyces wine yeasts in associationwith Saccharomyces strains has been suggested to also enhance theglycerol content (Ciani and Ferraro, 1996, 1998; Soden et al., 2000),deacidify the grape juice or wine (Ciani, 1995; Magyar and Panyik,1989), reduce the acetic acid content (Bely et al., 2008; Rantsiouet al., 2012), and enhance the total acidity of wines (Gobbi et al.,2013; Kapsopoulou et al., 2007; Mora et al., 1990). The presenceon the market of new yeast formulas of mixed starter culturescontaining both Saccharomyces and non-Saccharomyces strains isevidence of the strong commercial interest in mixedfermentations.

Non-Saccharomyces yeasts could influence not only the poly-saccharide concentration in the wine, as previously reported, butcould also influence the final chemical composition of these mac-romolecules, leading to functional differences in the resultant wine.Previous analyses of mannoproteins released in wine during alco-holic fermentation have been conducted mainly using strainsbelonging to Saccharomyces (Escot et al., 2001; Llaubères et al.,1987). The resulting macromolecule composition was found to besimilar to that of the yeast cell wall, with a molecular mass between50 and 500 kDa (Villetaz et al., 1980).

The Saccharomyces mannoproteins are found in the outmostlayer of yeast cell wall (Fleet, 1991; Klis et al., 2006), and consist of85e90% carbohydrates, mainly mannose, and 10e15% proteins. Therelease of these macromolecules during the growth of yeast cells(Guilloux-Benatier et al., 1995; Llaubères et al., 1987) is due to acontrolled hydrolysis of the mother cell wall to allow bud emer-gence (Charpentier et al., 1986; Fleet, 1991).

The glycans are attached to the protein via an asparagine res-idue; these N-linked glycans represent one form of glycosylationpreviously described in yeasts (Ballou, 1976). Previous works haveshown that S. cerevisiae N-linked glycans are composed of a core ofMan8e14GlcNAc2, (Mannose, N-acetyl glucosamine) and a highlybranched outer chain (50e200 mannose units) of a-linkedmannose and mannose-phosphate residues (Ballou, 1990; Trimbleand Atkinson, 1992).

Glycans and glyco-conjugates are an interesting and diverseclass of molecules with many important bioactive functions;however their inherent structural complexity and diversity rendersthem challenging to study in a comprehensivemanner and requiresthe use of intensive purification prior to analytical characterization.

The recent use of accurate mass spectrometry techniques suchas the soft ionization method known as MALDI (Matrix-assistedlaser desorption ionization) as well as the ESI (electrospray ioni-zation) method, in combination with microchip-based nano-liquidchromatographic separation, have allowed isolation and charac-terization of new oligosaccharides in both red and white wine(Bordiga et al., 2012; Ducasse et al., 2010).

In the present study we evaluated eight non-Saccharomycesstrains, selected because of desirable enological attributes, for theirability to release mannoproteins into a synthetic grape juice. Wereport the chemical characterization of these components in termsof carbohydrate composition analyzed by HPLC, protein profile bygel electrophoresis as well as detailed investigation of the degree of

polymerization of released N-glycans (8e15 units) by massspectrometry.

2. Materials and methods

2.1. Yeast strains

Eight non-Saccharomyces strains from the yeast culture collec-tion of the Department of Agricultural, Food and Forestry Systems(GESAAF, University of Florence, Italy) and of the Department of thePolytechnic University of Marche SAIFET (Ancona, Italy) were used.These yeast strains, all belonging to different genera (Table 1), havebeen isolated from grape and must of different origins, and alreadywere selected for their enological attributes in mixed fermentationin grape juice at the laboratory scale (Comitini et al., 2011; Domizioet al., 2011). A commercial strain, Lalvin EC1118 (Lallemand Inc.,Montreal, Canada), was used as reference strain for Saccharomycesand for comparison determination.

2.2. Fermentation trials

The fermentations were carried out in duplicate at 28 �C in200 mL Erlenmeyer flasks containing 150 mL of a synthetic grapejuice medium “Minimal Must Medium” (MMM) (Spiropoulos et al.,2000). The medium was sterilized by filtration. The flasks wereinoculated at optical density of 0.1 (OD600 nm), with 48-h pre-cultures grown in 10 mL of YPD medium (10 g/L yeast extract,20 g/L peptone, 20 g/L glucose) (Oxoid Unipath Ltd, Hampshire,UK), at 25 �C in a roller drum.

The levels of sugar and total assimilable nitrogen were 220 g/Land 208 mg/L, respectively. The assimilable nitrogen concentrationwas obtained by using 0.2 g/L of L-arginine and 0.5 g/L of ammo-nium phosphate. The flasks, continuously agitated at 150 rpm, werestopperedwith silicone vent bungs (Ferm-Rite), allowing the CO2 to“burp” out and escape. The flasks were weighed every day until theend of fermentation (as a constant weight for two consecutivedays) to monitor the fermentation kinetics.

2.3. Analysis

2.3.1. Biomass determinationSamples were taken from each flask during the alcoholic

fermentation to monitor the growth kinetics by OD600 nm. The cellviability, expressed as number of colony forming units (cfu), ofS. cerevisiae and non-Saccharomyces yeast strains, was measured atthe beginning and at the end of the alcoholic fermentation by

P. Domizio et al. / Food Microbiology 43 (2014) 5e15 7

plating 100 mL aliquots of serial dilutions of each sample onto theYPD medium. At the end of the fermentation, the dry weightbiomass (dw) was determined gravimetrically by filtering 10 ml ofsample from each culture on pre-weighed 0$45 mm nitrocellulosemembranes. Membranes were washed twice with distilled waterand placed in oven at 100 �C for 24 h before being weighed again,and the difference in starting and final weight was used to calculatebiomass.

2.3.2. Analytical determinations of the fermentation productsEthanol and residual sugars were determined by high perfor-

mance liquid chromatography (HPLC) using an Agilent 1100 seriesHPLC system (Agilent, Palo Alto, CA, USA) coupled with a refractiveindex detector (Hewlett Packard HP-1047A). After filtrationthrough 0.45 mm nitrocellulose membranes and after appropriatedilution in water, 20 mL of each sample was injected into the HPLCapparatus. Isocratic separation was performed at 65 �C on a(300 � 7.7 mm) Hi-Plex H column (Agilent, Palo Alto, CA, USA). Themobile phase was 5 mM H2SO4 at a flow rate of 0.6 mL/min.Quantification was performed in comparison with an externalcalibration curve with concentrations of each compound rangingfrom 0.1 g/L to 20 g/L.

The total polysaccharides were determined using the sameHPLC apparatus, according to the method of Peyron et al. (1993),with some modifications. Separation of the polysaccharides wasperformed at 65 �C, using a TSKGEL G-Oligo-PW (Supelco 808031)column, and a TSK-GEL Oligo (Supelco 808034) guard column withisocratic elution (0.2 M NaCl; 0.8 mL/min). Quantification wasperformed by comparison with an external calibration curve ofmannan (SigmaeAldrich), at concentrations from 50 mg/L to1500 mg/L.

The areas of the peaks of interest were integrated using theChemStation Data Analysis System, version A.10.2 (1757) (Agilent,Palo Alto, CA, USA). All the analyses were carried out in duplicate.

2.3.3. Polysaccharide purificationAfter 13 days of alcoholic fermentation, the fermented juice was

centrifuged (8000 g, 4 �C, 10 min) to separate the yeast cells. Thepellets were discarded and the supernatants were filtered through0.45 mm cellulose membranes (Corning Incorporated, NY, USA).Four volumes of cold 95% ethanol containing 0.3 M HCl was addedto the permeate; the permeate was then kept under 4 �C for 24 h toprecipitate the polysaccharide. The precipitated fraction was thenseparated by centrifugation (4255 g, 4 �C, 30min). The supernatantswere discarded and the pellets were washed two times with fourvolumes of 95% of cold ethanol and finally dried at 27 �C in a SpeedVacuum concentrator (MiVac Quattro Concentrator, Genevac).

The dried pellets were then rehydrated with a known amount ofnanopure water for the successive analytical determinations.

2.3.4. Polysaccharide characterization

2.3.4.1. Monosaccharide composition. The monosaccharide compo-sition of the polysaccharides was determined according to themethod reported by Quirós et al. (2011). Briefly, 20 mL of therehydrated pellets were dried at 35 �C in the speed vac and then re-suspended into 100 mL of 1 M H2SO4 and incubated in an oven at100 �C for 4 h in sealed vials. After hydrolysis, the tubes were brieflyspun down and the samples were diluted 10 times by adding 900 mLof nanopurewater. Finally, the samples were filteredwith a 0.45 mmTeflon filter (MSI, Westboro, MA, USA) and were injected into HPLCjust after the hydrolysis to quantify the amount of monosaccharidesreleased during hydrolysis. The HPLC analysis of sugars was per-formed as reported above, with an external calibration curve pre-pared using an aqueous solution of 100 mM H2SO4 to mimic the

conditions after the hydrolysis; concentrations of each compoundranged from 250 mg/L to 1 g/L.

2.3.4.2. Protein quantification. The protein concentration wasdetermined by dye-binding Bradford assay (Bradford, 1976) usingbovine serum albumin (BSA) (SigmaeAldrich) and dye reagents(Bio-Rad) for the calibration curve.

2.3.4.3. Proteins profiling by gel electrophoresis. Electrophoresiswas performed by using 10% Sodium Dodecyl Sulfate poly-acrylamide gel electrophoresis (SDS-PAGE) (Laemmli, 1970). Basedon the protein concentration, the rehydrated pellets obtained fromthe precipitated polysaccharides were further diluted with ultra-pure MilliQ water to have a protein concentration ranging between0.2 and 0.3 mg/mL. Finally, 20 mL of the diluted sample was treatedwith 6.65 mL of 4� Laemmlli buffer (Bio-Rad) and 2.75 mL of 1 Mdithiothreitol (DTT) (Acros Organics) and was heated at 95 �C for5 min. Then 15 mL of the mixture was loaded onto the gel.

CandyCane molecular weight standard (a mixture of glycosy-lated and un-glycosylated protein standards) (Invitrogen, Carlsbad,CA, USA) at a concentration of 750 ng/lane, was used. In addition,10 mL of Blue precision plus protein standard (Bio-Rad) was loaded.

The SDS-PAGE was developed in a Mini Protean II apparatus(Bio-Rad) at 45 V within the stacking gel and at 104 V in thedeveloping gel (until the tracking dye bromophenol blue ran off thegel).

Gels were stained first with a Pro-Q Emerald 488 glycoproteinstain (Molecular Probes, Inc., Eugene, OR, USA) to detect the gly-coproteins according to the manufacturer’s protocol. The resultinggreen-fluorescent signal produced by the Pro-Q Emerald 488 dyewas visualized by a TyphoonTrio Plus variable mode imager (GEHealthcare) with a 488 nm excitation laser and a 520 nm long-passemission filter. After detecting glycoproteins, total protein profileswere subsequently detected using Bio-Safe Coomassie G-250 stain(Bio-Rad).

2.3.4.4. N-glycan isolation2.3.4.4.1. From the purified polysaccharide fraction. To charac-

terize the mannoproteins released in the media at the end ofalcoholic fermentation by the different non-Saccharomyces yeaststrains, analysis of the N-glycans released from the glycoproteindeglycosylated by PNGase F enzyme was performed. 200 mL of10 mM dithiothreitol (DTT) in an equal volume of 200 mMNH4HCO3, was added to the fraction obtained by ethanol precipi-tation of 10 mL of fermented juice and then boiled for 5 min. Toeach sample 2 mL of PNGase F (New England Biolab Inc. IpswichMA)was added and the deglycosylation reaction was performed over-night at 37 �C. Finally, the samples were prepared for furtherMALDI-TOF analysis as described below.

2.3.4.4.2. In situ digestion. The bands present in the SDS-PAGEgel, stained only with Bio-Safe Coomassie G-250 stain, wereexcised and the gel slices were chopped into small pieces and driedin speed vac. The gel pieces were then rehydrated by adding 200 mLof 10 mM dithiothreitol (DTT) in 100 mM NH4HCO3 and thenincubated at 55 �C for 1 h. After cooling to room temperature, thesupernatant, which contained DTT, was discarded. The gel pieceswere then treated with 200 ul of iodoacetamide (IoAA) and incu-bated in dark for 45 min. The gel pieces were treated with 100 mL of100 mM NH4HCO3 with agitation for 10 min and washed with thesame amount of pure acetonitrile (ACN) with agitation for 10 min.These steps were repeated twice, after which a 1:1 mixture of100 mM NH4HCO3 and ACN was added. The supernatant was dis-carded and the gel pieces were dried in a speed vac. In order torelease the glycan moieties from mannoproteins, a PNGase F so-lution (1 mL of PNGase F enzyme in 30 mL of 200mMNH4HCO3), was

Fig. 1. Saccharomyces and non-Saccharomyces growth ( ), fermentation kinetics ( ), and total polysaccharide content (columns) after 13 days of alcoholic fermentation.Data are representative of two independent experiments. Error bars represent standard deviation of two independent experiments, each carried out in duplicate.

Table 2The fermentative performance of the eight non-Saccharomyces yeast strains and thecommercial strain of S. cerevisiae in terms of ethanol production and amounts ofresidual sugars after 13 days of alcoholic fermentation.

Strain Glucose (% w/v) Fructose (% w/v) Ethanol (% v/v)

S. cerevisiae# Ec1118 0.01 � 0.00 0.01 � 0.00 13.08 � 0.70P. fermentans# 4 3.23 � 0.20 8.63 � 0.70 6.61 � 0.30S. bacillaris# 22 9.22 � 0.60 0.01 � 0.00 7.80 � 0.50H. osmophila# 32 2.07 � 0.01 4.81 � 0.20 10.21 � 0.90Z. florentinus# 42 1.58 � 0.07 5.36 � 0.02 9.73 � 0.60M. pulcherrima# 46 3.88 � 0.02 7.19 � 0.50 7.04 � 0.00S. ludwigii# 64 0.03 � 0.00 1.39 � 0.10 12.27 � 0.60T. delbrueckii# 92 2.46 � 0.10 5.82 � 0.30 8.90 � 0.40L. thermotolerans# 101 3.36 � 0.03 5.69 � 0.30 7.66 � 0.40

P. Domizio et al. / Food Microbiology 43 (2014) 5e15 9

added to the dried gel pieces. After the gel pieces had completelyabsorbed the enzyme solution in an ice bath, 100 mL of the 100 mMNH4HCO3 solution, was added. The digestion was carried out at37 �C overnight. Glycans were extracted from the gel pieces byremoving the incubation buffer and washing three times with100 mL of 5% formic acid in 60% acetonitrile solvent. The solvent,which contained N-glycans, was dried in a speed vac. Finally, thesamples were suspended in water and prepared for further MALDI-TOF analysis as described below.

2.3.4.5. MALDI-TOF analysis. The N-glycans released from both thepool of proteins and the single band of the gel were detected ac-cording to the method reported in Bordiga et al. (2012), with somemodifications. A two-step solid-phase extraction (SPE) was per-formed using a C-8 cartridge to remove the PNGase F enzyme andother proteins, and a carbograph to remove residual salts andmonosaccharides that would interfere with MS analysis. C-8 car-tridges (3-mL SupelCleanTM LC-8 SPE tubes, Supelco, PA, USA)were conditioned with three volumes of 80% ACN containing 0.1%Trifluoroacetic acid (TFA) and three volumes of water. Graphitizedcarbon cartridges (150 mg carbon, 4 mL tube capacity, Alltech,Deerfield, IL, USA) were conditioned following the protocoldescribed by Ninonuevo et al. (2006). TheN-glycans retained by thegraphitized carbon were then eluted stepwise with three cartridgevolumes of an 80:20 deionized watereACN solution and then driedin a vacuum centrifuge. Malto-oligosaccharides isolated from beerwere used to calibrate the instrument as a molecular referencestandard for oligosaccharides consisting of hexose (Hex) residues.One mL of each samplewasmixedwith 0.4 mL of 1mMNaCl and 1 mLof DHB (20 mg/mL of 2,5-dihydroxybenzoic acid in 70% ACN and0.1% TFA) and finally, 0.5 mL of the mixture was spotted ontostainless steel sample plate (Applied Biosystems, Foster City, CA,USA). The spots were then allowed to dry under vacuumprior toMSanalysis. MALDI-TOF analysis was performed in the m/z scan rangefrom 400 Da to 6000 Da on a Bruker Ultraflex II MALDI-TOF (Bru-kerDaltonics, Bremen, Germany) in positive ion mode. Each samplewas spotted in triplicate from biological duplicate extractions andthe mass spectra were collected by adding individual spectraresulting from 100 laser shots. All mass spectra were analyzed us-ing the Flex-Analysis software from Bruker Daltonics (Bremen,Germany).

3. Results and discussion

3.1. Fermentation performance and polysaccharide release

The eight non-Saccharomyces strains utilized in the presentwork were selected based on their ability to modulate the finalconcentrations of various volatile compounds and to persist with S.cerevisiae in mixed inocula fermentations of grape juice (Comitiniet al., 2011; Domizio et al., 2011). Because native grape poly-saccharides can impact the wine polysaccharide profiles, apolysaccharide-free synthetic grape juice was utilized. The abilityof these yeasts to release polysaccharides could then be verifiedand the polysaccharides could be chemically characterized. As ex-pected, the growth rate and fermentation capacity of all non-Saccharomyces strains were low compared to a commercial strain ofS. cerevisiae EC1118 (Fig. 1). However, the strains Saccharomycodesludwigii, Zygosaccharomyces florentinus, and Hanseniaspora osmo-phila reached an ethanol concentration greater than or equal to 10%after 13 days of alcoholic fermentation (Table 2). However, all thenon-Saccharomyces strains, with the exception of H. osmophila andStarmerella bacillaris, showed a higher level of total polysaccharides(mg/L) in the relevant fermented media compared to S. cerevisiaeEC1118 (Fig. 1). In particular, an increase of polysaccharides in the

media ranging from 52% to 61% was observed for the strainsMetschnikowia pulcherrima, Torulaspora delbrueckii and Pichia fer-mentans, and a remarkable increase of 118% was observed for thestrain S. ludwigii.

These results are in agreement with previous findings where thehigher capacity of the non-Saccharomyces yeasts to release poly-saccharides compared to S. cerevisiae was evaluated in naturalgrape juice (Comitini et al., 2011; Domizio et al., 2011). Moreover,Romani et al. (2010) observed a wide intergenic and intragenicbiodiversity among non-Saccharomyces yeasts belonging to 10different genera with regards to the capacity to release poly-saccharides. With regards to the quantity of polysaccharidesreleased in the media per gram of yeast biomass (dry weight),S. cerevisiae EC1118 produced the lowest amount (65 mg/g) (Fig. 2)throughout the experiment. In contrast, M. pulcherrima, S. ludwigiiand P. fermentans produced the highest amounts of polysaccharides(mg/g dw) (133, 139, and 170 mg/g, respectively). Interestingly, thehigh ratio value of polysaccharides per gram of cell dry weight (mg/g dw) for the two yeast strains M. pulcherrima and P. fermentanscould be due to an early enzymatic cell wall degradation, typicallyobserved during the autolytic process.

During the first day of the alcoholic fermentation, each strainshowed a different ability to release polysaccharide, which wasgrowth-independent. In this contest, while the strain P. fermentansshowed the highest ratio polysaccharide (mg/L)/growth (measuredas optical density at 600 nm), the H. osmophila showed the lowestone. This different behavior could be explained as a differentresponse, yeast strain-dependent, to external stimuli that cause cellwall stress. Indeed, various authors (Aguilar-Uscanga and François,2003; Klis et al., 2002, 2006) report that yeast cell walls represent adynamic protective shells, undergoing continuous changesdepending on culture conditions and environmental stresses,involving the cell wall proteins presented at the cell surface.

In contrast, during the next 3 days of the alcoholic fermentation,the polysaccharide content was growth-dependent. Indeed, thecorrelation between growth (OD600 nm) and polysaccharide con-centration (mg/L) in the medium ranged from R2 ¼ 0.9571 toR2 ¼ 0.9985 for all the strains, with the exception of the strainZ. florentinus (R2 ¼ 0.8968) (data not shown). In this case, thepolysaccharide release could be explained as the result of activelygrowing cells, as already observed by other authors (Boivin et al.,1998; Llaubères et al., 1987) for strains to the genus Saccharo-myces. During the growth phase the cell produces b-glucanases toweaken the cell wall and allow bud emergence, a process thoughtto lead to the release of polysaccharides. At the end of the expo-nential phase or early in the stationary phase, most of the non-Saccharomyces strains had already released most of the poly-saccharides into the media. In particular, between the 5th and 13thday of fermentation, an increase of the polysaccharide levels in themedia, ranging from 13% to 29%, was observed for most of the non-Saccharomyces strains. Interestingly, a higher increase (35%), was

Fig. 2. Cell dry weight in mg/mL (,) and the ratio of polysaccharides (mg) to grams of cell dry weight (-) (mg/g) for all yeast strains. Error bars represent the standard deviation oftwo independent experiments, each carried out in duplicate.

P. Domizio et al. / Food Microbiology 43 (2014) 5e1510

observed for the strains S. ludwigii, which after 13 days was stillslowly fermenting and showing the highest cell viability(8 � 106 cell/mL) compared to all the other yeast strains (Fig. 3).This non-Saccharomyces yeast strain showed the highest quantityof polysaccharides released in the media (558 mg/L).

In contrast, a poor correlation (R2 ¼ 0.583) between growth andpolysaccharides released in the media during the exponentialphase was observed for the commercial strain S. cerevisiae EC1118.For this strain, the quantity of polysaccharides released was around100 mg/L after the first day of fermentation and remained almostconstant at this level for the next five days, when the fermentationactivity, measured as CO2 released, stopped and the cells were inthe stationary phase of growth. However, after three more days(8th day of alcoholic fermentation), an increase of 55% of

Fig. 3. Viable cell at time 0 (,) and after 13 days (-) of alcoholic fermentation. Error baduplicate.

polysaccharides was observed, remaining constant through the13th day of fermentation when the viability of the strain was stillhigh (104 cell/mL).

In contrast, polysaccharides are released continuously duringthe growth of most of the non-Saccharomyces yeast strains; thiscould reflect higher mannoproteins production levels in theseyeasts or indicate a greater turnover of cell wall componentsduring growth or during quiescent phases. In native fermentation,non-Saccharomyces yeasts are often present at high concentrationsat the beginning of alcoholic fermentation (Heard and Fleet, 1985;Fleet et al., 1984) and therefore they could contribute to a higherquantity of polysaccharides in the relevant wine. However,because native fermentations are not always predictable in theproducts being produced, using non-Saccharomyces wine yeast

rs represent standard deviation of two independent experiments, each carried out in

Fig. 4. The composition ewith regards to sugars and protein e of the polysaccharides released by each yeast strain after 13 days of alcoholic fermentation. Protein ( ) and sugars (mannose, glucose) were measured as a percent of the rehydrated pellets derived from the polysaccharides precipitated with ethanol. Error bars represent standard deviation oftwo independent experiments, each carried out in duplicate.

Fig. 5. SDS-PAGE electrophoresis of the proteins released into the media from Saccharomyces and non-Saccharomyces yeasts. Glycoprotein visualization was performed using a two-step procedure. The glycoproteins were first stained on electrophoretic gel with Pro-Q Emerald 488 glycoprotein gel stain kit (5a, b). Then, the proteins were stained in the same gelwith Bio safe Coomassie (5c, d). Std1: Blue precision plus molecular weight standard; Std2: CandyCane molecular weight standard. (For interpretation of the references to color inthis figure legend, the reader is referred to the web version of this article.)

P. Domizio et al. / Food Microbiology 43 (2014) 5e15 11

Fig. 6. MALDI-TOF-MS profiles of N-glycans prepared by PNGase F digestion. Digestion was conducted on the proteins released by each yeast strain during alcoholic fermentation.The chromatograms are representative of the analyses performed on each sample in triplicate from biological duplicate extractions. HexNAc: N-acetyl hexosamines.

P. Domizio et al. / Food Microbiology 43 (2014) 5e15 13

strains together with Saccharomyces strains in mixed fermenta-tions could be a tool to obtain the advantages of spontaneousfermentation; this would also reduce the risk of stuck fermenta-tion (Bisson and Kunkee, 1993). Based on the wine target, thewinemaker could choose the appropriate mixed Saccharomyces/non-Saccharomyces starter to enhance specific characteristics ofwine.

Although further studies will need to be performed to optimizethe process of mixed-culture fermentations, the results obtained sofar indicate that selected non-Saccharomyces yeast strains couldbecome innovative tools to increase wine complexity.

Moreover, the high production of polysaccharides after onlythree days of growth and fermentation makes them an easy sourceof exogenous polysaccharides. This quality of non-Saccharomycesyeast strains would eliminate the need of extracting these macro-molecules from yeast cell walls, which is currently a commonpractice for commercial mannoproteins.

3.2. Polysaccharides characterization

Previous work (Llaubères et al., 1987; Villetaz et al., 1980) re-ported that the structure of the mannoproteins released into winedepends on the yeast strain but is still similar to that of the yeastcell wall. Because of the different cell wall composition at the

Fig. 7. MALDI-TOF-MS profiles of N-glycans prepared by PNGase F digestion. Digestion was cthe analyses performed on each sample in triplicate from biological duplicate extractions.

genera, species, and strain level (Ballou, 1976; Yoshifumi andTetsuji, 1999), differences in the relevant released mannoproteinscomposition are predictable.

The analysis of sugar composition confirmed that these poly-saccharides released in the media are indeed of cell wall origin andtherefore essentially mannoproteins. As a matter of fact, a highpercentage of mannose was found among the monosaccharidesidentified by HPLC (Fig. 4), with a content ranging from 86% to 95%for most of the non-Saccharomyces strains, with the exception ofS. bacillaris (80%). A small percentage of glucose was also present(Fig. 4), ranging from 10% to 20%, with the exception of T. delbrueckiiand S. ludwigii (7% and 5%, respectively). The protein percentage ofthe polysaccharides ranged from 2.4% to 3.9% with the exception ofT. delbrueckii and M. pulcherrima (1.5% and 1.2%, respectively)(Fig. 4). These values are in agreement with previous findingsregarding the polysaccharides released during fermentation byS. cerevisiae (Gonçalves et al., 2002; Rosi et al., 2000).

The overall pattern of the extracted proteins resolved by SDS-PAGE was very different among all the strains (Fig. 5). A wide va-riety of sizes, ranging from 25 to greater than 250 kDa, wasobserved for all eight strains. Moreover, several proteins werefound in the stacking gel most likely due to their high molecularmasses. Glycoprotein visualizationwas performed using a two-stepprocedure. First, the glycoproteins were stained on electrophoretic

onducted on the bands excised from the gel. The chromatograms are representative ofHexNAc: N-acetyl hexosamines.

P. Domizio et al. / Food Microbiology 43 (2014) 5e1514

gel with a Pro-Q Emerald 488 glycoprotein gel stain kit (Fig. 5a, b).All the proteins were subsequently stained in the same gel byCoomassie (Fig. 5c, d). It is evident from the figure that most of theproteins were glycosylated and those with a molecular weighthigher than 250 Kda were particularly abundant (darker bands) inthe strains belonging to the genera Metschinikowia, Saccharomyc-odes, and Torulaspora. A similar range of glycoproteins with mo-lecular weights from 50 to 500 kDa was observed amongS. cerevisiae strains (Llaubères et al., 1987; Villetaz et al., 1980).Moreover, Palmisano et al. (2010) found a total of 15 different yeastglycoproteins on a Chardonnay wine fermented with a commercialSaccharomyces yeast strain.

The presence of mannoproteins in the synthetic grape juicefermented by the different non-Saccharomyces yeast strains wasconfirmed by characterizing the released N-linked glycans byMALDI-TOF. Fig. 6 shows the MALDI-TOF spectra of neutral gly-cans ranging from 1600 to 4000 Da. The N-glycans were observedto vary in size starting with the Man8GlcNAc2 and hexose typesugar chains present in a different proportion for each strain. Thepresence of these peaks was confirmed with duplicate extrac-tions and triplicate spotting and analysis. While the absoluteamount of glycans cannot be determined, the relative abundanceof the various glycans can be delineated from the mass spectra. Inparticular, all strains showed unique profile with a different in-tensity for each glycans and a core of N-glycans with massranging in size from 8 mannoses up to 15 mannoses(Man8GlcNAc2 to Man15GlcNAc2). The analysis of the N-glycanswith MALDI-TOF does not allow for the assignment of definedglycan structures, but does provide a snapshot of the composi-tions present.

To further understand the protein origin of N-glycans, selectedbands (that were stained by Pro-Q Emerald staining) were excisedfrom the gel (indicated with arrow in Fig 5c, d) and treated withPNGase F to release the glycans and compared with the profileobtained on the pool proteins. Fig. 7 shows the MALDI-TOF-MSprofiles of neutral N-glycans prepared by in-gel PNGase F diges-tion of two bands belonging to T. delbrueckii, one band toS. bacillaris, and one band to Z. florentinus. Interestingly, a similartrend was seen for N-glycans present in the single glycoproteinband and in the total pool of proteins. However this phenomenonwas restricted to within the individual strains; inter-strain vari-ability was observed, confirming what already observed in theanalysis of pooled proteins.

Because the different chemical composition of the glycoproteinscould determine a different functional effect on the wine, furtheranalyses will be necessary to evaluate a possible correlation/effecton the wine attributes derived from the different mannoproteinreleased by the non-Saccharomyces strains. For example,Charpentier et al. (2004b) found a direct correlation betweenmannose/glucose ratio of yeast glycoproteins and their capacity toprevent tannin aggregation in a model solution.

The utilization of more informative mass spectrometry tech-niques, such as ESI in combination with microchip-based nano-liquid chromatographic separation, could enable the full charac-terization of novel important glycoproteins, providing insights intothe structural complexity of the components released from non-Saccharomyces yeasts.

Acknowledgment

Financial support for this research was provided by the UC Davisstart-up fund of D. Barile.

The authors acknowledge technical support from Carolyn Doyle,Dr. Annabelle LeParc Le Bouedec, Dr. David Dallas, Dr. HyeyoungLee, Allison Dycaico and Aaron Hsu.

References

Aguilar-Uscanga, B., François, J.M., 2003. A study of the yeast cell wall compositionand structure in response to growth conditions and mode of cultivation. Lett.App. Microbiol. 37, 268e274.

Anfang, N., Brajkovich, M., Goddar, M.R., 2009. Co-fermentation with Pichia kluyveriincreases varietal thiol concentrations in Sauvignon Blanc. Aust. J. Grape WineRes. 15, 1e8.

Ballou, C., 1976. Structure and biosynthesis of the mannan component of the yeastcell envelope. Adv. Microb. Physiol. 14, 93e158.

Ballou, C.E., 1990. Isolation, characterization, and properties of Saccharomyces cer-evisiae mnn mutants with non-conditional protein glycosylation defects.Methods Enzym. 185, 440e470.

Bely, M., Stoeckle, P., Masnuef-Pomarède, I., Dubourdieu, D., 2008. Impact of mixedTorulaspora delbrueckii-Saccharomyces cerevisiae culture on high-sugarfermentation. Int. J. Food Microbiol. 122, 312e320.

Bisson, L.F., Kunkee, R.E., 1993. Microbial interactions during wine production. In:Zeikus, J.G., Johnson, E.A. (Eds.), Mixed Cultures in Biotechnology. McGraw-Hill,New York, NY, pp. 37e68.

Boivin, S., Feuillat, M., Alexandre, H., Charpentier, C., 1998. Effect of must turbidityon cell wall porosity and macromolecules excretion of Saccharomyces cerevisiaecultivated on grape juice. Am. J. Enol. Vitic. 49 (3), 325e332.

Bordiga, M., Travaglia, F., Meyrand, M., German, J.B., Lebrilla, C.B., Coisson, J.D.,Arlorio, M., Barile, B., 2012. Identification and characterization of complexbioactive oligosaccharides in white and red wine by a combination of massspectrometry and gas chromatography. J. Agric. Food Chem. 60, 3700e3707.

Bradford, M.M., 1976. A rapid and sensitive method for the quantitation of micro-gram quantities of protein utilizing the principle of proteinedye binding. Anal.Biochem. 72, 248e254.

Brown, S., Stockdale, V., Pettolino, F., Pocock, K., De Barros Lopes, M., Williams, P.,Bacic, A., Fincher, G., Høj, P., Waters, E., 2007. Reducing haziness in white wineby overexpression of Saccharomyces cerevisiae genes YOL155c and YDR055w.Appl. Microbiol. Biotechnol. 73, 1363e1376.

Caridi, A., 2007. New perspectives in safety and quality enhancement of winethrough selection of yeasts based on the parietal adsorption activity. Int. J. FoodMicrobiol. 120, 167e172.

Chalier, P., Angot, B., Delteil, D., Doco, T., Gunata, Z., 2007. Interactions betweenaroma compounds and whole mannoprotein isolated from Saccharomyces cer-evisiae strains. Food Chem. 100, 22e30.

Carvalho, E., Mateus, N., Plet, B., Pianet, I., Dufourc, E., De Freitas, V., 2006. Influenceof wine pectic polysaccharides on the interactions between condensed tanninsand salivary proteins. J. Agric. Food Chem. 54, 8936e8944.

Charpentier, C., Dos Santos, A.M., Feuillat, M., 2004. Release of macromolecules bySaccharomyces cerevisiae during ageing of French flor sherry wine “Vin jaune”.Int. J. Food Microbiol. 96 (3), 253e262.

Charpentier, C., Escot, S., González, E., Dulau, L., Feuillat, M., 2004. The influence ofyeast glycosylated proteins on tannins aggregation in model solution. J. Int. Sci.Vigne Vin. 38, 209e218.

Charpentier, C., Feuillat, M., 1993. Yeast autolysis. In: Fleet, G.H. (Ed.), WineMicrobiology and Biotechnology. Harwood Academic Publishers, Switzerland,pp. 225e242.

Charpentier, C., N’guyen Van Long, T., Bonaly, R.C., Feuillat, M., 1986. Alteration ofcell wall structure in Saccharomyces cerevisiae and Saccharomyces bayanusduring autolysis. Appl. Microbiol. Biotechnol. 24, 405e413.

Ciani, M., 1995. Continuous deacidification of wine by immobilized SchizoSacchar-omyces pombe cells: evaluation of malic acid degradation rate and analyticalprofiles. J. Appl. Bacteriol. 79, 631e634.

Ciani, M., Comitini, F., Mannazzu, I., Domizio, P., 2010. Controlled mixed culturefermentation: a new perspective on the use of non-Saccharomyces yeasts inwinemaking. FEMS Yeast Res. 10 (2), 123e133.

Ciani, M., Ferraro, L., 1996. Enhanced glycerol content in wines made with immo-bilized Candida stellata cells. Appl. Environ. Microbiol. 62, 128e132.

Ciani, M., Ferraro, L., 1998. Combined use of immobilized Candida stellata cells andSaccharomyces cerevisiae to improve the quality of wines. J. Appl. Microbiol. 85,247e254.

Comitini, F., Gobbi, M., Domizio, P., Romani, C., Lencioni, L., Mannazzu, I.,Ciani, M., 2011. Selected non-Saccharomyces wine yeasts in controlled mul-tistarter fermentations with Saccharomyces cerevisiae. Food Microbiol. 28,873e882.

Domizio, P., Lencioni, L., Ciani, M., Di Blasi, S., Pontremolesi, C., Sabatelli, M.P., 2007.Spontaneous and inoculated yeast population dynamics and their effect onorganoleptic characters of Vinsanto wine under different process conditions.Int. J. Food Microbiol. 115, 281e289.

Domizio, P., Romani, C., Comitini, F., Gobbi, M., Ciani, M., Lencioni, L., Mannazzu, I.,2010. Potential spoilage non-Saccharomyces yeasts in mixed cultures withSaccharomyces cerevisiae. Ann. Microbiol. 61 (1), 137e144.

Domizio, P., Romani, C., Lencioni, L., Comitini, F., Gobbi, M., Mannazzu, I., Ciani, M.,2011. Outlining a future for non-Saccharomyces yeasts: selection of putativespoilage wine strains to be used in association with Saccharomyces cerevisiae forgrape juice fermentation. Int. J. Food Microbiol. 147, 170e180.

Ducasse, M.A., Williams, P., Meudec, E., Cheynier, V., Doco, T., 2010. Isolation ofCarignan and Merlot red wine oligosaccharides and their characterization byESI-MS. Carbohydr. Polym. 79, 747e754.

P. Domizio et al. / Food Microbiology 43 (2014) 5e15 15

Escot, S., Feuillat, M., Dulau, L., Charpentier, C., 2001. Release of polysaccharides byyeasts and the influence of released polysaccharides on colour stability andwine astringency. Aust. J. Grape Wine Res. 7, 153e159.

Fleet, G.H., 1991. Cell wall. In: Rose, A.H., Harrison, J.S. (Eds.), The yeasts; YeastOrganelles, vol. 4. Academic Press, London, pp. 199e277.

Fleet, G.H., Lafon-Lafourcade, S., Ribéreau-Gayon, P., 1984. Evolution of yeasts andlactic acid bacteria during fermentation and storage of Bordeaux wines. Appl.Environ. Microbiol. 48, 1034e1038.

Giovani, G., Rosi, I., 2007. Release of cell wall polysaccharides from Saccharomycescerevisiae thermosensitive autolytic mutants during alcoholic fermentation. Int.J. Food Microbiol. 116, 19e24.

Gobbi, M., Comitini, F., Domizio, P., Romani, C., Lencioni, L., Mannazzu, I., Ciani, M.,2013. Lachancea thermotolerans and Saccharomyces cerevisiae in simultaneousand sequential co-fermentation: a strategy to enhance acidity and improve theoverall quality of wine. Food Microbiol. 33 (2), 271e281.

Gonçalves, F., Heyraud, A., De Pinho, M.N., Rinaudo, M., 2002. Characterization ofwhite wine mannoproteins. J. Agric. Food Chem. 50, 6097e6101.

Gonzalez-Ramos, D., Cebollero, E., Gonzalez, R., 2008. A recombinant Saccharomycescerevisiae strain overproducing mannoproteins stabilizes wine against proteinhaze. Appl. Environ. Microbiol. 74, 5533e5540.

Gonzalez-Ramos, D., Gonzalez, R., 2006. Genetic determinants of the release ofmannoproteins of enological interest by Saccharomyces cerevisiae. J. Agric. FoodChem. 54 (25), 9411e9416.

Guilloux-Benatier, M., Guerreau, J., Feuillat, M., 1995. Influence of initial coll.oidcontent on yeast macromolecule production and on the metabolism of winemicroorganisms. Am. J. Enol. Vitic. 46, 486e492.

Guadalupe, Z., Palacios, A., Ayestarán, B., 2007. Maceration enzymes and man-noproteins: a possible strategy to increase colloidal stability and colorextraction in red wines. J. Agric. Food Chem. 55 (12), 4854e4862.

Heard, G.M., Fleet, G.H., 1985. Growth of natural yeast flora during the fermentationof inoculated wines. Appl. Environ. Microbiol. 50, 727e728.

Jolly, N.P., Augustyn, O.P.H., Pretorius, I.S., 2003. The use of Candida pulcherrima incombination with Saccharomyces cerevisiae for the production of Chenin blancwine. South Afr. J. Enol. Vitic. 24, 63e69.

Jolly, N.P., Augustyn, O.P.H., Pretorius, I.S., 2006. The role and use of non-Saccha-romyces yeasts in wine production. South Afr. J. Enol. Vitic. 27, 15e38.

Kapsopoulou, K., Mourtzini, A., Anthoulas, M., Nerantzis, E., 2007. Biological acidi-fication during grape must fermentation using mixed cultures of Kluyveromycesthermotolerans and Saccharomyces cerevisiae. World J. Microbiol. Biotechnol. 23,735e739.

Klis, F.M., Mol, P., Hellingwerf, K., Brul, S., 2002. Dynamics of cell wall structure inSaccharomyces cerevisiae. FEMS Microb. Rev. 26, 239e256.

Klis, F.M., Boorsma, A., De Groot, P.W., 2006. Cell wall construction in Saccharomycescerevisiae. Yeast 2 (3), 185e202.

Laemmli, U.K., 1970. Cleavage of structural proteins during the assembly of the headof bacteriophage T4. Nature 227 (5259), 680e685.

Llaubères, R.M., Dubourdieu, D., Villetaz, J.C., 1987. Exocellular polysaccharides fromSaccharomyces cerevisiae in wine. J. Sci. Food Agric. 41, 277e286.

Magyar, I., Panyik, I., 1989. Biological deacidification of wine with Schizosacchar-omyces pombe entrapped in Ca-alginate gel. Am. J. Enol. Vitic. 40, 233e240.

Mora, J., Barbas, J.I., Mulet, A., 1990. Growth of yeast species during the fermenta-tion of musts inoculated with Kluyveromyces thermotolerans and Saccharomycescerevisiae. Am. J. Enol. Vitic. 41, 156e159.

Moreno-Arribas, V., Pueyo, E., Nieto, F.J., Martın-Álvarez, P.J., Polo, M.C., 2000. In-fluence of the polysaccharides and the nitrogen compounds on foamingproperties of sparkling wines. Food Chem. 70 (3), 309e317.

Moruno, E.G., Sanlorenzo, C., Boccaccino, B., Di Stefano, R., 2005. Treatment withyeast to reduce the concentration of Ochratoxin A in red wine. Am. J. Enol. Vitic.56, 73e76.

Ninonuevo, M.R., Park, Y., Yin, H., Zhang, J., Ward, R.E., Clowers, B.H., German, J.B.,Freeman, S.L., Killeen, K., Grimm, R., 2006. A strategy for annotating the humanmilk glycome. J. Agric. Food Chem. 54 (20), 7471e7480.

Nunez, Y., Carrascosa, A., González, R., Polo, M., Martinez-Rodriguez, A., 2006.Isolation and characterization of a thermally extracted yeast cell wall fractionpotentially useful for improving the foaming properties of sparkling wines.J. Agric. Food Chem. 54, 7898e7903.

Palmisano, G., Antonacci, D., Larsen, M.R., 2010. Glycoproteomic profile in wine: a‘Sweet’ molecular renaissance. J. Proteome Res. 9 (12), 6148e6159.

Peyron, D., Boukharta, M., Cuby, A., Feuillat, M., 1993. Dosage des polysaccharidesdans les vins rouges. Interactions avec les composes phénoliques. Sci. Aliments13, 761e767.

Quirós, M., Morales, P., Pérez-Través, L., Barcenilla, J.M., González, R., 2011. A newmethodology to determine cell wall mannoprotein content and release in wineyeasts. Food Chem. 125, 760e766.

Rantsiou, K., Dolci, P., Giacosa, S., Torchio, F., Tofalo, R., Torriani, S., Suzzi, G., Rolle, L.,Cocolin, L., 2012. Candida zemplinina can reduce acetic acid produced bySaccharomyces cerevisiae in sweet wine fermentations. Appl. Environ. Microbiol.78, 1987e1994.

Renouf, V., Claisse, O., Lonvaud-Funel, A., 2007. Inventory and monitoring of winemicrobial consortia. Appl. Microbiol. Cell Physiol. 75, 149e164.

Rojas, V., Gil, J.V., Pi�naga, F., Manzanares, P., 2001. Studies on acetate esterproduction by non-Saccharomyces wine yeasts. Int. J. Food Microbiol. 70,283e289.

Romani, C., Domizio, P., Lencioni, L., Gobbi, M., Comitini, F., Ciani, M., Mannazzu, I.,2010. Polysaccharides and glycerol production by non-Saccharomyces wineyeasts in mixed fermentation. Quad. Vitic. Enol. Univ. Torino 31, 185e189,2009e2010.

Rosi, I., Gheri, A., Domizio, P., Fia, G., 2000. Production de macromolecules pariétalesde Saccharomyces cerevisiae au cours de la fermentation et leur influence sur lafermentation malolactique. Rev. Des. Oenol. 94, 18e20.

Soden, A., Francis, I.L., Oakey, H., Henschke, P.A., 2000. Effect of co-fermentationwith Candida stellata and Saccharomyces cerevisiae on the aroma and compo-sition of Chardonnay wine. Aust. J. Grape Wine Res. 6, 21e30.

Spiropoulos, A., Tanaka, J., Flerianos, I., Bisson, L.F., 2000. Characterization ofhydrogen sulfide formation in commercial and natural wine isolates ofSaccharomyces. Am. J. Enol. Vitic. 51, 233e248.

Swiegers, H., Bartowsky, E.J., Henschke, P.A., Pretorius, I.S., 2005. Yeast and bacterialmodulation of wine aroma and flavour. Aust. J. Grape Wine Res. 11, 139e173.

Trimble, R.B., Atkinson, P.H., 1992. Structural heterogeneity in the Man8e13GlcNAcoligosaccharides from log-phase Saccharomyces yeast: a one- and two-dimensional 1H NMR spectroscopic study. Glycobiology 2, 57e75.

Vidal, S., Francis, L., Williams, P., Kwiatkowski, M., Gawel, R., Cheynier, V., Waters, E.,2004. The mouth-feel properties of polysaccharides and anthocyanins in a winelike medium. Food Chem. 85, 519e525.

Villetaz, J.C., Amado, R., Neukom, H., Horisberger, M., Horman, I., 1980. Comparativestructural studies of the d-mannans from a rosé wine and Saccharomycesuvarum. Carbohydr. Res. 81, 341e344.

Yoshifumi, J., Tetsuji, O., 1999. Mannosylphosphate transfer to yeast mannan. Bio-chim. Biophys. Acta-Gen. 1426 (2), 335e345.