OCarbonylTable wineOenococcus oeniMalolactic fermentation

eye itrgann. Mdied during MLF in wine with 12 commercially available strains of Oenococcus oeni.

t presether foet a

International Journal of Food Microbiology 155 (2012) 153157

Contents lists available at SciVerse ScienceDirect

International Journal o

sevSO2 an ideal and cost effective food stabilizer. Despite its value andpotency, a majority of consumers view sulte additions to wine asunnatural and unhealthy (Stolz and Schmid, 2008). These apprehen-sions about SO2 utilization may stem from historical sulte abuses insome grocery stores and restaurants (Martin et al., 1986) and sultewarning labels on wine bottles. The US FDA estimates that 1% of theU.S. population show an increased degree of sensitivity to sultes(Papazian, 1996). Asthmatics appear to be especially susceptible,with estimates that up to 5% may risk adverse reactions upon sulte

ing high product quality. An important strategy has been to decreaseconcentrations of carbonyl or keto compounds that bind with SO2,thus decreasing its preservative activity compared to free SO2(Rankine, 1968; Ribreau-Gayon et al., 1998a). Acetaldehyde is an im-portant SO2 binding compound found in wine, primarily due to thelow dissociation constant of its sulfonate (Kd=1.5106) (Ribreau-Gayon et al., 1998a). Other SO2 binding compounds including pyruvate,-ketoglutaric and galacturonic acids may also have a signicant effecton bound SO2 levels (Ribreau-Gayon et al., 1998a).exposure (Snelten and Schaafsma, 1992; Vallprocessed foods and wines with little or no suingly popular with consumers (Azabagaoglu

Corresponding author at: Department of Food ScienExperiment Station, Cornell University, Geneva, NY 1445fax: +1 3157872284.

E-mail address: rm369@cornell.edu (R. Mira de Ord

0168-1605/$ see front matter 2012 Elsevier B.V. Alldoi:10.1016/j.ijfoodmicro.2012.01.025l., 2002), anti-oxidantain and Morris, 1991;w concentrations make

organic wine market grew by 1015% (Richter and Padel, 2007).Tomeet increasing consumer demands, winemakers are challenged

to restrict or even eliminate sultes during vinication, whilemaintain-

(Danilewicz, 2003) and anti-enzymatic (MWedzicha et al., 1991) functionalities at lo1. Introduction

Sulfur dioxide (SO2) is an importanin winemaking and the production of o2007). Its anti-microbial (Carretturonic acid was not. At the time of malic acid depletion, percent degradation of pyruvate, -ketoglutaricacid and acetaldehyde was 49%, 14% and 30%, respectively. During MLF, the decrease in average bound SO2levels, as calculated from carbonyl metabolism, was 22%. The largest reduction in wine carbonyl contentoccurred in the week after completion of MLF and was 53% (107 mg/L to 34 mg/L) calculated as boundSO2. Prolonged activity of bacteria in the wines (up to 3 weeks post malic acid depletion) resulted only inreduced additional reductions in bound SO2 levels.The results suggest that microbiological wine stabilization one week after malic acid depletion is an effectivestrategy for maximum removal of SO2 binders while reducing the risk of possible post-ML spoilage by O. oenileading to the production acetic acid and biogenic amines.

2012 Elsevier B.V. All rights reserved.

rvative commonly usedods (Doyle and Beuchat,

sales of certied organic foods have increased from $78 million in1980 to approximately $6 billion in 2000, with an average annualincrease of 24% during the 1990s (Hughner et al., 2007). The organicwine sector has also followed this trend, and in 2005 the worldSulfur dioxideAcetaldehydePyruvic acid, acetaldehyde and -ketoglutaric acid were degraded to various degrees by O. oeni, but galac-

Keywords: and malic acid were followeMetabolism of SO2 binding compounds bymalolactic fermentation in white wine

J.N. Jackowetz, R. Mira de Ordua Cornell University, NYSAES, Department of Food Science, Geneva, NY 14456, USA

a b s t r a c ta r t i c l e i n f o

Article history:Received 25 October 2011Received in revised form 23 January 2012Accepted 29 January 2012Available online 16 February 2012

Sulfur dioxide (SO2) is the kcan bind to SO2 and decreasmand for low sulte and omanaged during vinicatiobut detailed time course stu

j ourna l homepage: www.e ly et al., 1999). Minimallyltes have been increas-et al., 2007). In the U.S.,

ce, New York State Agricultural6, USA. Tel.: +1 3157872266;

ua).

rights reserved.enococcus oeni during and after

additive for the preservation of wines. Carbonyl and keto compounds in wines efcacy, resulting in higher total SO2 requirements. Increased consumer de-ic wines pose production challenges if SO2 binders have not been properlyalolactic fermentation (MLF) has been known to reduce bound SO2 levels

s are not available. In this work, the kinetics of major SO2 binding compounds

f Food Microbiology

i e r .com/ locate / i j foodmicroPrevious studies have recognized the importance of malolactic fer-mentation (MLF) to reduce the pool of SO2 binders in wine (Flaminiet al., 2002; Radler, 1986; Zaunmuller et al., 2006), especially acetal-dehyde (Osborne et al., 2000). The aim of this work was to providethe comprehensive analysis of the kinetics of acetaldehyde, pyruvate,-ketoglutaric acid and galacturonic acid concentrations during MLFwith 12 commercial strains of heterofermentative Oenococcus oeniin wine. This thorough analysis of SO2 binders during and after

malic acid metabolism, with concurrent modeling of bound SO2levels, offers new insights for sulte management during vinication.

2. Materials and methods

2.1. Microorganisms and chemicals

Saccharomyces cerevisiae strain CY3079 and 12 commercial O. oenistrains were provided by Lallemand Inc. (Montral, Canada) and

directly. Separation occurred on a Phenomenex Kinetex C18 sta-tionary phase (1003.0 mm) with 2.6 m particle size held at 37 Cwith a ow rate of 0.75 mL min l. The analytes were quantied at

eters

Malic acid(g/L)

Concentration of SO2 binding compounds before MLF (mg/L)

AcHO Pyr -KG GA

3.020.04 84.61.0 11.70.5 24.60.3 612.417.2

Table 2Gradient program for the chromatographic separation of DNPH derivatized SO2binding compounds using a Phenomenex Kinetex C18 column (1003.0 mm).

Time (min) Solvent A Solvent B

0.00 85 154.25 82 184.75 75 256.50 62 3812.00 40 60

154 J.N. Jackowetz, R. Mira de Ordua / International Journal of Food Microbiology 155 (2012) 153157stored according to the manufacturer's recommendations. All chemi-cals, unless cited otherwise, were of analytical grade from ThermoFisher Scientic (Fairlawn, NJ).

2.2. Grape must and general inoculation procedures

A ash pasteurized German Riesling (Kamil EX-IM, Canada) mustwas used as medium for all fermentations in this study. The analyticalprole of the must and the wine after alcoholic fermentation is sum-marized in Table 1. Yeast inoculation was conducted in accordancewith supplier's recommendations; 250 mg/L of the yeast (CY3079)and yeast hydration nutrient (GoFerm, Lallemand, Canada) wereeach added to the grape must. Alcoholic fermentation was completedin 2.0 U.S. gallon jugs with air locks to dryness (b5.0 g/L sugars).Bentonite (2.0 g/L Ca-Granulat, Erbslh Geisenheim, Germany) wasadded after the completion of alcoholic fermentation to aid in yeastsettling and clarication. Wines were then cold stabilized for 5 daysat 4 C and racked in preparation for sterile ltration using nylonmembrane lters (0.45 m, Millipore, Ireland). 200 mL aliquots ofRiesling wine were aseptically transferred into previously sterilized250 mL glass bottles for inoculation with O. oeni. O. oeni strains werepre-grown in sterile MRS media (BD, Franklin Lakes, NJ) (pH 4.5) untilstationary phase, and then centrifuged at 2000 g for 10 min to concen-trate biomass (~1.0108 CFU/mL). Cells were washed using a sodiumhydrogen tartrate buffer containing 7.5 g/L tartaric acid, 495 mg/LMgSO4, 345 mg/L MnSO45H2O and 0.005% (w/v) Tween 80 adjustedto pH 4.5 with 5 N NaOH. O. oeni suspensions thus obtained werethen inoculated into each container at a rate of 1% v/v and held atroom temperature (201 C) until completion of MLF.

2.3. Sampling

During MLF, samples were taken regularly and immediately fro-zen (18 C) for future analysis. During sampling, food grade nitro-gen (Airgas, NY) was used to ush the headspace of fermentationbottles and limit oxygen ingress.

2.4. Analytical methods and statistical analysis

Glucose, fructose and malic acid were measured enzymaticallyusing a commercial test kit (Megazyme, Ireland). Titratable aciditywas determined by titration of must with 0.1 M NaOH, using phenol-phthalein as an indicator. Total yeast assimilable nitrogen (YAN) wascalculated as the sum of primary amino acids and total ammonia. Pri-mary amino acid content was quantied using the NOPA method de-ned by Dukes and Butzke (1998), while ammonia was quantiedusing an ion selective electrode (Ammonia ISE electrode, Cole Palmer,

Table 1Analytical prole of Riesling grape must and wine (average of duplicate samplesSE).

Must parameters Wine param

Solublesolids (Brix)

pH Titratable acidity(tartaric acid g/L)

Total YAN(mg/L)

Ethanol(%v/v)

20.6 3.4 7.2 106.2 110.1AcHO (acetaldehyde), Pyr (pyruvate), -KG (-ketoglutaric acid), GA (galacturonic acid).Vernon Hills, IL) according to the method of McWilliams and Ough(1974). Alcohol content (%v/v) was quantied by near infrared spec-troscopy (Alcolyzer, Anton Paar, Germany).

2.5. Chemicals and preparation of reagents for SO2 binder analysis

2,4-Dinitrophenylhydrazine (DNPH) had a purity of 97% (as dryweight) and contained 3040% water to reduce the explosion hazard.ASTM Class I water was prepared using a water purication system(Arium 611UV, Sartorius, Germany) which was used as a solvent forall reagents and UHPLC solvent A (Table 2).

The derivatizing reagent was prepared by dissolving DNPH in ace-tonitrile adjusted to pH 1.5 with perchloric acid to obtain an 11 mMDNPH solution. For sample pre-treatments, an aqueous solution of86 mM EDTA in 1 M sodium hydroxide was prepared and degassed(Aquasonic Model 150D, VWR International, PA) for 20 min. Solutionswere stored up to one week at 4 C.

2.6. Derivatization procedure

Derivatizations were conducted in 2.0 mL glass HPLC vials withTeon caps (National Scientic, TN). For sulfonate hydrolysis, 200 Lof sodium hydroxide/EDTA solution were added to the vial, followedby 100 L of sample wine andmixing. After 10 min, 200 L of 1 M per-chloric acid was added followed by 800 L of the derivatizing reagent.After mixing, the solution was allowed to react for exactly 30 h at30.00.1 C and then promptly cooled to 4 C until analysis. Deriva-tized samples thus prepared were stable for up to ve days.

2.7. HPLC Analysis

An ultra high pressure liquid chromatography system (Shimadzu,Japan) consisting of a binary LC-20AD XR pumping unit, a DGU-20A3degasser, a SIL-20 AC XR autosampler, a CTO-20AC column oven, anda SPD-20A UV/VIS detector were used for separation and analysis ofDNPH derivatized wine carbonyls. Data acquisition and analysiswere performed with the LCSolution software (1.23). Solvent A con-sisted of water acidied to pH 2.500.01 using perchloric acid.Solvent B was HPLC grade acetonitrile. All solvents were lteredprior to utilization (0.22 m, nylon, Millipore, Ireland). Sampleswere held at 4 C in the autosampler and 5.0 L of sample injected

12.50 10 9013.25 85 15

365 nm using external calibration standards with linear regressionanalysis. The separation gradient is shown in Table 2.

2.8. Statistical analysis

Student's t-tests were conducted with JMP 7.0 (SAS, North Carolina)to determine statistical signicance between sample populations atthe 0.05 condence level.

3. Results

Complete malic acid metabolism was achieved during all MLFusing 12 commercial strains of O. oeni. Fig. 1 illustrates the concentra-tions of malic acid and four SO2 binding compounds during MLF intwo exemplary fermentations. The start of malic acid degradationand the time point of its completion was strain dependant. The aver-age duration for malic acid depletion was 25 days after inoculation,with minima and maxima of 16.9 and 43.8 days (R1108 and R1032,respectively, Table 3).

With the exception of galacturonic acid, the concentrations of all

further and steady decrease in calculated bound SO2 concentrations.The mean bound SO2 level reductions were 75%, 81% and 83% mea-sured at one, two and three weeks after malic acid depletion, respec-tively (Table 5).

4. Discussion

Management of SO2 binding compounds is integral for optimalsulte usage in winemaking. Lower SO2 concentrations in wines canreduce adverse consumer perceptions and reactions to wine, and en-sure adherence to domestic and international sulte regulations. Thisstudy is the rst to closely investigate the kinetics of SO2 bindingcompounds throughout and after MLF. Twelve commercial strains ofO. oeni showed equivalent abilities to degrade three important SO2binders, but the extent of their metabolism was strain dependant.

Galacturonic acid is a wine carbonyl of grape origin (Ribreau-Gayon et al., 1998a) found at higher concentrations in wines withsignicant skin contact, i.e. red wines (Sponholz and Dittrich, 1985),and wines produced from botrytized grapes (Francioli et al., 1999).Galacturonic acid was not degraded by any of the O. oeni strainstested in this work.

-Ketoglutaric acid is an important intermediate for amino acidmetabolism in the cell (Berg et al., 2007) and it has been suggestedthat lactic acid bacteria (LAB) use -ketoglutaric acid as an electron

155J.N. Jackowetz, R. Mira de Ordua / International Journal of Food Microbiology 155 (2012) 153157Fig. 1. Kinetics of the concentrations of malic acid ( ), galacturonic acid ( ),-ketoglutaric acid ( ), pyruvic acid ( ) and acetaldehyde ( ) during MLF in Rieslingwine using O. oeni strains R1076 and R1108. Estimated bound SO2 levels wereother SO2 binders studied were strongly affected following inocula-tion with O. oeni. Acetaldehyde concentrations decreased rapidlyduring MLF and followed the metabolism of malic acid with slightdelay. A rapid initial acetaldehyde degradation phase could be distin-guished from a second slower phase (Fig. 1). Across all strains, therapid metabolism of acetaldehyde was completed between 2.0 and8.4 days after malic acid depletion with an average delay of 4.9 days(Table 3). In all fermentations, acetaldehyde degradation continuedafter depletion of malic acid. From an average of 30% degradation atthe time of malic acid depletion, 75% and 87% degradations were ob-served at the end of the rapid acetaldehyde metabolism phase, andthree weeks after malic acid depletion, respectively (Table 4).

A clear distinction of an initial rapid, and a subsequent slower deg-radation phase was also observed for pyruvic acid. However, a tran-sient increase of pyruvic acid concentrations that occurred towardsthe end of malic acid metabolism could be observed in most fermen-tations studied (Fig. 1). In contrast to acetaldehyde, the rapid degra-dation of pyruvic acid was not delayed in comparison with themetabolism of malic acid. On average, the end of the rapid pyruvicacid degradation was found to occur 3 days before the depletion ofmalic acid (Table 3). Across all strains, degradation of pyruvic acidcalculated based on 30 mg/L free SO2.increased from 49% at the time of malic acid depletion to 85% threeweeks post-MLF (Table 4).

The onset of the -ketoglutaric acid degradation was similar tothe start of acetaldehyde degradation, but the degradation rateswere lower thereafter (Fig. 1). At the time of malic acid depletion6%25% (=14.2) of -ketoglutaric acid had been depleted increas-ing to 28%72% (=48.2) three weeks post-MLF (Table 4).

For any given time point, the concentrations of SO2 binders wereused to estimate the total level of bound SO2 if the wines were to bestabilized. For this calculation, the dissociation constants reportedby Ribreau-Gayon et al. (1998b) were used, and the wine was as-sumed to have 30 mg/L of free SO2 and no other SO2 binders. Acetal-dehyde was found to be the most inuential SO2 binding compoundfor the overall calculated level of bound SO2 (Fig. 1). Metabolism ofSO2 binding compounds by O. oeni concomitantly reduced calculatedmean bound SO2 levels by 22% at the time-point of malic acid deple-tion. Wine contact with the bacteria after malic acid depletion led to a

Table 3Comparison between the time point of malic acid depletion and the end of the rapidmetabolism phase of the carbonyls acetaldehyde and pyruvic acid during MLF in Ries-ling by 12 strains of O. oeni. Mean valuesSE are given.

Strain Time ofmalic aciddepletion [d]

Time until end of rapidcarbonyl metabolismphase [d]

Time difference betweenmalic acid depletion andend of rapid carbonylmetabolism phase [d]

AcHO Pyr AcHO Pyr

R1032 43.80.0 48.82.1 38.41.6 5.02.1 5.361.6R1054 24.80.0 28.31.5 23.40.6 3.51.5 1.350.6R1075 43.80.0 48.82.1 38.41.6 5.02.1 5.361.6R1076 38.41.6 41.91.9 36.80.0 3.41.9 1.641.6R1077 22.80.0 24.80.0 17.40.5 2.00.0 5.40.5R1098 18.40.0 26.80.0 17.40.5 8.40.0 1.00.5R1101 19.90.5 28.31.5 17.40.5 8.41.5 2.50.5R1105 16.90.0 19.90.0 14.40.5 3.00.0 2.50.6R1106 17.40.0 24.80.0 14.40.6 7.40.0 3.00.6R1108 16.90.5 22.32.4 14.40.6 5.42.4 2.50.6R1118 16.90.0 19.90.0 14.40.6 3.00.0 2.50.6R1124 19.90.0 24.80.0 17.40.5 4.90.0 2.50.5Mean 25.02.1 29.92.1 22.07.0 4.90.6 3.00.4Range 16.943.8 19.948.8 14.438.4 2.08.4 5.36 to -1.0

AcHO (acetaldehyde), Pyr (pyruvic acid).acceptor during MLF (Zhang and Ganzle, 2010). A study by Hegazi

Table 4Percent degradation of SO2 binding compounds displayed at the time of malic acid depletionMean values are givenSE.

Strain Percent degradation

At time of malic acid depletion [%] At end ophase [%

AcHO Pyr -KG GA AcHO

R1032 28.711.3 68.12.9 20.91.4 0.00.0 58.818R1054 31.612.4 70.82.2 24.53.5 0.00.0 60.917R1075 24.30.0 63.74.3 21.22.2 0.00.0 78.02.R1076 13.310.0 50.91.1 9.92.1 0.00.0 79.72.R1077 48.915.6 54.73.3 15.73.7 0.00.0 82.11.R1098 18.84.8 8.71.7 12.11.1 0.00.0 79.27.R1101 16.413.1 49.65.4 8.92.1 0.00.0 68.312R1105 24.68.2 20.15.1 8.12.1 0.00.0 68.510R1106 29.16.2 42.44.6 12.91.9 0.00.0 77.96.R1108 21.07.7 81.32.3 5.93.1 0.00.0 83.03.R1118 43.910.1 59.42.6 10.62.6 0.00.0 79.91.R1124 64.516.7 56.90.4 19.90.9 0.00.0 82.21.Mean 30.43.7 48.95.7 14.21.3 0.00.0 74.92.Range 13.364.5 20.181.3 5.924.5 N/A 58.883.

AcHO (acetaldehyde), Pyr (pyruvate), -KG (-ketoglutaric acid), GA (galacturonic acid).

156 J.N. Jackowetz, R. Mira de Ordua / International Journal of Food Microbiology 155 (2012) 153157and Abo-Elnaga (1980) found that 112 LAB belonging to the generaLeuconostoc, Streptococcus and Lactobacillus could not degrade -ketoglutaric acid, whereas work by Radler (1986) using O. oeni inwine suggested that concentrations could be decreased. In thiswork, all organisms tested degraded -ketoglutaric acid. However,on average less than half of the initial -ketoglutaric acid was degrad-ed and strain variation was large.

In contrast to -ketoglutaric acid, pyruvic acid levels decreasedrapidly and almost completely during MLF, across all strains studied.Among the SO2 binders tested, pyruvate was the only one whoseconcentration decreased prior to malic acid metabolism. The earlydegradation may be attributable to its signicance in a number of bio-chemical pathways related to energy conservation by ATP production(Henick-Kling, 1993) and NAD(P)+ regeneration (Maicas et al., 2002;Zaunmuller et al., 2006). The transient increase of pyruvic acid levelsduring the latter stages of malic acid degradation may result fromits formation as an intermediate during the metabolism of citrate(Hugenholtz, 1993), which has been observed to commence towardsthe end of MLF (Bartowsky and Henschke, 2000; Nielsen andRichelieu, 1999). Among all O. oeni strains studied, R1105 producedthe highest transient pyruvic acid concentrations during MLF, and

subsequently displayed the highest nal residues.

Table 5Estimated bound SO2 levels during MLF with 12 commercial strains of O. oeni. BoundSO2 were calculated based on the concentration of SO2 binding compounds, thereported dissociation constants of their sulfonates, and assuming 30 mg/L free SO2.

Strain Bound SO2 (mg/L)

Before O. oeniinoculation

At time ofmalic aciddepletion

1 week aftermalic aciddepletion

2 weeks aftermalic aciddepletion

3 weeks aftermalic aciddepletion

R1032 130.010.0 152.012.0 33.35.7 28.91.1 25.10.9R1054 149.219.2 136.211.2 38.71.3 31.51.5 28.91.1R1075 142.52.5 106.32.3 33.23.8 27.12.1 23.82.8R1076 136.54.5 128.88.8 29.70.3 26.61.6 23.80.8R1077 138.25.2 34.215.8 25.21.8 22.30.7 20.30.7R1098 140.320.3 120.05.0 44.18.1 22.30.3 18.50.5R1101 138.65.6 131.213.8 65.07.0 25.02.0 19.61.4R1105 123.616.4 107.08.0 28.71.3 25.50.5 23.30.3R1106 137.24.2 90.910.9 28.30.7 27.50.5 24.32.3R1108 140.310.3 118.78.7 25.87.2 22.52.5 20.21.8R1118 132.82.2 89.012.0 29.911.1 26.04.0 22.31.7R1124 129.17.9 70.414.6 28.95.1 25.92.1 22.60.4Mean 136.52.7 107.16.8 34.22.5 25.90.7 22.70.6Range 123.6149.2 34.2152.0 25.265.0 22.331.5 18.528.9In wine, acetaldehyde can be rapidly metabolized by O. oeni (DiazMaroto et al., 2001; Mayer et al., 1976) leading to the production ofethanol and acetic acid (Osborne et al., 2000). Osborne et al. (2006)observed that acetaldehyde metabolism was simultaneous withmalic acid degradation for two O. oeni strains tested in white wine.However, this interpretation may have resulted from infrequent sam-plings. In this work, the degradation of acetaldehyde followed malicacid metabolism after a denitive delay, with a rapid phase of acetal-dehyde metabolism completed 4.9 days after malic acid depletion. Itis possible that the initial concentrations of malic acid and acetalde-hyde may also inuence the relative durations of malic acid and acet-aldehyde depletion. More work would be needed to elucidate thispossible relationship.

Across most tested strains, the largest decrease in calculatedbound SO2 concentrations (53%) occurred within one week followingmalic acid depletion coinciding with signicant metabolism of acetal-dehyde, whose sulfonate has a very low dissociation constant (Kd=1.5106, (Ribreau-Gayon et al., 1998a)). Together with thedecrease that occurred during MLF (22%), this one week delayresulted in a total reduction of 75% of the calculated bound SO2. Wait-ing for additional two or three weeks only led to a further reduction

, at the end of the rapid carbonyl degradation phase and 3 weeks after MLF completion.

f rapid metabolism]

3-weeks post malic acid depletion [%]

Pyr AcHO Pyr -KG GA

.9 97.03.1 82.71.5 100.00.0 54.74.3 0.00.0

.0 83.40.3 92.30.1 83.40.5 56.31.5 0.00.07 81.61.9 82.81.4 87.50.1 71.82.8 0.00.06 84.32.7 84.10.6 84.60.8 52.43.5 0.00.02 80.41.9 88.80.5 86.50.7 50.23.4 0.00.00 79.81.2 91.80.7 91.70.4 53.41.6 0.00.0.9 77.92.4 91.40.3 88.91.5 35.21.9 0.00.0.4 66.50.9 86.90.2 67.80.1 47.31.9 0.00.01 71.20.4 85.80.2 71.70.3 52.21.8 0.00.00 95.22.9 83.21.3 100.00.0 34.31.3 0.00.06 67.40.4 88.30.4 78.61.6 28.12.3 0.00.00 76.90.7 87.70.2 80.90.3 43.31.2 0.00.05 81.11.9 87.11.0 85.14.7 48.22.0 0.00.00 66.595.2 82.892.3 67.8100.0 28.171.8 N/Aof 6% or 8% respectively.Post-ML activity of wine LAB may cause heterofermentation of

sugars and decarboxylation of amino acids, leading to wine qualitydegradation by the production of acetic acid (Lonvaud-Funel, 1999),biogenic amines (Gonzalez-Marco and Ancin-Azpilicueta, 2006;Martin-Alvarez et al., 2006) and citrulline, the precursor to the carci-nogenic ethyl carbamate (Mira de Ordua et al., 2000). This worksuggests that a one week period following malic acid depletion maybe effective to reduce SO2 binders, while still minimizing the risk ofmetabolic transformations that are detrimental to overall wine qual-ity. In wines with a higher risk for such quality degradation (highpH, residual sugars), stabilization decisions should be based on themeasurement of acetaldehyde or bound SO2 levels (following benchtests with previously sulted wine). Finding the adequate timepoint for stabilization with SO2 and/or removal of residual lees iscritical for both the reduction of SO2 binders and maintenance ofwine quality.

5. Conclusions

Recently, a new method was presented and patented, thatwould allow the reduction of SO2 binding carbonyls in wines usingan insoluble resin (Blasi et al., 2008; Deleuze et al., 2007a, 2007b).

The method seems to have been especially intended for wines fromSauternes, which tend to have high SO2 binder concentrations andSO2 requirements. So far, the method has not been scientically scru-tinized and is not permitted. The current work shows that within thecurrently legal winemaking tools, MLF remains likely the most effec-tive method to achieve reduced wine SO2 additions. Where suitableby wine type and style, MLF can be considered within the strategiesto improve SO2 utilization.

Appendix A. Supplementary data

Supplementary data to this article can be found online at doi:10.1016/j.ijfoodmicro.2012.01.025.

Hughner, R.S., McDonagh, P., Prothero, A., Shultz, C.J., Stanton, J., 2007. Who are organicfood consumers? A compilation and review of why people purchase organic food.Journal of Consumer Behaviour 6, 94110.

Lonvaud-Funel, A., 1999. Lactic acid bacteria in the quality improvement and depreci-ation of wine. Antonie van Leeuwenhoek International Journal of General andMolecular Microbiology 76, 317331.

Maicas, S., Ferrer, S., Pardo, I., 2002. NAD(P)H regeneration is the key for heterolacticfermentation of hexoses in Oenococcus oeni. Microbiology 148, 325332.

Main, G.L., Morris, J.R., 1991. Color of Riesling and Vidal wines as affected by bentonite,Cufex, and sulfur dioxide juice treatments. American Journal of Enology andViticulture 42, 354357.

Martin, L.B., Nordlee, J.A., Taylor, S.L., 1986. Sulte residues in restaurant salads. Journalof Food Protection 49, 126129.

Martin-Alvarez, P.J., Marcobal, A., Polo, C., Moreno-Arribas, M.V., 2006. Inuence oftechnological practices on biogenic amine contents in red wines. European FoodResearch and Technology 222, 420424.

Mayer, K., Pause, G., Vetsch, U., 1976. Gehalte an SO2-bindenden Stoffen in Wein:Einuss von Grung und biologischem Sureabbau. Schweizerische Zeitschrift frObst- und Weinbau 112, 309313.

McWilliams, D.J., Ough, C.S., 1974. Measurement of ammonia in musts and wines usinga selective electrode. American Journal of Enology and Viticulture 25, 6772.

157J.N. Jackowetz, R. Mira de Ordua / International Journal of Food Microbiology 155 (2012) 153157Azabagaoglu, M.O., Akyol, A., Ozay, A., 2007. The demand for organic wines and organicwine marketing. Journal of Environmental Protection and Ecology 8, 171178.

Bartowsky, E.J., Henschke, P.A., 2000. Management of malolactic fermentation for thebuttery diacetyl avour in wine. Australian Grapegrower & Winemaker 438a,5867.

Berg, J.M., Tymoczko, J.L., Stryer, L., 2007. Biochemistry. W. H. Freeman and Company,New York.

Blasi, M., Barbe, J.C., Dubourdieu, D., Deleuze, H., 2008. New method for reducing thebinding power of sweet white wines. Journal of Agricultural and Food Chemistry56, 84708474.

Carret, R., Vidal, M.T., Bordons, A., Constanti, M., 2002. Inhibitory effect of sulfur dioxideand other stress compounds in wine on the ATPase activity of Oenococcus oeni. FEMSMicrobiology Letters 211, 155159.

Danilewicz, J.C., 2003. Review of reaction mechanisms of oxygen and proposed inter-mediate reaction products in wine: central role of iron and copper. AmericanJournal of Enology and Viticulture 54, 7385.

Deleuze, H., Blasi, M., Barbe, J.C., Dubourdieu, D., Maillard, B., 2007a. Extraction ofcarbonyl compounds from beverages using an inert solid support. French PatentNr. FR2898891.

Deleuze, H., Blasi, M., Barbe, J.C., Dubourdieu, D., Maillard, B., 2007b. Process forextracting carbonylated compounds from a drink by liquidsolid extraction witha functionalized inert support. World Patent WO 2007/107603 A1.

Diaz Maroto, C., Garcia, E., Hermosin, I., Cabezudo, M.D., 2001. Changes in Cencibel(Tempranillo) red wines following malolactic fermentation. Alimentaria 321,143151.

Doyle, M., Beuchat, L., 2007. Food Microbiology: Fundamentals and Frontiers. ASMPress, Washington, DC.

Dukes, B.C., Butzke, C.E., 1998. Rapid determination of primary amino acids in grapejuice using an o-phtalaldehyde/N-acetyl-L-cysteine spectrophotometric assay.American Journal of Enology and Viticulture 49, 127134.

Flamini, R., de Luca, G., di Stefano, R., 2002. Changes in carbonyl compounds inChardonnay and Cabernet Sauvignon wines as a consequence of malolacticfermentation. Vitis 41, 107112.

Francioli, S., Buxaderas, S., Pellerin, P., 1999. Inuence of Botrytis cinerea on the poly-saccharide composition of Xarel.lo musts and cava base wines. American Journalof Enology and Viticulture 50, 456460.

Gonzalez-Marco, A., Ancin-Azpilicueta, C., 2006. Inuence of lees contact on evolutionof amines in Chardonnay wine. Journal of Food Science 71, C544C548.

Hegazi, F.Z., Abo-Elnaga, I.G., 1980. Degradation of organic acids by dairy lactic acidbacteria. Zentralblatt fuer Bakteriologie 2. Abteilung 135, 212222.

Henick-Kling, T., 1993. Malolactic fermentation. In: Fleet, G.H. (Ed.), Wine Microbiologyand Biotechnology. Harwood Academic Publishers, Chur, Switzerland, pp. 289326.

Hugenholtz, J., 1993. Citrate metabolism in lactic acid bacteria. FEMS MicrobiologyReviews 12, 165178.Mira de Ordua, R., Liu, S.-Q., Patchett, M.L., Pilone, G.J., 2000. Ethyl carbamate precursorcitrulline formation from arginine degradation bymalolactic wine lactic acid bacteria.FEMS Microbiology Letters 183, 3135.

Nielsen, J.-C., Richelieu, M., 1999. Control of avour development in wine during andafter malolactic fermentation by Oenococcus oeni. Applied and EnvironmentalMicrobiology 65, 740745.

Osborne, J.P., Mira de Ordua, R., Liu, S.-Q., Pilone, G.J., 2000. Acetaldehyde metabolismby wine lactic acid bacteria. FEMS Microbiology Letters 191, 5155.

Osborne, J.P., Dub Morneau, A., Mira de Ordua, R., 2006. Degradation of free andsulphur-dioxide-bound acetaldehyde by malolactic lactic acid bacteria in whitewine. Journal of Applied Microbiology 101, 474479.

Papazian, R., 1996. Sultes: safe for most, dangerous for some. FDA Consumer Magazine30, 1015.

Radler, F., 1986. Microbial biochemistry. Experientia 42, 884893.Rankine, B.C., 1968. The importance of yeast in determining the composition and

quality of wines. Vitis 7, 2249.Ribreau-Gayon, P., Dubourdieu, D., Donche, B., Lonvaud-Funel, A., 1998a. Trait

d'Oenologie, I. Dunod, Paris.Ribreau-Gayon, P., Glories, Y., Maujean, A., Dubourdieu, D., 1998b. Trait d'Oenologie,

II. Dunod, Paris.Richter, T., Padel, S., 2007. The world of organic agriculture. Statistics and Emerging

Trends 2006. International Federation of Organic Agriculture Movements(IFOAM) & Research Institute of Organic Agriculture (FiBL). Last accessed: 15.Sept. 2011.

Snelten, H.J., Schaafsma, G., 1992. Health aspects of oral sulphite, and sulphite in wine.Voeding 53, 8890.

Sponholz, W.R., Dittrich, H.H., 1985. Origin of gluconic, 2-oxo-gluconic and 5-oxo-gluconic, glucuronic and galacturonic acids in musts and wines. Vitis 24, 5158.

Stolz, H., Schmid, O., 2008. Consumer attitudes and expectations of organic wine. Lastaccessed: 15. Sept. 2011.

Vally, H., Carr, A., El Saleh, J., Thompson, P., 1999. Wine-induced asthma: a placebo-controlled assessment of its pathogenesis. The Journal of Allergy and ClinicalImmunology 103, 4146.

Wedzicha, B.L., Bellion, I., Goddard, S.J., 1991. Inhibition of browning by sultes. Nutri-tional and Toxicological Consequences of Food Processing 289, 217236.

Zaunmuller, T., Eichert, M., Richter, H., Unden, G., 2006. Variations in the energy metabo-lism of biotechnologically relevant heterofermentative lactic acid bacteria duringgrowth on sugars and organic acids. Applied Microbiology and Biotechnology 72,421429.

Zhang, C., Ganzle, M.G., 2010. Metabolic pathway of alpha-ketoglutarate in Lactobacillussanfranciscensis and Lactobacillus reuteri during sourdough fermentation. Journalof Applied Microbiology 109, 13011310.References

Metabolism of SO2 binding compounds by Oenococcus oeni during and after malolactic fermentation in white wine1. Introduction2. Materials and methods2.1. Microorganisms and chemicals2.2. Grape must and general inoculation procedures2.3. Sampling2.4. Analytical methods and statistical analysis2.5. Chemicals and preparation of reagents for SO2 binder analysis2.6. Derivatization procedure2.7. HPLC Analysis2.8. Statistical analysis

3. Results4. Discussion5. ConclusionsAppendix A. Supplementary dataReferences