Continuous pulsed electric field treatments’ impact on the microbiota of red Tempranillo wines aged in oak barrels

Continuous pulsed electric field treatments’ impact on the microbiota of red Tempranillo wines aged in oak barrels

Food Bioscience xxx (xxxx) xxx–xxx Contents lists available at ScienceDirect Food Bioscience journal homepage: www.elsevier.com/locate/fbio Continu...

NAN Sizes 0 Downloads 0 Views

Food Bioscience xxx (xxxx) xxx–xxx

Contents lists available at ScienceDirect

Food Bioscience journal homepage: www.elsevier.com/locate/fbio

Continuous pulsed electric field treatments’ impact on the microbiota of red Tempranillo wines aged in oak barrels ⁎

Lucía González-Arenzanaa, Isabel López-Alfaroa, , Ana Rosa Gutiérreza, Noelia Lópezb, Pilar Santamaríaa, Rosa Lópeza a

Instituto de Ciencias de la Vid y del Vino, ICVV (Gobierno de La Rioja, Universidad de La Rioja and CSIC), Finca La Grajera, Ctra. Burgos, Km 6, 26007 Logroño, La Rioja, Spain b Centro Nacional de Tecnología y Seguridad Alimentaria (CNTA), Ctra. NA 134, km. 53, 31570 San Adrián, Navarra, Spain

A R T I C LE I N FO

A B S T R A C T

Keywords: Continuous flow Wine Oak barrel ageing Brettanomyces/Dekkera Lactic acid bacteria

The effects of pulsed electric field (PEF) technology on the microbiota of wines aging in oak barrels without additional chemical preservatives were studied. Two differently aged wines without sensory deviations, naturally having Brettanomyces/Dekkera, were PEF-treated. Results showed that the PEF treatment in continuous flow inactivated the most abundant spoilage microbial population of each wine. However, the recovery of some microorganisms after treatment was also observed. Therefore, to determine PEF as an alternative to the use of SO2, additional studied are needed with a wider sampling set.

1. Introduction The use of oak barrels to age Rioja wines has led to improvement of wine composition, flavor and even microbial stability (Rubio et al., 2015). Oak barrels need to be disinfected between uses although this is not easy due to the porous structure of wood. This provides niches for the survival of some spoilage microorganisms. These microorganisms belong to lactic acid bacteria (LAB), acetic acid bacteria (AAB) and yeast genera. On the other hand, few of the LAB species involved in winemaking are really important for the malolactic fermentation (MLF) development. In contrast, strains of other species such as Pediococcus and even Lactobacillus genera are related to wine spoilage because of the production of biogenic amines or exopolysaccharides (Beneduce, Spano, Vernile, Tarantino, & Massa, 2004; Walling, Gindreau, & Lonvaud-Funel, 2001). Having an oxidative metabolism with wine that enhances volatile acidity and unpleasant odors such as acetic acid, the AAB presence in wine has always been considered negative for wine quality (Bartowsky & Henschke, 2008). Among the yeasts, the Brettanomyces/Dekkera genera are the most undesirable as they produce ethylphenols, which are responsible for unpleasant odors such as animal leather or black olive. (Romano, Perello, Lonvaud-Funel, Sicard, & de Revel, 2009). These yeasts tolerate high ethanol content in wines and are even able to resist the commonly used doses of SO2 (Longin et al., 2016). Other non-Saccharomyces yeasts may have either a



negative or positive influence on wine quality (Sun, Gong, Jiang, & Zhao, 2014). The facilities, tools and tanks of wineries are not sterile (Garijo et al., 2009; Ocón, Gutiérrez, Garijo, López, & Santamaría, 2010). In effect, not only are the oak barrels the origin of spoilage microorganism contamination, but also the alcoholic fermentation (AF) and the MLF. Spoilage microorganisms can grow even during wine aging (Wedral, Shewfelt, & Frank, 2010). The wine industry uses several methods to prevent spoilage with SO2 being the most common (Ribéreau-Gayon, Dubourdieu, Donèche, & Lonvaud-Funel, 2007). Despite the effectiveness and utility of this compound, it causes allergic reactions and sensory deviations of wines when used with incorrect doses. Alternative new technologies are being tested to reduce the SO2 doses used in winemaking (Guerrero & Cantos-Villar, 2014). Some of these new technologies have shown variable effectiveness in inactivating diverse wine-borne microorganisms (Yang, Huang, Lyu, & Wang, 2016). Pulsed electric field (PEF) technology leads to the electroporation of the cells, thus other authors studied the possibility of applying PEF to ageing to increase the phenolic extraction in grapes and must, achieving promising results (Garde-Cerdán, Marsellés-Fontanet, Arias-Gil, Ancín-Azpilicueta, & Martín-Belloso, 2008; Puértolas, Saldaña, Álvarez, & Raso, 2010). Previous research based on optimization of the PEF technique showed the efficiency of a PEF-treatment designed to electroporate the cells of some wine-borne microorganisms inoculated in a previously

Corresponding author. E-mail address: [email protected] (I. López-Alfaro).

https://doi.org/10.1016/j.fbio.2018.10.012 Received 17 January 2018; Received in revised form 19 October 2018; Accepted 21 October 2018 2212-4292/ © 2018 Published by Elsevier Ltd.

Please cite this article as: González-Arenzana, L., Food Bioscience, https://doi.org/10.1016/j.fbio.2018.10.012

Food Bioscience xxx (xxxx) xxx–xxx

L. González-Arenzana et al.

(glucose yeast peptone medium) (20 g l−1 glucose (Scharlab SL., Barcelona, Spain), 5 g l−1 yeast extract (Panreac AppliChem), 5 g l−1 mycological peptone (Biolife Italiana SL., Milano, Italy), 20 g l−1 agar (Panreac AppliChem) and 75 mg l−1 chloramphenicol (Sigma, St. Louis, MO, USA)] (Rodrigues et al., 2001). Moreover, samples were spread on plates with an aerobic atmosphere at 30 °C for two days on MRS (Man Rogosa agar) for detecting LAB (52 g l−1 MRS broth (Scharlab SL.), 20 g l−1 agar and 50 mg l−1 pymaricine (VGP SL., Barcelona, Spain)) and on modified MRS (M30) anaerobically (BDGasPak™ Becton, Dickinson and Co., Wickoff, NJ, USA) for 10 days at 30 °C (52 g l−1 MRS broth, 6 g l−1 fructose (Panreac Applichem), 5 g l−1 D,L malic acid (Panreac AppliChem), 0.5 g l−1 HCl cysteine (Sigma), 100 ml l−1 tomato juice prepared in the laboratory from fresh tomatoes (as a source of pantothenic acid, which is needed to help the Oenococcus oeni grow), 20 g l−1 agar and 50 mg l−1 pymaricine (VGP SL.), which also helps O. oeni growth). In the case of AAB, Mann culture media plates were incubated with an aerobic atmosphere at 25 °C for two days (25 g l−1 D-mannitol (Sigma), 3 g l−1 bacteriological peptone (Panreac AppliChem), 5 g l−1 yeast extract, 20 g l−1 agar and 3 U ml−1 penicillin (Sigma)).

sterilized wine (González-Arenzana, Portu et al., 2015) and in wines treated just after AF (González-Arenzana et al., 2018). Recently, Delsart et al. (2016) described favorable inactivations for PEF treatments applied in a static chamber with commercially aged wines inoculated with three spoilage microorganisms. With this background, testing the PEF effects with real wines holding natural spoilage microorganisms, before and during ageing and after bottling might provide a possible alternative procedure. A continuous flow system was used. This study was done to evaluate the changes produced with PEF in two wines ageing in oak barrels naturally infected with Brettanomyces/Dekkera. 2. Materials and methods 2.1. Sampling Samples obtained from a total of 54 oak barrels (220 L) from the experimental winery of the ICVV Research Centre of the Spanish Northern Region of La Rioja, were inoculated into Sniff-Brett® liquid medium (Intelli'oeno, Les Gouvernaux, Chabeuil, France) to determine their natural and contaminant population of Brettanomyces/Dekkera as described by Portugal et al. (2015). This liquid culture media allows the prediction of CFU/ml potentially able to produce volatile phenol compounds by these spoilage yeasts. In 27 oak barrels the Brettanomyces/Dekkera population was not found, in 21 oak barrels the Brettanomyces/Dekkera population was predicted to be very low, between 5 and 50 colony forming units/ml (CFU/ml). Only in 6 oak barrel were the estimated Brettanomyces/Dekkera population > 104 CFU/ml, which could spoil wine quality. A sensorial analysis of visual, olfactory (intensity and quality) and taste (intensity and quality) properties, was done with these 6 wines using professional testers who had sensory faculties for describing color, taste, smell or feel of wines. Only in two wines (referred to as wine 1 and wine 2) were there no pre-existing sensory properties related to spoilage. After the MLF, wine 1 has been stored in the barrel for two years whereas wine 2 has been stored only 5 months. They were each distributed into 6 stainless steel tanks of 5 l. The experiment was done in triplicate, so that 3 stainless tanks of each wine served as the control while the others were PEF-treated. Sampling was done at 4 stages; the first one was immediately after PEF treatments. Five months after applying the PEF treatment, the second sampling was carried out. After this stage, wine from control and PEF-treated replicates were bottled. The third sampling was 9 months after PEF treatment and the fourth sampling was 25 months after the PEF treatment (Supplementary Fig. 1).

2.3.2. Counts and genera identification After incubation, the CFU ml−1 were counted with each control and PEF-treated replicated of each wine. Then, the average count and standard deviation for each sample was determined and the results were then expressed in logarithmic units. In samples with counts higher than one log unit, 5 colonies were isolated for identification. For that purpose, the DNA of each isolate was obtained and molecular biology methods allowed the genera identification. For yeast identification, partial 26S rRNA genes were amplified using the primers NL1 and NL4 (Cocolin, 2000); for LAB identification, the PCR was done with primer pairs WLAB1 and WLAB2 targeted the V4 and V5 16S rDNA regions and for AAB identification amplification of the V7 to V8 region of 16S rDNA gene with WBAC1 and WBAC2 primers described by Lopez et al. (2003). The PCR amplicons were sequenced by Macrogen Inc. (Seoul, South Korea) and then sequences were compared to those in the GenBank database (https://www.ncbi.nlm.nih.gov/nucleotide/) using the basic local alignment search tool (BLAST) (https://blast.ncbi.nlm.nih. gov/Blast.cgi?PROGRAM=blastn&PAGE_TYPE=BlastSearch&LINK_ LOC=blast) to obtain the correct identification (Altschul, Gish, Miller, Myers, & Lipman, 1990). All the information about the primers can be found in Lopez et al. (2003). 2.4. PEF equipment and treatments An ELCRACK-HVP5 PEF (DIL, German Institute of Food Technologies, Quakenbrück, Germany) was used. The treatment was chosen as the best optimized PEF-treatments for the inactivation of ~2.5 log units of 10 yeast species involved in winemaking, 3.5 log units of 12 LAB and 4 log units of AAB (González-Arenzana, Portu et al., 2015; González-Arenzana et al. 2018). The treatment was done using continuous flow (12 ± 0.4 l h−1) with a peristaltic pump (Millipore, Bedford, MA, USA) and calibrated with wine before each use. The device was sterilized before treatments to avoid external contaminations using a circulating solution of 10% sodium hypochlorite for 3 min and sterile water for 10 min. The circuit configuration generated square bipolar pulses. The PEF treatment chamber, developed by Dr. Álvarez and Dr. Raso (University of Zaragoza, Aragón, Spain) was collinear with a 0.45 cm diameter and 0.56 cm between electrodes. A cooling jacket was connected to the equipment to quickly cool down the wine after leaving the electrode system. The inlet temperature was always 18 °C and the outlet temperature was always < 22 °C. The electric field strength was assessed numerically using the finite element method, with the COMSOL Multiphysics software (COMSOL Inc., Stockholm, Sweden) in the core of the treatment chamber. Thus, the specific electric field strength in the center point of the treatment zone (Toepfl,

2.2. Oenological analysis of initial wines The two selected wines were characterized physically and chemically by determining the alcohol by volume (ABV), volatile acidity, total SO2 and pH, according to the European Community Official Methods Protocols (1990). 2.3. Microbiological analysis 2.3.1. Culture media At each sampling, serial decimal dilutions of the replicates (n = 3) of the control and PEF-treated wine samples were plated on 5 different culture media to detect viable and cultivable cells. The DBDM (Dekkera and Brettanomyces differential medium) plates used an anaerobic atmosphere and were incubated at 25 °C for 21 days for Brettanomyces and Dekkera detection (Rodrigues, Gonçalves, Pereira-da-Silva, Malfeito-Ferreira, & Loureiro, 2001) (6.7 g l−1 yeast nitrogen base (Sigma), 60 ml l−1 ethanol (Panreac AppliChem, Darmstadt, Germany), 10 mg l−1 cycloheximide (Sigma), 100 mg l−1 p-coumaric (Sigma), 22 mg l−1 bromocresol green (Sigma) and 20 g l−1 agar (Panreac Aplichem)). Yeasts were grown aerobically for two days at 25 °C on GYP 2

Food Bioscience xxx (xxxx) xxx–xxx

L. González-Arenzana et al.

Heinz, & Knorr, 2007) was 23 kV cm−1 and the specific energy was 95 ± 3 KJ Kg−1, with a pulse width of 8 µs, a repetition rate of 330 Hz and 17.6 s of effective time of PEF treatment.

Table 1 Initial characteristics of wine 1 and wine 2, before the PEF treatment. Characterising

2.5. Volatile composition Ageing period (months) Predicted CFU ml−1 Brettanomyces bruxellensis (Sniff-Brett®) ABV Volatile acidity (mg l−1) Total SO2 (mg l−1) pH

The volatile composition of samples was assessed at 25 months using continuous liquid-liquid extraction for 1 h (Ortega, López, Cacho, & Ferreira, 2001) with some modifications. Dichloromethane used for gas chromatography and gradient grade ethanol for liquid chromatography were from Merck (Darmstadt, Germany). The reference compounds 4-ethylguayacol and 4-ethylphenol with purities of 98% and 99%, respectively, were from Sigma. Water was obtained from a Milli-Q purifying system (Millipore, Darmstadt, Germany). An internal standard solution was prepared for quantification in 100 ml of ethanol with 40 mg of 2-octanol (Sigma). Four ml of wine samples or of the internal standard solution, 4.5 g of (NH4)2SO4, 7 ml of water, 15 µl of internal standard solution and 0.2 ml of dichloromethane were added into 15 ml screw-capped centrifuge tubes. The tube was shaken for 1 h at 50 rpm and then centrifuged at 5000 g for 10 min in a centrifuge (5810R, Eppendorf, Hamburg, Germany). Once the phases had separated, the dichloromethane phase was recovered and 2 µl injected into a HewlettPackard (Palo Alto, CA, USA) 6890 Series II Gas Chromatograph equipped with an automatic injector (6890 Series Injector, Agilent Santa Clara, CA, USA) and a Hewlett-Packard FID detector. Separation was carried out with a DB-Wax column (60 m × 0.32 mm i.d., J&W Scientific, Folsom, CA, USA). The injection was in split mode. The temperature program was as follows: 40 °C for 5 min, then raised at 3 °C min−1 to 220 °C. Carrier gas was N2 at 2 ml min−1. Split flow was 30 ml min−1, injector temperature was 220 °C and detector temperature was 280 °C. Identification of the two compounds of interest was carried out by comparing their retention times with those of pure reference standards and using a Hewlett-Packard GCD Series II Gas Chromatograph Electron Ionization Detector with the same chromatographic conditions. Quantification of volatile compounds was done with an internal standard method because not all samples were assumed to have the same response on a weight or molecule basis using the software provided with the instrument.

Wine 1

2

24 104 14.5 0.85 50 3.6

5 104 13.6 0.69 86 3.8

2.6. Statistical analyses The log CFU ml−1 of control and PEF treated samples at the same and at different sampling stages were statistically analyzed, for each one of the three control and PEF-treated replicates (n = 3) of wine 1 and wine 2. No letters mean no significant differences. For that, the statistical software SPSS (IBM® SPSS Statistic version 23, Armonk, NY, USA) was used and the standard deviation and the analyses of the variance (ANOVA) were calculated. Differences were significant when p value was < 0.05. For the statistical analysis of the concentration of aromatic compounds in control and PEF-treated samples the same statistical software was used with the student t-test. Fig. 1. Average counts (CFU ml−1) and standard deviation (error bars) of the three replicates in logarithmic units done individually for control (C in dark grey) and PEF-treated (PEF in light grey) wines 1 and 2, and in different culture media (A: DBDM for Brettanomyces/Dekkera growth, B: GYP for total yeast growth, C: MRS for LAB growth and D: M30 for O. oeni growth) after PEF treatment. M mean months. Different capital letters mean significant differences (p < 0.05) between control and PEF treated samples at the same sampling time. Different lower case letters mean significant differences (p < 0.05) between samples at different sampling times. No letters mean no significant differences.

3. Results The characteristics of initial wines 1 and 2 before the PEF treatment are shown in Table 1. Wine 1 had been ageing in the oak barrel for 24 months while wine 2 had aged 5 months in the oak barrel. In both wines, the initially predicted Brettanomyces/Dekkera bruxellensis population was ~104 CFU ml−1. Wine 1 had the higher ABV and volatile acidity while wine 2 had the highest SO2 and pH. The average log CFU ml−1 and the standard deviation obtained using different culture media are shown in Fig. 1. Average DBDM counts for wine 1 controls were ~2 log units at the beginning. After 5 months, the population of this control was similar and after 9 months it did not vary at all. A non-significant increase in the population was 3

Food Bioscience xxx (xxxx) xxx–xxx

L. González-Arenzana et al.

observed at 25 months (lower case letters in Fig. 1A). No colonies were found immediately after the PEF-treatment for wine sample 1 DBDM plates (capital letters in Fig. 1A). Nevertheless, 5 months later, the viable population was 1.7 log units. The viable population, 9 and 25 months after the PEF treatment, was higher than one log unit. The differences between control and PEF-treated samples were only significant in the initial sampling immediately after the PEF treatment. The DBDM control and PEF-treated wine 2 had no viable population growing until the last sampling with the control being non-significantly higher than the PEF-treated. The colonies of wine 1 control with GYP were < 1 log unit. After the PEF treatment, no significant population was found. A significant decrease was only observed in control wine 1 between the initial sampling and sampling 5 months later (lower case letters Fig. 1B). The control GYP of wine 2, had no significant population but 9 months later, the population increased to > 1 log units, but no GYP growth was found at 25 months. Wine 2 GYP after PEF treatment was < 1 log unit although all GYP differences were non-significant. Initially, no colonies were found on MRS for control wine 1 and remained < 1 log units throughout. After PEF treatment of wine 1, no colonies were detected throughout. For control wine 2, no colonies were detected with MRS initially, but at 5 months 5 log units were observed. The counts 4 months later were similar and decreased at the final sampling stage (lowercase letters in Fig. 1C). Initially after the PEF-treatment, no colonies were detected with wine 2; however, the counts increased at 5 and 9 months, and decreased at the final sampling stage. Significant differences were observed at 5 months after PEF treatment (capital letters in Fig. 1C). A similar result was obtained with culture media M30. The control wine 1 control samples had no initial population. After PEF treatment, no colonies were found with wine 1 throughout the sampling. Control wine 2 had 1.5 log units which increased over time. Wine 2 directly after the PEF treatment had no colonies, which was significant (capital letters in Fig. 1D). Five months later the population was 1.3 log units which was significantly different from the wine 2 control (lowercase in Fig. 1D). After that, the wine 2 controls and treated samples were not significantly different. No growth was observed on the mannitol plates (data not shown). Table 2 shows the genera identified during the current study. Three genera of yeasts and three of LAB were found initially in control wine 1. At 5 months, Brettanomyces and Dekkera were found in the control samples while only Dekkera were found with PEF treatment. At 9 month only Brettanomyces was found with PEF treatment and this also was the case at 25 months. Control wine 2 initially had the genera Oenococcus while nothing was found after PEF treatment. The control samples 5 months had Lactobacillus and Pediococcus and the PEF-treated had only Pedicoccus. In the control samples after 9 months, the genera Pichia and Pediococcus were found while the PEF-treated still only had Pedicoccus. At 25

Fig. 2. Volatile phenol content (mg l−1) of control (C) and PEF-treated (PEF) replicates of wine 1 and 2, 25 months after the PEF treatment of A) 4-ethylguaicol; B) 4-ethylphenol and statistical analyses done individually for each wine. **Significant differences between content of C and PEF samples at the same time (p < 0.05).

months both the control and the PEF-treated samples had Pediococcus. The results for volatile phenols after 25 months are shown in Fig. 2. For wine 1, significant differences between control and PEF-treated samples were observed with lower concentrations of 4-ethylguaiacol and 4-ethylphenol after PEF treatment. For wine 2, no significant differences were observed. 4. Discussion Both wines had no sensory deviation due to spoilage microorganisms but both had Brettanomyces at around 104 CFU ml−1 as predicted using the Sniff-Brett® culture media, which made them interesting for applying PEF technology. Wine 1 was an appropriate oak barrel aged with a higher ABV content and high volatile acidity. In contrast, wine 2 had medium levels of ABV and volatile acidity and higher SO2 concentration. Their yeast and bacteria cultivable populations were also different.

Table 2 Genera of microorganisms detected (+) and not detected (−) in the control (C) and PEF-treated (PEF) replicates of wine 1 and wine 2 with counts higher than 1 log units, at different stages – just after PEF treatment (0 m), 5 months later (5 m), 9 months later (9 m), and 25 months later (25 m). Genera

Wine 1

Wine 2

0m

Yeasts Brettanomyces Dekkera Pichia Lactic acid bacteria Lactobacillus Oenococcus Pediococcus

5m

9m

25 m

0m

5m

9m

25 m

C

PEF

C

PEF

C

PEF

C

PEF

C

PEF

C

PEF

C

PEF

C

PEF

– + –

– – –

+ + –

– + –

+ + –

+ – –

+ + –

+ – –

– – –

– – –

– – –

– – –

– – +

– – –

+ – –

+ – –

– – –

– – –

– – –

– – –

– – –

– – –

– – –

– – –

– + –

– – –

+ – +

– – +

– – +

– – +

– – +

– – +

4

Food Bioscience xxx (xxxx) xxx–xxx

L. González-Arenzana et al.

phenols (Saez, Lopes, Kirs, & Sangorrín, 2011). This recovery was also confirmed after 25 months. Therefore, the ability of Pediococcus to survive PEF treatment, bottling, and ageing was shown. In previous studies several species of this genus had been found to be more resistant than others to PEF treatments (González-Arenzana, Portu et al., 2015), so the resistance of this LAB should be taken into consideration to avoid problems during ageing. Even cultivable Brettanomyces/Dekkera cells were found for first time with wine 2 and, consequently, resulted in its recovery in wine 2 similar to wine 1. Both wines were initially tested by several professional testers and did not have any remarkable sensory deviations although 4-ethylphenol and the 4-ethylguaicol content was analyzed. These two compounds are considered to be mainly responsible for wine spoilage odors during wine ageing. Considering that the threshold perception was 0.620 µg l−1 for 4-ethylphenol and 0.140 µg l−1 for 4-ethylguaiacol (Chatonnet, Dubourdie, Boidron, & Pons, 1992), the concentrations of both found in this experiment for control and PEF-treated samples could be negatively influencing the wines aromatic profile. However, wine 1 had the higher viable population of Brettanomyces/Dekkera, which might explain the ethylphenol synthesis in both control and PEFtreated samples. In contrast, a significant lower content of ethylphenols was observed in PEF-treated samples, which might reflect the significant initial inactivation of Brettanomyces/Dekkera populations in these same samples. Moreover, in wine 2 the ethylphenols synthesis of control and PEF treated samples could be caused by Brettanomyces/ Dekkera and also by other microorganisms. In fact, some LAB such as Pediococcus and some yeasts such as Pichia have been reported to be responsible for the formation of these types of off-odors (Saez et al., 2011; Silva, Campos, Hogg, & Couto, 2011). The ethylphenol content of this wine was lower than wine 1, probably because it was a younger wine; moreover, the content found in control and PEF-treated samples was not significantly different and did not seem to be influenced by the microbial inactivation of LAB after PEF at early sampling stages.

Yeast and LAB populations vary due to different nutrient concentrations and chemical preservatives which may explain the differences in both yeast and bacterial populations. Wine 1 had 2 log units of yeast and 1 log unit of bacteria. The viable yeast population was identified as Dekkera yeast, which is the anamorph form of Brettanomyces, which with adverse environmental conditions generates resistance cells to survive. This makes it potentially more dangerous than Brettanomyces (Romano, Perello, de Revel, & Lonvaud-Funel, 2008). Wine 2, on the other hand, had no yeasts and only a low population of culturable bacteria, mainly the LAB Oenococcus genera. This LAB is the main biological agent of MLF, so finding it after a recent MFL in oak barrels is a common result. Interestingly, wine 2 had an active population of Brettanomyces according to the Sniff-Brett® culture medium, but no yeast were found with DBDM plates. This result is in accordance with some authors such as Capozzi et al. (2016) who observed that viable Brettanomyces may not be able to grow on some culture media such as DBDM. Monitoring the microbial inactivation of PEF-treated samples of wine 1 comparing with control samples, a significant reduction of Dekkera could be observed. Five months later, in control samples not only Dekkera was found but also Brettanomyces which could be due to the transfer from the oak barrel to a smaller stainless steel tank and to the intrinsic aeration during transference as other authors have found (Garijo et al., 2008, 2009). At this stage, no significant differences between control and PEF-treated samples of wine 1 were obtained because Dekkera was recovered. 9 months after the beginning of the study, the microbiology of control and PEF-treated wine was similar to that described for the previous sampling. Thus, bottling must not have significantly affected the Brettanomyces/Dekkera population of control and PEF-treated wine 1. Finally, in the last sampling, a clear prevalence of the viable population of Brettanomyces/Dekkera in control samples and of Brettanomyces in PEF-treated samples was observed, increasing the population and even overtaking the initial population. The recovery of Brettanomyces/Dekkera was corroborated as also reported by other authors (Capozzi et al., 2016). Furthermore, their ability to survive PEF treatment, bottling and ageing was in accord with Coulon, Perello, Lonvaud-Funel, de Revel, and Renouf (2010) who pointed out the necessity of protecting wines from aeration in the cellar, and more before bottling. The microbial populations in control and PEF-treated samples just after the PEF treatment of wine 2 showed significant LAB inactivation and a very small reduction of yeasts. In effect, the LAB genus Oenococcus was not immediately found after the PEF treatment. Five months later, nearly 5 log units identified as LAB, Lactobacillus and Pediococcus, were found in control samples. The presence of Lactobacillus in this stage was unusual because these LAB are usually found at the initial or middle stages of spontaneous MLF as was observed in other studies, where these were not identified after MLF depletion (González-Arenzana, Santamaría, Gutiérrez, López, & LópezAlfaro, 2017). The genus Pediococcus sometimes has been found along with Brettanomyces/Dekkera (Smith & Divol, 2016) and is linked to wine spoilage because of the production of exopolysaccharides and therefore ropy wines (Bartowsky, 2009). Pediococcus is a LAB that has been found in wine in some cases after MLF (Lucas, Claisse, & Lonvaud-Funel, 2008). Both genera (Lactobacillus and Pediococcus) have been described as producers of biogenic amines which decrease wine quality (GarcíaRuiz, González-Rompinelli, Bartolomé, & Moreno-Arribas, 2011; Landete, Ferrer, & Pardo, 2007). The detection of some LAB species in controls before bottling could be linked to the changes in the environmental conditions such as the transfer of wine from oak barrel to a small steel tank or the unavoidable slight aeration and movement of lees. On the other hand, the PEF-treated samples had lower growth; in fact, only the LAB Pediococcus was found in the PEF-treated wines. 9 months after initial sampling, besides Pediococcus, the yeast Pichia was found. This yeast is a flor yeast that using an oxidative metabolism in wine when oxygen is present but it is also able to produce volatile

5. Conclusion The PEF treatment led to an immediate and significant microbial inactivation of most important spoilage microorganisms in aged wines, which also led to a significant reduction of volatile phenols in the wine with the highest Brettanomyces/Dekkera population. This would directly translated into a longer shelf life of wine until bottling without using additional preservatives. The recovering of some of the main microorganism involved in ageing, in both control and PEF-treated wines, suggested it would be beneficial to continue to study PEF treatment to learn more about the microbial, chemical and sensorial changes of ageing wines. Acknowledgements This work has been supported by the Instituto Nacional de Investigación y Tecnología Agraria y Alimentaria project RTA201100070-00-00 and the Navarra Government project IIQ14037.RI1 and by a pre-doctoral grant (Boletín Oficial del Estado 12th May 2012). Appendix A. Supporting information Supplementary data associated with this article can be found in the online version at doi:10.1016/j.fbio.2018.10.012. References Altschul, S. F., Gish, W., Miller, W., Myers, E. W., & Lipman, D. J. (1990). Basic local alignment search tool. Journal of Molecular Biology, 215(3), 403–410. Bartowsky, E. J. (2009). Bacterial spoilage of wine and approaches to minimize it. Letters in Applied Microbiology, 48(2), 149–156. Bartowsky, E. J., & Henschke, P. A. (2008). Acetic acid bacteria spoilage of bottled red wine – A review. Vinegars and Acetic Acid Bacteria, 125(1), 60–70 (2005).

5

Food Bioscience xxx (xxxx) xxx–xxx

L. González-Arenzana et al.

Microbiology, 69(11), 6801–6807. Lucas, P. M., Claisse, O., & Lonvaud-Funel, A. (2008). High frequency of histamine-producing bacteria in the enological environment and instability of the histidine decarboxylase production phenotype. Applied and Environmental Microbiology, 74(3), 811–817. Ocón, E., Gutiérrez, A. R., Garijo, P., López, R., & Santamaría, P. (2010). Presence of nonSaccharomyces yeasts in cellar equipment and grape juice during harvest time. Food Microbiology, 27(8), 1023–1027. Ortega, C., López, R., Cacho, J., & Ferreira, V. (2001). Fast analysis of important wine volatile compounds. Development and validation of a new method based on gas chromatographic-flame ionisation detection analysis of dichloromethane microextracts. Journal of Chromatography A, 923, 205–214. Portugal, C., Pinto, L., Ribeiro, M., Tenorio, C., Igrejas, G., & Ruiz-Larrea, F. (2015). Potential spoilage yeasts in winery environments: Characterization and proteomic analysis of Trigonopsis cantarellii. International Journal of Food Microbiology, 210, 113–120. Puértolas, E., Saldaña, G., Álvarez, I., & Raso, J. (2010). Effect of pulsed electric field processing of red grapes on wine chromatic and phenolic characteristics during aging in oak barrels. Journal of Agricultural and Food Chemistry, 58(4), 2351–2357. Ribéreau-Gayon, P., Dubourdieu, D., Donèche, B., & Lonvaud-Funel, A. (2007). Handbook of enology. The microbiology of wine and vinifications. 2. Hoboken, New Jersey, USA: John Willey and Son. Rodrigues, N., Gonçalves, G., Pereira-da-Silva, S., Malfeito-Ferreira, M., & Loureiro, V. (2001). Development and use of a new medium to detect yeasts of the genera Dekkera/Brettanomyces. Journal of Applied Microbiology, 90(4), 588–599. Romano, A., Perello, M. C., de Revel, G., & Lonvaud-Funel, A. (2008). Growth and volatile compound production by Brettanomyces/Dekkera bruxellensis in red wine. Journal of Applied Microbiology, 104(6), 1577–1585. Romano, A., Perello, M. C., Lonvaud-Funel, A., Sicard, G., & de Revel, G. (2009). Sensory and analytical re-evaluation of Brett character. Food Chemistry, 114(1), 15–19. Rubio, P., Garijo, P., Santamaría, P., López, R., Martínez, J., & Gutiérrez, A. R. R. (2015). Influence of oak origin and ageing conditions on wine spoilage by Brettanomyces yeasts. Food Control, 54(0), 176–180. Saez, J. S., Lopes, C. A., Kirs, V. E., & Sangorrín, M. (2011). Production of volatile phenols by Pichia manshurica and Pichia membranifaciens isolated from spoiled wines and cellar environment in Patagonia. Food Microbiology, 28(3), 503–509. Silva, I., Campos, F. M., Hogg, T., & Couto, J. A. (2011). Factors influencing the production of volatile phenols by wine lactic acid bacteria. International Journal of Food Microbiology, 145(2), 471–475. Smith, B. D., & Divol, B. (2016). Brettanomyces bruxellensis, a survivalist prepared for the wine apocalypse and other beverages. Food Microbiology, 59, 161–175. Sun, S. Y., Gong, H. S., Jiang, X. M., & Zhao, Y. P. (2014). Selected non-Saccharomyces wine yeasts in controlled multistarter fermentations with Saccharomyces cerevisiae on alcoholic fermentation behaviour and wine aroma of cherry wines. Food Microbiology, 44, 15–23. Toepfl, S., Heinz, V., & Knorr, D. (2007). High intensity pulsed electric fields applied for food preservation. Chemical Engineering and Processing: Process Intensification, 46(6), 537–546. Walling, E., Gindreau, E., & Lonvaud-Funel, A. (2001). Exopolysaccharide biosynthesis by Pediococcus damnosus strains isolated from wine: Elaboration of molecular detection tools. Lait, 81(1–2), 289–300. Wedral, D., Shewfelt, R., & Frank, J. (2010). The challenge of Brettanomyces in wine. Wine: Parameters Impacting Composition and Quality, 43(10), 1474–1479. Yang, N., Huang, K., Lyu, C., & Wang, J. (2016). Pulsed electric field technology in the manufacturing processes of wine, beer, and rice wine: A review. Food Control, 61, 28–38.

Beneduce, L., Spano, G., Vernile, A., Tarantino, D., & Massa, S. (2004). Molecular characterization of lactic acid populations associated with wine spoilage. Journal of Basic Microbiology, 44(1), 10–16. Capozzi, V., Di Toro, M. R., Grieco, F., Michelotti, V., Salma, M., Lamontanara, A., & Spano, G. (2016). Viable but not culturable (VBNC) state of Brettanomyces bruxellensis in wine: New insights on molecular basis of VBNC behaviour using a transcriptomic approach. Food Microbiology, 59, 196–204. Chatonnet, P., Dubourdie, D., Boidron, J., & Pons, M. (1992). The origin of ethylphenols in wines. Journal of the Science of Food and Agriculture, 60(2) (no pages). Cocolin, L. (2000). Direct profiling of the yeast dynamics in wine fermentations. FEMS Microbiology Letters, 189(1), 81–87. Coulon, J., Perello, M. C., Lonvaud-Funel, A., de Revel, G., & Renouf, V. (2010). Brettanomyces bruxellensis evolution and volatile phenols production in red wines during storage in bottles. Journal of Applied Microbiology, 108(4), 1450–1458. Delsart, C., Grimi, N., Boussetta, N., Miot Sertier, C., Ghidossi, R., Vorobiev, E., & Mietton Peuchot, M. (2016). Impact of pulsed-electric field and high-voltage electrical discharges on red wine microbial stabilization and quality characteristics. Journal of Applied Microbiology, 120(1), 152–164. European Community (1990). Community methods for the analysis of wine. Commission Regulation no. 26/76/90. Ninth Commission Regulation, 2676/9, 1–191. García-Ruiz, A., González-Rompinelli, E. M., Bartolomé, B., & Moreno-Arribas, M. V. (2011). Potential of wine-associated lactic acid bacteria to degrade biogenic amines. International Journal of Food Microbiology, 148(2), 115–120. Garde-Cerdán, T., Marsellés-Fontanet, A. R., Arias-Gil, M., Ancín-Azpilicueta, C., & Martín-Belloso, O. (2008). Effect of storage conditions on the volatile composition of wines obtained from must stabilized by PEF during ageing without SO2. Innovative Food Science & Emerging Technologies, 9(4), 469–476. Garijo, P., López, R., Santamaría, P., Ocón, E., Olarte, C., Sanz, S., & Gutiérrez, A. R. (2009). Presence of lactic bacteria in the air of a winery during the vinification period. International Journal of Food Microbiology, 136(1), 142–146. Garijo, P., Santamaría, P., López, R., Sanz, S., Olarte, C., & Gutiérrez, A. R. (2008). The occurrence of fungi, yeasts and bacteria in the air of a Spanish winery during vintage. International Journal of Food Microbiology, 125(2), 141–145. González-Arenzana, L., López-Alfaro, I., Garde-Cerdán, T., Portu, J., López, R., & Santamaría, P. (2018). Microbial inactivation and MLF performances of Tempranillo Rioja wines treated with PEF after alcoholic fermentation. International Journal of Food Microbiology, 269, 19–26. González-Arenzana, L., Portu, J., López, R., López, N., Santamaría, P., Garde-Cerdán, T., & López-Alfaro, I. (2015). Inactivation of wine-associated microbiota by continuous pulsed electric field treatments. Innovative Food Science & Emerging Technologies, 29, 187–192. González-Arenzana, L., Santamaría, P., Gutiérrez, A. R. A. R., López, R., & López-Alfaro, I. (2017). Lactic acid bacteria communities in must, alcoholic and malolactic Tempranillo wine fermentations, by culture-dependent and culture-independent methods. European Food Research and Technology, 243(1), 41–48. Guerrero, R. F., & Cantos-Villar, E. (2014). Demonstrating the efficiency of sulphur dioxide replacements in wine: A parameter review. Trends in Food Science & Technology, 42(1), 27–43. Landete, J. M., Ferrer, S., & Pardo, I. (2007). Biogenic amine production by lactic acid bacteria, acetic bacteria and yeast isolated from wine. Food Control, 18(12), 1569–1574. Longin, C., Degueurce, C., Julliat, F., Guilloux-Benatier, M., Rousseaux, S., & Alexandre, H. (2016). Efficiency of population-dependent sulfite against Brettanomyces bruxellensis in red wine. Food Research International, 89, 620–630. Lopez, I., Ruiz-Larrea, F., Cocolin, L., Orr, E., Phister, T., Marshall, M., & Mills, D. A. (2003). Design and evaluation of PCR primers for analysis of bacterial populations in wine by denaturing gradient gel electrophoresis. Applied and Environmental

6