Influence of inoculation time of an autochthonous selected malolactic bacterium on volatile and sensory profile of Tempranillo and Merlot wines

Influence of inoculation time of an autochthonous selected malolactic bacterium on volatile and sensory profile of Tempranillo and Merlot wines

International Journal of Food Microbiology 156 (2012) 245–254 Contents lists available at SciVerse ScienceDirect International Journal of Food Micro...

875KB Sizes 0 Downloads 16 Views

International Journal of Food Microbiology 156 (2012) 245–254

Contents lists available at SciVerse ScienceDirect

International Journal of Food Microbiology journal homepage: www.elsevier.com/locate/ijfoodmicro

Influence of inoculation time of an autochthonous selected malolactic bacterium on volatile and sensory profile of Tempranillo and Merlot wines Pedro Miguel Izquierdo Cañas a, b, Fátima Pérez-Martín c, Esteban García Romero a, Susana Seseña Prieto c,⁎, María de los Llanos Palop Herreros c a b c

Instituto de la vid y el Vino de Castilla-La Mancha, Crta. Toledo-Albacete, s/n, 13700, Tomelloso (Ciudad Real), Spain Parque Científico y Tecnológico de Albacete, Paseo de la Innovación 1, 02006 Albacete, Spain Departamento de Química Analítica y Tecnología de los Alimentos, Facultad de Ciencias del Medio Ambiente, Universidad de Castilla-La Mancha, Avda. Carlos III, s/n, 45071, Toledo, Spain

a r t i c l e

i n f o

Article history: Received 24 November 2011 Received in revised form 22 March 2012 Accepted 30 March 2012 Available online 5 April 2012 Keywords: Malolactic fermentation Co-inoculation Sequential inoculation Volatile compounds Red wine

a b s t r a c t A study was carried out to determine the effect of the inoculation time of the lactic acid bacteria (LAB) on the kinetic of vinification and on chemical and sensory characteristics of Tempranillo and Merlot wines. Traditional vinifications, with LAB inoculated after completion of AF, were compared with vinifications where yeast and bacteria were co-inoculated. Two commercial yeast strains and an autochthonous Oenococcus oeni strain (C22L9) previously identified and selected at our laboratory were used. Monitoring of alcoholic and malolactic fermentations was carried out by yeast and lactic acid bacteria counts and by measuring contents of glucose + fructose, malic acid and lactic acid. The implantation rate of O. oeni C22L9 was calculated by typing isolates obtained from count plates using the RAPD-PCR (Randomly Amplified Polymorphic DNA-Polymerase Chain Reaction) technique. Wines were chemically characterised and analysed for biogenic amine and volatile compound contents. A sensory analysis, consisting in a descriptive and a triangular test was also carried out. Results from this study showed that for both grape varieties, the concurrent yeast/bacteria inoculation of musts produced a significant reduction in duration of the process, without a pronounced degradation of malic acid during AF, nor an excessive increase in volatile acidity. Biogenic amine content was also lower in wines produced by co-inoculation. Important differences in volatile compound contents were observed, although there was little impact on the sensorial profile of wines. These results suggest that co-inoculation using O. oeni C22L9 is a worthwhile alternative compared to traditional post AF inoculation for Tempranillo and Merlot winemaking. © 2012 Elsevier B.V. All rights reserved.

1. Introduction Winemaking is a complex process frequently involving two successive fermentations; an alcoholic fermentation (AF), conducted by yeasts, and a subsequent malolactic fermentation (MLF) carried out by wine lactic acid bacteria. MLF results in deacification and microbial stabilization of wine, and in addition it produces changes at the organoleptic profile of wines which have important consequences for the final quality (Bauer and Dicks, 2004; Lonvaud-Funel, 1999). Both, AF and MLF may occur spontaneously from the activity of yeasts and bacteria naturally present in musts and wines, although in the last few years winemakers are starting to recognise the benefits of inoculating with dried commercial starter cultures of yeast and LAB for better control of how and when these fermentations take place. Timing of inoculation of starter cultures is an important factor influencing the success of induced fermentations, and various studies have been carried out to determine the effect of bacterial inoculation

⁎ Corresponding author. Tel.: + 34 925 265 716; fax: +34 925 268 840. E-mail address: [email protected] (S.S. Prieto). 0168-1605/$ – see front matter © 2012 Elsevier B.V. All rights reserved. doi:10.1016/j.ijfoodmicro.2012.03.033

time on vinification kinetics, chemical composition and sensory and sanitary attributes of wines (Abrahamse and Bartowsky, 2012; Alexandre et al., 2004; Antalick, 2010; Edwards et al., 1999; Henick-Kling and Park, 1994; Mendoza et al., 2011; Rosi et al., 2003). Results from some of these studies have shown that simultaneous yeast/bacteria inoculation poses important risks, such as the development of undesirable/ antagonistic interactions between the two microorganisms, stuck AF, interruption of AF before sugar depletion, wines with increased concentrations of acetic acid that render them unacceptable for consumption, or the production of possible off-odours. Consequently, simultaneous inoculation has not been a very common practice and the addition of bacterial starter culture after the completion of AF has been largely adopted by wineries. On the contrary, other authors (Krieger et al., 2007; Sieczkowski, 2004) have recommended simultaneous addition of the yeast and the bacteria to the must, on the basis of a better performance of the bacteria, due to the low alcohol concentration and the higher nutrient availability present in musts, and to better sensory characteristics of the wines. More recent studies carried out with different grape varieties (Azzolini et al., 2010; Jussier et al., 2006; Massera et al., 2009; Zapparoli et al., 2009) have reported a reduction in total fermentation time and better

246

P.M.I. Cañas et al. / International Journal of Food Microbiology 156 (2012) 245–254

control of the MLF, due to the early dominance of the inoculated bacterial strain. However, co-inoculation is not yet a very common practice and new and more exhaustive studies are needed to gain more knowledge of the influence of early bacterial inoculation on the winemaking process and on wine quality. The aim of this study was determine the influence of LAB inoculation time, and of the pair yeast/bacteria inoculated, on vinification kinetics and on the chemical composition and sensory characteristics of Tempranillo and Merlot wines. To that end traditional vinifications with LAB inoculated after completion of AF were compared with vinifications where yeast and bacteria were inoculated concurrently. Two commercial yeast strains and an autochthonous Oenococcus oeni strain, previously identified and selected at our laboratory (Ruiz et al., 2010) and never before assayed in co-inoculation with yeasts, were used in this study. Tempranillo wine was chosen because it is the most important red grape cultivar in Spain while Merlot is a low pH wine variety where growth of LAB and consequently the success of MLF is more difficult and unpredictable. During fermentation, the implantation rate of inoculated LAB was determined, and after MLF a sensory analysis and an exhaustive comparative analysis of the chemical composition of the resulting wines were carried out. To our knowledge, works reporting such comprehensive survey on the effect of timing of inoculation of the LAB in the vinification process of red wines have not been published to date. 2. Experimental 2.1. Microorganisms Saccharomyces cerevisiae strains VRB and VN and the O. oeni C22L9 were obtained as dried culture from Lallemand (Montreal, Canada). For each must the pairs VRB-O. oeni C22L9 and VN-O. oeni C22L9 were used in microvinification trials.

At the end of MLF wines were decanted and sulphited to reach 25 mg/L of free SO2 and then clarified, stabilised and filtered through 0.2 μm filters before bottling. 2.3. Microbiological analysis Samples were taken under aseptic conditions after inoculation and until the end of MLF. For yeast counts, plates of Malt Extract Agar (Cultimed, Barcelona, Spain) were incubated for 48 h at 28 °C. LAB counts were carried out by plating on MLO Agar (MLOA, Oenococcus oenos Medium) (Scharlab, Barcelona, Spain) supplemented with 10% (v/v) tomato juice, 50 mg/mL sodium azide and 100 mg/mL cycloheximide. Plates were incubated under anaerobic conditions (Gas Pack System, Oxoid Ltd., Basingstoke, UK) at 30 °C for 5 days. Counts were expressed as colony forming units (cfu) per mL of wine. In order to assess the implantation rate of the inoculated O. oeni C22L9, fifty isolated colonies from MLOA plates were picked at random from three stages during fermentation: after inoculation, at the middle of MLF (approximately 50% L-malic acid depleted) and at the end of MLF (L-malic acid content lower than 0.2 g/L). Colonies were purified by successive streaking on the same medium prior to typing by RAPDPCR as described below. DNA extraction for RAPD-PCR analysis was carried out as described by Rodas et al. (2003). A colony was suspended in 10 μL sterile milli-Q water and 1 μL of the suspension was used for PCR reactions. RAPDPCR reactions were performed in a total volume of 20 μL according to the procedure described by Ruiz et al. (2008). The genetic profiles of the isolates were compared with the profile of the inoculated strain and the implantation rate was calculated as the number of colonies with the specific RAPD-PCR profile divided by the total number of colonies picked, expressed as a percentage (RuizBarba et al., 1994). 2.4. Chemical analysis

2.2. Microvinifications Tempranillo and Merlot grapes from La Mancha wine region (Spain) were harvested during the 2009 vintage. Musts were obtained in the experimental cellar of the Institute of Vine and Wine of Castilla-La Mancha (IVICAM). The chemical composition of musts is shown in Table 1. 50 mg/L of SO2 were added to musts for sequential inoculation (SEQ) assays while 40 mg/L SO2 were added to those for co-inoculation (COI). It is compatible with bacterial growth provided that the bacteria are inoculated some hours after SO2 addition to allow the SO2 to combine to compounds such as the carbonyl compounds(Ribéreau-Gayon et al., 2006). Fermentations (in triplicate) were carried out following the standard protocols for Tempranillo and Merlot vinifications using 100 kg of grapes in each one. The yeasts, S. cerevisiae VRB or S. cerevisiae VN, and the malolactic bacteria O. oeni C22L9 were inoculated according to the manufacturer´s instructions. LAB was inoculated either after the completion of AF, when glucose+ fructose (G+ F) content was below 1 g/L (SEQ), or 24 h after yeast inoculation (COI), when free SO2 concentration was less than 10 mg/L. Table 1 Chemical composition of musts obtained during 2009 vintage.

Glucose + Fructosea Total acidityb pH L-malic acida Citric acida a b

g/L. g/L tartaric acid.

Tempranillo

Merlot

234.30 ± 1.40 3.99 ± 0.07 3.63 ± 0.06 2.55 ± 0.26 0.33 ± 0.19

233.30 ± 1.26 5.13 ± 0.09 3.51 ± 0.04 1.15 ± 0.29 0.13 ± 0.24

Musts and wines were chemically characterised by determining alcohol content, total acidity, expressed as tartaric acid, pH, volatile acidity, expressed as acetic acid, L-malic acid, L-lactic acid and citric acid, following the official analytical methods established by the International Organisation of Vine and Wine (OIV, 2009). The progress of AF was monitored for decline of glucose + fructose content (G+ F) and MLF was monitored following L-malic acid degradation. Both, glucose + fructose and L-malic acid contents were determined by enzymatic methods. Biogenic amine (histamine, tyramine, putrescine, cadaverine and phenylethylamine) contents in wines were determined by liquid chromatography (HPLC) following the method proposed by Gómez-Alonso et al. (2007). The compounds analysed were identified on the basis of the aminoenone derivative retention times of the corresponding standards (Sigma-Aldrich Chemie, Steinheim, Germany) and were quantified using the internal standard method. L-2-aminoadipic acid from Sigma was used as internal standard. 2.5. Volatile compound analysis Volatile compounds were analysed by GC/MS using a ThermoQuest GC2000 gas chromatograph and a DSQII mass detector with quadrupole analyser. All masses were obtained in electronic impact mode at 70 eV. In both methods, for major and minor compounds, a BP21 column (SGE) 50 m × 0.32 mm internal diameter and 0.25 mm thick of FFAP phase (polyethylenglycol treated with TPA) was used. For the major volatile compounds 200 mL of wine was steam distilled up to a volume of 200 mL. 1 μL of distilled wine with 4-methyl2-pentanol as internal standard was directly injected. The chromatographic conditions were as follows: carrier helium gas (1.7 mL/min,

P.M.I. Cañas et al. / International Journal of Food Microbiology 156 (2012) 245–254

247

split 1/25); injector temperature, 220 °C ; and oven temperature, 43 °C (5 min)–4 °C/min-100 °C–20 °C/min–190 °C–(1 min). For analysis of the minor volatile compounds, 500 mL of wine containing 100 μL of 10 g/L 4-nonanol as internal standard were extracted over 24 h with 250 mL of a 60:40 mixture of pentane dichloromethane. The extracts were concentrated by distillation in a Vigreux column to 2 mL and then kept at −20 °C until analysis. 2 μL of the extract was injected. The chromatographic conditions were: oven temperature, 43 °C (15 min)–2 °C/min–125 °C–1 °C/min–150 °C–4 °C/min–200 °C (45 min) and carrier gas helium (1.4 mL/min, split 1/15, splitless time 0.5 min). Separated compounds were identified by their mass spectra and their chromatographic retention times, using commercial products ACS grade from Sigma-Aldrich Chemie (Steinheim, Germany) and Merck (Darmstard, Germany) as a standard. Quantification was performed following the procedure described by Izquierdo et al. (2008a). 2.6. Sensory analysis Sensory analyses were performed to investigate the differences among treatments, always comparing within the wine variety. Both descriptive and triangular tests were carried out. Descriptive sensory analysis was performed by 14 selected tasters following the Sensory Profile method according to ISO Standard, 11035, 1994. The descriptors were scored on a scale of 0 to 8 (0 absence of the descriptor and 8 maximum intensity). The flavour attributes evaluated were aromatic intensity, red fruit, ripe fruit, spicy, floral, vegetable and dairy odour, and the taste attributes were acidity, astringency and body. A triangular test to evaluate the differences in aroma and taste was conducted in dark wine-tasting glasses to avoid judgements being influenced by the color of the wine. Sets containing 3 samples were analysed by 18 tasters in three different sessions, according to ISO Standard 4120, 1983. 2.7. Statistical analysis The paired Student t-test was used to determine whether there were significant differences between the results from chemical analysis. Multivariate data analysis (PCA) was used to obtain a more comprehensible overview of the chemical compounds and to investigate possible correlations amongst wines. The SPSS 12.0 software was used for both analysis.

Fig. 1. Yeasts counts from sequential inoculation and co-inoculation assays in Tempranillo (A) and Merlot (B) wines. Arrows indicate bacterial inoculation in sequential treatment. Values are mean of triplicates± SE.

3. Results and discussion 3.1. Evolution of microbial populations and fermentation kinetics Viable yeast populations (Fig. 1) followed a similar evolution at sequential and COI assays, until beginning of MLF. As reported by other authors (Massera et al., 2009; Mendoza et al., 2011), the presence of O. oeni C22L9 during active AF in COI assays seems not to influence the viable yeast population, since the counts obtained from both time inoculation assays were similar. On the contrary, evolution of LAB population (Fig. 2) was different for SEQ and COI assays. In COI assays, although with slight differences between must varieties, counts around 10 6 cfu/mL were obtained from the beginning of the assays and up to the end of AF, while in SEQ inoculation assays, counts of native LAB were lower until the inoculation of O. oeni C22L9. It is important to highlight that in SEQ inoculation assays a decrease in LAB population was observed after inoculation in both wine varieties, which was sharper in Merlot wine, probably as a consequence of the lower pH of this wine at the time of inoculation: 3.57 for Merlot wine and 3.86 for Tempranillo wine. At the end of MLF, LAB counts at both assays reached similar values.

Results for G + F analysis in SEQ assays (Fig. 3) showed that at both wine varieties a depletion of G + F, with concentrations for these compounds lower than 5 g/L, had already occurred when O. oeni C22L9 was inoculated. The trend in L-malic acid and L-lactic acid contents was similar in both wine varieties although in Merlot wine degradation of L-malic acid started 6 days after inoculation, thus lengthening the duration of the MLF. In both wine varieties, values around 0.1 g/L of malic acid, the level generally recognised as the threshold for complete MLF, were reached some days earlier by the pair VN/C22L9. In contrast with the report by Massera et al. (2009), results from malic acid analysis showed that native LAB present in the SEQ assays before inoculation of O. oeni C22L9, did not consume malic acid, since its concentration remained constant up to that moment. In COI assays (Fig. 4) there was a slow decrease in L-malic acid content and an increase in L-lactic acid, almost immediately following the inoculation. As reported by Jussier et al. (2006), results from G + F and malic acid analysis showed that the degradation of all these compounds was complete regardless of the time of inoculation of LAB (Figs. 3 and 4).

248

P.M.I. Cañas et al. / International Journal of Food Microbiology 156 (2012) 245–254

Fig. 2. LAB counts from sequential inoculation and co-inoculation assays in Tempranillo (A) and Merlot (B) wines. Arrows indicate bacterial inoculation in sequential treatment. Values are mean of triplicates± SE.

Fig. 3. Time courses of glucose + fructose and malic and lactic acid concentrations during AF and MLF of Tempranillo (A) and Merlot (B) wines obtained from sequential inoculation of VN or VRB yeasts and O. oeni C22L9. Glucose + fructose content is plotted against the right-hand axis. Values are mean of triplicates ± SE.

Major differences in overall fermentation (AF + MLF) duration were observed between SEQ and COI assays, with values ranging between 32 and 12 days, respectively (Figs. 3 and 4). For each must variety, the time required for sugar concentration to fall below 1 g/L and L-malic acid concentration below 0.1 g/L, was much shorter when COI was used. The length of MLF itself, measured as time elapsing between LAB inoculation and depletion of malic acid, was longer in SEQ assays than the total fermentation time in COI treatments. These results are consistent with reports by some authors (Azzolini et al., 2010; Jussier et al., 2006; Massera et al., 2009) and they are significant from a technological point of view because of the shorter time required to conclude the vinification process and the early microbiological stability conferred on wines. A total of 600 isolates were obtained from countable MLOA plates from wine samples taken during fermentation in all the assayed microvinifications. A cluster analysis of the RAPD-PCR patterns obtained for the isolates from each stage and that of the inoculated strain was carried out (data not shown). The value for the similarity coefficient (r = 80%) obtained from the reproducibility study was applied to group the isolates and calculate the implantation value corresponding to each stage. As reported by other authors (Combina et al., 2008; Sturm et al., 2008) implantation of O. oeni C22L9 was 100% in all samples taken after LAB inoculation. These results display that in co-inoculation assays

a dominance of the selected bacterial strain was reached early in the fermentation, which allows to control growth of the spontaneous bacterial populations, including spoilage microorganisms, without any negative effect on the yeast population and performance of AF, in concordance with data from other authors (Henick-Kling and Park, 1994; Massera et al., 2009; Rosi et al., 2003). 3.2. Chemical analysis of wines Table 2 summarises the mean values ± standard deviation of the chemical parameters analysed in the wines from all the assayed microvinifications. Values for citric acid concentrations in wines from coinoculation assays were significantly higher than those from sequential inoculation. It is important to highlight that in Tempranillo wine from COI, citric acid is not used by O. oeni C22L9. It would suggest that diacetyl would be potentially not being produced from citric acid in this treatment. Total acidity and lactic acid content were higher in wines from coinoculation assays of both varieties, with statistically significant differences for lactic acid depending on the yeast used. Values for volatile acidity, expressed as acetic acid, obtained in this study ranged between 0.30 and 0.51 g/L, which, as reported by some authors (Azzolini et al., 2010; Ribéreau-Gayon et al., 1989), are

P.M.I. Cañas et al. / International Journal of Food Microbiology 156 (2012) 245–254

249

considered to conform to the standard quality parameter for volatile acidity in red table wine. Statistically higher values for volatile acidity were obtained in Tempranillo wines produced by co-inoculation, though they were within the range normally found for Tempranillo wine (Izquierdo et al., 2008a). On the contrary, volatile acidity values of co-inoculated Merlot wines were equal to or significantly lower than those of Merlot wines produced by SEQ inoculation. In all wines, histamine and phenylethylamine concentrations were very low ranging between 0.01 and 0.06 mg/L. In contrast with the results from Massera et al. (2009), cadaverine and tyramine concentrations were significantly lower in wines produced by co-inoculation, except for tyramine in Merlot wines inoculated with S. cerevisiae VRB. A conclusive explanation for these results has not been found but their practical importance make valuable future researches in order to confirm them. Cadaverine has been reported to enhance the toxicity of histamine, tyramine and phenylethylamine, and can also have a detrimental effect on wine quality by imparting flavours of rotten meat (Palacios, 2006). As reported by other authors (Izquierdo et al., 2008b; Massera et al., 2009) putrescine was the most abundant amine in all analysed wines, with concentrations ranging between 2.92 and 4.66 mg/L. No significant differences, except for VRB Merlot wines, were found in the values for this amine in wines obtained from SEQ or COI. 3.3. Volatile compounds

Fig. 4. Time courses of glucose + fructose and malic and lactic acid concentrations during AF and MLF of Tempranillo (A) and Merlot (B) wines obtained from co-inoculation of VN or VRB yeasts and O. oeni C22L9. Glucose + fructose content is plotted against the righthand axis. Values are mean of triplicates± SE.

Table 3 shows the mean and the standard deviation values for the volatile compounds determined by gas chromatography–mass spectrometry. With respect to lineals alcohols, C6 alcohols and bencenic alcohols, wines produced by co-inoculation contained more propanol, isobutanol and isoamilic alcohols and less 1-pentanol and syringol, with significant differences in some of these compounds. Likewise, there were higher concentrations of 1-hexanol, c-3-hexen-1-ol and benzyl alcohol, compounds which contribute significantly to wine aroma (Ugliano and Henschke, 2008). The remaining compounds of these groups behaved differently depending on both the grape variety and the yeast strain inoculated. Thio-alcohols concentrations were significantly lower in wines produced by co-inoculation, except for Merlot wines inoculated with VN yeast. This could be connected with the lower concentration of SO2 added to the musts at the outset of AF and is extremely important from a sensory standpoint in that at high concentrations these compounds impart notes of boiled vegetables, onion, etc. (Flanzy, 2000).

Table 2 Chemical composition of wines at the end of MLF. Tempranillo Yeast

Merlot

VRB

VN

VRB

VN

Inoculation method

SEQ

COI

SEQ

COI

SEQ

COI

SEQ

COI

Alcohol (% v/v) Total aciditya pH Volatile acidityb L-malic acidc L-lactic acidc Citric acidc Histamined Tyramined Putrescined Cadaverined Phenylethylamined

13.62 ± 0.16 3.03 ± 0.17 4.23 ± 0.01 0.30 ± 0.03 0.11 ± 0.02 1.25 ± 0.00 0.07 ± 0.01 0.03 ± 0.01 1.09 ± 0.29 4.43 ± 0.23 0.59 ± 0.04 0.05 ± 0.01

13.78 ± 0.23 3.12 ± 0.04 4.22 ± 0.03 0.36 ± 0.01 ⁎ 0.11 ± 0.00 1.27 ± 0.02 0.31 ± 0.01 ⁎ 0.02 ± 0.02 0.43 ± 0.07 ⁎

13.93 ± 0.16 3.08 ± 0.04 4.28 ± 0.00 0.46 ± 0.00 0.15 ± 0.00 1.51 ± 0.01 0.04 ± 0.04 0.01 ± 0.01 1.09 ± 0.54 4.49 ± 0.06 0.59 ± 0.02 0.06 ± 0.01

13.56 ± 0.33 3.53 ± 0.01 ⁎ 4.23 ± 0.03 ⁎ 0.51 ± 0.01 ⁎ 0.11 ± 0.01 ⁎ 1.68 ± 0.03 ⁎ 0.32 ± 0.02 ⁎

13.87 ± 0.04 3.80 ± 0.02 3.66 ± 0.02 0.27 ± 0.01 0.10 ± 0.03 0.67 ± 0.02 0.11 ± 0.00 0.02 ± 0.01 0.77 ± 0.14 3.02 ± 0.06 0.18 ± 0.01 0.02 ± 0.00

13.55 ± 0.27 ⁎ 4.23 ± 0.04 ⁎ 3.66 ± 0.07 0.27 ± 0.01 0.10 ± 0.01 0.69 ± 0.06 0.13 ± 0.00 ⁎ 0.03 ± 0.00 0.69 ± 0.06 2.93 ± 0.02 ⁎ 0.14 ± 0.00 ⁎

13.95 ± 0.11 3.65 ± 0.05 3.65 ± 0.01 0.38 ± 0.01 0.10 ± 0.05 0.60 ± 0.02 0.09 ± 0.01 0.02 ± 0.02 1.28 ± 0.79 2.92 ± 0.08 0.22 ± 0.00 0.02 ± 0.00

13.79 ± 0.15 3.99 ± 0.00 ⁎ 3.68 ± 0.01 ⁎ 0.35 ± 0.01 ⁎ 0.12 ± 0.01 ⁎ 0.72 ± 0.04 ⁎ 0.12 ± 0.00 ⁎

a

4.21 ± 0.11 0.50 ± 0.00 ⁎ 0.04 ± 0.01

0.04 ± 0.02 0.49 ± 0.04 ⁎ 4.66 ± 0.29 0.52 ± 0.04 ⁎ 0.04 ± 0.00 ⁎

g/L tartaric acid. b g/L acetic acid. c g/L. d mg/L. ⁎ Denotes statistically significant differences (p ≤ 0.05) between the different inoculation method.

0.02 ± 0.01

0.01 ± 0.01 0.42 ± 0.10 ⁎ 2.95 ± 0.03 0.15 ± 0.01 ⁎ 0.01 ± 0.00

250

P.M.I. Cañas et al. / International Journal of Food Microbiology 156 (2012) 245–254

Table 3 Volatile compounds of wines at the end of MLF. Tempranillo

Merlot

Yeast

VRB

Inoculation method

SEQ

COI

SEQ

158.62±0.19 75.52±1.89 38.31±1.43 54.08±4.79 271.3±1.34

146.83±14.79 82.94±7.39 48.16±0.23 ⁎ 53.50±1.76 280.58±11.19

154.70 ± 3.16 148.94 ± 6.75 49.56 ± 0.27 54.85 ± 0.75 ⁎ 32.88 ± 0.79 36.69 ± 0.94 ⁎ 61.00 ± 10.35 41.68 ± 0.64 ⁎ 217.11 ± 3.02 271.30 ± 10.14 ⁎

3.00±0.19 92.94±6.25 17.42±0.55 0.22±0.00

3.76±0.41 ⁎ 102.56±7.12 18.62±0.88 ⁎ 0.44±0.04 ⁎

3.69 ± 0.26 107.02 ± 4.69 50.24 ± 0.34 0.24 ± 0.00

4.06 ± 0.05 ⁎ 111.15 ± 0.39 35.66 ± 5.90 ⁎ 0.37 ± 0.03 ⁎

0.29±0.02 14.67±0.11 25.15±4.01 2.82±0.33 0.30±0.01 1.59±0.09

0.34 ± 0.02⁎ 17.29±2.01 ⁎ 19.44±4.88 4.26±0.30 ⁎ 0.25±0.02 ⁎ 1.96±0.28 ⁎

0.29 ± 0.00 15.89 ± 0.25 25.24 ± 9.21 3.64 ± 0.11 0.43 ± 0.01 2.38 ± 0.11

0.31 ± 0.01 ⁎ 25.44 ± 2.34 ⁎ 16.81 ± 3.77 3.45 ± 0.36 0.40 ± 0.01 ⁎

119.28±13.12 0.41±0.03 20.90±0.79

67.01±32.62 ⁎ 0.21±0.11 ⁎ 6.67±3.24 ⁎

16.14±1.02 2.26±0.78 34.19±2.23

Lineals alcohols Methanol 1 Propanol 1 Isobutanol 1 1-pentanol 2 Active amyl+ isoamyl 1 C6 Alcohols 1-hexanol 1 t-3-hexenol 2 t-2-hexenol 2 c-3-hexen-1-ol 1 Bencenic alcohols Benzyl alcohol 1 2-Phenylethanol 1 Syringol 2 Eugenol 2 t-Isoeugenol 2 Methoxyeugenol 2 Thio-alcohols 2-Methyl-thio-etanol 2 3-Methyl-thio-propanol 1 3-Ethyl-thio-propanol 2 Furans Furfuryl alcohol 2 Furfural 2 Furaneol 2 Acids Hexanoic acid 1 Octanoic acid 1 Decanoic acid 1 Aldehydes and ketones Ethanal 1 2,3-butanodione 1 3-Hydroxy-2-butanone 1 1-Hydroxy-2-propanona 2 3-Hydroxy-2-pentanona 3 Esters Ethyl acetatea Isoamyl acetatea Isobutyl acetateb Hexyl acetateb Benzyl acetateb 2-Phenylethyl acetateb Ethyl butiratea Ethyl lactatea Ethyl hexanoatea Ethyl octanoatea Ethyl decanoatea Diethyl succinatea Terpenes α-Terpineolb Citronellolb Nerolb Geraniolb Linaloolb Volatile phenols 4-Ethyl-phenolb 4-Vinylphenolc 4-Ethyl-guaiacolb 4-Vinyl-guaiacolc Phenolb Vanillate derivates Vanillinb Methyl vanillateb Acetovanilloneb Zingeroneb Syringaldehydeb Acetosyringoneb Norisoprenoids Damascenoned 3-Hydroxy-βdamasconed

VN

VRB COI

VN

SEQ

COI

SEQ

188.40±5.35 29.29±0.07 43.10±0.75 75.71±4.76 306.04±1.21

194.11 ± 14.00 40.91 ± 2.15 ⁎ 61.34 ± 3.30 ⁎

157.58 ± 4.39 203.03 ± 12.87 ⁎ 23.69 ± 1.25 27.85 ± 0.42 ⁎ 53.92 ± 2.14 61.79 ± 3.76 ⁎ 63.01 ± 14.85 39.57 ± 24.73 342.27 ± 20.35 429.04 ± 24.97 ⁎

70.62 ± 8.21 401.59 ± 20.48 ⁎

COI

3.63 ± 0.37 5.39 ± 0.72 ⁎ 182.50 ± 18.41 173.59 ± 21.75 17.05 ± 3.23 20.81 ± 7.74 0.05 ± 0.01 0.08 ± 0.01 ⁎

2.90 ± 0.68 189.89 ± 25.92 46.63 ± 0.84 0.08 ± 0.00

3.97 ± 1.88 75.38 ± 41.56 ⁎ 12.54 ± 8.14 ⁎ 0.32 ± 0.19 ⁎

2.33 ± 0.28

0.40 ± 0.00 35.83 ± 0.24 40.18 ± 3.30 0.95 ± 0.06 0.19 ± 0.08 1.01 ± 0.16

0.54 ± 0.09 ⁎ 67.96 ± 7.36 ⁎ 20.01 ± 0.18 ⁎ 1.85 ± 0.29 ⁎ 0.42 ± 0.04 ⁎ 0.62 ± 0.07 ⁎

0.38 ± 0.01 75.13 ± 13.15 38.80 ± 6.39 1.24 ± 0.06 0.29 ± 0.09 1.49 ± 0.23

0.60 ± 0.41 13.28 ± 12.14 ⁎ 19.62 ± 5.78 ⁎ 3.65 ± 2.55 0.37 ± 0.14 2.19 ± 1.32

93.76 ± 10.84 0.47 ± 0.02 36.67 ± 0.95

69.75 ± 8.25 ⁎ 0.38 ± 0.01 ⁎ 26.28 ± 4.19 ⁎

99.03 ± 10.47 1.42 ± 0.10 30.28 ± 0.35

77.60 ± 3.89 ⁎ 1.12 ± 0.09 ⁎ 18.99 ± 0.65 ⁎

87.15 ± 9.47 1.60 ± 0.21 71.29 ± 0.01

53.16 ± 34.72 0.21 ± 0.15 ⁎ 5.76 ± 4.37 ⁎

6.60±3.55 ⁎ 4.02±0.43 ⁎ 21.91±11.08 ⁎

10.58 ± 0.89 2.89 ± 1.31 33.55 ± 2.97

6.63 ± 0.14 ⁎ 3.52 ± 0.26 19.83 ± 2.48 ⁎

20.60 ± 0.85 4.53 ± 0.25 22.40 ± 3.82

8.45 ± 0.31 ⁎ 8.53 ± 0.53 ⁎ 21.55 ± 0.08

20.71 ± 4.56 3.15 ± 0.66 26.47 ± 2.68

5.20 ± 3.79 ⁎ 5.77 ± 2.87 17.96 ± 11.41

2.91±0.01 3.75±0.07 0.72±0.22

3.34±0.27 ⁎ 3.85±0.27 0.69±0.13

3.14 ± 0.03 4.32 ± 0.33 0.70 ± 0.45

4.03 ± 0.48 ⁎ 5.62 ± 1.06 ⁎ 1.03 ± 0.44

2.62 ± 0.10 3.03 ± 0.07 0.37 ± 0.24

4.04 ± 0.39 * 3.49 ± 0.30 ⁎ 0.65 ± 0.21

2.88 ± 0.36 4.14 ± 0.87 0.53 ± 0.40

2.24 ± 1.50 2.66 ± 1.74 0.47 ± 0.23

4.97±0.23 9.85±0.28 2.82±0.09 25.19±8.73 0.11 ± 0.02

3.66±0.51 ⁎ 10.42±1.01 2.16±0.40 ⁎ 12.59±1.79 ⁎ 0.03±0.01 ⁎

2.58 ± 0.14 5.36 ± 0.08 1.63 ± 0.14 17.78 ± 6.51 0.02 ± 0.00

4.35 ± 1.33 ⁎ 6.60 ± 0.02 ⁎ 0.75 ± 0.21 ⁎ 6.96 ± 3.03 ⁎ 0.01 ± 0.00 ⁎

4.72 ± 0.45 5.79 ± 0.18 1.87 ± 0.01 18.44 ± 6.81 0.11 ± 0.00

5.33 ± 0.37 8.76 ± 1.68 ⁎ 1.63 ± 0.23 ⁎ 16.05 ± 1.63 0.08 ± 0.00 ⁎

6.76 ± 0.54 5.91 ± 0.34 2.44 ± 0.01 8.25 ± 2.93 0.09 ± 0.00

5.43 ± 0.34 ⁎ 9.02 ± 0.19 ⁎ 1.84 ± 0.03 ⁎

38.84 ± 1.82 1.16 ± 0.02 62.66 ± 13.56 41.67 ± 4.77 0.39 ± 0.01 76.61 ± 2.37 0.23 ± 0.00 20.80 ± 0.38 0.20 ± 0.01 0.24 ± 0.03 0.03 ± 0.02 0.33 ± 0.00

43.00 ± 1.77 ⁎ 0.99 ± 0.04 ⁎ 93.89 ± 13.51 ⁎

42.87±12.13 66.37±9.04 ⁎ 2.23±0.65 2.02±0.09 138.90±0.57 110.58±5.85 ⁎ 155.70±14.43 66.13±3.46 ⁎ 0.86 ± 0.11 0.30±0.00 ⁎ 114.84±1.97 56.07±3.75 ⁎ 0.24±0.05 0.42±0.03 ⁎ 21.57±0.95 27.87±1.29 ⁎ 0.18±0.04 0.19±0.03 0.32±0.03 0.36±0.02 0.05±0.01 0.05±0.02 0.17±0.01 0.27±0.02 ⁎

64.30 ± 0.54 73.47 ± 3.03 ⁎ 1.71 ± 0.00 1.91 ± 0.24 127.61 ± 11.29 109.41 ± 7.66 ⁎ 74.03 ± 3.52 58.62 ± 2.13 ⁎ 0.37 ± 0.04 0.25 ± 0.06 ⁎ 63.31 ± 1.18 72.00 ± 0.90 ⁎ 0.25 ± 0.02 0.31 ± 0.03 ⁎ 24.45 ± 1.47 26.94 ± 0.83 ⁎ 0.24 ± 0.05 0.25 ± 0.03 0.41 ± 0.05 0.52 ± 0.06 0.04 ± 0.03 0.06 ± 0.02 0.27 ± 0.02 0.39 ± 0.03 ⁎

39.19 ± 1.82 0.26 ± 0.00 ⁎ 88.27 ± 7.07 ⁎ 0.30 ± 0.02 ⁎ 25.86 ± 5.64 ⁎ 0.24 ± 0.03 ⁎ 0.34 ± 0.04 ⁎ 0.04 ± 0.02 0.47 ± 0.02 ⁎

10.14 ± 6.56 0.04 ± 0.02 ⁎

39.19 ± 13.45 44.88 ± 0.32 1.20 ± 0.35 1.20 ± 0.01 106.81 ± 44.38 181.87 ± 6.90 ⁎ 40.10 ± 2.58 54.70 ± 34.51 0.36 ± 0.03 0.12 ± 0.17 ⁎ 162.64 ± 8.64 38.44 ± 28.01 ⁎ 0.23 ± 0.08 0.25 ± 0.01 21.38 ± 1.28 26.99 ± 0.45 ⁎ 0.18 ± 0.06 0.14 ± 0.07 0.33 ± 0.07 0.25 ± 0.14 0.04 ± 0.03 0.03 ± 0.01 0.16 ± 0.00 0.40 ± 0.10 ⁎

0.57±0.01 8.36±0.35 2.39±0.16 9.81±0.58 0.99±0.08

0.67±0.05 ⁎ 16.38±1.48 ⁎ 3.23±0.24 ⁎ 11.85±0.05 ⁎ 0.76±0.14 ⁎

0.49 ± 0.01 6.51 ± 0.04 2.96 ± 0.06 12.94 ± 0.99 0.89 ± 0.13

0.62 ± 0.03 ⁎ 8.84 ± 1.59 ⁎ 3.32 ± 0.58 16.65 ± 1.29 ⁎ 0.72 ± 0.06 ⁎

1.50 ± 0.00 13.04 ± 0.03 2.96 ± 0.38 7.41 ± 0.29 2.80 ± 0.42

1.29 ± 0.04 ⁎ 21.05 ± 3.86 ⁎ 4.47 ± 0.77 ⁎ 9.72 ± 2.62 2.02 ± 0.37 ⁎

1.79 ± 0.02 4.82 ± 4.23 2.86 ± 1.27 8.47 ± 6.40 2.82 ± 0.38

0.58 ± 0.00 ⁎ 14.83 ± 91.14 ⁎ 2.03 ± 1.49 13.29 ± 0.99 0.59 ± 0.40 ⁎

0.13±0.01 0.08±0.03 0.09±0.03 0.54±0.05 1.79±0.16

0.15±0.00 ⁎ 0.05±0.04 0.09±0.03 0.42±0.24 1.48±0.16 ⁎

0.15 ± 0.01 0.06 ± 0.01 0.09 ± 0.01 0.28 ± 0.04 1.61 ± 0.21

0.15 ± 0.03 0.03 ± 0.01 ⁎ 0.08 ± 0.05 0.14 ± 0.07 ⁎ 1.29 ± 0.07 ⁎

0.45 ± 0.01 0.03 ± 0.01 0.11 ± 0.02 0.53 ± 0.10 1.99 ± 0.09

0.61 ± 0.11 ⁎ 0.05 ± 0.00 ⁎ 0.20 ± 0.01 ⁎ 0.81 ± 0.03 ⁎ 2.40 ± 0.13 ⁎

0.55 ± 0.05 0.05 ± 0.01 0.12 ± 0.01 0.77 ± 0.20 1.99 ± 0.15

0.13 ± 0.02 ⁎ 0.24 ± 0.12 ⁎ 0.11 ± 0.06 1.25 ± 0.74 1.18 ± 0.76

1.63±1.36 3.53±0.41 63.17±7.62 5.76±0.50 4.94±4.12 11.59±1.62

2.35±0.51 4.54±0.04 ⁎ 68.28±4.84 8.25±0.11 ⁎

2.85 ± 0.00 4.36 ± 0.02 ⁎ 59.07 ± 6.08 10.29 ± 0.10 ⁎

9.07±2.21 14.44±1.04 ⁎

2.97 ± 0.45 4.22 ± 0.06 66.05 ± 3.32 7.19 ± 0.37 8.86 ± 2.31 11.47 ± 0.94

8.82 ± 0.71 11.46 ± 0.40

1.76 ± 1.28 16.11 ± 1.65 67.75 ± 8.09 5.28 ± 0.49 4.40 ± 2.83 10.83 ± 1.47

4.03 ± 1.02 ⁎ 28.53 ± 3.10 ⁎ 113.09 ± 11.95 ⁎ 15.51 ± 2.80 ⁎ 10.36 ± 2.65 ⁎ 21.47 ± 2.04 ⁎

2.07 ± 0.23 22.43 ± 2.00 93.26 ± 6.65 8.03 ± 0.69 4.51 ± 1.29 17.11 ± 2.10

4.31 ± 0.30 ⁎ 10.23 ± 6.61 ⁎ 87.77 ± 14.18 11.81 ± 0.60 ⁎ 13.42 ± 4.36 ⁎

0.48±0.08 14.37±2.37

0.54±0.11 24.56±2.25 ⁎

0.48 ± 0.02 18.95 ± 0.89

0.55 ± 0.07 10.55 ± 0.56 ⁎

1.04 ± 0.06 68.47 ± 5.19

1.69 ± 0.21 ⁎ 120.45 ± 13.27 ⁎

2.34 ± 0.31 96.87 ± 9.56

3.10 ± 0.22 ⁎ 40.94 ± 26.32 ⁎

18.21 ± 3.40

P.M.I. Cañas et al. / International Journal of Food Microbiology 156 (2012) 245–254

251

Table 3 (continued) Tempranillo Yeast

VRB

Inoculation method

SEQ

Norisoprenoids dihydro-α-iononad 3-oxo-α-ionolc 3-oxo-7,8-dihydro-αionold β-iononeb

Merlot VN COI

SEQ

VRB COI

SEQ

27.33 ± 1.03 0.63 ± 0.03 40.46 ± 2.41

33.31 ± 2.06⁎ 0.79 ± 0.02⁎ 55.16 ± 1.46⁎

26.06 ± 0.39 0.70 ± 0.02 46.46 ± 2.81

26.89 ± 1.47 0.68 ± 0.00 50.99 ± 0.22 ⁎

0.24 ± 0.01

0.35 ± 0.03⁎

0.27 ± 0.07

0.27 ± 0.06

VN COI

19.90 ± 15.80 15.94 ± 2.64 0.80 ± 0.09 1.76 ± 0.27 ⁎ 63.48 ± 8.79 127.56 ± 15.17 ⁎ 0.26 ± 0.06

0.51 ± 0.07 ⁎

SEQ 10.29 ± 2.25 1.12 ± 0.18 96.61 ± 16.87 0.17 ± 0.03

COI 30.42 ± 16.68 ⁎ 1.12 ± 0.05 74.87 ± 1.40 ⁎ 0.31 ± 0.28

a

mg/L. b μg/L. c Area compound/area IS. d Area compound/area ISX1000. ⁎ Denotes statistically significant differences (p ≤ 0.05) between the different inoculation method.

The furan group behaved differently depending on the compounds. Furfuryl alcohol concentrations were significantly lower in all COI wines, while furaneol concentrations, though lower, significant differences were obtained only for Tempranillo wines. On the contrary, furfural concentrations were higher in COI wines, with significant differences when VRB yeast was used. Concentrations of the acids analysed varied not only depending on the type of inoculation (SEQ or COI) but also on the grape variety and yeast used. For instance, concentrations of hexanoic and octanoic acids were higher in wines produced by co-inoculation, except for Merlot wine inoculated with strain VN. This acids impart notes herbaceous and fruity, fatty or rancid to wine (Gómez-Mínguez et al., 2007; Mansfield et al., 2011) and even in low concentrations, their presence makes a significant contribution to wine aroma because of their low perception threshold (Rodríguez et al., 1990). Wines produced by co-inoculation also contained higher concentrations of 2,3-butanodione and lower concentrations of 3-hydroxy-2butanone, 3-hydroxy-2-pentanone and 1-hydroxy-2-propanone, with the exception of the Merlot wines inoculated with VN yeast. Concentrations of most of these compounds differed significantly from concentrations found in SEQ wines. Low concentrations of these compounds give the wine aromatic complexity, with buttery notes, contributing positively to aroma and organoleptic quality. Esters are important for determining wine aroma, and the presence of some short-chain esters, such as ethyl acetate, isobutyl acetate, isoamyl acetate and hexyl acetate, contributes imparting fruity flavours. So, ethyl acetate at concentrations lower than100 mg/L provides fruity notes, but it is responsible of an undesirable solvent/nail varnish-like aroma when it is present at high concentrations. Others such as diethyl succinate and ethyl lactate are beneficial in that they impart fruity, buttery and creamy notes to wines and contribute to mounthfeel (Izquierdo et al., 2008a; Lerm et al., 2010; Peinado et al., 2004). Ester synthesis and hydrolysis during MLF are due to the esterase activity of lactic bacteria and there is disagreement among authors as to the influence of MLF on the final ester content (Boido et al., 2009; Maicas et al., 1999). The wines produced by COI contained higher concentrations of ethyl acetate, ethyl butirate, ethyl lactate and diethyl succinate. Concentrations of the most of these compounds differed significantly from the concentrations in SEQ wines. On the contrary, benzyl acetate concentrations were significantly lower in COI wines. Terpenes are compounds present in free and glycosylated form in grapes. During AF, content of free terpenes often increases due to the β-glucosidase activity of yeasts (Gil et al., 1996) and during MLF a decrease of glycosidically bound volatile compounds, terpenes included, occurs (Ugliano and Moio, 2006). The ability of O. oeni strains to release terpenes from glycosidic precursors has been described by some authors (D´Incecco et al., 2004; Hernández-Orte et al., 2009; Ugliano et al., 2003) with the degree to which the enzymatic hydrolysis takes place being dependent on the bacterial strain, the chemical structure

of the substrate and the growth phase of the bacteria (Lerm et al., 2010; Ugliano and Moio, 2006). Results from this study show that wines produced by COI presented statistically higher concentrations of citronellol and statistically lower concentrations of linalool. Concentrations of geraniol were also higher in these wines though statistically significant differences were found only for Tempranillo wine. Concentrations of the remaining terpenes analysed, varied both with the grape variety (e.g. α-terpineol) and the yeast used (e.g. nerol in Merlot wine). In contrast with these results Knoll et al. (2012) have reported for Riesling wine that the SEQ wines have higher content of α-terpineol while COI wines have higher content of linalool. Of the group of volatile phenols, ethyl-phenols are particularly important because they contribute negatively to the final quality of wine, being responsible for the ‘phenolic’, ‘animal’ and ‘stable’ off-odours found in certain red wines (Couto et al., 2006; Gerbaux et al., 2009; Nelson, 2008). This study shows differences, statistically significant in some cases, in the concentrations of some of these compounds in wines produced by COI or SEQ, although their behaviour differed depending on the grape variety. For example, concentrations of 4-vinylphenol and 4-vinyl-guaiacol were lower in Tempranillo wines produced by COI and higher in Merlot wines. Of the compounds belonging to the vanillate derivates group, which give wines their spicy and smoked characteristics (Ferreira et al., 1995), it is worth noting the results from zingerone, concentrations of which were significantly higher in the wines produced by COI. The behaviour of the remaining compounds in this group was dependant on the yeast used or on the grape variety. The last of the groups analysed, norisoprenoids, also exert a significant influence on the sensory quality of wines (Izquierdo et al., 2008a), contributing fruity, floral or spicy notes. Concentrations of some of these, such as 3-Hydroxy-β-damascone and 3-oxo-7,8-dihydro-αionol, differed significantly although any trend was observed.

3.4. Multivariate data analysis Principal component analysis (PCA) was applied to the results from chemical and volatile compounds analysis to investigate possible correlations amongst wines. Table 4 shows the variables that correlated best with principal component 1 (PC1) and principal component 2 (PC2). 56. 25% of the variance was explained by the first two principal components. Fig. 5 shows the distribution of wines on the plane formed by the two principal components PC1 and PC2. Two different groups were evident: the Tempranillo wines on the left of the PC1 and the Merlot wines located on the positive side of this axis. The latter had lower pH, total acidity and putrescine content and higher content of methyl vanillate. The principal component 2 separated the wines according to the time of inoculation of LAB with COI wines located in the negative side of this axis. These wines contained lower content of furfuryl alcohol and tyramine and higher content of ethyl lactate.

252

P.M.I. Cañas et al. / International Journal of Food Microbiology 156 (2012) 245–254

Table 4 Principal component analysis (PCA) applied to the data from chemical and volatile compounds analysis. Principal component

Variance explained (%)

Total variance (%)

1

33.69

33.69

2

22.56

56.25

attributable to the higher astringency of Tempranillo wines produced by SEQ and to the more intense red fruit aroma of Merlot wines produced by COI.

Variables highly correlated with the axis and their loadings

4. Conclusions Cadaverine (− 0.938) Putrescine (−0.927) Isobutanol (0.909) Active amyl + isoamyl (0.901) pH (− 0.898) Total acidity (0.897) Methyl vanillate (0.891) 3-Hydroxy-β-damascone (0.886) Acetovanillone (0.885) 3-oxo-7,8-dihydro-α-ionol (0.875) Phenylethylamine (− 0.834) Acetosyringone (0.794) L-lactic acid (−0.770) 4-Ethyl-phenol (0.763) 4-Ethyl-guaiacol (0.730) 2-Phenylethanol (0.724) Furfuryl alcohol (0.906) Ethyl lactate (− 0.841) 3-Ethyl-thio-propanol (0.814) 2-Phenylethyl acetate (0.790) Linalool (0.760) 3-Methyl-thio-propanol (0.751) Tyramine (0.746) c-3-hexen-1-ol (− 0.732)

3.5. Sensory analysis Fig. 6 shows the results from descriptive sensory analysis of Tempranillo and Merlot wines. For each variety, sensory profiles of wines were similar with only slight differences attributable both to the yeast used and to the inoculation time. So, Tempranillo wines produced by COI had less astringency and body and a more intense spicy, vegetable and dairy aroma while COI Merlot wines had a more intense red fruit and vegetable aroma. Results from the triangular test are shown in Table 5. Tasters detected statistically significant differences in the flavour of Tempranillo wines and in the aroma of Merlot wines. These differences may be

The current study has shown that simultaneous inoculation of the lactic bacteria strain O. oeni C22L9 and a commercial yeast strain is a winemaking alternative to the traditional vinification, for making Tempranillo and Merlot variety wines. It represents a real saving of time for wineries, between 9 and 20 days depending on the grape variety and the pair yeast/bacteria, and no evidence of a negative impact on fermentation success or on final wine parameters has been found. These are important advantages for wineries, in terms of process efficiency, and, in addition, the presence of a selected LAB strain from the outset of microvinification keeps out other undesirable spontaneous bacteria, thus conferring sanitary benefits. Our results also have confirmed the findings of other authors (Azzolini et al., 2010; Jussier et al., 2006; Semon et al., 2001) for different grape varieties, demonstrating the possibility of simultaneous induction of alcoholic and malolactic fermentation without a pronounced degradation of malic acid during AF or an excessive increase in volatile acidity. From a sanitary point of view it is notable that concentrations of some biogenic amines like cadaverine and tyramine were lower in wines produced by co-inoculation, which should be evaluated in future researches. Exhaustive chemical analysis revealed numerous differences as regards volatile compounds composition, and multivariate analysis of these results separated wines according to the inoculation time. However, the impact of these differences on sensory profiles was limited according to the results of descriptive sensory analysis. The behaviour of the two yeast strains (VRB and VN) used in this study was similar at each must variety and both yeast/bacteria pairs assayed showed good compatibility, although the behaviour of the pair VN/C22L9 was slightly better in both grape varieties. Future research should be addressed to investigate vinification performance of other commercial O. oeni strains used in co-inoculation with commercial yeast strains and the possibility of using this winemaking practice with other grape varieties. Likewise, it would be necessary to confirm these findings to an industrial scale.

2

PC2 1,5 1 0,5 0 -0,5 -1 -1,5 -2 -1,5

-1

-0,5

0

0,5

1

1,5

2

PC1 Tempranillo VRB-SEQ

Tempranillo VRB-COI

Tempranilo VN-SEQ

Tempranillo VN-COI

Merlot VRB-SEQ

Merlot VRB-COI

Merlot VN-SEQ

Merlot VN-COI

Fig. 5. Plotting of the samples on the plane defined by the two principal components obtained by principal component analysis (PCA) of the data from chemical and volatile compounds.

P.M.I. Cañas et al. / International Journal of Food Microbiology 156 (2012) 245–254

A

B

Aromatic intensity

Aromatic intensity 8

8

Body

Red fruit

Body

Red fruit 6

6

4

4

Astringency

253

Astringency

Ripe fruit

2

Ripe fruit

2 0

0

Acidity

Acidity

Spicy

Spicy

SEQ-VRB Dairy

Floral

COI-VRB

Dairy

SEQ-VN

Vegetable

Floral Vegetable

COI-VN Fig. 6. Descriptive sensory analysis. A = Tempranillo wine. B = Merlot wine.

Table 5 Results of the sensory triangular test comparing aroma and flavour of wines produced using sequential inoculation and co-inoculation. Tempranillo

VRB Correct responses VN Correct responses

Merlot

Aroma

Flavour

Aroma

Flavour

6

10⁎

10⁎

8

5

11⁎

10⁎

9

Number of tasters: n = 18. ⁎ Denotes statistically significant differences for (p ≤ 0.05).

Acknowledgements The authors wish to thank the Consejería de Educación y Ciencia of the Junta de Comunidades de Castilla-La Mancha (JCCM) for project PCC 05-003-2 and the Ministerio de Educación y Ciencia (INIA) for project RM 2006-00011-C02-02. F. Pérez-Martín is supported by a grant of the Council of Communities of Castilla-La Mancha. P. M. Izquierdo acknowledges the Fondo Social Europeo and INCRECYT for cofunding his contract. We also thank J. M. Heras at Dantars Ferment A. G. for donation of the Oenococcus oeni and yeast strains. References Abrahamse, C.E., Bartowsky, E.J., 2012. Timing of MLF in Shiraz grape must and wine: influence on chemical composition. World Journal of Microbiology and Biotechnology 28, 255–265. Alexandre, H., Costello, P.J., Remize, F., Guzzo, J., Guilloux-Benatier, M., 2004. Saccharomyces cerevisiae–Oenococcus oeni interaction in wine: current knowledge and perspectives. International Journal of Food Microbiology 93, 141–154. Antalick, G., 2010. Bilan biochimique et sensoriel des modifications de la note fruitée des vins rouges lors de la fermentation malolactique: rôle particulier des esters. Thèse de doctorat en Sciences. École Doctorale en Sciences de la Vie et de la Santé (Bordeaux). Azzolini, M., Tosi, E., Vagnoli, P., Krieger, S., Zapparoli, G., 2010. Evaluation of technological effects of yeast–bacterial co-inoculation in red table wine production. Italian Journal of Food Science 3 (22), 257–263. Bauer, R., Dicks, L.M.T., 2004. Control of malolactic fermentation in wine. A review. South African Journal of Enology and Viticulture 25 (2), 74–88. Boido, E., Medina, K., Faria, L., Carrau, F., Versini, G., Dellacassa, E., 2009. The effect of bacterial strain and aging on the secondary volatile metabolites produced during malolactic fermentation of Tannat red wine. Journal of Agriculture and Food Chemistry 57, 6271–6278. Combina, M., Mellimacci, A., Sturm, M.E., Massera, A., Mercado, L., Sari, S., Catania, C., 2008. Malolactic fermentation in red wines: early inoculation with lactic bacteria. Proceedings of XII Jornadas Argentinas de Microbiología, Rosario, Argentina, p. 356. Couto, J.A., Campos, F.M., Figueiredo, A.R., Hogg, T.A., 2006. Ability of lactic acid bacteria to produce volatile phenols. American Journal of Enology and Viticulture 57, 166–171. D´Incecco, N., Bartowsky, E., Kassara, S., Lante, A., Spettoli, P., Henschke, P., 2004. Release of glycosidically bound flavour compounds of Chardonnay by Oenococcus oeni during malolactic fermentation. Food Microbiology 21, 257–265.

Edwards, C.G., Reynolds, A.G., Rodriguez, A.V., Semon, M.J., Mills, J.M., 1999. Implication of acetic acid in the induction of slow/stuck grape juice fermentations and inhibition of yeast by Lactobacillus spp. American Journal of Enology and Viticulture 50, 204–210. Ferreira, V., Fernández, P., Gracia, J.P., Cacho, J.F., 1995. Identification of volatile constituents in wines from Vitis vinifera var. Viradillo and sensory contribution of the different wine flavour fraction. Journal of the Science of Food and Agriculture 69, 229–310. Flanzy, C., 2000. Enología: fundamentos científicos y tecnológicos. A. Madrid Vicente, Madrid. Gerbaux, V., Briffox, C., Dumont, A., Sibylle, K., 2009. Influence of inoculation with malolactic bacteria on volatile phenols in wines. American Journal of Enology and Viticulture 60, 233–235. Gil, J.V., Mateo, J.J., Jiménez, M., Pastor, A., Huerta, T., 1996. Aroma compounds in wine as influenced by apiculate yeasts. Journal of Food Science 61, 1247–1249. Gómez-Alonso, S., Hermosín, I., García-Romero, E., 2007. Simultaneous HPLC analysis of biogenic amines, amino acids and ammonium ion as aminoenones derivatives in wine and beer samples. Journal of Agriculture and Food Chemistry 55, 608–613. Gómez-Mínguez, M.J., Cacho, J.F., Ferreira, V., Vicario, I.M., Heredia, F.J., 2007. Volatile components of Zalema white wines. Food Chemistry 100, 1464–1473. Henick-Kling, T., Park, Y.H., 1994. Considerations for the use of yeast and bacterial starter cultures: SO2 and timing of inoculation. American Journal of Enology and Viticulture 45, 464–469. Hernández-Orte, P., Cersosimo, M., Loscos, N., Cacho, J., García-Moruno, E., Ferreira, V., 2009. Aroma development from non-floral grape precursors by wine lactic acid bacteria. Food Research International 42, 773–781. ISO Standard 11035, 1994. Sensory analysis. Identification and selection of descriptors for establishing a sensory profile. ISO Standard 4120, 1983. Sensory Analysis. Methodology for triangular test. Izquierdo, P.M., García, E., Gómez, S., Palop, M.Ll., 2008a. Changes in the aromatic composition of Tempranillo wines during spontaneous malolactic fermentation. Journal of Food Composition Analysis 21, 724–730. Izquierdo, P.M., García, E., Gómez, S., Fernández, M., Palop, M.L.L., 2008b. Amino acids and biogenic amines during spontaneous malolactic fermentation in Cencibel red wines. Journal of Food Composition and Analysis 21, 731–735. Jussier, D., Morneau, A.D., Mira de Orduña, R., 2006. Effect of simultaneous inoculation with yeast and bacteria on fermentation kinetics and key wine parameters in cool climate Chardonnay. Applied and Environmental Microbiology 72, 221–227. Knoll, C., Fritsch, S., Schell, S., Grossmann, M., Krieger-Weber, S., du Toit, M., Rauhut, D., 2012. Impact of different malolactic fermentation inoculation scenarios on Riesling wine aroma. World Journal of Microbiology and Biotechnology 28, 1143–1153. Krieger, S., Zapparoli, G., Veneri, G., Tosi, E., Vagnoli, P., 2007. Comparison between simultaneous and sequential alcoholic and malolactic fermentations for partially dried grapes in the production of Amarone style wine. Australian NZ grapegrower winemaker 517, 71–77. Lerm, E., Engelbrecht, L., du Toit, M., 2010. Malolactic fermentation: the ABC's of MLF. South African Journal of Enology and Viticulture 31 (2), 186–212. Lonvaud-Funel, A., 1999. Lactic acid bacteria in the quality improvement and depreciation of wine. Antonie Van Leeuwenhoek 76, 317–331. Maicas, S., Gil, J.V., Pardo, I., Ferrer, S., 1999. Improvement of volatile composition of wines by controlled addition of malolactic bacteria. Food Research International 32, 491–496. Mansfield, A.K., Schirle-Keller, J.P., Reineccius, G.A., 2011. Identification of odor-impact compounds in red table wine produced from Frontenac grapes. American Journal of Enology and Viticulture 62 (2), 169–176. Massera, A., Soria, A., Catania, C., Krieger, S., Combina, M., 2009. Simultaneous inoculation of Malbec (Vitis vinifera) musts with yeast and bacteria: effects on fermentation performance, sensory and sanitary attributes of wines. Food Technology and Biotechnology 47, 192–201. Mendoza, L.M., Merín, M.G., Morata, V.I., Farías, M.E., 2011. Characterization of wines produced by mixed culture of autochthonous yeasts and Oenococcus oeni from

254

P.M.I. Cañas et al. / International Journal of Food Microbiology 156 (2012) 245–254

the northwest region of Argentina. Journal of Industrial Microbiology and Biotechnology 38, 1777–1785. Nelson, L., 2008. The production of volatile phenols by wine microorganisms. Thesis, Stellenbosch University, Private Bag X1, 7602 Matieland (Stellenbosch), South Africa. OIV, 2009. In: OIV (Ed.), Compendium of International Methods of Wine and Must Analysis, Edition 2009. Paris. Palacios, A., 2006. In: Morenzoni, R. (Ed.), Malolactic fermentation in wine: understanding the science and the practice. Lallemand, Montreal, pp. 71–77. Peinado, R.A., Moreno, J., Bueno, J.E., Moreno, J.A., Mauricio, J.C., 2004. Comparative study of aromatic compounds in two young white wines subjected to pre-fermentative cryomaceration. Food Chemistry 84, 589–590. Ribéreau-Gayon, J., Peynaud, E., Sudraud, P., Ribereau-Gayon, P., 1989. Tratado de Enología, Ciencias y Técnicas del Vino. Hemisferio Sur, Argentina . Ribéreau-Gayon, P., Dubourdieu, D., Donèche, B., Lonvaud-Funel, A., 2006. Handbook of Enology: The Microbiology of Wine and Vinifications, second ed. Wiley and Sons, Chichester, UK. Rodas, A.M., Ferrer, S., Pardo, I., 2003. 16S-ARDRA, a tool for identification of lactic acid bacteria isolated from grape must and wine. Systematic and Applied Microbiology 26, 412–422. Rodríguez, S.B., Amberg, E., Thornton, R.J., 1990. Malolactic fermentation in Chardonnay: growth and sensory effects of commercial strains of Leuconostoc oenos. Journal of Applied Bacteriology 68, 139–144. Rosi, I., Fia, G., Canuti, V., 2003. Influence of different pH values and inoculation time on the growth and malolactic activity of a strain of Oenococcus oeni. Australian Journal of Grape and Wine Research 9, 194–199. Ruiz, P., Izquierdo, P.M., Seseña, S., Palop, M.Ll., 2008. Intraspecific genetic diversity of lactic acid bacteria from malolactic fermentation of Cencibel wines as derived from combined analysis of RAPD-PCR and PFGE patterns. Food Microbiology 25, 942–948.

Ruiz, P., Izquierdo, P.M., Seseña, S., Palop, M.Ll., 2010. Selection of autochthonous Oenococcus oeni strains according to their oenological properties and vinification results. International Journal of Food Microbiology 137, 230–235. Ruiz-Barba, J.L., Cathcart, D.P., Wamer, P.J., Jiménez-Díaz, R., 1994. Use of Lactobacillus plantarum LPCO10, a bacteriocin producer, as a starter culture in Spanish-style green olive fermentations. Applied and Environmental Microbiology 60, 2059–2064. Semon, M.J., Edwards, C.G., Forsyth, D., Dinn, C., 2001. Inducing malolactic fermentation in Chardonnay musts and wines using different strains of Oenococcus oeni. Australian Journal of Grape and Wine Research 7, 52–59. Sieczkowski, N., 2004. Maîtrise et intérêts de la co-inoculation levures-bactéries. Revue Française d'Oenologie 207, 24–28. Sturm, M.E., Massera, A., Casassa, F., Krieger, S., Catania, C., Combina, M., 2008. Inoculación temprana de bacterias lácticas pre-aclimatadas en mostos Malbec. Enología 6, 1–8. Ugliano, M., Henschke, P.A., 2008. In: Moreno-Arribas, M.V., Polo, C. (Eds.), Wine chemistry and biochemistry. Springer, New York, pp. 328–348. Ugliano, M., Moio, L., 2006. The influence of malolactic fermentation and Oenococcus oeni strain on glycosidic aroma precursors and related volatile compounds of red wine. Journal of the Science of Food and Agriculture 86, 2468–2476. Ugliano, M., Genovese, A., Moio, L., 2003. Hydrolysis of wine aroma precursors during malolactic fermentation with four commercial starter cultures of Oenococcus oeni. Journal of Agriculture and Food Chemistry 51, 5073–5078. Zapparoli, G., Tosi, E., Azzolini, M., Vagnoli, P., Krieger, S., 2009. Bacterial inoculation strategies for the achievement of malolactic fermentation in high alcohol wines. South African Journal of Enology and Viticulture 30, 49–55.