Research Article Received: 8 November 2014
Revised: 27 April 2015
Accepted article published: 3 June 2015
Published online in Wiley Online Library: 6 July 2015
(wileyonlinelibrary.com) DOI 10.1002/jsfa.7293
Sequential culture with Torulaspora delbrueckii and Saccharomyces cerevisiae and management of fermentation temperature to improve cherry wine quality Shu Yang Sun,a* Han Sheng Gong,a Yu Ping Zhao,b Wen Li Liua and Cheng Wu Jina Abstract BACKGROUND: There has been limited research on the use of non-Saccharomyces yeasts for the production of cherry wines. This work used an autochthonous Torulaspora delbrueckii strain 49 (TD49) in association with a commercial S. cerevisiae RC212 yeast, to investigate the effect of multi-starter culture (sequential inoculation and simultaneous inoculation) and fermentation temperature on the quality of cherry wines. RESULTS: Both TD49 and RC212 proliferated during alcoholic fermentation (AF) under sequential inoculation conditions, whereas in the case of simultaneous inoculation, TD49 increased slowly at first and then declined sharply near the fermentation end. The analytical profile showed that both mixed fermentations produced lower levels of volatile acidy and higher levels of aromatic compounds than those from RC212 mono-culture. During sensory analysis, wines from sequential fermentation obtained the highest score, mainly due to the higher intensity in ‘fruity’ and ‘floral’ characters. As for the influence of temperature, a low temperature (20∘ C) enhanced TD49 persistence during AF, but the sensory quality decreased anyway; 30∘ C resulted in decreases in most measured descriptors. Therefore, 25∘ C was selected as the best culture temperature. CONCLUSION: TD49/RC212 sequential inoculation and fermentation at 25∘ C significantly enhanced the cherry wine quality. © 2015 Society of Chemical Industry Keywords: cherry wine; Torulaspora delbrueckii; Saccharomyces cerevisiae; inoculation modality; sensory analysis
INTRODUCTION
1880
Cherries are one of the most popular fruits cultivated and consumed worldwide. This fruit is known for its very high amount of nutrients and antioxidant components, such as vitamin C, vitamin A, vitamin E, anthocyanins and carotenoids.1 However, the fruit is relatively perishable, and is often not suitable to be used as fresh table fruit after transportation; thus, it is often processed to juice and wine. Cherry wine has a unique colour and flavour, and contains most nutrients derived from cherry fruit. Nowadays, this kind of product is very popular in China, and it is a certainty that the special characteristic and distinct flavour of cherry wines will receive more scrutiny in the future.2,3 Aroma is one of the most important distinguishing features of cherry wines. It is derived from hundreds of volatile compounds from cherry itself, others formed during wine-making by yeasts and the ageing process.4,5 As most of the odorous compounds are produced via fermentation, the selection of proper yeast strains is critical for the style and quality of cherry wine.6 Today, a number of Saccharomyces cerevisiae wine yeasts are commercially available, offering the opportunity to explore suitable strains to assure a rapid and reliable fermentation process, and giving wines a consistent quality.7 However, such practices also have drawbacks, especially in the way of masking or reducing the characteristics and J Sci Food Agric 2016; 96: 1880–1887
differences among wines made from diverse varieties, which quite contradict the growing consumer demand for wines with more complexity. In order to deal with the dilemma, the application of combined fermentation using different yeast species is a reasonable and practical approach. Spontaneous wine-making, where both S. cerevisiae and non-Saccharomyces yeasts co-exist, leads to more complex and aromatic wines due to the functions and interactions of the various microorganisms.8 – 10 Over the past decades, a few non-Saccharomyces yeasts have been tested and intensively investigated for the production of grape wines, and the most frequently used species include Torulaspora delbrueckii, Metchnikowia pulcherrima, Candida zemplinina, Lachancea thermotolerans and Hanseniaspora uvarum.8 – 17 These
∗
Correspondence to: Shu Yang Sun, School of Food Engineering, Ludong University, Yantai, Shandong 264025, P.R. China. E-mail: [email protected]
a School of Food Engineering, Ludong University, Yantai, Shandong 264025, P.R. China b Institute of Food Science and Engineering, Yantai University, Yantai, Shandong 264005, P.R. China
www.soci.org
© 2015 Society of Chemical Industry
Bacterial cultures to improve cherry wine
www.soci.org
strains could promote the release of certain aromatic components, such as higher alcohols, esters, acids and terpenes. However, the inappropriate use of non-Saccharomyces yeasts may result in a series of serious fermentative problems, for instance, stuck/sluggish fermentation, excessive accumulation of several undesirable metabolites (e.g. acetic acid, ethyl acetate and acetoin), lack of reproducibility, etc. Therefore, in order to highlight the positive effect of non-Saccharomyces wine yeasts and simultaneously reduce its negative impact, several technological factors should be considered and intensively studied. To date, there have been few reports available on the use and effect of non-Saccharomyces yeasts for production of cherry wines. In our previous investigation,18 we found the introduction of Torulaspora delbrueckii to the fermenting cherry musts could evidently enhance the wine’s fruity characters, but the inoculation modality still remained unresolved. In addition, some significant technological parameters affecting the cherry wine quality, such as culture temperature also needed elucidation. For these purposes, the current study was divided into two sections, with the first part aiming at the impact of inoculation modality, and the second part focusing on the influence of fermentation temperatures.
MATERIALS AND METHODS Cherry and cherry must preparation Cherry fruits of a tart variety ‘Early Richmond’ were selected as the materials for the experiment. They were harvested during the 2012 vintage from a local cherry orchard in Yantai city (Shandong, China). After transportation to the laboratory, the cherries were immediately crushed and manually deseeded to acquire the cherry musts. Then, the musts were pasteurised for 10 min at 102∘ C, cooled down and poured into 10-L bottles where they were treated with a pectinolytic preparation, Lallzyme EX-V pectinase (20 mg L−1 ), for 10 h at 20∘ C. The initial cherry must composition was: total reducing sugars (TRS) 181 g L−1 , total acidity (TA) 7.44 g L−1 , L-malic acid 2.5 g L−1 , pH 4.04. Microorganisms and medium The commercial S. cerevisiae strain Lalvin RC212 (Lallemand Inc., Montreal, Canada), and autochthonous strain of Torulaspora delbrueckii 49 (TD49) isolated from the local cherry wineries, were used in the present investigation. TD49 was previously identified and characterised by restriction fragment length polymorphism PCR analysis (unpublished data). TD49 was sub-cultured on YPD medium (20 g L−1 glucose, 20 g −1 L peptone, 10 g L−1 yeast extract, 20 g L−1 agar) at 3-month intervals, and maintained at 4∘ C. The media used to evaluate viable cell counts during mixed fermentations were WL nutrient agar (Oxoid, Unipath Ltd, Basingstoke, UK) and lysine agar (Oxoid). WL nutrient agar is used for the putative identification of wine yeast on the basis of the colour and morphology of their colonies.19 Lysine agar medium is a selective medium used to differentiate S. cerevisiae from non-Saccharomyces yeasts,20 and was therefore used for the viable count of the non-Saccharomyces yeasts cultured in mixed fermentation. The enumeration of yeast cells in the single culture fermentations was carried out using YPD agar plates.
J Sci Food Agric 2016; 96: 1880–1887
Evaluation of culture temperatures TD49/RC212 sequential fermentations were carried out at 20, 25 and 30∘ C to elucidate the effect of culture temperature on the quality of cherry wines. Before inoculation, both yeast strains were precultured in YPD broth medium at 25∘ C for 48 h. For fermentation, TD49 was inoculated at 107 cells mL−1 and 24 h prior to RC212 which was inoculated at 106 cells mL−1 . Analysis of fermentative progression During fermentation, a sample was withdrawn and different aliquots were used for the determination of viable cell counts and TRS. TRS was measured by the official methods21 after the removal of yeast cells by centrifugation. Viable cells was determined by plating onto corresponding medium as stated above using the successive dilution method, and counting visible colonies after incubation at 28∘ C for 2–5 days. After fermentation, in both pure and mixed cultures, the fermented musts were centrifuged (10 min at 5439 g) to remove yeast cells. Then 50 mg L−1 of SO2 was added to fermented musts, which were kept at 4∘ C before analysis. Analytical determinations TRS, TA (expressed as g L−1 of malic acid), pH, ethanol and volatile acidity (VA), were determined according to the official methods.21 Extraction and quantification of volatile compounds The aromatic compounds were extracted by a headspace solid-phase microextraction (HS-SPME) method, conducted on a 50/30 μm DVB/CAR/PDMS (Supelco, Bellefonte PA, USA) fibre, and separated using an Agilent 6890N GC (Agilent, Santa Clara, CA, USA) on a DB-Wax column (60 m × 0.25 mm i.d., 0.25 μm film thickness; J&W Scientific, Folsom, CA, USA) equipped with an Agilent 5975 MS detector. Sample preparation and GC analysis was performed as described previously.22 All analyses were carried out in triplicate. Identification of compounds was determined by comparing mass spectral data from Database (Agilent) and confirmed by comparing Kovats RIs to those of the standards. For quantification of volatile compounds, standard curves were constructed and used, as stated previously.22 Standards were prepared in a synthetic wine (12 v/v % ethanol:water solution, pH 4.10) with 3-octanol as the internal standard. An aliquot of 5 mL of synthetic solution containing various concentrations of volatile standards was diluted in a step-wise manner with synthetic solution using a series of 1:1 dilutions. And the standard curve for individual volatile compounds was obtained by plotting the response ratio of target compound and internal standard against the concentration ratio.
© 2015 Society of Chemical Industry
wileyonlinelibrary.com/jsfa
1881
Evaluation of the inoculation modalities Triplicate fermentations were carried out under static conditions at 25∘ C in 10-L sterile Erlenmeyer flasks which contained 7 L
pasteurised cherry must supplemented with sucrose (to 210 g L−1 ) and potassium metabisulfite (to obtain 50mg L−1 of total SO2 ). The inoculations were carried out as follows: (1) pure culture; (2) sequential fermentation (inoculation of TD49 was 24 h ahead of RC212); and (3) co-culture (simultaneous inoculation of TD49 and RC212). Before inoculation, both yeast strains were precultured in YPD broth medium at 25∘ C for 48 h, then the biomass was centrifuged at 7104 g for 10 min and re-suspended in water. The non-Saccharomyces strain was inoculated at 107 cells mL−1 , while the S. cerevisiae starter was inoculated at 106 cells mL−1 .
www.soci.org Sensory analysis Duplicate sensory evaluation was performed on the cherry wine samples resulting from different inoculations (mono-culture of RC212 and mixed fermentation of TD49/ RC212) and culture temperature (20, 25 and 30∘ C). The aroma descriptors were made by an expert panel comprising 11 tasters (seven women and four men, ranging in ages from 24 to 45 years), and included floral, fruity, sweet, green, almond, fatty and overall aroma. Two samples were evaluated for aroma characteristics in each session. Panellists were required to grade the samples on the basis of a 5-point intensity scale, where 0 = none, 1 = very low, 2 = low, 3 = medium, 4 = high and 5 = very high. Statistical analysis To evaluate the statistical significance of the chemical and sensory evaluation data of the wines, these were subjected to one-way analysis of variance using the SPSS version 16.0 statistical package for windows (SPSS Inc., Chicago, IL, USA). The significant differences among the data were determined using the Duncan test, at P < 0.05.
RESULTS AND DISCUSSION
1882
Effect of inoculation modality on the evolution of yeasts and fermentation kinetics The growth kinetics of biomass in either single or multi-starter fermentations carried out at 25∘ C are presented in Figs 1–3. For the pure cultures (Fig. 1), RC212 finished alcoholic fermentation (AF) within 5 days and generated a maximum viable population of 1.05 × 108 cells mL−1 , while TD49 took 14 days for AF and led to a maximal cell population of 1.84 × 108 cells mL−1 . In comparison with RC212 mono-culture, a much longer fermentation period by the single inoculation of TD49 was linked to a much slower consumption rate of reducing sugar. When AF was conducted with mixed cultures, both simultaneous and sequential fermentations progressed to completion in 5–6 days. Under sequential condition (Fig. 2), both yeast strains co-dominated the fermentation process. TD49 showed greater persistence during the first 5 days and achieved a higher cell population of 1.39 × 108 cells mL−1 . The increase in TD49 viable cells led to a limited reduction in the number of viable cells of RC212, and 18% lower viable population was obtained relative to its mono-culture. This may be ascribed to the enhanced competition of the non-Saccharomyces yeast in the delayed inoculation of S. cerevisiae strain,8,23,24 which will probably be caused by competition for nitrogen, as has been previously noted.25 However, when the two starters were inoculated at the same time (Fig. 3), RC212 quickly adapted to the cherry musts, and maintained a relatively higher growth rate for approximately 3 days. It reached a viable population of 9.95 × 107 cells mL−1 on the third day, and maintained this level almost until the end of fermentation. In contrast, TD49 strain was significantly influenced by the proliferation of RC212, as seen by the fact that TD49 slowly increased during the initial AF and dropped dramatically after 5 days. These observations suggest that certain interaction may occur between the two yeast strains, which led to a sharp decline of TD49 near the end of AF. This would probably be ascribed to a fierce competition for the use of nutrients (sugar, nitrogen, oxygen) and/or space,15,26,27 but other possibilities, such as inhibition caused by certain killer toxins or some other metabolites synthesised by S. cerevisiae,10,28 – 30 cannot completely be eliminated. From the performance of TD49
wileyonlinelibrary.com/jsfa
SY Sun et al.
over AF either in the case of co-culture or sequential culture, it was found that TD49, like many other non-Saccharomyces yeasts, does not strongly compete with other fermenting yeasts, as many commercial Saccharomyces strains do, thus enabling the maintenance of higher yeast diversity during the fermentation process. Effect of inoculation modality on the compositional and volatile profiles Table 1 (columns 2–5) shows the compositional profiles from pure cultures and from those mixed fermentations under different inoculation conditions. It was seen that the physico-chemical parameters of these wines were all within the rank of values accepted for regular wines. As expected, cherry wines produced from TD49 single inoculation were characterised by the highest TRS and the lowest ethanol content, whereas RC212 single culture resulted in the opposite result (lowest TRS and highest ethanol). In comparison with S. cerevisiae mono-culture, the mixed fermentation based cherry wines contained more residual sugar and consequently lower ethanol levels. Also, the TA and L-malic acid displayed higher levels in those two wines, particularly in that with the sequential inoculation. Another important feature that needs to be highlighted is the marked decrease in the volatile acidity by multi-starter cultures, as evidenced by the fact that co-culture led to a level of 0.33 g L−1 , and sequential inoculation generated a value of 0.34 g L−1 , both evidently lower than that from RC212 pure culture (0.42 g L−1 ). The non-Saccharomyces yeasts have been historically considered as wine-making contaminants due to their production of high amounts of acetic acid.7,31 Comitini and Renault have found that, unlike many apiculate strains, isolates of T. delbrueckii did not form high levels of VA.9,16 Our results, using strain TD49 for the production of cherry wines, was in agreement with their reports, and at the same time, clearly show the promising oenological feature of this strain. For volatile compounds determined (Table 2, columns 3–6), pure culture of TD49 produced the lowest total amount, then follows RC212 mono-culture and RC212/TD49 co-fermentation, and the sequential mixed culture based cherry wine showed the highest total content. Monoculture wines demonstrated that RC212 was a stronger producer of volatile esters, alcohols and acids, which were 1.2-, 1.6- and 1.7-fold greater than those found in TD49 single culture, respectively. With regard to mixed fermentations, the wines from sequential culture and simultaneous culture showed similar values in the total amount of volatile components, but significant differences between their particular profiles were observed. For specific compounds identified, sequential culture markedly enhanced the production of several volatile alcohols, esters and varietal compounds, mainly represented by ethyl 3-methylbutanoate, ethyl hexanoate, ethyl hex-3-enoate, ethyl octanoate, 2-methyl-1-propanol, 3-methyl-1-butanol, 𝛽-phenylethanol and linalool. Of note, the yield of 𝛽-phenylethanol was 2- and 1.2-fold higher in this wine than that in RC212 mono-culture and co-fermentation wines, respectively. This observation was in agreement with previous findings, as reported by Comitini who found that in grape wines, mixed fermentation involving T. delbrueckii would significantly enhance the production of this compound.9 Descriptive analysis of cherry wines resulting from different inoculation strategies Figure 4 shows the mean rating of aroma attributes of cherry wines obtained from different inoculation modalities. The RC212
© 2015 Society of Chemical Industry
J Sci Food Agric 2016; 96: 1880–1887
Bacterial cultures to improve cherry wine
www.soci.org
8.5
Log CFU mL–1
200 7.5 150
RC212 ( ) TD49 ( ) Sugar consumption by RC212 ( ) Sugar consumption by TD49 ( )
100
6.5
Recucing sugar (g L–1)
250
50
5.5 0
1
2
3
4
5
6
7 8 9 Time (days)
10
11
12
13
14
15
0
Figure 1. Growth kinetics and sugar consumption during cherry wine-making by single cultures (25∘ C). Values are mean of triplicates ± SE.
6
5
50
0
1
2
3
4
5
6
7
8
200
8
150 7
RC212 ( ) TD49 ( ) TRS( )
6
5
0
50
0
1
Time (days) Figure 2. Growth kinetics and sugar consumption during cherry wine-making by sequential cultures at 25∘ C. Values are mean of triplicates ± SE.
mono-culture based cherry wine exhibited moderate intensities for fruity and floral aroma, and low scores for fatty, almond, sweet and green notes. In the case of mixed fermentations, it was found that both multi-starter cultures markedly increases the wines ‘floral’ and ‘fruity’ descriptors which was particularly evident in the
100
2
3 4 5 Time (days)
6
7
8
Reducing sugar (g L–1)
100
Log CFU mL–1
150
RC212 ( ) TD49 ( ) TRS( )
7
Reducing sugar (g L–1)
Log CFU mL–1
200
8
250
9
250
9
0
Figure 3. Growth kinetics and sugar consumption during cherry wine-making by co-cultures (25∘ C). Values are mean of triplicates ± SE.
case of sequential fermentation trials, and simultaneously reduced the wines ‘green’ and ‘fatty’ nuances that showed no significant differences between co-culture and sequential fermentation. ‘Almond’ is the only character that remained nearly unchanged when a mixed fermentation strategy was applied, when compared to a S. cerevisiae single fermentation.
Table 1. The analytical profile of cherry wines resulting from different inoculations (columns 2–5) and different temperatures (columns 6–8)
Parameter
RC212 (25∘ C)
TD49 (25∘ C)
Sequential culture (25∘ C)
Fermentation time (days) Ethanol (%, v/v) TRS (g L−1 ) pH TA (g L−1 ) L-Malic acid (g L−1 ) VA (g L−1 )
5 10.81 ± 0.20b 0.96 ± 0.02a 4.07 ± 0.02a 6.49 ± 0.04a 2.24 ± 0.02a 0.42 ± 0.04b
14 8.19 ± 0.18a 13.83 ± 0.05d 4.04 ± 0.02a 6.74 ± 0.06c 2.49 ± 0.02c 0.29 ± 0.02a
6 10.30 ± 0.20a 3.25 ± 0.04c 4.06 ± 0.02a 6.60 ± 0.04bc 2.31 ± 0.02b 0.34 ± 0.02ab
Co-culture (25∘ C) 5 10.36 ± 0.23a 1.44 ± 0.02b 4.06 ± 0.02a 6.52 ± 0.04ab 2.29 ± 0.02b 0.33 ± 0.03a
20∘ C (Sequential)
25∘ C (Sequential)
30∘ C (Sequential)
10 10.28 ± 0.20A 3.62 ± 0.04C 4.06 ± 0.02A 6.58 ± 0.06B 2.26 ± 0.02A 0.35 ± 0.04A
6 10.30 ± 0.20A 3.25 ± 0.04B 4.06 ± 0.02A 6.60 ± 0.04B 2.31 ± 0.02B 0.34 ± 0.02A
5.5 10.33 ± 0.22A 0.98 ± 0.02A 4.07 ± 0.02A 6.46 ± 0.04A 2.28 ± 0.02AB 0.38 ± 0.05A
a–d In
J Sci Food Agric 2016; 96: 1880–1887
© 2015 Society of Chemical Industry
wileyonlinelibrary.com/jsfa
1883
columns 2–5 (cherry wines resulting from different in inoculations), values in the same row with different lower-case superscript letters are significantly different, according to Duncan’s test (P < 0.05). A–C In columns 6–8 (cherry wines obtained at different temperatures), values in the same row with different upper-case superscript letters are significantly different, according to Duncan’s test (P < 0.05). TA was expressed as g L−1 of malic acid.
www.soci.org
SY Sun et al.
Table 2. Concentrations of volatile compounds in cherry wines resulting from different inoculations (columns 3–6) and different temperatures (columns 7–9) (in milligrams per litre, n = 3)
RI
Compound
RC212 (25∘ C)
Sequential TD49 (25∘ C) culture (25∘ C)
Esters 0.86 ± 0.10b Ethyl acetate* Ethyl butyrate 0.14 ± 0.01a Ethyl 3-methylbutanoate 0.03 ± 0.01a 3-Methylbutyl acetate 0.37 ± 0.04ab Ethyl hexanoate 0.11 ± 0.02a Ethyl hex-3-enoate 0.39 ± 0.04a Ethyl hex-2-enoate 0.41 ± 0.05a Ethyl lactate 7.48 ± 0.81b Ethyl octanoate 0.66 ± 0.07a 2-Phenylethyl acetate 0.13 ± 0.02ab Sub-total 10.58 Alcohols 1029 1-Propanol 6.94 ± 0.72b 1076 2-Methyl-1-propanol 4.04 ± 0.44b 1136 1-Butanol 0.34 ± 0.03a 1193 3-Methyl-1-butanol 2.98 ± 0.0.31b 1342 1-Hexanol 7.55 ± 0.84b 1868 Benzyl alcohol 5.45 ± 0.52b 1896 𝛽-Phenylethanol 4.96 ± 0.50a Sub-total 32.26 Acids 1445 Acetic acid 14.38 ± 1.55c 1556 2-Methylpropanoic acid 1.28 ± 0.14b 1620 Butanoic acid 0.35 ± 0.03c 1662 2-Methylbutanoic acid 0.58 ± 0.05c 1840 Hexanoic acid 1.6 ± 0.17c 2051 Octanoic acid 1.08 ± 0.11b Sub-total 19.27 Aldehydes
Revised: 27 April 2015
Accepted article published: 3 June 2015
Published online in Wiley Online Library: 6 July 2015
(wileyonlinelibrary.com) DOI 10.1002/jsfa.7293
Sequential culture with Torulaspora delbrueckii and Saccharomyces cerevisiae and management of fermentation temperature to improve cherry wine quality Shu Yang Sun,a* Han Sheng Gong,a Yu Ping Zhao,b Wen Li Liua and Cheng Wu Jina Abstract BACKGROUND: There has been limited research on the use of non-Saccharomyces yeasts for the production of cherry wines. This work used an autochthonous Torulaspora delbrueckii strain 49 (TD49) in association with a commercial S. cerevisiae RC212 yeast, to investigate the effect of multi-starter culture (sequential inoculation and simultaneous inoculation) and fermentation temperature on the quality of cherry wines. RESULTS: Both TD49 and RC212 proliferated during alcoholic fermentation (AF) under sequential inoculation conditions, whereas in the case of simultaneous inoculation, TD49 increased slowly at first and then declined sharply near the fermentation end. The analytical profile showed that both mixed fermentations produced lower levels of volatile acidy and higher levels of aromatic compounds than those from RC212 mono-culture. During sensory analysis, wines from sequential fermentation obtained the highest score, mainly due to the higher intensity in ‘fruity’ and ‘floral’ characters. As for the influence of temperature, a low temperature (20∘ C) enhanced TD49 persistence during AF, but the sensory quality decreased anyway; 30∘ C resulted in decreases in most measured descriptors. Therefore, 25∘ C was selected as the best culture temperature. CONCLUSION: TD49/RC212 sequential inoculation and fermentation at 25∘ C significantly enhanced the cherry wine quality. © 2015 Society of Chemical Industry Keywords: cherry wine; Torulaspora delbrueckii; Saccharomyces cerevisiae; inoculation modality; sensory analysis
INTRODUCTION
1880
Cherries are one of the most popular fruits cultivated and consumed worldwide. This fruit is known for its very high amount of nutrients and antioxidant components, such as vitamin C, vitamin A, vitamin E, anthocyanins and carotenoids.1 However, the fruit is relatively perishable, and is often not suitable to be used as fresh table fruit after transportation; thus, it is often processed to juice and wine. Cherry wine has a unique colour and flavour, and contains most nutrients derived from cherry fruit. Nowadays, this kind of product is very popular in China, and it is a certainty that the special characteristic and distinct flavour of cherry wines will receive more scrutiny in the future.2,3 Aroma is one of the most important distinguishing features of cherry wines. It is derived from hundreds of volatile compounds from cherry itself, others formed during wine-making by yeasts and the ageing process.4,5 As most of the odorous compounds are produced via fermentation, the selection of proper yeast strains is critical for the style and quality of cherry wine.6 Today, a number of Saccharomyces cerevisiae wine yeasts are commercially available, offering the opportunity to explore suitable strains to assure a rapid and reliable fermentation process, and giving wines a consistent quality.7 However, such practices also have drawbacks, especially in the way of masking or reducing the characteristics and J Sci Food Agric 2016; 96: 1880–1887
differences among wines made from diverse varieties, which quite contradict the growing consumer demand for wines with more complexity. In order to deal with the dilemma, the application of combined fermentation using different yeast species is a reasonable and practical approach. Spontaneous wine-making, where both S. cerevisiae and non-Saccharomyces yeasts co-exist, leads to more complex and aromatic wines due to the functions and interactions of the various microorganisms.8 – 10 Over the past decades, a few non-Saccharomyces yeasts have been tested and intensively investigated for the production of grape wines, and the most frequently used species include Torulaspora delbrueckii, Metchnikowia pulcherrima, Candida zemplinina, Lachancea thermotolerans and Hanseniaspora uvarum.8 – 17 These
∗
Correspondence to: Shu Yang Sun, School of Food Engineering, Ludong University, Yantai, Shandong 264025, P.R. China. E-mail: [email protected]
a School of Food Engineering, Ludong University, Yantai, Shandong 264025, P.R. China b Institute of Food Science and Engineering, Yantai University, Yantai, Shandong 264005, P.R. China
www.soci.org
© 2015 Society of Chemical Industry
Bacterial cultures to improve cherry wine
www.soci.org
strains could promote the release of certain aromatic components, such as higher alcohols, esters, acids and terpenes. However, the inappropriate use of non-Saccharomyces yeasts may result in a series of serious fermentative problems, for instance, stuck/sluggish fermentation, excessive accumulation of several undesirable metabolites (e.g. acetic acid, ethyl acetate and acetoin), lack of reproducibility, etc. Therefore, in order to highlight the positive effect of non-Saccharomyces wine yeasts and simultaneously reduce its negative impact, several technological factors should be considered and intensively studied. To date, there have been few reports available on the use and effect of non-Saccharomyces yeasts for production of cherry wines. In our previous investigation,18 we found the introduction of Torulaspora delbrueckii to the fermenting cherry musts could evidently enhance the wine’s fruity characters, but the inoculation modality still remained unresolved. In addition, some significant technological parameters affecting the cherry wine quality, such as culture temperature also needed elucidation. For these purposes, the current study was divided into two sections, with the first part aiming at the impact of inoculation modality, and the second part focusing on the influence of fermentation temperatures.
MATERIALS AND METHODS Cherry and cherry must preparation Cherry fruits of a tart variety ‘Early Richmond’ were selected as the materials for the experiment. They were harvested during the 2012 vintage from a local cherry orchard in Yantai city (Shandong, China). After transportation to the laboratory, the cherries were immediately crushed and manually deseeded to acquire the cherry musts. Then, the musts were pasteurised for 10 min at 102∘ C, cooled down and poured into 10-L bottles where they were treated with a pectinolytic preparation, Lallzyme EX-V pectinase (20 mg L−1 ), for 10 h at 20∘ C. The initial cherry must composition was: total reducing sugars (TRS) 181 g L−1 , total acidity (TA) 7.44 g L−1 , L-malic acid 2.5 g L−1 , pH 4.04. Microorganisms and medium The commercial S. cerevisiae strain Lalvin RC212 (Lallemand Inc., Montreal, Canada), and autochthonous strain of Torulaspora delbrueckii 49 (TD49) isolated from the local cherry wineries, were used in the present investigation. TD49 was previously identified and characterised by restriction fragment length polymorphism PCR analysis (unpublished data). TD49 was sub-cultured on YPD medium (20 g L−1 glucose, 20 g −1 L peptone, 10 g L−1 yeast extract, 20 g L−1 agar) at 3-month intervals, and maintained at 4∘ C. The media used to evaluate viable cell counts during mixed fermentations were WL nutrient agar (Oxoid, Unipath Ltd, Basingstoke, UK) and lysine agar (Oxoid). WL nutrient agar is used for the putative identification of wine yeast on the basis of the colour and morphology of their colonies.19 Lysine agar medium is a selective medium used to differentiate S. cerevisiae from non-Saccharomyces yeasts,20 and was therefore used for the viable count of the non-Saccharomyces yeasts cultured in mixed fermentation. The enumeration of yeast cells in the single culture fermentations was carried out using YPD agar plates.
J Sci Food Agric 2016; 96: 1880–1887
Evaluation of culture temperatures TD49/RC212 sequential fermentations were carried out at 20, 25 and 30∘ C to elucidate the effect of culture temperature on the quality of cherry wines. Before inoculation, both yeast strains were precultured in YPD broth medium at 25∘ C for 48 h. For fermentation, TD49 was inoculated at 107 cells mL−1 and 24 h prior to RC212 which was inoculated at 106 cells mL−1 . Analysis of fermentative progression During fermentation, a sample was withdrawn and different aliquots were used for the determination of viable cell counts and TRS. TRS was measured by the official methods21 after the removal of yeast cells by centrifugation. Viable cells was determined by plating onto corresponding medium as stated above using the successive dilution method, and counting visible colonies after incubation at 28∘ C for 2–5 days. After fermentation, in both pure and mixed cultures, the fermented musts were centrifuged (10 min at 5439 g) to remove yeast cells. Then 50 mg L−1 of SO2 was added to fermented musts, which were kept at 4∘ C before analysis. Analytical determinations TRS, TA (expressed as g L−1 of malic acid), pH, ethanol and volatile acidity (VA), were determined according to the official methods.21 Extraction and quantification of volatile compounds The aromatic compounds were extracted by a headspace solid-phase microextraction (HS-SPME) method, conducted on a 50/30 μm DVB/CAR/PDMS (Supelco, Bellefonte PA, USA) fibre, and separated using an Agilent 6890N GC (Agilent, Santa Clara, CA, USA) on a DB-Wax column (60 m × 0.25 mm i.d., 0.25 μm film thickness; J&W Scientific, Folsom, CA, USA) equipped with an Agilent 5975 MS detector. Sample preparation and GC analysis was performed as described previously.22 All analyses were carried out in triplicate. Identification of compounds was determined by comparing mass spectral data from Database (Agilent) and confirmed by comparing Kovats RIs to those of the standards. For quantification of volatile compounds, standard curves were constructed and used, as stated previously.22 Standards were prepared in a synthetic wine (12 v/v % ethanol:water solution, pH 4.10) with 3-octanol as the internal standard. An aliquot of 5 mL of synthetic solution containing various concentrations of volatile standards was diluted in a step-wise manner with synthetic solution using a series of 1:1 dilutions. And the standard curve for individual volatile compounds was obtained by plotting the response ratio of target compound and internal standard against the concentration ratio.
© 2015 Society of Chemical Industry
wileyonlinelibrary.com/jsfa
1881
Evaluation of the inoculation modalities Triplicate fermentations were carried out under static conditions at 25∘ C in 10-L sterile Erlenmeyer flasks which contained 7 L
pasteurised cherry must supplemented with sucrose (to 210 g L−1 ) and potassium metabisulfite (to obtain 50mg L−1 of total SO2 ). The inoculations were carried out as follows: (1) pure culture; (2) sequential fermentation (inoculation of TD49 was 24 h ahead of RC212); and (3) co-culture (simultaneous inoculation of TD49 and RC212). Before inoculation, both yeast strains were precultured in YPD broth medium at 25∘ C for 48 h, then the biomass was centrifuged at 7104 g for 10 min and re-suspended in water. The non-Saccharomyces strain was inoculated at 107 cells mL−1 , while the S. cerevisiae starter was inoculated at 106 cells mL−1 .
www.soci.org Sensory analysis Duplicate sensory evaluation was performed on the cherry wine samples resulting from different inoculations (mono-culture of RC212 and mixed fermentation of TD49/ RC212) and culture temperature (20, 25 and 30∘ C). The aroma descriptors were made by an expert panel comprising 11 tasters (seven women and four men, ranging in ages from 24 to 45 years), and included floral, fruity, sweet, green, almond, fatty and overall aroma. Two samples were evaluated for aroma characteristics in each session. Panellists were required to grade the samples on the basis of a 5-point intensity scale, where 0 = none, 1 = very low, 2 = low, 3 = medium, 4 = high and 5 = very high. Statistical analysis To evaluate the statistical significance of the chemical and sensory evaluation data of the wines, these were subjected to one-way analysis of variance using the SPSS version 16.0 statistical package for windows (SPSS Inc., Chicago, IL, USA). The significant differences among the data were determined using the Duncan test, at P < 0.05.
RESULTS AND DISCUSSION
1882
Effect of inoculation modality on the evolution of yeasts and fermentation kinetics The growth kinetics of biomass in either single or multi-starter fermentations carried out at 25∘ C are presented in Figs 1–3. For the pure cultures (Fig. 1), RC212 finished alcoholic fermentation (AF) within 5 days and generated a maximum viable population of 1.05 × 108 cells mL−1 , while TD49 took 14 days for AF and led to a maximal cell population of 1.84 × 108 cells mL−1 . In comparison with RC212 mono-culture, a much longer fermentation period by the single inoculation of TD49 was linked to a much slower consumption rate of reducing sugar. When AF was conducted with mixed cultures, both simultaneous and sequential fermentations progressed to completion in 5–6 days. Under sequential condition (Fig. 2), both yeast strains co-dominated the fermentation process. TD49 showed greater persistence during the first 5 days and achieved a higher cell population of 1.39 × 108 cells mL−1 . The increase in TD49 viable cells led to a limited reduction in the number of viable cells of RC212, and 18% lower viable population was obtained relative to its mono-culture. This may be ascribed to the enhanced competition of the non-Saccharomyces yeast in the delayed inoculation of S. cerevisiae strain,8,23,24 which will probably be caused by competition for nitrogen, as has been previously noted.25 However, when the two starters were inoculated at the same time (Fig. 3), RC212 quickly adapted to the cherry musts, and maintained a relatively higher growth rate for approximately 3 days. It reached a viable population of 9.95 × 107 cells mL−1 on the third day, and maintained this level almost until the end of fermentation. In contrast, TD49 strain was significantly influenced by the proliferation of RC212, as seen by the fact that TD49 slowly increased during the initial AF and dropped dramatically after 5 days. These observations suggest that certain interaction may occur between the two yeast strains, which led to a sharp decline of TD49 near the end of AF. This would probably be ascribed to a fierce competition for the use of nutrients (sugar, nitrogen, oxygen) and/or space,15,26,27 but other possibilities, such as inhibition caused by certain killer toxins or some other metabolites synthesised by S. cerevisiae,10,28 – 30 cannot completely be eliminated. From the performance of TD49
wileyonlinelibrary.com/jsfa
SY Sun et al.
over AF either in the case of co-culture or sequential culture, it was found that TD49, like many other non-Saccharomyces yeasts, does not strongly compete with other fermenting yeasts, as many commercial Saccharomyces strains do, thus enabling the maintenance of higher yeast diversity during the fermentation process. Effect of inoculation modality on the compositional and volatile profiles Table 1 (columns 2–5) shows the compositional profiles from pure cultures and from those mixed fermentations under different inoculation conditions. It was seen that the physico-chemical parameters of these wines were all within the rank of values accepted for regular wines. As expected, cherry wines produced from TD49 single inoculation were characterised by the highest TRS and the lowest ethanol content, whereas RC212 single culture resulted in the opposite result (lowest TRS and highest ethanol). In comparison with S. cerevisiae mono-culture, the mixed fermentation based cherry wines contained more residual sugar and consequently lower ethanol levels. Also, the TA and L-malic acid displayed higher levels in those two wines, particularly in that with the sequential inoculation. Another important feature that needs to be highlighted is the marked decrease in the volatile acidity by multi-starter cultures, as evidenced by the fact that co-culture led to a level of 0.33 g L−1 , and sequential inoculation generated a value of 0.34 g L−1 , both evidently lower than that from RC212 pure culture (0.42 g L−1 ). The non-Saccharomyces yeasts have been historically considered as wine-making contaminants due to their production of high amounts of acetic acid.7,31 Comitini and Renault have found that, unlike many apiculate strains, isolates of T. delbrueckii did not form high levels of VA.9,16 Our results, using strain TD49 for the production of cherry wines, was in agreement with their reports, and at the same time, clearly show the promising oenological feature of this strain. For volatile compounds determined (Table 2, columns 3–6), pure culture of TD49 produced the lowest total amount, then follows RC212 mono-culture and RC212/TD49 co-fermentation, and the sequential mixed culture based cherry wine showed the highest total content. Monoculture wines demonstrated that RC212 was a stronger producer of volatile esters, alcohols and acids, which were 1.2-, 1.6- and 1.7-fold greater than those found in TD49 single culture, respectively. With regard to mixed fermentations, the wines from sequential culture and simultaneous culture showed similar values in the total amount of volatile components, but significant differences between their particular profiles were observed. For specific compounds identified, sequential culture markedly enhanced the production of several volatile alcohols, esters and varietal compounds, mainly represented by ethyl 3-methylbutanoate, ethyl hexanoate, ethyl hex-3-enoate, ethyl octanoate, 2-methyl-1-propanol, 3-methyl-1-butanol, 𝛽-phenylethanol and linalool. Of note, the yield of 𝛽-phenylethanol was 2- and 1.2-fold higher in this wine than that in RC212 mono-culture and co-fermentation wines, respectively. This observation was in agreement with previous findings, as reported by Comitini who found that in grape wines, mixed fermentation involving T. delbrueckii would significantly enhance the production of this compound.9 Descriptive analysis of cherry wines resulting from different inoculation strategies Figure 4 shows the mean rating of aroma attributes of cherry wines obtained from different inoculation modalities. The RC212
© 2015 Society of Chemical Industry
J Sci Food Agric 2016; 96: 1880–1887
Bacterial cultures to improve cherry wine
www.soci.org
8.5
Log CFU mL–1
200 7.5 150
RC212 ( ) TD49 ( ) Sugar consumption by RC212 ( ) Sugar consumption by TD49 ( )
100
6.5
Recucing sugar (g L–1)
250
50
5.5 0
1
2
3
4
5
6
7 8 9 Time (days)
10
11
12
13
14
15
0
Figure 1. Growth kinetics and sugar consumption during cherry wine-making by single cultures (25∘ C). Values are mean of triplicates ± SE.
6
5
50
0
1
2
3
4
5
6
7
8
200
8
150 7
RC212 ( ) TD49 ( ) TRS( )
6
5
0
50
0
1
Time (days) Figure 2. Growth kinetics and sugar consumption during cherry wine-making by sequential cultures at 25∘ C. Values are mean of triplicates ± SE.
mono-culture based cherry wine exhibited moderate intensities for fruity and floral aroma, and low scores for fatty, almond, sweet and green notes. In the case of mixed fermentations, it was found that both multi-starter cultures markedly increases the wines ‘floral’ and ‘fruity’ descriptors which was particularly evident in the
100
2
3 4 5 Time (days)
6
7
8
Reducing sugar (g L–1)
100
Log CFU mL–1
150
RC212 ( ) TD49 ( ) TRS( )
7
Reducing sugar (g L–1)
Log CFU mL–1
200
8
250
9
250
9
0
Figure 3. Growth kinetics and sugar consumption during cherry wine-making by co-cultures (25∘ C). Values are mean of triplicates ± SE.
case of sequential fermentation trials, and simultaneously reduced the wines ‘green’ and ‘fatty’ nuances that showed no significant differences between co-culture and sequential fermentation. ‘Almond’ is the only character that remained nearly unchanged when a mixed fermentation strategy was applied, when compared to a S. cerevisiae single fermentation.
Table 1. The analytical profile of cherry wines resulting from different inoculations (columns 2–5) and different temperatures (columns 6–8)
Parameter
RC212 (25∘ C)
TD49 (25∘ C)
Sequential culture (25∘ C)
Fermentation time (days) Ethanol (%, v/v) TRS (g L−1 ) pH TA (g L−1 ) L-Malic acid (g L−1 ) VA (g L−1 )
5 10.81 ± 0.20b 0.96 ± 0.02a 4.07 ± 0.02a 6.49 ± 0.04a 2.24 ± 0.02a 0.42 ± 0.04b
14 8.19 ± 0.18a 13.83 ± 0.05d 4.04 ± 0.02a 6.74 ± 0.06c 2.49 ± 0.02c 0.29 ± 0.02a
6 10.30 ± 0.20a 3.25 ± 0.04c 4.06 ± 0.02a 6.60 ± 0.04bc 2.31 ± 0.02b 0.34 ± 0.02ab
Co-culture (25∘ C) 5 10.36 ± 0.23a 1.44 ± 0.02b 4.06 ± 0.02a 6.52 ± 0.04ab 2.29 ± 0.02b 0.33 ± 0.03a
20∘ C (Sequential)
25∘ C (Sequential)
30∘ C (Sequential)
10 10.28 ± 0.20A 3.62 ± 0.04C 4.06 ± 0.02A 6.58 ± 0.06B 2.26 ± 0.02A 0.35 ± 0.04A
6 10.30 ± 0.20A 3.25 ± 0.04B 4.06 ± 0.02A 6.60 ± 0.04B 2.31 ± 0.02B 0.34 ± 0.02A
5.5 10.33 ± 0.22A 0.98 ± 0.02A 4.07 ± 0.02A 6.46 ± 0.04A 2.28 ± 0.02AB 0.38 ± 0.05A
a–d In
J Sci Food Agric 2016; 96: 1880–1887
© 2015 Society of Chemical Industry
wileyonlinelibrary.com/jsfa
1883
columns 2–5 (cherry wines resulting from different in inoculations), values in the same row with different lower-case superscript letters are significantly different, according to Duncan’s test (P < 0.05). A–C In columns 6–8 (cherry wines obtained at different temperatures), values in the same row with different upper-case superscript letters are significantly different, according to Duncan’s test (P < 0.05). TA was expressed as g L−1 of malic acid.
www.soci.org
SY Sun et al.
Table 2. Concentrations of volatile compounds in cherry wines resulting from different inoculations (columns 3–6) and different temperatures (columns 7–9) (in milligrams per litre, n = 3)
RI
Compound
RC212 (25∘ C)
Sequential TD49 (25∘ C) culture (25∘ C)
Esters 0.86 ± 0.10b Ethyl acetate* Ethyl butyrate 0.14 ± 0.01a Ethyl 3-methylbutanoate 0.03 ± 0.01a 3-Methylbutyl acetate 0.37 ± 0.04ab Ethyl hexanoate 0.11 ± 0.02a Ethyl hex-3-enoate 0.39 ± 0.04a Ethyl hex-2-enoate 0.41 ± 0.05a Ethyl lactate 7.48 ± 0.81b Ethyl octanoate 0.66 ± 0.07a 2-Phenylethyl acetate 0.13 ± 0.02ab Sub-total 10.58 Alcohols 1029 1-Propanol 6.94 ± 0.72b 1076 2-Methyl-1-propanol 4.04 ± 0.44b 1136 1-Butanol 0.34 ± 0.03a 1193 3-Methyl-1-butanol 2.98 ± 0.0.31b 1342 1-Hexanol 7.55 ± 0.84b 1868 Benzyl alcohol 5.45 ± 0.52b 1896 𝛽-Phenylethanol 4.96 ± 0.50a Sub-total 32.26 Acids 1445 Acetic acid 14.38 ± 1.55c 1556 2-Methylpropanoic acid 1.28 ± 0.14b 1620 Butanoic acid 0.35 ± 0.03c 1662 2-Methylbutanoic acid 0.58 ± 0.05c 1840 Hexanoic acid 1.6 ± 0.17c 2051 Octanoic acid 1.08 ± 0.11b Sub-total 19.27 Aldehydes
