Scaling-up in industrial winemaking using low ... - Wiley Online Library

Journal of Applied Microbiology ISSN 1364-5072

ORIGINAL ARTICLE

Scaling-up in industrial winemaking using low electric current as an alternative to sulfur dioxide addition G. Lustrato1, G. Alfano1, C. Belli1, L. Grazia2, M. Iorizzo1 and G. Ranalli1 1 DISTAAM, Universita` del Molise, Campobasso, Italy 2 DIPROVAL, Universita` di Bologna, Reggio Emilia, Italy

Abstract

Keywords grape must, industrial scale, low electric current, sulfur dioxide, winemaking. Correspondence G. Ranalli, DISTAAM, Facolta` di Agraria, University of Molise, Via De Sanctis 46, 86100 Campobasso, Italy. E-mail: [email protected]

2005/0362: Received 6 April 2005, revised 11 January 2006 and accepted 22 January 2006 doi:10.1111/j.1365-2672.2006.02931.x

Aims: To better understand the outcome of employing low electric current (LEC) technology as a new preservation and alternative in wine technology, and to contribute to its development. It is used in industrial-scale winemaking with commercial yeast (Saccharomyces cerevisiae) during the grape must fermentation. Methods and Results: LEC (200 mA, time 16 days) was applied to fresh grape must as an alternative method to the usual sulfur dioxide addition used in the industrial process; two tanks, each 30 000 l, were employed for parallel fermentations. The results show that LEC decreased the survival time and increased the death rate of apiculate yeasts, whereas it did not affect the growth and survival of S. cerevisiae. A comparison was made of the main chemical and sensory parameters of the wines obtained. Conclusions: The results have demonstrated that the low-voltage treatment had a positive effect on the grape juice fermentation (yeast microflora) during the early stages of winemaking. Significance and Impact of the Study: These results could be of significant importance in developing, for ‘biological wine’, new winemaking technologies for an innovative control process of yeast fermentation.

Introduction In recent years, much publicity has been given to consumer concerns about the addition of chemicals to food. One such concern is the interaction of additives, and the possible risk to human health of the joint effects of the cocktail of additives consumed everyday (Adams 1997). Sulfur dioxide (SO2) in its various forms is used to help preserve all types of foods, including fruits, vegetables and derived drinks. Sodium metabisulfite (MBS) is used as a preservative in food and wine and frequently triggers asthma attacks (Wright et al. 1990; Vally et al. 2000; Vally and Thompson 2001). In Canada, there have been reports of more than 100 sulfite (SO3)–related reactions and at least one death (Yang 1989). A few individuals appear to have an IgEmediated sensitivity to SO3, but others are thought to be hyper-reactive to inhaled SO2 (Taylor and Dormedy 682

1998), or have a deficiency in SO3 oxidase, an enzyme responsible for oxidizing SO3 to inactive sulfate (SO4). Indeed, in hypersensitive individuals, levels of inhaled SO3 as low as 1 ppm have been known to trigger reactions, but it is generally believed that most SO3-sensitive individuals are able to tolerate levels of up to 10 ppm in a food matrix (Simon 1986). Reactions vary, depending on the form of the SO3, the amount present and the mechanism of SO3 sensitivity (Wedzicha 1984; Taylor et al. 1988), and reaction severity ranges from nausea, abdominal pain, diarrhoea and asthma to anaphylactic shock (Yang and Purchase 1985; Yang 1989). It is extremely rare for nonasthmatic subjects to have adverse reactions to SO3, but the risk of SO3 reaction for asthmatics has been estimated to be almost 4% of such patients (Bush et al. 1986). In the United States and other countries, SO3 use in fresh and processed foods is limited by regulations (Anderson 1994), and in Europe, the recent European

ª 2006 The Authors Journal compilation ª 2006 The Society for Applied Microbiology, Journal of Applied Microbiology 101 (2006) 682–690

G. Lustrato et al.

Scaling-up in industrial winemaking using LEC as an alternative to SO2 addition

regulation CE 2003/897 has made it compulsory for labels to indicate SO3 presence and amounts >10 mg l)1. In winemaking, the addition of SO2 to grape juice for controlling oxidation reactions and restricting the growth of indigenous microflora is a well-established practice (Usseglio-Tomasset 1992; Romano and Suzzi 1993). As an antimicrobial agent, SO2 can be expected to have the following effects on the kinetics of yeast growth during fermentation: (i) increase the lag phase and delay the onset of fermentation; (ii) decrease the growth rate and increase the time to complete fermentation; (iii) accelerate the decline/death phase; (iv) have selective effects on the species or strains that grow and contribute to the fermentation. The growth studies of Goto (1980) and Heard and Fleet (1988) clearly show that the addition of 50 mg l)1 SO2, or more, to grape juice decreases, about tenfold, the initial yeast population, and results in a 1–2-day lag phase before the onset of fermentation. The principal yeast of wine fermentation, Saccharomyces cerevisiae, is not predominant in must flora, which consists mostly of apiculate yeasts (genera Kloekera and Hanseniaspora) and various species of other yeasts (Zambonelli et al. 1989; Rainieri and Pretorius 2000). One of the desired objectives of adding SO2 to the must is to suppress the growth of these non-Saccharomyces spp. and to selectively encourage the growth and dominance of S. cerevisiae (Romano and Suzzi 1993). The growth of indigenous yeasts (Kloeckera and Candida spp.) has been found consistently in commercial wine fermentations to which customary SO2 concentrations (50–100 mg l)1 total) had been added to the juice (Fleet et al. 1984; Mora et al. 1990; Fleet and Heard 1993). Indeed, more specific studies have revealed that the growth of Kloekera apiculata is not inhibited by SO2 concentrations of 100–150 mg l)1 (Heard and Fleet 1988). These findings raise questions about the efficacy of SO2 in controlling indigenous yeasts, and challenge one of the basic reasons for using SO2 in winemaking. Furthermore, wines fermented in the presence of SO2 have been found to have higher concentrations of acetaldehyde than those fermented in its absence (Herraiz et al. 1989, 1990). Acetaldehyde is extremely reactive and can react with amino acids to generate various flavour compounds (Griffith and Hammond 1989). To minimize the occurrence of such SO3-binding compounds, an improved understanding of the factors affecting SO2 binding in grape juice formulations and fermentation processes is needed. In response to public health issues (WHO) related to the presence of SO2 in food and beverages, the maximum permitted level in wine has, over recent decades, been progressively decreased, and some winemakers have reduced the use of this preservative quite significantly

(Ough 1983; Stockley et al. 1993; Peterson et al. 2000). However, this seems to have increased the risk of microbial spoilage of the wine after bottling (Godden 2000). Let us take a brief look at the situation in a few countries. In France, it was once quite common for white wines to have SO2 levels of approximately 300 mg l)1, but nowadays the values are much lower (Blouin 1993). More than 4000 American wines in a recent survey were revealed to have an average total SO2 concentration of 74 mg l)1 (Peterson et al. 2000). In Australian wines, especially red, the general increase in the maximum pH (from pH 3Æ6–3.7 to 3Æ9–4Æ5) in conjunction with low total SO2 (30–80 mg l)1) has been found to seriously reduce the effective molecular SO2 concentration in these wines (Bruer et al. 1999). Indeed, for some wines, the risk of permissive conditions for microbial growth has greatly increased, especially for non-S. cerevisiae, lactic acid and acetic acid bacteria (Bartowsky et al. 2003). Given the dramatic increase in consumer demand for new products free of additives, and with high sensory organoleptic and nutritional qualities, the search for new food processing alternatives has gained impetus. The most common way of increasing shelf life and maintaining food safety in the food industry has been to adopt processing treatments aimed at inactivating spoilage and pathogenic micro-organisms (Barbosa-Canovas et al. 2001; Devlieghere et al. 2004). In recent decades, the use of electrical treatment in food processing has progressively increased: in fact, more and more research studies are being directed towards achieving the inactivation of micro-organisms in the food system (Vega-Mercado et al. 1997; Knorr and Heinz 2001; Bendicho et al. 2002; Ulmer et al. 2002; Abram et al. 2003). Indeed the inactivation of bacteria and yeast cells by electrochemical means is now well documented (Davis et al. 1994; Gaskova et al. 1996; Grahl and Markl 1996; Velizarov 1999; Wouters et al. 1999, 2001; Heinz et al. 2003). In the food industry, pulsed electric-field treatment has become of increasing interest as it offers some attractive advantages over methods currently used in processing raw materials and foods (Heinz et al. 2002; Abram et al. 2003). In fact, the application of external electric fields to biological cells causes alteration in the membrane structure, leading to pore formation. Under mild pulsation conditions, membrane pore formation is reversible, whereas more drastic conditions lead to the irreversibility of the phenomenon that eventually results in cell death (Weaver and Chizmadzhev 1996; Barbosa-Canovas et al. 1999). The factors crucial to inactivation can essentially be classified as the processing parameters, medium properties and microbial characteristics (Wouters et al. 2001). In a recent study involving winemaking, our objective was to reduce the risk of wine spoilage caused by indi-


683


genous yeast growth. Our previous research had demonstrated, on both the laboratory scale and in pilot plants, that low-voltage treatment has a positive effect on grape juice fermentation. This result prompted us to apply, in the absence of SO2, an LEC to grape juice to obtain fermentation with a highly active S. cerevisiae starter culture (Ranalli et al. 2000, 2002). Our findings reveal the very evident advantages correlated with such treatment and raise the strong possibility of it providing a valid alternative to SO2 addition (Lustrato et al. 2003). The objectives of the present work were to verify the use, on the industrial scale, of LEC to control the process of winemaking, to assess the potential of this approach as an alternative to the traditional addition of SO2, to monitor the microbial activity of the fermenting grape must process during the LEC treatment and to characterize the finished wine by chemical and sensory parameters. Materials and Methods Starter yeast strain A commercial yeast, S. cerevisiae (Australys white arom, Oliver Ogar Italia Verona, Italy), was used as the inoculum for the industrial winemaking process. Industrial winemaking process For the 2003 vintage using Trebbiano white grapes, fermentation was carried out by an industrial winemaker, the Cooperativa Vitivinicola Miglianico (Miglianico, Italy), located in the Abruzzo region, southern Italy. The industrial fermentation took place, according to the usual vinification of white wines, in two stainless steel (AISI 304 quality) cylindrical fermentation tanks with isolation jackets: diameter 4Æ0, height 3Æ8, full capacity 30 000 l (Di Zio, Spoltore, Italy). We set up two tests: must subjected to LEC treatment (Test A) and must with added SO2 (Test B). Figure 1 shows the flowchart of the adopted winemaking process. After 16 days, the samples were clarified at 4C for 24 h, decanted and analysed immediately. The main chemical characteristics of the grape must were: pH 3Æ5, sugars 175Æ4 g l)1, total acidity 5Æ25 g l)1 and malic acid 2Æ98 g l)1. LEC power device The LEC power device is an operating electrolytic unit (R. De Ponti Electronics, Treviglio, Italy) composed of two independently working sections. Each section has a pair of electrodes that can be positioned as required and in which the electric power can be regulated. The two 684

G. Lustrato et al.

electrodes work independently (Ranalli et al. 2002; Zanardini et al. 2002; Lustrato et al. 2003). Type of electrode For our trial, the electrode adopted was of titanium, which is chemically resistant and mechanically robust. This titanium electrode/anode was coated with mixed metal oxide (MMO) that has excellent electro-catalytic properties (cylindrical shape and 200-cm length) (Metakem GmbH, Usingen, Germany). Microbial monitoring of fermentative process Microbiological counts of the yeast populations were made at three different times on yeast peptone dextrose (YPD) agar, Wallerstein (WL) nutrient agar and Lysine agar media (Oxoid), in petri dishes. The WL nutrient medium was also used in the selection and counting of elliptical and apiculated yeasts (Ranalli et al. 2002). The non-Saccharomyces yeast spp. were isolated selectively and enumerated on lysine agar (Heard and Fleet 1986). Yeast classification into genera and species was done on the basis of morphological and physiological properties (Kurtzman 1990). Ten yeast colonies were then randomly isolated on each of the YPD plates and characterized according to the criteria and methods described by Yarrow (1998). ATP assay on the first fermentative process was performed using a specific enzymatic kit modified by Jago et al. (1989) (NRM/Lumit-Qm, code 9332-l; Lumac B.V., Landgraaf, the Netherlands) (Ranalli et al. 2002). A Biocounter 1500 P luminometer (Lumac B.V.) equipped with a photomultiplier tube set at 7200 RLU with 200-pg ATP in 100 ll of Lumit buffer and Lumit-QM reagent was used (Ranalli et al. 1996, 1998). Chemical analyses From start-up and throughout the fermentative process, the pH, titratable acidity, ethanol and sugar concentration of the fresh grape juice and the wines were assessed using standard methods (Anonymous 1990). The quantification of glycerol, acetic, malic and succinic acids was done by enzyme assay (TC glycerol no. 148270, TC acetic acid no. 148261, TC malic acid no. 139168, TC succinic acid no. 176281; Boehringer Mannheim, Germany). Sensory analysis After fermentation was complete, the wines were evaluated by a group of seven expert panellists. All the panellists were DISTAAM University staff members, aged from 20


G. Lustrato et al.


Grape harvesting

Transportation

Destemming

Crushing

Juice separation Test A Yeast inoculation LEC

Figure 1 Flowchart of winemaking process. Must submitted to low electrochemical current treatment (LEC; 200 mA, Test A); must with added sulfur dioxide (SO2; 80 mg l)1, Test B).

to 40 years and experienced in wine sensory analysis. The wines were presented in random order at 10C in coded standard 3951 glasses (ISO 1977), covered with a watch glass to minimize the escape of volatile components. The wines were first compared in a blind tasting, in which the panellists developed a list of descriptors. Sampling was done in randomized complete blocks, each treatment replicate being sampled twice by each panellist. The wines’ sensory profile (floral, reduced, oxidized, sweaty, acetic and body) was conducted at 21–23C in individual booths under white illumination. Twenty millilitre of each sample was presented in a random order at 10C in coded, clear, 170-ml wine-tasting glasses covered with a watch glass in each session. Mineral water was provided at each wine sampling session, to rinse the mouth between wines. Sensory characteristics were quantified on a 0–10 scale using descriptors developed through free choice profiling (Meilgaard et al. 1991; Egli et al. 1998). Samples were compared using

Test B Yeast inoculation

Juice clarification Fermentation

Fermentation

Racking

Racking

Clarification

Clarification

Stabilization

Stabilization

Filtration

Filtration

Aging

Aging

Bottling

Bottling

Storage

Storage

SO2

the so-called spider web diagrams. In this diagram, the centre of the figure represents low intensity, with the intensity of each attribute increasing to an intensity of ten at the perimeter. All the samples were evaluated in a fully equipped tasting room according to ISO standard 8589 (ISO 1988). Statistical analysis All the data concerning microbial counts and chemical responses underwent statistical analyses using the SAS statistical software package (SAS Institute Inc. 1989). Results Industrial scale process In the direct inoculation method, the commercial yeast S. cerevisiae was suspended in warm tap water at 35C


685


Survival (log10 CFU ml–1)

7 6 5 4 3 2 1

0

1

2

3 4 Time (day)

5

6

Figure 3 Survival dynamics of Saccharomyces cerevisiae cells in fresh must at different times and for different treatments. ( ) Fresh grape must; ( ) reactor A: grape must + electric current at 200 mA; (h) reactor B: grape must + addition of sulfur dioxide (SO2), 80 mg l)1.

14 12 ATP (ng ml–1)

according to manufacturer’s instructions (Oliver Ogar Italia, Verona, Italy). After 20 min, the yeast viable cell suspension was divided into two homogeneous aliquots (540 l, 2% v/v, given an initial inoculation of log 5Æ8 ± 0Æ2 CFU ml)1) that were then added simultaneously to two parallel fermentors A and B with a working volume of 27 000 l each. The first fermentor (A) was equipped with an MMOcoated-titaniumanode electrode located perpendicularly in the centre of the tank and suspended from the top of the steel-tank reactor. The current intensity applied was 200 mA and the dispersion area was 200 cm2 with a treatment time of 16 days. The second electrode (cathode/anode) was considered to be the inner surface of the same steel reactor. To the second fermentor (B) was added 80 mg l)1 of SO2 and no electrode/electric current was applied. The reactors were supplied with a programmed temperature control system and the temperature of the must fermenting in reactors A and B was monitored throughout the trial period. The average temperature of the must over the 2-week period was 16C. The main microbiological characteristics of the grape must were: non-Saccharomyces yeasts 5 · 105 CFU ml)1 (Hanseniaspora guilliermondii 4 · 105 CFU ml)1, Kloeckera apis 4 · 105, Candida stellata 2 · 105 CFU ml)1); S. cerevisiae yeasts 5 · 102 CFU ml)1. Figures 2–5 and Table 1 show the results of the present trial performed in the plant on the industrial scale. Figure 2 reveals that there was no significant variation between the two treatments (reactor A, electric current at 200 mA and reactor B, addition of SO2, 80 mg l)1) in the viable cell count of non-Saccharomyces over time. Both the LEC treatment and the SO2 addition to the must resulted in a constant decrease in viable cells, although there were quite marked effects at 24 h and after 5 days.

G. Lustrato et al.

10 8 6 4 2 0

0

1

2

3 4 Time (day)

5

6

Figure 4 Mean ATP content in must fermentation at different times and for different treatments. ( ) Fresh grape must; ( ) reactor A: grape must + electric current at 200 mA; (h) reactor B: grape must + addition of sulfur dioxide (SO2), 80 mg l)1.

Survival (log10 CFU ml–1)

7 6

Floral 8

5 4

Diacetyl

3 2

Body

1 0 0

1

2

3 4 Time (day)

5

6

Acetic Herbaceus vegetative

Figure 2 Survival dynamics of non-Saccharomyces cells in fresh must at different times and for different treatments. ( ) Fresh grape must; ( ) reactor A: grape must + electric current at 200 mA; (h) reactor B: grape must + addition of sulfur dioxide (SO2), 80 mg l)1.

686

6 4 2 0

Reduced

Oxidized

Sweaty Astringent phenolic

Figure 5 Spider web diagrams of the mean scores of the sensory attributes of the Trebbiano white wines. ( ) Wine + low electric current (LEC); (r) wine + sulfur dioxide (SO2).


G. Lustrato et al.


Table 1 Chemical characteristics of wines harvested in 2004 pH

Total acids Volatile acids Malic Succinic (g l)1 tartaric Sugars Alcohol (g l)1 acetic acid acid Glycerol acid) (g l)1) (% vol.) acid) (g l)1) (g l)1) (g l)1)

Trials + SO2 3Æ52 4Æ8 Trials + 200 mA 3Æ60 3Æ86