SPONTANEOUS FERMENTATION IN WINE PRODUCTION AS A CONTROLLABLE TECHNOLOGY

This study focuses on the isolation of a consortium of microorganisms from spontaneously fermenting must that naturally contain lactic acid bacteria, non-saccharomyces yeasts, and saccharomyces yeasts. To collect the greatest diversity of microorganisms, the consortium was taken from the point of micro-sparkling. Based on the growth curves, isolation was performed using individual special nutrient media, and the isolates were subsequently multiplied in the nutrient medium. Individual isolates were then used for fermentation tests to monitor the percentage of fermented sugar and hydrogen sulphide production. The highest fermentation abilities were achieved in the isolates containing Saccharomyces cerevisiae and Zygosaccharomyces bailii. The smallest amount of ethanol was formed from the isolates containing Hanseniaspora uvarum, while Candida sake isolate produced the lowest amount of hydrogen sulphide and Zygosaccharomyces bailii produced the highest. The other isolates produced an average amount. Based on these results, a consortium containing the given isolates in a certain ratio was compiled.


INTRODUCTION
Saccharomyces cerevisiae (S. cerevisiae) is a type of yeast that performs alcohol fermentation and is widely used as a fermentation starter. During the alcoholic fermentation of grape must, S. cerevisiae becomes the dominant species with the increasing ethanol concentration (Mas, Guillamón and Beltran, 2016). Therefore, the isolation of natural S. cerevisiae is generally carried out from must be fermented by spontaneous fermentation Oenococcus oeni is generally the main species used as an MLF starter due to its easy adaptation to wine conditions. New strains such as L. plantarum are currently being discovered, however, that can also survive and adapt well to the viticulture. These strains also have more favorable biological properties compared to O. oeni, such as higher growth rate, creation of a more complex aromatic profile, and the prevention of undesirable by-product formation (du Toit et al., 2011;Brizuela et al., 2017).
Therefore, the aim of this study was to isolate yeast and lactic bacteria from the vineyard and carry out a fermentation experiment using these isolates.

Scientific hypothesis
Each vineyard has its own wine of microbiota, which affects the quality of the resulting wine. Separation and cultivation methods can be used to characterization and multiplying individual microorganisms. Based on the fermentation and H 2 S production tests, the resulting consortium can be assembled, which can be used for fermentation, thereby promoting both the fermentation process and the terroir of the wine.

MATERIAL AND METHODOLOGY
The aim was to obtain a functional consortium of wine microorganisms that was characterized in terms of biotechnology and taxonomy. This consortium was separately cultivated to achieve the required amounts and then used to inoculate a drained batch of Hibernal must in which fermentation and H 2 S production were monitored.

Materials
MEA+T Malt Extract Agar with Tetracycline (a broadspectrum antibiotic against both gram-positive G+ and gram-negative G-bacteria). Only eukaryotic microorganisms grow on this medium.

Isolation of yeasts
The must of the Hibernal variety was fermented spontaneously. During this spontaneous fermentation, 20 mL of the matrix (must, fermentation must, wine) was taken at selected monitoring points (must, micro-sparkling, 4 and 8 vol. % ethanol, end of fermentation). The sample obtained was subsequently diluted using the so-called decimal series. From each dilution, 250 mL was pipetted onto Petri dishes with MEA+T, WLN, and MRS culture medium, and a microbiological rod smear was performed. The Petri dishes were then placed in a thermostat (30 °C; WLN and MEA+T -3 days; MRS -7 days).
At the end of the cultivation, the total number of microorganisms and individual colonies was enumerated. Based on a combined analysis of phenotypic characters (macroscopic and microscopic properties), a sampling point was selected (the most suitable was the microsparkling point -the diversity of technologically important microorganisms was evaluated -saccharomyces and nonsaccharomyces yeasts and lactic bacteria) to serve as a source for consortium acquisition.

Determination of growth characteristics of individual
isolates of the 2018 Wine Microorganism Consortium (growth curves) The aim was to determine the growth characteristics of individual isolates. Individual isolates of microorganisms were pre-cultured in standard media (yeast -YPD; lactic acid bacteria -MRS) with the following culture conditions: 30 °C; shaking 120 rpm; yeast 24 h; and lactic acid bacteria 72 h. The obtained cell suspensions were centrifuged (10 mins; 10 °C; 10,000 rpm), washed with saline solution, and then resuspended in the selected media (yeast -YPD, ME, YPD mod ; lactic acid bacteria -MRS, YPD, and YPD mod ) so that the resulting optical density value of the suspension was 0.2 at a wavelength of 600 nm.
The obtained suspension was then pipetted onto Bioscreen C culture plates (Oy Growth Curves Ab Ltd). Each arrangement (microorganism x medium) consisted of five repetitions to ensure the achievement of relevant results.

Volume 14
The culture conditions of the Bioscreen C device were set as follows: 30 °C; shaking every 3 mins; duration of one shaking cycle = 1 min; and the so-called wide band of wavelengths -WB (420 -620 nm). The maximum growth rate µ (h -1 ) was then calculated from the measured data, and the maximum optical density OD MAX was determined. These data, along with the course of growth curves, served to assign culture media to individual isolates.

Identification of microorganisms
The MALDI TOF (matrix-assisted laser desorption/ionization coupled with time of flight mass spectrometry) method was used to identify microorganisms. It is a very accurate, simple method, able to determine high molecular weight substances, proteins, peptides, lipids, nucleic acids, carbohydrates (Huong et al., 2014). An essential part of the MALDI TOF measurement was the preparation of fresh α-cyano-4-hydroxycinnamic acid. The organic solvent was prepared by mixing 500 µL of acetonitrile (100%), 475 µL of distilled water, and 25 µL of trifluoroacetic acid. Before use, 250 µL of organic solvent was added to the plastic tube. The contents of the tube were vortexed until the complete dissolution of the crystals. α-cyano-4-hydroxycinnamic acid was stored in the dark place and its preparation is ideal the day before the measurement.
The cultures were applied to the clean metal plate for MALDI TOF and the culture was allowed to dry on the plate. It was then covered with 1 microliter of α-cyano-4hydroxycinnamic acid. At the same time, it was important to homogenize the sample and matrix (Jarolímková, 2017). Unlike the analysis of bacteria, preprocessing of the yeast isolates was required to extract fungal proteins. The protein extraction method used to process yeast isolates for MALDI-TOF MS was adapted directly from established methods used to identify difficult bacterial isolates. Specifically, 1 to 5 colonies of an isolate were inactivated in 75% ethanol, pelleted, and then suspended in a 1:1 mixture of 70% formic acid and acetonitrile. The resulting supernatant was then analyzed by MALDI-TOF MS  Table 5.

Optaining pure cultures: Separation and lyophilisation
This procedure aimed to cultivate individual isolates of technologically important microorganisms and preserve them using the lyophilization method. Separate cultivation of individual microbial isolates was performed based on the information obtained from the growth characteristics. Individual taxa were first pre-cultured in 250 mL Erlenmeyer flasks (100 mL medium volume; orbital stirring 120 rpm; 20 °C). The media and the culture times are shown in (Table 3). The pre-cultured cell suspension was examined microscopically (cell morphology, elimination of contamination) and centrifuged (10,000 rpm; 10 mins; 10 °C).
After separating the supernatant, the pellet was washed with saline solution and re-centrifuged (10,000 rpm; 10 mins; 10 °C) and resuspended in the pure culture medium. The prepared suspension served as the inoculum for the second cultivation stage, which was carried out in 2,000 mL Erlenmeyer flasks (1,000 mL medium volume; orbital stirring 100 rpm; 30 °C). The media and the culture times used for the individual isolates are shown in (Table 5).
After the cultivation was complete, the suspension was repeatedly centrifuged and washed as described in the previous step. The obtained biomass was mixed with cryoprotective medium and shock-frozen (70 °C; 24 h). The frozen suspension was then lyophilized. The viability of the obtained dehydrated biomass was then determined, and according to the qualitative and quantitative microbiological analysis (Table 2) and the cell viability in the lyophilisate, the 2018 Wine Microorganism Consortium was compiled.

Fermentation tests
Individual yeast isolates were initially cultured in 50 mL Erlenmeyer flasks (25 mL medium volume; orbital stirring 120 rpm; 30 °C). The media and the culture times used for the individual isolates are shown in (Table 5). In the obtained cell suspension, the cell density was determined by microscopic cell counting in a so-called Bürker chamber. The calculated amount of this suspension was then pipetted to a final concentration of 10 8 cells.mL -1 in a 250 mL Erlenmeyer flask (100 mL YPDm medium volume; without shaking; 25 °C). Fermentation was monitored by the gravimetric method, and weight loss due to the metabolic conversion of fermentable sugars to carbon dioxide and ethanol was observed.

H 2 S production
Individual isolates were inoculated onto a BIGGY agar identification medium using a microbiological loop. Individual Petri dishes were statically cultured at 30 °C for 3 days. Based on the visual evaluation, the individual isolates were marked as low, medium, and high producers of H 2 S.
BiGGY, Bismuth Sulphite Glucose Glycine Yeast Agar, is based on the formulation developed by Nickerson (1953) and mainly used for the isolation and presumptive identification of Candida species. In a study of sulphite reduction by yeasts, the ability of many yeasts to reduce a bismuthyl hydroxy polysulphite was noted. Growth on an acidic or neutral medium containing bismuth sulphite produced black colonies because of the extra-cellular reduction of the bismuth sulphite, to bismuth sulphide. The bismuth sulphite complex confers a high degree of selectivity to the medium, and most strains of bacteria are inhibited on BIGGY Agar. In this study, BIGGY agar was used as a simple and rapid method to compare the rate of H 2 S production between pure yeast isolates.

Statistical analysis
Statistical analyses and figures were generated using Excel 2007 software packages (manufactured by Microsoft Office, USA) and Statistica 10 statistical software (Copyright © StatSoft). The Statistica 10 software was used to process growth curves data and create their line graphs.

RESULTS AND DISCUSSION Isolation of yeasts
The results of the microbiological analysis are shown in (Table 1). These data are comparable with the normal course of fermentation of the grape must. Based on the phenotypic analysis, a consortium was selected from the point of micro-sparkling.
The quantitative parameters of the individual taxa of the 2018 Wine Microorganism Consortium are given in (Table  2). The individual values in (Tables 1 and Table 2) were averaged from three measurements.
Through the application of microbiological techniques, the 2018 Wine Microorganism Consortium was obtained from the spontaneous batch, which was characterized in qualitative and quantitative terms.

Growth curves
The growth curve courses are shown in (Figures 1 and  Figure 2), while the numerical parameters of the growth characteristics are shown in (Tables 3 and Tables 4). The growth characteristics of individual isolates were determined based on the growth curves of different types of media. These characteristics were used to assign culture media and culture times to individual isolates (Table 3).

Obtaining pure cultures: Separation and lyophilisation
To culture and lyophilise the 2018 Wine Microorganism Consortium, 30 g of the consortium was prepared and used for fermentation for 100 litres of must. The representation of the individual isolates can be seen in Table 4.

Taxonomic identification LAB01
Lactobacillus brevis LAB02 Lactobacillus plantarum Y01 Hanseniaspora gulliermondi Y02 Saccharomyces cerevisiae 1 Y03 Hanseniaspora uvarum 1 Y04 Hanseniaspora uvarum 2 Y05 Saccharomyces cerevisiae 2 Y06 Zygosaccharomyces bailii Y07 Candida sake    Volume 14 The growth curves, fermentation tests, and H 2 S production results were used to inform the composition of the consortium so that the entire mixture exhibited the best fermentation capabilities, including the desired strains of lactic acid bacteria. Isolates Y03, Y06, and Y07 had the largest mass representation as characterized by a looser fermentation process but had the ability to ferment alcohol to 6.98, 9.37, and 7.20% vol. alc., respectively. These isolates showed low, medium, and high H 2 S-producing ability. Isolates Y02 and Y05 had the second largest mass representation of the resulting consortium as characterized by high sugar fermentability up to an alcohol content of 13.28 and 13.30% vol. alc., respectively. The fermentation process was not gradual, however, which could negatively affect the quality of the wine. Both isolates showed mean H 2 S production capacity. The isolate Y04 had a low mass representation. It showed the lowest sugar fermentability to an alcohol content of only 2.98 and had a medium ability to produce H 2 S. Table 5 contains the identification of individual isolates. The consortium consists of 2 strains of lactic acid bacteria and 7 strains of saccharomyces and non-saccharomyces yeast. The most abundant yeast strains of Consortium are Hanseniaspora uvarum, Zygosaccharomyces bailii and Candida sake.

Fermentation tests
Based on the results of the fermentation tests ( Figure 3 and Table 6), we can conclude that there was some similarity between isolates from the 2018 Wine Microorganism Consortium. For example, between isolates Y02 and Y05 and isolates Y01 and Y07. For the Y03 and Y06 strains, there was always a certain deviation that distinguished them from the other isolates, and isolate Y04 showed a completely different course of fermentation compared to the other yeast microorganisms tested. Based on these data, we were also able to divide the strains into high fermentation strength strains (Y02 and Y05) and medium fermentation strength strains (Y01, Y03, Y06, and Y07). Figure 3 shows the different fermentation progress of individual isolates. The consortium mixture was composed of isolates that showed different fermentation ability. The selection also included the incorporation of saccharomyces and non-saccharomyces yeast, which contribute to the sensory expression of the wine. Isolates Y03, Y06, and Y07 are characterized by a slower fermentation process. Isolates Y02 and Y05 were characterized by high sugar fermentability up to an alcohol content of 13.28 and 13.30% vol. alc., i.e. with a faster fermentation process. Y04 isolate showed the lowest sugar fermentability to an alcohol content of only 2.98, but the fermentation process was the most gradual.

H 2 S production
Low, medium, and high sulfane productivity occurred during the monitoring of H 2 S production by yeast isolates (Figure 4, Table 7). Most isolates (Y01, Y02, Y03, Y04, Y05) had medium sulfane productivity. H 2 S production was also monitored during the testing of a suitable fermentation strain, and strains with low, medium and high H 2 S production were found. The Y07 strain showed low production; the Y01, Y02, Y03, Y04, and Y05 strains showed medium production; and the Y06 strain showed high production. Y02 and Y05 also showed high fermentation strength and medium H 2 S production, while the Y07 strain produced a low amount of H 2 S and had a medium fermentation capacity.

Discussion
This study focused on the isolation of yeasts and lactic acid bacteria representing the given vineyard. From the isolated microorganisms and growth curves, the most suitable culture media and required culture times were determined. The isolation was realized from the point of micro-sparkling of spontaneous fermentation due to the strain diversity. The study found that at an ethanol concentration stage of 4.5 -5%, naturally, present nonsaccharomyces yeasts die and ethanol-tolerant S. cerevisiae begins to act. Previous studies have also reported that many different strains occur at the beginning of fermentation, but only a few dominate in the later stages of wine fermentation (Torija et al., 2001). Subsequently, fermentation tests were carried out to monitor the fermentation process and the fermentability of sugars to ethanol in individual isolates. Differences were found between individual isolates due to the isolation of microorganisms from the point of micro-sparkling when yeast diversity was greatest. During the fermentation tests, the differences in fermentability were shown, allowing us to select the yeast strain most suitable for fermentation. Some similarities were found in the sugar fermentability of different isolates, specifically isolates Y02 and Y05 and Y01 and Y07. In the strains Y03 and Y06, there was always a certain deviation that distinguished them from the second group, and isolate Y04 showed a completely different course of fermentation compared to the other yeast microorganisms tested. Based on these data, we were also able to divide the strains according to strains with high fermentation strength (Y02 and Y05) and those with medium fermentation strength (Y01, Y03, Y06).
During spontaneous fermentation, different yeast species and strains interact with each other differently depending on the changing conditions of the fermenting must The study by Ganucci et al. (2018) further found that independent of the grape variety, five of the six wineries in the study only had one predominant S. cerevisiae strain with an isolation frequency ranging from 32 to 74%, while the variable number of strains (from four to 14) was characterized by an isolation frequency of less than 10%. This finding is consistent with those reported by other authors (Versavaud et al., 1995; Gutiérrez et al., 1997;Egli et al., 1998;Sabate et al., 1998), although in some cases the predominant strains of S. cerevisiae were not found by the fermentation process (Vezinhet et al., 1992).
To select the most suitable yeast strain for fermentation, H 2 S production was also monitored on a special nutrient medium: BIGGY. The obtained isolates showed varying degrees of H 2 S production, ranging from low, medium to high H 2 S production. These results confirm those presented by Perrone et al. (2013) and Pérez-Torrado et al. (2017), which state, inter alia, that the dominant behavior of yeast strains is due to differential H 2 S production and killer factor resistance. It is noteworthy that in the high-frequency strains that were tested by Ganucci, Guerrini et al. (2018), no killer factor was detected and no significant differences in H 2 S production were found. The degree of competition of each strain, which determines the ability of one strain to compete with another, is influenced by other factors, however, including pH, temperature, ethanol, osmotic pressure, and available nitrogen (Ciani et al., 2016). Ganucci et al. (2018) study looking at the effect of ethanol and temperature on growth performance and condition advantage of high-frequency S. cerevisiae strains showed that these two factors can play an important role in determining the dominance of one strain over another during wine fermentation. A single action of ethanol on the growth performance led to the high-frequency strains showing significantly lower inhibition than the lowfrequency strains.
According to Arroyo-López, Querol and Barrio (2009), an even more accurate indicator of total yeast growth is the percentage of inhibition as this parameter is indirectly related to the delayed phase but linearly related to both the maximum specific growth rate (µmax) and the maximum cell density at the end of growth. Consequently, there is an advantage of condition, which according to Salvadó et al. (2011), represents the difference in µmax between competitors for specific environmental conditions. This leads to higher concentrations of highfrequency strains, indicating their enhanced adaptability to increasing ethanol concentrations compared to lowfrequency strains. Each S. cerevisiae strain can exhibit different stress reactions to ethanol because the effects of increasing ethanol concentrations on the yeast cell include various changes, such as membrane composition and gene expression, synthesis of heat shock proteins, increase in chaperone proteins, etc. (Ding et al., 2009).
Another study of four commercial wine yeast strains recently highlighted that fermentation temperature may be an important factor in determining the dynamics of a population of S. cerevisiae strains (García-Ríos et al., 2014). Ethanol and high temperature synergistically affect membrane integrity and permeability, causing a decrease in yeast population growth (Alexandre, Rousseaux and Charpentier, 1994; Albergaria and Arneborg, 2016).

CONCLUSION
The outcome of this study was the 2018 Wine Microorganism Consortium, which was obtained from the spontaneous fermentation that characterizes the given vineyard and supports the 'terroir' of the wine. At the same time, inoculation with this mixed culture helps to prevent problems with stagnant fermentation, which is often associated with spontaneous fermentation. Isolates of lactic acid bacteria and non-saccharomyces and saccharomyces yeasts were obtained and the resulting consortium was formed from these isolates. The obtained consortium was then used for fermentation tests where the percentage of fermented sugar and hydrogen sulphide production were monitored.
The Y03, Y06, and Y07 isolates had the largest mass representation and were characterized by a looser fermentation, but with the ability to ferment alcohol to 6.98, 9.37, and 7.20 vol. % alc. These isolates showed low, medium, and high ability to produce H2S. Y02 and Y05 isolates had the second largest mass representation of the resulting consortium and were characterized by high sugar fermentability up to an alcohol content of 13.28 and 13.30% vol. alc., but the fermentation process was not gradual, which could negatively affect the quality of the wine. Both isolates showed medium ability to produce H2S. Y04 isolate also has a low mass representation and showed the lowest ability to ferment sugar to an alcohol content of only 2.98 and a medium ability to produce H2S.