**2. Pulsed-field gel electrophoresis (PFGE) for the study of yeast population**

PFGE as a system encompasses a series of techniques in which the intact chromosomes of microorganisms like yeasts and filamentous fungi are submitted to the action of a pulsing electric field in two orientations that is changing direction, in a matrix of agarose. The bestknown PFGE modality is the CHEF system (Contour-clamped Homogeneous Electric Fields); this consists of a hexagon of 24 electrodes surrounding the gel that produce a homogeneous electric field alternating between two directions orientated at 120º with respect to each other. Using this system, Chu et al. (1986) resolved the electrophoretic karyotype of *Saccharomyces cerevisiae* in 15 bands in a size range of 200-2200 kb. Before performing the electrophoresis the yeast cells must be suitably treated, avoiding the direct

2009). In the industry, knowledge of specific strains of these microorganism species is important for (i) their selection; (ii) their use as starter cultures; and (iii) improving the

During the 1990's the development of molecular techniques has enabled the identification and characterization of different strains belonging to the same species of yeast, and it has been possible to establish the ecology of spontaneous fermentations in many of the world's winemaking regions (Fleet, 2008). These techniques also constitute a powerful tool not only for the selection of the most suitable yeast, since they tell us which yeasts are the most representative in the fermentation process, but also for obtaining information on the addition to the must of particular strains of yeast in the case of inoculated fermentations

Two of the approaches most often used for the molecular characterization of industrial yeast are analysis of the electrophoretic karyotypes by pulsed-field gel electrophoresis (PFGE) and analysis of the restriction fragment length polymorphism of the mitochondrial DNA (mtDNA-RFLP). We have used PFGE in winemaking to analyse the diversity of wild yeasts in spontaneous fermentation of a white wine produced in a winery in SW Spain with the object of selecting the most suitable autochthonous starter yeast; and from the results of the inoculation, we were able to make decisions for improving the efficiency of the process and to establish procedures for the proper performance of the inoculation (Rodríguez et al., 2010). We have also applied the analysis of the karyotypes to characterize natural yeasts in biodynamic red wines in another region of Spain. In this chapter we also evaluate the use of the mtDNA-RFLP technique for quick monitoring of the dominance of inoculated strains in industrial fermentation, without any need for the prior isolation of yeast colonies

Another electrophoretic technique has been used to show substantial changes in protein levels in selected wine yeasts under specific growth conditions. It has recently been stated that the proteome is "the relevant level of analysis to understand the adaptations of wine yeasts for fermentation" (Rossignol et al., 2009). Following this, in-depth studies are now being made of the proteome of wine yeast strains and the relationship between the proteome and wine quality and winery processes. We are now exploring more generally the relevance of proteomics to wine improvement. In this chapter, we will summarize the efforts being made by the proteomics research community to obtain the knowledge needed on

**2. Pulsed-field gel electrophoresis (PFGE) for the study of yeast population**  PFGE as a system encompasses a series of techniques in which the intact chromosomes of microorganisms like yeasts and filamentous fungi are submitted to the action of a pulsing electric field in two orientations that is changing direction, in a matrix of agarose. The bestknown PFGE modality is the CHEF system (Contour-clamped Homogeneous Electric Fields); this consists of a hexagon of 24 electrodes surrounding the gel that produce a homogeneous electric field alternating between two directions orientated at 120º with respect to each other. Using this system, Chu et al. (1986) resolved the electrophoretic karyotype of *Saccharomyces cerevisiae* in 15 bands in a size range of 200-2200 kb. Before performing the electrophoresis the yeast cells must be suitably treated, avoiding the direct

fermentation process.

(Rodríguez et al., 2010).

(Rodríguez et al., 2011).

proteins in the post-genomics era

manipulation of the genetic material to prevent possible rupture of the chromosomes. The cells are then embedded in blocks of agarose which are subsequently treated with a reducing agent and K proteinase to destabilize the wall and cytoplasmatic membranes, respectively (Figure 1), thus facilitating the release of the DNA when submitted to the action of an electric field.

This methodology for correctly obtaining the karyotype of *S. cerevisiae* is based on the procedure described by Carle & Olson (1985) and optimized by Rodríguez et al. (2010). It also depends on the concentration of the agarose gel (1%), buffer (0.5 x TBE), initial and final switch (60-120 seconds respectively), run time (24 hours), voltage (6 V/cm) and buffer temperature (14 ºC).

The analytical results given by this technique are the number and size of the yeast chromosomes, and it allows specific strains of *Saccharomyces* to be differentiated because their karyotypes show distinct bands running below the 500-kb marker. It also allows the differentiation between *S. cerevisiae* and *Saccharomyces bayanus* var. *uvarum* (*S. uvarum*) species (Naumov et al., 2000, 2002).

Fig. 1. Methodology for characterizing yeast strains, using pulsed field gel electrophoresis to obtain the karyotype.

In previous research the PFGE technique has been used to analyse the dynamics of the yeast population during the spontaneous fermentations of wine (Demuyter et al., 2004; Martínez et al., 2004; Naumov et al., 2002; Raspor et al., 2002; Rodríguez et al., 2010), and it has also been used to characterize other industrial yeasts including baker's and brewer's yeast (Codón et al., 1998). Another relevant application of PFGE has been to characterize the yeast population which is present in the *flor velum* that grows on the surface of fino-type sherry wines in the barrel, during their biological ageing process (Mesa et al., 1999, 2000). The results revealed an interesting correlation between the yeast genotypes and the different blending stages.

One disadvantage of the technique is that it is laborious, expensive and requires specialized personnel; increasingly, therefore, analysts are resorting to other simpler and faster techniques to discriminate between yeast clones, like, for example, interdelta analysis of sequences or microsatellite analysis (Cordero-Bueso et al., 2011; Le Jeune et al., 2007; Schuller et al., 2007). However, the methodology proposed in Figure 1 enables a large

Application of Gel Electrophoresis Techniques to

only the strain with pattern V has been used.

vintage campaign.

fermentation had occurred.

the Study of Wine Yeast and to Improve Winemaking 7

In our work on the analysis of the karyotype, it has been possible to monitor the yeast population under industrial conditions for several years when fermentations of the white wine were inoculated with selected autochthonous yeast strains. This has allowed inoculation strategies to be designed for the correct development of inoculated yeast while retaining the unique regional character of the finished wine (Rodríguez et al., 2010). The strains with patterns I, II, III and V were the most representative during the spontaneous fermentation process (Figure 2) and they could be isolated at the late fermentation phase. These autochthonous yeasts show valuable traits of enological interest, such as high fermentative capacity, ethanol tolerance, and they had a killer phenotype. The capacity of each strain to compete within the mixed population was also tested under semi-industrial conditions by PFGE. The results show that the strains with karyotypes II and V were the most vigorous competitors, followed by the strain with patterns III and I, which were detected in lower proportions (Rodríguez et al., 2010). Therefore, the strains with patterns II, III and V were used to inoculate the fermentations in the year 2001; strains with patterns II and V were used in 2002, 2003 and 2004; and from the year 2005 until the present (2011),

The inoculation of industrial vessels of the winery of this study presented several peculiarities. For example: (i) in each vintage year several vessels with a total capacity of 400 000-l were inoculated; (ii) the inoculums comprising the selected autochthonous yeast strains were prepared from fresh YEPD plates (1% yeast extract, 2% glucose, 2% peptone and 2% agar) by preparing a starter in which each scaling-up round was performed when the ºbé reached a value between 1-2, and in each round, the fermentation volume was increased tenfold to give high initial levels of inoculum (> 60 x 106 viable cells/ml) and ensure the correct development of the inoculated strain; (iii) once the starter cultures were scaled-up and added to a 400 000-l container, partial volumes were withdrawn and used for the inoculation of other 400 000-l vessels of the winery; and (iv) the 400 000-l vessels received random additions of fresh must until reaching the final volume. The frequency and timing of these additions depended on the production yield of fresh must during the

In our results (Table 1) only the strain with pattern V, called P5, was dominant under industrial conditions and, for this reason, the number of the starter strains were reduced over the years and currently only the strain with karyotype V is used for the inoculation of the industrial fermentations. In spite of this strain's good capacity for achieving dominance,, in some years a high degree of polymorphism was detected in the fermentations; and the cause of the unexpected predominance of wild yeast karyotype was linked to several factors, including: a sudden decrease in the temperature of one of the vessels during the scaling-up process of the inoculation in 2002; the method of inoculation and the scaling-up process, which were changed in 2003, whereby the inoculums of pure culture did not represent 10% and each scaling-up round was performed when the inoculums had a high sugar content (around 3-5 ºBé); and the storage of must in the vessels in which spontaneous

When karyotype V was dominant in the fermentations, the wine obtained had fruity characteristics, with well-balanced acidity, that satisfied the wine producer. Although we did not obtain a comprehensive aromatic characterization of the wine, the panel of wine tasters (Figure 3) considered the wine produced in these fermentations better than the wine

number of yeast isolates to be processed, and PFGE is considered a most suitable technique for discriminating between yeast clones (Schuller et al., 2004).

In our laboratory, this technique has been used to characterize the wine yeast population responsible for the spontaneous fermentation of a white wine produced in a winery in SW Spain (Rodríguez et al., 2010). Analyses of industrial-scale fermentations (in 400 000-l fermentation vessels) were carried out during two consecutive vintages. In 1999 and 2000 a total of 211 and 228 yeast colonies, respectively, from different vessels, were characterised by karyotyping. The degree of polymorphism observed was high, and 17 different karyotypic patterns were detected in 1999, and 21 patterns in 2000. In the two campaigns, we also found patterns belonging to non-*Saccharomyces* yeasts, the karyotypes of which did not show the four bands running below 500-kb. During the fermentation, this population was displaced by *S. cerevisiae* strains; patterns I, II, III and V were predominant during entire fermentation process in 1999, whereas in 2000 patterns II and V were predominant (see Figure 2). Those were the yeast strains selected for inoculating the industrial fermentations, as will be explained bellow.

Fig. 2. Frequencies of the majority karyotype patterns (%) obtained in the spontaneous fermentations of 1999 and 2000. NS corresponds to non-*Saccharomyces* yeasts.

The results of the characterization of the yeasts also showed that the different strains changed their proportion, and there was a sequential substitution of strains during the fermentation; this gave a valuable indication of the dynamics of the yeasts population throughout the process. Some of these changes were specific to a particular fermentation phase, suggesting that the yeast strains with different electrophoretic karyotypes also differ in their adaptation to the evolving environment at different phases of the fermentation process.

Although the diversity of wild yeast can contribute to high-quality and unique flavour in the finished wine, spontaneous fermentation is often unpredictable and might introduce less desirable traits to the product, sometimes even spoiling a production batch. Other risks associated with spontaneous fermentation include either slow or arrested fermentation. To avoid these problems, winemakers often add cultures of selected yeasts, in the form of active dried yeasts or autochthonous yeasts. Nevertheless, in some cases, these yeasts used as starters are not able to displace the wild yeasts present in the must, since the wild yeasts can be very competitive (Esteve-Zarzoso et al., 2000; Lopes et al., 2007).

number of yeast isolates to be processed, and PFGE is considered a most suitable technique

In our laboratory, this technique has been used to characterize the wine yeast population responsible for the spontaneous fermentation of a white wine produced in a winery in SW Spain (Rodríguez et al., 2010). Analyses of industrial-scale fermentations (in 400 000-l fermentation vessels) were carried out during two consecutive vintages. In 1999 and 2000 a total of 211 and 228 yeast colonies, respectively, from different vessels, were characterised by karyotyping. The degree of polymorphism observed was high, and 17 different karyotypic patterns were detected in 1999, and 21 patterns in 2000. In the two campaigns, we also found patterns belonging to non-*Saccharomyces* yeasts, the karyotypes of which did not show the four bands running below 500-kb. During the fermentation, this population was displaced by *S. cerevisiae* strains; patterns I, II, III and V were predominant during entire fermentation process in 1999, whereas in 2000 patterns II and V were predominant (see Figure 2). Those were the yeast strains selected for inoculating the industrial fermentations,

for discriminating between yeast clones (Schuller et al., 2004).

NS

% of the patterns

I

fermentations of 1999 and 2000. NS corresponds to non-*Saccharomyces* yeasts.

can be very competitive (Esteve-Zarzoso et al., 2000; Lopes et al., 2007).

II

Fig. 2. Frequencies of the majority karyotype patterns (%) obtained in the spontaneous

III

The results of the characterization of the yeasts also showed that the different strains changed their proportion, and there was a sequential substitution of strains during the fermentation; this gave a valuable indication of the dynamics of the yeasts population throughout the process. Some of these changes were specific to a particular fermentation phase, suggesting that the yeast strains with different electrophoretic karyotypes also differ in their adaptation to the evolving environment at different phases of the fermentation

Although the diversity of wild yeast can contribute to high-quality and unique flavour in the finished wine, spontaneous fermentation is often unpredictable and might introduce less desirable traits to the product, sometimes even spoiling a production batch. Other risks associated with spontaneous fermentation include either slow or arrested fermentation. To avoid these problems, winemakers often add cultures of selected yeasts, in the form of active dried yeasts or autochthonous yeasts. Nevertheless, in some cases, these yeasts used as starters are not able to displace the wild yeasts present in the must, since the wild yeasts

V

Others

Vintage 1999

Vintage 2000

as will be explained bellow.

process.

In our work on the analysis of the karyotype, it has been possible to monitor the yeast population under industrial conditions for several years when fermentations of the white wine were inoculated with selected autochthonous yeast strains. This has allowed inoculation strategies to be designed for the correct development of inoculated yeast while retaining the unique regional character of the finished wine (Rodríguez et al., 2010). The strains with patterns I, II, III and V were the most representative during the spontaneous fermentation process (Figure 2) and they could be isolated at the late fermentation phase. These autochthonous yeasts show valuable traits of enological interest, such as high fermentative capacity, ethanol tolerance, and they had a killer phenotype. The capacity of each strain to compete within the mixed population was also tested under semi-industrial conditions by PFGE. The results show that the strains with karyotypes II and V were the most vigorous competitors, followed by the strain with patterns III and I, which were detected in lower proportions (Rodríguez et al., 2010). Therefore, the strains with patterns II, III and V were used to inoculate the fermentations in the year 2001; strains with patterns II and V were used in 2002, 2003 and 2004; and from the year 2005 until the present (2011), only the strain with pattern V has been used.

The inoculation of industrial vessels of the winery of this study presented several peculiarities. For example: (i) in each vintage year several vessels with a total capacity of 400 000-l were inoculated; (ii) the inoculums comprising the selected autochthonous yeast strains were prepared from fresh YEPD plates (1% yeast extract, 2% glucose, 2% peptone and 2% agar) by preparing a starter in which each scaling-up round was performed when the ºbé reached a value between 1-2, and in each round, the fermentation volume was increased tenfold to give high initial levels of inoculum (> 60 x 106 viable cells/ml) and ensure the correct development of the inoculated strain; (iii) once the starter cultures were scaled-up and added to a 400 000-l container, partial volumes were withdrawn and used for the inoculation of other 400 000-l vessels of the winery; and (iv) the 400 000-l vessels received random additions of fresh must until reaching the final volume. The frequency and timing of these additions depended on the production yield of fresh must during the vintage campaign.

In our results (Table 1) only the strain with pattern V, called P5, was dominant under industrial conditions and, for this reason, the number of the starter strains were reduced over the years and currently only the strain with karyotype V is used for the inoculation of the industrial fermentations. In spite of this strain's good capacity for achieving dominance,, in some years a high degree of polymorphism was detected in the fermentations; and the cause of the unexpected predominance of wild yeast karyotype was linked to several factors, including: a sudden decrease in the temperature of one of the vessels during the scaling-up process of the inoculation in 2002; the method of inoculation and the scaling-up process, which were changed in 2003, whereby the inoculums of pure culture did not represent 10% and each scaling-up round was performed when the inoculums had a high sugar content (around 3-5 ºBé); and the storage of must in the vessels in which spontaneous fermentation had occurred.

When karyotype V was dominant in the fermentations, the wine obtained had fruity characteristics, with well-balanced acidity, that satisfied the wine producer. Although we did not obtain a comprehensive aromatic characterization of the wine, the panel of wine tasters (Figure 3) considered the wine produced in these fermentations better than the wine

Application of Gel Electrophoresis Techniques to

temperature of the fermentation was kept at 17 ºC.

karyotypes cIII, cVI, cXI and cXII.

pattern V.

the Study of Wine Yeast and to Improve Winemaking 9

By using PFGE to study the yeast population of the inoculated fermentations, the producer was able to make informed decisions for improving the process; the common factors in the vintages of 2001, 2004, 2005, 2007 and 2008, in which the inoculated strain was dominant, can be highlighted. These factors were the following: (i) the culture was not scaled-up to the next volume until the yeast had fully depleted the sugar to less than 1 ºBé (one degree is equivalent to 18 g/l of fermentable sugars in the must); therefore all cultures reached a high alcohol content before the addition of fresh must; (ii) the inoculum was always diluted less than 10-fold in each scaling-up round; (iii) the

We think that these criteria favoured the adaptation of the inoculums to the conditions of the must obtained in each vintage and to the final conditions within the 400 000-l industrial vessels. In addition, these criteria favoured the predominance of the inoculated strain with

In another study with biodynamic red wines, carried out in the Ribera del Duero D. O. Region (Valladolid, Spain), spontaneous fermentations were also analysed applying PFGE. We studied seven fermentations in three phases during the fermentation process: initial (IF), middle (MF) and final (EF), and 20 isolated strains per sample were characterized by applying PFGE (417 strains in 2008, and 412 strains in 2009). The results for two consecutive vintages studied showed the presence of different types of the yeast during the fermentations that were grouped in three populations. The first population was formed by non-*Saccharomyces* yeast, whose strains showed patterns with the absence of bands running below the region of 500 kb, which are specific to *S. cerevisiae* strains as reported above. The second population comprised *Saccharomyces bayanus* var. *uvarum* (*S. uvarum*); and the third population included *Saccharomyces cerevisiae* yeast. The strains of *S. uvarum* were differentiated from the *S. cerevisiae* strains by the presence of two small chromosomes in the region of 245-370 kb, instead of three as for S*. Cerevisiae*, as reported by Naumov et al. (2000, 2002). Non-*Saccharomyces* (NS) yeasts were dominant in the initial phase of fermentation but were displaced in the subsequent and final phases of the process by another population of yeasts. *S. uvarum* yeasts were present mainly in the phase mid-way through the fermentation; then the population of *S. cerevisiae* yeasts displaced the NS and *S. uvarum* yeasts, and remained dominant until the end of the fermentation, in the majority of the deposits analysed. The low frequency of detection of *S. uvarum* at the end of the fermentation could be indicative of its lower ethanol tolerance compared to *S. cerevisiae*. Within each population yeast strains were also found with different karyotyping patterns, and the distribution (by %) of these varied in the seven deposits analysed during the two consecutive years studied. Thus, for *S. uvarum*, considerable variability of strains and a total of 12 different electrophoretic patterns were detected (Figure 4): uI-uVII for vintage 2008; and uI-uIII, uV, uVIII-uXII in 2009. The strains with patterns uI, uII, uIII and uV, followed by uIV (in 2008) and uIX (in 2009) were the most representative in two years studied. Within the population of the *S. cerevisiae* yeasts, the variability of the patterns was higher than in *S. uvarum* ; 29 (cI-cXXIX) and 27 (cI-cVII, cX-cXII, cXV, cXVII, cXIX, cXXII, cXXIV, cXXX-cXLI) electrophoretic karyotype patterns were detected for 2008 and 2009 respectively. The *S. cerevisiae* yeast strains most representative of the fermentation process in these years were those that showed the



Table 1. Frequencies (%) of the inoculated strains in the industrial fermentations for seven vintage years. This analysis was not performed in the vintage year of 2006. NS: non-*Saccharomyces.*


\*Strain group II was inoculated in the years 2001 to 2004; the strain with pattern III was inoculated only in 2001; and strain group V was inoculated in all vintages

Fig. 3. Composition of the wine yeast population in each vintage, and its relationship with the quality of the final product. For each year the size of the yeast cell shown is proportional to the contribution of each strain to the total wine yeast population of the winery. The x symbol indicates that the proportion of the strain(s) within the population was below 5%. Wine quality was evaluated by a panel of expert wine-testers from the winery, who graded the final product on a scale from 1 to 5 based on fruity wine with well-balanced acidity desired by the producer (5 indicates highest quality and 1 lowest quality). Predominance of the strain with pattern V corresponded to a better quality of the wine.

obtained either in the spontaneous fermentations of 1999 and 2000 or in the vintage years of 2002 and 2003, when the karyotype V (strain P5) was not detected in high proportion in the

Patterns 2001 2002 2003 2004 2005 2007 2008 II 7.6 9.2 3.6 0.4 0 10.7 0 III 8.5 0 0 0 0 0 0 V 83 0.6 19.3 99.6 79.8 50 100 (NS) 0.2 0 0.4 0 0.4 0 0 Others 0.7 90.2 76.7 0 19.8 39.3 0

> 235

Table 1. Frequencies (%) of the inoculated strains in the industrial fermentations for seven vintage years. This analysis was not performed in the vintage year of 2006. NS: non-

**X X**

**X X X**

\*Strain group II was inoculated in the years 2001 to 2004; the strain with pattern III was inoculated only

Fig. 3. Composition of the wine yeast population in each vintage, and its relationship with the quality of the final product. For each year the size of the yeast cell shown is proportional to the contribution of each strain to the total wine yeast population of the winery. The x symbol indicates that the proportion of the strain(s) within the population was below 5%. Wine quality was evaluated by a panel of expert wine-testers from the winery, who graded the final product on a scale from 1 to 5 based on fruity wine with well-balanced acidity desired by the producer (5 indicates highest quality and 1 lowest quality). Predominance of

**X X** 5

**X X X** 5

 278  142

**X**

 240

4

3

5

5

3

**wine quality**

yeast population.

Total yeast analysed 423

*Saccharomyces.*

**2001**

**2002**

**2003**

**2004**

**2005**

**2007**

**2008**

 174  223

**vintage II III V others Inoculated Yeast Strains\***

**X X**

**X X**

the strain with pattern V corresponded to a better quality of the wine.

in 2001; and strain group V was inoculated in all vintages

By using PFGE to study the yeast population of the inoculated fermentations, the producer was able to make informed decisions for improving the process; the common factors in the vintages of 2001, 2004, 2005, 2007 and 2008, in which the inoculated strain was dominant, can be highlighted. These factors were the following: (i) the culture was not scaled-up to the next volume until the yeast had fully depleted the sugar to less than 1 ºBé (one degree is equivalent to 18 g/l of fermentable sugars in the must); therefore all cultures reached a high alcohol content before the addition of fresh must; (ii) the inoculum was always diluted less than 10-fold in each scaling-up round; (iii) the temperature of the fermentation was kept at 17 ºC.

We think that these criteria favoured the adaptation of the inoculums to the conditions of the must obtained in each vintage and to the final conditions within the 400 000-l industrial vessels. In addition, these criteria favoured the predominance of the inoculated strain with pattern V.

In another study with biodynamic red wines, carried out in the Ribera del Duero D. O. Region (Valladolid, Spain), spontaneous fermentations were also analysed applying PFGE. We studied seven fermentations in three phases during the fermentation process: initial (IF), middle (MF) and final (EF), and 20 isolated strains per sample were characterized by applying PFGE (417 strains in 2008, and 412 strains in 2009). The results for two consecutive vintages studied showed the presence of different types of the yeast during the fermentations that were grouped in three populations. The first population was formed by non-*Saccharomyces* yeast, whose strains showed patterns with the absence of bands running below the region of 500 kb, which are specific to *S. cerevisiae* strains as reported above. The second population comprised *Saccharomyces bayanus* var. *uvarum* (*S. uvarum*); and the third population included *Saccharomyces cerevisiae* yeast. The strains of *S. uvarum* were differentiated from the *S. cerevisiae* strains by the presence of two small chromosomes in the region of 245-370 kb, instead of three as for S*. Cerevisiae*, as reported by Naumov et al. (2000, 2002). Non-*Saccharomyces* (NS) yeasts were dominant in the initial phase of fermentation but were displaced in the subsequent and final phases of the process by another population of yeasts. *S. uvarum* yeasts were present mainly in the phase mid-way through the fermentation; then the population of *S. cerevisiae* yeasts displaced the NS and *S. uvarum* yeasts, and remained dominant until the end of the fermentation, in the majority of the deposits analysed. The low frequency of detection of *S. uvarum* at the end of the fermentation could be indicative of its lower ethanol tolerance compared to *S. cerevisiae*. Within each population yeast strains were also found with different karyotyping patterns, and the distribution (by %) of these varied in the seven deposits analysed during the two consecutive years studied. Thus, for *S. uvarum*, considerable variability of strains and a total of 12 different electrophoretic patterns were detected (Figure 4): uI-uVII for vintage 2008; and uI-uIII, uV, uVIII-uXII in 2009. The strains with patterns uI, uII, uIII and uV, followed by uIV (in 2008) and uIX (in 2009) were the most representative in two years studied. Within the population of the *S. cerevisiae* yeasts, the variability of the patterns was higher than in *S. uvarum* ; 29 (cI-cXXIX) and 27 (cI-cVII, cX-cXII, cXV, cXVII, cXIX, cXXII, cXXIV, cXXX-cXLI) electrophoretic karyotype patterns were detected for 2008 and 2009 respectively. The *S. cerevisiae* yeast strains most representative of the fermentation process in these years were those that showed the karyotypes cIII, cVI, cXI and cXII.

Application of Gel Electrophoresis Techniques to

southern Spain (Rodriguez et al., 2011).

inoculated yeast strain can be assured.

DNA (Figure 5).

for use by industry.

**inoculated yeast strains in wine fermentations** 

the Study of Wine Yeast and to Improve Winemaking 11

Although PFGE has been reported to be the most efficient in discriminating between different strains of *S. cerevisiae*, the mtDNA-RFLP technique is frequently used to differentiate between yeast isolates of the same species (González et al., 2007) because it enables a larger number of strains to be analyzed in a shorter time; it is a fast, simple, reliable and economic method, which does not require sophisticated material or specialized personnel (Fernández-Espinar et al., 2006). For these reasons, it is a very suitable technique

Most of the mitochondrial DNA in yeasts does not code proteins, and contains a high proportion of AT bases. Analysts can take advantage of this characteristic to characterize yeasts; it involves measuring variation in sequences in the mtDNA affecting the restriction sites of several endonucleases. Endonucleases such as *Alu*I, *Hinf*I or *Rsa*I, recognise the very frequent restrictions in the chromosomal DNA but not in the mitochondrial DNA, leading to a total cleavage of the chromosomal DNA in small pieces. These pieces can be easily differentiated from the mitochondrial fragments, which appear as bands with an electrophoretic mobility corresponding to molecules greater than 2 kb, generating

In previous studies applying this technique it has been demonstrated that the population of a fermentation vessel is "taken over" by wild yeasts, which displace the inoculated yeast strain, reducing it to a minority presence (Esteve-Zarzoso et al., 2000; Lopes et al., 2007; Raspor et al., 2002). In our research, when we have analyzed the inoculated fermentation of white wine as described above, we have found several examples of real situations that led to a significant decrease in the proportion of the inoculated strain (pattern V) and, in consequence, the quality of the wine was reduced. In order to minimize the impact of unwanted yeasts, wineries need a simple method for rapid diagnosis of the degree of dominance of inoculated strains, a method that could be performed routinely during the fermentation process (Ambrona et al., 2006, López et al., 2003). With this object we have used RFLP analysis of mtDNA for the rapid monitoring of the dominance, or otherwise, of inoculated yeast strains in industrial fermentations of white and red wines in a winery in

We apply this technique directly to samples of fermenting wine without previously isolating yeast colonies. For white wine fermentations, a rapid assay is performed consisting of taking a sample of fermenting must, purifying the DNA from harvested cells, and obtaining the restriction patterns by digestion with endonuclease *Hinf*I. The same protocol is applied to red wine fermentation, but an overnight cultivation step is added before purification of the

The criterion for considering the result of the rapid test to be positive was obtaining restriction patterns of mtDNA that were identical for the total cells and the inoculated strain; when this is the case, the starter yeast can be taken as being dominant in the fermentation. The result was considered negative when additional bands, or absence of bands, were observed in the patterns; in this case neither the dominance, nor even the presence, of the

polymorphisms that allow the characterization between yeast strains.

**3. Application of mtDNA-RFLP as a rapid method for monitoring the** 

The yeast population dynamics presented in this biodynamic red wine were different from those observed in other studies of white wines in which *S. uvarum* was dominant during spontaneous alcoholic fermentation (Demuyter et al., 2004).

Although *S. uvarum* has been found in other producing regions of the world, such as Alsace (Demuyter et al., 2004), at the moment there are no studies about the population dynamics of *S. uvarum* in Ribera del Duero, Spain.

The use of the PFGE technique allows analysts to detect a high degree of polymorphism in the population of the yeast and to monitor the dynamics of yeast ecology during the fermentation; this is because it is able to show the occurrence of gross chromosomal rearrangements, which is the phenomenon that mainly accounts for the rapid evolution of yeast clones subjected to industrial conditions (Infante et al., 2003). The technique also shows the most representative yeast strains in the industrial winemaking process, which are partially but significantly responsible for the finished wine's quality. Knowledge of these main strains can be used as a criterion for making a first selection of the autochthonous yeast. Later, with these previously selected strains, PFGE can be applied to study the features of these strains that are of enological interest, as described in recent years, which fall into three main categories: (i) properties that affect the performance of the fermentation process; (ii) properties that determine the quality of the wine; and (iii) properties associated with the commercial production of yeast (reviewed in Fleet, 2008). The yeasts selected by these means can then be used for inoculation in the fermentation process, thus improving winemaking.

Fig. 4. Electrophoretic karyotype of 14 colonies isolated from the sample taken in 2009 from a vessel in the phase mid-way through the fermentation process. Colonies 3-14 correspond to different karyotype patterns (uI-uIX) of yeast strains found in the *S. uvarum* species. Isolates 1-2 correspond to non-*Saccharomyces* strains which show the same pattern (NSI), with absence of bands running below 1225 kb. The chromosomes of the *S. cerevisiae* YNN295 strain were used as reference (M).

The yeast population dynamics presented in this biodynamic red wine were different from those observed in other studies of white wines in which *S. uvarum* was dominant during

Although *S. uvarum* has been found in other producing regions of the world, such as Alsace (Demuyter et al., 2004), at the moment there are no studies about the population dynamics

The use of the PFGE technique allows analysts to detect a high degree of polymorphism in the population of the yeast and to monitor the dynamics of yeast ecology during the fermentation; this is because it is able to show the occurrence of gross chromosomal rearrangements, which is the phenomenon that mainly accounts for the rapid evolution of yeast clones subjected to industrial conditions (Infante et al., 2003). The technique also shows the most representative yeast strains in the industrial winemaking process, which are partially but significantly responsible for the finished wine's quality. Knowledge of these main strains can be used as a criterion for making a first selection of the autochthonous yeast. Later, with these previously selected strains, PFGE can be applied to study the features of these strains that are of enological interest, as described in recent years, which fall into three main categories: (i) properties that affect the performance of the fermentation process; (ii) properties that determine the quality of the wine; and (iii) properties associated with the commercial production of yeast (reviewed in Fleet, 2008). The yeasts selected by these means can then be used for inoculation in the fermentation process, thus improving

**M 1 2 3456 7 8 9 10 11 13 12 14**

Fig. 4. Electrophoretic karyotype of 14 colonies isolated from the sample taken in 2009 from a vessel in the phase mid-way through the fermentation process. Colonies 3-14 correspond to different karyotype patterns (uI-uIX) of yeast strains found in the *S. uvarum* species. Isolates 1-2 correspond to non-*Saccharomyces* strains which show the same pattern (NSI), with absence of bands running below 1225 kb. The chromosomes of the *S. cerevisiae* YNN295

**NSI NSI uIII uIX uI uI uIuIX uV uIII uII uVIII uI uI**

spontaneous alcoholic fermentation (Demuyter et al., 2004).

of *S. uvarum* in Ribera del Duero, Spain.

winemaking.

2200 1600

strain were used as reference (M).

680

825

1020 1225
