Saccharomyces bayanus var. uvarum and Saccharomyces cerevisiae succession during spontaneous fermentations of Recioto and Amarone wines

Annals of Microbiology, 53 (4), 411-425 (2003) Saccharomyces bayanus var. uvarum and Saccharomyces cerevisiae succession during spontaneous fermentations of Recioto and Amarone wines F. DELLAGLIO 1*, G. ZAPPAROLI 1, P. MALACRINÒ 1, 2, G. SUZZI 3, S. TORRIANI 1 1 Dipartimento Scientifico e Tecnologico, Università degli Studi di Verona, Strada Le Grazie, 15, 37134 Verona; 2 Laboratori Agroalimentari, Consorzio Zai, Verona, Via Sommacampagna 61, 37137 Verona, Italy; 3 Dipartimento di Scienze degli Alimenti, Università degli Studi di Teramo, 64023 Mosciano S. Angelo, Teramo, Italy Abstract - This study was undertaken to evaluate the biodiversity of the indigenous Saccharomyces sensu stricto population during traditional vinification processes of Recioto and Amarone wines using molecular typing techniques. In total 109 isolates, collected from eight wineries during spontaneous fermentations, were identified and characterised by conventional tests and by molecular methods, i.e. PCR fingerprinting using the primer (GTG) 5, mtdna restriction and karyotype analyses. Sixty per cent of the isolates were assigned to Saccharomyces bayanus var. uvarum and 40% to Saccharomyces cerevisiae. A succession between S. bayanus var. uvarum, which dominated the first fermentation, and S. cerevisiae, which appeared after the wine drawing-off operation, was observed during a traditional Amarone winemaking. An extensive polymorphism was found between the isolates, however a few specific genetic biotypes prevailed in the different wineries. This ecotaxonomic survey constitutes a basic step to safeguard and exploit the oenological potential of the yeast biodiversity in the Recioto and Amarone wine ecosystems. Such biodiversity could be further explored to correlate the genetic patterns of the isolates with oenologically useful characteristics, with the ultimate goal to carry out selection programmes of typical strains of the Valpolicella area. Key words: Saccharomyces bayanus; Saccharomyces cerevisiae; mtdna restriction analysis; karyotyping; Recioto and Amarone wines. INTRODUCTION Extensive ecological surveys of the natural variability of Saccharomyces populations within specific wine-producing areas have generally shown a wide polymorphism among species and strains (Pretorius et al., 1999). This geographical biodiversity has been and still is explored with the ultimate goal to select new indigenous isolates and provide appropriate starter cultures per wine type and per area of * Corresponding author. Phone: +39-0458027930. Fax: +39-0458027051. E-mail: franco.dellaglio@univr.it 411

production (Sipiczki et al., 2001). The yeast communities associated with wines produced in diverse regions differ considerably in relation to the climatic and geographic conditions of the area, the grape variety and the winery environment (Versavaud et al., 1995; Mortimer and Polsinelli, 1999). Therefore, to achieve a complete picture of the indigenous yeast populations that promote the diversity of style of specific wines, systematic ecological and taxonomic studies are required. A deeper basic knowledge of these populations can arise from the application of molecular fingerprinting techniques such as karyotype analysis by Pulsed-Field Gel Electrophoresis (PFGE), mitochondrial DNA (mtdna) restriction assay and Randomly Amplified Polymorphic DNA (RAPD) - PCR analysis (Vezinhet et al., 1992; Paffetti et al., 1995; Guillamon et al., 1996; Fernandez-Espinar et al., 2001; Sabate et al., 2002). Electrophoretic karyotyping has been one of the first techniques utilized to discriminate yeast species and is considered as a reference method with high taxonomic potential (Bidenne et al., 1992; Vaughan-Martini et al., 1993). Restriction analysis of mtdna was shown to be invaluable for investigating the diversity of the genus Saccharomyces, because the mtdna of these yeasts is polymorphic and stable during vegetative multiplication (Pis kur et al., 1998; Gasent-Ramirez et al., 1999). Several studies have largely documented the usefulness of comparison of PFGE chromosomal patterns and of mtdna restriction profiles to identify Saccharomyces sensu stricto strains, to obtain an intraspecific differentiation, and to monitor the evolution of inoculated and native strains (Querol et al., 1992b; Guillamon et al., 1994; Schutz and Gafner, 1994; Pramateftaki et al., 2000; Naumov et al., 2002). In particular, these molecular techniques allow the distinction between the species S. bayanus and S. cerevisiae, that often co-operate to realize the spontaneous fermentations of wines produced at low temperature (Naumov et al., 2000; Torriani et al., 1999). Composite associations of S. bayanus and S. cerevisiae strains are also involved in the natural fermentations that occur during production of Recioto and Amarone wines (Usseglio Tomasset et al., 1980; Torriani et al., 1999). Such wines are produced in the Valpolicella area (Verona, Italy) from partially dried grapes following a similar winemaking process. For both wines, the must fermentation is performed at low temperature (3-5 C) by yeasts originating from grapes and winery equipment. The alcoholic fermentation is not completed with either wine stiles. The Recioto wine is characterised by a light sweet taste due to the presence of 20-50 g/l residual sugars. In contrast, Amarone wine undergoes a re-fermentation, during summer months of the following year, that reduces the sugar content to less than 1 g/l. These red wines are then aged for at least two years in oak barrels. This study was undertaken to evaluate the biodiversity of the indigenous Saccharomyces sensu stricto population that participate to the production of Recioto and Amarone wines. One hundred and nine isolates, obtained from eight wineries during spontaneous alcoholic fermentations, were identified and characterised by conventional tests and by a range of molecular techniques, i.e. PCR fingerprinting using the primer (GTG) 5, mtdna restriction and karyotype analyses. The succession of Saccharomyces sensu stricto species and strains during a traditional vinification of Amarone wine was investigated. 412 F. DELLAGLIO et al.

MATERIALS AND METHODS Wine samples and yeast isolation. The must and wine samples used in this study were provided by eight wineries located in the Valpolicella area, that produce Recioto and Amarone wines by traditional technological practices. The samples were collected from tanks and oak barrels during vinification processes from February to August 2001. In particular, two samples of Amarone wine were taken from each winery, i.e. one sample during first fermentation and the other one during refermentation, except the winery W4, from which only one sample was collected in winter. Moreover, a total of 20 samples were collected during the entire Amarone winemaking process conducted at the winery W2. The samples of Recioto wine from four different wineries were taken during alcoholic fermentation. Yeast isolation was performed on YEPD agar (1% yeast extract, 2% peptone, 2% glucose and 2% agar), after incubation at 30 C for 2-4 days. A representative number of yeast colonies was isolated at random from plates and further analysed. Yeast identification. The isolates were characterised following the conventional criteria for spore formation and physiological tests according to Kurtzman and Fell (1998). Their ability to ferment glucose, maltose, sucrose, galactose and melibiose as carbon sources, was determined in Durham tubes, containing YEPD with 2% of the appropriate sugars, after 10 days at 30 C. The capacity to grow at 6, 34 and 37 C was tested in YEPD medium. The identification of the isolates was established by PCR fingerprinting using the microsatellite oligonucleotide primer (GTG) 5 (Baleiras Couto et al., 1996). Extraction of DNA, amplification, electrophoresis and pattern recognition were performed as previously reported by Torriani et al. (1999). Four type strains of the Saccharomyces sensu stricto group, namely S. bayanus DBVPG 6171 T, S. cerevisiae DBVPG 6173 T, S. paradoxus DBVPG 6411 T and S. pastorianus DBVPG 6047 T, obtained from the Industrial Yeasts Collection of Perugia (Italy), were used for comparison. Yeast differentiation. mtdna restriction analysis. Total DNA from overnight YEPD cultures was extracted and purified by the method of Querol et al. (1992a), except that lytic enzyme (25 mg/ml) from Rhizoctonia solani (Sigma-Aldrich GmbH, Seelze, Germany) was used to digest the cell wall. DNA restriction was performed with HinfI and RsaI restriction endonucleases (Takara Shuzo Co., Otsu, Shiga, Japan) using the instructions provided by the manufacturer. Restriction fragments were separated by electrophoresis on 0.7% (w/v) agarose gels in 0.5 TAE buffer (45 mm Trisacetate, 1 mm EDTA, ph 8). Karyotype analysis. Chromosomal DNA was prepared from stationary phase cultures in low-melting-point agarose plugs as described by Schwartz and Cantor (1984). The chromosome-size DNA molecules were separated using a CHEF MAPPER XA Electrophoresis System (Bio-Rad Laboratories, Hercules, CA) in 1% agarose gel in 0.5 TBE buffer (45 mm Tris-borate, 1 mm EDTA, ph 8) at 200 V and 14 C, with the program: 40 s switch time for 5 h, 60 s switch time for 9 h, 90 s switch time for 11 h. Ann. Microbiol., 53 (4), 411-425 (2003) 413

Numerical analysis of bands profiles. Photographs of the mtdna and PFGE banding patterns were scanned using a ScanJet IIcx scanner (Hewlett-Packard Co., Palo Alto, CA). Conversion, normalization and further processing of the patterns were performed using the GelCompar software version 4.0 (Applied Maths, Kortrijk, Belgium). For each isolate, the mtdna restriction patterns obtained with HinfI and RsaI endonucleases and the karyotype pattern were combined into one single profile. The similarity between all pairs of combined banding profiles was expressed by a Dice coefficient correlation, and a dendrogram was obtained using the Unweight Pair Group Method using Arithmetic Average (UPGMA) clustering algorithm (Vauterin and Vauterin, 1992). RESULTS Identification and monitoring of indigenous yeasts in wine Atotal of 109 yeast strains was isolated from wine samples provided by the eight wineries in the Valpolicella area. The wineries were chosen because of the high quality of the Recioto and Amarone wines, which were produced following traditional procedures without the use of commercial yeast starter cultures. Identification of the isolates was achieved by conventional taxonomic methods, as well as by PCR fingerprinting analysis. According to standard assays based on morphological, physiological and biochemical tests, the isolates were assigned to the Saccharomyces sensu stricto complex. PCR amplification with the microsatellite oligonucleotide primer (GTG) 5 yielded species-specific products that allowed the visual distinction of two groups of strains corresponding to the S. bayanus and S. cerevisiae species, in compliance with the banding profiles of the type strains (data not shown). Table 1 shows the number of the identified isolates from each winery, their fermentation patterns and ability to grow at different temperatures. All of the 65 S. bayanus isolates grew at 6 C and exhibited an homogeneous sugar fermentation profile; 45% of the S. bayanus isolates did not grow at 34 C and none at 37 C. In regards to the S. cerevisiae, all of the 44 isolates grew at 34 and 37 C, but not at 6 C. Melibiose was not fermented by any isolate, while the capacity to utilize galactose was variable: only 34% of the isolates fermented this sugar. A different distribution of S. bayanus and S. cerevisiae species in the wineries was observed: both species were detected in six wineries, only S. bayanus in cellar W4 and only S. cerevisiae in cellar W6. These differences could be correlated to the sampling times and could reflect the evolution of these species during the production processes. Indeed, S. bayanus was mainly isolated from samples taken during alcoholic fermentation of Amarone and Recioto wines, while S. cerevisiae was principally found in samples of Amarone wine that underwent re-fermentation. To better investigate the diversity and population change that occur during the production of Amarone wine, a traditional winemaking process was examined in quantitative and qualitative detail. The growth of yeasts and the S. bayanus and S. cerevisiae species succession throughout the fermentation period in the winery W2 are depicted in Fig. 1. The majority of the S. bayanus strains (29 out of 31 isolates) were isolated during the alcoholic fermentation phase. On the contrary, S. cerevisiae dominated the re-fermentation process. However, S. cerevisiae occurred also in the other phases of the fermentations at a lower numerical level. 414 F. DELLAGLIO et al.

TABLE 1 Physiological characteristics of the Saccharomyces bayanus and Saccharomyces cerevisiae isolates from Recioto and Amarone wines produced in eight wineries of the Valpolicella area Winery Wine No. of identified Fermentation of a : Growth at: isolates GAL MEL 6 C 34 C 37 C W1 Amarone 6 S. bayanus + + + + 2 S. bayanus + + + 1 S. cerevisiae + + + 1 S. cerevisiae + + W2 Amarone 19 S. bayanus + + + + 12 S. bayanus + + + 3 S. cerevisiae + + + 23 S. cerevisiae + + W3 Recioto 1 S. bayanus + + + 2 S. cerevisiae + + + Amarone 3 S. cerevisiae + + + W4 Amarone 1 S. bayanus + + + + 2 S. bayanus + + + W5 Recioto 2 S. bayanus + + + 1 S. bayanus + + + + 1 S. cerevisiae + + + Amarone 3 S. bayanus + + + 1 S. cerevisiae + + + 1 S. cerevisiae + + W6 Recioto 2 S. cerevisiae + + + Amarone 3 S. cerevisiae + + + 1 S. cerevisiae + + W7 Recioto 2 S. bayanus + + + + 3 S. bayanus + + + Amarone 4 S. bayanus + + + + 2 S. bayanus + + + 1 S. cerevisiae + + W8 Amarone 3 S. bayanus + + + + 2 S. bayanus + + + 2 S. cerevisiae + + a : all strains ferment glucose, maltose and sucrose; GAL, galactose; MEL, melibiose. Yeast differentiation by molecular genetic methods The genetic diversity of 72 representative yeast isolates from the eight wineries was examined by two high-resolution molecular techniques, i.e. mtdna restriction and karyotype analyses. Such isolates were selected on the basis of origin and differences in phenotypic traits (Table 1), in order to reduce the probability to analyse duplicates of the same culture. Ann. Microbiol., 53 (4), 411-425 (2003) 415

Alcoholic fermentation beginning end Re-fermentation Log CFU/ml Sulfiting Drawing off Transfer in barriques Time (h) FIG. 1 Dynamics of Saccharomyces sensu stricto population during a production of Amarone wine at winery W2. The numbers in bold and in italic indicate the isolates assigned to the species S. bayanus and S. cerevisiae, respectively, by conventional taxonomic methods and PCR fingerprinting using primer (GTG) 5. The arrows indicate the main alcoholic fermentation phases. FIG. 2 Mitochondrial DNA restriction profiles obtained with HinfI and RsaI endonucleases and total DNA template from representative isolates of Saccharomyces bayanus (A, HinfI; B, RsaI) and Saccharomyces cerevisiae (C, HinfI; D, RsaI). M, molecular marker (Lambda/HindIII marker, Stratagene, Germany). 416 F. DELLAGLIO et al.

Restriction analysis of mtdna with HinfI and RsaI endonucleases revealed a high genetic polymorphism between the isolates, especially for S. cerevisiae species. In fact, of the 72 isolates examined, HinfI and RsaI generated 11 and 12 different banding profiles for S. cerevisiae, 7 and 5 for S. bayanus, respectively (Fig. 2). Hence, HinfI was more discriminative than RsaI for S. bayanus, while the opposite occurred for S. cerevisiae. In addition, the two species did not have common HinfI and RsaI restriction patterns and the differences between the band patterns of the two taxa were substantial. The distribution of the isolates from each winery, as grouped by their mtdna restriction patterns, is reported in Table 2. The profiles H2 and H8 obtained with HinfI for S. bayanus and S. cerevisiae isolates, respectively, and the profiles R1 and R6 obtained with RsaI for S. bayanus and S. cerevisiae isolates, respectively, were the most common in the different wineries. Several unique mtdna restriction profiles were observed for both S. bayanus and S. cerevisiae. The distribution of the mtdna restriction profiles in the different wineries often reflects the origin of the isolates. Indeed, most of the profiles were characteristic of a single winery: for example, profile H5 was recognised only in winery W5 and profile H9 only in winery W2. The individual HinfI and RsaI profiles obtained for each yeast were combined to achieve a more reliable differentiation of the isolates. Twelve and 11 combinations were recognised for S. cerevisiae and S. bayanus, respectively (Table 3). Several of the combined profiles were constituted by unique HinfI and RsaI restriction patterns. However, a particular restriction profile can share different patterns obtained with the other enzyme. For example, H1 profile shared R1, R2, R3 and R4 profiles or R1 profile was combined with H1, H2, H4, H5 and H7 profiles. Electrophoretic karyotyping was performed on 57 representative S. cerevisiae and S. bayanus isolates chosen on the basis of the results of mtdna restriction analysis. The 17 different PFGE profiles recognised (labelled from P1 to P17) are depicted in Fig. 3. The examined yeasts displayed similar karyotypes composed of 12-16 bands in the range of chromosome size 2200 to 200 kb, which are typical of the Saccharomyces sensu stricto group (Vaughan-Martini et al., 1993). The isolates of the two species differed in the number of bands in the region between 365 and 225 kb: all the S. bayanus isolates showed two bands in that region, while all the S. cerevisiae isolates had three bands. These findings allowed a clear distinction of the two species in agreement with previous observations (Giudici et al., 1998). Furthermore, all the Mel + cryotolerant isolates from Recioto and Amarone wines showed a karyotype profile typical for the taxon S. bayanus var. uvarum, recently established by genetic analysis (Naumov, 2000). It was noted that, for both species, some isolates with different combined mtdna profiles exhibited an identical karyotype (Table 3). In addition, for some S. bayanus isolates, different PFGE profiles shared the same combined mtdna profiles: for example, H2R1, H4R1 and H5R1 corresponded to two (P2 and P7), two (P5 and P8) and three (P6, P7 and P9) PFGE profiles, respectively. On the contrary, for the majority of the S. cerevisiae isolates each combined restriction profile corresponded to only one PFGE pattern except for isolates with combined patterns H9R13, H10R12, H11R10, H12R11 and H18R9 that were indistinguishable when analyzed by PFGE, as they showed a unique profile (P12). Therefore, the two fingerprinting techniques displayed different discriminative power and furnished complementary results useful for the differentiation of the isolates. Ann. Microbiol., 53 (4), 411-425 (2003) 417

TABLE 2 Distribution of the Saccharomyces bayanus and Saccharomyces cerevisiae isolates with different mtdna restriction profiles obtained with HinfI and RsaI endonucleases in relation to the winery of origin MtDNA profile No. of isolates Total a W b 1 W2 W3 W4 W5 W6 W7 W8 S. bayanus HinfI H1 4 1 1 6 (14) H2 8 4 3 2 3 1 21 (49) H3 2 2 (4) H4 1 1 7 1 10 (24) H5 1 1 (2) H6 1 1 (2) H7 1 1 2 (4) 43 (100) RsaI R1 8 5 1 3 5 9 3 34 (79) R2 4 4 (10) R3 1 1 (2) R4 1 2 3 (7) R5 1 1 (2) 43 (100) S. cerevisiae HinfI H8 1 1 1 1 4 (14) H9 6 6 (20) H10 1 1 2 (7) H11 3 3 (11) H12 2 2 (7) H13 1 1 (3) H14 2 2 (7) H15 1 1 (3) H16 1 2 3 (11) H17 4 4 (14) H18 1 1 (3) 29 (100) RsaI R6 1 1 1 1 4 (14) R7 4 4 (14) R8 1 1 (3) R9 1 1 (3) R10 3 3 (11) R11 2 2 (7) R12 1 1 2 (7) R13 6 6 (21) R14 2 2 (7) R15 1 1 (3) R16 1 1 2 (7) R17 1 1 (3) a : isolates with the same mtdna restriction pattern (%); b : winery. 29 (100) 418 F. DELLAGLIO et al.

TABLE 3 Comparison of combined HinfI and RsaI restriction profiles with karyotyping profiles obtained in selected Saccharomyces bayanus and Saccharomyces cerevisiae isolates from the eight wineries and frequencies of genetic profiles in the isolates from winery W2 Species Combined mtdna PFGE profiles Combined mtdna profiles (no. of isolates) and PFGE profiles (no. of isolates) (no. of isolates from winery W2) S. bayanus S. cerevisiae H1R1 (2) P4 (2) H1R2 (2) P1 (2) H1R2P1 (2) H1R3 (1) P9 (2) H1R4 (1) P1 (2) H1R4P1 (1) H2R1 (19) P2 (5), P7 (1) H2R1P2 (5) H2R4 (2) P3 (2) H3R2 (2) P1 (2) H3R2P1 (2) H4R1 (10) P5 (1), P8 (1) H4R1P8 (1) H5R1 (1) P6 (1), P7 (1), P9 (1) H6R5 (1) P4 (2) H7R1 (2) P4 (2) H8R6 (4) P11 (2) H8R6P11 (1) H9R13 (6) P12 (6) H9R13P12 (6) H10R12 (2) P12 (2) H11R10 (3) P12 (2) H12R11 (2) P12 (2) H13R8 (1) P10 (2) H14R14 (2) P15 (2) H15R15 (1) P17 (2) H16R16 (2) P13 (2) H16R17 (1) P16 (2) H17R7 (4) P14 (4) H17R7P14 (4) H18R9 (1) P12 (2) H18R9P12 (1) Comparison by numerical analysis of the combined electrophoretic patterns generated with both karyotyping and mtdna restriction techniques and observed in at least one isolate, resulted in the dendrogram shown in Fig. 4. The band at 2200 kb in PFGE profiles corresponding to chromosome XII for S. cerevisiae was excluded from the analysis because of its non-reproducible nature (Carle and Olson, 1985). Two well separated clusters with a similarity level of 30% corresponding to the S. bayanus and S. cerevisiae species were discerned. The considerable genetic polymorphism of the isolates, especially of those belonging to the S. cerevisiae species, was confirmed by the low similarity values obtained between the profiles comprised in each cluster. On the basis of the molecular clustering results, a reliable estimation of the distribution and frequencies of the individual isolates in all stages of the fermentations carried out in the winery W2 could be drawn. The number of the S. bayanus and S. cerevisiae isolates having the same combined patterns is presented in Table 3. The marked genetic diversity observed among the S. bayanus and S. cerevisiae isolates indicates that the fermentation processes were conducted by several different well- Ann. Microbiol., 53 (4), 411-425 (2003) 419

FIG. 3 Electrophoretic karyotypes of representative Saccharomyces bayanus (A) and Saccharomyces cerevisiae (B) isolates. M, DNA Size Markers-Yeast Chromosomal (Bio-Rad Laboratories). adapted strains. The S. bayanus isolates with the combined profile H2R1P2 predominated during the fermentation of must, while the S. cerevisiae isolates with profile H9R13P12 were found in larger numbers during re-fermentation. DISCUSSION In the present study, the biodiversity of the indigenous Saccharomyces sensu stricto population throughout the traditional vinification of Recioto and Amarone wines were revealed by using molecular typing techniques. Samples of must and wine collected during spontaneous fermentations in eight wineries located in the Valpolicella area were found to contain strains of the species S. bayanus and S. cerevisiae. This finding is in agreement with previous investigations concerning ecological surveys of such wines (Usseglio-Tomasset et al., 1980; Torriani et al., 1999). Phenotypic and molecular analyses showed that all of the S. bayanus isolates 420 F. DELLAGLIO et al.

% Similarity S. bayanus S. cerevisiae FIG. 4 UPGMA dendrogram derived from comparison of the combined electrophoretic patterns (karyotype and mtdna restriction with HinfI and RsaI endonucleases) obtained for Saccharomyces sensu stricto isolates. belong to the variety uvarum. The genetic heterogeneity of the S. bayanus species has been revealed by several investigations, which have lead to the division of such species into two natural subgroups, i.e. uvarum and bayanus (Nguyen and Gaillardin, 1997; Rainieri et al., 1999; Nguyen et al., 2000; Naumov et al., 2001). Recently, Naumov (2000) proposed the taxon S. bayanus var. uvarum, which comprises melibiose-fermenting strains with distinctive physiological and biochemical properties: cryotolerance, production of high quantities of glycerol, succinic acid and malic acid, low production of acetic acid and high sporulation ability. The ecological niche of S. bayanus var. uvarum is found in winemaking at low temperatures (Naumov et al., 2000). Mel + Saccharomyces strains have been associated with several wines produced at 5-15 C, such as Tokaj (Slovakia), Muscat (Ukraine) wines and some sweet type of French wines (Naumov et al., 2002). All these wines, as well as Recioto and Amarone wines, are produced from Botrytis cinerea contaminated grape musts, which were fermented at low temperature. Such conditions determine a selective pressure on natural Saccharomyces population that favour the cryophilic yeast S. bayanus var. uvarum. Ann. Microbiol., 53 (4), 411-425 (2003) 421

The dynamics of the Saccharomyces sensu stricto species was analyzed in depth throughout the fermentation period in a winery that produces Amarone wine of an excellent quality without the use of commercial yeast starter cultures. Saccharomyces bayanus var. uvarum dominated the first fermentation, but it was replaced by strains belonging to S. cerevisiae during summer when the cellar temperature increases. Hence, in Amarone wine, re-fermentation was conducted by more ethanol-tolerant strains of S. cerevisiae, which were detected during the first fermentation, albeit in low numbers, and succeeded cryophilic yeast completing the alcoholic fermentation. Such re-fermentation is detrimental for Recioto wine, which has to retain a certain amount of residual sugars, therefore it has to be avoided (Usseglio-Tomasset et al., 1980). To study the genetic diversity and the sequential growth of individual yeast strains during Recioto and Amarone winemaking, a detailed molecular analysis was conducted on representative isolates from each winery. Both mtdna restriction patterns and karyotype data indicated an extensive genetic variation between the isolates and proved that a number of strains are involved in the spontaneous fermentations of Recioto and Amarone wines. In previous researches, several authors have highlighted the coexistence of many strains during the course of natural fermentations and the variation of their relative proportions during the phases of fermentation (Polsinelli et al., 1996; Sabate et al., 1998; Ribereau-Gayon et al., 2000). Generally, it was shown that few strains dominated the fermentation process and the most frequent strains could be detected from the same location in consecutive vintages, suggesting the presence of a characteristic yeast population in each cellar (Versavaud et al., 1995; Sabate et al., 1998). The findings of the present study confirm the co-operation of various S. bayanus var. uvarum and S. cerevisiae strains in the spontaneous vinification processes of Recioto and Amarone wines and the prevalence of specific genetic biotypes in the different wineries. Results suggest that electrophoretic karyotyping is suitable for typing individual strains of Saccharomyces sensu stricto complex, which is consistent with several previous observations. Karyotype analysis by PFGE was performed to corroborate the genetic diversity of the indigenous yeast isolates as shown by mtdna restriction assay. Our data underlined that the results from the two genetic assays were not fully equivalent. Indeed, comparative analysis of genetic diversity of the yeast isolates showed a different discriminatory capacity of the two molecular techniques since strains with different mtdna restriction profiles showed identical PFGE chromosomal patterns and vice versa. This aspect was also noted by Nadal et al. (1996) who attributed the appearance of S. cerevisiae strains with the same mtdna patterns but different karyotype profiles mainly to a hypervariability of low-mobility bands and suggested that these strains originated from a pre-existent population of different, related yeast clones. More recently, Mesa et al. (2000), studying the yeasts involved in the ageing of Sherry wine belonging to the physiological races of S. cerevisiae found that several strains shared the same karyotype or mtdna pattern. However, a preferential association between karyotype and mtdna appeared. The yeast isolates showing different mtdna restriction patterns and identical karyotype, or the contrary, most likely originated from independent genetic lineages. Thus, the discrimination of each individual isolates was possible by combining the results obtained by the two typing techniques. Furthermore, a quantification of the genetic relatedness between the isolates could be carried out 422 F. DELLAGLIO et al.

by performing a clustering analysis on combined mtdna restriction and karyotyping patterns: a major diversity among the isolates of S. cerevisiae in respect to S. bayanus var. uvarum could be observed. Relationships between genetic diversity of yeast populations and ecological origin, which result from the presence of certain strains in a particular area (winemaking region, vineyard, winery, cellar, etc.), have been previously demonstrated (Vezinhet et al., 1992; Querol et al., 1994; Querol et al., 2003). It was supposed that the adaptation of specific strains to a microenvironment could have important implications not only from an ecological point of view but also for winemaking processes. In this study, the polyclonal character of the Saccharomyces sensu stricto population responsible of Recioto and Amarone wine fermentations was recognized and the prevalence of some strains established. The association of strain and winery could stimulate further research to define the influence of each strain on the quality of Recioto and Amarone wines produced by local wineries. Moreover, the correlation between a specific electrophoretic profile and a phenotypic character might facilitate the selection of wild yeasts with desired oenological traits. In conclusion, this study allowed, by means of molecular approaches, the recognition of genetically different strains of S. bayanus var. uvarum and S. cerevisiae associated with Recioto and Amarone wines and monitoring their succession during fermentative processes. The existence of such biodiversity in this peculiar ecosystem could be further explored to correlate the genetic patterns of the isolates with oenologically useful characteristics, with the ultimate goal to carry out selection programmes of these representative strains. Acknowledgements We would like to thank the oenologists of the wineries in the Valpolicella area for generously providing the wine samples for the analysis. REFERENCES Baleiras Couto M.M., Eijsma B., Hofstra H., Uis in t Velt J.H., van der Vossen J.M.B.M. (1996). Evaluation of molecular typing techniques to assign genetic diversity among Saccharomyces cerevisiae strains. Appl. Environ. Microbiol., 62: 41-46. Bidenne C., Blondin B., Dequin S., Vezinhet F. (1992). Analysis of the chromosomal DNA polymorphism of wine strains of Saccharomyces cerevisiae. Curr. Gen., 22: 1-7. Carle G.F., Olson M.V. (1985). An electrophoretic karyotype for yeast. Proc. Nat. Ac. Sci. USA, 82: 3756-3760. Fernandez-Espinar M.T., Lopez, V., Ramon D., Bartra E., Querol A. (2001). Study of the authenticity of commercial wine yeast strains by molecular techniques. Int. J. Food Microbiol., 70: 1-10. Gasent-Ramirez J.M., Castrejon F., Querol A., Ramon D., Benitez T. (1999). Genomic stability of Saccharomyces cerevisiae baker s yeasts. Syst. Appl. Microbiol., 22: 329-340. Giudici P., Caggia C., Pulvirenti A., Zambonelli C., Ranieri S. (1998). Electrophoretic profile of hybrids between cryotolerant and non-cryotolerant Saccharomyces strains. Lett. Appl. Microbiol., 27: 31-34. Guillamon J., Barrio E., Huerta T., Querol A. (1994). Rapid characterization of four species of the Saccharomyces sensu stricto complex according to mitochondrial DNA patterns. Int. J. Syst. Bacteriol., 44: 708-714. Ann. Microbiol., 53 (4), 411-425 (2003) 423

Guillamon J., Barrio E., Querol A. (1996). Characterization of wine yeast strains of the Saccharomyces genus on the basis of molecular markers. Relationship between genetic distance and geographic or ecological origin. Syst. Appl. Microbiol., 19: 122-132. Kurtzman C.P., Fell J.W. (1998). The Yeasts, A Taxonomic Study. Elsevier Science Publishers, Amsterdam. Mesa J.J., Infante J.J., Rebordinos L., Sanches J.A., Cantoral J.M. (2000). Influence of the yeast genotypes on enological characteristics of Sherry wines. Am. J. Enol. Vitic., 51: 15-21. Mortimer R., Polsinelli M. (1999). On the origins of wine yeast. Res. Microbiol., 150: 199-204. Nadal D., Colomer B., Pina B. (1996). Molecular polymorphism distribution in phenotypically distinct populations of wine yeast strains. Appl. Environ. Microbiol., 62: 1944-1950. Naumov G.I (2000). Saccharomyces bayanus var. uvarum comb. nov. a new variety established by genetic analysis. Mikrobiologiya, 69: 410-414. Naumov G.I., Masneuf I., Naumova E.S., Aigle M., Dubourdieu D. (2000). Association of Saccharomyces bayanus var. uvarum with some French wines: genetic analysis of yeast population. Res. Microbiol., 151: 683-691. Naumov G.I., Naumova E.S., Antunovics Z., Sipiczki M. (2002). Saccharomyces bayanus var. uvarum in Tokaj wine-making of Slovakia and Hungary. Appl. Microbiol. Biotechnol., 59: 727-730. Naumov G.I., Nguyen H.V., Naumova E.S., Michel A., Aigle M., Gaillardin C. (2001). Genetic identification of Saccharomyces bayanus var. uvarum, a cider-fermenting yeast. Int. J. Food Microbiol., 65: 163-171. Nguyen H.V., Gaillardin C. (1997). Two subgroup within the S. bayanus species evidenced by PCR amplification and restriction polymorphism of the non-transcribed spacer 2 in the ribosomal DNA unit. Syst. Appl. Microbiol., 20: 286-294. Nguyen H.V., Lepingle A., Gaillardin C.A. (2000). Molecular typing demonstrates homogeneity of Saccharomyces uvarum strains and reveals the existence of hybrids between S. uvarum and S. cerevisiae, including the S. bayanus type strain CBS 380. Syst. Appl. Microbiol., 23: 71-85. Paffetti D., Barberio C., Casalone E., Cavalieri D., Fani R., Fia G., Mori E., Polsinelli M. (1995). DNA fingerprinting by random amplified polymorphic DNA and restriction fragment length polymorphism is useful for yeast typing. Res. Microbiol., 146: 587-594. Pis kur J., Smole S., Groth C., Petersen R.F., Pedersen M.B. (1998). Structure and genetic stability of mitochondrial genomes vary among yeasts of the genus Saccharomyces. Int. J. Syst. Bacteriol., 48: 1015-1024. Polsinelli M., Romano P., Suzzi G., Mortimer R. (1996). Multiple strains of Saccharomyces cerevisiae on a single grape vine. Lett. Appl. Microbiol., 23: 110-114. Pramateftaki P.V., Lanaridis P., Typas M.A. (2000). Molecular identification of wine yeasts at species or strain level: a case study with strains from two wine-growing areas of Greece. J. Appl. Microbiol., 89: 236-248. Pretorius I.S., Van der Westhuizen T.J., Augustyn O.P. (1999). Yeast biodiversity in vineyards and wineries and its importance to the South African wine industry. A review. S. Afr. J. Enol. Vitic., 20: 61-74. Querol A., Barrio E., Barrio D. (1992a). A comparative study of different methods of yeast strain characterization. Syst. Appl. Microbiol., 15: 439-446. Querol A., Barrio E., Huerta T., Ramon D. (1992b). Molecular monitoring of wine fermentations conducted by active dry yeast strains. Appl. Environ. Microbiol., 58: 2948-2953. 424 F. DELLAGLIO et al.

Querol A., Barrio E., Ramon D. (1994). Population dynamics of natural Saccharomyces strains during wine fermentation. Int. J. Food Microbiol., 21: 315-323. Querol A., Fernandez-Espinar M.T., del Olmo M., Barrio E. (2003). Adaptive evolution of wine yeast. Int. J. Food Microbiol., 86: 3-10. Rainieri S., Zambonelli C., Hallsworth J.E., Pulvirenti A., Giudici P. (1999). Saccharomyces uvarum, a distinct group within Saccharomyces sensu stricto. FEMS Microbiol. Lett., 177: 177-185. Ribereau-Gayon P., Dubourdieu D., Doneche B., Lonvaud A. (2000). Handbook of Enology. The Microbiology of Wine and Vinifications. Vol. 1., Wiley & Sons, Chichester, UK. Sabate J., Cano J., Esteve-Zarzoso B., Guillamon J.M. (2002). Isolation and identification of yeasts associated with vineyard and winery by RFLP analysis of ribosomal genes and mitochondrial DNA. Microbiol. Res., 157: 267-274. Sabate J., Cano J., Querol A., Guillamon J.M. (1998). Diversity of Saccharomyces strains in wine fermentations: analysis for two consecutive years. Lett. Appl. Microbiol., 26: 452-455. Schutz M., Gafner J. (1994). Dynamics of the yeast strain populations during spontaneous alcoholic fermentation determined by CHEF gel electrophoresis. Lett. Appl. Microbiol., 19: 253-257. Schwartz D.C., Cantor C.R. (1984). Separation of yeast chromosomal-sized DNAs by pulsed-field gradient gel electrophoresis. Cell, 37: 67-75. Sipiczki M., Romano P., Lipani G., Miklos I., Antunovics Z. (2001). Analysis of yeasts derived from natural fermentation in a Tokaj winery. Antonie van Leeuwenhoek, 79: 97-105. Torriani S., Zapparoli G., Suzzi G. (1999). Genetic and phenotypic diversity of Saccharomyces sensu stricto strains isolated from Amarone wine. Antoine van Leeuwenhoek, 5: 207-215. Usseglio-Tomasset L., Bosia P.D., Delfini C., Ciolfi G. (1980). I vini Recioto e Amarone della Valpolicella. Vini d Italia, 22: 85-97. Vaughan-Martini A., Martini A., Cardinali, G. (1993). Electrophoretic karyotyping as a taxonomic tool in the genus Saccharomyces. Antoine van Leeuwenhoek, 63: 145-156. Vauterin L., Vauterin P. (1992). Computer-aided objective comparison of electrophoresis patterns for grouping and identification of microorganisms. Eur. Microbiol., 1: 37-41. Versavaud A., Courcoux P., Roulland C., Dulau L., Hallet J.W (1995). Genetic diversity and geographical distribution of wild Saccharomyces cerevisiae strains from the wine-producing area of Charentes, France. Appl. Environ. Microbiol., 61: 3521-3529. Vezinhet F., Hallet J.N., Valade M., Poulard, A. (1992). Ecological survey of wine yeast strains by molecular methods of identification. Am. J. Enol. Vitic., 43: 83-86. Ann. Microbiol., 53 (4), 411-425 (2003) 425