Use of WL Medium to Profile Native Flora Fermentations

198 Pallman et al. Use of WL Medium to Profile Native Flora Fermentations Christina L. Pallmann, 1 James A. Brown, 1 Tammi L. Olineka, 2 Luca Cocolin, 3 David A. Mills, 4 and Linda F. Bisson 4 * Vineyard, winery, barrel, and controlled temperature fermentation samples from a single commercial winery (Luna Vineyards) conducting fermentations with indigenous organisms were plated onto Wallerstein Laboratory Nutrient Agar (WL) to evaluate colony diversity. Seventeen unique colony morphologies were identified. Sequence analysis of the DNA encoding a portion of the large ribosomal 26S rrna indicated that the colony types defined members of six genera: Hanseniaspora uvarum (Kloeckera apiculata), Saccharomyces cerevisiae, Issatchenkia orientalis, Pichia kluyveri, Candida olephelia, and Metschnikowia. Distinct colony subtypes were identified within the pulcherrimin producers traditionally classified as a single species, Metschnikowia pulcherrima. Sequence analysis of the D1/D2 region of the 26S rdna of these biotypes showed a high degree of divergence, suggesting that these organisms might define separate species. Analysis of fermentations revealed that colony type as indicated on this medium could be used to monitor the yeast population dynamics. Key words: WL media, Saccharomyces, Kloeckera, Metschnikowia, native fermentation Many yeast species can be present during a native flora fermentation, but members of three genera are dominant: Hanseniaspora uvarum (Kloeckera apiculata), Metschnikowia pulcherrima (Candida pulcherrima), and Saccharomyces cerevisiae or S. bayanus [3,5,16]. The end products of metabolism of the first two organisms may or may not be desirable, depending upon the style of wine being produced. Numerous environmental parameters, such as temperature of fermentation, addition of nutrients or antimicrobial agents, settling, and clarification practices, can impact the ratio and numbers of the members of these species in the fermenting must [6]. It is desirable, therefore, to have a reliable means of quantifying the relative numbers of organisms present under enological situations. Differential media for the enumeration of Saccharomyces and non-saccharomyces strains has been proposed [9], but involves plating on two or more selective media to obtain quantitative information. Wallerstein Laboratory Nutrient Agar (WL) medium was designed for use in brewing and industrial fermentation processes to observe microbial populations and is not a very selective medium [8]. Recently, Cavazza et al. [4] showed that the majority of yeast species typically found in wine fermentations could be distinguished on the basis of colony color and/or morphology on WL medium. Single isolates from a culture collection were evaluated in that study. Our goal was to determine whether plating on WL medium could be used to follow changes in the populations of native yeast fermentations. Different yeast populations were identified initially on the basis of unique colony 1 Graduate Student, 2 Postgraduate Researcher, 3 Postdoctoral Fellow, and 4 Faculty, Department of Viticulture and Enology, University of California, Davis, One Shields Ave, Davis, CA USA 95616-8749 *Corresponding author [Email: lfbisson@ucd.edu; Tel: 530-752-3835; Fax: 530-752-0382] Acknowledgments: This research was supported by a grant from the American Vineyard Foundation. Manuscript submitted October 2000; revised July 2001 Copyright 2001 by the American Society for Enology and Viticulture. All rights reserved. morphology. Genomic DNA was isolated from representatives of each colony type and the genus and species determined from sequence comparisons of 26S rdna [12,13]. Materials and Methods Fermentation conditions. Chardonnay grapes from the Napa appellation were pressed at Luna Vineyards and Winery during the harvest of 1998. From tank, the musts were transferred to sterilized 5-gallon carboys and moved to the fermentation facilities of the Department of Viticulture and Enology, University of California, Davis (UCD). No additions were made to the must. The starting Brix was 26.2 with a ph of 3.64. The musts were fermented at 13 and 18 C, with three replicate carboys for each temperature. Carboys were stirred five times immediately prior to sampling. Samples were aseptically collected on a daily basis for eight days. Dilutions were plated in duplicate on WL medium with a target of 20 to 100 colonies per plate. Plates were counted after five days of incubation at 28 C. Colony forming units represent the average of the duplicate plates of all three replicate fermentations. Samples were also plated from the vineyard (harvested, crushed clusters) and winery equipment (juice sample passed through equipment early in harvest) as well as from new and old barrel fermentations conducted at Luna Vineyards using the same Chardonnay juice. Isolation of representative colony types. Unique colony types were restreaked from each condition and colonies purified on WL medium. Where possible, at least four of each type from each condition were restreaked. With some of the rarer colony morphologies, fewer than four colonies were obtained and all were restreaked. DNA preparation. Tubes containing 10 ml of YPD (yeast extract [10 g/l], peptone [20g/L], glucose [2%]) were inoculated from the purified colonies and incubated on a roller drum 198

Use of WL Medium to Profile Native Flora Fermentations 199 for 24 hr. Two ml of culture were then spun in microcentrifuge tubes at 12,000 rpm for 10 min. The pellet was resuspended in sterile deionized water and re-centrifuged. Following washing, the pellet was resuspended in 300 ml of breakage buffer. Cells were disrupted using glass beads [2] and DNA purified according to [2]. Sequence analysis. Genomic DNA was amplified by polymerase chain reaction (PCR) using the NL-1 and NL-4 primers developed by O Donnell [15]. These primers amplify the D1/ D2 domain (nucleotides 63-642) of the 5 end of the large subunit (LSU) 26S rrna gene. This region has been sequenced extensively across the yeast phyla and is currently being used in the taxonomic identification of yeast species. The PTC-100 Programmable Thermal Controller (MJ Research, Inc., Watertown, MA) was used. The PCR reaction was performed according to [12]. The PCR products were purified using the QIAGEN QIAquick PCR Purification Kit following manufacturer s instructions. PCR products were sequenced by Davis Sequencing (Davis, CA). Sequencing results were analyzed using the NCBI (www.ncbi.nlm.nih.gov). Sequences of the isolates were compared to each other using the CLUSTAL W Multiple Sequence Alignment Program version 1.7, available from http:// www.clustalw.genome.ad.jp/. The multiple sequences alignment and phylogenetic tree calculations were performed using the Jalview multiple alignment editor, available at http:// www2.ebi.ac.uk/~michele/jalview/. The species in the database used for comparison are designated M. pulcherrima, M6148, M6344, and M7275. The isolates obtained in this study are designated by UCD accession numbers and are available from the yeast culture collection of the Department of Viticulture and Enology. The M. pulcherrima colony morphology considered typical of the type species was that displayed by all isolates that generated a 100% match to the D1/D2 sequence of the type species as listed in the database. Atypical isolates were those with a noticeably different colony morphology. Results and Discussion Identification of yeast colony types on WL. WL medium was originally developed for monitoring yeast populations during brewing fermentations [8]. Cavazza et al. [4] published detailed descriptions of the color and colony topography of wine yeasts on WL Nutrient Agar (translated in Table 1). The main species present in native yeast fermentations, Hanseniaspora uvarum, Metschnikowia pulcherrima, and Saccharomyces cerevisiae/s. bayanus, are distinguishable on this medium, which suggested that it might be useful for following yeast population dynamics under enological conditions. It is also possible to identify minor yeast species and to determine the extent to which such species persist in the fermentations. To assess the utility of this medium in monitoring native yeast flora Chardonnay fermentations, samples were plated from the vineyards, from winery equipment during harvest, from in-house barrel fermentations at Luna Vineyards, and during the course of small-lot fermentations conducted at the University of California, Davis using the same Chardonnay juice. Pilot scale fermentations were conducted at two temperatures, 13 and 18 C. Several distinct colony types were observed (Table 2). Some colony types were found at all sampling sites (vineyard, winery equipment, barrel, and small-lot fermentations), while some were not as widespread. Hanseniaspora (intense green) and Metschnikowia (brown/red pulcherrimin pigment production) were isolated from all vineyard, winery, and fermentation samples. Saccharomyces (pale green to cream colonies) was occasionally observed in vineyard and winery equipment platings, but was predominant in the later stages of all fermenting samples. The identity of the isolates was confirmed by DNA sequence analysis of the D1/D2 region of the 26S rdna [12,13]. All of the intense green colonies yielded DNA sequences with a 100% match to the sequence of this region from Hanseniaspora uvarum in the NCBI database (data not shown, sequences available on request). Similarly, the DNA sequence of the 26S region from all of the pale green to cream-colored colonies with a distinct dome was identical to that of Saccharomyces cerevisiae (100% match). Three other colony types also matched known sequences Table 1 Translation of descriptions of yeast colonies on WL medium from Cavazza et al. [4]. Strain Colony color Colony topography Saccharomyces Cream to green Knoblike, surface: smooth, opaque cerevisiae Torulaspora Cream, can have Knoblike, surface: smooth, opaque delbrueckii a hint of green Hanseniaspora uvarum Intense green Flat, surface: smooth, opaque (Kloeckera apiculata) Consistency of butter Saccharomycodes Bright green Knoblike, convex, surface: smooth, opaque ludwigii Schizosaccharomyces Intense green Very small size, surface: smooth, opaque pombe Consistency of butter Rhodutorula species Red Knoblike, convex surface: smooth, mucoid Consistency of butter Metschnikowia Cream with hint of red; Small size, convex pulcherrima red-brown from bottom Consistency of flour Pichia Gray-green with hint of blue Elevated and convex, surface: wrinkled membranefaciens Consistency of flour Hansenula anomala Cream to blue gray; Flat, surface: smooth blue after 8 days Brettanomyces Cream, appears after Small, elevated to a dome, surface: smooth intermedius 8 days Zygosaccharomyces Cream Small, elevated dome, surface: smooth bailii

200 Pallman et al. Table 2 Description of isolates. Strain Colony description Source UCD2102 Brown with green center and white rim, Vineyard small dome UCD2103 Cream, knoblike 18 C UCD2105 Light brown, small dome Barrel UCD2106 Brown/red with small dome 13 C UCD2107 Brown/red with white rim small dome Vineyard UCD2108 Brown/red with brown/green middle 13 C UCD2110 Intense green, flat Vineyard UCD2111 Intense green, flat 13 C UCD2112 Intense green, flat 18 C UCD2113/ Green middle with white rim, flat Vineyard UCD2119 and 18 C UCD2115 White, wrinkled and rough surface, 13 C flat with volcanic center UCD2116 White, fuzzy with volcanic center 13 C UCD2117 Light brown with white rim, intense Vineyard green center UCD2118 White with greenish hue, knoblike 13 C UCD2120 White with dome 13 C UCD2121 Greenish white with green center, flat 18 C UCD2122 White with greenish hue and dome 13 C (100% match) in the database. UCD2113 (UCD2119), found in the vineyard as well as in the 18 C fermentations, was identified as Candida oleophilia. UCD2115, which gave distinct volcanic-looking colonies isolated from the 13 C fermentations, was identified as Pichia kluyveri, and the colony type that was similar was identified as Issatchenkia orientalis (UCD2116). There were noticeable differences in the appearance of some of the minor yeast species in the fermentations conducted at the two temperatures. Apiculate yeasts (Hanseniaspora/Kloeckera) initially colonize many fruit surfaces and are dominant at early stages of ripeness [1,14]. As the fruit deteriorates, yeasts capable of using a broader spectrum of carbon and energy sources become apparent on the surface [1,14]. Issatchenkia orientalis and members of the genera Pichia and Candida are in this latter category [1,14]. While this study of yeast populations was reported on fruit other than grape, our data are consistent with these observations. Interestingly, by using WL medium and by virtue of the fact that these yeasts produce distinctive colonies, we were able to UCD2108 11 (+1N) 6 5 M pul 8 (+1N) 3 demonstrate the persistence of UCD2106 7 (+1N) these minor species in the controlled temperature fermentations. The fact that Pichia and Issatchenkia were only found at 13 C while Candida was present in samples from only the 18 C condition suggests that juice fermentation temperature will play an important role in the nature of minor yeast species present in native flora fermentations. However, this conclusion needs to be interpreted with caution as these yeasts are present in low numbers, typically 0 to 2 colonies per plate of roughly 100 total colonies. Accurate quantification is difficult with this range of diversity in relative numbers of colony forming units. Diversity of Metschnikowia species in native yeast fermentations. In contrast to the other isolates, there was significant 26S rdna sequence diversity in the D1/D2 region among the red pigment (pulcherrimin) producers. Currently, only one species of pigment producers is recognized: Metschnikowia pulcherrima [11]. In general, Metschnikowia isolates display high sequence divergence in the D1/D2 region. Opposite mating types of M. agaves differed by as much as five nucleotides [13]. Typically, conspecific species vary by no more than two to three nucleotides over 580 bp [13]. M. pulcherrima is frequently present in native flora fermentations [3,5,6,7,10,16]. We noted several distinct colony types producing a pigment (Table 2). There were three colony types producing a light brown color distinguishable on the bases of a tinge of green in the center (UCD2102), or a distinct white rim with an intense green center (UCD2117), or solid brown (UCD2105). These three colony types yielded identical sequencing data from the 26S rdna analysis, indicating that they are the same species (Figure 1). They displayed from 2 to 10 base pair differences with the other isolates, having 5 nucleotide differences from M. pulcherrima (Table 3). Table 3 also includes information on the number of bases for each organism that were not able to be unequivocally determined by sequencing that are traditionally indicated by N for nucleotide. The UCD2102 cluster showed two base pair mismatches with Metschnikowia species 6148 (AF017401). This is similar enough to suggest that they might be the same species. Interestingly, this strain was found in the vineyard and barrel samples, but not from the fermentation samples. While further studies are necessary, it is possible that a distinct group of Metschnikowia subspecies may Table 3 Analysis of divergent base pairs in Metschnikowia species. UCD2107 UCD2106 M pul UCD2108 UCD2105 M6148 UCD2117 UCD2102 M6344 M7275 6 (+5N) a 3 (+4N) 6 (+4N) 8 (+4N) 8 (+4N) 8 (+4N) 8 (+4N) 8 (+4N) 4 (+4N) M6344 11 (+1N) 6 5 8 8 8 8 8 UCD2102 10 (+1N) 6 5 2 0 2 0 UCD2117 10 (+1N) 6 5 2 0 2 M6148 10 (+1N) 6 7 4 2 UCD2105 10 (+1N) 6 5 2 a N indicates the number of bases that could not be unequivocally determined from the sequence analysis.

Use of WL Medium to Profile Native Flora Fermentations 201 gin (UCD2107), one had a distinct green pigmentation in the middle of the colony (UCD2108) similar to UCD2117, and the third was a more uniform dark brown (UCD2106). The DNA sequences of these isolates were not identical to each other (Figure 1). One isolate, UCD2108, ranged from 2 to 11 base pair differences with the other isolates and was closest in sequence identity to the UCD2102 cluster. UCD2108 displayed four nucleotide differences with M. species 6148, so the relationship among these strains is not clear. UCD2106 displayed six base pair differences with most other isolates. UCD2106 yielded the fewest nucleotide differences (three) when compared to the type species, Metschnikowia pulcherrima. However it also showed only three differences versus the entered sequence for M. species 7275. The sequence for M. species 7275 in the database displays many N or undetermined bases, making analysis of divergence ambiguous. However this species displays six nucleotide differences from M. pulcherrima, suggesting it is a distinct biotype. UCD2107 was highly divergent with six nucleotide differences from the nearest neighbor M. species 7275. A phylogenetic tree was prepared displaying the relationships between these isolates (Figure 2). This analysis suggests that there is considerable diversity among the pulcherrimin-producing members of the Metschnikowia genus. Analysis of the yeast flora of native fermentations. WL medium was used to monitor changes in different types of colonies over the course of fermentation at two different temperatures. For fermentation monitoring, colonies were divided into four categories: intense green (Hanseniaspora/ Kloeckera); pulcherrimin producers (the Metschnikowia species); white to pale green (Saccharomyces); and other. The latter group contained a mixture of colony types and was only rarely observed. At both temperatures, Hanseniaspora populations persisted until the end of fermentation (Figure 3). Population density UCD2107 M6344 M7275 Mpulcherrima UCD2108 Figure 1 Sequence alignment of the D1/D2 region of Metschnikowia isolates. Fully conserved base pairs are indicated by a dash (-), divergent bases are indicated by letter. be selected during fermentative growth. Indeed, it has been reported previously that Metschnikowia strains are found in a higher proportion of total yeast species in fermenting must as compared to the surface of the grapes [5,6]. Three other colony types produced significantly more pigment and were a deep brown/red in color. As with the previous isolates, one of the colony types had a distinct white rim or mar- M6148 UCD2102 UCD2117 UCD2105 UCD2106 Figure 2 Multiple sequence alignment and phylogenetic tree of D1/D2 regions of the Metschnikowia species. Calculations were performed with the Jalview multiple alignment editor.

202 Pallman et al. A A Colony forming units (log value x 10-4 ) B Colony forming units (log value x 10-3 ) B Time (day) Figure 3 Changes in colony types over time at two fermentation temperatures: 18 C (panel A) and 13 C (panel B). Data is plotted as the log of the colony forming unit value, following division by a factor of 10,000. Kloeckera/Hanseniaspora (intense green) colonies: ; Metschnikowia (pigment producing) colonies: ; Saccharomyces (cream colored) colonies:. decreased more at the higher temperature. Metschnikowia populations were roughly the same density in both fermentations through day five, at which point significant loss of population was evident at the higher temperature (18 C) (Figure 3). At the lower temperature, the population was able to persist for seven days. Saccharomyces was initially undetectable, with colonies beginning to appear after day two at 18 C and after day three at 13 C, and persisted throughout the rest of the fermentation (Figure 3). Population densities were also compared at three time points of fermentation in old and new barrels (Figure 4). There was no significant difference in the Hanseniaspora populations in either barrel type. Likewise, Metschnikowia isolates were present under both conditions at a similar level. Saccharomyces levels appeared to rise more quickly in the old barrels, but terminal cell densities were identical. Conclusions This study demonstrated the utility of plating on WL medium as a tool to monitor yeast population diversity during native flora fermentations. The colony morphologies of the yeasts typically present on grape surfaces and in fermenting musts are sufficiently unique to be identified to genus/species by plating. This method allowed identification of yeasts present in the fermentations at Time (day) Figure 4 Colony types isolated from barrel fermentations. Panel A: old (used) barrels; panel B: new (unused) barrels. Kloeckera/Hanseniaspora (intense green) colonies: ; Metschnikowia (pigment-producing) colonies: ; Saccharomyces (cream colored) colonies:. low numbers members of the genera Pichia, Issatchenkia, and Candida due to their dramatically different morphologies on this medium. While minor species, the metabolic activities of these yeasts may impact wine composition and quality. Identities of isolates can be confirmed by DNA sequence analysis of the D1/D2 region of the 26SrRNA. A surprising finding was the level of diversity among the pulcherrimin producers found in a single winery with respect to sequence of the D1/D2 region. Several distinct biotypes were found based upon colony morphology differences. Some strains displaying distinctly different colonies, the UCD2102 cluster, were found to be identical in sequence while others were not the same. A more systematic analysis of diversity among the pulcherrimin producers not biased by colony morphology is needed. Further analysis will reveal if the distinct biotypes observed herein represent independent species. Literature Cited 1. Abranches, J., M.J.S. Vital, W.T. Starmer, L.C. Mendonca-Hagler, and A.N. Hagler. The yeast community and mycocin producers of guava fruit in Rio de Janeiro, Brazil. Mycologia 92:16-22 (2000). 2. Ausubel, F.M., R. Brent, R.E. Kingston, D.D. More, J.G. Seidman, J.A. Smith, and K. Struhl. Current Protocols in Molecular Biology. Wiley & Sons, New York (1995). 3. Bisson, L.F., and R.E. Kunkee. Microbial interactions during wine production. In Mixed Cultures in Biotechnology. J.G. Zeikus and A.E. Johnson (Eds.), pp. 37-68. McGraw Hill, New York (1993).

Use of WL Medium to Profile Native Flora Fermentations 203 4. Cavazza, A., M.S. Grando, and C. Zini. Rilevazione della flora microbica di mosti e vini. Vignevini 9:17-20 (1992). 5. Fleet, G.H. The microorganisms of winemaking: Isolation, enumeration and identification. In Wine Microbiology and Biotechnology. G.H. Fleet (Ed.), pp. 1-25. Harwood Academic Publishers, Switzerland (1993). 6. Fleet, G.H., and G.M. Heard. Yeasts: Growth during fermentation. In Wine Microbiology and Biotechnology. G. H. Fleet (Ed.), pp. 27-54. Harwood Academic Publishers, Switzerland (1993). 7. Gao, C., and G.H. Fleet. The effects of temperature and ph on the ethanol tolerance of the wine yeasts, Saccharomyces cerevisiae, Candida stellata and Kloeckera apiculata. J. Appl. Bacteriol. 65:405-409 (1988). 8. Green, S.R., and P.P. Gray. A differential procedure applicable to bacteriological investigation in brewing. Wallerstein Comm. 13:357-366 (1950). 9. Heard, G.M., and G H. Fleet. Evaluation of selective media for enumeration of yeasts during wine fermentations. J. Appl. Bacteriol. 60:477-481 (1986). 10. Heard, G.M., and G.H. Fleet. The effects of temperature and ph on the growth of yeast species during the fermentation of grape juice. J. Appl. Bacteriol. 65:23-28 (1988). 11. Kurtzman, C.P., and J.W. Fell. The Yeasts: A Taxonomic Study. Elsevier, Amsterdam (1998). 12. Kurtzman, D.P., and C.J. Robnett. Identification of clinically important ascomycetous yeasts based on nucleotide divergence in the 5 end of the large subunit (26S) ribosomal DNA gene. J. Clin. Microbiol. 35:1216-1223 (1997). 13. Kurtzman, C.P., and C.J. Robnett. Identification and phylogeny of ascomycetous yeasts from analysis of nuclear large subunit (26S) ribosomal DNA partial sequences. Ant. Van Leeuwen. 73:331-371 (1998). 14. Morais, P.B., M.B. Martins, L.B. Klacxko, L.C. Mendonca-Hagler, and A.H. Hagler. Yeast succession in the amazon fruit Parahancornia amapa as resource partitioning among Drosophila spp. Appl. Environ. Microbiol. 61:4251-4257 (1995). 15. O Donnell, K. Fusarium and its near relatives. In The Fungal Holomorph: Mitotic, Meiotic and Pleomorphic Speciation in Fungal Systematics. D.R. Reynolds and J. W. Taylor (Ed.), pp. 225-233. CAB International, Wallingford, UK (1993). 16. Schütz, M., and J. Gafner. Analysis of yeast diversity during spontaneous and induced alcoholic fermentation. J. Appl. Bacteriol. 75:551-558 (1993).