Selection and improvement of wine yeasts

Transcription

1 Annals of Microbiology, 50, (2000) Selection and improvement of wine yeasts S. RAINIERI, I.S. PRETORIUS* Institute for Wine Biotechnology, University of Stellenbosch, Stellenbosch, Private Bag X1, Matieland 7602, South Africa Abstract - The selection of wine yeasts is usually carried out within the species Saccharomyces cerevisiae. It aims at identifying the yeast strains that, besides fermenting grape juice vigorously and producing high ethanol yield, can also positively influence the composition and the sensorial characteristics of wine. The natural availability of yeast strains possessing an ideal combination of oenological characteristics is highly improbable. Moreover, selected S. cerevisiae wine strains usually produce wines with a plain aromatic profile. The extension of the selection of wine yeasts to S. cerevisiae not growing in oenological environments or to non-saccharomyces yeasts has provided strains possessing novel and interesting oenological characteristics. Nevertheless, these strains cannot be directly used as starter cultures in wine fermentations, mainly because they are not vigorous or competitive in oenological conditions. Wine strains possessing innovative oenological traits that can influence the sensorial characteristics of wine can be constructed using genetic or molecular methods. Intraspecific S. cerevisiae hybridisation has provided useful oenological strains. Nevertheless, the traits of oenological interest that can be exchanged or introduced using this technique are only those commonly found in the species S. cerevisiae. Innovative oenological traits can be introduced or exchanged by hybridising strains belonging to different species but with a sufficient genetic affinity for them to mate. Interspecific Saccharomyces hybrids were found to be stable, vigorous and possessing the parental oenological traits in novel and interesting combinations. Nevertheless, they are sterile; the genetic improvement cannot therefore be taken further than the first generation. Moreover, the combination of the parental traits cannot be specifically programmed and a combination of positive traits is often the result of chance. The recent development of recombinant DNA technology has overcome the limitations of traditional genetic techniques as well as broadening the potential of wine yeast improvement. Key words: genetic improvement, Saccharomyces, selection, wine. INTRODUCTION The term wine is most commonly used to refer to the fermented product of grape juice. Similar products can be obtained from apples and pears (e.g. British cider * Corresponding author. Phone ; Fax ; isp@maties.sun.ac.za. 15

2 and perry), dates (e.g. Sudanese sherbote), bananas (e.g. Kenian urwaga), prickly pear cacti (Opuntia spp.; e.g. Mexican colonche), Agave (e.g. Mexican plaque), palm (e.g. Nigerian emu, and Indian kallu), jackfruit (Artocarpus heterophyllus; e.g. Indian jackfruit wine) and other plants (for a review see Steinkraus, 1996). Many of these drinks are produced on a small scale using basic, traditional methods. By contrast, a global high-tech industry has developed to produce wine from grape juice. Before wine production was extensively commercialised, wines were the result of natural fermentation carried out by the microflora of grapes. The microorganisms responsible for ethanol production were identified as yeasts during the second half of the 19 th century (Demain and Solomon, 1981). Countless studies have since confirmed that yeasts play a critical role in determining the body, viscosity, colour, flavour and aroma of wines. This review summarises the work carried out during the last 20 years to improve wine quality by selection for, and development of, improved yeast strains. SPONTANEOUS AND GUIDED FERMENTATION Grape juice can support the growth of a multitude of microorganisms. However, its low ph (3 to 3.5) and high sugar content (average 200 g l -1 ) exert a selective pressure on microbial populations. Wine has traditionally been produced by allowing the microorganisms naturally present on grapes to grow. Yeasts, fungi, acetic acid bacteria and lactic acid bacteria are the main microorganisms able to grow in grape juice. Once fermentation is under way, the anaerobic conditions that are created contribute to the selective pressure, and microorganisms incapable of fermentative metabolism (such as fungi and acetic acid bacteria) are inhibited. As the fermentation progresses nutrients are depleted and ethanol concentration increases, thereby inhibiting ethanol-sensitive species (Henschke, 1997). Yeasts of the genera Hanseniaspora (Kloeckera) generally start the fermentation of grape juice. Although these yeasts are vigorous, they tend not to be ethanol-tolerant. As the ethanol concentration reaches 3 to 4% (vol/vol) more ethanoltolerant species predominate; most notably Saccharomyces cerevisiae (Fleet and Heard, 1993; Mortimer et al., 1994). There are, however, several other yeasts that can grow, such as Torulaspora delbrueckii, Zygosaccharomyces bailii and some Schizosaccharomyces spp. Saccharomycodes ludwigii, Metschnikowia pulcherrima and Brettanomyces spp. are also occasionally present (Fleet and Heard, 1993; Henschke, 1997). S. cerevisiae plays the main role in the spontaneous fermentation of grape juice. However, Saccharomyces yeasts are not necessarily widespread on grapes. The predominance of S. cerevisiae in wine is a consequence of the conditions, especially the high ethanol concentration, that inhibit other microbial species more severely. After the importance of yeasts for wine quality was established, a technique to isolate yeasts derived from a single cell was developed. In 1890, Muller-Thurgau produced wine by inoculating grape juice with a pure yeast culture (Kunkee and Amerine, 1970). Yeast strains have since been selected and commercialised to be used as starter cultures in fermentations. The use of starter cultures has improved the reproducibility and the predictability of the quality of wines. During the last thirty years, wine producing countries such as Australia, South Africa and the USA modified the traditional winemaking practices and introduced inoculation of 16

3 grape juice using S. cerevisiae starter cultures, especially for large-scale wine productions. Nevertheless, many winemakers still prefer wines produced by spontaneous fermentation. According to Dittrich (1995) natural fermentations are still carried out in most European wineries, and 80% of the wines, worldwide, are produced using this method (Heard, 1999). The main critics of the practice of guided fermentations (using starter cultures) dislike the fact that the commercial wine strains, despite being numerous, possess very ordinary characteristics. Commercial yeast strains produce wines with average qualities and do not enhance the aromatic traits that characterise many yeasts isolated from specific geographical areas. Studies on the improvement and the selection of wine yeasts to overcome this problem have recently been carried out. THE SELECTION OF WINE YEASTS The selection of yeasts for winemaking consists of identifying those cultures that can ferment grape juice efficiently and produce good quality wines. The selection is generally carried out within the genus Saccharomyces. Yeast cultures are preferably isolated from grape juice or wine. Saccharomyces strains growing in these substrates are, in fact, well adapted to the oenological environment and can therefore ferment grape juice very efficiently. Nevertheless, Saccharomyces yeasts are scarcely present on grapes. Their isolation on solid media (Malt agar or Sabouraud dextrose agar), may therefore not be suited, especially if a relevant number of cultures need to be collected. The use of an enrichment technique is generally preferred (Frezier and Dubourdieu, 1992; Fleet, 1993; Querol et al., 1994; Schütz and Gafner, 1994; Versavaud et al., 1995; Constantí et al., 1997). This method consists of creating the conditions that favour the growth of some microorganisms in a mixed population, and inhibit the growth of the rest of the population. The high concentration of ethanol that accumulates in grape juice during fermentation is the main factor favouring the growth of Saccharomyces strains. Isolations are therefore carried out in fermenting grape juice. After a relevant number of strains have been isolated, they are oenologically characterised. The oenological traits of S. cerevisiae have been divided by Zambonelli (1998) into two groups, i.e. technological and qualitative. The technological traits to be considered in the selection of wine yeasts are listed in Table 1. These traits influence the efficiency of the fermentation process. S. cerevisiae strains generally possess the technological characteristics required to perform an efficient fermentation. The determination of these traits is, however, necessary, since most of these characteristics can vary among the strains. The qualitative traits of S. cerevisiae to be evaluated in the selection of wine yeasts are listed in Table 2. These traits help to determine the chemical composition and influence the sensorial characteristics of wines. During grape juice fermentation yeasts convert sugars to ethanol and CO 2. They also produce numerous compounds derived from the sugars fermented (such as glycerol, succinic acid and acetic acid) and from the other grape juice components (such as acetic aldehyde, higher alcohols, malic acid, nitrogen compounds, H 2 S, sulfites, and other sulfuric compounds). The concentration of these compounds varies within yeast species and among yeast strains. 17

4 TABLE 1 Technological characteristics to be considered in the selection of wine strains Ethanol tolerance Fermentation vigour Resistance to SO 2 Type of growth in liquid media Dispersed cells Aggregates cells Flocculence Foam formation Film formation Sedimentation speed Growth at high and low temperatures Presence of killer factor TABLE 2 Qualitative characteristics to be considered in the selection of wine strains Fermentation by-products Glycerol Succinic acid Acetic acid Acetaldehyde n-propanol Iso-butanol Isoamyl alcohol β-phenylethanol Production of sulfuric compounds H 2 S SO 2 Action on malic acid Enzymatic activity β-glucosidase Esterase Proteolytic enzymes Autolysis 18

5 The oenological traits can be evaluated by carrying out small-scale fermentations in synthetic media and eventually in grape juice. Technological traits can be assessed by monitoring the progression of the fermentation; quantitative traits can be assessed by determining the concentration of the compounds found in wine post-fermentation. Large scale fermentations in grape juice from different grape cultivars can then be carried out with the strains showing the best combination of oenological traits. Sensorial analysis of the wines obtained can also be carried out to improve the description of the strains selected. The natural availability of S. cerevisiae strains possessing an ideal combination of technological and qualitative traits is highly improbable. Nevertheless, wine strain characteristics can be improved and ideal combinations of oenological traits can be artificially achieved by means of genetic and molecular techniques. The selection process is, however, a fundamental step in any yeast improvement programme. SELECTION OF ALTERNATIVE WINE YEASTS S. cerevisiae strains selected for winemaking ferment grape juice vigorously, leave small amounts of unfermented sugars in the medium and produce wines with high ethanol concentrations. However, they generally produce wines with ordinary and plain aromatic profiles. S. cerevisiae strains isolated from non-oenological environments, as well as non-s. cerevisiae strains, have been found to possess interesting and novel oenological characteristics that exert a positive influence on the sensorial profile of wines. Nevertheless, these strains do not usually ferment grape juice vigorously or efficiently, and for this reason they are not employed as starter cultures in wine fermentations. However, they can be successfully employed as parental strains in yeast improvement programmes. S. cerevisiae isolated from non-oenological environments The potential oenological use of yeast strains isolated from non-oenological environments has recently been explored. Rainieri et al. (1996, 1998b) isolated and oenologically characterised S. cerevisiae strains from Parmesan whey. The outstanding characteristic of these strains is their ability to degrade the malic acid present in the grape juice. Indeed these strains can decrease the starting concentration of this compound up to 50%. Degradation of malic acid is a highly desirable trait, especially for the production of wines in areas with a cool to cold climate, where the acidity of grape juice is usually excessive. S. cerevisiae wine strains are usually unable to cause such an efficient deacidification of grape juice. S. cerevisiae strains isolated from whey are not suitable for employment as starter cultures for wine fermentations. Their fermentation vigour is low at temperatures of oenological interest, they often leave high concentration of unfermented sugars (1.5-2%) and they produce excessive amounts of acetic acid. However, they have been successfully used in yeast improvement programmes (Rainieri et al., 1998a; 1999b). Non-S. cerevisiae strains Besides S. cerevisiae, the Saccharomyces strains that can easily be isolated from 19

6 grape juice and wine belong to the group Saccharomyces uvarum (Rainieri et al. 1999a). These strains can ferment vigorously at low temperatures (6-10 C) and are often responsible for starting fermentations in cold-stored grape juices (Castellari et al., 1992). The major oenological characteristic of these strains is their ability to synthesise malic acid. An increase in malic acid can contribute to improving the acidity of wines produced in areas with hot climates, where grape juice acidity is usually insufficient (Castellari et al., 1994). These strains also produce very low concentrations of acetic acid, and high concentrations of glycerol and succinic acid, important traits for the improvement of wine aromatic profiles (Kishimoto et al., 1993; Castellari et al., 1994). S. uvarum strains produce wines with lower amounts of ethanol compared with S. cerevisiae wine strains. They also produce high concentrations of higher alcohols, especially β-phenylethanol, that spoil the aromatic characteristics of wine (Bertolini et al., 1996). S. uvarum strains have been used as starter cultures in wine fermentations, but the optimal application of their characteristics can be achieved by improving them genetically or using molecular methods. Use of wine indigenous non-saccharomyces yeasts as starter cultures The proliferation of some non-saccharomyces yeast species, even though indigenous to grapes, is not always desirable during grape juice fermentation. Non- Saccharomyces yeasts are not ethanol-tolerant, they are sensitive to SO 2 and produce undesirable high concentrations of acetic acid and ethyl acetate (Fleet and Heard, 1993). For these reasons they are regarded as unsuitable for winemaking and they have not been considered in studies concerning the selection of wine yeasts. Only fairly recently has the potential application of non-saccharomyces indigenous wine yeasts in winemaking been explored. Several studies have pointed out that some yeast species belonging to the genus Candida, Kloeckera or Hanseniaspora can positively influence the overall character of the wine, mainly by enhancing aromatic properties and imparting complex and novel flavour profiles (Fleet and Heard, 1993; Fleet, 1997; Romano, 1997). Non-Saccharomyces indigenous wine yeasts usually produce high concentrations of esters, higher alcohols, aldehydes and glycerol; compounds that play a major role in determining the sensorial profile of wine. The production of such compounds varies within the different yeast species. For instance, strains belonging to the species Candida stellata and Kloeckera apiculata were found to produce high concentrations of glycerol; strains belonging to the species Candida colliculosa were found to produce high concentrations of acetaldehyde and high concentrations of n-propanol (Heard, 1988). Soden et al. (2000) noticed an increase in the concentration of glycerol, acetic acid and ethyl acetate in wines obtained by inoculating grape juice with a strain of C. stellata. They carried out a sensory analysis on this wine and compared the results with those of a wine obtained using a S. cerevisiae wine strain. The C. stellata product showed a more intense honey, apricot, sauerkraut and ethyl acetate aroma and a diminished lime, banana and floral aroma ascribed to the S. cerevisiae wine strain. Garcia et al. (2000) increased the aroma compounds exploiting the high levels of β-glucosidase of a strain of Debaryomyces vanriji. Species such as K. apiculata and C. stellata are naturally found in wine at the early stage of fermentation, but as ethanol is produced to concentrations of 4 to 5% their growth is inhibited. Due to the sensitivity to ethanol, indigenous non- 20

7 Saccharomyces wine strains can be employed as starter cultures in winemaking only in conjunction with more ethanol-tolerant strains such as S. cerevisiae, ensuring the completion of fermentation. Grape juice can either be inoculated with a mixed culture of non-saccharomyces yeasts and a vigorous S. cerevisiae wine strain, or with a sequential inoculation of a number of non-saccharomyces yeasts followed by inoculation of an S. cerevisiae wine strain. Non-Saccharomyces yeasts are increasingly being used for commercial wine production in the USA and Australia (Goldfarb, 1994; Ramey, 1995; Price, 1996). This practice represents a good alternative to the relatively unpredictable and potentially problematic spontaneous fermentation. Indeed, since these non-saccharomyces yeasts are indigenous to wine, they can contribute in enhancing the wine nuance that is typical of a specific area. IMPROVEMENT OF WINE YEASTS The characteristics of yeast strains can be modified to accommodate the requirements of the fermentation industry. Several yeast improvement techniques were developed over the last century. Induced mutation and selection, hybridisation, rare-mating, spheroplast fusion, and gene cloning and transformation are among the most widely used techniques. Detailed description of these and other improvement techniques have been given by several authors (Pretorius and van der Westhuizen, 1991; Barre et al., 1993; Zambonelli, 1998; Pretorius, 2000). Recent results as well as strategies being applied in the improvement of wine strains by using the hybridisation technique and recombinant DNA technology, are discussed in the present work. HYBRIDISATION Cells of S. cerevisiae proliferate by multilateral budding. When nutritional starvation occurs, diploid cells undergo meiosis generating four haploid spores inglobated into the mother cell (ascus). When released from the ascus the spores can grow to originate haploid cultures known as single spore cultures. Spores possess a specific sexual type. Spore sexuality is determined by a gene occupying the MAT locus which can be expressed in the allelic forms of α or a. Yeasts are termed heterothallic when spores possess a stable sexual type, either a or α, and the cultures originating from these spores are permanently haploid. Conjugation between spores belonging to the same culture, and consequently to the same sexual type, are not possible. Heterothallic strains are therefore permanently haploid. Yeasts are termed homothallic when some of the spores belonging to a culture can modify their sexual status through a phenomenon known as switching of the mating type. In this case a single culture contains spores with opposite sexual types. These spores can conjugate originating a diploid strain (autodiploidisation) homozygous for all the traits but the sexual type. Most S. cerevisiae strains in nature are homothallic, whereas laboratory strains are generally heterothallic. The first studies on yeast hybridisation were carried out by Winge and Laustsen (1938). Hybridisation can be carried out between heterothallic yeast strains 21

8 (by conjugation of haploid cells), as direct spore-cell mating (by hybridisation between spores obtained from a homothallic strain and cells of a heterothallic strain), and between homothallic yeast strains (by spore conjugation) (Pretorius and van der Westhuizen, 1991; Barre et al., 1993; Zambonelli, 1998). The conjugation of haploid cells of selected strains is a simple and fast method. Its use in the improvement of wine strains is restricted by the limited availability of heterothallic strains of oenological interest. Nevertheless, this technique has been successfully used in the construction of killer yeast strains resistant to SO 2 by hybridising a killer haploid strain with an SO 2 resistant strain possessing good winemaking qualities (Thornton, 1983). Direct spore-cell mating was used by Eschenbruch to construct yeast strains not producing foam (Eschenbruch et al., 1982); by Thornton to construct strains with an increased fermentative efficiency and an improved resistance to SO 2 (Thornton, 1982); and to produce flocculent wine strains (Thornton, 1985). The hybridisation of homothallic strains requires a complex procedure, but wine strains being generally homothallic, it is usually the most widely used method. Homothallic strains are diploid, being able to autodiploidise. A natural diploid strain could be heterozygous for a few characteristics. The progenitor of a hybridisation programme must be homozygous, especially for the traits of interest. Cultures obtained from the single spore cultures of diploid homothallic strains are homozygous for all traits and can be used as parental strains. Single spore cultures of the strains to be hybridised are grown separately in nutritive medium. Asci are broken, either enzymatically (using the enzyme Zymoliase) or mechanically (using a micromanipulator), and individual spores are separated using a micromanipulator. Spores are then put together and conjugation is assessed by microscopic observation. Intraspecific hybridisation When yeasts belonging to the same species are hybridised, the hybrid culture can sporulate; its spores are isolated and analysed to establish which culture possesses the desired characteristics. Yeast strains of oenological interest frequently present abnormalities in their ploidy and as a consequence a poor sporulation ability (Stewart, 1981). It is therefore necessary to select a very high number of potential parental strains when planning to construct strains with specific characteristics. This technique was successfully employed by Romano et al. (1985). They constructed a flocculent S. cerevisiae strain not producing H 2 S to be used in the production of sparkling wines by hybridising a flocculent strain with a H 2 S non-producing strain. One of the limitations of hybridising strains belonging to the same species is that the traits that can be exchanged or introduced in the hybrid culture and in its progeny are limited to the characteristics typical of this species. To obtain innovative results in the improvement of wine strains using the hybridisation technique, an alternative approach should be evaluated. Interspecific hybridisation Single spore cultures obtained from strains belonging to different species can be hybridised if they possess a sufficient genetic affinity. The hybrid culture obtained is sterile and therefore cannot sporulate. As a consequence, the result of the genetic improvement is represented by the hybrid itself. The characteristics of 22

9 TABLE 3 Results recently achieved by using the recombinant DNA technology in the improvement of wine strains OBJECTIVE STRATEGY Improvement of fermentation performance Efficient sugar utilisation Overexpression of HXT permease gene family in S. cerevisiae to enhance sugar uptake. Improved nitrogen assimilation Construction of mutated wine yeast containing a recessive ure2 gene (repressor of the genes involved in the conversion of proline to glutamate) to increase the assimilation of the abundant supply of proline of grape juice. Increased tolerance to antimicrobial compounds Increase of the resistance to copper by integrating the CUP1 gene (encoding for a copper binding protein) at multiple sites in the genome to enable wine yeasts to tolerate higher copper residue in grape juice. Clarification of wine Reduction of protein hazes Overexpression of PEP4 gene, encoding for the vacuolar protease A, to activate the protease itself, normally inactive. Prevention of polysaccharide hazes Co-expression of pectate-lyase gene (pele) from Erwinia chrysanthemi and polygalacturonase gene (peh1) from Erwinia carotovora in wine yeasts to enable the yeast to degrade polypectate efficiently. Co-expression of the endo-β-1,4-glucanase gene (END1) from Butyrivibrio fibrisolvens, the endo-β-1,3-1,4-glucanase gene (BEG1) from Bacillus subtilis, the cellodextrinase gene (CEL1) from Ruminococcus flavefaciens, the cellobiohydrolase gene (CBH1) from Phanerochaeta chrysosporium and the cellobiase gene (BGL1) from Saccharomycopsis fibuligera in a cassette that was introduced in S. cerevisiae enabled the yeast to degrade glucans efficiently. REFERENCES Riou et al. (1999) Salmon and Barre (1998) Fogel et al. (1983) Henderson et al. (1985) Lourens (1992) Laing and Pretorius (1993) van Rensburg et al. (1997, 1998) (Continued) 23

10 TABLE 3 Results recently achieved by using the recombinant DNA technology in the improvement of wine strains (follow the previous page) OBJECTIVE STRATEGY REFERENCES Improvement of wine flavour and other sensory qualities Release of grape terpenoids Expression of the β-glucosidase gene (BGL1) from Saccharomycopsis fibuligera in S. cerevisiae to increase wine aroma. Expression of β-1,4-glucanase gene from Trichoderma longibranchiatum in wine yeast to increase aroma intensity of wine. Overexpression of the exo-β-1,3 glucanase gene from S. cerevisiae and introduction in the same yeast of the endo-β-1,4-glucanase gene (END1) from Butyrivibrio fibrisolvens, the endo-β-1,3-1,4-glucanase gene (BEG1) from Bacillus subtilis, and the α-arabinofuranosidase gene (ABF2) to enhance the varietal character of Muscat wines. van Rensburg et al. (1997, 1998) Pérez-Gonzáles et al. (1993) Crous et al. (1996) Enhanced production of desirable volatile esters Overexpression of the acetyltransferase gene (ATF1), fundamental in the synthesis of esters to increase level of ethyl-acetate, iso-amyl acetate and ß-phenylethyl acetate, to improve the fruity aromatic character in wine and wine distillates. Lilly et al. (2000) Enhanced glycerol production Overexpression of GPD1 gene (cytosolic glycerol-3-phospate dehydrogenase), the key limiting enzyme of glycerol formation, constitutive expression of FPS1, encoding for a channel protein that act as a glycerol transporter facilitator, to increase glycerol production, and disruption of ALD6 and ALD7, genes encoding for acetaldehyde dehydrogenase to avoid the excessive concentration of acetate formed to balance the overproduction of glycerol. Remize et al. (1999) (Continued) 24

11 TABLE 3 Results recently achieved by using the recombinant DNA technology in the improvement of wine strains (follow the previous page) OBJECTIVE STRATEGY REFERENCES Bio-adjustment of wine acidity Reduction of acidity for wines produced in temperate climates Co-expression of the malolactic gene (mles) from Lactococcus lactis and of the malate permease gene (mae1) from Schizosaccharomyces pombe in S. cerevisiae to induce malo-lactic fermentation. Co-expression of the malic enzyme gene (mae2) from S. pombe and malate permease gene (mae1) from S. pombe in S. cerevisiae to induce maloethanolic fermentation. Bony et al. (1997) Volschenk et al. (1997) Acidification of high ph wines produced in warmer climates Construction of a S. cerevisiae strain containing the lacticodehydrogenase gene (LDH) from Lactobacillus casei expressed under the control of the yeast alcohol dehydrogenase gene to promote the conversion of glucose to lactic acid. Dequin et al. (1999) Wine yeasts producing antimicrobial peptides Expression of two bacteriocins in S. cerevisiae; one encoding a pediocin (peda) from Pediococcus acidilactici and one encoding a leucocin (lcab) from Leuconostoc carnosum to construct a bactericidal yeast strain. Schoeman et al. (1999) 25

12 hybrids obtained by crossing strains of different species are difficult to predict; nevertheless this is one of the ways by which novel traits can be introduced into wine strains. Moreover interspecific hybrids are usually stable and more vigorous than the parental strains (Naumov et al., 1992). Interspecific hybrids were constructed by crossing S. cerevisiae wine strains with S. uvarum strains. These hybrids produced wines with fermentation by-products at concentrations that were midway between those obtained using the parental strains (Zambonelli et al., 1993; 1997; Kishimoto, 1994). The basic traits of the S. uvarum strains, such as the ability to synthesise malic acid and to produce low amounts of acetic acid, remained present in the hybrids, but to a lesser degree. These hybrids have been successfully used to replace S. uvarum strains that produced wine with excessive acidity or unacceptably high concentrations of glycerol and β-phenylethanol (Zambonelli et al., 1997; Caridi et al., 1997). Interspecific hybrids were also constructed by crossing S. cerevisiae strains isolated from whey with S. uvarum strains. The cultures obtained possessed a combination of all the desirable oenological characteristics coming from both parental strains. Furthermore, all the traits that made the parental strains unsuitable for winemaking were apparently absent. These hybrids maintained the most useful trait of the whey strain of S. cerevisiae: the considerable capacity for reducing malic acid; and one of the most useful traits of S. uvarum: the production of very low concentrations of acetic acid (Rainieri et al., 1998a, 1999b). The genetic improvement of yeast strains through interspecific hybridisation cannot be taken further than the first generation; the hybrid is in fact sterile. Moreover the combination of the parental traits is not programmable, and the combination of desirable traits is often the result of chance. GENE CLONING AND TRANSFORMATION The traditional techniques used in the genetic improvement of yeasts have been fundamental in providing strains with novel and useful characteristics for winemaking. Nevertheless, they lack the specificity required to construct a strain with an exact combination of characteristics. Though carefully planned, mutation, hybridisation or spheroplast fusion are likely to produce wine yeast strains improved for the desired characteristic, but also, in most cases, presenting unwanted traits as a consequence of the introduced modification. The use of recombinant DNA technology can overcome this limitation. By cloning and inserting a specific gene into a wine strain, it is possible to introduce or modify one or more traits without altering the existing characteristics of the yeast (Pretorius, 1999; 2000). The basic steps of gene cloning and transformation are: (i) identification of a target gene; isolation of the corresponding DNA fragment; (ii) identification and linearisation, using specific restriction enzymes, of a suitable plasmid vector; (iii) joining of the DNA fragment containing the target gene to the linearised plasmid generating recombinant DNA molecules; (iv) insertion of the recombinant DNA molecules into host cells by transformation; and (v) screening of the transformed cells and selection of those containing the target gene. The applications of recombinant DNA techniques can go far beyond introducing specific genes into yeasts: (i) The phenotypic effect produced by a gene can be increased to a desired level by producing numerous copies of the genes (ampli- 26

13 fication of gene expression). (ii) The synthesis of enzymes can be directed to be released from a particular metabolic control or it can be diverted to a new one. (iii) The secretion of a specific gene product into the medium can be obtained by in-frame splicing of a structural gene to a secretion signal. (iv) Gene products with modified characteristics can be obtained by site-directed mutagenesis. (v) Specific undesirable characteristics can be eliminated by gene disruption. (vi) Genetic information from organisms such as fungi, bacteria, animals and plants can be incorporated into yeasts. Some examples of the results achieved applying the recombinant DNA technology to the improvement of wine strains are given in Table 3. CONCLUSIONS The availability of traditional genetic techniques and especially the development of more sophisticated molecular methods have provided or made feasible the construction of yeast strains with the specific characteristics required by the fermentation industry. However, yeast strain selection remains an important and fundamental step in any yeast improvement programme. Through the selection and the oenological characterisation of yeast strains, it is in fact possible to identify innovative and important oenological traits that can then be introduced into wine strains. Improvement programmes are now directed towards the construction of strains possessing innovative, useful properties for winemaking, while guaranteeing the preservation of the genetic biodiversity typical of every wine producing areas (Pretorius et al., 1999). The application of recombinant DNA technology has so far provided the most promising results in the improvement of wine yeasts. These results have confirmed that it is now possible to produce wine yeasts designed for any specific purpose, preserving the tradition of winemaking. REFERENCES Barre P., Vézinhet F., Dequin S., Blondin B. (1993). Genetic improvement of wine yeast. In: Fleet G.H., ed., Wine Microbiology and Biotechnology, Harwood Academic Publishers, Singapore, pp Bertolini L., Zambonelli C., Giudici P., Castellari L. (1996). Higher alcohol production by cryotolerant Saccharomyces strains. Am. J. Enol. Vitic., 47: Bony M., Bidart F., Camarasa C., Ansanay V., Dulau L., Barre P., Dequin S. (1997). Metabolic analysis of S. cerevisiae strains engineered for malolactic fermentation. FEBS Lett., 410: Caridi A., Rainieri S., Passarelli P., Zambonelli C. (1997). Effects of hybrids between cryo and non-cryotolerant Saccharomyces strains on the composition of wines from southern Italy. Vignevini, 24: Castellari L., Pacchioli G., Zambonelli C., Tini V., Grazia L. (1992). Isolation and initial characterisation of cryotolerant Saccharomyces strains. Ital. J. Food Sci., 3: Castellari L., Ferruzzi M., Magrini A., Giudici P., Passarelli P., Zambonelli C. (1994). 27

14 Unbalanced wine fermentation by cryotolerant vs. non-cryotolerant Saccharomyces strains. Vitis, 33: Constantí M., Poblet M., Arola L., Mas A., Guillamón J.M. (1997). Analysis of yeast populations during alcoholic fermentation in a newly established winery. Am. J. Enol. Vitic., 48: Crous J.M., Pretorius I.S., van Zyl W.H. (1996). Cloning and expression of the α-l-arabinofuranosidase gene (ABF2) of Aspergillus niger in Saccharomyces cerevisiae. Appl. Microbiol. Biotechnol., 46: Dequin S., Baptista E., Barre P. (1999). Acidification of grape musts by Saccharomyces cerevisiae wine yeast strains genetically engineered to produce lactic acid. Am. J. Enol. Vitic., 50: Demain A.L., Solomon N.A. (1981). Industrial microbiology: introducing an issue of how products useful to man are manufactured by microorganisms. Sci. Am., 245: Dittrich H.H. (1995). Wine and Brandy. In: Rehm H.J., Reed G., Pühler A., Stadler P., Vol. 9 (Reed G., Nagodawithana T.W., eds, Biotechnology, Enzymes, Biomass, Food and Feed), VCH, Weinheim, pp Eschenbruch R., Cresswell K.J., Fisher B.M., Thornton R.J. (1982). Selective hybridisation of pure culture wine yeasts. 1. Elimination of undesirable winemaking properties. Europ. J. Appl. Microbiol. Biotechnol., 14: Fleet G.H. (1993). The microorganisms of winemaking Isolation, enumeration and identification. In: Fleet G.H., ed., Wine Microbiology and Biotechnology, Harwood Academic Publishers, Singapore, pp Fleet G.H., Heard G.M. (1993). Yeasts Growth during fermentation. In: Fleet G.H., ed., Wine Microbiology and Biotechnology. Harwood Academic Publishers, Singapore, pp Fleet G.H. (1997). Wine. In: Doyle M.P., Beuchat L.R., Montville T., eds., Food Microbiology Fundamentals and Frontiers, American Society for Microbiology, Washington DC, pp Fogel S., Welch J.W., Cathala G., Karin M. (1983). Gene amplification in yeast: CUP1 copy number regulates copper resistance. Curr. Genet., 7: Frezier V., Dubourdieu D. (1992). Ecology of yeast strain Saccharomyces cerevisiae during spontanteous fermentation in a Bordeaux winery. Am. J. Enol. Vitic., 43: Garcia A., Dalau L., Samson A., Aguera E., Agosin E., Günata Z. (2000). Influence of a mixed culture of Debaryomyces vanriji and Saccharomyces cerevisiae on the volatiles of a Muscat wine. Appl. Environ. Microbiol. (submitted). Goldfarb A. (1994). Wild or natural, native yeasts have a role in modern winemaking. Wines and vines., 75: Heard G.M. (1988). Occurrence and growth of yeasts during the fermentation of Australian wines. PhD Thesis, University of New South Wales. Heard G.M. (1999). Novel yeasts in winemaking looking to the future. Food Australia, 51: Henderson R.C.A., Cox B.S., Tubb R. (1985). The transformation of brewing yeasts with a plasmid containing the gene for copper restistance. Curr. Genet., 9: Henschke P.A. (1997). Wine yeasts. In: Zimmermann F.K., Entian K.D., eds, Yeast Sugar Metabolism, Technomic Publishing Co., Lancaster, Pennsylvania, pp Kishimoto M., Shinohara T., Soma E., Goto S. (1993). Identification and enological characteristics of cryophilic wine yeasts. J. Brew. Soc. Japan, 88:

15 Kishimoto M. (1994). Fermentation characteristics of hybrids between the cryophilic wine yeast Saccharomyces bayanus and the mesophilic wine yeast Saccharomyces cerevisiae. J. Ferment. Bioeng., 77: Kunkee R.E., Amerine A.M. (1970). Yeast in winemaking. In: Harrison J.S., Rose A.H., eds, The Yeast. Vol. 3, Academic Press, New York, pp Laing E., and Pretorius I.S. (1993). Co-expression of an Erwinia chrysanthemi pectate lyase-encoding gene (pele) and an Erwinia carotovora polygalacturonase-encoding gene (peh1) in Saccharomyces cerevisiae. Appl. Microbiol. Biotechnol., 39: Lilly M., Lambrechts M.G., Pretorius I.S. (2000). The effect of increased yeast alcohol acetyltransferase activity on the sensorial quality of wine and brandy. Appl. Environ. Microbiol., 66: Lourens K. (1992). Genetic engineering of Saccharomyces cerevisiae for efficient expression and secretion of proteases. MSc Thesis, University of Stellenbosch. Mortimer R.K., Romano P., Suzzi G., Polsinelli M. (1994). Genome renewal: a new phenomenon revealed from a genetic study of 43 strains of Saccharomyces cerevisiae derived from natural fermentation of grape musts. Yeast, 10: Naumov G.I., Naumova E., Korhola M. (1992). Genetic identification of natural Saccharomyces sensu stricto yeasts from Finland, Holland and Slovakia. Antonie van Leeuwenhoek, 61: Pérez-González J.A., González R., Querol A., Sendra J., Ramon D. (1993). Construction of a recombinant wine yeast expressing β-1,4-endoglucanase and its use in microvinification processes. Appl. Environ. Microbiol., 59: Pretorius I.S. (1999). Engineering designer genes for wine yeasts. Aust. N.Z. Wine Indust. J., 14: Pretorius I.S. (2000). Tailoring wine yeast for the new millennium: novel approaches to the ancient art of winemaking. Yeast, 16: (in press). Pretorius I.S., van der Westhuizen T.J. (1991). The impact of yeast genetics and recombinant DNA technology on the wine industry a review. S. Afr. J. Enol. Vitic., 12: Pretorius I.S., van der Westhuizen T.J., Augustyn O.P.H. (1999). Yeast biodiversity in vineyards and wineries and its importance to the South African wine industry. A review. S. Afr. J. Enol. Vitic., 20: Price E. (1996). Natural yeast fermentations Amberley Estate. In: Allen M., Leske P., Baldwin G., eds, Proceedings of the ASVO Oenology seminar, Advances in Juice Clarification and Yeast Inoculation. Melbourne, Australia, Australian Society of Viticulture and Enology, Australia, pp Querol A., Borrio E., Ramon D. (1994). Population dynamics of natural Saccharomyces strains during wine fermentation. Int. J. Food Microbiol., 21: Rainieri S., Zambonelli C., Passarelli P., Castellari L., Franzoni, S. (1996). Initial characterisation of thermotolerant strains of Saccharomyces sensu stricto. Ind. Bevande, 25: Rainieri S., Zambonelli C., Giudici P., Castellari L. (1998a). Characterisation of thermotolerant Saccharomyces cerevisiae hybrids. Biotechnol. Lett., 20: Rainieri S., Zambonelli C., Tini V., Castellari L., Giudici P. (1998b). The enological traits of thermotolerant Saccharomyces strains. Am. J. Enol. Vitic., 49: Rainieri S., Zambonelli C., Hallsworth J.E., Pulvirenti A., Giudici P. (1999a). Saccharomyces uvarum, a distinct group within Saccharomyces sensu stricto. FEMS Microbiol. Lett., 177: Rainieri S., Zambonelli C., Tini V., Castellari L., Giudici P. (1999b). Oenological properties of an interspecific Saccharomyces hybrid. S. Afr. J. Enol. Vitic., 20:

16 Ramey D. (1995). Low input winemaking let nature do the work. In: Stockley C.S., Sas A.N., Johnstone R.S., Lee T.H., eds, Proceedings of the Ninth Australian Wine Industry Technical Conference, Adelaide, Australia. Winetitles, Adelaide, pp Remize F., Roustan J.L., Sblayrolles J.M., Barre P., Dequin S. (1999). Glycerol overproduction by engineered Saccharomyces cerevisiae wine strains leads to substantial changes in by-product formation and to a stimulation of fermentation rate in stationary phase. Appl. Environ. Microbiol., 65: Riou C., Luyten K., Blondin B. (1999). Hexose transporter (HXT) gene expression in Saccharomcyces cerevisiae during wine fermentation. XIX International Conference on Yeast Genetics and Molecular Biology, Rimini, Italy, May Curr. Genet., 35: 314. Romano P., Soli M.G., Suzzi G., Grazia L., Zambonelli C. (1985). Improvement of a wine Saccharomyces cerevisiae strain by a breeding program. Appl. Environ. Microbiol., 50: Romano P. (1997). Metabolic characteristics of wine strains during spontaneous and inoculated fermentation. Food Tech. Biotech., 35: Salmon J.M., Barre P. (1998). Improvement of nitrogen assimilation and fermentation kinetics under enological conditions by derepression of alternative nitrogen-assimilatory pathways in an industrial Saccharomyces cerevisiae strain. Appl. Environ. Microbiol., 64: Schoeman H., Vivier M.A., du Toit M., Dicks L.M.T., Pretorius I.S. (1999). The development of bactericidal yeast strains by expressing the Pediococcus acidilactici pediocin gene (peda) in Saccharomyces cerevisiae. Yeast, 15: Schütz M., Gafner J. (1994). Dynamics of the yeast strain population during spontaneous alcoholic fermentation determined by CHEF gel electrophoresis. Lett. Appl. Microbiol., 19: Soden A., Francis I.L., Oakey H., Henschke P.A. (2000). The effect of cofermentation with Candida stellata and Saccharomyces cerevisiae on the aroma and composition of Chardonnay wine. Aust. J. Grape Wine Res., 6: (In press). Steinkraus K.H. (1996). Handbook of Indigenous Fermented foods. Marcel Dekker, Inc., New York. Stewart G.G. (1981). The genetic manipulation of industrial yeast strains. Can. J. Microbiol., 27: Thornton R.J. (1982). Selective hybridisation of pure culture wine yeasts. Eur. J. Appl. Microbiol. Biotechnol., 14: Thornton R.J. (1983). New yeast strains from old the application of genetics to wine yeasts. Food Technol. Australia, 35: Thornton R.J. (1985). The introduction of flocculation into a homothallic wine yeast. A practical example of the modification of winemaking properties by the use of genetic techniques. Am. J. Enol.Vitic., 36: van Rensburg P., van Zyl W.H., Pretorius I.S. (1997). Over-expression of the Saccharomyces cerevisiae exo-β-1,3-glucanase gene together with the Bacillus subtilis endo-β-1,3-1,4-glucanase gene and the Butyrivibrio fibrisolvens endo-ß-1,4-glucanase gene in yeast. J. Biotechnol., 55: van Rensburg P., van Zyl W.H., Pretorius I.S. (1998). Engineering yeast for efficient cellulose degradation. Yeast, 14: Versavaud A., Courcoux P., Roulland C., Dalau L., Hallet J.-N. (1995). Genetic diversity and geographical distribution of wild Saccharomyces cerevisiae strains from wineproducing area of Chanrentes, France. Appl. Environ. Microbiol., 61: Volschenk H., Viljoen M., Grobler J., Petzold B., Bauer F.F., Subden R.E., Young R.A., Lonvaud A., Denayrolles M., van Vuuren H.J.J. (1997). Engineering pathways for malate degradation in Saccharomyces cerevisiae. Nature Biotechnol., 15:

17 Winge O., Laustsen O. (1938). Artificial species hybridization in yeast. C.R. Trav. Lab. Carlsberg, sér physiol., 22: Zambonelli C., Passarelli P., Rainieri S., Giudici P. (1993). Taxonomic and technological implication of sterility in hybrids from cryotolerant and non cryotolerant Saccharomyces strains. Ann. Microbiol. Enzimol., 43: Zambonelli C., Passarelli P., Rainieri S., Bertolini L., Giudici P., Castellari L. (1997). Technological properties and temperature response of interspecific Saccharomyces hybrids. J. Sci. Food Agric., 74: Zambonelli C. (1998). Microbiologia e biotecnologia dei vini. Edagricole, Bologna. 31