Feature

Tailoring wine yeast for the third millennium: novel approaches to the ancient art of winemaking

Isak S. Pretorius

Institute for Wine Biotechnology, University of Stellenbosch, Stellenbosch, ZA-7600, South Africa

1. Winemaking is an ancient art
The history of winemaking parallels that of civilisation: historians believe that wine was being made in the Caucasus and Mesopotamia as early as 6000 BC. References to wine have been found in Egypt and Phoenicia dating as far back as 5000 BC, and by 2000 BC wine was being produced in Greece and Crete. Colonising by the Romans spread winemaking all around the Mediterranean; by 500 BC wine was being produced in Sicily, Italy, France, Spain, Portugal and in northern Africa. Cultivation of the vine also spread into the Balkan States, and the Romans took it into Germany and other parts of northern Europe, eventually reaching as far as Britain. 

European explorers in the 16th century introduced the vine into the New World. In 1530 the Spanish conquistadors planted Vitis vinifera in Mexico, Argentina, Peru and Chile. In 1655 Dutch settlers in South Africa planted French vine cuttings on the lower slopes of the Cape of Good Hope's majestic Table Mountain. Planting in California followed soon thereafter, and in Australia and New Zealand more than a century later in 1813. 
Disaster struck the world of wine in the 1870's, when the root-eating insect Phyloxera vastatrix threatened almost every vine in Europe and the New World. They were pulled up and replaced with new V. vinifera vines grafted onto phyloxera-resistant rootstocks from the native American vine. Despite attacks by phyloxera and the spread of other diseases, such as downy mildew (Plasmopara viticola) and powdery mildew (Oidium tuckerii), the Office International de la Vigne et du Vin in Paris reports that today there are some 8 million hectares of vineyards across the world, mainly concentrated within the earth's temperate zones. Each of these vineyards reflects the terroir, history, culture and traditions of its region.

Until the early years of the 17th century wine was considered to be the only wholesome readily storable (to a point) beverage, accounting for the rapid global increase of wine fermentation technology. Today wine is synonymous with culture and a convivial lifestyle around the world, complementing food, entertainment and the arts. Wine plays a major role in the economies of many nations, which produce more than 26 billion litres of wine annually. Modern winemakers supply a wide variety of wines year round independent of location and time of consumption. Fierce competition for market share has led to increased diversity and innovation within the wine industry, much to the benefit of the consumer. 

A look at the early days of winemaking makes it obvious that while different techniques produced varied styles of wine, the basic principles changed very little. During the last 150 years or so, however, the scientific basis of winemaking has gradually become clearer, and many practices once thought impossible have now become routine. 

In 1863 Louis Pasteur revealed for the first time the hidden world of microbial activity during wine fermentation. He proved conclusively that yeast is the primary catalyst in wine fermentation, basing his work upon Antonie van Leeuwenhoek's first microscopic observation in 1680 of yeast cells and the claims by three other independent pioneers, Cagniard-Latour, Kützing and Schwann, in the late 1830's, that these cells are living organisms. With the knowledge that yeast was responsible for the biotransformation of grape sugars (mainly glucose and fructose) into alcohol and carbon dioxide, winemakers could control the process from the vineyard to bottling plant. Pasteur also made it possible to destroy unwanted souring bacteria. As a result, the quality and quantity of wine production was vastly improved.

These fundamental innovations in winemaking practices revolutionised the wine industry, and today the forces of market-pull and technology-push continue to challenge the tension between tradition and innovation. There will continue to be further improvements in winemaking by refining viticultural and oenological practices. These factors will remain important to the improvement of the overall quality and endless variety of wine. 
But today there is a new, and for the moment controversial, focal point for innovation in winemaking - the genetic modification of the two main organisms involved, the grape cultivar and wine yeast. The diversity of yeast species associated with winemaking, the tailoring of wine yeast and the possible use of strains expressing novel designer genes make possible exciting new approaches to winemaking in the 21st century.

2. Yeast diversity associated with winemaking
2.1 Composition and dynamics of yeast populations during spontaneous wine fermentations

The fermentation of grape juice into wine is a complex microbiological process involving progressive development and interactions among yeasts, bacteria and filamentous fungi. Of these microbes, yeasts play the central role.

Traditionally, yeasts associated with grapes and cellar equipment were simply allowed to ferment the sugars to ethanol, carbon dioxide and other minor, but important, metabolites. Yeasts of the genera Kloeckera and Hanseniaspora are the predominant species on the surface of the grape, accounting for about 50 to 75% of the total yeast population. Together with Candida, they dominate the early stages of spontaneous fermentations, followed by several species of Cryptococcus, Kluyveromyces, Metschnikowia and Pichia (including those species that were previously assigned to the genus Hansenula) in the middle stages when the ethanol level rises to 3-4%. The last stages of natural wine fermentations are invariably dominated by the alcohol-tolerant strains of Saccharomyces. S. cerevisiae is universally known as the wine yeast; it is ultimately responsible for the alcoholic fermentation. However, other non-Saccharomyces yeasts, such as species of Brettanomyces, Schizosaccharomyces, Torulaspora and Zygosaccharomyces, also may be present during the fermentation and can occur in the resultant wine.

The diverse and dynamic activities of these yeasts impact on many aspects of winemaking. Some species play beneficial roles in the production of wine; others act detrimentally as spoilage organisms. The individual and collective contribution to wine of these indigenous yeasts vary according to their numbers and diversity of species present in must and wine. The combined effect of several factors affects the composition of the indigenous yeast population in spontaneous wine fermentations, thereby contributing to the considerable variation in the quality and organoleptic characteristics of spontaneously fermented wines from one year to another. These factors include the grape variety; temperature, rainfall and other climatic influences; soil, fertilization, irrigation and viticultural practices (e.g., vine canopy management); development stage at which grapes are examined; physical damage caused by mould, insects and birds; and fungicides applied to vineyards. It is also important to note that harvesting equipment, including mechanical harvesters, picking baskets and other infrequently cleaned delivery containers could also represent sites for yeast accumulation and microbiological activity before grapes reach the winery. This becomes more important as travel time to the winery increases. The microbiota of grape must are affected indirectly by all these factors influencing the indigenous grape microflora and the winery flora. Added to these factors is must pretreatment (e.g., cellar hygiene, aeration, enzyme treatment, sulphite addition, clarification method, temperature, inoculation with yeast starter cultures). 

2.2 Spontaneous versus inoculated fermentations

Notwithstanding the fact that spontaneous fermentations usually take longer than most winemakers are willing to admit and that the outcome is highly unpredictable, many boutique wineries that depend on vintage variability accept the potentially staggering risks involved to achieve stylistic distinction. The indigenous yeasts present in spontaneous wine fermentations are believed to produced higher concentrations of glycerol and other polyols, thereby yielding wines with a distinct sensorial quality often described as a fuller, rounder palate structure. Another contributing factor to this characteristic style of wine is the fact that the extended lag phase before the onset of vigorous fermentation allows for the reaction of oxygen with anthocyanins and other phenols in the absence of ethanol, which is thought to enhance color stability in red wines as well as accelerating phenol polymerization.

In modern, large-scale wineries however, where rapid and reliable fermentations are essential for consistent wine flavor and predictable quality, the rule, rather than the exception, is the use of specially selected starter cultures of S. cerevisiae of known ability. The pure cultures, which are inoculated in grape must following SO2 applications to suppress natural microflora, reduce lag phases and complete sugar conversion to alcohol much faster than un-inoculated fermentations. As a result of this practice, the wine reproducibility exhibits, in successive vintages, sensory properties typical of the wines produced in the region of origin.

It should be noted that the use of S. cerevisiae starter cultures may not necessarily prevent the growth and metabolic activity of indigenous, winery-associated S. cerevisiae strains or other natural non-Saccharomyces yeasts. It is generally accepted that both spontaneous and inoculated wine fermentations are affected by the diversity of yeasts associated with the vineyard and winery. Since the wealth of this yeast biodiversity in the various wine-producing regions of the world is still largely untapped for its oenological potential, the possibility exists that in future some winemakers might prefer to use mixtures of S. cerevisiae strains and non-Saccharomyces species as starter cultures. Such mixed starter cultures might even be tailored to reflect the yeast biodiversity of a given region.

3. Active dried wine yeast starter cultures
3.1 The need for new wine yeast strains

The concept of inoculating wine fermentations with pure yeast starter cultures dates back to 1890 when Müller-Thurgau convinced some German winemakers of the benefits of the concomitant rapid and even rates of fermentation. About 75 years later, the first two commercial active dried wine yeast strains were produced for a Californian winery, and these two strains were offered worldwide as all-purpose yeasts. Their success was limited; the need soon became evident for separate S. cerevisiae strains with specific characteristics for different types and styles of wine. In addition to the primary role of wine yeast to catalyze the efficient and complete conversion of grape sugars to alcohol without the development of off-flavors, today's winemakers demand starter culture strains with a whole range of specialized properties (Table 1), their importance differing with the type and style of wine to be made and the technical requirements of the winery. There is a growing need for S. cerevisiae strains better adapted to various wine-grape varietals, viticultural practices and winemaking techniques.

Table 1. Desirable characteristics of wine yeast.

This quest for wine yeast strains that are optimized for specific tasks set by winemakers has engendered the development of rapid molecular methods for wine yeast strain differentiation (Table 2) and dedicated breeding programs. This in turn, has spotlighted the genetic make-up of wine yeasts, genetic techniques for strain development (Table 3), and the possible use of genetically modified active dried wine yeast cultures.

Table 2. Molecular methods for wine yeast strain differentiation.

3.2 Genetic constitution of wine yeasts

The yeast nucleus is a round-lobate organelle of about 1.5 µm diameter and the nucleoplasm is separated from the cytoplasm by a double membrane containing pores 20-100 nm in diameter. The nuclear membranes are occasionally contiguous with the endoplasmic reticulum and, unlike most eukaryotic cells, the yeast nuclear membrane is not dismantled during mitosis. The nucleoplasm contains chromatin that consists of condensed basic nucleoprotein material, comprising double-helical DNA-histone complexes, which are organized in chromosomes. 

Most laboratory-bred strains of S. cerevisiae are either haploid or diploid, whereas industrial wine yeast strains are predominantly diploid or aneuploid and occasionally polyploid. In general, S. cerevisiae is considered as having a relatively small genome, a large number of chromosomes, little repetitive DNA and few introns. Besides the chromosomal DNA, several non-Mendelian genetic elements are known to exist in the nucleus (e.g., 35 to 55 copies of Ty retrotransposons per haploid genome and 50 to 100 copies of the 6.3-kb 2mm plasmid DNA), mitochondria (75-kb mtDNA) and cytoplasm (e.g., killer viral-like particles containing dsRNA genomes, and prion-like elements such as [Psi], Eta] and [URE2]). However, with the exception of the mtDNA, there seems to be no effect on the quality of wine produced by S. cerevisiae strains containing any of these non-chromosomal genetic elements.

Haploid strains contain approximately 12 to 13 megabases (mb) of nuclear DNA, distributed along 16 linear chromosomes. Each chromosome is a single DNA molecule approximately 200 to 2200 kb long. The genome of a laboratory strain of S. cerevisiae has been completely sequenced and found to contain roughly 6000 protein-encoding genes, 275 tRNA genes, 140 rRNA genes and 20 genes encoding small nuclear RNA (snRNA). Almost 70% of the more than 12 million bp consists of open reading frames (ORFs), indicating that there is a protein-encoding gene about every 2 kb. Therefore, the S. cerevisiae genome, which is relatively rich in guanine and cytosine content (%G+C of 39-41) is much more compact even when compared with the genomes of other yeasts and fungi. Of its 6000 protein-encoding genes, only 4% are interrupted by non-coding intervening sequences. 

Chromosomal DNA of S. cerevisiae contains relatively few repeated sequences and, with the exception of the tRNA and rRNA genes, most genes appear to be present as single copies in the haploid genome. Less than half the genes are currently classified as 'functionally characterized'. However, technology has been developed to provide a direct link between the genome sequence and the transcriptome (a complete set of mRNA that yeast is capable of synthesizing). The genomic sequence has been used to design and synthesize high-density oligonucleotide arrays for monitoring the expression levels of nearly all genes of yeast cells grown under a large number of different cultural conditions. Furthermore, several laboratories are systematically screening disrupted genes in deletion (knock-out) mutants to functionally analyze the yeast genome. The deciphering of the function of the 6000 genes will make the complete proteome (the full set of proteins that a yeast is capable of synthesizing) accessible. This information about the genome, transcriptome and proteome of a laboratory strain of S. cerevisiae can then be gradually expanded to the much more complex genomes of industrial wine yeast strains. 

3.3 Genetic techniques for the analysis and development of wine yeasts strains

S. cerevisiae can be genetically manipulated in many ways. Some techniques alter limited regions of the genome; others are used to recombine or rearrange the entire genome. Techniques having the greatest potential in genetic programming of wine yeast strains are: clonal selection of variants, mutation and selection, hybridization, rare-mating, spheroplast fusion as well as gene cloning and transformation (Table 3). The combined use of classical genetic techniques and recombinant DNA methods have dramatically increased the genetic diversity that can be introduced into yeast cells.

Selection of variants is a simple direct means of strain development that depends on the genetic variation normally present in all wine yeast strains. Genetic heterogeneity in wine yeast strains is due mainly to mitotic recombination during vegetative growth and spontaneous mutation. Successful isolation of variants depends on the frequency at which they occur and the availability of selection procedures to isolate strains containing the improved characteristic. Dramatic improvements in most characteristics cannot be expected; nevertheless intra-strain selection has been used for decades to improve wine yeast strains.

Table 3. Some methods employed in genetic research and development of wine yeasts.

The average spontaneous mutation frequency in S. cerevisiae at any particular locus is approximately 10-6 per generation. The use of mutagens greatly increases the frequency of mutations in a wine yeast population. Mutation and selection appear to be a rational approach to strain development when a large number of performance parameters are to be kept constant while only one is to be changed. However, mutation of wine yeasts can lead to improvement of certain traits with the simultaneous debilitation of other characteristics. Although mutations are probably induced with the same frequency in haploids, diploids or polyploids, they are not as easily detected in diploid and polyploid cells because of the presence of non-mutated alleles. Only if the mutation is dominant is a phenotypic effect detected without the need for additional alterations. Therefore, haploid strains of wine yeasts are preferred, though not essential, when inducing mutations. Successful mutation breeding is usually associated with mutations in meiotic segregants, where the two mating parents of a genetically stable hybrid provide a good basis for the introduction of recessive mutations. Mutagenesis has the potential to disrupt or eliminate undesirable characteristics and to enhance favorable properties of wine yeasts. Though the use of mutagens for directed strain development is limited, the method could be applied to isolate new variants of wine yeast strains prior to further genetic manipulation.

Intra-species hybridization involves the mating of haploids of opposite mating-types to yield a heterozygous diploid. Recombinant progeny are recovered by sporulating the diploid, recovering individual haploid ascospores and repeating the mating/sporulation cycle as required. Haploid strains from different parental diploids possessing different genotypes can be mated to form a diploid strain with properties different from that of either parental strain. Thus, theoretically speaking, crossbreeding can permit the selection of desirable characteristics and the elimination of undesirable ones. Unfortunately many wine yeasts are homothallic and the use of hybridization techniques for development of wine yeast strains has proved difficult. This problem can be circumvented, however, by direct spore-cell mating, where four homothallic ascospores from the same ascus are placed into direct contact with heterothallic haploid cells using a micromanipulator. Mating takes place between compatible ascospores and cells. Elimination or inclusion of a specific property can thus be achieved relatively quickly by hybridization, provided that it has a simple genetic basis, for example one or two genes. However, many desirable wine yeast characteristics are specified by several genes or are the result of several gene systems interacting with one another.
Wine yeast strains that fail to express a mating-type can be force-mated (rare-mating) with haploid MATa and MATa strains. Typically, a large number of cells of the parental strains are mixed and a strong positive selection procedure is applied to obtain the rare hybrids formed. For instance, industrial strains with a defective form or lack of mtDNA (respiratory-deficient mutants) can be force-mated with auxotrophic haploid strains having normal respiratory characteristics. Mixing of these non-mating strains at high cell density will generate only a few respiratory-sufficient prototrophs. These true hybrids with fused nuclei can then be induced to sporulate for further genetic analysis and crossbreeding. Rare-mating is also used to introduce cytoplasmic genetic elements into wine yeasts without the transfer of nuclear genes from the non-wine yeast parent. This method of strain development is termed cytoduction. Cytoductants (or heteroplasmons) receive cytoplasmic contributions from both parents but retain the nuclear integrity of only one. Cytoduction requires a haploid mating strain carrying the kar1 mutation; that is, a mutation that impedes karyogamy (nuclear fusion) after mating. 

This more specific form of strain construction can, for example, be used to introduce the dsRNA determinants for the K2 zymocin and associated immunity into a particular wine yeast. Cytoduction can also be used to substitute the mitochondrial genome of a wine yeast or to introduce a plasmid encoding desirable genetic characteristics into specific wine yeast strains. Mating between strains, one of which carries the kar1 allele, occasionally generates progeny that contain the nuclear genotype of one parent together with an additional chromosome from the other parent. The donation of a single chromosome from an industrial strain to a haploid kar1 recipient is termed single-chromosome transfer, and is used to examine individual chromosomes of industrial yeast strains in detail.

Spheroplast fusion is a direct, asexual technique that can be used in crossbreeding as a supplement to mating. Like rare-mating, spheroplast fusion can be used to produce either hybrids or cytoductants. Both these procedures overcome the requirement for opposite mating types to be crossed, thereby extending the number of crosses that can be done. Cell walls of yeasts can be removed by lytic enzymes in the presence of an osmotic stabilizer to prevent osmolysis of the resulting spheroplasts. Spheroplasts from the different parental strains are mixed in the presence of a fusion agent, polyethylene glycol (PEG) and calcium ions, and then allowed to regenerate their cell walls in an osmotically stabilized, selective agar medium. Spheroplast fusion of non-sporulating industrial yeast strains serves to remove the natural barriers to hybridization. The desirable (and undesirable) characteristics of both parental strains will recombine in the offspring. Cells of different levels of ploidy can be fused. For instance, a diploid wine yeast strain can be fused to a haploid strain to generate triploid strains. Alternatively, two diploid wine yeasts with complementing desirable characteristics can be fused to generate a tetraploid wine yeast strain containing all of the genetic backgrounds of the two parental wine yeasts.

While clonal selection, mutagenesis, hybridization, rare-mating and spheroplast fusion are valuable to strain development programmes, these methods lack the specificity required to modify wine yeasts in a well-controlled manner. It may not be possible, for example, to precisely define the change required, and a new strain may bring an improvement in some aspects, while compromizing other desired characteristics. Yeast geneticists must be able to alter the characteristics of wine yeasts in specific ways: an existing property must be modified, or a new one introduced without adversely affecting other desirable properties. Molecular-genetic techniques capable of this are now available. Gene cloning and recombinant DNA technology offer exciting prospects for improving wine yeasts.

Gene cloning and transformation is analogous to cutting a printed page in half, inserting a new paragraph in the middle and repeatedly photocopying the altered version to reproduce the new material along with the old. Genetic transformation is the change of the genetic set-up of a yeast cell by the introduction of purified DNA. By using such procedures it should be possible to construct new wine yeast strains differing from the original only in single specific characteristics. In principle, there are five major steps in the cloning of a gene: (i) identifying the target gene and obtaining the DNA fragment [obtained from a genomic or cDNA library or by amplification using the polymerase chain reaction (PCR)] to be cloned (passenger DNA) by enzymatic fragmentation of the donor DNA using restriction endonucleases; (ii) identifying and linearizing a suitable plasmid vector; (iii) joining the passenger DNA fragments to the linearized vector DNA, thereby generating recombinant DNA molecules, designated a gene library; (iv) inserting the recombinant DNA molecules into host cells by transformation; (v) screening transformed cells and selecting those cells containing the target gene.

A number of options are available at each of these stages and decisions depend on a number of factors, not the least of which is the extent of information available about the target gene product and the gene itself. Free DNA molecules, however, are not taken up by normal yeast cells; their entry requires the generation of a permeable spheroplast. DNA is added in the presence of calcium ions and polyethylene glycol that makes the plasma membrane permeable, encouraging the passage of DNA. The methods involving spheroplasts yield high transformation efficiencies; however, transformation is somewhat laborious and is associated with a high frequency of cell fusion. Also, different strains vary considerably in their transformation competence, that seems to be inherited in a polygenic manner. A simpler method has been developed using intact yeast cells and alkali cations, especially lithium acetate (or lithium sulfate) and polyethylene glycol to induce DNA uptake. Currently, the lithium method seems to be the most commonly used, despite its disadvantage of giving a slightly lower transformation efficiency than the spheroplast method. Yeast cells can also be transformed by electroporation and biolistic bombardment.

To become a heritable component of the yeast cell, the transforming DNA normally suffers one of two fates: either it is maintained as a self-replicating plasmid, physically separated from the endogenous yeast chromosomes, or it must integrate into a chromosome and thus be maintained by the functions of the chromosome. A wide range of E. coli-S. cerevisiae shuttle vectors (YIp, YEp, YRp, YCp, YTp, YLp/YAC, YXp, etc.) containing bacterial and yeast marker genes and origin of replication sequences were developed. The introduction of recombinant plasmids into a wine yeast strain requires either that the strain be made auxotrophic before transformation or that the plasmid used for transformation carry a marker that is selectable against a wild-type diploid or polyploid background. Positive selectable markers include: (i)the kanamycin-resistance gene; (ii) the gene encoding resistance to the antibiotic geneticin G418; (iii) the copper-resistance (CUP1) gene; (iv) hygromycin B-resistance; (v) resistance to chloramphenicol; (vi) methotrexate-resistance; (vii) resistance to the herbicide sulfometuron methyl (SMR1 gene); (viii) resistance to methylglyoxal; (ix) the L-canavanine-resistance (CAN1) gene; and (x) the ability to utilize melibiose. Recombinant plasmids with positive selectable markers, containing a particular target gene, are usually either integrated into a chromosome or maintained as a stable minichromosome in industrial yeast strains. Such minichromosomes should preferably be stripped of all non-relevant bacterial DNA sequences before transformation into industrial yeast strains.

In addition to the introduction of specific genes into wine yeasts, recombinant DNA approaches offer wider applicability. Some of the applications provided by recombinant-DNA techniques include: (i) amplification of gene expression by maintaining a gene on a multi-copy plasmid, integration of a gene at multiple sites within chromosomal DNA or splicing a structural gene to a highly efficient promoter sequence; (ii) releasing enzyme synthesis from a particular metabolic control or subjecting it to a new one; (iii) in-frame splicing of a structural gene to a secretion signal to engineer secretion of a particular gene product into the culture medium; (iv) developing gene products with modified characteristics by site directed mutagenesis; (v) eliminating specific undesirable strain characteristics by gene disruption; (vi) incorporation of genetic information from diverse organisms such as fungi, bacteria, animals and plants.

The genetic techniques of mutation, hybridization, cytoduction and transformation discussed in this section will most likely be used in combination for commercial wine yeast improvement. Procedures centered around DNA transformation have revolutionized strategies for strain modification, but it remains difficult to clone unidentified genes. Thus, mutation and selection will persist as an integral part of many breeding programmes. Furthermore, although recombinant DNA methods are the most precise way of introducing novel traits encoded by single genes into commercial wine yeast strains, hybridization remains the most effective method for improving and combining traits under polygenic control.

4. Targets for strain development

Due to technical difficulties and the fact that the requirements of the wine industry have not been defined in genetic terms, little genetic research on S. cerevisiae has been undertaken by this industry when compared to that by the baking and brewing industries. Historically, development of wine yeast strains has almost exclusively relied on strain selection. Changes in winemaking technology to improve the reliability of fermentation, wine quality and economics of production have placed new demands (Table 4) on the performance of selected wine yeast strains. 

The efficiency of fermentation would be markedly improved by better sugar utilization, increased tolerance to ethanol, resistance to zymocins and heavy metals, reduced formation of foam, induced flocculance at the end of fermentation and production of extracellular enzymes. Moreover, wine yeast strains with improved nitrogen and sulfur metabolism have the potential to reduce or eliminate the need for ammonium salts, and copper salts for the removal of H2S. Yeast strains secreting pectinases, glucanases, xylanases and proteases are being proposed to improve wine clarity and filtration efficiency. Secretion of glucanases and glucosidases may also enhance wine flavor by hydrolysis of flavor precursor glycosides. Over-expression of the yeast's own alcohol acetyltransferase has been shown to be the first step towards enhanced ester production, thereby adjusting the aroma profile of wine considerably. Furthermore, genetically modified wine yeasts have the potential to correct wine acidity by the metabolism of malic acid. Rapid autolytic characteristics of strains used for the production of sparkling wines can affect the time and cost of wine production. 

Table 4. Targets for, and aims of, strain development.

We are also investigating the possibility of developing wine yeast strains with antimicrobial activity that have been suggested as a partial replacement for the bioinhibitory properties of sulfur dioxide. External preservatives can be significantly reduced if wine yeast strains secrete natural antimicrobial peptides (such as bacteriocins) during fermentation, thereby playing an autosterilizing role. To this end, the pediocin and leucocin genes from lactic acid bacteria and the lysozyme gene from chicken egg white were successfully expressed in S. cerevisiae. Preliminary results indicate that it is indeed possible to develop bactericidal wine yeast strains that could be useful in the production of wine with reduced levels of potentially harmful chemical preservatives.

5. Complying with statutory regulation and consumer demands

Genetic engineering, lauded as a spectacular achievement in science, has unfortunately been repackaged in an emotive, fear-mongering wrapping. Critics are whipping up public alarm, often to fuel political agendas and to protect agricultural markets. The myths of genetically modified (GM) 'Frankenfood' and global havoc caused by genetically modified organisms (GMOs) have been spread far enough to masquerade in the cultural folklore as truth. Fears about food and environmental safety spread more readily than good sense or wise science; this has evoked a plethora of strict legislation and regulatory guidelines based far more on emotion than on science.

This mostly irrational debate began with questions about the morality ('unnatural,' interfering with evolution, playing God, etc.) and safety (GMOs and GM products are inherently dangerous, toxic to humans and bad for the environment) of genetic engineering. Over the years it has become generally agreed that these fears are largely groundless; to date, no scientifically reputable test has shown any of the GM foods currently on the shelf to be in the least toxic.

The initial problems with statutory approval and negative public perception of genetically engineered organisms in food and beverages are now slowly being dissolved by a growing consensus that risk is primarily a function of the characteristics of a product, rather than the use of genetic modification per se. Scientists have reached a broad consensus that organisms modified by modern molecular and cellular methods respond to the same physical and biological laws as any other. Therefore, no conceptual distinction exists between modification of yeast and grapevine by classical methods or by molecular techniques that modify DNA and transfer genes. 
Regulations, although differing in detail, are broadly similar in most countries. Guidelines for approval of GM products and the release of GMOs usually require a number of obvious guarantees. These include a complete definition of the DNA sequence introduced and the elimination of any sequence that is not indispensable for expression of the desired property; the absence of any selective advantage conferred on the transgenic organism that could allow it to become dominant in natural habitats; no danger to human health and/or the environment from the transformed DNA; and a clear advantage to both the producer and the consumer. 

The concept of 'substantial equivalence' is widely used in the determination of safety by comparison with analogous conventional food and beverage products. When substantial equivalence can be demonstrated, no further safety considerations are usually necessary. When substantial equivalence is not convincingly shown, the points of difference must be subjected to further safety scrutiny. However, to date, regulatory authorities appear more willing to approve the use of GMOs than the public is to use them. A significant proportion of the public still suspects that GM food will prove unhealthy in the long term, and that the escape of GMOs with transplanted genes will damage the environment and result in the loss of biodiversity. However, their questions now appear to be growing more specific. Is the product safe to the consumer and the environment? Is there sufficient legal and practical protection against accidents involving GMOs? Will genetic engineering reduce biodiversity and concentrate economic power in the hands of a few large producers? Do patents on living organisms confer an unfair advantage on certain producers? Should products produced by gene technology be specifically labelled?

It is clear that consumer education is essential to remove their fear of the unknown. Scientists must consistently inform the public and remain open about experiments, research and products. The consumer should be reassured of first-class, transparent regulatory systems and the meticulous implementation of biosafety legislation with clear technical standards and definitions with respect to GM products. The consumer should be persuaded by proper risk assessment and clear demonstration of safety, and thus empowered to make informed decisions.

There are a number of activities that must be avoided at all cost: conducting obviously risky experiments; misusing scientific data and consumer confusion to justify trade bans and technical barriers to free trade; riding the 'backlash' market with labels stating that a particular product is 'GM free'; suppressing 'inconvenient' scientific data or simply lying about food safety (as has been the case for some governments with distressingly bad biosafety records); and 'force-feeding' GM products and GMOs down consumers' throats for profit when there is no clear advantange for the consumer. 

Successful application of recombinant DNA technology in the wine industry will depend on assuring commercial users of genetically modified wine yeasts that existing desirable characteristics have not been damaged, that the requirements of beverage legislation are met, and that the engineered strain will be stable in practice, with suitable procedures for monitoring. The first recombinant wine products should unequivocally demonstrate organoleptic, hygienic and economic advantages for the wine producer and the consumer. Furthermore, wine's most enthralling and fascinating aspect, its diversity of style, should never be threatened by the use of tailored wine yeast strains. In fact, gene technology should rather be harnessed to expand the diversity of high quality wines and other grape-derived products. 

There is vast potential benefit to the wine consumer and industry alike in the application of this exciting new technology. That benefit will be realized, though, only if the application is judicious, systematic, and done with high regard for the unique nature of the product. In vino veritas!

6. Acknowledgements

I would also like to acknowledge the South African wine industry (Winetech) and the National Research Foundation (NRF) for financial support of our research programme.

7. Literature cited
This article is condensed from a comprehensive review by Pretorius (2000), and all references are not listed here. The reader is referred to the following review articles and references therein:
1. Bauer, F.F., and I.S. Pretorius. Yeast stress response and fermentation efficiency: How to survive the making of wine - a review. S. Afr. J. Enol. Vitic. 21: 27-51 (2000).
2. Du Toit, M., and I.S. Pretorius. Microbial spoilage and preservation of wine: Using weapons from nature's own arsenal - a review. S. Afr. J. Enol. Vitic. 21: 74-96 (2000).
3. Lambrechts, M.G., and I.S. Pretorius. Yeast and its importance to wine aroma - a review. S. Afr. J. Enol. Vitic. 21: 97-129 (2000).
4. Pretorius, I.S., and T.J. van der Westhuizen. The impact of yeast genetics and recombinant DNA technology on the wine industry - a review. S. Afr. J. Enol. Vitic. 12:3-31 (1991). 
5. Pretorius, I.S. Utilization of polysaccharides by Saccharomyces cerevisiae. In: Yeast Sugar Metabolism: Biochemistry, Genetics, Biotechnology and Applications. F.K. Zimmermann and K.-D. Entian (Eds.). pp 459-501, Technomic Publishing Co., Inc., Lancaster, Pennsylvania (1997).
6. Pretorius, I.S., T.J. van der Westhuizen, and O.P.H. Augustyn. 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 (1999). 
7. Pretorius, I.S. Engineering designer genes for wine yeasts. Austral. NZ Wine Ind. J. 14:42-47 (1999). 
8. Pretorius, I.S. Tailoring wine yeast for the new millennium: novel approaches to the ancient art of winemaking. Yeast 16:1-55 (2000). 
9. Rainieri, S., and I.S. Pretorius. Selection and improvement of wine yeasts. Ann. Microbiol. 50:15-31 (2000).
10. Van Rensburg, P., and I.S. Pretorius. Enzymes in winemaking: Harnessing natural catalysts for efficient biotransformations - a review. S. Afr. J. Enol. Vitic. 21: 52-73 (2000).
11. Vivier, M.A., and I.S. Pretorius. Genetic improvement of grapevine: Tailoring grape varieties for the third millennium - a review. S. Afr. J. Enol. Vitic. 21: 5-26 (2000).

Science in Africa - Africa's First On-Line Science Magazine

Copyright  2001, Janice Limson. All Rights Reserved