0% found this document useful (0 votes)

104 views55 pages

Yeast - 2000 - Pretorius - Tailoring Wine Yeast For The New Millennium Novel Approaches To The Ancient Art of Winemaking

This document summarizes a review article about tailoring wine yeast for new approaches to winemaking in the 21st century. It discusses the history of winemaking and yeast's role in fermentation. It highlights opportunities to select specialized wine yeast strains with improved properties to meet modern demands. The review cautions that fully realizing potential benefits from genetically modified yeast will require overcoming regulatory hurdles and building understanding of wine yeast genomes.

Uploaded by

Tatiana Rivera Correa

We take content rights seriously. If you suspect this is your content, claim it here.

Available Formats

Download as PDF, TXT or read online on Scribd

0% found this document useful (0 votes)

104 views55 pages

Yeast - 2000 - Pretorius - Tailoring Wine Yeast For The New Millennium Novel Approaches To The Ancient Art of Winemaking

Uploaded by

Tatiana Rivera Correa

We take content rights seriously. If you suspect this is your content, claim it here.

Available Formats

Download as PDF, TXT or read online on Scribd

You are on page 1/ 55

Yeast

Yeast 2000; 16: 675±729.

Review

Tailoring wine yeast for the new millennium: novel

approaches to the ancient art of winemaking
Isak S. Pretorius*
Institute for Wine Biotechnology, University of Stellenbosch, Stellenbosch, ZA-7600, South Africa

* Correspondence to: Abstract

I. S. Pretorius, Institute for Wine
Biotechnology, University of Yeasts are predominant in the ancient and complex process of winemaking. In spontaneous
Stellenbosch, Victoria Street, fermentations, there is a progressive growth pattern of indigenous yeasts, with the ®nal
ZA-7600 Stellenbosch, stages invariably being dominated by the alcohol-tolerant strains of Saccharomyces
South Africa. cerevisiae. This species is universally known as the `wine yeast' and is widely preferred for
E-mail: [email protected]. initiating wine fermentations. The primary role of wine yeast is to catalyze the rapid,
complete and ef®cient conversion of grape sugars to ethanol, carbon dioxide and other
minor, but important, metabolites without the development of off-¯avours. However, due to
the demanding nature of modern winemaking practices and sophisticated wine markets,
there is an ever-growing quest for specialized wine yeast strains possessing a wide range of
optimized, improved or novel oenological properties. This review highlights the wealth of
untapped indigenous yeasts with oenological potential, the complexity of wine yeasts'
genetic features and the genetic techniques often used in strain development. The current
status of genetically improved wine yeasts and potential targets for further strain
development are outlined. In light of the limited knowledge of industrial wine yeasts'
complex genomes and the daunting challenges to comply with strict statutory regulations
and consumer demands regarding the future use of genetically modi®ed strains, this review
cautions against unrealistic expectations over the short term. However, the staggering
potential advantages of improved wine yeasts to both the winemaker and consumer in the
third millennium are pointed out. Copyright # 2000 John Wiley & Sons, Ltd.
Keywords: Saccharomyces cerevisiae; wine yeast; genetic improvement

Contents the Caucasus and Mesopotamia as early as 6000 BC

[131]. References to wine have been found in Egypt
Introduction . . . . . . . . . . . . . . . . . . . . . . . 675 and Phoenicia dating as far back as 5000 BC and, by
Yeast diversity associated with grapes and wine- 2000 BC, wine was being produced in Greece and
making . . . . . . . . . . . . . . . . . . . . . . . . . 678 Crete. Colonization by the Romans spread wine-
Genetic constitution of wine yeasts . . . . . . . 681 making all around the Mediterranean; by 500 BC,
Genetic techniques for the analysis and develop- wine was being produced in Sicily, Italy, France,
ment of wine yeast strains . . . . . . . . . . . . 685 Spain, Portugal and northern Africa. Cultivation of
Targets for wine yeast strain development . . 691 the vine also spread into the Balkan States, and the
Complying with statutory regulation and consumer Romans took it into Germany and other parts of
demands . . . . . . . . . . . . . . . . . . . . . . . . 721 northern Europe, eventually reaching as far as
Conclusions . . . . . . . . . . . . . . . . . . . . . . . 723 Britain (Figure 1) [131].
References . . . . . . . . . . . . . . . . . . . . . . . . 723 European explorers in the sixteenth century
introduced the vine into the New World. In 1530
Introduction the Spanish conquistadors planted Vitis vinifera in
Mexico, Argentina, Peru and Chile. In 1655 Dutch
The history of winemaking parallels that of civiliza- settlers in South Africa planted French vine cuttings
tion: historians believe that wine was being made in on the lower slopes of the Cape of Good Hope's

Copyright # 2000 John Wiley & Sons, Ltd.

10970061, 2000, 8, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/1097-0061(20000615)16:8<675::AID-YEA585>3.0.CO;2-B by Cochrane Colombia, Wiley Online Library on [24/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
676 I. S. Pretorius

Figure 1. The early spreading and world distribution of the vine and winemaking technology

majestic Table Mountain. Planting in California Until the early years of the seventeenth century,
followed soon thereafter, and in Australia and wine was considered to be the only wholesome
New Zealand more than a century later, in 1813 readily storable (to a point) beverage, accounting
[131]. for the rapid global increase of wine fermentation
Disaster struck the world of wine in the 1870s, technology. Today wine is synonymous with culture
when the root-eating insect Phyloxera vastatrix and a convivial lifestyle around the world, comple-
(Dactylasphaera vitifoliae) threatened almost every menting food, entertainment and the arts. Wine
vine in Europe and the New World. They were plays a major role in the economies of many
pulled up and replaced with new V. vinifera vines nations, which produce more than 26 billion litres
grafted onto phyloxera-resistant rootstocks from of wine annually. Modern winemakers supply a
the native American vine. Despite attacks by wide variety of wines year round independent of
phyloxera and the spread of other diseases, such as location and time of consumption. Fierce competi-
downy mildew (Plasmopara viticola) and powdery tion for market share has led to increased diversity
mildew (Oidium tuckerii), the Of®ce International (Figure 2) and innovation within the wine industry,
de la Vigne et du Vin (OIV) in Paris reports that much to the bene®t of the consumer.
today there are some 8 million hectares of vineyards A look at the early days of winemaking makes it
across the world, mainly concentrated within the obvious that, while different techniques produced
earth's temperate zones. Each of these vineyards varied styles of wine, the basic principles changed
re¯ects the terroir, history, culture and traditions of very little (Figure 3). During the last 150 years or
its region. so, however, the scienti®c basis of winemaking has

Copyright # 2000 John Wiley & Sons, Ltd. Yeast 2000; 16: 675±729.
10970061, 2000, 8, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/1097-0061(20000615)16:8<675::AID-YEA585>3.0.CO;2-B by Cochrane Colombia, Wiley Online Library on [24/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Tailoring wine yeast for the new millennium 677

Figure 2. Diversity of natural (table) and forti®ed wines produced in South Africa

gradually become clearer, and many practices once and fructose) into alcohol and carbon dioxide,
thought impossible have now become routine. winemakers could control the process from vine-
In 1863, Louis Pasteur revealed for the ®rst time yard to bottling plant. Later, yeasts with improved
the hidden world of microbial activity during wine characteristics were selected and, by 1890, MuÈller-
fermentation. He proved conclusively that yeast is Thurgau introduced the concept of inoculating wine
the primary catalyst in wine fermentation, basing fermentations with pure yeast cultures. As a result,
his work upon Antonie van Leeuwenhoek's ®rst the quality and quantity of wine production were
microscopic observation (in 1680) of yeast cells and vastly improved.
the claims by three other independent pioneers, These fundamental innovations in winemaking
Cagniard-Latour, KuÈtzing and Schwann (in the late practices revolutionized the wine industry, and
1830s) that these cells are living organisms [6]. With today the forces of market-pull and technology-
the knowledge that yeast was responsible for the push continue to challenge the tension between
biotransformation of grape sugars (mainly glucose tradition and innovation [118]. There will continue

yeasts to ferment the grape sugars spontaneously to

ethanol, carbon dioxide and other metabolites, is
today vastly more complex and sophisticated. The
fermentation of grape must and production of
premium quality wines is a complex ecological and
biochemical process involving the sequential devel-
opment of microbial species, as affected by a
particular environment. The process includes the
interaction of fungi, yeasts, lactic acid bacteria,
acetic acid bacteria, as well as the mycoviruses and
bacteriophages affecting these grape-associated
microorganisms [35,36]. Of all these, yeasts are at
the heart of the biochemical interaction with the
musts derived from the varieties of V. vinifera and
other grape species.
Yeasts are de®ned as unicellular ascomycetous or
basidiomycetous fungi whose vegetative growth
results predominantly from budding or ®ssion, and
which do not form their sexual states within or
upon a fruiting body [80]. Of the 100 yeast genera
representing over 700 species described in the latest
edition of the monographic series, The Yeasts, A
Taxonomic Study [79], 15 are associated with wine-
making: Brettanomyces and its sexual (`perfect')
equivalent Dekkera; Candida; Cryptococcus; Debar-
Figure 3. The main steps in wine production (adapted from yomyces; Hanseniaspora and its asexual counterpart
Walker [176]) Kloeckera; Kluyveromyces; Metschnikowia; Pichia;
Rhodotorula; Saccharomyces; Saccharomycodes;
to be further improvements in winemaking by Schizosaccharomyces; and Zygosaccharomyces
re®ning viticultural and oenological practices. [121]. Despite the striking growth in the number of
These factors will remain important to the improve- described yeast species over the last 50 years,
ment of the overall quality and endless variety of [77,79,94,95], it is generally accepted that the
wine. wealth of yeast biodiversity with hidden oenological
But today there is a new (and, for the moment, potential is still largely untapped.
controversial) focal point for innovation in wine-
makingÐthe genetic modi®cation of the two main Yeast ecology
organisms involved, the grape cultivar and wine With respect to the vineyard and winery niche
yeast. The diversity of yeast species associated with habitats, some of these yeasts are considered as
winemaking, the tailoring of wine yeast, and the `autochthonous' (essential) and others as `allochtho-
possible use of strains expressing novel designer nous' (transient or fortuitous) members of the
genes make possible exciting new approaches to communities found in these environments. Their
winemaking in the twenty-®rst century. successful coexistence depends on the sum of all
physical, chemical and biotic factors that pertain to
vineyards and wineries [84]. `Generalist' yeasts are
Yeast diversity associated with grapes
endowed with a broad niche and occupy many
and winemaking
habitats, whereas `specialist' yeasts occur in unique
habitats [176].
Yeast biodiversity and ecology
The micro¯ora of grapes vary according to the
Yeast diversity and important taxa grape variety; temperature, rainfall and other
Louis Pasteur's simple biochemical process of climatic in¯uences; soil, fertilization, irrigation and
converting grape juice into wine by allowing the viticultural practices (e.g. vine canopy manage-

ment); development stage at which grapes are degree to which it has been cleaned and sanitized.
examined; physical damage caused by mould, Irregular, unpolished surfaces such as cracks and
insects and birds; and fungicides applied to vine- welds may support dense populations of winery
yards [121]. It is also important to note that yeasts [41].
harvesting equipment, including mechanical har- The microbiota of grape must are affected
vesters, picking baskets and other infrequently indirectly by all the factors in¯uencing the indigen-
cleaned delivery containers can also represent sites ous grape micro¯ora and the winery ¯ora. Added to
for yeast accumulation and microbiological activity these factors are the following: method of grape
before grapes reach the winery [41]. This becomes harvest (handpicked or mechanical), grape tempera-
more important as travel time to the winery ture, transport from vineyard to cellar (distance/
increases. time, initial grape temperature, air temperature,
Kloeckera (e.g. K. apiculata) and Hanseniaspora sulphite addition), condition of grapes (time, tem-
(e.g. H. uvarum) are the predominant species on the perature, sulphite addition) and must pretreatment
surface of grape berries, accounting for roughly (cellar hygiene, aeration, enzyme treatment, sulphite
50±75% of the total yeast population [35,36]. addition, clari®cation method, temperature, inocu-
Numerically less prevalent than these apiculate lation with yeast starter cultures) [61,121].
yeasts are species of Candida (e.g. C. stellata and Though grape must is relatively complete in
C. pulcherrima), Brettanomyces (e.g. B. intermedius, nutrient content, it can support the growth of only
B. lambicus and B. custeri), Cryptococcus, Kluyvero- a limited number of microbial species [61]. The low
myces, Metschnikowia (e.g. the sexual equivalent of pH and high sugar content of grape must exert
C. pulcherrima, M. pulcherrima), Pichia (e.g. the so- strong selective pressure on the microorganisms,
called ®lm yeast, P. membranaefaciens, as well as such that only a few yeast and bacterial species can
those species that were previously assigned to the
proliferate. Concentrations of sulphur dioxide,
genus Hansenula, i.e. H. anomala) and the pink
added as an antioxidant and antimicrobial preser-
yeast Rhodotorula (e.g. R. minuta) [35,36,161].
vative, impose additional selection, particularly
Contrary to popular belief, fermentative species
against undesirable oxidative microbes [61]. The
of Saccharomyces (e.g. S. cerevisiae) occur at
selectivity of fermenting must is further strength-
extremely low populations on healthy, undamaged
ened once anaerobic conditions are established;
grapes and are rarely isolated from intact berries
certain nutrients become depleted and the increas-
and vineyard soils [100]. In fact, the origin of S.
ing levels of ethanol start to eliminate alcohol-
cerevisiae is quite controversial; one school of
sensitive microbial species [61]. Spontaneous fer-
thought claims that the primary source of this
commercially important yeast is the vineyard, and mentation of grape juice into wine can be regarded
the presence or absence of S. cerevisiae differs with as a heterogeneous microbiological process involv-
each plant and grape cluster [157]. Others believe ing the sequential development of various yeasts
the evidence points to a direct association with and other microbiological species, affected by the
arti®cial, man-made environments such as wineries prevailing fermentation conditions in a particular
and fermentation plants, and that a natural origin vat or tank.
for S. cerevisiae should be excluded [100,168]. In When must is used as a culture medium, selective
contrast to its low occurrence in natural habitats pressures always favour the yeasts with the
such as vineyards, S. cerevisiae is abundant on the most ef®cient fermentative catabolism, particularly
grape juice and must-coated surfaces of winery strains of S. cerevisiae and perhaps strains of closely
equipment, forming an important component of a related species such as Saccharomyces bayanus. For
so-called `residential' or `winery' yeast ¯ora [37]. In this reason, S. cerevisiae is almost universally
fact, S. cerevisiae is by far the most dominant yeast preferred for initiating alcoholic fermentation, and
species colonizing surfaces in wineries, demonstrat- has earned itself the title of `the wine yeast'. This,
ing the selective effects of grape juice and wine as however, does not preclude the possibility that, in
growth substrates. The extent of the development the future, some winemakers might prefer to use a
of a winery yeast ¯ora, usually comprising species mixture of indigenous yeast species and strains as
of Saccharomyces, Candida and Brettanomyces, starter cultures tailored to re¯ect the yeast bio-
depends upon the nature of the surface and the diversity of a given region [58].

Wine yeast starter cultures undoubtedly yield wines with distinct sensorial
quality, often described as wine with a fuller,
Spontaneous vs. inoculated wine fermentations
rounder palate structure. This may well be the
Originally, all wine was made by taking advantage
consequence of higher concentrations of glycerol
of natural micro¯ora for spontaneous fermentation;
and other polyols produced by indigenous yeasts.
no deliberate inoculation was made to start the
Furthermore, the extended lag phase before the
process. Various yeasts found on the surface of
onset of vigorous fermentation allows for the
grape skins and the indigenous microbiota asso-
reaction of oxygen with anthocyanins and other
ciated with winery surfaces participate in these
phenols in the absence of ethanol, which is thought
natural wine fermentations. Yeasts of the genera
to enhance colour stability in red wine as well as
Kloeckera, Hanseniaspora and Candida predominate
accelerating phenol polymerization [189].
in the early stages, followed by several species of However, the cast of stylistic characters and the
Metschnikowia and Pichia in the middle stages, individual and collective contribution of indigenous
when the ethanol rises to 3±4% [37,105]. The latter yeasts to the wine vary. The outcome of sponta-
stages of natural wine fermentations are invariably neous fermentation depends not only on the
dominated by the alcohol-tolerant strains of S. numbers and diversity of yeasts present in must,
cerevisiae. Other yeasts, such as species of Bretta- but also upon grape chemistry and processing
nomyces, Kluyveromyces, Schizosaccharomyces, Tor- protocol. The combined effect makes the outcome
ulaspora and Zygosaccharomyces, may also be dif®cult to predict. This lack of predictability/
present during the fermentation and subsequently reproducibility is most troublesome when compar-
in the wine, some of which are capable of adversely ing spontaneous fermentation with that resulting
affecting sensory quality. from active dried yeast starters.
A breakthrough was made when Hansen of the Notwithstanding the fact that spontaneous fer-
Carlsberg Brewery in Denmark isolated a pure mentations usually take longer than most wine-
culture derived from a single yeast cell and, in 1890, makers are willing to accept, and that the outcome
MuÈller-Thurgau from Geisenheim introduced the is not always what was anticipated, there is no
concept of inoculating wine fermentations with pure consensus among the world winemakers about
yeast starter cultures [119]. In 1965, the ®rst two using yeast starters [41]. At one extreme are those
commercial active dried wine yeast (ADWY) strains who continue to use solely indigenous yeasts,
were produced for a large Californian winery [27]. believing that the unique contributions of diverse
These two strains, Montrachet and Pasteur Cham- yeast species confer a complexity upon wine not
pagne, were offered worldwide as all-purpose seen in inoculated and guided fermentations.
yeasts, with limited success. Today, several yeast- Others prefer to begin with native yeasts and
manufacturing companies market a wide variety of later inoculate with a commercial yeast starter.
dehydrated cultures of various S. cerevisiae strains. Still others initiate their wine fermentation with
In guided fermentations, the actively growing starters but at lower than recommended inoculum
starter culture dominates the native yeast species levels.
present in grape must. To achieve this, a cell density In large-scale wine production, however, where
upon inoculation of 1±3 million colony forming rapid and reliable fermentations are essential for
units (cfu)/ml is usually recommended [41]. consistent wine ¯avour and predictable quality, the
The practice of spontaneous fermentations use of selected pure yeast inocula of known ability
remained prevalent in `Old World' wine-producing is preferred. These large wineries will be the main
areas until the 1980s because of the popular belief bene®ciaries of programmes aimed at producing
that superior yeast strains associated with speci®c new yeast strains with even more reliable perfor-
vineyards gave a distinctive style and quality to mance, reducing processing inputs, and facilitating
wine. Even today, winemakers at many `boutique' the production of affordable high-quality wines.
wineries accept the potentially staggering risks
involved in spontaneous fermentations to achieve Industrial-taxonomic relationship for wine yeast
stylistic distinction and vintage variability. Conver- starter cultures
sion of grape sugars to alcohol and other end The ®rst `all-purpose' commercial wine yeast strains
products by mixed populations of yeast may were only partially successful. As the modern

winemaking community was slowly adopting the recognize the importance of speci®c S. cerevisiae
use of starter cultures in preference to the `gam- starter culture strains to the type and style of
bling' tradition of spontaneous fermentations, the product. With the importance of S. cerevisiae's role
need became evident for separate Saccharomyces in winemaking now ®rmly established, there is an
strains with speci®c characteristics for different ever-growing demand for new and improved wine
types of wine. The strains of Saccharomyces yeast strains. In addition to the primary role of
became classi®ed into several different species or wine yeast to catalyze the ef®cient and complete
varieties, including S. bayanus, S. beticus, S. conversion of grape sugars to alcohol without the
capensis, S. chevaleri, S. ellipsoideus, S. fermentati, development of off-¯avours, starter culture strains
S. oviformis, S. rosei and S. vini [169]. of S. cerevisiae must now possess a range of other
The characteristics of some of the yeasts used for properties, such as those listed in Table 2. The
the production of speci®c wine types were so importance of these additional yeast characteristics
marked that a strong taxonomic linkage was differs with the type and style of wine to be made
believed to exist [61]. For example, while S. and the technical requirements of the winery. The
ellipsoideus was widely used for the production of need is for S. cerevisiae strains that are better
dry wine, ethanol-tolerant and ¯occulent strains adapted to the different wine-producing regions of
with autolytic properties (e.g. S. bayanus and S. the world with their respective grape varietals,
oviformis) were preferred for the production of viticultural practices and winemaking techniques.
bottle-fermented sparkling wine, ®lm-forming Leading winemakers are now translating the adage
strains with strong oxidative capabilities (e.g. S. ``horses for courses'' into ``special yeasts for special
beticus and S. capensis) for the production of ¯or treats''.
sherry, and osmotolerant strains forming little or no
volatile acids (e.g. S. rosei) for sweet wines [61].
Over the years, successive editions of The Yeasts, Genetic constitution of wine yeasts
A Taxonomic Study re-de®ned and re-grouped
species of Saccharomyces so that the 16 and 21 Wine yeast cytology and reproduction
species that were described in the ®rst and second S. cerevisiae cells are generally ellipsoidal in shape.
editions, respectively, were assigned to a single The subcellular compartmentalization in this
species in the third edition and designated as S. eukaryote is schematically presented in Figure 4.
cerevisiae [77,94,95]. In the fourth edition, the single Under optimal nutritional and cultural conditions,
species has been separated into four species [79]. S. cerevisiae doubles its mass every 90 min. The cell
Despite the considerable phenotypic differences division cycle is presented in Figure 5 and the basic
among the various wine yeast strains, most of them life cycles of heterothallic and homothallic strains
are now considered to be physiological strains of S. are shown in Figure 6.
cerevisiae. Of all these wine yeasts, only S. fermentati
and S. rosei were not re-classi®ed as Saccharomyces
Chromosomal DNA and ploidy
but rather as Torulaspora delbrueckii [79]. The
assignment of most of the traditional wine yeast S. cerevisiae has a relatively small genome, a large
strains to a single species does not, however, imply number of chromosomes, little repetitive DNA and
that all strains of S. cerevisiae are equally suitable few introns [116]. Haploid strains contain approxi-
for the various wine fermentations; they differ mately 12±13 megabases (mb) of nuclear DNA,
signi®cantly in their fermentation performance and distributed along 16 linear chromosomes. Each
their contribution to the ®nal bouquet and quality of chromosome is a single deoxyribonucleic acid
wine and distillates [61,121]. It is therefore not (DNA) molecule approximately 200±2200 kilobases
surprising that several molecular methods have (kb) long. The genome of a laboratory strain of S.
been developed for wine yeast strain differentiation cerevisiae has been completely sequenced and found
[25,26,160] (Table 1). to contain roughly 6000 protein-encoding genes.
The S. cerevisiae genome, which is relatively rich in
A quest for new wine yeast strains guanine and cytosine content (%G+C of 39±41) is
While the old saying `the best wines are made in the much more compact when compared with the
vineyard' is still valid, oenologists have also come to genomes of other eukaryotic cells.

Table 1. Molecular methods for wine yeast strain differentiation (adapted from Walker [186])
Method Description

Chromatography Pyrolysis-gas chromatography or gas chromatography of long-chain fatty acid

methyl esters
Polyacrylamide gel electrophoresis (PAGE) Total soluble yeast proteins are electrophoresed and banding patterns analyzed
by computer
Restriction enzyme analysis (DNA ®ngerprinting) Total, ribosomal or mitochondrial DNA is digested with restriction
endo-nucleases and speci®c fragments hybridized after electrophoretic
separation with multi-locus DNA probes such as the Ty1 retrotransposon;
restriction fragment length polymorphisms (RFLPs) are detected
Electrophoretic karyotyping (chromosome ®ngerprinting) Whole yeast chromosomes are separated electrophoretically using pulse-®eld
techniques; chromosome length polymorphisms (CLPs)
Polymerase chain reaction (PCR) Speci®c DNA sequences are exponentially propagated in vitro and the ampli®ed
products analyzed after electrophoretic separation. Randomly ampli®ed
polymorphic DNA (RAPD) and ampli®ed fragment length polymorphisms
(AFLPs) can also be analyzed by PCR
Genetic tagging Speci®c genetic sequences, including selectable markers, are introduced into
yeasts to facilitate their recognition (e.g. replacement of chloramphenicol
resistance sequences with a `tag' which confers sensitivity to the antibiotic)

Most laboratory-bred strains of S. cerevisiae are formance [56]. However, other researchers claim
either haploid or diploid. However, industrial wine that the polyploid state might enable industrial
yeast strains are predominantly diploid or aneu- yeasts to harbour a high dosage of genes important
ploid, and occasionally polyploid [142]. It is not yet for ef®cient fermentation. When a heat-induced
clear whether polyploidy in industrial yeast strains endomitotic polyploidization procedure was used to
is advantageous. When a series of homozygous and construct an isogenic ploidy series (2N to 4N) from
heterozygous strains with ploidy from one to eight an industrial wine yeast strain, it was found that the
were constructed, it was found that the heterozy- physical and metabolic differences observed among
gous triploids and tetraploids were more ef®cient in these strains were due to differences in gene dosage
fermentation than the homozygous strains of higher alone, and not to heterosis [132]. These reports only
or lower ploidy. Based on these results it was emphasize the fact that the relationship between the
concluded that heterosis rather than ploidy is fermentation ability and the ploidy of a yeast strain
responsible for improvement of fermentation per- is rather complicated.

Table 2. Desirable characteristics of wine yeast

Fermentation properties Technological properties

Rapid initiation of fermentation High genetic stability
High fermentation ef®ciency High sulphite tolerance
High ethanol tolerance Low sulphite binding activity
High osmotolerance Low foam formation
Low temperature optimum Flocculation properties
Moderate biomass production Compacts sediment
Flavour characteristics Resistance to desiccation
Low sulphide/DMS/thiol formation Zymocidal (killer) properties
Low volatile acidity production Genetic marking
Low higher alcohol production Proteolytic acitivity
Liberation of glycosylated ¯avour precursors Low nitrogen demand
High glycerol production Metabolic properties with health implications
Hydrolytic activity Low sulphite formation
Enhanced autolysis Low biogenic amine formation
Modi®ed esterase activity Low ethyl carbamate (urea) potential

Figure 4. A schematic representation of the subcellular compartmentalization of a wine yeast cell (adapted from Pretorius
and Van der Westhuizen [119]). The cell envelope, comprising a cell wall, periplasm and plasma membrane, surrounds and
encases the yeast cytoplasm. The structural organization of the intracellular milieu, containing organelles such as the nucleus,
endoplasmic reticulum, Golgi apparatus, mitochondria and vacuoles, is maintained by a cytoskeleton. Several of these
organelles derive from an extended intramembranous system and are not completely independent of each other

Figure 6. A schematic representation of the life cycle of

Figure 5. A schematic representation of the cell cycle of a
heterothallic and homothallic wine yeast strains (adapted
budding wine yeast cell. Buds may arise at any point on the
from Hammond [56]). Haploid cells of the MATa mating type
mother cell surface, but never again at the same site.
produce a 13 amino acid-long a factor; while the a mating
Branched chaining (pseudohyphae) may occasionally follow
type cells produce a peptide of 12 amino acids, the a factor.
multilateral budding when buds fail to separate. Under
A MATa/MATa diploid cell formed by mating by two haploid
optimal nutritional and cultural conditions S. cerevisiae
cells of opposite mating types can neither produce nor
doubles its mass every 90 min. The cell division cycle
respond to mating pheromones and will under satisfactory
consists of four phases, G1, S, G2 and M
nutritional and cultural conditions grow and divide, main-
taining the diploid state. Upon nutritional starvation, the
Extrachromosomal elements MATa/MATa diploid cell undergoes meiosis, generating four
Several non-Mendelian genetic elements are known haploid ascospores (two MATa and two MATa ascospores)
encapsulated within an ascus. When released from the ascus,
to exist in the nucleus (e.g. transposons and 2 mm
the ascospores germinate to commence new rounds of
plasmid DNA), mitochondria and cytoplasm (e.g.
haploid existence. Strains that can be maintained stably for
viral-like particles and prion-like elements such as many generations as haploids are termed heterothallic.
y, g and URE3). The genome of S. cerevisiae Strains in which sex reversals, cell fusion and diploid
contains approximately 35±55 copies of retrotrans- formation occur are termed homothallic
posons (Ty elements). These transposable elements
move from one genomic location to another via an
majority of the enzymes involved in the generation
RNA intermediate using reverse transcriptase. The
of ATP during aerobic growth. Unlike the replica-
2 mm plasmid DNA is the only naturally occurring,
stably maintained, circular nuclear plasmid in S. tion of nuclear DNA, replication of mtDNA is not
cerevisiae. This 6.3 kb extrachromosomal element is limited to the S phase; it takes place throughout the
also inherited in a non-Mendelian fashion and, cell cycle. The mtDNA polymerase also lacks
although most strains of S. cerevisiae contain 50- proofreading (exonuclease) activity, resulting in a
100 copies of 2 mm DNA per cell, its biological much higher mutation rate within the mtDNA than
function has not yet been discovered. within nuclear genes, so mtDNA can evolve
extremely rapidly [55]. This lack of an error repair
mechanism during mtDNA replication is partly
Mitochondrial DNA compensated for by the abundance of mitochon-
Mitochondria possess their own genetic system and drial DNA molecules in a single cell [179].
their own protein synthetic machinery. S. cerevisiae With a genome that is much larger than required,
has among the largest mitochondrial DNAs the yeast mtDNA encodes proteins that perform
(mtDNAs) of any organism, consisting of 75 kb only a few activities. One explanation for the
circularly permuted molecules [55]. However, the persistence of this large mitochondrial genome is
mitochondrial genome of S. cerevisiae is adenine- that, in yeast, it could play the additional role of a
thymine (A±T) rich, carrying the genetic informa- reservoir of genetic diversity, capable of serving the
tion for only a few, essential, mitochondrial nuclear genome by contributing evolved sequences
components, and does not even code for the [55]. This could be one contributing factor to the

observed genetic heterogeneity of pure cultures of yeast strains are immune to these zymocins but do
wine yeasts. not produce active toxin. These so-called neutral
Unlike other eukaryotic cells, yeasts can survive strains contain L and M dsRNA genomes, but the
without its mtDNA. Mitochondrial mutants usually M genomes code only for the production of the
lack vital oxidative enzymes, rendering them unable immunity factor and not for the production of an
to generate ATP oxidatively. As a result, mitochon- active zymocin. Zymocidal strains isolated from
drial mutants grow slowly and form smaller (petite) fermenting grape musts are usually of the K2 or K28
colonies on solid agar surfaces than the wild-type type, most probably because of the low pH
(grande) cells. Petite mutants are respiratory- optimum for their zymocidal activity which is
de®cient and unable to utilize non-fermentable favoured by the acidic conditions of grape must
substrates. The term `cytoplasmic petite mutant' (pH 2.8±3.8) [136,182]. Apart from sporadic reports
describes respiratory-defective strains with cytoplas- which imply that the presence of killer contami-
mically inherited mutations, ranging from point nants could, under certain conditions, contribute
mutations (Mitx) through deletion mutations towards stuck (incomplete) or sluggish wine fer-
(Rhox) to complete elimination of the mtDNA mentations [167], no real effects on the whole-
(Rhou) [55]. To distinguish cytoplasmic petite someness and sensorial quality of wines are
mutants from respiratory-de®cient strains with produced by zymocidal S. cerevisiae strains.
genetic lesions in nuclear genes, the latter are
referred to as nuclear petite or pet mutants [55].
Genetic techniques for the analysis and
The mitochondrial genome is also involved in cell
development of wine yeast strains
functions other than respiratory metabolism. Since
the generation of petite mutants of wine yeasts
S. cerevisiae can be modi®ed genetically in many
occurs spontaneously at quite high rates, it is
ways. Some techniques alter limited regions of
important to note that yeasts with different
the genome, whereas other techniques are used
mtDNAs could differ in their ¯occulation character-
to recombine or rearrange the entire genome
istics, lipid metabolism, higher alcohol production
[8,56,119,123]. Techniques having the greatest
and formation of ¯avour compounds [176]. Thus,
potential in genetic programming of wine yeast
although wine yeasts are not required to respire
strains are: clonal selection of variants, mutation
during fermentation of grape must, some mtDNA-
and selection, hybridization, rare-mating, sphero-
encoded functions are important and for this reason
plast fusion and gene cloning and transformation.
petite strains are not used for winemaking.
The combined use of tetrad analysis, replica-
plating, mutagenesis, hybridization and recombi-
Killer factors
nant DNA methods have dramatically increased the
The killer phenomenon in S. cerevisiae is associated
genetic diversity that can be introduced into yeast
with the presence of non-infectious, intracellular
cells.
virus-like particles (VLP). VLPs in killer (zymoci-
dal) yeasts that are cytoplasmically inherited con-
Selection of variants
tain two major linear double-stranded ribonucleic
acid (dsRNA) types, the L and M genomes [182]. Genetic drift
The L genome encodes an RNA-dependent RNA The maintenance of the genetic identity of strains in
polymerase and the viral coat protein that encapsu- a pure culture is complex. Although the term `pure
lates both genomes. The M genome encodes both a culture' denotes that it has been derived from a
proteinaceous toxin (zymocin) and an immunity single cell, it does not imply that the culture is
factor. The zymocin is secreted by the zymocidal genetically uniform [142]. Even under closely con-
strains and is lethal to sensitive strains of the same trolled conditions of growth, a yeast strain reveals
species. Five types of S. cerevisiae killers, K1, slow but distinct changes after many generations.
K2, K3, K28 and K3GR1 have been described This might be due to a number of different
[40,182,185,186,187]. It now seems that K3 and processes, including spontaneous mutation, Ty-
K3GR1 are only variants from the K2 type. The size promoted chromosomal translocations and, more
of the L genome is 4.5 kb, whereas the different M frequently, mitotic crossing-over or gene conver-
dsRNA genomes vary between 1.3 and 2 kb. Some sion. Over the years, the phenomena of hetero-

geneity in pure yeast cultures and genetic drifting and selected to be used as commercial starter
were harnessed to improve wine yeast strains [142]. cultures. It is now believed that strains of S.
It was shown that successive single-cell cultures of cerevisiae indigenous to vineyards and wineries
commercial wine yeast strains could result in strains tend to be homozygous for most of the genes by a
with considerably improved characteristics. It is process known as `genome renewal' [105]. This
well known that the sporulation and spore viability phenomenon is based on the ability of homothallic
of pure yeast cultures are generally poor and that haploid S. cerevisiae cells to switch their mating-
there is considerable variation in growth rate type from a to a and vice versa through the HO-
between spore clones [142]. Some of this genetic controlled cassette model (Figure 7), and to con-
heterozygosity of pure cultures is undoubtedly due jugate with cells of the same single-spore colony.
to segregation of aneuploid chromosome comple- Continued propagation of yeast cells in their
ments from a polyploid or aneuploid parental natural (e.g. vineyard) or man-made (e.g. winery)
strain; the remaining variation probably re¯ects habitats leads to strains of S. cerevisiae accumulat-
the segregation of lethal genes or genes compromising heterozygous recessive mutations, and the
ing ef®cient growth. Mating between MATa and concomitant heterozygotes can change to com-
MATa ascospores, generated by sporulation, can pletely homozygous diploids by sporulation and
also cause genetic instability. Increased homozy- homothallic switching of individual haploid spores
gosity in polyploid yeasts is expected to confer [105].
greater genetic stability. It has also been reported This process would eliminate the recessive lethal
that the rate of genetic drift of yeast strains increases or deleterious genes that adversely affect yeast
with ploidy. This ®nding is contrary to a long-held ®tness (e.g. slower growth, lower fermentation
belief that the polyploid state protects against rate, reduced spore viability, etc.). Genome renewal
mutation and genetic variability. Since wine yeasts could also be responsible for the replacement of the
most probably harbour recessive mutations, genetic parental heterozygous strains by the new homo-
stability is likely to be a function of the frequency of zygous diploids bearing new recessive alleles that
segregational events leading to expression of mutant increase ®tness (Figure 8). Perhaps this is the reason
genes, rather than the frequency of mutation itself why most indigenous strains of S. cerevisiae isolated
[56]. It has also been established that, although at a from grapes, wineries and must are homothallic,
much lower frequency than in laboratory strains since homothallism, together with the capability to
under strong selective pressure, Ty-driven ectopic sporulate, would provide the yeast community with
recombinations are more commonly responsible for a mechanism by which cells carrying deleterious
karyotypic variability in wine yeast than was initially recessive mutations could be eliminated, thereby
expected [124]. It would seem unwise to assume a enabling them to adapt ef®ciently to changing
priori that all wine yeast strains are genetically stable. environmental conditions [121].
It is not yet clear what the contributions of The practical implications of genome renewal and
mitochondrial genomes are to the genetic drift in yeast population dynamics in the vineyards and
wine yeasts. However, this genetic heterogeneity wineries (and even within yeast starter cultures) are
serves as a rich, natural gene pool from which far-reaching, whether winemakers rely on sponta-
desirable variants can be selected and further neous fermentation of grape juice or whether they
improved. Selection of variants is, therefore, a inoculate grape must with selected wine yeast
direct means of strain development and successful strains. Although dramatic improvements in most
isolation of variants with desirable characteristics characteristics cannot be expected, intra-strain
depends on the frequency at which mitotic recombi- selection has been used for decades to obtain
nation, Ty-driven remodelling of the genome and improved wine yeast strains [119].
spontaneous mutation occur, as well as the avail-
ability of selection procedures to identify those
Mutagenesis and selection
variants exhibiting improved oenological traits.
The average spontaneous mutation frequency in S.
Genome renewal cerevisiae at any particular locus is approximately
For many years, strains of S. cerevisiae were 10x6 per generation [119]. The use of mutagens
isolated from vineyards and wine fermentations, greatly increases the frequency of mutations in a

Figure 7. The cassette model of mating-type switching in homothallic wine yeast strains (adapted from Pretorius and Van der
Westhuizen [119])

wine yeast population. Mutation and selection genetically stable hybrid provide a good basis for
(often through replica-plating on selective agar the introduction of recessive mutations. Mutagen-
media) appear to be a rational approach to strain esis has the potential to disrupt or eliminate
development when a large number of performance undesirable characteristics and to enhance favour-
parameters are to be kept constant while only one is able properties of wine yeasts. Though the use of
to be changed. mutagens for directed strain development is limited,
However, mutation of wine yeasts can lead to the method could be applied to isolate new variants
improvement of certain traits with the simultaneous of wine yeast strains prior to further genetic
debilitation of other characteristics [56]. Although manipulation.
mutations are probably induced with the same
frequency in haploids, diploids or polyploids, they
Hybridization
are not as easily detected in diploid and polyploid
cells because of the presence of non-mutated alleles. Mating
Only if the mutation is dominant is a phenotypic Intra-species hybridization involves the mating of
effect detected without the need for additional haploids of opposite mating-types to yield a hetero-
alterations. Therefore, haploid strains of wine zygous diploid (Figure 9). Recombinant progeny
yeasts are preferred, though not essential, when are recovered by sporulating the diploid, recovering
inducing mutations. Successful mutation and breed- individual haploid ascospores and repeating the
ing is usually associated with mutations in meiotic mating/sporulation cycle as required [56]. Haploid
segregants, where the two mating parents of a strains from different parental diploids, possessing

Figure 8. A hypothetical scheme to describe a possible succession of events leading to the replacement of one Saccharomyces
population by another through the process of `genome renewal' (adapted from Mortimer et al. [105])

different genotypes, can be mated to form a diploid and the use of hybridization techniques for devel-
strain with properties different from that of either opment of wine yeast strains has proved dif®cult.
parental strain. Thus, in theory, crossbreeding can However, this problem can be circumvented by
permit the selection of desirable characteristics and direct spore-cell mating using a micromanipulator,
the elimination of undesirable ones [8]. in which four homothallic ascospores from the
Unfortunately, many wine yeasts are homothallic same ascus are placed into direct contact with

Figure 10. The isolation of haploid strains from a homo-

thallic wine yeast by spore-cell mating. Four ascospores from
the same ascus are micromanipulated into direct contact
with heterothallic haploids yeast cells. Mating takes place
between compatible spores and cells. The resulting diploid is
sporulated. Since two spores in each ascus are homothallic
and two spores are heterothallic, stable haploids can be
isolated from the sporulated diploids (adapted from Pretor-
ius and Van der Westhuizen [119])

applied to obtain the rare hybrids formed. For

instance, industrial strains that have a defective
form of, or lack, mtDNA (respiratory-de®cient
mutants) can be force-mated with auxotrophic
haploid strains having normal respiratory charac-
teristics [56]. Mixing of these non-mating strains at
high cell density will generate only a few respira-
tory-suf®cient prototrophs. These true hybrids with
fused nuclei can then be induced to sporulate for
Figure 9. Hybridization (mating) between haploids of two further genetic analysis and crossbreeding.
opposite mating-types of wine yeast (adapted from Ham-
mond [56]) Cytoduction
Rare-mating is also used to introduce cytoplasmic
heterothallic haploid cells (Figure 10). Mating takes genetic elements into wine yeasts without the
place between compatible ascospores and cells. transfer of nuclear genes from the non-wine yeast
Elimination or inclusion of a speci®c property can parent (Figure 11). This method of strain develop-
thus be achieved relatively quickly by hybridization, ment is termed cytoduction. Cytoductants (or
provided that it has a simple genetic basis, for heteroplasmons) receive cytoplasmic contributions
example one or two genes [119]. Unfortunately, from both parents but retain the nuclear integrity of
many desirable wine yeast characteristics are speci- only one [56]. Cytoduction requires that a haploid
®ed by several genes or are the result of several gene mating strain carry the kar1 mutation; i.e. a
systems interacting with one another. mutation that impedes karyogamy (nuclear fusion)
after mating. This more speci®c form of strain
Rare mating construction can, for example, be used to introduce
Wine yeast strains that fail to express a mating-type the dsRNA determinants for the K2 zymocin and
can be force-mated (rare-mating) with haploid associated immunity into a particular wine yeast.
MATa and MATa strains (Figure 11). Typically, a Cytoduction can also be used to substitute the
large number of cells of the parental strains are mitochondrial genome of a wine yeast or to
mixed and a strong positive selection procedure is introduce a plasmid encoding desirable genetic

Figure 11. Rare-mating between industrial and laboratory

strains of wine yeast. Industrial strains that fail to show a
mating type are force-mated with haploid strains, exhibiting a
or a mating type. A large number of cells of the parental
strains are mixed and the rare hybrids or cytoductants are
selected as respiratory-suf®cient prototrophs from crosses
between a respiratory-de®cient mutant of the industrial
strains and an auxotrophic haploid laboratory strain (adapted
Figure 12. Spheroplast fusion between two different yeast
from Hammond [56])
cells is a direct asexual technique to produce either hybrids
or cytoductants. Spheroplasts are formed by removing the
characteristics into speci®c wine yeast strains. cell wall with an appropriate lytic enzyme preparation in an
Mating between strains, one of which carries the osmotically stabilized medium to prevent lysis. Spheroplasts
kar1 allele, occasionally generates progeny that from two different strains are mixed together in the
contain the nuclear genotype of one parent together presence of polyethylene glycol and calcium ions to fuse.
with an additional chromosome from the other The fused cells are allowed to regenerate their cell walls
parent [119]. The donation of a single chromosome in an osmotically stabilized agar medium (adapted from
from an industrial strain to a haploid kar1 recipient Hammond [56])
is termed `single-chromosome transfer', and is used
to examine individual chromosomes of industrial crossed, thereby extending the number of crosses
yeast strains in detail [56]. that can be done [56]. Cell walls of yeasts can be
removed by lytic enzymes in the presence of an
osmotic stabilizer to prevent osmolysis of the
Spheroplast fusion resulting spheroplasts. Spheroplasts from the differ-
Spheroplast fusion is a direct, asexual technique ent parental strains are mixed together in the
employed in crossbreeding as a supplement to presence of a fusion agent, polyethylene glycol and
mating. Like rare-mating, spheroplast fusion can calcium ions, and then allowed to regenerate their
be used to produce either hybrids or cytoductants cell walls in an osmotically-stabilized selective-agar
(Figure 12). Both these procedures overcome the medium.
requirement for opposite mating types to be Spheroplast fusion of non-sporulating industrial

yeast strains removes the natural barriers to Genetic techniques of mutation, hybridization,
hybridization. The desirable (and undesirable) cytoduction and transformation will most likely be
characteristics of both parental strains will recom- used in combination for commercial wine yeast
bine in the offspring. Cells of different levels of improvement. Strain modi®cation has been revolu-
ploidy can be fused. For instance, a diploid wine tionized by DNA transformation strategies, but it
yeast strain can be fused to a haploid strain to remains dif®cult to clone unidenti®ed genes. Thus,
generate triploid strains. Alternatively, two diploid mutation and selection will persist as an integral
wine yeasts with complementing desirable charac- part of many breeding programmes. Furthermore,
teristics can be fused to generate a tetraploid strain although recombinant DNA methods are the most
with all the genetic backgrounds of the two parental precise way of introducing novel traits encoded by
wine yeasts [119]. single genes into commercial wine yeast strains,
hybridization remains the most effective method for
improving and combining traits under polygenic
Gene cloning and transformation
control.
Clonal selection, mutagenesis, hybridization, rare-
mating and spheroplast fusion all have value in
Targets for wine yeast strain
strain development programmes, but these methods
development
lack the speci®city required to modify wine yeasts in
a well-controlled fashion [119]. It may not be
Some of the requirements listed in Table 2 are
possible to de®ne precisely the change required
complex and dif®cult to de®ne genetically without a
using these genetic techniques, and a new strain
better understanding of the biochemistry and
may bring an improvement in some aspects, while
physiology involved. To date, no wine yeast in
compromising other desired characteristics. Gene
commercial use has all the characteristics listed, and
cloning and transformation (Figure 13) offer the
it is well established that wine yeasts vary in their
possibility of altering the characteristics of wine
winemaking abilities. While some degree of varia-
yeasts with surgical precision: the modi®cation of
tion can be achieved by altering the fermentation
an existing property, the introduction of a new
conditions, a major source of variation is the
characteristic without adversely affecting other
genetic constitution of the wine yeasts.
desirable properties, or the elimination of an
unwanted trait (Figure 14). By using such proce-
Improved quality control and strain handling
dures, it is possible to construct new wine yeast
strains that differ from the original only in single Strain maintenance
speci®c characteristics. One of the main objectives for using pure cultures
In addition to the introduction of speci®c genes in winemaking is to ensure reproducible fermenta-
into wine yeasts, recombinant DNA approaches tion performance and product quality. It is there-
offer wider applicability, including: (a) ampli®ca- fore important to maintain the genetic identity of
tion of gene expression by maintaining a gene on a wine yeasts and to slow down the rate of strain
multi-copy plasmid, integration of a gene at multi- evolution caused by sporulation and mating, muta-
ple sites within chromosomal DNA or splicing a tions, gene conversions and genetic transpositions.
structural gene to a highly ef®cient promoter Total prevention of heterogeneity in pure cultures is
sequence; (b) releasing enzyme synthesis from a impossible, since homothallism, inability to spor-
particular metabolic control or subjecting it to a ulate and mate, and polyploidy (multiple gene
new one; (c) in-frame splicing of a structural gene to structure) only protect against genetic drift caused
a secretion signal to engineer secretion of a by sexual reproduction and mutation, but not
particular gene product into the culture medium; against that caused by gene conversion and trans-
(d) developing gene products with modi®ed char- position. Even stringently controlled conditions for
acteristics by site-directed mutagenesis; (e) eliminat- maintenance of culture collections (i.e. freeze-dried
ing speci®c undesirable strain characteristics by cultures, cultures preserved in liquid nitrogen or in
gene disruption; (f) incorporation of genetic infor- silica gel) will not provide full protection against
mation from diverse organisms such as fungi, genetic drift in pure yeast cultures. In fact, freeze-
bacteria, animals and plants [56,119]. drying (lyophilization) for long-term maintenance

Yeast 2000; 16: 675±729.

I. S. Pretorius

Figure 13. Gene cloning and transformation are used to introduce recombinant DNA molecules (e.g. possessing a useful

Copyright # 2000 John Wiley & Sons, Ltd.

gene) into wine yeasts
692
10970061, 2000, 8, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/1097-0061(20000615)16:8<675::AID-YEA585>3.0.CO;2-B by Cochrane Colombia, Wiley Online Library on [24/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Tailoring wine yeast for the new millennium 693

relative to that of the culture medium, thereby

counteracting the deleterious effects of dehydration
on lyophilized yeast cells.
A better understanding of precisely how yeast
cells acquire cryotolerance and osmotolerance may
lead to genetic modi®cation of starter culture
strains with greater robustness for industrial fer-
mentations. However, at present, fermentation
trials, continuous strain evaluation, and early
detection of genetic changes using comparative
molecular techniques are the only practical ways
to limit possible economic loss.

Molecular marking
The potential for exploiting genetic markers in wine
yeast identi®cation has been recognized, and delib-
erately marked oenological strains were developed
as an aid to monitor the kinetics of yeast popula-
tions during wine fermentations [116,172]. Genetic
labelling could also be regarded as a quality control
tool in general yeast culture management as well as
in trouble-shooting, particularly for wineries using
more than one yeast strain. The genomes of
commercial wine yeasts can be tagged so as to
discourage illegitimate use of (patented) commercial
wine yeast strains by `pirate' yeast and wine
producers.
The marking of wine yeast strains usually entails
the integration of speci®c genetic markers into their
genomes. This could take the form of synthetic
oligonucleotides or foreign genes of known nucleo-
tide sequences. These DNA sequences can then be
Figure 14. Homologous recombination can be used to used as `diagnostic probes' to identify speci®c wine
transfer mutations into and out of a given locus on a yeast yeast strains. In one instance a wine yeast was
chromosome by the process of gene transplacement double-marked with diuron and erythromycin resis-
tance genes [172]. A more sophisticated manner of
of yeast stock cultures causes phase transitions in marking was the expression of the Escherichia coli
membrane lipids and cell death during freeze- b-glucuronidase (GUS) gene (uidA) under control
thawing and may also induce respiratory de®cient of the yeast alcohol dehydrogenase I promoter and
variants. In an attempt to improve resistance to terminator sequences [116]. The GUS construct was
cryo-damage, anti-freeze peptides from polar ®sh integrated into the ILV2 gene of S. cerevisiae and a
were successfully expressed in S. cerevisiae [31]. simple assay procedure was devised to detect GUS
Cryoprotectants and osmoprotectants, such as activity in yeast cells or colonies. In a GMO
cellular trehalose and glycerol, also alleviate free- (`genetically modi®ed organism') risk assessment
ze±thaw and water stress [176]. Trehalose appears experiment, this yeast is currently being used to
to stabilize cell membranes of lyophilized or monitor the dissemination of transgenic yeast
cryopreserved stock cultures as well as their cellular strains on vines cultivated in a biologically con-
proteins by replacing water and forming a hydra- tained glasshouse. This will undoubtedly provide an
tion shell around proteins. Glycerol accumulation insight into the kinetics of transgenic and native
seems to control intracellular solute potential yeast populations on vines.

Improvement of fermentation performance factors affect the fermentation performance of wine

yeasts. Apart from a successful inoculation with the
The primary selection criteria applied to most strain
appropriate starter culture strain, the physiological
development programmes relate to the overall
condition of such an active dried wine yeast culture,
objective of achieving a better than 98% conversion
and its ability to adapt to and cope with nutritional
of grape sugar to alcohol and carbon dioxide, at a
de®ciency and the presence of inhibitory substances,
controlled rate and without the development of off-
are of vital importance to fermentation performance.
¯avours [61]. The growth and fermentation proper-
Successful yeast cellular adaptation to changes in
ties of wine yeasts have, however, yet to be
extracellular parameters during wine fermentation
genetically de®ned. What makes the genetic de®ni-
requires the timely perception (sensing) of chemical
tion of these attributes even more complex is the or physical environmental parameters, followed by
fact that lag phase, rate and ef®ciency of sugar accurate transmission of the information to the
conversion, resistance to inhibitory substances and relevant compartments of the cell (Figure 15).
total time of fermentation are strongly affected by Chemical signals emanating during wine fermenta-
the physiological condition of the yeast, as well as tions include the availability/concentration of cer-
by the physicochemical and nutrient properties of tain nutrients (e.g. fermentable sugars, assimilable
grape must [61]. nitrogen, oxygen, vitamins, minerals, ergosterol,
Generally, sugar catabolism and fermentation and unsaturated fatty acids) and the presence of
proceed at a rate greater than desired, and are inhibitory substances (e.g. ethanol, acetic acid, fatty
usually controlled by lowering the fermentation acids, sulphite, agrochemical residues, and killer
temperature [62]. Occasionally, wine fermentation toxins). Signals of a physical nature include factors
ceases prematurely or proceeds too slowly. Mea- such as temperature, pH, agitation and osmotic
sures to rescue such `sluggish' or `stuck' fermenta- pressure. As an example, physiological and mor-
tions include the increase of fermentation phological modi®cations in response to a limited
temperature, addition of vitamin supplements, supply of essential nutrients, such as carbon and
limited aeration by pumping over, and re- nitrogen sources, include a shift in transcription
inoculation [62]. The commercial implications of patterns, the modi®cation of the cell cycle, a change
`runaway' wine fermentations arise from the fact in budding pattern and strongly polarized growth.
that fermentor space is reduced because of foaming It is becoming clear that a complex network of
and volatile aroma compounds are lost by entrain- interconnected and cross-talking signal transduction
ment with the evolving carbon dioxide [61]. Con- pathways, relying on a limited number of signalling
versely, ®nancial losses through sluggish or modules, governs the required adaptive responses to
incomplete wine fermentations are usually attribu- changes that occur as the fermentation progresses
ted to inef®cient utilization of fermentor space and [9].
wine spoilage resulting from the low rate of This complexity explains why it is so dif®cult to
protective carbon dioxide evolution and high de®ne all the key genetic determinants of a yeast's
residual sugar content [62]. Optimal performance fermentation performance that may be candidates
of wine yeasts in white wine fermentations, con- for genetic engineering. However, general targets
ducted at cooler temperatures (10±15uC) so as to include increased tolerance to desiccation and
minimize the loss of aromatic volatiles, and red viability of active dried yeast; improved grape
wine fermentations, performed at higher tempera- sugar uptake and assimilation; increased ethanol
tures (18±30uC) to enhance extraction of anthocya- tolerance; improved nitrogen assimilation; en-
nin pigments, is therefore of critical importance to hanced resistance to microbial metabolites and
wine quality and cost-effectiveness [61]. toxins; resistance to heavy metals and agrochemical
Fermentation predictability and wine quality are residues; tolerance to sulphite; and reduced foam
directly dependent on wine yeast attributes that formation.
assist in the rapid establishment of numerical and
metabolic dominance in the early phase of wine Improved viability and vitality of active dried wine
fermentation, and that determine the ability to yeast starter cultures
conduct an even and ef®cient fermentation with a Both the genetic and physiological stability of stock
desirable residual sugar level. A wide range of cultures of seed yeast and wine yeast starter cultures

Figure 15. A schematic representation of how environmental signals are transmitted through a network of interconnected
signal transduction pathways within a yeast cell. The ability of yeast cells to adapt to changing environmental conditions during
yeast manufacturing or alcoholic fermentation is essential for their survival and physiological activity

are essential to optimal fermentation performance. cytometry which can rapidly determine cell viability
Physiological stability and `®tness' of active dried and other aspects of yeast physiology (e.g. stress
wine yeast cultures relate to the maintenance of cell responses) [176]. These techniques generally show
viability and vitality during the process of yeast varying degrees of correlation with fermentative
manufacturing, including desiccation and storage. performance and none of them, alone, can accu-
The differentiation between yeast `viability' and rately predict the physiological activity of an active
`vitality' is based upon the fact that cells which dried wine yeast starter culture.
irreversibly lose their ability to reproduce may still The manufacturers of active dried wine yeast
be capable of active metabolism. Therefore `via- starter cultures can positively in¯uence the degree
bility' is de®ned as the relative proportion of living of viability and vitality, as well as the subsequent
cells within an active dried starter culture, whereas fermentation performance of their cultures, by the
`vitality' refers to the measure of metabolic activity way they cultivate their yeasts [27]. Industrial
and relates to the ®tness or vigour of a starter cultivation of wine yeasts can have a profound
culture [176]. Yeast viability can be assessed directly effect on the microbiological quality, fermentation
by determining loss of cell reproduction/division rate, production of hydrogen sulphide, ethanol yield
(e.g. plate and slide counts) and indirectly by and tolerance, resistance to sulphur dioxide as well
assessing cellular damage (e.g. vital staining with as tolerance to drying and rehydration. For
bright-®eld or ¯uorochrome stains) or loss of example, if a protein to phosphate ratio (P2O5:N)
metabolic activity (e.g. ATP bioluminescence and of 1 : 3 in a yeast cell is exceeded it would result in
NADH ¯uorescence). Yeast vitality can be indir- an excess of water linked to the protein which
ectly assessed by measuring metabolic/fermentative would, in turn, negatively affect the drying pro-
activity (e.g. CO2 evolution in mini-scale fermenta- cedure, viability and ®nal activity of the dry yeast
tions), storage molecules (e.g. glycogen), intracellu- [27]. Due to the roles that trehalose and glycogen
lar/extracellular pH (acidi®cation power) and play in a yeast cell's response to variations in
gaseous exchange coef®cients (e.g. respiratory quo- environmental conditions, it is generally recom-
tients or RQ). Automatic in-line monitoring of mended that the manufacturers of active dried wine
yeast cell viability in fermentation plants can be yeast starter cultures cultivate their yeast in such a
achieved with electrosensors such as capacitance way that the maximum amount of these storage
probes or with ¯uorescent probes coupled with ¯ow carbohydrates is accumulated in the yeast cells.

In S. cerevisiae, trehalose (a-D-glucopyranosyl a- Glycogen, another carbohydrate reserve whose

D-glucopyranoside) is synthesized from glucose-6- accumulation by yeast propagated for drying has
phosphate and UDP-glucose by the TPS1-encoded been linked to enhanced viability and vitality upon
trehalose-6-phosphate synthetase and converted to reactivation, provides a readily mobilizable carbon
trehalose by the TPS2-encoded trehalose-6-phos- and energy source during the adaptation phase. The
phate phosphatase (Figure 16) [39]. The regulation biosynthesis of glycogen (a-1,4-glucan with a-1,6
of trehalose synthesis and degradation (by treha- branches) is effected by glycogen synthase, which
lase) is mediated by cAMP-dependent phosphoryla- catalyzes the sequential addition of glucose from
tion mechanisms [152,153]. Trehalose is associated UDP-glucose to a polysaccharide acceptor in a
with nutrient-induced control of cell cycle progres- linear a-1,4 linkage, while branching enzymes are
sion; control of glucose sensing, transport and responsible for the formation of a-1,6 branches
initial stages of glucose metabolism; as well as (Figure 16) [176]. There are two forms of glycogen
stress protection against dehydration, freezing, synthase in S. cerevisiae, Gsy1p and Gsy2p. The
heating and osmo-stress; toxic chemicals, such as GSY1 gene is expressed constitutively at a low level
ethanol, oxygen radicals and heavy metals [152]. along with growth on glucose, while the level of the
This storage carbohydrate plays an important role GSY2-encoded glycogen synthase increases at the
during sporulation, nutrient starvation, growth end of the exponential phase of growth when
resumption and growth rate. Trehalose content in glycogen accumulates [39]. This indicates that
the yeast cell is probably one of the most important GSY2 encodes the major glycogen synthase. Glyco-
factors affecting the resistance of yeasts to drying gen breakdown (catalyzed by glycogen phosphor-
and subsequent rehydration [27]. The accumulation ylase) quickly following depletion of nutrients at the
of this disaccharide on both sides of the plasma end of fermentation, is accompanied by sterol
membrane is thought to confer stress protection by formation (Figure 17) [39]. Since sterol is essential
stabilizing the yeast's membrane structure. for yeast vitality, low levels of accumulated glyco-

Figure 16. A scheme of the biosynthesis and degradation of glycogen and trehalose (adapted from Francois et al. [39])

pyruvate via the glycolytic pathway. Pyruvate is

decarboxylated to acetaldehyde, which is then
reduced to ethanol. The rate of fermentation and
the amount of alcohol produced per unit of sugar
during the transformation of grape must into wine
is of considerable commercial importance. During
wine yeast glycolysis, one molecule of glucose or
fructose yields two molecules each of ethanol and
carbon dioxide. However, the theoretical conver-
sion of 180 g sugar into 92 g ethanol (51.1%) and
88 g carbon dioxide (48.9%) could only be expected
in the absence of any yeast growth, production of
other metabolites and loss of ethanol as vapour
[13]. In a model fermentation, about 95% of the
sugar is converted into ethanol and carbon dioxide,
1% into cellular material and 4% into other
products such as glycerol [13].
The ®rst step to ensure ef®cient utilization of
grape sugar by wine yeasts is to replace any mutant
alleles of genes encoding the key glycolytic enzymes,
namely hexokinase (HXK), glucokinase (GLK),
Figure 17. An outline of the main steps leading to phosphoglucose isomerase (PGI), phosphofructo-
ergosterol biosynthesis (adapted from Walker [176]) kinase (PFK), aldolase (FBA), triosephosphate
isomerase (TPI), glyceraldehyde-3-phosphate dehy-
gen in active dried wine yeast starter cultures may drogenase (TDH), phosphoglycerate kinase (PGK),
result in insuf®cient yeast sterols which, in turn, phosphoglycerate mutase (PGM), enolase (ENO),
may impair yeast performance upon inoculation pyruvate kinase (PYK), pyruvate decarboxylase
into grape juice [176]. In this regard, it is important (PDC) and alcohol dehydrogenase (ADH). The
to note that the overexpression of the SUT1 and genes encoding PGI, TPI, PGM and PYK appear
SUT2 genes has been shown to promote the uptake to be present in single copy in a haploid genome,
of sterol from the medium under fermentative while multiple forms exist for TDH (three iso-
conditions. zymes), ENO (two isozymes) and GLK (three
Owing to its multiple roles in increasing survival isozymes) (Figure 18) [10].
of S. cerevisiae cells exposed to several physical and The assumption that an increase in the dosage of
chemical stresses, trehalose and glycogen have genes encoding these glycolytic enzymes would
important implications for the viability, vitality result in an increase in the ef®ciency of conversion
and physiological activity of active dried wine of grape sugar to alcohol has been disproved; it has
yeast starter cultures upon reactivation. Therefore, been demonstrated that overproduction of the
there is a strong incentive to develop wine yeast enzymes has no effect on the rate of ethanol
strains with a superior trehalose and glycogen formation [135]. This indicates that the step of
accumulation ability. However, due to the complex- sugar uptake represents the major control site for
ity of yeast viability, vitality and physiological the rate of glycolytic ¯ux under anaerobic condi-
activity, it is unclear at this stage whether the tions, whereas the remaining enzymatic steps do not
modi®cation of the expression levels of the TPS1, appear to be rate limiting [13]. In other words, the
TPS2, GSY1, GSY2, SUT1 and/or SUT2 genes rate of alcohol production by wine yeast is
would contribute to yeast ®tness and fermentation primarily limited by the rate of glucose and fructose
performance of starter culture strains. uptake. Therefore, in winemaking, the loss of
hexose transport towards the end of fermentation
Ef®cient sugar utilization may result in reduced alcohol yields [176].
In S. cerevisiae, glucose and fructose, the main Sugars enter yeast cells in one of three ways:
sugars present in grape must, are metabolized to simple net diffusion, facilitated (carrier-mediated)

Figure 18. Enzymatic steps of the glycolytic pathway in wine yeast (adapted from Boulton et al. [13])

diffusion and active (energy-dependent) transport. unclear, some aspects about glucose and fructose
In grape must fermentations where sugar concentra- uptake can be noted. Glucose uptake is rapid down
tions above 1 M are common, free diffusion may a concentration gradient, reaching an equilibrium
account for a very small proportion of sugar uptake and is therefore not accumulative [10,19]. Several
into yeast cells. However, since the plasma mem- speci®c, energy-dependent glucose carriers mediate
branes of yeast cells are not freely permeable to the process of facilitated diffusion of glucose and
highly polar sugar molecules, various complex proton symport is not involved. Phosphorylation by
mechanisms are required for ef®cient translocation the HXK1- and HXK2-encoded hexokinases and the
of glucose, fructose and other minor grape sugars GLK1-encoded glucokinase is linked to high-af®nity
into the cell. The hexose transporter family of S. glucose uptake. Glucose transporters, encoded by
cerevisiae consists of more than 20 proteins compris- HXT1-HXT18 and SNF3, are stereospeci®c for
ing high, intermediate and low af®nity transporters certain hexoses and will translocate glucose, fruc-
and at least two glucose sensors. Many factors affect tose and mannose. Some members of this multigene
both the abundance and intrinsic af®nities for permease family affect glucose, galactose, glucose
hexoses of these transporters present in the plasma and mannose, or glucose, fructose and galactose
membrane of wine yeast cells, among them glucose uptake, but thus far none has been described
concentration, stage of growth, presence or absence as speci®cally affecting fructose uptake [176]. It
of molecular oxygen, growth rate, rate of ¯ux appears that in S. cerevisiae, fructose is transported
through the glycolytic pathway and nutrient avail- via facilitated diffusion rather than active transport,
ability (particularly of nitrogen) [13]. whereas related species (S. bayanus and S. pastor-
Although the precise mechanisms and regulation ianus) within the Saccharomyces sensu stricto group
of grape sugar transport of wine yeast are still do possess fructose±proton symporters.

Based on the spectacular increase in the amount mance and sluggish or stuck fermentations [71,72].
of information on sugar sensing and their entry into Such problematic and incomplete fermentations
yeast cells that has come to the fore over the last occur because nitrogen depletion irreversibly arrests
few years, several laboratories have identi®ed this hexose transport. Other problems related to the
main point of control of glycolytic ¯ux as one of the nitrogen composition of grape must include the
key targets for the improvement of wine yeasts. For formation of reduced-sulphur compounds, in parti-
example, in some instances, certain members of the cular hydrogen sulphide, and the potential for-
HXT permease gene family are being overexpressed mation of ethyl carbamate from metabolically
in an effort to enhance sugar uptake, thereby produced urea [62]. A schematic overview of
improving the fermentative performance of wine nitrogen assimilation by S. cerevisiae is presented
yeast strains. However, more in-depth details are in Figure 19, while Figure 20 outlines the degrada-
required about the complex regulation of glucose tion of nitrogenous compounds in wine yeast.
and fructose uptake as well as glycolysis as it occurs Unlike grape sugars that are usually present in
in grape juice (especially in the presence of high large excess (often exceeding 20% w/v) to that
sugar levels during the early phase of fermentation needed for maximal yeast growth, the total nitrogen
and during the ®nal stages of sugar depletion content of grape juices ranges 40-fold from 60 to
coupled to nutrient limitation) before it will be 2400 mg/l and can therefore be growth-limiting [62].
possible to devise novel strategies to improve wine The assimilable content of grape must is dependent
yeast's fermentation performance and to prevent upon grape cultivar and root stock, as well as upon
sluggish or stuck fermentations. several aspects of vineyard management, including
nitrogen fertilization, berry maturation, vine water
Improved nitrogen assimilation status, soil type and fungal infection [62]. Grape
Of all nutrients assimilated by yeast during wine juices with nitrogen levels below 150 mg/l have a
fermentations, nitrogen is quantitatively second high probability of becoming problem ferments due
only to carbon [62]. Carbon-nitrogen imbalances to inadequate yeast growth and poor fermentative
and, more speci®cally, de®ciencies in the supply of activity.
assimilable nitrogenous compounds, remain the There are two basic strategies to circumvent
most common causes of poor fermentative perfor- problems linked to nitrogen de®ciency: prevention

Figure 19. A schematic overview of nitrogen assimilation by wine yeast (adapted from Walker [176])

lation in grape must is associated with grapevine

stress, in particular with low moisture, whereas high
levels of c-aminobutyrate, another nitrogen com-
pound, may be formed in the grape berries most
probably post-harvest and prior to processing of the
grapes [13].
S. cerevisiae is incapable of adequately hydrolyz-
ing grape proteins to supplement nitrogen-de®cient
musts, and relies therefore on the ammonium and
amino acids present in the juice. Wine yeasts can
distinguish between readily and poorly used nitro-
gen sources. Ammonium is the preferred nitrogen
source and, as it is consumed, the amino acids are
taken up in a pattern determined by their concen-
tration relative to yeast's requirements for biosynth-
esis and to total nitrogen availability [133]. When a
readily used nitrogen source (such as ammonium,
glutamine, or asparagines) is present, genes
involved in the uptake and catabolism of poorly
utilized nitrogen sources (including proline) are
Figure 20. A schematic representation of the degradation repressed. This nitrogen catabolite repression
of nitrogenous compounds by wine yeast (Henschke & exerted upon non-preferred nitrogenous compounds
Jiranek [62]) rigorously impairs the assimilation of proline as
well as arginine since both amino acids depend on
the proline utilization pathway [133]. Since the
of nitrogen de®ciency in grape juice by optimizing
proline content of wine is generally not less than
vineyard fertility, and more commonly, supplemen- grape juice, it appears that proline is not taken up
tation with ammonium salts such as diammonium by wine yeast under anaerobic fermentative condi-
phosphate (DAP). However, the injudicious use of tions [13]. Proline is transported into S. cerevisiae
DAP supplements often contravenes the wine by the general amino acid permease and the PUT4-
industry's desire to minimize its use of additives encoded proline-speci®c permease (Figure 21). Once
while producing wines of high quality. Moreover, inside the yeast cell, proline is converted to
excessive addition of inorganic nitrogen often glutamate in mitochondria by the PUT1-encoded
results in excessive levels of residual nitrogen, proline oxidase and PUT2-encoded pyrroline-5-
leading to microbial instability and ethyl carbamate carboxylate dehydrogenase. The expression of both
(and phosphate in the case of DAP) accumulation PUT1 and PUT2 is regulated by the PUT3-encoded
in wine [71,72,73]. The degree of supplementation activator and the URE2-encoded repressor. Ure2p
of inorganic nitrogen in grape juice is therefore represses transcription of PUT1 and PUT2 under
often regulated. This implies that knowledge of the nitrogen-limiting conditions, while the GLN3-
nitrogen content of grape juice and the requirement encoded regulator has no effect on these genes
for nitrogen by yeast are important considerations [133].
for optimal fermentation performance and the Since wine yeast strains vary widely in their
production of wines that comply with the demands nitrogen requirement, an obvious target for strain
of regulatory authorities and consumers. improvement is to select or develop starter strains
The major nitrogenous compounds in the average that are more nitrogen ef®cient for use in low-
grape must are proline, arginine, alanine, gluta- nitrogen musts [61,71,72,73]. To achieve this, a
mate, glutamine, serine and threonine [13], while the thorough understanding of the regulation of nitro-
ammonium ion levels may also be high, depending gen assimilation by yeast under fermentative condi-
on grape variety and vineyard management. Proline tions is required. In an effort to develop wine yeast
and arginine account for 30±65% of the total amino strains that are relieved from nitrogen catabolite
acid content of grape juices. High proline accumu- repression and that are capable of utilizing proline

Figure 21. A schematic representation of the pathway for proline degradation in wine yeast (adapted from Boulton et al.
[13])

more ef®ciently under winemaking conditions, a underlying cause of sluggish or stuck fermentations.
mutant containing a ure2 recessive allele was Apart from the inhibitory effect of excessive sugar
constructed [133]. It was demonstrated that this content on yeast growth and vini®cation fermenta-
mutation strongly deregulates the proline utilization tion, the production of excessive amounts of
pathway, thereby improving the overall fermenta- ethanol, coming from harvest of over-ripe grapes,
tion performance of the ure2-carrying yeast. This is known to inhibit the uptake of solutes (e.g. sugars
may be the ®rst step towards the development of and amino acids) and to inhibit yeast growth rate,
wine yeasts that are able to ef®ciently assimilate the viability and fermentation capacity [13].
abundant supply of proline in grape juice under The physiological basis of ethanol toxicity is
fermentative conditions. complex and not well understood, but it appears
that ethanol mainly impacts upon membrane
Improved ethanol tolerance structural integrity and membrane permeability
The winemaker is confronted with the dilemma [13]. The chief sites of ethanol action include the
that, while ethyl alcohol is the major desired yeast cell's plasma membrane, hydrophobic pro-
metabolic product of grape juice fermentation, it is teins of mitochondrial membranes, nuclear mem-
also a potent chemical stress factor that is often the brane, vacuolar membrane, endoplasmic reticulum

and cytosolic hydrophilic proteins [176]. Increased ¯uidity to synthesis of detoxi®cation enzymes.
membrane ¯uidity and permeability due to ethanol Responses include a decrease in membrane satu-
challenge seem to result in futile cycling of protons rated fatty acids (e.g. palmitic acid); an increase
and dissipation of ATP energy. However, the in membrane unsaturated long-chain fatty acids
dissipation of the proton gradient across the (e.g. oleic acid); phosphatidylinositol biosynthesis
membrane and ATP is not only affected by (thereby increasing the phospholipid : protein ratio
increased permeability to protons, but ethanol may in the membrane); elevated levels of cellular
also directly affect the expression of the ATPase- trehalose that neutralize the membrane-damaging
encoding genes (PMA1 and PMA2) and membrane effects of ethanol; stimulation of stress protein
ATPase activity [61]. This explains the interference biosynthesis; enhanced mitochondrial superoxide
of ethanol with energy-coupled solute transport in dismutase activity that countereffects ethanol-
yeast cells. induced free radical synthesis; increased synthesis
Several intrinsic and environmental factors are of cytochrome P450, alcohol dehydrogenase activity
known to synergistically enhance the inhibitory and ethanol metabolism [13,176].
effects of ethanol. These factors include high From this, it is clear that the genetics of ethanol
fermentation temperatures, nutrient limitation tolerance are polyvalent and very complex. It is
(especially oxygen, nitrogen, lipids and magnesium speculated that more than 250 genes are involved in
ions) and metabolic by-products, such as other the control of ethanol tolerance in yeast [13].
alcohols, aldehydes, esters, organic acids (especially Nevertheless, some reports claim that continuous
octanoic and decanoic acids), certain fatty acids and culture of yeasts in a feedback system in which the
carbonyl and phenolic compounds [32]. By manip- ethanol was controlled by the rate of carbon
ulating the physicochemical environment during the dioxide evolution, enabled the selection of viable
cultivation and manufacturing of active dried wine mutants with improved ethanol tolerance and
yeast starter cultures and during the actual vini®ca- fermentation capabilities [14]. Dramatic increases
tion process, the yeast cells' self-protective adapta- in ethanol tolerance, however, seem to elude
tions can be promoted. Prior exposure of yeast cells researchers. It therefore appears that, for the time
to ethanol (physiological pre-conditioning) elicits being, ethanol tolerance in wine yeasts will be
adaptive stress responses that confer a degree of addressed by `cell engineering' rather than `genetic
resistance to subsequent exposure to high levels of engineering'.
ethanol. Furthermore, osmotic pressure, media
composition, modes of substrate feeding and by- Increased tolerance to antimicrobial compounds
product formation play important roles in dictating Besides the various yeast metabolites such as
how yeast cells tolerate ethanol during vini®cation alcohols, acetic acid and medium chain length
[176]. Most of the so-called survival factors (e.g. fatty acids (e.g. decanoic acid) that can interfere
certain unsaturated long chain fatty acids and with ef®cient grape must fermentations, there are
sterols) are formed only in the presence of mole- several other antimicrobial compounds that can
cular oxygen, which in part explains the great impede the fermentation performance of wine
success in the use of commercial starter cultures yeasts. These compounds include killer toxins,
that are cultivated under highly aerobic conditions chemical preservatives (especially sulphite) and
and in low glucose concentrations. These starter agrochemicals containing heavy metals (e.g.
yeast cells contain high levels of the survival factors copper). Since S. cerevisiae strains vary widely in
that can be passed onto the progeny cells during the their ability to resist or tolerate these compounds,
six or seven generations of growth in a typical wine the differences may lend themselves as targets for
fermentation [13]. strain development.
Wine yeast strains usually contain higher levels of Killer toxins are proteins produced by some
survival factors than non-wine Saccharomyces yeasts that are lethal to sensitive wine yeast strains.
strains [13]. The physiological response of wine The killers themselves, however, are immune to
yeast to ethanol challenge is also greater than is the these mycovirus-associated toxins. It remains con-
case with non-wine strains. These defensive adapta- troversial whether the growth and zymocidal activ-
tions of wine yeasts, conferring enhanced ethanol ity of some wild killer yeasts have the potential to
tolerance, range from alterations in membrane delay the onset of fermentation, cause sluggish or

stuck fermentations and produce wines with suppress the growth of unwanted microbes, and
increased levels of acetaldehyde, lactic acid, acetic tolerance to sulphite forms the basis of selective
acid and other undesirable sensory qualities [138]. implantation of active dried wine yeast starter
However, it appears that under certain conditions cultures into grape must [61]. Membrane transport
(e.g. inef®cient inoculation with highly sensitive of sulphite in wine yeasts is by simple diffusion of
starter cultures in low-nitrogen musts) that favour liberated sulphur dioxide rather than being carrier-
the development of killer yeast contamination of mediated [176]. SO2 dissociates within the cell to
grape juice, potent zymocidal yeasts may indeed SO32x and HSO3x and the resulting decline in
contribute to incomplete fermentations. While intracellular pH forms the basis of the inhibitory
zymocidal toxins produced by killer strains (K1, action. Although S. cerevisiae tolerates much higher
K2, K3, K28) of S. cerevisiae are lethal only to levels of sulphite than most unwanted yeasts and
sensitive strains of the same species, those produced bacteria, excessive SO2 dosages may cause sluggish
by non-Saccharomyces killer species (K4 to K11) or stuck fermentations [13].
may be toxic to a wider range of wine yeast strains Wine yeasts vary widely in their tolerance of
and other wild yeasts. The killing of sensitive wine sulphite, and the underlying mechanism of toler-
yeasts by the two S. cerevisiae killer toxins that ance as well as the genetic basis for resistance are
function at wine pH, K2 and K28, occurs via two still unclear. Once these have been better de®ned, it
different mechanisms: the K2 toxin acts as an may be advantageous to engineer wine yeast starter
ionophore affecting membrane permeability and strains with elevated SO2 tolerance. This, however,
leakage of protons, potassium cations, ATP and should not replace efforts to lower the the levels of
amino acids, whereas the K28 toxin inhibits DNA chemical preservatives in wine.
synthesis [136]. Improper application of copper-containing
An unfortunate consequence of ignorance regard-
fungal pesticides (copper oxychloride) to control
ing the role of killer yeasts in wine fermentations is
downy mildew (Plasmopara viticola) and, to a lesser
that some winemakers use co-cultures to inoculate
extent, dead arm (Phomopsis viticola) and anthrac-
fermentations, one strain being a killer and the
nose (Gloeosporium ampelophagum) could lead to
other a sensitive strain. The advantage of using
copper residues in musts that may cause lagging
killer or neutral wine yeasts should therefore not be
fermentation and affect wine quality detrimentally
underestimated. For this reason, the aim of many
[158]. Copper toxicity towards wine yeast cells
strain development programmes is to incorporate
involves the disruption of plasma membrane integ-
the mycoviruses from killer yeasts into commercial
rity and perhaps also intracellular interaction
wine strains. Mycoviruses are readily transmitted by
cytoplasmic fusion and have been used to transfer between copper and nucleic acids and enzymes [5].
the killer character into commercial yeasts. In most Several copper uptake, ef¯ux and chelation strate-
cases, however, the mixing of the genomes of gies have been developed by yeasts to control
commercial strains and donor strains containing copper ion homeostasis. One such protective
the killer character would prove undesirable even mechanism relates sequestration of copper by the
though repeated back-crossing could be used to CUP1-encoded copper-binding protein, copper-
minimize the unwanted effects. chelatin. Such metallothionein proteins are gener-
One way to circumvent this problem is cytoduc- ally synthesized when S. cerevisiae cells are exposed
tion between a donor killer strain de®cient in to potentially lethal levels of toxic metals. The
nuclear fusion and a haploid derived from a copper resistance level of a given yeast strain
commercial wine strain. Another means is to cross correlates directly with the CUP1 copy number
a haploid derived from a killer wine yeast with [38]. One way to engineer wine yeasts resistant to
haploid cells or ascospores from a sensitive wine copper would be to clone and integrate the CUP1
yeast [159]. An alternative to the use of cytoduction gene at multiple sites into their genomes [59]. This
and hybridization to develop broad spectrum would enable the wine yeast to tolerate higher
zymocidal resistance into wine yeasts would be to concentrations of copper residues in musts. Copper-
clone and introduce the toxin-immunity genes from resistant wine yeasts should, however, not be used
non-Saccharomyces killer yeasts into wine yeasts. to encourage disrespect for recommended fungicide
Sulphur dioxide is widely used in wineries to withholding periods.

Reduced foam formation generally thought to produce insigni®cant composi-

Excessive foaming, caused by certain wine yeast tional changes, ®ning is intended to bring about
strains during the early stages of wine fermentation, changes that will prevent further precipitation.
can result in the loss of grape juice. Moreover, Fining can therefore be used to modify the sensory
formation of a froth-head can reduce fermenter attributes of wine even though existing clarity may
capacity, as part of the fermentation vessel may not be a problem.
have to be reserved to prevent the froth from Fining reactions include the removal of colloids
spilling over [142,156]. In some cases, foaming may such as partially soluble, haze-forming proteins,
also reduce the suspended yeast cell density in the ®lter-clogging polysaccharides as well as complexes
fermenting must [61]. between proteins and phenols, and between proteins
Froth generation varies widely among S. cerevi- and polysaccharides. The removal of tannic or
siae strains. Genetic analysis of the foaming brown polymeric phenols is usually achieved by
characteristic suggests that this trait is under the proteinaceous ®ning agents (e.g. casein, isinglass,
control of at least two dominant genes, FRO1 and albumin and gelatin), whereas the depletion of
FRO2 [74,154,155]. Apparently these genes, located monomeric and small polymeric phenols is reached
on chromosome VII, code for proteins that interact by treatment with polyamide materials (e.g. poly-
with the grape juice thereby causing foaming [104]. vinylpolypyrrolidone or PVPP) [13]. Haze-forming
Several researchers have successfully used intra- proteins are removed by exchanging clays such as
genomic hybridization to cross out the genes that bentonites, while the removal of ®ne colloidal
are responsible for foaming [33,171]. However, the particles and incipient precipitates is achieved by
FRO1 and FRO2 genes have yet to be cloned and the sieving effect of other gelatinous materials [13].
their encoded proteins characterized. Once this is The slow development of protein hazes in white
done, the regulation of FRO1 and FRO2 can wine is considered to be the next most common
be unravelled. Gene disruption through targeted physical instability after the precipitation of potas-
homologous recombination would then also sium bitartrate. Protein instability, occurring after
become possible, which would eliminate the foam- bottling and shelf storage, is induced by high
ing characteristic of wine yeast strains without ethanol and low pH. Protein haze is not dependent
changing the remainder of their genetic back- upon total protein content but rather upon speci®c
grounds. grape-derived proteins, whose size or isoelectric
properties make them particularly susceptible to
solubility limitations [13]. Protein instability is
Improvement of processing ef®ciency
presumably associated with pathogenesis-related
Improved protein and polysaccharide clari®cation (PR) proteins produced in grape berries when
The main objectives of ®ning and clari®cation challenged by fungal attack. Although the removal
during wine processing include the removal of of these haze-forming proteins by bentonite treat-
excess levels of certain components to achieve ment is effective, the non-speci®c nature of this
clarity and ensure the physicochemical stability of diatomaceous clay can result in the loss of impor-
the end product. The need for ®ning and clari®ca- tant aroma and ¯avour compounds, thereby alter-
tion depends on the composition of the must and ing the sensory characteristics of the wine.
the winemaking practices that have been employed. Furthermore, bentonite ®ning is an expensive and
Fining of wine entails the deliberate addition of an laborious practice that generates large volumes of
adsorptive compound, followed by the settling or lees for disposal and causes a 5±20% loss of wine
precipitation of partially soluble components from [15].
the wine [13]. Further clari®cation is usually To omit the bentonite treatment, an application
achieved by sedimentation and racking, centrifu- of an appropriate acid protease to hydrolyze the
gation and ®ltration. Wine ®ltration involves a wide grape PR-proteins has been suggested. However,
range of objectives, from the partial removal of the search for fungal enzymes that could degrade
large suspended solids by various grades of diato- these haze-forming proteins has so far remained
maceous earth or ®lter sheets to the complete unsuccessful.
retention of microbes by perpendicular ¯ow poly- We have investigated the feasibility of engineer-
meric membranes [13]. While clari®cation of wine is ing a proteolytic wine yeast that could facilitate

protein haze reduction. Proteolytic activities of S. neither is it due to covalently bound sugars
cerevisiae include the acid endoprotease, protease (glycosylation) or associated phenolic compounds.
ysc A; the endo serine-sulphydryl endoprotease, It appears that protein conformation bestows
protease ysc B; the serine exopeptidase, carboxy- stability to these PR-proteins and that appropriate
peptidase ysc Y; and the four metallo exo- viticultural practices, rather than post-harvest pro-
peptidases, namely carboxypeptidase ysc S, cessing, may hold the key to controlling the
aminopeptidase ysc I, aminopeptidase ysc II and concentrations of protein in wine.
aminopeptidase ysc Co [15]. However, the vacuolar Like grape proteins, polysaccharides also in¯u-
protease A, encoded by the PEP4 gene, is the only ence the clari®cation and stabilization of must and
one that is active at the low pH of wine. wine. Polysaccharides, found in wines at levels
Furthermore, it has been reported that the pro- between 300 and 1000 mg/l, originate in the grape
longed storage of wine on the lees after the itself, the fungi on the grape and the microorgan-
completion of the alcoholic fermentation renders a isms present during winemaking. The main poly-
wine more protein stable. This phenomenon was saccharides responsible for turbidity, viscosity and
attributed to the action of proteinase A during ®lter stoppages are pectins, glucans (a component
autolysis. of cellulose) and, to a lesser extent, hemicellulose
This acid endoprotease is synthesized as a (mainly xylans). Grape pectic substances are hetero-
preprotein in S. cerevisiae. The prepeptide is cleaved polysaccharides consisting of partially methylated
early in the secretory pathway and the propeptide is a-1,4-D-galacturonan chains linked to L-rhamnopyr-
cleaved upon entrance of proteinase A into the anose units carrying neutral side chains [120].
vacuole. The propeptide contains the vacuolar Glucans such as b-1,3-1,6-glucan produced by the
targeting information and serves as an inhibitor to
grey mould Botrytis cinerea in botrytized grape
keep protease A inactive during transport through
juice, comprise b-D-glucopyranose units with a high
the secretory pathway. The PEP4 gene was cloned
degree of polymerization [120]. Xylans are complex
and expressed in a wine yeast by using different
polymers consisting of a b-D-1,4-linked xylopyra-
combinations of several promoter, leader and
noside backbone substituted with acetyl, arabinosyl
termination sequences. Northern blot analysis indi-
and glucuronosyl side chains [120]. Enzymatic
cated the presence of these PEP4 transcripts in the
breakdown of pectic polymers occurs by the de-
various transformants. Upon replacing the PEP4-
esterifying action of pectinesterase, releasing the
encoded prepro-region (vacuolar localization signal)
methyl ester groups of the pectin molecule, and by
with the yeast mating pheromone a-factor (MFa1-
encoded) prepro-region (secretion signal), no extra- the hydrolase or lyase action of the depolymerases
cellular protease activity was detected. However, (pectin lyase, pectate lyase and polygalacturonase),
Western blot analysis revealed the presence of splitting the a-1,4-glycosidic linkages in the poly-
extracellular protease A when the PEP4 gene was galacturonate chain. Glucans are hydrolyzed by
overexpressed under control of the constitutive endoglucanases (b-1,4-D-glucan glucanohydrolase),
yeast alcohol dehydrogenase I promoter (ADH1P) exoglucanases (b-1,4-D-glucan cellobiohydrolase),
and terminator (ADH1T) signals. Casein agar test cellodextrinases and cellobiases (b-1,4-D-glucoside
plates con®rmed that these transformants secreted glucohydrolase, a member of the b-glucosidase
biologically active protease A. Overexpression of family). Enzymatic degradation of xylans is cata-
PEP4 in S. cerevisiae seems to have saturated the lysed by the synergistic actions of endo-b-1,4-D-
vacuolar targeting machinery and resulted in secre- xylanases, b-D-xylosidases and a-L-arabinofuranosi-
tion of biologically active protease. dases [120].
Later, it became known, however, that bentonite The endogenous pectinase, glucanase, xylanase
®ning is unlikely to be replaced by the addition of and arabinofuranosidase activities of grapes and
proteolytic enzymes to wine or by engineering a yeasts are often neither ef®cient nor suf®cient under
proteolytic wine yeast. This is not because these winemaking conditions to prevent polysaccharide
proteases are inactive in must and wine, but because hazes and ®lter stoppages [15]. Industrial enzyme
the haze-forming proteins in wine are inherently preparations are widely used to supplement these
resistant to proteolysis. Their resistance is not due polysaccharide-degrading activities [20]. Most com-
to protection by other wine components in wine, mercial pectinase and glucanase preparations are

derived from Aspergillus and Trichoderma, respec- gene (ABF2) [22,23,81,82,96,103]. The xlnA and
tively [15]. xlnB genes from Aspergillus nidulans were also
Since the addition of these commercial enzyme reported to be expressed in S. cerevisiae [114].
preparations can be quite expensive, some research- It is hoped that these efforts will lay the
ers are looking at the native pectinases and foundation for developing pectolytic, glucanolytic
glucanases of S. cerevisiae. Certain strains of S. and xylanolytic wine yeasts that would contribute
cerevisiae were reported to produce pectinesterase, to the clari®cation of wine and replace or reduce the
polygalacturonase and pectin lyase [48], while all levels of commercial enzyme preparations needed.
strains of S. cerevisiae show some form of glucanase Furthermore, polysaccharide-degrading enzymes
activity [120]. All of these glucanase genes have secreted by wine yeasts may also improve liquefac-
been cloned and characterized. The EXG1 (BGL1) tion of the grapes, thereby increasing the juice yield.
gene encodes a protein whose differential glycosyla- Since many of the ¯avour compounds are trapped
tion accounts for the two main extracellular exo-b- in the grape skins, pectolysis, glucanolysis and
1,3-glucanases (EXGI and EXGII), while EXG2 xylanolysis may release more of these aromatic
encodes a minor exo-b-1,3-glucanase (EXGIII). compounds during skin contact in red wine fermen-
BGL2 encodes a cell wall associated endo-b-1,3- tations and make a positive contribution to the
glucanase, while SSG1 (SPR1) codes for a sporula- wine bouquet [119].
tion-speci®c exo-b-1,3-glucanase.
Since these endogenous pectinolytic and glucano- Controlled cell sedimentation and ¯occulation
lytic activities of S. cerevisiae are not suf®cient to S. cerevisiae adapts its growth pattern in response
avoid clari®cation and ®ltration problems, we have to a wide range of physical and chemical signals
introduced a wide variety of heterologous pectinase, sensed by the cells. These changes include yeast
glucanase, xylanase and arabinofuranosidase genes ®lamentation, agglomeration, ¯occulation and ¯o-
into S. cerevisiae. A pectinolytic wine yeast was tation, in¯uenced by a variety of genetic, physiolo-
developed by co-expressing the Erwinia chrys- gical and biochemical factors which are not always
anthemi pectate lyase gene (pelE) and the Erwinia clearly understood.
carotovora polygalacturonase gene (peh1) in S. Filamentous growth and the formation of pseu-
cerevisiae [85,86,87]. Both these bacterial genes dohyphae and hyphal-like structures often result in
were inserted in the ADH1P-MFa1S-ADH1T yeast dimorphism, known to be affected by nutrient
expression-secretion cassette. The pectinase gene limitation and the availability of oxygen [52,177].
cassette, consisting of ADH1P-MFa1S-pelE-ADH1T The phenomenon of agglomeration involves an
(designated PEL5) and ADH1P-MFa1S-peh1- extensive, non-reversible cell aggregation process;
ADH1T (designated PEH1) enabled wine yeast ¯occulation refers to an asexual cellular aggregation
strains of S. cerevisiae to degrade polypectate when yeast cells adhere, reversibly, to one another
ef®ciently [87]. Likewise, our laboratory has also to form microscopic ¯ocs which sediment out of
constructed a glucanase gene cassette comprising suspension. Yeast cell ¯otation, the converse of
the Butyrivibrio ®brisolvens endo-b-1,4-glucanase ¯occulation, de®nes the ability of non-aggregated
gene (END1), the Bacillus subtilis endo-b-1,3-1,4- yeast cells to trap CO2 bubbles in a fermenting
glucanase gene (BEG1), the Ruminococcus ¯ave- liquid and form a ®lm or vellum at the top of
faciens cellodextrinase gene (CEL1), the Phan- fermentation vessels. All these phenomena are
erochaete chrysosporium cellobiohydrolase gene highly relevant to the production of several yeast-
(CBH1) and the Saccharomycopsis ®buligera cello- fermented products. Grittiness, caused by agglom-
biase gene (BGL1) [162,163,164,165,166]. Upon erated baker's yeast strains and the concomitant
introduction of this glucanase gene cassette, S. appearance of granular material, is detrimental,
cerevisiae transformants were able to degrade since it results in inadequate mixing into bread
glucans ef®ciently. We have also successfully dough leading to limited leavening ability. Yeast
expressed in S. cerevisiae the endo-b-xylanase ¯occulation, on the other hand, is often exploited in
genes from Aspergillus kawachii (XYN1), Aspergil- the production of lager beer and wine (especially
lus niger (XYN4 and XYN5) and Trichoderma reesei bottle-fermented sparkling wine). The ¯ocs that
(XYN2), as well as the Bacillus pumilus xylosidase settle to the bottom of the fermentor by the end of
(XLO1) and the A. niger a-L-arabinofuranosidase the primary fermentation can easily be removed

from the fermentation product, thereby allowing for MUC1-mediated ®lamentous growth and STA1-
rapid and ef®cient clari®cation and reduced hand- STA3-facilitated starch assimilation [46,47,88,181].
ling of wine. Yeast ¯otation is important for the The overall structure of the FLO11/MUC1-
production of traditional ale beer by top-fermenting encoded cell wall associated protein is similar to
strains, and ¯or sherry by vellum-forming strains. those of the Flo1p, Flo5p and Flo10p. All these
Flocculation in S. cerevisiae is thought to be ¯occulins comprise an amino-terminal domain
mediated by speci®c calcium-activated lectins, the containing a hydrophobic signal sequence and
FLO-gene encoded ¯occulins which are surface a carboxyl-terminal domain with homology to
glycoproteins capable of directly binding manno- the glycosyl-phosphatidyl-inositol-anchor-containing
proteins of adjacent cells [151]. Proteinaceous proteins, separated by a central domain of highly
`hairy' protrusions called `®mbriae' often emanate repeated sequences rich in serine and threonine
from the cell surface of ¯occulant S. cerevisiae cells residues [88,93]. Of all the ¯occulation genes, FLO1
[139]. Cell surface charge and hydrophobicity have is considered to be the best studied and perhaps
also been implicated in a primary or complementary most important, capable of conferring ¯occulation
role with lectins to facilitate the onset of ¯occula- when transformed into non-¯occulant S. cerevisiae
tion [150,183]. Environmental factors that may strains [56,69,178].
in¯uence the level of ¯occulant S. cerevisiae strains Regulated expression of the ¯occulation genes is
include temperature, pH, calcium and zinc ions, important in wine production, because yeast must
certain inhibitors, oxygen content, sugar and inosi- perform con¯icting roles; during fermentation of
tol depletion, growth phase and cell density [57]. grape must, a high suspended yeast count ensures a
Several dominant, semi-dominant and recessive rapid fermentation rate, while at completion of
genes are known to be involved in ¯occulation, and sugar conversion, ef®cient settling is needed to
distinct ¯occulation phenotypes have been identi®ed minimize problems with wine clari®cation [61].
based on their sensitivities to sugar inhibition Moreover, ¯occulation has also been linked to
and proteolylitic enzymes [149]. These phenotypes, enhanced ester production. For these reasons we
designated Flo and NewFlo, also display different have linked the FLO1 gene to the HSP30 gene
sensitivities to yeast growth conditions, most nota- promoter [170]. It is known that the HSP30
bly temperature, acidity of the culture medium and promoter induces high gene expression during late
glucose availablity [143,144,148]. The ¯occulation stationary phase [130]. We have shown that the
genes include FLO1, FLO2, ¯o3, FLO4, FLO5, ¯o6, expression of FLO1, linked to the late-fermentation
¯o7, FLO9, FLO10 and FLO11/MUC1 [88,93,151]. HSP30 promoter, can be induced by a heat-shock
The FLO11/MUC1 gene was also shown to be treatment, con®rming that controlled ¯occulation is
involved in pseudohyphal development and invasive indeed possible during fermentation [170].
growth [88], while FLO8 was reported to encode a
transcriptional activator of FLO1 and FLO11/ Controlled cell ¯otation and ¯or formation
MUC1 [46,47,92]. Apparently, Flo8p inactivates Flor sherry is produced using certain strains of S.
the TUP1 and CYC8/SSN6-encoded cascade which cerevisiae (formerly known as S. beticus and S.
represses ¯occulation and pseudohyphal differentia- capensis) capable of forming a yeast ®lm on the
tion in certain strains [76,92]. However, when the surface (¯or) of a base wine exposed to air. These
expression of FLO1 and FLO11/MUC1 was inves- strains are known for their high ethanol tolerance,
tigated in 25 commercial wine yeast strains, it was superior ®lm-forming ability and desirable oxidative
found that they are not co-regulated [18]. Further- metabolism [61]. Flor sherry is characterized by a
more, it is unclear what the advantage would be to high ethanol (>15%), low sugar and high aldehyde
the yeast cell of co-regulating the expression of content. The typical nutty character of ¯or sherry
FLO11/MUC1 and three glucoamylase-encoding can be ascribed to the partial oxidation of ethanol
genes (STA1, STA2 and STA3) involved in starch to acetaldehyde and to the speci®c contribution
metabolism [174]. In fact, the unusually long made by the ¯or strains of S. cerevisiae.
(>3 kb) promoter sequences of FLO11/MUC1 and Although initial reports indicated that the
STA1-STA3 are almost identical, and we have vellum-forming trait segregated according to Men-
shown that several transcriptional activators (e.g. delian rules in asci of sherry yeasts, it now seems
Flo8p, Msn1p and Mss11p) co-regulate FLO11/ unlikely that the ¯or trait is controlled by a single

dominant gene. Several genes encoding cell-wall- exceptions (e.g. terpenes in the aromatic varieties
associated, hydrophobic proteins have been impli- and alkoxypyrazines in the vegetative or herbaceous
cated in vellum formation. Since few yeasts capable cultivars), perceived ¯avour is the result of speci®c
of growth on wine are suitable for ¯or sherry ratios of many compounds rather than being
production, the genotype of sherry yeasts is likely to attributable to a single `impact' compound
be more complex than originally expected. How- [21,107]. In wines and brandies, the major products
ever, once the most important genes responsible for of yeast fermentation, esters and alcohols
®lm formation and the characteristic nutty bouquet (Figure 22), contribute to a generic background
have been identi®ed, the relevant genetic and ¯avour, whereas subtle combinations of trace
metabolic mechanisms that would allow for con- components derived from the grapes usually elicit
trolled vellum formation in ¯or sherry production the characteristic aroma notes of these complex
may be brought to light. beverages.

Improvement of wine ¯avour and other

Enhanced liberation of grape terpenoids
sensory qualities
The varietal ¯avour of grapes is mainly determined
The single most important factor in winemaking is by the accumulation and pro®le of volatile second-
the organoleptic quality of the ®nal product. A ary metabolites in V. vinifera [61]. However, a high
wine's bouquet is determined by the presence of percentage of these metabolites occur as their
desirable ¯avour compounds and metabolites in a respective, non-volatile O-glycosides. Several stu-
well-balanced ratio, and the absence of undesirable dies have shown that increased enzymatic hydro-
ones. lysis of aroma precursors present in grape juice can
Many variables contribute to the distinctive liberate the aglycone to intensify the varietal
¯avours of wine, brandy and other grape-derived character of wines [15]. For instance, terpenols
alcoholic beverages. Grape variety, viticultural such as geraniol and nerol can be released from
practices, and terroir affect vine development and terpenyl-glycosides by the grape-derived b-D-glyco-
berry composition, and exert major in¯uences on sidase activity present in muscat grape juice.
the distinctiveness of wine and brandy ¯avours However, grape glycosidases are unable to hydro-
[21,107]. Oenological practices, including the yeast lyze sugar conjugates of tertiary alcohols such as
and fermentation conditions, have a prominent linalool [15]. Moreover, these grape enzyme activ-
effect on the primary ¯avours of V. vinifera wines. ities are inhibited by glucose and exhibit poor
The volatile pro®le of wines is dominated by those stability at the low pH and high ethanol levels of
components that are formed during fermentation, wine [61]. Thanks to these limiting characteristics of
since these compounds are present in the highest grape-derived glycosidases and the fact that certain
concentrations [54]. In brandy, the character is processing steps during the clari®cation of must and
further changed as distillation alters the absolute wine profoundly reduce their activity, these endo-
and relative amounts of volatiles. genous enzymes of grapes have a minimal effect in
The ¯avours of wine and brandy immediately enhancing varietal aroma during winemaking [15].
after fermentation or distillation only approximate As an alternative to the inef®cient grape glycosi-
those of the ®nished product [16]. After the sudden dases, aroma-liberating b-glucosidases from Asper-
and dramatic changes in composition during fer- gillus and other fungal species have been developed
mentation and distillation, chemical constituents as components of commercial enzyme preparations
generally react slowly during aging to move to their to be added to fermented juice (as soon as the
equilibria, resulting in gradual changes in ¯avour. glucose has been consumed by the yeast) or to
The harmonious complexity of wine and brandy young wine [15]. The addition of exogenous enzyme
can subsequently be further increased by volatile preparations to wine, however, is an expensive
extraction during oak barrel aging [16]. practice, and is viewed by many purists as an
Despite the extensive information published on `arti®cial' or `unnatural' intervention by the wine-
¯avour chemistry, odour thresholds and aroma maker. This has led to renewed interest in the more
descriptions, the ¯avour of complex products such active b-glucosidases produced by certain strains of
as wine and brandy cannot be predicted. With a few S. cerevisiae and other wine-associated yeasts such

Figure 22. A schematic representation of derivation of ¯avour compounds from sugar, amino acids and sulphur metabolism
by wine yeast (adapted from Henschke and Jiranek [62])

as Candida, Hanseniaspora and Pichia (formerly yeast sterol biosynthetic pathway are able to
Hansenula) species. produce monoterpenes (geraniol, citronelol and
Unlike the grape b-glycosidases, yeast b-glucosi- linalool) similar to those of the muscat grape
dases are not inhibited by glucose, and the libera- cultivars.
tion of terpenols during fermentation can be
ascribed to their action on the terpenyl-glycoside Enhanced production of desirable volatile esters
precursors [15]. Since these b-glucosidases are During the primary or alcoholic fermentation of
absent in most S. cerevisiae starter culture strains, grape sugars, wine yeast produces ethanol, carbon
we have functionally expressed the b-glucosidase dioxide and a number of by-products including
gene (BGL1) of the yeast Saccharomycopsis ®buli- esters, of which alcohol acetates and C4±C10 fatty
gera in S. cerevisiae [166]. When the b-1,4-glucanase acid ethyl esters are found in the highest
gene from Trichoderma longibrachiatum was concentration in wine and brandy (Figure 22)
expressed in wine yeast the aroma intensity of [63,64,101,108,145,147]. Although these compounds
wine increased, presumably due to the hydrolysis of are ubiquitous to all wines and brandies, the level of
glycosylated ¯avour precursors [115]. Likewise, we esters formed varies signi®cantly. Apart from
have overexpressed the S. cerevisiae exo-b-1,3- factors such as grape cultivar, rootstock and grape
glucanase gene (EXG1) and introduced the endo-b- maturity, the ester concentration produced during
1,4-glucanase gene (END1) from Butyrivibrio ®bri- fermentation is dependent on the yeast strain,
solvens, the endo-b-1,3-1,4-glucanase (BEG1) from fermentation temperature, insoluble material in the
Bacillus subtilis and the a-arabinofuranosidase grape must, vini®cation methods, skin contact,
(ABF2) in S. cerevisiae [164,165,166]. Further trials must pH, the amount of sulphur dioxide, amino
are under way to determine the effect of these acids present in the must and malolactic fermenta-
transgenic yeasts on the varietal character of muscat tion [21,65]. Furthermore, the ester content of
wines. distilled beverages is greatly dependent on whether
Another intriguing discovery gives yeast the the yeast lees is present during distillation
potential to modify the `impact' compound pro®le [16,97,108].
of low-¯avoured wines [70]. Certain mutants of the The characteristic fruity odours of wine are

primarily due to a mixture of hexyl acetate, ethyl Optimized fusel oil production
caproate and caprylate (apple-like aroma), isoamyl Alcohols with carbon numbers greater than that of
acetate (banana-like aroma), and 2-phenylethyl ethanol, such as isobutyl, isoamyl and active amyl
acetate (fruity, ¯owery ¯avour with a honey note). alcohol, are termed fusel oils. These higher alcohols
The synthesis of acetate esters such as iso-amyl are produced by wine yeasts during alcoholic
acetate and ethyl acetate in S. cerevisiae is ascribed fermentation from intermediates in the branched
to at least three acetyltransferase activities: alcohol chain amino acids pathway leading to production of
acetyltransferase (AAT), ethanol acetyltransferase isoleucine, leucine and valine by decarboxylation,
(EAT) and iso-amyl alcohol acetyltransferase (IAT) transamination and reduction [180]. At high con-
[98,99]. These acetyltransferases are sulphhydryl centrations, these higher alcohols have undesirable
(SH) enzymes which react with acetyl co-enzyme A ¯avour and odour characteristics [53]. Higher
(acetyl-CoA) and, depending on the degree of alcohols in wines, however, are usually present at
af®nity, with various higher alcohols to produce concentration levels below their threshold values
esters [43,44,45]. It has also been shown that these and do not affect the taste of wine unfavourably. In
enzymatic activities are strongly repressed under some cases, they may even contribute to wine
aerobic conditions and by the addition of unsatu- quality [78]. However, since higher alcohols are
rated fatty acids to a culture. concentrated by the distilling process, their reduc-
The ATF1-encoded alcohol acetyltransferase tion in wines that are to be distilled for brandy
activity (AAT) is the best-studied acetyltransferase production is of great importance [142].
in S. cerevisiae. It has been reported that the Initial attempts to use Ilex, Leux and Valx
61 kDa ATF1 gene product (Atf1p) is located auxotrophic mutants succeeded in lowering the
within the yeast's cellular vacuoles, and that it levels of isobutanol, active amyl alcohol and
plays a major role in the production of iso-amyl isoamyl alcohol production in fermentations, but
acetate and to a lesser extent ethyl acetate during these mutants were of no commercial use as their
beer fermentation. To investigate the role of AAT growth and fermentation rates were compromised
in wine and brandy composition, we have cloned [67,68]. A Leux mutant derived from the widely
the ATF1 gene from a widely used commercial wine used Montrachet wine yeast was reported to
yeast strain (VIN13) and placed it under control of produce more than 50% less isoamyl alcohol
the constitutive yeast phosphoglycerate kinase gene during fermentation than the prototrophic parent
(PGK1) promoter and terminator [91]. Integration [142]. It will be of great interest to see whether
of this modi®ed copy of ATF1 into the genomes of integrative disruption of speci®c ILE, LEU and
three commercial wine yeast strains (VIN7, VIN13 VAL genes of wine yeasts will result in lower levels
and WE228) resulted in the over-expression of AAT of fusel oil in wine for distillation.
activity and increased levels of ethyl acetate, iso-
amyl acetate and 2-phenylethyl acetate in wine and Enhanced glycerol production
distillates. The concentration of ethyl caprate, ethyl Due to its non-volatile nature, glycerol has no
caprylate and hexyl acetate showed only minor direct impact on the aromatic characteristics of
changes, whereas the acetic acid concentration wine. However, this triol imparts certain other
decreased by more than half. These changes in the sensory qualities; it has a slightly sweet taste, and
wine and distillate composition had a pronounced owing to its viscous nature, also contributes to the
effect on the solvent/chemical (associated with ethyl smoothness, consistency and overall body of wine
acetate and iso-amyl acetate), herbaceous and [122,134].
heads-associated aroma of the ®nal distillate and The amount of glycerol in wines depends on
the solvent/chemical and fruity/¯owery character of many factors: grape variety, nitrogen composition,
Chenin blanc wines [91]. This study established the degrees of ripeness (sugar levels) and mould infec-
concept that the over-expression of acetyltransfer- tion (during which glycerol is produced), sulphite
ase genes such as ATF1 could profoundly affect the levels and pH of grape must, fermentation tempera-
¯avour pro®les of wines and distillates de®cient in ture, aeration and choice of starter culture strain
aroma, thereby paving the way for the production and inoculation level [129,134]. Typically, under
of products maintaining a fruitier character for controlled conditions, glycerol concentrations are
longer periods after bottling. higher in red wines than in white wines, varying

from 1 to 15 g/l [134]. The threshold taste level of signal transduction pathway when cells are exposed
glycerol is observed at 5.2 g/l in wine, whereas a to hyperosmotic stress [134].
change in the viscosity is only perceived at a level of Conversely, the utilization of glycerol is coupled
25 g/l [134]. Wine yeast strains overproducing to respiration via a glycerol kinase. This GUT1-
glycerol would therefore be of considerable value encoded glycerol kinase converts glycerol to
in improving the organoleptic quality of wine glycerol-3-phosphate, which is then oxidized to
[102,129]. dihydroxyacetone phosphate by the GUT2-encoded,
In addition, the overproduction of glycerol at the ¯avin-dependent and membrane-bound mitochon-
expense of ethanol could ful®l a growing need for drial glycerol-3-phosphate dehydrogenase [134].
table wine with lower levels of ethanol. About GUT2 is strongly repressed in the presence of
4±10% of the carbon source is usually converted to glucose. FPS1 that encodes a channel protein
glycerol, resulting in glycerol levels of 7±10% of that belonging to the MIP family, was shown to act as
of ethanol [134]. Redirecting more of the grape a glycerol transport facilitator controlling both
sugars to glycerol would provide a desirable glycerol in¯ux and ef¯ux [134].
alternative to the current physical ethanol-removing Slight increases in glycerol production in wine
processes that non-speci®cally alter the sensorial can be achieved by using yeast strains selected or
properties of the ®nal product [129]. Conversely, bred for high glycerol production, and by optimiz-
wine yeasts in which the glycerol pathway has been ing fermentation conditions. More recently it was
minimized would yield more alcohol, which would reported that the overexpression of GPD1, together
be of great value for the production of brandy and with constitutive expression of FPS1, successfully
other distilled products [102]. redirected the carbon ¯ux towards glycerol and the
The physiological functions of glycerol synthesis extracellular accumulation of glycerol. Depending
are related to redox balancing, resistance to on the genetic background in these engineered
hyperosmotic and oxidative stress, recycling of strains, 1.5 to four-fold increases in glycerol levels
cytosolic inorganic phosphate and nitrogen metab- were obtained [102,129]. As a result of redox
olism [122]. Futhermore, glycerol-3-phosphate, the imbalances resulting from glycerol overproduction,
precursor of glycerol, is an essential intermediate in ethanol formation was decreased and the metabolite
the biosynthesis of membrane lipids. It is also pattern of these recombinant wine yeasts was
noteworthy that glycerol is not only produced by considerably changed (Figure 23). A lower biomass
yeasts, but can also serve as carbon source in concentration was attained in the GPD1-overex-
aerobically grown cultures. pressing strains, probably due to high acetaldehyde
During wine fermentations, the main role of production during the growth phase [129]. Interest-
glycerol synthesis is to supply the yeast cell with ingly, the fermentation rate during the stationary
an osmotic stress responsive solute and to equili- phase of wine fermentation was stimulated in these
brate the intracellular redox balance by converting strains, suggesting that the availability of NAD
the excess NADH generated during biomass for- may be a factor controlling the rate of glycolytic
mation to NAD+ [134]. Glycerol formation entails ¯ux [129]. Other side-effects of these glycerol-
the reduction of dihydroxyacetone phosphate to overproducing yeasts included the accumulation of
glycerol-3-phosphate, a reaction catalyzed by by products such as pyruvate, acetate, acetoin, 2,3-
glycerol-3-phosphate dehydrogenase and followed butanediol and succinate [102,129].
by the dephosphorylation of glycerol-3-phosphate A method was recently devised to overcome the
to glycerol by glycerol-3-phosphatase [129]. Two most negative side-effect of glycerol overproduc-
cytosolic glycerol-3-phosphate dehydrogenases, con- tion, namely a marked increase in acetate forma-
sidered the key limiting enzymes for glycerol tion. Since acetaldehyde dehydrogenases were
formation in wine, are encoded by GPD1 and shown to play a prominent role in acetate forma-
GPD2 [134]. The expression of GPD1 is usually tion, the ALD6 and ALD7 genes encoding a
increased by hyperosmotic stress, whereas GPD2 cytosolic Mg2+-activated, NADP-dependent and a
expression is increased by anaerobic conditions mitochondrial K+-activated, NAD(P)-dependent
[134]. The level of glycerol in S. cerevisiae, and the acetaldehyde dehydrogenase, respectively, were dis-
expression of both these genes, are partially rupted. A wine yeast strain in which GPD1 was
controlled by the HOG (high osmotic glycerol) overexpressed in conjunction with the deletion of

Figure 23. A schematic representation of the overproduction of glycerol in wine yeast resulting in the production of acetate,
acetoin and butanediol

ALD6 produced two- to three-fold more glycerol of anti-oxidants, antimicrobial compounds and
and a similar amount of acetate compared to the enzyme additions, the solubility of proteins and
untransformed strain [129]. The redox balance was tartrate salts, the effectiveness of bentonite treat-
maintained in these recombinant wine yeasts by ment, the polymerization of the colour pigments,
increasing the formation of succinate and 2,3- and the oxidative and browning reactions [17].
butanediol to concentrations remaining in the Wine contains a large number of organic and
range of that found in wine. These yeasts offer inorganic acids. The predominant organic acids are
new prospects to improve the quality of wine tartaric and malic acid, accounting for 90% of the
lacking in smoothness and body, and to production titratable acidity of grapes. The main features of
of low-alcohol wines [129]. wine acidity include the acids themselves, the extent
of their dissociation, the titratable acidity and pH.
Bio-adjustment of wine acidity Factors affecting the pH and titratable acidity of
The acidity of grape juice and wine plays an grapes include soil potassium and soil moisture; the
important role in many aspects of winemaking and nature of the rootstock and characteristics of the
wine quality, including the sensory quality of the root system; viticultural practices such as canopy
wine and its physical, biochemical and microbial management and irrigation; climatic conditions and
stability. The juice and wine acidity, in particular prevailing temperature during ripening; and the
the pH, has a profound in¯uence on the survival cultivar and ®nal berry volume at harvest [17,90].
and growth of all microorganisms; the effectiveness Of all these factors the climatic conditions and

ambient temperature have a critical effect on grape were produced [49]. Attempts to fuse wine yeasts
maturation and resulting acidity of the fruit [90]. with malate-assimilating yeast also failed [127]. The
Under certain climatic conditions, the development application of high-density cell suspensions of
of acidic compounds in the grape during matura- several yeasts including S. cerevisiae, in an effort
tion and the subsequent physical and microbial to increase the rate at which L-malate was degraded
modi®cation of these compounds during the process during fermentation, was unsuccessful [50].
of winemaking can cause imbalances in the acidity Their lack of success forced investigators back to
of wines. Unless the acidity of such wines with the wine yeast itself. The ability of S. cerevisiae
suboptimal pH values is adjusted, the wines will be strains to assimilate L-malate acid varies widely.
considered as unbalanced or spoilt. In cooler Unlike S. pombe, S. cerevisiae lacks an active
climates (northern Europe, Canada, north-eastern malate transport system and L-malate enters wine
USA) chemical adjustment generally means a yeast by simple diffusion. Once inside the cell, S.
reduction in titratable acidity by physicochemical cerevisiae's own constitutive NAD-dependent malic
practices such as blending, chemical neutralization enzyme converts L-malate to pyruvate, which, under
by double salting (addition of calcium carbonate) anaerobic conditions, will be converted to ethanol
and precipitation. These procedures often reduce and carbon dioxide. Aerobically, malic acid is
wine quality and require extensive labour and decarboxylated into water and carbon dioxide.
capital input. Although the biochemical mechanism for malate
In the warmer viticultural regions of southern degradation in S. cerevisiae is the same as in S.
Europe, California, South Africa and Australia, pombe, the substrate speci®city of the S. cerevisiae
blessed with adequate sunshine during the growing malic enzyme is about 15-fold lower than that of
season and grape ripening period, malic acid is the S. pombe malic enzyme [175]. This low substrate
catabolized at a faster rate. Here, adjustment of speci®city, together with the absence of an active
wine acidity generally entails increasing the titrata- malate transport system, is responsible for S.
ble acidity, or more critically, lowering the pH by cerevisiae's inef®cient metabolism of malate.
the addition of tartaric acid, and sometimes malic Genetic engineering of wine yeast to conduct
and citric acids, depending on the laws of the alcoholic fermentation and malate degradation
country. Since the addition of calcium carbonate simultaneously has been explored by several
and acids are highly contentious practices that groups. In order to engineer a malolactic pathway
sometimes affect free trade in wine, several labora- in S. cerevisiae the malolactic genes (mleS) from
tories explored biological alternatives in order to Lactococcus lactis, [2,3,11,28] Lactobacillus del-
minimize such chemical intervention. brueckii [184] and the mleA gene from O. oeni [83]
At present, biodeacidi®cation of wine is mediated were cloned and expressed in S. cerevisiae. The mleS
by lactic acid bacteria, in particular Oenococcus oeni gene encodes a NAD-dependent malolactic enzyme
(formerly Leuconostoc oenos). During malolactic that converts L-malate to L-lactate and carbon
fermentation, L-malic acid is decarboxylated to L(+)- dioxide [28]. However, due to the absence of an
lactic acid and carbon dioxide. Malolactic fermen- active malate transport system in S. cerevisiae, these
tation not only reduces the total acidity of wine, it engineered strains could still not metabolize malate
also enhances microbiological stability and presum- ef®ciently [175]. Ef®cient malolactic fermentation
ably improves the organoleptic quality of wine [24]. was achieved only when the L. lactis mleS gene was
However, owing to nutrient limitation, low tem- co-expressed with the S. pombe mae1 gene encoding
perature, acidic pH, and high alcohol and sulphur malate permease [175].
dioxide levels, the malolactic bacteria often grow Similarly, an ef®cient malo-ethanolic S. cerevisiae
poorly in wine, thereby complicating the manage- was constructed by co-expressing the mae1 per-
ment of this process. Stuck or sluggish malolactic mease gene and the mae2 malic enzyme gene from
fermentation often leads to spoilage of wine. S. pombe in S. cerevisiae [175]. A functional
Several alternatives were explored, including the malolactic wine yeast could replace the unreliable
possible use of malate-degrading yeasts. During bacterial malolactic fermentation, whereas a malo-
malo-ethanolic fermentations conducted by the ethanolic strain of S. cerevisiae would be more
®ssion yeast Schizosaccharomyces pombe, malate is useful for the production of fruity ¯oral wines in
effectively converted to ethanol but off-¯avours the cooler wine-producing regions of the world.

Conversely, acidi®cation of high-pH wines pro- formed in grapes during ripening; dusting of vines
duced in the warmer regions with a wine yeast with fungicides containing elemental sulphur pro-
would be an inexpensive and convenient biological vides another source. During the winemaking pro-
alternative. The formation of high levels of L(+)- cess, sulphite is deliberately added to most wines as
lactic acid by S. cerevisiae during alcoholic fermen- an antioxidant and antimicrobial agent. Health
tation would be useful for reducing the pH. In concerns and an unfavourable public perception of
addition to its acidi®cation properties, L(+)-lactic sulphite have led to demands for restriction of its use
acid, the main product of the metabolism of lactic and reassessment of all aspects of sulphite accumula-
acid bacteria, is stable. Due to its pleasant acidic tion in wine. Consequently, the production of
¯avour and its properties as a preservative, lactic sulphur-containing compounds by wine yeast itself
acid is widely used as a food acidulant. Moreover, it has become a focal point of research.
is naturally present in most fermented products, Sulphur is essential for yeast growth and S.
including wine, where it may be present in amounts cerevisiae can use sulphate, sulphite and elemental
of up to 6 g/l after malolactic fermentation [30]. sulphur as sole sources. The formation of sulphite
Due to the inef®ciency of the mitochondrial and sulphide by wine yeasts greatly affects the
lactate dehydrogenases under fermentation condi- quality of wine. Unlike sulphur dioxide (SO2),
tions, natural S. cerevisiae strains produce only which when properly used, has some bene®cial
traces of lactic acid during alcoholic fermentation effects, hydrogen sulphide (H2S) is one of the most
[51]. In an attempt to redirect glucose carbon to undesirable of yeast metabolites, since it causes,
lactic acid in S. cerevisiae, the lactate dehydrogen- above threshold levels of 50±80 g/l, an off-¯avour
ase-encoding genes from Lactobacillus casei [29] and reminiscent of rotten eggs [142]. Sulphite is only
bovine muscle [117] were expressed in laboratory formed from sulphate, while sulphide is formed
yeast strains. Encouraged by the fact that the L. from sulphate, sulphite, from elemental sulphur
casei lactate dehydrogenase gene, expressed under applied as a fungicide, and from cysteine
control of the yeast alcohol dehydrogenase gene, (Figure 24) [128,142]. The formation of sulphite
converted 20% of the glucose into L(+)-lactic acid, and sulphide is affected by many factors, including
this construct was also introduced into eight wine the composition of the fermentation medium.
yeast strains [30]. Wines obtained with these Apart from strain effect, the nutrient composition
engineered lactic acid±alcoholic fermentation of grape juice, the concentration of sulphate, must
yeasts were shown to be effectively acidi®ed, but clari®cation, the initial pH and temperature all
the fermentation rate was slower [30]. affect sulphite formation by wine yeasts [128].
Defects in sulphate uptake and reduction, which is
Elimination of phenolic off-¯avour normally regulated by methionine via its metabo-
Excessive amounts of volatile phenols such as 4- lites methionyl-tRNA and S-adenosylmethionine,
vinylphenol, 4-vinylguaiacol, 4-ethylphenol and 4- can result in excessive sulphite production [61].
ethylguaiacol often confer undesirable organoleptic During investigations into the regulation of sulphur
attributes on wine. These phenolic off-¯avours can metabolism in high and low sulphite-producing
be described as smoky, woody, clove-like, spicy and wine yeast strains, considerable differences in the
medicinal [61]. The POF1 gene in some strains of S. levels of activity of sulphate permease, ATP-
cerevisiae encodes a substituted cinnamic acid sulphurylase and sulphite reductase were reported
carboxylase that is able to decarboxylate grape [128]. Sulphate permease, mediating the uptake of
hydroxycinnamic acids in a non-oxidative fashion sulphate by the yeasts, was shown not to be
to vinylphenols. Perhaps the disruption of POF1 repressed by methionine in high sulphite-producing
could provide a way to reduce the content of strains. ATP-sulphurylase and ADP-sulphurylase
volatile phenols in, at least, white wines [61]. are not regulated by sulphur intermediates in
high or low sulphite-producing strains. Unlike the
Reduced sulphite and sulphide production high sulphite-producing strains, the low sulphite-
Owing to their high volatility, reactivity and producing strains showed an increased biosynthesis
potency at very low threshold levels, sulphur- of NADPH-dependent sulphite reductase, O-acetyl-
containing compounds have a profound effect on serine sulphydrylase and O-acetylhomoserine sul-
the ¯avour of wine [128]. These substances are phydrylase during the exponentional growth phase

Figure 24. A schematic representation of the pathway for sulphate reduction (adapted from Boulton et al. [13])

in the presence of sulphate, sulphite and djencolic- nitrogen (typically in the form of diammonium
acid [128]. Methionine and cysteine are known to phosphate) during active fermentation. However, it
prevent an increase in the levels of sulphite has been reported that impaired membrane trans-
reductase, O-acetylserine sulphydrylase and O-acet- port function and intracellular de®ciency of certain
ylhomoserine sulphydrylase [128]. Since sulphite vitamins can also cause H2S accumulation [61].
production is very energy-dependent, the cellular The amount of H2S produced can also be
metabolism of high SO2-forming yeast strains is affected by the addition of a high level of SO2 to
reduced, explaining the decreased production of the must shortly before inoculating with yeast, and
biomass and slow fermentation rate [128]. by the strain of yeast involved. Certain yeasts more
The formation of H2S by yeast during fermenta- readily reduce sulphate and SO2 to H2S when
tion is largely in response to nutrient depletion, deprived of nitrogen, in a futile effort to synthesize
especially assimilable nitrogen and possibly certain and supply sulphur-containing amino acids to the
vitamins such as pantothenate or pyridoxine [61]. In growing yeast cell [62,71,72,73]. The addition of
the absence of the H2S sequestering molecules O- ammonium salts prevents H2S accumulation in
acetylserine and O-acetylhomoserine, as caused by wine, not by stopping its formation but by enabling
nitrogen starvation, free H2S accumulates and the yeast to synthesize amino acid precursor
diffuses from the cell (Figure 25) [71]. Depending compounds which react with H2S to form sulphur-
on soil type and vintage conditions, some grape containing acids [62]. Due to higher fermentation
varieties (e.g. Riesling, Chardonnay and Syrah), temperatures in hot climate red wine production,
tend to have a low nitrogen content. This problem yeast cells use more nitrogen during rapid fermen-
can usually be suppressed by the addition of tations and tend to develop sulphidic smells.

Figure 25. A schematic representation of the repressive regulation of the sulphate reduction sequence and
methionine±cysteine biosynthetic pathway (adapted from Henschke and Jiranek [62])

Fortunately, H2S is highly volatile and can usually in wine yeasts. The MET3 gene encoding ATP
be removed by the stripping action of CO2 sulphurylase (the ®rst enzyme in the conversion of
produced during these rapid high-temperature intracellular sulphate to sulphite) has been cloned
fermentations [61]. However, H2S formed towards and shown to be regulated at the transcriptional
the end of, or after, fermentation can react with level [61]. This may lead to the elucidation of
other wine components to form mercaptans, thiols sulphite and sulphide formation by wine yeasts.
and disulphides, which have pungent garlic, onion H2S production also appears to be closely related to
and rubber aromas [62]. the activity of sulphite reductase [71,72,73] and this
Yeast strains differ widely in their ability to could also provide a target for down regulation of
produce sulphite and sulphide [62]. One way to take H2S formation in wine yeasts.
advantage of this fact is to select or develop a wine
yeast strain that will either produce less H2S or that
will retain most of the H2S produced intracellularly.
Improvement of wine wholesomeness
It was amply demonstrated in several laboratories Until the eighteenth century, wine played a pivotal
that yeast strains with low H2S production and role in medical practice, not least because it was a
improved winemaking properties can be bred by safer drink than most available water. Thanks to its
hybridization. In addition to exploiting the genetic alcohol and acid content, wine inhibits the growth
heterogeneity in sulphite and sulphide formation, of many spoilage and pathogenic microorganisms.
the deliberate introduction of mutations in certain By the second half of the twentieth century, though,
enzymes of the sulphur, sulphur amino acids, alcohol consumption, including wine drinking, had
pantothenate and pyridoxine pathways might well become a target of some health campaigners, who,
enable a stepwise elimination of these characteristics with some success, demanded warning labels on

wine bottles. By the 1990s medical science was violet radiation. Resveratrol is synthesized particu-
reporting that moderate consumption, especially of larly in the skin cells of grape berries and only
red wine, can reduce the incidence of heart disease. traces are found in the fruit ¯esh. Red wine
Today, it is generally accepted that moderate wine therefore contains a much higher resveratrol con-
drinking can be socially bene®cial, and that it can centration than white wine due to skin contact
be effective in the management of stress and during the ®rst phase of fermentation.
reducing the risk of coronary heart disease. The One way to increase the levels of resveratrol in
prudent wine drinker, however, continues to keep a both red and white wine is to develop wine yeasts
close eye on what and how he or she drinks to able to produce resveratrol during fermentation. To
ensure that the bene®ts exceed the risks [131]. The achieve this goal, the phenylpropanoid pathway in
worldwide decrease of alcohol consumption testi®es S. cerevisiae will have to be modifed to produce p-
to this effect. coumaroyl-coenzyme A, one of the substances for
In developing wine yeast strains, it is therefore of resveratrol synthesis. This can be done by introdu-
the utmost importance to focus on these health cing the phenylalanine ammonia-lyase gene (PAL),
aspects and to develop yeasts that may reduce the cinnamate 4-hydroxylase gene (C4H) and the
risks and enhance the bene®ts. It is therefore no coenzyme A ligase gene (4CL216) in S. cerevisiae.
surprise that, since glycerol and ethanol are inver- The introduction of the grape stilbene synthase gene
sely related, part of the objective in developing (Vst1) may then catalyze the addition of three
glycerol-overproducing S. cerevisiae strains is to acetate units from malonyl-coenzyme A, already
reduce alcohol content in the end product. Like- found in yeast, to p-coumaryl-coenzyme A, result-
wise, research in several laboratories around the ing in the formation of resveratrol. At this stage,
world is directed towards the elimination of however, there is little indication of the chances for
suspected carcinogenic compounds in wine, such as success in developing resveratrol-producing wine
ethyl carbamate, and asthmatic chemical preserva- yeast strains.
tives, such as sulphites.
It might even be possible to develop wine yeasts Reduced formation of ethyl carbamate
that could increase the levels of phenolic and anti- Ethyl carbamate (also known as urethane) is a
oxidative substances (e.g. resveratrol) associated suspected carcinogen that occurs in most fermented
with the so-called `French paradox', in which, foods and beverages. Given the potential health
despite the high dietary fat intake of the cheese- hazard, there is a growing demand from consumers
loving population of southern France, the death and liquor control authorities to reduce the allow-
rate from coronary heart disease is signi®cantly able limits of ethyl carbamate in wines and related
lower than in comparable industrialized countries. products. Although young wines do not contain
Several possible explanations have been offered, measurable levels (<10 mg/l) of ethyl carbamate,
but the best case for resolving this paradox has the required precursors are present which can
been made for red wine phenolics that chemically generate a considerable amount of this mutagenic
modify blood lipoproteins in cholesterol-furred compound when wine is aged or stored at elevated
arteries. temperatures [112]. High-alcohol beverages such as
sherries, dessert wines and distilled products also
Resveratrol production tend to contain much higher levels of ethyl
Phytoalexins, including stilbenes such as resvera- carbamate than table wine. It is believed that ethyl
trol, have been shown to reduce the risk of carbamate forms in ageing wines, forti®ed wines,
coronary heart disease. By acting as an antioxidant and brandies by reaction between urea and ethanol
and as an antimutagen, resveratrol shows cancer [111]. For this reason, excessive application of urea-
chemopreventive activity, as well as the ability to containing fertilizers to vines and spraying of urea
induce speci®c enzymes which metabolize carcino- shortly before harvest to remove leaves are not
genic substances. recommended. Furthermore, the use of urea-
Stilbenes are secondary plant products produced containing nutrient supplements for yeast during
through the phenylalanine±polymalonate pathway. wine fermentations to avoid stuck or sluggish
Resveratrol is a stress metabolite produced by V. fermentations is also prohibited. Apart from these
vinifera during fungal infection, wounding or ultra factors that could lead to high urea levels and

concomitant transgression of ethyl carbamate must with a low-urea producing wine yeast strain
limits, S. cerevisiae strains also vary widely with when the juice has a high arginine content.
regard to their urea-forming ability [113]. Strain selection is only one way of reducing the
In S. cerevisiae urea is formed during the break- accumulation of urea in wine. As an alternative
down of arginine, one of the main amino acids in means of curbing ethyl carbamate formation in the
grape juice, by the CAR1-encoded arginase end product, successive disruption of the CAR1
(Figure 26). During this reaction, arginine is con- arginase gene in an industrial sakeÂ yeast proved to
verted to ornithine, ammonia and carbon dioxide, be successful in eliminating urea accumulation in
while urea is formed as an intermediate product. rice wine [75]. This arginase deletion mutation
Certain yeast strains secrete urea into wine and, resulted in a yeast strain that could not metabolize
depending on fermentation conditions, may be arginine but it also impeded growth, thereby limit-
unable to further metabolize the external urea. ing the commercial use of such a strain.
Although all S. cerevisiae strains secrete urea, the Another possibility is adding commercial pre-
extent to which they re-absorb the urea differs [1]. parations of acidic urease, enabling the hydrolysis
S. cerevisiae secretes more urea at higher fermenta- of urea in wine [110]. This practice has recently been
tion temperatures, whereas high ammonia concen- approved by the OIV and is used in some wine-
trations suppress the re-absorption of urea by the producing countries to lower ethyl carbamate levels
yeast. It is therefore important to inoculate grape in their wines and related products. A less expensive

Figure 26. A schematic representation of arginine catabolism and urea formation in wine yeast (adapted from Henschke and
Jiranek [62])

route to lower levels of ethyl carbamate would be to is associated with the presence of 2,4,6-trichloro-
develop a wine yeast that produces an extracellular, anisole in bottled wine [89].
acidic urease. In one such attempt a novel urease Spoilage yeasts include species from Brettano-
gene was constructed by fusing the a, b and c myces, the osmotolerant yeast Zygosaccharomyces
subunits of the Lactobacillus fermentum urease and the ®lm-forming yeast species Pichia and
operon [173]. In addition, jack bean urease linker Candida. Brettanomyces intermedius is known to
sequences were inserted between the a and b, as well produce haze, turbidity, volatile acidity and a
as the b and c subunits. Both gene constructs were mousy taint; Zygosaccharomyces balii causes tur-
successfully expressed under the control of the S. bidity after re-fermentation during storage of wine
cerevisiae PGK1 promoter and terminator signals in or after bottling, resulting in sediment formation
the yeasts S. cerevisiae and S. pombe. Although the and reduction in acidity [146]. Wines spoiled by
level of transcription in S. cerevisiae was much Pichia membranaefaciens and Candida krusei taste
higher than in S. pombe, the secretion of urease oxidized and less acidic [146].
peptides was extremely low [173]. Unlike the S. Without underestimating the degree of wine
pombe urease, the S. cerevisiae-derived urease was spoilage that can be caused by moulds and yeasts,
unable to convert urea into ammonia and carbon it is widely accepted that bacteria are the primary
dioxide. The absence of recombinant urease activity culprits, especially acetic acid and lactic acid
in transformed S. cerevisiae cells is probably due to bacteria. A vinegary taint in wine is often associated
the lack of the essential auxiliary proteins present with the activity of acetic acid bacteria, such as
only in urease-producing species such as S. pombe. Acetobacter aceti, Acetobacter pasteurianus and
Without these proteins, S. cerevisiae is unable to Gluconobacter oxydans [146]. Although some lactic
assemble the various subunits into an active urease. acid bacteria play a key role in the malolactic
It seems, therefore, that accessory genes of L. fermentation of wine, others may cause serious
fermentum will also have to be cloned and expressed faults. Excessive volatile acidity, mannitol taint,
in addition to the structural urease genes to enable ropiness, mousiness, acrolein formation and bitter-
S. cerevisiae to express an active urease. ness, tartaric acid degradation, diacetyl overproduc-
tion and rancidness, as well as the very unpleasant
geranium off-¯avour, are often the consequence of
Improved biological control of wine spoilage
uncontrolled growth of some species of Lactabacil-
microorganisms
lus (e.g. L. brevis, L. hilgardii, L. plantarum),
Uncontrolled microbial growth before, during or Leuconostoc (e.g. L. mesenteroides), Streptococcus
after wine fermentation can alter the chemical (S. mucilaginosus) and Pediococcus (e.g. P. cerevi-
composition of the end product, detracting from siae) [146].
its sensory properties of appearance, aroma and Healthy grapes, cellar hygiene and sound oeno-
¯avour. In severe cases of microbial spoilage, the logical practices (e.g. appropriate pH, fermentation
wine becomes unpalatable. Owing to the high initial temperature, ®ltration, application of ®ning agents,
sugar content, low pH, anaerobic fermentation etc.) will remain the corner stones of the wine-
conditions and high alcohol levels at the end of maker's strategy against uncontrolled proliferation
fermentation, only a few spoilage yeasts and of spoilage microbes. But the use of ef®cient S.
bacteria can survive the strong selective pressures cerevisiae and O. oeni starter cultures at appropriate
present in fermenting grape must and in wine [61]. inoculation levels will usually outcompete undesir-
Moulds usually spoil wine by infecting the grapes able contaminants, thereby limiting the risk of poor
or spoiling cork slabs. These include species of quality wine and concomitant ®nancial loss [41].
Penicillium, Anahanocladium, Mucor, Monilia, Tri- For additional safety, chemical preservatives such
choderma, Oidiodendron, Botrytis, Rhizopus, Clado- as sulphur dioxide and dimethyl dicarbonate are
sporium and Paecilomyces. Penicillium glabrum is commonly added to control the growth of
considered the major mould on cork slabs, while unwanted microbial contaminants. However, the
some strains of Botrytis cinerea are associated with excessive use of these chemical preservatives is
grey rot (`pourriture grise') of grapes [89]. They deleterious to the quality of wine and related
confer mouldiness and cork taints to wine. This forti®ed and distilled products, and is confronted
earthy, musty, sometimes mushroom-like aroma by mounting consumer resistance.

Consumer concerns have spurred a worldwide was successfully expressed in E. coli and S.
search for safe, food-grade preservatives of biologi- cerevisiae. In E. coli, the bactericidal action of the
cal origin [141]. A major focus of these investiga- recombinant lysozyme against Gram-negative bac-
tions into novel biopreservatives includes the teria was enhanced when a pentapeptide was
identi®cation and application of effective antimicro- inserted into C-terminus [66]. Research is underway
bial enzymes (e.g. lysozyme) and peptides (e.g. to express a modi®ed lysozyme gene in wine yeast
zymocins and bacteriocins). These efforts have that would avoid hyperglycosylation and broaden
been encouraged by the successful application of its activity to effectively eliminate spoilage by lactic
lysozyme and nisin to protect beer, wine and fruit and acetic acid bacteria.
brandies from spoilage lactic acid bacteria
[4,60,109,125,126]. But wine is a market-sensitive Wine yeasts producing antimicrobial peptides
commodity, and large-scale industrial application of The killer phenomenon is widespread among grape,
puri®ed antibacterial enzymes and bacteriocins is must, and wine-related yeast genera, including
expensive, resulting in an increase in retail costs, as Candida, Cryptococcus, Debaryomyces, Hansenia-
observed in the case of beer production [109]. This spora, Kloeckera, Kluyveromyces, Pichia and Rho-
may be overcome by developing wine yeast starter dotorula [138]. Most zymocidal strains of S.
culture strains producing appropriate levels of cerevisiae associated with wine fermentation pro-
ef®cient antimicrobial enzymes and peptides. duce the K2 or K28 zymocins which are functional
at the low pH of grape must and wine. Zymocidal
Wine yeasts producing antimicrobial enzymes yeast contaminants are implicated as one of the
Antimicrobial enzymes are ubiquitous in nature, causes of sluggish or stuck fermentations, but they
playing a pivotal role in the defense mechanisms of are also promoted for inhibiting the proliferation of
host organisms against infection by fungi and unwanted yeast contaminants. However, their ef®-
bacteria [42]. Hydrolytic antimicrobial enzymes cacy under winemaking conditions has yet to be
such as chitinases, b-glucanases and lysozyme demonstrated. Furthermore, zymocins produced by
function by degrading key structural components S. cerevisiae are lethal only to sensitive strains of
of the cell walls of moulds and bacteria. Chitinases S. cerevisiae, whereas those produced by non-
and b-glucanases synergistically attack the main Saccharomyces species may be toxic to S. cerevisiae
components of fungal cell walls, chitin and as well as non-Saccharomyces species [138].
b-1,3-glucan. Lysozyme, an N-acetylhexosamini- Several attempts have been made over the years
dase, lyses the cell walls of certain Gram-positive to expand the zymocidal activity of S. cerevisiae so
species of bacteria lacking an outer membrane by that it could also eliminate other yeast contami-
hydrolyzing the b-1,4-glucosidic linkages of pepti- nants. In some instances, different killer types of S.
doglycan in the cell wall. Its alkaline nature cerevisiae were hybridized by mating, cytoduction
contributes to the antibacterial activity of lysozyme. and spheroplast fusion, while (in another case) a
Furthermore, Gram-negative bacteria containing an DNA copy of the K1 dsRNA was introduced in a
outer membrane are more sensitive to lysozyme in K2 strain of S. cerevisiae [12]. However, even a K1/
combination with a chelating agent such as EDTA K2 double killer S. cerevisiae is very limited as to
or when lysozyme is modi®ed by perillaldehyde [42]. the variety of yeast contaminants that can be
Conjugation to galactomanan also increases the eliminated. Rather, attention is now focused on
potency of lysozyme towards Gram-negative bac- the identi®cation of genes encoding more effective
teria by enabling diffusion of the enzyme across the zymocins in other yeasts such as Pichia and
outer membrane of the target cell [42]. Hanseniaspora and their possible introduction into
The OIV has recently approved the use of S. cerevisiae.
commercial lysozyme preparations to control mal- We have investigated the feasibility of controlling
olactic fermentation and to stabilize wine after- spoilage bacteria during wine fermentations by
wards. However, the general use of lysozyme in engineering bactericidal strains of S. cerevisiae. To
winemaking is limited because of its low cost- test this novel concept, we have successfully
ef®ciency. This has encouraged efforts to develop expressed two bacteriocin genes in yeast, the one
lysozyme-producing S. cerevisiae strains [106]. The encoding a pediocin and the other a leucocin [137].
lysozyme-encoding gene from chicken egg white The pediocin gene originates from Pediococcus

acidilactici PAC1e0 [188] and the leucocin gene domesticated organism made possible the world's
from Leuconostoc carnosum B-Ta11a [34]. ®rst biotechnological processes.
The pediocin operon of P. acidilactici consists of With the emergence of modern molecular genet-
four clustered genes, namely pedA (encoding a 62 ics, S. cerevisiae has again been harnessed to shift
amino-acid precursor of the PA-1 pediocin), pedB the frontiers of mankind's newest revolution,
(encoding an immunity factor), pedC (encoding a genetic engineering. The ®rst approved human
PA-1 transport protein) and pedD (encoding a vaccine (against hepatitis B) and food product
protein involved in the transport and processing of (calf chymosin for cheese making) resulting from
PA-1) [188]. The leucocin operon of L. carnosum recombinant DNA technology were produced with
comprises two genes: lcaB (encoding a 61 amino- transgenic S. cerevisiae strains [7]. S. cerevisiae was
acid precursor of the B-Ta11a leucocin) and lcaB1 also the ®rst genetically modi®ed organism (GMO),
(encoding a 113 amino-acid immunity factor) [34]. as distinguished from a genetically modi®ed pro-
Both the P. acidilactici pedA and L. carnosum lcaB duct, to be cleared for food use, as a baking and
genes were inserted into a yeast expression-secretion brewing strain [176]. The genetically modi®ed
cassette and introduced as multicopy episomal baking strain containing constitutively expressed
plasmids into laboratory strains of S. cerevisiae. maltose permease and maltase genes, produces CO2
Northern blot analysis con®rmed that the pedA and faster than conventional baker's yeasts, thereby
lcaB structural genes in these constructs (ADH1P- ensuring that dough rises more rapidly [140]. The
MFa1S-pedA-ADH1T, designated PED1 and novel engineered feature of the pioneer GMO
ADH1P-MFa1S-lcaB-ADH1T, designated LCA1), brewer's yeast is a glucoamylase-encoding gene
were ef®ciently expressed under the control of the that allows partial hydrolysis of maltodextrins,
yeast alcohol dehydrogenase I gene promoter yielding a lower-carbohydrate beer [140]. Although
(ADH1P) and terminator (ADH1T). Secretion of not yet cleared for commercial use, considerable
the PED1-encoded pediocin and LCA1-encoded progress has been made, as detailed in this review,
leucocin was directed by the yeast mating phero- in genetically tailoring wine yeast for speci®c wine-
mone a-factor's secretion signal (MFa1S). The making processes and products of the vine.
presence of biologically active antimicrobial pep- While the scienti®c case for use of genetic
tides produced by the S. cerevisiae transformants modi®cation in the improvement of food organisms
was indicated by agar diffusion assays against is strong and persuasive, genetically modi®ed
sensitive indicator bacteria (e.g. Listeria monocyto- baking, brewing, and wine yeasts have not, as yet,
genes B73). The heterologous peptides were present been used commercially. The public perception of
at relatively low levels in the yeast supernatant but risk with regard to GM food has, so far, out-
pediocin and leucocin activities were readily weighed its view of possible bene®ts.
detected when intact yeast colonies were used in Genetic engineering, lauded as a spectacular
sensitive strain overlays. These preliminary results achievement in science, has unfortunately been
indicate that it is indeed possible to develop repackaged in an emotive, fear-mongering wrap-
bactericidal wine yeast strains that could be useful ping. Critics are whipping up public alarm, often to
fuel political agendas and to protect agricultural
in the production of wine with reduced levels of
markets. The myths of `Frankenfood' and global
potentially harmful chemical preservatives.
havoc caused by GMOs have been spread far
enough to masquerade in the cultural folklore as
truth. Fears about food and environmental safety
Complying with statutory regulation and spread more readily than good sense or wise
consumer demands science; this has evoked a plethora of strict
legislation and regulatory guidelines based far
S. cerevisiae has enjoyed a long and distinguished more on emotion than on science.
history in the fermented food and beverage indus- This mostly irrational debate began with ques-
tries; it is without doubt the most important tions about the morality (`unnatural', interfering
commercial microorganism with GRAS (`generally with evolution, playing God, etc.) and safety
regarded as safe') status. By brewing beer, leavening (GMOs and GM products are inherently danger-
bread dough, and sparkling wine, mankind's oldest ous, toxic to humans and bad for the environment)

of genetic engineering. Over the years, it has suf®cient legal and practical protection against
become generally agreed that these fears are largely accidents involving GMOs? Will genetic engineering
groundless; to date, no scienti®cally reputable test reduce biodiversity and concentrate economic
has shown any of the GM foods currently on the power in the hands of a few large producers? Do
shelf to be in the least toxic. patents on living organisms confer an unfair
The initial problems with statutory approval and advantage on certain producers? Should products
negative public perception of genetically engineered produced by gene technology be speci®cally
organisms in food and beverages are now slowly labelled?
being dissolved by a growing consensus that risk is It is clear that consumer education is essential to
primarily a function of the characteristics of a remove their fear of the unknown. Scientists must
product, rather than the use of genetic modi®cation consistently inform the public and remain open
per se. Scientists have reached a broad consensus about experiments, research and products. The
that organisms, whether modi®ed by modern consumer should be reassured of ®rst class, trans-
molecular or cellular methods or not, respond to parent regulatory systems and the meticulous
the same physical and biological laws. Therefore, implementation of biosafety legislation with clear
no conceptual distinction exists between modi®ca- technical standards and de®nitions with respect to
tion of yeast and grapevine by classical methods GM products. The consumer should be persuaded
and that by molecular techniques that modify DNA by proper risk assessment and clear demonstration
and transfer genes. of safety, and thus empowered to make informed
Regulations, although differing in detail, are decisions.
broadly similar in most countries. Guidelines for There are a number of activities that must be
approval of GM products and the release of GMOs scrupulously avoided: conducting obviously risky
usually require a number of obvious guarantees. experiments; misusing scienti®c data and exploiting
These include a complete de®nition of the DNA consumer confusion to justify trade bans and
sequence introduced, and the elimination of any technical barriers to free trade; riding the `backlash'
sequence that is not indispensable for expression of market with labels stating that a particular product
the desired property; the absence of any selective is `GM-free'; suppressing `inconvenient' scienti®c
advantage conferred on the transgenic organism data or simply lying about food safety (as has been
that could allow it to become dominant in natural the case for some governments with distressingly
habitats; no danger to human health and/or the bad biosafety records); and `force-feeding' GM
environment from the transformed DNA; and a products and GMOs down consumers' throats for
clear advantage to both the producer and the pro®t when there is no clear advantage for the
consumer. consumer.
The concept of `substantial equivalence' is widely Successful application of recombinant DNA
used in the determination of safety by comparison technology in the wine industry will depend on
with analogous conventional food and beverage assuring commercial users of genetically modi®ed
products [140]. When substantial equivalence can be wine yeasts that existing desirable characteristics
demonstrated, then usually no further safety con- have not been damaged, that the requirements of
siderations are necessary. When substantial equiva- beverage legislation are met, and that the engi-
lence is not convincingly shown, the points of neered strain will be stable in practice, with suitable
difference must be subjected to further safety procedures for monitoring. The ®rst recombinant
scrutiny. However, to date, regulatory authorities wine products should unequivocally demonstrate
appear more willing to approve the use of GMOs organoleptic, hygienic and economic advantages for
than the public is to use them. A signi®cant the wine producer and consumer. Furthermore,
proportion of the public still suspects that GM wine's most enthralling and fascinating aspect, its
food will prove unhealthy in the long term, and that diversity of style, should never be threatened by the
the escape of GMOs with transplanted genes will use of tailored wine yeast strains. In fact, gene
damage the environment and result in the loss of technology should rather be harnessed to expand
biodiversity. However, their questions now appear the diversity of high quality wines and other grape-
to be growing more speci®c. Is the product safe to derived products.
the consumer and the environment? Is there There is vast potential bene®t to the wine

consumer and industry alike, in the application of Acknowledgements

this exciting new technology. That bene®t will be
The author thanks Mr Terry Plantinga and Dr Maret du
realized, though, only if the application is judicious,
Toit for critical reading of this manuscript. Thanks are also
systematic, and done with high regard for the due to Mr Yoshi Sho for his professional design of the
unique nature of the product. In vino veritas! graphics. The research conducted in the Institute for Wine
Biotechnology is supported by grants from the National
Research Foundation (NRF) and the South African Wine
Conclusions Industry (Winetech).

Over the last few years considerable progress has

been made in developing new wine yeast strains. References
However, the wine industry, two decades after the
1. An D, Ough CS. 1993. Urea excretion and uptake by wine
®rst successful yeast transformation, has entered the
yeasts as affected by various factors. Am J Enol Vitic 44:
third millennium without a transgenic wine yeast 34±40.
used on a commercial scale to produce wine. 2. Ansanay V, Dequin S, Blondin B, Barre P. 1993. Cloning,
Successful commercialization of transgenic wine sequence and expression of the gene encoding the malolactic
yeasts will depend on a multitude of scienti®c, enzyme from Lactococcus lactis. FEBS Lett 332: 74±80.
technical, economic, marketing, safety, regulatory, 3. Ansanay V, Dequin S, Camarasa C, Schaeffer V, Grivet
J-P, Blondin B, Salmon J-M, Barre P. 1996. Malolactic
legal and ethical issues. Therefore, it would be fermentation by engineered Saccharomyces cerevisiae as
foolish to entertain unrealistic expectations about compared with engineered Schizosaccharomyces pombe.
rapid commercialization and short-term bene®ts. Yeast 12: 215±225.
The information and technology that currently exist 4. Arima H, Ibrahim HR, Kinoshita T, Kato A. 1997.
for laboratory strains of S. cerevisiae must be Bactericidal action of lysozyme attached with various sizes
of hydrophobic peptides to the C-terminal using genetic
expanded to the much more complex genomes of modi®cation. FEBS Microbiol Lett 415: 114±118.
industrial wine yeast strains before dramatic break- 5. Avery SV, Howlett NG, Radice S. 1996. Copper toxicity
throughs can be expected. Simultaneously, innova- towards Saccharomyces cerevisiae: dependence on plasma
tive technical strategies are required to comply with membrane fatty acid composition. Appl Environ Microbiol
the strict statutory regulations that pertain to the 62: 3960±3966.
6. Barnett JA. 1998. A history of research on yeasts: work by
use of GMOs. Credible means must be found to chemists and biologists, 1789±1850. Yeast 14: 1439±1451.
effectively address the concerns of traditionalists 7. Barr PJ, Gibson HL, Bathurst IC, Sabin EA, Medina-Selby
within the wine industry and the negative over- A, Coit D, Valenzuela P. 1989. Recombinant subunit
reaction of some consumer groups. vaccines from yeast. In Yeast Genetic Engineering, Barr J,
These are daunting challenges that will have to be Brake AJ, Valenzuela P (eds). Butterworth: London;
319±339.
overcome during the next few years. But given the 8. Barre P, VeÂzinhet F, Dequin S, Blondin B. 1993. Genetic
availability of the complete genome and proteome improvement of wine yeast. In Wine Microbiology and
of S. cerevisiae, the current energetic effort to Biotechnology, Fleet GH (ed). Harwood Academic: Read-
decipher the function of the roughly 6000 yeast ing; 421±447.
9. Bauer FF, Pretorius IS. 2000. Pseudohyphal and invasive
genes, and the eventual understanding of the
growth in Saccharomyces cerevisiae. Appl Microbiol (in
interaction of gene networks within the yeast cell, press).
there is no question of new and innovative devel- 10. Bisson LF. 1993. YeastÐmetabolism of sugars. In Wine
opments that will be of great bene®t to both the Microbiology and Biotechnology, Fleet GH (ed). Harwood
winemaker and the wine consumer. Academic: Reading; 55±75.
11. Bony M, Bidart F, Camarasa C, Ansanay V, Dulau L,
With their broad experience in yeast-based
Barre P, Dequin S. 1997. Metabolic analysis of S. cerevisiae
fermentations, winemakers are well placed to strains engineered for malolactic fermentation. FEBS Lett
explore new opportunities offered by the exciting 410: 452±456.
age of molecular yeast genetics and modern 12. Boone C, Sdicu AM, Wagner J, DegreÂ R, Sanchez C,
biotechnology. I am con®dent that leading-edge Bussey H. 1990. Integration of the yeast K1 killer toxin
gene into the genome of marked wine yeasts and its effect
gene technology will be sensibly applied to wine
on vini®cation. Am J Enol Vitic 41: 37±42.
yeast strains and that it will help the wine industry 13. Boulton B, Singleton VL, Bisson LF, Kunkee RE. 1996.
meet the technical challenges of the twenty-®rst Yeast and biochemistry of ethanol fermentation. In Princi-
century. Wine yeast toasts the new millennium! ples and Practices of Winemaking, Boulton B, Singleton VL,

Bisson LF, Kunkee RE (eds). Chapman and Hall: New IM, Verbakel JMA. 1995. Expression and secretion of
York; 139±172. antifreeze peptides in the yeast Saccharomyces cerevisiae.
14. Brown SW, Oliver SG. 1982. Isolation of ethanol-tolerant Yeast 11: 849±864.
mutants of yeast by continuous selection. Eur J Appl 32. Edwards CG, Beelman RB, Bartley CE, McConnell AL.
Microbiol Biotechnol 16: 119±122. 1990. Production of decanoic acid and other volatile
15. Canal-Llauberes R-M. 1993. Enzymes in winemaking. In compounds and the growth of yeast and malolactic bacteria
Wine Microbiology and Biotechnology, Fleet GH (ed). during vini®cation. Am J Enol Vitic 41: 48±56.
Harwood Academic: Reading; 477±506. 33. Eschenbruch R, Rassel JM. 1975. The development of non-
16. Cantagrel R, Lurton L, Vidal JP, Galy B. 1995. From vine foaming yeast strains for wine-making. Vitis 14: 43±47.
to Cognac. In Understanding Natural Flavours, Piggott JR, 34. Felix JV, Papathanasopoulos MA, Smith AA, Von Holy A,
Patterson A (eds). Blackie Academic and Professional: Hastings JW. 1994. Characterization of leucocin B-Ta11a: a
Glasgow; 208±228. bacteriocin from Leuconostoc carnosum Ta11a isolated from
17. Caputi A Jr, Ryan T. 1996. Must and wine acidi®cation. meat. Curr Microbiol 29: 207±212.
Presentation at a meeting of the OIV Expert Group 35. Fleet GH. 1993. The microorganisms of winemakingÐ
Technologie du Vin, Paris. isolation, enumeration and identi®cation. In Wine Micro-
18. Carstens E, Lambrechts MG, Pretorius IS. 1998. Floccula- biology and Biotechnology, Fleet GH (ed). Harwood Aca-
tion, pseudohyphal development and invasive growth in demic: Reading; 1±25.
commercial wine yeast strains. S Afr J Enol Vitic 19: 52±61. 36. Fleet GH. 1998. The microbiology of alcoholic beverages.
19. Ciriacy M, Reifenberger E. 1997. Hexose transport. In In Microbiology of Fermented Foods, vol 1, Wood BJB (ed).
Yeast Sugar Metabolism, Zimmermann FK, Entian K-D Blackie Academic and Professional: Glasgow; 217±262.
(eds). Technomic Publishing Lancaster: PA; 45±62. 37. Fleet GH, Heard GM. 1993. YeastsÐgrowth during
20. Colagrande O, Silva A, Fumi MD. 1994. Recent applica- fermentation. In Wine Microbiology and Biotechnology,
tions of biotechnology in wine production. Biotechnol Prog Fleet GH (ed). Harwood Academic: Reading; 27±54.
10: 2±18. 38. Fogel S, Welch JW, Cathala G, Karin M. 1983. Gene
21. Cole VC, Noble AC. 1995. Flavor chemistry and assess- ampli®cation in yeast: CUP1 copy number regulates copper
ment. In Fermented Beverage Production, Lea AGH, Piggott resistance. Curr Genet 7: 347±355.
JR (eds). Blackie Academic and Professional: London; 39. Francois J, BlaÂzquez MA, Arino J, Gancedo C. 1997.
361±385. Storage carbohydrates in the yeast Saccharomyces cerevi-
22. Crous JM, Pretorius IS, Van Zyl WH. 1996. Cloning and
siae. In Yeast Sugar Metabolism, Zimmermann FK, Entian
expression of the a-L-arabinofuranosidase gene (ABF2) of
K-D (eds). Technomic Publication: Lancaster, PA; 285±311.
Aspergillus niger in Saccharomyces cerevisiae. Appl Micro-
40. Franken DB, Ariatti M, Pretorius IS, Gupthar AS. 1998.
biol Biotechnol 46: 256±260.
Genetic and fermentation properties of the K2 killer yeast,
23. Crous JM, Pretorius IS, Van Zyl WH. 1995. Cloning and
Saccharomyces cerevisiae T206. Antonie van Leeuwenhoek
expression of an Aspergillus kawachii endo-1,4-b-xylanase
J Microbiol 73: 263±269.
gene in Saccharomyces cerevisiae. Curr Genet 28: 467±473.
41. Fugelsang KC. 1997. Wine Microbiology. Chapman and
24. Davis CR, Wibowo W, Eschenbruch R, Lee TH, Fleet GH.
Hall: New York.
1985. Practical implications of malolactic fermentation: a
42. Fuglsang CC, Johansen C, Christgau S, Adler-Nissen J.
review. Am J Enol Vitic 36: 290±301.
1995. Antimicrobial enzymes: applications and future
25. De Barros Lopes M, Laden A, Henschke PA, Langridge P.
potential in the food industry. Trends Food Sci Technol 6:
1996. PCR differentiation of commercial yeast strains using
390±396.
intron splice site primers. Appl Environ Microbiol 62:
43. Fujii T, Kobayashi O, Yoshomoto H, Furukawa S, Tamai
4514±4520.
Y. 1997. Effect of aeration and unsaturated fatty acids on
26. De Barros Lopes M, Rainieri S, Henschke PA, Langridge
P. 1999. AFLP ®ngerprinting for analysis of yeast genetic expression of the Saccharomyces cerevisiae alcohol acetyl-
variation. Int J Syst Bacteriol 49: 915±924. transferase gene. Appl Environ Microbiol 63: 910±915.
27. DegreÂ R. 1993. Selection and commercial cultivation of 44. Fujii T, Nagasawa N, Iwamatsu A, Bogaki T, Tamai Y,
wine yeast and bacteria. In Wine Microbiology and Hamachi M. 1994. Molecular cloning, sequence analysis,
Biotechnology, Fleet GH (ed). Harwood Academic: Read- and expression of the yeast alcohol acetyltransferase gene.
ing; 421±447. Appl Environ Microbiol 60: 2786±2792.
28. Denayrolles M, Aigle M, Lonvaud-Funel A. 1995. Func- 45. Fujii T, Yoshimoto H, Tamai Y. 1996. Acetate ester
tional expression in Saccharomyces cerevisiae of the production by Saccharomyces cerevisiae lacking the ATF1
Lactococcus lactis mleS gene encoding the malolactic gene encoding the alcohol acetyltransferase. J Ferment
enzyme. FEMS Microbiol Lett 125: 37±44. Bioeng 81: 538±542.
29. Dequin S, Barre P. 1994. Mixed lactic acid±alcoholic 46. Gagiano M, Van Dyk D, Bauer FF, Lambrechts MG,
fermentation by Saccharomyces cerevisiae expressing the Pretorius IS. 1999a. Msn1p/Mss10p, Mss11p and Muc1p
Lactobacillus casei L(+)-LDH. Bio/Technol 12: 173±177. are part of a signal transduction pathway downstream of
30. Dequin S, Baptista E, Barre P. 1999. Acidi®cation of grape Mep2p regulating invasive growth and pseudohyphal
musts by Saccharomyces cerevisiae wine yeast strains differentiation in Saccharomyces cerevisiae. Mol Microbiol
genetically engineered to produce lactic acid. Am J Enol 31: 103±116.
Vitic 50: 45±50. 47. Gagiano M, Van Dyk D, Bauer FF, Lambrechts MG,
31. Driedonks RA, Toschka HY, Van Almkerk JW, Schaffers Pretorius IS. 1999b. Divergent regulation of the evolution-

ary closely related promoters of the Saccharomyces cerevi- 66. Ibrahim HR, Yamada M, Kobayashi K, Kato A. 1992.
siae STA2 and MUC1 genes. J Bacteriol 181: 6497±6508. Bactericidal action of lysozyme against Gram-negative
48. Gainvors SA, FreÂzier V, Lemaresquier H, Lequart C, Aigle bacteria due to insertion of a hydrophobic pentapeptide
M, Belarbi A. 1994. Detection of polygalacturonase, pectin- into its C-terminus. Biosci Biotech Biochem 56: 1361±1363.
lyase and pectin-esterase activities in a Saccharomyces 67. Ingraham JL, Guymon JF. 1960. The formation of higher
cerevisiae strain. Yeast 10: 1311±1319. aliphatic alcohols by mutant strains of Saccharomyces
49. Gallander JF. 1977. Deacidi®cation of eastern table wines cerevisiae. Arch Biochem Biophys 88: 157±166.
with Schizosaccharomyces pombe. Am J Enol Vitic 28: 68. Ingraham JL, Guymon JF, Crowell EA. 1961. The pathway
65±68. of formation of n-butyl and n-amyl alcohols by a mutant
50. Gao C, Fleet GH. 1995. Degradation of malic and tartaric strain of Saccharomyces cerevisiae. Arch Biochem Biophys
acids by high density cell suspensions of wine yeasts. Food 95: 169±175.
Microbiol 12: 65±71. 69. Ishida-Fujii K, Goto S, Sugiyama H, Tagaki Y, Saiki T,
51. Genga AM, Tassi F, Ferrerro I. 1983. Mitochondrial NAD, Tagaki M. 1998. Breeding of ¯occulent industrial yeast
L-lactate dehydrogenase and NAD, D-lactate dehydrogenase strains by self-cloning of the ¯occulation gene FLO1 and
in the yeast Saccharomyces cerevisiae. Microbiologica 1: 1±8. repeated-batch fermentation by transformants. J Gen Appl
52. Gimeno CJ, Ljungdahl PO, Styles CA, Fink GR. 1992. Microbiol 44: 347±353.
Unipolar cell division in the yeast Saccharomyces cerevisiae 70. Javelot C, Girard P, Colonna-Ceccaldi B, Vladescu B. 1991.
leads to ®lamentous growth: regulation by starvation and Introductions of terpene-producing ability in a wine strain
RAS. Cell 68: 1077±1090. of Saccharomyces cerevisiae. J Bacteriol 21: 239±252.
53. Giudici P, Romano P, Zambonelli C. 1990. A biometric 71. Jiranek V, Langridge P, Henschke PA. 1995a. Regulation
study of higher alcohol production in Saccharomyces of hydrogen sul®de liberation in wine-producing Saccharo-
cerevisiae. Can J Microbiol 36: 61±64. myces cerevisiae strains by assimilable nitrogen. Appl
54. Gomez E, Laencina J, Martinez A. 1994. Vini®cation effects Environ Microbiol 61: 461±467.
on changes in volatile compounds of wine. J Food Sci 59: 72. Jiranek V, Langridge P, Henschke PA. 1995b. Amino acid
406±409. and ammonium utilization by Saccharomyces cerevisiae
55. GueÂrin B. 1991. Mitochondria. In The Yeasts, 2nd edn, vol wine yeasts from a chemically de®ned medium. Am J Enol
4, Yeast Organelles, Rose AH, Harrison JS (eds). Academic Vitic 46: 75±83.
Press: London; 541±600. 73. Jiranek V, Langridge P, Henschke PA. 1995c. Validation of
56. Hammond JRM. 1996. Yeast genetics. In Brewing Micro- bismuth-containing indicator media for predicting H2S-
biology, Priest FG, Campbell I (eds). Chapman and Hall: producing potential of Saccharomyces cerevisiae wine yeast
London; 45±82. under enological conditions. Am J Enol Vitic 46: 269±273.
57. Hansen J, Kielland-Brandt MC. 1997. Brewer's yeast. In 74. Kasahara H, Ouchi K, Kurata K, Ishido T, Nunokawa T.
Yeast Sugar Metabolism, Zimmermann FK, Entian K-D 1974. Genetic characters of non-foaming mutants of sakeÂ
(eds). Technomic Publishing: Lancaster, PA; 503±526. yeast. J Ferment Technol 52: 348±351.
58. Heard G. 1999. Novel yeasts in winemakingÐlooking to 75. Kitamoto K, Oda K, Gomi K, Takahashi K. 1991. Genetic
the future. Food Australia 51: 347±352. engineering of a sake yeast producing no urea by successive
59. Henderson RCA, Cox BS, Tubb R. 1985. The transforma- disruption of arginase gene. Appl Environ Microbiol 57:
tion of brewing yeasts with a plasmid containing the gene 301±306.
for copper resistance. Curr Genet 9: 133±138. 76. Kobayashi O, Suda H, Ohtani T, Sone H. 1996. Molecular
60. Henning S, Metz R, Hammes WP. 1986. New aspects for cloning and analysis of the dominant ¯occulation gene
the application of nisin to foods based on its mode of FLO8 from Saccharomyces cerevisiae. Mol Gen Genet 251:
action. Int J Food Microbiol 3: 135±140. 707±715.
61. Henschke PA. 1997. Wine yeast. In Yeast Sugar Metab- 77. Kreger-Van Rij NJW. 1984. The Yeasts, A Taxonomic
olism, Zimmermann FK, Entian K-D (eds). Technomic Study, 3rd edn. Elsevier Science: Amsterdam.
Publishing: Lancaster, PA; 527±560. 78. Kunkee RE, Amerine MA. 1970. Yeast in wine-making. In
62. Henschke PA, Jiranek V. 1993. YeastsÐmetabolism of The Yeast, vol 3, Yeast Technology, Rose AH, Harrison JS
nitrogen compounds. In Wine Microbiology and Biotechnol- (eds). Academic Press: New York; 5±57.
ogy, Fleet GH (ed). Harwood Academic: Reading; 77±164. 79. Kurtzman CP, Fell JW. 1998a. The Yeasts, A Taxonomic
63. Herraiz T, Ough CS. 1993. Formation of ethyl esters from Study, 4th edn. Elsevier Science: Amsterdam.
amino acids by yeasts during the alcoholic fermentation of 80. Kurtzman CP, Fell JW. 1998b. De®nition, classi®cation and
grape juice. Am J Enol Vitic 44: 41±48. nomenclature of the yeasts. In The Yeasts, A Taxonomic
64. Herraiz T, Reglero G, Herraiz M, Martin-Alvarez PJ, Study, 4th edn. Kurtzman CP, Fell JW (eds). Elsevier
Cabezudo MD. 1990. The in¯uence of the yeast strain and Science: Amsterdam; 3±5.
type of culture on the volatile composition of wine 81. La Grange DC, Pretorius IS, Van Zyl WH. 1996. Expres-
fermented without sulfur dioxide. Am J Enol Vitic 41: sion of the Trichoderma reesei b-xylanase gene (XYN2) in
313±318. Saccharomyces cerevisiae. Appl Environ Microbiol 62:
65. Houtman AC, Marais J, Du Plessis CS. 1980. Factors 1036±1044.
affecting the reproducibility of fermentation of grape juice 82. La Grange DC, Pretorius IS, Van Zyl WH. 1997. Cloning
and of the aroma composition of wines. I. Grape maturity, of the Bacillus pumilus b-xylosidase gene (xynB) and its
sugar, inoculum concentration, aeration, juice turbidity and expression in Saccharomyces cerevisiae. Appl Microbiol
ergosterol. Vitis 19: 37±54. Biotechnol 47: 262±266.

83. Labarre C, Guzzo J, Cavin J-F, DivieÁs C. 1996. Cloning acetyltransferase in Saccharomyces cerevisiae during fer-
and characterization of the genes encoding the malolactic mentation. J Am Soc Brew Chem 49: 47±53.
enzyme and the malate permease of Leuconostoc oenos. 100. Martini A. 1993. The origin and domestication of the wine
Appl Environ Microbiol 62: 1274±1282. yeast Saccharomyces cerevisiae. J Wine Res 4: 165±176.
84. Lachance M-A, Stramer WT. 1998. Ecology and yeasts. In 101. Mateo JJ, Jimenez M, Huerta T, Pastor A. 1992. Compar-
The Yeasts, A Taxonomic Study, 4th edn. Kurtzman CP, ison of volatiles produced by four Saccharomyces cerevisiae
Fell JW (eds). Elsevier Science: Amsterdam; 21±30. strains isolated from Monastrell musts. Am J Enol Vitic 43:
85. Laing E, Pretorius IS. 1992. Synthesis and secretion of an 206±209.
Erwinia chrysanthemi pectate lyase in Saccharomyces cere- 102. Michnick S, Roustan J-L, Remize F, Barre P, Dequin S.
visiae regulated by different combinations of bacterial and 1997. Modulation of glycerol and ethanol yields during
yeast promoter and signal sequences. Gene 121: 35±45. alcoholic fermentation in Saccharomyces cerevisiae strains
86. Laing E, Pretorius IS. 1993a. The primary structure and overexpressed or disrupted for GPD1 encoding glycerol 3-
expression of an Erwinia carotovora polygalacturonase- phosphate dehydrogenase. Yeast 13: 783±793.
encoding gene (peh1) in Escherichia coli and yeast. J Appl 103. Moes CJ, Pretorius IS, Van Zyl WH. 1996. Cloning and
Bacteriol 75: 149±158. expression of the Clostridium thermosulfurogenes D-xylose
87. Laing E, Pretorius IS. 1993b. Co-expression of an Erwinia isomerase gene (xylA) in Saccharomyces cerevisiae. Biotech-
chrysanthemi pectate lyase-encoding gene (pelE) and an nol Lett 18: 269±274.
Erwinia carotovora polygalacturonase-encoding gene (peh1) 104. Molan PS, Edwards M, Eschenbruch R. 1982. Foaming in
in Saccharomyces cerevisiae. Appl Microbiol Biotechnol 39: wine making: II, separation and partial characterization of
181±88. foam inducing proteins excreted by pure culture wine
88. Lambrechts MG, Bauer FF, Marmur J, Pretorius IS. 1996. yeasts. Eur J Appl Microbiol Biotechnol 16: 110±113.
Muc1, a mucin-like protein that is regulated by MSS10, is 105. Mortimer RK, Romano P, Suzzi G, Polsinelli M. 1994.
critical for pseudohyphal differentiation in yeast. Proc Natl Genome renewal: a new phenomenon revealed from a
Acad Sci U S A 93: 8419±8424. genetic study of 43 strains of Saccharomyces cerevisiae
89. Lee TH, Simpson RF. 1993. Microbiology and chemistry of derived from natural fermentation of grape musts. Yeast 10:
cork taints in wine. In Wine Microbiology and Biotechnol- 1543±1552.
ogy, Fleet GH (eds). Harwood Academic: Reading; 106. Nakamura S, Takasaki H, Kobayashi K, Kato A. 1993.
353±372. Hyperglycosylation of hen egg white lysozyme in yeast.
90. Leske P, Lee T. 1996. Acidi®cation des mouts et des vins J Biol Chem 268: 12706±12712.
(Acidi®cation of musts and wines). Presentation at a 107. Noble AC. 1994. Wine ¯avour. In Understanding Natural
meeting of the OIV Expert Group Technologie du Vin, Flavours, Piggott JR, Patterson A (eds). Blackie Academic
Paris. and Professional: Glasgow; 228±242.
91. Lilly M, Lambrechts MG, Pretorius IS. 2000. The effect of 108. NykaÈnen L. 1986. Formation and occurrence of ¯avor
increased yeast alcohol acetyltransferase activity on the compounds in wine and distilled alcoholic beverages. Am
sensorial quality of wine and brandy. Appl Environ J Enol Vitic 37: 84±96.
Microbiol 66: 744±753. 109. Ogden KM, Waites J, Hammond JRM. 1988. Nisin and
92. Liu H, Styles CA, Fink GR. 1996. Saccharomyces cerevisiae brewing. J Inst Brew 94: 233±243.
S288C has a mutation in FLO8, a gene required for 110. Ough CS, Trioli G. 1988. Urea removal from wine by an
®lamentous growth. Genetics 144: 967±978. acid urease. Am J Enol Vitic 39: 303±307.
93. Lo W-S, Dranginis AM. 1996. FLO11, a yeast gene related 111. Ough CS, Crowell EA, Gutlove BR. 1988a. Carbamyl
to the STA genes, encodes a novel cell surface ¯occulin. compound reactions with ethanol. Am J Enol Vitic 39:
J Bacteriol 178: 7144±7151. 239±242.
94. Lodder JL. 1970. The Yeasts, A Taxonomic Study, 2nd edn. 112. Ough CS, Crowell EA, Mooney LA. 1988b. Formation of
Elsevier North-Holland: Amsterdam. ethyl carbamate precursors during grape juice (Chardon-
95. Lodder JL, Kreger-Van Rij NJW. 1952. The Yeasts, A nay) fermentation. I. Addition of amino acids, urea and
Taxonomic Study, 1st edn. Elsevier North-Holland: Amster- ammonia: effects of forti®cation on intracellular and
dam. extracellular precursors. Am J Enol Vitic 39: 243±249.
96. Luttig M, Pretorius IS, Van Zyl WH. 1997. Cloning of two 113. Ough CS, Huang Z, An D, Stevens D. 1991. Amino acid
b-xylanase-encoding genes from Aspergillus niger and their uptake by four commercial yeasts at two different tempera-
expression in Saccharomyces cerevisiae. Biotechnol Lett 19: tures of growth and fermentation: effects on urea excretion
411±415. and readsorption. Am J Enol Vitic 41: 26±40.
97. Maarse H, Van Den Berg F. 1994. Flavour of distilled 114. PeÂrez-GonzaÂlez JA, De Graaf LH, Visser J, Ramon D.
products. In Understanding Natural Flavours, Piggott JR, 1996. Molecular cloning and expression in Saccharomyces
Patterson A (eds). Blackie Academic and Professional: cerevisiae of two Aspergillus nidulans xylanase genes. Appl
Glasgow; 243±267. Environ Microbiol 62: 2179±2182.
98. Malcorps P, Dufour JP. 1987. Ester synthesis by Sac- 115. PeÂrez-GonzaÂlez JA, GonzaÂlez R, Querol A, Sendra J,
charomyces cerevisiae: localisation of acetyl-CoA: iso- Ramon D. 1993. Construction of a recombinant wine
amylalcohol acetyltransferase (AT). Proc Eur Brew Conv yeast strain expressing b-(1,4)-endoglucanase and its use in
21: 377±384. microvini®cation processes. Appl Environ Microbiol 59:
99. Malcorps P, Cheval JM, Jamil S, Dufour JP. 1991. A new 2801±2806.
model for the regulation of ester synthesis by alcohol 116. Petering JE, Henschke PA, Langridge P. 1991. The

Escherichia coli b-glucuronidase gene as a marker for assimilatory pathways in an industrial Saccharomyces
Saccharomyces yeast strain identi®cation. Am J Enol Vitic cerevisiae strain. Appl Environ Microbiol 64: 3831±3837.
42: 6±12. 134. Scanes KT, Hohmann S, Prior BA. 1998. Glycerol produc-
117. Porro DL, Brambilla L, Ranzi BM, Martegani E, Alber- tion by the yeast Saccharomyces cerevisiae and its relevance
ghina L. 1995. Development of metabolically engineered to wine: a review. S Afr J Enol Vitic 19: 17±22.
Saccharomyces cerevisiae cells for the production of lactic 135. Schaaff I, Heinisch J, Zimmerman FK. 1989. Overproduc-
acid. Biotechnol Prog 11: 294±298. tion of glycolytic enzymes in yeast. Yeast 5: 285±290.
118. Pretorius IS. 1999. Engineering designer genes for wine 136. Schmitt MJ, Klavehn P, Wang J, SchoÈnig I, Tipper DJ.
yeasts. Aust NZ Wine Indust J 14: 42±47. 1996. Cell cycle studies on the mode of action of yeast K28
119. Pretorius IS, Van der Westhuizen TJ. 1991. The impact of killer toxin. Microbiology (UK) 142: 2655±2662.
yeast genetics and recombinant DNA technology on the 137. Schoeman H, Vivier MA, Du Toit M, Dicks LMT,
wine industryÐa review. S Afr J Enol Vitic 12: 3±31. Pretorius IS. 1999. The development of bactericidal yeast
120. Pretorius IS. 1997. Utilization of polysaccharides by strains by expressing the Pediococcus acidilactici pediocin
Saccharomyces cerevisiae. In Yeast Sugar Metabolism, gene (pedA) in Saccharomyces cerevisiae. Yeast 15: 647±656.
Zimmerman FK, Entian K-D (eds). Technomic Publishing: 138. Shimizu K. 1993. Killer yeasts. In Wine Microbiology and
Lancaster, PA; 459±501. Biotechnology, Fleet GH (ed). Harwood Academic: Read-
121. Pretorius IS, Van der Westhuizen TJ, Augustyn OPH. 1999. ing; 243±264.
Yeast biodiversity in vineyards and wineries and its 139. Smart KA, Boulton CA, Hinchcliffe E, Molzahn S. 1995.
importance to the South African wine industry. S Afr Effect of physiological stress on the surface properties of
J Enol Vitic 20: 61±74. brewing yeasts. J Am Soc Brew Chem 53: 33±38.
122. Prior BA, Hohmann S. 1997. Glycerol production and 140. Smith JE. 1998. The impact of genetic engineering on food
osmoregulation. In Yeast Sugar Metabolism, Zimmermann and beverage fermentation. In Microbiology of Fermented
FK, Entian K-D (eds). Technomic Publishing: Lancaster, Foods, vol 2, Wood BJB (ed). Blackie Academic and
PA; 313±337. Professional: Glasgow; 838±852.
123. Querol A, Ramon D. 1996. The application of molecular 141. Smith JL, Pintauro ND. 1981. New preservatives and future
techniques in wine microbiology. Trends Food Sci Technol trends. In Development in Food Preservatives, Tilbury RH
7: 73±78. (ed). Elsevier Applied Science: Barking; 137±160.
124. Rachidi N, Barre P, Blondin B. 1999. Multiple Ty-mediated 142. Snow R. 1983. Genetic improvement of wine yeast. In Yeast
chromosomal translocations lead to karyotype changes in a GeneticsÐFundamental and Applied Aspects, Spencer JFT,
wine strain of Saccharomyces cerevisiae. Mol Gen Genet 261: Spencer DM, Smith ARW (eds). Springer-Verlag: New
841±850. York; 439±459.
125. Radler F. 1990a. Possible use of nisin in winemaking. I. 143. Soares EV, Mota M. 1996. Flocculation onset, growth
Action of nisin against lactic acid bacteria and wine yeasts phase and genealogical age in Saccharomyces cerevisiae.
in solid and liquid media. Am J Enol Vitic 41: 1±6. Can J Microbiol 42: 539±547.
126. Radler F. 1990b. Possible use of nisin in winemaking. II. 144. Soares EV, Teixeira JA, Mota M. 1994. Effect of cultural
Experiments to control lactic acid bacteria in production of and nutritional conditions on the control of ¯occulation
wine. Am J Enol Vitic 41: 7±11. expression of Saccharomyces cerevisiae. Can J Microbiol 40:
127. Radler F. 1993. Yeast metabolism of organic acids. In Wine 851±857.
Microbiology and Biotechnology, Fleet GH (ed). Harwood 145. Soles RM, Ough CS, Kunkee RE. 1982. Ester concentra-
Academic: Reading; 165±182. tion differences in wine fermented by various species and
128. Rauhut D. 1993. YeastÐproduction of sulfur compounds. strains of yeasts. Am J Enol Vitic 33: 94±98.
In Wine Microbiology and Biotechnology, Fleet GH (ed). 146. Sponholz W-R. 1993. Wine spoilage by microorganisms. In
Harwood Academic: Reading; 183±223. Wine Microbiology and Biotechnology, Fleet GH (ed).
129. Remize F, Roustan JL, Sablayrolles JM, Barre P, Dequin S. Harwood Academic: Reading; 395±420.
1999. Glycerol overproduction by engineered Saccharo- 147. Stashenko H, Macku C, Shibamato T. 1992. Monitoring
myces cerevisiae wine yeast strains leads to substantial volatile chemicals formed from must during yeast fermenta-
changes in by-product formation and to a stimulation of tion. J Agric Food Chem 40: 2257±2259.
fermentation rate in stationary phase. Appl Environ Micro- 148. Stratford M. 1996. Induction of ¯occulation in brewing
biol 65: 143±149. yeasts by change in pH value. FEMS Microbiol Lett 136:
130. Riou C, Nicaud J, Barre P, Gaillardin C. 1997. Stationary 13±18.
phase gene expression in Saccharomyces cerevisiae during 149. Stratford M, Assinder S. 1991. Yeast ¯occulation: Flo1 and
wine fermentation. Yeast 13: 903±915. NewFlo phenotypes and receptor structure. Yeast 7:
131. Robinson J. 1994. The Oxford Companion to Wine. Oxford 559±574.
University Press: Oxford. 150. Straver MH, Kijne JW, Smit G. 1993. Cause and control of
132. Salmon J-M. 1997. Enological fermentation kinetics of an ¯occulation in yeast. Trends Biotechnol 11: 228±232.
isogenic ploidy series derived from an industrial Saccharo- 151. Teunissen AW, Steensma HY. 1995. Review: The dominant
myces cerevisiae strain. J Ferment Bioeng 83: 253±260. ¯occulation genes of Saccharomyces cerevisiae constitute a
133. Salmon J-M, Barre P. 1998. Improvement of nitrogen new subtelomeric gene family. Yeast 11: 1001±1013.
assimilation and fermentation kinetics under enological 152. Thevelein JM. 1996. Regulation of trehalose metabolism
conditions by derepression of alternative nitrogen- and its relevance to cell growth and function. In The

Mycota III, Brambl R, Marzluf GA (eds). Springer-Verlag: legends on the prime industrial microorganism. J Ind
Berlin and Heidelberg; 395±420. Microbiol 14: 514±522.
153. Thevelein JM, De Winde JH. 1999. Novel sensing mechan- 169. Vaughan-Martini A, Martini A. 1998. Saccharomyces
isms and targets for the cAMP-protein kinase A pathway in Meyen ex Reess. In The Yeasts, A Taxonomic Study, 4th
the yeast Saccharomyces cerevisiae. Mol Microbiol 33: edn. Kurtzman CP, Fell JW (eds). Elsevier Science:
904±918. Amsterdam; 358±371.
154. Thornton RJ. 1978a. Investigations on the genetics of 170. Verstrepen K, Bauer FF, Michiels C, Delvaux F, Pretorius
foaming in wine yeasts. Eur J Microbiol Biotechnol 5: IS. 2000. Controlled expression of FLO1 in Saccharomyces
103±107. cerevisiae. EBC Monogr (in press).
155. Thornton RJ. 1978b. The mapping of two dominant genes 171. VeÂzinhet F. 1989. Obtention by intragenomic recom-
for foaming in wine yeasts. FEMS Microbiol Lett 4: bination of non-foaming clones from an enological strain
207±209. of the yeast Saccharomyces cerevisiae. Sci Aliments 9:
156. Thornton RJ. 1983. New yeast strains from oldÐthe 253±265.
application of genetics to wine yeasts. Food Technol 172. VeÂzinhet F, Lacroix S. 1984. Marquage geÂneÂtique de
Australia 35: 46±50. levures: outil de controle des fermentations en souche
157. ToÈroÈk T, Mortimer RK, Romano P, Suzzi G, Polsinelli M. pure. Bull OIV 643-644: 759±773.
1996. Quest for wine yeastsÐan old story revisited. J Ind 173. Visser JJ. 1999. Cloning and expression of the Lactobacillus
Microbiol 17: 303±313. fermentum acid urease gene in Saccharomyces cerevisiae.
158. Tromp A, De Klerk CA. 1988. Effect of copperoxychloride MSc thesis, University of Stellenbosch.
on the fermentation of must and on wine quality. S Afr 174. Vivier MA, Lambrechts MG, Pretorius IS. 1997. Co-
J Enol Vitic 9: 31±36. regulation of starch-degradation and dimorphism in the
159. Van der Westhuizen TJ, Pretorius IS. 1992. The value of yeast Saccharomyces cerevisiae. Crit Rev Biochem Mol Biol
electrophoretic ®ngerprinting and karyotyping in wine yeast 32: 405±435.
breeding programmes. Antonie van Leeuwenhoek J Microbiol 175. Volschenk H, Viljoen M, Grobler J, Petzold B, Bauer FF,
61: 249±257. Subden R, Young RA, Lonvaud A, Denayrolles M, Van
160. Van der Westhuizen TJ, Augustyn OPH, Pretorius IS. 1999. Vuuren HJJ. 1997. Engineering pathways for malate
The value of long-chain fatty acid analysis, randomly degradation in Saccharomyces cerevisiae. Nature Biotech
ampli®ed polymorphic DNA and electophoretic karyotyp- 15: 253±257.
ing for the characterisation of wine yeast strains. S Afr 176. Walker GM. 1998. Yeast Physiology and Biotechnology.
J Enol Vitic 20: 3±10. Wiley: New York.
161. Van der Westhuizen TJ, Augustyn OPH, Pretorius IS. 2000. 177. Walker GM, O'Neill JD. 1990. Morphological and meta-
Geographical distribution of indigenous Saccharomyces bolic changes in the yeast Kluyveromyces marxianus var.
cerevisiae strains isolated from South African vineyards in marxianus NRRLy 2415. J Chem Technol Biotechnol 49:
the cooler, coastal regions of the Western Cape in South 75±89.
Africa. S Afr J Enol Vitic (in press). 178. Watari J, Nomura M, Sahara H, Koshino S. 1993.
162. Van Rensburg P, Van Zyl WH, Pretorius IS. 1994. Construction of ¯occulent brewer's yeast by chromosomal
Expression of the Butyrivibrio ®brisolvens endo-b-1,4- integration of the yeast ¯occulation gene FLO1. J Inst Brew
glucanase gene together with the Erwinia pectate lyase and 100: 73±77.
polygalacturonase gene in Saccharomyces cerevisiae. Curr 179. Watson JD, Hopkins NH, Roberts JW, Steitz JA, Weiner
Genet 27: 17±22. AM. 1987. Molecular Biology of the Gene, Benjamin
163. Van Rensburg P, Van Zyl WH, Pretorius IS. 1995. Cunnings: Menlo Park; CA.
Expression of the Ruminococcus ¯avefaciens cello- 180. Webb AD, Ingraham JL. 1963. Fusel oil: a review. Adv
dextrinase gene in Saccharomyces cerevisiae. Biotech Lett Appl Microbiol 5: 317±353.
17: 481±486. 181. Webber AL, Lambrechts MG, Pretorius IS. 1997. MSS11, a
164. Van Rensburg P, Van Zyl WH, Pretorius IS. 1996. novel yeast gene involved in the regulation of starch
Co-expression of a Phanerochaete chrysosporium cellobio- metabolism. Curr Genet 32: 260±266.
hydrolase gene and a Butyrivibrio ®brisolvens endo-b-1,4- 182. Wickner RB. 1996. Double-stranded RNA viruses of
glucanase gene in Saccharomyces cerevisiae. Curr Genet 30: Saccharomyces cerevisiae. Microbiol Rev 60: 250±265.
246±250. 183. Wilcocks KL, Smart KA. 1995. The importance of surface
165. Van Rensburg P, Van Zyl WH, Pretorius IS. 1997. charge and hydrophobicity for the ¯occulation of chain-
Over-expression of the Saccharomyces cerevisiae exo-b-1,3- forming brewing yeast strains and resistance of these
glucanase gene together with the Bacillus subtilis endo- parameters to acid washing. FEMS Microbiol Lett 134:
b-1,3-1,4-glucanase gene and the Butyrivibrio ®brisolvens 293±297.
endo-b-1,4-glucanase gene in yeast. J Biotechnol 55: 43±53. 184. Williams SA, Hodges RA, Strike TL, Snow R, Kunkee RE.
166. Van Rensburg P, Van Zyl WH, Pretorius IS. 1998. 1984. Cloning the gene for the malolactic fermentation of
Engineering yeast for ef®cient cellulose degradation. Yeast wine from Lactobacillus delbrueckii in Escherichia coli and
14: 67±76. yeasts. Appl Environ Microbiol 47: 288±293.
167. Van Vuuren HJJ, Wing®eld BD. 1986. Killer yeast-cause of 185. Wing®eld BD, Southgate VJ, Pretorius IS, Van Vuuren
stuck fermentations in a wine cellar. S Afr J Enol Vitic 7: HJJ. 1990a. A neutral Saccharomyces cerevisiae strain
113±118. contains a mutation in the M genome which affects toxin
168. Vaughan-Martini A, Martini A. 1995. Facts, myths and production but not immunity. Yeast 6: 159±170.

186. Wing®eld BD, Van der Meer LJ, Pretorius IS, Van Vuuren Behaviour of Listeria monocytogenes in wiener exudates in
HJJ. 1990b. K3 killer yeast is a mutant K2 killer yeast. the presence of Pediococcus acidilactici H or pediocin AcH
Mycol Res 94: 901±906. during storage at 4 or 25uC. Appl Environ Microbiol 57:
187. Wing®eld BD, Van Vuuren HJJ, Pretorius IS. 1989. Size 1461±1467.
differentiation of M2 genomes among K2 killer yeasts. 189. Zoecklein BW, Fugelsang KC, Gump BH, Nury FS. 1995.
Mycol Res 92: 364±367. Wine Analysis and Production. Chapman and Hall: New
188. Yousef AE, Luchansky JB, Degnan AJ, Doyle MP. 1991. York.

Fermentation of Wine
No ratings yet
Fermentation of Wine
5 pages
3.experimental Approaches To The Study of Metabolism
100% (1)
3.experimental Approaches To The Study of Metabolism
8 pages
Industrial Microbiology Lec 10
No ratings yet
Industrial Microbiology Lec 10
41 pages
Week 1 - Student
No ratings yet
Week 1 - Student
82 pages
Enzyme and Bioprocess PBL
No ratings yet
Enzyme and Bioprocess PBL
24 pages
1 Literature Review Cactuse
No ratings yet
1 Literature Review Cactuse
31 pages
Fermentation in Winemaking - Wikipedia PDF
No ratings yet
Fermentation in Winemaking - Wikipedia PDF
30 pages
IHM 2nd Year FNB Service Notes
90% (29)
IHM 2nd Year FNB Service Notes
163 pages
Food Beverage Service
No ratings yet
Food Beverage Service
55 pages
In The Vineyard: Viticulture
No ratings yet
In The Vineyard: Viticulture
66 pages
Wild Yeasts Selection For High Quality Patagonian Wines
No ratings yet
Wild Yeasts Selection For High Quality Patagonian Wines
32 pages
Wines: Krisel Barcenas Angelica Zoilo
100% (1)
Wines: Krisel Barcenas Angelica Zoilo
37 pages
Wine Introduction & History
No ratings yet
Wine Introduction & History
19 pages
F & B Project Works
No ratings yet
F & B Project Works
22 pages
BHM 602BT
No ratings yet
BHM 602BT
110 pages
This Article Is About The Beverage. For Other Uses, See .: From Wikipedia, The Free Encyclopedia
100% (1)
This Article Is About The Beverage. For Other Uses, See .: From Wikipedia, The Free Encyclopedia
23 pages
Wine Terminology
100% (1)
Wine Terminology
22 pages
Estimating Economic and Environmental Impacts of Red-Wine-Making Processes in The USA
No ratings yet
Estimating Economic and Environmental Impacts of Red-Wine-Making Processes in The USA
14 pages
Project Based Learning Assignment: Topic: Wine Production
No ratings yet
Project Based Learning Assignment: Topic: Wine Production
10 pages
Example of Fermentation Process
100% (3)
Example of Fermentation Process
34 pages
Wine Knowledge Handbook
No ratings yet
Wine Knowledge Handbook
27 pages
Diversity of Microbial World
No ratings yet
Diversity of Microbial World
16 pages
Introduction To Wines
No ratings yet
Introduction To Wines
12 pages
Sns College of Technology: Ganapathy Post, Coimbatore-641006
No ratings yet
Sns College of Technology: Ganapathy Post, Coimbatore-641006
54 pages
Wine and Spirits-1
No ratings yet
Wine and Spirits-1
10 pages
IND. PROD. of WINES
No ratings yet
IND. PROD. of WINES
16 pages
Mycology One
No ratings yet
Mycology One
114 pages
Wines - An Introduction (1.1 To 1.3)
No ratings yet
Wines - An Introduction (1.1 To 1.3)
10 pages
Selection and Improvement of Wine Yeasts
No ratings yet
Selection and Improvement of Wine Yeasts
17 pages
Art of Food and Wine Pairing-2
No ratings yet
Art of Food and Wine Pairing-2
65 pages
10 .Wines
No ratings yet
10 .Wines
6 pages
Chapter 07 Let's Party
No ratings yet
Chapter 07 Let's Party
8 pages
Microbiology of Winemaking
No ratings yet
Microbiology of Winemaking
4 pages
A Chemical Industry
No ratings yet
A Chemical Industry
4 pages
Wine Knowledge Course Notes
100% (1)
Wine Knowledge Course Notes
26 pages
Wine Report
No ratings yet
Wine Report
18 pages
Wine Yeasts For The Future 7294 PDF
No ratings yet
Wine Yeasts For The Future 7294 PDF
17 pages
Chapter 6 - Fungi Editted
No ratings yet
Chapter 6 - Fungi Editted
92 pages
Wine 101 Basics & Beyond
No ratings yet
Wine 101 Basics & Beyond
63 pages
Prelim Myco Trans PPT Based
No ratings yet
Prelim Myco Trans PPT Based
98 pages
Oenology - Lesson 1
No ratings yet
Oenology - Lesson 1
2 pages
Introduction To Mycology
No ratings yet
Introduction To Mycology
2 pages
Introduction of Wine
No ratings yet
Introduction of Wine
6 pages
1
No ratings yet
1
2 pages
Introduction To Wine
No ratings yet
Introduction To Wine
21 pages
Introduction To Wines
No ratings yet
Introduction To Wines
12 pages
?wine?
No ratings yet
?wine?
31 pages
Ascomycotina MCQ
No ratings yet
Ascomycotina MCQ
9 pages
Sas 1
No ratings yet
Sas 1
12 pages
The History of Wine
No ratings yet
The History of Wine
10 pages
Yeast - 2000 - Pretorius - Tailoring Wine Yeast For The New Millennium Novel Approaches To The Ancient Art of Winemaking
No ratings yet
Yeast - 2000 - Pretorius - Tailoring Wine Yeast For The New Millennium Novel Approaches To The Ancient Art of Winemaking
55 pages
Fungi
No ratings yet
Fungi
38 pages
Microbes Associated With Food Fermentation Part 2
No ratings yet
Microbes Associated With Food Fermentation Part 2
16 pages
Mycology W1-3
No ratings yet
Mycology W1-3
22 pages
Syllabus: of Plant Families
No ratings yet
Syllabus: of Plant Families
74 pages
The Process of Winemaking
No ratings yet
The Process of Winemaking
8 pages
PFD Train
No ratings yet
PFD Train
51 pages
P1 BSMLS Mycology Introduction
No ratings yet
P1 BSMLS Mycology Introduction
86 pages
Wines - 1-1-12
No ratings yet
Wines - 1-1-12
12 pages
Biology Practical Full - Removed
No ratings yet
Biology Practical Full - Removed
113 pages
0 CH 13
No ratings yet
0 CH 13
59 pages
Morphology Fungi
No ratings yet
Morphology Fungi
37 pages
Larone S Medically Important Fungi - 2018 - Walsh - Glossary
No ratings yet
Larone S Medically Important Fungi - 2018 - Walsh - Glossary
12 pages
Growth of Yeasts During Wine Fermentations
No ratings yet
Growth of Yeasts During Wine Fermentations
14 pages
Aspergillus 2
No ratings yet
Aspergillus 2
4 pages
Wine Presentation
No ratings yet
Wine Presentation
19 pages
Reproduction of Fungi
No ratings yet
Reproduction of Fungi
2 pages
Cycle of Wine
No ratings yet
Cycle of Wine
3 pages
Bacte - Medically Significant Fungi
No ratings yet
Bacte - Medically Significant Fungi
4 pages
Taphrina
No ratings yet
Taphrina
7 pages
Teknologi
No ratings yet
Teknologi
12 pages
3 Spores Shapes, Structure and Ornamentation
No ratings yet
3 Spores Shapes, Structure and Ornamentation
15 pages
Cross Breeding of Novel Cordyceps Strains With High Contents of Cordycepin by Using MAT Genes As Selectable Markers
No ratings yet
Cross Breeding of Novel Cordyceps Strains With High Contents of Cordycepin by Using MAT Genes As Selectable Markers
7 pages
Ascomycetes XXX Full Description
No ratings yet
Ascomycetes XXX Full Description
18 pages
Sexual Reproduction in Fungi-1
No ratings yet
Sexual Reproduction in Fungi-1
18 pages
AP Lab 10 - Meiosis and Tetrad Analysis 2013
No ratings yet
AP Lab 10 - Meiosis and Tetrad Analysis 2013
6 pages
3.1. Ainsworth-Classification-of-Fungi
No ratings yet
3.1. Ainsworth-Classification-of-Fungi
11 pages
STUDY349@493396
No ratings yet
STUDY349@493396
13 pages
Act Quickly Molecular Wine Microbiology Google Drive Download
100% (15)
Act Quickly Molecular Wine Microbiology Google Drive Download
22 pages
Xylaria of Maharastra 12508-26635-1-SM
No ratings yet
Xylaria of Maharastra 12508-26635-1-SM
10 pages
Wild Yeast Tales
No ratings yet
Wild Yeast Tales
5 pages
Physiology and Genetics Selected Basic and Applied Aspects 2nd Edition by Timm Anke, Anja SchÃ Ffler ISBN 3319717391 978-3319717395 Instant Download
100% (2)
Physiology and Genetics Selected Basic and Applied Aspects 2nd Edition by Timm Anke, Anja SchÃ Ffler ISBN 3319717391 978-3319717395 Instant Download
78 pages
Irjnst V4i4p101
No ratings yet
Irjnst V4i4p101
6 pages
Wine2 2
No ratings yet
Wine2 2
26 pages
Bio 1810 Essay
No ratings yet
Bio 1810 Essay
3 pages
The Encyclopedia of Home Winemaking
From Everand
The Encyclopedia of Home Winemaking
André Vanasse
No ratings yet
Grape Vines - With Chapters on Pot Culture, Propagation and Varieties
From Everand
Grape Vines - With Chapters on Pot Culture, Propagation and Varieties
A. Ward
No ratings yet
Wine For Beginners: A Complete Guide On How To Taste, Serve, Choose And Store The Best Wines
From Everand
Wine For Beginners: A Complete Guide On How To Taste, Serve, Choose And Store The Best Wines
Jake Brown
No ratings yet
Wine for Beginners: Essential Wine Guide For Newbies: Wine & Spirits
From Everand
Wine for Beginners: Essential Wine Guide For Newbies: Wine & Spirits
Amy Maia Parker
No ratings yet
Beginner's Bottle: A First-Time Guide to the World of Wine
From Everand
Beginner's Bottle: A First-Time Guide to the World of Wine
Robert Jakobsen
No ratings yet

Yeast - 2000 - Pretorius - Tailoring Wine Yeast For The New Millennium Novel Approaches To The Ancient Art of Winemaking

Uploaded by

Yeast - 2000 - Pretorius - Tailoring Wine Yeast For The New Millennium Novel Approaches To The Ancient Art of Winemaking

Uploaded by

Yeast

Yeast 2000; 16: 675±729.

Tailoring wine yeast for the new millennium: novel

* Correspondence to: Abstract

Contents the Caucasus and Mesopotamia as early as 6000 BC

Copyright # 2000 John Wiley & Sons, Ltd.

yeasts to ferment the grape sugars spontaneously to

Chromatography Pyrolysis-gas chromatography or gas chromatography of long-chain fatty acid

Table 2. Desirable characteristics of wine yeast

Fermentation properties Technological properties

Figure 6. A schematic representation of the life cycle of

Figure 10. The isolation of haploid strains from a homo-

applied to obtain the rare hybrids formed. For

Figure 11. Rare-mating between industrial and laboratory

Yeast 2000; 16: 675±729.

Copyright # 2000 John Wiley & Sons, Ltd.

relative to that of the culture medium, thereby

Improvement of fermentation performance factors affect the fermentation performance of wine

In S. cerevisiae, trehalose (a-D-glucopyranosyl a- Glycogen, another carbohydrate reserve whose

pyruvate via the glycolytic pathway. Pyruvate is

lation in grape must is associated with grapevine

Reduced foam formation generally thought to produce insigni®cant composi-

Improvement of wine ¯avour and other

consumer and industry alike, in the application of Acknowledgements

Over the last few years considerable progress has

You might also like