Yeast and bacterial modulation of wine aroma and flavour

Abstract

Wine is a highly complex mixture of compounds which largely define its appearance, aroma, flavour and mouth-feel properties. The compounds responsible for those attributes have been derived in turn from three major sources, viz. grapes, microbes and, when used, wood (most commonly, oak). The grape-derived compounds provide varietal distinction in addition to giving wine its basic structure. Thus, the floral monoterpenes largely define Muscat-related wines and the fruity volatile thiols define Sauvignon-related wines; the grape acids and tannins, together with alcohol, contribute the palate and mouth-feel properties. Yeast fermentation of sugars not only produces ethanol and carbon dioxide but a range of minor but sensorially important volatile metabolites which gives wine its vinous character. These volatile metabolites, which comprise esters, higher alcohols, carbonyls, volatile fatty acids and sulfur compounds, are derived from sugar and amino acid metabolism. The malolactic fermentation, when needed, not only provides deacidification, but can enhance the flavour profile. The aroma and flavour profile of wine is the result of an almost infinite number of variations in production, whether in the vineyard or the winery. In addition to the obvious, such as the grapes selected, the winemaker employs a variety of techniques and tools to produce wines with specific flavour profiles. One of these tools is the choice of microorganism to conduct fermentation. During alcoholic fermentation, the wine yeast Saccharomyces cerevisiae brings forth the major changes between grape must and wine: modifying aroma, flavour, mouth-feel, colour and chemical complexity. The wine bacterium Oenococcus oeni adds its contribution to wines that undergo malolactic fermentation. Thus flavour-active yeasts and bacterial strains can produce desirable sensory results by helping to extract compounds from the solids in grape must, by modifying grape-derived molecules and by producing flavour-active metabolites. This article reviews some of the most important flavour compounds found in wine, and their microbiological origin.

Full text

Swiegers, Bartowsky, Henschke & Pretorius Microbial modulation of wine aroma and flavour 139
1. Introduction
1.1 Wine quality and perceived value
It is surprising that the wine industry attracts so much
attention, given its relative size. Vines cover less than
0.5% of the world’s crop land and wine accounts for just
0.4% of global household consumption – but for millions
of investors and hundreds of millions of consumers, the
industry’s products command intense interest, often
bordering on obsession (Anderson 2004). Although there
is value addition to the grapevine through the production
of wine and its subsequent enticing packaging and inten-
sive marketing, it is still surprising how fascinated people
are with this product, one cloaked in mysticism and
romanticism. Wine writers often collapse into language
like ‘the artistry of bottling poetry, the science of bottling
sunshine, and the economics of bottling sustainable con-
sumer satisfaction – a unique artistic masterpiece of
individual creativeness, innovative technology and smart
‘bottomlining’ when writing about the product and not
the process. Why is this the case? There are as many
hypotheses to this question as there are labels on the
shelves of a good wine merchant.
Whatever the explanation, the wine industry has come
to realise that centre stage is a market-driven space (Figure
1), and today’s consumers vote with their wallets for those
wine producers who offer a pleasurable and recognisable
‘sensory experience’. They expect a safe product produced in
an environmentally sustainable manner and enjoyable in
all sensory aspects (Bisson et al. 2002). The wine indus-
try’s challenge is to respond to these consumer sentiments
and deliver products at superior quality/price ratios.
The terms quality and value are widely used in refer-
ence to wine; the International Standards Organisation
defines quality as the ‘degree to which a set of inherent
characteristics fulfils requirements’ (Francis et al. 2005). It
is instructive to relate this definition to the different ele-
ments involved in the wine production chain. To the con-
sumer who compares wines for purchase, fulfilling require-
ments is associated with the ‘intrinsic’ sensory quality of
the wine, i.e. how the wine pleasures on appearance, the
nose and the palate, as well as the perceived value.
Value is related to both intrinsic quality and image – the
latter derived from many aspects such as how the wine is
marketed, awards received, winery environmental sus-
tainability record – and cost (Francis et al. 2005). Thus, a
wine with delightful and recognisable sensory attributes
and a high perceived image at a competitive price would
be considered by consumers as high in value.
Yeast and bacterial modulation of wine aroma and flavour
J.H. SWIEGERS, E.J.BARTOWSKY, P.A. HENSCHKE and I.S. PRETORIUS
The Australian Wine Research Institute, PO Box 197,Glen Osmond,Adelaide, SA 5064, Australia
Corresponding author: Professor Isak Pretorius, facsimile: +61 8 8303 6601, email Sakkie.Pretorius@awri.com.au
Abstract
Wine is a highly complex mixture of compounds which largely define its appearance, aroma, flavour and
mouth-feel properties. The compounds responsible for those attributes have been derived in turn from
three major sources, viz. grapes, microbes and, when used, wood (most commonly, oak). The grape-
derived compounds provide varietal distinction in addition to giving wine its basic structure. Thus, the floral
monoterpenes largely define Muscat-related wines and the fruity volatile thiols define Sauvignon-related
wines; the grape acids and tannins, together with alcohol, contribute the palate and mouth-feel properties.
Yeast fermentation of sugars not only produces ethanol and carbon dioxide but a range of minor but
sensorially important volatile metabolites which gives wine its vinous character. These volatile metabolites,
which comprise esters, higher alcohols, carbonyls, volatile fatty acids and sulfur compounds, are derived
from sugar and amino acid metabolism. The malolactic fermentation, when needed, not only provides
deacidification, but can enhance the flavour profile.
The aroma and flavour profile of wine is the result of an almost infinite number of variations in
production, whether in the vineyard or the winery. In addition to the obvious, such as the grapes
selected, the winemaker employs a variety of techniques and tools to produce wines with specific flavour
profiles. One of these tools is the choice of microorganism to conduct fermentation. During alcoholic
fermentation, the wine yeast Saccharomyces cerevisiae brings forth the major changes between grape must
and wine: modifying aroma, flavour, mouth-feel, colour and chemical complexity. The wine bacterium
Oenococcus oeni adds its contribution to wines that undergo malolactic fermentation. Thus flavour-active
yeasts and bacterial strains can produce desirable sensory results by helping to extract compounds from
the solids in grape must, by modifying grape-derived molecules and by producing flavour-active
metabolites. This article reviews some of the most important flavour compounds found in wine, and their
microbiological origin.
Keywords: acids, aroma, bacteria, esters, higher alcohols, terpenes, thiols, wine, yeast

140 Microbial modulation of wine aroma and flavour Australian Journal of Grape and Wine Research 11, 139–173, 2005
ences of individuals and populations in their target markets
and produce distinctive wines accordingly (Figure 3).
1.2 Organoleptic quality of wine
Above all, wine is supposed to be enjoyed. Four senses are
involved in defining the organoleptic quality of wine:
sight, smell, taste and touch. Wine jargon has arisen to
describe what a wine drinker senses (Forrestal 2000):
appearance (sight – e.g. cloudy, hazy, deposit in the bottom
of the glass, depth of colour, hue, mousse), nose (smell –
e.g. aroma and bouquet) and palate (taste and touch –
flavour and mouth-feel). The term aroma is normally used
to describe the smell of a young, fresh wine; primary
aromas originate during fermentation – typically youthful
with upfront fresh fruit notes. Bouquet is the term for an
older wine, less fresh but more complex; secondary aromas
stem from oak maturation and tertiary aromas originate
during bottle ageing – developed fruit showing more age,
with stewed or dried fruit and other smells also seeking
attention. Wine flavour involves sweetness, acidity, bit-
terness, saltiness and the taste of umami, mouth-feel relates
to the body and texture of wine influenced by factors
such as alcohol content (sensation of warmth) and tannins
(drying sensation). The structure of a wine includes acidity,
sweetness, bitterness (occasionally), tannin (in red wine),
alcohol, palate weight and length, mouth-feel, mousse
(in sparkling wine), as well as the intensity of fruit aroma
and flavour, and complexity (diversity and layers of
flavour). These structural elements should be in balance
and harmony – they are not assessed in isolation but in
relationship to each other. The myriad terms (Forrestal
2000) used to describe different wine styles demonstrate
how complex it is to assess and define professionally the
organoleptic quality of wine; the scope of this article is
limited to the contribution of the most important, but
not the only, flavour-active compounds with a stronger
emphasis on the aromatic compounds. For the sake of
simplicity, the terms aroma and flavour are used inter-
changeably.
The aroma of wine is due to chemical compounds
with low boiling points, which are, therefore, volatile,
escaping the glass and detectable by the human nose.
Small differences in the concentration of these volatile
aroma compounds can mean the difference between a
world-class wine and an average, ‘run of the mill drop’. To
date, more than 680 volatile compounds have been iden-
tified, an indication of the potential complexity of wine
aroma (Schreier 1979, Maarse and Vissher 1994, Rapp
1998, Guth and Sies 2002).
1.3 Wine aroma and flavour
The aroma and flavour of wine are one of the main char-
acteristics that define the differences among the vast array
of wines and wine styles produced throughout the world.
They are affected by the innumerable possible variations
in wine’s production, both in viticulture and in wine-
making. For example, while some wines have a barely
detectable odour, others have fragrance that leaps from
the glass.
One of the numerous tools that can assist winemakers
Figure 1. Wine producers are facing intensifying competition brought
about by a widening gap between wine production and wine
consumption; a shift of consumer preferences away from basic
commodity wine to premium quality wine; and economic
globalisation. While modern-day consumers look for
quality
, the
wine industry globally has achieved
quantity
with about 29 billion
litres produced annually from some 8 million hectares of vines, which
results in about 5 billion litres – or between 15% and 20% of total
production – without a ready market. Rather than the ‘commodity’
wine buyer of yesteryear, the consumer today is quality-focused,
image-conscious and price-sensitive and this has led to a change in
the rules of the marketplace to a degree where quality is defined as
‘sustainable customer and consumer satisfaction’. The process of
transforming the wine industry from a production-orientated to a
market-driven industry results in an increasing dependence on,
amongst others, technological innovation.
This quality/value equation underlies the successes of
the Australian wine industry and Brand Australia; pro-
ducers listen to the consumer. However, in a globalised,
market-driven sector with more than 34 countries pro-
ducing hundreds of thousands of different wine labels
and a global annual surplus of about 6 billion litres of
unsaleable wine (Pretorius and Bauer 2002), no wine
industry can afford to entertain complacency. Meeting
quality requirements in the future will require a better
understanding of the biology of human perception, olfac-
tory and flavour preferences, the relationship between
composition and the sensorial quality of wine, and the
production of wine to changing market specifications and
sensory preferences.
Wine drinkers’ senses are not uniformly sensitive to the
subtle assortment of changing sensations. Some of the
diversity in sensory perception and preferences for different
wine styles among individuals and populations is cultural,
some learned, some genetic (Pretorius et al. 2004). But
preferences are also influenced by factors such as gender
and age (Figure 2). Fierce competition is forcing wine pro-
ducers to understand better the expectations and prefer-

Figure 2. A hypothetical representation of the possible diversity of consumer perceptions and preferences (e.g. tropical flavours versus
herbaceous characters in Sauvignon Blanc wines) in different wine markets in the world. The diversity in perceptions and preferences for certain
flavours and wine styles between individuals and populations might be due to age, gender, cultural, ethnic and geographic factors. Discovering
these preferences and why individuals are more receptive to different tastes and smells is an area that will enable winemakers to capture
opportunities in a changing global marketplace.
Figure 3. A diagrammatic representation of the need to meet consumer demands by generating and improving a responsive supply chain to
deliver wines of appropriate quality to consumers in different world market segments. It is increasingly important to produce wine consistently
to definable aroma and flavour specifications. With this comes the need to integrate better grape and wine research and to focus on the
management of sensory features of wine that attract the consumer. Some of the key fundamental questions that relate to satisfying consumer
preferences across the wine production chain that need to be answered are: (i) What are the grape-derived precursors to the impact flavour
compounds in wine? (ii) How are these flavour-active compounds transformed during winemaking? (iii) How do viticultural and winemaking
techniques influence the concentrations of these compounds in wine? (iv) How do these compounds influence wine flavour and aroma?
(v) How do consumers respond in combination with other marketing cues?
Swiegers, Bartowsky, Henschke & Pretorius Microbial modulation of wine aroma and flavour 141
in producing wines with specific flavour profiles and to
market specifications is the choice of microbial starter cul-
ture strains to conduct fermentation (Figure 4). During the
alcoholic fermentation, yeasts do not only convert sugars
to ethanol and carbon dioxide; they also produce a range
of minor but sensorially important volatile metabolites
that gives wine its vinous character (Schreier 1979,
Etiévant 1991, Guth 1998, Rapp 1998, Lambrechts and
Pretorius 2000, Romano et al. 2003) (Figure 5). Similarly,
during the malolactic fermentation, bacteria do not only
provide deacidification when needed but they can also
enhance the flavour profile (Henick-Kling 1993, Henschke
1993, Laurent et al. 1994, Bartowsky et al. 2002a).
Wines made from specific grape varieties typically dis-
play varietal character, e.g. distinctive aromas which evoke
that variety (Dubourdieu 2000, Lambrechts and Pretorius
2000, Guth and Sies 2002, Swiegers and Pretorius 2005).
However, while some volatile aroma compounds arise
directly from chemical components of the grapes, many
grape-derived compounds are released and/or modified by
the action of flavour-active yeast and bacteria, and a
further substantial portion of wine flavour substances
result from the metabolic activities of these wine microbes
(Schreier 1979, Simpson 1979, Williams et al. 1989,
Etiévant 1991, Guth 1998, Boulton et al. 1998, Rapp
1998, Dubourdieu 2000, Ferreira et al. 2000, Lambrechts

142 Microbial modulation of wine aroma and flavour Australian Journal of Grape and Wine Research 11, 139–173, 2005
and Pretorius 2000, Ribéreau-Gayon et al. 2000a,b, Ortega
et al. 2001, Guth and Sies 2002). That is why wine has
more flavour than the grape juice it was produced from.
Therefore, the importance of yeast and, to a lesser extent,
bacteria are central to the development of wine flavour.
Many biosynthetic pathways in wine yeast and malo-
lactic bacteria are involved in the formation of wine aroma
and are affected by various factors such as the composition
and pH of the grape must and the nature and prevailing
temperature of the fermentation. In addition, viticultural
factors influencing the quality of the grapes and the wine
include the cultivar, soil quality, water management, vine
canopy management and the ripeness of the grapes.
Technological aspects and vinification methods like the
method of grape crushing, must treatment and skin con-
tact time also significantly influence the final aroma
(Houtman et al. 1980a,b, Henick-Kling 1993, Boulton et
al. 1998, Lambrechts and Pretorius 2000, Ribéreau-Gayon
et al. 2000a,b, Bartowsky et al. 2002a).
Modern advances in scientific research are giving
winemakers tools to shape their wines toward predeter-
mined aroma outcomes. Today, wine yeast and bacteria
can be selected to optimally biosynthesise flavour-active
compounds, and to release grape-derived flavour com-
pounds and/or modify grape-derived flavour compounds
without affecting the general fermentation performance.
2. The modulation of wine flavour by yeast
Though grape must is relatively complete in nutrient con-
tent, it can support the growth of only a limited number
of microbial species (Henschke 1997). The low pH and
high sugar content of grape must exert strong selective
pressure on the microorganisms, such that only a few
yeast and bacterial species can proliferate. Concentrations
of sulfur dioxide, added as an anti-oxidant and anti-micro-
bial preservative, impose additional selection, particular-
ly against undesirable oxidative microbes. The selectivity
of fermenting must is further strengthened once anaerobic
conditions are established; certain nutrients become
depleted and the increasing levels of ethanol start to elim-
inate alcohol-sensitive microbial species (Henschke 1997).
Yeasts are predominant during the complex process of
winemaking. Of the 100 yeast genera representing over
700 species, 16 are associated with winemaking: Brettano-
myces and its sexual (‘perfect’) equivalent Dekkera, Candida,
Cryptococcus, Debaryomyces, Hanseniaspora and its asexual
counterpart Kloeckera, Kluyveromyces, Metschnikowia, Pichia,
Rhodotorula, Saccharomyces, Saccharomycodes, Schizosaccharo-
myces, Torulaspora and Zygosaccharomyces (Pretorius et al.
1999).
In spontaneous fermentations there is a progressive
growth pattern of indigenous yeasts: yeasts of the genera
Kloeckera, Hanseniaspora and Candida predominate in the
early stages, followed by several species of Metschnikowia
and Pichia in the middle stages when the ethanol rises to
3–4% (Fleet and Heard 1993). The latter stages of natur-
al wine fermentations are invariably dominated by the
alcohol-tolerant strains of Saccharomyces cerevisiae. Other
yeasts, such as species of Brettanomyces, Kluyveromyces,
Schizosaccharomyces, Torulaspora and Zygosaccharomyces,
might also be present during the fermentation and sub-
sequently in the wine, some of which are capable of
adversely affecting sensory quality.
The selective pressures prevailing during the wine-
making process always favour the yeasts with the most
efficient fermentative catabolism, particularly strains of
Figure 4. A diagrammatic representation of the microbial modulation of the profile of volatile compounds in wine (based upon Swiegers and
Pretorius 2005). The relationship between grape juice composition and the final wine product that reaches the consumer depends on a myriad
of influences beyond the vineyard. It is pivotal to integrate research in viticulture and oenology with a ‘grape-to-glass-to-consumer’ focus.
Flavour-active wine yeasts and malolactic bacteria can produce desirable sensory results by helping to extract compounds from the solids in
grape must, by modifying grape-derived molecules and by producing flavour-active metabolites.

Swiegers, Bartowsky, Henschke & Pretorius Microbial modulation of wine aroma and flavour 143
Saccharomyces cerevisiae. For this reason, Saccharomyces cere-
visiae is almost universally preferred for initiating alcoholic
fermentation, and has earned itself the title of the wine
yeast. The primary role of wine yeast is to catalyse the
rapid, complete and efficient conversion of grape sugars to
ethanol, carbon dioxide and other minor, but sensorially
important metabolites without the development of off-
flavours (Figure 5) (Pretorius 2000). A secondary role
concerns the modification of grape-derived constituents
such as glyco- and cysteine-conjugates, which enhance
the wines’ varietal character. In this section, specific atten-
tion will be given to the contribution of yeast-derived
acids, alcohols, carbonyl compounds, phenols, esters, sul-
fur compounds and monoterpenoids to the aroma and
flavour profile of wine.
2.1 Acids
2.1.1 Non-volatile acids
The acidity of grape juice and wine has a direct impact on
its sensory quality and physical, biochemical and microbial
stability (Fowles 1992, Jackson 1994, Boulton et al. 1998).
Acids can have both positive and negative impacts on
aroma and flavour, depending on concentration and the
type and style of wine. This acidity, particularly pH, influ-
ences (i) the survival and growth of all microorganisms;
(ii) the effectiveness of anti-oxidants, antimicrobial com-
pounds and enzyme additions; (iii) the solubility of pro-
teins and tartrate salts; (iv) the effectiveness of bentonite
treatment; (v) the polymerisation of the colour pigments;
(vi) the oxidative and browning reactions; and (vii) the
freshness of some wine styles. Wine contains a large
number of organic and inorganic acids. The predominant
non-volatile organic acids are tartaric acid and malic acid,
accounting for 90% of the titratable acidity (TA) of grape
juice. Citric acid and lactic acid also contribute to the acid-
ity of grape juice; succinic and keto acids are present only
in trace amounts in grapes, but concentrations are high-
er in wines as a result of fermentation (Whiting 1976,
Fowles 1992, Radler 1993, Boulton et al. 1998).
The main features of wine acidity include the types
and concentrations of the acids, the extent of their disso-
ciation, the titratable acidity and pH. Acidity imbalances
can result, under certain climatic conditions, from the
development of acidic compounds in the grapes and the
physical and microbial modification of these compounds
during the process of winemaking. Without adjustment of
acidity, the wines will be regarded as unbalanced or spoilt.
Chemical adjustment in cooler climates generally
means reducing titratable acidity by blending, chemical
neutralisation by double salting (addition of calcium car-
bonate) and precipitation. In warmer viticultural regions
with adequate sunshine during the growing season and
grape ripening period, malic acid is catabolised at a faster
rate than tartaric acid. Adjustment of wine acidity gener-
ally entails increasing the titratable acidity, or more criti-
cally, lowering the pH by the addition of tartaric acid, and
sometimes malic acid and citric acid, depending on the
laws of the country.
Figure 5. A schematic representation of derivation and synthesis of flavour-active compounds from sugar, amino acids and sulfur metabolism
by wine yeast.

144 Microbial modulation of wine aroma and flavour Australian Journal of Grape and Wine Research 11, 139–173, 2005
As tartaric acid is essentially stable to microbial activ-
ity, little change in its concentration occurs during fer-
mentation. Most yeasts can utilise significant concentra-
tions of malic acid. Whereas wine strains of Saccharomyces
cerevisiae typically degrade 3–45% of malic acid during
fermentation most strains of Schizosaccharomyces pombe and
Schizosaccharomyces malidevorans can completely degrade
it to ethanol and CO2(Rankine and Fornachon 1964,
Radler 1993). Whilst the formation of malic acid is
restricted to only some strains of Saccharomyces cerevisiae, it
appears to be widespread amongst Saccharomyces uvarum
strains (Radler 1993, Guidici et al. 1995). In one study, it
was reported that a commercial wine strain of Saccharo-
myces cerevisiae, Enoferm M2, increased the malic acid con-
centration of Cabernet Sauvignon wines by up to 1.5 g/L,
whereas another commercial wine yeast, ICV D254, con-
sumed 0.5 g/L (Holgate 1997).
The production of succinic acid is common amongst
yeasts and is the main carboxylic acid produced during
fermentation, where it typically accumulates to 2 g/L
(Thoukis et al. 1965, Radler 1993, Coulter et al. 2004). Its
production is highly variable amongst strains of Saccharo-
myces cerevisiae but Saccharomyces uvarum or Saccharomyces
bayanus strains tend to produce higher concentrations
(Heerde and Radler 1978, Giudici et al. 1995, Eglinton et
al. 2000). Succinic acid has been reported to have an
‘unusual salty, bitter taste’ in wine (Whiting 1976). The
most likely pathway for its formation appears to involve
the reductive branch (via oxaloacetate and malate) of the
tri-carboxylic acid (TCA) cycle during anaerobic fermen-
tation (Roustan and Sablayrolles 2002, Camarasa et al.
2003). As for malic acid, abnormal production of succinic
acid is sometimes observed, which can be a problem for
the winemaker since it affects the expected wine TA value,
and therefore requires further correction after fermenta-
tion (Holgate 1997). Abnormal succinic acid accumulation
during fermentation has been associated with yeast strain,
fermentation temperature, aeration, must clarity and com-
position, including sugar concentration, nutrient content,
pH, titratable acidity and sulfur dioxide concentration
(Coulter et al. 2004). γ-Amino butyric acid, whose con-
centration in must can be affected by post-harvest factors,
has been suggested to account for abnormal concentra-
tions of succinic acid in wine (Bach et al. 2004).
The keto acids, principally pyruvic and α-ketoglutaric
acid, have implications for wine stability and quality due
to their abilities to bind sulfur dioxide and to react with
phenols (Rankine 1967, Rankine 1968a,b, Rankine and
Pocock 1969). The keto acids are produced either during
the early stages of fermentation via sugar metabolism, or
from the corresponding amino acids, alanine and gluta-
mate, by the Ehrlich pathway. Strain is the most important
factor in determining keto acid production, but nitrogen
type and content of the medium also affects the concen-
tration of α-ketoglutaric acid produced (Rankine 1968
b). When nitrogen is adequate, α-ketoglutaric acid typi-
cally accumulates in wine to less than 50–100 mg/L but,
when nitrogen is limited, several hundred mg/L can be
produced by yeast (Rankine 1968b, Radler 1993).
Due to its pleasant acidic flavour and its properties as
a preservative, lactic acid, the main product of the metab-
olism of lactic acid bacteria, is widely used as a food acidu-
lant. Lactic acid is stable and, in wine, it might be present
in amounts of up to 6 g/L after malolactic fermentation.
Due to the inefficiency of the mitochondrial lactico-
dehydrogenases under fermentation conditions, natural
Saccharomyces cerevisiae strains produce only traces of lac-
tic acid during alcoholic fermentation (Dequin and Barre
1994).
Researchers have successfully employed genetic engi-
neering to construct Saccharomyces cerevisiae strains capable
of modulating the concentrations of lactic acid and malic
acid. In an attempt to redirect glucose carbon to lactic
acid in Saccharomyces cerevisiae, the lacticodehydrogenase-
encoding genes from Lactobacillus casei and bovine sources
were expressed in laboratory yeast strains (Dequin and
Barre 1994, Porro et al. 1995, Skory 2003). Encouraged by
the fact that the Lactobacillus casei lacticodehydrogenase
gene, expressed under control of the yeast alcohol dehy-
drogenase gene, converted 20% of the glucose into lactic
acid, this construct was also introduced into eight wine
yeast strains (Dequin et al. 1999). Although the fermen-
tation rate was slower, wines obtained with these engi-
neered lactic acid-alcoholic fermentation yeasts were
effectively acidified.
Unlike Schizosaccharomyces pombe, Saccharomyces cere-
visiae lacks an active malate transport system; malate
enters wine yeast by simple diffusion. Once inside the
cell, Saccharomyces cerevisiae’s own constitutive NAD-
dependent malic enzyme converts malate to pyruvate,
which, under anaerobic conditions, is converted to
ethanol and carbon dioxide. Aerobically, malic acid is
decarboxylated into water and carbon dioxide. Although
the biochemical mechanism for malate degradation in
Saccharomyces cerevisiae is the same as in Schizosaccharomyces
pombe, the substrate specificity of the Saccharomyces cere-
visiae malic enzyme is about 15-fold lower than that of the
Schizosaccharomyces pombe malic enzyme (Radler 1993,
Ansanay et al. 1996). This low substrate specificity togeth-
er with the absence of an active malate transport system
is responsible for Saccharomyces cerevisiae’s inefficient
metabolism of malate.
An attempt to produce a yeast for the rapid utilisation
of malic acid from highly acidic musts involved the isola-
tion of a mutant strain of Schizosaccharomyces malidevorans
(Rodriguez and Thornton 1988). The mutant, which was
induced by exposure to ultra-violet irradiation, consumed
malic acid at a higher rate than the wild-type and had
reduced utilisation of glucose in the presence of malic
acid (Rodriguez and Thornton 1989). The conversion of
malic acid to ethanol and CO2under anaerobic condi-
tions means that a greater level of deacidification can be
achieved than that by malolactic fermentation. Commercial-
scale trials under winemaking conditions have shown
that the mutant could degrade 3.5–10 g/L malic acid in
juices over a 21–73 hour period (Thornton and Rodriguez
1996).
Several groups of researchers have explored genetic
engineering of wine yeast to conduct alcoholic fermenta-
tion and malate degradation simultaneously (Dequin and

Swiegers, Bartowsky, Henschke & Pretorius Microbial modulation of wine aroma and flavour 145
Barre 1994, Ansanay et al. 1996, Volschenk et al. 1997).
To engineer a malo-ethanolic wine yeast, the mae1 malate
permease gene and the mae2 malic enzyme gene from
Schizosaccharomyces pombe were co-expressed in Saccharo-
myces cerevisiae. Similarly, to engineer a malolactic pathway
in Saccharomyces cerevisiae, the malolactic genes (mleS) from
Lactococcus lactis were co-expressed with the Schizosaccharo-
myces pombe mae1 permease gene (Bony et al. 1997). Malo-
ethanolic wine yeast would be favoured for low pH wines
from the cooler wine-producing regions, while the malo-
lactic wine yeast would provide the best solution for high
pH wines from warmer regions.
During vinification trials, it was shown that these
malo-ethanolic and malolactic wine yeasts could degrade
all the malic acid in must within three days with no
off-flavour (Dequin et al. 1999). The ‘malolactic yeast’ is
the first wine yeast to be commercialised by a yeast
manufacturing company; it was tested in 2002/2003 in
Moldavia. To the best of our knowledge, this represents
the first large-scale (20,000 litre) winemaking trial with a
genetically modified (GM) wine yeast. Wider use of this
GM wine yeast in commercial winemaking is likely to
be delayed by several years due to current anti-GMO
sentiments.
2.1.2 Volatile acids
Volatile acidity (VA) describes a group of volatile organic
acids of short carbon chain-length. The volatile acid con-
tent of wine is usually between 500 and 1000 mg/L
(10–15% of the total acid content) and of this, acetic acid
usually constitutes about 90% of the volatile acids (Fowles
1992, Henschke and Jiranek 1993, Radler 1993). The rest
of the volatile acids, principally propionic and hexanoic
acids, are produced as the result of fatty acid metabolism
by yeast and bacteria.
Acetic acid is of particular importance. At elevated
concentrations it imparts a vinegar-like character to wine.
Acetic acid becomes objectionable at concentrations of
0.7–1.1 g/L, depending on the style of wine; the optimal
concentration is 0.2– 0.7 g/L (Corison et al. 1979, Dubois
1994).
Acetic acid production by the strains of Saccharomyces
cerevisiae used in winemaking has been reported to vary
widely and, during fermentation, as little as 100 mg/L and
up to 2 g/L are produced (Radler 1993). Strains in current
use tend to produce acetic acid concentrations at the lower
end of the range for dry wines but tend to higher values
for sweet wines (Monk and Cowley 1984, Henschke and
Dixon 1990, Millan et al. 1991, Bely et al. 2003, Erasmus
et al. 2004). Strains of the related cryotolerant species,
Saccharomyces bayanus and Saccharomyces uvarum, typically
produce less acetic acid than Saccharomyces cerevisiae (Giudici
et al. 1995, Eglinton et al. 2000).
Acetate is produced by yeast as an intermediate of the
pyruvate dehydrogenase (PDH) bypass, a pathway res-
ponsible for the conversion of pyruvate into acetyl-CoA
through a series of reactions catalysed by pyruvate decar-
boxylase (PDC), acetaldehyde dehydrogenase and acetyl-
CoA synthase. The PDH bypass supplies the cell with
cytosolic acetyl-CoA, which is needed for anabolic process-
es such as lipid biosynthesis (Flikweert et al. 1996, Pronk
et al. 1996). The reaction catalysed by acetaldehyde dehy-
drogenase also generates reducing equivalents, which are
needed in many synthetic pathways and for redox reac-
tions (involving NAD(P)H). Acetaldehyde dehydrogenase
forms acetate by oxidising the acetaldehyde produced
from pyruvate during the fermentation. The cytosolic
acetaldehyde dehydrogenases are encoded by ALD6, ALD2
and ALD3, whereas the mitochondrial isoforms are encod-
ed by ALD4 and ALD5 (Navarro-Avino et al. 1999).
Recently, it has been shown that Ald6p, Ald5p and Ald4p
are the main enzymes responsible for acetate formation
during the more wine-like anaerobic growth on glucose
(Saint-Prix et al. 2004).
Although Saccharomyces can produce acetic acid, exces-
sive concentrations in wine are largely the result of the
metabolism of ethanol by aerobic acetic acid bacteria, a
topic discussed below.
2.2 Alcohols
2.2.1 Ethanol
Modern ‘bottled sunshine’, to continue the metaphor, is
usually characterised by full fruit and intense varietal
flavours, for which it is becoming common practice to
harvest fully ripened grapes (de Barros Lopes et al. 2003).
However, sunshine can also burn, and it is the application
of smart technology and viticulture that harnesses its good
points while negating detrimental influences. To ‘bottle
the sunshine’, grape must is typically prepared from fully
matured grapes, making for high flavour intensity, but also
a considerable concentration of sugar. This much sugar
invariably leads to the production of wines with high
levels of ethanol, sometimes reaching concentrations
above 15% (v/v) (de Barros Lopes et al. 2003).
The presence of ethanol is essential to enhance the
sensory attributes of other wine components. Excessive
ethanol, however, can produce a perceived ‘hotness’ and
mask the overall aroma and flavour of the wine (Guth and
Sies 2002). This, along with heightened health con-
sciousness, stricter drinking and driving laws, and
increased tax rates associated with high ethanol wines,
have increased the demand for wines with reduced alco-
hol concentrations, putting pressure on wine producers,
particularly those in warm climates where grape sugar
levels can become high (Day et al. 2002).
The removal or reduction of alcohol in wine can be
achieved by various physical processes, which are some-
times used in combination, and include reverse osmosis,
adsorption, distillation, centrifugation, evaporation,
extraction, freeze concentration, membrane, and partial
fermentation. These methods, which are generally effec-
tive and allow easy control of the amount of alcohol being
removed can, however, involve expensive equipment and
processing. There are restrictions on the use of some of
these techniques in some countries according to their
food laws and regulations. Loss or modification of aroma
and flavour compounds during processing is an important
consideration for several of these techniques.
Several biological solutions are being developed to
overcome some of the limitations imposed by the physi-

146 Microbial modulation of wine aroma and flavour Australian Journal of Grape and Wine Research 11, 139–173, 2005
cal techniques. As the initial sugar concentration of grape
must is an important target for achieving wines with
lower alcohol content, glucose oxidase (GOX) provides
one approach for reducing the glucose content of must
(Pickering 1999 a,b,c). Glucose is converted by GOX to D-
glucono-δ-lactone and gluconic acid, rendering it unavail-
able for alcohol formation during fermentation. Pure
enzyme preparations are effective on an industrial scale,
but further work is needed to establish the impact on
wine sensory properties. The high cost of enzyme prepa-
rations would be a deterrent, at least, at this stage.
Genetic modification of wine yeast offers another
approach to decreasing ethanol concentration in wine. A
decrease of almost 2% in ethanol concentration has been
achieved by the expression of the GOX1 glucose oxidase
gene of a food-grade fungus (Aspergillus niger) in yeast
(Malherbe et al. 2003). Similarly, a significant decrease in
ethanol concentration (up to 2%) and a concomitant
increase in extracellularly accumulated glycerol have been
achieved by the overexpression of either of the authentic
GPD1- or GPD2-encoded glycerol-3-phosphate dehydro-
genase isozymes of Saccharomyces cerevisiae (Michnick et al.
1997, Remize et al. 1999, de Barros Lopes et al. 2000).
Theoretically, a combination of these two strategies should
lower the alcohol content by more than 4% (v/v). This
research is in progress.
2.2.2 Glycerol
Glycerol is a major product of alcoholic fermentation
(Gancedo et al. 1968, Pronk et al. 1996, Scanes et al.
1998). Chemically, glycerol is a polyol with a colourless,
odourless and highly viscous character. It tastes slightly
sweet, as well as having an oily and heavy mouth-feel.
Glycerol is present in dry and semi-sweet wines in con-
centrations ranging from 5 to 14 g/L. Red wines typical-
ly have higher concentrations of glycerol than white wines
(6.82 g/L versus 10.49 g/L; Nieuwoudt et al. 2002), and
botrytised wines frequently have concentrations up to 25
g/L (Rankine and Bridson 1971, Ough et al. 1972,
Nieuwoudt et al. 2002). Although this non-volatile triol
has no direct impact on the aromatic characteristics of
wine, glycerol can, depending on its concentration and the
style of wine, have a noticeable effect on apparent sweet-
ness (Noble and Bursick 1984). Sensory tests have shown
that glycerol imparts sweetness at a threshold of about 5.2
g/L in dry white wine.
Only at high concentration does glycerol affect the
apparent viscosity of wine; a concentration of more than
25.8 g/L would be needed to produce a difference in
viscosity (Noble and Bursick 1984). Furthermore, no
relationship exists between glycerol concentration and
the tears that sometimes form on the inside of a wine
glass. In the usual concentrations that it is found in dry
wine, glycerol has no effect on the perceived aroma inten-
sity of wine (Nieuwoudt 2004).
Glycerol metabolism by yeasts plays several impor-
tant roles during the anaerobic fermentation of sugars:
(i) it provides precursors for the synthesis of phospho-
lipids, which are components of cell membranes, during
the period of yeast growth; (ii) glycerol formation helps
to maintain the cell’s redox balance, necessary for ATP
energy generation and cell growth; and (iii) glycerol pro-
tects yeast from high osmotic stress caused by high sugar
concentrations (Pronk et al. 1996). This latter role explains
the higher concentrations of glycerol found in sweet
wines.
As mentioned, the overexpression of either of the
authentic GPD1 or GPD2 genes of Saccharomyces cerevisiae
redirects the carbon flow to glycerol formation with the
concomitant overproduction of glycerol at the expense of
ethanol (Michnick et al. 1997, Scanes et al. 1998, Remize
et al. 1999, de Barros Lopes et al. 2000). However, it was
found that these high-glycerol-producing prototype strains
also increased acetic acid concentrations to unacceptable
levels. This negative side effect was circumvented by delet-
ing the ALD6-encoded acetaldehyde dehydrogenase activ-
ity, the main contributor to the oxidation of acetaldehyde
during fermentation. For example, a laboratory strain of
Saccharomyces cerevisiae over-expressing GPD2 and lacking
ALD6 had the desired effect of producing more glycerol
and less ethanol, without an increase in acetic acid (Remize
et al. 2000, Eglinton et al. 2002).
2.2.3 Higher alcohols
Higher alcohols (also known as fusel alcohols) are sec-
ondary yeast metabolites, and can have both positive and
negative impacts on the aroma and flavour of wine
(Figure 6). Excessive concentrations of higher alcohols
can result in a strong, pungent smell and taste, whereas
optimal levels impart fruity characters (Table 1) (Nykänen
et al. 1977, Lambrechts and Pretorius 2000, Swiegers and
Pretorius 2005).
Higher alcohols are divided into two categories,
aliphatic and aromatic alcohols, and are also extremely
important in wine and distillates (Nykänen et al. 1977).
The aliphatic alcohols include propanol, isoamyl alcohol,
isobutanol and active amyl alcohol. The aromatic alcohols
consist of 2-phenylethyl alcohol and tyrosol. It has been
reported that concentrations below 300 mg/L add a desir-
able level of complexity to wine, whereas concentrations
that exceed 400 mg/L can have a detrimental effect (Rapp
and Versini 1991).
The use of different yeast strains during fermentation
contributes considerably to variations in higher alcohol
profiles and concentrations in wine (Rankine 1968b,
Giudici et al. 1990). The concentration of amino acids
(the precursors for higher alcohols) in the must also influ-
ences higher alcohol production, where the total produc-
tion of higher alcohols increases as concentrations of the
corresponding amino acids increase (Schulthess and
Ettlinger 1978). Furthermore, ethanol concentration, fer-
mentation temperature, the pH and composition of grape
must, aeration, level of solids, grape variety, maturity and
skin contact time also affect the concentration of higher
alcohols in the final product (Fleet and Heard 1993). Non-
Saccharomyces yeast can also contribute to the levels of
higher alcohols. For example, mixed fermentation with
Pichia fermentans and Saccharomyces cerevisiae produced a
substantial increase in higher alcohols such as 1-propanol,
n-butanol and 1-hexanol compared to fermentation with

Swiegers, Bartowsky, Henschke & Pretorius Microbial modulation of wine aroma and flavour 147
Figure 6. A schematic representation of the biosynthesis of higher alcohols in wine yeast (based upon Boulton et al. 1998).
Table 1. Esters, higher alcohols and other aroma and flavour compounds commonly found in wine (Soles et al. 1982,
Etiévant 1991, Fleet and Heard 1993, Martineau et al. 1995, Guth 1997, Lambrechts and Pretorius 2000, Swiegers and
Pretorius 2005, Siebert et al. 2005, Smyth 2005).
Compound Concentration in wine Aroma threshold Aroma descriptor
(mg/L) (mg/L)
Ethyl acetate 22.5–63.5 7.5* VA, nail polish, fruity
Isoamyl acetate 0.1–3.4 0.03* Banana, pear
2-Phenylethyl acetate 0–18.5 0.25* Flowery, rose, fruity
Isobutyl acetate 0.01–1.6 1.6**** Banana, fruity
Hexyl acetate 0–4.8 0.7** Sweet, perfume
Ethyl butanoate 0.01–1.8 0.02* Floral, fruity
Ethyl hexanoate 0.03–3.4 0.05* Green apple
Ethyl octanoate 0.05–3.8 0.02* Sweet soap
Ethyl decanoate 0–2.1 0.2***** Floral, soap
Propanol 9.0–68 500** Pungent, harsh
Butanol 0.5–8.5 150* Fusel, spiritous
Isobutanol 9.0–174 40* Fusel, spiritous
Isoamyl alcohol 6.0–490 30* Harsh, nail polish
Hexanol 0.3–12.0 4** Green, grass
2-Phenylethyl alcohol 4.0–197 10* Floral, rose
Acetic acid 100–1150 280* VA, vinegar
Acetaldehyde 10–75 100** Sherry, nutty, bruised apple
Diacetyl <5 0.2** / 2.8*** Buttery
Glycerol 5–14 g/L 5.2 g/L** Odourless (slightly sweet taste)
Linalool 0.0017–0.010 0.0015******/0.025***** Rose
Geraniol 0.001–0.044 5******/30* Rose-like
Citronellol 0.015–0.042 8******/100* Citronella
2-acetyl-1-pyrroline (ACPY) Trace 0.0001****** Mousy
2-acetyltetrahydropyridine (ACPTY) 0.0048–0.1 0.0016****** Mousy
4-ethylphenol 0.012–6.5 0.14*/0.6*** Medicinal, barnyard
4-ethyl guaiacol 0.001–0.44 0.033*/0.11*** Phenolic, sweet
4-vinyl phenol 0.04–0.45 0.02****** phamaceutical
4-vinyl guaiacol 0.0014–0.71 10****** Clove-like, phenolic
* 10% ethanol, ** wine, *** red wine,**** beer, ***** synthetic wine, ****** water

148 Microbial modulation of wine aroma and flavour Australian Journal of Grape and Wine Research 11, 139–173, 2005
Saccharomyces cerevisiae alone (Clemente-Jimenez et al.
2005).
Branched-chain higher alcohols, isoamyl alcohol,
active amyl alcohol and isobutanol, are synthesised in the
yeast cell through the Ehrlich pathway, which involves
the degradation of the branched-chain amino acids,
leucine, isoleucine and valine (Ehrlich 1904). The uptake
of branched-chain amino acid by Saccharomyces cerevisiae is
mediated by at least three transport proteins: the general
amino acid permease Gap1p, the branched-chain amino
acid permease Bap2p, and one or more unknown perme-
ases (Didion et al. 1996).
The first step in the synthesis of higher alcohols
involves the synthesis of α-keto acids, which are formed
via the catabolic or Ehrlich pathway or an anabolic path-
way involving synthesis of branched-chain amino acids
through their biosynthetic pathway from glucose. The
first step in the catabolism of branched-chain amino acids
is transamination to form the respective α-keto acids (α-
ketoisocaproic acid from leucine, α-ketoisovaleric acid
from valine, and α-keto-β-methylvaleric acid from iso-
leucine) (Dickinson and Norte 1993). This reaction is
catalysed inside the yeast by mitochondrial and cytosolic
branched-chain amino acid aminotransferases encoded
by BAT1 and BAT2 (Eden et al. 1996, 2001, Kispal et al.
1996). Interestingly, the BAT1 gene is highly expressed
during the logarithmic growth phase and down-regulated
during the stationary phase, while the BAT2 gene shows
an inverse pattern of expression (Eden et al. 1996). A
pyruvate decarboxylase converts the resulting α-keto acid
to the corresponding branched-chain aldehyde with one
carbon-less atom, and the alcohol dehydrogenase catalyses
the NADH-dependent reduction of this aldehyde to the
corresponding fusel alcohol. Alternatively, the aldehyde
might be oxidised to an acid (Derrick and Large 1993).
Recently, researchers looked at the effect of increased
yeast branched-chain amino acid transaminase activity, in
particular Bat1p and Bat2p, on the production of higher
alcohols in the flavour profiles of wine and distillates. The
BAT1 and BAT2 genes were overexpressed under the con-
trol of the constitutive PGK1 (phosphoglycerate kinase I)
regulatory sequences in a widely used commercial wine
yeast strain (VIN13). It was found that wines and distil-
lates prepared by strains overexpressing BAT1 increased
the concentration of isoamyl alcohol and its correspond-
ing ester isoamyl acetate but to a lesser extent. The con-
centration of the higher alcohol isobutanol and isobutyric
acid also increased. The overexpression of the BAT2 gene
resulted in a substantial increase in the concentration of
isobutanol, isobutyric acid and propionic acid production,
while the deletion of this gene led to a decrease in the pro-
duction of these compounds. Sensory analyses indicated
that the wines and distillates produced with the strains in
which the BAT1 and BAT2 genes were overexpressed
individually had more fruity characteristics (peach and
apricot aromas) than the wines produced by the wild-
type strains (Lilly 2004).
2.3 Carbonyl compounds
Acetaldehyde is the major carbonyl compound found in
wine with concentrations ranging from 10 mg/L to 75
mg/L and a sensory threshold value of 100 mg/L
(Schreier 1979, Berg et al. 1955) (Table 1). Aldehydes
contribute to flavour with aroma descriptors such as
‘bruised apple’ and ‘nutty’ but can also be a sign of wine
oxidation. During fermentation, the most rapid accumu-
lation of acetaldehyde occurs when the rate of carbon
dissimilation is at its maximum, after which it falls to a low
level at the end of fermentation and then slowly increas-
es over time. Fermentation conditions such as medium
composition, nature of insoluble material used to clarify
the must, and extreme aerobic growth conditions greatly
affect acetaldehyde concentrations (Bennetzen and Hall
1982, Denis et al. 1983, Delfini and Costa 1993).
In wine, the amount of acetaldehyde can increase over
time due to oxidation of ethanol, activity of film yeast and
aeration (Fleet and Heard 1993). It has also been shown
that the use of high concentrations of sulfur dioxide in
grape must can result in an accumulation of acetalde-
hyde by the yeast. Furthermore, it was found that sulfite-
resistant strains of Saccharomyces cerevisiae produce much
more acetaldehyde than parental non-resistant strains
(Casalone et al. 1992). Acetaldehyde also increases with
increasing fermentation temperature: e.g. a fermentation
carried out at 30ºC resulted in a significantly higher con-
centration of acetaldehyde (Romano et al. 1994). How-
ever, in previous studies it was shown that temperature
does not affect aldehyde concentration at all (Amerine and
Ough 1980). Acetaldehyde concentration can also vary
considerably (from 6 to 190 mg/L) depending on the
yeast strain (Then and Radler 1971).
As the last precursor before ethanol is formed, acet-
aldehyde is one of the major metabolic intermediates in
yeast fermentation. Pyruvate, the end-product of glyco-
lysis, is converted to acetaldehyde through pyruvate
decarboxylase enzymes encoded by three genes, PDC1-3
(Pronk et al. 1996). Acetaldehyde is then converted to
ethanol through alcohol dehydrogenase enzymes, the
main one being encoded by the ADH1 gene. This step is
crucial for maintaining a redox balance in the cell, as it re-
oxidises NADH to NAD+, which is required for glycolysis
(Pronk et al. 1996).
The presence of acetaldehyde in white wines is an
indication of wine oxidation. The process of converting
ethanol to acetaldehyde in the presence of oxygen is also
referred to as ‘madeirisation’ and this produces a slightly
almondy flavour that resembles the fortified sweet wine,
Madeira. It is usually facilitated by prolonged storage in
a barrel at high temperatures and the resulting wine
lacks freshness and has a musty taste known as rancio
(Robinson 1999). Acetaldehyde in red wines can con-
tribute to aroma complexity as long as the concentration
does not exceed 100 mg/L.
Acetaldehyde, and indeed other yeast products, includ-
ing pyruvic acid and vinyl phenol, are involved in reac-
tions with anthocyanins and various wine phenolic com-
pounds to produce stable pigments in red wine (Bakker
and Timberlake, 1997, Benabdeljalil et al. 2000, Hayasaka
and Asenstorfer 2002, Eglinton et al. 2004). Acetaldehyde
can react with malvidin 3-O-glucoside to form the vinyl

Swiegers, Bartowsky, Henschke & Pretorius Microbial modulation of wine aroma and flavour 149
adduct of malvidin 3-O-glucoside (Bakker and Timberlake,
1997, Benabdeljalil et al. 2000). Via the enolic form,
acetaldehyde can form an ethyl-linked conjugate of
anthocyanin with proanthocyanidin (Timberlake and
Bridle 1976b). Yeast strain plays a role in these reactions,
presumably by producing different concentrations of
carbonyl and other metabolites during fermentation
(Benabdeljalil et al. 2000, Eglinton et al. 2003, Bartowsky
et al. 2004, Eglinton et al. 2004).
Another important carbonyl compound in wine is
diacetyl, which produces a ‘butter’ or ‘butterscotch’
aroma. At low concentrations it can be described as nutty
or toasty, but it becomes objectionable at concentrations
between 1 and 4 mg/L (Sponholz, 1993). Although yeasts
biosynthesise some diacetyl (0.2–0.3 mg/L) in wine, most
of it originates from the metabolic activities of lactic acid
bacteria, as discussed later (Laurent et al. 1991, Bartowsky
and Henschke 2004).
2.4 Volatile phenols
Volatile phenols (formed from the hydroxycinnamic acid
precursors in the grape must) have a relatively low
detection threshold and are, therefore, easily detected.
Although volatile phenols can contribute positively to the
aroma of some wines, they are better known for their
contribution to off-flavours such ‘Band-aid’, ‘barnyard’ or
‘stable’, which results from high concentrations of ethyl-
phenols (Dubois 1983). The prominent ethylphenols are
are 4-ethylguaiacol and 4-ethylphenol. Vinylphenols,
especially 4-vinylguaiacol and 4-vinylphenol, produce
a pharmaceutical odour, particularly in white wines
(Ribéreau-Gayon et al. 2000b).
Trace amounts of volatile phenols are present in grape
must, but they are predominantly produced by yeast dur-
ing fermentation (Baumes et al. 1988). The nonflavonoid
hydroxycinnamic acids, such as p-coumaric acid and fer-
ulic acid, are decarboxylated in a non-oxidative process by
Saccharomyces cerevisiae to form the volatile phenols 4-
vinylguaiacol and 4-vinylphenol, respectively (Chatonnet
et al. 1993). The Brettanomyces/Dekkera spp. yeasts are
well-known for their ability to form volatile phenols in
wine (Chatonnet et al. 1995, du Toit and Pretorius 2000).
These yeasts are associated with the more unpleasant
odourous ethylphenols, and are therefore regarded as
spoilage organisms resulting in aromas described as
‘Band-aid’, ‘medicinal’, ‘pharmaceutical’, ‘barnyard-like’,
‘horsey’, ‘sweaty’, ‘leathery’, ‘mouse urine’, ‘wet dog’,
‘smoky’, ‘spicy’, ‘cheesy’, ‘rancid’ and ‘metallic’ (Chatonnet
et al. 1995).
Phenolic acids can also be decarboxylated into volatile
phenols, usually first into 4-vinyl derivatives and then
reduced to 4-ethyl derivatives through enzymes called
phenolic acid decarboxylases (Cavin et al. 1993). Several
bacteria and fungi have been found to contain the genes
encoding phenolic acid decarboxylases and these genes
include PAD1 (also known as POF1) from Saccharomyces
cerevisiae, fdc from Bacillus pumilus, pdc from Lactobacillus
plantarum, padC from Bacillus subtilis and padA from
Pediococcus pentosaceus (Clausen et al. 1994, Zago et al.
1995, Cavin et al. 1997, Cavin et al. 1998, Barthelmebs et
al. 2000b). These enzymes are not inhibited by other
grape phenolics and they result in a high transformation
of the vinylphenol derivatives to the ethylphenol deriva-
tives.
In addition to the metabolic activity of yeast and
bacteria, other factors such as oak maturation can also
increase the amount of volatile phenols in wine (Pollnitz
et al. 2000). In particular, 4-ethylguaiacol and the 4-ethyl-
phenol concentrations showed a marked increase during
oak maturation.
The production of phenolic off-flavours by Saccharo-
myces uvarum brewing strains is dependent on the presence
of a functional allele of the PAD1/POF1 phenylacrylic acid
decarboxylase gene (Meaden and Taylor 1991, Hwang
1992, Clausen et al. 1994, Shinohara et al. 2000). The
phenylacrylic acid decarboxylase activity in yeast is
localised in the cytoplasm and its activity confers resistance
to cinnamic acids (Clausen et al. 1994).
However, Saccharomyces cerevisiae, unlike other organ-
isms, does not use its phenolic acid decarboxylase as its
sole defence against phenolic acid toxicity, probably
explaining why phenolic acid decarboxylase activity is so
low in most Saccharomyces cerevisiae strains (Barthelmebs et
al. 2000a).
Tannins do, however, inhibit the cinnamate decar-
boxylase enzyme of Saccharomyces cerevisiae, which might
be a factor in red wines where tannins are abundant
(Chatonnet et al. 1993). Pad1p does not commonly cause
odour formation in wine, whereas odour formation is
very common in beer, which might be due to a lack of
tannin inhibitors. Wine yeasts with optimised decarboxy-
lation activity on phenolic acids were developed by over-
expressing the Bacillus subtilis phenolic acid decarboxylase
gene (padC), the Lactobacillus plantarum p-coumaric acid
decarboxylase gene (pdc) and the Saccharomyces cerevisiae
phenylacrylic acid decarboxylase gene (PAD1/POF1) in
a laboratory strain of Saccharomyces cerevisiae (Σ1278b)
(Smit et al. 2003). The overexpression of padC and pdc in
Saccharomyces cerevisiae showed high enzyme activity, how-
ever, this was not the case for the PAD1/POF1-encoded
enzyme activity. Subsequently, the padC and pdc genes
were also overexpressed in the VIN13 commercial yeast
and these strains were compared with both the original
VIN13 host strain and a VIN13 mutant in which both
alleles of PAD1/POF1 were disrupted. Strains overexpress-
ing padC and pdc respectively, gave an approximate two-
fold increase in volatile phenol formation in a laboratory
strain of Saccharomyces cerevisiae (Σ1278b). Surprisingly, it
was also found that the overexpression of the padC gene
in wine yeasts resulted in wine with elevated levels of
favourable monoterpenes. In wine made with commercial
wine yeast VIN13 in which the PAD1/POF1 gene was dis-
rupted no volatile phenols could be detected (Smit et al.
2003). Further work might lead to ways through which
the concentration of volatile phenols in wine can be con-
trolled.
2.5 Esters
The production of esters by the yeast during fermentation
can have a significant effect on the fruity flavours in wine

150 Microbial modulation of wine aroma and flavour Australian Journal of Grape and Wine Research 11, 139–173, 2005
(Figure 7). The most significant esters are ethyl acetate
(fruity, solvent-like), isoamyl acetate (isopentyl acetate,
pear-drops aromas), isobutyl acetate (banana aroma),
ethyl caproate (ethyl hexanoate, apple aroma) and 2-
phenylethyl acetate (honey, fruity, flowery aromas)
(Table 1) (Thurston et al. 1981). Commercial wine strains
produce variable amounts of esters, such as isoamyl
acetate, hexyl acetate, ethyl hexanoate and ethyl
octanoate, which have a potential impact on the aroma
profile (Rankine 1977, Soles et al. 1982, Lambrechts and
Pretorius 2000, Marais 2001). However, there are sever-
al non-Saccharomyces wine yeasts that can contribute to the
ester aromas of wine. For example, mixed culture fer-
mentations by wild yeasts, such as Hanseniaspora guillier-
mondii and Pichia anomala, together with Saccharomyces
cerevisiae showed increased acetate ester concentrations
compared to fermentations with Saccharomyces cerevisiae
alone, without significantly affecting acetaldehyde, acetic
acid, glycerol and total higher alcohols (Rojas et al. 2003).
Although esters in wine are mainly produced by yeast
metabolism (through lipid and acetyl-CoA metabolism),
their production can be influenced by the grape variety. In
Pinot Noir wines the characteristic fruity flavours of plum,
cherry, strawberry, raspberry, blackcurrant and blackberry
characters were shown to be influenced by four distinct
esters: ethyl anthranilate, ethyl cinnamate, 2,3-dihydro-
cinnamate, and methyl anthranilate (Moio and Etiévant
1995). These esters are synthesised by the yeast from
grape precursors and have distinct aromas: sweet-fruity
and grape-like odour (ethyl anthranilate) and cinnamon-
like, sweet-balsamic, sweet-fruity, plum and cherry-like
flavour (ethyl cinnamate). The aroma of ethyl 2,3-
dihydrocinnamate is very similar to ethyl cinnamate, but
its contribution to the overall aroma is smaller (Moio and
Etiévant 1995).
It has been shown that Chardonnay wines character-
istically contain ethyl esters such as ethyl-2-methyl
propanoate, ethyl-2-butanoate, 3-methyl butanoate, ethyl
hexanoate, ethyl octanoate, ethyl decanoate, and the
acetate esters hexyl acetate, 2-methylbutyl acetate and
3-methylbutyl acetate. Although, Riesling wines contained
similar esters, 3-methyl butanoate and ethyl hexanoate
were found to be unimportant to the final aroma of the
wines (Smyth et al. 2005). Ester concentrations differed
among wine types, and there appears to be a synergy
between the grape and the yeast metabolism in estab-
lishing the characteristic ester blueprint of different grape
varieties.
The synthesis of acetate esters by Saccharomyces cere-
visiae is catalysed by a group of enzymes called alcohol
acetyltransferases (AAT) by utilising alcohols and acetyl-
CoA as substrates (Peddie 1990). Different Saccharomyces
AATase encoding genes have been cloned, namely ATF1,
ATF2, and LgATF (Fujii et al. 1994, Nagasawa et al. 1998,
Yoshimoto et al. 1998, Lilly et al. 2000, Mason and Dufour
2000).
The ATF1 gene was first cloned from Saccharomyces cere-
visiae and a brewery lager yeast, Saccharomyces uvarum
(Fujii et al. 1994). In this research, yeast strains with mul-
tiple copies of the Saccharomyces uvarum ATF1 gene and
subsequently elevated enzyme activity showed a 27-fold
increase in isoamyl acetate concentration and a nine-fold
Figure 7. A schematic representation of the
formation of ethyl acetate and isoamyl acetate in
wine yeast (based upon Swiegers and Pretorius
2005).

Swiegers, Bartowsky, Henschke & Pretorius Microbial modulation of wine aroma and flavour 151
increase in ethyl acetate concentration while the produc-
tion of ethanol and other higher alcohols did not change.
Furthermore, when the ATF1 gene was disrupted in Sac-
charomyces cerevisiae, AATase assays using isoamyl alcohol
as substrate indicated that, although the AATase activity of
the null mutant was dramatically reduced, 20% of the
activity was retained. However, when ethanol was used as
the substrate in the AATase assays, more than 80% of
the activity was retained. In support, it has been shown
that when overexpressing an ATF1 gene in the VIN13
commercial wine yeast, the concentrations of ethyl
acetate, isoamyl acetate and 2-phenylethyl acetate in wine
made with this yeast increased up to 10-fold, 12-fold,
and 10-fold, respectively (Lilly et al. 2000). These changes
in the wine composition had a pronounced effect on the
solvent or chemical and fruity or flowery characters of the
wines.
Researchers recently investigated the effect of
increased ester-synthesising (the ATF1- and ATF2-encod-
ed alcohol acetyltransferases and EHT1-encoded ethanol
hexanoyl transferase) and ester-degrading (the IAH1- and
TIP1-encoded esterases) enzyme activities on the flavour
profile of wine and distillates (Verstrepen et al. 2003, Lilly
2004). Esterases catalyse the reaction RCOOR1+ H2O →
R1OH + RCOOH (Peddie 1990). The balance between
ester-synthesising enzymes and esterases is important for
the net rate of ester accumulation. The ATF1, ATF2, EHT1,
IAH1 and TIP1 genes were overexpressed under the con-
trol of the constitutive PGK1 regulatory sequences in the
VIN13 commercial wine yeast. When the ester concen-
trations and aroma profiles of wines and distillates pre-
pared with these transformants were compared, it was
found that the overexpression of ATF1 and ATF2 increased
the concentrations of ethyl acetate, isoamyl acetate, 2-
phenylethyl acetate and ethyl caproate, while the over-
expression of IAH1 resulted in a significant decrease in the
concentrations of ethyl acetate, isoamyl acetate, hexyl
acetate and 2-phenylethyl acetate. The overexpression of
EHT1 resulted in a marked increase in the concentrations
of ethyl caproate, ethyl caprylate and ethyl caprate. The
overexpression of TIP1 did not decrease the concentrations
of any of the esters. The modification of the ester metab-
olism by the modified yeast had a pronounced effect on
the solvent/chemical and fruity/flowery characters of the
wines and distillates. The estery/synthetic fruit flavour
was overpowering in the wines fermented with the yeast
Figure 8. A schematic representation of the sulfur metabolism of wine yeast (based upon Wang et al. 2003).

152 Microbial modulation of wine aroma and flavour Australian Journal of Grape and Wine Research 11, 139–173, 2005
in which ATF1 was overexpressed, but much more subtle
in the strain overexpressing ATF2. An intense apple aroma
was detected in the wines produced by the yeast in which
EHT1 was overexpressed (Lilly 2004). These investiga-
tions into flavour-active esters represent progress towards
laying the foundation for the possible development of
wine yeast starter strains with optimised ester-producing
capabilities.
2.6 Sulfur compounds
Sulfur-containing flavour compounds typically occur in
wine at very low concentrations, have very low detection
thresholds and generally confer a negative sensory con-
tribution to wine (Table 2) (Peppard 1988, Mestres et
al. 2000, Vermeulen et al. 2005). On the basis of their
chemical structure, sulfur compounds in wine fall into
five different categories: sulfides, polysulfides, heterocyclic
compounds, thioesters and thiols. These compounds vary
widely in their sensory properties. Many sulfur com-
pounds are associated with negative descriptors, which
include cabbage, rotten egg, sulfurous, garlic, onion and
rubber (Rauhut 1993, Mestres et al. 2000, Vermeulen et
al. 2005), whereas some can contribute positive aromas to
wine, such as strawberry, passionfruit and grapefruit
(Tominaga et al. 1996, 1998a,b).
A variety of biochemical as well as chemical mecha-
nisms are involved in the formation of sulfur compounds
in wine and foods, however many of these mechanisms
are still poorly defined (Rauhut 1993, Mestres et al. 2000,
Vermeulen et al. 2005). The development of these sulfur
compounds by yeasts (Figure 8) include (i) the degrada-
tion of sulfur-containing amino acids; (ii) the degradation
of sulfur-containing pesticides; and (iii) the release and/or
the metabolism of grape-derived sulfur-containing pre-
cursors (Mestres et al. 2000).
2.6.1 Sulfides
Probably the best known sulfur compound in wine is
hydrogen sulfide, a highly volatile thiol which imparts a
‘rotten egg’ aroma and has a very low odour threshold.
Due to the frequent occurrence of this compound and
the low aroma threshold (50–80 µg/L), it is probably one
of the most common problems associated with the winery
(Rankine 1963, Acree et al. 1972, Eschenbruch 1974, Vos
and Gray 1979, Monk 1986, Henschke and Jiranek 1991,
Rauhut 1993). However, the problem is relatively easily
dealt with through the use of copper (which results in the
formation of copper sulfide) or aeration (resulting in oxi-
dation of the sulfide) (Monk 1986). Nevertheless, elimi-
nation of the use of copper salts by wineries is a desirable
food processing goal and the presence of oxidised sulfur
compounds in young wine could be related to the reduc-
tive character in bottled wine.
Hydrogen sulfide can be formed metabolically by yeast
from either inorganic sulfur compounds, sulfate and sulfite,
or organic sulfur compounds, cysteine and glutathione
(Rankine 1963, Eschenbruch 1974, Eschenbruch et al.
1978, Monk 1986, Henschke and Jiranek 1993, Rauhut
1993, Hallinan et al. 1999, Spiropoulis et al. 2000). Cell
growth creates a metabolic requirement for the organic
sulfur compounds, including cysteine, methionine, S-
adenosyl methionine and glutathione. When these organ-
ic compounds are absent, the cell must synthesise them
from inorganic sulfur compounds accumulated from must.
Under certain conditions, sulfide is liberated during the
reduction of inorganic sulfur to become detectable by the
winemaker. The concentration of hydrogen sulfide pro-
duced varies with the availability of sulfur compounds,
yeast strain and fermentation conditions, and the nutri-
tional status of the environment (Henschke and Jiranek
1991, Rauhut 1993, Spiropoulis et al. 2000). However,
Table 2. Sulfur compounds, including thiols, commonly found in wine (Acree et al. 1972, Rauhut 1993, Tominaga et
al. 1996, 1998a,b, Dubourdieu 2000, Mestres et al. 2000, Ribéreau-Gayon et al. 2000b, Murat 2001a,b, Vermeulen et
al. 2005).
Compound Concentration Aroma threshold Aroma descriptor
in wine (µg/L) (µg/L)
Hydrogen sulfide Trace–>80 10–80 rotten egg
Methanethiol (methyl mercaptan) 5.1, 2.1 0.3 cooked cabbage, onion, putrefaction, rubber
Ethanethiol (ethyl mercaptan) 1.9–18.7 1.1 onion, rubber, natural gas
Dimethyl sulfide 1.4–61.9 25 asparagus, corn, molasses
Diethyl sulfide 4.1–31.8 0.93 cooked vegetables, onion, garlic
Dimethyl disulfide 2 15, 29 cooked cabbage, intense onion
Diethyl disulfide Trace–85 4.3 garlic, burnt rubber
3-(Methylthio)-1-propanol (methionol) 140–5000 500 cauliflower, cabbage, potato
Benzothiazole 11 50 rubber
Thiazole 0–34 38 popcorn, peanut
4-Methylthiazole 0–11 55 green hazelnut
2-Furanmethanethiol 0–350 ng/L 1 ng/L roasted coffee, burnt rubber
Thiophene-2-thiol 0–11 0.8 burned, burned rubber, roasted coffee
4-Mercapto-4-methylpentan-2-one (4MMP) 0–30 ng/L 3 ng /L cat urine, box tree/ blackcurrant, broom
3-Mercaptohexan-1-ol (3MH) 50–5000 ng/L 60 ng/L passionfruit, grapefruit
3-Mercaptohexyl acetate (3MHA) 1–100 ng/L 4 ng /L Riesling-type note, passionfruit, box tree

Swiegers, Bartowsky, Henschke & Pretorius Microbial modulation of wine aroma and flavour 153
some strains appear to form unregulated amounts of
hydrogen sulfide and presumably represent metabolic
defects, at least in the wine environment (Jiranek et al.
1995a,b, Spiropoulis et al. 2000, Mendes-Ferreira et al.
2002).
Two distinct phases of hydrogen sulfide production
are evident during fermentation (Henschke and Jiranek
1991, Thomas et al. 1993, Henschke and de Kluis 1995,
Park et al. 2000). Hydrogen sulfide produced during the
early to middle stages of fermentation is associated with
yeast growth and typically responds to nutrient addition,
especially diammonium phosphate (DAP) (Vos and Gray
1979, Monk 1982, Stratford and Rose 1985, Henschke
and Jiranek 1991, Giudici and Kunkee 1994, Jiranek et al.
1995a,b, Rauhut et al. 1996, Hallinan et al. 1999, Park et
al. 2000, Mendes-Ferreira et al. 2002, Moreira et al. 2002,
Wang et al. 2003). When availability of pantothenic acid
is limited, addition of this vitamin can reduce hydrogen
sulfide formation (Eschenbruch et al. 1978, Slaughter and
McKernan 1988, Wang et al. 2003). Pantothenic acid is a
cofactor of CoA, which is needed for the synthesis of O-
acetylserine and O-acetylhomoserine (OAS/OAH), the
nitrogen-containing compounds that combine with
hydrogen sulfide.
The mechanism(s) for hydrogen sulfide formation
during the final stages of fermentation are not clear
(Henschke 1996, Park et al. 2000). In white wine fer-
ments, hydrogen sulfide formation is inversely correlated
with initial total nitrogen, and glutathione measured after
fermentation (Park et al. 2000). During the final stages of
fermentation in red wine ferments, however, hydrogen
sulfide production appears to be unresponsive to DAP
addition, but, at least in several cases, some evidence sug-
gests that aeration and vitamin addition can moderate
hydrogen sulfide production (Henschke 1996). The
involvement of a deficiency of certain vitamins or the
degradation of S-reserves, which is derepressed by a defi-
ciency of some nutrients, is suggested by the observation
that nutrient levels are typically very low during the final
stages of fermentation in some viticultural regions
(Henschke and de Kluis 1995). The impaired ability of
yeast to take up the low concentrations of nutrients dur-
ing the final stages of fermentation in the presence of
inhibitory concentrations of ethanol could be another
explanation (van Uden 1989, Iglesius et al. 1991).
Although many factors have been reported to affect
hydrogen sulfide production, including yeast strain, sulfur
source, nitrogen composition and nitrogen concentration
of the test medium, the strain of yeast is the key to deter-
mining hydrogen sulfide production; surveys show that
strains vary widely in their ability to produce hydrogen
sulfide (Zambonelli 1964, Rankine 1968, Eschenbruch et
al. 1978, Thornton and Bunker 1989, Thomas et al. 1993,
Giudici and Kunkee 1994, Jiranek et al. 1995b, Rauhut et
al. 1996, Spiropoulis et al. 2000, Mendes-Ferreira et al.
2002).
Grape must is typically deficient in organic sulfur com-
pounds (less than 10 mg/L cysteine and methionine),
which signals yeast to synthesise organic sulfur com-
pounds from inorganic sources, normally plentiful in
grape must (Henschke and Jiranek 1993, Park et al. 2000,
Moreira et al. 2002). Hydrogen sulfide is, therefore, a
metabolic intermediate in the reduction of sulfate or sul-
fite needed for the synthesis of organic sulfur compounds
(Figure 8). When these reactions proceed in the presence
of a suitable nitrogen supply, hydrogen sulfide is seques-
tered by O-acetyl serine and O-acetyl homoserine, which
are derived from nitrogen metabolism, to form the organ-
ic sulfur compounds. Under some conditions, however,
when insufficient or unsuitable nitrogen sources are avail-
able, free hydrogen sulfide can accumulate in the cell and
diffuse into the fermenting must (Vos and Gray 1979,
Stratford and Rose 1985, Henschke and Jiranek 1991,
Giudici and Kunkee 1994, Jiranek et al. 1995a, Jiranek et
al. 1996).
In Saccharomyces cerevisiae, hydrogen sulfide is the prod-
uct of the Sulfate Reduction Sequence (SRS) pathway and
acts as an intermediate in the biosynthesis of sulfur-
containing amino acids (Yamagata 1989, Rauhut 1993,
Thomas and Surdin-Kerjan 1997). The ability of a strain
to produce hydrogen sulfide is, at least, partly genetic,
since hydrogen sulfide production by different wine
strains varies under the same conditions (Thornton and
Bunker 1989, Henschke and Jiranek 1993, Jiranek et al.
1995b, Spiropoulis et al. 2000). Mendes-Ferreira et al.
(2002) recently screened a large selection of commercial
wine yeast, in addition to non-Saccharomyces yeasts, which,
when tested under identical physiological conditions, all
had the same growth characteristics but varied in sulfite
reductase (the enzyme producing hydrogen sulfide) activ-
ity. After fermentation in grape musts, yeast strains could
be classified as nonproducers of hydrogen sulfide, must-
composition-dependent producers and invariable pro-
ducers (Mendes-Ferreira et al. 2002).
The first step of the SRS metabolic pathway involves
the transport of sulfate from the medium into the yeast
cell through the sulfate permease (Figure 8). Sulfate is
then reduced to sulfide through a series of steps using the
enzymes ATP-sulfurylase (using two ATP molecules) and
sulfite reductase. The next step leads to the sequestering
of the sulfide: O-acetylserine (from the amino acid serine)
combines with sulfide to form cysteine, and O-acetylhomo-
serine (from the amino acid aspartate) to form homo-
cysteine, which can then be converted to methionine.
A study has recently investigated the role of the
bifunctional O-acetylserine/O-acetylhomoserine sulf-
hydrylase as means to modulate hydrogen sulfide pro-
duction by industrial yeast. Overexpression of the MET17
gene, which encodes O-acetylserine/O-acetylhomoserine
sulfhydrylase, in a strain of Saccharomyces cerevisiae results
in greatly reduced hydrogen sulfide formation. However,
this was not the case with another strain, indicating that
O-acetylserine/O-acetylhomoserine sulfhydrylase activity
is not directly related to hydrogen sulfide formation
(Spiropoulos and Bisson 2000).
Overexpression of the two genes MET14 and SSU1
have been shown to increase the formation of sulfite
(Donalies and Stahl 2002). It has, therefore, been postu-
lated that the deletion of the MET14 adenosylphospho-
sulphate kinase gene or the MRX1 methionine sulfoxide

154 Microbial modulation of wine aroma and flavour Australian Journal of Grape and Wine Research 11, 139–173, 2005
reductase gene might be the most effective way to prevent
wine yeast from producing hydrogen sulfide (Pretorius
and Bauer 2002, Pretorius 2003, 2004).
A novel genetic approach that involved modifying the
activity of the key enzyme, sulfite reductase, by protein
engineering one of the enzyme subunits has been in-
vestigated (Sutherland et al. 2003). Sulfite reductase is a
heterotetramer, consisting of two α- and two β-subunits,
which are encoded by MET10 and MET5 genes, respec-
tively (Kobayashi and Yoshimoto 1982, Hansen et al.
1994, Sutherland et al. 2003). The enzyme, a hemoflavo-
protein, binds the cofactors flavin adenine dinucleotide,
flavin mononucleotide and siroheme. Mutations were
introduced into the MET10 gene such that the α-subunit
could no longer bind cofactor but could still form a het-
erotetramer protein complex with the β-subunit. In this
way, overexpression of the mutant met10 gene would pro-
duce a nonfunctional subunit which could reduce the
proportion of functional sulfite reductase in the cell, and
hence reduce sulfide formation. Further work is required
to demonstrate whether this genetic strategy will be effec-
tive in the wine fermentation.
Because the concentrations of the amino acids cys-
teine and methionine in grape juices are typically not suf-
ficient to meet the metabolic needs of growing cells, the
SRS metabolic pathway is induced in order to meet this
demand (Henschke and Jiranek 1993). When adequate
nitrogen is also present in the medium, sufficient precur-
sors for these amino acids (O-acetylserine and O-acetyl-
homoserine) will be available to sequester the sulfide.
However, if nitrogen is limiting, insufficient precursors
will be available. Therefore, the SRS pathway will be acti-
vated and sulfide will accumulate due to the lack of pre-
cursors. Surplus sulfide is then liberated from the cell as
hydrogen sulfide (Thomas and Surdin-Kerjan 1997). For
some strains, the problem can be worse when sulfite is
present in the ferment because extracellular sulfite read-
ily diffuses into the cell, resulting in a steady production
of hydrogen sulfide. Therefore, in conditions of nitrogen
depletion, high and continuous production of hydrogen
sulfide is observed in the presence of sulfite (Stratford
and Rose, 1985, Jiranek et al. 1995a,b). Sulfate availabil-
ity can also influence sulfide formation (Hallinan et al.
1999).
Other environmental factors that can affect hydrogen
sulfide production include: (i) high residual levels of
elemental sulfur; (ii) presence of sulfur dioxide; (iii) pres-
ence of sulfur-containing organic compounds; (iv) panto-
thenate deficiency; (v) high threonine content relative to
other amino acids; and (vi) relative methionine to ammo-
nium concentrations (Monk 1986, Henschke and Jiranek
1991, Rauhut 1993, Spiropoulos et al. 2000).
Cells which undergo autolysis after fermentation can
also release hydrogen sulfide (Suomalainen and Lehtonen
1979). It has been suggested that sulfur-containing amino
acids are degraded, but the mechanism involved is unclear
(Henschke and Jiranek 1993). Cells accumulate gluta-
thione during growth and it has been estimated that this
peptide can contribute up to 40% of the sulfide produced
by nitrogen-starved cells (Hallinan et al. 1999). It has also
been suggested that conditions that produce unhealthy
cells, such as during a sluggish or stuck fermentation, are
more likely to promote autolysis (Berry and Watson
1987). Uninoculated (feral) fermentations, in which there
is a considerable population of low alcohol tolerant non-
Saccharomyces species, could be another example. These
yeasts become inhibited by the increasing concentration of
ethanol, lose viability and presumably autolyse during
the early to mid phases of fermentation (Henschke and
Jiranek 1991, Fleet and Heard 1993).
Hydrogen sulfide is a highly reactive species, which can
take part in a range of reactions to generate compounds
that impact on the flavour of a wine (Vermeulen et al.
2005). For example, mercaptans such as ethanethiol can
be formed by the reaction of hydrogen sulfide with ethanol
or acetaldehyde (Amerine et al. 1980, Rauhut 1993).
The formation of dimethyl sulfide (DMS), which elic-
its odours described as ‘asparagus’, ‘corn’ and ‘molasses’,
is not clear. It could be formed in a similar way to other
mercaptans. The concentration of DMS found in wine is
well above the sensory threshold of 25 µg/L (white wine)
and 60 µg/L (red wine). DMS is considered a beneficial
compound in low concentrations, contributing to the
aroma of bottle age. The formation of dimethyl sulfide
happens during wine maturation, through a yeast mech-
anism by cleavage of S-methyl-L-methionine to homo-
serine and dimethyl sulfide. In beer production, heat
decomposition during malting of S-methylmethionine
produces dimethyl sulfoxide that can reduce to dimethyl
sulfide, presumably during storage (Rauhut 1993). DMS
formation during fermentation has also variously been
linked to cysteine, cystine or glutathione metabolism in
yeast (Rauhut 1993, Ribéreau-Gayon et al. 2000b).
One mechanism for formation of the polysulfides,
dimethyl disulfide, dimethyl trisulfide and dimethyl tetra-
sulfide is believed to involve oxidation of the mercap-
tans, e.g. oxidation of methyl mercaptan to form dimethyl
disulfide. Yeast can reduce disulfides to mercaptans. These
compounds, which elicit a ‘rubber’ or ‘garlic’ odour can-
not be removed by copper fining.
The mercaptans, including methyl mercaptan and
ethyl mercaptan, are highly reactive compounds with low
aroma thresholds. The aroma of ethyl mercaptan is
described as ‘onion’ or ‘rubber’ with a threshold value of
1.1 µg/L in wine (Goniak and Noble 1987). These mer-
captans are observed to form during fermentation in asso-
ciation with hydrogen sulfide. Their suppression by DAP
suggests that they are produced as by-products of yeast
metabolism of methionine (Rauhut 1993). Thioacetic acid
esters of these mercaptans are also observed to form dur-
ing fermentation, and these can slowly hydrolyse to the
parent mercaptan at a later stage (Rauhut et al. 1996).
Disulfides can be reduced to mercaptans by the action
of sulfite ions, which can then be removed by copper or
silver (not permitted in some countries) fining (Bobet et
al. 1990). However, the disulfides left as by-products of the
reaction cannot be removed by copper ions, and not all of
the off-flavours can, therefore, be removed in this way.

Swiegers, Bartowsky, Henschke & Pretorius Microbial modulation of wine aroma and flavour 155
2.6.2 Sulfur-containing fusel alcohols
Methionine can be metabolised by yeast to form sulfur-
containing fusel alcohol, methionol or 3-methylthio-1-
propanol, which has ‘cauliflower’ and ‘cabbage’ odours
(Mestres et al. 2000). This compound can be converted
further to 3-methylthiopropyl acetate, which has a ‘mush-
room’ or ‘garlic’ odour. It has also been proposed that 4-
methylthio-1-butanol with an onion/garlic odour and 2-
mercapto-1-ethanol with a ‘poultry’/‘farmyard’ odour can
be biosynthesised by yeast in the same way by using the
amino acids homocysteine and cysteine, respectively
(Mestres et al. 2000).
2.6.3 Thiols
The volatile thiols are one of the most potent groups of
aroma compounds found in wine, some imparting nega-
tive aromas, others contributing positively. Furfurylthiol is
a potent aroma compound identified in Bordeaux red
wines, white Petite Manseng, and also in toasted barrel
staves (Tominaga et al. 2000b). Furfurylthiol has also been
found in roasted coffee, meat, wheat bread and popcorn,
with a perception threshold of 0.4 ng/L (Tominaga et al.
2000b). Its presence in wine has been shown to be the
result of yeast transformation of furfural released from
toasted oak staves during fermentation (Blanchard et al.
2001). These authors showed that fermentations that
have an added nitrogen source, such as asparagine, do not
produce as much furfurylthiol. Therefore, production of
furfurylthiol is linked to the production of the HS–anion,
which is not produced when ammonium sulfate is added
in sufficient quantities in a fermentation (Blanchard et al.
2001).
The volatile thiols 4-mercapto-4-methylpentan-2-one
(4MMP), 3-mercaptohexan-1-ol (3MH) and 3-mercapto-
hexyl acetate (3MHA) are of particular importance to
wine aroma. These sulfur-containing compounds (thiol
referring to the SH functional group) have extremely low
perception thresholds: 3 ng/L (4MMP), 60 ng/L (3MH)
and 4 ng/L (3MHA). In Sauvignon Blanc wine, these
compounds are of particular importance to the varietal
character as it imparts box tree (4MMP), passionfruit,
grapefruit, gooseberry and guava aromas (3MH and
3MHA) (Dubourdieu et al. 2000). However, 4MMP, 3MH
and 3MHA have also been identified in wines made from
Colombard, Riesling, Semillon, Merlot and Cabernet
Sauvignon in varying concentrations and can, therefore,
potentially impact the aroma (Tominaga et al. 2000a,
Murat et al. 2001b).
The volatile thiols are almost non-existent in the grape
juice and only develop during fermentation. Therefore, it
has been proposed that the wine yeast, Saccharomyces cere-
visiae, is responsible for the formation of volatile thiols.
However, it has been shown that 4MMP and 3MH do
exist in the grapes but in the form of non-volatile, cysteine
bound conjugates and that yeast are responsible for the
cleaving of the thiol from the precursor (Darriet et al.
1995).
A mechanism of thiol release was proposed on the
basis of experiments showing that a cell-free enzyme
extract of the bacteria Eubacterium limosum containing
Figure 9. A schematic representation
of the biosynthesis of sterols and
terpenes in wine yeast (based upon
Carrau et al. 2005).

156 Microbial modulation of wine aroma and flavour Australian Journal of Grape and Wine Research 11, 139–173, 2005
carbon-sulfur lyase enzymes can release 4MMP from its
precursor S-4-(4- methylpentan-2-one)-L-cysteine (Cys-
4MMP) (Tominaga et al. 1995). Therefore, it was sug-
gested that the amplification of Sauvignon Blanc varietal
aromas during fermentation occurs through the action of
yeast carbon-sulfur lyases (Tominaga et al. 1998a,b). Work
at AWRI has investigated how yeasts are affected in their
ability to release 4MMP from Cys-4MMP when genes
encoding putative yeast carbon sulfur lyases are deleted.
Four genes that influence the release of the volatile thiol
4MMP in a laboratory strain were identified, indicating
that the mechanism of release probably involves multiple
genes. These findings were confirmed in a homozygous
derivative of the commercial wine yeast, VL3, showing
that deletion of the genes leads to a decrease in the
amount of 4MMP released (Howell et al. 2005). In relat-
ed thiol research, these workers showed that the volatile
thiol 3MHA is formed by yeast from 3MH by the action of
the ester forming alcohol acetyltransferase, encoded by the
ATF1 gene (Swiegers et al. 2005). Hereby, the link in ester
and volatile thiol metabolism in yeast was established for
the first time.
The laboratory of Dubourdieu has shown in model
ferments that when the chemically synthesised precursor,
S-3-(hexan-1-ol)-L-cysteine (Cys-3MH) decreases in con-
centration, 3MH increases. However, only a small fraction
(1.6% at day 6 of fermentation) of the cysteine-bound
precursor was released as 3MH (Dubourdieu et al. 2000).
In Cabernet Sauvignon and Merlot musts, it was shown
that the amount of 3MH released was proportional to the
Cys-3MH concentration. Therefore, the higher the con-
centration of the cysteine conjugate thiol precursors in the
must are, the higher the volatile thiol concentration in the
resulting wine will be (Murat et al. 2001a). However, on
average, only 3.2% of the precursor was released during
fermentation. It is, therefore, clear that there is a huge,
untapped flavour potential remaining in the wine after
fermentation but that this source of flavour is not fully
utilised due to the metabolic limitations of the yeast cell.
The amount of 4MMP released in wine ferments is
dependent on which yeast strain is used to conduct the
fermentation (Dubourdieu et al. 2000). Therefore, the
genetic and physiological characteristics of the yeast strain
have a huge effect on its ability to release thiols. Com-
mercially available wine strains Saccharomyces cerevisiae
VL3 and EG8 release more thiols than strains VL1 and
522d. Additionally, Saccharomyces bayanus strains release
more 4MMP than Saccharomyces cerevisiae strains VL3 and
EG8. Wines made with hybrids produced between
Saccharomyces bayanus and Saccharomyces cerevisiae have
been shown to contain more of the volatile thiols (Murat
et al. 2001b). Work at the AWRI has confirmed these
findings by showing that different commercial wine strains
have variable abilities to release of 4MMP from the Cys-
4MMP precursor in model ferments. Commercial wine
yeast strains that release even more thiols than VL3 were
identified (Howell et al. 2004). Furthermore, the ability of
different commercial wine yeast to bioconvert 3MH to
3MHA was also investigated. Large variations in the
ability of commercial wine yeast to convert 3MH were
observed and in most cases this did not correspond to the
ability to release 4MMP (Swiegers et al. 2005). Therefore,
yeast strain selection is an important tool that can assist
winemakers in creating specific wine styles according to
consumer preferences.
2.7 Monoterpenoids
Monoterpenoids are potent aroma compounds that are
produced by higher plants, algae, fungi and even some
yeast, from a common precursor, geranyl pyrophosphate
(GPP) (Figure 9). In particular, two of the plant species
that produce monoterpenoids are V. vinifera (grapes) and
Humulus lupulus (hops) (King and Dickinson 2000). Some
fungal (Penicillium) and yeast species are also able to pro-
duce monoterpenoids (Larsen and Frisvad 1994, 1995).
Yeast species that produce terpenoids include Kluyvero-
myces lactis, Torulaspora delbrueckii (formerly Saccharomyces
fermentati) and Ambrosiozyma monospora (Drawert and
Barton 1978, Fagan et al. 1981, Klingenberg and Sprecher
1985). These compounds could have significant value to
the winemaker and brewer.
Although mutant strains of Saccharomyces with a genet-
ic defect in the sterol pathway have been reported, native
strains of Saccharomyces cerevisiae are capable of producing
only trace amounts relative to concentrations present in
wines. A recent survey of native wine strains, isolated in
Uruguay, has shown that several are capable of significant
production. In order to avoid interference from grape-
derived monoterpenes, the experiments were conducted
in chemically-defined media free from terpenes or their
glycosides. Furthermore, fermentation conditions could be
used to enhance monoterpene production. High nitro-
gen – 400 mg N/L compared with 180 mg N/L – which
stimulated fermentation rate but not biomass yield, also
stimulated monoterpene production. Interestingly, for-
mation of the sesquiterpenes, nerolidol and farnesol, was
not stimulated (Carrau et al. 2005).
To explain these results, Carrau and colleagues (2005)
searched the Saccharomyces genome database (www.
yeastgenome.org) for the presence of appropriate genes
using the web-based BLAST (Basic Local Alignment
Search Tool; Altschul et al. 1990) search procedure. From
these searches, they hypothesised that monoterpene
biosynthesis could proceed by an alternative pathway
which does not involve the sterol pathway from which
sesquiterpenes appear to be derived. This alternative path-
way, which is located in the mitochondrion, involves the
conversion of leucine to mevalonic acid. This fact could
explain the non-coordinated synthesis of the mono-
terpene and sesquiterpene groups (Carrau et al. 2005).
This work suggests that some strains of Saccharomyces
yeast could contribute to the floral aroma of wine by de
novo synthesis of monoterpenes, and this contribution
could be augmented by certain fermentation conditions
such as musts with higher concentrations of assimilable
nitrogen like the ammonium ion, in combination with
microaerobic fermentation (Carrau et al. 2005). A
schematic representation of the biosynthesis of sterols
and terpenes in wine yeast is shown in Figure 9.
The study of the aromatic potential of some fruits,

Swiegers, Bartowsky, Henschke & Pretorius Microbial modulation of wine aroma and flavour 157
such as grape, passionfruit, papaya, raspberry, as well as of
their fermented products (juice and wine) has revealed
that, besides a free fraction of volatile terpenoids, naturally
non-odourous and non-volatile precursors exist that rep-
resent an important source of fragrant compounds.
The aglycone moiety of the precursor glucoside can be
linked to β-D-glucose or to the disaccharides 6-O-α-L-
arabinofuranosyl-β-D-glucopyranose, 6-O-α-L-rhamno-
pyranosyl-β-D-glucopyranose and 6-O-β-D-apiofuranosyl-
β-D-glucopyranose (Günata et al. 1985, Voirin et al.
1990). Terpenols such as linalool, nerol, geraniol, α-ter-
pineol, citronellol, and in some cases linalool oxides and
terpene diols and triols, can act as aglycone precursors.
Aliphatic or cyclic alcohols, such as hexanol, 2-phenyl-
ethanol, benzylalcohol, C13-norisoprenoids and and
volatile phenols such as vanillin are also possible precur-
sors (Günata et al. 1985, Park and Noble 1993).
During winemaking, bound terpenoids can be released
by the action of glycosidase enzymes which are produced
by the grapes, yeast and bacteria. Therefore, increasing
glucosidase enzyme activity is a tool for enhancing the ter-
penoid aromas in wines.
The aromatic grape varieties, such as Muscat, Riesling
and Gewürztraminer, contain large amounts of the
monoterpenes geraniol and nerol. Geraniol has aromas
described as rose-like and linalool aromas described as
rose, whereas linalool oxides are described as camphorous
and nerol oxides as vegetative (Simpson 1979). In general,
more bound glycosides are found than the free terpenoids,
and the ratios of bound to free terpenoids can also vary
amongst different grape cultivars. Muscat of Alexandria
grapes, for example, have a ratio of 5:1, whereas some
non-Muscat varieties have a ratio of 1:1 (Williams et al.
1984).
Besides enzymatic hydrolysis, chemical acid hydro-
lysis can be used for the release of the monoterpenes
from their glycosidically-bound, non-volatile precursors.
Previously, acid hydrolysis was thought to be an effective
method for monoterpene liberation, but studies have
shown that high temperature acid hydrolysis results in
unwanted rearrangement of the monoterpene aglycones
(Usseglio Tomasset and Di Stefano 1980, Williams et al.
1982 a). However, enzymatic hydrolysis is an efficient
method to release monoterpenes and it does not result in
modification of the aromatic character (Günata et al.
1985). Enzymatic hydrolysis of monoterpenes involves
two steps. In the first step, an α-L-rhamnosidase and an α-
L-arabinofuranosidase or a β-D-apiofuranosidase (depend-
ing on the structure of the aglycone moiety) cleave the
1,6-glycosidic linkage, and in the second step the mono-
terpenols are liberated from the monoterpenyl β-D-gluco-
sides by the action of a β-glucosidase (Günata et al. 1988,
1990).
Interestingly, the origin of the enzyme and the
structure of the aglycone determine the efficiency of the
hydrolysis of monoterpenyl β-D-glucosides by β-glucosi-
dases. Maturation of grapes results in the cleavage of
monoterpenyl β-D-glucosides by endogenous grape β-
glucosidases. However, these enzymes exhibit almost non-
existent activity towards grape terpenyl-glycosides in must
and wine, probably because they are inhibited by glucose
and exhibit poor stability at the low pH and high ethanol
concentration (Bayonove et al. 1984, Aryan et al. 1987).
Some strains of Saccharomyces cerevisiae possess β-glucosi-
dase activity. However, their activity towards glycoside
precursors seems to be very low (Günata et al. 1986,
Delcroix et al. 1994, Hernández et al. 2003).Therefore, the
addition of functional exogenous β-glucosidase to a
fermentation is the most effective way to improve the
hydrolysis of the glycoconjugated aroma compounds to
enhance wine flavour (Aryan et al. 1987, Shoseyov et al.
1990, Vasserot et al. 1993).
Non-Saccharomyces yeasts such as Brettanomyces/Dekkera,
Candida, Debaryomyces, Hanseniaspora and Pichia have been
screened for novel β-glucosidases with the desired prop-
erties (Vasserot et al. 1989, Rosi et al. 1994, McMahon et
al. 1999, Fernández et al. 2000, Garcia et al. 2002). For
wine purposes, these glycosidases need to have: (i) high
affinity for grape-derived terpenoid aglycones; (ii) optimal
activity at wine pH (pH 2.5–3.8); (iii) resistance to glucose
inhibition; and (iv) high tolerance to ethanol (Riou et
al. 1998). Recently, the β-glucosidase from Debaryomyces
pseudopolymorphus was found to be suitable for use under
wine conditions (Cordero Otero et al. 2003). It exhibits
resistance to wine-associated inhibitory compounds such
as glucose, ethanol and sulfur dioxide. Its optimum pH lies
within the wine spectrum (pH 2.5–3.8), and it has high
substrate affinity and large aglycone-substrate recogni-
tion. Addition of the enzyme to Chardonnay ferments
resulted in increased concentrations of citronellol, nerol
and geraniol. Due to the financial cost of the addition of
exogenous aroma-liberating enzyme preparations to wine
efforts have been made to express heterologous glucosi-
dase enzymes in wine yeast. Indeed, the aroma intensity
of wine made with a yeast expressing the β-1,4-glucanase
gene from Trichoderma longibrachiatum was shown to be
more intense than the wild-type (Vilaneuva et al. 2000).
For the same reason the β-glucosidase genes (BGL1 and
BGL2) of Saccharomycopsis fibuligera, the β-L-arabinofura-
nosidase (ABF2) of Aspergillus niger and a glucanase-encod-
ing gene cassette consisting of several glucanase genes
(BEG1,END1 and EXG1) were expressed in wine yeast
(Pretorius 2000, van Rensburg and Pretorius 2000,
Pretorius 2003, Pretorius and Bauer 2002, Pretorius 2004).
Surprisingly, the wines produced by the VIN13 commer-
cial wine yeast transformed with the Saccharomycopsis fibu-
ligera BGL1 and BGL2 β-glucosidase genes also contained
increased ester concentrations (van Rensburg et al. 2005).
Many of these fragrant compounds, when produced in
appropriate concentrations, would contribute to the fer-
mentation bouquet of wine. It is unclear at the moment
whether this acquired capacity of the transformed wine
yeast is of practical significance in large-scale wine pro-
duction, but it is fertile soil for further investigation.
C13-norisoprenoids are also considered to be important
to the aroma of wine. Although there is no ‘concrete
proof’, the role of yeast and bacteria in the release of C13-
norisoprenoids is very probable. C13-norisoprenoids are
secondary metabolites formed in the grape berry and
many accumulate as non-volatile glycosides (Winterhalter

158 Microbial modulation of wine aroma and flavour Australian Journal of Grape and Wine Research 11, 139–173, 2005
and Schreier 1994). These C13-norisoprenoids are released
from their glycosidic precursors during winemaking. As
with monoterpenes, precursor analysis of grapes is often
used to identify norisoprenoids in grapes and wine
(Ribéreau-Gayon 2000b, Sefton et al. 1993, 1994, 1996,
Sefton 1998). Structurally, norisoprenoids can be divided
into two main groups, megastigmane and non-megastig-
mane (Ribéreau-Gayon et al. 2000b).
The megastigmanes have complex aromas. A particu-
larly significant megastigmane, β-damascenone, is thought
to have the aroma of flowers, tropical fruit, and stewed
apple. β-Damascenone has a particularly low threshold
concentration and is thought to be present in all varieties
of grapes, often above its threshold concentration.
Consequently, this compound is thought to play a role in
the aroma of some wines. Analysis of 52 young red wines
found that β-damascenone was present in all wines above
its threshold concentration (Ferreira et al. 2000). Other
megastigmanes such as β-ionone and α-ionone are
thought to be important to the aroma of some wines
(Ribéreau-Gayon et al. 2000b, Kotseridis et al. 1998). In
one study, β-ionone was found above its threshold con-
centration in a range of young red wines, though α-
ionone was found to be below its threshold concentration
(Ferreira et al. 2000).
Researchers have identified non-megastigmane noriso-
prenoids as particularly active aroma compounds in wine;
the most important of these is 1,1,6-trimethyl-1,2-dihydro-
napthalene (TDN), which has a distinctive kerosene aroma
(Winterhalter 1991). TDN plays an important role in the
aroma of certain old Riesling wines. It is thought that
TDN is also present in other wine varieties, although at
concentrations below its threshold (Etiévant 1991,
Ribéreau-Gayon et al. 2000b, Marais et al. 1992, Simpson
and Miller 1983).
3. The modulation of wine flavour by bacteria
Wine is a chemically hostile environment for bacteria.
Acetic acid and lactic acid bacteria are the only families of
bacteria commonly found in grape juice and wine: lactic
acid bacteria play a more important role in winemaking,
whereas the acetic acid bacteria are only considered to be
spoilage organisms due to the formation of major oxi-
dised products, such as acetaldehyde and acetic acid. Only
four genera of the lactic acid bacteria genera – Lactobacillus,
Leuconostoc, Oenococcus and Pediococcus – are able to survive
the unfavourable conditions (low pH, high ethanol
concentration and low nutrients) present in wine to
any extent. Oenococcus oeni is the most well adapted wine-
associated species and is used almost exclusively for the
induction of malolactic fermentation (MLF) in red, white
and sparkling base wines (Wibowo et al. 1985, Henick-
Kling 1993, Henschke 1993).
Research in progress is showing that these bacteria
can modify some of the components and sensory proper-
ties of wine, providing a new opportunity to alter the
chemistry and possibly the aroma and flavour perception
of wine (Figure 10) (Henick-Kling 1993, Bartowsky et al.
2002b, Matthews et al. 2004). Despite this concentration
Figure 10. A schematic representation of the biosynthesis and modulation of flavour-active compounds by malolactic bacteria.

Swiegers, Bartowsky, Henschke & Pretorius Microbial modulation of wine aroma and flavour 159
on Oenococcus oeni, the contribution by the other wine
genera, particularly Lactobacillus species, should not be
underestimated.
3.1 Acids
3.1.1. Non-volatile acids
Together with tartaric acid, malic acid is one of the major
organic acids in wine and is usually present in the concen-
tration range of 2–5 g/L, dependent upon geographical
location and climatic conditions. Malic acid metabolism
forms the basis of malolactic fermentation and can be car-
ried out by Oenococcus oeni and wine-associated species of
Lactobacillus and Pediococcus. The metabolism of organic
acids during malolactic fermentation can have a significant
impact on the flavour of the wine (Henick-Kling 1993,
Bartowsky et al. 2002a).
The metabolism of sugars and organic acids during
malolactic fermentation can be divided into three phases
(Krieger et al. 2000). During the growth phase (Phase I),
sugar catabolism occurs with little production of acetic
and lactic acid; minimal citric and malic acid are
metabolised in this phase. As the bacterial cell numbers
increase above 5 × 106cfu/mL during Phase II, the cata-
bolism of sugar ceases and malic acid metabolism pro-
ceeds accompanied by production of lactic acid; citric acid
remains untouched at this stage, and there is no acetic acid
produced during malic acid degradation. Phase III is char-
acterised by the metabolism of citric acid accompanied
by an increase in acetic acid. The increase of lactic acid
content in the wine results in a softer mouth-feel and
the acetic acid contributes to the volatile acidity of the
wine (Figure 10).
The decarboxylation of malic acid to lactic acid forms
the basis of malolactic fermentation. This reaction is cata-
lysed by the enzyme malate decarboxylase, often referred
to as the malolactic enzyme, with the requirement of
cofactors NAD+and Mn++ (Lonvaud-Funel and Strasser de
Saad 1982, Caspritz and Radler 1983, Spettoli et al. 1984,
Naori et al. 1990, Kunkee 1991) (Figure 10). All wine
lactic acid bacteria are able to perform the malolactic
reaction; however, Oenococcus oeni, the best adapted species
to highly acidic wine conditions, is the preferred species.
Tartaric acid is relatively stable to bacterial activity and
can only be metabolised aerobically by some Lactobacillus
species with the production of acetic acid, lactic acid and
succinic acid (Kandler 1983, Dittrich 1987). When tartar-
Figure 11. A schematic representation of citric acid metabolism and the synthesis of diacetyl in malolactic bacteria (based upon Ramos and
Santos 1996).

160 Microbial modulation of wine aroma and flavour Australian Journal of Grape and Wine Research 11, 139–173, 2005
ic acid degradation occurs, it tends to appear as part of a
general spoilage scenario.
Citric acid can be metabolised by numerous genera of
the lactic acid bacteria and results in the production of
acetic acid and diacetyl, both of which can have an impor-
tant effect on wine flavour (Figure 11) (Bartowsky and
Henschke 2004). Citric acid, a grape-derived organic acid,
is commonly present in wine in the range of 0.1–0.7 g/L
and most strains of Oenococcus oeni are able to metabolise
this acid during malolactic fermentation. The metabolism
of citric acid normally occurs after that of malic acid, as
does its depletion from wine. Higher peak concentrations
of diacetyl (a flavour-active metabolite of citric acid cata-
bolism) generally correlate with an elevated concentration
of citric acid; however, the magnitude of the relationship
depends on many other factors. Though the addition of
citric acid to wine can be accompanied by an increase of
diacetyl, the formation of other flavour metabolites,
particularly acetic acid, can also result (Henick-Kling and
Park 1994). Furthermore, since some yeasts can meta-
bolise citric acid, unexpected changes in the concentration
of diacetyl and titratable acidity could result. Therefore,
addition of citric acid to grape must or wine for the pur-
pose of elevating diacetyl accumulation should be
approached with caution.
In some countries, sorbic acid, a short-chained unsat-
urated fatty acid, is used as a chemical preservative in
sweetened wines at bottling to prevent fermentation
occurring in the bottle. It inhibits the proliferation of some
yeast (including Saccharomyces) and moulds, but it is inef-
fective against Dekkera/Brettanomyces, acetic acid bacteria
and lactic acid bacteria. In fact, sorbic acid can be meta-
bolised by lactic acid bacteria, including Oenococcus oeni.
Sorbic acid is reduced to sorbyl alcohol, which undergoes
a chemical rearrangement at wine pH to 2,4-hexadien-1-
ol and can, in turn, react with ethanol to give rise to the
ether 2-ethoxyhexa-3,5-diene (Crowell and Guymon
1975). The final compound has an odour reminiscent of
geranium leaves with a reported sensory threshold of 100
ng/L (Riesen 1992).
3.1.2. Volatile acids
Injudicious aeration during and/or after the winemaking
process can result in the growth and activity of acetic acid
bacteria, high volatile acidity and a vinegary taint in wine.
These bacteria are classified into the genera Acetobacter,
Acidomonas, Gluconobacter and Gluconacetobacter; of these,
Gluconobacter oxydans, Acetobacter aceti, Acetobacter pasteuri-
anus, Gluconacetobacter liquefaciens and Gluconacetobacter
hansenii are normally associated with grapes and wine
(Drysdale and Fleet 1988). The oxidation of ethanol to
acetic acid is the best-known characteristic of these
wine-associated acetic acid bacteria. In this reaction, a
membrane-bound alcohol dehydrogenase oxidises ethanol
to acetaldehyde, and is further oxidised to acetate by a
membrane-bound aldehyde dehydrogenase. The concen-
tration of oxygen required for metabolic activity and sur-
vival in wine is now much lower than previously thought;
acetic acid bacteria can survive in wine barrels for long
periods of low oxygen tension and, somewhat unexpect-
edly, spoilage of bottled red wine by acetic acid bacteria
has been reported (Drysdale and Fleet 1988, Bartowsky et
al. 2003).
A small increase in VA is often observed after the com-
pletion of malolactic fermentation conducted by malo-
lactic bacteria. Two pathways can be involved. Acetic acid
can be produced from residual sugar through heterolactic
metabolism (phosphoketolase pathway) (Henick-Kling
1993, Ribéreau-Gayon et al. 2000a) (Figure 10), and the
first step in citric acid metabolism produces acetic acid
(Cogan 1987, Ramos et al. 1995, Ramos and Santos 1996)
(Figure 11). This latter metabolism is discussed further in
Section 3.3.2.
3.2 Alcohols (polyols)
Bacteria can modulate the concentrations of alcohols such
as glycerol, mannitol and erythritol, and affect wine
flavour. Metabolism of glycerol is not widespread amongst
the wine lactic acid bacteria (Lactobacillus brevis, Lacto-
bacillus hilgardii and Pediococcus pentosaceus) and results in
wine spoilage (Sponholz 1993, Claisse and Lonvaud-Funel
2001, Vizoso Pinto et al. 2004). Glycerol can be degraded
by two pathways, either via glycerol dehydrase or glycerol
kinase. The glycerol dehydrase converts glycerol to 3-
hydroxypropionaldehyde. Spontaneous chemical dehy-
dration of the aldehyde, by heating or by long-term stor-
age in acidic solution, results in the formation of acrolein,
which reacts with wine phenolics, particularly in red
wines, to form a bitter complex (Figure 10).
Mannitol spoilage of wine (also referred to as mannite
disease) can be caused by some heterolactic bacteria.
Under certain conditions, these bacteria bring about slim-
iness and produce a vinegary-estery, slightly sweet taste.
A six-carbon sugar alcohol, or polyol, is the end product
of fructose reduction (Pilone et al. 1991, Veiga-da-Cunha
et al. 1993, von Weymarn et al. 2002, Wisselink et al.
2002). This reduction is catalysed by mannitol-1-
phosphatase in homofermentative lactic acid bacteria
(Lactobacillus groups I and II and Pediococcus species),
whereas in heterofermentative lactic acid bacteria
(Lactobacillus Group III, Leuconostoc species and Oenococcus)
the mannitol dehydrogenase catalyses the formation of
mannitol. In heterofermentative lactic acid bacteria, the
reduction of fructose is a mechanism by which the cell can
regenerate NAD+, particularly under anaerobic conditions.
The homofermentative lactic acid bacteria are also able to
utilise mannitol, transporting it into the cell via the man-
nitol-specific phosphotransferase system and phosphory-
lating it to mannitol 1-phosphate, which is then oxidised
by mannitol 1-phosphate dehydrogenase to the glycolysis-
intermediate fructose 6-phosphate (Figure 10). In general,
the homofermentative lactic acid bacteria will only
produce small amounts of mannitol, whereas some
heterofermentative lactic acid bacteria produce and export
substantial amounts of mannitol (Wisselink et al. 2002).
Mannitol can serve as a sole carbon source for the
homofermentative lactic acid bacteria Lactobacillus
plantarum (Davis et al. 1988, Liu et al. 1995, Wisselink et
al. 2002).
In Oenococcus oeni fructose can be metabolised by two

Swiegers, Bartowsky, Henschke & Pretorius Microbial modulation of wine aroma and flavour 161
different pathways: heterolactic fermentation or mixed
heterolactic/mannitol fermentation (Richter et al. 2003).
Heterolactic fermentation is carried out by bacteria in the
late-exponential growth phase, whereas mixed hetero-
lactic/mannitol fermentation is common during expo-
nential growth. The switch from one fermentation type to
the other apparently occurs at the metabolic level. The
shift is related to the growth rate, or to the high metabol-
ic rates present during high growth rates (Richter et al.
2003). The regeneration of NAD(P)H is the key, where in
glucose fermentation, the erythritol pathway takes over
this role and in fructose fermentation the mannitol path-
way is used. That is, Oenococcus oeni uses the heterolactic
fermentation pathway during growth on fructose as a
substrate and the mannitol pathway when using fructose
as an electron acceptor (Richter et al. 2003) (Figure 10).
3.3 Carbonyls
Acetaldehyde and diacetyl are two of the more important
flavour-active carbonyls in wine and both can be
metabolised by wine bacteria.
3.3.1 Acetaldehyde
Acetaldehyde, which mainly originates from yeast metab-
olism (Figure 10), is a highly volatile compound with an
apple-like and nutty aroma. It enhances the colour devel-
opment of red wine by promoting condensation reactions
between anthocyanins and catechins to tannins, forming
stable polymeric pigments resistant to sulfur dioxide
bleaching (Timberlake and Bridle 1976a, Somers and
Wescombe 1987). It is, therefore, inevitable that any bac-
terial activity that affects the concentration of acetalde-
hyde in wine potentially can affect its colour and flavour.
Some strains of Oenococcus oeni and Lactobacillus (but
not Pediococcus) can metabolise acetaldehyde to acetic acid
and ethanol (Osborne et al. 2000). The ability to
metabolise acetaldehyde bound to sulfur dioxide can
inhibit the growth of bacteria by releasing sulfur dioxide,
which accumulates to form an inhibitory concentration
(Hood 1983) (Figure 10). The chemical and sensory
impact of the ethanol and acetic acid formed by the
metabolism of acetaldehyde by lactic acid bacteria is
believed to be limited, but the reduction in the acetalde-
hyde pool in wine is believed to influence final wine
colour. It has also been suggested that the degradation of
acetaldehyde-bound sulfur dioxide by sulfur dioxide-sen-
sitive malolactic bacteria could lead to an incomplete or
prolonged malolactic fermentation.
3.3.2 Diacetyl
2,3-Butanedione, commonly referred to as ‘diacetyl’, is a
major flavour compound in dairy products. Extensive
research has been devoted to this topic, some of which is
applicable to the lactic acid bacteria associated with wine-
making (Figure 10). When present at a high concentration
(exceeding 5–7 mg/L) in wine, diacetyl is regarded by
many to be undesirable (Rankine et al. 1969, Davis et al.
1985). At around 1–4 mg/L, however, depending on the
style and type of wine, it is considered to contribute a
desirable ‘buttery’ or ‘butterscotch’ flavour. The sensory
perception of diacetyl in wine is also highly dependent
upon the presence of other compounds in the wine, and
is influenced by the age, the style and origin of the wine
(Rankine et al. 1969, Martineau et al. 1995, Bartowsky et
al. 2003).
Yeast and bacteria contribute to the diacetyl content of
wine (Figure 10), though the concentration of diacetyl
produced by yeast during alcoholic fermentation is usually
below its detection threshold (Martineau and Henick-
Kling 1995). In contrast, bacteria can produce significant
amounts of diacetyl during malolactic fermentation, and
diacetyl is one of the most important flavour compounds
produced by Oenococcus oeni.
The formation and degradation of diacetyl is directly
related to the growth of malolactic bacteria and the metab-
olism of sugar, malic acid and citric acid. It is formed as an
intermediate metabolite in the reductive decarboxylation
of pyruvic acid to 2,3-butanediol (Figure 11). Pyruvic acid
is derived essentially from the metabolism of sugar and
citric acid, and the formation of 2,3-butanediol might
contribute to the redox balance of cellular metabolism.
Theoretically, 1 mol of citrate produces 1 mol of acetic
acid, 2 mol of carbon dioxide and 0.5 mol of a mixture
consisting of diacetyl, acetoin and 2,3-butanediol (Cogan
1987, Ramos et al. 1995, Ramos and Santos 1996).
A variety of factors, including some that the wine-
maker can control, affect the concentration of diacetyl in
wine, including oxygen exposure, fermentation tempera-
ture, sulfur dioxide levels and duration of malolactic fer-
mentation (reviewed by Bartowsky and Henschke 2004).
The conversion of α-acetolactate to diacetyl is a non-
enzymatic decarboxylation, enhanced by the presence of
oxygen (Figure 10). Although malolactic fermentation is
essentially an anaerobic process, it is not greatly affected
by limited exposure to air. However, the amount of
diacetyl can vary from 2 mg/L under anaerobic conditions
to 12 mg/L under semi-aerobic conditions (Nielsen and
Richelieu 1999).
Although the optimum temperature in laboratory
media for the growth of malolactic bacteria is about 27ºC,
growth in wine is restricted to 15–25ºC with an optimum
for most Oenococcus oeni cultures of about 20–22ºC (Kelly
et al. 1989). Malolactic fermentation conducted at lower
temperatures, perhaps 18ºC rather than 25ºC, tends to be
slower, but wines accumulate a higher concentration of
diacetyl (Lonvaud-Funel et al. 1984, Hart 1997).
The properties of sulfur dioxide, such as antioxidant
and antimicrobial, are important to the winemaking
process. Sulfur dioxide can interact with carbonyl
compounds, including acetaldehyde and diacetyl in a
reversible manner. In the presence of sulfur dioxide, the
concentration of free diacetyl in wine is lowered, however
as the sulfur dioxide content decreases, as for example
during ageing, the ratio of free diacetyl will increase again,
thus increasing its sensory impact (Nielsen and Richelieu
1999).
The duration of malolactic fermentation is dependent
upon numerous factors including bacterial strain, chemi-
cal composition of the wine and wine temperature. Wines
that undergo a prolonged malolactic fermentation, for

162 Microbial modulation of wine aroma and flavour Australian Journal of Grape and Wine Research 11, 139–173, 2005
whatever reason, tend to have a higher diacetyl content
(McCarthy 2000, Bartowsky et al. 2002a).
Thus, increasing the ‘buttery’ diacetyl impact of a wine
can be achieved by using a lower than usual inoculum of
a high diacetyl producing strain in the absence of active
yeast, such as after racking wine off yeast lees. The diacetyl
content should then be stabilised by filtering to remove
bacteria (and yeast if present) and prevent re-metabo-
lism, and by adding sufficient sulfur dioxide to prevent
further microbial activity. A low diacetyl content can be
achieved by using an appropriate strain inoculated during
the late stage of alcoholic fermentation, and if necessary
maintaining the wine on stirred lees until the diacetyl
becomes undetectable (Bartowsky and Henschke 2004).
3.4 Esters
The majority of wine esters are produced by yeast during
alcoholic fermentation (Figure 7). Esters can, however,
also be derived from the grape, from the chemical esteri-
fication of alcohols and from acids during wine ageing
(Rapp and Mandery 1986, Etiévant 1991, Younis and
Stewart 1998, Lambrechts and Pretorius 2000). Esterase
activity of wine-associated bacterial species is not well
understood; esterases of dairy-associated species of Lacto-
bacillus and Pediococcus have been observed, and it appears
that their growth in grape juice or wine might modify the
ester profile of wine (Matthews et al. 2004).
In a survey of wine lactic acid bacteria, over two-thirds
of Oenococcus oeni strains, Lactobacillus species and Pedio-
coccus parvulus strains examined demonstrated esterase
activity by the hydrolysis of an ester substrate (Davis et al.
1988). Researchers observed increases in ethyl ester con-
centration in wine following malolactic fermentation,
including ethyl acetate, ethyl hexanoate, ethyl lactate,
and ethyl octanoate, as well as decreases in some esters
(Zeeman et al. 1982, Dittrich 1987, Laurent et al. 1994, de
Revel et al. 1999, Delaquis et al. 2000, Gambaro et al.
2001). These variances in ester concentrations during
grape vinification suggest that esterases are involved in
both the synthesis and hydrolysis of esters. Changes in
ester concentration following malolactic fermentation
may either enhance or degrade the wine quality, depend-
ing on the ester metabolised.
3.5 Sulfur-containing compounds
Amino acids are the most important source of nitrogen,
carbon and sulfur for sulfur-containing amino acids
among the wine substrates metabolised by lactic acid bac-
teria. Except for biogenic amine formation and the catab-
olism of arginine, the metabolism of amino acids by
Oenococcus oeni has not been studied extensively. Though
amino acids play an important role in cheese flavour,
their role in wine flavour has not been examined to
date. Methionine metabolism has been shown recently
by Oenococcus oeni strains and wine-associated Lactobacillus
species producing methanethiol, dimethyl sulfide, 3-
(methylsulfanyl)propan-1-ol and 3-(methylsulfanyl)-
propanoic acid (Pripis-Nicolau et al. 2004). In trials with
malolactic fermentation induction by four commercial
Oenococcus oeni cultures in Merlot wine, elevated con-
centrations of 3-(methylsulfanyl)propanoic acid were
observed, suggesting that Oenococcus oeni can metabolise
methionine to form volatile sulfur compounds.
Cysteine can be the precursor of S-containing hetero-
cycles, such as thiazoles. The sulfur-containing amino acid
cysteine and the tri-peptide glutathione stimulated the
growth of Oenococcus oeni whereas methionine did not
(Rauhut et al. 2004). However, none of the three com-
pounds influenced the metabolism of malic acid to lactic
acid.
Sulfur-containing amino acids have a high chemical
reactivity with carbonyl compounds, in particular with
sugars, according to the Maillard reaction. This well-
known mechanism occurs at high temperatures and is
favoured by dry conditions. However, carbonyl groups
have a greater reactivity with electrophiles other than
those of sugars, in aqueous media and at ambient tem-
perature.
Cysteine is a particularly interesting amino acid
because of its involvement in the varietal flavours of
Sauvignon Blanc wines (Tominaga et al. 1998b). A reac-
tion in wine can occur between α-dicarbonyl compounds
(including diacetyl) and amino acids, in particular cysteine
(Pripis-Nicolau et al. 2004). Various aromas can arise,
including sulfury notes, floral, fruity, toasted and roasted
notes, depending upon the amino acid involved in the
reaction. It has been found that the decomposition rate of
cysteine in the presence of diacetyl at pH 3.5 and 25ºC is
approximately 70% over seven days whereas at pH 8.0,
cysteine is completely decomposed within an hour-and-a-
half.
These findings highlight the importance of possible
reactions arising from carbonyl compounds in the deriva-
tion of aromatic products even under unfavourable con-
ditions of low pH and temperatures, such as those encoun-
tered during storage and ageing of wine. Of particular
interest are reactions involving cysteine and diacetyl that
could occur after alcoholic or malolactic fermentations.
These reactions have been demonstrated and newly
formed compounds identified under wine storage and
ageing conditions (low pH and low temperature)
(Marchand et al. 2000). Odours developed in solutions
with an α-dicarbonyl compound (e.g. diacetyl) were more
intense than those developed with a hydroxy ketone (e.g.
acetoin). Many of the compounds produced in this way
have been identified in wine, and because of their low
olfactory thresholds could play an important role in wine
aroma and flavour.
3.6 Glycosidic conjugates
The complex aroma and flavour compounds found in
wine largely originate from the grape, from yeast metab-
olism during alcoholic fermentation and from oak when
used. Bacterial metabolism during malolactic fermentation
might contribute to wine flavour by the formation of
additional compounds and the modification of grape-,
yeast- and oak-derived compounds.
Grape-derived glycoconjugates constitute a latent pool
of volatile aglycones that can be another source of wine
aroma and flavour compounds (Williams et al. 1989).

Swiegers, Bartowsky, Henschke & Pretorius Microbial modulation of wine aroma and flavour 163
These aglycones are grouped broadly by structure; noriso-
prenoids (e.g. damascenone), volatile phenols and other
benzene derivatives (e.g. raspberry ketone), monoter-
penes (e.g. linalool, nerol and geraniol) and aliphatics
(e.g. hexanol). In Chardonnay, a non-floral variety, almost
200 different aglycones have been identified in grape juice
prior to fermentation (Sefton et al. 1993).
The flavourless glycoconjugates are glucosides or dis-
accharide or trisaccharide glycosides. These all contain a
glucosyl moiety, but for the disaccharide glycosides, the
glucose is further substituted with α-L-arabinofuranosyl,
α-L-rhamnopyranosyl, β-D-xylopyranosyl or β-apiofura-
nosyl sugars. In the grape, the disaccharide glycosides are
the dominant storage form of aroma substances (Williams
et al. 1982b). Many of the wine volatile compounds can
be released from their flavourless glycoconjugate precur-
sors by either acid or enzymatic hydrolysis.
The glycosidases involved with the enzymatic cleavage
of the disaccharide glycosides include α-L-arabinofura-
nosidase, α-L-rhamnopyranosidase, β-D-xylopyranosi-
dase, β-apiofuranosidase and β-D-glucopyranosidase (also
referred to as α-D-glucosidase). The liberation of the
volatile aglycone from the disaccharide by yeast involves
the sequential release of the sugar moieties, where the first
step is the hydrolysis of the inter-sugar link and the sec-
ond step is a β-glucosidase activity for the cleavage of the
remaining β-glucosidic moiety (Günata et al. 1988). This
has also been shown to be the case for Oenococcus oeni
(D’Incecco et al. 2004).
Wine bacteria, in particular Oenococcus oeni, are able to
cleave the glucose moiety from the major red wine antho-
cyanin, malvidin-3-glucoside, and use it as a carbon
source (Vivas et al. 1997). Oenococcus oeni strains possess
various glycosidase activities; however, these activities on
synthetic glycosides were dependent on wine conditions
such as pH, ethanol and residual sugar content (Grimaldi
et al. 2000). Another study using arbutin was unable to
demonstrate Oenococcus oeni glycosidic activity (McMahon
et al. 1999). More recent studies using Tannat wine or an
isolated Chardonnay wine glycosidic extract in synthetic
wine medium demonstrated that there was some limited
release of glycosylated wine volatiles by Oenococcus oeni
during malolactic fermentation (Boido et al. 2002,
D’Incecco et al. 2004).
The ability of Oenococcus oeni to liberate aroma com-
pounds bound to sugar moieties might depend on the
grape variety and conditions of malolactic fermentation.
No release could be demonstrated, for example, in a
Viognier wine (Mansfield et al. 2002), but release could be
shown in a highly aromatic Muscat variety (Ugliano et al.
2003). The degree of release of glycosidically bound aroma
compounds appears to be very much strain-dependent
(Grimaldi et al. 2000, Boido et al. 2002, Ugliano et al.
2003, D’Incecco et al. 2004). Recently, it was found that
species of Lactobacillus and Pediococcus show varying
degrees of β- and α-D-glucopyranosidase activity, which in
turn is influenced by exposure to ethanol and/or sugars,
temperature and pH (Grimaldi et al. 2005).
3.7 Phenols
Phenolic compounds are abundant in wine, originating
from the grape (skin, seeds and stalks) and from the wood
used for ageing and maturation of the wine. The major
groups of phenolic compounds found naturally in white
and red grapes are: phenolic acids (hydroxycinnamic and
hydroxybenzoic acids and their conjugates), flavanols
Figure 12. A schematic representation of the formation of potent and unpleasant nitrogen-heterocycle ‘mousy’ off-flavour compounds
(2-acetyltetrahydropyridine (ACTPY) and 2-acetyl-1-pyrroline (ACPY)) by some lactic bacteria (based upon Costello and Henschke 2002).

164 Microbial modulation of wine aroma and flavour Australian Journal of Grape and Wine Research 11, 139–173, 2005
(catechins), proanthocyanidins (grape tannins or con-
densed tannins) and flavonols (quercetin). Red grapes
also contain the pigmented polyphenolics and antho-
cyanins. During fermentation of red must many of the
grape phenolics compounds are extracted and some are
modified, so that red wine contains, in addition to the
grape phenolics, pyroanthocyanins (e.g. vitisins), poly-
meric pigments (conjugates of anthocyanins and proan-
thocyanidins) and wine tannins (modified proantho-
cyanidins). Hydrolysable tannins will also be present when
wood has been used in the fermentation process.
Of the phenolic compounds present in wine, the phe-
nolic acids are most susceptible to metabolism by many
wine lactic acid bacteria and acetic acid bacteria. Phenolic
acids can be transported into bacterial cells by active trans-
port, decarboxylated to the vinyl derivatives by hydroxy-
cinnamic acid decarboxylases and enzymatically reduced
to the ethyl derivatives (Cavin et al. 1993, 1994) (Figure
10). Identity of the transport systems for the vinyl and
ethyl derivatives are not clear. Laboratory studies have
demonstrated the ability of various wine lactic acid bac-
teria strains to produce both vinyl and ethyl phenols from
p-coumaric and ferulic acids (Cavin et al. 1993, 1994).
However, in white wine, which only contains a low con-
centration of flavanols, only the vinylphenol was pro-
duced as the major product, and at much diminished con-
centration compared to that produced by the reference
Brettanomyces bruxellensis strain (Chatonnet et al. 1997).
Furthermore, the only strain of Oenococcus oeni tested pro-
duced much lower concentrations of volatile phenols
compared to Pediococcus damnosus and Lactobacillus plan-
tarum. The concentration of (seed) procyanidic tannins
was shown to inhibit the formation of volatile phenols by
a Lactobacillus plantarum strain, but not by the Brettanomyces
bruxellensis strain in wine. Thus, together with
Saccharomyces yeast, some lactic acid bacteria appear to
have a low capacity to contribute to the accumulation of
vinyl phenols but probably not ethyl phenols during wine
production.
Studies in a synthetic wine-like medium have shown
that hydroxycinnamic acids inhibit the growth of Oeno-
coccus oeni more than do hydroxybenzoic acids, whereas
Lactobacillus hilgardii is less affected by these phenolic com-
pounds, with the exception of p-coumaric acid (Campos et
al. 2003). Even though numerous Lactobacillus species
have genes for the hydroxycinnamic acid (p-coumaric
acid) decarboxylase, Lactobacillus hilgardii is unable to
decarboxylate p-coumaric acid and/or ferulic acid to 4-
vinylphenol and/or 4-vinylguaiacol, respectively (van
Beek and Priest 2000). Supporting the inhibitory effect of
p-coumaric acid on Lactobacillus hilgardii growth and
survival, as well as for Oenococcus oeni, is the inability of
these bacteria to decarboxylate, and presumably detoxify,
p-coumaric acid.
On the other hand, catechin and gallic acid, which
were metabolised by Lactobacillus hilgardii strain 5w, stim-
ulated growth (Alberto et al. 2001, 2004). The production
of pyrogallol from gallic acid and catechin metabolism is
known to serve as an oxygen scavenger and to reduce the
redox potential of media (Vivas and Glories 1995, Alberto
et al. 2001). This effect might promote growth in the
absence of oxygen. The presence of phenol carboxylic
acids (caffeic, ferulic, p-coumaric and gallic acids) and
catechin appear to stimulate the growth of Oenococcus oeni,
particularly by reducing the initial lag phase and enhanc-
ing the metabolism of citric acid (Vivas et al. 1997, Rozes
et al. 2003). Though these studies indicate that Oenococcus
oeni is stimulated by the presence of ferulic and p-coumar-
ic acids, it remains unclear whether these acids can be
metabolised by wine strains of this species to produce
sensorially important concentrations of 4-vinylguaiacol
and 4-vinylphenol.
3.8 Amino acids and peptides
Oenococcus oeni is a fastidious organism that requires sev-
eral amino acids and short peptides for growth. In addition
to amino acid transport systems Oenococcus oeni has
transport systems for peptides, albeit uncharacterised.
Proteolytic activity has been observed in selected
Oenococcus oeni isolates (Manca de Nadra et al. 1997,
Manca de Nadra et al. 1999, Farias and Manca de Nadra
2000). Though the uptake of short peptides by other
species of the lactic acid bacteria has been well charac-
terised, both physiologically and genetically, that is not the
case with Oenococcus oeni (Kok and de Vos 1994,
Christensen et al. 1999, Peltoniemi et al. 2002). Peptides
are an important source of amino acids, but they can also
contribute to bitterness (Habibi-Najafi and Lee 1996,
Desportes et al. 2001).
The metabolism in wine of certain amino acids,
notably lysine and ornithine, can lead to the formation
of several extremely potent and unpleasant nitrogen-
heterocycle ‘mousy’ off-flavour compounds (Heresztyn
1986, Costello et al. 2001) (Figure 12). The compounds
are perceived on the back palate as a persistent after-
taste. The heterofermentative lactic acid bacteria, Oeno-
coccus oeni, Leuconostoc mesenteroides and some Lactobacillus
species were capable of synthesising the three known
sensorially important nitrogen-heterocycle compounds:
2-acetyltetrahydropyridine (ACTPY), 2-acetyl-1-pyrroline
(ACPY) and 2-ethyltetrahydropyridine (ETPY) (Costello et
al. 2001). The heterofermentative lactobacilli favoured
the formation of ACTPY, Oenococcus oeni the least flavour
active ETPY, whereas the homofermentative pediococci
favoured the highest flavour active compound, ACPY.
Oenococcus oeni strains varied in their ability to form the
nitrogen-heterocycles. The homofermentative species,
Pediococcus and Lactobacillus plantarum produced little or no
detectable off-flavour compounds.
In general, the heterofermentative lactic acid bacteria
showed the highest ability to produce nitrogen-hetero-
cycles and mousy-off flavour. Working with a Lactobacillus
hilgardii strain it was shown that the presence of ethanol,
a fermentable carbohydrate (D-fructose) and Fe2+, in addi-
tion to L-lysine and L-ornithine, were necessary for the
formation of these nitrogen-heterocycle compounds
(Costello and Henschke 2002). ACTPY was formed in the
absence of the two amino acids but was greatly stimulat-
ed by the presence of L-lysine. L-ornithine was required
for ACPY formation. Ethanol and acetaldehyde are

Swiegers, Bartowsky, Henschke & Pretorius Microbial modulation of wine aroma and flavour 165
involved in nitrogen-heterocycle compounds formation,
possibly by increasing the C-2 acylating pool as side chain
precursor (Figure 12).
4. Concluding remarks
One of the grandest research and development aspira-
tions of the Australian wine industry is to develop objec-
tive measures for grape and wine quality. Despite good
progress, that goal remains elusive (Francis et al. 2005).
Not only is the composition of wine challenging to unrav-
el, but consumer and market preferences will dictate the
type of measurement that will be important. Nevertheless,
it is a common fact that wine flavour is one of the key
drivers of consumer choice. Therefore, we need to inte-
grate our research in viticulture and oenology with a
‘grape-to-glass-to-consumer’ focus.
The research outlined in this article demonstrates that
the strains of wine yeast and malolactic bacteria with
which grape must and wine are inoculated can have an
important impact on the aroma and flavour profile of the
final product. As our knowledge develops, it will become
possible to select specific yeast and bacteria to produce
wines of any chosen style and meet the changing demands
of consumers. The choice of yeast and bacterial strain to
modulate wine flavour according to specifications of a
target market will differ with (i) the type and style of
wine to be made; (ii) the grape variety and viticultural
practices; as well as (iii) winemaking techniques and tech-
nical requirements of the winery (reviewed in Pretorius
2000, 2003, 2004, de Barros Lopes et al. 2005, Pretorius et
al. 2005, Swiegers and Pretorius 2005).
Future research at The Australian Wine Research
Institute and collaborating research and industry partners
will focus stronger on the following key issues: (i) devel-
opment of methodologies to identify sensory characteris-
tics and consumer preferences for aroma and flavour in
wine; (ii) development of methodologies to predict wine
choice behaviour in key potential export markets; (iii)
determination and quantification of grape precursors for
key flavour and aroma compounds; (iv) identification of
key viticultural factors that influence the levels of grape
flavour precursors; and (v) determination of winemaking
factors that influence transformation of grape precursors
into potent flavour-active compounds and their stability in
wine. With this research endeavour we hope to deliver
improved consumer preference information, microbial
strains, viticultural management and winemaking tech-
nologies to the Australian wine industry.
Acknowledgements
We thank the numerous colleagues who have been
involved in the AWRI work outlined, particularly, Dimi
Capone, Miguel de Barros Lopes, Gordon Elsey, Mark
Sefton, Tracey Siebert and Robyn Willmot. We are grate-
ful to Rae Blair for her assistance in preparing and editing
this manuscript. The yeast and bacteria research at the
AWRI is financially supported by Australia’s grapegrowers
and winemakers through their investment body the Grape
and Wine Research and Development Corporation, with
matching funding from the Australian Government.
References
Acree, T.E., Sonoff, E.P. and Spittstoesser, D.F. (1972) Effect of yeast
strain and type of sulfur compound on hydrogen sulfide production.
American Journal of Enology and Viticulture 23, 6–9.
Alberto, M.R., Farias, M.E. and Manca de Nadra, M. (2001) Effect of
gallic acid and catechin on Lactobacillus hilgardii 5w growth and
metabolism of organic compounds. Journal of Agricultural Food
Chemistry 49, 4359–4363.
Alberto, M.R., Gomez-Cordoves, C. and Manca de Nadra, M.C. (2004)
Metabolism of gallic acid and catechin by Lactobacillus hilgardii from
wine. Journal of Agricultural and Food Chemistry 52, 6465–6469.
Altschul, S.F., Gish, W., Miller, W., Myers, E.W. and Lipman, D.J.
(1990) Basic local alignment search tool. Journal of Molecular
Biology 215, 403–410.
Amerine, M.A., Berg, H.V., Kunkee, R.E., Ough, C.S., Singleton, V.L.
and Webb, A.D. (1980) The Technology of Winemaking (AVI
Technical Books Inc. 4th ed.: Westport, CT).
Amerine, M.A. and Ough, C.S. (1980) Alcohols. In: Methods for
Wine and Must Analysis. Eds M.A. Amerine and C.S. Ough (John
Wiley and Sons: New York) pp. 74–127.
Anderson, K. (2004) The World’s Wine Markets. Globalization at
Work (Edward Elgar Publishing Limited: Cheltenham, Glos, UK) p.
335.
Ansanay, V., Dequin, S., Camarasa, C., Schaeffer, V., Grivet, J.-P.,
Blondin, B., Salmon, J.-M. and Barre, P. (1996) Malolactic fermen-
tation by engineered Saccharomyces cerevisiae as compared with engi-
neered Schizosaccharomyces pombe. Yeast 12, 215–225.
Aryan, A.P., Wilson, B., Strauss, C.R. and Williams, P.J. (1987) The
properties of glycosidases of Vitis vinifera and a comparison of their
β-glucosidase activity with that of exogenous enzymes – an assess-
ment of possible applications in enology. American Journal of
Enology and Viticulture 38, 182–188.
Bach, B., Camarasa, C., Rossignol, T., Blondin, B. and Dequin, S.
(2004) γ-Aminobutyrate assimilation in Saccharomyces cerevisiae dur-
ing wine fermentation. Proceedings of Physiology of Yeasts and
Filamentous Fungi (PYFF2), Anglet, France. p. 185 (abstract).
Bakker, J. and Timberlake, C.F. (1997) Isolation, identification and
characterization of new color-stable anthocyanins occurring in some
red wines. Journal of Agricultural and Food Chemistry 45, 35–43.
Barthelmebs, L., Diviès, C. and Cavin, J. F. (2000a) Knockout of the
p-coumarate decarboxylase gene from Lactobacillus plantarum reveals
the existence of two other inducible enzymatic activities involved in
phenolic acid metabolism. Applied and Environmental Microbiology
66, 3368–3375.
Barthelmebs, L., Lecomte, B., Diviès, C. and Cavin, J.F. (2000b)
Inducible metabolism of phenolic acids in Pediococcus pentosaceus is
encoded by an autoregulated operon which involves a new class of
negative transcriptional regulator. Journal of Bacteriology 182,
6724–6731.
Bartowsky, E.J. and Henschke, P.A. (2004) The ‘buttery’ attribute of
wine – diacetyl – desirability, spoilage and beyond. International
Journal of Food Microbiology 96, 235
–
252.
Bartowsky, E., Costello, P. and Henschke, P. (2002a) Management
of malolactic fermentation – wine flavour manipulation. The
Australian Grapegrower and Winemaker 461a, 7–12.
Bartowsky, E.J., Francis, I.L., Bellon, J.R. and Henschke, P.A. (2002b)
Is buttery aroma perception in wines predictable from diacetyl con-
centration? Australian Journal of Grape and Wine Research 8,
180–185.
Bartowsky, E.J., Xia, D., Gibson, R.L., Fleet, G.H. and Henschke, P.A.
(2003) Spoilage of bottled red wine by acetic acid bacteria. Letters in
Applied Microbiology 36, 307–314.
Bartowsky, E.J., Dillon, S.J., Henschke, P.A., Markides, A.J., Dumont,
A., Pretorius, I.S., Ortiz-Julien, A. and Herderich, M. (2004) The
potential of Saccharomyces cerevisiae wine yeast to improve red wine
colour. Australian and New Zealand Grapegrower and Winemaker
490, 83–85.
Baumes, R., Bayonove, C., Barillere, J.M., Escudier, J.L. and
Cordonnier, R. (1988) La macération pelliculaire dans la vinification
en blanc – Incidences sur la composante volatiles des moûts.
Connaissance de la Vigne et du Vin 22, 209–223.

166 Microbial modulation of wine aroma and flavour Australian Journal of Grape and Wine Research 11, 139–173, 2005
Bayonove, C.L., Günata, Y.Z. and Cordonnier, R.E. (1984) Mise in evi-
dence de l’intervention des enzymes dans le développement de
l’arôme du jus de Muscat avant fermentation: La production des ter-
pénols. Bulletin de l’OIV 643, 741–758.
Bely, M., Rinaldi, A. and Dubourdieu, D. (2003) Influence of assim-
ilable nitrogen on volatile acidity production by Saccharomyces cere-
visiae during high sugar fermentation. Journal of Bioscience and
Bioengineering 96, 507–512.
Benabdeljalil, C., Cheynier, V., Fulcrand, H., Hakiki, A., Mosdaddak,
M. and Moutounet, M. (2000) Evidence of new pigments resulting
from reactions between anthocyanins and yeast metabolites. Science
des Aliments 20, 203–220.
Bennetzen, J.L. and Hall, B.D. (1982) The primary structure of the
Saccharomyces cerevisiae gene for alcohol dehydrogenase – I. Journal
of Biological Chemistry 257, 3018–3025.
Berg, H., Filipello, F., Hinreiner, E. and Webb, A. (1955) Evaluation of
thresholds and minimum difference concentrations for various con-
stituents of wines: I. Water solutions for pure substances. Food
Technology 9, 23–26.
Berry, D.R. and Watson, D.C. (1987) Production of organoleptic com-
pounds. In: Yeast Biotechnology. Eds D.R. Berry, I. Russell and G.G.
Stewart (Allen and Unwin: London) pp. 345–368.
Bisson, L.F., Waterhouse, A.L., Ebeler, S.E., Walker, M.A. and Lapsley,
J.T. (2002) The present and future of the international wine indus-
try. Nature 418, 696–699.
Blanchard, L., Tominaga, T. and Dubourdieu, D. (2001) Formation of
furfurylthiol exhibiting a strong coffee aroma during oak barrel fer-
mentation from furfural released by toasted staves. Journal of
Agricultural and Food Chemistry 49, 4833–4835.
Bobet, R.A., Noble, A.C. and Boulton, R.B. (1990) Kinetics of the
ethanethiol and diethyl disulfide interconversion in wine-like
solutions. Journal of Agricultural and Food Chemistry 38, 449–452.
Boido, E., Lloret, A., Medina, K., Carrau, F. and Dellacasa, E. (2002)
Effect of β-glycosidase activity of Oenococcus oeni on the glycosylated
flavor precursors of Tannat wine during malolactic fermentation.
Journal of Agricultural Food Chemistry 50, 2344–2349.
Bony, M., Bidart, F., Camarasa, C., Ansanay, V., Dulau, L., Barre, P. and
Dequin, S. (1997) Metabolic analysis of S. cerevisiae strains engi-
neered for malolactic fermentation. FEBS Letters 410, 452–456.
Boulton, R.B., Singleton, V.L., Bisson, L.F. and Kunkee, R.E. (1998)
Principles and Practices of Winemaking (Aspen Publishers, Inc:
USA).
Camarasa, C., Grivet, J. and Dequin, S. (2003) Investigation by 13C-
NMR and tricarboxylic acid (TCA) deletion mutant analysis of
pathways for succinate formation in Saccharomyces cerevisiae during
anaerobic fermentation. Microbiology 149, 2669–2678.
Campos, F.M., Couto, J.A. and Hogg, T.A. (2003) Influence of phenolic
acids on growth and inactivation of Oenococcus oeni and Lactobacillus
hilgardii. Journal of Applied Microbiology 94, 167–174.
Carrau, F.M., Medina, K., Boido, E., Farina, L., Gaggero, C., Dellacassa,
E., Versini, G. and Henschke, P.A. (2005) De novo synthesis of
monoterpenes by Saccharomyces cerevisiae wine yeasts. FEMS Micro-
biology Letters 243, 107–115.
Casalone, E., Colella, C.M., Daly, S., Gallori, E., Moriani, L. and
Polsinelli, M. (1992) Mechanism of resistance to sulfite in Saccharo-
myces cerevisiae. Current Genetics 22, 435–440.
Caspritz, G. and Radler, F. (1983) Malolactic enzyme of Lactobacillus
plantarum. Purification, properties, and distribution among bacteria.
Journal of Biological Chemistry 258, 4907–4910.
Cavin, J.F., Andioc, V., Etievant, P.X. and Diviès, C. (1993) Ability of
wine lactic-acid bacteria to metabolize phenol carboxylic-acids.
American Journal of Enology and Viticulture 44, 76–80.
Cavin, J.F., Guzzo, J., Gasporotto, L., Logier, E. and Diviès, C. (1994)
Etude de la décarboxylation de l’acide para-coumarique chez
Lactobacillus plantarum. Proceedings of the SFM Congress: Gestion
des Populations Microbiennes dans les Industries Agro-Alimentaires,
Dijon, France. p. 36.
Cavin, J.F., Barthelmebs, L. and Diviès, C. (1997) Molecular charac-
terization of an inducible p-coumaric acid decarboxylase from
Lactobacillus plantarum: gene cloning, transcriptional analysis, over-
expression in Escherichia coli, purification, and characterization.
Applied and Environmental Microbiology 63, 1939–1944.
Cavin, J.F., Dartois, V. and Diviès, C. (1998) Gene cloning, tran-
scriptional analysis, purification, and characterization of phenolic
acid decarboxylase from Bacillus subtilis. Applied and Environmental
Microbiology 64, 1466–1471.
Chatonnet, P., Dubourdieu, D., Boidron, J.N. and Lavigne, V. (1993)
Synthesis of volatile phenols by Saccharomyces cerevisiae in wines.
Journal of the Science of Food and Agriculture 62, 191–202.
Chatonnet, P., Dubourdieu, D. and Boidron, J.N. (1995) The influence
of Brettanomyces/Dekkera sp yeasts and lactic acid bacteria on the
ethylphenol content of red wines. American Journal of Enology and
Viticulture 46, 463–468.
Chatonnet, P., Viala, C. and Dubourdieu, D. (1997) Influence of
polyphenolic components of red wines on the microbial synthesis of
volatile phenols. American Journal of Enology and Viticulture 48,
443–448.
Christensen, J.E., Dudley, E.G., Pederson, J.A. and Steele, J.L. (1999)
Peptidases and amino acid catabolism in lactic acid bacteria. Antonie
van Leeuwenhoek 76, 217–246.
Claisse, O. and Lonvaud-Funel, A. (2001) Detection of lactic acid
bacteria producing 3-hydropropionaldehyde (acrolein precursor)
from glycerol by molecular tests. Lait 81, 173–181.
Clausen, M., Lamb, C.J., Megnet, R. and Doerner, P.W. (1994) PAD1
encodes phenylacrylic acid decarboxylase which confers resistance
to cinnamic acid in Saccharomyces cerevisiae. Gene 142, 107–112.
Clemente-Jimenez, J.M., Mingorance-Cazorla, S., Martinez-
Rodriguez, S., Las Heras-Vazquez, F.J. and Rodriguez-Vico, F. (2005)
Influence of sequential yeast mixtures on wine fermentation.
International Journal of Food Microbiology 98, 301–308.
Cogan, T.M. (1987) Co-metabolism of citrate and glucose by
Leuconostoc spp.: Effects on growth, substrates and products. Journal
of Applied Bacteriology 63, 551–558.
Cordero Otero, R.R.C., Iranzo, J.F.U., Briones-Perez, A.I., Potgieter, N.,
Villena, M.A., Pretorius, I.S. and van Rensburg, P. (2003)
Characterization of the β-glucosidase activity produced by enolog-
ical strains of non-Saccharomyces yeasts. Journal of Food Science 68,
2564–2569.
Corison, C.A., Ough, C.S., Berg, H.W. and Nelson, K.E. (1979) Must
acetic acid and ethyl acetate as mold and rot indicators in grapes.
American Journal of Enology and Viticulture 30, 130–134.
Costello, P., Lee, T.H. and Henschke, P.A. (2001) Ability of lactic acid
bacteria to produce N-heterocycles causing mousy off-flavour in
wine. Australian Journal of Grape and Wine Research 7, 160–167.
Costello, P.J. and Henschke, P.A. (2002) Mousy off-flavour of wine:
Precursors and biosynthesis of the causative N-heterocycles 2-ethyl-
tetrahydropyridine, 2-acetyltetrahydropyridine, and 2-acetyl-
1-pyrroline by Lactobacillus hilgardii DSM 20176. Journal of
Agricultural Food Chemistry 50, 7079–7087.
Coulter, A.D., Godden, P.W. and Pretorius, I.S. (2004) Succinic acid –
how it is formed, what is its effect on titratable acidity, and what fac-
tors influence its concentration in wine? Australian and New
Zealand Wine Industry Journal 19, 16–20, 21–25.
Crowell, E.A. and Guymon, J.F. (1975) Wine constituents arising
from sorbic acid addition and identification of 2-ethoxyhexa-3,5-
diene as source of geranium-like off-odor. American Journal of
Enology and Viticulture 26, 97–102.
Darriet, P., Tominaga, T., Lavigne, V., Boidron, J. and Dubourdieu, D.
(1995) Identification of a powerful aromatic compound of Vitis
vinifera L. var. Sauvignon wines: 4-Mercapto-4-methylpentan-2-
one. Flavour and Fragrance Journal 10, 385–392.
Davis, C.R., Wibowo, D., Eschenbruch, R., Lee, T.H. and Fleet, G.H.
(1985) Practical implications of malolactic fermentation: A review.
American Journal of Enology and Viticulture 36, 290–301.
Davis, C.R., Wibowo, D., Fleet, G.H. and Lee, T.H. (1988) Properties
of wine lactic acid bacteria: their potential enological significance.
American Journal of Enology and Viticulture 39, 137–142.
Day, R.E., Gishen, M., Godden, P.W. and Hughes, J.B. (2002)
Producing unique wine styles for overseas markets. Proceedings of
the Eleventh Australian Wine Industry Technical Conference
(Australian Wine Industry Technical Conference Inc.: Adelaide, SA)
pp. 27–32.

Swiegers, Bartowsky, Henschke & Pretorius Microbial modulation of wine aroma and flavour 167
de Barros Lopes, M.A., Rehman, A-.U., Gockowiak, H., Heinrich,
A.J., Langridge, P. and Henschke, P.A. (2000) Fermentation prop-
erties of a wine yeast overexpressing the Saccharomyces cerevisiae
glycerol 3-phosphate dehydrogenase gene (GPD2). Australian
Journal of Grape and Wine Research 6, 208–215.
de Barros Lopes, M.A., Eglinton, J., Henschke, P.A., Høj, P.B. and
Pretorius, I.S. (2003) The connection between yeast and alcohol
reduction in wine: managing the double-edged sword of bottled sun-
shine. Australian and New Zealand Wine Industry Journal 18,
17–18, 20, 22.
de Barros Lopes, M.A., Bartowsky, E.J. and Pretorius, I.S. (2005) The
application of gene technology in the wine industry. In: Handbook
of Food Science, Technology and Engineering; Volume 1 – Properties
and Products. Ed. H.Y. Hui (CRC Press: Florida/Marcel Dekker: New
York, USA) pp. 40.1–40.21.
Delaquis, P., Cliff, M., King, M., Girard, B., Hall, J. and Reynolds, A.
(2000) Effect of two commercial malolactic cultures on the chemi-
cal and sensory properties of Chancellor wines vinified with differ-
ent yeasts and fermentation temperatures. American Journal of
Enology and Viticulture 51, 42–48.
Delcroix, A., Günata, Z., Sapis, J.C., Salmon, J.M. and Bayonove, C.
(1994) Glycosidase activities of 3 enological yeast strains during
winemaking – Effect on the terpenol content of Muscat wine.
American Journal of Enology and Viticulture 45, 291–296.
Delfini, C. and Costa, A. (1993) Effects of grape must lees and insol-
uble materials on the alcoholic fermentation rate and the production
of acetic acid, pyruvic acid and acetaldehyde. American Journal of
Enology and Viticulture 44, 102–107.
Denis, C.L., Ferguson, J. and Young, E.T. (1983) mRNA levels for the
fermentative alcohol dehydrogenase of Saccharomyces cerevisiae
decrease upon growth on a nonfermentable carbon source. Journal
of Biological Chemistry 258, 1165–1171.
Dequin, S. and Barre, P. (1994) Mixed lactic acid-alcoholic fermenta-
tion by Saccharomyces cerevisiae expressing the Lactobacillus casei L(+)-
LDH. Bio/Technology 12, 173–177.
Dequin, S., Baptista, E. and Barre, P. (1999) Acidification of grape
musts by Saccharomyces cerevisiae wine yeast strains genetically engi-
neered to produce lactic acid. American Journal of Enology and
Viticulture 50, 45–50.
Derrick, S. and Large, P.J. (1993) Activities of the enzymes of the
Ehrlich pathway and formation of branched-chain alcohols in
Saccharomyces cerevisiae and Candida utilis grown in continuous-
culture on valine or ammonium as sole nitrogen-source. Journal of
General Microbiology 139, 2783–2792.
de Revel, G., Martin, N., Pripis-Nicolau, L., Lonvaud-Funel, A. and
Bertrand, A. (1999) Contribution to the knowledge of malolactic fer-
mentation influence on wine aroma. Journal of Agricultural Food
Chemistry 47, 4003–4008.
Desportes, C., Charpentier, M., Duteutre, B., Maujean, A. and
Duchiron, F. (2001) Isolation, identification, and organoleptic char-
acterization of low-molecular-weight peptides from white wine.
American Journal of Enology and Viticulture 52, 376–380.
Dickinson, J.R. and Norte, V. (1993) A study of branched-chain amino
acid aminotransferase and isolation of mutations affecting the cata-
bolism of branched-chain amino acids in Saccharomyces cerevisiae.
FEBS Letters 326, 29–32.
Didion, T., Grauslund, M., Kielland Brandt, M.C. and Andersen, H.A.
(1996) Amino acids induce expression of BAP2, a branched-chain
amino acid permease gene in Saccharomyces cerevisiae. Journal of
Bacteriology 178, 2025–2029.
D’Incecco, N., Bartowsky, E.J., Kassara, S., Lante, A., Spettoli, P. and
Henschke, P.A. (2004) Release of glycosidically bound flavour com-
pounds of Chardonnay by Oenococcus oeni during malolactic fer-
mentation. Food Microbiology 21, 257–265.
Dittrich, H.H. (1987) Mikrobiologie des Weines. Handbuch der
Lebensmitteltechnologie.2nd edn. (Ulmer: Stuttgart).
Donalies, U.E.B. and Stahl, U. (2002) Increasing sulfite formation in
Saccharomyces cerevisiae by overexpression of MET14 and SSU1. Yeast
19, 475–484.
Drawert, F. and Barton, H. (1978) Biosynthesis of flavor compounds
by micro-organisms. 3. Production of monoterpenes by the yeast
Kluyveromyces lactis. Journal of Agricultural and Food Chemistry 26,
765–767.
Drysdale, G.S. and Fleet, G.H. (1988) Acetic acid bacteria in wine-
making – A review. American Journal of Viticulture and Enology 39,
143–154.
Dubois, P. (1983) Volatile phenols in wine. In: Flavour of Distilled
Beverages, Origins and Developments. Ed. J.R. Pifgott (Ellis
Horwood: Chichester, UK) pp. 110–119.
Dubois, P. (1994) Les arômes des vins et leurs défauts. Revue Française
d’Oenologie 145, 27–40.
Dubourdieu, D. (2000) The role of yeasts in grape flavour develop-
ment during fermentation: The example of Sauvignon Blanc.
Proceedings of the ASEV 50th Anniversary Annual Meeting, Seattle,
USA (American Society of Enology and Viticulture: Davis, CA) pp.
196–203.
Dubourdieu, D., Tominaga, T., Masneuf, I., Peyrot de Gachons, C. and
Murat, M.L. (2000) The role of yeasts in grape flavor development
during fermentation: The example of Sauvignon Blanc. Proceedings
American Society of Enology and Viticulture 50th Annual Meeting,
Seattle, USA (American Society of Enology and Viticulture) p. 37.
du Toit, M and Pretorius, I.S. (2000) Microbial spoilage and preser-
vation of wine: Using weapons from Nature’s own arsenal – A
review. South African Journal for Enology and Viticulture 21, 74–96.
Eden, A., Simchen, G. and Benvenisty, N. (1996) Two yeast homologs
of ECA39, a target for c-Myc regulation, code for cytosolic and mito-
chondrial branched-chain amino acid aminotransferases. Journal of
Biological Chemistry 271, 20242–20245.
Eden, A., Van Nedervelde, L., Drukker, M., Benvenisty, N. and
Debourg, A. (2001) Involvement of branched-chain amino acid
aminotransferases in the production of fusel alcohols during fer-
mentation in yeast. Applied Microbiology and Biotechnology 55,
296–300.
Eglinton, J.M., McWilliam, S.J., Fogarty, M.W., Francis, I.L.,
Kwiatkowski, M.J., Høj, P.B. and Henschke, P.A. (2000) The effect
of Saccharomyces bayanus-mediated fermentation on the chemical
composition and aroma profile of Chardonnay wine. Australian
Journal of Grape and Wine Research 6, 190–196.
Eglinton, J.M., Heinrich, A.J., Pollnitz, A.P., Langridge, P., Henschke,
P.A. and de Barros Lopes, M. (2002) Decreasing acetic acid accu-
mulation by a glycerol overproducing strain of Saccharomyces cerevisiae
by deleting the ALD6 aldehyde dehydrogenase gene. Yeast 19,
295–301.
Eglinton, J., Henschke, P., Høj, P. and Pretorius, I. (2003) Winemaking
properties and potential of Saccharomyces bayanus wine yeast – har-
nessing the untapped potential of yeast biodiversity. Australian and
New Zealand. Wine Industry Journal 18, 16–19.
Eglinton, J., Griesser, M., Henschke, P.A., Kwiatkowski, M.J., Parker,
M. and Herderich, M. (2004) Yeast-mediated formation of pig-
mented polymers in red wine. Proceedings Red Wine Color:
Revealing the Mysteries (ACS Symposium Series 886: Oxford
University Press) pp. 7–21.
Ehrlich, F. (1904) Über das naturliche isomere des leucins. Berichte
Deutsche Chemische Gesellschaft 37, 1809–1840.
Erasmus, D.J., Cliff, M. and van Vuuren, H.J.J. (2004) Impact of
yeast strain on the production of acetic acid, glycerol, and the
sensory attributes of icewine. American Journal of Enology and
Viticulture 55, 371–378.
Eschenbruch, R. (1974) Sulfite and sulfide formation during wine
making. A review. American Journal of Enology and Viticulture 25,
157–161.
Eschenbruch, R., Bonish, P. and Fisher, B.M. (1978) The production
of H2S by pure culture wine yeasts. Vitis 17, 67–74.
Etiévant, P.X. (1991) Wine. In: Volatile Compounds of Food and
Beverages. Ed. H. Maarse (Marcel Dekker Inc.: New York, USA) pp.
483–546.
Fagan, G.L., Kepner, R.E. and Webb, A.D. (1981) Production of
linalool, cis-nerolidol and trans-nerolidol, and trans,trans-farnesol
by Saccharomyces fermentati growing as a film on simulated wine. Vitis
20, 36–42.
Farias, M.E. and Manca de Nadra, M.C. (2000) Purification and
partial characterization of Oenococcus oeni exoprotease. FEMS
Microbiology Letters 185, 263–266.
Fernández, M., Ubeda Iranzo, J.F. and Briones Pérez, A.I. (2000)
Typing of non-Saccharomyces yeasts with enzymatic activities of inter-

168 Microbial modulation of wine aroma and flavour Australian Journal of Grape and Wine Research 11, 139–173, 2005
est in winemaking. International Journal of Food Microbiology 59,
29–36.
Ferreira, V., Lopez, R. and Cacho, J.F. (2000) Quantitative determi-
nation of the odorants of young red wines from different grape
varieties. Journal of the Science of Food and Agriculture 80,
1659–1667.
Fleet, G.H. and Heard, G.M. (1993) Yeasts: Growth during fermenta-
tion. In: Wine Microbiology and Biotechnology. Ed. G.H. Fleet
(Harwood Academic Publishers: Chur, Switzerland) pp. 27–54.
Flikweert, M.T., Van der Zanden, L., Janssen, W.M.T.M., Steensma,
H.Y., Van Dijken, J.P. and Pronk, J.T. (1996) Pyruvate decarboxylase:
An indispensable enzyme for growth of Saccharomyces cerevisiae on
glucose. Yeast 12, 247–257.
Forrestal, P. (2000) The Global Encyclopedia of Wine (Harper Collins
Publishers: Sydney, Australia).
Fowles, G.W.A. (1992) Acids in grapes and wines: A review. Journal
of Wine Research 3, 25–41.
Francis, I.L., Høj, P.B., Dambergs, R.G., Gishen, M., de Barros Lopes,
M.A., Pretorius, I.S., Godden, P.W., Henschke, P.A, Herderich, M.J.
and Waters, E.J. (2005) Objective measures of grape quality – are
they achieveable? Proceedings of the Twelfth Australian Wine
Industry Technical Conference, Melbourne, Australia (Australian
Wine Industry Technical Conference Inc.: Adelaide, SA) pp. 85–89.
Fujii, T., Nagasawa, N., Iwamatsu, A., Bogaki, T., Tamai, Y. and
Hamachi, M. (1994) Molecular cloning, sequence analysis, and
expression of the yeast alcohol acetyltransferase gene. Applied and
Environmental Microbiology 60, 2786–2792.
Gambaro, A., Boido, E., Zlotejablko, A., Medina, K., Lloret, A.,
Dellacassa, E. and Carrau, F. (2001) Effect of malolactic fermentation
on the aroma properties of Tannat wine. Australian Journal of
Grape and Wine Research 7, 27–32.
Gancedo, C., Gancedo, J.M. and Sols, A. (1968) Glycerol metabolism
in yeast. Pathways of utilization and production. European Journal
of Biochemistry 5, 165–172.
Garcia, A., Carcel, C., Dulau, L., Samson, A., Aguera, E., Agosin, E.
and Günata, Z. (2002) Influence of a mixed culture with Debaryo-
myces vanriji and Saccharomyces cerevisiae on the volatiles of a Muscat
wine. Journal of Food Science 67, 1138–1143.
Giudici, P., Romano, P. and Zambonelli, C. (1990) A biometric study
of higher alcohol production in Saccharomyces cerevisiae. Canadian
Journal of Microbiology 36, 61–64.
Giudici, P., Zambonelli, C., Passarelli, P. and Castellari, L. (1995)
Improvement of wine composition with cryotolerant Saccharomyces
strains. American Journal of Enology and Viticulture 46, 143–147.
Giudici, P. and Kunkee, R.E. (1994) The effect of nitrogen deficiency
and sulfur-containing amino acids on the reduction of sulfate to
hydrogen sulfide by wine yeasts. American Journal of Enology and
Viticulture 45, 107–112.
Goniak, O.J. and Noble, A.C. (1987) Sensory study of selected volatile
sulfur compounds in white wine. American Journal of Enology
and Viticulture 38, 223–227.
Grimaldi, A., McLean, H. and Jiranek, V. (2000) Identification and par-
tial characterization of glycosidic activities of commercial strains of
the lactic acid bacterium, Oenococcus oeni. American Journal of
Enology and Viticulture 51, 362–369.
Grimaldi, A., Bartowsky, E.J. and Jiranek, V. (2005) Screening of
Lactobacillus spp. and Pediococcus spp. for glycosidase activities that are
important in oenology. Journal of Applied Microbiology (in press).
Günata, Y.Z., Bayonove, C.L., Baumes, R.L. and Cordonnier, R.E.
(1985) The aroma of grapes 1. Extraction and determination of free
and glycosidically bound fractions of some grape aroma compo-
nents. Journal of Chromatography 331, 83–90.
Günata, Y.Z., Bayonove, C.L., Baumes, R.L. and Cordonnier, R.E.
(1986) Stability of free and bound fractions of some aroma compo-
nents of grapes cv. muscat during the wine processing – Preliminary
results. American Journal of Enology and Viticulture 37, 112–114.
Günata, Y.Z., Bitteur, S., Brillouet, J.-M., Bayonove, C. and Cordon-
nier, R. (1988) Sequential enzymatic hydrolysis of potentially aro-
matic glycosides from grape. Carbohydrate Research 184, 139–149.
Günata, Y.Z., Bitteur, S., Baumes, R., Sapis, J.C. and Bayonove, C.
(1990) Activites glycosidases en vinification. Perspectives d’ex-
ploitation des precurseurs d’arôme du raisin, de nature glycosidique.
Revue Française d’Oenologie 122, 37–41.
Guth, H. (1997) Quantitation and sensory studies of character impact
odorants of different white wine varieties. Journal of Agricultural
and Food Chemistry 45, 3027–3032.
Guth, H. (1998) Comparison of different white wine varieties by
instrumental analysis and sensory studies. In: Chemistry of Wine
Flavour. Eds L.A Waterhouse and S.E. Ebeler (ACS Symposium
Series 714: San Francisco, USA) pp. 39–52.
Guth, H. and Sies, A. (2002) Flavour of wines: Towards an under-
standing by reconstitution experiments and an analysis of ethanol’s
effect on odour activity of key compounds. Proceedings of the
Eleventh Australian Wine Industry Technical Conference, Adelaide,
Australia. (Australian Wine Industry Technical Conference Inc.:
Adelaide, SA) pp. 128–139.
Habibi-Najafi, M.B. and Lee, B.H. (1996) Bitterness in cheese – A
review. Critical Reviews in Food Science and Nutrition 36, 397–411.
Hallinan, C.P., Saul, D.J. and Jiranek, V. (1999) Differential utilization
of sulfur compounds for H2S liberation by nitrogen-starved wine
yeast. Australian Journal of Grape and Wine Research 5, 82–90.
Hansen, J., Cherest, H. and Kielland-Brandt, M.C. (1994) Two diver-
gent MET10 genes, one from Saccharomyces cerevisiae and one from
Saccharomyces carlsbergensis, encode the alpha subunit of sulfite reduc-
tase and specify potential binding sites for FAD and NADPH. Journal
of Bacteriology 176, 6050–6058.
Hart, A. (1997) The influence of malolactic fermentation on the con-
centration of diacetyl by individual strains of Oenococcus oeni. Honours
thesis (The University of Adelaide, Adelaide, Australia).
Hayasaka, Y. and Asenstorfer, R.E. (2002) Screening for potential
pigments derived from anthocyanins in red wine using nano-
electrospray tandem mass spectrometry. Journal of Agricultural and
Food Chemistry 50, 756–761.
Heerde, E. and Radler, F. (1978) Metabolism of the anaerobic forma-
tion of succinic acid by Saccharomyces cerevisiae. Archiv für Mikro-
biologie 117, 269–276.
Henick-Kling, T. (1993) Malolactic fermentation. In: Wine
Microbiology and Biotechnology. Ed. G.H. Fleet. (Harwood:
Camberwell, Vic.) pp. 289–326
Henick-Kling, T. and Park, Y.H. (1994) Considerations for the use of
yeast and bacterial starter cultures: SO2and timing of inoculation.
American Journal of Enology and Viticulture 45, 464–469.
Henschke, P.A. (1993) An overview of malolactic fermentation
research. Australian and New Zealand Wine Industry Journal 8,
69–79.
Henschke, P.A. (1996) Hydrogen sulfide production by yeast during
fermentation. Proceedings Eleventh International Oenological
Symposium, Sopron, Hungary (International Association for Winery
Technology and Management: Breisach, Germany) pp. 83–102.
Henschke, P.A. (1997) Wine yeast. In: Yeast Sugar Metabolism:
Biochemistry, Genetics, Biotechnology, and Applications. Eds F.K.
Zimmermann and K.-D. Entian (Technomic: Lancaster, PA) pp.
527–560.
Henschke, P.A. and Dixon, G.D. (1990) Effect of yeast strain on acetic
acid accumulation during fermentation of Botrytis affected grape
juice. Proceedings of the Seventh Australian Wine Industry Technical
Conference, Adelaide, Australia (Australian Industrial Publishers:
Adelaide SA) pp. 253–254.
Henschke, P.A and Jiranek, V. (1991) Hydrogen sulfide formation
during fermentation: Effect of nitrogen composition in model grape
must. Proceedings of the International Symposium on Nitrogen in
Grapes and Wine, Seattle, USA (American Society for Enology and
Viticulture: Davis, CA) pp. 172–184.
Henschke, P.A. and Jiranek, V. (1993) Yeast – Growth during fer-
mentation. In: Wine Microbiology and Biotechnology. Ed. G.H.
Fleet (Harwood Academic Publishers: Chur, Switzerland) pp. 27–54.
Henschke, P.A. and de Kluis, F.M. (1995) Origin and control of hydro-
gen sulfide produced by yeast during fermentation. Proceedings of
the XXI World Congress of grapes and wines, Punte del Este,
Uruguay (Office International de la Vigne et du Vin: Paris, France)
pp. 165–178.
Heresztyn, T. (1986) Formation of substituted tetrahydropyridines
by species of Brettanomyces and Lactobacillus isolated from mousy
wines. American Journal of Viticulture and Enology 37, 127–132.

Swiegers, Bartowsky, Henschke & Pretorius Microbial modulation of wine aroma and flavour 169
Hernández, L.F., Espinosa, J.C., Fernandez-Gonzalez, M. and Briones,
A. (2003) β-Glucosidase activity in a Saccharomyces cerevisiae wine
strain. International Journal of Food Microbiology 80, 171–176.
Holgate, A. (1997) The influence of yeast strain on the acid balance
of red wine. Proceedings Advances in Juice Clarification and Yeast
Inoculation, Melbourne, Australia (Australian Society of Viticulture
and Oenology: Adelaide, SA) pp. 39–41.
Hood, A. (1983) Inhibition of growth of wine lactic-acid bacteria by
acetaldehyde-bound sulphur dioxide. Australian Grapegrower and
Winemaker 232, 34–43.
Houtman, A.C., Marais, J. and Du Plessis, C.S. (1980a) The possibil-
ities of applying present-day knowledge of wine aroma compo-
nents: Influence of several juice factors on fermentation rate and
ester production during fermentation. South African Journal of
Enology and Viticulture 1, 27–33.
Houtman, A.C., Marais, J. and Duplessis, C.S. (1980b) Factors affect-
ing the reproducibility of fermentation of grape juice and of the
aroma composition of wines 1. Grape maturity, sugar, inoculum con-
centration, aeration, juice turbidity and ergosterol. Vitis 19, 37–54.
Howell, K.S., Swiegers, J.H., Elsey, G.M., Siebert, T.E., Bartowsky, E.J.,
Fleet, G.H., Pretorius, I.S. and de Barros Lopes, M.A. (2004)
Variation in 4-mercapto-4-methyl-pentan-2-one release by Saccharo-
myces cerevisiae commercial wine strains. FEMS Microbiology Letters
240, 125–129.
Howell, K.S., Klein, M., Swiegers, J.H., Hayasaka, Y., Elsey, G.M.,
Fleet, G.H., Høj, P.B., Pretorius, I.S. and de Barros Lopes, M.A.
(2005) Genetic determinants of volatile thiol release by Saccharomyces
cerevisiae during wine fermentation. Applied and Environmental
Microbiology (in press).
Hwang, A. (1992) Restriction mapping of the POF1 gene. Journal of
the Institute of Brewing 98, 467–470.
Iglesias, R., Ferreras, J.M., Arias, F.J., Munoz, R. and Girbes, T. (1991)
Effect of continued exposition to ethanol on activity of the ammo-
nium and fructose transport systems in Saccharomyces cerevisiae var.
ellipsoideus. Biotechnology and Bioengineering, 37, 389–391.
Jackson, R.S. (1994) Wine Science. Principles and Applications
(Academic Press: London, UK).
Jiranek, V., Langridge, P. and Henschke, P.A. (1995a) Amino acid
and ammonium utilization by Saccharomyces cerevisiae wine yeasts
from a chemically defined medium. American Journal of Enology
and Viticulture 46, 75–83.
Jiranek, V., Langridge, P. and Henschke, P.A. (1995b) Validation of bis-
muth-containing indicator media for predicting H2S producing
potential of Saccharomyces cerevisiae wine yeasts under enological
conditions. American Journal of Enology and Viticulture 46,
269–273.
Jiranek, V., Langridge, P. and Henschke, P.A. (1996) Determination of
sulphite reductase activity and its response to assimilable nitrogen
status in a commercial Saccharomyces cerevisiae wine yeast. Journal of
Applied Bacteriology 81, 329–336.
Kandler, O. (1983) Carbohydrate metabolism in lactic acid bacteria.
Antonie van Leeuwenhoek 49, 209–224.
Kelly, W.J., Asmundson, R.V. and Hopcroft, D.H. (1989) Growth of
Leuconostoc oenos under anaerobic conditions. American Journal of
Enology and Viticulture 40, 277–282.
King, A. and Dickinson, J.R. (2000) Biotransformation of monoter-
pene alcohols by Saccharomyces cerevisiae, Torulaspora delbrueckii and
Kluyveromyces lactis. Yeast 16, 499–506.
Kispal, G., Steiner, H., Court, D.A., Rolinski, B. and Lill, R. (1996)
Mitochondrial and cytosolic branched-chain amino acid transami-
nases from yeast, homologs of the myc oncogene-regulated Eca39
protein. Journal of Biological Chemistry 271, 24458–24464.
Klingenberg, A. and Sprecher, E. (1985) Production of monoterpenes
in liquid cultures by the yeast Ambrosiozyma monospora. Planta
Medica 3, 264–265.
Kobayashi, K. and Yoshimoto, A. (1982) Studies on yeast sulfite
reductase. Biochimica and Biophysica Acta 709, 46–52.
Kok, J. and de Vos, W.M. (1994) The proteolytic system of lactic acid
bacteria. In: Genetics and Biotechnology of Lactic Acid Bacteria.
Eds M.J. Gasson and W.M. de Vos (Blackie: Glasgow) pp. 169–210.
Kotseridis, Y., Baumes, R. and Skouroumounis, G.K. (1998) Synthesis
of labelled [H-2(4)] β-damascenone, [H-2(2)]2-methoxy-3-
isobutylpyrazine, [H-2(3)] a-ionone, and [H-2(3)] β-ionone, for
quantification in grapes, juices and wines. Journal of Chromato-
graphy A 824, 71–78.
Kunkee, R.E. (1991) Some roles of malic acid in the malolactic fer-
mentation in wine making. FEMS Microbiology Reviews 88, 55–72.
Krieger, S.A., Lemperle, E. and Ernst, M. (2000) Management of
malolactic fermentation with regard to flavour modification in wine.
Paper presented at the 5th International Symposium on Cool
Climate Viticulture and Oenology, Melbourne, Australia.
Lambrechts, M.G. and Pretorius, I.S. (2000) Yeast and its importance
to wine aroma. South African Journal of Enology and Viticulture 21,
97–129.
Larsen, T.O. and Frisvad, J.C. (1994) A simple method for collection
of volatile metabolites from fungi based on diffusive sampling from
petri dishes. Journal of Microbiological Methods 19, 297–305.
Larsen, T.O. and Frisvad, J.C. (1995) Characterization of volatile
metabolites from 47 Penicillium-taxa. Mycological Research 99,
1153–1166.
Laurent, M.-H., Henick-Kling, T. and Acree, T.E. (1994) Changes in he
aroma and odor of Chardonnay wine due to malolactic fermenta-
tion. Viticultural and Enological Science 49, 3–10.
Laurent, M.H., Lavin, E.H., Henick-Kling, T. and Acree, T.E. (1991)
Changes in the odor-active compounds found in grape wine due to
malolactic fermentation. Abstracts of Papers American Chemical
Society 202, 137.
Laurent, M.-H., Henick-Kling, T. and Acree, T.E. (1994) Changes in
the aroma and odor of Chardonnay wine due to malolactic fer-
mentation. Wein-Wissenschaft 49, 3–10.
Lilly, M., Lambrechts, M.G. and Pretorius, I.S. (2000) Effect of
increased yeast alcohol acetyltransferase activity on flavor profiles of
wine and distillates. Applied and Environmental Microbiology 66,
744–753.
Lilly, M. (2004) The development of yeasts for the optimal production
of flavour-active esters and higher alcohols in wine and distillates.
PhD Thesis (Stellenbosch University: South Africa).
Liu, S.-Q., Davis, C.R. and Brooks, J.D. (1995) Growth and metabo-
lism of selected lactic acid bacteria in synthetic medium. American
Journal of Enology and Viticulture 46, 166–174.
Lonvaud-Funel, A. and Strasser de Saad, A.M. (1982) Purification and
properties of a malolactic enzyme from a strain of Leuconostoc
mesenteroides isolated from grapes. Applied and Environmental
Microbiology 43, 357–361.
Lonvaud-Funel, A., Zmirou, C. and Larue, F. (1984) Le métabolisme
de l’acide citrique par les bactéries lactiques de la fermentation mal-
olactiques des vins. Sciences des Aliments 4, 81–85.
Maarse, H. and Vissher, C. (1994) Volatile Compounds in Food,
Qualitative and Quantitative Data. 6th edn (Central Institute for
Nutrition and Food Research, TNO: Zeist, The Netherlands).
Malherbe, D.F., du Toit, M., Cordero Otero, R.R., van Rensburg, P.,
Pretorius, I.S. (2003) Expression of the Aspergillus niger glucose oxi-
dase gene in Saccharomyces cerevisiae and its potential applications in
wine production. Applied Microbiological Biotechnology 61,
502–511.
Manca de Nadra, A.C., Farías, M.E., Moreno-Arribas, V., Pueyo, E. and
Polo, M.C. (1999) A proteolytic effect of Oenococcus oeni on the
nitrogenous macromolecular fraction of red wine. FEMS Micro-
biology Letters 174, 41–47.
Manca de Nadra, M.C., Farías, M.E., Moreno-Arribas, M.V., Pueyo, E.
and Polo, M.C. (1997) Proteolytic activity of Leuconostoc oenos. Effect
on proteins and polypeptides from white wine. FEMS Microbiology
Letters 150, 135–139.
Mansfield, A.K., Zoecklein, B. and Whiton, R.S. (2002) Quantification
of glycosidase activity in selected strains of Brettanomyces bruxellensis
and Oenococcus oeni. American Journal of Enology and Viticulture 53,
303–307.
Marais, J. (2001) Effect of grape temperature and yeast strain on
Sauvignon Blanc wine aroma composition and quality. South
African Journal of Enology and Viticulture 22, 47–51.
Marais, J., Versini, G., van Wyk, C.J., Rapp, A. (1992) Effect of region
on free and bound monoterpene and C13-norisoprenoid concen-
trations in Weisser Riesling wine. South African Journal of Enology
and Viticulture 13, 71–77.

170 Microbial modulation of wine aroma and flavour Australian Journal of Grape and Wine Research 11, 139–173, 2005
Marchand, S., de Revel, G. and Bertrand, A. (2000) Approaches to
wine aroma: Release of aroma compounds from reactions between
cysteine and carbonyl compounds in wine. Journal of Agricultural
Food Chemistry 48, 4890–4895.
Martineau, B., Acree, T.E. and Henick-Kling, T. (1995) Effect of wine
type on the detection threshold for diacetyl. Food Research
International 28, 139–143.
Martineau, B. and Henick-Kling, T. (1995) Formation and degradation
of diacetyl in wine during alcoholic fermentation with Saccharomyces
cerevisiae strain EC 1118 and malolactic fermentation with Leuconostoc
oenos strain MCW. American Journal of Enology and Viticulture
46, 442–448.
Mason, A.B. and Dufour, J.P. (2000) Alcohol acetyltransferases and
the significance of ester synthesis in yeast. Yeast 16, 1287–1298.
Matthews, A., Grimaldi, A., Walker, M., Bartowsky, E., Grbin, P. and
Jiranek, V. (2004) Lactic acid bacteria as a potential source of
enzymes for use in vinification. Applied and Environmental
Microbiology 70, 5715–5731.
McCarthy, H. (2000) Evaluations of the ability of the chemical sensor
to determine the aroma profiles of wines which have undergone
malolactic fermentation. Honours thesis (The University of Adelaide:
Adelaide, Australia).
McMahon, H., Zoecklein, B.W., Fugelsang, K. and Jasinski, Y. (1999)
Quantification of glycosidase activities in selected yeasts and lactic
acid bacteria. Journal of Industrial Microbiology and Biotechnology
23, 198–203.
Meaden, P.G. and Taylor, N.R. (1991) Cloning of a yeast gene which
causes phenolic off flavours in beer. Journal of the Institute of
Brewing 97, 353–357.
Mendes-Ferreira, A., Mendes-Faia, A. and Leao, C. (2002) Survey of
hydrogen sulphide production by wine yeasts. Journal of Food
Protection 65, 1033–1037.
Mestres, M., Busto, O. and Guasch, J. (2000) Analysis of organic sul-
fur compounds in wine aroma. Journal of Chromatography A 881,
569–581.
Michnick, S., Roustan, J.-L., Remize, F., Barre, P. and Dequin, S.
(1997) Modulation of glycerol and ethanol yields during alcoholic
fermentation in Saccharomyces cerevisiae strains over-expressed or
disrupted for GPD1 encoding glycerol-3-phosphate dehydrogenase.
Yeast 13, 783–793.
Millan, M.C., Moreno, J., Medina, M. and Ortega, J.M. (1991)
Influence of the physiological state of the inoculum on fermentation
of musts from Pedro Ximenez grapes by Saccharomyces cerevisiae.
Microbios 65, 87–95.
Moio, L. and Etievant, P.X. (1995) Ethyl anthranilate, ethyl cinnamate,
2,3-dihydrocinnamate, and methyl anthranilate – 4 important odor-
ants identified in Pinot Noir wines of Burgundy. American Journal
of Enology and Viticulture 46, 392–398.
Monk, P.R. (1982) Effect of nitrogen and vitamin supplements on
yeast growth and rate of fermentation of Rhine Riesling grape juice.
Food Technology in Australia 34, 328–332.
Monk, P.R. (1986) Formation, utilization and excretion of hydrogen
sulphide by wine yeast. Australian Wine Industry Journal,1, 10–16.
Monk, P.R. and Cowley, P.J. (1984) Effect of nicotinic acid and sugar
concentration of grape juice and temperature on accumulation of
acetic acid during yeast fermentation. Journal of Fermentation
Technology 62, 515–521.
Moreira, N., Mendes, F., Pereira, O., Guedes de Pinho, P., Hogg, T. and
Vasconcelos, I. (2002) Volatile sulphur compounds in wines related
to yeast metabolism and nitrogen composition of grape musts.
Analytica Chimica Acta 458, 157–167.
Murat, M.L., Masneuf, I., Darriet, P., Lavigne, V., Tominaga, T. and D.
Dubourdieu (2001a) Effect of Saccharomyces cerevisiae yeast strains on
the liberation of volatile thiols in Sauvignon blanc wine. American
Journal of Enology and Viticulture 52, 136–139.
Murat, M.L., Tominaga, T. and Dubourdieu, D. (2001b) Assessing
the aromatic potential of Cabernet Sauvignon and Merlot musts
used to produce rose wine by assaying the cysteinylated precursor
of 3-mercaptohexan-1-ol. Journal of Agricultural and Food
Chemistry 49, 5412–5417.
Nagasawa, N., Bogaki, T., Iwamatsu, A., Hamachi, M. and Kumagai,
C. (1998) Cloning and nucleotide sequence of the alcohol acetyl-
transferase II gene (ATF2) from Saccharomyces cerevisiae Kyokai No. 7.
Bioscience Biotechnology and Biochemistry 62, 1852–1857.
Naori, P., Chagnaud, P., Arnaud, A. and Galzy, P. (1990) Purification
and properties of a malolactic enzyme from Leuconostoc oenos ATCC
23278. Journal of Basic Microbiology 8, 577–585.
Navarro-Avino, J.P., Prasad, R., Miralles, V.J., Benito, R.M. and
Serrano, R. (1999) A proposal for nomenclature of aldehyde dehy-
drogenases in Saccharomyces cerevisiae and characterization of the
stress-inducible ALD2 and ALD3 genes. Yeast 15, 829–842.
Nielsen, J.C. and Richelieu, M. (1999) Control of flavor develop-
ment in wine during and after malolactic fermentation by Oenococcus
oeni. Applied and Environmental Microbiology 65, 740–745.
Nieuwoudt, H.H. (2004) Glycerol and wine. PhD Thesis, University of
Stellenbosch, p. 115.
Nieuwoudt, H.H., Prior, B.A., Pretorius, I.S. and Bauer, F.F. (2002)
Glycerol in South African table wines: An assessment of its rela-
tionship to wine quality. South African Journal of Enology and
Viticulture 23, 22–30.
Noble, A.C. and Bursick, G.F. (1984) The contribution of glycerol to
perceived viscosity and sweetness in white wine. American Journal
of Enology and Viticulture 35, 110–112.
Nykänen, L.L., Nykänen, I. and Suomalainen, H. (1977) Distribution
of esters produced during sugar fermentation between the yeast cell
and the medium. Journal of the Institute of Brewing 83, 32–34.
Ortega, C., López, R., Cacho, J., and Ferreira, V. (2001) Fast analysis
of important wine volatile compounds: Development and validation
of a new method based on gas chromatographic flame ionisation
detection analysis of dichloromethane microextracts. Journal of
Chromatography A 923, 205–214.
Osborne, J.P., Mira de Orduna, R., Pilone, G.J. and Liu, S.-Q. (2000)
Acetaldehyde metabolism by wine lactic acid bacteria. FEMS
Microbiology Letters 191, 51–55.
Ough, C., Fong, D. and Amerine, M. (1972) Glycerol in wine: deter-
mination and some factors affecting. American Journal of Enology
and Viticulture 23, 1–5.
Park, S.K. and Noble, A.C. (1993) Monoterpenes and monoterpene
glycosides in wine aromas. ACS Symposium Series 536, 98–109.
Park, S.K. Boulton, R.B. and Noble, A.C. (2000) Formation of hydro-
gen sulfide and glutathione during fermentation of white grape
musts. American Journal of Enology and Viticulture 51, 91–97.
Peddie, H.A.B. (1990) Ester formation in brewery fermentations.
Journal of the Institute of Brewing 96, 327–331.
Peltoniemi, K., Vesanto, E. and Palva, A. (2002) Genetic characteri-
zation of an oligopeptide transport system from Lactobacillus del-
brueckii subsp. bulgaricus. Archives of Microbiology 177, 457–467.
Peppard, T. (1988) Analytical measurements of volatile sulfur com-
pounds in beer. In: Beer Analysis. Eds H.F. Linkens and J.F. Jackson
(Springer-Verlag: Berlin, Germany) pp. 241–253.
Pickering, G.J., Heatherbell, D.A. and Barnes, M.F. (1999a) The pro-
duction of reduced-alcohol wine using glucose oxidase treated juice.
Part I. Composition. American Journal of Enology and Viticulture
50, 291–298.
Pickering, G.J., Heatherbell, D.A. and Barnes, M.F. (1999b) The pro-
duction of reduced-alcohol wine using glucose oxidase treated juice.
Part II. Stability and SO2-binding. American Journal of Enology
and Viticulture 50, 299–306.
Pickering, G.J., Heatherbell, D.A. and Barnes, M.F. (1999c) The pro-
duction of reduced-alcohol wine using glucose oxidase treated juice.
Part III. Sensory. American Journal of Enology and Viticulture 50,
307–316.
Pilone, G.J., Clayton, M.G. and van Duivenboden, R.J. (1991)
Characterization of wine lactic acid bacteria: single broth culture for
tests of heterofermentation, manitol from fructose and ammonia
from arginine. American Journal of Enology and Viticulture 42,
153–157.
Pollnitz, A.P., Pardon, K.H. and Sefton, M.A. (2000) 4-Ethylguaiacol
and oak lactones in Australian red wines. Australian Grapegrower
and Winemaker 438, 45, 47–50.
Porro, D.L., Brambilla, L., Ranzi, B.M., Martegani, E. and Alberghina,
L. (1995) Development of metabolically engineered Saccharomyces
cerevisiae cells for the production of lactic acid. Biotechnology
Progress 11, 294–298.

Swiegers, Bartowsky, Henschke & Pretorius Microbial modulation of wine aroma and flavour 171
Pretorius, I.S. (2000) Tailoring wine yeast for the new millennium:
novel approaches to the ancient art of winemaking. Yeast 16,
675–729.
Pretorius, I.S. (2003) The genetic analysis and tailoring of wine yeasts.
In: Genetics and Genomics of Industrial Yeasts. Ed. J.H. de Winde
(Springer-Verlag: Heidelberg, Germany) pp. 99–134.
Pretorius, I.S. (2004) The genetic improvement of wine yeasts. In:
Handbook of Fungal Biotechnology. Eds D.K. Arora, P.D. Bridge
and D. Bhatnagar (Marcel Dekker: New York, USA) pp. 183–223.
Pretorius, I.S. and Bauer, F.F. (2002) Meeting the consumer chal-
lenge through genetically customized wine-yeast strains. Trends in
Biotechnology 20, 426–432.
Pretorius, I.S., van der Westhuizen, T.J. and Augustyn, O.P.H. (1999)
Yeast biodiversity in vineyards and wineries and its importance to
the South African wine industry. South African Journal of Enology
and Viticulture 20, 61–74.
Pretorius, I.S., de Barros Lopes, M.A., Francis, I.L., Swiegers, J.H.
and Høj, P.B. (2004) The genomics of taste and smell – cracking the
codes that determine who likes what on the palate and on the
nose. Australian and New Zealand Wine Industry Journal 19, 13–18.
Pretorius, I.S., Bartowsky, E.J., Bauer, F.F., de Barros Lopes, M., du
Toit, M., van Rensburg P. and Vivier, M.A. (2005) The tailoring of
designer grapevines and microbial starter strains for a market-direct-
ed and quality-focussed wine industry. In: Handbook of Food
Science, Technology and Engineering. Volume 4 – Food Technology
and Food Processing. Ed. H.Y. Hui (CRC Press: Florida/Marcel
Dekker: New York, USA) pp. 174.1–174.24.
Pripis-Nicolau, L., de Revel, G., Bertrand, A. and Lonvaud-Funel, A.
(2004) Methionine catabolism and production of volatile sulphur
compounds by Oenococcus oeni. Journal of Applied Microbiology 96,
1176–1184.
Pronk, J.T., Steensma, H.Y. and van Dijken, J.P. (1996) Pyruvate
metabolism in Saccharomyces cerevisiae. Yeast 12, 1607–1633.
Radler, F. (1993) Yeast: Metabolism of organic acids. In Wine
Microbiology and Biotechnology. Ed. G.H. Fleet (Harwood Academic
Publishers: Chur, Switzerland) pp. 165–182.
Ramos, A. and Santos, H. (1996) Citrate and sugar cofermentation in
Leuconostoc oenos, a 13C nuclear magnetic resonance study. Applied
and Environmental Microbiology 62, 2577–2585.
Ramos, A., Lolkema, J.S., Konings, W.N. and Santos, H. (1995)
Enzyme basis for pH regulation of citrate and pyruvate metabolism
by Leuconostoc oenos. Applied and Environmental Microbiology 61,
1303–1310.
Rankine, B.C. (1963) Nature, origin and prevention of hydrogen
sulphide aroma in wines. Journal of the Science of Food and
Agriculture 14, 79–91.
Rankine, B.C. (1967) Influence of yeast strain and pH on pyruvic acid
content of wines. Journal of the Science of Food and Agriculture 18,
41–44.
Rankine, B.C. (1968a) The importance of yeasts in determining the
composition and quality of wines. Vitis 7, 22–49.
Rankine, B.C. (1968b) Formation of ketoglutaric acid by wine yeasts
and its oenological significance. Journal of the Science of Food and
Agriculture 19, 624–627.
Rankine, B.C. (1977) Modern developments in selection and use of
pure yeast cultures for winemaking. Australian Wine, Brewing and
Spirit Review 96(8), 31–33 and 96(9), 32–34.
Rankine, B.C. and Fornachon, J.C.M. (1964) Schizosaccharomyces
malidevorans sp.n., a yeast decomposing L malic acid. Antonie Van
Leeuwenhoek 30, 73–75.
Rankine, B.C. and Pocock, K.F. (1969) Influence of yeast strain on
binding of sulphur dioxide in wines, and on its formation during fer-
mentation. Journal of the Science of Food and Agriculture 20,
104–109.
Rankine, B.C., Fornachon, J.C.M. and Bridson, D.A. (1969) Diacetyl
in Australian dry red wines and its significance in wine quality.
Vitis 8, 129–134.
Rankine, B. and Bridson, D. (1971) Glycerol in Australian wines and
factors influencing its formation. American Journal of Enology and
Viticulture 23, 6–12.
Rapp, A. (1998) Volatile flavour of wine: Correlation between instru-
mental analysis and sensory perception. Nahrung 42, 351–363.
Rapp, A. and Mandery, H. (1986) Wine aroma. Experientia 42,
873–884.
Rapp, A. and Versini, G. (1991) Influence of nitrogen compounds in
grapes on aroma compounds in wine. Proceedings of the
International Symposium on Nitrogen in Grapes and Wine, Seattle,
USA (American Society of Enology and Viticulture: Davis, CA) pp.
156–164.
Rauhut, D. (1993) Yeasts – Production of sulfur compounds. In: Wine
Microbiology and Biotechnology. Ed. G.H. Fleet (Harwood Academic
Publishers: Chur, Switzerland) pp. 183–223.
Rauhut, D., Kürbel, H., Dittrich, H.H. and Grossmann, M. (1996)
Properties and differences of commercial yeast strains with respect
to their formation of sulfur compounds. Viticulture and Enology
Science 51, 187–192.
Rauhut, D., Gawron-Scibek, M., Beisert, B., Kondzior, M., Schwartz,
R., Kürbel, H., Grossmann, M. and Krieger, S.A. (2004) Impact of S-
containing manio acids and glutathione on growth of Oenococcus
oeni and malolactic fermentation. Paper presented at the Wine
Quality and Malolactic Fermentation – Les XVI Entretiens
Scientifiques Lallemand, Porto, Portugal.
Remize, F., Roustan, J.L., Sablayrolles, J.M., Barre, P. and Dequin, S.
(1999) Glycerol overproduction by engineered Saccharomyces cere-
visiae wine yeast strains leads to substantial changes in byproduct for-
mation and to a stimulation of fermentation rate in stationary phase.
Journal of Applied and Environmental Microbiology 65, 143–149.
Remize, F., Andrieu, E. and Dequin, S. (2000) Engineering of the
pyruvate dehydrogenase bypass in Saccharomyces cerevisiae: Role of
the cytosolic Mg2+ and mitochondrial K+acetaldehyde dehydroge-
nases Ald6p and Ald4p in acetate formation during alcoholic fer-
mentation. Journal of Applied and Environmental Microbiology
66, 3151–3159.
Ribéreau-Gayon, P., Dubourdieu, D., Donéche, B. and Lonvaud, A.
(2000a) Handbook of Enology, Volume 1: The Microbiology of
Wines and Vinification (John Wiley & Sons Ltd: Chichester, UK).
Ribéreau-Gayon, P., Glories, Y., Maujean, A. and Dubourdieu, D.
(2000b) Handbook of Enology Volume 2: The Chemistry of Wine
Stabilisation and Treatments (John Wiley & Sons Ltd: Chichester,
UK).
Richter, H., Hamann, I. and Unden, G. (2003) Use of the mannitol
pathway in fructose fermentation of Oenococcus oeni due to limiting
redox regeneration capacity of the ethanol pathway. Archives of
Microbiology 179, 227–233.
Riesen, R. (1992) Undesirable fermentation aromas. Paper presented
at the ASEV/ES workshop – Wine aroma defects, Corning, NY.
Riou, C., Salmon, J.M., Vallier, M.J., Günata, Z. and Barre, P. (1998)
Purification, characterization, and substrate specificity of a novel
highly glucose-tolerant β-glucosidase from Aspergillus oryzae. Applied
Environmental Microbiology 64, 3607–3614.
Robinson, J. (1999) The Oxford Companion to Wine. 2nd edn
(Oxford University Press: Oxford).
Rodriguez, S.B. and Thornton, R.J. (1988) Rapid utilisation of malic
acid by a mutant of Schizosaccharomyces malidevorans. Proceedings of
the Second International Symposium for Cool Climate Viticulture
and Oenology, Auckland, New Zealand. (New Zealand Society of
Viticulture and Oenology: Auckland, New Zealand) pp. 313–315.
Rodriguez, S.B. and Thornton, R.J. (1989) A malic acid dependent
mutant of Schizosaccharomyces malidevorans. Archives of Microbiology
152, 564–566.
Rojas, V., Gil, J.V., Pinãga, F., Manzanares, P. (2003) Acetate ester for-
mation in wine by mixed cultures in laboratory fermentations.
International Journal of Food Microbiology 86, 181–188.
Romano, P., Suzzi, G., Turbanti, L. and Polsinelli, M. (1994) Acet-
aldehyde production in Saccharomyces cerevisiae wine yeasts. FEMS
Microbiology Letters 118, 213–218.
Romano, P., Fiore, C., Paraggio, M., Caruso, M. and Capece, A. (2003)
Function of yeast species and strains in wine flavour. International
Journal of Food Microbiology 86, 169–180.
Rosi, I., Vinella, M. and Domizio, P. (1994) Characterization of β-
glucosidase activity in yeasts of oenological origin. Journal of Applied
Bacteriology 77, 519–527.
Roustan, J.L. and Sablayrolles, J.M. (2002) Impact of the addition of
electron acceptors on the by-products of alcoholic fermentation.

172 Microbial modulation of wine aroma and flavour Australian Journal of Grape and Wine Research 11, 139–173, 2005
Enzyme and Microbial Technology 31, 142–152.
Rozes, N., Arola, L. and Bordons, A. (2003) Effect of phenolic com-
pounds on the co-metabolism of citric acid and sugars by Oenococcus
oeni from wine. Letters in Applied Microbiology 36, 337–341.
Saint-Prix, F., Bonquist, L. and Dequin, S. (2004) Functional analysis
of the ALD gene family of Saccharomyces cerevisiae during anaerobic
growth on glucose: The NADP(+)-dependent Ald6p and Ald5p iso-
forms play a major role in acetate formation. Microbiology 150,
2209–2220.
Scanes, K.T., Hohmann, S. and Prior, B.A. (1998) Glycerol production
by the yeast Saccharomyces cerevisiae and its relevance to wine: A
review. South African Journal of Enology and Viticulture 19, 17–24.
Schreier, P. (1979) Flavour composition of wines. CRC Critical
Reviews in Food Science and Nutrition, 59–111.
Schulthess, D. and Ettlinger, L. (1978) Influence of the concentration
of branched chain amino acids on the formation of fusel alcohols.
Journal of the Institute of Brewing 84, 240–243.
Sefton, M.A. (1998) Hydrolytically-released volatile secondary
metabolites from a juice sample of Vitis vinifera grape cvs Merlot and
Cabernet Sauvignon. Australian Journal of Grape and Wine
Research 4, 30–38.
Sefton, M.A., Francis, I.L. and Williams, P.J. (1993) The volatile com-
position of Chardonnay juices: A study by flavor precursor analysis.
American Journal of Enology and Viticulture 44, 359–370.
Sefton, M.A., Francis, I.L. and Williams, P.J. (1994) Free and bound
secondary metabolites of Vitis vinifera grape cv. Sauvignon Blanc.
Journal of Food Science 59, 142–147.
Sefton, M.A., Francis, I.L. and Williams, P.J. (1996) The free and
bound volatile secondary metabolites of Vitis vinifera grape cv.
Semillon. Australian Journal of Grape and Wine Research 2,
179–183.
Shinohara, T., Kubodera, S. and Yanagida, F. (2000) Distribution of
phenolic yeasts and production of phenolic off flavours in wine
fermentation. Journal of Bioscience and Bioengineering 90, 90–97.
Shoseyov, O., Bravdo, B.A., Ikan, R. and Chet, I. (1990) Immobilized
endo-β-glucosidase enriches flavor of wine and passion fruit juice.
Journal of Agricultural and Food Chemistry 27, 1973–1976.
Siebert, T.E., Smyth, H.E., Capone, D.L., Neuwohner, C., Pardon,
K.H., Skouroumounis, G.K., Herderich, M.J., Sefton, M.A. and
Pollnitz, A.P. (2005) Stable isotope dilution analysis of wine
fermentation products by HS-SPME-GC-MS. Analytical and
Bioanalytical Chemistry 381, 937–947.
Simpson, R.F. (1979) Some important aroma components of white
wine. Food Technology Australia 31, 516–522.
Simpson, R.F. and Miller, G.C. (1983) Aroma composition of aged
Riesling wine. Vitis 22, 51–63.
Skory, C.D. (2003) Lactic acid production by Saccharomyces cerevisiae
expressing a Rhizopus oryzae lactate dehydrogenase gene. Journal of
Industrial Microbiology and Biotechnology 30, 22–27.
Slaughter, J.C. and McKernan, G. (1988) The influence of pan-
tothenate concentration and inoculum size on the fermentation of
a defined medium by Saccharomyces cerevisiae. Journal of the Institute
of Brewing 94, 14–18.
Smit, A., Cordero Otero, R.R.C., Lambrechts, M.G., Pretorius, I.S.
and Van Rensburg, P. (2003) Enhancing volatile phenol concentra-
tions in wine by expressing various phenolic acid decarboxylase
genes in Saccharomyces cerevisiae. Journal of Agricultural and Food
Chemistry 51, 4909–4915.
Smyth, H.E. (2005) The compositional basis of the aroma of Riesling
and unwooded Chardonnay wines. PhD thesis (University of Adelaide,
Adelaide, Australia).
Smyth, H., Cozzolino, D., Herderich, M.J., Sefton, M.A. and Francis,
I.L. (2005) Relating volatile composition to wine aroma:
Identification of key aroma compounds in Australian white wines.
Proceedings of the Twelfth Australian Wine Industry Technical
Conference, Melbourne, Australia (Australian Wine Industry
Technical Conference Inc.: Adelaide, SA) pp. 31–33.
Soles, R.M., Ough, C.S. and Kunkee, R.E. (1982) Ester concentration
differences in wine fermented by various species and strains of
yeasts. American Journal of Enology and Viticulture 33, 94–98.
Somers, T.C. and Wescombe, L.G. (1987) Evolution of red wines. II.
An assessment of the role of acetaldehyde. Vitis 26, 27–36.
Spettoli, P., Nuti, M.P. and Zamorani, A. (1984) Properties of malo-
lactic activity purified from Leuconostoc oenos ML34 by affinity chro-
matography. Applied and Environmental Microbiology 9, 184–189.
Spiropoulos, A. and Bisson, L.F. (2000) MET17 and hydrogen sulfide
formation in Saccharomyces cerevisiae. Applied and Environmental
Microbiology 66, 4421–4426.
Spiropoulos, A., Tanaka, J., Flerianos, I. and Bisson, L.F. (2000)
Characterization of hydrogen sulfide formation in commercial and
natural wine isolates of Saccharomyces. American Journal of Enology
and Viticulture 51, 233–248.
Sponholz, W.-R. (1993) Wine spoilage by microorganisms. In: Wine
Microbiology and Technology. Ed. G.H. Fleet (Harwood Academic
Publishing: Amsterdam) pp. 395–420.
Stratford, M. and Rose, A.H. (1985) Transport of sulfur dioxide by
Saccharomyces cerevisiae. Journal of General Microbiology 132, 1–6.
Suomalainen, H. and Lehtonen, M. (1979) The production of aroma
compounds by yeast. Journal of the Institute of Brewing 85,
149–156.
Sutherland, C.M., Henschke, P.A., Langridge, P. and de Barros Lopes,
M. (2003) Subunit and cofactor binding of Saccharomyces cerevisiae
sulfite reductase – towards developing wine yeast with lowered
ability to produce hydrogen sulfide. Australian Journal of Grape and
Wine Research 9, 186–193.
Swiegers, J.H. and Pretorius, I.S. (2005) Yeast modulation of wine
flavour. Advances in Applied Microbiology 57, 131–175.
Swiegers, J.H., Willmott, R., Hill-Ling, A., Capone, D.L., Pardon, K.H.,
Elsey, G.M., Howell, K.S., de Barros Lopes, M.A., Sefton, M.A.,
Lilly, M. and Pretorius, I.S. (2005) Modulation of volatile thiol and
ester aromas in wine by modified wine yeast. Proceedings of the
Weurman Flavour Research Symposium, Roskilde, Denmark.
Then, R. and Radler, F. (1971) Vergleichende Untersuchung der
Acetaldehydbildung bei der aroben Vergärung von Glucose bei ver-
scheidenen Stämmen von Saccharomyces cerevisiae und Saccharomyces
carlsbergensis. Monatsschrift für Brauerei 24, 127–130.
Thomas, C.S., Boulton, R.B., Silacci, M.W. and Gubler, W.D. (1993)
The effect of elemental sulfur, yeast strain, and fermentation medi-
um on hydrogen sulfide production during fermentation. American
Journal of Enology and Viticulture 44, 211–216.
Thomas, D. and Surdin-Kerjan, Y. (1997) Metabolism of sulfur amino
acids in Saccharomyces cerevisiae. Microbiology and Molecular Biology
Review 61, 503–532.
Thornton, R.J. and Bunker, A. (1989) Characterisation of wine yeasts
for genetically modifiable properties. Journal of the Institute of
Brewing 95, 181–184.
Thornton, R.J. and Rodriguez, S.B. (1996) Deacidification of red and
white wines by a mutant of Schizosaccharomyces malidevorans under
commercial winemaking conditions. Food Microbiology 13,
475–482.
Thoukis, G., Ueda, M. and Wright, D. (1965) The formation of succinic
acid during alcoholic fermentation. American Journal of Enology
and Viticulture 16, 1–8.
Thurston, P.A., Taylor, R. and Ahvenainen, J. (1981) Effects of lino-
leic acid supplements on the synthesis by yeast of lipids and acetate
esters. Journal of the Institute of Brewing 87, 92–95.
Timberlake, C.F. and Bridle, P. (1976a) The effect of processing and
other factors on the colour characteristics of some red wines. Vitis 15,
37–49.
Timberlake, C.F. and Bridle, P. (1976b) Interactions between antho-
cyanins, phenolic compounds, and acetaldehyde and their signifi-
cance in red wines. American Journal of Enology and Viticulture 27,
97–105.
Tominaga, T., Masneuf, I. and Dubourdieu, D. (1995) A S-cysteine
conjugate, precursor of aroma of white sauvignon. Journal
International des Sciences de la Vigne et du Vin 29, 227–232.
Tominaga, T., Darriet, P. and Dubourdieu, D. (1996) Identification of
3-mercaptohexyl acetate in Sauvignon wine, a powerful aromatic
compound exhibiting box-tree odor. Vitis 35, 207–210.
Tominaga, T., Peyrot des Gachons, C. and Dubourdieu, D. (1998a) A
new type of flavour precursor in Vitis vinifera L. cv. Sauvignon Blanc:
S-Cysteine conjugates. Journal of Agricultural and Food Chemistry
46, 5215–5219.
Tominaga, T., Furrer, A., Henry, R. and Dubourdieu, D. (1998b)

Swiegers, Bartowsky, Henschke & Pretorius Microbial modulation of wine aroma and flavour 173
Identification of new volatile thiols in the aroma of Vitis vinifera L.
var. Sauvignon blanc wines. Flavour and Fragrance Journal 13,
159–162.
Tominaga, T., Baltenweck-Guyot, R., Peyrot de Gachons, C. and
Dubourdieu, D. (2000a) Contribution of volatile thiols to the aromas
of white wines made from several Vitis vinifera grape varieties.
American Journal of Enology and Viticulture 51, 178–181.
Tominaga, T., Blanchard, L., Darriet, P. and Dubourdieu, D. (2000b).
A powerful aromatic volatile thiol, 2-furanmethanethiol, exhibiting
roast coffee aroma in wines made from several Vitis vinifera grape
varieties. Journal of Agricultural Food Chemistry 48, 1799–1802.
Ugliano, M., Genovese, A. and Moio, L. (2003) Hydrolysis of wine
aroma precursors during malolactic fermentation with four com-
mercial starter cultures of Oenococcus oeni. Journal of Agricultural and
Food Chemistry 51, 5073–5078.
Usseglio Tomasset, L. and Di Stefano, R. (1980) Profilo aromatico del
Moscato bianco del Piemonte. Rivista di Viticoltura e di Enologia 33,
3–13.
van Beek, S. and Priest, F.G. (2000) Decarboxylation of substituted
cinnamic acids by lactic acid bacteria isolated during malt whisky
fermentation. Applied and Environmental Microbiology 66,
5322–5328.
van Rensburg, P. and Pretorius, I.S. (2000) Enzymes in winemaking:
harnessing natural catalysts for efficient biotransformations. South
African Journal of Enology and Viticulture 21, 52–73.
van Rensburg, P., Stidwell, T., Lambrechts, M.G., Cordero Otero, R.R.
and Pretorius, I.S. (2005) Development and assessment of a recom-
binant Saccharomyces cerevisiae wine yeast producing two aroma-
enhancing β-glucosidases encoded by the Saccharomycopsis fibuligera
BGL1 and BGL2 genes. Annals of Microbiology 55, 33–42.
van Uden, N.J. (1989) Effects of alcohols on membrane transport in
yeasts. In: Alcohol Toxicity in Yeasts and Bacteria. Ed. N.J. van
Uden (CRC Press: Boca Raton) pp. 135–146.
Vasserot, Y., Arnaud, A. and Galzy, P. (1993) Evidence for marc
monoterpenol glycosides hydrolysis by free or immobilized yeast β-
glucosidase. Bioresource Technology 43, 269–271.
Vasserot, Y., Christiaens, H., Chemardin, P., Arnaud, A. and Galzy, P.
(1989) Purification and properties of a β-glucosidase of Hanseniaspora
vineae with the view to its utilization in fruit aroma liberation.
Journal of Applied Bacteriology 66, 271–279.
Veiga-da-Cunha, M., Santos, H. and van Schaftingen, E. (1993)
Pathway and regulation of erythritol in Leuconostoc oenos. Journal of
Bacteriology 175, 3941–3948.
Vermeulen, C., Gijs, L. and Collin, S. (2005) Sensorial contribution
and formation pathways of thiols in foods: A review. Food Reviews
International 21, 69–137.
Verstrepen, K., Derdelinckx, G., Dufour, J.-P., Winderickx, J.,
Thevelein, J.M., Pretorius, I.S. and Delvaux, F.R. (2003) Flavour-
active esters: Adding fruitiness to beer – A practical review. Journal
of Bioscience and Bioengineering 96, 110–118.
Villaneuva, A., Ramon, D., Valles, S. Lluch, M. A. and MacCabe, A. P.
(2000) Heterologous expression in Aspergillus nidulans of a
Trichoderma longibrachiatum enoglucanase of enological relevance.
Journal of Agricultural and Food Chemistry 48, 951–957.
Vivas, N. and Glories, Y. (1995) Vinification et élévage des vins.
Potential d’oxydoréduction en oenologie. Revue des Oenologues et
des Techniques Vitivinicoles et Oenologiques 76, 10–14.
Vivas, N., Lonvaud-Funel, A. and Glories, Y. (1997) Effect of pheno-
lic acids and anthocyanins on growth, viability and malolactic activ-
ity of a lactic acid bacterium. Food Microbiology 14, 291–300.
Vizoso Pinto, M.G., Pasteris, S.E. and Strasser de Saad, A.M. (2004)
Glycerol catabolism by Pediococcus pentosaceus isolated from beer.
Food Microbiology 21, 111–118.
Voirin, S.G., Baumes, R.L., Bitteur, S.M., Günata, Z.Y. and Bayonove,
C.L. (1990) Novel monoterpene disaccharide glycosides of Vitis
vinifera grapes. Journal of Agricultural and Food Chemistry 38,
1373–1378.
Volschenk, H., Viljoen, M., Grobler, J., Petzold, B., Bauer, F. F., Subden,
R., Young, R.A., Lonvaud, A., Denayrolles, M. and van Vuuren,
H.J.J. (1997) Engineering pathways for malate degradation in
Saccharomyces cerevisiae. Nature Biotechnology 15, 253–257.
von Weymarn, N., Hujanen, M. and Leisola, M. (2002) Production of
D-mannitol by heterofermentative lactic acid bacteria. Process
Biochemistry 37, 1207–1213.
Vos, P.J.A. and Gray, R.S. (1979) The origin and control of hydrogen
sulphide during fermentation of grape must. American Journal of
Enology and Viticulture 30, 187–197.
Wang, X.D., Bohlscheid, J.C. and Edwards, C.G. (2003) Fermentative
activity and production of volatile compounds by Saccharomyces
grown in synthetic grape juice media deficient in assimilable nitro-
gen and/or pantothenic acid. Journal of Applied Microbiology 94,
349–359.
Whiting, G.C. (1976) Organic acid metabolism of yeasts during fer-
mentation of alcoholic beverages – A review. Journal of the Institute
of Brewing 82, 84–92.
Wibowo, D., Eschenbruch, R., Davis, C.R., Fleet, G.H. and Lee, T.H.
(1985) Occurrence and growth of lactic acid bacteria in wine: A
review. American Journal of Viticulture and Enology 36, 302–313.
Williams, P.J., Strauss, C.R., Wilson, B. and Massywestropp, R.A.
(1982a) Studies on the hydrolysis of Vitis vinifera monoterpene pre-
cursor compounds and model monoterpene β-D-glucosides ratio-
nalizing the monoterpene composition of grapes. Journal of
Agricultural and Food Chemistry 30, 1219–1223.
Williams, P.J., Strauss, C.R., Wilson, B. and Massy-Westropp, R.A.
(1982b) Novel monoterpene disaccharide glycosides of Vitis vinifera
grapes and wines. Phytochemistry 21, 2013–2020.
Williams, P.J., Strauss, C.R., Wilson, B. and Dimitriadis, E. (1984)
Recent studies into grape terpene glycosides. In: Progress in Flavour
Research. Ed. J. Adda (Elsevier: Amsterdam, The Netherlands) pp.
349–357.
Williams, P.J., Sefton, M.A. and Wilson, B. (1989) Nonvolatile con-
jugates of secondary metabolites as precursors of varietal grape
flavour components. In: Flavour Chemistry, Trends and
Developments, ACS Symposium Series No 388. Eds R. Teranishi,
R.G. Buttery and R. Shahisi (American Chemical Society:
Washington, DC) pp. 35–48.
Winterhalter, P. (1991) 1,1,6-Trimethyl-1,2-dihydronaphthalene
(TDN) formation in wine .1. Studies on the hydrolysis of 2,6,10,10-
tetramethyl-1-oxaspiro[4.5]dec-6-ene-2,8-diol rationalizing the ori-
gin of TDN and related C-13 norisoprenoids in Riesling wine.
Journal of Agricultural and Food Chemistry 39, 1825–1829.
Winterhalter, P. and Schreier, P. (1994) C13-Norisoprenoid glyco-
sides in plant tissues: an overview on their occurrence, composition
and role as flavour precursors. Flavour and Fragrance Journal 9,
281–287.
Wisselink, H.W., Weusthuis, R.A., Eggink, G., Hugenholtz, J. and
Grobben, G.J. (2002) Mannitol production by lactic acid bacteria: A
review. International Dairy Journal 12, 151–161.
Yamagata, S. (1989) Roles of O-acetyl-L-homoserine sulfhydrylases in
micro-organisms. Biochimie 71, 1125–1143.
Yoshimoto, H., Fujiwara, D., Momma, T., Ito, C., Sone, H., Kaneko, Y.
and Tamai, Y. (1998) Characterization of the ATF1 and Lg-ATF1
genes encoding alcohol acetyltransferases in the bottom fermenting
yeast Saccharomyces pastorianus. Journal of Fermentation and
Bioengineering 86, 15–20.
Younis, O.S. and Stewart, G.G. (1998) Sugar uptake and subsequent
ester and higher alcohol production by Saccharomyces cerevisiae.
Journal of Institute of Brewing 104, 255–264.
Zago, A., Degrassi, G. and Bruschi, C. V. (1995) Cloning, sequencing,
and expression in Escherichia coli of the Bacillus pumilus gene for fer-
ulic acid decarboxylase. Applied and Environmental Microbiology
61, 4484–4486.
Zambonelli, C. (1964) Ricerche biometriche sulla produzione di
idrogeno solforato da solfati e solfiti in Saccharomyces cerevisiae var.
ellipsoidesus. Annuals of Microbiology 14, 129.
Zeeman, W., Snyman, J.P. and van Wyk, C.J. (1982) The influence of
yeast strain and malolactic fermentation on some volatile bouquet
substances and on quality of table wines. Paper presented at the
Grape and Wine Centennial Symposium, University of California,
Davis.
Manuscript received: 23 May 2005
Revised manuscript received: 27 May 2005