~
, IIIER[)!r::. EK(T i.' A.N\ l.iA ~ 0'; •.~!: \l
I \ : C'p. ('.~ST. T[;lG}:::'!:'[ LIlu.:
1
~\ I!,![',lnr:-rl' r; 4)")'''1l
\l".\':D •. \ \ •• I, "I , I \\
\L_ ...
.,.,·
..,_..
·_"
U 0 V
. .
..
S
-
B
IBL
t
OTEEK
*198706292201220000019*
I~~~~~~~~~m~rn~~~~~~~~~~~rn~~~~~~~rn~rn~rn~rn~rn~~~~m
* 1 Q R 7nA ~ Q ~ ~ n 1 ~ ~n n n n n 1 Q*_.
f':.l~q.
t'j'j\'l\.t
'\~t
...
_
• •
•
.
_
-.:---....
.
__
..
_
----
-
...
by
METABOLITE PATTERNS OF YEASTS ASSOCIATED WITH WINE
Hendrik Gabriël Tredoux
Submitted in fulfilment of the requirements for the degree
MAGISTER SCIENTIAE
in the
DepartMent of Microbiology, Faculty of Science, University of the Orange Free State, Bloemfontein
9300, Republic of South Africa
Promotor: Dr.
Instead, He left it as raw material - to tease us, to tantalize us, to set us thinking and experimenting and risking and adventuring, and therein we find our supreme interest in living.
God gave us the world unfinished so that we might share in the joys and satisfactions of creation - He left the oil in the rock, He left the forests un-felled and the cities un-built, He left the music un-sung and the dramas un-played, He left the poetry un-dreamed in order that men and women might not become bored but engage in stimulating, exciting and experiencing all the joys and durable satisfactions of achievement. He
gave us the challenge of raw materials, not the satisfaction of perfect finished things. Works, thought, creation - these give life its stimulus, its real satisfaction, its intriguing value.
LOUIZE
CONTENTS
ACKNOWLEDGEMENTS
CHAPTER 1
CHAPTER 2
INTRODUCTION
1.1 The need for a yeast identification system in the wine industry.
1.2 Definition of yeasts.
1.3 Development of yeast taxonomy.
1.4 Problems encountered with the conventional taxonomic system.
1.5 Purpose of the research.
1.5.1 The value of long-chain fatty acid composition in the taxonomy of wine- and related yeasts.
The use of volatile metabolites in 1. 5. 2
the identification of yeasts associated with wine.
THE VALUE OF CELLULAR LONG-CHAIN FATTY ACID COMPOSITION IN THE TAXONOMY OF WINE- AND RELATED YEASTS
2.1 Introduction
2.1.1 The use of cellular long-chain fatty acid composition in the identification of yeasts associated with the wine industry.
The value of cellular long-chain fatty
2.1.2
acid composition in the taxonomy of wine- and related yeasts.
2.1.2.1 The relation between long-chain fatty acid composition and the degree of mycelium formation. 1 1 2 2 8 10 12 21 23 24 24 24
CHAPTER 3
CHAPTER 4
24
2.i.2.2 The value of long-chain fatty acid composition in the
phylo-geny of the genus Kluyveromyces 24
2.1.2.3 The value of long-chain fatty acid composition in the tax-onomy of the genus Saccharo-myces
2.2 Materials and Methods
2.3 Results and Discussion
2.3.1 The use of cellular long-chain fatty acid composition in the identification of yeasts associated with the wine industry.
The value of cellular long-chain fatty acid composition in the taxonomy of wine- and related yeasts.
2.3.2.1 The relation between long-chain fatty acid composition and the complexity of mycelium
forrnation.
2.3.2.2 The value of long-chain fatty acid composition in the
phylogeny of the genus
Kluyveromyces.
2.3.2.3 The value of long-chain fatty acid composition in the tax-onomy of the genus
Saccharo-2.3.2
myces
THE USE OF VOLATILE METABOLITES IN THE IDENTIFICA-TION OF YEASTS ASSOCIATED WITH WINE
GENERAL DISCUSSION AND CONCLUSIONS
4.1 The value of cellular long-chain fatty acid composition in the taxonomy of wine- and related yeasts. 24 26 26 31 31 34 38 51 61
64
4.1.1 The use of cellular long-chain fatty acid composition in the identification of yeasts associated with the wine industry.
4.1.2 The value of cellular long-chain
fatty acid composition in the taxonomy and phylogeny of some wine- and
related yeasts.
61
4.2 The use of volatile metabolites in
the'iden-tification of yeasts associated with wine. 65
4.3 Future Research 65
SUMMARY 69
\
-000-I wish to sincerely thank the following for contributing towards the successful completion of this study:
Dr. J.L.F. Kock, Department of Microbiology, University of the Orange Free State, for his able guidance and constructive criticism during the course of this study;
Prof. P.M. Lategan, Head of the Department of Microbiology, University of the Orange Free State, for his interest and advice;
The executive, and in particular Mr. O.P.H. Augustyn, of· the Viticultural and Oenological Research Institute, Stellenbosch, for their support and interest in this study;
The Department of Agriculture for their permission to use the results of this study for scription purposes;
Susan Erasmus, for her valuable technical assistance;
-Mr. P.J. Bates, for assistance during the gas chromatographic analyses of the cellular fatty acids;
The personnel and fello~ students of the Microbiology Department, U.O.F.S., for their support and friendship;
Mr. J. Marais, for his assistance in analysing the volatile metabolites; Mr. and Mrs. F. Weldhagen and Mr. and Mrs. A. Conradie of Bloemfontein for
their support during the course of this study; Carol Viljoen, for the typing· of this dissertation;
My mother and my parents-in-law for their love, interest and encouragement; My wife and daughter, Marthelize, for their love, patience and endurance.
and
Finally, to the ALMIGHTY, \vithout whom this study would not have been possible.
INTRODUCTION
1.1 THE NEED FOR A YEAST IDENTIFICATION SYSTEM IN THE WINE INDUSTRY
The South African wine industry annually ferments approximately 10 million hectolitres of grape must wf.th a total value of 243 million rands (KW
annual report, 1985). This fermentation relies almost exclusively on the inoculation of selected Saccharomyces cerevisiae strains. The inoculation with these strains (supplied as active dried wine yeast) should ensure
swift fermentations and the production of quality wines minimizing negative qualities such as off-odours and -tastes.
The practice of inoculating grape must, instead of allowing spontaneous fermentation by the natural flora, prevails in many of the newer wine producing countries such as Australia and South Africa and is also gaining ground in the traditional wine producing areas of France and Germany
(Rankine, 1968).
The presence of contaminants in the inoculum and later during the fermentation may hamper the fermentation process and reduce the quality of the wine (Rankine, 1968; RadIer, 1973; Heard and Fleef, 1985) • Sacch. cerevisiae strains are selected for their fermentative capability and ability to produce wines without off-odours such as H2S or ethylacetate . Contaminants in the inoculum can produce these off-flavours in such quantities that the quality of the wine is severely affected. Since these contaminants, so-called wild yeasts, are almost always of lesser fermentative capability (Rankine, 1968), they can reduce the rate of fermentation - leading to sluggish or stuck fermentations which again leads to a reduction in wine quality.
Stuck or lagging fermentations have lately become a major problem, causing concern in the local wine industry (Tromp, 1980). Although this is a multi-faceted problem, one of the causes could be related to contaminated inocula or the domination, during fermentation, of the selected yeast strain by another that does not possess the ability to conduct a satisfactóry fermentation.
Consequently, in order to conduct a satisfactory fermentation, there is a need for a yeast, identification system which can detect contamination by wild yeasts. In order to accomplish this, it is important that yeast
species and strains are well defined and that an appropriate yeast identification and classification system exists.
1.2 DEFINITION OF YEASTS
In order to give a definition of the term "yeast"· it is important to retrace some aspects of the historical development.
Antonie van Leeuwenhoek (1680) defined yeasts as globular to spherical bodies which were found in beer and are able to multiply by budding (Phaff
et al., 1978). Eventually these yeasts were termed "zuckerpilz" or "sugar fungus" from which the name Saccharomyces originates (Brock, 1961).
Since these definitions were unsatisfactory, yeasts were described by other investigators as unicellular organisms that reproduce asexually by budding, fission or both and produce ascospores under suitable conditions within a naked ascus, originating either from a zygote or parthenogenically
from a single cell (Alexopoulos and Mims, 1979; Gorin and Spencer, 1970; Kreger-van Rij, 1969 and Phaff et al., 1966). In the above definitions, mainly morphological aspects were taken into consideration.
Flegel (1977), on the other hand, defin~d yeasts as assimilative growth forms which are un Lce LluLa r and reproduce by budding or fission. Finally, in 1985, van der WaIt and van Arx stated that "yeasts are hyaline microfungi which, with numerous exceptions, reproduce asexually by budding, ferment at least glucose and form naked asci".
1.3 DEVELOPMENT OF YEAST TAXONOMY
The techniques currently in use for the classification of yeasts to species level are based upon morphological, physiological, sexual and biochemical characteristics (Barnett et al., 1983). This system has evolved through the years as a result of the work of several authors.
Reess (1870) observed endospores in different yeasts and described their shapes apd mode of germination. He also suggested the name Saccharomyces
for.the spore-forming yeasts and they were included in the Ascomycetes. De Bary stated in 1884 that yeast spores and ascospores were produced by "free cell formatt on'", These spores were free from attachment to the cell wall in contrast with spore formation in other classes of fungi. Hansen perfected Pasteur's methods for obtaining pure cultures and also studied some morphological and physiological aspects of these cultures. He also attempted the first comprehensive system of yeast taxonomy in 1896. Many of the spec Les differentiated by Han sen are still recognized today
(Phaff et al., 1978).
From 1920 to 1928 Guilliermond expanded the field of taxonomy with additional information on physiology, sexuality and phylogenetic relations. He also devised various dichotomous keys for identifying yeast species (Guilliermond, 1920; 1928).
The Delft school of taxonomists under the inspiration of Kluyver produced mainly six leading contributions on taxonomy from 1931 to 1984:
1931: Stelling-Dekker produces a scheme of classification for the sporulating yeasts.
Wickerham (1951) introduced new techniques and principles e. g. synthetic media for the study of morphology and assimilation tests with more carbon compounds and vitamins.. He also put greater emphasis on the chromosomal state of the yeast in nature and the existence of heterothallic mating types.
Kudrjawzew (1954) classified the yeasts in a new order, the Unicellomy-cetales, later changed to Saccharomycetales (Kudrjawzew, 1960). The Saccharomycetales was divided into three families according to their mode of vegetative reproduction, i.e. Saccharomycetaceae (budding), Schizosaccharomycetaceae (fission) and Saccharomycodaceae (bud-fission).
In 1956, Wickerham and Burton proposed the genus Dekk.eromucee , which is similar to the genera Eaboepora and Zygofabospora. These three genera are now merged into the genus Kluyveloomyces (van der Walt emend. van der Halt).
Phaff et al. (1978) proposed three yeast families which contained the known ascosporogenous yeasts namely the Saccharomycetaceae, Sporobolomy-cetaceae and Cryptococcaceae. The first group was divided into six
1934: 1942: 1952: 1970: 1984: There were subfamilies:
Lodder publishes a volume on non-sporeforming yeasts. Diddens and Lodder publishes a second volume on non-spore forming yeasts.
Lodder and Kreger-van Rij produces a comprehensive
classification of both sporogenous and asporogenous yeasts.
Lodder edits a comprehensive volume on yeasts.
Kreger-van Rij edits a comprehensive volume on yeasts.
also other important contributions to yeast taxonomy.
Schizosaccharomycetoideae, Endomycetoideae, Uptomycetoideae, Nematosporoideae, Saccharomycetoideae and Eremascoideae.
In a more recent classification system (Von Ar x , 1981), the ascosporogenous yeasts were divided into six families (Table 1).
The yeasts associated w ith the wine industry (Barnett et al . , 1983) are given in Table 2.
NematospoY'a Asbya
CY'ebY'othecium EY'emothecium
Dipodascaceae Endomycetaceae Saccharomycodaceae Saccharomycetaceae Metchnikowiaceae Schizosaccharomycetaceae
Dipodascus Endomyces SacchaY'omycodes SacchaY'omyces. Metchnikowia SchizosacchaY'omyces Ascoidea Cephaloascus AmbY'osiozyma HOY'moascus BotY'yoascus Hyphopichia Stephanoascus Pichia Hansenula pachysolen DekkeY'a HanseniaspoY'a Nadsonia CYickeY'hamia ZygosacchaY'omyces TOY'ulaspoY'a DebaY'yozyma Issatchenkia PachytichospoY'a SpoY'opachydeY'ma KluyveY'omyces LoddeY'omyces ClavispoY'a WickeY'hamiella CiteY'omyces Wingea WiUiopsis Schwanniomyces Endomycopsella SacchaY'omycopsis AY'thY'oascus Cyniclomyces Lipomyces
.10-Table 2: Yeasts associated with wine and wine-making (Barnett et al.,,
1983)
Brettanomyces cZaussenii Custers
Brettanomyces.custersii Florenzano
Brettanomyces Zambicus Kufferath & van Laer
*Candida aZbicans (Robin) Berkhout
Candida apicoZa (Hajsig) Meyer & Yarrow
Candida boidinii Ramirez
Candida cantarel.l.i
i
(van der I~alt & van Kerken) Meyer & YarrowCandida catenuZata Diddens& Lodder
Candida divevea Oha ra et al., ex van Uden & Buckley
Candida gZabrata (Anderson) Meyer & Yarrow
Candida incommunis Ohara et al.,
Candida inconspicua (Lodder & Kreger-van Rij) Meyer & Yarrow
Candida intermedia (Cifferri & Ashford) Langeron & Guerra
Candida norvegica (Reiers61) Meyer & Yarrow
Candida parapsiZosis (Ashford) Langeron & Talice
*Candida ruqosa (Anderson) Diddens & Lodder
Candida sake (Saito & Ota) van Uden & Buckley
Candida soZani Lodder & Kreger-van Rij
*Candida steatoZytica Yarrow
Candida steZZata (Kroemer & Krumbholz) Meyer & Yarrow
*Candida tenuis Diddens & Lodder
Candida tropicaZis (Castellani) Berkhout
Candida vandeY'WaZtii (Vidal-Leiria) Meyer & Yarrow
Candida veronae Florenzano ex van Uden & Buckley
Candida vereatii.l ie (Etchells & Bell) Heyer & Ya rrow
Candida vini (Desmazieres) van Uden & Buckley
Candida zeyZanoides (Castellani) Langeron & Guerra
Ci.tieromucee matritensis (Santa Maria) Santa Haria
*Cl?yptococcus al.bidue (Saito) Skinner
Cryptcoccus humicoZus (Daszewska) Golubev
Cl~ptococcus Zaurentii (Kufferath) Skinner
Cryptococcus ZuteoZus (Saito) Skinner
*Debaryomyces hansenii (Zopf) Lodder & Kreger-van Rij
Debaryomyces poZymorphus (Klocker) Price & Phaff
Dekkera bruxeZZensis van der Walt
Dekk.era i.ntermedia van der 1\7alt
Filobasidie Tla neoformans Kwon -Chung
*Filobasidium capsuligenvm Rodrigues de Miranda
Geotir-ichum ['ermentiane (Diddens & Lodder) von Ar x
Hanseniaspora occidentalis Smith
Haneeniaepora osmophila (Niehaus) Phaff et al. "Baneen iaepora uvarum (Niehaus) Shehata et al. *Hanseniaspora valbyensis Kloeker
Hanseniaspora vineae van der Walt & Tscheuschner
*Hyphopichia burtonii (Boidin et al.) von Arx & van der Walt
*Issatchenkia orientalis Kudrjawzew
*
Kluyveromyces marxianus (Hansen) van der Walt*
Kluyveromyces thermotolerans (Phillippov) YarrowLeucoepor-idium scottii Fell et al.
Lipomyces starkeyi Lodder
&
Kreger-van Rij*Lodderomyces elongisporus (Recca
&
Mrak) van der Walt*Metschnikowia reukauif-i.i. Pitt & Miller
Nadsonia elongata Konokotina
Paahytispora transvaalensis (van der Walt) van der Walt
*Piahia anomala (Hansen) Kurtzman comb. nov.
*Piahia canadensis (Wickerham) Kurtzman comb.
Piahia carsonii Phaff & Knapp
~iahia etchellsii Kreger-van Rij
Piahia farinosa (Lindner) Hansen
*Piahia fermentans Lodder
*Piahia guilliermondii Wickerham
Piahia humboldtii Rodrigues de Miranda
&
Torok nov.Piahia jadinii (A. et R. Sartory, Weill et Meyer) Kurtzman comb. nov.
*Piahia membranaefaci.ene (Han se n) Hansen
Piahia silvicola (Wickerham) Kurtzman comb. nov.
*Piahia subpelliculosa Kurtzman sp. nov.
Rhodotorula aurantiaca (Saito) Lodder
Rhodotorula bogoriensis (Deinema) von Arx
&
WeijmanRhodotorula glutinis (Fresenius) Harrison
Rhodotorul-a minuta (Saito) Harrison
*RhodotoruZa muci.laginosa (Jorgenson) Harrison
Rhodatorul-a pal.l ida Lodder
*
Saccharomyces cerevisiae Meyen ex Hansen*Sacc-haromu ces "Saccharomucee "Saccharomucee
exiguus Reess ex Hansen
kluyveri Phaff et al. uni.eporue .Jêr ge ns on
*
Saccharomucodee l.udiai.qi.i. (Hansen) HansenSch.i.zoeaccharomucee [aponi.cue Yukawa & Maki
*Schizosacchaloomyces mal-ideverene Rankine & Fornachon
*Schizosaccharomyces octosporus Beijerinck
*Schizosaccharomyces pombe Lindner
Spoi-i diobo lue pararoeeue Fell & Tallman
SporidioboZus saZmonicoZor Fell & Tallman
Sporobol.omuces looseus Kluyver & van Niel
*ToruZaspora deZbrueckii (Lindner) Lindner
Torul.aepora ql.oboea '(Kloeker) van der Walt & Johannsen
Trichosporon beigeZii (Kuchenmeister & Rabenhorst) Vuillemin
Trichosporon puUuZans (Lindner) Diddens & Lodder
*T-lickerhc:mieUadomercqi.ae van der Walt & Liebenberg
h1i l-liope ie cal-iforni.ca (Lodder) von Arx
*WiZZiopsis saturnus (Kloeker) Zender
*Zygosaccharomyces baiZZii (Lindner) Guilliermond
Zygosaccharomyces bisporus Naganishi
Zygosaccharomyces [Lorent inue Castelli ex Kud r j awz ew
*Zygosaccharomyces mioroel-l-ipeo idee (Osterwalder) Yarrow
*
Zygosacchal°omyces rouxi-i (Boutroux) YarrowThe family Saccharomycetaceae comprises yeasts whi~h are associated with the wine industry and include species of the genera Saccharomyces~ Zygosaccharomyces~ Iorul aspooa, Ieeatohenk.ia, Kluyveromyces~ Lodderomqoee , Iripomucee, WickerhamieZZa and ~liZZiopsis.
This family also includes Il other genera which are separated on the basis of the shape, number and mode of ascospore formation (Kreger-van Rij, 1984) .
The family Endomycetaceae contains the wine associated species Dekkera~ Endomuoee , Hyphopichia~ Pielria and Haneenula; The latter two genera are now combined in Pielria Hansen emend. Kurtzman (1984). This family also contains 7 other genera and differs from the Dipodascaceae in producing a small and generally def LnIte number of ascospores (one to eight) in each ascus. The mycelium is composed of well-developed, typical hyphae. Asexual reproduction is by means of arthrospores or blastospores.
The family Dipodascaceae comprises of only one genus, Dipodascus,
characterized by elongated asci, borne singly from two mating hyphae and containing a large number of single cell hyaline ascospores surrounded by a sheath (van Arx, 1972).
The family Saccharomycodaceae comprises 4 genera of which species of
Saccharomycodes and Hanseniaspora are associated with wine. The genera of this family are characterized by bipolar budding (von Arx, 1972) and the formation of occasional pseudomycelium (Kreger-van Rij, 1984).
In the family Metchnikowiaceae only one of the five genera, namely
Metchnikowia is associated with the wine environment (Barnett et al.,
1983) • The genera of this family are generally characterized by non-septate hyphae and multilateral budding and were originally placed in
the family Spermophthoraceae by Lodder (1970) and Phaff et al. (1978).
Schizosaccharomyces, a wine associated yeast and only genus belonging to the family Schizosaccharomycetaceae, is mainly characterized by fission of the vegetative cells and the formation of true hyphae and arthrospores.
1.4 PROBLEMS ENCOUNTERED WITH THE CONVENTIONAL TAXONOMIC SYSTEM
The conventional system of species differentiation is based upon morphological, physiological, sexual and biochemical characteristics and (Barnett et al., 1983) has certain limitations. The ascospore shape of a species, long considered to be a constant character, proved to be variable when Wickerham and Burton (1954) reported the formation of both spherical and hat-shaped ascospores by strains of Pichia ohmeri.
simultaneously or at different growth stages (Gorin and Spencer, 1970). It is thus evident that different strains of the same species may differ in their ability to produce pseudohyphae, making this characteristic invalid as a differentiating criterion.
In the same way, Hansenula and Pichia were separated only by the ability or lack of ability to utilize nitrate. This phenotypic characteristic became invalid when Kurtzman (1984) combined these genera on the basis of results of DNA hybridization studies.
Another limitation is the problem of the instability of physiological characters of yeasts. Scheda and Yarrow (1966) observed enough variation in the fermentation and carbon assimilation patterns of a number of Saccharomyces species to cause difficulties in the assignment of yeast strains to different species. Another problem regarding the limitations of the conventional taxonomic system is the relation of the biochemical tests to the metabolism of the organisms. Originally it was not taken into consideration that various carbon sources are not necessarily assimilated independently but may be metabolized by common pathways. This suggests that yeasts that assimilate one carbon compound can also assimilate a structurally related one by the same metabolic pathway (Gorin and Spencer, 1970).
A problem that mainly concerns taxonomists in the wine making and brewing industries is the rapidly changing nomenclature of yeasts (Barnett,
1986). These changes are most inconvenient to these scientists who have to serve an industry where changes are not accepted easily.
An example of such changes is the "lumping" of different wine-making and brewery strains of Saccharomyces cereui eiae; Sacch, bauanue, Sacch . aarl eberqenei:e, Sacch . uvarum and Sacch , logos to one species, namely
Sacch, oereui eiae . This "lumping" pr oce as has obvious advances for the pure taxonomist, but the wine- and brewing taxonomist are required to distinguish between these yeasts (Hough et al., 1982).
For instance, Sacch . bayanus is known for its high alcohol tolerance, making it a most suitable yeast to reinoculate stuck fermentations (Rosini
et al., 1982) or secondary fermentations in champagne production.
Although some problems are encountered with the current system of classification, it must be recognized that phenotypic classification does serve its purpose and that not all characters utilized are unstable.
In the search for supplementary taxonomic characteristics, a number of new, more stable cri teria have been proposed which include comparison of the
ascospore surfaces by scanning electron microscopy (Kurtzman et aZ., 1972,
1975); serology (Campbell, 1971; Tsuchiya et aZ., 1974); proton magnetic resonance spectra of cell wall mannans (Gorin and Spencer, 1970); classification of the Coenzyme Q system (Yamada et aL, 1973, 1976, 1977) ; DNA hybridization studies (Kurtzman and Smiley, 1979; Kurtzman, 1984); electrophoretic enzyme patterns (Baptist and Kurtzman, 1976) and genome comparisons (Price et aZ., 1978).
1.5 PURPOSE OF THE RESEARCH
It has been found that chemical compounds, such as DNA, RNA enzyme proteins and mannose containing polysaccharides of yeast cell components, vary from species to species. This has led to a great interest in the chemotaxonamy of yeast cells, using as criteria chemical compounds as well as the
physical and immunological properties of macromolecules (Gorin and Spencer, 1970) .
1.5.1 The value of long-chain fatty acid composition in the taxonomy of wine- and related yeasts (See Chapter 2)
Lipid analyses are a well established criterion in bacterial taxonomy and have also provided suitable characteristics for the classificati.on and identification of many Coryneform and Actinomycete genera (Collins and Shah, 1984 and Athalye et aL, 1985).
Long-chain fatty acids are considered chemically as non-volatile acids ranging from C8 to C30 and can be divided into odd- and "even-chain fatty acids. The fatty acids of yeast lipids consist mainly of C16 and C18 acids, although a variety of other acids have been observed. A total of 33 acids, ranging from C8 to C22, including significant amounts of isopre-noid-type acids, have been detected in Saceh. cerevisiae (Rattray et aZ.,
1975). Welch and Burlingame (1973), however, found that C20 to C30 acids accounted for only 1 to 2% of the total fatty acid components. A minor poLythe no Ld acid component, as well as C8 to C12 acids were found in baker's yeast (Suomalainen and Keranen, 1968). All these fatty acids were and intracytoplasmic elements such as lipid particles (Rozijn and Tonino, lndge, 1968; Holley and Kidby, 1973; located in membranous structures
nuclei, vacuoles, mitochondria and 1964; Matile and Wiemken, 1967; Clausen et aL, 1974).
The pathways of fatty acid (Hunter and Rose, 1971) but
synthesis in yeasts have been documented the mechanisms of regulation are less
well-defined. The initial step ef the de novo biesynthesis ef fatty acids invelving acetyl-ceenzyrne A (CeA) carbexylase has been suggested as being under negative feedback centrel by long-chain fatty acyl CeA (Gill and Ratledge, 1973a, 1973b; Sumper, 1974). This is again influenced by the extent ef fatty acyl CeA incerperatien into. membraneeus systems (Sumper, 1974). It has been note d that the presence ef leng-chain fatty acids reduces the cellular centent ef acetyl CeA carbexylase (Kamirye and Numa, 1973) and may be significant in the ebserved inhibition (Mishina et al.,
1973) ef fatty acid biesynthesis by higher edd-chain fatty acids. The ability ef acetyl CeA synthetase to. ferm CeA esters frem shert-chain acids in Saceh. cerevisiae grewn aerebically, is inhibited markedly by leng-chain fatty acyl CeA (Satyanarayana and Klein, 1973). It was also. feund by these é1uthers that different preteins were invelved in the synthetase activity in aerobic as well as anaerebic cells. It has been shewn that Candida tropicalis, grewn on n-tetradecane, requires feur different types of acyl CoA synthetase, each having specific substrate requirements
intracellular lecatien.
Studies en yeasts grewing en different n-alkanes shewed that two mechanisms occur in fatty acid synthesis (Mishina et al., 1973). Odd-chain
fatty acids eriginated frem the elengatien ef edd-chain fatty acid p re cur s o r s and even-chain fatty acids by de novo synthesis. A similar elongatien system was recegnized (arme et al., 1972) in a mutant ef Saceh. cerevisiae that ceuld synthesize higher acids frem C13 to. C17 acid
supplements and ceuld net perferm de novo synthesis. Erwin (1973) discussed the fermatien ef unsaturated fatty acids which are influenced especially by the presence or absence ef exygen (Rattray et al., 1975).
The cellular lipid centent and -cempesitien is influenced by numereus factors, i. e. the grewth cycle (Daws on and Craig, 1966; McMurreugh and Rese, 1971); sperulation (Illingwerth et al., 1973); nutrients such as nitrogen and pho s ph oru s (Ratledge, 1968; .Johns on et al., 1972); grewth factors such as inositel (Lewin, 1965; Jehnsten and Paltauf, 1970; Paltauf and Jehnston, 1972), vitamin B6 (Haskell and Snell, 1965) and biotin (Suemalainen and Ker ánen , 1968): s od i um chleride (Combs et al.,
1968); cheline (Palmer, 1971); benzopyrene (Ba r aud et al., 1973); propanediel (Suzuki and Hasegawa, 1974): exygen (Hunter and Rese, 1972; Kovác et al., 1967); temperature (Hunter and Rese, 1972; Kates and Paradis, 1973) and pH (Rattray et al., 1975).
Abel et al. (1963) was the first to. empley gas-liquid chromategraphy fer the c La s s Lf LcatLon ef bacteria en the b a s Ls ef their cellular fatty acid cempositien. Since then a number of studies en the cellular fatty acid and
composition and the taxonomic relationship h~ve been reported (Shaw, 1974; Kaneko et al., 1976;
1979; Chen , 1981;
alo, 1985; Kock et
It was found that the fatty acid composition of microorganisms varies betwe en species of a genus and also with culture age, medium composition
Hossack and Spencer-Martins, 1978; Nishimura et al. ,
Moss et al. , 1982; Athalye et al. , 1985; Cottrell et al. , 1985; Kock et al. , 1986).
and growth temperature (Deinema, 1961; Merdinger and Devine, 1965; McMurrough and Rose, 1967; Brown and Rose, 1969; Hunter and Rose, 1972; Drucker and Veazey, 1977; Tornabene, 1985 and Viljoen et al., 1986). It
is therefore of utmost importance to use standardized conditions for growth in order to obtain reproducible results in a taxonomic study.
In this thesis the long-chain fatty acid compositions of yeasts associated with wine environments were investigated as an aid in identifi-cation and classification.
1.5.2 The use of volatile metabolites in the identification of yeasts associated with wine
Many of the volatile constituents associated with the bouquet and flavour of wines are produced by yeasts. The isolation and identification of these yeast metabolites (mainly carbonyl compounds, alcohols and fatty acid esters) have been studied in considerable detail using gaschromatography with subsequent mass-spectrometry (Hardy and Ramsh aw , 1970; Killian and Ough, 1979; Schreier et al , , 1980).
It was found that the volatile metabolites produced by yeasts vary between different yeast species and strains (Wenzel, 1966; Di Stefano et al., 1981; Soles et al., 1982) and could therefore have taxonomic
implications.
A method to determine these volatile metabolites on a qualitative-, as well as quantitative basis will have notable advantages over the organoleptic tests (Lodder,
classification system.
In this investigation, the use of different volatile metabolites in the 1970) used in the present conventional
identification of some wine yeasts was investigated in a preliminary study.
REFERENCES
Abel, K., De Schmertzing, H. and Peterson, J.I. (1963) Classification of micro-organisms by analysis of chemical composition. I. Feasibility of utilizing gas chromatography. J.·Bacteriol. 85: 1039-1044.
Alexopoulus, C.J. and Mims, C.W. (1979) Introductory Mycology. Third edition. John Wiley and Sons, Inc., New York.
Athalye, M., Noble, W.C. and Minnikin, D.E. (1985) Analysis of cellular fatty acids by gas chromatography as a tool in the identification of medically important coryneform bacteria. J. Appl. Bacterial. 58: 507-512.
Baptist, J.N. and Kurtzman, C.P. (1976) Comparative enzyme patterns in
CY'yptococcus laurent-i-i and its taxonomic varieties.
1195-1203.
Mycologia 68:
Baraud, J., Maurice, A., Napias, C. and Velours, J. (1973) Actions de carbures concéro genes sur les lipides des membranes protoplastiques et des mitochondries du Saceharomucee oerevi.eiae, .Biochim. Biophys. Acta 296: 481-492.
Barnett, J.A., Payne, R.W. and Yarrow, D. (1983) Yeasts: Characteristics and identification, Cambridge University Press, New York.
Barn~tt, J.A. (1986) The stability of biological nomenclature: yeasts. Nature 322: 599.
Broek, T.D. (1961) Milestones in Microbiology. Prentice-Hall, Englewood Cliffs, N.J.
Brown, C.M. and Rose, A.H. (969) Fatty-acid composition of Candida
utilis as affected by growth temperature and dissolved-oxygen tension. J. Bacteriol. 99: 371-378.
Campbell, I. (1971) Comparison of serological and physiological
classification of the genus Saccharomucee-~, J. Gen'. Hicrobiol. 63: 189-198.
Chen, EC-H. (1981) Fatty acid profiles of some cultured and wild yeasts in brewery. J. ASBC 39(4): 117-124.
Clausen, M.K., Christiansen, K., Jensen, P.K. and Behnke,
o.
(1974)Isolation of lipid particles from baker's yeast.
176-179.
FEBS Lett. 43:
Collins, M.D. and Shah, H.N. (1984) F5tty acid menaquinone and polar
lipid composition of Rothia dent.ooar-ioea- Arch. Microbio.!. 137 (3): 247-249.
Combs, T.J., Guarneri, J.J. and Pisano, M.A. (1968) The effect of sodium chloride on the lipid content and fatty acid composition of Candida albicans. Mycologia 60: 1230-1239.
Cottrell, M., Kock, J.L.F., Lategan, P.M., Botes, P.J. and Britz, T.J. (1985) The long-chain fatty acid composition of species representing the genus Kluyveromyces. FEMS Microbiol. Lett. 30: 373-376.
Dawson, P.S.S. and Craig, B.M. (1966) Lipids of Candida utilis: changes with growth. Can. J. Microbiol. 12: 775-785.
Deinema, M.H. (1961) Intra- and extracellular lipid production by Ph.D Thesis.
yeasts. Laboratory of Microbiology, Agriculture University, Wageningen, The Netherlands.
Diddens, H.A. and Lodder, J. (1942) Die anaskosporogenen Hefen, II Halfte. North-Holland Publ. Co., Amsterdam.
Di Stefano, R., Ciolfi, G. and Delfini, C. (1981) Composti volatili prodotti dai lievitti. Riv. Vitic. Enol. 34(8): 342-355.
Drucker, D.B. and Veazey, F.J. (1977) Fatty acid fingerprints of
Streptococcus mutans NCTC 10832 grown at various temperatures. Appl. Environ. Microbiol. 33(2): 221-226.
Flegel, T.W. (1977) Let's call a yeast a yeast. Can. J. Microbiol. 23: 945-946.
Gill, C.O. and Ratledge, C. (1973a) Inhibition of glucose assimilation and transport by n-decane and other n-alkanes in Candida 107. J. Gen. Microbiol. 75: 11-22.
Gill, C.O. and Ratledge, C. (1973b) Regulation of de novo fatty acid biosynthesis in the n-alkane utilizing yeast Candida 107. Microbiol. 78: 337-347.
Gorin, P.A.J. and Spencer, J.F.T. (1970) Proton magnetic resonance
J. Gen.
spectroscopy - an aid in identification and chemotaxonomy of yeasts. Adv. Appl. Microbiol. 13: 25-89.
Guilliermond, A. (1920) The yeasts (trans. F.W. Tanner). John Wiley, New York.
Guillermond, A. (1928) Clef dichotomique pour la determination des levures. Librairie le Francois, Paris.
Hardy, P.J. and Ramshaw, E.H. (1970) Analysis of minor volatile constituents of wine. J. Sci. Food Agric. 21: 39-41.
Haskell, B.E. and Snell, E.E. (1965) Effect of vitamin B6 deficiency on the composition of yeast lipids. Arch. Biochem. Biophys. 112: 494-555.
Heard, G.M. and Fleet, G.H. (1985) Growth of natural yeast flora during the fermentation of inoculated wines. Appl. Environ. Microbiol.
50 (3): 727-728.
Holley, R.A. and Kidby, D.K. (1973) Role of vacuoles and vesicles in extracellular enzyme secretions from yeast. Can. J. Microbiol. 19:
113-117.
Hossack, J.A. and Spencer-Martins, I. (1978) Lipid composition of the yeast Li.pomucee kononenkoae . Eur. J. Appl. Microbiol. Biotechnol. 5:
273-278.
Hough, J.S., Briggs, D.E., Stevens, .R. and Young, T.W. (1982) Malting and Brewing Science. Vol. 2: Hopped ~70rt and Beer. Second Edition. Chapman and Hall Ltd. London and New York.
Hunter, K. and Rose, A.H. (1971) Yeast lipids and membranes.
In
The yeasts 2 (Ed. Rose, A.H. and Harrison, J.S.). Academic Press. London and New York.Hunter, K. and Rose, A.H. (1972) Lipid composition of Saccharomyces
cereui.eiae as inf luenced by growth tempera tur e , Biochim. Biophys. Acta 260: 639-653.
Illingworth, R.F., Rose, A.H. and Beckett, A. (1973) Changes in the lipid composition and fine structure of Saccharomyces cerevisiae during ascus formation. J. Bacteriol. 113: 373-386.
Indge, K.J. (1968) The isolation and properties of the yeast cell vacuole. J. Gen. Microbiol. 51: 441-446.
Johnson, B., Nelson, S.J. and Brown, C.M .. (1972) Influence of glucose concentration on the physiology and lipid composition of some yeasts. Antonie van Leeuwenhoek 38: 129-136.
Johnston, J.M. and Paltauf, F. (1970) Lipid metabolism in
inositol-deficient yeast, Saccharomyces carl.eberqenei:e , II Incorporation of labelled precursors into lipids by whole cells and activities of some enzymes involved in lipid formation. Biochim. Biophys. Acta 218: 431-440.
K.W.V. Annual Report (1985) Statistical report of the Kooperatiewe Wijnbouwers Vereniging, Suider-Paarl, Republic of South Africa. Kamiryo, T. and Numa, N. (1973) Reduction of the acetyl Coenzyme A
carboxylase content of Saccharomyces cereui eiae by exogenous fatty acids. FEBS Lett. 38: 29-32.
Kaneko,~ H., Hosohara, M., Tanaka, M. and Itoh, T. (1976) Lipid composition of 30 species of yeasts. Lipids 11: 837-844.
Kates , M. and Paradis, H. (1973) Phospholipid desaturation in
Candida
lipolytica
as a function of temperature and growth. Can. J. Biochem. 51: 184-197.Killian, E. and Ough, C.S. (1979) Fermentation esters - formation and retention as affected by fermentation temperature.
Vitic. 30(4): 301-305.
Kock, J.L.F., Lategan, P.M., Botes, P.J. and Viljoen, B.C. (1985)
Am. J. Enol.
Developing a rapid statistical identification process for different yeast species. J. Microbiol. Methods 4: 147-154.
Kovác, L., Subik, J., Russ, G. and Kollár, K. (1967) On the relationship between respiratory activity and lipid composition of the yeast cells. Biochim. Biophys. Acta 144: 94-101.
Kreger-van Rij, N.J.W. (1969) Taxonomy and systematics of yeasts.
In
The yeasts (Ed. Rose, A.H. and Harrison, J.S.), Academic Press. London and New York.Kreger-van Rij, N.J.W. (1984) The yeasts: a taxonomic study. Third revised and enlarged edition. Elsevier Science Publishers. B.V. Amsterdam.
Kudrjawzew, W.I~ (1960) Die systematik der Hefen. Akademie-Verlag, Berlin (originally published in Moscow in Russian, 1954).
Kurtzman, C.P. (1984) Synonomy of the yeast genera
Hansenula
andPichia
demonBtrated through comparisons of deoxyribonucleic acid relatedness. Antonie van Leeuwenhoek 50: 209-217.
Kurtzman, C.P. and Smiley, M.J. (1979) Taxonomy of
Pichia carsonii
and its synonymsPichia vini
andP. vini
var.mel ibioei.:
comparison by DNA reassociation. Mycologia 71: 658-662.Kurtzman, C.P., Smiley, M.J. and Baker, F.L. (1975) Scanning
electron-microscopy of ascospores of
Debaryomyces
andSaccharomyces.
Mycopathol. Mycol. Appl. 55: 29-34~
Lewin, L.M. (1965) Effects of meso-inositol deficiency on some important biological and chemical characteristici of yeast. J. Gen. Microbiol. 41: 215-224.
Lodder, J. (1934) Die anaskosporogenen Hefen, I. Halfte. Verh. Kon. Ned. Akad. Wet., Afd. Natuurk., Sect. II, 32: 1-56.
Lodder, J. (1970) The yeasts: a taxonomic study. (Ed. Lodder, J.)
North-Holland Publishing Co., Amsterdam.
Lodder, J. snd Kreger-van Rij, N.J.W. (1952) The yeasts - A taxonomie Study. North-Holland Publishing Co., Amsterdam.
Lodder, J., Slooff, W.Ch. and Kreger-van Rij, N.J.W. (1958) The
classification of yeasts. In The chemistry and biology of yeasts. (Ed. Cook, A.H.) Academic Press, New York.
Matile, P. and Wiemken, A. (1967) The vacuole as the lysosome of the yeast cell. Arch. Mikrobiol. 56: 148-155.
McMurrough, I. and Rose, A.H. (1967) Effect of growth rate and substrate limitation on the composition and structure of the cell wall of
Saccharomucee oereui-eiae . Biochem. J. 105: 189-203.
McMurrough, I. and Rose, A.H. (1971) Effects of temperature variation on the fatty acid composition of a psychrophilic Candida species. J. Bacteriol. 107: 753-758.
Merdinger, E. and Devine, E.M. Jr. (1965) Lipids of Debaryomyces hansenii. J. Bacteriol. 89(6): 1488-1493.
Mishina, M., Yanagawa, S., Tanaka, A. and Fukui, S. (1973) Effects of chain-length of alkane substrate on fatty acid composition and biosynthetic pathway in some Candida yeasts. Agric. BioI. Chemo 37: 863-870.
Moss, C.W., Shinoda, T. and Samuels, J.W. (1982) Determination of
cellular fatty acid compositions of various yeasts by gas-liquid chromatography. J. Clin. Microbiol. 16(6): 1073-1079.
Nishimura, Y., Yamamoto, H. and Iizuba, H. (1979) Taxonomical studies of
Acinetobacter species - cellular fatty acid composition. Zeitschrift fur Algemeine Mikrobiologie 19(4): 307-308.
Orme, T.W., Mclntyre, J., Lynen, F., Klihn, L. and Schweizer, E. (1972) Fatty-acid elongation in a mutant of Saccharomyces oereirieiae
deficient in fatty acid synthetase. Eur. J. Biochem. 24: 407-415. Palmer, F.B.St.C. (1971) The extraction of acidic phospholipids ~n
organic solvent mixtures containing water. Biochim. Biophys. Acta 231: 134-144.
Paltauf, F. and Johnston, J.M. (1972) Lipid metabolism in
inositol-deficient yeast, Saccharomyces carlsbergensis. 1. Influence of different carbon sources on the lipid composition of deficient cells. Biochim. Biophys. Acta 218: 424-430.
Phaff, H.J., Miller, M.W. and Mrak, E.M. (1966) The life of yeasts.
Their nature, activity, ecology and relation to mankind. Harvard University Press, Cambridge and Massachusetts.
Phaff, H.J., Miller, M.W. and Mrak, E.M. (1978) The life of yeasts. Second edition, revised and enlarged.
Cambridge, Massachusetts and London.
Price, C.W., Fuson, G.B. and Phaff, H.J. (1978) Genome comparison in yeast systematics: delimitation of species within the genera
Schiaann-iomucee , Saccharomuces , Debaruomuaes and Pichia. Micr. Rev. 42: 161-193.
Rad Ier , F. (1973). Bedeutung und mdg lLchke Lten der verwendung von
reinkulturen von hefen bei der weinbereitung. Weinberg KeIler 29: 3. Rankine, B.C. (1968) The importance of yeasts in determining the
composition and quality of wines. Vitis 7: 22-49.
Ratledge, C. (1968) Production of fatty acids and lipids by a Candida sp. growing on a fraction of n-alkanes predominating in tridecane. Biotechnol. Bioeng. 10: 511-533.
Rattray, J.B.M., Schibeci, A. and Kidby, D.K. (1975) Lipids of yeasts. Bacteriol. Rev. 39(3): 197-231.
Rosini, G., Federici, F., Vaughan, A.E. and Martini, A. (1982)
Systematics of the species of the yeast genus Saccharomyces associated with the fermentation industry. Eur. J. Appl. Microbiol. Biotechnol.
15: 188-193.
Rozijn, Th.H. and Tonino, G.J.M. (1964) Studies on yeast nucleas. I. The isolation of nuclei. Biochim. Biophys. Acta 91: 105-112.
t
Satyanarayana, T. and Klein, H.P. (1973) Studies on acetyl-coenzyrne A synthetase of yeast: inhibitio~ by long-chain acyl-coenzyme A esters. J. Bacteriol. 115: 600-606.
Scheda, R. and'Yarrow, D. (1966) The instability of physiological properties used as criteria in the taxonomy bf yeasts. Mikrobiol. 55: 209-225.
Schreier, P., Drawert, F. and Abraham, K.O. (1980) Identification and Arch.
determination of volatile constituents in Burgundy Pinot Noir wines. Lebensm. Wiss. Technol. 13: 318-323.
Shaw, N. (1974) Lipid composition as a guide to the classification of bacteria. Adv. Appl. Microbiol. 17: 63-108.
Soles, R.M., Ough, C.S. and Kunkee, R.E. (1982) Ester concentration
differences in wine fermented by various species and strains of yeasts. Am. J. Enol. Vitic. 33(2): 94-99.
Stelling-Dekker, N.H. (1931) Di.e Hefesammlung des "Centraal-Bureau voor Schimmelcultures" . Beitrage zu einer Monographie der Hefearten. I. Die sporogenen Hefen. Dr ukkerLj Holland, Ams terdam .
Sumper, M. (1974) Control of fatty-acid biosynthesis by long-chain acyl CoAls and lipi.d membranes. Eur. J. Biochem. 49: 469-475.
Suomalainen, H. and Keranen, A.J.A. (1968) The fatty acid composition of baker's and brewer's yeast. Chemo Phys. Lipids 2: 296-315.
Suzuki, T. and Hasegawa, K. (1974) Diol lipids in the phospholipid
fraction of Lipomyces starkeyi grown in the medium containing l,2-propanediol. Agric. BioI. Chemo 38: 613-620.
Tornabene, T.G. (1985) Lipid analysis and relationship to chemotaxonomy. Methods in Microbiology 18: 209-234.
Tromp, A. (1980) The influence of various factors on fermentation and wine quality with special reference to lagging fermentation. Short course in enology and cellar planning, November 1980, V.O.R.I., Private Bag X5026, 7600 Stellenbosch, Republic of South Africa.
Tsuchiya, T., Fukazawa, Y., Taguchf, M., Nakase, T. and Shinoda, T. (1974) Serological aspects on yeast classification. Mycopathol. Mycol. Appl. 53: 77-91.
Van der Walt, J.P. and von Arx, J.A. (1985) Use of ultrastructural chatacters in yeast systematics. S.A. J. of Science 81: 156-159. Von Arx, J.A. (1972) On Endomyces~ Endomycopsis and yeast-like fungi.
Antonie van Leeuwenhoek 38: 289-309.
Von Arx, J.A. (1981) The genera of fungi sporulating in pure culture. Third revised edition. J. Cramer. Braunschweig.
Viljoen, B.C., Kock, J.L.F. and Lategan. P.M. (1986) The influence of culture age ón the cellular fatty acid composition of four selected yeasts. J. Gen. Microbiol. 132(7): 1895-1898.
Welch, J.W. and Burlingame, A.L. (1973) Very long-chain fatty acids in yeast. J. Bacteriol. 115: 464-466.
WenzeI, K.O. (1966) Gaschromatografiese analise vir geurstowwe in wyn. M.Sc. thesis. University of Stellenbosch, Stellenbosch. Republic of South Africa.
Wickerham, L.J. (1951) Taxonomy of yeasts. U.S. Department of Agriculture, Techn. Bull. No. 1029: 1-56.
Wickerham, L.J. and Burton, K.A. (1954) A clarification of the
relationships of Candida guilliermondii to other yeasts by a study of their mating types. J. Bacteriol. 68: 594-597.
Wickerham, L.J. and Burton, K.A. (1956) Hybridization studies involving
Saccharomucee laet-ieand Zygosaccharomyces ashbyii. J. Bac teriol. 71: 290-295.
Yamada, Y., Ari~oto, M. and Kondo, K. (1973) Coenzyme ~ system in the classification of the ascosporogenous yeast genus Schi~osaccharomyces
and the yeast-like genus Endomuces , J. Gen. Appl. Microbiol. 19: 353-358.
Yamada, Y., Arimoto, M. and Kondo, K. classification of apiculate
Saccharomucodee , Hanseni aepora,
Appl. Microbiol. 22: 293-299.
Yamada, Y., Arimoto, M. and Kondo, K. (1977) Coenzyme Q system in the (1976) Coenzym~ Q system in the yeast in the genera Nadsonia~ Kl.oeckera and Wickel·hamia. J. Gen.
classification of some ascosporogenous yeast genera in the families Saccharomycetaceae and Spermophthoraceae. Antonie van Leeuwenhoek 43: 65-72.
CHAPTER 2
THE VALUE OF CELLULAR LONG-CHAIN FATTY ACID COMPOSITION IN THE TAXONOMY OF
WINE-AND RELATED YEASTS
'Condensed versions of different sections in this chapter have been accepted for publication:
Tredoux H.G., J.L.F. Kock, P.M. Lategan and H.B. Muller. ,1987. A rapid i.dentification technique to differentiate between Saccharomyces cerevisiae
strains and other yeast species in the wine industry. Am. J. Enol. Vitic.
Tredoux H.G., J.L.F., Kock and P.M. Lategan. 1987. The use of cellular. long-chain fatty acid composition in the identification of some yeasts associated with the wine industry. Syst. Appl. Microbiol.
CHAPTER 2
THE VALUE OF CELLULAR LONG-CHAIN FATTY ACID COMPOSITION IN THE TAXONOMY OF
WINE-AND RELATED YEASTS
ABSTRACT
The cellular long-chain fatty acid compositions of 103 yeast strains representing 38 species associated with the wine industry were determined by gas-liquid chromatography.
It was possible to differentiate between all the species examined except
Scni zoeaccharomucee mal idevorane and S. pombe which had a similar fatty acid composition, as well as between some strains within Saccharomyces cerevie iae and within other species examined. Of importance to the wine industry is the fact that sacch . cerevi eiae had an unique fatty acid profile.
This method resulted in identification within three days which- compares favourably with the seven to ten days and longer required for conventional methods.
A general correlation was found between the presence of linoleic- and 1inolenic acid and the complexity of cell differentiation. Two phylogene-tic lines were obtained in Kluyveromyces by comparing long-chain fatty acid compositon, genetic recombination, pseudomycelium formation and carbon source and ethylamine utilization. These lines correspond with the proposed conventional phylogenetic scheme for Kluyve2'omyces. A correlation was found in this genus between the long-chain fatty acid composition and the ability to form pseudomycelium, to utilize carbon sources as well as ethylamine and the ability to hybridise. In the genus Saccharomyces, a similar correlation was found- between the presence of linoleic- and linolenic acid and the ability to utilize a large number of carbon sources. A developmental line was found which corresponds with a sequential acquirement of the ability to utilize carbon sources, the ability to form pseudomycelium, loss of resistance to cycloheximide as well as the acquirement of linoleic- and linolenic acid.
2.1 INTRODUCTION
The present yeast classification system aims to assign yeast strains to species and genera on the basis of their morphological characteristics, sexual reproduction
(Phaff et al., 1978;
and certain physiological Barnett et al., 1983).
and biochemical features
Certain difficulties are encountered when the above mentioned criteria are applied. For instance, the genera Candida and Torulopsis are separated only on the ability of the former to produce pseudohyphae (Lodder et al.,
1958) • It was, however, observed that these species can produce two or more types of pseudomycelium simultaneously or at different stages of growth (Van Uden and Buckley, 1970). Wickerham and Burton (1954) reported the presence of both spherical and hat-shaped ascospores in strains of
pichia ohmeri at a time when it was thought that spore shape was a constant eh ar ac teristic of a species. Stelling-Dekker (1931) proposed that
Hansenula and Pichia be separated primarily on their ability to assimilate nitrate as a sole source of nitrogen. Since this criterion cannot always be successfully applied,
disappear and one generic (1966) observed enough assimilation patterns of
difficulties in the assignment of their yeast strains to specific species. Since some morphological differences are unreliable for taxonomy, and biochemical and physiological criteria are also sometimes variable, new the difference between these two genera could
name wil L have to be used. Scheda and Yarrow variability in the fermentation and carbon
a number of Saccharomyces species to cause
criteria, which are more stable, should be examined. These include a number of rnacromolecular comparisons such as proton magnetic resonance
(Gorin and Spenc~r, 1970), s~rology (Campbell, 1971), classification of the isoprenoid quinones in the electron transport system (Yamada et al., 1977),
electrophoretic patterns of isozymes and enzymes (Baptist and Kurtzman, 1976), DNA hybridization (Kurtzman and Smiley, 1979; Kurtzman, 1984), genome comparisons (Price et al., 1978) and scanning electron microscopy
(Kurtzman et al., 1975). Taxonomic schemes based on Adansonian analyses of the traditional, phenotypic characters have also been proposed (Campbell, 1974).
Since the introduction of lipid analyses by gas-liquid chromatography (GLC) , various 'investigations concerning the identification and classification of bacteria and fungi (Shaw, 1974; Miura et al., 1983;
Collins and Shah, 1984; Athalye et al., 1985) and yeasts were undertaken (Moss and Dees, 1975; Rattray et al., 1975; Chen, 1981; Nos s et al.,
composition (Rattray et al., 1975) it is not possible to utilize the results of the above mentioned studies for taxonomic purposes.
Recent work in this field resulted in a reproducible technique for the cultivation of yeasts and analysis of cellular fatty acids (Cottrell et al., 1985; Kock et al., 1985; Kock et al., 1986; Viljoen et al., 1986).
In this study, the fatty acid compositions of yeast strains representing 38 species of 21 genera associated with the wine industry were determined by using the method of Kock et al. (1985).
The application of long-chain fatty acid compositions in the identifica-tion and classification of wine-associated and related yeasts is discussed as follows:
2.1.1 The use of cellular long-chain fatty acid composition in the identification of yeasts associated with the wine industry. The value of cellular long-chain fatty acid composition in the taxonomy of wine- and related yeasts. This includes:
2.1. 2
2.1.2.1 The relation between long-chain fatty acid composition and the degree of mycelium formation.
The value of long-chain fatty acid composition in the phylogeny of the genus Kluyveromyces.
The value of long-chain fatty acid composition in the taxonomy of the genus Saccharomyces.
2.1.2.2
2.1.2.3
2.2 MATERIALS AND METHODS
Strains: One hundred and three strains comprising 38 species were obtained from the Centraalbureau voor Schimmelcultures , Yeast division, Delft, The Netherlands (CBS); Professor J.P. van der Walt, Council for Scientific and Industrial Research, Pretoria, South Africa (CSIR-Y); the Viticultural and Oenological Research Institute, Stellenbosch, South Africa
(N) and the American Type Culture Collection (ATCC) (Table 1).,
CuI tiva tion of strains: The inoculum was prepared from stock cu Itures maintained on YM (Hickerham, 1951) slants. These were then cultured in triplicate for 16 h at 30°C on a rotary shaker at 160 rpm (throw = 50 mm) in 150 mE Erlenmeyer flasks. Each flask contained 40 mE of medium, consisting of 80 g/E glucose (Merc-k) and 6.7 g/E yeast nitrogen base (YNB) (Difco) .
Ten m£ quantities of the precultured strains (Klett = 200) were then inoculated into 400
mR
of glucose YNB liquid medium in 1R
conical flasks and cultured for 2 days under the conditions described. Since a constant and reproducible fatty acid comp0sition was found in the yeast cells during stationary phase by Vilj oen et al. (1986), the cells we re then harvested during this phase by centrifugation at 8000 x g for 5 min at 4°C. The sediment was washed three times with cold 0.85% saline solution and lyophilized.Extraction of the fatty acids and preparation of methyl esters: Fatty acids were extracted from 0.12 g lyophilized yeast cells suspended in 5
mR
of 15% KOH in 50% methanol. The suspension (in sealed screw-capped test tubes) was heated in a boiling waterbath for 1 h, the saponified material cooled to room temperature and the pH adjusted to 2.0 with 6N HCl. The free fatty acids were then methylated with 3
mR
of 20% borontrifluoride in methanol (Merck, Darmstadt) in a boiling waterbath for 15 min while shaking. Again the suspension was cooled to room temperature and 0.25mR
of a saturated NaCl solution was added. The methyl esters were then extracted by vigorous shaking with three 6
mR
portions of a 1:4 chloroform-hexane mixture. The chloroform-hexane mixtures were recovered by centrifugation at approximately 500 rpm for 3 minutes. The solvent mixture was evaporated by means of nitrogen gas and the dried methyl esterfraction dissolved in 1.8
mR
hexane.Separation of fatty acids by gas-liquid chromatography: The methyl esters of the total fatty acids were analysed by GLC on a Hewlett Packard model 5830A gas chromatograph equipped with dual flame-ionization detectors. Identification of the esters was based on the comparison of retention times \.".ithknown standards of C14: 0 (myristic acid), C14: 1 (myristoleic acid), C16:0 (palmitic acid), C16:1 (palmitoleic acid), C18:0 (stearic acid), C18:1 (oleic acid), C18:2 (linoleic acid) and C18:3 (linolenic acid) (Serva, Heidelberg, Germany). All analyses were carried out using glass columns (4 mm I.D. x 1.5 m) packed with 5% diethyleneglycol succinate on Chromosorb W (80-100 mesh). The flow rate of the carrier gas
(nitrogen) was 30 cm3 min -1 at a column temperature of 160°C. Relative
amounts of given fatty aci ds were calculated from their respective peak areas.
Pseudomycelium and mycelium formation: This morphological characterstic was determined using the Dalmau plate technique as described by Van der Wal t (Lodder, 1970).
2.3 RESULTS AND DISCUSSION
2.3.1 The use of cellular long-chain fatty acid composition in the identification of yeasts associated with the wine industry
It was possible to differentiate between these organisms within 3 days which is a marked improvement on the usual 7 to
la
days and longer required with the conventional methods of Barnett et al. (1983). The high resolution, sensitivity and speed of this identification system can also complement the physiological, morphological and serological techriiques conventionally used for yeast differentiation.The results obtained were reproducible when the strains were grown under standard conditions. The standard deviation for triplicates was between 2% and 7%. The 38 species are characterized by the presence (or absence) of varying amounts of myristic acid (CI4:0), myristoleic acid (CI4:1), palmitic acid (C16:0), palmitoleie acid (CI6: 1), stearic acid (C18:0), oleic acid (CI8:1), linoleic acid (C18:2) and linolenic acid (C18:3) calculated as relative percentages.
It was possible to distinguish between most of these species as is shown in Table 1. The fatty acids used for differentiation are also highlighted. A detailed discussion concerning the differentiation of these species is now presented.
Division into groups
The strains can be divided into seven major groups according to their fatty acid content (Table 1). Groups I to IV are characterized by the absence of linoleic acid (C18:2) and linolenic acid (C18:3) while strains in group V contain linoleic acid (C18:2) and strains in groups VI and VII contain both linoleic acid (C18:2) and linolenic acid (C18:3).
Group I, characterized by the presence of palmitoleic acid (C16:1) and oleic acid (C18: 1) as major fatty acids and the absence of linoleic acid (C18:2) and linolenic acid (C18: 3), comprises the 41 strains of
Saccharomqcee cereui.ei.ae ,
Group II: The strains representing this group, have fatty acid compositions similar to group I, but are differentiated by a significantly
(P < 0.05) lower mean percentage palmitoleie acid (C16: 1) and a significantly higher mean. percentage oleic acid (C18:1) (P < 0.05) compared
to groups I, III and IV (student's t-test). This group includes the strains of Schizosaccharomyes malidevorans~ S. octosporus and S. pombe.
Group III includes strains of Hanseniaspora uvarum~ H. valbyen8is~ Saccharomyces exi.quue , Sacch . unisporus and Saacharomucodee ludwigii and
they have higher mean percentages of palmitoleic acid (C16:1) and significantly smaller mean percentage of oleic acid (CIB: 1), compared to groups I, II and IV. These strains contain no linoleic acid (C18: 2) or linolenic acid (C18:3).
Group IV comprises only ~!ickerhamie l-La domercqiae which contains a lower percentage palmitoleic acid (C16:1) compared to groups I and III and a lower percentage oleic acid (C18:1) compared to group II.
Group V is characterized by the presence of linoleic acid (C18: 2) and the absence of linolenic acid (C18: 3) . This group comprises strains of
Endomyces fibuliger3 Pichia etchellsii3 Torulaspora delbrueckii~
Zygosaccharomyces microellipsoides and Z. rouxii.
Group VI: The strains representing this group contain linolenic acid (C18:3) which is not present in the previous groups. This group includes strains of Candida tenui.e , Cryptococcus al.bi dus , Debaruomuces haneeni.i , Filobasidium capsuligenum~ Hansenula anomala~ H. canadensis3 H.
eubpel.l-ioul oea, Netchrcikoun.a reukauf'[i.i , Pichia qui.l.l iermondi.i: and P. membranae faci.ene ,
Group VII includes the strains of Candida albicans~ C. Y'ugosa~ C. e teatol.ut-ica, Debaruomucee haneeni:i , Hyphopichia bur toni.i , Issatchenkia te ri-icol.a, KluyveY'omyces marxianus , K. thermotol.evane , Lodderomucee e Lonqiepomcs , Pichia fel'lnentans3 Rhodotorul.a muci.l.aqinoea, Saccharomyces
kluyveri3 Williopsis saturnus and Zygosaccharomyces baillii. These strains
contain lower mean percentages linoleic acic (C18:2) compared to those in group VI and also contain linolenic acid (CIS:3).
Subdivision within groups
Group I: This group of 41 Saccharomyces cerevi.ei.ae strains can be divided into five different subgroups according to their oleic acid (C18:1) content (Table 1).
a lower mean percentage palmitic acid (C16:0) and a higher mean percentage palmitoleic acid (C16:l) compared to strain N34.
Subgroup b. Strain CSIR-Y2 contains the lowest and strain N18 contains the highest mean percentage palmitic acid (C16:0). Strain N13 contains the highest, while N29 contains the lowest mean percentage palmitoleic acid
(C16:l). Differentiation b e tv..een the remainder of the strains was not attempted.
Subgroup c. The fatty acid compositions of strains CBS 1907, NI, N3, N4, N7, N8, N9, N14, N19, N23, N25, N32 and N4l are similar and they contain a lower mean percentage palmitic acid (C16:0) compared to the other s trains in the subgroup. Further subdivision may therefore be possible, but was not attempted.
Subgroup d. In this subgroup, .strains N26 and N27 are similar and may be distinghuished from the other strains on the basis of their lower palmitoleic acid (C16: 1) content. Strain N5 contains the highest mean percentage palmitic acid (C16:0), while strain N31 contains the highest mean percentage palmitoleic acid (C16:l) in the group.
Group II. In this group, the strains of Schi.zoeaccharomucee mal.i.devorane and S. pombe contain similar fatty acid compositions. These
strains contain lower mean percentages palmitic acid (C16: 0) and higher mean percentages palmitoleic acid (C16:1) and oleic acid (C18:1) than strains of S. octosporus.
Group III. The Saccharomyces exiguus strain contains a lower mean percentage of palmitoleic acid (C16: 1) and also a higher mean percentage
stearic acid (C18:0) compared to the other strains. The strains of
Haneeni.aepora uvarum and H. valbyensis contain the highest mean percentage palmitoleic acid (C16: 1) and also the lowest mean percentage oleic acid (C18:1). H. uval'Um contains a lower mean percentage palmitoleic acid (C16: 1) compared to H. valbyensis. Saccharomyces unisporus contains the lowest mean percentage palmitic acid (C16: 0) compared to the strains of
Saccharomycodes ludwigii which contain the highest mean percentage of this fatty acid. S. Luduiqi.i. CSIR-Y8 contains a higher mean percentage oleic acid (C18:1) compared to strain CSIR-Y22.
Group V. In this group the strains of Endomuces fibuliger contain the highest, while Torul.aepora del.bruecki.i. N30 and Zygosaccharomyces rouxii
CSIR-Y364 contain the lowest mean percentage palmitic acid (C16:0). Strain CSIR-Y643 of E. fibuliger contains a high mean percentage lino] eic acid
(CI8: 2) compared to strain CSIR-Y269. T. delbruecki
i
CSIR-YI38 contains the lowest mean percentage linoleic acid (CI8:2), while T. delbrueckii N30 contains the lowest mean percentage oleic acid (CI8:1). The strain representing Z. microellipsoides CSIR-Y263 contains the highest menn percentage oleic acid (CI8:1) while Z. rouxii CSIR-Y364 contains the highest mean percentage linoleic acid (CI8:2) within this group.Group VI. Candida tenuis strain CSIR-Y604 contains the highest mean percentage of linoleic acid (CI8: 2) and also the lowest mean percentage oleic acid (CI8:1) while strain CSIR-Y565 contains the highest mean percentage stearic acid (CI8:0) in the group. Cryptococcus albidus
contains the highest mean percentage ~leic acid (CI8:1) in the group while the strains of Debaryomyces hansenii contains the lowest mean percentage of linoleic acid (CI8:2) in the group.
Pilobasidium capeul iqenum is characterized by the highest mean percentage of palmitic acid (CI6: 0) and the lowest mean percentage of palmitoleic acid (CI6:1) and linolenic acid (CI8:3) in the group.
Hansenula canadensis contains the highest mean percentage linolenic acid (CI8:3) in the group.
H. subpelliculosa contains, next to Metchnikowia reukauffii, the lowest mean percentage stearic acid (CI8:0)\ in the group and is differentiated
from H. anomala also by the higher palmitoleic acid (CI6:1) content.
M. reukauffii contains the highest mean percentage palmitoleic acid (C16:1) while Pichia membranaefaciens contains the lowest mean percentage palmitic acid (C16:0) in the group.
P. guilliermondii is characterized by its linolenic acid (C18:3) content which is lower than H. canadensis but higher than the rest of the group.
Group VII. This group is divided into 5 subgroups on the basis of their oleic acid (CI8:1) content.
Subgroup a. Kluyveromyces marxianus CBS 2745 contains the highest mean percentage oleic acid (C18: 1) and the lowest mean percentage palmitoleic acid (C16: 1) and stearic acid (C18:0), while K. thermotolerans N48 is characterized by the highest mean
palmitoleic acid (C16: 1) and the (C18:2) and linolenic acid (C18:3).
percentage palmitic acid (C16: 0) and lowest mean percentage linoleic acid The strain of Saccharomyces kluyveri
contains the lowest mean palmitic acid (C16:0) and highest mean per.centage stearic acid (C18:0) and linolenic acid (C18:3).
Subgroup b. Palmitoleic acid (C16: 1) is present in the highest amount in Rl.uuveromucee matxcianus CBS 4857 while strain CSIR-Y293 contains the
