Compagno, Dipartimento di Scienze Biomolecolari e Biotecnologie, Universita` degli Studi di Milano, via Celoria, 26 20133 Milan, Italy Fax: +39 02503 14912 E-mail: concetta.compagno@unim
Trang 1Saccharomyces complex
Annamaria Merico1, Pavol Sulo2, Jure Pisˇkur3and Concetta Compagno1
1 Dipartimento di Scienze Biomolecolari e Biotecnologie, Universita` degli Studi di Milano, Milan, Italy
2 Department of Biochemistry, Faculty of Natural Sciences, Comenius University, Bratislava, Slovakia
3 Department of Cell and Organism Biology, Lund University, Sweden
The concentration of oxygen in the environment is
one of the most important factors that regulate
energy conversion in living cells Organisms have
developed multiple processes to optimize the
utiliza-tion of oxygen when its availability is reduced
Accord-ing to the role of oxygen in their metabolism, yeasts
can be classified as: (a) obligate aerobes, displaying
an exclusively respiratory metabolism; (b) facultative
fermentatives, displaying both respiratory and
fermenta-tive metabolism; and (c) obligate fermentafermenta-tives The
ability of yeasts to grow in very oxygen-limited condi-tions is strictly dependent on the ability to perform alcoholic fermentation, allowing reoxidation of NADH generated during glycolysis In Saccharomyces cerevisiae, fermentation predominates over respiration when glucose concentrations are high, even under aerobic conditions Depending on this characteristic, yeasts are classified as positive or Crabtree-negative Thus, in Crabtree-positive yeasts, such as
S cerevisiae, NADH is mainly oxidized in
glucose-Keywords
evolution; fermentation; petite mutants;
redox metabolism; respiration
Correspondence
C Compagno, Dipartimento di Scienze
Biomolecolari e Biotecnologie, Universita`
degli Studi di Milano, via Celoria,
26 20133 Milan, Italy
Fax: +39 02503 14912
E-mail: concetta.compagno@unimi.it
(Received 9 October 2006, revised 24
November 2006, accepted 11 December
2006)
doi:10.1111/j.1742-4658.2007.05645.x
The yeast Saccharomyces cerevisiae is characterized by its ability to: (a) degrade glucose or fructose to ethanol, even in the presence of oxygen (Crabtree effect); (b) grow in the absence of oxygen; and (c) generate respir-atory-deficient mitochondrial mutants, so-called petites How unique are these properties among yeasts in the Saccharomyces clade, and what is their origin? Recent progress in genome sequencing has elucidated the phylo-genetic relationships among yeasts in the Saccharomyces complex, providing
a framework for the understanding of the evolutionary history of several modern traits In this study, we analyzed over 40 yeasts that reflect over
150 million years of evolutionary history for their ability to ferment, grow
in the absence of oxygen, and generate petites A great majority of isolates exhibited good fermentation ability, suggesting that this trait could already
be an intrinsic property of the progenitor yeast We found that lineages that underwent the whole-genome duplication, in general, exhibit a fermentative lifestyle, the Crabtree effect, and the ability to grow without oxygen, and can generate stable petite mutants Some of the pre-genome duplication lin-eages also exhibit some of these traits, but a majority of the tested species are petite-negative, and show a reduced Crabtree effect and a reduced abil-ity to grow in the absence of oxygen It could be that the abilabil-ity to accumu-late ethanol in the presence of oxygen, a gradual independence from oxygen and⁄ or the ability to generate petites were developed later in several line-ages However, these traits have been combined and developed to perfection only in the lineage that underwent the whole-genome duplication and led to the modern Saccharomyces cerevisiae yeast
Abbreviation
EtBr, ethidium bromide.
Trang 2rich media by fermentation rather than by respiration,
even in the presence of oxygen This has been
attrib-uted to a limited capacity and⁄ or saturation of the
respiratory route of pyruvate dissimilation [1,2]
Glu-cose metabolism and oxygen can also be related by
the Pasteur effect, which has been defined as the
inhi-bition of fermentative metabolism by oxygen, but in
S cerevisiae this phenomenon is only observable at
low glycolytic fluxes [3] In the Kluyver effect, the
absence of oxygen impairs the utilization of particular
disaccharides, although one or both of the
monosac-charide components can be used anaerobically in
fer-mentation [4] This characteristic seems to be
determined mainly by the activity of sugar carriers
[5] The inhibition of fermentation of glucose as well
as other sugars in the absence of oxygen has been
described as the Custer effect, found in Brettanomyces
intermedius and in Candida utilis [6], and has been
proposed to be due to a redox imbalance The
regula-tory mechanisms behind these phenomena appear to
influence energy metabolism in different ways among
different yeast species
Apart from alcoholic fermentation, the ability to
grow under anaerobic conditions also determined by
other factors Some metabolic pathways require the
presence of molecular oxygen This is true to various
extents for the biosynthesis of sterols and fatty acids,
heme⁄ hemoproteins, NAD, and uracil [7,8] The
abil-ity to translocate ATP produced in the cytoplasm into
mitochondria, and the ability to adjust the redox
bal-ance, play a very important role in independence
from oxygen [9–14] In S cerevisiae, three genes
encode for ATP transporters, AAC1, AAC2 and
AAC3 Deletion of AAC2 and AAC3 is anaerobically
lethal [9–11] Under anaerobic conditions, yeast cells
can achieve redox balance by production of glycerol
[12–14] This means that the nutritional conditions
also have a strong influence on the ability to grow
anaerobically Comparison of species belonging to
several yeast genera for their ability to grow
anaerobi-cally in complex and synthetic minimal media
revealed a superiority of S cerevisiae for growth
under restrictive conditions in terms of strict
anaero-biosis and minimal presence of organic nutrients [15]
The use of cDNA arrays recently provided new
insights into gene networks and pointed out the
essen-tial role of the regulation of gene expression
underly-ing the physiologic response of S cerevisiae to oxygen
deprivation [16,17]
Saccharomyces cerevisiae constantly produces
mutants that are stable during vegetative reproduction
and are characterized by a reduced colony size on
solid media in which a fermentable carbon source is
the limiting factor [18] These mutants are called
‘petites’, and are a special class of respiratory-deficient mutants characterized by large deletions in their mtDNA or a complete lack of the mitochondrial gen-ome [19,20] Several Saccharomyces yeasts readily give rise to petites [21], but a majority of other yeasts fail
to yield stable petite mutants, and are therefore called
‘petite-negative’ yeasts [22] So far, the origin of and the biochemical and physiologic requirements for the occurrence of petites in yeast have been unclear It has previously been suggested that the petite-positive character might coincide with the ability to grow
in the absence of oxygen [22–24] However, Saccharo-myces kluyveri is an example of a yeast that can grow anaerobically, but cannot generate true petite mutants [25]
The origin of different responses to the pres-ence⁄ absence of oxygen has so far been poorly under-stood [26] Among the reasons are that few yeasts have been studied, and that the phylogenetic relation-ships among these yeasts were unclear at the time Recently, phylogenetic relationships among yeasts have been determined from a multigene sequence ana-lysis, which placed 75 species of the Saccharomyces complex into 14 well-supported clades [27] In many cases, these clades do not correspond to the circum-scribed genera: species of Kluyveromyces as well as of Zygosaccharomyces are found in different clades, indi-cating the polyphyly of these genera as presently defined According to this analysis, it was proposed
to reassign the species into five new genera [28] The
S cerevisiae lineage underwent a whole-genome dupli-cation about 100 million years ago [29–31], and the Saccharomyces clade can therefore be subdivided into pre- and post-genome duplication lineages Appar-ently, the duplication took place after the separation
of Saccharomyces, Kazachstania, Naumovia, Nakesimia and Tetrapisispora from the rest of Saccharomyces complex genera (Fig 1)
Another problem for comparative studies on the regulation of energy metabolism in aerobic and anaer-obic growth is caused by differences among the experi-mental conditions used, such as composition of media and adequate control of anaerobic conditions The purpose of the present work was to study the fermen-tative capacity, the ability to grow in anaerobic condi-tions and the occurrence of the petite phenotype in a large set of strains belonging to the ‘Saccharomyces complex’ Our study includes more than 40 strains, and provides a basis for speculation on how these metabolic traits evolved within the Saccharomyces clade, which originated approximately 150 million years ago
Trang 3Glucose metabolism and ethanol production
in aerobiosis (Crabtree effect)
In order to look for the presence of the Crabtree effect
in species belonging to the Saccharomyces complex, we
performed batch cultivations in a fermenter under
well-controlled aerobic conditions As a consequence of
respirofermentative glucose metabolism (Crabtree
effect), leading to the production of ethanol and other
byproducts (pyruvate, acetate, succinate, and glycerol),
S cerevisiaegrowing in batch on glucose under aerobic
conditions gave a low biomass yield (Table 1) Species
belonging to the genera Naumovia (Saccharomyces
castellii) and Nakaseomyces (Candida glabrata) showed
(Table 1) high specific ethanol production rates
(20.5 mmolÆg)1Æh)1 and 16.8 mmolÆg)1Æh, respectively)
as well as a low biomass yield (0.08 gÆg)1and 0.11 gÆg)1)
during the exponential phase of growth, with values very
similar to those reported for S cerevisiae [35] These
data indicate that these species behave as typical
Crab-tree-positive yeasts In the Torulaspora genus, we found
one species, T globosa, that showed a typical Crabtree
effect, with a high specific ethanol production rate
(18.6 mmolÆg)1Æh)1) and a low biomass yield
(0.08 gÆg)1) On the other hand, T delbrueckii showed a
less pronounced Crabtree effect, with a lower specific
ethanol production rate (6.13 molÆg)1Æh)1) and a higher
biomass yield (0.27 gÆg)1), as previously observed in
S kluyveri(Table 1) [36] A similar situation was
detec-ted in species belonging to the Hanseniaspora genus
Hanseniaspora vinaeand Hanseniaspora occidentalis did
in fact exhibit the ability to produce ethanol under
aero-bic conditions, but to a lower extent than S cerevisiae (Table 1) In the Zygosaccharomyces genus, Z bailii has been reported to show a reduced Crabtree effect [37] In our experiments, Z rouxii showed the lowest ethanol production rate (1.51 mmolÆg)1Æh)1) of all tested species Species belonging to the Kluyveromyces genus, such
as K wickerhamii, behaved like the Crabtree-negative yeast K lactis, being quite unable to produce ethanol under aerobic conditions (Table 1), in spite of high glucose consumption rates As a consequence of a purely respiratory metabolism, the two Kluyveromyces species showed the highest biomass yields (0.45 gÆg)1 and 0.4 gÆg)1, respectively) In conclusion, even though
a limited number of species was tested, our data indi-cate that the Crabtree effect is present in several spe-cies of the Saccharomyces complex, but is expressed at significantly different levels
Growth in aerobic conditions in the presence
of antimycin A
To further assess fermentative capacity, we tested for growth when respiration becomes gradually more impaired, by increasing the concentration of anti mycin A This drug is a well-known inhibitor of elec-tron transfer from quinone to cytocrome b [38] Yeast strains were cultivated in aerobic conditions on plates containing rich or synthetic minimal medium (Table 2) All but two of the species analyzed grew
on rich medium plus antimycin A, indicating that they are able to grow through fermentative metabo-lism, and most likely produce ethanol Most of the species, 29 out of 49, were able to grow on synthetic minimal medium at the highest antimycin A
concen-Table 1 Occurrence of respirofermentative metabolism in aerobic batch cultures: specific rates of growth (l),specific consumption rates of glucose (fructose) [qGlu (Frt)], specific production rates of ethanol (qEtOH) and the yields of biomass and ethanol relative to consumed glucose (fructose) for several yeasts of the Saccharomyces complex The data for S cerevisiae, Z bailii and S kluyveri are from the literature [35–37].
Strain
Carbon source
l (h)1)
qGlu (Frt) (mmolÆg)1Æh)1)
qEtOH (mmolÆg)1Æh)1)
Biomass yield (gÆg)1)
Ethanol yield (gÆg)1)
Trang 4Table 2 Analysis of growth under aerobic conditions in the presence of increasing concentrations of antimycin A The analysis refers to the Saccharomyces complex: the species are listed according to their phylogenetic relationship with S cerevisiae (the lowest species in the col-umn is the least related) as reported by Kurtzman & Robnett [27] –, no growth; +, growth within 7 days; NT, not tested Numbers indicate the maximal tolerated dose of antimycin A.
Antimycin A concentration (l M )
Rich medium:
5
Synthetic minimal medium:
0.5–25
Synthetic minimal medium plus lysine and glutamic acid:
0.5–25
Synthetic minimal medium plus acetoin: 0.5–25
Trang 5tration tested, whereas the rest could grow at lower
levels of the drug For some of these species, such as
Z bailii, T globosa, Zygosaccharomyces
microellipso-ides, K lactis, and H occidentalis, the addition of
acetoin, as well as the addition of lysine plus
glutam-ate, restored growth in the presence of high
concen-trations of antimycin A This suggests that for these
species the inability to grow in the presence of
anti-mycin A is mainly due to an impaired redox balance
This balance is substantially affected on synthetic
minimal medium because of the high level of NADH
generation, due to amino acid biosynthesis Much of
the generation of NADH during amino acid
biosyn-thesis takes place in the mitochondria Because of the
block in the respiratory chain caused by the addition
of antimycin A, NADH should then be reoxidized
through shuttle mechanisms with the cytoplasm [14]
Acetoin acts as a redox sink at the cytoplasmic level,
being reduced to 2,3-butanediol by the cytosolic
NAD+-linked 2,3-butanediol dehydrogenase [39] In
some species (Saccharomyces bayanus, Saccharomyces
servazii, Hanseniaspora valbyensis), we observed that
the inability to grow in the presence of high
concen-trations of antimycin A was actually due to an
impairment in the reoxidation of NADH at the
mit-ochondrial level, because in this case the addition of
acetoin did not help to restore the redox balance
(Table 2) This could indicate that, in these yeasts,
the mechanisms for shuttling NADH reducing
equiva-lents from mitochondria to cytosol are inefficient For
other species, such as S barnettii, Z rouxii, K
wick-erhamii, Eremothecium gossypii, and Kloeckera
lindner-i, the inability to grow on synthetic minimal medium
when respiration is at least partially impaired was not
alleviated by the addition of acetoin or of amino
acids In this case, the very low fermentative capacity
does not provide sufficient energy for growth when
respiration is limited
These data seem to indicate that most of the species belonging to the Saccharomyces complex possess a good fermentative capacity, being able to generate suf-ficient energy to grow when respiration is impaired Nevertheless, we observed that redox problems can, in some cases, limit the ability of the yeast to grow when the respiration chain is blocked
Growth under strict anerobic conditions All strains were cultivated on plates containing rich or synthetic minimal medium, and incubated under strict anerobic conditions Under these conditions, most of the species were able to grow after 7 days on both complex and synthetic minimal media (Fig 1) All an-alyzed species belonging to the Saccharomyces, Kaz-achstania, Naumovia, Nakaseomyces and Tetrapisispora genera were able to grow under the most stringent conditions, i.e on synthetic minimal medium under strict anaerobiosis (Fig 1, species in red)
Z microellipsoides (Torulaspora genus) and S kluyveri (Lachancea genus) were able to grow after 7 days only
on rich medium However, the addition of acetoin⁄ amino acids restored growth on synthetic minimal medium after 14 days of incubation (Fig 1, in blue) This suggests that the growth problems of these strains
on synthetic minimal medium are mainly caused by inefficient homeostasis of the redox cofactors under these conditions
Species belonging to the genera Zygosaccharomyces (Z bailii), Torulaspora (T globosa), Kluyveromyces (K lactis, K marxianus) and Hanseniaspora (H guiller-mondii and H occidentalis) showed growth on rich medium only after 14 days of incubation, but failed to grow on synthetic minimal medium, even in the pres-ence of acetoin (Fig 1, in green) This may reflect a strong redox problem that can completely impair growth in anaerobic conditions on synthetic minimal
Table 2 (Continued).
Antimycin A concentration (l M )
Rich medium:
5
Synthetic minimal medium:
0.5–25
Synthetic minimal medium plus lysine and glutamic acid:
0.5–25
Synthetic minimal medium plus acetoin: 0.5–25
Trang 6medium, where NADH production is high Z bailii is
known to produce more ethanol on fructose than on
glucose [37], and fructose is taken up by facilitated
transport [40] We then tested whether the presence of
fructose (instead of glucose) as carbon source could
allow for growth in anaerobic conditions However,
this was not the case
Other species belonging to the genera
Zygosaccha-romyces (Z rouxii, Z bisporus), Zygotorulaspora
(Z mrakii), Kluyveromyces (K aestuarii, K
nonfermen-tans, K wickerhamii), Eremothecium (E gossypii) and
Hanseniaspora (K lindneri) (Fig 1, in black) were
unable to grow on both rich and synthetic minimal
media in anaerobic conditions, even after addition of acetoin
The ability of some species to grow under anaerobic conditions on synthetic minimal medium was also tes-ted in batch cultures K lactis was used as a negative control, because it was previously found to be unable
to grow under these conditions [13] The species ana-lyzed, S castellii and C glabrata, showed the same behavior as observed in plate experiments, and were able to grow at high specific growth rates: 00.18 h)1 and 0.2 h)1, respectively (Fig 2)
In short, the upper five genera on the phylogenetic tree (post-genome duplication genera) showed a clear
Fig 1 Growth under strict anaerobic
condi-tions: yeast species in red grow both on rich
and on synthetic minimal medium within
7 days; species in blue grow on rich
med-ium within 7 days and on synthetic minimal
medium enriched with lysine and glutamic
acid or acetoin within 14 days; species in
green grow on rich medium within 14 days,
but fail to grow on the synthetic minimal
medium; species in black do not grow on
either rich or on synthetic minimal medium.
The phylogenetic tree is adapted from
Kurtzman & Robnett [27] The timing,
approximately 100 million years ago, of the
whole-genome duplication [29] is indicated
by an arrow.
Trang 7potential to grow under strictly anaerobic conditions.
On the other hand, the lower genera (pre-genome
duplication species) represent a mosaic of phenotypes;
some species being able and others being unable to
grow in the absence of oxygen
Petite generation
The ability to generate respiratory-deficient mutants
with grossly rearranged mtDNA molecules, sometimes
referred to as ‘the petite phenotype’, has often been
associated with the ability to grow anaerobically [25]
The following species, belonging to the Saccharomyces
clade, have previously been studied in detail for petite
generation ability and mtDNA structure: several
Sac-charomyces spp sensu stricto, Kazachstania genus
(S servazzii, S unisporus, S transvaalensis, S exiguus),
Naumovia genus (S castellii and S dairenensis), and
Nakeseomyces genus (C glabrata) They were found
to be petite-positive [21,41] On the other hand,
S kluyveri (belonging to the Lachancea genus) and
K lactis (belonging to the Kluyveromyces genus) do
not easily produce viable and stable petite clones [25]
Over 30 species⁄ strains, mainly belonging to the
groups that have so far not been tested for petite
generation, were analyzed in at least two independent
experiments (Fig 3) The aim of this experiment was
to determine whether a certain strain⁄ species can exist
as a petite mutant (which represents a special physio-logic state) and not to study the mechanisms behind the generation of petite mutants Kazachstania species (Arxiozyma telluris, S transvaalensis, K africanus,
S spencerorum, K lodderae, K piceae, S barnettii and
C humilis) could all generate spontaneous respiratory-deficient colonies, and also generated petites at a high frequency when exposed to ethidium bromide (EtBr)
In the Nakeseomyces genus, C glabrata and K bacilli-sporus generated spontaneous petites and EtBr-induced petites, but petites could not be detected in
K delphensis
In the Tetrapisispora genus, two species, T phaffii and T iriomotensis, were sensitive to EtBr and could therefore not be tested for petite induction, but
K blattae easily generated petites upon exposure to EtBr T phaffii and T iriomotensis did not generate spontaneous petites, or induced petites at lower EtBr concentrations The tested members of the genera Zygosaccharomyces (Z bisporus, Z rouxii), Zygotoru-laspora (Z mrakii), Torulaspora (T delbrueckii, T glo-bosa), Lachancea (Z fermentati, K thermotolerans and
S kluyveri) and Kluyveromyces (K aestuarii, K nonfer-mentans and K lactis) did not generate any sponta-neous or induced petites under the employed conditions, and are therefore considered to be petite-negative However, two species, Z florentinus and
K wickerhamii, generated petites upon prolonged exposure (10 days) to EtBr A few K wickerhamii petites were analyzed, and were found to contain grossly rearranged mtDNA with an elevated A + T content (data not shown) E gossypii was very sensi-tive to EtBr, and its ability to produce petites could therefore not be tested, but spontaneous petites could not be detected In the Hanseniospora genus, H occi-dentalis and H vinae did not generate petites spontane-ously or upon induction with EtBr, but H osmophila generated petites upon prolonged exposure to EtBr Again, post-genome duplication species, except for the Tetrapisispora group, showed an almost uniform phe-notype with regard to the ability to generate petite mutants On the other hand, a majority of the pre-gen-ome duplication species could not generate viable petites, except for three species belonging to three dif-ferent genera (Fig 3)
Discussion
The fundamental physiologic characteristics of the yeast S cerevisiae can be summarized as the ability to: (a) degrade glucose or fructose to ethanol, even in the presence of oxygen (Crabtree effect); (b) grow in the
A
B
Fig 2 Anaerobic batch cultures on glucose synthetic minimal
med-ium of (A) S castellii and (B) C glabrata: r, biomass measured as
D 600 ⁄ mL; j, glucose; m, ethanol; h, glycerol Both species show
behavior similar to that of S cerevisiae [25].
Trang 8absence of oxygen; and (c) generate
respiratory-deficient mitochondrial mutants, so-called petites [42]
However, how unique are these properties among
clo-sely related yeasts, and what is their origin? Recent
progress in genome sequencing has elucidated
phylo-genetic relationships among yeasts belonging to the
Saccharomyces clade, and thereby provides a
frame-work for an understanding of the evolutionary history
of several modern traits For example, the
whole-gen-ome duplication took place approximately 100 million
years ago in the S cerevisiae lineage [29–31], and we
can therefore talk about pre- and post-whole-genome
duplication yeasts within the Saccharomyces clade In
this study, we analyzed over 40 yeasts for their ability
to ferment, grow in the absence of oxygen, and
gener-ate stable petites, and we attempted to determine
whether these traits were expressed in the progenitor
yeasts, and whether they are related to the whole-genome duplication
A good fermentative capacity is the condi-tio sine qua non for the development of the ability to grow in strictly anaerobic conditions Under anaerobic conditions, the respiration-based biochemical pathways are shut down, and substrate-level phosphorylation is the only way for the cell to produce energy However, homeostasis of the redox cofactors is also important for continuation of metabolic activities Under anaer-obic conditions, yeast cells achieve such a redox bal-ance through the production of glycerol, mainly through the action of glycerol 3-phosphate dehydrogen-ase (Gpd2) [12], and through the production of succi-nate, by fumarate reductase [43] Under these conditions, the mitochondria do not play a role in energy metabolism, but they are still essential for some
Fig 3 Distribution of petite-positive and
petite-negative species in a phylogenetic
tree of the Saccharomyces complex,
adap-ted from Kurtzman & Robnett [27] The
examined species are indicated by different
colors: red, petite-positive species; green,
petite-negative species The timing,
approxi-mately 100 million years ago, of the
whole-genome duplication [29] is indicated by an
arrow.
Trang 9assimilatory reactions, as in amino acid biosynthesis,
and the generation of NADH [44] The ability to grow
in anaerobic conditions is therefore also strictly
dependent on the nutritional conditions
In our experiments, all but two of the analyzed
spe-cies belonging to the Saccharomyces complex could
grow on rich media when mitochondrial respiration
was partially impaired with antimycin A (Table 2)
Thus, the progenitor of the Saccharomyces complex
yeast probably had a well-developed fermentative
meta-bolism, which was sufficient to support growth in the
absence of oxygen When we made the conditions more
stringent, by increasing the concentration of
antimy-cin A and testing on the synthetic minimal medium
(Table 2), different yeast groups showed different
growth properties A high fermentative activity is, in
fact, essential in this case to cope with this situation If
the fermentative activity is too low, energy problems
can arise ATP is consumed by glucose uptake in the
case of yeasts having H+-symport mechanisms for
glu-cose transport, and in all cases ATP is used for
phos-phorylation of the hexose before ATP can be produced
in later metabolism Moreover, glycerol production
leads to reduced ATP production In these cases, the
presence of alternative respiration mechanisms, such as
cyanide-resistant salicyl hydroxamate-sensitive
respir-ation associated with the presence of complex I, can
operate and provide additional ATP when respiration
is blocked by antimycin A [45] Nevertheless, in some
cases we observed that the main problem for growth,
when respiration is impaired, seems to be insufficient
homeostasis of redox cofactors In these cases, the
addition of a redox sink, at the cytosolic as well as at
the mitochondrial level, efficiently promoted growth
This means that, in addition to high-level fermentative
metabolism, efficient mechanisms to maintain redox
balance are important for the ability to grow at low
levels of oxygen
Among the analyzed species belonging to the genera
Saccharomyces, Kazachstania, Naumovia,
Nakaseomy-ces and Tetrapisispora, those that showed high
resist-ance to antimycin A were also able to grow under the
most stringent conditions, i.e on the synthetic minimal
medium and under strict anaerobic conditions (Table 2
and Fig 1) Interestingly, some species, such as S
bay-anus, S servazii, and S barnettii, which showed
severely impaired growth in aerobic conditions in the
presence of antimycin A, were perfectly able to grow
under strict anaerobic conditions This seems to reflect
an inhibitory effect exerted by oxygen on fermentative
activity, the so-called Pasteur effect [3] Such an
inhibi-tory effect of oxygen could be a more recently
acquired trait, originating independently in several
yeast lineages In contrast, whereas the upper four post-genome duplication genera generated respiratory-deficient petite mutants, Tetrapisispora exhibited a transition petite phenotype This group deserves more study to determine the details of respiratory, fermenta-tive and mtDNA metabolism
In the other yeast groups (pre-genome duplication genera), the situation is more heterogeneous Among the analyzed species belonging to the genera Zygotoru-laspora, ToruZygotoru-laspora, Lachancea and Hanseniaspora, some of those that, in aerobic conditions, showed good resistance to antimycin A were able to grow under strict anaerobic conditions, like the above-mentioned genera (Table 2 and Fig 1) Some species belonging to the Zygosaccharomyces, Torulaspora, Kluyveromyces and Hanseniaspora groups were able to grow in anaer-obic conditions, but only on rich media, where the presence of amino acids can remedy the redox imbal-ance problems, and at low growth rates (detection requiring 14 days) Other species belonging to the gen-era Zygosaccharomyces, Zygotorulaspora, Kluyveromy-ces, Eremothecium and Hanseniaspora showed a much reduced level of resistance to antimycin A, and were quite unable to grow in anaerobic conditions, both on rich and on synthetic minimal media In these cases, the main growth problem appeared to be lack of energy, because an insufficient amount of ATP could
be generated by fermentation This interpretation is supported by the fact that S cerevisiae mutants in which glycolytic enzyme levels are low, such as gcr1 or gcr2, or in which hexose transport is inefficient, are sensitive to low concentrations of antimycin A and are unable to grow in anaerobic conditions [46,47]
The ability to grow in anaerobic conditions is a result of fine-tuning of several metabolic pathways This trait is not only dependent on the presence of genes encoding specific enzyme activities; these must also be a part of a well-regulated network The phylo-genetic tree (Fig 1) suggests that lineages that under-went whole-genome duplication exhibit a fermentative lifestyle, the presence of the Crabtree effect, the ability
to grow without oxygen, and the ability to generate petites (Table 1, Figs 1 and 3) Whereas a majority of pre-genome duplication species showed a reduced Crabtree effect, could not generate viable petite mutants, and needed some oxygen for their growth, some lineages exhibited similar traits as the post-gen-ome duplication lineage (Fig 4) However, it should
be noted that none of the pre-genome duplication spe-cies had all these traits expressed to the same quantita-tive level as the post-genome duplication species The presence of these traits in at least one species in each genus suggests that the Saccharomyces complex
Trang 10progenitor had the basic capacity to ferment, and this
was probably an adaptation to an environment with a
low oxygen concentration The mosaic distribution of
the studied phenotypes in the phylogenetic tree may,
then, reflect independent adaptations to changes in
environmental conditions that occurred many millions
of years ago The end of the Cretaceous period
provi-ded an excess of fruits, and thereby increased amounts
of sugars Different lineages of yeast, able to ferment,
entered into a fierce competition for these sugars with
different bacteria The independence from oxygen and
the ability to generate spontaneous petites, which can
only ferment and therefore produce ethanol, were
likely to strengthen the competitive advantages of
yeast Horizontal transfer of bacterial genes could also
have contributed to the increase in level of oxygen
independence [48] The ability to accumulate ethanol
in the presence of oxygen was exploited by several
yeasts as an additional weapon to inhibit the growth
of other microbes The appearance of an elevated
fre-quency of spontaneous petites helped to increase the
production of ethanol However, other evolutionary
strategies could also have contributed to the evolution
of these traits in yeasts [49,50]
Alternatively, it could be that the progenitor was
already Crabtree-positive, petite-positive and able to
grow without oxygen, but these properties were later
independently lost in several pre-genome duplication
lineages However, it is difficult to find a rationale for
this and imagine environmental conditions that would
promote this evolutionary scenario
Experimental procedures
Yeast strains The yeast species analyzed in this study belong to the Saccharomycescomplex described by Kurtzman & Robnett [27] Most of these strains were kindly provided by
C Kurtzman (Microbial Genomics and Bioprocessing Research Unit, US Department of Agriculture, Peoria, IL, USA) A majority of the studied species are represented
by their type strains: A telluris NRRL-YB-4302 (CBS 2685),
C glabrata Y-65 (CBS 138), C humilis NRRL-Y-17074 (CBS 5658), H guillermondii NRRL-Y-1625 (CBS 465), H occidentalis NRRL-Y-7946 (CBS 2592),
H osmophila NRRL-Y-1613 (CBS 313), H valbyensis NRRL-Y-1626 (CBS 479), H vineae NRRL-Y-17529 (CBS 2171), Klo lindneri NRRL-Y-17531 (CBS 285),
K aestuarii NRRL-YB-4510 (CBS 4438), K africanus NRRL-Y-8276 (CBS 2517), K bacillisporus NRRL-Y-17846 (CBS 7720), K blattae NRRL-Y-10934 (CBS 6284), K del-phensis 2379 (CBS 2170), K lodderae
NRRL-Y-8280 (CBS 2757), K marxianus NRRL-Y-8281 (CBS 712),
K nonfermentans NRRL-Y-27343 (JCM 10232), K piceae 17977 (CBS 7738), K thermotolerans
NRRL-Y-8284 (CBS 6340), K waltii NRRL-Y-8285 (CBS 6430),
K wickerhamii NRRL-Y-8286 (CBS 2745), S barnettii NRRL-Y-27223 (CBS 6946), S bayanus NRRL-Y-12624 (CBS 380), S castellii NRRL-Y-12630 (CBS 4309),
S dairensis NRRL-Y-12639 (CBS 421), S exiguus NRRL-Y-12640 (CBS 379), S kluyveri NRRL-Y-12651 (CBS 3082), S paradoxus NRRL-Y-17217 (CBS 432),
S pastorianus NRRL-Y-27171 (CBS 1538), S servazii
Fig 4 A simple phylogenetic relationship
between the yeasts analyzed in aerobic
batch cultures is shown, and the size of
their Crabtree effect is quantified as the
yields of biomass in relation to consumed
glucose (in brackets, gÆg)1) The S cerevisiae
and K lactis lineages separated more than
100 million years ago; the S cerevisiae and
S pombe lineages separated more than
200 million years ago The timing,
approxi-mately 100 million years ago, of the
whole-genome duplication [29] is indicated by an
arrow.