Yeast
Yeast 2005; 22: 177–192.
Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/yea.1200
Research Article
Inferences of evolutionary relationships from a
population survey of LTR-retrotransposons and
telomeric-associated sequences in the Saccharomyces
sensu stricto complex
Gianni Liti1 *, Antonella Peruffo1# , Steve A. James2 , Ian N. Roberts2 and Edward J. Louis1
1 Department of Genetics, University of Leicester, Leicester LE1 7RH, UK
2 National Collection of Yeast Cultures, Institute of Food Research, Norwich
*Correspondence to:
Gianni Liti, Department of
Genetics, University of Leicester,
University Road, Leicester LE1
7RH, UK.
E-mail: gl23@le.ac.uk
# Present Address: Department
of Experimental Veterinary
Science, University of Padova,
Legnaro, Italy.
Received: 18 August 2004
Accepted: 8 November 2004
Research Park, Norwich NR4 7UA, UK
Abstract
The Saccharomyces sensu stricto complex consists of six closely related species and
one natural hybrid. Intra- and inter- species variability in repetitive elements can
help elucidate the population structure and evolution of these close relatives. The
chromosome positions of several telomeric associated sequences (TASs) and LTRretrotransposons have been determined, using PFGE, in 112 isolates. Most of the
repetitive elements studied are found in multiple copies in each strain, although
in some subpopulations these elements are present in low copy number or are
absent. Hybridization patterns and copy numbers of the repetitive elements correlate
with geographic distribution. These patterns may yield interesting clues as to the
origins and evolution of some TASs and retrotransposons, e.g. we can infer that
Y originated on the left end of chromosome XIV. There is strong evidence for
horizontal transfer of Ty2 between S. cerevisiae and S. mikatae. Ty1 and Ty5 are
either lost easily or frequently horizontally transferred. We have also found some
gross chromosomal rearrangements in isolates within species and a few new natural
hybrids between species, indicating that these processes occur in the wild and are not
limited to conditions of human influence. DNA sequences have been deposited with
the EMBL/GenBank database under Accession Nos AJ632279–AJ632293. Copyright
 2005 John Wiley & Sons, Ltd.
Keywords:
evolution
Saccharomyces; GCR; LTR-retrotransposon; TAS; ITS1 ; hybrids;
Introduction
The Saccharomyces sensu stricto complex, demonstrated on the basis of DNA–DNA reassociation
and hybrid sterility, was originally represented
by S. cerevisiae, S. paradoxus, S. bayanus and
the sterile hybrid S. pastorianus (Martini and
Martini, 1987; Vaughan Martini, 1989; Naumov,
1987). Recently, three new species S. cariocanus,
S. mikatae, S. kudriavzevii, showing reproductive
isolation with the previous four, have been added
(Naumov et al., 1995a, 1995b, 2000a).
Copyright  2005 John Wiley & Sons, Ltd.
The relatively recent speciation makes this group
an attractive model to investigate important biological aspects, such as the molecular mechanism of
speciation (Fischer et al., 2000; Greig et al., 2002,
2003; Delneri et al., 2003; Hunter et al., 1996),
and to identify functional elements and consensus
non-coding sequences (Kellis et al., 2003; Cliften
et al., 2001, 2003). However, this similarity makes
their taxonomic classification difficult and often
controversial.
In this study, strains were obtained from several yeast collections, and were selected for their
geographical origin of isolation. S. cerevisiae
178
strains are available in large number. However,
almost all of the strains are derived from human
activity associated with fermentation processes
(Vaughan-Martini and Martini, 1995). The same
situation is found in the hybrid species S. pastorianus, associated with lager beer, and S. bayanus,
where many strains also appear to be hybrid
(Casaregola et al., 2001; de Barros Lopes et al.,
2002). We only used S. bayanus strains which displayed no evidence of hybrid nature. Strains used
in industrial fermentation processes undergo continuous culturing and exposure to different stress
conditions (i.e. temperature, low pH, high sugar
and ethanol concentrations), which could result in
recombination, transposition and gross chromosomal rearrangements (GCRs). Spontaneous reciprocal translocations involving chromosomes VIII and
XVI are selected under fermentation conditions,
such as the use of sulphite as a wine preservative (Perez-Ortin et al., 2002). All these events
could affect the number and position of repetitive
sequences.
Although rarely associated with human activity (Naumov et al., 1998), S. paradoxus has been
frequently isolated from tree bark, flux exudates
(mainly Quercus spp.) and associated soil. Numerous S. paradoxus strains have been isolated from
geographically distinct regions: northern and southern Europe, UK, Far East Asia, South Africa and
North America (Greig et al., 2003; Sniegowski
et al., 2002; Johnson et al., 2004; Naumov et al.,
1997). These isolates are nearly all wild-type,
homothallic (HO) diploids. These two features
make S. paradoxus an attractive model to assess
genetic variation independent of the influences
human activity may have had on S. cerevisiae
(Johnson et al., 2004).
Fewer isolates are available for the other three
species. Two strains for each species isolated in
Japan, S. mikatae and S. kudriavzevii, and two
strains from Brazil (S. cariocanus) have been previously described (Naumov et al., 2000a). There are
likely to be more strains belonging to these species
hidden in the various yeast collections but classified
as other species of Saccharomyces sensu stricto,
due to their high degree of relatedness. Two additional strains of S. kudriavzevii and 12 of S. mikatae
were obtained from the Japanese NBRC yeast collection (formerly the IFO), following their reclassification in light of the recent species descriptions
(Naumov et al., 2000a).
Copyright  2005 John Wiley & Sons, Ltd.
G. Liti et al.
In order to assess the degree of genetic variation within this complex, we surveyed the genomes
of 112 different strains. Our approach utilized
chromosomal separation and probe hybridization.
Since variability in eukaryotic organisms increases
toward telomeres, repetitive elements, including
telomeric associated sequences (TASs), as well
as LTR-retrotransposons, were chosen for detecting genetic variation. We show that the chromosome hybridization pattern correlates well with
geographic distribution. In addition, we can now
suggest a scenario concerning the origin and evolution of specific repetitive elements in Saccharomyces genomes.
Materials and methods
Yeast strains
Yeast strains have been provided from a number of
different laboratories and yeast culture collections
and are listed in Table 1. No novel strains have
been isolated for this study.
Probe cloning and amplification
Genes used as probes were amplified by PCR
either from DNA of strains S288C and Y55 or
subcloned from other plasmids (Table 2). Products were cloned using the TOPO TA Cloning Kit
(Invitrogen) according to the manufacturer’s protocol. Plasmids were purified with QIAprep Miniprep
Kit (QIAGEN) and digested with EcoRI in order
to check the size of the cloned fragment. Plasmids
showing the right size of insertion were sequenced
at the internal nucleic acid services PNACL, University of Leicester, UK.
CHEF gel and Southern hybridization
We used seven different probes for a set of four
membranes containing all 112 isolates, and three
additional probes in 30 selected strains. All probes
are listed in Table 2. Genomic DNA samples,
CHEF gels, Southern blot and hybridization were
performed as previously described (Louis, 1998).
PCR amplification of the internal transcribed
spacer (ITS) region
The entire ITS region was amplified from 51 of
the Saccharomyces sensu stricto isolates listed in
Yeast 2005; 22: 177–192.
Repetitive sequences in Saccharomyces
Table 1. Saccharomyces sensu stricto strains analysed for
repetitive sequences
Species
S. paradoxus
DBVPG 6565,
6566, 6698
CBS 432T
CBS 5829
DBVPG 4650
DBVPG 4652
DG1768
DBVPG 6303–4,
6037
DBVPG 6045
Q4.1, Q32.3,
Q59.1, Q70.8, T8.1,
T21.4, T32.1, T62.1,
T76.6
DBVPG 6701
N-8, N-9, N-11,
N-12, N-15, N-17,
N-18, N-25, N-34,
N-36
N-42–N-50
YPS125, 138, 145,
150, 151, 152, 155,
158
S. bayanus
CBS 7001
VKM Y-361, 508
NRRL Y-969
VKM 1146–6B
S. mikatae
NBRC 1815T –16,
10 992–11 003
S. cariocanus
UFRJ 50 791,
50 816T
S. kudriavzevii
NBRC 1802T –3,
10 990–1
NCYC 1379
S. cerevisiae
DBVPG 7054,
7062
DBVPG 6295
DBVPG 6763–5,
6907 (var. boulardii)
DBVPG 6693
DBVPG 6696
DBVPG 6861
DBVPG 6044
DBVPG 6041
DBVPG 1788
DBVPG 3591
DBVPG 3051
DBVPG 1849,
3049
Sources
179
Table 1. Continued.
Species
Sources
Origin
DBVPG 1339
DBVPG 1794
DBVPG 1373
DBVPG 1853
DBVPG 4651
DBVPG 1378
DBVPG 1133,
1135
SK1
S288c
Y55
YPS128, 129
Grape must
Soil
Soil
White Tecc
Truffle
Grape must
Cherries
The Netherlands
Viik, Finland
The Netherlands
Ethiopia
Italy
Sardinia, Italy
Sicily, Italy
Laboratory
Rotting fig
Wine
Associated with
Quercus sp.
USA
USA
France
Pennsylvania, USA
Beer
Beer
The Netherlands
Copenhagen,
Denmark
Denmark
The Netherlands
Unknown
Copenhagen,
Denmark
Copenhagen,
Denmark
Origin
Spoiled mayonnaise
Unknown
Soil
Mor soil
Guano
Soil
Laboratory
Drosophila sp.
Russia
Denmark
Italy
Italy
Unknown
USA
Exudates of Quercus sp. The Netherlands
Exudates of Quercus sp. London
S. pastorianus
DBVPG 6258
DBVPG 6285
Tree exudates
Russia
Exudates of Quercus sp. Russia
Exudates of Quercus sp. Far East Asia
Associated with
Pennsylvania,
Quercus sp.
USA
Mesophylax adoperus
Wine
Unknown
Grape berries
Spain
Czech Republic
Unknown
Russia
Soil and decayed leaf
Japan
Drosophila sp.
Brazil
Decayed leaf
Japan
Brewery
New Zealand
Beer
Czech Republic
Grape must
Unknown
South Africa
Unknown
Beer
Banana wine
Polluted stream water
Bili wine
Faeces of man
Soil
Cocoa beans
Grape must
White Tecc
Belgium
Burundi
Brazil
West Africa
Unknown
Turku, Finland
Unknown
Israel
Ethiopia
Copyright  2005 John Wiley & Sons, Ltd.
DBVPG 6560
DBVPG 6047
DBVPG 6033
DBVPG 6283
Brewery
Hansen’s 1888
Unknown
Unknown
DBVPG 6282
Unknown
Geographical origin and source have been included. DBVPG,
Dipartimento Biologia Vegetale Perugia, Yeast Industrial Collection,
Perugia, Italy; NCYC, National Collection of Yeast Culture, Norwich,
UK; CBS, Centraalbureau voor Schimmelcultures, Utrecht, The
Netherlands; NBRC (ex-IFO), NITE Biological Resource Centre,
Chiba, Japan; VKM, National Collection of Microorganisms, Moscow,
Russia; UFRJ, Universidade Federal do Rio de Janeiro, Brazil; NRRL,
ARS Culture Collection, Peoria, IL, USA; YPS, Yeast P. Sniegowski,
University of Pennsylvania, Philadelphia, PA, USA; N-, strains isolated
by G. Naumov, Moscow, Russia; Q and T, strains isolated by A. Burt,
Imperial College, London, UK.
Table 1, using the conserved fungal primers ITS1
and ITS4 (White et al., 1990) and a Biometra T1
thermocycler. The protocol used was previously
described (James et al., 1996), except that PCR
amplification was carried out on genomic DNA and
not directly using whole cells. The amplified products were purified using a QIAquick PCR purification kit (QIAGEN) according to the manufacturer’s
protocol.
Sequence determination and analysis
The ITS1 sequence of each Saccharomyces sensu
stricto isolate was determined using the primers
ITS1 and ITS2 (White et al., 1990). Direct sequencing was performed using a Taq DyeDeoxy terminator cycle sequencing kit (Applied Biosystems)
Yeast 2005; 22: 177–192.
180
G. Liti et al.
Table 2. Probes used in this study
Name
Gene
Notes and reference
pEL50
pRED552
ERR1
PAU6
pRED551
Core X
Enolase-related repeat (Pryde et al., 1995)
Cloned from PCR amplification of genomic DNA of Y55 and sequenced (FW:
TAACTTCAATCGCTGCTG RE: CCGTCCTTGGATAGAGC)
400 bp subcloned from pEL89 (XI-L) via PCR, TA cloned and sequenced (FW:
TAGTGGTGATTTTGTGGG T, RE: TCACATGCCATACTCACC)
TA cloning of 4.6 kbp of internal fragment of Y shared by the short and long form ending
before the 36 bp repeat (FW: TCACTGTATTGCATGCTGGA, RE: atgaatgcacgtgtcgctgt)
Ty1-specific probe was a 1.3 kbp EcoRI–SalI fragment (Naumov et al., 1992)
Ty2-specific probe was a 1.7 kbp ClaI–ClaI fragment (Naumov et al., 1992)
Cloned from PCR amplification of genomic DNA of S288C and sequenced (FW:
AGTAATGCTTTAGTATTG, RE: TAGTAAGTTTATTGGACC)
Cloned from PCR amplification of genomic DNA of S288C and sequenced (FW:
AGTAATGCTTTAGTATTG, RE: TAGTAAGTTTATTGGACC)
270 bp amplified with gradient PCR (FW: GTGACATGAGTTGCTATGG, REV:
AACACACGCCGATTGGTC)
2100 bp amplified with gradient PCR (FW: GTACTGATCATGAACCAGG, REV:
CACCTCTGCAGACTATCC)
Y
pKS4
pJH80
pJH81
pRED555
Ty1
Ty2-917
YCLW Ty5-1
pRED556
YCLW ω1
LTR4
TY4-1
and a Biometra T1 thermocycler, according to the
manufacturers’ recommendations. Sequences were
edited with the program DNAMAN, version 5.0
(Lynnon BioSoft). The CLUSTAL W (Thompson
et al., 1994) algorithm included in the DNAMAN
package was used to align the ITS1 sequences
and to construct a neighbour-joining tree with
1000 bootstrap iterations. BLAST searches of the
known genome sequences from the sensu stricto
group (Cliften et al., 2003; Kellis et al., 2003)
for homologies to S. cerevisiae Ty elements, Y s,
core X elements, ERR and PAU genes were also
performed.
Tetrad dissection
Sporulation was induced for 3–5 days at room temperature in 1% potassium acetate media and spores
were dissected as previously described (Naumov
et al., 1994).
Results
Gross chromosomal rearrangements (GCRs)
are found at low frequencies in wild-type
isolates
The karyotypes of the 112 strains listed in
Table 1 have been characterized. Strains belonging
to the same species are largely homogeneous
with few chromosome-size polymorphisms (not
Copyright  2005 John Wiley & Sons, Ltd.
shown). Rearrangements consistent with GCRs
have been detected in four isolates of S. paradoxus
(Figure 1A) and one of S. bayanus (Figure 1C).
Specifically, in S. paradoxus CBS 5829 and
N-43, these rearrangements are consistent with
de novo reciprocal translocation involving two
different chromosomes (see legend of Figure 1
for details), as inferred by their spore viability
(<50%) when crossed with other S. paradoxus
strains (Greig et al., 2003; Naumov et al., 1995b).
In S. paradoxus T21.4 a different rearrangement
occurred. A fragment for chromosome I is
apparently missing and a signal of double intensity
corresponding to chromosome VI indicates their
co-migration (Figure 1A, lane 4). In this isolate,
no other chromosome seems to be affected in size,
excluding any reciprocal translocation. A possible
explanation for this increase in size (approximately
70 kbp) is single or multiple duplications. A similar
but smaller (25 kb) increase at chromosome I of S.
paradoxus N-17 (Figure 1, lane 6) is seen.
S. mikatae is an exception to this general rule
of karyotype stability. Strains NBRC 1815T and
NBRC 1816, previously analysed, show one and
two translocations, respectively, compared with S.
cerevisiae (Fischer et al., 2000). Twelve of the
S. mikatae strains analysed in this study show
a great number of chromosome polymorphisms
(Figure 1D) and some of these changes in size are
consistent with GCRs such as translocations. Some
chromosomes are more stable than others, such as
the four smaller ones, I, III, IX and VIII, which
Yeast 2005; 22: 177–192.
Repetitive sequences in Saccharomyces
181
Figure 1. Electrophoretic karyotypes of Saccharomyces sensu stricto strains. The strains selected here exhibit GCRs or
are potentially hybrids. In parentheses are indicated chromosomes involved in rearrangements and H indicates potential
hybrids. First lane of each panel shows a conventional karyotype for these species. (A) S. paradoxus. Lanes: 1, N-44; 2, N-43
(VIII, XV-R); 3, CBS 5829 (V, XI); 4, T21.4 (I); 5, DBVPG 6566 (II, XIII); 6, N-17 (I); 7, DBVPG 6701 (H). (B) S. cerevisiae.
Lanes: 8, S288C; 9, DBVPG 1849 (H); 10, DBVPG 1373; 11, DBVPG 1853; 12, NCYC1379 (H). (C) S. bayanus. Lanes: 13,
VKMY 508; 14, VKMY 361 (XI). (D) Electrophoretic karyotype of 14 strains of S. mikatae showing an unusual high degree
of chromosome polymorphism. Lanes: 15, NBRC 1815T ; 16, NBRC 1816; 17–28, NBRC 10 992–11 003
appear constant in size. The karyotype of S. mikatae
appears to be unstable.
A limited number of S. cerevisiae strains show
an unconventional karyotype with additional bands
(Figure 1B). Finally, a potential hybrid karyotype
has been found in S. paradoxus DBVPG 6701
(Figure 1A, lane 7), DBVPG 4652 and DBVPG
6045 (result not shown).
LTR-retrotransposons outline differences
between geographical subpopulations
We assess the presence/absence of four different S. cerevisiae classes of LTR-retrotransposons.
Hybridization results are summarized in Table 3.
The two most distant species from S. cerevisiae, S.
bayanus and S. kudriavzevii, show an absence of
hybridization signals in most isolates for all four
Ty elements. A faint hybridization signal using a
complete Ty4 sequence was detected in these distant species (Table 3), confirming its presence in
the common Saccharomyces sensu stricto ancestor
Copyright  2005 John Wiley & Sons, Ltd.
(Neuveglise et al., 2002). In these species, LTRretrotransposons could either have degenerated
after speciation, and therefore cannot be detected
using genes cloned from S. cerevisiae as probes, or
they are absent. The recent sequencing of a number of the sensu stricto genomes will be useful
for investigating repetitive sequences in these less
related species (Cliften et al., 2003; Kellis et al.,
2003). BLAST results with Ty1 indicate 87% identity in S. mikatae over a large part of the element
and 74% identity over a smaller region in S. kudriavzevii. This drops to limited homology over a
small domain in S. bayanus, which could be due
to homologies between the Ty families. A similar
result is seen for Ty2, except that the percentage
identity is much higher for S. mikatae (see below).
Ty4 exhibits reasonable percentage identity over a
larger fragment size across all the species. With
Ty5 there is no homology seen in the S. bayanus
sequences.
In S. mikatae we detected an extreme variation in terms of presence and abundance of
Yeast 2005; 22: 177–192.
182
G. Liti et al.
Table 3. Prevalence of LTR-retrotransposons in the sensu
stricto complex
Subgroups
or species
S. bayanus
S. kudriavzevii
S. mikatae
NBRC 10 992–4
NBRC 10 996
NBRC 10 999
NBRC 11 000,
11 002–3
NBRC 1815–6,
10 995, 10 997–8,
11 001
S. paradoxus
Far East
Pennsylvania
DG 1768
DBVPG 6303-4
Most isolates
(including var. douglasii)
DBVPG 4652
DBVPG 6701
DBVPG 6045
S. cariocanus
S. cerevisiae
Most isolates
var. boulardii
DBVPG 6044 (T of S.
mangini)
Ty1
Ty2
Ty4
Ty5
A
A
P
H
1
P
P
A
A
P
L
H
P
A
P
P
P
P
P
P
P
A
A
P
P
A
A
P
P
P
P
P
P
L
A
A
P
A
A
A
A
A
H
H
A
P
P
A
A
A
A
P
H
1
H
P
H
P
H
A
H
?
H
P
A
A
A
A
P
1
P
P
P
1
P
P
P
P
P
?
P, presence; A, absence; H, high, present in many chromosomes; 1,
present only in one chromosome; ?, uncertain results.
LTR-retrotransposons (Table 3), e.g. strain NBRC
10 993 is rich in Ty1 elements (in at least 10 chromosomes; results not shown) and poor in Ty2 (only
two chromosomes). On the other hand, NBRC
10 996 has only one chromosome carrying Ty1 and
11 hybridizing with the Ty2 probe; 11/14 strains
of S. mikatae show a presence of Ty2 similar to
S. cerevisiae. This result is unexpected, since the
two closest related species to S. cerevisiae, S. paradoxus and S. cariocanus, do not hybridize with this
retro-element (Table 3). Two strains of S. mikatae,
NBRC 10 996 and NBRC 10 999, do not show
any Ty5 hybridization, while the others have Ty5
(Table 3).
The availability of different geographical isolates of S. paradoxus offers an opportunity to
look at variation in subpopulations. Both Far East
and Pennsylvanian isolates are missing different
classes of LTR elements (Table 3). Specifically,
both groups are missing Ty5 elements. Previous
Copyright  2005 John Wiley & Sons, Ltd.
reports showed lack of Ty5 elements in some North
America isolates used in this study (Zou et al.,
1995). In addition, in the Pennsylvanian isolates,
Ty1 is absent or present at very low copy number,
as previously reported (Sniegowski et al., 2002).
Both Far East and Pennsylvanian isolates have elevated numbers of chromosomes with Ty4. A few
strains have divergent features compared with the
majority of S. paradoxus isolates. Specifically, isolates DBVPG 4652, DBVPG 6701 and DBVPG
6045 show the presence of Ty2 and absence of
Ty5. However, these three strains have qualities of
more than one species as discussed below. Furthermore, DBVPG 6045 shows an unusually large
number of Ty1, Ty2, Ty4 and Y elements in all its
chromosomes (Table 3).
Most S. cerevisiae isolates have all the LTR
transposon classes present with a high degree of
variation and only a few strains have a complete
loss of entire Ty classes. The four strains classified
as S. cerevisiae var. boulardii (DBVPG 6763-5
and DBVPG 6907) show Ty1 elements only in
chromosome IV (Figure 2). We have found the
same results in a further 10 strains not related by
their source or geographical isolation. Chromosome
IV could potentially be the original location of
Ty1 in S. cerevisiae. Ty2 is present in a single
chromosome in isolate DBVPG 6044, previously
classified as S. mangini. Some isolates seem to lack
Ty5, as previously described (Zou et al., 1995).
Distribution of Y elements among the
Saccharomyces sensu stricto complex
The Y elements, which contain a RNA helicase, are widely distributed and variable at telomere ends in all sensu stricto species with the
exception of S. bayanus (Louis et al., 1994; Naumov et al., 1992, 2000a). Strains listed in Table 1
have been tested for its presence and distribution
(Figure 3, Table 4). Wild-type strains of S. paradoxus, indicated as Far East isolates, are extremely
poor in terms of abundance of this telomericassociated sequence (TAS). In the N-44 isolate,
no Y sequence has been detected (Figure 3A, lane
26). The other eight Far East isolates show at least
one copy located at chromosome XIV, and in four
of them a second copy has been detected at doublet
XIII + XVI (Figure 3A). In analysing other geographical subpopulations, a Y in chromosome XIV
is consistently shared by isolates from the Russia,
Yeast 2005; 22: 177–192.
Repetitive sequences in Saccharomyces
183
Figure 2. PFGE hybridization of S. cerevisiae isolates using an internal fragment of Ty1 as probe (pJH80). Lanes: 1, S288c;
2, SK1; 3, Y55; 4, DBVPG 6763; 5, DBVPG 6764; 6, DBVPG 6765; 7, DBVPG 6907; 8, DBVPG 6044; 9, DBVPG 6041; 10,
DBVPG 7054; 11, DBVPG 6295; 12, DBVPG 7062; 13, DBVPG 6693; 14, DBVPG 6696; 15, DBVPG 6861; 16, DBVPG
1788; 17, DBVPG 3591; 18, DBVPG 3051; 19, DBVPG 1849; 20, DBVPG 1339; 21, DBVPG 1794; 22, DBVPG 3049; 23,
DBVPG 1373; 24, DBVPG 1853; 25, DBVPG 1378; 26, DBVPG 1135; 27, DBVPG 1133
UK, USA (var. douglasii) and other EU countries,
with few exceptions. Strains recently isolated in
Pennsylvania present a relatively high number of
Y s but only four out of eight show hybridization
in chromosome XIV (Figure 3B). Strains DBVPG
6303, DBVPG 6304 and DBVPG 6037, isolated
in North America, share a low number of Y
sequences but none in chromosome XIV (lanes
9–11, Figure 3B). However, these strains present
several features divergent with the rest of the S.
paradoxus isolates studied, including their ITS1
sequence (Figure 6). Eleven strains of different
geographical location show only two Y copies, one
at XIV and an additional hybridization in chromosome V. This suggests that chromosome XIV
may have been the first location of Y and that
chromosome V may have been the first stage of
telomere–telomere recombination involving Y s in
S. paradoxus, but alternative explanations are also
possible.
All S. cerevisiae strains tested displayed a large
number of chromosomes with Y ends. There is
no evident pattern shared by the different strains,
so it is quite difficult to trace back its route
of spread, although they all have Y at XIV.
The only strains showing a limited number of
Y s are isolates classified as S. cerevisiae var.
boulardii and these include chromosomes XIV and
Copyright  2005 John Wiley & Sons, Ltd.
V, consistent with the scenario outlined for S.
paradoxus (Figure 3B).
Y hybridization has been already characterized in the few isolates available for the three
new species (Naumov et al., 2000a). Two Brazilian isolates of S. cariocanus show a translocation involving chromosome XIV–L. In this rearranged chromosome Y is also present (Naumov
et al., 2000a). More Japanese isolates, recently reassessed by NBRC, have been analysed. S. mikatae
strains show very high variation in terms of Y
(Figure 3B). Three strains lack the presence of this
element (NBRC 10 995, NBRC 10 997 and NBRC
10 999). Four strains exhibit hybridization in chromosome VIII and IX, while other strains show
large numbers of this TAS. The low number of
Table 4. Presence and absence of TAS in Saccharomyces
sensu stricto complex
Species
CoreX
Y
PAU
ERR
A
A
P
P
P
P
A
P
P
P
P
P
P
P
P
P
P
P
A
A
A
P
P
P
S. bayanus
S. kudriavzevii
S. mikatae
S. paradoxus
S. cariocanus
S. cerevisiae
P, presence; A, absence.
Yeast 2005; 22: 177–192.
184
G. Liti et al.
Figure 3. PFGE hybridization using an internal fragment of Y as probe (pKS4). (A) Different S. paradoxus isolates. Lanes:
1, CBS 5829 (chromosome V is rearranged in this strain); 2, DBVPG 6565; 3, DBVPG 6566; 4, DBVPG 6698; 5, Q4.1;
6, Q32.3; 7, Q59.1; 8, Q70.8; 9, T8.1; 10, T21.4; 11, T32.1; 12, T62.1; 13, T76.6; 14, N-8; 15, N-9; 16, N-11; 17, N-15;
18, N-17; 19, N-18; 20, N-25; 21, N-34; 22, N-36; 23–32, N-42–N-50. (B) S. paradoxus. Lanes: 1, YPS125; 2, YPS138; 3,
YPS145; 4, YPS150; 5, YPS151; 6, YPS152; 7, YPS155; 8, YPS158; 9, DBVPG 6303; 10, DBVPG 6304; 11, DBVPG 6037. S.
cerevisiae. Lanes: 12–14, DBVPG 6763–5, 15, DBVPG 6907. S. mikatae. Lanes: 16–27, NBRC 10 992–11 003
isolates available for S. kudriavzevii and S. cariocanus do not allow a comparative analysis. We
confirm absence of Y in S. bayanus (Table 4).
Subtelomeric structure of the Saccharomyces
sensu stricto complex
Using the same approach several subtelomeric
sequences have been investigated (Table 4). Core X
Copyright  2005 John Wiley & Sons, Ltd.
is present at all chromosome ends in S. cerevisiae.
We did not find any strain variation in this species,
except for isolates DBVPG 7054 and DBVPG
7062, as well as in few S. paradoxus isolates
that have an unusual Y pattern. However, X elements have been detected in a few chromosomes
in S. mikatae and S. cariocanus (the four smaller
ones) but cannot be detected in the less related
Yeast 2005; 22: 177–192.
Repetitive sequences in Saccharomyces
species, S. bayanus and S. kudriavzevii (Table 4).
These elements are highly variable, even within
chromosomes of the same strain (80–95% identity in the sequenced strain of S. cerevisiae). We
do see 60–70% identity with the core X element
by BLAST search in these less related species but
this is below the detection level by hybridization.
PAU genes are the largest multigene family
found in S. cerevisiae and are located preferentially at subtelomeric positions (Viswanathan et al.,
1994; Pryde and Louis, 1997). A role in fermentative metabolism has been suggested (Rachidi et al.,
2000). Hybridization results using PAU6 reveal its
presence among all species of the sensu stricto
complex (Table 4). Strong hybridization signals
are detected in all 112 strains, indicating a high
degree of homology shared by all the Saccharomyces sensu stricto species. As these are embedded in the highly variable subtelomeric regions, this
gene family could represent an excellent tool in
DNA fingerprinting of Saccharomyces sensu stricto
strains.
Finally, in a selected number of strains, we
investigated the presence of enolase related repeat
(ERR). ERRs share homology with ENO1 and
ENO2 genes. This gene is within a subtelomeric region shared between chromosome XIII-R
(ERR3 ), XV-R (ERR1 ) and XVI-L (ERR2 ) (Pryde
et al., 1995). All S. cerevisiae strains analysed
reveal the same hybridization profile using ERR1
probe (Figure 4). A faint signal corresponding to
chromosome XV, indicating a low level of homology, is also exhibited by some strains of S. paradoxus and S. cariocanus but is not detected in
the other species (Figure 4). This chromosome
end could be where this gene was located originally before undergoing multiple duplications in
S. cerevisiae. These results are confirmed by the
recent release of S. paradoxus sequences: a 500
185
bp sequence has been found, with a homology
of 79% corresponding to ERR1 located in XV-R.
However, the subtelomeric duplication of ERR1 is
found solely in S. cerevisiae.
Spontaneous hybridization occurs in natural as
well as industrial Saccharomyces
Saccharomyces sensu stricto species can generate
viable hybrids but a post-zygotic barrier results
in only rare viable spores (Naumov et al., 1994).
Several strains have been reported to arise from
spontaneous hybridization between different Saccharomyces species (de Barros Lopes et al., 2002;
Groth et al., 1999; Masneuf et al., 1998). PFGE
analysis of the strains in this study indicates a
potential hybrid nature of several strains (Figure 1
and results not shown). These strains present an
unusual karyotype with an additional number of
bands compared to non-hybrid strains. We have
found five strains with features that strongly support a hybrid nature.
In NCYC 1379, isolated in a New Zealand
brewery and identified as S. cerevisiae by conventional chemotaxonomy, numerous indications
suggest a hybrid origin of this strain. Subsequent
26S rDNA sequencing identified NCYC 1379 as
a strain of S. kudriavzevii (100% sequence identity to the type strain NBRC 1802T ), whereas
hybridization results indicate an S. cerevisiae-like
donor contributing to part of the genome. Furthermore, Southern hybridization reveals the presence of LTR-retrotransposons and TASs, such as
Ty2 and ERR1, present in S. cerevisiae but not
in S. kudriavzevii. Analysis of viability of spores
strongly supports a hybrid origin for NCYC 1379,
since only 2.5% (2/144) spores are viable, despite
normal levels of sporulation.
S. cerevisiae strain DBVPG 1849 exhibits additional bands in its electrophoretic karyotype (lane 9,
Figure 4. PFGE hybridization using the ERR1 probe. ERR1 hybridizes in a single position on chromosome XV in S. paradoxus
strains. Lanes: 1, DBVPG 6044; 2, DBVPG 6763; 3, YPS128; 4, T32.1; 5, DBVPG 6566; 6, DBVPG 6303; 7, N-44; 8, YPS
125; 9, UFRJ 50 791; 10, UFRJ 50 816; 11, NBRC 1816; 12, NBRC 1802T ; 13, CBS7001; 14, DBVPG 4652
Copyright  2005 John Wiley & Sons, Ltd.
Yeast 2005; 22: 177–192.
186
Figure 1B) that were supported by different Southern hybridization patterns. A very low spore viability, 3.75% (3/80), is consistent with more than
one species contributing to its genome. The ITS1
region of DBVPG 1849 displays 99.7% sequence
identity with S. cerevisiae CBS382. The other parent species contributing to this strain is unknown.
Isolate DBVPG 4652 has been isolated and
classified as S. paradoxus by assimilation tests, but
shows several atypical features compared to the
majority of S. paradoxus strains. Ty2 is abundant in
DBVPG 4652 despite its absence in most isolates
of S. paradoxus. Furthermore, a strong indication of
its hybrid origin is derived from a Southern analysis
using ERR1 as a probe, since only one band
corresponding to doublet XIII + XVI has been
detected (Figure 4). The signal is very strong and
comparable to S. cerevisiae strains. Spore viability
is 48.5% (38/80) and higher than expected for a true
hybrid. However, this strain could derive from a
fertile allotetraploid, as previously shown (Naumov
et al., 2000b), or from one of the rare viable spores
of the original hybrid which overcame the meiotic
defect. New species have been created in this way
(Greig et al., 2002). The ITS1 region could not be
sequenced directly from a PCR product.
A similar situation has been found in S. paradoxus DBVPG 6045, with presence of additional
bands in the karyotype and the presence of Ty2.
This strain does not generate spores in sporulation
media.
Finally, S. paradoxus DBVPG 6701 is also a
potential hybrid. This strain has Ty2, additional
chromosome bands in PFGE, spore viability of
67.5% (54/80) and an ITS1 sequence similar to
that in S. kudriavzevii strains. Interestingly, all five
potential hybrids resulted from crosses between
species with co-linear genomes (DBVPG 6701 S.c.
× S.k.; DBVPG 4652 S.c. × S.p.; NCYC1379 S.c.
× S.k.; DBVPG 1849 S.c. × unknown; DBVPG
6045 S.c. × S.p.).
Eight strains of S. pastorianus have also been
investigated in this screening for repetitive sequences. These strains show extensive differences
in repetitive sequence and fall into two classes
(Figure 5). This is consistent with at least two
independent events leading to the hybrids, rather
than a single event from which all isolates were
derived. However, because of their hybrid nature,
analysis is difficult and other approaches need to
be employed.
Copyright  2005 John Wiley & Sons, Ltd.
G. Liti et al.
Figure 5. Hybridizations pattern of different isolates of S.
pastorianus. Strains: 1, DBVPG 6033; 2, DBVPG 6047; 3,
DBVPG 6258; 4, DBVPG 6285; 5, DBVPG 6560; 6, DBVPG
6283; 7, DBVPG 6282
ITS1 sequences reveal novel single nucleotide
polymorphisms (SNPs)
In order to determine genetic variation, outlined by
repetitive elements, at the sequence level we investigated the ITS1 region. This region is more likely
to carry nucleotide variability than the more conserved 26S rDNA. Fifty-one isolates were chosen
for ITS1 sequencing in order to test the hypothesis
outlined below (Figure 6).
First, some geographical subpopulations, such
as the Far East and Pennsylvanian isolates of
S. paradoxus, have been sequenced in order to
test whether this sequence is diverging after a
geographical isolation. We have found a unique
Yeast 2005; 22: 177–192.
Repetitive sequences in Saccharomyces
Base pair position
187
27
167
231
357
Sp3 DBVPG 6303-4, Sc DG 1768, Sp YPS4
Sp YPS125
. . A . . . . . . . . . . . . A . . . . . . . . . . - . . . . . .
. . A . . . . . . . . . . . . A . . . G . . . . . . - . . . . . .
Sc
Sc5
Sc
Sc
Sc
Sc
.
.
G
.
.
.
DBVPG 1849 (Hybrid)
DBVPG 6044, YPS 128-9
Y55, SK1
DBVPG 1373
DBVPG 6763-4
(var. boulardii)
DBVPG 6765
.
.
.
.
.
.
A
A
A
A
A
A
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
G
.
.
.
.
.
.
.
.
.
.
.
.
.
C
C
C
C
C
C
.
.
.
.
.
.
.
.
.
.
.
G
A
A
A
A
A
A
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
T
T
T
T
T
T
.
.
.
.
.
.
.
.
.
.
.
.
.
.
G
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
T
-
.
.
.
.
.
.
.
.
.
.
.
.
G
G
G
G
G
G
.
.
.
.
.
.
.
.
.
C
C
C
–
. . A . - - . . T . . G . A . A . . T T . . . . . T - T . G A . .
. . A . - - . . . . . G . A . A . . T T . . . C . T - T . G A . .
. . A . - - . . . . . G . A . A G . T T . . . C . T - T . G A . .
Sp DBVPG 6701 (Hybrid)
Sk7 NBRC 10990-1
. G A G . . T . . . C . . A C A . C T T . . . . C T - T G G A . .
. G A G . . T . . G C . . A C A . C T T C . . . C T - T G G A . .
Sm Sk
Sm NBRC 10992-4, 10999
Sm6 NBRC 10995-6, 10998, 11000-3
Sm NBRC 10997
Sce
A T G A A G A A C A T C T T T G T G C A T T A T T C - A T A G T A
. . . . . . . . . . . . . . . . . . . . . C . . . . - . . . . . .
Sp Sca
Sp1 Different isolates
Sp FAR EAST2
Figure 6. Base position based on the ITS1 sequence of S.paradoxus CBS 432T . Colour patterns indicate taxonomic clusters
on the basis of ITS1 sequence. Strains with GCRs are underlined. 1 100% sequence identity to S. paradoxus CBS 432T . This
group includes: CBS 5829, T21.4, T 32.1, DBVPG 6565, DBVPG 6566, N-17, YPS 158. 2 Far East: N-43, N-44, N-47, N-50.
3
100% S. cariocanus UFRJ 50 816T. 4 YPS138, YPS145, YPS150-2, YPS155. 5 100% S. cerevisiae HA6. 6 100% S. mikatae NBRC
1815T . 7 100% S. kudriavzevii NBRC 1802T
SNP within the ITS1 sequence of Far East (T →
C in position 181; Figure 6, white arrow). This
transition has not been found in any of the other
isolates and seems to be unique to Far East isolates
and correlates with their geographical isolation.
S. paradoxus isolated in Pennsylvania presents
an unexpected result; 6/8 isolates show 100%
sequence homology with the S. cariocanus type
strain. The same result has also been found in S.
paradoxus DBVPG 6303-4 and DG 1768. This is
in good agreement with their atypical hybridization
pattern, such as with the X-element and PAU,
since they were closer to S. cariocanus than
other isolates of S. paradoxus. Interestingly, the
transition in position 44 (Figure 6, black arrow),
G → A, is shared within all sensu stricto isolates
sequenced except S. paradoxus strains outside of
North America.
Among S. cerevisiae isolates we also find variation in ITS1 sequences. In this case, however, the
SNPs do not correlate with geographic location.
Strains SK1 and Y55 share the same SNPs relative
to the S. cerevisiae type strain and this is consistent with their phylogenetic relationship determined
by CGH (Winzeler et al., 2003). The strains previously characterized as S. boulardii have different
SNPs (Figure 6), although they are clearly closer
to the S. cerevisiae ITS1-type sequence. Variation
here has also been detected in S. mikatae. The
ITS1 sequences of the 14 S. mikatae strains fall
into three different groups and reveal three novel
SNPs not present in any of the other strains analysed. We were also able to sequence ITS1 from
Copyright  2005 John Wiley & Sons, Ltd.
two hybrid strains, DBVPG 6701 and DBVPG
1849 (Figure 6). ITS1 of DBVPG 6701 shares
sequences of both parents, whereas DBVPG 1849
has a unique 1 bp gap in position 231.
Finally, all four S. paradoxus strains with obvious GCRs (Figure 1) have been sequenced in order
to test whether these strains are now diverging
within the same geographical subpopulation, as
translocations can lead to partial reproductive isolation. These strains do not appear to be diverging
from their geographical subpopulation, since we
have found an identical ITS1 sequence with their
closest geographically related strain (Figure 6). It
would be interesting to determine whether the
sequence of genes located in regions involved in
these GCRs are now subject to a different rate of
sequence evolution.
Discussion
GCRs and Ty distributions
In S. cerevisiae multiple pathways play an active
role in maintaining genome stability resulting in
low frequencies of spontaneous mutations and
GCRs (Myung et al., 2001). This stability seems to
be active during most evolutionary time including
speciation events, since only a limited number of
rearrangements occurred in the sensu stricto complex (Fischer et al., 2000; Kellis et al., 2003). Several GCRs have previously been characterized in
industrial and wine strains of S. cerevisiae (PerezOrtin et al., 2002; Rachidi et al., 1999). These
Yeast 2005; 22: 177–192.
188
rearrangements arise mainly from ectopic recombination events involving repetitive sequences, such
as Ty elements or tRNAs. With the exception of
S. mikatae, few rearrangements have been found
from the analysis of unrelated geographical subpopulations. These results suggest that the genomes
of wild-type isolates are stable. Moreover, these
hybridization and ITS1 sequencing results do not
show an increasing divergence of strains carrying
GCRs and these strains share similar patterns with
isolates from the same geographical origin. The
unusual situation in S. mikatae has to be further
characterized. One possibility is the abundance of
LTR elements that increase the number of GCRs.
Interestingly, this is the only species where Ty2
elements have been found outside the S. cerevisiae
species (Table 3). Perhaps Ty2 is a recent acquisition by S. mikatae and this has led to genome instability that has yet to stabilize after long term coevolution, but a transfer from S. mikatae to S. cerevisiae is equally likely. Indeed, there is good evidence of horizontal transfer of Ty2 between S. cerevisiae and S. mikatae, as the sequence similarity of
S. mikatae Ty2 and S. cerevisiae Ty2 (95%, determined here) is much greater than Ty1 (85%) and
other sequences (83.6% genome average (Cliften
et al., 2001), indicating that the Ty2s have a more
recent common ancestor than the two species themselves. Another possibility is that S. mikatae either
lost or did not develop efficient genome stability
machinery present in all other sensu stricto species.
This instability seems to result in high variability
in terms of repetitive elements and ITS1 sequence,
despite their common geographical and source origin. This is in contrast with the trend found in S.
paradoxus, where genetic variation strongly correlates with geographical distribution.
The absence of Ty elements in some groups
of geographical isolates leads to interesting ideas
about how this could have arisen. They could either
have been lost or were acquired after speciation
in some geographical lineages. Analysing remnant
LTRs has recently shown that a Ty-less strain of
S. paradoxus lost all transposons (Moore et al.,
2004). These results also suggest abundant presence of one class of Ty that often correlates with
lack or absence of other Ty classes, indicating a
homeostatic control in total number of repetitive
elements (e.g. in the Far East isolates; Table 3).
Copyright  2005 John Wiley & Sons, Ltd.
G. Liti et al.
TAS origins and distributions
The expression of the RNA-helicase encoded in
Y has been detected only in meiosis and in the
absence of telomerase (Louis, 1995; Yamada et al.,
1998). It may be involved in maintaining telomeres
via this expression in the absence of telomerase,
as Y s are used in one survival pathway (Yamada
et al., 1998). Isolate N-44 in the absence of telomerase generates survivors and goes through meiosis
at the same rate as other well-characterized strains,
despite its complete absence of Y s, perhaps indicating that its expression is not essential or functional (Liti and Louis, unpublished results). Furthermore, the phylogenetic inferences made from
locations of these elements allow us to determine
the most likely origin of Y elements, as we can
arrange a series of strains with similar patterns to
show that all strains have one location occupied
by Y , including those with a single unique copy
(Figure 3). The simplest explanation is that this was
the original location and the other sites of occupancy came about through recombinational spread
to the other telomeres, with variation among different lineages. Core X hybridization results outline a
paradoxical situation involving the two most distal
TASs. Y and X element are situated almost in the
same chromosome positions. Core X seems to be
involved in chromatin structure and gene silencing at native ends (Pryde and Louis, 1999) but
its sequence is very divergent, even between different ends of the same strain (Pryde and Louis,
1997). Alternatively, Y s appear to be very well
conserved between strains and species, despite an
apparent lack of function (Louis and Haber, 1992;
Pryde and Louis, 1997). As previously suggested,
duplication within subtelomeric regions of gene
families involved in secondary metabolism (MAL,
MEL, SUC ) could derive from an adaptive response
to different environmental conditions (Pryde et al.,
1995). Using wild-type isolates of S. paradoxus
and S. cariocanus, the gene ancestor of ERR has
been found at the right telomere of chromosome
XV (ERR1 ). These strains of S. paradoxus and S.
cariocanus may have not been exposed to the same
selective pressure as strains of S. cerevisiae. These
results suggest that certain subtelomeric duplications are conserved within the same species but
are not shared by different species.
Yeast 2005; 22: 177–192.
Repetitive sequences in Saccharomyces
189
Speciation in progress
The presence of S. paradoxus in many different environmental conditions could determine the
generation of many variants, as different subpopulations, and generate new species. Some S. paradoxus strains have ITS1 sequences identical to S.
cariocanus and high sequence homology between
these two species has been found (Cliften et al.,
2001). Maybe these strains that share features
with both S. paradoxus and S. cariocanus are
an example of speciation in progress. Likewise,
previous reports have shown partial reproductive
isolation between North American and European
isolates, resulting in a decrease of spore viability of 42–67% (Greig et al., 2003; Sniegowski
et al., 2002). The analysis presented here is consistent with S. cariocanus being a subpopulation
of S. paradoxus, with four reciprocal translocations
Ty5
S. cariocanus
S.p. North America
Ty5
S.p. Far East
Ty1
Ty5
ERR1
S. p. Europe
S.c. var. boulardii
Y (XIV-L)
S.c. S288c
S.c. var. manginii
Ty2
DBVPG 4652
DBVPG 6045
DBVPG1849
S.c. SK1
S.m. NBRC1815
NCYC1379
DBVPG6701
S.m. NBRC10997
PAU
Ty4
S.m. NBRC10992
S.k. NBRC1802
S.k. NBRC1803
S. pastorianus
S. bayanus
HYBRIDS
4.5
4
2
Nucletide substitution (x100)
0
Figure 7. Saccharomyces sensu stricto phylogeny and origins of repetitive elements. The ITS1 sequences can be used to build
a phylogenetic relationship amongst the strains and species analysed in this study. The topology presented is consistent
with previous phylogenies of these species (Fischer et al., 2000). The first occurrence of various repetitive elements can
be mapped onto this phylogeny using the data presented here. The PAU gene family as well as Ty4 elements are found
in all six Saccharomyces sensu stricto species and therefore they most likely existed in the common ancestor of this group.
Y s, on the other hand, are not found in S. bayanus, except in hybrids with S. cerevisiae (Casaregola et al., 2001; Louis et al.,
1994), and are also not found in any more distant yeasts (Jager and Philippsen, 1989; Zakian and Blanton, 1988), therefore
the first occurrence was likely to be in the common ancestor of the non-bayanus sensu stricto species, as it is found in
most isolates of the other five species. Furthermore, the most likely original location can be inferred from the conserved
XIV left occupancy by Y in almost all extant isolates, even those with a single Y . Ty1 and Ty5 are found in very distant
yeast species (Neuveglise et al., 2002). If they entered the sensu stricto group by horizontal transfer, then we can similarly
determine the likely origin in the common ancestor to the group, which split from S. kudriavzevii. Perhaps chromosome IV
and chromosome V are the first locations of Ty1 and Ty5, respectively. For Ty5 we have to invoke a loss of the element in
some subpopulations of S. paradoxus. The ERR gene family first arose on the S. cerevisiae–S. paradoxus–S. cariocanus lineage
and was triplicated in the S. cerevisiae lineage. Ty2 is only found in S. cerevisiae and S. mikatae and the sequences in the two
species are much more similar than other gene divergences, including Ty1. A possible explanation is horizontal transfer
between these two species. Which species first had Ty2 is still to be determined. Within the S. cerevisiae, S. paradoxus and S.
kudriavzevii lineages there is a great deal of divergence between some subpopulations, particularly in the S. paradoxus group.
The Ty element, subtelomeric element and ITS data presented here place S. cariocanus within one of these subpopulations
of S. paradoxus and its designation as a species may have to be revisited. Finally, a number of hybrids between various
species were identified (grey box)
Copyright  2005 John Wiley & Sons, Ltd.
Yeast 2005; 22: 177–192.
190
enhancing reproductive isolation. Further sequence
analysis will test this hypothesis.
Natural hybrids
Furthermore, we have found five strains that
have arisen from hybridization between different
species, excluding S. pastorianus. These results
suggest that natural hybridization within the Saccharomyces sensu stricto complex is possible and
maybe underestimated. The S. pastorianus isolates appear to be the result of independent hybrid
formations, given the level of variation in TAS
and LTR-transposon copy number and location
(Figure 5). At least one of the apparent hybrids,
DBVPG 6701, has a reasonable level of spore viability, which could be due to factors described
above or may be similar to the rare aneuploid
hybrids seen that can yield viable spores (Delneri
et al., 2003). These hybridization events could be
important in the evolution of new gene combinations and species. Taking advantage of the recent
release of sensu stricto genomes, a genome-wide
approach could help to define the importance of
hybridization and horizontal gene transfer in yeast
genome evolution.
Conclusions
In Figure 7 we put forward a likely scenario for
the relationship amongst the Saccharomyces sensu
stricto species. Most strains fall clearly into one
lineage and the distribution of repetitive elements
is generally consistent with the topology defined by
ITS1 sequences. However, it is clear that the tree
is not simple when considering the overall picture.
Several strains within a species exhibit wide variation in these elements. Generally this correlates
with geographic distribution of the isolates. This
is true for S. paradoxus, S. cerevisiae and also S.
kudriavzevii. Five strains share features of divergent lineages and are likely to be hybrids. In a few
cases we can infer where in the phylogenetic tree
a repetitive element first arose, although in at least
one case, Ty2, there is clear evidence for horizontal
transfer between S. cerevisiae and S. mikatae. Even
if the S. paradoxus species lost Ty2 and a common
ancestor of the three species had it, the sequence
homology between Ty2s of the two species is much
higher than expected for the evolutionary distance
Copyright  2005 John Wiley & Sons, Ltd.
G. Liti et al.
between the species, based on other information
including another major LTR-retrotransposon, Ty1.
In general, the phylogeny of Saccharomyces sensu
stricto species is a reasonable and consistent tree,
yet there are clearly some cross-branch interactions
and continuous diversification going on. This collection of strains with natural variation in repetitive
elements and in ITS1 sequences can be used as the
basis of further studies.
Acknowledgements
We are grateful to A. Vaughan-Martini and A. Martini
for helping in selection and a gift of yeast strains; P.
Sniegowski and A. Burt for S. paradoxus North America
and UK isolates respectively. We also thank D. Barton
and N. Guzman for technical help, and A. Contento, K.
Straatman and M. Marvin for comments on the manuscript.
This work was funded by the Wellcome Trust (E.J.L.) and
the BBSRC (I.N.R.).
References
Casaregola S, Nguyen HV, Lapathitis G, Kotyk A, Gaillardin C.
2001. Analysis of the constitution of the beer yeast genome by
PCR, sequencing and subtelomeric sequence hybridization. Int
J Syst Evol Microbiol 51: 1607–1618.
Cliften P, Sudarsanam P, Desikan A, et al. 2003. Finding
functional features in Saccharomyces genomes by phylogenetic
footprinting. Science 301: 71–76.
Cliften PF, Hillier LW, Fulton L, et al. 2001. Surveying Saccharomyces genomes to identify functional elements by comparative DNA sequence analysis. Genome Res 11: 1175–1186.
de Barros Lopes M, Bellon JR, Shirley N J, Ganter PF.
2002. Evidence for multiple interspecific hybridization in
Saccharomyces sensu stricto species. FEMS Yeast Res 1:
323–331.
Delneri D, Colson I, Grammenoudi S, et al. 2003. Engineering
evolution to study speciation in yeasts. Nature 422: 68–72.
Fischer G, James SA, Roberts IN, Oliver SG, Louis EJ. 2000.
Chromosomal evolution in Saccharomyces. Nature 405:
451–454.
Greig D, Louis EJ, Borts RH, Travisano M. 2002. Hybrid
speciation in experimental populations of yeast. Science 298:
1773–1775.
Greig D, Travisano M, Louis EJ, Borts RH. 2003. A role for
the mismatch repair system during incipient speciation in
Saccharomyces. J Evol Biol 16: 429–437.
Groth C, Hansen J, Piskur J. 1999. A natural chimeric yeast
containing genetic material from three species. Int J Syst
Bacteriol 49(4): 1933–1938.
Hunter N, Chambers SR, Louis EJ, Borts RH. 1996. The
mismatch repair system contributes to meiotic sterility in an
interspecific yeast hybrid. EMBO J 15: 1726–1733.
Jager D, Philippsen P. 1989. Many yeast chromosomes lack
the telomere-specific Y sequence. Mol Cell Biol 9:
5754–5757.
Yeast 2005; 22: 177–192.
Repetitive sequences in Saccharomyces
James SA, Collins MD, Roberts IN. 1996. Use of an rRNA
internal transcribed spacer region to distinguish phylogenetically
closely related species of the genera Zygosaccharomyces and
Torulaspora. Int J Syst Bacteriol 46: 189–194.
Johnson LJ, Koufopanou V, Goddard MR, et al. 2004. Population
genetics of the wild yeast Saccharomyces paradoxus. Genetics
166: 43–52.
Kellis M, Patterson N, Endrizzi M, Birren B, Lander ES. 2003.
Sequencing and comparison of yeast species to identify genes
and regulatory elements. Nature 423: 241–254.
Louis EJ. 1995. The chromosome ends of Saccharomyces
cerevisiae. Yeast 11: 1553–1573.
Louis EJ. 1998. Whole chromosome analysis. In Yeast gene
analysis, Brown AJP and Tuite MF (eds). Academic Press: New
York; 15–31.
Louis EJ, Haber JE. 1992. The structure and evolution of
subtelomeric Y repeats in Saccharomyces cerevisiae. Genetics
131: 559–574.
Louis EJ, Naumova ES, Lee A, Naumov G, Haber JE. 1994. The
chromosome end in yeast: its mosaic nature and influence on
recombinational dynamics. Genetics 136: 789–802.
Martini AV, Martini A. 1987. Three newly delimited species of
Saccharomyces sensu stricto. Antonie Van Leeuwenhoek 53:
77–84.
Masneuf I, Hansen J, Groth C, Piskur J, Dubourdieu D. 1998.
New hybrids between Saccharomyces sensu stricto yeast species
found among wine and cider production strains. Appl Environ
Microbiol 64: 3887–3892.
Moore SP, Liti G, Stefanisko KM, et al. 2004. Analysis of a Ty1less variant of Saccharomyces paradoxus: the gain and loss of
Ty1 elements. Yeast 21: (in press).
Myung K, Chen C, Kolodner RD. 2001. Multiple pathways
cooperate in the suppression of genome instability in
Saccharomyces cerevisiae. Nature 411: 1073–1076.
Naumov GI. 1987. Genetic basis for classification and identification of the ascomycetous yeasts. Stud Mycol 30: 469–475.
Naumov GI, James SA, Naumova ES, Louis EJ, Roberts IN.
2000a. Three new species in the Saccharomyces sensu
stricto complex: Saccharomyces cariocanus, Saccharomyces
kudriavzevii and Saccharomyces mikatae. Int J Syst Evol
Microbiol 50(5): 1931–1942.
Naumov GI, Naumova ES, Hagler AN, Mendonca-Hagler LC,
Louis EJ. 1995a. A new genetically isolated population of the
Saccharomyces sensu stricto complex from Brazil. Antonie Van
Leeuwenhoek 67: 351–355.
Naumov GI, Naumova ES, Lantto RA, Louis EJ, Korhola M.
1992. Genetic homology between Saccharomyces cerevisiae and
its sibling species S. paradoxus and S. bayanus: electrophoretic
karyotypes. Yeast 8: 599–612.
Naumov GI, Naumova ES, Louis EJ. 1995b. Two new genetically
isolated populations of the Saccharomyces sensu stricto complex
from Japan. J Gen Appl Microbiol 41: 499–505.
Naumov GI, Naumova ES, Masneuf I, et al. 2000b. Natural
polyploidization of some cultured yeast Saccharomyces sensu
stricto: auto- and allotetraploidy. Syst Appl Microbiol 23:
442–449.
Naumov GI, Naumova ES, Sniegowski PD. 1997. Differentiation
of European and Far East Asian populations of Saccharomyces
paradoxus by allozyme analysis. Int J Syst Bacteriol 47:
341–344.
Copyright  2005 John Wiley & Sons, Ltd.
191
Naumov GI, Naumova ES, Sniegowski PD. 1998. Saccharomyces
paradoxus and Saccharomyces cerevisiae are associated with
exudates of North American oaks. Can J Microbiol 44:
1045–1050.
Naumov GI, Nikonenko TA, Kondrat’eva VI. 1994. [Taxonomic
identification of Saccharomyces from yeast genetic stock
centers of the University of California]. Genetika 30:
45–48.
Neuveglise C, Feldmann H, Bon E, Gaillardin C, Casaregola S.
2002. Genomic evolution of the long terminal repeat
retrotransposons in hemiascomycetous yeasts. Genome Res 12:
930–943.
Perez-Ortin JE, Querol A, Puig S, Barrio E. 2002. Molecular
characterization of a chromosomal rearrangement involved
in the adaptive evolution of yeast strains. Genome Res 12:
1533–1539.
Pryde FE, Huckle TC, Louis EJ. 1995. Sequence analysis of the
right end of chromosome XV in Saccharomyces cerevisiae: an
insight into the structural and functional significance of subtelomeric repeat sequences. Yeast 11: 371–382.
Pryde FE, Louis EJ. 1997. Saccharomyces cerevisiae telomeres. A
review. Biochemistry (Mosc) 62: 1232–1241.
Pryde FE, Louis EJ. 1999. Limitations of silencing at native yeast
telomeres. EMBO J 18: 2538–2550.
Rachidi N, Barre P, Blondin B. 1999. Multiple Ty-mediated
chromosomal translocations lead to karyotype changes in a
wine strain of Saccharomyces cerevisiae. Mol Gen Genet 261:
841–850.
Rachidi N, Martinez MJ, Barre P, Blondin B. 2000. Saccharomyces cerevisiae PAU genes are induced by anaerobiosis. Mol
Microbiol 35: 1421–1430.
Sniegowski PD, Dombrowski PG, Fingerman E. 2002. Saccharomyces cerevisiae and Saccharomyces paradoxus coexist in
a natural woodland site in North America and display different levels of reproductive isolation from European conspecifics.
FEBS Yeast Res 1: 299–306.
Thompson JD, Higgins DG, Gibson TJ. 1994. CLUSTAL W:
improving the sensitivity of progressive multiple sequence
alignment through sequence weighting, position-specific gap
penalties and weight matrix choice. Nucleic Acids Res 22:
4673–4680.
Vaughan Martini A. 1989. Saccharomyces paradoxus comb. nov.,
a newly separated species of the Saccharomyces sensu stricto
complex based upon nDNA/nDNA homologies. System Appl
Microbiol 12: 179–182.
Vaughan-Martini A, Martini A. 1995. Facts, myths and legends
on the prime industrial microorganism. J Ind Microbiol 14:
514–522.
Viswanathan M, Muthukumar G, Cong YS, Lenard J. 1994.
Seripauperins of Saccharomyces cerevisiae: a new multigene
family encoding serine-poor relatives of serine-rich proteins.
Gene 148: 149–153.
White TJ, Bruns T, Lee S, Taylor J. 1990. Amplification and direct
sequencing of fungal ribosomal RNA genes for phylogenetics. In PCR Protocols: A Guide for Methods and Applications,
Innis MA, Gelfand DH, Sninsky JJ, Taylor TJ (eds). Academic
Press: New York; 315–322.
Winzeler EA, Castillo-Davis CI, Oshiro G, et al. 2003. Genetic
diversity in yeast assessed with whole-genome oligonucleotide
arrays. Genetics 163: 79–89.
Yeast 2005; 22: 177–192.
192
Yamada M, Hayatsu N, Matsuura A, Ishikawa F. 1998. Y -Help1,
a DNA helicase encoded by the yeast subtelomeric Y element,
is induced in survivors defective for telomerase. J Biol Chem
273: 33 360–33 366.
Zakian VA, Blanton HM. 1988. Distribution of telomereassociated sequences on natural chromosomes in Saccharomyces
cerevisiae. Mol Cell Biol 8: 2257–2260.
Copyright  2005 John Wiley & Sons, Ltd.
G. Liti et al.
Zou S, Wright DA, Voytas DF. 1995. The Saccharomyces Ty5
retrotransposon family is associated with origins of DNA
replication at the telomeres and the silent mating locus HMR.
Proc Natl Acad Sci USA 92: 920–924.
Yeast 2005; 22: 177–192.