SED1 polymorphism within the genus Saccharomyces

Abstract

The SED1 gene is characterised by abundant length and sequence polymorphisms within the species Saccharomyces cerevisiae, due to the expansion and contraction of minisatellite-like sequences located within the ORF. A survey of the SED1 ORFs of 26 yeasts ascribed to the species S. cerevisiae, S. bayanus, S. pastorianus, S. paradoxus, S. cariocanus, S. kudriavzevii and S. mikatae revealed SED1 gene length and sequence variations between the species of the genus. Moreover, results obtained by Neighbour-Joining analysis of a dataset comprising the partial predicted amino acid sequences of SED1 ORFs agreed with the phylogenetic relationships of the seven species. Thus, the SED1 gene may represent a further molecular target for the identification of Saccharomyces isolates.

SED1, Saccharomyces, Gene length polymorphism, PCR-RFLP, Yeast identification

1 Introduction

The genus Saccharomyces, which includes strains important in yeast-based industries, consists of seven species, namely Saccharomyces cerevisiae, S. paradoxus, S. pastorianus, S. bayanus, S. cariocanus, S. kudriavzevii and S. mikatae[1]. Of these, S. cariocanus, S. kudriavzevii and S. mikatae have recently been included within the Saccharomyces sensu stricto complex [2], and S. pastorianus has been recognised as a hybrid between S. cerevisiae and S. bayanus or the brewing yeast S. monacensis[3–8].

Results obtained in different studies have indicated that the species of the genus are phylogenetically closely related. Electrophoretic karyotyping [9], mtDNA restriction analysis [10], ribosomal DNA sequence analysis [11] and RFLP of ITS and IGS spacers [12] have all shown S. cerevisiae and S. paradoxus to be highly-related species. The use of RAPD-PCR has confirmed the close relationship between S. cerevisiae and S. paradoxus on one side, and between S. bayanus and S. pastorianus on the other [13]. S. cariocanus is recognised as being closely related to S. cerevisiae and S. paradoxus[2,14], while S. kudriavzevii and S. mikatae are included in separate clusters [14] and S. bayanus is considered to be the most divergent species [2,4,12,15,16].

The yeasts belonging to the genus Saccharomyces also occupy common ecological niches. This is particularly true for strains belonging to the species S. cerevisiae and S. paradoxus which colonise natural environments and can be isolated from exudates of broad-leaved trees, soils and insects [17], and from oak-associated soils [18], but also co-exist in man-made environments, such as vineyards [19]. Moreover, S. cerevisiae co-occurs with S. bayanus in fruit juice and winery environments, and with S. pastorianus in beer [20].

According to several authors the cultivated yeast S. cerevisiae is the most heterogeneous among the species of the genus [14,21]. In contrast, if compared to S. cerevisiae, S. paradoxus is characterised by intraspecific homogeneity as shown by enzyme electrophoresis [21,22] and by a slower evolution of coding regions [23], even though microsatellite fingerprinting and RAPD analysis has revealed genetic variability among different strains of this species [24].

We have previously observed that S. cerevisiae genes encoding cell wall proteins are highly polymorphic in length [25,26]. In particular the SED1 gene is characterized by abundant length and sequence polymorphism within this species, due to the expansion and contraction of minisatellite-like sequences within two distinct regions of the gene [25].

With the aim of gaining information on the inter- and intra-specific sequence variations of the SED1 gene within the genus Saccharomyces, also in view of its possible use as a further molecular marker suitable for yeast identification, we analysed the SED1 ORFs of twenty-six strains ascribed to the species S. cerevisiae, S. paradoxus, S. bayanus, S. cariocanus, S. kudriavzevii, S. mikatae and S. pastorianus.

2 Materials and methods

2.1 Strains and media

The 26 Saccharomyces strains analysed are given in Table 1. They were all cultivated in YEPD (2% glucose, 1% yeast extract, 2% peptone) solidified with agar (2%) when needed, and incubated at 25 °C.

1

The Saccharomyces strains used

Species	Strain designation
S. bayanus	DBVPG 6348	NCYC 374-2
S. bayanus	DBVPG 6730
S. bayanus	DBVPG 6131	CBS 431
S. bayanus	DBVPG 6179	CBS 395
S. bayanus	DBVPG 6253	CBS 1546
S. bayanusa	DBVPG 6171	CBS 380
S. bayanus	DBVPG 6259	CBS 1604
S. bayanus	DBVPG 6260	CBS 1505
S. pastorianus	DBVPG 6282	Carlsberg BK 2233
S. pastorianus	DBVPG 6283	Carlsberg BK 2230
S. pastorianus	DBVPG 6284	Carlsberg AJL 248
S. pastorianus	DBVPG 6258	CBS 1486
S. pastorianusa	DBVPG 6261	CBS 1503
S. pastorianus	DBVPG 6033	CBS 1513
S. pastorianusa	DBVPG 6047	CBS 1538
S. pastorianus	DBVPG 6257	CBS 1260
S. paradoxus	DBVPG 6489
S. paradoxus	DBVPG 6490
S. paradoxus	DBVPG 6491
S. paradoxus	DBVPG 6304	UCD 51-186
S. paradoxus	DBVPG 6466	CBS 5829
S. paradoxusa	DBVPG 6411	CBS 432
S. cerevisiaea	DBVPG 6173	CBS 1171
S. mikataea		NCYC 2888a
S. kudriavzeviia		NCYC 2889a
S. cariocanusa		NCYC2890a

Culture collections are abbreviated as follows: DBVPG, Dipartimento di Biologia Vegetale, Università di Perugia, Italy; NCYC, National Collection of Yeast Cultures, Institute for Food Research, Norwich, UK; CBS, Centraal bureau voor Schimmelcultures, Utrecht, The Netherlands; Carlsberg, Department of Yeast Genetics, Carlsberg Laboratory, Copenhagen, Denmark; UCD, Department of Food Science and Technology, University of California, Davis, California, USA.

a

Type strain.

2.2 DNA extraction, PCR conditions and restriction analyses

Total genomic DNA for PCR analysis was isolated from 24 h old cultures as described by Ushinsky et al. [27]. The PCR primers were SED1FOR (5′-ATGAAATTATCAACTGTCCTATTATCTGCCGG-3′) and SED1REV (5′-TTATAAGAATAACATAGCAACACCAGCCAAACC-3′) [25]. The PCR amplification reactions were performed on a Gene AMP PCR System 9700 (Perkin–Elmer, Boston, MA, USA) in 25 μl reaction mixture containing 2 μl template DNA (containing approximately 10 ng μl⁻¹), 0.6 U Taq polymerase (Amersham Biosciences, Piscataway, NJ, USA), 1× reaction buffer [MgCl₂ free], 3 mM MgCl₂, 200 μM of each dNTP, and 30 pmoles each of SED1FOR and SED1REV primers. The reactions were performed as follows: 25 cycles “touch-up PCR” with denaturation at 94 °C for 1 min, an initial annealing temperature of 52 °C (with an increase in the annealing temperature of 0.5 °C per cycle), and elongation at 72 °C for 45 s. The touch-up cycles were followed by 10 cycles with an annealing temperature of 64 °C. An initial denaturation step at 94 °C for 5 min, and a final 7 min extension at 72 °C were used. The PCR products were analysed by electrophoresis using 1.5% agarose gels in 0.5× TBE buffer. The gel images were visualized using a BioRad Gel DOC 1000, and acquired using the Multi-Analyst software (BioRad Laboratories Inc., Nazareth Eke, Belgium). Twenty μl of the PCR products were digested overnight with excess HpaII or KpnI in a final volume of 50 μl. The restriction fragments were analysed in 2% agarose gels in 0.5× TBE buffer.

2.3 DNA sequencing and sequence analyses

The PCR products were purified using GFX™ PCR DNA and the Gel Band Purification Kit (Amersham Biosciences, Piscataway, NJ, USA) and sent for sequencing to MWG Biotech S.r.l. (Florence, Italy). The partial amino acid sequences were deduced by translation of the partial nucleotide sequences in all of the possible reading frames. The primary sequence alignment was performed using Clustal View (1.82), available at http://www.ebi.ac.uk/Tools/. Minor adjustments of the output alignment were performed by hand. The nucleotide sequences of S. bayanus SED1-1 and SED1-2, S. paradoxus SED1, S. mikatae SED1, S. kudriavzevii SED1, S. cariocanus SED1, S. pastorianus SED1-1 and SED1-2 were deposited in GenBank with Accession Nos: AY589777, AY589778, AY589779, AY589780, AY589781, AY589782, AY601036 and AY601037, respectively. The GenBank Accession No. for S. cerevisiae SED1-1 was previously reported [25].

2.4 Cluster analysis

Cluster analysis was conducted using PAUP^∗4.0b10 (http://www.paup.csit.fsu.edu). The unrooted tree was obtained by neighbour-joining analysis [28] based on the mean character difference of the dataset of the partial amino acid sequences predicted from the nucleotide sequences. Bootstrap values were based on 1000 replications.

3 Results and discussion

3.1 SED1 analysis within the genus Saccharomyces

Six to eight strains belonging to the species S. paradoxus, S. bayanus and S. pastorianus were analysed with the aim to study the variation of the SED1 gene in species of the genus Saccharomyces which most frequently occur in natural and man-made environments. In addition, to reveal the presence of possible interspecific SED1 polymorphisms, the type strains of S. cariocanus, S. kudriavzevii and S. mikatae[2] and S. cerevisiae were also included (Table 1).

The PCR analysis used to study variation in the SED1 gene required the development of a “touch-up PCR” protocol which, by means of a gradual increase of the annealing temperature in the first 25 cycles, allows the primer to anneal to the target sequence even when there is not full homology between primer and target sequences. As shown in Fig. 1, two different PCR profiles were observed within S. bayanus. One profile, occurring in strains DBVPG 6253 and DBVPG 6260, consists of two length variants, named SbSED1-1 (short) and SbSED1-2 (long). The coexistence of two SED1 alleles of different length in the two S. bayanus strains may be due to heterozygosity for this gene, similar to what has been observed in S. cerevisiae wine isolates [25]. The other PCR profile of S. bayanus is made up of a single amplification product as can be expected in individuals homozygous for this gene [25]. This single amplification product, showing the same size as SbSED1-2, was observed in the type strain of S. bayanus CBS380 (DBVPG 6171) and in the remaining S. bayanus strains (Fig. 1).

1

SED1 gene polymorphism in the species of the genus Saccharomyces. PCR primers designed on the SED1 sequence were used to amplify the SED1 genes of: S. bayanus DBVPG 6253 and DBVPG 6260 (lane 1); S. bayanus CBS 380 (DBVPG 6171), DBVPG 6348, DBVPG 6730, DBVPG 6131, DBVPG 6179 and DBVPG 6259 (lane 2); S. pastorianus DBVPG 6258 (lane 3); S. pastorianus DBVPG 6047, DBVPG 6282, DBVPG 6283, DBVPG 6284, DBVPG 6261, DBVPG 6033 and DBVPG 6257 (lane 4); S. paradoxus DBVPG 6489, DBVPG 6490, DBVPG 6491, DBVPG 6304, DBVPG 6466, DBVPG 6411 (lane 5); S. cerevisiae (lane 6), S. mikatae (lane 7); S. kudriavzevii (lane 8); S. cariocanus (lane 9); L: 50-bp DNA Ladder (Amersham-Pharmacia, Piscataway, NJ, USA).

Similarly, the S. pastorianus strains investigated in this study harboured two different PCR profiles. One, made up of a single SED1 amplicon, was present in strain DBVPG 6258. The other PCR profile consisting of two SED1 length variants, named SpaSED1-1 (short) and SpaSED1-2 (long), occurred in the type strain DBVPG 6047 and in DBVPG 6261 (syn. S. monacensis), DBVPG 6033 (syn. S. carlsbergensis) and the remaining S. pastorianus strains. SpaSED1-1 showed the same length as SbSED1-1 and SpaSED1-2 had the same length as the allele of S. cerevisiae. Thus, apart from strain DBVPG 6258, all the S. pastorianus strains analysed seemed to harbour S. cerevisiae and S. bayanus alleles, which is in accordance with the hybrid nature of this taxon [8,29–31].

The type strain of S. cerevisiae CBS1171 yielded a single amplicon of 975 bp corresponding to the SED1-1 allele [25] which, although it is the shortest S. cerevisiae allele sequenced to date from a large collection of laboratory and wine strains [25,26], is clearly longer than that of S. bayanus, S. paradoxus, S. cariocanus, S. mikatae and the shortest known allele of S. pastorianus (Fig. 1).

No length polymorphism was observed between the six S. paradoxus strains which produced a single amplicon, named SpxSED1, measuring approximately 950 bp (Fig. 1). The lack of variation of the SED1 gene in this species is in accordance with the findings of other authors [21,22], who have shown that S. paradoxus is characterised by a high intraspecific homogeneity as evidenced by enzyme electrophoresis. Moreover, this result is in agreement with the indications of Kellis et al. [23], who have found that coding regions of S. paradoxus evolved slower than those of S. cerevisiae.

The type strains of the species S. mikatae, S. cariocanus and S. kudriavzevii, included by Naumov and coworkers [2] within the Saccharomyces sensu stricto complex, produced single SED1 amplicons measuring from 900 to 1000 bp. These are longer than the two alleles of S. bayanus (SbSED1-1 and SbSED1-2) and the shorter allele of S. pastorianus (SpSED1-1).

Thus, the SED1 gene is characterised by length variations within the genus Saccharomyces, and a PCR-based analysis of SED1 length polymorphism allows discrimination between the most common species in natural and man-made environments, namely S. cerevisiae, S. pastorianus, S. paradoxus and S. bayanus.

3.2 RFLP analysis of the SED1 amplicons

To further investigate the structure of the gene, the SED1 amplification products were digested with the enzymes HpaII and KpnI. In the case of the S. cerevisiae SED1 gene, these two enzymes can be conveniently used to delimit the regions of the gene subjected to length variations [25].

According to the results obtained, the HpaII restriction sites are not conserved in the SED1 genes of the species ascribed to the genus Saccharomyces. Indeed, HpaII cut the S. cerevisiae and the S. kudriavzevii SED1 genes, and was unable to digest those of S. paradoxus, S. bayanus, S. mikatae and S. cariocanus. In the case of S. pastorianus, SpaSED1-1 remained uncut, as expected, due to its hypothesized homology with SbSED1-1, whereas the PCR profile consisting of SpaSED1-1 and SpaSED1-2 exhibited a mixed restriction pattern. SpaSED1-1 was not digested and SpaSED1-2 showed the same restriction profile as S. cerevisiae SED1-1 (Fig. 2). Thus, HpaII digestion confirmed the hybrid nature of S. pastorianus.

2

HpaII restriction analysis of the SED1 gene. S. bayanus DBVPG 6253 and DBVPG 6260 (lane 1); S. bayanus CBS 380 (DBVPG 6171), DBVPG 6348, DBVPG 6730, DBVPG 6131, DBVPG 6179 and DBVPG 6259 (lane 2); S. pastorianus DBVPG 6258 (lane 3); S. pastorianus DBVPG 6047, DBVPG 6282, DBVPG 6283, DBVPG 6284, DBVPG 6261, DBVPG 6033 and DBVPG 6257 (lane 4); S. paradoxus DBVPG 6489, DBVPG 6490, DBVPG 6491, DBVPG 6304, DBVPG 6466, DBVPG 6411 (lane 5); S. mikatae (lane 6); S. kudriavzevii (lane 7); S. cariocanus (lane 8); S. cerevisiae (lane 9). L: GeneRuler™ 100-bp DNA Ladder (MBI Fermentas GmbH, St. Leon-Rot, Germany); L1: GeneRuler™ 50-bp DNA Ladder (MBI Fermentas GmbH, St. Leon-Rot, Germany).

The results obtained by the use of KpnI were more informative. Indeed, the use of this enzyme confirmed the occurrence of SED1 gene variations between the species of the genus. Moreover, the restriction profiles of the seven Saccharomyces species analysed differed from each other, thus allowing the discrimination between all species analysed (Fig. 3).

KpnI restriction analysis of the SED1 gene. S. cerevisiae (lane 1); S. bayanus CBS380 (DBVPG 6171), DBVPG 6348, DBVPG 6730, DBVPG 6131, DBVPG 6179 and DBVPG 6259 (lane 1); S. bayanus DBVPG 6253 and DBVPG 6260 (lane 3); S. pastorianus DBVPG 6047, DBVPG 6282, DBVPG 6283, DBVPG 6284, DBVPG 6261, DBVPG 6033 and DBVPG 6257 (lane 4); S. pastorianus DBVPG 6258 (lane 5); S. paradoxus DBVPG 6489, DBVPG 6490, DBVPG 6491, DBVPG 6304, DBVPG 6466, DBVPG 6411 (lane 6); S. mikatae (lane 7); S. cariocanus (lane 8); S. kudriavzevii (lane 9). L: GeneRuler™ 100-bp DNA Ladder (MBI Fermentas GmbH, St. Leon-Rot, Germany).

3.3 Sequence polymorphism of the SED1 gene

To further understand the results of the restriction analysis and to obtain a detailed structure of the coding region of the gene present in the seven Saccharomyces species, the PCR-amplified SED1 alleles were sequenced and the predicted amino acid sequences subjected to multiple alignment (Fig. 4).

4

Multiple alignment of partial predicted amino acid sequences of the SED1 alleles. The repeat sequences are underlined. Dashes indicate gaps introduced to optimise the alignment; dots indicate identical residues. The beginning and the end of each repeat unit are indicated by arrows (↓).

The SED1 alleles of the species analysed in this work presented minisatellite-like sequences like in S. cerevisiae[25]. Moreover, it was observed that the general structure of the gene is conserved within the genus (Fig. 4). However, the S. bayanus alleles SbSED1-1 and SbSED1-2 presented large deletions within the repeat containing part of the gene. In addition, SbSED1-1 was characterized by an extensive deletion located downstream of the region containing the repeat sequences. These deletions are responsible, at least in part, for the short size of the SbSED1-1 allele (Fig. 4).

Alignment of partial amino acid sequences showed that SpaSED1-1 and SpaSED1-2 are identical to SbSED1-1 and S. cerevisiae SED1-1, respectively, thus confirming the hybrid nature of S. pastorianus. Indeed, SpaSED1-1, which is identical to SbSED1-1 and present in the homozygous or heterozygous state in all the S. pastorianus strains tested, appeared to be not very common in S. bayanus, as it was found in only two of the strains analysed. These two sequence types could have been involved in the generation of the hybrid species S. pastorianus.

The predicted partial amino acid sequence of SED1 in S. cerevisiae showed the highest homology with those of S. paradoxus and S. cariocanus (90% and 88%, respectively), which is in accordance with the close relationship between these taxa, as was shown by Kurtzman and Robnett [32] on the basis of multigene sequence analysis. Lower levels of similarity were observed between the sequences of S. kudriavzevii and S. mikatae (78%) and between these two sequences and that of S. cerevisiae (75% and 85%, respectively). The two alleles of S. bayanus, which is the most divergent species within the genus [2,4,12,15,16], had the lowest level of identity with all the others (data not shown).

The unrooted tree derived from neighbour-joining (NJ) analysis of the partial predicted amino acid sequence dataset of the nine SED1 alleles is in accordance with these results (Fig. 5). Indeed, SbSED1-1 and SbSED1-2 are the two alleles showing the highest mean character difference and are therefore located in a separate cluster. Moreover, the two S. pastorianus alleles, due to their complete identity with S. cerevisiae SED1-1 and SbSED1-1, are located in the two clusters containing these alleles.

5

neighbour-joining tree based on the comparison of the partial predicted amino acid sequences of the SED1 alleles. Bootstrap values based on 1000 replications are given in brackets at the nodes.

In conclusion, contrarily to what has been observed for S. cerevisiae[25,26], the SED1 gene showed limited length polymorphism within the species S. bayanus, S. pastorianus and S. paradoxus. However, significant SED1 length and sequence polymorphisms were observed between all the species of the genus Saccharomyces. A simple PCR analysis of the SED1 gene allowed the clear discrimination between S. cerevisiae, S. bayanus, S. pastorianus and S. paradoxus, which occur most frequently in natural and man-made environments. A further KpnI restriction analysis of the SED1 amplicons, however, led to the discrimination between all species occurring within the genus.

Moreover, the NJ tree based on the partial amino acid sequences of the nine alleles was in accordance with suggested phylogenetic relationships among the species within the genus [2,4,12–16]. Thus, the SED1 gene can be used as a further molecular marker suitable for typing and clustering of Saccharomyces isolates.

Acknowledgements

The authors thank Madan Thangavelu for critical reading of the manuscript and Gianluigi Cardinali for useful suggestions.