当前位置: 首页 > 期刊 > 《分子生物学进展》 > 2005年第4期 > 正文
编号:11176539
Comparative Genomics in Hemiascomycete Yeasts: Evolution of Sex, Silencing, and Subtelomeres
http://www.100md.com 《分子生物学进展》
     Unité de Génétique Moléculaire des Levures, (URA2171 CNRS, UFR Université Pierre et Marie Curie), Département Structure et Dynamique des Génomes, Institut Pasteur, 75724 Cedex Paris, France

    Correspondence: E-mail: efabre@pasteur.fr; cfair@pasteur.fr

    Abstract

    The recent release of sequences of several unexplored yeast species that cover an evolutionary range comparable to the entire phylum of chordates offers us a unique opportunity to investigate how genes involved in adaptation have been shaped by evolution. We have examined how three different sets of genes, all related to adaptative processes at the genomic level, have evolved in hemiascomycetes: (1) the mating-type genes that govern sexuality, (2) the silencing genes that are connected to regulation of mating-type cassettes and to telomere position effect, and (3) the gene families found repeated in subtelomeric regions.We report new combinations of mating-type genes and cassettes in hemiascomycetous species; we show that silencing proteins diverge rapidly. We have also found that in all species studied, subtelomeric gene families exist and are specific to each species.

    Key Words: mating type ? gene families ? silencing of chromatin ? subtelomeres ? hemiascomycetes

    Introduction

    The question of whether the gene content of a genome can be correlated to adaptative properties can be addressed by comparing genome sequences from species with different lifestyles. In addition to partial genome sequences of closely related Saccharomyces species (Cliften et al. 2003; Kellis et al. 2003), the complete sequences of six unexplored yeast species (i.e., Candida glabrata, Kluyveromyces waltii, Kluyveromyces lactis, Ashbya gossypii, Debaryomyces hansenii and Yarrowia lipolytica) are now available (Dietrich et al. 2004; Dujon et al. 2004; Kellis, Birren, and Lander 2004). These species cover a vast range of lifestyles and exhibit a variety of life cycles and mating mechanisms (table 1).

    Table 1 Habitats and Life Cycles of Yeasts Presented Here

    We have examined, in the new hemiascomycete genomic sequences available, three different sets of genes, all related to adaptative processes at the genomic level. First, we analyzed genes involved in sexual reproduction because of the major role played by sex, or the loss of it, in a species' evolution. Second, we analyzed genes involved in silencing, because the silencing of sexual genes is an important feature in S. cerevisiae's and other yeasts' sexual cycles and because silencing is involved in regulation of gene expression that is linked to an organism's direct adaptation to the environment. Third, we studied subtelomeric genes because of the plasticity of these regions and their capacity for harboring large gene families and because expression of some genes in subtelomeres is regulated through silencing mechanisms (Ai et al. 2002; De Las Penas et al. 2003; Halme et al. 2004).

    Most ascomycetes have only two different mating types, as opposed to certain other fungi, in which many more mating types exist (Coppin et al. 1997). In all ascomycetes, the MAT locus encodes transcription factors that regulate mating-type–specific genes involved in pheromone production, pheromone sensing, and signal transduction (Fraser and Heitman 2004). Haploid-specific gene products are involved in repression of meiosis and mating-type switching, such as the HO endonuclease, at least in S. cerevisiae (Herskowitz and Oshima 1981; Haber 1998). In the case of S. cerevisiae and Schizosaccharomyces pombe, a very distant ascomycete species, homothallism is caused by gene conversion between the MAT locus and two MAT-like loci during cellular division of haploid cells (Herskowitz, Rine, and Strathern 1992; Klar 1992; Haber 1998).

    The duplicated MAT-like cassettes must be transcriptionally repressed, and this is achieved through a process called silencing, which consists of the formation of a specialized compacted chromatin structure. In S. cerevisiae, the MAT-like cassettes, HMR and HML, are both surrounded by "silencers," short specific sequences that are binding sites for DNA-binding proteins and are also involved in transcriptional activation and DNA replication (for recent reviews, see Gasser and Cockell [2001]. Grewal and Moazed [2003], and Rusche, Kirchmaier, and Rine [2003]). The other known proteins involved in this process are the SIR proteins, Sir1p through Sir4p, for which an enzymatic activity has only been described for the deacetylase Sir2p. Deacetylation of amino-terminal tails of histones (H3 and H4) can lead to nuclesome compaction correlated with transcriptional repression.

    Silencing mechanisms have also been described in the regions close to telomeres and in telomeric repeats themselves. Such mechanisms exhibit the telomere position effect (TPE; for review, see Tham and Zakian [2002]). TPE is a repressor effect observed when reporter genes are inserted in a terminal position in a chromosome. It also occurs, to different extents, at about half of natural telomeres (Pryde and Louis 1999). In S. cerevisiae, chromosome ends consist of telomerase-dependent (C1-3A) imperfect repeats, upstream of which several particular features are found. A retrotransposon-like Y' element is present in two third of the chromosomes in strain S288C and a 500-bp core X element is found at all chromosome ends (Louis and Haber 1992; Louis 1995). In natural telomeres, silencing involves binding to the core X domain of almost the same proteins that also bind to MAT-like silent cassettes (Pryde and Louis 1999). Additionnal proteins that bind telomeric repeats, such as the Ku complex, participate in TPE (Laroche et al. 1998). Subsequent to the binding of the Sir complex and hypoacetylation of amino-terminal tail of histone H3, nucleosome compaction spreads along the chromatin up to 5 to 8 kb from telomeres (Kimura, Umehara, and Horikoshi 2002; Suka, Luo, and Grunstein 2002). Histone acetyltransferase activities, as well as displacement of nucleosomes, are supposed to counteract such a spreading (Rusche, Kirchmaier, and Rine 2003).

    We have further analyzed the gene content in larger subtelomeric regions than those subject to TPE. Recent characterization of subtelomeric regions from a variety of organisms from yeast to man has led to the realization that many chromosome ends are similar in structure (Mefford and Trask 2002). These subtelomeric regions correspond to about 30 kb of AT-rich, gene-poor sequences that contain no essential genes but contain some unique genes and genes present in multiple copies in other subtelomeric region. These subtelomeric regions are compacted in a heterochromatin domain extending from 10 to 25 kb from the telomere end (Robyr et al. 2002; Martin et al. 2004). Heterochromatin-like structure is then dependent on hypoacetylation and methylation of H3 but not on the Sir complex (Wyrick et al. 1999; Robyr et al. 2002; Martin et al. 2004). Industrial and "wild" yeasts are polymorphic for characters such as chromosome sizes and often heterozygous for sugar-utilization genes (Johnston, Baccari, and Mortimer 2000; Carro et al. 2003). These phenomena have been linked to plastic subtelomeric regions, which maintain gene families that are positively selected for in various brewing or baking strains of yeast.

    By comparing genomic evolution of mating-type cassettes, we were able to describe further the evolutionary relationship between heterothallism and homothallism, corroborating the paradigm that heterothallism is the ancestral mode of sexual reproduction in ascomycetes (Coppin et al. 1997). We confirm that changes in modes of sexual reproduction are frequent when hemiascomycetous species adapt, indicating that sexual dysfunction is tolerated, possibly leading to complete loss of sexuality. We also show that characteristics of subtelomeric regions are probably shared widely among hemiascomycetes, although subtelomeres in each species present some specificities. As for silencing, we show that the evolutionary divergence of silencing mechanisms is far greater than the genes that are silenced themselves, thus, reinforcing the notion that transcription factors diverge faster than other proteins.

    Materials and Methods

    Genomes Analyzed in This Work

    Six genomes of non- Saccharomyces hemiascomycetes were compared with S. cerevisiae in this work, four genomes from the Genolevures II effort (Dujon et al. 2004) (Candida glabrata, Kluyveromyces lactis, Debaryomyces hansenii and Yarrowia lipolytica) and two others (Kluyveromyces waltii [Kellis, Birren, and Lander 2004] and Ashbya gossypii [Dietrich et al. 2004]). For some analyses, partial sequences of six Saccharomyces genomes (Cliften et al. 2003; Kellis et al. 2003) (S. paradoxus, S. mikatae, S. kudriavzevii, S. bayanus, S.castellii, S. kluyveri), partial unpublished K. thermotolerans sequences (E. Talla et al., 2005), and partial Z. rouxii from the Genolevures I program (Souciet et al. 2000) were also examined. C. albicans data is from Magee and Magee (2000), Tzung et al. (2001), and our own Blast searches. The phylogenetic relationship between species is represented in figures and tables by the order of their names in rows and columns.

    Comparative Gene Analysis

    Analyses pertaining to genes encoding proteins involved in mating and in silencing and to subtelomeric gene families were performed by Blast searches using S. cerevisiae sequences as queries, and hits were usually found within annotated genes. To search for specific subtelomeric families, intraspecies Blast results for individual subtelomeric gene were examined. Genes in MAT loci and genes encoding pheromones, being smaller than average and sometimes containing introns, have sometimes been overlooked in annotation. Their identification is detailed below.

    S. cerevisiae query sequences were retrieved from the Saccharomyces Genome Database (SGD, http://www.yeastgenome.org/). Most Blast searches, annotation and synteny data examination were done at SGD for sensu stricto and sensu latto Saccharomyces and for A. gossypii. Searches were done at the Genolevures site (http://cbi.labri.fr/Genolevures/) for Candida glabrata, Kluyveromyces lactis, Debaryomyces hansenii, and Yarrowia lipolytica and for Genolevures I and C. albicans sequences. Were also used the A. gossypii Database (http://data.cgt.duke.edu/ashbya/Blast.html, http://agd.unibas.ch/), partial genomic sequences of S. mikatae, S. kudriavzevii, S. bayanus, S. castellii, S. kluyveriat from the Genome Sequencing Center Database (http://genome.wustl.edu/projects/yeast/), and Blast on fungi at NCBI (http://www.ncbi.nlm.nih.gov/sutils/genom_table.cgi?organism=fungi). For K. waltii, contig and protein sequences were downloaded from (http://www.nature.com/nature/journal/v428/n6983/extref/nature02424-s1.htm) and analyzed locally. Depending on the site used and the type of sequences, BlastP, BlastN, tBlastN, or tBlastX was used, usually with default parameters (Gish, W. [1996–2004] http://Blast.wustl.edu). For some subtelomeric sequences, BlastP was used with the filters off, to take into account the low-complexity sequences of amino acid segments in certain proteins.

    Homologs found by Blast searches were examined visually to check whether alignments covered at least roughly two thirds of the length of each protein. Homologs were classified in three categories as follows: "highly similar" (more than 80% similarity in amino acid sequence), "similar" (50% to 80% similarity), and "weakly similar/some similarities with" (less than 50% similarity and similarity spread throughout sequence/short, highly conserved segments).

    Some MAT genes had already been characterized (Kurischko et al. 1992; Kurischko et al. 1999; Srikantha, Lachke, and Soll 2003; Wong et al. 2003), but some new MAT loci were retrieved with genes described in other species than S. cerevisiae. Multiple hits indicated the presence of extra cassettes. Allocation of MAT and HML/HMR nomenclature is based on subtelomeric localization of extra cassettes and synteny analysis of neighboring genes when conserved. In K. waltii, only two loci are identifiable; they are not yet mapped and cannot be classified. MAT and HML/HMR-like loci were aligned so they could be subdivided into smaller boxes: the X and Z1 boxes are present at all three loci, the W and Z2 are present in two copies, and the Y box determine mating type. In A. gossypii, all three loci code for MATa information and therefore, the X and Z1 boxes cannot be identified. The genes encoding mating factor a were not annotated in the new genomes, except in A. gossypii. We found them by Blast search and because of the presence of a conserved CAAX motif (Chen et al. 1997). The genes encoding alpha pheromones were annotated as such in databases and are characterized by the presence of an octapeptidic motif repeated several times in the sequence (Singh et al. 1983).

    Synteny Analyses and Search for Gene Relics

    Syntenic clusters as defined in Dujon et al. (2004) were used to confirm the identity of a hit found by Blast analyses. In this case, neighboring genes of a putative ortholog are also homologs to neighboring genes of the query in S. cerevisiae. When no hit could be found by Blast analyses, syntenic clusters were also used to confirm that the lack of detection of an ortholog was not caused by accumulation of mutations in the ancient open reading frame. The search of such a relic was performed according to Lafontaine et al. (2004). When all attempts at finding an ortholog failed, we cannot evidently exclude that functional homologs, with no easily detectable sequence similarity, are nonetheless present in the considered species (this is the case for SIR4 in K. lactis, [see Results]).

    Results

    Mating-Related Genes

    Genes involved in the primary steps of the sexual cycle—pheromone production, pheromone sensing, signal transduction, meiosis, and mating-type switching—are regulated by the MAT locus that encodes transcription factors. As seen on table 2, genes coding for the sexual pheromones and their receptors have homologs in all six genomes, except D. hansenii and Y. lipolytica, which lack a recognizable mating factor a gene (see Materials and Methods) but have homologs of the genes encoding maturation proteins of the missing pheromone. In Y. lipolytica, two potential homologs are found for the mating factor alpha gene. A putative duplicate pair is also found in A. gossypii, with one gene annotated as such, but the octapeptidic repeat is absent (see Materials and Methods). Even when the gene coding for one of the pheromones seems to be missing in some species, genes encoding modification factors of the pheromone and genes encoding some of the proteins from the specific signal transduction cascade are present. This absence could result in the pathway not being functional or, alternatively, the actual gene is too diverged to be recognized. The conservation of genes encoding transcription factors at MAT-like loci is shown in table 3.

    Table 2 Putative Homologs of S. cerevisiae Genes Involved in Sexual Cycle in Hemiascomycetes

    Table 3 Homologs of Genes in MAT-like Loci

    FIG. 2.— A putative HO gene relic in K. lactis. (A) Coordinates of the sequence segment of K. lactis, including the putative HO relic on chromosome 5 (KLLA0E). (B) and (C) Dotplot graphical outputs from DOTTER that compare nucleotide sequences translated into the three possible amino acid sequences (Sonnhammer and Durbin 1995) (sliding window of 16, threshold score of 35, BLOSUM62 matrix); x-axis same as in (A); y-axis in (B): HO (YDL227c); y-axis in C: intein from TFP1 (YDL185w) (Chong et al. 1996).

    FIG. 1.— Representation of MAT and MAT-like loci in six yeast genomes in comparison to S. cerevisiae. Loci were identified from genome sequences, as described in Materials and Methods. On the left are shown the HML loci and homologs when present, in the center are the MAT loci, and on the right are the HMR loci and homologs. In S. cerevisiae, HML, HMR, and MAT itself are located on the same chromosome and contain identical sequence elements called W, X, Y, Z1, and Z2 "boxes." The Y box encodes either the "a" or the "alpha" information. We have noted all genes as a and alpha in reference to S. cerevisiae, even when genes had a different nomenclature, such as in Y. lipolytica (Barth and Gaillardin 1997). Black lines represent chromosomes, the number of which are indicated by roman numerals. Boxes represent identical elements between cassettes from the same species. Genes are represented by arrows beneath boxes or arrows on black line for genomes without duplicated cassettes and triangles represent putative introns. Telomeres are represented by arrows at ends of chromosomes. At the MAT locus, two alleles, or idiomorphs, exist in most species and are shown one above the other, where the top one is the one from the sequenced strain. Drawing is almost to scale. Sizes of MATa loci are, in S. cerevisiae, in C. glabrata, in K. waltii, and in K. lactis, respectively, 2.4 kb, 1.4 kb, 1.3 kb, and 3 kb long. Sizes of MATalpha loci are, in the same order, 2.5 kb, 1.6 kb, 2.5 kb, and 5.9 kb long. In A. gossypii, the two identical cassettes are approximately 4 kb long. In Y. lipolytica, MATa and MATalpha are approximately 3 kb long. In D. hansenii, the single locus is approximately 3.5 kb long. Mata2 is probably not a true gene in S. cerevisiae (Johnson 1995), but homologs are found in K. waltii and in C. glabrata, where it has no start codon. The Mata2 gene in all the other fungi has no obvious homology to Mata2 from S. cerevisiae; all encoded proteins contain an HMG domain, and expression of this gene has been shown to prevent mating in Y. lipolytica (Kurischko et al. 1999). In K. lactis, an additional gene, alpha3 is found; it is expressed from both MAT and HML (Astrom et al. 2000) and the double deletant is unable to mate.

    Overall, we show that most of the genes presented here are conserved across species, with levels of divergence increasing with phylogenetic distance.

    MAT-like and HMR/HML-like Loci

    We examined the homologs of the MAT and MAT-like loci in all species (fig. 1). These were already described for Y. lipolytica, K. lactis, and C. glabrata (Kurischko et al. 1992; Astrom et al. 2000; Butler et al. 2004), and we now extend this description to the other sequenced species. As shown on table 3, all have at least one MAT locus homolog; however, the sizes of these loci are variable. In S. cerevisiae, the MAT locus is less than 3 kb long, but in A. gossypii, it is almost 5 kb long (fig. 1). Variation not only in gene composition but also in gene size explains this phenomenon. Most MAT loci have either "a" or "alpha" type information, but interestingly, D. hansenii has a MAT-like locus that seems to be a mosaic of MATa and MATalpha genes. We also searched for the existence of silent cassettes. Y. lipolytica and D. hansenii have a single MAT-like locus, whereas K. lactis and C. glabrata, as well as A. gossypii, like S. cerevisiae, each contain two additional loci with similar information. Only two copies are found in the sequence available for K waltii, but a third one may have not yet been assembled from the sequence (8X coverage). Both C. glabrata and K. lactis have two copies on the same chromosome, and the third is elsewhere. A. gossypii has each copy on a different chromosome, and K. waltii has two copies on the same contig. All sequenced strains contain opposite mating-type information in the duplicated cassettes, except A. gossypii, in which all three copies contain the MATa type information. The examination of duplicated cassettes from all species containing them shows that MAT, HML, and HMR-like loci are composed of repeated boxes, denoted W, X, Y, Z1, and Z2 in reference to S. cerevisiae (see Material and Methods), but that the sizes of the different boxes, their combination between loci, and even their presence is not conserved across species (fig. 1).

    The presence of duplicated cassettes suggests that corresponding species undergo mating-type switching, a phenomenon driven by the HO endonuclease in S. cerevisiae. As shown on table 2, we have not found any HO homolog in species that have only one MAT-like cassette, such as Y. lipolytica and D. hansenii, but we have also not found it in either A. gossypii or K. waltii, which contain duplicated cassettes. We have found the HO homolog in C. glabrata, as published (Butler et al. 2004), and discovered a potential relic of the HO gene in K. lactis. As shown in figure 2, this relic is highly degenerate, but the dodecapeptide motif of HO-type endonucleases is still recognizable (Gimble 2000). This family of endonucleases includes the intein VDE, which can be confused with HO homologs and is inserted in VMA1 (or TFP1) in S. cerevisiae. Because the VDE intein gene is present in K. lactis in a VMA1 homolog, we can hypothesize that the sequence we have found is a relic of an ancient HO gene. Finally, we have found a dodecapeptide motif endonuclease homolog in K. thermotolerans that is not located inside the vacuolar ATPase gene and which we propose to be an HO homolog (fig. 3).

    FIG. 3.— Intein gene, HO gene, and MAT-like cassettes in various yeast genomes. Approximate phylogenic relationships among the hemiascomycetes (adapted from Cai, Roberts, and Collins [1996], Kurtzman and Robnett [2003], and Dujon et al. [2004]) is represented on the left. The intein gene information is given for clarification of HO gene status (see text). A minus sign in the intein column means there is a vacuolar ATPase gene homolog in the genome, without the intein gene inserted in it. S. kluyverii was omitted because of negative search results in available sequences. The superscript 1 indicates negative search results in the incomplete sequences available. The superscript 2 means see table 1 and text for details. The appearance of the HO gene, MAT-like cassettes, and other features are indicated on the phylogenetic representation.

    The HO recognition sites on chromosome III of S. cerevisiae are located at the border between the Y and Z1 boxes in all three cassettes, but only the one in the MAT locus is cut when switching occurs. The sequence from C. glabrata exhibits the three potential sites with only minor variations, as shown on figure 4. In K. thermotolerans and in K. lactis, the site is not recognizable, which is consistent with the absence of a functional HO gene.

    FIG. 4.— Y-Z1 junctions in yeast species with cassettes. Sequences from defined Y-Z1 limits are compared with the sequences of S. cerevisiae containing the HO recognition site. The sequences from C. glabrata are those that align best with S. cerevisiae, and it has been shown experimentally that the site present at the MATalpha locus from C. glabrata is cut by the HO endonuclease from S. cerevisiae (Nickoloff, Singer, and Heffron 1990).

    Genes Involved in Silencing of MAT-like Loci, Telomeres, and Subtelomeres

    We have analyzed how a set of proteins involved in mating-type silencing and the telomere position effect in S. cerevisiae were conserved in hemiascomycetes. Table 4 shows that proteins required for mating-type silencing and directly bound to HMR and HML silencers (i.e., Orc1p, Rap1p and Abf1p) present two behaviors. Rap1p and Abf1p are not detected in D. hansenii or Y. lipolytica; neither could be gene relics. This suggests either that they are too diverged to be recognized or that they have appeared in species close to S. cerevisiae. On the contrary, Orc1p is found conserved across species, perhaps because it belongs to the ORC complex also required for initiation of DNA replication, a strong selective pressure operating on this complex. The DNA-binding domain of Rap1p, which recognizes telomeric sequences, is, however, found conserved in yeast species in which Rap1p is detected (fig. 5), suggesting a conserved three-dimensional structure. Accordingly, a 6-bp AC-rich motif resembling the sequence recognized by Rap1p in S. cerevisiae, is present at least in C. glabrata and K. lactis telomeres (McEachern and Blackburn 1994; Dujon et al. 2004). Table 4 also shows that a homolog of Sir3p is found in C. glabrata, despite its weak sequence similarity. The lack of detection of a Sir3p homolog in the other hemiascomycetes and the resemblance between Orc1p and Sir3p in S. cerevisiae, in which they are paralogs, suggest that a duplication event occurred in the ancestor common only to these two species and that these duplicated copies of S. cerevisiae and C. glabrata have evolved rapidly (Kellis, Birren, and Lander 2004; Fabre et al., unpublished data). As shown in table 4, two additional proteins, Sir1p and Sir4p, have peculiar evolutive characteristics. They could not be detected in most of the yeasts studied here and neither could gene relics (I.L. and B.D., unpublished data). Indeed, we have failed to detect any structural homolog of Sir1p in any of the yeast species considered, including C. glabrata. A putative Sir1p was found in S. castellii, in which the amino acids required for binding to Orc1p are conserved (Bose et al. 2004, and data not shown), but because the phylogenetic position of S. castellii is ambigous (Kurtzman and Robnett 2003; Dujon et al. 2004), we cannot determine whether the lack of SIR1 in C. glabrata is caused by rapid divergence of this gene, the loss of this gene in this species, or the de novo creation of SIR1 in Saccharomyces species. It has to be noted that synteny between S. cerevisiae and C. glabrata is disrupted at the SIR1 locus. In the case of Sir4p, we found it weakly conserved at the primary amino acid sequence in C. glabrata, but it belongs to the same syntenic block as in S. cerevisiae. In K. lactis, K. waltii, and A. gossypii, we could not find any obvious homolog unless synteny was examined (Materials and Methods). In fact, the syntenic KLLA0F13420g corresponds to the functional K. lactis homolog isolated by transcomplementation of a sir4 S. cerevisiae mutant but has no similarity to the gene from S. cerevisiae (Astrom and Rine 1998). Similarly, we found two putative homologs of SIR4, tandemly repeated, in A. gossypii. This is not the case in D. hansenii, in which the two genes that surround SIR4 in S. cerevisiae in the same orientation, contiguous and separated by 50 bp, leave no space for a putative SIR4 gene. Because the carboxy-terminal coiled-coil of Sir4p is essential for silencing at telomeres and silent mating loci (Chang et al. 2003; Murphy et al. 2003), we examined whether the putative orthologs of Sir4p show a conserved coiled-coil structure by using the algorithm multicoil (Wolf, Kim, and Berger 1997). We found a conserved coiled-coil at the carboxy-terminal part of these proteins (fig. 6), suggesting a conserved functional role. In the case of A. gossypii, the two copies show nonoverlapping coiled-coils. X-ray crystal structure of Sir4p coiled-coil reveals two interfaces, one formed between homodimeric, parallel coiled-coils and the other between pairs of coiled-coils to form a large hydrophobic interface (Murphy et al. 2003). How each of these independent coiled-coils in A. gossypii participates in each of these interactions remains an open question.

    Table 4 Putative Homologs of S. cerevisiae Genes Involved in Chromatin Silencing in Hemiascomycetes

    FIG. 5.— RAP1and homolog sequences in different species. Multiple alignment between S. cerevisiae RAP1 (YNL216w) and structural homologs are performed with ClustalW (Thompson, Higgins, and Gibson 1994). Sequences from C. glabrata (CAGL0K04917g), K. waltii (Kwal_4491), A. gossypii (ABL180W), K. lactis (KLLA0D19294g), and D. hansenii (DEHA0C13959g) are shown. On the top, DNA-binding domain (residues 361 to 596) of Rap1p, shown to crystallize with an 18-pb telomeric fragment and recognize DNA tandem repeats containing the ACACCA sequence (Konig et al. 1996), is surrounded by two black arrows. Black indicates identical residues; gray indicates similar residues.

    FIG. 6.— Coiled-coil structure prediction in putative Sir4p orthologs. The multicoil software (Wolf, Kim, and Berger 1997) was used to analyze the structure of putative SIR4 orthologs as defined by conservation of synteny. A coiled-coil is predicted in the carboxy-terminal domain of each predicted protein from S. cerevisiae (Sir4p), C. glabrata (CAGL0K11396g), K. lactis (KLLA0F13420g), K. waltii (11611), and A. gossypii (AGR188w and AGR189w). The x-axes correspond to protein length in amino acid and the y-axes correspond to coiled-coil probability.

    All the other silencing proteins examined in table 4, such as the DNA-binding proteins yKu70p and yku80p, histone deacetylases, acetylases, or methyltransferases, were found in all hemiascomycetes, with a level of divergence that follows phylogenetic distances.

    Subtelomeric Genes in Hemiascomycetes

    We have first analyzed the conservation of subtelomeric genes of S. cerevisiae in other hemiascomycetes. The following set of genes were examined (table 5): gene families that encode proteins involved in carbohydrate metabolism, SUC2, MAL, MEL, RTM; the FLO family of adhesins; and families of uncharacterized proteins, COS and PAU, the largest subtelomeric family in S288C. We have extended our search to Saccharomyces species because previous partial data in these species have been central to the notion that these gene families are dynamic (see references in table 5).

    Table 5 Subtelomeric Gene Families in S. cerevisiae and Their Putative Homologs

    The FLO family deserves special mention, as it has been the subject of many studies, some very recent (Halme et al. 2004). FLO genes encode serine/threonine-rich GPI-anchored cell wall proteins (Caro et al. 1997) that act as flocculins. The sequenced S288C strain has four copies of FLO1 homologs and additional subtelomeric FLO1 gene fragments and the nonsubtelomeric FLO11 (Lo and Dranginis 1998). The FLO family has been directly implicated in molecular rearrangements of subtelomeric sequences (Carro et al. 2003). In C. glabrata, the EPA family of proteins, which has homology to the FLO family, has been shown to be involved in cellular adhesion (Cormack, Ghori, and Falkow 1999). Our C. glabrata data confirms that many entire copies and fragments of EPA genes exist, mainly in the last few kilobases close to the ends of the chromosomes, and, in fact, all genomes examined here contain multiple copies of genes and/or gene fragments with stretches of low-complexity serine/threonine amino acid sequences.

    Overall, table 5 shows that many subtelomeric genes are conserved in Saccharomyces species. The situation is more complex outside of this group, as conservation or absence does not match evolutionary distance between the species. D. hansenii has many homologs of S. cerevisiae subtelomeric genes, whereas the phylogenetically closer C. glabrata has almost none, apart from the EPA family that has resemblance to the FLO family. This confirmed the high plasticity of these regions, and we, therefore, looked for the presence of specific families of genes that would be indicative of genes adapted to the species' environment. Indeed, this is what we found, by taking a direct look at the similarities between subtelomeric sequences of chromosomes in a given species, using data from the four Genolevures genomes (Materials and Methods). We have found several families in Y. lipolytica with one subtelomeric member; all others are internal. This situation exists in S. cerevisiae, for example, for the HXT genes (http://www.le.ac.uk/ge/ejl12/research/telostruc/ClustersLarge.html). In K. lactis, we have identified a sequence fragment of approximately 9 kb containing three genes that is repeated on seven different chromosome ends, in the same orientation relative to the telomere, two of which are represented on figure 7. This is a higher degree of redundancy at chromosome ends than what is observed in S. cerevisiae S288C. In all species examined so far, we found evidences of specific gene families at chromosome ends.

    FIG. 7.— Subtelomeric families in K. lactis. Dot-matrix representation of nucleotide identity (15 identities per 23 nucleotides) between subtelomeric sequences from the left end of chromosome I (AL) and the left end of chromosome V (EL). Telomeric repeat alignments form the black square at top left corner (TEL), diagonal indicates the presence of an approximately 9-kb duplication between subtelomeres, encompassing three annotated genes on each sequence, schematized as gray, white, and black arrows, with their names indicated.

    Discussion

    In this work, we have examined three sets of genes directly linked to the divergence of species and their adaptation to an evolutionary niche: genes involved in the variation in the sexual transmission of genomes, genes that encode proteins involved in the variation of sexual mating type and in the variation of expression of some gene families, and polymorphic gene families that lead to, and are created by, chromosomal rearrangements in genomes and that encode products needed in large quantities for adaptation to the environment. Many pathogens such as Plasmodium, Trypanosoma, and Pneumocystis generate variations of cell surface molecules by the presence of large subtelomeric families encoding surface glycoproteins, and human pathogens are prone to becoming asexual. Although several species here are animal or plant pathogens, these phenomena may represent a general situation in the hemiascomycetes' evolutionary branch.

    The genomes presented here exhibit a great variety of situations pertaining to sexuality, mating-type switching, and homothallism, and evolution of sexuality among species is revealed by the genomic data (fig. 3). Two asexual species were analyzed, A. gossypii and C. glabrata. The sequenced type strain of A. gossypii contains three type "a" cassettes, which could explain the fact that it is described as a haploid nonmater. Such cassette configurations also exist in C. glabrata strains (Srikantha, Lachke, and Soll 2003) and in S. cerevisiae. Opposite mating types of strains of A. gossypii could however be searched for in nature, which would allow determination of whether asexuality is a peculiarity of the type strain possibly selected for loss of sexuality-induced clonality or whether it is related to the pathogenicity of this fungus in cotton plants. C. glabrata seems to have all the elements needed for mating and switching (Brockert et al. 2003; Wong et al. 2003; Butler 2004; H. Muller, and C. Fairhead, unpublished data), but only haploid cells have ever been isolated, and no mating has yet been observed. Nonetheless, two haploid types exist (Srikantha, Lachke, and Soll 2003; H. Muller, and C. Fairhead, unpublished data), and there are reports of mating-type switching (Brockert et al. 2003; Butler et al. 2004). Even though all three cassettes are not located on the same chromosome, switching is still theoretically possible; a slightly similar configuration has been shown in S. cerevisiae to allow switching in MATa cells (Wu, Wu, and Haber 1997). Nonetheless, cells that have lost the capacity to switch mating types need not have lost the capacity to mate; these two characters are independent, as illustrated on figure3. In C. glabrata, we have observed that a synthetic alpha factor, deduced from the sequence from C. glabrata, has the property of arresting growth of MATa cells of S. cerevisiae, but not of MATa cells of C. glabrata, in a pheromone-mediated growth arrest assay (data not shown). The structure of the pheromone, although slightly diverged, is still recognized by the receptors in S. cerevisiae, raising the possibility of interspecies crosses. Because C. glabrata cells fail to respond, some component of the haploid pheromone-sensing pathway or of the mating response must be defective. It is possible that C. glabrata, like many other fungal pathogens, has lost its sexual mode of reproduction, a fact that may be correlated with the absence of transposable elements in this genome, as expected from theory (Hickey 1982; Wright and Finnegan 2001). Other asexual fungal pathogens include C. albicans (Magee and Magee 2000), which still contains mating-related genes but has lost meiosis-related genes (Tzung et al. 2001), Aspergillus fumigatus (Dyer, Paoletti, and Archer 2003), Pneumocystis carnii (Fraser and Heitman 2003), and Cryptococcus neoformans, in which pathogenicity is linked to the haploid state (Lengeler et al. 2000).

    Most sexual species presented in figure 3 have HML/HMR-like cassettes, which appear after the split from Y. lipolytica and the D. hansenii/C. albicans branch. The MAT locus of D. hansenii was previously undescribed and seems to be a mosaic of type a and type alpha information, in accordance with the description of this species as homothallic. This is, to our knowledge, the first molecular description of a hemiascomycetous species that is homothallic without mating-type switching. The appearance of this new locus may result from the fusion of a and alpha loci, possibly in diploid cells, a mechanism speculated for the distant ascomycete Cochliobolus (Yun et al. 1999). No homolog of MATalpha2 was found, and roles of transcriptional factors encoded by genes at the MAT locus may have diverged so as to control mating-related genes in a novel way. It must be noted that D. hansenii has genes for all maturation factors and receptors of both a and alpha pheromones, even though we have found no genes encoding a factor precursors.

    In the branches of Kluyveromyces and Saccharomyces, the presence of the two duplicated MAT-like cassettes (fig.1), and their alternate transposition at the MAT locus, theoretically allows self-fertility in these species by cassette switching and HO recruitment. The HO gene, encoding the endonuclease that increases the frequency of switching, has been hypothetized to have appeared in the ancestor common to S. cerevisiae and C. glabrata but after the divergence from K. lactis (Butler et al. 2004). We have found a relic of the HO gene in the genome of K. lactis and a potentially functional homolog in K. thermotolerans; this implies that HO appeared before the separation with Saccharomyces species. K. lactis must have inherited this gene and lost it, and, therefore, become mostly heterothallic, as reported (table 1). K. waltii is described as homothallic and having a sexual cycle, and we identify no HO homolog and only two cassettes of opposite mating types. K. waltii may be a homothallic mater by a novel regulation of the two cassettes. Transcriptional analysis of these loci, in this and other species, would yield precious information on such processes.

    A common characteristics of triplicated mating loci include the fact that they contain boxes of identical sequences and the fact that in all genomes in which the data was available, two cassettes are located in subtelomeric regions of chromosomes. This is compatible with the notion that the general organization of such loci was inherited from a common ancestor. However, the borders of the cassettes differ between species, and even the gene content of these loci varies, suggesting that they have evolved independently in each species, although in a concerted fashion between cassettes of the same genome. This may be the result of gene conversion mechanisms that are involved in mating-type switching. The cassettes could be homogenized by the recombination process, the lowest percentage of mismatches between sequences itself allowing the most efficient recombination (Mezard, Pompon, and Nicolas 1992).

    The second set of genes examined is involved in silencing. It is remarkable that along evolution, heterochromatin formation remains a conserved feature of eukaryotic genomes, allowing for genome stability and correct chromosome segregation (Perrod and Gasser 2003). Heterochromatin also plays a central role in the regulation of gene expression during development and cellular differentiation (Grewal and Moazed 2003). Heterochromatin-like structures are indeed involved in the stable inactivation of developmental regulators such as the homeotic gene clusters in Drosophila and mammals and, in S. cerevisiae and S. pombe, the mating-type genes.

    In this work, we have analyzed how some central players of heterochromatin formation in S. cerevisiae have evolved in hemiascomycetes, and we have determined that a number of these silencing factors have appeared in branches close to S. cerevisiae. Because of this protein divergence, the question remains as to whether multiple mating-type cassettes are silenced in the species where they exist (fig.3). The first partners required in initial steps of establishment of silencing of mating-type cassettes in S. cerevisiae are DNA-binding proteins that include transcription factors. Studies in K. lactis have shown that mating-type silencing can occur through different transcription factors but that they belong to the same Myb domain transcription factor family (Astrom et al. 2000). Silencers have evolved quickly, and some transcription factors may have been substituted by others of the same family with more or less similar DNA-binding affinities. In subsequent steps of the formation of silenced chromatin, a linker protein containing a coiled-coil domain has a pivotal role in S. cerevisiae. We have shown that putative Sir4p homologs with such a conserved structural domain can be found in each species where more than one mating-type cassette is present, suggesting that silencing can be functional. Indeed, this has been proved for K. lactis (Astrom and Rine 1998). As for Sir3p, in most of the yeast species, we have no evidence for an ortholog, but it is possible that this function has been taken over by Orc1p or other proteins, although we cannot exclude that Sir3p function is not required for silencing in these yeasts. All HML/HMR-like loci may be silenced by a mechanism similar to that of S. cerevisiae and may corroborate the fact that these yeasts have a sexual cycle (except A. gossypii [see above]). The notable exception is C. glabrata, in which all the proteins required for mating-type silencing are present, except Sir1p (table 4). It could be hypothesized that the triplicate MAT-like loci of these yeast are, therefore, not silenced and would explain the failure to mate C. glabrata cells of opposite mating types, as this yeast would behave as a diploid.

    As for subtelomeric silencing, it has been experimentally demonstrated that expression of genes involved in surface adhesion is regulated by an SIR3/RAP1-dependent mechanism in C. glabrata (De Las Penas et al. 2003), suggesting that subtelomeric silencing in this yeast species follows molecular mechanisms close to those of S. cerevisiae. Subtelomeric silencing in K. lactis is also dependent on Rap1p fixation on telomeric repeats (Gurevich et al. 2003). In the distant species D. hansenii and Y. lipolytica, the question as to how and whether subtelomeric silencing is achieved remains open. As we have observed that silencing proteins diverge rapidly, it is possible that subtelomeric regions in novel species are not subjected to silencing; this observation is consistent with the ability of these regions to harbor highly expressed gene families (see below).

    Understanding how transcriptional inactive chromatin is established and maintained remains a challenge, but histones and their posttranslational modifications play a pivotal role in the assembly of heterochromatin-like structures (Grewal and Moazed 2003). Accordingly, histones and trans-modifying activities of amino-terminal histone H3 and H4 tails, such as the prominent histone deacetylases, acetylases, and methyltransferases, are also conserved in the hemiascomycetous analyzed here (table 4). However, the way heterochromatin complexes are targeted to specific domains are different and were probably invented several times, because in hemiascomycetous yeasts close to S. cerevisiae, silencers and DNA-binding proteins are the first key markers, whereas in metazoans, establishment of inactive chromatin implicate repetitive DNA and noncoding RNAs. Study of establishment of inactive chromatin in distant yeast species such as D. hansenii and Y. lipolytica, in which both DNA-binding proteins and Sir proteins and components of the RNAi pathway are not conserved (not shown), should, thus, provide precious information.

    Finally, we have seen that subtelomeric gene families are species specific: even though sometimes orthologs do exist across species (i.e., the protein activity is conserved), they do not exhibit the same characteristics in terms of copy number and subtelomeric localization. This was noticeable in partial data on genes encoding sugar-utilizing enzymes in Saccharomyces species, because 69% of species-specific genes were reported to be subtelomeric (Kellis et al. 2003). Our preliminary analysis shows that the presence of subtelomeric gene families in all species examined is a general rule, even if there are specific characteristics that need further investigation.

    Even though subtelomeres are regions where transcriptionally silenced genes have been described, one of the roles of subtelomeric DNA, at least in S. cerevisiae, is to allow amplification of gene families, the expression of which gives an evolutionary advantage to the cell (Turakainen, Aho, and Korhola 1993). We have seen that all species contain homologs of FLO-like proteins with low-complexity amino acid sequence segments, and FLO-like GPI- anchored cell wall proteins (GPI-CWP) have previously been characterized in the very distant hemiascomycete Y. lipolytica (Jaafar and Zueco 2004). The FLO-like families may have evolved as a reservoir of surface protein genes that can be expressed alternatively and genetically shuffled. The evolutionary mechanism may be of the same sort as described for interspecies divergence by Kellis, Birren, and Lander (2004) and Wolfe (2004), in which pairs of genes surviving from the whole-genome duplication have diverged so much as to not be recognizeable by BlastP analysis but are nonetheless related.

    Apart from gene family amplification, another characteristic of subtelomeric sequences is the plasticity they confer onto chromosome ends. In S. cerevisiae, chromosome length polymorphism is mainly caused by illegitimate recombination between subtelomeric sequences and between Ty's (Rachidi et al. 1999). This sometimes results in the emergence of new, mosaic gene sequences (Kobayashi et al. 1998). Other properties of these repeated sequences in S. cerevisiae include allowing survival of a small number of cells when the telomerase complex is defective (Huang et al. 2001), where survival depends on Y' sequence amplification, recombination between chromosome ends, and chromosome end fusion. Subtelomeric repeats also allow more cells to survive to a double-strand break than when the break is central, (Ricchetti, Dujon, and Fairhead 2003) because of break-induced replication and gene-conversion events occurring between COS and FLO genes. Even though Y'-type sequences are absent from the genomes of species outside the Saccharomyces group, other common features suggest that subtelomeric properties may be shared among ascomycetes and need experimental testing such as telomerase mutant analysis.

    Mating-type cassette transposition and chromosomal rearrangements between subtelomeric sequences both rely on the presence of the mitotic recombination machinery, which is shown to be conserved throughout hemiascomycetes (Richard et al., in press), thereby demonstrating the essential character of this machinery and emphasizing the generality of its involvement in numerous mitotic gene-conversion and genomic-rearrangement pathways.

    Supplementary Material

    Web Sites

    Genolevures home page: http://cbi.labri.fr/Genolevures/

    Blast on Ashbya gossypii at Duke: http://data.cgt.duke.edu/ashbya/blast.html

    Ashbya gossypii sequence home page: http://agd.unibas.ch/

    Partial genomic sequences of S. mikatae, S. kudriavzevii, S. bayanus, S. castellii, S. kluyveriat Genome Sequencing Center: http://genome.wustl.edu/projects/yeast/

    Blast on fungi at NCBI: http://www.ncbi.nlm.nih.gov/sutils/genom_table.cgi?organism=fungi

    Blast and sequences from K. waltii in Supplementary Information of Kellis, Birren, and Lander (2004): http://www.nature.com/nature/journal/v428/n6983/extref/nature02424-s1.htm

    Washington University Blast home page: http://blast.wustl.edu

    Maps of Subtelomeres of S. cerevisiae: http://www.le.ac.uk/ge/ejl12/research/telostruc/ClustersLarge.html

    Saccharomycesgenome database: http://www.yeastgenome.org/

    Acknowledgements

    We thank Emmanuel Talla for access to K. thermotolerans sequences and Christophe Hennequin for help with mating-type gene analysis. We thank members of the Unité de Génétique Moléculaire des Levures and of the Génolevures consortium for general discussions and for annotation of the novel yeast sequences and the LABRI team for initiation and maintenance of the Génolevures database. This work was supported by the Consortium National de Recherche en Génomique (to Génoscope and to Institut Pasteur Génopole), the CNRS (GDR2354, Génolevures sequencing consortium), the Ministère de la Jeunesse, de l'Education et de la Recherche (ACI IMPBio n°IMPB114 "Génolevures en ligne") and the ‘Conseil Régional d'Aquitaine’ ("Génotypage et Génomique Comparée"). H.M. and P.T. are recipients of a doctoral fellowship of the Ministère de la Recherche through the Université Paris 6. B.D. is member of the Institut Universitaire de France.

    References

    Ai, W., P. G. Bertram, C. K. Tsang, T. F. Chan, and X. F. Zheng. 2002. Regulation of subtelomeric silencing during stress response. Mol. Cell. 10:1295–1305.

    Astrom, S. U., A. Kegel, J. O. Sjostrand, and J. Rine. 2000. Kluyveromyces lactisSir2p regulates cation sensitivity and maintains a specialized chromatin structure at the cryptic alpha-locus. Genetics 156:81–91.

    Astrom, S. U., and J. Rine. 1998. Theme and variation among silencing proteins in Saccharomyces cerevisiae and Kluyveromyces lactis. Genetics 148:1021–1029.

    Barth, G., and C. Gaillardin. 1997. Physiology and genetics of the dimorphic fungus Yarrowia lipolytica. FEMS Microbiol. Rev. 19:219–237.

    Bose, M. E., K. H. McConnell, K. A. Gardner-Aukema, U. Muller, M. Weinreich, J. L. Keck, and C. A. Fox. 2004. The origin recognition complex and Sir4 protein recruit Sir1p to yeast silent chromatin through independent interactions requiring a common Sir1p domain. Mol. Cell. Biol. 24:774–786.

    Brockert, P. J., S. A. Lachke, T. Srikantha, C. Pujol, R. Galask, and D. R. Soll. 2003. Phenotypic switching and mating type switching of Candida glabrata at sites of colonization. Infect. Immun. 71:7109–7118.

    Butler, G., C. Kenny, A. Fagan, C. Kurischko, C. Gaillardin, and K. H. Wolfe. 2004. Evolution of the MAT locus and its Ho endonuclease in yeast species. Proc. Natl. Acad. Sci. USA 101:1632–1637.

    Cai, J., I. N. Roberts, and M. D. Collins. 1996. Phylogenetic relationships among members of the ascomycetous yeast genera Brettanomyces, Debaryomyces, Dekkera, and Kluyveromyces deduced by small-subunit rRNA gene sequences. Int. J. Syst. Bacteriol. 46:542–549.

    Caro, L. H., H. Tettelin, J. H. Vossen, A. F. Ram, H. van den Ende, and F. M. Klis. 1997. In silicio identification of glycosyl-phosphatidylinositol-anchored plasma-membrane and cell wall proteins of Saccharomyces cerevisiae. Yeast 13:1477–1489.

    Carro, D., J. Garcia-Martinez, J. E. Perez-Ortin, and B. Pina. 2003. Structural characterization of chromosome I size variants from a natural yeast strain. Yeast 20:171–183.

    Chang, J. F., B. E. Hall, J. C. Tanny, D. Moazed, D. Filman, and T. Ellenberger. 2003. Structure of the coiled-coil dimerization motif of Sir4 and its interaction with Sir3. Structure (Camb.) 11:637–649.

    Chen, P., S. K. Sapperstein, J. D. Choi, and S. Michaelis. 1997. Biogenesis of the Saccharomyces cerevisiae mating pheromone a-factor. J. Cell Biol. 136:251–269.

    Chen, X. J., and G. D. Clark-Walker. 1994. sir2 mutants of Kluyveromyces lactis are hypersensitive to DNA-targeting drugs. Mol. Cell. Biol. 14:4501–4508.

    Chong, S., Y. Shao, H. Paulus, J. Benner, F. B. Perler, and M. Q. Xu. 1996. Protein splicing involving the Saccharomyces cerevisiae VMA intein: the steps in the splicing pathway, side reactions leading to protein cleavage, and establishment of an in vitro splicing system. J. Biol. Chem. 271:22159–22168.

    Cliften, P., P. Sudarsanam, A. Desikan, L. Fulton, B. Fulton, J. Majors, R. Waterston, B. A. Cohen, and M. Johnston. 2003. Finding functional features in Saccharomycesgenomes by phylogenetic footprinting. Science 301:71–76.

    Coppin, E., R. Debuchy, S. Arnaise, and M. Picard. 1997. Mating types and sexual development in filamentous ascomycetes. Microbiol. Mol. Biol. Rev. 61:411–428.

    Cormack, B. P., N. Ghori, and S. Falkow. 1999. An adhesin of the yeast pathogen Candida glabrata mediating adherence to human epithelial cells. Science 285:578–582.

    Cosentino, S., M. E. Fadda, M. Deplano, A. F. Mulargia, and F. Palmas. 2001. Yeasts associated with Sardinian ewe's dairy products. Int. J. Food Microbiol. 69:53–58.

    Critchlow, S. E., and S. P. Jackson. 1998. DNA end-joining: from yeast to man. Trends Biochem. Sci. 23:394–398.

    De Las Penas, A., S. J. Pan, I. Castano, J. Alder, R. Cregg, and B. P. Cormack. 2003. Virulence-related surface glycoproteins in the yeast pathogen Candida glabrata are encoded in subtelomeric clusters and subject to RAP1- and SIR-dependent transcriptional silencing. Genes Dev. 17:2245–2258.

    Dietrich, F. S., S. Voegeli, S. Brachat et al. (14 co-authors). 2004. The Ashbyagossypii genome as a tool for mapping the ancient Saccharomyces cerevisiaegenome. Science 304:304–307.

    do Carmo-Sousa, L., and C. Barroso-Lopes. 1969. A comparative study of the extracellular and cell wall polysaccharides of some Candida species. Antonie Van Leeuwenhoek 35(Suppl.):A7-8.

    Dujon, B., D. Sherman, G. Fischer et al. (67 co-authors). 2004. Genome evolution in yeasts. Nature 430:35–44.

    Dyer, P. S., M. Paoletti, and D. B. Archer. 2003. Genomics reveals sexual secrets of Aspergillus. Microbiology 149:2301–2303.

    Fraser, J. A., and J. Heitman. 2003. Fungal mating-type loci. Curr. Biol. 13:R792–795.

    ———. 2004. Evolution of fungal sex chromosomes. Mol. Microbiol. 51:299–306.

    Gasser, S. M., and M. M. Cockell. 2001. The molecular biology of the SIR proteins. Gene 279:1–16.

    Gimble, F. S. 2000. Invasion of a multitude of genetic niches by mobile endonuclease genes. FEMS Microbiol Lett. 185:99–107.

    Goncalves, P. M., K. Maurer, W. H. Mager, and R. J. Planta. 1992. Kluyveromyces contains a functional ABF1-homologue. Nucleic Acids Res. 20:2211–2215.

    Grewal, S. I., and D. Moazed. 2003. Heterochromatin and epigenetic control of gene expression. Science 301:798–802.

    Gurevich, R., S. Smolikov, H. Maddar, and A. Krauskopf. 2003. Mutant telomeres inhibit transcriptional silencing at native telomeres of the yeast Kluyveromyces lactis. Mol. Genet. Genomics 268:729–738.

    Haber, J. E. 1998. Mating-type gene switching in Saccharomyces cerevisiae. Annu. Rev. Genet. 32:561–599.

    Halme, A., S. Bumgarner, C. Styles, and G. R. Fink. 2004. Genetic and epigenetic regulation of the FLO gene family generates cell-surface variation in yeast. Cell 116:405–415.

    Haw, R., A. D. Yarragudi, and H. Uemura. 2001. Isolation of a Candida glabrata homologue of RAP1, a regulator of transcription and telomere function in Saccharomyces cerevisiae. Yeast 18:1277–1284.

    Herman, A., and H. Roman. 1966. Allele specific determinants of homothallism in Saccharomyces lactis. Genetics 53:727–740.

    Herskowitz, I., and Y. Oshima. 1981. Control of cell type in Saccharomyces cerevisiae: mating type and Mating-type interconversion. Pp. 181–209 in The molecular biology of the yeast Saccharomyces: Life cycle and inheritance. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

    Herskowitz, I., J. Rine, and J. N. Strathern. 1992. Mating-type determination and mating-type interconversion in S. cerevisiae. Pp. 583–656 in J. R. Broach, J. R. Pringle, and E. W. Jones, eds. The molecular biology of the yeast Saccharomyces: Gene expression. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

    Hickey, D. A. 1982. Selfish DNA: a sexually-transmitted nuclear parasite. Genetics 101:519–531.

    Huang, P., F. E. Pryde, D. Lester, R. L. Maddison, R. H. Borts, I. D. Hickson, and E. J. Louis. 2001. SGS1 is required for telomere elongation in the absence of telomerase. Curr. Biol. 11:125–129.

    Jaafar, L., and J. Zueco. 2004. Characterization of a glycosylphosphatidylinositol-bound cell-wall protein (GPI-CWP) in Yarrowia lipolytica. Microbiology 150:53–60.

    Johnson, A. D. 1995. Molecular mechanisms of cell-type determination in budding yeast. Curr. Opin. Genet. Dev. 5:552–558.

    Johnston, J. R., C. Baccari, and R. K. Mortimer. 2000. Genotypic characterization of strains of commercial wine yeasts by tetrad analysis. Res. Microbiol. 151:583–590.

    Kellis, M., B. W. Birren, and E. S. Lander. 2004. Proof and evolutionary analysis of ancient genome duplication in the yeast Saccharomyces cerevisiae. Nature 428:617–624.

    Kellis, M., N. Patterson, M. Endrizzi, B. Birren, and E. S. Lander. 2003. Sequencing and comparison of yeast species to identify genes and regulatory elements. Nature 423:241–254.

    Kimura, A., T. Umehara, and M. Horikoshi. 2002. Chromosomal gradient of histone acetylation established by Sas2p and Sir2p functions as a shield against gene silencing. Nat. Genet. 32:370–377.

    Klar, A. J. S. 1992. Molecular genetics of fission yeast cell type: mating type and mating-type interconversion. Pp. 745–777 in J. R. Broach, J. R. Pringle, and E. W. Jones, eds. The molecular biology of the yeast Saccharomyces: Gene expression. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

    Kobayashi, O., N. Hayashi, R. Kuroki, and H. Sone. 1998. Region of FLO1 proteins responsible for sugar recognition. J. Bacteriol. 180:6503–6510.

    Konig, P., R. Giraldo, L. Chapman, and D. Rhodes. 1996. The crystal structure of the DNA-binding domain of yeast RAP1 in complex with telomeric DNA. Cell 85:125–136.

    Kooistra, R., P. J. Hooykaas, and H. Y. Steensma. 2004. Efficient gene targeting in Kluyveromyces lactis. Yeast 21:781–792.

    Krauskopf, A., and E. H. Blackburn. 1996. Control of telomere growth by interactions of RAP1 with the most distal telomeric repeats. Nature 383:354–357.

    Kurischko, C., P. Fournier, M. Chasles, H. Weber, and C. Gaillardin. 1992. Cloning of the mating-type gene MATA of the yeast Yarrowia lipolytica. Mol. Gen. Genet. 232:423–426.

    Kurischko, C., M. B. Schilhabel, I. Kunze, and E. Franzl. 1999. The MATA locus of the dimorphic yeast Yarrowia lipolytica consists of two divergently oriented genes. Mol. Gen. Genet. 262:180–188.

    Kurtzman, C. P. 1994. Molecular taxonomy of the yeasts. Yeast 10:1727–1740.

    Kurtzman, C. P., and C. J. Robnett. 2003. Phylogenetic relationships among yeasts of the ‘Saccharomyces complex’ determined from multigene sequence analyses. FEMS Yeast Res. 3:417–432.

    Lafontaine, I., G. Fischer, E. Talla, and B. Dujon. 2004. Gene relics in the genome of the yeast Saccharomyces cerevisiae. Gene 335:1–17.

    Laroche, T., S. G. Martin, M. Gotta, H. C. Gorham, F. E. Pryde, E. J. Louis, and S. M. Gasser. 1998. Mutation of yeast Ku genes disrupts the subnuclear organization of telomeres. Curr. Biol. 8:653–656.

    Lengeler, K. B., P. Wang, G. M. Cox, J. R. Perfect, and J. Heitman. 2000. Identification of the MATa mating-type locus of Cryptococcus neoformans reveals a serotype A MATa strain thought to have been extinct. Proc. Natl. Acad. Sci. USA 97:14455–14460.

    Lo, W. S., and A. M. Dranginis. 1998. The cell surface flocculin Flo11 is required for pseudohyphae formation and invasion by Saccharomyces cerevisiae. Mol. Biol. Cell. 9:161–171.

    Louis, E. J. 1995. The chromosome ends of Saccharomyces cerevisiae. Yeast 11:1553–1573.

    Louis, E. J., and J. E. Haber. 1992. The structure and evolution of subtelomeric Y' repeats in Saccharomyces cerevisiae. Genetics 131:559–574.

    Magee, B. B., and P. T. Magee. 2000. Induction of mating in Candida albicans by construction of MTLa and MTLalpha strains. Science 289:310–313.

    Martin, A. M., D. J. Pouchnik, J. L. Walker, and J. J. Wyrick. 2004. Redundant roles for histone H3 N-terminal lysine residues in subtelomeric gene repression in Saccharomyces cerevisiae. Genetics 167:1123–1132.

    McEachern, M. J., and E. H. Blackburn. 1994. A conserved sequence motif within the exceptionally diverse telomeric sequences of budding yeasts. Proc. Natl. Acad. Sci. USA 91:3453–3457.

    Mefford, H. C., and B. J. Trask. 2002. The complex structure and dynamic evolution of human subtelomeres. Nat. Rev. Genet. 3:91–102.

    Mezard, C., D. Pompon, and A. Nicolas. 1992. Recombination between similar but not identical DNA sequences during yeast transformation occurs within short stretches of identity. Cell 70:659–670.

    Murphy, G. A., E. J. Spedale, S. T. Powell, L. Pillus, S. C. Schultz, and L. Chen. 2003. The Sir4 C-terminal coiled coil is required for telomeric and mating type silencing in Saccharomyces cerevisiae. J. Mol. Biol. 334:769–780.

    Naumov, G. I., E. S. Naumova, R. A. Lantto, E. J. Louis, and M. Korhola. 1992. Genetic homology between Saccharomyces cerevisiae and its sibling species S. paradoxus and S. bayanus: electrophoretic karyotypes. Yeast 8:599–612.

    Naumov, G. I., E. S. Naumova, and E. J. Louis. 1995. Genetic mapping of the alpha-galactosidase MEL gene family on right and left telomeres of Saccharomyces cerevisiae. Yeast 11:481–483.

    Naumov, G. I., E. S. Naumova, and C. A. Michels. 1994. Genetic variation of the repeated MAL loci in natural populations of Saccharomyces cerevisiae and Saccharomyces paradoxus. Genetics 136:803–812.

    Naumov, G. I., E. S. Naumova, E. D. Sancho, and M. P. Korhola. 1996. Polymeric SUC genes in natural populations of Saccharomyces cerevisiae. FEMS Microbiol. Lett. 135:31–35.

    Ness, F., and M. Aigle. 1995. RTM1: a member of a new family of telomeric repeated genes in yeast. Genetics 140:945–956.

    Nickoloff, J. A., J. D. Singer, and F. Heffron. 1990. In vivo analysis of the Saccharomyces cerevisiae HO nuclease recognition site by site-directed mutagenesis. Mol. Cell. Biol. 10:1174–1179.

    Perrod, S., and S. M. Gasser. 2003. Long-range silencing and position effects at telomeres and centromeres: parallels and differences. Cell Mol. Life Sci. 60:2303–2318.

    Pryde, F. E., and E. J. Louis. 1999. Limitations of silencing at native yeast telomeres. EMBO J. 18:2538–2550.

    Rachidi, N., M. J. Martinez, P. Barre, and B. Blondin. 2000. Saccharomyces cerevisiae PAU genes are induced by anaerobiosis. Mol. Microbiol. 35:1421–1430.

    Rachidi, N., P. Barre, and B. Blondin. 1999. Multiple Ty-mediated chromosomal translocations lead to karyotype changes in a wine strain of Saccharomyces cerevisiae. Mol. Gen. Genet. 261:841–50.

    Ricchetti, M., B. Dujon, and C. Fairhead. 2003. Distance from the chromosome end determines the efficiency of double strand break repair in subtelomeres of haploid yeast. J. Mol. Biol. 328:847–862.

    Richard, G. E., A. Kerrest, L. Lafontaine, and B. Dujon, 2005. comparative genomics of hemiascomycete yeasts: genes involved in DNA replication, repair, and recombination. Mol Biol Evol, in press.

    Robyr, D., Y. Suka, I. Xenarios, S. K. Kurdistani, A. Wang, N. Suka, and M. Grunstein. 2002. Microarray deacetylation maps determine genome-wide functions for yeast histone deacetylases. Cell 109:437–446.

    Rusche, L. N., A. L. Kirchmaier, and J. Rine. 2003. The establishment, inheritance, and function of silenced chromatin in Saccharomyces cerevisiae. Annu. Rev. Biochem 72:481–516.

    Singh, A., E. Y. Chen, J. M. Lugovoy, C. N. Chang, R. A. Hitzeman, and P. H. Seeburg. 1983. Saccharomyces cerevisiae contains two discrete genes coding for the alpha-factor pheromone. Nucleic Acids Res. 11:4049–4063.

    Sonnhammer, E. L., and R. Durbin. 1995. A dot-matrix program with dynamic threshold control suited for genomic DNA and protein sequence analysis. Gene 167:GC1–10.

    Souciet, J., M. Aigle, F. Artiguenave et al. (24 co-authors). 2000. Genomic exploration of the hemiascomycetous yeasts: 1. A set of yeast species for molecular evolution studies. FEBS Lett. 487:3–12.

    Spode, I., D. Maiwald, C. P. Hollenberg, and M. Suckow. 2002. ATF/CREB sites present in sub-telomeric regions of Saccharomyces cerevisiae chromosomes are part of promoters and act as UAS/URS of highly conserved COS genes. J. Mol. Biol. 319:407–420.

    Srikantha, T., S. A. Lachke, and D. R. Soll. 2003. Three mating type-like loci in Candida glabrata. Eukaryot. Cell. 2:328–340.

    Suka, N., K. Luo, and M. Grunstein. 2002. Sir2p and Sas2p opposingly regulate acetylation of yeast histone H4 lysine16 and spreading of heterochromatin. Nat. Genet. 32:378–383.

    Taddei, A., F. Hediger, F. R. Neumann, C. Bauer, and S. M. Gasser. 2004. Separation of silencing from perinuclear anchoring functions in yeast Ku80, Sir4 and Esc1 proteins. EMBO J. 23:1301–1312.

    Teunissen, A. W., and H. Y. Steensma. 1995. Review: the dominant flocculation genes of Saccharomyces cerevisiae constitute a new subtelomeric gene family. Yeast 11:1001–1013.

    Tham, W. H., and V. A. Zakian. 2002. Transcriptional silencing at Saccharomyces telomeres: implications for other organisms. Oncogene 21:512–521.

    Thompson, J. D., D. G. Higgins, and T. J. Gibson. 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.

    Turakainen, H., S. Aho, and M. Korhola. 1993. MEL gene polymorphism in the genus Saccharomyces. Appl. Environ. Microbiol. 59:2622–2630.

    Tzung, K. W., R. M. Williams, S. Scherer et al. (13 co-authors). 2001. Genomic evidence for a complete sexual cycle in Candida albicans. Proc. Natl. Acad. Sci. USA 98:3249–3253.

    Wolf, E., P. S. Kim, and B. Berger. 1997. MultiCoil: a program for predicting two- and three-stranded coiled coils. Protein Sci. 6:1179–1189.

    Wolfe, K. 2004. Evolutionary genomics: yeasts accelerate beyond BLAST. Curr. Biol. 14:R392–394.

    Wong, S., M. A. Fares, W. Zimmermann, G. Butler, and K. H. Wolfe. 2003. Evidence from comparative genomics for a complete sexual cycle in the ‘asexual’ pathogenic yeast Candida glabrata. Genome Biol. 4:R10.

    Wright, S., and D. Finnegan. 2001. Genome evolution: sex and the transposable element. Curr. Biol. 11:R296–299.

    Wu, X., C. Wu, and J. E. Haber. 1997. Rules of donor preference in Saccharomyces mating-type gene switching revealed by a competition assay involving two types of recombination. Genetics 147:399–407.

    Wyrick, J. J., F. C. Holstege, E. G. Jennings, H. C. Causton, D. Shore, M. Grunstein, E. S. Lander, and R. A. Young. 1999. Chromosomal landscape of nucleosome-dependent gene expression and silencing in yeast. Nature 402:418–421.

    Yun, S. H., M. L. Berbee, O. C. Yoder, and B. G. Turgeon. 1999. Evolution of the fungal self-fertile reproductive life style from self-sterile ancestors. Proc. Natl. Acad. Sci. USA 96:5592–5597.(Emmanuelle Fabre, Hélo?se)