The whole-genome duplication (WGD) may provide a basis for the emergence of the very characteristic life style of Saccharomyces cerevisiae—its fermentation-oriented physiology and its capacity of growing in anaerobiosis. Indeed, we found an over-representation of oxygen-responding genes in the ohnologs of S. cerevisiae. Many of these duplicated genes are present as aerobic/hypoxic(anaerobic) pairs and form a specialized system responding to changing oxygen availability. HYP2/ANB1 and COX5A/COX5B are such gene pairs, and their unique orthologs in the ‘non-WGD’ Kluyveromyces lactis genome behaved like the aerobic versions of S. cerevisiae. ROX1 encodes a major oxygen-responding regulator in S. cerevisiae. The synteny, structural features and molecular function of putative KlROX1 were shown to be different from that of ROX1. The transition from the K. lactis-type ROX1 to the S. cerevisiae-type ROX1 could link up with the development of anaerobes in the yeast evolution. Bioinformatics and stochastic analyses of the Rox1p-binding site (YYYATTGTTCTC) in the upstream sequences of the S. cerevisiae Rox1p-mediated genes and of the K. lactis orthologs also indicated that K. lactis lacks the specific gene system responding to oxygen limiting environment, which is present in the ‘post-WGD’ genome of S. cerevisiae. These data suggested that the oxygen-responding system was born for the specialized physiology of S. cerevisiae.
Citation: Fang Z-A, Wang G-H, Chen A-L, Li Y-F, Liu J-P, Li Y-Y, et al. (2009) Gene Responses to Oxygen Availability in Kluyveromyces lactis: an Insight on the Evolution of the Oxygen-Responding System in Yeast. PLoS ONE 4(10): e7561. https://doi.org/10.1371/journal.pone.0007561
Editor: Dana Davis, University of Minnesota, United States of America
Received: June 1, 2009; Accepted: September 16, 2009; Published: October 26, 2009
Copyright: © 2009 Fang et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: The authors received no specific funding for this study. Z.-A.F. is a recipient of a fellowship from the French-Chinese joint doctoral program (French Embassy in China). G.-H.W. is supported by a post-doctoral fellowship from the Franco-Chinese Foundation for Basic and Applied Science (FFCSA) and the China Scholarship Council. And W.-G.B. acknowledges support from the French National Centre for Scientific Research (CNRS). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
The yeast Saccharomyces cerevisiae, as well as other sensu stricto Saccharomyces species and a few of their close relatives, has a characteristic physiology that is obviously different from that of other eukaryotic cells and even most yeast species. One of the specialized features of S. cerevisiae is its regulatory network responding to oxygen availability, which allows this species to live on an exclusively fermentative mode even in fully aerated conditions, and also to grow vigorously in the complete absence of oxygen. Such a particular life style must rely on rapid and efficient gene responses to oxygen availability. The regulation of the oxygen-responding system, including both aerobic and hypoxic(anaerobic) genes that function either in presence or in absence of oxygen molecules, has been extensively investigated in S. cerevisiae, and the major transcription activators and repressors such as Hap1p, Rox1p and Mot3p have been identified and characterized –.
It has been proposed that a whole-genome duplication (WGD) event occurred after the separation of S. cerevisiae and Kluyveromyces lactis from a common ancestral ‘pre-WGD’ yeast –. Some recent analyses suggested that the WGD could provide the basis for the evolution that led to the special physiological properties of the modern S. cerevisiae yeast containing a ‘post-WGD’ genome , . For example, an increase in the number of some genes resulted in an enhanced glycolytic flux, which is necessary for the fermentation-oriented metabolism of S. cerevisiae . This change was regarded as an outcome of the WGD event followed by a selection in glucose-rich environments .
Among massive duplicated genes of S. cerevisiae (most of which were formed by whole-genome duplication, and called ohnologs-a subset of paralogs), many gene pairs were combined into a network system specially involved in oxygen response. HYP2/ANB1 and COX5A/COX5B are thought to be such gene pairs. While HYP2 and COX5A work in aerobic condition, ANB1 and COX5B function in hypoxic and anaerobic conditions –. This oxygen-dependent alternative expression pattern of aerobic versus hypoxic(anaerobic) genes can also be found in many other cases of the duplicated gene pairs, such as CYC1/CYC7 , PET9/AAC3 , and HMG1/HMG2 . A trait shared with these gene pairs is that their hypoxic(anaerobic) partners are commonly under the repression of Rox1p in an oxygen-dependent way . Some hypoxic(anaerobic) genes such as ANB1 and HEM13 are also regulated by another repressor Mot3p , –.
As opposed to S. cerevisiae, K. lactis shows a quite different oxygen response. It is Crabtree effect-negative  and Kluyver effect-positive , like many other yeast species, whilst S. cerevisiae is not. Further, K. lactis contains a ‘non-WGD’ genome  in which there are sets of genes comparable to those of S. cerevisiae . The genomic and physiological features of K. lactis suggest that this yeast may have retained some major characteristics of the oxygen response patterns of the ancestral genome that gave rise to S. cerevisiae and K. lactis. The regulation of a few genes involved in respiration and haem synthesis in K. lactis has been reported to be dependent on oxygen –. Our previous work revealed that the target genes and regulatory modes of the oxygen-dependent regulator KlHap1p in K. lactis were different from those of the S. cerevisiae Hap1p . However, the information is still limited about the gene response to oxygen availability and, in particular, about its regulators in K. lactis. We therefore selected some K. lactis genes, which appeared equivalent to the well-known oxygen-responding genes in S. cerevisiae such as HEM13, HYP2/ANB1 and COX5A/COX5B, and investigated the gene response to oxygen in K. lactis, in an attempt to understand how the oxygen-responding system has evolved from a ‘pre-WGD’ to a ‘post-WGD’ genome.
Oxygen-responding genes are over-represented in ohnologs of S. cerevisiae
It has been reported that hundreds of genes respond to oxygen in S. cerevisiae . In this ‘post-WGD’ genome, 554 ohnolog pairs have been identified (http://wolfe.gen.tcd.ie/ygob/) , –. To estimate the role of the ohnologs in oxygen response, we checked the expression of these genes under aerobic and anaerobic conditions according to the published microarray survey including 6020 ORFs  (also see http://transcriptome.ens.fr/ymgv/publi_desc.php?pub_id=34). The results are included in Supplementary Table S1. With respect to a factor of 1.5, 2 or 3 in the change of the gene expression level between aerobic and anaerobic conditions, a significantly higher proportion of genes responding to oxygen were always found in ohnologs than the genes scattered over the whole genome. This is the case for both up- and down- regulated genes (Supplementary Table S2-A). The results indicated that oxygen-responding genes arose from ohnologs more frequently than from non-ohnologs (Supplementary Table S2-B) in the evolution of the S. cerevisiae genome. For example, considering a factor of 2, 18.7% of ohnologs were oxygen responsive, but only 10.4% in non-ohnologs (Figure 1). The former is 78.9% higher than the latter (P<e−15). There are 174 ohnolog pairs among 554, in which at least one gene has oxygen response, indicating that nearly one third of genes (31.4%) became involve in response to oxygen availability after the WGD event. These data suggested a connection between the formation of efficient oxygen-responding system in S. cerevisiae and the whole-genome duplication event, which has been previously discussed , .
Genes are counted that showed more than 2-fold change of expression level (up- or down-regulation) in response to oxygen. Hatched and blank boxes represent the percentages of oxygen-responding genes in onhologs and non-onhologs respectively.
Aerobic/hypoxic(anaerobic) gene pairs in S. cerevisiae usually appear unique in K. lactis
There are many aerobic/hypoxic(anaerobic) paralog pairs in S. cerevisiae, such as HYP2/ANB1 and COX5A/COX5B , , . Either gene in each pair functions differently when environmental oxygen concentration changes, constituting an optimized oxygen response network. Most of these paralogs are in the list of ohnolog pairs, and might be formed as an outcome of selection for oxygen response after the whole-genome duplication. There are also a few exceptions among the aerobic/hypoxic(anaerobic) paralog pairs. For example, OYE3/OYE2  and AAC1/AAC3 (a triplet including PET9 (AAC2))  could derive from individual gene duplication. Sequence similarity search revealed that such aerobic/hypoxic(anaerobic) pair mostly has only one ortholog in K. lactis (Table 1). We may ask how these singular genes behave to respond to oxygen availability in K. lactis and question whether K. lactis and S. cerevisiae share similar gene categories involved in oxygen response.
KlHEM13, KlHYP2(ANB1) and KlCOX5A(5B) show differential responses to oxygen availability in K. lactis
HEM13, HYP2/ANB1 and COX5A/COX5B are well-known genes responding to oxygen in S. cerevisiae , , , –. These genes play important roles in heme synthesis, translational initiation and mitochondrial respiratory chain biogenesis, respectively. We therefore investigated the expression of their K. lactis orthologs under both aerobic condition and during the shift to hypoxic condition. The results of Northern hybridization are shown in Figure 2A and 2B.
Cells were grown to OD600 ∼4 at 28°C in complete YP glucose medium supplemented with ergosterol and Tween 80 under aerobic (O2) condition. A half of the culture was used for hypoxic treatment (N2) as described in the “Materials and Methods”, and then incubated at 28°C for a further 6 hours. Total RNA was extracted and Northern hybridization was performed to probe KlHEM13, KlHYP2(ANB1) and KlCOX5A(5B) transcripts (see Materials and Methods). Panel A: the K. lactis wild type MW270-7B and its isogenic mutant Δklrox1. Panel B: MW270-7B and its isogenic mutant Δklmot3. The RNA samples of MW270-7B in panels A and B were independent preparations. The rRNA was used for quantifying sample loads.
HEM13 of S. cerevisiae encodes coproporphyrinogen III oxidase . KlHEM13, its counterpart in K. lactis, is a singleton gene. The transcript level was slightly increased under hypoxic condition, though not as much as that of HEM13 in S. cerevisiae. This response of KlHEM13 was the same as that reported previously , in which cells were grown under hypoxic condition instead of here a shift from aerobic to hypoxic conditions.
Both HYP2/ANB1 and COX5A/COX5B, encoding translation initiation factor elF-5A  and subunit V of cytochrome c oxidase  respectively, are ohnolog pairs in S. cerevisiae , –. Each of these two gene pairs has only one unique ortholog, KlHYP2(ANB1) and KlCOX5A(5B) respectively, in the K. lactis genome. Analyses of sequence similarity (Table 1) and genomic context (http://wolfe.gen.tcd.ie/cgi/browser/ygob.pl?gene), however, could not decide which gene in the pair was closest to the K. lactis ortholog. For example, KlCOX5A(5B) shows more similarity to COX5A in sequence, but is more related to COX5B in terms of synteny and by presence of an intron. The situation is the same for KlHYP2(ANB1). Northern hybridization showed that the transcription of both KlHYP2(ANB1) and KlCOX5A(5B) was markedly reduced under hypoxic condition (Figure 2A and 2B), very much like the aerobic genes HYP2 and COX5A in the S. cerevisiae pairs, but unlike the hypoxic(anaerobic) ANB1 and COX5B. It is worth noting that KlCOX5A(5B) showed two transcripts of different size. The meaning of this observation is unknown, but a similar phenomenon was also reported for KlCYC1 ,  and HGT1 .
Simultaneous deletion of both HYP2 and ANB1 is lethal for S. cerevisiae . Therefore the singular KlHYP2(ANB1) could be also necessary for viability of K. lactis. In this sense, the decreased expression of KlHYP2(ANB1) under oxygen limited conditions in K. lactis may be a reason for its inability to grow in anaerobiosis. The cox5acox5b double deletion mutants of S. cerevisiae are completely non-respiratory and only grow by fermentation . Inactivation of respiration in K. lactis (for example disruption of the single cytochrome c gene KlCYC1) leads to severe growth defect even on glucose because of its limited fermentation capacity . Thus, KlCOX5A(5B) must be important for growth of K. lactis, even under aerobic condition. Therefore, the growth impact of a reduced KlCOX5A(5B) expression under hypoxic condition could be another possible reason of the distinct oxygen response of K. lactis.
Analysis of putative ROX1 orthologs suggests that there was a transition from the K. lactis-type ROX1 to the S. cerevisiae-type ROX1
ROX1 encodes a major repressor of the expression of many oxygen-responding genes such as HEM13, ANB1 and COX5B in S. cerevisiae , . The differential oxygen responses of KlHEM13, KlHYP2(ANB1) and KlCOX5A(5B) led us to question whether there exists a ROX1 ortholog in K. lactis. By a BLAST search using the amino acid sequence of Rox1p, we found in the K. lactis genome (Genolevures: http://www.genolevures.org/) a possible equivalent of ROX1: KLLA0B11495g, but with a limited E-value (7e−13). We designated the gene KlROX1. It predicts a protein of 393 amino acids, 45-kDa, in which a HMG (high-mobility group) domain was well conserved (44% identical) while other parts have diverged from Rox1p (368 amino acids). The HMG domain is localized at C-terminal part of KlRox1p, while it is at N-terminal part of Rox1p (Figure 3A and 3B). Also, there is no syntenic relation for these two genes. YPR063C and UBA3(YPR066W) are at upstream and downstream of ROX1 in S. cerevisiae, whilst the neighbour genes for KlROX1 are orthologous to MRPS5(YBR251W) and AFG3(YER017C) respectively (Figure 3A and 3B).
Panels A, B and C: the blank box indicates the HMG-encoding gene in which the solid box shows the location of the HMG-domain. The sizes are in proportion. Other types of boxes represent the chromosomal neighbors of ROX1 or KlROX1 and their orthologs in different yeast species. Panel D: alignment of the HMG-domains from various yeasts. The numbers indicate the amino-acid positions in the proteins. The asterisk and dot below the alignment show identity or conservative replacement of amino acid, respectively.
As described above, we defined KlROX1 solely on the basis of BLAST search. Its sequence similarity to ROX1 is limited. We may ask: do there exist such two kinds of ROX1 orthologs among yeasts? To examine this possibility, we analyzed putative ROX1 orthologs in other yeast species. In the yeasts S. paradoxus, S. mikatae and S. bayanus which are closely related to S. cerevisiae, a sequence significantly similar to the S. cerevisiae ROX1 (E value <e−88) was found in each of these genomes. The contexts of upstream YPR063C and downstream UBA3(YPR066W) genes are also conserved in these three yeasts (Figure 3A). Although BLAST results showed that the similarity was restricted to the HMG part in the Rox1p orthologs of S. castellii (1.5e−27) and S. kluyveri (5e−32), some sequence features such as glutamine stretches are still found in these proteins. Analysis of HMG localization and synteny indicated that these two genes are likely orthologs of the S. cerevisiae ROX1 (Figure 3A). However, these two yeasts contain another gene that may be orthologous to KlROX1 (4.1e−15 and 7e−19) (Figure 3B). The K. lactis-type gene, instead of the S. cerevisiae-type gene, is also present in the genomes of C. glabrata (2e−20) and K. waltii (3e−15) (Figure 3B). In the taxonomically distant yeast species, D. hansenii (9e−24), C. albicans (1.7e−20), Y. lipolytica (5e−18) and Schiz. pombe (3.9e−13), there still exist genes encoding HMG-containing proteins. However, except for a conserved HMG domain (Figure 3D), these genes differ much in their structural and syntenic features (Figure 3C).
Thus, in the process of the yeast evolution from the ancestral ‘pre-WGD’ species to the ‘post-WGD’ species, the HMG-encoding genes could have undergone a development from the K. lactis-type ROX1 gene to the S. cerevisiae-type ROX1 gene, through an intermediary co-existence of KlROX1 and ROX1. Importantly, we found that the formation of the S. cerevisiae-type ROX1 gene could connect with the development of the capability of anaerobic growth in the yeast evolution, as shown in Table 2.
Oxygen-dependent regulation of KlHEM13, KlHYP2(ANB1) and KlCOX5A(5B) is not mediated by KlROX1 in K. lactis
The analysis of possible ROX1 orthologs found in various yeast species has raised the question whether the changes from the K. lactis-type ROX1 to the S. cerevisiae-type ROX1 accompany a functional alteration of these HMG-encoding genes. In order to see whether KlROX1 has a transcriptional repression function similar to that of the S. cerevisiae ROX1, a Δklrox1 null mutant was constructed and used in the investigation of oxygen response in K. lactis. Northern hybridization showed that the klrox1 mutation had no obvious effect in the expression of KlHEM13, KlHYP2(ANB1) and KlCOX5A(5B), neither in aerobic nor hypoxic conditions (Figure 2A). Therefore, KlROX1 is not involved in the transcriptional repression of these three K. lactis genes that are orthologous to the well-known oxygen-responding genes of S. cerevisiae.
KlRox1p is not functionally equivalent to Rox1p in the repression of HEM13
In S. cerevisiae, the repression of Rox1p on the HEM13 expression has been extensively investigated , , . Exactly as reported, deletion of ROX1 derepressed the expression of HEM13 under aerobic growth (Figure 4A). The KlROX1 gene on a 2μ-based vector was then transformed into the S. cerevisiae Δrox1 mutant, and expression of KlRox1p was confirmed by Western blot (data not shown). As compared to the empty vector, introduction of KlROX1 led to an activation rather than to a repression on HEM13 expression (Figure 4B), suggesting different molecular functions between KlRox1p and Rox1p.
Cells were grown to OD600 1.5 to 2.0 at 28°C in minimal glucose medium supplemented with ergosterol and Tween 80 under aerobic (O2) or hypoxic (N2) conditions. Total RNA was extracted and Northern hybridization was performed to probe HEM13 transcript (see Materials and Methods). Panel A: the S. cerevisiae wild type W303-1B and its isogenic mutant Δrox1. Panel B: the Δrox1 mutant transformed with the empty vector pCM262 and the plasmid carrying KlROX1 in which three independent KlROX1 clones were included. The rRNA was used for sample loading quantification.
Search for the Rox1p-binding site YYYATTGTTCTC in the upstream regions of the K. lactis genes orthologous to the S. cerevisiae Rox1p target genes
In S. cerevisiae, Rox1p controls its target genes through binding to cis-acting elements with consensus sequence YYYATTGTTCTC . The HMG domain is responsible for the protein-DNA interaction ,  and its binding induces a topological change of 90° bending of DNA . Up to now, other genes coding for HMG-containing proteins (Figure 3C) have been characterized in Schiz. pombe ,  and C. albicans –. They encode either a mating-type M-specific polypeptide or a repressor of filamentous growth, the functions of which are completely divergent from that of Rox1p in S. cerevisiae. Although not involved in oxygen response, both proteins of Schiz. pombe and C. albicans could bind specifically to the YYYATTGTTCTC site, and the binding was dependent on the HMG domain , , .
Since the HMG domain is well conserved in KlRox1p (Figure 3D), we may expect that it would possess some of the Rox1p functions. KlRox1p could activate, but not repress, the expression of HEM13 in the S. cerevisiae Δrox1 mutant (Figure 4). We therefore supposed that the K. lactis protein, like the HMG proteins of other yeast species, could bind the YYYATTGTTCTC site. We then investigated the possible site in the upstream regions of the K. lactis genes, which are orthologous to the S. cerevisiae oxygen-responding genes mediated by Rox1p according to the data previously reported , , , . While DAN1, YHK8, TIR2, YGR035C, YAR028W and IRC23 are absent, 37 genes have their orthologs in K. lactis as revealed by BLAST search, in which 4 genes may have 2 putative K. lactis orthologs (Supplementary Table S3). 7 sites perfectly matched with the YYYATTGTTCTC motif present in 6 promoters of the S. cerevisiae genes, but no such site was found in any upstream sequences of the K. lactis genes. When 1 mismatch was allowed, only 2 upstream regions were detected 1 site each in K. lactis, much less than that in S. cerevisiae (13 promoters containing total 15 sites). The situation was similar when 2 mismatches were allowed (K. lactis: 13 sites in 11 upstream regions, S. cerevisiae: 43 in 27) (Supplementary Table S3 and Figure 5A).
Panel A: Comparison of the upstream regions (1 kb) containing the site YYYATTGTTCTC between the Rox1p-mediated hypoxic genes of S. cerevisiae (green circle) and their orthologs of K. lactis (red square). Panel B: stochastic model about the presence of the YYYATTGTTCTC site in a system containing 37 1-kb DNA sequences. Series white, prefect match; series yellow, 1 mismatch allowed; series blue, 2 mismatches allowed. Green bars represent the probabilities corresponding to the situations in S. cerevisiae and red in K. lactis.
KlMOT3 is dispensable for the regulation of the KlHEM13, KlHYP2(ANB1) and KlCOX5A(5B) in K. lactis
Our data above suggested that the Rox1p-mediated oxygen-responding system might not exist in K. lactis. Besides ROX1, MOT3 has also been reported to be involved in repressing a subset of hypoxic genes such as HEM13 and ANB1 in S. cerevisiae –, . By BLAST search using the amino-acid sequence of the S. cerevisiae Mot3p, we found a candidate ortholog KLLA0E18645g (designated KlMOT3) in K. lactis, with a moderate E-value (7.6e−28). The similarity between KlMot3p and Mot3p is concentrated in the C-terminal part containing two Cys2-His2 zinc fingers. But the gene synteny is conserved among different yeast species (http://wolfe.gen.tcd.ie/cgi/browser/ygob.pl?gene=YMR070W). To address whether K. lactis contains a functional MOT3 ortholog or not, a Δklmot3 null mutant was constructed to investigate its possible role in oxygen response of K. lactis. Northern hybridization indicated that the expression of KlHEM13, KlHYP2(ANB1) and KlCOX5A(5B) was not affected in the Δklmot3 mutant, under either aerobic or hypoxic conditions (Figure 2B). The result suggested that KlMOT3, as well as KlROX1, is not involved in the transcriptional repression of these three K. lactis genes expression and that KlMOT3 and MOT3 are not functional equivalents in their role of oxygen response.
The pre-WGD genome of K. lactis might lack a specific gene system responding to oxygen-limiting environment
In S. cerevisiae, the transcriptional repressor Rox1p and its target genes such as ANB1 and COX5B form a network devoted to respond to low oxygen environments. This repressor seems absent in K. lactis and the genes orthologous to ANB1 and COX5B showed a down-regulation response to low oxygen condition. Our results suggested that the Rox1p-mediated response to oxygen is not operating in this non-WGD yeast species. This proposition is consistent with a recent observation in which the hypoxic AAC3 gene of S. cerevisiae was expressed constitutively in K. lactis under both aerobic and hypoxic conditions .
The upstream YYYATTGTTCTC site connects the Rox1p target genes to the oxygen response in S. cerevisiae. We found that this site is much less abundant in the upstream regions of the forty-one K. lactis genes orthologous to the thirty-seven Rox1p target genes of S. cerevisiae. To evaluate the significance of such difference in the site occurrence, we calculated the stochastic probabilities that the YYYATTGTTCTC site is present at 0, 1, 2, 3, ……, n 1-kb DNA sequences in a system of thirty-seven members (Figure 5B) (the result was similar for a system of forty-one members). There was not any the perfectly matched site in the upstream regions of the K. lactis genes, the corresponding probability in the stochastic system was 0.97-almost a certain event. In the cases of 1 and 2 mismatches allowed, the probabilities were 0.08 and 3.17e−4 for K. lactis. On the contrary, all the probabilities for S. cerevisiae were 1.59e−12, 3.33e−15 and 2.11e−20 respectively. These data indicated that the K. lactis system is almost (or close to) a random site distribution while the S. cerevisiae system is highly selected with respect to the YYYATTGTTCTC site, suggesting that these K. lactis genes have not associated into a regulatory network which can respond synergistically to oxygen availability.
Appearance of the S. cerevisiae-type ROX1 gene could be a hallmark that links the ability of anaerobic growth in some yeasts
Some metabolites can be produced through either oxygen-dependent pathway or oxygen-independent pathway. Utilization of oxygen independent or less-dependent pathway is the reasonable choice for growth under oxygen limiting conditions. For example, NAD is synthesized from both tryptophan and niacin by the oxygen-requiring kynurenine pathway and the alternative oxygen-independent salvage pathway in S. cerevisiae. The salvage pathway has been proved to be necessary for the anaerobic growth for S. cerevisiae , but the variation in the modes of NAD acquisition in yeasts seems to have no direct link to the ability of anaerobic growth , since the salvage pathway for NAD synthesis exists in all yeast species including strict aerobes such as Y. lipolytica and Schiz. pombe (Table 2). However, it has been found that some metabolic processes are reconfigured to avoid dependence on oxygen-requiring reactions or respiration in the evolution towards anaerobic or hypoxic growth. For example, the enzyme dihydroorotate dehydrogenase (DHODase) in pyrimidine synthesis pathway was converted from a mitochondrial component into a cytoplasmic protein by horizontal gene transfer –. Also, birth of DAL gene cluster and loss of genes encoding urate oxidase (UOX) and urate permease (UAP) led a switch from urate to allantoin utilization . These biochemical reorganizations in the yeast evolution would economize oxygen consummation to adapt the life under oxygen-poor or –depleted environments.
Our analyses indicated that the S. cerevisiae-type ROX1 is specific for anaerobes including S. castellii and S. kluyveri , ,  and it is absent in aerobes (Table 2). ROX1 and its target hypoxic genes form a specialized network that actively takes challenges of oxygen-limiting or -absent conditions.
Formation of the Rox1p-mediated hypoxic genes and whole-genome duplication
The transcription analyses in this work revealed the expression of some important genes for cell viability such as KlHYP2(ANB1) and KlCOX5A(5B) was significantly repressed by hypoxic condition in the pre-WGD species K. lactis. Yeast has to devise a gene version capable of hypoxic(anaerobic) expression in order to survive in the absence of oxygen. Among 37 targets of Rox1p, 16 (43.2%) genes have an ohnolog and 13 (35.1%) singular genes are located within the duplication blocks (Supplementary Table S3), suggesting that as high as 78% of Rox1p-mediated genes may originate from the whole-genome duplication event. The whole-genome duplication would provide a basis for the construction of a hypoxic(anaerobic) working system with many genes. After the duplication, acquisition or creation of Rox1p-binding site could render one gene of a duplicated pair to become a hypoxic(anaerobic) version (Figure 6). Ideally, the paired genes function concertedly under aerobic and hypoxic(anaerobic) conditions. The capability of anaerobic growth of S. cerevisiae can be understood as a consequence of the whole-genome duplication that allowed acquisition of a new physiological property.
Materials and Methods
Strains and media
Yeast strains are listed in Table 3. Yeast cells were routinely grown at 28°C in a complete YP medium (1% yeast extract, 1% peptone, and 2% glucose), or synthetic minimal medium (0.67% Yeast Nitrogen Base without amino acid, and 2% glucose) supplemented with auxotrophic nutrients. The antibiotic G418 was added to the complete medium when required (200 µg ml−1). The media for hypoxic growth were complemented with 30 µg ml−1 of ergosterol and 0.2% Tween 80 (polyoxyethylene sorbitan monooleate), and the hypoxic condition was established through 5-minutes air evacuation and 5-minutes nitrogen filling, repeated three times in sealed flasks.
Deletion of the S. cerevisiae ROX1 gene was performed by a PCR (polymerase chain reaction)-based one step disruption procedure. A disruption cassette containing a KanMX gene was PCR-amplified (primers: 5′-TACTAATACTTCTTCACACAAAAGAACGCAGTTAGACAATCAACAGACATGGAGGCCCAGAATAC-3′ and 5′-ATAGTATAATATAATATAACGGAAAGAAGAAATGGAAAAAAAAAAGTATCGAATCGACAGCAGT-3′), and transformed into the S. cerevisiae strain W303-1B. Correct integration was verified by Southern hybridization: the entire open reading frame of ROX1, exactly from the initiation codon ATG to the stop codon TGA, was deleted and replaced by the KanMX gene to result in a Δrox1 mutant.
K. lactis Δklrox1 and Δklmot3 null mutants were constructed by “split-marker recombination” . The DNA fragments corresponding to the upstream and downstream flanking regions of KlROX1 or KlMOT3 were amplified by PCR (2 pairs of primers for KlROX1: 5′-CGGGATCCGATCTATTCTCATAACTTCGGG-3′, 5′-GGGGTACCTGACAAACCGACAGACTCATAC-3′ and 5′-CGGGATCCGAGAAACTCTGGTTCTAGTCCG-3′, 5′-TACCATTCGGCATATCACAG-3′; for KlMOT3: 5′-CGGGATCCGTTCAGCCGTGTGCATCTTC-3′, 5′-GGGGTACCATTCGGACCTATATCAGCATCC-3′ and 5′-CGGGATCCGTGTAGCTAACACCGCGTTGG-3′, 5′-GGGGTACCTGTCTTGCGTTTGGTATTGC-3′), and cloned into pKA and pAN vectors  respectively. The resulting plasmids were co-transformed into the K. lactis strain MW270-7B. Expected structure of integration was confirmed by Southern hybridization: the sequences from the 7th codon to the 373rd codon of KlROX1 and from the 15th codon to the 429th codon of KlMOT3 were deleted and replaced by a KanMX selection marker respectively, to give a Δklrox1 mutant and a Δklmot3 mutant.
Cloning of the KlROX1 gene into the expression vector
The open reading frame of KlROX1 was amplified by PCR (primers: 5′-GCGCGGCCGCATGAGTCTGTCGGTTTGTCATAGAC-3′ and 5′-GCCTGCAGTTTGATTTTGGGGATTGCTCT-3′) from K. lactis genomic DNA, and inserted into the NotI-PstI site of a 2μ-based multi-copy vector pCM262 . The resulting plasmid pCM262/KlROX1 contained a KlROX1-HA (a haemoagglutinin epitope) fusion expression cassette.
Cell-free protein extraction and Western blot
Cells grown to an early stationary phase in minimal medium were harvested, washed and resuspended in a buffer containing 20 mM Tris-HCl, pH 8.0, 50 mM ammonium acetate, 2 mM EDTA and 2 mM phenylmethysulfonyl fluoride (PMSF). One volume of 20% trichloroacetic acid (TCA) was added. The cells were then disrupted with glass beads four times by a vigorous shaking for 1 minute followed by a cooling on ice for 1 minute. The mixture was centrifuged and the protein pellet was washed with acetone and dissolved in gel loading buffer. Proteins were separated in 10% SDS-polyacrylamide gels, transferred to Hybond-C Extra membrane (Amersham), and probed with specific monoclonal antibody 12CA5 for the HA epitope. The reacted protein band was visualized using the ECL chemiluminescence detection system (Amersham).
RNA extraction and Northern hybridization
Total RNA was isolated as described , fractionated on an agarose/formaldehyde denaturing gel, and immobilized onto a Hybond-N+ membrane (Amersham). Hybridization was performed at 65°C in a buffer containing 7% SDS, 0.5 M sodium phosphate buffer, pH 7.2, and 10 mM EDTA. Probes were synthesized by PCR (oligonucleotides for HEM13 amplification: 5′-GATCCAAGGAATCTTCCAAT-3′ and 5′-TAACCATGAAGCATGTTCAG-3′; for KlHEM13: 5′-TCCATTCGACTCACCAACTG-3′ and 5′-TTCAAACCCATTCAACAGGG-3′; for KlHYP2(ANB1): 5′-CCAAACGCATTAAACAAATCA-3′ and 5′-CTTCTTTCCATTTATCCAGGG-3′, and for KlCOX5A(5B): 5′-CCACTTGCAATTATCGTCTGA-3′ and 5′-AGAAGAGAGGGAGAAAATGCA-3′), and labelled with 32P using the ‘Ready to Go DNA Labelling Kit’ (Pharmacia).
Comparative genome analysis in yeasts
BLAST search was performed using tools implemented at the database websites of the yeast species (S. cerevisiae, Saccharomyces paradoxus, Saccharomyces mikatae, Saccharomyces bayanus, Saccharomyces castellii, Saccharomyces kluyveri, and Kluyveromyces waltii: http://db.yeastgenome.org/cgi-bin/FUNGI/showAlign or/and http://seq.yeastgenome.org/cgi-bin/blast-fungal.pl?name; Candida glabrata, K. lactis, Debaryomyces hansenii and Yarrowia lipolytica: http://cbi.labri.fr/Genolevures/blast/index.php; Candida albicans: http://www.candidagenome.org/cgi-bin/nph-blast; Schizosaccharomyces. pombe: http://www.genedb.org/genedb/pombe/blast.jsp).
Synteny comparison was carried out manually or viewed at http://db.yeastgenome.org/cgi-bin/FUNGI/FungiMap?locus (S. cerevisiae, S. paradoxus, S. mikatae and S. bayanus) and http://wolfe.gen.tcd.ie/cgi/browser/ygob.pl?gene (S. cerevisiae, S. bayanus, C. glabrata, S. castellii, K. waltii, S. kluyveri, and K. lactis).
Cumulative binomial distribution.
To determine the significance of the difference of the fraction of a certain property between two samples (i.e., datasets), we used the formula below to calculate P-values:where N is the total number of genes in the testing sample, c is the number of genes with a specific property and the number of observed genes with this property, and p is the probability of finding a gene with the same property randomly (picking from the entire genome) or in the control sample.
Stochastic model of the site YYYATTGTTCTC.
According to a previous report about the DNA sequence requirements of the consensus Rox1p-binding site Y1Y2Y3A4T5T6G7T8T9C10T11C12 , the sequence T5T6G7T8 is absolutely required for the affinity of Rox1p binding; A4 can be relatively tolerated to substitution by T, T9 by C or A and Y3 by A; A4 and T9 can not be permitted to change at the same time. With these limitations of mismatches allowed, we searched for the possible binding sites in the upstream sequences (1 kb) of both the thirty-seven Rox1p-target genes of S. cerevisiae and their forty-one orthologs of K. lactis (see Results). And we also calculated the stochastic probabilities that the site YYYATTGTTCTC appears in thirty-seven and forty-one 1-kb DNA sequences as described below:
Lemma 1 : Let , …, denote the events in some probability space. Then, and holds for all , where the sum is taken over all nonnegative integers with .
Let denote the string “”, where , , and . Define , , , , , , and . Let and (prefect match, 1 mismatch allowed, 2 mismatches allowed).
DNA is double-stranded. For complementary strand, let denote the string “”, where , and . Define , , , , , , and . Let . Define (prefect match, 1 mismatch allowed, 2 mismatches allowed).
Let be the probability space consisting of all the character strings of length on set with equality probability. Now we can choose a string from at random. For , let () be the event that contains substring () and () be the event that the substring of consisting of st character to st character is (). Then and .
By Lemma 1, we can have that and . By symmetry, . Since , we can know that , in which and .
Let . If , , that is ; If , , that is ; If , , that is, .
Now we can choose strings of length from . For , let be the number of the strings that contains substring or . Let , then , that is , where .
List of oxygen-responding genes in the ohnologs of S. cerevisiae.
(0.48 MB XLS)
Significance of distribution of oxygen-responding genes in the ohnologs of S. cerevisiae.
(0.10 MB DOC)
We are very grateful to Dr Hiroshi Fukuhara (Orsay) for critical reading and suggestions to improve the manuscript. And we would like to thank Dr. Richard S. Zitomer (New York) for kindly providing us with his Δrox1 and Δmot3 strains, Mr. Christophe Leplat (Orsay) for his assistance in hypoxic treatment, and Dr. Li Liu (Lanzhou University) for his help in computing.
Conceived and designed the experiments: WGB. Performed the experiments: ZAF YFL JPL YYL WGB. Analyzed the data: WGB. Wrote the paper: MBF WGB. Performed the mathematical analyses. Performed the mathematical analyses: AC.
- 1. Hon T, Dodd A, Dirmeier R, Gorman N, Sinclair PR, et al. (2003) A mechanism of oxygen sensing in yeast. Multiple oxygen-responsive steps in the heme biosynthetic pathway affect Hap1 activity. J Biol Chem 278: 50771–50780.
- 2. Kastaniotis AJ, Mennella TA, Konrad C, Torres AM, Zitomer RS (2000) Roles of transcription factor Mot3 and chromatin in repression of the hypoxic gene ANB1 in yeast. Mol Cell Biol 20: 7088–7098.
- 3. Kwast KE, Lai LC, Menda N, James DT 3rd, Aref S, et al. (2002) Genomic analyses of anaerobically induced genes in Saccharomyces cerevisiae: functional roles of Rox1 and other factors in mediating the anoxic response. J Bacteriol 184: 250–265.
- 4. Ter Linde JJ, Steensma HY (2002) A microarray-assisted screen for potential Hap1 and Rox1 target genes in Saccharomyces cerevisiae. Yeast 19: 825–840.
- 5. Zhang L, Hach A (1999) Molecular mechanism of heme signaling in yeast: the transcriptional activator Hap1 serves as the key mediator. Cell Mol Life Sci 56: 415–426.
- 6. Zitomer RS, Lowry CV (1992) Regulation of gene expression by oxygen in Saccharomyces cerevisiae. Microbiol Rev 56: 1–11.
- 7. Wolfe KH, Shields DC (1997) Molecular evidence for an ancient duplication of the entire yeast genome. Nature 387: 708–713.
- 8. Keogh RS, Seoighe C, Wolfe KH (1998) Evolution of gene order and chromosome number in Saccharomyces, Kluyveromyces and related fungi. Yeast 14: 443–457.
- 9. Kellis M, Birren BW, Lander ES (2004) Proof and evolutionary analysis of ancient genome duplication in the yeast Saccharomyces cerevisiae. Nature 428: 617–624.
- 10. Merico A, Sulo P, Piskur J, Compagno C (2007) Fermentative lifestyle in yeasts belonging to the Saccharomyces complex. Febs J 274: 976–989.
- 11. Ishtar Snoek IS, Yde Steensma H (2007) Factors involved in anaerobic growth of Saccharomyces cerevisiae. Yeast 24: 1–10.
- 12. Bolotin-Fukuhara M, Casaregola S, Aigle M (2006) Genome evolution: Lessons from Genolevures. In: Sunnerhagen P, Piskur J, editors. Comparative genomics: using fungi as models. Heidelberg: Springer. pp. 165–196.
- 13. Conant GC, Wolfe KH (2007) Increased glycolytic flux as an outcome of whole-genome duplication in yeast. Mol Syst Biol 3: 129.
- 14. Lowry CV, Lieber RH (1986) Negative regulation of the Saccharomyces cerevisiae ANB1 gene by heme, as mediated by the ROX1 gene product. Mol Cell Biol 6: 4145–4148.
- 15. Schnier J, Schwelberger HG, Smit-McBride Z, Kang HA, Hershey JW (1991) Translation initiation factor 5A and its hypusine modification are essential for cell viability in the yeast Saccharomyces cerevisiae. Mol Cell Biol 11: 3105–3114.
- 16. Hodge MR, Kim G, Singh K, Cumsky MG (1989) Inverse regulation of the yeast COX5 genes by oxygen and heme. Mol Cell Biol 9: 1958–1964.
- 17. Burke PV, Raitt DC, Allen LA, Kellogg EA, Poyton RO (1997) Effects of oxygen concentration on the expression of cytochrome c and cytochrome c oxidase genes in yeast. J Biol Chem 272: 14705–14712.
- 18. Kolarov J, Kolarova N, Nelson N (1990) A third ADP/ATP translocator gene in yeast. J Biol Chem 265: 12711–12716.
- 19. Thorsness M, Schafer W, D'Ari L, Rine J (1989) Positive and negative transcriptional control by heme of genes encoding 3-hydroxy-3-methylglutaryl coenzyme A reductase in Saccharomyces cerevisiae. Mol Cell Biol 9: 5702–5712.
- 20. Sertil O, Kapoor R, Cohen BD, Abramova N, Lowry CV (2003) Synergistic repression of anaerobic genes by Mot3 and Rox1 in Saccharomyces cerevisiae. Nucleic Acids Res 31: 5831–5837.
- 21. Klinkenberg LG, Mennella TA, Luetkenhaus K, Zitomer RS (2005) Combinatorial repression of the hypoxic genes of Saccharomyces cerevisiae by DNA binding proteins Rox1 and Mot3. Eukaryot Cell 4: 649–660.
- 22. Mennella TA, Klinkenberg LG, Zitomer RS (2003) Recruitment of Tup1-Ssn6 by yeast hypoxic genes and chromatin-independent exclusion of TATA binding protein. Eukaryot Cell 2: 1288–1303.
- 23. Fukuhara H (2003) The Kluyver effect revisited. FEMS Yeast Res 3: 327–331.
- 24. Scannell DR, Butler G, Wolfe KH (2007) Yeast genome evolution-the origin of the species. Yeast.
- 25. Bolotin-Fukuhara M, Toffano-Nioche C, Artiguenave F, Duchateau-Nguyen G, Lemaire M, et al. (2000) Genomic exploration of the hemiascomycetous yeasts: 11. Kluyveromyces lactis. FEBS Lett 487: 66–70.
- 26. Freire-Picos MA, Hollenberg CP, Breunig KD, Cerdan ME (1995) Regulation of cytochrome c expression in the aerobic respiratory yeast Kluyveromyces lactis. FEBS Lett 360: 39–42.
- 27. Gonzalez-Dominguez M, Freire-Picos MA, Ramil E, Guiard B, Cerdan ME (2000) Heme-mediated transcriptional control in Kluyveromyces lactis. Curr Genet 38: 171–177.
- 28. Fontanesi F, Viola AM, Ferrero I (2006) Heterologous complementation of the Klaac null mutation of Kluyveromyces lactis by the Saccharomyces cerevisiae AAC3 gene encoding the ADP/ATP carrier. FEMS Yeast Res 6: 414–420.
- 29. Bao WG, Guiard B, Fang ZA, Donnini C, Gervais M, et al. (2008) Oxygen-dependent transcriptional regulator Hap1p limits glucose uptake by repressing the expression of the major glucose transporter gene RAG1 in Kluyveromyces lactis. Eukaryot Cell 7: 1895–1905.
- 30. ter Linde JJ, Liang H, Davis RW, Steensma HY, van Dijken JP, et al. (1999) Genome-wide transcriptional analysis of aerobic and anaerobic chemostat cultures of Saccharomyces cerevisiae. J Bacteriol 181: 7409–7413.
- 31. Byrne KP, Wolfe KH (2005) The Yeast Gene Order Browser: combining curated homology and syntenic context reveals gene fate in polyploid species. Genome Res 15: 1456–1461.
- 32. Wong S, Butler G, Wolfe KH (2002) Gene order evolution and paleopolyploidy in hemiascomycete yeasts. Proc Natl Acad Sci U S A 99: 9272–9277.
- 33. Dietrich FS, Voegeli S, Brachat S, Lerch A, Gates K, et al. (2004) The Ashbya gossypii genome as a tool for mapping the ancient Saccharomyces cerevisiae genome. Science 304: 304–307.
- 34. Chen H, Xu L, Gu Z (2008) Regulation dynamics of WGD genes during yeast metabolic oscillation. Mol Biol Evol 25: 2513–2516.
- 35. Piskur J (2001) Origin of the duplicated regions in the yeast genomes. Trends Genet 17: 302–303.
- 36. Kwast KE, Burke PV, Poyton RO (1998) Oxygen sensing and the transcriptional regulation of oxygen-responsive genes in yeast. J Exp Biol 201: 1177–1195.
- 37. Niino YS, Chakraborty S, Brown BJ, Massey V (1995) A new old yellow enzyme of Saccharomyces cerevisiae. J Biol Chem 270: 1983–1991.
- 38. Amillet JM, Buisson N, Labbe-Bois R (1995) Positive and negative elements involved in the differential regulation by heme and oxygen of the HEM13 gene (coproporphyrinogen oxidase) in Saccharomyces cerevisiae. Curr Genet 28: 503–511.
- 39. Verdiere J, Gaisne M, Labbe-Bois R (1991) CYP1 (HAP1) is a determinant effector of alternative expression of heme-dependent transcribed genes in yeast [corrected]. Mol Gen Genet 228: 300–306.
- 40. Hodge MR, Singh K, Cumsky MG (1990) Upstream activation and repression elements control transcription of the yeast COX5b gene. Mol Cell Biol 10: 5510–5520.
- 41. Trueblood CE, Poyton RO (1988) Identification of REO1, a gene involved in negative regulation of COX5b and ANB1 in aerobically grown Saccharomyces cerevisiae. Genetics 120: 671–680.
- 42. Trueblood CE, Wright RM, Poyton RO (1988) Differential regulation of the two genes encoding Saccharomyces cerevisiae cytochrome c oxidase subunit V by heme and the HAP2 and REO1 genes. Mol Cell Biol 8: 4537–4540.
- 43. Zagorec M, Buhler JM, Treich I, Keng T, Guarente L, et al. (1988) Isolation, sequence, and regulation by oxygen of the yeast HEM13 gene coding for coproporphyrinogen oxidase. J Biol Chem 263: 9718–9724.
- 44. Schwelberger HG, Kang HA, Hershey JW (1993) Translation initiation factor eIF-5A expressed from either of two yeast genes or from human cDNA. Functional identity under aerobic and anaerobic conditions. J Biol Chem 268: 14018–14025.
- 45. Cooper CE, Nicholls P, Freedman JA (1991) Cytochrome c oxidase: structure, function, and membrane topology of the polypeptide subunits. Biochem Cell Biol 69: 586–607.
- 46. Freire-Picos MA, Lombardia-Ferreira LJ, Ramil E, Gonzalez-Dominguez M, Cerdan ME (2001) The KlCYC1 gene, a downstream region for two differentially regulated transcripts. Yeast 18: 1347–1355.
- 47. Seoane S, Guiard B, Rodriguez-Torres AM, Freire-Picos MA (2005) Effects of splitting alternative KlCYC1 3′-UTR regions on processing: metabolic consequences and biotechnological applications. J Biotechnol 118: 149–156.
- 48. Billard P, Menart S, Blaisonneau J, Bolotin-Fukuhara M, Fukuhara H, et al. (1996) Glucose uptake in Kluyveromyces lactis: role of the HGT1 gene in glucose transport. J Bacteriol 178: 5860–5866.
- 49. Herrmann JM, Funes S (2005) Biogenesis of cytochrome oxidase-sophisticated assembly lines in the mitochondrial inner membrane. Gene 354: 43–52.
- 50. Chen XJ, Clark-Walker GD (1993) Mutations in MGI genes convert Kluyveromyces lactis into a petite-positive yeast. Genetics 133: 517–525.
- 51. Kastaniotis AJ, Zitomer RS (2000) Rox1 mediated repression. Oxygen dependent repression in yeast. Adv Exp Med Biol 475: 185–195.
- 52. Balasubramanian B, Lowry CV, Zitomer RS (1993) The Rox1 repressor of the Saccharomyces cerevisiae hypoxic genes is a specific DNA-binding protein with a high-mobility-group motif. Mol Cell Biol 13: 6071–6078.
- 53. Deckert J, Torres AM, Hwang SM, Kastaniotis AJ, Zitomer RS (1998) The anatomy of a hypoxic operator in Saccharomyces cerevisiae. Genetics 150: 1429–1441.
- 54. Deckert J, Rodriguez Torres AM, Simon JT, Zitomer RS (1995) Mutational analysis of Rox1, a DNA-bending repressor of hypoxic genes in Saccharomyces cerevisiae. Mol Cell Biol 15: 6109–6117.
- 55. Kelly M, Burke J, Smith M, Klar A, Beach D (1988) Four mating-type genes control sexual differentiation in the fission yeast. Embo J 7: 1537–1547.
- 56. Dooijes D, van de Wetering M, Knippels L, Clevers H (1993) The Schizosaccharomyces pombe mating-type gene mat-Mc encodes a sequence-specific DNA-binding high mobility group box protein. J Biol Chem 268: 24813–24817.
- 57. Kadosh D, Johnson AD (2001) Rfg1, a protein related to the Saccharomyces cerevisiae hypoxic regulator Rox1, controls filamentous growth and virulence in Candida albicans. Mol Cell Biol 21: 2496–2505.
- 58. Kadosh D, Johnson AD (2005) Induction of the Candida albicans filamentous growth program by relief of transcriptional repression: a genome-wide analysis. Mol Biol Cell 16: 2903–2912.
- 59. Khalaf RA, Zitomer RS (2001) The DNA binding protein Rfg1 is a repressor of filamentation in Candida albicans. Genetics 157: 1503–1512.
- 60. van Beest M, Dooijes D, van De Wetering M, Kjaerulff S, Bonvin A, et al. (2000) Sequence-specific high mobility group box factors recognize 10-12-base pair minor groove motifs. J Biol Chem 275: 27266–27273.
- 61. Panozzo C, Nawara M, Suski C, Kucharczyka R, Skoneczny M, et al. (2002) Aerobic and anaerobic NAD+ metabolism in Saccharomyces cerevisiae. FEBS Lett 517: 97–102.
- 62. Li YF, Bao WG (2007) Why do some yeast species require niacin for growth? Different modes of NAD synthesis. FEMS Yeast Res 7: 657–664.
- 63. Nagy M, Lacroute F, Thomas D (1992) Divergent evolution of pyrimidine biosynthesis between anaerobic and aerobic yeasts. Proc Natl Acad Sci U S A 89: 8966–8970.
- 64. Gojkovic Z, Knecht W, Zameitat E, Warneboldt J, Coutelis JB, et al. (2004) Horizontal gene transfer promoted evolution of the ability to propagate under anaerobic conditions in yeasts. Mol Genet Genomics 271: 387–393.
- 65. Piskur J, Langkjaer RB (2004) Yeast genome sequencing: the power of comparative genomics. Mol Microbiol 53: 381–389.
- 66. Hall C, Brachat S, Dietrich FS (2005) Contribution of horizontal gene transfer to the evolution of Saccharomyces cerevisiae. Eukaryot Cell 4: 1102–1115.
- 67. Wong S, Wolfe KH (2005) Birth of a metabolic gene cluster in yeast by adaptive gene relocation. Nat Genet 37: 777–782.
- 68. Moller K, Olsson L, Piskur J (2001) Ability for anaerobic growth is not sufficient for development of the petite phenotype in Saccharomyces kluyveri. J Bacteriol 183: 2485–2489.
- 69. Fairhead C, Llorente B, Denis F, Soler M, Dujon B (1996) New vectors for combinatorial deletions in yeast chromosomes and for gap-repair cloning using ‘split-marker’ recombination. Yeast 12: 1439–1457.
- 70. Rodriguez-Navarro S, Llorente B, Rodriguez-Manzaneque MT, Ramne A, Uber G, et al. (2002) Functional analysis of yeast gene families involved in metabolism of vitamins B1 and B6. Yeast 19: 1261–1276.
- 71. Schmitt ME, Brown TA, Trumpower BL (1990) A rapid and simple method for preparation of RNA from Saccharomyces cerevisiae. Nucleic Acids Res 18: 3091–3092.
- 72. Feller W (1970) An introduction to probability theory and its applications. New York: John Wiley & Sons. v. p.