Search

Saccharomyces bayanus var. uvarum investigated here is the species closest to Saccharomyces cerevisiae. Random sequence tags (RSTs) allowed us to identify homologues to 2789 open reading frames (ORFs) in S. cerevisiae, ORFs duplicated in S. uvarum but not in S. cerevisiae, centromeres, tRNAs, homologues of Ty1/2 and Ty4 retrotransposons, and a complete rDNA repeat. Only 13 RSTs seem to be homologous to sequences in other organisms but not in S. cerevisiae. As the synteny between the two species is very high, cases in which synteny is lost suggest special mechanisms of genome evolution. The corresponding RSTs revealed that S. uvarum can exist without any S. cerevisiae DNA introgression. Accession numbers are from AL397139 to AL402278 in the EMBL databank.

Since its completion more than 4 years ago, the sequence of Saccharomyces cerevisiae has been extensively used and studied. The original sequence has received a few corrections, and the identification of genes has been completed, thanks in particular to transcriptome analyses and to specialized studies on introns, tRNA genes, transposons or multigene families. In order to undertake the extensive comparative sequence analysis of this program, we have entirely revisited the S. cerevisiae sequence using the same criteria for all 16 chromosomes and taking into account publicly available annotations for genes and elements that cannot be predicted. Comparison with the other yeast species of this program indicates the existence of 50 novel genes in segments previously considered as 'intergenic' and suggests extensions for 26 of the previously annotated genes.

We have analyzed the evolution of chromosome maps of Hemiascomycetes by comparing gene order and orientation of the 13 yeast species partially sequenced in this program with the genome map of Saccharomyces cerevisiae. From the analysis of nearly 8000 situations in which two distinct genes having homologs in S. cerevisiae could be identified on the sequenced inserts of another yeast species, we have quantified the loss of synteny, the frequency of single gene deletion and the occurrence of gene inversion. Traces of ancestral duplications in the genome of S. cerevisiae could be identified from the comparison with the other species that do not entirely coincide with those identified from the comparison of S. cerevisiae with itself. From such duplications and from the correlation observed between gene inversion and loss of synteny, a model is proposed for the molecular evolution of Hemiascomycetes. This model, which can possibly be extended to other eukaryotes, is based on the reiteration of events of duplication of chromosome segments, creating transient merodiploids that are subsequently resolved by single gene deletion events.

The identification of molecular evolutionary mechanisms in eukaryotes is approached by a comparative genomics study of a homogeneous group of species classified as Hemiascomycetes. This group includes Saccharomyces cerevisiae, the first eukaryotic genome entirely sequenced, back in 1996. A random sequencing analysis has been performed on 13 different species sharing a small genome size and a low frequency of introns. Detailed information is provided in the 20 following papers. Additional tables available on websites describe the ca. 20000 newly identified genes. This wealth of data, so far unique among eukaryotes, allowed us to examine the conservation of chromosome maps, to identify the 'yeast-specific' genes, and to review the distribution of gene families into functional classes. This project conducted by a network of seven French laboratories has been designated 'Génolevures'.

We have evaluated the degree of gene redundancy in the nuclear genomes of 13 hemiascomycetous yeast species. Saccharomyces cerevisiae singletons and gene families appear generally conserved in these species as singletons and families of similar size, respectively. Variations of the number of homologues with respect to that expected affect from 7 to less than 24% of each genome. Since S. cerevisiae homologues represent the majority of the genes identified in the genomes studied, the overall degree of gene redundancy seems conserved across all species. This is best explained by a dynamic equilibrium resulting from numerous events of gene duplication and deletion rather than by a massive duplication event occurring in some lineages and not in others.

We explored the biological diversity of hemiascomycetous yeasts using a set of 22000 newly identified genes in 13 species through BLASTX searches. Genes without clear homologue in Saccharomyces cerevisiae appeared to be conserved in several species, suggesting that they were recently lost by S. cerevisiae. They often identified well-known species-specific traits. Cases of gene acquisition through horizontal transfer appeared to occur very rarely if at all. All identified genes were ascribed to functional classes. Functional classes were differently represented among species. Species classification by functional clustering roughly paralleled rDNA phylogeny. Unequal distribution of rapidly evolving, ascomycete-specific, genes among species and functions was shown to contribute strongly to this clustering. A few cases of gene family amplification were documented, but no general correlation could be observed between functional differentiation of yeast species and variations of gene family sizes. Yeast biological diversity seems thus to result from limited species-specific gene losses or duplications, and for a large part from rapid evolution of genes and regulatory factors dedicated to specific functions.

Comparisons of the 6213 predicted Saccharomyces cerevisiae open reading frame (ORF) products with sequences from organisms of other biological phyla differentiate genes commonly conserved in evolution from 'maverick' genes which have no homologue in phyla other than the Ascomycetes. We show that a majority of the 'maverick' genes have homologues among other yeast species and thus define a set of 1892 genes that, from sequence comparisons, appear 'Ascomycetes-specific'. We estimate, retrospectively, that the S. cerevisiae genome contains 5651 actual protein-coding genes, 50 of which were identified for the first time in this work, and that the present public databases contain 612 predicted ORFs that are not real genes. Interestingly, the sequences of the 'Ascomycetes-specific' genes tend to diverge more rapidly in evolution than that of other genes. Half of the 'Ascomycetes-specific' genes are functionally characterized in S. cerevisiae, and a few functional categories are over-represented in them.

The primary analysis of the sequences for our Hemiascomycete random sequence tag (RST) project was performed using a combination of classical methods for sequence comparison and contig assembly, and of specifically written scripts and computer visualization routines. Comparisons were performed first against DNA and protein sequences from Saccharomyces cerevisiae, then against protein sequences from other completely sequenced organisms and, finally, against protein sequences from all other organisms. Blast alignments were individually inspected to help recognize genes within our random genomic sequences despite the fact that only parts of them were available. For each yeast species, validated alignments were used to infer the proper genetic code, to determine codon usage preferences and to calculate their degree of sequence divergence with S. cerevisiae. The quality of each genomic library was monitored from contig analysis of the DNA sequences. Annotated sequences were submitted to the EMBL database, and the general annotation tables produced served as a basis for our comparative description of the evolution, redundancy and function of the Hemiascomycete genomes described in other articles of this issue.

Using genetic hybridization analysis, electrophoretic karyotyping and PCR-RFLP of the MET2 gene, we found that the yeast Saccharomyces bayanus var. uvarum is associated with certain types of wines produced in the Val de Loire, Sauternes, and Jurancon regions. The average frequency of appearance of this yeast in the three regions of France was 41, 7 and 77%, respectively. In contrast, we did not find S. bayanus var. uvarum in red wines produced in the Bordeaux area. The results of this study, as well as the findings already reported on Tokay (Slovakia), Muscat (Crimea, Ukraine) and Amarone (Italy) wines, lead us to consider that distribution of S. bayanus var. uvarum yeast is connected with low temperature climatic conditions and/or wine technologies in which must fermentation is at least partially carried out at low temperatures (10-15 degrees C).

The Rvs161p and Rvs167p proteins of Saccharomyces cerevisiae, homologues of higher eukaryotes' amphiphysins, associate with actin and appear to be involved in several functions related to the actin cytoskeleton. In order to identify partners of the Rvsp proteins, yeast libraries constructed in two-hybrid vectors were screened using either Rvs167p or Rvs161p as a bait. The selected candidates, representing 34 ORFs, were then tested against both Rvsp proteins, as well as domains of Rvs167p or Rvs161p. Among the most significant ones, 24 ORFs were specific preys of Rvs167p only and two gave interactions with Rvs161p only. Interestingly, five ORFs were preys of both Rvs161p and Rvs167p (RVS167, LAS17, YNL094w, YMR192w and YPL249c). Analysis of putative functions of the candidates confirm involvement of the Rvsp in endocytosis/vesicle traffic, but also opens possible new fields, such as nuclear functions., (Copyright 2000 John Wiley & Sons, Ltd.)

The rvs mutants display phenotypes close to those described for the actin mutants: disorganization of the actin cytoskeleton, random budding of the diploids, loss of polarity and sensitivity to salt. Mutations in the RVS genes lead to synthetic lethality with a set of mutations in the actin gene, ACT1. This synthetic lethality is allele-specific regarding the act1 mutations, pointing to a region on the actin molecule where contacts with the myosin head have been described. The possible involvement of a myosin in a vital function fulfilled both by the Rvsp proteins and actin is strengthened further by the fact that the double mutants rvs167, myo1 and rvs167, myo2 are lethal and severely affected in growth respectively. These data support the idea that actin, myosin and Rvsp proteins are linked in a common functional pathway in yeast.

As previous studies indicated that the RVS161 and RVS167 gene products of Saccharomyces cerevisiae seem to be involved in the same cellular function, we considered the possibility of a complex between the proteins encoded by these two genes. Using the two hybrid system, we have shown that Rvs167p interacts with Rvs161p, through its N-terminal domain which contains predicted coiled-coil structures. Moreover, if a tagged Rvs 167p protein was immuno adsorbed under non-denaturing conditions, it brought along the Rvs161p protein. These results confirmed the hypothesis of an in vivo complex between the two proteins.

Mutations in RVS161 and RVS167 yeast genes induce identical phenotypes associated to actin cytoskeleton disorders. The whole Rvs161 protein is similar to the amino-terminal part of Rvs167p, thus defining a RVS domain. In addition to this domain, Rvs167p contains a central glycine-proline-alanine rich domain and a SH3 domain. To assess the function of these different domains we have expressed recombinant Rvs proteins in rvs mutant strains. Phenotype analysis has shown that the RVS and SH3 domains are necessary for phenotypical complementation, whereas the GPA domain is not. Moreover, we have demonstrated that the RVS domains from Rvs161p and Rvs167p have distinct roles, and that the SH3 domain needs the specific RVS domain of Rvs167p to function. These results suggest that Rvs161p and Rvs167p play distinct roles, while acting together in a common function.

The rvs161 and rvs167 mutant cells exhibit several identical phenotypes including sensitivity to several different growth conditions and morphological defects such as alteration of the actin cytoskeleton and budding patterns. The selection of genes that, when overexpressed, are able to suppress the reduced viability upon carbon starvation of the rvs167 mutant strain, has allowed the cloning of the SUR7 gene (Accession Number Z46729x11). We showed that the suppressive ability of the overexpressed SUR7 gene concerns all the rvs167 phenotypes. However, this suppression is only partial since the rvs167-suppressed strain is not of wild-type phenotype. Moreover, SUR7 is also able to suppress partially the phenotypes exhibited by the rvs161 and rvs167 and rvs161 mutant strains. The SUR7 gene encodes a putative integral membrane protein with four transmembrane domains. Furthermore, sequence comparisons revealed that Sur7p and two other proteins, Yn1194p and Yd1222p, present significant sequence and structural similarities. Taken together, these results strongly suggest that the Rvs161 and Rvs167 proteins act together in relation with Sur7p. Moreover, the putative transmembranous character of Sur7p suggests a membrane localization of the Rvs function, a localization which is consistent with the different rvs phenotypes and the actin-Rvs167p interaction.

Several yeast strains of the species Saccharomyces cerevisiae, S. bayanus and S. paradoxus, first identified by hybridization experiments and measurements of DNA/DNA homology, were characterized using polymerase chain reaction/restriction fragment length polymorphism (PCR/RFLP) analysis of the MET2 gene. There was no exception to the agreement between this method and classical genetic analyses for any of the strains examined, so PCR/RFLP of the MET2 gene is a reliable and fast technique for delimiting S. cerevisiae and S. bayanus. Enological strains classified as S. bayanus, S. chevalieri, and S. capensis gave S. cerevisiae restriction patterns, whereas most S. uvarum strains belong to S. bayanus. Enologists should no longer use the name of S. bayanus for S. cerevisiae Gal strains, and should consider S. bayanus as a distinct species.

The aim of this paper is to gather and complete data about four members of a new gene family. Mutation in SUR4 gene was originally selected as a suppressor of defects caused by mutations in RVS161 or RVS167 genes. Cloning and sequencing of the SUR4 gene were performed. The deduced protein contains six putative transmembrane domains. Sequence comparison revealed that two yeast genes, FEN1 and JO343, shared significant similarities with SUR4. Mutants for SUR4 and FEN1 have the same pleiotropic phenotype, including bud localization defects, resistance to an immunosuppressor and resistance to ergosterol biosynthesis inhibitors. The double inactivation of SUR4 and FEN1 genes is lethal. These data and other aspects implicating SUR4 in glucose metabolism, suggest an involvement of these genes in the dynamics of cortical actin cytoskeleton in response to nutrient availability. Moreover, the existence of a fourth homologous gene in C. elegans extends the family to pluricellular organisms.

The actin cytoskeleton cells is altered in rvs161 mutant yeast, with the defect becoming more pronounced under unfavorable growth conditions, as described for the rvs167 mutant. The cytoskeletal alteration has no apparent effect on invertase secretion and polarized growth. Mutations in RVS161, just as in RVS167, lead to a random budding pattern in a/alpha diploid cells. This behavior is not observed in a/a diploid cells homozygous for the rvs161-1 or rvs167-1 mutations. In addition, sequence comparisons revealed that amphiphysin, a protein first found in synaptic vesicles of chicken and shown to be the autoantigen of Stiff Man syndrome, presents similarity with both Rvs proteins. Furthermore, limited similarities with myosin heavy chain and tropomyosin alpha chain from higher eukaryotic cells allow for the definition of a possible consensus sequence. The finding of related sequences suggests the existence of a function for these proteins that is conserved among eukaryotic organisms.

Malolactic fermentation, a crucial step in winemaking, results mostly in degradation by lactic acid bacteria of L-malic acid into L-lactic acid. This direct decarboxylation is catalysed by the malolactic enzyme. Recently we, and others, have cloned the mleS gene of Lactococcus lactis encoding malolactic enzyme. Heterologous expression of mleS in Saccharomyces cerevisiae was tested to perform simultaneously alcoholic and malolactic fermentations by yeast. mleS gene was cloned in a yeast multicopy vector under a strong promoter. Malolactic activity was present in crude extracts of recombinant yeasts. Malic acid degradation was tested during alcoholic fermentation in synthetic media and must. Yeasts expressing the mleS gene actually produced L-lactate from L-malate; nevertheless malate degradation was far from complete.

The nucleotide sequence of a 31,352 bp fragment from chromosome II of Saccharomyces cerevisiae has been determined and analysed. The fragment originates from the right arm of chromosome II, located between the GAL7,10,1 and the PHO3,5 loci, at a distance of about 130 kb from the centromere. The sequence contains a tRNA tandem repeat and 17 open reading frames (ORFs) larger than 100 amino acids. One of them extends into adjacent DNA and is incomplete. The two tRNA genes, coding for a tRNA(asp) and a tRNA(arg), and three of the ORFs, had been sequenced previously, i.e. HSP26, SEC18, and UBC4. Four other ORFs showed similarity with yeast genes; amino acid transporter genes, the RAD54, SNF2 and STH1 family, the SPS2 gene and the bromodomain of SPT7, respectively. Two showed homology with sequences from other organisms, i.e. with a Plasmodium falciparum gene encoding a surface antigen and with a gene from Saimirine herpes virus respectively. Three ORFs, YBR0726, YBR0735 and YBR0740 are completely contained in YBR0727, YBR0734 and YBR0739 respectively, and thus probably do not represent real genes. Two ORFs, YBR0727 and YBR0745 most likely contain an intron.

Malolactic enzyme is the key enzyme in the degradation of L-malic acid by lactic acid bacteria. Using degenerated primers designed from the first 20 N-terminal amino acid sequence of lactococcal malolactic enzyme, a 60-bp DNA fragment containing part of the mleS gene was amplified from Lactococcus lactis in a polymerase chain reaction. This specific probe was used to isolate two contiguous fragments covering the gene as a whole. The 1.9-kb region sequenced contains an open reading frame of 1623 bp, coding a putative protein of 540 amino acids. The deduced amino acid sequence reveals that lactococcal putative protein (Mlep) is highly homologous to the malic enzyme of other organisms. Expression of the mleS gene in Escherichia coli results in malolactic activity.

A Leuconostoc oenos plasmid, pLo13, was studied to analyze its genetic organization and to define its functions. The nucleotide sequence (3948 bp) revealed three major open reading frames. Features commonly found in plasmids that replicate via a rolling-circle mechanism were identified within pLo13: first, a sequence coding for a protein with an amino acid sequence homologous to the plasmid recombination enzymes (Pre), but for which a specific target site similar to those previously described was not found; second, a sequence probably encoding a replication protein (Rep). The putative pLo13 Rep protein amino acid sequence is divergent from the pC194-pUB110 family Rep proteins. However, the consensus sequence of the Rep protein active site was found, as well as the Rep protein consensus target site. No sequence similar to the previously described minus-origins (SSOs) are present in pLo13; nevertheless, a 200-bp sequence rich in imperfect palindromes may be involved in the minus-strand replication. These overall differences are in agreement with the previously reported important phylogenetic distance existing between Ln. oenos and other lactic acid bacteria.

The imidazole and triazole fungicides inhibit cytochrome P450 14 alpha-lanosterol demethylase (P45014DM) implicated in the ergosterol biosynthesis pathway, which is specific to fungi and yeasts. Two plasmids were obtained which allow triazole and imidazole resistance in Saccharomyces cerevisiae. The low copy number plasmid (pFD 383) encodes cytochrome P450 C14 alpha lanosterol demethylase under the control of phospho-glycerate-kinase promoter. S. cerevisiae transformed by the pFD 383 plasmid are resistant to imidazoles and triazoles. Moreover, this transformed strain shows increased levels of P45014DM mRNA and of cytochrome P450. A second low copy number plasmid (pFD 384) carries a mutant cytochrome P450 14 alpha lanosterol demethylase gene, which increases imidazole and triazole resistance. These constructions can be used on a dominant selection marker to transform wild-type yeasts and to confer imidazole and triazole resistance in industrial fermentation.

The DNA sequence of a 6794 bp fragment located at about 100 kb from the right telomere of chromosome II from Saccharomyces cerevisiae has been determined. Sequence analysis reveals five open reading frames. One is the ARO4 gene encoding the 3-deoxy-D-arabinoheptulosonate 7-phosphate synthase. Another presents strong homology with the S5 ribosomal protein from bacteria. The open reading frame YBR1705 shows significant homology with dUTPase, suggesting for the first time the existence of such an enzyme in S. cerevisiae.

We have isolated a mutant (rvs272) of the yeast (Saccharomyces cerevisiae) that displays an altered phenotype in stationary phase. It shows a high proportion of large-budded cells with two non-segregated nuclei staying in the mother cell. This phenotype is also expressed in various conditions when cells are synchronized, energy depleted or treated with the antimitotic drug benomyl. The RVS272 gene has been identified as the NUM1 gene already described. This gene presents a 192 bp tandemly repeated motif and we show that the number of repeats can vary from 1 to about 24 among different strains, without apparently affecting the function of the encoded protein. We suggest that this protein could be involved in polymerization catalysis and/or stabilization of microtubules.

We report here the sequence of a 19,482 bp DNA segment of chromosome II of Saccharomyces cerevisiae. The fragment contains 16 open reading frames (ORFs) covering 74% of the sequence. Four predicted products present homology with known proteins. The ORF YBR1732 exhibits a strong homology to serine hydroxymethyl transferase; the best score is 53.1% identity in 458 amino acids overlap with the serine hydroxymethyl transferase from rabbit liver. YBR1724, which shows homology with riboflavin synthase of Bacillus subtilis, is probably the RIB5 gene implied in riboflavine synthesis and mapped in this region. YBR1733 is homologous to rab protein and YBR1728 is presumably a GTPase activating protein.

We report here the isolation of temperature-sensitive mutants of the yeast Saccharomyces cerevisiae which exhibit cdc phenotypes. The recessive mutations defined four complementation groups, named ore1, ore2, ore3 and ore4. At the non-permissive temperature, strains bearing these mutations arrested in the G1 phase of the cell cycle. The wild-type allele of the gene altered in ore2 mutants was cloned. The nucleotide sequence of a fragment which can complement the mutation showed the presence of an open reading frame capable of encoding a protein with 286 amino acid residues. The deduced amino acid sequence showed 25% identity with that of the Escherichia coli delta 1-pyrroline-5-carboxylate reductase, an enzyme of the pathway for the biosynthesis of proline. The ore2 mutants, correspondingly, were found to be capable of growing at the non-permissive temperature on a synthetic medium supplemented with proline. In addition, the chromosomal location of the gene and its restriction map were compatible with those previously reported for the PRO3 gene which encodes the S. cerevisiae delta 1-pyrroline-5-carboxylate reductase.

The complete nucleotide sequence of the D10H fragment (10850 bp) was determined. The D10H fragment is located on the right arm of chromosome III near the centromere and contains the SUF2 gene. Six open reading frames (ORFs) larger than 300 bp were found. One of them is the CIT2 gene encoding the cytoplasmic citrate synthase. The others are new putative genes and show no significant similarity with any known gene. In addition two tRNA genes (Asn and Pro) and a solo delta element were identified. Two ORFs were disrupted; no peculiar phenotype was observed.

In yeast, nutrient starvation leads to entry into stationary phase. Mutants that do not respond properly to starvation conditions have been isolated in Saccharomyces cerevisiae. Among them the rvs161 mutant (RVS for Reduced Viability upon Starvation) is sensitive to carbon, nitrogen and sulphur starvation. When these nutrients are depleted in the medium, mutant cells show cellular viability loss with morphological changes. The mutation rvs161-1 is very pleiotropic, and besides the defects in stationary phase entry, the mutant strain presents other alterations: sensitivity to high salt concentrations, hypersensitivity to amino acid analogs, no growth on lactate or acetate medium. The addition of salts or amino acid analogs leads to the same morphological defects observed in starved cells, suggesting that the gene could be implicated mainly in the control of cellular viability. The gene RVS161 was cloned; it codes for a 30,252 daltons protein. No homology was detected with the proteins contained in the databases. Moreover, Southern analysis revealed the presence of other sequences homologous to the RVS161 gene in the yeast genome.

In this paper we report a rapid method to screen yeast mutants exhibiting reduced viability directly on plates. This method avoids the need for replica plating and is based on the addition of the vital dye erythrosine B in nutrient medium. After 2 or 3 days of culture, colonies containing a large proportion of dead cells show a pink or a dark pink color whereas normal colonies are practically white.

Some species of yeasts contain naturally-occurring circular DNA plasmids. The most studied of these plasmids is the 2 microns circle of Saccharomyces cerevisiae. Three variants of this plasmid, Scp1, Scp2 and Scp3, have been described according to their restriction maps [Cameron et al., Nucleic Acids Res. 4 (1977) 1429-1448; Livingston, Genetics 86 (1977) 73-84]. The entire nucleotide (nt) sequence of the Scp1 variant from strain A364A has been published [Hartley and Donelson, Nature 286 (1980) 860-864]. We report here the nt sequence of the 2 microns plasmid REP1 gene from S. cerevisiae strain SKQ2n. According to the restriction analysis, this plasmid is the Scp3 variant previously described. The only observed differences between the Scp1 and Scp3 variants were the loss of one EcoRI restriction site and an apparent deletion in Scp3. The nt sequence we report differs significantly from the previously published one for Scp1. The differences correspond to 128 (about 8.5%) substituted, deleted or additional nt of 1510 nt compared. These differences affect the coding region (8%) as well as the noncoding regions (9.7%). Regarding the putative encoded proteins, 38 (about 10%) amino acids (aa) are modified or deleted in our sequence and 11 are added. Most of these aa modifications are not randomly distributed but are concentrated in certain regions. These observations are indicative of important intraspecific evolution between the two 2 microns plasmid variants considered, as well as of conservative selection pressure on some domains of the REP1 protein.

The yeast nuclear gene ATP4, encoding the ATP synthase subunit 4, was disrupted by insertion into the middle of it the selective marker URA3. Transformation of the Saccharomyces cerevisiae strain D273-10B/A/U produced a mutant unable to grow on glycerol medium. The ATP4 gene is unique since subunit 4 was not present in mutant mitochondria; the hypothetical truncated subunit 4 was never detected. ATPase was rendered oligomycin-insensitive and the F1 sector of this mutant appeared loosely bound to the membrane. Analysis of mitochondrially translated hydrophobic subunits of F0 revealed that subunits 8 and 9 were present, unlike subunit 6. This indicated a structural relationship between subunits 4 and 6 during biogenesis of F0. It therefore appears that subunit 4 (also called subunit b in beef heart and Escherichia coli ATP synthases) plays at least a structural role in the assembly of the whole complex. Disruption of the ATP4 gene also had a dramatic effect on the assembly of other mitochondrial complexes. Thus, the cytochrome oxidase activity of the mutant strain was about five times lower than that of the wild type. In addition, a high percentage of spontaneous rho- mutants was detected.

A plasmid containing the gene coding for the Saccharomyces cerevisiae F0F1 ATPase subunit 4 was isolated from a yeast genomic DNA library using the oligonucleotide probe procedure. The gene and the surrounding regions were cloned into M13 tg 130 and M13 tg 131 phage vectors. A 732-base-pair open reading frame encoding a 244-amino-acid polypeptide is described. The nucleotide sequence predicts that subunit 4 is probably derived from a precursor protein with a hydrophilic and basic 35-amino-acid leader sequence. Mature subunit 4 contains 209 amino acid residues and the predicted molecular mass is 23250 Da. This subunit presents amphiphilic behaviour with two distinct domains. A high alpha-helix content of 77% was predicted from the sequence. Subunit 4 shows homology with the b subunit of Escherichia coli ATP synthase.

By using two chimeric plasmids containing yeast ura3 gene and 2-micron yeast DNA linked to the bacterial plasmid pCR1, yeast transformation of a high frequency has been achieved. The first plasmid is such that the 2-micron DNA part, in which the ura3 gene is incorporated, can be removed in one step and thus the 2-micron-ura3 sequence can be considered as a "transposable" block. In contrast, the second one bears the entire 2-micron plasmid and the ura3 gene is inserted in the bacterial plasmid part. As shown through hybridization experiments and genetic studies, the ura3 gene was maintained as a cytoplasmic element. Plasmids recovered from the yeast transformants were used to transform Escherichia coli. Their analysis by EcoRI showed that in many cases the vector had recombined with the endogenous 2-micron DNA of the recipient strain. The specific activity of orotidine 5'-monophosphate decarboxylase (coded by ura3) in yeast transformants was 10- to 30-fold higher than in the wild type.

The structure of ATP synthase subunit 4 was determined by using the oligonucleotide probe procedure. This subunit is the fourth polypeptide of the complex when classifying subunits in order of decreasing molecular mass. Its relative molecular mass is 25 kDa. The ATP4 gene was isolated and sequenced. The nucleotide sequence predicts that subunit 4 is probably derived from a precursor protein 244 amino acids long. Mature subunit 4 contains 209 amino acid residues and the predicted molecular mass is 23250 kDa. Subunit 4 shows homology with the b-subunit of Escherichia coli ATP synthase and the b-subunit of beef heart mitochondrial ATP synthase. By using homologous transformation, a mutant lacking wild subunit 4 was constructed. This mutant is devoid of oxidative phosphorylation and F1 is loosely bound to the membrane. Our data are in favor of a structural relationship between subunit 4 and the mitochondrially-translated subunit 6 during biogenesis of F0.

[URE3], a non-mitochondrial non-mendelian mutation which modifies drastically yeast nitrogen metabolism has been genetically studied. Cytoduction experiments show definitely that the inheritance of the determinant is not linked to the nucleus. The maintenance of the [URE3] determinant seems controlled by the product of a conventional nuclear gene (ure2) which is itself involved in nitrogen metabolism. The (ure2) mutation alone gives the same phenotype as [URE3] but it is impossible to obtain a stable recombinant containing simultaneously the (ure2) mutation and the [URE3] determinant. Application of the Newcombe respreading experiment demonstrates that the [URE3] mutational event occurs before the selection procedure and is therefore not strictly adaptative. Nevertheless, the nature of the selection medium changes considerably the frequency of the [URE3] mutants recovered.