Short telomeres in yeast are highly recombinogenic

We report that recombination rates specifically increase by up to 10(3) near shortened telomeres in K. lactis cells. This occurs in cells lacking telomerase that undergo growth senescence as well as in cells with stably shortened telomeres that cause little effect on cell growth. The high rates of gene conversion allowed a subtelomeric marker, initially present at a single telomere, to efficiently spread to most or all other telomeres in the cell. We propose that short telomeres in K. lactis are not fully competent at capping chromosome ends and hence are occasionally processed by proteins that normally act to repair broken DNA ends through recombination. This helps explain how recombination can be frequent enough to permit maintenance of telomeres in yeast cells lacking telomerase.

Telomeres and their control

Telomeres are DNA and protein structures that form complexes protecting the ends of chromosomes. Understanding of the mechanisms maintaining telomeres and insights into their function have advanced considerably in recent years. This review summarizes the currently known components of the telomere/telomerase functional complex, and focuses on how they act in the control of processes occurring at telomeres. These include processes acting on the telomeric DNA and on telomeric proteins. Key among them are DNA replication and elongation of one telomeric DNA strand by telomerase. In some situations, homologous recombination of telomeric and subtelomeric DNA is induced. All these processes act to replenish or restore telomeres. Conversely, degradative processes that shorten telomeric DNA, and nonhomologous end-joining of telomeric DNA, can lead to loss of telomere function and genomic instability. Hence they too must normally be tightly controlled.

Telomere fusions caused by mutating the terminal region of telomeric DNA

Mutations in the template region of a telomerase RNA gene can lead to the corresponding sequence alterations appearing in newly synthesized telomeric repeats. We analyzed a set of mutations in the template region of the telomerase RNA gene (TER1) of the budding yeast Kluyveromyces lactis that were predicted to lead to synthesis of mutant telomeric repeats with disrupted binding of the telomeric protein Rap1p. We showed previously that mutating the left side of the 12-bp consensus Rap1p binding site led to immediate and severe telomere elongation. Here, we show that, in contrast, mutating either the right side of the site or both sides together leads initially to telomere shortening. On additional passaging, certain mutants of both categories exhibit telomere-telomere fusions. Often, six new Bal-31-resistant, telomere repeat-containing bands appeared, and we infer that each of the six K. lactis chromosomes became circularized. These fusions were not stable, appearing occasionally to resolve and then reform. We demonstrate directly that a linear minichromosome introduced into one of the fusion mutant strains circularized by means of end-to-end fusions of the mutant repeat tracts. In contrast to the chromosomal circularization reported previously in Schizosaccharomyces pombe mutants defective in telomere maintenance, the K. lactis telomere fusions retained their telomeric DNA repeat sequences.

Telomeric sequence diversity within the genus Saccharomyces

Conservation of telomeric DNA repeat sequences has been found across evolutionarily diverse eukaryotes. Here we report on a marked telomeric sequence diversity within the budding yeast genus Saccharomyces. Cloning and sequencing of telomeric repeat units from S. castellii, S. dairensis, S. exiguus and S. kluyveri showed a length variation between 8 and 26 bp, as well as a distinct variation in the degree of homogeneity, among the species. In S. castellii and S. dairensis, TCTGGGTG constituted a majority of the telomeric repeat units. However, the character of the variant repeats differed: in S. castellii the major class of variant repeats contained additional TG dinucleotides per repeat unit, [TCTGGGTG(TG)1-3], whereas in S. dairensis the major variant repeat is the shorter, uniform sequence TCTGGG. This result suggests mechanistic differences in the action of the telomerases of these closely related yeasts. Despite their length and homogeneity differences, all the Saccharomyces telomeric sequences show a conserved core which is also shared by the Candida glabrata telomeric sequence. This evolutionary similarity may be partly explained by the preservation of a binding site for the RAP1 protein.

Cap-prevented recombination between terminal telomeric repeat arrays (telomere CPR) maintains telomeres in Kluyveromyces lactis lacking telomerase

Deletion of the telomerase RNA gene (TER1) in the yeast Kluyveromyces lactis results in gradual loss of telomeric repeats and progressively declining cell growth capability (growth senescence). We show that this initial growth senescence is characterized by abnormally large, defectively dividing cells and is delayed when cells initially contain elongated telomeres. However, cells that survive the initial catastrophic senescence emerge relatively frequently, and their subsequent growth without telomerase is surprisingly efficient. Survivors have lengthened telomeres, often much longer than wild type, but that are still subject to gradual shortening. Production of these postsenescence survivors is strongly dependent on the RAD52 gene. We propose that shortened, terminal telomeric repeat tracts become uncapped, promoting recombinational repair between them to regenerate lengthened telomeres in survivors. This process, which we term telomere cap-prevented recombination (CPR) may be a general alternative telomere maintenance pathway in eukaryotes.

Runaway telomere elongation caused by telomerase RNA gene mutations

The ribonucleoprotein enzyme telomerase adds telomeric DNA onto chromosome ends and is normally regulated so that telomeric DNA lengths are kept within defined bounds. In the telomerase RNA gene from the yeast Kluyveromyces lactis, specific mutations that alter telomeric DNA sequences result in telomeres elongating to up to 100 times their normal length and impair cell growth. Some mutations cause immediate elongation whereas others behave like genetic time bombs, causing elongation only after a latent period of hundreds of generations.

A conserved sequence motif within the exceptionally diverse telomeric sequences of budding yeasts

Telomeric DNA sequences have generally been found to be remarkably conserved in evolution, typically consisting of repeated, very short sequence units containing clusters of G residues. Recently however the telomeric DNA of the asexual yeast Candida albicans was shown to consist of much longer repeat units. Here we report the identification of seven additional telomeric sequences from sexual and asexual budding yeast species. The telomeric repeat units from this group of relatively closely related species show more phylogenetic diversity in length (8-25 bp), sequence, and composition than has been seen previously throughout a wide phylogenetic range of other eukaryotes. We also show that certain strains of the asexual diploid species Candida tropicalis have two forms of telomeric repeats, which appear to differ by a single base pair. Despite their great diversity, the telomeric repeat units of C. albicans, Saccharomyces cerevisiae, and all of the species we have examined in this report share a conserved approximately 6-bp motif of T and G residues resembling more typical telomeric sequences.

Unusually large telomeric repeats in the yeast Candida albicans

We have identified sequences at the telomeres of the yeast Candida albicans and have found that they are composed of tandem copies of a 23-bp sequence. Through the cloning of native telomeric ends and the characterization and cloning of a "healed" end, we demonstrate that these repeated sequences are sufficient to function as a telomere. All copies of the 23-bp repeat that have been sequenced from a number of C. albicans strains are identical. In contrast, adjacent subtelomeric sequences are variable both between strains and within the WO-1 strain. In the WO-1 strain, the lengths of the telomeres are dependent upon growth temperature and are substantially longer at higher temperatures. Telomere growth is accompanied by increases in the number of the 23-bp repeats present on the telomeric fragments. These results suggest that either telomerase-maintained telomeres can be more complex in structure than was previously imagined or that Candida telomeres are maintained via a telomerase-independent mechanism.

A G-protein alpha subunit from asexual Candida albicans functions in the mating signal transduction pathway of Saccharomyces cerevisiae and is regulated by the a1-alpha 2 repressor

We have isolated a gene, designated CAG1, from Candida albicans by using the G-protein alpha-subunit clone SCG1 of Saccharomyces cerevisiae as a probe. Amino acid sequence comparison revealed that CAG1 is more homologous to SCG1 than to any other G protein reported so far. Homology between CAG1 and SCG1 not only includes the conserved guanine nucleotide binding domains but also spans the normally variable regions which are thought to be involved in interaction with the components of the specific signal transduction pathway. Furthermore, CAG1 contains a central domain, previously found only in SCG1. cag1 null mutants of C. albicans created by gene disruption produced no readily detectable phenotype. The C. albicans CAG1 gene complemented both the growth and mating defects of S. cerevisiae scg1 null mutants when carried on either a low- or high-copy-number plasmid. In diploid C. albicans, the CAG1 transcript was readily detectable in mycelial and yeast cells of both the white and opaque forms. However, the CAG1-specific transcript in S. cerevisiae transformants containing the C. albicans CAG1 gene was observed only in haploid cells. This transcription pattern matches that of SCG1 in S. cerevisiae and is caused by a1-alpha 2 mediated repression in diploid cells. That is, CAG1 behaves as a haploid-specific gene in S. cerevisiae, subject to control by the a1-alpha 2 mating-type regulation pathway. We infer from these results that C. albicans may have a signal transduction system analogous to that controlling mating type in S. cerevisiae or possibly even a sexual pathway that has so far remained undetected.

Dosage of the smallest chromosome affects both the yeast-hyphal transition and the white-opaque transition of Candida albicans WO-1

The WO-1 strain of Candida albicans is capable of alternating between two highly distinct yeast cell types termed white and opaque (E. H. A. Rikkerrink, B. B. Magee, and P. T. Magee, J. Bacteriol. 170:895-899, 1988; B. Slutsky, M. Staebell, J. Anderson, L. Risen, M. Pfaller, and D. R. Soll, J. Bacteriol. 169:189-197, 1987). We have isolated WO-1 mutants that show a marked deficiency at being able to switch from the white form to the opaque form under conditions normally favorable for this transition. Pulsed-field electrophoresis demonstrated that one of the initial two spontaneous nonswitching mutants lacked the smallest chromosome that is normally present in WO-1. The availability of a WO-1 derivative whose only functional ADE2 gene is located on this small chromosome made possible, through the induction of chromosome nondisjunction, the isolation of numerous new mutants missing this chromosome as well as mutants containing two copies of the chromosome. Mutants missing the smallest chromosome showed a greatly diminished ability to produce opaque sectors and to produce germ tubes in the presence of human serum. Mutants containing two copies of the small chromosome showed an increased ability to produce germ tubes. These results indicate that this small chromosome carries one or more genes involved in both the white-opaque switch and the yeast-hyphal switch.

N-terminal truncated forms of the bifunctional pi initiation protein express negative activity on plasmid R6K replication

The replication initiation protein pi of the Escherichia coli plasmid R6K is a dual regulator in the control of plasmid copy number, functioning both as a specific initiator and inhibitor of replication. While the biochemical basis of these activities is not known, initiator activity requires binding of the protein to the seven 22 bp direct repeats within the gamma-origin region. By deleting C-terminal segments of the pi coding region, we have found that the N-terminal polypeptides of pi that are produced, corresponding to the first 117 and 164 amino acids, respectively, retain the negative activity of the bifunctional protein, i.e. these truncated pi proteins specifically inhibit R6K replication in vivo. These negatively acting polypeptides, however, are incapable of initiating replication in vivo and fail to bind to the gamma-origin of the R6K DNA in vitro. A correspondence between the observed negative activity of the N-terminal peptide and the negative regulatory activity of the intact pi protein is supported by the finding that point mutations introduced into the 164 amino acid N-terminal peptide that result in a decrease in its inhibitory activity also produce a plasmid high-copy phenotype when these mutations are incorporated into the full-length pi protein. These findings demonstrate that the negative domain of pi resides in the N-terminal segment of the protein. Furthermore, the data obtained suggest that inhibition of R6K replication by pi does not require direct binding to DNA.

Negative control of plasmid R6K replication: possible role of intermolecular coupling of replication origins

The gamma origin binding sites of the replication initiator pi protein, composed of seven 22-base-pair (bp) direct repeats and previously shown to be essential for replication of plasmid R6K, can also act as an inhibitor of R6K replication in Escherichia coli cells if provided in trans. Inhibition is dependent upon the ability of these repeats to bind the R6K-encoded pi protein but is not overcome by increasing the intracellular pi level. The insertion of a second repeat cluster in close proximity to the gamma origin also can markedly inhibit replication. The severity of this effect is dependent upon the position, orientation, and number of repeats present in the extra cluster. As few as six extra repeats can result in a completely nonfunctional gamma origin. However, this inactive gamma origin plasmid containing the six extra repeats is functional when placed in a strain that underproduces the wild-type pi protein or when placed in the presence of any of several copy-up mutant pi proteins. On the basis of these observations, we propose that the nucleoprotein structures formed by the binding of pi protein to the seven 22-bp direct repeats at the gamma origin are capable of coupling with each other in vivo and that replication initiation is prevented at such coupled origins. In support of this model of replication control, we demonstrate by electron microscopy analysis that the pi protein has the ability to associate two DNA molecules containing gamma origin sequences and also show that pi enhances the DNA ligase-catalyzed multimerization of a DNA fragment containing the gamma origin.

DNA and protein interactions in the regulation of plasmid replication

As for bacterial and animal viruses that employ different mechanisms for their duplication in a host cell, plasmids have evolved different strategies to assure their hereditary stability or maintenance at a specific copy number during cell growth and division. A characteristic feature of plasmid replication control, however, is an involvement of one or more negatively controlling elements. Furthermore, a majority of the bacterial plasmids examined to date contain direct nucleotide sequence repeats at their origin of replication and encode a replication protein that binds to these repeat sequences. The binding of the replication protein (pi protein) specified by the antibiotic resistance plasmid R6K to a set of 22 base pair direct nucleotide sequence repeats is required for the initiation of replication at each of three origins of replication (alpha, beta and gamma) within a 4 Kb segment of R6K. The pi initiation protein is multifunctional in that it has both positive and negative activities in both controlling the initiation of replication and autoregulating its own synthesis. Similarly, the direct repeats of plasmid R6K and several other plasmid systems play more than one role in plasmid replication. These repeats, termed iterons, are not only required for origin activity but also exert a negative effect on plasmid copy number possibly as a result of their 'titration' of a plasmid encoded replication protein. The properties of plasmid replication proteins and direct nucleotide sequence repeats that are important for their opposing positive and negative roles in the regulation of the initiation of replication are described with particular emphasis on plasmid R6K of Escherichia coli.

Positive and negative roles of an initiator protein at an origin of replication

The properties of mutants in the pir gene of plasmid R6K have suggested that the pi protein plays a dual role; it is required for replication to occur and also plays a role in the negative control of the plasmid copy number. In our present study, we have found that the pi level in cell extracts of Escherichia coli strains containing R6K derivatives is surprisingly high (approximately equal to 10(4) dimers per cell) and that this level is not altered in cells carrying high copy number pir mutants. The wild-type and a high copy mutant (Cos405) pir gene were inserted downstream of promoters of different strengths to measure the copy number of an R6K gamma replicon as a function of a 1000-fold range of intracellular pi concentrations. The data demonstrate that reducing the intracellular level of pi to 5% of its normal value can result in a substantial increase in copy number of a gamma origin replicon and that a pi level less than 1% of normal is still permissive for replication. Conversely, increasing the pi level even a few-fold above normal results in a marked inhibition of replication of plasmids containing a single, two, or all three of the R6K origins (alpha, beta, and gamma). We have also shown that the replication inhibition mediated by excess pi is greatly reduced by the pir405 Cos mutation. These results demonstrate that the total level of pi protein is not rate-limiting for a gamma replicon. We have also determined the sensitivity of the pir gene promoter to a wide range of pi concentrations. The activity of this promoter is stimulated by very low pi levels and is almost entirely inhibited when the protein is overproduced 2-fold.

Mutations in direct repeat sequences and in a conserved sequence adjacent to the repeats result in a defective replication origin in plasmid R6K

Plasmid pMM3 is a pBR322 derivative carrying the gamma origin of replication of the naturally occurring plasmid R6K. We have produced a gamma-origin mutant bank of this plasmid using the single-strand-specific mutagen sodium bisulfite. Members of this bank contain single or multiple mutations in the seven direct repeats and the flanking sequences in the gamma origin. Three mutants with defective gamma origins have been isolated from this mutant bank. Two of these direct repeat mutants, gamma 117 and gamma 120, are unable to replicate and also have lost the ability to bind the R6K initiation protein pi in vitro at one of the seven 22-base-pair direct repeats within their respective origins. Precise deletion of the damaged repeat of either of these mutants restores origin function, suggesting that the primary defect of these mutants involves a disruption of the normal spacing of pi binding and flanking sequences within the gamma origin. The third mutant, gamma 111, binds pi normally but replicates at a greatly reduced copy number due to a mutation near the seventh repeat. This mutation falls within a short sequence that appears to be conserved among a number of other plasmids that contain direct repeats within their origins of replication.