Modulation of telomere terminal structure by telomerase components in
http://www.100md.com
《核酸研究医学期刊》
Department of Microbiology and Immunology, W. R. Hearst Microbiology Research Center, Weill Medical College of Cornell University 1300 York Avenue, New York, NY 10021, USA
*To whom correspondence should be addressed. Tel: +1 212 746 6506; Fax: +1 212 746 8587; Email: nflue@med.cornell.edu
ABSTRACT
The telomerase ribonucleoprotein in Candida albicans is presumed to contain at least three Est proteins: CaEst1p, CaEst2p/TERT and CaEst3p. We constructed mutants missing each of the protein subunit of telomerase and analyzed overall telomere dynamics and single-stranded telomere overhangs over the course of many generations. The est1- mutant manifested abrupt telomere loss and recovery, consistent with heightened recombination. Both the est2- and est3- mutant exhibited progressive telomere loss, followed by the gradual emergence of survivors with long telomeres. In no case was telomere loss accompanied by severe growth defects, suggesting that cells with short telomeres can continue to proliferate. Furthermore, the amount of G-strand terminal overhangs was greatly increased in the est2- mutant, but not others. Our results suggest that in addition to their well-characterized function in telomere elongation, both CaEst1p and CaEst2p mediate some aspects of telomere protection in Candida, with the former suppressing excessive recombination, and the latter preventing excessive C-strand degradation.
INTRODUCTION
Telomeres are nucleoprotein structures located at the ends of chromosomes that are crucial for maintaining chromosome stability (1–3). In the majority of organisms, telomeric DNA consists of short repetitive G-rich sequences that end with a single-stranded 3' overhang. In mammalian cells, the overhang can invade a more proximal region of telomeres to form a specialized D-loop, known as T-loop (4,5). Functionally important proteins are recruited to telomeres through both DNA–protein and protein–protein interactions. Both the DNA and protein components of telomeres are essential to the maintenance of chromosome stability.
Loss of telomere DNA can occur with each round of replication due to the so-called ‘end replication problem’ (6,7). Two compensatory mechanisms that counteract telomere loss have been identified. The first entails the addition of nucleotides to the G-rich strand of telomeres by a specialized reverse transcriptase named telomerase (8–11). Telomerase is a ribonucleoprotein complex whose enzymatic core consists of a catalytic protein (named TERT in general and Est2p in yeast) and a template RNA (named TLC1 in yeast) (9). In the budding yeast Saccharomyces cerevisiae the telomerase complex contains at least two additional protein subunits, named Est1p and Est3p (12–14). Both subunits are required for telomere extension in vivo, but not essential for telomerase activity in vitro (9). Est1p has been subjected to detailed analysis and shown to mediate the recruitment of the telomerase complex to telomere ends by interacting with a G-strand telomere-binding protein named Cdc13p (15,16). It also appears to perform a post-recruitment or activating function (14,16). A recent study further suggests a role for the Candida Est1p in defining the substrate preference of telomerase in vitro (17). The second mechanism of telomere elongation entails recombination, and is primarily observed in telomerase-deficient cells (18–20). For example, in several budding yeasts, loss of telomerase is initially accompanied by progressive telomere shortening and senescence. However, rare populations of cells (called ‘survivors’) are able to activate a recombination-based mechanism(s) for telomere elongation and regain normal growth.
Equally important in telomere maintenance is protection of telomeric DNA from excessive nucleolytic degradation. Both single- and double-stranded telomere-binding proteins have been implicated in telomere protection. A well-studied example of single-stranded telomere-binding protein is Cdc13p in S.cerevisiae. Evidence for the protective function of Cdc13p came from analysis of a cdc13ts mutant, which suffers from extensive nucleolytic degradation of the C-rich strand of telomeres (21). C-strand degradation in this mutant is EXO1-dependent, implying a role for Cdc13p in antagonizing the activity of EXO1 (22). A family of proteins that are similar in function to Cdc13p have been identified in diverse organisms, and are known as Pot1 (23,24). In fission yeast, loss of Pot1 function leads to telomere degradation and telomere–telomere fusion (24). In humans, RNAi-mediated knock down of Pot1 provokes changes in telomere structure and induces chromosome instability (23). Among double-stranded telomere-binding proteins, the heterodimeric Ku complex, in particular, has a well-characterized role in telomere protection. Initially identified as a key mediator of non-homologous end joining, the Ku complex was shown recently to perform multiple functions at telomeres (25,26). Like the cdc13ts mutant, yku knockout strains exhibit elevated levels of G-strand overhangs at telomeres (27,28), which can also be attributed to aberrant degradation of the C-strand (22,27,28). Interestingly, several recent studies suggest that the telomerase complex, aside from its well-known function in telomere elongation, may have an additional role in telomere protection. For example, in the absence of telomerase, telomeres in the budding yeast are not only lost gradually, but also exhibit increased fusion to double-strand breaks (29). The fusion rate was even higher if the telomerase mutation was combined with the loss of Tel1p, a telomere-binding checkpoint protein. As another example, expression of telomerase mutants in human fibroblasts with shortened telomeres can prolong their life span without causing bulk telomere elongation (30).
In this study, we used Candida albicans as a model system for investigating telomerase function and regulation. C.albicans is an opportunistic fungal pathogen that can cause systemic infection in immuno-compromised individuals. It is an attractive model system because of its possession of longer telomere tracts (2–5 Kb in the BWP17 wild-type strain), a 23 bp regular telomere repeat (31), and the availability of genome sequence. In our previous work we showed that loss of each of the individual protein component of the telomerase complex in C.albicans caused specific defects in telomere maintenance (32). However, we failed to observe either cell senescence, as evidenced by growth retardation, or emergence of survivors, as evidenced by improved growth and sudden telomere elongation after severe shortening. To confirm and extend these findings, we reconstructed the mutants using a different set of disruption cassettes and assessed telomere length and G- and C-strand overhangs over many more generations. Here we report that with sufficient number of passages, telomere shortening followed by the apparent emergence of survivors can be observed in the telomerase mutants, albeit without frank senescence. The est1- mutant manifested abrupt telomere loss and recovery, consistent with heightened recombination. In addition, loss of Est2p was accompanied by a substantial increase in the amount of G-strand overhangs, consistent with a role for this protein in preventing aberrant degradation of the recessed C-strand. Our results suggest that in addition to their well-characterized functions in telomere elongation, both CaEst1p and CaEst2p mediate some aspects of telomere protection in Candida.
MATERIALS AND METHODS
Plasmids and strains
C.albicans BWP17 (ura3-::-imm434/ura3-::-imm434 his1::hisG/his1::hisG arg4::hisG/arg4::hisG, a gift from A. Mitchell (Columbia University)), was derived from CAI4 as described previously (33). The double knockout mutants est1-, est2- and est3- were created using the ‘URA-blaster’ method (34). For the disruption of EST1, a 426 bp fragment of the noncoding sequence upstream of the EST1 open reading frame (ORF) was inserted 5' to the URA-blaster cassette, and a 920 bp fragment downstream of the ORF was inserted 3' to the cassette. For the disruption of EST2, a 768 bp fragment of the noncoding sequence upstream of the EST2 ORF was inserted 5' to the URA-blaster cassette, and an 860 bp fragment downstream of the ORF was inserted 3' to the cassette. For the disruption of EST3, a 560 bp fragment of the noncoding sequence upstream of the EST3 ORF was inserted 5' to the URA-blaster cassette, and a 980 bp fragment downstream of the ORF was inserted 3' to the cassette.
C.albicans transformations were carried out as described previously (34). Transformants were first selected on SD-Ura plates and tested for the correct integration of the URA-blaster cassette by Southern blots. EST/est clones that lost uracil prototrophy were selected for their ability to grow on 5-FOA plates and used for the second round of transformation with the same cassette. The loss of both alleles of EST genes was verified by Southern blots. All strains were grown in the YPD medium, and supplemented with 80 μg/ml uridine when necessary.
Detection of double-stranded telomeres
For analysis of telomeres over multiple generations, cells were streaked on YPD plates, grown for 2 days at 30°C, and one colony was picked for further re-streaking and for inoculation into 5 ml YPD. The cultures were grown at 30°C and the cells collected after 16–20 h of growth. Chromosomal DNA was isolated by the ‘Smash and Grab’ method (35). The DNA (5 μg) was digested with a combination of AluI and NlaIII. Restriction fragments were fractionated on a 0.9% agarose gel and transferred to a nylon membrane. Telomeric DNA was detected by hybridization at 65°C to a 5' end-labeled 46mer oligonucleotide containing two copies of the C.albicans telomere repeat. Subsequently, the nylon membrane was stripped of the telomeric probe and hybridized to a 32P-labeled fragment of the RAD52 gene. To compare the total telomeric DNA content in different samples, the signals from the telomeric and RAD52 blots were quantified using a PhosphorImager system (Molecular Dynamics), and the telomeric DNA signal was normalized against the RAD52 signal.
Plating efficiency
Colonies of the est2- mutant at different points during the propagation were inoculated in YPD supplemented with uridine (80 μg/ml) and grown for 16 h. Cell cultures were diluted with water and the cell densities determined using a Nuebauer hemocytometer. An aliquot containing 300 cells was plated on a YPD plate supplemented with uridine (80 μg/ml). Plates were incubated for 2 days at 30°C and the number of colonies counted. Plating efficiency was calculated as the ratio of the number of colonies on the plate after 2 days of incubation to the colony forming units calculated using the hemocytometer (expressed as percentages), and the results were plotted. All experiments were performed in duplicates. Note that because clumped cells were not counted, the actual number of cells used for plating was probably higher than that determined by the hemocytometer, thus resulting in plating efficiency that was higher than 100%.
In-gel hybridization assay
In-gel hybridization assays were performed as described previously with some modifications (36). Briefly, genomic DNA isolated from the wild-type and mutant Candida strains was digested with AluI and NlaIII, and the resulting fragments were resolved on a 0.9% agarose gel. The gel was dried on a vacuum drier for 1 h at room temperature, and subjected to pre-hybridized in the Church mix (supplemented with 100 μg/ml of boiled calf thymus DNA) for 1 h at 55°C. Hybridization was done for 3 h at 55°C using 32P-labeled 23mer oligonucleotides complementary to either the G- or C-strand of telomeres. Subsequently, the gel was washed three times for 20 min in 4x SSC at room temperature and three times for 20 min in 4x SSC, 0.1% SDS at 55°C and exposed to a PhosphorImager screen. After an image was acquired, the gel was denatured by immersing in 0.6 M NaCl, 0.2 M NaOH for 1 h, neutralized in 1.5 M NaCl, 0.5 M Tris, pH 7.4, for 1 h, and then washed with water for 30 min. The denatured DNA in the gel was re-probed using the same oligonucleotide. The relative levels of single-stranded DNA were calculated as follows. (i) Hybridization signals for single-stranded and double-stranded telomeric DNA in each sample were quantified using a PhosphorImager system (Molecular Dynamics Inc.) and the background subtracted. (ii) The ratios for the single-stranded versus double-stranded signals were obtained for each sample. (iii) The ratios for the mutant samples were normalized against that for a wild-type sample.
Telomere oligonucleotide ligation (T-OLA) assay
The T-OLA assay was performed as described previously except for a few minor modifications (37). Briefly, 10 pmol of a 23mer oligonucleotide complementary to the G-strand or to the C-strand of telomeres (ACACCAAGAAGTTAGACATCCGT and ACGGATGTCTAACTTCTTGGTG, respectively) were 5' end-labeled using 20 U of OptiKinase (Amersham) and 10 pmol of ATP (6000 Ci/mmol, 10 mCi/ml) (Perkin Elmers) in 50 μl reaction containing 1x OptiKinase buffer (Amersham) for 30 min at 37°C, at which point 1 μl of 0.1 M cold ATP and 10 U of OptiKinase were added, followed by 15 min incubation at 37°C. Reaction was terminated by addition of 50 μl TE and phenol/chloroform extraction. Labeled oligonucleotides were purified using Nick columns (Amersham), and concentrated by ethanol precipitation and resuspension in 40 μl water. Chromosomal DNA (5 μg) was mixed with 0.5 pmol of labeled oligonucleotides in 1x Taq DNA ligase buffer (NEB) in 20 μl volume, and incubated at 55°C for 14–16 h. At the end of incubation, 40 U of Taq DNA ligase (NEB) were added and ligation was allowed to proceed for 5 h at 55°C. Ligation was terminated by addition of 80 μl water and phenol/chloroform extraction. DNA was precipitated by the addition of ethanol and dissolved in 55 μl water. As a loading control, one-eleventh of the sample was removed for DNA quantification by electrophoresis and Ethidium Bromide staining. The remainder of the sample was dried in a vacuum dryer, dissolved in 1x sample buffer and analyzed on a 6% denaturing polyacrylamide gel. Radioactive signals were quantified using a PhosphorImager system (Molecular Dynamics), normalized against the loading control, and the relative levels of single-stranded DNA calculated as described before (38).
Slot blot
The slot/dot blot procedure was basically performed as described previously (21). Briefly, 4 μg of AluI and NlaIII-digested genomic DNA was used as the starting material. The DNA was diluted in water to a volume of 660 μl, from which a 60 μl aliquot was transferred to a fresh tube and subjected to alkali-heat denaturation (addition of NaOH to 0.2 M, followed by 30 min incubation at 65°C). Both native and denatured DNA samples were further diluted to 1200 μl in 5x SCP buffer (0.5 M NaCl, 0.15 M Na2HPO4 and 0.01 mM EDTA). Aliquots of DNA (300 μl per slot, four slots per sample) were applied to a Hybond-N membrane (Amersham Biosciences) by using a Microfiltration Apparatus (Bio-Rad) and immediately washed three times with equal volume of 10x SCP each. The moist membrane was exposed to UV light (Stratagene) to effect crosslinking and then divided into two equal halves. Each half of the membrane contained two slots of native and two slots of denatured DNA from each sample. Membranes were probed separately with 32P-labeled 46mer oligonucleotides complementary to either the G- or C-strand of telomeres. Relative levels of single-stranded DNA were calculated as follows. (i) Hybridization signals for single-stranded DNA and double-stranded DNA in each sample were quantified and the background subtracted. (ii) The ratios for the single-stranded versus double-stranded signals were obtained for each sample. (iii) The ratios for the mutant samples were normalized against that for a wild-type sample.
Nuclease treatment
Chromosomal DNA was isolated by the ‘Smash and Grab’ method (35) and was further purified using a Nick column (Amersham). To remove the 3' single-stranded telomere overhang, ExoI (USB) was used in a reaction containing 1 U of enzyme per 50 ng of DNA in 67 mM glycine (pH 9.5), 10 mM 2-mercaptoethanol, and 6.7 mM MgCl2, and the reaction was allowed to proceed for 24 h at 37°C. Alternatively, ExoI (Epicentre) was used in a reaction that contained 50 mM Tris–HCl (pH 7.9), 10 mM MgCl2, 100 mM NaCl and 1 mM DTT. To remove both protruding 3' and 5' single-stranded telomere overhangs, Exonuclease VII (ExoVII, USB) was used in a reaction containing 2 U of enzyme per μg of DNA in 50 mM Tris–HCl (pH 7.9), 50 mM potassium phosphate (pH 7.6), 8.3 mM EDTA, 10 mM 2-mercaptoethanol and the reaction was performed at 37°C for 30 min. Reactions were terminated by phenol/chloroform extraction and processed for subsequent assays.
RESULTS
Construction of the C.albicans est mutants
Previously, we reported results of a study in which the telomere dynamics of C.albicans mutants was followed through 250–300 generations (32). Though some defects in telomere maintenance were observed, we failed to detect senescence or the emergence of survivors, as judged by changes in the growth rate of cultures. To explore the possibility that more extended passaging of the mutants may reveal additional phenotypes, we reconstructed the mutants, propagated them for 800 generations, and analyzed telomere lengths and structure during the passages. For the current study we deleted the EST genes using the ‘URA-blaster’ method. In comparison with the ‘One-Step Transformation’ method used in the earlier study, the ‘URA-blaster’ technique has the advantage of precluding the homozygosis of genes located distal to the disrupted gene (32,39). Transformation cassettes in the current study were created by inserting short DNA fragments that flank the ORF of the target gene upstream and downstream of the ‘URA-blaster’ cassette (Materials and Methods). Double deletion mutants were derived from two rounds of transformation and selection and confirmed by Southern blots (data not shown). The mutants were propagated by re-streaking every 2 days for isolated colonies. It is estimated that each restreak corresponds to 25 generations of growth.
Growth and telomere dynamics of the est mutants during extended passage
We next investigated the overall telomere dynamics of the wild type and mutants during extended passage. For the wild-type parental strain used in this study (BWP17), the majority of NlaIII and AluI digested telomere restriction fragments are between 2 and 5 kb in size. Consistent with the earlier report, we found that the size of telomere fragments and their distribution in the wild-type strain differed from one streak to the next (Figure 1A). However, the variations between the total intensity of the telomeric signals over time are no more than 20%. We have passaged the BWP17 for a total of 25 streaks without observing noticeable changes in telomeric content (data not shown). These data are in agreement with our earlier observation that some changes in telomere length occur normally during the propagation of C.albicans cells, but a homeostatic mechanism must be acting to maintain the telomeric content of the cell within a range.
Figure 1 Telomere dynamics in the wild-type and est mutant strains. Telomeres dynamics was analyzed in successive generations of the wild-type and the est mutant strains by Southern blots. The blots were stripped and re-probed using a labeled RAD52 fragment and the results shown at the bottom. The strains and passages analyzed are (A) BWP17, streaks 2–12; (B) est1-, streaks 1–33; (C) est2-, streaks 1–35; (D) est3-, streaks 1–35. Instances of sudden appearance and disappearance of telomere restriction fragments are marked by arrows.
For analysis of mutants, we examined the telomere dynamics of two or three clones (three for est2- and two each for est1- and est3-) for each telomerase mutant, which yielded similar results. The newly derived est1- strains were passaged for >30 restreaks after the selection of the mutant (Figure 1B). Significant telomere loss was observed during passage, with the total telomere content showing an almost 50% decrease at streak 11 (data not shown). However, large telomeric fragments were still present at streak 17. Presence of long telomeres in combination with overall telomere loss in the est1- strain suggests that telomeres in the mutant were more heterogeneous in size than the wild type. Long telomeres became absent for several passages (streak 19–25) but reappeared thereafter. This was accompanied by restoration of telomere content to wild-type levels. Such an abrupt increase of telomere length en masse is reminiscent of type II survivors that arose after senescence in telomerase-negative S.cerevisiae strains. We have thus adopted the name ‘survivors’ to describe the Candida cells, notwithstanding some phenotypic differences between the Saccharomyces and Candida est1 mutant. Most notably, in the C.albicans est1- clone there was no decrease in the growth rate during passage (data not shown), and the telomere content of the mutant never fell below 20% of the wild type. Even in the population with the shortest average telomere length, it was possible to detect telomeric restriction fragments that were at least 2 kb long. After the emergence of ‘survivors’, telomere content of the est1- clone became comparable with that of the wild type and was maintained during the remaining passages. It is unclear if additional cycles of telomere loss and restoration can occur in these cells with even more extended subculturing.
In contrast to Est1p, loss of Candida Est2p resulted in progressive telomere attrition (Figure 1C). Notably, while occasional lengthening and shortening events were observed in the est2- mutant, they were not as frequent as in the case of est1- mutant (compare Figure 1B and C, see also Supplementary Figure 1), nor did telomere length distribution in any est2- clone became as heterogeneous as est1-. All telomeres became shorter than 1 kb at about the 19th streak, but subsequently underwent significant elongation, which was followed by a second round of contraction. This dynamics, which is similar to that of S.cerevisiae telomerase mutants, was not observed in the previous study of another Candida est2- mutant, most likely due to the shorter duration of the earlier experiment (32). It is also possible that the earlier mutant may have suffered some additional genetic changes due to the ‘UAU1’ knockout protocol. The greater reliability of the URA-blaster transformation protocol and the more extended passaging of the mutant suggest that the current results are more reflective of the true consequences of deleting EST2 in Candida. Like the Candida est1- mutant, the est2- ‘survivors’ exhibited features that were distinct from their S.cerevisiae counterparts. First, loss of Est2p functions in C.albicans did not result in severe growth defects even after prolonged propagation. Mild growth retardation was detected in the est2- strain in liquid media (Figure 2A). However, this was true for cultures of early and late passages, and does not appear to be due primarily to telomere attrition. From the exponential phase of the growth curves, it was estimated that the doubling time of the mutant was increased by 20%. Growth of est2- was further investigated by measurement of plating efficiency. As shown in Figure 2B, the plating efficiency of the mutant was slightly reduced (by 20%) when monitored at three different points during passage. Thus, the growth retardation of est2- may be due to loss of viability of a small fraction of the cells. Another distinct feature of C.albicans est2- mutant relates to the kinetics of telomere elongation in the cell population. At streak 21, there was an abrupt broadening in the length distribution of telomeres, indicating that the population of cells contained both very long and short telomeres. Whether the long and short telomeres co-exist in a single cell is unclear. Over the next few streaks the relative intensity of the long telomeres increased steadily and became dominant. These observations suggest that cells with shortened telomeres in the est2- mutant culture were able to proliferate and co-exist with apparent ‘survivors’. Lastly, the telomere lengths of Candida est2- survivors are comparable with those of wild-type cells, whereas the Saccharomyces type II survivors have telomeres that are dramatically longer than normal. Thus, as in the case of est1 mutants, there are apparent similarities as well as significant differences between the Candida and S.cerevisiae est2 survivors.
Figure 2 Growth of the est2- mutant. (A) The doubling time for the wild-type and est2- strain during exponential growth was determined at three different points during passage and plotted. (B) The plating efficiency of the wild-type and est2- strain during passage was determined as described in Materials and Methods and plotted.
Telomeres in the new est3- mutants experienced progressive loss (Figure 1D), just as described for the previous est3- mutant (32). Similar to est1- and in contrast to est2-, no growth retardation was evident during passage of the est3- mutant (data not shown). At streak 21, after severe telomere attrition, there was an abrupt appearance of long telomeres, which became pre-dominant during the later streaks. Interestingly, unlike the est2- strain, bulk telomere content after recovery was preserved in 1 est3- survivor for 10 streaks (Figure 1D), suggesting potential variability in the dynamics of the survivors. Analysis of more clones will be necessary to confirm the prevalence of this variability. We conclude that telomere attrition followed by the apparent emergence of survivors can be observed in Candida following prolonged passage of telomerase-deficient clones.
Detection of G- and C-strand overhangs at Candida telomeres using in-gel hybridization
It has been suggested that telomerase may have a protective function that is independent of its telomere extension function. Loss of telomere protection has been linked to changes in the level of terminal overhangs. We therefore investigated the impact of Candida telomerase mutations on the amount of telomere overhangs. Several methods have been developed previously for the purpose of quantifying single-stranded telomere overhangs (21,37,39). For analysis of Candida telomeres, we adopted in-gel hybridization as our primary assay because this technique has been used widely by many laboratories to analyze telomeres in different organisms. As described later, we also confirmed the key conclusions of the in-gel analysis using two additional assays.
We first investigated samples of chromosomal DNA from BWP17 using in-gel hybridization, and found that low levels of G- and C-strand signals could be detected at telomeres (Figures 3 and 4 and data not shown). The G-strand signal was abolished by pre-treatment of DNA with ExoI, a single-stranded 3'5' exonuclease (Figure 3A, lanes 3 and 4). In contrast, the bulk telomere signals obtained after denaturation of the DNA were unaffected by this treatment (Figure 3A, lanes 5 and 6). Both the G-strand and C-strand signals for native DNAs were also sensitive to ExoVII, a single-strand specific 3'5' and 5'3' exonuclease (Figure 3B and data not shown). These results indicate that most of the native in-gel signals were due to terminal overhangs. We note that the native in-gel signals were generally less discreet in size than the signals obtained after denaturation and re-probing. One potential explanation is that DNA fragments with significant overhangs may represent a small fraction of the telomere restriction fragments, and may behave more heterogeneously due to variable amount of single-stranded overhangs. The native and denatured signals, however, do exhibit similar size distributions in a given sample, consistent with their similar origins (Figures 3 and 4).
Figure 3 The effect of nuclease treatment on the amount of unpaired telomeric overhangs. (A) DNA isolated from BWP17 was incubated with or without E.coli ExoI, and then treated with AluI and NlaIII. The resulting samples were first analyzed by gel electrophoresis and ethidium bromide staining (left panel). The gel was subsequently dried and subjected to in-gel hybridization using a probe complementary to the G-strand of Candida telomere repeat (middle panel). The DNA in the gel was then denatured and hybridized again to the same probe (right panel). (B) DNA isolated from BWP17 was incubated with or without E.coli ExoVII, and then treated with AluI and NlaIII. The resulting samples were first analyzed by gel electrophoresis and ethidium bromide staining (left panel). The gel was subsequently dried and subjected to in-gel hybridization using a probe complementary to the C-strand of Candida telomere repeat (middle panel). The DNA in the gel was then denatured and hybridized again to the same probe (right panel).
Figure 4 Analysis of G-strand overhangs in the wild-type and est2- mutant clones. (A) (Upper panel) Genomic DNA was isolated from overnight cultures of the wild-type strain and the indicated streaks of the est2- mutant, and subjected to in-gel hybridization analysis using an oligonucleotide probe complementary to the G-strand of Candida telomeres. (Middle panel) After probing the native DNA with strand-specific probes, the DNA in the gel was denatured and hybridized again with the same probe. (Lower panel) The hybridization signals for single- and double-stranded telomeres were quantified, and the ratios plotted. (B) Sameas (A) except that the probe was designed to anneal to the C-strand of Candida telomeres. (C) The intensity traces for the lanes labeled ‘*’ and ‘**’ in Figure 4A were obtained using the Imagequant program (Molecular Dynamics Inc.). The native and denatured profiles for each sample are plotted alongside each other.
G- and C-strand overhangs in est mutants
Chromosomal DNAs derived from Candida est mutants were first subjected to in-gel hybridization analysis using a probe complementary to the G-strand. For each mutant, we investigated two independent clones over 10 successive streaks. Remarkably, loss of Est2p had a strong impact on the level of unpaired G-strand: there was an up to 20-fold increase in the amount of G-strand overhangs during passage, when the G-strand signals were normalized against total telomeric DNA (Figure 4A). The increase was observed in both mutant clones and in multiple samples, and the signals were sensitive to ExoI and ExoVII treatment (Supplementary Figure 2). These results suggest that loss of Candida Est2p may result in aberrant degradation of the 5' end-containing C-strand, thus leading to the accumulation of unpaired G-strand. Comparison of the signals in native and denatured gel images indicates that within a given sample, shorter DNA fragments possess higher levels of G-strand overhangs than longer fragments. For example, the ‘*’ sample yielded two major peaks in both the native and denatured analysis (3.6 and 2.3 kb,Figure 4A and C). The signal for the lower peak was stronger in the native analysis, but weaker in the denatured analysis (Figure 4C, left). Likewise, the lower peak in the ‘**’ sample was comparable in intensity to the upper peak in the native analysis, but much weaker in the denatured analysis (Figure 4C, right). Quantitative analysis of other samples yielded similar results (data not shown). Thus, shorter telomeres appear to be more susceptible to C-strand degradation in the est2- mutant. In contrast to est2-, the amount of G-strand overhang in the est1- and est3- mutant were quite low and indistinguishable from that found in the parental strain (Figure 5A and B).
Figure 5 Analysis of G-strand overhangs in the est1- and est3- mutant clones. (A and B) (Upper panel) Genomic DNA was isolated from overnight cultures of the wild-type strain and the indicated streaks of the est1- or est3- mutant, and subjected to in-gel hybridization analysis using an oligonucleotide probe complementary to the G-strand of Candida telomeres. (Middle panel) After annealing the native DNA with the G-strand-specific probe, the DNA in the gel was denatured and hybridized again with the same probe. (Lower panel) The ratios of G-strand signals to double-stranded telomere signals were calculated for both the wild-type and mutant samples and the averages and standard deviations plotted.
Unpaired C-strand in the est mutants was also investigated by in-gel hybridization using a probe that corresponds to the G-strand repeat. Again, loss of Est2p had the most impact on the amount of unpaired C-strand (Figure 4C). There was a moderate decrease (40–70%) in the amount of C-strand overhang in the two clones of est2- mutant during the course of 10 streaks. In contrast to est2-, the amount of C-strand overhangs in the est1- and est3- mutant was similar to that found in the parental strain (data not shown). Thus, loss of individual subunits of the Candida telomerase complex has different impacts on the telomere overhang structure.
Analysis of telomere overhangs by the T-OLA and slot blot assay
Because changes in telomere overhangs have not been observed in pre-senescent telomerase mutants before, we sought to verify our in-gel hybridization results for est2- using two other assays, namely the slot blot and T-OLA assay (21,36,37). These assays detect single-stranded telomere overhangs by hybridizing native DNAs on nylon membranes and in solution, respectively, to strand-specific probes. In the case of T-OLA, multiple copies of the probe that annealed to the single-stranded overhangs are further ligated and subsequently analyzed by gel electrophoresis. Chromosomal DNAs derived from wild-type and est2- mutant were subjected to both assays. As shown in Figures 6 and 7, the est2- mutant consistently displayed higher levels of G-strand signals in comparison with the wild-type samples in both T-OLA and slot blot assays. The average increases were 7- and 10-fold in the T-OLA and slot blot assays, respectively (Figures 6B and 7). In contrast, the C-strand signals yielded by the mutant samples were somewhat lower than the wild type (Figure 7 and data not shown). These findings are entirely consistent with the results of the in-gel hybridization analysis. Thus, alterations in the amount of telomere overhangs in the est2- mutant can be detected by three different assays.
Figure 6 T-OLA analysis of unpaired G-strand overhangs in the wild-type and est2- strains. (A) G-strand overhang dynamics was analyzed in successive generations of the wild type and the est2- mutant strain by T-OLA. Selected T-OLA products are designated by the number of repeats that they contain. The strains and passages analyzed in the different parts of the figure are BWP17, streaks 4–10; est2-, streaks 1–9. (B) The relative levels of G-strand overhangs were quantified for the wild-type (streaks 4, 6, 8 and 10) and est2- samples (streaks 1–9) as described in Materials and Methods, and the averages and deviations plotted.
Figure 7 Slot blot analysis of unpaired G- and C-strand overhangs in est2- mutant. Genomic DNA was isolated from overnight cultures of the wild-type strain and the indicated streaks of the est2- mutant, and subjected to slot blot analysis. Note that 10 times more native DNA than denatured DNA was applied to the slots in order to generate appreciable signals in these analyses (Materials and Methods). In all cases analyzed, the denatured DNA yielded a signal that is at least 30 times stronger than the native DNA when corrected for this difference in application. Relative ratios of single-stranded to double-stranded telomere DNA for wild-type (duplicate samples for streak 2) and est2- (streaks 1–7) in the slot blots were calculated (Materials and Methods) and plotted at the bottom.
Accumulation of G-strand overhangs in the est2- clones was due to the loss of Est2p
Even though we observed accumulation of unpaired G-strand in several independent est2- clones, it remains formally possible that this phenotype was due to unrelated genetic changes that occurred during the multiple steps required for mutant strain derivation. To definitively rule out this possibility, we re-integrated one wild-type copy of the EST2 gene at a disrupted locus, and analyzed the phenotypes of the resulting strain. Telomere length measurements through successive streaks indicate that the strain is capable of maintaining telomere lengths over many generations (data not shown). Importantly, the level of G-strand overhangs in the re-integrant is comparable with the wild-type strain and much lower than the est2- mutant from which the re-integrant was derived (Figure 8A and C). As expected, the overall telomeric content of the re-integrant was higher than the est2- mutant (Figure 8B). We conclude that the accumulation of G-strand overhangs in the est2- mutant must be due to the loss of Est2p rather than some other unintended genetic changes.
Figure 8 Analysis of G-strand overhangs in wild type, est2- and est2- with a re-integrated copy of EST2. (A) Genomic DNA was isolated from overnight cultures of a wild-type strain, an est2- mutant, and an est2- mutant with a re-integrated copy of EST2, and subjected to in-gel hybridization analysis using an oligonucleotide probe complementary to the G-strand of Candida telomeres. Lanes 1–3 and 4–6 represent duplicate assays of the same samples. (B) After probing the native DNA with strand-specific probes, the DNA in the gel was denatured and hybridized again with the same probe. (C) Relative ratios of single-stranded to double-stranded telomere signals for all three samples were calculated and plotted.
DISCUSSION
In this study, we investigated the total amount and length of telomeric DNA in Candida mutants that are missing individual components of the telomerase complex during extended passage. In addition, we analyzed the levels of single-stranded telomere overhangs in the same strains. Significant findings include (i) apparent observation of telomerase-deficient survivors, (ii) increased levels of telomere recombination in the est1- mutant, (iii) accumulation of G-strand overhangs in the est2- mutant, and (iv) detection of C-strand overhangs. We also compared three different methods for assessing telomere overhangs and found that they yielded qualitatively similar results.
Telomerase-deficient C.albicans survivors
In our earlier analysis of Candida telomerase mutants, we did not observe the emergence of survivors (32), most likely due to insufficient number of passages. Here we show that with extended subculturing, phenotypes that are reminiscent of S.cerevisiae survivors can indeed be observed in Candida mutants. Specifically, both the est2- and est3- strains suffered progressive telomere attrition, followed by the emergence of survivors with greatly elongated telomeres. Some interesting differences, however, can be noted between the S.cerevisiae and Candida survivors. In S.cerevisiae, survivors do not arise until the culture has suffered very significant growth defects. The appearance of survivors is correlated with the detection of longer than wild-type telomeres in the vast majority of cells and is accompanied by an increase in the growth rate of the culture (19,40–42). In C.albicans, the emergence of survivors does not coincide with a large change in the growth rate of the culture. Thus, either telomere attrition has little effect on C.albicans growth, or the activation of the recombination pathway occurs before the degree of telomere attrition that is necessary to induce large growth defects. Also in contrast to S.cerevisiae, the Candida survivors initially have telomeres that are comparable in lengths with the wild type. The formation and behavior of Candida survivors appear to resemble more closely those of another budding yeast Kluyveromyces lactis, where senescence can also be accompanied by relatively mild growth defects (43). Indeed, a recent study of K.lactis telomeres demonstrates that senescence in telomerase-deficient cells can be suppressed by the introduction of a single elongated telomere (44). Given the greater heterogeneity of Candida telomere length distribution, the possibility that a few long telomeres might be stochastically present in telomerase-negative mutants even after prolonged passage to suppress senescence must also be considered. One potentially important distinction between the current analysis and previous studies of other budding yeasts is that all of our work was performed using diploid cells, whereas almost all of the published work on S.cerevisiae and K.lactis was done using haploid yeast. It remains to be seen if any of the distinct features of Candida with regard to telomere maintenance can be attributed primarily to its diploid nature.
Heightened telomere recombination in the Candida est1- mutant
The telomere dynamics of the C.albicans est1- mutant differs substantially from that of the other mutants and its S.cerevisiae counterpart (40,45). Long telomeres can persist upon extended subculturing, and there are instances of sudden telomere loss and acquisition (Figure 2B). While instances of sudden telomere loss and acquisition were also observed in est2- and est3- mutants, they were not as frequent, and telomeres in these mutants never reached the level of heterogeneity seen in est1- before the emergence of survivors. These results suggest that in est1- recombinational elongation of relatively long telomeres occurs at somewhat higher frequencies than in other mutants. This hypothesis can explain the absence of severe telomere shortening in early generations, and the maintenance of telomere content at the wild-type levels in later generations. It is tempting to speculate that loss of Est1p may increase the accessibility of telomeres to recombination proteins. In other word, Est1p appears to ‘protect’ telomeres against recombination that otherwise can occur only on severely shortened telomeres (46). A protective function for Est1 at telomeres was also suggested by recent analyses of the human and Schizosaccharomyces pombe homologs (47,48).
If the dynamics of telomeres in est1- were due to heightened recombination, then it ought to be dependent upon recombination proteins. Several attempts were made to generate an est1- rad52- mutant without success. Because unlike S.cerevisiae, loss of Rad52p alone in Candida causes severe growth defects (49), it is unclear whether our failure was due to this growth defect or possibly a synthetic interaction between telomerase and RAD52 in Candida. Additional studies are underway to address this issue.
Prevention of aberrant C-strand resection by Candida Est2p
Perhaps the most interesting finding from our investigation of telomere overhangs is the substantial buildup of such overhangs in the est2- mutant. This buildup can be seen even in early generations of the mutant. While accumulation of unpaired G-strand has been reported previously for an S.cerevisiae est1- mutant, the accumulation was only observed after severe telomere attrition and is likely to be mechanistically distinct (50). Study of S.cerevisiae yku and cdc13 mutants demonstrates that abnormally long G-strand is created through resection of the C-strand, at least in part by the EXO1 nuclease (22,50,51). That this phenotype is observed with the loss of Est2p (but not Est1p or Est3p) in C.albicans suggests that Est2p may have a similar role in protecting telomeres against excessive nucleolytic degradation. The apparently increased sensitivity of short telomeres to C-strand degradation suggests that the protective function of Est2p may be more important for short telomeres. Why should the Est2p component of telomerase alone possess this unique activity? One speculative explanation can be proposed based on earlier findings in S.cerevisiae. In S.cerevisiae, Est2p is localized to telomeres throughout the cell cycle, at least in part through an indirect association with the Ku complex (i.e. Est2p binds telomerase RNA, which in turn binds Ku) (52–54). The Est2p–Ku interaction may thus enhance the ability of the entire complex to prevent nuclease attack. In contrast, there is no evidence that the association of Est1p and Est3p with telomeres is stable throughout the cell cycle (16). Because telomerase RNA is also presumed to be part of the protective complex, it is tempting to speculate that deleting the RNA gene may also cause accumulation of G-strand overhangs. Further studies will be necessary to determine the validity of this hypothesis.
C-strand overhangs in C.albicans
Presence of C-strand overhangs has only been described previously for human telomeres using the T-OLA assay (55), and not for any fungi. Our detection of such overhangs in C.albicans is therefore unexpected. Nevertheless, the biology of C.albicans (e.g. the obligate diploid genome) is sufficiently distinct from S.cerevisiae to make such differences in telomere structure far from implausible. The most obvious change in the Candida C-strand overhangs associated with telomerase mutations was a moderate decrease observed in the est2- samples. The functional significance of this finding is unclear. It is worth noting that because the wild-type samples yielded stronger signals for both single-stranded and double-stranded telomeres, if the hybridization signals are not strictly linear with respect to the levels of single and double-stranded DNA, then the quantitative differences we calculated may be more apparent than real. Additional studies are necessary to address this possibility. (On the other hand, our conclusion with respect to G-strand alterations in the est2- mutant is not subject to the same caveat, because the mutant samples consistently yielded weaker signals for double-stranded telomeres, but stronger signals for single-stranded telomeres.)
CONCLUSIONS
Altogether, our analysis of the telomere length dynamics and G-strand variations in the mutant strains support a role for both Candida Est1p and Est2p in telomere protection. In the case of est1-, loss of protection is mainly manifested as heightened recombination at telomeres. In the case of est2-, loss of protection is mainly manifested as accumulation of G-strand overhangs. That the phenotypic consequences are different for the two mutants implies that the precise mechanisms of protection must also be different.
SUPPLEMENTARY DATA
Supplementary Data are available at NAR Online.
ACKNOWLEDGEMENTS
We are grateful to Dr Shu-Chun Teng for constructive comments on the manuscript. We also thank Sunitha Singh and You-Chin Lin for help with the construction of mutant strains, Alain Dandjinou for help with in-gel hybridization, and Drs E. D'Ambrosio, S. Stewart and I. Ben-Porath for advice on T-OLA. This work was supported by a New Investigator Award in Pathogenic Mycology from Burroughs Wellcome Fund and a grant from NIH. Funding to pay the Open Access publication charges for this article was provided by the Irma Hirschl Monique Weill-Caulier Trust.
REFERENCES
Blackburn, E. (2001) Switching and signaling at the telomere Cell, 106, 661–673 .
Ferreira, M., Miller, K., Cooper, J. (2004) Indecent exposure: when telomeres become uncapped Mol. Cell, 13, 7–18 .
Cech, T. (2004) Beginning to understand the end of the chromosome Cell, 116, 273–279 .
Griffith, J.D., Comeau, L., Rosenfield, S., Stansel, R.M., Bianchi, A., Moss, H., de Lange, T. (1999) Mammalian telomeres end in a large duplex loop Cell, 97, 503–514 .
de Lange, T. (2002) Protection of mammalian telomeres Oncogene, 21, 532–540 .
Olovnikov, A. (1973) A theory of marginotomy. The incomplete copying of template margin in enzymic synthesis of polynucleotides and biological significance of the phenomenon J. Theor. Biol, . 41, 181–190 .
Lingner, J., Cooper, J., Cech, T. (1995) Telomerase and DNA end replication: no longer a lagging strand problem? Science, 269, 1533–1534 .
Greider, C.W. and Blackburn, E.H. (1985) Identification of a specific telomere terminal transferase activity in Tetrahymena extracts Cell, 43, 405–413 .
Nugent, C.I. and Lundblad, V. (1998) The telomerase reverse transcriptase: components and regulation Genes Dev, . 12, 1073–1085 .
Kelleher, C., Teixeira, M., Forstemann, K., Lingner, J. (2002) Telomerase: biochemical considerations for enzyme and substrate Trends Biochem. Sci, . 27, 572–579 .
Harrington, L. (2003) Biochemical aspects of telomerase function Cancer Lett, . 194, 139–154 .
Lundblad, V. (2003) Telomere replication: an Est fest Curr. Biol, . 13, R439–R441 .
Hughes, T.R., Evans, S.K., Weilbaecher, R.G., Lundblad, V. (2000) The Est3 protein is a subunit of yeast telomerase Curr. Biol, . 10, 809–812 .
Evans, S. and Lundblad, V. (2002) The Est1 subunit of Saccharomyces cerevisiae telomerase makes multiple contributions to telomere length maintenance Genetics, 162, 1101–1115 .
Pennock, E., Buckley, K., Lundblad, V. (2001) Cdc13 delivers separate complexes to the telomere for end protection and replication Cell, 104, 387–396 .
Taggart, A., Teng, S., Zakian, V. (2002) Est1p as a cell cycle-regulated activator of telomere-bound telomerase Science, 297, 1023–1026 .
Singh, S. and Lue, N. (2003) Ever shorter telomere 1(EST1)-dependent reverse transcription by Candida telomerase in vitro: evidence in support of an activating function Proc. Natl Acad. Sci. USA, 100, 5718–5723 .
Neumann, A. and Reddel, R. (2002) Telomere maintenance and cancer—look, no telomerase Nature Rev. Cancer, 2, 879–884 .
Lundblad, V. (2002) Telomere maintenance without telomerase Oncogene, 21, 522–531 .
Kass-Eisler, A. and Greider, C. (2000) Recombination in telomere-length maintenance Trends Biochem. Sci, . 25, 200–204 .
Garvik, B., Carson, M., Hartwell, L. (1995) Single-stranded DNA arising at telomeres in cdc13 mutants may constitute a specific signal for the RAD9 checkpoint Mol. Cell. Biol, . 15, 6128–6138 .
Maringele, L. and Lydall, D. (2002) EXO1-dependent single-stranded DNA at telomeres activates subsets of DNA damage and spindle checkpoint pathways in budding yeast yku70Delta mutants Genes Dev, . 16, 1919–1933 .
Yang, Q., Zheng, Y., Harris, C. (2005) POT1 and TRF2 cooperate to maintain telomeric integrity Mol. Cell. Biol, . 25, 1070–1080 .
Baumann, P. and Cech, T.R. (2001) Pot1, the putative telomere end-binding protein in fission yeast and humans Science, 292, 1171–1175 .
Bertuch, A. and Lundblad, V. (2003) Which end: dissecting Ku's function at telomeres and double-strand breaks Genes Dev, . 17, 2347–2350 .
Fisher, T. and Zakian, V. (2005) Ku: A multifunctional protein involved in telomere maintenance DNA Repair, 4, 1215–1226 .
Gravel, S., Larrivee, M., Labrecque, P., Wellinger, R.J. (1998) Yeast Ku as a regulator of chromosomal DNA end structure Science, 280, 741–744 .
Polotnianka, R.M., Li, J., Lustig, A.J. (1998) The yeast Ku heterodimer is essential for protection of the telomere against nucleolytic and recombinational activities Curr. Biol, . 8, 831–834 .
Chan, S. and Blackburn, E. (2003) Telomerase and ATM/Tel1p protect telomeres from nonhomologous end joining Mol. Cell, 11, 1379–1387 .
Kim, M., Xu, L., Blackburn, E. (2003) Catalytically active human telomerase mutants with allele-specific biological properties Exp. Cell Res, . 288, 277–287 .
McEachern, M.J. and Hicks, J.B. (1993) Unusually large telomeric repeats in the yeast Candida albicans Mol. Cell. Biol, . 13, 551–560 .
Singh, S.M., Steinberg-Neifach, O., Mian, I.S., Lue, N.F. (2002) Analysis of telomerase in Candida albicans: potential role in telomere end protection Eukaryot Cell, 1, 967–977 .
Wilson, R.B., Davis, D., Mitchell, A.P. (1999) Rapid hypothesis testing with Candida albicans through gene disruption with short homology regions J. Bacteriol, . 181, 1868–1874 .
Fonzi, W. and Irwin, M. (1993) Isogenic strain construction and gene mapping in Candida albicans Genetics, 134, 717–728 .
Hoffman, C. and Winston, F. (1987) A ten-minute DNA preparation from yeast efficiently releases autonomous plasmids for transformation of Escherichia coli Gene, 57, 267–272 .
Dionne, I. and Wellinger, R.J. (1996) Cell cycle-regulated generation of single-stranded G-rich DNA in the absence of telomerase Proc. Natl Acad. Sci. USA, 93, 13902–13907 .
Cimino-Reale, G., Pascale, E., Battiloro, E., Starace, G., Verna, R., D'Ambrosio, E. (2001) The length of telomeric G-rich strand 3'-overhang measured by oligonucleotide ligation assay Nucleic Acids Res, . 29, E35 .
Stewart, S., Ben-Porath, I., Carey, V., O'Connor, B., Hahn, W., Weinberg, R. (2003) Erosion of the telomeric single-strand overhang at replicative senescence Nature Genet, . 33, 492–496 .
Enloe, B., Diamond, A., Mitchell, A. (2000) A single-transformation gene function test in diploid Candida albicans J. Bacteriol, . 182, 5730–5736 .
Lundblad, V. and Blackburn, E.H. (1993) An alternative pathway for yeast telomere maintenance rescues est1-senescence Cell, 73, 347–360 .
Teng, S., Chang, J., McCowan, B., Zakian, V. (2000) Telomerase-independent lengthening of yeast telomeres occurs by an abrupt Rad50p-dependent, Rif-inhibited recombinational process Mol. Cell, 6, 947–952 .
Teng, S. and Zakian, V. (1999) Telomere-telomere recombination is an efficient bypass pathway for telomere maintenance in Saccharomyces cerevisiae Mol. Cell. Biol, . 19, 8083–8093 .
McEachern, M.J. and Blackburn, E.H. (1996) Cap-prevented recombination between terminal telomeric repeat arrays (telomere CPR) maintains telomeres in Kluyveromyces lactis lacking telomerase Genes Dev, . 10, 1822–1834 .
Topcu, Z., Nickles, K., Davis, C., McEachern, M. (2005) Abrupt disruption of capping and a single source for recombinationally elongated telomeres in Kluyveromyces lactis Proc. Natl Acad. Sci. USA, 102, 3348–3353 .
Lendvay, T.S., Morris, D.K., Sah, J., Balasubramanian, B., Lundblad, V. (1996) Senescence mutants of Saccharomyces cerevisiae with a defect in telomere replication identify three additional EST genes Genetics, 144, 1399–1412 .
McEachern, M. and Iyer, S. (2001) Short telomeres in yeast are highly recombinogenic Mol. Cell, 7, 695–704 .
Beernink, H., Miller, K., Deshpande, A., Bucher, P., Cooper, J. (2003) Telomere maintenance in fission yeast requires an est1 ortholog Curr. Biol, . 13, 575–580 .
Reichenbach, P., Hoss, M., Azzalin, C., Nabholz, M., Bucher, P., Lingner, J. (2003) A human homolog of yeast est1 associates with telomerase and uncaps chromosome ends when overexpressed Curr. Biol, . 13, 568–574 .
Ciudad, T., Andaluz, E., Steinberg-Neifach, O., Lue, N., Gow, N., Calderone, R., Larriba, G. (2004) Homologous recombination in Candida albicans: role of CaRad52p in DNA repair, integration of linear DNA fragments and telomere length Mol. Microbiol, . 53, 1177–1194 .
Hackett, J. and Greider, C. (2003) End resection initiates genomic instability in the absence of telomerase Mol. Cell. Biol, . 23, 8450–8461 .
Bertuch, A. and Lundblad, V. (2004) EXO1 contributes to telomere maintenance in both telomerase-proficient and telomerase-deficient Saccharomyces cerevisiae Genetics, 166, 1651–1659 .
Fisher, T., Taggart, A., Zakian, V. (2004) Cell cycle-dependent regulation of yeast telomerase by Ku Nature Struct. Mol. Biol, . 11, 1198–1205 .
Stellwagen, A., Haimberger, Z., Veatch, J., Gottschling, D. (2003) Ku interacts with telomerase RNA to promote telomere addition at native and broken chromosome ends Genes Dev, . 17, 2384–2395 .
Smith, C., Smith, D., De, R.J., Blackburn, E. (2003) Telomeric protein distributions and remodeling through the cell cycle in Saccharomyces cerevisiae Mol. Biol. Cell, 14, 556–570 .
Cimino-Reale, G., Pascale, E., Alvino, E., Starace, G., D'Ambrosio, E. (2003) Long telomeric C-rich 5'-tails in human replicating cells J. Biol. Chem, . 278, 2136–2140 .(Olga Steinberg-Neifach and Neal F. Lue*)
*To whom correspondence should be addressed. Tel: +1 212 746 6506; Fax: +1 212 746 8587; Email: nflue@med.cornell.edu
ABSTRACT
The telomerase ribonucleoprotein in Candida albicans is presumed to contain at least three Est proteins: CaEst1p, CaEst2p/TERT and CaEst3p. We constructed mutants missing each of the protein subunit of telomerase and analyzed overall telomere dynamics and single-stranded telomere overhangs over the course of many generations. The est1- mutant manifested abrupt telomere loss and recovery, consistent with heightened recombination. Both the est2- and est3- mutant exhibited progressive telomere loss, followed by the gradual emergence of survivors with long telomeres. In no case was telomere loss accompanied by severe growth defects, suggesting that cells with short telomeres can continue to proliferate. Furthermore, the amount of G-strand terminal overhangs was greatly increased in the est2- mutant, but not others. Our results suggest that in addition to their well-characterized function in telomere elongation, both CaEst1p and CaEst2p mediate some aspects of telomere protection in Candida, with the former suppressing excessive recombination, and the latter preventing excessive C-strand degradation.
INTRODUCTION
Telomeres are nucleoprotein structures located at the ends of chromosomes that are crucial for maintaining chromosome stability (1–3). In the majority of organisms, telomeric DNA consists of short repetitive G-rich sequences that end with a single-stranded 3' overhang. In mammalian cells, the overhang can invade a more proximal region of telomeres to form a specialized D-loop, known as T-loop (4,5). Functionally important proteins are recruited to telomeres through both DNA–protein and protein–protein interactions. Both the DNA and protein components of telomeres are essential to the maintenance of chromosome stability.
Loss of telomere DNA can occur with each round of replication due to the so-called ‘end replication problem’ (6,7). Two compensatory mechanisms that counteract telomere loss have been identified. The first entails the addition of nucleotides to the G-rich strand of telomeres by a specialized reverse transcriptase named telomerase (8–11). Telomerase is a ribonucleoprotein complex whose enzymatic core consists of a catalytic protein (named TERT in general and Est2p in yeast) and a template RNA (named TLC1 in yeast) (9). In the budding yeast Saccharomyces cerevisiae the telomerase complex contains at least two additional protein subunits, named Est1p and Est3p (12–14). Both subunits are required for telomere extension in vivo, but not essential for telomerase activity in vitro (9). Est1p has been subjected to detailed analysis and shown to mediate the recruitment of the telomerase complex to telomere ends by interacting with a G-strand telomere-binding protein named Cdc13p (15,16). It also appears to perform a post-recruitment or activating function (14,16). A recent study further suggests a role for the Candida Est1p in defining the substrate preference of telomerase in vitro (17). The second mechanism of telomere elongation entails recombination, and is primarily observed in telomerase-deficient cells (18–20). For example, in several budding yeasts, loss of telomerase is initially accompanied by progressive telomere shortening and senescence. However, rare populations of cells (called ‘survivors’) are able to activate a recombination-based mechanism(s) for telomere elongation and regain normal growth.
Equally important in telomere maintenance is protection of telomeric DNA from excessive nucleolytic degradation. Both single- and double-stranded telomere-binding proteins have been implicated in telomere protection. A well-studied example of single-stranded telomere-binding protein is Cdc13p in S.cerevisiae. Evidence for the protective function of Cdc13p came from analysis of a cdc13ts mutant, which suffers from extensive nucleolytic degradation of the C-rich strand of telomeres (21). C-strand degradation in this mutant is EXO1-dependent, implying a role for Cdc13p in antagonizing the activity of EXO1 (22). A family of proteins that are similar in function to Cdc13p have been identified in diverse organisms, and are known as Pot1 (23,24). In fission yeast, loss of Pot1 function leads to telomere degradation and telomere–telomere fusion (24). In humans, RNAi-mediated knock down of Pot1 provokes changes in telomere structure and induces chromosome instability (23). Among double-stranded telomere-binding proteins, the heterodimeric Ku complex, in particular, has a well-characterized role in telomere protection. Initially identified as a key mediator of non-homologous end joining, the Ku complex was shown recently to perform multiple functions at telomeres (25,26). Like the cdc13ts mutant, yku knockout strains exhibit elevated levels of G-strand overhangs at telomeres (27,28), which can also be attributed to aberrant degradation of the C-strand (22,27,28). Interestingly, several recent studies suggest that the telomerase complex, aside from its well-known function in telomere elongation, may have an additional role in telomere protection. For example, in the absence of telomerase, telomeres in the budding yeast are not only lost gradually, but also exhibit increased fusion to double-strand breaks (29). The fusion rate was even higher if the telomerase mutation was combined with the loss of Tel1p, a telomere-binding checkpoint protein. As another example, expression of telomerase mutants in human fibroblasts with shortened telomeres can prolong their life span without causing bulk telomere elongation (30).
In this study, we used Candida albicans as a model system for investigating telomerase function and regulation. C.albicans is an opportunistic fungal pathogen that can cause systemic infection in immuno-compromised individuals. It is an attractive model system because of its possession of longer telomere tracts (2–5 Kb in the BWP17 wild-type strain), a 23 bp regular telomere repeat (31), and the availability of genome sequence. In our previous work we showed that loss of each of the individual protein component of the telomerase complex in C.albicans caused specific defects in telomere maintenance (32). However, we failed to observe either cell senescence, as evidenced by growth retardation, or emergence of survivors, as evidenced by improved growth and sudden telomere elongation after severe shortening. To confirm and extend these findings, we reconstructed the mutants using a different set of disruption cassettes and assessed telomere length and G- and C-strand overhangs over many more generations. Here we report that with sufficient number of passages, telomere shortening followed by the apparent emergence of survivors can be observed in the telomerase mutants, albeit without frank senescence. The est1- mutant manifested abrupt telomere loss and recovery, consistent with heightened recombination. In addition, loss of Est2p was accompanied by a substantial increase in the amount of G-strand overhangs, consistent with a role for this protein in preventing aberrant degradation of the recessed C-strand. Our results suggest that in addition to their well-characterized functions in telomere elongation, both CaEst1p and CaEst2p mediate some aspects of telomere protection in Candida.
MATERIALS AND METHODS
Plasmids and strains
C.albicans BWP17 (ura3-::-imm434/ura3-::-imm434 his1::hisG/his1::hisG arg4::hisG/arg4::hisG, a gift from A. Mitchell (Columbia University)), was derived from CAI4 as described previously (33). The double knockout mutants est1-, est2- and est3- were created using the ‘URA-blaster’ method (34). For the disruption of EST1, a 426 bp fragment of the noncoding sequence upstream of the EST1 open reading frame (ORF) was inserted 5' to the URA-blaster cassette, and a 920 bp fragment downstream of the ORF was inserted 3' to the cassette. For the disruption of EST2, a 768 bp fragment of the noncoding sequence upstream of the EST2 ORF was inserted 5' to the URA-blaster cassette, and an 860 bp fragment downstream of the ORF was inserted 3' to the cassette. For the disruption of EST3, a 560 bp fragment of the noncoding sequence upstream of the EST3 ORF was inserted 5' to the URA-blaster cassette, and a 980 bp fragment downstream of the ORF was inserted 3' to the cassette.
C.albicans transformations were carried out as described previously (34). Transformants were first selected on SD-Ura plates and tested for the correct integration of the URA-blaster cassette by Southern blots. EST/est clones that lost uracil prototrophy were selected for their ability to grow on 5-FOA plates and used for the second round of transformation with the same cassette. The loss of both alleles of EST genes was verified by Southern blots. All strains were grown in the YPD medium, and supplemented with 80 μg/ml uridine when necessary.
Detection of double-stranded telomeres
For analysis of telomeres over multiple generations, cells were streaked on YPD plates, grown for 2 days at 30°C, and one colony was picked for further re-streaking and for inoculation into 5 ml YPD. The cultures were grown at 30°C and the cells collected after 16–20 h of growth. Chromosomal DNA was isolated by the ‘Smash and Grab’ method (35). The DNA (5 μg) was digested with a combination of AluI and NlaIII. Restriction fragments were fractionated on a 0.9% agarose gel and transferred to a nylon membrane. Telomeric DNA was detected by hybridization at 65°C to a 5' end-labeled 46mer oligonucleotide containing two copies of the C.albicans telomere repeat. Subsequently, the nylon membrane was stripped of the telomeric probe and hybridized to a 32P-labeled fragment of the RAD52 gene. To compare the total telomeric DNA content in different samples, the signals from the telomeric and RAD52 blots were quantified using a PhosphorImager system (Molecular Dynamics), and the telomeric DNA signal was normalized against the RAD52 signal.
Plating efficiency
Colonies of the est2- mutant at different points during the propagation were inoculated in YPD supplemented with uridine (80 μg/ml) and grown for 16 h. Cell cultures were diluted with water and the cell densities determined using a Nuebauer hemocytometer. An aliquot containing 300 cells was plated on a YPD plate supplemented with uridine (80 μg/ml). Plates were incubated for 2 days at 30°C and the number of colonies counted. Plating efficiency was calculated as the ratio of the number of colonies on the plate after 2 days of incubation to the colony forming units calculated using the hemocytometer (expressed as percentages), and the results were plotted. All experiments were performed in duplicates. Note that because clumped cells were not counted, the actual number of cells used for plating was probably higher than that determined by the hemocytometer, thus resulting in plating efficiency that was higher than 100%.
In-gel hybridization assay
In-gel hybridization assays were performed as described previously with some modifications (36). Briefly, genomic DNA isolated from the wild-type and mutant Candida strains was digested with AluI and NlaIII, and the resulting fragments were resolved on a 0.9% agarose gel. The gel was dried on a vacuum drier for 1 h at room temperature, and subjected to pre-hybridized in the Church mix (supplemented with 100 μg/ml of boiled calf thymus DNA) for 1 h at 55°C. Hybridization was done for 3 h at 55°C using 32P-labeled 23mer oligonucleotides complementary to either the G- or C-strand of telomeres. Subsequently, the gel was washed three times for 20 min in 4x SSC at room temperature and three times for 20 min in 4x SSC, 0.1% SDS at 55°C and exposed to a PhosphorImager screen. After an image was acquired, the gel was denatured by immersing in 0.6 M NaCl, 0.2 M NaOH for 1 h, neutralized in 1.5 M NaCl, 0.5 M Tris, pH 7.4, for 1 h, and then washed with water for 30 min. The denatured DNA in the gel was re-probed using the same oligonucleotide. The relative levels of single-stranded DNA were calculated as follows. (i) Hybridization signals for single-stranded and double-stranded telomeric DNA in each sample were quantified using a PhosphorImager system (Molecular Dynamics Inc.) and the background subtracted. (ii) The ratios for the single-stranded versus double-stranded signals were obtained for each sample. (iii) The ratios for the mutant samples were normalized against that for a wild-type sample.
Telomere oligonucleotide ligation (T-OLA) assay
The T-OLA assay was performed as described previously except for a few minor modifications (37). Briefly, 10 pmol of a 23mer oligonucleotide complementary to the G-strand or to the C-strand of telomeres (ACACCAAGAAGTTAGACATCCGT and ACGGATGTCTAACTTCTTGGTG, respectively) were 5' end-labeled using 20 U of OptiKinase (Amersham) and 10 pmol of ATP (6000 Ci/mmol, 10 mCi/ml) (Perkin Elmers) in 50 μl reaction containing 1x OptiKinase buffer (Amersham) for 30 min at 37°C, at which point 1 μl of 0.1 M cold ATP and 10 U of OptiKinase were added, followed by 15 min incubation at 37°C. Reaction was terminated by addition of 50 μl TE and phenol/chloroform extraction. Labeled oligonucleotides were purified using Nick columns (Amersham), and concentrated by ethanol precipitation and resuspension in 40 μl water. Chromosomal DNA (5 μg) was mixed with 0.5 pmol of labeled oligonucleotides in 1x Taq DNA ligase buffer (NEB) in 20 μl volume, and incubated at 55°C for 14–16 h. At the end of incubation, 40 U of Taq DNA ligase (NEB) were added and ligation was allowed to proceed for 5 h at 55°C. Ligation was terminated by addition of 80 μl water and phenol/chloroform extraction. DNA was precipitated by the addition of ethanol and dissolved in 55 μl water. As a loading control, one-eleventh of the sample was removed for DNA quantification by electrophoresis and Ethidium Bromide staining. The remainder of the sample was dried in a vacuum dryer, dissolved in 1x sample buffer and analyzed on a 6% denaturing polyacrylamide gel. Radioactive signals were quantified using a PhosphorImager system (Molecular Dynamics), normalized against the loading control, and the relative levels of single-stranded DNA calculated as described before (38).
Slot blot
The slot/dot blot procedure was basically performed as described previously (21). Briefly, 4 μg of AluI and NlaIII-digested genomic DNA was used as the starting material. The DNA was diluted in water to a volume of 660 μl, from which a 60 μl aliquot was transferred to a fresh tube and subjected to alkali-heat denaturation (addition of NaOH to 0.2 M, followed by 30 min incubation at 65°C). Both native and denatured DNA samples were further diluted to 1200 μl in 5x SCP buffer (0.5 M NaCl, 0.15 M Na2HPO4 and 0.01 mM EDTA). Aliquots of DNA (300 μl per slot, four slots per sample) were applied to a Hybond-N membrane (Amersham Biosciences) by using a Microfiltration Apparatus (Bio-Rad) and immediately washed three times with equal volume of 10x SCP each. The moist membrane was exposed to UV light (Stratagene) to effect crosslinking and then divided into two equal halves. Each half of the membrane contained two slots of native and two slots of denatured DNA from each sample. Membranes were probed separately with 32P-labeled 46mer oligonucleotides complementary to either the G- or C-strand of telomeres. Relative levels of single-stranded DNA were calculated as follows. (i) Hybridization signals for single-stranded DNA and double-stranded DNA in each sample were quantified and the background subtracted. (ii) The ratios for the single-stranded versus double-stranded signals were obtained for each sample. (iii) The ratios for the mutant samples were normalized against that for a wild-type sample.
Nuclease treatment
Chromosomal DNA was isolated by the ‘Smash and Grab’ method (35) and was further purified using a Nick column (Amersham). To remove the 3' single-stranded telomere overhang, ExoI (USB) was used in a reaction containing 1 U of enzyme per 50 ng of DNA in 67 mM glycine (pH 9.5), 10 mM 2-mercaptoethanol, and 6.7 mM MgCl2, and the reaction was allowed to proceed for 24 h at 37°C. Alternatively, ExoI (Epicentre) was used in a reaction that contained 50 mM Tris–HCl (pH 7.9), 10 mM MgCl2, 100 mM NaCl and 1 mM DTT. To remove both protruding 3' and 5' single-stranded telomere overhangs, Exonuclease VII (ExoVII, USB) was used in a reaction containing 2 U of enzyme per μg of DNA in 50 mM Tris–HCl (pH 7.9), 50 mM potassium phosphate (pH 7.6), 8.3 mM EDTA, 10 mM 2-mercaptoethanol and the reaction was performed at 37°C for 30 min. Reactions were terminated by phenol/chloroform extraction and processed for subsequent assays.
RESULTS
Construction of the C.albicans est mutants
Previously, we reported results of a study in which the telomere dynamics of C.albicans mutants was followed through 250–300 generations (32). Though some defects in telomere maintenance were observed, we failed to detect senescence or the emergence of survivors, as judged by changes in the growth rate of cultures. To explore the possibility that more extended passaging of the mutants may reveal additional phenotypes, we reconstructed the mutants, propagated them for 800 generations, and analyzed telomere lengths and structure during the passages. For the current study we deleted the EST genes using the ‘URA-blaster’ method. In comparison with the ‘One-Step Transformation’ method used in the earlier study, the ‘URA-blaster’ technique has the advantage of precluding the homozygosis of genes located distal to the disrupted gene (32,39). Transformation cassettes in the current study were created by inserting short DNA fragments that flank the ORF of the target gene upstream and downstream of the ‘URA-blaster’ cassette (Materials and Methods). Double deletion mutants were derived from two rounds of transformation and selection and confirmed by Southern blots (data not shown). The mutants were propagated by re-streaking every 2 days for isolated colonies. It is estimated that each restreak corresponds to 25 generations of growth.
Growth and telomere dynamics of the est mutants during extended passage
We next investigated the overall telomere dynamics of the wild type and mutants during extended passage. For the wild-type parental strain used in this study (BWP17), the majority of NlaIII and AluI digested telomere restriction fragments are between 2 and 5 kb in size. Consistent with the earlier report, we found that the size of telomere fragments and their distribution in the wild-type strain differed from one streak to the next (Figure 1A). However, the variations between the total intensity of the telomeric signals over time are no more than 20%. We have passaged the BWP17 for a total of 25 streaks without observing noticeable changes in telomeric content (data not shown). These data are in agreement with our earlier observation that some changes in telomere length occur normally during the propagation of C.albicans cells, but a homeostatic mechanism must be acting to maintain the telomeric content of the cell within a range.
Figure 1 Telomere dynamics in the wild-type and est mutant strains. Telomeres dynamics was analyzed in successive generations of the wild-type and the est mutant strains by Southern blots. The blots were stripped and re-probed using a labeled RAD52 fragment and the results shown at the bottom. The strains and passages analyzed are (A) BWP17, streaks 2–12; (B) est1-, streaks 1–33; (C) est2-, streaks 1–35; (D) est3-, streaks 1–35. Instances of sudden appearance and disappearance of telomere restriction fragments are marked by arrows.
For analysis of mutants, we examined the telomere dynamics of two or three clones (three for est2- and two each for est1- and est3-) for each telomerase mutant, which yielded similar results. The newly derived est1- strains were passaged for >30 restreaks after the selection of the mutant (Figure 1B). Significant telomere loss was observed during passage, with the total telomere content showing an almost 50% decrease at streak 11 (data not shown). However, large telomeric fragments were still present at streak 17. Presence of long telomeres in combination with overall telomere loss in the est1- strain suggests that telomeres in the mutant were more heterogeneous in size than the wild type. Long telomeres became absent for several passages (streak 19–25) but reappeared thereafter. This was accompanied by restoration of telomere content to wild-type levels. Such an abrupt increase of telomere length en masse is reminiscent of type II survivors that arose after senescence in telomerase-negative S.cerevisiae strains. We have thus adopted the name ‘survivors’ to describe the Candida cells, notwithstanding some phenotypic differences between the Saccharomyces and Candida est1 mutant. Most notably, in the C.albicans est1- clone there was no decrease in the growth rate during passage (data not shown), and the telomere content of the mutant never fell below 20% of the wild type. Even in the population with the shortest average telomere length, it was possible to detect telomeric restriction fragments that were at least 2 kb long. After the emergence of ‘survivors’, telomere content of the est1- clone became comparable with that of the wild type and was maintained during the remaining passages. It is unclear if additional cycles of telomere loss and restoration can occur in these cells with even more extended subculturing.
In contrast to Est1p, loss of Candida Est2p resulted in progressive telomere attrition (Figure 1C). Notably, while occasional lengthening and shortening events were observed in the est2- mutant, they were not as frequent as in the case of est1- mutant (compare Figure 1B and C, see also Supplementary Figure 1), nor did telomere length distribution in any est2- clone became as heterogeneous as est1-. All telomeres became shorter than 1 kb at about the 19th streak, but subsequently underwent significant elongation, which was followed by a second round of contraction. This dynamics, which is similar to that of S.cerevisiae telomerase mutants, was not observed in the previous study of another Candida est2- mutant, most likely due to the shorter duration of the earlier experiment (32). It is also possible that the earlier mutant may have suffered some additional genetic changes due to the ‘UAU1’ knockout protocol. The greater reliability of the URA-blaster transformation protocol and the more extended passaging of the mutant suggest that the current results are more reflective of the true consequences of deleting EST2 in Candida. Like the Candida est1- mutant, the est2- ‘survivors’ exhibited features that were distinct from their S.cerevisiae counterparts. First, loss of Est2p functions in C.albicans did not result in severe growth defects even after prolonged propagation. Mild growth retardation was detected in the est2- strain in liquid media (Figure 2A). However, this was true for cultures of early and late passages, and does not appear to be due primarily to telomere attrition. From the exponential phase of the growth curves, it was estimated that the doubling time of the mutant was increased by 20%. Growth of est2- was further investigated by measurement of plating efficiency. As shown in Figure 2B, the plating efficiency of the mutant was slightly reduced (by 20%) when monitored at three different points during passage. Thus, the growth retardation of est2- may be due to loss of viability of a small fraction of the cells. Another distinct feature of C.albicans est2- mutant relates to the kinetics of telomere elongation in the cell population. At streak 21, there was an abrupt broadening in the length distribution of telomeres, indicating that the population of cells contained both very long and short telomeres. Whether the long and short telomeres co-exist in a single cell is unclear. Over the next few streaks the relative intensity of the long telomeres increased steadily and became dominant. These observations suggest that cells with shortened telomeres in the est2- mutant culture were able to proliferate and co-exist with apparent ‘survivors’. Lastly, the telomere lengths of Candida est2- survivors are comparable with those of wild-type cells, whereas the Saccharomyces type II survivors have telomeres that are dramatically longer than normal. Thus, as in the case of est1 mutants, there are apparent similarities as well as significant differences between the Candida and S.cerevisiae est2 survivors.
Figure 2 Growth of the est2- mutant. (A) The doubling time for the wild-type and est2- strain during exponential growth was determined at three different points during passage and plotted. (B) The plating efficiency of the wild-type and est2- strain during passage was determined as described in Materials and Methods and plotted.
Telomeres in the new est3- mutants experienced progressive loss (Figure 1D), just as described for the previous est3- mutant (32). Similar to est1- and in contrast to est2-, no growth retardation was evident during passage of the est3- mutant (data not shown). At streak 21, after severe telomere attrition, there was an abrupt appearance of long telomeres, which became pre-dominant during the later streaks. Interestingly, unlike the est2- strain, bulk telomere content after recovery was preserved in 1 est3- survivor for 10 streaks (Figure 1D), suggesting potential variability in the dynamics of the survivors. Analysis of more clones will be necessary to confirm the prevalence of this variability. We conclude that telomere attrition followed by the apparent emergence of survivors can be observed in Candida following prolonged passage of telomerase-deficient clones.
Detection of G- and C-strand overhangs at Candida telomeres using in-gel hybridization
It has been suggested that telomerase may have a protective function that is independent of its telomere extension function. Loss of telomere protection has been linked to changes in the level of terminal overhangs. We therefore investigated the impact of Candida telomerase mutations on the amount of telomere overhangs. Several methods have been developed previously for the purpose of quantifying single-stranded telomere overhangs (21,37,39). For analysis of Candida telomeres, we adopted in-gel hybridization as our primary assay because this technique has been used widely by many laboratories to analyze telomeres in different organisms. As described later, we also confirmed the key conclusions of the in-gel analysis using two additional assays.
We first investigated samples of chromosomal DNA from BWP17 using in-gel hybridization, and found that low levels of G- and C-strand signals could be detected at telomeres (Figures 3 and 4 and data not shown). The G-strand signal was abolished by pre-treatment of DNA with ExoI, a single-stranded 3'5' exonuclease (Figure 3A, lanes 3 and 4). In contrast, the bulk telomere signals obtained after denaturation of the DNA were unaffected by this treatment (Figure 3A, lanes 5 and 6). Both the G-strand and C-strand signals for native DNAs were also sensitive to ExoVII, a single-strand specific 3'5' and 5'3' exonuclease (Figure 3B and data not shown). These results indicate that most of the native in-gel signals were due to terminal overhangs. We note that the native in-gel signals were generally less discreet in size than the signals obtained after denaturation and re-probing. One potential explanation is that DNA fragments with significant overhangs may represent a small fraction of the telomere restriction fragments, and may behave more heterogeneously due to variable amount of single-stranded overhangs. The native and denatured signals, however, do exhibit similar size distributions in a given sample, consistent with their similar origins (Figures 3 and 4).
Figure 3 The effect of nuclease treatment on the amount of unpaired telomeric overhangs. (A) DNA isolated from BWP17 was incubated with or without E.coli ExoI, and then treated with AluI and NlaIII. The resulting samples were first analyzed by gel electrophoresis and ethidium bromide staining (left panel). The gel was subsequently dried and subjected to in-gel hybridization using a probe complementary to the G-strand of Candida telomere repeat (middle panel). The DNA in the gel was then denatured and hybridized again to the same probe (right panel). (B) DNA isolated from BWP17 was incubated with or without E.coli ExoVII, and then treated with AluI and NlaIII. The resulting samples were first analyzed by gel electrophoresis and ethidium bromide staining (left panel). The gel was subsequently dried and subjected to in-gel hybridization using a probe complementary to the C-strand of Candida telomere repeat (middle panel). The DNA in the gel was then denatured and hybridized again to the same probe (right panel).
Figure 4 Analysis of G-strand overhangs in the wild-type and est2- mutant clones. (A) (Upper panel) Genomic DNA was isolated from overnight cultures of the wild-type strain and the indicated streaks of the est2- mutant, and subjected to in-gel hybridization analysis using an oligonucleotide probe complementary to the G-strand of Candida telomeres. (Middle panel) After probing the native DNA with strand-specific probes, the DNA in the gel was denatured and hybridized again with the same probe. (Lower panel) The hybridization signals for single- and double-stranded telomeres were quantified, and the ratios plotted. (B) Sameas (A) except that the probe was designed to anneal to the C-strand of Candida telomeres. (C) The intensity traces for the lanes labeled ‘*’ and ‘**’ in Figure 4A were obtained using the Imagequant program (Molecular Dynamics Inc.). The native and denatured profiles for each sample are plotted alongside each other.
G- and C-strand overhangs in est mutants
Chromosomal DNAs derived from Candida est mutants were first subjected to in-gel hybridization analysis using a probe complementary to the G-strand. For each mutant, we investigated two independent clones over 10 successive streaks. Remarkably, loss of Est2p had a strong impact on the level of unpaired G-strand: there was an up to 20-fold increase in the amount of G-strand overhangs during passage, when the G-strand signals were normalized against total telomeric DNA (Figure 4A). The increase was observed in both mutant clones and in multiple samples, and the signals were sensitive to ExoI and ExoVII treatment (Supplementary Figure 2). These results suggest that loss of Candida Est2p may result in aberrant degradation of the 5' end-containing C-strand, thus leading to the accumulation of unpaired G-strand. Comparison of the signals in native and denatured gel images indicates that within a given sample, shorter DNA fragments possess higher levels of G-strand overhangs than longer fragments. For example, the ‘*’ sample yielded two major peaks in both the native and denatured analysis (3.6 and 2.3 kb,Figure 4A and C). The signal for the lower peak was stronger in the native analysis, but weaker in the denatured analysis (Figure 4C, left). Likewise, the lower peak in the ‘**’ sample was comparable in intensity to the upper peak in the native analysis, but much weaker in the denatured analysis (Figure 4C, right). Quantitative analysis of other samples yielded similar results (data not shown). Thus, shorter telomeres appear to be more susceptible to C-strand degradation in the est2- mutant. In contrast to est2-, the amount of G-strand overhang in the est1- and est3- mutant were quite low and indistinguishable from that found in the parental strain (Figure 5A and B).
Figure 5 Analysis of G-strand overhangs in the est1- and est3- mutant clones. (A and B) (Upper panel) Genomic DNA was isolated from overnight cultures of the wild-type strain and the indicated streaks of the est1- or est3- mutant, and subjected to in-gel hybridization analysis using an oligonucleotide probe complementary to the G-strand of Candida telomeres. (Middle panel) After annealing the native DNA with the G-strand-specific probe, the DNA in the gel was denatured and hybridized again with the same probe. (Lower panel) The ratios of G-strand signals to double-stranded telomere signals were calculated for both the wild-type and mutant samples and the averages and standard deviations plotted.
Unpaired C-strand in the est mutants was also investigated by in-gel hybridization using a probe that corresponds to the G-strand repeat. Again, loss of Est2p had the most impact on the amount of unpaired C-strand (Figure 4C). There was a moderate decrease (40–70%) in the amount of C-strand overhang in the two clones of est2- mutant during the course of 10 streaks. In contrast to est2-, the amount of C-strand overhangs in the est1- and est3- mutant was similar to that found in the parental strain (data not shown). Thus, loss of individual subunits of the Candida telomerase complex has different impacts on the telomere overhang structure.
Analysis of telomere overhangs by the T-OLA and slot blot assay
Because changes in telomere overhangs have not been observed in pre-senescent telomerase mutants before, we sought to verify our in-gel hybridization results for est2- using two other assays, namely the slot blot and T-OLA assay (21,36,37). These assays detect single-stranded telomere overhangs by hybridizing native DNAs on nylon membranes and in solution, respectively, to strand-specific probes. In the case of T-OLA, multiple copies of the probe that annealed to the single-stranded overhangs are further ligated and subsequently analyzed by gel electrophoresis. Chromosomal DNAs derived from wild-type and est2- mutant were subjected to both assays. As shown in Figures 6 and 7, the est2- mutant consistently displayed higher levels of G-strand signals in comparison with the wild-type samples in both T-OLA and slot blot assays. The average increases were 7- and 10-fold in the T-OLA and slot blot assays, respectively (Figures 6B and 7). In contrast, the C-strand signals yielded by the mutant samples were somewhat lower than the wild type (Figure 7 and data not shown). These findings are entirely consistent with the results of the in-gel hybridization analysis. Thus, alterations in the amount of telomere overhangs in the est2- mutant can be detected by three different assays.
Figure 6 T-OLA analysis of unpaired G-strand overhangs in the wild-type and est2- strains. (A) G-strand overhang dynamics was analyzed in successive generations of the wild type and the est2- mutant strain by T-OLA. Selected T-OLA products are designated by the number of repeats that they contain. The strains and passages analyzed in the different parts of the figure are BWP17, streaks 4–10; est2-, streaks 1–9. (B) The relative levels of G-strand overhangs were quantified for the wild-type (streaks 4, 6, 8 and 10) and est2- samples (streaks 1–9) as described in Materials and Methods, and the averages and deviations plotted.
Figure 7 Slot blot analysis of unpaired G- and C-strand overhangs in est2- mutant. Genomic DNA was isolated from overnight cultures of the wild-type strain and the indicated streaks of the est2- mutant, and subjected to slot blot analysis. Note that 10 times more native DNA than denatured DNA was applied to the slots in order to generate appreciable signals in these analyses (Materials and Methods). In all cases analyzed, the denatured DNA yielded a signal that is at least 30 times stronger than the native DNA when corrected for this difference in application. Relative ratios of single-stranded to double-stranded telomere DNA for wild-type (duplicate samples for streak 2) and est2- (streaks 1–7) in the slot blots were calculated (Materials and Methods) and plotted at the bottom.
Accumulation of G-strand overhangs in the est2- clones was due to the loss of Est2p
Even though we observed accumulation of unpaired G-strand in several independent est2- clones, it remains formally possible that this phenotype was due to unrelated genetic changes that occurred during the multiple steps required for mutant strain derivation. To definitively rule out this possibility, we re-integrated one wild-type copy of the EST2 gene at a disrupted locus, and analyzed the phenotypes of the resulting strain. Telomere length measurements through successive streaks indicate that the strain is capable of maintaining telomere lengths over many generations (data not shown). Importantly, the level of G-strand overhangs in the re-integrant is comparable with the wild-type strain and much lower than the est2- mutant from which the re-integrant was derived (Figure 8A and C). As expected, the overall telomeric content of the re-integrant was higher than the est2- mutant (Figure 8B). We conclude that the accumulation of G-strand overhangs in the est2- mutant must be due to the loss of Est2p rather than some other unintended genetic changes.
Figure 8 Analysis of G-strand overhangs in wild type, est2- and est2- with a re-integrated copy of EST2. (A) Genomic DNA was isolated from overnight cultures of a wild-type strain, an est2- mutant, and an est2- mutant with a re-integrated copy of EST2, and subjected to in-gel hybridization analysis using an oligonucleotide probe complementary to the G-strand of Candida telomeres. Lanes 1–3 and 4–6 represent duplicate assays of the same samples. (B) After probing the native DNA with strand-specific probes, the DNA in the gel was denatured and hybridized again with the same probe. (C) Relative ratios of single-stranded to double-stranded telomere signals for all three samples were calculated and plotted.
DISCUSSION
In this study, we investigated the total amount and length of telomeric DNA in Candida mutants that are missing individual components of the telomerase complex during extended passage. In addition, we analyzed the levels of single-stranded telomere overhangs in the same strains. Significant findings include (i) apparent observation of telomerase-deficient survivors, (ii) increased levels of telomere recombination in the est1- mutant, (iii) accumulation of G-strand overhangs in the est2- mutant, and (iv) detection of C-strand overhangs. We also compared three different methods for assessing telomere overhangs and found that they yielded qualitatively similar results.
Telomerase-deficient C.albicans survivors
In our earlier analysis of Candida telomerase mutants, we did not observe the emergence of survivors (32), most likely due to insufficient number of passages. Here we show that with extended subculturing, phenotypes that are reminiscent of S.cerevisiae survivors can indeed be observed in Candida mutants. Specifically, both the est2- and est3- strains suffered progressive telomere attrition, followed by the emergence of survivors with greatly elongated telomeres. Some interesting differences, however, can be noted between the S.cerevisiae and Candida survivors. In S.cerevisiae, survivors do not arise until the culture has suffered very significant growth defects. The appearance of survivors is correlated with the detection of longer than wild-type telomeres in the vast majority of cells and is accompanied by an increase in the growth rate of the culture (19,40–42). In C.albicans, the emergence of survivors does not coincide with a large change in the growth rate of the culture. Thus, either telomere attrition has little effect on C.albicans growth, or the activation of the recombination pathway occurs before the degree of telomere attrition that is necessary to induce large growth defects. Also in contrast to S.cerevisiae, the Candida survivors initially have telomeres that are comparable in lengths with the wild type. The formation and behavior of Candida survivors appear to resemble more closely those of another budding yeast Kluyveromyces lactis, where senescence can also be accompanied by relatively mild growth defects (43). Indeed, a recent study of K.lactis telomeres demonstrates that senescence in telomerase-deficient cells can be suppressed by the introduction of a single elongated telomere (44). Given the greater heterogeneity of Candida telomere length distribution, the possibility that a few long telomeres might be stochastically present in telomerase-negative mutants even after prolonged passage to suppress senescence must also be considered. One potentially important distinction between the current analysis and previous studies of other budding yeasts is that all of our work was performed using diploid cells, whereas almost all of the published work on S.cerevisiae and K.lactis was done using haploid yeast. It remains to be seen if any of the distinct features of Candida with regard to telomere maintenance can be attributed primarily to its diploid nature.
Heightened telomere recombination in the Candida est1- mutant
The telomere dynamics of the C.albicans est1- mutant differs substantially from that of the other mutants and its S.cerevisiae counterpart (40,45). Long telomeres can persist upon extended subculturing, and there are instances of sudden telomere loss and acquisition (Figure 2B). While instances of sudden telomere loss and acquisition were also observed in est2- and est3- mutants, they were not as frequent, and telomeres in these mutants never reached the level of heterogeneity seen in est1- before the emergence of survivors. These results suggest that in est1- recombinational elongation of relatively long telomeres occurs at somewhat higher frequencies than in other mutants. This hypothesis can explain the absence of severe telomere shortening in early generations, and the maintenance of telomere content at the wild-type levels in later generations. It is tempting to speculate that loss of Est1p may increase the accessibility of telomeres to recombination proteins. In other word, Est1p appears to ‘protect’ telomeres against recombination that otherwise can occur only on severely shortened telomeres (46). A protective function for Est1 at telomeres was also suggested by recent analyses of the human and Schizosaccharomyces pombe homologs (47,48).
If the dynamics of telomeres in est1- were due to heightened recombination, then it ought to be dependent upon recombination proteins. Several attempts were made to generate an est1- rad52- mutant without success. Because unlike S.cerevisiae, loss of Rad52p alone in Candida causes severe growth defects (49), it is unclear whether our failure was due to this growth defect or possibly a synthetic interaction between telomerase and RAD52 in Candida. Additional studies are underway to address this issue.
Prevention of aberrant C-strand resection by Candida Est2p
Perhaps the most interesting finding from our investigation of telomere overhangs is the substantial buildup of such overhangs in the est2- mutant. This buildup can be seen even in early generations of the mutant. While accumulation of unpaired G-strand has been reported previously for an S.cerevisiae est1- mutant, the accumulation was only observed after severe telomere attrition and is likely to be mechanistically distinct (50). Study of S.cerevisiae yku and cdc13 mutants demonstrates that abnormally long G-strand is created through resection of the C-strand, at least in part by the EXO1 nuclease (22,50,51). That this phenotype is observed with the loss of Est2p (but not Est1p or Est3p) in C.albicans suggests that Est2p may have a similar role in protecting telomeres against excessive nucleolytic degradation. The apparently increased sensitivity of short telomeres to C-strand degradation suggests that the protective function of Est2p may be more important for short telomeres. Why should the Est2p component of telomerase alone possess this unique activity? One speculative explanation can be proposed based on earlier findings in S.cerevisiae. In S.cerevisiae, Est2p is localized to telomeres throughout the cell cycle, at least in part through an indirect association with the Ku complex (i.e. Est2p binds telomerase RNA, which in turn binds Ku) (52–54). The Est2p–Ku interaction may thus enhance the ability of the entire complex to prevent nuclease attack. In contrast, there is no evidence that the association of Est1p and Est3p with telomeres is stable throughout the cell cycle (16). Because telomerase RNA is also presumed to be part of the protective complex, it is tempting to speculate that deleting the RNA gene may also cause accumulation of G-strand overhangs. Further studies will be necessary to determine the validity of this hypothesis.
C-strand overhangs in C.albicans
Presence of C-strand overhangs has only been described previously for human telomeres using the T-OLA assay (55), and not for any fungi. Our detection of such overhangs in C.albicans is therefore unexpected. Nevertheless, the biology of C.albicans (e.g. the obligate diploid genome) is sufficiently distinct from S.cerevisiae to make such differences in telomere structure far from implausible. The most obvious change in the Candida C-strand overhangs associated with telomerase mutations was a moderate decrease observed in the est2- samples. The functional significance of this finding is unclear. It is worth noting that because the wild-type samples yielded stronger signals for both single-stranded and double-stranded telomeres, if the hybridization signals are not strictly linear with respect to the levels of single and double-stranded DNA, then the quantitative differences we calculated may be more apparent than real. Additional studies are necessary to address this possibility. (On the other hand, our conclusion with respect to G-strand alterations in the est2- mutant is not subject to the same caveat, because the mutant samples consistently yielded weaker signals for double-stranded telomeres, but stronger signals for single-stranded telomeres.)
CONCLUSIONS
Altogether, our analysis of the telomere length dynamics and G-strand variations in the mutant strains support a role for both Candida Est1p and Est2p in telomere protection. In the case of est1-, loss of protection is mainly manifested as heightened recombination at telomeres. In the case of est2-, loss of protection is mainly manifested as accumulation of G-strand overhangs. That the phenotypic consequences are different for the two mutants implies that the precise mechanisms of protection must also be different.
SUPPLEMENTARY DATA
Supplementary Data are available at NAR Online.
ACKNOWLEDGEMENTS
We are grateful to Dr Shu-Chun Teng for constructive comments on the manuscript. We also thank Sunitha Singh and You-Chin Lin for help with the construction of mutant strains, Alain Dandjinou for help with in-gel hybridization, and Drs E. D'Ambrosio, S. Stewart and I. Ben-Porath for advice on T-OLA. This work was supported by a New Investigator Award in Pathogenic Mycology from Burroughs Wellcome Fund and a grant from NIH. Funding to pay the Open Access publication charges for this article was provided by the Irma Hirschl Monique Weill-Caulier Trust.
REFERENCES
Blackburn, E. (2001) Switching and signaling at the telomere Cell, 106, 661–673 .
Ferreira, M., Miller, K., Cooper, J. (2004) Indecent exposure: when telomeres become uncapped Mol. Cell, 13, 7–18 .
Cech, T. (2004) Beginning to understand the end of the chromosome Cell, 116, 273–279 .
Griffith, J.D., Comeau, L., Rosenfield, S., Stansel, R.M., Bianchi, A., Moss, H., de Lange, T. (1999) Mammalian telomeres end in a large duplex loop Cell, 97, 503–514 .
de Lange, T. (2002) Protection of mammalian telomeres Oncogene, 21, 532–540 .
Olovnikov, A. (1973) A theory of marginotomy. The incomplete copying of template margin in enzymic synthesis of polynucleotides and biological significance of the phenomenon J. Theor. Biol, . 41, 181–190 .
Lingner, J., Cooper, J., Cech, T. (1995) Telomerase and DNA end replication: no longer a lagging strand problem? Science, 269, 1533–1534 .
Greider, C.W. and Blackburn, E.H. (1985) Identification of a specific telomere terminal transferase activity in Tetrahymena extracts Cell, 43, 405–413 .
Nugent, C.I. and Lundblad, V. (1998) The telomerase reverse transcriptase: components and regulation Genes Dev, . 12, 1073–1085 .
Kelleher, C., Teixeira, M., Forstemann, K., Lingner, J. (2002) Telomerase: biochemical considerations for enzyme and substrate Trends Biochem. Sci, . 27, 572–579 .
Harrington, L. (2003) Biochemical aspects of telomerase function Cancer Lett, . 194, 139–154 .
Lundblad, V. (2003) Telomere replication: an Est fest Curr. Biol, . 13, R439–R441 .
Hughes, T.R., Evans, S.K., Weilbaecher, R.G., Lundblad, V. (2000) The Est3 protein is a subunit of yeast telomerase Curr. Biol, . 10, 809–812 .
Evans, S. and Lundblad, V. (2002) The Est1 subunit of Saccharomyces cerevisiae telomerase makes multiple contributions to telomere length maintenance Genetics, 162, 1101–1115 .
Pennock, E., Buckley, K., Lundblad, V. (2001) Cdc13 delivers separate complexes to the telomere for end protection and replication Cell, 104, 387–396 .
Taggart, A., Teng, S., Zakian, V. (2002) Est1p as a cell cycle-regulated activator of telomere-bound telomerase Science, 297, 1023–1026 .
Singh, S. and Lue, N. (2003) Ever shorter telomere 1(EST1)-dependent reverse transcription by Candida telomerase in vitro: evidence in support of an activating function Proc. Natl Acad. Sci. USA, 100, 5718–5723 .
Neumann, A. and Reddel, R. (2002) Telomere maintenance and cancer—look, no telomerase Nature Rev. Cancer, 2, 879–884 .
Lundblad, V. (2002) Telomere maintenance without telomerase Oncogene, 21, 522–531 .
Kass-Eisler, A. and Greider, C. (2000) Recombination in telomere-length maintenance Trends Biochem. Sci, . 25, 200–204 .
Garvik, B., Carson, M., Hartwell, L. (1995) Single-stranded DNA arising at telomeres in cdc13 mutants may constitute a specific signal for the RAD9 checkpoint Mol. Cell. Biol, . 15, 6128–6138 .
Maringele, L. and Lydall, D. (2002) EXO1-dependent single-stranded DNA at telomeres activates subsets of DNA damage and spindle checkpoint pathways in budding yeast yku70Delta mutants Genes Dev, . 16, 1919–1933 .
Yang, Q., Zheng, Y., Harris, C. (2005) POT1 and TRF2 cooperate to maintain telomeric integrity Mol. Cell. Biol, . 25, 1070–1080 .
Baumann, P. and Cech, T.R. (2001) Pot1, the putative telomere end-binding protein in fission yeast and humans Science, 292, 1171–1175 .
Bertuch, A. and Lundblad, V. (2003) Which end: dissecting Ku's function at telomeres and double-strand breaks Genes Dev, . 17, 2347–2350 .
Fisher, T. and Zakian, V. (2005) Ku: A multifunctional protein involved in telomere maintenance DNA Repair, 4, 1215–1226 .
Gravel, S., Larrivee, M., Labrecque, P., Wellinger, R.J. (1998) Yeast Ku as a regulator of chromosomal DNA end structure Science, 280, 741–744 .
Polotnianka, R.M., Li, J., Lustig, A.J. (1998) The yeast Ku heterodimer is essential for protection of the telomere against nucleolytic and recombinational activities Curr. Biol, . 8, 831–834 .
Chan, S. and Blackburn, E. (2003) Telomerase and ATM/Tel1p protect telomeres from nonhomologous end joining Mol. Cell, 11, 1379–1387 .
Kim, M., Xu, L., Blackburn, E. (2003) Catalytically active human telomerase mutants with allele-specific biological properties Exp. Cell Res, . 288, 277–287 .
McEachern, M.J. and Hicks, J.B. (1993) Unusually large telomeric repeats in the yeast Candida albicans Mol. Cell. Biol, . 13, 551–560 .
Singh, S.M., Steinberg-Neifach, O., Mian, I.S., Lue, N.F. (2002) Analysis of telomerase in Candida albicans: potential role in telomere end protection Eukaryot Cell, 1, 967–977 .
Wilson, R.B., Davis, D., Mitchell, A.P. (1999) Rapid hypothesis testing with Candida albicans through gene disruption with short homology regions J. Bacteriol, . 181, 1868–1874 .
Fonzi, W. and Irwin, M. (1993) Isogenic strain construction and gene mapping in Candida albicans Genetics, 134, 717–728 .
Hoffman, C. and Winston, F. (1987) A ten-minute DNA preparation from yeast efficiently releases autonomous plasmids for transformation of Escherichia coli Gene, 57, 267–272 .
Dionne, I. and Wellinger, R.J. (1996) Cell cycle-regulated generation of single-stranded G-rich DNA in the absence of telomerase Proc. Natl Acad. Sci. USA, 93, 13902–13907 .
Cimino-Reale, G., Pascale, E., Battiloro, E., Starace, G., Verna, R., D'Ambrosio, E. (2001) The length of telomeric G-rich strand 3'-overhang measured by oligonucleotide ligation assay Nucleic Acids Res, . 29, E35 .
Stewart, S., Ben-Porath, I., Carey, V., O'Connor, B., Hahn, W., Weinberg, R. (2003) Erosion of the telomeric single-strand overhang at replicative senescence Nature Genet, . 33, 492–496 .
Enloe, B., Diamond, A., Mitchell, A. (2000) A single-transformation gene function test in diploid Candida albicans J. Bacteriol, . 182, 5730–5736 .
Lundblad, V. and Blackburn, E.H. (1993) An alternative pathway for yeast telomere maintenance rescues est1-senescence Cell, 73, 347–360 .
Teng, S., Chang, J., McCowan, B., Zakian, V. (2000) Telomerase-independent lengthening of yeast telomeres occurs by an abrupt Rad50p-dependent, Rif-inhibited recombinational process Mol. Cell, 6, 947–952 .
Teng, S. and Zakian, V. (1999) Telomere-telomere recombination is an efficient bypass pathway for telomere maintenance in Saccharomyces cerevisiae Mol. Cell. Biol, . 19, 8083–8093 .
McEachern, M.J. and Blackburn, E.H. (1996) Cap-prevented recombination between terminal telomeric repeat arrays (telomere CPR) maintains telomeres in Kluyveromyces lactis lacking telomerase Genes Dev, . 10, 1822–1834 .
Topcu, Z., Nickles, K., Davis, C., McEachern, M. (2005) Abrupt disruption of capping and a single source for recombinationally elongated telomeres in Kluyveromyces lactis Proc. Natl Acad. Sci. USA, 102, 3348–3353 .
Lendvay, T.S., Morris, D.K., Sah, J., Balasubramanian, B., Lundblad, V. (1996) Senescence mutants of Saccharomyces cerevisiae with a defect in telomere replication identify three additional EST genes Genetics, 144, 1399–1412 .
McEachern, M. and Iyer, S. (2001) Short telomeres in yeast are highly recombinogenic Mol. Cell, 7, 695–704 .
Beernink, H., Miller, K., Deshpande, A., Bucher, P., Cooper, J. (2003) Telomere maintenance in fission yeast requires an est1 ortholog Curr. Biol, . 13, 575–580 .
Reichenbach, P., Hoss, M., Azzalin, C., Nabholz, M., Bucher, P., Lingner, J. (2003) A human homolog of yeast est1 associates with telomerase and uncaps chromosome ends when overexpressed Curr. Biol, . 13, 568–574 .
Ciudad, T., Andaluz, E., Steinberg-Neifach, O., Lue, N., Gow, N., Calderone, R., Larriba, G. (2004) Homologous recombination in Candida albicans: role of CaRad52p in DNA repair, integration of linear DNA fragments and telomere length Mol. Microbiol, . 53, 1177–1194 .
Hackett, J. and Greider, C. (2003) End resection initiates genomic instability in the absence of telomerase Mol. Cell. Biol, . 23, 8450–8461 .
Bertuch, A. and Lundblad, V. (2004) EXO1 contributes to telomere maintenance in both telomerase-proficient and telomerase-deficient Saccharomyces cerevisiae Genetics, 166, 1651–1659 .
Fisher, T., Taggart, A., Zakian, V. (2004) Cell cycle-dependent regulation of yeast telomerase by Ku Nature Struct. Mol. Biol, . 11, 1198–1205 .
Stellwagen, A., Haimberger, Z., Veatch, J., Gottschling, D. (2003) Ku interacts with telomerase RNA to promote telomere addition at native and broken chromosome ends Genes Dev, . 17, 2384–2395 .
Smith, C., Smith, D., De, R.J., Blackburn, E. (2003) Telomeric protein distributions and remodeling through the cell cycle in Saccharomyces cerevisiae Mol. Biol. Cell, 14, 556–570 .
Cimino-Reale, G., Pascale, E., Alvino, E., Starace, G., D'Ambrosio, E. (2003) Long telomeric C-rich 5'-tails in human replicating cells J. Biol. Chem, . 278, 2136–2140 .(Olga Steinberg-Neifach and Neal F. Lue*)