Dynamics of Telomeric DNA Turnover in Yeast

Corresponding author: Michael J. McEachern, Life Sciences Bldg., University of Georgia, Athens, GA 30602-7223., mjm{at}arches.uga.edu (E-mail)

Communicating editor: L. S. SYMINGTON

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Telomerase adds telomeric DNA repeats to telomeric termini using a sequence within its RNA subunit as a template. We characterized two mutations in the Kluyveromyces lactis telomerase RNA gene (TER1) template. Each initially produced normally regulated telomeres. One mutation, ter1-AA, had a cryptic defect in length regulation that was apparent only if the mutant gene was transformed into a TER1 deletion strain to permit extensive replacement of basal wild-type repeats with mutant repeats. This mutant differs from previously studied delayed elongation mutants in a number of properties. The second mutation, TER1-Bcl, which generates a BclI restriction site in newly synthesized telomeric repeats, was indistinguishable from wild type in all phenotypes assayed: cell growth, telomere length, and in vivo telomerase fidelity. TER1-Bcl cells demonstrated that the outer halves of the telomeric repeat tracts turn over within a few hundred cell divisions, while the innermost few repeats typically resisted turnover for at least 3000 cell divisions. Similarly deep but incomplete turnover was also observed in two other TER1 template mutants with highly elongated telomeres. These results indicate that most DNA turnover in functionally normal telomeres is due to gradual replicative sequence loss and additions by telomerase but that there are other processes that also contribute to turnover.

TELOMERES, the protein and DNA complexes at the ends of eukaryotic chromosomes, function to protect chromosome ends from terminal sequence losses and fusions and also appear to be important for meiotic chromosome segregation (reviewed in ZAKIAN 1995 ; LINGNER and CECH 1998 ; MCEACHERN et al. 2000B ). Telomeric DNA in the great majority of eukaryotic species is composed of tandem repeats of 5- to 26-bp sequence units. These repeats contain binding sites for proteins required for telomere function. Because DNA polymerases cannot fully replicate DNA ends, telomeres use specialized mechanisms to ensure their complete replication. In eukaryotic cells with telomeres composed of short tandem repeats, this involves the enzyme telomerase.

Telomerase is a reverse transcriptase composed of both protein and RNA subunits. Part of the RNA subunit serves as a template for synthesis of telomeric repeat units. Telomerase binds to 3' single strand tails of telomeric DNA, partly through base pairing interactions involving nucleotides in the templating domain of the telomerase RNA and partly through other interactions of telomerase with the 3' overhang of the telomeric DNA (LEE and BLACKBURN 1993 ; PRESCOTT and BLACKBURN 1997A ). Without telomerase activity, the telomeres of dividing yeast and mammalian cells gradually shorten. In yeast cells, this leads to telomere shortening, loss of telomere function, and eventual death of the great majority of cells (LUNDBLAD and SZOSTAK 1989 ; SINGER and GOTTSCHLING 1994 ; MCEACHERN and BLACKBURN 1995 ). Many human somatic cells naturally lack or have low levels of telomerase activity. In contrast, most human cancers express high levels of telomerase (KIM et al. 1994 ). This has led to the suggestion that telomere shortening in somatic cell lineages may be an adaptation that limits the proliferative capacity of developing cancers. All immortal human cell lines appear to have an active pathway of telomere maintenance, through either the presence of telomerase activity or, less frequently, the less understood telomerase-independent pathway termed alternative lengthening of telomeres (ALT; REDDEL et al. 1997 ). Yeast cells lacking telomerase that survive beyond the initial phase of growth senescence emerge with telomeres that have been elongated through mechanisms involving RAD52-dependent homologous recombination (LUNDBLAD and BLACKBURN 1993 ; MCEACHERN and BLACKBURN 1996 ; TENG and ZAKIAN 1999 ).

Telomerase RNA genes are particularly useful for studying telomere function because mutating the template region allows experimental alteration of the telomeric sequences. Typically, such mutated sequences cause net telomere lengthening or shortening (YU et al. 1990 ; SINGER and GOTTSCHLING 1994 ; MCEACHERN and BLACKBURN 1995 ; PRESCOTT and BLACKBURN 1997B , PRESCOTT and BLACKBURN 2000 ) and, in some cases, telomeric fusions, aberrant chromosome segregation, and nuclear division (YU et al. 1990 ; KIRK et al. 1997 ; SMITH and BLACKBURN 1999 ; MCEACHERN et al. 2000A ). Certain altered template sequences in Tetrahymena and the yeast Saccharomyces cerevisiae also cause misincorporation of bases as well as other defects in the enzymatic activities of telomerase in vitro and in vivo (YU and BLACKBURN 1991 ; GILLEY et al. 1995 ; GILLEY and BLACKBURN 1996 ; PRESCOTT and BLACKBURN 1997B ).

The majority of single or double template base substitutions in Kluyveromyces lactis led to an abnormal telomere length phenotype soon after replacement of the wild-type TER1 gene with a mutant TER1 gene carrying the mutant template (MCEACHERN and BLACKBURN 1995 ). For two such mutants (ter1-Acc and ter1-Bsi) the degree of immediate telomere lengthening correlated with the reduction in Rap1p binding affinity in vitro (KRAUSKOPF and BLACKBURN 1996 ). Rap1p regulates telomere length in S. cerevisiae and K. lactis (CONRAD et al. 1990 ; LUSTIG et al. 1990 ; KRAUSKOPF and BLACKBURN 1996 ). Two other K. lactis TER1 template mutants, ter1-Bgl and ter1-Kpn, which initially had short telomeres, produced greatly elongated telomeres only after a lag period of >100 cell divisions (MCEACHERN and BLACKBURN 1995 ), although the repeats made by ter1-Bgl and ter1-Kpn did not display weakened affinity to Rap1 in vitro (KRAUSKOPF and BLACKBURN 1996 ). During the lag period, the mutant telomeric repeats were confined to the outer tips of telomeres; only when most of the basal wild-type repeats had been replaced with mutant repeats did telomeres elongate and lose length control. Such extensive replacement could occur either through gradual turnover over prolonged cell passaging or by transforming the mutant TER1 gene directly into a TER1 deletion strain with very short telomeres.

A phenotypically silent telomerase TER1-Bcl template mutation has been a valuable tool for studying functions of both telomeres and telomerase (KRAUSKOPF and BLACKBURN 1998 ; ROY et al. 1998 ; SMITH and BLACKBURN 1999 ; TZFATI et al. 2000 ; MCEACHERN and IYER 2001 ). This class of mutant may not exist for most species as typical very short telomeric repeats must bind multiple different telomeric proteins and may have little or no tolerance for mutational change. The unusually long (25 bp) telomeric repeats of K. lactis (MCEACHERN and BLACKBURN 1994 ) are likely to tolerate mutations better than the short repeats in most other species.

Here we report a detailed examination of two K. lactis telomerase RNA template mutations, TER1-Bcl and ter1-AA, that each initially produce normal-length telomeres. The TER1-Bcl mutant generates a restriction site in newly synthesized telomeric repeats, allowing repeat incorporation to be monitored. We investigated long-term telomeric repeat turnover both in TER1-Bcl cells and in the immediate elongation mutants ter1-Acc and ter1-Bsi. We concluded that the TER1-Bcl mutant behaves indistinguishably from wild type. In both normal-length and elongated telomeres, telomeric turnover was found to involve all but the innermost few repeats. We propose that normal telomeric turnover occurs primarily by gradual replacement involving replicative sequence losses and telomerase action in small increments, but that other processes are also involved.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Strains:
All strains used in this study are derivatives of K. lactis 7B520 (ura3-1 his2-2 trp1; WRAY et al. 1987 ). The his2-2 allele is complemented by the S. cerevisiae HIS3 gene. Use of the term "wild type" refers to this parental strain. Construction of TER1 template base changes to make the TER1-Bcl, ter1-Acc, ter1-Bsi, and ter1-AA mutants was done through a plasmid loop-in, loop-out procedure (MCEACHERN and BLACKBURN 1995 ). Construction of TER1^ex mutants was done by transforming a ter1 deletion allele (lacking 300 bp of TER1, including the template) with pTER-BX-UA containing URA3 and either TER1-Bcl or ter1-AA. These plasmids integrated at the ter1- allele to form a plasmid loop-in with one functional telomerase RNA gene. Two slightly different types of TER1^ex mutants were utilized in this study. One type, shown as the ter1-AA^ex clones in Fig 2, were loop-in strains still containing the integrated URA3 vector flanked by the functional TER1-Bcl or ter1-AA allele on one side and the nonfunctional ter1- allele on the other side. The second type, shown as the ter1-AA^ex clones in Fig 3, were loop-out clones lacking the integrated URA3 plasmid vector and the nonfunctional ter1- allele.

View larger version (14K): In this window In a new window Download PPT slide   Figure 1. Diagram of the K. lactis TER1 template region and the position of alterations in four mutants. Shown in gray is the sequence of the 30-nucleotide (nt)-long template region of TER1. The arrows indicate the positions of the 5-nt direct repeats. The sequence shown in black is that of the 25-bp telomeric repeat unit of K. lactis. The solid bar indicates the position of the Rap1 protein binding site within telomeric sequences.

View larger version (116K): In this window In a new window Download PPT slide   Figure 2. Southern blot showing the telomeric EcoRI fragments of ter1-AA and TER1-Bcl mutants. (A) DNA prepared after long-term passaging of exWT and ex (see text for explanation) derivatives of ter1-AA and TER1-Bcl alongside DNA from a wild-type K. lactis (WT). Numbers on top indicate different clonal lineages. The exWT clones were grown for 120 passages and the ex clones had been grown for 90 passages. The probe used in A is a telomeric oligonucleotide Klac 19-7 at 38°. The faint bands above 4 kb in size are mostly from nontelomeric sequences due to the relatively low stringency hybridization conditions. One faint band, part of a doublet near 7 kb, is the TER1 EcoRI fragment. The relatively weak signal from the TER1exWT cells is due to the poor ability of the oligonucleotide probe used to hybridize to AA repeats. (B) exWT derivatives of ter1-AA and TER1-Bcl mutants after five streaks. Temporary elongation of certain telomeres is evident in ter1-AA clones, particularly clone 3. Size markers (in kilobases) are shown at sides of panels. The probe used is the oligonucleotide Klac 1-25 at 50°.

View larger version (79K): In this window In a new window Download PPT slide   Figure 3. Telomeric length changes in ter1-AAex cells. Southern blot showing the telomeric EcoRI fragments of ter1-AAex mutants. (A) Fluctuating telomere lengths in DNA prepared from 21 consecutive streaks of a ter1-AAex clone. DNA from a wild-type K. lactis (WT) is shown to the left. B and C show jumps in telomere lengths that occurred in other ter1-AAex clones. Numbers above lanes refer to the number of streaks cells have gone through from the point the mutant was generated. The bracketed parts of C show the same region of the same filter hybridized with two different probes. The probe used for the upper part of C is the AA repeat-specific oligonucleotide KLAA at 40°. The probe used for all other parts is the telomeric oligonucleotide Klac 9-22 at 40°. Size markers (in kilobases) are shown between panels.

Long-term passaging of mutant cells was carried out by serial streaking on rich media (YPD plates) at 30°. Strains were streaked twice weekly down to single cells, which grew into colonies. Each streak is estimated to represent 20–25 cell divisions. Transformations of K. lactis were performed using procedures and media identical to those used with S. cerevisiae.

K. lactis linear vectors and cloning telomeres:
Minichromosome vectors were created combining K. lactis CEN + ARS vectors pKL313 (HIS3) and pKL316 (URA3) (ROY et al. 1998 ) with a restriction fragment containing two telomeres oriented head to head that are separated by a URA3 gene. The two-telomere fragment was created by inserting a DpnI telomere fragment into the BamHI and BglII sites in the polylinker of pMH3, flanking the URA3 gene (HOLLINGSWORTH and BYERS 1989 ). An EcoRI fragment containing the two telomeres was then excised and inserted into the EcoRI site of pKL313 and pKL316. The DpnI telomere fragment was derived from a cloned K. lactis telomere (MCEACHERN and BLACKBURN 1994 ) containing 12 telomeric repeats. One DpnI site was provided by a polylinker BamHI site next to the end of the telomere and the second DpnI site was provided by a BglII-linked XbaI site present 120 bp internal from the start of the telomeric repeats. The resulting linear vectors were designated pHISLIN1, pHISLIN2, pURALIN1, and pURALIN2 depending upon the marker present and the orientation of the two-telomere + URA3 insert. In pHISLIN1 and pURALIN1, the URA3 gene between the two telomeres is transcribed toward the KpnI site in the polylinker.

To clone mutant telomeres, two techniques were employed. In one, TER1-Bcl K. lactis cells were transformed with pHISLIN1 after cleavage with BamHI and XhoI to linearize the plasmid and excise the URA3 gene separating the two telomeres. DNA from a transformant was then isolated, cleaved with SmaI, which cleaves off one telomere, treated with T4 polymerase to blunt ends, and circularized with ligase. Plasmid clones were then recovered by transformation into Escherichia coli.

In the second telomere cloning protocol, a plasmid containing 0.6 kb of K. lactis subtelomeric DNA was transformed into long-term TER1-Bcl cells where it integrated next to a telomere. This plasmid, pSubtelHis, was generated by cloning an EcoRI-XbaI fragment (located in K. lactis 120 bp from the innermost telomeric repeats and homologous to 11 of the 12 telomeres) into pRS423 (CHRISTIANSON et al. 1992 ) and allows for the recovery of transformants containing the plasmid integrated into any one of most of the telomeres in the cell. The transforming plasmid either was in circular form or was first linearized within its subtelomeric sequence using NcoI. DNA from transformants was isolated, cut with XhoI (to cleave the plasmid + telomere away from other chromosomal DNA) treated with T4 polymerase to ensure blunt ends, ligated, and then transformed into E. coli.

The permutation of the K. lactis telomeric repeat present at the subtelomere/telomere junction is as indicated for the strand running 5' to 3' toward the terminus: subtelomere, GGTGTACGGATTTGATT*AGGTATGT. The T residue marked with an asterisk denotes the position of the Bcl mutation. The number of copies of the sequence shown is what was counted as basal wild-type repeats.

Hybridizations:
Southern blotting was performed using Hybond N⁺ membrane (Amersham, Pharmacia Biotech, Piscataway, NJ). All hybridizations were carried out in Na₂HPO₄ and SDS (CHURCH and GILBERT 1984 ). Oligonucleotides used as hybridization probes are as follows: Klac 9-22 (GATTAGGTATGTGG), Klac 17-8 (ATGTGGTGTACGGATTT), Klac 1-25 (ACGGATTTGATTAGGTATGTGGTGT), and KLAA (GTATGTAATGTACG). Underlined nucleotides of KLAA show the mutant-specific changes. Other oligonucleotides are perfect matches to wild-type K. lactis telomeric repeats. All washes for hybridizations were carried out at the same temperature as hybridizations in Na₂HPO₄ and SDS (200 mM Na⁺ and 2% SDS).

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Characterization of two telomerase RNA template mutants initially displaying normal telomere length:
Fig 1 shows the sequences of the ter1-AA and TER1-Bcl mutations. The expected sequence change in the mutant AA repeats is located within the conserved binding site of the Rap1 protein (COHN et al. 1998 ), and a duplex telomeric repeat DNA oligonucleotide containing the AA base changes binds Rap1 with reduced affinity in vitro (KRAUSKOPF and BLACKBURN 1996 ). In contrast, the sequence change predicted to be synthesized in Bcl telomeric repeats lies in a region of the repeat that is not conserved among telomeric sequences from a number of related yeasts (COHN et al. 1998 ) and is outside the predicted Rap1 binding site.

The ter1-AA and TER1-Bcl mutants retained normal telomere lengths soon after replacement of the wild-type TER1 gene (Fig 2). To test whether either mutant underwent delayed telomere elongation, the mutant strains were extensively passaged by serial colony streaking and telomere lengths were monitored periodically. Each mutant was constructed in a genetic background that initially contained a wild-type TER1 gene and normal-length telomeres, by first integrating ("looping in") a plasmid containing the TER1 template mutant gene adjacent to the native TER1 locus, followed by plasmid loop-out and screening for retention of the mutant-template ter1 allele. Telomeric profiles of clonal lineages of mutants generated in this way (referred to as "exWT") are shown in Fig 2A for four ter1-AA^exWT and two TER1-Bcl^exWT clones after 120 consecutive streaks; each streak represented 20–25 cell divisions. Even after this very long period of growth, the telomeres in TER1-Bcl clones remained indistinguishable from wild type, and ter1-AA clones showed only slight net elongation. Thus, in this test, neither mutant behaved like the previously studied delayed elongation ter1-Bgl or ter1-Kpn mutants, which in comparable experiments eventually exhibited extreme telomere elongation. However, during this prolonged passaging additional subtle changes in length regulation appeared in the ter1-AA^exWT telomeres. After only 5 streaks, some individual telomeres, particularly in ter1-AA^exWT clone 3, were appreciably longer than normal (Fig 2B). After 5–10 more streaks, these elongated telomeres had gradually shortened back to normal length. Other sporadic instances of modest temporary telomere elongation occurred over the 120 consecutive streaks of the ter1-AA^exWT clones. The fluctuations in telomere length, such as those seen in Fig 2B and Fig 3, were distinctly greater than those in either wild-type or TER1-Bcl^exWT cells and are suggestive of subtly altered telomere function.

In a second type of test of telomere function, the ter1-AA and TER1-Bcl mutant genes were each transformed into a senescing TER1 deletion (ter1-) strain at the point when telomeres were considerably shortened. We refer to these TER1 deletion transformed cells as ter1-AA^ex and TER1-Bcl^ex. These "ex" strains can exist in two forms, a plasmid loop-in form and a plasmid loop-out form, which appear indistinguishable with regard to telomere function (see MATERIALS AND METHODS for details). It was shown previously that this procedure could shorten the lag before telomeres elongate in the ter1-Kpn mutant, apparently because much of the inner tract of basal wild-type repeats was eliminated (MCEACHERN and BLACKBURN 1995 ). For each mutant (ter1-AA^ex and TER1-Bcl^ex), two independent transformants were examined. In both TER1-Bcl^ex clones the telomere length was wild type and remained so even after 10 additional streaks (data not shown). In contrast, while the telomeres in the ter1-AA^ex clones were initially slightly longer than wild type (M. MCEACHERN, T. FULTON and E. BLACKBURN, data not shown), eventually telomere lengths grew to many times normal length and exhibited the smeared appearance on Southern blots characteristic of telomere length deregulation in other K. lactis ter1 mutants (MCEACHERN and BLACKBURN 1995 ; SMITH and BLACKBURN 1999 ; Fig 2A). The telomeres in ter1-AA^ex clones gradually lengthened over at least 80 streaks (an estimated 1600–2000 cell divisions; data not shown). Unlike some TER1 template mutations that produce extreme telomere lengthening (MCEACHERN and BLACKBURN 1995 ; MCEACHERN and BLACKBURN 1997 ; SMITH and BLACKBURN 1999 ), the ter1-AA^ex and TER1-Bcl^ex mutants at no stage exhibited abnormal colonies or cell morphology. Also, neither mutant exhibited the fusions between telomeric ends that have been observed in certain other TER1 template mutants (MCEACHERN et al. 2000A ).

Four additional clonal ter1-AA^ex lineages were analyzed over at least 20 consecutive streaks, beginning shortly after the creation of the strains. Telomere elongation was not always continuous or gradual. In one clonal ter1-AA^ex lineage followed for 21 consecutive streaks, overall net telomere length did not increase (average telomere lengths remained roughly double the wild-type length; see Fig 3A), but did fluctuate considerably. Some fluctuations were gradual and occurred over several streaks, with some telomeres gradually shortening as others concurrently gradually lengthened. In other clonal lineages of ter1-AA^ex cells, some telomere lengths showed large and rapid increases over shorter growth periods (Fig 3B and Fig C). In one lineage, within 1–2 streaks (20–50 cell divisions) all telomeres in the cell grew by 1.5 kb to over 10 kb (Fig 3B, compare lanes 2 and 3). Later (lanes 4 and 5), the telomeric fragments became relatively stably and uniformly elongated at greater than 5 kb. In another clonal lineage (Fig 3C), several telomeres (migrating at 1.2–1.8 kb) gradually lengthened by 2 bp per cell division. In contrast, the largest telomeric fragment (initially at 4 kb) showed two large jumps in size (between lanes 14 and 15 and lanes 18 and 19), accompanied by increases in the hybridization to an AA repeat-specific probe but not to a wild-type repeat-specific probe (compare Fig 3C, top, to Fig 3C, bottom). No gradual lengthening was apparent after each jump (see bracketed region in Fig 3C). The largest jump (1 kb) minimally represents a mean lengthening rate of 40 bp per cell division. These results are consistent with bursts of elongation by the mutant telomerase. Although we have not tested ter1-AA, the more severe telomere elongation phenotypes in both immediate and delayed mutants have been shown to be RAD52 independent (D. UNDERWOOD and M. MCEACHERN, unpublished data). This indicates that telomerase and not recombination is responsible for the telomere elongation in both types of mutants. In summary, AA mutant repeats can cause defects in telomere length regulation that can vary stochastically between telomeres in ways that can persist for many cell divisions.

Repeats synthesized by the TER1-Bcl telomerase contain only the anticipated sequence change:
Mutating the template region of a telomerase RNA gene has been shown to produce the equivalent sequence changes in the telomeric repeats synthesized by the mutant telomerase in ciliated protozoans, yeasts, and mammalian cells (YU et al. 1990 ; SINGER and GOTTSCHLING 1994 ; MCEACHERN and BLACKBURN 1995 ; MARUSIC et al. 1997 ). However, in vitro and in vivo evidence has also shown that altering the telomerase template sequence can greatly lower the fidelity of the enzyme, leading in some cases to most newly synthesized repeats having nonpredicted sequences (YU and BLACKBURN 1991 ; GILLEY et al. 1995 ; MCCORMICK-GRAHAM et al. 1997 ; D. H. UNDERWOOD and M. J. MCEACHERN, unpublished data). Therefore, we sequenced the telomeric repeats synthesized by the TER1-Bcl mutant by cloned telomeres from mutant cells. A linear minichromosome vector capable of replicating in K. lactis (see MATERIALS AND METHODS) was introduced into TER1-Bcl cells so that repeats synthesized by the mutant TER1-Bcl telomerase became added onto the ends of the telomeres of the vector. Total DNA isolated from these yeast cells was treated with SmaI to cleave off one telomeric end and then treated with T4 polymerase to create a blunt end on the other telomere. The plasmid DNA was then circularized with ligase and used to transform E. coli. The two Bcl telomeres cloned by this method both contained the expected wild-type repeats immediately internal to the predicted mutant repeats. One clone terminated with seven, and the other with four, tandem Bcl repeats, all with the expected single base substitution (see Fig 1A) and no additional alterations. This same result was also observed for five additional telomeres cloned from long-term TER1-Bcl cells (see below). Thus, as also reported in another study (TZFATI et al. 2000 ), the Bcl telomerase faithfully synthesized repeats with the predicted mutation. These results, together with the unchanged telomere behavior described above, showed that the TER1-Bcl telomerase and Bcl mutant telomeric DNA are functionally equivalent to wild type.

Long-term turnover of telomeric repeats in TER1-Bcl cells:
The terminal regions of telomeric DNA are highly dynamic, through both elongation by telomerase and shortening due to incomplete replication, and possibly other processes. If telomeric DNA turnover is mediated solely by telomerase, replacement by Bcl repeats is expected to occur progressively from the ends inward. In contrast, recombinational events within telomeric tracts might be expected to cause Bcl repeats to be recombined from outer to inner portions of the telomeric repeat tract. The phenotypically silent TER1-Bcl template mutation allowed us to study the dynamics of telomeric repeat turnover in functionally wild-type cells, by following the incorporation of BclI restriction sites in telomeric repeats. Double digestion with EcoRI plus BclI generates telomeric fragments with the mutant Bcl repeats at the telomeric termini removed (Fig 4C). Telomeres in two clonal lineages of a TER1-Bcl mutant were analyzed over 130 consecutive colony streaks (a year of continuous growth, or an estimated 2600–3250 cell divisions; Fig 4A and Fig B). The results were very similar for both lineages. As with other TER1 template mutants (MCEACHERN and BLACKBURN 1995 ), Bcl telomeric repeats were present at the first time point after isolation of the mutant strain and, by the 10th streak, had penetrated further into the telomeric repeat arrays. Thereafter, penetration was slower, but detectable as late as the 120th streak. After 130 streaks, an average of 300 bp of mutant Bcl repeats (12 repeats) were present on each telomere. Size analyses of restriction digests that produced smaller telomeric fragments (our unpublished data) and measurement of hybridization intensities from the filter shown in Fig 4 indicated that after 130 streaks, an average of 3–4 basal wild-type repeats remained. Consistent with this, seven telomeres cloned from TER1-Bcl cells at 120 streaks contained from one to six full-length basal wild-type repeats with all the more distal repeats containing the Bcl mutation (Fig 5A). Thus, on average, 75–80% of the telomeric repeat array was replaced during the prolonged passaging. This indicates either that telomeres are occasionally truncated to very short lengths or that Bcl repeats are moved internally by a process other than by telomerase addition to ends. One cloned telomere, LTBcl6E2-31, contained only about five complete repeats. Whether this clone represents a rare naturally truncated telomere or a cloning artifact is unclear. A summary of the patterns of turnover in TER1-Bcl cells is shown in Fig 5B.

View larger version (72K): In this window In a new window Download PPT slide   Figure 4. Southern blot examining the extent of penetration of Bcl repeats in a time course of growth of two lineages of TER1-BclexWT cells (A–B). Controls in the left three lanes and far right lane show EcoRI digests of wild-type (WT) or TER1-BclexWT DNA, showing telomere fragments with all telomeric repeats present. Other lanes (marked by bar) show double digests (EcoRI + BclI) of DNA prepared from the TER1-BclexWT cells. Numbers above lanes indicate the number of streaks since the isolation of the mutant. Arrow indicates one instance where there was an increase in the size of a BclI-cleaved telomeric fragment. Note that one telomere, at 2.4 kb in the EcoRI digests, contains a subtelomeric BclI site and is cut to 2 kb in size in the absence of Bcl telomeric repeats. This band runs as a doublet with another band. Probe used is the K. lactis telomeric oligonucleotide Klac 17-8 at 47°. Size markers are given on the right. (C) A map of the 12 K. lactis telomeres with the position of the EcoRI site nearest the telomere indicated (RI). Positions of a BclI site and a BsiWI site within individual telomeric EcoRI fragments are also indicated. The diagram illustrates the shortening of telomeric fragments in EcoRI + BclI digests that results from incorporation of Bcl repeats at telomeric termini in TER1-Bcl cells.

View larger version (32K): In this window In a new window Download PPT slide   Figure 5. Turnover in long-term TER1-Bcl cells. (A) Sequence of seven telomeres cloned from long-term (120 streaks) TER1-Bcl mutants. Two clones from lineage 6E1 (Fig 4A) and five from lineage 6E2 (Fig 4B) are shown. Sequences are shown 5' to 3' extending to the end of the telomeric sequence homology. Numbers of complete wild-type (WT) and Bcl repeats present in each cloned telomere are indicated. Sequences of each type of repeat are indicated above, with the Bcl mutation shown underlined. (B) Summary of telomeric turnover in the phenotypically wild-type TER1-Bcl mutant. Shortly after the wild-type TER1 gene has been replaced by TER1-Bcl, only the terminal few repeats (of 10–20 total) are Bcl repeats. After a few hundred cell divisions have passed, approximately the outer halves of the telomeres have been replaced. After a few thousand cell divisions, 80% of the telomeres have been turned over. Open boxes indicate wild-type repeats and shaded boxes indicate Bcl repeats.

Two observations on the long-term TER1-Bcl lineages are inconsistent with a model in which sequence loss from ends and sequence addition by telomerase are the only mechanisms affecting telomeric structure in the clones we examined. First, in one instance an EcoRI + BclIcleaved telomeric restriction fragment increased in size between the 50th and 60th streak (arrow in Fig 4B). As this fragment has had all mutant repeats removed by cleavage with BclI, it must have acquired either additional basal wild-type repeats or additional subtelomeric sequences. Second, this same telomeric band initially existed in the clonal lineage as a doublet representing two K. lactis telomeres, but became a singlet between streaks 10 and 20 in one lineage (Fig 4A) and 20 and 30 in the other (Fig 4B and phosphorimaging data not shown). This suggests that one of the telomeres of the doublet underwent a telomeric or subtelomeric gene conversion, events common in a short telomere mutant but rare in wild-type cells (MCEACHERN and IYER 2001 ).

Turnover of telomeric repeats in TER1 mutants with elongated telomeres:
Similar analyses of telomeres over long-term passaging by single-colony streaks were done on mutants with greatly elongated telomeres. In the two previously characterized TER1 template mutants, ter1-Bsi and ter1-Acc, telomeres rapidly elongate to many kilobases, beginning immediately after replacement of the wild-type TER1 gene despite the presence of nearly full-length arrays of basal wild-type repeats (MCEACHERN and BLACKBURN 1995 ).

Repeat turnover in the ter1-Bsi and ter1-Acc mutants was monitored by restriction digestion, as each mutant telomerase generates a novel restriction site (BsiWI and AccI, respectively) in newly made telomeric repeats. In a clonal ter1-Bsi lineage followed over 60 streaks, telomeres remained at several kilobases, but turnover of the basal wild-type repeats continued for up to 50 streaks, at a rate similar to that observed with TER1-Bcl strains (Fig 4C and Fig 6 and data not shown). Similar results were obtained with the ter1-Acc mutant (our unpublished data), which has even greater telomere elongation than ter1-Bsi (MCEACHERN and BLACKBURN 1995 ). We conclude that both functionally wild-type and highly elongated telomeres undergo similar patterns of turnover that affect all but the innermost few telomeric repeats.

View larger version (101K): In this window In a new window Download PPT slide   Figure 6. Southern blot examining the extent of penetration of mutant repeats in a time course of growth of ter1-Bsi cells. On the left side are EcoRI digests of DNAs from wild-type (WT) and ter1-Bsi cells from each of multiple time points (indicated by numbers above lanes). These lanes provide an indication of total telomeric fragment lengths. The right side of the figure shows EcoRI + BsiWI digests of the DNAs from ter1-Bsi cells that show telomeric fragment lengths after removal of all Bsi repeats. Note that the largest telomere in the EcoRI digests contains a subtelomeric BsiWI site and is present in the double digests as the fastest migrating band below 1 kb in size (see map of Fig 4C). The probe used is a K. lactis telomeric oligonucleotide, Klac 1-25 at 50°. Size markers are shown on both sides.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

The ter1-AA mutation causes a cryptic defect in telomere length regulation:
The detailed examination reported here of the K. lactis TER1-Bcl and ter1-AA mutants, each of which initially produces normal-length telomeres, has uncovered novel features of telomere dynamics. The ter1-AA mutant is similar in some respects to the previously described delayed elongation TER1 template mutants (MCEACHERN and BLACKBURN 1995 ). In those mutants, while telomeres eventually grew very long, they did so only after a protracted latent period during which telomeres were stable and short. The latent period required the presence of internal wild-type repeats to maintain length regulation; only after most wild-type repeats had been replaced by mutant repeats did abrupt lengthening ensue. Similarly, ter1-AA mutants generated by transforming a strain initially containing wild-type TER1 (ter1-AA^exWT) maintained near-normal telomere length even after thousands of cell divisions. However, transforming the ter1-AA gene into senescing TER1 deletion cells with short telomeres (ter1-AA^ex) caused considerable telomere lengthening. Because ter1-AA^exWT telomeres did not lengthen even after thousands of cell divisions, we term ter1-AA a cryptic elongation mutant rather than a delayed elongation mutant. These results with ter1-AA highlight the importance of the continued presence of sufficient basal wild-type repeats in telomeres in the compensation for defective telomere function caused by a TER1 template mutation. This compensation is long lasting in ter1-AA^exWT cells, apparently because the innermost few wild-type repeats are highly resistant to turnover.

One hypothesis for why telomeres elongate in ter1-AA^ex but not ter1-AA^exWT cells is that in ter1-AA^ex cells the elongated telomeres retain fewer basal wild-type telomeric repeats than those of even very long-term ter1-AA^exWT cells. Alternatively, the physiology of a senescing TER1 deletion cell could affect telomere function in some ways that allows generation of elongated telomeres by a newly transformed ter1-AA gene, in spite of the presence of a number of basal wild-type repeats that, in a ter1-AA^exWT cell, would be sufficient to block elongation. It is also conceivable that elongation of a small number of telomeres containing very few wild-type repeats somehow induces the elongation of other telomeres containing greater numbers of basal wild-type repeats. This latter possibility could help explain the sudden extensive elongation occurring in some clonal lineages of ter1-AA^ex cells, such as the example shown in Fig 3B.

Four distinct differences between ter1-AA and the previously identified ter1-Bgl and ter1-Kpn delayed lengthening mutants (MCEACHERN and BLACKBURN 1995 ) suggest that ter1-AA represents a distinct type of allele. First, ter1-AA cells initially produced normal-length telomeres, while ter1-Bgl and ter1-Kpn mutants initially produced shorter-than-normal telomeres. Second, the latent period of ter1-AA cells before telomere lengthening (>3000 cell divisions), if finite, was minimally several times longer than that of ter1-Bgl and ter1-Kpn mutants. Third, the telomere lengthening in ter1-AA cells was typically more gradual and less extensive than that seen in the other mutants. Finally, the altered telomeric repeats specified by the ter1-AA telomerase have a moderately reduced in vitro binding affinity for Rap1 protein, while the measured in vitro affinity of the mutated repeats of the ter1-Bgl and ter1-Kpn mutants for Rap1 was normal or slightly elevated (KRAUSKOPF and BLACKBURN 1996 ). This is consistent with the fact that the base changes in ter1-AA lie within the consensus Rap1p binding site while those of ter1-Bgl and ter1-Kpn lie outside it (Fig 1). It was shown previously that the degree of immediate telomere elongation caused by the ter1-Acc and ter1-Bsi mutations in vivo correlated with the reduction in binding affinities of Acc and Bsi repeats to Rap1p in vitro (KRAUSKOPF and BLACKBURN 1996 ). These same studies also found that AA telomeric repeats had a more moderate reduction in the ability to bind Rap1p. The C terminus of Rap1 protein negatively regulates telomere length in both S. cerevisiae and K. lactis (CONRAD et al. 1990 ; LUSTIG et al. 1990 ; KRAUSKOPF and BLACKBURN 1996 ), in part through "counting" the number of Rap1p C termini present at any given telomere (MARCAND et al. 1997 ). We propose that the cryptic length regulation defect of ter1-AA cells is due to the weakened interaction of Rap1p with AA repeats and that the presence of a minimum number of basal wild-type repeats can prevent telomere elongation from occurring. The properties of ter1-AA mutants support the possibility that the delayed elongation phenotypes of ter1-Bgl and ter1-Kpn stem from an unknown defect in the interaction between the mutant repeats and Rap1p. Possibilities for this defect include effects on the known DNA bending properties of Rap1p (VIGNAIS and SENTENAC 1989 ; MULLER et al. 1994 ) or other alteration(s) in the higher-order protein-DNA telomeric complex.

The TER1-Bcl allele behaves indistinguishably from wild type:
In contrast to the ter1-AA mutant, the TER1-Bcl mutant was indistinguishable from wild type by several stringent criteria: unchanged telomere length even in very long-term TER1-Bcl^exWT and TER1-Bcl^ex clones, no abnormal colony or cellular phenotypes, no detectable telomeric fusions, and apparently normal telomerase fidelity. In TER1-Bcl cells constructed to contain one telomere composed entirely of Bcl repeats (UNDERWOOD and MCEACHERN 2001 ), this telomere also remains wild type in length (our unpublished data), further reinforcing the conclusion that Bcl repeats are completely normal in their length regulation.

A crucial role of telomeres in cells is to prevent chromosome ends from eliciting responses from enzymes involved in the repair of broken DNA ends. Mutational alteration of telomeric sequences might therefore be expected to cause high rates of recombination or end-to-end ligations, two major pathways by which cells are known to repair DNA double-strand breaks. A number of TER1 template mutants, but not TER1-Bcl mutants, are highly prone to one or both of these processes (MCEACHERN et al. 2000A ; MCEACHERN and IYER 2001 ). The fact that telomeres in TER1-Bcl mutants cause no abnormal growth phenotype, remain unfused, and undergo normal levels of recombination (MCEACHERN and IYER 2001 ) provides strong evidence that telomeres with Bcl repeats provide a wild-type degree of protection for chromosome ends.

Processes besides replicative sequence loss and telomerase-mediated sequence addition at telomeres:
The results reported here, together with other short-term studies on altered telomerase template mutants (YU et al. 1990 ; SINGER and GOTTSCHLING 1994 ; MCEACHERN and BLACKBURN 1995 ; PRESCOTT and BLACKBURN 1997B ), support the hypothesis that short-term turnover of telomeric repeats is attributable primarily to the combined actions of two processes: sequence addition by telomerase and gradual terminal sequence loss attributable to the failure of DNA polymerases to fully replicate ends. These processes can account for the incorporation of mutant repeats only at telomeric termini soon after replacement of wild-type TER1 with the phenotypically wild-type TER1-Bcl. The relative stability of the more internal part of telomeres was also inferred from work in both S. cerevisiae (WANG and ZAKIAN 1990 ; FORSTEMANN et al. 2000 ) and Plasmodium falciparum (PONZI et al. 1992 ).

We also found evidence for other processes that contribute to telomeric turnover. In wild-type telomeres and also in the very long and deregulated telomeres of ter1-Bsi and ter1-Acc cells, turnover typically extends into all but the innermost few repeats. During the very protracted growth of TER1-Bcl cells, turnover of the original wild-type repeats eventually penetrated into all but the innermost (most basal) 1–5 repeats, i.e., 125–225 bp further in than the observable lower limit (10 repeats) of telomere size (see Fig 4). This result cannot be explained solely by the combination of gradual sequence loss and addition by telomerase. If such turnover were the result solely of incomplete replication at DNA ends, failure to elongate the shortest telomeres by telomerase would have to continue for up to 45 consecutive cell divisions, based upon the observed terminal sequence loss rate of 5 bp per cell division in K. lactis cells lacking telomerase (MCEACHERN and BLACKBURN 1995 , MCEACHERN and BLACKBURN 1996 ). Given that short telomeres are the ones most likely to be elongated by telomerase (MARCAND et al. 1999 ), it is likely that turnover of the more internal repeats in TER1-Bcl cells results from some other, albeit infrequent, process acting at telomeres.

This deep turnover into telomeres must involve both a mechanism for loss of large amounts of telomeric sequence and a mechanism to preserve the sequence of the innermost few repeats. Sizable abrupt shortening of telomeres has been reported previously. During macronuclear development in the ciliated protozoan Euplotes crassus, new telomeres are shortened by 50 bp prior to DNA replication occurring, indicating that nucleolytic cleavage must occur (VERMEESCH et al. 1993 ). Dramatic telomere shortening also is observed in S. cerevisiae cells constructed to contain some normal-length and some very long telomeres. In this case, long telomeres were sometimes abruptly shortened to wild-type length, a process called telomere rapid deletion (TRD; LI and LUSTIG 1996 ). Similarly, telomeres in human cell lines, trypanosomes, and some K. lactis mutants with elongated telomeres sometimes undergo deletions too large to be accounted for by gradual replicative sequence loss (PAYS et al. 1985 ; MURNANE et al. 1994 ; MCEACHERN and BLACKBURN 1995 ; STRAHL and BLACKBURN 1996 ). Multiple mechanisms, including recombination, replicational slippage, and nucleolytic cleavage could potentially produce large deletions of telomeric sequence.

In our mutants, the internal repeats' resistance to turnover could simply be the result of biases in the mechanism generating telomeric deletion. However, it is necessary to explain how long-telomere ter1-Bsi and ter1-Acc mutants can replace kilobases of telomeric sequence yet specifically retain only their innermost few repeats as wildtype sequences. The mechanism recently suggested to account for TRD could potentially account for our results. It was proposed that the 3' end of a telomere strand invades into a more internal part of the same telomere and promotes recombinational deletion of the intervening telomeric repeats (BUCHOLC et al. 2001 ). However, it is not clear whether such a mechanism could be biased to preserve the innermost few repeats.

Another factor that could contribute to the retention of basal wild-type repeats relates to the likelihood that two distinct mechanisms, telomerase and recombinational repair, can repair shortened telomeres. Which mechanism extended a given truncated telomere likely depends upon the number of telomeric repeats. Telomeres retaining enough repeats to efficiently bind proteins required for telomerase function, including Cdc13p and Est1p (NUGENT et al. 1996 ; VIRTA-PEARLMAN et al. 1996 ), could be readily relengthened by telomerase. In contrast, any telomeres shortened to a greater extent might be less able to bind telomerase and would thus become subject to recombinational repair, the pathway favored in yeast for processing DNA ends lacking telomeric repeats. Also, if the telomere shortening that drives deep turnover occurs as the result of TRD or other recombination events, a newly generated truncated telomere might be better poised to undergo recombinational repair rather than bind telomerase. Some data support the possibility of recombinational repair contributing to telomeric turnover. K. lactis telomeres significantly shorter than normal, but still retaining several telomeric repeats, are known to experience greatly elevated gene conversion (MCEACHERN and IYER 2001 ). Recombinational repair is also known to be highly efficient in yeast, capable of repairing up to 99% of chromosomes with a double-strand break through gene conversion (INBAR and KUPIEC 1999 ). Severely truncated telomeres, as mimics of double-strand breaks, would presumably be repaired with similar efficiency. Because at least 11 of the 12 K. lactis telomeres share substantial subtelomeric homology, a very short or missing telomere could be readily regenerated using a telomere from another chromosome as a template in a gene conversion. Such conversion, initiating as a replication fork that extends to the end of the chromosome has been called break-induced replication or break copy duplication (MALKOVA et al. 1996 ; MORROW et al. 1997 ; BOSCO and HABER 1998 ). Thus, any greatly shortened telomere could be regenerated by copying another telomere rather than by telomerase, making the innermost wild-type repeats appear resistant to being turned over. Use of another telomere as a template in a gene conversion could explain both the increase in size of the telomeric EcoRI + BclI fragment we observed during passaging of TER1-Bcl cells (arrow in Fig 4B) and the apparent loss of one telomere in the doublet of 2-kb EcoRI fragments (Fig 4B).

While the most basal few repeats of K. lactis telomeres are generally resistant to turnover even after 3000 cell divisions, they are unlikely to be stable over an evolutionary time scale. The repeats in basal regions of all the cloned telomeres analyzed to date from a variety of yeast species with different telomeric repeat sequences do not differ in sequence from the rest of the telomere (MCEACHERN and BLACKBURN 1994 ; COHN et al. 1998 ). Significantly, this has been shown for both K. lactis and its close relative Candida pseudotropicalis, which has telomeric repeats that differ from those of K. lactis at a single position (MCEACHERN and BLACKBURN 1994 ; COHN et al. 1998 ). This suggests that over the very long term, recombination eventually homogenizes all parts of yeast telomeres after the occurrence of a telomerase RNA template mutation. Without exchanges between the inner and outer parts of telomeric repeat arrays, the innermost repeats would be expected to accumulate mutations. Interestingly, in the ciliate Tetrahymena thermophila, the innermost tracts of the telomeres of the germline nucleus consist of tandem arrays of a homogeneous repeat variant, TTTGGGG, located next to a shared region of subtelomeric homology (KIRK and BLACKBURN 1995 ). This repeat variant differs from the homogeneous TTG GGG repeat sequences known to be templated by the telomerase RNA of this species, suggesting that it is maintained by a nontelomerase-mediated, likely recombinational, mechanism.

In summary, our data suggest that there are multiple mechanisms that contribute to telomeric repeat turnover in K. lactis that act with widely different frequencies. Primarily, gradual replicative shortening countered by telomerase addition acts during most or all cell cycles. Second, as identified here, other processes, perhaps including nucleolytic truncation of telomeres, act much less frequently and contribute to turnover of all but the innermost few repeats. Finally, on evolutionary time scales, recombination events may homogenize all repeats of telomeres.

	ACKNOWLEDGMENTS

We gratefully acknowledge support for this work from the National Institutes of Health (grants GM26259 and DE11356) and the American Cancer Society (grant RPG-00-082-01-GMC). Dana Hager Underwood was supported by a University of Georgia University Wide Fellowship.

Manuscript received August 17, 2001; Accepted for publication October 18, 2001.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

BOSCO, G. and J. E. HABER, 1998  Chromosome break-induced DNA replication leads to nonreciprocal translocations and telomere capture. Genetics 150:1037-1047[Abstract/Free Full Text].

BUCHOLC, M., Y. PARK, and A. J. LUSTIG, 2001  Intrachromatid excision of telomeric DNA as a mechanism for telomere size control in Saccharomyces cerevisiae. Mol. Cell. Biol. 21:6559-6573[Abstract/Free Full Text].

CHRISTIANSON, T. W., R. S. SIKORSKI, M. DANTE, J. H. SHERO, and P. HIETER, 1992  Multifunctional yeast high-copy-number shuttle vectors. Gene 110:119-122[Medline].

CHURCH, G. M. and W. GILBERT, 1984  Genomic sequencing. Proc. Natl. Acad. Sci. USA 81:1991-1995[Abstract/Free Full Text].

COHN, M., M. J. MCEACHERN, and E. H. BLACKBURN, 1998  Telomeric sequence diversity within the genus Saccharomyces. Curr. Genet. 33:83-91[Medline].

CONRAD, M. N., J. H. WRIGHT, A. J. WOLF, and V. A. ZAKIAN, 1990  RAP1 protein interacts with yeast telomeres in vivo: overproduction alters telomere structure and decreases chromosome stability. Cell 63:739-750[Medline].

FORSTEMANN, K., M. HOSS, and J. LINGNER, 2000  Telomerase-dependent repeat divergence at the 3' ends of yeast telomeres. Nucleic Acids Res. 28:2690-2694[Abstract/Free Full Text].

GILLEY, D. and E. H. BLACKBURN, 1996  Specific RNA residue interactions required for enzymatic functions of Tetrahymena telomerase. Mol. Cell. Biol. 16:66-75[Abstract].

GILLEY, D., M. S. LEE, and E. H. BLACKBURN, 1995  Altering specific telomerase RNA template residues affects active site function. Genes Dev. 9:2214-2226[Abstract/Free Full Text].

HOLLINGSWORTH, N. M. and B. BYERS, 1989  HOP1: a yeast meiotic pairing gene. Genetics 121:445-462[Abstract/Free Full Text].

INBAR, O. and M. KUPIEC, 1999  Homology search and choice of homologous partner during mitotic recombination. Mol. Cell. Biol. 19:4134-4142[Abstract/Free Full Text].

KIM, N. W., M. A. PIATYSZEK, K. R. PROWSE, C. B. HARLEY, and M. D. WEST et al., 1994  Specific association of human telomerase activity with immortal cells and cancer. Science 266:2011-2015[Abstract/Free Full Text].

KIRK, K. E. and E. H. BLACKBURN, 1995  An unusual sequence arrangement in the telomeres of the germ-line micronucleus in Tetrahymena thermophila.. Genes Dev. 9:59-71[Abstract/Free Full Text].

KIRK, K. E., B. P. HARMON, I. K. REICHARDT, J. W. SEDAT, and E. H. BLACKBURN, 1997  Block in anaphase chromosome separation caused by a telomerase template mutation. Science 275:1478-1481[Abstract/Free Full Text].

KRAUSKOPF, A. and E. H. BLACKBURN, 1996  Control of telomere growth by interactions of RAP1 with the most distal telomeric repeats. Nature 383:354-357[Medline].

KRAUSKOPF, A. and E. H. BLACKBURN, 1998  Rap1 protein regulates telomere turnover in yeast. Proc. Natl. Acad. Sci. USA 95:12486-12491[Abstract/Free Full Text].

LEE, M. S. and E. H. BLACKBURN, 1993  Sequence-specific DNA primer effects on telomerase polymerization activity. Mol. Cell. Biol. 13:6586-6599[Abstract/Free Full Text].

LI, B. and A. J. LUSTIG, 1996  A novel mechanism for telomere size control in Saccharomyces cerevisiae. Genes Dev. 10:1310-1326[Abstract/Free Full Text].

LINGNER, J. and T. R. CECH, 1998  Telomerase and chromosome end maintenance. Curr. Opin. Genet. Dev. 8:226-232[Medline].

LUNDBLAD, V. and E. H. BLACKBURN, 1993  An alternative pathway for yeast telomere maintenance rescues est1- senescence. Cell 73:347-360[Medline].

LUNDBLAD, V. and J. W. SZOSTAK, 1989  A mutant with a defect in telomere elongation leads to senescence in yeast. Cell 57:633-643[Medline].

LUSTIG, A. J., S. KURTZ, and D. SHORE, 1990  Involvement of the silencer and UAS binding protein RAP1 in regulation of telomere length. Science 250:549-553[Abstract/Free Full Text].

MALKOVA, A., E. L. IVANOV, and J. E. HABER, 1996  Double-strand break repair in the absence of RAD51 in yeast: a possible role for break-induced DNA replication. Proc. Natl. Acad. Sci. USA 93:7131-7136[Abstract/Free Full Text].

MARCAND, S., E. GILSON, and D. SHORE, 1997  A protein-counting mechanism for telomere length regulation in yeast. Science 275:986-990[Abstract/Free Full Text].

MARCAND, S., V. BREVET, and E. GILSON, 1999  Progressive cisinhibition of telomerase upon telomere elongation. EMBO J. 18:3509-3519[Medline].

MARUSIC, L., M. ANTON, A. TIDY, P. WANG, and B. VILLEPONTEAU et al., 1997  Reprogramming of telomerase by expression of mutant telomerase RNA template in human cells leads to altered telomeres that correlate with reduced cell viability. Mol. Cell. Biol. 17:6394-6401[Abstract].

MCCORMICK-GRAHAM, M., W. J. HAYNES, and D. P. ROMERO, 1997  Variable telomeric repeat synthesis in Paramecium tetraurelia is consistent with misincorporation by telomerase. EMBO J. 16:3233-3242[Medline].

MCEACHERN, M. J. and E. H. BLACKBURN, 1994  A conserved sequence motif within the exceptionally diverse telomeric sequences of budding yeasts. Proc. Natl. Acad. Sci. USA 91:3453-3457[Abstract/Free Full Text].

MCEACHERN, M. J. and E. H. BLACKBURN, 1995  Runaway telomere elongation caused by telomerase RNA gene mutations. Nature 376:403-409[Medline].

MCEACHERN, M. J. and E. H. BLACKBURN, 1996  Cap-prevented recombination between terminal telomeric repeat arrays (telomere CPR) maintains telomeres in Kluyveromyces lactis lacking telomerase. Genes Dev. 10:1822-1834[Abstract/Free Full Text].

MCEACHERN, M. J., and E. H. BLACKBURN, 1997 Consequences of mutations that alter telomeres in the yeast K. lactis, pp. 111–127 in Pezcoller Foundation Symposia No. 8: Genomic Instability and Immortality in Cancer, edited by E. MIHICH, and L. HARTWELL. Plenum Press, New York.

MCEACHERN, M. J. and S. IYER, 2001  Short telomeres in yeast are highly recombinogenic. Mol. Cell 7:695-704[Medline].

MCEACHERN, M. J., S. IYER, T. B. FULTON, and E. H. BLACKBURN, 2000a  Telomere fusions caused by mutating the terminal region of telomeric DNA. Proc. Natl. Acad. Sci. USA 97:11409-11414[Abstract/Free Full Text].

MCEACHERN, M. J., A. KRAUSKOPF, and E. H. BLACKBURN, 2000b  Telomeres and their control. Annu. Rev. Genet. 34:331-358[Medline].

MORROW, D. M., C. CONNELLY, and P. HIETER, 1997  "Break copy" duplication: a model for chromosome fragment formation in Saccharomyces cerevisiae.. Genetics 147:371-382[Abstract].

MÜLLER, T., E. GILSON, R. SCHMIDT, R. GIRALDO, and J. SOGO et al., 1994  Imaging the asymmetrical DNA bend induced by repressor activator protein 1 with scanning tunneling microscopy. J. Struct. Biol. 113:1-12[Medline].

MURNANE, J. P., L. SABATIER, B. A. MARDER, and W. F. MORGAN, 1994  Telomere dynamics in an immortal human cell line. EMBO J. 13:4953-4962[Medline].

NUGENT, C. I., T. R. HUGHES, N. F. LUE, and V. LUNDBLAD, 1996  Cdc13p: a single-strand telomeric DNA-binding protein with a dual role in yeast telomere maintenance. Science 274:249-252[Abstract/Free Full Text].

PAYS, E., M. GUYAUX, D. AERTS, N. VAN MEIRVENNE, and M. STEINERT, 1985  Telomeric reciprocal recombination as a possible mechanism for antigenic variation in trypanosomes. Nature 316:562-564[Medline].

PONZI, M., T. PACE, E. DORE, L. PICCI, and E. PIZZI et al., 1992  Extensive turnover of telomeric DNA at a Plasmodium berghei chromosomal extremity marked by a rare recombinational event. Nucleic Acids Res. 20:4491-4497[Abstract/Free Full Text].

PRESCOTT, J. and E. H. BLACKBURN, 1997a  Functionally interacting telomerase RNAs in the yeast telomerase complex. Genes Dev. 11:2790-2800[Abstract/Free Full Text].

PRESCOTT, J. and E. H. BLACKBURN, 1997b  Telomerase RNA mutations in Saccharomyces cerevisiae alter telomerase action and reveal nonprocessivity in vivo and in vitro.. Genes Dev. 11:528-540[Abstract/Free Full Text].

PRESCOTT, J. C. and E. H. BLACKBURN, 2000  Telomerase RNA template mutations reveal sequence-specific requirements for the activation and repression of telomerase action at telomeres. Mol. Cell. Biol. 20:2941-2948[Abstract/Free Full Text].

REDDEL, R. R., T