Telomerase-Independent Proliferation Is Influenced by Cell Type in Saccharomyces cerevisiae

Corresponding author: Lorraine Pillus, 9500 Gilman Dr., University of California, San Diego, CA 92093-0347., lpillus{at}biomail.ucsd.edu (E-mail)

Communicating editor: F. WINSTON

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Yeast strains harboring mutations in genes required for telomerase function (TLC1 and the EST genes) exhibit progressive shortening of telomeric DNA and replicative senescence. A minority of cells withstands loss of telomerase through RAD52-dependent amplification of telomeric and subtelomeric sequences; such survivors are now capable of long-term propagation with telomeres maintained by recombination rather than by telomerase. Here we report that simultaneous expression in haploid cells of both MATa and MAT information suppresses the senescence of telomerase-deficient mutants, with suppression occurring via the RAD52-dependent survivor pathway(s). Such suppression can be mimicked by deletion of SIR1–SIR4, genes that function in transcriptional silencing of several loci including the silent mating-type loci. Furthermore, telomerase-defective diploid strains that express only MATa or MAT information senesce at a faster rate than telomerase-defective diploids that are heterozygous at the MAT locus. This suggests that the RAD52-dependent pathway(s) for telomere maintenance respond to changes in the levels of recombination, a process regulated in part by the hierarchy of gene control that includes MAT regulation. We propose that cell-type-specific regulation of recombination at human telomeres may similarly contribute to the tissue-specific patterns of disease found in telomerase-deficient tumors.

TELOMERES, the physical ends of chromosomes, are composed of unusual chromatin and are required to prevent such catastrophic cellular events as chromosome loss, degradation, and end-to-end fusions. In most organisms, telomeres are composed of tandemly arrayed short sequence repeats flanked on the centromere-proximal side by middle-repetitive sequence elements (reviewed in LOUIS 1995 ). In the yeast Saccharomyces cerevisiae, telomeric DNA is composed of TG_1–3 repeats totaling 300–500 bp (SHAMPAY et al. 1984 ). These repeats are often abutted by one to four copies of subtelomeric sequences called Y' elements, each separated by 50–130 bp of TG_1–3 repeats (WALMSLEY et al. 1984 ), and by a mosaic of more centromere-proximal subtelomeric sequences, collectively referred to as X elements (FLINT et al. 1997 ; PRYDE et al. 1997 ). The functions of Y' and X repeats are unknown, but their presence indicates that recombination has occurred at telomeres (LOUIS and HABER 1990 ; LOUIS et al. 1994 ).

Maintenance of telomeres normally requires the enzyme telomerase, a reverse transcriptase complex containing an RNA molecule that serves as an internal template for the synthesis of new telomeric DNA repeats (reviewed in NUGENT and LUNDBLAD 1998 ). In S. cerevisiae, the RNA and catalytic protein components of the core enzyme are encoded by TLC1 (telomerase component) and EST2 (ever shorter telomeres), respectively (SINGER and GOTTSCHLING 1994 ; COUNTER et al. 1997 ; LINGNER et al. 1997 ). Three additional yeast genes, EST1, EST3, and CDC13 (EST4), can be mutated to yield phenotypes identical to those of tlc1 and est2 mutants and function in the same pathway as components of telomerase (LUNDBLAD and SZOSTAK 1989 ; LENDVAY et al. 1996 ; NUGENT et al. 1996 ; MORRIS and LUNDBLAD 1997 ). Of these, Est1p and Est3p are components of the telomerase holoenzyme whereas Cdc13p serves both to protect the telomere from nuclease activity and to recruit telomerase to the telomere (EVANS and LUNDBLAD 1999 ; HUGHES et al. 2000 ; PENNOCK et al. 2001 ).

Disruption of telomerase function in yeast is not immediately detrimental because each telomere loses only a few base pairs of DNA upon each cell division (LUNDBLAD and SZOSTAK 1989 ; SINGER and GOTTSCHLING 1994 ; LENDVAY et al. 1996 ). However, progressive shortening of telomeres in tlc1 and est mutants correlates with an eventual increase in cell death, often referred to as replicative senescence. It is noteworthy that after extended growth, a small proportion of tlc1, est2, est1, est3, or cdc13-2 mutants escape senescence (LUNDBLAD and BLACKBURN 1993 ; SINGER and GOTTSCHLING 1994 ; LENDVAY et al. 1996 ). These "survivors" can be classified into two categories that are distinguished by general differences in the rearrangements that occur at chromosome ends (reviewed in LUNDBLAD 2002 ). The more common class, sometimes referred to as type I survivors, has rearranged chromosomal termini that contain massive amplifications of the subtelomeric sequence element Y' capped with very short tracts of TG_1–3 repeats. The growth rate and viability of these survivors fluctuates dramatically and is coupled with the reappearance of rare senescent subpopulations. In the second class, referred to as type II survivors, little Y' amplification is seen. Instead, chromosomes have extremely long terminal TG_1–3 repeats and exhibit long-term viability and healthy growth rates. Upon sustained outgrowth, type I survivors frequently transform to type II survivors, but the reverse is not observed (TENG and ZAKIAN 1999 ). Notably, however, in the absence of RAD52, neither type of survivor can form, indicating that maintenance of telomeres in the absence of telomerase requires homologous recombination (LUNDBLAD and BLACKBURN 1993 ; SINGER and GOTTSCHLING 1994 ; LENDVAY et al. 1996 ).

Recent studies have further emphasized the role of recombination in extending cell survival in the absence of telomerase. For example, a partial loss of telomerase, resulting in stably short telomeres but no obvious senescence phenotype, can still confer a growth disadvantage (MORRIS and LUNDBLAD 1997 ). Such shortened telomeres can also be highly recombinogenic, even prior to becoming critically short. In Kluyveromyces lactis, recombination-mediated exchanges among subtelomeric repeats are increased by up to 200-fold in cells in which telomerase is still partially active (MCEACHERN and IYER 2001 ). Alterations that increase the frequency of genetic exchanges at chromosomal termini can also influence telomerase-independent survival. Defects in mismatch repair, which relieve the normal inhibition of recombination between homeologous DNA sequences, enhance the growth of strains of either K. lactis or S. cerevisiae that also lack telomerase (RIZKI and LUNDBLAD 2001 ). In fact, a mismatch repair defect can influence growth during the early stages following loss of telomerase, well before telomeres become critically short and the strain displays a noticeable barrier to proliferation. These studies collectively suggest that at least one pathway for conversion to a telomerase-independent survivor may depend on multiple rounds of recombination and that recombination contributes to telomere maintenance even during the early stages of growth of a telomerase-minus strain (LUNDBLAD and BLACKBURN 1993 ). The enhanced recombinogenic nature of telomeres that have undergone only intermediate shortening may be a reflection of a partial loss of telomeric end protection, which increases the risk that chromosomal termini will be exposed to the types of DNA repair activities that normally act on double-strand breaks (MCEACHERN and BLACKBURN 1996 ; MCEACHERN and IYER 2001 ).

DNA repair and recombination activities are also influenced by changes in cell identity or mating type (LOVETT and MORTIMER 1987 ; HEUDE and FABRE 1993 ; SCHILD 1995 ; YAN et al. 1995 ; FRANK-VAILLANT and MARCAND 2001 ; KEGEL et al. 2001 ; OOI et al. 2001 ; VALENCIA et al. 2001 ; MORGAN et al. 2002 ). In S. cerevisiae, a cell's identity is established by the mating-type information at the MAT locus that encodes transcription factors affecting the expression of a variety of haploid and diploid specific genes. Haploid cells express either MATa or MAT information, whereas diploids have and express both types of information simultaneously. Additional copies of the mating-type information are at HML and HMR, and these loci are ordinarily transcriptionally silenced through the action of a variety of cis-acting sequences and trans-acting factors (reviewed in FREEMAN-COOK et al. 2000 ). This latter category includes Sir2–4p, along with Sir1p. The SIR2–4 (silent information regulator) genes encode components of telomeric chromatin and are required to maintain both transcriptional silencing and wild-type length of the telomeres (APARICIO et al. 1991 ; PALLADINO et al. 1993 ; HECHT et al. 1996 ; GOTTA et al. 1997 ; STRAHL-BOLSINGER et al. 1997 ). In contrast, the function of SIR1 appears most significant at the silent mating-type loci, although a modest role has also been noted for it within the subtelomeric repeats (FOUREL et al. 1999 ; PRYDE and LOUIS 1999 ). In the absence of SIR gene functions, a haploid cell gains MATa/MAT characteristics and, consequently, it is nonmating and transcription of haploid specific genes is repressed.

Here we present evidence that cell mating-type identity influences the ability of cells to survive in the absence of telomerase. Specifically, simultaneous expression of MATa and MAT information in a haploid cell suppresses the senescence phenotype of telomerase mutants. This suppression is also induced by deletion of SIR1–4, but only when such deletions are accompanied by a state of MAT heterozygosity. Suppression of senescence is dependent on RAD52, indicating that coexpression of MATa and MAT enhances the formation of telomerase-independent survivors. This effect is also observed in telomerase-defective diploid strains, where the severity of the telomere replication defect exhibits a clear correlation with the MATa/MAT program of gene expression. Together, our findings suggest that changes in cell identity that lead to alterations in gene expression enhance the efficiency of recombination-dependent telomere maintenance pathways in strains that lack telomerase.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Yeast strains and media:
The strains used in this study are listed in Table 1. Since yeast harboring mutations in the TLC1 or EST genes undergo senescence, tlc1 and est1 mutants were covered by a plasmid copy of the appropriate gene prior to performing experiments. All strains are in the W303 background. The sir3::TRP1 (STONE et al. 1991 ), sir4::HIS3 (MARSHALL et al. 1987 ), tlc1-::LEU2 (LENDVAY et al. 1996 ), est1-1::HIS3 (LUNDBLAD and SZOSTAK 1989 ), rad52::LEU2 (SCHILD et al. 1983 ), and hml::TRP1 alleles were introduced into these strains through crosses or standard disruption techniques (ROTHSTEIN 1983 ; BAUDIN et al. 1993 ). Both sir1::TRP1 and sir2::TRP1 are null alleles (gifts of J. Rine) that were generated by completely deleting the open reading frames of SIR1 or SIR2, respectively, replacing them with TRP1, and introducing them into strains through crosses. LPY4737, a diploid MATa/MAT est1/est1 mutant strain, was generated by crossing LPY3143 with LPY3149. Subsequently, LPY4737 was subjected to low doses of UV irradiation to stimulate recombination at MAT. This treatment resulted in the production of est1/est1 diploid mutants that were homozygous for either MAT (LPY7568) or MATa (LPY7571) as confirmed by mating assays and molecular amplification using MATa- and MAT-specific primers. The control strain LPY7507 was likewise constructed and confirmed.

We observed that strains in the W303 background spontaneously produce pet mutants at a high frequency, a situation complicating analysis of the mutant strains. To circumvent this problem, all of the strains used in this study, except LPY2691, LPY4745, LPY4749, LPY4884, LPY4994, and LPY4999, were made rho° as described (FOX et al. 1991 ), and results in otherwise isogenic rho° and rho⁺ strains were comparable.

YPD, YPG, YPAD, supplemented synthetic medium lacking specific nutrients used to maintain plasmid selection, and liquid sporulation media were prepared as described (ADAMS et al. 1998 ). 5-Fluoroorotic acid (5-FOA; Toronto Research Chemicals) was added to supplemented synthetic medium at a concentration of 0.1% (BOEKE et al. 1987 ).

DNA manipulations:
The plasmids used in this study are listed in Table 2. pSD120 (a gift of S. Diede and D. Gottschling) was constructed by cloning the 2.6-kb HpaII fragment of TLC1 into pRS316. pVL308 was made by cloning the 2.6-kb BamHI/SalI fragment of EST1 into YCplac33 opened with BamHI and SalI. pLP923 was created by inserting the 4.5-kb SalI fragment of SIR3 into pVL308 opened with SalI. pLP1185 was constructed by inserting an 4-kb PstI/ClaI fragment of MATa into pRS314 opened with PstI and ClaI. Yeast transformations were performed as described (SCHIESTL and GIETZ 1989 ). Disruption plasmids pVL154 and pSM20 were used as described (SCHILD et al. 1983 ; LUNDBLAD and SZOSTAK 1989 ), and disruptions were confirmed by molecular amplification and/or by genomic DNA blotting.

Senescence assays:
The growth phenotypes of haploid and diploid mutant and control strains were assayed for senescence using a protocol previously established for analysis of est mutant strains (LENDVAY et al. 1996 ). In the experiments presented here, tlc1 and est1 haploid single-mutant strains or est1/est1 diploid mutants were generated by a plasmid shuffle in which spontaneous loss of covering plasmids with wild-type genes was selected by plating on 5-FOA (BOEKE et al. 1987 ). This plasmid shuffle results in some variability in the onset and severity of the senescence phenotype of tlc1 and est1 mutants (presumably a reflection of variability in the time at which the covering plasmid was lost). To address this variability, large numbers of isolates of each mutant and mutant combination were analyzed.

Fresh isolates of each strain were streaked onto synthetic complete medium containing 5-FOA and incubated for 96 hr at 30° to select for the loss of plasmids containing the relevant wild-type gene (pSD120, pVL308, or pLP923). In a typical experiment, 4–12 single, small colonies (<1 mm in diameter) of each strain were streaked from 5-FOA onto YPAD, incubated for 96 hr at 30°, examined for growth phenotypes, and photographed (representing 25 generations). Single colonies from the first passage were restreaked and analyzed similarly (50 generations). This process was repeated a third time (75 generations). In experiments conducted with haploid strains for the majority (75%) of tlc1 and est1 mutant isolates examined, the most extreme senescence was observed at 75 generations after which time RAD52-dependent survivors overwhelmed the population. All tlc1 sir and est1 sir double-mutant isolates behaved comparably, and senescence among double mutants was only infrequently observed over the course of an experiment. Examples of typical experiments are presented in the RESULTS.

For strains harboring pRS314 or pLP1185 in addition to pSD120, pVL308, or pLP923, senescence assays were performed as described above except that the medium used lacked tryptophan.

Telomeric DNA analysis:
In parallel with the senescence experiments described above, individual colonies grown for 25 generations were inoculated into 8 ml of YPAD and grown at 30° for 48 hr. Genomic DNA was prepared (HOFFMAN and WINSTON 1987 ), digested with XhoI, separated on a 0.7% agarose gel, transferred to nylon membrane, and probed with poly d(GT/CA) as described (LUNDBLAD and SZOSTAK 1989 ). The relative amplification of Y' elements in tlc1 and tlc1 sir mutants was quantified by PhosphorImager analysis and the "Imagequant" software package (Molecular Dynamics, Sunnyvale, CA) by comparing the intensity of the signal corresponding to both Y' elements relative to a nontelomeric 4-kb band also detected by the poly d(GT/CA) probe. The data presented in Fig 3 represent multiple independent experiments. Genomic DNA from a total of 27 tlc1 or est1 single-mutant isolates and 44 sir tlc1 or sir est1 double-mutant isolates at early cultivation times was examined by Southern blotting. From these analyses, indication of Y' amplification was observed in only eight tlc1/est1 mutants, but was clearly apparent in 38 of the sir tlc1/sir est1 double-mutant isolates.

View larger version (74K): In this window In a new window Download PPT slide   Figure 1. The senescence of haploid telomerase mutants is suppressed by the coexpression of MATa and MAT. Viability of (A) wild-type (LPY4424), tlc1 (LPY4421), and est1 (LPY4427) strains harboring a vector-control plasmid and (B) wild-type (LPY4422), tlc1 (LPY4419), and est1 (LPY4425) strains harboring plasmid-borne MATa was assayed by successive streak-outs. Each streak-out represents 25 generations of growth.

View larger version (299K): In this window In a new window Download PPT slide   Figure 2. The senescence of telomerase mutants is suppressed by sir1, sir2, sir3, and sir4 mutations. Viability of (A) wild-type (LPY3107), sir1 (LPY3478), sir2 (LPY3473), sir3 (LPY3109), and sir4 (LPY3111) control strains; (B) tlc1 (TVL249), tlc1 sir1 (LPY3479), tlc1 sir2 (LPY3474), tlc1 sir3 (LPY3105), and tlc1 sir4 (LPY3085) mutant strains; and (C) est1 (LPY3143), est1 sir1 (LPY3480), est1 sir2 (LPY3477), est1 sir3 (LPY3146), and est1 sir4 (LPY3147) mutant strains was assayed by successive streak-outs. Each streak-out represents 25 generations of growth.

View larger version (121K): In this window In a new window Download PPT slide   Figure 3. Telomeric rearrangements occur early in tlc1 sir3 and tlc1 sir4 mutants. A Southern blot of genomic DNA prepared from isolates of wild-type (LPY3107), sir3 (LPY3109), sir4 (LPY3111), tlc1 (TVL249), tlc1 sir3 (LPY3105), and tlc1 sir4 (LPY3085) grown in the absence of TLC1 for 25 generations was probed with a telomere-specific probe [poly d(GT/CA)]. Telomeric TG1–3 DNA from Y'-containing telomeres is bracketed. The two major classes of Y' elements are marked by arrows, and examples of non-Y'-containing telomeres are indicated by arrowheads. In addition, lanes marked by asterisks correspond to isolates that have an appearance characteristic of type II survivors. Note that the telomeric TG1–3 DNA from a tlc1 sir4 double mutant is heterogeneous in length; although we do not understand the basis for this phenomenon, it is apparently genotype independent (C. NUGENT and V. LUNDBLAD, unpublished results; see also Fig 1A in NUGENT et al. 1996 for further examples).

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Coexpression of MATa and MAT suppresses senescence of haploid telomerase mutants:
S. cerevisiae containing a mutation either in TLC1 or in any of the EST genes exhibit progressive shortening of telomeric TG_1–3 DNA and an accompanying senescence phenotype (LUNDBLAD and SZOSTAK 1989 ; SINGER and GOTTSCHLING 1994 ; LENDVAY et al. 1996 ). At its MAT locus, a haploid cell ordinarily expresses either "a" or "" information, but not both simultaneously. In fact, the coexpression of MATa and MAT information leads to the production of an a1/2 heterodimer that blocks the transcription of a variety of haploid-specific genes. To test whether a cell's identity influences its response to loss of telomerase, we introduced plasmids containing MATa (pMATa) or a vector control (vector) into a MAT tlc1 mutant or MAT wild-type strains. Senescence was monitored by the successive restreaking of multiple isolates of each strain, with each streak-out representing 25 generations of growth (see MATERIALS AND METHODS). As expected, a MAT tlc1 mutant harboring a vector-control plasmid exhibited senescence, showing a moderate growth defect at 50 generations and a severe loss of viability by 75 generations (Fig 1A). Strikingly, however, a MAT tlc1 mutant harboring pMATa did not exhibit notable levels of inviability by this colony assay (Fig 1B), suggesting that changing a haploid cell's identity from to a/ results in the suppression of tlc1 senescence. In control experiments, neither pMATa nor vector plasmids affected the growth of a MAT wild-type strain (Fig 1A and Fig B).

To test the specificity of the suppression, pMATa or vector-control plasmids were introduced into a MAT est1 mutant. In the presence of the vector-control plasmid, MAT est1 senesced (Fig 1A), whereas in the presence of pMATa, suppression was again observed (Fig 1B). Thus, suppression of senescence that results from the coexpression of MATa and MAT is not restricted to loss of function of the RNA component of telomerase. This finding supports previous epistasis analysis demonstrating that tlc1 and est1, along with est2, est3, and cdc13-2 mutants, share identical phenotypes, consistent with their involvement in the same process of telomere maintenance and replication (LENDVAY et al. 1996 ; NUGENT et al. 1996 ). We predict that coexpression of MATa and MAT would have comparable effects on est2, est3, and cdc13-2 mutants, although these double-mutant combinations have not been tested.

Mutations in SIR genes suppress the senescence of telomerase mutants:
The silent mating-type loci are transcriptionally silenced by a combination of cis-acting elements and trans-acting factors including the Sir1–4 proteins (FREEMAN-COOK et al. 2000 ). Loss of SIR function leads to the expression of the mating-type information at HML and HMR, resulting in a nonmating, a/ haploid. On the basis of our observation that simultaneous expression of MATa and MAT suppresses the senescence of haploid telomerase mutants, we hypothesized that loss of SIR function would similarly suppress senescence. To address this hypothesis, we evaluated genetic interactions between sir and tlc1/est mutants.

As expected, wild-type, sir1, sir2, sir3, and sir4 strains did not display a decline in viability following extensive propagation (Fig 2A). To examine if the absence of SIR function suppresses the senescence of telomerase mutants, we introduced sir1, sir2, sir3, or sir4 mutations into tlc1 mutant strains. By this colony assay, all tlc1 sir double mutants failed to senesce over the course of 75 generations, indicating that the loss of SIR1, SIR2, SIR3, or SIR4 function suppressed tlc1 senescence (Fig 2B). To test whether the sir-mediated suppression of senescence was specific for tlc1, we similarly introduced sir mutations into an est1 strain. All sir mutations suppressed est1 senescence in a manner comparable to their suppression of senescence in a tlc1 mutant (compare B and C in Fig 2). Suppression of senescence was also observed in sir3 est2 and sir4 est2 mutant strains (data not shown). Thus, by the criterion that sir suppression of senescence was not specific for a single telomerase-defective strain, sir suppression appeared comparable to that of the coexpression of MATa and MAT.

On the basis of growth rates and variability in colony size and morphology, a subset of tlc1 sir, est1 sir, and est2 sir double-mutant isolates could be distinguished from wild-type strains (data not shown). In fact, est1 sir double-mutant isolates analyzed by restreaking for up to 250 generations fluctuated through periods of senescence and periods of viability much like est1 single-mutant type I survivors (LUNDBLAD and BLACKBURN 1993 ). This suggests that suppression occurs by enhancement of the pathway(s) that mediates the formation of survivors in tlc1 and est single-mutant strains (see also below).

Loss of SIR function promotes the formation of a telomerase-independent telomere maintenance pathway:
In seeking to identify the mechanism by which sir mutations suppress tlc1 and est senescence, we hypothesized at least two possibilities. Suppression might occur such that characteristic shortening of telomeric TG_1–3 DNA in telomerase-defective strains was prevented. Alternatively, the absence of SIR might facilitate recombination at telomeres, resulting in an increased frequency of appearance of telomerase-independent survivors. To test these possibilities, we examined telomeric DNA from multiple isolates of wild-type, sir3, sir4, tlc1, tlc1 sir3, and tlc1 sir4 strains. The strains were grown in parallel for 25 generations, a time at which amplification of either subtelomeric or telomeric sequences is not normally observed in survivors (LUNDBLAD and BLACKBURN 1993 ; V. LUNDBLAD, unpublished data). As described previously (SINGER and GOTTSCHLING 1994 ), telomeric DNA in tlc1 mutants was 150 bp shorter than telomeric DNA prepared from wild-type strains after 25 generations (Fig 3). Likewise in agreement with previous results (PALLADINO et al. 1993 ), sir3 and sir4 single mutants had telomeric DNA repeats 50 and 100 bp shorter than those found in wild type. In tlc1 sir3 and tlc1 sir4 double mutants, telomeric TG_1–3 DNA was shortened to an extent similar to that observed for tlc1 single mutants. These data thus demonstrated that sir3 and sir4 did not block the telomeric shortening of tlc1 mutants.

Although the length of the telomeric TG_1–3 DNA did not differ, examination of the subtelomeric DNA structure in tlc1 sir3 and tlc1 sir4 mutants indicated that the telomerase-independent pathway occurred earlier than it did in tlc1 single-mutant strains. In S. cerevisiae, the appearance of this telomerase-independent telomere maintenance pathway is heralded not only by a wild-type-like growth phenotype but also by changes in telomeric DNA composition that are visualized easily on genomic blots (LUNDBLAD and BLACKBURN 1993 ; TENG and ZAKIAN 1999 ). One change characteristic of this pathway is an increase in the copy number of two size classes of the Y' subtelomeric repeats (LUNDBLAD and BLACKBURN 1993 ; TENG and ZAKIAN 1999 ). These elements are not amplified early in the outgrowth of tlc1 or est mutant strains, but survivors generated after >100 generations of growth of a telomerase-defective strain exhibit increases in Y' copy number ranging from 10- to 200-fold, in parallel with amplification of TG_1–3 DNA (LUNDBLAD and BLACKBURN 1993 ; LENDVAY et al. 1996 ). As shown in Fig 3, amplification of Y' elements was observed even earlier in the outgrowth of tlc1 sir3 and tlc1 sir4 mutant strains relative to tlc1, sir3, sir4, and wild-type strains. Quantitation showed that Y' copy number was only marginally increased in tlc1 strains (by a factor of 1.9-fold, relative to wild type) at this early time point. In contrast, Y' copy number was increased by 4.3- and 6.5-fold in tlc1 sir3 and tlc1 sir4 strains, respectively, relative to wild-type and sir control strains (see MATERIALS AND METHODS).

A second change in telomeric structure that is diagnostic of the Y' amplification survivor pathway is the acquisition of Y' elements by telomeres that did not previously have this subtelomeric repeat. This rearrangement in telomeric structure can be detected by monitoring individual telomeric restriction fragments that lack Y' elements (indicated by arrowheads in Fig 3); such bands diminish or disappear in survivors (LUNDBLAD and BLACKBURN 1993 ). Consistent with the observed increase in Y' copy number, these bands were absent or weaker in intensity in many of the tlc1 sir3 and tlc1 sir4 double mutants (Fig 3). Analysis of telomeric DNA prepared from est1, est1 sir3, and est1 sir4 mutants prepared after 25 generations of growth revealed similar results: Y' elements in est1 sir3 and est1 sir4 double mutants were amplified and rearranged relative to est1 single-mutant isolates and control strains (data not shown). Although the predominant genomic change detected in tlc1 sir3 and tlc1 sir4 double-mutant strains was amplification and dispersal of Y' elements, we also observed changes characteristic of the appearance of type II survivors (see lanes marked with asterisks, Fig 3).

Thus, telomeric rearrangements occurred earlier in tlc1 sir or est1 sir double mutants than in tlc1 or est1 single mutants. These results suggest that suppression of tlc1 and est senescence in sir3 and sir4 mutants occurs through the same processes by which survivors ordinarily form. Previous work demonstrated that this telomerase-independent pathway is mediated via recombination, as elimination of RAD52 function blocks both the formation (LUNDBLAD and BLACKBURN 1993 ; LENDVAY et al. 1996 ) and maintenance (TENG and ZAKIAN 1999 ) of survivors. We therefore tested whether the suppression of est1 senescence by a sir3 mutation is similarly dependent upon RAD52. Fig 4 shows that the absence of RAD52 function had no effect on the growth of a sir3 mutant strain, and as expected, the senescence phenotype exhibited by the est1 mutant at 50 generations was suppressed in an est1 sir3 double mutant. However, both the est1 rad52 double mutant and the est1 sir3 rad52 triple mutant were inviable by 50 generations, and the growth patterns of these two strains were indistinguishable. Similarly, est1 sir1 rad52, est1 sir2 rad52, and est1 sir4 rad52 triple mutants exhibited the same rapid senescence phenotype of an est1 rad52 mutant (data not shown). Therefore, the suppression of senescence observed in est sir mutant strains was recombination mediated. This observation, in combination with the telomeric rearrangements observed in tlc1 sir and est sir strains, strongly suggests that sir suppression utilizes the same RAD52-dependent mechanisms originally defined for survivor formation (LUNDBLAD and BLACKBURN 1993 ).

View larger version (41K): In this window In a new window Download PPT slide   Figure 4. sir-mediated suppression of senescence is RAD52 dependent. Viability of (A) wild-type (LPY3409), est1 (LPY3410), rad52 (LPY3412), and est1 rad52 (LPY3414) control strains and (B) sir3 (LPY3411), est1 sir3 (LPY3413), sir3 rad52 (LPY3415), and est1 sir3 rad52 (LPY3416) mutant strains was assayed by successive streak-outs. To ensure that telomere lengths were comparable in both SIR3 wild-type and sir3 mutant strains at the start of this experiment, all strains initially harbored a URA3-marked plasmid bearing both EST1 and SIR3 (pLP923), which was lost upon growth on 5-FOA prior to the first streak-out depicted here.

Suppression of senescence by coexpression of MATa and MAT occurs through a RAD52-dependent mechanism:
To determine whether the suppression of senescence observed by coexpression of MATa and MAT occurred by stimulating the survivor pathway(s), we asked whether suppression was similarly dependent on RAD52. We introduced either a vector control or pMATa into est1, rad52, or est1 rad52 mutants. As shown in Fig 5A, the vector control did not block the senescence of est1 or est1 rad52 mutants and had no effect on the growth of a rad52 mutant. Fig 5B reveals that pMATa, as observed previously (see Fig 1B), suppressed est1 senescence and had no effect on a rad52 single mutant. Importantly, however, in the absence of RAD52, pMATa was unable to suppress the senescence of an est1 mutant. From these data, we conclude that suppression by coexpression of MATa and MAT requires RAD52-dependent homologous recombination and therefore is due to enhanced survivor formation.

View larger version (54K): In this window In a new window Download PPT slide   Figure 5. Cell-type-specific suppression of est1 senescence is RAD52 dependent. Viability of (A) est1 (LPY5020), rad52 (LPY5024), and est1 rad52 (LPY5028) strains harboring a vector-control plasmid and (B) est1 (LPY5022), rad52 (LPY5026), and est1 rad52 (LPY5030) strains harboring plasmid-borne MATa was assayed by successive streak-outs. We note that est1 and est1 rad52 mutants harboring the vector-control plasmid senesced especially rapidly in this experiment.

Coexpression of MATa and MAT information is required for the sir mutant-mediated suppression of senescence:
The suppression of senescence observed in telomerase-defective strains by sir mutations bears remarkable similarity to that of the suppression of tlc1 and est1 mutants by the coexpression of MATa and MAT. Notably, the suppression phenotypes are indistinguishable from one another and are likewise RAD52 dependent. However, given that Sir2–4p not only are required for silencing at the silent mating-type loci but also are components of telomeric chromatin (reviewed in THAM and ZAKIAN 2002 ), it was formally possible that although MATa and MAT coexpression is sufficient for suppression of telomerase-defective mutants, loss of SIR could directly promote survivor formation in telomerase-defective strains. To test this possibility, sir3 was combined with est1 in a genetic background in which the silent mating-type locus HML was deleted (hml), resulting in sir3 est1 cells that expressed only MATa information. If sir mutations had effects on suppression unrelated to the simultaneous expression of both forms of mating-type information, suppression of senescence in telomerase-defective strains would still be observed. On the other hand, if sir suppression required coexpression of MATa and MAT, then sir est1 strains would be predicted to senesce. The hml single and hml sir3 double mutants were mating competent, as expected, and did not display any loss in viability over the course of a senescence assay (Fig 6A and data not shown). The sir3 mutation, however, was no longer capable of suppressing est1 senescence in an hml background. These strains displayed a marked loss in viability by 50–75 generations (Fig 6A). Moreover, sir1 and sir2 mutations also failed to suppress the senescence of an hml est1 strain (Fig 6B), indicating that coexpression of MATa and MAT is a requirement of sir suppression of senescence.

View larger version (80K): In this window In a new window Download PPT slide   Figure 6. Suppression of senescence requires coexpression of MATa and MAT information. (A) Viability of MATa hml (LPY2691), MATa hml est1 (LPY4745), MATa hml sir3 (LPY4884), and MATa hml est1 sir3 (LPY4749) and (B) MATa hml est1 sir1 (LPY4994) and MATa hml est1 sir2 (LPY4999) strains were assayed by successive streak-outs.

Senescence of diploid telomerase mutants is also influenced by MAT expression:
The results above demonstrated that in haploid cells, MAT heterozygosity can modulate the onset of a recombination-dependent mechanism for telomere maintenance. This finding suggests that in diploid strains coexpression of both MAT loci, when compared to strains expressing only one type of mating-type information, should similarly influence survival in the absence of telomerase. To test this prediction, we generated isogenic est1/est1 diploid mutants that were either heterozygous (MATa/MAT) or homozygous (MATa/MATa or MAT/MAT) at the mating-type locus (see MATERIALS AND METHODS for details of strain construction). Diploid EST1/EST1 and haploid est1 strains were included as controls (Fig 7). As expected, diploid strains in which telomerase was intact did not exhibit any signs of senescence regardless of whether they were heterozygous or homozygous for MAT. In contrast, a MATa/MAT est1/est1 diploid displayed a clear senescence phenotype, although the onset of senescence in this strain was slightly delayed relative to est1 haploid controls (75 generations vs. 50 generations). Therefore, in a diploid strain, simply expressing both MATa and MAT is not sufficient to suppress senescence. By contrast, both MATa/MATa est1/est1 and MAT/MAT est1/est1 diploid mutants exhibited accelerated senescence relative to either the heterozygous MATa/MAT est1/est1 diploid or the haploid est1 mutant controls. Thus, similarly to the observations described above for haploid telomerase mutants, expression of both MATa and MAT information also facilitates survival in a diploid strain in the absence of telomerase.

View larger version (73K): In this window In a new window Download PPT slide   Figure 7. Senescence of diploid telomerase mutants is influenced by MAT. Viability of diploid control strains MATa/MAT (LPY7501) and MATa/MATa (LPY7507) and diploid mutant strains MATa/MAT est1/est1 (LPY4737), MATa/MATa est1/est1 (LPY7571), and MAT/MAT est1/est1 (LPY7568) was assayed by successive streak-outs and compared to haploid mutant controls MATa est1 (LPY3149) and MAT est1 (LPY3143). Each streak-out represents 25 generations of growth.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Over the past two decades, a mechanistic understanding of how organisms with linear chromosomes solve the "end replication" problem (WATSON 1972 ) has increased significantly. Components of telomerase, the end-replicating enzyme, have been identified in an enormous range of organisms (reviewed in NUGENT and LUNDBLAD 1998 ), implying that this mechanism of telomere replication is evolutionarily conserved. Initially, however, the discovery that telomeric DNA is highly repetitive and the observation that conserved, subtelomeric DNA sequences are frequently dispersed among chromosome ends was interpreted to suggest that recombination might play a central role in telomere replication. This viewpoint has recently gained renewed interest on the basis of the discovery that in S. cerevisiae, K. lactis, Schizosaccharomyces pombe, and certain human tumors and tumor cell lines, telomeres can sometimes be maintained in the absence of telomerase (LUNDBLAD and BLACKBURN 1993 ; MCEACHERN and BLACKBURN 1996 ; BRYAN et al. 1997 ; NAKAMURA et al. 1998 ).

The process by which telomerase-independent "survivors" arise in the budding yeast has been the subject of numerous studies (reviewed in LUNDBLAD 2002 ). Insight into the mechanism promoting this process came from the observation that telomerase-independent survivors could not form in the absence of RAD52, the gene required for most homologous recombination (LUNDBLAD and BLACKBURN 1993 ; MCEACHERN and BLACKBURN 1996 ). Not only did loss of RAD52 function ultimately block the formation of survivors, but also, in this genetic context, the senescence characteristics of a telomerase-defective strain were exaggerated, indicating that recombination contributed to telomere maintenance immediately after the loss of telomerase. In contrast, the viability of telomerase mutants harboring defects in the mismatch-repair pathway was enhanced, implying that a system that safeguards genomic stability interferes with survivor formation (RIZKI and LUNDBLAD 2001 ).

MAT heterozygosity suppresses senescence in telomerase mutants:
This work demonstrates that the senescence phenotype of telomerase mutants can also be influenced by the status of MAT in both haploid and diploid strains. A comparison of a MAT tlc1 or a MAT est1 strain to the identical strain harboring a pMATa plasmid revealed a striking phenotype: over the course of 75 generations, the senescence phenotype evident in the telomerase-defective strains was suppressed in those strains containing pMATa (see Fig 1). Consistent with the change in the status of MAT that accompanies mutations in SIR, we were able to mimic this phenotype by combining mutations in SIR1–4 with mutations in TLC1 or EST1 (see Fig 2). Suppression is not seen in the absence of RAD52 function (Fig 4), indicating that suppression requires homologous recombination. Such RAD52 dependence was also true for experiments with episomal expression of MAT (Fig 5). These data strongly suggest that the suppression we observed was due to an early arrival of survivors in a population of cells that would ordinarily be undergoing senescence.

It was noteworthy that simple loss of SIR function was not adequate to suppress senescence (Fig 6). Instead, mutations in SIR suppressed senescence only with simultaneous MAT heterozygosity (Fig 2), implying that a/ coexpression was necessary. Thus, this shows that the effects of mutations in the SIR genes are most likely indirect, rather than the consequence of a change in telomeric chromatin due to loss of the Sir complex.

These observations in haploid strains were also supported by an examination of the senescence phenotypes in telomerase-defective est1/est1 diploid strains, where a clear correlation between the severity of the telomere replication defect and the MATa/MAT program of gene expression was observed. Notably, however, senescence was not suppressed in an est1/est1 diploid expressing both MATa and MAT information, in contrast to the suppression observed in a haploid est1 strain expressing both MAT loci. This might be attributable to the twofold increase in the number of telomeres that must be processed to sustain viability, which could exceed the capacity of the recombination machinery in a diploid strain.

Recombination and repair phenotypes are associated with MAT heterozygosity:
Maintenance of silencing at the silent mating-type loci ensures that a haploid cell will be able to conjugate with a cell of its opposite mating type, thereby promoting the exchange of genetic material when the resulting diploid subsequently undergoes meiosis and sporulation. When a haploid cell becomes heterozygous for MAT, the resulting a1/2 heterodimer represses the expression of haploid-specific genes and the cell is unable to mate. Microarray techniques have facilitated the identification of genes whose expression is affected by the state of MAT (PRIMIG et al. 2000 ; additional microarray data cited in GALITSKI et al. 1999 ; KEGEL et al. 2001 ; OOI et al. 2001 ; VALENCIA et al. 2001 ), underscoring the fact that many differences exist between haploid and diploid yeast. The finding that at least one of these genes, NEJ1, is involved in nonhomologous end joining (NHEJ) is in accordance with previous observations that haploid and diploid cells rely on different types of recombination as their primary means of repairing double-strand DNA breaks. However, the suppressing effects we observe are clearly distinct from NHEJ mechanisms, since it has previously been reported that est1 mutants in which NHEJ is disrupted due to a mutation in LIG4 senesce with the same properties as est1 mutants alone (HACKETT et al. 2001 ). Moreover, MAT heterozygosity increases homologous recombination in diploid cells (LEE et al. 1999 ; CLIKEMAN et al. 2001 ), suppresses the X-ray-sensitive phenotype of a rad52-20 mutant (SCHILD 1995 ) and a rad54 null mutant (LOVETT and MORTIMER 1987 ), bypasses the ATP hydrolysis requirement for Rad51p (MORGAN et al. 2002 ), and influences the ability of a cell to withstand an unrepaired, double-strand break (BENNETT et al. 2001 ). Thus, MAT heterozygosity has widespread implications on processes that extend beyond the ability of a cell to mate.

We propose that changes in gene expression that accompany an alteration in cell identity can alter the frequency of recombination events at chromosomal termini as a consequence of either increased expression of genes whose products promote recombination or, alternatively, decreased expression of genes responsible for keeping recombination in check (Fig 8). Future identification of the specific genes responsible for the cell-type-specific effects on telomere maintenance will further advance our understanding of the mechanisms that promote proliferation in the absence of telomerase.

View larger version (20K): In this window In a new window Download PPT slide   Figure 8. A model for influence of cell identity on telomerase-independent proliferation. (A) In haploid telomerase mutants (MAT tlc1), the silent mating-type loci (here represented by HMRa) are transcriptionally silenced and homologous recombination DNA repair systems are ordinarily intact (indicated by the transcription of RAD52). As-yet-unidentified genes that facilitate recombination in diploid cells (GENE F) or inhibit recombination in haploid cells (GENE D) are, respectively, repressed and transcribed. Telomerase mutants senesce, but over time survivors form. (B) When a sir mutation is combined with a telomerase mutation, HMRa is transcribed, resulting in the formation of the a1/2 transcription factor. This change in cell-type transcription may then activate GENE F and/or repress GENE D. The resulting, new transcriptional program can then promote RAD52-dependent recombination and, hence, survivors form earlier than in telomerase single-mutant cells. (C) In the absence of RAD52, homologous recombination cannot occur and survivors do not form, regardless of cell mating-type identity.

Alternative telomere maintenance in mammalian cells and tumors:
Although 90% of immortal mammalian cell lines and tumors are positive for telomerase activity (KIM et al. 1994 ), indicating that they have escaped crisis by reactivating telomerase, numerous cell lines that are deficient in telomerase activity and yet contain grossly amplified tracts of telomeric repeats have also been identified (BRYAN et al. 1995 , BRYAN et al. 1997 ). This phenomenon, known as alternative telomere maintenance (ALT), exhibits similarities to the properties exhibited by survivors recovered from both S. cerevisiae and K. lactis strains (reviewed in HENSON et al. 2002 ; LUNDBLAD 2002 ), including evidence for recombination between telomeres in ALT cell lines (DUNHAM et al. 2000 ; VARLEY et al. 2002 ). Further parallels between ALT in human cells and tumors and survivor formation in yeast may well exist. Our observation that loss of silencing at the silent mating-type loci has indirect consequences in survivor formation in yeast hints that there may be cell-type-specific regulation of recombination at human telomeres. Ultimately, such cell-type differences may contribute to the tissue-specific patterns of disease found in individual telomerase-deficient tumors.

	FOOTNOTES

¹ Present address: Laboratory of Molecular Parasitology, Box 185, The Rockefeller University, 1230 York Ave., New York, NY 10021.

	ACKNOWLEDGMENTS

We thank A. Salinger for technical assistance with Fig 3 and Fig E. J. Spedale and S. Jacobson for help with strain construction. S. Diede, D. Gottschling, J. Rine, V. Zakian, K. Friedman, and R. Sternglanz generously provided strains, plasmids, and other reagents. We thank F. Winston and the anonymous reviewers of this and an earlier version of this work for their suggestions and insight. We also thank E. Stone and S. Jacobson for critical comments on the manuscript. J. Lowell was supported in part by a National Institutes of Health Training grant (GM-07135-22). This work was supported by grants from the National Institutes of Health to the labs of L.P. and V.L.

Manuscript received September 15, 2002; Accepted for publication March 26, 2003.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

ADAMS, A., D. E. GOTTSCHLING, C. A. KAISER and T. STEARNS, 1998 Methods in Yeast Genetics: A Cold Spring Harbor Laboratory Course Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

APARICIO, O. M., B. L. BILLINGTON, and D. E. GOTTSCHLING, 1991  Modifiers of position effect are shared between telomeric and silent mating-type loci in S. cerevisiae.. Cell 66:1279-1287.[Medline]

BAUDIN, A., O. OZIER-KALOGEROPOULOS, A. DENOUEL, F. LACROUTE, and C. CULLIN, 1993  A simple and efficient method for direct gene deletion in Saccharomyces cerevisiae.. Nucleic Acids Res. 21:3329-3330.[Free Full Text]

BENNETT, C. B., J. R. SNIPE, J. W. WESTMORELAND, and M. A. RESNICK, 2001  SIR functions are required for the toleration of an unrepaired double-strand break in a dispensable yeast chromosome. Mol. Cell. Biol. 21:5359-5373.[Abstract/Free Full Text]

BOEKE, J. D., J. TRUEHEART, G. NATSOULIS, and G. R. FINK, 1987  5-Fluoroorotic acid as a selective agent in yeast molecular genetics. Methods Enzymol. 154:164-175.[Medline]

BRYAN, T. M., A. ENGLEZOU, J. GUPTA, S. BACCHETTI, and R. R. REDDEL, 1995  Telomere elongation in immortal human cells without detectable telomerase activity. EMBO J. 14:4240-4248.[Medline]

BRYAN, T. M., A. ENGLEZOU, L. DALLA-POZZA, M. A. DUNHAM, and R. R. REDDEL, 1997  Evidence for an alternative mechanism for maintaining telomere length in human tumors and tumor-derived cell lines. Nat. Med. 3:1271-1274.[Medline]

CLIKEMAN, J. A., G. J. KHALSA, S. L. BARTON, and J. A. NICKOLOFF, 2001  Homologous recombinational repair of double-strand breaks in yeast is enhanced by MAT heterozygosity through yKU-dependent and -independent mechanisms. Genetics 157:579-589.[Abstract/Free Full Text]

COUNTER, C. M., M. MEYERSON, E. N. EATON, and R. A. WEINBERG, 1997  The catalytic subunit of yeast telomerase. Proc. Natl. Acad. Sci. USA 94:9202-9207.[Abstract/Free Full Text]

DUNHAM, M. A., A. A. NEUMANN, C. L. FASCHING, and R. R. REDDEL, 2000  Telomere maintenance by recombination in human cells. Nat. Genet. 26:447-450.[Medline]

EVANS, S. K. and V. LUNDBLAD, 1999  Est1 and Cdc13 as comediators of telomerase access. Science 286:117-120.[Abstract/Free Full Text]

FLINT, J., G. P. BATES, K. CLARK, A. DORMAN, and D. WILLINGHAM et al., 1997  Sequence comparison of human and yeast telomeres identifies structurally distinct subtelomeric domains. Hum. Mol. Genet. 6:1305-1313.[Abstract/Free Full Text]

FOUREL, G., E. REVARDEL, C. E. KOERING, and E. GILSON, 1999  Cohabitation of insulators and silencing elements in yeast subtelomeric regions. EMBO J. 18:2522-2537.[Medline]

FOX, T. D., L. S. FOLLEY, J. J. MULERO, T. W. MCMULLIN, P. E. THROSNESS et al., 1991 Analysis and manipulation of yeast mitochondrial genes, pp. 149–169 in Guide to Yeast Genetics and Molecular Biology, edited by C. GUTHRIE and G. R. FINK. Academic Press, San Diego.

FRANK-VAILLANT, M. and S. MARCAND, 2001  NHEJ regulation by mating type is exercised through a novel protein, Lif2p, essential to the ligase IV pathway. Genes Dev. 15:3005-3012.[Abstract/Free Full Text]

FREEMAN-COOK, L., R. KAMAKAKA and L. PILLUS, 2000 Chromatin contributions to epigenetic transcriptional states in yeast, pp. 203–227 in Chromatin Structure and Gene Expression, edited by S. C. R. ELGIN and J. L. WORKMAN. Oxford University Press, London/Oxford/New York.

GALITSKI, T., A. J. SALDANHA, C. A. STYLES, E. S. LANDER, and G. R. FINK, 1999  Ploidy regulation of gene expression. Science 285:251-254.[Abstract/Free Full Text]

GOTTA, M., S. STRAHL-BOLSINGER, H. RENAULD, T. LAROCHE, and B. K. KENNEDY et al., 1997  Localization of Sir2p: the nucleolus as a compartment for silent information regulators. EMBO J. 16:3243-3255.[Medline]

HACKETT, J. A., D. M. FELDSER, and C. W. GREIDER, 2001  Telomere dysfunction increases mutation rate and genomic instability. Cell 106:275-286.[Medline]

HECHT, A., S. STRAHL-BOLSINGER, and M. GRUNSTEIN, 1996  Spreading of transcriptional repressor SIR3 from telomeric heterochromatin. Nature 383:92-96.[Medline]

HENSON, J. D., A. A. NEUMANN, T. R. YEAGER, and R. R. REDDEL, 2002  Alternative lengthening of telomeres in mammalian cells. Oncogene 21:598-610.[Medline]

HEUDE, M. and F. FABRE, 1993  a/-control of DNA repair in the yeast Saccharomyces cerevisiae: genetic and physiological aspects. Genetics 133:489-498.[Abstract]

HOFFMAN, C. S. and F. WINSTON, 1987  A ten-minute DNA preparation from yeast efficiently releases autonomous plasmids for transformation of Escherichia coli.. Gene 57:267-272.[Medline]

HUGHES, T. R., S. K. EVANS, R. G. WEILBAECHER, and V. LUNDBLAD, 2000  The Est3 protein is a subunit of yeast telomerase. Curr. Biol. 10:809-812.[Medline]

KEGEL, A., J. O. SJOSTRAND, and S. U. ÅSTRÖM, 2001  Nej1p, a cell type-specific regulator of nonhomologous end joining in yeast. Curr. Biol. 11:1611-1617.[Medline]

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]

LEE, S. E., F. PAQUES, J. SYLVAN, and J. E. HABER, 1999  Role of yeast SIR genes and mating type in directing DNA double-strand breaks to homologous and non-homologous repair paths. Curr. Biol. 9:767-770.[Medline]

LENDVAY, T. S., D. K. MORRIS, J. SAH, B. BALASUBRAMANIAN, and V. LUNDBLAD, 1996  Senescence mutants of Saccharomyces cerevisiae with a defect in telomere replication identify three additional EST genes. Genetics 144:1399-1412.[Abstract]

LINGNER, J., T. R. HUGHES, A. SHEVCHENKO, M. MANN, and V. LUNDBLAD et al., 1997  Reverse transcriptase motifs in the catalytic subunit of telomerase. Science 276:561-567.[Abstract/Free Full Text]

LOUIS, E. J., 1995  The chromosome ends of Saccharomyces cerevisiae.. Yeast 11:1553-1573.[Medline]

LOUIS, E. J. and J. E. HABER, 1990  The subtelomeric Y' repeat family in Saccharomyces cerevisiae: an experimental system for repeated sequence evolution. Genetics 124:533-545.[Abstract]

LOUIS, E. J., E. S. NAUMOVA, A. LEE, G. NAUMOV, and J. E. HABER, 1994  The chromosome end in yeast: its mosaic natur