Checkpoint Proteins Influence Telomeric Silencing and Length Maintenance in Budding Yeast

Corresponding author: Maria Pia Longhese, Dipartimento di Biotecnologie e Bioscienze, Università degli Studi di Milano-Bicocca, P.zza della Scienza 2, 20126 Milano, Italy., mariapia.longhese{at}unimib.it (E-mail)

Communicating editor: A. G. HINNEBUSCH

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

A complex network of surveillance mechanisms, called checkpoints, interrupts cell cycle progression when damage to the genome is detected or when cells fail to complete DNA replication, thus ensuring genetic integrity. In budding yeast, components of the DNA damage checkpoint regulatory network include the RAD9, RAD17, RAD24, MEC3, DDC1, RAD53, and MEC1 genes that are proposed to be involved in different aspects of DNA metabolism. We provide evidence that some DNA damage checkpoint components play a role in maintaining telomere integrity. In fact, rad53 mutants specifically enhance repression of telomere-proximal transcription via the Sir-mediated pathway, suggesting that Rad53 might be required for proper chromatin structure at telomeres. Moreover, Rad53, Mec1, Ddc1, and Rad17 are necessary for telomere length maintenance, since mutations in all of these genes cause a decrease in telomere size. The telomeric shortening in rad53 and mec1 mutants is further enhanced in the absence of SIR genes, suggesting that Rad53/Mec1 and Sir proteins contribute to chromosome end protection by different pathways. The finding that telomere shortening, but not increased telomeric repression of gene expression in rad53 mutants, can be suppressed by increasing dNTP synthetic capacity in these strains suggests that transcriptional silencing and telomere integrity involve separable functions of Rad53.

IN eukaryotic cells, the consequences of DNA damage are minimized by the simultaneous activation of DNA repair mechanisms and signal transduction pathways, called checkpoints, which lead to a transient delay of cell cycle progression when alterations in DNA structure are detected. Checkpoint pathways ensure the proper order and timing of cell cycle events and contribute to the maintenance of genetic integrity as they increase the repair capacity of a damaged cell and prevent replication and segregation of damaged chromosomes (reviewed in HARTWELL and WEINERT 1989 ; PAULOVICH et al. 1997 ). In Saccharomyces cerevisiae, a number of genes have been identified that control the ability of cells to arrest the cell cycle and/or to activate the transcriptional response of DNA repair genes. Upstream components acting at the early steps in the DNA damage checkpoint regulatory network include the RAD9, RAD17, RAD24, MEC3, and DDC1 genes (reviewed in LONGHESE et al. 1998 ; WEINERT 1998 ). Conversely, the DNA replication proteins Pol, Dpb11, Rfc2, and Rfc5 appear to sense DNA alterations specifically during DNA synthesis, thus linking entry into mitosis to a proper completion of S phase (ARAKI et al. 1995 ; NAVAS et al. 1995 ; SUGIMOTO et al. 1996 , SUGIMOTO et al. 1997 ; NOSKOV et al. 1998 ). Once DNA alterations are sensed the signaling process involves a protein phosphorylation cascade propagated through the two protein kinases Mec1 and Rad53 (reviewed in LONGHESE et al. 1998 ; WEINERT 1998 ), indicating that these two gene products play a central role in the checkpoint signal transduction cascade. Rad53 is an essential Ser/Thr/Tyr protein kinase (ZHENG et al. 1993 ), whereas Mec1 belongs to the PI-3 kinase motif family, including Tel1 (GREENWELL et al. 1995 ; MORROW et al. 1995 ) and Schizosaccharomyces pombe Rad3 (BENTLEY et al. 1996 ), as well as human DNA-PK (DNA-dependent protein kinase) (JEGGO et al. 1995 ) and the ATM (ataxia-telangiectasia mutated gene) gene product (SAVITSKY et al. 1995 ). In response to DNA insults, Mec1 is required to phosphorylate and activate several substrates, including Rad53 (ZHENG et al. 1993 ; SANCHEZ et al. 1996 ), Ddc1 (PACIOTTI et al. 1998 ), and Rad9 (EMILI 1998 ; SUN et al. 1998 ; VIALARD et al. 1998 ), suggesting a pivotal role for Mec1 in the checkpoint signal transduction pathway. Moreover, Mec1 and Rad53 induce transcription of the genes encoding ribonucleotide reductase (RNR), which catalyzes the rate-limiting step in dNTP synthesis that is necessary for both replication and repair (ZHOU and ELLEDGE 1993 ; ALLEN et al. 1994 ).

Since DNA is packaged into chromatin, chromatin is the context in which any alteration in DNA structure is recognized by the checkpoint pathways. In S. cerevisiae, transcriptionally silent chromatin is found at telomeres and at the HML and HMR loci, encoding the cryptic mating-type cassettes (APARICIO et al. 1991 ). Transcription of genes located at telomeres and at HML or HMR loci is subject to reversible but mitotically inheritable repression (GOTTSCHLING et al. 1990 ; reviewed in LOO and RINE 1995 ). A number of proteins are required for telomeric repression. These include repressor activator protein 1 (Rap1; KYRION et al. 1993 ), the regulators Sir2-4 (APARICIO et al. 1991 ), and the N termini of histones H3 and H4 (KAYNE et al. 1988 ; THOMPSON et al. 1994 ), suggesting that these factors may contribute to establish a higher-order repressed state of chromatin. Associated with telomeres are also the yeast homologs of the DNA end-binding Ku proteins, yKu70 and yKu80. yKu70 was shown to bind to Sir4 by two-hybrid assay (TSUKAMOTO et al. 1997 ) and, similar to sir mutants, deletion of YKU70 or YKU80 causes the loss of telomere-adjacent gene silencing (BOULTON and JACKSON 1998 ; LAROCHE et al. 1998 ; NUGENT et al. 1998 ). While the absence of Sir proteins does not cause loss of Ku proteins from telomeres, the absence of Ku proteins results in loss of telomere-associated Sir proteins, indicating a prominent role of Ku proteins in heterochromatin assembly at telomeres. The investigation of the relationships between telomeres, chromatin, and checkpoints is crucial for the understanding of how chromatin affects DNA repair and checkpoint controls and how telomeres avoid being recognized as broken DNA ends. In addition to exerting profound influences on the transcription of adjacent genes, yeast telomeres are required for the maintenance of chromosome integrity. In fact, one of the essential functions of telomeres is to protect chromosomal termini from degradation and fusion that can lead to DNA changes and genomic instability. These protective structures also provide a mechanism to replicate chromosomal ends (reviewed in ZAKIAN 1995 ). Telomere length is kept within a narrow size range in a cell population and can be viewed as the result of a balance between elongation and shortening. Telomerase, a ribonucleoprotein complex responsible for the addition of DNA sequences to telomeres of eukaryotic chromosomes, is specifically required to maintain telomere length, preventing progressive shortening of chromosomal ends (reviewed in NUGENT and LUNDBLAD 1998 ).

Recent lines of evidence in budding and fission yeast suggest that proteins involved in DNA damage checkpoints have a role in telomere length maintenance. First, in fission yeast, DNA replication checkpoint mutants display altered telomeres (DAHLEN et al. 1998 ; MATSUURA et al. 1999 ), and mutations in rad3⁺, encoding the homologue of S. cerevisiae Mec1, and tel1⁺ cause dramatically shortened telomeres (NAITO et al. 1998 ), suggesting that telomere synthesis is at least partially dependent on both kinases. This appears to be a conserved mechanism, since in budding yeast, strains carrying mutations in both MEC1 and TEL1 undergo continual loss of telomeric repeats that is associated with a senescence phenotype (RITCHIE et al. 1999 ). Moreover, in budding yeast, Mec3 shows genetic and physical interaction with Set1, a protein required for transcriptional telomeric silencing and telomere integrity (CORDA et al. 1999 ). Mec3 and Set1 have antagonistic effects on both telomere length maintenance and repression of telomere-proximal transcription. Finally, Ku, which is involved in double-strand break repair and telomere maintenance, and Sir proteins residing at telomeres relocalize in response to DNA damage and this process is under the control of the DNA damage checkpoint pathways (MARTIN et al. 1999 ; MILLS et al. 1999 ). To further investigate the role of the checkpoint proteins in modulating chromatin and telomere structure, we studied the effects of mutations in the RAD53 and MEC1 genes and in other checkpoint components on telomere-directed transcriptional silencing and telomere length maintenance.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Strains and media:
All the oligonucleotides mentioned below and used for PCR amplification are reported in Table 1. The strains used in this study are listed in Table 2. Strains UCC3537, UCC1001, UCC3511, and UCC3515 were kindly provided by D. Gottschling (Fred Hutchinson Cancer Research Center, Seattle), and strains Y300 and Y300tel1 were kindly provided by S. Elledge (Howard Hughes Medical Institute, Houston, TX). Strains DMP2696/3D, DMP2696/4A, DMP2698/1D, and DMP2694/2C were derivatives of W303 (MATa or MAT ade2-1 trp1-1 leu2-3, 112 his3-11, 15, ura3; PACIOTTI et al. 1998 ). One-step replacement of RAD53 with the rad53K227A allele was carried out by transforming strains UCC3537, UCC1001, W303, Y300, Y300tel1, UCC3511, and UCC3515 with the EcoRI-digested pCH8 plasmid (PELLICIOLI et al. 1999 ), kindly provided by M. Foiani (University of Milan, Italy), to generate strains YLL410, YLL392, DMP2760/1A/DMP2760/3B, YLL678, YLL677, YLL416, and YLL422, respectively. Strains YLL430, DMP2840/3D, DMP2831/1B, and YLL439, in which the URA3 gene has been integrated into the telomeric region of chromosome VII, were generated by transforming strains W303, DMP2760/1A, DMP2696/3D, and DMP2698/1D, respectively, with SalI-EcoRI-digested pVII-L URA3-TEL plasmid, kindly provided by D. Gottschling. Strain DMP3141/2B is a meiotic segregant from a cross between strains YLL488 and DMP2831/1B. Strains DMP2952/2B and DMP2952/2C were meiotic segregants from a cross between strains YLL430 and DMP2854/2B. Strains DMP2955/4A and DMP2955/4C were derived from a cross between strains YLL430 and DMP2855/7C. To construct the SIR3 chromosomal deletion, sir3::HIS3 and sir3::KanMX4 cassettes were constructed by PCR using, respectively, pFA6a-HIS3 and pFA6a-kanMX4 plasmids (WACH et al. 1994 ) as templates and oligonucleotides PRP109 and PRP110 as primers. One-step replacement of 2866 bp of the SIR3 coding region with the Kluyveromyces lactis HIS3 cassette was carried out by transforming strains W303, YLL410, YLL392, DMP2760/1A, and DMP2696/3D with the sir3::HIS3 PCR product to give rise to strains YLL521, YLL465, YLL462, YLL522, and YLL525, respectively. One-step replacement of 2866 bp of the SIR3 coding region with the KanMX4 cassette was carried out by transforming strains UCC1001 and UCC3537 with the sir3::KanMX4 PCR product to give rise to strains YLL387 and YLL405, respectively. To construct the SIR4 chromosomal deletion, a sir4::HIS3 cassette was constructed by PCR using plasmid pFA6a-HIS3 as a template and oligonucleotides PRP105 and PRP106 as primers. One-step replacement of 3808 bp of the SIR4 coding region with the K. lactis HIS3 cassette was carried out by transforming strains W303, DMP2760/1A, and DMP2696/3D with the sir4::HIS3 PCR product to give rise to strains YLL541, YLL543, and YLL546, respectively. To generate the SML1 chromosomal deletion, a sml1::HIS3 cassette was constructed by PCR using pFA6a-HIS3 as a template and oligonucleotides PRP148 and PRP149 as primers. One-step replacement of 269 bp of the SML1 coding region with the K. lactis HIS3 cassette was carried out by transforming strains UCC1001, YLL392, DMP2760/1A, and DMP2696/3D with the sml1::HIS3 PCR product to give rise to strains YLL606, YLL607, YLL590, and YLL623, respectively. Strains YLL488 and DMP2947/1C were constructed by transforming, respectively, strains W303 and DMP2694/2C with the sml1::kanMX4 PCR product, obtained by PCR using pFA6a-kanMX4 as a template, and oligonucleotides PRP148 and PRP149 as primers. Strain DMP2950/3B was a meiotic segregant from a cross between strains YLL244 (LONGHESE et al. 1997 ) and DMP2818/1B. Strains DMP2928/9A and DMP2932/4D were meiotic segregants from a cross between strains DMP2760/1A and DMP2694/2C or YLL244, respectively. To construct the RAD53 chromosomal deletion, a rad53::HIS3 cassette was constructed by PCR using plasmid pFA6a-HIS3 as a template and oligonucleotides PRP158 and PRP159 as primers. One-step replacement of 2346 bp of the RAD53 coding region with the K. lactis HIS3 cassette was carried out by transforming strain YLL488 with the rad53::HIS3 PCR product to give rise to strain YLL509. To construct the MEC1 chromosomal deletion, a mec1::HIS3 cassette was constructed by PCR using plasmid pUC19His3⁺ as a template and oligonucleotides PRP64 and PRP65 as primers. One-step replacement of 6505 bp of the MEC1 coding region with the HIS3 cassette was carried out by transforming strain YLL488 with the mec1::HIS3 PCR product to give rise to strain YLL490. Strain DMP2854/2B was derived from a cross between strains W303 (MAT) and YLL490. Strain DMP2818/1B was a meiotic segregant from a cross between strains YLL488 and W303 (MAT). Strain DMP2855/7C was a meiotic segregant from a cross between strains W303 (MAT) and YLL509. To construct the DDC1 chromosomal deletion, a ddc1::kanMX4 cassette was constructed by PCR using plasmid pFA6a-kanMX4 as a template and oligonucleotides PRP21 and PRP22 as primers. One-step replacement of 1752 bp of the DDC1 coding region with the KanMX4 cassette was carried out by transforming strain UCC1001 with the ddc1::kanMX4 PCR product to give rise to strain YLL388. To construct the RAD17 chromosomal deletion, a rad17::kanMX4 cassette was constructed by PCR using plasmid pFA6a-kanMX4 as a template and oligonucleotides PRP119 and PRP120 as primers. One-step replacement of 1057 bp of the RAD17 coding region with the KanMX4 cassette was carried out by transforming strain UCC1001 with the rad17::kanMX4 PCR product to give rise to strain YLL389. To construct the RAD24 chromosomal deletion, a rad24::kanMX4 cassette was constructed by PCR using plasmid pFA6a-kanMX4 as a template and oligonucleotides PRP127 and PRP128 as primers. One-step replacement of 1917 bp of the RAD24 coding region with the KanMX4 cassette was carried out by transforming strain UCC1001 with the rad24::kanMX4 PCR product to give rise to strain YLL390. To construct the RAD9 chromosomal deletion, a rad9::kanMX4 cassette was constructed by PCR using plasmid pFA6a-kanMX4 as a template and oligonucleotides PRP131 and PRP132 as primers. One-step replacement of 3880 bp of the RAD9 coding region with the KanMX4 cassette was carried out by transforming strain UCC1001 with the rad9::kanMX4 PCR product to give rise to strain YLL424. Strain YLL409 was derived from strain UCC3537 by transformation with NdeI-digested pML54 plasmid DNA (LONGHESE et al. 1996 ) to obtain one-step replacement of the 877-bp PvuII-SacI fragment of the chromosomal MEC3 coding region with an EcoRI-BglII fragment containing the TRP1 gene. The accuracy of all gene replacements was verified by Southern blot analysis or PCR.

Strains W303, YLL521, YLL541, YLL244, DMP2760/1A, DMP2694/2C, and DMP2696/4A were transformed with a TRP1 empty vector (pBAD54; DESANY et al. 1998 ) giving rise to strains YLL719, YLL721, YLL723, YLL725, YLL727, YLL729, and YLL731, respectively. Strains W303, YLL521, YLL541, YLL244, DMP2760/1A, DMP2694/2C, and DMP2696/4A were transformed with a TRP1 plasmid carrying the GAP-controlled RNR1 gene (pBAD70; DESANY et al. 1998 ), giving rise to strains YLL720, YLL722, YLL724, YLL726, YLL728, YLL730, and YLL732, respectively.

Cells were grown at 25° in YEP medium (1% yeast extract, 2% bactopeptone) containing 2% glucose (YEPD), 2% raffinose (YP-raf), or 2% raffinose and 2% galactose (YP-gal). Transformants carrying the KanMX4 cassette were selected on YEPD plates containing 400 µg/ml G418 (U.S. Biological).

Plasmids:
To construct plasmid pML225 (URA3 CEN4 GAL1-MEC1), where the 7437-bp fragment spanning from the MEC1 ATG to the SacI site and containing the whole MEC1 coding region is fused to the GAL1 promoter, the 1302-bp XbaI-BamHI MEC1 fragment from plasmid pML79 (LONGHESE et al. 1997 ) was cloned into XbaI-BamHI sites of plasmid SP1 (S. Piatti, University of Milan, Italy), giving rise to plasmid pML195; plasmid SP1 carried the EcoRI-BamHI fragment containing the GAL1-10 promoter cloned in the EcoRI-BamHI sites within the YCplac33 polylinker region (GIETZ and SUGINO 1988 ). A MEC1 fragment spanning from position +1 to position +99 from the translation initiation codon was then amplified by PCR using plasmid pML79 as a template and oligonucleotides PRP134 and PRP75 as primers and then cloned into the BamHI site of plasmid pML195, followed by insertion of the 6049-bp AatII-SacI fragment of MEC1 coding region. Plasmids pBAD54 and pBAD70 containing, respectively, the GAP promoter and the RNR1 open reading frame under the control of the GAP promoter were kindly provided by S. Elledge.

Southern analysis of telomere length:
Yeast DNA was prepared according to standard methods (GUTHRIE and FINK 1991 ) and digested with an appropriate enzyme (XhoI or EcoRV). The resulting DNA fragments were separated by gel electrophoresis in 0.8% agarose gel and transferred to a GeneScreen nylon membrane (New England Nuclear, Boston) followed by hybridization with a poly(GT) probe or with a 1166-bp HindIII URA3 fragment purified from plasmid YEp24. Standard hybridization conditions were used.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Mutations in RAD53 gene increase transcriptional silencing at telomeres:
We first assayed the effect of a rad53 mutation on the expression of the ADE2 reporter gene integrated into a telomeric region (GOTTSCHLING et al. 1990 ). Yeast cells that do not express ADE2 form red-pigmented colonies on medium containing low levels of adenine, whereas white colonies result from cells expressing the ADE2 gene. When the ADE2 gene was placed at the telomere such that ADE2 transcription was directed toward the telomere, cells developed red-sectored colonies due to heritable Sir-dependent transcriptional silencing (GOTTSCHLING et al. 1990 ; APARICIO et al. 1991 ). We introduced the kinase-defective rad53K227A allele (ZHENG et al. 1993 ) in strain UCC-3537, carrying the only copy of the ADE2 gene inserted into the telomeric region of chromosome V-R (HUANG et al. 1997 ) and scored for the colony-developed color. As shown in Fig 1, most wild-type colonies were completely white or red-sectored, suggesting that the ADE2 gene was expressed in most cells. Conversely, a large proportion of the rad53 mutant colonies were either completely red or contained few white sectors, indicating that the telomere-associated ADE2 gene was transcriptionally silent (Fig 1). According to previous observations, mec3 also increased transcriptional silencing at telomeres compared to wild type (Fig 1; CORDA et al. 1999 ).

View larger version (46K): In this window In a new window Download PPT slide   Figure 1. Telomeric position effect on ADE2 expression in rad53 mutants. Isogenic wild type (UCC3537), sir3 (YLL405), rad53K227A (YLL410), mec3 (YLL409), and rad53K227A sir3 (YLL465) cells, carrying the ADE2 gene at telomere V-R, were streaked onto YEPD plates. After incubation at 25° for 3 days, the plates were placed at 4° for 1 wk before being photographed. Comparable results were obtained in three independent experiments.

The telomeric position effect (TPE) is dependent on the known Rap1-interacting proteins Sir3 and Sir4, and null mutations of SIR3 or SIR4 abolish silencing at telomeres (IVY et al. 1986 ; APARICIO et al. 1991 ). To determine whether the increase in silencing caused by the rad53 mutation was dependent on the function of the Sir proteins and not on a general bypass of the silencing machinery, we examined the silencing defects in sir3 rad53K227A double mutants by using the above ADE2 system. As shown in Fig 1, both sir3 rad53K227A double and sir3 single mutants developed white colonies and were indistinguishable from each other, indicating that the enhanced silencing in the rad53K227A mutants is still dependent on the normal machinery that controls repression of gene expression at telomeres.

To further assess the enhanced transcriptional silencing at telomeres in rad53 mutants, we assayed the telomere position effect in strains in which the only copy of the URA3 gene (URA3-TEL) had been integrated into a telomeric region (GOTTSCHLING et al. 1990 ; APARICIO et al. 1991 ). Repression of URA3-TEL expression was then measured by the ability of cells to grow on media lacking uracil and to form colonies on 5-fluoroorotic acid (5-FOA), which kills cells expressing the URA3 gene. We introduced the rad53K227A mutation in strain UCC1001, carrying URA3 gene inserted next to the left telomere of chromosome VII (GOTTSCHLING 1992 ). As expected, the sir3 mutant was unable to form colonies on 5-FOA, indicating derepression of the URA3-TEL gene (Fig 2A). Conversely, we reproducibly observed a consistent increase in URA3-TEL silencing in the rad53K227A mutant compared to wild type, as indicated by both the increase in the frequency of 5-FOA-resistant clones and the decrease in the number of cells growing on media lacking uracil (Fig 2A). The enhanced repression of URA3-TEL in rad53 mutants requires the Sir-mediated pathway. In fact, the increased repression observed in rad53K227A mutation was abolished by deletion of SIR3 gene, since the ability to grow on 5-FOA of rad53K227A sir3 double mutants was similar to that observed for an otherwise isogenic sir3 single mutant strain (Fig 2A).

View larger version (70K): In this window In a new window Download PPT slide   Figure 2. rad53 mutations increase silencing at telomeres. Isogenic strains carrying the URA3 gene at telomere VII-L were grown overnight in rich (YEPD) medium, and then 10-fold serial dilution was spotted onto complete synthetic medium (SC) and onto the same medium lacking uracil (SC-Ura) or supplied with 5-FOA to assay for expression of the URA3-TEL gene. Isogenic strains were as follows: (A) wild type (UCC1001), sml1 (YLL606), rad53K227A (YLL392), rad53K227A sml1 (YLL607), sir3 (YLL387), and rad53K227A sir3 (YLL462). (B) Wild type (YLL430), rad53-K227A (DMP2840/3D), sad1-1 (YLL439), rad53 sml1 (DMP-2955/4A), and sml1 (DMP2955/4C). Comparable results were obtained in three independent experiments.

To generalize our results and to assess whether the transcriptional silencing defects were a peculiarity of the rad53K227A allele or a common feature of rad53 mutants, we analyzed the transcriptional silencing defects of different mutations in the RAD53 gene in the W303 background, which is different from the UCC1001 genetic background, by integrating the URA3 gene into the telomeric region of chromosome VII (GOTTSCHLING et al. 1990 ) in a W303 derivative strain. As shown in Fig 2B, when such rad53K227A, sad1-1 (a RAD53 mutation described by ALLEN et al. 1994 ), and rad53 mutants were analyzed for the ability to express the URA3-TEL gene, we detected an enhanced telomere-directed transcriptional silencing in all of them. Therefore, different mutations in the RAD53 gene, including its deletion, improve telomeric silencing, suggesting that Rad53 might modulate the accessibility to repressive chromatin in the telomeric silent domains.

The involvement of Rad53 in silencing prompted us to analyze the effect of mutations in MEC1 on transcriptional silencing. Very recently it was reported that the mec1-21 allele specifically decreases telomeric silencing (CRAVEN and PETES 2000 ). As shown in Fig 3A, we did not detect any transcriptional silencing defect in strains carrying either the deletion of the MEC1 gene or the mec1-14 allele (PACIOTTI et al. 1998 ). The apparent slight growth defect of mec1-14 cells on media lacking uracil was likely due to the slow-growth phenotype associated with this allele, which was suppressed by deletion of SML1 gene. Since mec3 cells enhance transcriptional silencing at telomeres (Fig 1; CORDA et al. 1999 ), we also analyzed the silencing defects in strains lacking DDC1, RAD24, RAD17, and RAD9. As shown in Fig 3B, ddc1, rad17, rad24, and rad9 checkpoint mutants did not show any silencing defects, thus implying a functional difference between these genes and MEC3.

View larger version (70K): In this window In a new window Download PPT slide   Figure 3. Silencing assay in mec1, ddc1, rad17, rad24, and rad9 mutants. Isogenic strains carrying the URA3 gene at telomere VII-L were grown overnight in rich (YEPD) medium, and then 10-fold serial dilution was spotted onto complete synthetic medium (SC) and onto the same medium lacking uracil (SC-Ura) or supplied with 5-FOA to assay for expression of the URA3-TEL gene. Isogenic strains were as follows: (A) wild type (YLL430), mec1-14 (DMP2831/1B), mec1-14 sml1 (DMP3141/2B), mec1 sml1 (DMP2952/2B), and sml1 (DMP2952/2C). (B) Wild type (UCC1001), ddc1 (YLL388), rad17 (YLL389), rad24 (YLL390), and rad9 (YLL424).

rad53 mutations do not affect transcriptional silencing at mating-type loci:
Since several regulatory proteins function in silencing at both telomeres and HM loci, we tested whether rad53 mutations affected silencing also at the mating-type loci by introducing the rad53K227A allele in strains UCC3511 and UCC3515 containing the URA3 gene inserted at the HMR and HML loci, respectively (SINGER et al. 1998 ). Silencing at both HMR and HML loci was not enhanced in rad53 mutants, since the rad53K227A mutation did not affect the ability of cells to grow on media lacking uracil or in the presence of 5-FOA compared to the isogenic wild-type strain (Fig 4). These data suggest that rad53 mutants have primarily a telomere-specific silencing defect.

View larger version (51K): In this window In a new window Download PPT slide   Figure 4. Silencing at mating-type loci. Isogenic strains carrying the URA3 gene inserted into the HML or HMR locus were grown overnight in rich (YEPD) medium, and then 10-fold serial dilution was spotted onto complete synthetic medium (SC) and onto the same medium lacking uracil (SC-Ura) or supplied with 5-FOA to assay for expression of the URA3 gene. Isogenic strains were as follows: hmr::URA3 (UCC3511), hmr::URA3 rad53K227A (YLL416), hml::URA3 (UCC3515), and hml::URA3 rad53K227A (YLL422). The more efficient silencing of the URA3 gene at HML than at HMR has been previously reported (SINGER et al. 1998 ) and depends on the difference in the way the URA3 gene was inserted within the two HM loci.

Rad53 and Mec1 are involved in telomere length maintenance:
Many mutations that influence telomeric silencing also influence telomere length. For example, sir3 and sir4 mutants, where telomeric silencing is disrupted, have telomeric repeats that are 50–100 bp shorter than wild type (PALLADINO et al. 1993 ). Since RAD53 gene modulates TPE, we asked whether it also has a role in telomere length regulation. To this purpose, we first examined the effect of the rad53K227A mutation on the length of the left telomere of chromosome VII in UCC1001 strain, carrying the URA3-TEL gene as described above. As shown in Fig 5, Southern blot analysis with a URA3 probe revealed that telomeres were shorter in rad53K227A mutant than in isogenic wild-type cells, and this telomere shortening was comparable to that observed for sir3 mutants. The rad53 K227A sir3 double mutants showed further decrease in telomere length compared to both rad53K227A and sir3 single mutants (Fig 5), suggesting that Rad53 and Sir3 proteins contribute to chromosome end protection by different pathways.

View larger version (71K): In this window In a new window Download PPT slide   Figure 5. rad53 deficient strains display defects in telomere VII-L length. Genomic DNA from wild type (UCC1001), rad53K227A (YLL392; two independent transformants), rad53K227A sir3 (YLL462; two independent transformants), and sir3 (YLL387) isogenic strains, all carrying a subtelomeric URA3 gene on chromosome VII-L, was prepared after 45 generations of growth, digested with EcoRV, separated on a 0.8% agarose gel, and hybridized to an URA3 probe.

To confirm the requirement for Rad53 function in telomere length maintenance and to analyze the effect of mutations in the MEC1 gene on the same process, we examined the effects of rad53 and mec1 mutations in W303 genetic background by measuring the length of the telomeric (C_1–3A)_n repeat. To this end, we introduced the mutations under analysis in the W303 strain and analyzed the genomic DNA recovered from each strain by Southern hybridization analysis with a poly (GT) probe. As shown in Fig 6A, the rad53K227A mutant showed telomere shortening, thus confirming the above indication that Rad53 contributes to telomere integrity. Moreover, not only rad53K227A sir3, but also rad53K227A sir4 double mutants showed a more pronounced telomere shortening compared to each single mutant. According to what was previously observed in the mec1-21 mutant (RITCHIE et al. 1999 ), a comparable telomere length defect was also detectable in the mec1-14 mutant. In fact, as shown in Fig 6B, mec1-14 mutant cells displayed telomere shortening that was further enhanced when the mec1-14 allele was combined with the deletion of the SIR3 or SIR4 genes. Therefore, both Rad53 and Mec1 proteins are required for controlling telomere length by a mechanism that appears to be different from that involving the Sir3 and Sir4 proteins. We also analyzed telomere length in a rad53K227A mec1-14 double mutant and we found that it was undistinguishable from that observed in the most defective single mutant (rad53K227A; data not shown), suggesting that rad53 and mec1 mutations impair telomere length control through the same mechanism.

View larger version (56K): In this window In a new window Download PPT slide   Figure 6. Telomere length defects in rad53 and mec1 deficient strains. Genomic DNA from each cell culture was prepared after 45 generations of growth, digested with XhoI, separated on a 0.8% agarose gel, and hybridized to a poly(GT) telomere-specific probe. Isogenic strains were as follows: (A) wild type (W303), rad53K227A (DMP2760/1A; DMP2760/3B), sir3 (YLL521), rad53K227A sir3 (YLL522), sir4 (YLL541), rad53K227A sir4 (YLL543), sml1 (YLL488), rad53K227A sml1 (YLL590), and rad53 sml1 (YLL509). (B) Wild type (W303), mec1-14 (DMP2696/3D; DMP2696/4A), sir3 (YLL521), mec1-14 sir3 (YLL525), mec1-14 sml1 (YLL623), sir4 (YLL541), mec1-14 sir4 (YLL546), sml1 (YLL488), and mec1 sml1 (YLL490).

It has been previously observed that strains carrying mutations in both MEC1 and TEL1 genes exhibit a senescence phenotype and a more pronounced shortening of telomeres compared to each single mutant (RITCHIE et al. 1999 ), suggesting that Tel1 and Mec1 contribute to telomere length control by different pathways. Since we showed that Rad53, together with Mec1, is necessary to maintain telomere integrity, we analyzed the telomere length defect in the rad53 tel1 double mutants. To this purpose, we introduced the kinase-defective rad53K227A allele in a strain carrying the deletion of TEL1 gene. As shown in Fig 7, tel1 rad53K227A double mutants exhibited a telomere shortening indistinguishable from that observed in a tel1 single mutant, suggesting that the rad53K227A mutation was not able to further decrease telomere length in the absence of Tel1. Moreover, in contrast to what was observed in tel1 mec1 double mutants (RITCHIE et al. 1999 ; our unpublished observation), deletion of the TEL1 gene did not affect the growth rate of rad53K227A mutants (data not shown).

View larger version (65K): In this window In a new window Download PPT slide   Figure 7. Genetic interaction between TEL1 and RAD53. Genomic DNA from each cell culture was prepared after 45 generations of growth, digested with XhoI and hybridized to a poly(GT) telomere-specific probe. Isogenic strains were as follows: wild type (Y300), rad53K227A (YLL678), tel1 (Y300tel1; two independent transformants), and rad53K227A tel1 (YLL677; three independent transformants).

Telomere length and transcriptional silencing controls are separable functions of Rad53:
Telomere shortening usually has been found associated with reduced TPE, whereas abnormally long telomeres can hyper-repress telomere-adjacent genes (KYRION et al. 1992 ). Since the rad53 mutant displayed increased telomeric silencing concomitantly with decreased telomere length, the enhanced TPE observed in this mutant could be unrelated to the telomere length defect. In addition to its involvement in the DNA damage checkpoint pathway, Rad53, together with Mec1, is essential for cell viability. Their essential function can be bypassed by increasing expression of genes encoding ribonucleotide reductase (DESANY et al. 1998 ) or by deleting the SML1 gene (ZHAO et al. 1998 ), which negatively affects dNTP pools likely through post-translational regulation of ribonucleotide reductase activity. On the basis of the above observations, the telomere shortening observed in rad53 mutants might be caused by defective telomere synthesis in the presence of a reduced dNTP pool. If this were the case, we would expect to suppress the telomere length defects in rad53K227A and mec1-14 mutants by deleting the SML1 gene. As shown in Fig 6A and Fig B, telomere length in sml1 rad53K227A and sml1 mec1-14 double mutants was comparable to that observed in a wild-type strain and was indistinguishable from that observed in a sml1 single mutant. Therefore, telomere shortening in rad53K227A and mec1-14 single mutants might result from defective DNA replication caused by nucleotide depletion. This hypothesis is further supported by the finding that strains carrying deletions of the RAD53 or MEC1 genes, whose viability depends on the lack of Sml1, did not show any telomere shortening (Fig 6A and Fig B). Deletion of SML1 did not affect transcriptional silencing in rad53 mutants. In fact, as shown in Fig 2A and Fig B, both rad53 sml1 and the rad53K227A sml1 double mutants showed an increase in the frequency of 5-FOA-resistant cells and a decrease in the number of cells growing on media lacking uracil, compared to the wild type, which were similar to those observed in rad53K227A single mutants. This finding suggests that the role of Rad53 in transcriptional silencing can be separated from that in telomere size control, implying distinct functions of Rad53 in modulating these processes.

Effects of other DNA damage checkpoint mutants on telomere length:
We also asked whether other DNA damage checkpoint mutants displayed defects in telomere length regulation. To this purpose, we examined the telomere length in ddc1, rad17, rad24, and rad9 strains. Southern blot analysis using the telomere probe showed that ddc1 and rad17 mutations caused shortening of telomeres, although we reproducibly found that the extent of shortening was less than that caused by rad53 and mec1 mutations (Fig 8A and Fig B). The combination of ddc1 with rad17 did not enhance the telomere shortening (data not shown), suggesting that Ddc1 and Rad17 proteins belong to the same epistatic group not only with respect to their involvement in DNA damage checkpoint response but also with respect to their role in chromosome end protection. Conversely, we did not detect any reproducible telomere shortening in rad9 and rad24 strains (data not shown). In contrast to what was observed for the rad53 and mec1 mutants, the telomere length defects in ddc1 and rad17 mutants were not suppressed by deleting the SML1 gene, since rad17 sml1 and ddc1 sml1 double mutants still showed telomere shortening (Fig 8A and Fig B). These data suggest that the role of Ddc1 and Rad17 in telomere length control is different from that of Rad53 and Mec1. However, the telomere shortening was not more severe by combining rad17 and ddc1 mutations with the rad53K227A allele (Fig 8A and Fig B), suggesting that Rad17/Ddc1 and Rad53/Mec1 belong to the same epistatic group with respect to telomere length control.

View larger version (88K): In this window In a new window Download PPT slide   Figure 8. Telomere length in ddc1 and rad17 mutants. Genomic DNA from each cell culture was prepared after 45 generations of growth, digested with XhoI, and hybridized to a poly(GT) telomere-specific probe. Isogenic strains were as follows: (A) wild type (W303), ddc1 (YLL244), ddc1 sml1 (DMP2950/3B), sml1 (YLL488), rad53K227A (DMP2760/1A), and rad53K227A ddc1 (DMP2932/4D). (B) Wild type (W303), rad17 (YLL244), rad17 sml1 (DMP2950/3B), sml1 (YLL488), rad53K227A (DMP2760/1A), and rad53K227A rad17 (DMP2928/9A).

RNR1 overexpression suppresses the telomere shortening in rad53 and mec1 mutants:
To provide further evidence that telomere shortening in mec1 and rad53 mutants is caused by a defective DNA replication caused by nucleotide depletion, we tested whether the telomere length defects might be suppressed by providing additional nucleotides through upregulation of ribonucleotide reductase activity. To this purpose, we analyzed whether overexpression of RNR1 gene, encoding the large subunit of ribonucleotide reductase (RNR), the rate-limiting enzyme of deoxyribonucleotide synthesis, could suppress the telomere length shortening in rad53 and mec1 mutants. As shown in Fig 9, rad53K227A and mec1-14 mutants overexpressing RNR1 gene did not show any telomere shortening, while ddc1 and rad17 mutants still showed telomere length defects. According to the finding that Rad53/Mec1 and Sir proteins contribute to telomere end protection by different pathways, overexpression of RNR1 did not suppress the telomere shortening in sir3 and sir4 mutants.

View larger version (118K): In this window In a new window Download PPT slide   Figure 9. Effect of RNR1 overexpression on telomere length control. Strains were transformed with a TRP1 plasmid carrying GAP-controlled RNR1 (pBAD70) or empty vector (pBAD54; DESANY et al. 1998 ). Cell cultures were grown in SC-Trp before preparing DNA. DNA was digested with XhoI and hybridized to a poly(GT) telomere-specific probe. Isogenic strains were as follows: wild type (YLL719), wild type [pGAP-RNR1] (YLL720), rad53K227A (YLL727), rad53K227A [pGAP-RNR1] (YLL728), mec1-14 (YLL731), mec1-14 [pGAP-RNR1] (YLL732), rad17 (YLL729), rad17 [pGAP-RNR1] (YLL730), ddc1 (YLL725), ddc1 [pGAP-RNR1] (YLL726), sir3 (YLL721), sir3 [pGAP-RNR1] (YLL722), sir4 (YLL723), and sir4 [pGAP-RNR1] (YLL724).

Overexpression of the MEC1 gene affects telomere length independently of SML1:
Since mutations in genes affecting telomere length often display the same phenotype observed when the corresponding genes are overexpressed (APARICIO et al. 1991 ; SINGER et al. 1998 ), we examined telomere length in strains carrying the MEC1 and RAD53 genes expressed from the GAL1 promoter. As shown in Fig 10, we observed telomere shortening in strains overexpressing the MEC1 gene compared to wild-type cells. Since the lack of Sml1 suppressed the telomere shortening in mec1 mutants, we then analyzed the effect of MEC1 overexpression in an sml1 background. Surprisingly, MEC1 overexpression also reduced telomere length in sml1 strains, indicating that telomere shortening caused by Mec1 overproduction occurs independently of the presence of Sml1. Conversely, overexpression of the RAD53 gene did not affect telomere integrity in either wild-type or sml1 strains (data not shown).

View larger version (50K): In this window In a new window Download PPT slide   Figure 10. Overexpression of the MEC1 gene causes telomere shortening. Cell cultures logarithmically growing in YP-raf were transferred to YP-gal (1 x 104 cells/ml) and allowed to reach the final concentration of 2 x 107 cells/ml before preparing genomic DNA. DNA was digested with XhoI and hybridized to a poly(GT) telomere-specific probe. Isogenic strains were as follows: wild type (YLL634; two independent transformants), GAL-MEC1 (YLL632; two independent transformants), sml1 (YLL640; two independent transformants), and sml1 GAL-MEC1 (YLL638; two independent transformants).

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

DNA strand breaks existing within telomeres at the end of linear eukaryotic chromosomes are not normally recognized as DNA damage. One way in which new masking may be achieved is through telomeric DNA being sequestered into a unique type of chromatin organization. In fact, yeast telomeres contain tandem arrays of the repeated sequence C_1–3A, to which a number of telomere-associated proteins critical for their replication and maintenance are bound.

In this study we analyzed the role of the checkpoint genes in maintaining telomere integrity. Different lines of evidence implicate DNA damage checkpoint components in chromosome end protection. Mutations in RAD53 and MEC1 genes affect the length of the telomere repeat sequences. One question is how a defective Rad53 or Mec1 alters telomere length control. MEC1 and RAD53 genes are essential for cell viability, and cell lethality in mec1 and rad53 cells is rescued by upregulation of ribonucleotide reductase activity or by deletion of the SML1 gene that causes an increase in dNTP synthetic capacity, which is necessary for efficient DNA replication and repair. This suggests that the lethality of mec1 and rad53 null mutants is due to accumulation of dNTP precursors insufficient to support DNA replication. Telomere length is maintained by a balance between opposing processes such as telomere polymerization and shortening, the latter of which might involve lagging-strand DNA synthesis and an exonuclease activity (WELLINGER et al. 1996 ). Since the deletion of the SML1 gene and upregulation of the ribonucleotide reductase activity suppress the telomere length defects observed in rad53 and mec1 mutants, telomere shortening observed in these mutants might be caused by a defective DNA replication at telomeric ends. In this view, the loss of Rad53 and Mec1 might decrease dNTP pools, which might result in a defective DNA replication at telomeres. Rad53 and Mec1 seem to be required to support telomere replication by the same mechanism since the telomere length defect of the rad53K227A mec1-14 double mutant is indistinguishable from the most defective single mutant (rad53K227A). In addition, we show that Ddc1 and Rad17 checkpoint proteins are required to maintain wild-type telomere structure. In fact, the lack of either Ddc1 or Rad17, inferred from genetic studies to operate in one pathway and to interact biochemically (PACIOTTI et al. 1998 ; KONDO et al. 1999 ), causes telomere shortening. The requirement for Rad17 in telomere metabolism appears to be conserved during evolution, since deletion of the S. pombe rad1⁺ gene, encoding the homologue of S. cerevisiae Rad17, also causes telomere shortening (DAHLEN et al. 1998 ; MATSUURA et al. 1999 ). In contrast to what was observed in rad53 and mec1 mutants, the lack of Sml1 does not suppress the telomere length defects of rad17 and ddc1 mutants, suggesting that the absence of Rad17 and Ddc1 affects the telomere ends by a mechanism different from that impaired by the rad53 and mec1 mutations. However, combination of the rad53 allele with rad17 or ddc1 does not further decrease telomere length, suggesting that Rad17/Ddc1 and Rad53/Mec1 function in the same pathway controlling telomere length, although their roles in maintaining telomere integrity might be different.

The evolutionary conservation of the checkpoint pathways raises the possibility that the checkpoint components play similar roles in other organisms. Major similarities between the checkpoint pathways include structural similarities between S. cerevisiae Mec1 and S. pombe Rad3 and between S. cerevisiae Rad53 and S. pombe Cds1. Like the S. cerevisiae RAD17, the telomere shortening of S. pombe rad1 mutants is not further affected by combination of rad1 with rad3 or with other mutants affecting the DNA damage checkpoint pathway. In contrast to what was observed for rad53 mutants, S. pombe cds1 cells apparently are not affected in telomere structure (MATSUURA et al. 1999 ). This difference might not be due necessarily to a different role of the two proteins in modulating telomere length, but might be explained by the way dNTP synthesis occurs in the two yeasts. In fact, unlike the S. cerevisiae MEC1 and RAD53, the S. pombe rad3⁺ and cds1⁺ genes are not essential for cell viability, suggesting that the nucleotide levels in rad3 and cds1 mutants are not rate limiting for some vital process like DNA replication. However, S. pombe rad3 still shows telomere shortening, which is further increased by deleting the Tel1⁺ gene (NAITO et al. 1998 ; MATSUURA et al. 1999 ), suggesting a specific role for Rad3 in maintaining telomere integrity. Similarly, in S. cerevisiae, tel1 mec1 double mutants show a more pronounced reduction in telomere length compared to each single mutant (RITCHIE et al. 1999 ), while inactivation of Rad53 does not enhance the telomere shortening in tel1 strains. This suggests that S. pombe Rad3, and possibly its S. cerevisiae counterpart Mec1, might have additional functions in controlling telomere integrity compared to the Cds1 and Rad53 kinases. Interestingly, overexpression of MEC1 leads to telomere shortening in both wild-type and sml1 background, indicating that MEC1 overexpression affects telomere integrity by a mechanism different from that impaired by its loss of function. Since reduction in telomere length can be achieved by reducing the rate of telomere elongation or by increasing the rate of telomere degradation, an excess of Mec1 might affect the balance between the rate of telomere elongation and shortening, altering the activity of components required to maintain telomere integrity or their accessibility to the telomeric sequences. Mec1 shows homology with Tel1, whose overexpression suppresses the DNA damage sensitivity of mec1 mutants, suggesting that the two proteins may have partially overlapping functions. In this view, overproduction of Mec1 might affect telomere length by competing with Tel1 and/or altering its activity in phosphorylating target proteins involved in maintaining telomere structure.

rad53 mutants specifically increase repression of the telomere-adjacent gene expression, suggesting that Rad53 might be involved in the relief of silencing at telomeres. The finding that telomere shortening, but not increased telomeric repression in rad53 mutants, is suppressed by deleting the SML1 gene suggests that the role of Rad53 in controlling telomere length might be separable from its role in modulating chromatin structure. If a defective DNA synthetic capacity likely accounts for the telomere shortening in rad53 mutants, several possible explanations might be proposed for the increased transcriptional repression of telomere-proximal genes observed in these mutants. For example, it has been suggested that the establishment of silencing can be enhanced by slowing down cell cycle progression (LAMAN et al. 1995 ). Since the rad53 sml1 mutant exhibits a slow-growth phenotype, it is possible that the enhanced transcriptional silencing might be caused by a defective cell cycle progression. However, many DNA replication mutants are unable to restore repression of a silencing-deficient HMR-E silencer, suggesting that the slowing of cell cycle progression alone is not sufficient to reestablish silencing (EHRENHOFER-MURRAY et al. 1999 ). In fact, only mutations in a restricted set of replication proteins that have been implicated also in the S phase checkpoint response, such as proliferating cell nuclear antigen (PCNA), RF-C, DNA polymerase , CDC45, CDC7, restore silencing at HM defective loci (AXELROD and RINE 1991 ; EHRENHOFER-MURRAY et al. 1999 ). Therefore, the enhanced reestablishment of silencing appears to be related to specific defects in DNA metabolism.

In addition to rad53 mutants, mec3 null mutants also show altered silencing. In fact, although to a lesser extent than rad53 mutants, mec3 cells enhance transcriptional silencing at telomeres (CORDA et al. 1999 ; Fig 1). Conversely, the mec1-14, rad17, ddc1, rad24, and rad9 checkpoint mutants do not show any increase in telomeric repression. This would suggest that, among the analyzed checkpoint proteins, only Rad53 and Mec3 are involved in modulating the accessibility to telomeric sequences. However, recent observations indicate that the mec1-21 allele causes a decrease in telomeric silencing, suggesting that Mec1 has a role in telomeric position effect that is detectable only in the presence of specific mutations (CRAVEN and PETES 2000 ).

The finding that Mec3 shows genetic and physical interaction with Set1, a protein required for transcriptional telomeric silencing (CORDA et al. 1999 ), might account for the involvement of Mec3 in modulating chromatin accessibility. Whether Rad53 directly regulates chromatin structure remains an open question. The finding that deletion of the SML1 gene does not suppress the increased telomere silencing of rad53 mutants suggests that Rad53 has additional functions compared to Mec1 and to the other DNA damage checkpoint components. This hypothesis is also supported by the observation that SML1 deletion restores viability in mec1 and rad53 strains differently. In fact, while mec1 sml1 double mutants grow as well as the wild-type cells, rad53 sml1 double mutants still show severe growth defects (ZHAO et al. 1998 ; our unpublished observations), implying additional functions for Rad53 in cell metabolism. Interestingly, the enhanced repression observed in rad53 mutants is specific for the telomeric sequences. On the basis of these observations, it is tempting to speculate that Rad53 could be specifically recruited to telomeric transcriptionally inactive chromatin, where it might modulate the accessibility of chromatin by influencing the activities of proteins specifically required to establish silencing at telomeres, possibly through phosphorylation events. Since some checkpoint functions may be activated during normal DNA replication (LONGHESE et al. 1997 ; PACIOTTI et al. 1998 ), it is reasonable that the checkpoint pathway genes affect telomeres even in an unperturbed cell cycle. In this scenario, in rad53 mutant Ku and/or Sir proteins might be more tightly bound to telomeres, resulting in more efficient telomeric silencing. The previous finding that Ku and Sir proteins are released from telomere in response to DNA double-strand breaks in a checkpoint-dependent manner further supports this hypothesis (MARTIN et al. 1999 ; MILLS et al. 1999 ).

Altogether, our observations that gene products implicated in DNA damage checkpoint response in S. cerevisiae are also required for the integrity of yeast telomere ends and chromatin structure provides further evidence that checkpoint mechanisms, telomeres, and chromatin are connected by complex relationships, whose elucidation will be a challenging subject for future work.

	ACKNOWLEDGMENTS

We thank S. Elledge, D. Gottschling, and M. Foiani for gifts of strains and plasmids, S. Piatti for critical reading of the manuscript, V. Geli, E. Gilson, and all the members of our laboratory for useful discussions and criticisms. This work was supported by grants from Associazione Italiana Ricerca sul Cancro and Cofinanziamento 1997 MURST-Università di Milano to G.L. and by CNR Target Project on Biotechnology Grant CT.97.01180.PF49(F). V.P. was supported by a fellowship from Fondazione Italiana per la Ricerca sul Cancro.

Manuscript received December 7, 1999; Accepted for publication April 17, 2000.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

ALLEN, J. B., Z. ZHOU, W. SIEDE, E. C. FRIEDBERG, and S. J. ELLEDGE, 1994  The SAD1/RAD53 protein kinase controls multiple checkpoints and DNA damage-induced transcription in yeast. Genes Dev. 8:2416-2428[Abstract/Free Full Text].

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].

ARAKI, H., S. H. LEEM, A. PHONGDARA, and A. SUGINO, 1995  Dpb11, which interacts with DNA polymeraseII () in Saccharomyces cerevisiae, has a dual role in S-phase progression and at a cell cycle checkpoint. Proc. Natl. Acad. Sci. USA 92:11791-11795[Abstract/Free Full Text].

AXELROD, A. and J. RINE, 1991  A role for CDC7 in repression of transcription at the silent mating-type locus HMR in Saccharomyces cerevisiae.. Mol. Cell. Biol. 11:1080-1091[Abstract/Free Full Text].

BENTLEY, N. J., D. A. HOLTZMAN, G. FLAGGS, K. S. KEEGAN, and A. DEMAGGIO et al., 1996  The Schizosaccharomyces pombe rad3 checkpoint gene. EMBO J. 15:6641-6651[Medline].

BOULTON, S. J. and S. P. JACKSON, 1998  Components of the Ku-dependent non-homologous end-joining pathway are involved in telomeric length maintenance and telomeric silencing. EMBO J. 17:1819-1828[Medline].

CORDA, Y., V. SCHRAMKE, M. P. LONGHESE, T. SMOKVINA, and V. PACIOTTI et al., 1999  Interaction between Set1p and checkpoint protein Mec3p in DNA repair and telomere functions. Nat. Genet. 21:204-208[Medline].

CRAVEN, R. J. and T. D. PETES, 2000  Involvement of the checkpoint protein Mec1p in silencing of gene expression at telomeres in Saccharomyces cerevisiae.. Mol. Cell. Biol. 20:2378-2384[Abstract/Free Full Text].

DAHLEN, M., T. OLSSON, G. KANTER-SMOLER, A. RAMNE, and P. SUNNERHAGEN, 1998  Regulation of telomere length by checkpoint genes in Schizosaccharomyces pombe.. Mol. Biol. Cell 9:611-621[Abstract/Free Full Text].

DESANY, B. A., A. A. ALCASABAS, J. B. BACHANT, and S. J. ELLEDGE, 1998  Recovery from DNA replicational stress is the essential function of the S-phase checkpoint pathway. Genes Dev. 12:2956-2970[Abstract/Free Full Text].

EHRENHOFER-MURRAY, A. E., R. T. KAMAKAKA, and J. RINE, 1999  A role for the replication proteins PCNA, RF-C, Polymerase and Cdc45 in transcriptional silencing in Saccharomyces cerevisiae.. Genetics 153:1171-1182[Abstract/Free Full Text].

EMILI, A., 1998  MEC1-dependent phosphorylation of Rad9p in response to DNA damage. Mol. Cell 2:183-189[Medline].

GIETZ, R. D. and A. SUGINO, 1988  New yeast-Escherichia coli shuttle vectors constructed with in vitro mutagenized yeast genes lacking six base-pair restriction sites. Gene 74:527-534[Medline].

GOTTSCHLING, D., 1992  Telomere-proximal DNA in Saccharomyces cerevisiae is refractory to methyltransferase activity in vivo.. Proc. Natl. Acad. Sci. USA 89:4062-4065[Abstract/Free Full Text].

GOTTSCHLING, D. E., O. M. APARICIO, B. L. BILLINGTON, and V. A. ZAKIAN, 1990  Position effect at S. cerevisiae telomeres: reversible repression of Pol II transcription. Cell 63:751-762[Medline].

GREENWELL, P. W., S. L. KRONMAL, S. E. PORTER, J. GASSENHUBER, and B. OBERMAIER et al., 1995  TEL1, a gene involved in controlling telomere length in S. cerevisiae, is homologous to the human Ataxia Telangiectasia gene. Cell 82:823-829[Medline].

GUTHRIE, C., and G. R. FINK (Editors), 1991 Guide to Yeast Genetics and Molecular Biology. Academic Press, Inc., San Diego.

HARTWELL, L. H. and T. WEINERT, 1989  Checkpoints: controls that ensure the order of cell cycle events. Science 246:629-634[Abstract/Free F