Telomere Folding Is Required for the Stable Maintenance of Telomere Position Effects in Yeast (original) (raw)

Mol Cell Biol. 2000 Nov; 20(21): 7991–8000.

Derik de Bruin

Department of Molecular Biology, Princeton University, Princeton, New Jersey 08544,2 and Laboratory of Molecular Genetics and Immunology, The Rockefeller University, New York, New York 100211

Sara M. Kantrow

Department of Molecular Biology, Princeton University, Princeton, New Jersey 08544,2 and Laboratory of Molecular Genetics and Immunology, The Rockefeller University, New York, New York 100211

Rachel A. Liberatore

Department of Molecular Biology, Princeton University, Princeton, New Jersey 08544,2 and Laboratory of Molecular Genetics and Immunology, The Rockefeller University, New York, New York 100211

Virginia A. Zakian

Department of Molecular Biology, Princeton University, Princeton, New Jersey 08544,2 and Laboratory of Molecular Genetics and Immunology, The Rockefeller University, New York, New York 100211

*Corresponding author. Mailing address: Department of Molecular Biology, Princeton University, Princeton, NJ 08544. Phone: (609) 258-6770. Fax: (609) 258-1701. E-mail: ude.notecnirp.oiblom@naikazv.

†Present address: Vanderbilt University, School of Medicine, Nashville, TN 37232.

Received 2000 Jun 1; Revisions requested 2000 Jul 18; Accepted 2000 Aug 8.

Abstract

Yeast telomeres reversibly repress the transcription of adjacent genes, a phenomenon called telomere position effect (TPE). TPE is thought to result from Rap1 and Sir protein-mediated spreading of heterochromatin-like structures from the telomeric DNA inwards. Because Rap1p is associated with subtelomeric chromatin as well as with telomeric DNA, yeast telomeres are proposed to form fold-back or looped structures. TPE can be eliminated in trans by deleting SIR genes or in cis by transcribing through the C1–3A/TG1–3 tract of a telomere. We show that the promoter of a telomere-linked URA3 gene was inaccessible to restriction enzymes and that accessibility increased both in a sir3 strain and upon telomere transcription. We also show that subtelomeric chromatin was hypoacetylated at histone H3 and at each of the four acetylatable lysines in histone H4 and that histone acetylation increased both in a sir3 strain and when the telomere was transcribed. When transcription through the telomeric tract occurred in G1-arrested cells, TPE was lost, demonstrating that activation of a silenced telomeric gene can occur in the absence of DNA replication. The loss of TPE that accompanied telomere transcription resulted in the rapid and efficient loss of subtelomeric Rap1p. We propose that telomere transcription disrupts core heterochromatin by eliminating Rap1p-mediated telomere looping. This interpretation suggests that telomere looping is critical for maintaining TPE.

The physical termini of linear eukaryotic chromosomes consist of DNA-protein structures called telomeres (57). In addition to their functions in promoting chromosome stability, yeast telomeres affect the behavior of nearby DNA. Saccharomyces telomeres repress the basal transcription of adjacent genes, a phenomenon called telomere position effect (TPE) (19), and they also affect both the timing of DNA replication (13) and the rate of mitotic recombination (48). Yeast telomeres are often clustered near the nuclear periphery, and it has been suggested that this positioning is important for transcriptional silencing (17).

Yeast telomeric DNA is assembled into a nonnucleosomal chromatin structure, the telosome, which encompasses the entire terminal tract of telomeric DNA (56). The major protein in the telosome is the sequence-specific DNA binding protein Rap1p (11), but Sir2p, Sir3p, and Sir4p, proteins required for TPE (2), also bind telomeres in vivo (5). The Sir proteins are most likely recruited to the telomere by their ability to interact with Rap1p (10, 36), and their appearance there is thought to be the initiating event in the establishment of TPE (31). Subtelomeric chromatin propagates continuously from the telomere inward (41), presumably through interactions between the Sir proteins and subtelomeric nucleosomes (24).

Although telomere-adjacent DNA is packaged into nucleosomes (56), these subtelomeric nucleosomes differ in several respects from nucleosomes elsewhere in the genome. First, the DNA in subtelomeric nucleosomes is less accessible to DNA-modifying enzymes such as Escherichia coli dam methyltransferase (20). Second, Sir proteins associate as a complex with subtelomeric nucleosomes (25, 49) by virtue of their ability to bind the N-terminal tails of histones H3 and H4 (24). Third, the N-terminal tails of histones H3 and H4 in subtelomeric nucleosomes are hypoacetylated compared to those of histones in most other regions of the genome (6, 35). These structural features are also characteristics of the yeast silent mating type or HM loci (22). At both telomeres and the HM loci, these structural features are lost in strains with sir or histone mutations that eliminate silencing, suggesting that transcriptional repression is a consequence of their uncommon chromatin structure (6, 30, 46, 52).

There is also evidence for a higher-order organization of yeast telomeric chromatin. Chromatin immunoprecipitation experiments reveal that Rap1p is associated in vivo not only with telomeric DNA (11) but also with subtelomeric chromatin (49). Since Rap1p, unlike the Sir proteins, does not interact directly with histones, this result led to the proposal that the yeast telomere folds back onto the subtelomeric regions to form a ∼3-kb region of core heterochromatin (22, 49). Similarly, mammalian telomeres are known to end in large 10- to 20-kb duplex “t loops” (21). Telomere loops might play an important role in regulating TPE (39).

The transcriptional state of a telomere-linked gene is reversible, and once established both the transcribed and the repressed states are stable for many cell generations (19, 35). The reversibility of TPE suggests a competition between the establishment of an active transcription complex and the assembly of the repressive chromatin structure, and this competition appears dependent on cell cycle progression. Although DNA replication is not needed for certain yeast genes to switch from a repressed to a transcriptionally active state (43), reactivation of regionally silenced chromatin, which involves the reconfiguration of an entire chromosomal domain, is thought to require DNA replication. For example, the accessibility of a repressed telomere-linked URA3 gene to its _trans_-activator Ppr1p occurs only during a limited interval late in the cell cycle (3) that corresponds roughly to the time when telomeric DNA is replicated (55).

In earlier studies on telomeric heterochromatin, loss of silencing was achieved using mutations in SIR or histone genes. However, these mutations affect all telomeres (24, 52) and also influence basal transcription (29), recombination (18), and chromosome stability (38). In this paper, we use a system in which TPE can be eliminated in cis at a single telomere by inducing transcription through the telomeric tract of C1–3A/TG1–3 DNA. This telomeric transcription has no effect on the stability of the affected chromosome (42). We show that switching this conditional telomere from a repressed to a transcribed state was accompanied by all of the chromatin changes previously associated with transcriptional activation, such as increased accessibility to the dam methyltransferase and a conversion from hypo- to hyperacetylated histones. The chromatin changes that accompany the switch from a repressed to a transcribed state in a telomere-linked gene, as well as transcription itself, occurred in G1-arrested cells. Although by nearly all criteria the chromatin structure of the transcribed telomere was indistinguishable from that of a telomere in a sir3 strain, telomere transcription did not result in the complete loss of subtelomeric Sir3p. However, telomere transcription rapidly and efficiently disrupted subtelomeric Rap1p interactions, suggesting that it caused loss of the core heterochromatin fold-back telomeric structure.

MATERIALS AND METHODS

All experiments were done with Saccharomyces cerevisiae yeast strains derived from either YPH499 (MATa ura3-52 lys2-801 ade2-101 trp1-Δ63 his3-Δ200 leu2-Δ) or YPH500 (a MATα version of YPH499) (45). Strains containing the UT, PT, and XT telomere modifications were made by transforming with _Eco_RI- and _Sal_I-cleaved pADH4UCA, ADH4UGT, and pADH4UGT-TATA, respectively (19, 42). E. coli dam methyltransferase was inserted at LYS2 by transforming YPH499 with _Xho_I-digested pDP6-dam to create YPH499LD (20). UTLD, PTLD, and XTLD were derived from YPH499LD by transformation as described above. With the exception of the dam methyltransferase-expressing strains, yeast strains carried ura3::LEU2 and had an 845-bp deletion in the ura3 transcription unit generated by transforming with Hin_dIII-digested pULA (19). UT_sir3 (sir3::LYS2) was made by transforming UT with _Eco_RI-digested pJR317 (28). The UINT strain was constructed by inserting URA3 at lys2 using _Aat_II-cleaved pCF116 (15). The _ura3_ΔEB minigene was made by first deleting the 78-bp _Eco_RV-_Bst_BI fragment from pVZ_URA3_-1 (1.1-kb _URA3 Hin_dIII fragment in pVZ1 (J. Stavenhagen, unpublished) and then recircularizing the plasmid to generate pVZ_ura3_ΔEB. The ∼1-kb _Xba_I _ura3_ΔEB minigene fragment was recovered from pVZ_ura3_ΔEB, the ends were blunted, and then the fragment was ligated into _Sma_I-digested pDP6 (14). The resulting plasmid, pDP_ura3_ΔEB, was digested with _Xho_I and transformed into _ura3_Δ yeast to generate DdB5ΔEB, thus integrating _ura3_ΔEB at the LYS2 locus. UTΔEB, PTΔEB, and XTΔEB were made by transforming DdB5ΔEB as described above. UT_sir3_ΔEB was derived from UTΔEB by transformation with Eco_RI-digested pKL3_sir3::HIS3 (J. Stavenhagen, unpublished). For the G1 arrest experiments, strain aDdB6PTb was used. This strain carried bar1::HIS3 and was constructed by transforming a MATa version of PT with _Pvu_II-digested pUCB14HIS3 (from D. Pederson).

Yeast complete synthetic medium (YC), 5-fluoro-orotic acid (FOA) medium, and rich medium (yeast extract-peptone [YEP]) were prepared and supplemented with either 2% glucose, 3% raffinose, or 3% galactose (Ultrapure; Sigma) as described previously (4, 19). For TPE analyses, 10-fold serial dilutions from each of four to six colonies from a YEP-peptone-dextrose plate were plated (∼300 cells per plate) on YC, YC-uracil, and YC plus FOA medium (47). Colonies were counted after 4 days of growth at 30°C. Methods for dam methylation protection assays (35) were described previously. Nuclei were prepared for in vivo restriction enzyme mapping from spheroplasts by Ficoll lysis as described previously (56), except that phenylmethylsulfonyl fluoride (PMSF) and iodoacetic acid were replaced with Complete, EDTA-free protease inhibitors (Boehringer Mannheim) plus 0.7 μg of pepstatin/ml. Nuclei were washed and resuspended in 1 ml of reaction buffer (10 mM Tris [pH 7.5], 50 mM NaCl, 10 mM MgCl2, 0.1 mM EDTA [pH 8.0], 0.2 mM EGTA [pH 8.0], 1 mM dithiothreitol, 0.1 mM PMSF) for every 7.5 × 109 starting cells. Enzyme digestion was carried out essentially as described previously (1). Aliquots of nuclei (200 μl) were removed, 100 to 200 U of enzyme was added, and the reaction mixture was incubated at 37°C for 30 min. Reactions were stopped by the addition of 1/10 volume of stop buffer (2% Sarkosyl, 0.4 M EDTA [pH 8], 10 mg of proteinase K/ml), and the DNA was recovered and analyzed by Southern blotting (56). The URA3 Southern hybridization probe was the 248-bp _Eco_RV-_Stu_I fragment from pVZ_URA3_-1.

ChIP experiments were carried out essentially as detailed elsewhere (12, 25). Antibodies against either tetra-acetylated histone H4, diacetylated histone H3, unacetylated histone H3 (gifts from David Allis or purchased from Upstate Biotechnology), or specific monoacetylated histone H4 lysine residues (H4KAc5, H4KAc8, H4KAc12, and H4KAc16; Serotec) were used. Production of antihistone antibodies, their specificities and their efficacies in chromatin immunoprecipitation (ChIP) have been described elsewhere (6, 9). The anti-Sir3p and anti-Rap1p antisera were previously described (11, 38). Recovered DNA was analyzed by PCR (94°C for 3 min followed by 25 cycles of 45 s at 94°C, 45 s at 55°C, and 75 s at 72°C), and the amount of template DNA for PCRs was determined empirically. PCR primers were URA3 5′ (URA3F1, GGA AAC GAA GAT AAA TCA TGT C; URA3R1, AGG CCT CTA GGT TCC TTT GTT AC), URA3 3′ (URA3F2, GTC CCA AAA TTT GTT TAC TAA AAA C; URA3R2, CTA CCT TAG CAT CCC TTC CC), AND TEL VI (TEL-300.fwd and TEL-300.rev) (34). PCR products were separated on 2% MetaPhor agarose (FMC) Tris-borate-EDTA-buffered gels, stained with ethidium bromide, and quantified by image analysis (Bio-Rad Molecular Analyst software).

For the G1 arrest experiments, aDdB6PTb cells were grown in YEP–3% raffinose media to an optical density at 660 nm of 0.4. Next, α-factor (Sigma) was added to 20 nM and potassium hydrate phthalate (pH 5.5) was added to 50 mM, and the culture was incubated at 30°C for 3 h to induce cell cycle arrest, as indicated by the presence of <2% budded cells (3, 8). Upon arrest, cultures were split into aliquots and either were allowed to continue growth with α-factor in YEP-raffinose or were harvested by centrifugation and suspended in an equal volume of warmed YEP–3% galactose medium either with or without α-factor and potassium hydrate phthalate. To the cultures without α-factor 1 mg of pronase/ml was added to degrade residual mating pheromone (8). Throughout this experiment the cells with α-factor remained at <2% buds, whereas the cells without α-factor lost the arrested phenotype and the bud index increased.

RNA (20 μg of DNase I-treated total yeast RNA) was transferred by slot blotting onto 0.2-μm Nytran+ nylon membranes (Schleicher & Schuell) as described previously (4). The ACT1 probe was the 282-bp _Kpn_I-_Pst_I DNA fragment from pYST122 (6), the SWI5 probe was a 2.1-kb labeled PCR product of the full-length gene amplified from YPH499 genomic DNA using the primers _SWI5F_F (TGG ATA CAT CAA ACT CTT GGT T) and _SWI5_R (CCT TTG ATT AGT TTT CAT TGG C), and the URA3 probe was a full-length antisense transcript from _Bam_HI-cut pVZ_URA3_-1 transcribed in vitro by T3 RNA polymerase.

RESULTS

Chromatin structural changes in a telomeric URA3 gene without loss of TPE.

Yeast cells expressing Ura3p die on media containing FOA. A telomeric URA3 gene is a convenient reporter for TPE, as the fraction of cells in which the telomeric URA3 is repressed can be measured by determining the fraction of cells that grow on FOA media (19). Three previously described _URA3_-based reporter constructs (42) were used in the present study. In each construct, URA3 was positioned near the left telomere of yeast chromosome VII (Fig. 1A). In the UT construct, the 5′ end of the URA3 transcription unit is ∼1.1 kb from the chromosome end (19). To make the XT construct, the GAL1,10 upstream activation sequence (UASG) was inserted between URA3 and the telomere (42). UASG is located immediately downstream of URA3 and contains four wild-type Gal4p binding sites but lacks endogenous TATA sequences. When UASG is positioned downstream of a gene, as at the XT telomere, it does not activate transcription (23, 50). The PT construct is similar to XT but contains the CYC1 TATA between UASG and the telomere. Because the CYC1 TATA-UASG is located 3′ of URA3 in the PT strain and because all of the strains carried PPR1, URA3 remains under the transcriptional control of its own promoter and _trans_-activator, Ppr1p. Previous studies demonstrated that growth of PT cells in galactose media results in Gal4p-mediated transcription through the telomeric tract of C1–3A/TG1–3 DNA and loss of TPE (42) but that TPE at other telomeres (W. H. Tham and V. A. Zakian, unpublished) and the chromosome stability of the transcribed telomere (42) were not affected. Thus, in this system, TPE at a single telomere can be controlled in cis without affecting other aspects of chromosome behavior.

Chromatin from galactose-grown PT and XT cells is accessible to dam methyltransferase. (A) A conditional telomere alleviates TPE in cis. Shown is a schematic of the UT, XT, and PT telomere modifications at the ADH4 locus on the left arm of chromosome VII (42). Quantitative TPE data (percentages of FOA-resistant cells) are means ± standard deviations. The URA3 promoter is located ∼1.1, 1.4, and 1.7 kb from the end of the chromosome in UT, XT, and PT cells, respectively. ←, C1–3A/TG1–3 telomeric repeat tract. Relevant restriction enzyme recognition sites are marked (D, _Dpn_I; H, _Hin_dIII; B, _Bam_HI). ∗, telomeric GATC sites (D) in URA3 that were protected from in vivo dam methylation. (B) Yeast cells expressing E. coli dam DNA methyltransferase (499LD, UTLD, XTLD, and PTLD) were grown in YEP media supplemented as shown. Purified DNA was digested with _Bam_HI and _Hin_dIII, and some was also digested with _Dpn_I (+), while the rest was not (−). The DNA was analyzed by Southern blotting using a URA3 probe (19). The XT and PT strains contain two _Dpn_I sites near the telomere. The _Dpn_I site closest to the telomere was within the nuclease-sensitive region that is fully accessible to dam methyltransferase (56). In this experiment the accessibility of the more-internal _Dpn_I site is monitored. Lanes 499, internal control.

The UT, XT, and PT constructs were placed in the YPH499 strain background. In raffinose media, Gal4p binds UASG but cannot activate transcription, whereas in galactose media the bound Gal4p activates transcription. As predicted from previous results (42), the three strains had equivalently high TPE when grown in FOA-raffinose medium. Although TPE was lost in galactose-grown PT cells, TPE remained high in the galactose-grown UT and XT strains (Fig. 1A).

The E. coli dam DNA methyltransferase gene can be expressed in yeast, and the accessibility of a given GATC sequence to dam methylation in vivo is thought to reflect the degree of openness in the surrounding chromatin structure (46). For example, as measured by the degree of methylation-sensitive Dpn_I cleavage, the single GATC site within the UT URA3 open reading frame (ORF) is inaccessible to dam methyltransferase in wild-type UT cells but is wholly accessible in TPE-negative UT_sir2 or UT_sir4_ cells (20). We determined if the transcriptional state of the telomeric URA3 gene in XT and PT cells also correlated with access to in vivo dam methyltransferase (20). The ura3-52 allele in these strains, which is near the centromere of chromosome V, was used as an internal control for full dam methyltransferase accessibility (Fig. 1B, lanes 499). DNA isolated from raffinose- and galactose-grown cells was analyzed by Southern hybridization either with or without cleavage by _Dpn_I (Fig. 1B).

The internal GATC site in telomeric URA3 genes was largely inaccessible to modification by dam methyltransferase in raffinose-grown UT, PT, and XT cells and in galactose-grown UT cells (Fig. 1B). However, in both PT and XT galactose-grown cells, the GATC site was completely accessible to the dam methyltransferase. The level of accessibility of PT and XT in galactose medium was comparable to that of the same site in a UT_sir3_ strain (Fig. 1B). These results demonstrate that URA3 transcription is not necessary for the chromatin alterations that allow dam methyltransferase access, as this site was methylated efficiently in FOA-resistant (Fig. 1A) galactose-grown XT cells (Fig. 1B). Additionally, chromatin immunoprecipitation analysis revealed that Gal4p bound the downstream UASG in both raffinose and galactose media but not in glucose medium (data not shown). Thus, the chromatin remodeling seen in XT galactose-grown cells was dependent on Gal4p activation rather than Gal4p binding. These data suggest that accessibility to the dam methyltransferase reflects an intermediate state in chromatin remodeling that is not itself sufficient for transcriptional activation.

Restriction enzyme accessibility reveals promoter blocking in a repressed, telomere-linked URA3.

To obtain a more detailed description of the chromatin structure of repressed and transcribed telomeric URA3 genes, we developed a restriction enzyme accessibility assay to monitor the openness of specific DNA sequences. Purified yeast nuclei from UT and UT_sir3_ cells were incubated with five different restriction enzymes with recognition sites throughout the ORF or within the promoter region of URA3 (Fig. 2A). DNA was then purified, digested with restriction enzymes to release the telomeric fragments, and analyzed by Southern blotting using a URA3 probe (Fig. 2A). In UT nuclei, the telomeric URA3 gene was largely inaccessible to digestion by all restriction enzymes tested (Fig. 2B). A similar pattern was seen for UT_sir3_ nuclei, except that the _Nde_I and _Dde_I restriction sites flanking the URA3 TATA were much more accessible to restriction enzyme cleavage (Fig. 2B). The increased accessibility of the _Nde_I and _Dde_I sites was not due to a loss of a neighboring nucleosome, as the nucleosome immediately adjacent to the URA3 TATA contains a recognition site for _Pst_I (Fig. 2A) and this site was inaccessible to cleavage in both sir3 and wild-type UT cells (Fig. 2B).

The URA3 TATA region is nuclease accessible when a telomeric gene is transcribed. (A) Cartoon of URA3 transcription unit showing six positioned nucleosomes (ovals) (51). The URA3 ORF (hatched box), URA3 TATA (gray box), relevant restriction enzyme recognition sites (H, _Hin_dIII; D, _Dde_I; N, _Nde_I; P, _Pst_I; E, _Eco_RV; S, _Stu_I; Sa, _Sau_3A; B, _Bam_HI), probe, and C1–3A/TG1–3 telomeric repeat tract (←) are shown. The _Sau_3A (_Dpn_I) site is 203 bp from the Bam_HI site on the edge of the penultimate nucleosome. Vertical arrows denote restriction enzymes with recognition sites throughout the ORF or within the promoter region of URA3. (B) Restriction enzyme mapping of telomeric URA3 loci. For these experiments, the internal ura3-52 locus was deleted, so that the only URA3 sequences in the strain came from the telomere-linked gene. Isolated nuclei from UT and UT_sir3 cells were incubated with either _Dde_I (D), _Nde_I (N), _Eco_RV (E), _Sau_3A (Sa), or _Pst_I (P). Recovered DNA was digested with _Hin_dIII and _Stu_I (lanes −, D, N, P, and E) or with _Bam_HI and _Hin_dIII (lane Sa) prior to Southern analysis using the URA3 probe. The Southern blots were quantitated by image analysis. The experiment was performed twice, and representative data are shown. (C) In vivo restriction enzyme mapping of UT, PT, and XT cells. Prior to nucleus isolation, cells were grown overnight in either raffinose or galactose media. Chromatin was analyzed as described for panel B. This experiment was performed twice, and representative data are shown.

To determine if transcription through the telomere had the same effects on restriction enzyme susceptibility as mutating sir3, raffinose- and galactose-grown UT, XT, and PT cells were examined using the same methods. For all three strains, the URA3 TATA was not accessible to either _Dde_I or Nde_I cleavage in raffinose-grown cells or in galactose-grown UT cells (Fig. 2C). However, galactose-grown PT cells were accessible to cleavage by both enzymes to an extent comparable to that for UT_sir3 cells (86 and 37% accessibility for the _Dde_I and Nde_I sites, respectively, in galactose-grown PT cells versus 90 and 22%, respectively at the same sites for UT_sir3 cells). In the XT strain, there was a small but reproducible increase in the level of _Dde_I accessibility in galactose-grown cells (Fig. 2C). It has been proposed that proximity to a telomere diminishes the accessibility of the promoter of a telomeric gene (3). Our data provide the first evidence for this model, as well as a new monitor for the chromatin structure of a telomere-linked gene. On the basis of promoter accessibility, telomeric transcription and deletion of SIR3 had similar effects on telomeric chromatin.

Telomeric chromatin is hypoacetylated at each acetylatable histone H4 lysine residue as well as in histone H3.

Previous studies examining histone acetylation patterns in yeast showed that the silent HM loci are associated with hypoacetylated histones H3 and H4 (6, 7). However, at the HM loci, histone H4 K12 appears hyperacetylated relative to the other lysine residues but hypoacetylated with respect to a sir2 mutant (7). When antiserum that recognizes a tetra-acetylated histone H4 is used, subtelomeric chromatin is hypoacetylated and this hypoacetylation is SIR dependent (6, 35). Here we used a quantitative ChIP to examine the acetylation status at specific lysine residues in telomeric histone H4 as well as the general acetylation state of histone H3.

The strains used for ChIP analysis contained, in addition to the telomeric URA3 gene, a ura3 minigene called _ura3_ΔEB that was inserted at the nontelomeric LYS2 locus. The _ura3_ΔEB gene had a 78-bp deletion within the URA3 ORF. After formaldehyde fixation, a soluble chromatin fraction was prepared from UTΔEB and UT_sir3_ΔEB cells. Antibodies specific for various acetylated and unacetylated forms of histones H3 and H4 were used to immunoprecipitate chromatin fragments (6, 25). The DNA in the immunoprecipitate was purified and subjected to quantitative competitive PCR (12) to determine the relative levels of URA3 and _ura3_ΔEB present in the immunoprecipitated chromatin.

Representative PCR results as well as the quantitation of these and other experiments obtained using _URA3_-specific primers are shown in Fig. 3A. The antisera specific for tetra-acetylated histone H4 (Fig. 3A) showed that telomeric URA3 chromatin is hypoacetylated compared to the internal _ura3_ΔEB chromatin) in wild-type cells but not in sir3 cells. In addition, telomeric chromatin was hypoacetylated at histone H3, and this hypoacetylation was SIR3 dependent (Fig. 3A, lanes 3+ and 3−). By using antisera specific for each of the monoacetylated forms of histone H4 (anti-H4KAc16, anti-H4KAc12, anti-H4KAc8, and anti-H4KAc5), telomeric chromatin was shown to be hypoacetylated at each H4 acetylatable lysine residue, including H4 K12, and that at each residue hypoacetylation was SIR3 dependent (Fig. 3A). As a control for the telomere ChIP experiments, we examined the histone acetylation state of the HMR locus. We found that histone H4 at HMR was hypoacetylated relative to ACT1 in a _SIR3_-dependent manner and that there was slight enrichment in H4 monoacetylated at lysine 12 at HMR chromatin compared to the level of acetylation of other H4 lysines (data not shown).

Histone acetylation analysis of UTΔEB and UT_sir3_ΔEB cells. Antibodies specific for acetylated and unacetylated histones were used to immunoprecipitate chromatin. The recovered DNA was analyzed by PCR using primers from either the 5′ or 3′ end of URA3. PCR products were quantitated by image analysis. To determine the relative level of histone acetylation, the _URA3/ura3_ΔEB ratio was calculated from an aliquot of the soluble chromatin extract and this value was used to normalize the _URA3/ura3_ΔEB ratios of the specific immunoprecipitates. This correction accounts for the relative immunoprecipitation efficiency of each antiserum (12). (A) (Top) Inverse of an ethidium bromide-stained agarose gel showing representative ChIP PCRs using the URA3 5′ primers. T, total soluble chromatin aliquot; 4+, anti-tetra-acetylated histone H4; 3+, anti-di-acetylated histone H3; H3−, anti-unacetylated histone H3; 16+, 12+, 8+, 5+, antibodies specific for specific acetylated N-terminal lysine residues of histone H4. (Bottom) Quantitated ChIP data for subtelomeric chromatin. _URA3/ura3_ΔEB ratios are mean values ± standard deviations (SD) from three experiments using the 5′ URA3 primers. (B) Quantitated ChIP analysis of histone acetylation state in raffinose- and galactose-grown UTΔEB, PTΔEB, and XTΔEB cells. Yeast cells were grown in the indicated media and analyzed by ChIP as described for panel A using either the anti-tetra-acetylated histone H4 antibody (anti-H4+) or the anti-unacetylated histone (anti-H3−) antisera. _URA3/ura3_ΔEB ratios are the mean values ± SD from three independent assays using either the 5′ URA3 or 3′ URA3 primer pairs.

To determine if Gal4p binding alone and/or telomeric transcription promoted changes in histone acetylation, we examined the 3′ and 5′ regions of the telomeric URA3 gene in UTΔEB, PTΔEB, and XTΔEB strains using quantitative ChIP and either the tetra-acetylated histone H4 antisera or the unacetylated histone H3 antisera (Fig. 3B). In UT_sir3_ and PT galactose-grown cells, URA3 activation correlated with a 4- to 5-fold increase in histone H4 acetylation in the 5′ region and a 11- to 14-fold increase in the 3′ region of the gene. In contrast, in UT and XT cells shifted from raffinose to galactose, histone H4 acetylation was relatively unchanged. Likewise, the acetylation of histone H3 increased 3- to 12-fold in PT galactose-grown and UT_sir3_ cells, whereas galactose-grown UT and XT cells showed little change. Thus, telomere transcription and sir3 mutation had similar effects on histone acetylation. However, the binding of activated Gal4p in XT galactose-grown cells was not sufficient to promote histone acetylation of subtelomeric chromatin.

A switch from repressed to active transcriptional states does not require DNA replication.

Earlier experiments suggested that DNA replication is necessary for a telomere-linked gene to switch from a repressed to a transcribed state (3). In particular, if ppr1 cells with a telomeric URA3 gene are arrested in G1 and Ppr1p expression is then induced from a plasmid, then the telomeric URA3 gene remains silenced. In contrast, Ppr1p can induce URA3 expression in G2/M-arrested cells (3). We reconsidered this observation by asking if the repressed URA3 gene in galactose-grown PT cells could be reactivated during G1 arrest.

PT cells were grown in raffinose media to which α-factor was added to arrest cells in late G1 phase (26). Two criteria were used to verify that cells remained in G1 phase, a bud index of <2% (data not shown) and a lack of SWI5 RNA (Fig. 4A) (SWI5 is expressed only from S through M phases) (3, 37). Once cells were arrested, a time zero aliquot was removed for RNA analysis (Fig. 4A, sample 1). The rest of the culture was washed, resuspended in medium containing 3% galactose, and then split in two. One culture was maintained in G1 by addition of more α-factor. In the second, α-factor was removed and the cells allowed to progress through the cell cycle (Fig. 4A). Aliquots of cells were taken for RNA analysis at 1-h intervals (Fig. 4A).

RNA slot blot analysis and in vivo restriction enzyme mapping of α-factor-arrested PT cells. (A) PT cells (aDdB6PTb) were grown in raffinose media to which α-factor was added to arrest cells in late G1 phase. An aliquot of cells was removed (sample 1) for RNA isolation. The remainder of the culture was resuspended in galactose medium and then split into two; α-factor was added to one culture (+), and pronase was added to the other (−). Aliquots were removed at 1-h intervals for RNA preparation (time values are hours in galactose media for samples 2 to 11). Total RNA slot blots were hybridized with URA3, ACT1, and SWI5 probes. (B) (Top) Cells were grown in raffinose media and arrested in late G1 with α-factor mating pheromone. An aliquot of arrested cells (R0) was removed, and nuclei were prepared for restriction enzyme mapping. The rest of the culture was split into two parts; galactose was added to one culture, and fresh α-factor was added to both. Aliquots were removed for preparation of the nuclei from the cultures at the indicated number of minutes. Nuclei were prepared and then either incubated without restriction enzyme (−) or with either _Dde_I (D) or _Nde_I (N). DNA was recovered and analyzed by Southern blotting. (Bottom) Total RNA was prepared from PT cells harvested from an α-factor arrest experiment identical to the one described above. In addition to the raffinose- or galactose-grown cultures containing α-factor (RAF+α or GAL+α, respectively), a culture with galactose but without α-factor (GAL-α) was examined. The amounts of URA3, SWI5, and ACT1 RNA were determined by slot blot analysis. The URA3 and SWI5 values were normalized to those for ACT1, and these values were plotted.

Within 1 h of galactose addition, RNA slot blot analysis showed that the telomeric URA3 gene was transcribed in both G1-arrested (with α-factor) and cycling (without α-factor) cells. By the criteria of SWI5 expression and budding index, the culture without α-factor progressed through the cell cycle while the cells with α-factor remained in G1 (Fig. 4A). These data suggest that a repressed telomere-linked gene can be reactivated without undergoing DNA replication.

An alternative explanation for these data is that the telomeric URA3 gene was transcribed in only a small fraction of cells. To test this possibility, we examined the chromatin structure of the telomeric URA3 gene in G1-arrested PT cells after addition of galactose. If only a small fraction of telomeric genes were transcribed, then we would not expect to see the changes in chromatin structure that characterized the switch from a repressed to a transcribed state. However, by the criterion of promoter accessibility, the chromatin structure of URA3 was that of a transcribed gene by 45 min after galactose addition (Fig. 4B). Furthermore, using the anti-tetra-acetylated histone H4 antibody to immunoprecipitate chromatin, we found that the acetylation state of histone H4 in both G1-arrested and cycling galactose-grown PT cells increased (data not shown). These results indicate that transcriptional activation of a telomeric gene does not require cell cycle progression.

Subtelomeric Rap1p interactions are lost upon telomere transcription.

By the experimental criteria used so far, the chromatin structure of galactose-grown PT cells was indistinguishable from that of a strain lacking Sir3p (Fig. 1 to 3). Overproduction of Sir3p, furthermore, could partially restore silencing in galactose-grown PT cells (data not shown). Moreover, earlier work showed that the HM loci are transcribed in the absence of DNA replication when cells are depleted of Sir3p (33). These data suggested that the loss of silencing in galactose-grown PT cells might be due to the loss of Sir3p from subtelomeric chromatin. ChIP analysis was used to test directly if Sir3p was lost from subtelomeric chromatin in galactose-grown PT cells.

PTΔEB cells were grown for 16 h (∼10 cell divisions) in either raffinose or galactose media; UT_sir3_ΔEB cells were grown for 16 h in raffinose media. Soluble chromatin was precipitated using antisera for tetra-acetylated histone H4, Sir3p (anti-Sir3p), or Rap1p (anti-Rap1p). Precipitated DNA was analyzed by multiplex PCR using primers for URA3 and a unique subtelomeric region on chromosome VI (34). The _URA3/ura3_ΔEB ratio was calculated to determine changes in histone acetylation relative to that of the ura3 minigene. The URA3/TEL VI ratio was calculated to determine loss of Sir3p or Rap1p at the transcribed telomere relative to the level of Sir3p or Rap1p at an unmodified subtelomeric region on chromosome VI (Fig. 5A). For presentation purposes, the absolute ratios have been normalized to show the percentage of change.

ChIP analysis of subtelomeric Sir3p and Rap1p in PT and UT_sir3_ yeast. ChIP analysis was done using either an anti-tetra-acetylated histone H4 antiserum (anti-H4+), an anti-Sir3p antiserum (anti-Sir3p), or an anti-Rap1p antiserum (anti-Rap1p) as described for Fig. 3. (A) PTΔEB and UT_sir3_ΔEB yeast cells were grown overnight (16 h) in either raffinose (PT + RAF and UT_sir3_) or galactose (PT + GAL) media prior to analysis. (Left) Inverse of ethidium bromide-stained gel showing representative multiplex PCRs using the URA3 5′ primers (URA3 and _ura3_ΔEB) and the unique chromosome VI subtelomeric primers (TEL VI). Lanes: T, total; 4+, anti-H4+; S, anti-Sir3p; R, anti-Rap1p. (Right) Quantitated ChIP data. The mean values of the URA3/ura3_ΔEB and URA3/TEL VI ratios from three independent assays were calculated. These ratios were divided by either the UT_sir3 value for anti-H4+ (3.2) or by the PT + RAF values for anti-Rap1p (1.1) and anti-Sir3p (1.2), and the normalized ratios were plotted. (B) Time course ChIP analysis of PTΔEB cells. Cells were grown in raffinose media and then switched to galactose media. Aliquots of cells were removed at the indicated time points and analyzed by ChIP as described for panel A.

As expected, PT cells grown in galactose for 16 h showed increased histone H4 acetylation at the PT telomere (Fig. 5A) but not at the VI R telomere (data not shown) whereas the sir3 strain showed increased acetylation at both (Fig. 5A and data not shown). Although the level of subtelomeric Sir3p was reduced in galactose-grown PT cells, the level of subtelomeric Sir3p at 16 h was still ∼35% of the preinduction level (Fig. 5A, lanes S). In contrast, the level of subtelomeric Rap1p was <20% of that seen in raffinose-grown cells (Fig. 5A, lanes R).

To assess the relative rates of Rap1p and Sir3p loss, the amounts of subtelomeric Rap1p and Sir3p as well as the extent of histone H4 acetylation were monitored shortly after the switch of PT cells to galactose medium. Within 20 min of galactose addition, the level of subtelomeric Rap1p dropped to <50% of the preinduction level, whereas the levels of Sir3p and histone H4 acetylation were unchanged (Fig. 5B). Although the amount of Sir3p gradually decreased between 20 min and 2.5 h, at 2.5 h Sir3p was still at ∼65% of the preinduction level. By 1 h, URA3 RNA was at high levels (Fig. 4A), even though Sir3p was still at the telomere (Fig. 5B). In contrast, at 2.5 h, the amount of subtelomeric Rap1p was reduced to background levels (Fig. 5B, 16 h). Thus, telomere transcription rapidly and efficiently eliminates subtelomeric Rap1p but only partially and more slowly removes subtelomeric Sir3p.

Induction of telomeric URA3 is not sufficient to disrupt subtelomeric Rap1p and Sir3p.

Since telomere transcription resulted in the efficient loss of subtelomeric Rap1p, we asked if this loss was a general feature of the transcriptional reactivation of a telomeric URA3 gene. Using ChIP analysis, UT cells growing in complete medium, where TPE is high (Fig. 1A), were compared to UT cells growing in medium lacking uracil, where TPE is eliminated (19). RNA slot blot analysis confirmed that the amounts of steady-state URA3 mRNA in UT cells lacking uracil and galactose-grown PT cells were similar (Fig. 6A). However, URA3 induction in the UT strain did not result in loss of either Rap1p or Sir3p from subtelomeric chromatin, although, as expected, histone H4 acetylation increased (Fig. 6B). Thus, transcriptional activation of a telomeric gene can occur without loss of either subtelomeric Rap1p or Sir3p.

RNA and ChIP analysis of an activated telomeric URA3 gene. (A) RNA slot blot analysis of steady-state URA3 transcripts. Total yeast RNA was isolated from yeast strains with the endogenous ura3-52 sequences deleted and grown in the indicated media. The UINT strain contains a nontelomeric URA3 gene. DNase I-treated RNA was probed with either ACT1 or URA3 probes. The URA3/ACT1 ratios are means ± standard deviations obtained from three independent experiments. (B) ChIP analysis of subtelomeric histone H4, Sir3p, and Rap1p in repressed and activated UTΔEB cells. UTΔEB cells were grown overnight in the indicated media, and the chromatin was immunoprecipitated as described for Fig. 3. (Left) Inverse of ethidium bromide-stained gel showing multiplex PCRs using the URA3 5′ primers (URA3 and _ura3_ΔEB) and the unique chromosome VI subtelomeric primers (TEL VI). Lanes, T, total; 4+, anti-H4+; S, anti-Sir3p; R, anti-Rap1p. (Right) Quantitated ChIP data. The _URA3/ura3_ΔEB and URA3/TEL VI ratios were calculated, and the average ratios from two independent experiments were plotted.

DISCUSSION

It has been suggested that the transcriptional silencing of telomere-linked genes occurs because telomeric heterochromatin occludes the promoter, thereby preventing access to transcription factors (3). This paper provides the first direct evidence for this hypothesis. Using restriction enzyme accessibility, we show that the promoter of a telomere-linked URA3 gene was inaccessible to restriction enzymes and that this effect was SIR3 dependent (Fig. 2). When URA3 was repressed, restriction enzyme sites flanking the URA3 TATA sequence were inaccessible to in vivo cleavage (∼5% access for repressed UT cells), whereas these sites were ∼90% accessible in a UT_sir3_ strain.

We also examined changes in the chromatin structure of a telomere-linked URA3 gene using a system wherein TPE is controlled in cis at a single telomere by transcription through the C1–3A/TG1–3 telomere repeat tract. Telomere transcription resulted in changes in chromatin structure at the PT telomere similar to those resulting from deletion of SIR3: we observed enhanced access of subtelomeric DNA to both dam methyltransferase (Fig. 1) and to restriction enzymes (Fig. 2). In contrast, the XT telomere in galactose-grown cells appeared to have a chromatin state intermediate between repressed and transcribed states. In galactose medium, the XT telomere was repressed (Fig. 1A), yet accessibility to the dam methyltransferase was as high as that for a transcribed telomeric gene (Fig. 1B). These data show that an inaccessible chromatin structure at the 3′ end of a telomere-linked gene is not essential for maintaining transcriptional repression. The URA3 promoter in galactose-grown XT cells also appeared to be in an intermediate state, as monitored via its accessibility at the _Dde_I site (Fig. 2). Yet despite the changes in dam methyltransferase and promoter accessibility, the acetylation of histones H3 and H4 did not increase in galactose-grown XT cells (Fig. 3B). The 5′ and 3′ URA3 chromatin changes could represent preliminary steps in URA3 reactivation, such as the Gal4p-mediated recruitment of a preinitiation complex to the URA3 TATA (40). Once the complex is recruited to the URA3 TATA, elongation of the assembled transcription complex past the _Nde_I site may be blocked either by the core heterochromatin structure (44) or by telomere-telomere or telomere-nuclear envelope interactions. These data suggest that certain changes in telomeric chromatin structure, such as the reorganization that allows accessibility to the dam methyltransferase, are induced prior to transcription but that others, such as histone acetylation, are consequences of transcription.

Although yeast HM, Schizosaccharomyces pombe centromeric, and Drosophila melanogaster heterochromatins are generally hypoacetylated, there is a significant level of histone H4 lysine 12-specific acetylation (H4KAc12) present in all three (7, 12, 53). Subtelomeric chromatin had previously been shown to have low levels of acetylated histone H4 (35). We extend this result, showing that subtelomeric histone H4 was hypoacetylated at each acetylatable lysine, including H4KAc12, as well as at histone H3 (Fig. 3). Like the deletion of SIR3, telomere transcription in galactose-grown PT cells resulted in a 4- to 11-fold increase in subtelomeric histone acetylation relative to that for raffinose-grown cells (Fig. 3). Thus, the effects of telomere transcription on telomeric chromatin structure mimic those caused by the deletion of SIR3, demonstrating that the conditional telomere was a legitimate system for the study of TPE.

Telomere position effects are reversible (19). Since telomeric chromatin blocked a telomere-linked promoter (Fig. 2), DNA replication may provide an opportunity for an activator to find a newly exposed promoter, explaining why switching from a repressed to a transcribed state is limited to a discrete part of the cell cycle (3). However, this paper demonstrated that TPE was lost in G1-arrested PT cells when these cells were switched from raffinose to galactose medium. Both URA3 expression (Fig. 4A) and chromatin remodeling (Fig. 4B) occurred in the absence of DNA replication. This switch was accompanied by rapid and efficient loss of subtelomeric Rap1p and a slower and less-complete loss of Sir3p (Fig. 5A). After ∼10 generations of continuous telomere transcription, about 35% of the wild-type level of subtelomeric Sir3p was still associated with the PT telomere, whereas the level of subtelomeric Rap1p was ∼15% of the preinduction level (Fig. 5A). Additional ChIP experiments revealed that 20 min of telomere transcription was sufficient to remove about 60% of the subtelomeric Rap1p from the PT telomere, but in contrast the level of subtelomeric Sir3p was essentially unchanged at this time point (Fig. 5B). This loss of subtelomeric Rap1p was not seen in two other situations where TPE was disrupted, i.e., deletion of sir3 (Fig. 5A) and URA3 induction by growth of UT cells in uracil-deficient medium (Fig. 6), nor is subtelomeric Rap1p lost in galactose-grown XT cells (data not shown).

We interpret these data as indicating that telomere transcription alters the higher-order structure of telomeric chromatin, eliminating the loop postulated to form by the telomere folding back onto subtelomeric nucleosomes (49) (Fig. 7). We propose that DNA replication also causes telomere unfolding, and that this unfolding, not DNA replication per se (3), is a prerequisite for a switch from a repressed to a transcribed state. In addition, we further suggest that, when looping is disrupted by telomere transcription, the telomere is dislocated away from the nuclear periphery (32, 38).

Model of how telomere transcription eliminates TPE. Growth of PT cells in galactose media eliminates telomere looping. Rap1p, in addition to being bound at the telosome, is also associated with subtelomeric chromatin for a distance of 3 to 4 kb from the telomere (49). This observation led to the model that yeast telomeres form fold-back or looped structures (22). In addition, it has been suggested that subtelomeric Rap1p may result from the spreading of Rap1p from the telomere through interactions with SIR protein complexes (16). We suggest that the rapid and efficient loss of Rap1p from the subtelomeric region of galactose-grown PT cells reflects the loss of telomere folding, which in turn eliminates Rap1p spreading and silencing at URA3.

Telomere looping may help explain the observed increase in accessibility of the URA3 promoter to restriction enzymes in galactose-grown XT cells. Since UASG does not function when positioned downstream of the promoter of a nontelomeric gene (23, 50), telomere folding could bring UASG-bound Gal4p to the URA3 promoter. Indeed in the absence of the _GAL4_-negative regulator GAL80, Gal4p can activate URA3 from downstream and eliminate TPE in XT cells (D. de Bruin, Z. Zaman, R. A. Lieratore, and M. Ptashne, submitted for publication), thus supporting the existence of yeast telomere loops in vivo.

The analysis of mammalian telomeres suggests that looping may be a general property of eukaryotic telomeres (21). However, mammalian telomere loops appear to be critical for chromosome stability, as conditions that disrupt these loops in vivo promote end-to-end fusions (27, 54). In contrast, in yeast, telomere transcription disrupts looping but does not decrease chromosome stability (42). However, yeast telomere looping may still serve important functions. For example, telomere looping might be critical for telomeric silencing as well as for other telomeric position effects, such as blocking undesirable telomere-telomere recombination events (48).

ACKNOWLEDGMENTS

We thank D. Allis for antihistone antisera, L. Pillus for the anti-Sir3p antiserum, and D. Peterson for the pUCB14HIS3 plasmid. We also thank M. A. Osley, J. Broach, E. Monson, and X. Bi for their helpful comments on the manuscript and P. Kaloudis for help with the figures. Finally, we are grateful to J. Ravetch, The Rockefeller University, for his extremely generous support of D.D.B., S.M.K., and R.A.L.

This work was supported in part by National Institutes of Health grant GM43265 to V.A.Z. and by American Cancer Society grant PF-4236 to D.D.B.

REFERENCES

1. Almer A, Horz W. Nuclease hypersensitive regions with adjacent positioned nucleosomes mark the gene boundaries of the PHO5/PHO3 locus in yeast. EMBO J. 1986;5:2681–2687. [PMC free article] [PubMed] [Google Scholar]

2. Aparicio O M, Billington B L, Gottschling D E. Modifiers of position effect are shared between telomeric and silent mating-type loci in S. cerevisiae. Cell. 1991;66:1279–1287. [PubMed] [Google Scholar]

3. Aparicio O M, Gottschling D E. Overcoming telomeric silencing: a trans-activator competes to establish gene expression in a cell cycle-dependent way. Genes Dev. 1994;8:1133–1146. [PubMed] [Google Scholar]

4. Ausubel F N, Brent R, Kingston R E, Moore D D, Seidman J G, Smith J A, Struhl K, editors. Current Protocols in Molecular Biology. 1 to 3. New York, N.Y: John Wiley and Sons, Inc.; 1994. [Google Scholar]

5. Bourns B D, Alexander M K, Smith A M, Zakian V A. Sir proteins, Rif proteins, and Cdc13p bind Saccharomyces telomeres in vivo. Mol Cell Biol. 1998;18:5600–5608. [PMC free article] [PubMed] [Google Scholar]

6. Braunstein M, Rose A B, Holmes S G, Allis C D, Broach J R. Transcriptional silencing in yeast is associated with reduced nucleosome acetylation. Genes Dev. 1993;7:592–604. [PubMed] [Google Scholar]

7. Braunstein M, Sobel R E, Allis C D, Turner B M, Broach J R. Efficient transcriptional silencing in Saccharomyces cerevisiae requires a heterochromatin histone acetylation pattern. Mol Cell Biol. 1996;16:4349–4356. [PMC free article] [PubMed] [Google Scholar]

8. Brewer B J, Fangman W L. The localization of replication origins on ARS plasmids in S. cerevisiae. Cell. 1987;51:463–471. [PubMed] [Google Scholar]

9. Clarke D J, O'Neill L P, Turner B M. Selective use of H4 acetylation sites in the yeast Saccharomyces cerevisiae. Biochem J. 1993;294:557–561. [PMC free article] [PubMed] [Google Scholar]

10. Cockell M, Palladino F, Laroche T, Kyrion G, Liu C, Lustig A J, Gasser S M. The carboxy termini of Sir4 and Rap1 affect Sir3 localization: evidence for a multicomponent complex required for yeast telomeric silencing. J Cell Biol. 1995;129:909–924. [PMC free article] [PubMed] [Google Scholar]

11. Conrad M N, Wright J H, Wolf A J, Zakian V Z. RAP1 protein interacts with yeast telomeres in vivo: overproduction alters telomere structure and decreases chromosome stability. Cell. 1990;63:739–750. [PubMed] [Google Scholar]

12. Ekwall K, Olsson T, Turner B M, Cranston G, Allshire R C. Transient inhibition of histone deacetylation alters the structural and functional imprint at fission yeast centromeres. Cell. 1997;91:1021–1032. [PubMed] [Google Scholar]

13. Ferguson B M, Fangman W L. A position effect on the time of replication origin activation in yeast. Cell. 1992;68:333–339. [PubMed] [Google Scholar]

14. Fleig U N, Pridmore R D, Philippsen P. Construction of LYS2 cartridges for use in genetic manipulations of Saccharomyces cerevisiae. Gene. 1986;46:237–245. [PubMed] [Google Scholar]

15. Freudenreich C H, Kantrow S M, Zakian V A. Expansion and length-dependent fragility of CTG repeats in yeast. Science. 1998;279:853–856. [PubMed] [Google Scholar]

16. Gotta M, Cockell M. Telomeres, not the end of the story. Bioessays. 1997;19:367–370. [PubMed] [Google Scholar]

17. Gotta M, Laroche T, Formenton A, Maillet L, Scherthan H, Gasser S M. The clustering of telomeres and colocalization with Rap1, Sir3, and Sir4 proteins in wild-type Saccharomyces cerevisiae. J Cell Biol. 1996;134:1349–1363. [PMC free article] [PubMed] [Google Scholar]

18. Gottlieb S, Esposito R E. A new role for a yeast transcriptional silencer gene, SIR2, in regulation of recombination in ribosomal DNA. Cell. 1989;56:771–776. [PubMed] [Google Scholar]

19. Gottschling D E, Aparico O M, Billington B L, Zakian V A. Position effect at S. cerevisiae telomeres: reversible repression of Pol II transcription. Cell. 1990;63:751–762. [PubMed] [Google Scholar]

20. Gottschling D E. Telomere-proximal DNA in Saccharomyces cerevisiae is refractory to methyltransferase activity in vivo. Proc Natl Acad Sci USA. 1992;89:4062–4065. [PMC free article] [PubMed] [Google Scholar]

21. Griffith J D, Comeau L, Rosenfield S, Stansel R M, Bianchi A, Moss H, de Lange T. Mammalian telomeres end in a large duplex loop. Cell. 1999;97:503–514. [PubMed] [Google Scholar]

22. Grunstein M. Yeast heterochromatin: regulation of its assembly and inheritance by histones. Cell. 1998;93:325–328. [PubMed] [Google Scholar]

23. Guarente L, Hoar E. Upstream activation sites of the CYC1 gene of Saccharomyces cerevisiae are active when inverted but not when placed downstream of the “TATA box.” Proc Natl Acad Sci USA. 1984;81:7860–7864. [PMC free article] [PubMed] [Google Scholar]

24. Hecht A, Laroche T, Strahl-Bolsinger S, Gasser S M, Grunstein M. Histone H3 and H4 N-termini interact with SIR3 and SIR4 proteins: a molecular model for the formation of heterochromatin in yeast. Cell. 1995;80:583–592. [PubMed] [Google Scholar]

25. Hecht A, Strahl-Bolsinger S, Grunstein M. Spreading of transcriptional repressor SIR3 from telomeric heterochromatin. Nature. 1996;383:92–96. [PubMed] [Google Scholar]

26. Hereford L M, Hartwell L H. Sequential gene function in the initiation of Saccharomyces cerevisiae DNA synthesis. J Mol Biol. 1974;84:445–461. [PubMed] [Google Scholar]

27. Karlseder J, Broccoli D, Dai Y, Hardy S, de Lange T. p53- and ATM-dependent apoptosis induced by telomeres lacking TRF2. Science. 1999;283:1321–1325. [PubMed] [Google Scholar]

28. Kimmerly W J, Rine J. Replication and segregation of plasmids containing _cis_-acting regulatory sites of silent mating-type genes in Saccharomyces cerevisiae are controlled by the SIR genes. Mol Cell Biol. 1987;7:4225–4237. [PMC free article] [PubMed] [Google Scholar]

29. Lenfant F, Mann R K, Thomsen B, Ling X, Grunstein M. All four core histone N-termini contain sequences required for the repression of basal transcription in yeast. EMBO J. 1996;15:3974–3985. [PMC free article] [PubMed] [Google Scholar]

30. Loo S, Rine J. Silencers and domains of generalized repression. Science. 1994;264:1768–1771. [PubMed] [Google Scholar]

31. Lustig A J, Liu C, Zhang C, Hanish J P. Tethered Sir3p nucleates silencing at telomeres and internal loci in Saccharomyces cerevisiae. Mol Cell Biol. 1996;16:2483–2495. [PMC free article] [PubMed] [Google Scholar]

32. Maillet L, Boscheron C, Gotta M, Marcand S, Gilson E, Gasser S M. Evidence for silencing compartments within the yeast nucleus: a role for telomere proximity and Sir protein concentration in silencer-mediated repression. Genes Dev. 1996;10:1796–1811. [PubMed] [Google Scholar]

33. Miller A M, Nasmyth K A. Role of DNA replication in the repression of silent mating type loci in yeast. Nature. 1984;312:247–251. [PubMed] [Google Scholar]

34. Mills K D, Sinclair D A, Guarente L. _MEC1_-dependent redistribution of the Sir3 silencing protein from telomeres to DNA double-strand breaks. Cell. 1999;97:609–620. [PubMed] [Google Scholar]

35. Monson E K, de Bruin D, Zakian V A. The yeast Cac1 protein is required for the stable inheritance of transcriptionally repressed chromatin at telomeres. Proc Natl Acad Sci USA. 1997;94:13081–13086. [PMC free article] [PubMed] [Google Scholar]

36. Moretti P, Freeman K, Coodly L, Shore D. Evidence that a complex of SIR proteins interacts with the silencer and telomere-binding protein RAP1. Genes Dev. 1994;8:2257–2269. [PubMed] [Google Scholar]

37. Nasmyth K, Seddon A, Ammerer G. Cell cycle regulation of SW15 is required for mother-cell-specific HO transcription in yeast. Cell. 1987;49:549–558. [PubMed] [Google Scholar]

38. Palladino F, Laroche T, Gilson E, Axelrod A, Pillus L, Gasser S M. SIR3 and SIR4 proteins are required for the positioning and integrity of yeast telomeres. Cell. 1993;75:543–555. [PubMed] [Google Scholar]

39. Park Y, Lustig A J. Telomere structure regulates the heritability of repressed subtelomeric chromatin in Saccharomyces cerevisiae. Genetics. 2000;154:587–598. [PMC free article] [PubMed] [Google Scholar]

40. Ranish J A, Yudkovsky N, Hahn S. Intermediates in formation and activity of the RNA polymerase II preinitiation complex: holoenzyme recruitment and a postrecruitment role for the TATA box and TFIIB. Genes Dev. 1999;13:49–63. [PMC free article] [PubMed] [Google Scholar]

41. Renauld H, Aparicio O M, Zierath P D, Billington B L, Chhablani S K, Gottschling D E. Silent domains are assembled continuously from the telomere and are defined by promoter distance and strength, and by SIR3 dosage. Genes Dev. 1993;7:1133–1145. [PubMed] [Google Scholar]

42. Sandell L, Gottschling D E, Zakian V A. Transcription of a yeast telomere alleviates telomere position effect without affecting chromosome stability. Proc Natl Acad Sci USA. 1994;91:12061–12065. [PMC free article] [PubMed] [Google Scholar]

43. Schmid A, Fascher K D, Horz W. Nucleosome disruption at the yeast PHO5 promoter upon PHO5 induction occurs in the absence of DNA replication. Cell. 1992;71:853–864. [PubMed] [Google Scholar]

44. Sekinger E A, Gross D S. SIR repression of a yeast heat shock gene: UAS and TATA footprints persist within heterochromatin. EMBO J. 1999;18:7041–7055. [PMC free article] [PubMed] [Google Scholar]

45. Sikorski R S, Hieter P. A system of shuttle vectors and yeast host strains designed for efficient manipulation of DNA in Saccharomyces cerevisiae. Genetics. 1989;122:19–27. [PMC free article] [PubMed] [Google Scholar]

46. Singh J, Klar A J S. Active genes in budding yeast display enhanced in vivo accessibility to foreign DNA methylases: a novel in vivo probe for chromatin structure of yeast. Genes Dev. 1992;6:186–196. [PubMed] [Google Scholar]

47. Stavenhagen J B, Zakian V A. Internal tracts of telomeric DNA act as silencers in Saccharomyces cerevisiae. Genes Dev. 1994;8:1411–1422. [PubMed] [Google Scholar]

48. Stavenhagen J B, Zakian V A. Yeast telomeres exert a position effect on recombination between internal tracts of yeast telomeric DNA. Genes Dev. 1998;12:3044–3058. [PMC free article] [PubMed] [Google Scholar]

49. Strahl-Bolsinger S, Hecht A, Luo K, Grunstein M. SIR2 and SIR4 interactions differ in core and extended telomeric heterochromatin in yeast. Genes Dev. 1997;11:83–93. [PubMed] [Google Scholar]

50. Struhl K. Genetic properties and chromatin structure of the yeast gal regulatory element: an enhancer-like sequence. Proc Natl Acad Sci USA. 1984;81:7865–7869. [PMC free article] [PubMed] [Google Scholar]

51. Tanaka S, Livingstone-Zatchej M, Thoma F. Chromatin structure of the yeast URA3 gene at high resolution provides insight into structure and positioning of nucleosomes in the chromosomal context. J Mol Biol. 1996;257:919–934. [PubMed] [Google Scholar]

52. Thompson J S, Ling X, Grunstein M. Histone H3 amino terminus is required for telomeric and silent mating locus repression in yeast. Nature. 1994;369:245–247. [PubMed] [Google Scholar]

53. Turner B M, Birley A J, Lavender J. Histone H4 isoforms acetylated at specific lysine residues define individual chromosomes and chromatin domains in Drosophila polytene nuclei. Cell. 1992;69:375–384. [PubMed] [Google Scholar]

54. van Steensel B, Smogorzewska A, de Lange T. TRF2 protects human telomeres from end-to-end fusions. Cell. 1998;92:401–413. [PubMed] [Google Scholar]

55. Wellinger R J, Wolf A J, Zakian V A. Origin activation and formation of single-strand TG1–3 tails occur sequentially in late S phase on a yeast linear plasmid. Mol Cell Biol. 1993;13:4057–4065. [PMC free article] [PubMed] [Google Scholar]

56. Wright J H, Gottschling D E, Zakian V A. Saccharomyces telomeres assume a non-nucleosomal chromatin structure. Genes Dev. 1992;6:197–210. [PubMed] [Google Scholar]

57. Zakian V A. Structure, function, and replication of Saccharomyces cerevisiae telomeres. Annu Rev Genet. 1996;30:141–172. [PubMed] [Google Scholar]

Articles from Molecular and Cellular Biology are provided here courtesy of Taylor & Francis